University of Pardubice - 9th International Seminar NTREM (2006)

proceedings of
New Trends in Research

of Energetic Materials
Pardubice, April 1921 2006

University of Pardubice, Faculty of Chemical Technology
Institute of Energetic Materials
Sponsors
University of Pardubice, Faculty of Chemical Technology

proceedings of the 9th Seminar
New Trends in Research

held at the University of Pardubice

Pardubice, Czech Republic
April 1921 2006
intended as a meeting of students, postgraduate students, university teachers,

and young research and development workers, concerned from the whole world
This publication has not been submitted to language corrections and contributions have not
been reviewed.
Contributions of the Proceedings will be quoted in the Chemical Abstracts.
The only distributor of the present publication is the
University of Pardubice
532 10 Pardubice
Czech Republic
where the publication can be ordered or gained by exchange of similar pulications.
NTREM 06
Proceedings of New Trends in Research of Energetic Materials
Jan Ottis, Miloslav Krupka (editors).
Published by University of Pardubice, publication number UP 06-19.
Czech Republic, April, 2006.
828 pages, 150 copies.
Printed by Press Center of University of Pardubice.
Typeset by authors, checked by editor, and completed by Vt Zka
(www.typokvitek.com).
C Copyright to all papers are retained by the authors.

ISBN 80-7194-849-7
9th Seminar of the New Trends in Research of Energetic Materials

Seminar is supported by:
Schlumberger Technology Corporation, Rosharon, Texas,
European Office of Aerospace Research and Development of the USAF,
Austin Detonator, Inc., Vsetn,
Indet Safety Systems, Inc., Vsetn, a member of Nippon Kayaku group,
Explosia, Ltd., Pardubice,
BORGATA, Ltd., Praha 5,
Faculty of Chemical Technology, University of Pardubice,
OZM Research, Hrochuv Tnec.
Chairman: Prof. Svatopluk Zeman, D.Sc.
Scientific Committee:
Chairman
Dr. Adam Cumming
Members
Prof. Ang How-Ghee
Dr. Anthony J. Bellamy
Dr. Yurii M. Burov
Dr. Stanislaw Cudzilo
Prof. Manfred Held
Dr. David Jones
Prof. Mikhail Ilyushin
Prof. Thomas Klapoetke
Prof. Michel Lefebvre
Prof. Frantiek Ludvk
Prof. Andrzej Maranda
Dr. Henric stmark
Prof. Peter Politzer
Dr. William G. Proud
Prof. Valerii P. Sinditskii
Prof. Igor Tselinski
Prof. Yuanjie Shu
Dr. Muhamed Sueska
Prof. Waldemar A. Trzciski
Assoc. Prof. Pavel Vvra
Dr. Woodward Waesche
Organizing Committee:
Chairman
Assoc. Prof. Betislav Janovsk
Members
Assoc. Prof. Ladislav Lehk
Dr. Jan Jakubko
Dr. Marcela Jungov
Dr. Pavel Valenta
Dr. Iva Ulbrichov
DSTL, Sevenoaks, U.K.

National Univesity of Singapore, Singapore
Cranfield University, U.K.
Russian Academy of Sciences, Chernogolovka, Russia
Military Univ. Technol., Warsaw, Poland
EADS/TDW, Schrobenhausen, Germany
CERL, Ottawa, Canada
St. Petersburg State Inst. of Technol., Russia
Ludwig-Maximilians-Universitt, Mnchen
Royal Military Academy, Belgium
Univ. of Defence, Brno, Czech Republic
FOI, Tumba, Sweden
Univ. of New Orleans, USA
Cavendish Laboratory, Cambridge, U.K.
Mendeleev Univ. of Chem. Technol., Moscow
St. Petersburg State Inst. of Technol., Russia
Inst. of Chem. Materials, CAEP, Sichuan, China
Brodarski Inst., Zagreb, Croatia
Univ. of Pardubice, Czech Republic
SAIC, Gaineswille, USA
IEM, Univ. of Pardubice, Czech Republic

Explosia, Ltd., Pardubice, Czech Republic
Indet Safety Systems, Vsetn, Czech Republic
Univ. of Pardubice, Czech Republic
Austin Detonator, Vsetn, Czech Republic
University of Pardubice, Czech Republic
1
New Trends in Research of Energetic Materials
Czech Republic
2006
[Content]
Contents
Part 1
Invited lectures
1
European Govermental Collaboration in Energetics Research The Future

Adam Cumming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
New Diagnostic Techniques in Blast Waves

Manfred Held . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
The Thermal Behaviour of BTAW, a High Nitrogen Fuel

David Jones, K. Armstrong, T. Parekunnel, Q. Kwok . . . . . . . . . . . . . . . . . . . . . . . . . 42
New Trends in Detection of Explosives

Petr Mostak, Miroslav Stancl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Molecular Surface Electrostatic Potentials in the Computational

Characterization of Energetic Compounds
Peter Politzer, Jane Murray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Lectures
The Influence of Geometry and Material Properties on an Explosives

Gurney Velocity and Energy
Joseph Backofen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Computer Simulation of Thermal Decomposition Mechanism for

Compounds with Nitroguanidine Fragment
Eugeniya Bakhmatova, Vyacheslav Korolev, Aleksey Porollo, Tatyana Pivina . . . . . 90
Thermal Decomposition of Solid Energetic Materials

Yurii Burov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Use of Compressed Gas Gun to Investigate Response of Explosives to a

Range of Initiation Stimuli
Michael Cartwright, Andrew Lewis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
10
The Relationship between Shock Sensitivity and Morphology in Granular

RDX
Helen Czerski, William Proud, John Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
11
Guanidinium Nitroformate Salts: Possible New Oxidizers for High

Performance, Halogen Free Solid Propellants
Michael Gbel, Thomas Klaptke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Czech Republic
2006
[Content]
12
Preliminary Study of Superfine Metal Powders Effect on the Decomposition

Alexander Gromov, Elena Popenko, Alexey Sergienko, Young-Soon Kwon,
Ulrich Teipel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
13
Preparation, Properties and Application of Energetic Complex Perchlorates

of d-Metals
Mikhail Ilyushin, Igor Tselinsky, Irina Bachurina, Yurii Gruzdev,
Valery Sinditskii, Valery Serushkin, Vyacheslav Egorshin . . . . . . . . . . . . . . . . . . . . . 143
14
Balistic Stability of a Spherical Propellant: Comparison with a Flattened

Spherical Propellant
Laurence Jeunieau, Michel Lefebvre, Pierre Guillaume . . . . . . . . . . . . . . . . . . . . . . . 150
15
Synthesis and Characterization of 1,2-Dinitroguanidine (DNG) and its

Derivatives
Nikolaj Latypov, Martin Johansson, Larisa Yudina-Wahlstrm, Stefan Ek,
Carina Eldster, Patrick Goede . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
16
Energetic Characteristics of Solid Composite Propellants and Ways for

Energy Increasing
David Lempert, G. Nechiporenko, G. Manelis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
17
Mechanism of Low-Frequency Non-Acoustic Instability Phenomena From

the Point of View of the Concept of Spatial-Periodic
Alexander Lukin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
18
Stable Salts of Methylated 5-Aminotetrazoles

Carles Mir, Michael Gbel, Konstantin Karaghiosoff, Thomas Klaptke,
Jan Welch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
19
Applicability of Dynamic Mechanical and Thermal Methods in Investigation

of Ageing Processes of Double Based Propellants
Sanja Muani, Muhamed Sueska, Sanko Baka . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
20
High Energy Density Materials (HEDM): Overview, Theory and Synthetic

Efforts at FOI
Henric stmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
21
Aerobic Biodegradation of Mononitrotoluenes in Different Reactor

Configuration
Jan Pca, Martin Halecky, Rakesh Bajpai . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
22
Polyfunctional N-nitramines Structure Differentiation as a Basis for

Simulation of Their Decomposition Mechanism
Tatyana Petukhova, Victor Ivshin, Vyacheslav Korolev, Tatyana Pivina . . . . . . . . 260
23
Process Hazard Evaluation for Explosives Manufacturing Using

TEMCLEV-EX Procedures
Tadeusz Piotrowski, Tomasz Saaciski, Micha Frczak . . . . . . . . . . . . . . . . . . . . . . 268
24
Determination of the Thermal Resistance of the Emulsion Explosives

Tomasz Saaciski, Waldemar Witkowski, Andrzej Maranda,
Johann Kasperski, Barbara Gobek, Henryka Boniuk . . . . . . . . . . . . . . . . . . . . . . . . 279
3
Czech Republic
2006
[Content]
25
Reduction of Nitroaromatic Explosives by Oxygen-insensitive

NAD(P)H:Nitroreductases: Implications for their Cytotoxicity
Jonas Sarlauskas, Lina Miseviciene, Henrikas Nivinskas,
Zilvinas Anusevicius, Ausra Nemeikaite-Ceniene, Evaldas Maldutis,
Narimantas Cenas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
26
Complex Research of Al Nanopowders Produced by the Exploding Wire

Method
Valentin Sedoi, Galina Volkova, Mirslan Osmonaliev, Victor Ivanov,
Ludmila Chemezova, Alexander Alekseev . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
27
Ab Initio MD Computation of the Vibrational Relaxation Time in HE

Molecules
Alexander Selezenev, Aleksey Aleynikov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
28
Unusual Thermal Decomposition of NTO: Is It a Result of Very Strong

Bonds or Other Reasons Exist?
Valerii Sinditski, Sergei Smirnov, Valerii Egorshev . . . . . . . . . . . . . . . . . . . . . . . . . . 314
29
The principal approach to the estimation of cookoff temperature using FEM

Jakub eleovsk, Miloslav Krupka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
30
Low Temperature Activation Energy of Propellants

Stephan Wilker, Uldis Ticmanis, Gabriele Pantel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
31
Theoretical Study of the Tetrazines C-N Heterocyclic Derivatives

Zhou Yang, Shu Yuan-Jie, Wang Xin, Long Xin-Ping, Tan An-Min
32
Self-aggregation of Thermal Decomposition Products

Tomasz Zielenkiewicz, Teresa Ksiczak, Andrzej Ksiczak
. . . . . . . . . . . 356
. . . . . . . . . . . . . . . . . . 368
Only in Proceedings
33
Study of the Routes for Producing Nitroguanidine(NQ) in Aqueous Nitric

Acid Solutions. Development of the Efficient Meth
Alexandr Astratev, Dmitry Dashko, Andrei Stepanov . . . . . . . . . . . . . . . . . . . . . . . . 377
34
Fuel Dispersion in Air by Detonation Wave

Esmaeil Bazrafshan, Seyed Pourmortazavi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
35
Preliminary Study of the Behavier of Al Nano Powders Water Slurry

Compositions at Near Room Temperatures
Mikhail Laritchev, Olga Laricheva, Ilia Leipunsky, Pavel Pshechenkov,
Alexey Jigatch, Michael Kuskov, Valentin Sedoi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
36
Carbon Nanotubes as Support for Co/TiO2 Electrocatalysts for Hydrogen

Evolution
Orce Popovski, P. Paunovi, A. Dimitrov, D. Slavkov, D. Smileski,
M. Tasev, S. Hadi Jordanov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416
37
Sliding Shock Wave Causing Deformation of the Metal Plate Attached to

the Cover
Huiying Zhao, Zhaowu Shen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
Czech Republic
2006
[Content]
Part 2
Posters
38
Simulation of Coffee Mill Explosion Accident Underatex 137 Directive

Vladislav Adamk, Betislav Janovsk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
39
Explosion Hazard of Water Gel-like Systems on a Base of High Explosives

Alexei Apolenis, Ekaterina Aleshkina, Vladimir Annikov, Georgii Kozak . . . . . . . . 464
40
Heat of Melting of Nitrocompounds Measuring by Means of the Differential

Scanning Calorimetry Method
Svetlana Arinina, Georgii Kozak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473
41
Characterization and some Properties of 1-Nitroamidino-3-amino-5nitrimino-1,2,4-triazole

Alexander Astachov, Vitaliy Revenko, Ludmila Kruglyakova,
Andrew Nefedov, Eduard Buka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
42
The Use of a Kenics Static Mixer for the Reaction of TNT with NaOCl
Anthony Bellamy, Victoria Lomax, Kate Read . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491
43
Investigation of Stabilizing Properties of Triphenylamine (TPA)

Jan Bladek, Stanislaw Cudzilo, Sylwia Pietrzyk, Stephan Wilker . . . . . . . . . . . . . . . 500
44
Some Properties of a Composite Propellant Premix

Daniel Buczkowski, Bogdan Florczak, Tomasz Saaciski . . . . . . . . . . . . . . . . . . . . . 507
45
Thermophysical Properties of Thermoplastic Composite Propellants Based

on Modified PVC Binder
Vladica Bozic, Marko Milos, Djordje Blagojevic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513
46
Thermal Decomposition of the trans-1,4,5,8-Tetranitro-1,4,5,8tetraazadecalin

Yurii Burov, Irina Kucherova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
47
Observation of Changes in Chosen Properties of Plastic Explosives During

Artificial Ageing
Martina Chovancov, Peter Oko, Jozef Lopch, Albta Pechov . . . . . . . . . . . . . 529
48
Hexanitrostilbene - Density Dependend Sensitivity

Dominik Clment, Karl Rudolf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540
49
Self-sustaining reductive defluorination of (CF)n and characterization of the

reaction products
Stanislaw Cudzilo, Andrzej Huczko, Mateusz Szala, Michal Bystrzejewski . . . . . . . 544
50
Insensitive Environmentally-compatible and Non-toxic Replacements for

RDX in EMs New Method Of Synthesis Of DAT
Chaza Darwich, Thomas Klaptke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551
51
The Nitration Reaction of Isowurtzitane Acyl Derivatives

Maciej Duda, Pawe Maksimowski, Wincenty Skupiski . . . . . . . . . . . . . . . . . . . . . . 561
5
Czech Republic
2006
[Content]
52
Initiation of Deflagration and Detonation in the Metal Dust/Air Systems

Irina Egorova, Boris Kondrikov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569
53
JET Velocity and the Effects of the Shaped Charge Perforator for Use in
the Oil and Gas Well
Zvonimir Ester, Mario Dobrilovi, Branimir Jankovi, Zvonimir Tomii,
Mile Odak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581
54
Thermochemical Calculations and Experimantal Study of Composite

Propellants Containing DADNE or NTO
Bogdan Florczak, Andrzej Orzechowski, Tomasz Saaciski . . . . . . . . . . . . . . . . . . . 587
55
Implementation of the Absolute Method in the Ballistic Analisys of

Gunpowder
Tomasz Gawor, Tomasz Wolszakiewicz, D. Korsze . . . . . . . . . . . . . . . . . . . . . . . . . . 598
56
Green Polymer-Bound Explosive (PBX) for Laser

Mikhail Ilyushin, Igor Tselinsky, Irina Shugalei, Anatoliy Chernay,
Valeria Toftunova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602
57
Synthesis of 2,4-dinitro-2,4-diazapentane from methylnitramine and

dihalogenmethanes
Zdenk Jalov, Pavel Mareek, Kamil Dudek, Mat Ondrejkovi, Frantiek Lika
608
58
Application of Supercritical Fluid Extraction for the Isolation of

Nitrocompounds
Vra Jeov, Tom Fojtk, Ale Eisner, Martin Adam, Karel Ventura . . . . . . . . . 613
59
Possibilities of Degradations Process of Nitroaromatic Compounds

Vra Jeov, Tom Fojtk, Karel Ventura, Vt Matj, Eva Kyslkov,
Simona Voshlov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 618
60
Preparation of Low-melting Compositions Based on 1,3,3-Trinitroazetidine

Dmitrii Katorov, N. Vyalova, G. Rudakov, A. Ladonin, V. Zhilin, E. Veselova . . 624
61
Salts OF 5-Nitrotetrazolate
Thomas Klapoetke, Peter Mayer, Kurt Polborn, Jan Wlech . . . . . . . . . . . . . . . . . . . 631
62
5,5-Bis-(1H-tetrazolyl)amine (H2bta): A Promising Ligand in Novel

Copper Based Priming Charges (PC)
Thomas Klaptke, Peter Meyer, Kurt Polborn, Jrg Stierstorfer, Jan Weigand . . 641
63
Emulsive Explosive Sensitized by Inorganic Peroxides

Igor L. Kovalenko, Alexandr V. Kuprin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 652
64
Detonability of Nitromethane Mixtures with Azideethanol

George Kozak, Vlada Raikova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 657
65
Effect of Rubbing Surfaces Velocity on the Results of Friction Sensitivity

Testing of Explosives
Miloslav Krupka, Jan Horkel, Jakub Selesovsky . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668
Czech Republic
2006
[Content]
66
Thermal Effects of a Slow Thermal Decomposition and Inflammation of

Nitramines and Its Mixtures with Al, B and Zr Powder
Valeriy Lashkov, A. Strikanov, A. Tikhonova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 673
67
Sensitivity Research of Explosive Mixtures Containing Nitroguanidine and

3-Nitro-1,2,4-triazol-5-one
Andrzej Maranda, Jerzy Nowaczewski, Andrzej Orzechowski, Dorota Powaa . . . . 680
68
Cytotoxicity of Novel Nitrotriazole and Trinitroaniline Explosives: The Role

of Oxidative Stress
Ausra Nemeikaite-Ceniene, Jonas Sarlauskas, Henrikas Nivinskas,
Zilvinas Anusevicius, Narimantas Cenas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689
69
Optimalization of Crystallization of 1,1-Diamino-2,2-dinitroethylene

Andrzej Orzechowski, Dorota Powaa, Bogdan Florczak, Andrzej Maranda,
Jerzy Nowaczewski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693
70
Selective Modifiers of Thermal Decomposition and Combustion of

Composite Solid Rocket Propellants Containing HMX
E. Popenko, A. Sergienko, E. Sokolov, D. Petrov, V. Shandakov,
Y. Zharinov, A. Lobanova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701
71
Crystal Structures of Selected Energetic Materials Calculated by Molecular

Simulations
Miroslav Pospil, Pavel Vvra, Petr Kov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712
72
Comparison of Coating with Various Deterrents

Pavel Prchal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 719
73
Treatment of WastewatersContaining Nitrate Esters

Jan Skldal, Martin Kouba, Martina Mudrukov, Miloslav Slezk . . . . . . . . . . . . . 723
74
The Use of Low Pressure Closed Vessel and Rocket Motor for Measurements
of Burning Rate of Rocket Solid Propellants.
Petr Stojan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 730
75
Experimental Rating of Surge Withstand Capability of Reinforced Concrete

to the Standard Concrete with Dynamic Load
Ji toller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 736
76
An Experimental Study of Detonation Transmissionfrom Hydrogen-Oxygen

to Hydrogen-Air Mixture
Andrzej Teodorczyk, A. Dabkowski, M. Bratowski . . . . . . . . . . . . . . . . . . . . . . . . . . . 742
77
On Some Methods of Determination of the Detonation Energy of Explosives

Waldemar A. Trzciski, Jzef Paszula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 748
78
Regarding the Fundamental Problem Solving of Interior Ballistics in the

Case of Flegmatised Powders
Titica Vasile, Cristian Barbu, Doru Safta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 756
79
Numerical Description of the Effects in Gas Generator

Tomasz Wolszakiewicz, Tomasz Gawor, Zbygniew Walenta . . . . . . . . . . . . . . . . . . . . 764
Czech Republic
2006
[Content]
80
The Mechanisms of Thermal Decomposition of Simple Hydronitrogen

Compounds
Ying Xiong, Yuanjie Shu, Ge Zhou, Xinfeng Wang, Anming Tian,
Ning-Bew Wong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769
81
Influencing Parameters on HE Projectiles with Natural Fragmentation

Berko Zecevic, Jasmin Terzic, Alan Catovic, Sabina Serdarevic-Kadic . . . . . . . . . . 780
82
New Findings in the Electric Spark Sensitivity of Nitramines. Part I.

Aspects of Molecular Structure
Svatopluk Zeman, Ji Ko, Vojtch Pelikn, Ji Majzlk, Zdenk Friedl . . . . . . . 796
83
New Findings in the Electric Spark Sensitivity of Nitramines. Part II.

A Problem of Hot Spots
Svatopluk Zeman, Ji Ko, Vojtch Pelikn, Ji Majzlk . . . . . . . . . . . . . . . . . . . . 810
84
Theoretical Study on the Decomposition of 1,1-Diamino-2,2-Dinitroethylene

Hehou Zong, Yuanjie Shu, Yigang Huang, Xinfeng Wang . . . . . . . . . . . . . . . . . . . . 816
Abstract
85
Polymorfic Transformations and Thermal Decomposition of Ammonium

Nitrate at Thermal Analysis Conditions
Valeriy Lashkov, E. Egorychev, A. Strikanov, A. Tikhonova, V. Rybakov . . . . . . . 825
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 826
situation of the middle European countries, Balkan and former Soviet Union, no fee was and
is asked from the participants. This concept has been maintained up to now.
Czech Republic
2006
[Content]
In the year that has passed from the eigtth seminar, we lost some of the well-know
figures in the international energetic community. The greatest loss for us is the demise of Dr.
Preface
Fred Volk in September 2005; he was a very familiar person at international meetings and
cooperation.
Fred, who was know by his unselfish help given young scientific-research
This is the ninth seminar New Trends in Research of Energetic Materials and the eighth inworkers ternational
in the filed
ofthe
energetic
materials
wasDepartment
actively ofengaged
in Technology
the process of
one in
series organized
by theand
former
Theory and
of
Explosives,
now
the
Institute
of
Energetic
Materials
(IEM)
at
the
University
Parinvolving the Czech workplaces of this branch into cooperation within Europeanofstructures,
dubice.
The
original
purpose
of
these
meetings
was
to
teach
young
research
workers
how
to him
contributed significantly to moulding the character of our April meetings. We will miss
present their results in front of scientific audience. With regard to economic situation of the
as not only
excellent
scientist
but
a Soviet
very Union,
good nofriend.
loss
middlean
European
countries,
Balkan
andalso
former
fee was Another
and is asked
fromfor
the us in
December
2005 wasThis
theconcept
death of
of Czech
of Blasting Techniques and
participants.
haspresident
been maintained
up toAssociation
now.
Pyrotechnics and external IEM assistant professor Dr. Vclav Tamchyna. In his case, IEM
the year that has passed from the eigtth seminar, we lost some of the well-know figures in
has lost In
direct
connection with the blaster community of the Czech Republic and Europe as
the international energetic community. The greatest loss for us is the demise of Dr. Fred Volk
well. It need
not
be emphasised
losses
areatappreciable
it will
some time to
in September
2005; he was athat
verythese
familiar
person
international and
meetings
andtake
cooperation.
know
by his unselfish
help given
young continue,
scientific-research
workers
in the
field of is its
get overFred,
themwho
at was
least
partially.
However,
life must
and our
ninth
Seminar
energetic materials and was actively engaged in the process of involving the Czech workplaces of
constituent.
this branch into cooperation within European structures, contributed significantly to moulding
the character
of ourseminars
April meetings.
will miss
as not only
an excellent
scientist but
also
Realization
of our
wouldWenever
behim
possible
without
the generous
support
of
a very good friend. Another loss for us in December 2005 was the death of president of Czech
many sponsoring institution and individuals. In the case of this ninth Seminar a very
Association of Blasting Techniques and Pyrotechnics and external IEM assistant professor Dr.
significant
financial
support
provided
by with
the the
Schlumberger
Reservoir
Vclav
Tamchyna.
In hishas
case,been
IEM regularly
has lost direct
connection
blaster community
Completions
theItEuropean
of Aerospace
Research
&
of the from
Czech Rosharon
Republic and(Texas),
Europe asby
well.
need not beOffice
emphasised
that these losses
are
appreciable
and
it will in
take
some time
get overDetonator
them at least
partially.
However,
mustSafety
Development
of the
USAF
London,
bytoAustin
Comp.
(Vsetn),
bylife
Indet
ninth Seminar is its constituent.
Systems continue,
Comp.and
(a our
member
of Nippon Kayaku group in Vsetn), by Explosia, Ltd.,
Pardubice,
by the ofCzech
company
OZM
in Hrochv
and by
the of
BORGATA,
Realization
our seminars
would
never Ltd.
be possible
without Tnec
the generous
support
many
sponsoring
institution
and individuals.
In the casesupport
of this ninth
a very significant
fiLtd., Prague.
Almost
traditionally,
some financial
has Seminar
been provided
by Dr. Oldrich
nancial
support
has
been
regularly
provided
by
the
Schlumberger
Technology
Corporation
from
Machacek, president of the Universal Tech. Corporation, Dallas, who (as chance would have
Rosharon (Texas), by the European Office of Aerospace Research & Development of the USAF
it) is onein of
the first
PhD Detonator
graduatesComp.
from (Vsetn),
IEM. Also
ourSafety
Faculty
of Chemical
Technology
London,
by Austin
by Indet
Systems
Comp. (a member
markedlyof Nippon
contributed
the costs
of thisLtd.,
Seminar.
The
efficient
help inOZM
ensuring
Kayakutowards
group in Vsetn),
by Explosia,
Pardubice,
by the
Czech company
Hrochv Tnec
and by
Prague.from
Almost
some fi- and
smooth Ltd.
and insuccessful
course
of the
theBORGATA,
meeting Ltd.,
obtained
all traditionally,
these institutions
nancial support has been provided by Dr. Oldrich Machacek, president of the Universal Tech.
individuals
is gratefully acknowledged. We greatly appreciate it that thanks to this
Corporation, Dallas, who (as chance would have it) is one of the first Ph.D. graduates from
sponsoring
all
theour
above-mentioned
specifics
of the
previous
seminars
canthe
be costs
maintained.
IEM. Also
Faculty of Chemical
Technology
markedly
contributed
towards
of this
Seminar. The efficient help in ensuring smooth and successful course of the meeting obtained
Finally,
I wish to thank the members of the Scientific Committee, the authors of all the
from all these institutions and individuals is gratefully acknowledged. We greatly appreciate it
seminar that
papers
and,
but not least,
the participants
seminar,seminars
for itscan
success
thanks
to last
this sponsoring
all theyou,
above-mentioned
specificsofofthis
the previous
maintained.
and its be
influence
on the continued success and growth of all future meetings at our
University of young people and university teachers working in the field of all kinds of
Finally, I wish to thank the members of the Scientific Committee, the authors of all the seminar
energeticpapers
materials.
and, last but not least, you, the participants of this seminar, for its success and its
influence on the continued success and growth of all future meetings at our University of young
people and university teachers working in the field of all kinds of energetic materials.
Pardubice, March 25th, 2006
Pardubice, March 25, 2006
Svatopluk
eman
SvatoplukZZeman
Invited lectures
Czech Republic
2006
[Content]
EUROPEAN GOVERNMENTAL COLLABORATION IN

ENERGETICS RESEARCH THE FUTURE
A S Cumming
Dstl Fort Halstead
Sevenoaks
Kent
UK
ascumming@dstl.gov.uk
Abstract
The level of joint working in Energetics has increased significantly in the last few
decades. It has developed from informal meetings and discussions to formal
intergovernmental agreements on joint programmes. The need to and benefits of
working together have been obvious and this has produced greater interactions. This is
changing again with the setting up of the European Defence Agency, and while there is
some confusion over what this means, it will provide new routes to collaboration. This
paper will describe some of the achievements as well as discussing the possibilities for
the future.
INTRODUCTION
In the mid-eighties formal collaboration was limited. There were links, usually for
information exchange, and scientists met at conferences where they presented on work in
progress and discussed common problems. There were formal links within NATO and also
some other agreements in place. One reason for this was that the most nations were able to
sustain their own needs through their own investment. The needs differed as did the
investments, but there was no awareness of major limitations.
At the same time equipment was becoming increasingly complex and expensive so that
while is was possible to have national main battle tanks for example in Germany, France the
UK and the US, continuing that development was becoming more difficult to do. This
increasingly led to joint projects and the same time policy looked towards the privatisation of
defence manufacturing. The UK privatised the Ordnance Factories in the mid eighties and
also turned the research establishments into an agency aimed at improving efficiency and
providing value for money. One aim was to assist in the transfer of technology for civil use.
It was realised that better value for many could be gained by working with partners on
joint research programmes, and so existing links were developed with Memoranda of
1
Adam Cumming
11
Czech Republic
2006
[Content]
Understanding to do much more than exchange information but to include staff exchanges
and joint programmes with financial and delivery commitment.
However the existing links did not meet all the perceived needs, and a new Group the
Independent European Programme Group, IEPG, was set up to develop joint research
programmes as well as trying to harmonise what would now be called capability
requirements.
Energetics played a major part in this with the formation of Technical Area (TA) 12,
looking at gun propellants, TA25 with three programmes covering extrusion of energetic
compositions, pressed shaped charges and synthesis and characterisation of new materials.
Each programme required a separate MOU leading to delays and complications so that
attempts were made to provide a more flexible route to collaboration in Europe. This led to
the CEPAs and to WEAG.
WEAG and the CEPAs
The first group of CEPAs did not include Energetics. The made use of the EUCLID
MOU, which was based round competition and the assumption of industrial participation. It
had the draw back of not allowing closed groups and making the participation of government
agencies more difficult.
However when CEPA 14 was started it was possible to devise and begin programmes on
new Rocket Propellant Technology and Insensitive Munition Technology. While these were
started two further developments made collaboration easier. First the creation of the
Research Cell in Brussels to support the activity of each CEPA and also to place central
contracts for the agreed programmes, and second the signature of the THALES MOU aimed
at Joint Programmes amongst government laboratories. This was rapidly followed by the
SOCRATE MOU allowing other nations, non-NATO members such as Sweden to take part.
A third factor was the introduction of the EUROFINDER process where industries were
encouraged to propose programmes for inclusion. There were partially industrially funded
and closed, being limited to the proposing nations only.
Finally the EUROPA MOU was developed and signed, streamlining the processes and
providing for closed programmes amongst groups of members. This provided the final and
most effective set of tools for use in collaboration.
Within CEPA 14 this provided a spur to new work and the development of joint
understanding and information exchange for the development of joint programmes. For
nearly ten years the CEPA acted as a focus a clearing house for most of European Energetics
research. One effect was the reduction of activity within many of the older agreements,
though specific items might still be developed by those methods.
12
Czech Republic
2006
[Content]
The main problem was always the limited amount of collaboration possible when there
was no harmonisation in requirements. That harmonisation might include capability need,
finance, timescales and other factors. The more nations involved the harder is remains to
agree programmes beyond a limited level. This is illustrated by the Venn diagram below
where the degree of total overlap is very small. (Fig 1)
The limitations in the WEAG process made it harder to improve on this, and it meant a
practical limit of about 5% of the available research funding. The only effective way to
improve on this is by improving the coordination of the capability requirements so that the
factors listed above are dealt with at an early stage. If this is done then the level of mutual
involvement will increase significantly,
R&T Priorities Multilateral
Fig. 1.
ACHIEVEMENTS OF CEPA 14
Despite all the problems outlined above the CEPA achieved a significant amount and
even though it has now officially ended, that achievement continues through its legacy.
Two programmes were mentioned above. The IM programme added significantly to the
European capability to manage Munition vulnerability. Within the UK it provided the
technology to ensure that a new and major system met our IM Policy requirements. In the
same the way the programme on LOVA ammunition assisted in the decision to reinvest in
such technology with the UK.
The CEPA has provided the means of holding Workshops including the European
Survivability Workshop; Workshops on Environmental Effects on Munitions of Out of Area
Operations, High Energy Density Materials, The Effect of Shock on Energetics; Hybrid
Missile Propulsion and others. Many of these have led to proposals for new joint research
programmes.
Adam Cumming
13
Czech Republic
2006
[Content]
In addition there are running programmes on Flares, the development of a European

Virtual Underwater Testing Centre for munitions, Fragment and Bullet attack on Munitions,
again supporting IM, Particle Size optimisation and of course the Synthesis of Nitro
compounds, involving the Czech Republic. All these will continue and older ones have
already proposed new programmes to continue the relevant research. This is the legacy of
the CEPA.
The Future
The European Defence Agency is in its infancy, and has yet to develop it processes. It is
an intergovernmental organisation, not a supra-governmental one, and exists to support the
member states. There remain issues over the dealing with nations who are signatories of
EUROPA but not in the EU, and hence EDA (Norway, Turkey and Romania) or have
refused to join the EDA (Denmark). The EDA has inherited most of the existing WEAG
programme and is attempting to produce mechanisms to develop this. The major mechanism
is by means of Capability/Technology Areas or CapTechs, with national coordinators
(CNCs) supported by an EDA Moderator and by national and industrial experts.
The EDA will have very limited central funding for the foreseeable future and it
therefore dependent on the requirements and policies of the member states. One of its main
roles is to begin the harmonisation of national requirements so that more extensive and
effective collaboration can be undertaken. This is not a simple task, and is made more
complex by the variation in money available for R&T. The relative levels are shown in
Figure two below.
5000
MUSD
97
%
77
%
4000
3000
2000
1000
0
UK
FR
GE
IT
SP
SW
PL
NL
TU
GR
NO
CZ
HU
AU
BE
FI
PO
DK
LU
Fig. 2. R& Figure 2 Expenditure Across WEAG
14
Czech Republic
2006
[Content]
The EUROPA MOU still exists and will continue to exist having been chosen for certain
other activities outside the EDA. It requires some modification but that is being done. It
may be extended to cover other non-EDA countries.
There are, therefore mechanisms for collaboration in Europe, at least as good as those
that formerly existed. Within the Energetics community, those involved with CEPA 14 have
proposed an Expert Group covering the same area and supporting both EDA CapTech
activities as well as those outside it. How this develops depends on the way the various
needs. However, for most nations it is agreed that the EDA will be the prime method of
operation, but other methods may be employed if they are more appropriate.
In short it will be possible to work as closely together and opportunities should be
sought and developed. The mechanism employed to deliver the result may vary but does
exist and the processes will continue to develop to meet the need to provide results for the
national MODs. It may look complex and uncertain, but the same was true when WEAG
was created and that grew and developed to meet real requirements. For that reason it is
worth looking back to see how we arrived where we now are, and to reassure ourselves that
there has been real progress. No mechanism is perfect or remains useful in all
circumstances. It must evolve to meet new needs or be replaced by methods that do meet
requirements, learning from the experience.
In this changing Europe there is no reason to suppose that these changes will not assist in
providing the defence capabilities we require.
Adam Cumming
15
Czech Republic
2006
[Content]
NEW DIAGNOSTIC TECHNIQUES IN BLAST WAVES
Prof. Dr. Manfred Held

TDW/EADS Schrobenhausen
E-Mail: manfred.held@eads.com
Tel.: 49-8252-996-345;
Fax: 49-8252-996-126
Abstract
The author has developed and used methods to measure the blast contour of not
spherical charges and the momentum distribution of antitank mines under different
ground conditions lying on the ground, levelled to the ground or buried in the ground.
Further he has defined a diagnostic method to measure the combined transferred
impulse of fragments and blast loads on steel-plates.
1.
IMPULSE LOAD IS DOMINANT
The pressure durations of the positive pressure history phase (Fig. 1) of detonating high
explosive charges in the near field at less than two (Z < R/W1/3) are shorter than 2 ms (Fig.
2). In such short times the maximum pressure is not responsible for any damage, because the
movement of the target is minimal in this short time interval. But the transferred momentum
to the target structure can be compared to a ballistic pendulum. Therefore the transferred
momentum has to be measured. The maximum pressure has no relevance to the damage
caused by small charges history in the fireball or if the charge is lying in sand or is confined.
High frequency pressure gauges have also some prices. They can not be used in large
quantities for high angle resolutions, what is easily possible for the momentum gauges which
the author has used and is using as a diagnostic tool.
2.
DIAGNOSTIC WITH THE MOMENTUM METHOD
Momentum gauges are arranged on a steel band around the charge (Fig. 3). The heights
of the momentum gauges and of the high explosive charge has to be exactly levelled, so that
only horizontal forces are acting on the rectangular shaped gauges. At closer distances of 0,5
m and 0,75 m the gauges of 2,5 cm5 cm are oriented with the smaller surface of 12,5 cm to
the charge. At 0,5 m distance 60 steel momentum gauges are used with 490 g weight in the
semi-circle, which gives an angle resolution of 3, and at 0,75 m distance 90 aluminium
16
Czech Republic
2006
[Content]
momentum gauges with 172 g weight, which results in 2 angle resolution (Fig. 3). They
were supported by a 3 mm thick and 30 mm wide steelband, where the cross section to the
charge was sharpened with an angle of about 45.
At the 1,0 m and 1,5 m radii the same gauges are oriented with their quadratic cross
sections of 25 cm to the high explosive charge in the centre, which gives also the angle
resolutions of 3, respectively 2 at 1,0 m and 1,5 m radius. A typical picture of the test setup
with the cylindrical charge in the centre of the two half circles is shown in Fig. 4. By the
horizontally acting blast wave the momentum blocks are thrown radially away, where the
first impacts in the sand are measured on the levelled ground.
To get the transferred momentum the velocity of the momentum gauges has to be found
out. The velocity can be easily measured from their displacement D. The falling time t from
the height H is constant and is given by the equation t = 2 H / g , independent from the
horizontally transferred velocities.
For the height H of 1 m the falling time t is 0,45 sec (t =
2 1(m) / 9,81(m / sec ) ).
The trajectories of exact horizontally loaded momentum gauges with different initial
velocities are shown in Fig. 5. Therefore the horizontal displacement D is direct proportional
to the velocity v (Fig. 6).
This method allows to measure easily the transferred momentum M, respectively the
impulse of non-spherical charges with very high angle resolution and with very low
investment costs. Also the momentum distribution of confined charges can be very well
measured with this method.
It should be shortly summarised the by the author used blast wave terminologys with
the corresponding dimensions.
Pressure p = force/area = F/A
(1 kp/cm = 1 bar ~ 10 kgm/s/cm = 105 kgm/s/m = 105 Pa)

Impulse I, respectively momentum M
Impulse density
= mass x velocity = mv
(kgm/s)
= force x time
(kps)
= Ft
ID = I/A = mv/A = Ft/A
(1 kp s/cm ~ 10 kgm/s/cm = 105 kg m/s/m = 105 Pa s)

Scaled impulse density IDS = ID/W1/3
Manfred Held
kgm/s/(cm W1/3) = 105 Pa s/ kg1/3
17
3.
Czech Republic
2006
[Content]
TYPICALLY NEAR FIELD BLAST MOMENTUM RESULTS OF

NON-SPHERICAL CHARGES
The thrown distances of the momentum blocks on the ground are shown, as an example
in Fig. 7, where the comparison is made between plastic bonded charges of HMX charge
(left side) to RDX (right side). The impression is correct, that the differences for the 1 kg
high explosive charges are only marginal.
The results can be much better compared in the diagrams of the impulse density as a
function of the polar angle in the semi-circles, where 0 is the axis of the charge in
detonation direction, 90 exactly the radial load and 180 the axis of the charge opposite to
the detonation direction. In these diagrams the characteristics of the blast momentum of the
cylindrical charges with the length to diameter ratio of 2/1 are nearly identical as shown at
1,0 m (Fig. 8) and 1,5 m distance (Fig. 9). In the axial direction (1) surprisingly high values
are achieved, but only in a very narrow angle of less than 10. The minima (2) from 15 to
30 are lying beneath 0,01 IDS. Values beneath 0,01 IDS are not taken into account, because
no damage will be more caused. By the interaction of the axial and radial wave a small
maximum (3) in the range between 40 and 50 is generated. It follows again a minimum (4)
in the angle around 65. Then the impulse density rises exponentially (linear in the log
diagram) to the radial maximum (5) at the angle of 85 90. The rearwards first minima (6)
are typically around 120, followed by a small maximum (7) at 130 and the next minima (8)
at around 155 and finally the rearward impulses (9) against the detonation directions.
4.
BLAST MOMENTA OF DIFFERENT CHARGE COMPOSITIONS
The direct comparison of the charges KS 33 (90% HMX and 10% Binder) and KS 13
(88% RDX and 12% Binder) shows, that the HMX charge just presents a little higher
impulse densities in some regions. Generally the advantage of the more powerful HMX
charge in the blast impulse output seems to be only very small. It has to be mentioned, that
the logarithmic scale, which is necessary for the large differences of the impulse densities in
the different directions is strongly levelling out small differences.
As standard charge was used the pasty flexible or formable charge SK DM 12 B 1,
which consists of 87% PETN and 13% lubricant (Fig. 10). At 1 m distance the cast
TNT/RDX 35/65 charge seems very similar, if one takes out some random noises of
individual momentum gauges. The comparison at 1,5 m gives a slight advantage for the
TNT/RDX 35/65 charge.
A comparison was also made with aluminised high explosive charges with different
binder systems. Both charges had 67% RDX; 28% aluminium and 15% binder. The KS 22a
charge had an HTPP and the KS 22S a silicon binder type. The blast contours at 1,0 m and
1,5 m seems to show that the silicon binder system gives slightly higher impulse density
values (Fig. 11).
The composition B type TNT/RDX 35/65 charge is also compared to a plastic bonded
aluminised high explosive charge from type KS 22 A, which contains a RDX/Al/B 67/28/15
18
Czech Republic
2006
[Content]
(HTPP) composition. At 1 m distance (Fig. 12) the impulse densities look very similar, but
at 1,5 m distance the aluminised charge has definitely less impulse density values.
A very interesting aspect was found on 4 tests with not aluminised and aluminised high
explosive charges, once containing 90% RDX and 10% binder and the other type 70% RDX,
20% aluminium and also 10% binder. If the charges are initiated by a booster with 37 mm
diameter, the impulses were generally lower compared to a stronger initiation. In 1 m
distance the aluminised charge looks a little beneath the not aluminised charge (Fig. 13) but
stronger or more pronounced difference exists at 1,5 m distance. If the charges are strongly
initiated by a 10 mm thick disc of PBX N9 over the full end surface of 75 mm, the
aluminised RH 29 charge looks a bit better at 1 m distance, but definitely better at 1,5 m
distance (Fig. 14).
For cylindrical charges the impulse densities are not regularly decreasing with increasing
distance (Fig. 15 and Fig. 16), as the simple impulse equation for spheres would predict. The
narrow axial blast fan in detonation direction has to be specially treated. The bridge wave
sectors seem to be more filled up on costs of the axial and radial blast waves at larger
distances. The radial blast wave seems to be stronger reduced than linear. This are few
general remarks on the distance dependence of the blast impulse and finally for the damage
behaviour.
5.
CUBICAL CHARGE AND CHARGES WITH DIFFERENT L/D

RATIOS
Cubical charges
1 kg heavy cubical charges made from the pasty high explosive type Seismoplast or SK
DM 12 B1 are also fired under two test conditions in half circles of 1,0 m and 1,5 m
distance. Once the cube was initiated on the flat side in the centre and one at the corner point
oriented in the semi-circle in such a way, that in the first case the flat areas were oriented to
0, 90 and 180 (reverse direction), respectively to the diagonal directions.
Corresponding to the two different orientations, the momentum contours are remarkably
different (Fig. 17 and Fig. 18). In the case 1 the maxima are oriented in the flat areas. By the
interactions of these strong blast waves smaller maxima are just build-in the diagonal
directions at 45 and 135. Extreme low impulse density values are registered between these
maxima (see Fig. 17). If the cube is oriented in the semicircles in a diagonal direction, then
no maxima are measured and generally a lower level of impulse density values can be
analysed (see Fig. 18).
Different L/D ratios

Here also the pasty, high explosive charge of 86% PETN and 14% lubricant were taken.
The 1 kg heavy charges are formed in cylinders of the L/D ratios of 0,5, 1, 2 and 4. All these
charge Types are one end centrally initiated by a number 8 electric detonator (Fig. 19).
Manfred Held
19
Czech Republic
2006
[Content]
In all cases the specific impulse density has a sharp maximum around 0 or in the
forward axial direction at 0,5 m distance (Fig. 20). The contours are described in detail for
the L/D = 1 charge. The value falls steeply down around to 0,03 at around 30. The
interaction of the axial and radial blast waves give a small maximum with 0,07 in the bridge
wave around 40. These then follows a second minimum, lower than 0,02. From this very
low value the impulse rises to a shoulder at 78 with a value of 0,27. Then it increases
further to 0,4 at around 90. After this the pressure decreases in the rearward direction,
where the total contour is relatively smooth around 0,05 to 0,06 in the range of 135 to 165.
Then it rises to the rearward direction near the axis.
The contours look surprisingly very similar, still some differences exist. Some main
features are presented in the following 2 tables and in the Figures 21 and 22 at the distances
of 0,50 m and 0,75 m.
Table 0,50 m distance

Geometry
L/D
M radial
-1
-2
Cone angles for IDS 0,1
kg_m_s cm _kg
-1/3
Forward F
Radial Ra
Rearward R
0,5
0,16
2 x 37
30
2 x 43
0,40
2 x 20
50
2 x 10
0,40
2 x 20
38
2 x 15
0,20
2 x 25
52
2 x 40
These values are influenced by the scatter of the data, but they show the clear general
trends. The radial Mradial impulse is the largest for the L/D = 1 and L/D = 2 ratios with 0,4 at
0,5 m distance, as expected. The larger peak value in the case of 0,75 m distance for the L/D
= 1 is a little surprising.
Table 0,75 distance

Geometry
L/D
20
M radial
-1
-2
kg_m_s cm _kg
Cone angles for IDS 0,1

-1/3
Forward F
Radial Ra
Rearward R
0,5
0,16
2 x 29
12
2 x 33
0,40
2 x 10
27
0,25
2 x 7
27
2 x 12
0,18
2 x 25
13
2 x 20
Czech Republic
2006
[Content]
As expected, the forward and rearward cone angle has the largest values for disc shaped
charge with the L/D = 0,5. Surprising is, that the second largest angles in these directions are
achieved by the charge with the L/D = 4.
The radial blast angle around the 90 direction is wider for the L/D = 1 and L/D = 2. The
large opening angle of 52 is not very well understood for the L/D = 4 at 0,5 m distance up
to now.
For high radial blast loads the shapes L/D = 1 and L/D = 2 are the favourites, which have
also large radial cone angles.
6.
CHARGES WITH DIFFERENT CONFINEMENTS
Also the pasty compound, called Seismoplast for commercial applications or SK DM

12 B1 in military use, with L/D ratios of 2:1, respectively 1:1, with 75 mm diameter and
150 mm length, respectively 95 mm diameter and 95 mm length were used for the high
explosive charge. The charge weight was constantly 1 kg. The charges were fired unconfined
and confined with steel casings of 1 mm, 3 mm and 10 mm (Fig. 23). The charges were
initiated in the axis at one end-surface (Fig. 24).
The fragments of casings naturally hit these frames, which were distorted after each test
with confined charges (Fig. 25). A number of momentum gauges in forward and rearward
axial directions, and in the radial direction were also hit by fragments (Fig. 26).
In figure 27 the specific impulse densities of the confined cylindrical charges are
compared to bare cylindrical high explosive charges in the distance of 0,50 m and the
geometry L/D 2:1.
These detailed structures of the blast wave in the near field from the forward direction
with the narrow axial blast fan, the large minimum in the forward bridge-wave direction, the
maximum of the radially expanding blast wave to the second minimum in rearward bridgewave direction are already discussed.
If casings of steel are attached, the blast contour is changed by the additional fragment
impacts on the momentum gauges, which was expected.
This is caused in the axial forward and rearward directions by the endplates fragments of
the charge, and by the fragments of the cylindrical confinement, expanding radially in
perpendicular direction to the charge axis. By the Taylor angle they are pushed a little in the
Manfred Held
21
Czech Republic
2006
[Content]
detonation direction which means in the range of 80. The transferred momenta in these
areas covered by fragment covered areas are much higher by the combined momenta of the
fragment impacts and the blast wave, although the blast wave impulses are partially reduced
in these areas (Held 2003b). But the blast waves are continuously increasing, in the bridge
wave fields, as shown by the impulse density diagrams from 1 mm to 3 mm and 10 mm
confinement.
The reason for this is, that the products of explosives cannot expand fast axially and
radially due to the confinements, as on unconfined charges, because then the bridge wave
gaps are much more filled up with the expanding reaction products. Therefore the blast loads
are strongly increasing with confinements in these directions (Fig. 28).
In summary: confined cylindrical charges are increasing the momenta in all directions by
the fragments in axial and radial, and by the enhanced blast wave in the diagonal directions.
This investigation shows that the damage potential of confined charges in free air is
remarkably increased by confinements, because in the fragment zones the momenta are
increased by the larger loads, caused by the fragments, and the bridgewave areas are filled
up by factors of 5 to 10 to higher blast momenta, compared to unconfined charges.
7.
COMBINED FRAGMENT AND BLAST LOAD
Momentum sleds were used to measure the combined blast and fragment loads. For all
described tests cylindrical C4 charges of 60 mm diameter and 290 mm length were used,
giving an explosive weight of about 1,2 kg. These were surrounded by steel casings with
three different thicknesses 2,5 mm, 5 mm and 10 mm. For reference of the airblast load
alone, a bare charge in a very thin plastic tube was also used. The radially transferred
momenta of the blast and fragments were measured with sleds. The front of each sled
consisted of a 15 mm thick mild steel plate, 500 mm in height and 600 mm in width. The
sleds were fixed on roller bearings. The friction of the sleds was found to be less than 10 N.
The velocities of the 51,5 kg sleds were measured with electronic displacement transducers
(wire position sensors) and with video cameras with 25 frames/sec, but 0,1 ms exposure
times on an extended rod with distance marks. The vertical positions of the target plates with
respect to the charge positions were arranged according to the Taylor angles of the fragment
trajectories ( = arc sin Fr/2d) to ensure maximum fragment impacts. For each casing
thickness, two tests were conducted, one with two sleds at 1,0 m and 1,5 m and the second at
2,0 m and at 3,0 m stand-off distances (Fig. 29). The sled rails were mounted on a heavy
concrete block, as shown in Fig. 30. The front side of the concrete block and the edge of the
tracks were protected by a massive steel plate (Fig. 31).
The measured velocities of the bare and the 3 different cased charges at the 4 distances
are shown in Table1:
22
Czech Republic
2006
[Content]
Table 1. Measured initial sled velocities by the blast and fragment loads
Casing
Velocities in m/s at the four investigated distances
thickness
1,0 m
1,5 m
2,0 m
3,0 m
0 mm
5,5
2,65
1,2
0,36
2,5 mm
6,4
3,6
2,2
0,93
5,0 mm
7,1
4,6
3,2
1,8
10,0 mm
10,4
6,8
5,0
2,5
These values have to be multiplied with the sled weights of 51,5 kg to get the achieved
momenta in kgm/s or Ns of the combined blast and fragment loads (Table 2).
Table 2. Measured sled momenta by the combined blast and fragment loads
Casing
Momenta in kg m/s or N s at the four distances
thickness
1,0 m
1,5 m
2,0 m
3,0 m
0 mm
283
137
62
19
2,5 mm
330
185
113
48
5,0 mm
366
237
145
93
10,0 mm
536
350
258
129
Individual blast and fragment loads

The fragment impulses can be calculated by using the fragment velocities, which are
predicted by the Gurney equation with the Gurney constant 2 E of 2165 m/s for the used
C4 charges (5), and the fragment masses in the azimuth hit sector which is covered by the
600 mm wide plates. It is assumed that at 1 m distance the 500 mm vertical height of the
target plate captures 100% of the vertical spread of the fragments. Then linear reduction of
the fragment hits is assumed with increasing distance 0% at 1 m, 2,5% at 1,5 m, 5% at 2 m
and 10% at 3 m distance. The initial fragment velocities are taken as constant along the four
short distances between 1 m and 3 m.
Table 3 gives the maximum mass of the fragments that can impact the plates assuming
100% hit the 600 mm wide target plates. This is calculated with the following equation:
mfrag.sector = (De - Di) /4L Steel(arc tan 0,3/d)2/360
mfrag.sector = (De - 6) /4297,85(arc tan 0,3/d)2/360
8
Manfred Held
23
Czech Republic
2006
[Content]
where De and Di are the external and internal diameters of the steel casings, L their
length of 29 cm, Steel the steel density and d the distance of the plates to the charge.
Table 3. Fragment masses in grams impacting the 600 mm wide target plates
Casing
De
(mm)
(cm)
= c/m
1,5
2,5
6,5
0,93
104
70
53
35
7,0
1,94
2156
146
110
74
10
8,0
4,17
464
316
237
159
Distance d (m)
Table 4. Calculated fragment momenta in the given azimuth angles on the 600 mm
wide target plates with 100% fragment hits at 1 m, 97,5% at 1,5 m, 95% at
2 m and 90% at 3 m distance.
Casing
= m/c
(mm)
VFragment
Momenta (kgm/s or Ns)
Distance
1,0 m
1,5 m
2,0 m
3,0 m
Hit %
100%
97,5%
95%
90%
2,5
0,93
1805 m/s
188
123
91
57
1,94
1386 m/s
299
196
145
92
10
4,17
1002 m/s
465
309
226
143
Approximate values of the blast loads alone can be calculated by subtracting the
calculated fragment loads in Table 4 from the measured momenta in Table 2. These values
are summarised in Table 5.
Table 5. Estimated contribution of blast load alone in kgm/s or Ns

Casing
24
Differences between the measured and calculated fragment momenta
Thickness
1,0 m
1,5 m
2,0 m
3,0 m
0 mm
283
137
62
19
2,5 mm
330188 = 142
185123 = 62
117-91 = 26
48-57 = -9
5,0 mm
366-299 = 67
237-196 = 41
145-145 = 0
95-92 = +3
10 mm
536-465 = 71
350-309 = 41
258-226 = 32
129-143 = -14
Czech Republic
2006
[Content]
These are the blast contributions of the momenta in kgm/s or Ns. These values can be
also expressed in percentages to the blast loads of the corresponding (table 6) bare charges.
Table 6. Fraction of the blast loads of the confined charges to the bare charge
Casing
Ratio of the blast loads of confined to bare charges
Thickness
1,0 m
1,5 m
2,0 m
3,0 m
2,5 mm
142/283 = 0,50
62/137 = 0,45
26/62 = 0,42
(-9/19 = -0,47)
5 mm
67/283 = 0,24
41/137 = 0,30
0/62 = 0
(3/19 = +0,16)
10 mm
71/283 = 0,25
41/137 = 0,30
32/62 = 0,52
(-14/19 = -0,74)
From Table 6 the conclusion can be drawn that a greater portion of momentum is
transferred by the fragment impacts of encased charges. Furthermore these data demonstrate
that, if a combined blast and fragment load impacts a target, the blast part in radial direction
is more reduced in the case of more heavily confined charges.
Radial blast reduction by confinement

A number of equations exists for predicting the reduction of the blast effect by radial
confinements of the detonating high explosive charge. One of these is the well known FANO
equation which is described for the positive impulse as an extension of the Gurney formula.
W/W =
0,2 + 0,8/(1+2 m/c)
From the former Naval Ordnance Labortaory NOL the following equation was given for
the positive impulse of the positive phase for confined charges:
W/W =
(1+m/c(1- m ))/(1+m/c)
where m is equal to m/c, if m/c<1 and 1, if m/c>1.

The reflected specific impulse density can be expressed with the following equation:
IDS,refl/W1/3 = CW0,84/R0,84
where C is a constant, W the charge weight and R the distance:
IDS,refl = CW0,84W0,333/R0,84 = CW1,173R-0,84
A reduction factor F, generally called casing factor, has to be used for the charge weight
W if the charges are confined. Therefore the above equation can be written in the following
form:
IDS,refl = C(FW)1,173/R1,25
10
Manfred Held
25
Czech Republic
2006
[Content]
This means the impulse values have to be corrected by the casing factors F1,173 for the
different confinements. In the following Table 7 the calculated casing factors, using F alone
and their F1,173 values, are summarised for IDS,refl with the confinements for the above
equations.
Table 7. Reduction factors of the radial momenta without and with their exponential
values
Equation/Thickness
2,5 mm / = 0,93
5 mm / = 1,94
10 mm / = 4,17
FANO
0,480 / 0,422
0,364 / 0,302
0,286 / 0,320
NOL Impulse
0,552 / 0,498
0,344 / 0,286
0,193 / 0,146
These values are independent of the distance between the charge and the target. Reasonable
experimental values were found only in the distances of 1,0 m and 1,5 m. Therefore these
values are only used for the prediction of the blast load alone in radial direction of
cylindrical charges with the FANO and NOL impulse load reductions with different
confinements in Fig. 32.
The experimentally achieved blast reduction factors by confinements are only rough
indications. But they give the same trend. Shortcomings in these considerations are the
assumptions of the accuracy of the calculated fragment velocities and also the percentages of
fragment hit in the elevation directions spalling fragments on the end faces of the charge
may have angles which are not captured by the vertical heights of the target plates.
Combined fragment and blast load

In an EMI report [6] the specific impulse densities are summarised for blast and
fragment loads of cylindrical confined charges with different m/c ratios of 1, 2 and 4 against
perpendicular targets in the dimensions kgm/skg-1/3m-2 (Fig. 6). These lines solid can
be expressed with the following exponential equations:
bare charge:
IDS,bare[6] = 370(R/W1/3)-1
confined charge:
IDS,conf. [6] = 1500(R/W1/3)-1 0,6
In this diagram the results achieved in the momentum sled tests are added as dashed
lines. the values, also expressed in the dimensions kgm/skg-1/3m-2 are summarised in Table
8. They give similar trends in the log log diagram although the values are generally lower
than the published data in [6].
26
11
Czech Republic
2006
[Content]
Table 8. Specific impulse densities of the combined blast and fragment loads in
kgm/skg1/3m-2 of the described test series
Casing
= m/c
Distances
1,0 m
1,5 m
2,0 m
3,0 m
0 mm
888
430
194
60
2,5 mm
0,93
1035
580
354
151
5 mm
1,94
1148
783
455
292
10 mm
4,17
1681
1098
809
405
These data can be described with the following equations, which have generally the
same trends.
370(R/W1/3)-1,25
bare charge:
IDS,bare[6] =
confined charge:
IDS,conf. [6] = 1010(R/W1/3)-1,25 0,33
The impulse density needs a selection of target area. This can lead to some problems. If
a larger area with less dense fragment hits is selected, then lower values will be calculated
for the specific impulse density.
With the sled diagnostic tests the combined effects of blast and fragment loads were
measured for m/c ratios of about 1, 2 and 4 and compared to bare cylindrical charges in
distances form 1 m to 3 m or in scaled distances Z of about 1 to 3. The momenta were
measured with a new diagnostic technique, namely with the involved velocity of momentum
sleds, on which the target plates were arranged perpendicularly. The measured results
generally fit the data of the reflected impulse of bare charges referenced in the literature and
also approximately the casing factors of confined charges, using the FANO equation.
8.
BLAST CONTOURS AT MEDIUM DISTANCES
To get the blast contour at scaled distances Z of 3 and 6, tests are performed for the 1 kg
charges at 3 m and 6 m radius. From the impulse density over the scaled distance diagram
was predicted impulse density values around 0,02 ID and 0,008 ID (Fig. 34).
To get displacements of a few meters wooden momentum gauges were selected (Fig.
35). In the first test it was fired a 1 kg sphere, made from Seismoplast. The momentum
gauges, arranged at 3 m radius, are laying in less than 0,5 m distance, instead of more than 3
m (Fig. 36). In 6 m radius they were still partially lying on the holding fixtures. The
achieved results of impulse densities in 0,5, 3 and 6 m compared to the expected values are
shown in Fig. 37 for 1 kg charge.
12
Manfred Held
27
Czech Republic
2006
[Content]
The momentum gauges are loaded by the positive and also the negative phase of the
pressure history (see Fig. 1). People are normally not taken into account the negative
pressure phase. The author has measured the pressure and Impulse history at 3 m and 4 m
distance with pressure transducer and has found out that the total impulse over the integrated
time of 30 ms has given only a very small value (Fig. 38 and Fig. 39). It is rarely mentioned
in the literature ([..] and [..]) that the resulting impulse at larger distances is remarkably
reduced by the succeeding wave (Fig. 40).
Beside the 1 kg charges also 2,83 kg and 8 kg heavy spheres of Seismoplast were fired.
The analyzed impulse density values over the polar angle are shown in Fig. 41. The impulse
density values over the scaled distances - which means the distances have to be divided by
the cubic root of the charge weights and the impulse density by the 2/3 root of the charge
weight are presented in Fig. 42.
The impulse density contours for a 1 kg Seismoplast charge of a L/D ratio 2/1 from
0,5 m until 4 m distance are presented in Fig. 43. It seems that the axial load in detonation
direction is constant form 0,5 to 1,5 m distance and is remarkably reduced from 3 m on. The
blast loads in the bridge wave direction around 45 have a similar behaviour and looks
constant between 0,5 m and 1,5 m distances. From 4 m distance on the blast contour is more
or less levelled out.
All the tests are done in open air with sometimes small wind conditions where such
influences cannot be fully neglected. For better controlled conditions, such tests should be
done in large enough buildings where all such details can be better controlled.
9.
GHOST EVENT
Video pictures of the detonation event of cast TNT-charge with 1 kg weight and L/D
ratio of 1 has given a surprising phenomena. At first the author has thought, that this is an
optical reflection in the camera from the bright detonation light (Fig. 44). After some
discussion of this surprising effect with experts of the proofing ground Meppen [..] he has
got the picture of a 100 kg cast TNT charge with also a L/D ratio of 1 (Fig. 45).
Very narrow blast effects with very high impulse densities happens generally for
cylindrical charges in the detonation directions up to scaled distances of 3. But the ghost
effect is in around 3 m altitude above the fireball of the cylindrical charge.
This surprising event happens in distances of 20 to 30 times charge diameter above the
charge.
Reasonable explanation was not found up to now and is an open point for a research and
some investigations
28
13
10.
Czech Republic
2006
[Content]
DIAGNOSTIC OF THE MOMENTUM DISTRIBUTION OF ANTITANK MINES
In the very near field of detonating high explosive charges scaled distances beneath
one Z < 1 - , again the damage is only caused by the momentum transferred to the target.
Therefore it is sufficient to measure directly the momentum and not the pressure time history
which is at least very difficult, if not really impossible in a mostly very nasty environment.
The pressure duration of spherical charges of a few kilograms weight in a distance of around
0,5 m is in the range of 0,1 ms [6]. Therefore, the blast wave only induces the movement to
the target which reacts with frequencies of fractures of seconds, similar to a ballistic
pendulum.
The author could use flash X-ray units which were installed in a Leopard 1 chassis at the
Proving Ground Meppen, Germany [10]. To level up the observation height to the flash Xray tube altitude concrete blocks were arranged to a quadrat, which was then filled up with
sand. The high explosive charge was layed up in the middle (Fig. 46). 500 mm above the
levelled sand the momentum gauges were arranged. They were lying on two steel bands of 5
mm width and 50 mm height. The charge axis was oriented exactly on the symmetry planes.
To provide momentum gauges, plates with 25 mm width and 100 mm length and the
following heights of thicknesses were arranged: 150, 100, 70, 50, 25, 20, 10, 5 mm, flush
with the symmetry plane in blocks of always 4 with equal thickness (Fig. 47). The distance
between the flash X-ray tubes and the charge was about 4,0 m and between the charge and
the film 1,5 m (Fig. 48). This provides a magnification factor of 1.6. With two flash X-ray
units first a still picture of the arrangement was made on a separate film and then two
exposures of the dynamic event, typically 2 ms and 10 ms after the detonation of the charge
(Fig. 49).The delay generators were triggered by an ionization pin, a twisted isolated copper
wire in contact to the charge.
The flash X-ray pictures show the displacements of the momentum gauges at the two
time differences. The velocities are obtained by dividing the measured distances by the
magnification factor and then divided by the time difference.
A horizontal plate arrangement perpendicularly to the blast wave would not undergo
horizontal stretching forces by the blast load and therefore this vector component is not taken
into account.
The achieved impulse densities ID on the line of the momentum gauges are presented in
Fig. 50 as a function of the horizontal distance for the high explosive disks lying on the
ground, levelled to the ground and buried 100 mm deep.
These results are a cut to the circular load on a plate. These values have to be multiplied
by ring areas for the individual 25 mm wide momentum gauges (Fig. 51). The values of the
100 mm deep buried charge are a little jumping, caused by the sand fountains. The load on
circles as a function of their radii, are given by the sum of the individual ring zones (Fig. 52).
Over the analyzed distance of 700 mm, the charge levelled to the ground gives an about 50%
14
Manfred Held
29
Czech Republic
2006
[Content]
and the charge 100 mm deep buried a 90% higher value than the charge laying on the
ground.
This test method gives reproducible results of the blast load by the diagnostic of the
achieved/measured momenta. With the high space resolution in this differential method
they are ideal for the calibration and verification of blast models with the charges arranged
on top of sand, levelled to the sand and buried 100 mm deep.
This method allows to measure the real loads of existing mines. Any change of the high
explosive charge type and the shape render different results, which was expected.
11.
DIAGNOSTIC OF THE PLATE ACCELERATION BY ANTI-TANK

MINES
The bulge of different thick steel plates under anti-tank mine blast loads should be
measured with high time and space resolution. For this purpose a special device was
designed to get the displacement, the velocity and the acceleration. The diagnostic tool had
to be fitted in the dome device of a special German mine-protection test rig (MinenschutzSprengtisch) (Fig. 53). A height of 380 mm was available. This diagnostic tool could not be
fixed in the upper part, because it had to be brought in by lifting up the test plate above the
ground.
The drawing of the displacement device is shown in Fig. 54. A 20 kg heavy cylinder
acting as a drag mass and was held via 3 shear pins of 2 mm diameter on the tripod. The
static cutting force of one shear pin is about 3 KN. In the dynamic case a force of 20 KN can
be expected for all three pins. This leads to an acceleration of the 20 Kg heavy mass to 1.000
m/s (a = F/m = 20.000 Kg m/s2/20 Kg = 1.000 m/s).
This force causes a movement of the drag mass after the equation s = 0,5 g t to 510-3
mm after 100 s time difference, what is a negligible value.
The drag mass had a hole of 20,2 mm diameter in the axis, in which a tube of 20 mm
outside and 16 mm inside diameter with a sharp knife edge on top was guided. This moving
up tube was cutting and shortening the wires of a flat band cable. This is tightened from the
side in the drag mass. The signals were recorded with transient recorders. A tube was
selected so that the weight is small and the cut flat band pieces can fall down inside and are
out of the way. Fig. 55 shows the displacement device assembled with the flat band cables,
standing on the testing plate, just before the plate is brought up into the test rig (Fig. 56).
An 8 kg heavy cylindrical cast TNT-charge, with 267 mm diameter and 89 mm length,
given in a so-called steel pit arrangement, was exactly positioned to the centre of the plate
and displacement device. It had 500 mm distance to the 25 mm thick high graded steel plate
(Fig. 57). At this test a 64 fold flat band cable was used, where only the first 16 contacts
where measured with transient recorders at 0,05 s time resolution. The copper bundles have
1,25 mm distances from one another. The shortening times as function of distance are shown
30
15
Czech Republic
2006
[Content]
in the diagram of Fig. 58. The spacing of 1,25 mm divided by the time differences between
two contacts gives their velocities, which are presented as a function of displacement
distance in Fig. 59. These velocities as a function of the shortening times are presented in
Fig. 60. But the velocities are not values at the individual end times. They represent always
the mean velocities between two contacts. Therefore they are drawn up over the mean time
of two contacts respectively (Fig. 60 left). The velocities rise linear with the time, which
means constant acceleration. Only the tangent to the first point cannot be very well
predicted, because it has, with 40 s in the starting phase, a long interval time. Nevertheless
the acceleration is constant with 920 000 m/s between 20 s and 140 s or 1,25 mm and
125 mm. From this time and distance on there cannot be made any further predictions with
the gained data.
The knife tube moves on with its maximum velocity because it is not mechanically
connected to the plate and has nearly no friction forces by cutting the wires and is therefore
not decelerated. The maximum velocity has achieved around 140 m/s after 12,5 mm bulge
and 140 s time delay.
This is an autonomous displacement device or acceleration gauge which allows to
measure accurately the increasing velocity and the magnitude of accelerations of the bulge
on a steel plate caused by a mine blast. Such devices can be also fitted in real tanks or
vehicles. It is independent, how the total system reacts on the mine blast, because the heavy
drag mass is virtually stable during the measuring time.
SUMMARY
The author tried to summarize some new, partially unusual diagnostic methods to get
answers to the damage levels by blast waves. Partially they are low costs in investment, but
partially a little labour intensive. The methods with momentum gauges around the energy
source has given new results on the momentum distribution of not spherical charges in the
near, respectively in the damage distances. Surprisingly he has found out, that the negative
phase, which has a great influence on the transferred momenta at medium distances is mostly
neglected.
The ghost events in the axial direction of a detonating cylindrical high explosive charges
is also a new task.
The diagnostic of anti-tank mine blast distribution, using flash X-ray equipments, allows
to calibrate the numerical codes.
The diagnostic of the acceleration of a bulging plate, caused by a detonating blast mine,
was also a special challenge.
16
Manfred Held
31
32
Czech Republic
2006
[Content]
17
18
Manfred Held
Czech Republic
2006
[Content]
33
34
Czech Republic
2006
[Content]
19
20
Manfred Held
Czech Republic
2006
[Content]
35
36
Czech Republic
2006
[Content]
21
22
Manfred Held
Czech Republic
2006
[Content]
37
38
Czech Republic
2006
[Content]
23
24
Manfred Held
Czech Republic
2006
[Content]
39
40
Czech Republic
2006
[Content]
25
26
Manfred Held
Czech Republic
2006
[Content]
41
Czech Republic
2006
[Content]
THE THERMAL BEHAVIOUR OF BTAW, A HIGH NITROGEN FUEL

D.E.G. Jones*, K. Armstrong, T. Parekunnel, Q.S.M. Kwok
Canadian Explosives Research Laboratory, Natural Resources Canada
555 Booth Street, Ottawa, ON, Canada K1A 0G1
djones@nrcan.gc.ca
Abstract:
BTAw (bis-(1(2)H-tetrazole-5yl)-amine monohydrate) has recently been considered for use
as a low-smoke pyrotechnic fuel. There is relatively little information available in the
literature concerning the thermal properties of BTAw or its precursors. In the present
work, various thermoanalytical experiments were performed on BTAw and BTA (bis(1(2)H-tetrazole-5yl)-amine) in an effort to better characterize the thermal stability and
decomposition of these compounds.
Variable heating rate studies were carried out on BTAw samples in a helium atmosphere
using DSC and TG. Two steps were seen in the results: dehydration followed by
decomposition. Kinetic parameters were determined for both of these steps using a number
of methods. Experiments using simultaneous TG-DTA coupled with FTIR and MS were
performed on BTAw in both helium and dry and CO2 free air atmospheres, and evolved gas
analysis was used to determine the gaseous decomposition products. The thermal stability
of BTAw and BTA was examined using accelerating rate calorimetry (ARC).
INTRODUCTION
Typically, pyrotechnic formulations consist of carbonaceous fuels with metal salts added to
oxidize the fuel and colour the flame. These combinations tend to produce an abundance of
noxious smoke and ash, the result of partial combustion, and the generation of non-gaseous
products, particularly metal oxides. Chlorinated polymers were introduced for colour
enhancement but they still produce large amounts of smoke. High-nitrogen compounds were
also introduced with more success. Chavez and Hiskey have reported work on 3, 6-dihydrazinos-tetrazine (DHT) in which it was mixed with non-metallic oxidizers and 5 mass % coloring
agents, resulting in the production of little smoke and ash [1].
Other high-nitrogen compounds studied for use as low smoke pyrotechnic fuels are BTAw,
shown in Figure 1, and 5,5-bis-1H-tetrazole (BT). These compounds are diprotic acids that
react with basic amines to form mono- and di-aminated salts, and react with metal carbonates or
hydroxides to form metal salts. Chavez et al have recently reported that, by preparing BTAw
and BT with different ratios and compositions, it is possible to achieve greater control and
variability in burn rate and ignitability of high-nitrogen pyrotechnic formulations than with DHT
alone. Various characteristics of these two molecules, their hydrates and salts, were examined
by Chavez, including impact sensitivity, thermal behaviour by thermogravimetry (TG) as well
as the dehydration of BTAw to BTA [2]. It was noted that BTA exhibited greater impact
sensitivity than the hydrated form, BTAw. Nedelko et al. studied the kinetics and
42
Czech Republic
2006
[Content]
decomposition products of BTA using thermogravimetry, volumetry, calorimetry, IRspectroscopy, and mass spectroscopy [3].
Earlier results from our laboratory showed that BTAw was relatively insensitive to impact
and friction, thermally stable at 75 C and not reactive when exposed to flames.
The present paper includes a detailed study of the current work on BTAw. Two variable
heating rate studies on BTAw, one using DSC and the other using TG, a study of the thermal
decomposition using simultaneous TG-DTA coupled with FTIR and MS, and an investigation of
the thermal stability of BTAw and BTA in air and Ar atmospheres using accelerating rate
calorimetry (ARC) are described in this paper.
EXPERIMENTAL
Samples of BTAw containing 20 % water were used as starting materials. These asreceived samples were used in the DSC heating rate study. Two methods were used to pre-dry
the BTAw. One method involved drying the BTAw in situ, i.e. in the ARC studies, for 3 hours
at 120 C in an open system. The other method consisted of drying BTAw in an oven at
temperatures of either 50 C, in order to remove excess water, or at 120 C to remove water of
hydration, for several hours. With pre-drying at 120 C, the end result was BTA. The pre-dried
sample at 50 C was used in the TG and TG-DTA-FTIR-MS studies. Samples obtained from
both methods were used in the ARC studies.
A TA Instrument (TAI) 2910 DSC was used for a variable heating rate study of BTAw
following ASTM standard test method E 698-01 [4]. Hermetic pin-hole Al pans (75 m hole)
containing about 0.25 mg of BTAw were heated from 30 C to 350 C at the rates of 0.5, 1, 2, 5,
8 and 10 C min-1. The DSC was calibrated for heat flow [5] and temperature [6]. An empty
pan was used as the reference.
The TG studies on BTAw were conducted using two different modules: the TAI 2950 and
the TAI Q500. The consistency of results between both modules was verified by performing
repeat measurements at the same heating rate. Sample sizes of 1 mg placed in aluminum pans
were run in helium at a flow rate of 100 mL min-1, split 60:40 between the furnace and the
balance. The samples were heated from 30 C to 350 C at rates of 1, 2, 5 and 10 C min-1. The
TG instruments were calibrated for mass using the procedures recommended by the
manufacturer and for temperature using the Curie point method as described in ASTM method
E1582 [7].
Details of the instruments used to obtain simultaneous TG-DTA-FTIR-MS data as well as
calibration information can be found in an earlier publication [8]. The TG-DTA-FTIR-MS data
were acquired simultaneously to study the thermal behaviour of BTAw and to identify the gases
that evolved while the BTAw sample was heated. Samples and reference (Pt foil) of 5 mg were
heated in alumina pans at a rate of 5 C min-1. Samples were run in helium and in dry and CO2free air atmospheres, purged at rates of 100 and 50 mL min-1, respectively.
The ARC measurements were performed on two instruments: one adiabatic calorimeter was
distributed by TIAX LLC (formerly known as Arthur D. Little Inc.) while the other was a
Thermal Hazard Technology (THT) instrument. Samples of about 0.5 g were placed in
lightweight spherical titanium vessels. The ARC experiments were started at an ambient
David Jones, K. Armstrong, T. Parekunnel, Q. Kwok
43
Czech Republic
2006
[Content]
pressure of either air or argon. The standard ARC procedure of heat-wait-search (HWS) was
used [9,10] with a 5 C heat step.
RESULTS AND DISCUSSION

Dehydration
DSC Results
Typical thermal curves of BTAw obtained at various heating rates are shown in Figure 2.
The inset shows the entire range of the thermal curve obtained at 5 C min-1, including an
endotherm in the 125 to 150 C range, which is attributed to the dehydration of BTAw. The
average enthalpy of dehydration is 268 15 J g-1 or 43 2 kJ per mol of BTAw. Hydrogen
bonding energies are generally found to be between 8 and 42 kJ mol-1. This result suggests that
there is strong hydrogen bonding present between the water and BTA molecules.
Kinetic parameters for the dehydration step were determined from a series of variable
heating rate experiments following the ASTM method E698 [4]. The results were determined
from Figure 3, which is a plot of ln(T-2/K-1min-1) vs. 103 K/T for the peak temperatures (T) of
the dehydration endotherm at different heating rates (). The activation energy, E, was found to
be 197 5 kJ mol-1, indicating that the water molecule is strongly bound to BTA. Using this
value of E in conjunction with the calculated pre-exponential factor, ln(Z/min-1) = 46 1, the
rate constant was determined to be ln[k(144 C)/min-1] = -11 2.
TG Results
A two-step mass loss was observed in all experiments, the first step represents dehydration
and the second step is a result of decomposition. Figure 4 shows a typical TG curve obtained
from the variable heating rate study. An initial 9 10 % mass loss was observed in the 130
160 C range. This corresponds to the loss of 1 mol of H2O from 1 mol of BTAw, for which the
calculated mol % H2O is 10.5.
In a study [2] by Chavez et al, it was reported that hydrated water in a BTAw sample was
lost from 74 C to 116 C. In this particular study the heating rate used during the TG
experiment was 0.1 C min-1, which is substantially lower than the heating rates used in the
present work. Thus, the loss of hydrated water at lower temperatures was expected because of
the much slower heating rate.
Kinetic parameters, E/kJ mol-1 and ln(Z/min-1), were obtained for the dehydration step using
ASTM method E 1641 [11], which is based on the assumption that the dehydration obeys firstorder kinetics. As recommended by the ASTM method, calculations were carried out at several
different levels of conversion: 5, 10, 15 and 20 %. The results can be seen in Figure 5, which is
a plot of ln vs. 103 K/T, where T is the temperature of these different conversion levels, ,
obtained at various heating rates (). Table 1 shows that the values of E do not vary between
the range of conversion level examined and an average value of 193 5 kJ mol-1 was
determined. This result is in good agreement with the DSC results in this study, as seen from
44
Czech Republic
2006
[Content]
Figure 3. An average value of the pre-exponential factor was also calculated, giving ln(Z/min-1)
= 57 2.
With the peak temperatures from the DTG data, ASTM method E 698 [4] was employed as
another technique to determine kinetic parameters. The results are shown in Figure 3, which is a
plot of ln(T-2/K-1min-1) vs. 103 K/T, where T is the peak temperature from the DTG data.
Calculations of the kinetic parameters yielded E = 182 4 kJ mol-1 and ln(Z/min-1) = 41 1.
Using the calculated kinetic parameters, E and lnZ, previously mentioned, the values for lnk
were calculated for the dehydration step at a common temperature of 144 C in order to
compare the results obtained by ASTM E698 and ASTM E1641. The value for the DTG data
was ln(k/ min-1) = 11 1, which is in agreement with the value calculated from the DSC data.
From the TG data, an average was taken of the rate constants and found to be ln(k/min-1) = 1
2, which is significantly higher than those found using the DSC and DTG data. This was
expected given the difference in pre-exponential factors. This difference could be due to the
fact that the values calculated using the TG data were obtained in the early stages of the
dehydration in comparison with the values from the DSC and DTG data, which were obtained at
much later stages.
The software IsoKin [12] was then used to perform a model free kinetic (MFK) analysis on
the first mass loss from the TG results. The model-free approach makes use of an
isoconversional method that allows for evaluating a dependence of the effective activation
energy on the extent of conversion [13]. These results can be found in Figure 6, which shows
the dependence of the activation energy on the extent of reaction. The MFK results show that
the activation energy remains consistent throughout the extent of the dehydration step,
indicating a single-stage process, with a mean value of 178 3 kJ mol-1 for 0.1 < < 0.8. From
the MFK analysis, the mean value of ln[f().(Z/min-1)] is 49.7 0.8.
Decomposition
DSC Results
The results of the heating rate study, using ASTM method E698 [4], on the decomposition
of BTAw are shown in Figure 7, which is a plot of ln(T-2/K-1min-1) vs. 103 K/T. This plot was
used to determine the values of the kinetic parameters, E and ln(Z/min-1), for BTAw. Only
results from the 0.5, 1, 2 and 5 C min-1 heating rates were included as the heat generated for the
8 and 10 C min-1 runs exceeded the maximum heat generation limit of 8 mW recommended in
the ASTM method E968 [5]. The kinetic parameters determined are E = 222 16 kJ mol-1 and
ln(Z/min-1) = 38 4. A ln(k/min-1) value of 12 6 was calculated at 265 C.
To confirm the kinetic results, a sample of BTAw was aged at 234 C for 64 minutes, the
calculated half-life of BTAw at this temperature from analysis of the data, and then immediately
quenched to a temperature at least 50 C lower. This aged sample was then heated at 5 C min-1
and its thermal curve recorded. The peak area for this curve was 737 5 J g-1. When compared
to 1.48 0.01 kJ g-1, the peak area for the unaged sample of BTAw run at the same heating rate,
it is seen that the peak area of the aged sample shows good agreement with the predicted value.
45
Czech Republic
2006
[Content]
Thus, the values determined for the kinetic parameters are good estimates for BTAw. The DSC
thermal curves for the aged and unaged samples of BTAw are shown in Figure 8.
Using the data collected from the DSC, a MFK analysis was performed with the IsoKin
software package [12]. The results for E and ln[f().(Z/min-1)] are shown in Figure 9. E has
increasing values as the extent of conversion, , increases to 0.6. This observation suggests that
parallel reactions are occurring from 0.1 < < 0.6.
TG Results
The second observed mass loss corresponds to thermal decomposition. This step occurred
with a mass loss of approximately 60 % in the 250 290 C range, as shown in Figure 4. When
using ASTM method E1641 [11] to evaluate kinetic parameters for the decomposition step, the
calculated values for E and ln Z were observed to increase as the extent of conversion increased
(Table 1 and Figure 10).
Decomposition of BTAw was reported [2] to commence at approximately 175 C using a
heating rate of 0.1 C min-1. With a heating rate significantly lower than those used in this
study, it was expected that the onset temperature in the literature work would also be lower due
to the dependence of the onset temperature on the heating rate.
Nedelko et al reported [3] kinetic parameters corresponding to both first order
decomposition and a first order catalytic reaction. For the first order decomposition of BTAw,
kinetic parameters of E = 197 5 kJ mol-1 and ln(Z/min-1) = 40 5 were given. A value for lnk
= -6 5 was calculated from this data at 242 C. In order to compare this value with the rate
constants from this study, values for lnk were also calculated at 242 C. From the TG data at 5
% conversion, it was found that ln(k/min-1) = 1 2, and at 10 % conversion, ln(k/min-1) = 0 3.
These values agree with the results from Nedelko within the margin of error. From the DSC
data, a value of ln(k/min-1) = -14 5 was found. This value was also within the margin of error
in comparison with Nedelkos results; however, using data from the DTG results to calculate the
rate constant gave ln(k/min-1) = -14 2, which did not fall within the range of error.
Using the DTG peak temperatures and ASTM method E 698 [4], kinetic parameters of E =
236 5 kJ mol-1 and ln(Z/min-1) = 41 1 were calculated. Both of these values are consistent
with those obtained from the DSC heating rate study and provide a good estimate of the kinetic
parameters for BTAw. Figure 7 is a plot of ln(T-2/K-1min-1) vs. 103 K/T for both DTG and
DSC data. Values for ln(k/min-1) were calculated using a common temperature of 265 C. The
DTG lnk value was found to be 12 1 min-1 while the TG value at 10 % conversion was 2 3
min-1. This difference was not unexpected as the results from ASTM method E 1641 (Figure
10) showed that the kinetic parameters vary with extent of conversion. The DTG value for lnk
was consistent, however, with that obtained from the DSC results.
The MFK results for this second step, seen in Figure 11, revealed that E slowly increases for
< 0.6. In this range, MFK analysis also showed that the values of ln[f().(Z/min-1)] slowly
increased from about 30 to 50. The variation of E with is consistent with the results obtained
using ASTM method E 1641 (Figure 10). As in the DSC decomposition results, this
dependence of E on and its increasing nature suggest that parallel reactions are occurring in
the range 0.1 < < 0.6.
46
Czech Republic
2006
[Content]
TG-DTA-FTIR-MS Results
The thermal decomposition of BTAw was studied using TG-DTA-FTIR-MS. A TG-DTAFTIR-MS plot for the thermal decomposition in helium is shown in Figure 12. A 10 15 %
mass loss is observed in the 125 to 150 C range. This was attributed to dehydration, verified
by the evolved gas analysis: the absorbance peak in the FTIR spectrum at 3854 cm-1 and the ion
current peak for m/e = 18 in the MS. The mass of the water lost indicates that there is
approximately 1 mol of water per mol of BTAw.
Hydrogen azide (2153 and 1165 cm-1), ammonia (967 cm-1) and hydrogen cyanide (714 cm) appear as the major products in the FTIR trace in helium beginning at approximately 240 C.
This is supported by the MS data with peaks appearing for m/e = 17, 29, and 43. A strong peak
in the MS for m/e = 28 could be due to N2 formed either by electron impact fragmentation of
HN3, or by the thermal decomposition of BTAw. The total mass loss for this decomposition
slowly approaches 80 % by 350 C. These results for mass loss are in good agreement with
those obtained by TG.
1
A TG-DTA-FTIR-MS plot for the thermal decomposition of BTAw in air is shown in

Figure 13. The same dehydration is seen in the 125 to 150 C range. However, at
approximately 262 C, a sharp mass loss totalling almost 100 % was observed. This was
accompanied by the evolution of mainly hydrogen cyanide and ammonia, as well as small
amounts of carbon monoxide, carbon dioxide and various other gases for which the
identification is inconclusive. FTIR absorbance in the 2280 to 2285 cm-1 region suggests the
evolution of a nitrile compound in both He and air. The presence of m/e = 42 suggests this
nitrile may be cyanamide but further work is needed for a definitive identification. Nitrogen gas
is thought to be evolved based on previous studies of the thermal decomposition of 5aminotetrazole done by Levchik et al [14]. However, there is no supporting evidence at this
time due to the presence of nitrogen in the purge gas.
Tetrazole and 5-aminotetrazole both exhibit tautomerism, which is known to affect the
mechanism of their thermal decomposition. When the thermal decomposition of tetrazole occurs
in a melt, nitrogen is evolved as a result of breakdown of the azido-form of the molecule and
approximately 5 % of the sample decomposes with the elimination of hydrogen azide from the
ring form of the molecule [15,16]. With 5-aminotetrazole, fragmentation of the imino form, the
main isomer of the molecule in the solid state, most likely gives HN3 [14, 17]. However, upon
heating the imino form converts to the amino form and decomposition results in the elimination
of nitrogen gas and ammonia. Using these two molecules for comparison, BTAw may undergo
a tautomeric isomerization in which one form of the molecule has one or both tetrazole rings in
an imino form and results in the elimination of hydrogen azide. If this is correct, the lack of
hydrogen azide production in a reactive atmosphere would suggest that the imino form is not
present in air. The decomposition of the other form of the molecule results in the formation of
nitrogen, hydrogen cyanide and ammonia.
ARC Results
The thermal behaviour of BTAw and BTA using ARC was studied in atmospheres of air
and argon at ambient pressure. Figure 14 is an example of the results for a typical ARC
47
Czech Republic
2006
[Content]
experiment, showing temperature and pressure vs. time. A comparison of ln(R/C min-1 g-1)
versus reciprocal temperature for the various methods used is shown in Figure 15. Table 2
presents a summary of the ARC results using the various drying methods.
BTAw samples were dried in-situ at 120 C followed by a HWS performed in an ambient
air atmosphere using an initial temperature of 150 C. In this set of experiments, each reaction
concluded with an exothermic reaction with a self-heating rate greater than the set maximum
rate of 1 C min-1. The experiments displayed consistent results, having onset temperatures of
213 and 215 5 C and each ending in a runaway reaction.
In addition, samples of BTAw were dried externally at 50 C. HWS experiments were
performed on these samples in both ambient argon and air. The results in argon exhibited
similar behaviour with onset temperatures of 205 and 203 5 C. These runs also showed some
unusual behaviour in which the heating rate attained the pre-determined self-heating rate limit, 1
C min-1, which triggered cooling of the ARC, but after cooling had begun, the heating rates
were in excess of 10 C min-1 for a short time before cooling resumed. Similar experiments in
ambient air did not exhibit this unusual behaviour and the onset temperatures for these
experiments were 211 and 201 5 C. In spite of the potential oxidizing potential in air, there is
no apparent difference between the onset temperatures for the ARC experiments conducted in
inert gas or air.
Finally, HWS runs in ambient argon were carried out on samples of BTAw that were predried at 120 C. This method of pre-drying removed all water content and resulted in BTA
(verified by conducting a TG study on the sample). Onset temperatures of 198 and 203 5 C
were observed, which are similar to, but slightly lower than, the previous experiments conducted
in argon on a sample pre-dried at 50 oC, suggesting that BTA is less thermally stable than
BTAw. In addition, when comparing this sample to the sample dried in-situ at 120 C, a
significant decrease was seen in the onset temperature. This indicated that an increase in drying
time approximately 16 hours for the external drying compared to 3 hours for in-situ drying
results in a decrease in the thermal stability and thus, the onset temperature.
From Figure 15, it is apparent that all the experiments demonstrate similar variation of R
with reciprocal temperature, with the exception of the sample dried in-situ. For this sample, the
onset temperature is slightly higher and the values of R are significantly larger than the other
experiments throughout the entire temperature range. It should be noted that the initial
temperature was also higher for this sample and this may have had an effect on these
observations.
CONCLUSIONS
A DSC heating rate study was performed on the dehydration and decomposition of BTAw
using the rates of 0.5, 1, 2 and 5 C min-1. This yielded, for the dehydration step, an activation
energy of 197 5 kJ mol-1 and a ln(Z/min-1) value of 46 1. For the decomposition step, values
of 222 16 kJ mol-1 and 38 4 min-1 were calculated for the activation energy and the preexponential factor, respectively. The results from the half-life test indicated that the ASTM
method E698 [4] provided a good estimate of the kinetic parameters for BTAw.
A second heating rate study, this time using TG, was performed on BTAw using 1, 2, 5, and
10 C min-1. This yielded an activation energy of 236 5 kJ mol-1 and a ln(Z/min-1) value of 41
48
Czech Republic
2006
[Content]
1. These results, in conjunction with the DSC results, provide a good estimate of the kinetic
parameters for decomposition of BTA.
Results from DSC, TG, TG-DTA-FTIR-MS and ARC all suggest that BTAw first loses
water before undergoing any exothermic reaction and that BTAw is more thermally stable than
BTA. The enthalpy for the dehydration step is consistent with the energy of strong hydrogen
bonding. Results from TG-DTA-FTIR-MS experiments indicate that BTAw decomposes with
similar gaseous products as those from tetrazole and 5-aminotetrazole, which may suggest a
similar mechanism of decomposition is involved. However, further studies are required,
including an analysis of the solid residue left over after decomposition.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
1 D.E. Chavez and M. A. Hiskey, J. Pyrotechnics, 7 (1998) 11.

2 D.E. Chavez, M. A. Hiskey and D. L. Naud, J. Pyrotechnics, 10 (1999) 1.
3 V.V. Nedelko, B.L. Korsounskii, A.V. Shastin, N.V. Chukanov, T.S. Larikova, in: Proceedings
of the 29th Annual Conference of ICT on Energetic Materials and Technology, Karlsruhe,
Germany, 30 June 3 July 1998.
4 ASTM E 698 01, Standard Test Method for Arrhenius Kinetic Constants for Thermally
Unstable Materials, American Society for Testing and Materials, West Conshohocken, PA, U.S.A.
5 ASTM E 968, Standard Practice for Heat Flow Calibration of Differential Scanning
Calorimeters, American Society for Testing and Materials, West Conshohocken, PA, U.S.A.
6 ASTM E 967, Standard Practice for Temperature Calibration of Differential Scanning
Calorimeters and Differential Thermal Analyzers, American Society for Testing and Materials,
West Conshohocken, PA, U.S.A.
7 ASTM E 1582 00, Standard Practice for Calibration of Temperature Scale for
Thermogravimetry, American Society for Testing and Materials, West Conshohocken, PA, U.S.A.
8 R. Turcotte, R.C. Fouchard, A.-M. Turcotte and D.E.G. Jones, J. Therm. Anal. Cal., 73 (2003)
105.
9 D.I. Townsend and J.C. Tou, Thermochim. Acta, 37 (1980) 1.
10
ASTM E 1981-98, Standard Guide for Assessing the Thermal Stability of Materials by
Methods of Accelerating Rate Calorimetry, American Society for Testing and Materials, West
Conshohocken, PA, U.S.A.
11
ASTM E 1641 04, Standard Test Method for Decomposition Kinetics by
Thermogravimetry, American Society for Testing and Materials, West Conshohocken, PA, U.S.A.
12
C.A. Wight, Isoconversional Data Analysis Program, Center of Thermal Analysis,
University of Utah, Version 1.42, 2000.
13 S. Vyazovkin and D. Dollimore, J. Chem. Inf. Comp. Sci., 39 (1996) 42.
14
S.V. Levchik, O. A. Ivashkevich, A. I. Balabanovich, A. I. Lesnikovich, P. N. Gaponik
and L. Costa, Thermochim. Acta, 207 (1992) 115.
15
A.I. Lesnikovich, O.A. Ivashkevich, G.V. Printsev, P.N. Gaponik and S.V. Levchik,
Thermochim. Acta, 171 (1990) 207.
16
A.I. Lesnikovich, S.V. Levchik, A.I. Balabanovich, O.A. Ivashkevich and P.N. Gaponik,
Thermochim. Acta, 200 (1992) 427.
17 A.I. Lesnikovich, O.A. Ivashkevich, S.V. Levchik, A.I. Balabanovich, P.N. Gaponik and A.A.
Kulak, Thermochim. Acta, 388 (2002) 233.
49
Czech Republic
[Content]
N
H
H
N
N
N
2006
N
C
N
H
H2O
Figure 1 Structure for BTAw.
Figure 2 DSC curves for BTAw at 5 C min-1 (____); 2 C min-1 (__ __ __); 1 C min-1 ();
..
0.5 min-1 (
).
50
Czech Republic
2006
[Content]
-2
-1
-1
-ln(
T /K min )
9.0
10.0
11.0
12.0
DSC
DTG
13.0
2.30
2.35
2.40
2.45
2.50
10 K/T
100
80
60
40
-1
120
20
50
100
150
200
250
300
DTG/% C
TG/mass %
Figure 3 Plot of ln( T-2/K-1 min-1) vs. 103 K/T for the dehydration step using data from DTG
and DSC.
-1
350
Temperature/C
Figure 4 TG and DTG mass loss curves for BTAw for = 5 C min-1.
10
51
Czech Republic
2006
[Content]
2.5
-1
ln(
/K min )
2.0
1.5
1.0
0.5
0.0
Conversion/%
5
10
15
20
-0.5
2.35
2.40
2.45
2.50
10 K/T
Figure 5 Plot of ln(/K min-1) vs. 103 K/T for the dehydration step with various heating rates.
Figure 6 BTAw (pre-dried at 50 C) IsoKin TG dehydration results ( = 1 to 10 C min-1).
52
11
Czech Republic
2006
[Content]
-2 -1
-1
-ln(
T /K min )
10
11
12
13
DSC
DTG
14
1.75
1.80
1.85
1.90
1.95
103 K/T
Figure 7 Plot of ln(T-2/K-1 min-1) vs. 103K/T for the thermal decomposition step using data
from DTG and DSC.
endo
Heat Flow/W g
-1
BTA
Aged at T = 234 C
Unaged
200
225
250
275
300
Temperature/C
Figure 8 Comparison of the DSC thermal curve of an aged BTA sample with that of an unaged
sample.
12
53
Czech Republic
2006
[Content]
Figure 9 BTAw IsoKin DSC results for = 0.5 to 5 C min-1.
2.5
-1
ln(
/K min )
2.0
1.5
1.0
0.5
0.0
Conversion/%
5
10
15
20
-0.5
1.80
1.85
1.90
1.95
2.00
103 K/T
Figure 10. Plot of ln(/K min-1) vs. 103 K/T for the thermal decomposition step with various
heating rates.
54
13
Czech Republic
2006
[Content]
Figure 11. BTAw (pre-dried at 50 C) IsoKin TG decomposition results for = 1 to 10 C min-1.
14
55
Czech Republic
100
1.5
1.0
= 5 C min -1
5 mg
60
0.5
0.0
20
7
6
CO + HN3
CO2
H2O
HCN
NH3
4
3
HN3
[N C]
1
0
10000
MS Ion Current/pA
c)
[Content]
-0.5
FTIR 10 Absorbance
b)
2006
DTA
T/C
a)
TG Mass %
1000
m/e
17
18
28
29
30
42
43
44
100
10
1
50
100
150
200
250
300
350
Temperature/C
Figure 12 5 mg BTAw heated in helium at 5 C min-1: a) TG-DTA; b) FTIR; c) MS.
56
15
a)
Czech Republic
100
80
= 5 C min
5 mg
60
-1
1
0
40
-1
20
0
b)
FTIR Absorbance
0.6
0.5
2006
[Content]
DTA
T/C
TG Mass %
-2
0.25
0.20
CO + HN3
CO2
0.15
0.4
H2O
0.10
HCN
NH3
0.3
0.05
N2O
0.2
[N C]
0.00
0.1
50
100
150
200
250
300
350
0.0
c)
MS Ion Current/nA
100
10
m/e
1
17
18
42
43
44
52
61
0.1
0.01
0.001
50
100
150
200
250
300
350
Temperature/C
Figure 13 5 mg BTAw heated in air at 5 C min-1: a) TG-DTA; b) FTIR; c) MS.
Figure 14 Example of ARC results for BTAw: pre-dried at 50 C, and run in a closed system of
ambient Ar.
16
57
Czech Republic
2006
[Content]
ln(R/C min -1 g-1)
-1
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5.0
10 K/T
Figure 15 Comparison of rate vs. temperature for various methods. Dried in-situ at 120 C and
then run in ambient air ( ); Pre-dried externally at 50 C and then run in ambient Ar (- - -);
Pre-dried externally at 50 C and then run in ambient air ( ); Pre-dried externally at 120 C
and then run in ambient Ar ( ).
58
17
18
Method
Conversion % E/kJ mol
DSC ASTM E698
197 5
DTG
182 4
MFK
178 3
TG ASTM E1641
5
192 6
10
194 5
15
194 5
20
193 5
a)
Calculated at 144 C
b)
Calculated at 265 C
-1
ln(Z/min )
46 1
41 1
57 2
57 2
57 1
57 1
-1
Dehydration
ln(k/min )
-11 2
-11 1
-1 1
23
13
12
12
-1 a)
Table 1 Comparison of kinetic parameters from the variable heating rate studies.
E/kJ mol
222 16
236 5
180 6
197 9
203 10
209 8
-1
ln(Z/min-1)
38 4
41 1
43 1
46 2
47 2
48 2
Decomposition
ln(k/min-1) b)
-12 6
-12 1
23
23
13

Czech Republic
2006
[Content]
59
60
air
Atmosphere Sample Mass/g
Ti/C
0.48
150
0.49
150
Externally at 50 C
Ar
0.53
100
0.53
100
Externally at 50 C
air
0.51
100
0.49
100
Externally at 120 C
Ar
0.50
100
0.50
100
a)
Uncertainty from the extrapolation of the onset temperature.
b)
Pressure recorded at ambient temperature following the run.
c)
lnk calculated at 220 C.
In-situ at 120 C
Drying Method
213
215
205
203
211
201
203
198
(To 5)/C a)
Table 2 Comparison of ARC results from the various drying methods.
100 2
101 2
77.2 0.7
76.2 0.7
87.2 0.7
75 1
84 5
78.1 0.6
E/kJ mol-1 ln(Z/min-1)
0.37 @ 20 C 418 7
0.30 @ 22 C 425 10
0.27 @ 38 C 322 3
0.68 @ 21 C 319 3
0.52 @ 20 C 365 3
0.51 @ 26 C 315 4
0.48 @ 25 C 351 2
0.53 @ 39 C 326 2
Pf/MPa b)
-2 3
-3 3
-1 1
-2 1
-2 1
-2 1
-2 5
-1.4 0.8
ln(k/min-1) c)

Czech Republic
2006
[Content]
19
Czech Republic
2006
[Content]
NEW TRENDS IN DETECTION OF EXPLOSIVES

P. Mostak, M. Stancl
Research Institute of Industrial Chemistry, Explosia
53217 Pardubice-Semtin, Czech Republic
Abstract:
Detection methods used in screening procedures looking for hidden explosive charges
and improvised explosive devices (IED) are presented and factors influencing the
detection efficiency discussed. The simultaneous use of x-ray screening and electronic
detectors adds the substantial synergy to the efficiency of explosive detection.
Exploitation of dogs and colour reactions are another important detection methods. The
improvement of imaging ability of X-ray systems and the increasing sensitivity of
electronic detectors should be expected in near future. The progress in neutron
technologies should be also achieved. LIBS, MW and radar systems have a chance to
fulfil the gap in the stand- off detection. The detection on distance is still the serious
problem in the defence against bomb attacks performed by suicide bombers and big
explosive charges situated in cars. The effort in search for principally new detection
methods is required.
Keywords:
detection
1.
detection of explosives, improvised explosive devices, marking for
INTRODUCTION
The experience obtained in the fight with terrorist attacks by explosives have proved,
that the tools used by terrorists in bomb attacks are more sophisticated, more dangerous,
effective countermeasures complicated and in some cases not adequate.
The broad use of remote control of IED initiation by mobile phones gives to terrorists
the choice of exact moment of the bomb attack and also the attacks performed by suicide
bombers opened new challenge to efficient countermeasures including quick detection of
explosive charge or/and IED on distance.
Therefore, more effective detection systems and following countermeasures are
required and the research and development leading to substantial improvement of detection
methods and a development of new detection technologies is a permanent important task
The aim of this presentation is to analyse the efficiency of detection techniques and
discuss main factors, which influence the explosive detection. The important tendencies,
leading to the improvement of detection efficiency, will be discussed.
Petr Mostak, Miroslav Stancl
61
2.
Czech Republic
2006
[Content]
MAIN DETECTION METHODS
The important task in screening operations is covered by the visualisation of the

checked object content by X-rays or other imaging techniques. In fact X-ray equipment is
the most frequently used tool for detection of hidden explosives at stabile checking points,
the visualisation enables also the search of further contraband as weapons and narcotics.
Electronic detectors of explosives are further effective method for detection of
explosive charges by analysing vapours and/or particles of explosives.
The very important tools of explosive detection are special trained dogs. The
sensitivity of the dog nose is unique and electronic detectors have not yet overcome the high
detection ability of dog. Further technologies as nuclear methods, NMR, NQR are in
development stage or in field tests.
The simultaneous use of different detection systems usually gives the substantial
synergy to the detection efficiency, typical example is integration of X-ray and electronic
detectors at stabile checking points in airports, other transport terminals and important
buildings.
3.
FACTORS INFLUENCING THE DETECTION EFFICIENCY
IED design
There are many types of explosives used in IEDs and many possible designs, differing
in initiation chain, starting mechanism and packing arrangement. The range of explosive
mass used in IEDs should be very different, starting with several grams in letter bombs and
ending with the mass of several tons of explosive in a lorry. There are also many different
scenarios in which IEDs are used.
Very dangerous are mobile or radio controlled IEDs situated in transport means, on
roads, streets and places with high concentration of people.
Extremely dangerous are:
-
IEDs fixed on bodies or carried in personal luggage of suicide bombers
IEDs in cars containing big amount of explosives or ammunition
Electronic detectors
Electronic detectors based on IMS and GC are able to detect, by analysing vapours
and/or particles, most of military and civil explosives and also organic peroxides. The
sensitivity of this equipment is fully sufficient in particle detection, collection of 1 particle
usually enables the positive result.
The vapour detection is dependent on conditions at detection, from which most
important are the vapour pressure of explosive, barrier properties of packing, the
temperature, air movement and the sampled air quantity.
In unfavourable conditions, the vapour detection is not fully reliable method.
Therefore, the systems using simultaneously vapour and particle detection are effective and
62
Czech Republic
2006
[Content]
often used. The detection efficiency is increased by pre-concentration of explosive traces

from sampled air, some equipment is able to work with pre-concentration factor as high as
100 000. Such high pre-concentration factor is typical for stabile screening equipment.
Dogs
The dog remains the unique tool for trace detection. The dog is able to detect
explosives even at low temperatures at which most detectors are not fully effective. No
exact knowledge about the mechanism of dog sniffing is at disposal, but it seems, that dog is
able to sniff not only vapours but also particles of explosives and even explosives fixed on
the surface of dust.
Dogs have their place in search for hidden IEDs in various scenarios and the efficiency
of dogs has been confirmed by steady rise of dog numbers used in security missions.
Colour reactions
Explosives produce colour reactions with specific reactants, and this knowledge has
been exploited in trace detection. The sensitivity of this detection method is lower in
comparison with electronic detectors, but when we are able to collect several particles of
explosive, the detection is reliable. Usually, collection of 1 average particle is sufficient for
positive detection. Detection sets for colour reaction can detect also AN, chlorates and
perchlorates and organic peroxides, it means important components of some civil and
improvised explosives.
Imaging and radiation technologies

X-ray systems are broadly used as the basic screening tool for control of luggage,
postal sending and further objects. The progress of X-ray technology is steady and enables
the better imaging of the object content, scanning and 3 dimensional screening. New
developed X-ray method can identify the chemical compounds present in suspicious part of
object by angle diffraction spectrum. Diffraction spectrum is typical for each compound and
substantially increases the possibility of detection and identification of explosive charge.
Further radiation and wave technologies as neutron in-gamma out detection, LIBS
spectroscopy, radar, NMR, QMR and other wave systems are in development phase and/or
in
field tests.
Detection of marking agents in plastic explosives
Practical importance among marking agents, specified in the Technical Annex of
Montreal Convention on the Marking of Plastic Explosives for the Purpose of Detection,
have only following compounds:
-
2,3-Dimethyl-2,3-dinitrobutane (DMNB)
para-Mononitritoluene (p-MNT).
Vapour detection of DMNB, the most popular marking agent for plastic explosives, by
IMS is not very effective. This is caused by the fact, that cluster ions formed from molecules
of DMNB by irradiation have the low thermal stability and therefore, the yield obtained after
Petr Mostak, Miroslav Stancl
63
Czech Republic
2006
[Content]
acceleration in drifting tube is low. The detection of DMNB by GC is excellent.

Unfortunately, most detectors used in detection practice at this time are based on IMS. The
reason are some advantages of IMS in comparison with GC such as lower price, lower
weight, absence of carrying gas, and a very short time of analysis. This discrepancy is to be
solved by some acceptable way, but the solution will be not easy.
4.
STAND- OFF DETECTION
Extremely difficult problem, which emerged in last years, is the massive use of IEDs
controlled on distance by radio or mobile phone and attacks by bombs fixed oh the body of
suicide bombers. Also attacks by cars filled by big amount of explosives present high danger
and consequences of such attacks are very heavy.
The experience, obtained in Iraq, Chechnya and other countries, confirms, that the
detection of these terrorist tools on distance and effective countermeasures is a very uneasy
task. No effective, service ready technology for stand-off detection is at disposal now.
Therefore, research and development of effective detection technology for stand-off
detection should have the highest priority.
5.
EXPECTED PROGRESS IN DETECTION TECHNOLOGIES

The progress in electronic detectors should be expected in following directions:
-
increasing sensitivity of vapour and particle detectors
higher efficiency of pre-concentrators
broader use of micro and nano technologies to decrease mass, size and cost
of instruments
Further increase of performance of x-ray screening systems should be attained and

broader application of angle diffraction spectra will be achieved in the near future.
Some progress is to be expected in the detection using neutron technologies. The
research and development in imaging and wave technologies should lead to efficient
detection methods.
Strong effort is concentrated on techniques enabling the stand-off detection of
explosives charges. It is probable, that the results obtained in the field of LIBS, MW and
possibly radar technology will be able to bring the feasible detection method to detect
explosive charges on distance.
We can also expect, that an intensive investigation of new principles and methods
suitable for detection of explosives and IEDs will bring new knowledge and experience. This
should lead to improvement of the efficiency in the fight with the bomb terrorism and better
safety to people.
64
Czech Republic
2006
[Content]
MOLECULAR SURFACE ELECTROSTATIC POTENTIALS IN THE

COMPUTATIONAL CHARACTERIZATION OF ENERGETIC
COMPOUNDS
Peter Politzer and Jane S. Murray
Department of Chemistry
University of New Orleans
New Orleans, LA 70148
USA
Abstract:
The electrostatic potential of a molecule is a fundamental determinant of its properties
and behavior. When computed on the molecular surface of an energetic compound, it is
characterized by strong and highly variable regions of positive potential, which can be
related to the compounds impact sensitivity. The features of molecular surface
electrostatic potentials can also be used to establish quantitative expressions for heats
of sublimation and vaporization, thus permitting gas phase heats of formation of
energetic (and other) compounds to be converted to the more useful solid and liquid
values. Since these procedures are carried out computationally, they can be applied to
the assessment of proposed target compounds as well as those that have already been
prepared.
Keywords:
1.
molecular electrostatic potentials, impact sensitivities, solid and liquid

heats of formation, heats of sublimation, heats of vaporization
THE ELECTROSTATIC POTENTIAL
The remarkable Hohenberg-Kohn theorem[1] established rigorously that the

properties of a system of nuclei and electrons are determined by its electronic density (r).
Thus its energy, E, is a functional of (r):
E = [(r ) ]
(1)
The electronic density is in turn related by Poissons equation to the electrostatic potential
V(r) that is created by the systems nuclei and electrons:
2 V(r ) = 4(r ) 4
Z (r R )
A
(2)
V(r) is simply the Coulomb potential,
V(r ) =
R
A
Peter Politzer, Jane Murray
ZA
(r)dr
r r
A r
(3)
65
Czech Republic
2006
[Content]
in which ZA is the charge on nucleus A, located at RA. V(r) is a physical observable; it can
be obtained experimentally, by diffraction techniques,[2-4] as well as computationally. The
sign of V(r) in any region depends upon whether the positive contribution of the nuclei or
the negative one of the electrons is dominant there.
On the basis of the Hohenberg-Kohn theorem in conjunction with eq. (2), it can be
argued that the electrostatic potential is as fundamental a determinant of the properties of a
system as is the electronic density.[5,6] There have been derived exact expressions for atomic
and molecular energies[6-9] and the chemical potential (electronegativity)[5,6] in terms of the
electrostatic potentials at the nuclei. Functional relationships between the energies and V(r)
have been developed,[9] as well as effective formulations for ionic and covalent radii.[10,11]
In this paper, we will review work that has shown how electrostatic potentials can
help to quantify and predict two important properties of energetic compounds: impact
sensitivity and condensed phase (solid or liquid) heat of formation. For these purposes, we
compute V(r) on the surfaces of the respective molecules, which we define, following the
suggestion of Bader et al,[12] as the 0.001 electrons/bohr3 three-dimensional contour of the
electronic density (r). The resulting surface potential is labeled VS(r).
In seeking to develop quantitative relationships involving VS(r), we have found it
useful to characterize it in terms of certain statistically-defined features. The ones that will
be relevant to the present discussion include:
(1) the most positive and negative values of VS(r), the VS,max and the VS,min;
(2) the positive and negative averages, VS+ and VS- , given by
VS+ =
1 m +
VS (ri )
m i =1
VS- =
1 n VS (rj )
n j=1
(4)
(3) the total variance of VS(r), 2tot , and its positive and negative components, 2+ and
2 ,
2tot = 2+ + 2 =
2
2
1 m
1 n
VS+ (ri ) VS+ + VS- (rj ) VS-
m i =1
n j=1
(5)
(4) an electrostatic balance parameter,

=
2+ 2
2tot
(6)
The first summations in eqs. (4) and (5) are over the points where VS(r) is positive, the
second over those where it is negative, on a grid encompassing the molecular surface.
The quantities 2tot , 2+ and 2 are measures of the variabilities of the total, positive
and negative surface potentials, reflecting how extensive are the ranges that they cover. The
parameter is an indicator of the degree of balance between the positive and negative
potentials; when 2+ = 2 , then reaches its maximum value of 0.25.
We have shown, in a series of studies, that the quantities in terms of which we
characterize VS(r) (which include also its average deviation) provide a basis for developing
analytical relationships for a variety of condensed phase physical properties that depend
66
Molecular Surface Electrostatic Potentials in the Computational . . .
Czech Republic
2006
[Content]
upon noncovalent interactions: boiling points and critical constants, heats of phase
transitions, solubilities and solvation energies, partition coefficients, diffusion constants,
viscosities, surface tensions, etc. For reviews of this work, see Murray et al.[13-15]
SENSITIVITY
2.
2.1
Background
In the present context, sensitivity denotes the vulnerability of an energetic compound

to accidental detonation, i.e. caused by an unintended external stimulus. Such stimuli can
include impact, shock, heat, friction and electrostatic charge.[16,17] These various types of
sensitivity do show some approximate similarities in trends. For example, Storm et al have
found correlations between impact and shock sensitivity, and betweeen the latter and the
critical temperature at which thermal decomposition becomes self-sustaining.[18] Zeman et
al obtained relationships involving impact and electric spark sensitivities.[19]
Minimizing sensitivity while maximizing detonation performance is a major
challenge, to which a great deal of attention and effort is being devoted. Energetic
compounds are inherently metastable. The introduction of external energy, e.g. via impact,
can lead to decomposition, first endothermic and then highly exothermic, releasing gaseous
products and a large amount of energy and producing a shock front that moves through the
compound at supersonic speed causing continuing and self-sustaining exothermic
decomposition (detonation).[16,17,20,21]
The likelihood that such a series of events will take place for a particular compound,
i.e. its sensitivity, depends upon several factors. These include its molecular composition
and structure, crystal properties, and physical state, e.g. sizes of particles. Thus, polymorphs
of a given energetic solid may differ in sensitivity, as may crystals of various sizes and
shapes.
Since crystal and physical features can have such importance, it is somewhat
remarkable that quite good correlations have also been obtained between sensitivity
(especially toward impact) and diverse properties of individual molecules (usually of a
specific chemical type, e.g. nitramines). Among these properties are different indicators of
the stabilities of CNO2 and NNO2 bonds (trigger linkages), electronic excitation
processes, rates of energy transfer into certain vibrational modes, and stoichiometry. There
have been several reviews of these studies.[22-24] A cautionary note was sounded by Brill and
James,[25] who pointed out that the existence of these correlations does not necessarily imply,
in each instance, any fundamental significance.
2.2
VS(r) and Sensitivity
As was discussed above, sensitivity has been found to correlate quantitatively with
some rather disparate molecular features, and it has been linked qualitatively to yet others,
such as the ability to form strong intermolecular hydrogen bonds.[22-24] One possible
interpretation of these observations is that they reflect different aspects of the overall global
nature of the molecule, and that a useful approach might be to try to characterize it as a
whole, globally, instead of focusing upon certain bonds, vibrational modes, etc. We have
explored, with encouraging success, the possibility of doing this in terms of the electrostatic
potential, the fundamental nature of which has already been demonstrated.
There is a great deal of empirical evidence indicating a relationship between the
electrostatic potential and molecular stability. Paulings recognition of this is embodied in
67
Czech Republic
2006
[Content]
his electronegativity postulate,[26] which states essentially that the charge distributions in
molecules tend to be such as to make each atom as close to electrically neutral as possible.
This translates into the electrostatic potential on the molecular surface, VS(r), being
relatively uniform and fairly weak; deviations from this promote instability.
This reasoning has long been widely applied in chemistry; delocalization of charge is
commonly associated with enhanced stability. For example, relative pKa values of
carboxylic acids are often interpreted on the basis of how well the charge of the resulting
anion can be delocalized, thereby increasing its stability and lowering the acids pKa. We
have recently shown that the magnitudes of the most positive potentials on the surfaces of
carbocations, the VS,max, are good measures of their stabilities;[27] the less positive is VS,max,
the more has the charge delocalized and the more stable is the carbocation.
Before examining the VS(r) of the molecules of energetic compounds, it is useful to
gain some perspective by looking at a group of more representative ones. In Table 1 are
listed, for a diverse series of organic molecules, some of the computed quantities that we use
to characterize VS(r): (a) its most positive and negative values, VS,max and VS,min, (b) the
positive and negative averages, VS+ and VS- , and (c) the variances, 2tot , 2+ and 2 . Also
included are the postive and negative surface areas, AS+ and AS .
An overall trend is immediately apparent. For most organic molecules, the regions of
positive surface potential are larger in area but weaker than are the negative ones. There are
some exceptions, such as anthracene and 1,4-dichlorobenzene (see Table 1), but these are
unusual. Thus, it is normally the case that AS+ > AS , but VS,min > VS,max , VS- > VS+ and 2 >
2+ . The differences are often quite large, as can be seen in Table 1. A more extensive data
set confirms that these are general patterns.[29]
Table 1. Computed surface quantities (HF/STO-5G*//STO-3G*) for some representative

molecules.a,b
Molecule
AS+
A-S VS,max VS,min
VS+
VS 2+
-2
benzene
50.6
58.0
9.5
10.2
4.6
4.9
7.3
8.5
butylbenzene
123.6
78.3
8.5
12.4
3.7
4.5 4.9
13.0
anthracene
100.1 112.6 10.6
9.4
5.3
5.2
8.8
6.8
dimethylsulfide
68.3 32.4 11.2
17.0
4.1
5.7
7.2
25.7
1-hexanol
131.2
51.7 12.6
36.7
5.0
9.2 9.9
132.5
cyclohexanol
118.7
26.7 26.7
43.5
5.6
15.9 18.6
207.5
methanol
44.8 21.5 29.9
41.6 10.3
18.4 48.9
181.9
ammonia
27.7 19.7 18.1
29.0
9.2
12.8 27.6
73.7
methylamine
52.0 20.9 21.4
53.2
9.3
20.2 34.6
263.0
piperazine
92.2
38.5 22.9
47.8
9.3
17.4 26.3
207.2
phenol
65.0
63.0 34.9
36.6
8.6
8.5 64.0
73.1
acetamide
62.7 31.8 32.0
40.6 12.7
20.6 68.0
150.8
1,4-dichlorobenzene 95.8 56.7 15.2
11.1
6.2
7.0 18.0
10.2
3-nitrochlorobenzene 101.9 54.2 22.3
37.5 12.3
16.7 21.7
124.4
________________________________________________________________________
a
Units: AS+ and A-S are in A2; VS,max, VS,min , VS+ and VS- are in kcal/mole; 2+ and -2 are in
(kcal/mole) 2.
b
Data taken from reference 28.
68
Czech Republic
2006
[Content]
The situation is quite different for molecules of energetic compounds, as is shown in

Table 2. The available data are more limited, but it is clear that the positive regions are now
stronger than the negative. In every instance, VS+ > VS- and 2+ > 2 . The striking contrast
between energetic and typical organic molecules is due to the electron-attracting powers of
the nitro groups[30] and the ring nitrogens[31] of the former. These withdraw electronic
charge from the central portions of the molecules, thereby creating strong and extensive
positive regions, sometimes having several local VS,max. Thus, the negative VS(r) above the
center of the ring in benzene, associated with its electrons, is not present in TNB, TNT and
picric acid. The only negative surface potentials of energetic molecules are likely to be
above parts of their peripheries, e.g. NO2 oxygens, ring nitrogens, the carbonyl oxygen in
NTO, etc. These various features of the VS(r) of energetic molecules can be seen in color
figures in several earlier reports, by us[24,32] and by Rice and Hare.[33] It should be pointed
out, however, that a single NO2 or ring nitrogen is not sufficient to produce these effects, as
is evident from the examples of nitrobenzene, which has VS+ < VS- and 2+ < 2 [34], and
piperazine (Table 1). It is necessary that these components constitute some critical portion
of the entire molecule, just as is required for energetic performance.
Table 2. Computed surface quantities (HF/STO-5G*//STO-3G*) for some energetic

molecules.a,b
Molecule
AS+
A-S VS,max VS,min
VS+
VS 2+
-2
1,3,5-trinitrobenzene,
TNB
---
---
39.2
---
23.9
15.3
109.0
55.3
2,4,6-trinitrotoluene,
TNT
125.0
77.8
37.5
28.8
20.6
14.7
104.3
53.2
2,4,6-trinitrophenol,
picric acid
---
---
---
---
24.9
17.0
70.7
53.1
---
---
21.0
96.3
73.8
---
---
21.8
3-nitro-1,2,4-triazol-5-one,
NTO
1,3,3-trinitroazetidine,
TNAZ
18.0
12.2
174.3
111.0
142.0
87.5
81.0
42.6
octanitrocubane,
ONC
203.3 110.3
----23.8
6.9
407.3 16.5
________________________________________________________________________
a
Units: AS+ and A-S are in A2; VS,max, VS,min , VS+ and VS- are in kcal/mole; 2+ and -2 are in
(kcal/mole) 2.
b
Data are from references 28 and 34 and unpublished results.
In the present context, it is also relevant to mention a particular unusual buildup of

positive potential, with a local VS,max, that is found above the CNO2 bonds in
nitroaromatics, nitroheterocycles and nitroalkanes.[32,35,36] This is not true of chemical bonds
5
69
Czech Republic
2006
[Content]
in general. We have shown that there is a limited inverse correlation between this VS,max and
the CNO2 bond energy,[35,36] which is consistent with early observations of a relationship
between nitroaromatic impact sensitivities and the roughly estimated electrostatic potentials
of the CNO2 bonds.[36] (Another interesting aspect of the positive buildup above CNO2
bonds is that it has been demonstrated to provide a channel for the initial approach of
nucleophiles.[37,38])
Can it be inferred from the preceding discussion that the metastabilities of energetic
compounds are reflected in their characteristic molecular surface electrostatic potentials,
specifically the unusually strong and highly variable positive components? If so, can this
provide a basis for ranking their crystalline sensitivities? We have shown that this is indeed
possible.[34] We focused upon the imbalance between the positive and negative surface
potentials. There are of course a number of ways to express this imbalance, e.g. in terms of
VS,max VS,min , VS+ VS- , VS+ / VS- , 2+ -2 , 2+ / -2 , etc., or combinations of these. We
were most successful when we used a different combination for each class of compounds
considered nitroaromatics, nitramines and nitroheterocycles. This produced very good
correlations with impact sensitivity, as obtained by the drop-weight method; [16-18] they all
had R2 0.980, with standard deviations of 8 to 14 cm, for experimental databases that
covered, respectively, 284, 307 and 256 cm. The detailed relationships are presented in the
original paper.[34] We certainly do not view these as final. As larger databases become
available, or with more judicious choices of variables, they are likely to be modified, perhaps
simplified. Nevertheless, it is gratifying that such good correlations are already attainable.
2.3
Discussion
In would clearly be preferable to be able to include all chemical classes in a single

correlation. This would be desirable both conceptually, since it might provide insight into
key factors determing impact sensitivity, and also from a practical standpoint, because there
can occasionally be some ambiguity concerning the chemical class to which a compound
should be assigned. Developing one general relationship remains an important objective.
What is very significant, however, is the unifying concept that the impact sensitivities
of energetic compounds are related to specific features of their molecular surface
electrostatic potentials: their characteristic anomalously strong and variable positive
components. The extensive study by Rice and Hare supports this concept, although they feel
that the key positive regions are those above covalent bonds within the inner molecular
framework.[33]
In section 1 of this paper, we noted that a number of condensed phase physical
properties have been expressed analytically as functions of quantities characterizing their
molecular VS(r). It is therefore at least plausible that by linking sensitivity to VS(r), we are
taking some account of the effects of crystal properties upon sensitivity, which was also
mentioned earlier (section 2.1).
3.
SOLID OR LIQUID PHASE HEAT OF FORMATION
The heat of formation Hf is a key property of an energetic compound. It allows the

determination of the energy release upon combustion or decomposition, the latter in turn
being used to calculate detonation pressure and velocity,[17,21,39] two primary measures of
explosive performance.
70
Czech Republic
2006
[Content]
Since experimental Hf are often not available, or may be unreliable,[40,41] a great

deal of effort has gone into developing procedures for producing them computationally. One
option is to utilize an ab initio or density functional technique to obtain H for either the
formation of the molecule from its elements or for its separation into the constituent atoms.
The first yields Hf directly; the second may require appropriate inclusion of some
dissociation energies. Due to the remarkable advances in methodology, software and
processor technology in recent years, these approaches can give quite accurate results for
molecules of practical interest, e.g. propellant combustion products,[42] although size is
certainly a limitation. For larger molecules, it is common to introduce empirical correction
terms. Another widely-used route to Hf is by calculating H for isogyric, isodesmic or
homodesmic reactions, which are designed to achieve some cancellation of error. For
comparisons and assessments of these and other procedures, see Stewart,[40] Irikura and
Frurip,[43] Thiel[44] and Politzer et al.[45]
The various schemes that have been mentioned typically lead to predictions of gas
phase heats of formation. What is normally relevant for energetic compounds, however, is
Hf for the corresponding solid, or less often the liquid. One can arrive at these from the gas
phase value if the heats of sublimation and vaporization, Hsub and Hvap, are known:
H f (solid) = H f (gas) H sub
(7)
H f (liquid) = H f (gas) H vap
(8)
Our present interest is not in the different ways of obtaining H f (gas) computationally, but
rather in H sub and H vap . If these are not known, as is often the case, then some method
for estimating them is needed.
It was already brought out in section 1 that heats of phase transitions are among the
physical properties that can be expressed in terms of the quantities that characterize VS(r),
defined by eqs. (4) (6). The relationships for H sub [46] and H vap [47] are:
2
H sub = 1AS,tot
+ 2 ( 2tot ) + 3
0.5
H vap = 1 ( AS,tot ) + 2 ( 2tot ) + 3

0.5
0.5
(9)
(10)
In eqs. (9) and (10), AS,tot is the total surface area of the molecule, AS,tot = AS+ + AS- . The
parameters i and i , i = 1-3, can be evaluated by fitting to experimental H sub and H vap .
Our most recent parametrizations of eqs. (9) and (10), at the density functional B3PW91/631G** level, was with general databases of 66 H sub and 30 H vap .[48] We were able to
reproduce these properties for 105 and 44 compounds, respectively, including many that
71
Czech Republic
2006
[Content]
were not involved in the parameter fittings, with average absolute deviations from
experiment of 2.8 kcal/mole for H sub and 2.0 kcal/mole for H vap .
We have used eqs. (9) and (10) in computing heats of formation for numerous
energetic solids (and one liquid),[45,46,49-53] many of them being proposed target compounds.
Rice et al have carried out very extensive and detailed calculations of the gas, liquid and
solid phase heats of formation of energetic compounds, utilizing eqs. (9) and (10) to find
H sub and H vap .[54,55] They parametrized their entire procedure with databases containing
exclusively C/H/N/O energetic compounds. Their root-mean-square deviations in the solid
and liquid phase heats of formation of compounds not included in the fitting were 5.6 and
3.2 kcal/mole, respectively.[55]
The focus in this section has thus far been entirely upon molecular solids and liquids.
For an ionic solid, the heat of formation can be obtained by converting its lattice energy to
H for the process[56]
MmXn(s) mM+a(g) + nXb(g)
(11)
and combining it properly with the experimental or computed Hf of the gas phase ions. We
have shown, for three different cations (NH4+, Na+ and K+), that the lattice energies of their
salts can be represented in terms of the quantities characterizing the VS(r) of the anions.[57]
In this manner, we have predicted Hf for NH4O2 and NH4O3.[58]
4.
CONCLUDING REMARKS
We have shown how the electrostatic potential, a fundamental determinant of

molecular properties and behavior, can be applied to the characterization of energetic
compounds, with respect to their impact sensitivities and their solid or liquid phase heats of
formation. A point to be emphasized is that the procedures that have been described are
entirely computational in nature. They can accordingly be used to predict the impact
senstivities and heats of formation of proposed target compounds as well as others that may
have been prepared but are not readily amenable, for whatever reason, to experimental
analysis. These procedures can therefore play key roles in the assessment of potential
energetic performance.
REFERENCES
[1] P. Hohenberg and W. Kohn, Phys. Rev. B 136, 864 (1964).
[2] R. F. Stewart, J. Chem. Phys. 57, 1664 (1972).
[3] P. Politzer and D. G. Truhlar, eds., Chemical Applications of Atomic and Molecular
Electrostatic Potentials, Plenum Press, New York, 1981.
[4] G. Naray-Szabo and G. G. Ferenczy, Chem. Rev. 95, 829 (1995).
[5] P. Politzer and J. S. Murray, in Molecular Electrostatic Potentials: Concepts and
Applications, J. S. Murray and K. Sen, eds., Elsevier, Amsterdam, 1996, ch. 16.
[6] P. Politzer and J. S. Murray, Theor. Chem. Acc. 108, 134 (2002).
[7] P. Politzer and R. G. Parr, J. Chem. Phys. 61, 4258 (1974).
[8] P. Politzer, P. Lane and M. C. Concha, Int. J. Quantum Chem. 90, 459 (2002).
72
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
Czech Republic
2006
[Content]
P. Politzer, Theor. Chem. Acc. 111, 395 (2004).

K. D. Sen and P. Politzer, J. Chem. Phys. 90, 4370 (1989); 91, 5123 (1989).
P. Politzer, J. S. Murray and P. Lane, J. Comp. Chem. 24, 505 (2003).
R. F. W. Bader, M. T. Carroll, J. R. Cheeseman and C. Chang, J. Am. Chem. Soc. 109,
7968 (1987).
J. S. Murray and P. Politzer, J. Mol. Struct. (Theochem) 425, 107 (1998).
P. Politzer and J. S. Murray, Trends Chem. Phys. 7, 157 (1999).
P. Politzer and J. S. Murray, Fluid Phase Equilib. 185, 129 (2001).
S. Iyer and N. Slagg, in Structure and Reactivity, J. F. Liebman and A. Greenberg, eds.,
VCH Publishers, New York, 1988, ch. 7.
J. Khler and R. Meyer, Explosives, 4th ed., VCH Publishers, New York, 1993.
C. B. Storm, J. R. Stine and J. F. Kramer, in Chemistry and Physics of Energetic
Materials, S. N. Bulusu, ed., Kluwer, Dordrecht (The Netherlands), 1990, ch. 27.
S. Zeman, Z. Kamensky, P. Valenta and J. Jakubko, Proceedings of the Study Days on
the Sensitivity of Energetic Components and Substances to Electrostatics, Aussois,
France, 1996, p. 197.
P. A. Urtiew and B. Hayes, J. Energet. Mater. 9, 297 (1991).
C. L. Mader, Numerical Modeling of Explosives and Propellants, 2nd ed., CRC Press,
New York, 1998.
P. Politzer and H. E. Alper, in Computational Chemistry: Reviews of Current Trends,
vol. 4, J. Leszczynski, ed., World Scientific, Singapore, 1999, ch. 6.
P. Politzer and S. Boyd, Struct. Chem. 13, 105 (2002).
P. Politzer and J. S. Murray, in Energetic Materials. Part 2. Detonation, Combustion,
P. Politzer and J. S. Murray, eds., Elsevier, Amsterdam. 2003, ch. 1.
T. B. Brill and K. J. James, Chem. Rev. 93, 2667 (1993).
L. Pauling, J. Chem. Soc. 1461 (1948).
A. M. Robbins, P. Jin, T. Brinck, J. S. Murray and P. Politzer, Int. J. Quantum Chem.,
in press.
P. Politzer, J. S. Murray and Z. Peralta-Inga, Int. J. Quantum Chem. 85, 676 (2001).
J. S. Murray, T. Brinck, P. Lane, K. Paulsen and P. Politzer, J. Mol. Struct. (Theochem)
307, 55 (1994).
O. Exner, Correlation Analysis of Chemical Data, Plenum Press, New York, 1988.
P. Lane, J. S. Murray and P. Politzer, J. Mol. Struct. (Theochem) 236, 283 (1991).
J. S. Murray, P. Lane and P. Politzer, Mol. Phys. 85, 1 (1995).
B. M. Rice and J. J. Hare, J. Phys. Chem. A 106, 1770 (2002).
J. S. Murray, P. Lane and P. Politzer, Mol. Phys. 93, 187 (1998).
P. Politzer and J. S. Murray, Mol. Phys. 86, 251 (1995).
P. Politzer and J. S. Murray, J. Mol. Struct. 376, 419 (1996), and references cited.
P. Politzer, P. R. Laurence, L. Abrahmsen, B. A. Zilles and P. Sjoberg, Chem. Phys.
Lett. 111, 75 (1984).
J. S. Murray, P. Lane and P. Politzer, J. Mol. Struct. (Theochem) 209, 163 (1990).
M. J. Kamlet and S. J. Jacobs, J. Chem. Phys. 48, 23 (1968).
J. J. P. Stewart, J. Mol. Model. 10, 6 (2004).
P. Politzer, P. Lane and M. C. Concha, Struct. Chem. 15, 473 (2004).
P. Politzer, P. Lane and M. E. Grice, J. Phys. Chem. A 105, 7473 (2001).
K. K. Irikura and D. J. Frurip, in Computational Thermochemistry, K. K. Irikura and D.
J. Frurip, eds., ACS Symposium Series 677, American Chemical Society, Washington,
1998.
W. Thiel, in Computational Thermochemistry, K. K. Irikura and D. J. Frurip, eds., ACS
Symposium Series 677, American Chemical Society, Washington, 1998.
73
Czech Republic
2006
[Content]
[45] P. Politzer, P. Lane and M. C. Concha, in Energetic Materials. Part 1. Decomposition,

Crystal and Molecular Properties, P. Politzer and J. S. Murray, eds., Elsevier,
Amsterdam, 2003, ch. 9.
[46] P. Politzer, J. S. Murray, M. E. Grice, M. DeSalvo and E. Miller, Mol. Phys. 91, 923
(1997).
[47] J. S. Murray and P. Politzer, in Quantitative Treatments of Solute/Solvent Interactions,
J. S. Murray and P. Politzer, eds., Elsevier, Amsterdam, 1994, ch. 8.
[48] P. Politzer, Y. Ma, P. Lane and M. C. Concha, Int. J. Quantum Chem. 105, 341 (2005).
[49] P. Politzer, J. S. Murray and M. E. Grice, Mater. Res. Soc. Symp. Proc. 418, 55 (1996).
[50] P. Politzer and P. Lane, Adv. Mol. Struct. Res. 3, 269 (1997).
[51] P. Politzer, M. C. Concha, M. E. Grice, J. S. Murray, P. Lane and D. Habibollahzadeh,
J. Mol. Struct. (Theochem) 452, 75 (1998).
[52] P. Politzer, P. Lane and J. J. M. Wiener, in Carbocyclic and Heterocyclic Cage
Compounds and Their Building Blocks, K. K. Laali, ed., JAI Press, Stamford, CT, 1999,
p. 73.
[53] P. Politzer, M. E. Grice and J. S. Murray, Recent. Res. Devel. Phys. Chem. 3, 95
(1999).
[54] B. M. Rice, S. V. Pai and J. Hare, Combust. Flame 118, 445 (1999).
[55] E. F. C. Byrd and B. M. Rice, J. Phys. Chem. A 110, 1005 (2006).
[56] D. R. Lide, ed., Handbook of Chemistry and Physics, 78th ed., CRC Press, New York,
1997.
[57] P. Politzer and J. S. Murray, J. Phys. Chem. A 102, 1018 (1998).
[58] P. Politzer, J. S. Murray, M. E. Grice and T. Brinck, J. Energ. Mater. 18, 89 (2000).
74
Lectures
Czech Republic
2006
[Content]
THE INFLUENCE OF GEOMETRY AND MATERIAL PROPERTIES

ON AN EXPLOSIVES GURNEY VELOCITY AND ENERGY
Joseph E. Backofen
BRIGS Co.
2668 Petersborough Street, Oak Hill, VA, USA 20171-2443
Abstract:
Experimental data from many scientific publications reveal that the copper cylinder test
used internationally to measure an explosives Gurney Velocity and Gurney Energy falls
within a unique combination of geometry and material properties factors affecting an
explosives measured performance. These data also support the need to use two
separate propulsion events to model detonation-driven propulsion: a brisant first stage
and a gas-dynamic second stage.
Keywords:
1.
detonation performance, detonation, explosive, cylinder test,

Gurney models, Gurney Velocity, Gurney Energy
INTRODUCTION
This paper presents and employs a practical means by which to model an explosives
detonation-driven propulsive effects by separating these into two parts: 1) initial motion
imparted by a brisant shock-dominated process that depends upon intimate contact of an
explosive with the propelled material, and 2) subsequent acceleration by a gas-push (gasdynamic) process. Initial motion is envisioned as being caused by the higher-pressure region
of a detonation front (i.e. envision the von Neumann spike or reaction zone region as being a
finite thickness of solid material squeezed at high pressure). The gas-push process is
envisioned similar to that assumed by Gurney modeling. The two-stage modeling employs
an engineering methodology consistent with the Zeldovich-von Neumann-Doering (ZND)
detonation model.
The need for a new model was experienced more than thirty years ago when studying
the effects of explosives during explosive welding, explosive compaction of powders, and
the development of commercial and military hardware. However, this need remained
unsatisfied with proprietary rules-of-thumb and practical experience being used as
substitutes for a rigorous well-defined model. For example, in the shaped charge
community, a liners initial velocity was presumed equal to zero in acceleration models
derived both analytically and through the use of hydrocode analyses even though the initial
free-surface velocity imparted by an explosives detonation against a plate formed the basis
of many equation-of-state measurements.[1-5] Similarly, the motion of material driven by a
grazing detonation (a detonation traveling along the surface rather than impacting
perpendicular to the surface) was taken to occur along the direction of half of the angle
through which a plate was bent a result of practical experience rather than analysis.[4,6]
During development of the BRIGS methodology for explosive device design and
analysis, old rules-of-thumb were used in its acceleration and Gurney models. An
eyeballed initial free-surface velocity (sometimes called the jump-off velocity) started
76
The Influence of Geometry and Material Properties on an Explosives Gurney . . .
Czech Republic
2006
[Content]
an exponential curve rising to that calculated using 86% of an explosives Gurney Velocity
with the asymmetric-sandwich formula when the travel distance was 5-mm and linearly from
there to a published 100% Gurney Velocity at 19-mm travel. These parameters represented
the two travel distances associated with the one-inch cylinder test at approximately 2 and 6.5
gas-volume expansions.[4,7] Although this model was only roughly approximate and even
discontinuous in nature, it generally worked well enough for more than twenty years.[8]
Some questions and problems still affect the use of Gurney modeling and development
of an analytical acceleration model. What is an explosives Gurney Velocity? Why is the
Gurney Velocity inconsistent with different materials? What should the initial free-surface
velocity be for grazing in contrast to head-on (perpendicular-impact) detonation waves?
How might these affect the Gurney Velocity? Should the factor be 86% at two gas-volume
expansions? This paper presents answers to some of these questions.
2.
THEORY, DATA, AND DISCUSSION

2.1
Gurney Models and the Gurney Velocity Constant
Gurney Energy modeling is commonly used for predicting the velocity of boundary
material debris from explosive-filled devices of various geometry assuming that a fraction of
an explosives chemical detonation energy is converted into the kinetic energy of both
boundary material debris and the explosives detonation products.[9] A negligible amount of
energy is generally assumed lost to deforming and fracturing the boundary material.[10,11]
The methodology assumes that the explosive was instantaneously converted to static
homogeneous all-burned high-pressure gaseous products which can expand at uniform
density with a linear velocity gradient within a system that expands from a momentum massbalanced virtual point or virtual plane at which some detonation products have zero velocity.
The explosives chemical detonation energy continues to be converted into kinetic energy
until the boundary debris attains a steady-state velocity a velocity from which one then can
calculate a Gurney Velocity and a Gurney Energy. These Gurney constants are now
typically obtained during experiments using a standard-size copper cylinder and then used to
approximate velocities which could be imparted to other boundary materials when using the
same explosive in different geometrical arrangements.[12] However, it has long been
recognized, as shown in Table 1, that Gurney constants are not single-value constants.[13]
Table 1. Comparison of Gurney Velocities derived from experiments using steel
cylinders versus those using copper cylinders [14,15]
Steel (m/s)
Copper (m/s)
Comp. A-3 (RDX)
2416
2630
Cyclotol (75/25 cast)
2320
2790
Comp. B
2310
2700
TNT (cast)
2040
2370
Tetryl
2209
2500
Joseph Backofen
77
Czech Republic
2006
[Content]
The Gurney formulas for simple, symmetric geometry and a relationship to approximate
the Gurney Velocity (Vg or (2Eg)1/2) using the detonation rate (D) in km/sec and adiabatic
expansion constant () for an explosives gaseous detonation products are:
Vf cyl = (2Eg) 1/2 [M/C + 1/2] -1/2 (cylinder)
(1)
Vf plate = (2Eg) 1/2 [M/C + 1/3] -1/2 (symmetric sandwich)
(2)
Vf sph = (2Eg) 1/2 [M/C + 3/5] -1/2
(sphere)
(3)
(Roths formula)
(4)
(2Eg)1/2 0.605 D / [ - 1]
Where M and C represent the masses of the inert boundary material and explosive,
respectively.[9,16,17] (Cooper provides an alternative approximation that the Gurney Velocity
is about 0.337 times an explosives detonation rate [3].)
One of the problems with so-called standard tests, such as the copper cylinder test, is
that once a procedure is accepted as a standard, very little other data is produced
examining variations of materials or geometry. Figure 1 contains Gurney Velocity data for
Composition B explosive which were derived using final steady-state cylinder expansion
velocities in equation (1) showing that the data also depend upon the cylinders material and
wall thickness.[18] These data clearly show that experiments using steel cylinders with a
cylinder wall areal density (tcyl cyl, in units of g/cm2) equal to that of copper cylinders yield
a lower Gurney Velocity. The data also show that aluminum cylinders behave similar to
steel cylinders.
7.0
Aluminum in Air
6.5
Gurney Velocity (km/s)
6.0
Aluminum in Helium
5.5
5.0
Steel in Air
4.5
Copper in Air
4.0
3.5
3.0
2.5
2.0
0
Cylinder Wall Areal Density (g/cm2)
Fig 1. Gurney Velocity versus Cylinder Wall Areal Density for Comp. B Explosive
The data in Fig.1 imply that explosive energy is somehow lost when steel and aluminum
are used instead of copper or that at least more energy is lost to a process not included in
Gurney modeling.
78
1.2
Czech Republic
2006
[Content]
Gurney Velocity Trends affected by Thickness and Material
To date, fifteen published papers have been found to contain additional, useful cylinder
test data.[19] Mostly, the data were from so-called standard, scaled copper-cylinder
expansion experiments taken at the time at which the gaseous detonation products had
expanded to about seven times their initial volume. However, some data were available for
copper cylinders of greater or lesser thickness as well as for other materials. Where the data
were good enough, the final steady-state velocity was used to calculate the Gurney
Velocity (Vg) if it was not explicitly provided in the paper.
The useful Gurney Velocity data were examined to see if a pattern might emerge
revealing how either cylinder geometry or materials may have affected the experimental
results. Figure 2 presents a portion of this data where the cylinder wall areal density is
normalized by an expression for explosive areal density (Rex ex) using the cylinders inner
radius in cm and explosive density in g/cm3 along a radius through the cylinder wall. (Only
some representative data were chosen from the many standard cylinder tests available for
modern explosives and their mixtures with binders since most copper-cylinder tests have
been performed at (tcyl cyl / Rex ex) 1). Figure 2 reveals that the data suggest a trend.
0.001
0.010
Gurney Velocity / Detonation Rate (Vg / D)
Comp.B in Aluminum
0.100
1.000
1.000
Vg / D = 0.302 ( tcyl cyl / R ex ex ) (--
10.000
5 / 30 )
Comp.B in Steel
Comp.B in Copper
RDX in Steel
RDX in Copper
Octol 75/25 in Copper
LX-14 in Copper
PBX9404 in Copper
LX-04 in Copper
TATB in Copper
LX-17 in Copper
PETN in Copper
Trendline
0.100
Ratio of Cylinder Wall Mass to Explosive Mass in the Radial Direction ( tcyl cyl / Rex ex )
Fig 2. Ratio of an Explosives Gurney Velocity to Its Detonation Rate Compared to the
Normalized Cylinder Wall Areal Density
The ideal situation assumed for using Gurney modeling is rarely attained in engineering
practice. For example, premature bursting of a cylindrical or spherical shell can permit the
detonation products to escape before they can perform the same level of gas-dynamic work
that is usually found during standard copper cylinder tests. However, data exhumed from
scientific publications show in Fig.3 that similar material property and geometry effects
appear in fragmentation test data even though the cylinder wall velocities were measured at
4
Joseph Backofen
79
Czech Republic
2006
[Content]
time of cylinder fracture rather than after a fully-contained gas-dynamic volumetric

expansion to a steady-state velocity.[20-23] These data from tests involving five explosives
(TNT, Comp B, Octol, Baratol, and an unknown explosive) and five materials (aluminum,
brass, copper, steel, and lead) trend parallel to the previous trend equation. (For clarity,
Fig.3 is a semi-log plot containing only a subset of the data.)
0.40
Gurney Velocity / Detonation Rate (Vg / D)
Vg / D = 0.302 ( tcyl cyl / Rex ex ) ( -- 5 / 30 )
0.35
0.30
Comp B in Aluminum 2S
Comp B in Unk. Aluminum
Unknown in Aluminum D16
0.25
Comp B in Copper
Unknown in Copper
0.20
Unknown in Brass
Unknown in Lead
Calc. Comp B in Aluminum
0.15
0.10
1.00
10.00
Normalized Cylinder Wall Areal Density ( tcyl cyl / Rex ex )
Fig 3. Gurney Velocity Data from Measurments of Cylinder Wall Velocity at Fracture
Similar effects from material properties and geometry were found in experiments using
spherical charges. Figure 4 presents data extracted from four papers in comparison to data
from cylinder tests using comparable explosives and materials.[24-27] (Filled symbols
represent data from cylinder tests while open symbols present data from experiments
involving spheres.) If one were to disregard effects caused by the dimensions and properties
of the explosive and confining material, one might estimate that the ratios of the Gurney
Velocities to the detonation rates fall within +/- 17% of 0.350 with the spread representing
experimental error. (This is not much different from 0.337 found by Cooper from cylinder
test data representing sixteen military explosives.[3]) However, it is rather clear in Fig.4 that
the Gurney Velocities derived from the experiments using spherical charges demonstrate a
distribution and a trend similar to those trends found for cylindrical charges. (The trend for
spherical charges has been marked using 0.250 vice 0.302 in the trend formula.) Thus, these
data appear also to confirm that the relative dimensions and properties of the confining
material and explosive play a role beyond that expressed only by the mass-to-charge ratio
available in Gurney formulas.
80
Czech Republic
2006
[Content]
0.45
Comp B in Al 2S Cyl
Comp B in Al Sphere
Gurney Velocity / Detonation Rate (Vg/D)
TG 50/50 in Al Sphere
PBX9404 in Al Sphere
LX-04 in Al Sphere
0.40
RDX/TNT in Al Sphere
Comp B in Cu Cyl
PBX9404 in Cu Cyl
LX-04 in Cu Cyl
TG 50/50 in Cu Cyl
0.35
RDX/TNT in Cu Sphere
Comp B in Steel Cyl
LX-04 in Mild Steel Cyl
LX-04 in Stainless Steel Cyl
Unk Expl, Stainless St. Sphere
TG 50/50 in Steel Sphere
0.30
RDX/TNT in Zn Sphere
RDX/TNT in Pb Sphere
Unk Expl in U Sphere
Calc. Comp B in Al Cyl
Calc. Comp B in Al Sphere
0.01
0.10
0.25
1.00
Normalized Boundary Areal Density (tdevice matl / Rex ex)
Fig 4. Normalized Gurney Velocity Data for Some Explosives Detonated in Cylinders and
Spheres of Different Materials
The following equation can be used to include the trends identified in Figs.2-4 into
Gurney modeling by expanding Roths formula so that the currently known effects of
geometry and material properties can affect a final steady-state velocity calculation:
Vg / D = A (tcyl cyl / Rex ex ) B
(5)
Where:
A = 0.302 for Comp B explosive in aluminum cylinders (or for = 3 explosives),
A 0.605 / [ - 1] for explosives in copper cylinders (Roths approximation), and
B = (-5 / 30).
For spherical and plate geometry, sph, tsph and pl, tpl are substituted for cyl and tcyl with
Tex also replacing Rex for plate geometry.
It must be noted, however, that extensive experimental data must yet be collected in
order to fully identify and affirm the relationships between geometry (planar, cylindrical,
spherical, wall thickness and explosive thickness), material properties, and explosives that
affect the total propulsion imparted by an explosives detonation.
Joseph Backofen
81
Czech Republic
2006
[Content]
Brisant Propulsion the 1st Stage Propulsion Effect Producing the

Initial Free-Surface Velocity
1.3
The fact that both material properties and the normalized areal density appear to affect
detonation propulsion performance brought into consideration the initial shock effects
measured long ago during equation-of-state experiments. As previously mentioned, the
initial free-surface velocity (Vi) which is affected by shock waves during such experiments
appears to be important to the acceleration of a boundary material.
ETRi (head-on) Vi/D * (RhoP/RhoE) ^ 0.5
In ref.[5], the initial velocity imparted by the shock action of perpendicular impact of a
detonation wave with plates was found to be predictable by two simple equations. These
equations resulted from analysis of over 600 summarized data-sets representing data from
thousands of experiments involving ten materials and nineteen explosives as plotted in Fig.5.
0.7
Uranium
Plexiglass
Reasonable Charges &

Very Thin Plates
Zinc
0.6
Magnesium
Brass
0.5
Dural
Thick Plates &

Thin Charges
Copper
Dural
0.4
Reasonable Plates &

Very Long Charges
0.3
0.417 * (t/L) ^ -3/40

Aluminum
Reasonable Plates &

Reasonable Charges
0.380 * (t/L) ^ -0.30
0.2
0.001
0.01
0.1
Plate Thickness / Charge Length
Brass
10
(t/L)
Fig 5. Free-Surface Velocity Ratio (ETRi) versus the Ratio of Plate Thickness to Charge
Length for Detonation Waves Impacting Perpendicular to a Plate
As described in ref.[5], the initial effort attempted to use a conventional momentum
balance concept in order to create a two-stage propulsion model suitable for describing
coupling of detonation waves to inert boundaries as a function of the angle between the
detonation front and the boundary. However, the following Energy Transference Ratio
(ETR) was found to best represent the wide range of experimental data shown in Fig.5 in the
region within which explosives might be employed in practical devices.
ETRi = (Vi / D) (cyl / ex) 1/2 = 0.417 (tcyl / Rex) -3/ 40
(6)
(For plate geometry, pl, tpl, and Tex are used for plate density, plate thickness, and explosive
thickness, respectively.)
In contrast to the perpendicular impact of a detonation wave, there is a paucity of freesurface initial velocity data driven by a grazing detonation simply because of two factors: 1)
no practical applications seemed to require such free-surface velocity data, and 2) such data
82
Czech Republic
2006
[Content]
were very difficult to acquire and to interpret before reliable Fabry-Perot interferometry was
applied to detonation wave coupling experiments. Very simply, it was difficult to determine
the instantaneous free-surface velocity while material rotated through the initial bend
angle. The limited amount of data available in scientific literature, however, has been
summarized in Fig.6 [28]. As noted in ref.[28] and Fig.6, the data can be approximately
represented by equation (6) using 0.2085 (half of 0.417). This also provides a simple means
by which to estimate the initial free-surface velocity as a function of the angle between the
detonation wave and the driven boundary as it changes from a gazing to perpendicular
impact wave.
ETRi (grazing)
Vi/D (RhoP/RhoE)^0.5
0.40
0.35
Magnesium
Tantalum
0.30
0.189 (tp/Te) ^-3/8

0.25
0.2085 (tp/Te) ^-3/40

0.20
0.15
Aluminum
Uranium, Lead
Plexiglas, Antimony
Copper
0.10
Iron
0.05
0.00
0.001
0.01
0.1
10
Plate Thickness / Explosive Thickness (tp/Te)
Fig 6. Free-Surface Velocity Ratio (ETRi) versus the Ratio of Thickness to Charge
Thickness for Grazing Detonation Waves Running Along a Plate or in a Cylinder
The initial free-surface velocity equation can also be used to analyze the effect that the
initial acceleration might play in overall detonation-driven propulsion. Substituting
equations (5) and (6) into (1) and (2) provides the following ratios between the initial
velocity (Vi) and the final steady-state velocity (Vf) for a = 2.75 explosive:
(Vi/Vf) cyl = (0.2085 / 0.3457) (tcyl /Rex) -3/ 40 [(tcyl /Rex)2 + 2 (tcyl /Rex) + 0.5 (ex / cyl)] 1/2
(Vi/Vf) plate = (0.2085 / 0.3457) (tpl /Tex)
-3/ 40
[2 (tpl /Tex) + 0.333 (ex / pl)]
1/2
(7)
(8)
Equation (6) is also used in Fig.7 using 0.2085 in order to estimate the initial velocity for
some cylinder tests for which only final velocity data were available. Since aluminum was
used in most of the experiments from which equation (6) was derived, it was used to
calculate Vinitial for the aluminum cylinders. Values calculated for steel cylinders were
simply divided by 2 to approximate the decreased initial velocities observed in currently
available experiment data. As can be seen in Fig.7, this simple approximation for the energy
lost to the iron to phase transition an effect well known since the 1950s produces
data which nearly overlays some experiment data. (See Table 2 for some other materials
8
Joseph Backofen
83
Czech Republic
2006
[Content]
commonly used in explosive-driven devices and susceptible to phase transitions at pressures

less than an explosives Chapman-Jouguet detonation pressure.)
Table 2. Phase Transition Pressure for Some Materials Commonly Used with
Explosive Charges
Transition Pressure (GPa)
Aluminum
20.5
Carbon (pressed graphite)
23
Iron (0.2 wt% Carbon)
14.7
Iron (0.5 wt% Carbon)
13
Titanium
9.4
Zirconium
23
0.9
Copper
0.8
Iron
0.7
Magnesium
Tantalum
V initial / V final
0.6
Fe (est.Vi/2)
Al (est.Vi)
0.5
Cu Cyl Calc.
Fe Cyl Calc.1.9
0.4
Al Cyl Calc.
Cu (BN03-5)
0.3
Steel (BN03-5)
Ta (BN03-5)
0.2
Al (BN03-5)
0.1
0
0.1
0.2
0.3
0.4
0.5
Ratio of Cylinder Wall to Explosive Radius ( tcyl / Rex ) or

Plate Thickness to Sandwich Explosive Thickness ( tp / Tex )
0.6
0.7
= 2.75 for Calc.
Fig 7. Vinitial / Vfinal Data for Cylinders and Plates of Various Inert Materials Subjected to
Grazing Detonation by Various Explosives
Fig.7 reveals that the initial free-surface velocity is not only a major portion of the total
final steady-state velocity but also their ratio (Vi/Vf ) is affected by the properties and
thickness of the boundary materials. As shown in Fig.7, approximately 50% of the final
velocity (representing approximately 25% of the final kinetic energy) is imparted to the
cylinder wall by the initial brisant process during experiments having the geometry ratios
employed during typical standardized cylinder tests the regions of 0.1 and 0.2 thickness
ratio. Other authors have also recorded similar observations for plates accelerated by
84
Czech Republic
2006
[Content]
grazing detonations. [29-31] It remains for future research, however, to show whether this is
the reason why a shaped charge liners motion follows a trajectory at approximately one half
of the total bend angle (Taylor angle).
Gas-Dynamic Propulsion the 2nd Propulsion Process Responsible

for Most of the Final Kinetic Energy
1.4
The findings for the ratio Vi/Vf suggest that propulsion during the gas-dynamic (gaspush) stage should also be modeled using Vf to normalize the instantaneous velocity (Vx,
Vt, or Ver) respectively coordinated to distance, time or the explosives gaseous product
expansion ratio.[32] Figure 8 presents data for sixteen (ideal and non-ideal) explosives
demonstrating the similarity by which copper cylinders are propelled when Ver is
normalized by Vf as an explosives gaseous products expand from 1.0 < Expansion Ratio
(ExR) < 13. However, the similarity might actually be much better than shown in Fig.8
since its data are normalized using published Vf values values mostly measured at arbitrary
final expansion ratios (ExRf), such as 6.5 or 7. Such published Vf do not necessarily
represent the maximum velocity that could have been reached during experiments at larger
expansion ratios.
1.05
Velocity / Final Velocity ("Steady-State")
0.95
0.85
0.75
TNT
Comp B
Cyclotol
PETN
HMX
LX-04
LX-07
LX-09
LX-10
LX-11
LX-14
H-6
H-6 fine
9404
9011
Nitromethane
0.5_ExpR10
0.45_ExpR10
0.65
0.55
0.45
0.39_ExpR10
0.35
1
11
13
Expansion Ratio (Volume / Initial Volume)
Fig 8. Normalized Velocity Data for Sixteen Explosives Driving Copper Cylinders
The data shown in Fig.8 demonstrate a rapid velocity rise from the initial free-surface
velocity during 1< ExR < about 1.5. After this, the velocity rises at a less steep slope until
eventually reaching Vf at ExRf. This implies that two terms should be chosen to represent
such velocity data:
Ver / Vf = V1 / Vf + V2 / Vf
V1 / Vf = (Vi / Vf) {(e -- ExR e ExR ^ 3 ) + ([ExR / ExRf ] -- 0.5 / [1 / ExRf ] -- 0.5 )}
10
Joseph Backofen
(9)
(10)
85
Czech Republic
2006
[Content]
V2 / Vf = [1 {(Vi / Vf) / (1 / ExRf) -- 0.5 }](ExR / ExRf) -- 0.333 [log (ExR) / log (ExRf)] (11)
The first term (V1/Vf) appears to represent a lingering effect from the brisant 1st
propulsion stage imparting the initial free-surface velocity; and the second term (V2/Vf)
appears to represent acceleration by detonation product gas pressure during their gaseous
expansion. Some experimental data exhumed from published literature appear to support
that typical ideal explosives continue pushing out to expansions beyond 10 and nonideal explosives towards 14 as in the case of some aluminized explosives. (See also Fig.9.)
This means that most data in textbooks and technical papers probably do not provide a sound
basis for accurately modeling an explosives gas-dynamic propulsion since data are almost
always only measured up to or normalized to ExRf = 6.5. (In other words, cylinder wall
movement was normalized to travel of 19 mm during a standard 1-inch copper cylinder
test.) Nevertheless, equation (9) offers a means to estimate velocity versus time, expansion
ratio or boundary movement by means of direct measurement of Vi and Vf at ExRf. (The
former velocity can be measured using a Fabry-Perot interferometer and the latter velocity
using a streak camera.)
Velocity / Final Velocity ("Steady-State")
1.05
0.95
0.85
Nitromethane
0.75
Pourex
Aquanal
0.65
Unigel
ANFO
0.55
0.39_ExpR10
0.45_ExpR 8
0.45
0.35
1
11
13
Expansion Ratio (Volume / Initial Volume)
Fig 9. Normalized Velocity Data for Five Commercial Explosives Driving Copper
Cylinders
Figure 10 shows that the time derivative of equation (9) creates a smoothly varying
equation comparable to a JWL equation-of-state (EoS).[33] As shown in Fig.10, both are
found to be comparable for military explosives such as LX-07. However, there are distinct
differences. Firstly, the new equation only represents the gas-push 2nd stage propulsion
cycle whereas the JWL model was formed to describe the entire acceleration provided by the
explosive. Thus, the JWL EoS must initially employ a high gas pressure in order to play
catch-up since the JWL method assumes an initial velocity of zero rather than recognizing
86
Czech Republic
2006
[Content]
an initial free-surface velocity. The JWL EoS pressure also drops more quickly to a
negligible pressure well before the time ExR reaches 6.5 whereas the time derivative of
equation (9) continues a decaying pressure until the expansion ratio at which the final
steady-state velocity is achieved in experiments.
Pressure (JWL Along Isentrope) Compared To Acceleration

Equation
0.14
0.12
Time Derivative of Eqn 4; Vi / Vf = 0.52; ExRf = 6.5

0.10
JWL EoS for LX-07 Explosive

JWL EoS for Comp B Explosive
0.08
Time Derivative of Eqn 4; Vi / Vf = 0.6; ExRf = 6.5

0.06
0.04
0.02
0.00
1.0
2.0
3.0
4.0
5.0
6.0
7.0
Expansion Ratio
Fig 10. Comparison of JWL Calculated Pressures for Two Explosives to the Time
Derivative of Equation (9) for These Explosives Driving Copper Cylinders
3.
SUMMARY AND CONCLUSIONS
This paper has shown that experimental data reveal that geometry and material
properties affect an explosives performance beyond the mass-to-charge ratio used in Gurney
modeling. As a result, these factors affect the measurement of an explosives Gurney
Velocity and its corresponding Gurney Energy parameters used not only in analytical
formula Gurney modeling but also in the formulation of gas-dynamic equations-of-state.
This paper has also provided a new model a two-stage detonation propulsion model
capable of describing a materials velocity beginning from its initial free-surface velocity,
during its gas-dynamic-driven acceleration, and at its final steady-state velocity. However,
the experiments and their data upon which this model and previous Gurney modeling in
scientific literature are based are not well described, are not statistically significant in and by
themselves, and contain measurement limitations and errors that affect the accuracy of work
based on them. Nevertheless, the new model and its equations can be used to guide
experiments to create a consistent, robust database from which future Gurney Velocity and
12
Joseph Backofen
87
Czech Republic
2006
[Content]
Gurney Energy data can be extracted. Such work to clarify, to fully understand, and to
accurately describe how a detonating explosive actually provides propulsion needs to:
Measure the brisant 1st stage propulsions initial free-surface velocity,
Thoroughly measure the gas-push 2nd stage propulsion out to large expansion ratios,
Conduct the experiments over a wide range of explosives, materials, geometry

(planar, cylindrical, spherical), and normalized areal densities.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
88
G. RANDERS-PEHRSON: An Improved Equation for Calculating Fragment Projection Angles,

Proc. of 2nd Int. Symp. Ballistics, 9-11 March, Daytona Beach, FL, American Defense
Preparedness Assoc., Washington, D.C., 1976
P.C. CHOU, J. CARLEONE, E. HIRSCH, W.J. FLIS, R.D. CICCARELLI: Improved Formulas for
Velocity, Acceleration, and Projection Angles, Proc. Sixth Int. Symp. Ballistics, J.E. Backofen,
Jr., Ed., 27-29 Oct., Orlando, FL., American Defense Preparedness Assoc., pp. 286-296, 1981
P.W. COOPER: Explosives Engineering, VCH Publishers, Inc., New York, 1996
J.A. ZUKAS, W.P. WALTERS, Eds: Explosive Effects and Applications, Springer-Verlag, New
York, 1998
J.E. BACKOFEN, C.A. WEICKERT: The Effects of Plate Thickness and Explosive Properties on
Projection from the End of a Charge, Proc. 16th Int. Symp. Ballistics, 23-28 Sept., San
Francisco, CA, American Defense Preparedness Assoc., Rosslyn, VA, pp. 641-650, 1996
W.P. WALTERS, J.A. ZUKAS: Fundamentals of Shaped Charges, John Wiley & Sons, New
York, 1989
M. SUCESKA: Test Methods for Explosives, Springer-Verlag, New York, 1995
J.E. BACKOFEN: The Use of Analytical Computer Models in Shaped Charge Design,
Propellants, Explosives, Pyrotechnics, 18, pp. 247-254, 1993
J.E. KENNEDY: Explosive Output for Driving Metal, (Behavior and Utilization of Explosives in
Engineering Design, Proc. 12th Annual Symp. of the ASME (New Mexico Section), L.
Davison, J.E. Kennedy, F. Coffey, Eds.), Albuquerque, NM, pp. 109-124, 1972
J.E. REAUGH, P.C. SOUERS: A Constant-Density Gurney Approach to the Cylinder Test,
Propellants, Explosives, Pyrotechnics, 29(2), pp. 124-128, 2004
W.J. STRONGE, MA XIAOQING, ZHAO LANTING: Fragmentation of Explosively Expanded Steel
Cylinders, Int. J. Mech. Sci., 31(11/12), pp. 811-823, 1989
J.E. KENNEDY: The Gurney Model of Explosive Output for Driving Metal, (Explosives Effects
and Applications, J.A. Zukas, W.P. Walters, Eds.), Springer-Verlag, New York, pp.221-257,
1998
D.R. KENNEDY: The Elusive (2E) , 21st Annual Bomb & Warhead Section Meeting,
American Ordnance Assoc., Picatinny Arsenal, NJ, 22 Oct. 1969
D.R. PRICE: Dependence of Damage Effects Upon Detonation Parameters of Organic High
Explosives, Chem. Rev., pp. 801-825, 1959
B.M. DOBRATZ, P.C. CRAWFORD: LLNL Explosives Handbook: Properties of Chemical
Explosives and Explosive Simulants, UCRL-52997, Lawrence Livermore National Lab., 31 Jan.
1985
J. ROTH: Correlation of the Empirical Gurney Constant with Detonation Parameters of the
Driver Explosive, Proc. Int. Symp. Pyrotechnics and Explosives, 12-15 Oct., Beijing, China,
China Academic Publishers, pp. 629-634, 1987
J. ROTH: The Adiabatic Exponent a Key to the Numerical Evaluation of the CJ State, Proc.
24th Int. Pyrotechnics Seminar, Chicago, pp. 469-481, 1998
Czech Republic
2006
[Content]
[18] J.E. BACKOFEN, C.A. WEICKERT: Effect of an Inert Materials Thickness and Properties on the
Ratio of Energies Imparted by a Detonations 1st and 2nd Propulsion Stages, (Shock
Compression of Condensed Matter-2001, M.D. Furnish, N.N. Thadhani, Y. Horie, Eds.), Amer.
Inst. Physics, pp. 954-957, 2002
[19] J.E. BACKOFEN: The Effects of Cylinder Geometry and Material on Gurney Velocities and GasPush Gurney Velocities Measured During Cylinder Test Experiments, BRIGS Report 02-1,
BRIGS Co., Oak Hill, VA, 24 Feb. 2002
[20] J.E. BACKOFEN: Confirmation of the Effects of Cylinder Wall Thickness and Material
Properties on Measurement of an Explosives Gurney Velocity, BRIGS Note 03-1, BRIGS Co.,
Oak Hill, VA, 9 Feb. 2003
[21] J.E. BACKOFEN: Additional Information on How a Cylinders Wall Thickness and Material
Properties Can Affect the Measurement of an Explosives Gurney Velocity, BRIGS Note 03-2,
BRIGS Co., Oak Hill, VA, 23 March 2003
[22] M. SINGH, H.R. SUNEJA, M.S. BOLA, S. PRAKASH: Dynamic Tensile Deformation and
Fracture of Metal Cylinders at High Rates of Strain, Int. J. Impact Engng., 27(9), Oct., pp. 939954, 2002
[23] A.I. VOROBEV, M.S. GAINULLIN, G.V. ZLYGOSTEV, A.P. RYBAKOV: Experimental
Investigation of the Motion of Cylindrical Shells Under the Action of the Products of an
Explosion in a Cavity, J. Appl. Mech. Tech. Phys., pp. 872-877, 1976 (trans. Zh. Prik. Mekh.
Tekh. Fiz., No.6, Nov.-Dec., pp. 165-169, 1974)
[24] P.M.B. SLATE, M.J.W. BILLINGS, P.J.A. FULLER: The Rupture Behaviour of Metals at High
Strain Rates, J. Inst. Metals, 95, pp. 244-251, 1967
[25] E.L. LEE, H.C. HORNIG, J.W. KURY: Adiabatic Expansion of High Explosive Detonation
Products, UCRL-50422, TID-4500, UC-4, Chem., Lawrence Radiation Lab., Univ. California,
Livermore, CA, 2 May 1968
[26] F. OLIVE, A. NICAUD, J. MARILLEAU, R. LOICHOT: Rupture Behaviour of Metals in Explosive
Expansion of Shells, Inst. Phys. Conf. Series No.47: Ch.2, pp. 243-251; and M. STELLY:
Metallurgical Aspects of the Dynamic Expansion of Shells, ibid., pp. 252-253, (Mechanical
Properties at High Rates of Strain, J. Harding, Ed.), Inst. Phys., Bristol and London, 1979
[27] A.I. IVANOV, M.A. SYRAININ, A.G. FEDORENKO, A.P. TSOI: Fragmentation of Spherical Shells
Under Blast Loading, Strength of Materials, 33(2), pp. 150-156, 2001 (trans. Prob. Proch.,
No.2, Mar.-Apr., pp. 78-87, 2001).
[28] J.E. BACKOFEN, C. WEICKERT: Initial Free-Surface Velocities Driven by Grazing Detonation
Waves, (Shock Compression of Condensed Matter-1999, M.D. Furnish, L.C. Chhabildas, R.S.
Hixon, Eds.), Amer. Inst. Physics, pp. 919-922, 2000
[29] O.B. DRENNOV, A.I. MIKHAILOV: Initial Stage in the Acceleration of Thin Plates in the
Grazing Detonation Mode of a High Explosive, Fiz. Gor. Vzry. 15(4), July-Aug., pp. 143-146,
1979, (trans. Comb. Expl. Shock Waves, 15(4), pp. 539-542, 1980)
[30] V.V. KISELEV: Estimation of the Properties of Metal Plates Propelled by the Sliding
Detonation of Charges of Condensed Explosives in the Initial Phase of the Process, Fiz. Gor.
Vzry. 31(1), Jan.-Feb., pp. 138-142, 1995, (trans. Comb. Expl. Shock Waves, 31(1), pp. 134137, 1995)
[31] A.A. DERIBAS: Acceleration of Metal Plates by a Tangential Detonation Wave, Prik. Mekh.
Tekh. Fiz., 41(5), Sept.-Oct., pp. 68-74, 2000, (trans. J. Appl. Mech. Tech. Phys., 41(5), pp.
824-830, 2000)
[32] J.E. BACKOFEN: Modeling a Materials Instantaneous Velocity during Acceleration Driven by
a Detonations Gas-Push, in publication (Shock Compression of Condensed Matter-2005),2006
[33] H. HORNBERG: Determination of Fume State Parameters from Expansion Measurements of
Metal Tubes, Propellants, Explosives, Pyrotechnics, 11, pp. 23-31, 1986
14
Joseph Backofen
89
Czech Republic
2006
[Content]
COMPUTER SIMULATION OF THERMAL DECOMPOSITION

MECHANISM FOR COMPOUNDS WITH NITROGUANIDINE
FRAGMENT
Eugeniya A. Bakhmatova*, Vyacheslav L. Korolev*, Aleksey A. Porollo**,
and Tatyana S. Pivina*
*
Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences,

Leninsky Prospect 47, Moscow 119991, RUSSIA
Fax: +7 (495) 135 5328, E-mail: tsp@ioc.ac.ru
**
Biomedical informatics, Childrens Hospital Research Foundation,

3333 Burnet Av, Cincinnati, OH 45229, USA
Abstract:
The application efficiency of energetic compounds depends on the set of their
characteristics. The most important of them is a thermal stability, which is connected
with the thermal decomposition mechanism of compounds.
Nitroguanidine and his analogs have been of interest as an example of energetic
compounds. However, currently there is no general view for the thermolysis of
nitroguanidine analogs. Having labile hydrogen atoms, theoretically, nitroguanidine
and his analogs may exist in different tautomeric forms. Meanwhile, there are no
experimental evidences if this or other tautomeric form domination during a
decomposition process.
In order to fill this gap, the simulation of mechanism of all nitroguanidine tautomers
and 3-nitramine-1,2,4-triazole thermal decomposition was carried out. Subsequent
evaluation of different tautomeric forms in terms of thermodynamic stability and
activation energy for initial steps of their decomposition reactions has been conducted
using DFT approach (B3LYP/6-31G*). Thermo-chemical preferences of some
decomposition pathways have been determined.
Keywords: nitroguanidine, 1,2,4-triazole, simulation of mechanism, decomposition

reaction.
1.
INTRODUCTION
As is known, experimental investigations of multi-stepped thermolysis processes for

energetic materials are complicated. That is why in a set of works a theoretical approach to
simulate mechanisms of organic compounds thermolysis processes has been proposed /1-5/.
In the present work, this approach has been used for nitroguanidine (1) and 3-nitramino1,2,4-triazole (2) thermolysis channels modeling (the latter compound was considered
because it contains nitroguanidine in the nitrimine form as a fragment of the structure).
90
Computer Simulation of Thermal Decomposition Mechanism for Compounds . . .
H2 N
H2 N
Czech Republic
N NO2
N
NH
2.
[Content]
NH NO2
2006
METHODS OF INVESTIGATIONS
Based on analysis of experimental data on mechanisms of thermal decomposition

reactions for nitroguanidine and 1,2,4-triazoles, a generator of hypothesis for the
mechanisms of the compounds thermolysis has been elaborated. At the heart of the
methodology was the representation of decomposition processes as Recombination Reaction
Nets (RRN). The approach was realized in the CASB computer program /1-2/ and allows
one to get the whole spectrum of reactions occurring in the course of compounds thermal
decomposition. Having probable mechanism, to evaluate the thermochemical preference of
these or other pathways of decay, quantum-chemical calculations of the activation energy for
reactions at the first step of decomposition were carried out. All calculations were
accomplished by the method of density functional (DFT) using hybrid potential B3LYP and
standard valence-split basic 6-31G* /6/. Calculations were executed using GAUSSIAN 98
/7/ in the Computational Centre of Zelinsky Institute of Organic Chemistry, Russian
Academy of Sciences.
3.
RESULTS ANF DISCUSSION
For nitroguanidine (1) it is possible ten tautomeric forms and rotational isomers.
Experimental evidences of one or other tautomer prepotency under thermolysis conditions
are lacking. Apparently, the opportunity of one or another tautomer or their mixture
existence depends on particular conditions. That is why the simulation of nitroguanidine
thermolysis mechanism was carried out for all (1-1) tautomeric forms.
As the measure of isomer stability the calculated tautomerisation energy /9/ of
different forms we used. The results (Table 1) show that in gas phase the tautomer (1b) is the
most thermodynamic stable: its total energy is the least relatively to all others forms.
Table 1.
Energy characteristics of nitroguanidine tautomers (1a-c).
Total
energy
(E), ..
Compounds
Tautomerization
energy,
kcal/mol
Total
energy
(E), ..
Tautomerization
energy,
kcal/mol
23.4
1c4 O
409.739156
Compounds
NH2H
HN
HN
C
H2 N
H
N
1a
NO2
409.75874
11.09
O
N
Eugeniya Bakhmatova, Vyacheslav Korolev, Aleksey Porollo, Tatyana Pivina
91
Czech Republic
2006
[Content]
NH
H2 N
1b
H2 N
NO2
H2N
0.00
O
N
1c5
409.77642
N
O
20.7
409.743446
0
NH
NH2
HN
O
N
1c1
N
O
H2N
20.8
O
N
409.74323
1c6
NH2
O
1c2
H2N
20.7
O
N
1c7
409.74336
23.9
409.738276
NH H
O
1c3
NH2
N
409.743129
NH
HN
HN
20.9
H2N
24.0
1c8
409.73813
O
N
25.3
409.736059
The energy of tautomerization was calculated by formula [(E + ZPE) (E +

ZPE)]627.5 kcal/mol /9/, where ZPE is a correction for the energy of zero-point
oscillations, x the data for the tautomer (1b) and y the data for any other isomers.
Computer generation of possible pathways for (1b) decomposition gave three

hypotheses of the initial stage (Fig. 1): homolytic cleavage of nitro-group (channel 1), NH2
group breakage (channel 2), and nitro-nitrite rearrangement (channel 3).
H 2N
H 2N
NO2
3
H2N
H2N
Fig. 1.
92
NO2
H2 N
NO2
NH2
H2 N
NO
H2 N
The initial stages of nitroguanidine (1b) thermal decomposition.
Czech Republic
2006
[Content]
The results of activation energy (Ea) calculations of the presented decomposition

reactions shown the thermochemical preference of (1b) destruction along the channel 1 with
= 46.4 kcal/mol. The radical NH2 cleavage energetic input (channel 2) is higher (=56.1
kcal/mol). Unfortunately, we failed to localize the nitro-nitrite rearrangement transition state
(channel 3). However, according to /10/ for N-nitro derivatives this rearrangement is
thermochemicaly less favorable than homolytic cleavage of nitro-group.
At the Fig. 2 the fragment of consequent stages for nitroguanidine (1b)
decomposition along the channel 1 is presented as an illustration of generated net
intermediates and of the order of their transformations.
H2N
NO2
H2N
H2N
H2N
H2 N
+HONO
C
NH2
O
H2N
+M
+M
ONO
HN
HNO O
+M
H
N
H2O
NH2
HN
C
H
N
N2
NH
NH2
O O
N2
H2 O
CO
NH3
+HONO
H2O
H2O
NH2
NH2 N
+M
NH3
NH2
CO2
CO2
HNO
-NO
H2O
+ONO
CO
+NO2
N2 O
NH2
+NO
CO2
N2O
+NO2-NO
O
+M
N2
H2 N
OH
+M
H2O
H2O
+M
+NO2
N2O
CHO
+R
CH2O
+N O
CO2
H2N
N
OH
H2 N
+M
NH2
NH3
HCN
+M
+M
+2NO
H2O HNO NO2 N2O
NH2
H2N
+HONO
N2O OH
NO
HNO O
+ONO
O
HN
H2O
NH2
CN
NO
N
NO
+M
NH2
+H2O
N2O
H2 O N 2
C
O
+NO2
+M
NH2
HN
NH2
NH3
NH
HONO
OH
NH
H2 N
+M
ONO
H2N
+R
HN
Condensed residuum
N2
Fig. 2. The fragment of consequent stages for nitroguanidine (1b, channel 1)

thermal decomposition.
The generated mechanism of (1b) thermolysis allows one to describe the whole
spectrum of reactions occurring in the course of thermal decomposition and of the
experimentally confirmed final products /11/ of thermolysis: N2O, NH3, CO2, HNCO,
cyanamide, dicyanamide, urea, cyanuric acid, melamine.
For the second compound that we present in the paper (C-nitramino-1,2,4-triazole (2))
30 tautomeric forms and rotational isomers are possible. The same as in the case of
nitroguanidine, for the compound (2) there is no evidence of thermochemical preferences of
this or other form. The results of our calculations (DFT/B3LYP/6-31G*) done for all
isomers (Table 2) show that in gas phase the tautomer (2a) is the most thermodynamic stable
one.
4
93
N
H2
Czech Republic
2006
[Content]
Table 2. Energy characteristics of C-nitramino-1,2,4-triazole tautomers (2a-h).

Compounds
Total
Energy
(E), ..
Tautomerization
energy, kcal/mol
Compounds
N NH
N NH
-502.0856
0.0
N N
NH
-502.0854
N
H
0.3
NO 2
2b
N
O
NH
N
H
NO2
-502.0748
9.5
-502,0597
15.6
-502,0639
13.1
-502,0637
13.3
-502,0662
11.6
-502,0593
15.9
-502,0637
13.0
-502,0672
10.6
6.6
O
H
2c
HN N
-502.0835
1.4
N
O
2d1
N
O
N
O
-502.0842
N O
H
0.9
2h4
2d2
O
N
HN NH
N
2h3
N
O
HN N
N NH
N
H
2h2
O
NH
HN N
O
-502.0581
16.8
2e1
H O
N
N O
2h5
HN NH
HN N
N
N
O
-502.0472
23.8
O H
N
N O
2h6
2e2
HN N
N NO2
NH
-502.0699
O
HN N
HC
+
NH N
9.0
2h1
N N
-502,0711
HN N
NH N
N
H
2g7
2a
Tautomerization
energy, kcal/mol
NO 2
NH
Total
Energy
(E), ..
-502.0460
24.5
O
N NO H
2h7
2f1
NH N
NH
+
2f2
94
N N
N
NO2
-502.0602
15.5
N
H
H O
N
N O
2i1

N NH
Czech Republic
O H
N
N O
N N
-502,0771
N
H O
5.0
N
H
2006
[Content]
-502,0531
19.7
-502,0540
18.9
-502,0559
17.8
-502,0604
15.1
-502,0611
14.7
-502,0523
19.8
2i2
2g1
N NH
N N
N
O O
H
-502,0637
13.3
NN O
N O
N N
H
H
2i3
2g2
N NH
O
N H
O
NN
N
N
O
-502,0657
12.1
N
H
2i4
2g3
N NH
N N
N
-502,0672
N O
O
11.3
2i5
O H
N N
N N
-502,0635
13.3
N
H
2g5
N NH
N
N
O
N O
2i6
N N
O
N N
H
2g6
N O
H
O
2g4
N NH
N
H
-502,0716
8.7
N
H
N
N O
O
H
2i7
Therefore in the paper we present the result of simulation of thermal decomposition

mechanism for tautomer (2). Generation of possible pathways of (2) decomposition gave
three probable channels of the initial stage (Fig. 3): homolytic cleavage of nitro-group
(channel 1), N-N bond breakage in heterocycle (channel 2), and C-N out-of-cycle bond
breakage (channel 3).
In nitro-1,2,4-triazoles /12/ it takes 23.7-34.3 kcal/mol more energy for heteroaromatic
C-N bond breakage than for transannular N-N bond breakage (channel 2). According to our
calculations for (2) tautomer at the initial step the most benefit is the C-NO2 homolysis
(channel 1). This mechanism is estimated to have the activation energy of 27.9 kcal/mol. The
initial steps of nitro-1,2,4-triazole decomposition processes of other channels (2 and 3) are
almost twice as many (= 68.1 and 87.1 kcal/mol, correspondingly) compared with the
channel 1 process.
95
Czech Republic
2006
[Content]
N NH
. .
N NH
N HN
N
N
H
N
H
II
NO2
NO2
III
NH
N NH
ONO
+M
+M
OH
.
N
N
H2N
+M
H2N
H2N
NH2
.C
N2O H2O
+M
NH3
CO2
N
N
+R
HN
H2 N
NH3
O
HN
NH3
NH2
CN
+ONO
N N
OH N
NO
N
N NH
NO2
+NH2
N NH
NH
+CN
N N
O
N
+M
NH2
ONO HCN
+HONO
NO
CH N N OH
+M
N NH
+OH
N NH
OH
N2O
+M
H2O
Condensed residuum
CHO
+NO
O
NH
NH2
HNO
H
N
NH2
NH2
+M
+NO2
H2O
+M
+M
H2O
+NO
NH2
NH2
+M
H2NNO2
.
NH2
N2
N NH
H2N
NO2
N NH
+NO2
H2N
HNO
.NH
+R
H2N
N2O
+M
H2O
HN
+2 NO
NO
+M
NH2
OH
HONO
N NH
N2O
+NO2
CO2
+R
CO
N2
+NO2
HNO
+NO
-NO
N2O
CO2
N2
OH
O
+M
+M
H2O
Fig. 3. Consequent stages for 3-nitramino-1,2,4-triazole (2a) thermal decomposition.

The computer generation of the possible pathways of (2) thermal decomposition (Fig.
3) provided a wide number of reactions and the final decomposition products. It is
significant that during HS-thermolysis of 3-nitramino-1,2,4-triazole /13/ only NO, NO2,
HCN, CO, N2O have been detected experimentally. However, in this study we shown (Fig.
3) the possibility of formation for a set of some other products being experimentally not
found: H2O, N2, O2, CO2, HNCO, ammonia, cyanamide, dicyanamide, dicyane, urea,
96
Czech Republic
2006
[Content]
cyanuric acid, melamine, 3(5)-amino-1,2,4-triazole, 4-amino-1,2,4-triazole, 1,2,4-triazole,

1,2,4-triazolon, 3(5)-nitro-1,2,4-triazole, 3(5)-nitrozo-1,2,4-triazole.
4.
CONCLUSION
As the result of nitrimine form for nitroguanidine and of 3-nitramino-1,2,4triazole thermal decomposition mechanisms simulation the comprehensive spectrum of
possible pathways for their thermal decay has been suggested. Based on the suggested
hypotheses on the mechanisms of thermal decomposition the estimation of thermo-chemical
preferences of decomposition pathways have been carried out. The data obtained can be used
to propose the thermal stability of compounds and for understanding of experimental thermal
decomposition kinetic investigations.
REFERECES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
POROLLO A.A., LUSHNIKOV D.E., PIVINA T.S., IVSHIN V.P. J. Mol. Struct.
(TEOCHEM), Vol. 391, 1997, 117.
POROLLO A.A., LUSHNIKOV D.E., PIVINA T.S., IVSHIN V.P., ZEFIROV N.S., Izvestiya
Akademii nauk, 48, 1845, 1999 (in Russian).
POROLLO A., PETUKHOVA T., IVSHIN V., PIVINA T., and LUSHNIKOV D. Azetidine
and its mono-, di- and tri-nitro substituted derivatives: computer modeling of decomposition
reactions, in Proceedings of the 30th International Annual Conference of ICT, Karlsruhe,
Germany, p. 15 /1-3, 1999.
PIVINA T. S., POROLLO A. A., PETUKHOVA T. V., and IVSHIN V. P. Basic scheme for
computer simulation of decomposition reactions for energetic compounds, in Proceedings of
the 5th International Symposium on Special Topics in Chemical Propulsion (5-ISICP), Italy, p.
343-344, 2000.
PETUKHOVA T. V., POROLLO A. A., KOROLEV V. L., IVSHIN V. P. V. P., SURIKOVA
Y. N., and PIVINA T. S. Computer Modeling of Decomposition Reactions for Varies
Representatives of Nitroamines, in Proceedings of the 33rd International Annual Conference of
ICT, Karlsruhe, Germany, p. 60/1-10, 2002.
KOCH W. and HOLTHAUSEN M. C. A Chemistss Guide to Density Functional Theory,
Wiley-VCH, Weinheim, p.300, 2001.
CLARK ., Computational Chemistry (In Russian), Mir, Moscow, 383 p., 1990.
Programme Complex Gaussian-98 of the Computer Centre for Chemical Investigations,
Russian Academy of Sciences.
POLITZER P., GRICE M. E., SEMINARIO J. M., J. Quantum Chem., Vol. 61, p. 389, 1997.
MANELIS G. B., NAZIN G. M., RUBTSOV Yu. I., and STRUNIN V. A., Thermal
Decomposition and Combustion of Explosives and Powders, Nauka, Moscow, p. 223, 1996.
DAVIS T. L., ABRAMS A.-J.J., Proc. Am. Acad. Sci., Vol. 61.p. 437-457, 1926.
BAKHMATOVA E.A., PETUKHOVA T.V., KOROLEV V.L., PIVINA T.S., IVSHIN V.P., in
Proceedings of the 8th Seminar "New Trends in Research of Energetic Materials", Pardubice,
the Czech Republic, 2005.
GAO A., REINGOLD A.L., BRILL T.B. //Propellants, Explosives, Pyrotechnics, 1991, Vol.
16, 3, 97.
97
Czech Republic
2006
[Content]
THERMAL DECOMPOSITION OF SOLID ENERGETIC MATERIALS
Yurii M. Burov
Institute of Problems of Chemical Physics, Russian of Sciences, 142432
Chernogolovka, Moscow Region, Russian Federation.
E-mail: yuburov@icp.ac.ru
Abstract:
Decomposition of the organic compounds in the solid phase occurs much more slowly
than in the liquid and gas phases. The retardation effect of the crystalline lattice (REL),
which can be defined as the ratio of rate constants of decomposition in the liquid phase
to that in the solid phase (kliq/ks), is often the only factor that causes the stability of the
substance and its suitability for use after prolong storage. This situation is
characteristic of medicines, explosives, and initiators of chain processes. Experimental
determination of REL and development of methods suitable to predict the properties of
new substances are an important aspect of the general theory of the stability of organic
compounds. Development in this field is restricted because the date on ks values are
limited and the theoretical models of monomolecular reactions in the solid phase cannot
be verified. Solid phase decomposition reactions are usually accompanied by several
side processes, such as evaporation and fast decomposition of compounds in the gas
phase, melting on admixture and products autocatalysis at the early stages of the
process, effects of premelting near points of phase transitions, and others. Due to side
reactions, the observed ks values often exceed the true values by one and even two
orders of magnitude. Therefore, the purpose of this work is to obtain, firstly, a
sufficiently representative array
of correct ks and REL = kliq/ks values and, secondly, the dependence of REL on the
physical properties of the crystal and other parameters used in theoretical models. The
REL values for 22 compounds calculated using the correct ks values are presented in
this work. The REL values vary within very wide limits from 4 to 104. The reaction on
defects limits the kliq/ks ratio to 104, and the reaction in the bulk can be observed only if
REL does not exceed 100. Reaction in the bulk of the crystal requires the formation of a
cavity with a volume exceeding the activation volume s V , so that the leaving group
does not experience forces of interatomic attraction. Cavities can be formed by two
mechanisms energetically or entropically. Accordingly, two mechanisms were
observed. Both of them are in agreement with experiment.
Keywords: crystalline lattice, thermal decomposition, kinetics
1.
INTRODUCTION
Decomposition of the organic compounds in the solid phase occurs much more slowly
than in the liquid and gas phases [1-3]. The retardation effect of the crystalline lattice
(REL), which can be defined as the ratio of rate constants of decomposition in the liquid
phase to that in the solid phase (kliq/ks), is often the only factor that causes the stability of the
98
Czech Republic
2006
[Content]
substance and its suitability for use after prolong storage. This situation is characteristic of
medicines, explosives, and initiators of chain processes. Experimental determination of REL
and development of methods suitable to predict the properties of new substances are an
important aspect of the general theory of the stability of organic compounds. Development in
this field is restricted because the date on ks values are limited and the theoretical models of
monomolecular reactions in the solid phase cannot be verified.
Solid phase decomposition reactions are usually accompanied by several side processes,
such as evaporation and fast decomposition of compounds in the gas phase, melting on
admixture and products autocatalysis at the early stages of the process, effects of premelting
near points of phase transitions, and others. Due to side reactions, the observed ks values
often exceed the true values by one and even two orders of magnitude. The methods for
taking into account secondary factors [2] are used to a full extent only for several
compounds.
Therefore, the purpose of this work is to obtain, firstly, a sufficiently representative array
of correct ks and REL = kliq/ks values and, secondly, the dependence of REL on the physical
properties of the crystal and other parameters used in theoretical models.
2.
EXPERIMENTAL
Decomposition rates were measured by the manometric method. The reaction course
was monitored at conversions of 0.01 1.00%, which allowed one to avoid the influence of
autocatalytic processes and topochemical regime of the reaction. To eliminate unstable or
catalytic admixtures, the substances were purified by sublimation onto a heated support or y
recrystallization from different solvents with thorough drying. The purification was carried
until a constant decomposition rate was achieved.
To increase the sensitivity of the method, we used such loadings of the substance that the
ratio of the sample weight to the volume of the reaction vessel was 0.5 g cm-3. Thus, the
amount of the substance in the vapor phase was insignificant as compared to the weight of
the solid sample, and the reaction in vapor could be neglected even at the highest possible
difference in the rates in the gas and solid phases (four orders of magnitude) [4 6].
The ks values were calculated from the time of achievement of the degree of
decomposition of 0.1%. For calculation of the conversion, the stoichiometric coefficient of
gas release determined by the decomposition of the substance in melt was used.
3.
The REL values for 22 compounds calculated using the correct ks values are presented in
Table 1. The kinetic data using for composing Table 1 are presented in Table 2. The rate
constants obtained for the decomposition in melt (noncatalytic stages) or in inert solvents
ware used as kliq. When several data were available, the minimum ks and kliq values taken for
the calculation of REL
The REL values ware calculated at temperatures 200C lower then the melting point of
the substance when pre-melting effects do not affect the ks value. The melting points of the
substances and the volume compressibility ( ) (the parameter characterizing the force of
interatomic interaction in the crystal) are also presented in Table 1. The values were
calculated by the Rao method [19], and the group increments were borrowed from the
literature date [20].
2
Yurii Burov
99
Czech Republic
2006
[Content]
The mechanisms of initial monomolecular stages of decomposition are known for all
compounds presented in Table 1: bond cleavage of C-NO2 (2, 3, 5 - 7, 10, 13), N-NO2 (1, 14,
16, 17, 19 22), O-NO2 (18), and N-N (4) and elimination of N2 (11) via formation of the
linear transition state or elimination of N2 (9) and CO2 (12, 15) molecules through the cyclic
transition state.
Table 1.
Retardation effect of the crystalline lattice and physical properties of the

substances
Compound
kliq/ks
m.p., 0
.1010 V0,
, P-1
cm3/mol
[FC(NO2)2CH2]2NNO2
86
3,530
173,9
[F2NC(NO2)2CH2]2NNO2
102
198,0
C(NO2)3C(NO2)3
140
5,500
169,4
PhN=N-NHPh
101
[C(NO2)3CH2]2NNO2
10
95,5
1,050
199,0
F2NC(NO2)2CH2NNO2(CH2)2C(NO2)2NF2
10
120
[NF2(NO2)2CCH2N(NO2)CH2]2
20
158
25
158
1,3,5-Trinitro-2,4,6-triazidobenzene
28
131
10
1,3,5-Tris(trinitromethyl)benzene
38
113
11
N3(CH2NNO2)4CH2N3
87
177
12
HOOCCH2COOH
90
135
13
[(NO2)3CCH2N(NO2)CH2]2
92
180
1,060
254,5
14
2,4,6-(NO2)3C6H2N(NO2)Me
100
130
1,200
164,5
15
HOOC-COOH
122
189
16
1,4-Dinitro-1,4-diazacyclohexane
210
213
1,310
107,3
17
[MeC(NO2)2CH2]2NNO2
230
177
0,968
189,5
18
C(CH2ONO2)4
360
142
178,5
19
1,1,3,5,5,7-Hexanitro-1,4-diazacyclohexane
500
250
0,925
203,2
20
1,4,6,9-Tetranitro-1,4,6,9-tetraazadecalin
1000
236
21
1,3,5-Trinirro-1,3,5-triazacyclohexane
1400
202
0,809
122,0
22
1,3,5,7-Tetranitro-1,3,5,7-tetraazacyclooctane ( -modification)
8380
277
0547
155,8
.
100
Czech Republic
2006
[Content]
The majority of reactions are characterized by a positive activation volume, i. e., they
occur with a volume increase in the transition state ( V ). For compounds 9 and 11, one
can expect V values close to zero, whereas for 12 and 15, they can even be negative.
It is seen from the data in Table 1 that the REL values vary within very wide limits from
4 to 104. To explain such a broad interval of REL variation, the theory of solid phase
reactions should be considered.
Table 2. Kinetic parameters of decomposition of organic compounds in liquid and

solid states
Com-
Mediuma
pound
Solution in TNT
1
Solid phase
T / 0C
Eb
170-210
66
40.0
lgA /s-1
14.92
k c /s-1
References
1.310-
4
29
11
3.01012
2
3
4
5
6
7
8
9
10
11
12
13
14
15
4
Yurii Burov
Melt
Solid phase
Gas
Solid phase
Melt
Solid phase
Melt
Solution
Solid phase
Solution in DNB
Solid phase
Solution in TNT
Solid phase
Solution in NB
Solid phase
Solution in xylene
Solid phase
Solution in TNT
Solid phase
Solution in DNB
Solid phase
Melt
Solid phase
Solution in DNB
Solid phase
Melt
Melt
Melt
Solid phase
Gas
105-120
82
90-135
120
100-160
80
110-150
110-165
75
100
100
130-170
100-140
114
114
70-115
50-100
95
95
120-180
120-175
136-160
115
130-180
158
140-160
131-155
150-175
80-120
150-190
38.5
15.70
35.8
16.52
39.9
17.50
36.8
36.1
15.59
15.06
40.3
42.6
26.0
28.8
36.3
36.7
32.2
40.7
16.9
16.4
12.10
12.20
14.59
12.86
13.50
16.80
35.2
36.0
35.9
38.9
35.2
13.50
13.80
13.50
13.20
13.80
310-8
8.810-9
4.010-4
6.610-5
7.210-8
1.210-8
2.910-8
2.410-8
2.410-9
1.310-7
1.310-8
310-4
1.510-6
3.310-4
1.310-4
1.610-3
5.710-5
1.210-3
3.510-5
1.410-4
1.610-4
2.310-5
2.610-7
2.610-4
2.810-4
2.610-7
1.810-7
1.010-7
1.010-9
2.510-4
29
29
2
29
5
29
4
4
29
29
29
29
29
29
29
6
29
29
29
7
29
29
29
4
29
8
9
10
2
29
101
16
17
18
19
20
21
22
Solid phase
Melt
Solution in TNB
Solution in NB
Solid phase
Solution in DNB
Solid phase
Melt
Solid phase
Solution in TNT
Solid phase
Gas
Solid phase
Solid phase
Solution in DNB
Solid phase
Solution in DNB
Solid phase
120-180
135-200
120-260
225-245
185
145-170
150
120-130
170-210
230
216
216
204-234
160-200
140-190
171-215
130-180
Czech Republic
2006
38.6
37.1
37.9
37.1
13.40
13.90
14.00
12.00
39.7
15.7
40.0
39.0
38.0
15.8
12.7
14.5
50.1
39.7
39.8
44.9
37.9
18.9
14.3
11.2
16.0
9.2
2.110-6
1.510-8
8.110-8
2.010-6
9.510-9
3.310-5
2.310-7
4.710-7
1.310-9
110-2
210-5
210-3
2.110-5
3.210-4
1.410-5
110-8
3.110-3
3.710-7
[Content]
29
11
12
12
29
13
29
14
2
29
29
29
29
15
16
17
16
17
Based on the published date [1], a generalized physical model of monomolecular

reactions in the solid phase can be envisaged. A reaction can proceed in the bulk of the
crystal or on its surface and on defects of the crystalline lattice. The reaction can proceed in
the bulk of the crystal or on its surface and on defects of the crystalline lattice. The reaction
occurring on defects has the same activation energy (E) as in the liquid phase (or close to it),
but the pre-exponential factor includes the coefficient that takes into account the fraction of
molecules in disordered sites of the lattice. According to estimates [1, 24, 25] based on the
calculation of the number of molecules arranged on the network of dislocations that separate
microblocks of the crystal (the linear sizes of the microblocks are 10-3 10-5 cm), the
fraction of such reactive molecules is 0.01 1.00%. Thus, the reaction on defects limits the
kliq/ks ratio to 104, and the reaction in the bulk can be observed only if REL does not exceed
100.
Reaction in the bulk of the crystal requires the formation of a cavity with a volume
exceeding the activation volume Vs , so that the leaving group does not experience forces
of interatomic attraction. Cavities can be formed by two mechanisms energetically or
entropically.
The macroscopic approach [22, 31 - 33], considering the crystal as an elastic continuous
medium, leads to the equation
E = E s Eliq = Vs
) /(2V ) = RT ln k
2
liq
,
k s
(1)
where V0 is the molar volume of the substance, and Vs is the activation volume in the
0
solid phase. This volume is not equal to the value V . According to the published data
[21], it can be estimated as an increase in the volume of the cell occupied by a molecule due
to elongation (by 10 15%) of the cleaving bond in the transition state [27], i. e.,
approximately by 0.2 . Extension of the cell by 0.2 results in weakening of
intermolecular interactions and allows atoms of molecule to converge freely, forming cyclic
102
Czech Republic
2006
[Content]
transition states. Thus, the Vs value depends slightly on the reaction type and on the value
and sign of the true activation volume.
The pre-exponential factor of the reaction occurring in the bulk should be the same as
that for the reaction in the liquid phase [22]. In [23] added influence of solid state on the preexponential factor.
The calculation of REL by Eq. (1), assuming that the cell volume is V0 /(6.02 10 23 ) and
its expansion is 0.2 , gives values that often coincide by an order of magnitude with the
experimental values. For compounds 1, 3, 5, 13, 14, 16, 17, 19, 21, and 22, i. e., when
kinetic data for calculations were available, the calculated REL value was 8, 3, 100, 386, 10,
453, 31, 250, 400, 400, and 3040, respectively. The highest deviations from the experimental
values (see Table 1) are observed for molecules with a long chain. In this case, local motions
of, not the whole molecule, but only of its fragments containing the reaction center have a
substantial effect on the REL values.
1
10 13 / Pa mol cm 3
V0
Fig. 1. Dependence REL on volume compressibility.
As it's shown in the works [31 - 33], the main assumptions of this theory are contrary to
the principles of the thermodynamics. Nevertheless, this theory rather well describes the
experiment (see Fig. 1 and 2). Therefore, the new approach - "free volume model" was
offered in the works [31 - 33].
At temperatures is higher 2 , where - characteristic temperature, the crystal

properties are in the best way described in frameworks cell model. Within the framework of
v
the Lennard Jones cell model [29], the "free volume" f , which one can be computed by a
method atom atom potential [30]
6
Yurii Burov
103
Czech Republic
2006
[Content]
U (q ) U (q )
(2)
dq
kT
q - generalized coordinates, U (q ) - energy of crystal lattice, U (q) - interact energy of a
molecula with the neighbours in the supposition, that the adjacent moleculas take
midpositions, the integrating is carried out on volume basis cells.
v f = exp(
Probability of originating of a cavity by the size Vs
P (Vs ) = exp(
Vs
vf
Supposing, that [31 33]

k s = k liq P(Vs )
(3)
the problem of an estimation REL is reducible to finding of a free volume, which one
can be computed by Eq. (2), or can be estimated from the degree of thermal expansion
T
v f 3 (T )V0 dT ,
0
(T ) - is coefficient of thermal expansion.

The absence of a method of exact estimation of Vs restricts the possibility of using
Eq. (1) and (3) for practical calculations, although there is some doubt that considered
models adequately reflect the real pattern. Equation (1) is important because it predicts, first,
low REL, i.e., the possibility of the reaction occurring purely in the bulk of the sample, and,
second, the dependence of REL on the elastic properties of the crystal. Both predictions
agree with the experimental data. Only the nonconstant character of the Vs values disturbs
the linearity of the dependence of REL on 1
(V0 ) , which is demonstrated by the data in
Fig. 1.
The dependence of REL on the melting point of the substance, which is also a good
measure of forces of intermolecular interaction, especially for short-chain molecules [28], is
presented in Fig. 2, and points for substances with a high molecular weight lie closer to the
lower boundary in Fig. 1. For compounds with m.p. < 150 0C, the REL values are mainly
lower than 100 and, hence, we may consider that the reaction in the volume predominates in
this case as well.
For the substances with higher melting points, the reaction in the volume is changed
by the decomposition on crystal defects, which, as shown above, can limit REL by the
interval of 102 104, which is experimentally observed. Compound 3 (it is not shown in Fig.
1), which represents plastic crystals, falls out of the general dependence. For long-chain
compounds 6 and 7, according to the absolute REL value, decomposition in the bulk is also
possible, despite sufficiently high melting points.
The experimentally determined E value is low and does not usually exceed the
measurement error, which is 3 4 kcal/mol. Therefore, it is difficult to use this value to
estimate REL at different temperatures and to separate volume and local reactions. The latter
can be performed using absolute values of REL or melting points of the substances. The
experimental data in Table 1 agree with considered models and suggest that for organic
compounds with m. p. 100 0C the irreversible monomolecular reaction occurs in the ideal
104
Czech Republic
2006
[Content]
part of the crystal lattice with REL < 10, and at m. p. > 200 0C the reaction proceeds on
defects only. Within the interval m. p. = 150 200 0C, REL is equal to 102 103, and the
indicated reactions compete with each other. These conclusions give a clear semiquantitative pattern of changing REL; however, they need further verification.
Fig. 2. Dependence REL on melting point.
v
It is necessary to obtain new data on f , Vs and REL, including those for reaction
with a negative activation volume, as well as to identify directly localized reactions and to
compare the decomposition rate with the defectiveness of crystal.
4.
REFERENCES
[1]
[2]
Chemistry of Solid State, Ed. W. E. Garner, Butterworths, London, 1955.

Yu. M. Burov, G. B. Manelis, and G. M. Nasin, Dokl. Akad. Nauk SSSRR, 1984, 279, 1142
[Dokl. Chem., 1984 (Engl. Transl.)]
[3] O. N. Karpukhin, T. V. Pokholok, and V. Ya. Shlyapintokh, Vysokomol. Sedin. A, 1971, 13,
(Engl. Transl.)
[4] B. L. Korsunski, V. G. Matveev, L. D. Nazina, and G. M. Nazin, 29th Intern. Annual
Conference of ICT (June 3- - July 3, 1998), Karlsruhe, 1998, 60.
[5] N. Okawara, H. Jamazaki, and E. Imato, J Chem. Soc. Jpn., Ind. Chem. Sect., 1955, 58, 991.
[6] B. L. Korsunski, T. A. Apina, and F. I. Dubovitskii, Izv. Akad. Nauk SSSR, Ser. Khim., 1970,
2080 [Bull. Acad. Sci. USSR, Div. Chem. Sci., 1971, 20 (Engl. Transl.)].
[7] Yu. M. Burov and G. M. Nazin, 29th Intern. Annual Conference of ICT (June 3- - July 3, 1998),
Karlsruhe, 1998, 129.
[8] F. I. Dubovitskii, G. B. Manelis, and L. P. Smirnov, Zh. Fiz. Khim., 1961, 35, 521 [J. Phys.
Chem. USSR, 1961, 35 (Engl. Transl.)].
[9] F. I. Dubovitskii, Yu. I Rubtsov, and G. B. Manelis, Izv. Akad. Nauk SSSR, Otd. Khim. Nauk
[Bull. Acad. Sci. USSR, Div. Chem. Sci.], 1960, 1763 (in Russian).
[10] R. S. Stepanov, V. N. Shan'ko, I. P. Medvetskaya, and V. m. Gorodetskaya, in Khimicheskaya
fisika protsessov goreniya i vsryva. Kinetika khimicheskikh reaktsii [Chemical Physics of
8
Yurii Burov
105
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
106
Czech Republic
2006
[Content]
Combustion and Explosion. Kinetics of Chemical Reaction], OIKhF AN SSSR,

Chernogolovka, 1977, 56, (in Russian).
G. V. Sitonina, B. l. Korsunskii, N.F. Pyatakov, V. G. Shvaiko, I. Sh. Abdrakhmanov, and F. I.
Dubovtskii , Izv. Akad. Nauk SSSR, Ser. Khim., 1979, 311 [Bull. Acad. Sci. USSR, Div.
Chem. Sci., 1979, 28 (Engl. Transl.)].
B. A. Lur'e and V. N. Ivakhov, in Khimicheskaya fizika kondensirovannykh vzryvchatykh
veshcestv [Chemical Physics of Condensed Explosives], Moscow, 1979, 12 (Tr. MkhTI, 1979,
No 104) (in Russian).
B. L. Korsunskii, L.Ya Kiseleva, V. I. Ramushev, and F. I. Dubovitskii, Izv. Akad. Nauk
SSSR, Ser. Khim., 1974, 1778 [Bull. Acad. Sci. USSR, Div. Chem. Sci., 1974, 23 (Engl.
Transl.)].
K. K. Andreev and Kaidymov, , Zh. Fiz. Khim., 1961, 35, 2676 [J. Phys. Chem. USSR, 1961,
35 (Engl. Transl.)].
H. Ronqzu, Y. Zhengounan, and L. Yanjun, Thermochim. Acta, 1988, 123, 135.
Yu. Ya. Maksimov, in Teoriya vzryvchatykh veshcestv [Theory of Explosives], Vysshaya
Skola, Moscow, 1967, 73, (Tr. MkhTI, 1967, No.53) (In Russian).
Yu. M. Burov and G. M. Nazin, Khim. Fiz., 1984, 3, 1126 [Chem. Phys., 1984, 3. No. 7 (Engl.
Transl.)].
Yu. M. Burov, G. M. Manelis, and G. M. Nazin, Khim. Fiz., 1985, 4, 956 [Chem. Phys., 1985,
4. No. 7 (Engl. Transl.)].
M. R. Rao, J. Chem. Phys., 1941, 9, 682.
I. M. Voskoboinikov, A. N. Afanasenkov, and V. M. Bogomolov, Fizika Goreniya I Vzryva,
1967, 3, 585 [Phys. Comb. Explos., 1967, 3,4 (Engl. Transl.)].
G. M. Shutov, Zh. Fiz. Khim., 1967, 39, 2817 [J. Phys. Chem. USSR, 1961, 39 (Engl. Transl.)].
G. B. Manelis, in Problemy kinetiki elementarnykh khimicheskikh reaktsii [Problems of
Kinitics of Elementary Chemical Reactions], Nauka, Moscow, 1973, 93 (in Russian).
L. D. Zusman and A. B. Gel'mam, Zh. Strukt. Khim., 1980, 21, 72 [J. Struct. Chem. (USSR),
1980, 21 (Engl. Transl.)].
M. S. Belyaeva, G. K. Klimenko, L. T. Babaitseva, and N. P. Stolyarov, in Khimicheskaya
fisika protsessov goreniya i vsryva. Kinetika khimicheskikh reaktsii [Chemical Physics of
Combustion and Explosion. Kinetics of Chemical Reaction], OIKhF AN SSSR,
Chernogolovka, 1977, 47, (in Russian).
A. I. Kitaygorodskii, Moleculyarnye kristally [Molecular Crystals], Nauka< 424 pp. (in
Russian).
D. A. Charman, S. Rozak, P. B. Keegstra, R. C. Hariharan, J. J. Kaufman, and R. S. Buenker,
Int. J. Quant. Chem., 1991, 38, 541.
M. G. Gonikberg, Khimicheskoe ravnovesie i skorost' reaktsii pri vysokikh davleniyakh
[Chemical Equilibrium and Reaction Rate at High Pressures], Khimiya, Moscow, 1969, 427 pp.
(in Russian).
M. S. Westwell, M. S. Searle, D. J. Walles, and D. H. Williams, J. Am. Chem. Soc., 1995, 117,
5013.
J. Lennard-Jones, Proc. Phys. Soc., 52, 729, 1940.
A. I. Kitaigorodskii, Usp. Fiz. Nauk, 127, 391, 1979, (In Russian).
Yu. M. Burov, XIV Symposium Sovremennaya Khimicheskaya Fizika [ Modern Chem.
Physics], 18 29 Sept. 2002, Tuapse, 2002, 59, (in Russian)
Yu. M. Burov, Zh. Fiz. Khim., 1961, 35, 2676 [Russian J. Phys. Chem., 78, No. 4, 2004, 579
(Engl. Transl.)].
Yu. M. Burov, XVII Symposium Sovremennaya Khimicheskaya Fizika [ Modern Chem.
Physics], 18 29 Sept. 2005, Tuapse, 2005, 32, (in Russian).
Czech Republic
2006
[Content]
USE OF COMPRESSED GAS GUN TO INVESTIGATE RESPONSE OF

EXPLOSIVES TO A RANGE OF INITIATION STIMULI
Michael Cartwright and Andrew Lewis
Department of Environmental and Ordnance Systems,
CRANFIELD UNIVERSITY
At DEFENCE ACADEMY of the U.K.
Shrivenham
Oxfordshire, U.K.
SN6 8LA
Abstract
Low velocity Impacts are a source of explosive accidents. Several different
mechanisms may be concurrently responsible for the initiation. Studying the effects of
low velocity, comparatively long duration, impacts on energetic materials can assist
in the development of explosive safety procedures and Insensitive Munitions. Careful
design of the target configuration can eliminate some of the problems associated with
the original Steven tests and enable the various mechanisms such as friction, shear
and pinch to be isolated from other factors. In the new configuration the impact is
delivered to the explosive by a spigot driven into the target by collision from a gun
launched sabot. The spigot geometry i.e. flat or hemi-spherically nosed was varied as
was the surface roughness and diameter of the spigot. The rate of energy delivery was
varied, by varying the mass of the sabot and its velocity. The gun, 50 mm diameter,
operated in the normal atmosphere conditions, NAG. Results show that pinch may be
the most critical stimulus but also that friction is more significant than shear.
Examination of the target debris following an event indicated that for some materials
the limited confinement arising during the firing may result in a burn to detonation
condition whereas for other materials the confinement was insufficient for the DDT to
occur. Further work on the sample perfection and homogeneity are planned.
Keywords accidents, impact , initiation, explosives, energy, output
Introduction
In the past a number of costly accidents with munitions, both in peacetime and
war zones has resulted in the loss of five USN warships, four ammunition storage
depots and two Soviet submarines all with significant loss of human lives
More recent accidents during EOD work have demonstrated that there is a gap in test
procedures. Operators sent out to sample failed munitions prior to disposal have
variously taken chisels, screwdrivers and broom handles in conjunction with large (3
kg) hammers. In one accident a rocket motor ignited and left the chisel in the
operators skull. In Afghanistan a multinational team were trying to remove an
explosive warhead from a Soviet SAM missile when an ignition occurred with a
resulting deflagration or low order detonation event killing five people1. In these two
incidents the maxim elicited by F.P. Bowden 2 hitting a solid with a hammer or
1
Michael Cartwright, Andrew Lewis
107
Czech Republic
2006
[Content]
rubbing it with sandpaper is a proper experiment for a Carpenter than for a physicist
appears to have been forgotten. Nevertheless impact either by dropping munitions
from modest heights or striking munitions with implements is a major source of
accidental initiation in explosives and thus requires further study.
As a result of accident initiation the following series of safety tests for
explosive fillings using different forms of impact have been developed in the NATO
countries
Bullet Attack (STANAG 4241)
Fragment Attack (STANAG 4296)
Shaped Charge Attack(STANAG 4526)
Sympathetic reaction (STANAG 4625)
Some of these tests are particularly applied to completed weapons in their
packaging. Impact test methodologies are almost as numerous as the test houses using
them and quantitative agreement between them is rare3. JSP 333 4quotes 38 separate
sensitivity tests of which a number are impact in one form or another. Impact test
sample size ranges from the mg level of the Rotter and Drop or Fall Hammer tests to
the 200 g of the Susan and Steven tests. In the Susan test5, the explosive, contained in
a specified projectile, is fired at an armoured target using a propellant gun and in the
Steven6 test the explosive, contained in a target, is struck by a gun launched projectile.
Basic principles of the Steven test and modified assembly trialled are shown in figure
1. Computational fluid dynamic calculations on the Steven test7 predict that ignition
occurs from a variety of mechanisms and the response can be sensitive to preparation
procedures. Modelling and prediction are thus difficult.
Ignition occurs in accidents as a result of hot spot generation when the rate of
heat liberation is greater than the rate of heat loss. The major stimuli responsible for
hot spot creation are :Compressive effects - adiabatic heating of trapped gases and pore collapse
Frictional effects - both inter-crystalline and at surface boundaries
Mechanical effects - shear, tribochemical and viscous flow
Most of the accidental ignitions occur at low velocities so that the shock wave
impact mechanism is not operative. Driving a spigot into explosive samples could
produce systems in which the various ignition mechanisms could each be given
prominence and thus an estimate of the response of materials to different stimuli
could be obtained. A target vehicle, which would confine an explosive and deliver a
108
Use of Compressed Gas Gun to Investigate Response of Explosives to a Range of . . .
Czech Republic
2006
[Content]
measured quantity of energy through an impact by a projectile, launched from a gas

gun, has been designed manufactured and used.
Experimental
Equipment
The gun used was 50 mm calibre firing in air against a mounted target NAG.
Initially mechanical rupture of a diaphragm was used to fire the gun, but later gas
pressurisation behind the sabot, through a venting valve was used as shown below
Figure 1 Schematic diagram of Gas gun arrangement with Sabot loaded.

This latter had the advantage of providing more sensitive pressure
control..Three sabots were used see figure ;
Figure 2 Typical sabots used in Tests.

The simple near spherical sabot was an attempt to reduce the effect of yawing
of the projectile on impact energy and later the extended projectiles with the widely
3
109
Czech Republic
2006
[Content]
spaced O rings to obscure the venting orifices between the gas storage chamber and
the barrel were used. Both flat and round nosed projectiles had masses of 1.5 Kg.
Lighter projectiles, 0.8 Kg, consisted of front and rear steel sections bolted together
through a PVC tube which carried the obturation O rings as shown in the gun
schematic. Sabot velocity was determined from the time taken break of a series of
equally spaced wires stretched across a plastic tube through which the sabot passed
after it had exited the barrel and before striking the target
Target Design
A range of targets assemblies was used. Initially simple and modified Steven
test assemblies were used see figure 3
Figure 3 Modified Steven test Assembly showing increased confinement

Notice that the confining ring in the modified test was extended considerably over the
cover plate leaving an uncovered area slightly larger than the sabot impact area and
also down the sides of the body thus ensuring significant increase in confinement of
the explosive over the original target design. More recent tests used an assembly
shown below figure 4 .
Figure 4 Schematic Target Assembly using moving piston mounted spigot

in which a piston, with attached spigot, was driven into the explosive by impact of a
gas gun launched sabot with the top of the piston.
110
Czech Republic
2006
[Content]
Mounted within these assemblies were pellets of the explosive under test. Two pellet
sizes were used, 70 mm and 25mm diameter, by 15 mm thick Columns up to three
pellets thick could be mounted in the target assembly and held in place by PTFE or
PVC rings. Spigot penetration into the explosive was controlled by shims in front or
behind the pellets, figure .
Rear closure
Sample ring
washer
Spigot
Piston
Sample
shim
Sample holder
Mounting
plate
Figure 5 Target assembly (schematic)

This enabled the ignition mechanisms i.e. friction shear and nip sensitivities to be
investigated. Some pellets had dimples machined into one face allowing air to be
trapped between two pellets to investigating aidiabatic heating effects.
In house manufactured PVDF pressure gauges were enclosed within the
targets between pellets and also between pellets and housing surfaces. The intention
was to monitor the pressure as the projectile struck and after reaction commenced.
Experiments investigated the effect of spigot characteristics, diameter, shape and
surface texture as well as penetration and nip on sensitivity. The range layout for the
normal air gun NAG are shown in the figure
111
Czech Republic
2006
[Content]
Figure 6 Schematic range layout for 50 mm NAG

Events were monitored by high speed video camera and data recorded on fast
recording oscilloscopes. All firings were performed remotely on the COTEC ranges
on the edge of the Salisbury Plain Training Area at West Lavington
Explosive compositions investigated were a PBX formulation based on HMX
and PETN/WAX (92:8), both supplied by B.Ae. systems Glascoed, U.K. The PBX
was hot, 363 K, pressed into 70 mm diameter 13 mm thick pellets at 100 MPa. The
PETN/wax was pressed, at ambient temperatures without additional binder, into 25
mm diameter pellets 13 mm thick at 50 Mpa. The lower pressure was used for the
PETN to reduce the risk of events even though the pressing was done remotely on the
range facilities at the college. Some PETN pellets were pressed after removing the
wax by solvent washing but these showed poor mechanical strength and so were not
used beyond the preliminary stages.
112
Czech Republic
2006
[Content]
Results and Discussion

(a) PBX
Some of the observed events occurring from the modified Steven tests were
recorded at velocities >100 m s-1 by direct impact of the projectile. The induced
reactions appeared to be deflagrations and metallographic examination of the broken
edges confirmed ductile failures. In the spigot impact tests the velocities of the spigot
were less than the velocity of the impacting projectile but nevertheless events
occurred at lower sabot speeds. This is an indication of the extent of the energy lost
during the deformation of the steel cover of the Steven test. The results from a series
of tests on the PBX composition with the spigot intrusion are shown in the table 1.
Table 1 Impact trial results PBX showing spigot configuration sabot
velocity nature of event and mass of explosive recovered
Trial
Velocity
ms
event
-1
mass %
Recovered
13 mm flat spigot
100
Yes def
42
13 mm flat spigot
65
No
92
13 mm diam. Flat Spigot air bubble
53
Yes def.
48
As above
45
Yes def
45
As above
36
No smell
96
13 mm diam round nosed spigot
86
No sl. evi
95
103
yes
52
4 mm round nose
115
No smell
97
Flat nosed
115
yes
30
When events occurred they were deflagrations confirmed my metalographic

examination of the fragments. In these PBX tests not all the explosive was consumed
in the deflagration. As soon as the pressure generated exceeded the tensile strength of
the bolts holding the target together the vehicle separated, the pressure released and
the burning process was extinguished. About 50% of the contents were recovered
from the floor of the test facility and the fragments of the target assembly. Even
addition of a reinforcing plate over the piston housing, although it resulted in
apparently more violent reaction, which distorted the covering plate, it was still a
deflagration. The interesting point is that the smaller diameter piston seemed to
7
113
Czech Republic
2006
[Content]
require more energy to be deposited by the spigot since events only occurred at higher
impact velocities. This is opposite to the effect predicted by Lee8 based on energy
density. The same energy over a smaller area should have an increased ignition effect.
The reason for this may come from the motion of the spigot. Not all the energy
imparted to the piston is delivered to the explosive. The piston is arrested by the
carrier rather than the spigot in the explosive causing the stop. The explosive sample
offers less resistance to the smaller spigot and hence the same energy may not be
delivered for the same impact velocity. Methods for measuring the energy transferred
to the explosive by the spigot rather than just the piston K.E. are under evaluation.
Simple conservation of momentum principles need modification.
Tests to confirm the effects of detonation were performed by mounting two
explosive pellets which had been impacted but failed to initiate into a target assembly
and inserting an electric detonator L2A1 in the cavity with the firing wires passing out
through the gauge slots. The piston was inserted into the holder but was free to move.
The whole assembly was mounted on the target plate in the normal mode and the
detonator fired. Typical fragments are shown in the figure below.
Figure 7 Showing Rear Plate fragments recovered from full detonation in target
The back plate was destroyed and the confining ring fragmented with the
bottom plate perforated. The surprising effect was the piston was retained in the
carrier housing because the shock wave had mushroomed the end protruding beyond
114
Czech Republic
2006
[Content]
the carrier before the piston had chance to move backwards in the carrier. Notice how
similar the fragments are to those obtained from the PETN samples mentioned below
where the impact induced a run to detonation
(b) PETN/wax
The results from these firings are summarised in the table below.
Table 2 PETN/wax Results Summary
Shot
No
(a)
1
2
3
4
5
6
7
8
9
10
Trial Type
(b)
10mm intrusion
10mm intrusion
10mm intrusion
10mm intrusion
10mm intrusion
10mm intrusion
15mm Shear
15mm Shear
Pinch
Pinch
Spigot
Surface
(c)
Plain
Plain
220 grit
plain
80 grit
80 grit
plain
plain
plain
plain
Impact
Velocity
Ms-1
(d)
28.2
47.77
41.8
42.857
42.25
37.1
43.86
47.78
47.32
84.74
Response
(e)
Event ( old o rings)
Spigot damaged
No reaction
Event
Confinement loss
No reaction
No reaction
Event
Event
The points of interest are that the spigots surface roughness appeared to
influence the sensitivity. Shear was not found to be a particularly exciting stimulus in
that none of our shear experiments produced an event even though we were firing at
velocities at which both friction and pinch had given violent deflagrations. Pinch was
deemed to be the most sensitive since reaction occurred at similar velocities to simple
intrusion experiments but the event at the higher velocity of 87 m s-1 was a detonation
which did not occur with any of the other stimuli at that velocity. Examination of the
target fragments recovered showed brittle fracture
115
Czech Republic
2006
[Content]
Figure 8 Firing 10 against PETN/wax showing target destruction

The added wax is a desensitiser for the PETN but the sensitivity has not been reduced
to the level of the PBX tested. Examination of the traces from the PVDF gauges show
typical pressure increase leading to DDT following mechanical impact with increased
velocity but below the shock threshold
Figure 9 Showing Oscilloscope traces recorded from PVDF gauges mounted at

front and rear of the target pellet.
116
Czech Republic
2006
[Content]
Conclusions
The most significant stimuli for accidental initiation are :pinch or nip of the explosive between end of spigot and rear steel plate
gas compression within the explosive
Friction between explosive and spigot surface
Increasing surface roughness of spigot increases response to impact
Shape of the spigot is important. Flat fronted spigot impact results in events at lower
velocities than hemispherical fronted spigots.
Shear does not appear to be a significant source of initiation but may be a contributing
factor
PETN when suitably confined can run to detonation before confinement is lost if the
energy input is above the simple deflagration threshold
Acknowledgements
Staff at COTEC ranges for assembly of the protected firing facilities and providing a
TCO and pickets during the firing.
Andrew Jones, Peter Williams and other members of staff from AWE, Aldermaston,
who provided both the instruments and the technical expertise to operate the fast
video cameras and recording oscilloscopes
Lt. Col. T B Rapple HQ Land Command, Wilton, private communication

F.P. Bowden and A.D. Yoffe, Initiation and Growth of Explosions in Liquids and Solids,
Cambridge University Press, 1952
3
Encyclopaedia of Explosives and related Items 8, 102, 1978, R and D Command, US Army
4
JSP333, Services Text book of Explosives, chapter 15, 2nd edition, HMSO(U.K.) 1989
5
P.W. Cooper & S.R.Kurowski, ch 4, Introduction to the technology of explosives Wiley VCH 1996
6
S.K. Chidester, L.G. Green & C.G. Lee, A frictional work predictive model for the Initiationof Solid
High explosives from Low Pressure Impacts, Proc. 10th Int. Det. Symposium
7
Alec Milne private communication
8
P.R. Lee, Critical Power Density: A Universal Quantitative Initiation Criteria, Proc 10 Int. Symp on
Ballistics, Peking, 1987
2
11
117
Czech Republic
2006
[Content]
THE RELATIONSHIP BETWEEN SHOCK SENSITIVITY AND

MORPHOLOGY IN GRANULAR RDX
Helen Czerski, William G. Proud, John E. Field
PCS Group, Cavendish Laboratory, J.J. Thomson Avenue
Cambridge, CB3 0HE, UK
Abstract: It is known that batches of the secondary explosive RDX from different
manufacturers show significant variation in their shock sensitivity.
No obvious
correlation between shock sensitivity and either chemical composition or morphology
has previous been identified which explains this. We use a range of techniques to study
the microstructure of RDX crystals and the bulk morphology of granular beds in order
to assess which hotspot mechanisms tend tot be dominant. Crystals were characterized
using mercury porosimetry, environmental scanning electron microscopy (ESEM) and
optical microscopy. This range of methods yields quantitative and qualitative data on
internal void size and number and surface structure. Shock sensitivity is quantified
using small-scale gap tests, and this demonstrates the clear differences in sensitivity
between batches from different manufacturers.
The samples used are from three
manufacturers, produced by both the Woolwich and Bachmann processes, and all have
an average particle size of approximately 1200 m.
Keywords:
1.
RDX, gap tests, sensitivity, morphology
INTRODUCTION
Explosives can be found in many different forms such as liquids, pressed powder compacts,
polymers, and two-phase mixtures such as polymer-bonded explosives. These can be
divided into homogenous and heterogeneous materials and this division provides a basis for
describing the details of their response to mechanical stimuli. A homogenous explosive is
simpler to understand since it is a continuous medium; each region has the same material
properties as those adjacent to it and parameters such as temperature, pressure and density
vary continuously across the bulk. In general the mechanisms of ignition and initiation in
these materials are well-understood. For example, ignition in a liquid containing a number
of bubbles is likely to start as the bubbles collapse to form hotspots(1).
Most commonly used explosives fall into the second category. Solid explosives such as
RDX and HMX perform very well, releasing a considerable amount of energy per unit
volume and having high detonation pressures and velocities. However, they are produced in
a crystalline form and are usually either pressed into high-density compacts or are mixed
with a polymer binder to produce a polymer-bonded explosive.
In either case, the
heterogeneities in the system make the details of ignition or initiation much harder to
understand than is the case for homogenous explosives. These heterogeneities lead to a
highly non-uniform temperature and stress field in the shocked material, and it is the highesttemperature regions within this distribution which will form critical hotspots(2) and lead to
the reaction of the bulk.
118
The Relationship between Shock Sensitivity and Morphology in Granular RDX
Czech Republic
2006
[Content]
The commonly used secondary explosive RDX is produced by many different manufacturers
and it has been found that there are considerable variations in its sensitivity(3). The batches
are generally found to be chemically identical and so the explanation for these differences
seems to lie in the crystal and bed morphology. In this research, four batches of RDX of a
single size from three different manufacturers were examined. Their sensitivity to shock was
assessed using a small-scale gap test, and their morphology was examined using a wide
range of techniques.
The exact processes causing ignition and initiation in granular
explosives are not yet completely understood and so this study should shed light on these
details. We have the opportunity to compare a wide range of large crystals with varying
morphologies.
The processes which occur when a shock wave passes through a granular material do not
depend only on the features of individual crystals. A granular bed consists of individual
particles, supported by each other, with a small number of contact points per particle. Low
level stresses are not supported by all the particles equally, but are transmitted by force
chains which depend on how the particles fit together. Considerable work has been done on
the structure of granular materials and how they transmit stress, but almost all of it relates to
the quasi-static regime. During a shock, there will be a qualitatively different response(4)
which will significantly modify the material behaviour, with the result that quasistatic
studies may be of limited use.
In the past, many studies have been done to try to link particular RDX particle features, for
example closed internal pores, to sensitivity. There has been some success(5,6,7,8), but a
complete explanation has not been found. It is possible that part of the explanation relates
to how the particles interact with each other through the contact points and not only how
particles respond individually. In general, the mechanisms which occur will contribute to
the inhomogeneity of the temperature field and critical hotspot formation are friction, viscoplastic work, jetting, and adiabatic compression of gas-filled pores(9).
The approach that will be taken here is to examine the morphology in a general sense,
including how particles are arranged and how they interact as a bulk material, as well as
studying the features of individual particles.
It should be noted that the sensitivity differences between different batches of RDX have
mostly been observed for material cast in a polymer binder. We have chosen to test the
sensitivity of free-poured material in order to study the simplest possible system, and to
investigate how much the reduced sensitivity effects are due to the behaviour of the asreceived material and not due to interaction with the binder or damage caused by pressing.
Helen Czerski, William Proud, John Field
119
2.
Czech Republic
2006
[Content]
EXPERIMENTAL METHODS
2.1
Samples
Granular RDX from three different manufacturers in a single size class was acquired. The
particles had an average size of 1200 m and all fell within the range 600-1500m. Laser
particle size analysis confirmed the size distributions.
To make the samples, as-received crystalline material was poured into the confinements in
small increments and tapped. This resulted in a very reproducible porosity for each sample
batch.
For the purposes of this paper, samples are labelled in order of sensitivity within
each size class, with (1) being the most sensitive material.
2.2
Sensitivity tests
Shock sensitivity was measured using a small scale gap test(10) (see figure 1). The detonator
generates a reproducible shock wave which is then attenuated by a PMMA gap. By
conducting tests with various gaps, the critical gap can be found the largest gap
(corresponding to the lowest pressure) at which ignition will occur.
Figure 1. Diagram of the small-scale gap test. The charge was 9mm in diameter and
25mm long, contained within a 25mm diameter PMMA confinement. The shock pressure
from the detonator alone was approximately 15GPa.
The gap test results are shown in figure 2. It can be seen that there are significant variations
in shock sensitivity and that some samples require twice the input pressure of others in order
for reaction to start.
The criterion that must be satisfied for a go result in a gap test is that part of the charge
detonates. In these materials, almost all the no go samples ignited, but did not transition to
a detonation. From the streak photography used to follow how the reaction progressed with
time, it can be seen that there was a steady increase in the burn speed up to the point where
the material actually detonated. All detonations were prompt detonations and the input
pressure necessary to cause a go result is therefore the detonation pressure. This is
shock-to-detonation or SDT. It is likely that the sensitivity hierarchy would be the same
120
Czech Republic
2006
[Content]
if longer charge lengths had been used to allow deflagration-to-detonation processes to

occur, but this has not been demonstrated for these samples.
5
4.5
4
Gap /mm
3.5
3
2.5
2
1.5
1
0.5
0
0
Figure 2: Gap test results for the four batches. Filled symbols represent go results and
open symbols represent no go results. An additional 2mm of PMMA gap attenuates the
shock by approximately a factor of 2. These results show significant differences in shock
sensitivity, since nearly twice the critical pressure for sample (1) is required to initiate
sample (4).
2.3
Investigation of morphology
A range of techniques are available to study individual particle morphology. Environmental

scanning electron microscopy yields information on surface features of 1m and above in
size. Optical microscopy of particles which are surrounded by a refractive index-matched
fluid can be used to observe closed internal voids down to 1m in size, and also shows the
general particle morphology. Mercury porosimetry provides surface roughness data in the
form of specific surface area (with a resolution down to 0.01m).
Figures 3 and 4 show typical crystals viewed using optical microscopy and ESEM.
discernable surface or internal features correlate with sensitivity in the material.
No
The contents of the closed internal voids seen is unknown, but it has been shown(5) that it is
likely to be a mixture of water, solvent and air. Gas-filled voids are expected to affect shock
sensitivity the most as the gas is compressible and allows the void to collapse and the
material around it to plastically deform.
Previous studies(11) have shown that shock
sensitivity can decrease with void content, i.e a higher pressure is required for the system to
detonate. This can be understood if a variety of mechanisms can cause critical hotspots and
there is no single ignition process for all particle morphologies.
121
Czech Republic
2006
(1)
(2)
(3)
(4)
[Content]
Figure 3 Optical microscopy showing typical crystals from the four batches. The field of
view in each case is 1.3mm wide.
Quantifying bulk morphology is far more difficult, but an indication of intergranular friction
(which may affect the packing and co-ordination number) is provided by measuring the
angle of repose. The angle of repose is the angle between the horizontal and the slope of a
poured pile of the granular material (see figure 5). Particles which interlock more and have
greater intergranular friction will be able to support a steeper slope.
For all of these samples, the measured bulk properties (porosity and angle of repose) were
very similar in spite of the large differences in overall shape. However, they are very
different to the measured quantities for batches of RDX with smaller particle sizes(5), so this
may be relevant in a wider context.
122
Czech Republic
2006
(1)
(2)
(3)
(4)
[Content]
Figure 4. ESEM images of the surface of each of the four crystal batches. Since any
critical hotspot is expected to be 0.1-10m in size, it might be expected that features of
approximately that size would be potential sites for critical hotspots.
However, no
correlation is seen here between any surface feature and sensitivity. (1) is the most
sensitive and (4) is the least sensitive material.
Table 1. All quantifiable data for each batch. Samples are shown with the most sensitive
at the top of the list and the least sensitive at the bottom.
Samples from different
manufacturers were supplied with up to 4% of HMX as an impurity. No average void
content is shown because this data does not capture the highly irregular void shape and size
in some samples, and such a comparison would be misleading. The percentage of the
theoretical maximum density (% TMD) data relates to the free-poured and tapped samples
which were used for the sensitivity tests. No trends can be seen in any single parameter as
the shock sensitivity decreases.
Sample
Critical Gap
/mm
%
HMX
(1)
(2)
(3)
(4)
3.6 0.2
3.2 0.4
2.9 0.3
1.9 0.4
0
4.1
0
0.3
Specific
Surface
Area m2/g
0.0044
0.0087
0.0018
Angle of
repose
% TMD
35 2
35 1
35 2
32 2
62 2
62 1
59 1
63 1
123
Czech Republic
2006
[Content]
Figure 5. Angle of repose of a powder.
3.
DISCUSSION
Am empirical study of the morphology and shock sensitivity of a material cannot prove in
absolute terms whether any one particular mechanism alone is responsible for sensitivity.
Direct observation of these mechanisms causing critical hotspots in a shocked material is not
possible in these systems. However, any correlation seen would be a strong indication of
the importance of a particular feature and a corresponding mechanism.
These results show that there are significant differences in sensitivity between different RDX
batches. Some manufacturers claim to produce a "reduced sensitivity" product, but no
causal explanation of these effects has been found. The batches tested here are a mixture of
"reduced sensitivity" products and standard products from manufacturers. Much of the
current effort to quantify and explain the difference has made the assumption that there is
one dominant property or mechanism which extends across all the products. No clear
correlation is seen here between any measured property and the shock sensitivity.
Large particles are more likely to fracture than smaller particles, which tend to plastically
deform, and it seems that some of the stress in the shock could be relieved by particle
fracture before ignition. The samples studied here contained large grains of millimetre
dimensions, and so are also likely to contain large internal cracks and defects initially. This
means that fracture is more likely than plastic deformation to relieve the initial stress.
Fracture surface energy in these materials is not sufficient to cause critical hotspots, but the
fractured morphology could respond differently to the continued loading.
Jetting from surfaces concave to the incident shock can be a source of critical hotspots. As
the radius of curvature of the concave region increases, the resulting jet increases in speed.
Jets causing critical hotspots are most likely from cavities with a radius of a few microns or
less. Significant numbers of external particle features which meet this criterion were not
observed in any one sample more than the others and only the least sensitive material had
significant numbers of closed internal voids of this size. It seems that jetting alone is not a
likely critical mechanism in this case.
124
Czech Republic
2006
[Content]
There are large closed internal voids in the most sensitive material examined, but their
contents is not known. Hotspots caused by adiabatic heating of trapped gas are more likely
when the gas spaces are larger, but in this case the closed internal voids are extensively
interconnected. No other sample contained significant numbers of larger internal voids and
there is no correlation between internal void content and sensitivity here.
The most
important factor here could be the number of gas-filled spaces only, since closed internal
voids could also contain water or solvent.
No clear correlations were seen between the bulk material properties and the sensitivity.
Quantities such as co-ordination number are difficult to measure directly, but could be very
relevant, since the lower the number of interparticle contacts, the higher the stress
concentration at each contact. The importance of such parameters has not been investigated,
but it seems very likely that the nature of the packing and the interparticle contacts must play
a large role in localising stress. Contacts such as these will experience considerable stress
and plastic deformation at such regions could cause very high temperatures locally, making
these potential sites for critical hotspots.
The fact that the go results were all prompt detonations implies that the input pressures
necessary to cause a go were above or equal to the detonation pressure. However, it
seems likely that the resulting sensitivity hierarchy will be the same as one measured with
longer charges.
4.
CONCLUSIONS
In the case of free-poured granular samples, there is no correlation between any individual
morphological feature and shock sensitivity for these large particle sizes. Previous work(7)
has shown that there is no such correlation for smaller particle sizes either. This suggests
that there is no single quality leading to the differences in sensitivity between different RDX
products.
The critical pressure for these samples varies by a factor of two between the most and least
sensitive, so there are significant sensitivity differences between samples of the raw material
from different manufacturers. The observed variation in shock sensitivity may be explained
by qualitative differences in how a shock interacts with the bulk material as well as with
individual particle features. It may be the case that the combination of material properties in
RDX mean that several critical hotspot mechanisms become significant at a similar shock
pressure. Separating out specific mechanisms and quantifying their contribution would be
extremely difficult at this stage. For the samples examined here, we cannot identify which
mechanisms are dominant in causing critical hotspots, but we can make some suggestions.
Bulk morphology is difficult to quantify since properties such as average co-ordination

number cannot be directly measured. Only angle of repose and the particle size analysis
provide quantitative data on bulk morphology and neither of these quantities correlates
directly with sensitivity. However, it seems likely that the interaction of particles in the
granular bed must make some contribution to the inhomogeneity of the stress field. A nonuniform stress field will cause a non-uniform temperature field, because of the
micromechanical response of the material to shock. There are many processes that could
8
125
Czech Republic
2006
[Content]
cause critical hotspots (viscoplastic deformation, jetting, friction, adiabatic gas collapse) and
all are directly affected by both the microstructure and how the particles are arranged.
Further investigation of how the interaction of bulk and individual particle morphology
might affect shock sensitivity in free-poured granular beds is required. Future papers will
deal with samples which has smaller particle sizes than those reported here.
5.
ACKNOWLEDGEMENTS
We would like to acknowledge Adam Cumming, Dave Tucker and Richard Biers of [dstl]
for funding this research and Ron Hollands or BAe Land Systems for providing samples and
background knowledge. We are grateful for technical assistance from Dave Powell of the
Cavendish Laboratory.
6.
REFERENCES
[1]
N. K.BOURNE and J. E. FIELD : Explosive ignition by the collapse of cavities, Proceedings of

the Royal Society series A, 455(1987): 2411-2426 Jul 8 (1999)
[2] F. P.BOWDEN and A. D. YOFFE : Initation and Growth of Explosion in Liquids and Solids,
Cambridge University Press, London, 1952
[3] F. PEUGEOT and D. WATT: RS-RDX, Literature Review and Discussions, NIMIC report, 2004
[4] V. F.NESTERENKO: Dynamics of Heterogeneous Materials, Springer Series: High Pressure
Shock Compression of Condensed Matter, 2001
[5] L. BORNE: Explosive Crystal Microstructure and Shock Sensitivity of Cast Formulations, 11st
Detonation Symposium, 1998, p657-663
[6] L. BORNE and A. BEAUCAMP: Effects of Explosive Crystal Internal Defects on Projectile
Impact Initiation, 12th Detonation Symposium, 2002, p35
[7] H. MOULARD, J.W.KURY, and A. DELCLOS : The effect of RDX particle size on the shock
sensitivity of cast PBX formuations, 8th Detonation Symposium, 1986
[8] L. BORNE and J. PATEDOYE: Quantitative Characterization of Internal Defects in RDX
Crystals, Propellants, Explosives, Pyrotechnics 24, 255-259, 1999
[9] J. E. FIELD, N.K. BOURNE, S.J.P.PALMER, S.M. WALLEY, J.M. SMALLWOOD: Hotspot ignition
mechanisms for explosive and propellants, Proceedings of the Royal Society series A,
339(1654): 269-283 May 15 1992
[10] A. CHAKRAVATY, M. J. GIFFORD, M. W. GREENAWAY, W.G. PROUD and J. E. FIELD : Factors
affecting shock sensitivity of energetic materials, Shock Compression of Condensed Matter,
1007-1010, 2001
[11] H. CZERSKI, M.W.GREENAWAY, W.G. PROUD and J. E. FIELD : Links between the morphology
of RDX crystals and their shock sensitivity, Shock Compression of Condensed matter, 2006
[12] H. CZERSKI, W. G. PROUD and J. E. FIELD: Aspects of Material Insensitivity, Submitted to
Insensitive Munitions & Energetic Materials Technology Symposium, Bristol 2006
126
Czech Republic
2006
[Content]
GUANIDINIUM NITROFORMATE SALTS:

POSSIBLE NEW OXIDIZERS FOR HIGH PERFORMANCE,
HALOGEN FREE SOLID PROPELLANTS
Michael Gbel, Thomas M. Klaptke and Peter C. Thumbs
Department of Chemistry and Biochemistry
Ludwig-Maximilians University of Munich
Butenandtstr. 5 - 13 (Haus D)
D-81377 Munich, Germany
Abstract:
Energetic salts of the nitroformate anion with several guanidinium cations were
investigated, including guanidinium nitroformate monohydrate (GNFH),
aminoguanidinium nitroformate (AGNF), diaminoguanidinium nitroformate
(DAGNF) as well as triaminoguanidinium nitroformate (TAGNF). Guanidinium
nitroformate was obtained according to the literature procedure as the
monohydrate (GNFH), whereas AGNF, DAGNF and TAGNF are reported for the
first time. All salts were characterized using vibrational spectroscopy (IR,
Raman), mass spectrometry, multinuclear NMR spectroscopy and elemental
analysis. In addition, the impact, friction and electrostatic sensitivity data were
determined for all compounds. The crystal structures of GNFH, AGNF, DAGNF
and TAGNF were determined using single crystal X-ray diffraction. The thermal
decomposition of the salts was monitored using differential scanning calorimetry.
Keywords:
1.
nitroform, X-ray crystallography, GNFH, AGNF, DAGNF, TAGNF
INTRODUCTION
Although several energetic salts containing the nitroformate anion have been reported in the
literature, so far only one compound has gained practical application. Since its discovery in
1951, more than eighty publications [1] have since appeared, showing that there is continued
interest in investigating the properties of hydrazinium nitroformate (HNF). Furthermore, it
has been shown that a formulation containing HNF, aluminum (Al) and glycidyl azide
polymer (GAP) has improved properties in comparison with other composite propellants,
and a theoretically predicted performance that is superior to even advanced propellants
based on the oxidizer ADN. Investigations into the use of HNF and HNF containing
propellants continues to be undertaken by various companies including Nobel's Explosives
Company, ICI (Great-Britain), Fiat Avio (Italy) and SNPE (France) [2].
Recently,
two
derivatives
of
hydrazinium
nitroformate
(HNF),
namely
monomethylhydrazinium nitroformate (MMHNF) and dimethylhydrazinium nitroformate
(DMHNF) were synthesized and suggested to be not only promising new high performance
energetic materials, but also potential ingredients for melt castable high explosive
formulations [3].
In the course of our investigations into high energy density materials (HEDM) [4-13], we
recently focussed our attention to various guanidinium nitroformate salts, in order to explore
their potential as possible new oxidizers for high performance, halogen-free solid
propellants, which would avoid the problematic formation of hydrogen chloride when used
as oxidizers in rocket propellant formulations.
1
Michael Gbel, Thomas Klaptke
127
2.
Czech Republic
2006
[Content]
SYNTHESIS
NHR3
NO2
K/Ag
C
O2N
NO2
R1HN
NHR2
NO2
Cl
O2N
NO2
NHR3
R1HN
NHR2
KCl/AgCl
GNFH: R1 = R2 = R3 = H
AGNF: R1 = NH2, R2 = R3 = H
DAGNF: R1 = R2 = NH2, R3 = H
TAGNF: R1 = R2 = R3 = NH2
All of the compounds were obtained as bright yellow solids according to a slightly modified
procedure [14] in high yields by the reaction of either the potassium or silver salt of nitroform
with the corresponding guanidinium chloride in acetonitrile.
3.
CRYSTAL STRUCTURES
Fig 1. Asymmetric unit of Guanidinium nitroformate hydrate (GNFH) (1) (left) and
Aminoguanidinium nitroformate (AGNF) (2) (right).
Fig 2. Asymmetric unit of Diaminoguanidinium nitroformate (DAGNF) (3) (left) and

Triaminoguanidinium nitroformate (TAGNF) (4) (right).
128
Guanidinium Nitroformate Salts: Possible New Oxidizers for High Performance, . . .
TABLE 1.
2006
[Content]
Composition of Crystallographic data of GNFH, AGNF, DAGNF and

TAGNF.
compound
formula
formula weight
temperature / K
crystal system
space group
a/
b/
c/
/
/
/
volume / 3
Z
absorption coefficient /mm-1
density exptl. / g / cm3
F(000)
2 theta /
index ranges
reflections collected
reflections unique
parameters
GOOF
R1 / wR2 [I > 2(I)]
R1 / wR2 (all data)
Czech Republic
GNFH (1)
C2 H8 N6 O7
228.14
200(2)
monoclinic
C 2/c (no. 15)
8.1859(4)
14.3541(7)
7.7556(5)
90.0
101.132(2)
90.0
894.15(8)
4
0.167
1.695
472
54.88
- 10 h 9,
- 18 k 17,
- 9 l 10
1021
811
87
1.072
0.0513 / 0.1293
0.0645 / 0.1395
AGNF (2)
DAGNF (3)
TAGNF (4)
C2 H8 N8 O6
C2 H9 N9 O6
C2 H7 N7 O6
225.15
240.16
255.18
200(2)
200(2)
200(2)
triclinic
monoclinic
triclinic
P 21/n (no. 14)
P -1
P-1
5.0690(7)
10.980(1)
8.021(2)
7.559(1)
7.7524(8)
8.347(2)
11.320(1)
11.416(1)
8.515(2)
84.567(9)
90.00
105.49(2)
84.425(9)
105.336(8)
95.03(2)
79.61(1)
90.00
111.10(2)
423.29(9)
937.1(2)
501.7(2)
2
4
2
0.170
0.162
0.159
1.766
1.702
1.689
232
496
264
52
52
52
- 6 h 6,
-13 h 13,
-9 h 9,
- 9 k 9,
-9 k 9,
-10 k 9,
- 13 l 13
- 14 l 14
- 6 l 10
1657
1832
1962
1515
1573
1617
164
178
190
1.086
1.168
1.099
0.0312 / 0.0748 0.0467 / 0.1064 0.0391 / 0.1014
0.0346 / 0.0773 0.0604 / 0.1155 0.0491 / 0.1113
129
Czech Republic
2006
[Content]
Considering the relationship of the orientation of the nitro groups relative to the central C-N3
plane it appears that the relation of bond lengths vs. dihedral angles of the nitro groups works
well [15]. A correlation between the length of the C-N and N-O bonds and the corresponding
dihedral angles can be observed as expected, whereby the C-N bonds of out-of-plane nitro
groups are elongated and C-N bonds of in-plane nitro groups are shortened. There is also a
reverse tendency for N-O distances. The structural characteristics such as bond distances and
dihedral angles of the guanidinium salts of nitroform are presented in Table 2, and fall within
the range of expected values.
TABLE 2.
Selected structural parameters of the nitroformate anion in the guanidinium,

aminoguanidinium, diaminoguanidinium and triaminoguanidinium salts.
nitro groups of:

GNFH
1
2
3
AGNF
1
2
3
DAGNF
1
2
3
TAGNF
1
2
3
130
dihedral angle []
C-N distance
[]
N-O distance []
80.9(3)
4.8(6)
4.8(6)
1.454(2)
1.363(5)
1.363(5)
1.217(4)
1.246(9)
1.246(9)
16.1(2)
35.7(2)
24.4(2)
1.385(2)
1.418(5)
1.409(6)
1.243(2)
1.228(1)
1.235(1)
24.5(5)
31.7(8)
9.7(8)
1.397(2)
1.411(2)
1.396(2)
1.235(5)
1.229(5)
1.238(2)
2.6(7)
95.2(7)
2.7(2)
1.373(2)
1.461(2)
1.367(2)
1.251(6)
1.217(6)
1.254(2)
4.
Czech Republic
2006
[Content]
NMR SPECTROSCOPY
An investigation of the NF anion and the corresponding guanidinium cation in

solution was performed using 1H, 13C and 15N NMR spectroscopy. The 13C and 15N
spectra of the anion showed one signal at room temperature. A comparison of the
obtained signals can be found in Table 3. The signals of the NF anion fit well with
those reported by A.A. Gakh et al. [15] who reported the 13C chemical shifts of the
cesium and the tetrabutylammonium salts of the nitroformate anion at 150.3 ppm and
151.6 ppm (both recorded in acetone-d6), respectively.
Table 3.
NMR Spectroscopic data of GNFH, AGNF, DAGNF and TAGNF.

13
GNFH
AGNF
5.
C (DMSO-d6) / :
158.0 (C(NH2)3+)
150.3 (C(NO2)3-)
158.9(H2N-NH=C(NH2)2+)
150.3 (C(NO2)3-)
DAGNF
159.9 (H2N-C(NH-NH2)2+)
150.4 (C(NO2)3-)
TAGNF
161.8 (C(NH-NH2)3+)
150.6 (C(NO2)3-)
15
N (DMSO-d6) / :
-306 (NH2)
-31 (NO2)
-326 (NH-NH2)
-311 (NH-NH2)
-284 (C-NH2)
-30 (NO2)
-328 (NH-NH2)
-313 (NH-NH2)
-288 (C-NH2)
-31 (NO2)
-249 (NH-NH2)
-95 (NH-NH2)
-31 (NO2)
H (DMSO-d6) / :
6.9 (NH2)
3.39 (H2O)
8.52
6.94
4.28
8.50 (NH-NH2)
4.48(C-NH2,NH-NH2))
9.79 (C(NH-NH2)3+)
3.53 (C(NH-NH2)3+)
DECOMPOSITION EXPERIMENTS
Differential Scanning Calorimetry (DSC) was used to determine the relative thermal
stabilities of compounds 1-4. Figure 3 shows the DSC curves that were obtained for 2-4.
Whereas 2 shows only one decomposition signal (80C (onset)), 3 and 4 show melting
points ((78C (onset)) and (83C (onset))) as well as subsequent decomposition ((110C
(onset)) and (138C (onset)) using a heating rate of 20C/min. The thermal stability
increases from compounds 2-4. Since 1 is a hydrate, it was not included in Figure 3. Table 3
summarizes the thermal behaviour of 1-4 using heating rates of 2/min. IR spectroscopy was
used to show that slow decomposition of 2 and 3 occurs at room temperature. AGNF and
DAGNF were converted to the corresponding nitrates on standing at room temperature and
are believed to decompose in much the same way as potassium nitroformate which is
converted to potassium nitrate [16].
131
Czech Republic
2006
[Content]
Fig 3. DSC curves of AGNF, DAGNF and TAGNF ( = 20/min).
The sensitivity data of GNFH, AGNF, DAGNF and TAGNF were measured in order to
establish safe handling procedures for these compounds. Two different instruments were
used to determine the friction and impact sensitivity: the Bundesanstalt fr
Materialforschung und Prfung (BAM) drop hammer (BAM fh (Fallhammer)) and friction
tester (BAM ft) [17]. GNFH was found to meet the United Nations (UN) recommendations
for the transport of dangerous goods, with a friction sensitivity of greater than 360 N and an
impact sensitivity of greater than 29.6 J. In contrast, great care should be taken when
handling the other salts (Table 2) which are considerably more sensitive.
Table 4.
Comparison of the sensitivity data of GNFH, AGNF, DAGNF and TAGNF.

The values of the melting and the decomposition points correspond to a
heating rate of = 2/min.
Parameter
GNFH
AGNF
DAGNF
TAGNF
C2H8N6O7
C2H7N7O6
C2H8N8O6 C2H9N9O6
formula
228.12
225.15
240.16
255.18
weight [g/mol]
1.70
1.77
1.70
1.6
density [g/cm3]
110-117
80
84
melting point [C]
71
82
104
decomp. point [C]
deflagration deflagration deflagration deflagration
flame
negative
negative
negative
negative
electrostatic discharge
> 360
144
32-40
20-24
friction sensitivity [N]
30
10
5-6
2-2.5
impact sensitivity [Nm]
132
6.
Czech Republic
2006
[Content]
CONCLUSION
Although guanidinium nitroformate monohydrate (GNFH) has already been mentioned in

the literature,[14] the characerization reported in the original paper was limited to the melting
point as well as IR spectroscopy and mass spectrometry. This compound has now been
extensively characterized, as well as the related anhydrous species aminoguanidinium
nitroformate
(AGNF),
diaminoguanidinium
nitroformate
(DAGNF)
and
triaminoguanidinium nitroformate (TAGNF), which have been synthesized for the first time.
The sensitivity data and the thermal behaviour of the compounds appears to follow a general
trend. Correlations of analytical, sensitivity and performance properties of these compounds
within the series of homologous compounds, as well as between these compounds and HNF
have been established. Triaminoguanidinium nitroformate (TAGNF) seems to be a more
promising candidate for practical application than the GNFH, AGNF and DAGNF salts.
Acknowledgements
We thank Priv.-Doz. Dr. K. Karaghiosoff for carrying out the NMR experiments. We are
grateful to the Cusanuswerk for a scholarship to MG. Ludwig-Maximilan University and the
Fonds der Chemischen Industrie are thanked for generous financial support.
133
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
134
SciFinder enquiry 11/2005

www.appbv.nl/hnf.html
Synthesis, characterization and thermal behaviour of hydrazinium nitroformate (HNF)
and its new N-alkyl substituted derivatives. Jadhav, H. S.; Talawar, M. B.; Dhavale,
D. D.; Asthana, S. N.; Krishnamurthy, V. N., Indian Journal of Chemical Technology
2005, 12(2), 187.
Mononitrobiuret (MNB) and 1,5-Dinitrobiuret (DNB): Pyrolysis Experiments and
Thermochemistry, J. Geith, G. Holl, T. M. Klaptke, J. J. Weigand, Combustion and
Flame 2004, 139, 358.
1,4-Bis-[1-methyltetrazol-5-yl)-1,4-dimethyl-2-tetrazene: A Stable, Highly Energetic
Hexamer of Diazomethane (CH2N2)6 ,T. M. Klaptke, P. Mayer, A. Schulz, J. J.
Weigand, Propellants Explosives and Pyrotechnics 2004, 29, 325.
Improved Synthesis and X-Ray Structure of 5-Aminotetrazolium Nitrate, M. v.
Denffer, T. M. Klaptke, G. Kramer, G. Spie, J. M. Welch, G. Heeb, Propellants
Explosives and Pyrotechnics, 2005, 30(3), 191.
1,5-Diamino-4-methyltetrazolium dinitramide, T. M. Klaptke, P. Mayer, A. Schulz,
J. J. Weigand, J. Am. Chem. Soc., 2005, 127, 2032.
Derivatives of 1,5-Diamino-1H-tetrazole A New Family of Energetic Heterocyclicbased Salts, J. C. Glvez-Ruiz, G. Holl, K. Karaghiosoff, T. M. Klaptke, K.
Lhnwitz, P. Mayer, H. Nth, K. Polborn, Ch. J. Rohbogner, M. Suter, J. J. Weigand,
Inorg. Chem. 2005, 44(12), 4237.
A study on the thermal decomposition behavior on derivatives of 1,5-diamino-1Htetrazole (DAT): a new family of energetic heterocyclic-based salts G. Fischer, G.
Holl, T, M. Klaptke, J. J. Weigand, Thermochim Acta 2005, 437, 168.
The Elimination of NO2 from Mixtures of the Nitramines HMX, RDX and CL20 with
the Energetic Binder Glycidyl Azide Polymer (GAP) - A Computational Study I, M.
A. Bohn, A. Hammerl, K. Harris, T. M. Klaptke, Central European Journal of
Energetic Materials, 2005, 2(2), 29.
Azidoformamidinium and Guanidinium 5,5'-Azotetrazolate Salts, M. A. Hiskey, A.
Hammerl, G. Holl, T. M. Klaptke, K. Polborn, J. Stierstorfer, J. J. Weigand,
Chemistry of Materials, 2005, 17, 3784.
The Dianion of 5-Cyanoiminotetrazoline: C2N62-, T. M. Klaptke, C. Kuffer, P.
Mayer, K. Polborn, A. Schulz, J. J. Weigand, Inorg. Chem., 2005, 44(16), 5949.
Further Decomposition Pathways of Mixtures of the Nitramines HMX, RDX and
CL20 with the Energetic Binder Glycidyl Azide Polymer (GAP) - A Computational
study II, M. A. Bohn, A. Hammerl, K. Harris, T. M. Klaptke, Central European
Journal of Energetic Materials, 2005, 2(3), 3.
Ashwin M. Krishnan, Per Sjoberg, Peter Politzer and Joseph H. Boyer, J. Chem. Soc.
Perkin Trans. II, 1989, 1237.
A. A. Gakh, J. C. Bryan, M. N. Burnett, P. V. Bonnesen, J. Mol. Struct. 2000, 520,
221.
M. Gbel, T. M. Klaptke, P. Mayer, ZAAC, 2006, in press.
Test methods according to the UN Recommendations on the Transport of Dangerous
Goods, Manual of Tests and Criteria, fourth revised edition, United Nations
Publications, New York and Geneva, 2003, ISBN 92-1-139087-7, Sales No.
E.03.VIII.2.; 13.4.2 Test 3(a)(ii) BAM Fallhammer.
Czech Republic
2006
[Content]
PRELIMINARY STUDY OF SUPERFINE METAL POWDERS EFFECT

ON THE DECOMPOSITION OF ENERGETIC MATERIALS
Alexander Gromov *, Elena Popenko **, Alexey Sergienko**
Young-Soon Kwon***, Ulrich Teipel****
* Tomsk Polytechnic University, 30, Lenin Ave., 634050, Tomsk, Russia
** Biysk Technologic Institute, 29, Trofimov Str., 659315, Biysk, Russia
*** Research Center for Machine Parts and Materials Processing, School of Materials and
Metallurgical Engineering, University of Ulsan, Mugeo-2Dong, Ulsan 680-749, South Korea
**** Fraunhofer Institute for Chemical Technology, P.O. Box 1240, D-76318 Pfinztal, Germany
Abstract:
The preliminary study describes the decomposition of energetic materials (AP, HMX) in
the presence of superfine metal powders (SFP) of Al, Ni, Cu-Ni, Fe, Cu and W. The
decomposition temperature for the energetic substances could be lowered because of
catalytic effect of superfine metal powders. The effect of several types of metal particles
produces by electro-explosive method to the rate of decomposition, temperature of
decomposition of modern energetic substances have been discussed. The of DTA-TG
investigation were conducted under non-linear heating of mixes (SFP-HMX) and (SFPAP) in air.
Keywords: decomposition, HMX, AP, catalysis, superfine metal powders, DTA-TG.
1.
INTRODUCTION
It is known that the usage of superfine Al powders (SFAP, as ~ 100 nm) instead of
industrial Al powders (as ~ 10 m) results in achievement of the double increase in the
burning rate for aluminized solid propellant, which contain 55 % of nitrocellulose (13,5 %
N), 30 % of dioxyethylnitramine and 15 % of Al [1]. The complete substitution of industrial
Al powders to superfine Al powders does lead to an increase in the burning rate of
combustion for aluminized solid propellant on the basis of active binder (20 %), AP (27 %),
HMX (35 %) and Al (18 %) [2,3]. The value of pressure exponent is reduced in the burning
law u=Ap, at the same time the combustion efficiency of metal increases and the degree of
agglomeration is reduced [4]. It is known that decomposition temperature for the energetic
substances could be lowered because of catalytic effect of superfine metal powders [5], but
the experimental data needs to be confirmed as well as the mechanism of such catalytic
process has to be developed. In this work we tested the effect of several types of metal
nanopowders, produces by electro-explosive method [6]. The purpose of this work is a
preliminary study of catalytic activity of superfine metal powders in their mixtures with AP
and HMX. The three types of superfine Al powders, SFP Fe, SFP W, SFP Ni, SFP Cu and
SFP (Cu-Ni) composition as well as micron sized Al powder (ASD-4) was tested.
2.
EXPERIMENTAL RESULTS AND DISCUSSION
Production of the samples of superfine metal powders was carried out by using an electrical
explosion of wires (EEW) machine UDP-4G which was constructed for the production of
1
Alexander Gromov, Elena Popenko, Alexey Sergienko, Young-Soon Kwon, . . .
135
Czech Republic
2006
[Content]
metal powders at the High Voltage Research Institute, Tomsk Polytechnic University,
Russia.
Non-passivated superfine metal powders self-ignite if exposed to air, making
passivation for them essential. Hence, after the EEW machine stopped, the collector with
prepared powders was placed into a separate hermetic passivation chamber immediately
after powder production. Ten samples, studied within this work and their specific surface
area, determined by BET method, as well as metal content, measured by DTA analysis, are
shown in table 1.
Table 1. Properties of nano-metals
BET, as,
m2/g m
Cmetal,
wt. %
1. SFAP-Al2O3
Initial wire Gas media in Passivation

condition
composition explosive
chamber
Al
Ar
Air
7.0
0.3
76.1
2. SFAP -AlB2
Al (B)
Ar
Air
17.0
0.1
78.2
3. SFAP -NC
Al
Ar
0.4
67.5
4. ASD-4
Industrial powder
Nitrocellulose 5.0
in ethanol
0.4
5.6
98.5
5. Fe
Fe
Ar
Air
18.2
0.1
88.9
6. Ni
Ni
Ar
Air
20.3
0.1
91.0
7. Cu
Cu
Ar
Air
16.8
0.1
93.2
8. W
Ar
Air
5.2
0.4
98.0
9. Cu-6 % Ni
Cu-6 % Ni
Ar
Air
5.9
0.4
96.5
10. Cu-45 % Ni
Cu-45 % Ni
Ar
Air
6.4
0.3
96.7
Sample code
The powders, passivated with AlB2 (SFAP-AlB2), nitrocellulose (SFAP-NC) and

SFAP, which particles had the oxide passivation layer (SFAP-Al2O3) as well as other metals
were studied by SEM (Fig.1). The application of the AlB2 passivation coating was
performed for the increase in the stability of particles to the oxidation and for the decrease in
the agglomeration of Al particles during production [7]. The specific surface area of powder
with the AlB2 coating is twice as much as for the powder, which has the oxide passivating
layer (table 1). The application of NC coating was proceed with the purpose of the study of
aluminum oxidation activation due to the high exothermic effect of NC decomposition at
T=180oC. Coating of powders by nitrocellulose, which contains 12.4 wt. % of nitrogen, was
used by the method of SFAP treatment in the 10 wt. % NC solution in acetone with the
subsequent evaporation of acetone at room temperature. As a result SFAP had the content of
NC - 33 wt. %. The other metals were passivated by air in standard procedure [8] and had the
oxide layers.
136
Preliminary Study of Superfine Metal Powders Effect on the Decomposition of . . .
Al-Al2O3
Al-AlB2
Ni
Fe
W
Fig 1.
Czech Republic
Cu-6 % Ni
2006
[Content]
Al-NC
Cu
Cu-45 % Ni
SEM images of superfine metal powders (JSM-6500 F JEOL, Japan).
The superfine metal powders were tested by the method of differential-thermal

analysis (DTA). DTA-TG (Universal 2.4 F TA Instruments) was used for testing of
components non-isothermal oxidation in air with heating rate 10 K/min. DTA results are
given in fig. 2. The temperature of the oxidation onset (Ton, oC) and the degree of the
oxidation (, %), calculated under the (eq. 1) of the studied Al powders were different
depending on the particle size.
+m
(1),
(AlAl2O3) = ---------------- 100%
Al 0.89
+m, % - weight increase on TG curve;
Al, % - Al metal content in the powder.
Industrial powder ASD-4 had Ton = 580oC, while for SFAP-Al23 this temperature
was 510 (fig. 2, a). For the SFAP-Al23 the considerably higher degree of oxidation
(=28.3 % up to T=660oC) is observed in comparison with the industrial powder ASD-4
(=1.8 %) (fig. 2, b). SFAP with the coating of AlB2 had higher Ton = 560oC in comparison
with the SFAP-Al23 (510). Probably, this is caused by the protective film of refractory
AlB2. The degrees of the oxidation of the samples SFAP-Al23 and SFAP-AlB2 are higher at
3
137
Czech Republic
2006
[Content]
660oC and approximately identical. In the DTA curve for sample Al-NC two exothermic
effects are characteristic, whose are caused by NC decomposition in the temperature range of
180-280 with the 25.0 % weight decrease and Al metal oxidation (510-620). The
gasification of the carbonaceous species, which were being formed during the NC
decomposition, occurs in the range of temperatures 280-510. The total weight loss at the
stage of NC decomposition (on TG) it achieved 31.2 %. The intensive Al oxidation for the
sample SFAP-NC occurs with T=525-620 and the degree of oxidation in this case was
=19.4 %. Thus, the processes of NC and Al decomposition under the heating in air proceed
independently of each other (with DTA curves). The parameters of thermal decomposition of
NC and the parameters of SFAP oxidation remain at the same level as for individual
substances.
The DTA results of mixtures (SFAP-Al23 + 50 % HMX) showed that the
temperature of the beginning of the decomposition (Td) of HMX in the studied mixture is
Td=260oC (fig. 2, c), which is 28oC lower than for the pure HMX (Td=288oC). Hence, SFAPAl23 initiates the decomposition of HMX at the lower temperature. The parameters of the
oxidation of SFAP-Al23 sample in presence of HMX do not change. SFAP, covered by
AlB2 and NC, have an analogous effect on the Td of HMX, i.e., the Td is decreased down to
240 and 250, respectively. The small decrease of Td for the HMX (to 270) was
observed also on the DTA curve of mixture (ASD-4 + 50 % HMX). The degree of ASD-4
oxidation (=1.1 % to 660) in the mixture with the HMX was considerably lower in
comparison with the oxidation of the sample SFAP -Al23, for which =13.3 % at 660
(fig. 2, d). The parameters of AP decomposition are practically not change in the mixture
(SFAP-Al23+50 % AP). Oxidation of SFAP-Al23 in the mixture with AP begins on 30
lower (d=480) in comparison with the oxidation of pure SFAP -Al23 (fig. 1, e,f).
The results of DTA-TG investigation of the other metal SFPs are presented in table 2.
Compared to ASD-4 and SFAP, the samples 5-10 have the very low temperatures of the
oxidation onset (174-352oC), while oxid for all metals were relatively high, except for SFP
W.
The effect of superfine metal particles on the HMX decomposition was studied with
DTA-TG (fig. 3). The mixtures (SFP Me + 50 % HMX) was suspended in the ethanol
(C2H6O), dried in air and than analysed. Pure HMX had a Td=288oC. The Cu, Fe and (Cu-6
% Ni) powder (samples 7, 5 and 10, table 3) strongly effect on HMX decomposition. The
decomposition temperature of HMX was 96oC lower as a result of the Fe powder interaction
with HMX. Hence, Fe and Cu powder has been selected for the future study.
3.
CONCLUSION
The results of experimental studies of superfine metal powder effect on the HMX and
AP decomposition shows the strong effect of Fe, Cu and Cu-Ni superfine powders on the
HMX decomposition temperature it is lowered on more than on 90oC. Superfine aluminum
powders coated by AlB2 react with HMX at the same temperatures, but with twice higher
heat release in comparison with SFAP, coated by Al2O3 and NC.
4.
ACKNOWLEDGEMENT
This work has been financially supported by INTAS grant YSF 03-55-671 and Russian
President grant 1812. 2005.8.
138
70
m, %
150
150
a)
450
300
SFAP-AlB2
c)
450
ASD-4 (Micron-sized)
SFAP-NC
SFAP-Al2O3
300
NC decomposition
SFAP-AlB2
ASD-4 (Micron-sized)
SFAP-NC
SFAP-Al2O3
600
600
, o
, o
750
750
-45
-35
-25
-15
-5
m, %
-20
20
40
60
80
DTA, mV
300
300
SFAP-Al2O3+HMX
SFAP-NC+HMX
ASD-4+HMX
SFAP-AlB2+HMX
150
150
d)
b)
450
450
600
600
, 0
SFAP-AlB2+HMX
ASD-4+HMX
SFAP-NC+HMX
SFAP-Al2O3+HMX
750
750
, 0
2006
-40
-30
-20
-10
DTA, mV
Czech Republic
10
20
30
40
-20
-10
10
20
30
40
50
60
80

[Content]
139
140
450
600
0
750
-100
-80
-60
AP
SFAP-Al2O3 + AP
750
8.
284
Fig 2.
1766*
2408*
611
4040
4870
4180
220
5790
6232
4730
oxid Me, J/g
DTA and TG curves of the studied SFAP.
*two peaks of oxidation (total).
174
181
10. Cu-45 % Ni
600
Cu
7.
337
-40
450
Ni
6.
352
239
300
Fe
5.
Cu-6 % Ni
150
300
9.
150
580
525
560
-20
m, %
ASD-4
SFAP-NC
3.
4.
SFAP-AlB2
SFAP-Al2O3 510
1.
AP
2.
Sample code on, o
SFAP-Al2O3 + AP
Table 2. DTA-TG study of metal SFPs
16
11
14
23
19
35
19
22
28
+m (to 600oC), %
Czech Republic
2006
f)
e)
-5
15
25
DTA, mV

[Content]
Czech Republic
2006
[Content]
Table 3. DTA-TG study of mixes (SFP Me + 50 % HMX)

Sample code
on, o
T, oC
oxid Me, J/g
-m (to 300oC), % Remark
HMX
288
3240
70
5.
Fe
192
96
1500
48
6.
Ni
283
747
45
7.
Cu
228
60
2270
34
8.
284
611
14
9.
Cu-6 % Ni
193
95
1800
39
10.
Cu-45 % Ni
196
92
4831
48
No effect
No effect
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
B. BASCHUNG, D. GRUNE, H.H. LICHT, SAMIRANT. Combustion Phenomena of a Solid

Propellant
Based
on
Aluminum
Powders.
Internet
Site:
www.argonide.com/Gun_Propellant_Report.pdf
V.N. SIMONENKO, V.E. ZARKO. Comparative Studying of the Combustion Behavior of
Composite Propellants Containing Ultra Fine Aluminum. In: Energetic materials. 30th Int.
Annual Conf. of ICT, 1999. Karlsruhe, Germany. Report 21, 14 P.
O.G. GLOTOV, V.E. ZARKO, M.W. BECKSTEAD. Agglomerate and Oxide Particles Generated
in Combustion of Alex Containing Solid Propellants. In: Energetic materials. Analysis,
diagnostics and testing. 31th Int. Annual Conf. of ICT, 2000. Karlsruhe, Germany, Report 130,
14 P.
U. TEIPEL, I. MIKONSAARI. Size Reduction of Particulate Energetic Material, Propellants,
Explosives, Pyrotechnics. 2002, Vol. 27, P. 168-174.
LEILI LIU, FENGSHENG LI, LINGHUA TAN, LI MING, YANG YI. Effects of Nanometer Ni, Cu, Al
and NiCu Powders on the Thermal Decomposition of Ammonium Perchlorate, Propellants,
Explosives, Pyrotechnics. 2004, Vol. 29, P. 34-38.
Y.S. KWON, Y.H. JUNG, N.A. YAVOROVSKY, A.P. ILYIN AND J.S. KIM: Ultrafine Metal
Powders by Wires Electric Explosion Method, Scripta mater. 2001, Vol. 44, P. 2247-2251.
Y.S. KWON, A.A. GROMOV, A.P. ILYIN. Reactivity of Superfine Aluminum Powder Stabilized
by Aluminum Diboride, Combustion and Flame. 2002, Vol. 131. P. 349352.
Y. S. KWON, A.A. GROMOV, A.P. ILYIN AND G.H. RIM: Passivation Process for Superfine
Aluminum Powders Obtained by Electrical Explosion of Wires, Applied Surface Science, 2003,
Vol. 211, P.57-67.
141
Fig 3.
142
Czech Republic
2006
[Content]
DTA and TGA curves for superfine metal powders mixtures with HMX under nonisothermal heating in air (m = 2.6 mg, vheat= 10 K/min, etalon -Al2O3).
Czech Republic
2006
[Content]
PREPARATION, PROPERTIES AND APPLICATION OF ENERGETIC
COMPLEX PERCHLORATES OF D-METALS

Mikhail A. Ilyushin, Igor V. Tselinsky, Irina V. Bachurina, Yurii A. Gruzdev,
Valery P. Sinditskii#, Valery V. Serushkin# and Vyacheslav Yu. Egorshin#
Saint-Petersburg State Institute of Technology (Technical University), 26 Moskovsky pr., SaintPetersburg, Russia
#
D.I.Mendeleev Russian University of Chemical Technology, 9 Miusskaya sq., Moscow, Russia.

Abstract:
Methods for the synthesis of novel energetic complex perchlorates of d-metals have been
developed. According to derivatographic analysis data, the complexes are thermally
stable compounds. One of the promising ways of regulating ballistic characteristics of
highly filled energetic formulations is the introduction of burn modifiers. Complex
compounds of d-metals are known to be effective modifiers of the process of burning.
Ballistic tests were made in a constant pressure bomb and demonstrated that the
prepared compounds are promising modifiers of the burn rate and can be used in the
developed formulations of rocket propellants.
Keywords:
complex perchlorates, 1,5-pentamethylenetetrazole, modifiers of
burning, derivatographic analysis, drop hammer test, ballistic test, bomb of constant
pressure.
INTRODUCTION
1.
The report is devoted to the synthesis and characterization of coordination compounds in

the series of perchlorates of d-metals containing substituted tetrazole as ligand. Considerable
attention to this class of energetic materials is caused by their application in safe detonators
and rocket propellants.
2.
DISCUSSION
2.1
Synthesis and properties of the perchlorate complexes
We have carried out the synthesis of new energetic d-metal complex perchlorates with
1,5-pentamethylenetetrazole (PMT) as ligand.
Mikhail Ilyushin, Igor Tselinsky, Irina Bachurina, Yurii Gruzdev, . . .
143
Czech Republic
CH2
CH2
CH2
CH2
2006
[Content]
N
N
H2C
1,5-Pentamethylenetetrazole (PMT)
Preparation of metal complexes was performed according to the following reactions:
For Cu(II) complexes
Cu(H3COO)2 + n(CH2)5CN4 + 2HClO4 [Cu((CH2)5CN4)n](ClO4)2 + 2H3COOH
n = 2, 4
For Co(III) complex
[Co(NH3)5H2O]3+(ClO4-)3 + P {Co(NH3)5[P]}3+(ClO4-)3 + H2O
The following compounds were prepared and investigated: Cu(PMT)4(C1O4)2 (1),
Co(NH3)5(PMT)(ClO4)3 (2). The composition and structure of complexes 1 and 2 were
supported by the results of elemental analyses; IR-, UV-, and NMR(H1)-spectroscopy. The
results of thermochemical studies of complex perchlorates by means of derivatographic
method are presented in Tables 1 and 2.
Table 1. Derivatographic analysis of complex 1
Products of
thermal decomposition
[Cu((CH2)5CN4)4] (ClO4)2
T 0C
Effect
30-175
Endo
Residue, %
Found Calculated
100
-
Cu ((CH2)5CN4))3 (ClO4)
175-240
Exo
29,3
29,6
Cu((CH2)5CN4))2
240-305
Exo
29,3
29,6
CuCxNy
350
40,9
Table 2. Derivatographic analysis of complex 2

Products of thermal decomposition T 0C
144
Residue, %
Found Calcul-ated
100
[Co(NH3)5(C6H10N4)](ClO4)3
60-245
(max 243)
Co(C6H10N4)(ClO4)3
245-360
85.0
(max 275, 295, 320)
85.3
1/2Co2O3
360-500
14.3
14.0
Preparation, Properties and Application of Energetic Complex Perchlorates of . . .
Czech Republic
2006
[Content]
The data listed in Tables 1 and 2 shows that complexes 1 and 2 are decomposed in
several steps. Decomposition of the complexes begins when temperature rises above 200 0C.
Removal of PMT molecule is the first step of decomposition of perchlorate complex 1. For
complex 2 during the first stage the process of inactivation and removal of NH3 molecules
from inner sphere is observed. Subsequently the thermal decomposition of the tetrazole
ligand occurs with the involvement of the perchlorate anion into oxidation reactions at the
late stages of the complex destruction. According to the data of derivatographic analysis, the
complexes are thermally stable compounds.
The drop hammer test of impact sensitivity according to GOST 4545-88 demonstrated
that the sensitivity to impact of perchlorate complex 2 is similar to that of PETN. Sensitivity
to impact of copper complex 1 was at the level of tetryl. Complex 2 has a crystal density
near 1,82 g/cm3. Sensitivity to impact of complex 2 (drop hammer -44-II, mass of hammer
2 kg, H=25 cm, % of explosion) is 16% (the impact sensitivity of PETN is 12% under
similar conditions). Complex 2 has a short distance of deflagration-to-detonation transition
and can initiate charges of RDX in blasting cap No 8. Minimal charge of complex 2 for
blasting RDX is near 0.4 g under this condition [1]. Detonation velocity of complex 2 at
maximum density is near 6.98 mm*s-1. (Detonation velocity calculated by method of A.A.
Kotomin [2]). Consequently, complex 2 may serve as an initiating material with the
detonation velocity close to that of TNT and sensitivity to mechanical stimuli like other
modern high explosives. Perchlorate complex 1 does not initiate HE.
200
100
80
60
Burn rate, m m /s
40
20
10
8
6
4
1
0.1
0.2
0.4
0.6 0.8 1
8 10
20
Pressure, MPa
Fig 1.
The rate of deflagration of complex 2 as function of pressure
The burn rates of perchlorate complexes 1 and 2 in the bomb of constant pressure (BP
400) were investigated. Charges of the perchlorate complexes were pressed into
polymethylmethacrylate tubes with the internal and external diameters equal to 4 and 7 mm,
respectively. Photoregister FR-14 was used for measuring the burn rates of complex 2. Fig. 1
demonstrates the rate of deflagration of complex 2.
145
Czech Republic
2006
[Content]
The pressure exponent in the equation for the burn rate of perchlorate complex 2 ()
equals to 0,874. This value is a little lower than the pressure exponent in equation for the
burn rate of hexaammine cobalt (III) perchlorate ([Co(NH3)6](ClO4)3) (=1) and
aquapentaammine cobalt (III) perchlorate ([Co(NH3)5(H2O)](ClO4)3) (=1) which have the
structures analogous to that of complex 2 [3]. It is noteworthy that perchlorate complex 2
burns partly in the condensed phase. This effect may be the result of presence of CH2groups as a fuel in the ligand.
Complex 1 has a lower burn rate. Digital camera JVC GR-DVL 9800 was used to
measure burn rates of complex.1. Fig. 2 shows the rate of deflagration of complex 1.
100
80
60
40
Burn rate, mm/s
20
10
8
6
1
0.1
0.2
0.4
0.6 0.8 1
8 10
20
Pressure, MPa
Fig 2.
The rate of deflagration of complex 1 as function of pressure.
The lower pressure limit of self-sustaining burning of perchlorate complex 1 is 0.5 MPa.
In the interval from 0.5 MPa to 2 MPa the pressure exponent in the equation of burning ()
equals to 1.089 and shows that the process of burning takes place entirely in the gas phase.
The line illustrating the dependence of the burn rate on pressure for values greater than 2
MPa changes the slope ( =0,383) because of the change in the rate limiting stage. As the
pressure rises, the processes in the condensed phase become dominant.
Experimental values of burn rate were used to calculate the burn temperatures and
composition of burning products for complexes 1 and 2 at the studied pressures . The
operation was carried out by means of REAL computer program. The program was
developed by G.V. Belov for computer simulating of complex chemical reactions taking
place at high temperatures and pressures.
The experemental formulae of perchlorate complexes (Cu1C24H40N16Cl2O8 (1) and
Co1N9H25C6Cl3O12 (2)) and the values of enthalpies of formation (f0 = 439 kJ/kg for
complex 1, f0= - 883 kJ/kg for complex 2) were used to perform calculations. The
146
Czech Republic
2006
[Content]
calculated burn temperatures and composition of burning products of perchlorate complexes

at pressure 10 MPa are shown in Tables 3-5.
Table 3. The calculated burn temperatures for perchlorate complexes 1 and 2
Complex
[Cu((CH2)5CN4)4] (ClO4)2
[Co(NH3)5(C6H10N4)](ClO4)3
Burn temperature, K
1451
2329
Table 4. Table 4.The calculated composition of principal burning products of

perchlorate complex 1
Burning product
C(condensed)
Cu(condensed)
HCl
N2
H2
CH4
CO
Concentration, mol/kg
18.270
1.218
2.444
9.788
17.459
2.436
8.534
Trace amounts of the following burning products: CO2 -0.184 mol/kg, H2O-0.910 mol/kg, HCN0.013 mol/kg, NH3 -0.040 mol/kg are shown to be present
Table 5. The calculated composition of principal burning products of perchlorate

complex 2
Burning product
Co(condensed)
HCl
N2
H2O
H2
CO
CO2
Concentration, mol/kg
1.468
4.684
7.749
8.957
10.206
8.964
1.369
Trace amounts of the following burning products: NH3 0.003 mol/kg, CoCl2 0.235 mol/kg are
shown to be present
The above regularities in the calculated composition of main burning products of

perchlorate complexes 1 and 2 may be the result of strongly negative oxygen balance of the
salts (the oxygen balance of perchlorate complex 1 equals to -117.77 %, the oxygen balance
of perchlorate complex 2 equals to - 33.07 %)
2.2
Perchlorate complexes as the modifiers of burning processes in

solid rocket propellants.
One of the promising ways of regulating ballistic characteristics of highly filled

energetic formulations is the addition of burn modifiers. Complex compounds of d-metals
are effective modifiers of the burning [4]. An energetic copolymer of 2-methyl-5-
147
Czech Republic
2006
[Content]
vinyltetrazole with methacrylic acid (PVMT) (Technical Specification 38-403-208-88) was

used as a high-molecular binder for a model solid rocket propellant formulation.
-CH-CH 2| N
N
CH 3
CH 3
|
-CH 2-C|
COOH
PVMT
A liquid aliphatic azide, 1,5-diazido-3-nitro-3-azapentane (3) was used as active
plastisizer of the polymer base (Technical Specification 130176)[4,5]..
N3-(CH2)2-N-(CH2)2-N3
|
NO2
1,5-Diazido-3-nitro-3-azapentane
Plastisizer 3 has the density of to 1.344 g/cm3, and value of the enthalpy of formation
f, = 3025 kJ/kg
Model formulations had the following composition: ammonium perchlorate of bimodal
granulometric composition:grade (<50 m) - 32%; :grade A according to Technical
Specification -84-942-82 (Sspecific = 12000 cm2/g) - 32%, fuel - aluminum powder of grade
ACD - 6 (Sspecific = 6000 cm2/g) - 6%, binder and plastisizer - 28%, additions of the
complexes were introduced into the formulations in amount of 2%. Ballistic tests were made
in a constant pressure bomb. Results of ballistic tests are presented in Table 6.
Table 6. Effect of additions of perchlorate complexes 1 and 2 on the burning of a
model propellant
Index of formulation
Basic formulation
[Cu((CH2)5CN4)4))](ClO4)2
[Co(NH3)5((CH2)5CN4)](ClO4)3
Equation of deflagration
U=AP
11,1
0,71
16,1
0,56
10,5
0,60
U10,
mm/s
47,2
58,4
41,8
U4,
mm/s
26,6
35,0
24,1
U10- rate of burning at 10 MPa pressure;

U4- rate of burning at 4 MPa pressure;
The results obtained permit to make the following conclusions:

-The pressure exponent of burn rate law in the presence of additions decreases from 0,71
to 0,60-0,56 and this is an indirect confirmation of the fact that these additions affect the
burning processes in the condensed phase.
-omplex perchlorate 1 increases the burn rate of a model propellant by 35.5%.
148
Czech Republic
2006
[Content]
Therefore, the prepared energetic complexes are promising modifiers of the burn rate
and can be used in the development of solid rocket propellant formulations.
ACKNOWLEDGMENTS
This work was performed under financial support of RF Ministry of Education and
Science via scientific and technical program Development of scientific potential of higher
school, Subprogram II Applied research and developments of priority directions of
science and engineering, Project 0120.0504746 and State Contract
02.444.11.7051(RI-111.0/001/028)
REFERENCES
[1] M.A.Ilyushin, A.Yu. Zhilin, I.V.Tselinsky, Yu.A. Nikitina, A.S.Kozlov and I.V Shugalei.
Energetic ammine cobalt (III) perchlorates with cyclopentamethilentetrazole as ligand. //Russian
Journal of Applied Chemistry. 78. No.2. 195-199 (2005) (in Russian).
[2] M.A.Ilyushin, A.V.Smirnov, A.A.Kotomin and I.V Tselinsky. Coordination complexes as
inorganic explosives for initiation systems //Hanneng Cailiao=Energetic materials. 2. No 1. 1620.(1994)
[3] A.E. Fogelzang, V.Yu. Egorshev, V.P.Sinditsky,.V.V. Serushkin, A.R.Saclanty. A study of
burning behavior of chemical analogs of explosive CP// Proceedings of the 18-th International
Pyrotechnics Seminar. Chicago. Illinois. 13-15 July 1992. 1992. P. 303- 320.
[4] A.S. Dudyrev, I.V. Tselinsky and M.A. Ilyushin Chapter 6.Technology of energetic materials and
products on their base in Chemical technologies, edited by P.D. Sarkisov. Ministry of Education of
Russia. D.I.Mendeleev Russian University of Chemical Technology. Moscow. 2003. pp. 403-488. (in
Russian).
[5] M.A.Ilyushin, I.V.Tselinsky, I.V.Bachrina, L.O.Novocelova, E.N.Konyushenko, A.S.Kozlov ,
Yu.A.Gruzdev. Application of energy-saturated complex perchlorate.//International Conference
Zababakhin scientific talks. September 5-10.2005. Snezhinsk. Chelyabinsk Region. Russia.2005.
p.83-84.
149
Czech Republic
2006
[Content]
BALISTIC STABILITY OF A SPHERICAL PROPELLANT:

COMPARISON WITH A FLATTENED SPHERICAL PROPELLANT.
L. Jeunieau, M. H. Lefebvre,
Ecole Royale Militaire, Avenue de la Renaissance 30, 1000 Bruxelles (BEL)
e-mail: laurence.jeunieau@rma.ac.be
P. Guillaume
PB Clermont SA, Rue de Clermont 176, 4480 Engis (BEL)
_____________________________________________________________________
Abstract
The physical ageing of a spherical deterred propellant has been monitored by infra-red
microscopy, closed vessel tests and ballistic firing. This result has been compared with the
ageing of a flattened spherical propellant; it has been observed that the ageing of the
spherical propellant is far less important.
Keywords: spherical propellant, flattened propellant, deterrent, ageing
_____________________________________________________________________
INTRODUCTION
Rolled ball propellants are used widely in small calibre weapons. They are easy to
manufacture and their properties can be specially adjusted for different ammunition types by
chemically modifying their surface composition. This surface modification usually consists
of an impregnation by a deterrent, which is only present along the surface up to a certain
depth and with a certain gradient. This deterrent gradient gives a nearly constant pressure
generation during the combustion process, compensating the reduction of surface by an
increase of the burning rate with time.
The disadvantage of this special design is the possible migration of the deterrent
throughout the propellant. In this case, the burning rate increases at the beginning of the
burning process, which leads to higher pressures in the weapon. This can damage the
weapon or give rise to malfunctions of the ammunition.
The service life of propellants consists of two components: the chemical shelf life and
the functional or ballistic shelf life. The chemical shelf life covers the period of time during
which the propellant can be safely stored without representing a hazard to its environment.
This one is limited by the extent of chemical ageing reactions, such as decomposition of
nitric esters and reactions of the decomposition products with the stabilizer. The functional
life or ballistic shelf life is the period of time during which the propellant or ammunition can
be used safely or during which the internal requirements remain fulfilled. In this work only
the ballistic shelf life has been investigated.
Previously, we have extensively investigated the ageing of a flattened spherical
propellant [1-4]. In this work, the ageing of a spherical propellant has been studied. To ease
the comparison between the results, the same conditions of ageing have been used at 80C,
70C and 65C.
150
Balistic Stability of a Spherical Propellant: Comparison with a Flattened . . .
Czech Republic
2006
[Content]
EXPERIMENTAL
2.1 Samples
The propellant, which is used for the investigation is the double base ball powder
propellant SPH27. It is a spherical propellant with an average diameter of 756 m. It
contains around 10% of nitroglycerin. It is stabilized with DPA. The deterrent applied on
the surface is dibutylphthalate (DBP). The average concentration of DBP is 7.4%.
The flattened propellant K5810 which has been previously studied can be
geometrically modelled by an ellipsoid with a great axis of 660 m and two small axis of
360m. It contains 10% of nitroglycerin and 4.6% of DBP.
The two types of propellant differ by their geometry and their DBP concentration.
2.2 Experiments and work program
Three different areas of investigation are covered: migration analysis, closed vessel
tests and ballistic firing.
Infrared microscopy
Migration of the deterrent was made visible by IR microscopy: the propellant grain is
placed into an adhesive, cut by a microtome into small slices (thickness 7m) and analyzed
by IR microscopy. A Bruker Hyperion infrared microscope mounted on a Vector 33 Fourier
transform spectrometer is used in this study. A medium-band MCT (HgCdTe) detector in
the microscope gives high sensitivity in the 4000-600cm-1 range. A 15X cassegrain mirror
objective is used to obtain the infrared spectra. The IR spectrometer is operated at a
resolution of 4cm-1 and 32 scans are acquired for each measurement posision. The
measuring window has a aperture of 10m x 50m, the great side of the aperture is placed
perpendicularly to the measured diameter. The step for each data point is 3m. The DBP
concentration is obtained using the ratio of two IR bands, one typical of the DBP (1720cm-1),
one typical of the nitrocellulose (1160cm-1).
Closed vessel tests
Closed vessel tests experiments were carried out in a vessel of 132cm3 using a
piezoelectric transducer (Kistler 6215) coupled to a charge amplifier (Kistler 5011A) for
recording the pressure. The output voltage of the pressure gauge was transferred to a data
recording acquisition system (Nicolet Multipro, resolution 12 bit, sampling frequency 250
KHz). The ignition system consists of two electrodes, which are connected with a nickeline
hot wire. The vessel is equipped with a valve to introduce the gaseous ignition mixture. The
partial pressures of this mixture are measured with a piezoelectric transducer (Kistler 4070).
The composition of the used ignition mixture is 1bar CH4 1.4 bar O2. For the actual
composition the initial atmospheric pressure must be added. The used loading density is
0.15g/cm3.
Laurence Jeunieau, Michel Lefebvre, Pierre Guillaume
151
Czech Republic
2006
[Content]
Ballistic firing
For the spherical propellant, the experimental work is carried out in a .50inch weapon
instrumented with a piezoelectric transducer (Kistler 6215) coupled to a charge amplifier
(Kistler 5011A) in order to record the pressure. The pressure measurement is located in the
cartridge case. The output charge of the pressure is amplified and filtered to get an output
voltage which is transferred to a data acquisition system (resolution 12bit, sampling rate
1MHz). 14g of the propellant are poured in a standard 12.7 x 99 mm cartridge. The
corresponding loading density is 0.79g/cm3.
The ballistic firings for the flattened propellant have been performed in a 5.56
calibre, for more details see [3].
Propellant ageing
The ageing times at 80C are nearly identical for the flattened and the spherical
propellant.
The ageing times of the spherical propellant are presented in Table 1. Throughout
this paper S stands for the spherical propellant ageing. The two numbers behind the first
letter describe the ageing temperature whereas the last letter describes the ageing state (see
table 1). By example, S80B corresponds to a propellant aged at 80C during 1.81 days. For
the ballistic firing, the aged propellant is poured into the cartridge.
The ageing times for the flattened propellant are presented in table 2. For these
propellant K stands for the flattened propellant ageing.
T
[C]
80
70
65
Table 1: Ageing plan of the spherical propellant.

C
D
E
F
G
H
1.20
1.81
2.28
9.82
5.53 7.22
21.74
9.02
10.91 20
J
15.73
29.8
T
[C]
80
70
65
Table 2: Ageing plan of the flattened propellant.

B
C
D
E
F
G
1.20
4.83
1.81
7.25
2.28
9.66
3.61 5.41 7.22 9.02 10.83

14.50 21.74 29.00 36.23 43.48
30.00
3. RESULTS
3.1 Deterrent migration
Figure 1 shows some deterrent concentration profiles of the spherical propellant aged
at 80C. It can be observed that the deterrent concentration decreases in surface with ageing
and increases in the centre part of the propellant with ageing.
152
Czech Republic
[Content]
w ithout ageing
S80B
S80C
S80E
S80H
S80J
14
12
10
%DBP
2006
8
6
4
2
0
0
50
100
150
Figure 1: Deterrent concentration profiles of DBP in the spherical propellant aged at 80C.
Figure 2 shows the deterrent concentrations profiles that have been previously
obtained for the flattened spherical propellant for the same ageing time. It can be observed
that the variation of the deterrent concentration profiles is far more important for the
flattened spherical propellant than for the spherical propellant.
14
w ithout ageing
K80B
K80C
K80E
K80H
12
%DBP
10
8
6
4
2
0
0
50
100
150
Figure 2 : Deterrent concentration profiles of DBP in the flattened spherical deterred

propellant.
3.2 Closed vessel tests
The two combustion phases resulting for the chemical composition gradient of a
deterred propellant can be observed in closed vessel tests if an appropriate gaseous ignition
system is used. The influence of the ignition system on the obtained combustion rate has
been extensively studied elsewhere [5-6]. Figure 3 shows the obtained combustion rate of
the spherical propellant before ageing and the corresponding derivative of the pressure
versus time. The pressure derivative permits to distinguish clearly two phases in the
combustion process, as two different slopes can be observed during the combustion process.
4
153
Czech Republic
20
2006
[Content]
100
dP/dt (Mpa/ms)
combustion rate (cm/s)
15
10
5
80
60
40
20
0
0
0
50
100
Pressure (MPa)
150
-5
0
time (ms)
Figure 3: Combustion rate of the spherical propellant without ageing and the corresponding
derivative of the pressure versus time
Figure 4 shows the obtained combustion rate of some of the samples aged at 80C. It
can be observed a decrease of the combustion rate for the high pressure value. In fact, an
increase of the combustion rate pressure for the low pressure value is also present but is not
visible in this figure.
For this reason, figure 5 shows the variation of the combustion rate at 20MPa and at
100MPa as a function of ageing time.
These variations can be related to the propellant migration: the deterrent concentration
decreases near the surface which is correlated with the increase of the combustion rate at the
low pressure values and the concentration increases in the centre part of the propellant;
which is correlated with the decrease of the combustion rate at the high pressure values.
20
Combustion rate (cm/s)
18
16
14
12
10
without ageing
8
6
S80J
S80B, S80C,
S80E, S80H
2
0
0
20
40
60
80
Pressure (MPa)
100
120
140
Figure 4: Combustion rate of the spherical propellant aged at 80C for different ageing time.
154
2.2
flattened
propellant
spherical
propellant
1.8
1.6
1.4
1.2
1
0
10
20
Czech Republic
Combustion rate @ 100MPa (cm/s)
Combustion rate @ 20 MPa (cm/s).
2006
[Content]
flattened
propellant
12.5
12
11.5
11
10.5
10
9.5
9
8.5
8
spherical
propellant
ageing time @ 80C (day)
10
20
ageing time @ 80C (days)
Figure 5: Variation of the combustion rate at 20MPa and at 100MPas a function of ageing
time at 80C for the spherical and the flattened propellant.
Figure 5 shows the variation of the combustion rate at 20MPa and at 100MPa as a
function of ageing time for the spherical and the flattened propellant. It can be observed that
the variations are greater for the flattened propellant than for the spherical one. This can be
correlated to the more important deterrent migration in the flattened propellant. As the
variation of the deterrent concentration is greater with ageing, the variation of the
combustion rate is also more important.
3.3 Ballistic firing
Ballistic experiments were performed with aged and unaged cartridge. The pressure in
the cartridge and the velocity of the bullet have been determined. No variation of the bullet
velocity at the muzzle position have been observed. This shows that the ageing time has not
modified significantly the energy content of the propellant.
Figure 6 shows the pressure recorded in the cartridge for a double base propellant and a
deterred propellant at a same loading density. When the propellant begins to burn, the
pressure increases. When a threshold pressure is reach, the force applied on the projectile is
enough to engrave the rotating band. The movement of the projectile down the gun bore
causes the volume of the reaction chamber to increase. As the propellant continue to burn,
the pressure still increases, but this increase is slower as the volume increases. After a
certain time the increase of the volume has a more important effect than the increase of the
gas mol number and the pressure decreases. It can be observed that for a deterred propellant
the pressure increase is more progressive and the pressure stay at its maximal value for a
greater time than for a not deterred propellant. This is due to the presence of the deterrent in
surface. At the beginning of the combustion, the deterred part of the propellant burnt, the
burning rate is low but the propellant surface is high, when the deterred part of the propellant
is burnt the combustion rate is higher but the propellant surface is smaller and the gas
production stay more or less constant. For a not deterred propellant the combustion rate is
high at the beginning of the combustion and as the propellant surface is important, the
pressure increase is rapid and by consequence the maximal pressure in the cartridge is high.
155
Czech Republic
2006
[Content]
350
double base propellant
Pressure (MPa)
300
deterred
propellant
250
200
150
100
50
0
0
time (ms)
Figure 6: Variation of the pressure in the cartridge as a function of time for a double base
propellant and a deterred propellant in a .50inch weapon for a loading density of 0.6g/cm3.
Figure 7 shows the pressure recorded in the cartridge for different aged propellants. It
can be observed that the maximal pressure increases with ageing and that the pressure peak
for an aged propellant has a shape which resembles more to the one of a not deterred
propellant. Due to deterrent migration, for an aged propellant the transition between the
deterred and the not deterred part of the propellant is smoother. By consequence, the
pressure increases is more rapid at the beginning of the combustion and the maximal
pressure is higher.
300
Pressure (MPa)
250
without ageing
200
150
S80E and
S80H
100
S80B
S80C
50
0
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
tim e (m s)
Figure 7: Variation of the pressure in the cartridge for the spherical propellant at different
ageing time.
The same type of variation has been obtained for the flattened propellant (fig.8) but
in this case the pressure increase is more important for the same ageing time. This greater
variation is emphasized by fig.9 which shows the variation of the maximal pressure as a
function of ageing time for the two types of propellant.
156
Czech Republic
2006
[Content]
300
Pressure (MPa)
250
without ageing
200
150
S80E and
S80H
100
S80B
S80C
50
0
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
tim e (m s)
Maximal pressure in the cartridge

(MPa)
Figure 8: Variation of the pressure in the cartridge for the flattened spherical propellant at
different ageing time.
flattened propellant
500
spherical propellant
450
400
350
300
250
200
150
0
10
15
20
ageint time @ 80C (day)
25
Figure 9: Variation of the maximal pressure in the cartridge as a function of ageing time at
80C for a flattened and a spherical propellant.
3.4 Correlation between the different experiments

3.4.1 Correlation between infrared measurement and closed vessel tests
The combustion rate at 20MPa corresponds to the beginning of the combustion and
can be related to the deterrent concentration in surface. In fig.10, the combustion rate at
20MPa has been plotted as a function of the deterrent concentration in surface for the
flattened and the spherical propellant. It can be observed that the combustion rate decreases
with the deterrent concentration and that a single linear regression can be drawn for the
flattened and the spherical propellant. This shows that the combustion rate at 20MPa
depends only of the deterrent concentration in surface and that the geometrical difference
between the two propellants is appropriately treated in the combustion rate determination.
157
combustion rate @ 20MPa
Czech Republic
2006
[Content]
2.5
2
1.5
1
0.5
0
0
10
%DBP @ 0m
15
Figure 10: Variation of the combustion rate at 20MPa as a function of the DBP
concentration in surface for the flattened and the spherical propellant.
3.4.2 Correlation between infrared microscopy and ballistic firing
Figure 11 shows the variation of the maximal pressure measured in the cartridge as a
function of the DBP concentration at the outer surface of the propellant grain for the
flattened and spherical propellant. It can be observed that the maximal pressure in the
cartridge decreases with the DBP concentration in surface. This confirms that the variation
of the maximal pressure in the cartridge depends on the deterrent migration.
Maximal pressure in the

cartridge (MPa)
The apparent difference between the two types of propellant is due to the different
uses: the flattened propellant is used in a 5.56 mm cartridge and the spherical propellant is
for a .50inch cartridge.
500
450
400
350
300
250
200
0
10
%DBP @ 0m
15
Figure 11: Variation of the maximal pressure in the cartridge as a function of the DBP
concentration in surface for the flattened and spherical propellant.
4
CONCLUSIONS
The ageing of a spherical deterred propellant has been monitored by infrared

microscopy, closed vessel tests and ballistic firing. These results have been compared to the
one obtained for a flattened spherical propellant. It has been observed that the effect of
ageing is more important for the flattened propellant than for the spherical. This can be due
to the effect of stress during the flattening operation. Some crack may appear in the
propellant during this operation. A microscopic observation should confirm or infirm this
hypothesis.
A good correlation has been observed between the combustion rate at 20MPa and the
deterrent concentration in surface and between the maximal pressure in the cartridge and the
deterrent concentration in surface.
158
Czech Republic
2006
[Content]
BIBLIOGRAPHY
[1] S.Wilker, P. Guillaume, M.H. Lefebvre, S. Chevalier, L. Jeunieau, Ballistical and
chemical stability of rolled ball propellants, 7 Seminar New Trends in Research of
Energetic Materials , Pardubice, April 2004.
[2] S. Wilker, P. Guillaume, M.H. Lefebvre, S. Chevalier, L. Jeunieau, G. Pantel, U.
Ticmanis, T. Stottmeister, Stability analyses of rolled ball propellants Part I :
Microcalorimetric studies and stabilizer deplation, 34th International Annual Conference of
ICT, Karlsruhe, June, 2003.
[3] L. Jeunieau, M.H. Lefebvre, P. Guillaume, S. Wilker, S. Chevalier, S. Eibl, Stability
analyses of rolled ball propellants. Part II- Ballistic stability, 35rd International Annual
Conference of ICT, Karlsruhe, June, 2004.
[4] L. Jeunieau, M.H. Lefebvre, P. Guillaume, S. Wilker, S. Chevalier, Stabillity analyses of
rolled ball propellants Part III- Correlation between closed vessel tests, ballistic firing and
deterrent migration, 36th International Annual Conference of ICT & 32rd International
Pyrotechnics Seminar, June, 2005
[5] L. Jeunieau, M.H. Lefebvre, A. Papy, M.C. Pirlot, C. Reynaud, Closed vessel test:
influence of the ignition method on the combustion rate, 33rd International Annual
Conference of ICT, Karlsruhe, June 2002.
[6] L. Jeunieau, M.H. Lefebvre, M.C. Pirlot, P. Guillaume, Characterization of deterred
propellants by closed vessel tests: importance of the ignition method, 8th Seminar New
Trend in Research of Energetic Materials, Pardubice, April 2005.
10
159
Czech Republic
2006
[Content]
SYNTHESIS AND CHARACTERIZATION OF

1,2-DINITROGUANIDINE (DNG) AND ITS DERIVATIVES
Nikolaj V. Latypov, Martin Johansson, Larisa N. Yudina-Wahlstrm, Stefan Ek,
Carina Eldster and Patrick Goede
FOI, Swedish Defence Research Agency, Department of Energetic Materials,
S-147 25 Tumba, Sweden
Abstract:
1,2-Dinitroguanidine (DNG), (1), is a novel explosive with high performance and
moderate sensitivity1,2. Synthesis of DNG and its salts has recently been reported3. X-ray
studies of the crystal structure of DNG can also be found in the literature4. In this paper
we present complementary studies on the synthesis, reactivity and thermal
characteristics of this compound and some of its derivatives.
Keywords:
1.
1,2-Dinitroguanidine, DNG, synthesis, characterization
INTRODUCTION
N,N-Dinitroguanidines have been known compounds for a rather long time5. Despite the
extensive studies6,7, which resulted in the synthesis of several interesting substances in this
series, the un-substituted 1,2-dinitroguanidine itself had never been reported until recently
when different groups of Russian researchers presented simultaneously both the synthetic
approach3 and basic properties of the compound3,4.
Judging by the molecular composition of DNG, which is an oxidizer (+5,4%) one can
treat the substance and its derivatives as attractive ingredients in both explosive and
propellant formulations, - surprisingly, very little information about their properties as
energetic materials has so far been published1,2.
For this reason we have performed a preliminary estimate of the most important
properties of these compounds.
160
Synthesis and Characterization of 1,2-Dinitroguanidine (DNG) and its . . .
2.
Czech Republic
2006
[Content]

2.1 Synthesis
1,2-dinitroguanidine (DNG) was easily synthesised by the reported method3 with
some minor modifications: higher content of nitric acid in the nitration mixture, in
comparison with the published procedure, was used to ensure a controllable reaction
course. In addition to the observations made by Astratev et al.3, showing that the
nitration of mono-nitro guanidine (MNG) to DNG is a slow process taking place in a
very reactive nitrating media (figure1), we
Fig 1.
UVspectra of the reaction mixtures for the nitration of MNG to DNG;
nitrating mixture HNO3: oleum (65% SO3) = 0, 55: 0, 45.
7
6
Absorbency
5
4
3
2
1
0
-1
150
200
1__after 2,5 hrs
1__after 6 hrs
1__after 20 hrs
250
300
350
400
450
Wavelength (nm)
have also noticed the instability of DNG in such mixtures so that its concentration is
strongly dependent on the reaction time (figure 2) and the composition of the nitrating
media. The yield of DNG has never exceeded 80%.
Nikolaj Latypov, Martin Johansson, Larisa Yudina-Wahlstrm, Stefan Ek, . . .
161
Czech Republic
2006
[Content]
Fig 2.
Formation of DNG by nitration of MNG in different nitration mixtures;
dependence of the yield on the reaction time.
%
80
1
2
60
1__ HNO3 : oleum (65% SO3) = 1,0:1,5

2__ HNO3 : oleum (65% SO3) = 1,0:0,9
3__ HNO3 : oleum (30% SO3) = 1,0:1,0
40
20
12
16
20
24
72
Hrs
For this reason, the reaction course was in all cases monitored by UV spectroscopy to
detect the full conversion of MNG to DNG and the maximum concentration of the
latter. Despite these efforts, the product separated by dilution of the nitration
mixtures and extraction by appropriate solvent always contained small amounts (25%) of MNG.
All these findings can be easily understood in view of the known fact that N-nitration
is a reversible process and nitramines are prone to irreversible decomposition in
strongly acidic media8 (Scheme1).
Scheme 1.
Reaction sequence in the nitration of MNG to DNG.
NH
H2N
NO2+, -H+
NH2
-NO2+, H+
H2N
NO2
NO2+, -H+
NH2
-NO2+, H+
H+
O2N
N
H
NO2
NH2
H+
N2O + (NH2)2CO + H2O
2.2
Physicochemical properties.
Samples of DNG produced by this method needed further purification which was
achieved by dissolution of the raw product in diisopropylether, followed by removal of the
insoluble MNG by filtration. The purified product was identified by elemental analysis and
mass-spectroscopy; its spectral characteristics (UV, NMR) were identical with those
published in the literature3.
162
Czech Republic
2006
[Content]
DNG is indeed a very acidic product: our measurements resulted in the pKa value of
0.95, which agrees reasonably good with the figure (1.11), reported by Astratev et al.3 The
density of crystalline DNG (1.895), estimated by pycnometry in the present work, is also
rather close to that (1.884) obtained from X-ray measurements4. Salts of DNG were easily
obtained by careful neutralisation of its solutions in an alcohol with an appropriate base. The
properties of DNG and some of its salts are summarized in Table1.
Table 1. Properties of DNG and its salts.

Density
3
g/cm
DNG
H f,
Sensitivity
Sensitivity
kcal/mol
Friction, N
Impact, cm/J
90
22 / 4,5
1.906*/
0,0 ***
1.884**
ADNG
1,86*
- 36,60 *
250
78 / 14,2
K salt
n.m.
n.m.
130
28 / 7,3
ADN
1,805
-34,5
153
12 / 3,0
RDX
1,802
14,7
195
32 / 6,4
- present work
** - reference [4]
*** - reference [2]
Unlike alkylation and substitution reactions3, the reactivity of DNG, or , strictly
speaking, its anion in reactions of the Henry and Michael type is quite low (cf. Table 2); this
is in line with the behaviour of some anions of primary nitramines, which undergo such
reactions only with difficulty9,10.
Table 2. Attempted reactions of DNG.
Michael donor Michael acceptor
DNG, NaDNG
Methyl acrylate
DNG, NaDNG
Acrylic acid
DNG
Methyl vinyl ketone
NaDNG
Methyl vinyl ketone
DNG
Acrylonitrile
NaDNG
Acrylonitrile
Temperature
RT, reflux
RT, Reflux
RT
RT
RT
RT
Result
Denitration
Denitration
Denitration
Addition
Denitration
Product mixture
Several N-alkyl-N, N-dinitroguanidines has been also synthesized by nitration6,7of

corresponding mono-nitro-compounds, which in turn were obtained by the known method 11.
Of these, iso -butyl-DNG showed interesting properties as a potential plasticizer. (fig.3)
163
Czech Republic
2006
[Content]
Fig 3. DSC thermogram showing the low temperature properties of i-butyl-DNG.

^exo
!&SEK_iBuDN G
SEK_iBuDNG, 20,0000 mg
Step
Onset
Inflect. Pt.
Ends et
Midpoint
-140
-120
-0,21 Wg ^- 1
-4,24 mW
-130,29 C
-125,71 C
-123,56 C
-127,03 C
-100
FOI: Vapen och Skydd.
2.3
0,5
Wg^-1
-80
-60
-40
-20
20 C
METTLER TOLEDO STARe SW 7. 01
Thermal stability
According to DSC measurements 1, 2-DNG has a rather good thermal stability and
decomposes without melting. Similar behaviour was observed for salts of DNG, though their
thermal stability varies significantly (cf. Figures 4-6).
Fig 4.
164
DSC thermogram of DNG and ADNG.
Czech Republic
Fig 5.
DSC thermogram of HDNG and LiDNG.
Fig 6.
DSC thermogram of NaDNG and KDNG.
2006
[Content]
These promising results prompted us to further investigate the long term stability of the
products. This was performed according to standard method (STANAG4582). As it can be
seen from fig. 7-8 DNG itself has excellent thermal stability. On the contrary, salts of DNG
(Fig.8-9) and iso -butyl-DNG (Fig.10) did not pass the test: in all cases violent, and, most
probably, autocatalytic decomposition was observed.
165
Czech Republic
2006
[Content]
Fig 7.
Heat flow produced by DNG in isothermal heat flow calorimetry
measurements at 75 C for 19 days. The signal is jagged due to large
amplification.
Fig 8. Heat flow evolved during isothermal heat flow calorimetry measurements on
different samples of ADNG at 75 C for 19 days.
166
Czech Republic
2006
[Content]
Fig 9.
Heat flow evolved during isothermal heat flow calorimetry measurements on
DNG and salts at 75 C for 19 days.
Fig 10. Heat flow evolved during isothermal heat flow calorimetry measurements on
iso-butyl-DNG at 75 C for 19 days.
These dissapointing properties , such as enormously high acidity of DNG itself and
unacceptable thermal stability of its derivatives made us to suspend further work on this
interesting group of compounds.
167
3.
Czech Republic
2006
[Content]
CONCLUSIONS
1,2-dinitroguanidine is an interesting energetic material; unfortunately, any

formulation work with this highly acidic compounds does not seem feasible. On the
contrary, this namely property favours synthesis of its derivatives with variable
characteristics.
Several compounds of this series, - salts of DNG and also iso-butyl-DNG- synthesized and
characterized in the present work, are not stable enough to find some application.
Further work in this area in search of more stable derivatives of DNG might be interesting.
Acknowledgements
The authors would like to thank the Swedish Defence Forces for financial support of this
research.
4.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
168
REFERENCES
V.I. Pepekin, L.M. Kostikova, G.T.Afanasev, Explosive properties of 1,2dinitroguanidine, 35th International Annual Conference of ICT, June29-July2, 2004.
V.I. Pepekin, G.T. Afanasev, Explosive properties of 1,2-dinitroguanidine and its
compositions with graphite Chimitcheskaya fyzicka, v.24, N4, p.p. 66-68, 2005(in Russian).
A. A. Astratev, D.V. Dashko, L.L. Kuznetsov, Synthesis and some properties of 1, 2dinitroguanidine, Russian Journal of Organic Chemistry, 2003, v.39, N4, p.p.537-548. (in
Russian ).
A. D. Vasilev, A.M. Astachov, M.S. Molokeev, L.A. Kruglyakova and R.S. Stepanov, 1, 2dinitroguanidine, Acta Crystallographica Section C, Crystal Structure Communications,
2003, C59, o550-o552
A.F.McKay Nitroguanidines, Chemical Reviews, 1952, 51, 301-346.
A.F.McKay and G.F.Wright, The nitration products of 2-nitramino-2-1,3diazacycloalkenes, JACS, 1948, v.70, p.p. 3990-3994.
R.H.Meen and G.F.Wright, The Scission of New 1-alkyl-1,3-dinitroguanidines and an
analog, JACS, 1952, v.74, p.p.2077-2079.
K.Schofield, Aromatic nitration, Cambridge University Press, 1980, part 15,N-Nitration
and Nitramines rearrangement and references cited therein.
L.W.Kissinger and M.Schwartz, Some Michael-Like Additions of Primary Nitramines,
JOC , 1958, v.23, p.p. 1342-1344.
David Woodcock, The reactions of Nitramines with Formaldehyde, J.Chem.Soc., 1949,
p.1635-1637.
Lawrence Fishbein and John A.Gallaghan The preparation and reactions of 2-Alkyl-1(or3)nitro-2-thiopseudourea, JACS, 1954, v.76, p.p.1877-1879.
Czech Republic
2006
[Content]
ENERGETIC CHARACTERISTICS OF SOLID COMPOSITE

PROPELLANTS AND WAYS FOR ENERGY INCREASING
Lempert D.B., Nechiporenko G.N., Manelis G.B.
Institute of Problems of Chemical Physics of Russian Academy of Sciences, (IPCP RAS),
Chernogolovka, Semenov av.1. Moscow reg. 142432. Russia
lempert@icp.ac.ru
Abstract: The creation of solid composite propellants is one of the most important
applications of high energetic materials. If in the very beginning of the development of
this branch (the middle of the 20th century) achieved energetic level increased
considerably every decade, up to 70-80 years it seamed that energy had already come to
maximal possible value, that is the power contained in chemical compounds (in
individual substances as well as in their mixtures) was already limited. Detailed
investigations, that have been carrying out in IPCP RAS, showed that there is a definite
potential for further energy increase though this potential is rather limited. The growth
of energy capacity of chemical substances is attended with the degradation of many
operating characteristics thermal stability decreases, impact sensitivity and friction
sensitivity increase, compability of compounds degrades, probability to transfer the
combustion process into explosion decreases, the cost increases drastically etc.
However, the increasing of propellants power is necessary therefore we have to agree to
a compromise energy growth against definite degradation of other performances.
Hereat investigations of thermal stability began to develop strongly, ways to increase
thermal stability of many substances (which could not be used in propellants earlier)
have been found, many other branches (polymer chemistry for new binders creation,
chemistry of inorganic and organic oxidizers, studying of combustion low etc) began to
develop.
In the presentation energetic characteristics of solid composite propellants of different
type and application are described metal-free compositions containing an oxidizer
and a fuel-binder only; compositions containing additionally an energetic compound
(Al, Be, Mg, B), metal hydrides (AlH3, BeH2, BxNyHz). A considerable attention is
attracted to analysis of oxidizers (ammonium perchlorate, hydroxylammonium
perchlorate, ammonium dinitramide etc) because oxidizer occupies the greatest mass
fraction in propellant and mostly that is the oxidizer that determines energetic
characteristics of the propellant. Solid composite propellant contains always a fuelbinder (for achievement of necessary physico-mechanical characteristics), therefore a
considerable attention is attracted to principles of the using of different kinds of binder.
It was shown that the competence to choose the binder (even among the set of accessible
ones) to given oxidizer can increase considerably the energetic characteristics even
without creation of new compounds. The ways to optimize solid composite propellants
destined for missile complexes of different mass characteristics are described in the
presentations. Different ways to further development of propellants energetic
characteristics are presented optimization of formulations basing on accessible
components; creation of new more effective binders, that have additional functions of
oxidizer or gasifying component etc. Ecological problems of solid composite propellants
are also a topic of the presentation, i.g. perchlorates outthrow a great amount of HCl,
that consumes greenery at large radius from the launch location. Berillium-containg
propellants are extremely dangerous for humanity, fluorine-containing propellants are
dangerous too. Ways to decrease ecological danger while solid composite propellants
using are discussed in the presentation.
David Lempert, G. Nechiporenko, G. Manelis
169
Czech Republic
2006
[Content]
The creation of solid composite propellants (SCP) is one of the most important
applications of high energetic materials. In the very beginning (the middle of the 20th
century) of the SCP development the achieved energetic level has been increasing
considerably every decade up to 70-80 years; however later it seamed that energy had
already come to the maximal possible value, that is the power contained in chemical
compounds (in individual substances as well as in their mixtures) was already limited. The
fact is that the growth of energy capacity of chemical material is attended with the
degradation of many operating characteristics thermal stability degrades, impact sensitivity
and friction sensitivity increase, compability of compounds degrades, probability of
combustion becoming a detonation decreases, the cost increases drastically etc. So, when we
try to increase energetic characteristics we must be ready to a compromise - energy growth
against definite degradation of other performances.
Rocket flight range and mass of the useful load depend mainly on specific impulse (Isp)
value (though many other parameters are influence too), and this dependence is very
considerable. For example, when the specific Isp is increased from 255 to 260, or 2%, useful
load is increased by 10% at the same flight range. Which is a significant achievement both
for military and non-military rockets.
Specific impulse, together with the content of combustion products is easily calculated
nowadays with computers if the brut-formula and enthalpy value of propellant, and the
pressure in the combustion chamber and nozzle section are known. When we compare
different types of propellant, we do not compare experimental parameters, but the theoretical
(i.e. calculated) ones, because experimental determination is very expensive (the sample
mass must be at least several kg) and even impossible when there is a lack of any compound.
The calculation of Isp is a regular procedure using current computers and necessary
software. It was shown that the difference between the calculated and real Isp values is 4-6
% and this difference depends much more on the type of rocket engine than on the type of
propellant.
BASIC CHEMICAL CONTENT AND BASIC COMPONENTS OF SCP.

The major chemical elements from which one can create a real SCP are H, O, F, C, N,
and Cl. The most important for thrust creation are H, O and F atoms. The value of specific
impulse definitely increases as the enthalpy of the composition or pressure in the combustion
chamber increases. The specific impulse usually also increases if the percentage of H, O or F
increases. However in real SCP one cant exclude C and N, because without them its
impossible to create molecules in solid state. It is very useful to take notice that a simple
presence of high energetic components in composition does not guarantee a high value of
Isp, the optimal ratio among components is always necessary.
Oxidizer is one of the most important components of SCP because it takes the greatest
mass fraction in propellant and mostly that is the oxidizer that determines energetic
characteristics of the propellant. Solid composite propellant contains always a fuel-binder
(for achievement of necessary physico-mechanical characteristics). This binder often plays
function of combustible when the heat of its oxidation by the oxidizer is the main source of
energy. The introduction of metal (Be, Al, B, sometimes Mg) into composition is one of
ways to increase the Isp value because of high combustion heat of these metals. There is
another kind of component, which properly speaking can be named neither oxidizer nor fuel
because these components play the both functions actually they may be named gas
170
Energetic Characteristics of Solid Composite Propellants and Ways for Energy . . .
Czech Republic
2006
[Content]
generating components. Cyclotetramethylene tetranitroamine (octogen or HMX) is the most

well-known and most used substance of this type.
REQUIREMENTS TO SOLID COMPOSITE PROPELLANTS, A

COMPROMISE AMONG BASIC PROPERTIES.
While in liquid-propellant rocket engines all components are kept in tanks and are
injected with pumps into the combustion chamber, solid-propellant rocket engines look
simpler - the propellant fills up almost entirely the whole interior of the rocket body and
burns from the surface. But this simplicity requires many specific properties for the
propellant. Since the propellant is subjected to much tension during acceleration - it has to be
both strong and elastic, it has to burn uniformly with a given rate, the combustion rate has
not to depend considerably on the pressure (in the formula U=A*P, where U is combustion
rate, P pressure, the exponent usually must be lower than 0.6), otherwise during the
combustion a sharp increase in pressure is possible and consequently the destruction of the
rocket body. It is necessary to prevent the possibility of combustion becoming a detonation.
Other restrictions include: low hygroscopicity, enough thermal stability, rather low impact
and friction sensitivity, not too high temperature in the combustion chamber (<3800-3900K)
and a lot of other restrictions. All these restrictions hamper developers of new more powerful
compositions. For example, a substitution of any hydrazonium salt (as oxidizer) for its
ammonium analogue would definitely increase the Isp value, however it would significantly
degrade thermal stability, impact sensitivity and compatibility of components. Introducing of
the most effective elements (beryllium, active fluorine) into the propellant formulation
damages the environment. Increasing of the enthalpy of formation increases the Isp
(Isp/Hf 0.06 s/(kcal/kg)), but the thermal stability usually degrades etc.
DENSITY AND BALLISTIC EFFECTIVENESS.

The density is one of the important properties determining ballistic effectiveness of
propellant. It happens often that density is in a compromise with specific impulse. As draft
estimation one uses some special functions for comparing ballistic characteristics of
propellants with different Isp and density values. One of these functions is a so-called
ballistic effectiveness (Efn) of the propellant on the nth stage of the multistage rocket
complex. Efn can be written as Efn =Isp*da, where d is the density, the value of the exponent
a depends on the mass characteristics of the rocket complex and the definite stage number
n. Usually for three-staged rocket complexes the values a are the following: ~0.6 for
the first stage, ~0.4 for the second, and ~0.2 for the third one. Thus, on the upper stages the
density has less influence on the ballistic effectiveness. For example if we have five different
compositions, where the first has Isp=260 with d=1.95; the second Isp=270 with d=1.8; the
third Isp=275 with d=1.7; the forth Isp=295 with d=1.0, and the fifth Isp=230 with
d=2.3, the best composition for the first stage is #1; for the second stage - #2; for the third
stage -#3. Though the composition #4 is of the highest Isp value its low density makes it less
effective than ## 1-3 on all stages. Thus, developing ballistic effectiveness of SCP, we have
to find on what stage of the rocket complex the SCP under investigation can be used with the
most effectiveness.
171
Czech Republic
2006
[Content]
THE SIMPLEST SCP BINDER+OXIDIZER

The set of oxidizers is not too large, we can divide them conditionally on three basic
categories: a) oxidizers free of halogen; b) perchlorates; and c)oxidizers containing active
fluorine.
Oxidizers of type a are ecological safe, of type c are very dangerous ecologically,
oxidizers of the type b occupy the intermediate position they are wide used though their
combustion product (hydrogen chlorine) damages environment.
a) SCP basing on halogen-free oxidizers.
The cheapest oxidizer of this type is ammonium nitrate (AN). It has the best element
content, however its too low enthalpy of formation (-1080 kcal/kg) makes it low effective as
oxidizers for SCP. In table 1 several oxidizers are represented for give an impression about
Isp values and difference among Isp values of different compositions. All Isp values
represented in this paper are calculated in condition when the combustion chamber pressure
is 40 atm and nozzle section pressure is 1 atm. Dimension of a Isp value is (kg-force)(kgmass)-1s (that is Isp is the resultant thrust by propellant burning with the rate 1 kg/s), or
simply s. All compositions represented in this paper contain at least oxidizer and a binder
with the optimal content of the binder, but in condition that binder percentage is not lower
than 18-20 volume % (that is about 8-10 mass% for compositions with total density ~1.8 and
binder density ~0.9) otherwise its impossible to produce a composite with satisfactory
physico-mechanical properties. Actually in binary compositions (oxidizer+binder) a binder
is necessary not only as a binder itself, but as a combustible too. In binary compositions
simple hydrocarbon binders may be rather optimal.
Table 1.
Energetic characteristics of SCP basing on the most important NOHoxidizers.
NH4NO3
Hf,
kcal/kg
-1080
d,
Additional info
g/cm3
1,725 M.p.
160C,
stable.
N2H5NO3
-661
1,685
M.p 71C,
hygroscopic
NH4N(NO2)2
-270
1,82
M.p.~89C,
enough stable
Oxidizer
NO2
NH4 N
NO2
ADNA
172
Composition, mass
%%
12%CH*
12%PMVT**
10%CH +20%Al
11%CH
12%PMVT
17%Act ***
15%Act +20%Al
12%PMVT+20%Al
10%CH +20%Al
12%CH
12%PMVT
10%CH +20%Al
15%Act +20%Al
12%CH +13%Be
15%CH +13%Be
34%Act +13%Be
34%Act +14%Be
Isp, s
211
220
246
219
230
240.5
256.5
255.5
255
247
250
262
264
281
293
291
294.5
d,
g/cm3
1.57
1.60
1.70
1.54
1.62
1.64
1.78
1.7
1.67
1.63
1.72
1.76
1.86
1.63
1.34
1.42
1.40
N2H5N(NO2)2
-52
1,80
Czech Republic
2006
[Content]
M.p.76-80C,
hygroscopic
12%CH
251
1.62
16%PMVT
255
1.70
19%Act
241.3 1.64
10% CH +20%Al
267
1.75
C(NH2)3N(NO2)2 -229
1,71
M.p
135- 10%CH
208
1.58
139C, enough 18%Act
233
1.67
stable
17%Act +20%Al
255
1.80
*-CH hydrocarbon binder (#1, table 3); ** PMVT - polymethylvinyltetrazole
(#2, table 3); *** Act- active binder (#3, table3).
The most interesting among NHO-oxidizers is ammonium dinitramide (ADNA). It was
first synthesized in the former USSR in 1972 (however first published in 1992 [1]). ADNA
is enough stable and compatible with many compounds and nowadays it is used in real
compositions. ADNA is almost the best oxidizer for SCP for upper stages while because of
its rather low density it is not enough powerful for first stage SCP, where ammonium
perchlorate (see below) is more appropriate.
b) SCP basing on perchlorates
Among all perchlorates (we consider only oxidizers free of metal, because mostly metal
salts are considerably less effective) only ammonium perchlorate (AP) is really used.
Nitronium perchlorate (NP) is the best energetic oxidizer, but it is so much chemical active,
that it can not be used in mixture with any other compound. Hydroxylammonium perchlorate
(HAP), hydrazine monoperchlorate (HMP), and hydrazine diperchlorate (HDP) are much
more effective than AP, especially for lower stages (because of high density). However these
high-density oxidizers have some serious drawbacks they are less thermostable than AP,
less compatible with other compounds, and, finally, they are too hygroscopic. In IPCP RAS
methods of thermal stabilization of HAP, HMP, HDP have been found [2], methods of
stabilizing mixtures of HAP with some compounds have been developed [3], but the
hygroscopicy can not be decreased yet.
Table 2. Energetic characteristics of SCP basing some perchlorates.
Oxidizer
Hf,
kcal/k
g
d,
g/cm3
Additional info
NH4ClO4
-597,5
1,95
Stable
NH3OHClO4
- 494
2,07
N2H5ClO4
- 320
1,94
Formulation, mass
%%
10%CH*
20%Act***
18%PMVT**
9%CH+20%Al
m.p.
89C, 13%CH
hygroscopic
20%Act
21%PMVT
9%CH+20%Al
m.p.
137C, 9%CH
hygroscopic,
10%CH
high
impact 26%Act
sensitivity .
9%CH+20%Al
Isp, s
d,
g/cm3
240
204
238
250
250
204
247
253.5
251
246.5
240
259
1.75
1.83
1.78
1.86
1.77
1.92
1.82
1.94
1.76
1.74
1.70
1.86
173
N2H6(ClO4)2
- 297
2,21
C(NH2)3ClO4
-465
1,75
C(NHNH2)3ClO
-116,5
1,56
65
2,22
NO2ClO4
Czech Republic
hygroscopic,
high
impact
sensitivity
m.p.240C,
stable,
hygroscopic
high imp. sensit.,
mp.132C.
Extremally
reactive
2006
[Content]
12%CH
9%CH+20%Al
251
254
1.88
2.02
18%Act
15%Act+20%Al
226
246.4
1.70
1.82
18%Act
15%Act+20%Al
20%CH
14%CH+20%Al
233.5
255
261
258
1.54
1.69
1.72
1.91
c) Oxidizers containing active fluorine

The fluorine is a more active oxidizer than oxygen is, the heat of H2O formation is much
lower than the heat of HF formation. There are some inorganic oxidizers with active fluorine
(e.g. FClO4, some complex salts (NF4)aMeFx) but their physical and chemical properties do
not allow us to use them in SCP. The only real way to introduce active fluorine into
molecule is the using of >N-F or -NF2 groups because if fluorine is introduced with a
component containing C-F bond no power increase is gained because of high energy of the
C-F bond. Nowadays many NF2-derivatives of organic substances are known, some of them
are used in energetic compositions. If such compounds are used as oxidizer its possible to
create SCP with Isp 280 and even higher. However these substances have some serious
drawbacks they are not ecologically soft, and their impact and friction sensitivity are rather
high.
SCP, CONTAINING METALS.

Only four metals (Be, Al, B and Mg) can increase the Isp of the pair binder+oxidizer.
The effectiveness of metal is due to the great heat of their oxidation into oxides. In case of
other metals the heat of their combustion does not compensate the loss of gas fraction in
combustion products. The effectiveness of metal introduction falls with the range
BeAlBMg, and it depends much on the oxidizer enthalpy. In fig.1 the dependence of
the effectiveness of metal introduction (dIsp) on oxidizer enthalpy is demonstrated. dIsp =
IspMe - Isp0, where IspMe is the Isp value of the optimal composition
metal+oxidizer+hydrocarbon binder; Isp0 is the Isp value of the optimal binary formulation
oxidizer+hydrocarbon binder. Circlets on the figure represent very different oxidizers, with
different oxygen balance, organic and inorganic, real and designed. Thats why all circlets
for each metal do not represent an absolute straight lines. However, the tendency is
evidential effectiveness of metal introduction falls with the growth of the oxidizer
enthalpy. Magnesium introduction may be useful only if Hf of the oxidizer is rather low
(see fig.1), while the introduction of beryllium is useful until Hf of oxidizer is about +600800 kcal/kg. However Be-containing SCP are extraordinarily toxic because of BeO toxicity,
thats why their development ended about 30 years ago. Regardless the introduction of other
metals always decreases the Isp values, some metals with high density (Zr, Ti) can be useful
in definite cases when the propellant density is of higher importance (see above), that is
when the exponent a in the formula Efn=Isp*da is 0.7-0.8 and higher.
174
Czech Republic
2006
[Content]
dIsp
60
NH4NO3
ADNA
50
Be
40
30
N2H5N3O4
Al
20
10
0
Mg
-1000
-500
500
1000
oxidizer's specific enthalpy, kcal/kg
Fig. 1. The dependence of the dIsp on CNHO-oxidizer enthalpy [4].
OPTIMAL BINDERS.
We found that regardless binder does not occupy a big fraction in the composition
(usually 9-12 mass % if binder density is 0.9-1.0 and about 20-25 mass % if the density is
high, up to 1.5-1.55) the choice of binder type is very important for the maximal energy
achievement. Element content of the binder has to compensate excess or lack of oxygen in
the leftover formulation (the summary of compounds except binder). Thus, if binary
formulation contains a binder and an oxidizer with a high oxygen balance ( =
O/(2C+0.5H)), e.g. ADNA or HAP, the best binder is usually a simple hydrocarbon binder,
because it contains a lot of hydrogen. As soon we begin to add so called gasifying
component, e.g. HMX, the value of the mixture oxidizer + HMX begins to decrease and
the hydrocarbon binder can not any longer secure maximal energetic potential, that is kept in
these compounds we have to use more active binders with higher values. Because of
the same reason active binders are necessary if the formulation contains metal hydrides (see
below). Thus, when we study a new compound (mainly oxidizer, gasifying or energetic
compound) we have to test formulations with different kinds of binder. For example, in our
investigations we use a listing like that represented in table 3. Binder #2 is a representative
of high-enthalpy binders, binder #4 is a representative of binders with a middle value, but
higher enthalpy and hydrogen content than #3. In some cases such binders as #4 may
become the optimal binders among all other types.
As for gasifying compounds the best of them is widely used well-known HMX. Actually
we may consider a gasifying compound as a part of a binary oxidizer. Regardless gasifying
compound is not usually a good oxidizer itself, its mixture with the main oxidizer can be
more effective than if the main oxidizer is used alone. For example in the system 15.5
mass% (20 volume %) active binder + 20 mass %Al +AP + HMX the maximum value of Isp
is 260.5 at 57 % HMX, while in the system 10 mass % (20 volume %) hydrocarbon binder +
20 %Al +AP + HMX the maximum value of Isp is 254 at 25 % HMX.
175
Czech Republic
2006
[Content]
Table 3. Energetic properties of some binders.

#
Name, formula
Hydrocarbon, C73,17H120,9
Polymethylvinyltetrazole, C37,34H56,007N32,69O2,32
Active, containing 20% polyvinylmethyltetrazole,
nitroglycerine with some additives. C18,96H34,64N19,16O29,32
3
4
Polyvinylmethoxydiazen-N-oxid, C3H6N2O2
Hf,
kcal/kg
d,
g/cm3
-94
0,91
+300
1,28
~0
-181
1,49
0.53
-14
1,31
0.22
SCP WITH METAL HYDRIDES.

The using of metal hydrides (BeH2, AlH3) in SCP offers great opportunity to increase
much Isp values. In table 3 its evident that Isp values can be increased on 10-15s using the
same oxidizer, because hydrides serve not only as metal fuel, but as a rich source of
hydrogen too. However new SCP with hydrides required new binders so called active
binders (see above). If in propellants oxidizer+binder and oxidizer+binder+metal a binder
was necessary not only as properly binder, but as a fuel (in system oxidizer+binder) and a
gas source (both quoted systems) too, in the presence of hydride SCP has already a powerful
heat source as well as a gas source. So, binder is not energetically necessary any more.
However so long as a polymer binder is still necessary, it has to possess new qualities,
namely to serve as oxidizer too. So, new active binders, containing NO2, ONO2, NNO2
groups etc have been created. Simultaneously active binders may serve as energetic
component if they contain high energetic functional fragments, e.g. N-heterocycles. The
creation of new SCP with metal hydrides was a great step in development of high energy
SCP. Isp value of the composition containing 25% AlH3, 25% active binder and ADNA is
276-277 (d=1.64), of the composition containing 14% BeH2, 34% active binder and ADNA
is 293.5 (d1.40). However AlH3 and especially BeH2 have low density (1.48 and ~0.7 g/cm3
respectively), therefore the SCP density is lower than in similar SPC with free metals. Thus,
SCP with BeH2 is energetically useful only for upper stages (really for the third stage only)
when the exponent a in the formula Efn=Isp*da is 0.2 and lower; with AlH3 - if the
exponent a is ~0.4 and lower, i.e. for the second and third stages. Moreover, BeH2 has the
same obstacle than free Be has, this is an extraordinary high toxicity of the combustion
product (BeO).
SCP BASING ON HIGH-ENTHALPY OXIDIZERS.

This kind of SCP is not developed yet. Regardless many organic oxidizers with Hf
+500 kcal/kg and even higher are known, their application in SCP is limited by many factors
(high impact and friction sensitivity, sometimes low thermal stability, high price, etc.). The
first relatively high-enthalpy component that has been widely investigated during last 15
years or so is hexanitrohexaazaisowurtzitane (CL-20, Hf =+185 kcal/kg, d=2.044 [5]).
Propellants containing aluminum, active binder, CL-20, and AP are a bit more effective on
the lowest stage (Isp~258 at Pc:Pa=40:1, d=~2.00; combustion temperature 3970K;
temperature of products leaving nozzle section 2770K) than the similar propellant containing
HMX instead of CL-20. Actually this higher effectiveness is the result of higher density of
CL-20 against HMX.
176
Czech Republic
2006
[Content]
There are many organic substances (containing mainly high-enthalpy N-heterocyclic

frame with fragments-oxidizers, such as NO2, NNO2, C(NO2)3, etc) that can be used as highenthalpy oxidizers (Hf >500 kcal/kg), e.g. ditrinitromethylfuroxane (DTNMF, Hf=+570
kcal/kg ); 1,2,5-oxodiazolo-[3,4-e]-[1,2,3,4]-tetrazine 4,6-dioxide (ODATDO, Hf=+730
kcal/kg); 5-(3-nitro-furoxan-4-yl)-1H-[1,2,3]triazolo[4,5-c]-furoxan inner salt (NFTAF,
Hf=+760 kcal/kg) [6,7]. Using such oxidizers with appropriate binders its possible to
accomplish Isp values up to 268-270s.
(O2N)3C
NO2
H2C
O2N N
H2C
CH2
O2NN
O2NN
N NO2
N NO2
N
CH2
C(NO2)3
NNO2
N
N
N
O
O2NN
NNO2
O
N
N
O
NO2
N
NO2
HMX
CL-20
DTNMF
ODATDO
NFTAF
Anyway propellants with such new oxidizers as DTNMF, NFTAF or ODATDO will
cost much more than propellants basing on accessible PA and HMX. Moreover, as it was
already remarked, further increasing of energy power of solid composite propellants is
attended with an appreciable aggravation of other properties (impact and friction sensitivity,
thermal stability etc.).
Anyway such high-enthalpy oxidizers are still rather promising, moreover they do not
need metals (see above) and therefore they have no loss of Isp because of biphasic
combustion products while metal-containing SCP have them (we mean here the loss as the
difference between real Isp value and the calculated one). The reason of so-called biphasic
loss is the following: when in combustion products there is a solid phase some extra heat is
carried away from the nozzle section with these solid particles - they are a bit superheated in
comparison with the gas phase because the thermal equilibrium between solid and gas phase
can not be established during so small time when combustion products expand trough
nozzle. Thermodynamic calculation of Isp does not include this kind of thrust loss, therefore
if calculated Isp values of two different formulations are equal, but the first formulation
contains metal while the second does not, it means that other characteristics being equal the
second formulation procures higher real thrust. Actually biphasic loss depends on solid
particles dimension and of the rocket engine type and may average 5% or so.
WAYS FOR FURTHER SOLID COMPOSITE PROPELLANTS

DEVELOPMENT.
Recently the geopolitical situation changed considerably because of the cold war end.
Therefore the creation of cost-to-cost ballistic missiles and their propellants have not such a
great importance any longer while the portion of SCP civil applications grows constantly in
importance.
If we estimate briefly the succeed energy level of SCP that are already produced in
reality it is evident that the most powerful are the following ones:
propellants based on aluminum, HMX (~55%) and active binder - Isp~260 at
Pc:Pa=40:1, d=1.91; combustion temperature (Tc) =3780K; temperature of products leaving
nozzle section (Ta) =2550K - this kind of propellant is the most effective on lower stages
because of high density;
177
Czech Republic
2006
[Content]
propellants based on aluminum hydride, ADNA and active binder (Isp~276 at

Pc:Pa=40:1, d=1.64; Tc=3550K; Ta=2330K), this kind of propellant is the most effective on
upper stages;
propellants basing on CL-20 (see above) that are developed recently are more effective
on the lowest stage only because of its higher density, on upper stages HMX is more
effective.
SCP basing on new high-enthalpy oxidizers (see above) have rather higher Isp values,
but this kind of propellants is not developed yet, and in any case they will cost much higher
than propellants with PA, HMX, ADNA. Moreover, further increasing of SCP power is
attended with an appreciable aggravation of other properties. For example, application of
high-enthalpy CHNO-substances (>+500 kcal/kg) would decrease the stability; would
degrade impact and friction sensibility, would increase values of exponent in combustion
law U=A*P. Anyway maybe SCP of this type (with high-enthalpy CHNO-substances) can
be used too, but this perspective is too far.
In the case of civil applications one puts a considerable accent on ecological problems.
The HCl effusion ruins greenery and atmospheric ozone coat. The solution of these problems
may be carried by different ways, using chlorine free inorganic (e.g. ADNA, Table 4, #4) or
organic oxidizers, e.g. HMX (#3), RDX (hexogen), as well as using propellants containing
potassium or sodium salts (e.g. sodium nitrate or hyponitrate) together with AP (Table 4,
#15,16) when equivalent amount of alkaline metal is a bit higher than one of chlorine, thus
almost all chlorine undergoes into the most thermodynamically stable alkaline metal chloride
[8].
Table 4. Energetic characteristics and concentration of toxic gases and condensed
phase in combustion products
Propellant
Isp
Tc, K Ta,K
HCl+Cl2,
mol/kg
CO,
mol/kg
Cond. phase,
mass.%
251 1.84 3610 2520
38
PA+CH*+Al
AP+CH
240 1.76 3000 1900
7.8
0.8
HMX+ Act**+Al
257 1.95 3700 2550
12
38
ADNA+ CH+Al
262 1.76 3580 2330
38
ADNA+ CH
249 1.64 2970 1580
3.6
N2H5NO3+ CH+Al
254
7.7
38
N2H5NO3+ CH
930
NH4NO3+ CH+Al
244 1.68 2880 1800
7.3
38
NH4NO3+ CH
205 1.56 1870
920
10 N2H5C(NO2)3+CH+Al 265 1.81 3580 2360
11
38
178
1.6
2910 1770
215 1.53 1950
12
N2H5C(NO2)3+ CH
254
13
NaClO4+ CH +Al
14
NaClO4+ CH
1.7
Czech Republic
[Content]
225 2.05 3650 2610
1.1
10
66
211
3080 2030
0.25
44
15 AP+NaNO3+ CH +Al
230 1.96 3665 2650
0.9
4.7
37.7
16 AP:Na2N2O3+CH+Al
237 1.94 3675 2600
0.98
5.3
37.7
2.0
3070 1600
2006
SOME NEW CIVIL APPLICATIONS OF SOLID COMPOSITE

COMPOSITIONS.
Multiyear experience of chemists in the area of military propellants creation and
production may help to resolve currently some tasks of exceptionally civil application of
energetic materials, i.e. in development of gas-generating compositions for airbag inflators.
In this task the thrust value is no longer necassary, but quite other attributes become crucial.
Moreover, very high energy makes even cardinally worse the characteristics of the toppriority. New propellants must be free of drawbacks of currently used propellants. The main
drawbacks are toxic ingredients, presence of toxic gases (CO, NOx) and solid phase in
combustion products, too high combustion temperature, insufficient stability. By the way
propellants for airbag inflators require much higher thermal stability level than propellants
for solid rocket engines do. Main requirements for compositions for airbags and the
characteristics that SCP can provide differ much (Table 5). Thus, for improve characteristics
of propellants for their application in airbags, we have to decrease their power, that is to
decease considerably combustion temperature of stoichiometric formulation (#1, table 5),
and on the other hand we have to increase the stability and combustion rate.
It was shown that it is possible to decrease combustion temperature of stoichiometric
formulations (this is CaHbNxO2a+0.5b, otherwise combustion products contain either CO or
NOx) by introducing low-enthalpy oxygen-containing groups (-OH, -O-, -COOH, >CO, COO-, -OCO-COO- etc.) into oxidizer containing regular groups-oxidizers (-NO2, -ONO2, NNO2). Later groups provide high oxygen amount in the formulation while the first ones
provide considerable enthalpy decreasing, and consequently Tc decreasing at minimal
oxygen amount decreasing [9].
Table 5. Requirements to propellants for airbags and characteristics that can be
reached with accessible SCP.
#
Characteristics
Element ratio
Condensed phase in
comb.products
Requirement to
propellants for
airbags
2C+0.5H~ =O (atoms
ratio)
Absence of solid, at
least very low amount
Combustion Temperature
~ 2000-2400K
11
Characteristics that can

be reached with
accessible SCP.
No requirements
Usually 37 mass % of
Al2O3 if the formulation
contains 20% Al. If there
is no metal in the
formulation no
condensed phase
>2900 K
179
Combustion rate
Exponent in combustion law

(U=A*P)
Thermal stability
toxic products and hydrogen
content combustion products
6
7
Minimal melt.point of
ingredients
9 sensitivity to impact and friction
10 requirement to dioxines amount
Czech Republic
2006
[Content]
25-30mm/s at 200
atm
~0.5-0.6
~10-20 mm/s
at 200 atm
0.3-0.5
~ 2 weeks at 110C
No toxics, no HCl, no
Cl2.
Low H2 content, very
low CO and NOx
content.
110C
One week at ~60-70C

[CO] =3 mol/kg (even if
Tc~1900K) and higher.
[H2] =0.5-15 mol/kg.
Many HCl if the
formulation contains AP.
80C
Low
No data
Rather low
Accordingly international
ecological requirements<10-10g/m3
High energetic compositions may be used more widely as gas generating compositions
for different goals. Of the most importance are generators producing gases (air, oxygen,
nitrogen and other gases) with given content with a necessary combustion rate.
REFERENCES.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
180
Lukyanov O.A., Gorelik V.P., Tartakovskii V.A., Izvestia of Russ.Ac.Sci., ser.chem., 1994,
P.94-97.
Lempert D.B., Dolganova G.P., Nechiporenko H.N.// Hydroxylammonium perchlorate thermal
decomposition on low decomposition level. (1998) J.Appl.Chem. (Russ) v.98, N3, P.365-370.
Lempert D.B., Dolganova G.P., Nechiporenko H.N.// Thermal stability of mixture
hydroxylammonium perchlorate with HMX. (1998). J.Appl.Chem. (Russ) v.98, N10, P.16231628.
Lempert D.B., Nechiporenko H.N., Dolganova G.P.//Chemical Physics (Russ), 1998. V.17, N7,
P.114-120
R.L. Simpson et al.//Propellants, Explosives, Pyrotechnics 22, 249-255 (1997)
Lempert D.B., Nechiporenko H.N. et al. Chemical Physics (Russ), 2003. V.22. N4. P.64-69.
Nechiporenko H.N., Lempert D.B., Soglasnova S.I. Chemical Physics (Russ), 2005. V.24. N3.
P.74-80.
Lempert D.B., Dolganova G.P., Nechiporenko H.N., Stesik L.N.// Chemical Physics (Russ),
1997. V.16. N9. P.91-100.
Lempert D.B., Manelis G.B., Nechiporenko H.N., Shteinberg V.G., Zyuzin I.N.// Principles of
Creation of Smokeless Gas Generating Compositions Free of Toxic Combustion Products. 7th
International Airbag-2004, 2004, Karlsruhe, Germany, P.43.
Czech Republic
2006
[Content]
MECHANISM OF THE LOW-FREQUENCY NON-ACOUSTIC

INSTABILITY PHENOMENA FROM THE POINT OF VIEW OF THE
CONCEPT OF THE SPATIAL-PERIODIC MICRO-STRUCTURES
EXCITATION AT THE ENERGETIC MATERIALS COMBUSTION
Alexander N. Lukin
Institute of Applied Mechanics, Ural Branch of the Russian Academy of Sciences,
426067, Izhevsk, Udmurtia Republic, Russian Federation
Abstract:
Understanding of the mechanism of the low-frequency non-acoustic instability
phenomena at the solid rocket propellants combustion - one of the most complicated
problems in the solid propulsion physics. Several theories have been proposed
highlighting one or other process as the dominant mechanism, but a unifying theory is
yet to emerge. The present work focuses on one of the critical events, namely the spatialperiodic micro-structures (SPMS) excitation in the evaporated energetic materials (EM)
liquid-viscous layer (LVL). The low-frequency non-acoustic combustion instability
phenomenon, the chuffing phenomenon and the accompanying physico-chemical effects
have received a new explanation within the concept based on the data of optical
visualization of the physico-chemical processes on the EM burning surface. This
concept connected, mainly, with excitation of the synergetic dissipative SPMS in the thin
LVL and on the EM burning surface and determining the burning wave spatial
instability. At heating from above in the thin LVL occurs the thermo-electric convection
excitation, that induce cellular movement and formation of the synergetic SPMS.
Besides the velocity cells, in the LVL arise the electric field cellular structures.
Preliminary results of the investigation show that mechanism of the low-frequency nonacoustic combustion instability of the evaporated EM is connected with the phenomenon
of the existential micro-structures excitation in the EM LVL. Comparison of the scales of
the EM heated up layer, the LVL, the SPMS and the torches (burning cells) on the
burning surface also shows that the torches (cells) are formed from the set of SPMS from the set of thermo-electric convection cells. The experimental data have shown that
torch combustion phenomenon develops irrespective of the properties and structure of
the specific EM. By other words, on the EM burning surface is observed the process of
self-organizing of the dynamic dissipative synergetic SPMS into the torch macrostructures. In this connection, on the burning surface the process of moving (wandering)
of the torches is observed. The torch macro-structures on the EM burning surface can
be considered as independent synergetic structures. Besides, the torch structures on the
burning surface can exist steadily in the event that they form a pair of two torches. This
phenomenon reflects universal law of the SPMS excitation on the EM burning surface.
Within suggested mechanism becomes possible to give a new explanation of the
phenomenon of the end-burning solid propellant charges rhythmic extinctions. An
attempt has been made to indicate future direction of research that can be lead to more
accurate prediction of starting transients in the solid propulsion systems.
Keywords:
evaporated energetic materials, combustion instability, chuffing,

torch macro-structures, ionic fusion with thermo-electric properties,
thermo-electric convection excitation, carbon grid
Alexander Lukin
181
1.
Czech Republic
2006
[Content]
INTRODUCTION
The problem of combustion instability and anomalies of burning of the energetic

materials (EM) traditionally remains one of actual problems in the theory of combustion. [1]
The principles of the EM combustion anomalies theory were established by outstanding
Russian scientist Ya.B. Zel'dovich. [2]
Prediction of ignition transients and low-frequency non-acoustic combustion instability
in the solid propulsions systems (SPS) has remained a topic of active research for several
decades, yet there appears to be no model that can describe the roles played by all the
complex physico-chemical processes. The development of larger and more sophisticated
SPS have emphasized the need to model the ignition transients accurately as they do not lend
themselves to costly trial and error development techniques. The radical difference in size
and technology of these SPS defy extrapolation of the empirical knowledge gained in the
development of earlier, more conventional SPS.
With regard to the physical mechanisms of the solid rocket propellant combustion, the
combustion zone is quite complicated even under steady state conditions, with multiple
flamelets attached to different parts of the burning surface. The complications are, in part,
due to the statistically random nature of the distribution of the reactants in the condensed
phase, and, in part, due to poor understanding of the combustion zone owing primarily to its
microscopic nature and the hostility of the environment to clear investigation.
Understanding of the mechanism of the low-frequency non-acoustic instability
phenomena at the solid rocket propellants combustion - one of the most complicated
problems in the solid propulsion physics. Several theories have been proposed highlighting
one or other process as the dominant mechanism, but a unifying theory is yet to emerge. The
present work focuses on one of the critical events, namely the spatial-periodic microstructures (SPMS) excitation in the evaporated EM liquid-viscous layer (LVL).
2.
THE LOW-FREQUENCY NON-ACOUSTIC INSTABILITY IS THE

CHARACTERISTIC PROPERTY OF COMBUSTION PROCESS OF
THE SOLID ROCKET PROPELLANTS
SPS with a small characteristic length of the chamber L* may exhibit spontaneous
oscillations in the chamber pressure. If the oscillations occur simultaneously in the bulk of
the chamber, such an instability has been variously called the L*, bulk mode or non-acoustic
type. The phenomenon is typically characterized by the amplifying low-frequency pressure
oscillations of below some 300Hz, leading to the extinction of the SPS. In extreme cases,
when the L* is very small, extinction occurs almost immediately after ignition, followed by a
sequence of periodic pressure build-up and extinction: such a process has been called
chuffing.
The phenomenon of chuffing involves large pressure fluctuations, sometimes with
widely varying frequencies. Although the order of magnitude of chuffing frequencies
suggests a dominant solid-phase energy release mechanism at low pressures, the underlying
process is not convincingly explained yet. Such kind of combustion instability in the
combustion chamber is most frequently observed for the end-burning solid propellant
charges and for channel-shaped propellant grains. According to the data presented in the
182
Mechanism of Low-Frequency Non-Acoustic Instability Phenomena From the . . .
Czech Republic
2006
[Content]
paper [3], for the channel-shaped charges the boundary of low-frequency combustion
instability depends not from the channel volume and the chamber volume, but is determined,
mainly, by the pre-nozzle volume. Therefore, if the volume of the SPS combustion chamber
with end-burning charge coincides with the pre-nozzle volume of the SPS with a channelshaped charge, the boundaries of their low-frequency stability coincide.
In the paper [4] on the basis of large number of the fire stand tests has been made
conclusion that periodic processes - the chuffing processes and the low-frequency nonacoustic combustion instability are the characteristic properties of combustion process of the
solid rocket propellants. Usually, these properties appears in the conditions of low pressure
in the combustion chamber and (or) in the conditions of the low ambient temperatures. As a
matter of fact, for the first time has been suggested assumption that the burning process of
each specific solid propellant can be characterized by the set of own frequencies of
pulsations of the burning surface that appears in the critical burning conditions. And also has
been mentioned the existence of some universal law connected with unstable processes on
the burning surface of the solid propellants. However how it is possible to explain this
property and what physical phenomena appears here not enough clearly till now. Existence
of such problems has been indicated in the recent dissertations. [3, 5] The hypothesis that the
determining mechanism of excitation of the low-frequency combustion instability and
chuffing are the reactions in the gas phase has been suggested in the paper [6]. The similar
hypothesis about existence of the gas-phase oscillatory chemical reactions which excite lowfrequency combustion instability and chuffing has been suggested in 2003, in the dissertation
[5]
. However the papers [7, 4] have convincingly shown that the determining mechanism of
excitation of the chuffing processes and the EM low-frequency combustion instability are the
reactions in the condensed phase (the mechanism of thermal explosion of the condensed
phase), but not the reactions in the gas phase. Frequency of pulsations at the burning process
of the propellant laboratory samples in the air appeared same as at burning of the samples in
the conditions of the nitrogen flow above the propellant burning surface.
At the same time, the heat feedback from the hot gases trapped in the chamber when the
chamber and ambient pressure equalized due to extinction, could not cause re-ignition of the
propellant. Re-ignition after extinction, and hence chuffing, could not be obtained with a
surface reaction model.
For successful solution of the problems, connected with suppression of the SPS
combustion instability are necessary to have detailed understanding about essence of
mechanisms of the EM unstable burning on the new qualitative level. Obviously the
possibilities of understanding of laws of this complex phenomenon are closely connected
with excitation and formation of the cellular micro-structures on the EM burning surface.
Certainly, the new level in understanding of the fundamental laws of the EM lowfrequency non-acoustic combustion instability can be provided on the basis of use of the
newest technologies of visualization of zones of burning of the EM.
Recently with use of new interferometric and shadow technologies of optical
visualization of zones of burning of the EM, the unique results overturning traditional
representations about the mechanisms of cellular-pulsating burning and low-frequency nonacoustic instability at the solid rocket propellants combustion have been received [8 - 10].
Alexander Lukin
183
Czech Republic
2006
[Content]
Some fragments of optical visualization of the process of burning of the standard

ballistite propellants are shown on the Fig. 1 and Fig. 2.
Fig 1. Sequential images of dynamics of the carbon spatial-periodic micro-structures on the

burning surface of the standard ballistite propellant.
On the Fig. 1 are shown the fragments of the process of throwing away of the carbon
grid, formed at the burning surface.
The experimental data have shown that torch combustion phenomenon develops
irrespective of properties and structure of the specific EM. At realization of this burning
mode the sizes of burning cells are not connected with sizes of components and the sizes and
structure of researched EM. Besides, according to the experimental data, ignition of long
propellant channels occurs in the cellular or in the torch mode. [11]
Also the experimental data has shown that active (reaction-capable) SPMS (the burning
cells) on the burning surface are united into the more large-scale structures - into the torch
macro-structures. By other words, on the EM burning surface is observed the process of selforganizing of the dynamic dissipative synergetic SPMS into the torch macro-structures. The
torch micro-structures on the EM burning surface can be considered as independent
synergetic structures. This phenomenon reflects universal law of the spatial-periodic microstructures excitation on the burning surface of the EM.
In this connection, on the burning surface the process of moving (wandering) of the
torches is observed.
184
Czech Republic
2006
[Content]
Fig 2. The effect of formation of the pair of torch structures. Sequential images of dynamics
of two torch structures on the burning surface of the standard ballistite propellant.
Fig. 2 shows dynamics of the two torch structures on the burning surface of the standard
ballistite propellant. The processes in each of the burning cells develops independently from
each other. The torch structures on the burning surface can exist steadily in the event that
they form a pair of two torches. This phenomenon is observed also on the shadow photos of
torches on the burning surface.
The Fig. 2 (1) shows a luminescence of gaseous products of reaction in the one torch. At
this time occurs formation of the carbon grid on the burning surface. After formation of the
carbon grid on the burning surface the luminescence of the torch disappears. Further the
process starts to develop on the burning surface and appears luminescence of the carbon grid
on the burning surface (Fig. 2 (2)). The combustion products flowing from the burning
surface throw away a carbon grid from the burning surface ( Fig. 2 (3)). After throwing
away of the carbon grid the luminescence of the torch above the burning surface appears
again ( Fig. 2 (4)).
Separate researches of the torch structures were executed, with use of shadow method,
both for the gun-powder "N" and for standard gun-powders on the basis of nitrocellulose
with the lead oxide. [8 - 10] In the researches were used the gun-powder plates with thickness
of 3 mm (on the course of the light beam) and with width of 12 mm.
Fig. 3 and Fig. 5 contain the information only of two categories. Firstly, this is jet stream
of the gasification products from the burning surface and secondly, the dependence of the
sizes (and accordingly, the numbers) of the burning cells from the pressure.
5
Alexander Lukin
185
Czech Republic
2006
[Content]
Fig 3. Burning of the ballistite gunpowder "N" in the nitrogen

atmosphere at the pressure 0.4
MPa. Two torch macro-structures
on the burning surface.

MPa. Four torch macro-structures

MPa. Six torch macro-structures
At execution of the researches in the field of high pressures it was impossible to

differentiate the cells on the burning surface. In these conditions on the burning surface there
was so many burning cells, that at transverse passing of the light beam through the several
jets it was impossible to differentiate behavior of the separate burning cell on the
background of others.
3.
EXCITATION OF THE SPATIAL-PERIODIC MICROSTRUCTURES IN THE IONIC FUSION WITH THERMO-ELECTRIC

PROPERTIES ON THE BURNING SURFACE
In accordance with extensive experimental data [12 - 14] burning of the evaporated EM is
accompanied by occurrence of electric conductivity of the burning surface (of the liquidviscous layer) and by chemical ionization of the gas layers adjoining to the burning surface.
186
Czech Republic
2006
[Content]
In conditions of burning wave where the temperature in the condensed phase increasing by
exponential law, the thin reactionary LVL can be considered as the molten mass with ionic
properties. Concentration of the ions increases with the burning rate and determines electric
conductivity of the burning surface. At burning in conditions of threshold-low pressures,
periodic pulsations of the electric conductivity are observed.
The low-frequency non-acoustic combustion instability phenomenon, the chuffing
phenomenon and the accompanying physical-chemical effects have received a new
explanation within the following universal concept based on the unique data of optical
visualization of physico-chemical processes on the EM burning surface.
This concept connected, mainly, with excitation of the synergetic dissipative SPMS in
the thin LVL and on the EM burning surface and determining the burning wave spatial
instability. At heating from above in the thin LVL occurs the thermo-electric convection
excitation, that induce cellular movement and formation of the synergetic SPMS. [15] Besides
the velocity cells, in the LVL arise the electric field cellular structures.
Fig. 6 schematically shows the distribution of the physico-chemical processes in the thin
LVL in the EM reaction zone.
Fig 6. Thin liquid-viscous layer in the EM reaction zone, having thermo-electric properties
at heating from above.
In accordance with the experimental data, the ratio of the longitudinal and cross-sectional
sizes of the elementary SPMS are the most stable characteristic of the cellular movement
excitation. Change of the boundary conditions practically does not influence this parameter.
The reason of stability of the sizes of the SPMS is connected with the thermo-electric
mechanism of excitation of such structures.
Alexander Lukin
187
Czech Republic
2006
[Content]
In other words, stability of the SPMS is provided by the electro-magnetic field cellular
structures in the LVL. The experimental data shows the boundaries of the electric field
structures coincide with the boundaries of the convection cells (the structures of velocity).
And on the LVL surface under influence of thermo-electric field is excited the electric
charge.
4.
TWO SCALE OF THE SPATIAL-PERIODIC STRUCTURES ON THE

BURNING SURFACE
Comparison of the scale of zone of the EM cellular-pulsating burning shows that

reaction-capable SPMS excited in the LVL are the initial existential structures that are united
in the set of aggregates. Each of such aggregates represents a torch or the burning cell on the
burning surface.
The burning process localized in each of the SPMS, is supported by the process of selforganizing of the SPMS into the groups. In other words, each of such aggregates, formed
from two and more SPMS with identical properties it is possible to consider as a cluster.
TORCH (CELLULAR) SPATIAL-PERIODIC MICRO-STRUCTURES ON THE

BURNING SURFACE
Scale of the torch (the burning cell) on the burning surface - (10 - 15) h
(Diameter of the torch - (1 - 2) mm)
CARBON GRID (CARBON SPATIAL-PERIODIC MICRO-STRUCTURES)

ON THE BURNING SURFACE
HEATED-UP LAYER
OF THE ENERGETIC
MATERIAL
Reaction Liquid-Viscous Layer
The Thickness of the
Heated-up Layer
(The Conductive Size)
h a / uC
(0.010 0.070) mm
The thickness of the reaction liquid-viscous layer determines

scale of the SPMS (cells) of the thermo-electric convection.
For estimation of the average thickness of the reaction liquidviscous layer (the effective thickness of the zone of electric
conductivity) in the EM heated-up layer can be used the
following expression: S a / ( 10 u C )
The thickness of the reaction liquid-viscous layer
- (0.0010 0.0035) mm
The Inert Heated-up Layer of the Energetic Material
188
Czech Republic
2006
[Content]
The torch structures on the burning surface forms the non-uniform burning front: on the
burning surface appears the local discontinuity.
In the conditions of low pressures the sizes of the SPMS and the cluster structures
(torches) will increase, but the total number of the torch structures on the burning surface
will decrease. And vice versa, in the conditions of high pressures the sizes of the SPMS and
the cluster structures (torches) will decrease, but the total number of the torch structures on
the burning surface will increase.
Besides for estimation of the S value can be used D.A. Frank-Kamenetskiis
characteristic temperature interval [16]. Let's consider that the main part of reactions takes
place within the limits of the quadruple specified temperature interval:
S =
4 R T S2
(1)
Ea 0
With taking into account equality for the temperature gradient:
0 =
CuC
( T R T0 )
(2)
We can write down:

S =
4 R T S2
E a u C ( TR T0
4 R TS
a
u C E a (T R T0
2
(3)
For example, at T 0 = 293 K and P = 4 MPa, S = 0.0035 mm; and at T 0 = 293 K and P =
10 MPa, S = 0.0022 mm.
5.
THE PHENOMENON OF THE SPATIAL-PERIODIC MICROSTRUCTURES EXCITATION ON THE ENERGETIC MATERIAL

BURNING SURFACE AT THE INFLUENCE OF THE LASER
RADIATION ENERGY
The experimental data have shown that the SPMS develops on the EM burning surface
under influence of the laser radiation energy. Both the experiment, and the theory confirm,
that the SPMS formation is rather universal phenomenon.
In the researches was used radiation of the 2 - laser with power of 60 watt and with
wave length 10.6 micrometers. The time of influence of the laser radiation energy on the
gun-powder surface gradually increased up to the moment of ignition of the burning surface
of the sample. It has allowed to observe development of the micro-structures on the burning
surface in various phases of the process. Fig. 7 shows the image of the burning surface of the
standard ballistite propellant under influence of the laser radiation energy in the nitrogen
atmosphere at the pressure of 0.4 MPa.
Alexander Lukin
189
Czech Republic
2006
[Content]
Fig 7. The structure of the surface of the sample after influence of laser radiation during 120
milliseconds.
Excitation and formation of the SPMS on the burning surface is accompanied by
separation of the LVL on the local zones with different viscosity and density. Local zones
with the increased viscosity are formed on the boundaries between cellular micro-structures.
The Carbon
Particle
Fig 8. Formation of the carbon micro-structures on the burning surface.

190
Czech Republic
2006
[Content]
In such zones with increased viscosity occurs formation of the carbon micro-structures
(Fig. 8). Actually, the carbon grid reflects one of sequential stages of development of the
cellular micro-structures on the EM burning surface. In this case, excitation of the thermoelectric convection in the layer is a main structures-forming factor. At further increase of the
duration of influence of the laser radiation energy on the surface of the sample, the carbon
structures will unite with each other and will cover all a greater and greater surface. Fig. 9
show gradual increase in quantity of carbon structures on the burning surface.
Fig 9. Increase in quantity of the carbon micro-structures on the EM burning surface.
Fig 10. Formation of the carbon micro-structures on the burning surface.

11
Alexander Lukin
191
Czech Republic
2006
[Content]
When the carbon micro-structures covers the surface of the sample approximately on 60
%, transition from ignition to burning will be observed. Fig. 10 shows the threshold time
moment, after which the surface of the sample will be ignited.
The EM burning surface gradually becomes covered by a carbon grid. During formation
of the carbon grid on the burning surface the burning process is gradually reduced and
eventually the torch luminescence disappears. At this stage of the process the luminescence
of the carbon grid is observed only (Fig. 2, (2) ). Further, the combustion products flowing
from the burning surface throw away a carbon grid from the burning surface Fig. 11.
Carbon Grid
Fig 11. The process of throwing away of the carbon grid, formed at the burning surface
accompanied by extinction of the burning surfaces.
After throwing away of a carbon grid the luminescence of the torch above the burning
surface appears again. Formation of the carbon grid are determined by formation of the
cellular micro-structures on the EM burning surface.
6.
THE PROCESS OF SELF-ORGANIZING OF THE DYNAMIC

DISSIPATIVE SYNERGETIC SPATIAL-PERIODIC MICROSTRUCTURES INTO THE TORCH MACRO-STRUCTURES
On the EM burning surface is observed the process of self-organizing of the dynamic

dissipative synergetic SPMS into the torch macro-structures. In this connection, on the
burning surface the process of moving (wandering) of the torches is observed. Besides, the
192
Czech Republic
2006
[Content]
torch structures on the burning surface can exist steadily in the event that they form a pair of
two torches.
The SPMS (Cells) of The

Thermo-Electric Convection
SELF-ORGANIZING
OF THE SPMS
FORMATION OF THE CARBON
MICRO-STRUCTURES
SELF- RGANIZING OF THE SPMS INTO

THE TORCH MACRO -STRUCTURES
7.
THE PHENOMENON OF THE END-BURNING SOLID

PROPELLANT CHARGES RHYTHMIC EXTINCTIONS

phenomenon of the end-burning solid propellant charges rhythmic extinctions. This
phenomenon is observed enough frequently both for small-sized (Fig. 12) and for large-sized
end-burning charges of the gas-generators and SPS (engines for control on a roll channel, the
nose cone engines, having diameter up to 0.4 m.). This anomaly of burning for the first time
was described in details by Dr. Valentin N. Fomenko in 1990 (the Federal Centre of Dualuse Technologies "SOYUZ", Dzerzhinsky, Moscow region, Russia), on the basis of use of
the results of large number of the fire stand tests. The phenomenon is typically observed only
for charges with insulation layer.
13
Alexander Lukin
193
Czech Republic
2006
[Content]
Fig 12. Sectional view of the solid propellant gas-generator. [17]
Fig 13. Combustion products pressure developnent at the end-burning solid propellant
charges rhythmic extinctions. Curves 1-4 corresponds to the solid propellant rocket
motor charge, manufactured from the standard ballistite propellant NMF-2 220/0400. Curves 5 and 6 corresponds to the gas generator propellant charge,
manufactured from the standard ballistite propellant NDT-OM 60/0-500. Figure
shows curves at different ambient temperatures: 1, 4 293 K; 2, 5 223 K; 3,6 323
K; Curve 4 corresponds to the situation at flame propagation under the charge
armoring cover.
In the conditions of increased pressure, inflow of the inhibitor of burning from the
insulation layer on the end-burning surface will increase viscosity and density of the LVL
and will lead to increasing of sizes of the SPMS and the burning cells. In this situation the
quantity of the burning cells will decrease. Accordingly, more dense and steady carbon grid
on the burning surface will be formed (as in conditions of the low pressures). Periodical
throwing away of the carbon grid, formed at the burning surface causes extinction and the
subsequent flashes that is recorded on the pressure curve.
194
Czech Republic
2006
[Content]
It is important to note, that in the majority of theoretical models of the low-frequency

non-acoustic instability phenomenon in the SPS, the processes of excitation of the SPMS on
the burning surface are not taken into account. Thus, suggested mechanism and theoretical
model opens possibilities for understanding the essence of the EM unstable burning
phenomena on the new qualitative level. Such an improved model will be a step towards
detailed interpretation of the large amount of experimental data on L* instability that is
available.
8.
CONTROLLING OF THE ABNORMAL PHYSICO-CHEMICAL

PROCESSES AT THE ENERGETIC MATERIALS BURNING
Actuality of the researches connected with controlling of the abnormal physico-chemical

processes at the EM burning is evidently demonstrated by continuous perfection of the
designing-technological development in the given scientific-technical area. Let us examine
some new Russian technologies intended for effective suppression of the unstable
combustion and abnormal modes of the SPS operation. These technologies are created on the
basis of traditional theoretical ideas about the EM combustion instability (Table 1).
Table 1. Technologies for the combustion instability suppression in the SPS.
Design Schemes and Brief Description
The SPS with acoustic
damper, manufactured
as the cylindrical twolevel chamber. Device
provides
effective
suppression
of
vibrating burning in
the wide frequencies
band. [18]
The SPS with largelengthened casing. The
engine design provides
effective suppression of
longitudinal
pressure
pulsations
(longitudinal
acoustic oscillations) with
low frequencies. [19]
The SPS contains main and additional
chambers, connected by subsonic nozzle. The
engine design provides effective suppression
and stabilization of the low and highfrequency combustion products oscillations at
the ignition system operation and at
appearance of the charge unstable burning. [20]
15
Alexander Lukin
195
Czech Republic
2006
[Content]
The SPS contains the acoustic cavity of given

sizes, and at the solid propellant charge end
faces are installed the elastic boots. The engine
design provides effective suppression of the
high- and low-frequency pressure oscillations at
In the SPS, between the front bottom and
the engine casing the perforated thin
diaphragm is installed, and between the
diaphragm and the front bottom are
located the acoustic cavity. Oscillations
of the front diaphragm provide additional
suppression of the combustion products
acoustic oscillations. [22]
The SPS contains a sectional solid propellant

charge and the acoustic cavities, located
between the charge sections. The engine design provides stability of the internal ballistics
characteristics by suppression of the pressure
oscillations in the wide frequencies band, at
Mechanical device installed in the

SPS charge channel: longitudinal
and cross perforated plates and
diaphragms. [5]
Method of ignition provides increased

reliability of the solid propellant rocket
motor with multigrain charge with large
relative elongation in wide temperature
range. [24]
Suppression of the combustion instability can use various mechanical devices installed in
the combustion chamber or in the charge channel: the resonant rods of various forms;
196
Czech Republic
2006
[Content]
longitudinal and cross perforated plates and diaphragms; various screens. Other versions of
technologies for suppression of combustion instability are purposeful change or regulation of
form and geometrical characteristics of the combustion chamber; exact tuning of geometrical
characteristics of the chamber or its separation on several volumes; creation of resonant
acoustic cavities in the combustion chamber; installation of damping rings (diaphragms) and
also internal and external acoustic (gas dynamics) dampers (Helmholtz resonators). Besides,
in some cases for suppression of the combustion instability there may be used high-porosity
cellular materials in the form of disks installed in the combustion chamber. Such disks are
usually manufactured by means of powder metallurgy.
At the same time there are deficiencies of these technologies. The efficiency of the
mechanical damper decreases after burning out a portion of the propellant charge, in
connection with change of the cavity sizes, because the mechanical means of suppression of
unstable burning operates in a narrow frequencies band. Also, damper design elements
induce the energy losses due to braking of the combustion products flow. Use of mechanical
dampers leads to increase of the engine inert weight. In some cases there may be damper
destruction and ejection of its separate elements from the engine. Absorbers of resonant type
the acoustic dampers (Helmholtz resonators) operates effectively in a narrow frequencies
band and can induce transition of unstable mode to other frequency or lead to destabilization
of steady burning process in the engine.
The main source of instability is a solid propellant pulsating burning surface. However,
the existing technologies for suppression of the solid propellant combustion instability do
not take into account the influence of the synergetic micro-structures on the propellant
burning surface.
Further progress in this vital area may be connected with understanding of mechanisms
of formation and destruction of the synergetic micro-structures on the EM burning surface.
In the SPS of a new generation with high energy and mass characteristics, the problem
of prevention of development of combustion instability and anomalies of burning have
extreme importance. Solution of this problem can be provided with using of suggested
mechanism and with using of plasma physics technologies for control of the process of
excitation of the SPMS on the burning surface.
For control of the process of the SPMS excitation on the EM burning surface the effect
of acoustic cavitation in the LVL can be used. This effect can be organized in the LVL under
influence of the acoustic field with certain frequency.
9.
CONCLUSION
The low-frequency non-acoustic combustion instability phenomenon, the chuffing

phenomenon and the accompanying physico-chemical effects have received a new
explanation within the concept based on the data of optical visualization of the physicochemical processes on the EM burning surface. On the EM burning surface is observed the
process of self-organizing of the dynamic dissipative synergetic SPMS into the torch macrostructures.
Understanding of the mechanism of the torch combustion phenomenon of the solid
rocket propellants - one of the most complicated problem in the solid propulsion physics.
17
Alexander Lukin
197
Czech Republic
2006
[Content]
The torch micro-structures on the EM burning surface can be considered as independent

synergetic structures and can exist steadily in the event that they form a pair of two torches.
Excitation of the spatial-periodic micro-structures in the ionic fusion with thermoelectric properties on the burning surface is a main source of development of the synergetic
phenomena in the EM burning zone.
phenomenon of the end-burning solid propellant charges rhythmic extinctions.
The existing technologies for suppression of the solid propellant combustion instability
do not take into account the influence of the synergetic micro-structures on the propellant
burning surface. Further progress in this vital area may be connected with understanding of
mechanisms of formation and destruction of the synergetic micro-structures on the EM
burning surface.
Design community and quality assurance people should continue to give due importance
to the above problem of controlling low-frequency non-acoustic combustion instability of
the energetic materials in practical applications so that their guidelines will be realistic for
safety.
Acknowledgments:
Author would like express deepest gratitude for the extensive experimental data
provided by Prof. Valery S. Abrukov (Chief of the Chair of Thermo-Physics of the Chuvash
State University, Cheboksary, Russia), by Dr. Valery D. Kochakov (Associate Professor of
the Chair of Thermo-Physics of the Chuvash State University, Cheboksary, Russia), by Prof.
Vladimir N. Marshakov (Chief of the Group of Non-Stationary Combustion of the
N.N.Semenov Institute of Chemical Physics of the Russian Academy of Sciences, Moscow,
Russia) and by Prof. Nickolay M. Pivkin (Chief of the Department No. 012 of the ScientificResearch Institute of Polymeric Materials, Perm, Russia).
Nomenclature
a
C
Ea
FL
FST
FTE
g
h
k0
P
q0
198
- Thermal diffusivity of the liquid-viscous layer;

- Specific heat capacity of the condensed phase;
- Activation energy of the condensed phase reactions;
- Buoyancy (lifting) force (the difference between Archimedean force and the
force of gravity), (Rayleigh, 1916);
- Surface tension force (thermo-capillary force), (G.K.A. Pearson, 1958), (the
convection cells arising under influence of this mechanism is known as
Maringoni cells);
- Thermo-electric (electrostatic) force, Coulombs force, (I.V.Ioffe and
E.D.Eidelman, 1976);
- Acceleration due to gravity;
- Characteristic size of fluctuation movement (the thickness of the liquidviscous layer, cross-sectional size of the cell);
- Pre-exponential factor of chemical reaction rate;
- Pressure;
- Heat release distribution due to chemical reactions in the solid phase and
liquid-viscous layer of the cell;
Czech Republic
2006
[Content]
liquid-viscous layer of the cell;

- Heat flux issuing from the gaseous phase (flame) to the surface of the
burning cell;
- Gas constant;
- Time of the process;
- Temperature;
- Initial temperature;
Q0
R
t
T
T0
- Temperature of the cold boundary surface of the liquid-viscous layer;

- Temperature of the hot boundary surface of the liquid-viscous layer (on the
interface of liquid-viscous layer and gasification zone);
- Temperature of beginning of the effective exothermic reactions;
TC
Th
TR
- Temperature on the burning surface;

- Linear burning rate of the energetic material sample;
- Linear burning rate of the energetic material sample in the cell with number
(i, j);
- Space coordinate, perpendicular to the liquid-viscous layer;
TS
uC
u (Ci , j)
Greek Symbols
0
C
- Average thickness of the reaction liquid-viscous layer (the effective thickness

of the zone of electric conductivity);
- Temperature gradient;
- Conductivity;
- Density of the liquid-viscous layer, density of the condensed phase;
Subscripts and Superscripts

c
g
s
0
Condensed phase;
Gas phase;
Surface;
Parameters of equilibrium state;
REFERENCES
[1]
A.M. LIPANOV, A.N.LUKIN: Abnormal Physics-Chemical Processes in the Technical Systems,

Based on Combustion of the Solid Propellants and Gun-powders, The Concept of Development
of Combustion and Explosion as a Field of Scientific-Technical Progress, Ed. by Acad.
A.G.Merzhanov, Publ. House Territoriya, Chernogolovka, p. 124, 2001, (in Russian)
[2]
YA.B. ZEL'DOVICH: About Stability of Burning of Gun-powder in the Semi-Closed Cavity, J. of

Appl. Mech. and Tech. Phys., No. 1 , p. 67-76, 1963, (in Russian)
[3]
I.G.ASSOVSKII: The Critical Phenomena and Non-Stationary Burning of the Energetic

Materials, Dr. of Sci. Thesis (Phys.-Math., Spec. 01.04.17 Chem. Phys., incl. Physics of
19
Alexander Lukin
199
Czech Republic
2006
[Content]
Combustion and Explosion), N.N.Semenovs Institute of Chemical Physics of the RAS,

Moscow, Russia, p. 360, 2001, (in Russian)
[4]
R.A. YOUNT and T.A.ANGELUS: Chuffing and Non-acoustic Instability Phenomena in Solid
Propellant Rockets, AIAA Journal, vol. 2, No. 7, p. 1307-1313, 1964
[5]
N.M. PELYKH: Non-Stationary Burning of the Solid Propellant Charges and Use of This Effect
in the National Economy, Dr. of Sci. Thesis (Tech., Spec. 01.02.05 Mech. of the Liquid, Gas
and Plasma and 01.04.17 Chem. Phys., incl. Physics of Combustion and Explosion),
Scientific-Research Institute of Polymeric Materials, Perm, Russia, p. 244, 2003, (in Russian)
[6]
C.M. HUGGETT: Combustion Processes, High Speed Aerodynamics and Jep Propulsion,
Princeton University Press, Princeton, N.J., vol. II, Sec. 4, 1956
[7]
G.B. COOK: Some Developments in the Theory of Thermal Explosions, Sixth Symposium
(International) on Combustion, Reinhold Publishing Corp., New York, p. 621-631, 1957
[8]
V.S. ABRUKOV, S.V.ILYIN, V.M.MALTSEV: Interferometric Techniques and Other Optical

Methods in Combustion Research. A New Approach, Optical Techniques in Fluid, Thermal, and
Combustion Flow, San-Diego, USA, Ed. Soyong Stephen Cha and James D. Trolinger, Proc.
Of the SPIE's Int. Symp., vol. 2546, p. 420-426, 1995
[9]
V.S. ABRUKOV, S.V.ILYIN, V.M.MALTSEV, I.V.ANDREEV: Interferometric Technique in

Combustion, Gas Dynamic and Heat Transfer Research. New Results and Technologies. CDROM Proc. of the VSJ-SPIE98 Int. Conference on Optical Technologies and Image Processing
in Fluid, Thermal, and Combustion Flow, Yokohama, JAPAN, AB076, p. 13, 1998
[10] V.S. ABRUKOV, A.V. KHRISTOFOROV, V.YE. NIKONOROV, I.V. ANDREEV: Interferometric and
Shadow Technique in Analysis of Combustion Wave Shaping and Development During Ignition
of Solids, Optical Diagnostics for Fluids/Heat/Combustion and Photo-mechanics for Solids, Ed.
by Soyoung S. Cha, Peter J. Bryanston-Cross, Carolyn R. Mercer, Proc. Of the SPIE's Int.
Symp., USA, Denver, vol. 3783, p. 5, 1999
[11] V.T. KUZNETSOV, V.S. AVERIN: Ignition of Heterogeneous System in the Dead-end Channel
with Variable Cross-Section Area, Proceedings of the Fourth International Workshop on IntraChamber Processes, Combustion and Gas Dynamics of Dispersed Systems, (June 28 - July 3,
2004, Baltic State Technical University, Saint-Petrsburg, Russia), vol. 1, Baltic State Technical
University, Saint-Petrsburg, Russia, p. 52-54, 2004, (in Russian)
[12] YU. S. IVASHENKO, A. L. SADYRIN, and V. L. PAVLENKO: Investigation of the Electric
Conductivity of the Ammonium Perchlorate Burning Surface, J. Physics of Combustion and
Explosion, vol. 22, No. 4, p. 30-33, 1986, (in Russian)
[13] G. V. MELIK-GAIYKAZOV, V. N. MARSHAKOV: Investigation of the Electric Conductivity of
the Ammonium Perchlorate at Burning, J. Physics of Combustion and Explosion, vol. 23, No.
4, p. 15-20, 1987, (in Russian)
[14] G. V. MELIK-GAIYKAZOV: Experimental Investigations of the Gunpowder Non-Stationary
Burning Modes in the Semi-Closed Cavity, Ph.D. Thesis (Phys.-Math. Sci., Spec. 01.04.17
Chem. Phys., incl. Physics of Combustion and Explosion), N.N.Semionov Inst. of Chem. Phys.
RAS, Moscow, p. 147, 1998, (in Russian)
200
Czech Republic
2006
[Content]
[15] A. N. LUKIN: Concept of the Spatial-Periodic Micro-Structures Excitation at the Evaporated

Energetic Materials Transient Combustion as a Way for Improvement of the Solid Propulsion
Technologies, in: Proc. 8th Int. Seminar New Trends in Research of Energetic Materials, Part
I, University of Pardubice, Czech Republic, p. 267-295, 2005
[16] D. A. FRANK-KAMENETSKII: Diffusion and Heat Transfer in Chemical Kinetics, Publishing
House "Nauka", Moscow, p. 502, 1987, (in Russian)
[17] A. M. LIPANOV, A.YU. LESCHEV: Ignition System for Gas-Generator, Pat. 2 239 474 of
Russia, A 62 C 13/22, Application 2002133993/02, Priority from December 17, 2002, (in
Russian)
[18] N. N. GLUKHAREV ET AL: The Solid Propellant Rocket Motor, Pat. 2 056 519 of Russia, F
02 K 9/08, Application 93035598/06, Priority from July 8, 1993, Registered at March 20,
1996, The Inventions Bulletin (Russia) No. 8, March 20, 1996, (in Russian).
[19] M. I. SOKOLOVSKII ET AL: The Solid Propellant Rocket Motor, Pat. 2 173 783 of Russia, F
02 K 9/26, Application 2000111590/06, Priority from May 10, 2000, Registered at
September 20, 2001, The Inventions Bulletin (Russia) No. 26, September 20, 2001, (in
Russian).
[20] O. G. BORISOV ET AL: The Solid Propellant Rocket Motor, Pat. 2 125 173 of Russia, F 02 K
9/08, Application 98103166/06, Priority from February 20, 1998, Registered at January 20,
1999, The Inventions Bulletin (Russia) No. 2, January 20, 1999, (in Russian).
[21] V. R. ALYAZHEDINOV ET AL: The Solid Propellant Rocket Motor, Pat. 2 147 342 of Russia,
F 02 K 9/26, Application 98119256/06, Priority from October 23, 1998, Registered at April
10, 2000, The Inventions Bulletin (Russia) No. 10, April 10, 2000, (in Russian).
[22] P. I. TRAPEZNIKOV ET AL: The Rocket, Pat. 2 170 910 of Russia, F 42 B 15/00, Application
2000103008/02, Priority from February 10, 2000, Registered at July 07, 2001, The
Inventions Bulletin (Russia) No. 20, July 07, 2001, (in Russian).
21
Alexander Lukin
201
Czech Republic
2006
[Content]
STABLE SALTS OF METHYLATED 5-AMINOTETRAZOLES

Michael Gbel*, Konstantin Karaghiosoff*, Thomas M. Klaptke*, Carles Mir*, Jan
M. Welch*
*University of Munich, Butenandstr. 5-13 (D), D-81377 Munich, Germany
Abstract:
Several salts of 1-methyl-5-aminotetrazole (2), 2-methyl-5-aminotetrazole (3) and 1-4dimethyl-5-aminotetrazole (free base of 4) with a variety of anions: nitrate (5, 7, 10),
perchlorate (6, 8, 11), picrate (9,15), azide (12), dinitramide (13) and azotetrazolate (14) as
the dihydrated species have been successfully synthesized in reasonable to good yields and
characterized by analytical (elemental analysis and melting point) and spectroscopic (IR,
Raman and 13C-, 1H-, 15N-NMR) methods. The structures of all compounds have been
determined using X-ray diffraction techniques. All compounds were found to be stable at room
temperature and moisture and air stable. An insight into their sensitivity was gained by either
throwing them into a flame or by submitting them to standardized BAM tests for friction,
impact and electrostatics sensitivity. All perchlorates (6, 8, 11) and the dinitramide (13)
exploded, however the azide (12), surprisingly, and all other compounds subject of this study
remained mainly unmodified after the test.
Keywords:
1.
methylated 5-aminotetrazoles, high energy density material (HEDMs),

crystal structure, vibrational spectroscopy, NMR spectroscopy
(multinuclear)
INTRODUCTION
The synthesis of energetic, non-nuclear materials for possible military application has
been a long term goal in our research group [1,3]. In the last years there has been great
interest in investigating the chemistry of new high energy density materials (HEDMs) with a
high-nitrogen content and very high positive heats of formation (directly attributable to the
large number of inherently energetic N-N and C-N bonds) that show insensitivity to shock,
friction, electrostatic discharge [4] and thermal stimuli. These compounds combust ideally to
give only gaseous products (i.e. CO, CO2, and N2). Most of the energy obtained from
modern energetic materials comes either from oxidation of the carbon backbone (such is the
case for traditional energetic materials as for example HMX or TNT) [5] or from their high
positive heats of formation, as in for tetrazole azide [6].
Recently, the synthesis of new heterocyclic-based energetic, low melting salts were
reported [7]. These materials that are salt-based often possess advantages over nonionic
structures since they normally exhibit very low vapor pressures, eliminating, thus, the risk of
exposure through inhalation, and they generally have higher densities than the nonionic
parents [8].
202
Czech Republic
2006
[Content]
Salts containing a tetrazole derivative as the cation (which adds a high nitrogen-content
to the molecule) show high densities and interesting energetic properties when combined
with typical energetic anions such as nitrate in 5-aminotetrazolium nitrate [9] or dinitramide
(N(NO2)2) in 1,5-diamino-4-methyltetrazolium dinitramide [10]. Therefore, tetrazoles and
aminotetrazoles, in particular, are prospective materials for the generation of gases, as
blowing agents, solid propellants, and other combustible and thermally decomposing
systems. Based on these results 5-aminotetrazole 1 (with a 82.3% nitrogen-content) was
considered as a valuable starting material for the synthesis of nitrogen-rich HEDMs. The
relative basicity of 1 allows to prepare HEDMs by simple protonation of one of the ringnitrogens with a strong acid such as nitric [10] or perchloric acid [11]. Unfortunately, several
attempts to generate the interesting azide, dinitramide and azotetrazolate parents failed,
apparently due to the higher acidity of 5-aminotetrazole in comparison to the conjugated acid
of the above-mentioned anions. In order to overcome this difficulty and to make more stable
5-aminotetrazole derivatives, introduction of a small alkyl chain was thought to be of use.
Methylation of 1 according to literature procedure [12] with methyl iodide in acetone/water
afforded a mixture of 1-methyl-5-aminotetrazole 2 and 2-methyl-5-aminotetrazole 3 which
can be separated. Salts 5-9 were synthesised and again the methylated 5-aminotetrazoles
turned out to be too acidic to yield the azide, dinitramide or azotetrazolate salts. For this
reason, the quaternization of the tetrazole ring with great excess methyl iodide in acetonitrile
[13]
was carried out to yield dimethyl-5-aminotetrazolium iodide 4 of which a variety of salts
(10-15) with typical energetic anions were prepared by metathesis.
NH2
NH2
Pic-
PicHN
N
N
N
N
9
Picric acid
H2O reflux
NH2
NH2
NH
N
MeI
Acetone/H2O
reflux, 2h
MeI
CH3CN
reflux, 10h
N
N
HX
NH2
HN
AgY
or
AgN(NO2)2*2Py
NH2
N
N
HN
MeOH
NH2
N
N
3
HX
MeOH
reflux
NH2
i) AgClO4
ii) KPicrate
NH2
15
5. X = NO 3-
7. X = NO 3-
6. X = ClO4-
8. X = ClO4-
10. Y = NO 311. Y = ClO412. Y = N3
13. Y = (NO2)2N14. Y = [N4C-N=N-CN4]2-
Scheme 1. Reaction equations for the synthesis of the methylated 5-aminotetrazole salts
Carles Mir, Michael Gbel, Konstantin Karaghiosoff, Thomas Klaptke, . . .
203
2.
Czech Republic
2006
[Content]
EXPERIMENTAL
2.1
Synthesis
Caution! Although no problems came up during the synthesis and handling of all
materials studied in this work, silver azide, silver dinitramide, aminotetrazolaes and their
derivatives are energetic materials and tend to explode under certain conditions. All
HEMDs should be treated with respect and appropriate safety precautions should be taken
at all times. Laboratories and personnel should be properly grounded and safety equipment
such as Kevlar gloves, leather coat, face shield and ear plugs are necessary, specially for
perchlorates 6, 8 and 11 as well as for dinitramide 13, in particular when working with big
amounts.
All chemicals and solvents (analytical grade) were used as supplied by Sigma-Aldrich
Fine Chemicals Inc.. 2-Methyl-5-amino-1H-tetrazole, 1-methyl-5-amino-1H-tetrazole [12],
silver azide and dinitramide [14] as well as disodium azotetrazolate [15] were prepared
according to literature procedure.
2.1.1.
2-Methyl-5-amino-1H-tetrazolium nitrate (5):
Method 1: 2-methyl-5-amino-1H-tetrazole (0.100 g, 1.0 mmol) was dissolved in the

minimum amount of concentrated nitric acid. The clear, colorless mixture was stirred until
all 1-methyl-5-amino-1H-tetrazole had dissolved. After cooling in the fridge overnight,
colourless plates were formed, filtered off cold and washed with diethyl ether until acid
free. No further purification was necessary (0.081 g, 49 %).
Method 2: Alternatively, 2-methyl-5-amino-1H-tetrazole (0.229 g, 2.3 mmol) was
dissolved in little concentrated nitric acid and 2 mL acetonitrile were added. The clear,
colorless mixture was stirred until 2-methyl-5-amino-1H-tetrazole had completely
dissolved and the product was precipitated by addition of 200 mL pentane. The yellow
powder could be washed with pentane and the out-coming white powder recrystallized
from methanol by diffusion of ether to yield clear colourless crystals, which were filtered
off and washed with ether. (0.337 g, 90 %). These crystals were suitable for X-ray
structure determination.
2.1.2
2-Methyl-5-amino-1H-tetrazolium perchlorate (6):
2-methyl-5-amino-1H-tetrazole (0.180 g, 1.8 mmol) was dissolved in little

concentrated perchloric acid. 1 mL ether and 0.5 mL acetone were added to the clear
yellowish solution and ether was diffused into it over 2 days. Little colorless prismic
crystals separated from the mother solution which allowed us to measure the crystal
structure. After filtering and washing with ether the compound required no further
purification (0.267 g, 70 %)
204
2.1.3
Czech Republic
2006
[Content]
To 2 mL concentrated nitric acid were added 1.620 g (10.0 mmol) 1-methyl-5-amino1H-tetrazole. The clear, colorless reaction mixture was stirred until all 1-methyl-5-amino1H-tetrazole had dissolved. The solution was then cooled to 0 C overnight resulting in the
formation of clear colorless crystals, which were filtered off cold and washed until dry and
acid free with diethyl ether. No further purification was necessary and the crystals were
already suitable for X-ray analysis (1.206 g, 73 %).
2.1.4
1-Methyl-5-aminotetrazole (0.142 g, 1.8 mmol) was dissolved with little concentrated

perchloric acid. The mixture was layered with ether and cooled in the fridge overnight to
yield colorless plates which were filtered off and washed with ether. No further purification
was necessary. Alternatively, ether can be diffused into the mixture over 2 days yielding nice
colorless crystals suitable for X-ray analysis (0.204 g, 63 %).
2.1.5
1-Methyl-5-amino-1H-tetrazolium picrate (9):
A solution of 1-methyl-5-aminotetrazole (0.179 g, 1.8 mmol) in 2 mL hot water was

added to a solution of picric acid (0.398 g, 1.7 mmol) in 3 mL hot water. The mixture was
boiled together for a few minutes and let to cool down slowly to room temperature.
Immediately yellow prismic crystals suitable for X-ray structure determination precipitated.
The crystals were filtered off, washed with cold acetone and ether. A second crop could be
obtained from cooling down the mother liquors in the fridge overnight and filtering off cold.
This was treated like the first one. Both crops were pure by EA (0.521 g, 91 %).
2.1.6
1,4-Dimethyl-5-aminotetrazolium nitrate (10):
Silver nitrate (10.878 g, 64.0 mmol) was added to a suspension containing 1,4-dimethyl5-aminotetrazolium iodide (15.436 g, 64.0 mmol) in 50 mL methanol. Immediate
precipitation of yellow silver iodide was observed and the mixture was stirred for 2 hrs.
under exclusion of light. The solid was filtered off and the solvent reduced under high
vacuum to about 30 mL. The solution was divided into 10 equal portions into which diethyl
ether was allowed to diffuse over 2 days yielding clear prismic crystals the structure of
which was determined by single crystal X-ray analysis. No further purification was
necessary (5.912 g, 53 %).
205
2.1.7
Czech Republic
2006
[Content]
1,4-Dimethyl-5-aminotetrazolium perchlorate (11):
To 20 mL methanol were added 4.820 g (20.0 mmol) 1,4-dimethyl-5aminotetrazolium Iodide. The solution was stirred until all solid had dissolved. A second
solution containing 4.140 g anhydrous silver perchlorate weighed out in a dry-box
dissolved in 5 mL methanol was added to the solution of 1,4-dimethyl-5-aminotetrazolium
iodide. White silver iodide precipitated immediately and the temperature of the reaction
mixture rose near to the boiling point of methanol. After stirring for 30 minutes the
solution was filtered, divided into 12 equal portions and recrystallized by diffusion of ether
into the methanol solutions. The recrystallized material was used to measure the X-ray
structure in the solid state. No further purification was necessary (2.812 g, 71 %).
2.1.8
1,4-Dimethyl-5-aminotetrazolium azide (12):
Method 1: To a suspension of 0.200 g (2.5 mmol) potassium azide in 2 mL of

methanol were added 0.430 g (2.0 mmol) of 1,4-dimethyl-5-aminotetrazolium perchlorate.
The mixture (suspension) was covered and stirred for 3 hrs. After 3 hrs. the remaining
white solid was filtered off and the solution was divided into two equal portions. Diethyl
ether was allowed to diffuse into both solutions over 2 days, yielding clear prismic crystals.
No further purification was necessary (0.258 g, 66 %).
Method 2: Alternatively, to a solution of 0.240 g (1.0 mmol) of 1,4-dimethyl-5aminotetrazolium iodide in 3 mL methanol were added 0.200 g (1.33 mmol) silver azide.
This suspension was stirred for 1 hour and then filtered. The remaining solution was
divided into three 1 mL portions into which diethyl ether was allowed to diffuse over
several days yielding again clear prismic crystals suitable for X-ray crystal structure
determination. No further purification was necessary (0.122 g, 78 %).
2.1.9
1,4-Dimethyl-5-aminotetrazolium dinitramide (13):
Method 1: To a suspension of 0.220 g (1.5 mmol) potassium dinitramide in 2 mL

methanol were added 0.210 g (1.0 mmol) 1,4-dimethyl-5-aminotetrazolium perchlorate.
The suspension was stirred for 3 hours then filtered. The resulting clear solution was
divided into two equal portions and diethyl ether was allowed to diffuse into the methanol
solutions overnight yielding clear colorless crystals suitable for X-ray structure
determination (0.220 g, 67 %).
Method 2: Alternatively, to a suspension of 0.370 g (1.0 mmol) of silver bis-pyridine
dinitramide in 2 mL methanol were added 0.240 g (1.0 mmol) of 1,4-dimethyl-5aminotetrazolium iodide. The reaction was stirred under exclusion of light for 30 minutes,
the silver iodide was then removed by filtration and the remaining solution was divided
into two equal fractions and diethyl ether was allowed to diffuse into the solutions
overnight yielding clear, colorless needles (0.157 g, 71 %).
206
2.1.10
Czech Republic
2006
[Content]
1,4-Dimethyl-5-aminotetrazolium azotetrazolate*2H2O (14):
1,4-Dimethyl-5-aminotetrazolium iodide (0.378 g, 1.6 mmol) was dissolved in the

minimum amount of hot water and was added to a solution of disodium azotetrazolate (0.140
g, 0.7 mmol) in little hot water. The reaction mixture was refluxed for one hour and leave to
slowly cool down to room temperature. A feathery yellow powder precipitated on standing
in the oil bath and the yield could be increased by cooling down the solution in the fridge
(0.235 g, 78 %). X-ray structure determination from some of the material precipitated in the
fridge confirms the presence of two water molecules in the unit cell.
2.1.11
1,4-Dimethyl-5-aminotetrazolium picrate (15):
A fresh solution of 1,4-dimethyl-5-aminotetrazolium perchlorate prepared from reacting

over 1 hour 1,4-dimethyl-5-aminotetrazolium iodide (0.845 g, 3.5 mmol) with silver
perchlorate (0.730 g, 3.5 mmol) in 15 mL methanol was added (after filtering off yellow
silver iodide) to a suspension of orangey potassium picrate prepared from stirring during 30
min. a mixture of picric acid (0.803 g, 3.5 mmol) and potassium hydroxide (0.197 g, 3.5
mmol) in 15 mL methanol. The reaction mixture was stirred under reflux for 3 hours. Yellow
potassium perchlorate was filtered off hot and the reaction mixture left to cool down to room
temperature. Yellow needles precipitated immediately. The solvent was left overnight to
evaporate yielding a yellow crystalline material (0.894 g, 75 %) which could be used to
measure the crystal structure.
2.2
Crystal structures
Crystals were obtained as already mentioned above. X-ray quality crystals were mounted
on a glass fibre, and the X-ray crystallographic data were collected on a Oxford Xcalibur
diffractometer with a CCD area detector with graphite-monochromated Mo K radiation (
= 0.71073 ). The structures were solved by direct methods and refined by means of fullmatrix least-squares procedures using the SHELX-97 program. Fig. 1 to Fig. 4 show the
molecular structure of selected synthesized compounds in the crystalline state (DIAMOND
plot, 50 % probability):
207
2.2.1
Czech Republic
2006
[Content]
Table 1. Crystallographic data for 5
2MATN
Formula
Form. weight (g mol-1)
Crystal system
Space-group
a ()
b ()
c ( )
()
()
()
Cell volume (3)
Calc. density (g cm-3)
Z
R1(F>4 (F))
wR2 (all data)
C2H6N6O3
162.11
monoclinic
P 21/n
6.340(1)
8.704(2)
12.223(2)
90
99.88(3)
90
664.6(3)
1.620
4
0.0514
0.1502
Fig. 1. Molecular structure for 5
2.2.2

1MATP
Formula
Crystal system
Space-group
a ()
b ()
c ( )
()
()
()
Cell volume (3)
Z
R1(F>4 (F))
wR2 (all data)
C2H6N5O4Cl
199.55
monoclinic
P 21/n
5.2741(8)
20.696(3)
7.149(1)
90
106.60(1)
90
747.8(3)
1.772
4
0.0535
0.1167
208
2.2.3
Czech Republic
2006
[Content]
1,4-Dimethyl-5-aminotetrazolium dinitramide (13):
DMATDN
13
Formula
Crystal system
Space-group
a ()
b ()
c ( )
()
()
()
Cell volume (3)
Z
R1(F>4 (F))
wR2 (all data)
C3H8N8O4
220.15
orthorhombic
P 2 12 12 1
5.3985(6)
12.926(2
13.336(2)
90
90
90
930.5(2)
1.513
4
0.0622
0.1375
2.2.4
1,4-Dimethyl-5-aminotetrazolium picrate (15):

DMATPic
15
Formula
Crystal system
Space-group
a ()
b ()
c ( )
()
()
()
Cell volume (3)
Z
R1(F>4 (F))
wR2 (all data)
C9H10N8O7
342.25
monoclinic
P 21/c
14.862(1)
5.7352(4)
16.839(1)
90
104.96(1)
90
1386.7(2)
1.639
4
0.0649
0.1475
209
2.3
Czech Republic
2006
[Content]
Sensitivity data
The following table indicates the results obtained from submitting the studied
compounds to different stimuli. The figures in the boxes indicate the sensitivity obtained
from the minimum amount of weight and height that detonates a compound at least two
times out of six for the drop-hammer (impact sensitivity test, in Joules) and the minimum
force which needs to be applied to detonate a compound two out of four times for the
friction test (determined using a BAM friction tester [16], in Newtons). For the impact tests
about 30-40 mg of the sample were placed in a brass cup and hit with a weight (BAM
Fallhammer) in accordance to the BAM method [17]:
Table 5. Impact, friction, electrostatics and flame sensitivity for the studied compounds*
Impact
Friction
Spark
Flame
5
>30
>360
n.e.
D
6
n.m.
n.m.
n.e.
D
7
>30
>360
n.e.
D
8
2
10
n.e.
D
9
B
>360
n.e.
B
10
>30
>360
n.e.
B
11
6
96
n.e.
D
12
>30
>360
n.e.
B
13
5
>360
n.e.
D
14
n.m.
n.m.
n.e.
D
15
>30
>360
n.e.
B
*n.e. = no explosion, D = deflagration, B = the compound burned, n.m. = not measured
3.
5-aminotetrazole salts are really interesting compounds for the synthesis of HEDMs
due to their high-nitrogen-content and to the fact that they usually exhibit good thermal
stability. In order to increase the nitrogen content and/or the oxygen balance of these salts,
highly energetic anions such as dinitramide, azide, azotetrazolate ([N4C-N=N-CN4]2-),
nitrate, perchlorate or picrate where selected as candidates. Unfortunately, for the first
three anions, from our experience, the acidity of a hydrogen attached to a 5-aminotetrazole
nitrogen makes it impossible to synthesize their 5-aminotetrazole salts unless
quaternization of the tetrazole ring is carried out. In this respect, the synthesis of these salts
is limited to the alkylated derivatives which increases considerably the stability of the salts
but it has the drawback that diminishes the nitrogen-content which would readily combine
to give molecular nitrogen under conditions which initiate the explosion. This relates, in
turn, with the performance of a HEDM.
210
Czech Republic
2006
[Content]
Several stable salts of the 5-aminotetrazole methylated derivatives (2-4) with a nitrogencontent varying between 34.4 % for tetrazole 9 and 71.8 % for 12 have been prepared in
reasonable to good yields by either protonation of the tetrazole ring or metathesis reactions.
Their sensitivity to different stimuli has been assessed by using the drop-hammer test, a
friction tester, initiating an electrical spark or throwing the compounds into a flame. Most of
them showed to be stable and, in consequence, safe to handle apart from perchlorates 8 and
11 (perchlorate 6 has not been yet tested) which with impact sensitivity values of 2 J and 6 J
turned out to be rather unstable. On the other side, the high sensitivity of 8 to friction (10 N)
is in contrast with a medium friction sensitivity for 11 of 96 N. This just puts into
perspective what would have been expected: the introduction of small alkyl groups on the
tetrazole ring increases the stability of the compounds to stimuli as well as the melting points
of the salts, which are 123 C and 220 C (for 8 and 11 respectively). Regardless of the wellknown explosive character of many azides, compound 12 showed to be really stable and
with a melting point of 186 C (well far-away from its decomposition temperature above 350
C) and a high-nitrogen content (71.8 %) could be considered as a candidate for further
studies. Dinitramide 13, which is quite impact sensitive (5 J) although friction unsensitive
(>360 N), showed much better friction stability than perchlorate 11 (96 N), regardless of the
similar impact sensitivity values (6 J for the perchlorate). The high-melting points of the
prepared salts, well-apart from its decomposition temperature (at least 50 C, only 16 C in
the case of nitrate 7), make the synthesised compounds prospective materials for further
studies.
4.
CONCLUSIONS
From this experimental study the following conclusions can be drawn:

(i)
An easy synthesis for several methylated 5-aminotetrazole salts was found

which yields the compounds in reasonable to good yields.
(ii)
The studied compounds were characterised analytically and by means of

vibrational spectroscopy (IR, Raman), NMR spectroscopy (13C, 1H, 15N) and by
single crystal X-ray diffraction
(iii) An insight into their sensitivity was gained by submitting them to standard
BAM tests.
10
211
Czech Republic
2006
[Content]
REFERENCES
[1] W.-H. WALTHER, T.M. KLAPTKE, AND G. HOLL: Physical Chemical Characterization
and Explosive Properties of Hydrazine Azide, in: Proceedings 29th International Annual
Conference of ICT, June 30 - July 3, 1998, Energetic Materials Production, Processing and
Characterization, Fraunhofer Institut fr Chemische Technologie: Karlsruhe, Germany, 1998,
pp. 134/1
[2] T. M. KLAPTKE, B. KRUMM, G.HOLL, AND M. KAISER: Synthesis, Characterization
and Quantum Chemical Computation of Hexanitrohexaaza-Isowurtzitane, in: Proceedings
30th Int. Annual Conference of ICT, June 30 July 3, 1999, Energetic Materials Modelling
of Phenomena, Experimental Characterization, Environmental Engineering, Fraunhofer
Institut fr Chemische Technologie: Karlsruhe, Germany, 1999, pp. 120/1
[3] T. HABEREDER, A. HAMMERL, G. HOLL, T. M. KLAPTKE, J. KNIZEK, P. MAYER,
AND H. NTH: Methylsubstituted Hydrazinium Azide Derivatives, in: Proceedings 31th Int.
Annual Conference of ICT, June 27 30, 2000, Energetic Materials Analysis, Diagnostics
and Testing, Fraunhofer Institut fr Chemische Technologie: Karlsruhe, Germany, 2000, pp.
150/1-150/7
[4] CHENGFENG YE, JI-CHANG XIAO, BRENDAN TWAMLEY AND JEANNE M.
SHREEVE. Energetic Salts of Azotetrazolate, Iminobis(5-tetrazolate) and 5,5bis(tetrazolate), Chem. Commun., 2005, 2750-2752
[5] J. KHLER AND R. MAYER: Explosivstoffe, 9th ed.; Wiley-VCH: Weinheim, 1998.
[6] A. HAMMERL, T. M. KLAPTKE AND M. WARCHHOLD: Synthesis, Structure,
Molecular Orbital and Valence Bond Calculations for Tetrazole Azide, CHN7, Propellants,
Explosives, Pyrotechnics 2003, 28(4), 163-173
[7] G. DRAKE, T. HAWKINS, A. BRAND, L. HALL, V. A. MCKAY AND I. ISMAIL:
Energetic, Low-Melting Salts of Simple Heterocycles, Propellants, Explosives, Pyrotechnics
2003, 28(4), 174-180. G. W. DRAKE: U. S. Patent 6,509,473, Jan. 21, 2003. G. DRAKE
AND T. HAWKINS: AFOSR Ionic Liquids Workshop, Dulles, VA, October 2002; Air Force
Office of Scientific Research: Arlington, VA. G. DRAKE: AFOSR Ionic Liquids Workshop,
Aberdeen, MD, September 2003; Air Force Office of Scientific Research: Arlington, VA.
[8] HONG XUE, YE GAO, BRENDAN TWAMLEY AND JEANNE M. SHREEVE: New
Energetic Salts Based on Nitrogen-Containing Heterocycles, Chem. Mater. 2005, 17, 191-198
[9] MORITZ VON DENFFER, GERHARD HEEB, THOMAS M. KLAPTKE, GERNOT
KRAMMER, GUNNAR SPIESS AND JAN M. WELCH: Improved Synthesis and X-Ray
Structure of 5-Aminotetrazolium Nitrate, Propellants, Explosvies, Pyrotechnics 2005, 30(3),
191-195
[10] THOMAS M. KLAPTKE, PETER MAYER, AXEL SCHULZ AND JAN J. WEIGAND:
1,5-Diamino-4-Methyltetrazolium Dinitramide, J. Am. Chem. Soc., 2005, 127, 2032-2033
[11] MORITZ VON DENFFER AND THOMAS M. KLAPTKE, unpublished
[12] R. A. HENRY AND W. G. FINNEGAN: Mono-alkylation of Sodium 5-Aminotetrazole in
Aquesous Medium, J. Am. Chem. Soc., 1954, 76, 923-926
[13] THOMAS M. KLAPTKE, KONSTANTIN KARAGHIOSOFF, PETER MAYER,
ALEXANDER PENGER AND JAN M. WELCH: Synthesis and Characterization of 1,4Dimethyl-5-Aminotetrazolium 5-Nitrotetrazolate, Propellants, Explosvies, Pyrotechnics, 2005,
in press
212
11
Czech Republic
2006
[Content]
[14] HOW-GHEE ANG, WOLFGANG FRAENK, KONSTANTIN KARAGHIOSOFF, THOMAS

M. KLAPTKE, PETER MAYER, HEINRICH NTH, JOANNA SPROTT AND MARCUS
WARCHHOLD: Synthesis, Characterization and Crystal Structures of Cu, Ag and Pd
Dinitramide salts, Z. Anorg. Allg. Chem., 2002, 628, 2894-2900
[15]
JOHANNES THIELE UND J. T. MARAIS: Tetrazolderivate aus Diazotetrazotsure, Justus
Liebigs Ann. Chem., 1893, 273, 144-160
[16]
Reichel & Partner GmbH, http://www.reichel-partner.de/
[17]
Test methods according to the UN Recommendations on the Transport of Dangerous Goods,
Manual of Tests and Criteria, 4th rev. ed.; United Nations Publication: New York, 2003
Acknowledgments:
Financial support of this work by the University of Munich (LMU),
the Fonds der Chemischen Industrie and the European Research Office (ERO)
of
the
U.S.
Army
Research
Laboratory
(ARL)
under
contract
no.
N 62558-05-C-0027 is gratefully acknowledged.
12
213
Czech Republic
2006
[Content]
APPLICABILITY OF DYNAMIC MECHANICAL AND THERMAL

METHODS IN INVESTIGATION OF AGEING PROCESSES OF DOUBLE
BASED ROCKET PROPELLANTS
Sanja Matei Muani*, Muhamed Sueska*, Sanko Bakija **
*Brodarski Institute, Av. V. Holjevca 20, 10 020 Zagreb, Croatia
**Ministry of Defence, Bauerova 33, 10000 Zagreb, Croatia
Abstract:
Ageing of double based rocket propellants (DBR), as a consequence of chemical
decomposition reaction, as well as physical processes (migration of low molecular
constituents: plasticizers, etc.) has significant influence on their relevant properties (e.g.
decrease of molar mass of nitrocellulose, decrease of stabiliser content, decrease of
specimen mass, change of mechanical and thermal properties, etc.).
The change of relevant properties of DBR propellants limits their safe and reliable service
life. Even more, under certain conditions decomposition of a DBR propellant may become
autocatalytic, which can lead to self-ignition of the DBR propellant. Because of that, it is
very important to find out reliable methods for determination of propellant stability at a
given moment of time, as well as to predict self-ignition probability under given conditions.
In this work we studied dynamic mechanical and thermal properties of DBR propellants
artificially aged for different period of time at 90C, in order to detect and quantify changes
in dynamic mechanical and thermal properties that can be used in the propellants stability
assessment. Dynamic mechanical properties were studied by dynamic mechanical analysis
(DMA), while thermal properties were studied using differential scanning calorimetry
(DSC) and thermogravimetry (TGA).
The obtained results showed that the ageing caused significant changes of DMA, DSC and
TGA curves shape. The changes are quantified by following of some characteristic points
on DMA, DSC and TGA curves. It was found out that the most sensitive
parameters/characteristic points to the ageing process at DMA curves are: storage modulus
at 25 C, storage modulus at softening region, peak width and height on loss modulus curve,
glass transition temperature, loss modulus at viscoelastic region, tan maximum at
viscoelastic region, while the most sensitive parameters at DSC and TGA curves are: peak
height and width, heat of reaction, max. heat flow rate, mass loss at a given time, and rate
of mass loss at a given time.
Keywords:
ageing, DBR propellant, dynamic mechanical analysis, differential
scanning calorimerty, mechanical properties, thermal properties,
termogravimetry
1.
INTRODUCTION
Double based rocket propellants, as well as other materials and living organisms, are subject
to the ageing processes. There are a number of parameters that can influence degradation of
214
Applicability of Dynamic Mechanical and Thermal Methods in Investigation of . . .
Czech Republic
2006
[Content]
relevant material properties, such as temperature, time, moisture, light, atmospheric conditions,
stresses produced during production and use of materials, etc.
Nitrocellulose and other nitrate ester (nitro-glycerine, etc.), which are the main ingredients
of the DBR propellants, because of relatively low bound energies (155-163 kJ/mol), and a low
value of activation energy (120-190 kJ/mol) are subject to a slow chemical decomposition even
at the room temperatures [2, 4-9].
The thermal decomposition of nitrocellulose and nitro-glycerine starts with the homolytic
breakdown of the O-NO2 bond of the aliphatic nitrate esters, thus forming nitrogen dioxide and
corresponding alkoxyl radical. The released NO2 radical immediately undergoes consecutive
reactions with other decomposition products or with other propellant ingredients. During this
process NO2 is reduced to NO, N2O, N, HNO2 and HNO3.
Another main decomposition pathway is the neutral to acid hydrolysis of the nitrate esters.
The reaction is catalysed by moisture and residual acids, or by water and acids formed during
decomposition process [10].
The summary reaction of the thermal decomposition is autocatalytic, and accompanied by
the heat generation. The heat released, due to very low propellant conductivity, can accumulate
in the propellant grain, and under certain conditions can lead to the propellant self-ignition [3,
11].
Apart from chemical ageing, DBR propellants are subjected to ageing due to physical
processes such as migration of low molecular constituents (e.g. nitro-glycerine), or crack
formation/propagation that can be initiated by residual stress at rocket grain [4,12] Ageing due
to chemical reactions, as well as due to physical processes, changes their relevant properties
(e.g. nitrocellulose mean molar mass, stabiliser content, specimen mass, mechanical and thermal
properties, ballistic properties, etc.). During use, these changes can result in dangerous failures,
such as explosion of rocket motors. Therefore, it is necessary to find out reliable methods for
determination of propellant stability at a given moment of time, as well as to predict selfignition probability under given conditions.
By now, a number of methods have been used for determination of propellant stability at a
given moment of time. In order to predict propellant stability during some period of time
(service life time), it is necessary to find out the processes which have greatest influence on the
ageing of the investigated propellant, to determine their rates and to quantify them as accurately
as possible [1].
Some quantitative methods based on the measurements of propellants relevant parameters
(such as stabiliser content, decrease of mean molar mass of nitrocellulose, specimen mass loss,
gas formation, heat generation, etc.) can be used to predict the safe service life time [10, 13-15].
In this work we studied dynamic mechanical and thermal behaviour of artificially aged
DBR propellants by dynamic mechanical analysis (DMA), differential scanning calorimetry
(DSC) and thermogravimetry (TGA) in order to check applicability of these methods in
propellants stability studies. The results of our previous investigation of these topics are reported
in papers [1-2, 16-18].
Sanja Muani, Muhamed Sueska, Sanko Baka
215
2.
Czech Republic
2006
[Content]
EXPERIMENTAL
The study was performed on the double based rocket propellant containing ~54% of
nitrocellulose, ~35% of nitro-glycerine, ~3% of dietilftalate, ~3% centralite, and ~5% of other
additives. In order to quantify changes of mechanical and thermal properties of the propellant,
caused by the ageing processes, the samples of artificially aged propellant were tested by DMA,
DSC, and TGA analysis.
The propellant was cut up into rectangular shape of the following dimensions: 50 x 10 x 2.5
mm, and then subjected to artificial ageing in glass tubes at 90 C (single-temperature aging).
The samples for DMA, DSC and TGA measurements were taken periodically (table 2-4, 6-7),
while the total ageing time was 82 days.
The dynamic mechanical measurements (DMA) were carried out using TA Instruments
DMA, Model 983. The DMA measurements were carried out using samples of the rectangular
bar shape (50 x 10 x 2.5 mm), while the measuring conditions were:
heating rate:
frequency of an oscillatory load:
amplitude of deformation:
length to thickness ratio (L/T):
temperature range:
2 C/min.
1 Hz
0.2 mm
10
-120C to +80C
The differential scanning calorimetry (DSC) measurements were carried out using TA
Instruments DSC, Model 2910. The DSC measurements were carried out using samples
weighing 1.0 0.2 mg. The samples were tested in aluminium sample pans covered by
perforated cap, with a heating rate of 5 C/min, and under nitrogen atmosphere with flow rate of
50 ml/min.
Isothermal thermogravimetry (TGA) measurements were carried out using TA Instruments
SDT, Model 2960. The TGA measurements were carried out using samples weighing 1.0 0.2
mg. The samples were tested in aluminium sample pans, at 100 C, and under nitrogen
atmosphere with a flow rate of 50 ml/min.
The degree of change of some parameter was calculated using the following equation:
yP =
( P0 Pt )
100
P0
(1)
where: yP degree of change of some parameter, P0 parameter value at the beginning of

the ageing, and Pt parameter value after some time of ageing
The ageing times at 90 C (t90C) were, with assumption that ageing mechanism at 90 and
30C are the same, transformed into corresponding ageing times at 30 C (t30C), in accordance
with the following equation [1]:
t 30 0 C = t 90 0C a10
[(90 30 ) / 10 ]
(2)
where reaction rate accelerating factor (10) is taken to be equal 3.
216
3.
Czech Republic
2006
[Content]
Samples of non aged and aged double based rocket propellants were subjected to DMA,
non-isothermal DSC and isothermal TGA tests. For this preliminary investigation, because of
the limitation of the experiment duration time, the selected ageing temperature was 90 C,
which is much higher than the normal storage temperature.
3.1.
Dynamic mechanical properties
Change of storage modulus with ageing

Storage modulus vs. temperature curves of a non aged DBR propellant and the DBR
propellants aged for different period of time at 90 C are given at Fig. 1.
10000

8000
Flex Storage Modulus (MPa)

6000

4000

0
-120
-100
-80
-60
-40

2000
Non - aged
Aged at 90C for 2 days
-20

20
Temperature (C)

40
60
80
100
Universal V3.8B TA Instruments
Fig. 1. Storage modulus curves of a non-aged propellant and the propellants aged for different
period of time at 90 C as a function of temperature
From Fig. 1. is evident that E-T curve was shifted with the ageing time to higher
temperatures, as well as to higher values of modulus. An increase of the storage modulus is
indicator of reduced flexibility of nitrocellulose macromolecules. Reduced flexibility of the
macromolecular chain, in this case is a result of reducing amount of the energetic plasticizer
(nitro-glycerine). The plasticizer migration from the propellant grain interior to the surface, and
their vaporization, causes shortening of distances between macromolecules, and an increase of
intermolecular forces. These processes have significant effect on flexibility of the nitrocellulose
macromolecule units.
In order to quantify changes in the storage modulus, caused by artificial ageing of the DBR
propellant, several characteristic points on E-T curves are analysed, Fig. 2, Tab. 1, 2.
217
Czech Republic
2006
[Content]
Fig. 2. Typical E-T curve of DBR propellant, along with derivative ET curve
Table 1. Some characteristic points/parameters on DMA curves
No.
Characteristic points on E-T curves

o
Denotation
Unit
E (-115C) MPa
MPa
E(25C)
1.
2.
3.
4.
5.
6.
storage modulus at -115 C,

storage modulus at 25 oC
extrapo. onset temp. at the beginning of transition from glassy to viscoela. stage
extrapo. onset temp. at the end of transition from glassy to viscous stage
storage modulus at max. decreasing rate of storage modulus in softening region
temperature at maximum decreasing rate of storage modulus in softening region
T(EON1)
T(EON2)
E1
T(E1)
C
C
MPa
C
No.
Characteristic points on E-T curves
Denotation
Unit
loss modulus at -115 C

loss modulus at 25 oC
peak width at half height
peak height
glass transition temperature (peak maximum temperature)
loss modulus at viscoelastic region (close to softening point)
extrapo. onset temp. at begining of transition from viscoelastic to viscous stage
extrapo. onset temp. at the end of transition from viscoelastic to viscous stage
E(-115C)
E(25C)
Wpeak
hpeak
Tg
E1
T(EON1)
T(EON2)
MPa
MPa
C
MPa
C
MPa
C
C
No.
Characteristic points on tan -T curves
Denotation
Unit
1.
2.
3.
4.
5.
6.
7.
tan at -115 oC
extrapolated onset temperature at begin of glass transition
tan maximum in viscoelastic region
extrapolated onset temperature at end of glass transition
tan at -115 oC
extrapo. onset temp. at begining of transition from viscoelastic to viscous stage
extrapo. onset temp. at the end of transition from viscoelastic to viscous stage
tan (-115 C)
1.
2.
3.
4.
5.
6.
7.
8.
218
T(tan ON1)
tanmax
T(tanON2)
tan (25C)
T(tanON3)
T(tanON4)
C
C
C
C
Czech Republic
2006
[Content]
In Tabs. 2-4 and 6-7, the values of the characteristic points at the DMA, DSC and TGA
curves according to time of ageing at 90 C are shown. Also, it is specified degree of change,
and measurement uncertainty for every characteristic point.
Table 2. Values of some characteristic points on E-T curves of DRB propellants aged for
different period of time at 90C
Ageing time
at 90 C,
days
Corresponding
ageing time at
30C, years**
E (-115C)
MPa
T(EON1)
C
T(EON2)
C
E(25C)
MPa
E1
MPa
T(E1)
C
Non aged
2
4
10
25
45
67
82
Non aged
4
8
20
50
90
134
164
8248
8063
7997
8961
8043
7655
7327
7540
-67.18
-69.48
-67.99
-68.13
-67,65
-71.27
-77.07
-91.37
-16.43
-15.67
-16.46
-16.40
-14.84
-6.60
-
1213
1097
1048
1194
1181
1580
2260
3044
572.4
582.6
493.5
513.5
621.1
790.1
1248.0
1438.0
42.42
39.90
42.50
43.70
41.80
43.70
47.20
60.70
8.59
36.00
105.48
132.28
151.22
34.89
8249 369
-67 1.51
-16.43 0.60
1218 97.8
572,4 26.8
42.42 1.79
Degree of change (after 82 days

of ageing at 90C) / % *
Mean value experimental
standard deviation
E at maximum decreasing rate of storage

modulus in softening region, E1, MPa
7000
Extrapolated onset temperature at the end of

Storage modulus at 25C
the transition from glassy to viscous stage,
(E(25
C)),
MPa
(T(E
ON2)),C
*- Calculated according to Eq. 1, **- Calculated according to Eq. 2
3500
6000
3000
peak ve
5000
4000
2500
2000
3000
1500
2000
1000
1000
500
E1
E(25C)
0
0
20
40
60
80
0
100
0
-2
-4
-6
-8
-10
-12
-14
-16
-18
-20
0
Ageing time at 90C, days
20
40
60
80
100
Ageing tine at 90C, days
Fig. 3. Changes of some characteristic points on E-T curves of DBR propellants aged for
different period of time at 90 C
From Tab. 2 and Fig. 3 it is evident that some of characteristic points at E-T curve
(E(25C), T(EON2, E1) are change significantly with the time of ageing. For example, degree
of change of storage modulus at the maximum decreasing rate in the softening region, E1, after
82 days of ageing at 90 C is more than 150%.
The changes of storage modulus at the maximum decreasing rate in the softening region,
E1, becomes higher than measurement uncertainty limits (Tab. 2) after 10 days of ageing at 90
C (i.e. after ~20 years of ageing at 30 oC), while the changes of other parameters become
219
Czech Republic
2006
[Content]
higher than measuring uncertainty limits after 25 days of ageing at 90oC (i.e. after ~50 years of
ageing at 30 oC). This means that, on the basis of the E-T curve measurements, a reliable
conclusion on DBR propellant stability, i.e. its age, may be set out only after this period of
ageing.
The accelerated change of all characteristic points at the E-T curves begins after 45 days of
ageing at 90 C. This is related to the intensive degradation of nitrocellulose chain.
Change of loss modulus with ageing

From the E-T curves of DBR propellants aged for different period of time at 90C (Fig. 4)
is evident that ageing causes significant quantitative and qualitative changes on the E-T curves.
For example, peak height at the glass transition region decrease (from ~ 420 MPa for non-aged
propellant to ~180 MPa for propellant aged 82 days at 90 C), peak width increases, and the
complete E-T curve is shifted to higher temperatures.
Very intensive changes on the E-T curve begin after 45 days of ageing at 90 C. After that
time of ageing, begins intensive decrease of peak height, and very intensive increase of the peak
width, that after 67 days of ageing even resulted with the splits of the main peak into two
separate peaks.
500
400
Flex Loss Modulus (MPa)

300
200
100

0
-120
Non - aged

-100
-80
-60
-40
-20
Temperature (C)
20
40
60
80
100
Fig. 4. Loss modulus curves of non-aged propellant and the propellants aged for different period
of time at 90 C as a function of temperature
From Fig. 4 and Tab. 3 is visible that the temperature of the peak maximum at the glass
transition region, Tg, shifts to the lower temperatures, while at the same time, the temperature of
the peak maximum at the viscoelastic region increases.
Increase of the peak width and presence of two peaks at the E-T curves, after 67 days of
ageing at 90 C, indicates an increase in the heterogeneity of system. This is probably a
consequence of the very intensive decomposition reaction of low - molecular components of the
DBR propellant (nitro-glycerine, stabiliser, etc.), as well as the splits of the nitrocellulose
macromolecules.
220
Czech Republic
2006
[Content]
In order to quantify these changes in the loss modulus, caused by artificial ageing of DBR
propellant at 90 C, several characteristic points on the E-T curves were analysed, Fig. 5, Tab.
1, 3.
Fig. 5.Typical E-T curve of DBR propellant, along with derivative E-T curve
Table 3. Values of some characteristic points on E-T curves of DRB propellant aged for
Ageing time
at 90 C,
days
Corresponding
ageing time at
30C, years**
E(-115C)
MPa
Wpeak
C
Tg
C
hpeak
MPa
E(25C)
MPa
E1
MPa
T(EON1)
C
T(EON2)
C
Non aged
2
4
10
25
45
Non aged
4
8
20
50
90
64.0
63.4
63.7
70.6
78.3
76.5
59.28
57.45
57.16
57.84
61.09
78.52
111.7
112.1
112.1
120.2
127.5
155.2
44.76
43.68
44.56
45.08
44.91
46.51
62.90
63.49
62.68
63.61
64.84
68.83
134
102.8
140.68
180.9
180.7
54.26
78.92
82
164
126.2
166.8
421.68
401.30
399.40
431.00
378.20
329.10
229.0
161.7*
179.9
115.2*
123.4
126.1
130.2
136.0
147.1
162.7
67
-41.98
-43.12
-41.62
-42.00
-39.90
-41.79
-45.59
-11.50*
-51.89
4.70*
223.0
213.7
69.53
91.54
Degree of change (after 82 days

of ageing at 90C) / % *
97.13
181.38
30.59
72.68
58.39
91.35
45.64
36.02

standard deviation
64.02
6.7
59.28
1.12
-41.98
0.72
421.68
18.47
123.4
6,7
111.7
5.1
44.76
62.90
1.80
2.05
It is evident from Tab. 3 and Fig. 6 that characteristic values at the E-T curves (loss
modulus at -115 oC, peak width at half height, peak height, loss modulus at viscoelastic
region, etc.) are sensitive to ageing processes. The changes of majority of parameters become
higher than measuring uncertainty limits after 10 days of ageing at 90 oC (i.e. after ~20 years of
ageing at 30 oC).
221
450
400
350
peak ve
-10
-20
-30
-40
300
250
200
150
100
-50
50
-60
20
40
350
300
250
200
150
100
hpeak
160
140
120
100
80
60
40
peak ve
20
50
Wpeak
0
100
20
100
40
60
80
Aging time at 90C, days
Ageing time at 90C, days
[Content]
180
80 0
60
400
2006
Peak width (W peak ), C
500
450
Loss modulus at viscoelastic region

(E1), MPa
500
Loss modulus at -115C, MPa
Tg
E1
E (-115C)
10
Peak height (hpeak ), MPa
Temperature of glass transition

(Tg), C
20
Czech Republic
Fig. 6. Changes of some characteristic points on E-T curves of DBR propellant aged different
period of time at 90C
From Tab. 3 and Fig. 6 can be seen that a significant change of characteristic parameters
begins after 45 days of ageing at 90 C. The change of the DBR propellants properties after 45
days of ageing was especially influenced at characteristic parameters related with softening
point (T(EON1), T(EON2)), which before that time of ageing were almost constant. Intensive
increase of softening point after 45 days of ageing is probably the result of increased
intermolecular forces, as a consequence of decreased amount of plasticizer (nitro-glycerine) and
increased heterogenity of the system (nitrocellulose chain breakdown).
Change of tan with ageing

Tan vs. temperature curves of a non-aged the DBR propellant and DBR propellants aged
for different period of time at 90 C are given in Fig. 7.
0.6

Non - aged
0.4

Tan Delta

0.2

0.0
-120

-100
-80

-60
-40
-20
Temperature (C)
20

40
60
80
100
Fig. 7. Tan -T curves of non-aged DBR propellant aged for different period of time at 90 C
From the tanT curves of DBR propellants aged for different period of time at 90C,
shown in Fig. 7, is obvious that the shape of tan -T curves changes considerably with the
222
Czech Republic
2006
[Content]
ageing. At temperatures below the glass transition region, the value of tan remains almost
unchanged, while in the glass transition region, viscoelastic region, and in the softening
temperature region the value of tan decreases significantly. From Fig. 7, it can be seen that
with the ageing time the tan- T curve shifts to higher temperatures. All these changes,
especially increased softening point, are the indication that by the ageing the amount of
plasticizer in the propellant decreases.
In order to quantify these changes in the tan, several characteristic points on tan-T curves
were analysed, Fig. 8, Tab. 1, 4.
Fig. 8. Typical tan -T curve of DBR propellant, along with derivative tan-T curve
Table 4. Values of some characteristic points on tan -T curves of DRB propellant aged for
Ageing
time at 90
C, days
Corresponding
ageing time at
30C, years**
Non aged
2
4
10
25
45
67
82
Non aged
4
8
20
50
90
134
164
Degree of change (after 82

days of ageing at 90C) / % *
standard deviation
tan
T(tan
T(tan
(-115C)
ON1)
ON2)
MPa
0.0078
0.0079
0.0080
0.0079
0.0097
0.0103
0.0139
0.0167
C
-76.90
-75.09
-76.37
-76.38
-75.34
-78.43
-94.46
-98.21
C
-36.38
-36.87
-36.89
-36.12
-35.29
-35.10
-45.47
-56.67
113.74
27.72
104.7
0.0078
-76.90
-36.38
0.001
3.04
1.09
tan max
MPa
0.1074
0.1059
0.1074
0.1035
0.1024
0.0909
0.0604
0.0481
tan
(25 C)
MPa
0.1164
0.1150
0.1242
0.1139
0.1246
0.1030
0.0800
0.0733
T(tan ON3)
C
35.50
32.99
34.22
35.28
35.45
36.61
43.18
55.15
T
(tan ON4)
C
57.32
55.22
54.71
55.29
56.41
59.19
70.16
89.16
55.21
24.71
43.25
59.19
0.1074
0.0025
0.1164
0.0145
35.50
1.98
57.32
2.52
Although, changes of the characteristic parameters at the tanT curves are evident
(Figs. 8 and 9, Tab. 4), for the majority of the parameters, they still remain within measurement
10
223
Czech Republic
2006
[Content]
uncertainty up to 45 days of ageing at 90 C (i.e. ~ 90 years of ageing at 30 C). Exceptions is

aximum in the viscoelastic region (tanmax), which seems to be very sensitive to
the tan m
ageing. The values of the tan maximum in viscoelastic region, reaches the measurement
uncertainty limits after 10 days of ageing at 90 C (i.e. after ~20 years of ageing at 30 oC), so
that parameters should be considered to one of the parameters that can be used for prediction of
the DBR propellant life time.
0,08
0,06
0,04
0,02
0
0
20
0,019
0,017
-10
0,015
peak ve
0,013
at -115C
0,1
10
-30
0,011
Tan
Extrapolated onset temperature at end of

glass transition T(tan ON2), C
Tan maximum in viscoelastic

region, tan max
0,12
-40
0,009
-20
peak ve
-50
tand max
0,007
tand (-115C)
-60
0,005
400
6020
80 40 100 60
Ageing time at 90C, daysAgeing time at 90C, days
80
100
Fig. 9. Changes of some characteristic points on tan-T curves of DBR propellant aged for
3.2.
Thermal properties
Change of non-isothermal DSC curves with ageing

Non-isothermal DSC curves of DBR propellants aged for different period of time at 90 C
are given in Fig. 10.
6
Heat Flow (W/g)
Non-aged

-1
100
Exo Up

125
150
175
200
Temperature (C)
225
250
275
300
Universal V3.8B TA Instrum ents
Fig. 10. Non-isothermal DSC curves of DBR propellants aged for different period of time at 90
C
224
Czech Republic
2006
[Content]
Fig. 11. Typical non-isothermal DSC curve of DBR propellant, along with derivative heat flow
curve and heat flow curve running integral (heating rate 5 oC/min)
Table 5. Some characteristic points/parameters on DSC and TGA curves
12
No.
Characteristic points on DSC curves
Denotation
Unit
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
extrapolated peak onset temperature

peak maximum temperature
extrapolated peak end-set temperature
heat of reaction
peak height
peak width at half-height
conversion at maximum heat flow rate
conversion at peak maximum
maximum heat flow rate
maximum actual heating rate
Tie
Tp
Tfe
Hr
hp
wp
(d/dt)max
(Tp)
(d/dt)max
max
C/min
No.
Characteristic points on TGA curves
Denotation
Unit
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
maximum decomposition rate

time to reach maximum decomposition rate
degree of decomposition at maximal decomposition rate
time to reach 10% sample mass loss
mass loss after 60 min
mass loss after 200 min
rate of decomposition after 10 min
rate of decomposition after 20 min
(dm/dt)max
t [(dm/dt)max]
m [(dm/dt)max]
t (10 %)
t (15 %)
t (20 %)
m (60 min)
m (200 min)
(dm/dt)10 min
(dm/dt)20 min
%/min
J/g
W/g
o
%
%
W/(g min)
min
%
min
min
min
%
%
%/min
%/min
225
Czech Republic
2006
[Content]
From Fig. 10, is clear that ageing has caused significant changes of the DSC curves shape
and position. For example, DSC peak height increases, peak width decreases, peak onset
temperature increases, heat of reaction increases, maximum heat flow rate increases, etc. In
order to quantify these changes in the DSC curves, several characteristic points on DSC curves
were analysed, Fig. 11, Tab. 5, 6.
Table 6. Values of some characteristic points on DSC curves for DBR propellants aged for
different period of time at 90oC
Ageing
time at
90 oC,
days
Non aged
2
4
10
45
60
67
72
82
Tfe,
o
C
Hr,
J/g
193.71
196.11
196.80
194.45
195.97
195.34
191.43
195.46
195.51
224.76
224.74
225.08
222.97
221.45
224.83
222.57
225.24
225.21
4.25
0.93
0.20
172.16
3.07
194.94
2.00
220.71
7.38
Corresponding
ageing time
at 30 oC, year**
Tie,
o
C
Tp,
o
C
Non aged
4
8
20
90
119
134
144
164
168.38
168.42
167.98
169.76
172.71
171.21
173.81
175.15
175.55

days of ageing at 90C) / %
*
standard deviation
hp,
W/g
wp,
o
C
1731
1795
1873
1910
1946
1901
1980
2178
2093
3.82
4.09
4.33
4.57
4.90
4.88
5.46
5.32
5.15
31.91
30.51
30.28
27.53
28.61
28.46
25.46
28.16
27.88
20.91
34.82
12.63
1810
252
4.62
1.13
27.54
5.16
(d/dt)max,
%
(d/dt)max,
(Tp),
%
W/(g min)
26.57
26.59
26.39
27.27
20.44
28.25
28.50
29.82
31.61
46.25
46.08
46.55
45.92
45.33
45.67
46.54
42.96
44.67
0.888
0.884
0.922
1.087
1.536
1.296
1.511
1.489
1.814
18.97
3.42
104.28
24.84
5.04
45.86
3.97
1.2
0.38
*- Calculated according to Eq. 1, **- Calculated according to Eq. 2,
From Tab. 6 and Figs. 10, 12, it is visible that the changes are more intensive during the first
15 days of ageing at 90 oC (i.e. ~30 years of ageing at 30 oC), almost constant between 15 and
45 days of ageing, and then intensive again after that period. According to our findings intensive
changes in the first 15 days of ageing at 90 oC are primarily due to nitro-glycerine evaporation
and decomposition, while changes after 45 days of ageing are connected with intensive
degradation of nitrocellulose (chain breakdown) [2].
2300
2.000
6.00
1.800
wp
hp
5.00
28.00
4.50
26.00
4.00
24.00
3.50
(d /dt )max ,%min
30.00
2200
1.600
5.50
Peak height, W/g
Peak width, C
32.00
1.400
2100
1.200
2000
1.000
1900
0.800
0.600
0.400
(df/dt)max
0.200
Hr
0.000
22.00
0
20
40
60
Ageing time at 90 o C, days
80
3.00
100
20
40
60
80
1800
Reaction heat, J/g
34.00
1700
1600
100
Ageing time at 90 C, days
Fig. 12. Changes of some characteristic points/parameters on DSC curve of DBR propellant
with the ageing time at 90 oC
226
Czech Republic
2006
[Content]
Although changes of the DSC curves shape with the ageing, and consequently characteristic
points on the DSC curves, are evident they still remain within the measuring uncertainty limits
( ) up to 67 days of ageing at 90 oC (i.e. ~135 years of ageing at 30 oC). This means that, on
the basis of non-isothermal DSC tests, a reliable conclusion on DBR propellant stability, i.e. its
age, may be set out only after this period of ageing.
Change of isothermal TGA curves with ageing

Isothermal TGA curves of DBR propellants aged for different period of time at 90 oC are
given in Fig. 13.

100

Weight (%)
95
90
85
80

Non-aged




75
0
100
200
Time (min)
300
400
Fig. 13. Isothermal TGA curves of DBR propellants aged for different period of time at 90C
In order to quantify changes in the TGA curves caused by artificial ageing of the DBR
propellant at 90 C, several characteristic points on TGA curves were analysed (Fig. 14, Tabs. 5,
7).
Fig. 14. Typical isothermal TGA and derivative TGA curves of NC propellant
14
227
Czech Republic
2006
[Content]
Table 7. Values of some characteristic points on TGA curves for DBR propellants aged for
different period of time at 90 oC
Ageing
time at
90 oC
Nonaged
2
10
25
45
72
82
Corresponding
ageing time
at 30 oC**
(dm/dt)max, t((dm/dt)max), m(dm/dt)max,

%/min
min
%
0.0
4.0
20.0
49.9
89.9
143.8
163.8

days of ageing at 90C) / %
*
Mean value
experimental standard
deviation
t(10 %),
min
t(15 %),
min
m(60 min), m(200 min),

%
%
(dm/dt)10 min,
%/min
(dm/dt)20 min,
%/min
0.997
1.019
0.986
0.701
0.644
0.497
0.270
2.47
2.47
2.47
2.39
1.84
1.81
2.55
1.55
1.58
1.73
1.03
0.84
0.72
0.60
23.47 66.81 14.51

33.63 118.00 12.37
27.15 82.60 13.66
132.40
7.57
399.20
5.78
4.80
1.77
34.13
16.90
18.22
13.48
8.41
7.18
2.49
0.4312
0.3396
0.3679
0.2228
0.1344
0.1078
0.0301
0.24
0.19
0.21
0.16
0.08
0.08
0.16
72.92
3.24
61.29
1600.89
23.63
87.80
92.70
93.02
33.33
1.248
0.62
2.471
0.42
2.052
1.12
13.865 30.117
9.11
33.42
19.921
5.81
23.636
5.18
0.663
0.27
0.358
0.11
It is evident from Fig. 13 and Tab. 7 that the ageing causes significant changes of TGA
curves. For example, time to reach the same degree of mass loss increases with the ageing, mass
loss after the same time period decreases, the rate of mass loss decreases (Fig. 14), etc. The
changes of majority of parameters become higher than measuring uncertainty limits (Tab. 7)
after 10 days of ageing at 90 oC (i.e. after ~20 years of ageing at 30 oC).
TGA experiments were carried out using opened aluminium sample pans and under 100 oC
isothermal temperature. Under such conditions the sample mass loss is predominantly a
consequence of nitro-glycerin evaporation, and partly degradation of nitro-glycerine and
nitrocellulose. Our previous experiments have shown that mass loss of NC propellant at 100 oC
is less than 2 % after 400 minutes [1], which is almost negligible comparing with mass loss due
to nitro-glycerine evaporation and degradation.
40.00
t(10 %)
400.00
(dm/dt)max
1.000
1.80
m((dm/dt)max
1.60
300.00
25.00
250.00
20.00
200.00
15.00
150.00
10.00
100.00
(dm/dt )max , %/min
30.00
m (200 min), %
t (10 %), min
350.00
2.00
1.200
35.00
m(200 min)
1.20
0.600
0.80
50.00
5.00
0.00
0.00
100
0.000
20
40
60
o
Ageing time at 90 C, day
80
1.00
0.400
0.200
1.40
0.800
0.60
m (dm/dt )max , %
450.00
0.40
0.20
0
20
40
60
80
0.00
100
Ageing time at 90 C, day
Fig. 15. Changes of some characteristic points/parameters on TGA curve of DBR propellant
with the ageing time at 90 oC
228
Czech Republic
2006
[Content]
The results of isothermal TGA measurements (Fig. 15) clearly explain the conclusion that
the dominant process during artificial ageing at 90 oC is nitro-glycerine evaporation and
degradation (simultaneously). Because of that, the TGA curve of DBR propellant aged for 82
days at 90 oC becomes very close to TGA curve of NC propellant.
Taking into account all DSC and TGA experiments carried out on the DBR propellant, it is
obvious that the ageing causes changes of their thermal properties. However, the changes of
DSC parameters still remains within the measuring uncertainty limits during 67 days of ageing
at 90 oC (i.e. ~135 years of ageing at 30 oC), while the changes of TGA parameters remain
within the measuring uncertainty limits during the first 10 days of ageing at 90 oC (i.e. 20 years
of ageing at 30 oC). This means that TGA is more sensitive than DSC to the ageing of DBR
propellants.
4.
CONCLUSION
The results presented in this paper have shown that artificial single-temperature ageing (at
90 C) of DBR propellants causes significant changes of their mechanical and thermal
properties. Obtained results have confirmed that dynamic mechanical analysis, differential
scanning calorimetry and termogravimetry can be used to follow and to quantify these changes.
Obtained changes on DMA, DSC and TGA curves indicate that ageing processes have two
stages. In the primary stage of ageing dominant processes are migration and evaporation of
nitro-glycerine, while in the secondary stage of ageing a dominant process is intensive
degradation of nitrocellulose polymer chain.
It was found out that the most sensitive parameters/characteristic points to the ageing
process at DMA curves are: storage modulus at 25 C, storage modulus at the softening region,
peak width and height on loss modulus curve, glass transition temperature, loss modulus at the
viscoelastic region, tan maximum in viscoelastic region, extrapolated onset temperature at the
end of the glass transition, while the most sensitive parameters at the DSC and the TGA curves
are: peak height and width, heat of reaction, maximum heat flow rate, mass loss at a given time,
and rate of mass loss at a given time.
However, because of the relatively high measuring uncertainty and limited sensitivity
(especially for DSC measurements) to the age of the propellants, it follows that DMA, DSC and
TGA are not too sensitive techniques to the propellants age, but they still may help in the
propellants stability evaluation. In the future work we will try to found out optimal testing
conditions which will give minimal measurement uncertainty.
5.
REFERENCES
[1]
M. SUESKA, S. MATEI MUANI, M. RAJI LINARI, Applicability of Thermal Methods in

Stability Studies of Homogeneous Propellants, IASPEP Conference, Beijing, China, 2005.
S. MATEI MUANI, Thermal decomposition of double base rocket propellant and its
influence on viscoelestic properties, Masters thesis, University of Zagreb, Faculty of Chemical
Engineering, Zagreb, 2005. (in Croatian)
M. SUESKA, Baruti, MORH, 1996.
TAGER, Physical Chemistry of Polymers, Mir Publishers, Moscow, (1972)
M. A. BOHN, Kinetic Modelling of the Ageing of Gun and Rocket Propellants for the Improved
And Time-Extended prediction of Their Service Lifetime, Proceed., 1998 Life Time Cycles of
[2]
[3]
[4]
[5]
16
229
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
230
Czech Republic
2006
[Content]
Energetic Materials, Fullerton, California, USA, 29 March-1 April, (1998)

M.A. BOHN, Methods and Kinetic Models for the Lifetime Assessment of Solid Propellants,
Paper 2 on the 87th Symp. Of the Propulsion and Energetic Panel (PEP) of the AGARD, Service
Life of Solid Propellant Systems , Athens, Greece, May 6-10, (1996), AGARD-ConferenceProceedings 586 (AGARD-CP-586), F-92200 Neuilly sur Seine, France, May (1997).
F. VOLK, G. WUNSCH, Determination of the Decomposition Behaviour of Double Base
Propellants at Low Temperature, Propell. Expl. Pyrot. 10, 181 (1985).
A. PFEIL, H.H. KRAUSE, N. EISENREICH,, The Consequences of Beginning Slow Thermal
Decomposition on the Molecular Weight of Nitrated Cellulose, Thermochim Acta 85, 395
(1985).
F. VOLK, M. A. BOHN, G. WUNSCH, Determination of the Chemical and Mechanical Properties
of Double Base Propellants During Aging, Propell. Expl. Pyrot. 12, 81 (1987).
B. VOGELSANGER, Chemical stability, Compatibility and Self Life of Explosives, Chimia 58,
401-408, 2004
M. A. BOHN, F. VOLK, Prediction of the Lifetime of Propellants by Stabilizer Consumption and
Molar Mass Decrease, Proceed. ADPA Predictive Technology Symp., June 22-24, (1993),
Orlando, FL, US-Army ARDEC, Picatinny Arsenal, New Jersey, USA.
T. MURAYAMA, Dynamic Mechanical Analysis of Polymeric Material, Elsevier Scientific
Publishing Company, Amsterdam, Oxford, New York, (1978)
BOHN M. A, Kinetic Modelling of the Ageing of Gun and Rocket Propellants for the Improved
and Time-Extended Prediction of Their Service Lifetime, Proc of the 4th Life Cycles of Energetic
Materials Conference, March 29 April 1, Fullerton, USA, 1998, p. 1-38.
M. A. BOHN, F. VOLK, Ageing Behaviour of Propellants Investigated by Heat Generation,
Stabiliser Consumption, and Molar Mass Degradation, Propellants, Explos. Pyrotech., No. 17,
(1992), p. 171-178.
R. R. Rogers, J. L., Rogers. Explosive science, Los Alamos, NM, http://home.att.net/~mrogers,
2002.
M. SUESKA, S. MATEI MUANI, M. RAJI LINARI, Determination of Arrhenius kinetic
constants for DB propellants by non-isothermal DSC measurements. Influence of some testing
conditions, Proc. of 6th Int. Seminar New trends in research of energetic materials, Pardubice,
2003, pp. 374-391.
M. SUESKA, M. RAJI, Thermal characterisation of stable and unstable nitrocellulose
propellant, Proc. of 33rd Int. Annual conference of ICT, Karlsruhe; ICT, 2002. 98.1-98.17
M. RAJI LINARI, M. SUESKA, S. MATEI MUANI, Changes of some thermophysical
properties of nitrocellulose propellants during accelerated aging, Proc. of 8th Int. Seminar
New trends in research of energetic materials, Pardubice, Czech Republic, 2005., pp.758-764.
Czech Republic
2006
[Content]
HIGH ENERGY DENSITY MATERIALS (HEDM):

OVERVIEW, THEORY AND SYNTHETIC EFFORTS AT FOI
Henric stmark
Department of Energetic Materials
FOI, Swedish Defence Research Agency
Grindsjn Research Centre, 147 25 Tumba, Sweden
Abstract:
This paper presents a literature overview of HEDM. A number of potential HEDMs with
performances theoretically up to more than 10 times HMX have been identified, ranging
from ionic all-nitrogen compounds like N5- and N5+ with suitable counter ions (N5+N5gives a performance of ~1.6 times HMX) to polymeric nitrogen (10 times HMX) and
metallic hydrogen. In between there are for example metastable nitrogen clusters (Nx,
x4 to 60). Calculations show that N4 has approximately 3.2 times the performance of
HMX. Both metallic hydrogen and polymeric nitrogen are high-pressure compounds
(several Mbar) and thus probably hard to synthesize in larger amounts. The nitrogen
clusters offer an interesting alternative or complement in that they are all normal
chemical compounds and they all have high heat of explosion, and probably also high
density (2.0-2.7 g/cm3).
This overview covers conventional chemistry, novel chemistry (all-nitrogen
compounds, new oxidizers in the form of NxOy compounds) and exotic physics (eg.
metallic hydrogen, cold fusion and anti-matter). The time in which these concepts can
be realised ranges from near future to many years ahead and some may possibly never
be made at all or perhaps only in lab scale. The overview is complemented with
theoretical and experimental results from our institute in our effort to synthesize new all
nitrogen compounds, e.g. Pentazolate ion (N5-) and tetraazatetrahedrane (N4(Td) .
Keywords:
1.
HEDM, synthesis, QM, performance, all nitrogen compounds
INTRODUCTION
The three most important reasons for the development of energetic materials for military
purposes today are the needs for increased performance, increased safety and certain tailored
properties. This is important for warhead, propulsion and launch applications. Several factors
influence the development directly or indirectly. The most important are increased use of
insensitive munitions (IM), the need for more precise weapons effects, higher performance,
weapon specific properties, environmental concerns, internationalisation, standardisation,
and increased system integration.
In this paper we will discuss the possibilities to develop materials with performance
considerably better than what is used today, commonly referred to as High Energy Density
Materials (HEDM). The aim of this paper, which is based on the available literature and on
work done at our institute, is to give an overview of the area, to see the possibilities of what
could be achieved with different concepts, soon or in a longer time perspective.
When dealing with new energetic materials a term that has been used in the Energetic
Material community is High Energy Density Materials (HEDM). The limit between
1
Henric stmark
231
Czech Republic
2006
[Content]
conventional energetic materials and HEDM has never been defined and may therefore differ
between research groups. In the first European Workshop on High Energy Density
Materials, held at FOI in October 2005, the participants agreed on a common definition for
HEDM as energetic materials better than the best that are in use in systems today. The
baseline will vary with the application and we end up with different reference substances for
different applications. The European definition gives the following baselines; HEDMs are
explosives with higher performance than HMX, solid rocket propellants with higher
performance than AP/Al/HTPB and gun propellants better than JA2 (NC/NG) propellants.
Computational modelling, using mainly quantum chemical methods (numerically solving
the Schrdinger equation at different level of approximation), plays an integral part in the
development of new high energy density materials. This is the case both for the synthesis,
detection and characterization of new all-nitrogen compounds, and in the design and
synthesis of conventional type explosives.
The performance of a substance is readily calculated from its heat of formation and
density; two properties that can be estimated by means of quantum chemical computations.
Apart from performance, some practical aspects such as stability, sensitivity and state of
aggregation are important. In order to make experimental work meaningful it is also essential
to calculate properties that make it possible to detect and identify the material once it has
been made. It is also very useful to use quantum mechanics to evaluate synthesis routes.
This paper is mainly based on a more comprehensive overview of HEDM research,
including chapters on QM and performance calculations High Energy Density Materials
(HEDM) A Literature Survey [1]. This report was a joint effort by many people whose
contribution is hereby acknowledged.
2.
HISTORIC REVIEW
Yesterday, today
The earliest known energetic material was black powder and Greek fire, both with origins
held in some mystery but probably invented about 2000 years ago, Greek fire in ancient
Greece and black powder in China. Black powder was first used only for fireworks and
mainly not for war. The first real use for warfare was probably in the 14th century as
propellant for guns and firearms. It was then in use until the little revolution of the midnineteenth century when the black powder era ended, and when modern-day propellants and
high explosives came into use. The development of energetic materials has since then
proceeded relatively slowly. Most of the explosives found in todays weapons systems are
still TNT, PETN, HMX or RDX, i.e. energetic materials developed in the 19th century or, in
one case, the 1940s (HMX).
Table 1 shows an overview of the history of energetic materials development. The
column for Performance gives values relative to the performance of HMX. The performance
increases about 70% from picric acid (60) to HMX (100) between World War One and
today. This can be compared to the threefold increase at the end of the black powder era.
The slow development during the greater part of the 20th century was partly due to a lack
of understanding of energetic molecular structures rather than a poorly developed synthesis
technique. For instance, scientists believed for a long time that HNB, hexanitrobenzene, was
the optimal high explosive. The relative performance of HNB vs. RDX is 1.23. After
Russian scientists first reported the successful synthesis of this compound, US researchers
232
High Energy Density Materials (HEDM): Overview, Theory and Synthetic . . .
Czech Republic
2006
[Content]
spent 10 years trying to find the synthetic path. This was finally achieved, after which it
became apparent that the substance was unstable and highly sensitive to light.
tomorrow
During the last decade, the development of computers and thus the increased capability
for quantum chemical calculations, as well as a broadened view of the characteristics of
energetic molecules and their applications, have yielded new opportunities for the
development of energetic materials. It was generally believed that useful energetic materials
would be CHNO-substances, with molecules containing carbon, hydrogen, nitrogen and
oxygen. Today, several completely different types of compounds have been added to the list
of possible energetic materials. Among these are caged molecules that contain more energy
per unit of mass and volume, molecular composites where fuel and oxidizer are mixed at a
molecular level, nano-systems of reactive materials and meta-stable compounds. Even very
extreme forms of energy storage, such as nuclear isomers and anti-matter are investigated.
In other words, the last decade has seen an explosive development in the number of
theoretical candidates for useful energetic materials. Some of these have already been
synthesised and produced in useful quantities.
Today, the development of energetic materials follows two distinct lines, one
evolutionary and one revolutionary. There is also the development of reactive materials that
combined with energetic materials can increase the performance of future weapons systems.
Table 1 Historical overview of development of energetic materials.
Invented
In Application
Energetic Material
Energy
Density
3
Performance
(MJ/kg)
(kg/m )
(relative to HMX=100)
1000?
1600-1870
Black Powder
20
1847
1870-2010
Nitroglycerine
6.5
1.6
70
1867
Not in weapons
Dynamite
1.45
50
1771
WW1
Picric Acid
1.6
60
1863
1900-2020
TNT
1.55
60
1899
WW2
RDX
6.6
1.8
90
1894
WW2
PETN
6.3
1.8
80
1940s
1960-
HMX
6.5
1.9
100
1986
2005
CL-20
7.5
2.1
120
2006
3.
HEDM POSSIBLE BY THE USE OF CONVENTIONAL

CHEMISTRY
The evolutionary development of new energetic materials concerns CHNO-substances

prepared by conventional chemistry. The two main driving forces for the development of
explosives are lower sensitivity and increased performance. Until now, new energetic
materials have either been less sensitive with retained performance compared to current
materials or more powerful but also more sensitive (Fig 1). A long-term objective is of
course to merge these two qualities and produce energetic molecules that are both more
3
Henric stmark
233
Czech Republic
2006
[Content]
powerful and less sensitive than current materials. Compounds with the performance of CL20 and the sensitivity of TATB would be desirable long-term goals.
To increase performance compared to current explosives the molecules need to contain
particularly energetic structures that give a high heat of formation (e.g. caged molecules,
strained rings, nitramines, furazanes, tetrazines) and/or have high-density functional groups
(e.g. fluorine, cage, urea, guanidine). Computational methods allow quite accurate prediction
of density and heat of formation, which makes it possible to evaluate large numbers of
candidate molecules before their preparation is attempted.
It is more difficult to predict the sensitivity of new energetic materials, which leads to a
more qualitative analysis of the proposed target molecules. Factors that influence sensitivity
are for instance functional groups (e.g. nitrate esters), crystal structure and hydrogen
bonding. It is also possible to influence the sensitivity of explosives by perfecting their
crystal shape. Round, flawless crystals are desired, both because of their easier processability
and their decreased sensitivity. Recently the long studied crystallization of RDX yielded the
product I-RDX [2] which is equal to normal RDX performance-wise, but has improved
sensitivity characteristics towards shock. This breakthrough has sparked an interest for the
preparation of I-HMX [3] and perhaps even I-CL-20.
The most promising new candidate molecules are CL-20, ADN, and FOX-7, which have
all been produced in sufficient quantities to allow them to be used for various applications.
TATB and NTO are two low sensitivity energetic materials that are already in use.
Current
RDX, HMX
TNT
Sensitivity
FOX-7, NTO, TATB
Performance
CL-20
HEDM
Lower sensitivity
Increased
Performance
Fig 1. Lines of development of energetic materials.

A number of potential HEDM molecules have been theoretical or experimental
investigated. In some of the most promising or interesting shown . A short description of
each is also given below.
CL-20 or Hexanitrohexaazaisowurtzitn contains more energy than earlier high
performance energetic materials, e.g. HMX. The purely chemical theoretical increase in
performance is 20%. In shaped charges, because of the higher energy content in the
explosive charge, different materials can be used in the liner yielding an increase in
performance of about 60%. CL-20 is, however, as sensitive as PETN regarding impact and
friction. As for HMX there are several polymorphs of the substance adding further
complexity to its handling. Due to a difficult synthesis the price is also quite high.
Octanitrocubane or ONC was first envisioned by Gilbert in the beginning of the 1980s
as a compound with a perfect oxygen balance and exceptionally high heat of formation and
density [4]. As the compound was a C-Nitro explosive good sensitivity properties were also
234
Czech Republic
2006
[Content]
expected [4]. These predictions were also supported by advanced calculations. The
compound was finally prepared in the year 2000 after a very long and laborious synthesis but
proved to be a disappointment with a lower than expected density [4]. It turned out that the
heptanitrocubane was a superior explosive due to its higher density, there are however hopes
to find a higher density polymorph of ONC which might live up to the predictions.
O2N
NO2
N NO
2
O2N
N
O2N
NO2
O2N
NO2
O2 N
NO2
N
O2N
O2 N
NO2
NO2
Octanitrocubane (ONC)
CL-20
O2N
NO2
NO2
NO2
O2N
O2N
O
N
O2N
HHTDD
O
N
N
N
O
NH
NO2
NO2
Tetranitro-tetraaza-cubane
O
NO2
N
N
HN
TTTO
Oxadiazolo-oxadiazole-dioxide
Fig 2. HEDM molecules possible by the use of conventional chemistry

2,4,6,8-Tetranitro-1,3,5,7-tetraaza-cubane a proposed compound with a structure
similar to that of ONC. The proposed high density is derived from the cubane structure
which also contributes to its heat of formation. The compound would also be perfectly
oxygen balanced. The calculated performance is 1.4 times that of HMX.
HHTDD or cis-syn-cis-2,6-Dioxo-1,3,4,5,7,8-hexanitrodecahydro-1H, 5H-diimidazo[4,5b:4,5-e] pyrazine. The dinitrourea moiety confers high density to molecules containing it.
A property, which has been realized and utilized in a whole series of experimental energetic
materials. HHTDD [5] is supposed to be the explosive with the best performance ever
prepared, it lacks however both in sensitivity and hydrolytic stability something it shares
with its close analogue Sorguyl (TNGU) [6]. Our calculations based on experimental density
and heat of formation implies a performance slightly below that for CL-20.
TTTO or [1,2,3,4]Tetrazino[5,6-e][1,2,3,4]tetrazine 1,3,5,7-tetraoxide is a compound
that has not been prepared but which has been evaluated by computational techniques and
found to be extremely interesting. Calculations propose a density of 2.38 g/cm3 and a heat of
5
Henric stmark
235
Czech Republic
2006
[Content]
formation of 190 kcal/mole, leading to a relative performance of 2.2 times HMX. This
compound can be seen as a possible performance limit to normal explosives and what could
be achieved by a focused, well funded research program.
1H,4H-[1,2,5]Oxadiazolo[3,4-c][1,2,5]oxadiazole 3,6-dioxide is another proposed
bicyclic heterocycle similar to TTTO. The two nitrogenatoms are N-oxidized and the
compound has a proposed high density (2.3 g/cm3) which would give it excellent
performance as a high explosive (1.7 times HMX).
In Table 2 the calculated densities, heats of formation and performances of the described
compounds are compiled. Data were calculated by the methods described in Reference 1.
The values for ONC were taken directly from the Cheetah 2.0 database.
Density (g/cc)
Heat of formation (kJ/mol)
Detonation velocity (mm/s)
Detonation pressure (GPa)
Energy (HMX=100, V=2.2 V0)
Table 2. Calculated performance of HEDM prepared or possible to prepare by ordinary

chemistry
CL-20
2.04*
393*
10.0
47.8
121
ONC
1.92
381
9.2
37.1
105
Tetranitrotetraaza-cubane
2.02
912.4
10.5
50.1
137
HHTDD
2.07*
153.4*
10.2
46.1
118
TTTO
2.38
794.6
10.9
131
220
Oxadiazolooxadiazoledioxide
2.30
263.8
11.3
71.3
167
Energetic material
* Experimental data
In conclusion, the evolutionary development of energetic materials during the last decade
gives a potential for future increase in systems performance of up to 100%. The foundation
has been laid, but more applied research is needed in order to fill the requirements of specific
weapons systems. This would also make it possible to tailor energetic materials and high
explosives for various applications.
4.
ALL-NITROGEN COMPOUNDS
Polynitrogen compounds have received considerable attention in recent years as potential

candidates for high energy density materials (HEDMs). The fact that most of these have a
very high calculated energy content is a consequence of the large bond energy of 228
kcal/mol of the triple bond in molecular nitrogen (N2) compared to the bond energies of
nitrogen-nitrogen single and double bonds, which typically amounts to 39 and 100 kcal/mol,
respectively. In addition to the favorable energetics of the polynitrogen compounds, they
also have the advantage that the energy releasing decomposition into N2 leads to no
deposition of harmful substances. This makes them particularly suitable for propulsion of
236
Czech Republic
2006
[Content]
spacecrafts, since the commonly used propellants of today results in large depositions of
harmful substances in the atmosphere. The special properties of polynitrogen compounds
also make them of great interest for energy storage. The high heat of formation and large
density makes polynitrogens nearly ideal for use as high explosives.
Even though theoretical studies have identified a number of potentially stable structures,
the experimentally observed nitrogen compounds are still few. The azide anion, N3-, was first
synthesized in 1890 by Curtius. Christe and coworkers have since 1999 reported the
preparation and isolation of N5+ together with several different counter ions [7]. A few other
species, such as N3, N3+ and N4+, have been observed only as gaseous or matrix-isolated ions
or radicals. The formation of N5- from arylpentazoles in mass spectrometry experiments has
recently been reported from our lab [8, [9] and from Air Force Research Laboratory (Edwards
Air Force Bas) in the US [10]. The detection of HN5 and a zinc pentazolate salt in solution
has also been reported [11]. These results have however been questioned [12].
Quantum mechanical studies of stability and energy content have been carried out on a
number of nitrogen compounds. For a recent review on the structure and stability of N4 and
N5 systems see Nguyen [13] and references therein. An overview of some calculated
structures and stabilities of N2 to N10 species as well as their spectroscopic characteristics
can be found in a report by Bartlett [14]. Many all-nitrogen structures have also been treated
by Glukhovtsev el al. [15]. Theoretical calculations lay the foundation for ongoing attempts to
synthesise such compounds and increases the possibility of successfully producing one or a
few useful substances and formulations. A possible development in this field is substances
allowing an increase in performance as high explosives of 25 times the values of today. The
impulse of for example N4 is comparable to H2/O2 but has the enormous advantage of a
much higher density which would allow for smaller and thereby lighter rockets.
Quantum mechanical calculations to determine the heat of formation for a number of
poly-nitrogen molecules (Fig 3) have been made at FOI using Gaussian 98 at B3LYP/631G(d) level of theory.
Fig 3. Calculated structures of N4(D2h), N4(Td), N6(D3h), N8(Oh), N10(D5h) and N12(D6h)
The difference between Hf(g) and Hf(s) is the sublimation energy, i.e. the energy
gained by the molecules by forming a solid or liquid instead of gas. The size of this
difference is one indication of a substances inclination to be in solid phase.
The estimation of the state of aggregation for a substance from calculations is a difficult
task. In Table 3 one can see that very little is gained for N4(Td) so this may not be expected
to be in solid state at ambient temperature and pressure. Absolute estimations are very
difficult to make. Use of the methods described by Murray [16] give boiling points. The
accuracy of these calculations varies with substance. For comparison, the calculated boiling
point for hydrazine is 338 K (exp 386 K) and for tetrazole 434 K (exp 429 K mp sublimes).
The trend is clear; N8, N10 and N12 are liquids at room temperature, N10 and N12 may even be
solids. N4 is a gas at room temperature, but would be a liquid or even probably solid at liquid
nitrogen temperature.
Interesting from a point of view of usefulness of new materials are properties such as
energy content (H), density, energy barrier towards dissociation, i.e. stability and
aggregation form. Some of these properties such as aggregation form are more difficult to
estimate than others. From an experimentalists point of view in trying to make these
7
Henric stmark
237
Czech Republic
2006
[Content]
molecules it is also necessary to have reasonable synthesis routes, something that can often
also be studied with quantum mechanics. One of the most important questions for an
experimentalist to answer is how to detect the successful synthesis of the molecule. Quantum
chemistry is very useful in supplying answers.
Table 3 Heats of formation from QM calculations (Gaussian 98, hs95fo), densities
calculated with MOLPAK and detonation velocities, pressures and performances (except Isp)
calculated with Cheetah 2.0 (BKWC), Isp calculated with NASA CEA.
Impulse (sec)
Tc (K)
Force (J/g)
121
289
313
408
7563
3932
N4(Td)
202
201
2.3
15.7
125
200
327
422
7795
4212
N6(D3h)
346
342
2.5
17.6
169
285
428
439
8079
5193
N8(Oh)
480
473
2.7
19.0
206
331
498
444
8162
N10(D5h)
551
544
2.7
19.2
208
364
484
433
7971
4628
N12(D6h)
712
703
2.8
19.7
224
407
526
443
8140
Hf(s) (kcal/mol) a
(HMX=100, V/V0=2.2)
15.5
Energy
2.2
Boiling point (K) c
Detonation Pressure (GPa)
Density (g/cm3)
189
(kcal/mol)
194
Sublimation energy
N4(D2h)
Hf(g) (kcal/mol) a
Detonation velocity (km/s)
Molecule
Heat of formation is calculated at B3LYP/6-31G(d) level of theory. This level has been chosen for
comparative purposes and does not give the most accurate values calculated. The differences to heats of
formation calculated at higher levels of theory are however small and do not affect the calculated performance
in an important way.
b
Results by Karle [17] indicate a higher sublimation energy for N4(Td) (5 kcal/mol) which would of course mean
a higher boiling point.
c
The boiling point has been calculated using a lower level of theory, STO-5 using the method described in ref
There are many papers describing quantum mechanical calculations on various allnitrogen molecules. Calculations on larger nitrogen molecules are mainly focused on
geometry and finding stable forms. Little has been done on determining chemical or thermal
stability and performance. For any of these molecules to be of experimental interest, there
must also exist ideas on synthesis routes. All of these compounds have high energies (based
on the number of single or double bonds), but in order to assess their performance, their
density must also be calculated. Only a few, such as N4 and N5-, have been extensively
studied with quantum mechanics and other calculations. In this paper N4(Td) and N5- will
serve as example on how to proceed towards the synthesis of HEDM with the aid of QM
calculations.
238
4.1
Czech Republic
2006
[Content]
Tetraazatetrahedrane, N4(Td)
One of the most interesting candidates for a HEDM among the polynitrogen compounds
is tetraazatetrahedrane, N4(Td). Phosphorous (below nitrogen in the periodic table) exists in,
among other forms, the form of P2 as well as P4(Td) (white phosphorous) whereas nitrogen is
only known in the form of a dimer (N2).
Energy content and stability
In 1990, Francl and Chesick [18] calculated the energy of formation of N4 (814 kJ/mol) as
well as its dissociation barrier (315 kJ/mol). They concluded that the calculated barrier was
large enough to suggest a significant lifetime and metastability if N4 could be made and
trapped. The energy release involved (heat of reaction) in the dissociation of N4(Td) into N2
has been calculated in many studies, giving similar results. Lee and Martin [19] give a value
of 763 2 kJ/mol.
The dissociation barrier has also been discussed in many studies with results ranging from
118 to 315 kJ/mol. The height of the dissociation barrier, i.e. the activation energy, can be
calculated with much higher accuracy now than only ten years ago. The first studies used
very simple approximations to estimate the height. Francl and Chesick [18] used the lowest
energy point on the ridge of intersection between the potential energy surface (PES) of
ground state N4 and the PES of two N2 molecules, giving the high value of 315 kJ/mol.
Yarkony [20] used a much more refined method taking into account that the potential energy
surfaces of other states may cross the investigated PES (Fig 4). This resulted in the low value
of 118 kJ/mol. A later study taking in more information about the system but essentially
using the same principle has given a higher value of 214 kJ/mol [21]. Despite the incredibly
high energy content, this computed dissociation barrier indicates that it is sufficiently stable
for most potential applications.
TS
Ea
N4 GS
H
2 N2
Fig 4. Schematic picture of a potential energy curve (blue) for the dissociation of a
substance via a transition state (TS) releasing the energy H. A potential energy curve
(green) from a different state crosses it, effectively lowering the activation energy Ea.
The high energy content of N4(Td) makes it hard to make by traditional methods in
chemical synthesis. All attempts of experimental preparation and detection of N4(Td) have so
far been unsuccessful. Radziszewski and coworkers reported to have detected N4(Td) in 2000
[22]
, but it was later shown that the reported isotope effect on the observed IR-band is not
compatible with N4(Td) [19]. An open chain N4 isomer in the triplet state was recently
detected in mass-spectrometry experiments [23, [24, [25]. However, this species is expected to
be too unstable and also too low in energy to be of any interest as a HEDM.
Henric stmark
239
Czech Republic
2006
[Content]
Suggested synthesis routes

A number of synthesis routes have been suggested, mainly by our institute, and in some
cases the chances of success using these routes have been assessed by quantum chemical
calculations. Francl and Chesick [18] suggested that prolonged irradiation of liquid nitrogen
with radiation of wavelength less than 140 nm might yield evidence of N4 formation. (Note
that the calculated wavelength here is too high. The energy of the barrier has to be added to
the photon energy which gives ~106 nm.) This suggestion was of course based purely on an
energetic basis. In order to increase the chances of success, more refined arguments and
investigations can be made. In Fig 5 an overview of possible synthetic routs to N4(Td) is
shown.
I
N4(Td)
N2* + N2
N(2D) + N2
III
(alt.)
N(2D)
cyc-N3(C2v)
h
ClN3
N4(Td)
O
t-Bu
t-Bu
VI
N
N3* + cyc-N3(C2v)
NXY + cyc-N3+(D3h)
t-Bu
N N
t-Bu
t-Bu
t-Bu
-N2
IV
N4(Td)
N4(Td)
cyc-N3(C2v)
-Cl
+ cyc-N3(C2v)
t-Bu
N2(A3+u) + N2(A3+u)
II
N N
t-Bu
h
t-Bu
t-Bu
N4(Td)
t-Bu
t-Bu
N
N N
- NX-1Y+1
Actual synthesis
route for
tetratertbutylhedrane
and suggested route
for N4(Td)
X=O,S,Se,NH,..
Fig 5. Proposed routes for synthesis of N4(Td). Excited states are marked by *.
I. The collision of a ground-state N2 with an excited-state N2 to form N4 requires that
their total energy is high enough to form the molecule and reach over its dissociation barrier.
The triplet excited states generally have longer lifetimes than the singlet excited states and
are therefore most interesting for this type of reaction. The result would be a N4 molecule of
triplet character. Calculations [26] show that metastable excited triplet states exist, which
increases the chances of success. However, the electronic structure of ground-state N2 is not
optimal for formation of new N-N bonds. The required excited states can be produced by
excitation of nitrogen in liquid or solid form. Excited nitrogen molecules can also be
obtained by RF (microwave cavity) or electrical discharges (hollow cathode or pulsed nozzle
discharge), and the reaction products can be trapped on a cold surface. Ion bombardment of
solid nitrogen or nitrogen in rare gas matrices is another possibility.
II. Another suggested route is N2(A3+u) + N2(A3+u) N4(Td). This reaction has the
advantages that the orbital occupancy of N2(A3+u) is favorable for formation of new N-N
bonds, and that this state has a long life time (1.9s). The collision of two molecules in an
240
Czech Republic
2006
[Content]
excited state is generally very difficult to achieve but there are ideas on how to succeed. Ion
or -particle bombardment of solid nitrogen in argon has been shown to excite nitrogen to
the A3+u state [27]. It is likely that molecules are excited primarily along the trajectory of the
ions or -particles through the matrix. This enhances the chances of two molecules next to
each other being excited to the correct state.
III. On the basis of quantum chemical calculations we have suggested that N4(Td) may be
synthesized from molecular nitrogen and atomic nitrogen in the 2D state in a two step
process [28]. In the first step a cyclic N3 isomer, cyc-N3(C2v) is formed by the addition of N2
to N(1D). Addition of a second N(1D) to N can lead to the formation of N4(Td). Nitrogen
atoms in the 2D state have a long lifetime in cryogenic matrices (37 s) and can be generated
at high concentrations. The main drawback of this route seems to be side reactions. Linear
N3, N3(D) can be formed in the first step and 2 N2 in the second step. The existence of cycN3(C2v) has been confirmed in a recent experimental study, where it was produced by
photolysis of ClN3 [29]. An experiment could be setup where the photolysis products are
deposited in a nitrogen matrix with a high concentration of N(2D). In addition, recent
unpublished computations from our group indicate that a cyclic N3+ isomer, cyc-N3+(D3h),
exists, which can be obtained directly by ionization of cyc-N3(C2v). These two new species
opens up new potential pathways for synthesis of N4(Td) and other polynitrogen allotropes.
IV. It can be noted that both the production of N3 from N(2D) + N2 and ClN3 is likely to
lead to a mixture of N3(D) and cyc-N3(C2v). (A 0.8/0.2 mixture was obtained from ClN3.)
The reaction between N3(D) and cyc-N3(C2v) to form N4(Td) and N2 should be considered.
Preliminary calculations indicate that this reaction could have a too high barrier to be
feasible. However, the reaction between cyc-N3(C2v) and an excited state of N3 may be
barrierless. This could be achieved experimentally by simultaneous application of two laser
sources, one tuned for photolysis of ClN3 and the other to the proper excitation wavelength.
V. Unpublished calculations show that a stable cyclic N3+ isomer, cyc-N3+(D3h), can be
formed by ionization of cyc-N3(C2v). Although it is higher in energy than ground state linear
N3+, it is very stable towards unimolecular decomposition. The reaction between cycN3+(D3h) and N3- should be investigated. It may be possible to deposit both ions into a
matrix. Also reactions between cyc-N3+(D3h) and other neutral and ionic states of N and N3
should be explored by computations to find possible pathways for formation of N4(Td).
VI. Tetratertbutyltetrahedrane can be formed by photolysis of tetra-tertbutylcyclopentadienone in argon matrices [30]. It may be possible to form tetraazatetrahedrane
(N4(Td)) by a similar method starting from tetrazole-5-one. This starting material is not
known from the literature and low level (B3LYP/6-311G) calculations indicate that it is not
stable, but higher level calculations should be performed to test the stability of this and
alternative starting materials. Another option is to use arylpentazoles as starting material.
This type reaction is suitable for mass spectrometric studies with laser-induced
decomposition of suitable precursors.
Detection
The symmetric nature of N4(Td) means that IR spectroscopy alone is insufficient. The
only line available gives no identification capabilities. The difficulties are clearly illustrated
by the questionable assignment of the origin of an IR line as N4(Td) [22], made by
Radszisewski et al. Lee and Martin showed that the experimental isotopic shift upon 15Nlabeling unfortunately does not match the theoretical predictions for N4(Td) [19].
11
Henric stmark
241
Czech Republic
2006
[Content]
Raman spectroscopy is a much better choice of detection method for N4(Td). It gives a
fingerprint of the molecule with three lines. The detection limit in a matrix has been
determined by combining experimental signal strength and noise levels of N2 in solid
nitrogen with ab initio quantum mechanical calculations of Raman intensities of N2 and
N4(Td) [31]. Using that particular experimental setup 80 ppm is necessary for detection using
8 h measurement time and 1W laser power. This detection limit is too high considering that
no optimization of a signal can be made during the measurement. An improvement of the
experimental setup, for example by using a Raman probe, could lower the detection limit
rendering Raman spectrometry a feasible detection method.
A more sensitive Raman method, Surface Enhanced Raman Spectroscopy (SERS), has
been evaluated for the purpose of N4 detection using white phosphorous, P4(Td) as a model
substance [32, [33]. No SERS effect was observed for P4 which may indicate that SERS is not
a promising detection method for N4.
Detection with Laser-Induced Fluorescence (LIF) would be ideal due to its sensitivity.
However, accurate calculations of excitation energies and geometry optimizations of the
excited states are necessary to determine laser and fluorescence wavelengths. An excited
state of N4(Td) has been identified by EOM-CCSD theory which is 10.89 eV above the
ground state, and which is optically active for one and two-photon absorption [34]. However,
it has not yet been possible to optimize the excited state geometry. A similar study of
rectangular N4, N4(D2h), has successfully given a useful route for LIF detection [35]. The
vertical absorption and emission energies for the first excited state were determined to 1.58
and 1.55 eV respectively (laser and detection wavelengths 785 and 800 nm respectively).
It may be possible to use the previously discussed state at 10.89 eV to detect N4(Td) by
UV absorption at around 114 nm. This is just outside the relatively easily accessed range
where PM-tubes, MgF2 windows and deuterium lamps work so extra attention to the
experimental setup would be necessary.
Mass spectroscopy is another option for N4 detection, but it is difficult to tell whether an
m/z 56 fragment is due to a bound D2h or Td structure or is the result of a weakly bound (N2)2
complex. There is also a need for calculations to determine whether ionized N4(Td) is bound
and stable enough to detect in a mass spectrometer.
Isotopic labeling can be used in conjunction with the above techniques to confirm or rule
out certain origins of spectral lines and mass spectrometric fragments.
4.2
Roadmap to pentaaza HEDM:s ( N5+/N5-/N5+N5-)
The making of High Energy Density Materials with fivemembered nitrogen compounds
starts with the synthesis N5+ or N5-. Recently the novel homoleptic polynitrogen ion, N5+ (Fig
6) was synthesized [7]. This is the first new stable all nitrogen ion in more than 100 years. So
far no useful counter ion has been found. It has been suggested that cyclo-N5- (Fig 7) may be
combined with N5+ to form a N10 salt.
Fig 6. N5+ the first new all-nitrogen

compound in 100 years
242
Fig 7. Pentazolate anion, cyclo-N5-
Czech Republic
2006
[Content]
The nitrogen pentazole ring system has been studied since 1903, when Hantzsch made the
first attempts to prepare phenylpentazole [36]. The first preparation of substituted pentazoles
was however achieved by Huisgen and Ugi [37] in 1956. The making of pentazolate
compounds starts with decomposition of arylpentazoles (
Fig 8). A breakthrough was achieved when the feasibility of this process was confirmed
in a laser induced mass spectrometry experiment at our institute [8], in which the pentazolate
anion was detected from selective cleavage of p-dimethylaminophenylpentazole.
2N
N
N
Ti
N
N
N
N
N
N
N
R
N
N
N
N
Me
N
N
sol
N
N
N
N
N
N
Me=Fe, Mn, Co..........

N
N
N
Zn 2+
N
N
N
N
N
N
Fig 8. How to make pentazolate HEDMs.

Calculations have shown that the aromatic pentazole anion (N5-) should be stable enough
to isolate, especially if trapped as a metal complex [38, [39, [40, [41, [42, [43]. This complex could,
if a suitable metal were used as the complexing agent, be highly energetic. More important,
it could serve as a N5--containing compound, stable enough for an ion exchange to be
performed. Some metals are believed to stabilize the pentazole ring, both in the
arylpentazole and in pure ionic form. In a recent study [44] to investigate if amination of
halobenzenes is a viable way to isolate the pentazolate anion, the reactivities of halobenzenes
were compared to phenylpentazole. The reaction energetics for phenylpentazole became
favorable after complexation with Zn2+. The study indicated that addition of Zn2+ lowers the
free energy of activation by more than 6 kcal/mol, which may be sufficient to make the
reaction useful for synthesis.
Once a stable complex with N5- has been made, an ion exchange can be attempted in
order to find a suitable counter ion that makes a good HEDM. Calculations show that a salt
with the recently discovered [7] N5+ ion would be highly energetic, with a performance as an
explosive about 1.6 times that of HMX [45], provided of course that it is stable. QM
calculations on the stability of N5+N5- are not conclusive. There are three studies of its
stability, two of which predict that N5+N5- is stable and one that predicts its instability.
The first publication, by Fau et al. [46, [47], concludes that N5+N5- is a stable compound
with a dissociation barrier of 19 kcal/mol and high density (1.9 g/cm3) and heat of formation
13
Henric stmark
243
Czech Republic
2006
[Content]
(296 kcal/mol, 1239 kJ/mol). They conclude that its performance as a rocket propellant is
about the same as that of hydrazine but that its high density (about twice that of hydrazine)
will allow for smaller and lighter rockets.
The second publication, by Evangelisti et al. [48], concludes that N5+N5- is stable and that
(N5+N5-)2 is even more stable. They also conclude that the dimer stability suggests the
possibility of a N5+N5- crystal. Worth noticing here is that even though no crystal structure is
calculated, calculations indicate that (N5+N5-)2 has a planar, layered structure. This is very
interesting since it is believed that such structures give less sensitive energetic materials [49].
The third publication, by Dixon et al. [50], concludes that both N5+N5- and N5+N3- are
unstable. They have also experimentally confirmed that N5+N3- really is unstable.
Interestingly enough, despite their negative results, they state that a hypothetical
polynitrogen compound, such as N5+N5-, would be an excellent monopropellant for rocket
propulsion or explosives, if it could be synthesized in a stable form.
The fourth publication by Gagliardi et al. [51] concludes that the reaction to form a
bicyclic form of N10 from free ions proceeds without a barrier. If we, however, allow the
two ions to approach each other along the energetically most favourable path, an ion pair
can be formed that is bound with 117 kcal/mol with respect to the free ions. The ion pair is
24 kcal/mol higher in energy than the bicyclic form.
Calculations using Cheetah 2.0 give Isp=281 s for a HTPB/N5+N5- mixture with 70 vol%
solids.
5.
METASTABLE STATES: POLYMERIC NITROGEN
The natural state (low energy form) of the elements nitrogen and hydrogen is as diatomic
molecules, N2 (triple bond) and H2 (single bond). Much energy can therefore be stored in
nitrogen molecules with single or double bonds. A form of nitrogen with all single bonds
and therefore high energy content is polymeric nitrogen. It also has high density (compare
with carbon in the form of graphite versus diamonds). Theory suggests that if a volume of
fluid N2 is sufficiently squeezed, the nitrogen will give up its diatomic structure and form a
continuous network of single covalent bonds [52]. This chemical structure is similar to that of
its chemically similar group V neighbours phosphorus and arsenic. The energy released
when the densely packed polymeric nitrogen is released to the diatomic state would be
pollution free and enough to qualify it as an interesting future HEDM. Theoretical
calculations first predicted the existence of a metastable polymeric nitrogen phase at the
pressure 65 GPa [52]. In high pressure experiments several new phases of nitrogen have been
found, among them a non-molecular phase that could be recovered to ambient pressure [53].
The most promising cubic gauge structure (cg-N) was predicted to have high energy and
high density compared to -N2.
The theoretically estimated performance is Isp 516s, density 3.9 g/cm3, and the
performance as an explosive is 10.6 times the performance of HMX. This is based on
thermochemical calculations with Cheetah 2.0 made with BKWC equations of state and with
data from Mailhiot [52]. Notice that the equation of state used (BKWC) only is calibrated up
to 40 GPa so this is a very large extrapolation.
244
Czech Republic
2006
[Content]
Fig 9. Polymeric cg-N-structure: Each nitrogen atom is connected to three neighbours by

three single covalent bonds [54].
A group under Mikhail Eremets has managed to prove its existence [55]. To synthesize it,
a diamond anvil cell was used (Fig 10). It is a robust device with a small (~20 m diameter)
sample chamber within which pressures exceeding that at the centre of the earth (~350 GPa)
have been reached. In this experiment 110 GPa at temperatures above 2000 K was reached
by heating of a laser. The existence of cubic gauge structure was experimentally verified by
X-ray and Raman measurements.
Fig 10. A cross-section of the sample arrangement. The heated nitrogen sample is coloured
blue and untransformed molecular nitrogen is red. The laser beam is absorbed by a black,
thin, boron plate resting on c-BN pieces that thermally insulate the plate from the bottom
anvil. The sample squeezed between the anvils is surrounded by the c-BN/epoxy gasket
followed by the metallic (Re) supporting ring [55].
To get a useful substance it needs to have a huge hysteresis enabling it to be recovered to
ambient pressures. In the experiment above the structure was stable down to 42GPa at room
temperature. At low temperatures the pressure could be further released, at 140K it reached
25GPa before the gasket ruptured. Its a typical problem of the current technique and a novel
method of recovering the sample to ambient pressure needs to be developed. Theory predicts
that this structure is metastable at atmospheric pressures [52, [56]. It would then be an analogue
to carbon that is stable as graphite at atmospheric pressure but also metastable as diamond
created at very high pressures and temperature. If cg-N is not metastable at ambient pressure,
it can perhaps be stabilized in compounds with other elements or by introducing impurities
[54]
.
15
Henric stmark
245
6.
Czech Republic
2006
[Content]
CONCLUSIONS
The available energy in exotic materials such as nuclear isomers and antimatter is orders
of magnitude larger than the energy of HMX (Fig 11). These have very high potential but the
risk of failing is also very high. On a more intermediate level, materials such as all-nitrogen
compounds have high potential and lower (but still high) risk. Other materials such as
reactive materials based on nanotechnology and HEDM by conventional chemistry have
good potential, will probably be accessible in shorter time and involve much less risk of
failure. In Table 4 a summary of potential HEDM is given. An increase of performance of
three to six times the performance of HMX would revolutionize warfare. This can be
compared to the performance increase of three times at the end of the black powder era
which is known to have changed the battlefield enormously.
Fig 11. Energy content of some HEDMs. The figure is from LLNL (internet). The point for
Anti-matter has been added to the figure. The comment in the figure refers to nuclear
isomers.
246
Czech Republic
2006
[Content]
Detonation pressure (GPa)
Energy (HMX=100, V=2.2 V0)
Energy (HMX=100, V=40 V0)
-708
7.3
21.2
60
65
226
992
1.65
-63
6.9
19.6
55
63
210
897
1.905
74.75
9.1
38.5
105
103
266
1397
2.04
393
10
47.8
121
116
273
1380
1.9
1239
12.1
62
162
156
347
~2000
2.3
761
15.5
121
308
288
422
3700
N60
1.97
6780
12.3
65
161
150
331
2296
Polymeric nitrogen
3.9
290
30
660
1060
-a
513
-a
2.38
795
10.9
131
220
-a
288
-a
Density (g/cc)
Force (J/g)
Detonation velocity (mm/s)
1.65
Energetic material
Impulse (s)
Heat of formation (kJ/mol)
Table 4. Comparison of performance for common high explosives, high nitrogen compounds
and some theoretical HEDMs, calculations using Cheetah 2.0 with BKWC EOS.
Reference compounds:
NC, Nitrocellulose
O 2N
NO 2
TNT
O2N
NO 2
N
O 2N
NO 2
HMX
NO 2
O2N
NO2
N NO
2
O2N
N
CL-20 O N
2
NO2
HEDM:
N5+N5N
N4(Td)
O
N
N
N
O
TTTO
a
17
N
N
N
O
Calculations do not converge.
Henric stmark
247
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
248
WALLIN, S., STMARK, H., WINGBORG, N. et al. (2004) High Energy Density Materials
(HEDM) - A Literature Survey (Stockholm, Swedish Defence Research Agency).
FRECHE, A., AVILES, J., DONNIO, L. & SPYCKERELLE, C. (2000) Insensitive RDX, I-RDX,
Paper presented at the Insensitive Munitions and Energetic Materials Technology
Symposium, San Antonio.
FRECHE, A., LECUME, S., SPYCKERELLE, C., AVILES, J. & DONNIO, L. (2003) Insensitive
Nitramines, Paper presented at the Europyro 8e Congrs International de Pyrotechnic, SaintMalo.
EATON, ZHANG, M.-X., GILARDI, R. et al.: Octanitrocubane: A New Nitrocarbon,
Propellants, Explosives, Pyrotechnics, 27, p. 1-6.2002
VEDACHALAM, M., RAMAKRISHNAN, V. T., BOYER, J. H. et al.: Facile Synthesis and
Nitration of cis-syn-cis-2,6-dioxo-1,3,4,5,7,8-hexanitrodekahydro-1H,5H-diimidazo[4,5b:4',5'-e]pyrazin, Journal of Organic Chemistry, 56, p. 3413-3419.1991
PAGORIA, P. F., LEE, G. S., MITCHELL, A. R. & SCHMIDT, R. D.: Thermochimica Acta, 384,
p. 187-204.2002
CHRISTE, K. O., WILSON, W. W., SHEEHY, J. A. & BOATZ, J. A.: N5+: A Novel Homoleptic
Polynitrogen Ion as a High Energy Density Material, Angewandte Chemie International
Edition, 38, p. 2004-2009.1999
STMARK, H., WALLIN, S., BRINCK, T. et al.: Detection of pentazolate anion (cyclo-N5-)
from laser ionization and decomposition of solid p-dimethylaminophenylpentazole.,
Chemical Physics Letters, p.2003
HAHMA, A., HOLMBERG, E., HORE, N. et al. (2002) Synthesis and Characterization of
Phenylpentazoles, Paper presented at the 33rd International Annual Conference of ICT,
Karlsruhe.
VIJ, A., PAVLOVICH, J. G., WILSON, W., VIJ, V. & CHRISTE, K. O.: Experimental Detection
of the Pentaazacyclopentadienide (Pentazolate) Anion, cyclo-N5-, Angewandte Chemie
International Edition, 41, p. 3051-3053.2002
BUTLER, R. N., STEPHENS, J. C. & BURKE, L. A.: First generation of pentazole (HN5,
pentazolic acid), the final azole, and a zinc pentazolate salt in solution: A new N-dearylation
of 1-(p-methoxyphenyl) pyrazoles, a 2-(p-methoxyphenyl) tetrazole and application of the
methodology to 1-(p-methoxyphenyl) pentazole, Chemical Communications, p. 10161017.2003
SCHROER, T., HAIGES, R., SCHNEIDER, S. & CHRISTE, K. O.: The race for the first generation
of the pentazolate anion in solution is far from over, Chemical Communications, p. 16071609.2005
NGUYEN, M. T.: Polynitrogen compounds 1. Structure and stability of N4 and N5 systems,
Coordination Chemistry Reviews, 244, p. 93-113.2003
BARTLETT, R. J. Structure and Stability of Polynitrogen Molecules and Their Spectroscopic
Characteristics.
GLUKHOVTSEV, M. N., JIAO, H. & SCHLEYER, P. V. R.: Besides N2, What Is the Most Stable
Molecule Composed Only of Nitrogen Atoms? Inorganic Chemistry, 35, p. 7124-7133.1996
MURRAY, J. S., LANE, P., BRINCK, T. et al.: Relationships of Critical Constants and Boiling
Points to Computed Molecular Surface Properties, Journal of Physical Chemistry, 97, p.
9369-9373.1993
KARLE, J. (2000) A discussion of azacubane and some peripheral remarks about other N8
compounds, Paper presented at the HEDM contractors meeting, Park City, Utah.
FRANCL, M. M. & CHESICK, J. P.: The N4 Molecule and Its Metastability, Journal of Physical
Chemistry, 94, p. 526-528.1990
LEE, T. J. & MARTIN, J. M. L.: An accurate quartic force field, fundamental frequencies, and
binding energy for the high energy density material TdN4, Chemical Physics Letters, 357, p.
319.2002
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
19
Czech Republic
2006
[Content]
YARKONY, D. R.: Theoretical Studies of Spin-Forbidden Radiationless Decay in Polyatomic

Systems: Insights from Recently Developed Computational Methods, Journal of the American
Chemical Society, 114, p. 5406-5411.1992
BRINCK, T., BITTEREROVA, M. & STMARK, H. (2003) Electronic structure calculations as a
tool in the quest for experimental verification of N4, in: Politzer, P. A. & Murray, J. S. (Eds.)
Energetic Materials, Part 1: Decomposition, Crystal and Molecular Properties, pp. 421-439
(Elsevier).
ZHENG, J. P., WALUK, J., SPANGET-LARSEN, J., BLAKE, D. M. & RADZISZEWSKI, J. G.:
Tetrazete (N4). Can it be prepared and observed? Chemical Physics Letters, 328, p. 227233.2000
CACACE, F.: From N2 and O2 to N4 and O4: Pneumatic Chemistry in the 21st Century,
Chemistry - A European Journal, 8, p. 3838-3847.2002
CACACE, F., PETRIS, G. D. & TROIANI, A.: Experimental Detection of Tetranitrogen, Science,
295, p. 480-481.2002
NGUYEN, M. T., NGUYEN, T. L., MEBEL, A. M. & FLAMMANG, R.: Azido-Nitrene is
Probably the N4 Molecule Observed in Mass Spectrometric Experiments, Journal of Physical
Chemistry A, 107, p. 5452-5460.2003
BITTEREROV, M., BRINCK, T. & STMARK, H.: Theoretical study of the triplet N4 potential
energy surface, Journal of Physical Chemistry A, 104, p. 11999-12005.2000
SCULLMAN, R., WALLIN, S., LAUBERSHEIMER, F. & REHM, O.: Luminescence from Matrices
of Nitrogen and Argon Excited by -Particles, Physica Scripta, 52, p. 172-177.1995
BITTEREROV, M., STMARK, H. & BRINCK, T.: A theoretical study of the azide(N3) doublet
states. A new route to tetraazatetrahedrane (N4); N+N3->N4, Journal of Chemical Physics,
116, p. 9740.2002
HANSEN, N. & WODTKE, A. M.: Velocity Map Ion Imaging of Chlorine Azide Photolysis:
Evidence for Photolytic Production of Cyclic-N3, Journal of Physical Chemistry A, 107, p.
10608-10614.2003
MAIER, G., PFRIEM, S., SCHFER, U. & MATUSCH, R.: Angewandte Chemie, International
Edition, England, 17, p. 520.1978
STMARK, H., LAUNILA, O., WALLIN, S. & TRYMAN, R.: On the possibility of detecting
tetraazaterahedrane (N4) in liquid or solid nitrogen by Fourier transform Raman
spectroscopy, Journal of Raman Spectroscopy, 32, p. 195-199.2001
STMARK, H., BITTEREROV, M., BRINCK, T. et al. (2002) The N4 project. Annual report for
the year 2001, pp. 69 (Swedish Defence Research Agency).
STMARK, H., WALLIN, S., HORE, N. & LAUNILA, O.: Raman spectra of P4 at low
temperatures, Journal of Chemical Physics, 119, p. 5918-5922.2003
BITTEREROV, M., BRINCK, T. & STMARK, H.: Theoretecal study of the singlet
electronically excited states of N4, Chemical Physics Letters, 340, p. 597-603.2001
BITTEREROV, M., STMARK, H. & BRINCK, T.: Ab initio study of the ground state and the
first excited state of the rectangular (D2h) N4 molecule, Chemical Physics Letters, 347, p.
220-228.2001
HANTZSCH, A.: Ber. d. Chem. Ges., 36, p. 2056.1903
HUISGEN, R. & UGI, I.: Zur Lsung eines klassischen Problems der organischen StickstoffChemie, Angewandte Chemie, 68, p. 705-706.1956
NGUYEN, M. T., MCGINN, M. A., HEGARTY, A. F. & ELGURO, J.: Can The Pentazole Anion
(N5-) be Isolated and/or Trapped in Metal Complexes, Polyhedron, 4, p. 1721-1726.1985
GLUKHOVTSEV, M. N., SCHLEYER, P. V. R. & MAERKER, C.: Pentaaza- and
Pentaphosphacyclopentadienide Anions and Their Lithium and Sodium Derivatives:
Structures and Stabilities, Journal of Physical Chemistry, 97, p. 8200-8206.1993
BURKE, L. A., BUTLER, R. N. & STEPHENS, J. C.: Theoretical characterization of pentazole
anion with metal counter ions. Calculated and experimetnal 15N shifts of aryldiazonium, azide and -pentazole systems, Journal of the Chemical Society, Faraday Transactions 2, p.
1679-1684.2001
LEIN, M., FRUNZKE, J., TIMOSHKIN, A. & FRENKING, G.: Iron Bispentazole Fe(5-N5)2, a
Theoretically Predicted High-Energy Compound: Structure, Bonding Analysis, Metal Henric stmark
249
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
250
Czech Republic
2006
[Content]
Ligand Bond Strength and a Comparison with the Isoelectronic Ferrocene, Chemistry - A
European Journal, 7, p. 4155-4163.2001
STRAKA, M. & PYYKK, P.: One Metal and Forty Nitrogens. Ab Initio Predictions for
Possible New High-Energy Pentazolides, Inorganic Chemistry, 42, p. 8241-8249.2003
TSIPIS, A. C. & CHAVIARA, A. T.: Structure, Energetics, and Bonding of First Row
Transition Metal Pentazolato Complexes: A DFT Study, Inorganic Chemistry, 43, p. 12731286.2004
CARLQVIST, P., STMARK, H. & BRINCK, T.: Computational Study of the Amination of
Halobenzenes and Phenylpentazole. A Viable Route to Isolate the Pentazolate Anion?
Journal of Organic Chemistry, 69, p. 3222-3225.2004
STMARK, H., GOEDE, P., WALLIN, S. et al. (2004) Development of energetic materials over
the time span 2004-2025, Special forecast (Stockholm, Swedish Defence Research Agency,
FOI).
FAU, S., WILSON, K. J. & BARTLETT, R. J.: On the Stability of N5+N5-, Journal of Physical
Chemistry, 106, p. 4639-4644.2002
FAU, S., WILSON, K. J. & BARTLETT, R. J.: Correction: On the Stability of N5+N5-, Journal
of Physical Chemistry, 108, p. 236.2004
EVANGELISTI, S. & LEININGER, T.: Ionic nitrogen clusters, Journal of Molecular Structure
(Theochem), 621, p. 43-50.2003
BEMM, U. & STMARK, H.: Structure of 1,1-diamino-2,2-dinitroethylene: A novel Energetic
Material with Infinite Layers in Two Dimensions, Acta Crystalography, C54, p. 19971999.1998
DIXON, D. A., FELLER, D., CHRISTE, K. O. et al.: Enthalpies of Formation of Gas-Phase
N3,N3-,N5+, and N5- from Ab Initio Molecular Orbital Theory, Stability Predictions for N5+N3and N5+N5-, and Experimental Evidence for the Instability of N5+N3-, Journal of the American
Chemical Society, 126, p. 834-843.2004
GAGLIARDI, L., ORLANDI, G., EVANGELISTI, S. & ROOS, B. O.: A theoretical study of the
nitrogen clusters formed from the ions N3-, N5+ and N5-, Journal of Chemical Physics, 114, p.
10733-10737.2001
MAILHIOT, C., YANG, L. H. & MCMAHAN, A. K.: Polymeric nitrogen, Physical Review B,
46, p. 14419-14435.1992
EREMETS, M. I., HEMLEY, R. J., MAO, H.-K. & GREGORYANZ, E.: Semiconducting nonmolecular nitrogen up to 240 GPa and its low-pressure stability, Nature, 411, p. 170174.2001
MCMAHAN, A. & LORENZANA, H. E. (1994) Polymeric Nitrogen: A potential compound to
store energy Energy & Technology Review, LLNL periodical, pp. 10-11.
EREMETS, M. I., ET AL: Single-bonded cubic form of nitrogen, Nature materials, 3, p. 558562.2004
BARBEE III, T. W.: Metastability of atomic phases of nitrogen, Physical Review B, 48, p.
9327-9330.1993
Czech Republic
2006
[Content]
AEROBIC BIODEGRADATION OF
MONONITROTOLUENES IN DIFFERENT REACTOR
CONFIGURATION
Jan Paca*1, Martin Halecky1, Rakesh Bajpai**
*
Department of Fermentation Chemistry and Bioengineering, Institute of Chemical

Technology, Prague, Technicka 5, 166 28, Czech Republic
Tel: +420-2-24353785, Fax: +420-2-24355051, Email: Jan.Paca@vscht.cz
**
Department of Chemical Engineering, University of Missouri, Columbia, MO 65211, USA
Tel: (573) 882 3708, Fax: (573) 884 4940, Email: bajpair@missouri.edu
1
Presenting and corresponding author
Abstract:
Degradations of nitrotoluenes, individually and in a mixture, were carried out in batch and
continuous aerobic reactors by a defined mixed microbial culture. The degradation rates and
efficiencies of the isomers were evaluated in batch and continuous reactors. The results
demonstrated that all the three-mononitrotoluene isomers were degraded simultaneously and
completely in presence of excess oxygen. The nitro group position on the benzene ring
influenced the degradation rates of the individual MNT isomers in batch systems. In the
continuous biofilm reactor with a sufficient biocatalyst quantity, quality and excessive
oxygen supply rate, the degradation rates of the mononitrotoluenes were almost identical as
long as the compounds were present individually and their loading did not exceed the
capacity of the catabolic master reaction. The microbial composition of the biofilm changed
qualitatively and quantitatively during long-term continuous operation under aerobic nonaseptic conditions. This complex investigation resulted in data that can be applied for the
scale-up procedure for field experiments.
Keywords: Degradation rate and efficiency, individual and mixtures of MNT, interactions in
catabolism, biofilm composition and surface structure.
1. INTRODUCTION
Explosives and other nitrated compounds are widely distributed environmental
contaminants [1]. Nitrophenols and nitrotoluenes are used extensively as raw materials and
intermediates in industry, especially in ammunition plants.
Extensive research has focused on the microbial removal of nitroaromatic compounds
in batch [3] and continuous reactors [4]. Although nitrotoluenes are not easily
biodegradable, several microorganisms with ability to use nitrotoluenes as the sole carbon,
energy and nitrogen sources have been reported [5]. Biodegradation of nitrotoluenes has been
demonstrated under both aerobic and anaerobic conditions [3].
[2]
The objective of this work was to study degradation of the MNTs individually and in a
mixture in batch and continuous reactor systems. The results of biodegradation by a mixed
1
Jan Pca, Martin Halecky, Rakesh Bajpai
251
Czech Republic
2006
[Content]
culture are reported here. The mixed culture was isolated in our laboratory from
contaminated soil from an ammunition plant (Explosia Co., Pardubice, Czech Rep.).
Analyses of the microorganisms in the mixed culture in the inocula and in biofilm after four
months of continuous operation are presented. Observation of the packing material and
biofilm surfaces from a packed bed reactor (PBR) were carried out using an electron
microscope during a long-term continuous experiment and these are also reported here.
2. MATERIALS AND METHODS

2.1. Microorganisms
Soil contaminated with a mixture of mononitrotoluenes, 2,4-DNT and 2,6-DNT was
collected from a Czech Ammunition plant in Pardubice, Czech Republic. A mixed culture
was enriched from this contaminated soil and several pure cultures capable of growing on a
mixture of mono- and dinitrotoluenes were isolated from this mixed culture.
2.2. Media, Culture Conditions, and Reactor Systems

A basal salt medium (BSM) containing 3.40 g KH2PO4,
4.30 g K2HPO4, 0.80 g KNO3, 0.34 g MgCl2.6H2O, and 1 mL
trace element solution per liter was used for preparation of
inocula and for the degradation experiments. Batch degradation
tests were performed in triplicate in 500 mL Erlenmeyer flasks
with working volumes of 100 mL at 26 C in a rotary shaker at
120 rpm. All the flasks for batch experiments were wrapped with
an aluminum foil in order to avoid any photocatalytic (abiotic)
degradation. All the experiments were conducted in triplicate.
For continuous degradation studies, a packed bed reactor
(PBR) with an internal diameter of 22 mm and a bed height of
240 mm was used (Figure 1). The cooling jacket of the reactor
was wrapped in an aluminum foil to prevent light exposure. The
packing material was expanded slate with a particle size 4 to 8
mm. The selected defined mixed culture (Table 1) was
immobilized on the packing material. The reactor was operated in
cocurrent up-flow mode. Inlet pH of the medium was 7.2.
Concentrations of individual MNTs in the feed stream to the
packed bed reactor were 8, 15, 20, and 30 mg.L-1.
Figure 1. Schematic of PBR. 1 packing, 2 medium influent, 3 medium effluent, 4 air
inlet, 5 air outlet, 6 tempering jacket, 7 combined electrode for dissolved oxygen
concentration and temperature, 8 sampling port.
2.3. Analytical methods

Sample analyses were conducted by HPLC (System DeltaChrom, Watrex Prague Ltd.,
Prague, Czech Republic) using WATREX column (250x4 mm Nucleosil, 120-5 C18,
Watrex Praha, s.r.o. Czech Republic), a mobile phase of methanol/water (50:50) at 30 C
and a flow rate of 1 mL min-1, and measuring absorbance with a diode array detector (Model
UV 6000 LP, Thermo Separation Products Inc., San Jose, CA, USA) at 230 and 238 nm.
Under these conditions, retention times of 2-NT, 3-NT and 4-NT were 22.2, 26.2, and 23.7
min.
pH of the medium was measured at the end of batch degradations and in the outflow
252
Aerobic Biodegradation of Mononitrotoluenes in Different Reactor Configuration
Czech Republic
2006
[Content]
from the PBR using a pH meter, model MFD 77 (INSA Corp., Czech Rep.).
An oxygen probe (Type CSOT44LL, INSA Corp., Czech Rep.) was used to measure
dissolved oxygen concentration (DOC) in the outflow from PBR. The data were collected
and stored in a computer. DOC was maintained above 5 mg.L-1 in PBR effluent by
manipulating airflow rate.
2.4. Microbial analyses

Identification of the individual strains in free cells or in biofilm was performed using
several biochemical tests. Fungi were identified by their morphological characteristics
especially by the shape of their fructification organs. The cells of biofilm, immobilized on
expanded slate, were suspended in BSM by vigorous mixing before any microbial analysis.
Table 1. Identification of the inoculum strains.
G- bacteria
Pseudomonas putida A1
Pseudomonas veronii B1
Chryseobacterium sp. D1
Stenotrophomonas maltophilia D2
G+ bacterium
Paenibacillus glucanolyticus D1/B
2.5. Microscopy
Samples of the packing material with and without biofilm were used for scanning
electron microscopy (SEM). The packing material with biofilm was taken from a depth of 2
cm from the top of PBR after 4 months of operation.
2.6. Calculations
Degradation rate (qi) of individual nitrotoluenes (NT) in a batch process:
S0 S
t
Where, S final NT concentration (mgL-1); S0 starting NT concentration; t time
of the batch process (days).
qi =
The degradation rate of NTs in the packed bed reactor
F ( Sin S out )
VL
Where, qi degradation rate of the individual NT; Sin NT concentration in the
medium influent (mgL-1); Sout NT concentration in the medium effluent (mgL-1); F flow
rate of the medium (mLday-1); VL reactor liquid volume (mL). The removal efficiencies
(RE) of individual NTs, the overall degradation rate of the NT mixture (qw), and hydraulic
retention time (HRT, min) were calculated as
S
RE = (1 out ) 100
Sin
qw = qi
qi =
253
Czech Republic
2006
[Content]
VL
F
Where F (mLmin-1) is the volumetric flow rate of fluid through PBR, and VL is the
liquid volume in the reactor (mL).
HRT =
Standard deviations of the concentrations in batch experiments were calculated from

parallel replicates and these are shown as bars in Figures 2-3. Degradation rates were
calculated from the experiments using Microsoft Excel statistic program and these exceeded
the reliability limit of 0.9.
3. RESULTS AND DISCUSSION
3.1. Degradation in a Batch System

The mixed culture was able to utilize the mononitrotoluenes (MNTs) as sole sources of
carbon and energy (Figure 2). From Figure 3 it is clear that at least 2-NT and 3-NT are
degraded simultaneously when present together in solution. 4-NT was still degraded the
fastest.
Another important finding throughout all the degradation experiments was that no
dead-end intermediates accumulated in the medium. Since all the analyses were carried out
in UV range using a diode array detector, all of the MNTs were considered degraded to
aliphatic compounds devoid of any unsaturated bonds.
Since a buffered medium was used, pH changes during the degradation of MNTs
individually and in a mixture were very low. The C/N/P ratio in experiments where MNTs
were the only C-, N-, and energy source was about 1/0.17/73.0.
3
2
1
0
t (days)
Figure 2. Degradation of the individual

NTs without any other C-,
energy, and N-sources in the
medium. () 2-NT; () 3-NT;
(); 4-NT.
254
-1
S (mg.L )
-1
S (mg.L )
t (days)
Figure 3. Courses of the individual NT

concentrations during a degradation
of the NTs mixture (without any
other C-, energy, and N-sources in
the medium). () 2-NT; () 3-NT;
(); 4-NT.
Czech Republic
2006
[Content]
Differences in the degradation rate of MNTs when present individually and in the
mixture are shown in Table 2. Taking into account the statistical error from the parallel
experiments, it is obvious that regardless to the S0 value, the presence of 4-NT and 2-NT in
the medium strongly suppressed the degradation rate of 3-NT. The degradation rates of 4-NT
and 2-NT remained the same as those reached when present individually in the medium.
From this finding it can be concluded that at such low starting NT concentrations the rates of
the master reaction of catabolism of both 4-NT and 2-NT are not affected by the presence of
the other MNTs in the medium. This behavior is also probably influenced with a large
excess of oxygen in the medium (DOC >5 mgL-1 in the outflow).
Table 2. Differences in the uptake rate of NTs when present individually and in a mixture.
Initial Concentration,
mg.L-1
5.0
4-NT
3-NT
2-NT
q (mg. L-1. day-1)

Individual NTs
Mixture of NTs
36.8 5
38.9 7
12.5 1
4.4 0.5
3.7 0.5
3.6 0.5
3.2. Continuous Degradation in a Packed Bed Reactor
-1
RE (%)
-1
q (mg.L .day )
Figure 4 shows the loading characteristics of individual MNTs in the PBR at a constant
HRT of 88 min. Increasing MNT concentrations in the influent changed the loading. 100%
degradation efficiency for 4-NT, and 3-NT was maintained at inflow concentrations of up to
22 and 20 mgL-1, respectively; the corresponding removal rates were 370 and 320 mg.L1.
day-1. The maximum removal efficiency of 2-NT was above 95 % but never 100% up to
feed concentration of 20 mgL-1 at this HRT. Above these threshold feed concentrations, the
removal efficiencies decreased with increasing concentration (4-NT and 2-NT dropped to
RE of 85 % and that of 3-NT to 95 %) in the experiment. In the feed concentration range
studied, the degradation rate continued to increase with feed concentration; the highest
degradation rates for 4-NT and 2-NT
100
800
was 460 mgL-1day-1 and it was 400
700
mgL-1day-1 for 3-NT. pH of the medium
80
decreased from 7.20 in the inflow to
600
7.03 in outflow with 3-NT and 2-NT;
500
60
the outflow pH remained above 7.15 for
400
4-NT.
40
300
200
20
100
0
0
0
10
4-NT (RE)
4-NT (q)
15
20
-1
25
S (mg.L )
3-NT (RE)
3-NT (q)
30
35
40
2-NT (RE)
2-NT (qs)
Figure 4. Comparison of loading characteristics

for the individual NTs in PBR. () 2-NT; ()
3-NT; () 4-NT; RE empty points; q full
points.
5
Measurements of the dissolved

oxygen concentration (DOC) in the
effluent showed that the airflow rate
had to be increased in order to keep
DOC above 5 mgL-1 during the loading
tests. This suggested an increase in
oxygen uptake rate of cells with
increased loading of MNTs.
This
observation supports assertion that the
MNTs were perhaps mineralized, rather
than just transformed into some
incompletely oxidized intermediates.
255

100
700
85
400
80
300
75
200
70
100
65
0
0
10
15
20
25
100
700
600
90
500
-1
85
400
-1
qw (mg.L .day )
95
RE (%)
-1
500
-1
90
qw (mg.L .day )
600
80
300
75
200
70
100
65
0
0
10
15
20
25
100
700
90
500
-1
600
85
400
-1
95
80
300
75
200
70
100
65
qw (mg.L .day )
RE (%)
95
RE (%)
Czech Republic
0
0
RE 4-NT
10
RE 3-NT
15
20
RE 2-NT
25
qw
Figure 5. Comparison of NT mixtures

degradations in PBR. RE: () 2-NT; () 3NT; () 4-NT; () qw. (a) Loading by 4NT; Sin = 8 mgL-1 of 3-NT and 2-NT each.
(b) Loading by 2-NT; Sin = 8 mgL-1 of 4NT and 3-NT each. (c) Loading by 3-NT;
Sin = 8 mgL-1 of 4-NT and 2-NT each
256
2006
[Content]
Due to changes in the gas hold-up,

it was necessary to correct the
medium flow rate to maintain
HRT constant.
A second set of experiments
focused on loading tests with a
mixture of nitrotoluenes in the
medium, again at the constant
HRT of 88 minutes. The loading
of 4-NT was changed by changing
its inflow concentration in a
medium containing 8 mgL-1 of 2NT and 3-NT. The results of
degradation rate (qw) of 4-NT and
removal efficiencies of all the
MNTs are shown in Fig. 5. The
removal efficiency of 4-NT was
100% in the entire loading range
tested (up 20 mgL-1). Nevertheless,
the removal efficiencies of 2-NT
and 3-NT did not reach 100% even
at the lowest 4-NT concentration
in the medium ( 8 mgL-1). All the
MNTs
were
degraded
simultaneously with no apparent
inhibition of 4-NT degradation by
either of the other MNTs.
Fig. 5 shows the same
parameters for the loading by 2NT in presence of 3-NT and 4-NT
in the packed bed reactor. Under
these conditions also, 100%
removal efficiency of 4-NT was
achieved in the entire loading
range. Although the RE of 3-NT
was 100%, it was similar to the
values reported in Figure 8a with
variable loading of 4-NT. For 2NT, the removal efficiency
dropped with increase in its
loading.
When loading of 3-NT was
changed at constant concentrations
of the other MNTS, the removal
efficiency of 2-NT and 3-NT was
strongly and negatively impacted
(Figure 5). At feed concentration
of 20 mg.L-1, the removal
efficiency of 4-NT also started to
drop.
Czech Republic
2006
[Content]
In presence of mixture of MNTs, simultaneous degradation of all the MNTs was

confirmed in the continuously operated reactor too and no dead-end intermediates were
observed in the medium. pH of the medium remained unchanged from inflow to outflow in
these experiments.
3.3. Microbial Analyses

An identification of the strains used for the selected mixed culture inoculum is shown
in Tab. 1. All the selected strains except Stenotrophomonas maltophilia D2 were able to
grow on a mixture of MNTs (10 mgL-1 each) as the sole C-, N-, and energy source. The
Stenotrophomonas maltophilia D2 was involved into the mixed culture as it was found in
almost all mixed cultures degrading a very broad spectrum of aromatic hydrocarbons.
The same microbial identification tests were carried out with resuspended cells taken
from the biofilm on the packing particles after four months of PBR operation. The results are
shown in Table 3. A comparison of Tables 1 and 3 shows that Pseudomonas putida A1 was
the only organism able to survive the strong selection pressure of various loading modes for
over four months of continuous operation and sustained degradation of MNTs.
Table 3. Identification of the strains in biofilm after four months of MNT degradation.
G- bacteria
Pseudomonas putida A1
Unidentified rods PCN 2
Kluyvera cryocrescens PCN 4
Sphingomonas sp. PCN 3
Fungi
Penicillium albicans
Aspergillus oryzae
Since the reactor operation was not aseptic, three other gram-negative bacteria and two
fungi were also found in the reactor. These supposedly were also primary MNT degraders,
i.e. they were able to grow on a mixture of MNTs as the sole C-, N-, and energy source.
To get a better idea of the packing material and the biofilm surface, scanning electron
microscopy was employed. Electron micrograph of a particle with the microbial film after
four months is shown in Figure 6. The biofilm surface is almost homogenous and the
bacterial cells appear to be trapped in some extracellular polymeric material. The cell
attachment to the packing surface is rather strong.
257
Czech Republic
2006
[Content]
Figure 6. SEM of the biofilm surface taken after four months of the reactor performance.
4. CONCLUSIONS
Much higher entire degradation rates of NT mixtures in PBR comparing with that one
from the batch process result from different catalyst composition and concentration in PBR.
Degrading various MNT mixtures in PBR, the metabolic interactions have been
demonstrated (cf Figures 5 and 3). Under conditions of the 4-NT loading the removal rates
of both 2-NT and 3-NT were not affected but their RE was not 100%. The total 4-NT
mineralization follows from a high oxygen supply rate necessary to maintain DOC > 5 mg.L1
and from no presence of any intermediate in the outflow medium. This finding coincides
well with that of Walia et al. [3].
With respect to the fact that 2-NT and 4-NT are precursors of DNT and are significant
contaminants at TNT manufacturing facilities while 3-NT is found only in trace amounts the
RE and degradation activities, that can be reached during NT mixture degradations, could be
high efficient (cf Figure 5).
Optimization of NT mixture degradations in the production wastewater could be
achieved using aerated continuously operated bioreactors with an immobilized enrichment
culture on a convenient packing material. In order to reach the complete oxidation
(mineralization) of NT mixtures from the influent wastewater, the important performance
parameters affecting the process optimization are organic load (OL), hydraulic retention time
(HRT), and dissolved oxygen concentration (DOC).
Tha data can be applied for the scale-up procedure for field experiments.
258
Czech Republic
2006
[Content]
REFERENCES
[1] J. C. Spain: Introduction. In: Biodegradation of Nitroaromatic Compounds and
Explosives, pp. 1-5 ( Spain, J. C., Hughes, J. B., Knackmuss, H. J., Eds.), Boca Raton,
FL, Lewis Publishers, 2000
[2] D. F. Carpenter, N. G. McCormick, J. H. Cornell, A. M. Kaplan: Microbial
transformation of 14C-labeled 2,4,6-trinitrotoluene in an activated-sludge system, Appl.
and Environ. Microbiol., 35, pp. 949-954, 1978
[3] S. K. Walia, S. Ali-Sadat, G. R. Chaudry: Influence of nitro group position on
biotransformation of nitrotoluenes in Pseudomonas strain OU83, Pesticide Biochem.
Physiol., 76, pp. 73-81, 2003
[4] C. M. Peres, B. Van Aken, H. Naveau, S. N. Agathos: Continuous degradation of
mixtures of 4-nitrobenzoate and 4-aminobenzoate by immobilized cells of Burkholderia
cepacia strain PB4, Appl. Microbiol. Biotechnol., 52, pp. 440-445, 1999
[5] B. E. Haigler, J. C. Spain: Biodegradation of 4-nitrotoluene by Pseudomonas sp. strain
4NT, Appl. Environ. Microbiol., 59, pp. 2239-2243, 1993
Acknowledgment:
The study was financially supported by the Czech Science Foundation Project
104/04/0686.
259
Czech Republic
2006
[Content]
POLYFUNCTIONAL N-NITRAMINES STRUCTURE

DIFFERENTIATION AS A BASIS FOR SIMULATION OF THEIR
DECOMPOSITION MECHANISM
Tatyana V. Petukhova1, Victor P. Ivshin1,

Vyacheslav L. Korolev2, and Tatyana S. Pivina2
1
Mari State University, Lenin square 1, Yoshkar-Ola 424000,

Mari El Republic, RUSSIA, E-mail: organic@marsu.ru
Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences,

Leninsky Prospect 47, Moscow 119991, RUSSIA
ABSTRACT
The structural classification of different chemical classes of N-nitramines has been
developed and the differentiation of their types throughout the functional surroundings
has been proposed. Basing on it and using up-to-date experimental data on their
thermal decomposition mechanisms the set of generators for chemical reactions that are
typical during nitramines decomposition process has been elaborated. The original
schemes for different types of N-nitramines thermolysis reactions have been designed
and the initial stage activation energy of it has been calculated by using the B3LYP/631G* level of density functional theory. As the result the most favorable pathways of
compounds decomposition have been displayed.
The suggested methodology for thermochemical processes simulation can be used for
a set of practical problems solution, including the investigations of the mechanisms of
decomposition, ignition, combustion, and detonation of energetic materials as well as
for structure-properties relationships study and for the search of prospect highenergy substances structure.
INTRODUCTION
The systematical investigations of structure and properties relationships for energetic
materials (aliphatic and aromatic nitrocompounds, nitramines and nitroethers, heterocycles
and others) are in progress from the end of the fifties last century. Both individual
compounds and the ascertainment of common features of structure-property relationships
have been investigated. N-nitramines take the first place among energetic materials (EM)
and the basic property of them is the thermostability concerned with the processes of
compounds decomposition.
260
Polyfunctional N-nitramines Structure Differentiation as a Basis for Simulation . . .
Czech Republic
2006
[Content]
Although a large number of experimental investigations concerning the kinetics of

(EM) thermolysis were performed, the problem of possible pathways prediction of the
decomposition mechanisms remains quite topical. The problem is also of interest, from the
viewpoint of assessment of the capability of potentially energetic compounds, which have
not yet been synthesized. Meanwhile, theoretical methods for the solution of such a problem
are very insufficient at present. In addition, they mostly deal with the interpretation of
experimental data rather than with predictions. In this connection, we have formulated [1,2]
generalized problem: computer generation of possible homolysis pathways for a given
compound on the basis of its structural formula, as well as correction of these pathways on
the basis of some formal rules, which are specified in accordance with the experimental data.
The framed set of rules allows one to differentiate of the thermal decomposition
pathways of nitramines according to its molecular structure.
During energetic materials thermal decomposition process in gas phase thermolysis
reactions can pass practically throughout all known mechanisms: molecular dissociation with
free radicals formation, elimination, rearrangement with the subsequent decay, disclosing of
a cycle, etc. The prevalence one or other way of transformation during decomposition
process depends on molecular structure, first of all [3].
Against amine nitrogen surroundings there are possible to mark out four Nnitrocompounds groups:
..
O2N
..
CH2CH2R
..
O2N
CH2CH2R
CH2X
O2N
II
..
.. X
.. C X
O2N
III
IV
The specimens of the first group (I), where individed pair of amine nitrogen does not
take part in conjugation or other nonvalent orbital interactions [4] are characterized by the
highest thermal stability. Their decomposition begins with NNO2 bond homolysis.
In contrast of the first group of N-nitrocompounds, where are not the effects of
undivided electronic pair, the effects of heminal interaction can be came out in substances of
the group (II). This is a reason of the wide ranged variability in thermal constancy of the
second group members. In the case of symmetric heminal systems of the group (II), when
heteroatoms >N and X are approximately equal in basicity and inductive effect, there is
practically no difference in action with compounds of the group (I), since effects of heminal
interaction are multiply degenerated. In the case of assimetric systems (II) effects of heminal
interaction are fairly appeared. It deeply reduces thermal stability and changes the kinetics of
initial stage of thermolysis [5-7].
In compounds of the third class (III) the amine nitrogen of N-nitro group is directly
bonded with the unsaturated fragment. When there are no steric constraints for pN*C=X
conjunction, the sharp decrease of N-N-bond energy is observed that facilitate its hemolytic
scission or 1,3-sigmatropic shift of nitro group from nitrogen to X-atom.
The group (IV) has been formed by N-nitrocompounds containing amine nitrogen
directly connected with heteroatom. The close contact of heterosubstitute with N-nitro group
is accompanied by the weakening of bonds in N-nitro fragment according to inductive
scheme and scheme of exchange interactions [8,9].
Tatyana Petukhova, Victor Ivshin, Vyacheslav Korolev, Tatyana Pivina
261
Czech Republic
2006
[Content]
Basic scheme of thermal induced transformations of N-nitro compounds are presented at

the Fig. 1.
R' = Alk
R' ..
N
R
R' = CH2X
R N CH2+ X -
(I)
R'
NO2
RN(NO2)CH2
R' =
C C
C X
N2O CH2O
RNH C C NO2
RN
(III)
(IV)
RX
NO2
(II)
R
NO2
C C NO2
R' = H, C(O)R", C(O)OR",

CH=CH2, SiR3
RN
..
R' = Hal, N R
R
R' ..
N
R
OR'
O-
ROR' + N2O
NO2
Fig. 1. Thermal induced transformations of N-nitro compounds.
METHODS
Mechanism of chemical reactions during the decomposition process we presented as a
topologic structure, which is a function of source substances, intermediate and final
products. It can be invariant for any concrete chemical system. To define a mechanism of
decomposition we used a graph theory and formulated the general approach to the computerassisted modeling of the mechanism of homolytical reactions for nitrocompounds. A set of
chemical transformation generators based on experimental data has been formulated to
follow up the more probable decomposition reactions for compounds of different chemical
classes.
Screening of the most advantageous pathways from the energy standpoint were carried
out by calculations of the initial stage activation energy (B3LYP/6-31G* [11] level of
density functional theory [10]). Calculations were executed using GAUSSIAN 98 /12/ in the
Computational Centre of Zelinsky Institute of Organic Chemistry, Russian Academy of
Sciences.
As the result the most favorable pathways of compounds decomposition have been
displayed.
262
Czech Republic
2006
[Content]

As a model compound of the first group (I) we studied N, N-dimethylnitramine. Fig.2
represents four possible initial stages of decomposition of corresponding substance: (1) N-N
bond homolysis, (2) HNO2 elimination, (3) nitro-nitrite rearrangement, and (4) homolysis of
the C-N bond. As a result of computer generation of probable N, N-dimethylnitramine
thermolysis pathways we have got a whole series of formally probable reactions. The data
obtained can be used to understand the mechanism of thermal decomposition of
corresponding compound as well as to clarify the formation of nitrogen, protoxide of
nitrogen, H2O, carbon monoxide and carbonic acid as the final decomposition products.
CH3
CH3
CH2
CH2
NO2
+R
NO2
CH3
CH3
+M
+ONO
ONO
CH3
O.
CH3
+NO
N O CH3
+M
+R
+R
CHO HCN
+ONO
+R
. CH2ONO
CH3NO
+M
CH3OH HNO NO CH2O
+ONO
CH3NO2
CH3
CH3
+M
CH4
.
CH3
N OH
CH3
CH3
+NO
+R
+R
NO
+M
N CH3 N N OH + CH2O
+M
H2O HNO
CH2 N OH
CO CO2 + HNO
CO2
.
OH
N OH .
CH3 H2C
N2
CH3NO HCN
-NO
HONO
NO CH2 N CH3
+HONO
CH2
O
+M
+M
CH3
+M
. .
N2O CH2O H C N ONO CH3O . .NO
2
.O
CH3 N CH3
CH3
CH3ONO
CH4
NO2
+HONO
CH3OH
+R
+ONO
+R
H2O HCN O C N N OH
CH3O CO CO2 + HNO

+R
N2
H2CO2 CH2O
+NO
CHO
-NO
OH + N2O
+M
H2 O
CO2
Fig. 2. The fragment of N, N- dimethylnitramine thermal decomposition

scheme.
The calculated activation energy () of homolytic reaction of NN bond cleavage is
180.7 kJ/mol, compared to 336.0 kJ/mol for the CN bond cleavage and to 282.2 kJ/mol for
nitro-nitrite rearrangement reactions. In this connection, one can conclude that two lastnamed processes are unlikely in comparison with the N-N bond homolysis reaction in N, Ndimethyl nitramine. The calculated activation energy () of the reaction of HNO2
elimination is 190.2 kJ/mol (197.8 kJ/mol [13]) that is only for 9.5 kJ/mol higher than for the
homolytic cleavage of NN bond. Therefore the HNO2 elimination can be consider as the
alternative pathway of decomposition for secondary nitramines.
The simulation of the second group (II) nitramines was performed by the example of Nmethyl-N-chlorine methylnitramine. Fig. 3 illustrates the reaction scheme of the compound
decomposition process of the compound.
Here and at other schemes of decomposition processes the known experimentally reactions and products are
marked by the bold type.
263

Cl CH2
N ONO
CH3
Cl CH2
CH3
CH2Cl
.CH2
Cl
N O
+M
+R
CH3
+M
+M
CH3 N
CH4
HNO
+M
+HONO
+M
CH3NO NO
+NO
+R
N2O + OH
CO
+M
H2O
+ONO
CO2 + HNO
+NO2 -NO
+M
+M
HONO
+R
CO CO 2 + HNO
+M
CO2
H2O
+M
NO
CHO
+M
+NO2 -NO
N2O + OH CO 2
+R
+R
OH
H2O HNO
+R
HCN HONO
+ONO
N2O + OH
CO CO2 + HNO OH
+NO +NO -NO

2
+M
H2O
+M
N 2O H2CO H2C N ONO HCl
+ONO
Cl
CH2 N NO 2
HNO
+NO
CH4
+R
NO
+M
N2 CH3 OH CHO
+M
ONO
CHO HCN
+R
N OH CH2O HCl OH
Cl CH2 N NO 2
CH3
CH2 Cl HONO N2O H2CO H2C N
[Content]
CH2 N NO 2 CH3Cl
+M
CH3Cl
CH3
Cl CH2 N CH3 ONO

N NO 2
.
HCl CH
2006
N NO 2
CH3
Cl CH2
Czech Republic
CO2
+M
H2O
+M
NO
+M
H2O HNO
+NO
N 2O + OH
+M
H2O
Fig. 3. The fragment of N-methyl-N-chlorinemethylnitramine decomposition

scheme.
Considerations of the presented scheme allow the conclusion that under the atom of
chlorine introduction into the molecule of N, N-dimethylnitramine the subsidiary channels
of thermal decomposition arise through the cyclic complex with the subsequent elimination
of methylchloride and homolytic cleavage of the CCl bond. These opportunities provide the
explanation for HCl and H3Cl presence among the final decomposition products (the origin
of which was not quite clear before).
The energy of the NN bond cleavage in N-methyl-N-chlorine methylnitramine is 180.7
kJ/mol, for the CN homolysis the energy is 361.1 kJ/mol, for the CCl cleavage bond
295.7 kJ/mol, and (Ea) for nitro-nitrite rearrangement is 211.6 kJ/mol. Unfortunately in our
calculations the transition state localization of the cyclic activated complex formation was
not a success:
CH3
O2N N
Cl
CH2
CH3Cl + CH2
N NO2
The experimental energy activation of the initial stage of N-methyl-N-chlorine methyl

nitramine thermal decomposition is 128.5 kJ/mol [15].
The reactionary scheme of thermal decomposition process for N-vinyl-N-methyl
nitramine, as one of the third type compounds representatives, is shown in Fig. 4.
Analysis of the results of decomposition processes simulation follows up that under
introducing an unsaturated fragment direct fixed with N-nitro group into a molecule, a set of
rearrangements may be observed which are absent in the case of (I) and (II) groups
compounds. At the same time, the possible thermo decompositions channels and resulting
intermediates increase in number.
264
CH2
CH
. ONO
N CH3
+M
+R
CH2 CH
+R
HONO
CH2
CH2 N
CH2
NO2
+HONO
CH2
CH2O CH2 CH N N OH OH
+M
CHO
+R
N2 CH3 CH O
H2O
+ONO
+R
+M
+R
HNO
HCN
+ONO
CH
CH CH2
CH
+M
CH3
+NO2 -NO
CO2
N2
CHO
+R
CO
+NO2 -NO
CO2
+ONO
+
CO2 HNO
CH3 NH CH CH NO2
NO2
CH3
N O
+
+M
CH2
CH
CH2
CH3 N
CH2 CH
N OH
CH3
CH4 CH2 CH .NO

CH2 HNO
_
O
N
O CH CH2
N OH
CH2 CH
+M
+M
+NO
CH
N O
CH3 CH2 CH N O CH2
.
OH + N2O O C N N OH CH2
CH4 CO CO2 +HNO

H2O
CH2
+M
CO2
N ONO
CH3
+HONO
+NO
CO CO2 +HNO CH3 CHO

+NO2 -NO
.
.NO . ONO CH2 N
[Content]
N NO2
CH2
CH3 N NO2
2006
CH
CH3
CH2
Czech Republic
CH3
+R
OH CH2
CH
CH2
OH
+M
H2O
OH + N2O +M
+M
H2O
H2O
+HONO
CH2O
CH2
N OH
CH3 CH O
N2
CHO
+R
CO
CH
+ONO
CO2 + HNO CH3 CHO
+NO2 -NO
CO2
+R
CH4
CO
+ONO
CO2 +HNO
+NO2 -NO
CO2
Fig. 4. The fragment of N-vinyl-N-methylnitramine decomposition scheme.

The calculated activation energy of the NN bond cleavage in N-vinyl-N-methyl
nitramine is 107.1 kJ/mol. That fits the assumption on the sharp NN bond energy decrease
for the (III) group compounds in comparison with the (I) group ones. The activation energy
of the unsaturated substitute NO shift is 226.2 kJ/mol. In case of CH3 elimination the
energy of CN bond homolysis is 271.8 kJ/mol and is 397.1 kJ/mol in case of vinyl radical
elimination. We failed to localize the nitro-nitrite rearrangement and the 1,3-sigmatropic
shift transition states. Though, the nitro-nitrite rearrangement is not likely since the molecule
has a plain configuration.
As a model of the fourth group (IV) compounds we selected N-methyl-N-chlorineamine.
Fig. 5 represents the scheme of its thermal decomposition process.
265

Cl
Czech Republic
2006
[Content]
N NO2
H3C
H3C
Cl
. ONO
+M
+R
H2C
Cl
Cl
HCl
+M
+R
HCN
H2O
N2O H2CO H2C N ONO Cl
HNO
O C N N OH N2O + OH
+R
+R
N2
+ONO
OH
+M
CO CO2 + HNO H2O

+NO2 -NO
CO 2
H2O
Cl
CH4 NO
+M
H3C
HNO
Cl
OH
N 2O HO
Cl
O
OCl
N O N O CH OCl
2
3
CH3 N O
CH4
+M
N2O + OH HCl
+M
+M
+M
HOCl
+ONO
H3C N
+M
O
+NO
HCN HONO
CO CO 2 + HNO OH
CO2
N ONO
H3C
+M
CHO
+M +NO2 -NO
CHO
+R
+R
+M
+M
Cl NH NO 2
NO 2 HCl
Cl
CH3
+M
NO
+NO
+HONO
CH2
N NO 2
Cl
+M
+R
HONO
N OH
H2C
+M
H3C N NO2 Cl
NO
+M
H2O HCl
+M
H2O
CH3
CH4
NO
+M
OH
+M
+M
HNO H2O
Cl
+M
HCl
+NO
H2O HNO
+NO
N 2O + OH
N 2O + OH
+M
H2O
+M
H2O
Fig. 5. The fragment of N-methyl-N-chlorinenitramine decomposition scheme.

As one can see the introducing of the heteroatom, directly fixed with the NO2 group into
the N-nitramine molecule, makes the competitive channel decomposition rising through the
C-X bond dissociation.
For the N-methyl-N-chlorinenitramine the activation energy values are: 105.8 kJ/mol
(for NN bond homolysis), 188.8 kJ/mol (for NCl bond homolysis), 302.2 kJ/mol (for CN
bond homolysis), and 233.8 kJ/mol (for the chlorine atom NO shift).
As a whole, the data obtained is evidence of the radical mechanism of N-methyl-Nchlorinenitramine decomposition through the N-N bond homolytic cleavage.
CONCLUSION
Thus, the structural classification of various chemical classes of N-nitramines has been
developed.
The results of thermolysis mechanisms simulations of the compounds have allowed to
predict the whole spectrum of formally possible reactions.
The activation energy calculations of decomposition reactions mark out the thermo
chemically most preferable ones.
As is shown, the change of the amine atom surroundings has an influence upon the
molecular structure that determines the modification of compounds thermostability and
makes for the appearance of subsidiary channels of decomposition.
266
Czech Republic
2006
[Content]
The authors are grateful to Dr. A.A. Porollo for the assistance in this work.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
A.A. POROLLO, D.E. LUSHNIKOV, T.S. PIVINA, V.P. IVSHIN, J. Mol. Struct.
(TEOCHEM), Vol. 391, p. 117-124, 1997.
A.A. POROLLO, D.E. LUSHNIKOV, T.S. PIVINA, V.P. IVSHIN, Izvestiya Akademii nauk,
10, p. 1869-1880,1999 (in Russian).
T.S. PIVINA, A.A. POROLLO, T.V. PETUKHOVA, and V.P. IVSHIN, In Proceedings of the
5th International Symposium on Special Topics in Chemical Propulsion (5-ISICP), Italy, p.
343-344, 2000
V.P. IVSHIN, O.A. Yashukova, V.A. Shlyapochnikov, Izvestiya Akademii nauk, 6, . 12951298, 1986.
F.I. DUBOVITSKY, B.L. KORSUNSKY, ADVANCES OF CHEMISTRY, Vol. 50, 1828-1871,
1981.
G.B. MANELIS, G.M. NAZIN, Yu.I. RUBTSOV, and V.A. STRUNIN: Thermal
decomposition and combustion of explosives and powders, Nauka, Moscow, 223 p., 1996.
Yu. SHU, B.L. KORSUNSKY, G.M. NAZIN, ADVANCES OF CHEMISTRY, 2004, Vol. 73, .
321-335, 2004.
M. GRAFF, C. JR GOTZMER, W.E. MCQUISTION J. Org. Chem., 10, . 3827-3830,
1967.
M. GRAFF, C. JR GOTZMER, W.E. MCQUISTION, J. Chem. Eng. Data, 14, 4, . 513514, 1969.
W. KOCH and M.C. HOLTHAUSEN: A Chemistss Guide to Density Functional Theory,
Wiley-VCH, Weinheim, 300 p., 2001.
T. KLARK: Computational chemistry (In Russian), Mir, Moscow, 383 p., 1990
Programme Complex Gaussian-98 of the Computer Centre for Chemical Investigations,
Russian Academy of Sciences.
M.A. JOHNSON, T.N. TRUONG, J. Phys. Chem., Vol. 103, p. 8840-8846, 1999.
B.L. KORSUNSKY, F.I. DUBOVITSKY, V.I. Losenova, Jurnal Physicheskoy Khimii (in
Russ.), 4, p.1159 1162, 1969.
267
Czech Republic
2006
[Content]
PROCESS HAZARD EVALUATION FOR EXPLOSIVES

MANUFACTURING USING TEMCLEV-EX PROCEDURES.
Tadeusz Piotrowski, Tomasz Saaciski, Micha Frczak
Instytut Przemysu Organicznego

Annopol 6 str. 03-236 Warszawa, Poland
(+48 22) 811-12-31 ext. 298, piotrowski@ipo.waw.pl
Abstract
The paper presents the rules of the process hazard indices formation: technological T,
protection TZAB and total process hazard ZP. The method of dividing the whole
installation into the process hazard classes KZP has been presented. The evaluation
carried out contains so called material part ensued from the dangerous properties of
the media used in process as well as from the process conditions, devices (apparatuses)
used and even work organization and safety management.
Key words:
Process safety, process hazard indices (material, technology,
protection), hazard evaluation and classification.
1.
INTRODUCTION
System called Temclev-Ex offers a rapid hazard assessment method for use in the
industry of explosives and pyrotechnics or in plant design / modernization. The indexes used
in it are primarily concerned with the fire and explosion problems. Toxicity is considered
only as an additional, partial factor. A brief description of the base Temclev System and its
modification Temclev-Ex has been presented at the 7-th Seminar two years ago. This
time we would like to present the most important part of the system, a method of the process
hazard indices and protection indices formation.
2.
EVALUATION OF PROCESS HAZARDS
The detailed work of hazard evaluation and classification is best done by a team, ideally
including personnel familiar with the technology used (plant and process), as well as a
member with experience of using the system details. As it was earlier presented the equation
describing a hazard level of a process unit ZP in Temclev-Ex for manufacturing and
processing explosives presents as follow:
ZPEx = [(ARw + BRz) SEx] [T/Tzab],
268
(1)
Process Hazard Evaluation for Explosives Manufacturing Using TEMCLEV-EX . . .
Czech Republic
2006
[Content]
where: A, B the notes of Rz,w parameters,

RW,Z indices describing explosives parameters according to Polish legislation.
T - technological index, SEx - specific material properties index, TZAB- protection index.
A description of so called material part and the first version of material (explosives)
ranking has been given in [1].
3.
PROCESS HAZARD (TECHNOLOGICAL) INDICES
An assumption has been made that processing a substance causes hazard increasing.
Dangerous properties of processed media may be activated in running process. It is because
of the stimuli acting on them at the process conditions like: pressure, temperature, flows,
atmosphere, etc. and apparatuses / devices which can generate strokes, friction, get catalytic /
inhibiting agents, ignition sources, static electricity, emergency states, corrosion and so on.
In the given equation (1) that role plays technological index T. It has been constructed using
numerical notes given for 4 groups of a process characteristics described by equation (2)
given below.
General characteristics of a process (operation) TO. In this group the information
about dangerous material, process/operation type, a way of acting, devices/apparatuses used
and detailed process/operation parameters have been taken into account.
Detailed characteristics of unit operations TF, containing a specific operation menu
for explosives manufacturing including such types of operations as: static (storage,
conditioning), in solvent (filtration, washing, crystallization), in bulk (supplying, emptying,
weighting), in cloud (milling, sieving, granulating, dust removal), pressing (different types,
rolling), in elevated temperature (heating, casting, melting), hand-made, mechanical working
and general unit operations as: transport, mixing, drying, mass transfer operations and others.
Chemical process characteristics TP. In this group all chemical processes have been
divided into 4 types in the respect of thermal effect, system homogeneity, catalytic agent
presence and reaction phase (gaseous, liquid, solid, mono and multi-phase).
Specific hazards characteristics TS. In this group such information have been taken
into account as: plant area, apparatuses height, potential emission sources, corrosion, static
electricity, extremely difficult processes, fire and/or explosion zone presence, specific low
requirements fulfilling and many others.
The groups 2 and 4 above have been changed a lot comparing to the basic Temclev
System because of the specific operation types used in explosives and pyrotechnics
production. The others have been only slightly changed.
T = TO + TF + TP + TS
(2)
The use of protective measures decreases process hazard level. According to equation
(1) resultant note is diminished TZAB note. The index TZAB is very important in the system. It
shows the areas where a hazard should be reduced and helps acting in that direction. In the
range of using protective measures, which may be technical, technological and
organizational, any selected unit should be analyzed also in 4 thematic groups [2]. They
2
269
Czech Republic
2006
[Content]
cover both: specific and general types of measures used in explosives industry for protecting
workers and processes.
Monitoring emergency states KS,
Failures preventive measures ZA,
Emergency systems and control SA,
Organization and safety management DO.
The index TZAB is computed using equation (3) as described above a product of 4
components:
TZAB = KS ZA SA DO
(3)
The components: KS, ZA, SA, DO are computed using equations presented in [3, 8].
The notes for components of equations (2, 3) may be taken from System Manual [9] or
computer program System_Ex, which has been elaborated in details for use in MS Windows
environment.
Some examples of the Evaluating Tables (1 5) used in Manual have been presented
below.
Table 1. Pressing
Unit Operation (physical) - Pressing
Rolling, pressing of product
Pressing in die and pushing out
Two-sided pressing
Operations in elevated temperature, with vibration, long lasting overpressure and/or,
repeatedly, with mould venting, other types every element additionally
Rating tII/1.6
3
4
5
+3
Table 2. . Mixing
Mixing Procedures
Without moving mechanical elements/devices:

Combustible and non-combustible substances
Combustible substances
Combustible substances and oxidizers
Explosives
In case using mechanical elements/devices like: paddle mixer, stirrer,
helical agitator perpetual etc. additionally
270
Rating
tII/1,6
1
2
5
8
+2
Czech Republic
2006
[Content]
Table 3. Specific conditions

Unit Operation (physical) Specific conditions
Hand-made operations like: packing, covering by plastic foil, a lot of

product forming, dosage etc.) additionally
Operations run in elevated temperatures like: melting, casting, heat
treatment, operations run under lower pressurefor every specific
condition additionally
Mechanical working (cutting, machining, milling, propellant/blowing
charge grinding)
Rating
tII/1.4
2
3
5
Table 4. Cut-off systems

Rating tZIII/1
Cut off system of media flow
Flow blocking outside the unit:
- cut-off valves on a flow-line (pipeline) remote operated from operating
1
room, long distance from unit area
Flow blocking inside the unit (installation):
- hand-operated cut-off valve on the work place
2
- operator activity in the operating room at the unit area
1
- the valves positioned automatically in safety positions
+2
- complete emergency system for process stoppage, flows cut-off and
5
vessels emptying activated with use of one switch or automatically after
exceeding a fixed limitation
Table 5. Process parameters control
Operation / process parameters control
Periodic
Continuous:
Under process operating computer supervising
Steering automatics, self-steering devices or remotely operated systems
meeting requirements of Polish law regulation MGPiPS (Dz. U. Nr163
poz.1577 z 2003 r.)
Manually from operating room
Manually by operator of a device on the work-place
When control includes more than 1 parameter, every next additionally
With possibility to start safety systems - additionally
Recording process parameters and control signals meeting requirements
of Polish law regulation MGPiPS (Dz. U. Nr163 poz.1577 z 2003 r.) additionally
4.
Rating tZ,I/1
0.5
3
2
1
0,5
+0,5
+1
+1
PROCESS UNIT CLASSIFICATION
Every selected unit may be classified to one of the five process hazard classes KZP as shown in
Table 6 below.
271
Czech Republic
2006
[Content]
Table 6. Division into KZP classes according to ZP index value.

Class KZP
5.
Hazard level
KZP1
KZP2:
2/III
2/II
2/I
Catastrophic (Very high)

High:
upper range
center range
lower range
KZP3:
3/III
3/II
3/I
Middle:
upper range
center range
lower range
KZP4:
4/II
4/I
KZP5:
5/II
5/I
Moderate:
upper range
lower range
Low:
upper range
lower range
ZP index values
ZP > 300
200 < ZP 300
270 < ZP 300
230 < ZP 270
200 < ZP 230
100 < ZP 200
170 < ZP 200
130 < ZP 170
100 < ZP 130
50 < ZP 100
75 < ZP 100
50 < ZP 75
0 < ZP 50
25 < ZP 50
0 < ZP 25
PROCESS RISK ASSESSMENT METHOD PRAM
PRAM is the Risk Score method modified into the process conditions. It is based on the
equation (4) shown below. The meaning of all indices and their digital values have been
detailed described in [10]. Process risk level Rp values for PRAM have been given in Table
7. Potential consequences of hazard level (failure) Sp values are connected with Temclev
System through KZP classes.
Rp = Sp x Ep x Pp
(4)
where:
Rp process risk level index,
Sp potential consequences of hazard level (failure) index,
Ep hazard exposure time (process realization type)
Pp hazard (failure) occurrence probability index
Table 7. Process risk level Rp values for PRAM
Rp index
values
Rp < 20
20 Rp < 70
70 Rp < 200
200 Rp < 400
Rp 400
272
Description of risk level and preventive/protective measures

needed
Very low (risk acceptable), no action required
Low (risk acceptable), a need of giving attention to eliminating
potentially accidental events and initial events
Essential, a need of rapid improvement of safety level and lowering
the risk value (risk temporary tolerable, ALARP area)
High, a need of immediate improvement of safety level and lowering
the risk value (risk temporary tolerable, ALARP area)
Very high, (risk not acceptable), consider to stop an operation or use
the immediate actions for significant decrease the risk value (at least to
ALARP area)
6.
Czech Republic
2006
[Content]
ALGORITHM OF PROCESS HAZARD EVALUATION
Fig. 1. Fig. 1. Process hazard evaluation algorithm.
273
7.
Czech Republic
2006
[Content]
PROCESS HAZARD AND RISK ASSESSMENT RESULTS
In the frames of this work 7 selected process units in 4 plants have been analyzed and
assessed using Temclev-Ex and PRAM methods and System-Ex software. Some of them the
authors presented below as the examples of elaborated assessment procedures.
7.1
Emulsion explosives. Cartridging unit.
Emulsion explosives are much safer materials than traditional explosives in

manufacturing process as well as in further use. Their every ingredient is not a real
explosive. They become explosives after sensibilization at the very end of production
process.
Their manufacturing (Fig.2, Fot.1) consist mainly on :
Making a mixture of oil and emulsifier,
Making a mixture of wax and paraffin,
Making a solution of oxidizers (ammonium nitrate and sodium nitrate in water),
Mixing everything together in a high-speed mixer adding aluminium powder, glass
microspheres and granulated ammonium nitrate.
A mixture made in such a way is then given into the head of the cartridging machine to
make blasting cartridges. The end of that is a cooling process, drying, labelling and packing.
The whole production process is controlled by a computer. Every disturbance of the process
causes automatically stoppage of the installation.
Fig. 2. Cartridging operation scheme.
274
Czech Republic
2006
[Content]
Fot.1. Cartridging of emulsion explosives using Filipack EX 1100.
7.2
Nitroglycerin explosives for mines. Cartridging unit.
Manufacturing of nitroglycerin explosives for mines Metanit S2H consists on mixing

their ingredients: ammonium nitrate, DNT, nitro-esters and sodium chloride in mixers, then
supplying the mass into the cartridging machines where the cartridging process runs
continuously.
The cartridging oparation itself consists on proportioning the mixture by volume into a
core made by paraffin paper, closing both ends by sticking and giving a ready charge onto
the rolls for packing. The process is fully automatic, runs using cartridging machine
Niepmann A63-IV (not modernized).
7.3
Manufacturing of ANFO type explosive. Mixing unit.
Manufacturing of ANFO type explosive is realized in a big rotary mixer (Fot.2).From

its one side a weighted portion of granulated ammonium nitrate is supplied and from its
opposite side an oil is injected by injection nozzle. After the time fixed in a process
instruction the product is ready and is automatically pour out into the uotlet tube from which
falls down into the multiwall paper bags (Fot.3).
Fot.2. Rotary mixer for ANFO type explosive.
Fot.3. Product packing.
275
Czech Republic
2006
[Content]
Risk assessment results for: emulsion explosives Emulinit 2 and nitroglycerin mining
explosive Metanit S2H cartridging and ANFO type explosive Saletrol manufacturing
have been shown in Table 8 below.
Table 8. Table 8. Comparison of risk assessment results for cartridging process
(two different explosive types, two different machines) and ANFO
manufacturing.Temclev-Ex and PRAM methods used.
Assessed
process unit
Emulsion
explosive
Emulinit 2
cartridging using
Filipack EX 1100
Nitroglycerin
mining explosive
Metanit S2H
cartridging using
Niepmann A63IV (before
modernization)
ANFO type
explosive
manufacturing
using rotary
mixer
8.
ZP value
and
KZP
Class
Sp index
value
Ep index
value
Pp index
value
Rp index
value
89,25
18
Very low
4/II
(risk acceptable)
162,08
126
Essential
3/II
(risk temporary
tolerable, ALARP
area)
85,9
4/II
18
Very low
(risk acceptable)
FINAL REMARKS AND CONCLUSIONS
The proposition of applying a semi-quantitative method for hazard and risk assessment
of process installations presented in this paper allows to compare the risk levels for different
media processed in given installation or / and for different installations on site.
The method described above allows also the simulations using for example: different
technical solutions (devices, protective measures) and different technological solutions
(process conditions). As a result of such simulations we can find the answers on questions
like: what to do for significant lowering a risk level of analyzed process? or is it
profitable or not?.
On the basis of assessment procedures carried out for selected manufacturing processes
the authors can conclude that:
1. The system Temclev-Ex it is developed TEMCLEV method (Technology & Media
Classification and Evaluation System) elaborated in the Institute of Industrial Organic
Chemistry, which aim was the assessment of explosion and fire hazard level in chemical
276
Czech Republic
2006
[Content]
industry, especially for installations working with flammable gases and liquids as well as
solids in state of dusts or powders.
2. The results of assessment procedures received during Temclev-Ex system testing,
show the correctness of assumptions taken and the way of computing the indices used in the
system. However, to get more experience and improve the system, the work should be
continued and the system itself implemented into chosen plant of explosives industry.
3. On the base of assessments made the authors may state that Temclev-Ex method
fulfill requirements for correct evaluation of fire and explosion hazards created in different
operations of explosives manufacturing.
4. Elaborated both: the system and the software open a new area on the field of risk
assessment in production of explosives in the Polish industry and create a new quality on the
area of process hazard and risk evaluation tools.
9.
REFERENCES
[1]
Papliski A., Piotrowski T., Maranda A., Frczak M., Saaciski T. Evaluation of Fire and
Explosion Hazard and Risk of Environment Injury for Condensed Explosives by the Method of
Temclev-Ex. Materiay V-th International Conference Scientific Aspects of Armament
Technology, Waplewo, Poland, 6-8 October 2004., 176-177. ISBN 83-921491-0-6
[2] Piotrowski T i inni - Final Report of Project PBZ 038-07, vol.1.: Opracowanie metod i
procedur identyfikacji, oceny i klasyfikacji zagroe procesowych w przemyle chemicznym
dla zapobiegania awariom technologicznym, powodujcym szczeglnie niebezpieczne poary i
wybuchy technologiczne (Elaboration of methods and procedures for identification,
evaluation and classification of process hazards in chemical industry, preventing technological
failures causing extremely dangerous fires and explosions), Biblioteka Instytutu Przemysu
Organicznego, Warszawa, 1998.
[3] Piotrowski T., Hancyk B., Gowiski J., Seweryniak M. Materiay II Kongresu Technologii
Chemicznej, Wrocaw 15 18.09.1997, tom IV, 1798-1808, Dolnolskie Wydawnictwo
Edukacyjne, Wrocaw, 1998.
[4] Piotrowski T., Hancyk B., Gowiski J. - Organika - Prace Nauk. Inst. Przem. Org., 1997-98,
cz. II, 89-99.
[5] Piotrowski T. Przemys Chemiczny 78/12 (1999), 419-421.
[6] Piotrowski T., Hancyk B. Atest Ochrona Pracy 2/2000 (633), 4-7.
[7] Piotrowski T., Haat A. Organika Prace Nauk. Inst. Przem. Org., 1999-2000, 149-163.
[8] Piotrowski T i inni - Final Report of Research Project No 4T09B 06325 vol. I, Adaptacja
systemu Temclev do oceny zagroe poarowo-wybuchowych w procesach produkcji
materiaw wybuchowych, (Adaptation of Temclev System to Fire and Explosion Hazard
Assessment in Manufacturing of Explosives), Biblioteka Instytutu Przemysu Organicznego,
Warszawa, 2005.
[9] Piotrowski T i inni - Final Report of Research Project No 4T09B 06325 vol. II, System
Manual, (Podrcznik Oceny), Biblioteka Instytutu Przemysu Organicznego, Warszawa,
2005.
[10] Piotrowski T., Gowiski J. Technology&Media Classification and Evaluation System
Temclev. System Construction and Examples of Its Practical Use. 11-th International
Symposium Loss Prevention and Safety Promotion in the Process Industries Loss Prevention
2004, Praha, Czech Republic, 31 May 3 June 2004. Paper Full texts, thematic Section A,
1268-1277, PCHE PetroChemEng, Praha 2004, ISBN 80-02-01574-6.
10
277
10.
Czech Republic
2006
[Content]
ACKNOWLEDGEMENT
This project was sponsored by the State Committee for Scientific Research as a grant
number 4 T09B 06325 entitled: Adaptation of Temclev System to Fire and Explosion
Hazard Assessment in Manufacturing of Explosives.
278
Czech Republic
2006
[Content]
DETERMINATION OF THE THERMAL RESISTANCE OF THE

EMULSION EXPLOSIVES
Tomasz Saaciski*, Waldemar Witkowski*, Andrzej Maranda**,
Johann Kasperski***, Barbara Gobek***, and Henryka Boniuk****
* Instytut Przemysu Organicznego, ul. Annopol 6, 03-236 Warszawa, PL
** Wojskowa Akademia Techniczna, ul. Kaliskiego 2, 00-908 Warszawa, PL
*** BLASTEXPOL Sp. z o. o., 59-145 Chocianowiec (Duninw), PL
**** Politechnika Warszawska, ul. Noakowskiego 3, 00-664 Warszawa, PL
Abstract:
In this paper the results of two methods of determination of temperature
of decomposition of modern variations of emulsion explosives have been compared.
First kind of the results are obtain from a method which has been use in the assessment
of hazards during transportation by public roads. The second source of results
is a thermogravimetry method. The influence of the heating with the rate of 5 oC/min in
the range from 100 oC to 400 oC on the stability of emulsion explosives has been
presented. The differences in the characteristics of decomposition of the emulsion
explosives tested have been indicate.
Keywords:
1.
thermal decomposition, emulsion explosives
INTRODUCTION
The main blasting materials that have been use in the mining since for years,
are explosives which contain highly processed technical grade ammonium nitrate(V). Their
first modification were ammonium nitrate(V)-fuel oil (ANFO) explosives. Ammonium
nitrate(V) is present in the ANFO in porous state [1]. Next modifications were slurry [2] and
emulsion [3] explosives. Especially in the latest years, emulsion explosives became very
popular modification of blasting materials [4]. It comes from their undoubted advantages,
e. g.: practically the lack of sensitivity to mechanical stimuli and ability to automatic
loading. In some practical applications, emulsion explosives get explosives properties after
finishing of the loading them into the blasting bore. It means, that during transportation
by public road that variations of emulsion explosives are not explosives. In such causes,
that threatens concerned with the range of explosives being transported by roads
are decreasing significantly.
Emulsion explosives are multicomponent and two- or three-phase explosive systems.
Detonation parameters as well as the structure of emulsion explosives are determine
by the content and by physical and chemical properties of particular components. Basic
ingredients of emulsion explosives are: oxidants, combustible materials, water
and substances which guarantee appearance of such structure that is able to detonation
process. In emulsion explosives, water creates structure of this material
and this is the component that assure presence of ammonium nitrate(V) in the completely
soluble form. The presence of water in slurry and emulsion explosives determines
1
Tomasz Saaciski, Waldemar Witkowski, Andrzej Maranda, . . .
279
Czech Republic
2006
[Content]
the influence of temperature on their ability to detonation in a considerable degree.

Especially, temperatures below zero influence negatively. For example, in the case
of emulsion explosives, if the temperature is below 35 oC, it takes place a precipitation
of ammonium nitrate(V) crystals followed by inversion of its emulsion structure
from water-in-oil into oil-in-water. As a result of this physical processes, emulsion
explosives lose their ability to detonation are no longer explosives.
Even more negatively can act elevated temperatures. In this case, in emulsion explosives
can proceed exothermal reaction that leads to spontaneous and unintended explosion.
This is why, in this paper, we have analyzed the behavior of the emulgites
one of the modification of emulsion explosives - in the elevated temperatures. Within
a framework of the experiments, presented in this paper, the temperature of decomposition
according to the Polish Standard BN-76/6091-08 and to differential thermogravimetry
analysis (DTA) has been determined.
2.
EXPERIMENTAL
2.1
Characteristics of explosives tested
Emulgites were tested during the experiments. Emulgites are a modification of emulsion
explosives. They were manufactured by Blastexpol (Duninw, Poland). General
compositions of the emulgites tested are presented in Table 1.
Table 1. General compositions of the emulsion explosives tested.
Component
Name of explosives
Emulgite
42GP
Emulgite
LWC
Emulgite LWC Emulgite LWC

AN20
ALAN3
Content of the component [%]

Oxidizers
85,0
89,3
87,1
87,4
Organic phase
6,5
6,0
6,0
4,2
Aluminium powder
5,0
Water
8,5
4,7
3,7
3,2
Modification agent
3,2
Oxidizing phases of the emulgites consists of ammonium and sodium nitrates(V).

Organic phase consists of oil and of emulgent. There were any solid components
in the emulgite 42GP. Additionally, emulgites of the type LWC (Low Water Composition)
incorporated not dissolve granule of ammonium nitrate(V). All the emulgites tested were
make sensitive by a chemical method. Recent results of determination of detonation
parameters of emulgite LWC ALAN3 are presented in [5].
280
2.2
Czech Republic
2006
[Content]
Testing methods of sensitivity of explosives on elevated

temperature
Within a framework of the experiments the temperature and character of decomposition

of emulsion explosives, according to the procedure presented in the Polish Standard
BN-76/6091-08 has been determined. The sketch of the experimental set is presented
in Fig. 1. Signal from the thermocouple of the temperature sensor (3) is transmitted
to the data acquisition module ADAM-4018. In all of the tests, three samples (2) of emulsion
explosives, each with the mass of 0,5 g, were tested simultaneously. Samples were heated
in open test tubes (4). The rate of heating was constant in the range from 100 oC to 350 oC
and it was 5 oC/min.
DATA
ACQUISITION
MODULE
4
5
3
6
2
EXAMINER
1
Fig 1.
Scheme of the experimental stand for determination of the temperature

of decomposition, according to the Standard BN-76/6091-08.
Descriptions: 1 heating device; 2 - sample tested; 3 temperature sensor;
4 - test tube; 5 protective glass tube for the sensor (3); 6 Woods alloy;
7 baths.
In DTA, derivatogrametry apparatus PC-MON, made in Hungary, has been used.

Samples of approximately 25 mg were heated in open platinum vessels with a constant rate
of 5 oC/min, in the range of temperatures from 25 oC to 500 oC and in the argon atmosphere.
In the paper, the presentation of the experimental results is restricted to the processes that
appear within the temperature range from 100 oC to 400 oC. It is because, for all the
emulgites tested, there were no evidence of decomposition in the range over 400 oC.
3.
EXPERIMENTAL RESULTS
Results obtained during determination of the temperature of decomposition of emulsion

explosives tested, according to the test procedure presented in the Polish Standard
BN-76/6091-08 and results obtained according to the thermogravimetry differential analysis,
respectively in Table 2 and in Figures 2 4, have been shown. As the evidence
of decomposition of emulsion explosives tested, the occurrence of the white smoke has been
approved.
281
Czech Republic
2006
[Content]
Table 2. Characterisation of decomposition processes of emulgites tested.

Emulsion
explosives
Temperature in which, in the surrounding of the

samples tested, occurred [oC]
Emulgite
white matt
deposit
colorless clear
liquid
white smoke
42 GP
117
147
241
LWC
113
148
236
Temperature in which
occurred the
excitation of the
sample [oC]
281
(calm burning out of
some of the samples)*
284
(violent liberation of
nitrogen oxides)
LWC AN20
120
160
LWC ALAN3
133
159
* - the rest of samples no firing until 350 oC.
226
no firing until 350
226
no firing until 350
2,5
274,5
2
-10
-20
283,7
249,0oC (19,73%)
1,5
egzo
-30
Loss of mass TG [%]
DTA
256,1
0,5
-40
0
-50
-0,5
-60
DTA
235,5
-1
-70
-1,5
TG
-80
-2
-90
-2,5
-100
-3
100
200
300
400
Temperature [oC]
Fig 2.
4.
TG-DTA curves of emulgite LWC ALAN3
DISCUSSION OF RESULTS
In support of the results obtained during determination of the temperature

of decomposition (Table 2), one can draw the conclusion, that the influence of heating
on the emulsion explosives tested is pointed out much below their temperature
of decomposition (temperature in which white smoke occurred). Despite the similar content
of particular groups of components (Table 1), there can be different effects accompanied
with the decompositions of the samples (from calm burning out, through violent flameless
282
Czech Republic
2006
[Content]
Loss of mass TG [%]
-20
o
222,8 C (20,5%)
-40
o
266,4 C (52%)
LWC 20AN
LWC ALAN 3
-60
LWC
42GP
-80
-100
100
200
300
400
Temperature [oC]
Fig 3.
TG curves of emulgites 42GP, LWC, LWC AN20 and LWC ALAN3
DTA
25
20
egzo
15
222,8
42GP
10
LWC
LWC 20AN
5
235,5
0
LWC ALAN 3
-5
100
200
300
400
Temperature [oC]
Fig 4.
DTA curves of emulgites 42GP, LWC, LWC AN20 and LWC ALAN3
- liberation of nitrogen oxides, to lack of visible decomposition effects). Effects that

appeared during decomposition of emulgites LWC do not lead to ignition, unlike
to emulgite 42GP. However ignition of emulgite 42GP did not occurred every time.
Basing on comparisoning of the data presented in Fig. 2 4, one can point out,
that the start of decomposition of emulgite 42GP becomes the earliest. Decomposition
of emulgites LWC begins later than for 42GP and the total exothermic effect is lower.
Between the emulgites tested, decomposition of emulgite LWC AN20 starts as the latest.
In 100 oC, loss of mass of emulgite LWC is approximately 1,5 % and of emulgite 42GP
is approximately 5,2 %. Visual observations founded during liberation of components
from the samples tested (white matt deposit in temperature ca. 120 oC and colorless clear
5
283
Czech Republic
2006
[Content]
liquid in temperature ca. 150 oC) were not confirm in the running of TG-DTA curves.
It comes from it, that one of the factors which decide about results of thermal stability testing
of emulsion explosives is the mass of samples. It determine the rate of the process of heat
exchanging between a sample and surroundings.
5.
CONCLUSIONS
1. The heating of emulgites 42GP and LWC is accompanying by various effects.
The effects are observable in both methods by visual observation (character of
decomposition) and by DTA testing (mass losses).
2. The scheme of decomposition obtained in respect to the visual observations of
samples with an order of magnitude of the weight 500 mg differ from results
obtained in DTA method, in which samples were of the weight 25 mg.
3. Emulgites LWC tested decompose much calm and later than emulgite 42GP.
4. Generally, the presence of aluminum powder do not cause differ character of
decomposition of emulgite LWC ALAN3, in comparison to the rest of materials
tested.
5. The emulsion explosives tested are thermally stable in the aspect of their use in the
deep mines.
REFERENCES
[1] V. L. BARON, V. KH. KANTOR: Tekhnika i tekhnologia vzryvnykh rabot v SshA, Izd.
Nedra, Moskva, 1989.
[2] M. A. COOK, H. E. FARNAM: PAT. USA2 930 685, 1960.
[3] H. F. BLUHM: PAT. USA 3 447 978, 1969.
[4] W. THUM: Blasting techniques in the German quarry industry. Proc. of 1st World.
Conf. on Explosives Blasting Technique, Munich, 2000.
[5] J. PASZULA, A. MARANDA, A. PAPLISKI, B. GOBEK, J. KASPERSKI: An analysis of
blast waves parameters and underwater explosion test of emulsion explosives and
dynamites. Proc. of 8th Seminar New Trends in Research of Energetic Materials Part II,
Pardubice, 2005.
Research work financially supported within the state research founds for science within the
years 2005 2007.
284
Czech Republic
2006
[Content]
REDUCTION OF NITROAROMATIC EXPLOSIVES BY OXYGENINSENSITIVE NAD(P)H:NITROREDUCTASES: IMPLICATIONS FOR

THEIR CYTOTOXICITY AND BIODEGRADATION
Jonas arlauskas*, Lina Miseviien*, Henrikas Nivinskas*, ilvinas Anuseviius*,
Aura Nemeikait-nien**, Evaldas Maldutis***, and Narimantas nas*
* Institute of Biochemistry, Mokslinink 12, LT-08662 Vilnius, Lithuania

** Institute of Immunology of Vilnius University, Molt Pl. 29, LT-08409, Vilnius, Lithuania
*** The General J. emaitis Military Academy of Lithuania, ilo 5a, LT-10322 Vilnius, Lithuania
Abstract:
Nitroaromatic explosives are toxic and mutagenic to humans and other mammalian
species. The first step(s) in the biodegradation/bioremediation of the explosive residues
in soil or groundwater is their reduction by bacterial oxygen-insensitive nitroreductases
to the relatively stable metabolites. Here we analyze the quantitative structure-activity
relationships in the reduction of nitroaromatic explosives and model nitroaromatic
compounds by mammalian DT-diaphorase (NQO1) and Enterobacter cloacae
NAD(P)H:nitroreductase (NR), which perform the four-electron reduction of
nitrogroups to corresponding hydroxylamines, and by Enterobacter cloacae PB2
NADPH: pentaerythritol tetranitrate
reductase (PETNR),
which performs
nitroreduction and reduction of benzene ring with the formation of hydrideMeisenheimer adducts. Our data show that in all the cases the reduction rate of
nitroaromatics mainly depends on the energetics of the charge transfer.
Keywords:
1.
nitroaromatic explosives; enzymatic reduction of; biodegradation of;
INTRODUCTION
Nitroaromatic explosives are toxic and mutagenic to humans and other mammalian
species. Their toxicity is partly attributed to the single-electron enzymatic reduction to
radicals accompanied by the formation of the reactive oxygen species (oxidative stress type
of cytotoxicity).[1,2] Single-electron reduction of nitroaromatics is performed by
flavoenzymes dehydrogenases-electrontransferases, e.g., NADPH:cytochrome P-450
reductase (P-450R, EC 1.6.2.4) or ferredoxin:NADP+ reductase (FNR, EC 1.18.1.2). In this
enzyme group, the flavin adenine dinucleotide (FAD) or flavin mononucleotide (FMN)
cofactors are characterized by the high stability of their neutral (blue) radicals at the
equilibrium.[3] The reduction of nitroaromatic compounds by P-450R and FNR follows an
outer-sphere electron-transfer model, i.e., the reactivity increases with an increase in their
single-electron reduction potential (E17), and is relatively insensitive to their structural
pecularities.[2,3]
Another important mechanism of cytotoxicity of nitroaromatic compounds is their two(four)-electron reduction to nitroso and hydroxylamino metabolites, which covalently bind
to proteins and DNA: [4]
ArNO2 + 2e- + 2H+ ArNO + H2O,
1
Jonas Sarlauskas, Lina Miseviciene, Henrikas Nivinskas, . . .
(1)
285
Czech Republic
2006
ArNO + 2e- + 2H+ ArNHOH
[Content]
(2)
In mammalian organism, this type of reactions is catalyzed by NAD(P)H:quinone

oxidoreductase (DT-diaphorase, NQO1, EC 1.6.99.2), which is localized mainly in the liver.
Another source of the formation of toxic aromatic hydroxylamines in mammals is the
gastrointestinal tract, where the nitroaromatic compounds are efficiently reduced by the
nitroreductases of Escherichia and Enterobacter sp. Because of the relative stability of the
nitroreduction products under aerobic conditions, these enzymes are named the oxygeninsensitive nitroreductases.
It is important to note that the reactions (1,2) being catalyzed by the oxygen-insensitive
nitroreductases of numerous bacteria, microorganisms, and fungi, also make a first step in
the biodegradation of nitroaromatic explosives in the environment.[5] In parallel, certain
nitroreductases reduce the aromatic ring with the formation of Meisenheimer-type hydride
adducts, which further undergo denitration:[5]
Ar(NO2)3 + H- ArH(=NO2-)(NO2)2
ArH(=NO2-)(NO2)2 + H- ArH(=NO2-)2NO2 NO2- + unidentified products
(3)
(4)
Typically, all the oxygen-insensitive nitroreductases including NQO1 possess the flavin
cofactor with the destabilized anionic (red) semiquinone state at the equilibrium. One may
note that in contrast to the reactions of single-electron reduction of nitroaromatics by
flavoenzymes, the mechanisms and the substrate specificity of their two-(four)-electron
reduction still remain insufficiently understood.
In this paper, we summarize the data of our previous[2,6-10] and ongoing studies on the
mechanisms of two-electron reduction of nitroaromatic explosives and model compounds by
oxygen-insensitive nitroreductases: rat NQO1 and Enterobacter cloacae NAD(P)H:
nitroreductase (NR) which reduce the nitrogroup(s) into hydroxylamine(s) (reactions (1,2)),
and Enterobacter cloacae PB2 NADPH:pentaerythritol tetranitrate (PETN) reductase
(PETNR), which in parallel performs the aromatic ring reduction with the formation of
Meisenheimer-type adducts (reactions (3,4)). The methods of synthesis of nitroaromatic
explosives and model nitroaromatic compounds, purification of NQO1 and NR, enzyme
kinetics and cytotoxicity studies, and quantum mechanical calculations were reported in our
previous papers.[2,7-10] Because the directly determined E17 values are unavailable for a large
number of nitroaromatic explosives, we also used their E17 values calculated from the data of
their single-electron enzymatic reduction (E17(calc.)) (www.bchi.lt/potential.pdf). PETNR was
obtained from Professor Nigel S. Scrutton and Dr. Richard J. Harris (University of Leicester,
UK).
2.
NITROREDUCTASE REACTIONS OF RAT LIVER NQO1
Mammalian NQO1 is a dimeric (2 x 31 kD) enzyme containing one molecule of FAD per
subunit. The midpoint potential of FAD at pH 7.0 is -0.159 V, whereas the potentials of
FAD/FAD-. and FAD-./FADH- couples are -0.200 and -0.118 V, respectively.[11] This
provides ca. 7 % FAD-. stabilization at the equilibrium. Quinone reduction by NQO1
follows a ping-pong scheme and often proceeds at a high maximal rate, > 1000 s-1. The
reactivity of quinones increases with an increase in their single-electron reduction potential
and decreases with an increase in their Van der Waals volume (VdWvol) above 200 3 , due
to the sterical hindrances.[12]
Recent X-ray and computer modeling studies provided
evidence that mono- and bicyclic quinones, an inhibitor of NQO1, dicumarol, and
286
Reduction of Nitroaromatic Explosives by Oxygen-insensitive . . .
Czech Republic
2006
[Content]
dihydronicotinamide ring of NADPH bind at the same domain in the active center.[13] These
compounds interact with tyrosine-126, histidine-161, and phenylalanine-106, and form -
complex with the isoalloxazine ring of FAD at a distance of 3.5 by displacement of
tyrosine-128.
Nitroaromatic compounds, although being less reactive than quinones, comprise
another important group of substrates for NQO1. NQO1 plays a considerable role in
mammalian cell cytotoxicity and genetoxicity of nitroaromatic pollutants - the components
of exhaust gases or the explosive residues in the environment. However, the mechanism and
substrate specificity of nitroreductase reactions of NQO1 are still poorly understood. There
is no X-ray data on the complexes of nitroaromatic compounds in the active center of
NQO1, except several computer modeling studies.[14,15]
We examined NQO1-catalyzed reduction of 39 nitroaromatic compounds, including 16
explosive compounds (TNT and its metabolites, tetryl, pentryl, dipentryl, hexyl,
tetranitrocarbazole, nitrobenzimidazolones, nitrobenzofuroxanes, and nitrotriazoles). The
obtained multiparameter regressions show that the reactivity of compounds increases with an
increase in their E17(calc.) with the coefficient log rate constant /E17(calc.) ~ 10 V-1, and the
highest torsion angle between the nitrogroup(s) and the aromatic ring. Another although less
certain factor increasing the reactivity of nitroaromatic compounds, is their affinity to the
oxidized form of NQO1. However, the use of the above parameters resulted in the relatively
uncertain (r2 ~ 0.800) correlations. Another important but poorly understood factor
influencing the reactivity of nitroaromatics is their ability to bind at the different domains in
the active center of NQO1. Our data partly contradict the computer modeling data[14,15]
which suggest that nitroaromatic compounds should bind at the dicumarol/quinone binding
domain, because: i) the entropies of activation (S) of nitroreductase reactions of NQO1 are
much less negative than S of quinone reduction, thus demonstrating the less efficient
electronic coupling of reactants, ii) increase in VdWvol of nitroaromatics up to 250-300 3
does not decrease their reactivity, but increases it, and iii) dicumarol acts as incomplete
inhibitor towards several nitroaromatic substrates, including tetryl. It shows that a number of
nitroaromatic compounds may bind outside the dicumarol/quinone binding domain.
However, dicumarol acts as a complete (linear) inhibitor with respect to 4,5,6,7tetranitrobenzimidazolone, whose reactivity is much higher than expected. This is probably
caused by its binding at the dicumarol/quinone binding domain. Another important aspect of
catalysis of NQO1 is that it catalyzes reductive N-denitration of tetryl in a single-electron
way, with the formation of N-methylpicramide, nitrite, superoxide, and oxidation of excess
NADPH:[2]
e-
Tetryl
O 2-.
3.
[Tetryl]-.
O2
e-, H+
N-Methylpicramide
NO 2
(5)
REACTIONS OF E. CLOACAE NITROREDUCTASE
Enterobacter cloacae NAD(P)H:nitroreductase (NR, EC 1.6.99.7) is a dimeric (2 x 24.5

kD) protein, containing a molecule of FMN per subunit. The midpoint potential of FMN at
pH 7.0 is -0.190 V, with less than 1 % FMN-. stabilization at the equilibrium.[16] E. cloacae
NR shares a significant sequence homology with nitroreductases from Escherichia coli,
Salmonella typhimurium, and FMN reductase from Vibrio fischeri. E. cloacae NR reduces
3
287
Czech Republic
2006
[Content]
nitroaromatic compounds according to a ping-pong scheme with strong competitive to

NAD(P)H inhibition by the nitroaromatic substrates. It shows that NAD(P)H and
nitroaromatic compounds bind at the same domain. NR reduces nitroaromatic compounds at
high maximal rates, exceeding 100 s-1 or even 1000 s-1 for tetryl and pentryl.[7] The reduction
of polynitroaromatic compounds (tetryl, TNT, dinitrobenzenes) occurs biphasically, with a
fast consumption of 2 moles NAD(P)H per mole of nitrocompound, and subsequent slower
oxidation of further excess NAD(P)H. The second phase of reaction is accompanied by O2
uptake. It shows that polynitroaromatic compounds are first reduced to
monohydroxylamines. According to the studies of E. coli NR, the reduction of nitroso
compounds into hydroxylamines (reaction (2)) is 104 times faster than the reaction (1).[17]
We found that monohydroxylamine metabolites of TNT are further reduced by E. cloacae
NR, but at slower rates than TNT.[8] The reduction of TNT is not accompanied by nitrite
formation. E. cloacae NR and NQO1 reduce tetryl in different ways: NR reduces tetryl to
unidentified product(s) other than N-methylpicramide, nitrite is formed not simultaneously
with the NAD(P)H oxidation, but after a lag-period.
In general, the reactivity of nitroaromatic compounds towards E. cloacae NR increases
with an increase in their electron-accepting potency, which may be expressed either as
E17(calc.), or as the enthalpies of free radical (Hf(ArNO2-.)) or hydride adduct (anionic N,Ndihydroxylamine) formation (Hf(ArN(OH)O-)).[7] The observed parabolic log rate constant
vs. potential dependence is relatively uncertain. However, at present we are unable to
characterize the structural peculiarities of nitroaromatic compounds prerequisite for their
efficient reduction by E. cloacae NR. The X-ray data of E. cloacae and E. coli NR show that
pyrimidine ring of FMN is close to the surface of protein globule.[18] The X-ray data of the
complex of oxidized E. coli NR with nitroaromatic compound nitrofurazone shows that the
compound binds at the isoalloxazine ring of FMN displacing phenylalanine-70 and
phenylalanine-124, but with an unproductive orientation.[17] These factors may impose
flexibility and multiple orientations in the binding of nitroaromatic substrates. It is also
possible that hexyl and hexyl-N-ethylnitrate may bind at several domains of NR, e.g., at the
catalytic and intersubunit domain, because their strong substrate inhibition (Ki 1 M at 200
M NADH) is characterized by the nonlinear (parabolic) Dixon plots. On the other hand, the
reactivity of 4,6-dinitrobenzofuroxane is much lower than one may expect from its E17(calc.)
value.[10] A possible reason for its low reactivity is a loss of aromaticity of benzene ring,
because this compound exists in a form of Meisenheimer-type hydroxy adduct.
4.
REACTIONS OF E. CLOACAE PB2 PETN REDUCTASE
E. cloacae PB2 NADPH:PETN reductase (PETNR) is a monomeric 40 kD protein,

containing a molecule of FMN per subunit. The midpoint potential of FMN at pH 7.0 is
-0.267 V, with less than 1 % FMN-. stabilization at the equilibrium.[19] PETNR belongs to a
structurally separate group of flavoenzymes, the Old Yellow Enzyme (OYE, Saccharomyces
carlsbergensis NADPH:oxidase) family, which also includes Pseudomonas putida M10
morphinone reductase (MR), Pseudomonas fluorescens xenobiotic reductase B (XenRB),
and E. coli N-ethylmaleimide reductase (NEMR).[20] All the enzymes of the OYE family
perform two independent reactions, the reduction of aromatic nitrogroup(s) into
hydroxylamine(s) (reactions (1,2)), and the aromatic ring reduction with the formation of
Meisenheimer-type hydride adducts, which further undergo denitration (reactions (3,4)).
The moles of nitrite formed per mole of TNT consumed vary from 0.25 (NEMR, XenRB)
and 0.18 (PETNR), to 0.03 (OYE, MR).[20] The pathways of nitrite formation during the
288
Czech Republic
2006
[Content]
reductive degradation of TNT require more thorough studies. It has been proposed that NO2may be liberated during the nonenzymatic reaction between 4-hydroxylamino-DNT and
Meisenheimer-type TNT-hydride diadduct.[21] However, according to the latter studies, the
denitration occurs in the absence of 4-hydroxylamino-DNT, and its rate depends on enzyme
concentration.[20] A specific feature of the OYE family members is that apart from the
reduction of nitroaromatics, they perform the reductive denitration of PETN and glycerol
trinitrate, although with different substrate preference. The transgenic tobacco plants
expressing PETNR perform the biodegradation of nitroaromatic and nitroaliphatic
explosives.[22]
We examined the reduction of 28 nitroaromatic compounds by PETNR, including 20
explosive compounds. For 22 compounds with available E17(calc.) values, there exists an
uncertain linear log rate constant vs. E17(calc.) relationship (r2 = 0.6535) with the coefficient
log rate constant /E17(calc.) ~ 12 V-1. It points to the important role of charge transfer
energetics in the reactivity of compounds. The maximal reduction rates of the most reactive
oxidants, tetryl, pentryl and hexyl, are close to 35 s-1, and their bimolecular reduction rate
constants are above 107 M-1s-1. In contrast to E. coli NR, the reactivity of 4,6dinitrobenzofuroxane also follows this relationship. At present, we cannot identify the
structural peculiarities of nitroaromatics prerequisite for their efficient reduction by PETNR.
According to the X-ray data, the binding of picric acid and TNT to PETN involves their
parallel orientation towards the isoalloxazine of FMN, and the interaction of 1-hydroxy or 1methyl groups with histidines-181,184.[23] This causes their favourable orientation for
hydride transfer with the Meisenheimer-type adduct formation, i.e., the proximity of C-5
carbon to the flavin N-5. However, our data show that the nature of the substituent in 1position of TNT, picramide, N-methylpicramide, 2,4,6-trinitroanizole, or its absence in
2,4,6-trinitrobenzene does not strongly influence the reactivity of nitroaromatics. In contrast,
the maximal reduction rates of tetryl and pentryl are by 20 times higher, and their
bimolecular reduction rate constants by 500 times higher than those of TNT, although the
above compounds possess bulky substituents in 1-position. This may be explained by higher
electron-accepting potency of tetryl and pentryl. The strongest inhibitors of PETNR, hexyl
(Ki < 2 M) and tetranitrocarbazole (Ki = 3.5 M) are structurally different. The presence of
bulky N-ethylnitrate group in hexyl-N-ethylnitrate decreases its affinity to PETNR (Ki = 70
M) and reactivity by 10 times as compared to hexyl. However, the reactivities of hexyl and
pentryl are similar, although pentryl is much weaker inhibitor of PETNR (Ki = 150 M).
This shows that nitroaromatic compounds may bind in the active center of PETNR in
different ways, and that their efficient interaction with histidines-181,184 is not obligatory.
We found that nitrite was formed during the reduction of structurally different compounds,
such as TNT, picramide, tetryl, pentryl, tetranitrobenzimidazolone and tetranitrocarbazole.
The reductive denitration of tetryl did not follow the single-elctron transfer (reaction (5)),
and yielded other than N-methylpicramide reaction product(s). However, the reduction of pdinitrobenzene was not accompanied by the production of nitrite, yielding
phydroxylaminonitrobenzene. We found that this was not due to the structural peculiarities of
p-dinitrobenzene, but due to the more favourable energetics of reduction of nitrogroup
instead of aromatic ring, which was reverse for the compounds forming nitrite.
5.
CONCLUSIONS
The data of this work show that apart from the currently insufficiently understood
structural peculiarities, an increase in the electron-accepting potency of nitroaromatic
explosives, preferably determined in the aqueous medium, increases their reactivity towards
5
289
Czech Republic
2006
[Content]
two-electron transferring oxygen-insensitive nitroreductases. In general, the mammalian

cell cytotoxicity of nitroaromatic explosives increases with an increase in their E17 or
E17(calc.), [2] except for the amino- and hydroxylamino- metabolites of TNT, which possess
higher than expected cytotoxicity.[8,24,25] In this work, we demonstrate the existence of a
certain parallelism between the reactivity of nitroaromatic explosives towards the singleelectron transferring flavoenzymes and two-electron transferring NQO1. Irrespectively of
this, the positive relationship between the electron-accepting properties of nitroaromatics and
their cytotoxicity should point to the leading role of the oxidative stress-type cytotoxicity,
because the inhibitor of NQO1, dicumarol, equivocally affects the cytotoxicity of
nitroaromatic compounds.[2] Our data also indicate that the order of reactivity of certain
nitroaromatic explosives (nitrotriazoles < TNT < tetryl, pentryl and hexyl) towards two
different bacterial nitroreductases, NR and PETNR, is also in line with their relative
electron-accepting potency. On the other hand, the enzyme specificity strongly influences
the
reduction
rate
of
dinitrobenzofuroxane,
tetranitrobenzimidazolone
and
tetranitrocarbazole. Besides, certain nitroaromatic explosives, e.g., hexyl, may act as a strong
inhibitors of both nitroreductase types, thus impeding their own biodegradation. Because the
nitro- and/or the aromatic ring reduction rates may influence the overall rate of
biodegradation of nitroaromatic explosives,[5] the data presented may provide some
guidelines for the design of specifically adjusted biodegradation systems.
Acknowledgements
This work was supported in part by the Lithuanian State Science and Studies
Foundation. We thank Professor Nigel S. Scrutton and Dr. Richard J. Harris (University of
Leicester, UK) for the fruitful collaboration.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
290
KONG, Q. JIANG and Q. QU: Formation of Superoxide Radical and Hydrogen Peroxide
Enhanced by Trinitrotoluene in Rat Liver, Brain, Kidney, and Testicle in Vitro and Monkey
Liver in Vivo, Biomed. Environ. Sci., 2, 72-77, 1989
N. NAS, A. NEMEIKAIT-NIEN, E. SERGEDIEN, H. NIVINSKAS, .
ANUSEVIIUS and J. ARLAUSKAS: Quantitative Structure-Activity Relationships in
Enzymatic Single-Electron Reduction of Nitroaromatic Explosives: Implications for their
Cytotoxicity, Biochim. Biophys. Acta, 1528, 31-38, 2001
M.V. ORNA and R.P. MASON: Correlation of Kinetic Parameters of Nitroreductase Enzymes
with Redox Properties of Nitroaromatic Compounds, J. Biol. Chem., 264, 12379-12384, 1989
K.H. LEUNG, M. YAO, R. STEARNS and S-H.L. CHIU: Mechanism of Bioactivation and
Covalent Binding of 2,4,6-Trinitrotoluene, Chem.-Biol. Interact., 97, 37-51, 1995
J.C. SPAIN: Biodegradation of nitroaromatic compounds, Annu. Rev. Microbiol., 49, 523-555,
1995
J. ARLAUSKAS, E. DIKANCAIT, A. NEMEIKAIT, . ANUSEVIIUS, H.
NIVINSKAS, J. SEGURA-AGUILAR and N. NAS: Nitrobenzimidazoles as Substrates for
DT-Diaphorase and Redox Cycling Compounds: Their Enzymatic Reactions and Cytotoxicity,
Arch. Biochem. Biophys., 346, 219-229, 1997
H. NIVINSKAS, R.L. KODER, . ANUSEVIIUS, J. ARLAUSKAS, A.-F. MILLER and N.
NAS: Quantitative Structure-Activity Relationships in Two-Electron Reduction of
Nitroaromatic Compounds by Enterobacter Cloacae NAD(P)H:Nitroreductase, Arch.
Biochem. Biophys., 385, 170-178, 2001
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
Czech Republic
2006
[Content]
J. ARLAUSKAS, A. NEMEIKAIT-NIEN, . ANUSEVIIUS, L. MISEVIIEN, M.

MARTINEZ-JULVEZ, M. MEDINA, C. GOMEZ-MORENO and N. NAS: FlavoenzymeCatalyzed Redox Cycling of Hydroxylamino- and Amino Metabolites of 2,4,6-Trinitrotoluene:
Implications for their Cytotoxicity, Arch. Biochem. Biophys., 425, 184-192, 2004
J. ARLAUSKAS, A. NEMEIKAIT-NIEN, . ANUSEVIIUS, L. MISEVIIEN, A.
MAROZIEN, A. MARKEVIIUS and N. NAS: Enzymatic Redox Properties of Novel
Nitrotetrazole Explosives: Implications for their Toxicity, Z. Naturforsch., 59c, 399-404, 2004
A. NEMEIKAIT-NIEN, J. ARLAUSKAS, L. MISEVIIEN, . ANUSEVIIUS, A.
MAROZIEN and N. NAS: Enzymatic Redox Reactions of the Explosive 4,6Dinitrobenzofuroxan (DNBF): Implications for its Toxic Action, Acta Biochim. Polon., 51,
1081-1086, 2004
G. TEDESCHI, S. CHEN and V. MASSEY: DT-Diaphorase. Redox Potential, Steady-State,
and Rapid Reaction Studies, J. Biol. Chem., 270, 1198-1204, 1995
. ANUSEVIIUS, J. ARLAUSKAS and N. NAS: Two-electron Reduction of Quinones
by Rat Liver NAD(P)H:Quinone Oxidoreductase: Quantitative Structure-activity relationships,
Arch. Biochem. Biophys., 404, 254-262, 2002
M.A. BIANCHET, M. FAIG and L.M. AMZEL: Structure and Mechanism of
NAD(P)H:Quinone Acceptor Oxidoreductase (NQO), Meth. Enzymol., 382B, 144-174, 2004
J.V. SKELLY, M.R. SANDERSON, D.A. SUTER, U. BAUMANN, M.A. READ, D.S.J.
GREGORY, M. BENNETT, S.M. HOBBS and S. NEIDLE: Crystal Structure of Human DTDiaphorase: A Model for Interaction with the Cytotoxic Prodrug 5-(Aziridin-1-yl)-2,4dinitrobenzamide (CB 1954), J. Med. Chem., 42, 4325-4330, 1999
S. CHEN, K. WU, D. ZHANG, M. SHERMAN, R. KNOX and C.S. YANG: Molecular
Characterization of Binding of Substrates and Inhibitors to DT-Diaphorase: Combined
Approach Involving Site-Directed Mutagenesis, Inhibitor-Binding Analysis, and Computer
Modeling, Mol. Pharmacol., 56, 272-278, 1999
R.L. KODER, C.A. HAYNES, M.E. RODGERS, D.W. RODGERS and A.F. MILLER: Flavin
Thermodynamics Explain the Oxygen Insensitivity of Enteric Nitroreductases, Biochemistry,
41, 14197-14205, 2002
P. R. RACE, A.L. LOVERING, R.M. GREEN, A. OSSOR, S.A. WHITE, P.F. SEARLE, C.J.
WRIGHTON and E.I. HYDE: Structural and Mechanistic Studies of Escherichia Coli
Nitroreductase with the Antibiotic nitrofurazone. Reversed Binding Orientations in Different
Redox States of the Enzyme, J. Biol. Chem., 280, 13256-13264, 2005
C.A. HAYNES, R.L. KODER, A.F. MILLER and D.W. RODGERS: Structures of
Nitroreductase in Three States: Effects of Inhibitor Binding and Reduction, J. Biol. Chem.,
277, 11513-11520, 2002
H. KHAN, T. BARNA, R.J. HARRIS, N.C. BRUCE, I. BARSUKOV, A.W. MUNRO, P.C.E.
MOODY and N.S. SRUTTON: Atomic Resolution Structures and Solution Behavior of
Enzyme-Substrate Complexes of Enterobacter cloacae PB2 Pentaerythritol Tetranitrate
Reductase, J. Biol. Chem., 279, 30563-30572, 2004
R.E. WILLIAMS, D.A. RATHBONE, N.S. SCRUTTON and N.C. BRUCE: Biotransformation
of Explosives by the Old Yellow Enzyme Family of Flavoproteins, Appl. Environ. Miocrobiol.,
70, 3566-3574, 2004
J.W. PAK, K.L. KNOKE, D.R. NOGUERA, B.G. FOX and G.H. CHAMBLISS:
Transformation of 2,4,6-Trinitrotoluene by Purified Xenobiotic Reductase B from
Pseudomonas fluorescens I-C, Appl. Environ. Microbiol., 66, 4742-4750, 2000
C.E. FRENCH, S.J. ROSSER, G.J. DAVIES, S. NICKLIN, N.C. BRUCE: Biodegradation of
Explosives by Transgenic Plants Expressing Pentaerythritol Tetranitrate Reductase, Nat.
Biotechnol., 17, 491-494, 1999
H. KHAN, R.J. HARRIS, T. BARNA, D.H. CRAIG, N.S. BRUCE, A.W. MUNRO, P.C.E.
MOODY and N.S. SCRUTTON: Kinetic and Structural Basis of Reactivity of Pentaerythritol
291
Czech Republic
2006
[Content]
Tetranitrate Reductase with NADPH, 2-Cyclohexenone, Nitroesters, and Nitroaromatic

Explosives, J. Biol. Chem., 277, 21906-21912, 2002
[24] B. LACHANCE, P.Y. ROBIDOUX, J. HAWARI, G. AMPLEMAN, S. THIBOUTOT and G.I.
SUNAHARA: Cytotoxic and Genotoxic Effects of Energetic Compounds on Bacterial and
Mammalian Cells in Vitro, Mutat. Res., 444, 25-39, 1999
[25] M.E. HONEYCUTT, A.S. JARVIS and V.A. McFARLAND: Cytotoxicity and Mutagenicity of
2,4,6-Trinitrotoluene and Its Metabolites, Ecotoxicol. Environ. Saf., 35, 282-287, 1996
292
Czech Republic
2006
[Content]
COMPLEX RESEARCH OF AL NANOPOWDERS PRODUCED BY

THE EXPLODING WIRE METHOD
Valentin Sedoi *, Galina Volkova **, Mirslan Osmonaliev *, Victor Ivanov**,
Ludmila Chemezova *, and Alexander Alekseev **
* Institute of High Current Electronics RAS, 634055, Tomsk, Russia
** Institute of Petroleum Chemistry RAS, 634055, Tomsk, Russia
Abstract:
The impetus for this research came from fundamental and applied questions connected
with specified characteristics of Al nanopowders, theirs controlled syntheses and
passivation, and theirs activity as high energetic ingredients in solid propulsion and in
aluminum-water gels.
Different conditions of powder syntheses by the exploding wire method and powder
passivation were used: syntheses in low-pressure gases (Ar, N2, CO2), passivation in
atmospheric-pressure gases (air, N2, CO2).
Investigations of particles size and morphology have been accomplished with a 125-kV
transmission electron microscope. The specific surface area S has been determined by
BET method. Particles are predominately spherical, BET surface area was of 15 40
m2/g. A composition and surface properties of nanopowders produced under various
conditions were studied by mass-spectroscopy techniques at temperatures of 0 450 0C,
electron and X-ray difraction methods, and by element analysis methods.
To determine activities of the powder samples in the combustion reactions, the
Differential Thermal Analysis, Thermogravimetry, and reaction with water were
conducted. Active metal contents have been determined from amount of impurities and
hydrogen extracting in direct thermohydrolysis of nanopowders by distilled water at 60
0
C. Kinetics of gas evolution was determined therewith and shows two steps. Active
aluminum content is within range of 16 80 %. A composition of adsorbed gases and
dynamics of desorption in vacuum have been investigated too.
Correlation between conditions of nanopowders syntheses, their specific surface and
chemical activity in reaction with water and maximum oxidation degree have been
determined.
Keywords:
1.
nanopowders, characterization, passivation, oxidation, combustion
INTRODUCTION
Different methods are developed and used for nanosized metal powder production: laser,
electron- beam, Gen-Miller, electrical explosion, plasma-chemical, and others. Among them
the exploding wire method has the advantage of direct conversion of electrical energy into
heat with high efficiency [1]. Besides, the pulsed heating provides high and uniform energy
densities, high velocities of expansion and high rates of the substance cooling. As a result,
dimensional, structural, phase, defect, and other energy-saturated states are realized in the
particles formed.
Growing interest in nanopowders is stimulated by the possibility to increase the
inflammability, the burning rate, and the completeness of metal combustion and to decrease
1
Valentin Sedoi, Galina Volkova, Mirslan Osmonaliev, Victor Ivanov, . . .
293
Czech Republic
2006
the agglomeration. The area of investigations includes metallized solid

metallized gelled propellants [4].
[Content]
[2]
, liquid
[3]
, and
The purpose of the present paper came from fundamental and applied questions
connected with specified characteristics of nanopowders produced by the exploding wire
method, and their activity as high energetic ingredients in solid propulsion [2] and aluminumwater gels [5-8].
The structure of the paper is as follows. Section 2 describes the methods and techniques
of investigation, different apparatus and physical-chemical methods used for analysis of
powders. Section 3 dwells on a size of particles and crystallites. Section 4 is devoted to
constitution of powder samples. Section 5 describes the results of investigation of powder
activity in reactions with air and hot water. Powder fineness effect is described as well.
Section 6 demonstrates powder fineness effect on combustion of Al/gelled water mixtures. It
has been established that the combustion rate sufficiently grows when a high-dispersed
powder is used.
2.
EXPERIMENTAL METHODS
Experimental setups include the installation for powder production and facilities for
characterization of the powder produced.
2.1
Engineering conditions
Engineering conditions for production of powders by the exploding wire method with
given requirements are formulated in the papers [1, 9]. The investigations performed have
shown that the electrical explosion allows synthesizing the powders passivated by oxide or
nitride [9, 10, 11] as well as by oxycarbide and by the layers of absorbed gases [8]. With that, the
activity of powders remains high.
The major characteristic in powder production is the thermal energy introduced into the
wire material. In this study, the energy density was close to the sublimation energy of
aluminum, and the conditions of uniform Joule heating of a wire [9, 12] were used.
2.2
Determination of dispersivity of powders
The analysis of ultra-fine particles is a difficult problem. Measurements only of the

specific surface area of the powder provide accurate information about particle size if the
morphology, smoothness and degree of equiaxialness of the particle are known.
Transmission Electron Microscope (TEM) analysis is capable to give some information on
these properties.
The specific surface area SBET of the samples was measured by the low-temperature
adsorption method (BET).
The particle shape and size distributions were determined using the 125 kV TEM. In
preparation, the suspension powder - industrial alcohol was applied on a slide and dried.
Then a carbon layer was sputtered on the powder-slide using the vacuum deposition system.
The resulted carbon replica was isolated by the use of water solution of gelatin, gelatin was
dissolved, and the carbon replica with the sputtered powder was placed on the copper mesh.
The mesh with the powder was used as the subject for TEM investigations. The mean
diameters D and the standard deviation were calculated from TEM data. Medium sizes of
the powder particles and their size distribution were analyzed with the microphotographs
with (250 300) 103 magnification. The results are given below. The number of measuring
294
Complex Research of Al Nanopowders Produced by the Exploding Wire Method
Czech Republic
2006
[Content]
particles is averaged 300-500 depending on the powder size uniformity. The algebraic
number (Ni/N) of particles with the size of the given interval was plotted in histograms.
It is necessary to have in view that this procedure is limited by the particle size of more
than 5 micrometer, because such particles may be lost. So, the results will be distorted. In
this connection, common measurements by TEM and BET methods will be more
informative.
The X-ray diffraction (XRD) method was used to determine mean sizes of regions of
coherent scattering (DRCS) for major phases. The coherent scattering regions (RCS) for
produced powders were determined on the data of roentgen lines broadening at small angles.
2.3
Determination of powder composition
A composition and surface properties of nanopowders produced under various

conditions were studied by mass-spectroscopy techniques at the temperatures of 0 450 0C,
electron and X-ray diffraction methods (XRD).
The phase analysis and determination of the fine crystal structure parameters of powders
have been implemented by means of XRD apparatus DRON-UM1 with filtered copper
radiation K. A relative error in determination of the phase constitution in the powder was
under 10% and in determination of the parameters it was under 0.1%.
Active aluminum mass content, Al %, was determined by measuring the hydrogen
volume, which was formed in reaction of aluminum with solution of sodium hydrate.
The composition of adsorbed gases has been determined by heating of samples in the
vacuum of 10-7 at various temperatures as well as during linear heating with ramp function
of temperature.
2.4
Investigation of powder activity in oxidation reactions
To determine activities of the powder samples in the combustion reactions, the

Differential Thermal Analysis (DTA), Thermogravimetry (TG), and reaction with hot water
were conducted.
DTA and TG were implemented in air with the heating rate of 5 0C/min.
The special gasometric device was used for determination of gas volume extracted
during the reaction of aluminium powder with hot water. Mixtures 2.7 mg of Al and 10 g of
distilled water were placed in the cell thermostated under 60 0C, and the volume of gas
released has been calculated using the bubble recorder.
In reactions with hot water, aluminium oxidized and is transformed to the oxyhydroxide.
Final product has been analysed using BET and TG methods. TG was implemented in air
with the temperature range of 20 1000 0C and heating rate of 10 0C/min.
2.5
Investigation of combustion of nanoaluminum-gelled water

compositions
Experiments were conducted in a close manometric bomb of 5 liter filled with argon at a
constant controlled pressure. Mixtures were placed in quartz glasses of 10 mm diameter and
35 mm length and were ignited by a nichrome spiral. At least 3 tests or more were performed
for one experimental point with the given conditions.
The combustion rate was determined by two photodiodes with the distance between
them of 20 mm. Both photodiodes were coarsened and adapted through two observation slits
295
Czech Republic
2006
[Content]
to radiation from a combustion wave. A millisecond stopwatch was started by the first
photodiode and stopped by the second one.
3.
THE SIZE OF PARTICLES AND CRYSTALLITES
The following types of powder markings are assumed: the gas used as medium for a
wire explosion is noted first, and the gas used for the powder passivation is noted through
the slant line. For instance, N2/CO2 designation corresponds to the powder produced in
nitrogen and passivated by carbon dioxide etc. The specific surface of powders varies from
16 to 38 m2/g. The particles are predominantly spherical (Figs. 1, 2).
If the particles are spherical, one can determine the mean-surface diameter dS as
d S = 6 / S BET where is the density, for Al = 2.7 g/cm3. These data for some samples are
placed in Table 1 together with the mean diameters D, the standard deviations , determined
from TEM photos, and mean sizes of regions of coherent scattering (DRCS) evaluated from
XRD lines.
60
W, %
N2/CO2
30
0
0
63,6
127,2
190,8
254,4
318
381,6
D, nm
Fig 1.
TEM photo and size distribution of powder sample N2/CO2
60
W, %
(N2+CO2)/ N2
30
0
0
50 nm
Fig 2.
296
40
80 120 160 200 240 280 320

D, nm
TEM photo and size distribution of powder sample (N2+ CO2)/CO2
Czech Republic
2006
[Content]
Table 1. Size of particles and crystallites.

Sample
N2/air
N2/CO2
N2+CO2/N2
SBET,
m2/g
17.5
21.0
34.3
dS ,
D,
nm
nm
127 40.3
106 40.4
65 32.4
DRCS,
nm
nm
16.4 29
6.1 30
4.7 32.5
Detailed comparison between values D, , and DRCS, shows that the particles of
N2+CO2/N2 sample are monocrystal, and comparison between values dS and D shows that
the particles of this sample are slightly agglomerated. Ideal case, dS = D, may be observed
with oxide or nitride particles because particles of pure metals are easy agglomerated during
preparation and store.
4.
CONSTITUTION OF POWDER SAMPLES
Different conditions of powder synthesis by the exploding wire method and powder
passivation were used: synthesis in low-pressure gases (Ar, N2, CO2), passivation in
atmospheric-pressure gases (air, N2, CO2).
A constitution of gas shells and desorption dynamics were investigated by using the
mass spectrometry method and by measuring the pressure of releasing gases in the interval
of 0 450 C on heating in vacuum at constant temperatures and under the fast (to 50
K/min) linear dynamic heating.
The composition and some properties of powder samples produced and passivated in
various gas mediums are summarized in Table 2.
Table 2. Results of Al nanopowders analysis
Sample
N2/N2
N2/CO2
(N2+ CO2)/N2
S,
m2/g
18.2
21.0
34.3
Al, Water, Sum of gases,

mass %
%
mass %
75,4
6,4
1,7
70,7
0,4
2,9
60.7
1.5
5.2
XRD, volume %
Al Al2O3 AlN Al N5O14
96
4
90
5
~5
91
9
The percentage content of pure aluminum (Al %) was determined by a volumetric

method in which the volume of hydrogen, as a result of reaction of Al with solution of
sodium hydrate, has been measured. The percentage volume contents of Al, Al2O3, AlN, and
Al N5O14 were determined by XRD method.
It was founded that a composition of desorbed gas components is highly changed with
the temperature increase and depends on the surroundings of preparation and passivation.
Tables 3 and 4 give the composition of gases desorbed in vacuum under different
temperatures.
Kinetics of gas evolution shows two steps (Fig. 3).
297
Czech Republic
2006
[Content]
Table 3. Sample N2/CO2: Composition of desorbed gases under different

temperatures (T0C), volume %
T, 0C
Gas
N2
O2
Ar
CO2
H2 O
NH3
50
100
150
200
250
300
350
400
73.5
8.2
0.0
2.3
16.1
0.0
79.6
8.0
0.0
1.6
11.8
0.0
88.4
5.5
0.02
0.4
5.7
0.0
91.1
4.7
0.01
0.3
3.9
0.0
82.0
8.8
0.0
0.2
6.8
1.2
64.7
5.6
0.0
0.1
23.0
9.7
86.7
4.9
0.0
0.0
6.3
2.1
96.0
1.9
0.0
0.0
2.1
0.0
82.8
15.5
0.0
0.0
1.7
0.0
Table 4. Sample CO2/CO2: Composition of desorbed gases under different

temperatures (T0C), volume %
T, 0C
0
50
100
150
200
250
300
350
400
35.3
11.8
0.0
48.8
4.1
15.6
4.7
0.0
73.3
7.4
26.0
6.4
0.1
57.1
6.4
15.8
6.0
0.2
72.8
5.2
13.3
3.6
0.05
67.0
16.0
9.5
2.6
0.0
53.8
34.1
6.7
1.8
0.0
73.1
18.3
4.9
1.3
0.0
92.1
1.7
4.1
0.8
0.0
92.3
2.8
1,0
0,8
Relative Intensity
Gas
N2
O2
Ar
CO2
H2 O
0,6
0,4
0,2
T, C
0
Fig 3.
100
200
300
400
500
Kinetics of gas evolution
Gas inclusions (air components, water, CO2, et al.) that are the most loosely-bounded
with the surface of particles as well as argon adsorbed after completion of the electrical
explosion, are removed at 100 150 C along with the external particle shell formed in
passivation. The next portion of gases chemically bounded to the surface is desorbed at 250
400 C (Figure 3). The temperature of desorption finishing is about 400 0C.
298
5.
Czech Republic
2006
[Content]
OXIDATION OF POWDERS IN REACTIONS WITH AIR AND

WATER
5.1
DTA and TG
Table 5 demonstrates the data of beginning of the oxidation temperature Tonset during
DTA in air for different powders with the heating rate of 5 0C/min.
Table 5. Comparison of data of oxidation beginning temperature during DTA in air
Sample
Al,
SBET,
D,
Tonset,
m /g
nm
N2/N2
75,4
18.2
44,2
375
N2/air
74
17.5
40,3
400
N2/CO2
70,7
21.0
40,4
425
N2+CO2/ N2
60.8
34.3
32.4
397
CO2/CO2
16,4
38.0
30,8
400
Ar/air
79,8
16.7
54,6
420
It can be seen from Table 5 that Tonset is independent of powder fineness (size varies
twice and purity varies almost 5 times). For all cases the temperature of oxidation beginning
is practically the same: Tonset = (400 25) 0C.
It seems that the temperature of oxidation beginning corresponds to the finishing
temperature of gases desorption. The two processes are connected with each other. Adsorbed
and chemically bonded gases prevent particles from oxidation in air. Oxidation in water
gives another results.
5.2
Oxidation in reaction with hot water
Correlation between conditions of nanopowder synthesis, the specific surface and

chemical activity in reaction with hot water has been determined.
In this study, the temperature of reaction onset in independent experiments on evolution
of the first hydrogen bubbles, a pH change, and some self-heating of the mixture was
registered. Some results are presented in Table 6.
Table 6. Reaction with hot water
No
Sample
1
2
3
N2/air
N2+CO2/ N2
CO2/CO2
SBET
sample,
m2/g
13
34.3
38
of Al
content,
%
74.0
60,8
16.4
,
before/after
reaction
4.96/9.71
5.07/9.41
5.2/8.2
SBET
products,
m2/g
368
342
165
of
299
Czech Republic
2006
[Content]
The Table 6 shows that there exists the direct relation between the amount of active
aluminum content and the specific surface of final product. An ultrafine aluminium
oxyhydroxide with the specific surface SBET = 368 m2/g was produced by oxidation of
nanosized aluminium by hot water.
Fig. 4 demonstrates the hydrogen release during reaction with water at 60 0C.
Volume, mm3
1500
1200
900
600
1
300
0
0
500
1000
1500
2000
2500
Time, s
Fig 4.
Kinetics of hydrogen realise in oxidation of Al samples 1, 2, 3 (Table 6)
It can be seen from Fig. 4 that various types of Al nanopowders oxidized in two stages.
TG analysis of final products shows the aluminum oxyhydroxide dehydration. This result
allows to suppose that the reaction 2Al + 4H2O 2Al OOH + 3H2 is limitative for samples 1
and 2. As for sample 3, Table 6, alumina probably comes together with preudoboehmite in
less alkaline medium (pH 8.2).
The oxidation level of Al powders in reaction with hot water (60 0C), % masses degree
of oxidation, and the temperature of reactions onset, T 0C, are given in Fig 5.
Degree of Oxidation, %
Temperature of teaction onset, C
60
100
55
80
1
2
60
50
45
40
40
20
35
14 16 18 20 22 24 26 28 30 32 34 36 38 40
2
S, m /g
Fig 5.
300
The o degree of oxidation (1), % masses, and the temperature of reactions

onset, T 0C (2)
Czech Republic
2006
[Content]
An oxidation level in reaction with water is from 36 to 100% for various types of
powders and their specific surface.
The more active is the powder, the lower is the temperature of reaction onset with water
and the higher is the oxidation degree at 60 C. The experimental data on relative chemical
activity show that the value of specific surface is the major characteristic of powder.
6.
COMBUSTION OF NANOALUMINUM / GELLED WATER

COMPOSITIONS
6.1
Preface
The possibility of combustion of the nanoaluminum with water in the mode of a normal
combustion regime has been established in the experimental work [6]. The calculated heat of
combustion of the Al nanopowder stoichiometric mixture with water, recounting to the pure
metal (50/50 %), is 7570 kJ/kg thus attributing this system to high-energy material. Owing to
inert admixtures in the powders, this value is vastly less (by 10 15 %).
In the paper [5] the more detailed data on combustion of the stoichiometric aluminum
mixture with water is given. In particular, the law of combustion rate of the stoichiometric
mixture is: U = 0.183 (p/p0)0.4, cm/s. Also the incompleteness of aluminum combustion is
observed.
Report [8] shows, that, for using the aluminum/water mixture as propellant, the ratio of
40% Al/60 gelled water is preferable. In this case, the completeness of Al combustion
reaches 100 %. The combustion temperature and the combustion rate increase with
increasing the oxidation degree. The combustion rate values under the pressure of 7 MPa for
the aluminum concentrations from 38 to 57 % are from 0.8 to 1.8 cm/s. The temperature of
combustion reaches 2450 0K.
The present study is a continuation of
water 40/60 %.
6.2
[5-8]
, and we used the optimal ratio aluminum/
The combustion rate
Figure 6 demonstrates clearly how the concentration and dispersivity of Al powder in

Al/water slurry influence the combustion rate.
The first curve in figure 6 shows the dependence of combustion rate on pressure for the
sample of Ar/air having the specific surface of 16.7 m2/g (Table 5), in the stoichiometric
50/50 % composition of Al/gelled water.
Investigations confirmed the early-represented data [8] showing that aluminum
decreasing in the composition of 40/60 % increases the completeness of metal combustion
and increases the combustion rate (curve 2). The sample of Al powders was the same
Ar/air.
Curve 3 also corresponds to the mixture of 40/60 %, but the specific surface of
nanoaluminum was of 34.3 m2/g (sample N2+CO2/ N2, Tables 1, 2, and 5). The combustion
rate grew sufficiently and reached 2.4 cm/s under the pressure of 5 8 MPa.
It is important to point out that the dependence of combustion rate on pressure is very
weak (line 3), the exponent in the combustion law = 0.12, and is even independent of
pressure in the area of 5 8 Mpa. That area of pressure is the working range for propellants.
Visual observation of burning surface and analysis of combustion products point to the facts
301
Czech Republic
2006
[Content]
that burning is more uniform and efficient, and solid products are less agglomerated
consisting of nanosized corundum.
2
U, cm/c
1
2
3
P, MPa
Fig 6.
7.
7 8 9 10
(1) sample Ar/air, SBET= 16.7 m2/g, stoichiometric 50/50 % composition,

(2) sample Ar/air, composition of 40/60 % Al/water, (3) sample
N2+CO2/ N2, SBET =34.3 m2/g, composition of 40/60 % Al/water
CONCLUSION
Because of peculiar chemical features of the Al nanopowders, the oxidation degree of

the powder by water and the temperature of the reaction onset can be used as a criterion to
compare chemical activity of the powders produced under various conditions.
Phase composition and phase structure of powders produced under different conditions
of heating and environment are overwhelmingly important. Stability of phase constitution of
particles as well as stability of their energetic and other characteristics during storage time
and their behavior under external influences are also sufficient.
Inorganic coatings allow producing powders having monocrystal nanosized particles.
Though pure metal content averages out to 75 %, powders demonstrate high energetic
activity in oxidation reactions due to high dispersivity of powders.
Compared to conventional Alex [3], the specific surface of powder produced using
justified engineering conditions is three times higher. This allowed sufficiently improving
characteristics of gelled Al/water compositions, namely, the combustion rate became two
times higher and less agglomeration has been observed.
Adsorbed and chemically bonded gases prevent particles from oxidation in air. The
temperature of oxidation beginning corresponds to the finishing temperature of gases
desorption. Oxidation in water gives another results.
An ultrafine aluminium oxyhydroxide with the specific surface SBET = 368 m2/g was
produced by oxidation of nanosized aluminium by hot water. From the other hand, an
oxidation with water in combustion regime allows producing nanosized corundum.
Important is that the dependence of combustion rate on pressure is very weak, and is even
independent of pressure in the area of 5 8 Mpa.
It has been established that the combustion rate sufficiently grows when a high-dispersed
powder is used.
302
Complex Research of Al Nanopowders Produced by the Exploding Wire . . .
Czech Republic
2006
[Content]
In conclusion we can say that the exploding wire method is a very promising method of
ultra-fine powders production. It is possible to produce ultra-fine powders with controlled
phase compositions.
Real future success requires the coordinated joint efforts of a wide circle of researchers
and technologists, producers and users.
REFERENCES
[1]
L.I. CHEMEZOVA, V.G. IVANOV, YU.F. IVANOV, M.N. OSMONALIEV, V.S. SEDOI, I.A.
SOBOLEV: Production of Nano-Al Powders by the Exploding Wire Method, Proceedings of
International Conference on Combustion and Detonation, Zeldovich Memorial II, Moscow,
paper WS1-4 pp. 1-58, 2004
[2] L.T. DE LUCA, L. GALFETTI, F. SEVERINI, L. MEDA, G. MARRA, A.B. VOROZHTSOV, V.S.
SEDOI, AND V.A. BABUK: Burning of Nano-Aluminized Composite Rocket Propellants,
Combustion, Explosion, and Shock Waves, Vol. 41, No.6, pp. 680-692, 2005
[3] F. TEPPER AND L. KALEDIN: Combustion Characteristics of Kerosene Containing Alex NanoAluminum (Combustion of Energetic Material, Edited by K.K. Kuo and L.T. De Luca), Beggel
House Inc., New York. Wallingford, pp. 195-206, 2002
[4] B. PALASZEWSKI: Metallized gelled propellants: historical and future developments, (Rocket
Propulsion: Present and Future, L.T. De Luca, Eds.), Grafiche GSS, Bergamo, Italy, pp. 32(119), 2002
[5] V.G. IVANOV, S.N. LEONOV, G.L. SAVINOV, O.V. GLAZKOV, O.V. GAVRILUK: Combustion of
Mixtures of Ultrafine Aluminum With Gel-Like Water, Combustion, Explosion, and Shock
Waves, 30, pp. 569-570, 1994
[6] V.G. IVANOV, O.V. GAVRILYUK, O.V. GLAZKOV, and M.N SAFRONOV: Specific Features of
the Reactions Between Ultrafine Aluminum and Water in a Combustion Regime, Combustion,
Explosion, and Shock Waves, 36, pp. 213-219, 2000
[7] V. SEDOI, V. IVANOV, YU. IVANOV, M. OSMONALIEV, M. LARICHEV, and O. LARICHEVA:
Synthesis by the Exploding Wire Method and Characterization of Nanoaluminum Powders with
tInorganic Coatings, Proceedings of the 36th International Annual Conference of ICT,
Karlsruhe, Federal Republic of Germany, paper 98, pp. 1-11, June 28-July 1, 2005
[8] L. CHEMEZOVA, V. IVANOV, YU. IVANOV, M. OSMONALIEV, V. SEDOI, and I. SOBOLEV:
Complex Research of Al Nanopowders Produced by the Exploding Wire Method, European
Conference for Aero-Space Sciences, July 4-7, 2005, Moscow, pp. 5.3(1-7)
[9] YU. F. IVANOV, M,N. OSMONALIEV, V.S. SEDOI, et al.: Production of Ultra-Fine Powders
and Their Use in High Energetic Compositions, Propellants, Explosives, Pyrotechnics, Vol. 28,
No. 6, pp. 319-333, 2003
[10] V.S. SEDOI, YU.F. IVANOV, and M.N. OSMONALIEV: Study of Ultra-Fine Powders Produced
by the Exploding Wire Method (Novel Energetic Materials and Applications, L.T. De Luca, L.
Galfetti, and R..A. Pesce-Rodrigues, Eds.), Grafiche GSS, Bergamo, Italy, pp. 16 (1-10), 2004
[11] V.S. SEDOI, V.V. VALEVICH: Production of Highly Dispersed Metal Powders by Electrical
Explosion in Reduced-Pressure Nitrogen, Tech. Phys. Lett. Vol. 25, pp. 584-585, 1999.
[12] V.V. VALEVICH, V.S. SEDOI: Ultrafine powders obtaining by the fast electrical explosion,
Russian Physics Journal, No. 6, pp. 70-76, 1998
Acknowledgment:
This work was performed with financial support by INTAS (grant 03-53-5203)
11
303
Czech Republic
2006
[Content]
AB INITIO MD COMPUTATION OF THE VIBRATIONAL

RELAXATION TIME IN HE MOLECULES
A.A.Selezenev, A.Yu.Aleynikov
RFNC-VNIIEF, Sarov, Russia
Abstract:
Ab Initio MD simulation has been used to compute equilibrium time for the
intramolecular vibrational degrees of freedom in molecules of TNT, RDX, HMX and
PETN. Computation of intramolecular forces field was done utilizing the code
Gaussian with HF/STO-3G method.
Ab Initio MD simulations have been carried out for the decomposition of TNT and RDX
molecules as they collide. For the kinetic energy of molecular collisions, its threshold
values have been found where decomposition of the molecules should occur before the
equilibrium distribution of collision energy into vibrational degrees of freedom.
INTRODUCTION
For kinetic modeling of the HE decomposition by detonation, it is important to find the
conditions that can make direct scission of the chemical bonds occur in HE molecules [1-5].
One possible mechanism of a chemical bond scission in the shock-wave front may be by
direct transfer of a kinetic energy to this bond that is larger than its dissociation energy [1, 4, 5].
Then, if the intramolecular vibrational relaxation time vv is more than the characteristic
time ( X ) of kinetic energy transfer to the bond in question, the energy going to the bond
may be actually higher than the dissociation energy. The occurrence of such mechanism has
been proved to be possible by direct MD simulations of the HE molecule decomposition in
shock-wave front. Thus, the studies using MD simulation [6, 7] have observed for strong
shock waves that the time of chemical bond scission in HE molecules may be of
picoseconds scale. Based on this mechanism, the bond scission essential requirements can
be written as:
Ek > 2 U 0 ,
(1)
X < VV ,
(2)
X < TV ,
(3)
where - E k is the kinetic energy of molecular collision, U 0 - bond dissociation energy,

VV - intramolecular V-V relaxation time, X - characteristic time of kinetic energy
transfer to the bond in question, TV - natural period of the bond ( TV = 2 / 0 ).
Requirement (3) is that the interaction of a colliding molecule and the given vibrational
mode of its partner molecule should be nonadiabatic. There can be efficient transfer of
translational energy to the vibrational mode, provided Massey parameter (Me) is on the order
of or less than unity [6]. In the case of interest, the relation used for Massey parameter is
304
Ab Initio MD Computation of the Vibrational Relaxation Time in HE . . .

Me =
Czech Republic
2006
[Content]
X X 0
. There is a correlation established between the values in requirements

TV
2
(1)-(3) and HE detonation parameters. From (1)-(3), obtaining such correlation evidently
requires calculations of the HE molecule vibration spectra, bond dissociation energy (BDE)
and V-V relaxation time.
1.
COMPUTING VIBRATIONAL SPECTRA AND NO2 BOND ENERGY

OF HE MOLECULES
Thus, infrared spectra and bond dissociation energy have been calculated for the
molecules of six HE formulations: TNT, RDX, HMX, PETN, TATB and NM. These IR
spectrum and BDE calculations used Gaussian code [7]. The IR spectra were calculated for
preoptimized structure of the molecules. The molecules were optimized in structure and their
spectra calculated by density functional technique (using B3LYP/6-311++g** basis set).
Table 1 shows comparison between the vibration wave numbers of identical fragments
in different HE molecules. The Table gives wave numbers of the vibration modes of
molecules (cm-1) and the molecular fragments to which the vibration modes are related.
Table 1.
Typical vibration frequencies of HE molecules
HE
Fragment\
RDX
1658.71,
1657.96
Nitromethane
1622.88
PETN
TNT
HMX
1747.39,
1735.1,
1729.83
1602.52
1638.88,
1652.77,
1623.38
1589.02
1645.95
905.17,
905.514
918.015,
930.943,
904.622,
996.499
841.648,
844,42,
835.659
1345.04
1285.82,
1303.01,
1310.96,
1245.03,
1339.05
1291.81,
1293.3
1314.36,
1323.37
1369.64,
1379.77
As indicated by the calculations summarized in Table 1, the vibration modes of identical
fragments in different HE molecules have similar frequency values.
Alexander Selezenev, Aleksey Aleynikov
305
Czech Republic
2006
[Content]
For accuracy test of the applied numerical method, the calculations of the current study
were compared with the experimental wave number values of some vibration modes in
nitromethane [10]. From this comparison, the calculated wave numbers are found to fit the
experimental values with the relative accuracy of 0.6% or better.
The dissociation energy calculations for -NO2 N-NO2 bonds were also performed
using B3LYP/6-311++g** basis set. The value of bond dissociation energy was calculated as
a difference between the total energy of a molecule in initial state (with allowance for zero
vibration energy) and that of the dissociation products. For RDX molecule, for instance, the
expression for BDE calculation of N-NO2 is
BDE=E (C3H6N6O6) - [E (C3H6N5O4) + E (NO2)].
The BDE calculation was also conducted to find the bond energy dependence on
interatomic distance. The bond energy relation to interatomic distance was approximated by
Morse potential
U (r ) = U 0 [e 2 ( r r0 ) 2e ( r r0 ) ] .
Table 2 shows parameters of Morse potential obtained for R-NO2 bonds in the different
HE molecules. Table 2 also compares the calculated BDE values with the respective
calculations in [11, 12].
Table 2. Morse potential parameters
HE
Bond type
0
U 0 , eV
, 1/ A
(BDE)
TATB
3.8316
2.1483
C NO2
TNT
3.0804
2.2642
C NO2
NM
4.7198
1.5777
C NO2
HMX
1.8937
2.8467
N NO2
RDX
1.7432
2.8351
N NO2
PETN
1.7845
2.6684
O NO2
It is of interest to consider these bond energy
molecules in the shock front of detonation wave:
Ek =
r0 , A
BDE, eV
[11]
BDE, eV
[12]
1.4224
3.35
3.00
1.4484
2.90
2.55
1.5533
1.3799
1.86
1.3922
1.80
1.4082
1.73
values versus the kinetic energy of HE
M U s2
= 0.5182 10 2 U s2 , = M / mu ,
2
where E k is the kinetic energy of molecule, eV, - relative molecular mass, mu atomic mass unit (= 1.6605655 10 27 kg), U s - molecular collision velocity, km/s.
Considering that no more than the kinetic energy of collision [14] can change into
internal energy at molecular collision, Fig 1 makes the comparison with BDE easier as it
gives the value of 0.5Ek as a function of relative collision velocity.
306
Czech Republic
2006
[Content]
8
-1
-2
6
0.5Ek, eV
-3
-4
4
0
0
0.5
1.5
2.5
3.5
U, /
Fig 1. Kinetic energy of collisions as a function of collision velocity for HE molecules.

(1-PETN, 2-HMX, 3-TNT, 4-RDX).
Comparing the results given in Table 2 and Fig 1, one can see for the collision velocity
of HE molecules at U>2 km/s, that kinetic energy of molecular collision can be larger than
NO2 bond dissociation energy.
2.
COMPUTING INTRAMOLECULAR VIBRATIONAL RELAXATION

TIME IN HE MOLECULES
The following procedure was used for computation of intramolecular V-V relaxation
time in HE molecules. Prior to this, the molecule was structurally optimized using
B3LYP/6-311++g** basis set. For this optimized structure, one of R-NO2 bonds there was
brought out of equilibrium. Thus, initial disturbance of the bond length was specified. The
increase in potential energy of the bond was U 1 (U 1 1 2 eV ) . This was followed by MD
simulation of the energy redistribution by vibration modes of the molecule in time. The
intramolecular force field was calculated using Gaussian code, and SageMD [13] code was
used for MD simulations and data processing. During MD simulation, there were
computations done to find kinetic energy for each atom in molecule and the total kinetic
energy of all atoms (less translational energy of molecule).
A typical kinetic energy total of atoms with time is given in Fig 2 for RDX molecule.
RDX
0.9
0.8
0.7
Ekin [eV]
0.6
0.5
0.4
0.3
0.2
0.1
0
0
10
Tim e [ps]
Fig 2. The kinetic energy of atoms in RDX molecule with time.

4
307
Czech Republic
2006
[Content]
One way to estimate intramolecular V-V relaxation time is as follows. In fact, relative
variance of total kinetic vibration energy can be shown to be inversely proportional to
excited vibration modes of the molecule
DEK
EK
1
,
NK
(4)
where DEK is the dispersion of total kinetic vibration energy, EK
- the average of
total kinetic vibration energy, N K - excited vibration modes of the molecule.

Given relationship EK = U1 / 2 , expression (4) can be written as
4 DE
1
or N K (t ) U12 / 4 DE .
2
U1
NK
(5)
Thus, by dividing the analyzed time interval into subintervals << VV and calculating
the kinetic energy variance on each time interval , a relationship can be obtained for the
number of excited vibration modes versus time. Obviously, relationship (5) should yield the
total of molecular vibration modes for >> VV
U 12
= N K
4DE
>> vv
( N K = 3N 6) ,
where N number of atoms in molecule.

With the above discussed approach, it is possible to make lower estimate of vv
parameter, i.e. to find its minimum value. The actual value of vv could be only larger than
U 12
the value as obtained thereby. Table 3 includes the values obtained for 4DE
and the
ff vv
calculated values of
vv .
Table 3. Calculated vibrational V-V relaxation times

N
NK
U 12
4DE
ff vv
vv , ps
TNT
21
57
59
RDX
21
57
60
HMX
28
78
82
PETN
29
81
82
2.5
Parameter
HE
The MD simulations have found the excitation of different vibration modes in HE

molecules to be essentially non-simultaneous. One might expect (and simulations have
308
Czech Republic
2006
[Content]
proved it), that the vibration modes at near initially excited mode frequency would have
small relaxation times. Evidently, there would be resonant excitation of this vibration mode.
And conversely, the relaxation time would be larger for the vibration mode whose natural
frequency is much different from that of excited vibration mode. This effect is illustrated
below using TNT molecule as an example (Figs 3-8). Figs 3 and 4 show the bond length and
kinetic vibration energy with time for the initially excited -NO2 bond in TNT molecule.
Figs.5 and 6 show similar data for the -NO2 bond that was initially unexcited.
From the data in Figs 5-6, the vibrations of -NO2 bond are evidently excited without
any delay.
Figs 7-8 provide the bond length and kinetic vibration energy with time for the initially
unexcited C-H bond in TNT molecule.
From the data in Figs 7-8, excitation of - bond occurs with significant delay in time.
3.
MD SIMULATION OF HE MOLECULE COLLISIONS
The simulations of bond scission by collision were conducted for TNT and RDX
molecules. These simulations were performed using PM3 method. Collision geometry was
considered, which was most effective to scission of R NO2 bond (R=C for TNT and R=N
for RDX molecule). Orientation of the molecules was made such that one of R NO2 bonds
would be directed collinearly with the collision. Initial position of TNT molecules in space
and the configurations of molecules after collision are shown in Fig 9.
Fig 9. Fragmentation of TNT molecules after collision at 8.25 km/s velocity.

6
309
Czech Republic
2006
[Content]
The simulations have found a threshold velocity of collision - the point where scission of
the bonds, -NO2 and N-NO2, would occur as the molecules collide. For TNT and RDX
molecules, these values were 8.1 and 7.6 km/s, respectively. The value of Massey parameter
for threshold velocity of collision was 0.5.
Thus, as it has been observed by direct numerical MD simulations, the scission of
-NO2 and N-NO2 bonds in the HE molecules colliding with each other can occur, with
probability of near 1, if Massey parameter is ~0.5 (obviously, provided E k > 2 U 0 energy
criterion). The bond dissociation energy data have been used to calculate the particle velocity
U S for Massey parameter to be equal to 0.5. For RDX, HMX and PETN molecules, these
particle velocity values were found to be 7.58, 7.55 and 7.53 km/s, respectively. The
observed values for particle velocity are 2 or 2.5 times the particle velocity of substance at
Neumann peak of strong detonation waves. Given the law of particle velocity doubling upon
the shock wave reaching the free surface, one can expect the analyzed bond scission
mechanism to occur, e.g., when the shock wave is interacting with micropores of the HE
lattice. The high molecule velocities can be also achieved due to dispersion of the particle
velocity at the shock wave front.
CONCLUSION
Infrared spectra and dissociation energy of NO2 bonds have been calculated for six HE
formulations. Calculation of the kinetic energy distribution by intramolecular vibration
modes has been also performed for the HE molecules of interest (V-V-relaxation time).
Molecular dynamics simulation has been carried out to study decompositions of TNT and
RDX molecules by collisions of the molecules with each other. Threshold collision
velocities have been found, when molecular decomposition should occur straight at collision.
The study also includes estimates that have been made for the probability of
decomposition for HE molecules colliding in detonation shock front. One candidate
decomposition behavior, a so-called accumulation mechanism, is discussed in which the
kinetic energy of collision is assumed to have possible location on a molecular bond. The
analysis suggests that the occurrence of such mechanism in the detonation shock front of
condensed HE is limited by the following requirements:
1. Distribution of the kinetic energy of ordered molecular motion into intermolecular
vibrations energy should be much shorter in time than it takes to achieve balance
with the energy of intramolecular vibrations. Then, the relative collision velocity of
molecules would have the same order of magnitude as the HE particle velocity.
2. With energy criterion observed, the bond dissociation probability is substantially
dependent on Massey parameter, which in turn is determined by collision time and
the vibration period of analyzed chemical bond.
The authors would like to express their thanks to Dr. V.N.Lashkov for his support to the
study and valuable discussions in analysis of the results.
REFERENCES
[1]
310
Klimenko V.Yu, Dremin A.N. On the decomposition reaction kinetics in the shock wave
front, In A.G. Merzhanov, editor, Detonation, chemical Physics of combustion and explosion
processes, Russia, Chernogolovka, 1980, Nauka.
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Czech Republic
2006
[Content]
Klimenko V.Yu., Yakoventsev M.A., Dremin A.N. Multiprocessor detonation model,

Khimicheskaya Fizika, 12, 5, 1993, pp.671-680.
Tarver C.M., Chemical Energy in Self-Sustaining Detonation Waves in Condensed
Explosives, Combustion & Flame, Vol.46, 1982, pp. 157-176.
Walker F.E., Physical kinetics, J. Appl. Phys. Vol.63 (11), 1988, pp.5548 5554.
Walker F.E., Quantum Mechanics and Molecular Dynamics Calculations Provide New
Evidence for a Free-Radical Shock Initiation Model, Propellants, Explosives, Pyrotechnics
Vol.7, 2-7, 1982.
Manaa M.R., Fried L.E., Melius C.F., Elstner M. and Frauenheim Th., J. Phys. Chem. A 106,
2002, p.9024.
Alejandro Strachan, Adri C.T. van Duin, Debashis Chakraborty, Siddhard Dasgupta, and
William A. Goddard III, Shock Waves in High-Energy Materials: The initial Chemical Events
in Nitramine RDX, Physical Review Letters, Vol. 91, 9, 2003, pp. 09801-1 09801-4.
Zeldovich Ya.B., Raizer Yu.P. Physics of shock waves and high-temperature hydrodynamic
events. 2nd edition, M.: Nauka, 1976.
Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M.
Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.
A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida,
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B.
Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P.
M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, and J. A. Pople, Gaussian, Inc.,
Wallingford CT, 2004.
Zerilli J.F. and Toton E.T., Shock induced molecular excitation in solids, Physical Review,
B, Vol. 29, 10, 1984, pp. 5891 5902.
Wu C.J. and Fried L.E., First-Principles Study of High Explosive Decomposition Energetics,
Eleventh Symposium (International) on Detonation, Naval Surface Weapons Center NSWC 86194, Albuquerque, NM, 1998, p.490.
Rice B.M, Sahu S., Owens F.J., Density functional calculations of bond dissociation energies
for NO2 scission in some nitroaromatic molecules, Journal of Molecular Structure
(Theochem), Vol.573, 2002, pp.69-72.
Selezenev A.A.; Aleynikov A. Yu.; Gantchuk N.S.; Yermakov P.V.; Labanowski J. K.
Sage MD: Molecular Dynamics Software Package to Study Properties of Materials with
Different Models for Interatomic Interactions Computational Materials Science, Vol. 28,
Issue 2, October 2003, pp. 107-124.
.. . .:
, 1976. (Nikitin E.E. Theory of elementary atomic-molecular behaviors in gases. M.:
Khimia, 1976.)
311
Czech Republic
2006
[Content]
APPENDIX
1.4
2.6
1.2
2.4
2.2
Ek, eV
R, A
2
1.8
1.6
1.4
0.8
0.6
0.4
1.2
0.2
1
0
t, ps
0
0
t, ps
Fig 3. Bond length with time.

(initially excited -NO2 bond in TNT
molecule)
312
Fig 4. Kinetic vibration energy for initially excited

-NO2 bond.
R, A
0.5
1.5
2
t, ps
2.5
4
t, ps
3.5
4.5
0.5
1.5
2
t, ps
2.5
3.5
4.5
Fig 8. Kinetic vibration energy

for initially unexcited -H bond.
0
t, ps
0.05
0.1
0.2
0.15
0.25
Fig 6. Kinetic vibration energy for initially unexcited NO2 bond.
2006
Fig 7. C-H bond length with time

(initially unexcited -H bond).
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
Czech Republic
0.8
1.2
1.4
Fig 5. Bond length with time

(initially unexcited -NO2 bond).
1.2
1.3
1.4
1.5
1.6
1.7
1.8
Ek, eV
Ek, eV
10
R, A

[Content]
313
Czech Republic
2006
[Content]
UNUSUAL THERMAL DECOMPOSITION OF NTO: IS IT A RESULT OF

VERY STRONG BONDS OR OTHER REASONS EXIST?
V.P. Sinditskii, S.P. Smirnov, V.Y. Egorshev
Mendeleev University of Chemical Technology
9 Miusskaya Square, 125047, Moscow, Russia
Abstract:
Burn rate characteristics of low sensitive explosive 3-nitro-1,2,4-triazol-5-one (NTO) have
been investigated in pressure interval of 0.1-40 MPa. The temperature distribution in the
combustion waves of NTO has been measured at pressures 0.4-2.1 MPa. Based on burn rate
and thermocouple measurement data, rate constants of NTO decomposition in the molten
layer at 370-425C have been derived from a condensed-phase combustion model
( k = 1.14 1014 exp(39000 / RT ) , s-1), and NTO vapor pressure above liquid phase (lnP = 9914.4/T + 17.12) and solid phase (lnP = -12984.4/T + 22.78) has been calculated.
Taking into account vapor pressure data obtained, decomposition of NTO in the gas phase
at 240-250C has been studied. Comparison of decomposition rate constants in gaseous and
solid state shows that they are close together, therefore, partial decomposition in the gas
can not substantially increase the total rate. High values of the activation energy for solidstate decomposition of NTO are not likely to be connected with submelting effect, because
decomposition occurs at temperatures well below melting point. It has been suggested that
abnormally high activation energy in the interval of 230-270C is stipulated by peculiarities
of NTO transitional process rather than strong bonds in the molecule. In this area, NTO
molecule undergoes isomerization into the aci-form, followed by C3-N2 heterocyclic bond
rupture, resulting in abnormally high value of the observed activation energy.
3-nitro-1,2,4-triazol-5-one (NTO), burn rate, thermal decomposition,
Keywords:
temperature profiles in the combustion waves, mechanism of decomposition
1.
INTRODUCTION
The thermal decomposition of insensitive, high explosive 3-nitro-1,2,4-triazol-5-one (NTO)

was studied widely.[1-12] Various possible mechanisms were suggested, such as a direct scission
by the C-NO2 bond or initial hydrogen transfer with subsequent C-NO2H cleavage. Bimolecular
mechanism producing CO2 and mechanism involving NO2 rearrangement to ONO followed by
direct loss of NO were also reported. The activation energies of NTO ranged from 40.7 kcal/mol
in the interval of 110-140C to 77-88 kcal/mol in the interval of 200-260C. In spite of many
theoretical studies[13-18] proposing that high activation energies are connected with a very strong
C-NO2 bond, one may conclude that the description of the NTO decomposition processes is still
far from perfect. This is particularly obvious from the fact that activation energies of NTO
decomposition in acetone and water solutions (38.4 and 48.8 kcal/mol)[9] are far less than those
in the solid state.
314
Unusual Thermal Decomposition of NTO: Is It a Result of Very Strong Bonds . . .
Czech Republic
2006
[Content]
In order to resolve this large discrepancy in experimental results, smaller values of the
Arrhenius parameters were suggested in work[6] to be a result of predominant or partial
sublimation. Thermogravimetric analysis and differential scanning calorimetry[12] showed that
heating of NTO lead to sublimation competitive with the condensed-phase exothermic
decomposition. Sublimation occurred most readily in open vessel conditions and was described
by low activation energy.
In resent works of Kondrikov et al.,[19-21] the rate constants of NTO initial decomposition
stage were described by two straight lines intersecting at 230C. Within the interval of 230260C, the decomposition rate showed a very high temperature sensitivity (activation energy of
98 kcal/mol), a result observed in many studies. However in the interval of 200-230C, the
activation energy of decomposition was found to be substantially less (42.7 kcal/mol) and was
in a good correlation with data of work[3] in the interval of 110-140C (40.7 kcal/mol).
Experiments in the works[19-21] were conducted in closed glass manometers of the Bourdon
type that did not allow explanation of obtained activation energy by sublimation process. The
behavior at temperatures below 230C, therefore, was attributed to the first-order reaction of
neat crystalline NTO decomposition. In the high temperature area, according Kondrikov the
decomposition includes four macroscopic stages. It comprises fast initial thermal decomposition
stage, first-order decomposition of crystalline NTO, first-order decomposition of NTO in the
liquid or quasi-liquid layer on the surface of the crystals, and an autocatalytic reaction in this
layer. The fast initial decomposition is attributed to reactions of NTO vapors as well as to the
condensed-phase decomposition of a proposed NTO derivative.
It is well known that faster decomposition reactions in liquid and gas phases can distort
results of solid-phase decomposition.[22] However, effect of sample submelting is generally
observed in the vicinity of melting point and is strongly affected by sample purity. During NTO
decomposition a sharp increase in the rate is observed 40 degrees before melting. Additional
purification does not influence this temperature gap. Besides, kinetic parameters describing
NTO decomposition in the liquid phase[21] are quite different from experimental data on NTO
decomposition in solutions.[9] As concerning effect of the gas phase decomposition one may
note that the decomposition kinetics in the interval of 230-260C observed in works[19,20] at ratio
of the sample mass to the volume of the reaction vessel (m/V ratio) of about 1 - 2 mg/cm3 are
similar to kinetics reported in works[8,9] at m/V ratio of about 5 - 10 mg/cm3, suggesting a weak
role of the gas-phase processes in NTO decomposition.
One may conclude, therefore, that the description of NTO decomposition processes in the
interval of 230-270C still remains far from perfect. Data on decomposition of NTO in the gas
phase are lacking so far.
A previous study of NTO combustion[23] showed that the burning rate was governed by the
condensed-phase chemistry at low pressures. This may allow finding NTO decomposition rate
constants in a high-temperature interval of 350-500, provided detailed investigations of
temperature distribution in the combustion wave are conducted. The purpose of the present
work was to investigate mechanism of NTO decomposition in a wide temperature interval and
different phase states of the material. The kinetic results obtained may be hoped to clarify
mechanism of NTO thermolysis.
Valerii Sinditski, Sergei Smirnov, Valerii Egorshev
315
2.
Czech Republic
2006
[Content]
EXPERIMENTAL
Preparation. NTO was synthesized from semicarbazide hydrochloride and formic acid
followed by a mixed acid nitration.[24] Recrystallisation from hot water gave the target NTO as a
white crystalline solid with a melting point of 269C (decomposition).
Decomposition Study. Manometric experiments were carried out in thin-walled glass
manometer of the compensation type (the glass Bourdon gauge) at 240 and 250C. The ratio of
the sample mass to the volume of the reaction vessel was about 0.1 mg/cm3.
Combustion Study. Burning rates were measured in a window constant-pressure bomb of
1.5-liter volume in the 0.1-40 MPa pressure interval. Samples to test were prepared as pressed
cylinders of 0.98 TMD (1.87 g/cm3) confined in transparent acrylic tubes of 7 mm i.d. prior to
pressing, the material was thoroughly comminuted in order to produce samples with a minimum
possible pore size, thus minimizing the possibility of flame propagation between particles.
Temperature profiles in the combustion wave were measured using fine thermocouples. The
thermocouples were welded from 80%W + 20%Re and 95%W + 5%Re wires 25 m thick
followed by rolling in bands 5-7 m thick. The thermocouple signal was recorded with a Pico
ADC 216 digital oscilloscope.
Thermodynamic calculations were performed with the use of the REAL computer
simulation of chemical equilibrium code.[25]
3.

3.1
Burning behavior and flame structure of NTO
NTO in the form of samples pressed into PMMA tubes of 7 mm diameter starts burning at
0.4 MPa with the rate of 0.3 mm/s. At atmospheric pressure NTO can sustain combustion only
when external heating is continuously provided by an electrically heated nichrome wire. In
doing so, a decomposition reaction with formation of gaseous products occurs, but no gas-phase
flame is observed. The gas flame appears at pressures when NTO is capable of sustained
combustion, but again the burning is accompanied by formation of copious yellow smoke and
solid residue inside the tubes. The NTO burning rate-pressure dependence is expressed as rb =
0.727p0.976, mm/s for the pressure interval of 0.4-11.9 MPa, and rb = 1.56p0.667 for the interval of
11.9-40 MPa.
The burning rates of NTO are close to the burning rates of TNT. As compared with HMX,
NTO burns 2 times slower within the whole pressure interval (Fig 1). Seemingly, it is a rather
low burning rate as well as a low pressure exponent (~0.67) at high pressures that are
responsible for low sensitivity of the substance to mechanical and shock stimuli.
The yellow-brown residue formed was shown by IR-spectroscopy to contain NHCOfragments. An analysis of the gaseous combustion products at 0.5 MPa revealed the presence of
NO, N2O, CO and CO2 species.
316
Czech Republic
2006
[Content]
100
5
HMX
Burning rate, mm/s
3
2
10
5
TNT
3
2
1
NTO
5
3
2
2
6 8 1
6 8 10
Pressure, MPa
Fig 1.
Comparison of burning rates of NTO, TNT and HMX.
2100
T2
T2
2.1
1900
T2
Temperature, K
1700
T2
1.0
1500
0.5
1300
1100
T1
0.4
T1
T1
900
700
500
300
0.0
2.0
4.0
Distance, mm
Fig 2.
Typical temperature profiles for NTO recorded at 0.4, 0.5, 1.1, and 2.1 MPa.
317
Czech Republic
2006
[Content]
1000
1.0
0.5
900
Temperature, K
0.4
800
Tc
Ts
700
Tc
600
Tm
500
400
300
-0.8
-0.4
0.0
0.4
0.8
Distance, mm
Fig 3.
Temperature distribution in the condensed phase and adjacent gas phase at

combustion of NTO.
Typical temperature profiles in the combustion wave of NTO measured with fine -shaped
tungsten-rhenium thermocouples at 0.4, 0.5, 1.0 and 2.1 MPa are shown Fig 2 Fig 3. All
experimental data obtained are collected in Fig 4 and Table 1.
2400
Tad
2200
T2
Temperature, K
2000
1800
1600
1400
1200
T1
1000
Tc
800
Ts
600
400
200
3
Pressure, MPa
Fig 4.
Characteristic temperatures in the combustion wave of NTO: surface temperature

(Ts), temperature of carcass (Tc), temperature of first (T1) and second (T2) flame.
Dotted line is NTO boiling temperature.
Thermocouple-aided measurements allow the following typical zones to be distinguished on

NTO temperature profiles (Fig 2 and Fig 3):
318
Czech Republic
2006
[Content]
preheated zone in the condensed phase from the initial temperature (To) to the first break
on the profiles at temperatures of 510-540 (237-267), which are close to the melting
point of NTO (270C).[26] For this temperature interval, the thermal diffusivity
coefficient () was calculated from Michelsons distribution to be (1.90.2)10-3 cm2/s;
molten layer from the melting point (Tm) to the surface temperature (Ts), which depth
decreases from 0.27 mm to 0.15 mm when pressure increases from 0.4 to 2.1 MPa.
Michelsons temperature distribution yields the thermal diffusivity coefficient in the
liquid phase as (1.70.2)10-3 cm2/s;
adjacent-to-the surface gas zone of less than 0.1 mm thickness, which includes also a
solid carcass made perhaps of NTO polymeric decomposition products. The temperature
gradient in this zone differs little from gradient in the condensed phase. The carcass
boundary is visible on profiles taken at low pressures (0.4-0.5 MPa) as sharp breaks at
temperatures of 780-820 (507-547).
following is aerosol zone, clearly visible at 0.4-0.5 MPa, in which temperature rises from
carcass decomposition/evaporation temperature (Tc) to the first flame temperature (T1).
luminous flame zone with final temperature T2, which appears at a considerable distance
from the surface (~3 mm) at low pressures and approaches fast to the burning surface as
pressure grows (~0.25 mm at 1 MPa), resulting in profiles with one-stage flame structure
at pressures above 2.1 MPa. Oscillating behavior of temperature traces in this zone is
indicative of condensed particles presented in the flame.
The maximum flame temperature on NTO combustion increases from 1500 to ~2000 K as
pressure rises from 0.4 to 2 MPa, approaching to calculated adiabatic one, which is ranging
around 2200 K. This result may be indication of almost full energy release in NTO combustion
at relatively low pressures.
Table 1. Characteristic temperatures in the combustion wave of NTO (Confidence
intervals are given for three and more parallel runs).
Pressure, MPa
Ts , K
T1, K
T2, K
0.4
0.5
0.6
1.1
1.6
2.1
645
63915
624
67710
673
70018
980
103020
1100
1150
1075
-
1500
166761
1750
1780
1975
198020
Adiabatic flame
temperature, K
2223
2224
2224
2224
2225
2225
The dependence of NTO boiling temperature on pressure was calculated from known
thermodynamic equation:
-RT lnKg = (H-TS) evaporation, where Kg = P,
It was assumed that the surface temperatures fit with boiling points of NTO at corresponding
pressures. At pressure of 0.5 MPa it is 63915 K that has been calculated from seven parallel
runs.
319
Czech Republic
2006
[Content]
The heat of evaporation of 19.7 kcal/mole was obtained as the difference between heat of
sublimation of 26.4 kcal/mole and heat of melting of 6.7 kcal/mole. Heat of sublimation
calculated from enthalpies of formation of solid (-30.93 kcal/mole[26] and gaseous (-3.2 and 5.7
kcal/mole[27,28]) NTO, and enthalpy of melting of 6.7 kcal/mole calculated from solubility data
of NTO in nitric acid solutions.[29] Unusually high assessment of enthalpy of melting, 92.7
kJ/mole (22.2 kcal/mole), indicated in work[30] is, undoubtedly, an erroneous value.
The dependence of NTO boiling temperature on pressure is presented in Fig 4. Fig 4 shows a
fairly good correlation between experimental surface temperatures in the interval of 0.4-2.1
MPa and calculated dependence of NTO boiling temperature on pressure (lnP = -9914.4/T +
17.12, pressure in atm) based on thermodynamic parameters and selected temperature at 0.5
MPa.
1E+2
5
3
2
1E+1
5
3
2
Pressure, atm
1E+0
5
3
2
Tm
1E-1
5
3
2
1E-2
5
3
2
1E-3
5
3
2
1E-4
0.0012
2
0.0016
0.0020
0.0024
0.0028
1/T, K-1
Fig 5.
Vapor pressure as a function of reciprocal temperature:

1 (crosses and line) NTO surface temperature, 2 (points and line) vapor
pressure above solid NTO, calculated from[6], 3 (crosses) initial pressures in
glass Bourdon gauge in experiments on NTO decomposition.
The temperature dependence of vapor pressure above solid NTO can be obtained as follows.
The sublimation rate under nonisothermal conditions was determined in the interval of 407-423
K in work.[6] It is evident that sublimation rate depends on rate of heat supply, rate of gas-phase
moving off and vapor pressure under solid substances at a given temperature. Only small
amounts of NTO were used in experiments, suggesting that the rate of heat supply was not a
limiting process. Since there were no data on the rate of gas-phase moving off, the kinetics of
evaporation of solid NTO should not be considered as a fundamental independent-of-device
value. Nevertheless, the temperature dependence obtained is characteristic of substances only,
320
Czech Republic
2006
[Content]
since the rate of evacuation is temperature independent. The value obtained (25.8 kcal/mol) is
very close to the enthalpy of sublimation, 26.4 kcal/mole, calculated earlier from
thermodynamic data.[26,28] Using this value and taking into account equality of vapor pressure
above liquid and solid phases, one can obtain the temperature dependence of vapor pressure
above solid NTO as following:
lnP = -12984.4/T + 22.78
A good conformation to this expression is nearness of calculated vapor pressures to initial
pressures in glass Bourdon gauge in experiments on NTO decomposition (Fig 5).
The important feature of NTO combustion at low pressures is very low temperature
gradients (~8600 K/cm at 0.5 MPa) just above the burning surface. According to thermocouple
data, the heat transferred back from the gas phase is as low as 10-12 cal/g. This value is
considerably less than the heat required for warming-up the substance to the surface temperature
and melting it (133-136 cal/g). It can be assumed, therefore, that the leading reactions on NTO
combustion at low pressures occur in the condensed phase. After merging gas flames at elevated
pressures, the temperature gradient from the gas substantially increases, the heat flux from the
gas being, however, still insufficient.
Low values of the heat flux from the gas at pressures up to at least 2 MPa allow considering
the condensed-phase chemistry as determining the combustion of NTO. Rate constants of the
dominant combustion reaction in the pressure range of 0.4-10 MPa can be obtained from the
Zeldovich expression[31] for the mass burning rate (m) that is based on a dominant role of the
condensed-phase chemistry:
m=
2 2 Q
c p (Ts T0 + Lm / c p )
RTs2
E / RTs
) Ae
E
The average specific heat, cp, was calculated to be 0.35 cal/gK from experimental data on
thermal diffusivity () as 1.710-3 cm2/s, density of strand as 1.87 g/cm3 and thermal
conductivity of the condensed phase () as 0.00123 cal/cmsK taken from Ref.[32] The heat of
reaction, Q, was taken as 310 cal/g,[12] and the heat of melting, Lm, as 51.5 cal/g.[29]
As seen in Fig 6, kinetics of NTO decomposition derived from the combustion model,
k = 1.14 1014 exp( 39000 / RT ) s-1, do not exhibit very high activation energy, 39 kcal/mol,
and approach decomposition data of solid NTO when extrapolated to the low-temperature
region.
The correctness of these new data can be supported by following considerations. According
to the condensed-phase combustion model, the pressure exponent in the burning rate vs.
pressure dependence is defined by the activation energy of the leading reaction and the heat of
evaporation process, which controls the surface temperature. Consequently, the activation
energy of NTO decomposition reaction, which is the limiting reaction in the condensed phase,
can be found from expression E = 2nLv. Using heat of evaporation as 19.7 kcal/mole, calculated
from thermodynamic data,[26,28] and experimental value of the pressure exponent, n = 0.976, the
activation energy can be calculated as which 38.5 kcal/mole. This value is close to the activation
energy derived from the combustion model. In addition, the activation energy, which
321
Czech Republic
2006
[Content]
Rate Constant, s-1
corresponds to NTO decomposition in melt, is in a good agreement with activation energies of

NTO decomposition in solution,[9] though the rate constants in it are less a little.
1E+4
1E+3
1E+2
1E+1
1E+0
1E-1
1E-2
1E-3
1E-4
1E-5
1E-6
1E-7
1E-8
1E-9
1E-10
1E-11
1E-12
1E-13
0.00120
5
4
0.00160
0.00200
0.00240
0.00280
1/Ts, K-1
Fig 6.
3.2
Comparison of the rate constants in a wide temperature interval: 1 (crosses

and line)- the rate constants of the leading reaction in NTO combustion, 2
(line and dotted line) data of work[3], 3 (rhombs) data of work,[9] 4
(crosses) - data of work,[20] 5(triangles) data of work[1].
Thermal decomposition of NTO in the gas phase
Knowing NTO vapor pressure, experiments have been conducted on decomposition of the
substance in the gas phase. The experiments were carried out by means of the manometric
method using glass Bourdon gauges of ~13 cm3 volume at temperatures of 240 (#221) and
250C (#222) and sample weights of 1.3 and 1.4 mg, respectively. The weights of the samples
were decreased significantly as compared to analogous previous experiments.[19,20] This made it
possible to come to the ratio of the sample mass to the volume of the reaction vessel as little as
0.1 mg/cm3. In doing so, estimated vapor pressure in the vessel would be less than the saturated
vapor pressure of NTO.
The time curves of the decomposition extent for experiments at 240 and 250C are
presented in Fig 7 and Fig. 8 in comparison with decomposition curve at m/V ~ 1 mg/cm3.[19,20]
It was observed during experiments that decomposition is accompanied by formation of
brown transparent film on the internal walls of the vessel. At the same time there was no solid
residue at the bottom of the vessel observed after decomposition, although it was the case for
bigger m/V ratios.
322
Czech Republic
2006
[Content]
At 240C pressure in the vessel increases in direct proportion to time, reaching the
calculated value of the vapor pressure of NTO (21 Torr). Then the decomposition rate begins to
decrease. After a half of NTO has been decomposed, a self-acceleration of decomposition is
observed, which is weaker, however, than in experiments with large m/V ratios. The total
decomposition reaction was found to produce 2.85 mole gases per one mole of NTO, as it had
been observed in earlier experiments.[20] In the area after reaching vapor pressure and up to 20%
decomposition, the reaction rate decreases in accordance with the firs-order low and is close to
the rate of decomposition at big m/V ratio (Fig 7, test #97). The firs-order rate constants for
these experiments (#221 and #97) were calculated to be 0.0037 min-1 and 0.0013 min-1,
respectively. Increasing the temperature of experiments did not result in changing the kinetic
pattern: at 250C the self-acceleration in the test with small sample mass (#222) was even
weaker, and the formal rate constant was 0.013 min-1 as compared to 0.007 min-1 for large
sample mass (Fig 8, test #100).
1.0
conversion, %
conversion, %
1.0
1
0.5
#221
#97
2
1
0.5
#222
#100
0.0
0
0.0
0
200
400
40
80
120
160
, min
600
, min
Fig 7.
3.3
Comparison of two kinetic

curves at 240C, 1- m/V = 0.1
mg/cm3, 2 - m/V = 1 mg/cm3.
Comparison of two kinetic

Fig 8.
curves at 250C, 1- m/V = 0.1
mg/cm3, 2 - m/V = 1 mg/cm3.
Mechanism of NTO thermal decomposition
The kinetic data of NTO decomposition collected for a wide temperature interval
comprising solid, liquid, and gaseous phase states of the material, indicate that unusually high
activation energy in 230-270C temperature interval is likely to be connected with some
transitional process rather than with extraordinarily strong C-NO2 bond in the NTO molecule.
In temperature interval of 230-270C, NTO vapor pressure reaches considerable values
(0.03-0.3 atm) and could affect formal kinetic parameters of decomposition. However, the
present study shows that decomposition rate constants in both gaseous and solid states are close
together. Therefore, partial decomposition in the gas can not substantially increase the total rate.
NTO decomposition in the liquid state as well as in solution[9] was shown to proceed much
faster than in the solid state. However, the effect of sample submelting at temperatures well
below melting point, as it was already mentioned above, is improbable. Most likely the
decomposition rate is apt to drastically grow in the 230-270C temperature interval for another
reason. That may be isomerization of NTO into a less stable compound.
10
323
Czech Republic
2006
[Content]
Calculations made in works[3,27,14,18] show that the C-NO2 bond in the NTO molecule is the
weakest one, suggesting homolysis of this bond to be the most probable initial step in
unimolecular decomposition. However, Bruss et al.[1] and then Oxley et al. [9] found out the
deuterium kinetic isotope effects (DKIE), leading them to a conclusion that hydrogen transfer
was involved as a rate-limiting process. Decomposition of NTO was proposed[9] to be initiated
by hydrogen transfer from N4 to O11 of nitro group. This step features concerted rupture of the
N-H bond and formation of O-H bond simultaneously with reorganization of double bonds in
the molecule:
7HN2
3
4
HN
1N
HN
N6
10
N
H8
(1)
HONO
OH
O 11
Calculations conducted in work[18] show that this path resulting in subsequent formation of
HONO has considerably higher barrier than rupture of C-NO2 bond.
At the same time it is reasonable to suggest a structure isomer of NTO, which can be
formed without any bond ruptures as a result of hydrogen shift from N2 to O11:
HN
K1
(2)
O
NO2
N
H
NO2H
N
H
This NTO isomer contains nitro group in the form of quite ordinary aci-form as contrasted
with above-stated product of hydrogen transfer (1). Appearance of yellow color on heating of
NTO samples testifies to formation of aci-NTO. Conversion of nitro group into the aci-form
conjugated with double bond is well known to be usually accompanied by bathochromic shift of
absorption maximum in UV-spectra in visible area.[33]
Calculations based on semi-empirical method AM1 show that the longest bond among C-N
bonds in aci-NTO is the C3-N2 one. Therefore, it may be expected that rupture of this bond will
initiate NTO decomposition:
N
N
H
N2
k2
NO2H
.
O
(3)
.
N
H
NO2H
The radical center at C=O group makes it susceptible to intra or intermolecular attack by
nitro group oxygen, resulting in formation of CO2 at earlier stages of NTO decomposition that
was reported in Ref.[10].
The proposed scheme is in accordance with not only deuterium kinetic isotope effect data,
but also with decomposition of 15N isotopically labeled NTO,[11] which evidenced 54% N2 as
forming from nitrogen of cycle at positions 1 and 2.
324
Czech Republic
2006
[Content]
According to the suggested scheme the rate of the initial stage of NTO thermolysis (W) is
determined by the rate constant of the decomposition reaction of aci-isomer (k2) and its
concentration:
W = k2[aci-NTO]
The concentration of aci-NTO depends on the equilibrium constant (K1) of the reversible
reaction (2):
K1 = [aci-NTO]/[NTO]
Using the Arrhenius expression for k2
k2 = A exp(-E2/RT)
and expressing K1 from
-RTlnK1 = H1 - TS
as
K1 = exp(S/R)exp(-H1/RT),
the reaction rate can be written as:
W = k2K1[NTO] = A exp(S/R)exp-((E2+ H1)/RT)[NTO]
In doing so, one may conclude that an increase in temperature leads to increase in not only
the aci-NTO decomposition rate, but also in the aci-NTO concentration. As a result, the
observed value of the activation energy is abnormally high. Actually it is a sum of two values:
the activation energy of aci-NTO decomposition and activation enthalpy of isomerization
reaction, E2 + H1.
Under combustion conditions, the reaction of isomerization (2) is accomplished in the melt,
and the observed activation energy corresponds to that of C3-N2 bond rapture in the cycle (39
kcal/mol). When decomposing in solutions at 220-280C, the ratio of aci-NTO is also high,
resulting in the observed activation energy to be different greatly from that observed for
decomposition in the solid state. Rate constants of NTO decomposition in the gas at 240-250C
are close to those for the condensed phase, that may be explained by predominant
decomposition on the vessel walls.
4.
CONCLUSIONS
Burn rate characteristics of low sensitive explosive 3-nitro-1,2,4-triazol-5-one (NTO) have

been investigated in the pressure interval of 0.1-40 MPa. The temperature distribution in the
combustion wave of NTO has been measured at 0.4-2.1 MPa. An important feature of NTO
combustion at low pressures is very low temperature gradients above the burning surface,
suggesting that the burning rate is governed by the condensed-phase chemistry. Thermocoupleaided measurements in the combustion wave allowed finding the surface temperatures and, as a
result, deriving NTO decomposition reaction rates in the melt at 370-425C as well as NTO
vapor pressure above the liquid and solid phases.
12
325
Czech Republic
2006
[Content]
Taking into account vapor pressure data obtained, decomposition of NTO in the gas phase
has been studied and compared with decomposition in liquid and solid state. At temperatures
above 240-250C, decomposition rates in the gas are close to that in the solid state, whereas
decomposition in the melt and in solutions occurs significantly faster and shows low values of
the activation energy. Nevertheless, high value of the activation energy for solid-state
decomposition of NTO are not likely to be connected with submelting effect, because
decomposition occurs at temperatures well below melting point. It has been suggested that
abnormally high activation energy in the interval of 230-270C is stipulated by peculiarities of
NTO transitional process rather than strong bonds in the molecule. In this area, NTO molecule
undergoes isomerization into the aci-form, followed by C3-N2 heterocyclic bond rupture,
resulting in abnormally high value of the observed activation energy. Actually it is a sum of two
values: the activation energy of aci-NTO decomposition and activation enthalpy of
isomerization reaction.
5.
REFERENCES
[1]
J.A. MENAPACE, J. MARLIN, D. BRUSS, R. DASHER: Photochemical and Thermochemical

Decomposition of 3-Nitro-1,2,4-triazol-5-one and Perdeuterio-3-nitro-1,2,4-triazol-5-one in Neat
and Mixed Systems, J. Phys. Chem., Vol. 95, pp.5509-5517, 1991.
[2] E.F. ROTHGERY, D.E. AUDETTE, R. WEDLICH, D. CSEJKA: The Study of the Thermal
Decomposition of 3-Nitro-1,2,4-triazol-5one (NTO) by DSC, TGA-MS, and ARC, Thermochem.
Acta, Vol. 185, pp. 235-243, 1991.
[3] H. STMARK: Thermal decomposition of NTO, FOA Report D-20178 2.3, National Defense
Research Establishment, Sundbyberg, Sweden, November 1991.
[4] B.C. BEARD, J. SHARMA: Early Decomposition Chemistry of NTO, J. Energ. Mater., Vol.11,
pp.325-344, 1993.
[5] K.V. PRABHAKARAN, S.R. NAIDU, E.M. KURIAN: Spectroscopic and Thermal Analysis Studies on
3-Nitro-1,2,4-Triazole-5-one(NTO), Thermochem. Acta, Vol. 291, pp.199-212, 1994.
[6] G.K. WILLIAMS, T.B. BRILL: Thermal Decomposition of Energetic Materials. 68. Decomposition
and Sublimation Kinetics of NTO and Evaluation of Prior Kinetic Data, J. Phys. Chem., Vol.99,
No.33, pp.12536-12539, 1994.
[7] T.B. BRILL, P.E. GONGWER, G.K. WILLIAMS: Thermal Decomposition of Energetic Materials. 66.
Kinetic Compensation Effects in HMX, RDX, and NTO, J. Phys. Chem., Vol.98, No.47, pp. 1224212247, 1994.
[8] J.C. OXLEY, J.L. SMITH, K.E. YEAGER, E. ROGERS, X.X. DONG: NTO Decomposition Studies, in
Decomposition, Combustion and Detonation Chemistry of Energetic Materials, MRS Symp. Proc.,
Ed: Brill, T.B., Russell, T.P., Tao, W.C. and Wardle, R.B., 1995, 27-30 Nov., Boston,
Massachusetts, USA, Vol. 418, pp.135-142, 1995.
[9] J.C. OXLEY, J.L. SMITH, L.-Z. ZUN, R.L. MCKENNEY: Thermal Decomposition Studies on NTO
and NTO/TNT, J. Phys. Chem., Vol. 99, No.25, pp.10383-10391, 1995.
[10] T.R. BOCHER, D.J. BEARDALL, C.A. WIGHT, L. FAN, T. BURKEY: Thermal Decomposition
Mechanism of NTO, J.Phys.Chem., Vol.100, pp.8802-8806, 1996.
[11] J.C. OXLEY, J.L. SMITH, E. ROGERS, X.X. DONG: NTO Decomposition Products Tracked with
15
N-Labels, J. Phys. Chem., Vol. 101, pp.3531-3536, 1997.
326
Czech Republic
2006
[Content]
[12] G.T. LONG, B.A. BREMS, C.A. WIGHT: Thermal Activation of the High Explosive NTO:
Sublimation, Decomposition, and Autocatalysis, J. Phys.Chem. B, Vol.106, No.15, pp. 4022-4026,
2002.
[13] D.C. SORESCU, T.R.L. SUTTON, D.L. THOMPSON, D. BEARDALL, C.A. WIGHT: Theoretical and
Experimental Studies of the Structure and Vibrational Spectra of NTO, J. Mol. Struct., Vol. 384
(2/3), pp. 87-99, 1996.
[14] N.J. HARRIS, K. LAMMERTSMA: Tautomerism, Ionization, and Bond Dissociations of 5-nitro-2,4dihydro-3H-1,2,4-triazolone, J. Am.Chem.Soc., Vol.118, No.34, pp.8048-8055, 1996.
[15] Y.-M. WANG, C. CHEN, S-T. LIN: Theoretical Studies of the NTO Unimolecular Decomposition, J.
Mol. Struct., Vol. 460, No.1/3, pp. 79-102, 1999.
[16] Y. KOHNO, O. TAKAHASHI, K. SAITO: Theoretical Study of Initial Decomposition Process of NTO
Dimer, Phys. Chem. Chem. Phys., Vol. 3, No.14, pp.2742-2746, 2001.
[17] W.-L. YIM, Z.-F. LIU: Application of Ab Initio Molecular Dynamics for A Priori Elucidation of the
Mechanism in Unimolecular Decomposition: The Case of 5-Nitro-2,4-dihydro-3H-1,2,4-triazol-3one (NTO), J. Amer. Chem. Soc., Vol.123, No.10, pp. 2243-2250, 2001.
[18] C. MEREDITH, T.P. RUSSELL, R.C. MOWREY, J.R. MCDONALD: Decomposition of 5-nitro-2,4dihydro-3H-1,2,4-triazol-3-one: Energetic Associated with Several Proposed Initiation Routes, J.
Phys. Chem. A, Vol.102, No.2, pp.471- 477, 1998.
[19] B.N. KONDRIKOV, S.P. SMIRNOV: Mechanisms of Thermal Decomposition and Steady-State
Burning of NTO, Proc. 32nd Inter. Annual Conf. of ICT, Karlsruhe, Germany, 24- 27 June, paper
23, pp.1-14, 2003.
[20] B.N. KONDRIKOV, S.P. SMIRNOV, A.V. MINAKIN, R.M. DOHERTY: Chemical Kinetics of the
Thermal Decomposition of NTO, Propellants, Explosives, Pyrotechnics, Vol. 29(1), pp. 27-33,
2004.
[21] B.N. KONDRIKOV, S.P. SMIRNOV, YA.A. KANIKOVA, R.M. DOHERTY: Chemical Kinetics of the
Macroscopic Stages of NTO Thermal Decomposition, Proc. 35th Inter. Annual Conf. of ICT,
Karlsruhe, Germany, June 29 - July 2, paper 89, pp.1-14, 2004.
[22] G.B. MANELIS, G.M. NAZIN, YU.I. RUBTSOV, V.A. STRUNIN: Thermal Decomposition and
Combustion of Explosives and Propellants, Nauka, Moscow, 223p., 1996, (in Russian).
[23] A.E. FOGELZANG, V.P. SINDITSKII, V.YU. EGORSHEV, V.V. SERUSHKIN, V.I. KOLESOV:
Combustion of 3-Nitro-1,2,4-triazol-5-one and its Salts, Proc. Inter. Conference on Combustion,
Moscow, 12-17 Sept., pp.129-132, 1994.
[24] G.I. CHIPEN, R.P. BOKALDER, V.Y. GRINSHTEIN: 1,2,4-Triazol-3-one and its Nitro and Amino
Derivatives, Chemistry of Heterocyclic Compounds, Vol.2(1), pp.110-116, 1966.
[25] G.B. BELOV: Thermodynamic Analysis of Combustion Products at High Temperature and
Pressure, Propellants, Explosives, Pyrotechnics, Vol. 23, pp. 86-89, 1998.
[26] A. FINCH, P.J. GARDNER, A.J. HEAD, H.S. MAJDI: The Enthalpies of Formation of 1,2,4-triazol-5one and 3-Nitro-1,2,4-triazol-5-one, J.Chem.Thermodinamics, Vol. 23. No.12, pp.1169-1173,
1991.
[27] J.P. RITCHIE: Structures and Energies of the Tautomers and Conjugate Bases of Some 1,2,4Triazolones, J. Org. Chem., Vol. 54, (15), pp.3553-3560, 1989.
[28] P. POLITZER, J.S. MURRAY, M.E. GRICE: Computational Determination of Heats of Formation of
Energetic Compounds, in Decomposition, Combustion and Detonation Chemistry of Energetic
Materials, MRS Symp. Proc., Ed: Brill, T.B., Russell, T.P., Tao, W.C. and Wardle, R.B., 27-30
Nov., Boston, Massachusetts, USA, Vol. 418, pp.55-66, 1995.
14
327
Czech Republic
2006
[Content]
[29] V.L. ZBARSKY, A. BASAL, N.V. YUDIN, V.F. ZHILIN: Study on Solubility of 2,4-Dihydro-1,2,4Triazol-3-one and 5-Nitro-2,4- Dihydro-1,2,4-Triazol-3-one in Diluted Acid, Proc. 34th Inter.
Annual Conf. of ICT, June 24 - June 27, Karlsruhe, FRG, paper 139, pp.1-8, 2003.
[30] K.J. KIM, M.J. KIM, J.-M. LEE, S.-H. KIM, H.-S. KIM, B.-S. PARK: Solubility, Density, and
Metastable Zone Width of the 3-Nitro-1,2,4-triazol-5-one + Water System, J. Chem. Eng. Data,
Vol.43(1), pp. 65-68, 1998.
[31] Y.B. ZELDOVICH: Theory of Combustion of Propellants and Explosives, Zh. Eksperimentalnoy i
Teoreticheskoi Fiziki (Russ. J. Exper. and Theor. Physics), Vol.12, No. 11-12, pp.498-524, 1942
(in Russian).
[32] Y. XIE, R.-Z. HU, C.-Q. YANG, G.-F. FENG, J. ZHOU: Studies on the Critical Temperature of
Thermal Explosion for 3-Nitro-1,2,4-Triazol-5-One (NTO) and its Salts, Propellants, Explosives,
Pyrotechnics, Vol.17, pp. 298-302, 1992.
[33] The Chemistry of Nitro and Nitroso Groups. Edited by Feuer H., Interscience Publishers, New
York, part 1, p. 286, 1959.
328
Czech Republic
2006
[Content]
THE PRINCIPAL APPROACH TO THE ESTIMATION

OF COOKOFF TEMPERATURE USING FEM
Jakub eleovsk, and Miloslav Krupka
University of Pardubice, 532-10 Pardubice, CZ
Abstract
The practical approach to the estimation of cookoff temperature using FEM is presented
in this paper. The characterization of explosive material properties (density, thermal
conductivity, specific heat, heat of decomposition, the decomposition kinetics) is
discussed. LS-DYNA3D code is used for the fitting of appropriate kinetic model and for
the simulation of heat transfer during slow cookoff. The results of simulations are
compared to experimental values for the SEMTEX 1A plastic bonded explosive.
Keywords:
FEM, cookoff, simulation, heat transfer
INTRODUCTION
The cookoff tests are used for evaluating of a hazard connected with heating of energetic
materials (especially EIDS Extremely Insensitive Detonating Substances) or ammunition
(LOVA LOw Vulnerability Ammunition). In case of energetic materials the explosive is
placed into a steel confinement and heated. In case of LOVA the whole ammunition object is
tested. The temperature of decomposition and the violence of decomposition reaction are
evaluated.
Because of the cost of experimental procedures, the numerical simulation is often used in
the field of characterization of various loading effect on the material. One of the most
commonly employed numerical techniques is the Finite Element Method FEM. The mean
principle of this method is to divide the body into small elements and to calculate the effect
of applied loads only in discrete points nodes. This creates a large system of equations,
which can then be solved numerically.
Several research groups published results of their effort in the field of cookoff computer
modeling to investigate energetic materials response to thermal heating, in a confined
geometry. Sandia National Laboratory (USA) is working on a multidimensional fully coupled
thermal/chemical/mechanical code to calculate the response of reactive materials[1-5]. They
also tried to develop model for thermally degraded energetic materials[6]. The LLNL (USA)
reported work on a single computer code specialized only for the cookoff response[7-11] of
energetic materials. The cookoff was also simulated in TNO (Netherlands)[12], or in
Australia[13].
One off the most popular FE codes, utilized in the field of explosive effects modeling, is
LS-DYNA3D from Lawrence Livermore National Laboratory (USA). LS-DYNA3D is
intended for the simulations of fast dynamic processes, like car crash tests, detonation, or e.g.
explosive burn. LS-DYNA3D is also used in Department of Energetic Materials, University
of Pardubice for the simulations of detonation processes. The modeling of heat transfer is
also possible.
The complex approach to the simulation of the heat transfer during slow cookoff using
LS-DYNA3D code is presented in this paper. The characterization of required material
properties using laboratory scale tests is also described.
Jakub eleovsk, Miloslav Krupka
329
Czech Republic
2006
[Content]
THEORETICAL DESCRIPTION
The calculation of the temperature distribution during cookoff can be described as the
transient heat conduction with the heat source. The transient character of the problem is given
by consecutive heating of the confinement and also by the heat generation caused by
decomposition of the explosive. The problem is nonlinear because of exponential dependence
of the heat generation on the temperature.
The equation for transient heat conduction with heat source is given:
T T T
T
+ q& ;
=
+
+
(1)
t
x y z
where C is specific heat, is density, T is temperature, t is time, is thermal
conductivity, x, y, z are cartesian coordinates and q& is a heat source.
C
The source of heat (heat generated by decomposing explosive) can be in the simplest
form expressed by Arrhenius equation (for unit volume):
q& = H r A exp( E / RT ) ;
(2)
where H r is reaction heat, A is preexponential factor, f ( ) is the function of fraction

reacted, E is activation energy and R is universal gas constant.
The equation (1) with the q& expressed from equation (2) can not be solved analytically.
However numerical solution is possible, e.g. using the finite element method. Various
commercial FE codes can be used to solve the eq. (1), we have used the LS-DYNA3D code.
The LS-DYNA code is intended for the simulation of fast dynamic processes primary. It
is also possible to simulate the heat transfer in this code, but there are some limitations. The
LS-DYNA code is not able to calculate the kinetics of chemical reactions. The heat released
by decomposing explosive can be dependent only on the temperature, not on e.g. fraction
reacted. Therefore the only one applicable kinetic model is the zero order reaction ( f ( ) = 1
in eq. (2)).
The characterization of material properties required for the solution of equation (1) and
the numerical modeling is described in experimental part.
EXPERIMENTAL PART
All the measurements and simulations described bellow were carried out with the
SEMTEX 1A plastic bonded explosive. SEMTEX 1A is commercial PETN based high
explosive, manufactured by Explosia, a.s., Czech Republic.
The estimation of cookoff temperature is done in following steps:
1. Determination of material parameters from the equation (1) C, , ;
2. determination of the decomposition kinetics H r , A, E;
3. verification of the kinetic model;
4. numerical simulation of slow cookoff.
330
3.1
Czech Republic
2006
[Content]
Determination of material parameters from the equation (1)
The specific heat C of explosive material can be determined by calorimetric methods. We

used the SETARAM C80 (Setaram, France) heat flow calorimeter. The temperature
increment method was used.
The method of double weighting was used for the density determination. The sample
of explosive was weighed in air, the same sample was weighed in the water and the density
was calculated on the base of Archimedes law.
The thermal conductivity was calculated from the temperature profile inside the
sample subjected to constant ambient temperature. The flat sample (cylinder, 20 mm in
diameter and 4 mm high) with the constant starting temperature was put between two
aluminum blocks tempered to higher temperature. The schema of the apparatus JK-80
(manufactured by OZM Research, Czech Republic) is in the figure 1. The temperature profile
measured in the center of the sample was compared to analytical expression for the
temperature inside the infinite slab and was estimated.
Fig 1. The apparatus for the thermal conductivity measurement.
3.2
Determination of decomposition kinetics
The heat of decomposition H r was measured using the automatic combustion

calorimeter MS-10A (manufactured by Laget, Czech Republic) in nitrogen atmosphere. The
presence of additional substance (nitrocellulose propellant) was essential to initialize the
decomposition in inert atmosphere.
To determine the kinetic parameters A and E, the differential thermal analysis (DTA
551Ex, manufactured by OZM Research, Czech Republic) was used. The isothermal
measurements were carried out at the temperatures of 166, 168, 170 and 172 C with 50 mg
of sample. The activation energy and preexponential factor was determined by isoconverional
method. However the description of decomposition kinetics with the zero order reaction
model is not very good. The values of activation energy and preexponential factor vary with
the fraction reacted (fig. 2).
These kinetic parameters are the key variables in equation (2). Their values have a strong
influence on the amount of heat envolved by the decomposition of the sample. The selection
of a pair of kinetic parameters, which preferably describe the sample decomposition, follows.
3
331
Czech Republic
2006
[Content]
Fig 2. The dependence of A and E on the fraction reacted .
3.3
Verification of kinetic model
As was said, the kinetic parameters A and E vary with the fraction reacted. The
nonisothermal DTA was simulated using simple LS-DYNA model. The various values of A
and E (corresponding to various ) were used for the description of explosive decomposition.
The simulation was carried out with five different heating rates and the results were
compared to experimental DTA curves, measured with DTA 551Ex (manufactured by OZM
Research). The example of such comparison is plotted in figure 3. Based on the least squares
method, the most appropriate model (A and E calculated at = 0.07 ) was chosen.
3.4
Numerical simulation of slow cookoff
The heat transfer during the slow cookoff test is simulated using LS-DYNA3D finite
element code, used FE model is in figure 3. The model contains 332 565 elements of
SOLID164 type. The process was modeled as the transient heat conduction with heat source
eq. (1) with the convection boundary condition. The previously determined material
parameters of explosive were used. The main result of this simulation is the cookoff
temperature.
The configuration of cookoff test, used in the Institute of Energetic Materials, is
described below.
332
Czech Republic
2006
[Content]
Fig 3. The comparison of simulated and experimental DTA data (heating rate 2 C/min).
Fig 4. The FE model of slow cookoff test.
333
Czech Republic
2006
[Content]
Fig 5. The simulated time behavior of temperature in the center of the sample during cookoff.
Fig 6. The simulated temperature distribution before ignition.
334
Czech Republic
2006
[Content]
Fig 7. The schema of cookoff temperature estimation using LS-DYNA3D FE code.
335
3.5
Czech Republic
2006
[Content]
Slow cookoff experiment
The slow cookoff (SCOT 400S apparatus, manufactured by OZM Research, Czech
Republic) test was carried out in accordance with the UN Recommendation for transporting
dangerous goods[14] manual. The only change is the reduction of dimension of the whole set.
The sample is put in the seamless steel tube with inner diameter 21 mm, outer diameter 26.4
mm and length 90 mm. Both sides of the tube were enclosed by steel caps. The whole set is
heated with rate 3.3 C/hr until the decomposition. The temperature in the center of the
sample, the temperature in the outer side of the bomb and the temperature of the air in the
oven is measured during the test.
The average cookoff temperature from three experimental measurements is 136.9 C.

The simulated time temperature behavior in the center of the sample during cookoff is plotted
in figure 5 and the calculated cookoff temperature is 138.0 C (estimated as the onset of the
temperature growth from figure 5). The temperature distribution before the termination the
simulation is in figure 6. The increase of temperature in the center of the sample corresponds
to the initiation.
The agreement between measured and simulated cookoff temperature is quite good. The
small difference (1.1 C) is probably caused by using the zero order reaction model for the
decomposition of the sample.
The sequence of operations is schematically displayed in figure 7. This routing was
successfully used in case of SEMTEX 1A high explosive sample. The explosive material was
completely characterized with laboratory scale experiments and the cookoff temperature was
predicted. Authors believe that this routing can be an instruction for the modeling the heat
transfer during heating of other explosives.
CONCLUSION
The principle of numerical simulation of heat transfer during cookoff was introduced in
this paper. All the material properties of the sample were specified in laboratory scale
experiments. The heat transfer during cookoff was simulated using LS-DYNA finite element
code, and the calculated cookoff temperature is in good agreement with experimental results.
Acknowledgments
This work was created as a part of the project of Ministry of Education, Youth and Sports
of the Czech Republic No. MSM 0021627501.
Authors would like to thank to the OZM Research company for lending the instruments
JK-80 and SCOT 400, and also to Galina Sdovsk, PhD., from Department of Inorganic
Technology, University of Pardubice for the measurement of heat capacity.
336
Czech Republic
2006
[Content]
REFERENCES
[1] M.R. BAER, M.E. KIPP, R.G. SCHMITT, AND M.L. HOBBS. Towards assessing the violence
of reaction during cookoff of confined energetic materials. In JANNAF Combustion
Subcommittee and Propulsion Systems Hazards Subcommittee Joint Meeting, Monterey,
California, USA, November 4-8 1996.
[2] M.L. HOBBS AND M.R. BAER. Multidimensional fully-coupled thermal/chemical/mechanical response of reactive materials. In JANNAF Propulsion Systems Hazards
Subcommittee Meeting, Huntsville, Alabama, USA, October 23-27 1995.
[3] M.R. BAER, M.L. HOBBS, R.J. GROSS, D.K. GARTLING, AND R.E. HOGAN. Modeling
thermal/chemical/mechanical response of reactive materials. In 26th International Annual
Conference ICT, Karlsruhe, Germany, 1995.
[4] M.R. BAER, M.L. HOBBS, R.J. GROSS, AND G.R. SCHMITT. Cookoff of energetic materials.
In 11th International Detonation Symposium, Snowmass, Coloredo, USA, August 31 September 4 1998.
[5] W.W. ERIKSON, R.G. SCHMITT, A.I. ATWOOD, AND P.D. CURRAN. Coupled thermalchemical-mechanical modeling of validation cookoff experiments. In JANNAF 37th
Combustion and 19th Propulsion Systems Hazards Subcommittee Meeting, Monterey,
California, USA, November 2000.
[6] M.L. HOBBS, M.R. BAER, AND R.J. GROSS. A constitutive mechanical model for energetic
materials. In 20th International Pyrotechnic Seminar, Colorado Springs, Colorado, USA,
1994.
[7] A.L. NICHOLS, R. COUCH, R.C. MCCALLEN, I. OTERO, AND R. SHARP. Modelling
thermally driven energetic response of high explosives. In 11th International Detonation
Symposium, Snowmass, Colorado, USA, August 31 - September 4 1998.
[8] M.A. MCCLELLAND, T.D. TRAN, B.J. CUNNINGHAM, R.K. WEESE, AND J.L.
MAIENSCHEIN. Cookoff response of PBXN-109: Material characterization and ALE3D
thermal predictions. In 50th Joint Army-Navy NASA Air Force (JANNAF) Propulsion
Meeting, Salt Lake City, Utah, USA, July 11-13 2001.
[9] A.L. NICHOLS, A. ANDERSON, R. NEELY, AND B. WALLIN. A model for high explosive
cookoff. In 12th Internatioanl Detonation Symposium, San Diego, California, USA, August
11-16 2002.
[10] M.A. MCCLELLAND, J.L. MAIENSTEIN, A.L. NICHOLS, J.F. WARDELL, A.I. ATWOOD, AND
P.O. CURRAN. ALE3D model predications and materials characterization for the cookoff
response of PBXN-109. In Joint Army Navy NASA Air Force 38th Combustions
Subcommittee, 26th Airbreathing Propulsion Subcommittee, 20th Propulsion Systems
Hazards Subcommittee and 2nd Modeling and Simulation Subcommittee Joint Meeting,
Destin, Florida, USA, April 8-12 2002.
[11] J.J. YOH AND M.A. MCCLELLAND. Simulating the thermal response of high explosives on
time scales of days to microseconds. In 13th APS SCCM Conference, Portland, Oregon,
USA, 2003.
[12] G. SCHOLTES, J. MAKKUS, AND N. VAN DER MEER. Measuring and modeling the cook-off
process. In 28th International Annual ICT Conference, Karlsruhe, Germany, 1997.
[13] S.-Y. HO. Modeling of cook-off reaction violence of confined energetic materials. In
V. Yang, T.B. Brill, and W.-Z. Ren, editors, Solid Propellant Chemistry, Combustion, and
Motor Interior Ballistics, pages 591606. Reston, Virginia, USA, 2000.
[14] Recomendations on the transportation of dangerous goods, manual of tests and criteria,
3rd edition, United Nations, New York and Geneva 1999.
337
Czech Republic
2006
[Content]
LOW TEMPERATURE ACTIVATION ENERGY OF PROPELLANTS

Uldis Ticmanis, Stephan Wilker, Gabriele Pantel
WIWEB ASt Heimerzheim, Groes Cent, 53913 Swisttal (DE)
Abstract
This paper describes a new approach to determine the activation energy of propellant
decomposition reactions which quantifies the temperature dependence of a reaction rate
at temperatures below 50C by using an indirect microcalorimetric method.
Whereas the determination of life time and stability of propellants is usually performed
by artificial ageing at elevated temperatures with a temperature range usually between
70 and 130C only few and scattered data are available for temperatures below 50C.
Normally these activation energies for propellant decomposition reactions are not
known below temperatures of about 50C because (i) the reaction rates are very small
and thus ageing times have to be very long and (ii) because of slow reactions the
chemical change of the propellant becomes very small which in combination with too
short ageing times may lead to wrong estimations and calculations, because the
experimental error is relatively big in comparison with the real change of the
propellant.
To be on the safe side people usually assume a low value of activation energy. This
has the effect that measuring times at high temperatures are much longer than they
should be and that maybe good material is thrown away too early in fear of a thermal
instability.
To overcome this problem we have established a new approach by an indirect microcalorimetric method. This method evaluates the shift of typical points in the heat flow
curve to earlier times after artificial ageing at relatively low (30-70C) temperatures.
From the time shift of a specific point in the heat flow curve (compared to the unaged
material) and the temperature difference between ageing temperature and measuring
temperature activation energies of the whole temperature range between 89C and 30C
are available. We have verified the usability of this approach with six different
nitrocellulose based propellants.
Keywords: propellants, activation energy, microcalorimetry, stabilizer depletion
1.
INTRODUCTION
The determination of life time and stability of propellants is usually performed by

artificial ageing at elevated temperatures. The temperature range usually lies between 90 and
130C for classical stability tests whereas other tests like microcalorimetry, stabilizer
depletion and vacuum stability are usually performed at temperatures between 70 and 100C.
To calculate the real storage time that is covered by a high temperature experiment one must
assume an activation energy which quantifies the temperature dependence of a reaction rate.
Normally these activation energies for propellant decomposition reactions are not known
below temperatures of about 60C because the reaction rates are too small for a direct
measurement even in sensitive calorimeters or for achievement of measurable gas generation
in the vacuum stability test. Also stabilizer depletion is much too slow unless very long
measuring times are realized.
338
Czech Republic
2006
[Content]
We have established a new approach to solve this problem by an indirect microcalorimetric method. This method evaluates the shift of typical points of the heat flow curve
(maxima, minima, points of inflection) to earlier times after artificial ageing at relatively low
(30-70C) temperatures in an oven. From the time shift of a specific point in the
microcalorimetric curve (e.g. point of inflection, maximum, minimum, thermal runaway)
compared to the unaged material and the temperature difference between ageing temperature
and measuring temperature activation energies of the whole temperature range between 89C
and 30C are available. We have verified the usability of this approach with six different
nitrocellulose based propellants.
2.
PROPELLANTS TESTED
Five types of gun propellants and one rocket propellant (propellant D) were investigated by this low temperature activation energy (LTAE) approach. They are listed (including
their main ingredients) in table 1. Propellants A, K, P and V were taken as received.
Propellant D was cut into small dices (2 - 4 mm length). Propellant N was milled and sieved.
Samples with particle sizes between 1 and 5 mm were taken for this study.
Table 1. Propellants used for the LTAE study
Propellant
abbreviation
Propellant
type
Stabilizer type
Ageing
temperatures [C]
NG content
[%]
Age [y] at
beginning
single base
DPA + EC
30, 35, 40, 50, 70
20
double base
2-NDPA
30, 35, 40
34
double base
DPA
30, 35, 40, 50, 70
19
triple base
EC
30, 35, 40, 50
21
double base
DPA + EC
30, 35, 40, 50, 70
10
porous
DPA
30, 35, 40, 50
18
Explanations: DPA = diphenylamine, EC = ethyl centralite, 2-NDPA = 2-nitro-DPA

No akardite stabilized propellant was taken because their decomposition follows nearly
exactly zero order, which means that no typical points (maxima, minima) could be chosen
for evaluation. Also no TPA stabilized propellant was available in large enough amounts
although the typical shape of their heat generation rate[3] would make them ideal candidates
for this study.
3.
EXPERIMENTS
Microcalorimetric measurements were conducted with a Thermal Activity Monitor

(TAM 2277 and TAM III, Thermometric AB, Sverige). The measurements were performed
in 3 ml glass ampoules at 89C.
Because it was important to repeat the shape of the microcalorimetric curve as precisely
as possible[1] the ampoules were filled with exactly ( 1 mg ) the same amount of propellant.
Because the shape of the curve also depends on the moisture level[2], the filling of the
2
Stephan Wilker, Uldis Ticmanis, Gabriele Pantel
339
Czech Republic
2006
[Content]
ampoules was performed within one day for one propellant type. Samples for each ageing
step were weighed in triplicate. Two samples were taken for microcalorimetry, the third
sample was used for stabilizer depletion analysis.
Stabilizer depletion was measured by high performance liquid chromatography
(HPLC). The samples were prepared by dissolution in 20 ml of acetonitrile. The
nitrocellulose was precipitated by adding 5 ml of water. The solution was filtered. The
stabilizers were separated and detected with a suitable HPLC system containing a diode
array detector ( = 200-356 nm). For more details see[1].
4.
RESULTS
4.1
Classical iso evaluation from isothermal tests and its limit
In a very long measuring campaign, PB Clermont S.A. (Belgium) has conducted

isothermal microcalorimetry of double base spherical propellants[1]. The measurements were
performed at 89, 80, 70, 60, and 50C. The latter measurement is still running after now
2300 days of continuous measurement. Fig. 1 shows this important experiment. The lines
with slope -1 cut the heat flow curves at the same point for any temperature. This means that
the activation energy is nearly independent of the reaction degree and the temperature.
Overall the value for Ea is around 137 3 kJ/mole.
HFC OF FRESH SAMPLES TLP K6210-13 LOT 225 AT DIFFERENT TEMPERATURES
1.0E-03
P (W/g)
T = 89C
1.0E-04
T = 80C
1.0E-05
T = 70C
T = 60C
1.0E-06
T = 50C
1.0E-07
0.1
Fig 1.
10
100
1000
TIME (DAYS ) 10000
Microcalorimetry of propellant K at five different temperatures
Nevertheless, a continuous measurement at temperatures below 50C is hard to realize.

There are two reasons for that: (i) the heat generation rate is too low to be able to distinguish
it from base line noise and (ii) the time to reach a significant level of decomposition is much
too long to be realizable.
340
4.2
Czech Republic
2006
[Content]
Precursor tests at 70C
The propellants A, K and P were taken in a preliminary test which was performed to
demonstrate the suitability of the whole project using an ageing temperature of 70C. All
three propellants were subjected to the LTAE test. In addition the activation energy of all
three samples was estimated from continuous measurements at 89C and 70C and
subsequent iso evaluation. Table 2 shows the results of the calculations (for details see
chapter 4.4).
Table 2. Kinetic evaluation of precursor tests at 70C and comparison of activation
energy values with isothermal runs
Propellant
Ea from LTAE (89-70C)
Ea from isothermal runs (89-70C)
131 kJ/mole
135 kJ/mole
144 kJ/mole (from poi);

137 kJ/mole (from maximum)
137 kJ/mole
137 kJ/mole
141 kJ/mole
Explanation: Ea = activation energy, poi = point of inflection

The values of Ea do not differ significantly from each other. Therefore this preliminary
test was declared as being successful and the real LTAE was begun (see chapter 4.3).
4.3
LTAE studies at temperatures between 30 and 50C
After the successful preliminary test we started the ageing programme at 30, 35, 40 and
50C. As ageing times are extremely long, this programme began five years ago and is still
far away from being finished. As the stability of the propellants is different and the different
stabilizers which are responsible for the typical shape of the microcalorimetric curve we did
not set up an overall ageing plan but we decided to take out samples when we assumed that a
significant change would have happened. In the following sub-chapters the features of the
different propellants are presented in alphabetical order.
To be able to make a correct assumption of the heat generation rate of the unaged
material a five-fold determination of all samples were made. For evaluation the average
position of the individual points was taken. In addition, to control the ageing conditions all
ampoules were weighed after artificial ageing. If the weight loss was significantly higher
than that of the parallel samples we have excluded this sample from evaluation because it
was assumed that the seal has not been airtight all the time.
4.3.1.
Propellant A
This single base propellant caused some trouble due to instabilities in the past and was
therefore chosen to be one of the candidates for our LTAE test. Its microcalorimetric curve is
not very spectacular the only special point that can be evaluated is a point of inflection
during the increase into the second level of decomposition. To find the corresponding time
properly the first derivative of the heat generation rate was calculated and evaluated. Figure
2 shows typical derivated curves after ageing at 35C as an example.
341
Czech Republic
1269 d at 35C 1038 d at 35C 757 d at 35C
2006
[Content]
370 d at 35C
1.0
unaged
dP/dt [nW/(g*s)]]
0.8
0.6
0.4
0.2
0.0
0
t [days] at 89C
Fig 2.
Derivated heat generation rates of propellant A after ageing at 35C
4.3.2.
Propellant D
The only rocket propellant in this test also showed only one specific point of the
microcalorimetric curve its second maximum. It coincides with the complete decomposition of the stabilizer 2-NDPA and appears after ca. 10 days at 89C in unaged material.
Figure 3 shows typical results after ageing at 40C as an example.
150
610 d at 40C
140
762 d at 40C
130
1176 d at 40C
120
P [W/g]
351 d at 40C
unaged
1340 d at 40C
110
100
90
80
70
60
50
0
10
12
t [days] at 89C
Fig 3.
342
Heat generation rates of propellant D after ageing at 40C. Please note the
depletion of the first maximum which is almost zero after 1340 d at 40C
Czech Republic
2006
[Content]
4.3.3.
Propellant K
This propellant is probably the best investigated propellant concerning stability[1]. As
fig. 1 shows it has a first minimum, a second maximum and, in-between them, a point of
inflection. The point of inflection and the second maximum were chosen for evaluation,
because the first minimum is very flat (and therefore difficult to find) and it appears close
after the first maximum which belongs to oxidation reactions of the propellant with the air
around the grains in the closed ampoule. If once all oxygen is used up the first maximum
disappears and thus also no first minimum is present anymore. The depletion of the first
maximum can be seen in all propellants except for propellant N (this propellant does not
have one). Figure 4 shows typical results after ageing at 40C as an example.
300
1091 d at 40C
838 d at 40C
250
P [W/g]
200
unaged
155 d at 40C
150
332 d at 40C
478 d at 40C
623 d at 40C
100
687 d at 40C
50
0
0.0
0.5
1.0
1.5
2.0
t [days] at 89C
Fig 4.
Heat generation rates of propellant K after ageing at 40C. The heat

generation rate of the 2nd maximum remains nearly constant.
4.3.4.
Propellant N
This propellant has a very interesting heat generation rate curve, from which three points
were selected for evaluation: A sharp minimum after 0.75 days (the chemical reason for it is
unknown but it appears in all propellants containing nitroguanidine), a first maximum and a
second minimum. Typical results after ageing are shown in figure 5.
4.3.5.
Propellant P
This propellant is chemically comparable to propellant K with the main difference that
the presence of EC changes the heat generation rate curve in such a way that no second
maximum is observed. So, like with propellant A only the point of inflection could be
evaluated. Similar to propellant A is also the reason for the choice of this propellant for the
LTAE study: There were problems with instability with this type of propellant before this
study started. Typical derivated heat flow curves are shown in figure 6.
343
Czech Republic
2006
[Content]
160
140
P [W/g]
120
100
unaged
476 d at 30C
836 d at 30C
1133 d at 30C
1503 d at 30C
80
60
40
20
0
0
t [days] at 89C
Fig 5.
Heat generation rates of propellant N after ageing at 30C. Note that the
heat generation rate is reduced with longer ageing time.
10
unaged
1499 d
dP/dt [nW/(g*s)]
1248 d
1352 d
980 d
1085 d
830 d 695 d 584 d 332 d 151 d
6
4
2
0
-2
0.0
0.5
1.0
1.5
2.0
2.5
t [days] at 89C
Fig 6.
Derivated heat generation rates of propellant P after ageing at 40C. The

maximum decreases with ageing time.
4.3.6.
Propellant V
This propellant may be the most interesting one of the chosen candidates, as it shows a
thermal runaway at 89C after 10-11 days. The pattern of the first part is also very useful for
LTAE evaluation as a first minimum and a second maximum (and two points of inflection
in-between them) could be evaluated. The time to reach thermal runaway was also evaluated,
knowing that this may not be reached by natural ageing (according to [4] 10-11 days at 89C
are equivalent to 26 - 29 years at 25C). Nevertheless, a pre-ageing was suspected to make
this propellant less stable and so there was a likeliness that this thermal runaway could be
shifted to significantly earlier times. Figure 7 shows again as an example heat flow
344
Czech Republic
2006
[Content]
curves after ageing at 40C, whereas figure 8 depicts derivated microcalorimetric curves
after ageing at 35C.
230
210
980 d
633 d
158 d
190
1143 d
P [W/g]
170
828 d
345 d
unaged
150
130
110
90
70
50
0.0
0.5
1.0
1.5
2.0
2.5
3.0
t [days] at 89C
Fig 7.
Heat generation rates of propellant V after ageing at 40C

1205 d 1027 d 828 d
645 d
376 d
unaged
dP/dt [nW/(g*s)].
0.5
0
-0.5
-1
-1.5
-2
0.0
0.5
1.0
1.5
2.0
2.5
t [days] at 89C
Fig 8.
Derivated heat generation rates of propellant V after ageing at 35C
345
4.4
Czech Republic
2006
[Content]
Evaluation of activation energies
All microcalorimetric values were transferred into EXCEL. To guarantee a perfect

comparison between all curves they were recorded with the same frequency (each 300 sec).
They were numerically derivated, averaged over 15 data points.
The calculation of the activation energy is performed by linear regression of all suitable
data points from
td = a * te .
(eq. 1)
td = time difference of a certain point in the heat low curve between the aged and the
unaged material [d]
te = ageing time at low temperatures [d].
The slope a is
a =e
1
Ea 1
*
R T d T e
(eq. 2)
Ea = activation energy [kJ/mol]

R = gas constant (0.00831447 J/(Kmol))
Td = measuring temperature (362,15 K)
Te = ageing temperature [K]
The activation energy is then calculated from eq. 3:
Ea =
R * ln a
1 1

Td Te
(eq. 3)
Also the activation energy within the temperature range of ageing (50-30C) was
calculated. This was performed by replacing the single slope (from the td versus te evaluation
in eq. 3 by the ration of two different slopes (from the evaluation of one te versus another)
and Td by the higher of both ageing temperatures.
One of the most interesting questions was whether a change in activation energy would
occur at temperatures around 50-60C as indicated by many authors[5] and as installed in
STANAG 4582[4].
4.5
Stabilizer depletion
Stabilizer depletion was conducted on most of the aged samples. The reason to do this
was to control whether the ageing characteristic which could be observed by microcalorimetry would also be detectable by changes in the stabilizer content. The second idea behind
this part of the work was to check the distribution of stabilizer depletion products which
could give us information about the ageing conditions (closed/open ampoules).
The stabilizer depletion was also evaluated by the kinetic model which was set up during
the development of AOP-48, Edition 2[6].
The following figures give two examples of the results of stabilizer depletion on various
propellants.
346
Czech Republic
2006
[Content]
1,8
1,7
2-NDPA [%]
1,6
30C
35C
40C
1,5
1,4
1,3
1,2
1,1
1,0
0
500
1000
1500
2000
t [days]
Fig 9.
2-NDPA depletion of propellant D at 30, 35 and 40C
0,45
0,40
DPA [%]
0,35
0,30
0,25
0,20
0,15
30C
0,10
35C
0,05
40C
0,00
0
500
1000
1500
t [days]
Fig 10. DPA depletion of propellant K at 30, 35 and 40C
5.
DISCUSSION
5.1
Shift of characteristic points in the microcalorimetric curve
The shift of characteristic points in the microcalorimetric curve was presented in some
examples in figures 2 to 8. To give a better overview on all the results the most important
figures are presented in figures 11 to 17. They are sorted in alphabetical order.
5.1.1.
Propellant A
This propellant shows a very unusual behaviour. Within the first part of the ageing the shift
of the point of inflection is nearly proportional to the ageing time. If the ageing time
increases, also the shift increases. The main reason for this behaviour is a change in the type
of the point of inflection. If these points of inflection do not represent the same reaction
degree through all curves then the evaluation becomes impossible. Propellant A seems to be
10
347
Czech Republic
2006
[Content]
one of these examples, although no sign of this behaviour could be detected at 70C. Fig. 11
gives one example of this non-linear behaviour.
180,000
160,000
140,000
t [s]
120,000
100,000
80,000
60,000
40,000
20,000
0
0
500
1000
1500
2000
t [days] at 35C
Fig 11. Non-linear shift of point of inflection of propellant A at 35C
5.1.2.
Propellant D
Although not enough sample material was available to run a complete programme with
all temperatures the evaluation clearly revealed a linear correlation between shift of the
maximum and ageing time although the series at 35C shows some scattering (see fig. 12).
40C
35C
30C
Linear (40C)
Linear (35C)
Linear (30C)
140,000
120,000
100,000
t [s]
80,000
60,000
40,000
20,000
0
0
400
800
1200
1600
ageing [days]
Fig 12. Shift of maximum heat generation rates in propellant D
348
11
Czech Republic
2006
[Content]
5.1.3.
Propellant K
Both points show a nearly perfectly linear correlation between shift and ageing time for
all ageing temperatures. Figure 13 shows one example.
70,000
point of inflection
60,000
2nd maximum
Linear (point of inflection)
t [s]
50,000
Linear (2nd maximum)
40,000
30,000
20,000
10,000
0
0
400
800
1200
1600
t [days] at 35C
Fig 13. Shift of point of inflection and maximum heat generation rate (propellant K)
shift of minimum [days] .
5.1.4.
Propellant N
All selected points showed that they could be perfectly evaluated. Only the first sharp
minimum became smaller with ageing and disappeared after ca. 1000 days at 40C. The
correlation between shift and ageing time turned out to be almost linear. Also the ratio of the
slopes were nearly identical when 2 ageing temperatures are compared. The evaluation of the
2nd minimum is presented in figure 14.
0.9
40C
0.8
0.7
0.6
0.5
35C
0.4
0.3
0.2
30C
0.1
0.0
0
400
800
1200
1600
ageing time [days]
Fig 14. Shift of 2nd minimum with increasing ageing time at 40, 35 and 30C
(propellant N)
12
349
Czech Republic
2006
[Content]
t [s]
5.1.5.
Propellant P
The relation between ageing time and shift of the point of inflection was almost linear. It
is presented in figure 15.
160,000
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
40C
50C
35C
30C
0
400
800
1200
1600
ageing time [days]
Fig 15. Shift of point of inflection as a function of ageing at 4 temperatures

(propellant P)
5.1.6.
Propellant V
The evaluation was divided into two parts: The first part covers the shift of the first
maxima, minima and two inflection points whereas the second part deals with the
autocatalytic decomposition at the end of the measurement.
The four different parameters of the first part of the microcalorimetric curve work pretty
well together. The shift of all four points (first minimum, first point of inflection, second
maximum and second point of inflection) is nearly linear with time, see fig. 16.
difference [sec] .
60000
1. Min
50000
1. POI
2. Max
40000
2. POI
30000
20000
10000
0
0
500
1000
1500
ageing time at 35C [d]
Fig 16. Shift of times for first minimum, first point of inflection, second maximum
and second point of inflection after ageing at 35C (propellant V)
350
13
Czech Republic
2006
[Content]
This linear behaviour is not true for the thermal runaway. The shift in time of its
beginning (a value of 1600 W/g was taken for comparison) moves to later times after a preageing of more than 600 days. This is shown in figure 17 (later autocatalysis = negative time
shift in this figure). The reason for this behaviour is unknown. The samples aged at 35C and
30C show the same tendency although they have not yet reached the decomposition degree
which is equivalent to 600 days at 40C.
1.0
t autocatalysis [d] .
0.5
0.0
-0.5 0
400
800
1200
1600
-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
ageing at 40C [days]
Fig 17. Shift of the runaway reaction after ageing (negative times = later autocatalysis), propellant V
5.2
Activation energies between 89C and 30C as well as within the

low temperature area
The activation energies were calculated between 89C and the ageing temperature. In a
second step the activation energies between 50C and 30C were calculated as described in
chapter 4.4.
Table 3 summarises the results of the calculation of activation energies between 89C
and the storage temperature. Table 4 deals with the activation energies within the low
temperatures. Propellant A was excluded from this calculation due to its non-constant
behaviour (see fig. 11). How well the slopes fit onto a straight line is shown in fig. 18.
These data, although they are somehow scattered show that a change in activation
energy from the standard value (130-140 kJ/mole) which exists for most of the propellants
between 70 and 90C takes place at temperatures probably around 50C. If the 50C values
are more elaborate we will take them into account to get a clearer picture of the temperature
range where this change is located. No activation energy value below 80 kJ/mole was found
which is good in a sense that the test time / temperature scheme in STANAG 4582 still is a
very conservative approach and does not need to be changed.
With a more exact knowledge of decomposition activation energies it will be easy to
precisely predict the behaviour of propellants at normal or slightly elevated temperatures.
14
351
Czech Republic
2006
[Content]
Table 3. Activation energies of propellants D, K, N, P, and V between 89C and

ageing temperatures. All values in kJ/mole.
Propellant
point
Ea 89-50
Ea 89-40
Ea 89-35
Ea 89-30
maximum
129.9 3.9
127.5 4.3
129.7 5.8
poi
136.1 0.8
134.7 1.5
132.8 1.8
131.3 1.3
nd
2 maximum
134.7 0.8
134.8 1.6
133.2 1.7
131.4 1.5
maximum
154.2 3.7
141.1 2.3
137.9 3.0
142.1 4.8
2 minimum
135.4 2.4
133.3 4.1
132.4 5.1
poi
129.7 2.8
132.2 1.6
128.8 0.8
124.7 0.8
st
137
141.7 3.5
137.1 2.0
134.9 1.7
st
129
135.8 2.8
132.3 1.6
130.5 2.9
nd
128
134.1 2.6
131.7 1.5
130.6 2.2
nd
129
135.3 3.9
131.8 2.3
131.1 2.3
135.5 3.6
132.6 3.2
131.9 4.4
nd
1 minimum
1 poi
2 maximum
2 poi
average
all points, all 134.7

propellants
poi = point of inflection
- = too few data points to make a good estimation

30C. All values in kJ/mole.
Propellant
point
Ea 50-30
Ea 40-30
Ea 40-35
Ea 35-30
maximum
120
124
116
poi
123
115
112
118
nd
2 maximum
125
116
118
114
maximum
125
139*
109
169*
2 minimum
117
99
134
poi
116
95
98
91
st
100
89
111
st
104
104
104
nd
106
103
109
nd
103
84
120
nd
1 minimum
1 poi
2 maximum
2 poi
all points, all 109 9

104 12
113 12
propellants
Note: Some data for 50C are not yet available (the ageing has just been started at 50C).
The data with * were excluded from averaging.
average
352
15
Czech Republic
2006
[Content]
100000
y = 6.1843E+23e
2
R = 9.9916E-01
-1.5758E+01x
y = 7.8020E+23e
2
R = 9.9949E-01
-1.5830E+01x
a [s/d]
10000
1000
poi
2. max
100
Exponentiell (2. max)

Exponentiell (poi)
10
2.75
2.85
2.95
3.05
3.15
3.25
3.35
1000/T
Fig 18. slope a (see eq. 2) as a function of temperature in an Arrhenius plot. The
points nearly lie on a straight line (propellant K)
100000
a [s/d]
10000
1000
100
10
2.70
2.80
2.90
3.00
3.10
3.20
3.30
3.40
1000/T
Fig 19. slope a (see eq. 2) as a function of temperature in an Arrhenius plot. The
slope changes significantly at around 40C (propellant P)
5.3
Stabilizer depletion
There were no detailed studies of stabilizer depletion chemistry performed during this
study as the ageing conditions were exactly the same throughout the whole project. The only
interest was focused on the activation energy of stabilizer depletion. This is important as
many papers in literature refer to a change of activation energy below 50C which is
16
353
Czech Republic
2006
[Content]
supported by stabilizer depletion analysis. We therefore took the chance to use these samples
to compare directly microcalorimetry with stabilizer depletion. The results of the
calculations are presented in table 5.

30C (stabilizer depletion). All values in kJ/mole.
Propellant
stabilizer Ea 40/35/30 AgeKin
Ea 40-35 zero order
Ea 35-30 zero order
DPA
128
132
2-NDPA
123
68
96
DPA
111
105
107
EC
DPA
111
103
73
DPA
AgeKin = calculated by the AgeKin programme of armasuisse[7]
zero order = evaluation by simple application of a zero order kinetics

-
= no useful value available (too few data points / too much scattering)
As table 5 shows the values have a much broader scattering than the ones delivered by
microcalorimetry but in principle (except for D propellant 0 order) they lie in the same range
than the ones calculated by LTAE data. Data handling also showed that stabilizer depletion
is less reliable than heat generation, because there were several examples with increasing
stabilizer content after further ageing which makes evaluation more difficult. Probably the
ageing does not only occur in the oven at 30-50C but also soon thereafter in the laboratory.
Microcalorimetry though seems to be much more precise than HPLC.
6.
CONCLUSION
An indirect method to measure activation energies of propellant decomposition reactions

at temperatures below 50C was presented in this paper. It is reliable and it works well if
propellants with a very typical heat generation rate shape are present and if a typical point in
a heat flow curve remains an iso point throughout the ageing.
The calculated data for the activation energy lie in the range of 100-125 kJ/mole (for 40 /
30C) which is significantly lower than the usual value of about 130-140 kJ/mole but much
higher than the (estimated) 80 kJ/mole which is the low temperature part of STANAG 4582.
The extrapolation of artificial ageing at temperatures like 70 or 89C to room temperature is
possible if the activation energies gained by this LTAE approach are used. Although the
propellants used for this study differed a lot they more or less show the same behaviour
concerning activation energies which shows that nitrocellulose itself is responsible for the
activation energy of decomposition of propellants.
It must be noted that all Ea values discussed in this paper refer to the early stages of
propellant decomposition ( from 0 to max. 0,05). It can not be assumed that these Ea values
are valid for the main decomposition reaction as well.
354
17
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
18
PIERRE GUILLAUME, MAURICETTE RAT, GABRIELE PANTEL, STEPHAN WILKER, Heat flow
calorimetry of propellants Effects of sample preparation and measuring conditions, Propellants, Explosives, Pyrotechnics 26, 51-57, (2001). See also A.H. HEEMSKERK, Stability of
Nitrocellulose Propellants, Symp. Chem. Probl. Connected Stabil. Explos. 8, 137-148 (1988).
DAN S. ELLISON, ANTON CHIN, Understanding the true driving force in the auto-ignition of
single and double base propellants, Proc. Int. Pyrotech. Sem. 28, 245-261 (2001); ANTON
CHIN, DAN S. ELLISON, Mechanistic approach to study the moisture and acidity effect on the
stability of single and double base propellants, Proc. Int. Pyrotech. Sem. 28, 173-186 (2001).
STEPHAN WILKER, GERHARD HEEB, BEAT VOGELSANGER, JAN PETRLEK, JAN SKLDAL,
Triphenylamine a new stabilizer for nitrocellulose based propellants Part I : Chemical
stability studies, Propellants, Explosives, Pyrotechnics 32, 2006, in press.
STANAG 4582, Explosives, Nitrocellulose Based Propellants, Stability Test Procedures and
Requirements Using Heat Flow Calorimetry, 1st Edition, March 2004.
JIM DODDS, GAYNOR KAVANAGH, DAVID TOD, RICHARD WHITE, Bulk Ageing Effects in
double base propellants, Int. Annu. Conf. ICT 30, 81 (1999); Y.-J. YIM, B.-N. RYOO, Y.-C.
PARK, Stabilizer Consumption by Accelerated Aging of PEG/RDX Propellant, Int. Annu.
Conf. ICT 29, 45 (1998); ANDR SOPRANETTI, H.U. REICH, Comparison of high-temperature
stability tests with studies of stabilizers in single-base propellants, Symp.Chem.Probl.
Connected Stabil.Explos. 5, 459-477 (1979); JIM DODDS, Current UK Procedures for
Propellant Shelf-Life Determinations, Symp.Chem.Probl.Connected Stabil.Explos. 9, 247-258
(1992); THOMAS B. BRILL, P.E. GONGWER, Thermal Decomposition of Energetic Materials
69. Analysis of the kinetics of nitrocellulose at 50-500C, Propellants, Explosives,
Pyrotechnics 22, 38-44 (1997); MAX FREY, Mikrokalorimetrische Untersuchungen an Nitratestertreibstoffen, Symp.Chem. Probl.Connected Stabil.Explos. 8, 15-30 (1988); MANFRED A.
BOHN, Comparison of Surveillance Methods for Gun Propellants - Performances and
Limitations, Proc. Finnex Seminar (2002).
AOP-48, Ed. 2, Nitrocellulose based propellants - Stability test procedures and requirements
using stabilizer depletion, CNG Draft 8 (2005).
AgeKin Software, programmed by armasuisse for NATO-AC/326 SG I to support AOP-48,
Version 1.0 (2005).
355
Czech Republic
2006
[Content]
THEORETICAL STUDY OF THE TETRAZINES C-N

HETEROCYCLIC DERIVATIVES
Zhou Yang1 Shu Yuan-Jie1* Wang Xin2** Long Xin-Ping1 Tian An-Min2
(1 Institute of Chemical Materials, China Academy of Engineering Physics, Mianyang 621900)
(2 Faculty of chemistry, Sichuan University, Chengdu 610064)
Abstract: 31 tetrazines C-N heterocyclic derivatives have been investigated by density
functional theory. Their optimized geometry structures, electronic structures,
conjugation, energies and heats of formation (HOF) were calculated and predicted at
the B3LYP/6-311G(d,p) level. The results show that most of the species keep a planar
structure and exit considerable conjugation over the whole molecule, which enhances
the stability of these derivatives. There is a good linear relationship between the energy,
HOF and N atoms in these species. Our study shows most of the species have high
HOFs and relative stabilities. The HOFs of these compounds are between 571.2KJ/mol
and 827.2KJ/mol, so they may be potential candidates for energetic materials.
Keywords: tetrazine, C-N heterocycle, theoretical study, heat of formation
* E-mail: syjfree@sohu.com
** E-mail: wangxin@scu.edu.cn
This work was supported by the Foundation of NASF (NO.10576030)
1.
INTRODUCTION
Energetic materials are used extensively both for civil and military applications. In order
to meet the requirements of future military and space applications, there are continuous
researches on how to develop new energetic materials with higher explosive performance
and enhanced insensitivity to thermal or shock insults than the existing ones. In the past,
high explosive performance has always been a prime requirement in the field of research and
development of explosives and common explosives hexahydro-1,3,5-trinitro-1,3,5-triazine
(RDX), octahydro-1,3,5,7-tetranitro-1,3,5,7-etrazocine (HMX) were considered adequate for
all weapon applications. But in modern ordnance, because many catastrophic explosions
results from unintentional initiation of munition by either impact or shock, aboard ships,
aircraft carriers, and munition trains. The explosives with the good thermal stability, impact
and shock insensitivity, and better performance are strongly required. However, these
demands are somewhat mutually exclusive. The explosives with good thermal stability and
impact insensitivity usually exhibit poor explosive performance and vice versa. Therefore,
the foremost objective at the stage of synthesis of new explosives consists of finding the
molecule with both good energy capability and optimal safety (reduced vulnerability, shock
and impact insensitivity) to those in current use[1].
Modern high-energy-density materials (HEDM) derive most of their energies either (i)
from oxidation of the carbon backbone, as with traditional energetic materials [2,3]or (ii) from
their very high positive heat of formation. Examples for the first class are traditional
explosives such as TNT, RDX and HMX [4]. Modern nitro compounds such as CL-20 or the
recently reported hepta- and octanitrocubanes[5] possess very high densities and have
enhanced energies due to substantial cage strain. And the second class of compounds is the
high-nitrogen energetic materials, which derive most of their energy from their very high
positive heats of formation[6]. In recent years, high-nitrogen energetic organic compounds
356
Czech Republic
2006
[Content]
have attracted great attentions because of their novel properties, such as high density, high
positive heat of formation and high thermal stability. The large positive heats of formation of
nitrogen-rich compounds make them promising targets for the synthesis of highly energetic
materials, as illustrated by the recent examples of 3,3-azobis (6-amino-1,2,4,5-tetrazine) and
hydrazinium N,N-azobis tetrazolate[7]. Among the synthesized high-nitrogen C-N
compounds, the tetrazine derivatives are one kind of potential and luminous candidates for
high performance insensitive explosive. However, there are few theoretical studies on the
tetrazine derivatives.
To help find new stable nitrogen-rich compounds and obtain a more comprehensive
picture of the stability of such high-energy compounds, in this paper we studied the structure
and stability of a large number of substituted tetrazine compounds by density functional
theory (DFT).
2.
COMPUTATIONAL DETAILS
Theoretical calculations were performed to investigate 31 substituted tetrazine

compounds by density functional theory (DFT)[8] with hybrid B3LYP function[9,10]. The 6311G(d,p) basis set was used for all geometry optimizations. The self-consistent field (SCF)
convergence criterion is set to 10-8. The harmonic vibrational frequencies were calculated to
obtain the zero point energies (ZPE) and verify if it is a minimum on potential energy
surface (PES). The natural bond orbital (NBO) [11,12,13,14] analysis has been carried out using
the optimized geometries. The thermodynamics in a vacuum was computed by correcting the
differences in electronic energy to enthalpies at 298 K upon inclusion of zero-point energy
and thermal corrections. All these calculations were carried out using the Gaussian 98
program [15].
3.

3.1
Single five and six-membered C-N heterocyclic compounds
3.1.1. Geometry analyses

Firstly, we optimized geometries of tetrazine and triazine with different base sets to find
a suitable basis set. And the results were compared with the experimental data, which were
illustrated in table 1. The calculated bond lengths with the 6-311G(d,p) base set is close to
experimental values, and the calculation is relatively inexpensive. Thus, we use 6-311G(d,p)
basis set for all the five and six-membered C-N heterocyclic compounds. The optimized
structures were displayed in Figure 1.
All the molecules have a rigid plane structures, and bond lengths of these single rings
have a averaged tend. In the five-membered C-N heterocyclic molecules the shortest C-C
bond length is 1.370, the longest is1.424. These bonds are between the standard C-C
(1.54) and C=C (1.34). And the C-N bond lengths are between 1.306 and 1.379. The
shortest is approximate to the standard value (1.30) of C=N and the longest is less than
normal C-N bond length (1.47). The N-N bond lengths are between 1.286 and 1.384 .
However in the six-membered C-N heterocyclic molecules, the C-N bond lengths (1.333
~1.336) have smaller change. The N-N bond length is 1.322 in the molecule T, and
1.333 in the molecule 2l. These values were between the standard value (1.460) of N-N
and the standard value (1.25) of N=N. And the C-C bond lengths (about 1.39) are near to
the standard C=C bond.
357
Czech Republic
2006
[Content]
Table 1. The experimental data and theoretical data of 1,2,4,5-tetrazine and 1,3,5triazine
1,2,4,5-tetrazine 1,3,5-triazine
B3LYP
NN
CN
CN
6-311G(d,p)
1.322
1.336
1.334
6-311G++(d,p)
1.321
1.336
1.334
cc-pvdz
1.324
1.340
1.338
cc-pvtz
1.318
1.333
1.331
aug-cc-pvdz
1.323
1.340
1.338
aug-cc-pvtz
1.317
1.333
1.331
1.321
1.334
1.338
1.37
0
1 .3
96
2l
Fig 1.
114.1
N5
.9
11
5
2d
C6
2.3
N4
1.335
.3
1 09
115
.9
11
8
1
C2 .335
127.2
N3
12
3.7
.4
1.
33
N1
C4
1
C5 1.39
116.5
2j
N2
N3
C4
C3
.0
122
C5
4
0.
11
C5
134
1.394
N1
C6 126.4
C6
1.35
1.41
1.35
5
12
4.8
11
1 .3
11
113
.
10 7
.7
106.0
C2
N4
.4
11 N1
1.333
1.322
107
33
N2
N1
4.
4
N1
36
1.3
4
9.
N1
C2
C3
118.5
92
C4 1.3
C5
10
1.3
123.7 C6
C3
8 10
C5
1.373
6 N3 1.3
84
30
123
C6
34 C2
1.3
125.9
5
3.
1.36
10
5.5 06.2
1
N3 1.306 N4
42
1.381
11 C5
6 .9
8.
C4
1.5
C6
N1 1.336
117.2
1
C4
10
1.
C2
11
1.32
41
1
5.
7.
2 10
N3
C4
1.379
I2
1
7 N1 .32
5
32
1. 100.6
N2
C5
1.394
.7
8 .4
8.5
10
N1
C5
1.346
C4
.9
112
105
1.35
10
10
10
P1
N3 1.36
86
5
1.2 111.3
N2
N4
.1
1.379
P2
N2 1.3
4
99
1.2 107.0 9
N3
N1
1.36
C3
C3
C5
7
37 110.7
4.5
66
1.3
1.424
10
6
.0
C4
.4
7
10
1.37
8
1.35
C2
N1
12
1.3 105.4
C2
7 N1 1.32
34
104.2 9
C5
1.
1.
N2
2
11
74 N1
1.3 109.8
C5
111
.6
Experiment
[16,17]
N5
C3
6.8
11
N4
the theoretical data of the bond lengths and bond angles for C-N
heterocyclic compounds
3.1.2. Conjugation of the single rings

From the Figure 1it can be seen that the average trend of bond lengths for triazine is
the most obvious, of which all bond lengths are 1.334In addition, the stabilization
interaction energies E(2), are calculated by means of the second-order perturbation theory. In
the NBO analysis, E(2) is used to describe the delocalization trend of electrons from the
donor bond to the acceptor bond. Through the analyses, we find that there exist strong
donor-acceptor interactions within all the single rings. Interaction energies E(2) (38.23
kcal/mol) between bonding orbital and * antibonding orbital in the triazine ring is the
maximum within all the systems, and E(2) (11.38Kcal/mol) between BD N2-N3->BD* N4C5 in the 41 molecule is the minimum, where BD and BD* represent bonding and
antibonding natural bond orbital. As discussed above, we find that the triazine molecule
358
Czech Republic
2006
[Content]
contains the best conjugation system, which means this structure is stable. For the tetrazine
interaction energies E(2) (24.51kcal/mol) between bonding orbital and * antibonding
orbital also shows that the tetrazine is a stable molecule, too. Through molecular orbital
analysis, we find a delocalized occupied orbital composed purely of 2PZ orbitals of all
carbon and nitrogen atoms of the ring in all derivatives.
3.2
Substitute of the single ring

3.2.1 Geometry analyses
11
8.6
1.3
37
121
.9
4
1.33
.8
121
6.2
11
12
2.2
1.3
36
0
1.4
0
7.8
46
1 .3
1.3
11
7.1
91
11
6.0
11
1.3
36
.2
122
1.3
35
7.2
11
35
9.8
11
1 .3
02
1.3
3
11
8.0
6
122
.
1 .4
1.3
45
117
.5
6 .3
11
1.3
34
.2
11
4
.8
125
34
40
1.3
.0
117
3
1.3
3
12
3.5
117
.
36
1.3
12
3.8
87
1.3
1.3
44
1.3
41
1.3
.0
11
8
11
7.8
C7 126.0
11
6.9
C11
N12
.0
N9
117
11
7.1
1.4
0
1.3
38
1.3
7
1.3
3
1.3
44
36
11
7.2
12
11
8.5
46
1.3
11
6.8
3. 9
1.3
98
117
.9
.1
117
7.2
11
117
.0
117
.4
36
39
1.333
126.2 C10
.0
1. 3
N12
34
114.2N10
5.8
12
N4
1.3
C9
C7 125.8
34
1.3
36
12
3.5
8.0
11
34
7
11
7.2
95
1.3
1.
5
1.3
3
8.9
11
11
7.8
8.1
11
11
7.5
1.3
46
34
1 .3
1.3
36
35
1. 3
1.3
35
95
1. 3
35
1.3
4 .2
1.495
1.333
11
125.8 C3
1.319
N8
40
1 .3
N2
7.0
11
1 .3
94
1.3
N11
117
TT
7.0
6 .9
N4
11
36
1.3
11
39
1.488
1.325
T3
N8
1.3
36
1 .3
Fig 2.
1.319
N5
N11
1.331
1.319
C6 126.1
119.3 N10
T2l-b
N2
C3
N5
33
1.3
C12
N1
C9
9.8
11
N4
7.0
11
1.319
C7 116.6
04
1 .4
1.316
C6
45
7.7
11
36
N1
1.3
7.0
11
1.3
N5
1.393
C8
1.477
1.3
N12
T2l-a
N2
124.5 C3
C6 125.9
123.5C10
9.5
N4
C9
11
1.315
6. 8
1.316
37
N5
11
C11
1. 3
1.331
1.380
C8
124.8 C3 1.487 C7 123.2

7.1
11
N12
C6 125.9
1.336
N2
T2j-c
N1
C12 1.332 N11
42
1.3
115.7N10
37
1.3
N4
1.317
1.317
N1
1.335
7.1
12
37
7.2
11
1.3
C7 121.8
1.491
1.344
1.337
C9
6.7
11
.7
1.396
124.8 C3
N5
127.1C10
6.2
11
1.390
C8
6.4
11
6
11
C6 125.9
N9
T2j-b
N2
1.342
2.2
N4
12
1.315
1.332
C7 116.2
00
N5
1.473
1 .4
C11
124.3 C3
7 .8
11
36
1.317
N11
1.334
C8
46
1 .3
7.1
11
1.3
1.334
N2
T2j-a
N1
96
1.3
C12
6.4
11
N4
1.315
N1
C6 125.8
116.6C10
2 .2
12
N12
91
1.3
6.0
35
N4
C9
C7 127.0
1 .3
7.2
11
35
1.319
1.334
N8
11
41
1.3
.0
117
1.3
N5
1.495
121.8C10
T2d
N2
125.5 C3
C6 126.1
C9
C7 121.5
3
1.316
1.487
0
1.4
3.9
8.
11
35
1.3
12
1.319
7.6
11
124.7 C3
1.329
N8
.6
.5
98
1.3
7 .9
N5
C11
1.391
N2
T1-c
N1
123.6C10
C12 1.333 N11
6
11
7.4
37
8
11
46
11
36
C12
N4
C6 125.9
117.1N10
C7 118.2
1.479
1.317
N1
11
C9
1.3
1.3
.1
117
1.3
1.316
1.391
C8
124.2 C3
C6 125.8
N5
N2
C9
T1-b
45
1 .3
1.316
C7 118.0
03
N4
T1-a
N1
1.474
1 .4
7. 9
1.315
.7
118
124.0 C3
11
36
N5
C11
47
.1
1.3
.6
123
1.331
117
C6 125.8
1.387
C8
N2
1.3
9
1.3
8.4
7.3
N12
1.316
N1
C9
118.3C10
C7 123.1
39
1 .3
11
36
1.3
N4
1.490
11
124.4 C3
1.317
1.390
C8
45
1.3
6 .8
11
C6 125.8
N5
N2
1.4
00
1.317
N1
N11
the molecular structures and parameters of the tetrazine substituted by sixmembered C-N heterocycle
359
4
1.3
5
51
1.3
1.355
5
1.3
3
49
1.3
1.381
1.375
1.
37
1
1.
38
7
1.3
3
1.396
1.3
38
7
1.3
3
1.358
1.3
71
1. 3
35
1.321
1.3
35
N9
C8
T134-1
1.3
36
C3
C11
1.462 1
.3
18
C6
42
1.377
N9
N2
1. 3
N5
1.316
N4
T41-2
N7
1.339
N8
1.295
1.316
1.3
51
4
38
1.3
50
1.3
45
35
1.3
N10 1.35
N9
5
1
1.3
6
1.469
43
1.3
96
N10
N4
T42-2
N9
1.333
37
1. 3
6
1.3
3
6
1.365
1.3
3
37
1.3
1.381
1.387
1.363
96
1 .3
1. 3
43
1.
40
1.3
1.373
1.308
1.315
0
1.3
7
46
1.3
1.2
N7
N4
38
C11
1.3
34
1.4
31
34
8
1.3
4
35
1.3
1.336
1.3
80
7
1.3
4
1 .3
46
1.388
5
1.3
3
1.424
1 .3
80
1
1.3
4
8
1.3
4
1.
39
1.435
53
1.3
40
96
1.3
1.3
36
5
1.3
3
38
1 .3
37
1. 3
39
1.3
34
1.390
42
1.3
1.3
1.316
1.3
36
36
1. 3
N9
N8 .284
1
C11
4
1.3
C3
1. 3
C3
3
1.3
1.317
T123-1
1.315
N1
N2
N5
63
1.3
N7
77
1 .3
37
1 .3
42
6
N5
1.399
N4
N1
71
1.3
1. 3
51
N10
92
C11
N7
1
C11 .363
C10
C3
1.315
1
N9 .345
C10
1.
N10 300
N2
C6
N8
C9 1.347
TI2-2
C6
N10
N8 1.275
T41-1
1.314
N4
C6
1
C11 .302
N7
1.402
1.317
3
1.3
1.3
T42-1
1. 2
64
N7
C8
1.454
1.316
N1
N2
N5
N8
N9
1
1.29
3
1.
N4
1.317
1.3
4
N7
36
1.3
36
Fig 3.
C11
N8
N9
N4
C8
1.355
1
C10
T123-3
C3
N1
C11
N2
3
1.3
1.3
1.412
1.458
C6
N5
N7 1.316
81
.3
36
1.316
N10
N9 1.369
N2
N4
1.318
N1
C11
T134-2
C3
1.317
N5
3
1.3
N4
C6
N5
C3
1
N7 .357
5
31
1.
1.318
C6
35
C3 1.458
C8
N8
1.314
N1
N2
1.3
60
45
N1
N7 336
1.
T123-2
1.3
39
1.3
1.314
N9
C11
N2
C6
N5
N9
1.
1.448
C101.356
82
1. 3
1.317
N7
82
35
1 .3
N5
N4
1.316
1
C11 .364
C10
N5
N4
1.316
48
1. 3
4
1.3
1. 3
C3
C3
TI2-4
1.313
N1
N2
N1
1.441
TI2-3
C6
C6
N5
1.3
N4
C3
34
C8
N7 .306
1
79
1.318
C6
1 .3
C11
1.3
34
N5
1.315
N1
N2
38
1 .3
51
1.3
1.457
N9
52
1. 3
1
C10 .367
C3
N7
C8 1.299
1.311
N1
N2
1.3
1.313
N1
N2
C6
C11
TI2-1
TP2-4
1.32
2
TP2-3
C10 1.360
N4
N4
22
1.3
1.334
1.313
88
N5
1.314
38
1.3
N7
N5
N9
1.386
1.3
C3
C11
1
42
1.
C10
C3
1.449
48
1.3
C6
N7
C8 1.361
50
1.3
TP2-2
N2
C6
N8
N7 .335
1
92
1 .3
N8
N5
N4
1.315
C11
1.316
N2
N1
37
63
1.3
3
1.3
1.448
N5
N4
1.316
1.3
C3
1.4
8C10 05
C11
1.462
N1
1.316
7
1.33
8
1.3
C9
C3
41
1.3
4
1.3
C6
C9
1.448 1
.3
87
N5
N4
1.312
1
C10 .374
19
C9
1 .4
36
1.3
N2
C6
C11
N7 1.317
TP2-1
1.316
N2
N1
N8
1.392
1.314
N1
C3
TP1-3
1.34
3
1.3
N5
N4
1.316
1.3
66
C10
37
1 .3
C3
C9
4
36
1.
42
1.3
C6
C6
TP1-2
1.315
N2
C11
N7 60
1.3
1
C10 .369
C11
N2
1.3
78
TP1-1
N1
1.439 C8
N5
N4
1.314
1.316
N1
C10
3
1.3
C3
C9 1.410
1.3
C11
N4
C6
35
C10
N5
N2
[Content]
51
N7
1.381
1.311
2006
1. 3
N1
C9
1.3
C3
C6
1
C8 .363
54
1.3
5
1.34
1.314
N1
N2
Czech Republic
1. 3
N8
N7 .318
1
the molecular structures and parameters of the tetrazine substituted by fivemembered C-N heterocycle
As some investigations shown, many organic molecules exhibit much different property
when its group substituted. As discussed above, the conjugation of traizne is the best, but its
nitrogen content is no more than that of tetrazine, and at present many known high-nitrogen
energetic materials take the tetrazine as the parent structure because of its better conjugation
and high-nitrogen content. So we discuss the structures and the electronic structures of these
derivatives coming from C-N heterocycle directly substitute a hydrogen atom of the
tetrazine, and we gained the thirty-one molecules aggregately. At first, we optimize the
geometry of all the molecules at the B3LYP/6-311G(d,p) level. The results see the Figure 2
and Figure 3. As shown in the two Figures, two rings of the molecules T2j-a, T2l-a, T3, TT
360
Czech Republic
2006
[Content]
and T42-1 are not in a plane, their dihedral angels are 37.3,16.8,41.2 and 24.9, and the
other molecules have a plane structure. We fixed these non-plane molecules at a plane, then
optimize their structures, at last we find a imaginary frequency and higher molecular energy.
The results show that the non-plane structure of these molecules is the most stable.
From the two Figures, we can know that after a hydrogen atom of the tetrazine is
substituted, the NN bond lengths become shorter, the CN bond lengths adjacent to the
substituent become longer and the CN bond lengths far from them have no change basically.
From the Figure 2, we can find that for the same substituent the different place connected
with the tetrazine has a different influence on the C3-C7 bond length between the two rings,
such as these molecules T2j-a,T3-a and TT, their C3-C7 bond lengths are 1.495,1.495
and 1.488, which are more than that of T1-a,T2d,T2j-c and T2l-a, because a hydrogen
atom of the subsituent is closer to one nitrogen atom of the tetrazine in the latter; the other
molecules have two hydrogen atoms closer to the nitrogen atoms of the tetrazine, their C3C7 bond lengths are the shortest correspondingly. This trend is more obvious in the isomers
T2j-a,b,c. The distance between the hydrogen and the nitrogen of tetrazine ring is between
2.447~2.531, there maybe exist the weak interaction, but there are no the bond critical
point between the two atoms from the results of AIM analyses, namely there dont form the
hydrogen bonding. As the Figure 3 shown, the C-N bonds of the subsituent connected the
tetrazine in the molecules TP1-1,TP2-1,TI2-1,T123-1,T134-1, T41-1and T42-1 are close to
1.40, the shortest is 1.381 and the longest is 1.412; the C-C bonds of the other
molecules are more than 1.40 (between 1.439~1.469). In a word, these bonds are
between the standard single bond and double bond.
3.2.2 Electronic structures and conjugation
The net charge of all the atoms is calculated by the NBO method. A hydrogen atom of
the tetrazine ring is taken as a substituent and the C-N heterocycle is taken as a whole, the
total charge of all the substituents are calculated. Table 2 illustrates the NBO charge of seven
atoms on the tetrazine rings and the total charge of the tetrazine ring and the subsitituent.
Table 2. The atoms NBO charge distribution of the tetrazine ring and the total
charge of two rings/(a.u.)
T
T1-a
T1-b
T1-c
T2d
T2j-a
T2j-b
T2j-c
T2l-a
T2l-b
T3
TT
TP1-1
TP1-2
TP1-3
TP2-1
TP2-2
TP2-3
TP2-4
TI2-1
N1
N2
C3
N4
N5
C6
-0.205
-0.200
-0.193
-0.192
-0.195
-0.202
-0.189
-0.196
-0.198
-0.189
-0.197
-0.192
-0.188
-0.196
-0.201
-0.188
-0.202
-0.197
-0.193
-0.184
-0.205
-0.229
-0.226
-0.217
-0.181
-0.187
-0.220
-0.221
-0.228
-0.216
-0.182
-0.182
-0.243
-0.215
-0.228
-0.245
-0.223
-0.222
-0.206
-0.238
0.215
0.355
0.372
0.370
0.348
0.342
0.367
0.347
0.352
0.364
0.332
0.329
0.559
0.341
0.371
0.546
0.353
0.368
0.345
0.555
-0.205
-0.180
-0.220
-0.217
-0.223
-0.187
-0.220
-0.174
-0.174
-0.212
-0.182
-0.182
-0.243
-0.258
-0.230
-0.199
-0.192
-0.227
-0.240
-0.235
-0.205
-0.198
-0.192
-0.192
-0.196
-0.202
-0.189
-0.194
-0.193
-0.189
-0.197
-0.192
-0.188
-0.196
-0.202
-0.187
-0.200
-0.198
-0.193
-0.183
0.215
0.206
0.202
0.207
0.210
0.213
0.206
0.214
0.211
0.211
0.219
0.221
0.182
0.183
0.186
0.189
0.199
0.192
0.197
0.189
0.194
0.192
0.194
0.195
0.194
0.192
0.196
0.195
0.194
0.197
0.195
0.197
0.195
0.190
0.189
0.196
0.191
0.191
0.195
0.198
Total charge of
the tetrazine rings
-0.196
-0.054
-0.063
-0.046
-0.043
-0.031
-0.049
-0.029
-0.036
-0.034
-0.012
-0.001
0.074
-0.151
-0.115
0.112
-0.074
-0.093
-0.095
0.102
Total charge of
the substituents
0.194
0.055
0.063
0.048
0.044
0.030
0.050
0.029
0.035
0.033
0.010
-0.001
-0.074
0.150
0.115
-0.110
0.073
0.095
0.094
-0.102
361

TI2-2
TI2-3
TI2-4
T123-1
T123-2
T123-3
T134-1
T134-2
T41-1
T41-2
T42-1
T42-2
-0.197
-0.203
-0.193
-0.185
-0.188
-0.198
-0.179
-0.193
-0.180
-0.188
-0.184
-0.195
-0.251
-0.228
-0.208
-0.239
-0.201
-0.226
-0.232
-0.245
-0.233
-0.230
-0.191
-0.182
0.329
0.352
0.342
0.542
0.342
0.349
0.552
0.329
0.537
0.325
0.525
0.325
-0.181
-0.195
-0.253
-0.191
-0.236
-0.188
-0.232
-0.172
-0.189
-0.167
-0.190
-0.188
-0.194
-0.201
-0.194
-0.182
-0.189
-0.195
-0.179
-0.189
-0.177
-0.185
-0.183
-0.196
Czech Republic
0.196
0.194
0.190
0.198
0.204
0.201
0.196
0.206
0.204
0.216
0.209
0.212
0.193
0.190
0.193
0.199
0.197
0.193
0.201
0.197
0.203
0.200
0.201
0.196
2006
-0.105
-0.091
-0.123
0.142
-0.071
-0.064
0.127
-0.067
0.165
-0.029
0.187
-0.028
[Content]
0.104
0.091
0.123
-0.143
0.071
0.066
-0.129
0.068
-0.165
0.029
-0.187
0.027
As showed in the Table 2, after the C-N heterocycle substitutes a hydrogen atom
connected with the atom C3 of the tetrazine, the positive charge of C3 atom increases
obviously. The part electrons on the tetrazine rings shifts to the substituent. It indicates that
the C-N heterocycle as the substituent has an electron-absorbing characteristic. The electronabsorbing characteristic of the six-membered C-N heterocyclic substituent would boost up
along with the increase on the number of nitrogen atom. And the electron-absorbing
characteristic of the five-membered C-N heterocyclic substituent is related to not only the
number of nitrogen atom, but also the place of the C-N heterocycle connected with the
tetrazine. When the nitrogen atom of the C-N heterocycle connected with the tetrazine ring,
the electron-absorbing characteristic is stronger than the instance joining with the carbon
atom. As above discussed, the electron-absorbing characteristic of the substituent 42 is
strongest among all the whole system.
Table 3. the part stabilization interaction energies E(2) of the tetrazine substituted
by six-membered C-N heterocycle/(Kcal/mol)
T1-a
T1-b
T1-c
T2l-a
362
Donor
Acceptor
E(2)
LP C3
BD* N1-N2
122.24
LP C3
BD* N4-N5
109.61
LP C3
BD* C7-C8
32.67
BD N1-C6
BD* N2-C3
8.76
BD N1-C6
BD* N4-N5
21.38
T2d
T2j-a
Donor
Acceptor
E(2)
LP C3
BD* N4-N5
118.31
LP C3
BD* N1-N2
108.09
LP C3
BD* C7-C12
31.91
LP C3
BD* N1-N2
116.96
LP C3
BD* N4-N5
116.95
BD N2-C3
BD* N1-C6
20.69
LP C3
BD* C7-N8
28.96
BD N2-C3
BD* N4-N5
19.21
BD N1-N2
BD* C3-N4
24.09
BD N4-N5
BD* N1-C6
22.73
BD N1-N2
BD* N5-C6
22.74
BD N4-N5
BD* N2-C3
23.91
BD C3-N4
BD* N1-N2
19.32
BD C7-C12
BD* N2-C3
24.79
LP C6
BD* N1-N2
112.89
LP C6
BD* N4-N5
BD C7-C8
LP* C3
BD C3-N4
BD* N5-C6
20.35
BD N5-C6
BD* N1-N2
21.46
112.89
BD N5-C6
BD* C3-N4
19.10
52.06
BD C7-C12
BD* C3-N4
21.88
BD N1-C6
BD* N2-C3
19.15
BD N1-C6
BD* N4-N5
20.26
T2j-b
T2j-c
LP C3
BD* N1-N2
120.48
LP C3
BD* N5-N4
108.88
BD N2-C3
BD* N1-C6
19.36
LP C3
BD* C7-C8
33.61
BD N2-C3
BD* N4-N5
19.07
BD N1-C6
BD* N2-C3
19.52
BD N4-N5
BD* N1-C6
24.09
BD N1-C6
BD* N4-N5
20.35
T2l-b
BD N4-N5
BD* N2-C3
24.22
BD N2-C3
BD* C7-N12
10.60
Czech Republic
T3
2006
[Content]
BD N2-C3
BD* N1-C6
18.87
BD N2-C3
BD* N4-N5
20.92
BD C7-N12
BD* N2-C3
13.91
BD N4-N5
BD* N1-C6
24.75
BD N1-N2
BD* C3-N4
24.13
BD N4-N5
BD* N2-C3
24.32
BD N1-N2
BD* N5-C6
23.06
BD C3-N4
BD* N1-N2
19.70
TT
BD C3-N2
BD* C6-N1
18.57
BD C3-N2
BD* N4-N5
20.75
BD C3-N4
BD* N5-C6
19.90
BD C6-N1
BD* C3-N2
19.90
BD N5-C6
BD* N1-N2
21.31
BD C6-N1
BD* N4-N5
20.51
BD N5-C6
BD* C3-N4
19.24
BD N4-N5
BD* C3-N2
24.48
BD C7-C8
BD* C3-N4
21.59
BD N4-N5
BD* C6-N1
24.61
In order to study the interaction between the substitutent and the tetrazine ring, the
stabilization interaction energies E(2) are calculated by means of the second-order
perturbation theory. The selected stabilization interaction energies E(2) for the molecules
composed of six-membered C-N heterocycle and tetrazine ring at the B3LYP/6-311G(d,p)
level are summarized in the Table 3, where BD and BD* represent bonding and antibonding
natural bond orbital and LP represents lone pairs. From the Table 3, we find that the
interaction between the bonding and antibonding orbital of the tetrazine ring in the systems
T1-a,T2d,T2j-a and T2j-c becomes weak (their values smaller than 10Kcal/mol are not
included in Table3). But Interaction between the lone pairs of C atom and * antibonding
orbitals of the tetrazine ring becomes strong, E(2) values are more than 100Kcal/mol. And in
the other systems, interaction between the bonding and antibonding orbital has no change
basically. Except for T3 and TT, there exit strong interaction between the two rings. In
addition, the results of NBO analyses show that when five-membered C-N heterocycle
connected the tetrazine ring, there exit strong donor-acceptor interaction between the LP of
N atom in the C-N heterocycle and the CN * antibonding orbitals in the tetrazine ring (see
Table 4). Because of this interaction, the lengths of CN bonds connected the two rings are
less than the lengths of CC bonds. The results agree with the analyses of the structures.
Table 4. the part stabilization interaction energies E(2) of the tetrazine substituted
by five-membered C-N heterocycle/(Kcal/mol)
TP1-1
TP2-1
TI2-1
T123-1
T134-1
T41-1
T42-1
Donor
LP N7
LP N8
LP N9
LP N7
LP N7
LP N9
LP N8
Acceptor
BD* C3-N4
BD* N2-C3
BD*N2-C3
BD* N2-C3
BD* N2-C3
BD* N2-C3
BD* N2-C3
E(2)
56.10
48.97
52.95
45.61
49.74
42.89
32.48
In addition, from the analyses of molecular orbitals, we get much conjugate information
about the molecules of the tetrazine substituted by the single ring, Figure 4 shows the
delocalized occupied orbital in all the molecules. The figure shows that there exit a
delocalized occupied orbital composed of purely 2PZ orbital of all carbon and nitrogen
atoms in the molecules T3,TT and the isomers TP2,TI2,T123, T134, T41 and T42. Although
other molecules have no orbital over the whole molecule, interaction between the two rings
dont destroy the primary conjugation of the tetrazine ring, and the conjugate orbital expands
its range to some atoms of the substituent. As above analyzed, after a hydrogen atom of the
tetrazine is substituted by the C-N heterocycle, there exit strong interaction between the two
rings. This interaction dont destroy the delocalized orbital of the tetrazine, and in some
8
363
Czech Republic
2006
[Content]
case it can form a delocalized occupied orbital over the whole molecule. This delocalized
orbital is advantage to the stabilization of the whole molecule.
T1-a
T1-b
T1-c
T2d
T2j-a
T2j-b
T2j-c
T2l-a
T2l-b
T3
TT
TP1-1
TP1-2
TP1-3
TP2-1
TP2-2
TP2-3
TP2-4
TI2-1
TI2-2
TI2-3
TI2-4
T123-1
T123-2
T123-3
T134-1
T134-2
T41-1
T41-2
T42-1
T42-2
Fig 4.
3.3
The delocalized occupied orbitals of the tetrazine substituted by the C-N

heterocycle
Energy and Heat of formation(HOF)
3.3.1 Energy analyses

As showed in Table 6, with the increase of the nitrogen atoms in the C-N heterocyclic
substituent, the molecular energy decreased gradually. The isomer with the minimum energy
is considered as the steadiest structure. Figure 5 shows the relation of energy and nitrogen
atom numbers. The results show that the total energy will decrease gradually with the
increase of the nitrogen atoms, and it submitted good linearity relation. The total molecule
energy decreased about 16a.u when increasing a nitrogen atom.
Fig 5.
The relationship between the total energy and N atom number for C-N
compounds
3.3.2 Heat of formation analyses
364
Czech Republic
2006
[Content]
Heat of formation is a critical parameter of energetic materials and the calculation of the
velocity of detonation. It is a key to design and synthesize new energy materials. And highnitrogen energetic materials possess very high positive heat of formation, which attracts
many researchers attention. So the calculation of HOF is a significant matter for theoretical
study and experimental study. In this paper, HOF of the objective compounds are calculated
through atomic scheme. The different methods were adopted to calculate the academic
values of HOF for the single six-membered C-N heterocycle, and these results are compared
with the experimental data, which are illustrated in Table 5. The point energy is calculated at
the B3LYP/6-311+G(2df,2p) level, then it is corrected by the ZPE which is calculated at the
B3LYP/6-311G(d,p) level. Finally we take corrected molecular energy as a parameter to
calculate heat of formation; the results see the last column of Table 5. Form the data of Table
5, we find that the results of this method show a good agreement with the experimental data,
and the most error is 19.4KJ/mol. So we calculated the heat of formation for all the
molecules using the above method discussed. The results see the Table 6.
Table 5. Tthe experimental data and theoretical data of HOF for the single sixmembered C-N heterocycle
Compounds
(1) pyridine
(2d) pyradine
(2j) pyrimidine
(2l) pyridazine
(3) 1,3,5-triazine
(T) 1,2,4,5-tetrazine
B3LYP
Experimental
[16]
data
G3
144.77
196.23
196.65
278.24
225.52
464
143.0
209.0
190.3
285.3
229.0
489.7
6-311G(d,p)
6-311+G(2df,2p)
184.3
237.5
220.4
313.7
249.5
496.8
155.1
204.2
187.6
280.2
213.0
455.5
6-311+G(2df,2p)
(ZPE correction
145.8
196.2
179.5
272.2
206.1
450.1
Table 6. The molecular energy and HOF of all the compounds
T1-a
T1-b
T1-c
T2d
T2j-a
T2j-b
T2j-c
T2l-a
T2l-b
T3
TT
Energy
a.u.
-543.5430
-543.5466
-543.5458
-559.5758
-559.5787
-559.5864
-559.5823
-559.5454
-559.5492
-575.6211
-591.5778
fH(298K)
(KJ/mol)
639.1
631.1
632.5
692.4
683.5
667.8
676.4
769.9
761.7
712.9
960.1
TP1-1
TP1-2
TP1-3
TP2-1
TP2-2
TP2-3
TP2-4
TI2-1
TI2-2
TI2-3
TI2-4
Energy
a.u.
-505.4232
-505.4371
-505.4336
-521.4500
-521.4595
-521.4647
-521.4644
-521.4700
-521.4799
-521.4787
-521.4842
fH(298K)
(KJ/mol)
606.8
571.2
580.2
677.1
653.2
641.2
641.4
625.6
599.8
602.4
588.9
T123-1
T123-2
T123-3
T134-1
T134-2
T41-1
T41-2
T42-1
T42-2
Energy
a.u.
-537.4709
-537.4870
-537.4840
-537.4925
-537.5013
-553.4995
-553.5110
-553.4993
-553.5115
fH(298K)
(KJ/mol)
759.8
720.1
727.7
705.9
682.5
825.0
796.0
827.2
797.2
The results (Table 5) show that the different type of bond has different contributions to
HOF, the sequence is NN bonds>CN bonds>CC bonds. For example, the tetrazine (T) has
two NN bonds, the pyridazine(2l) has a NN bond, and the triazine(3) has no NN bond, so
HOF of tetrazine> HOF of pyridazine> HOF of triazine. And the triazine has three CN
bonds, the pyriazine(2d) or the pyrimidine(2j) has two CN bonds and the pyridine(1) has
only one CN bond, so HOF of triazine> HOF of pyriazine or pyrimidine> HOF of pyridine.
When C-N heterocycle as the substituent (see Table 6), the system follow the same rule. For
example, the tetrazole has the most number of NN bonds, so HOF of these isomers are the
most one among all the molecules. In addition, HOF of the isomer is related to the place of
the substituents connected with the tetrazine. When two rings are connected with the CN
10
365
Czech Republic
2006
[Content]
bond, its HOF is more than that of those molecules which two rings are connected with the
CC bond.
Fig 6.
The relationship between the HOF and the number of N atoms
We take out the HOF of TP1-1, TP2-1, T123-1 and T42-1 (because their HOF are the
maximum in each isomer) and analyze the relationship between the HOF and the number of
N atom in these molecules, the result sees the Figure 8. This curve has a good relativity. The
Figure 6 shows that the HOF would boost up linearly along with the increase on the number
of nitrogen atom when five- membered heterocycle connected the tetrazine ring with CN
bond.
4.
CONCLUSIONS
In this paper, after a hydrogen atom of the tetrazine is substituted by the C-N
heterocycle, we obtain 31 tetrazines C-N heterocyclic derivatives and their optimized
geometries, electronic structures, conjugations, energies, and heats of formation have been
investigated by the DFT methods. The results show:(1) When the nitrogen heterocycle
replaces a hydrogen atom on tetrazine, NN bonds on tetrazine rings are shorter and CN
bonds which are close to the substituent are longer, the other bonds change little. (2) In these
derivatives, the conjugations of the tetrazine rings are not distroyed. The part system forms a
delocalized occupied orbitals composed by 2PZ orbitals of all atoms, which enhances the
stability of the molecule. (3) The C-N heterocycles as the substituent have a strong
absorbing-electron characteristic, and the five- memebered ring 42 (2H-tetrazole) among
them has the strongest absorbing-electron characteristic. (4) There are a good linear
relationship between the energies, HOFs and the number of N atoms, the total molecule
energy decrease about 16a.u when increasing a nitrogen atom. The HOFs would boost up
linearly along with the increase on the number of nitrogen atom when five-membered
heterocycle connected the tetrazine ring with CN bond. Our study shows most of the species
have high HOFs and relative stabilities. The HOFs of these compounds are between
571.2KJ/mol and 827.2KJ/mol, so they may be potential candidates for energetic materials.
366
11
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
12
A.K. Sikder, Nirmala Sikder. A review of advanced high performance, insensitive and
thermally stable energetic materials emerging for military and space applications. Journal of
Hazardous Materials A112 (2004) 115
H. Feuer and A. T. Nielsen, Nitro Compounds; VCH, NewYork, Weinheim, Cambridge, 1990
A. T. Nielsen, Nitrocarbons; VCH, New York, Weinheim, Cambridge, 1995
J. Khler and R. Meyer, Explosivstoffe, 7. Aufl., Wiley-VCH, Weinheim, 1991
M.-X. Zhang, P. E. Eaton, and R. Gilardi, Hepta- and Octanitrocubanes, Angew. Chem. Int. Ed.
2000, 39, 401
Anton Hammerl, Thomas M. Klaptke, and Peter Schwerdtfeger.Azolylpentazoles as HighEnergy Materials: A Computational Study. Chem. Eur. J. 2003, 9, 5511-5519
Anton Hammerl, Thomas M. Klaptke, Heinrich Nth, et al. Synthesis, Structure, Molecular
orbital and Valence Bond Calculations for Tetrazole Azide, CHN7.Propellants, Explosives,
Pyrotechnics 28 (2003), No.4, 156-173
Parr, R. G.; Yang, W. Density-functional Theory of Atoms and Molecules; Oxford University
Press: New York, 1989.
Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B. 1988, 37, 785.
Carpenter, J. E.; Weinhold, F. J. Mol. Struct. (THEOCHEM) 1988, 169, 41.
Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899.
Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102, 7211.
Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735.
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,M. A.; Cheeseman, J. R.;
Zakrzewski, V. G.; Montgomery, J. A., Jr.;Stratmann, R. E.; Burant, J. C.; Dapprich, S.;
Millam, J. M.; Daniels, A.D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.;
Cossi,M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi,I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; AlLaham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. m.W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.;
Replogle, E. S.; Pople, J. A. Gaussian 98, Revision A.11;Gaussian, Inc.: Pittsburgh, PA, 2001.
Lias S.G, Libman J.F, Levine R.D. J.Phys.Chem.Ref.Data.1984,13,695
367
Czech Republic
2006
[Content]
SELF_AGGREGATION OF THERMAL DECOMPOSITION PRODUCTS

A.Ksiczak, T. Ksiczak, T. Zielenkiewicz
Departament of Chemistry, Warsaw University of Technology, Noakowskiego 3, 00-664 Warsaw,
Poland
Abstract:
Thermal decomposition during the induction period of six dinitrotoluene isomers was examined. To
define the production of contaminations, the liquid-solid phase equilibrium was used (criometry
method). The mean association (aggregation) number of decomposition products was appointed in the
melting temperature of each substance. The presence of macromolecule products of decomposition was
detected.
Keywords: dinitotoluene, thermal decomposition, mean association number.
1. Introduction
The exothermic decomposition of unstable substances can be a reason of the thermal
explosion during different technological operations. The generated heat rate determines
conditions of safe processes. This can be predicted on the bases of reliable kinetic parameters
of the thermal decomposition process. The thermal explosion occurs when the conversion
degree of unstable substance is small. The kinetic parameters for induction period of the
thermal decomposition process are needed for safety estimation. Our investigations were
concentrated on elaborating the methods, which allow analyzing decomposition process for
small conversion degrees of the thermal decomposition reaction. These methods were
developed using substances containing nitro groups like N,2,4,6-tetranitro-N-methylaniline
(TET) and 2,4,6-trinitrotoluene (TNT)
[1-6]
. The criometry method was used to study the
amount of thermal decomposition products. There was also used the method which exploit the
heat effect of partial (initial) decomposition process. It was noticed that the mean value of
decomposition rate during following measurements performed on the same sample in the same
nonisothermal conditions changes significantly in the oscillatory way. These changes were
ascribed to the concentration changes of volatile decomposition products possessing catalytic
368
Czech Republic
2006
[Content]
properties. The changes of volatile products concentration are connected with the release
processes from liquid reaction phase [2,3] .
The first stage of aromatic nitrocompounds decomposition depends on substituents
position. The nitro-group without any substituent in ortho position can separate from the
benzene ring in homolytic way and to form nitrite-group
[7]
. Two nitro groups in the ortho
position can react with each other forming benzofurazane and benzofuroxane
group in the ortho position to the methyl-group forms anthranil
Rogers
[12]
and Dacons
[13]
[10,11]
[8,9]
. The nitro-
have identified thermal decomposition products of TNT at
temperatures 300 and 200 C respectively. Rogers have isolated a number of substances, which
were oxidation reaction products or another intramolecular reaction. Dacons have identified 25
substances dissolvable in benzene. One of this was a product of intermolecular reaction, which
was made of two benzene rings bonded by nitrogen-atoms. Additionally Dacons received
insoluble in benzene so called explosive cake, which were observed earlier by Adams [14]. It
is probably a mixture of macromolecule substances. From the presented results follows that
thermal decomposition process is very complicated. In the case of multisubstituted substances
it comes a question about a role of each unstable substituent in the thermal decomposition
process. This information allows to predict the thermal stability of multisubstituted substances
in the thermal explosion aspect (safety).
For better knowledge of the substituents position influence on the thermal stability it is
necessary to compare the thermal decomposition process of substances with a similar structure
homological series. One of these series is dinitrotoluene isomers. We decided to examine
these substances using criometry technique applied earlier to TNT and TET analysis.
For better knowledge of the substituents position influence on the thermal stability it is
necessary to compare thermal decomposition parameters of substances with a similar structure
homological series. One of these series is dinitrotoluene isomers. In the literature there are
not systematic studies for these substances. The onset temperatures values of thermal
decomposition for three isomers: 2,4-dinitrotoluene (2,4-DNT), 2,6-dinitrotoluene (2,6-DNT)
and 3,4-dinitrotoluene (3,4-DNT) have been determined by Duh
DNT and 3,4-DNT) were examined by Ando
[16]
[15]
. Only two isomers (2,4-
The aim of this work is to study the thermal decomposition during the induction period and
to define an influence of nitro-groups position in relation to methyl-group on the thermal
369
Czech Republic
2006
[Content]
decomposition of dinitrotoluene isomers.
2. Experimental
Dinitrotoluene isomers were synthesized in the Warsaw Military University of
Technology. Melting temperatures (Tm), melting enthalpies (Hm) and purity (P) determined by
criometry method from DSC measurements were collected in the table 1. Literature values of
melting temperature and enthalpy were placed in the table 1 for comparison.
Table 1. Melting temperatures (Tm), melting enthalpies (Hm) and purity (P) of
dinitrotoluenes.
Isomer
2,3-DNT
2,4-DNT
2,5-DNT
2,6-DNT
3,4-DNT
3,5-DNT
Tm/K
332.7
343.7
324.5
338.3
332.4
366.0
Hm/(kJ/mol)
16.13
21.49
15.67
18.02
18.52
19.79
Tm.(lit)/K
329.7
[17]
340.7
[17]
323.6
[18]
327.4
[17]
329.4
[17]
366.1
[17]
Hm(lit)/(kJ/mol)
P/(%mol)
[17]
99.9
[17]
99.2
17.5
22.2
99.4
[17]
99.9
[17]
99.8
23.8
18.8
99.9
The purity of dinitrotoluenes was investigated by gas chromatography and for all isomers it
was about P=99.9% mass.
Measurements were made on a heat-flux DSC 605 UNIPAN calorimeter. The calibration
was performed with gallium, indium, cadmium, tin, naphthalene and benzoic acid. The metals
used had purities greater than 99.999%, and the organic substances had purities greater than
99.95%. The measurements were made using hermetic aluminium pans with heating rate 2
K/min. The samples were sealed under a reduced pressure of about 1.3 kPa. Measurements
were started from the temperature below melting process to the chosen temperature below the
exothermic decomposition process (there was no visible exothermic effect). Consecutive
measurements were made on the same sample.
370
Czech Republic
2006
[Content]
3. Results and Discussion
3.1. Application of the liquid-solid phase equilibrium to the estimation of the

production of the thermal decomposition products
The conversion of investigated substance increases during the thermal decomposition
process. This conversion can be observed by changes of the melting enthalpy. It was assumed
that the decompositn products do not form a solid solution and they are totally soluble in liquid
phase. The final conversion degree after each DSC measurement (f), can be calculate
according to the following equation:
f =
H m H n
H n
(1)
where: Hm is the melting enthalpy before a thermal decomposition process determined from
the first DSC run, Hn is the melting enthalpy in the consecutive measurement performed on
the same sample.
Criometry method can be used to predict the real mol fraction ( x ) of impurities according
with following equation
R(Tm0 ) 2 1
x
Tm = Tm0
H m F
(2)
where F is the melted fraction of a sample, Tm0 is the melting temperature of ideally pure
substance, Tm is the melting temperature of the sample, Hm is the melting enthalpy. The purity
of a sample is equal to
P = 1 x 100%
Fig. 1 shows DSC curves of the 2,3-DNT melting process in five consecutive runs made on
the same sample. Measurements were leaded to chosen temperature Tf. The increase of the
decomposition products ( x ) follows during the measurement. This increase ( x ) is equal to the
371
Czech Republic
2006
[Content]
DSC Signal
310
330 T/K
350
Fig. 1. DSC curves of the melting process for 5 consecutive runs performed on the same
sample of 2,3-DNT.
difference between the mole fraction of impurities of the sample taken to the investigation and
the mole fraction after the thermal decomposition during DSC measurements. This increase is
equal to real mole fraction of thermal decomposition products connected with the thermal
decomposition of the investigated sample during DSC measurements
x = x i
i
where xi is the mole fraction of decomposition product i. Assuming that during

decomposition from one molecule of investigated substance is formed one molecule of the
thermal decomposition product in the liquid phase and other volatile decomposition products
are in the gas phase, the value of real mole fraction of contaminations ( x ) should be equal to
the conversion degree f.
If values of f predicted from melting enthalpy changes and values of x calculated from
criometry analysis are not equal, then it means that from one molecule of analyzed substance
372
Czech Republic
2006
[Content]
appears different than 1 number of contaminations molecules. If the value of ( x ) parameter is

lower than f, then it means that molecules of decomposition products in the liquid phase are
connecting to each other to form associates (aggregation).
The conversion degree is equal to nominal mole fraction of thermal products.
f = x = ixi
(3)
where xi is the mole fraction of decomposition product with i number of mers in the
molecules. The size of these associates can be defined by so-called mean association number .
This parameter can be calculated using the equation [19] :
x = f/( - f+ f)
(4)
3.2. Results
In the table 2 are placed the melting enthalpy changes (DHm), the real mole fraction of
thermal products created during thermal decomposition ( x ) and mean association number of
decomposition products . In these DSC measurements the final temperature (Tf) was lower
than the onset temperature of the exothermic process.
Table 2. The final temperature (Tf), the real mole fraction x , the melting enthalpy changes
(DH) and the mean association number () of decomposition products of dinitrotoluenes.
isomer
2,3-DNT
2,4-DNT
2,5-DNT
3,4-DNT
3,5-DNT
2,6-DNT
Number
of run
1
1
2
1
1
2
3
4
1
1
2
3
Tf/K
x*100
547,7
545,8
545,8
561,7
532,8
532,8
532,8
532,8
591,0
562,3
562,3
562,3
2,38
1,38
1,45
1,82
1,07
1,63
1,44
1,01
1,55
0,86
1,78
1,46
DH/
(kJ/mol)
1,978
0,760
0,590
2,619
0,919
0,530
0,960
0,370
3,119
0,720
0,196
1,260
5,73
2,62
3,35
10,83
4,83
3,87
7,56
9,83
4,80
2,19
7,90
5,73
373
Czech Republic
2006
[Content]
Values of the mean association number indicate a presence of decomposition products,

which arise from a few molecules of DNT. These values change during a measurement series in
oscillatory way they rise and decrease in consecutive runs. This result shows that creating of
macromolecule substances doesnt depend on the substituents position in each isomer. The
sample after measurement was analyzed on the MALDI TOF apparatus what proved a presence
of substances with mass many times bigger than DNT molecule. The fig. 2 shows an exemplary
diagram of the mass spectrum of a 2,5DNT sample. This result indicates that in the sample of
conditioned 2,5DNT exist many different thermal products with a molecule mass even above
1000 Da. All of DNT isomers give similar MALDI TOF mass spectrum. These results
independently prove that products of DNT thermal decomposition have an ability to connect to
Peak intensity
each other.
500
1000
1500
m/z
Fig. 2. Diagram of the MALDI TOF mass spectrum of a 2,5-DNT sample.
The structure of thermal decomposition products depends strongly on the position of nitro-
374
Czech Republic
2006
[Content]
group in relation to methyl-group. Our results indicate that some thermal products have an
ability to form associates (aggregation).
5. Conclusions
Application of the liquid - solid phase equilibrium let us study a decomposition below the
onset temperature of the exothermic decomposition. Values of mean association number which
were obtained in the melting temperature of dinitrotoluene indicates that thermal
decomposition products create molecules with mass many times bigger than DNT molecule.
These values change in oscillatory way during consecutive measurements. The presence of
macromolecules was verified by MALDI TOF analysis.
Literature
[1] A. Ksiczak, T. Ksiczak. Thermochim. Acta, 275 (1996) 27.
[2] A. Ksiczak, T. Ksiczak, T. Zielenkiewicz. J. Thermal Anal., 77(2004) 233-242.
[3] A. Ksiczak, T. Ksiczak. J. Thermal Anal., 60(2000) 25-33.
[4] A. Ksiczak, T. Ksiczak. Thermochim. Acta, 268(1995) 95-104.
[5] A. Ksiczak, T. Ksiczak. J. Thermal Anal.., 41 (1994) 1153-1163.
[6] A. Ksiczak, T. Ksiczak. Thermochim. Acta, 284 (1996) 299-308
[7] Yu. Ya. Maksimov, S. B. Sorochkin, S. V. Titov, Tr. Mosk. Khim. Technol. Inst. im. D. I.
Mendeleeva 1980, 112, 26
[8] D. E. Bliss, S. L. Christian, W. S. J. Wilson, Energ. Mater. 1991, 9, 319
[9] E. K. Fields, S.Meyerson, J. Am. Chem. Soc. 1967, 89, 3224
[10] Tsang, W., Robaugh, D., Mallard, W.G.; J. Phys. Chem. 1986, 90, 5968
[11] He, Y.Z., Cui, J.P., Mallard, W.G., Tsang, W. J. Am. Chem. Soc. 1988, 110, 3754
[12] R.N. Rogers. Anal. Chem., 39 (1967) 730.
[13] J.C. Dacons, H.G. Adolph, M.J. Kamlet. J. Phys. Chem., 74 (1970) 3035.
[14] G.K. Adams, P.R. Rowland, L.A. Wiseman. Ministry of Supply Report A. C. 3982, Great
Britain, 1943.
[15] Y. Duh, C. Lee, C. Hsu, D. Hwang, C. Kao. J. Hazard. Mat., 53 (1997) 183
[16] T. Ando, Y. Fujimoto, S. Morisaki. J. Hazard. Mat., 28(3) (1991)251
[17] H.J. Borchardt, F. Daniels. J. Am. Chem. Soc., 79 (1957) 41.
[18] J.W. Beckmann, J.S. Wilkes, R.R. McGuire. Thermochim. Acta, 19 (1977) 111.
[19] J. Prigogine, R. Defay. Chemical Thermodynamics, London, 1954.
375
Only in Proceedings
Czech Republic
2006
[Content]
STUDY OF THE ROUTES FOR PRODUCING

NITROGUANIDINE(NQ) IN AQUEOUS NITRIC ACID SOLUTIONS.
DEVELOPMENT OF THE EFFICIENT METHOD FOR THE
SYNTHESIS OF NQ
A.A.Astratev, D.V.Dashko, A.I.Stepanov
Special Design and Construction Bureau SDCB Technolog

of the Saint-Petersburg State Institute of Technology (technical university)
190013, Russia, Saint Petersburg, Moskovskiy pr.26
Abstract:
The main routes of transformation of
guanidine and dicyandiamide in aqueous
solutions of nitric acid are considered. The nitration of guanidine is shown to be
catalyzed by lower nitrogen oxides. On the basis of studies of the quantitative
regularities of transformation of guanylnitrourea, guanylisocyanate and nitroguanidine,
the probable mechanism of reactions is suggested and the main factors determining the
reactivity of the substrates are addressed.
Keywords: Synthesis, nitration of guanidine and dicyandiamide, nitroguanidine.
1.
INTRODUCTION.
The structural feature of nitroguanidine derivatives associated with the planar structure
of the nitroguanyl moiety stabilized by intramolecular hydrogen bonds and a considerable
delocalization of -electron density, attracts for a long time the attention of researchers to
such compounds as potential objects for the development of insensitive high explosives
(IHE). Nitroguanidine (NQ) itself may be considered as a nitroimine analogue of a
promising IHE, DADNE (FOX-7) [1,2] . In general, NQ can be viewed as one of the amidine
derivatives, incorporated into the structure of a series of such IHE as NTO, DNGU,
DADNE.
NH N
O
N
H
NTO
NO2
H2N
NH2
H2N
NO2
NO2
NO2
DADNE
NH2
N
Nitroguanidine (NQ)
Alexandr Astratev, Dmitry Dashko, Andrei Stepanov
NO2
N
H
N
O
O
N
H
N
NO2
DNGU
377
Czech Republic
2006
[Content]
First synthesized by Jousselin in 1877, NQ for over a century remains an important

ingredient of a variety of energetic composite materials of various application such as rocket
propellants, high explosive formulations, gun powders etc. based on RDX and HMX and
lends significantly to decreasing their sensitivity to external stimuli [3, 4]. It is used also as
ingredient of gas-generating and fire extinguishing formulations [5-7].
During the past decade NQ aroused keen interest as a synthon in the preparation of other
promising N-nitroamines. A successful synthesis of dinitroguanidine [8] was performed on
the basis of NQ, its condensation with hexamine and formaldehyde afforded a series of
heterocyclic compounds such as 6-nitrimino-1,3,5-trinitro-1,3,5-triazacyclohexane, 1,5di(nitroguanidino)-2,4-di(nitroimino)-1,3,5- triazacyclohexane. [9].
An exceptional feature of NQ is associated with the fact that among the variety of IHE it
has readily available and cheap raw materials, i.e. urea, ammonium nitrate or dicyandiamide
(DCDA).
Introduction of the nitro group into the guanidine molecule is usually carried out by the
action of mixed acid or else by the dehydration of guanidine nitrate with sulfuric acid [10-14].
Still the existing methods suffer a relatively low efficiency and environmental problems. The
insufficient extent of working up the synthetic aspects of NQ preparation doesnt allow one
to make the ultimate conclusion on the most promising route for its synthesis. In this
connection it seems actual to study the behavior of the starting materials and products in
nitric acid solutions for a sensible choice of optimum conditions for NQ synthesis.
2.
RESULTS AND DISCUSSION.
The study of kinetics of guanidine nitration in aqueous nitric acid has shown that NQ
does not essentially undergo decomposition under such conditions (Fig.1).
, min
Fig 1.
A kinetic curve for NQ accumulation during the treatment of guanidine

nitrate with aqueous nitric acid, 25, =265 nm.
HNO3, weight% : 1-96.8% 2-91.1% 3-88.5%
The dependence of the effective rate constants for the nitration of guanidine nitrate on
the nitric acid concentration is given in Fig. 2.
378
Study of the Routes for Producing Nitroguanidine(NQ) in Aqueous Nitric Acid . . .
Czech Republic
2006
[Content]
HNO3, wt.%
Fig 2.
The dependence of the effective rate constants for the nitration of

guanidine nitrate on the nitric acid concentration at: 1-50 , 2 - 25 .
The regression line in this case (Fig. 2) follows the equation:

50
log Kapp =(0.1170.003) [HNO3] wt.% (13.40.2) (1)
25
log Kapp =(0.1370.004) [HNO3] wt.% (15.70.3) (2)
The slopes in eqs. 1 and 2 are close to those observed earlier for the kinetics of nitration
of carbamates and benzene sulfonamides [19], which implies the absence in the course of
nitration of protolytic equilibria and the resulting variations in concentrations of the active
forms of the substrate. Taking into account the high basicity of guanidine (pKBH+=13.6) [15],
we can suppose that during the nitration in aqueous nitric acid, similar to the nitration in
mixed acid [16], the active form of the substrate is its conjugated acid. This conclusion meets
the Schofields criterion for the threshold rate of a bimolecular reaction [17]. Hence, the
guanidine nitration process may be presented as follows:
NH2
H+
NH2
+
H2N C
HN
NH2
NH2
HNO3
NH
H +2
+N C
NH2
O2N
NH2
-2 H+
N
O2N
NH2
Scheme 1. Transformations of guanidine in solutions of nitric acid.

During the nitration of guanidine nitrate in concnd. nitric acid alongside with
nitroguanidine, judging by UV spectral data (max303nm) for the diluted reaction mass, the
formation of dinitroguanidine in 20-25% yield is also observed. To avoid the introduction of
the second nitro group, it was found necessary to lower the nitric acid concentration to 90%.
Such a dilution, however, makes it necessary to raise the reaction temperature in order to
intensify the nitration process; otherwise the reaction time is considerably lengthened. At the
same time the elevation of temperature over 60-70 is undesirable because of the partial
decomposition of the formed NQ which becomes appreciable at 70 in 88.5% nitric acid
(Fig. 3).
379
Czech Republic
2006
[Content]
, min
Fig 3.
The kinetic curve for accumulation and decomposition of nitroguanidine in

the course of treatment of guanidine nitrate with 88.5% nitric acid at 70,
max=265 nm.
The theoretical studies show that the nitrating activity function is rather ambiguous in
nature and depends both on the atom undergoing the attack by an electophile and the
structure of the substrate. This ambiguity is caused by the participation in the nitration
reaction, alongside with the nitronium ion, of some other electrophilic agents. During
nitration most of the highly basic amines undergo protonation which hampers the subsequent
electrophilic attack of the conjugated acids due to large differences in the energies of the
lowest vacant molecular orbital of the acceptor (NO2+) and the highest occupied molecular
orbital of the donor (guanidinium cation). Assuming that the solvation energies of the
nitrating and nitrosating agents are close to each other, it seems plausible that the energy
level of the lowest vacant molecular orbital of the nitrosating agents is considerably
lower[18], which make them sufficiently strong electrophiles. The significance of the
electrostatic contribution in this case is not so marked because of the presence of a partial
positive charge on the reaction center (the nitrogen atom). Besides, the nitration is known to
be sensitive primarily to the charge control whereas the nitrosation is the subject of the
orbital control. Therefore more soft electrophiles should be used to effect nitrosation (the
combination of real charges plays a negligible role; the primary positions determining the
stability of the intermediate occupy the frontier orbital energies, their difference in the
present case being considerably lower compared to that of the nitrating agents). It follows
that the catalytic action of nitrogen oxides in the nitration reaction will develop more
distinctly in the series of amines which undergo nitration in the protonated form.
Consequently we can anticipate the catalytic activity of nitrogen oxides for guanidine
nitration, which, as it was mentioned above, in acidic media exists exclusively in the
protonated form. Indeed, the introduction of nitrogen oxides accelerated the rate of nitration
of guanidine nitrate many times (Fig. 4).
380
Czech Republic
2006
[Content]
log Kapp * 10-4 s-1
* 102 mol/l. NaNO2

Fig 4.
The effect of NaNO2 on the rate of nitration of guanidine nitrate in 84.6%

nitric acid at 25oC.
It follows from Fig. 4 that the introduction of 1.5weight % of sodium nitrite is

equivalent to a 50 elevation in temperature. There is no point in using nitrogen oxides in
combination with the elevated nitration temperature because the catalytic effect drops and
the NQ stability lowers under these conditions (Fig. 5). The mechanism of NQ formation in
the presence of lower nitrogen oxides may be presented as shown in Scheme 2.
, min
Fig 5.
The kinetic curves for accumulation and decomposition of nitroguanidine

under treatment of guanidine nitrate with 88.5% nitric acid at 50
(=265 nm). 1- in the absence of NaNO2 ; 2- in the presence of 1 weight
% NaNO2.
381
NH2
Czech Republic
2006
[Content]
NH2
kn
H2N C
O2N
NH2
NH2
NQ
kn1
N2O4
NH2
ON
N
NH2
NsQ
W=kn[C+(NH2)3]+ kn1[NsQ]
Scheme 2. The reaction of catalytic nitration of guanidine .
It is necessary for the catalytic effect to manifest itself that both the rate of guanidine
nitrosation to form the N-nitroso derivative and the rate of its subsequent transformation be
considerably higher compared to that for the direct guanidine nitration.
The enhanced activity of N-nitrosoguanidine in comparison with the starting amine in
the acidic nitration (kn1>kn) implies that the reactive form of the substrate is the covalent Nnitrosoguanidine (NsQ) molecule. It follows from the above that the introduction of the
nitroso group contributes to the decrease in its basicity to a greater extent than to the
lowering of its reactivity in the electrophilic substitution. Such behavior agrees well with the
ambident nature of the nitroso group, its negative inductive effect being compensated for by
its positive mesomeric effect. Hence the nucleophilic nature of N-nitrosoguanidine is
displayed more distinctly than its basicity, which allows one to assume the presence of
sufficiently soft acceptors such as HNO3 molecule, as nitrating agents.
The treatment of a preliminarily prepared N-nitrosoguanidine with nitric acid also
invariably resulted in the formation of NQ in a quantitative yield. The kinetics of Nnitrosoguanidine nitration turned out to be identical to that of guanidine nitration in the
presence of an equivalent amount of nitrogen oxides. Hence we may conclude that the rate of
denitrosation of the N-nitroso compound is rather high because the concentration of NsQ is
linearly related to that of the nitrosating agents (Fig.4).
Addition of sodium nitrite to the nitric acid solution during the nitrating process causes
the immediate increase in the formation of NQ (Fig.6). Hence we can conclude that the
nitrosation stage is a quickly established equilibrium, which is shifted to the side of the
guanidinium cation.
382
Czech Republic
2006
[Content]
, min.
Fig 6.
The kinetic curve for the NQ accumulation under treatment of guanidine

nitrate with 80.6% nitric acid at 25oC.
Generally, two possible routes for NsQ nitrolysis may be imagined (Scheme 3).
NH2
N2O4
-H+
NH2
ON
H2N C
N C
H
NH2
NH2
NH2
ON
NH2
ON
NH2
N
N NO2
H
HNO3
HNO3
-NO
NH2
-NO
N
O2N
Scheme 3.
NH2
ON
NH2
+
N C
NH2
O2N
The possible routes for NsQ nitrolysis.
By analogy with the guanidine nitration, the most probable appears the routeA because
N-nitro-N-nitrosoguanidine is an unstable intermediate in acidic media and undergoes a
prompt denitrosation to produce the final product. The route B (nitrolysis), however, can
not be completely excluded because the attack of the nitrating agent on the nitrogen atom of
the nitrosamine group effectively facilitates the heterolysis of N-NO bond.
In accordance with the conclusions drawn as a result of the kinetic study regarding a
significant acceleration of the reaction of NQ formation under the action of nitrogen oxides,
a preparative study of the nitration of guanidine nitrate with concentrated nitric acid with
added N2O4 has been carried out.
7
383
Czech Republic
2006
[Content]
Table 1. Results of the preparative nitration of guanidine nitrate in mixtures of

HNO3 and N2O4.
N2O4
amount,
(weight parts)
Weight parts Temperature

of HNO3 per 1 ,
part
of
guanidine
0.15
0.12
1.7
0.1
0.1
1.5
1.2
Maximum yield
of NQ, %
Time
( ),
necessary
to
attain
the
maximum
yield, hours.
45
90
1.5
30
96
45
85
60
81
45
85
20
96
20
93
It follows from the data listed in Table 1 that nitration at temperatures over 20-300C in the
presence of N2O4 is inefficient which is apparently caused by the low concentration of
nitrogen oxides in the liquid phase at elevated temperatures due to their high volatility.
Reducing further the excess of nitric acid over the substrate concentration deteriorates the
results of nitration because of a noticeable decrease in the solubility of nitrogen oxides in the
acid phase.
Hence the suggested procedure, as distinct from the existing methods, is featured by high
efficiency and an essential absence of wastes and by-products.
As an alternative method for the preparation of NQ, an approach based on dicyandiamide
(DCDA) as starting material should be considered. This multistep procedure involves the
following stages:
-
Hydrolysis nitration of DCDA to produce guanylurea;
Acid-catalyzed decomposition of guanylurea to guanidine;
Nitration of guanidine to nitroguanidine.
Reasoning from the general theoretic conceptions and taking into account a strong electronaccepting character of the cyano group comparable to that of the nitro group, it is believed
that DCDA, similar to NQ which is not subjected to the double nitration in aqueous solutions
of nitric acid, would be inert in such media. It follows that the process of its nitration will
have a stepwise character involving on the first stage the DCDA hydrolysis to form
guanylurea and consequently its subsequent nitration. The overall rate of this process should
not exceed the rate of nitration of guanylurea itself. However, in the course of kinetic studies
of DCDA nitration it turned out that the combined process of hydrolysis-nitration took place
much more readily compared to the direct nitration of guanylurea, its product of hydrolysis.
So the apparent first-order rate constant for the formation of guanylurea from DCDA in 83%
nitric acid is 17 times as high as the corresponding constant for the production of NQ from
guanylurea. This discrepancy can be rationalized in terms of an assumption that the
384
Czech Republic
2006
[Content]
hydrolysis of the cyano group and nitration take place simultaneously, i.e. that in this case
either the nitric acid molecule or its protonated form play the role of the nitro group carrier.
At the same time the sole hydrolysis product is guanidine. So it may be concluded that in
this N-nitro compound the nitro group is attached to the urea fragment. Thus the combined
process of hydrolysis-nitration can be depicted by the following scheme involving the
transfer of the nitro group from guanidine to the urea moiety (Scheme 4).
NH2
-H+
NH2
NH2
H2N
C
NH
N
H
NH2
H2N
+ NH2
H
+
N C
NH2
N
N
HNO3
HN
O
N
H2N
O
H
H
+ N NO
2
C
NH
NO2
HN
OH
Scheme 4. The stage of hydrolysis-nitration of dicyandiamide to form guanylurea.

It is well known that cyanamide under the action of concnd. nitric acid transforms into
nitrocyanamide, no hydrolysis being observed. Such appreciable differences in the
susceptibility of the cyano group toward hydrolysis in strongly acidic media can be
rationalized in terms of the geometry of the corresponding molecule. As distinct from
DCDA, the formation of a six-membered cyclic transition state for cyanamide is difficult.
Hence the processes of nitration and hydrolysis take place successively and the isolation of
both nitrocyanamide and nitrourea is possible. An anomalously low (ca. 2 times) order of the
reaction in the nitric acid concentration for DCDA compared to that usually observed for
other amines may serve as an additional kinetic confirmation of a peculiar nature of the
nitrating agent in the considered case (Fig.7).
385
Czech Republic
2006
[Content]
HNO3, wt.%
.
Fig.7 Dependence of the apparent nitration rate constants on HNO3
concentration at 25 for: 1-urea; 2-guanylurea; 3-DCDA. The
corresponding linear correlation equations are as follows:
1. log Kapp =(0.190.01) [HNO3] wt.% (19.40.9)
2. log Kapp =(0.120.01) [HNO3] wt.% (13.51.2)
3. log Kapp =(0.0770.005) [HNO3] wt.% (8.80.5)
Formally the nature of the nitrating agent for the nitration of DCDA and other amides is
identical but there exist some basic differences: whereas in the process of amides nitration
the formation of cyclic 6-membered transition state takes place by means of an
intramolecular hydrogen bond, in the case of DCDA the main feature is the ambident nature
of the reagent. On the one hand, the nitrating agent acts as electrophile, but on the other it
plays the role of a nucleophilic partner and as such assists in the cyano group hydrolysis. It
follows from the above that the nucleophilic assistance presupposes the primary participation
of nitric acid as a potential carrier of the nitro group, an additional polarization of N-O bond
taking place under the action of the carbon atom of the cyano group.
It should be mentioned that as the nitric acid concentration drops under 70%, the yield
of guanylnitrourea from DCDA decreases whereas the formation of guanylurea prevails
because the corresponding rate of nitration is slower. With time the yield of guanylnitrourea
approaches the quantitative one, i.e. guanylurea is slowly nitrated to the final product
(Fig.8).
.
386
Czech Republic
2006
[Content]
, h.
Fig. 8 The kinetic curve for the accumulation of guanylnitrourea in the course of
DCDA interaction with 75% nitric acid at 25 .
The character of the process of an irreversible decomposition of N-nitrocarboxamides in

general is the same and consists in the heterolysis of -NHNO2 bond. (Scheme 5)
+
NH2NO2
H+
NHNO2
Nu
Nu
+
R
+ NH2NO2
Nu
R
O
Scheme 5. The stage of an irreversible acid-catalyzed decomposition of Nnitrocarboxamides.

However, in practice an apparently unexpected character of the dependence of log Kapp.
vs. [HNO3] wt.% for guanylnitrourea is observed (Fig 9).
11
387
Czech Republic
2006
[Content]
HNO3
wt.%.
Fig.9 Dependence of the logarithms of the pseudo first-order rate constants for the
decomposition of N-nitrocompounds on the nitric acid concentration at 25oC (1nitrourea, 2-guanylnitrourea). The correlation equations are as follows:
for nitrourea:
log Kapp = (0.0870.006) [HNO3] wt.% (11.30.6)
for guanylurea:
log Kapp = (0.210.02) [HNO3] wt.% (23.01.5)
Both processes are subject to acid catalysis which is evidenced by the increase in the
decomposition rate with the nitric acid concentration, though for nitrourea a more flattened
dependence of log Kapp vs. [HNO3] wt.% is observed.
A conclusion may be drawn from the above reasoning that different mechanisms for
the decomposition of these N-nitro compounds are realized (Scheme 6). The decomposition
of nitrourea is accomplished with the assistance of the water molecule at the rate-limiting
stage and, consequently, as the concentration of nitric acid increases, the content of the
reactive protonated form of the N-nitro compounds raises as well, with the simultaneous
decrease in the activity of available nucleophiles in the reaction system. It gives rise to the
fact that the dependence of the decomposition rate on the nitric acid concentration has a
lower slope compared to that for guanylurea where the nucleophile does not take part in the
rate-limiting stage whereas the substrate undergoes a monomolecular decomposition.
NO2 2H+
NH
H2N
NO2
NH2
+
H3N
O
Scheme 6 .
H2O
+
H3NCOOH
NH2NO2
H+
The acid-catalyzed decomposition of nitrourea.
Consequently, in the case of guanylnitrourea the acid-catalyzed decomposition results in

the formation of the intermediate guanylisocyanate which in sufficiently concentrated
solutions of nitric acid (over 95%) at 25oC remains a fairly stable intermediate (Scheme 7).
This conclusion is further supported by the identical values of slopes for the dependence of
the rates of decomposition of arylsulfonamides on sulfuric acid concentration, irrespective of
the site of heterolytic dissociation (either denitration or the formation of nitramide) [19].
388
Czech Republic
2006
[Content]
Under the conditions of kinetic experiments the formation of NQ is not observed even
over a continued period of time ( several days) whereas the dilution of the reaction mass with
water down to 85-86% of HNO3 results in the emergence of NQ (Scheme 8).
H2N
H2N
2H+
O2N
NH
N
O
O2N
H2N
H2N
NH
+
C N C
H O
H2N
Scheme 7.
+ NH2NO2
N
H2N
Formation of guanylisocyanate from guanylnitrourea .
H2N
H2O
H2N
H+
H2N
+
N
H2N
H2N
HNO3
H2N
H2N
C NH2
H2N
Scheme 8.
OH
-CO2
C NH2
H2N
NO2
N
H2N
The stage of nucleophilic decomposition of guanylisocyanate to

guanidine and its nitration to NQ.
The stability of guanylisocyanate formed during the nitration of DCDA, besides being a
function of concentration, is significantly affected by the molar excess of HNO3. When the
weight ratio HNO3 : DCDA is lowered under 15, the only isolable product is NQ. It follows
that not only water but also guanylnitrourea can act as a nucleophile in this reaction. This
conclusion is further supported by the results of preparative experiments which made it
possible, depending on the weight ratio HNO3 : DCDA at the constant nitric acid
concentration (taking into account its dilution during the decomposition of guanylnitrourea),
to record the preferable formation either of guanylisocyanate or NQ.
Thus, depending on the experimental conditions (concentrations, the weight ratio of
reactants, temperature), the rate-limiting stage of the reaction may be either the formation of
guanylisocyanate or its decomposition. In relatively diluted solutions of nitric acid (80-90%),
irrespective of the substrate concentration, its complete conversion takes place. The rate
13
389
Czech Republic
2006
[Content]
constants calculated by the consumption of guanylnitrourea and by the accumulation of NQ,

coincide.
Such behavior is explained by the enhanced activity of nucleophiles in the aqueous nitric
acid solutions, which determines the low stability of guanylisocyanate.
Based on the reported above quantitative regularities of the transformation of DCDA into
NQ, the main features of this process as compared to that of the catalytical nitration of
guanidine can be formulated.
In order to ensure the safe conditions of the process, the nitration should be carried out at
a relatively low weight ratio HNO3 : NQ and at elevated temperatures (40-60oC) which
allows to exclude the accumulation of intermediates in the process of DCDA nitration.
Besides, it seems expedient to provide for a constant removal of nitrogen oxides from the
reacting zone.
3. CONCLUSION.
The nitration of guanidine is shown to be catalyzed by lower nitrogen oxides. On the
basis of studies of quantitative regularities of transformation of guanylnitrourea,
guanylisocyanate and nitroguanidine, the probable mechanism of reactions is suggested and
the main factors determining the reactivity of the substrates are addressed. An effective and
environmentally friendly method for NQ preparation amenable to scale-up, has been worked
out.
Acknowledgments
The authors would like to thank the International Science and Technology Center for
giving financial support to this project.
4. EXPERIMENTAL.
The rates of the formation and decomposition of N-nitro compounds were monitored by
means of an aliquot method. A weighed sample of the starting amine was placed into a
temperature controlled cell containing 20-30 ml of aqueous nitric acid. In fixed time
intervals the aliquots of the reaction mass (0.5-0.8 ml) were quenched in 50-100 ml of water
or in a 1-2 molar solution of sodium acetate. The resulting solution was
spectrophotometrically analyzed for the concentration of N-nitro compound. The 265 nm
wavelength was used as an analytical wave for guanylnitrourea and nitroguanidine. Because
of a low spectral resolution the measuments were carried out at two different pH values,
namely 5-6 and 10-11. The molar extinction coefficient was as follows:
for GNU:
= 5-6
= 13500 l.mol-1cm-1
p= 10-11 = 6500 l.mol-1cm-1

for NQ
pH= 3-11
= 13500 l.mol-1cm-1
The percentage of these compounds in their mixtures was calculated according to

formula:
[NQ]=2D1-D2/ 13500 mole.l-1
[GNM]= D1-D2/ 6500 mol.l-1
390
Czech Republic
2006
[Content]
,where D1, D2 stand for optical densities at 265 nm at 5-6 and 10-11, respectively.
As a reference solution we used the solution of the quenched aliquot of initial nitric
acid. During the study of nitration in the presence of nitrogen oxides a portion of sodium
nitrite was introduced into the cell before adding the sample of the reagent.
Synthesis of the nitrate of guanylisocyanate.
Guanylnitrourea (1 g) was dissolved at 5-10oC in 25 ml of 100% nitric acid and left at
10 C for 3 hours. The nitric acid was removed in vacuo at the water bath temperature under
20oC. The residue was dried in a vacuum desiccator. The product was obtained in 95% yield
(0.95 g).
o
Found, % C 16.0; H 3.0; N 37.2; C3H4N4O4. Calculated, % C 16.2; H 2.9; N 37.8.

IR spectrum, , cm-1 1760(s) 1700(s) 1650(w) 1600(s) 1340(s) 1240(s) 11110(w) 1060(m)
890(m)
M.p.180 OC (decomp) ( an admixture of NQ less than 3%) .
REFERENCES.
1. Karlsson, S.; stmark, H; Eldsater, C.; Carlsson,T.; Bergman, H.; Wallin, S.; Pettersson, A.
Detonation and Properties of FOX-7 and formulations containing FOX-7.; San Diego, 2002, p
5.
2. Lochert, I. FOX-7- New Insensitive Explosive DSTO, 2001
3. Y. Tang : The effect of NG and TATB to the Insensitivity of HMX; 33 nd International Annual
Conference of ICT, Karsruhe, Germany,2002p.1-7.
4.
D. Powala.; A. Orzechowski .;A.Maranda; J. Nowaczewski. Spherical Nitroguanidine as
component of high explosives.; 7. International Seminar of Pardubice, Czech Republic, p.606612, 2004.
5. Pat. WO 200015584
6. Y. Yamato, A study of nitroguanidine based propellant. The Proceedings of 1997 Autumn
Symposium of Japan Explosives Society (1997), p.53.
7. Pat. EP 992473
8. A.A. Astratev, D.V. Dashko, L.L. Kuznetsov: Synthesis and some properties of 1,2Dinitroguanidine.: Russ. J.Gen.Chem.2003 v.39 4,p.537-548 (in Russian).
9. Yu Yongzhong.; Su Zhuang; Duan Baoru; Chen Fubo: Synthesis of Polynitrocompounds from
Nitroguanidine .; Propellants, Explosives, Pyrotechnics 14,p.150-152, 1989.
10. Davis T.L.: Nitroguanidine L.; J.Org.Syntheses.1927, 7,p.68-69.
11. Davis T.L. Elderfild R.C.: Alkyl-nitroguanidines. Rearrangement and Preparation by nitration;
J.Am. Chem.Soc.1933, 55, p. 731-740.
12. Pat. USA. 288290,1959
13. German Pat .1142860,1963
14. German Pat .2745774,1979
15. Angual S.J., Warbwiton: The basic strengths of methylated guanidines.; J.Chem.Soc.1951,
p.2492.
16. Williams.G., Simkins R.J.J.: The nitration of guanidine in sulfuric acid. II.. Kinetics of Nnitration of the guanidinium ion and denitration of nitroguanidine.; J.Chem.Soc.1953, p.13861392.
17. Moodie R.B., Schofield. K., Willinmson M.J.: Nitro compounds / Pergamon Press. Lon. 1964.
18. A.I.Titov, Nitration of aromatic and unsaturated compounds. Uspekhy Khim. (Advances of
Chemistry), 1958, v.27, p.845-890 (in Russian)
15
391
Czech Republic
2006
[Content]
19. O.A.Drozdova, A.A.Astratev, L.L.Kuznetsov, V.F.Selivanov, The effect of structure on the

kinetics and mechanism of acid-catalyzed decomposition of benzenesulfonitramides.
Russ.J.Gen.Chem., 1982, v.18, p.2335-2340 (in Russian).
20. A.A. Astratev, L.L. Kuznetsov: Formation and decomposition of nitrocyanamide in strong
mineral acids.: Russ. J.Org.Chem.2002 v.38 9,p.1252-1259. (in Russian).
392
Czech Republic
2006
[Content]
FUEL DISPERSION IN AIR BY DETONATION WAVE

BAZRAFSHAN Esmaeil, POURMORTAZAVI Seyed Mahdi
Faculty of Material and Manufacturing Technologies, Malek Ashtar University of Technology,
P. O. Box 16765-3454, Tehran, Iran
(bazrafshan1@yahoo.com)
Abstract:
The ratio of powdered substance to dispersal burster charge has considerable effects on
dispersion process in FAE type devices, it affects on expansion rate and concentration
of dust in dust-air cloud. Using central burster charge is a common method to disperse
powdered substances in air by detonation wave. Detonation wave, produced by the
burster charge, ruptures the thin wall of fuel container and accelerates the fuel particles
outward radially. When the particles of the powdered substance leave initial device,
they have maximum velocity or initial velocity u0 , then they are decelerated by air drag
force. Test samples were designed and assembled; three samples were the same except
the amount of burster charge. Selected fill-to-burster ratios were 100 to 1, 100 to 2, and
100 to 3. Fill-to-burster ratio was the same in two samples but their scales were
different. Aluminum powder (18 m) and presses TNT were used as fuel and burster
charge respectively. Expansion of powder clouds were photographed by high-speed
camera (1000-8000 Frames/sec) in each test. The shapes of clouds were pan-cake
shaped completely. Fill-to-burster ratios were investigated and optimum ratio was
obtained. Fill-to-burster ratio 100 to 1 is close to optimum ratio according to the test
results. Dust-air cloud is initiated by second charge. Delay time between detonation of
burster charge and second charge should be determined. If the amount of burster
charge increases, then delay time between two charges decreases. Delay time in actual
devices is 20-50 ms; delay times in test samples with fill-to-burster ratio 100 to 1 were
into the range. Initial velocity u0 of radial expansion were 110, 180, and 220 m/s for test
samples with fill-to-burster ratis 100 to 1, 100 to 2, and 100 to 3 respectively. Optimum
fill-to-burster ratio in a large scale device is a little less than it in a small scale device,
because average traveled distance by particles of powdered substance is longer than it
in small scale.
Keyword: Powdered substances, dispersion, powder cloud, burster charge, fill-toburster ratio
1.
INTRODUCTION
The FAE type munitions, contrary to the conventional munitions, have only fuel as
combustible, and the required oxidizer comes from air by dispersing the fuel into the
surrounding atmosphere. So, dissemination of fuel in air, for all kinds of the FAE type
devices, is necessary, because this process causes to form fuelair mixture. Concentration of
fuel in the fuel-air mixture should be between the lower and the upper explosion limits,
because only in this case, the fuel-air mixture is detonable. The following factors lead to
increase the destructive effects of the detonation of the fuel-air mixtures.
Homogeneity in Fuelair mixture.
Being the concentration of fuel close to its stoichimetric concentration.
Pan-cake shaped of the cloud
1
Esmaeil Bazrafshan, Seyed Pourmortazavi
393
Czech Republic
2006
[Content]
The above objects indicate the importance of the dispersion process in all kinds of the
FAE devices. A typical FAE device consists of a container filled of liquid or finely
powdered mixture and two separate explosive charges. The first charge bursts open the
container and scatters the contents forming an aerosol cloud and then the second charge
detonates the cloud. Dispersion of powdered fuel in air plays a key role on reliability of the
operation of the FAE device. Without designing a suitable system for dissemination of
powdered mixture into the air; achievement on a detonable dust-air mixture is impossible.
Ratio between amounts of fuel and burster charge is called fill-to-burster ratio. It is an
important factor for dispersion process in FAE type devices. Detonable fuel-air cloud is
produced by proper amount of burster charge respect to the amount of fuel which should be
dispersed. Remaly and Ragus researched about fill-to-burster ratio in 1983[1], but they didn't
published their experiments.
2.
EXPANSION OF POWDERED SUBSTANCES IN AIR BY

DETONATION WAVE
The process of powdered substances expansion in air by detonation wave can be divided
into three stages:
1-Particles acceleration
2-Particles deceleration
3-Particles random dispersion
Detonation wave and gases pressure produced by detonation of burster charge,
accelerate fuel particles radially and rupture the wall of cylindrical fuel container .Duration
of this stage is very short about 2 ms. The fuel particles reach their own maximum velocity
u0 at the end of first stage.
When the fuel particles reach maximum velocity u0 and the wall of the fuel container
ruptures, the second stage of fuel dispersion process begins. In this stage, the fuel particles
move radially, but their velocities decrease by air drag force, so that particles radial
movement comes to an end. Therefore, the radius of the dust-air cloud grows in second
stage. After that the fuel particles continue their movement in random direction in third
stage. The uniformity of the dust air cloud is created by the random movement of fuel
particles.
3.
METHOD
Use of central burster is common method to disperse powdered substances in air by

detonation wave. Fig.1 shows the central burster device which includes bellow parts:
1-Cylindrical fuel container
2-Burster tube
3-Lower and upper thick plate (aluminum or brass) to prevent fuel dispersion in air
longitudinally
4- Fuel, aluminum powder
5- Burster charge, TNT
6- Detonator, standard No.8
394
Czech Republic
2006
[Content]
7-Polyethylene piece to support the detonator

8-Suitable paste, Epoxy resin & hardener
Ratio of height to diameter for cylindrical fuel container is 1.5. There are two circular
furrows (4 mm depth) on the lower and upper plates to connect the end of the fuel container
and burster tube to them adhesively. The fuel (aluminum powder) is placed between the
walls of the fuel container and burster tube. The burster charge including pieces of pressed
TNT is put into the burster tube, and poly ethylene piece supports detonator to tangent on top
of TNT pieces column.
Fig. 1. Cross section view of test device (with central burster) and its items.
4.
TEST DESIGN
The first step for designing tests was choice of different fill-to-burster ratios as bellow:
1- Fill-to-burster ratio in test No.1 was 100/1
The choice of fuel weight is also important, because fuel dispersion process is a
quantitative process. For example, a small cloud is formed by a small FAE device and fuel
particles move a short distance on the average, but a large cloud is formed by a big FAE
device and fuel particles move a long distance contrary to small one. So at least, 4 kg
powdered substance should be used in each test device. Fill-to-burster ratio in tests No.1 and
No. 4 was 100/1, but their scales were different. Aluminum powder with 18 m average
sizes and pressed TNT with 14 mm diameter were used in practical tests. Details of four test
samples are shown in table1.
395
Table 1.
Czech Republic
2006
[Content]
Details of test samples
Test No.
The type of
fuel
The weight
of fuel (kg)
1
2
3
4
Al powder
Al powder
Al powder
Al powder
4.050
4.050
4.050
6.750
The size of
fuel
particles
(m)
18
18
18
18
The type of
burster
charge
TNT
TNT
TNT
TNT
The weight
of burster
charge
(g)
40.5
81.0
121.5
67.5
Fill-toburster
ratio
100 / 1
100 / 2
100 / 3
100 / 1
The density of aluminum is 2700 kg/m3, but tapped aluminum powder 18 m density is
1700 kg/m3. Height-to-diameter ratio for test devices was chosen 1.5. Fuel (aluminum
powder) is placed between the walls of the fuel container and burster tube. Fuel placed
volume can be calculated by bellow equation (1)
VF = ( / 4) [ (dFC)2 (dBT)2 ] hFC
(1)
where, dFC and hFC are the diameter and height of fuel container respectively, and dBT is
the diameter of the burster tube. The height of the fuel container and the diameter of the
burster tube respect to the diameter of the fuel container in test devices have ratios as:
hFC = 1.5 dFC
(2)
dBT = 0.2 dFC
(3)
If the volume of the required fuel was known, then by using equation (1), (2), and (3),
the amount of dFC , hFC and dBT cloud be determined. The volume of the fuel and the weight
of the fuel have bellow equation:
WF = VF . F
(4)
where, WF , VF and F are the weight , volume , and density of the fuel respectively . The
above equations were used to determine the dimension of the test devices. Table 2 shows the
dimension of test devices which were used in experimental tests.
Table 2. Dimension of test devices
Diameter of
Height of fuel
Diameter of
Volume of
Volume of
fuel container
container
burster tube fuel container
fuel
(mm)
(mm)
(mm)
(cm3)
(cm3)
1
130
195
26
2480*
2380
2
130
195
26
2480*
2380
3
130
195
26
2480*
2380
4
150
225
32
3970*
3790
* The volume of the fuel container should be a little more than the volume of the fuel.
Test No.
5.
TEST ARRANGEMENT
Sufficient safety should be considered in test site. High-speed camera is established on a

safe point and can be zoomed on the site of fuel dispersion. The test device is put on the pile
1 m upper than ground surface. There are four distance indicators to estimate dimension of
dust-air cloud. The first one is installed on the right of the test device 2 m far from it. The
396
Czech Republic
2006
[Content]
second one is installed 4 m far from it too. Two other ones are installed on the left
symmetrically. Fig.2 shows the arrangement of the test device and distance indicators. The
high-speed camera is set on 1000 frames/sec, and photography begins just stimulation of
detonator by electric source.
Fig. 2. Arrangement of test device and distance indicators:

a) Before test b) After test
6.
TESTS RESULTS
Table 3 shows the radius of dust-air clouds at different times after detonation of burster
charge in tests 1, 2, 3, and 4. Also Table 4 and Table 5 show the average thickness and
volume of dust-air clouds at different times respectively. You can see other tests results in
Table 6.
Fig.3.a, b, c, and d show dust-air cloud at 15 ms after detonation of burster in tests 1, 2,
3, and 4 respectively. The shape of dust-air clouds are pan-cake shaped, but the shape of
cloud in tests 1 and 4 are better. Fortunately, the dust-air clouds exploded in all tests, that
show proper fuel dispersion in air, and the concentration of aluminum powder was in
explosion limits. Explosion of dust-air clouds in tests 2 and 3 were not powerful, but in tests
1 and 4 were powerful. The color of the exploded clouds in tests 2 and 3 were dark, but in
tests 1 and 4 were bright, which indicate that combustion of aluminum powder in tests 2 and
3 was imperfect and in tests 1 and 4 was perfect. In tests1 and 4, the movement of fire was
rapid, uniform and powerful. The burnt effect and shock wave effect on the ground after the
test 4 was manifested. Therefore, fill-to-burster ratio 100/1 is better than other ratio such as
100/2 or 100/3, so the fill-to-burster ratio is close to optimum.
397
Czech Republic
2006
[Content]
Fig. 3. Explosion of dust-air clouds:

a) Test 1, 15 ms after detonation of burster
b) Test 2, 15 ms after detonation of burster
c) Test 3, 15 ms after detonation of burster
d) Test 4, 15 ms after detonation of burster
Table 3. Dust-air cloud radius (m) at different times after detonation of burster
charge
398
Time
Test No.
5 ms
10 ms
20 ms
30 ms
40 ms
50 ms
0.55
1.08
2.10
3.00
3.85
4.10
0.90
1.75
3.00
4.10
4.30
1.10
2.15
3.80
4.50
0.55
1.09
2.15
3.15
4.00
4.50
Czech Republic
2006
[Content]
Table 4. Average thickness (m) of dust-air cloud at different times after detonation
of burster charge
Time
Test No.
5 ms
10 ms
20 ms
30 ms
40 ms
50 ms
0.50
0.55
0.70
0.90
1.00
1.10
0.70
0.90
1.05
1.35
1.50
1.60
0.90
1.05
1.30
1.50
1.70
0.60
0.65
0.80
1.00
1.10
1.20
Table 5. Volume (m3) of dust-air cloud at different times after detonation of burster
charge
Time
Test No.
5 ms
10 ms
20 ms
30 ms
40 ms
50 ms
0.47
2.01
9.69
25.43
46.54
58.06
1.78
8.65
29.67
71.26
87.09
3.42
15.24
58.94
95.38
0.57
2.42
11.61
31.16
55.26
76.30
Table 6. Specification of dust-air cloud, clouds explosion and explosion effects on

the ground in the accomplished tests
7.
Test No.
The shape of
dust-air cloud
Explosion of
cloud
Power of cloud
explosion
Effects on the
ground
Burnt and
shocked
Excellent
Perfect (%100)
Powerful
Very good
Imperfect (%80)
Weak
No effect
Good
Imperfect (%90)
Weak
No effect
Excellent
Perfect (%100)
Very powerful
Burnt and
shocked
completely
DISCUSSION
Radius of dust-air clouds in Table 3 are used to draw curves of clouds radius versus time
for accomplished tests 1, 2, 3, and 4 in Fig.4. At first, the growth of clouds radius are high,
but then the rate of cloud growth decreases gradually. If burster charge increases, the rate of
cloud expansion increases, but duration of cloud expansion decreases. So, when the fill-toburster ratio changes, also the rate of cloud expansion changes, but the final radius of cloud
changes little.
399
Czech Republic
[Content]
5
4
Series1
Series2
Series3
Cloud radius , m
Cloud radius , m
2006
5
4
Series1
Series4
2
1
20
40
60
80
20
Time , ms
40
60
80
Time , ms
Fig. 4. Fig. 4 The curves of clouds radius vs time. The series 1, 2, 3, and 4 present
the tests No. 1, 2, 3, and 4 respectively
a) The curves which concern to tests No. 1, 2, and 3
b) b) The curves which, concern to tests No. 1, and 4.
When burster is detonated, at first, the volume of cloud is minimal, but then it grows
gradually. When the volume of cloud is minimal, its average fuel concentration is maximal.
So, if the cloud grows, its volume increases, and its average fuel concentration decreases.
When the fuel concentration is close to the stoichiometric concentration, explosion power of
fuel-air cloud would be at most level. Whatever the fuel concentration becomes far from the
stoichiometric concentration, the explosion power would be lower. If the fuel concentration
is more than the upper explosion limit or less than the lower explosion, in other words, if the
fuel concentration is out of explosion limits, the fuel-air cloud can not explode. Therefore,
delay time between detonation of burster charge and second charge is very important. When
the second charge is detonated to initiate the cloud explosion, the concentration of the fuel
should be close to its stoichiometric concentration in air.
90
120
3
Cloud volume , m
80
Series1
60
Series2
Series3
40
3
Cloud volume , m
80
100
70
60
50
Series1
40
Series4
30
20
10
20
0
0
0
0
20
40
60
80
20
40
60
80
Time , ms
Time , ms
Fig. 5. The curves of cloud volume vs time. Series 1, 2, 3, and 4 denote to tests
No. 1, 2, 3, and 4 respectively.
The volumes of clouds are presented in Table 5. Also Fig.5 shows the curves of the
clouds volume versus time after detonation of burster in tests No. 1, 2, 3, and 4. The cloud
volume VCloud in which, aluminum powder concentration receives on its stoichiometric
concentration in air is computed by bellow equations. The stoichiometric concentration of
aluminum powder in air is 330 gr/m3.
CAl =
400
W Al
VCloud
(5)
Czech Republic
2006
[Content]
The weight of fuel (aluminum powder) in test device 1, 2, and 3 were 4.050 kg. So, the
cloud volume in which, the concentration of aluminum powder receives to its stoichiometric
concentration can be computed:
VSt, 1 = VSt, 2 = VSt, 3 = 4.050 / 0.330 = 12.3 m3
Also, the stoichiometric volume of cloud in test No. 4 can be calculated:
VSt, 4 = 6.750 / 0.330 = 20.5 m3
Therefore, when the volume of cloud in tests No. 1, 2, and 3 were 12.3 m3, and in test
No. 4 was 20.5 m3, the average concentrations of aluminum powder have been the
stoichiometric concentration. The times at which, the average concentrations of aluminum
powder in four accomplished practical tests have been the stoichiometric concentration or
close to it, are shown in Table 7.
Table 7. The delay times after burster detonation in which, the experimental tests
reached to stoichiometric concentration.
Test No.
Time
22 ms
11 ms
8 ms
25 ms
The experimental data and photographs of tests (taken by high-speed photography) show
the fill-to-burster ratio 100/1 is closer to the optimum amount than others ratios such as
100/2 and 100/3. In other words, the weight of burster charge respect to the weight of
powder fuel should be %1 or less. Above subject is demonstrated by sufficient reasons:
1- The shapes of the clouds in tests No. 1 and 4 are more pan-cake shaped than others.
The thickness of the clouds in tests No. 1 and 4 is thin which, shows aluminum powder
dispersed radially.
2- The power of cloud explosion in tests 1 and 4 was very high. After the test 4, the test
site was photographed; it shows strong destructive effects on the ground surface.
3- The brightness of cloud explosion in tests 1 and 4 was more than other tests.
4- The colors of clouds after explosion were bright in test 1 and 4, but the colors of
clouds after explosion were dark relatively in tests 2 and 3. The brightness of clouds shows
that all of aluminum powders have combusted completely, but the darkness of clouds after
cloud explosion shows that parts of aluminum powder have combusted imperfectly.
5- The taken film from test 4 shows that dust-air cloud has expanded in two distinct
steps; its reason is that, in first step the dust-air cloud was expanded by detonation wave
produced by burster, and in second step, the cloud was expanded by shock wave produced
by the strong explosion of dust-air cloud. But in other tests especially tests 2 and 3, the
second step of cloud expansion didnt occur.
6- The real double-event FAE devices have two separated charge. The first charge
(burster charge) is used to disperse the fuel in air, and second charge is used to initiate cloud
explosion. The delay time between them is more important. This delay time in real FAE
devices is 20-50 ms [2]. The delay time is a time at which, the fuel concentration in the cloud
is close to stoichiometric concentration. Table 7 shows the time at which, the fuel
concentration is stoichiometry, so these times are proper to the delay time. Appropriate delay
401
Czech Republic
2006
[Content]
times for tests 1, 2, 3, and 4 are 22 ms, 11 ms, 8 ms, and 25 ms respectively. The result is
that the test device 1 and 4 are close to the real types according to their proper delay times.
Therefore, above reasons demonstrate that optimum fill-to-burster ratio was 100/1 in
tests 1 and 4, but their scales were different. If they are investigated exactly, the result will
be that:
1- Optimum fill-to-burster ratio for large FAE devices is 100/1.
2- Optimum fill-to-burster ratio for small FAE devices is 150/1 or 200/1.
Optimum fill-to-burster ratio changes in different scales, its main reason is that not
only the amount of fuel in large scale FAE device is more, but also the average distance
which, is traveled by the fuel particles, is longer. So, if the scale of FAE devices increases,
the fill-to-burster ratio decreases.
8.
CONCLUSION
Fill-to-burster ratio was first subject in this research. The ratios 100/1, 100/2, and 100/3
were chosen in experimental tests. Experiment results show the ratio 100/1 is close to
optimum ratio. Also it is proved that fill-to-burster ratio in large scale devices is a little less
than it in small scale devices, because in large scale device not only amount of fuel
(powdered substance) is more, but also the fuel particles travel longer distance on the
average to from a large cloud. Therefore, optimum fill-to-burster ratio is 200 in small scale,
but it decreases to 100 in large scale. Two thick aluminum or brass circular plates are used in
bottom end and upper end of cylindrical fuel container. They are more effective on cloud
shape, because they prevent dispersion of fuel particles longitudinally, but intensify fuel
dispersion radially; so pan-cake shaped cloud is formed.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
402
Remaly R., and Ragus E., "Explosive and cold gas dissemination of particulate matter", Deepak
Publishing Co., Hampton, Virginia, USA, 1983.
Warren L. McCabe, and Julian C. Smith, "Unit operation of chemical engineering", McGrawhill Book Company, New York, USA, 1967.
Ratzel A. C., and Constantineau E. J., "Aerosol cloud generation experiments", Sandia National
Laboratories Albuquerque, NM, 1993.
Gidaspow D., James L. Austing, "The explosive dissemination of particulate pyrotechnics and
explosive powder", Proceeding of the 13th International Pyrotechnics Seminar, Grand Junction,
Colorado, USA, 1988.
David C. Sayles, "Method of generating single-event, unconfined fuel-air detonation", United
States Patents, No 4463680, 1984.
Tulis A. J., "The role of nitro guanidine in the detonation of aluminum powder-air dispersions",
Proceeding of the 13th International Pyrotechnics Seminar, Karlsruhe, Germany, 1985.
Gidaspow D., James L. Austing, "The explosive dissemination of particulate pyrotechnics and
explosive powder", Proceeding of the 13th International Pyrotechnics Seminar, Grand Junction,
Colorado, USA, 1988.
Xiao Shao-qing, and Bai Chun-hua, "Investigation on control fuel dispersion of FAE for
ignition methods of burster", Beijing Institute of technology, China, 2001.
Austing J. L., Selman J. R., "The initiation shock wave in dispersed reactive particle
detonation. (III) Computation of Rankine-Hugoniot curve from real gas equations of state",
10
Czech Republic
2006
[Content]
Proceeding of the 13th International Pyrotechnics Seminar, Grand Junction, Colorado, USA,
1988.
[10] Labbe J. L. D. S., and M. F. Bardon, "Impulsive dissemination of liquid fuel in air", Paper of
M.S. Thesis, Department of Chemical Engineering, Royal Military College of Canada,
Kingston, Ontario, Canada, April 1991.
11
403
Czech Republic
2006
[Content]
PRELIMINARY STUDY OF THE BEHAVIER OF AL NANO

POWDERS - WATER SLURRY COMPOSITIONS AT NEAR ROOM
TEMPERATURES
Mikhail Laritchev *, Olga Laricheva **, Ilia Leipunsky *, Pavel Pshechenkov *,
Alexey Jigatch *, Michael Kuskov *, Valentin Sedoi ***
* Institute for Energy Problems of Chemical Physics RAS, 119334, Moscow, Russia
**Moscow Research Institute of Medical Ecology, 119638, Moscow, Russia
***Institute of High Current Electronics RAS, 634055, Tomsk, Russia
Abstract:
The mixtures of nano Al powders water slurries are considered now as an example of
perspective green propellants with relatively high energetic parameters. The
practical usage of Al-water compositions requires special knowledge of interaction of
components of mixtures at temperatures of composition manufacturing and storage
before utilization.
The subject of this study is the interaction between metal Al powders producing by
different ways and water in slurry and dispersion compositions. The temperature region
between room temperature and water boiling point was investigated.
Reactivity of different aluminum nanopowders of different origin with water were
compared by using the special technique developed for these investigations. Influence of
the origin of Al metal surface passivating coatings was evidenced. The existence of
different types of surface active centers for Al particles and possibility of passivation of
these centers by organic inhibitors was showed.
The special reactive passivating coating on the base of aluminum carbide was proposed
and realized for nanopowders planned for utilization with liquid water. This coating
permits operation with nanopowders at normal atmospheric condition protecting metal
surface from atmospheric oxygen and water vapors oxidation. But it produces the heat
and fuel gas-methane in the reaction with hot liquid water. Replacement of aluminum
oxide coating usually covering the surface of Al particles by aluminum carbide coating
permits increase the efficiency of energy applications of aluminum nanopowders.
Keywords:
nanopowders, water slurry, water dispersion, passivation, oxidation
1. INTRODUCTION
Fine and ultra-fine powders are actively used in pyrotechnics, explosives and
propellants. The area of investigations includes metallized solid [1], liquid [2], and metallized
gelled propellants [3]. Growing interest in nanopowders is stimulated by the possibility to
increase the inflammability, the burning rate, and the completeness of metal combustion and
to decrease the agglomeration.
Different methods are developed and used for nanosized metal powder production: laser,
electron- beam, Gen-Miller, electrical explosion, plasma-chemical, and others. The subject
of present study is the investigation of Al powders produced by the Gen-Miller method [4] in
the Institute for Energy Problems of Chemical Physics RAS (INEPCP RAS) and by the
method of electrical explosion [5] in the Institute of High Current Electronics SB RAS (IHCE
SB RAS) and by SIBTERMOCHIM Ltd, (Alex powders [6]).
404
Preliminary Study of the Behavier of Al Nano Powders Water Slurry . . .
Czech Republic
2006
[Content]
The interaction between metal Al powders and water has been investigated in the
temperature region between room temperature and water boiling point.
2. OBJECTS OF INVESTIGATION
Slurries and dispersions prepared on the base of Al nanopowders and distillated water
were investigated. Polyacrylamide was used as gelling agent. Average size of Al particles
was between 50-250 nm. Besides of widely used Al powders with the surface passivated by
aluminum oxide coatings were investigated the experimental powders passivated by coatings
on the base of aluminum carbide (with admixture of aluminum oxide) [7], aluminum nitride
and the mixture of aluminum carbide and aluminum nitride [8]. The last two types of
passivating coatings are reactive coatings. It means that these coatings can participate at high
temperature in oxidation (burning) processes with heat evolution but they can protect metal
Al from oxidation by atmospheric air at near room temperature. Such type of reactive
coatings on the base of Al carbide was proposed in INECP RAS [9] and successively realized
in INECP RAS and in IHCE SB RAS [7]. The Al carbide is capable to form the gasproof
coatings that are stable and can provide the effective protection of a metal core of aluminum
nanoparticles from the oxidation during contact with atmospheric air at room temperature
and natural humidity. The reaction of Al carbide with hot water leads to both: heat and fuel
gas methane evolution:
Al4C3 + 6H2O = 2Al2O3 + 3CH4 +
Two types of aluminum oxide coatings with hydrophobic properties were investigated in
this work: organosilicon origin [10] (type Al1-6, see Tabl.1), organic origin (types Alex and
AKP-1, see Tabl.1).
3. TECHNIQUES AND METHODS OF INVESTIGATION

Experimental setups include the installation for powder production [4, 5] and facilities for
characterization of the powder produced.
3.1 Determination of dispersion of powders

The specific surface area SBET of the samples was measured by the low-temperature
adsorption method (BET). The particle shape and size distributions were determined using a
transmission electron microscope (TEM). The mean diameters D and the standard deviation
were calculated from TEM data. The X-ray diffraction (XRD) analysis was used to
determine mean sizes of regions of coherent scattering (DRCS) for major phases.
3.2 Determination of metal aluminium content and passivating coating

composition of investigated Al powders
The high specific surface area of powders and the presence of passivating coating create
a need of measurement of metal aluminum content of investigated powders. This content
(Almet %) was determined by volumetric method, i.e. by measuring the hydrogen volume,
which was formed in reaction of aluminum with solution of sodium hydrate.
A composition and surface properties of nanopowders produced under various
conditions were studied by mass-spectroscopy techniques at the temperatures of 0 450 0C,
electron and X-ray diffraction methods. The composition of adsorbed gases has been
determined by heating of samples in the vacuum of 10-7 at various temperatures as well as
during linear heating with ramp function of temperature.
2
Mikhail Laritchev, Olga Laricheva, Ilia Leipunsky, Pavel Pshechenkov, . . .
405
Czech Republic
2006
[Content]
The presence of Al carbide in coating content was determined by the presence of

methane in composition of gaseous products of Al powder reaction with water. The gasliquid chromatographic analysis was used for this purpose.
3.3 Investigation of Al powders activity in oxidation reaction with liquid water

3.3.1 Study of evolution of gaseous reaction products for Al powder samples in H2O
slurry or dispersion composition heated linear in atmosphere of saturated water
vapour.
Fig.1 shows the scheme of apparatus for investigation of the reactivity of Al powder in
slurry or dispersion composition at a linear heating of sample in saturated water vapor up to
100oC. The last allowed preventing the change of water content in samples as the result of
water evaporation at least at the beginning step of reaction. The heating rate was of 5
degree/min. The combination of both method of differential thermal analysis (DTA) and
volumetric method was used. The possibility of simultaneous measurement of both: the
temperature of an investigating sample and the temperature of water heating this sample
allows to determine the temperature of a start of reaction between the metal aluminum and
the water. The increase of the value of these two temperatures residual gave the evidence of
reaction start. The possibility of measurement of amount of evolving hydrogen allows to
determinate both the start temperature of reaction of Al powder with water and fractional
conversion of metal aluminum as result of reaction.
9
10
8
7
11
5 4 3
Fig.1. The scheme of apparatus for investigation

of reactivity of Al powders + gel
composition at linear heating in saturated
water vapor up to 100oC.
1-external liquid thermostat, 2-distilleted
water, 3-internal liquid thermostat,
4-thermocooples, 5- sample (Al powder +
gel composition), 6-reactor, 7-cooler,
8-connective tube, 9-volumeter,
10-thermometer, 11-heater.
3.3.2 Study of reaction of Al powder samples in H2O slurry or dispersion composition

heated linear in open system.
The reaction was investigated by usage of Derivatograph C (MOM, Hungary)
equipped with an analyzer of hydrogen AH-01 (AlfaBassens, Russia [11]). The linear
heating of the samples was realized at atmospheric pressure in the flow of inert gas Ar. Inert
gas atmosphere was used to exclude the influence of ambient air on the oxidation process.
The concentration of water in investigated samples during the experiment was not stable due
to evaporation of water during the samples heating and also due to the reaction of water with
406
Czech Republic
2006
[Content]
metal Al. It decreases with the increase of the sample temperature. The closed crucible
containing investigated sample was used to minimize the water evaporation intensity. The
cover sealed the crucible nonhermetically.
The presence of water evaporation process makes difficult the quantitative interpretation of
experimental results. Only experiments performed at completely the similar conditions
(shape of crucible, rate of sample heating, mass of sample, velocity of inert gas flow and rate
of sampling flow for hydrogen analyzer) gave the comparable results. The experimental
results were the TG, DTG, DTA curves recording by derivatograph and dependence of
partial pressure of hydrogen in composition of gaseous reaction products evolved during the
sample heating recording by hydrogen analyzer. The special calibration experiments were
performed to calibrate the sensitivity of AH-1 to gaseous hydrogen in these measurements.
Fig.2 shows the scheme of using apparatus.
Samples of slurries or dispersions were investigated immediately after preparation
(manual mixing) and after several (1, 2, 3) days of samples exposure at room temperature in
atmosphere of saturated water vapour. Manual mixing was used to exclude mechanical or
ultrasonic activation of oxidation reaction of metal Al.
Fig.2. The scheme of apparatus
for
investigation
of
reactivity of Al powders
with liquid water down
to 130oC and water vapor
up
to
1000oC
at
atmospheric pressure.
1- inert gas (Ar), 2- gas
reducer, 3-flow meters,
4-thermometer,
5-saturator,
6-Derivatograph-C,
7-analyzer of hydrogen,
8-gas pump.
The experimental results allow determining the temperature ranges in which a failure of
the passivating coating continuousness was arose. Such collapse of coating continuousness
gives possible the interplay between the metal aluminum and liquid water accompanied with
both hydrogen and heat evolutions. The fine structure of maxima of heat and hydrogen
evolution permitted made the conclusion about the mechanism of passivating coating
destruction.
4. RESULTS
4.1. Study of evolution of gaseous reaction products for Al/H2O slurry samples
heated linear in atmosphere of saturated water vapour.
Figs.3 compares the temperature dependencies of a volume of gaseous reaction products
(mainly hydrogen) formed in reaction of Al powder/H2O slurries heated linearly in
atmosphere of the saturated water vapor with the heating rate approximately 5
4
407
Czech Republic
2006
[Content]
degree/minute. The duration of heating was 40 min. The samples composition was 30-32
mass percent of Al nanopowder and 68 -70 mass percent of water gel (including 3 % of
polyacrylamide). The surface of all Al powders was passivated using coatings on the base of
aluminum oxide, carbide and nitride. The shape of curves of temperature dependencies
testifies that the reaction between Al particles and gelled water passes at least through two
stages. The start temperature of the first stage of reaction (57-90.5) depends from the
composition of passivating coating.
140
120
Al2-Ox_1
N95
N2(air)
N2_2
Gel
-1
Al-CO2
100
VH2 (cm3)
80
60
40
20
0
-20
20
30
40
50
60
70
80
90
100
T ( C)
Fig.3. Temperature dependencies of

volume of gaseous reaction
products (mainly hydrogen)
evolved in reaction of Al
nanopowder / H2O slurries
heated linearly in atmosphere
of saturated water vapor. Curve
Gel pure 3% gel without Al
powder. Passivating coatings of
Al particles designed on the
base of a aluminium oxide,
carbide and nitrate
The aluminum oxide coatings are more stable with respect to hot water interaction than
the coatings on the base of Al carbide. The hydrophobization of Al particles surface
increases the start temperature of reaction metal core and hot water with respect to the
similar particles without the hydrophobic treatment. The more stable appears the
hydrophobic AKP-1 powder. The second stage of reaction is responsible for hydrogen
evolution at the temperature near 100.
A l- C O 2 - 1
A l- C O 2 - 3
A l- C O 2 - 4
A l- C O 2 - 5
A l- ( C O 2 - C O 2 )
A l- ( N 2 + C O 2 ) / N 2
A l- ( N 2 + C O 2 ) / N 2 ( H 2 O + S t )
H 2O
160
140
120
VH (CM )
100
80
60
40
20
0
-2 0
20
40
60
o
T ( C)
80
100
Fig.4. Temperature dependencies

of volume of gaseous
reaction products (mainly
hydrogen)
evolved
in
reaction of Al nanopowder /
H2O slurries heated linearly
in atmosphere of saturate
water vapour. Curve H2O
pure 3% gel without Al
powder.
Passivating
coatings of Al particles
designed on the base of
aluminium carbide [7].
Fig.4 shows the similar curves for Al powders passivated with coatings on the base of Al
carbide.
408
Czech Republic
2006
[Content]
Table 1. The data characterizing the reactivity of Al nanopowder/H2O slurry

samples (3% polyacrylamide). Al powders surface passivated with:
aluminum carbide coatings (lines 1-10); aluminum nitride coatings (lines
11-14); aluminum oxide coatings (lines 15-19). Hydrophobic aluminum
oxide coatings (lines 17, 19 organic hydrophobization; lines 18
organosilicon hydrophobization).
Al powder
Dispers
(nm)
Toin(oC)
Tofin(oC)
Almet1
(%)
AlmetO
(%)
Produser
(1)
1*
2*
3
4
5
6**
7
8
9**
10
11
12
13
14
(2)
(3)
(4)
(5)
(6)
Al-CO2-1
72.4+0.1 79.0+1.0 74.1+1.0
Al-CO2-1
72.5+0.2 80.0+1.0 96.0+4.0
100
Al-CO2-3
62.2
70.3
96.5
Al-CO2-4
59.6
69.8
98.8
Al-CO2-5
56.9
70.5
95.0
Al-CO2-5+St
80.0
96.7
85.3
Al-CO2-CO2
160
64.9
72.8
~100
Al(N2CO2)/N2
72.2
81.5
87.0
Al(N2CO2)/N2+St < 200
79.4
86.9
74.0
Al-N2-CO2
72.7
81.9
85.6
Al-N2 air
71.1
78.6
52.3
50-200
Al-N2_2
70.9
77.9
50.4
Al-N2_1
82.4
89.6
43.6
95
71.5
87.6
41.3
(7)
77.2
77.2
77.2
77.2
77.2
77.2
20.0
75.8
75.8
88.0
79.8
88.0
88.0
88.0
15
Al2-OX
88.0
INEPCP
RAS,
Moscow
SIBTER
MOCHIM
Ltd,
256
75.2+0.6 86.0+0.2
58.4+ 6
16
L-Alex - uncoated 70-150
75.0
89.2
100
88.2
17
L-Alex
78.0
90.6
97.3
88.0
18
Al 1-6
143
19
P-1
200
75.9+ 0.1 83.0+ 3.0 53.6+ 6.0

87.3
97.5
30.2
(8)
INEPCP
RAS,
Moscow
IHCE
SB RAS,
Tomsk
Tomsk
88.0
INEPCP
RAS,
Moscow
Moscow
* Sample Al-CO2-1 was investigated immediately after slurry preparation (line2) and 45 hours after slurry preparation (line 1).
** water with admixture of stabilizer (inhibitor)
Column (4) temperature of initiation of the first stage of reactions. Column (5)
temperature of finishing of the first stage of reactions. Column (6) percentage of metal Al
reacting with water during experiment (40 min). Column (7) initial percentage of metal Al
in aluminum powder.
Tabl.1 summarizes the data characterizing the reactivity of Al nanopowder / water slurry
samples including the powders covering with:
6
aluminum carbide coatings (with admixture of aluminum oxide) (lines 1-7),
409
Czech Republic
2006
[Content]
mixture of aluminum nitride and aluminum carbide coatings (lines 8-10),
aluminum nitride coatings (lines 11-14),
aluminum oxide coatings (lines 15-19), including hydrophobic aluminum oxide

coatings (lines 17,19 organic hydrophobization; lines 18 organosilicon
hydrophobization).
The data of tabl.1 show that reactivity of nanopowders passivated with aluminum
carbide coatings is higher than the reactivity of nanopowders passivated with aluminum
nitride and aluminum oxide coatings. The results of gas chromatographic analysis
demonstrate that the amount of Al carbide rises in content of coatings Al-CO2-1 Al-CO2-5
(lines 1-5). It means that the resistance of these coatings to hydrolysis (Toin value) is in direct
proportionality with the amount of admixture of Al oxide in coating content. Percentage of
metal Al reacting with water during the experiments (40 min) for these samples is near
100%.
Sample Al-CO2-1 was investigated immediately after slurry preparation (line 2) and
after 4-5 hours of preliminary exposition of slurry at room temperature in saturated water
vapor (line 1). The distinction of values Almet1 (column 6) for the line1 and for the lines 2
evidences that the reaction between liquid water and Al particles covered with aluminum
carbide is remarkable even at room temperature. The usage of stabilizer (inhibitor) increase
essentially the start temperature of reaction, that decreases the reactivity of Al nanopowder /
water slurry composition (compare lines 5 and 6 for nanopowder Al-CO2-5 producing by
Gen-Miller method and lines 8 and 9 for nanopowder Al(N2CO2)/N2 producing by method
of wire explosion).
The reactivity of coatings on the base of Al oxide is lower than the reactivity of coating
on the base of Al nitride (Toin = ~75oC and ~71oC respectively) with the exception of sample
Al-N2_1 kept during 6 months at normal atmospheric conditions (Toin = ~82.4oC).
Apparently long term contact of this sample with atmospheric air leaded to essential
strengthening of this passivating coating.
Another distinction between Al oxide and Al nitride coatings is the amount of metal Al
reacting during experiments (Almet1). For the powders with coatings on the base of Al
nitride Almet1=~40-50% and for Al oxide coating Almet1=~90% (for powders with the
similar average size of particles). The nature of this phenomena needs in additional
investigation.
The powders passivated by aluminum oxide coatings with hydrophobic properties
appear the higher stability to hydrolysis than the similar powders without hydrophobic
treatment (compare lines 16 and 17 for nanopowders Alex; line 19 for hydrophobic
nanopowder AKP-1 with lines 7-15 for nonhydrophobic powders).
4.2 Study of reaction of Al powder samples in H2O slurry or dispersion

composition heated linear in open system.
The properties of passivating coatings of Al nanoparticles appearing in water slurry or
dispersion composition were investigated.
Fig.5 shows the TG, DTA curves recording by derivatograph for fresh prepared sample
of Al nanopowder / water slurry. The surface of Al particles was covered with passivating
coating designed on the base of mixture of carbide and oxide of aluminum.
DTA curve demonstrates at least two maxima (shoulder + maximum) of heat evolution
in the temperature region 90-120oC.
410
Czech Republic
2006
[Content]
Fig.6 shows the curve of hydrogen evolution in this experiment and demonstrates also
shoulder and maximum of hydrogen partial pressure in the similar temperature region. It
means that the passivating coating for this sample began to collapse at temperatures higher
90oC and this collapse passes through two stages. The collapse of passivating coating is
accompanied by the hydrogen evolution. The value of temperature of main maximum of heat
and hydrogen evolution is 108oC. It is higher the boiling temperature of distillated water
because the polyacrylamide gelling agent was used in this sample. For the dispersions of Al
nanopowders in distillated water the collapse temperature is lower 100oC.
108
0,7
0,6
PH2 (mm Hg)
0,5
0,4
Al-CO2, Fresh
0,3
0,2
97.5
0,1
0,0
-0,1
9
10
11
12
13
14
15
16
17
18
19
Time (min)
Fig.5. TG, DTA, T curves for Al / H2O slurry Fig.6. Time (temperature) dependence of
linear heating in inert argon flow up to
hydrogen evolution for Al / H2O
700oC. Passivating coating reaches the
slurry linear heating in inert argon
collapse at temperatures 97 and 107oC
flow up to 700oC. The hydrogen
(maxima of DTA curve). Al
evolution starts at temperature
nanopowder was passivated with
97.5oC. Hydrogen concentration
aluminum carbide coating (sample Alreaches maximum at 108oC. Al
CO2-1).
nanopowder was passivated with
aluminum carbide coating (sample
Al-CO2-1).
The presence of fine structure (several maximum or shoulders) on the curves of heat and
hydrogen evolution lets to propose the existence of several types of active surface centers
hot points (defects of structure or composition of passivating coatings or something like
that) responsible for the destruction of passivating coatings contacting with water. These
hot points can exist on the surface of dry coatings or can be the result of the coating
interaction with water during the slurries or dispersions preparation and storage. The heating
of nanopowders in aqueous medium increases the rate of hot point formation. This process
results in the collapse of surface films which make possible the oxidation of metal Al
accompanied by heat and hydrogen evolution. Hot points are different for the passivating
coatings having the different origin or composition and for Al particles having the different
manner of production.
Contact of Al particles with water during several hours even at room temperature
changes the properties of passivating coatings and results also in decrease of metal content in
the powder.
Fig.7. shows the time (temperature) dependences of hydrogen evolution for Al powder
heated in fresh prepared water slurry composition (black curve) and for the same sample
8
411
Czech Republic
2006
[Content]
after 1 day ageing at room temperature (red curve) in atmosphere of saturated water vapor.
The ageing decreases the temperatures characterizing the collapse of passivating coating and
decrease the amount of evolving hydrogen. The origin of passivating coating is Al oxide.
121
0,08
ASD-4, fresh
ASD-4, old
113 C
PH2 (mm Hg)
0,06
0,04
0,02
99 C
160
0,00
150
9
10
11
12
13
14
15
16
Fig.7. Curve of 2 evolution for the fresh

prepared mixture micron powder
ASD-4 and 2% water-gel (1:2):
black curve and for the same mixture
after 1 day ageing at room
temperature red curve. Linear
heating in Ar flow with rate 10
degree/min. The origin of passivating
coating is Al oxide.
Time (min)
Fig.8a demonstrates the time (temperature) dependence of hydrogen evolution for Al

powder heated in fresh prepared water slurry composition. Al nanopowder (Al-CO2-5)
passivated with aluminum carbide coating was used in this experiment. Fig.8b shows the
time dependence of hydrogen evolution for the same sample ageing 24 hours in atmosphere
of saturated water vapor at room temperature. Comparison the results of these two
measurements shows that the contact of nanopowder with water leads to decrease the
temperatures characterizing the coating collapse. The total amount of evolving hydrogen
increases with the ageing of sample. Further contact of nanopowder with water (more 30
hours) leads to the reverse results. It increases the temperatures characterizing the process of
hydrogen evolution, decreases essentially the amount of evolving hydrogen. This behavior of
passivating coating can be explained by transformation of Al carbide into Al oxide as a
result of reaction with water. The content of Al metal in the powder drops dramatically
during the ageing of sample.
o
77 C
85 C
0,5
0,5
Al-CO2-5, fresh
0,4
0,4
PH2 (mm Hg)
PH2 (mm Hg)
Al-CO2-5, old
0,3
0,2
o
0,1
90 C
58 C
68 C
0,3
0,2
82 C
70.7 C
58 C
33 C
0,1
0,0
0,0
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Time (min)
10
12
14
16
18
20
22
24
26
28
Time (min)
Fig.8a. Time dependence of hydrogen

evolution for fresh prepared Al / H2O
slurry sample linear heating in inert
argon flow up to 700oC.
412
Fig.8b. Time dependence of hydrogen

evolution for Al / H2O slurry sample
ageing 24 hours at room temperature
before linear heating in inert argon
flow up to 700oC.
Czech Republic
2006
[Content]
Figs.9 and 10 demonstrate the time dependences of Al metal contents for Al nanopowder
Alex ageing in water slurry composition. Fig.9 shows data for hydrophobic Alex in
distillated water slurry. Even 3-4 days contact of nanopowder with distillated water
decreases essentially its metal content. Fig.10 shows the dependence of metal content of
nonhydrophobic Alex in distillated water slurry + inhibitor. It is clear that admixture of
inhibitor increases the value of Alex powders life time by a factor of 5. The metal Al
content was measured by means of X-ray structural analysis. It is necessary to note that the
stability of hydrophobic Alex with respect to contact with liquid water (without inhibitor)
is substantially higher (on the factor 3) the same characteristic of nonhydrophobic Alex
powder.
Al (meta)%l
Al (metal) %
100
Alex (hydrophobic) / H2O slurry
Alex (nonhydrophobic) / H2O slurry + inh.
105
100
90
95
80
90
85
70
80
60
75
50
70
65
40
60
55
30
0
10
Time (days)
50
0
9 10 11 12 13 14 15 16 17 18 19 20
Time (days)
Fig.9. Time dependence (days) of metal Al Fig.10. Time dependence (days) of metal Al
content of Alex (nonhydrophobic) /
content of Alex (hydrophobic) / H2O
slurry sample (distilleted H2O)
H2O slurry sample (distilleted H2O +
ageing at room temperature in
inhibitor) ageing at room temperature
saturated water vapor atmosphere.
in saturated water vapor atmosphere.
Al(N2-CO2)/N2 + H2O(+inh), fresh

Al(N2-CO2)/N2 + H2O(+inh), 2 days
0,8
0,4
PH (mm Hg/gr)
0,6
0,2
0,0
20
30
40
50
60
70
80
90
o
T ( C)
10
100 110 120 130 140 150 160
Fig.11. Temperature dependence of

hydrogen evolution for Al
nanopowder in distilleted water
( + inhibitor) dispersion
composition for fresh prepared
sample and for same sample
after two days of ageing ot
room temperature in saturated
water
vapor
atmosphere.
Surface of Al particles
passivated with coating on the
base of Al nitride and Al
carbide
413
Czech Republic
2006
[Content]
Fig.11 shows the temperature dependence of hydrogen evolution for Al nanopowder

passivated with coating on the base of Al nitride and Al carbide in distillated water with
admixture of inhibitor dispersion composition for fresh prepared sample and for same
sample after two days of ageing at room temperature in saturated water vapor atmosphere.
The ageing decreases the amount of hydrogen evolving during the experiment (this is
evidence of the coating strengthening) but dont changes remarkably the maximum of
hydrogen evolution temperature and Al metal percentage in the sample.
CONCLUSION
1. The interaction between Al nanopowders producing by different ways and distillated
water in slurry and dispersion compositions was investigated for the temperature region
between room temperature and water boiling point. The surface of Al particles was
passivated with coatings having the various origin and composition: Al oxide, Al nitride, Al
carbide and its mixture. Al oxide coatings were nonhydrophobic and hydrophobic.
2. The special technique was developed for comparable express analysis characterizing
the activity of different types of nanopowders with respect to water or water slurries.
3. More stable to the storage in water slurry or in water dispersion composition are
nanopowders passivated hydrophobic coatings on the base of Al oxide. The powders covered
with reactive Al carbide coatings are more active with respect to water or water slurries.
4. The manners of stabilization Al nanopowders in water slurry and water dispersion
composition at near room temperature were recognized.
Acknowledgment:
This work was performed with financial support by INTAS (grant 03-53-5203) and by RFBR
(grant 05-08-17928).
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
414
L.T. DE LUCA, L. GALFETTI, F. SEVERINI, L. MEDA, G. MARRA, A.B. VOROZHTSOV, V.S.

SEDOI, AND V.A. BABUK: Burning of Nano-Aluminized Composite Rocket Propellants,
Combustion, Explosion, and Shock Waves, Vol. 41, No.6, pp. 680-692, 2005
F. TEPPER AND L. KALEDIN: Combustion characteristics of kerosene containing Alex nanoaluminum, Combustion of Energetic material,. Edited by K.K. Kuo and L.T. De Luca. Beggel
House Inc., New York. Wallingford, pp. 195-206, 2002
V.G. IVANOV, O.V.GAVRILYUK, O.V. GLAZKOV, and M.N. SAFRONOV: Specific Features of
the Reactions between Ultrafine Aluminum and Water in a Combustion Regime, Combustion,
Explosion, and Shock Waves, Vol 36, No. 2, pp. 213-219, 2000
GEN M.YA., MILLER A.V.: A technique for generation of aerosols, Patent USSR N814432,
Inventions Bulletin, 1981, 11, 23.03.81, P.25.
YU.F. IVANOV, M.N. OSMONOLIEV, V.S. SEDOI, V.A. ARKHIPOV, S.S. BONDARCHUK, A.A.
VOROZHTSOV, A.G. KOROTKIKH, AND V.T. KUZNETSOV: Production of Ultra-Fine Powders
and Their Use in High Energetic Compositions, Propellants, Explosives, Pyrotechnics, Vol. 28,
No. 6, pp. 319-333, 2003
M.M. MENCH, K.K. KUO, C.L. YEH, Y.C.LU.: Comparison of Thermal Behavior of Regular
and Ultra-fine Aluminum Powders (Alex) Made from Plasma Explosion Process, Combustion
Sci and Tech, 135, pp. 269-292, 1998
M.N. LARITCHEV, I.O. LEIPUNSKY, O.O. LARICHEVA, P.A. PSHECHENKOV, A.N. JIGATCH,
M.L. KUSKOV, V.S. SEDOI: Nw reactive surface coatings for Al metal nanoparticles, Pros. of
Czech Republic
2006
[Content]
36th International Annual Conference of ICT & 32nd International Pyrotechnics Seminar, ICT,
Karlsruhe, Germany, June 28 - July 1, p 110(1-11), 2005.
[8] VALENTIN SEDOI, VICTOR IVANOV, YURI IVANOV, MIRSLAN OSMONOLIEV,
MIKHAIL LARICHEV, AND OLGA LARICHEVA: Synthesis by the exploding wire method
and characterization of nanoaluminum powder with inorganic coating, Pros. of 36th
International Annual Conference of ICT & 32nd International Pyrotechnics Seminar, ICT,
Karlsruhe, Germany, June 28 - July 1, p 98 (1-11), 2005.
[9] M.N.LARITCHEV, A.N.JIGATCH, I.O.LEIPUNSKY, M.L.KUSKOV, A.A.PSHECHENKOV: Air
oxidation of aluminum nano particles. Influence of size and surface coating of particles on the
mechanism of aluminium oxidation process, Pros. of 9 International Workshop on Combustion
and Propulsion Novel energetic materials and applications, September 14-18, 2003, La
Spezia, Italy.
[10] JIGATCH A.N., LEIPUNSKY I.O., KUSKOV M.L., PSHECHENKOV P.A., BEREZKINA N.G.,
LARITCHEV M.N., KRASOVSKY V.G.: Synthesis of coatings on the surface of ultra fine
aluminum particles, Khimicheskaya Fizika (in Russian)//, 2002, v.21, N4, P.72-78.
[11] Hydrogen Analyzer (AH-1, AlfaBassence), Russia: http://www.alfabassens.ru
12
415
Czech Republic
2006
[Content]
CARBON NANOTUBES AS SUPPORT FOR CO/TIO2

ELECTROCATALYSTS FOR HYDROGEN EVOLUTION
O. Popovski, P. Paunovi*, A. Dimitrov*, D. Slavkov*, R. Smileski,
M. Tasev* and S. Hadi Jordanov*
Military Academy "General Mihailo Apostolski", Skopje, R. Macedonia
*Faculty of Technology and Metallurgy, University Sts. Cyril and Methodius, Skopje,
R. Macedonia
Abstract:
The aim of this work was to apply multiwalled carbon nanotubes (MWCNTs) as a
catalyst support in composite Co/TiO2/C electrocatalysts for hydrogen evolution. In
order to elucidate the effect of MWCNTs, corresponding catalysts deposited on
traditional carbon material Vulcan XC-72 were prepared as well.
Structural characterization was performed by SEM, IR and XPS method. It is shown that
intrinsic catalytic activity of catalysts deposited on both MWCNTs and Vulcan XC-72
are almost the same.
It was shown that the surface area of the catalysts deposited on MWCNTs is twice
higher than the one on classical Vulcan XC-72 substrate.
Electrochemical characterization was performed by cyclic voltammetry and steady-state
galvanostatic methods in alkaline solution (3,5 M KOH). Introduction of MWCNTs was
shown to increase considerably catalysts activity. The overpotential (at i = 20 mAcm-2)
is for 45 mV lower than corresponding one with Vulcan XC-72.
These results are valuable per se and also are good indicators for further improvement
of this type of catalysts.
Key words:
composite catalyst, multiwalled carbon nanotubes (MWCNTs),
hydrogen evolution reaction (HER), Co, TiO2
1.
INTRODUCTION
One of the most important fields of modern electrocatalysis is engineering of new

electrode materials for hydrogen evolution and fuel cells. Electrode materials should be i)
catalytically active, ii) chemically and mechanically stable and iii) inexpensive. The
composite materials based on mixture of metal from the right side of the transition series
(hyper d) and metal (or its compound) from the left side of transition series (hypo d) appear
as effective to satisfy the above criteria. Physical background of hypo-hyper d-electrocatalysts has given by Brewers valence-bon theory[1] and electrochemically interpreted by
Jaki[2].
In our previous papers[3,4] hypo-hyper d-electrocatalysts containing amorphous TiO2 as a
hypo d-phase and non-platinum metals (Ni, Co or CoNi) as a hyper d phase were
structurally and electrochemically characterized. The catalyst based on Co has been shown
as the most active one, but we believe that this is not its highest performance.
416
Carbon Nanotubes as Support for Co/TiO2 Electrocatalysts for Hydrogen . . .
Czech Republic
2006
[Content]
It is well known that there are two approaches to improve the catalysts: i) increase of
intrinsic activity by rising of interaction between hypo and hyper d-phases and ii) developing
of real surface area of the electrode.
In the previous works[3,4], the effect of increasing of hypo-hyper d-interaction was
studied involving anatase instead amorphous TiO2 as a hypo phase. The aim of this work is
to determine the effect of developing catalysts surface area by employing multiwalled
carbon nanotubes (MWCNTs) as a carbon substrate instead Vulcan XC-72.
Iijimas discovery in 1991[5] has initiated intensive research on carbon nanotubes
(CNTs). Their unique and superior physical, mechanical and electrical properties allow wide
range of application as e.g. hydrogen or energy storage, electrode material in electrochemical
capacitors, field emission electron source, fillers in polymer composites etc. The
extraordinary conductivity, mechanical stability and inner geometry make as well CNTs
promising material as catalyst support replacing traditionally carbon blacks.
2.
EXPERIMENTAL
The catalysts were prepared by sol-gel procedure described elsewhere[3,6]. Two catalysts
of the same composition (10%Co + 18%TiO2) were prepared, but on different carbon
substrate (MWCNTs and Vulcan XC-72, resp.). The first one was deposited on Vulcan XC72 as a catalyst support, while the second one on multiwalled carbon nanotubes (MWCNTs).
Structural analysis was preformed by scanning electron microscopy (SEM), x-ray
photoelectron spectroscopy (XPS) and Fourie transformed infrared spectroscopy (FTIR).
The real vs. geometric surface area ratio was determined by cyclic voltammetry as
described elsewhere[4]. Polarization characteristics were measured by steady-state
galvanostatic method using gas-diffusion working electrodes[4] in alkaline solution (3,5 M
KOH). Reference electrode was Hg/HgO, while as counter electrode platinum wire was
used.
3.
In Fig 1, SEM photographies of both catalysts are shown. The catalysts particles
deposited onto Vulcan XC-72 are of a spherical shape and grouped in clusters sized of 100200 nm, so that good adherence was achieved. There is a number of holes between the
aggregates on the catalysts surface, higher specific surface area, and consequently higher
catalytic activity is created due to these features.
The catalysts particles grafted on MWCNTs are grouped in smaller clusters than those
on Vulcan XC-72. This causes holes between particles to appear. Due to intrinsic
geometrical features of MWCNTs (empty cylinders with higher developed surface area as
compared to the spherical particles of Vulcan XC-72), they possess inner holes so that inner
porosity of MWCNTs is considerably higher than the one of Vulcan XC-72. This is
important for better dispersion of other catalysts components of the on the surface,
especially of active catalytic centers metallic particles.
SEM analysis is not able to give any information on the nature of bonding (mechanical
or chemical) of different particles into the catalysts clusters. This information can be
obtained by further spectroscopic investigation, by e.g. infrared spectroscopy or X-ray
photoelectron spectroscopy (XPS), as shown bellow.
2
Orce Popovski, P. Paunovi, A. Dimitrov, D. Slavkov, D. Smileski, . . .
417
Czech Republic
2006
a)
[Content]
b)
Fig. 1. SEM photographs of the catalyst deposited on a) Vulcan XC-72 and

b) MWCNTs
In Fig.2 XPS spectra of Ti and Co are given. The values of binding energy on maximum
of the Ti2p1/2 and Ti2p3/2 peaks correspond to existence of TiO2 on the catalysts surface
(Fig. 2a). The maximum of Ti2p1/2 corresponds to the table value of TiO2, while the value of
Ti2p3/2 peak is for 0,1 eV different from the tabled one (458,9 eV vs. 459,0 eV), what
suggests that interaction between TiO2 phase with metallic phase exists. The difference
between the maxima is almost equal to the standard energy of binding of TiO2 (5,7 eV).
Ti2p3/2
458,9 eV
Co2p3/2
470
Intensity, arb. units
Intensity, arb. units
781,5 eV
Ti2p1/2
464,7 eV
465
460
455
Binding energy, eV
Co2p1/2
Sat.
810
797,2 eV
800
Sat.
790
780
770
Binding energy, eV
a)
b)
4+
Fig. 2. XPS spectra of a) Ti (Ti2p1/2 and Ti2p3/2) and b) Co2+ (Co2p1/2 and
Co2p3/2), taken at the catalysts surface
The values of maxima of Co2p1/2 and Co2p3/2 peaks (797,2 and 781,5 eV) indicate
existence of cobalt as Co(OH)2 phase on the catalyst surface. The peaks are spread and with
418
Czech Republic
2006
[Content]
low intensity, what suggest that amount of cobalt into elemental state is considerably lower
than in valence state (Co2+). Based on the determined values of the binding energies (Fig. 2),
it is very probable that, next to Co(OH)2, Co2+ also exists in small amount of CoO and
CoTiO3. This is supported with the satellite peaks on 787,3 and 803,3 eV, which is a
consequence of paramagnetism of CoO phase.
XPS analysis also indicates that some interaction between the hypo and hyper dcomponents takes place.
On the FTIR spectra of investigated catalysts the only band of interest originates by
TiO2[7]. To determine hypo-hyper d-interaction one has to compare values of maximum of
TiO2 band incorporated in the electrocatalysts and maximum of the band from pure TiO2
obtained in identical conditions as that in electrocatalyst. The shift of the maximum value is
almost the same for both compared catalysts. This shift is 55 to 57 cm-1 related to the
maximum of pure TiO2 (Table 1). So, it shows that there is an intrinsic interaction between
hypo oxide and hyper d-metallic phase into the catalysts. The level of hypo-hyper dinteraction is the same in both investigated systems.
Table 1. Change of wave number, double layer capacity, Cdl and ratio of real vs.
geometrical surface area, SR/SG
Catalyst
wave number/cm-1
Cdl/mFcm-2
SR/SG
10%Co + 18%TiO2 + Vulcan XC-72
57
153,7
2560
10%Co + 18%TiO2 + MWCNTs
55
396,00
6600
-1
0.003
1 mVs
-1
2 mVs
0.002
-1
5 mVs
-1
8 mVs
0.001
-1
i, Acm
-2
10 mVs
0.000
-0.001
-0.002
-0.003
-0.22
-0.20
-0.18
-0.16
-0.14
-0.12
E vs. Hg/HgO, V
Fig. 3. Cyclic voltammograms in the potential region of double layer charging for
different scan rates (10%Co + 18%TiO2 + Vulcan XC-72)
419
Czech Republic
2006
[Content]
In order to compare the surface roughness (the ratio of real vs. geometrical surface area,
SR/SG), the capacities of double layer (Cdl) of the matched catalysts were determined by
cyclic voltammetry in the potential region where only charging of double layer occurs. Scan
rates of 1 to 10 mVs-1 were applied. Current density of double layer charging icap was
adopted to be as an average value of anodic and cathodic current densities at potential at
middle of the scanned interval (Fig.3.) Linear dependence of the icap vs. scan rate (
E
) was
t
determined with a slope equal to the value of double layer capacity of the electrode surface:
C dl =
di cap
(1)
E
d
dt
The real vs. geometric surface area ratio was determined as a ratio between Cdl of the
catalyst versus double layer capacity of ideally flat oxide surface (60 Fcm-2) The values of
double layer capacity and the ratio of real versus geometrical surface area of investigated
catalysts are shown in Table 1. As can be seen, the surface area of both catalysts is highly
developed, but the one on MWCNTs is higher 2,5 times (see Table 1). Also, Cdl value of
MWCNTs are almost twice higher than that of Vulcan XC-72 (Cdl(MWCNT) = 331 mFcm-2
versus Cdl(Vulc.XC-72) = 179 mFcm-2), what means that real surface area of MWCNTs is twice
then the surface of Vulcan XC-72 [8]. So, carbon nanotubes are the crucial factor which
contributes for developing of the surface area of the catalyst as a whole.
10% Co + TiO2 + Vulcan XC-72

10% Co + TiO2 + MWCNTs
-50
, mV
-100
-150
-200
-250
-300
-350
-10
10
20
30
40
50
60
70
80
90
i, mAcm-2
Fig. 4. Polarization curves of the matched catalysts for hydrogen evolution in

alkaline solution
420
Czech Republic
2006
[Content]
Substitution of Vulcan XC-72 with MWCNTs as carbon substrate results in a decrease

of HER overpotential for as much as 40 mV. This increase of the catalytic activity is results
only of more developed surface area of the catalyst deposited on MWCNTs. Hypo-hyper dinteraction is the same in both systems, so there is no improvement in intrinsic activity. The
higher activity for hydrogen evolution is a result first of all of the nature of carbon nanotubes
(their inner structure, surface area and conductive characteristics). The high developed
surface area enables better dispersion of active catalytic centers across the catalysts surface.
Also, extra-conductive properties enable easier electron exchange with hydrogen protons,
what intensifies formation of adsorbed hydrogen atoms and further hydrogen molecules. On
the other side, its high inter- and trans-particle porosity as well as its geometry (empty
cylinders), does facilitates the escape of hydrogen molecules from catalysts surface.
The effect of substitution of Vulcan XC-72 with MWCNTs in case of Co-based
catalysts is not so pronounced as in case of Ni-based ones where the decrease of the
overpotential for HER is even more than 120 mV[9]. This is due to very small particles of
cobalt (lower than 2 nm)[3] as a active catalytic centers, so their dispersion in the on the
Vulcan substrate is very high.
4.
CONCLUSION
At least three main conclusions could be drawn:
1)
Applying the modified sol-gel procedure provides production of catalysts that

exhibit pronounced synergetic effect for the hydrogen evolution reaction.
2)
This effect is due to both high developed surface area as result of obtained
nanostructured components of the catalysts and the strong metal-support
interaction between hypo and hyper d- electronic phases, which is equal for both
investigated systems.
3)
Using MWCNTs as carbon substrate instead Vulcan XC-72 creates an increased

catalytic activity, as a result of higher developed surface area as well as better
dispersion of active catalytic centers over the catalysts surface.
ACKNOWLEDGEMENT
This paper has been supported by and carried out within the EU Project
PROMETHEAS PL ICA2-2001-10037 and Project 13-1650/4-02 of Ministry of Education
and Science of R. Macedonia.
The authors are grateful and indebted to Professor Milan Jaki, University of Belgrade,
for his bountiful and extensive help in the literature supply and valuable advices for both
theoretical consideration and practical aspects of catalysts preparation.
Special thank to the staff of the Institute of Electrochemistry and Energy Systems,
Bulgarian Academy of Science, Sofia for the fruitful collaboration and the hospitality in the
Academys laboratories during investigation.
421
Czech Republic
2006
[Content]
REFERENCES
[1]
L. BREWER, Thermodynamic Stability and Bond Character in Relation to Electronic

Structure and Crystal Structure (Electronic Structure and Alloy Chemistry of
Transition Elemenets, P.A. BECK eds., p. 221-235, Intercsience, New York (1963),
Dover New York (1965),
[2]
M. M. JAKI, Advances in Electrocatalysis for Hydrogen Evolution in the Light of

the Brewer-Engel Valence-Bond Theory, Int. J. of Hydr. Ener., 12, p. 727-752 (1987)
[3]
P. PAUNOVI, O. POPOVSKI, M. TASEV, R. SMILEVSKI and S. HADI

JORDANOV, Sophisticated Electrocatalysts for Economical Production of Hydrogen
Part A: Structural Characterization, New Trends in Research of Energetic Materials
Proceedings of the VIII. Seminar, p. 719-724, Pardubice Czech Republic, April 19-21
(2005)
[4]
O. POPOVSKI, P. PAUNOVI, R. SMILEVSKI and S. HADI JORDANOV,

Sophisticated Electrocatalysts for Economical Production of Hydrogen Part B:
Electrochemical Characterization, New Trends in Research of Energetic Materials
Proceedings of the VIII. Seminar, p. 739-744, Pardubice Czech Republic, April 19-21
(2005)
[5]
S. LIJIMA, Nature, 354, 56 (1991)
[6]
S. HADI JORDANOV, P. PAUNOVI, O. POPOVSKI, A. DIMITROV and D.

SLAVKOV, Electrocatalysts in the last 30 years - from precious metals to cheaper but
sophisticated complex systems, Bull. Chem. Technol. Macedonia, 23, No2, p. 101-112,
(2004)
[7]
O. POPOVSKI, Synthesis and characterization of composite electrocatalysts for

hydrogen evolution reaction based on cobalt, Ph.D. Thesis, University Sts. Cyril and
Methodius, Skopje (2005)
[8]
P. PAUNOVI, O. POPOVSKI, S. HADI JORDANOV, A. DIMITROV and D.

SLAVKOV, Modification for improvement of catalyst materials for hydrogen
evolution, J. Serb. Chem Soc. Vol. 71 (2006) (in press)
[9]
P. PAUNOVI, O. POPOVSKI, A. DIMITROV, D. SLAVKOV, E. LEFTEROVA

and S. HADI JORDANOV, Improvement of Performances of Complex Non-Platinum
Electrode Materials for Hydrogen Evolution, Electrochimica Acta, (to be published)
422
Czech Republic
2006
[Content]
SLIDING SHOCK WAVE CAUSING DEFORMATION OF THE METAL

PLATE ATTACHED TO THE COVER
Huiying Zhao, Zhaowu Shen
(Department of Modern Mechanics, University of Science and Technology of China, Hefei, China
230027)
Abstract: Shock wave parameters of the cover are obtained by using shock wave limiting
curve equation. Based on the laws of shock wave refraction and reflection between the
cover and the plate and attenuation of the power of the shock wave during the
propagating process, the formulas about the detonation press loaded on the surface of the
metal plate has been gained. At the same time the relationship between the thickness of
the cover and the deformation degree of the metal plate has been presented. Shaped
charge with lager apex angle liner attached to the cover has been initiated from the center
point; as a result an explosion formed projectile with fins has been recovered, which has
proved the validity of the theory furthermore.
Key words: sliding detonation, cover, shock wave, planar flyer plate, shaped charge, EFP
(explosion formed projectile).
Introduction
The values of the study on the infinitesimal substance under the load of sliding detonation
have been recognized. Many scientists have taken efforts on this aspect and many useful
semi-empirical equations have been presented, for example the Taylor equation deduced by
G.T.Taylor [1] in 1942 and improved by P.C.Chou [2]. The steady-state two-dimension
solution of the sliding detonation products flowing has been discussed in reference [3], and the
performance of the metal plate under the press of the sliding detonation has been studied in
reference [4-8]. However, there are few reports on the deformation of the metal plate attached
to the cover under sliding detonation.
If the cover is settled on the interface between the metal plate surface and the explosive,
and the explosive initiated at one side point, the deformation degree of the plate under sliding
detonation is weaker than that contacts with explosive directly. In the same way as the covers
distributed on the large apex angle liner or ball-type liner homogenously, the EFP with fins
can be produced and the stability and the attack ability of EFP are improved. So the
investigation on the metal performance loaded by sliding shock wave is a far-reaching work.
Since the process of the metal under sliding detonation is very complex, it is difficult to
gain the analysis solution as yet. This paper employs the polar curve to deduce the parameters
of the shock wave. The laws of the shock wave refracting and reflecting between the metal and
the cover has been used to analyze the deformation of the metal plate, which will provide the
theoretical proof for the cover design.
423
Czech Republic
2006
[Content]
In the shaped charge field, the model of planar flyer plate under detonation driving is
usually applied to solving the problem of liner element acceleration. So using shaped charge in
experiment is suitable to prove the affection of the thickness of cover to the deformation of
metal plate.
In this paper we analyzed the influence of the covers on the metal plate under sliding
detonation and examined it through experiments.
Theory and methods
1. Theoretical model of the shock wave parameters
The phase of detonation wave propagating determines the reflected and refracted shock
wave parameters. In most applications, the detonation of explosives usually progresses in its
characteristic form that can be well explained by Chapman-Jouguet (C-J) detonation theory
[9]. In Fig. 1, O is the contact point of the detonation wave with the substance interface, which
moves together with the detonation front wave with the same velocity as the initial particle
velocity; the detonation wave front OA contacts with the cover; OC is the shock wave front in
the cover. The interface between the explosive and the cover inflects to OB. Shock wave front
reaches the interface CC obliquely and the refraction and reflection accompanying with this
process and the shock wave front CD has been shaped. CC inflects to the position of CE. If
wave impedance of substance 2 is larger than that of substance 3, rarefaction wave will be
reflected; otherwise shock wave will be reflected [10].
1.1 Analysis of shock wave parameters.
The state solution of the detonation products is expressed by formula (1).
p = A (1a)
k +1
0
A=
k
0 D12
k +1
(1b)
where 0 is the density of the explosive, k is the polytropic index and D1 is the detonation
velocity of the explosive. The expanding process of the detonation products is in the
Prandtl-Mayer shape. The detonation parameters of section 1 can be expressed in following
formulas (2) and (3) [6].
p1 = 0 D12 /(k + 1) (2)

Ma 1 = 1
(3)
where P1 is the pressure after the wave and Ma1 is Mach number. Detonation products in
section 2 expand in Prandtl-Mayer shape. Equation (4) indicates the relationship between the
detonation pressure of 1 and that of 3.
424
Sliding Shock Wave Causing Deformation of the Metal Plate Attached to the . . .
Czech Republic
2006
[Content]
k 1
k 1
Ma21 + 1
p3 2
=
p 1 k 1
Ma132 + 1
2
(4)
And the detonation polar curve of section 2 has been shown as equation (5) [11].
k +1
k 1
arctan
Ma132 1 arctan Ma132 1
k +1
k 1
2 =
k +1
k +1
arctan
Ma132 1 arctan MaI21 1
1
k
(5)
The substance in the section II0 and II1 obeys mass conservation and momentum
conservation and shock compress empirical laws. The equation (6) is obtained based on
momentum conservation and shock compress empirical laws.
p 3 =
0 D12 sin 2
b2
a2
sin 2 (6)
D1
where a 2 and b2 are the shock compress parameters. Equation (7) can be acquired from
equation (4), (5) and (6).
2
3
M 12 +
(k + 1) 0 sin 2
b 0
2
k 1
a
sin 2 2
D1
k 1
k
2
(7a)
k 1
k +1
k +1
k 1
k 1
arctan
( Ma23 1) arctan Ma23 1
arctan
(Ma21 1) arctan Ma21 1
k +1
k +1
k 1
k 1
(7b)
a2
1
tan 2
D1 sin 2
= arctan
a2 2
tan 2
b2 + b2 1 +
D1 sin 2
The unknown parameters M 3 and 2 can be obtained by using equation (7), and then
p 3 is acquired using equation (6).
2. Theoretical analysis about the moving of the metal plate
425
Czech Republic
2006
[Content]
As Fig. 1, front shock wave in direction of OC impacts the surface of metal plate.
Assuming that the thickness of the metal plate has no change, and that the air friction is
ignorant, the metal plate collapses under the shock p ' 1 .
The pressure loading on the surface of the metal plate influences the plate velocity and
collapse angle. Fig. 2 presents the collapse process of the metal plate under shock. And Fig. 3
shows the acceleration process of the metal plate under shock.
OC denotes the front shock wave profile that impacts the surface of metal plate. Since the
thickness of the explosive is limited, the sparse wave from free surface of the explosive will
make the pressure of the detonation products decreasing. As the detonation pressure p 1
decreases to p '1 , the inflection angle of the metal plate changes from 0 to . When the
detonation pressure approaches to 0, the inflection angle of the metal reaches the maximum
value as shown in equation (8) [12].
p '1
= 1
8
p 1
max
where is inflection angle of the metal, max is the maximum value. Based on equation (4)
and (8), equation (9) can be obtained.
k
k 1
k 1
2
2 Ma 1 + 1
p3 =
)] (9)
[ p 1 (1
max
k 1 Ma132 + 1
2
Due to the continuous reflection interface, p 3 is equal to p1 . Shock wave is reflected

regularly on the interface of the cover and the metal. Shock pressure weakens in exponential
attenuation, as shown in (10).
p = p0 exp(ax)
(10)
where p is the pressure after attenuation at propagating distance x , p 0 is the initial

pressure and is attenuation coefficient. So after passing a cover of depth , the impact
pressure on the metal plate p' 1 is
pII' 1 = pII 1 exp(a / cos 2 ) (11)
In Fig. 3, one moving coordinate system is defined to move with shock wave with
r
r
velocity u . An element of real velocity v has the relative velocity q = v u in the coordinate
426
Czech Republic
2006
[Content]
system. The pressure is independent with coordinate system, so the velocity u is equal to
u 30 [13]. According to the Newton second law, the relationship between the collapse angle of
the metal plate and the time can be denoted as

k
k 1
k 1
Ma21 + 1
d 2
1
=
)] exp( / cos 2 ) (12)
m hm u 30
0 D12 (1
[
dt k 1
k +1
max
Ma132 + 1
2
where m is the density of the metal, hm is the thickness of the plate and u 30 is the
particle velocity. The free expansion angle of the detonation products can be obtained from
Prandtl-Mayer law [14], as shown in
k +1
1)
k 1
(13)
Richter et al have deduced that the expanding angle equals the maximum collapse angle
max [16].
The processing of the metal plate collapse causes the displacement dy in the direction y
can be obtained from equation
k
k 1 2 k 1
Ma1 +1
1
2 d
D2 (1
2mhmu30
sin cos = 2
[
)]exp( / cos2 ) (14)
k 1 2 k + 1 0 1
dy
2
max
Ma13 +1
2
Equations (12) and (14) indicate the influence of detonation parameters, material
properties
and
the
thickness
of
cover
to
the
metal
plate
deformation.
Experiments scheme
The covers were attached to the surface of the large apex angle liner and the central line of
the cover is parallel to the trunk of the liner. The way of the covers distributed on the liner is
shown by Fig.6. Fig. 7 is the section view of the shaped charge. Three experiments had been
conducted. The apex angle of each liner experiment was 125, the wall thickness was 3mm,
and the diameter was 50mm. The height of each shaped charge was 55mm. In experiment
No.1the charge density was 1.1g/cm3; in experiment No.2 and No.3 the shaped charge density
was 1.4g/cm3. And the cases of the shaped charges were PVC. The covers were all made by
aluminum LY12 2024 . And the shape of the covers was same. The number of the covers
used in each experiment was four. The thickness of the covers was different. No.1 the
thickness of the covers was 1mm, and No.2 was 2.5mm, and No.3 was 5mm. The recovery
equipment included an iron barrel and leak-tight materials and water and protection plate.
5
427
Czech Republic
2006
[Content]
Fig.7 describes the recovery equipment. In three experiments each shaped charge was
detonated on the center point and the standoff of the each experiment is 50mm, as shown in
Fig. 7. The EFP with fins can be recovered in water.
EFP fins have been presented by picture8. The experimental results show that the liner
element attached to the cover has less deformation than the element directly contacted with the
explosive has. The state of the fins of each EFP changes with the variety of the thickness of the
covers.
Results and discussion

After detonation the element of the liner under the covers forms fins of the EFP and the
element without covers has large deformation close to the center line of the projectile. The
experiment results indicate that the variety of the thickness of the covers leads to the variety of
the fins. The thicker the covers are the more obvious the fins are. Since the experiment No.1,
the charge of the density is smaller than two other experiments, the diameter of the EFP is
bigger than two other EFPs.
High explosive plays an important role in the fields of shaped charge. The problem of
detonation driven liner have been studied by many scientists [15-18]. Actually this problem is
about the planar flyer plate under detonation driving. So the results obtained in section 1 and 2
can indicate the deformation and acceleration of the liner element. The value of displacement
of the metal plate element in the normal direction of the plate surface is noted by y. Equation
(12) indicates the relationship between the collapse angle of the plate and the time of
detonation loading. Equation (13) indicates the influence of the thickness of the cover to the
displacement of plate. Figure 8 shows the red copper with thickness of 3mm attached to
diversity thickness of the aluminum LY12 2024 cover collapse angle at different time.
The coefficients of explosive and cover have been listed in table1. The shock wave
coefficients that deduced by using the explosive and cover parameters and equation (1)-(7) are
shown in table2. Fig. 8 shows that when the thickness of the cover rises, the mental plate
collapse angle reduces. And Fig. 9 displays that the displacement in direction y decreases
when the thickness of the cover increasing.
The differences between shaped charges that are composed by the liner attaching to the
covers and that are composed by the liner without the covers are the mass of the explosive on
unit area and the detonation pressure on each element. The first difference is originated from
the diameter of the shaped charge is consistent and existence of covers on the liner cuts down
the mass of explosive. The second difference generates owing to the attenuation of the
detonation pressure through the covers. Both differences will lead to the element under the
covers loaded less detonation pressure than that directly contacted with explosive. The
deformation degree of the mental element in unit time is determined by the pressure loaded on
the liner. So the difference of detonation pressure loaded on the liners results in the EFPs with
fins. The difference of pressure brought by explosive mass difference can be neglected. And
the difference of covers attached to the liners cause the unlikeness of EFPs with fins.
The thickness of the covers affects the fins of the EFP mainly because the detonation
pressure attenuation of same material is associated with the thickness of the material. From
equation (10), when the thickness of the covers increase, the pressure through the covers is
428
Czech Republic
2006
[Content]
weaken more. The thicker the covers are the deformation degree of the liner element attached
to covers is less and deformation comparison between liner element attached to covers and
that without covers is more obvious. So shaped charges constructed by the same liner but
different thickness of the covers produce EFP with different fins.
Conclusions
We have shown that the thickness of the covers is a factor on the fins of the EFP. The
cover material property is another factor that affects the deformation of the liner. From former
theory analysis, the bigger is the attenuation coefficient, the smaller is detonation pressure
loaded on the liner element and final deformation. We can further perform experiments to
examine the influence of the material to the deformation of EFP fins.
Reference:
[1]
Taylor G I. Analysis of the explosion of a long cylindrical bomb detonated at one

end[A].Scientific papers of G.I.Taylor,Vol[C].Cambridge: Cambridge University
Press,1963:277-286.
[2]
Chou P C,Carleone J, Hirsch E, el al. Improved formulas for velocity, acceleration, and
projection angle of explosively driven liners [J].Propellants, Explosive, Pyrotechnics, 1983, 8:
175-183.
[3]
Shao Binghuang, Zhang Dengxia, Li Guohao. The movement of the metal plate under sliding
detonation [J]. Explosion and Impact, 1986 6 1 24-29.
[4]
M. Defourneaux,Thorie hydrodynamique des charges creuses et delapntration des jets,

MmorialdArtillerie Franaise, 2/70, 1070.
[5]
W. P. Walters, J. A. Zukas, Fundamentals of shaped charges, A Wiley-Inter- science

Publication, New York, 1989.
[6]
K. P. Stanjukovic, etc., Fizika vzryva, Second Edition, Moscow, 1974
[7]
A. Stamatovic, Physics of explosion, Ivexy, Belgrade, 1996
[8]
Shushan Wang, Yanmei Zhang. Initiation of the cover explosive under sliding detonation [J].
Acta Armamentarii, 1999 20 3 233-237.
[9]
Fickett, W., Davis W.C., Detonation, University of California Press, Berkeley,CA, 1979, 14.
[10] Shouzhong Zhang. Explosion and Impact dynamics [M]. Beijing: Ordnance industry publication,
1993.
[11] The compiling group of the Explosion and Explosion Effect, Explosion and Explosion Effect (A)
[M]. Beijing: National defense publication, 1979.
[12] Zheming Zhen, Zhensheng Yang. Explosion machining [M]. Beijing:
publication, 1981, 342-343.
National defense
[13] Chengsen Qin, Yi Liu. The projection angle equation of flyer plate under detonation driving [J].
Explosion and Impact, 2005 25 1 1-5.
[14] Chengwei Sun. Application detonation physics [M]. Beijing: National defense publication,
2000:610-612.
429
Czech Republic
2006
[Content]
[15] Taylor G I. Analysis of the explosion of a long cylindrical bomb detonated at one end [A].
Scientific papers of G. I. Taylor, Vol III[C]. Cambridge; Cambridge University Press, 1963: 277
- 286.
[16] Richter H. On the theory of shaped charges: Motion of thin liners of plastic material on the
surface of a plane explosive (in German) [ R] . Note Technique ISL n6a/ 48 , 1948.
[17] Chou P C, Carleone J, Hirsch E, et al. Improved formulas for velocity, acceleration, and
projection angle of explosively driven liners [J] . Propellants, Explosives, Pyrotechnics, 1983,
[8]175- 183.
[18] Binghuang SHAO , Dengxia ZHANG , Guohao L I , et al. Motion of the flyer plate under
glancing detonation[J ] .Explosion and Shock Waves , 1985 ,5 (3),1-12.
Table1 The known parameters of explosive, cover and metal plate
D1
(mm/s
7.8
a2
3.0
2.785
5.328
30
b2
(g/cm3) km/s
(g/cm3)
1.4
a3
b3
(g/cm3)
1.338
8.93
mm-1
3.91 1.51
0.0583
Table 2 The results of shock wave parameters of the cover
430
p1 GPa
1 g/cm
2 (rad)
2 (rad)
15.47
3.21
0.067
0.658
Czech Republic
2006
[Content]
Figure Captions
Figure 1. D1- detonation velocity; OA-detonation front wave; OC-shock front wave in the
cover; C-intersection point of the shock wave with cover and the metal plate; 2 - angle
between OC and tangential direction surface of the cover; OB- interface between explosive
and inflecting cover surface; 2 - angle between OB and original surface of cover;
CC-original surface of the metal plate; CD- shock wave in metal plate; 3-angle between CC
and the CD; 3- angle between CC and the CE.
Figure 2.
- thickness of the cover; OB-interface of the inflected interface between the
explosive and cover; S-arc length of the plate; dS-infinitesimal arc length; -collapse angle
of the plate; d -infinitesimal collapse angle; R-radius of collapse round; P-pressure on the
metal plate.
Figure 3. O-original point moving with shock wave; y-direction of the normal line of primal
metal plate surface; u-vector velocity of point O; n-normal line of the curved metal plate
surface; v-vector velocity of the metal plate in static ordinate; q-metal particle velocity in
moving coordinate system.
Figure 4 . Figure a is front view of liner; figure b is top view of liner. 1-liner ; 2-cover;3-
trunk 4- central line of cover

Figure 5 .1- initiation point; 2-explosive; 3-liner; 4-cover.
Figure 6 The distance between the shaped charge and the recovery equipment is 50mm.
Figure 7. Picture (a) is experiment result of liner with 1mm thick covers; picture (b) is
experiment result of liner wiht5mm thick covers; picture(c) is experiment results of liner with
2.5mm thick covers. Parts of fins far from the central line of EFP are the deformation results of
the liner elements under covers.
Figure 8. Line1. Calculation result of without cover metal plate; Line 2.Calculation result of
metal plate with 2.5mm thick covers; Line3. Calculation result of metal plate with 3mm thick
covers; Line4 Calculation result of metal plate with 5mm thick coves.
Figure 9. Line1. Calculation result of metal plate without cover; Line2. Calculation result of
metal plate with 2.5mm thick cover; Line3. Calculation result of mental plate with 1mm thick
cover.
431
Czech Republic
2006
[Content]
Figure 1 Sliding detonation loaded on the cover and the metal plate
u11
u0=D1
u13
explosive
cover
u20=D1
o'
u21
B
u22
C
2
3
3 u
31
u30=D1
30 metal plate
C'
E
(31)
D
Figure 2 Collapse process of the metal plate
explosive
D1
o'
u20=D1
cover
C
S
R
dS
d
Figure 3. Acceleration of the metal plate
u
c
n
q
u
432
Czech Republic
2006
[Content]
Figure 4. Large apex angle liner and distribution way of covers
(a)
1
(b)
Figure 5. Section view of the shaped charge.
11
433
Czech Republic
2006
[Content]
Figure 6 Recovery equipment

detonation point
shell
explosive
liner
water
barrel
inside lining
protect plate
434
Czech Republic
2006
[Content]
Figure 7. Recovery EFP under different thickness of covers.
(a)
(b)
(c)
Figure 8. The relationship of the time and bent angle of the red copper attached by
the Covers with different thickness
13
435
Czech Republic
2006
[Content]
Figure 9. Relationship of time and displace towards y of red copper with different thick covers
436
Czech Republic
2006
[Content]
437
Posters
Czech Republic
2006
[Content]
SIMULATION OF COFFEE MILL EXPLOSION ACCIDENT UNDER

ATEX 137 DIRECTIVE
Vladislav Adamik, Bretislav Janovsky
University of Pardubice, Institute of Energetic Materials, Pardubice, Czech Republic
Abstract:
The paper deals with numerical 3D simulation, through the code LS-DYNA, of the
effects caused by coffee powder air explosion on a coffee mill cover. The procedure
applied follows strongly ATEX directives. From the analysis, it is concluded that the
current cover coffee mill design is not appropriate, the cover sealing failure occurs at
time around only 17 ms after the explosion when the loading overpressure was only
10% of the maximum overpressure value. On the base of this analysis, it has been
concluded that the 3D numerical simulations using LS-DYNA have been a very powerful
and useful tool for the design of a new coffee mill covers. However, in more, the
analysis performed illustrates quite general approach that can be used successfully for
many industrial explosion accidents.
Keywords:
1.
ATEX directives, coffee powder air explosion, numerical simulation,

FEM, LS-DYNA
INTRODUCTION
Explosive atmospheres can be caused by flammable gases, mists or vapours or by

combustible dusts. If there is enough of the substance, mixed with air, then all it needs is a
source of ignition to cause an explosion. Explosions can cause loss of life and serious
injuries as well as significant damage. Preventing releases of dangerous substances, which
can create explosive atmospheres, and preventing sources of ignition are two widely used
ways of reducing the risk. Using the correct equipment design can help greatly in this.
Many workplaces may contain, or have activities that produce, explosive or potentially
explosive atmospheres. Examples include places where work activities create or release
flammable gases or vapours, such as vehicle paint spraying, or in workplaces handling fine
organic dusts such as grain flour or wood, etc.
ATEX is the name commonly given to the framework for controlling explosive
atmospheres and the standards of equipment and protective systems used in them. It is based
on the requirements of two following European Directives.
1)
Directive 99/92/EC (also known as ATEX 137 or the 'ATEX Workplace

Directive) on minimum requirements for improving the health and safety
protection of workers potentially at risk from explosive atmospheres.
2)
Directive 94/9/EC (also known as ATEX 95 or the ATEX Equipment

Directive) on the approximation of the laws of Members States concerning
equipment and protective systems intended for use in potentially explosive
atmospheres.
Vladislav Adamk, Betislav Janovsk
451
Czech Republic
2006
[Content]
ATEX 137 Directive coming into force also in Czech Republic. Among others ATEX
137 requires employers to put mitigation measures in place. These measures should be
consistent with the risk assessment and appropriate to the nature of the activity or operation
and include:
reducing the number of employees exposed to the risk;

providing plant that is explosion resistant;
providing explosion suppression or explosion relief equipment;
taking measures to control or minimize the spread of fires or explosions;
providing suitable personal protective equipment.
The contribution concerns the description of LS-DYNA calculations used for design of
an explosion resistant coffee mill.
2.
PROBLEM DESCRIPTION
Potential explosive mixture, coffee powder air, exists in coffee mill industrial
machines. The best mitigation measures against this explosion are measures that contain the
explosion within the machine frame housing so that no explosion effects enter the
surrounding space.
However, usually the frame housing is enough robust but has various manipulation
handling openings covered by housing covers. Generally, these covers are weak parts of a
coffee mill and must be designed to withstand loading caused by possible coffee powder
air explosion.
In this paper, these problems are solved through two consecutive steps:
During the first step, experimental research is conducted to establish the parameters
of the coffee powder air explosion, especially the curve: overpressure respect to the
time.
In the second step, the covers response to this explosion curve is computed. When
the cover cannot contain the explosion inside the machine, a new covers design must
be proposed and verified.
In the following, the explosion curve used in this paper is introduced and a typical
housing cover is described. A typical explosion curve is introduced in fig 1. This curve was
obtained by means of experiments performed in the University of Pardubice, Institute of
Energetic Materials, Pardubice and it was used for the analysis presented.
452
Fig 1.
Czech Republic
2006
[Content]
Overpressure (GPa) versus time (ms)
The housing cover analyzed consists of the cap, two rotation locks, and a hinge
connected, see figs 2 and 3. The lamella is inserted behind the frame housing and the hinge
is connected to the cap through seven rivets. The hinge has a rear inclined part that is
supported by the angling edge (45 deg) of the frame housing.
Fig 2.
Housing cover front view
453
Fig 3.
Czech Republic
2006
[Content]
Housing cover rear view
A detail of the lock is introduced on fig 4. The lock consists of a lamella, nut, shaft and
front covering, see fig 4.
lamella
shaft
nut
covering
Fig 4.
454
The lock rear view
3.
Czech Republic
2006
[Content]
LS-DYNA NUMERICAL SIMULATION
Due to the cover symmetry, only one half of real cover was modeled. For the cover and
lock materials, the Johnson-Cook constitutive model was used. For numerical simulations,
the 3D code LS-DYNA [1] with Lagrangian computational mesh was used.
The cover cap was divided into two parts: loaded part and non-loaded part (outside the
opening), see fig 5. The fig 6 presents the numerical mesh used shell elements of
Belytschko-Tsay type were used (5 integration points across the shell thickness 4 mm).
Similarly, in fig 7, shell elements of the same type for connected hinge are introduced
(thickness 2.5 mm).
In fig 8, solid elements (brick elements) used for the lock nut, shaft, and covering are
illustrated.
In fig 9, shell elements used for the rotation formed lamella (thickness 1 mm) are
introduced. The same shell element type was used as for the cap.
Fig 5.
Cap loaded and non-loaded parts
455
456
Czech Republic
Fig 6.
The cap finite shell elements
Fig 7.
The hinge connected finite shell elements
2006
[Content]
The lock solid elements
Fig 8.
Czech Republic
Fig 9.
2006
[Content]
The lamella shell elements
For the cover construction, two materials were used: hard aluminum cap, connected
hinge, steel lock with lamella. For both materials, the constitutive model of Johnson-Cook
was used:
Y = (A + B n )(1.0 + C ln ( * ))(1.0 T *m )
T* =
(1)
(T Tr )
(2)
(Tmelt Tr )
Where Y is the yielding stress, is the plastic strain, * is the strain rate, T is the current
temperature, Tr is the initial temperature and Tmelt is the melting point.
The equation of state has the form:

Peqs = K ln
0
(3)
where Peqs is the pressure, K is bulk module, is density.

The following parameters were used for the hard aluminium [2]:
A = 360 MPa
Tr = 300 K
B = 317 MPa
Tmelt = 775 K
C = 0.019
0 = 2700 kg/m3
n = 0.289
m = 1.00
K = 60 GPa
and for the steel [2]:
457
A = 200 MPa
Tr = 300 K
B = 184 MPa
Tmelt = 1800 K
C = 0.006
0 = 7870 kg/m3
n = 0.30
m = 0.53
Czech Republic
2006
[Content]
K = 175 GPa
For numerical simulations, it was necessary to respect all possible contacts between
materials.
The cap part outside the housing opening can enter into contact with the outer part of the
frame housing. This contact was simulated as a contact of the cap with a stonewall defined
by the external part of the housing. Similarly, the contact of the connected hinge with the
angling housing edge was simulated through a stonewall.
The hinge connection with the cap through the rivets was simulated by the contact no.6
of the LS-DYNA (tied hinge nodes to the cap surface).
The contact lamella housing was modeled through a stonewall.
The contact covering-cap was simulated through the contact no.2 of LS-DYNA (tied cap
nodes to the covering surface).
4.
RESULTS
The numerical simulation was performed only up to the real process time of 30 ms,
because even at this time the cover deformations were so high that the cover could not be
able to fulfill its function.
The computation was performed using mm/kg/ms unit system, thus the pressure is in
GPa and forces are in kN.
The dynamics of the cover sealing failure was as follows. It is well described by figs 10
to 13 that illustrate displacement isolines in (in mm) z coordinate, i.e. the displacement into
the outer coffee mill space.
Fig 10.
458
The lock solid elements
Fig 11.
The lamella shell elements
Fig 12.
Displacement at time 21 ms
Czech Republic
Fig 13.
2006
[Content]
Displacement at time 30 ms
At first, (figs 10 and 11) the displacement is higher for the upper cap part due to the
lamellas deformation.
However, the connected hinge in the lower cap part is sliding along the angling housing
edge. At time around 17 ms, it is outside the frame housing, the lower cap part is free and is
rotating around the lamellas (figs 12 and 13). The maximum displacement is around 77 mm,
i.e. completely outside the frame housing.
The stresses of the cap are below the yielding stress and their distributions are in figs 14
to 17. Initially, stress concentrations near locks and rivets are evident. Then, the stress is
concentrated naturally near locks.
Fig 14.
Stress at time 7 ms
Fig 15.
Stress at time 14 ms
459
Fig 16.
Czech Republic
Fig 17.
2006
[Content]
As mentioned before, the cap and the hinge did not show plastic strains. On the other
hand, the locks suffered a very small plastic strains, see figs 18 to 21. The maximum plastic
strain at time 30 ms is around 0.0008.
Fig 18.
460
Plastic strains at time 7ms
Fig 19.
Plastic strains at time 14 ms
10
Fig 20.
Czech Republic
Fig 21.
2006
[Content]
As regards the velocities along the coordinate z for individual cover parts, the fig 22
presents their histories: 1 loaded cap, 2 non-loaded cap, 3 hinge 4 rotating lamella.
The hinge has the maximum velocity because it is placed in the lower cover part: up to
10 m/s at time 30 ms. Then, the cap parts follow with the velocity around 4 m/s at time 30
ms. The lamella has the velocity around 2 m/s because it serves as a joint.
Fig 22. Z velocity histories for individual parts:

1 loaded cap, 2 non-loaded cap, 3 hinge, 4 lamella
For cover design, the most important contacts are: hinge-frame housing contact and
lamella-frame housing contact.
In fig 23, contact force time history for the first mentioned contact is presented. This
force continually increases, but at time around 17 ms the hinge is quasi outside the housing
and the force fast decreases. This cover part is a weak point of the cover analyzed.
11
461
Czech Republic
2006
[Content]
In fig 24, contact force time history for the second mentioned contact is presented. This
contact force is a little lowered after time 17 ms because the cover is rotating around the
lamellas. Then, the force is increasing due to higher loading pressure (at this time however
this loading is not realistic, the pressure must be decreased due to the cover sealing failure).
Fig 23. Contact force time history hinge - housing
Fig 24. Contact force time history lamella-frame housing
5.
CONCLUSIONS
The cover dynamic analysis has shown that the current cover coffee mill design was not
appropriate, the cover sealing failure occurred at time around only 17 ms when the loading
overpressure was only 10% of the maximum overpressure value.
For this reason, a new cover design was proposed that would enable to mitigate the
explosion inside the coffee mill: the cover cap made of steel (thickness around 10 mm) and
462
12
Czech Republic
2006
[Content]
provided in the lower part by a rotating welded joint and in the upper part connected with the
robust housing through two welded bolts with female screw (made of steel, too).
The analysis performed illustrates quite general approach that can be used successfully
for many industrial explosion accidents.
The work was performed as a part of the MSM Project no. 0021627501.
REFERENCES
[1]
[2]
13
LS-DYNA USERS MANUAL: Nonlinear Dynamic Analysis of Structures, Version 950,

Livermore Software Technology Corporation, May 1999
J. BUCHAR: private communication, Mendel University Brno, January 2006.
463
Czech Republic
2006
[Content]
EXPLOSION HAZARD OF WATER GEL-LIKE SYSTEMS ON A BASE

OF HIGH EXPLOSIVES
Apolenis A.V., Aleshkina E.I., Annikov V.E., Kozak G.D1.
Mendeleev University of Chemical Technology Miusskaya Sq.9, Moscow, 125190.
Abstract
The main goal of this work is estimation of the real explosion hazard of water gel-like
explosives (WGE) that content high explosives. Conversion high explosives extracted
from ammunition (RDX, HMX and phlegmatized RDX and HMX) according to
disarmament policy are utilized for preparing of water gel-like explosives (WGE) for
different branches of industry. Gel-like base (phlegmatizer) of such systems is water
solution of ammonium and sodium nitrates that thickens by poliacrylamide (1%). Solid
high explosives are mixed with gel-like base. Impact sensitivity of such systems was
investigated, it was found that it was a minimal one at content of gel-like base Cg=2.5
5%. It could be seemed that and explosion hazard of gel-like explosives was a lowest at
this content of Cg. To check this assumption such properties of explosives as abilities to
burning, to detonation, and to transition of deflagration to detonation was investigated
and analyzed in a wide area of phlegmatizer content. It was shown that real explosion
hazard were connected and controlled by these properties, and minimal impact
sensitivity at content of gel-like base Cg=2.5-5% did not reflect the real explosion
hazard and connected with enlargement (gluing) of solid particles of high explosives at
small gel-like base content.
Keywords: sensitivity, detonation, water-gel, phlegmatizer, sensitizer, burning, high
explosive.
1.
INTRODUCTION
Conversion high explosives (sensitizer) extracted from ammunition (RDX, HMX and
manufactured phlegmatized RDX and HMX) according to disarmament policy are utilized
for preparing of water gel-like explosives (WGE) for different branches of industry. Gel-like
base (phlegmatizer) of such systems is water solution of ammonium and sodium nitrates that
thickens by poliacrylamide (1%). Solid high explosives are mixed with gel-like base. The
role of water is polyfunctional one: (1) it creates consistence that is convenient for loading of
explosives into a bore-hole, (2) it permits to obtain a solution of sodium and ammonium
nitrates that are able to react in detonation wave in such state most efficiently [1]. At last (3)
water is very effective and inexpensive phlegmatizer.
The main goal of this work is estimation of real explosion hazard of water gel-like
explosives that content high explosives. Estimation of the real explosion hazard of such
systems is important task for users. As usual for solution of the problem it is not enough to
investigate only sensitivity to impact, and complex method including estimation of ability to
burning, detonability and tendency deflagration to detonation transition (DDT) must be
applied.
1
KozakGD@nm.ru
464
2.
Czech Republic
2006
[Content]
EXPERIMENTAL.
2.1
Preparation of WGE.
Content of gel-like phlegmatizer is presented in Table 1. Micro photographs of high

explosives samples are printed in Fig 2-5, particle sizes of them are written in signatures to
figures. Poliacrylamide (1%) was mixed with powdered ammonium and sodium nitrates, and
this mixture was added to water. Gel-like base was ready to use, when nitrates dissolved and
poliacrylamide swelled. The gel-like phlegmatizer constituted a fluid liquid, but it was more
thick and viscous system than melted paraffin or wax. The requisite quantity of powdered
high explosives was mixed with sample of gel-like phlegmatizer up to homogeneous state of
WGE.
Table 1. Composition of gel-like plegmatizer
Substance
Content, %
Ammonium nitrate
50
Sodium nitrate
20
Water
30
Polyacrilamide
1*
Potassium phosphate
0,1*
Density, g/cm3
1,41
* in excess of 100%
2.2
Impact sensitivity.
Investigation of impact sensitivity was carried out according the sensitivity test
method[2,3] based on the State Standards GOST 4545-80, and GOST 4545-88 that was
officially approved in Russia. Tools N1 application (Fig.1) with the sample of WGE placed
on anvil of vertical fallhammer machine K-44-II. Falling weight was M = 10 kg, its height of
falling down was H = 25 cm, the mass of the sample was m = 0.05 g. The frequency of
explosions is estimated in 25 runs.
Fig 1.
2
Tools N1: 1 a pallet; 2 a muff; 3 a sample; 4 rollers.
Alexei Apolenis, Ekaterina Aleshkina, Vladimir Annikov, Georgii Kozak
465
466
Czech Republic
2006
[Content]
Fig 2.
The micro photograph of RDX particles

(particle sizes is 50 mcm). The diameter
of viewing field is D = 3.2 mm.
Fig 3.
The micro photograph of HMX

particles (average particle sizes is 250
mcm).
Fig 4.
The micro photograph of phlegmatized

RDX particles (average particle sizes is
250 300 mcm).
Fig 5.
The micro photograph of mixture of

phlegmatized RDX particles with
aluminum (average particle sizes is 250
300 mcm).
2.3
Czech Republic
2006
[Content]
Detonability.
Critical diameter (df) was measured by means of method go no go in glass cylinder

tubes at powerful initiation [4-6]. Detonation, failure of detonation or damping of detonation
was fixed by means of witness-plates on that the charge was placed.
2.4
Burning.
Investigation of water influence on burning of high explosives was begun in work [7],
some later one of us with co-authors investigated burning of WGE in bomb of constant
pressure in a large scale of pressure [8]. This data were utilized at discussion of explosion
hazard of WGE.
2.5
DDT.
As to DDT, the same situation took place: influence of phlegmatizer content (paraffin)
on distance of DDT of RDX, and HMX was thoroughly investigated in one of woks of the
laboratory of B.N. Kondrikov in Mendeleev University [9] and these data was successfully
used in work [10] at examination of explosion hazard questions in industry. DDT distance was
measured in steel closed thick-walled (=13 mm) tubes, canal of them (d=10 mm) was
wholly filled with a sample of explosive. Ignition was carried out from one of the ends of
tube by means of a sample of pyrotechnic composition. When DDT took place the tube
destroyed starting with some distance from igniter. This distance was distance of DDT.
3.
EXPERIMENTAL RESULTS.
The results, that were obtained at investigation of impact sensitivity of WGE on a base
of RDX , are presented in Fig. 6.
100
f, %
90
80
70
60
50
40
30
20
10
0
Fig 6.
10
20
30
40
Cg, %
50
60
70
80
Frequency of explosions of RDX. vs. contents of gel.
The frequency of explosion of net powdered RDX is f = 64%, that corresponds to

requirement of fallhammer machine tare. Introduction of gel-like phlegmatizer Cg=2-5%
4
467
Czech Republic
2006
[Content]
reduces it to f= ~10%. Augmenting of Cg leads to arising of frequency of explosion. When

the Cg reaches to Cg>15 % frequency of explosion reduces. The result was checked in three
series of tests (25 runs were carried out in which of them). The similar picture took place if
WGE was sensibilized by HMX or phlegmatized RDX and HMX.
Detonability of WGE versus gel-like phlegmatizer content is shown in Fig 7 and 8 One
can see that df of compositions was approximately constant one at low content of gel-like
phlegmatizer, WGE on a base HMX had df <4.4 mm at Cg=5%, density =1.2 g/cm3 (Fig. 8).
Strong augmenting of df begins at much more Cg (Cg=40%).
Burning low (burning rate versus pressure u(p)) of compositions WGE on a base of
manufactured phlegmatized RDX are shown in Fig.8. There is not sufficient decrease of
burning rate (Fig. 9) of high explosive and augmenting of minimal pressure at that it begins
at small (Cg=2.5-5%) content of gel-like phlegmatizer [8]. It takes place only at Cg30%.
Dependence the distance (L) of DDT of RDX vs. content of phlegmatizer [9] is shown in
Fig. 10.
20
df , mm
18
16
14
12
10
8
6
4
2
0
20
40
60
80
Cg, %
Fig 7.
Dependence of a critical diameter detonation of manufactured

phlegmatized RDX from the contents of gel-like phlegmatizer. Solid points
correspond to detonation of the charges, open circles indicate failures of
detonation, circles with line indicate detonation extinguishing.
It is very slow enlargement of L at content of phlegmatizer C= 1-3%. Rapid growth of L

begins at C>5%. In spite of that the runs in work [9] were carried out with paraffin, all
reasons was to apologize that character of dependence L(C) would not change at substitution
of it on gel-like phlegmatizer.
468
Czech Republic
2006
[Content]
df , mm
20
18
16
14
12
10
8
6
4
2
10
20
30
40
50
60
70
80
Cg, %
Fig 8.
Dependence of a critical diameter detonation of HMX from the contents of

gel-like phlegmatizer.
U, mm / s
30
25
20
15
10
5
P, MP
0
0
Fig 9.
10
20
30
Dependence of a burning rate versus pressure. 1. manufactured

phlegmatized RDX; 2. manufactured phlegmatized RDX with 50 % of
gel-like phlegmatizer.
469
Czech Republic
2006
The distance of DDT (L), s

25
[Content]
explosive
TNT
1
2
3
4
20
TNPh
TNR
15
Hexyl
TNB
Tetryl
DINA
PENT, RDX,
HMX
10
The contents of paraffin, %
Fig 10. The distance (L) of DDT vs. content of phlegmatizer (paraffin) (1 HMX,
2 RDX, 3 manufactured phlegmatized RDX, 4 - manufactured
phlegmatized HMX). Names of explosives, to that correspond the distance
of DDT of phlegmatized composition, are written near right axiss (TNPh
pycric acid, TNR threenitroresorcinol, Hexyl hexanitridiphenylamine,
TNB threenitrobenzene).
4.
DISCUSSION.
WGE are the systems that fill halfway place between solid and liquid explosives. The
explosion hazard of liquid and solid explosives is estimated at experimental and calculating
investigation of some parameters. It is important to know complex of the parameters: ability
to burning without access of air, detonability, and ability to DDT. Such approach is
necessary condition of industrial safety in the branch of manufacture and treatment of
explosives and dangerous goods. Applying of such approach was demonstrated on example
of organic peroxides in works [11-13].
As one can see, the data of fallhammer machine test does not attract real explosion
hazard of the composition of WGE. It would be seen the lowest one at content of gel-like
phlegmatizer Cg=2.5-5% according to this test (Fig. 6), but investigations of abilities to
detonation, to burning, and to DDT (Fig. 10) contradict to this estimation. All of these
abilities, that determine the real explosion danger, are lower at much more content of gel-like
phlegmatizer. As it was shown at microscope investigation (Fig. 11), the reason of lower
impact sensitivity of WGE at Cg=2.5-5% was in gluing of particles i.e. in enlargement of its
size. As it was described in work [2] impact sensitivity of powdery systems (WGE at
Cg=2.5-5% was powdery, that was slightly wetted by gel-like phlegmatizer) diminished.
470
Czech Republic
2006
[Content]
Growth of Cg changes WGE into gel-like system that can flow, heat, and ignite in tough
conditions of fallhammer machine test, and frequency of explosion enlarges.
In real conditions of WGE employment impact on it leads to free flow, that is not
accompanied by heating. Much more dangerous situation is fortuitous inflammation and
DDT of explosive that occurs much more easy in a case powders than in a case of real WGE
at greater content of Cg.
Fig 11. The micro photograph of RDX

(particle sizes is 50 mcm) mixture
with 5% of gel-like phlegmatizer.
One can compare it with Fig. 2.
5.
CONCLUSION.
The attempts to use WGE at gel-like phlegmatizer content Cg=2.5-5%, that are based
on the results only of fallhammer machine test, can lead to dangerous situations and to
incidents. They deprive WGE theirs main advantage over powder explosive: reduced
inflammability, and lower ability to DDT.
ACKNOWLEDGEMENTS
Authors are grateful to Foundation of Basic Research of RF, that supported the
investigation through grant 04-03-32713.
471
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
472
ANNIKOV V.E.,.KONDRIKOV B.N,.SMAGIN N.P,.PARFENOV L.P, SHABALINA L.M.:

The study of detonation of aluminum containing watergel explosives, Voprosy teorii
vzryvchatykh vestchestv (Problems of theory of explosives), Mendeleev Institute of Chemical
Technology, Moscow, 1974, v. 83, 147 153.
AFANASEV G.T., BOBOLEV V.K.: Initiation of the solid explosives by impact (Russ.),
Moscow, Nauka, 1968, 123 pp.
DUBOVIK A.V., BOBOLEV V.K.: Sensitivity of liquid explosive to impact (Russ.), Moscow,
Nauka, 1978, 232 pp.
KOZAK G.D., KONDRIKOV B.N., STARSHINOV A.V.: Critical conditions of low and high
velocity detonation regimes in liquid nitrocompounds, Proceedings 11th Int. Detonation
Symposium, Snowmass, Colorado, August 31-September 4, 1998, Bookcomp, Ampersand,
2000, p.81-86.
KOZAK G. D.: Factors augmenting the detonability of energetic materials, Propellants,
Explosives, Pyrotechnics, 30, N 4, 2005 p. 291-297.
KOZAK G.D., RAIKOVA V.M., ALESHKINA E.I.: Critical conditions of propagation and
photorecording of detonation processes. Text-book, Mendeleev University of Chem.
Technology, Moscow, 2005, 64 pp.
KONDRIKOV B.N., ANNIKOV V.E., POLYAKOVA N.A.: Effect of water, ammonium
carbonate and some other additives on the explosives combustion, Comb., Expl. & Shock
Waves, 1969, v.5, 1, 60-67.
EGORSHEV V.YU, KONDRIKOV B.N, YAKOVLEVA O.I.: Combustion of waterimpregnated explosive compounds, Comb., Expl. & Shock Waves, 1991, v. 27, 5, 565-572
SVIRIDOV E.M.: The study of burning of aromatic nitrocompounds and deflagration to
detonation transition, Cand. Sci. (Ph. D.) Thesis Part 2, Mendeleev University of Chemical
Technology, Moscow, 1970, p 40.
AKININ N.I.: Lowering of explosion hazard of power energetic materials. M. Mendeleev
University of Chemical Technology, Moscow, 2005, p. 60.
AKININ N.I., KOZAK G.D., RAIKOVA V.M., ARININA S.V.: The Parameters estimating an
explosion hazard of organic peroxides. Trans. of XI International conference of organic and
organoelement peroxides chemistry . IKhF RAN, 2003, p. 220-221.
KOZAK G.D., AKININ N.I., RAIKOVA V.M., ARININA S.V.: Explosion hazard of some
organic peroxides. Proc. of The 6-th seminar New trends in research of energetic materials,
Univ. Pardubice, CSR, 2003, pp.173-181.
AKININ N.I.,. ARININA S.V, KOZAK G.D. PONOMAREV I.N.: The explosion parameters
of benzoyl and cyclohexanone peroxides. The 7-th seminar new trends in research of energetic
materials, Univ. Pardubice, CSR, 2004, pp.409-418.
Czech Republic
2006
[Content]
HEAT OF MELTING OF NITROCOMPOUNDS MEASURING BY

MEANS OF THE DIFFERENTIAL SCANNING CALORIMETRY
METHOD
S.V. Arinina, G.D. Kozak1

Mendeleev University of Chemical Technology 125047, Miusskaja sq. 9, Moscow, Russia.
125047, Miusskaja sq. 9, Moscow, Russia.
Abstract
The differential scanning calorimetry method was applied last year to compare
explosion hazard of nitrocompounds and untraditional explosives peroxides. In this
work the literature data for c-nitrocompounds heat of melting (Hmelt) were compiled
and analysed. It was found that the rule: (H/T)melt=54.412.6 kJ/mol K is kept for cnitrocompounls, however, there is tendency to correlation (H/T)melt=f(M), where M is
molecular mass. Brilliant linear correlation of these values was found for nnitrocompounds (dimethylnitranine, RDX, and HMX) and o-nitrocompounds at M=75
110. The differential scanning calorimetry method was applied for investigation of heat
of melting of TNT, PETN, and ethylenedinitramine (EDN). Our values of HmeltTNT
coincided with literature data. Firstly received data
HmeltPETNand HmeltEDN
supplemented linear correlation of literature data (H/T)melt=f(M), for nitroesters and
nitramines, correspondingly.
Keywords: differential scanning calorimetry, n-nitrocompounds, o-nitrocompounds, cnitrocompounds, heat of melting.
1.
INTRODUCTION.
The method of differential scanning calorimetry (DSC) was applied in work [1], that was
presented at previous Seminar, to study exo- and endothermic effects, that took place while
heating of benzoil peroxide and cumene hydroperoxide (CH) in comparison with traditional
explosives TNT and PETN and insensitive explosive NTO.
The DSC method [2] is based on simultaneous heating (or cooling) of sample
investigating in comparison with standard one. The given speed was regulated by means of
computer of the arrangement, measurement of the compensating heat flux was supported
temperature of a sample within the limits of the given program. The DSC experimental
curves are dependencies of a heat flux (millijoules per second) versus temperature (in
isothermal runs -versus time). They permit to register the temperature and the heat of
melting, the temperature of very decomposition starting with great accuracy. Unfortunately
the heat of decomposition could not be measured with sufficient accuracy because of heat
dissipation at gas formation. However the heat of fusion of samples this method registers to
a high degree of accuracy. By the way this method is suitable to check of solubility one
substance in another by means of measuring of melting point of a mixture.
KozakGD@nm.ru
Svetlana Arinina, Georgii Kozak
473
2.
Czech Republic
2006
[Content]
EXPERIMENTAL.
In this work all experiment were carried out by means of Mettler Toledo DSC 30
calorimeter. Mass of a sample was 1-1.5 mg.
TNT was chosen as there were sure calorimetric data heat of melting (HTNTmelt) in
literature [3,4]. Melting enthalpy of TNT, obtained in our experiment, was coincided precisely
with literature DSC data [3] HTNTmelt = 103 J/g. This coinciding gave an opportunity to
judge about reliability of our measurements. According to earlier sources [4] this value was
92.3 J/g but it was rather close to enthalpy obtained by means of DSC method.
Two another objects of investigation were nitroester PETN and nitramine
ethylenediamine dinitrate (EDNA). The samples of explosives were refined by means of
recrystallization. Melting points of them were measured during DSC experiments.
3.
EXPERIMENTAL RESULTS.
Experimental calorimetric curves for TNT, PETN and EDNA are presented in Fig 1 ac.
There is considerable interval of temperature between endothermic effect of melting and
exothermal effect of decomposition in the case of TNT. This interval is smaller in the case of
PETN, and it is absent in the case of EDNA, that begins to decompose at once after melting.
Experimental calorimetric curves for mixture of EDNA (CEDNA=12.5 %) and TNT is
shown in Fig 1 d. One can see two endothermic effects of melting: the first one corresponds
to melting of TNT, i.e. melting point coincides with melting point of net TNT (Fig 1a).
Second one is closed to melting point of EDNA. Contrary to TNT mixtures with DNT or
PETN [4,5] melting point of TNT mixture with EDNA does not change in comparison with
TNT. In other words, solution of EDNA in TNT does not take place. Properties of this
mixture were investigated in connection with works [5-10], that dealt with investigation of
explosive systems capable to spin pulsating detonation. The heat of explosion of such
systems was Qv=5.3-5.4 MJ/kg, and calculated detonation velocity was D= 7.37.4 km/s.
Calculated Qv and D of the mixture TNT/EDNA 87.5/12.5 had the same values, and spin
pulsating detonation of cast charges of it was observed in experiments. The photograph of
witness-plate is presented in Fig. 2.
Results of measuring of the enthalpy of melting of nitrocompounds and data that were
found in literature are collected in Table 1.
4.
DISCUSSION
The analysis of the references dealing with melting enthalpy calorimetric data
measurements [3,4,11-13] and our data are shown in Fig 35.
It is known [14] that ratio (H/T)melt is the constant value (Rule of Valden), it is
(H/T)melt=54.412.6 kJ/mol K for organic substances. The values (H/T)melt for cnitrocompounds are plotted versus molecular mass (M) in Fig. 3. One can see that this rule is
kept in common for them at M= 60-290, all of the points settle between dash lines that are
the frame 12.6 kJ/mol K , however the tendency to growth of (H/T)melt with molecular
mass arising takes place.
474
Heat of Melting of Nitrocompounds Measuring by Means of the Differential . . .
10
dH/dt, mW
Czech Republic
10
2006
[Content]
dH/dt, mW
0
0
-5
-5
-10
T,C
-10
50
100
150
200
250
300
350
T,C
400
-15
50
12
15 dH/dt, mW
100
150
200
250
dH/dt, mW
300
10
10
6
4
-5
0
-10
-2
-15
-4
-6
-20
o
-25
60
T, C
80
100
120
140
160
180
200
220
240
-8
50
T, C
100
150
200
250
Fig. 1. Experimental calorimetric curves. The velocity of heating is v=10 degrees/

min. The mass of sample m=1.1 1.54 mg, a TNT 1, b PETN, c
EDNA, d EDNA/TNT 12.5/87.5.
Fig. 2. Photograph of the copper witness-plate after spin pulsating detonation of

EDNA/TNT 12.5/87.5 cast charge. Diameter of the charge is d=12.3 mm,
density is =1.55 g/cm3, frequency of pulsation is f=1.2 MHz.
475
Czech Republic
2006
[Content]
Table 1. Enthalpy of melting of nitrompounds.
Hmelt
Explosive
M,
Tmelt,
g/mol
J/g
(H/T)melt,
Refe-
J/mol
J/mol K
rence
Nitromethane
61.04
244.8
158.9
9700
39.6
[11]
Tetriyl
287.1
402.6
79.9
22930
56.9
[12]
Pycric
229
394
74.7
17100
43.4
[11]
182
343.3
110.5
20120
58.6
[11]
352
103
23490
66.7
[11]
354
92.2
20920
59.1
[4]
354
103
23430
66.2
Our
acid
4
2,4 DNT
5
TNT
227
data
Nitrobebzene
123.1
278.8
99.2
12200
43.8
[11]
Ethylnitrate
91.07
178.6
93.66
8529
47.76
[15]
Izo-Propyl-
105.09
190.9
96.1
10099
52.9
[15]
nitrate
9
Methylnitrate
77.04
190.2
106.9
8235
43.3
[11]
10
PETN
316
414
171
54036
130.5
Our
data
11
RDX
222
478.2
169.6
37660
78.8
[11]
12
HMX
296.2
553.2
235.9
69873
126
[13]
13
Dimethyl-
90.08
331
43.29
3899
11.78
[15]
150.09
449
190
28512
63.5
Our
nitramine
14
EDNA
data
15
476
DINA
240.13
325.5
208.2
49995
153.6
[11]
Czech Republic
2006
[Content]
100
(H/T)melt, J/mol K
90
80
70
60
50
40
30
20
10
M, g/mol
0
0
50
100
150
200
250
300
350
Fig. 3. The values (H/T)melt of c-nitrocompounds. The numbers of points

corresponds to numbers of explosives in Table 1.
Correlation between (H/T)melt and M in the case of nitramines: dimethylnitramine
(DMNA) [15], RDX [11], HMX [13], and for EDNA (data of this work) is demonstrated in Fig 4
. Correlative equation is:
(
H/T)melt= 27.9+0.52 M J/mol K.
200
180
(1).
(H/T)melt, J/mol K
15
160
140
120
12
100
80
14
11
60
40
20
0
50
13
100
M, g/mol
150
200
250
300
350
Fig. 4. The values (H/T)melt of nitramines.

5
477
Czech Republic
2006
[Content]
DINA (point 15), that is nitramine and nitroester at the same time, does not conform
to correlative dependence (1).
Correlation between (H/T)melt and M in the case of nitroesters : methylnitrate (MN) [11],
ethylnitrate (EN) [15], izo-propylnitrate (izo-PN) [15] and PETN (data of this work) is shown
in Fig. 5. Correlation is brilliant one in spite of M of PETN in 36 times greater than M of,
MN, EN, and izo-PN. Correlative equation is:
(
H/T)melt = 14.6+0.37 M J/mol K
200
(2).
(H/T)melt, J/mol K
180
160
15
140
10
120
100
80
60
8
9
40
7
M, g/mol
20
0
50
100
150
200
250
300
350
400
Fig. 5. The values (H/T)melt of nitroesters.

The rule (H/T)melt=54.412.6 kJ/mol K is kept for nitroesters that have M=75110
g/mol, and PETN does not follow to this rule.
Dina that has two groups O-NO2 and one group N-NO2, as and in case of nitramines,
does not conform to correlative dependence (2).
5.
CONCLUSION.
The correlation between ratio (H/T)melt and molecular mass of nitrocompounds were
found on a base of it measuring by means of differential scanning calorimetry. If in the case
of c-nitrocompounds it is only a tendency to correlation it is quit definite one in a case of
nitramines and it is even greater definite one in a case of nitroesters. It was shown by means
of analyzing of differential calorimetric curves, that ethylenedinitramine (CEDNA= 12.5%) is
insoluble in TNT, spin pulsating detonation of this mixture was observed experimentally, the
heat of explosion of such system was Qv=5.3. MJ/kg, and calculated detonation velocity was
478
Czech Republic
2006
[Content]
D=7.4 km/s. This result justifies the conclusion of works [5,8-10] in that it was shown:
calculated detonation parameters of all of the systems, that were capable to this process, had
such values.
ACKNOWLEDGEMENTS
investigation through grant 04-03-32713. Our students: I.G. Chumakova and T.V.
Masterkova took part in carrying out of some experiments.
REFERENCES
[1]
ARININA S.V., KOZAK G.D.: Comparative explosion hazard of peroxides and of

nitrocompounds. Proceedings of the 8th seminar New trends in research of energetic
materials, Univ. Pardubice, Czech. Republic, April, 2005, p. 412-428.
[2] BERSHTAIN V.A..,EGOROV V.M.: Differential Scanning Calorimetry in polymer

physical chemistry. Leningrad Chemistry 1990.
[3] ACREE W. E. Jr. Termochim Acta, 1991,37. (Internet. Data base of US National
Institute of Science and Technology. (NIST)).
[4] ORLOVA E.YU.: Chemistry and technology of high explosives. M.,Chemistry, 1973.
p. 687.
[5] LIN ZHOU: Surface phenomena at detonation of condensed explosives. Cand. Sci. (Ph. D.)
[6]
[7]
[8]
[9]
Thesis. Mendeleev University of Chemical Technology, Moscow, 2000.

KOZAK G.D., KONDRIKOV B.N., OBLOMSKY V.B.: Spin detonation in solid substances,
Comb., Expl. & Shock Waves, 1989, v.25, 4, 88-93.
OBLOMSKY V.B.: Non-ideal detonation of homogeneous and porous explosives. Cand. Sci.
(Ph. D.) Thesis. Mendeleev University of Chemical Technology, Moscow, 1988.
KOZAK G.D., LIN Z.: The condensed explosive systems capable of propagating detonation in
the spin-pulsating regime. New trends in research of energetic materials, proceedings of the 7th
seminar Univ. Pardubice, Czech. Republic, April 20-22, 2004, p. 516-523.
KOZAK G. D., LIN Z.: Spin pulsating detonation of cast charges. Trans. of the conf.
(Int). The theory and practice of energetic materials. V. 5, Science press. Beijing-New
York. 2003, pp.404-407.
[10] KOZAK, N.V., KOZAK G.D., LIN Z.: The spin-pulsating regime of detonation in solid and
liquid explosives. Proceedings of the 8th seminar New trends in research of energetic
materials, Univ. Pardubice, Czech. Republic, April, 2005, p.600-605.
[11] DOMALSKI E.S. and HEARING E.D. I. Phys. Chem. Ref. Data, 1996. (Internet.
Data base of US National Institute of Science and Technology. (NIST)).

[12] KRIEN G.,LICHT H.H., ZIERATH I.: Thermochemikal Investigation of nitramines.
Termochim Acta 1973, 6, 465-472. . (Internet. Data base of US National Institute of

Science and Technology. (NIST)).
[13] Maximov YU. YA.: Boiling points and enthalpy of boiling of liquid RDX and HMX, J.
Phys. Chem. (Rus.), 1992, 66 (2), 540-542. . (Internet. Data base of US National
Institute of Science and Technology. (NIST)).
[14] VIKTOROV M.M.: The methods of calculating of physic-chemical values and applied
calculations. (Rus.)Leningrad, Khimiya, 1977, 360 p.
[15] STAL D., WESTRAM E., ZINKE H.: Khimicheskaya termodinamika organicheskikh
soedinenyi (transl. into Rus.), M., Mir, 1971, 312 p.
7
479
Czech Republic
2006
[Content]
CHARACTERIZATION AND SOME PROPERTIES OF

1-NITROAMIDINO-3-AMINO-5-NITRIMINO-1,2,4-TRIAZOLE
Alexander M. Astachov, Vitaliy A. Revenko,
Ludmila A. Kruglyakova, Andrew A. Nefedov,
and Eduard S. Buka
Siberian State Technological University,
Prosp. Mira 82, 660049 Krasnoyarsk, Russia
Abstract:
The 1-nitroamidino-3-amino-5-nitrimino-1,2,4-triazole (ANRTZNQ) structure was
investigated with the help of quantum chemistry calculations. Kinetics of a thermal
decomposition was investigated under isothermal conditions by manometric method in
solid phase. The decomposition products were investigated by mass-spectrometric
method. The sensitivity to explosion by heat was determined as the temperature of flash
and time to explosion delay. The impact sensitivity of ANRTZNQ was investigated with
the help of fallhammer test. Energetic and detonation parameters of ANRTZNQ were
estimated using thermodynamic method with BKW equation of state and by simple
correlation methods.
Keywords: high explosives, nitramines, nitrimines, 1,2,4-triazoles, thermal
decomposition, impact sensitivity
1.
INTRODUCTION
Nitroguanidines and 1,2,4-triazoles are of interest as energetic compounds [13]. We

assumed, that joining nitroguanyl fragment with 1,2,4-triazole cycle in one molecule would
lead to new energetic compounds, which would possess positive properties of both classes of
compounds thus combining high density and thermal stability with low impact sensitivity.
1-Nitroamidino-3-amino-5-nitrimino-1,2,4-triazole (ANRTZNQ) is a high explosive
which we first synthesized in 1999 [4] conducting a search of new insensitive energetic
compounds in a number of derivatives of nitroguanidine.
H2N
NH
NNO2
ANRTZNQ
N N
NNO2
H2N
The data on study of ANRTZNQ properties: research of kinetics and thermal

decomposition mechanism, impact sensitivity (with Russian apparatus) and an estimation of
the compound energetic characteristics are submitted in the article.
2.
SYNTHESIS
ANRTZNQ can be easily and with a high yield synthesized from the 3,5-diamino-1,2,4triazole and 2-methyl-1-nitroisothiourea [4, 5]:
480
Characterization and some Properties of . . .

N
H2N
Czech Republic
2006
[Content]
NH2 HNO3
N N
H2 S
NNO2
95%
O4
H2N
H2N
HNO3
SO 4
NNO2
H2N
N
N NH
NH2
CH3S
NH2
H2N
/H 2
O3
HN
NH2
NNO2
N N
NNO2
~100%
H2N
ANRTZNQ
N N
CH3SH
NH
NNO2
91%
H2N
We have earlier discussed some questions of reaction ability of ANRTZNQ and

problems connected with its structure [4, 6].
3.
CALCULATED STRUCTURE, UV AND FTIR SPECTROSCOPY
ANRTZNQ is practically insoluble in water and usual organic solvents. It does not allow
to grow good crystals for X-ray analysis. Attempts to grow monocrystal samples of
necessary quality by slow crystallization from solutions of a nitric acid have not resulted to
success.
Therefore we investigated the structure of ANRTZNQ only by calculation methods of
quantum chemistry. In the article the results ab initio HF/6-31G calculations are presented.
The GAMESS package was chosen as the calculation tool. Calculated molecular structure is
shown in Figures 1, bond lengths and angles are represented in Table 1. The calculated
structure of ANRTZNQ corresponds to the structure of other nitrimines, experimentally
studied earlier nitroguanidine derivatives and nitrimino-1,2,4-triazoles [7-9]. As in latter, the
nitrimine fragments and 1,2,4-triazole cycle are plane. Because of delocalization of electron density bond lengths CN, NN and NO have intermediate values between the
values characteristic for corresponding single and double bonds (Table 1). Like other
nitrimines, the planar conformation of ANRTZNQ is stabilized by intramolecular hydrogen
bonds (Figure 1, Table 2).
O1
O2
H5
H1
N4
N3
H4
N6
C2
N5
C1
N2
N1
H3
N9
N7
C3
N8
H2
O3
O4
Fig 1.
Ab initio (6-31G basis set) calculated molecular structure of ANRTZNQ;

dashed line indicate intramolecular hydrogen bond.
Alexander Astachov, Vitaliy Revenko, Ludmila Kruglyakova, . . .
481
Czech Republic
2006
[Content]
Table 1. Calculated bond lengths () and angles () for ANRTZNQ

N1N2
1.4207
C2N1
1.3577
N6O1
1.2542
N2C1
1.2887
N5N6
1.3570
N6O2
1.2086
C1N3
1.3416
N1C3
1.4012
N8O3
1.2543
C1N4
1.3825
C3N7
1.3080
N8O4
1.2103
N4C2
1.3516
N7N8
1.3603
C2N5
1.3133
C3N9
1.3216
N1N2C1
104.1
N4C2N5
132.0
C3N7N8
120.4
N2C1N3
125.9
C2N5N6
119.3
N5N6O1
120.4
N2C1N4
111.7
C2N1C3
129.3
N5N6O2
116.1
N3C1N4
122.4
N2N1C3
119.9
O1N6O2
123.5
C1N4C2
108.4
N1C3N9
116.7
N7N8O3
121.7
N4C2N1
105.0
N1C3N7
113.1
N7N8O4
115.7
C2N1N2
110.8
N7C3N9
130.2
O3N8O4
122.6
Table 2. Intramolecular hydrogen-bonding geometry (,) for ANRTZNQ

DH...A
DH
H...A
D...A
DH...A
N4H1...O1
0.995
1.998
2.5774
114.7
N9H2...O3
0.995
1.920
2.5946
122.5
N9H3...N5
0.998
1.945
2.7023
130.6
More detailed calculated structure data of ANRTZNQ (with HF/6-31G and other levels)
will be considered in the other article.
The UV absorption spectra (Figure 2) were recorded on the Shimadzu UV-1601
spectrophotometer. Because of low solubility of ANRTZNQ the maximal concentration of
solution, with which it was possible to write down a spectrum, makes only 4105 M. The
UV spectrum of ANRTZNQ shows an absorption at wavelengths max1 = 213.5 nm, max2 =
271 nm, and max3 = 327 nm. The potassium salt of ANRTZNQ, which is a bit more soluble
in water, shows completely similar peaks. It proves that ANRTZNQ exists in a water
solution as an anion. The absorbtion peak at wavelengths 271 nm is associated with n *
transition in a nitrimine fragment. This is typical for nitroguanidine derivatives, they usually
have intensive maximum at wavelengths 260290 nm, for example, in a nitroguanidine max2
= 265 nm (Figure 2).
482
Czech Republic
2006
[Content]
NQ
A
1.4
Fig 2.
A N R T Z NQ _ P S
1.2
1
0.8
0.6
A NR T Z NQ
0.4
0.2
0
190
210
230
250
270
290
310
330
350
370
390
UV spectrums of
ANRTZNQ (C=4105 M),
its potassium salt
(C=5105 M) and NQ
(C=104 M) in water
solution (thickness of the
UV cell is 1cm).
nm
The infrared spectra (Figure 3) were obtained on the Nicolet IMPACT-400D FTIR
spectrometer with the IR microscope SpectraTech INSPECT IR+. Table 3 summarizes
tentative assignments for some frequencies by analogy with the assignments for
nitroguanidine [10, 11].
1246
0.75
0.70
0.65
0.60
1316
1432
877
0.10
729
682
0.15
774
995
0.20
949
1069
0.25
1474
1660
3432
0.30
1586
3313
0.35
3163
0.40
1627
3370
1687
0.45
3266
Absorbance
0.50
1497
0.55
0.05
0.00
3500
3000
2500
2000
1500
1000
Wavenumbers (cm-1)
Fig 3.
Table 3.
The FTIR spectrum of ANRTZNQ.
Some observed IR frequencies of ANRTZNQ and their tentative assignments

Wavenumber, cm1
Assignment
3432w, 3370m , 3317w, 3266m, 3163w
as and s(NH2), (NH)
1687m, 1660w
(NH2)
1627m, 1586w
(C=N)
1432m , 1246vs
as and s(NO2) nitroamidine
1497m or 1474m, 1316m
as and s(NO2) nitrimino-1,2,4-triazole
1069w, 1036w
(N-C-N) or rock(NH2)
949w
twist(NH2)
vs = very strong; s = strong; m = medium; w = weak; vw = very weak
4
483
4.
Czech Republic
2006
[Content]
THERMAL DECOMPOSITION AND EXPLOSION TEMPERATURE

TEST
Kinetics of ANRTZNQ thermal decomposition has been studied under isothermal

conditions by manometric method using glass compensation zero-manometers Bourdon type
(Soviet manometric method) [12]. In addition we performed TGA and DTA studies
(5C/min). Isothermal thermal decomposition has been studied only in solid phase because
ANRTZNQ hasnt melting point and doesnt solve in the solvents which are usually used for
study of thermal decomposition. The DTA curve of ANRTZNQ has only one sharp
exothermal peak with maximum at 213C.
Gaseous products of thermal decomposition were analyzed in a mass-spectrometer MI
12-01-V (USSR) with special system for sample input [13] under the following conditions:
pressure inside the ion source 1104 Pa; accelerating voltage of EI source 3 kV; electron
energy of EI source 35 eV; cathode current 0.6 A; multiplier voltage 2 kV; the amount of
used ANRTZNQ sample is about 2-3 mg. The mass-spectrometer resolution allows
distinguishing between molecules and molecular fragments with close values of m/e (N2O
and CO2, N2 and CO etc.), but does not allow to say about comparative quantitative contents
of water in decomposition products.
Isothermal decomposition of the solid substance has been investigated at 180220C,
kinetic curves are satisfactorily described by a kinetic equation of the first order to the extent
of transformation of 3040% (Figure 4). Ratio m/V (ratio of sample weight to volume of
vessel) from 4.8103 to 4.9104 g/cm3 and packing of vacant volume of reaction vessel with
glass capillaries for the purpose of ratio S/V (ratio of surface of vessel to volume of vessel)
do not affect decomposition rate. When the ratio m/V decreases to 6105 g/cm3, a rate
constant of a thermal decomposition considerably grows, probably because of the increase of
the contribution reaction in gas phase in the general rate of the decomposition.
220oC
Vt ,
3
cm /g
200oC
320
240
190oC
o
210 C
160
180oC
80
0
0
40
80
120
160
200
240
Fig 4. Kinetic curves of thermal

decomposition of ANRTZNQ at
various temperatures in a solid
phase at loading density m/V =
6103 g/cm3.
t, min
The observed rate constants and activation parameters of thermal decomposition are
shown in Table 4, and Arrhenius temperature dependences are shown in Figure 5.
The main products of the thermal decomposition of ANRTZNQ are N2O (m/e = 44),
HNCO (m/e = 43), HCN (m/e = 27, 26) and NH3 (m/e = 17) (Figure 6). There is NO2 (m/e =
46) at products of the thermal decomposition, however its amount is much less in
comparison with N2O ratio of maxima intensities N2O/NO2 = 6:1. The peak with m/e = 30
corresponds to NO. The latter is generated under the action of electron impact from N2O and
484
Czech Republic
2006
[Content]
NO2, but not as a result of secondary oxidation-reduction reaction passing with the assistance
of NO2, because CO and CO2 peaks, which usually accompany the oxidation processes [14],
were not discovered in primary products of thermal decomposition.
Table 4. Kinetic parameters of ANRTZNQ thermal decomposition in a solid phase
Constant rate k105, s1 (at T, C)
180
190
200
210
220
1.82
6.23
16.7
36.1
85.2
Ea, kJ/mol
lgA
175.9
15.59
0.9975
ln k
-6
-8
-1 0
-1 2
2 .0 0
Fig 6.
2 .1 0
1 0 0 0 / T, K -1
2 .2 0
Fig 5. Arrhenius plot for the rate

constants of ANRTZ
thermal decomposition in
a solid phase.
3-D diagram of mass-spectrum of decomposition products of ANRTZNQ in a

solid phase
The composition of the thermal decomposition products allows to assume two main
initial ways of ANRTZNQ thermal decomposition:
6
485

NH
H2N
NNO2
NH
H2N
NNO2
N N
N NH
NNO2
NC-NHNO2
ANRTZ
Czech Republic
H2N
NH
2006
[Content]
H2N
N N
N2O
NC-NHNO2
NNO2
NH
N NH
H2N
m/e=187
H2N
ANRTZNQ
The first way of decomposition represents removal of nitrocyanamide and formation of

3-amino-5-nitrimino-1.4H-1,2,4-triazole (ANRTZ), which further decomposition was
considered earlier [8, 15]. This way of decomposition coordinates with the mechanism of
thermal decomposition of of 3,5-diamino-1-nitroamidino-1,2,4-triazole [8, 15]. The second
parallel way represents formation of four-membered cycle transition state and removal of
N2O from nitrimine group of heterocycle. Maximum with m/e = 187 (M+44) proves it.
Besides, there are peaks m/e = 100, 99 (3-amino-1,2,4-triazol-5-one), and also peaks of
fragmentation 1,2,4-triazole cycle under electron impact: m/e = 71, 70, 69 (C2H3N2O,
C2H2N2O), 55, 54 (C2HNO, C2NO), 42 (NH2CN), 30 (HN = NH), 29 (HN2) at massspectra of ANRTZNQ decomposition products.
Sensitivity to explosion by heat was determined as the temperature of flash and time to
explosion delay. The known method is used [1618]. The sample of explosive (10 mg) is
placed in a 8 detonator shell, and the lower end of this shell is plunged into a bath of
molten Woods metal at a measured temperature (T). The time required to cause explosion or
flashing is noted with a stop watch. Each experiment was repeated ten times and average
values are fixed. Then the similar operation is repeated with another temperature. The
specified method allows most simply to find the activation parameters of thermal
decomposition of explosive compounds. However, using such way the accuracy of definition
of activation parameters is, usually, low.
The temperature time to explosion curve for ANRTZNQ is shown in Figure 7. To find
activation parameters from ratio of temperature time to explosion, data processing in
coordinates ln = f(1/T), shown in Figure 7, was applied.
20
2.5
12
ln[(s)]
Time to thermal explosion, s
16
2.0
1.5
1.0
0.5
0
180
Fig 7.
b)
3.0
a)
190
200
210
Temper at ur e, oC
220
2.00
2.04
2.08
2.12
2.16 2.20
1000/ T, K -1
Ratio between the temperature and time to explosion for ANRTZNQ (a) and plot
of ln vs. 1/T (b)
This ratio can be expressed by the following equation:

ln = 23.73 + 12115.8/T, R = 0.9896
The values of energy of activation (Ea) and pre-exponential factor (A) are found from
this equation. The calculation is based on the following expression:
486
Czech Republic
2006
[Content]
= {cpRT 2/(Qexpl. Ea A)}exp(Ea/RT),
where cp heat capacity (we used approximate values for heat capacity organic
explosives cp = 1.25 J/(gK)); R ideal gas constant, 8.314 J/(molK); Qexpl. heat
decomposition reaction (J/g), identical with heat of explosion (calculated in the chapter 6).
Calculated activation parameters of thermal explosion are represented below:
Texpl.( =5 c) = 205C; T = 184-216C; Ea = 100.7 kJ/mol; lg A = 7.978.03 (average 8.00)
The obtained values of activation parameters of ANRTZNQ and calculated out of them
rate constants visibly differ from obtained in manometric experiments parameters (Table 4).
This fact may be concerned by both large error of activation parameters values from
explosion temperature test and/or greatly discriminate experimental conditions.
5.
IMPACT SENSITIVITY
The sensitivity to impact of ANRTZ was investigated with K-44-II Russian drop-weight
impact machine [17]. Experimental conditions: 5 kg weight drop hammer; the sample of
explosive is placed in the plunger assembly 1 [16], consisting of two steel rollers with
diameter 10 mm and height 12 mm [16]; samples weight is 30 mg. The results of tests are
collected in Table 5, the curves of sensitivity are shown in Figure 8. Each point was obtained
by results of 25 trials. For comparison the data for PETN are given. Samples of RDX with
weight 50 mg are used as a standard for K-44-II apparatus (f = 488% for H = 25 cm).
Table 5.
Impact sensitivity of ANRTZNQ
Load level (drop height) H0, cm
10
12
14
15
Explosion probability f, %
24
52
88
96
100
From the obtained data it follows that the sensitivity of ANRTZNQ is higher than
PETN, i.e. this compound is the sensitive dangerous high explosive.
f,%
100
PETN
80
ANRTZNQ
60
RD X 50 mg
40
20
Fig 8.
12
16
20
24
H , cm
Explosion probability (f) for ANRTZNQ vs drop height of load (H); for comparison
the data for PETN are given.
487
Czech Republic
2006
[Content]
The best method of the explosive impact sensitivity determination is a method of critical
pressure of initiation [19], however it requires to use complicated apparatus. To estimate
critical parameters of initiation by impact [19] we have tried to use a numerical method [20] and
activation parameters of thermal decomposition wich we obtained (chapter 4 in this paper).
Calculated value of Pcr. for ANRTZNQ is less then 450 MPa if it hypothetical melting point
is above 350C (compound is decomposed in a solid phase without melting). In calculations
we used activation parameters of thermal decomposition from manometric data for solid
phase and approximate value for heat capacity cp = 1.25 J/(gK). If using activation
parameters from explosion temperature test we could not get reasonable values of Pcr..
Experimental and calculation values of Pcr. for PETN is 480 MPa [19, 20].
6.
ENERGETIC AND DETONATION PARAMETERS
Energetic and detonation parameters of ANRTZNQ were estimated using

thermodynamic method with BKW equation of state [21, 22] at various sets of parameters
(BKW-RDX, -R, -RR, -C) and, also, by simple correlation methods [22, 23]. The calculation
was performed at maximum density of ANRTZNQ wich had been also calculated [24]. The
condensed carbon formed by detonation was taken into consideration in diamond (BKWRDX, -R, -C) and ultra fine diamond (BKW-RR) phases. The experiment and
thermodynamic calculation show that diamond phase is preferable in Chapman-Jouguet
point for high density explosives [25]. Table 6 shows average values obtained by the above
mentioned methods. Experimental values of detonation parameters for RDX [26] are shown in
Table 6 for comparison.
Table 6. Some energetic and detonation parameters of ANRTZNQ and RDX.
Explosives
NH
H2N
NNO2
O2NN
NNO2
ANRTZNQ
NNO2
N
NO2
N N
H2N
RDX
Brutto-formula
C3H5N9O4
C3H6N6O6
Molecular weight
231.13
222.12
Oxygen balance, %
31.1
21.6
Hf , kJ/mol
0
0, g/cm3
Qexpl., MJ/kg
+171
[3, 4]
1.87
4.63
+61.5
[3]
1.6%
1.70
1.80
5.40
1.70 g/cm3
V0, m3/kg
0.7510.6%
0.7621.70 g/cm3
D, km/s
8.733.4%
8.39
8.77
31.4
36.1
3740
3520
2.83
2.83
PCJ, GPa
32.7
TCJ, K
a)
4.9%
1797 3407
3.379.6%
b)
Notes: a) BKW-RDX; b) BKW-C.
488
7.
Czech Republic
2006
[Content]
CONCLUSION
ANRTZNQ is an attractive energetic compound because of its simple method of

synthesize and high yield. However in spite of initial hopes ANRTZNQ has appeared a highsensitivity explosive. Subsequently, relationships structure-properties for energetic
nitrimines were established during systematic study of properties of a large number of these
compounds and the results were explained.
Mechanism of the thermal decomposition of many nitrimines helps to understand their
rather high sensitivity. As a rule, their decomposition proceeds by ionic mechanism, or
through four-membered cycle transition state. In both cases decomposition begins with lower
temperatures than thermal decomposition by the radical mechanism. Rather low ANRTZNQ
thermal stability in combination with high melting point (exceeding decomposition
temperature under atmospheric pressure) inevitably results in high impact sensitivity of its
compound. The validity of the given conclusion is proved by an estimation of Pcr. by
numerical model of impact initiation [20].
ANRTZNQ has worse oxygen balance than RDX and does not surpass RDX by
energetic and detonation parameters. Besides, ANRTZNQ is more sensitivity by impact and
has smaller thermal stability than RDX. Due to all these facts ANRTZNQ is worse than
RDX as an high explosive.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
10
M. S. PEVZNER: The 1,2,4-triazole derivatives high-energetic compounds. Russ. Chem. J.

(Ross. Khim. Zh.). Vol. 41. No. 2, p. 7383, 1997 (In Russian)
H. H. LICHT, S. BRAUN, M. SCHFER, B. WANDERS, AND H. RITTER: Nitrotriazole: chemische
struktur und explosive eigenschaften. Proc. 29th Int. Annual Conf. ICT Energetic Materials
Production, Processing and Charakterization, Karlsruhe, Germany, p. 47/115, 1998
A. F. MCKAY: Nitroguanidines. Chem. Rev. Vol. 51. No. 2, p. 301346, 1952
A. M. ASTACHOV, I. V. GELEMURZINA, A. D. VASILIEV, A. A. NEFEDOV, L. A.
KRUGLYAKOVA, AND R. S. STEPANOV: New 1,2,4-triazolyl and tetrazolyl derivatives of
nitroguanidine. Proc. 32nd Int. Annual Conf. ICT Energetic Materials Ignition, Combustion
and Detonation, Karlsruhe, Germany, p. 139/110, 2001
A. M. ASTACHOV, A. D. VASILIEV, I. V. GELEMURZINA, W. A. SOKOLENKO, L. A.
KRUGLYAKOVA, AND R. S. STEPANOV: Nitrimines. I. Synthesis, structure and properties 3,5diamino-1-nitroamidino-1,2,4-triazole. Russ. J. Org. Chem. Vol. 39. No. 1, p. 130134, 2003
(In Russian)
A. M. ASTACHOV, W. A. SOKOLENKO, V. E. ZORIN, O. V. FALALEEV, A. I. RUBAILO, AND R.
S. STEPANOV: 1,2,4-Triazolyl and tetrazolyl derivatives of nitroguanidine: 1H, 13C and 15N
NMR Characterization. Proc. 33rd Int. Annual Conf. ICT Energetic Materials Synthesis,
Production and Application, Karlsruhe, Germany, p. 56/114, 2002
A. M. ASTACHOV, A. D. VASILIEV, M. S. MOLOKEEV, AND R. S. STEPANOV: X-ray
investigations of nitraminoderivatives of 1,2,4-triazole and tetrazole. Proc. All-Russian
Scientific Conf. Energetic Condensed Systems , Chernogolovka, Russia, p. 1920, 2004 (In
Russian)
A. M. ASTACHOV, A. A. NEFEDOV, L. A. KRUGLYAKOVA, A. D. VASILIEV, M. S. MOLOKEEV,
S. V. TRUBIN, P. P. SEMYANNIKOV, AND R. S. STEPANOV: Structure thermal stability
relationship for derivatives of nitroguanidine. Proc. International Scientific and Techn. Conf.
Modern Problems of Technical Chemistry, Kazan, Russia, p. 2735, 2004 (In Russian)
489
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
490
Czech Republic
2006
[Content]
A. M. ASTACHOV, A. D. VASILIEV, M. S. MOLOKEEV, V. A. REVENKO, AND R. S. STEPANOV:

Nitrimines. II. On structure nitramino-1,2,4-triazoles. Russ. J. Org. Chem. Vol. 41. No. 6, p.
928933, 2005 (In Russian)
N. S. MOROZOVA, E. L. METELKINA, T. A. NOVIKOVA, V. A. SCHLYAPOCHNIKOV, AND O. I.
SERGIENKO: Interpretation of IR spectra of nitroguandine. Russ. J. Org. Chem. Vol. 19. No. 6,
p. 12281232, 1983 (In Russian)
Y. OYUMI, A. L. RHEINGOLD, AND T. B. BRILL: Thermal decomposition of energetic materials.
XXIV. A comparison of the crystal structures, IR spectra, thermolysis and impact sensitivities of
nitroguanidine and trinitroethylnitroguanidine. Propellants, Explosives, Pyrotechnics. Vol. 12.
No. 1, p. 4652, 1987
G. B. MANELIS, G. M. NAZIN, YU. I. RUBTSOV, AND V. A. STRUNIN: Thermal Decomposition
and Combustion of Explosives and Powders, (Edited by F.I. Dubovitsky), Nauka, Moscow,
1996 (In Russian)
V. M. GRANKIN, AND P. P. SEMYANNIKOV: The ion source and the high-temperature source of
molecular beam for mass-spectrometer MI 12-01-V. Instruments and Experimental Tachniques.
Vol. 34. No. 4, p. 129131, 1991 (In Russian)
YU. SHU, B. L. KORSOUNSKII, AND G. M. NAZIN: Mechanism of thermal decomposition of
secondary nitramines. Uspekhi Khimii (Russ. Chem. Rev.). Vol. 28. No. 3, p. 320335, 2004
(In Russian)
A. A. NEFEDOV: Thermal decomposition of derivatives of nitroguanidine, Ph.D thesis, Siberian
State Technological University, Krasnoyarsk, 2004 (In Russian)
K. K. ANDREEV, AND A. F. BELYAEV: Theory of explosives, Oborongiz, Moscow, 1960 (In
Russian)
S. G. ANDREEV, A. V. BABKIN, F. A. BAUM ET. AL.: Physics of explosion, Vol. 1 (Edited by L.
P. Orlenko third edition), Fizmatlit, Moscow, 2002 (In Russian)
A. I. GOLBINDER: Laboratory works on a course in the theory of explosives, Rosvuzizdat,
Moscow, 1963 (In Russian)
G. T. AFANASIEV, AND V. K. BOBOLEV: Initiation of solid explosives by impact, Nauka,
Moscow, 1968 (In Russian)
A. V. DUBOVIK: Calculation of impact sensitivity parameters of solid explosives. Dokl. Acad.
Nauk USSR (Doklady Chemistry USSR). Vol. 286. No. 2, p. 377380, 1986 (In Russian)
C. L. MADER: Numerical modeling of explosives and propellants (second edition), CRC Press,
Boca Raton, 1998
V. I. PEPEKIN, AND S. A. GUBIN: Methods of calculation of detonation parameters of
explosives. Khim. Fizika (Chem. Phys. Report). Vol. 22. No. 9, p. 7297, 2003 (In Russian)
A. M. ASTACHOV, V. I. VLASENKO, AND E. S. BUKA: Estimation of energetic characteristic
some nitrimines. Proc. All-Russian Scientific and Techn. Annual Conf. Modern Problems of
Technical Chemistry, Vol. 1., Kazan, Russia, p. 249253, 2002 (In Russian)
A. M. ASTACHOV, I. V. GELEMURZINA, A. D. VASILIEV, A. A. NEFEDOV, L. A.
KRUGLYAKOVA, AND R. S. STEPANOV: 1,2,4-Triazolyl and tetrazolyl derivatives of
nitroguanidine: synthesis, structure and properties. Proc. All-Russian Scientific and Techn.
Annual Conf. Modern Problems of Technical Chemistry, Kazan, Russia, p. 2940, 2003 (In
Russian)
V. V. ODINTSOV, S. A. GUBIN, V. I. PEPEKIN, AND L. N. AKIMOVA: The form and size of
diamond crystals definition behind a detonation wave in the condensed explosives. Khim.
Fizika (Chem. Phys. Report). Vol. 10. No. 5, p. 687695, 1991 (In Russian)
M. F. GOGULYA, AND M. A. BRAZHNIKOV: Temperature of detonation products of the
condensed explosive. 1.The solid explosives. Khim. Fizika (Chem. Phys. Report). Vol. 13.
No.1, p. 5263, 1994 (In Russian)
Czech Republic
2006
[Content]
THE USE OF A KENICS STATIC MIXER

FOR THE REACTION OF TNT WITH NaOCl
Anthony J. Bellamy, Victoria Lomax and Kate Read
Department of Environmental and Ordnance Systems, Cranfield University,
Defence Academy of the United Kingdom, Shrivenham, Swindon SN6 8LA, UK
Abstract:
In the standard Shipp-Kaplan batch synthesis of hexanitrostilbene (HNS), the mixing of a
solution of TNT in THF-MeOH with aqueous NaOCl is very exothermic. This causes
considerable problems with scale-up, particularly if the objective is to quench the reaction
at an early stage in order to obtain a good yield of the first intermediate, 2,4,6trinitrobenzyl chloride, with minimum contamination by HNS and HNBB. The use of a
Kenics Static Mixer immersed in ice-water has been used to overcome this problem.
Under these conditions the formation (and precipitation) of HNS is sufficiently delayed to
avoid blocking the mixer and the resultant mixture may be quenched to give trinitrobenzyl
chloride in yield and purity comparable to the material obtained by the Shipp-Kaplan
recipe.
Keywords:
1.
static mixer, hexanitrostilbene, trinitrobenzyl chloride
INTRODUCTION
The synthesis of hexanitrostilbene (HNS) by reacting trinitrotoluene (TNT) with NaOCl was first
[1]
[2]
reported by Shipp and Kaplan in 1964 and has been widely studied since . As described by Shipp
and Kaplan, the reaction conditions may be varied so that the isolated product is either HNS (by adding
[1a]
TNT to NaOCl yield normally 40-45%)
, hexanitrobibenzyl (HNBB, by adding NaOCl to TNT
[1b]
yield normally ~80%))
or trinitrobenzyl chloride, the initially formed reaction intermediate (TNBCl,
[1b]
by quenching the reaction mixture in acid a short time after mixing yield normally ~85%)
, (see
reaction scheme in Fig 1). One can envisage that HNS is formed by self condensation of TNBCl
(reaction between TNBCl and its conjugate base B), whilst HNBB is formed by cross condensation of
TNBCl with the conjugate base of TNT (A), the mode of addition generating conditions which favour
one or the other of these two products. In less controlled mixing regimes both products are likely to be
formed.
Corresponding author: fax 44-1793-783878, email: A.J.Bellamy@cranfield.ac.uk
Anthony Bellamy, Victoria Lomax, Kate Read
491
Czech Republic
2006
[Content]
In the batch synthesis of HNS, the reaction occurs in essentially two stages. In stage one (mixing
duration minutes), TNT reacts exothermically with NaOCl to form TNBCl. In stage two (ageing duration ~2h), the TNBCl is allowed to react with the basic medium (NaOCl-NaOH) generating HNS
which separates from the medium. The best selectivity in the reaction is achieved if the temperature on
[2a, 3]
mixing is kept below 15C and the pH during the ageing period is maintained around 10
. Shipp and
[1a]
Kaplan
kept the mixing temperature low by pre-cooling both solutions to 0C, but did not control the
pH during the ageing period.
In scaling up the batch process, one of the major problems is controlling the temperature at the point
of mixing. This is also likely to be a problem if the reaction is performed as a semi-continuous flow
system. Any excess temperature rise on mixing will cause HNS to form prematurely in the mixing tubes
and result in blocking. In order to minimise this problem we have explored the use of a Kenics Static
Mixer immersed in ice-water in the hope that any heat generated on mixing would be rapidly transmitted
through the metal walls of the mixer into the surrounding bath.
2.
EXPERIMENTAL
[4]
The mixer-reaction system employed consisted of a Kenics Static Tube Mixer 37-03-075
[material 316SS, length 7.5 (190mm), OD 3/16 (4.75mm)] with 27 mixing elements (see Fig 2 inset
shows 3 of 27 mixing elements), oriented vertically. The reactant solutions [TNT in THF-MeOH-water
and aqueous NaOCl] were delivered from two syringe pumps [KD Scientific 100 series], via Teflon
tubing [ID 0.030, OD 0.063] to the mixing tee [Anachem U490, material SS, ID 1.00mm, internal
volume 2.2l] and thence into the top end of the static mixer via a short length of SS tube [ID 0.25mm
(0.01), OD 1/16] and a Swagelok reducing union [SS-300-6-1 (3/16 to 1/16), ID 0.05]. The outflow
from the bottom of the static mixer was conveyed by Teflon tubing [ID 2.5mm, OD 4.5mm, length
normally 3m] to the ageing or quenching vessel. Part of the arrangement is shown in Fig 2. Before
commencing a reaction, the whole of the system [inlet tubes (to ensure that the reactant solutions were
delivered at 0C), mixing tee, static mixer, outlet tube] was immersed in ice-water contained in a stirred
Dewar vessel and the inlet tubes were primed to within a few cm of the mixing tee. Because the
combined volume of the static mixer and the outlet tubes was a significant fraction of the total liquid
volume transferred, it was necessary to empty the system after the syringe pumps had delivered their
specified volumes. This was done by replacing both syringes with air-filled syringes and continuing the
delivery at the same rate.
2.1
Experiments in which NaOCl equivalents were varied
TNT solution: TNT (2.30g) in THF (18ml), MeOH (9ml), water (3ml) total volume of solution
32ml. The solution (25.0ml) was delivered at 435ml/h; addition time 3.45min. Amount of TNT
delivered 1.80g.
NaOCl solution: the number of equivalents of NaOCl was varied (0.5-1.8) by varying the
concentration of the solution whilst keeping the volume of solution delivered constant at 7.15ml;
delivery rate 124.3ml/h.
The outflow from the system was delivered into a stirred conical flask without cooling. After ageing
for 2h, the product was filtered off, washed with MeOH and then water, and finally dried.
492
2.2
Czech Republic
2006
[Content]
Experiments with pH control
In order to give sufficient depth for the glass electrode, these experiments were on twice the scale of
those in (a), but with the same flow rates. All experiments used 1.2 equivalents of NaOCl. After
collecting all of the reaction mixture in a stirred conical flask, the pH of the solution was reduced from
12-13 to the desired value by addition of 25wt% H2SO4 and was then controlled by the addition of 6wt%
NaOH [Radiometer Copenhagen PHM 82 standard pH meter with a TTT titrator]. Product isolation
was as in (a).
2.3
Quenched reactions
The outflow was added to a solution of conc HCl (2.3ml) in water (230ml). The resultant mixture
was stirred for 1h before the solid was isolated and washed with water. The length of the outlet tube was
varied; 1.5m gave a higher TNT content in the product and therefore gave insufficient time for complete
reaction of the TNT.
Reactions on five times the scale (9.43g TNT) were run using peristaltic pumps instead of syringe
pumps; the same flow rates were used.
2.4
Product analysis
HPLC conditions: -Bondapak C-18 10m, 3.9x150mm, mobile phase 52.5% water 47.5%
acetonitrile, flow rate 1.0ml/min. The retention times were: TNT 5.6min, TNBCl 6.4min, HNS 11.6min,
HNBB 13.8min, UNKNOWN 16.4min.
1
H NMR (DMSO-d6) was also used to confirm product identity and distribution :
TNT 2.56 (s, CH3), 9.03 (s, 2H, Ar)

TNBCl 5.00 (s, CH2Cl), 9.09 (s, 2H, Ar)
HNS 7.14 (s, 2H, CH=CH), 9.11 (s, 4H, Ar)
HNBB 3.35 (s, 4H, CH2), 9.08 (s, 4H, Ar)
UNKNOWN (appears to correspond to HPLC peak at RT 16.4min): 4.20 (d, 2H, J = 6.9Hz), 5.90 (t, 1H,
J = 6.8Hz) due to the low abundance of this product the aromatic peaks (if present) were not visible.
3.

3.1
HNS vs HNBB formation
In the first series of experiments (Table 1), the yield and composition of the product from a fixed
amount of TNT was investigated as the amount of NaOCl was varied from 0.5 to 1.8 equivalents, with
no attempt being made to control the pH in the ageing period. In theory only one equivalent of NaOCl is
required to convert all of the TNT to TNBCl and thence HNS. In practice, the highest material yield does
occur in the region of 1 equivalent of NaOCl, but the product in that case contains a high proportion of
HNBB. The highest proportion of HNS (93-96%) occurs when 1.2-1.4 equivalents of NaOCl are used,
but by this stage the material yield has started to decrease. The proportion of HNBB in the product
steadily decreases from 86% when 0.5 equivalents of NaOCl are used to ~5% when >1.2 equivalents are
used.
493
3.2
Czech Republic
2006
[Content]
pH control
In the second series of experiments (Table 2), all experiments were conducted using 1.2 equivalents
of NaOCl (the amount that gave a low proportion of HNBB but without reducing the material yield
excessively see Table 1), and the pH during the ageing period was controlled by addition of aqueous
H2SO4 and NaOH. Consistently higher material yields were obtained with pH control, compared to the
uncontrolled reaction with 1.2 equivalents of NaOCl. The material yield and composition of the product
was not particularly sensitive to pH in the range 9.5 to 11.5, but slightly purer HNS was obtained in the
[1a]
lower part of this range. This is in agreement with Golding and Hayes observations
from a series of
batch reactions.
An unknown product (<5%) was observed in all of these experiments (and in the uncontrolled
reactions when 1.0-1.2 equivalents of NaOCl were used). There appeared to be a correspondence
between two characteristic 1H NMR signals (an A2X system) and the HPLC peak at RT 16.4min. It is
suggested that this product is either -chlorohexanitrobibenzyl (intermediate C in the reaction scheme,
Fig 1) or more probably its hydrolysis product -hydroxyhexanitrobibenzyl.
3.3
Quenched reactions
One of our main objectives was to be able to prepare TNBCl using a flow reactor. Since the highest
selectivity for HNS formation most probably reflects efficient conversion of TNT to TNBCl, we chose
those reactant proportions that gave the purest HNS viz 1.2-1.4 equivalents of NaOCl, for testing in
quenched reactions (Table 3). Reactions using 1.4 equivalents of NaOCl gave high selectivity (~95%)
and high material yield (~95%) of TNBCl. Indeed these results are similar to those one obtains using the
[1b]
Shipp and Kaplan procedure
viz ~98% purity and 87% material yield. Using these same conditions at
5 times the scale (peristaltic pumps instead of syringe pumps) also gave good results.
4.
CONCLUSIONS
In all experiments where a significant amount of HNS was the final product, the flow system did not
become blocked because further reaction of the intermediate TNBCl, and hence HNS formation, was
delayed by the external cooling until the reaction mixture had left the flow system.
This mixing system gave a convenient way of scaling up the synthesis of HNS and TNBCl on a
laboratory scale, and has the potential for use at even larger scales.
494
Czech Republic
2006
[Content]
REFERENCES
[1]
(a) K.G.SHIPP, Reactions of -substituted polynitrotoluenes. I. Synthesis of

2,2,4,4,6,6-hexanitrostilbene, J.Org.Chem., 29, p.2620-2623, 1964 (b) K.G.SHIPP and
L.A.KAPLAN, Reaction of -substituted polynitrotoluenes. II. The generation and reactions of
2,4,6-trinitrobenzyl anion, J.Org.Chem., 31, p.857-861, 1966.
[2]
(a) P.GOLDING and G.F.HAYES, A parametric study of the synthesis of 2,2,4,4,6,6hexanitrostilbene from trinitrotoluene and sodium hypochlorite, Propellants, Explosives,
Pyrotechnics, 8, p.35-39, 1983 (b) B.SINGH and H.SINGH, Synthesis of 2,2,4,4,6,6hexanitrostilbene, Defence Science Journal, 31, p.305-308, 1982 (c) B.SINGH and
R.K.MALHOTRA, Hexanitrostilbene and its properties, Defence Science Journal, 33, p.165-176,
1983 (d) E.G.KAYSER, An investigation of the Shipp hexanitrostilbene (HNS) process, J.Energetic
Materials, 1, p.325-348, 1983.
[3]
A.E.GALLO and N.TENCH, Commisioning of a production plant for hexanitrostilbene,

J.Hazardous Materials, 9, p.5-11, 1984.
[4]
From Chemineer Ltd, Cranmer Road, West Meadows, Derby DE21 6XT, UK;
www.chemineer.com
Acknowledgment:
We thank Anjum Agha for HPLC analyses and Eurenco (Sweden) for support.
495
Czech Republic
2006
[Content]
Table 1. Reaction of TNT with variable amounts of NaOCl

HPLC analysis
% by peak area
NaOCl
Conc
%wt/vol
Equivalents
4.56
0.5
6.47
7.29
0.6
0.8
9.11
1.0
10.0
10.9
12.8
16.4
1.1
1.2
1.4
1.8
Yield
g*
TNT
TNBCl
HNS
HNBB
Unknown
0.1
2.6
3.3
0.9
0.8
0.7
0.3
0.0
0.0
1.2
3.0
0.0
0.0
0.5
0.5
0.4
0.6
0.0
0.0
2.1
10.6
23.1
9.3
55.9
56.2
67.5
77.1
92.8
95.8
91.0
86.3
73.8
87.4
42.7
41.3
29.8
20.2
6.2
4.2
5.7
0.0
0.0
0.0
0.0
1.2
1.6
1.8
1.0
0.0
0.0
0.60
0.76
0.88
0.74
0.90
0.86
0.80
0.74
0.60
0.35
* Maximum yield of HNS possible 1.78g (from 1.80g TNT)
Table 2. Reaction of TNT with NaOCl under pH control
pH
9.5
10
10.5
11
11.5
TNT
0.4
0.0
0.0
0.0
0.0
0.5
0.6
HPLC analysis
% by peak area
TNBCl
HNS
HNBB
1.2
90.0
3.3
1.1
91.9
4.1
0.9
92.1
3.6
1.3
87.5
7.3
1.0
88.8
6.0
0.8
89.8
4.4
0.5
88.3
6.7
Yield
g*
Unknown
4.1
2.8
3.1
3.9
4.2
4.4
3.8
2.42
1.99
2.31
1.74
1.89
2.14
1.93
* Maximum yield of HNS possible 3.94g (from 3.97g TNT)
496
Czech Republic
2006
[Content]
Table 3. Quenched TNT-NaOCl reactions
HPLC analysis
% by peak area
Conditions
NaOCl
Exit tube
equivalents
length (m)
With syringe pumps
1.0
3
1.2
1.4
1.5
With peristaltic pumps
1.4
TNT
TNBCl
HNS
6.9
1.4
11.7
5.1
0.3
0.5
1.4
2.6
83.9
98.0
82.9
91.7
99.2
96.1
94.7
95.7
9.2
0.5
5.4
3.2
0.5
1.9
2.0
1.7
3.9
0.2
91.7
97.6
0.8
1.0
Yield
g*
HNBB
1.5
1.9
3.6
1.1
1.54
1.25
1.72
1.85
1.67
2.16
2.20
1.77
10.9**
11.3**
* Maximum yield of TNBCl 2.29g (from 1.985g TNT)

** Maximum yield of TNBCl 10.86g (from 9.43g TNT)
497
NO2
O2N
Czech Republic
2006
[Content]
NO2
CH3
HO
O2N
CH2
NO2
NO2
(A)
(TNT)
NaOCl
NO2
O2N
NO2
CHCl
HO
O2N
NO2
CH2Cl
NO2
(B)
(TNBCl)
O2N
NO2
TNBCl
-Cl-
O2N
CHCl
CH2
NO2
NO2
TNBCl
-Cl-
O2N
CH2
NO2
O2N
NO2
HO
HNS
-HCl
(C)
O2N
CH2
O2N
NO2
(HNBB)
Fig 1. Reaction scheme
498
Czech Republic
2006
[Content]
Fig 2. Mixing system
499
Czech Republic
2006
[Content]
INVESTIGATION OF STABILIZING PROPERTIES OF

TRIPHENYLAMINE (TPA)
J. Bdek*, S. Cudzio*, S. Pietrzyk*, and S. Wilker**
Military University of Technology, 00-908 Warsaw, Poland
**
WIWEB ASt Heimerzheim, Groes Cent, 53913 Swisttal, Germany
*
Abstract:
In many laboratories, investigations aimed at replacing diphenylamine (DPA,
propellant stabilizer) with other, less harmful substances are being carried out.
Good effects have been obtained in experiments with triphenylamine (TPA),
which cannot produce owing to its structure carcinogenic nitroso
derivatives. In this work, results of an investigation of the kinetics of TPA
transformation during its nitration with NxOy are presented. The possibility of
application of instrumented thin layer chromatography (TLC) to analyze TPA
and its nitro derivatives is shown. The analytical data obtained are used to
estimate the rate of TPA depletion.
Keywords:
1.
propellant stability, stabilizing properties, TPA, DPA, TLC

analysis
INTRODUCTION
The kinetics of the changes occurring in the currently used stabilizers under influence
of nitrogen oxides emitted in the process of ageing of smokeless propellants is well
known [1]. For example, diphenylamine (DPA) the basic stabilizer of single-base
propellants is initially transformed into N-nitrosodiphenylamine (N-NO-DPA) and next, as
the propellant ageing process advances, 2-nitro-, 4-nitro- and higher nitrated derivatives of
DPA are formed [2]. Since carcinogenic properties of N-NO-DPA are confirmed, its
generation is a drawback for DPA. It should also be mentioned that nitrosamines are formed
during the ageing process of the propellants containing other stabilizers, like centralite,
akardite or 4-nitro-N-methyl aniline, thus they are all suspected of being carcinogenic. This
means that stabilizers which are nowadays used in propellants produce very toxic products.
Taking this into consideration, S. Wilker and coworkers [3] proposed the application of a
new propellant stabilizer triphenylamine (TPA). This compound does not include NH
bonds, so that N-nitroso-compounds cannot be created. The authors have shown that TPA is
an effective stabilizer for nitrocellulose based propellants. Joining this research we
undertook an attempt to develop a methodology of measurements of the reaction kinetics
between TPA and nitrogen oxides. The results of measurements were compared with those
obtained under the same conditions for the well-known stabilizer of smokeless powders
diphenylamine (DPA).
The currently used test methods for the evaluation of propellant stability can be divided
into two categories, depending on the propellant components being analyzed. The first
category includes methods which estimate the degree of decomposition of esters contained
in a representative propellant sample, subjected to heating. The second category includes
500
Czech Republic
2006
[Content]
methods based on an assessment of the stabilizer contents and the determination of the
products of reactions the stabilizer undergoes in the stabilized propellant samples
subjected to natural or accelerated ageing.
Because of the fact that each of the tests is an expensive and time consuming method,
causing especially a lot of trouble while testing new stabilizers, we propose a new, simple,
fast and inexpensive method of evaluation of the stabilizing properties. The method consists
in a direct nitration of a stabilizer adsorbed on a silica gel (so that the stabilizer is dispersed
like in a propellant sample) with nitrogen oxides followed by the qualitative and quantitative
analyses of the formed nitration products.
2.
EXPERIMENTAL
All the analyses were performed by the means of instrumented thin layer
chromatography (TLC). We looked for chromatographic systems and techniques of the
development of chromatograms suitable for separation of the following compounds:

triphenylamine (TPA), 2-nitro-TPA, 4-nitro-TPA, 2,4'-dinitro-TPA and 4,4'dinitroTPA,

diphenylamine (DPA), N-nitrosodiphenylamine (N-NO-DPA), 2-nitrodiphenylamine (2-nitro-DPA) and 4-nitrodiphenylamine (4-nitro-DPA).
The investigations were also aimed at determination of the analytical parameters of the
compounds, development of a quantitative analysis method and evaluation of the
measurement errors. The samples for analyses were prepared by nitration of the stabilizers
(TPA and DPA) with nitrogen oxides. First, the stabilizers were adsorbed on a silica gel and
then it was subjected to a stream of air/NxOy mixture. The reaction products were eluted with
methylene chloride and analyzed.
The equipment, chemicals and solvents used were: applicator Linomat IV (Camag);
horizontal chromatographic chamber DS (UMCS Lublin); densitometer CS-9000
(Shimadzu); chromatoplates with silica gel 60 F254 HPLC (Merck, No. 1.05548) or silica gel
with chemically bounded octadecyl (Merck, No. 1.05559); solid phase extraction set (J.T.
Baker), stabilizer standards (WIWEB ASt Heimerzheim and Merck), organic solvents (J.T.
Backer and Merck).
Selection of the chromatographic systems. The first step of investigations included
a search for a chromatographic system making the separation of the analytes, their
identification and determination of the analytical parameters possible. The experiments were
conducted using methylene chloride solutions of the standards (TPA, DPA and their nitroderivatives) with concentration of 10 ng/L. Samples of the solutions of 10 L in volume
were applied (spraying technique, band width 4 mm) on the start lines of chromatoplates.
The chromatograms were developed to a length of 510 cm in the horizontal chamber DS.
The spots originating from the nitro-derivatives of TPA and DPA were seen in the visible
light and TPA, DPA and N-NO-DPA were visualized by observation of the fluorescence
quenching. The best separation conditions of the individual groups of analytes (Fig. 1a,b)
were gained in the following chromatographic system:
TPA (isocratic two-steps elution): stationary phase high-performance silica gel

(Merck, No. 1.05548); mobile phase diethyl ether:hexane 2:1; v/v (first step:
distance of elution - 4 cm), chloroform:hexane 1:2 (v/v) (second step: distance of
elution - 8 cm).
Jan Bladek, Stanislaw Cudzilo, Sylwia Pietrzyk, Stephan Wilker
501
Fig. 1.
Czech Republic
2006
[Content]
DPA (isocratic one-step elution): stationary phase high-performance silica gel

(Merck, No. 1.05548); mobile phase hexane:methylene chloride 7:5 (v/v).
Separation of the TPA (a) and DPA derivatives (b), A absorbance in relative units
of the apparatus; z range of the chromatographic band [mm]
Quantification. To establish the quantitative relations A = f(c) (where: A an area of

a densitometric peak in relative units of the apparatus, c a mass of an analyte in
a chromatographic band [ng]), variable volumes (5500 L) of the solutions of standards
were applied on the start line of the chromatoplates and the chromatograms were developed
in the described above systems. Next the absorption spectra of the analytes were measured in
situ to get information on the wavelengths corresponding to an absorbance maximum max.
Finally, the absorption measurements were performed and the obtained results were used to
determine the dependence of A = f(c), as well as to evaluate the limit of detection, and the
maximum range of linearity of the detector indications (Tabs 1a and 1b).
Table 1a.
Analyte
Analytical parameters of the TPA derivatives

max
[nm]
Calibration curve,
Correlation Range of linearity

coefficient
[ng]
TPA
285 A=50218c - 4925
0.964
2500
2-nitro-TPA
275 A=56469c - 2958
0.989
1200
4-nitro-TPA
410 A=74142c - 658
0.994
1200
2,4'-dinitro-TPA 390 A=56145c + 1555
0.995
1600
4,4'dinitro-TPA
415 A=73449c + 5633
0.980
1600
A absorbance, c the mass of an analyte in the chromatographic band [ng]
502
A = f(c)
Detection
limit
[ng]
200
100
100
100
100
Table 1b.
Analyte
Czech Republic
2006
Analytical parameters of the DPA derivatives

max
[nm]
Calibration curve
Correlation
Range of linearity
coefficient
[ng]
DPA
265
47.7c - 2422
0.990
2500
N-NO-DPA
282
32.7c + 1564
0.988
1800
2-NO2-DPA
430
51.3c + 2047
0.999
2000
4-NO2-DPA
380
0.999
2100
67.5c + 3973
A absorbance, c the mass of an analyte in the chromatographic band [ng]
Fig. 2a.
[Content]
A = f(c)*
Detection
limit
[ng]
150
120
90
80
Densitograms of the products of nitration of TPA with air/NxOy mixtures for

different concentration c (v/v) of NxOy
Transformations of stabilizers exposed to nitrogen oxides. Having developed an

analytical tool which could be used for the qualitative and quantitative analyses of TPA,
DPA and their nitration products, we made an attempt to assess the rate of the changes of the
503
Czech Republic
2006
[Content]
stabilizers (TPA and for comparison DPA) exposed to nitrogen oxides. To this end
a 0.1% (wt.) solution of each stabilizer in methylene chloride was prepared and then it was
mixed with silica gel. The mixture was kept for half an hour at room temperature to reach the
adsorption-desorption equilibrium and retention of the TPA or DPA on the adsorbent. After
that, the mixture was filtered off, the adsorbent was dried and divided into samples of 0.75 g
in mass. The samples were placed into typical SPE cartridges, and from a few cartridges, the
adsorbed stabilizer was eluted with two 1-mL portions of methylene chloride using SPE set.
The masses of eluted stabilizers were determined by analyzing the combined eluates, and in
this way information on the amount of stabilizer that was adsorbed on the silica gel was
obtained.
Through the other cartridges, filled with silica gel containing adsorbed TPA or DPA,
5-mL portions of air/NxOy mixture with increasing concentration of nitrogen oxides were
passed. The stabilizers were then nitrated with a rate depending on the concentration of
NxOy. After finishing the nitration process, remaining stabilizers and their nitration products
were eluted and analyzed. Results of densitometric measurements of the obtained
chromatograms are shown in Figs 2a and 2b.
Fig. 2b.
504
Densitograms of the products of nitration of DPA with air/NxOy mixtures for

different concentration c (v/v) of NxOy
3.
Czech Republic
2006
[Content]
DISCUSSION AND CONCLUSIONS
The obtained results of measurements show that the stabilizing properties of TPA are
comparable to that of DPA. This is indicated by the similar rate of depletion of the
stabilizers subjected to nitrogen oxides under the same conditions, Fig. 3.
Fig. 3.
Changes in the amount of unreacted TPA and DPA after their nitration with
nitrogen oxides versus concentration of NxOy in air (c), u relative mass fraction
of an analyte in the sample
The proposed method for observation of the transformation of propellant stabilizers

subjected to nitrogen oxides which relies on:
(i)
(ii)
(iii)
(iv)
adsorption (from a solution) of a known quantity of the stabilizer on a silica gel,

filling a column (applied in SPE) with the obtained gel,
passing a NxOy/air mixture of known composition through the column,
elution and quantitative analysis of the stabilizer and its nitration products,
seems to be an effective one. As a matter of fact the method cannot be used for a direct
evaluation of the propellant stability, but owing to its simplicity, inexpensiveness and
fastness it is especially preferred for a comparative assessment of the stabilizing performance
of various stabilizers. It makes the observation of direction and kinetics of the changes of
stabilizers possible without the necessity of long-lasting ageing of stabilized samples of
propellants, Fig. 4.
Fig. 4.
Results of nitration of TPA and DPA (adsorbed on silica gel) with nitrogen
oxides, u relative mass fraction of an analyte in the sample, c concentration of
NxOy in air
505
Czech Republic
2006
[Content]
Instrumented thin-layer chromatography proved to be a useful method for testing

propellant stabilizers. The instrumentation of the method provides a possibility of
quantitative analysis. The detection limit is low enough and the indications of the
densitometer detector are linear within a sufficiently wide range of the concentration of the
tested analytes.
Acknowledgment:
This research was supported by the Ministry of Education and Science through Military
University of Technology, Grant PBS 700.
REFERENCES
[1]
VOLK E., Determining the shelf life of propellants, Propellants, Explosives,

Pyrotechnics, 1, 59-65 (1976).
[2]
LUSSIER L.S., GAGNON, H., BOHN M.A., On the chemical reactions of DPA and its
derivatives with NO2 at normal storage temperature condition, Propellants, Explosives,
Pyrotechnics, 25, 117-125 (2000).
[3]
WILKER S., PETRILEK J., SKLDAL J., PANTEL G., STOTTMEISTER L., Stability analyses
of double base propellants containing new stabilizers, Proc. 33rd Int. Ann. Conf. ICT,
109 (112), Karlsruhe, Germany (2002).
506
Czech Republic
2006
[Content]
SOME PROPERTIES OF A COMPOSITE PROPELLANT PREMIX

Daniel Buczkowski, Bogdan Florczak and Tomasz Saaciski
Institute of Industrial Organic Chemistry, 6 Annopol Str.,03-236 Warsaw, Poland
Abstract:
In this paper some dangerous properties of a composite propellant premix (decomposition
temperature, sensitivity to mechanical stimuli, character of combustion and ability to
detonation) as well as thermal analysis (DTA), burning rate and heat of combustion are
determined. The premix consists of fine-grained ammonium perchlorate, dioctyl adipate and
a binder (PBAN). The results bring to the conclusion that the premix is a dangerous goods.
Keywords:
premix, composite solid rocket propellant, safety, sensitivity and utility
parameters
1.
INTRODUCTION
The manufacture of a composite propellant is made by batch processes. Processes like

grinding, premixing, mixing and casting are potentially hazardous[1]. They are usually operated
and performed in buildings designed to work with explosives and to protect workers. However,
decreasing of threatens that appear during manufacturing of solid rocket propellants is still a
very important aim of researches. In this area, dangerous properties of ammonium perchlorate
(AP), proofed by e. g. accident in 1988 in PEPECON[2], are still under investigation.
The mixing and casting processes are the most complex and have more influence than
other processes on properties of manufactured propellant[2,3]. The premixing is a typical
operation during manufacturing composite propellant. It is a very important process which
influences on parameters of composite propellant. It is also very dangerous.
The tested premix was a composition consisted of small sized (below 0.03 mm) fractions
of AP, a plasticizer (dioctyl adipate) and as a binder - PBAN (polybutadiene/acrylic acid/
acrylonitrile polymer).
2.
PERFORMANCE CALCULATIONS
The thermodynamic properties were calculated using ICT Thermodynamic Code. The ICTThermodynamic Code is based on a method developed by the National Aeronautics and Space
Administration (NASA)[4]. Thermodynamic equilibrium was calculated for constant volume
conditions. The calculation of the heat of explosion is of special interest, because the
experimental measurements using a calorimetric bomb are sometimes difficult due to high
temperatures and erosive reaction products. The calculations of physicochemical properties of
the premix are shown in Table 1.
Daniel Buczkowski, Bogdan Florczak, Tomasz Saaciski
507
Czech Republic
2006
[Content]
Table 1. Physicochemical properties of the premix (calculated).

Parameter
Value
Sum Formula:
C
15.2218
27.2935
7.0018
50.2753
Cl
6.5368
Mixture ratio, O/F
3.31
Oxygen balance [%]
-40.03
Density [g/cm ]
1.562
Enthalpy of formation, Hf [kJ/kg]
3.
-2150.13
Gas formation [mol/kg]
44.885
Heat of explosion [J/g]
3825.7
Volume of combustion gasses, without H2O

at 25 oC [dm3/kg]
863.7
EXPERIMANTAL METHODS AND RESULTS

3.1
Behaviour at elevated temperature
Decomposition temperature was determined by heating 0.5 g of the premix placed in

open glass test tubes, with a diameter ca. 15 mm. The test tubes were located in Woods alloy
bath which was heated at a constant rate of 5oC/min, started from 100oC. A temperature sensor
was placed in Woods alloy. Temperature at which the sample started to decompose, was
recorded as decomposition temperature[5,6]. Tests were carried out according to the Polish
Standard[7]. A sketch of the experimental set is presented in Fig. 1.
Differential thermal analysis (using DTA 551 Ex analyzer, manufactured by OZM
Research) was carried out by heating 5.0 mg of the sample at a rate 5 oC/min. Analysed samples
were placed in open test tubes. A thermocouple, protected by a glass sheath, was inserted
directly into the sample. Data were evaluated using the analyzer and MEAVY 2.0.0.4 software.
An example of DTA thermogram of the premix is shown in Fig. 2. Onsets (Tonset) and maximum
(Tm) temperatures of the first exothermal effect as well as deflagration temperature are listed in
Table 2.
508
Czech Republic
2006
DATA
ACQUISITION
MODULE
[Content]
4
5
3
6
2
EXAMINER
1
Fig. 1. Scheme of the experimental stand for determination of the temperature

of decomposition, according to the Polish Standard BN-76/6091-08.
Description: 1 heating device; 2 tested sample; 3 temperature sensor; 4 test tube; 5 protective glass tube; 6 Woods alloy; 7 metal block
Fig. 2. DTA thermogram of the premix, heating rate 5oC/min.
509
Czech Republic
2006
[Content]
Table 2. Some thermal and burning properties of the premix

Parameter
Decomposition temperature [oC]
Differential thermal analysis [oC]
Tonset
Tm
Heat of combustion [J/g]
Burning rate [mm/s] for p = 40 bar, t= 23oC
3.2
Value
264
261
367
4174
15.930.12
Heat of combustion and burning rate of the premix.
The adiabatic bomb IKA calorimeter system C 4000 was used to determine the heat of
combustion. Combustion experiments were conducted with an automatically controlled
adiabatic jacket and without oxygen atmosphere (pressure about 400 Pa). The calorimeter was
calibrated by burning standard propellant to determine its effective energy equivalent.
Crawford bomb was used to determine the burning rate of the premix[5]. The measurements
were carried out at temperature 23oC and constant pressure about 40 bar. The results of these
test as well as heat of combustion of the premix are presented in Table 2.
3.3
Sensitivity to mechanical stimuli
The BAM Fallhammer and BAM friction apparatus methods, acoording to[8], were used to
measure the impact and friction sensitivity of the premix. The results are shown in table 3.
Table 3. Some explosive properties of the premix, AP and a composite propellant

Result for
Ammonium Perchlorate,
Propellant*
fractions:
fine
coarse
< 0.03 mm > 0.06 mm
Impact sensitivity [J]
20
10
15
15
Friction sensitivity [N]
360
> 360
> 360
80
Test A.1: BAM 50/60 steel tube test
Tube was fragmented completely
Test 3 (d): Small-scale burning test
Sample didnt exploded
Test 2 (c) (ii): Internal ignition test
Tube didnt fragmented
* - propellant consists of the premix, listed in the table grades of AP and RDX
Parameter
510
Premix
3.4
Czech Republic
2006
[Content]
Field tests
In Table 3 there are presented the results of the mentioned below tests[8]:
- A.1 test BAM 50/60 steel tube test was used to evaluate the ability to detonation of the
premix initiated by booster; the premix was placed in a steel tube.
- 3 (d) test Small-scale burning test was used to determine the response of the premix to
fire; a sample was placed on a sheet of burning paper.
- 2 (c) (ii) test Internal ignition test was used to determine the violency of explosive
reaction; the premix was placed in a steel tube, closed in both ends and initiated by black
powder.
3.5
Measurements of behavior at elevated temperature showed that the decomposition

temperature of the premix is 261oC. Figure 2 shows thermal decomposition process of the
premix measured by DTA at heating rate of 5oC/min. An endothermic peak seen at 244oC, refers
to crystal structure lattice phase transition of AP from orthorhombic to cubic. An exothermic
reaction occurs between 261oC and 367oC (three exothermic peaks: 292oC, 321oC and 367oC).
In case of thermal decomposition process of AP an endothermic peak (247oC) and two
exothermic peaks (334 oC and 447 oC) are observed[9].
Heat of combustion and burning rate of the premix are appreciable and combustion runs in a
relatively non-violent way.
The premix may detonate if is initiated by a strong booster and is placed in a massive,
strong case.
Examined premix is rather low sensitive to mechanical stimuli.
Performed tests of premix point that this material is a dangerous goods.
4.
CONCLUSIONS
The results presented in the paper have shown that the premix is a kind of a dangerous
goods.
Qualitative results of ability to detonation, ability to ignition and deflagration in
confinement of the tested premix are the similar to the results obtained for ammonium
perchlorate and the tested propellant.
The tested premix characterises smaller impact sensitivity in comparison with pure
ammonium perchlorate and with a hexogen contained propellant. Impact sensitivity of the tested
premix is significantly lower than for fine ammonium perchlorate (fractions below 0.03 mm). It
is probably caused by phlegmatization fine AP grains by the binder.
511
Czech Republic
2006
[Content]
Acknowledgment
This research was supported by the State Committee of Scientific Research through Institute of
Industrial Organic Chemistry, Grant Nr 3 T09B 021 27.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
512
G. P. Sutton and O. Biblarz: Rocket Propulsion elements, John Wiley & Sons, INC, New York,
2001
J. G. Routley: Fire and Explosion at Rocket Fuel Plant Henderson, Nevada, Technical Report
Series, United Fire State Administration, 1988
A. Davenas: Solid Propulsion Technology, Pergamon Press, Oxford, 1993
S.Gordon and B. J. McBride: Computer Program for Calculation of Complex Chemical
Equilibrium Compositions and Applications, NASA Reference Publication. I. Analysis 1311,
October 1994
R. Meyer, J. Khler and A. Homburg: Explosives, Wiley-VCH Verlag GmbH, Weinheim, 2002
T. B. Fedoroff: Encyclopedia of Explosives and Related Items, Picating Arsenal, Dover, NJ, USA,
Vol I, p. XVI, 1960
Polish standard BN-76/6091-08
"Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria"
(ST/SG/AC.10/11/Rev.4) United Nations, New York and Geneva, 2003
N. KUBOTA, Propellants and Explosives, Thermochemical Aspects of Combustion, WILEYVCH GmbH, Weinheim, Germany, 2002
Czech Republic
2006
[Content]
THERMOPHYSICAL PROPERTIES OF THERMOPLASTIC

COMPOSITE PROPELLANTS BASED ON MODYFIED PVC BINDER
Vladica Bozic, Marko Milos, and Djordje Blagojevic
Faculty of Mechanical Engineering, University of Belgrade, Kraljice Marije 16, 11120 Belgrade 35
Serbia and Montenegro
Abstract:
Modern solid propellants have found wide application as power sources, from
commercial small gas-generator systems like airbag inflators, to large launch vehicles
in space programs. Complex phenomena, which exist during combustion of solid
propellants, require precise determination of burning rate for ballistic calculations with
exact models of combustion. To predict and/or simulate combustion processes,
knowledge of the thermophysical properties of solid propellants is necessary. Properties
of particular interest are density, specific heat, and thermal conductivity. This paper
reports values of these parameters, which are measured for three different thermoplastic
composite propellants based on ammonium perchlorate as oxidizer and modified PVC
as binder. Thermophysical properties of modified PVC binder have been determined
also. They can be used for prediction of specific heat in other PVC propellant
compositions. Results obtained in this paper may be quite useful in researching
combustion processes of these propellants, analysis and design of their function, and
industrial applications.
Keywords:
1.
AP/PVC composite rocket propellant, thermophysical properties
INTRODUCTION
A solid rocket propellant is a complex and stable mixture of different oxidizing and
reducing ingredients. When the solid rocket propellant is ignited, it burns in a controlled
manner during which it is transformed into high-temperature low molecular-weight gases at
the grain surface. These gases provide the kinetic energy, which is released through the
exhaust nozzle of the rocket motor. During combustion, the burning surface regresses
perpendicularly to itself in parallel layers. The rate of regression is called the burning rate.
Burning rate as well as specific impulse, density, suitable physical properties, and safety
characteristics are usually the most important propellant properties which are necessary in
the design of new solid propellant rocket motor.
Numerous kinds of chemicals are used in making solid propellants compositions, to
obtain optimized combustion characteristics for different purposes. Presented here are the
results of an experimental investigation of thermophysical properties of thermoplastic solid
composite propellants based on polyvinyl chloride-PVC binder: specific heat, and thermal
conductivity. Like other composite propellants, the examined propellants basically consist of
inorganic solids (ammonium perchlorate-AP as oxidizer, aluminum-Al as metallic fuel),
uniformly dispersed in a continuous matrix of organic thermoplastic elastomer binder, which
are based on polyvinyl chloride-PVC. As these parameters are necessary in heat transfer
analysis of the burning of solid propellants, the results obtained make possible better
understanding of the combustion phenomena and can be further used in models of combustion.
1
Vladica Bozic, Marko Milos, Djordje Blagojevic
513
2.
Czech Republic
2006
[Content]
THEORY
The combustion mechanism of solid propellants is very complex, as different

phenomena occur during combustion including change of phase, energy transfer, and mass
flow. They involve a large number of various chemical processes (solid state decomposition,
reaction of gaseous products with solids, gas-phase reactions) and physical processes
(diffusion, sublimation, adsorption-desorption). In order to understand the combustion
phenomena of solid propellants under steady or unsteady operations, it is necessary to
examine the thermal behavior of propellants and their ingredients. The knowledge of their
thermochemical and thermophysical properties is a prerequisite to accurate prediction and/or
simulation of combustion processes. Also, experimenters must know the fundamental
properties of burning solid propellants to handle and/or deduce other, more elusive
experimental parameters. Additionally, this is very important for safety and shelf life
assessment.
Using various weight and calorimetric techniques in high-temperature environments like
differential scanning calorimetry (DSC), differential thermal analysis (DTA), thermal
gravimetry analysis (TGA) in combination with mass spectrometry (MS), gas
chromatography (GC) and heat flow calorimetry (HFC), it is possible to investigate
propellant degradation kinetics and identify pyrolysis products. Thermophysical properties
of particular interest in understanding phenomena of burning solid propellants are density-,
specific heat-c, and thermal conductivity-. From these two important aggregates follow: the
thermal diffusivity-, and thermal responsivity-. All of these properties depend not only on
composition, but also on temperature and possible phase transition of each propellant
ingredient; subsequently, pressure effects are expected as well.
3.
EXPERIMENTS
This paper reports values of thermophysical parameters: specific heat and thermal
conductivity, which are measured for three different propellant compositions. These
compositions represent three main groups in the examined class of propellant based on
modified polyvinyl chloride binder: 1) low energy propellant denoted by A, 2) medium
energy propellant denoted by B and 3) high energy propellant denoted by C. Formulations of
these propellants are given in Table 1. The other compositions in this class are modifications
of the above, depending upon the specific requirements for each rocket motor.
Heat flow into a substance induces many physical and chemical changes, which can help
to identify and characterize a sample. Differential scanning calorimetry (DSC) monitors heat
effects associated with phase transformations and chemical reactions as a function of
temperature. In a DSC the difference in heat flow to the sample and a reference (inert
material) at the same temperature, is recorded as a function of temperature. The temperature
of both the sample and reference are increased at a constant rate. A flow of nitrogen gas is
maintained over the samples to create a reproducible and dry atmosphere, and eliminates air
oxidation of the samples at high temperatures. Since the DSC is conducted at constant
pressure, heat flow is equivalent to enthalpy changes, so measurements are used to determine
enthalpies of phase change, phase and glass transition temperatures, and heat capacities.
Specific heat of propellant formulations has been determined using Du Pont Model 910
DSC. Sapphire (Al2O3) was used as reference material, as it is a standard material of known
variation of specific heat with temperature. The sample is sealed into a small aluminum pan.
The specific heat of a test specimen was determined by comparing the thermal lag between
sample and reference material under identical conditions at the temperature of interest.
514
Thermophysical Properties of Thermoplastic Composite Propellants Based on . . .
Czech Republic
2006
[Content]
Table 1. Composition of examined propellants.

comp. A
comp. B
comp. C
AP [%]
69.00
75.00
69.75
Al [%]
0.50
0.50
10.00
PVC binder [%]
30.50
24.50
20.25
1650
1680
1760
Density- [kg/m ]
The thermal conductivity of thermoplastic propellant compositions given in Table 1 has

been determined using the test apparatus shown in Fig. 1. Experimental measurements have
been conducted over the temperature range from 20 to 100 oC starting from ambient
temperature in the following way. The test specimen whose thermal conductivity was to be
measured, approximately 5 mm thick and 70 mm in diameter, has been placed as a sandwich
between two cover plates, made of materials with similar properties. The whole assembly
has been placed between a source of heat and a cooling system. A stabilised heater with a
copper plate has been used as a source of heat, while the cooling system was based on flow
of water with constant temperature. Thermal conductivity is determined from precisely
measured temperatures at contacts between different materials, test specimen and cover, and
known values of thickness of these materials. The measuring system is first calibrated with
reference material under identical conditions at the temperature of interest. Lead was used as
the standard material with known value of thermal conductivity versus temperature.
2
1
5
3
5
70
1. Test sample
4. Cooling system
2. Cover plates
5. Copper heater
3. End insulator
Fig 1.
Temperature sensors
Details of the test section
515
4.
Czech Republic
2006
[Content]
RESULTS
4.1
Thermophysical properties of ingredients
Ammonium perchlorate (NH4ClO4), although has relatively low enthalpy of formation

(-2518 kJ/kg), due to properties like good oxygen balance (+34.04 wt %), high density (1950
kg/m3) and others, is used as a main oxidizer in composite propellants. As an inorganic salt,
it is anisotropic and nonconductive (dielectric) material, as opposed to metallic additives.
The different scanning calorimetry experiments were performed on ammonium perchlorate
samples of 2-3 mg in closed aluminum pans. Samples of AP were heated in a flowing
atmosphere of nitrogen (flow 40 ml/min) at a heating rate of 10 oC/min and ambient
pressure. At temperature around 242 oC DSC curve (Fig. 2.) shows an endothermic peak
corresponding to the solid phase transition of AP from orthorhombic to cubic form. The first
exothermic feature appeared at 275 oC, followed by a second exothermic feature resulting in
thermal decomposition.
This result corresponds to available data from literature: ammonium perchlorate has
orthorhombic crystalline phase in the range of 83 to 513 K (-190 to 240 oC) and cubic above
513 K [1]. Actually, the crystalline transition can already be observed at 475 K (202 oC).
Although sometimes debated, ammonium perchlorate melts [2] at 723 K (450 oC) absorbing [3]
c=249400 J/kg. Densities in the solid and liquid phases, respectively are ort=1957 kg/m3 and
cub=1756 kg/m3, and l=1710 kg/m3. Therefore AP crystals volumetrically expand by about
10% when passing from the orthorhombic to the cubic solid phase. This will affect the
propellant properties to an extent depending on the AP mass fraction.
The specific heats of ammonium perchlorate obtained in our measurements correspond
to values given in JANAF data [4], shown in Fig. 3. On the basis of these values two simple
relationships can be defined: The first is valid in the temperature range below 240 oC:
exothermic
3
2
HEAT FLOW (W/g)
1
0
-1
-2
-3
-4
-5
-6
-7
0
50
100
150
200
250
300
350
400
TEMPERATURE ( C)
Fig 2.
516
DSC curves of ammonium perchlorate at a heating rate of 10 K/min
Czech Republic
2006
[Content]
2200
cAP (measured and JANAF data)

cAP-equation (3)
cAP-equation (6)
2100
SPECIFIC HEAT (J/kg K)
2000
1900
1800
1700
1600
1500
1400
1300
1200
1100
1000
0
50
100
150
200
250
300
350
400
450
500
TEMPERATURE ( C)
Fig 3.
Specific heat of ammonium perchlorate
cort , AP = 1093.3 + 1.723 (T Tref ) [J/kg K]
(1)
where Tref=300 K, while the following is valid in the interval from 240 to 450 oC:
ccub, AP = 1167.6 + 1.713 (T Tref ) [J/kg K]
(2)
As the gradients are very similar in both previous equations connected with crystalline
phases, we suggest the following expression which can replace them, and also is valid in the
temperature range from 20 to 450 oC:
c AP = (1093.3 + A) + 1.718 (T Tref ) [J/kg K]
(3)
where A=0 for T515 K (242 oC), A=74.28 for 515<T<723 K (242<t<450 oC). This
relationship yields for the average value of specific heat of ammonium perchlorate in the
range from 20-450 oC the value cAPav=1486.7 J/kg K. This average value also can be used in
propellant burning diagnostics.
Specific heat of pure AP is widely reported in the literature. For orthorhombic crystal
structure is given the following relationship [5] which is valid up to 240 oC:
cort , AP = 577.5 + 1.72 T + 4.72 10 6 T 2 [J/kg K]
(4)
The following relationship is valid for cubic structure (in the interval from 240 to 450 oC):
ccub, AP = 725.4 + 1.5 T + 161.48 10 6 T 2 [J/kg K]
(5)
Data obtained using expressions (4) and (5) are very similar to data reported as JANAF data
and our measurements, but are more complicate than expression (3).
The simplest relationship found in the literature [6] is in relative agreement with the
experimental values in the range from 300 K (27 oC) to above 700 K (430 oC):
5
517
Czech Republic
2006
[Content]
c AP = 1093.6 + 1.884 (T Tref ) [J/kg K]
(6)
where Tref=300 K, but it is obvious that the above relationship has not compare well with
experimental data. It is seen in Fig. 3 that equations (3), or (4) and (5) are in better
agreement with experimental data than equation (6).
For aluminum, a metal fuel commonly used in rocket propulsion, there is only one solid
phase of interest, melting at 933 K (660 oC) with a latent heat [7] of ls=397200 J/kg. Densities
in the solid and liquid phases are, respectively, c,Al=2.702 kg/m3 and l,Al=2.380 kg/m3. The
following linear law [8] is valid for the specific heat of the solid phase:
cc , Al = 895.2 + 0.498 (T Tref ) ,
(7)
at Tref=300 K, while for the liquid-phase specific heat one has cl,Al = 1176.5 J/kgK.
Thermophysical properties of binders are difficult to evaluate, because too many
different ingredients exists in compositions. Its effect is tempered by the fact that its mass
fraction seldom goes between 16-30% of the propellant total mass. The binder which has
been used in the propellants under test mainly consists of polyvinyl chloride (PVC) and
dioctyl phthalate (DOP) as plasticizer. Samples of binder weighing around 8 mg were
examined in closed pan, and the measured DSC curve is shown in Figure 4. Samples were
heated at temperature scan rate of 10 K/min in a nitrogen atmosphere (flow 10 ml/min) at
ambient pressure. The curve has shown initially a very slow endothermic effect up to around
240 oC, where a first small peak arises. This part of the curve is due to melting of gelled
PVC binder. As the temperature increases the endothermic effect continues up to 272 oC,
after which the exothermic effect begins. Exothermic effect has two peaks: first at 289.5 oC,
which is completed at 291 oC, and a second at 306 oC.
Thermal degradation of the examined binder began after a temperature of 241 oC, and at
beginning is connected with vaporization of plasticizer. Transformation to exothermal effect
1.0
0.9
0.8
0.7
0.5
exothermic
HEAT FLOW (W/g)
0.6
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
-0.5
0
50
100
150
200
250
300
350
400
TEMPERATURE ( C)
Fig 4.
518
DSC curves of PVC binder at a heating rate of 10 K/min
Czech Republic
2006
[Content]
3000
cbinder
binder-equation (8)
SPECIFIC HEAT [J/kgK]
2800
2600
2400
2200
2000
1800
1600
1400
1200
0
25
50
75
100
125
150
175
200
225
250
275
300
TEMPERATURE [ C]
Fig 5.
Specific heat of PVC binder
at temperature 272 oC is probably due to its degradation. This process shows maximum rate
at the first exothermic peak, and is finished soon after this. Second exothermic peak is
connected only with decomposition of PVC in binder and its dehydrochlorination, i.e.
elimination of chlorine from PVC which is released as hydrogen chloride (HCl).
Dehydrochlorination of the PVC shows a maximum rate at the second exothermic peak, and
appears to be complete at around 350 C.
DSC examination of binder which is used in rocket propellants has shown that the
internal structure of thermoplastics, basically consists of PVC, and is strongly affected by
addition of plasticizer and stabilizer. Changes in thermal history of plasticized PVC binder
are evident, as it has not any endothermic peak up to 240 oC. Plasticizer-free PVC (as
amorphous polymer) typically has a glass transition with a relaxation peak at temperature Tg
of about 80 oC. Addition of plasticizer to PVC has lowered the glass-transition temperature
to lower values. Also, it is obvious that stability to heat is greatly improved by the addition
of stabilizers, as liberation of HCl is displaced to higher temperatures in comparison with
pure PVC.
Specific heat results are shown in Fig. 5 in the temperature range from 40 to 260 oC. On
the basis of these values, the temperature dependence of specific heat for the binder can be
expressed with the following linear relationships respectively, where Tref=300 K:
cbin = 1527.8 + 4.96 (T Tref ) [J/kg K]
4.2
(8)
Specific heat of composite PVC binder propellants
Specific heat of propellants has been measured using DSC in a temperature range from
293 (20 oC) to 383 K (110 oC), and the results obtained are shown in Fig. 6. Samples
weighing around 40 mg were closed in aluminum pans and heated in a flowing atmosphere
of nitrogen (flow 50 ml/min) at a heating rate of 10 oC/min and ambient pressure.
519
Czech Republic
2006
[Content]
1450
SPECIFIC HEAT [J/kg K]
1400
1350
1300
1250
propellant A-measured
propellant A-equation (9)
propellant B-measured
propellant B-equation (10)
propellant C-measured
propellant C-equation (11)
1200
1150
1100
10
20
30
40
50
60
70
80
90
100 110
120
TEMPERATURE [ C]
Fig 6.
Specific heat of composite propellants with PVC binder
Based on measurement results, the temperature dependence of specific heat can be

expressed for these three compositions with the following linear relationships, where
Tref=300 K:
4.3
Composition A: c A = 1212.6 + 2.45(T Tref ) [J/kg K]
(9)
Composition B: c B = 1209.3 + 2.33(T Tref ) [J/kg K]
(10)
Composition C: cC = 1184.1 + 2.46(T Tref ) [J/kg K]
(11)
Thermal conductivity of composite PVC propellants
The temperature dependence of the thermal conductivities of ingredients used in the

examined compositions was difficult to find in the literature. Based on various data [7], the
thermal conductivity of ammonium perchlorate was found to satisfy the following
expression in a temperature range from ambient to 513 K (240 oC):
ort , AP = 0.5087 1.73 10 4 (T Tref ) 1.53 10 6 (T Tref ) 2 [W/(m K)]
(12)
where Tref=300 K, while for PVC[9] the following expression is accepted to 368 K (95 oC):
PVC = 0.1636 + 1.056 10 4 (T Tref ) 8.92 10 7 (T Tref ) 2 [W/(m K)]
(13)
These values of thermal conductivity are shown in Fig. 7. For aluminum the following linear
expression can be adequate for limited purposes [7], where Tref=300 K:
c, Al = 203.4 + 0.116 (T Tref ) [W/(m K)]
(14)
We can see that aluminum is a much better thermal conductor than AP and binder.
520
Czech Republic
2006
[Content]
THERMAL CONDUCTIVITY [W/mK]
0.55
AP
PVC
0.50
0.45
0.40
0.35
0.20
0.15
0.10
-50
-25
25
50
75
100
125
150
175
200
225
250
TEMPERATURE [ C]
Fig 7.
Thermal conductivity of AP and PVC
Thermal conductivity of thermoplastic propellants compositions given in Table 1 has

been determined as described in Section 3, and results are shown in Fig. 8. We can say that
the temperature dependence of thermal conductivity can be considered to be approximately
linear for all three compositions in the test temperature range.
THERMAL CONDUCTIVITY [W/mK]
0.6
0.5
0.4
0.3
propellant A
propellant B
propellant C
0.2
0.1
0.0
10
20
30
40
50
60
70
80
90
100 110 120
TEMPERATURE [ C]
Fig 8.
Thermal conductivity of composite propellants with PVC binder
521
Czech Republic
2006
[Content]
Results have shown that thermal conductivity of examined propellants decreases with
increasing temperature from ambient value. On the basis of values and character of curves,
as the thermal conductivities of aluminum and PVC are increasing functions of temperature
(which is opposite for AP), we can say that thermal conductivity of propellants is basically
controlled by the amount of AP in a formulation. As we have expected, the presence of
aluminum in composition C, as a much better conductor, has increased thermal conductivity
in comparison with composition A (which have same amount of binder), but not to a level to
be expected from values in equation (12). If we compare propellants A and B, where
propellant B has less percentage of binder in formulation, we can see that thermal
conductivity of binder is less than thermal conductivity of AP and values indicate that it is
close to the thermal conductivity of pure PVC. Unfortunately, analysis has shown that
thermal conductivity of modified PVC binder cant be determined exactly from preceding
results, as we have excessive disagreement among results.
5.
DISCUSSION
As for composite propellants the temperature dependence of specific heat

evaluated as:
cc = ox c AP (T ) + me c Al (T ) + (1 ox me) cbin (T ) ,
[7]
can be
(15)
where ox is the oxygen mass, me is the metallic fuel mass, while the binder mass fraction
includes all other propellant ingredients. The previous equation is a good approximation if
the energy of binding is small compared with the sum of the energies associated with the
homogeneous pockets of ingredients; this may not be true for the smallest AP particle sizes.
Expression for linear increase of specific heat of HTPB binder with temperature can be
found in literature [6]. Assuming that the same form of expression is valid for examined PVC
binder, and at the basis of experimental results of three previous compositions and the
known value of specific heat for AP and Al, we have determined specific heat of binder in
each composition, and have obtained the following relationships, where Tref=300 K:
1) Specific heat of PVC binder in composition A:
cbin , A = 1487.7 + 4.13(T Tref ) [J/kg K]
(16)
2) Specific heat of PVC binder in composition B:

cbin , B = 1570.8 + 4.24(T Tref ) [J/kg K]
(17)
3) Specific heat of PVC binder in composition C:

cbin ,C = 1639.7 + 5.97(T Tref ) [J/kg K]
(18)
We have obtained the following general expression for specific heat of PVC binder, by
averaging preceding relationships for PVC binders in compositions A, B and C:
cbin , AV = 1566.1 + 4.78(T Tref ) [J/kg K]
(19)
Using this linear expression, we have calculated values of specific heat, and compared
with previously measured data for binder (Eq. 8). Results of this comparison are shown on
Fig. 9, and we can see that values of specific heat of PVC binder obtained from both
equations are approximately equal.
522
Czech Republic
2006
[Content]
3200
equation (16)
equation (17)
equation (18)
average value, equation (19)
equation (8)
3000
SPECIFIC HEAT [J/kg K]
2800
2600
2400
2200
2000
1800
1600
1400
0
25
50
75
100
125
150
175
200
225
250
275
300
TEMPERATURE [ C]
Fig 9.
5.4
Comparison of measured and calculated values of specific heat of binder
Deduced condensed-phase ballistic properties
Additional ballistic properties of the considered propellants: the thermal diffusivity, and
thermal responsivity were computed on the basis of the preceding measured results for
density, specific heat, and thermal conductivity. All of these are valid only for condensed
phase (AP cubic structure, before AP crystalline transition).
The thermal diffusivity- of the condensed phase for the tested composite propellants
are computed according to:
=
(20)
and values at various temperatures are collected in Table 2, and shown in Fig 10. All results
are valid at ambient pressure, and possible pressure effect has not been considered.
Table 2. Thermal diffusivity of examined propellants.
x 107 [m2/s]
11
Temperature [oC]
comp. A
comp. B
comp. C
20
2.148
2.505
2.262
50
1.786
1.953
2.009
80
1.497
1.73
1.746
100
1.415
1.597
1.542
523
Czech Republic
2006
[Content]
The preceding results show that thermal diffusivity was found to decrease linearly for
temperature increasing from ambient value. Also, as the average thermal diffusivity of pure
AP is 1.89x10-7 m2/s, thermal diffusivity of propellants is controlled basically by APs
presence. If we compare compositions A and C, which have the same amount of binder in
compositions, it is obvious that the presence of Al as a good thermal conductor perceivably
increases the average thermal diffusivity. If we compare compositions A and B, we can see
the effect of PVC binder; it diminishes thermal diffusivity in the orthorhombic AP region.
By definition, thermal responsivity- of the condensed phase is:
= c
(21)
and the values, calculated for the three composite propellants, are given in Table 3. These
values are also shown in Fig. 10.
Table 3. Thermal responsivity of examined propellants.
comp. B
comp. C
20
912.2
1002.1
974.2
50
887.2
940.3
980.6
80
858.1
930.3
965.9
100
863.6
926.4
942.9
1050
3.1
1000
3.3
2.9
2.7
950
900
2.5
850
THERMAL RESPONSIVITY
propellant A
propellant B
propellant C
2.3
2.1
800
750
1.9
THERMAL DIFFUSIVITY
propellant A
propellant B
propellant C
1.7
1.5
700
650
1.3
0
10
20
30
40
50
60
70
80
90
100
110
2 1/2
comp. A
THERMAL RESPONSIVITY [J/m s K]
Temperature [oC]
THERMAL DIFFUSIVITY x 10 [m /s]
[J/(m2 s1/2 K)]
600
120
TEMPERATURE [ C]
Fig 10. Thermal diffusivity and thermal responsivity of examined propellants

524
Czech Republic
2006
[Content]
The thermal responsivity of examined propellants is also found to decrease

approximately linearly as the temperature increases from ambient value, and it has obvious
great influence on thermal conductivity results. Generally, thermal responsivity of
propellants is increased with increasing level of AP in formulation. Addition of aluminum in
propellant formulation increased thermal responsivity more than AP.
6.
CONCLUSION
Two thermal analysis methods were applied to study thermodynamic properties of

thermoplastic composite propellants based on AP and modified PVC as binder. Specific heat
and thermal conductivity have been measured at various temperatures for three different
propellant compositions. At the base of this results, and known values for raw materials, an
expression for calculation of specific heat of modified PVC binder has been determined and
it is shown to be in good agreement with values obtained by direct measurement. Additional
ballistic properties of the considered propellants: the thermal diffusivity, and thermal
responsivity were computed at the basis of the preceding measured results. The results have
confirmed the importance of temperature effects on the condensed-phase thermophysical
properties, and can be very useful for modeling and understanding of propellant combustion
under various operating conditions.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
13
M. M. MARKOWITZ and D. A. BORYTA: Some Aspects of the Crystallographic Transition

of AP, ARS Journal, Vol. 32, No. 12, p.1941-1942, 1962
J. A. STARK and R. E. TAYLOR: Determination of Thermal Transport Properties in AP,
Journal of Propulsion and Power, Vol. 1, No. 5, p.409-410, 1985
C. GUIRAO and F. A. WILLIAMS: A Model for Ammonium Perchlorate Deflagration
between 20 and 100 atm, AIAA Journal, Vol. 9, No. 7, p.1345-1356, 1971
M. W. CHASE, JR., NIST-JANAF Thermochemical Tables, Fourth Edition, Journal of Physical
Chemistry, Chemical Reference Data, Monograph 9, 1998
C. F. PRICE, T. L. BOGGS and R. L. DERR: The Steady-State Combustion Behaviour of AP
and HMX, AIAA Paper 79-0164, 1979
F. S. BLOMSHIELD and J. R. OSBORN: Effects of Variable Solid Phase Thermal Properies
on Propellant Combustion, Acta Astronautica, Vol. 12, No. 12, p.1017-1025, 1985
C. ZANOTTI, A. VOLPI, M. BIANCHESSI and L. DE LUCA, Measuring Thermodynamic
Properties of Burning Propellants, (Nonsteady Burning and Combustion stability of Solid
Propellants, Vol. 143, L. De Luca, E. W. Price, and M. Summerfield, AIAA), New York, p.
145-196, 1992
R. H. PERRY and C. H. CHILTON, Chemical Engineers Handbook, 5th edition, Mc-GrawHill, New York, 1973
LJ. MASKOVIC, R. MAKSIMOVIC and V. JOVOVIC, Polymer Materials-Physical
Properties and Certain Aspects of their Application, Police Academy, Belgrade, p. 81, 1997
525
Czech Republic
2006
[Content]
THERMAL DECOMPOSITION OF THE

TRANS-1,4,5,8-TETRANITRO-1,4,5,8-TETRAAZADECALIN.
Yurii M. Burov, Irina S. Kucherova
Institute of Problems of Chemical Physics, Russian of Sciences,
142432 Chernogolovka, Moscow Region, Russian Federation.
E-mail: yuburov@icp.ac.ru
Abstract
During synthesis of molecules, loaded the large number of nitro groups, obtained
molecules, in which one cannot be attributed to known - flat, or pyramid-shaped frames. So,
in a molecule of the trans-1,4,5,8-tetranitro-1,4,5,8-tetraazadecalin (TND) because of a
steric repulsion of oxygen adjacent of nitro groups there is a turn around of the bond N-N.
Let's call such of nitramine groups as twist-groups. One molecule of the TND contains two
pair of such twist-groups. It was possible to suspect, that the decay TND will drive on the
same gear, as for earlier studied secondary nitramines the pre-exponential factor of
reacting will be peer 1014,3 -1, and statistical factor = 4, and the activation energy, equal
strength of the bond N-N will be reduced that nitramine group are deformed by turn around
of the bond N-N. This supposition has appeared fair only half. The thermal decomposition
of the TND was studied by a manometer method in a solid state, in solution in mdinitrobenzene, and in gas. The reaction rate in solution and in gas is described by an
equation of the first order. In solution in a temperature range 150-180 lgAliq = 11,151
(-1), liq=24,32 kcal/mol, in gas the constant rate was gauged only at temperature 1800
and it has compounded kg = 110-3 c-1. In a solid state of the constant rate were calculated
on an equation of the first order on initial velocity and V = 425 cm3/g.. The rate in a solid
state by a composite mode depends on degree of fullness of a vessel, and for obtaining
activation parameters the relation V/m was extrapolated on zero. The following values of
activation parameters of the decomposition of the TND in a solid state are obtained: lgAs =
8,141 (-1), s = 24,23 kcal /mol, the retardation effect of the crystalline lattice (REL) at
180 0 is peer kliq/ks = 103. The experimentally measured value of activation energy of the
decomposition of TND is lower than reference for the decomposition of the secondary
nitramines almost twice. This is very large value. The data, obtained in activity,
demonstrate, that the thermal decomposition of the secondary nitramines with twist-groups
considerably differs from decomposition studied before compounds with flat and pyramidshaped nitramine groups and demands further research.
Keywords: trans-1,4,5,8-tetranitro-1,4,5,8-tetraazadecalin, thermal decomposition, kinetics
526
Thermal Decomposition of the trans-1,4,5,8-Tetranitro-1,4,5,8-tetraazadecalin
1.
Czech Republic
2006
[Content]
INTRODUCTION
The secondary nitramines still are a source of new power stuffs and consequently the
researches of relation of thermal stability of these connections from their constitution prolong to
be actual. It is now considered demonstrated, that the decomposition secondary nitramines starts
with a homolytic bond cleavage N-N [1]. Offered in a number of papers the of reacting scheme
passing through a cyclical transient state, have not received an admission. The main reason for
the benefit of homolytic bond cleavage N-N is served by high value of the pre-exponential
factor, equal 1014,5 -1, reference for reacting separation of radicals. Was established, that there
are two kinds of the nitramine groups - flat and pyramid-shaped, it was showed, that flat groups
more strongly pyramid-shaped [2]. These deduction was followed in the late paper [3]. During
synthesis of molecules, loaded the large number of nitro groups, obtained molecules, in which
one cannot be attributed to known - flat, or pyramid-shaped frames. So, in a molecule of the
trans-1,4,5,8-tetranitro-1,4,5,8-tetraazadecalin (TND) because of a steric repulsion of oxygen
adjacent of nitro groups there is a turn around of the bond N-N. Let's call such of nitramine
groups as twist-groups. One molecule of the TND contains two pair of such twist-groups. It was
possible to suspect, that the decay TND will drive on the same gear, as for earlier studied
secondary nitramines the pre-exponential factor of reacting will be peer 1014,3 -1, and
statistical factor = 4, and the activation energy, equal strength of the bond N-N will be reduced
that nitramine group are deformed by turn around of the bond N-N. This supposition has
appeared fair only half.
2.
The thermal decomposition of the TND was studied by a manometer method in a solid
state, in solution in m-dinitrobenzeneand, and in gas. The reaction rate in solution and in gas is
described by an equation of the first order. In solution in a temperature range 150-1800 lgAliq =
11,151 (-1), liq=24,32 kkal / mol, in gas the constant rate was gauged only at temperature
1800 and it has compounded kg = 110-3 c-1. In a solid state of the constant rate were calculated
on an equation of the first order on initial velocity and V = 425cm 3 / g . The rate in a solid state
by a composite mode depends on degree of fullness of a vessel, and for obtaining activation
parameters the relation V/m was extrapolated on zero. The following values of activation
parameters of the decomposition of the TND in a solid state are obtained: lgAs = 8,141 (-1),
s = 24,23 kkal / mol, the retardation effect of the crystalline lattice (REL) at 180 0 is peer
kliq/ks = 103.
The experimentally measured value of activation energy of the decomposition of TND is
lower than reference for the decomposition of the secondary nitramines almost twice. This is
very large value. The nitramine group is rather conservative to an inductive effect of groups. So,
in 2,4,6-(NO2)3C6H2N(NO2)Me the benzene ring reduces D (C-N) all on 5 kkal / mol, while for
bonds -, -, C-Hal the value of this decrease is peer 15-20 kkal / mol [1]. The relevant
value has a geometrical constitution of the nitramine group. 1,3-Dinitro-1,3-diazacyclopentane
and 1,3,5-trinitro-1,3,5-triazacyclohexane (RDX) have till two pyramid-shaped nitramines
group, and their decomposition rate in vapors with inhibitor on the order is higher, than one for
dimethylnitramine, diethylnitramine, 1,3,5,7-tetranitro-1,3,5,7-tetraazacyclooctane (HMX) and
Yurii Burov, Irina Kucherova
527
Czech Republic
2006
[Content]
other connections having only the flat nitramine groups. D (N-N) in the flat nitramine groups is
peer 40,5 kkal / mol, and in pyramid-shaped - 37,7 kkal / mol 2. The distinction makes 2,8 kkal /
mol. In the twist-groups of the TND the activation energy is reduced as contrasted to flat
nitramine groups on 16,2 kkal / mol. Even more radical changes have taken place with the value
pre-exponential factor.
For all studied before secondary nitramines average the value of the pre-exponential factor
in calculation on one group is equal 1014,5 -1. In TND the pre-exponential factor in calculation
on one nitramine group is peer 1010,65 -1. Such low value of the pre-exponential factor can
indicate that the transient complex has cyclical frame.
The data, obtained in activity, demonstrate, that the thermal decomposition of the
secondary nitramines with twist-groups considerably differs from decomposition studied before
compounds with flat and pyramid-shaped nitramine groups and demands further research.
3.
[1]
[2]
[3]
528
REFERENCES
G. B. Manelis, G. M. Nazin, Yu. I Rubtsov, V. A. Strunin Thermal decomposition and combustion
of exolosives and pouders. M.: Nauka, 1996. 223p.
Yu. M. Burov, G. M. Nazin - Kinetics and catalysis, 1982, V. 23, p. 12-17.
J. C. Oxley, A. B. Koob, R. Szekers,W. Zheng J Phys. Chem., 1994. V98, p. 7004 7008.
Thermal Decomposition of the trans-1,4,5,8-Tetranitro-1,4,5,8-tetraazadecalin
Czech Republic
2006
[Content]
OBSERVATION OF CHANGES IN CHOSEN PROPERTIES OF

PLASTIC EXPLOSIVES DURING ARTIFICIAL AGEING
Martina Chovancov, Peter Oko, Jozef Lopch, Alzbeta Pechov
VTS (Military Technical and Testing Institute) Zhorie, 905 24 Senica, SK
Abstract
This work presents the results of military explosives stability, sensitivity and detonation
velocity that were investigated during artificial ageing. The biggest change after ageing
was recorded in friction sensitivity. The least changes of sensitivity were spotted on the
samples aged in the aluminium-polyethylene foil.
Keywords:
plastic explosives, stability, impact sensitivity, friction sensitivity,
detonation velocity, artificial ageing
1.
INTRODUCTION
Military explosives should make ammunition to work safely and properly and must not
worsen the ammunition quality even during peace time long-term storage where at any rate
certain reserves should be kept. Therefore the military explosives and ammunition must be
subjected to very rigorous supervision and demands relating to its durability, stability and
also safety. High explosives, as chemical substances blend, have got their durability limited
by several factors among that there are chemical composition, manufacturing type, purity,
storage conditions, ammunition application and related construction material compatibility.
Under these factors, there might be a gradual change of chemical and chemical-physical
properties of an explosive. Of course, also the sensitivity characteristic can be subject of
change resulting to safety and function degradation of ammunition in which is explosives
built in. Therefore there is a must to observe not only quality status but also quality
characteristic of high explosives and to eliminate problems that might have been occurred
during life-in service of ammunition with explosive built in. These problems can be as
follows: limited stability and lifetime, incompatibility with construction material, tendency
for dangerous changes of properties during ageing, too large a sensitivity increase etc.
An important part of military explosive evaluation is determination of ageing influence
on explosive properties while being determine its lifetime. The artificial ageing can be a
proper way how to observe ammunition lifetime process. It is the process of quick ageing
simulating a nature explosive ageing in ammunition system. Usually, it is realized by
temperature conditioning at higher temperatures that speed up the physical-chemical
processes which under normal storage condition would be running for several years in
explosive. Main goal of the process is that during several weeks to bring the explosive to the
status in which explosive might has been after couple of years of ageing. Artificial ageing
method is used for prediction of the changes of stability, sensitivity, mechanical and
operation explosive properties and obviously of their lifetime.
Long-time period physical and chemical stability of military explosives and related
material compatibility are involved between basic requirements of military explosives. The
requirements must be applied for each military explosive munitions-application.
The term of Explosive stability means an ability to keep up quality status and so not to
be a subject to any of chemical-physical changes that would tend to undesired changes under
Martina Chovancov, Peter Oko, Jozef Lopch, Albta Pechov
529
Czech Republic
2006
[Content]
normal condition. There must not be changes of explosive characteristics or unexpected selfignition or even explosion. Long-term stability observation during long-run storage has its
reasonable meaning mainly from point of view of keeping a manipulation safety, storability
and explosive functionality.[1]
2.
CHANGES IN EXPLOSIVE PROPERTIES DURING ARTIFICIAL

AGEING
Properties study of plastic explosives on the base of hexogen during artificial ageing has
pointed out on changes in physical and detonation characteristics of the explosive. Artificial
ageing conditions were as follows: 30 and 90 days at temperatures 120C, 180C and 195
C, 5 7,5 hours of dynamic and isothermal heating. There was a result of the tested sample
that at the isothermal heating the critical temperatures of fast decomposition, ignition and
explosion was 1520 C less than at dynamic heating.[2]
Japan authors was studying ageing of 6 high explosives PBX types for several years,
evaluated various tests during the ageing, for example detonation rate, explosion under
water, friction sensitivity, hammer impact sensitivity, DTA, TGA, VST. Most of explosives
tested were confirmed as well-power even after ageing process.[3]
American and Canadian scientists detected dynamic-mechanical properties of hexogen
explosives of PBX bound with Kraton (hydrogenated butadiene-styrene polymer) and
plasticized wax explosives X-0287 and oil explosive X-0298. Explosive samples were
ageing at T=23, 60 and 74 C for 3 years. At X-0298 explosive there was at fast ageing no
worsening of mechanical properties but molecular mass was at decrease by ageing increase
temperature because of irregular chain split. In by-wax-plasticized explosive (X-0287)
related to kinetic degradation of Kraton a significant difference was found over and under
wax melting point. This way created model predicts durability of both explosives as equals
as 20,5 years under the worst military conditions ever would be (74 C).[4]
Ageing process study of I-RDX (so called insensitive RDX) and cast PBX explosives on
base of I-RDX proved that artificial ageing process does not affect low impact sensitivity of
PBX composites. This low sensitivity is conditional to I-RDX content what means a big
contribution for insensitive ammunition development. I-RDX is the hexogen type produced
by special re-crystallization process, patented by EURENCO fy.[5]
Study of sensitivity changes of explosives wax-moderated after artificial ageing (113
days,70C) proved an increase of wax-moderated explosive sensitivity to impact, friction
and electrostatic spark in spite of their thermal characteristic (DTA, VST) remained
unchanged. These changes have occurred because of explosive crystal exposure under high
temperature ageing used. Although the temperature of 70 C is not extreme, it is
inappropriate for observing the sensitivity changes of wax-moderated explosives after
artificial ageing.[6]
3.
EXPERIMENTAL
Chemical and thermal stability, impact sensitivity, friction sensitivity and detonation
velocity were studied on the hexogen-based plastic explosive samples during artificial
ageing under various conditions. When comparing, there were except of plastic explosive
samples used also 2 wax-moderated explosive samples. The goal was to find the limits under
that the parameters are still subject of changes and assess proper ageing conditions.
Used explosives:
530
Observation of Changes in Chosen Properties of Plastic Explosives During . . .
Czech Republic
2006
Comp CHa 88 % RDX + 12 % Dutral (EPDM)
Comp CHc 84 % RDX + 12 % Dutral + 12 % SiO2
Comp CHe 80 % RDX + 12 % Krynac (NBR) + 8 % Al
Comp C4 91 % RDX + 9 % PIB
T/H/E 50 % RDX + 48 % TNT + 2 % wax
AOa 97 % HMX + 3 % wax
[Content]
Review of the performed tests: Tab 1.

Review of the test explosives: Tab 2.
Table 1. Performed tests list

Explosive property
Chemical stability
Thermal stability
Sensitivity
Power (performance)
Test to perform
VST
DTA
Temperature of ignition
Impact sensitivity
Friction sensitivity
Detonation velocity
Procedure by
Stanag 4556
TP-SHVV-1080-00-ZH,
Stanag 4491
Stanag 4489,4487
AOP-7
Table 2. Test explosives

High explosive
Comp Cha
Comp CHc
Comp CHe 1/97
Basic component Test to perform

VST, DTA, Temperature of ignition, Impact
RDX
sensitivity and Friction sensitivity
Comp CHe 2/00

Comp C4
RDX
THE
RDX, TNT
AOa
HMX

sensitivity and Friction sensitivity, Detonation
velocity
velocity
velocity
NOTES:
VST
vacuum stability test
DTA
differential thermal analysis
DMDNB
marking agent
RDX
hexogen
HMX
oktogen
TNT
- trinitrotoluene
531
Czech Republic
2006
[Content]
Artificial ageing conditions are enlisted in Tab 3.
Table 3. Artificial ageing conditions

Temperature(C)
Number of days
50
92, 138, 276
65
20, 40, 60
Package type
Glass NATO-test tube
Al-foil + sealed PE-foil
sealed PE foil
Glass NATO-test tube
Al-foil + sealed PE-foil
sealed PE foil
NOTES:
Al aluminum
PE - polyethylene
3.1
Chemical and thermal stability tests.
Chemical stability test of explosives was performed by Stabil 16-Ex vacuum stability
test equipment at constant temperature 100C for 40 hours. The test was done using all
samples after each ageing interval at both temperatures. Tab 4 enlists the results of not-aged
samples and also of samples after ageing completion.
Thermal stability was done using Differential thermal analyses method by DTA-550-Ex
apparatus and by determination temperature of ignition with all samples after ageing at both
temperatures. Heating rate at the two tests was 5C/min.Tab 20 enlists the review of thermal
stability results measured before/after ageing at both temperatures.
Table 4. VST (cm3) results
Ageing type
Sample
Comp
CHe1/97
Comp
CHe2/00
Comp CHa
Comp CHc
Comp C4
AOa
532
Non-ageing
0,526
60 days ageing at 65C

Tube
Al+PE PE
0,631
0,437
0,312
276 days ageing at 50C

Tube
Al+PE PE
0,500
0,376
0,281
0,287
0,141
0,176
0,119
0,111
0,188
0,192
0,759
0,822
0,333
0,245
0,427
0,673
0,129
0,132
0,410
0,613
0,115
0,186
0,468
0,471
0,151
0,130
1,664
0,484
0,129
0,195
1,468
0,481
0,148
0,272
1,071
0,677
0,149
0,215
Czech Republic
2006
[Content]
Table 5. DTA and temperature of ignition (C) results

Sample
Che1/97
60d/65C
276d/50C
Che2/00
60d/65C
276d/50C
Cha
60d/65C
276d/50C
CHc
60d/65C
276d/50C
C4
60d/65C
276d/50C
AOa
60d/65C
276d/50C
Non-ageing
Aged in tube
Mp St Max TI Mp St Max
204 198 218 209
201 208 213
204 198 216
197 194 214 210
202 197 214
203 199 216
204 210 215 210
204 190 217
205 196 217
207 211 216 212
206 213 219
206 214 220
198 210 213 209
201 213 216
204 214 218
181 243 264 258
189 253 272
190 252 272
TI
Aged in Al+PE foil

Mp St Max TI
Aged in PE foil
Mp St Max TI
210 201 193 213

208 204 198 215
210 202 196 213

208 204 197 215
210
208
211 202 209 215

213 204 198 216
211 201 198 215

213 204 199 217
211
213
213 205 196 217

213 205 193 217
213 204 213 218

213 204 198 218
212
213
214 205 212 220

213 205 212 220
214 207 213 219

213 206 214 220
214
213
213 204 214 218

211 205 214 219
211 207 215 218

211 205 216 219
212
211
261 190 252 271

262 188 252 272
261 190 254 272

261 192 253 273
260
262
NOTES:
Mp melting point
St start of thermal decomposition
Max maximal temperature of thermal decomposition
TI temperature of ignition
Results of the tests have proved that the tested plastic explosives and AOa waxmoderant are chemical and thermal stabile after artificial ageing had been applied.
3.2
Sensitivity and power tests
All explosives were tested on sensitivity of impact on the BAM apparatus in

accordance with Stanag 4489 and on sensitivity of friction on the BAM apparatus in
accordance with Stanag 4487 before/after ageing at 65C. Detonation velocity was
determined in accordance with AOP-7 for Comp Che2/00, Comp CHg and THE explosives
before and after ageing at 65C. The test results are listed in Tab 6.
Table 6. Impact sensitivity, friction sensitivity, detonation velocity results
Sample
CHe1/97 non-aged
Aged in tube
Aged in Al+PE foil
Aged in PE foil
CHe 2/00 non-aged
Aged in tube
5
Impact E50
[J]
33,18
27,12
36,65
37,21
36,06
25,94
Friction F50
[N]
303,60
218,25
262,80
234,00
243,00
281,25
Detonation velocity
[m.s-1]
7436,20
533
Aged in Al+PE foil

Aged in PE foil
CHc non-aged
Aged in tube
Aged in Al+PE foil
Aged in PE foil
CHa non-aged
Aged in tube
Aged in Al+PE foil
Aged in PE foil
C4 non-aged
Aged in tube
Aged in Al+PE foil
Aged in PE foil
THE non-aged
Aged in tube
Aged in Al+PE foil
Aged in PE foil
AOa non-aged
Aged in tube
Aged in Al+PE foil
Aged in PE foil
4.
23,34
23,15
30,42
29,98
30,20
31,66
33,66
28,11
28,31
25,81
25,63
27,31
40,49
32,14
24,19
24,02
31,66
24,19
8,55
8,98
9,57
8,74
Czech Republic
256,50
258,75
248,40
218,25
204,75
200,25
313,20
240,75
296,40
272,25
213,75
204,75
214,80
153,00
249,75
218,00
234,00
227,25
188,40
138,00
141,00
144,00
2006
[Content]
7479,50
7448,15
8285,50
8125,30
8222,55
7351,00
7278,20
7184,20
-
DISCUSSION
The results above mentioned prove that during explosive ageing process mainly friction
sensitivity was changing, for all the samples after ageing there was found a friction
sensitivity increase except Comp Che 2/00 explosive, where the sensitivity slightly
decreased. As far as the impact sensitivity either was not changed or (Comp CHc) slightly
decreased.
Impact sensitivity increase was occurred using ageing only at Comp Che 2/00, Comp
Cha explosives and Comp Che1/97 aged in a test tube. In THE, there is not so striking
difference in impact sensitivity before and after ageing process as e. g. in Comp C4, where
ageing process sensitivity has striking slumped. The reason of that is probably high volume
of deterrent in THE and high volume of highly explosive in Comp C4. The intermolecular
activity dominant influences to initiation of detonation by impact in system highly explosive
binding material and conformational stability of molecules.[7] In thermolysis and in
initiation of detonation, those nitramine-parties are primary homolysing that are, in definitive
way, participating with intermolecular potential in correspondent molecular crystals.[8] The
organic additives to individual high explosive are influencing to the sensitivity as well, and
that thanks to their specific properties as e. g. specific heat, capability to dissolve the
energetic material, capability to generate the radicals by impact.[9,10,11]
In terms of the packaging used, the least sensitivity changes were captured on the
samples aged in the Al+PE foil. Detonation velosity slightly increased using ageing of Comp
Che 2/00 while the other two explosive samples proved a slight sensitivity decrease after
ageing.
534
Czech Republic
2006
[Content]
The results of DTA measurements were processed by Kissinger method

following equation:
ln
Tm
[12]
with
E 1
*
+A
R Tm
On figures 1 to 3, there are presented approximate relations between sensitivity

characteristics and Kissinger guideline as well as between detonation velocity and Kissinger
slope. The linear dependencies of these relations for individual high explosives are known
from publications [12,13,9].
45
C-4
40
CHe1/97
35
THE
CHc
30
CHa
25
E (J)
CHe2/00
20
15
y = -0,0016x - 5,6325
2
R = 0,7714
10
AOa
0
-35000
-30000
-25000
-20000
-15000
-10000
Kissinger slope
Fig 1.
Approximative relation between impact sensitivity of aged explosives and

DTA by Kissinger
535
Czech Republic
2006
[Content]
400
350
F (N)
300
CHa
CHe1/97
CHe2/00
250
THE
C-4
CHc
200
y = -0,0203x - 259,21
2
R = 0,8894
150
AOa
100
-61000
-51000
-41000
-31000
-21000
-11000
-1000
Kissingerova smernica
Fig 2.
Approximative relation between friction sensitivity of aged explosives and

DTA by Kissinger
80
70
C-4-N
C-4-Al
C-4-Pe
60
Che-Al
Che-N
THE-Al
THE-Pe
D2 (m.s-2)
50
Che-Pe
THE-N
y = 0,0009x + 70,518
2
R = 0,7457
40
30
20
10
0
-23000
-22000
-21000
-20000
-19000
-18000
-17000
-16000
-15000
Kissingerova smernica
Fig 3.
536
Approximative relation between detonation velocity and DTA by Kissinger
Czech Republic
2006
[Content]
There is possible to establish from figures, that plastic and wax high explosives are not
allocating ideal linear dependency, but trend of linear dependency between impact and
friction sensitivity and results of DTA by Kissinger is evident here. The similar trend of
linear dependency of tested high explosives has expressed in relative relations between
impact sensitivity and friction sensitivity, and between detonation velocity and friction
sensitivity, the figures 4 and 5 are presented this.
300
CHe2/00
250
CHa
THE
CHe1/97
CHc
C-4
200
F (N)
y = 5,7545x + 73,602
2
R = 0,7986
150
AOa
Np10
100
NpD
50
0
0
10
15
20
25
30
35
E (J)
Fig 4.
Approximative relation between friction sensitivity and impact sensitivity of

aged explosives
537
Czech Republic
2006
[Content]
75
70
C-4-N
C-4-Pe
C-4-Al
D2 (m2.s-2)
65
60
Che-Al
55
Che-Pe
Che-N
THE-N
THE-Al
THE-Pe
50
y = 0,12x + 24,84
2
R = 0,8009
45
40
120
140
160
180
200
220
240
260
280
F (N)
Fig 5.
Approximative relation between detonation velocity and friction sensitivity

of aged explosives
It is difficult to reach the more accurately linear dependency in presented relations

especially by two basic reasons. Nevertheless, it is difference composition of blending high
explosives, and in case of the detonation velocity is important that it concerns the correlation
between two profoundly different methods. There are samples of several 100 grams amount
using in the one method (detonation velocity) and milligrams amount using in the second
method (DTA, friction sensitivity).
5.
CONCLUSIONS
On the basis of results of experiment, it is possible to establish, that the tested plastic and
wax phlegmatizated high explosives stayed chemical and thermally stable after artificial
ageing process, their detonation velocity as well as impact sensitivity have succumbed to just
small changes and friction sensitivity has considerably increased in majority of them after
artificial ageing process. The impact sensitivity of Comp Cha a Che has moderately
increased. In term of packing, the Al-PE foil seems the optimal for plastic high explosives as
well as for wax phlegmatized high explosives, which application with was reported the least
of negative changes of explosive sensitivity.
The thermolysis and initiation of blending high explosives detonation by impact and
friction, they are affected primarily by intermolecular interactions and inherence of organic
additives. The work has confirmed, that it is possible to apply the DTA method by Kissinger
to approximately prediction of blending high explosives sensitivity.
538
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
11
S. BREBERA: Vojensk trhaviny a technologie vroby trhavinovch nlo, Uebn texty,

Univerzita Pardubice, 2001
ZHBANOVA,N.N., VLASOV,JU.A., FOMICHEVA,L.V., SHEVLYAGIN,O.V.: Effect of
themperature and time on some characteristics of plasticized octogen,, Russia, Trudy RFYa TsVNIIEF (2004), 6 246-249
MIYOSHI,
HITOSHI.
YOSHII
PLANT, CHUGOKU KAYAKU CO.: Low
vulnerability, high performance and suoerior aging of plastic bonded explosives, Ltd,
Yoshii machi Tano gun, Gunma, Japan. Kayaku Gakkaishi (2002), 63(4), 137 143
CALEY, LEONARD E.; HOFFMAN, D. MARK.: Dynamic mechanical and molecular weight
measurements on polymer bonded explosives from thermally accelerated aging testsIII. Kraton
block copolymer binder and plasticisers, Organic Coatings and Plastics Chemistry (1981), 44
691-6.
SPYCKERELLE,C., FRECHE,A., ECK,G., BOUTRY,G.: Ageing of I-RDX and of
compositions based on I-RDX, Sorgues Plant, EURENCO France, Fr. Intarnational
annual conference of ICT (2004), 35th 38/1 38/10
SELEOVSK, PACHM, HANUS: Changes of sensitiveness of flegmatized figh
explosives after artificial aging, Proc. of the VI.Seminar New Trends in Research of Energetic
Materials, Pardubice, April 2003
S. ZEMAN: New Aspects of the Impact Reactivity of Nitramine.Propellants, Explos. Pyrotech.
18(2), 2000
S. ZEMAN, Z. JALOV: Hears of Fusion of Polynitro Derivatives of Polyazaisowurtzitane
Termochim. Acta 345, 31-38, 2000
S. ZEMAN, M. KRUPKA, J. KO, M. CHOVANCOV, Z. JALOV: A Description of
some Explosive Characteristics of Plastic Explosives. Part III. Proc. of the III.Seminar New
Trends in Research of Energetic Materials, Pardubice, April 2000
YU. YA. MAXIMOV: Termicheskoye razlozenie Geksogena i oktogena (Thermal
Decomposition of Hexogen and Octogen). Tr. Mosk. Khim.-Teknol. Inst. Im. Mendeleeva 53,
73. 1967
R. C. BOWERS, J. B. ROMANS and W. A. ZISMAN: Mechanisms Involved in Impact,
Sensitivity and Desenitization of RDX. Ind. Eng. Chem. Prod. Res. Develop. 12, 1-13, 1973
S. ZEMAN, P. KOHLEK, M. MARANDA: Thermochim. Acta, 398, 185, 2003
M. KRUPKA, M. CHOVANCOV, Z. JALOV, S. ZEMAN: A Description of some
Explosive Characteristics of Plastic Explosives. Part III. Proc. of the III.Seminar New Trends
in Research of Energetic Materials, Pardubice, April 2000
STANAG 4556
STANAG 4491
STANAG 4487
STANAG 4489
AOP-7
539
Czech Republic
2006
[Content]
HEXANITROSTILBENE DENSITY DEPENDENT SENSITIVITY

Clment, Dominik and Rudolf, Karl
Diehl BGT Defence GmbH, Fischbachstrasse 16, 90552 Rthenbach a.d. Pegnitz, Germany
Abstract:
The change in sensitivity of hexanitrostilbene (HNS) pellets with different densities was
investigated by performing standard gap-tests. From these measurements we found that
the sensitivity of the HNS strongly depends on the density of the pellet, as expected.
These results will be presented and discussed in this paper. The density of the HNS was
varied in a range that the pellet material can be expected to be insensitive. However, in
the case of HNS we observed that the material became insensitive at densities larger
than 1.65 g cm-3. Further, we found that the no go-gap-distances decreased from 15
mm to 12 mm inert barrier for densities varying from 1.65 g cm-3 to 1.70 g cm-3.
Keywords:
1.
hexanitrostilbene, HNS, insensitivity, density
INTRODUCTION
The research and development for high explosives nowadays is dedicated to improve the
insensitivity of the materials. It is known that the sensitivity strongly depends on the surface
characteristics of the material. Defects, roughness or cracks influence the sensitivity of the
material [1], but also the density plays an important role [2]. With decreasing density the
amount of cracks or spaces increases. As these defects may be filled with air, upon impact of
a shock wave this volume element may act as a hot-spot and lead to an initiation of the explosive. Therefore, the sensitivity should increase with decreasing density. However, no
systematic experimental study of this effect exists up to now.
2.
EXPERIMENTS
The material used for these experiments was HNS-HBD (hexanitrostilbene high bulk
density). We applied different pressures while preparing the HNS-pellets and used two different pressing tools. One of them was applying the pressure only from one side, while the
second was applying pressure from both sides. The following samples were produced with
the first pressing tool, applying pressure only from one side. This pressing tool is designed in
the way that the length of the pellet is always the same. Therefore, the density of the pellet is
depending on the initial mass filled in the pressing tool. This dependency is shown in figure
1. The density is linearly increasing with increasing mass, as expected from theory.
As mentioned before, we produced the HNS-pellets with two different tools. In figure 2
the differences in the density-pressure-relation is shown. Since one tool applies the pressure
from two sides, the compression and therefore the average density of the material is much
higher than for the other case, where the pressure is applied only from the top. This effect
becomes more important with decreasing pressure applied. However, the observed density of
540
Czech Republic
2006
[Content]
the pellet produced in the one-side-tool exceeds the density of the pellet formed in the
two-side-tool at relatively high specific pressure above ~ 4 kbar. We think that this unexpected behavior might result from a change of the press. To achieve the necessary specific
pressures of 3.8 kbar for the one-side-tool we had to take another press reaching higher
total pressures. To clarify this observation further experiments have to be performed. From
theory we would expect that both pressure-density-relations would converge to one density
limit.
1,71
1,70
Density (g/cm )
1,69
1,68
1,67
1,66
1,65
130
131
132
133
13 4
135
136
137
138
Mass (mg)
Figure 1: Mass-density-relation.
1,68
Density (g/cm )
1,66
1,64
1,62
1,60
pressure from both sides

pressure from one side
1,58
1,5
2,0
2,5
3,0
3 ,5
4,0
4,5
Pressure (kbar)
Figure 2: Higher densities were reached at same specific pressures for the tool applying the pressure
from both sides.
Dominik Clment, Karl Rudolf
541
Czech Republic
2006
[Content]
The change in sensitivity of HNS-pellets with different densities was investigated by performing standard gap-tests. The density of the HNS was varied in a range that the pellet can
be expected to be insensitive. By definition insensitivity means no go-distances of 15 mm
inert barrier (water at ambient temperature in our case) for the gap-tests. The donor charge, a
HWC pellet (94.5% hexogen, 4.5% wax and 1% graphite), is initiated by a detonator and the
resulting shockwave is released into the water. Water is used to ensure that the possibly initiated detonation of the HNS is not initiated by any thermal energy output of the donor
charge, but only by the propagating shockwave. From these measurements we found that the
sensitivity of the HNS strongly depends on the density of the pellet, as expected. Figure 2
shows the results of the gap-tests performed for samples with different densities.
18
Go-values
NoGo-v alues
linear fit (Go)
linear fit (NoGo)
Inert Barrier (mm)
16
14
12
10
1 ,65
1 ,66
1,6 7
1,6 8
1,69
1,70
1,71
Density (g/cm )
Figure 3: Go- and NoGo-gap-distances for samples with different densities.
None of our HNS-pellets detonated at 15 mm inert barrier. The most insensitive samples
showed densities higher than 1.70 g cm-3. For these samples gap-distances for no go of 12
mm inert barrier were found and the most dense sample with = 1.705 g cm-3 even did not
detonate at a gap-distance of 11 mm inert barrier. We tried to fit the measured gap-distances
by a linear relation to predict the densities at which the HNS-pellets will become sensitive
(go-distances > 15 mm inert barrier). We found that the limit should be reached for pellets
with densities around 1.65 g cm-3 (s. figure 3).
542
3.
Czech Republic
2006
[Content]
RESULTS
According to our measurements it is necessary to achieve densities of 1.65 g cm-3 to receive insensitive HNS-pellets. On the other side the HNS may reach densities where no initiation is possible anymore. To ensure that the HNS-pellet is insensitive but can also be initiated with standard techniques the density of the HNS should be between 1.65 g cm-3 and
1.70 g cm-3. In this range of the density the no go-gap-distances decrease from 15 mm to
12 mm inert barrier (water at ambient temperature).
REFERENCES
[1]
[2]
L. BORNE, M. HERRMANN, C. B. SKIDMORE: Microstructure and Morphology, (Energetic Materials Particle Processing and Characterization, U. Teipel, Eds.), Pfinztal, p.333-366, 2005
A. V. D. HEIJDEN, J. TER HORST, J. KENDRICK, K.-J. KIM, H. KRBNER, F. SIMON, U. TEIPEL:
Crystallization: Product Quality of Energetic Materials, (Energetic Materials Particle Processing and Characterization, U. Teipel, Eds.), Pfinztal, p.69-76, 2005
Dominik Clment, Karl Rudolf
543
Czech Republic
2006
[Content]
SELF-SUSTAINING REDUCTIVE DEFLUORINATION OF (CF)n AND

CHARACTERIZATION OF THE REACTION PRODUCTS
Stanislaw Cudzilo *, Mateusz Szala *, Andrzej Huczko **
Michal Bystrzejewski **
*
Military University of Technology, Kaliskiego 2, 00-908 Warsaw, Poland

**
Warsaw University, Pasteur 1, 02-093 Warsaw, Poland
Abstract:
Mixtures of fluorinated graphite (CF)n with different reductants, such as NaN3, B, Ti,
ZrTi, Si, CaSi2, AlSi, TiSi, MoSi2 and CrSi2 were investigated. Reactions between these
substrates are fast and exothermic enough to proceed in a high temperature, selfsustaining regime. Heat effects accompanying the reactions were measured and the
solid reaction products were analyzed. SEM observation revealed the presence of
exfoliated graphite in each case. Nanostructures, like nanofibers, nanorods,
nanospheres and nano-sized crystals, were also present in most of the products. Phase
(XRD) and elemental composition as well as porous structure (N2 adsorption) of chosen
reaction products were also determined.
Keywords:
1.
nanostructures, combustion synthesis
INTRODUCTION
Pyrotechnic mixtures containing powdered metals (e.g. Mg, Al, Ti) and fluorocarbons
polymers (mainly polytetrafluoroethylene, acr. PTFE) are characterized by comparatively
high heat and temperature of combustion. Moreover their combustion products are good
emitters of infrared radiation. Therefore the compositions are commonly applied in infrared
decoy flares and in igniter devices [14].
Less recognized, though, is the fact that they should also be considered as a source of
novel carbon and ceramic nanostructures, which are created in the ultrafast chemical
reactions at very high temperatures in the combustion wave. In particular, this refers to
systems, in which silicon or its compounds or alloys are used as reductants (i.e.
defluorinating agents). Then, the combustion products contain silicon carbide monocrystals
in an unique form of nanofibers [512].
Searching for new, more reactive fluorocarbon oxidizers, E-Ch. Koch has proposed to
use strained fluorocarbons (fluorinated derivatives of tetrahedrane, prismane, cubane and
[60]-fullerane) and graphite fluoride (poly(carbon monofluoride), (CF)n) as fluorine sources.
Owing to an additional energy contained in the strained carbon skeletons, and a lower CF
bond energy in (CF)n than in PTFE, they are much better oxidizers compared to PTFE in
terms of enthalpy and temperature of combustion [13]. The application of caged fluorocarbons
is not possible at present, because some of them are too expensive and others have not been
sensitized so far, but fluorinated graphites are readily available. They can be obtained by the
direct reaction of elemental fluorine with graphite or other carbon materials [14].
Depending on the synthesis conditions, the atomic ratio F/C changes from 0.6 to 1.2.
Poly(carbon monofluroide) (CF)n is a dark-gray, crystalline substance with a density of
544
Self-sustaining reductive defluorination of (CF)n and characterization of the . . .
Czech Republic
2006
[Content]
2.65 g/cm3. Its thermal decomposition starts at 480 oC [13] releasing extremely reactive
carbene (:CF2) and tetrafluoroethylene (C2F4) as the main initial gaseous product of
degradation in vacuum [14].
Fluorinated graphite was first applied as an oxidizer [15] or combustion modifier [16] in
the 80s. and 90s. of the XX century, but only E.-Ch. Koch was the first to publish in 2005
complete results of a thermodynamic analysis of (CF)n defluorination with magnesium,
a mechanism of the reactions as well as radiative and thermochemical properties of a
Mg/(CF)n pyrolant containing 55% (CF)n, 40% Mg and 5% Viton [17]. In the paper, the
author also reports on detection of single walled carbon nanotubes (SWCNT) and carbon
nano carpet rolls (CNCR) in the combustion products. This discovery of great importance
was patented [18] as a new method of synthesis of carbon allotropes (fulleranes, carbon
nanotubes) and various carbon encapsulates.
In the patent the application of several dozens of metals, metalloids, their alloys and
compounds for reduction of fluorinated graphite is claimed, however only results of studies
on reactions between magnesium and (CF)n have been published so far. In the present work,
the following substances: NaN3, B, Ti, ZrTi, Si, CaSi2, AlSi, TiSi, MoSi2 and CrSi2 were
used as reductants of poly(carbon monofluoride). They were selected on the basis of our
previous experience in testing of reductant/PTFE systems [512]. Reductants containing
silicon are numerously represented because the presence of silicon in the reacting mixture
results in synthesis of silicon carbide monocrystals in the form of nanofibers. Boron,
titanium and zirconium create hard and thermally resistant carbides, and under the synthesis
conditions, the compounds may be obtained in the form of nanosized particles. Sodium azide
decomposes in the combustion wave releasing metallic sodium, which violently reacts with
fluorinated graphite producing a lot reactive carbon particles/clusters possible precursors
of carbon nanostructures.
The condensed combustion products were analyzed by X-ray diffraction (XRD) for
crystal structure and characterized using scanning electron microscopy (SEM) for
morphology. Moreover elemental composition and BET specific surface area of the
carbonaceous material obtained by carbonization of fluorinated graphite with sodium azide
were determined.
2.
EXPERIMENT
The compositions of the tested mixtures corresponded to assumed equations of reaction

leading to fluoride of an element at its highest oxidation number, free carbon and carbides of
some elements (SiC, TiC), Table 2. To check whether the assumed reactions are
spontaneous, the standard free enthalpy of the reactions was calculated using standard
enthalpies and entropies of the reactants [19]. In all the cases, for which the needed data were
found, the reaction free enthalpies are highly negative, so that the predicted direction of
reactions is thermodynamically privileged.
Powders of silicon, TiSi and AlSi (Alfa Aesar, particle size approximately 2 m),
boron (Fluka, particle size below 5 m), MoSi2 and CrSi2 (Aldrich, particle size below 5
m), ZrTi (Avocado, particle size below 100 m) and CaSi2 (Fluka, particle size below 100
m) were used to reduce fluorinated graphite (Aldrich, average particle size approximately
5 m). The tested binary compositions were prepared by mixing the substrates taken in an
adequate proportion.
Approximately 5-g samples of the mixtures were placed in a steel crucible and put into
a calorimetric bomb (275 cm3 in volume), which was after that filled with argon, at an initial
pressure of 1.0 MPa. The combustion process was initiated with an electrically heated
Stanislaw Cudzilo, Andrzej Huczko, Mateusz Szala, Michal Bystrzejewski
545
Czech Republic
2006
[Content]
resistance wire. Using the calorimetric bomb as a reactor, it was possible not only to collect
solid combustion products, but also to measure the heat effects accompanying the reactions.
Table 1.
Composition of the tested mixtures, assumed reactions and their standard free enthalpies
Composition
[wt. %]
NaN3 (67.8) / CF (32.2)
B (36.7) / CF (63.3)
Ti (34.0) / CF (66.0)
ZrTi (42.8) / CF (57.2)
Si (31.2) / CF (68.8)
AlSi (26.4) / CF (73.6)
CaSi2 (34.0) / CF (66.0)
TiSi (38.1) / CF (61.9)
MoSi2 (45.0) / CF (55.0)
CrSi2 (33.3) / CF (66.7)
Assumed Reactions
2NaN3 + 2CF = 2C + 2NaF + 3N2
5B + 3CF = B4C + BF3 + 2C
Ti + 3CF = TiF3 + 3C
ZrTi + 7CF = ZrF4 + TiF3 + 7C
2Si + 4CF = SiF4 + SiC + 3C
AlSi + 7CF = AlF3 + SiF4 + SiC + 6C
CaSi2 + 6CF = CaF2 + SiF4 + SiC + 5C
2TiSi + 11CF = TiF3 + 2SiF4 + TiC + 10C
MoSi2 + 6CF = MoF2 + SiC + 5C + SiF4
CrSi2 + 7CF = CrF3 + 6C + SiC + SiF4
Standard Free Enthalpy,

rGo [kJ/kg]
- 2297
- 3762
- 3856
- 4874
- 8292
- 6692
The solid combustion products were removed from the bomb with water. The
suspension was filtered off and the deposit obtained was washed with ethanol. After drying,
the reaction products were analyzed to identify their composition (XRD, Diffractometer
Simens D5000, Cu K) and morphology (SEM). Moreover elemental analysis (PerkinElmer CHNS/O 2400) and nitrogen adsorption measurements (Micromeritics Gemini 2360)
were performed for reaction products of sodium azide with fluorinated graphite.
3.
The results of XRD analysis and the reaction heats (each of them is an average of three
experimental values, standard deviation below 20 kJ/kg) are presented in Table 2.
Table 2.
Heats of reactions and results of XRD analysis
Reactant Composition
Heat of Reaction
[wt. %]
Q [kJ/kg]
NaN3 (67.8) / CF (32.2)
3780
B (36.7) / CF (63.3)
3052
Ti (34.0) / CF (66.0)
4440
ZrTi (42.8) / CF (57.2)
3780
Si (32.1) / CF (68.8)
5030
AlSi (26.6) / CF (73.8)
7160
CaSi2 (66.0) / CF (34.0)
4840
TiSi (38.1) / CF (61.9)
4580
MoSi2 (45.0) / CF (55.0)
3680
CrSi2 (66.7) / CF (33.3)
4210
* (s) strong, (m) medium, (w) weak
Phases Identified
C(s)* NaF(s)
C(s), B4C(s)
C(s), TiC(s), TiF3(s)
C(s), TiC(w), ZrF4(s), TiF3(s), Fe3Zr(m)
C(s), Si(s), SiC(s)
C(s), SiC(m), AlF3(s), Si(w), Al3.21Si0.47(m)
C(s), CaF2(s), SiC(s), Fe3C(w), Si(m)
C(s), TiC(s), SiC(w), Si(m)
C(s), Mo3C2(s) Si(w), Mo4.8Si3C0.8(w)
-
All the reactions are fast and exothermic enough to proceed in a self-sustaining
regime under the experiment conditions. The heat effect varies from c.a. 3050 kJ/kg
(B/(CF)n) to 7160 kJ/kg (AlSi/(CF)n), and this well correlates with the changes in the
standard free enthalpy of the reactions, i.e. the more negative rGo is the higher the reaction
heat is. It means that the reactions are driven rather by decreasing enthalpy than by
increasing entropy.
546
Czech Republic
2006
[Content]
Phase composition of the solid combustion products suggests that the assumed simple
reactions between (CF)n and the reductants are a fair approximation of the real course of
reactions in the tested systems. In each case, the XRD scans showed the presence of graphite
with strong, narrow peaks, so that layered structure of poly(carbon monofluoride) was not
destroyed during defluorination.
When boron, titanium or molybdenum were present in the reacting mixture, the
reaction products contained as it was expected carbides of the elements (B4C, TiC and
Mo3C2). Similarly, the presence of silicon resulted in synthesis of silicon carbide SiC. Other
silicides and mixed silicides/carbides were also created. However silicon did not react
completely in any system it was detected in each sample. The products of reactions
between ZrTi and fluorinated graphite contain zirconium(IV) and titanium(III) fluorides, TiC
and other unidentified phases with high unit cell constants (around 10 nm).
XRD analysis was combined with measurements of atom fluorescence and this
technique revealed that iron was present in each sample in spite of the fact that iron was an
impurity contained only in one reductant CaSi2. In the other samples, iron appeared as a
result of the contact of the hot reacting substances with the crucible and bomb walls made of
stainless steel.
Figures 1 and 2 show SEM micrographs of the raw combustion products.
A
Fig. 1. SEM images of the products of reactions between: NaN3/(CF)n (A), B/(CF)n (B), Ti/(CF)n
(C) and ZrTi/(CF)n (D)
When starting mixture does not contain silicon, SEM images of the combustion
products reveal exfoliated graphite (EG) as a dominating structure, Fig. 1. Apart from this,
nanocrystallites of metal fluorides and carbides (the presence of these compounds was
confirmed with XRD method) are distributed randomly within exfoliated graphite.
In order to characterize the porous structure of the exfoliated graphite, the raw products
of reactions between sodium azide and (CF)n were washed with hot water (to remove NaF),
4
547
Czech Republic
2006
[Content]
and then an isotherm (at 77 K) of nitrogen adsorption for the obtained material was recorded.
The BET specific surface area of the product was calculated to be 125 m2/g. The elemental
analysis showed that the sample contains 90.7% carbon, 3.5% fluorine and 1-1.5% of
nitrogen and hydrogen (adsorbed components of atmosphere). The remaining part of the
sample (c.a. 4%) probably includes sodium (NaF was not removed completely) and oxygen
(moisture and oxygen bound chemically on the surface).
A
Fig. 2. SEM images of the products of reactions between: Si/(CF)n (A), AlSi/(CF)n (B), CaSi2/(CF)n
(C), TiSi/(CF)n (D), MoSi2/(CF)n (E), CrSi2/(CF)n (F).
The application of aluminum, titanium, calcium, molybdenum or chromium silicides

as reductants, leads to reaction products containing exfoliated graphite along with 1-D
nanostructures, Figs 2B-2F. The elongated structures vary in size and shape, however taking
into consideration the phase composition of the materials (Tab. 2) and the results of the
previous investigations of the reactions between PTFE and Si-containing reductants [510], it
can be stated that the observed fibers are SiC fibers. In this context, the morphology of
548
Czech Republic
2006
[Content]
reaction products in mixture of pure silicon with fluorinated graphite (Fig. 2A) is especially
unexpected. In any SEM micrographs of the products (even at low magnification), fibers
have not been detected. To find the reasons for this, more detailed study is needed, which is
in progress.
4.
CONCLUSIONS
Poly(carbon monofluoride) (CF)n, like PTFE, is an reactive oxidizer, capable of higly

exothermic, self-sustaining reactions with many reductants, e.g. NaN3, B, Ti, ZrTi, Si, CaSi2,
AlSi, TiSi, MoSi2 and CrSi2. Regardless the kind of reductant, the reaction products include
exfoliated graphite (EG). In spite of the differences in the reaction heat and in the
temperature of condensation of the formed fluorides, the layered structure of fluorinated
graphite is not significantly destroyed during defluorination in the combustion wave. The
reaction products do not contain carbon nanostructures in a form of nanotubes or nanorolls.
As a result of the self-sustaining reactions between fluorinated graphite and elements
(B, Ti, Si) capable of creating thermally resistant carbides (B4C, TiC, SiC), nanosized
particles of these compounds are formed. If silicon is present in the reacting mixture, the
products also contain 1-D nanostructures. The fibers are formed with the highest yield when
not pure silicon but metal silicides are used as reductnats. However, the fibers are not so
regular in shape and are not produced so efficiently as it was observed when PTFE was used
as an oxidizer [512].
The products of reactions of sodium azide with (CF)n contain above 90% of carbon,
in the form of exfoliated graphite. Its specific surface area is equal to 125 m2/g, and this is
even several times as large as the surface area of a product obtained by heating graphite
intercalates [20].
In summary, the obtained results demonstrate that the combustion synthesis in
reductant/fluorinated graphite systems offers great potential for generating materials having
unique phase composition and structure.
Acknowledgment
This research was supported by the Ministry of Education and Science (Poland) through grants
3 T08D 012 28 (Warsaw University) and PBS 700 (Military University of Technology).
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
E-CH. KOCH, Review on Pyrotechnic Aerial Infrared Decoy Flares, Propellants, Explosives,
Pyrotechnics, Vol. 26, p. 3, 2001.
E-CH. KOCH, Metal/fluorocarbon Pyrolants, III, Development and Application of MTV,
Propellants, Explosives, Pyrotechnics, Vol. 27, p. 262, 2002.
E-CH. KOCH, Metal/fluorocarbon Pyrolants, IV, Thermochemical and Combustion Behavior of
MTV, Propellants, Explosives, Pyrotechnics, Vol. 27, p. 340, 2002.
T. KUWAHARA, et all, Ignition Characteristics of Ti/Teflon Pyrolants: safety igniter with
partition, Propellants, Explosives, Pyrotechnics, Vol. 30, p. 425, 2005.
A. HUCZKO, H. LANGE, G. CHOJECKI, S. CUDZILO, Y. QIU ZHU, H. W. KROTO, D. R. M.
WALTON, Synthesis of novel nanostructures by metal-polytetrafluoroethene thermolysis,
Journal of Physical Chemistry B, Vol. 107, p. 2519-2524, 2003.
S. CUDZILO, J. BLADEK, W KICINSKI, A. ARCISZEWSKA, A. HUCZKO, Synteza Materiaw
Wglowych w Fali Spalania i Ich Wykorzystanie w Analizie Chemicznej. Cz. I. Otrzymywanie i
Badania Waciwoci Fizycznych Materiaw Wglowych, Biul. WAT, Vol. 53, p. 133, 2004.
549
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
550
Czech Republic
2006
[Content]
A. HUCZKO, H. LANGE, M. BYSTRZEJEWSKI, M. H. RUMMELI, S. CUDZILO, Studies on

Spontaneous Formation of 1D Nanocrystals of Silicon Carbide, Crystal Research and
Technology, Vol. 40, p. 334, 2005.
S. CUDZILO, M. BYSTRZEJEWSKI, H. LANGE. A. HUCZKO, Spontaneous Formation of CarbonCased Nanostructures by Thermolysis-Induced Carbonization of Halocarbons, Carbon, Vol.
43, p. 1778, 2005.
M. RUMMELI, E. BOROWIAK-PALEN, T. GEMMING, A. HUCZKO, M. KNUPFER, S. CUDZILO, R.
KALENCZUK, T. PICHLER, Modification of SiC Based Nanorods via a Hydrogenated
Annealing Process, Synthetic Metals, Vol. 153, p. 349, 2005.
A. HUCZKO, M. BYSTRZEJEWSKI, H. LANGE, A. FABIANOWSKA, S. CUDZILO, A. PANAS, M.
SZALA, Combustion Synthesis as a Novel Method for Production of 1-D SiC Nanostructures,
Journal of Physical Chemistry B, Vol. 109, p. 16244, 2005.
S. CUDZILO, J. CHOMA, M. SZALA, A. HUCZKO, Characterization of Carbonaceous Materials
Produced by Rreduction of Halocarbons with Sodium Azide, Polish Journal of Applied
Chemistry, Vol. 49, p. 205, 2005.
A. HUCZKO, S. CUDZILO, M. BYSTRZEJEWSKI, H. LANGE, M. SZALA, Otrzymywanie
Nanostrukturalnych Materiaw na Drodze Syntezy Spaleniowej, Biul. WAT, No. 54, 57, 2005.
E-CH. KOCH, Metal/Fluorocarbon Pyrolants: V. Theoretical Evaluation of the Combustion
Performance of MTV Pyrolants Based on Strained Fluorocarbons, Propellants, Explosives,
N. WATANBE, S. KOYAMA, H. IMOTO, Thermal Decomposition Products of Graphite Fluoride,
(CF)n in a Vacuum, Bull. Chem. Soc. Jpn., Vol. 53, p. 2731, 1980.
J. L. FIELDS, J. D. MARTIN, US Patent 6736912, 2004.
T. K. LIU, I. M. SHUY, Y. S. HSIA, Effect of Fluorinated Graphite on Combustion of Boron
and Boron Based Fuel Rich Propellants, J. Propul. Power, Vol. 26, p. 12, 1996.
E-CH. KOCH, Metal/fluorocarbon Pyrolants: VI. Combustion Bbehaviour and Radiation
Properties of Magnesium/Poly(Carbon Monofluoride) Pyrolant, Propellants, Explosives,
E-CH. KOCH, Deutsches Patent, DE 10122750, 2001.
J. BARIN, Thermochemical Data for Pure Substances, 3rd Edition, VCH, Weinheim, 1995.
J. H. HAN, K. W. CHO, K. H. LEE, H. KIM, Porous Graphite Matrix for Chemical Heat Pumps,
Carbon, Vol. 36, p. 1801, 1998.
Czech Republic
2006
[Content]
INSENSITIVE ENVIRONMENTALLY-COMPATIBLE AND NON-TOXIC

REPLACEMENTS FOR HEXOGEN IN HIGHLY ENERGETIC
MATERIALS - NEW METHOD OF SYNTHESIS OF
THE 1,5-DIAMINO-1H-TETRAZOLE (DAT)
Chaza Darwich, Thomas M. Klaptke
Department of Chemistry and Biochemistry
Ludwig-Maximilian University of Munich
Butenandtstr. 5 - 13 (Haus D)
D-81377 Munich, Germany
Abstract:
This paper consists of an attempt to improve the synthesis of 1,5-diaminotetrazole
(DAT) by aminating 5-aminotetrazole (5-AT) with a suitable aminating reagent.
First there are described the steps of the synthesis and the characterization of the
compounds involved in this attempt.
Keywords:
1.
Nitrophenoxyphtalimides, Nitrophenylhydroxylamines,
5-aminotetrazole, amination, characterization (Melting point, Raman,
NMR, MS, EA, X-RAY).
INTRODUCTION
1.1
Synthesis Overview
Because of the difficult accessibility of 1,5-diaminotetrazole (1,5-DAT), only few

synthetic methods for its preparation have been described in the literature. This compound,
as all aminotetrazoles, has the highest content of nitrogen among the organic substances:
about 84% and due to the aromaticity, it exibits a relatively high thermal stability.
In addition, 1,5-DAT presents a high energy of formation, for this reason it can be used
as a valuable intermediate in the preparation of high-energy-density materials.
NH2
5
NH2
4 N
N1
N
1,5-diamino-1H-tetrazole (1,5-DAT)
A milestone in the synthesis of DAT was described by Raap[1] who reacted the sodium
salt of 5-aminotetrazole with hydroxylamine-O-sulfonic acid and obtained a mixture of
1,5-diaminotetrazole (1,5-DAT) and 2,5-diaminotetrazole (2,5-DAT) where the first
constitutes 8,5% of the reaction yield.
Chaza Darwich, Thomas Klaptke
551

NH2
N
2006
H
NaOH / HOSA
- Na2SO4
N
N
[Content]
NH2
NH2
Czech Republic
NH2
N
+
H2N
5-AT
1,5-DAT (8,5%)
2,5-DAT
Previously to Raap, Gaponik & al.[2] improved the synthesis reported by Stolle & al[3].
By reacting thiosemicarbazide with lead oxide and sodium azide in a CO2 atmosphere, a
carbodiimide intermediate is formed and reacts in-situ with HN3 to lead to 1,5-DAT.
NH2
S
PbO /
NH
- PbS / H2O
H2 N
HN3
HN = C = N
NH2
NH2
NH2
Unfortunately, this reaction leads to large amounts of lead azide as the side-product
which makes this synthesis problematic for an industrial scale.
The most recent approach to this synthesis was made in our research group[4] and it
eliminates the formation of highly-explosive lead azide and produces a substancially greater
yield. After diazotation of diaminoguanodinium chloride, the reaction mixture is brought to
pH 8 to deprotonate the intermediate formed which cyclises leading to 1,5-DAT in a yield of
58%.
NH2
NH2
HN
NH
H2 N
NH2
Cl
NaNO2
HCl
N3
Na2CO3
40C
NH
H2 N
NH2
N
N
N
NH2
N
Cl
Diaminoguanidinium chloride
1,5-DAT (58%)
Nevertheless, the synthesis conditions must be perfectly controlled because the reaction
of nitrous acid with aminoguanidinium is strongly dependent of the pH value as well as the
reactants amounts, otherwise it might lead to the azide derivative which is a very explosive
by-product.
1.2
Aims
The aim of this project is to make the synthetic procedure for DAT safer and with less
steps. The idea is to find a suitable reagent which can simply aminate 5-aminotetrazole. An
efficient synthesis of two nitrophenylhydroxylamines is undergone:
552
2,4-dinitrophenylhydroxylamine (DNPH)
2,4,6-trinitrophenylhydroxylamine (TNPH).
Insensitive Environmentally-compatible and Non-toxic Replacements for RDX in . . .
Czech Republic
2006
[Content]
By comparing these two aminating reagents, theoratically the second one is more
advantageous because it has three nitro electron-withdrawing groups, so the amination effect
should be stronger and should lead to a higher amination rate and a higher yield.
2.
EXPERIMENTS
2.3
Synthesis of O-nitrophenoxyphtalimide derivatives
As a first step, the procedure requires the synthesis of the hydroxylamines precursors[5]
namely:
2,4-dinitro-O-phenyoxyphtalimide (DNPP)
2,4,6-trinitro-O-phenyoxyphtalimide (TNPP)
O
NO2
OH
O2N
(CH3)2-CO
O2N
NO2
Et3N, r.t.
O
O
O
NO2
OH
Cl
O2N
(CH3)2-CO
O2N
NO2
Et3N, 5C
2.4
NO2
NO2
Synthesis of O-(nitrophenyl)-hydroxylamine derivatives
The reaction between 2,4-dinitro-O-phenyoxyphtalimide and hydrazine hydrate in the

presence of HCl at OC rendered after deprotection two compounds A and B[6].
O
O2N
NO2
NH
NO2
N2H5OH
MeOH, CH2Cl2
HCl, OC
O2N
NH2
NH
A precipitates in the reaction medium and it is filtrated. Afterwards, B is extracted from

the aqueous phase with dichloromethane and the crude obtained after removing the solvent
can be purified by recrystallisation from acetonitrile leading to a pure sample of B:
2,4-dinitrophenylhydroxylamine (DNPH).
The preparation of 2,4,6-trinitrophenylhydroxylamine (TNPH) by the same
hydrazinolysis reaction (same conditions) does not seem to be successful because several
by-products are formed but not the desired compound. An attempt to improve the
deprotection under milder conditions is work in progress.
553
2.5
Czech Republic
2006
[Content]
Amination reaction
5-aminotetrazole
(5-AT)
is
espacially
soluble
in
alcohols
but
2,4-dinitrophenylhydroxylamine (DNPH) decomposes into 2,4-dinitrophenol when refluxing
in these solvents[7]. This makes the choice of the solvent relatively limited. Therefore, the
amination of 5-AT by DNPH was carried out in DMF at 60 to 70C for 50 hours and should
normally lead to these salts:
NH2
NO2
5
4
N
1N
N
2
H
H2N
N
3
NO2
5-AT
2,4-O-DNPH
DMF, 60-70C, 50 h
NH2
NH2
NO2
5
H2N
1
N
N
2
N
3
1,5 DATH
3.
NO2
H2N
+
NO2
5
4
N
1
N
N
2
N
3
2,5 DATH+
2,4-DNPO-
H
O
NO2
2,4-DNPO-

3.1
O-Nitrophenoxyphtalimide derivatives
3.1.1. 2,4-dinitro-O-phenyoxyphtalimide
DNPP is obtained in a yield of 85%, it was purified by recrystallization from EtOH and
fully characterized as follows:
Melting point (Bchi B-540): [188.0-188.8C] which agrees with the reported value of
187.5C
.
Mass spectrometry (DEI, 70eV): M+ : 328.9, m/z: 90(100%), 283(99), 104(71),

76(52.5), 130(50.7), 147(43.8), 63(36.5)
Elemental Analysis: C14H7N3O7 Calc.(%) N: 12.76, C: 51.07, H: 2.14
Found (%) N: 12.75, C: 50.97, H: 2.27
NMR shifts: (DMSO-d6)
O
O2N
H3
c
b
554
4
6
H6
NO2
5
H5
O
Insensitive Environmentally-compatible
and Non-toxic Replacements for RDX in . . .
Czech Republic
2006
[Content]
1H (ppm): H3 8.90 (1H, d, J3-5 = 2.9 Hz)

H5 8.46 (1H, dd, J5-3 = 2.9 Hz, J5-6 = 8.8 Hz)
H6 8.12 (1H, d, J6-5 = 8.8 Hz)
Harom 7.94 (4H, m)
13C (ppm): C=O 162.8, C1 155.7, C4 143.5, C2 137.1, Cc 135.9, C5 130.6, Cb 129.5,
Ca 124.5, C3 122.7, C6 117.1.
-NO2
1354
Raman:
-NO2
=C-H (disubst. arom)
DNFB
N=O
3088
836
1540
1615
1801
3090
1357
1607
1539
Int
1187 1020
837
DNPP
714
C=O
1778
3087
C=O
1714
478
1187
1017
1609
600
NHP
4000,0
3000
Fig 1.
2000
1500
Raman Shift / cm-1
1000
500
50,0
Raman spectra superposition (100mW, 100 scans, 5 mm tube, r.t.):

dinitrofluorobenzene (DNFB), N-hydroxyphtalimide (NHP) (starting
materials) and dinitrophenoxyphtalimide (DNPP) (reactions product).
3.1.2. 2,4,6-trinitro-O-phenyoxyphtalimide
TNPP is obtained in a yield of 82%, it was purified by recrystallization from ACN and
fully characterized as follows:
Melting point (Bchi B-540): [212.3-213.7C] matching with the literature value of
213.5C
.
Mass spectrometry (DEI, 70eV): M+ : 374.0, m/z : 90(100%), 104(55), 76(39.8),

162(32.4), 146(28.2), 30(27), 130(22.7)
Elemental Analysis: C14H6N4O9 Calc.(%) N: 14.97, C: 44.93, H: 1.61
Found (%) N: 15.00, C: 45.11, H: 1.89
O2N
b
a
c
N
H3
2
O
NO2
Chaza Darwich, Thomas bKlaptke

O
2
O2N
3
H3
555
Czech Republic
2006
[Content]
1H (ppm): Harom 7.92 (4H, m), H3 9.19 (2H,s)

13C (ppm): C=O 161.8, C4 143.3, Cc 136.0, C2 134.5, Cb 127.9, C1 125.1, Ca 124.4,
C3 122.9
C=O
1778
Raman:
3087
C=O
1714
NHP
3087
Int
1017
714
1187
1609
478
1390
1607 1551 1349 1179

1801
1012 826
697
TNPP
1350 -NO2
N=O
1604 1555
824
3089
TNCB
4500,0
4000
Fig 2.
3000
2000
1500
Raman Shift / cm-1
1000
500
40,0
Raman spectra superposition (100mW, 100 scans, 5 mm tube, r.t.):

trinitrochlorobenzene (TNCB), N-hydroxyphtalimide (NHP) (starting
materials) and Trinitrophenoxyphtalimide (TNPP) (reactions product).
X-RAY (Oxford Xcalibur 3 CCD):

Suitable crystals for X-RAY analysis were grown in acetonitrile and the structure
obtained is the following:
109.912(9)
Fig 3.
556
Formula unit and labeling scheme for TNPP (Diamond 3 Plot, thermal
ellipsoid represents 50% probability).
Czech Republic
Formula sum
C14H6N4O9
Formula weight
374.223 g/mol
Crystal system
triclinic
Space-group
P -1 (2)
Cell parameters
a = 8.105(1)
= 81.54(1)
Cell volume
720.3(5) 3
Calculated density
1.725 g/cm3
Rint
3.9%
R2
10.9%
3.2
O-(Nitrophenyl)-hydroxylamine derivatives
2006
b = 8.1725(9)
= 85.51(1)
[Content]
c = 12.591(1)
= 60.83(1)
3.2.1. 2,4-dinitrophenylhydroxylamine
The hydrazinolysis of DNPP or B presents some drawbacks: reprecipitation of A which
obliges to filtrate several times, this adds to the problem that B decomposes in the aqueous
phase and must be immediately extracted. In consequence, the yield of DNPH is only about
40%. DNPH was purified by recrystallisation from acetonitrile, its characterization gave the
following:
Melting point (Bchi B-540): [111.1-112.2C] which agrees with the reported value of
112C.
.
Mass spectrometry (DEI, 70eV): M+ : 199.0, m/z : 181(100%), 184(47), 63(40),

56(26.8), 77(22.4), 62(21)
H3
NO2
3
O2N
1
5
H5
NH2
H6
1H (ppm): H3 8.63 (1H, d)

H5 8.10 (1H, dd)
H6 7.88 (1H, d)
NH2 6.42 (2H, s)
13C (ppm): C1 156.7, C4 137.5,C2 134.5, C5 129.2, C3 123.6 ,C6 117.6
557
Czech Republic
Raman:
2006
[Content]
-NO2
1344
2,20
2,0
1316
1,8
1,6
DNPH
1,4
1370
N=O
1612
1,2
Int
1,0
0,8
1518
3261
0,4
3319
0,2
-0,02
4500,0
838
1285
N-H
0,6
1137
4000
Fig 4.
3106
3000
2000
1500
Raman Shift / cm-1
Raman spectra (200mW, 200 scans,

dinitrophenylhydroxylamine (DNPH).
1000
500
mm
tube,
50,0
r.t.)
of
X-RAY (Oxford Xcalibur 3 CCD):

The crystals were grown in acetonitrile and the structure obtained is the following:
113.45(11)
Fig 5.
Formula unit and labeling scheme for DNPH (Diamond 3 Plot, thermal
ellipsoid represents 50% probability).
Formula sum
C6H5N3O5
Formula weight
199.13 g/mol
Crystal system
monoclinic
Space-group
P 21/a (14)
Cell parameters
a = 10.0146(7) b = 14.558(1)
c = 10.6425(7)
= 91.77(1)
558
Czech Republic
Cell volume
1550.9(3) 3
Calculated density
1.706 g/cm3
R1
4.74%
wR2
14.54%
GOOF
1.109
2006
[Content]
3.2.2. 2,4,6-trinitrophenylhydroxylamine (picrylhydroxylamine)

We are currently working on the deprotection of TNPP in milder conditions and the
study is still outstanding.
3.3
DAT Salts
3.3.1. DATH+ DNPOAfter removing DMF under reduced pressure, the residue was analysed by Raman and
we obtained the spectrum below:
-NO2
1340
Raman:
2,15
2,0
1,8
1,6
1,4
1,2
1265
Int 1,0
840
0,8
C=N- 1540
1626
0,6
0,4
0,2
N-H
3318
3213
1145
434

3065
-0,02
4000,0
3000
Fig 6.
2000
1500
Raman Shift / cm-1
1000
500
50,0
Raman spectra (200mW, 200 scans, 5 mm tube, r.t.) of DATH+DNPO-.
This spectrum shows resembling reflexions compared to those of the starting materials.
The residue was also analysed by Mass Spectrometry (FAB+, xenon, 6 keV, m-NBA
matrix):
559
Czech Republic
2006
[Content]
Mass spectrometry: 101 [DAT+H]+, 284 [CH5N6C6H3N2O5]

We are currently working on growing suitable crystals for X-RAY measurement.
4.
CONCLUSION
In the literature, the amination effect of 2,4-dinitrophenylhydroxylamine (DNPH) is

reported as a slow reaction, however the amination of 5-aminotetrazole (5-AT) was
successful but it would be also interesting to try the amination with the
O-picrylhydroxylamine. The next steps would be also to prepare and to explore special salts
of DAT.
5.
ACKNOWLEDGMENTS
Financial support of this work by the University of Munich (LMU), the Fonds der
Chemischen Industrie and the European Research Office (ERO) of the U.S. Army Research
Laboratory (ARL) under contract no. N 62558-05-C-0027 is gratefully acknowledged. We
are indebted to and thank Dipl.-Chem. Jrg Stierstorfer for the X-ray work. CD also wants to
thank the LMU Munich for a postdoctoral scholarship.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
560
R.
RAAP:
Amination
of
tetrazoles
with
hydroxylamine-O-sulfonic
acid:
1- and 2-aminotetrazoles, Can. J. Chem, No. 47(19), 3677-81, 1969
P. N.GAPONIK and V. P. KARAVAI: Synthesis and properties of 1,5-diaminotetrazole,
Khim. Geterotsikl. Soedin., No. (12), 1683-6, 1984
R. STOLLE, H. NETZ, O. KRAMER, S. ROTHSCHILD, E. ERBE and O. SCHICK:
Derivatives of 1-aminotetrazoles, J. Prak. Chem., No. 138, 1-17, 1933
(a) J. C. GALVEZ-RUIZ, G. HOLL, K. KARAGHIOSOFF, T. M. KLAPTKE,
K. LHNWITZ, P. MAYER, H. NTH, K. POLBORN, C. J. ROHBOGNER, M. SUTER and
J. J. WEIGAND: Derivatives of 1,5-diamino-1H-tetrazole: a new family of energetic
heterocyclic-based salts, Inorg. Chem., No. 44(12), 4237-4253, 2005
(b) J. C. GALVEZ-RUIZ, G. HOLL, K. KARAGHIOSOFF, T. M. KLAPTKE,
K. LHNWITZ, P. MAYER, H. NTH, K. POLBORN, C. J. ROHBOGNER,
M. SUTER and J. J. WEIGAND: Derivatives of 1,5-diamino-1H-tetrazole: a new family of
energetic heterocyclic-based salts, Inorg. Chem., No.44(14), 5192-5208, 2005
A. ROUGNY and M. DAUDON: Use of N-hydroxylimides for the synthesis of primary
alkoxylamines, Bulletin de la Socit Chimique de France, No. 5-6, 833-8, 1976
C. LEGAULT and A. B. CHARETTE: Highly Efficient Synthesis of
O-(2,4-Dinitrophenyl)hydroxylamine. Application to the Synthesis of Substituted
N-Benzoyliminopyridinium Ylides, J. Org. Chem., No. 68(18), 7119-22, 2003
A. S. RADHAKRISHNA, G. M. LOUDON and M. J. MILLER: Amination of ester enolates
with O-(2,4-dinitrophenyl)hydroxylamine, J. Org. Chem., No. 44(26), 4836-41, 1979
Czech Republic
2006
[Content]
THE NITRATION REACTION OF ISOWURTZITANE ACYL

DERIVATIVES
Maciej Duda, Pawe Maksimowski and Wincenty Skupiski
Division of Highenergetic Materials, Faculty of Chemistry, Warsaw University of Technology
(Politechnika), Noakowskiego 3, 00-664 Warszawa, Poland
Abstract
Preparation methods via nitration of isowurtzitane acyl derivatives, physicochemical
and spectral properties of a new isowurtzitane compound - 4,6,8,10,12-pentanitro-2acetyl-2,4,6,8,10,12-hexaatetracyclo [5.5.0.05,903,11] dodecane ( PNAIW), are described.
INTRODUCTION
A quest for organic compounds which, owing to their structure, crystallize with the
highest possible spatial packing have resulted in the development of synthesis of compounds
of cage structure. Instances of such compounds are: adamantants, cubans, wurtzitans and
isowurzitans. Aside from their characteristic feature mentioned, strain bonds are present in
the structure of these compound which additionally exhibit some properties characteristic of
aromaticity.
The compounds has for long been am attractive research object, as they provide a great
many new data for theory of chemistry. Moreover, they already find practical uses among
others as high energy materials. Ranking in this group are nitro derivatives of isowurtzitan,
e.g: 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12,hexaatetracyclo[5.5.9.95,903,11] dodecane i.e.
HNIW. Numerous reports and patents are devoted to the preparation of this compound.
Our studies on the synthesis of this compound showed that under specific conditions of
the HNIW synthesis an intermediate product, not described yet in literature, can be obtained
in
variable
yield,
viz.
4,6,8,10,12-pentanitro-2-acetyl-2,4,6,8,10,12hexaatetracyclo[5.5.0.05,903,11]dodecane i.e. PNAIW. The synthesis conditions, along with
major properties of PNAIW and its spectral data, are reported in this communication.
Maciej Duda, Pawe Maksimowski, Wincenty Skupiski
561
Czech Republic
2006
[Content]
EXPERIMENTAL
Preparation of PNAIW as a reference compound
Reactions of Nitration with Conventional Nitrating Mixtures

The reactions of nitration of the isowurtzitane derivatives under study were carried
out in a round-bottom flask provided with a mechanical agitator, a dropping funnel and a
cooling bath. In each experiment 12 g (0.2 mol) 99% nitric acid was placed in the flask. On
starting the agitator the contents of the flask was cooled to 5C with a dry ice/ acetone
freezing mixture. At that temperature, in vigorous stirring, 0.002 mol of isowurtzitane
derivative was being added portionwise for 5 min.
Subsequently, a suitable amount of the dehydrating agent was added through the
dropping funnel for 10 min at 5C. Next, the system was brought to the adequate process
temperature and nitrated. The process parameters for individual varieties are gathered in
Table 1. On completion of the nitration process the reaction mixture was allowed to cool and
was poured over ice. The precipitate formed was filtered off and washed with methanol. Pure
PNAIW was obtained after the crude product recrystallization from acetone.
Elementary Analysis: C8H9N11O11; calculated: C 22.10%, H 2.12%, N 35.34%;
obtained: C 22.08%, H 2.08%, N 35.40%.
Table 1. . Parameters of the Reaction of Nitration of Isowurtzitane Derivatives
Extra
Reagent
Reaction Time
Weight of Derivative [g]
99% H2SO4
CH3COOH
20% oleum
N2O4
TADBIW
5h
1.03
0.05 mol (4.6 g)
Derivative
TADNOIW
5h
0.79
0.2 mol (19.6 g)
0.2 mol (12 g)
23.5 g
TAIW
1h
0.67
0.2 mol (19.6 g)
TADBIW- tetraacetyldibenzylhexaazaisowurtzitane
TADNOIW tetraacetyldinitrosohexaazaisowurtzitane
TAIW - tetracetylhexaazaisowurtzitane
562
Czech Republic
2006
[Content]
TADBIW Nitration with a Mixture of 99% Nitric Acid and 99% H2SO4 with
a 20-% Addition of NaNO2
In a reaction flask of a 50-cm3 capacity equipped with a thermometer, a mechanical
agitator, a dropping funnel and a cooling bath, 12 g nitric acid (0.2 mol) was placed. Once
the stirring was started the flask contents was cooled to 5C with the dry ice / acetone
mixture. At that temperature was added, on vigorous stirring, 1.03 g (0.02 mol) TADBIW
during 5 min. Then 19.6 g (0.2 mol) 99% H2SO4 was added dropwise. The sulfuric acid
adding was continued for 10 min while maintaining the reaction temperature below 5C.
On addition of reactants the funnel was replaced by a reflux condenser, the cooling bath
by a hemispherical heater and the reaction system was brought to a required temperature of
80C and then the reaction was carried out for 5 h. After that time the reaction mixture was
cooled to 5C (on a cooling bath) and 2.5 g NaNO2 was added. Then the temperature was
made to rise to 25C and under these conditions the reaction was carried out for 20 h.
On completion of the reaction the reaction mixture was diluted with water (with ice).
The reaction product was a light-yellow solid which was filtered off, washed with methanol
and dried in the air.
Nitration of TADBIW with NO2BF4 in Sulfolane

In a round-bottom flask of a 250-cm3 capacity equipped with a thermometer, a
mechanical stirrer, and a cooling bath were placed a solution of 5.2 g TADBIW in 100 cm3
sulfolane and 0.4 cm3 H2O. Once the agitator was started 21.6 g tetrafluoroborane was
quickly added while maintaining the reaction temperature within 25C and the reaction was
continued for 3 h at 25C. After that time the cooling bath was replace by a hemispherical
heater and for the next 3 h the reaction was conducted at 5560C.
Then the reaction mixture was cooled to 10C in a cooling bath and transferred into a
three-neck round-bottom flask of a 2-dm3 capacity equipped with a cooling bath, a
thermometer, a dropping funnel and a mechanical agitator. Into the post-reaction solution 1.5
dm3 H2O was added dropwise while maintaining the temperature below 10C. The product
obtained was filtered off.
Nitration of TADBIW Using N2O4 and Then with 99.1% HNO3

In a three-neck flask of a 50-cm3 capacity equipped with a mechanical agitator, a
thermometer and a cooling bath were placed 12.6 cm3 dinitrogen tetroxide and 0.5 cm3 H2O.
Then, on slow stirring at 0C for 10 min 3 g TADBIW was added portionwise. As the
reactants become mixed, the reaction was continued at the same temperature for 1 h, and
next the contents of the flask was heated to 20C and the reaction was conducted for 20 h at
a temperature not to exceed 25C.
After that time the reaction mixture was allowed to cool to 5C and at that temperature
was added dropwise 50 cm3 99.1% HNO3. As the nitrating agent was added the dropping
funnel was replaced by a reflux condenser and the cooling bath by the heater. The flask
contents was then heated to 80C and under a slow water flow through the reflux condenser
the excess N2O4 was driven off. As the nitrogen oxides were removed the reaction was
continued at the same temperature for 2 h (under a rapid flow of water through the reflux
3
563
Czech Republic
2006
[Content]
condenser), whereupon the flask contents was diluted by pouring it over ice. The precipitate
formed was filtered off and washed with methanol.
Analysis of Product
For the determination of the content of individual products of the reaction of nitration of
isowurtzitane derivatives liquid chromatography technique was used with a Shimadzu model
10A instrument and a Supelco 300-5 C018 150 mm column and a 50:50 acetonitrile/ water
eluent system using a UV detector at a wavelength of 245 nm.
IR measurements were carried out with a Specord M-80 spectrometer. The NMR
spectrum was recorded with a Gemini 200BB apparatus and the DTA TG thermograph
was recorded with a Derivatograph-C
Results
Pentanitroacetylisowurtzitane can be obtained after Reaction 1:
W(X)2 +
HNO3 (HNO3/B)
PNIAW + HNIW + TEX+Mono-OXA
(1)
X = H, CH2Ph, NO: B = H2SO4, CH3COOH; N2O4

from tetraacetylisowurtzitane derivatives of general formula W(X)2 fig.1, with the
following substituents at positions 4 and 10: H, CH2Ph or NO (Fig. 1) by the action of
nitrating mixtures on the derivatives. The nitrating mixtures, aside from 99% nitric acid, may
contain sulfuric acid, acetic acid, nitrogen dioxide. Nitric acid alone may also be taken as a
nitrating reagent.
O
O2N
XN
O2N
NX
NO2
N
NO2 O2N
O2N
N
NO2
HNIW
O2 N
NO2 O2N
O2 N
NO2
NO2
Mono-oxa
TEX
W(X)2
Fig. 1. Spatial structure of isowurtzitane W(X)2 and HNIW, Mono-oxa and TEX
564
Czech Republic
2006
[Content]
Results obtained are reviewed in Table 2.
Substrate
Nitrating
mixture
Reaction
temperature
[oC]
HNIW
[%]
TEX
[%]
NonoOXA
[%]
PentaNitro
[%]
NO2BF4
sulfolane
25 - 60
90,3
5,2
25 80
80,7
1,4
8,4
3,9
I HNO3
H2SO4
II NaNO2
25 -80
4,9
1,0
1,2
11%
HNO3
N2O4
40
5,4
2,8
28,4
HNO3
15,8
30,8
HNO3
H2SO4
75,3
12,4
HNO3
oleum
80,8
4,8
HNO3
CH3COOH
12,9
21,5
HNO3
H2SO4
20
10,2
__
__
40,3
HNO3
5.9
__
__
75.0
I N2O4
TADBIW
TADNOIW
TAIW
II HNO3
Table 2. Conditions and results of PNAIW preparation

Nitration of TADBIW proceeds with difficulty. If 99% nitric acid and a nitrating mixture
containing this acid and 96% H2SO4 is used, the mixture of the reaction products contains as
detected chromatographically up to 5% PNAIW. A nitrating mixture of HNO3/N2O4 allowed
to produce PNAIW in a 28.4-% yield when the reactions were carried out at 40C for 5 h.
Contaminations of the product, which were found to be HNIW and Nono-OXA, were
removed during crystallization from acetone. The contaminations do not occur if the
reactions are carried out at lower temperatures, yet the PNAIW is then obtained in a lower
yield.
A more convenient compound to obtain PANIW is TADNOIW which, if nitrated with a
99-% HNO3 alone at 0C, after 3 h affords 30.8% of this compound. This nitration reaction
is accompanied by the formation of HNIW (15.8% yield) which can be separated by
crystallization from acetone. A higher reaction temperature and the use of an HNO3/H2SO4
565
Czech Republic
2006
[Content]
mixture lowers the PNAIW yield in favor of HNIW. Analogous results are obtained if a
HNO3/CH3COOH nitrating mixture is used.
The best PNAIW yield (75%) is observed for the nitration of TAIW with 99-% HNO3
alone at 0C for 1 h. Here again a raised temperature and an addition of sulfuric acid reduces
the PNAIW yield.
PANIW is a crystalline solid decomposing over the temperature range of 258260C
fig.2.
DTA
TG
50
115
180
245
310
375
Fig. 2. . DTA-TG of PNAIW

Readily soluble in inorganic acids (nitric, sulfuric) or organic (acetic) acids, the compound
is soluble in organic solvents: THF, DMSO, DMF. dimethylacetamide, acetonitrile. It is
insoluble in acetone, xylene, chloroform, ethyl acetate, methanol, ethanol, isopropanol or
hexane.
As a representative of isowurtzitans, PANIW is built of a six-member ring in which
two nitroamine groups are connected with two dimethylene bridges. The bridges form a
basis of two five-member rings which are clamped with a dimethyene bridge. In these rings
at positions 2, 5 nitrogen atoms occur, one of which is attached to the acyl group and the
other to nitrogen groups
566
Czech Republic
2006
[Content]
12
N+
O
O
+
10
N+
O
N+
Fig. 3. Spatial structure of PNAIW

Its 1H - NMR spectrum fig.4.
Fig. 4.
H-NMR of PNAIW
is typical of isowurtzitanes where the signals from methylene bridge hydrogens lies in
the range of the signals of aromatic protons: of the six-ring bridges at = 8.3 singlet 4H and
the protons of the bridge that links the five-member rings at = 8.1 two singlets 1H. The
spectrum differs from the PMR spectrum of HNIW which is composed of two singlets 4H
and 2H. Singlets 4H are assigned to two dimethylene bridges at six-member ring, and singlet
2H corresponds to the dimethylene bridge that clamps the five-member rings. The presence
of these singlet is evidence for the fact that the four hydrogen atoms of the six-member ring
are in the same environment. The presence of singlet 4H and the presence of 2 singlet 1H in
the PMR spectrum of PNAIW suggests that the 4H hydrogen atoms are in the same
enivironment, whereas 1H hydrogen atoms are not equivalent.
567
Czech Republic
2006
[Content]
In the IR spectrum fig.5 - the band of asymmetric vibrations of methylene groups of

the isowurtzitan cage is shifted towards higher wavenumbers to a value of 3028 cm-1.
90
T%
60
30
0
4000cm -1
3400
2800
2200
1600
1000
400
Fig. 5. IR spectrum of PNAIW

Also, the methyl group of the acyl in the PMR spectrum gives the signal shifted to =
2.04 and in the IR spectrum the C=O group band is observed in a non-typical position at
1688 cm-1.
PNAIW at low temperatures not exceeding 20C undergoes nitration with 99-% HNO3
to HNIW in a 98-% yield - Reaction 2.
PNAIW + HNO3 HNIW
(2)
Nitration of PNAIW with this acid in the presence of acetanhydride results in a decay of the
isowurtzitane structure to afford a new cage structure -TEX in a 21-% yield -Reaction 3
PNAIW + HNO3 + (CH3CO)2O TEX
(3)
REFERENCES
[1]
[2]
[3]
[4]
[5]
568
Nielsen A.T., Nissan R.A., Vanderah D.J., J.Org.Chem. 55, 1459-1466, 1990.
Bellamy A.J., Tetrahedron 51, 4711 4722,1995.
Tomotsu K., Masahiro T., Masanori I., EP 0 753 519 A I, 1997.
Bellamy A.J., 30th Int.Ann.Conf.ICT, 30, 109.1.-109.14, 1999.
Shu-Sen C., Wen-Ge-Q., Yong-Zhong Y., Chin.J.Chem. 18(5), 756-758, 2000.
Czech Republic
2006
[Content]
INITIATION OF DEFLAGRATION AND DETONATION IN THE

METAL DUST/AIR SYSTEMS
I.V. Egorova, B.N. Kondrikov
Mendeleev University of Chemical Technology 125047, Miusskaja sq. 9,Moscow,Russia
125047, Miusskaja sq. 9, Moscow, Russia
Abstract
This work concerns to summarizing of literature experimental results of initiation,
deflagration and detonation in the metal dust/air systems. Results of works of the
Institute of Chemical Physics of Academy of Sciences of Russia on initiation of
detonation of dust/air systems and works of Novosibirsk Scientific Center on ignition of
aluminum and magnesium particles in a shock tube and influence on this ignition of
organic substances are in detail considered. The main conclusions is that particles of
magnesium are ignited in a shock wave much easier, than aluminum and the most
dangerous additives to aluminum from organic liquids are nitrates of fat spirits.
Keyword: Ignition, ignition delay, initiation of detonation
1.
INTRODUCTION
A problem of initiation and propagation of deflagration and detonation in air-dust and

air-liquid droplets clouds presents the great interest in many respects, especially, of course,
from the point of view of providing reliability and safety of technological processes and
avoiding of man-caused catastrophes. The problem attracted attention of specialists in the
field of burning and detonation still in the 50th due to intensive usage of the particulate
metals in formulations of solid rocket propellants.
The new direction of explosive systems development has given birth to many of theories
and experimental investigations that converted, in fact, this rather badly developed field of
science into one of the most advanced directions of theory of explosives, as well as, of
course, the theory of reliability and safety in chemical and other branches of modern
technology.
2.
ICP WORKS
The central position in a row of scientific materials related to ignition and initiation of
detonation in the metal dust/oxidizing gas clouds presumably must be assigned to the work
by A.A. Borisov, B.A. Khasainov et al. carried out in the Institute of Chemical Physics of
Academy of Sciences of Russia (ICP) [1].
The shock tube made of 122 mm caliber 4.2 m long gun barrel was used for
investigation of detonation processes in aluminum/air suspensions in [1]. The flaked
aluminum of Russian production, named PAP, was taken for the experiments. Again, it is
not said what kind of aluminum, PAP-1 or PAP-2, is used, but thickness of the flakes, about
1 m, is reported.
Irina Egorova, Boris Kondrikov
569
Czech Republic
2006
[Content]
The powders were placed in trays inside the barrel, with 25 cm distance between them, ~
0.5 s before the experiment, the barrel was blown off by compressed air that formed the
dust/air suspension. The quantity of the powder deposited on the walls during the time of ~
0.5 s was specially estimated and appeared to be only 10-20 % of the original mass of the
powder in the trays.
Initiation of detonation was realized by a detonating cap and a sample of RDX placed at
the open end of the tube. Registration of a process inside the tube was implemented by
means of the streak-camera through the windows in the walls on the last 2 meters of the tube,
and two pressure gages located at 1.25 and 3.25 m from the open end.
The wave velocity was measured in every run and if it were constant on the last two
meters of the tube the process was considered detonation. When it diminished the initiator
was estimated to be too small to produce the steady-state detonation, and in the subsequent
run it was increased, till the steady state process was achieved. The energy of the initiator
was calculated on the base of the theory of the point explosion, and as to the system of RDX
plus detonator taken for initiation it corresponded to about 3.3 MJ per gram of RDX.
E, MJ/m
0
100
2
200
300
400
G, g/m
Fig 1.
500
600
Initiation energy of aluminum particles/air suspension vs concentration of

particles. Open points are dying waves, dark points are waves of stationary
detonation. 1-PAP; 2-UDA.
Results of the experiments in the tube 122 mm dia are presented in Fig. 1. The critical
energy of initiation for the flaked aluminum/air suspension (curve 1) of 330 g/m3
concentration is 3.4 MJ/m2 what corresponds to 12 g of RDX used as the initiator. For the
ultra-dispersed aluminum, UDA (~ 1 m dia) the initiation energy is much less than for the
flaked aluminum (curve 2 in Fig. 1, the minimal energy of initiation is less than 0.3 MJ/m2)
what could be explained presumably by the fact that the flakes are covered by the thin film
of aliphatic acids that form an additive obstacle (in addition to the film of aluminum oxide)
to the heating and inflammation of the flakes, in comparison with the UDA particles
deprived of the carbooxyhydrates film. The photoregistrograms of detonation process (Fig.
2) show the well formed spin-detonation process with a length of the spin-wave about 0.4 m
independently on concentration of the flaked aluminum ranged from 200 to 400 g/m3.
570
Fig 2.
3.
Czech Republic
2006
[Content]
Photographs by means of streak camera of detonation processes in

aluminum/air suspensions: a-PAP, b-UDA, -170 g/m3; c-UDA,
=210 g/m3.
NOVOSIBIRSK SCIENTIFIC CENTER DATA
Ignition of particles aluminum and magnesium in reflected shock waves was

investigated in [2]. The shock tube consisted of two sections, high and low pressure,
respectively, 1.5 m and 5 m length, 56x56 mm cross-section. Helium was employed as a
driving gas, at a pressure ranged from 2.5 to 5 MPa. Initial pressure of oxygen P = 10-100
kPa. The powders were injected into the tube mainly by the flow of a moving gas. Four
samples of powders were employed:
1. Spheroidal aluminum 3-5 m dia, 99.2% of neat aluminum.
2. Al/Fe alloy (89.4/10.6) with particles of a complex form 0-20 m dimension.
3
571
Czech Republic
2006
[Content]
3. Aluminum powder 20x30 m dimensions with a thickness of 2 to 5 m (again, the

question is what sort of the flaked aluminum was used but no answer is contained in the
paper).
4. Spheroid powders of Magnesium 40-63 and 63-80 m dia.
Results of the experiments are shown in Figs. 3 and 4. Two strange facts follow from the
Fig. 3. (1) All the three sorts of aluminum give exactly the same basic curve (open points) in
the log vs 1/T plane.
However, especially for the sort 1 of the spheroid aluminum particles, the great scatter of
points was observed. Ignition delay of 1 ms was measured at a temperature not only 1800 K
(basic line) but 1300K and even 1000K (a big group of solid points in Fig. 3). It does mean
that ignition of aluminum in shock wave is a complex process that may be described hardly
by usual equations of heat exchange and reaction development (dashed line in Fig. 3)
Fig 3.
572
Ignition delay of Magnesium particles in clear oxygen atmosphere behind

the reflected shock wave
Fig 4.
Czech Republic
2006
[Content]
Ignition delay of Magnesium powders. 1-d=4063 m; 2-d=6380 m; 3calculation (T=1368K, d=75 m).
Ignition of the magnesium particles is more uniform process (Fig. 4). Ignition delay as
well as in the case of aluminum does not depend on dimensions of the particles, and no
specific deviations for the overall experimental dependence is noted. The calculated lines are
deviated from the experimental points in Fig. 4 but it rather moderate difference.
Veyssiere [3] gives the method of calculation of ignition temperature developed further in
and compares the results with a formula for calculation of ignition temperature of
aluminum particles in a hot gas [8]:
[4-7]
Ti = TmAl2O3-0.6( cox0.3 /g)exp(-0.85 d )
(1)
where TmAl2O3=2310 K is the melting point of Al2O3,

cox is the concentration of the oxidizer in gas and
g is the thermal conductivity of the gas calculated at the average temperature of the gas.
The ignition temperature calculated on the base of formula (1) is essentially lower than
the melting point Al2O3 and ignition delay calculated at this ignition temperature is about
two times less than theory [3] gives.
4.
INTENSIFICATION OF IGNITION OF ALUMINUM POWDERS BY

ADDING OF EASILY FLAMMABLE LIQUIDS
Addition of easily flammable liquids (EFL) is well known method of intensification of

ignition and burning of metal powders in shock waves. Significant moment of this approach
5
573
Czech Republic
2006
[Content]
consists in investigation of ignition of the EFL themselves. Voluminous experimental

material in this respect is contained in [9-12]. This material is limited, however, mostly by
investigation of behavior in shock waves of hydrocarbons of the type of kerosene. Important
study of ignition of a broad circle of EFC was performed in [13]. Influence of some possible
promoters on the process, and, the main point, action of the EFL on ignition of the
air/aluminum dust systems was investigated.
Experiments were carried out in shock tube in a reflected wave in air. Mach numbers in
the incident wave were taken M=2.3-4.0, at an initial pressure of P1=0.01-0.1 MPa. The
hydrocarbons of the sort of aviation fuels TA and T-15, fuel oil DTL and tridecane (C13H28
TD) were investigated. Concentration of droplets varied along the cloud of the droplets in
working section of the tube and the ignition point arises in the part of the mixture where
air/droplets concentration is most suitable for that. In most of the cases the distance of
displacement of the liquid from the reflecting wall L=73 mm, pressure in the reflected wave
p5=2.3 MPa.
Visualization of the process was implemented by means of the Schlieren laser and usual
streak-camera methods. In Fig. 5, dependence of ignition delay on inverse temperature of the
media is presented.
Fig 5.
Ignition delay of droplets of aviation fuels T-15 (1) and TA (2), tridecane
(TD) (3), fuel oil, DTL (4) and droplets of nitrates: cyclohexylnitate (5),
isopropylnitrate (6), nitrates of alyphatic alcohols (7).
Together with the hydrocarbons, the data for the potential promoters cyclohexylnitate
(CHN, C6H11ONO2), isopropylnitrate (IPN, C3H7ONO2), and a mixture of the nitrates of
alyphatic alcohols (NA, CnH2n+1ONO2, at n=611).
Ignition delays for the kerosene TA and T-15 and TD are about the same:
574
ign=2.5exp(7220/T),
s, EA=60 kJ/mol
Czech Republic
2006
[Content]
where EA is the apparent energy of activation of the process.

Pre-exponential factor for the fuel oil is two times less, 1.2 s. At T>1250 K is all of the
four cases ign<1 s but for the fuel oil it is two times less than for the kerosene and TD.
The consequent experiment were carried out with the fuel oil.
Dependence of ignition delay on temperature for the nitrates is more diverse:
IPN:
ign=
0.014exp(11900/T), s, (EA=99kJ/mol),
CHN:
ign=
0.0006exp(11900/T), s, (EA=99kJ/mol),
ign=
75exp(3000/T), s, (EA=25kJ/mol).
NA:
Promoting influence of the nitrates follows from Fig. 6. One can see that all of the
promoters works in about the same manner, at a high temperature, 1500 K. At a lower
temperature difference appears. At T=1000 K only NA and IPN allow the fuel oil be ignited
at a time less 1 ms, and NA, in visible contradiction with Fig. 10, works better than the IPN
does.
Ignition delays of the fuel oil being activated by addition of NA are presented in Fig. 7.
Not only ign decreases at increasing of the NA concentration, cNA, in the solution but the
apparent energy of activation dimensions, at that
cNA=0% EA=60 kJ/mol,
cNA=10% EA=46kJ/mol,
obviously being directed towards the value of 25 kJ/mol which is characteristic for the
neat NA.
Mixture of 70% fuel oil and 30% NA is characterized by the Arrhenius type dependence
of the form:
ign=6exp(4330/T),
s.
Addition of the mixture to spherical aluminum powder ASD1 (99.2% Al, 0.2% Fe,
0.02% H2O, fractions of 3-5 m, 10-14 m and 14-20 m are used in the experiments).
Streak-camera photographs are presented in Fig. 7. Aluminum of 3-5 m fraction burning is
a subject of Fig. 7 experiments. Mass of the liquid in the experiments was ml=7 mg, relation
=mal/ml has been changed. Temperature was T1500 K. At =0 (no aluminum) Fig. 8a
liquid burns with quickly estinguishing flame. Addition of aluminum (mal=10 mg Fig 8b and
mal=10 mg in Fig 8c) leads to formation of voluminous flame of aluminum burning.
575
576
Czech Republic
2006
[Content]
Fig 6.
Ignition delay of DTL with promoting additives vs concentration of additive:

Numbers over curves denote temperature of the gas in reflected wave.
Fig 7.
Ignition delay of fuel oil with addition of NA: 1-100% NA, 2-100% fuel
oil, 3-95%fuel oil+5%NA; 4-90% fuel oil+10% NA, 5-70% fuel oil+30%
NA.
Fig 8.
Czech Republic
2006
[Content]
Photographs of ignition and burning of aluminum powder/EFL mixtures:

T1500 K, ml=7 mg, a- =0, b- =1.5 at mAl=10 mg, c- =10 at mAl=100
mg.
Dependence of ignition delay on inverse temperature (Fig. 9) shows that influence of

EFL on ignition of aluminum is a complicated matter. At =0.3 and =1.5 (mal=10 mg, ml
is, correspondingly, 33 mg and 6 mg) delay tame versus inverse temperature is about the
same line (with essential scatter of points what is not a surprise bearing in mind the
heterogeneous structure of the sample that should be, at first, dispersed and then ignited by
the shock wave). The line for a big =10 (at mal=100 mg and ml=10 mg, i.e. almost neat
aluminum powder) is located in Fig. 9 noticeable higher than the line attributed to the small
was. It is shown in Fig. 9 in such a way that its lower part intersects the smaller lines.
Fig 9.
Ignition delay of aluminum powder/EFL mixtures: =0.3 at ml=33 mg,

mAl=10 mg, 2- =1.5 at ml=6 mg, mAl=10 mg, 3- =10 at ml=10 mg,
mAl=100 mg, 4-EFL (70% fuel oil+30%NA).
577
Czech Republic
2006
[Content]
One would suggest, however, that this lower part of the =10 line is perhaps imaginary.
But the upper part in comparison with the smaller dependence has absolutely definite
sense: the aluminum ignition is essentially promoted by mixing it with the small quantity of
the DTL/NA solution. The small line (points 1 and 2 in Fig. 9) corresponds to the
relation:
ign=14(4330/T),
s, (EA=36 kJ/mol).
The relation is close to that for fuel oil/NA (70/10) mixture but preexponential factor is
about two times more.
=10 line in Fig. 9 is described by the formula:
ign=1.210
-3
exp(20800/T), s (EA=173 kJ/mol).
M, kg Tetryl
The apparent activation energy for the almost dry aluminum powder is close to those for
heptane/oxygen (191 kJ/mol) [13] and methane/oxygen/nitrogen [187 kJ/mol] [13].
0.1
0.01
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
Fig 10. Comparison of experimental data on critical initiating charge for

ethane/air [14] (upper line and points) and flaked aluminum/air [7]
(lower line and points) systems. Open circles and triangles are attributed
to detonation, solid circles and triangles relate to experiments with no
detonation.
One may conclude that ignition of aluminum powder is strongly promoted by addition of
liquid promoters, and the final result, as to the ignition delay is concerned, is determined
almost exclusively by the ignition characteristics of the promoting liquid.
It would be relevant to compare in conclusion the initiation detonation data for
aluminum/air [1] and ethane/air [8] systems (Fig. 10). Close similarity of those, in principle,
absolutely different systems, is very impressive fact.
578
10
5.
Czech Republic
2006
[Content]
CONCLUSIONS
Possibility of initiation of detonation in the clouds of the particulate aluminum strongly

depends on the particles dimensions. Aluminum of the flaked type, with about one micron of
the reduced diameter of the particle, in spite of the presence of the layer of hydrocarbons on
the surface of the flake that obviously possess inhibiting influence on ignition of the particle
suspended in gas (the same is not at all obvious when burning of the particles is considered
in the mass of the condensed metal containing energetic materials (MCEM) of waterimpregnated gelled type) have very reasonable initiation limits. The spherical ultra-fine
aluminum of about 1 m diameter sphere needs still more weak initiating impulse than the
flaked aluminum requires. But the sorts of spherical aluminum having bigger diameters of
the sphere can hardly be ignited and in fact are not capable of initiating and propagating of
the detonation wave. It is clearly shown that parameters of initiation of the air/dust and
air/gas systems are absolutely analogous to each other and the proposal that consideration of
the initiation of the gas/gas and gas/liquid clouds is useful for the air/metal dust systems is
directly confirmed.
It has been shown that addition of some promoters of the row of alyphatic esters of nitric
acid, isopropylnitrate (IPN), cyclohexylnitate (CHN), and especially the nitric esters of a
mixture of alyphatic alcohols of intermediate chain length strongly diminishes the ignition
delay of the mixture in shock wave and, correspondingly, will decrease the limiting
parameters of the initiating shock wave.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
11
Borisov, A.A., Khasainov, B.A., Veysser, B., Saneev, E.L., Fomin, I.B., Khomik, S.V.,
Detonation of aluminum suspension in air and oxygen, Chemical physics, 1991, Vol. 10, 2,
pp.250-273.
Boiko, V.M., Lotov, V.V., Papyrin, A.N., Ignition of suspensions of metal powders in gas in
reflected shock waves, Combustion, Explosion and Shock Waves, 1989, Vol. 25, 2, pp. 6775.
Veyssiere, B., Ignition of aluminum particles in a gaseous detonation, Shock Waves,
Explosions, and Detonations, 1983, Vol. 87, 362-375.
Khasainov, B.A., Vessiere, B., Archivum Combustionis, 1987, Vol. 7, 3-4, p. 333.
Khasainov, B.A., Vessiere, B., Dynamics of explosions, AIAA Progress in Astronautics and
Aeronautics, New York, AIAA, 1988, Vol. 114, p. 284.
Veyssiere, B., Khasainov, B.A., Combustion and Flame, 1991 (in press).
Veyssiere, B., Dynamics of explosions, AIAA Progress in Astronautics and Aeronautic, NewYork, AIAA, 1986, Vol. 106, p. 522.
Ozerov, E.S., Skvortsov, I.I., Burning of magnesium particles conglomeration, Combustion,
Explosion and Shock Waves, 1971, Vol. 7, 2 pp. 227-232.
Wierzba, A., Kauffman, C.W., Nicholls, J.A., Ignition of partially shattered liquid fuel drops in
reflected shock-wave environment, Combustion Scientific Technology, 1974, Vol. 9, 5/6, pp.
233-245.
579
Czech Republic
2006
[Content]
[10] Borisov, A.A., Gel'fand, B.E., Eremenko, L.T., Features of ignition of inflammable liquids,
Report of Acad. of Science of USSR, 1979, Vol. 247, 5, pp. 1176-1179.
[11] Borisov, A.A., Gel'fand, B.E., Timofeev, E.I., et al., Self-ignition of atomized liquid fuel in
gaseous medium, Flames, Lasers, and Reactive Systems, Prog. Astronaut. Aeronaut, 1991, Vol.
132, pp. 205-219.
[12] Boiko, V.M., Lotov, V.V., Papyrin, A.N., Ignition of liquid fuel drops in shock wave,
Dynamics of Deflagration and Reactive Systems: Heterogeneous Combustion, Progress in
Astronaut. Aeronaut, 1991, Vol. 132, pp. 205-219.
[13] Boiko, V.M., Poplavsky, S.V., Ignition of aluminum powders/liquid hydrocarbon fuels
mixtures in air, Combustion, Explosion and Shock Waves, 1999, Vol. 35, 1, pp. 35-40.
[14] Atkinson, R. and Bull, D.C., Kinetic modeling of ethane/air detonability, Shock Waves,
Explosions, and Detonations, Bowen, Mabson, Oppenheim, and, Soloukhin, Editors, 1983, Vol.
87, pp.318-332.
580
12
Czech Republic
2006
[Content]
JET VELOCITY AND THE EFFECTS OF THE SHAPED CHARGE

PERFORATOR FOR USE IN THE OIL AND GAS WELL
Ester Zvonimir, Mario Dobrilovi, Branimir Jankovi, Zvonimir Tomii*, Mile Odak*
Faculty of Mining, Geology and Petroleum Engineering
University of Zagreb, Croatia, Pierottijeva 6, 10000 Zagreb, Croatia
*Crosco oil services d.o.o., Ulica grada Vukovara 18, 1000 Zagreb, Croatia
Abstract:
In paper are described measurement methods and results of the jet velocity
measurement of the perforator shaped charge for use in oil and gas drills for improving
of the fluid flow. Effects of perforation are tested on different model type and compared
with theoretical values.
Keywords:
explosives, oil drills, perforators, shaped charge.
1.
INTRODUCTION
1.1
Use of explosive in oil and gas production
Perforating is the process of creating holes when an oil or gas well is completed, casing
and cement isolate the well bore from surrounding formations. Perforating is used to
establish fluid communication between the well bore and formation so that production
operations can proceed through perforation-hole in casing and cement.
This process can be done with use of perforating guns-devices used to perforate wells.
Sample of perforating gun is given on fig. 1. A typical gun consists of a collection of
explosive charges, a charge carrier, a detonating cord and a detonator. The chargers are load
in the charge carrier and they are connected to the detonating cord, which explosively
couples the charges. The gun is lowered into the well by wire line to the depth at which
perforations are desired.
Fig 1.
Perforating gun
Zvonimir Ester, Mario Dobrilovi, Branimir Jankovi, Zvonimir Tomii, . . .
581
1.2
Czech Republic
2006
[Content]
Lined shaped charges
The most important component of perforating gun is the perforating charge. The charge
utilizes a secondary high explosive to propel a stream or jet of high speed metallic particles
that penetrate the casing, cement and rock formation.
Perforating charges are lined shaped charges. A cavity is shaped in the explosive and is
lined with a metallic material. Liner shaped charge have four components: the charge case,
the liner, the main explosive, and the primer.
The charge case is container that houses the other charge components. It must withstand
high temperatures and pressures. Common case materials include machined steel, cold
formed steel, die-cast aluminium and ceramics. The liner furnishes the mass needed for the
jet to penetrate casing, cement and formation. The pressures exerted on the liner when the
main explosivedetonates cause the linear to colalapse and form the jet. The shape of the
linear and its thickness and composition greatly influence the depth, diameter, and
effectivenes of the perforation. The linear is usually conical or parabolic in shape. Themain
explosive provides the energy needed to produce the jet. The mass, distribution, and
detonation velocity of the main explosive greatly affects charge performance. Granular, high
order, secondary explosives such RDX, HMX, HNS and PYX are used. The proper amount
of explosive is measured into the charge case; the lineris placed on the top of the explosive;
then pressure is applied to the case and linear to compact the explosive. The shapes of the
linear and of the case determine the distribution of the explosive. The amount of main
explosive ranges from 2 g in charges loaded in small diameter open carriers to over 30 g in
charges loaded in large diameter hollow carrers used in casing application. The primer, or
buster is composed of a small amount of explosive and its main purpose is to transfer the
shock wave from the detonating cord to main explosive. After firing, the liner collapses,
staring at the apex. The particles begin to converge along the cone axsis to form the needlike,
high speed jet. The jet shoots from the charge cone and, upon contact with the initial target,
can have pressures at the leading tip over 400 kbar. Fig. 2 shows scheme of perforator lined
charge.
Fig 2.
Perforator lined shaped charge
The perforation of the target is accomplished by the stream of metallic jet particles
radially displacing the impacted target material and thereis no fusing, glazing, or scorcing of
materials surrounding the perforation. In the rock formation, the material immediately
surrounding the perforation is crused and compacted in crused zone. Since the crused zone
Since the crused zone usually exibits reduced permeability when compared to the
582
JET Velocity and the Effects of the Shaped Charge Perforator for Use in the Oil . . .
Czech Republic
2006
[Content]
undisturbed formation, it can have significant effect upon the productivity of the well. The
jet energy is quickly expended and the penetration process ceases. The length and diameter
of the perforation depend upon the target material and upon the energy distribution in the jet.
Fig.3 shows effects of perforation in formation.
Fig 3.
2.
Perforating effects in formation
TESTING OF LINEAR SHAPED CHARGES- RAPTOR
Testing of the perforators have been done with the aim of establishing real velocity of
detonation of the cumulative stream-jet in relation to the declared values, and determining
relation between the velocity of detonation of the raptors and penetration depths of the
stream in the environment, through the bore-hole to surrounding formations.
2.3
Lined shape raptor 3 3/8, Owen oil tools
Technical data of lined shaped raptor, produced by Owen oil tools producer 3 3/8:
type: 3 3/8 TAG-6SPF, 60 DEGREE RAPTOR NT3
charge type: SDP 3375-411 NT 3
explosive weight (g): 25 HMX
entry hole (mm): 10,67
penetration (mm): 945
Lined shape raptor3 3/8, Owen is shown on fig.4.
Fig 4.
Lined shape raptor 3 3/8
583
2.4
Czech Republic
2006
[Content]
Perforator (raptor) performace evalulating
To establish a basis for comparing perforator performance American Petroleum Institute

(API) recommended procedures. These procedures provide standard means of evaluating
perforator perfomance under surface conditions as well as as evaluating the physical and
flow characteristic of the perforations under temperature and pressure. Two types of targets
are used in testing, a concrete target and Berea sandstone target
2.4.1. Target description
Concrete target consist of a section of tubing or casing around which concrete has been
poured in a metal form. The perforator being tested is fired at ambient temperature and
pressure, the tubing fluid is water.
In Barea test the target consist of sandstone core mounted in drillpipe. The perforator
and target are positioned in the test chamber so that the perforator fires upward into the
target end plate.
Basic principles of those tests are given on fig. 5 and 6.
Fig 5.
Concrete target
Fig 6.
Barea target
The results of the tests can serve as a basic for generalcomparasion of perforator
performance. Actual conditions at the time of perforating a well are most likely but not the
same as those existing during testing; casing, fluids, cements, formation types and pressures
are different. In order to evaluete relationship between the perforating parameter, deepth and
diameter of penetration and jet velocity of detonation, VOD of the jet was measured.
2.5
VOD measurement of raptor jet
2.5.2. Measurement equipment

Electro-optical method was used for the measurements. Velocity of the raptor jet is
measured with electronic stop-timer. The device is receiving impulses through the optical
584
Czech Republic
2006
[Content]
fibre cables. In the measuring, timer calculates velocity on basis of the time and the distance
between the probes. Type of the electronic stop-timer is Explomet-FO-2000 of the
following characteristics: max measurable velocity 10.000 m/s, time interval from 0,1 do 10
s, accuracy +/- 0,1 s.
2.5.3. Measurement description
For the measurements needs, device has been constructed with possibility of setting up of
optic probes on the exactly determined distance. The device consists of: steel tube with holes
for probes, base and bearing plate for the perforators. Distance between the optic probes is
56 mm. Distance between the bottom of perforators and the first optic probe is 56 mm.
Measurement with sample no. 1 has been done in one time interval on the distance of 560
mm in order to confirm the measuring method. Measurements 2-5 are done for the purpose
of the determining velocites distribution of the stream in respect to the length of stream.
Scheme of the measurement principle is given is on the picture 7 and the photo of the
measuring equipment and the perforators on the picture 8.
Fig 7.
Measurement scheme
Fig 8.
Measurement equipment
1. Liner shape raptor

2. Detonating cord
3. Electric detonator
4. VOD meter
5. Optical probes
6. Steel plate
7. Steel tube 120 mm
5
585
Czech Republic
2006
[Content]
Results of measurements are given in table 1.

Table 1. Mesurement resuls
Interval
1
Interval
1
2
3
4
5
Interval
1
2
3
4
5
3.
Measurement no. 1
L(mm)
t(s)
v (m/s)
560
90,2
6208
Measurement no. 2
L (mm)
t (s)
v (m/s)
56,00
10,2
5490
56,00
9,9
5656
56,00
56,00
56,00
Measurement no. 4
L (mm)
t (s)
v (m/s)
56,00
8,2
6829
56,00
8,9
6292
56,00
7,5
7466
56,00
65,0
861
56,00
-
Measurement no. 3
L (mm)
t (s)
v (m/s)
56,00
16,1
3478
56,00
8,6
6511
56,00
9,5
5894
56,00
51,5
1087
56,00
265,7
210
Measurement no. 5
L(mm)
t (s)
v (m/s)
56,00
8,7
6436
56,00
8,9
6292
56,00
8,0
7000
56,00
10,4
5384
56,00
7,9
7088
CONCLUSION
In the measurement no. 2, 3 and 4 deviation of the jet occurred. That caused failure of
the measurement in some intervals. For the measurement no. 5, some changes in
measurement equipment have been done and that result with VOD data for all of intervals.
The results of measurements pointing on the conclusion that applied devices are
appropriate for measurement of the jet velocity. Measured velocities are in the range that
response to the manufacturer declared values (6000-7000 m/s).
Measuring equipment used in testing have properties for measuring maximum jet length
toward 560 mm. With reconstruction of the device with maximum length up to 1000 mm
(range of declared perforation length) is possible to determinate effect of shape lined chargeraptor, on the simply way without performing complicated tests.
REFERENCES
[1]
[2]
[3]
[4]
586
WELEX: A introduction to perforating, A Halliburton company, 1986

TOMII Z.,ODAK M.: Lined shaped charge, Crosco oil services, 2003, ppt.
TOMII Z., ODAK M.: Types of perforation, Crosco oil services, 2003, ppt.
LABORATORY FOR EXPLOSIVE Faculty of Mining, Geology and Petroleum Engineering:
Reports of testing, 2005
Czech Republic
2006
[Content]
THERMOCHEMICAL CALCULATIONS AND EXPERIMANTAL

STUDY OF COMPOSITE PROPELLANTS CONTAINING DADNE OR
NTO AND EVALUATION OF THEIR PROPERTIES
Bogdan Florczak, Andrzej Orzechowski and Tomasz Saaciski
Institute of Industrial Organic Chemistry, 6 Annopol Str.,03-236 Warsaw, Poland
Abstract:
In this paper, the thermochemical calculations (thermochemical properties, reaction
products nozzle, heat of explosion) and experimental study results (deflagration point,
thermal analysis and sensitivities to friction and impact, heat of combustion) of
composite solid propellants containing DADNE or NTO particles are described. The
deflagration point, the impact sensitivity and the friction sensitivity were determined by
the UN methods. Differences in properties of the propellants with DADNE or NTO are
presented.
Keywords: composite solid propellant, DADNE, NTO, thermochemical software,
sensitivity of energetic materials
1.
INTRODUCTION
Composite (heterogeneous) propellant, contains of at least two substances as an oxidizer

and a fuel which are physically mixed. The most commonly used composite solid propellant
in model rocket motors is a mixture of ammonium perchlorate (AP), synthetic rubbers whose
structures are mainly C-C and C-H bonds, metallic fuels such as aluminium and the other
additives for controling burning rate.
Solid propellants are usually mixed up with explosives. It is because molecules of
explosives contain both fragments of oxidizer and fuel what makes it able to self-sustain
combustion. Propellants generate combustion gases by deflagration phenomena (subsonic
combustion process), whereas explosives generate combustion gases by explosion or
detonation phenomena (supersonic combustion processes). However, actually propellants
and explosives are fundamentally the same in terms of their chemical compositions.
The ideal properties of composite solid propellant should be as follow[1]:
1. Low molecular weight of combustion products (the lower is their molecular
weight, the greater can be the acceleration that applied to the gases).
2. Stability (resistance to chemical and physical changes) over a long period of
time.
3. Maximum density, which allows to store the greatest amount of energy that can
be packed into every unit of case volume of propellant.
4. Smokeless exhaust gasses to avoid deposition of smoke particles at operational
locations and detection in military usage.
5. Nonluminous, noncorrosive and nontoxic exhaust.
1
Bogdan Florczak, Andrzej Orzechowski, Tomasz Saaciski
587
Czech Republic
2006
[Content]
Composite solid propellants containing of HMX or RDX particles offer the advantages
of low flame temperature and low molecular mass combustion products, as well as reduced
infrared emission. The reduced infrared emission can be obtain by elimination of CO2 and
H2O from the combustion products[2].
Hence it was of importance to study the performance of propellants with other
nitrocompounds novel insensitive energetic materials i.e. 3-nitro-1,2,4-triazo-5-one (NTO)
or 1,1-diamino-2,2-dinitroethylene (DADNE, FOX-7).
In this work, results of thermochemical calculations and experimental study of
composite solid propellant (AP/PBAN-Epoxy resin) containing 20% of DADNE or NTO are
presented.
2.
PERFORMANCE CALCULATIONS
Performance calculations were carried out employing the ICT Thermodynamic

Code[3]. The ICT-Thermodynamic Code is based on a method developed by the National
Aeronautics and Space Administration (NASA). This method uses mass action and mass
balance expressions to calculate chemical equilibria. Thermodynamic equilibria can be
calculated for constant pressure conditions as well as for constant volume conditions.
Propellant formulations investigated include mainly: DADNE, NTO, AP as oxidizer and
PBAN-Epoxy resin as binder, dioctyl adipate (DOA) as plasticizer. The calculations were
performed for a chamber pressure of 70 bar, considering an adiabatic through a nozzle in
onedimensional flow at chemical equilibrium and with an expansion pressure 1.0 bar.
2.1.
The basic properties of ingredients selected for our studies are listed in Table 1. The
thermochemical values for the ingredients are taken from literature[4-9].
Table 1. Physicochemical properties of propellant ingredients.

Ingredient
Sum Formula
Oxygen Balance [%] Density

[g/cm3]
Hf
[kJ/mol]
PBAN/Epoxy resin
C64.466H85.819N1.421O7.457
-263.01
1.027
-929.00
DOA
C22H42O4
-263.37
0.925
-1215.03
AP
NH4Cl4
34.04
1.950
-296.77
DADNE[4]
C2H4N4O4
-21.61
1.885
-133.89
NTO[5]
C2H2N4O3
-24.60
1.930
-59.83
Aluminum
Al
-88.95
2.702
0.00
Ferrocene
C10FeH10
-223.61
1.490
154.89
588
Thermochemical Calculations and Experimantal Study of Composite . . .
Czech Republic
2006
[Content]
Compositions and thermochemical calculated properties of composite propellants tested

are shown in Table 2.
From Table 2, it is seen that at insensitive energetic materials levels of 20%, the specific
impulse values (Ieq) decreases from 2330 Ns/kg to 2269 Ns/kg for DADNE and to 2220
Ns/kg for NTO. An interesting phenomenon is it that the thrust coefficient and mole number
are almost identical for all of the propellants.
The composite propellants (Table 2) containing DADNE or NTO particles offer the
advantages of low flame temperature and reduces infrared emissions (the reduces of H2O in
combustion products chamber and nozzle) and the mole fraction in reaction products of HCl
as shown in Fig 1-3.
Table 2. Chemical compositions and thermochemical properties of propellants.
Compounds, %
AP
PBAN/Epoxy resin
DOA
Al
Ferrocene
DADNE
NTO
thermochemical properties
density, g/cm3
mixture ratio, O/F
equivalence ratio
oxygen balance, %
temperature chamber, K
mole number, mole/kg
Ifr/Ieq, Ns/kg
Ivol, Froz/Eq., Ns/dm3
characteristic velocity (C*), m/s
thrust coefficient, CF
Propellant
1
58.0
16.2
3.6
2.0
0.2
20.0
0.0
58.0
16.2
3.6
2.0
0.2
0.0
20.0
1.642
1.381
0.337
-38.89
2537
39.31
2232/2269
3666/3726
1397
1.598
1.649
1.381
0.333
-39.49
2482
38.75
2190/2220
3611/3660
1371
1.597
78.0
16.2
3.6
2.0
0.2
0.0
0.0
1.652
3.545
0.489
-27.760
2659
39.50
2281/2330
3767/3849
1431
1.593
589
Czech Republic
2006
[Content]
Reaction Products Nozzle

30
25
Weight %
20
15
10
0
CO2 (g)
H2O (g)
N2 (g)
CO (g)
NH3 (g)
CH4 (g)
HCl (g)
FeCl2 (g)
Al2O3 (s)
Fig. 1. Fig 1. Propellant without DADNE and NTO (Propellant 1).

35
30
Weight %
25
20
15
10
0
CO2 (g)
H2O (g)
N2 (g)
CO (g)
NH3 (g)
CH4 (g)
HCl (g)
Al2O3 (s)
FeCl2 (l)
Fig. 2. Propellant with DADNE (Propellant 2).
590
Czech Republic
2006
[Content]

35
30
Weight %
25
20
15
10
0
CO2 (g)
H2O (g)
N2 (g)
CO (g)
NH3 (g)
CH4 (g)
HCl (g)
FeCl2 (g)
Al2O3 (s)
Fig. 3. Propellant with NTO (Propellant 3).
3.
EXPERIMANTAL RESULTS
The propellant 1 with good processing and mechanical properties was chosen as
baseline for this investigation. The purpose of examination in this study were high-energy
composite propellants containing 80% of solid components. Grain size distribution of AP,
DADNE and NTO crystals were determined by IPS-U (Infrared Particle Size) analyzer
version 8.12. The principle of the analyzer consists in measuring of changes of infrared rays
steam that diffuse on moving particles of material tested.
Bimodal AP was used as an oxidizer (small sized and large sized). Content of the small
sized (fine-grained) AP was 17.4% of the formulations. Particle diameters of the fine-grained
AP were below 50 m (Fig 4). Particle diameters of the large sized (coarse) AP were below
520 m (Fig 5). As ballistic modifiers (combustion catalyst) was used ferrocene (0.2%). As
insensitive explosives (new energetic materials) were used DADNE or NTO with content in
propellant of 20%. Particle diameters of DADNE and NTO are shown in Fig 6 and Fig. 7,
respectively. The content of a plasticizer (DOA) was 3.6%. The binder was based on PBAN
cured with epoxy resin.
Propellant formulations were mixed in a 1.5 - liter vertical mixer at 75 oC. Curing
required 48 h at 75 oC for its completion.
591
Czech Republic
2006
[Content]
4649
4346
4143
3841
3538
3335
3033
2730
2527
2225
1922
1719
1417
1114
911
69
36
20
18
16
14
12
10
8
6
4
2
0
03
Particle concentration, %
Particle diameter, m
Fig. 4. Numeric particle size concentration as a function of particle size for fine
AP.
8
7
6
5
4
3
2
1
502515
476489
449462
423436
396409
370383
343356
316330
290303
263277
237250
210224
184197
157171
131144
104118
7891
5164
2538
011
Fig. 5. Numeric particle size concentration as a function of particle size for coarse
AP.
592
Czech Republic
2006
[Content]
25
20
15
10
5
8390
7583
6875
6068
5360
4553
3845
3038
2330
1523
715
07
Fig. 6. Numeric particle size concentration as a function of particle size for
DADNE.
10
8
6
4
151158
143151
136143
128136
121128
98105
9098
8390
7583
6875
6068
5360
4553
3845
3038
2330
1523
07
715
113121
2
105113
12
Fig. 7. Numeric particle size concentration as a function of particle size for NTO.
3.1
Deflagration point
Deflagration points were determined by heating of the sample in a glass test tube in
Woods metal bath at a heating rate of 5 oC/min. Temperature at which the sample got
ignited decomposed, was recorded as deflagration point[9-10]. Three samples of mass 0.5 g
were tested simultaneously. The results of the deflagration point study of propellants 1 3
are presented in Table 3.
593
3.2
Czech Republic
2006
[Content]
Heat of combustion
Heat of combustion represents the caloric equivalent of the total combustion energy
of a given substance. Its value depends only on the composition of the material and not on
any other propellants, such as for example loading density. Combustion experiments are
conducted in a stationary bomb calorimeter with an automatically controlled adiabatic jacket
and without oxygen atmosphere (pressure about 400 Pa). The adiabatic bomb IKA
calorimeter system C 4000 was used to determine the heat of combustion. The calorimeter is
calibrated by burning standard propellant to determine its effective energy equivalent. The
results of the heat of combustion study of propellants 13 are presented in Table 3.
3.3
Thermal analysis
Differential thermal analysis (DTA 551 Ex manufactured by OZM Research) was

carried out by heating 5.05.4 mg of the sample at a rate 5 oC/min in the presence of static
air. Analyses were carried out in open test tubes and a thermocouple protected by a glass
sheath was inserted directly into the sample. Data were evaluated using MEAVY 2.0.0.4
software of DTA 551 Ex analyzer. Onsets and maximum of the first exothermal effect were
evaluated at the thermograms. Figure 8 shows thermograms of research propellants. Onset of
decomposition peak (Tonset) and maximum peak (Tm) of research propellants are presented in
Table 3.
3.4
Impact sensitivity
The method (BAM Fallhammer) was used to measure the impact sensitivity[9-11] of
composite solid propellants to drop weight impact. A sample of the tested propellant was
placed in the piston device, which consists of two steel rollers and a hollow cylinder (collar).
The sample was scooped by a special 40 mm3 measuring cup. The measured sample was
placed into the open piston device. The sample response was observed. Drop weight having
a mass of 5 kg was used. The impact of the drop weight, via the piston device, onto sample
may result in either its initiation or noninitiation, depending on the sensitivity of the
propellant, the weight mass, and its drop height. The study results are shown in Table 3. The
samples were tested under the 50% probability of explosive reaction. The results obtained
according to the impact sensitivity test are shown in Fig. 9.
3.5
Friction sensitivity
The Julius Peters method (BAM Friction Apparatus) was used to measure the
friction sensitivity[9-11] of composite solid propellants. Approximately 10 mm3 of sample was
placed on the rough porcelain plate which was attached to the sliding carriage of the
apparatus. The occurrence of the initiation was observed through the sound effects
(crackling), appearance of smoke, or by the characteristic smell of the decomposition
products. The results of the friction sensitivity study of propellants 1 3 are presented in
Table 3. The samples were tested under the 50% probability of explosive reaction. The
results obtained according to the friction sensitivity test are shown in Fig. 10.
594
Czech Republic
2006
[Content]
Fig. 8. DTA thermograms of propellants, heating rate

5 oC/min.
595
Czech Republic
2006
[Content]
Table 3. Some thermal and explosive properties of composite solid propellants

Properties
Propellant
Deflagration point, oC
Differential thermal analysis, oC
Tonset
Tm
Heat of combustion, J/g
Heat of explosion, J/g
Impact sensitivity, J
Friction sensitivity, N
* - calculated value[3]
1
243
240
245
2
218
220
219
3
213
213
213
245
252
5038
4934*
35
168
226
232
4207
4574*
35
144
219
228
3919
4312*
25
168
Probability of explosive
reaction [%]
100
75
50
25
0
10
20
30
40
50
Impact energy [J]

propellant 1
propellant 2
propellant 3
Probability of explosive reaction

[%]
Fig. 9. Sensitivity to impact of propellants 1 3.

100
75
50
25
0
100
150
200
250
300
350
Friction energy [N]

propellant 1
propellant 2
propellant 3
Fig. 10. Sensitivity to friction of propellants 1 3.

4.
CONCLUSIONS
On the basis of the thermochemical calculations and experimental results the following
can be concluded:
596
Czech Republic
2006
[Content]
-For the composite propellants tested containing insensitive energetic materials

(DADNE or NTO) in amount of about 20%, the specific impulse values (Ieq) decreases about
2.6-4.7%. An interesting phenomenon is noticeable that the thrust coefficient and mole
number are almost identical for all the propellants tested.
-The composite propellants composed of DADNE or NTO particles produced of low
flame temperature and reduces mole fractions of H2O and HCl in combustion products.
-The deflagration point and calorific value of propellant with NTO is the lowest.
-The friction sensitivity of propellant with DADNE is 24 N lower than for propellant
with NTO and propellant without the insensitive energetic materials tested.
-Characters of growing of sensitivity to impact for propellants 1 - 3 are very fast.
-In the impact test, propellant 3 obtains 100% probability of explosive reaction under the
level of the stimuli (ca 30 J), for which sensitivities of propellants 1 and 2 are below 50%.
Clear difference between the propellants tested is shown in the level of 100% probability.
The lowest value of 100% probability has been obtain for propellant 3, medium for
propellant 2 and the highest for propellant 1.
-In the friction test, propellants 1 - 3 obtain 50% probability of explosive reaction under
this same level of the stimuli (ca 230 N).
Acknowledgment
This research was supported by the State Committee of Scientific Research through
Institute of Industrial Organic Chemistry, Grant Nr 3 T09B 021 27.
REFERENCES
[1]
G. P. SUTTON and O. BIBLARZ, Rocket Propulsion elements, John Wiley & Sons, INC, New
York, 2001
[2] N. KUBOTA, Propellants and Explosives, Thermochemical Aspects of Combustion, WILEYVCH GmbH, Weinheim, Germany, 2002
[3] H. BATHELT, F. VOLK, M.WEINDEL, ICT-Thermodynamic Code, Windows Fronted:
Version 1.00, Fraunhofer ICT, 1999
[4] H. STMARK, A. LANGLET, H. BERGMAN, N. WINGBORG, U. WELLMAR and U.
BEMM, FOX-7 a new explosive with low sensitivity and high performance, FOA
[5] B. JANZON, H. BERGMAN, C. ELDSATER, et al., FOX-7 - A Novel, High Performance,
Low Vulnerability High Explosive For Warhead Applications, Proc. 20th International
Symposium On Ballistic, Orlando, 2002.
[6] M. W. SMITH, M.D. CLIFF, NTO Based Explosive Formulations: A Technology Review,
MDSTO-TR-0796, DSTO, 2001
[7] US Patent No 4,733,610 (1988)
[8] H. BATHELT, F. VOLK AND M. WEINDEL, The ICT-Thermodynamic-Code (ICT Code)
Version 1.0, Fraunhofer ICT, 2000
[9] R. MEYER, J. KHLER, A. Homburg, Explosives, Wiley-VCH Verlag GmbH, Weinheim,
2002
[10] T. B. FEDOROFF, Encyclopedia of Explosives and Related Items, Picating Arsenal, Dover,
NJ, USA, Vol I, p. XVI, 1960
[11] "Recommendations on the Transport of Dangerous Goods, Manual of Tests and Criteria"
(ST/SG/AC.10/11/Rev.4) United Nations, New York and Geneva, 2003
11
597
Czech Republic
2006
[Content]
IMPLEMENTATION OF THE ABSOLUTE METHOD IN THE

BALLISTIC ANALISYS OF GUNPOWDER
T. Gawor, T. Wolszakiewicz, D. Korsze
Institute of Organic Industrial Chemistry, Annopol 6, 03 236 Warsaw, Poland
Abstract
The aim of the ballistic analysis of powder is to determine its combustible properties. All
the requirements of the way of calculating the preceding properties are included in the
MIL and STANAG international standards. Taking advantage of the MathCad program,
the authors created a program establishing characteristics and calculating the figures
included in the foregoing norms. The main focus was put on the study of the results
absolute measurements, not token into consideration in the software used hitherto.
Keyword : ballistic analysis of powder, absolute measurements
INTRODUCTION
The emphasis was put on constructing software (in the MathCad pack), enabling
determination of ballistic parameters of powder, taking into account the absolute methods,
which were not included in the so far applied MIL program. The advantage of this method is
the possibility of drawing conclusions concerning the purposefulness and all the properties
of the examined powder, without the necessity of providing the standard powder. This is
the only method of examining powders without using their standard. This work presents
the computations acquired on account of the authors own software, which is then compared
to the results achieved by using the MIL program.
RESULTS
The results were worked upon on the basis of the recording course of pressure in time
function, burning of three weight amounts of examined powder in a manometric bomb. The
recording courses of pressure p = f (t ) acquired in the MIL program are presented in Fig. 1.
Fig. 1. Graph of the function p = f (t ) from MIL program.
598
Implementation of the Absolute Method in the Ballistic Analisys of Gunpowder
Czech Republic
2006
[Content]
And the same courses p = f (t ) from MathCad program are presented in Fig. 2.
250
250
200
150
pb 1
100
50
0.005
0.01
tb 1
0.015
0.0186
Fig. 2. Graph of the function p = f (t ) from MathCad program.

p
dp
in order to analyse the examined
= f
dt
pm
powder. It describes the derivative of pressure versus in the function of the ratio of pressure
to maximum pressure in the function of the pressure derivative. This function makes it
possible to determine the absolute velocity of the examined powder. The graphs of this
function achieved thanks to MIL and MathCad programs are presented respectively in Fig. 3
and Fig. 4.
It is necessary to determine the function
Fig. 3. Graph of the function
p
dp
from MIL program.
= f
dt
pm
Tomasz Gawor, Tomasz Wolszakiewicz, D. Korsze
599

50
Czech Republic
2006
[Content]
50
40
30
dpb1
20
10
0.2
0.4
0.6
0.8
pb 1
1
1.01
max( spb 1)
Fig. 4. Graph of the function
p
dp
from MathCad program.
= f
dt
pm
The characteristics gained through this study make it possible to calculate among others:
absolute force AF
absolute velocity AQ
powder force f
powder covolumen .
The exemplified results of the absolute force AF and absolute velocity AQ, achieved
in MIL and MathCad programs are as follows:
Absolute force
Absolute velocity
MIL
240,8784 MPa
6,73031GPa2/s
MathCad
241,455 MPa
7,258 GPa2/s
The exemplary results of force an kovolumen are as follows:
Powder force
Powder covolumen
linear regression method
0,86535
MJ
kg
1,42445
cm 3
g
arithmetic mean
0,86538
MJ
kg
1,42422
cm 3
g
MIL
600
Implementation of the Absolute Method in the Ballistic Analisys of Gunpowder
Czech Republic
0,8653
MathCad
MJ
kg
2006
[Content]
1,4245
cm 3
g
Force and powder covolumen are calculated on the basis of Nobel-Abel Law. In order to
conduct the calculations, it is necessary to perform two burnings for two different charge
density, and proceed to drawing up the graph (pmax/)=f(pmax). The calculated points of the
specified charge densities and maximum pressures were used to calculate the force and
covolumen through linear regression.
1.25
1.2
max( pf n)
n
1.15
1.1
1.05
90
122
154
186
218
250
max( pf n)
Fig. 5. Graph of function (pmax/)=f(pmax).
CONCLUSIONS
1. The created software allows to use the absolute method in order to determine the
properties of the powder, which does not have its standard.
2. The results, with the accuracy of measurement at 0,25 %, are equal to the results
achieved thanks to the software used so far. The only difference of 8 % ratio in
case of absolute velocity results from application of different numeric methods
used in the calculations.
Tomasz Gawor, Tomasz Wolszakiewicz, D. Korsze
601
Czech Republic
2006
[Content]
GREEN POLYMER-BOUND EXPLOSIVE (PBX) FOR LASER

INITIATION
Mikhail A. Ilyushin, Igor V. Tselinsky, Irina V. Shugalei,
Anatoliy V. Chernay*and Valeria V.Toftunova
Saint-Petersburg State Institute of Technology, Russia
*National Mining University, Dnepropetrovsk, Ukraine
Abstract:
One of the modern and principal trends in the development of energetic materials is the
search of the so-called green environmentally friendly primary explosives. One of the
representatives of such green explosives is hexamethylenetriperoxidediamine
(HMTD). Thin-layer composite charges based on HMTD and optically transparent
polymeric matrix (PBX) are initiated by the beam of a pulse laser of low power. This
feature makes the application of this material in the safe optical ignition charges very
promising.
Keywords:
HMTD, polymer, laser initiation
INTRODUCTION
1.
Regular industrial primary explosives like lead azide, mercury fulminate, lead styphnate
all possess high toxicity and their application in blasting caps and pistons results in
environmental pollution with lead and mercury [1]. So the search for environmentally benign
initiation systems is one of the pressing current problems [2].
2.
DISCUSSION
2.1
Hexamethylenetriperoxidediamine, or
3,4,8,9,12,13-Hexaoxa-1,6-diaza-bicyclo-[4,4,4]-tetradecane (HMTD), is well known as

ecologically safe (green) primary explosive [3,4].
602
Czech Republic
2006
[Content]
O
O
O
N
O
O
Fig 1.
HMTD
The procedure for the synthesis of hexamethylenetriperoxidediamine is very simple. Its

preparation is relatively safe and can be carried out practically at any laboratory or chemical
enterprise. Starting materials can be purchased at any drugstore and no special equipment is
necessary [5]. In the 50-ties of XX century in the USSR HMTD was used as charge in
explosive clinches [3]. Charges of hexamethylenetriperoxidediamine were also used by
terrorist organizations in Israel, the USA, the United Kingdom [5]. Very often the know-how
of its production is mentioned on the web sites dealing with illegal production of explosives.
HMTD is usually prepared by mixing hexamine with 30%-hydrogen peroxide in the
presence of citric acid at room temperature for several hours according to the reaction given
below:
CH2
N
CH2
CH2
CH2
N
CH2
+ 6 H 2O2
2 NH
+ 3 H 2O +
CH2
+
CH2-O-O-CH 2
CH2-O-O-CH 2
CH2-O-O-CH 2
N
Scheme 1
White orthorhombic crystals of hexamethylenetriperoxidediamine have crystal density

of 1.57 g/cm3 and bulk density near 0.66 g/cm3. HMTD is essentially insoluble in water and
organic solvents (solubility in water 0.01%; in ether0.17%; in alcohol0.01%, in CCl4
0.013%; in acetic acid0.04%; in CHCl3 0.64%: in acetone - 0.33%). The hygroscopicity of
HMTD is negligible. Above 40 0C it slowly decomposes in the presence of moisture.
Hexamethylenetriperoxidediamine hydrolyzes in boiling water according to the reaction
given below:
Mikhail Ilyushin, Igor Tselinsky, Irina Shugalei, Anatoliy Chernay, . . .
603
Czech Republic
2006
[Content]
C6H12O6N2 + 3 H2O = 2 NH3 + 3 CH2O + 3 HCOOH

At room temperature it is hydrolyzed by acids and bases. HMTD does not react with
aqueous solutions of NaHCO3, Na2CO3 and NH3. Under storage both in air and in airtight
containers hexamethylenetriperoxidediamine is stable for a sufficiently long time. It is also
stable being kept under the layer of water for about 4 weeks at room temperature but after 4
months
of
storage
under
water
HMTD
completely
decomposes.
Hexamethylenetriperoxidediamine is readily reduced with Zn metal in HCl medium, with
aqueous and alcoholic solutions of SnCl2 and solutions of HBr or HI [5]. Dry HMTD does
not react with metals but in the presence of moisture HMTD easily corrodes metals. At high
temperature the product sublimates. For example, HMTD loses 0.5% of weight per day at 60
0
C, 1.3% at 75 0C, and 68% at 100 0C. The constant volume energy of formation of this
peroxide equals to 384.2 Kcal/kg (-1608 kJ/kg), Hf0 = -413.7 Kcal/kg (-1732 kJ/kg), heat
of explosion of HMTD (see reaction of explosion below) is equal to 1214 Kcal/kg (5080
kJ/kg):
C6H12O6N2 = 0,35 CO2 + 3,88 CO + 0,03 C2H2 + 0,55 CH4 + 2,44 H2

+ 1,42 H2O + 0,48 NH3 + 0,47 N2 + 0,58 HCN + 0,58 C
HMTD has the temperature of explosion 2370 0C, the volume of gaseous explosion
products 813 l/kg, and the rate of detonation 4510 m/s at the charge density equal to 0.88
g/cm3 and 5100 m/s at the charge density equal to 1.10 g/cm3 [6]. The ignition temperature
of the peroxide is ~200 0C (delay time is near 1 s). The temperature of the onset of intense
decomposition of HMTD depends on the weight of the sample. It is equal to 170 0C for the
samples weight under 1mg and to 125-140 0C for the samples weight of tens of mg under
heating rate 20 K/min [3,7]. HMTD is a primary explosive. Its minimum charge for
TETRYL initiation is near 0.05g. Fine crystals of hexamethylenetriperoxidediamine have
lower sensitivity to mechanical stimuli compared to Pb(N3)2 and Hg(CNO)2 and can be
pressed up to the density 1.3 g/cm3 but coarse crystals of HMTD exploded when being
pressed [5,7].
Traces of acids and metal ions slowly decompose the charges of HMTD. The chemical
stability of the charges was shown to be increased by addition of small amounts of MgO. But
this addition may increase the sensitivity of the explosive to mechanical stimuli. Acidic
impurities may be removed by washing HMTD with aqueous solutions of Na2CO3. The
subsequent addition of small amounts of aminoazoles results in coordination of the metal
ions capable of catalyzing the peroxide decomposition and so increases its stability [8].
Chemically stable HMTD can be prepared using such treatment.
2.2
Light-sensitive PBX based on hexamethylenetriperoxidediamine
Laser initiation is a comparatively new method for blowing up of explosives in blasting

caps. It is more safe than traditional means because false optical pulses which can initiate
explosives in detonators are absent. Laser initiation systems can be used in civil
construction, mining industry and drilling gas-oil bore-holes [9]. Our investigations showed
that HMTD was a light-sensitive primary explosive. The threshold of initiation of HMTD
charges with the beam of Q-switch neodymium glass laser ( = 1.06 m, = 30 ns) is near
604
Czech Republic
2006
[Content]
40 mJ/cm2. It is difficult to use pure HMTD in optical detonators because it possesses high
sensitivity to external stimuli and high corrosive ability. Using inert polymer matrix decides
both problems.
Optically transparent poly-2-methyl-5-vinyltetrazole with terminal
carboxylic groups (PVMT) and polymethylmethacrylate (PMMA) were used as inert
matricess in our studies.
CH2 CH
C N
N
N
N CH3
n
Fig 2. PVMT
COOCH3
|
-CH 2-C|
CH 3
Fig 3.
PMMA
We investigated thin-layer formulations of PBX based on HMTD which contained 1040% of a polymer, as light-sensitive explosives. PBX films of 1cm in diameter were used in
these experiments. 0.1 mm thick aluminimum foil sheets 20x20 mm in size were covered
with the PBX composition. Ballistic pendulum was used for measuring the momentum of
explosion products of PBX films. The laser beam was defocused in such a way that the
energy density in the center of the beam exceeded of the energy density at the edges of
samples by no more than 20%. The average energy density of the laser beam at the surface
of charges was twice as high as the threshold of initiation of PBX films. In this case planar
detonation wave of PBX films was realized. In the experiments the weights of PBX films
were varied from 35 to 110 mg/cm2. The dependence of momentum of explosion products
(J) on the weight of samples (ms) for HMTD film formulations containing 10% of PVNT
polymer is shown in Fig.4.
605
Fig 4.
Czech Republic
2006
[Content]
Dependence of the momentum of explosion products (J) on the weight of

samples (ms) for HMTD film formulations containing 10% of PVMT
polymer
The experimental results were aproximated by the first degree polynomial. The vertical
pillars mark 95% probability level. Part of the charge of PBX films burned and gave no
energy impulse.
3.
CONCLUSIONS
Thus, the light-sensitive formulations based on HMTD and optically transparent

polymers were suggested as enviromentally friendly charges for laser initiation systems [10].
Acknowledgements
This work was financially supported by the Ministry of Education and Science of
Russian Federation in the framework of the Intercollgiate Research and Development
Programm Development of Scientific Potential of Higher School.
REFERENCES
[1]
606
Falyakhov I. F., Gilmanov R. Z., Nesterov A.V., Fedorov O.V. and Khusainov R.M.
Technologies of preparation of enviromentally friendly primary explosives on the basis of
aromatic diazo compounds.// Proceedings of International Conference Modern problems of
technical chemistry 22-24 December 2004. Kazan. 2004. P.359-362. (in Russian).
Czech Republic
2006
[Content]
[2]
Zhoui S., Liu J. & Xie G. A Discussion on Explosive Materials and Safety // Theory and
practice of energetic materials./Ed. by Huang P., Li S. and Wang Y. Beijing/New York.
Science Press 2003. Vol. V. Part A. P. 183 185.
[3]
Bagal L.I. Chemistry and Technology of Primary Explosives. Moscow. Mashinostroenie. 1975.
455 p. (in Russian)
[4]
Bubnov P.F. Primary explosives. Moscow. GIOP.1940. 322 p. (in Russian)
[5]
Matyas
R.
Chemical
decomposition
of
triacetone
triperoxide
and
hexamethylenetpiperoxidediamine//Proceed. of the VI. Seminar New trends in research of
energetic materials. Pardubice. Chech Republic. April 22-24 2003. 2003. P. 241 247
[6]
Danilov Yu.N., Ilyushin M.A. and Tselinsky I.V. Industrial explosives. Part I. Primary
explosives. Text-book. Saint-Petersburg. Synthesis. 2004. 113 p.(in Russian)
[7]
Lefebvre M.H., Falmagne B and Smedys B. Sensitivities and performances of non-regular

explosives //Proceed. of the VII. Seminar New trends in research of energetic materials
Pardubice. Chech Republic. April 20-22 2004. 2004. Part 1.p 165 174.
[8]
Stroev Yu. P. Stabilization of peroxides for application in rocket propellants// VI International

Young Scientists Conference Man and Space. Abstracts. Dnepropetrovsk. NTsAOMU.2004.
p.370. (in Ukrainian).
[9]
Chernay A.V., Sobolev V.V., Chernay V.A., Ilyushin M.A. and Bunchuk Yu.P. Chapter 11.
Ignition of explosives by pulse lasers.// Physics of impulse treatment of materials/ Ed. Sobolev
V.V. Dnepropetrovsk. Art-Press. 2003.p.267-314. (in Russian)
[10] Ukrainian Patent No. 52997 . (Cl 06B45/15) 2003. Byul.1.
607
Czech Republic
2006
[Content]
SYNTHESIS OF 2,4-DINITRO-2,4-DIAZAPENTANE
FROM METHYLNITRAMINE AND DIHALOGENMETHANES
Zdenk Jalov*, Kamil Dudek**, Pavel Mareek**, Mat Ondrejkovi*,

Frantiek Lika***
* University of Pardubice, Institute of Energetic Materials, Studentsk 95,
CZ-532 10 Pardubice, Czech Republic
** Explosia, a.s., CZ-532 17 Pardubice-Semtn, Czech Republic
*** Institute of Chemical Technology, Department of Organic Chemistry, Technick 5
CZ-166 28 Prague, Czech Republic
Abstract:
Methylnitramine reacts with dihalogenmethanes in dimethylformamide in the presence
of potassium carbonate or bicarbonate as bases to produce 2,4-dinitro-2,4diazapentane (I). The same product I is formed in the reaction of potassium salt of
methylnitramine (potassium methylnitroamide, II) with dibromomethane.
Keywords:
1.
2,4-dinitro-2,4-diazapentane, methylnitramine, dibromomethane,

dichloromethane, synthesis
INTRODUCTION
2,4-Dinitro-2,4-diazapentane was first synthesized by Goodman in 1953 by

condensation of methylnitramine with formaldehyde in strong sulphuric acid.[1] Recently
Tartakovskij published further particulars of the influence of reaction conditions on the
reaction process.[2]
NO2
2 H3C N H + CH2O
H2SO4
NO2
NO2
H3C N CH2 N CH3

I
(I) has been also prepared by condensation of sodium or potassium methylsulphamates

with formaldehyde and subsequent nitrolysis of 2,4-diazapentane-2,4-disulfonic acid salts.[3]
SO3Na
2 H3C N H + CH2O
SO3Na
H3C N CH2 N CH3
SO3Na
NO2+
NO2
NO2
H3C N CH2 N CH3 +
Na2SO4
Reaction of N-(chloromethyl)methylnitramine and ammonium methylnitroamide in

acetone is another synthetic method of I.[4]
608
Synthesis of 2,4-dinitro-2,4-diazapentane from methylnitramine and . . .
NO2
2 H3C N CH2Cl +
H3C N NO2
Czech Republic
2006
NO2
NH4+
[Content]
NO2
H3C N CH2 N CH3

I
Dichloro- and dibromomethane represent other compounds in which carbon atom has
the same oxidation state as in formaldehyde. The reaction of above mentioned
dihalogenmethanes with alkylnitramines has not been described in the literature until now.
That's why we have tried to explore the reaction of methylnitramine with dihalogenmethanes
for the synthesis of I.
2.
RESULTS AND DISCUSSIONS
Acetone, butan-2-one, methanol, propan-1-ol and dimethylformamide (DMF) were tried

as solvents for reaction of methylnitramine with dibromomethane in the presence of
potassium carbonate. The target product I was detected in the reaction mixture using TLC
method only in the case of DMF and propan-1-ol. But only in the case of DMF the product I
was successfully isolated.
NO2
2 H3C N H + CH2X2
K2CO3
DMF
NO2
NO2
H3C N CH2 N CH3 +
KX
X = Cl, Br
Dichloro- and dibromomethane were chosen as halogenmethanes for the reaction with
methylnitramine. It is assumed that the reaction proceeds in three steps. The first step is the
formation of potassium methylnitramide (II). The second step is the substitution reaction of
II with dihalogenmethane affording N-(halogenmethyl)methylnitramine III. The final
product I is formed in step 3 by the reaction of the intermediate III with potassium
methylnitroamide II.
The results of experiments are summarized in Table 1. Higher yields of I are obtained
from the reactions of II with dibromomethane (examples 1-8) than with dichloromethane
(examples 9-10). The reaction of dihalogenmethane with methylnitramine takes place only in
the presence of a base. In the case when no base is used for the reaction of dibromomethane
with methylnitramine (example 1), only unreacted methylnitramine was detected. A base is
not necessary when potassium methylnitroamide II is employed (example 8). In most
examples, potassium carbonate was chosen as a base. Reaction takes place both with two
equivalents (examples 3-4) and with one equivalent of potassium carbonate to
methylnitramine (example 5). In the first case, the yields are higher. When
dihalogenmethane is used in overstoichiometric amount (examples 4 and 10), the yield of
product is higher than in the case of stoichiometric amount in corresponding experiments
(examples 2 and 9 respectively). No anticipated intermediate III was detected in the final
product. On the basis of the obtained results we suppose that the reaction of II with
2
Zdenk Jalov, Pavel Mareek, Kamil Dudek, Mat Ondrejkovi, . . .
609
Czech Republic
2006
[Content]
dibromomethane is the rate determining step of the whole reaction. The higher yields of I
were obtained when the reaction was carried at higher temperatures than at room
temperature (example 3). When potassium bicarbonate is used as a base, the reaction takes
place both with two equivalents (example 6) and with one equivalent (example 7) of
potassium bicarbonate to methylnitramine. Again, the yields are higher in the first case.
Step 1
NO2
H3C N H + K2CO3
H3C N NO2
K+
KHCO3
II
Step 2
H3C N NO2
NO2
H3C N CH2X + KX
K+ + CH2X2
II
III
Step 3
NO2
H3C N CH2X +
III
H3C N NO2
II
K+
NO2
NO2
H3C N CH2 N CH3 + KX
I
X = Cl, Br
3.
EXPERIMENTAL
A mixture of methylnitramine, dihalogenmethane, potassium carbonate (potassium

bicarbonate) in 50 ml of dimethylformamide was stirrred under the different conditions
given in Table 1. After the elapsed time, reaction mixture was cooled to room temperature,
filtered and filtrate was evaporated on a vacuum rotary evaporator. The crude residue was
treated with 30 ml of acetone, filtered to remove residual salts and concentrated in vacuo
leaving brownish oil which slowly crystallized on standing. The crude product contains
some amount of unreacted methylnitramine and solvent. The exact yield of 2,4-dinitro-2,4diazapentane was determined from HPLC analysis. Pure material can be obtained by
dissolving of crude sample in CH2Cl2 followed by extraction with water. Dichloromethane is
then evaporated yielding 2,4-dinitro-2,4-diazapentane, which is identical with standard
prepared by condensation of methylnitramine with formaldehyde in sulphuric acid[1,2],
melting 55-56C (lit.[2] 54C).
610
Czech Republic
2006
[Content]
Table I: Experimental results of reaction of methylnitramine with dihalogenmethanes

Example
Reaction Components
Molar
Ratio
Temperature, C Yield of I,
(Time, hour)
%
CH3NHNO2
CH2Br2
no base
5.7 g
6.5 g
2
1
0
70 (3)
CH3NHNO2
CH2Br2
K2CO3
5.7 g
6.5 g
10.4 g
2
1
2
70 (3)
58
CH3NHNO2
CH2Br2
K2CO3
5.7 g
6.5 g
10.4 g
2
1
2
20 (120)
18
CH3NHNO2
CH2Br2
K2CO3
5.7 g
10.4 g
10.4 g
2
2
2
CH3NHNO2
CH2Br2
K2CO3
5.7 g
6.5 g
6.2 g
2
1
1
CH3NHNO2
CH2Br2
KHCO3
5.7 g
6.5 g
7.5 g
2
1
2
CH3NHNO2
CH2Br2
KHCO3
5.7 g
6.5 g
3.8 g
2
1
1
[CH3NNO2]- K+
CH2Br2
no base
5.7 g
6.5 g
2
1
0
CH3NHNO2
CH2Cl2
K2CO3
5.7 g
3.2
10.4 g
2
1
2
10
CH3NHNO2
CH2Cl2
K2CO3
5.7 g
3.2
10.4
2
4
2
70 (3)
62
70 (3)
47
70 (3)
48
70 (3)
27
70 (3)
47
70 (3)
70C
3 hours
15 %
Zdenk Jalov, Pavel Mareek, Kamil Dudek, Mat Ondrejkovi, . . .
611
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
612
L. GOODMAN: J. Am. Chem. Soc. 75, 3019-3020, 1953.

V.A. TARTAKOVSKIJ, A.S. ERMAKOV, V.A. KOROBAN, F.R. ALIMOV, N.V. SIGAI: Izv. Ak.
Nauk, Ser. Khim., 1999-2001, 1993.
V.A. TARTAKOVSKIJ, A.S. ERMAKOV, N.V. SIGAI, O.N. VARFOLOMEEVA: Izv. Ak. Nauk, Ser.
Khim., 1085-1087, 2000.
J. DENKSTEIN, V. KADEBEK: Collection Czechoslov. Chem. Commun. 31, 2904-2911, 1966.
Czech Republic
2006
[Content]
APPLICATION OF SUPERCRITICAL FLUID EXTRACTION FOR

THE ISOLATION OF NITROCOMPOUNDS
Vra Jeov, Tom Fojtk, Ale Eisner, Martin Adam and Karel Ventura
Department of Analytical Chemistry, University of Pardubice, 532 10 Pardubice, CZ
Abstract:
2,4,6-trinitrotoluene (TNT) is a widely used military explosive and a predominant
contaminant of soil and ground water. TNT has been described to be toxic to aquatic
and terrestrial organisms in the environment. TNT can be degraded, mineralized or
conjugated in the presence of bacterium or fungus. The initial products formed from
TNT biotransformation are the reduced to amino derivatives (such as 4-amino-2,6dinitrotoluene or 2-amino-4,6-dinitrotoluene). 2,4,6-triaminotoluene (TAT) is observed
under strictly anaerobic conditions. To obtain nitrocompounds from soils is necessary to
use some extraction techniques. Supercritical fluid extraction (SFE) is one of modern
extraction techniques applied for isolation of analytes from different solid samples. The
combination of SFE with gas chromatography and mass detection (GC-MS) is able to
determinate low concentrations of analyzed compounds.
Keywords:
1.
2,4,6-trinitrotoluene, Supercritical fluid extraction, GC-MS
INTRODUCTION
The contamination of soils by all types of explosives is widely spread and the detection
of these compounds is a growing international problem. 2,4,6-trinitrotoluene (TNT) is one of
the most common explosives used in military munitions. These explosives have been
manufactured on a massive scale in the past. Presently soil and ground water is a worldwide
problem that started following intensive military activities (World War I and II, etc.). The
toxicity of the above energetic chemicals has been documented using different aquatic
organisms, terrestrial species such as earthworms.
TNT may be mixed with other explosives and it is constituent of many explosives. The
advantages of TNT include low cost, safety in handling, fairly high explosive power, good
chemical and thermal stability, compatibility with other explosives, a low melting point and
moderate toxicity. TNT is practically insoluble in water, but soluble in other solvents
(alcohol, acetone, etc.).
The present study describes the use of supercritical fluid extraction with gas
chromatography in combination with mass spectrometry for the isolation and determination
of TNT from soil samples. Other alternatives for the isolation analytes from solid matrix are
ultrasonic extraction and pressurised fluid extraction [1, 2].
Supercritical fluid extraction (SFE) is one of the extraction techniques usually applied
for the isolation of analytes from solid or semi-solid matrices. SFE used a supercritical fluid
as the extraction medium [3]. A supercritical fluid is a substance above its critical temperature
and pressure. Supercritical fluids can be seen as intermediates between gases and liquids.
1
Vra Jeov, Tom Fojtk, Ale Eisner, Martin Adam, Karel Ventura
613
Czech Republic
2006
[Content]
Most SFE extractions are performed using carbon dioxide (CO2) as the mobile phase. CO2
has a low supercritical temperature (31 C) and pressure (7,38 MPa). It is nontoxic and
available at high purity. Being nonpolar is a good solvent for the extraction of nonpolar and
moderately polar compounds (solvating power for polar solutes is rather poor). The
extraction efficiency of polar compounds by CO2 can be improved by the additions of small
quantities of polar organic solvents referred to as modifiers (methanol, ethanol, acetonitril,
acetone, hexane, toluene, chloroform, etc.). Other potential solvents for use in SFE (nitrous
oxide, ammonia, freons, methanol, etc.) are seldom used for reasons of safety, price,
impractical critical conditions, corrosivity, stability or purity.
Supercritical fluid extraction can be operated in two modes: static or dynamic [4]. In
static mode the supercritical fluid is held in an extraction cell for a certain amount of time
and then released to a collection device. In dynamic mode the supercritical fluid flows
continuously through the extraction cell and out into a collection device.
SFE can work in two modes: off- or on-line [4, 5]. In the on-line configuration, SFE is
coupled directly to an analytical instrument (such as gas chromatograph, supercritical fluid
chromatograph or high-performance liquid chromatograph). Off-line SFE is a stand-alone
extraction method independent of the analytical technique to be used. Off-line SFE is more
flexible and easier to perform than the on-line methods and the extract is available for
analysis by different methods. On-line coupling of supercritical fluid extraction with HPLC
is described as well for the determination of explosives in vapour phases too [6].
Two extraction techniques (ultrasonic extraction and supercritical fluid extraction) were
compared in the isolation of HMX (octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine) from
plant tissue [7]. Both extraction of plant tissues were found to be compatible in their
extraction efficiency of HMX (sonication was found to be slightly better at the higher
concentration >300 mg/kg).
This study interests in the optimalization of different parameters of SFE extraction and
GC-MS determination.
2.
EXPERIMENTAL PART
The parameters such as time, pressure, temperature and type of modifier influence the
extraction efficiency in the case of SFE. It is necessary to optimize these extraction
parameters for the higher proceeds.
The pressure and temperature of extraction were optimized in this study. The SFE
conditions for the pressure optimization were follows: 0,5 g of soil samples, extraction time
for 45 min in dynamic mode, temperature 100 C, no modifier and methanol as the
extraction solvent.
The extraction temperature was discovered at these conditions: 0,5 g of soil samples,
extraction time for 45 min in dynamic mode, pressure 25 MPa, no modifier and methanol as
the extraction solvent.
The extracts were transferred into 5-ml volumetric flask. The sample was 5 times diluted
and the internal standard (toluene, 0,02 l/ml) was added before analyzing by GC/MS.
614
Application of Supercritical Fluid Extraction for the Isolation of . . .
Czech Republic
2006
[Content]
The diluted extracts were analysed by using gas chromatograph GC 17A coupled with
mass spectrometry detector QP 5050A (both Shimadzu). Separations were performed on a
30m x 25m i.d. capillary column coated with a 0,25 m film of polymethylsiloxane (DB-5
MS).
The column oven was isothermally maintained at 50 C for 1 minutes and than
temperature increase for 40C/min to 200C that was kept for 3 minutes. The temperature of
injector was 180C and the temperature of interface was 230C.
It was used the electronic ionization in SCAN mode, the split injection 1:100 and the
helium as carrier gas.
The identification of compounds was based on the comparison their MS spectrum with
the spectrum in the library (NIST 62 and NIST 12, Shimadzu) and with MS spectrum of
standards.
3.
Since various parameters potentially affect the extraction process, the optimization of the
experimental conditions represents a critical step in the development of a SFE method. In
fact, pressure and temperature of the fluid, percentage of the modifier and the extraction
times are generally considered as the most important factors.
To evaluate the effects of pressure on the SFE efficiency of TNT, extraction were
performed at the pressures of 10 to 40 MPa by fixed temperature. As can be seen in Fig 1,
recoveries for TNT increased with increasing pressure. Because a better extraction efficiency
of TNT from soils was achieved at the extraction pressure of 25 MPa, this pressure was
chosen for latter experiments.
3
As/Aistd
0
10
15
20
25
30
35
40
Pressure [MPa]
Fig 1.
Effects of pressure on the SFE
615
Czech Republic
2006
[Content]
The variation temperature during SFE affects the density of fluid, the volatile property of
the analytes, and desorption of the analytes from the matrix. At a higher temperature, the
analytes become more volatile but the supercritical CO2 density decreases.
In this study different pressures (40 100 C) were evaluated to optimize the extraction
temperature by fixed pressure (25 MPa). Fig 2 shows the effect of temperature and higher
temperatures have no significant influence on the extraction efficiency. An extraction
temperature of 50 C was chosen for later experiments because of increased amount of TNT
from soil samples.
As/Aistd
2,5
1,5
1
40
50
60
70
80
90
100
Temperature [C]
Fig 2.
4.
Effects of temperature on the SFE recoveries
CONCLUSION
This study demonstrated that supercritical fluid extraction is suitable for the routine
extraction of solid samples. The optimalization of extraction parameters is necessary for the
high extraction efficiency. Extraction pressure and temperature were optimized at first. The
extraction time and the additions of modifier will be optimized in further experiments.
The SFE extracts were analysed with gas chromatography-mass spectrometry. It was
found that extraction temperature of 50 C and pressure of 25 MPa are sufficient conditions
for the extraction of 2,4,6-trinitrotoluene from soil samples.
Acknowledgement:
Experiments were performed thanks to financial support from the Ministry of Education,
Youth and Sports of the Czech Republic (Project MSM0021627502) and the Grant Agency of
Czech Republic (GAR 203/05/2106).
616
Application of Supercritical Fluid Extraction for the Isolation of . . .
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
M. E. WALSH: Determination of nitroaromatic, nitrateamine and nitrate ester explosives in

soil by gas chromatography and an electron capture detector, Talanta 54, 427-438, 2001
V. JEOV, A. EISNER, J. CHALNKOV, K. VENTURA: Analysis of nitroaromatic
compounds in soil environmental, Proceeding of the VIII. seminar: New trendes in research
of energetic material, 584-87, 2005
A. J. HANDLEY: Extraction Methods in Organic Analysis, Sheffield Academic Press,
Sheffield, 1999
J. R. DEAN: Extraction Methods for Environmental Analysis, John Wiley & Sons, Chichester,
1998
S. MITRA: Sample Preparation Techniques in Analytical Chemistry, John Wiley & Sons, New
Jersey, 2003
R. BATLLE, H. CARLSSON, E. HOLMGREN, A. COLMSJ, C. CRESCENZI: On-line
coupling of supercritical fluid extraction with high-performance liquid chromatography
for the determination of explosives in vapour phases, Journal of Chromatography A, 963,
73-78, 2002
A. HALASZ, C. GROOM, E. ZHOU, L. PAQUET, C. BEAULIEU, S. DESCHAMPS,
A. CORRIVEAU, S. THIBOUTOT, G. AMPLEMAN, C. DUBOIS, J. HAWARI:
Detection of explosives and their degradation products in soil environments, Journal of
Chromatography A, 963, 411-418, 2002
617
Czech Republic
2006
[Content]
POSSIBILITIES OF DEGRADATIONS PROCESS OF

NITROAROMATIC COMPOUNDS
Vra Jeov1, Ale Eisner1, Tom Fojtk1 and Karel Ventura1
Vt Matj2, Eva Kyslkov2 and Simona Voshlov2
1
Department of Analytical Chemistry, University of Pardubice, 532 10 Pardubice, CZ

2
ENVISAN-GEM, a.s., Radiov 7,102 31 Praha 10, CZ
Abstract:
One of major problems is the contamination of soils, ground water and air with
hazardous and toxic compounds. A large of the contamination comes from agricultural,
industrial and military activities. The possibility of microorganisms to degrade toxic
pollutants has been widely discussed.
Many synthetic organic compounds are mineralized by microbes and transformed into
CO2 and H2O. 2,4,6-trinitrotoluene (TNT) is a reactive molecule that biotransform
readily under both aerobic and anaerobic conditions to give aminodinitrotoluenes.
Various bacteria were applied into the solutions include different concentration of TNT.
The bacteria were removed with a centrifuge after defined time. The products from the
transformation of TNT were analysed with gas chromatography GC 17A with mass
spectrometry detector QP505A (both Shimadzu).
Keywords:
1.
degradations, nitro compounds
INTRODUCITON
A lot of aromatic compounds can be transformed or mineralized by microbes into CO2

and H2O. Aromatics substituted with nitro or amino groups can be used not only as a Csource but also as the N-source.
These compounds are generally recalcitrant to biological treatment and remain in the
nature, where there constitute sources of pollution due to both toxic and mutagenic effects on
human or other organisms. Relatively few microorganisms have been described as being
able to use nitroaromatic compounds as nitrogen or carbon and energy source.
As a major component of munitions, 2,4,6-trinirotoluene (TNT) has become a serious
environmental concern because of its widespread existence and persistence in soil and its
hazardous effects on a variety of biota. TNT is a military explosive and a predominant
contaminant of soil and ground water at many sites of TNT production and processing in
many countries. It is recalcitrant and has been shown to be toxic to aquatic and terrestrial
organisms in the environmental. Laboratory studies have demonstrated that TNT inhibits the
growth and survival of bacteria, fungi, soil microflora and plants [1].
TNT is a compound relatively refractory to biological degradation, although microbes
can metabolize it or conjugated into more complex product. The good reactivity of TNT is
triggered by the presence of nitro group (-NO2) functional groups, which can easily be
reduced to form reactive nitroso (-NO), hydroxylamino (-NHOH) and amino (-NH2)
618
Czech Republic
2006
[Content]
functional group. These functional groups react under both abiotic and biological conditions
to produce azo, azoxy, hydrazo derivates [2].
In almost all studies cases, the initial products formed from TNT biotransformation are
the reduced amino derivates (such as 4-amino-2,6-dinitrotoluene, 2-amino-4,6-dinitrotoluene
or 2,6-diamino-4-nitrotoluene). 2,4,6-triaminotolune is only observed under the anaerobic
condition [2].
This study describes different types of bacterium and their ability to degrade TNT under
various experimental conditions in the culture media. The selected types of bacterium were
applied into the soils contaminated with TNT. The nitrocompounds were isolated from soils
by two extraction techniques (ultrasonic extraction or pressurised fluid extraction). The
extracts were analyzed by gas chromatography and identificated by mass spectrometry.
Ultrasonic extraction (also known as sonication - USE) uses ultrasonic vibration to
ensure contact between the sample and the solvent [3]. Sonication is relatively fast, easy, it is
possibility extraction of more samples, noisiness of the apparatus and higher consume of
solvent. After extraction, the extract is filtrated or centrifuged, and some from of cleanup is
generally needed prior to analysis.
Pressurised fluid extraction (also known as accelerated solvent extraction PFE) used
convention solvents at elevated temperatures (80-150 C) and pressures (to 10 MPa) to
enhance the extraction of organic analytes from solids [4, 5]. The solvent boiling point is
increased under high pressure, so the extraction can be conducted at higher temperatures.
PFE is easy, fast technique (15-30 min), it uses minimal amount of solvent and sample size
(1-30 g). PFE system is fully automated.
2.
EXPERIMENTAL PART
2.1
Determination of TNT in the culture media
2.1.1. Conditions for degradation process

First of all were chosen the type of bacterium that is able to degrade TNT. Two groups
of samples for the degradation process of TNT were prepared. The first type included one
type of bacterium in different concentrations. The one tribe of bacterium was incubated in
culture media with shaking at 30 C for 72 hours. The second group contained the mixture
bacteriums type and the time of incubation was 148 hours at 30 C in centrifuge.
The samples were taken at defined intervals and the bacteria were struck dead. The
samples were shaken out into acetonitrile and analyzed by GC/MS.
2.1.2. Analysis of samples by GC/MS

The obtained extracts were analyzed by using the gas chromatograph GC 17A coupled
with mass spectrometry detector QP 5050A (Shimadzu) and GC/MS solution data system
(Shimadzu). The electronic ionization in SIM mode (for each nitrocompound were chosen
their characteristic ions) was used for the determination of the samples.
The identification of compounds was based on the comparison their mass spectrum with
the spectrum of standards solutions and also with the spectrum in the library (NIST 62 and
NIST 12, Shimadzu).
Vra Jeov, Tom Fojtk, Karel Ventura, Vt Matj, Eva Kyslkov, . . .
619
Czech Republic
2006
[Content]
Separations were performed on a 30m x 25m i.d. capillary column coated with a 0,25
m film of polymethylsiloxane (DB-5 MS). The volume of dosed extract was 1 l. It was
used the split injection 1:100 and the helium as carrier gas.
The column oven was isothermally maintained at 50 C for 2 minutes. Than the
temperature increase for 15 C/min to 170 C and for 5 C/min to 200 C that was kept for 2
minutes. The temperature of injector was 180 C and the temperature of interface was
230 C.
2.2
Determination of TNT in soil samples
2.2.1. Soil samples

Soils free of TNT were spiked with pure TNT six months before the analysis
(2 kg/100 kg). The spiked soil samples were used for the optimization of extraction
parameters (USE, PFE).
The best extraction conditions were observed for the analysis of soil samples included
some type of bacteria. After half a year to the soil samples were applied and different
nitrocompounds were isolated by using both extractions. The TNT and its derivates were
extracted from soil after the optimization and quantified by GC/MS.
2.2.2. Extraction techniques

Two extraction techniques were applied and compared for the isolation of
nitrocompounds. Type and volume of solvent and extraction time were optimized in the case
of ultrasonic extraction. Best conditions are followed: 20 ml of acetonitrile and extraction
time for 20 min. 5g of soils with solvent in beaker was put into the ultrasonic bath
(Ultrasonic Compact Cleaner). The extracts were evaporated to 2 ml and dosed into GC/MS.
These parameters such as extraction temperature, extraction time and type of solvent
influence the pressurised fluid extraction and were optimized. The real samples were
extracted on one PSE extractor and the best conditions were followed: 80 C extraction
temperature, 10 + 10 min extraction time, acetonitrile as extraction solvent and 5 g of soils.
The extracts were evaporated to 2 ml before the dosing into GC/MS.
2.2.3. Analysis of samples by GC/MS

The extracts of real samples were separated by same GC/MS system and column as in
the case of determination of TNT in culture media.
Other conditions on GC/MS were used for the optimization extraction parameters. It was
measured in the SCAN mode, the temperature of injector was 220 C and the temperature of
interface was 230 C. The column oven was isothermally maintained at 50 C for 1 minute
and than the temperature increase for 40 C/min to 200 C that was kept for 8 minutes.
The identification of nitro compounds was base on the comparison with MS spectrum of
standard and with MS spectrum in library.
The quantification of nitro compounds was done by using the calibration curve for each
of them.
620
3.
Czech Republic
2006
[Content]

3.1
Determination of TNT in the culture media
The mixture of all standards was analysed by GC/MS at first (Fig 1) and the parameters
of determination were found. The mixture of standards included the mononitrotoluenes,
dinitrotoluenes and 2,4,6-trinitrotoluene.
250000
3-MNT
4 4 0 0
4 3 0 0
200000
4 2 0 0
4 1 0 0
4 0 0 0
3 9 0 0
3 8 0 0
Intensity
150000
4-MNT
3 7 0 0
3 6 0 0
1 6 , 6
100000
2- MNT
1 6 , 7
1 6 , 8
1 6 , 9
1 7
2,3-DNT+2,4-DNT
50000
3,4-DNT
2,6-DNT
2,4,6-TNT
0
8
10
12
14
16
18
Retention tim e [m in]
Fig 1.
The chromatogram of standards
The liquid samples with various types of bacteria were obtained with using GC/MS in
SIM mode. A lot of degradation products were found in the first type of samples and one of
the chromatograms is shown at the Fig 2. Mass detector reported the formations of
mononitrotoluenes or dinitrotoluenes as one of major intermediate of TNT.
The same types of compounds were observed in the case of the analysis of second group
of samples with bacteria. The incubation time of the samples was longer and that is why it
were obtained less degradation products of TNT (Fig 3).
3,4-DNT
2,6-DNT
5000
7500
2,4,6-TNT
4750
2,4,6-TNT
Intensity
Intensity
6500
2,3-DNT+2,4-DNT
5500
3,4-DNT
2,6-DNT
3-MNT
4500
2-MNT 4-MNT
4500
4250
4000
3750
3500
3500
8
10
12
14
16
18
10
Fig 2.
The chromatogram of 1st bacterium sample
Fig 3.
The chromatogram of 2nd bacterium sample
12
14
16
18
621
3.2
Czech Republic
2006
[Content]
Determination of TNT in soil samples
Effective determination of TNT and some other degradation products of TNT is

achieved by the optimization of extraction factors extraction time, suitable solvent, etc.
The efficiency of PFE and USE was compared on one type of samples included different
nitrocompounds. The same nitrocompounds were identified and quantified in both extracts.
Fig 4 shows the using USE and PFE chromatogram is at Fig 5.
2,4,6-TNT
1000000
800000
Intensity
Intesity
1200000
2,3-DNT
600000
400000
3,4-DNT
2,6-DNT
200000
0
3
11
1200000
1000000
800000
600000
400000
200000
0
2,4,6-TNT
2,3-DNT
2,6-DNT
13
3,4-DNT
Retention time [min]
Retention time [min]
Fig 4. PFE chromatogram
Fig 5. USE chromatogram
11
13
The transformation of TNT in soils was studied by using ultrasonic extraction during
two months. The bacteria were added into the initial samples included only different
concentration of TNT. The extracts were prepared always after one-month incubation. The
effect of TNT transformation is shown at Fig 6.
1st month
2nd month
0,035
0,03
Amount [%]
0,025
0,02
0,015
0,01
0,005
TNT
3,4-DNT
2,3+2,4-DNT
2,6-DNT
TNT
3,4-DNT
2,3+2,4-DNT
2,6-DNT
TNT
3,4-DNT
2,3+2,4-DNT
2,6-DNT
Soil sample
Fig 6. Amount of nitrocompounds in different soil samples with using USE
622
4.
Czech Republic
2006
[Content]
CONCLUSION
The aim of this study was found some type of bacterium for the transformation of 2,4,6trinitrotluene to some nitrocompounds. The time and conditions of incubation were tested
and it was discover that longer time and different types of bacteria mean less degradation
products.
Soils for optimization of extraction parameters were spiked with pure TNT 6 month
before the analysis. The TNT and its derivates were extracted from real soils included
bacterium by ultrasonic extraction or pressurised fluid extraction and the extracts were
quantified by GC/MS. The same nitroaromatic compounds were presented in all samples and
the main metabolites of TNT such as mononitrotoluenes and dinitrotoluenes were found. The
concentrations of nitrocompounds were similar and the extraction efficiencies of extraction
techniques were comparable.
Acknowledgement:
Experiments were performed thanks to financial support from the Ministry of Education,
Youth and Sports of the Czech Republic (Project MSM0021627502) and the Grant Agency of
Czech Republic (GAR 203/05/2106).
REFERENCES
[1]
[2]
[3]
[4]
[5]
P. GONG, S. D. SICILIANO, Ch. W. GREER, L. PAQUET, J. HAWARI, G. I. SUNAHARA:

Effects and bioavailability of 2,4,6-trinitrotoluene in spiked and field-contaminated soils to
indigenous microorganisms, Environmental Toxicity and Chemistry, 18, 2681-2688, 1999
J. HAWARI, S. BEAUDET, A. HALASZ, S. THIBOUTOT, G. AMPLEMAN: Microbial
degradation of explosives-biotransformation versus mineralization, Appl Microbiol
Biotechnol, 54, 605-618,2000
S. MITRA: Sample Preparation Techniques in Analytical Chemistry, John Wiley & Sons, New
Jersey, 2003
J. R. DEAN: Extraction Methods for Environmental Analysis, John Wiley & Sons, Chichester,
1998
A. J. HANDLEY: Extraction Methods in Organic Analysis, Sheffield Academic Press,
Sheffield, 1999
623
Czech Republic
2006
[Content]
PREPARATION OF LOW-MELTING COMPOSITIONS BASED ON

1,3,3-TRINITROAZETIDINE
D.V. Katorov, N.A. Vyalova, G.F. Rudakov, A.V. Ladonin, V.F. Zhilin, E.V. Veselova
Mendeleev University of Chemical Technology, 9 Miusskaya Square, 125047 Moscow, Russia
Abstract:
Investigations of TNAZ/AzDNAZ mixtures with different TNAZ contents revealed an
eutectic composition TNAZ/AzDNAZ 35/65 with melting point of 61oC. ombined
introduction of nitro- and azido groups followed by nitration of tBuDNAZ/tBuAzNAZ
mixtures in N2O5/CH3CN and NH4NO3/(CH3CO)2O systems has been investigated.
Possibility of synthesis TNAZ/AzDNAZ mixtures of various contents has been shown.
Keywords:
1.
TNAZ, nitration, eutectic mixture.
INTRODUCTION
Recently intensive searches have been conducted of the new explosive materials
possessing high energy of explosive transformation, density and stability. The most
interesting compounds obtaining were CL-20, NTO, TNAZ, belonging to a class of cyclic
nitroamines.
1,3,3-Trinitroazetidine has been synthesized in 1984 in the USA and represents fourmembered heterocyclic nitroamine containing gem-dinitro group. Now works with TNAZ
are carried on in the USA, France, Germany, South Korea and China.
It is proposed to use TNAZ in artillery powder systems[7,10], as a component of
pyrotechnic mixtures[8,9], explosive compositions[6] and solid propellants. Also TNAZ may
be used as a softener in mixtures with CL-20 and in gunpowders containing 4% of TNAZ[7].
One of advantages of TNAZ is rather low melting point (98 - 101[1-4]) at sufficient
power (approximately as RDX) and high density (1.84 g/cm3[1,3]), that makes it attractive to
use in some special compositions.
TNAZ is capable of forming eutectic mixtures with high energetic substances, such as
HMX, 2,4-dinitroimidazole, tetryl. TNAZ/tetryl eutectic composition with the melting point
81.5 (test DSC)[5].
2.
RESULTS AND DISTCUSSIONS
Recently we synthesized a new energetic substance 3-azido-1,3-dinitroazetidine

(AzDNAZ). This substance has been obtained by nitration of 3-azido-1-tert-butyl-3nitroazetidine (tBuAzNAZ). The oxidative substitution was carried out in the conditions
similar to prepartion of gem-dinitrocompound[11] (scheme 1):
624

O
O2N
OH
NaOH
O2N
+
Na
N
Czech Republic
N3
NaN3
K3[Fe(CN)]6
2006
[Content]
O2N
+
NO2
N3
N
NO2
AzDNAZ
tBuAzNAZ
Scheme 1. 1. Synthesis AzDNAZ[11]

Nitration was carried out in systems N2O5/CH3CN and NH4NO3/(CH3CO)2O. The yield
AzDNAZ achieved 60 %, and the melting point was 77.5-80.[11]
DSC analyses of various mixtures of -nitroazide with TNAZ have shown that the
compounds form an eutectic mixture with mass ratio TNAZ/AzDNAZ as 35/65. (Fig. 1-4).
Wg^-1
0
-5
-10
50
Fig 1.
100
150
200
250
DSC analysis of TNAZ
Wg^-1
20
10
-10
30
Fig 2.
50
70
90
110
130
150
170
190
210
230 C
DSC analysis of AzDNAZ
Dmitrii Katorov, N. Vyalova, G. Rudakov, A. Ladonin, V. Zhilin, E. Veselova
625
Czech Republic
2006
[Content]
Wg^-1
-5
50
Fig 3.
100
150
200
250
DSC analysis of mixture TNAZ/AzDNAZ 35/65
105
melting, oC
95
85
75
65
55
0
20
40
60
80
100
TNAZ, %
Fig 4.
TNAZ/AzDNAZ melting-point diagram
As oxidizing azidation and nitration take place in similar conditions there is an

opportunity of preparation of mixture TNAZ/AzDNAZ simultaneously in one-pot.
Nitroisobutylglycerol has been chosen as the initial substance for synthesis of
TNAZ/AzDNAZ mixture (scheme 2).
626

O2 N
OH
NO2
HO
NH2 + CH2O
Czech Republic
O2 N
HCl
OH
N
OH
1
OH
NO2
O2N
N *HCl
N3
O2N
+
2) NaNO2 K3Fe(CN)6
NaN3
P(Ph)3
N *HCl
1)NaOH
4
O2N
[NO2]+
NO2
N3
O 2N
+
OH
O2 N
DEAD
NH *HCl
[Content]
OH
OH
O2 N
2006
N
NO2
N
NO2
TNAZ / AzDNAZ
tBuDNAZ / tBuAzNAZ
Scheme 2. Synthesis of TNAZ/AzDNAZ mixtures

Research of kinetics of oxidizing nitration and azidation reactions, under optimum
conditions found (5 mol NaOH, 5 mol NaNO2 or NaN3, 5 mol K3Fe (CN)6) has shown that
nitration proceeds 3 times faster than azidation.
4
tBuDNAZ, %
0
0
Fig 5.
50
100
150
200
time of reacnion, min
250
300
Kinetics of nitration
tBuAzNAZ, % .
0
0
Fig 6.
200
400
600
time of reaction, min
800
1000
Kinetics of azidation
627
Czech Republic
2006
[Content]
Thus we made an assumption that various compositions of geminal nitro- and

azidoazetidines can be obtained by not only changing the ratio NaN3 and NaNO2, but also
the way of their addition.
At a simultaneous addition of NaN3 and NaNO2 we obtained the mass ratio of the
products as 80/20 (tBuDNAZ/tBuAzNAZ), with the total yield of the target product
of 75%, which corresponds to the yield of oxidizing nitration.
No simultaneous addition of NaN3 and NaNO2 leads to increase in the content of
azide in comparison with gem-dinitrocompound without increase in sodium azide
(Fig 7).
100
contents tBuAzNAZ,%
80
60
40
20
0
0
Interval betw een addition NaN3 and NaNO2, h
Fig 7.
Dependence of tBuDNAZ/tBuAzNAZ mixture composition on the

interval between addition of NaN 3 and NaNO 2
It has been shown that the 2-hours pause between addition of NaN3 and NaNO2 gives
rise to formation of a mixture of 85% tBuAzNAZ and 15% tBuDNAZ. The composition of
products was determined by HPLC method.
The earlier researches have shown that the most convenient systems for preparation
TNAZ are: HNO3/Ac2O 6/1, NH4NO3/Ac2O, and also systems containing nitric anhydride.
However, the maximum yield of TNAZ, 82%, was achieved in system NH4NO3/Ac2O
(yields were 60-80% by on different sources).
It has been shown that nitration of tBuAzNAZ occurs in this system with a small yield
(20 30%) that has been explained by low thermal stability of geminal nitroazide in acetic
anhydride. This behaviour of tBuAzNAZ can essentially change composition of the final
mixture. Therefore, it has been carried out nitration of mixtures with high contents of
tBuAzNAZ. As a result solid and liquid phases were obtained and analysed by HPLC.
Results of the analysis have shown that the solid product represents a mixture with
AzDNAZ content 92%, and the composition of the liquid phase is close to eutectic
(TNAZ/AzDNAZ 67/37).
The material balance has shown that the yield of TNAZ does not vary, and the yield of
AzDNAZ is 48%, 2 times higher than at nitration of pure tBuAzNAZ.
628
Czech Republic
2006
[Content]
EXPERIMENTAL PART.
3.
Melting points were run at 3C/min on a Boatius heating microtable. Infrared spectra
were obtained on a Thermo Nicolet 360 FT IR spectrophotometer using KBr disks.
HPLC analysis was carried out on Milichrom-4 chromatograph, column UVM
3780/12, 2.060 mm, speed of submission of eluent 200 L/min. Acetonitrile/water 1/1,
acetonitrile/water 35/65 as eluents was used. Thin-layer chromatography (TLC) was carried
out using DC-Alufolien TLC plates from Merck.
Differential Scanning Calorimetry (DSC) experiments were performed on a Mettler
Toledo DSC 822e using aluminium sample pans. The instrument was calibrated daily using
indium and aluminium standards. Traces were recorded at ramp rate of 10C/min between
25 to 200C under a flowing N2 gas atmosphere (50 ml/min). The sample weight was 1.28
mg.
3.1
Synthesis of a tBuDNAZ/tBuAzNAZ mixture.
To a solution 2 g (8.9 mmol) (1-tert-butyl-3-nitroazetidin-3-yl)methanol hydrochloride

(4) in 4 ml of water, solution NaOH 1.78 g (44.5 mmol) in 6 ml of water is added at stirring.
The white solid precipitated. After dissolution of a deposit the solution NaN3 1.7 g (26.7
mmol) in 4 ml of water is added. The mixture is maintained 10 min. and added to solution
K3[Fe(CN)6] 14.6 g (44.5 mmol) in 50 ml of water. The mixture is maintained at stirring
within 2 hours. Then the solution of NaNO2 1.2 g (17.8 mmol) is added and maintained for 2
hours. Reaction mixture is extracted with dichloromethane (4 x 30 ml). The extract is dried
above MgSO4. Removal of the solvent gave a yellow oil (1.4 g, 75%). IR: 2969 (-CH3), 2128
(as-N3), 1574 (as-NO2), 1335 (s-NO2) cm-1.
3.2
Synthesis of a TNAZ/AzDNAZ mixture.
To well stirred solution tBuDNAZ/tBuAzNAZ (84/16 1.3 g, 6.5 mmol) in acetic

anhydride (10 ml 59 mmol), nitrate of ammonium 0.8 g (9.8 mmol) was added. The mixture
was heating up to 75 and stirring 1.5 hours. Then the reaction mixture is poured out in 15
ml of water. Then it is cooled to 0. The white solid precipitated out and was filtered,
washed with ice water (2 x 20 ml) and dried above 25. The white solid 0.2g (19.4%) is
obtained, m.p. 6072C. Filtrate was extracted with ethyl acetate (3 x 15 ml). The extract is
washed by 1% Na solution (40 ml) and the same volume of water. The extract is dried
above MgSO4 and removal of the solvent gave a yellow oil (0.4 g, 39%). A total yield 0.6 g
(58,4%). IR: 2136 (vas N3), 1604-1538 (vas CNO2) cm-1.
4.
CONCLUSION
New low-melting composition based on 1,3,3-trinitroazetidine has been proposed. The

melting-point diagram of mixtures of TNAZ/AzDNAZ has been obtained.
Method of preparation of mixtures TNAZ/AzDNAZ by introduction of nitro- and azidogroups followed by nitration has been developed.
629

5.
Czech Republic
2006
[Content]
REFERENCE
[1]
D.S. WATT AND M.D. CLIFF: Evaluation of 1,3,3-Trinitroazetidine (TNAZ) - A High

Performance Melt-Castable Explosive, Aeronautical and Maritime Research Laboratory.
DSTO-TR-1000. 2000.
[2] K. SCHMID AND D. KASCHMIEDER: Synthesis and properties of TNAZ, 31st International
Annual Conference of ICT. Karlsruhe, FRG. p.10, 2000.
[3] P. MARECHEK AND K. DUDEK Cast TNAZ mixtures, 5th International Seminar NTREM,
Pardubice, Czech Republic.p.165-169, 2002
[4] T.G.ARCHILBALD, R. GILARDI, K. BAUM AND C. GEORGE: Synthesis and X-ray Crystal
Structure of 1,3,3-Trinitroazetidine, J. Org. Chem., V55, p.2920-2924, 1990,
[5] R. D. CHAPMAN, (TPL, Inc., Albuquerque, NM USA). Report 1994, TPL-FR-2049; Order No.
AD - A 286 004, 69 pp. (Eng.).
[6] US patent 5,997,668 (1999).
[7] US patent 5,690,868 (1997).
[8] US patent 6,425,601 (2002).
[9] US patent 6,513,602 (2003).
[10] US patent 5,798,481 (1998).
[11] D.V. KATOROV, G.F. RUDAKOV, A.V. LADONIN AND V.F. ZHILIN: 3-Azido-1,3dinitroazetidine new polyfunctional nitriamine, Material of bulletin Russian scientific and
technical conference Progress in chemistry and chemical technology., Moscow, Part 1, p.3942, 2005
630
Czech Republic
2006
[Content]
SALTS OF 5-NITROTETRAZOLATE
Thomas M. Klapoetke*, Peter Mayer*, Kurt Polborn* and Jan M. Welch*
*University of Munich, Butenandtstr. 5-13 (D), D-81377 Munich, Germany
Abstract:
The previously well described sodium dihydrate (1) and silver (2) salts of 5nitrotetrazolate were prepared and characterized by Raman spectroscopy. Additionally
X-ray structure determination and measurement of the 13C and 15N solution NMR
spectra were carried out for 1. The previously incompletely described or hereto
unknown 5-nitrotetrazolate salts of 1,4-dimethyl-5-aminotetrazolium (3), guanidinium
(4), aminoguanidinium (5), diaminoguanidinium (6), ammonium (7), hydrazinium (8)
and tetraphenylphosphonium (9) by metathesis of the corresponding hydrohalide
(iodide, bromide or chloride) with silver 5-nitrotetrazolate and characterized Raman
spectroscopy and X-ray structure determination. Lastly the incompletely described
potassium salt of 5-nitrotetrazolate (10) was also synthesized and characterized by
Raman spectroscopy and X-ray structure determination. All compounds were found to
be air and moisture stable.
Keywords:
1.
Aminotetrazole, Explosives, Nitrotetrazole, X-ray
INTRODUCTION
The synthesis of energetic, non-nuclear materials for possible military application has
been a long term goal in our research group.[1-3] We have recently prepared and determined
the properties and structures a variety of compounds based on and derived from 5-amino(1H)-tetrazole.[4-6] The 5-nitrotetrazolate anion has been the subject of most recent
investigations. Although the 5-nitrotetrazolate anion is well established and its metal salts
have been extensively investigated for use as, initiators, primary explosives and energetic
sensitizers,[7-10] the weakly basic 5-nitrotetrazolate anion has only recently been investigated
as a component of nitrogen rich ionic energetic materials.[11-13] Here we present our findings
in the area of nitrogen-rich ionic materials.
Thomas Klapoetke, Peter Mayer, Kurt Polborn, Jan Wlech
631
2.
Czech Republic
2006
[Content]

2.1
Synthesis
The synthesis of 1 and 2 were carried out according to literature methods.[9] Compound
10 was synthesized by appropriate modification (Substitution of potassium hydroxide for
sodium hydroxide during digestion step) of the literature method for 1. Salts 3-9 were
synthesized generally according to eq. (1).
NO2
NO2
C
N Ag
N
N
MeOH
+ [Cat][X]
+
N Cat + AgX
(1)
NH2
Cat=
NH2
C
N
N
N
(4)
(3)
H2 N
NH2
NH2
NH2
H2 N
(5)
N
H
X= Cl-
NH2
H2 N
NH2
N
H
NH4
2.2
(6)
BrI-
N
H
(7)
H3NNH2
(8)
X-ray Structures
The molecular structures of 1 and 3-10 in the crystalline state were determined by X-ray.
Generally all structures show a planar 5-nitrotetrazolate moiety as well as when possible
hydrogen bonding between cations and anions. The shortest donor-acceptor distances are
most often observed between the donor and N1 of the tetrazole ring in 5-nitrotetrazolate.
However, there are also notable hydrogen bonding interactions between the oxygen atoms of
the nitro group and donor groups in several cases and less frequently at the N2 position of
the tetrazole ring.
2.2.1. Sodium 5-nitrotetrazolate Dihydrate (1)

Crystals suitable for X-ray structure analysis were obtained by recrystallization of crude
1 from a mixture of acetone and diethyl ether into which additional diethyl ether was
allowed to diffuse. 1 Crystallizes in triclinic space group P-1 with two formula units per unit
cell. Each unit cell has dimensions of a = 6.4192(3) , b = 7.8838(2) , c = 8.1896(3) ,
angles = 115.98, = 96.52 and = 110.19 and a volume of 332.0(6) 3. The structure
was determined from data with Rint = 0.0599 in P-1, R1 and wR2 for all data equal to 0.0573
and 0.1300 respectively and R1 and wR2 for I>2(I) equal to 0.0437 and 0.1198
respectively.
632
Fig 1.
Czech Republic
2006
[Content]
Ortep plot of a formula unit of 1 with thermal ellipsoids showing 50%

probability
2.2.2. 1,4-dimethyl-5-aminotetrazolium 5-nitrotetrazolate (3)

Crystals suitable for X-ray structure analysis were obtained by treating a solution of 3 in
methanol with diethyl ether until the solution was saturated, then allowing additional diethyl
ether to diffuse into the solution. 3 Crystallizes in the orthorhombic space group P212121
with four formula units per unit cell. Each unit cell has dimensions of a = 5.709(1) , b =
12.866(2) , c = 13.328(2) and a volume of 978.9(3) 3. The structure was determined
from data with Rint = 0.0304 in P212121, R1 and wR2 for all data equal to 0.0454 and 0.0973
respectively and R1 and wR2 for I>2(I) equal to 0.0389 and 0.0926 respectively.
Fig 2.

probability
2.2.3. Guanidinium 5-nitrotetrazolate (4)

Crystals suitable for X-ray structure analysis were obtained
crystallizes in the monoclinic space group C2/c with four formula
unit cell has dimensions of a = 9.8240(3) , b = 11.7601(4)
monoclinic angle = 117.604(2) and a volume of 703.72(4)
3
as for 3. Compound 4
units per unit cell. Each
, c = 6.8736(2) , a
3. The structure was
633
Czech Republic
2006
[Content]
determined from data with Rint = 0.0922 in C2/c, R1 and wR2 for all data equal to 0.0395 and
0.1039 respectively and R1 and wR2 for I>2(I) equal to 0.0370 and 0.1010 respectively.
Fig 3.

probability
2.2.4. Aminoguanidinium 5-nitrotetrazolate (5)

Crystals suitable for X-ray structure analysis were obtained as for 3. Compound 5
crystallizes in the monoclinic space group P21/n with four formula units per unit cell. Each
unit cell has dimensions of a = 6.9447(5) , b = 6.3716(6) , c = 17.320(2) , a monoclinic
angle = 99.152(8) and a volume of 756.6(2) 3. The structure was determined from data
with Rint = 0.0493 in P21/n, R1 and wR2 for all data equal to 0.0499 and 0.0872 respectively
and R1 and wR2 for I>2(I) equal to 0.0356 and 0.0831 respectively.
Fig 4.

probability
2.2.5. Diaminoguanidinium 5-nitrotetrazolate (6)

Crystals suitable for X-ray structure analysis were obtained by layering a solution of 6 in
methanol and water with diethyl ether and allowing it to stand for several days. 6
Crystallizes in the orthorhombic space group Pbca with 16 formula units per unit cell. Each
634
Czech Republic
2006
[Content]
unit cell has dimensions of a = 18.508(4) , b = 7.045(2) , c = 26.114(4) and a volume

of 3405(2) 3. The structure was determined from data with Rint = 0.____ in Pbca, R1 and
wR2 for all data equal to 0.0660 and 0.1087 respectively and R1 and wR2 for I>2(I) equal
to 0.0348 and 0.0924 respectively.
Fig 5.
Ortep plot of the two inequivalent formula units of in crystaline 6 with

thermal ellipsoids showing 50% probability
2.2.6. Ammonium 5-nitrotetrazolate (7)

crystallizes in the monoclinic space group P21/c with four formula units per unit cell. Each
unit cell has dimensions of a = 4.8436(2) , b = 13.7981(6) , c = 8.0853(4) , a
monoclinic angle = 97.256(2) and a volume of 536.03(4) 3. The structure was
determined from data with Rint = 0.1018 in P21/c, R1 and wR2 for all data equal to 0.0559
respectively.
635
Fig 6.
Czech Republic
2006
[Content]

probability
2.2.7. Hydrazinium 5-nitrotetrazolate (8)

respectively.
Fig 7.

probability
2.2.8. Tetraphenylphosphonium 5-nitrotetrazolate (9)

Crystals suitable for X-ray structure analysis were obtained by treating a solution of 9 in
acetone with diethyl ether until the solution was saturated, then allowing additional diethyl
ether to diffuse into the solution. 9 Crystallizes in the orthorhombic space group P212121
with four formula units per unit cell. Each unit cell has dimensions of a = 7.7417(1) , b =
13.6314(2) , c = 21.2740(3) and a volume of 2245.05(5) 3. The structure was
determined from data with Rint = 0.0304 in P212121, R1 and wR2 for all data equal to 0.0454
respectively.
636
Fig 8.
Czech Republic
2006
[Content]

probability
2.2.9. Potassium 5-nitrotetrazolate (10)

respectively.
Fig 9.
2.3
7

probability
Analytical and Spectroscopic Data
637
Czech Republic
2006
[Content]
2.3.1. Sodium 5-nitrotetrazolate dihydrate (1)

Table 1. Spectroscopic Data for (1)
13
C{1H} NMR (Acetone-d6, 100.63 MHz, 25C, TMS) /ppm: 167.5

N{1H} NMR (Acetone-d6, 40.55 MHz, 25C, MeNO2) /ppm: 5, -27, -73
Raman (200 mW, 90, 1064 nm) /cm1 (rel. int.): 1555 (12), 1452 (7), 1420 (100),
1322 (5), 1194 (2), 1173 (2), 1068 (60), 1054 (22), 843 (8), 774 (3), 544 (3), 456 (3),
263 (4), 186 (4), 150 (7)
15
2.3.2. Silver 5-nitrotetrazolate (2)

Table 2. Tabulated Raman signals for (2)
1432 (100), 1331 (13), 1203 (23), 1184 (10), 1088 (27), 1070 (48), 1042 (5), 836
(15), 771 (5) 549 (3), 250 (10), 169 (17)
2.3.3. 1,4-dimethyl-5-aminotetrazolium 5-nitrotetrazolate (3)

Table 3. Spectroscopic and analytical data for (3)
C4H8N10O2 (228.18), calcd., found / %: C 21.1, 20.9; H 3.5, 3.6; N 61.4, 61.2
Raman (200 mW, 90, 1064 nm) /cm1 (rel. int.):
2.3.4. Guanidinium 5-nitrotetrazolate (4)

1873 (0), 1643 (1), 1569 (3), 1541 (5), 1499 (1), 1419 (100), 1317 (3), 1179 (3),
1051 (41), 1042 (32), 1016 (25), 843 (6), 773 (1), 545 (9), 455 (2), 321 (2) 268 (3),
135 (7), 116 (3)
2.3.5. Aminoguanidinium 5-nitrotetrazolate

1674 (1), 1630 (2), 1559 (2), 1533 (7), 1490 (2), 1422 (100), 1319 (5), 1285 (1),
1204 (2), 1177 (3), 1159 (4), 1063 (45), 1041 (44), 959 (7), 838 (8), 773 (3) 732 (1),
626 (3), 539 (3), 497 (4), 450 (3), 348 (4), 261 (2), 243 (4), 206 (3), 133 (7)
2.3.6. Diaminoguanidinium 5-nitrotetrazolate (6)

1627 (1), 1546 (7), 1487 (2), 1447 (4), 1421 (100), 1321 (4), 1286 (1), 1180 (3),
638
Czech Republic
2006
[Content]
1159 (2), 1063 (30), 1041 (28), 1034 (14), 929 (5), 837 (5), 774 (3), 546 (3), 449 (3),
406 (2), 368 (2), 258 (3), 137 (2)
2.3.7. Ammonium 5-nitrotetrazolate (7)

1865 (0), 1543 (9), 1436 (11), 1419 (100), 1320 (5), 1183 (2), 1168 (4), 1065 (46),
1050 (58), 837 (7), 774 (3), 734 (1), 542 (2), 456 (2), 316 (2), 242 (5), 123 (10)
2.3.8. Hydrazinium 5-nitrotetrazolate (8)

1873 (2), 1633 (4), 1532 (6), 1446 (4), 1420 (100), 1316 (7), 1291 (1), 1177 (3),
1162 (5), 1122 (1), 1062 (58), 1046 (41), 1029 (5), 980 (7), 840 (9), 773 (3), 542 (3),
454 (3), 258 (4), 174 (2), 129 (3)
2.3.9. Tetraphenylphosphonium 5-nitrotetrazolate (9)

2964 (3), 1588 (78), 1577 (23), 1521 (12), 1487 (4), 1402 (100), 1194 (11), 1162
(15), 1110 (14), 1098 (28), 1039 (30), 1030 (51), 1003 (94), 939 (5), 835 (11), 746
(4), 728 (7), 679 (19), 616 (15), 533 (6), 463 (5), 282 (21), 262 (35), 245 (26), 207
(24), 185 (10), 127 (24)
2.3.10. Potassium 5-nitrotetrazolate (10)

1419 (100), 1319 (3), 1174 (3), 1164 (2), 1067 (10), 1052 (18), 1028 (39), 842 (6),
774 (2), 736 (1), 541 (3), 454 (2), 246 (3), 150 (5)
3.
CONCLUSIONS
Several previously known 5-nitrotetrazole salts have been structurally and

spectroscopically characterized. Additionally several new salts of 5-nitrotetrazolate have
been synthesized and structurally and spectroscopically characterized. Pending further
investigation into their energetic properties many of these new nitrogen rich 5-nitrotetrazole
salts might serve as promising new ionic, nitrogen-rich energetic materials.
639
4.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Czech Republic
2006
[Content]
REFERENCES
W.-H. Walther, T. M. Klaptke, and G. Holl, Physical Chemical Characterization and
Explosive Properties of Hydrazine Azide, in: Proceedings 29th International Annual Conference
of ICT, June 30 July 3, 1998, Energetic Materials Production, Processing and
Characterization, Fraunhofer Institut fr Chemische Technologie: Karlsruhe, Germany, 1998,
pages 134-1 to 134-1.
T. M. Klaptke, B. Krumm, G. Holl, and M. Kaiser, Synthesis, Characterization and Quantum
Chemical Computations of Hexanitrohexaaza-Isowurtzitane, in: Proceedings 30th International
Annual Conference of ICT, June 30 July 3, 1999, Energetic Materials - Modeling of
Phenomena, Experimental Characterization, Environmental Engineering, Fraunhofer Institut
fr Chemische Technologie: Karlsruhe, Germany, 1999.
T. Habereder, A. Hammerl, G. Holl, T. M. Klaptke, J. Knizek, P. Mayer, and H. Nth,
Methylsubstituted Hyrdazinium Azide Derivatives, in: Proceedings 31st International Annual
Conference of ICT, June 27 30, 2000, Energetic Materials Analysis, Diagnostics and
Testing, Fraunhofer Institut fr Chemische Technologie: Karlsruhe, Germany, 2000, pages 1501 to 150-7.
T. M. Klaptke, New highly energetic materials based on hydrazine and tetrazole compounds,
ACS Spring 2003 Award Symposium, New Orleans, March 23 27, 2003.
J. Geith, T. M. Klaptke, J. Weigand, G. Holl, Calculation of the Detonation Velocities and
Detonation Pressures of Dinitrobiuret (DNB) and Diaminotetrazole Nitrate (HDAT-NO3),
Propellants Explosives and Pyrotechnics 2004, 29, 3-8.
M. v. Denffer, T. M. Klaptke, G. Kramer, G. Spie, J. M. Welch, G. Heeb, Improved
Synthesis and X-Ray Structure of 5-Aminotetrazolium Nitrate, Propellants Explosives and
Pyrotechnics 2005, 30, 191-1195.
E. v. Herz, C-nitrotetrazole Compounds, US Patent 2066954 (1932)
L. R. Bates, J. M. Jenkins, Produciton of 5-nitrotetrazole salts, US Patent 4094879 (1978).
W. H. Gilligan, M. J. Kamlet, Synthesis of Mercuric 5-nitrotetrazole, Technical Report 76-146
(1976), Navel Surface Weapons Center, White Oak Labd, Silver Spring, MD, USA.
R. J. Spear, P. P. Elischer, Studies of Stab Initiation. Sensitization of Lead Azide by Energetic
Sensitizers, Australian Journal of Chemistry 1982, 35, 1-13.
H. Xue, Y. Gao, B. Twamley, Jeanne M. Shreeve, Energetic Azolium Azolate Salts, Inorganic
Chemistry 2005, 44, 5068-5072.
H. Xue, B. Twamley, Jeanne M. Shreeve, Energetic Quaternary Salts Containing Bi(1,2,4triazoles) Inorganic Chemistry 2005, 44, 7009-7013.
T. M. Klaptke, K. Karaghiosoff, P. Mayer, A. Penger, J. M. Welch, Synthesis and
Charaterization of 1,4-dimethyl-5-aminotetrazolium 5-nitrotetrazolate, Propellants, Explosives,
Pyrotechnics, In Press.
Acknowledgment:
Acknowledgments Financial support of this work by the University of Munich (LMU), the
Fonds der Chemischen Industrie and the European Research Office (ERO) of the U.S. Army
Research Laboratory (ARL) under contract no. N 62558-05-C-0027 is gratefully
acknowledged.
640
10
Czech Republic
2006
[Content]
5,5`-BIS-(1H-TETRAZOLYL)AMINE (H2BTA): A PROMISING

LIGAND IN NOVEL COPPER BASED PRIMING CHARGES (PC)
Thomas M. Klaptke *, Peter Mayer*, Kurt Polborn*, Jrg Stierstorfer*, and
Jan J. Weigand**
* Contribution from the Chair of Inorganic Chemistry, Ludwig-Maximilian University of Munich,
D-81377 Munich, Germany and
** Department of Chemistry, Dalhousie University, B3H 4J3 Halifax, NS Canada
Abstract:
The reaction between 5,5`-bis-(1H-tetrazolyl)amine (1) and CuX2yH2O (X = Cl, y = 2; X = ClO4, y
= 6; X = NO3, y = 2), performed in the corresponding acid and required ratio of 1:1 and 1:2
stoichiometry, afforded complexes with the formula [{CuCl2(H2bta)(H2O)}2H2O] (2),
[{CuCl(H2bta)2}Cl]
(3),
[{CuCl2(H2bta)2}2H2O]
(4),
[{Cu(H2bta)2}(ClO4)2]
(5),
[{Cu(H2bta)2}(ClO4)2H2O] (6), [{Cu(NO3)2 (H2bta)2}0.5H2O] (7). 3 crystallize in the monoclinic
space groups P-1, 3 and 4 in the monoclinic space groups P21/n and P21/c, respectively. A
description of the molecular arrangement of 3, 4 and 5 is given. The sensitivity of the investigated
compounds ranges from very insensitive (2, 4) to very (6) and extreme (5) sensitive toward impact
and friction. Except for the copper compounds 2 and 4, all investigated complexes show
photosensitive behavior and explode by laser irradiation (Laser: Nd/YAG 1064 nm).
Keywords:
1.
Primer charges (PC), aminotetrazoles, energetic copper complexes,

X-ray determination
INTRODUCTION
Neutral as well as ionic azides can undergo cycloaddition with cyano groups yielding a
huge variety of different tetrazole derivatives [1]. An interesting precursor for cycloaddition
reactions with azide is the non-linear pseudohalide dicyanamide anion (N(CN)2-), which
exhibits a rich variety of bonding modes for coordination in for example 3d-complexes.
Dicyanamide complexes have attracted much interest in recent years in the construction of
supramolecular aggregates due to their intriguing network topologies and potential functions
as new class of materials [2].
The reaction of sodium dicyanamide with sodium azide under acid catalyzed condition
yield the corresponding 5,5`-Bis-(1H-tetrazolyl)amine as monohydrate (abbreviated as
H2bta, 1) [3]. In the continuous search of novel energetic materials with high nitrogen content
for application, for example as low-smoke producing pyrotechnic compositions, gas
generators, propellants and primers in primer charges (PC), high-energy-capacity transition
metal complexes are of special interest and 1 might play an important role in future
investigations as well applications. Still modern primary explosives may be represented by
lead azide as the main filling detonator [4], and by lead styphnate as the main filling for
primers usually associated with tetrazene as a sensitizer [5]. Recently, coordination
compounds containing near stoichiometric fuel and oxidizer fractions have evinced great
interest [6]. The extensive study on the relationship between structure of coordination
compound and explosive properties has been reported by various research groups [7,8].
Nickel, copper and cobalt complexes appear suitable for detonator applications [9]. For
1
Thomas Klaptke, Peter Meyer, Kurt Polborn, Jrg Stierstorfer, Jan Weigand
641
Czech Republic
2006
[Content]
example, nickel hydrazinium nitrate (NHN) [10] may find wide ranging applications in
conventional detonators, whereas bis-(5-nitrotetrazolato-N2)-tetraamine cobalt perchlorate
(BNCP) [11] has emerged as an energy producing component for semi-conducting bridge
(SCB) initiator applications.
Low-smoke high energy materials (HNC) are thought as suitable fuel components in
new coordination compounds used in novel PCs. Derivatives of tetrazoles seem to be the
most promising substance class since they are characterized by extreme properties among
heterocycles e.g. by the highest NH-acidity, lowest basicity and the presences of several
pyridine-like nitrogen atoms important in new ligand systems as appropriate metal binding
sites. Tetrazole containing substituents with potential coordination centres are of interest
because of their added chelating properties and it is know that tetrazoles are well known
ligands for ions of many metals [12]. With respect to new PCs, 5,5`-bis-(1H-tetrazolyl)amine
(1) was investigated to see if it could serve as new high-nitrogen content (HNC) ligand
system for new copper-based PEs [8].
The synthetic routes for 1 as its monohydrate fall into three main types: (1) reaction of
sodium dicyanamide with hydrazoic acid, in situ prepared from sodium azide and a
corresponding weak acid like trimethylammonium chloride, boric acid, ammonium chloride
and the like [13]; (2) reaction of sodium dicyanamide with sodium azide in the presence of a
catalyst like zinc chloride, bromide or perchlorate, followed by an acid work-up [14]; and (3)
reaction of 5-amino-1H-tetrazole (5-AT) with cyanogen bromide under base-catalyzed
condition followed by a subsequent cycloaddition of HN3 (Fig 1.) [15].
Fig 1.
Synthesis procedures for the preparation of 1.
For reaction 1 and 2 (Scheme 1), the first reaction step yields certain metal complexes
(e.g. Zn2+, Na+) which can easily transformed to 1. Treatment of 5-AT under base condition
with cyanogen bromide results in the formation of the interesting and poorly described
dianion of 5-cyanoiminotetrazoline (CIT), as sodium salt. Further treatment of the CIT salt
with hydrazoic acid yields 1 [15].
As 1 is a bidentate chelating ligand with multiproton donor sites, it is able to coordinate
to a metal with three reversible types of protonated and deprotonated modes: neutral
(H2bta), mono-deprotonated (Hbta-), and di-deprotonated (bta2-) types (Fig 2.). Moreover,
H2bta can be utilized as a new bridging ligand for controlling the molecular architectures
642
5,5-Bis-(1H-tetrazolyl)amine (H2bta): A Promising Ligand in Novel Copper . . .
Czech Republic
2006
[Content]
which, in combination with appropriate outer- and inner-sphere ligands, allows the variation
of physicochemical and explosive properties within a wide range.
Fig 2.
1 with two reversible types of protonated and deprotonated mode.
The occurrence of energetic oxygen rich cations is comparable few but required to
assure an appropriate oxygen balance. Therefore, counterions such as nitrate (NO3-) or
perchlorate (ClO4-) anion are required in combination with the bta ligand system. In such
systems, of course, the bta needs to coordinate in its protonated from (H2bta). The reaction
of copper salts (e.g. Cl-, NO3-, ClO4-) in the corresponding acid and H2bta strongly depends
on the reaction conditions, like temperature, reaction time and concentration of the acid. By
variation of those parameters a serious of highly energetic bta-copper complexes were
obtained (Fig 3.). The complexes are crystalline, typically blue-colored compounds and
dissolves in water under decomposition. They were characterized by means of IR
spectroscopy, elemental analysis and X-ray structure determination.
Here we wish to report (i) the synthesis of a serious of energetic copper/H2bta salts (27), (ii) the X-ray determination of selected examples of the obtained compounds and (iii)
selected physical-chemical properties, like IR and sensitivity data [16].
643
2.
Czech Republic
2006
[Content]
SYNTHESIS
Caution: H2bta in its dehydrated form shows increased friction and impact sensitivity.
Although the copper(II) chloride complexes of H2bta are kinetically stable and in most cases
are insensitive to electrostatic discharge, friction, and impact, they are nonetheless
energetic materials and appropriate safety precautions should be taken. The perchlorates
and nitrates in contrast are energetic materials and tend to explode under certain condition,
appropriate safety precautions should be taken all time, especially when this compound are
prepared on a larger scale. Laboratories and personnel should be properly grounded and
safety equipment such as Kevlar gloves, leather coat, face shield and ear plugs are
necessary when manipulating the perchlorates 5 and 6 as well as the nitrate 7. Several
severe explosions occurred during the preparation of the solvent free perchlorate 5. 5, 6 and
7 tend to explode violently when heated or manipulated with a metal spatula.
The ligand N,N-bis(1(2)H-tetrazol-5-yl)-amine (1) was synthesized by the reaction of
sodium dicyanamide with sodium azide under acid-catalysed conditions, according to the
method previously described [8]. The H2bta ligand (1) was isolated as monohydrate
(H2btaH2O). 1 is completely insoluble in aliphatic hydrocarbons, acetonitrile and
dichloromethane. It is very slightly soluble in methanol and cold water but significantly
better soluble in hot water and warm dilute acids like HCl, HNO3 and HClO4. It is worth to
note that the monohydrate of 1 can easily be dehydrated under vacuum (24h, 10-3 Torr) at
elevated temperature (120C) accompanied with an increased friction and impact sensitivity.
The result of elemental analysis indicates that the reaction between 1 and CuX2yH2O (X =
Cl, y = 2; X = ClO4, y = 6; X = NO3, y = 2), performed in the corresponding acid and
required ratio of 1:1 and 1:2 stoichiometry, afforded complexes with the formula
[{CuCl2(H2bta)(H2O)}2H2O] (2), [{CuCl(H2bta)2}Cl] (3), [{CuCl2(H2bta)2}2H2O] (4),
[{Cu(H2bta)2}(ClO4)2]
(5),
[{Cu(H2bta)2}(ClO4)2H2O]
(6),
[{Cu(NO3)2
(H2bta)2}0.5H2O] (7) (Fig 3.).
Fig 3.
Synthesis of complexes 27.
The synthesis of the CuII complexes strongly depends on the volume of the solvents,
concentration of the acids, temperature, and the growing time of the crystals. Interestingly, it
is possible to obtained different complexes by changing certain reaction conditions. The 1:1
complex 2 was obtained after the addition of the copper solution to a warm (45 C) solution
644
Czech Republic
2006
[Content]
of the ligand in conc. HCl after 2 weeks. 2 is unstable during storage on air, and a few days
after the removal of the crystals from the mother liquor, 2 becomes cloudy indicating the
loss of crystal water. The solvent free complex 3 was successfully obtained when
CuCl22H2O was refluxed with the ligand in a small amount of conc. HCl for a short time. 3
crystallize already during refluxing from the solution and X-ray quality crystals can be
separated from the reaction mixture within 10 minutes. In the case of 4 we found that the
formation of 4 is favored when the reaction is performed in less concentrated solutions and
when no heat is applied while adding the copper salt to the solution of the ligand. From such
reaction mixtures 4 is obtained within a few days. 3 and 4 are air stable blue to blue-green
compounds. The reaction of Cu(ClO4)26H2O in conc. HClO4 with 2 eqivalents of the ligand
1 under harsh condition (70C) yielded in the formation of the water free complex 5.
Applying the same reaction in less concentrated perchloric acid the copper(II) perchlorate
salt 6 was obtained as monohydrate. It is important to mention that the preparation of the
salts 5 and 6 is aligned with a high risk of explosion. Never heat reaction mixtures were the
crystallization is already in progress. Especially in the case of 5, the solution with separated
crystals shows an increased shock sensitivity and tend to explode without any obvious
external impact. Special precautions have to be applied to synthesize and separate those
crystals form the reaction mixture. Crystals of 5 are very hygroscopic and unstable once
separated from the mother liquor. Crystals of 7 were obtained by the addition of a warm
copper nitrate solution to a warm H2bta solution in conc. HNO3. 7 crystallizes as beautiful
coarse plates suitable for X-ray determination within 1 day. Although we never determined
any problems with complex 7 we recommend to apply special precautions during the
synthesis and separation of those crystals.
The IR spectrum of 1 is characteristic for aminotetrazoles [17]. The changes in the bands
that are composed from combinations of the vibrations of the tetrazole ring components
indicates the formation of a coordination bond trough the nitrogen atoms of the two tetrazole
rings of 1 involved. Although those absorptions bands showing medium or weak intensities,
the shift and change in intensity can be used for the demonstration of the direct participation
of the tetrazole cycle in the complexation. The characteristic bands located at 1656 and 1556
cm-1 and assigned to the asym(CtetNCtet) and sym(CtetNCtet) stretching vibrations in 1
are shifted to lower frequencies by approximately 8 to 10 cm-1. The bands attributed to the
stretching vibrations of the N=N (1454 cm-1), C=N (1352, 1337 cm-1) and NN 1282, 1268
cm-1) bonds in the IR spectrum of the free H2bta ligand are shifted to higher wavenumbers
after complexation associated with a decrease of intensity [18]. The IR spectra of the
compounds have in common the bands for the stretching and bending modes of the NH
groups from the protons at the tetrazole ring and the sec. amino group. As it was found for
other coppertetrazole complexes the bands between 3500 and 3000 cm1 assigned to
stretching vibrations of the NH groups of 1 are shifted to higher frequencies with respect to
those of the free ligand. Furthermore it is observed that those bands in the complexes are
less well resolved compared to the free ligand 1. This can be best explained due to the
formation of intermolecular hydrogen bonds in the crystalline network. A red shift of the
stretching modes establish that the heteromolecular hydrogen-bond interaction between the
NH (donor) groups of the ligand 1 and corresponding acceptors (e.g. couteranion and
water) are strong. In all spectra the bands of the respective energetic anions (ClO4- (5, 6) and
NO3- (7)) were obvious and as these bands usually have characteristic fingerprints in IR
spectra, they could be identified easily. The presence of nonsplit, strong absorption bands at
1384 cm-1 (NO3-) [19] in 7, 1090 and 626 cm-1 (ClO4-) [20] in 5 and 6 points to a lack of strong
deviation of the trigonal planar and tetrahedral symmetry in the nitrate and perchlorate
anion, respectively [21].
5
645
3.
Czech Republic
2006
[Content]
CRYSTAL STRUCTURES
3.1
Molecular structure of 3
The asymmetric unit of chlorobis(bis(1H-tetrazol-5-yl)-amino-2N,N)copper(II) chlorid

(3) with the labelling scheme of the atoms is depicted in Fig 4. 3 crystallizes in the triclinic
space group P-1 with two formula units in the unit cell. The asymmetric unit contains one
[CuCl(H2bta)2] cation and one chloride anion. The copper atom is five coordinated by four
nitrogen atoms (N2, N6, N11 and N15) and one chloro ligand (Cl1). According to the
valence-bond theory, if a Cu (d9) ion is five-coordinated, there will be two probable
coordination geometries around the metal ion (viz., trigonal-bipyramidal and squarepyramidal). In the former, the Cu ion adopts dsp3 or sp3d hybridization and in the latter d2sp2
or sp2d2. These two configurations of a d9 ion possess approximately equal energy, and they
can interconvert.
Fig 4.
The coordination environment of the CuII ion in 3, showing the atom-labeling scheme. Displacement
ellipsoids are drawn at the 50% probability level and H atoms are shown as spheres of arbitrary radii. Selected
bond length []: CuN6 2.010(2), CuN2 1.996(2), CuN15 1.995(2), CuCl1 2.2525(7), CuN11 2.260(2);
Selected bond angle []: N2CuN15 173.21(8), N6CuCl1 158.79(7), N15CuN6 89.00(8), N6CuN2
85.63(8), N15CuCl1 88.88(6), N2CuCl1 94.62(6), N15CuN11 83.03(8), N2CuN11 101.37(8);
N6CuN11 92.95(8).
The angular structural index parameter, , of the coordination polyhedron for the CuII
ion in 3 has been calculated according to Addison et al.[22] and the distortion value for (Fig
4.; = N2CuN15, = N6CuCl1) accounts to 0.24. This indicates that the coordination
geometry around each CuII ion is a slightly distorted square pyramid and that the Cu (d9)
ions probably adopt sp2d2 hybrid orbitals to accept electrons from the ligands. The Cu(II)
being linked to three nitrogen atoms (N2, N6, N15) and one chloro ligand (Cl1) in the basal
plane, and another nitrogen atom (N11) in the apical site. The CuN bond length in the basal
plane ranges from 1.995(2) 2.010(2) , which is typical and also found in complexes with
e.g. 1,10-phenantroline as ligand system [23]. The basal CuCl1 bond distance (2.2525(7) )
is similar within experimental error, to those reported in related complexes [24]. The bond
646
Czech Republic
2006
[Content]
distance found for the apical CuN11 accounts for 2.260(2) . The mean plane in the
molecule is the six-membered H2bta chelate ring defined by atoms N2, C1, N4, C2, N6 and
Cu which is planar with negligible distortion (average r.m.s. deviation form the six-atom
plane is 0.0084 ). The plane defined by N2, N6, N15 and Cl1, deviates from planarity
(average r.m.s. deviation form the six-atom plane is 0.2463 ), since the chloro ligand lies
under this plane with a distance out of this least square plane of 0.8126 .
3.2
MOLECULAR STRUCTURE OF 4
Fig 5.
bond length []: CuN6 2.003(3), CuN2 2.016(3), CuCl1 2.822(6), Selected bond angle []: N2CuN6
84.61(1), N6CuCl1 87.52(3), N2CuCl1 96.59(3). [Symmetry code: (i) 1-x, -y, 2-z.].
Dichlorobis(bis(1H-tetrazol-5-yl)-amino-2N,N)copper(II) dihydrate (4) crystallizes in the

monoclinic space group P21/n with two formula units in the unit cell. The Cu atom in 4 lies
on an inversion center, and thus the coordinated N atoms of the bta ligands form a perfect
plane including the Cu2+ cation and are placed at distances of 2.003(3) and 2.061(2) . The
NCuN angles are close to 90, ranging from 84.61(1) and 87.52(3) but indicating a
significant large rhombic distortion. The Cu atom has an elongated octahedral coordination,
as shown in Fig 5. The nitrogen atoms of the ligand 1 coordinates in the equatorial plane,
whereas the axial position are occupied by Cl atoms. The CuCl bond distances is 2.822(6)
which is similar to the values found in analogous complexes in which Cl atoms adopt an
axial position in a distorted octahedron around the Cu(II) atom. This coordination geometry
of the elongated octahedron is typical for this kind of chromophores distorted by the JahnTeller effect and is similar to those found for other complexes of both 2-substituted and 1substituted tetrazoles [25].
3.3
MOLECULAR STRUCTURE OF 5
The molecular structure of bis(bis(1H-tetrazol-5-yl)-amino-2N,N)copper(II)

perchlorate (5) with the labelling scheme of the atoms is shown in Fig 6. 5 crystallizes in the
monoclinic space group P21/c with two formula units in the unit cell. The Cu atom displays
7
647
Czech Republic
2006
[Content]
in accordance to the above discussed compound 4 also a distorted octahedral geometry and
is positioned at (0,0,0). It lies on an inversion centre, bonded to four nitrogen atoms of the
two ligands in the equatorial plane (CuN1 1.996(3); CuN6 1.974(3) ). It is bonded more
weekly to two nitrogen atoms N3 of the tetrazole ring of adjacent molecules (CuN3ii and
CuN3iii 2.943(5) ; [Symmetry code: (ii) x, -0.5+y, 0.5-z; (iii) x, 0.5-y, -0.5+z]) in the
axial position. The weak CuN3 interactions link the molecules into layers parallel to [0,1,1]
(Fig 7.), whereas parallel stacking of these layers in the x direction is achieved via hydrogen
bonds between the NH groups and the oxygen atoms of the perchlorate anion (N5HO3i
2.854(4) , N9HO1iv 2.848(4) , N9HO3iv 2.964(4) , N4HO2vi 2.958(3) ,
N4HO1iv 3.004(4) [Symmetry code: (iv) 1-x, -y, -z; (v) x, -0.5-y, 0.5+z; (vi) x, y,
1+z; (vii) x, 0.5-y, 0.5+z]). It is worth to mention that the oxygen atom O4 of the perchlorate
anion is not involved in the hydrogen bonding.
Fig 6.
bond length []: CuN1 1.996(3), CuN6 1.974(3), CuN3i 2.943(5), Selected bond angle []: N1CuN6
85.6(3), N1CuN6i 94.4(4), N1CuN3ii 100.3(1), N6CuN3ii 79.8(2). [Symmetry code: (i) -x, -y, -z; (ii)
x, -0.5+y, 0.5-z; (iii) x, 0.5-y, -0.5+z].
648
Czech Republic
2006
[Content]
Fig 7.
[0,1,1].
A view of the weak axial contacts in 5 (dashed lines) linking the molecules to a 2D layer parallel to
4.
SAFETY TESTING
For initial safety testing, the impact and friction sensitivity was tested according to
BAM methods with the BAM Fallhammer and BAM friction tester [26]. Due to the
extreme hygroscopicity and sensitivity of compound 5 suitable values for the friction and
impact sensitivity can not be given but we experienced that compound 5 is extremely
sensitive toward friction and impact and several explosions during preparation occurred
even without obvious external impact. In the case of the monohydrate 6 (3.4 J; 14.7 N)
values comparable to those of Pb(N3)2 (2.5 J; 10 N) indicates that this salt belongs to the
group of primary explosives. Interestingly, changing the counteranion in 6 by the nitrate
anion in 7, the impact sensitivity of 7 (3.9 J) seems to be not affected whereas the friction
sensitivity considerably decreases (196 N). The chloride 3 shows a low impact (11 J) and
very low friction (353 N) sensitivity whereas 2 and 4 are insensitive toward impact (> 40 J)
and friction (> 360 N).
Raman investigations of colored transition metal complexes fail in most cases, since the
samples tend to show fluorescence. The attempt to record Raman spectra of the compounds
3, 5, 6 and 7 resulted in the immediate explosion of the sample. In the case of 2 and 4 no
reaction was observed. This observation is quite interesting, since modern concepts of the
initiation of explosives with pulsed-laser radiation is based on the ignition at centers formed
upon hypothesis of radiation absorption by optical microheterogeneities in separate crystals
[27]
.
649
5.
Czech Republic
2006
[Content]
CONCLUSION
In this contribution we presented our preliminary results on a series of copper BTA salts
(2-7). They might play an important role in the development of safe non-toxic PC
formulations, as possibly photosensitive compounds utilized in laser detonators as well as
colorants in pyrotechnic formulations. Since the starting materials are cheap and the
preparation is straight forward, 1 has a high potential as ingredient in explosive or propellant
formulations. Except for the copper compounds 2 and 4, all investigated complexes show
photosensitive behavior. They tend to explode by Laser irradiation (Laser: Nd/YAG 1064
nm), performed with different Laser power (20 to 200 mW). This observation is quite
interesting, since modern concepts of the initiation of explosives with pulsed-laser radiation
is based on the ignition at centres formed upon hypothesis of radiation absorption by optical
microheterogeneties in separate crystals. The sensitivity of the investigated compounds
ranges from very insensitive (2, 4) to very (6) and extreme (5) sensitive toward impact and
friction.
ACKNOWLEDGEMENTS
The authors are indebted to and thank Prof. Dr. P. Klfers for his generous allocation of
X-ray diffractometer time. Financial support of this work by the University of Munich
(LMU) and the Fonds der Chemischen Industrie is gratefully acknowledged. J.J.W. thanks
for a FCI Ph.D. fellowship (DO 171/46) and an Alexander von Humboldt postdoctoral
scholarship. The authors are also indebted to and like to thank Mr. Gunnar Spie for the
drop hammer and friction tests. Financial support by the WIWEB is gratefully
acknowledged.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
650
R. N. Butler, In Comprehensive Heterocyclic Chemistry; A. R. Katritzky, C. W. Rees and E.

F. V. Scriven, Eds.; Pergamon: Oxford, U.K., Vol. 4, 1996.
a) M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. OKeeffe and O. M. Yaghi,
Science 2002, 295, 469; b) H. Li, M. Eddaoudi, M. OKeeffe and O. M. Yaghi, Nature 1999,
402, 276; c) S. R. Batten, P. Jensen, B. Moubaraki, K. S. Murray and R. Robson, Chem.
Commun. 1998, 439.
W. P. Norris, and R. A. Henry, J. Am. Chem. Soc. 1963, 29, 650
J. Khler and R. Mayer, Explosivstoffe, Wiley-VCH, D-Weinheim, 9. Auflage, 1998.
J. Akhavan, The Chemistry of Explosives, RCS Paperbacks, Cambridge, 1998.
a) V. G. Leontjev and E K. Shesttakov, Proceedings of International Pyrotechnoology
Seminar, 520, 1995; b) K. C. Patil and V. R. Vernekar Pai, Synth. React. Inorg. Met. Org.
Chem. 1982, 12, 383; c) A. Yu Zhilin, M. A. Ilyshin, I. V. Tselinskii, A. S. Kozlov and N. E.
Kuzmina, Russ. J. Appl. Chem. 2002, 75(11), 1849; d) A. V. Smirnov, M. A. Ilyushin and I.
V. Tselinskii, Russ. J. Appl. Chem. 2004, 77(5), 794.
a) W. R. Tomlinson, K. G. Ottoson and L. F. Audriete, J. Am. Chem. Soc. 1949, 71, 375; b) T.
B. Joyner, Can. J. Chem. 1969, 74, 2729; c) S. V. Sinditskii and V. V. Serushkin, Def. Sci. J.
1996, 46, 371.
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
11
Czech Republic
2006
[Content]
a) M. Friedrich, J. C. Glvez-Ruiz, T. M. Klaptke, P. Mayer, B. Weber and J. J. Weigand,

Inorg. Chem, 2005, 44(22), 8044; b) J. J. Weigand, High Energy Density Materials Based on
Tetrazole and Nitramine Compounds Synthesis, Scale-up and Testing PhD thesis, LudwigMaximilian University, 2005.
a) S. Zhu, Y. Wu, W. Zhang and J. Mu, Propellants, Explosives, Pyrotechnics 1996, 22(6),
317; b) M. L. Liebarger, Ind. Eng. Chem. Prod. Res. Dev. 1985, 24, 438.
S. Zh, Y. Wu, W. Zhang, J. Mu, J. Propellants, Explosives Pyrotech. 1996, 22(6), 317.
M. Parry, P. M. Dickson and J. E. Field, 19th International Annual Conference ICT, Karlsruhe,
44/1, 1988.
a) R. Bronisz, Eur. J. Inorg. Chem. 2004, 3688; b) P. J. van Koningsbruggen, Y. Garcia, L.
Fourne`s, H. Kooijman, A. L. Spek, J. G. Haasnoot, J. Moscovici, K. Provost, A. Michalowicz, F. Renz and P. Gtlich, Inorg. Chem. 2000, 39, 1891; c) A. F. Stasse, W. L. Driesse,
J. G. Haasnoot and J. Reedijk, Inorg. Chim. Acta 2003, 350, 57; d) P. N. Gaponik, S. V.
Votekhovich, A. S. Lyakhov, V. E. Matulis, O. A. Ivashkevich, M. Quesada and J. Reedijk,
Inorg. Chim. Acta 2005, 358, 2549; e) P. N. Gaponik, M. M. Degtyarik, A. S. Lyakhov, V. E.
Matulis, O. A. Ivashkevich, M. Quesada and J. Reedijk, Inorg. Chim. Acta 2005, 358, 3949.
T. K. Highsmith, R. M. Hajik, R. B. Wardle, G. K. Lund and R. Blau, United State Patent
5,468,866, 1995.
P. Mareek, K. Dudek and F. Lika, New Trends in Research of Energetic Materials,
Proceedings of the Seminar, 7th, Pardubice, Czech Republic, Arp. 20-22, 2, 566, 2004.
T. M. Klaptke, C. Kuffer, P. Mayer, K. Polborn, A. Schulz and J. J. Weigand, Inorg. Chem.
2005, 44(16), 5949.
T. M. Klaptke, P. Mayer, K. Polborn, J. Stierstorfer and J. J. Weigand, 2006, manuscript
submitted.
M. M. Degtyarik, P. N. Gaponik, V. N. Naumenko, A. I. Lesnikovich and M. V. Nikanovich,
Spectrochim. Acta. 1987, 43A, 349.
P. N. Gaponik, V. P. Karavai and Y. V. Grigorev, Him. Geterotsikl. Soed. 1985, 1521.
a) K. Williamson, P. Li and J. P. Devlin, J. Chem. Phys. 1968, 48, 3891; b) J. R. Fernandes, S.
Ganguly and C. N. R. Rao, Spectrochim. Acta. 1979, 35A, 1013.
a) H. J. Cohn, Chem. Soc. 1952, 4282; b) P. Redlich and J. Holt, T. Biegeleisen, J. Am. Chem.
Soc. 1944, 66, 13; c) H. Grothe and H. Willner, Angew. Chem. 1996, 108, 816.
K. Nakamoto, Infrared and Raman spectra of Inorganic and Coordination Compounds, 4th
ed., Wiley, New York, 1986.
(a) A. W. Addison, T. N. Rao, J. Reedijk and J. van Rijn, J. Chem. Soc., Dalton Trans. 1984,
1349. (b) Since the angular structural index parameter, = ( - )/60, is evaluated by the two
largest angles ( < ) in the five-coordinate geometry, the value of indicates a regular
square-based pyramidal stereochemistry ( = 1.0) or a regular trigonal bipyramidal
stereochemistry ( = 0.0).
a) H. Nakai and Y. Deguchi, Bull. Chem. Soc. Jap. 1975, 48, 2557; b) H. E. Heldal and J.
Sletten, Acta Chem. Scand. 1997, 51, 122.
A. Johansson and M. Hkansson, Acta Cryst. 2004, E60, m1068.
a) R. Prins, M. Biagini-Cingi, R. A. G. De Graff, J. Haasnoot, A.-M. Manotti-Lanfredi, P.
Rabu, P. Reedijk and E. Ugozzli, Inorg. Chim. Acta 1996, 248, 35; b) R. Bronisz, Inorg. Chim.
Acta 2004, 357, 396; c) P. Brsting and J. J. Steel, Eur. J. Inorg. Chem. 2004, 376; d) P. E. M.
Wijnands, J. S. Wood, J. Reedijk and W. J. A. Maaskant, Inorg. Chem. 1996, 35, 1214.
(a) Test methods according to the UN Recommendations on the Transport of Dangerous
Goods, Manual of Tests and Criteria, fourth revised edition, United Nations Publication, New
York and Geneva, 2003, ISBN 92-1-139087-7, Sales No. E.03.VIII.2.; 13.4.2 Test 3(a)(ii)
BAM Fallhammer; (b) Reichel & Partner GmbH, http://www.reichel-partner.de/.
M. A. Ilyshin, I. V. Tselinskii and A.V. Chernai, Ross. Khim. Zh. 1997, 41(4), 81.
651
Czech Republic
2006
[Content]
EMULSION EXPLOSIVE SENSITIZED BY INORGANIC PEROXIDES

Igor L. Kovalenko, Alexandr V. Kuprin
Ukrainian state university of chemical technology, Gagarin av., 8, 49005 Dnepropetrovsk, UA
Abstract:
The opportunity of deriving of effective emulsive explosives is rotined at introduction in
emulsive compositions such as a solution of saltpeter in oil of water solutions
inorganic peroxides. Responsiveness to an originating impulse and the high indexes of
serviceability (velocity of a detonation 5200-5500 m/s, critical diameter of a unclosed
charge of 80 mm) are ensured as a result of formation in volume of an emulsion of
microblisters of oxygen.
Keywords:
emulsion, explosive, sensitization , inorganic peroxides, oxygen
1.
INTRODUCTION
It is known, that alternative to trotyl explosive traditionally applied for mining

operations, are the emulsive explosives.
In USUCT together with ECCOM Ltd (Ukraine) was designed an emulsive composition
for making emulsive explosives intended for a direct charging in wells.
The emulsive composition represents a concentrated revertive emulsion such as " a
solution of niters in oil ". Despite of high technological and operational indexes the received
composition had a series of deficiencies, namely: by low design heat of explosion (28002900 kJ/kg) and low receptivity to a detonation impulse. It has given in necessity of
sensitization of an emulsion and pinch of its energy parameters.
As the most effective methods of sensitization of emulsive explosives are considered
"aeration" and introduction of explosive responsive builders. Use of the most wide-spread
expedients of aeration of an emulsion has not given in anticipated effects. So, the
introduction of empty microspheres though has given in sensitization of an emulsion, but has
rendered negative influencing on serviceability of explosive. In turn, the applying of
chemical filling up by gas by means of traditionally applied reagents, such as nitrite of
sodium, can be bound to particular hazards. So, it is supposed, that as a result of chemical
destruction of nitrite of sodium in volume of an emulsion the blisters of nitrogen are formed,
which one execute a role of "hotspots" of a detonation. However it is necessary to mark, that
the applying of nitrite of sodium requires padding introduction of an azotic or acetic acid for
supply of process of gasification, which one for nitrite of sodium is possible only at pH < 4.
Thus thermodynamically interquartile the formation in volume of explosive not only
nitrogen is also, but also its oxides, that essentially raises hazard of reversion with such
explosives. Radiating from this in the subsequent examinations the opportunity no aeration
sensitization surveyed.
652
2.
Czech Republic
2006
[Content]
FEATURES NONAERATIONS OF SENSITIZATION

SILICONCONTAINING BY DUSTS
For pinch of serviceability and the sensitivities of an emulsion utilised dusts

siliconcontaining of materials, and, in particular, of ferrosilicon. As have shown
examinations, such materials not only have enough high combustion heat, that allows to
raise energy parameters of explosive, but also have particular influencing on inorganic
nitrates (nitrate of ammonium), essentially slashing characteristic temperature of their
destruction[1].
It was supposed, that the mechanism of activity siliconcontaining of dusts possessing
high hardness and melting point, will result in to initiation of a detonation on the mechanism
of "hotspots". Really, presence in an emulsion of solid particles it is irregular-shaped with
high melting point[2] the concentrations of energy of shock on acute angles of particles
promote, and at the expense of intensive internal friction in emulsive system can form the
multiple centers of a local heat-up, which one ensure an inconvertible detonation on the
mechanism "of thermal explosion".
By introduction in an emulsion up to 7 % of a dust of ferrosilicon were received by
explosive of the mark Ukrainit-HP-1, which one are characterized in high energy parameters
(Q=4100-4400 kJ/kg) and superb have recommended itself at mining proof and apart of
proof monolithic soils. However, at trials was marked, that the efficiency Ukrainit-HP-1 in
highfitchery grounds is essentially reduced.
With the purpose of definition of parameters of a detonation the examinations of
intensity of decomposition of an emulsive composition and Ukrainit-HP-1, ferrosilicon,
containing a dust were held. The examinations were conducted by the specialists of Institute
of chemical physics (Russia) by a method of inhibiting action of a demarcation.
In experiment the travelling speed of a demarcation an inert obex - studied matter was
continuously fixed, which one moved under activity of a shock wave of a rectangular profile
generated by an initial charge (TNT)[3]. The originating in an studied material under activity
of a shock wave of exothermic decomposition in this or that standard called inhibiting action
of a motion of a demarcation an inert obex - studied matter. On change of a profile of a wave
judged presence and character of course of exothermic processes in an studied material.
As have shown results of examinations, the originating of volumetric exothermic
decomposition Ukrainit-HP-1 is possible only at pressures in detonation front above
9,8 GPa. Besides was spotted, that in the field of pressures is lower 9,8 GPa Ukrainit-HP-1
behaves as homogeneous explosive, in volume which one I miss the centers of a local heatup.
Thus, guess that the particles of ferrosilicon in an emulsion implement to the full
mechanism of "hotspots" of a detonation has appeared erratic. Apparently, the positive
influencing of ferrosilicon on serviceability of explosive is explained by the energy
contribution of exothermal reactions of a material of the additive with yields of a detonation
behind a point Juge, and particular interaction with niters.
Use Ukrainit-HP-1 in fitchery grounds results in losses of pressure as a result of
explosion and essential decrease of serviceability of explosive.
3.
SENSITIZATION OF EMULSIVE EXPLOSIVES BY INORGANIC

PEROXIDES
For supply of a reliable detonation feeble fitchery grounds and decrease of a critical
diameter of a unclosed charge (for Ukrainit-HP-1 is peer 200 mm) the solution was accepted
653
Czech Republic
2006
[Content]
will return to traditional methods of sensitization, but, whenever possible, at a qualitatively

new level.
The basic idea of examination was looking up of an opportunity of saturation of an
emulsive composition by blisters reactive not of toxiferous gas, which one is capable to
participate in exothermal reactions of explosion.
As reagent, generating gas, the solution inorganic peroxides (hydrogen peroxide,
peroxide of sodium carbonate , peroxide of sodium borate etc) utilised 3-10 %[4].
Due to use of water solution peroxides the sensitization is carried out as a result of filling
up gas of an emulsive composition by microblisters of oxygen, which one is formed at
contact of reagent to an emulsion:
H2O2 = H2O + O2
Thus during filling up the gas forms non-toxic gases, and the sensitization of an
emulsion is reached as at the expense of complete embodying of the mechanism of
"hotspots" of a detonation and abatement of density of emulsive explosive, and at the
expense of presence at volume of microblisters of oxygen possessing particular reactivity.
Idiosyncrasy of the proposed expedient of aeration is that the gas-making takes place on
boundary surface a dispersed phase (solution of niters) / a dispersion medium (solution of
emulsifying agent in petroleum). Thus presence of a non-polar organic phase, which one
contains nonionic SAS - emulsifying agent, ensures uniform and intensive allocation in
volume of explosive of microblisters of formed oxygen.
Fig.1 Sensitized emulsion (medial size of a blister of oxygen 150-300 microns)

Chemical properties of oxygen and the character of allocation in volume of an emulsion
ensures practically 100 % a completeness of chemical reactions of explosive transmutation
of explosive, ensuring high speed of a detonation (5200-5500 m/s).
It is known, that the intensity of destruction of peroxides depends on temperature and
pH. Radiating from this, was are held examinations permitting to place(install) optimal
parameters of filling up by gas of an emulsion at introduction of water solutions of
peroxides.
In a fig. 2 dynamics of filling up by gas is given at different pH of an emulsion (have
utillized 10 % a solution of hydrogen peroxide , quantity of an inlet solution 1 %, reference
temperature of an emulsion 500). As it is visible from a fig. 2, the optimal gas-making takes
place in area pH > 7, that eliminates necessity of introduction in an emulsion of acids,
raising physical stability and safety of explosive.
654
Czech Republic
2006
[Content]
Fig.2 Dynamics of filling up by gas at different pH of an emulsion

As a result of the held operations the emulsive explosives of the mark Ukrainit-HP-2
were received.
Table 1. Basic characteristics of explosive Ukrainit-HP-2
Index
Oxygen balance, %
Energy of explosion, kJ/kg
Density, kg /dm3
Volumetric energy of explosion, J/dm3
Light-end products of explosive, m3/kg
Critical diameter of a unclosed charge, mm
Velocity of a detonation, m\s
Value
-0,05-0,18
2900-2960
1,25-1,30
3700-3850
0,762-0,892
80
5200-5500
As the received explosive is visible from the table has small values of design energy of
explosion. However, as have shown industrial trials, Ukrainit-HP-2 is characterized by high
indexes of serviceability at use both in fitchery grounds, and in soils of a rocky type.
As a result of a serial of industrial trials 5955 tons Ukrainit-HP-2, as a result of which
1189 thousand m3 mining and 3896 thousand m3 of rocky soils were blasted at
100 % of dampness of a mountain massif. Thus the oversize yield compounded no more than
0,01 %, and the specific consumption in rocky soils depending on a stability compounded
0,98-1,29 kg/m3.
Measuring gaseous composition of explosion products also were carried out at explosion
in a dummy underground working. Set, that the quantity of toxiferous gases in recalculation
on CO compounds 24,3 dm3/kg, that is much lower, than at explosion of regular explosives.
The high indexes of serviceability of explosive, at small values of design parameters
(energy of explosion of 2800-2900 kJ/kg) can be explained as follows. At first, during
sensitization in volume of an emulsion the microblisters of oxygen are formed, which one
accept fissile involvement in secondary oxidizing reactions in a gas phase, ensuring
practically 100 % a completeness of chemical reactions of explosive decomposition of
explosive. It results in practically complete embodying potentially of possible energy of
explosion of explosive except for thermal losses. Secondly, oxygen and the peroxides have
particular activity concerning a hydrocarbonaceous phase of an emulsion.
As is known, the acidification of the majority of organic compounds, and including oil
of hydrocarbons takes place on the radical mechanism[5]. The introduction in an emulsion of
solutions of inorganic peroxides results in initiation of radical acidification of a
hydrocarbonaceous phase of an emulsion, raising its reactivity and sensitivity to an
originating impulse.
655
a)
Czech Republic
2006
[Content]
b)
Fig 3. The DTA-curves of destructions: no sensitized emulsion (a, b-1);

Emulsion sensitized : by microspheres (a-2), by hydrogen peroxide (b-3)
As have shown results of thermal examinations of an emulsive composition (heat rate
20 0/min), the introduction in its composition of inorganic peroxides not only results in
magnification of intensity of exothermal reactions of decomposition of niters, but also
slashes characteristic temperature of destruction of a hydrocarbonaceous phase of an
emulsion on 50-70 0 (fig 3., b-3).
Thus, the results of laboratory and industrial trial demonstrate high performance of
applying of water solutions of inorganic peroxides for sensitization of emulsive explosives.
For embodying a designed expedient of sensitization of emulsive explosives intended for
a mechanical charging immediately in wells, the construction of a charger designed, which
one realizes delivering a sensitizer immediately in a warhead hose pipe. Such device ensures
high safety of a charging, as the filling up is carried out gas of an emulsion already after a
charging of explosive, i.e. immediately in well.
We express gratitude to the professor Shvedov K.K. for the help in examination of the
detonation performances of emulsive explosives Ukrainit.
REFERENCES:
[1]
[2]
[3]
[4]
[5]
656
I.L. Kovalenko, A.V. Kuprin//Probl. of chem. and chem. techn., 4, P. 72-76, 2004.
F.A. Baum, K.P. Stanykovich, B.I. Shechter: Physics of detonating, Moscow,1959
K.K. Shvedov, V.V. Lavrov// Chem. Physics., V.17. 3, P. 74, 1998.
Pat. UA 74500: Emulsion explosive Ukrainit-HP-2; 20040705318; 15.12.2005
A.P. Glazkova: Catalysis of combustion of explosives, Moscow, 1976
Czech Republic
2006
[Content]
DETONABILITY OF NITROMETHANE MIXTURES WITH

AZIDEETHANOL
George D. Kozak1, Vlada M. Raikova
Mendeleev University of Chemical Technology,
Miusskaya sq. 9, Moscow A-47, Russia
Abstract
Detonation propagation in liquid mixtures of Nitromethane (NM) with Azideethanol
(AE) was investigated. It was revealed that high velocity detonation (HVD) of the
mixtures of NM contained less 70% AE propagated steady in thick-walled steel tubes.
Addition of 20% AE to NM decreased df from 13 to 9 mm (in glass tubes). Critical
pressure of initiation of detonation in this mixture was measured. Detonability of jel-like
AE was investigated too. Effective kinetic parameters of reactions in detonation wave
for NM-AE mixtures were estimated.
Keywords: detonation, Nitromethane, Azideethanol.
1.
INTRODUCTION
Nitromethane (NM) is sufficient high capacity explosive (explosion heat Qv=5300

kJ/kg), though its energy is less, than energy of nitroglycerin (Qv=6200 kJ/kg) and another
nitroesters. High thermal stability and low shock and impact sensitiveness are important
advantages of NM in comparison with nitroesters. Detonation of NM has been investigated
thoroughly. This substance is classical model in Failure detonation diameter theory of liquid
explosives [1-4]. A comprehensive investigation of the detonation of NM and other liquid
substances was conducted by B.N. Kondrikov and his collaborators at the Mendeleev
University of Chemical Technology (Moscow) [5-10]. High velocity detonation (HVD)
propagation in NM was studied in detail. The main achievement of this work was the
demonstration of the strong influence of organic bases, mineral acids, and unsaturated
compounds on the critical diameter of liquid nitrocompounds [5-8]. Initiation and propagation
of low velocity detonation (LVD) in NM firstly were examined in works [9,10].
Chemical kinetic of thermal decomposition and combustion of organic azides including
Azideethanol N3C2H3OH were studied quite enough [11,12]. However it is not known about
detonation of this substance.
The main goal of this work was study of initiation and propagation of detonation in NM
mixtures with AE. Experimental data on HVD and LVD were obtained. Detonability of gellike systems on the base of AE was investigated too. Shock and detonation parameters as
well as characteristics of the reaction zone to determine reinitiation of detonation wave at
low diameters of the charge were calculated by means of SGKR computer package [13].
2.
1
EXPERIMENTAL
KozakGD@nm.ru
George Kozak, Vlada Raikova
657
Czech Republic
2006
[Content]
Commercial Nitromethane was purified by distillation at 101 0C. Neat Azideethanol was
extracted from aqueous solution containing 30% water by means of dichloromethane.
Solutions of AE in NM were prepared by mixing of components during 10-15 minutes.
Before every experiment the density of solution was measured by means of bottle method.
The dependence of solution density vs. content of AE in mixture is shown in Fig.1.
Porous material in the form of phenol-formaldehyde microspheres (PFS) was used for
decreasing of liquid AE density. Average size of microspheres was about 0.12 mm. 5% of
colloxylin was added in liquid AE for increasing of viscosity and stabilization of emulsion.
Fig 1.
Experimental dependence of density () vs. AE content (CAE) in NM-AE

mixtures. Points are experiment data. Curve is calculated dependence on
the assumption additivity law of specific volumes.
Experiments investigating detonability of NM-AE mixtures and gel-like AE were carried

out in steel tubes of 10 mm inner diameter, 13mm wall thickness, length l=250 mm, and 9
radial holes d=2 mm were drilled in wall. The bottom of tube was covered by mean of a
resin plug. The experiments were carried out either without photoregisration or with it. In the
first case the radial holes were glued up by a scotch, and a metal witness-plate was fastened
to them. The wood plate was arranged between witness-plate and massive steel plate, on
which the construction was fastened, to prevent intense deformation of a witness-plate at
detonation. The character of deformation of a tube and perforation of a witness-plate gave an
information about stability of a process. The high velocity detonation destroyed the steel
tube in fragments. The low velocity detonation didnt crash the tubes, but perforated the
witness-plate uniformly, if the process propagated without damping. At damping of low
velocity detonation process a degree of perforation diminished to the end of a tube.
The pressed pellets of phlegmatized RDX (d=12 mm, m=2 g, density =1.67-1.68
g/cm3) were used as powerful initiator. Mixtures of RDX/NaCl of different composition [14]
with PMMA attenuator between the bottom of the initiating charge and the surface of a
liquid were used as donor charges for determination of critical pressure of initiation of
substance under study.
Russian streak camera GFR-3 was employed to measure detonation velocity. The
luminosity at detonation was too weak for registration by means of photography, and to
658
Czech Republic
2006
[Content]
develop the process propagation on streak camera record we used modifying technique,
suggested in [14,15]. The radial holes were cowered by paraffin plugs near channel of tube; the
length of plugs was the same in all holes (8 mm). Small amount (~0.02 g) of lead azide was
promptly placed into each hole on paraffin plug, and then the holes were glued up with a
scotch. Luminosity of lead azide flashes fixing in streak camera record helped to judge about
a stability of process, and to measure its velocity. As it had been shown in works [14,15], the
line connecting points-flashes of lead azide in radial holes on a streak camera record was a
straight one, if low velocity process was not damp. If a propagating process damped the
slope of line diminished to the end of a charge.
3.
EXPERIMENTAL RESULTS
Experiment results of NM mixtures with AE detonation in steel tubes at powerful

initiation are shown in Fig. 2. Steady HVD propagated in the mixtures containing 20-60%
AE. Steel tubes were destroyed in fragments in these tests. HVD was observed in the
mixture with 70% AE only in two tests (D=5.36, 5.40 km/s), LVD was registered in three
tests, and transition HVD to LVD was observed in one test. Detonation did not propagate in
solution containing 80% AE, and neat AE.
Fig 2.
Experiment results of NM mixtures with AE detonation in steel tubes at

powerful initiation.
The detonation failure diameter (df) of NM mixture with 20% AE was measured in glass
tubes. Steady HVD propagated in tube by diameter 10.5 mm and did not propagate at d= 8,49 mm. Extinction of detonation was observed in tube by d=9.5 mm. The value df=9.3 was
determined as critical diameter of this mixture.
Critical pressure of initiation was defined for mixture of NM with 20% AE (Tabl.1).
HVD was initiated by mixture RDX/NaCl 40/60 ( =0.95 g/cm3). Corresponding value of
critical pressure of HVD was Pin = 5.6 GPa. LVD was initiated by mixture RDX/NaCl 50/50
( =0.86 g/cm3), that corresponded to Pin = 4.5 GPa. Detonation did not propagate at Pin =
1.6 Gpa.
659
Czech Republic
2006
[Content]
Table 1. Critical pressure of initiation (Pin) for mixture of NM with 20% AE

Characteristics of RDX/ NaCl initiator
Content of
Pin, GPa
, g/cm3
NaCl, %
Character of deformation
Tube
witness-plate
85
0.55
1.6
Intact
Intact
50
0.86
4.5
Perforation
60
0.95
5.6
Intact
Destruction
in fragments
along tube
Destruction
in twain
Result
No
explosion
LVD
HVD
It was interesting to determine detonation of neat AE. To this end the method of
sensitization of detonation for low detonatability explosives by introduction of microspheres
was used [16, 17]. Experimental density dependence on PFS content in gel-like AE could be
described by equation:
=1.202-0.059 PFS,
where PFS is content of PFS in gel-like AE in mass percent.

Experiment results on detonation of gel-like AE in steel tubes at powerful initiation are
presented in Fig.3. The density of gel-like AE non-contained PFS was greater than density of
liquid AE. Only LVD in this gel emulsion was observed. Introduction 1-2% of PFS into gellike AE did not affect on detonation regime. HVD was registered in gel-like AE containing
3-4% PFS.
Fig 3.
660
Influence of density () on detonability of gel-like AE contained PFS

(black points) and non-contained PFS (open points) in steel tubes at
powerful initiation. Numbers correspond to content of PFS in gel-like AE.
4.
Czech Republic
2006
[Content]
CALCULATION RESULTS
Detonation and shock parameters as well as characteristics of the reaction zone defining
the process of reinitiation of detonation wave in the zone of instability were calculated by
means of SGKR computer code developed in our Laboratory previously [13]. Computation
procedure was described in detail in our previous papers [13,18].
The input data required for the calculation were the enthalpy of formation for the
substances, experimental dependencies of density, sound speed, specific heat from initial
temperature, and shock adiabat of the substances. Physical-chemical and shock
characteristics of NM were studied well [19]. Experimental density of AE AE=1.145 g/cm3
closes to NM value NM=1.13 g/cm3. The dependence of density on AE content in mixtures
is described well by additivity law of specific volumes (Fig.1}. The same temperature
coefficient of density 0/dT0=0.0014 g/(cm3K) for NM [19], AE and mixtures was assumed.
Sound speed, specific heat of AE were not known. These parameters were calculated having
used experimental values for liquid and gaseous HN3 [20,21] and liquid ethanol. The
generalized Hugoniot of liquids U = U0+0.29+1.63u [13] was used for calculation. The same
velocity of the reaction failure wave v= 3.8 km/s for all mixtures, neat AE as well as for neat
NM [4] was accepted.
Detonation parameters were calculated for NM, AE and their mixtures. Calculated heat
of explosion (Qv,) decreased with increasing content of AE in the mixture, as expressed by
the linear equation:
Qv= 5290 8.47CAE, kJ/kg,
where CAE is content of AE in mixture in mass percent.

Calculated detonation velocity slightly rose with increasing of AE content in solution
from DNM=6.3 to DAE=6.4 km/s. Detonation presser pcj weakly increased with rise of
solution density from 12.84 to 13.05 GPa. Isentropic exponent was not dependence from
composition of solution and equal =2,5350,002.
Calculated shock adiabats T(p) for NM, AE and their mixtures are shown in Fig.4. At
the same pressure shock temperature for NM is higher than T for AE, so at p=10 GPa
temperature difference is about 270 K.
Fig 4.
Pressure dependencies of shock temperature (thin lines) and temperature

T3 (thick curve) for NM, AE and their mixtures. Numbers correspond to
content of AE in the mixture.
661
Czech Republic
2006
[Content]
From the beginning characteristics of reaction zone (pressure p3, temperature T3 and
reaction time 3) near the limit of detonation propagation were calculated for neat NM and
the mixture containing 20% AE by experimental df values (Tabl.2). Parameters pcj, D, p3 for
NM and the mixture differed slightly from each other. The T3 value for NM is 50 K above
value for the mixture. Reaction time 3 at addition of 20% AE to NM reduced according to
decreasing of df.
Table 2.
The failure detonation diameter (df), detonation parameters (D, pcj) and
characteristics of reaction zone (pressure p3, temperature T3 and reaction
time 3) near the limit of detonation propagation for NM, the mixture of
NM with 20%, and AE.
CAE, %
df, mm
Exper. Calcal.
D,
m/s
pcj,
GPa
p3 ,
GPa
3,
3 ,
mcs
13
13
6334
12.8
9.2
1061
0.24
20
9.3
6334
12.9
9.3
1012
0.17
100
133
6.421
13.4
9.6
809
2.45
Earlier we established linear correlations between the pressure P3 and the detonation
pressure Pcj, and between df and 3 for liquid mixtures [18,22]. These correlations allow to
evaluate parameters p3, T3, 3 and df over a wide range of compositions, if only two df
values (for neat substances or mixtures) were determined. Since detonation pressure Pcj did
not dependence from AE content in mixture, pressure P3 was invariable too. The T3 value
was defined by shock temperature dependency T(p) at correspondent value of p3 for each
mixture (Fig.4). The addition of AE to NM decreased T3 in mixture. The calculated
dependence of T3 on the concentration of AE in solutions is expressed by the linear equation:
T3 = 1060 2.502 CAE
To predict values of df for liquid mixtures in addition to dependency T3(CAE) the kinetic
parameters of detonation reaction are necessary.
5.
DISCUSSION
Experimental results on detonability of NM mixtures with AE were unexpected. Though

density, calculated detonation velocity and pressure of AE were closed to NM parameters
and yet exceeded ones, nevertheless liquid AE, and NM mixtures with a great content of AE
did not detonated in steel tubes. On the other hand the addition of 20% AE to NM reduced
critical diameter in glass tubes from 13 mm to 9.3 mm.
662
Czech Republic
2006
[Content]
For explanation of experimental data chemical kinetics of decomposition of NM, AE

and their mixtures in detonation wave was examined. Founders of the Failure diameter
theory supposed [1,2] that df of liquid substance was connected with induction period of
thermal explosion. Therefore kinetic parameters of monomolekular reaction obtained in
usual kinetic experiments were used for df calculations. Subsequent investigations shown
that mechanism of decomposition in detonation wave including break of bond C-N
(activation energy Ea=230 kJ/mol and pre-exponent factor B=1014 s-1[23]) did not agree with
experimental dependencies of initiation delay in shock wave [24,25]. Formation and
consequent decomposition of aci-form of NM (CH2NOOH) were assumed to be the leading
chemical processes in detonation wave. According to estimation of various researchers [25,26]
activation energy of decomposition of aci-form NM in shock or detonation wave was Ea=30
35 kkal/mol (125-146 kJ/mol). The same Ea value was defined for decomposition of NM at
static compression pressure 50 kbar [23].
Calculation of df values for NM-AE mixtures was performed by assumption that NM
and AE decomposed in detonation wave independently from each other. In this case
effective values of rate constant of reaction in the mixture were defined by additive function
[18]
:
kef = k11 + k22,
(1)
where k1 k2 are the rate constants of decomposition reaction in neat NM and AE corr.,
1 2 are their molar fractions in solution.
The values of activation energy and pre-exponent factor of thermal decomposition of

NM calculated by experimental dependence df on initial temperature [27,28] were used for
calculation of k1:
k1 = 2 1014 exp(-128000/RT), (2)
The values of kinetic parameters of thermal decomposition of NM obtained in usual

kinetic experiments [11] were used for calculation of k2:
k2 = 1014 exp(-158000/RT),
(3)
The dependencies of kinetic constant k1, k2, and kef vs. 1/T3 are shown in Fig.5. Constant
kef decreased from NM to AE. According to calculation the maximum value of kef occurred
at the concentration 20% of AE in mixture.
663
Fig 5.
Czech Republic
2006
[Content]
The calculated kinetic constant k vs. 1/T3 for decomposition of NM (k1),

AE (k2) and their mixtures (kef) in detonation wave.
The kinetic constant kef was used for calculation of df(CAE) dependence for NM-AE
mixtures (Fig.6). Calculated df values for NM, and the mixture of NM with 20% AE are 13
and 9 mm correspondingly. The results of calculations of df are in good agreement with
experimental data. Estimation of critical diameter of AE with using eq.3 shows that df is
about 130 mm. This value is greater than equivalent critical diameter of steel thick-walled
tube deq=805 mm [18]. This significant fact explains non-detonation of AE in steel thickwalled tube.
Conversion of liquid AE into porous gel by means of introduction 5% of colloxylin and
1-4% PFS facilitated propagation both HVD and LVD. LVD was initiated at gel density
=1.1-1.2 g/cm3 and HVD was initiated at =1g/cm3.
The addition of 20% AE to NM decreased df and as well as critical pressure of HVD
initiation from 10 to 5.6 GPa. Critical pressure of LVD initiation for the mixture conversely
was higher than for NM. So introduction of AE in NM facilitates HVD initiation but
impedes propagation of LVD in liquid.
664
Fig 6.
6.
Czech Republic
2006
[Content]
The calculated failure diameter (df) vs. content AE (CAE) in NM mixture

with AE. Black points correspond to experimental data for NM and the
mixture of NM with 20% AE, open point is estimated value df=133 mm for
AE.
CONCLUSION
Limited conditions of HVD and LVD in mixtures of NM with AE and jel-like AE were
investigated. It was revealed that NM mixtures containing 70% and less are detonable.
Addition of 20% AE to NM decreased critical diameter and initiation pressure of HVD.
Detonation reactions in the mixtures were considered in terms of A. Dremins theory and
SGKR computer code calculations. The df values of NM, AE and their mixtures were
estimated by chemical kinetic and correlative equations.
ACKNOWLEDGEMENTS
investigation through grant 04-03-32713. Our post graduated students V.V. Potapov and
E.A. Likholatov took part in carrying out the experiments and calculation
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
A.N DREMIN: The failure detonation diameter of liquid explosives, Proceedings of Akad. Sci.
USSR (Doklady), v.147, p. 870-873, 1962 (Russ.)
A.N. DREMIN, V.S. TROFIMOV: Calculation of Failure Detonation Diameter of Liquid
Explosives, Problems of mechanics and theoretical physics, 1, p. 126-131, 1964 (Russ.)
J.W.ENIG AND F.W. PETRONE: The failure diameter theory of Dremin, 5th Symposium (Int.) on
Detonation, Pasadena, Ca. Aug. 18-21, p. 99-104, 1970.
C.N. TARVER, R. SHAW AND M. COWPERTHWAITE: Detonation failure diameter studies of four
liquid nitrialkanes, J.Chem.Phys. v.64, 6, p. 2666-2673, 1976.
B.N. KONDRIKOV, G.D.KOZAK, V.M. RAIKOVA, A.V. STARSHINOV: Detonation of
Nitromethane, Proceedings of Akad. Sci.USSR (Doklady), v.233, 3, p. 402-405, 1977
(Russ.)
B.N. KONDRIKOV, G.D. KOZAK, , V.M. RAIKOVA, , A.V. STARSHINOV: Homogeneous
catalysis at detonation of nitromethane, Detonation, Akad. Sci. USSR, Chernogolovka, p. 7376, 1977 (Russ.)
665
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
666
Czech Republic
2006
[Content]
V.M. RAIKOVA, B.N. KONDRIKOV, G.D KOZAK: Experimental investigation of

detonation of nitric acid solutions, Comb., Expl., and Shock Waves, v.34, 1, p.77-84, 1998
G.D. KOZAK: Factors augmenting the detonability of energetic materials, Propellants,
Explos., Pyrotech., v. 30, 4, p. 291-297, 2005
G.D.KOZAK, V.V.KONDRATIEV, B.N.KONDRIKOV, A.V.STARSHINOV: Low velocity
detonation of nitromethane, Detonation. Institute of Chemical Physics of Akad. Sci. USSR,
Chernogolovka, 34-37,1978 (Russ.)
G. D. KOZAK, B. N. KONDRIKOV, AND A. V. STARSHINOV: Critical Conditions of Low
and High Velocity Detonation Regimes in Liquid Nitrocompounds, 11th Symposium (Int.) on
Detonation, Snowmass Village, Colorado, August 31 September 4, p.81,1998
S.P. KANASHIN, M.S. KOZHUKH, A.N. KULAGIN, N.P.TOKAREV: Limited conditions of
burning of -azideetnanole. Comb., Expl., and Shock Waves, v.11, 3, p.22-26, 1981
B.M. ISAEV, S.P. KANASHIN, M.S. M.S. KOZHUKH, N.P.TOKAREV: Investigation of
burning process of some organic azides, Chemical physics of combustion and explosion
processions. Chemical kinetics of reactions. Akad. Sci. USSR, Chernogolovka, p. 97-101,
1980 (Russ.)
A.I. SUMIN, V.N. GAMEZO, B.N. KONDRIKOV, V.M. RAIKOVA: Shock and detonation general
kinetics and thermodynamics in reactive systems computer package, Trans. of the 11-th (Int.)
Detonation Symp., Snowmass, Colorado, USA. August 31-September 4, 1998. Bookcomp,
Ampersand, p. 30-35, 2000
V.F.TYSHEVICH: Investigation of low velocity detonation initiation and propagation of cast
and pressed explosive charges in strength confinement, Cand. Sci. (Ph. D.) Thesis (Rus.).
Mendeleev University of Chem. Technol. Moscow, 1992 (Russ.)
I.V.BABAITSEV,. B.N.KONDRIKOV,. Z.V PAUKOVA, V.F TYSHEVICH: Detonation of
cast explosives with a low speed, Comb., Expl. & Shock Waves, v.5, 3, p.326-330, 1969
G. D. KOZAK AND B. N. KONDRIKOV: The peculiarities of detonation in nitrobenzene and
in propargyl alcohol, Comb., Expl., and Shock Waves v.35, 1, , p.89-97, 1999
G.D. KOZAK, N.I. AKININ, V.M. RAIKOVA, S.V. ARININA: Explosion hazard of
someorganic peroxides. Proc. of The Int 6th Seminar New trends in research of energetic
materials, Univ. Pardubice, CSR, p.173-182, 2003
V.M. RAIKOVA, G.D. KOZAK, E.A. LIKHOLATOV: The Chemical Kinetics at Detonation
of Nitroester Solutions, The 7th Int. Seminar New trends in research of energetic materials,
Univ. Pardubice, CSR April 23-25, , p. 284-293, 2003.
P.C. LYSNE, D.R. HARDESTY: Fundamental equation of state of liquid nitromethane to 100
kbar, J. Chem. Phys., v.59, 12, pp. 6512-6522, 1973
G.S. YAKOVLEVA: Detonation properties of liquid hydronitric acid and some hydronitric
derivatives, Cand. Sci. (Ph. D.) Thesis, Institute of Chemical Physics of Akad. Sci. USSR,
Chernogolovka, 1981 (Russ.)
D.A DOWS, G.C..PIMENTEL: Infrared spectra of gaseous and solid hydrazoic asid and
deutero-hydrazoic acid: the thermodynamic properties of HN3, J. Chem. Phis., vol. 23, p
1258-1263, 1955
V.M. RAIKOVA, E.A. LIKHOLATOV: Chemical kinetics of detonation in some liquid
mixtures. Propellants, Explos., Pyrotech., v. 30, 4, p. 291-297, 2005
G.B. MANELIS, G.M.NAZIN, YU.I. RUBTSOV, V.A. STRUNIN: Thermal decomposition
and combustion of explosives and powders, Moscow, Nauka, 1996
J.G. BERKE, R. SHAW, D. TEGG, L.B. SEELY: Shock initiation of nitromethane, methyl
nitrite and some bis difluoramino alkanes. 5th Symp.(Intl.) on Detonation, ONR Rept. DR-163,
pp. 237-245, 1970
CHAIKEN R.F: Kinetic theory of detonation of high explosives, 8th Symp. (Intl.) on
Combustion, Williams and Wilkins Co., Baltimore, Md., p.759, 1962
10
Czech Republic
2006
[Content]
[26] I.M. VOSKOBOYNIKOV, N.F. VOSKOBOYNIKOVA: Measuring of reaction time of

condensed explosives in detonation wave, Chemical Physics (Russia), v.7, 3, p. 406-411,
1988
[27] A.W.CAMPBELL, M.E.MALLIN, T.E.HOLLAND: Temperature dependence of nitromethane
failure diameter, J. Appl. Phys. v.27, 8, p.963, 1956
[28] V.N.GAMESO: The heat release kinetics in the reaction zone of detonation wave, Cand. Sci.
(Ph. D.) Thesis, Mendeleev University of Chemical Technology, Moscow, 1992 (Russ.)
11
667
Czech Republic
2006
[Content]
EFFECT OF RUBBING SURFACES VELOCITY ON THE RESULTS

OF FRICTION SENSITIVENESS TESTING OF EXPLOSIVES
Miloslav Krupka, Jan Horkel, Jakub eleovsk

Institute of Energetic Materials, University of Pardubice, CZ-532 10 Pardubice
Abstract:
This paper reports the relationship between friction sensitiveness data of explosive
substances obtained by the BAM friction apparatus using several different velocities of
rubbing surfaces. The CO and NO analyzer was used for automatic determination of
load adequate to the first reaction (initiation) of tested substances. The little rubber
chamber for keeping decomposition gases was developed during this research work.
Keywords: Friction sensitivity, rubbing surfaces velocity, explosives, BAM friction
apparatus
1.
INTRODUCTION
Friction sensitiveness is one of the fundamental safety characteristics of explosive

substances. It can be determined using various apparatus. BAM friction apparatus is one of
the most popular in Europe. The results obtained by this apparatus are heavily dependent on
several different factors, which can be divided to the three basic categories: sample
parameters, instrument parameters and visual and auditory ability of operator. In case of
passing over the effect of particle size, the influence of sample parameters can be eliminated
through keeping the sample in given constant temperature and humidity conditions prior and
during testing. Objectivity and reproducibility of test results can be improved by using of
decomposition gases detector1,2. Whereas test conditions are carefully monitored2,3,4 the
instrument parameters are often neglected. There are several standards5,6,7,8 regarding to
BAM friction test, but only three of them6,7,8 determine velocity of porcelain plate against
porcelain pin. The requested velocities are 4.548, 4.886 and 7.147 cm.s-1. Individual
apparatuses also differ from each other due to different manufacturers and different rate of
wear. In order to these facts the rubbing surfaces velocity can significantly differ in different
laboratories. The goal of this paper is studying of rubbing surfaces velocity as one of the
instrumental parameters influencing the results of friction sensitiveness determination of 4
samples RDX and PETN with different particle size.
2.
EXPERIMENTAL
The following explosive samples were studied:
The samples of PETN type NpS (Fig.1) and NpD (Fig.2) manufactured by Synthesia and
RDX type Hx E (Fig.3) and Hx C (Fig.4) manufactured by Dyno Nobel. The samples of
each explosive differ in particle size. The quantity of tested sample was 10 mm3.
668
Effect of Rubbing Surfaces Velocity on the Results of Friction Sensitivity . . .
Czech Republic
2006
Fig.1: PETN type NpS (resolution 20m)
Fig.2: PETN type NpD (resolution 100m)
Fig.3: RDX type Hx E (resolution 20m)
Fig.4: RDX type Hx C (resolution 100m)
[Content]
Measure was carried out using new developed apparatus FSKM-50/20K (Fig. 5)
manufactured by OZM Research. This instrument fully complies with UN Tests and Criteria
Manual - BAM Friction sensitivity test5. The using of intelligent compact drive with fieldbus
stepper motor is the main advantage compared to instruments driven by electromotor with
the gear box traditionally used. This solution allows to control not the speed with high
precision only, but also precise positioning of the motor to the starting positions. The
overshoots or operating cycle incompletion often occur using some older apparatuses.
Fig. 5 Friction sensitivity tester FSKM-50/20K
The velocity, acceleration and deceleration can be changed easily with operating
software. The maximal average velocity of rubbing plates of this apparatus is closed to 10
cm.s-1 because of mechanical reasons. The minimal velocity is not limited.
Miloslav Krupka, Jan Horkel, Jakub Selesovsky
669
Czech Republic
2006
[Content]
The average velocity corresponding to total time of working cycle was changed in three
levels during testing. The real average velocity of each level was verified by independent
method.
Rubbing surfaces
velocity
Time of 1 cycle
(s)
(cm.s-1)
Number of
revolutions per
minute (r.p.m)
3.33
0.600
100
4.66
0.429
140
10.00
0.200
300
Initiation of majority tested substances is not accompanied by crack or flash. It is quite

difficult to evaluate test results for inexperienced operator. Operator has to observe smoking,
odor, scorch marks, sparking or popping noises in case of lack reactions. We have decided to
use electrochemical analyzer of gaseous product (CO and NO) as a gauge of sample
initiation for better objectivity of results and for decreasing of operator exposure to toxic
fumes. The range of analyzer sensors was 0-500 ppm (CO) and 0-100 ppm (NO). It was
necessary to keep decomposition gases for analyzing. Although there is lot of moving parts
in the apparatus, the little chamber around of porcelain plate was developed. The rubber
chamber is not fully hermetic on upper side, but this design is sufficient for keeping gases
prior to analysis. The chamber is also easy removable because of cleaning and more
comfortable handling with samples. The analyzer is more sensitive compared to human
operator. Therefore it is quite easy to evaluate value of first reactions load. First reaction
load was defined as the smallest load when indicia of initiation appear. The concentration of
decomposition gases measured by analyzer during first reaction was determined 0-1 ppm for
CO and 0-5 ppm for NO. Measurement was carried out 5 times for every load level up to no
reactions in 5 tests occurs.
3.
The change of the rubbing surface velocity influences significantly the determined
values of friction sensitiveness. The difference of 1 cm.s-1 causes the difference of 6-20 % in
determined sensitiveness. The decrease of sensitiveness with the increase of friction velocity
was observed in case of all samples. The biggest decrease was observed in case of samples
containing big crystals, as can be seen in table 1 and in figure 6. The values of friction
sensitiveness obtained at maximal friction velocity are 25-50 % smaller against the values
obtained at minimal velocity.
Velocity cm/s
3,33
4,66
Total dif. [N] Total dif. [%]

10
PETN NpS Load [N]
62,8
59,0
47,3
15,5
24,7
PETN NpD Load [N]
59,0
53,2
41,4
17,6
29,8
RDX Hx E
Load [N]
78,5
70,6
53,2
25,3
32,2
RDX Hx C
Load [N]
78,5
62,8
39,3
39,2
49,9
Table 1. Sensitiveness in dependence to velocity
670
Czech Republic
2006
[Content]
90,0
PETN NpS
PETN NpD
RDX Hx E
RDX Hx C
80,0
Load [N]
70,0
60,0
50,0
40,0
30,0
0
10
12
Velocity [cm/s]
Fig. 6: Relationship between rubbing surfaces velocity and sensitiveness by First reaction
Design of BAM Friction sensitiveness tester has one big disadvantage. The sample is
triturated only by the top of spherical part of the pin. The friction area is therefore very
small. The less sensitive samples containing big particles give very scattered results. The pin
only with problems triturates the big crystals; their moving away is more probable. The
triturating and following ignition is very irregular and comes on only in case of suitable
crystal position and/or suitable crystal shape. The probability of ignition increases with
increasing friction velocity. This rule does not show itself very much in case of PETN
samples, due to the high PETN sensitiveness (see figure 6). Both PETN curves are almost
parallel. The both RDX samples have the same level of sensitiveness at minimal friction
velocity. The first reaction friction load of RDX Hx C sample, containing big crystals,
determined at maximal velocity, falls to one half of the load determined at minimal velocity.
This sample is more sensitive than PETN samples at the friction velocity of 10 cm.s-1 and
demonstrates the sensitivity increase with increasing friction velocity in case of less sensitive
samples containing big crystals.
PETN NpD 4,66 m/s
Load [N] CO/NO [ppm]
59,0
1/7
53,2
2/8
47,3
0/2
RDX Hx E 4,66 m/s
Load [N] CO/NO [ppm]
78,5
0/5
70,6
0/9
62,8
0/4
CO/NO [ppm]
CO/NO [ppm]
CO/NO [ppm]
CO/NO [ppm]
3/9
1/7
2/7
2/8
0/7
0/5
1/7
0/5
0/0
0/1
0/0
0/1
CO/NO [ppm]
CO/NO [ppm]
CO/NO [ppm]
CO/NO [ppm]
0/7
0/11
0/7
0/8
0/8
0/8
0/8
0/9
0/1
0/2
0/3
0/2
Table 2.: CO/NO concentration on load dependence
Miloslav Krupka, Jan Horkel, Jakub Selesovsky
671
Czech Republic
2006
[Content]
Some of the results closed to first reaction friction load for the samples PETN NpD and
RDX Hx E are stated in table 2 (friction velocity 4.66 cm-1). Despite the CO and NO
concentration measurement is not quantitative, it is sufficiently reproducible and it increases
the objectivity of sensitiveness determination without the respect to the experience of the
operator.
4.
CONCLUSION
The velocity of rubbing surfaces is an important factor which influences the results of
friction sensitivity testing. It was found that relationship between velocity of rubbing surface
and sensitiveness exists. The changes in sensitiveness are significant and can cause relevant
deviations of results. The big differences were observed in case of less sensitive samples
containing bigger crystals. This is one of the disadvantages of the BAM friction
sensitiveness apparatus design. The control of velocity and automated evaluation of test
results by toxic gas analyzer bring work simplification, safer and improve the reproducibility
of obtained results.
Acknowledgements
This work was created as a part of Ministry of Education, Youth and Sports f the Czeech
Republic No. MSM 0021627501.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
672
G. MILES, M.R. WILLIAMS, R.K. WHARTON, A.W. TRAIN: The Development of

an Elevated Temperature Fiction Sensitiveness Apparatus, 27th Int. Annu. Conf. Of
ICT, Karlsruhe, p. 31, 1996
L.R. SIMPSON, M.F. FOLTZ: LLNL Small-Scale Fiction Sensitivity (BAM) Test,
Informal Report No. UCRL-ID-124563, Lawrence Livermore National Laboratory,
1996
A. Bailey, G. Miles, A.W.Train, R.K. Wharton, M.R.Williams: The Effect of
Temperature on The Fiction Sensitiveness of Explosives Preliminary Study,
Nanotech, Houston, 2002
G. Miles: The development of an Elevated Temperature Friction Sensitiveness Test,
Explosives Engineering, 1996
Recommendations on the Transport of Dangerous Goods, Tests and Criteria,
St/SG/AC.10/11/Rev.3, Third Revised edition, United Nations, 1999
EN 13631-3:2004 Explosives for Civil Uses - High explosives Part 3: Determination
of sensitiveness to friction of explosives
Vyhlka B, kterou se stanov podrobnj podmnky pro povolovn vbunin,
vbunch pedmt a pomcek do obhu a jejich pezkuovn zk. 246/1996 Sb.
Vyhlka B ze dne 15.12.998, kterou se stanov metoda pro zjiovn vbunosti
chemickch ltek a ppravk, zk. 316/1998 Sb.
Czech Republic
2006
[Content]
THERMAL EFFECTS OF A SLOW THERMAL DECOMPOSITION

AND INFLAMMATION OF NITRAMINES AND ITS MIXTURES
WITH ALUMINUM, BORON AND ZIRCONIUM POWDERS
V. N. Lashkov, A. V. Strikanov, A. A. Tikhonova
RFNC-VNIIEF
37 Pr. Mira, Sarov, 607188, Nizhny Novgorod Region, Russia
Tel.: (831-30) 4-36-93; fax: (831-30) 4-49-53; e-mail: lashkov@dep19.vniief.ru
Abstract
At linear heating with rates from 2 up to 10 K/min the heats of thermal decomposition
of three HE of the nitramine class (cyclotetraethylentetranitramine, HMX; bistrinitroethylethylendinitramine, substance I; bis-trinitroethylnitramine - substance II),
distinguished by oxygen balance (-22 %; 0 %; +17 % correspondingly) have been
determined.
The values of thermal effects of the slow decomposition and inflammation are
approximately equal to half an ideal heat of explosive transformation of the given HEs.
The influence of the additives (of aluminum and boron nanopowders, zirconium powder)
on the decomposition process of the HE has been investigated. Additional release of
heat in mixtures with of aluminum nanopowder is realized only at decomposition of
HMX at maximum rate of heating (10 K /min). The zirconium powder (size of particles
(agglomerates) approximately 10 micrometers) does not increase the heat of
decomposition of the studied HEs at all rates of heating.
The additive of r boron nanopowder increases the heat of decomposition of all HEs at
all rates of heating except for 2 K/min for HMX.
From 4 factors influencing to the reactivity of the inflammable solid additives to the
energetic materials, namely: 1) temperature; 2) specific surface; 3) oxygen balance; 4)
diffusion resistance, first factor is main one (the 4-th factor is connected to the first).
The heats of thermal decomposition of three HEs of the nitramine class

(cyclotetraethylentetranitramine, HMX; bis-trinitroethylethylendinitramine, substance I; bistrinitroethylnitramin - substance II), distinguished by oxygen balance (-22 %; 0 %; +17 %
correspondingly) were determined with use the differential scanning calorimeter DSC-141
(Setaram).
The experiments with an opportunity of thermal explosion (inflammation) of the sample
were realized by using sealed indestructible crucibles and reduction of weight of the sample
to 0.001 g. At linear heating rates were from 2 up to 10 K/min.
For HE with formula such as CaHbNcOd oxygen balance = (d - (2a + b/2)) / M * 16 *
100 (%)
The basic properties of studied HEs are submitted in Table 1[1, 2, 3].
Valeriy Lashkov, A. Strikanov, A. Tikhonova
673
Czech Republic
2006
[Content]
Table 1. Properties of HEs

HE
Formula
Mol. mass
Oxygen
balance
Melting
temperature, C
Heat of
explosion, J/g
TMD,
g/ccm
HMX
48N88
296
-22
278(d)*
5632
1,9
II
44N814
388
90
5138
1,96
68N1016
476
+17
178(d)
6450
1,82
*(d) with decomposition

The influence of the additives (of aluminum and boron nanopowders, zirconium powder)
on the decomposition process of the HE has been investigated.
The aluminum nanopowder consists of spherical particles by a diameter 50-150
nanometers, contents 90% active aluminum, rest basically - alumina. The sizes of particles of
the boron nanopowder are close to the sizes of particles of aluminum nanopowder, however
the diffusion resistance of a oxide film is more lower than that of alumina.
The particles of the zirconium powder are porous durable agglomerates with a specific
surface up to 20 m2/g, the average size of agglomerates are ~ 5 micrometers. The content of
active zirconium is 75-80 %, the rest basically - zirconia.
674
Thermal Effects of a Slow Thermal Decomposition and Inflammation of . . .
Czech Republic
2006
[Content]
Table 2. Results of Thermal Analysis

Substance
1
R,
r , J /g
/m in
( v ) ( %)
HMX
10
2
HMX +
1 0 %A l
10
2
HMX +
5% B
10
2
HMX +
2 0 %Zr
10
II
10
3492
(2.4)
3476
(3.9)
3600
(4.9)
3063
(11.3)
2948
(21.9)
6696
(3.0)
3109
(5.7)
4632
(45.4)
5902
(8.1)
2318
3174
3298
3036
(24.9)
2716
(16.2)
4320
(0.78)
r / t , %
4
62
62
64
54
52
119
55
82
105
41
56
59
59
53
84
d (v),
0
( %)
max (v),
0
( %)
268.8
270.0
(0.39)
(0.26)
275.0
276.0
(0.25)
(0.23)
276.7
278.0
(0.36)
(0.33)
272.9
273.6
(0.37)
(0.46)
277.0
278.1
(0.14)
(0.15)
277.7
278.4
(0.09)
(0.09)
274.5
275.4
(0.22)
(0.17)
277.0
277.7
(0.13)
(0.31)
278.0
278.8
(0.23)
(0.24)
272.6
273.7
273.8
274.9
276.9
278.7
148.9
156.1
(0.40)
(1.1)
159.8
168.7
(0.64)
(0.21)
164.5
176.3
(0.06)
(0.28)
, s
36
12
7.8
21
13
4.2
27
8.4
4.8
33
13
11
216
107
71
675
2
2
II +
10%Al
10
2
II +
5%B
10
2
II +
2 0 % Zr
10
10
2
I +
10%Al
10
2
I+
5%B
10
+20%Z
r
5
10
676
3
3098
(17.6)
2941
(4.5)
3569
(2.5)
4503
(33.1)
4436
(17.1)
5189
(5.1)
3293
(6.8)
2160
(44.7)
4384
(12.4)
4197
(1.9)
3064
(20.8)
4193
(6.6)
3823
(6.8)
4078
(10)
4539
(21)
5143
(1.6)
5037
(0.12)
6446
(13.3)
4583
4100
4869
4
60
57
69
88
86
101
64
42
85
65
48
65
59
63
70
80
78
100
71
64
75
Czech Republic
148.9
156.6
(0.98)
(0.39)
158.9
167.9
(0.16)
(0.23)
162.1
176.1
(0.13)
(1.7)
148.4
158.3
(1.9)
(1.1)
154.4
165.9
(0.86)
(0.43)
158.1
172.1
(2.4)
(2.4)
147.7
158.7
(0.072)
(1.0)
161.1
169.3
(0.11)
(0.29)
167.6
176.9
(2.1)
(0.34)
182.5
184.8
(0.11)
(0.07)
188.6
191.4
(0.24)
(0.41)
193.7
197.4
(0.12)
(0.07)
183.4
185.7
(0.16)
(0.11)
189.6
192.3
(0.04)
(0.01)
194.2
197.6
(0.09)
(0.08)
190.1
191.4
(0.048)
(0.02)
193.7
195.2
(0.022)
(0.004)
197.0
198.5
(0.18)
(0.67)
184.0
185.9
185.8
189.5
193.0
197.1
2006
[Content]
231
108
84
297
138
84
330
98
56
69
34
22
69
32
20
39
18
9.0
57
44
25
Czech Republic
2006
[Content]
Crucible stainless steel, atmosphere - air

R - rate of heating;
s - real temperature;
r - real heat of process as average value or the integral of peak of a thermal flow
curve for "n" of experiments;
t - theoretical value of decomposition heat according to thermochemical calculation;
d temperature for the start of decomposition;
n - amount of experiments;
- temperature of the maximal rate of decomposition ;
- Time from the start of decomposition up to the maxima rate of decomposition;

v - factor of a variation.
From the Table 2 see it is seen:
. Substances without additive.
The thermal effect of decomposition for all studied substances, at all rates of heating are
approximately equal to half an value of heat of explosive transformation.
The decomposition occurs at the maximum rate in HMX: the time of the maximum rate
achievement is from 8 second for rate 10 K/min and about 50 second for rate 2 K/min, that is
characteristic for inflammation.
Character of decomposition of substance "II" is quiet, the time of the maximum rate
achievement makes 1-4 minutes. The substance "I" occupies an average position, the time of
the maximum rate achievement 20-70 seconds.
From types of decomposition of the given substances follows, that the character of
decomposition is determined by temperature and that phase, in where the decomposition
takes place: less stable chemically and having lower melting point substance "II" is
decomposed less intensively, than more thermo stable HMX since its decomposition goes on
in a liquid phase and at the temperature that is more than 100K lower. The less intensity of
the decomposition of substance "I" is caused just by a temperature factor, since its
decomposition, as well as for HMX, goes on simultaneously with a melting.
It should be noted that in same sets of experiments a point scattering of the results has
been extremely large (Kv more than 20%). It may be caused by instability of sample form.
B. Mixed substances
The interaction of aluminum nanopowder with the thermal decomposition products is
observed only for the decomposition of HMX at high rate of heating (10 K/min). In this case
the heat of HMX decomposition grows up more than 2 times. In mixtures of aluminum
nanopowder with substances "I" and "II" a growth the heat of decomposition is not observed.
This fact may be explained by the main role in heterogeneous reaction between an active
gas phase and aluminum particle of the diffusion resistance of the oxide film for reacting
species.
677
Czech Republic
2006
[Content]
The rate of diffusion process grows sharply up at higher temperature that is realized at
decomposition of HMX. In this case the sharp increase of heat effect at high rate of heating
in a mixture of HMX with aluminum nanopowder is observed.
Thermal decomposition of substance "I" and, especially, "II" goes on much more slowly,
despite of much higher contents of oxygen in its molecule. In decomposition products of
substance II diffusion difficulties are not overcome, the heterogeneous reaction does not
go on fast. This observation proves assumption stated in work[4] that at high temperatures in
relation to particles of aluminum covered by oxide film, water steam is more active oxidizer,
than the pure oxygen.
The fact, that the main role in reaction of aluminum particles with products of
decomposition of the studied nitramines plays a diffusion factor, proves results of thermal
analysis of mixtures with the boron nanopowder.
Boron as the chemical element is closed to aluminum in respect of energy of formation
of oxide and to chemical similarity in relation to products of decomposition of nitramines.
The specific surface of a studied boron nanopowder is also close to the specific surface of
aluminum nanopowder.
It was already mentioned, essential difference of these two powders in relation to
reaction with oxygen-containing gases (H2O, CO2, CH2O) are diffusion difficulties.
As it is seen from thermal analysis results, submitted in Table 2, the sharp reduction of
diffusion barrier in the case of boron results that additional realize of decomposition heat of
HMX mixtures begins already at rate 5 K/min and grows by 20 % at the increase of rate up
to 10 K/min
In boron mixtures with substances "I" and "II" additional heat release is observed
already at the minimum rate of heating. For substance "II" the rather additional heat release
remains constant at all rates of heating, but for substance "I", as well as for HMX additional
heat release grows at the maximal rate.
The process of decomposition of substance "II" goes on at lower temperatures, than of
substance "I" and, especially, for HMX, it is less intensive, is more protracted in time. The
losses of heat are significant, that results in the decrease of temperature at a maximum rate.
This is probably due to incomplete boron reaction. This reason explains also the lowered
heat release in boron mixture with substance "II" in comparison with HMX and substance
"I".
It confirms the assumption that a major factor determining the rate of interaction of an
active additive with products of thermal decomposition the nitramines, is the diffusion
resistance of oxide film. The resistance decreases at the increase of temperature, and this
reduction depends on properties of oxide. For boron the diffusion resistance is lower, than
for aluminum even at the lower temperatures. The high specific surface of particles results in
increase of reactivity. The low reactivity of the zirconium powder is caused by strong
agglomeration of its primary particles. So, despite of the high specific surface of pores in
agglomerates, the penetration of products of decomposition inside of large agglomerates is
difficult.
It is possible to assert that from four factors influencing to the reactivity of the
inflammable solid additives to the energetic materials, namely: 1) temperature; 2) specific
surface; 3) oxygen balance; 4) diffusion resistance the first factor is the main one (the 4-th
factor is connected to the first).
678
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
Chemical encyclopedia, in five volumes, " The Soviet Encyclopedia ", ., 1988, (in Russian).
G. B. Manelis, et al. Thermal decomposition and combustion of explosives and gunpowders,
"Science", ., 1996, (in Russian).
E. Yu. Orlova et al. HMX - heat-resistant explosives. "Nedra", ., 1975, (in Russian).
V.M. Maltsev, et al. The basic characteristics of combustion. "Chemistry", ., 1977, (in
Russian).
679
Czech Republic
2006
[Content]
SENSITIVITY RESEARCH OF EXPLOSIVE MIXTURES

CONTAINING NITROGUANIDINE
AND 3-NITRO-1,2,4-TRIAZOL-5-ONE
A. Maranda*, J. Nowaczewski*, A. Orzechowski**, D. Powaa**
* Military University of Technology, 2 Kaliskiego St, 00-908 Warsaw 49, PL
** Institute of Industrial Organic Chemistry, 6 Annopol St, 03-236 Warsaw, PL
Abstract:
In order to obtain mixtures of low sensitivity to external stimuli, the explosives like
octogene (HMX) or hexogene (RDX) were mixed with different quantities of
nitroguanidine (NQ) or 3-nitro-1,2,4-triazol-5-one (NTO). In this paper, a method of
preparation of charges to tests is described. The measurement results of the sensitivity
to friction and shock wave generated by water layer are presented. On the basis of the
results it was found that, the addition of NQ or NTO considerably decreases friction
sensitivity of the compositions, whereas the sensitivity to shock wave is decreased in a
lesser degree.
1.
INTRODUCTION
During last years the main subject in world development of the military technique field
we could notice that so many people are interested in explosives of low sensitivity to
external stimuli in comparison with widespread like: trinitrotoluene (TNT), hexogene,
octogene, pentaerythritol trinitrate (PETN). This is the result of concerning about soldiers
security.
Unexpected explosion and unsuitable handle of explosives can sometimes lead to
accidents. The experts claim that we should reduce the number of accidents in the army by
supplying modern low sensitivity ammunition.
The following materials belong to the group of less sensitive high explosives : 1,2,4triazol -5-one (NTO), 1,3,5-triamino-2,4,6-trinitrobenzene (TATB), 2,2,4,4,6,6heksanitrostilbene (HNS), tetranitrodibenzo- 1,3a,4,6a-tetraazapentalene (TACOT), and
nitroguanidine (NQ) [1 7].
2.
SELECTION AND PREPARATION OF EXPLOSIVE MIXTURES
For these studies, less sensitive mixtures of different chemical constitution were
selected. The mixtures included hexogene or octogene as components of high sensitivity and
NQ or NTO as components which decreased the sensitivity. In order to monitor the influence
of changes of chemical contents on explosive properties of these mixtures, we decided to
make changes in mutual percentage fraction of the particular components and finally we
received twelve mixtures. This set of compositions was put to various tests.
680
Sensitivity Research of Explosive Mixtures Containing Nitroguanidine and . . .
Czech Republic
2006
[Content]
Table 1. Chemical composition of tested explosive mixtures of reduced sensitivity

(IHE)
Type of mixture
Percentage
(RDX+0,5%C) / NQ
70/30
50/50
30/70
(RDX+0,5%C) / NTO
(HMX+0,5%C) / NQ
70/30
50/50
30/70
70/30
50/50
30/70
(HMX+0,5%C) / NTO
70/30
50/50
30/70
The majority of the tests was realized on the samples of mixtures which was pressed by
hydraulic press to the possible high density. In order to obtain the mixtures, the components
were weighed in portions (1000 g), and mixed by shaking in foil container for 8 10
minutes. The loose mixture was divided in portions (10 - 50 g) and was pressed under
pressure 200 MPa. The diameters of dies were 20 - 40 mm. The density of obtained blocks
was determined by measurements of weight and dimensions. The obtained blocks were
glued together by nitrocellulose lacquer. The columns formed in this way were tested.
3.
SENSITIVITY TO FRICTION OF TESTED MIXTURES
The measurements of the sensitivity to friction were made according to PN-C86019:1994 standard by using Joulus-Petri apparatus. The lower insensitivity point of
mixtures was determinated. It is the maximum pressure of stamp (in Newton), which causes
minimum one positive result (explosive process) during six trials. The loose mixtures were
used to the all tests. The results of our experiments are given in table 2 and figure 1.
Table 2. Measurement results of sensitivity to friction of tested mixtures
Explosive
RDX/NTO
RDX/NQ
HMX/NTO
HMX/NQ
RDX PETN
compositions 70/30 50/50 30/70 70/30 50/50 30/70 70/30 50/50 30/70 70/30 50/50 30/70 cryst. cryst.
Friction
247
283
353
235
283
353
235
247
318
212
235
318
120
60
sensitivity
[N]
Andrzej Maranda, Jerzy Nowaczewski, Andrzej Orzechowski, Dorota Powaa
681
Friction sensitivity [N]
Czech Republic
2006
[Content]
400
350
300
250
200
283
247
353
235
283
353
235
247
318
212
235
318
150
100
120
50
60
0
RDX/NTO RDX/NTO RDX/NTO
70/30
50/50
30/70
Fig 1.
4.
RDX/NQ
70/30
RDX/NQ
50/50
RDX/NQ
30/70
HMX/NTO HMX/NTO HMX/NTO HMX/NQ

70/30
50/50
30/70
70/30
HMX/NQ
50/50
HMX/NQ
30/70
RDX
PETN
The comparison of friction sensitivity of tested mixtures with sensitivity of

crystalline explosives.
SENSITIVITY TO SHOCK WAVE OF TESTED MIXTURES
The most widespread method of sensitivity examination of explosives on shock wave is

a gap test [8]. The various modifications of this test are known. For different cases, the kind
and diameter of standard explosive which generated shock wave and also the kind of barrier
are changed. The size of tested charges can also be changed. The plexiglas layer is the most
often applied as a barrier to the gap test. This material owns stable physical properties and
the known dependence of shock wave amplitude on thickness of layer. The hexogene/wax as
the standard explosive is used.
In this work, the studies of explosive sensitivities to shock wave were carried out
according to the Timborn method [9]. The advantages of the method are simplicity and
accuracy of measurements and also small weight of tested charges. In figure 2, the system
elements for measurements are presented.
The measurements were carried out according to the following method. The donor
charge (4) was put into the part of the PCV tube (5) of 20 mm inside diameter and coating by
lubricant from syringe(2). This charge was placed for suitable depth (measured by slide
caliper). The upper edge of the tube was also coated by lubricant and the joining tube (6) was
placed on the tube (5). Next, water was poured into the tube (5) and the second part of tube
(7) with tested charge was put into the joining tube. The tubes were held down together and
such system was fixed to the indicator plate (9) by using sticky tape. The detonation cord (3)
was fixed into the nest of tested charge and was also stuck to the indicator plate. This plate
was hung in bunker and then the donor charge was fused by detonator (1) and initiated. The
scheme of this system is shown in figure 3.
682
Czech Republic
2006
[Content]
2
3
8
5
Fig 2.
The system elements for measurments of sensitivity to shock wave of

explosives: 1 detonator, 2 syringe with lubricant, 3 detonating cord,
4 donor charge, 5 PCV tube for donor charge, 6 joining tube, 7 PCV
tube for tested charge, 8 tested charge, 9 indicator plate
683
Czech Republic
2006
[Content]
1
2
4
5
8
9
Fig 3.
The system for measurments of sensitivity to shock wave of explosives: 1 indicator plate, 2 - detonating cord, 3 upper PCV tube, 4 tested charge,
5 - joining tube, 6 water layer, 7 - donor charge, 8 - lower PCV tube, 9 detonator
After the initiation of detonation in tested charge (7), the shock wave permeates through
the water layer (6) and depending on wave intensity (regulated by thickness of water layer)
the charge detonates or it does not. Since the applied charges are small, only clear trace after
detonation of detonating cord (2) on indicator plate (1) confirms that the tested charge
(4)was detonated.
684
Fig 4.
2006
[Content]
The indicator plates after detonation of tested charges:

1 RDX/NQ 70/30-24 mm, 2 RDX/NQ 70/30-26 mm, 3 RDX/NQ
50/50-22 mm, 4 RDX/NQ 30/70-22 mm, 5 RDX/NQ 70/30-24 mm, 6
RDX/NTO 70/30-21 mm, 7 - RDX/NTO 30/70-23 mm, 8 HMX/NQ
70/30-21mm
Fig 5.
Czech Republic
The indicator plates and pieces of detonating cord unless tested charges
were detonated: 1 RDX/NQ 70/30-27 mm, 2 RDX/NQ 70/30-27 mm, 3
HMX/NTO 70/30-24 mm, 4 RDX/NTO 30/70-24 mm, 5 HMX/NQ
30/70-22 mm, 6 RDX/NTO 70/30 -5 mm
The appearance of indicator plates after the detonation of tested charges is presented
in figure 4, whereas the plates without detonation of the charges are shown in figure 5. The
number of millimeters which is presented in the description indicates thickness of water
layer which shock wave permeates through. The measurement results of sensitivity for
mixtures RDX/NQ, RDX/NTO, HMX/NQ, HMX/NTO are presented in table 3 and figure 6.
685
Czech Republic
2006
[Content]
Table 3. Measurement results of sensitivity to shock wave of tested mixtures

Composition of explosives
Density
[g/cm3]
HMX/NTO 70/30
HMX/NTO 50/50
HMX/NTO 30/70
HMX/NQ 70/30
HMX/NQ 50/50
HMX/NQ 30/70
RDX/NTO 70/30
RDX/NTO 50/50
RDX/NTO 30/70
RDX/NQ 70/30
RDX/NQ 50/50
RDX/NQ 30/70
1,75
1,77
1,77
1,69
1,66
1,62
1,66
1,71
1,73
1,64
1,62
1,60
P [GPa]
1,2
Tickness of water leyers

[mm]
Detonation No detonation
22
23
21
22
20
21
23
24
22
23
20
21
24
25
23
24
23
24
26
27
22
23
22
23
Shock wave
pressure [GPa]
1,00
1,05
1,20
1,00
1,05
1,20
0,95
1,00
1,00
0,75
1,05
1,05
1,20
1,20
1,05
1,05
1,00
1,05
1,05
1,00
1,00
1,00
0,8
0,95
0,6
0,75
0,4
0,2
0
30/70
HMX/NTO
HMX/NQ
50/50
RDX/NTO
RDX/NQ
Fig 6.
70/30
The sensitivity to shock wave of explosive mixtures containing NQ and

NTO permeating through variable water layers.
The comparison of results reported by Timborn [8] (six columns on the left) with
chosen results which was obtained for the most sensitive tested mixtures (six columns on the
right) is presented in figure 7.
686
Czech Republic
2006
[Content]
P [GPa]
1,80
1,60
1,70
1,40
1,20
1,20
1,20
1,20
1,20
1,00
1,05
1,00
1,00
0,80
0,75
0,70
0,60
0,40
0,20
0,00
RDX/TNT 60/40
Fig 7.
5.
RDX/WOSK
95/5
PETN/WOSK
97/3
TETRYL
PETN/TNT
50/50
TNTpras.
RDX/NQ 70/30
HMX/NTO
70/30
HMX/NTO
50/50
HMX/NTO
30/70
The comparison of capacity to detonation through the water layer of some

explosives.
SUMMARY
The comparison of measurement results shows that the addition of less sensitive
component causes considerable decrease of sensitivity to friction of tested mixtures. The
content of 30% NQ or NTO in mixture generates double decrease of the sensitivity. It can be
stated that, only at a 70% share of the additives these mixtures are less sensitive to friction.
A load of above 300 N is required to the initiation of these mixtures.
The studies of sensitivity to shock wave of tested mixtures were made according to
the modified gap test (Tab.3, Fig. 6-7). These tests show that the sensitivity of mixtures
with RDX is higher than the sensitivity of mixtures with HMX. In the range of changeable
quantity of the same additives, the sensitivity was changed according to expectations. Tested
mixtures have a sensitivity level such as pressed TNT, pentolite 50/50, or desensitized RDX
Only at a 70% share of low sensitive additive. At different content of the additive, the
sensitivity level correspond to the tetryl.
687
6.
Czech Republic
2006
[Content]
REFERENCES
[1]. G.I. CHIPEN, R.P. BOKALDER, V.YA. GRISHTEN, 1,2,4-Triazol-3-one and Its Nitro and
Amino Derivatives, Khimia Geterogennykh Soedinienii, 1966, 2, 1.
[2]. K-Y. LEE, M.D. COBURN, 3-Nitro-1,2,4-triazole-5-one a Less Sensitive Explosive, Los Alamos
National Laboratory Report No. LA-10303-MS, Feb 1985.
[3]. K-Y. LEE, M.D. COBURN, 3-Nitro-1,2,4-triazole-5-one a Less Sensitive Explosive, Journal of
Energetic Materials, 1987, 5, 27-33.
[4]. K-Y. LEE, M.D. COBURN, 3-Nitro-1,2,4-triazole-5-one a Less Sensitive Explosive, pat USA
4733610, 1988.
[5]. E. VON RIPPER, G. K, ber - und -Nitroguanidin, Explosivstoffe 1969, 17, 7, 145 151.
[6]. W. ENGEL, H. HEINISCH, Process for Preparation of Compact Nitroguanidine, pat. USA 4
544 769, 1985.
[7]. J.A. SANCHEZ, E.L. ROEMER, L.A. STRETZ, Spherical Nitroguanidine Process, pat. USA 4
967 000, 1990.
[8]. M. HELD, GAP TESTS, Energetic materials, Reactions of Propellants, Explosives and
Pyrotechnics, 34-th International Annual Conference of ICT, Karlsruhe, FRG, 24-27 June,
2003.
[9]. R. MEYER, J. KHLER, A. HOMBURG, Explosives, 5 rev. and extended ed., Wiley VCH
Verlag GmbH, Weincheim 2002.
688
Czech Republic
2006
[Content]
CYTOTOXICITY OF NOVEL NITROTRIAZOLE AND

TRINITROANILINE EXPLOSIVES: THE ROLE OF OXIDATIVE
STRESS AND ACTIVATION BY DT-DIAPHORASE
Aura Nemeikait-nien*, Jonas arlauskas**, Henrikas Nivinskas**,
ilvinas Anuseviius**, and Narimantas nas**
* Institute of Immunology of Vilnius University, Molt Pl. 29, LT-08409, Vilnius, Lithuania
** Institute of Biochemistry, Mokslinink 12, LT-08662, Vilnius, Lithuania
Abstract:
We have examined the enzymatic single- and two-electron reduction of nitrotriazole
explosives NTA, MNTA, and triazolo/trinitroaniline explosive PATO by
ferredoxin:NADP+ reductase (FNR) and DT-diaphorase (NQO1), respectively, and
determined their cytotoxicity in FLK cells. The cytotoxicity of NTA and MNTA was
comparable to that of nitrobenzene, whereas the cytotoxicity of PATO was comparable
to that of 2,4,6-trinitrotoluene. The order of cytotoxicity of compounds matched their
reactivity in FNR-catalyzed reactions, thus showing that the main mechanism of
cytotoxicity is redox cycling of free radicals. In line with this, the cytotoxicity was partly
decreased by the antioxidants. However, the protective effects of the inhibitor of
NQO1, dicumarol, show that this enzyme also plays a definite role in the cytotoxicity of
examined compounds.
Keywords:
1.
nitrotriazole explosives, enzymatic reduction of, cytotoxicity
INTRODUCTION
Conventional nitroaromatic explosives like 2,4,6-trinitrotoluene (TNT) are toxic to

mammalian species, causing damage to liver, kidney and spleen, as well as
methemoglobinemia and hemolytic crisis.[1] The mechanisms of toxicity of TNT and other
polynitroaromatics involve the oxidative stress caused by the flavoenzyme-catalyzed freeradical formation and their redox cycling, and the formation of nitroso- and/or
hydroxylamino products of two-electron reduction which may modify proteins and DNA, as
well as induce methemoglobin formation.[2,3]
Continuing our previous studies,[4-7] in this paper we analyze the relationship between
the reactivity of novel nitrotriazole explosives NTA, MNTA, and triazolo/trinitroaniline
explosive PATO towards the single- and two-electron transferring flavoenzymes, and their
mammalian cell cytotoxicity.
2.
MATERIALS AND METHODS
1-Methyl-3,5-dinitro-1,2,4-triazole (MNTA), 3,5-dinitro-1,2,4-triazole (NTA), and N(1,2,4-triazol-3-yl)-2,4,6-trinitrophenylamine (PATO) were synthesized according to the
published methods. [8-10] The purity of compounds has been determined using melting points,
TLC, NMR, IR, and elemental analysis. All the other compounds were obtained from Sigma
and used as received.
Ausra Nemeikaite-Ceniene, Jonas Sarlauskas, Henrikas Nivinskas, . . .
689
Czech Republic
2006
[Content]
The kinetic measurements were carried out spectrophotometrically using a Hitachi-557

spectrophotometer in 0.1 M K-phosphate buffer (pH 7.0) containing 1 mM EDTA at 25 oC.
Ferredoxin:NADP+ reductase (FNR, EC 1.18.1.2) was prepared as described,[5] and was a
generous gift of Dr. M. Martinez-Julvez and Professor C. Gomez-Moreno (Zaragoza
University, Spain). The enzyme concentration was determined using 459 = 9.4 mM-1cm-1.
Rat liver DT-diaphorase (NQO1, EC 1.6.99.2) was prepared as described,[4] its concentration
being determined using 460 = 11 mM-1cm-1. In the experiments with NQO1, 0.01 % Tween
20 and 0.25 mg/ml bovine serum albumin were added as activators. The rates of enzymatic
oxidation of NADPH by nitroaromatic compounds were determined according to 340 = 6.2
mM-1cm-1 in the presence of 100 M NADPH. The catalytic constant (kcat) and the
bimolecular rate constant of the reduction of nitroaromatics (kcat/Km) were calculated from
the Lineweaver-Burk plots. kcat is the number of NADPH molecules oxidized by the single
active center of an enzyme per second.
The enthalpies of nitroanion-radical formation (Hf(ArNO2-.)) by the AM1 and PM3
methods were calculated using PC Spartan Pro (cersion 1.0.1, Wavefunction).[4-6]
The culture of bovine leukemia virus-transformed lamb kidney fibroblasts (line FLK)
was grown and maintained in Eagles medium supplemented with 10 % fetal bovine serum
at 37 oC as described.[4-7] In the cytotoxicity experiments, cells (3.0 x 104/ml) were grown in
the presence of various amounts of compounds on glass slides for 24 h. Further, the cells
were washed several times with the phosphate-buffered saline. The viable (adherent) cells
were counted under a light microscope with the viability determined by the exclusion of
Trypan blue.
3.
In this work, we examined the enzymatic redox reactions of NTA, MNTA, and PATO
which may be important in their cytotoxicity: i) their single-electron reduction by
flavoenzyme electrontransferase ferredoxin:NADP+ reductase (FNR) which is frequently
used as a model system to study the free radical formation of nitroaromatic compounds and
their redox cycling ability.[4-7] The reactivity of FNR increases with the increase in electronaccepting potency of nitroaromatics (single-electron reduction potential), and is insensitive
to their structural peculiarities; and ii) their two-electron reduction by rat NAD(P)H:quinone
oxidoreductase (DT-diaphorase, NQO1), which is one of the main enzymes responsible for
the formation of toxic nitroso- and/or hydroxylamine reduction products in mammalian
cells.[4-7] We also determined the concentrations of the above compounds for 50 % survival
of FLK cells (cL50). The results obtained are presented in Table 1, together with the
previously determined enzymatic reduction rate constants and cL50 of other nitrotriazole
explosives, 5-nitro-1,2,4-triazol-3-one (NTO) and 5-nitro-1,2,4-triazol-3-amine (ANTA),
and model compounds, nitrobenzene and 2,4,6-trinitrotroluene (TNT).
According to the calculated Hf(ArNO2-.) values, the electron-accepting potency of
NTA and MNTA in vacuo is intermediate between those of TNT and nitrobenzene, whereas
the electron-accepting potency of PATO is close to TNT (Table 1). In general, the rate
constants of single-electron reduction of nitroaromatics by FNR increase with the decrease in
their Hf(ArNO2-.), although the dependence is uncertain (Table 1). It is evident that the
order of cytotoxicity of compounds, nitrobenzene ~ NTO, ANTA NTA, MNTA < PATO <
TNT almost parallels their reactivity towards FNR (Table 1), i.e, the rate of enzymatic free
radical formation. It points to the possibility of the oxidative stress-type cytotoxicity
mechanism of NTA, MNTA, and PATO. The prooxidant character of cytotoxicity is further
690
Cytotoxicity of Novel Nitrotriazole and Trinitroaniline Explosives: The Role of . . .
Czech Republic
2006
[Content]
evidenced by the partial protection by the low molecular weigth antioxidant N,N-diphenylp-phenylene diamine (DPPD), the iron-chelating agent desferrioxamine, the latter preventing
the Fenton reaction, and by the antioxidant enzyme catalase (Table 2). However, an inhibitor
of NQO1, dicumarol, also partly protects against the cytotoxicity (Table 2), thus pointing to
the toxic effects of the products of NQO1-catalyzed reduction.
Table 1. Enthalpies of the formation of nitroanion-radicals, rate constants of the
reduction of nitroaromatic compounds by FNR and NQO1, and their
concentrations for 50 % survival of FLK cells (cL50). aThe data taken from
our previous work;[6] bThe kcat values (s-1) are given in parentheses; ccL50
is close to the solubility limit.
Hf(ArNO2-.) (kJ/mol)
Compound
kcat/Km (M-1s-1)
cL50
b
(M)
AM1
PM3
FNR
NQO1
NTA
-280
-293
107
15 (0.03)
2700380
MNTA
-271
-245
240
40 (0.10)
2300300
PATO
-307
-318
46000
2400 (0.14)
50c
NTOa
-242
-236
55
27 (0.09)
3500c
ANTAa
-148
-159
260
29 (0.09)
3000400
-308
-316
11000
670 (1.0)
255.0
-168
-172
11
24 (0.06)
43701470
TNT
Nitrobenzene
Table 2. The effects of antioxidant N,N-diphenyl-p-phenylene diamine (DPPD),

Fe-ion chelator desferrioxamine, catalase, and inhibitor of NQO1,
dicumarol, on the survival of FLK cells in the presence of PATO, MNTA,
and NTO. aThe data taken from our previuos work.[6] bStatistically
insignificant (p > 0.05).
No.
Compound
Cell viability (%)

no
additions
PATO (50 M)
MNTA (2000 M)
2 M
DPPD
300 M
desferrioxamine
100 U/ml 20 M
catalase dicumarol
59.02.0
74.53.5 68.93.5b 71.93.0
88.24.0
58.42.7
88.83.9
98.02.5
82.23.0
84.12.8
3
NTO (3000 M)
55.03.7 75.83.5 78.23.5
n.d.
68.03.8b
[6]
This work and our previous study demonstrate that the presence of triazole group in
NTO, ANTA, NTA, MNTA, and PATO does not affect their mammalian cell cytotoxicity,
Ausra Nemeikaite-Ceniene, Jonas Sarlauskas, Henrikas Nivinskas, . . .
691
Czech Republic
2006
[Content]
i.e., these compounds follow the same electron-accepting potency vs. cytotoxicity
relationship as polynitrobenzenes.
Acknowledgements
This work was supported in part by Lithuanian Science and Studies Foundation. We
thank Dr. M. Martinez-Julvez and Professor C. Gomez-Moreno (Zaragoza University,
Spain) for their generous gift of ferredoxin:NADP+ reductase.
REFERENCES
[1]
S. LETZEL, T. GOEN, M. BADER, J. ANGERER and T. KRAUS: Exposure to Nitroaromatic

Explosives and Health Effects During Disposal of Military Waste, Occup. Environ. Med., 60,
483-488, 2003, and references therein.
[2] L. KONG, Q. JIANG and Q. QU: Formation of Superoxide Radical and Hydrogen Peroxide
Enhanced by Trinitrotoluene in Rat Liver, Brain, Kidney, and Testicle in Vitro and Monkey
Liver in Vivo, Biomed. Environ. Sci., 2, 72-77, 1989
[3] K.H. LEUNG, M. YAO, R. STEARNS and S-H.L. CHIU: Mechanism of Bioactivation and
Covalent Binding of 2,4,6-Trinitrotoluene, Chem.-Biol. Interact., 97, 37-51, 1995
[4] N. NAS, A. NEMEIKAIT-NIEN, E. SERGEDIEN, H. NIVINSKAS, .
ANUSEVIIUS and J. ARLAUSKAS: Quantitative Structure-Activity Relationships in
Enzymatic Single-Electron Reduction of Nitroaromatic Explosives: Implications for their
Cytotoxicity, Biochim. Biophys. Acta, 1528, 31-38, 2001
[5] J. ARLAUSKAS, A. NEMEIKAIT-NIEN, . ANUSEVIIUS, L. MISEVIIEN, M.
MARTINEZ-JULVEZ, M. MEDINA, C. GOMEZ-MORENO and N. NAS: FlavoenzymeCatalyzed Redox Cycling of Hydroxylamino- and Amino Metabolites of 2,4,6-Trinitrotoluene:
Implications for their Cytotoxicity, Arch. Biochem. Biophys., 425, 184-192, 2004
[6] J. ARLAUSKAS, A. NEMEIKAIT-NIEN, . ANUSEVIIUS, L. MISEVIIEN, A.
MAROZIEN, A. MARKEVIIUS and N. NAS: Enzymatic Redox Properties of Novel
Nitrotetrazole Explosives: Implications for their Toxicity, Z. Naturforsch., 59c, 399-404, 2004
[7] A. NEMEIKAIT-NIEN, J. ARLAUSKAS, L. MISEVIIEN, . ANUSEVIIUS, A.
MAROZIEN and N. NAS: Enzymatic Redox Reactions of the Explosive 4,6Dinitrobenzofuroxan (DNBF): Implications for its Toxic Action, Acta Biochim. Polon., 51,
1081-1086, 2004
[8] A.K. SIKDER, M. GEETHA, D.B. SARWADE and J.P. AGRAWAL: Studies on
Characterization and Thermal Behaviour of 3-Amino-5-Nitro-1,2,4-Triazole and its
Derivatives, J. Hazard. Mat., 82A, 1-12, 2001
[9] L.I. BAGAL, M.O. PEVZNER, A.N. FROLOV and N.I. SHELUDYAKOVA: Synthesis of
Nitroderivatives of 1,2,4-Triazole, 1,3,4-Thiadiazole, Tetrazole, 1,3,4-Oxadiazole and Pyrazole
by means of Noncatalytic Substitution of Diazo Group into Nitrogroup, Khim. Heterocykl.
Sojed., 2, 259-264, 1970
[10] P.F. POGORIA, G.S. LEE, A.R. MITCHELL and R.D. SCHMIDT: A Review of Energetic
Materials Synthesis, Thermochim. Acta, 384, 187-204, 2004
692
Cytotoxicity of Novel Nitrotriazole and Trinitroaniline Explosives: The Role of . . .
Czech Republic
2006
[Content]
OPTIMALIZATION OF CRYSTALLIZATION OF
1,1-DIAMINO-2,2-DINITROETHYLENE
Andrzej Orzechowski*, Dorota Powaa *, Bogdan Florczak*,
Andrzej Maranda**, Jerzy Nowaczewski**
Institute of Industrial Organic Chemistry, 6 Annopol St, 03-236 Warsaw, Poland
Military University of Technology, 2 Kaliskiego St, 00-908 Warsaw 49, Poland
Abstract:
In this paper, the method of synthesis and crystallizations of 1,1-diamino-2,2dinitroetylene was presented. The aim of the work was to obtain cubic crystals of
FOX-7. Several solvents for crystallization of FOX-7 were checked. The influence of
various parameters on crystallization process was investigated The mixture DMSO and
water as a solvent to crystallization was the best.. The particle size distribution and the
tap density of the obtained crystals are presented.
Keywords: crystallization, 1,1-diamino-2,2-dinitroetylene, FOX-7
1.
INTRODUCTION
Growth of terrorist threats and military technology development causes the necessity
of new insensitive explosives elaboration. Hexogen, nitropenta, octogen are commonly used
high explosives. They are characterized by high explosive energy and low sensitivity to
external stimuli. The increase of resistance to mechanical, electrical and thermal, without
worsening other parameters, is realized by the addition of such explosives like
nitroguanidine (NGU), 3-nitro-1,2,4-triazol-5-one (NTO) or 1,3,5-triamino-2,4,6trinitrobenzene (TATB) [1-3].
FOX-7 is an interesting new high explosive because of its low vulnerability [4]. In the
literature, various synthesis methods of FOX-7 were referred. The best methods of FOX-7
preparing from acetamidine are described in [5-7], and optimization of crystallization methods
of this compound are shown in [5-10].
In our laboratory, we conduct researches on crystallization of FOX-7 in order to obtain a
suitable form of this explosive for application in Plastic Bonded Explosives.
2.
EXPERIMENTAL
Properties of 1,1-diamino-2,2-dinitroetylene
FOX-7 is a yellow crystalline substance, the density is 1.885 g/cm3 [11]. The impact
sensitivity of this explosive is 24.7 Nm and the friction sensitivity is above 350 N. The
solubility of FOX-7 in popular organic solvents i.e. acetone, methyl and ethyl alcohols, ethyl
acetate, ether, carbon tetrachloride, toluene, is limited, except N,N- dimethylformamide,
dimethylsulfoxide, N-methylpyrrolidone.
2.1
Synthesis of 1,1-diamino-2,2-dinitroethylene
FOX-7 was prepared by reacting dioctyl malonate with acetamidine hydrochloride in the
solution of sodium methylate, and next nitration of obtained 2-methylpyrimidyne1
Andrzej Orzechowski, Dorota Powaa, Bogdan Florczak, Andrzej Maranda, . . .
693
Czech Republic
2006
[Content]
4,6(3H,5H)-dione (Fig 1). According to [8] 1 mole of acetamidine hydrochloride was added
to the 2,5 mole of the sodium methylate solution in methanol (concentration of CH3ONa was
97 %). At room temperature 1 mole of the dioctyl malonate solution in methanol was added.
The mixture was stirred at 50 oC for 2 hours and then, the one was left for 24 hours (20 oC).
After that hydrochloric acid was added to the mixture (to pH 6). The second step was
nitration of methylpyrimidyne-4,6(3H,5H)dione. This compound was dissolved in
concentrated sulfuric acid (about 11 mole) at 20 oC, next the nitric acid was added at the
same temperature for 2 hours (H2SO4/ HNO3 3/1 molar ratio).The yield of the reaction was
74 %.
12
10
Volume [%]
0
5,7
137
270
403
535
668
801
Particle diameter [m x 10-6]
Fig 1.
2.2
Crude crystals of FOX-7
Crystallization of 1,1-diamino-2,2-dinitroethylene
In order to obtain applicable form of FOX-7 we carried out several experiments of

crystallization. As a solvent we used N,N- dimethylformamide, dimethylsulfoxide, Nmethylpyrrolidone. We tested solubility of 1,1-diamino-2,2-dinitroetylene in above
mentioned solvents (Fig. 2). During the tests different parameters were changed, for
example: solvents, mass ratio of the compound, stirring speed and cooling rate.
The mixture consisted of the crude product and the suitable solvent. The mixture was
stirred and gradually heated until the crystals were dissolved. The solution was gradually
cooled with various cooling rate. The crystals were filtered and washed with water and
acetone. The shape and size of obtained crystals were depended on parameters of
crystallization.
100
90
Solubility [%]
80
70
DMSO
60
DMSO + H2O
DMF
50
NMP
40
30
20
10
0
20
30
40
50
60
70
80
90
Temperature [ C]
Fig 2.
694
Figure 2. Solubility of FOX-7

Czech Republic
2006
[Content]
Crystallization from NMP

FOX-7 was added to a flask with NMP. The mixture was heated with mechanical
stirring 800 rpm to 70 C. Acetone was added to the mixture. The product was filtered and
washed with water and acetone. In this test conditions we were not able to obtain the suitable
form of FOX-7. The tap bulk density of FOX-7 was 0.45 g/cm3. The crystals are presented in
Fig. 3.
20
18
16
Volume [%]
14
12
10
8
6
4
2
0
5,7
137
270
403
535
668
801
Fig 3.
Crystals of FOX-7 obtained by crystallization from NMP
Crystallization from DMF

Example 1
The mixture of FOX-7 and DMF was gradually heated with mechanical stirring (900
rpm) to 90 C. Then the mixture was gradually cooled to 5 C with cooling rate 5 C/min.
The product was filtered and washed with water and acetone. The bulk density of FOX-7
was 0.91 g/cm3. The product is shown in Fig. 4.
Volume [%]
0
5,7
137
270
403
535
668
801
Fig 4.
Crystals of FOX-7 obtained by crystallization from DMF (example 1)
Example 2
This test was carried out in the same way just like in example 1. The difference between
these two examples was cooling rate. In this case it was about 20 C/min. The obtained
695
Czech Republic
2006
[Content]
product is presented in Fig. 5. Tap bulk density 0.63 g/cm3. We obtained lover bulk density
of FOX-7 than above.
3,5
Volume [%]
2,5
1,5
0,5
0
5,7
137
270
403
535
668
801
Fig 5.
Example 3
FOX-7 was added to a flask with DMF. The mixture was heated with mechanical
stirring (800 rpm) to 90 C. Acetone was added to the mixture. The product was filtered and
washed with water and acetone. The bulk density of FOX-7 , produced in this way, was 0.90
g/cm3. The crystals are presented in Fig. 6.
12
10
Volume [%]
0
5,7
137
270
403
535
668
801
Fig 6.
Crystallization from DMSO

Example 1
The mixture of FOX-7 and DMSO was gradually heated with mechanical stirring (900
rpm) to 90 C. The mixture was cooled to 5 C with cooling rate 5 C/min. The product was
filtered and washed with water and acetone. The bulk density of FOX-7 was 0.96 g/cm3. The
product is shown in Fig. 7.
696
Czech Republic
2006
[Content]
3,5
Volume [%]
2,5
1,5
0,5
0
5,7
137
270
403
535
668
801
Fig 7.
Crystals of FOX-7 obtained by crystallization from DMSO (example 1)
Example 2
This test was carried out in the same way like previous (example 1). The difference
between these two examples was stirring rate. In this case it was about 400 rpm. The
obtained product is presented in Fig. 8. Tap bulk density 1.0 g/cm3.
4
3,5
Volume [%]
2,5
1,5
0,5
0
5,7
137
270
403
535
668
801
Fig 8.
Example 3
FOX-7 was added to a flask with DMSO. The mixture was heated with mechanical
stirring 850 rpm to 90 C. Then acetone was added to the mixture The product was filtered
and washed. Tap bulk density of FOX-7 was 1.04 g/cm3. The crystals are presented in Fig. 9.
9
8
7
Volume [%]
6
5
4
3
2
1
0
5,7
137
270
403
535
668
801
Fig 9.
5
697
Czech Republic
2006
[Content]
Crystallization from DMSO/H2O

FOX-7 was added to a flask with DMSO/H2O 4/1. The mixture was heated with
mechanical stirring 900 rpm to 90 C. Next acetone was added to the mixture. The product
was filtered and washed with water and acetone. The FOX-7 was tap bulk density 0.93
g/cm3. The crystals are presented in Fig. 10.
9
8
7
Volume [%]
6
5
4
3
2
1
0
5,7
137
270
403
535
668
801
Fig 10. Crystals of FOX-7 obtained by crystalization from DMSO/H2O

We also researched thermal behavior of crystals which was obtained by crystallization
from DMSO/H2O (Fig.11). Differential thermal analyzer DTA 551 Ex was used with the
following experimental conditions:
-
sample weight 2 mg,
heating rate - 5 C/min,
temperature range 20 350 C.
Analyses were carried out in open test tubes and thermocouple protected by a glass shield
was inserted into the sample. Data were evaluated using MEAVY 2.0.0.4 software of DTA
551 Ex analyser.
D TA
M ultigraf
36
Fox Pr 6 5 ( 5) 2 mg
34
32
30
28
26
Difference [C]
24
22
20
18
16
14
12
10
8
6
4
2
0
1 00
1 20
14 0
16 0
1 80
2 00
Te mp e r atur e [C]
2 20
2 40
260
2 80
Me a v y
Fig 11. Thermogram of FOX-7 obtained by crystalization from DMSO/H2O
698
Czech Republic
2006
[Content]
Table 1. Table 1. Result of experiments

Sample
Yield
[%]
Bulk density
[g/cm3]
Tap bulk density

[g/cm3]
FOX-7 (crude)
70 80
0.43
0.50
DMF; 5 oC/min; 900 rpm
50.7
0.91
1.02
DMF; 20 oC/min; 900 rpm
50.8
0.63
0.74
DMF; acetone; 800 rpm
72.7
0.90
0.99
DMSO; 5 oC/min; 900 rpm
40.0
0.96
1.10
DMSO; 5 oC/min; 400 rpm
40.0
1.00
1.06
DMSO; acetone; 850 rpm
58.0
1.04
1.11
DMSO; nitrobenzene; 800 rpm
84.0
0.86
0.94
DMSO/H2O; acetone; 900 rpm
48.0
0.93
1.04
NMP; acetone; 800 rpm
53.0
0.45
0.51
In order to obtain suitable form of FOX-7 we checked some solvents. The best results
we obtained for mixture of DMSO and water. The parameters of crystallization, obtained
bulk density, and yield are presented in table 1. After these tests, coarse FOX-7 (bulk density
~1 g/cm3) was obtained. The crystals and characteristics of grain size of FOX-7 are shown in
Fig. 3-10. The best yield we obtained for FOX-7 precipitated with mixture DMSO and other
solvents like nitrobenzene.
3.
SUMMARY
Crystals of FOX-7 with good aspect ratio of 1,1-diamine-2,2-dinitroethylene were

obtained.
The optimal parameters of crystallization process for different solvents were
determinated.
The best results with respect to particle size and bulk density of the recrystallized
product were obtained for DMSO/H2O mixture as solvent for crystallization.
Our experiments showed that the size and form of crystals could be regulated by
conditions of process. The monocrystals of FOX-7 were 50 800 m, and tap bulk density
is about 1 g/cm3.
The highest yield of crystallization process was 84%. This result was obtained by
precipitation of FOX-7 with nitrobenzene from DMF.
699
Czech Republic
2006
[Content]
Acknowledgments
This research was supported by the State Committee of Scientific Research through
Institute of Industrial Organic Chemistry, grant 3 T09B 021 27.
References
[1]
[2]
J.P. AGRAWAL, Recent Trends In High-Energy Materials, Prog. Energy Combust. Sci., 1998.
A. BECUWE, A. DELCLOS, Low-Sensitivity Explosive Compounds for Low Vulnerability
Warheads, Propellants, Explosives, Pyrotechnic, 18, 1-10, 1993.
[3] Y. TANG, The Effect of NQ and TATB to the Insensitivity of HMX, Proc. of 33nd International
Annual Conference of ICT, Karsruhe, Germany 2002
[4] Y.N. MATYUSHIN, G.T. AFANASEV, V.P. LEBEDEV, M.N. MAHOV, V.I. PEPEKIN,
TATB and FOX-7: Thermochemistry, Performance, Detonability, Sensitivity, 34th International
Annual Conference of ICT, Karlsruhe, Germany, 24-27. 06.2003.
[5] N. Latypov, J. Bergman, A. Langlet, U. Wellmar, U. Bemm, Synthesis and reactions of , 1,1diamino-2,2-dinitroetylene, Tetrahedron 54, pp. 11525-11536, 1998.
[6] N. LATYPOV, A. LANGLET, U. WELLMAR, Chemical Compound Suitable for Use as an
Explosive, Intermediate and Method for Preparing the Compound, US Patent 6312538 B1,
Nov. 6, 2001
[7] H. OESTMARK, H. BERGMAN, U. BEMM, P. GOEDE, E. HOLMGREN, M.
JOHANSSON, A LANGLET, N.V. LATYPOV, A. PETTERSSON, M-L PETTERSSON, N.
WINGBORG, C. Vorde, H. Stenmark, L. Karisson, 2,2-Dinitroetylene-1,1-diamine (FOX-7)Properties, Analisis and Scale-up, Proc. 32nd International Annual Conference of ICT,
Karlsruhe, Germany,3-6. 06.2001.
[8] Z. CHYEK, S. CUDZIO, Jan Bdek, Sylwia Pietrzyk, Optymalization of 1,1-diamino-2,2dinitroetylene Synthesis, New Trends in Research of Energetic Materials, Pardubice, Czech
Republic, 207-212, 2005.
[9] U. TEIPEL, I. FUHR, K. HARTLIEB, A. KJELLSTROM, C. ELDSATER, Crystallization of
1,1-diamino-2,2-dinitroethylene (FOX-7), 35th International Annual Conference of ICT,
Karlsruhe, Germany, 2004.
[10] U. TEIPEL, Energetic Materials. Particle Processing and Characterization, Weinheim, 2005,
WILEY-VCH Verlag GmbH & Co. KGaA
[11] H. OESTMARK, H. BERGMAN, A. LANGLET, N. WINGBORG, U. WELLMAR, U.
BEMM, FOX-7 A New Explosives with Low Sensitivity and High Performance, Proc. 11th
Symposium (international) on Detonation, Snowmass Village, 1998, 18-23.
700
Czech Republic
2006
[Content]
SELECTIVE MODIFIERS OF THERMAL DECOMPOSITION AND

COMBUSTION OF COMPOSITE SOLID ROCKET PROPELLANTS
CONTAINING HMX
Popenko E. *, Sergienko A.*, Sokolov E.*, Petrov D.*
Shandakov V.**, Zharinov Y.**, Lobanova A.**
* Biysk Technological Institute of the Altay State Technical University, Biysk, Russia
**FGUP FR&PC "Altai", Biysk, Russia
Abstract:
The reference direction of the present researches was the search of structures of the
selective catalytic additives labilizing pre-flame processes of HMX transformation. At
choice of frame of the additives outwent from HMX ability to derivate molecular
[1]
complexes with a number of chemical compounds . The investigated salts and nitro-
nitroso-compounds essentially reduce tbid of HMX (on 30-80 0C). The interaction
of the investigated additives with HMX implements also in thermal
decomposition process of fuel compositions.
All researched catalytic additives are modifying agents of burning of fuel
compositions "PU+HMX". The interrelation between activity of the additive in
the thermal decomposition process of HMX and influencing this additive on
burning rate and value of -index of fuel compositions with HMX is observed.
The most effective activators of thermal decomposition of HMX and modifying
agents of burning of HMX based propellants are Cu, K-EDNA, K-NAPN and
4-NA.
Keywords:
1.
HMX, thermal decomposition, additives, burning
INTRODUCTION
Cyclic nitramines are perspective components of rocket propellants. They have high
energy characteristics. At addition of these compounds to propellant the combustion
temperature and single impulse is considerably increased. Typical representative of
nitramines class is HMX, which widely applied in rocket propellants of different types. But
the application of formulas with nitramines is complicated by high dependence of burning
rate of such propellants on pressure that upsets jet engine work stability and reduces
efficiency of rocket.
Is established, that HMX behaves in K-phase (condensate phase) of combustion as the
inert component, it decreases burning surface temperature and development of heat in
reactionary layer of K-phase. Disadvantage of HMX is as major ignition delay time [2]. The
particles of HMX have not time to ignite in K-phase, are carried out in gas phase, where
burn down with major development of heat. The entrance of heat from the flame zone does
not compensate energy consumption conditioned by transformation of HMX in K-phase. In
outcome the burning rate of propellant is reduced and the -index in the burning rate law is
E. Popenko, A. Sergienko, E. Sokolov, D. Petrov, V. Shandakov, . . .
701
Czech Republic
2006
[Content]
increased.With relation to combustion model of composite solid rocket propellant (CSRP)

with basic reaction in a K-phase and with heat entrance from aerosol zone it is possible to
achieve lowering of dependence of burning rate on pressure by enlarging the contribution of
K-phase processes owing to intensification of exothermic transformation of components in
reactionary layer [3, 4, 5, 6].
Effective way of control of thermal decomposition and combustion processes is usage of
the catalytic additives. But the problem is that the known catalytic agents of CSRP
combustion practically are not effective in propellants containing HMX. In the literature the
process of a chemosorption to HMX of compounds having in the frame amine group with a
different degree of substitution of hydrogen is described. These compounds accelerate
decomposition of HMX. Their effect appears already during a propellant thermostatting at
the solidification stage. It eliminates possibility their application in CSRP. Apparently, as
HMX decomposition catalysts it is expedient to use such compounds, in which one amine
groups will be derivated only at decomposition, at temperatures superior temperature of
propellants storage and processing.
The reference direction of the present researches was the search of stuctures of the
selective catalytic additives labilizing pre-flame processes of HMX transformation.
2.
EXPERIMENTS
2.1
Materials and Sample Formulation
Nitramine salts and nitro-nitroso-compounds, as accelerators of HMX decomposition

was selected. Structure of accelerators represented in the Table 1 and 2.
Table 1. Structure of nitramine salts.
Title of compound
Copper-potassium salt of
ethylendinitamine
(Cu, -EDNA)
Copper salt
of methylnitrouretane
(Cu-MNU)
Copper-ammonia salt
of methylnitrouretane
(Cu,NH3-MNU)
Nitraminopropyonitrile
(NAPN)
Potassium salt of
nitraminopropyonitrile (NAPN)
Copper salt of
nitraminopropyonitrile
(Cu-NAPN)
702
Formula
Melting
point tm,C
Decomposition
point td,C
K2[Cu(CH2NNO2)2]H2O
225
u(H2O)[O2NNC(O)OCH3]2
100
u(NH3)[O2NNC(O)OCH3]2
200
NCCH2CH2NHNO2
49-50
190
NCCH2CH2NNO2K
51-54
180
[NCCH2CH2NNO2]2Cu
160
Selective Modifiers of Thermal Decomposition and Combustion of Composite . . .
Czech Republic
2006
[Content]
Table 2. Structure of nitro- nitroso-compounds.

Title
Formula
NO
1-Nitroso-3,5,7-trinitro-1,3,5,7tetraazacyclooctane (1-NO)
N
O2N N
N NO2
N
NO2
NO
1,5-Dinitroso-3,7-dinitro-1,3,5,7tetraazacyclooctane (2- NO)
N
O2N N
N NO2
N
NO
1,4,5,9-Tetranitroso-1,4,5,9-tetraazabicyclo
[4,4,0]decaline (4-NA)
NO
NO
NO
NO
In work was used high-dispersive HMX with a specific surface area equal 256 m2/g. The
size of 1-NO and 2-NO particles has compounded 100 to 200 microns, 4-NA - 4 microns.
For manufacturing model fuel compositions fissile was selected binding presenting by itself
polyurethane caoutchouck plasticized by blasting oil (PU). Model fuel contained 60 % mass.
crystalline explosives with a relation HMX: nitrosamines equal 3:1. The content of the
catalytic agent did not exceed 3 % mass. from mass of a sample. As a starting composition
the intermixture containing 60 % of HMX and 40 % of PU was used.
2.2
Thermal Decomposition Process Measurements
The thermal decomposition of individual explosives, their intermixtures and model fuel
compositions on their basis was explored on derivatograph Paulik - Paulik - Erdey,
atmosphere - air, heat rate - 10 0C / min., interval of temperatures - 20-6000C.
2.3
Burning Rate Measurements
The performances of burning of fuel compositions were determined with a chimney type
strand burner (CTSB) which was pressurized with nitrogen at pressure range 2,0 to 10,0
MPa. Not hardened samples placed in fluorine-plastic glasses were investigated. During
operation pressure in the device and intensity of a flame glow were automatically fixed. The
burning time was determined on inflections on the curve "pressure - time". Burning rate
depending on the pressures were approximated in logarithmic coordinates by the equation
U=. The statistical processing of results was conducted under the reference computer
programs. The precision of definition of burning rate were about 3 %, -exponent were 5 %
703
Czech Republic
2006
[Content]
at confidence probability 0,97. The ignition of each sample was conducted by an electrically
heated nichrom wire which was attached on the top of sample.
3.

3.1
Thermal Decomposition Characteristics
From surveyed salts of nitrosamines the greatest decrease of temperature of HMX

intensive decomposition beginning excite copper-potassium salt of ethylendinitramine (u,
K - EDNA) and potassium salt of nitraminopropionitrile (K - NAPN). In a case u, K EDNA HMX is decomposed completely at temperature 240 0, that on 30 0 (Table 3) is
lower, as contrasted to by decomposition point of HMX without the additives.
Table 3. Parameters of thermal decomposition of HMX with nitrosamine salts.
1
2
3
4
5
6
7
Composition
ts,
HMX
270
HMX+Cu,K-EDNA
240
HMX+Cu-MNU
265
HMX+Cu,NH3-MNU 265
HMX+NAPN
150
HMX+K-NAPN
190
HMX+Cu-NAPN
260
1 stage
tf,
350
280
280
320
210
210
300
, %
99
87
39
99
6
27
93
ts,
300
380
380
260
260
380
2 stage
tf, , %
540
89
430 100
530 100
310
94
280
81
480
95
ts,
340
3 stage
tf, , %
380
90
The note:
ts start temperature of stage;
tf final temperature of stage;
transmutation degree, determined on the equation:
= (ms-mi)/ ms100 %
where: ms start mass of a sample, g.
mi mass of the heels in the end of i-th stage of decomposition, g.
At presence K-NAPN during decomposition of HMX it is possible to bleed two stages
(figure 2):
low-temperature (190 to 210 0C);
high-temperature (260 to 280 0C).
Set, that the degree of HMX transmutation at low-temperature stage is proportional to

the content of the additive. At magnification of K-NAPN content up to 9 %, at lowtemperature stage 50 % of HMX (table 4) is decomposed.
704
Czech Republic
2006
[Content]
W, %
100
80
exo
endo
60
40
20
0
0
100
200
300
400
500
o
600 t, C
losses of mass (TG);

heat effect (DTA);
Fig 1.
Table 4.
Derivatogram of decomposition of HMX with K-NAPN addition

Dependence of a HMX transmutation degree on the content of K-NAPN in
an interval of temperatures 190 to 210 0
K-NAPN content, % mass.

Transmutation degree, % mass.
1
8
2
18
5
35
7
38
9
50
15
51
The intermixture of PU with HMX is decomposed in 3 stages which are flowing past in
temperature bands 160-2000, 260-2700 and 360-5000. All stages are exothermic (figure
2).
W, %
100
80
exo
endo
60
40
20
0
0
100
200
300
400
500
600 t,

heat effect (DTA);
Fig 2.
Derivatogram of decomposition of HMX and PU blend
705
Czech Republic
2006
[Content]
The comparative analysis of derivatograms of blasting oil, PU and HMX decomposition

allows to correlate 1 stage with nitroether decomposition, second - with HMX
decomposition. The third stage, apparently, introduces by itself thermal-oxidizing
destruction of intermediate yields of caoutchouck pyrolisys.
The activity of the explored additives is maintained also at decomposition intermixtures
of HMX with polyurethane binding, plasticized by blasting oil (PU). All explored linkings
displace peak of decompositions of HMX with binding in lower temperatures area (Table 5).
Table 5. Parameters of thermal decomposition of intermixtures PU with HMX at

the presence of nitramines salts
1
2
3
4
5
6
7
8
Composition
PU
PU+HMX
PU+HMX+Cu,-EDNA
PU+HMX+Cu-MNU
PU+HMX+Cu,NH3-MNU
PU+HMX+NAPN
PU+HMX+K-NAPN
PU+HMX+Cu-NAPN
t s,
170
160
150
165
160
140
170
180
1 stage
tf ,
260
230
215
230
190
210
210
210
,
%
80
29
37
32
30
31
32
26
t s,
300
260
230
240
220
250
220
240
2 stage
tf ,
420
320
275
280
260
300
290
290
,
%
92
91
91
90
90
91
99
89
t s,
320
330
390
370
390
360
3 stage
tf ,
,
%
510 100
510 98
510 100
520 96
510 100
510 93
On the basis of results of the gas-chromatographic analysis of light-end products and

analysis of IR-spectrums of the condensed heels the possible mechanism of influencing of
the additives surveyed. Its basis is the nucleophilic interaction of the additives decay yields
with electron - deficient centres in HMX.
Nucleophilic particles are anions formed at decomposition of salts having a lone
electronic pair on amines nitrogen. The mechanism of salts transmutation we shall consider
on an example of nitraminopropionitrile derivatives.
(1)
(2)
(3)
2Me+NO2 Me2O + NO + NO2
706
(4)
Czech Republic
2006
[Content]
As in explored temperature range the potassium nitrite is inconvertible and is solid, the
reaction (4) is not implemented at decomposition of potassium salt of NAPNA, and the
fractional oxidizing of propionitrilimine implements under the plan:
NCCH2CH=NH+NO2N2+N2O+CO+CO2+H2O.
(5)
In HMX always there is a reactionary centre with a deficit of electrons and on it there
can be a nucleophilic attack by anion, that results in acceleration of HMX decomposition.
In case of K-NAPN and Cu-NAPN salts, explored in our work, nucleophilic particles are
propionitrilimine and its oligomer having a lone electronic pair on amines nitrogen.
On the basis of results obtained by us, it is possible to suspect that the reason of
activation of HMX decomposition in a case K-NAPN is the chemical interaction between
HMX and yields of decay of the additive, that confirms by dependence of HMX lowtemperature transmutation degree on K-NAPN content. Thus, at use K-NAPN, the
considerable importance in activation of HMX have the ionic processes.
The additives from a class of nitro-nitrosamines have exhibited high activity in thermal
decomposition and burning processes of HMX contained propellants.The results of research
have shown, that unlike HMX, for which one temperature of beginning of intensive
decomposition (tbid) is 270 0C, nitrosamines are decomposed at lower temperatures: to 1NHMX - at 230 0C, 2-NHMX 220 0C, 4-NA - at 200 0C. Thus, with increment of amount of
nitroso-groups, tbid of nitrosamine is reduced. The intermixtures of HMX with nitrosamines
are decomposed by one peak in temperature range, typical for decomposition of nitrosamine.
The degree of transmutation thus makes 92 % for intermixtures of HMX with its nitrosoderivatives and 74 % for an intermixture of HMX with 4-NA (Table 6).
Table 6. Parameters of thermal decomposition of nitro-nitrosamines and
propellants on their basis
Composition
HMX
1-NO
HMX +1-NO
PU
PU+ HMX
PU+ HMX +1- NO
2-NO
HMX +2-NO
PU+ HMX +2- NO
4-
HMX +4-
PU+ HMX +4-
t s,
270
180
240
170
160
160
160
220
150
200
190
180
1 stage
tf ,
350
210
320
260
230
200
210
340
200
270
270
205
, %
99
6
92
80
29
33
3
93
31
74
74
63
t s,
230
320
300
260
240
220
340
220
370
510
210
2 stage
tf ,
310
520
420
320
320
330
510
320
560
560
340
, %
92
94
92
91
84
90
95
91
97
89
85
t s,
320
330
320
440
3 stage
tf ,
590
470
535
620
, %
100
96
98
98
707
Czech Republic
2006
[Content]
Hence, the yield of decay of nitrosamines, interacting with HMX, produces its
destruction at lower temperatures. The number of fissile centres generated during
decomposition of nitrosamines, apparently, proportional to amount of nitroso-groups.
Therefore, it is possible to suspect, that the fissile centres are formed during break of N-Nconnections in nitrosamines.
The interaction of nitrosamines with HMX occurs also in thermal decomposition process
of fuel compositions. In fuel compositions the intermixtures of HMX with nitrosamines are
decomposed by one peak in temperature range, typical for decomposition of respective
nitrosamine (Fig.3).
W, %
100
exo
endo
80
60
40
20
0
0
100
200
300
400
500
600
t, oC

heat effect (DTA);
Fig 3.
Derivatogram of decomposition of PU+HMX+4-NA blend
Thus, hence the analysis of the obtained experimental data it is possible to make
following deductions:
-the investigated nitrosamines reduce tbid of HMX owing to interaction of yields of their
decay with HMX;
-the value of a decrease tbid of HMX is proportional to amount of nitro-groups in a
molecule of the additive;
-the interaction of yields of decay of nitrosamines with HMX implements both at
decomposition of intermixtures of explosives, and in thermal decomposition process of
model fuel compositions.
708
3.2
Czech Republic
2006
[Content]
Burning Rate Characteristics
The investigated additives practically do not influence quantity of burning rate and index of clean binding. In intermixtures PU with HMX the catalytic activity displays Cu, KEDNA and K- NAPN. At their insertion the burning rate at 4,0 MPa is incremented in 1,2
times, -index is descended from 0,88 (original compositions) to 0,75 in case Cu, K-EDNA
and to 0,68 in case K-NAPN. Results of definition of burning rate and -exponent are listed
in Table 7.
Table 7. Basic performances of burning of fuel compositions with the additives of
nitramines salts.
Composition
PU
PU +HMX
PU +Cu,-EDNA
PU +HMX+Cu,K-EDNA
PU +Cu,NH3-MNU
PU +HMX+Cu,NH3-MNU
PU +Cu-NAPN
PU + HMX +Cu- NAPN
PU +Cu-MNU
PU + HMX +Cu-MNU
PU + NAPN
PU + HMX + NAPN
PU +- NAPN
PU + HMX +- NAPN
, mm/s
MPa
3,45
1,30
3,00
1,95
2,97
1,36
3,06
1,17
3,17
1,07
3,03
1,26
2,82
2,03
0,60
0,88
0,72
0,75
0,62
0,81
0,64
0,93
0,56
0,99
0,64
0,84
0,70
0,68
U, mm/s, at
P, MPa
4
10
7,9
13,7
4,4
10,0
8,2
15,8
5,5
10,9
7,0
12,4
4,2
8,8
7,4
13,4
4,3
10,0
6,8
11,4
4,2
10,4
7,3
13,1
4,1
8,8
7,5
14,3
5,2
9,7
On the basis of high catalytic activity K-NAPN and Cu, K-EDNA at thermal
decomposition of HMX and its intermixtures with PU, it was possible to expect more
essential influencing of these salts on the performances of propellants burning. But the
activity of the additives is reduced by effect of a number of the factors:
-short time of contact of HMX particles and additives;
-presence of layers of binding between particles;
-inadhesion binding to HMX;
Ability of HMX fractionally dissolve in PU results in formation on a surface of its
particles of liquid layer, that sharply reduces an possibility of adhesion interaction between
polymer and HMX.
The intensive gas-making during a burnup of PU results in an offset of HMX particles to
a gas phase. Therefore the residence time of HMX in reaction layer of K-phase and,
accordingly, time of its contact with yields of decomposition of the catalytic additive is
reduced. And so, the problem of activation of HMX transmutation in K-phase of burning
needs to be solved complexly, together with modification combustible - binding, in order to
ensure sufficient adhesion to filling agent.
9
709
Czech Republic
2006
[Content]
The results on definition of the performances of burning, introduced in the table 8, is

evidence, what the intermixtures of PU with nitroso-compounds have lower values of index as contrasted to by composition containing PU and HMX. There is apparent
dependence of values of -index on amount of nitroso-groups in a molecule of the additive:
minimum value of is obtained in case of insertion of 4-NA in the fuel. At replacement 15
% of HMX in fuel composition on nitrosamines takes place a decrease of dependence of
burning rate on pressure.
The insertion in an intermixture PU+HMX of the complex additive consisting from
nitrosamine, nitrosamine salt and soot provides maximal effect of a decrease of dependence
of burning rate on pressure. So, at use of the most effective activator of HMX decomposition
- potassium salt of nitraminopropionitrile in an intermixture with soot, decrement of makes
from 30 up to 50 % for the formulas with an intermixture of HMX and nitrosamines (Table
8).
Table 8. Basic performances of burning of propellants with nitro-nitrosamines
Composition
PU + HMX
PU +1-NO
PU + HMX+1-NO
PU + 2-NO
PU + HMX +2-NO
PU + HMX+1-NO+K-NAPN
PU + HMX+2-NO+K-NAPN
PU + HMX +2-NO +K-NAPN+C
PU + HMX +1- NO +K-NAPN+C
PU + HMX +4-NA
PU + HMX +4-NA+K-NAPN
4.
A,
mm/s MPa
1,30
1,51
1,74
2,53
1,63
1,69
1,91
2,40
2,37
2,38
2,58
0,88
0,75
0,81
0,50
0,70
0,74
0,71
0,57
0,60
0,49
0,45
U, mm/s at
P, MPa
4
10
4,4
10,0
4,3
8,5
4,5
9,5
5,1
8,1
4,3
8,2
4,7
9,3
5,1
9,8
5,3
9,0
5,5
9,5
4,7
7,3
5,2
7,1
CONCLUSIONS
1. The investigated salts of nitramines and nitro-nitroso-compounds essentially reduce

tbid of HMX (on 30-80 0C). The value of a tbid decrement is proportional to the content of the
additive and amount of nitroso-groups in nitroso-compound. The activation is conditioned
by interaction of yields of decay of the additives with HMX.
2. The interaction of the investigated additives with HMX implements also in thermal
decomposition process of fuel compositions. As a result, in fuel composition there is a
collateral decomposition of HMX and additives in low-temperature area (from 180 up to 200
0
C).
3. The salts of nitramines and nitroso-compounds are modifying agents of burning of
fuel compositions "PU+HMX". The interrelation between activity of the additive in the
thermal decomposition process of HMX and influencing this additive on burning rate and
value of -index of fuel compositions with HMX is observed. The most effective activators
of thermal decomposition of HMX and modifying agents of burning of HMX based
710
Czech Republic
2006
[Content]
propellants are Cu, K-EDNA, K-NAPN and 4-NA. The insertion of soot boosts catalytic
effect during burning. Combination in fuel HMX with nitramines and combined additive KNAPN+C allows to decrease -index on 30-50 % as contrasted to by HMX based propellant.
5.
REFERNCES
[1]
V.V Maltsev PH.D. tesis. Moscow: IHF AN USSR, 1962.
[2]
A.A. Zenin, V.M. Puchkov, S.V. Finjakov HMX combustion wave structures at a
different temperature and pressure//FGB Journal, Vol. 34, 2, p. 59-66, 1988.
[3]
V.A. Strunin, G.B. Manelis About stability of a stable process of burning

explosive, limit reaction in to condensed phase//FGV Journal, 7, p. 498-500,
1971.
[4]
P.F. Pohil About a leading phase of burning gun powders//FGB Journal, 5, p.

439-440, 1969.
[5]
C.E. Hermance Model of Composite Propellant Combustion Including Surface

Heterogeneity and Heat Generation // AIAA Journal, Vol. 4, No. 9, p.1629-1637.,
1966.
[6] A.A. Zenin About one model of reactionary bed of a condensed phase of ballistite
gun powder//tesis. Moscow: IHF AN USSR, Vol. 213, p. 1357-1360, 1973.
11
711
Czech Republic
2006
[Content]
CRYSTAL STRUCTURES OF SELECTED ENERGETIC MATERIALS

CALCULATED BY MOLECULAR SIMULATIONS
Miroslav Pospil*, Pavel Vvra**, Petr Kov%*
* Charles University Prague, Faculty of Mathematics and Physics,
Department of Chemical Physics and Optics, Ke Karlovu 3, 12116 Prague 2, Czech Republic
** University of Pardubice, Faculty of Chemical Technology,
Institute of Energetic Materials, Studentsk 95, 53210 Pardubice, Czech Republic
Abstract:
Molecular simulations methods were used for calculation and possible prediction of
crystal structures of energetic materials. Calculated results for selected energetic
material (RDX, TNA) were compared with real crystal structures. Mainly these items
were compared: arrangement of individual molecules, space group and energy
minimum for experimental and calculated models. Calculations were done in different
types of empirical force fields, in different modeling environment (Cerius2 modules
crystal builder, off setup, minimizer, crystal packer, Material Studio module builders,
discover, charge modification), under different conditions (1) with and without respect
to conformation of real crystal structure, (2) with the same space group for calculated
and experimental sample or with neglecting of space group symmetry, (3) with different
arrangement of individual molecules in the crystal structure etc. Crystal structures of
calculated models with minimum energy were compared with published experimental
crystal structures. Calculated and experimental crystal structures show significant
similarities with respect to arrangement of individual molecules and also with respect to
energy. This way of prediction of energetic materials show that on the base of general
information the crystal structure can be partially calculated by molecular simulations
method.
Keywords:
energetic materials, RDX, TNA, molecular simulations, classical
molecular dynamics simulations, hydrogen bond
1.
INTRODUCTION
This work is derived from a series of our previous articles describing and analyzing
energetic materials by molecular mechanics and classical molecular dynamics simulations
under different conditions [1]. The most important conclusion from these results for this paper
is that influence of non bond interactions is significant for structure parameters and crystal
packing [2-3]. Calculated non bond parameters include van der Waals, Coulomb and
Hydrogen bond terms. For example, hydrogen bond interactions play important role for
measurement of physical quantity sensitivity [4]. On the base of knowledge of
intermolecular properties the crystal structures of cyclotrimethylene-trinitramine, known as
RDX [5] and crystal structure of 1-amino-2,4,6-trinitrobenzene, known as TNA [6] were
calculated by molecular simulations methods in Cerius2 and Material studio modeling
environments [7]. Resulted minimized models were compared with experimental data. The
design of new and proposed energetic compounds by different computational characteristics
was described by P. Politzer et al [8]. Computational methods like molecular dynamics
712
Crystal Structures of Selected Energetic Materials Calculated by Molecular . . .
Czech Republic
2006
[Content]
simulation [9, 10], Monte Carlo methods [11], Ab initio study [12] are widely used to describe
decomposition processes of energetic materials. Attention was also focused for important
role of weak interaction in crystal engineering [13, 14]. Crystal structures of energetic materials
are described by intermolecular potentials which very well characterize the real structure [15,
16]
. Moreover, influence of edge dislocation during hydrostatic compression on the electronic
structure can be also calculated by the ab ititio calculation [17]. These several examples show
increasing influence of computational chemistry for description and characterization of
energetic materials and their properties and bring promises for design of new energetic
materials with desirable properties.
2.
MOLECULAR SIMULATIONS
The experimentally described crystal structures of RDX and TNA were built on the
base of crystallographic data [5, 6]. The total crystal energy of these experimental structures
was calculated in individual empirical force field (COMPASS [18, 19], PCFF [20, 21], CVFF [22],
UFF [23], Tripos [24], Dreiding[25]) to choose the most convenient force field for following
calculation. After the first total energy calculation of RDX and TNA, structures were
minimized and total crystal energy of minimized structures was established. RDX was tested
and calculated in Material Studio modeling environment in discover module. TNA was
calculated in Cerius2 in Crystal Packer module, which allows predicting crystal structure.
After force field selection follow these general steps: (i) the space group of real RDX
and TNA was changed to P1 and minimized. (ii) Individual molecules of RDX and TNA
were randomly arranged in the crystal cell with fixed parameters related to real crystal cell.
A series of model with different initial position of molecules were minimized and on the
base of minimum crystal energy were selected and compared with previously minimized real
crystal structure. Models with minimum energy were compared with minimized real
structures.
After first energy minimization the conformational search was done by classical
molecular dynamics under following conditions: quench type dynamics - the model is after
100 steps of dynamics minimized. These minimized models are saved into quench trajectory
file and dynamics continue from the point before minimization. Dynamics were performed
in NVT ensemble, the number of atoms, volume and temperature are constant. Temperature
was 300 K, dynamics time step was 1fs. The maximum length of calculation in ps was 80 ps.
Bonding tolerances were 1.150 and molecular dynamics were calculated with ideal
periodical crystal structure without any defects. Charges were calculated by Charge
equilibration method (Qeq) [26].
3.
Empirical force fields used in this calculation do not describe the real crystal structure
precisely, so we can see difference of total crystal energy of real sample and minimized
sample, see table 1. The lower difference was calculated for COMPASS force field.
COMPASS force field was used for subsequent calculation of RDX (see Fig. 1) in Material
Studio. In table 2 we can see comparison of force field for experimental and minimized
crystal structure of TNA (see Fig. 2). Dreiding force field was selected for subsequent
calculations although it is only moderately accurate for geometries, conformational energies,
intermolecular binding energies, and crystal packing.
2
Miroslav Pospil, Pavel Vvra, Petr Kov
713
Czech Republic
2006
Fig 1.
Crystal structure of RDX based on X ray diffraction data
Fig 2.
Crystal structure of TNA based on X ray diffraction data
[Content]
Table 1. Total crystal energy [kcal/mol] of experimental and minimized RDX

crystal structure
RDX
COMPASS
PCFF
CVFF
experimental
-748
-480
-327
minimized
-841
-803
-721
Table 2. Total crystal energy [kcal/mol] of experimental and minimized TNA

crystal structure
714
Czech Republic
2006
TNA
Dreiding
Universal
Tripos
experimental
-160
-151
-145
minimized
-167
-159
-186
[Content]
Crystal structure of RDX and TNA was changed to P1 space group. After minimization with
fix space groups parameters we obtain the same value of total crystal energy. Consequently
individual molecules of RDX and TNA were randomly placed in the space group with fixed
parameters and minimized. Results with minimum total energy for the best models are in the
table 3. Initial random arrangement of RDX and TNA molecules after minimization was
more organized than in initial phase. Calculated crystal structure of RDX - test1 is shown in
the Fig. 3 and for TNA - test1 in the Fig. 4.
Table 3. Total crystal energy [kcal/mol] of minimized RDX and TNA crystal
structures for series
test1
test2
test3
RDX
(COMPASS)
-804,65
-804,61
-803,01
TNA
(Dreiding)
-159,3
-140,0
-134,4
Fig 3.
Calculated crystal structure of RDX-test1
715
Fig 4.
4.
Czech Republic
2006
[Content]
Calculated crystal structure of TNA-test1
CONCLUSIONS
Molecular mechanics and classical dynamics simulations in this work were used for
showing several possible steps on the way of design new energetic materials on the base of
their general properties derived from non bond interactions. Calculated models can be
compared with real experimental structures in case, that this real crystal structures is
optimized in the same force field. The force field describing one crystal structure precisely is
not usually suitable for different type of energetic material. On the other hand force field
with very general parameters can describe a series of energetic material but with high error
in comparison with real crystal structure. The aim of this work was to find some consensus
among these two point of view for describing energetic materials by force field calculations.
REFERENCES
[1]
[2]
[3]
[4]
[5]
716
M. POSPIL, P. JAPKOV, P. VVRA AND S. ZEMAN: Characterization of Explosive Materials

Using Molecular Dynamics Simulations, Proc. of the 5th Seminar New Trends in Research of
Energetic Materials, University of Pardubice, Czech Republic, 5, p. 262-270, 2002.
P. VVRA, M. POSPIL, J. REPKOV: Effect of intermolecular forces on some properties of
explosives, Proceedings of the 5th Seminar New Trends in Research of Energetic Materials,
University of Pardubice, Czech Republic, 5, 357-368, 2002.
P. VVRA, M. POSPIL: Effect of Intermolecular Forces on Some Properties of Explosives II.
Influence of Non-covalent Hydrogen Bond, Proceedings of the 6th Seminar New Trends in
Research of Energetic Materials, University of Pardubice, Czech Republic, 6, 473-481, 2003.
M. POSPIL, P. VVRA: Study of Electron Density of Molecules, Intermolecular Forces and
Impact Sensitivity of Explosives, Proceedings of the 7th Seminar New Trends in Research of
Energetic Materials Editor J. Vgenknecht, University of Pardubice, 7, pp. 600-605, 2004.
C. S. CHOI AND E. PRINCE: The Crystal Structure of Cyclotrimethylene-Trinitramine, Acta
Cryst., B 28, p. 2857-2862, 1972.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
Czech Republic
2006
[Content]
J. R. HOLDEN, C. DICKINSON, C. M. BOCK: The Crystal Structure of 2,4,6-Trinitroaniline,

J. Phys. Chem., 76, p.3597-3602, 1972.
Cerius2 documentation, June 2000, San Diego: Molecular Simulations Inc, 2000.
www.accelrys.com.
P. POLITZER, J. S. MURRAY, J. M. SEMINARIO, P. LANE, M. E. GRICE AND M. C. CONCHA:
Computational Characterization of Energetic Materials, J. Mol. Struct. (Theochem), 573 p. 110, 2001.
Y. KOHNO, K. UEDA AND A. IMAMURA: Molecular Dynamics Simulations of Initial
Decomposition Process on the Unique N-N Bond in Nitramines in the Crystalline State, J.
Phys. Chem., 100, p. 4701-4712, 1996.
B. M. RICE, G. F. ADAMS, M. PAGE, D. L. THOMPSON: Classical Dynamics Simulations of
Unimolecular Decomposition of CH2NNO2: HONO Elimination vs N-N Bond Scission, J.
Phys. Chem., 99, p. 5016-5028, 1995.
D. V. SHALASHILIN AND D. L. THOMPSON: Monte Carlo Variational Transition-State Theory
Study of the Unimolecular Dissociations of RDX, J. Phys. Chem. A, 101, p. 961-966, 1997.
C. J. WU, L. E. FRIED: Ab Initio of RDX Decomposition Mechanisms, J. Phys. Chem. A., 101, p.
8675-8679, 1997.
J. M. A. ROBINSON, D. PHILP, K.D. M. HARRIS, B. M. KARIUKI: Weak interactions in crystal
engineering understanding the recognition properties of the nitro group, New J. Chem., 24,
p. 799-806, 2000.
R. S. ROWLAND, R. TAYLOR: Intermolecular Nonbonded Contact Distances in Organic Crystal
Structures: Comparison with Distances Expected from van der Waals Radii, J. Phys. Chem.,
100, p. 7384-7391, 1996.
D.C. SORESCU, B.M. RICE, D.L. THOMPSON: Intermolecular Potential for the Hexahydro-1,3,5trinitro-1,3,5-s-triazine Crystal (RDX): A Crystal Packing, Monte Carlo, and Molecular
Dynamics Study, J. Phys. Chem. B, 101, p. 798-808, 1997.
D.C. SORESCU, B.M. RICE, D.L. THOMPSON: Molecular Packing and Molecular Dynamics
Study of the Transferability of a Generalized Nitramine Intermolecular Potential to NonNitramine Crystals, J. Phys. Chem. A, 103, p. 989-998, 1999.
M. M. KUKLJA, A. B. KUNZ: Compression-induced effect on the electronic structure of
cyclotrimethylene trinitramine containing an edge dislocation, Journal of applied physics, 87,
p. 2215-2218, 2000.
H. SUN, D. RIGBY: Polysiloxanes: Ab Initio Force Field And Structural, Conformational And
Thermophysical Properties, Spectrochimica Acta (A), 53, p. 1301-1323, 1997.
D. RIGBY, H. SUN, B. E. EICHINGER: Computer Simulations of Poly(ethylene oxide): Force
Field, PVT Diagram and Cyclization Behaviour, Polymer International, 44, p. 311-330, 1997.
J. R. MAPLE, U. DINUR, A. T. HAGLER: Derivation of Force Fields for Molecular Mechanics
and Dynamics from ab initio Energy Surfaces, Proc. Natl. Acad. Sci., 85, p. 5350-5354, 1988.
J. R. MAPLE, M. J. HWANG, T. P. STOCKFISCH, U. DINUR, M. WALDMAN, C. S. EWIG, A. T.
HAGLER: Derivation Of Class II Force Fields .1. Methodology and Quantum Force Field for
the Alkyl Functional Group and Alkane Molecules, J. Comp. Chem., 15, p. 162-182, 1994.
P. DAUBER-OSGUTHORPE, V. A. ROBERTS, D. J. OSGUTHORPE, J. WOLFF, M. GENEST, A. T.
HAGLER: Structure and energetics of ligand binding to proteins: E. coli dihydrofolate
reductase-trimethoprim, a drug-receptor system, Proteins: Structure, Function and Genetics, 4,
p. 31-47, 1988.
A. K. RAPP, C. J. CASEWIT, K. S. COLWELL, W. A. GODDARD, W. M. SKIFF: UFF, a full
periodic table force field for molecular mechanics and molecular dynamics simulations, J.
Amer. Chem. Soc., 114, 10024-10035, 1992.
M. CLARK, R. D. CRAMER III, N. VAN OPDENBOSCH: Validation of the general purpose tripos
5.2 force field, J. Comp. Chem., 10, p. 982-1012, 1989.
717
Czech Republic
2006
[Content]
[25] S. L. MAYO, B. D. OLAFSON, W. A. GODDARD III: DREIDING: a generic force field for
molecular simulations, J. Phys. Chem., 94, p. 8897-8909, 1990.
[26] A. K. RAPP AND W. A. GODDARD III: Charge Equilibration for Molecular Dynamics
Simulations, J. Phys. Chem., 95, p. 3358-3363, 1991.
Acknowledgment:
M.P. and P.K. thanks to financial support from the Ministry of Education, Youth and
Sports of the Czech Republic (Project MSM0021620835).
P.V. thanks to financial support from the Ministry of Education, Youth and Sports of the
Czech Republic (Project MSM0021627501).
718
Czech Republic
2006
[Content]
COMPARISON OF COATING WITH VARIOUS DETERRENTS

Pavel Prchal
Research Institute of Industrial Chemistry, Explosia corp., 532 17 Pardubice-Semtn, Czech Republic
Abstract:
Several substances used in present as deterrents for coating of propellants were
compared in the sense of efficiency of utilisation, thickness and look of surface layer,
heat of explosion, gun chamber maximal pressure and bullet velocity.
Keywords:
1.
deterrent, coating, propellant
INTRODUCTION
At firing is generally required to obtain the highest bullet velocity and the lowest
maximal gun chamber pressure and temperature for the minimal abrasion and moderate
construction demands. The processes in gun barrel may be affected by several ways - one of
them is coating of propellant grain. This depends on the choice of suitable deterrent and on
the conditions and way of treatment (e. g. temperature, solvent system, time and mode of
deterrent adding). The aim of this contribution is to compare coating of propellants by
several substances - both substances used for this purpose for a longer time and substances
potentially applicable in future. For the possibility of simple comparing were the processes
made in the same way.
2.
EXPERIMENTAL
2.1
Sample preparation
The coating process was performed under stirring in water with ratio of uncoated grain
and water 1 : 4.2 at the temperature 80 C with adding of deterrent in six portions at time
intervals of 20 minutes with following 3-hour idle period [1].
The two series of samples were prepared from the two types of uncoated grain with
different grain size. For the possibility of comparing every set of samples includes one
standard sample coated with dibutylphthalate.
The set of samples 10, 11, 12 and 14/05 was prepared from uncoated grain with bigger
grain size, as deterrent were used DBP (dibutylphthalate), iNBP (isononylbenzylphthalate),
BuNENA (N-n-butyl-N-(2-nitroxy-ethyl)nitramine) [2,3] and DNDA57 (mixture of linear
nitramines 5, 6 and 7 - similar to DNDA7) [2,3] in amounts 5 % to uncoated grain weight.
The set of samples 21, 23, 24, 25, 26 and 27/05 and 02/06 was prepared from uncoated
grain with smaller grain size, as deterrent were used DBP, DOP (dioctylphthalate), DNT
(2,4-dinitrotoluene), DNDA7 (3,5-dinitro-3,5-diazaheptane) [2,3], CI (Centralite I), CII
(Centralite II) and DOTP (dioctylterephthalate) in amounts 6.7 % to uncoated grain weight.
Pavel Prchal
719
2.2
Czech Republic
2006
[Content]
Gun firing
All samples were fired in gun 5.56 NATO with bullet weight 4 g - every sample in five
shots. The chamber pressure was detected by piezo-sensor, bullet velocity was measured in
distance 9,5 m from muzzle.
3.
RESULTS
The deterrent amounts in samples were determined and efficiencies of deterrent

utilisation (i. e. ratio of amount of deterrent found in a sample and amount of deterrent used
in a batch) were calculated. The optical analysis of surface layer thicknesses, heats of
explosion measurements and shootings were performed. All results see in table 1 and
pictures 1-11.
Table 1. Analyses and shootings
sample
10/05-DBP
surface
layer
thickness
heat of
explosion
[%]
[
m]
[kJ.kg-1]
[MPa]
[m.s-1]
96
42
3750
188.6
723.6
gun
chamber
bullet
velocity
pressure
11/05-iNBP
4330
429.2
900.2
12/05-BuNENA
93
4220
370.5
865.6
14/05-DNDA57
85
4305
413.0
900.0
21/05-DBP
92
54
3928
247.3
769.7
23/05-DOP
4543
434.6
867.9
24/05-DNT
97
209
4368
412.2
849.0
25/05-DNDA7
95
214
4464
408.1
848.5
26/05-CI
97
44
3809
235.8
759.0
27/05-CII
89
199
3741
294.4
761.5
02/06-DOTP
17
4349
432.4
855.2
Fig 1. 10/05-DBP
720
efficiency of
utilisation
Fig 2. 11/05-iNBP
Czech Republic
2006
Fig 3. 12/05-BuNENA
Fig 4. 14/05-DNDA57
Fig 5. 21/05-DBP
Fig 6. 23/05-DOP
Fig 7. 24/05-DNT
Fig 8. 25/05-DNDA7
Fig 9. 26/05-CI
Fig 10. 27/05-CII
Pavel Prchal
[Content]
721
Czech Republic
2006
[Content]
Fig 11. 02/06-DOTP
4.
CONCLUSIONS
Classical deterrents (DBP, DNT, CI a CII) show good efficiency of utilisation between
90 and 100 %, surface layer thickness for CI is similar to standard with DBP, in the case of
DNT and CII was selected time of deterrent coating too long (diffusion practically through
the whole grain).
New compounds (BuNENA, DNDA7 a DNDA57) show good efficiency of utilisation
(about 90 %), in the case of surface layer thickness probably the diffusion through the whole
grain occurred - for BuNENA and DNDA57 is surface layer not visible, for DNDA7 is
diffusion through the whole grain visible.
The other tested phthalates with heavier molecules (iNBP, DOP a DOTP) have no
visible surface layer, analyses of amounts of phthalates in samples showed efficiency of
utilisation nearby zero for iNBP and DOP and the low 17 % efficiency for DOTP, i. e. they
are not suitable for coating in wet diffusion regime.
The values of heat of explosion, chamber pressure and bullet velocity are corresponding
to deterrent amount in sample, to depth of deterrent diffusion into grain and to
energetic/nonenergetic character of molecule of individual deterrents.
REFERENCES
[1]
[2]
[3]
722
K. LEINWEBER: Research report 140-740/05, VPCH, Explosia Corp., 2005

J. PETRLEK: Research report 185-785/02, VPCH, Explosia Corp., 2002
J. PETRLEK: Research report 759/03, VPCH, Explosia Corp., 2003
Czech Republic
2006
[Content]
TREATMENT OF WASTEWATERS CONTAINING

NITRATE ESTERS
Jan Skldal*, Martin Kouba*, Martina Mudrukov*, and Miloslav Slezk**

*Research Institute of Industrial Chemistry, Explosia a.s.,
532 17 Pardubice - Semtn, Czech Republic
**Institute of Environment Protection, University of Pardubice,
532 10 Pardubice, Czech Republic
Abstract:
This paper describes the methods of analysing nitroglycerin (NG) - ethylene glycol
dinitrate (EGDN) mixture present in wastewaters. In practice, for application reasons
the mixture of NG and EGDN is produced by nitration of glycerine - ethylene glycol
mixture. These substances have excellent explosive properties, they are, however, toxic
for living organisms. Techniques of chemical wastewaters treatment are based on
various chemical or physical-chemical operations. The aim is to clean or effectively preclean these waters.
At present situation, the utilization of chemical reactions seems to be the most suitable
method of treatment of wastewaters from nitrate esters production. Three methods were
used for nitrate esters degradation - reduction of nitrocompounds with fine-grained
iron, continual and discontinual, and reduction with lime milk (calcium hydroxide water
solution). Nitrate esters were separated, their quantity was determined by means of
high-pressure liquid chromatography with ultraviolet detection (HPLC/UV) with
internal standard used.
Keywords:
1.
wastewater, nitrate esters
INTRODUCTION
Wastewaters mostly have to be treated directly in defined areas of the companies

producing explosives or in rooms, where explosives are handled, mostly confined. This
enables to introduce partial or total waters recycling, to reduce contamination of
environment, or to reduce residual pollution in outlets of biological wastewater treatment
plants. Important is also a safe in-process handling with wastewaters, often saturated with
dissolved explosives. Under temperature variation the explosives can accumulate in channel
system, and act as permanent source of risk of fire or explosion.
Considering their pure aerobic biological degradability, the wastewaters from nitrate
esters production are difficult to treat biologically, and, in addition, they can cause
intoxication of insufficiently adapted activated sludge. Therefore, the biological treatment of
such waters is not sufficient enough and requires at least completing of common aerobic
treatment plants by anaerobic step or source of external carbon. These findings and
experience indicate the necessity to realize the units for pre-treatment of wastewaters prior to
biological treatment directly at the sources of wastewaters formation. Economic factor has to
be taken into consideration as well. This is the reason why there is a big progress in
Jan Skldal, Martin Kouba, Martina Mudrukov, Miloslav Slezk
723
Czech Republic
2006
[Content]
development of unconventional treatment methods for these types of wastewaters. This

paper brings description of chemical methods[1].
2.
EXPERIMENTAL
2.1
Chemical methods of degradation
2.1.1. Discontinual reduction with iron filings

Discontinual reduction with iron filings was used for degradation of nitrate esters present
in wastewaters from NG and EGDN production. Wastewater A with NG content 0.510 g/l
and EGDN content 3.810 g/l was used for experiments.
150 ml or 300 ml wastewater and 7.5 or 15 g iron filings were placed into reactor. After
15 minutes (T15) 25 ml of this way processed water was taken, defined quantity of internal
standard was added and made up with methanol to 50 ml. After dissolution of internal
standard, NG and EGDN contents were determined by means of HPLC/UV. In the same
way determinations of nitrate esters were carried out in defined time intervals T30 T165.
2.1.2. Continual reduction with iron filings
The possibility was experimentally tested of reduction of nitrate esters, present in
wastewaters from NG and EGDN production, with iron filings in continual regime in flowthrough column. For experiments wastewater B was used with NG content 0.310 g/l and
EGDN content 3.289 g/l. Glass column of internal diameter 32.5 mm with fritted glass and
glass cock on its bottom was used for experiments in continual regime. The column was
filled with iron filings in such a way that the filings were deposited evenly up to the height
of 145 mm.
After 5 minutes circulation (T5) 25 ml of this way processed wastewater was taken,
defined quantity of internal standard was added and made up with methanol to 50 ml. After
dissolution of internal standard, NG and EGDN contents were determined by means of
HPLC/UV. In the same way determinations of nitrate esters were carried out in defined time
intervals T33 T160.
2.1.3. Discontinual reduction with lime milk (calcium hydroxide water
solution)
The method of discontinual reduction with lime milk was used for degradation of nitrate
esters, present in wastewaters from NG and EGDN production. For the experiments 500 ml
of saturated water solution of NG/EGDN mixture of NG content 0.635 g/l and EGDN
content 3.551 g/l and 500 ml of lime milk were used.
The solution was vigorously shaken. After 15 minutes (T15) 100 ml of this mixture was
taken, transferred into separating funnel and shaken for three times always with 60 ml ether.
The ether extracts collected were washed with water saturated with sodium chloride and
dried with anhydrous sodium sulphate. Ether was evaporated and the extract obtained was
after desiccation over sulphuric acid in desiccator and weighing analyzed by means of
HPLC/UV method. In the same way determinations of nitrate esters were carried out in
defined time intervals T30 T10080.
2.2
HPLC/UV method of determination of NG and EGDN content
The content of nitrate esters was quantified using a high-performance liquid

chromatography (HPLC) with a diode array detector (DAD), the column used was Separon
SGX C18 7 m, 4 x 250 mm. For the analysis of nitrate esters the isocratic mobile phase was
724
Czech Republic
2006
[Content]
used, containing 60.5 % methanol and 39.5 % redistilled water, at flow rate of 0.6 ml/min
and absorbance was observed at 230 nm. Internal standard method was used[2].
2.3
Basic indicators of water pollution
2.3.1. Chemical Oxygen Demand CODCr - determination by oxidation with

K2Cr2O7
The method is based on oxidation of organic compounds contained in a wastewater by
K2Cr2O7 at two-hours boiling in strongly acid environment (H2SO4). Oxidation is catalyzed
with silver ions and proceeds in excess K2Cr2O7. To prevent oxidation of chlorides to Cl2,
which would cause a positive error at given determination conditions, HgSO4 is added.
The quantity of K2 Cr2O7 consumed for oxidation of organic compounds is measured by
means of titration with (NH4)2Fe(SO4)2.6H2O volumetric solution on feroin indicator.
Organic compounds are oxidized according to the following reaction:
Cr2O72- + 6 e- + 14 H+
2Cr3+ +
7 H2O
When the excess dichromate is determined using (NH4)2Fe(SO4)2.6H2O solution,

the following reaction takes place:
Cr2O72- + 6 Fe2+ + 14H+
2Cr3+ + 6Fe3+
+ 7 H2O
Maximum permitted CODCr value in the wastewaters entering the pre-treatment system
on wastewater treatment plant inlet is 2700 mg/l.
2.3.2. Biochemical Oxygen Demand BOD5 - Dilution and Inoculation Method
of Biochemical Oxygen Demand Determination with Nitrification
Suppression
(BOD5) - biochemical 5 days oxygen demand: mass concentration of dissolved oxygen,
consumed under defined conditions (5 days incubation at dark at 20C) by biochemical
oxidation of organic, possibly also inorganic compounds in water. As organic compounds
represent one of the major water pollutants, BOD5 is one of the key water pollution
indicators.
This method can be applied for all waters with BOD5 value ranging from 3 mg/l to
6000 mg/l. The results obtained are a combination of biochemical and chemical reactions.
This is why they do not represent as unambiguous and precise result as it is with some
simple and well-defined chemical process. Nevertheless it is important parameter, which
the water quality can be concluded from.
Principle of the method: BOD5 is determined as a difference between the concentrations
of dissolved oxygen in a wastewater sample prior and after biochemical oxidation.
Determined loss in dissolved oxygen (i.e. BOD5 value) is proportional to the content of
biochemically degradable compounds in water.
Procedure: The wastewater sample is diluted with various volumes of diluting water
with sufficient concentration of dissolved oxygen under addition of aerobic organisms
inoculum. Nitrification is inhibited by addition of allylthiourea. Incubation proceeds in dark,
in fully filled and closed vial at 20C. The quantity of dissolved oxygen is determined by
means of oxi-electrode prior and after incubation. The mass of oxygen consumed in 1 litre of
water is calculated.
725
Czech Republic
2006
[Content]
Maximum permitted CODCr / BOD5 ratio in wastewaters entering the pre-treatment

system on wastewater treatment plant inlet is 2.7.
3.

3.1
Discontinual reduction with iron filings discontinual process I
The possibilities were verified of discontinual reduction of nitrate esters by iron in

neutral medium in the samples of wastewaters from NG and EGDN production. In the tested
sample of wastewater (A, B) the values were determined shown in Table 1.
Table 1. Wastewaters A, B
Wastewater
A
B
NG [g/l]
0.510
0.310
EGDN [g/l]
3.810
3.289
CODCr [mg/l]
2175
1773
BOD5[mg/l]
--27
The experiments were carried out under various reduction times at ambient temperature
(Table 2). Agitation rate did not affect the values measured. The results are summarized in
Figure 1.
concentration [g/l]
4
3
NG
2
EGDN
1
0
0
50
100
150
200
tim e [m in]
Fig 1.
Residual NG and EGDN concentration /time dependence - discontinual

process I
Table 2. Discontinual wastewaters treatment I - Residual nitrate esters content

Time [min]
0
15
30
45
105
127
165
726
NG [g/l]
0.510
0.340
0.220
0.178
0.000
0.000
0.000
EGDN [g/l]
3.810
2.731
2.717
2.458
1.568
1.055
1.090
Czech Republic
2006
[Content]
Utilization of iron filings for nitrate esters degradation confirmed the previous finding,
that removing of NG from a wastewater requires reaction time 1.5 - 2 hours. NG is degraded
completely, while starting EGDN content decreases to 1/3 only within the same time.
3.2
Continual reduction with iron filings
Further, the possibility was tested of reduction of nitrate esters, present in wastewater B
from NG and EGDN production, with iron filings in continual regime in flow-through
column (Table 3). The experiments were carried out at ambient temperature. Again,
the decrease in nitrate esters content achieved depends on reaction time, i.e. on the time of
delay of the water in the column (Figure 2).
concentration [g/l]
4
3
NG
2
EGDN
1
0
0
50
100
150
200
tim e [m in]
Fig 2.
Residual NG and EGDN concentration / time dependence - continual

process
Table 3. Continual wastewaters treatment - Residual nitrate esters content

Time [min]
0
5
33
160
NG [g/l]
0.310
0.262
0.186
0.000
EGDN [g/l]
3.289
3.287
2.569
0.221
The best result so far was achieved in decrease in EGDN content to 7 % of its starting
content, with NG being completely removed.
As to operation, this arrangement of the future production equipment would be less
demanding than discontinual process.
3.3
Discontinual reduction with lime milk (calcium hydroxide water

solution) - discontinual process II
The possibilities were tested of discontinual reduction of nitrate esters in the sample of
model water using lime milk. NG content in this model water was 0.635 g/l and EGDN
content was 3.551 g/l.
727
Czech Republic
2006
[Content]
Agitation time was changed in experiments, reaction proceeded at ambient temperature

(Table 4). Survey of the results achieved is shown in Figure 3.
concentration [g/l]
4
3
NG
EGDN
1
0
0
100
200
300
tim e [m in]
Fig 3.
Residual NG and EGDN concentration / time dependence - discontinual

process II
Table 4. Discontinual wastewaters treatment II - Residual nitrate esters content

Time [min]
0
15
30
60
120
240
1440
2880
10080
NG [g/l]
0.635
0.443
0.313
0.124
0.026
0.002
0.003
0.000
0.000
EGDN [g/l]
3.551
2.980
2.874
2.870
2.686
2.658
2.603
2.607
2.605
Utilization of this method for nitrate esters degradation confirmed the previous finding,
that removing of NG from a wastewater requires reaction time 2 hours. NG is degraded
completely, while starting EGDN content decreases to 2/3 only within the same time and its
degradation rate depends on the presence of NG. After degradation of NG, EGDN content
remains unchanged even after 7 days (10080 minutes). This method is demanding for
preparation of water samples prior to analysis as well.
4.
CONCLUSION
The procedures described represent new possibilities of treatment of nitrate esters

containing wastewaters. The degradation products formed are easy to degrade biologically.
The best method of NG and EGDN degradation in wastewater seems to be continual
reduction with iron filings. When discontinual methods are used, NG degradation is
complete, but EGDN degradation is only partial.
728
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
J. YINON, S. ZITRIN: Modern Methods and Applications in Analysis of Explosives, John Wiley
& Sons, Chichester 1993
J. SKLDAL, V. JEOV, A. EISNER, M. KOUBA, K. VENTURA: Determination of Nitrate Esters
Content in Water Samples, Proc. of the 8th Seminar New Trends in Research of Energetic
Materials, Univ. of Pardubice, April 2005
729
Czech Republic
2006
[Content]
THE USE OF LOW PRESSURE CLOSED VESSEL AND ROCKET

MOTOR FOR MEASUREMENTS OF BURNING RATE OF ROCKET
SOLID PROPELLANTS
Petr Stojan
Explosia a.s., Research Institute for Industrial Chemistry,
CZ-532 17, Pardubice - Semtn, Czech Republic
Abstract:
This paper presents the results of testing of burning process of rocket solid propellants
when low pressure closed vessel or rocket motor are used for measurements.
The dependence is measured of burning rate of solid propellant on gas pressure in
combustion chamber.
Keywords:
1.
solid propellant, closed vessel, rocket motor, burning rate.
INTRODUCTION
Stable burning of the solid propellant used is an essential condition of proper function of
a rocket motor. The most important parameter indicating the stability of burning of solid
propellant is the dependence of burning rate on gas pressure in combustion chamber. This
paper presents the results of testing of burning process of rocket solid propellants when low
pressure closed vessel (LPCV) or rocket motor (RM) are used for measurements.
2.
THEORETICAL
The values of gas pressure in combustion chamber, burning temperature and gas flow
rate are important parameters to express the function of solid propellant rocket motor. These
values define the rate of burning, which can be expressed as follows [1,2]
u = u0 .g ( p ). f (TPH ).h( w) ,
(1)
where u0 is burning rate constant, g(p) is function of gas pressure in combustion

chamber, f(TPH) is function of initial propellant temperature TPH and h(w) define the effect of
gas flow rate w. As u0 and g(p) are the parameters of primary interest Eq.(1) is rewritten to
the form:
u ( p ) = u 0 .g ( p ) ,
(2)
Determination of burning rate - gas pressure dependence u(p) is usually carried out using
Crawford bomb. It is based on measurement of burning time of solid propellant grain of
defined dimension at defined pressures. Certain disadvantage is the necessity to use
complicated equipment, special modified solid propellant grain and time consumption. For
730
The Use of Low Pressure Closed Vessel and Rocket Motor for Measurements of . . .
Czech Republic
2006
[Content]
these reasons other methods were studied. As another possibility proved to be utilization of
LPCV or RM.
In this paper burning rate - gas pressure dependencies are compared, measured with
i)
Low Pressure Closed Vessel (LPCV) or
ii)
Rocket Motor (RM)
The main advantages of measuring with LPCV are: simple equipment, utilization of
standard solid propellant grains, time savings. The main advantages of measuring with
rocket motor are: complexity of measurements (i.e. possibility of measuring any parameter),
utilization of standard test procedures.
(i) Low Pressure Closed Vessel (LPCV)
Closed vessel (CV) is a standard apparatus for determination of gun propellant burning
rate. LPCV is a low-pressure variant (pmax << 100 MPa) of closed vessel, serving for
observation of rocket solid propellant. Evaluation of burning rate involves processing of
measured pressure-time curves.
Gas pressure in CV as well as in LPCV defined as function of proportionate quantity of
burned propellant 0.1 can be expressed as [4]
p( ) =
rTv
(W ( 1 ) )
(3)
where r is specific gas constant, Tv is burning temperature of the propellant, is propellant

weight, W is volume of closed vessel, is propellant density, is covolume of propellant
gas.
If the constant burning area is considered, then, then the burning rate can be expressed
from Eq. (3) in the form
u ( p) =
e0
dp
. ,
k B ( ) p max dt
(4)
where e0 is unit burning thickness of solid propellant grain, pmax is maximal pressure in CV
and coefficient kB() is equal to
k B ( ) =
(W )(W )
[W ( 1 ) ]2
(5)
As in LPCV /W0.03 kg/dm3 (see below) and therefore kB() 1, from Eq. (4) it can
be written
u ( p)
e0
dp
. ,
p max p z dt
(6)
where pz is ignition pressure.

Equation (6) is used for estimation of burning rate - gas pressure dependencies (see Eq. (2))
inside the interval pz - pmax.
Petr Stojan
731
Czech Republic
2006
[Content]
(ii) Rocket Motor

The course of pressure-time dependence in combustion chamber of rocket motor is one
of standard measurements.
Average burning rate u th for average pressure pth can be expressed as
u th ( pth ) =
e0
,
th t z
(7)
where e0 is unit burning thickness of propellant grain, time tz (start of burning) and time th
(end of burning) are shown in Fig. 1, and average pressure pth in time interval from tz to th is
defined as
pth =
1
th t z
th
p(t )dt .
(8)
tz
tz
Fig 1.
th
Pressure p - time t dependence in combustion chamber of rocket motor definition of time tz and th.
Equation (7) is used for determination of one point of burning rate-gas pressure
dependence (see Eq. (2)).
3.
EXPERIMENTAL
Basically, the measurements in test LPCV or RM involve measurements of pressure in

combustion chambers.
For the experiments typical double base solid propellants were used, containing
nitroglycerine (27%), nitrocellulose (58%), dinitrotoluene (9%), Centralite I (3%) and
catalysts (3%). The propellants contained various catalysts and were designated A, B, C.
Dimensions of the tested propellant grains were: external diameter D 0.025 m; internal
diameter d 0.013 m (i.e. the unit burning thickness e0 =(D-d)/4 0.003 m); length
L (0.120 or 0.100) m, weight was about 0.060 kg. The grains used in LPCV were fitted
with insulation on front surfaces. The insulation is needed to achieve constant burning area
of propellant grain surface (necessary to ensure legitimacy of Eq. (6)).
The volume of LPCV is about 2 dm3. Therefore the value of charge density
= /W 0.03 kg/dm3 is lower then usual value 0.20 kg/dm3 used if gun propellants are
732
Czech Republic
2006
[Content]
p [MPa]
tested in CV. Propellant grain is safely ignited by 5 g black powder and ignition pressure
pz = 1.8 MPa. Pressure p - time t curves for the tested propellants are shown in Fig. 2.
Burning times are approx. 0.25 s and the values of maximal pressures amount to 50 MPa.
Note that these values are very close to values achieved in rocket motor (see below).
50
Propellant B
45
Propellant C
40
Propellant A
35
30
25
20
15
10
5
0
0
Fig 2.
50
100
150
200
250
t [ms]
300
Pressure p - time t dependencies measured in LPCV - propelants A, B, C
p [MPa]
Gas pressure around the tested propellant grains in combustion chamber of the test
rocket motor are adjusted by critical diameter of output nozzle. Typical pressure p - time t
graphs for different nozzles for propellant B grains are shown in Fig. 3.
7,8 mm
14
7,4 mm
12
7,0 mm
16
6,7 mm
10
8
6
4
2
0
0
Fig 3.
Petr Stojan
100
200
300
400
t [ms]
500
Pressure p - time t curves measured in RM for four various nozzles propellant B grains
733
4.
Czech Republic
2006
[Content]
RESULTS
From the data of experiments in the LPCV (see Fig. 2) burning rate - pressure
dependencies were calculated by means of Eq. (6). Calculated curves for propellants A, B
and C are shown in Fig. 4.
u [m/s]
0,016
0,011
Propellant B
Propellant C
Propellant A
0,006
4
Fig 4.
10
12
14
16
p [MPa]
18
Burning rate u [m/s] - presure p [MPa] dependencies calculated from the

results of testing of propellants A, B, C in LPCV.
From the pressure p - time t curve measured in RM, burning rate was calculated (Eq. (7))
for corresponding average pressure calculated by (Eq. (8)). Calculated values for propellants
A, B and C are shown in Fig. 5.
u [m/s]
0,016
0,011
Propellant C
Propellant A
Propellant B
0,006
4
Fig 5.
734
10
p12
[MPa]
14
Burning rate u th [m/s] - presure pth [MPa] values calculated from results
of testing of propellants A, B and C in RM.
Czech Republic
2006
[Content]
Differences between the values of burning rates achieved in LPCV (see Fig. 4) and
in RM (see Fig. 5) at the same pressure levels are about 10%. There are several possible
reasons for the differences. One of them could be that real pressure course in LPCV is
influenced by thermal losses. Therefore values of maximal pressure and pressure rise rate are
lower. Calculation of burning rate gas pressure dependencies is influenced too. Another
reason could be acceleration of burning due to gas flow (i.e. erosive burning) in RM and
inaccurate definition of start-end burning time points (see Fig. 1).
5.
CONCLUSION
This paper presents the results of testing of burning process of rocket solid propellants
when low pressure closed vessel or rocket motor are used for measurements.
The dependence is measured of burning rate of various solid propellants on gas pressure
in combustion chamber. Differences between the values of burning rates achieved in LPCV
and in RM at the same pressure levels are about 10%.
REFERENCES
[1]
[2]
[3]
[4]
F. LUDVK, P. KONEN: Vnitn balistika raketovch motor na tuhou pohonnou hmotu,

Uebnice VA v Brn U-1153/1, Brno, 1999 (in Czech).
J. KUSK: Vnitn balistika raketovch motor na tuhou pohonnou hmotu, Licenn studium
Technologie a teorie vbunin, The University of Pardubice, 2005 (in Czech) .
M. IMEK, V. KUTTELWASCHER., P. STOJAN: Influence of velocity of gas flux to the burning
surface, Proc. 7th Seminar New Trends in Research of Energetic Materials, The University of
Pardubice, April 2004, p. 268.
M. VODA, S. SEDLEK, J. TMA: Vnitn balistika. Univ. Pardubice, 1978 (in Czech).
Petr Stojan
735
Czech Republic
2006
[Content]
EXPERIMENTAL RATING OF SURGE WITHSTAND CAPABILITY

OF STEEL FIBRE REINFORCED CONCRETE TO THE STANDARD
CONCRETE WITH DYNAMIC LOAD WITH PIHX 30 PLASTIC
EXPLOSIVE
Ji TOLLER*
* kpt. Ing. Ji TOLLER, Ph.D., University of Defence, K 203, Kounicva 65, 612 00 Brno, Czech
Republic
Abstract:
The article presents the results of experimental tests carried out on anti-spalling layer
made of steel fibre reinforced concrete, intended for protective structures. In particular,
the effects of plastic explosive PlHx 30 on slabs made of plain concrete or steel fibre
reinforced concrete are shown.
Keywords:
1.
steel fibre reinforced concrete, anti-spalling layer,
INTRODUCTION
Concrete, widely used material especially at construction business possesses two

fundamental disadvantages: low strength at tensile stress and fragility while damaged. Such
two disadvantages are partly removed in common constructions by placing in the
conventional reinforcement that has been inserted into building blocks as individual steel
rods, mesh and/or grids. Both mentioned disadvantages are eliminated using this
reinforcement because of rightly placed reinforcement that is capable of to take up resultant
force of tensile stresses as well as to avoid the failure of individual fragile construction
elements.
By employing the conventional reinforcement the structure behavior changes, however
the concrete properties remain the same. We can reach increase in tensile strength and
change of fragile character in concrete failure only by direct reinforcement of concrete
structure. To achieve such reinforcement there is a necessity to use reinforcing elements
meeting the size of components that create its structure.
The reinforcing element is a fiber. The important advantage of concrete is that all
production technologies enable its reinforcement by fibers. Not only fibers influence the
properties of hardened fiber-concrete but the fiber-concrete hardening in process and last but
not least the properties of fiber-concrete mixture of newly made steel fiber reinforced
concrete.
From this point of view is the conjunction of fiber reinforced concrete and armored
concrete very advantageous. Armored concrete itself is resistive nevertheless its structure
remains fragile especially in the armature area.
Presently, there is a research in the area of use of fiber concrete for protective
constructions at the department of Engineer Corps under way. As an experimental
736
Experimental Rating of Surge Withstand Capability of Reinforced Concrete to . . .
Czech Republic
2006
[Content]
examination of the dynamic load, the tests of fiber concrete slabs have been carried out with
the PIHx 30 plastic explosive.
The search for new materials used for building of protective constructions is a result of
development of new weapons that have been used against resistive targets. This is why it is
necessary to look for the new materials for new and existing structures as well and adapt
such structures to the new requirements.
2.
LOADING THE TEST SLABS BY PLHX 30 PLASTIC EXPLOSIVE

BLAST
Some test slabs were manufactured of conventional concrete and of fiber concrete, too.
To set the guaranteed concrete strength, the test objects for strength tests were being made
simultaneously such as 150 mm edge test cubes and for tensile strength tests, the 150 150
600 mm test squared logs. The test body dimensions were selected with respect to the used
DRAMIX RC-65/50-BN fiber.
The dimension of test slabs was determined with regard to material properties of fiber
concrete and slab manipulation while being tested. The tested slabs dimension was 1000
1000 150 mm and weighted approximately 370 kg.
2.1
Measuring of slab quality by impulse ultra sound method
The quality of used concrete and fiber concrete was stated prior to dynamic loading by
plastic explosive by impulse ultrasound method. The reason to determine quality of material
and changes to come after the slab has been loaded by plastic explosive lead to selection of
above mentioned method. This very method was also employed to learn about dynamic
modulus of elasticity of concrete and fiber concrete used. Measurement was taken twice
once just before the initial test of loading by blast and for the second time after the intrinsic
blast test. The test was carried out and evaluated according to standard CSN 73 1371.
Fig 1.
2
Ji toller
Tested fibre-concrete slab after being loaded by plastic explosive blast

737
Czech Republic
2006
[Content]
Considering the results of ultra sound measuring and the result of variation factor (for
concrete slabs v=1.9%, fiber concrete slabs v=1.87%) we can conclude the conventional
concrete slabs as well as the fiber concrete slabs were made of high quality material. In
respect to the variation factor we can say that despite the fact they were manufactured out of
lab they were of laboratory quality (up to v=2.5%).
Having the results taken after the loading of slabs by PIHx 30 plastic explosive blast, we
could assess what way the specific charge had influenced the existing slab. It was stated
there were great differences between the conventional and fiber concrete slabs.
2.2
Assessment of spalling
After preparation of slabs the loading was carried out by plastic explosive blasting.
Firstly, the conventional concrete slabs were loaded. After being assessed, the charges for
fiber slabs were set. Graphic assessment of relation between the spalling volume and the
used explosive is mentioned in Fig.2.
9000
Volume of fragments [cm3]
8000
90
7000
6000
5000
concrete
4000
200
70
SFRC
3000
2000
50
1000
0
0
20
40
50
60
120
90100
110 130
150
70
80 100 120 140 160 180 200 220
Explosive PlHx 30 [g]
Fig 2.
Relation between spalling volume and use explosive
Distribution of fibers on the spalling surface is somehow equal. The number of fibers
matches the size of spalling and indicates the distribution of fibers in the test body. When
conventional slabs were loaded by 50g of plastic explosive, the spalling did not occur. Only
some fissures with little fragments appeared on the slab surface. The spalling occurred after
the second test was performed. With increased load the volume and depth of spalling rose.
Looking at the graph of relation between spalling volume and used explosive (Fig.2) you
will find the spalling had risen up to the value of 200g after the destruction of slab took place
two times consecutively.
It also happened that the spalling did not occur in fiber concrete slabs while being loaded
by 50g and next by 70g of explosive. In the first case only micro fissures appeared and in the
738
Czech Republic
2006
[Content]
second case the fissures with little fragments showed up. Further, while being loaded the
spalling volume had increased rather slowly as seen in Fig.2
Fig 3.
3.
Fibre concrete slab being initiated by plastic explosives (in the center you
can see creation of spalling. The picture was taken by high-speed camera)
CONCLUSION
Fiber concrete is one of concrete materials with fiber reinforced structure. It is a concrete
that varies from others by structure reinforced by scattered steel fibers. This fact while
being reinforced by certain minimal amount of fibers is a ground for important change of
properties that have been marked as basic drawbacks of conventional concrete. Such
reinforcement results in more strength and resistance against dynamic load.
The results revealed the fiber concrete is not suitable for anti-spalling layer. When
loaded by blast strength there is considerable tension on inner face that the fiber concrete is
not capable to carry, thus resulting in spalling.
The right way to find new materials used for anti-spalling layer could be the sandwich
structure. It also applies to armored vehicles where basically the same effect at protective
constructions takes place while being hit by high explosive ammunition.
Well, it is necessary to conclude that fiber concrete probably will not replace the
armoring that has been used for reinforcing the protective constructions. However, being
compared with conventional concrete the fiber concrete has displayed up to three times
higher endurance. Such a fact predetermines afore mentioned material to be used at
protective constructions. More resistant structure (main supporting structure) will be created
by incorporation of reinforced concrete and fiber concrete. This way created structure should
Ji toller
739
Czech Republic
2006
[Content]
prove higher endurance against penetration of bombs and projectiles designed to destroy
resistant targets.
740
Fig 4.
Layout of izovel concrete slab prior to blast loading
Fig 5.
Layout of izovel concrete slab after blast loading
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
COUFAL, JOSEF KAPLAN, VROSLAV. Ochrann stavby. [Skripta] S-297/III, Vojensk

akademie v Brn, 1993.
KRTK, JI TRTK, KAREL VODIKA, JAN. Drtkobetonov konstrukce. 1. vyd.
Praha: Informan centrum KAIT, 1999. 108 s. ISBN 80-86364-00-3.etc.
TOLLER, JI. Drtkobeton stavebn materil pro prmyslov podlahy a jeho mon vyuit
pro ochrann stavby. In LOGVD 2004. 7. vdecko odborn konference s mezinrodn ast.
ilinsk univerzita v ilin, 2004, s.250-258. ISBN 80-8070-319-1.
KAPLAN, VROSLAV TOLLER, JI. Vyuit drtkobetonu v rmci ochrannch staveb. In
2. doktorandsk konference. Univerzita obrany, Brno, 2005, 6 stran. ISBN 80-85960-98-2.
KAPLAN, VROSLAV TOLLER JI. Experimentln oven vlastnost drtkobetonu pi
rzovm zaten. In SANACE 15. mezinrodn sympozium. Brno: Sdruen pro sanace
betonovch konstrukc SSBK, 2005, s.378- 382. ISSN 1211-3700.
Ji toller
741
Czech Republic
2006
[Content]
AN EXPERIMENTAL STUDY OF DETONATION TRANSMISSION

FROM HYDROGEN-OXYGEN TO HYDROGEN-AIR MIXTURE
M.Bratowski, A.Dabkowski, A.Teodorczyk
Warsaw University of Technology, ITC, Nowowiejska 21/25, 00-665 Warszawa, Poland
Abstract:
This paper presents experimental results on detonation transmission through sharp
interface between stoichiometric hydrogen-oxygen mixture and less reactive
stoichiometric hydrogen-air mixture. Experiments were conducted in detonation tube of
square cross-section 110110 mm and 8 m long. Detonation wave was monitored by
pressure transducers and smoked foils inserted into the tube. The effect of initial
pressure on detonation transmission was studied.
Keywords: detonation, transmission, hydrogen, gradient
1.
INTRODUCTION
The gradient transmission problem is that of a detonation moving from one mixture
into another, either through discontinuous change in mixture composition or a continuous
mixture composition variation. A detonation propagating from a more to less sensitive
mixture can continuously adjust and travel as a steady CJ wave corresponding to the local
conditions, fail and then re- initiate, or fail completely.
Paterson[1] performed an analysis of a detonation impinging on a contact surface
between a combustible and inert mixture using a simple model. He took the gas state behind
the incident wave as the CJ conditions of the detonation wave (i.e. infinite kinetic rates).
Paterson also noted that this assumption of a continuous region behind the wave at CJ
conditions is not realistic (neglecting the following Taylor expansion wave), but expected
this to be valid for calculating the initial transmitted shock strength. The analysis was
extended to combustible mixtures on both sides of the contact surface by introducing an ad
hoc one parameter characterizing the degree of reaction in the receptor mixture.
Thomas[2] used Patersons non-reacting receptor analysis and found excellent
agreement with the initial strength of the transmitted shock when compared experiments in
which a detonation impinged on an inert contact surface. The experimental observations
were of an initially decaying transmitted shock which leveled off a short time after the
contact surface interaction. This decay to a plateau value was accurately captured by Thomas
with computations taking the Taylor expansion following the detonation wave into account.
Experiments in which the contact surface was allowed time to diffuse demonstrated the
detonation wave velocity and cellular structure rapidly conforming to the local mixture
conditions.
Bjerketvedt[3], Bull[4], Engebretsen[5], Thomas[2] and Teodorczyk [6,7] have studied the
transmission of detonations across an air gap between combustible mixtures, often
determining critical air gap lengths and time/distance for reinitiation once the transmitted
shock entered the second combustible mixture. Thomas showed that gradual gradients
742
An Experimental Study of Detonation Transmissionfrom Hydrogen-Oxygen to . . .
Czech Republic
2006
[Content]
significantly assist the transmission process, whereas sharp gradients tend to cause
immediate decoupling of the shock and reaction zone.
Engebretsen[5], Strehlow[8] and Benoan[9] have conducted experiments of detonation
transmission through combustible mixture gradients, focusing on identifying conditions
under which the detonation could continue to propagate and investigating the effect of
mixture perturbation on the cellular structure.
This paper reports recent experimental study on detonation transmission through
sharp change in mixture composition from stoichiometric hydrogen-oxygen mixture into
stoichiometric hydrogen-air.
2.
EXPERIMENTAL
The detonation tube shown in Figs.1 and 2 was used in the experiments. It consists of
a 1 m long booster as well as 2 m and 4 m long square cross-section channels with internal
dimensions 110110 mm. The booster is filled with the oxy-acetylene stoichiometric
mixture at 0.5 bar, which ignited, by a 1 J electric spark rapidly detonates initiating in turn
detonation in the first 2 m long channel filled with the stoichiometric hydrogen-oxygen
mixture in all experiments.
The second 4 m long channel was separated by plastic membrane and filled with
weaker stoichiometric hydrogen-air mixture. The experiments were conducted for three
values of initial pressure: 0.1, 0.2 and 0.3 bar. The pressure was equal in both channels
before ignition. Detonation velocity in the first cha nnel was calculated from the signals of
two pressure transducers C1 and C2 and the behavior of combustion wave (fast deflagration
or detonation) in the second channel was monitored by four pressure transducers (C3, C4, C5
and C6) and photodiode. In additio n the structure of detonation wave was studied with the
use of smoked metal foils inserted along the bottom wall of second channel.
Fig 1.
Schematic of experimental setup. Numbers below transducers indicate

distance in mm from the first membrane
Andrzej Teodorczyk, A. Dabkowski, M. Bratowski
743
Fig 2.
3.
Czech Republic
2006
[Content]
The view of detonation tube
RESULTS
Table 1 presents the summary of the experimental results obtained for hydrogen-air
mixture in the second channel. Three experiments for each initial pressure are listed in the
table. From the velocities calculated from pressure signals it is evident that detonation wave
successfully transmits from stronger hydrogen-oxygen mixture to weaker hydrogen-air and
quickly adjusts its velocity to steady CJ value. The larger velocities between transducers C3
and C4 as well as pressure values in Fig. 3 show that the wave is at the beginning slightly
overdriven.
Table 1. Summary of experimental results
Channel A
Chanel B
H2 -O2
H2 -air
Initial
pressure
[bar]
CJ
detonation
velocity
[m/s]
Velocity
between C1
and C2
[m/s]
CJ
detonation
velocity
[m/s]
0.1
2716.40
2870.26
1918.30
1914.24
1879.70
0.1
2716.40
2912.06
1918.30
1965.41
1912.78
0.1
2716.40
2604.17
1918.30
1884.66
1876.88
0.2
2754.00
2597.40
1935.30
2017.76
1935.73
0.2
2754.00
2564.10
1935.30
1987.28
1947.80
No
744
Velocity
Velocity
between C3 between C5
and C4
and C6
[m/s]
[m/s]
Czech Republic
2006
0.2
2754.00
2568.05
1935.30
1933.49
1930.50
0.3
2776.20
2639.92
1944.90
2024.29
1934.98
0.3
2776.20
2653.93
1944.90
2099.08
1960.02
0.3
2776.20
2660.99
1944.90
2055.92
1987,28
[Content]
Fig.3. Pressure profiles from transducers C3, C4 and C5

located in the acceptor channel for experiment no 4
Fig.4. Detonation cells for hydrogen-air at 0.2 bar
745
Czech Republic
2006
[Content]
Fig.5. Cell sizes for transmitted detonation in hydrogen-air compared with cell size
reported in literature for steady wave
Figure 4 presents detonation cells recorded on smoked metal foil for transmitted
detonation in hydrogen-air mixture at 0.2 bar. This cells are shown statistically in Fig.5
together with average value (blue line) and steady detonation value taken from literature [].
The fact that all the cells are smaller than for the steady case also indicates that detonation
wave is slightly overdriven at that phase of propagation. Figure 6 presents cell sizes of the
transmitted detonation wave for all three pressures compared with steady values.
The important question for the problem of detonation transmission from one mixture to
another is: does the wave transmit smoothly with continuously growing cells or there is
some transition distance for reinitiation after failure of regular wave cellular structure. The
smoked foils revealed that after transmission cells start to appear after some distance (Table
2).
Table 2. Transition distance for detonation cell appearance on smoked foils
Initial pressure [bar]
Transition distance [mm]
0.1
0.2
500
0.3
230
For lowest initial pressure no detonation cells were observed on smoked foils which
suggests that transition to detonation occurred further away from diaphragm. The length of
smoke foil was only 600 mm. This conclusion is also supported by pressure profile recorded
by transducer C3 (Fig.6) which shows first strong shock wave followed by detonation wave.
746
Czech Republic
2006
[Content]
Fig.6. Pressure profiles from transducers C3, C4 and C5

located in the acceptor channel for initial pressure of 0.1 bar
4.
CONCLUSIONS
Experimental study of detonation transmission from stoichiometric hydrogen-oxygen

mixture to less reactive stoichiometric hydrogen-air mixture through discontinuous
composition interface revealed that the detonation wave first fails and then re- initiates after
some transition distance which depends on initial pressure.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
Patterson S.: Contact transmission of detonation, Proc.of the Combustion Institute, vol.4, 1953,
pp.468-471
Thomas G.O., Sutton P., Edwards D.H.: The behavior of detonation waves at concentration
gradients, Combustion and Flame, vol.84, 1991, pp.312-322
Bjerketvedt D., Sonju O.K., Moen I.O.: The influence of experimental condition on the
reinitiation of detonation across an inert region, Dynamics of Explosions, Prog.Astro.Aeron.,
vol.106, 1986, pp.109-130
Bull D.C., Elsworth J.E., McLeod M.A., Hughes D.: Initiation of unconfined gas detonations
in hydrocarbon-air mixtures by a sympathetic mechanism, Prog.Astro.Aeron., vol.75, 1981,
pp.61-72
Engebretsen T., Bjerketvedt D., Sonju O.K.: Propagation of gaseous detonations through
regions of low reactivity, Dynamics Aspects of Detonations. Prog.Astro.Aeron., vol.153, 1993,
pp.324-346
Teodorczyk A., Benoan F.: Interaction of Detonation with Inert gas Zone, Shock Waves 6, no
4, 1996, pp.211-223
Teodorczyk A., Thomas G.O., Ward S.M.: Transmission of a Detonation Across an Air Gap,
Proc. of the 20th Symposium on Shock Waves, 1996, pp.1095-1100
Strehlow R.A., Adamczyk A.A., Stiles R.J.: Transient Studies of Detonation Waves,
Astronautica Acta, 17, 1972, pp.509-527
Benoan F., Teodorczyk A.: Experimental Investigation of Transmission of DetonationThrough
Composition Gradients, Archivum Combustionis, 13, 1993, pp.21-32
747
Czech Republic
2006
[Content]
ON SOME METHODS OF DETERMINATION

OF THE DETONATION ENERGY OF EXPLOSIVES
Waldemar A. Trzciski and Jzef Paszula
Military University of Technology,
Kaliskiego 2, 00 908 Warsaw, Poland
Abstract:
Two methods of determination of the detonation energy of explosives are discussed in
the paper. The first one is a calorimetric method, in the second one the results of a socalled cylinder test are used. A calorimetric system for measurement of the heat of
detonation of explosive charges is applied in which compressed inert gas is used to stop
secondary reactions in the detonation products. The basis of the second method is the
result of a cylinder test. In this test a copper tube is accelerated by the detonation
products of explosive. The expansion of the tube is recorded. From the cylinder test
results the kinetic energy of the tube and that of detonation products can be calculated.
The kinetic energies for the tested and standard explosives enable us to determine the
detonation energy. The method is verified by the use of chosen cylinder test results.
Keywords:
1.
explosives, measurement techniques
INTRODUCTION
An influence of detonation conditions on the heat effect of explosives was investigated

by D. L. Ornellas[1] in the bomb of 5.28 litre in volume. Ornellas studies indicate that the
heat effects from cylindrical heavily confined charges correspond to the detonation heats of
explosives. The detonation products from confined charges are those on the ChapmanJouguet (CJ) isentrope in the freeze-out range of 1500 to 1800 K. The results obtained from
unconfined charges coincide with the heats of reactions proceeding under conditions of high
temperature and low pressure. Among others, the calorimetric heats of TNT and HMX were
measured. The values of 1093 and 632 cal/g were obtained for heavily confined and
unconfined TNT charges located in the calorimetric bomb with vacuum. The heat effects of
1479 and 1334 cal/g were measured for HMX charges under similar conditions.
The influence of gaseous filling of a steel chamber of 1.50 m3 in volume on the
composition of detonation products was studied in Ref.[2]. It was proved that compressed
argon behaved as a confining medium and it caused that the heat output of detonation
increased. In this work, inert gas filling the calorimetric bomb is applied to limit the effect of
secondary reactions in the detonation products [3-4].
The results of the expansion of a metal tube caused by the detonation products of an
explosive charge placed within it can provide considerable information concerning the
amount of chemical energy released from the explosive and the extent of its transfer to the
surroundings. Such a test is called the cylinder test. Some methods enabling an application
of the expanding tube records for determination of the acceleration capacity, the expansion
work and the detonation energy of high explosives, and the equation of state of the
748
Czech Republic
2006
[Content]
detonation products were proposed in Refs. [5-6]. In this paper the improved method of
determination of the detonation energy from the cylinder test results is presented.
2.
CALORIMETRIC DETERMINATION OF THE DETONATION HEAT
Figures 1 and 2 show the calorimetric system used for measurement of the heat of
detonation of explosive charges. The spherical steel bomb having an internal volume of
5.6 dm3 was placed in the steel calorimeter vessel. The vessel contained about 26.5 dm3 of
water. The water temperature was measured by a thermometer sensor and recorded by a
memory system of the control unit. The calorimeter was surrounded by a constanttemperature jacket.
Figure 1. Calorimetric bomb
Figure 2. The measurement set-up
The heat capacity (thermal equivalent) of the calorimeter was determined by burning
about 5 g of benzoic acid in oxygen at a pressure of 2 MPa. The thermal equivalent, taken as
the average of five tests, was 154 470 J/oC.
The heat of detonation was measured in the bomb filled with compressed nitrogen,
which replaced Ornellas solid confinement of a charge. The charge of explosive was
suspended in the centre of the bomb. Weighed samples of explosives were pressed in
matrices of an inner diameter of 30 mm. Each charge has a fuse cavity 8 mm in diameter and
15 mm long. The heat of explosion of an electrical fuse was estimated in Ref.[3] and it was
11720 J.
At first step, the influence of the initial pressure of gas filling the calorimetric bomb on
the heat effect of detonation was determined. TNT charges of 20 g mass were detonated in
nitrogen under pressure from 0.1 to 2.2 MPa. Measured heat effects are presented in Fig. 3.
Above the value of initial pressure of 1 MPa the heat changes in a interval 4530 210 J/g (
5%). This value is close to the detonation heat measured in Ref.[1] for heavily confined TNT
charges (average value of 4575 J/g). On that account, the heat of detonation of other
explosives was measured in a bomb filled with nitrogen at a pressure of 2 MPa.
Waldemar A. Trzciski, Jzef Paszula
749
Czech Republic
2006
[Content]
TNT of 1.58 g/cm3, RDXph (RDX/(CH2)n 94/6) of 1.64 g/cm3, HMXph (HMX/(CH2)n
96/4) of 1.74 density, FOX-7 of 1.78, and NTO of 1.77 g/cm3 were tested. The results
obtained for different mass of charges are presented in Table 1.
5500
5000
4500
Qd [J/g]
4000
3500
TNT
3000
2500
experiment
regresion
2000
1500
1000
0.0
0.5
1.0
1.5
2.0
2.5
p [MPa]
Figure 3. Dependence of the detonation heat of TNT on initial pressure of nitrogen filling
a calorimetric bomb
Table 1. Detonation heat of high explosives for different charge mass
TNT
RDXph
HMXph
FOX-7
NTO
m [g]
Qd [J/g]
m [g]
Qd [J/g]
m [g]
Qd [J/g]
m [g]
Qd [J/g]
m [g]
Qd [J/g]
15.018
19.980
20.073
20.095
25.073
4351
15.076
5341
15.110
5906
19,912
4837
19,994
3070
4316
20.073
5624
15.024
5912
19,894
4930
19,908
3227
4501
25.097
5557
19.989
5802
20,037
4814
19,447
3298
24.983
5839
4515
4512
(Qd)aver = 4439
(Qd)aver = 5508
(Qd)aver = 5865
(Qd)aver = 4860
(Qd)aver = 3199
For comparison, calorimetric detonation heats measured by Ornellas for heavily

confined charges are as follows 4573, 4480, 4576 and 4672 J/g for TNT, 6093 J/g for the
mixture RDX/HMX 91.35/8.65 and 6182 J/g for pure HMX. Therefore, quite good
agreement between Ornellas results and those obtained in this work enables us to formulate
the inference that the heat of detonation of high explosives can be measured by the use of
unconfined charge in a calorimetric bomb filled with inert gas at high pressure.
750
3.
Czech Republic
2006
[Content]
ESTIMATION OF THE DETONATION ENERGY FROM CYLINDER

TEST RESULTS
In the cylinder expansion test, a metal tube is driven by the detonation products of a
cylindrical charge of explosive. The expansion of the tube is monitored by precise methods
of measurements of the quickly changing processes, such as streak photography, laser
interferometry, X-ray photography. The cylinder test is mainly used in the methods in which
parameters of the Jones, Wilkins, Lee (JWL) equation of state of the detonation products of
explosives are determined. However, the cylinder test can be also applied for determining
some energetic characteristics of explosives, including the acceleration ability and
detonation energy of the explosive [6].
The result of the cylinder test is the experimental dependence of the radius of external
surface of the tube on the time or the photograph of the expanding tube. From the latter, the
dependence of the position of external surface on the axis co-ordinate can be obtained. On
assuming that the motion of the detonation products and the tube material is stationary, the
time-dependence of the external radius of tube can be calculated. At the end, the velocity of
the tube is determined [6].
During the process of tube acceleration, the work done by the detonation products is
converted to the kinetic energy of the tube, detonation products and air surrounding the tube,
and to the internal energy of tube material and air. If the kinetic energy of air and the heat
losses in the tube material and air are neglected, then the law of energy conservation can be
written in the form
e0 = ekL ( v ) + ekDP ( v ) + pi ( v ) dv
(2.1)
where e0 denotes the detonation energy, ekL and ekDP are the kinetic energy of the tube and
detonation products, respectively. The quantity pi(v) is the pressure on the isentrope starting
from the CJ point, and v is the specific volume of expanding detonation products. The
integral in Eq. (2.1) corresponds to the internal energy remained in the detonation products.
All quantities in Eq. (2.1) are related to unit mass of the explosive.
The sum of kinetic energies of the copper tube and the detonation products per unit mass
of the explosive is called the Gurney energy, and for cylindrical tube it can be expressed by
the following relation
1 u2
EG ekL + ekDP = + L
2 2
(2.2)
where is the ratio of the tube and explosive masses per unit area.
If we assume that the integral in (2.1), i.e., the internal energy of the detonation
products, is proportional to the Gurney energy in the identical way for all explosives, the
following relations are true for the given and standard explosives
e0 = A ( ekL + ekDP )
(2.3)
st
st
e0st = A ( ekL
+ ekDP
)
(2.4)
where A is constant for chosen volume v.
751
Czech Republic
2006
[Content]
Taking into account the dependence (2.2), the following equation can be obtained from
(2.3) and (2.4)
1
+
e0
2
=
e0st st + 1
2
uL
st
uL
(2.5)
where uL and uLst are velocities of the tubes driven by the detonation products of the given
and standard explosives, respectively.
If the detonation energy is related to unit volume of explosive, then Eq. (2.5) has the
form
1
+ 0
e0
2
=
e0st st 1 st
+ 0
2
uL
st
uL
(2.6)
where 0 and 0st denote the density of the given and standard explosives, respectively.
From Eqs. (2.5) or (2.6) the detonation energy of the given explosive can be estimated if
the detonation energy of standard explosive and the results of cylinder test for both
explosives are known.
To verify the method of determining the detonation energy, the experimental results of
the cylinder test given in Ref.[7] were used. The velocity of the tube was calculated from the
function approximating the time-dependence of radius of the external surface of the tube.
Dependence of the square of velocity on the reciprocal volume of the detonation products is
presented in Fig. 4. The values of uL2 for v0/v = 0.1 were taken into account. Pentryt (PETN)
was chosen as a standard explosive. It was assumed that the detonation energy of PETN was
equal to the energy E0 from the JWL equation of state[7]. The detonation energies of chosen
explosives from Ref.[7] calculated by using relation (2.6) are presented in Table 2.
From Table 2 it follows that, for high explosives, differences between the detonation
energies calculated by the method proposed and those given in literature do not exceed 12 %.
The highest one is for nitromethane. The detonation energy of TNT estimated from the
cylinder test data is 3865 J/g and that given in Ref.[7] is 4295 kJ/kg. Also the values of the
calorimetric heat of detonation of TNT given in literature differ significantly from each other
and, in reality, they depend on the measurements conditions.
The method proposed cannot be applied for nonideal explosives. From Table 2 it follows
that the discrepancy Rw exceeds considerably 10 % for these explosives (last four explosives
in Table 2). Such a high disagreement may be caused by the fact that heat releasing in
nonideal explosives takes place also behind the detonation zone. Probably, this energy is
ineffectively transferred to the tube during later stage of the acceleration.
752
Czech Republic
2006
[Content]
3.7
3.6
3.5
LX-09
uL2 [m2/s2]
3.4
LX-10
3.3
3.2
3.1
PETN
LX-07
3.0
LX-14
LX-04
2.9
2.8
PBX-9011
2.7
2.6
0.00
0.05
0.10
0.15
0.20
0.25
v0/v
Figure 4. Dependence of the square of velocity of the copper tube on reciprocal volume
of the detonation products
Table 2. Comparison of the detonation energy calculated from the cylinder test data,
E0cal, with the detonation energy given in literature, E0 [7],
Rw = [(E0cal-E0)/E0] 100 %
Explosive
PETN
Comp. B
Cyclotol
E0
[GPa]
10.1
8.5
E0cal
Rw
[GPa]
-
[%]
-
8.59
+1.1
9.2
9.28
+0.8
HMX
10.5
11.26
+7.2
LX-04
9.5
10.37
+1.7
LX-07
9.6
9.98
+4.0
LX-09
10.5
10.70
+2.8
LX-10
10.4
10.57
+1.6
LX-14
10.2
10.37
+1.7
PBX-9011
8.9
9.27
+4.1
PBX-9004
10.2
10.41
+2.1
TNT
7.0
6.30
-9.9
Nitromethane
5.1
4.50
-11.7
Aquanal
5.5
2.51
-54.4
Pourvex
4.5
3.82
-15.1
ANFO
2.9
2.21
-23.6
Unigel
5.1
3.91
-23.4
753
Czech Republic
2006
[Content]
To record the copper tube driven by the detonation products of explosives tested in
calorimetric experiments, Scandiflash X-ray set was used. The cooper tube filled with the
explosive tested was 300 mm in length. Its internal diameter and wall thickness were 25 mm
and 2.5 mm, respectively.
An exemplary radiograph of the copper tube driven by detonation products is shown in
Fig. 5. From the photograph, dependence of the external surface radius of the tube on the
axial co-ordinate was determined using graphical computer programs. The velocity of the
tube was determined by using the method described in Ref. [6].
Figure 5.
X-ray photograph of the copper tube driven by detonation products of HMXph
As a standard explosive, phegmatised RDX was chosen. Its heat of detonation (Qdet =
5508 J/g) was determined in a calorimetric bomb filled with compressed nitrogen. The heat
of detonation of RDXph was assumed to be close to the detonation energy (e0) Ref.[6] .
Results of the calculation of the detonation energy of TNT, HMXph, FOX-7 and NTO are
4190, 5980, 5060 and 3989, respectively. NTO, differences between the detonation energies
calculated from the cylinder test results and the calorimetric heats of detonation for these
explosives do not exceed 6 %. But NTO is extremely insensitive high explosives and the
detonation of charges of about 20 g mass in the calorimetric bomb is probably incomplete.
4.
SUMMARY
The calorimetric set-up for measurement of the heat of detonation of explosive was
tested in which inert gas filled a steel bomb. It was proved that compressed nitrogen stopped
secondary reactions in the detonation products and it behaved like a solid envelope confining
the explosive charge. The improved method of determination of the detonation energy from
the cylinder test results was also presented. The method was verified and used for
determination of the detonation energy of explosives tested in the calorimetric system. Quite
good agreement between the results of both methods was observed.
754
Czech Republic
2006
[Content]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
D. L. Ornellas, Calorimetric determinations of the heat and products of detonation for

explosives: October 1961 to April 1982, Report UCRL 52821, Lawrence Livermore National
Laboratory.
F. Volk, Detonation products as a function of initiation strength, ambient gas and binder
systems of explosives charges, Propellants, Explosives, Pyrotechnics 21, 155-159, (1996).
S. Cudzio, R. Trbiski, W. A. Trzciski, P. Wolaski, Comparison of heat effects of
combustion and detonation of explosives in a calorimetric bomb filled with inert gas or air (in
Polish), Biul. WAT, 47, 11 (1998).
D. Gazowski, W. A. Trzciski., M. Siwirski, Measurement of the detonation heat of
explosives in a calorimetric bomb filled with inert gas (in Polish), Biul. WAT, 54, 2-3 (2005).
R. Trbiski, W. Trzciski, Determination of an expansion isentrope for detonation products of
condensed explosives, Journal of Technical Physics, 40, 4 (1999).
W. A. Trzciski, Application of a cylinder test for determining energetic characteristics of
explosives, Journal of Technical Physics, 42, 2 (2001).
H. Hornberg, Determination of fume state parameters from expansion measurements of metal
tubes, Propellants, explosives, Pyrotechnics, 11, 23-31 (1986).
755
Czech Republic
2006
[Content]
REGARDING THE FUNDAMENTAL PROBLEM SOLVING OF

INTERIOR BALLISTICS IN THE CASE OF FLEGMATISED
POWDERS
Titica VASILE, Cristian BARBU, Doru SAFTA
Military Technical Academy, 81-83 George Cosbuc Avenue, Bucharest, Romania
Abstract
The fundamental problem solving of interior ballistics - the variation of the

powder gases pressure and the projectile velocity versus displacement inside of
barrel and versus time - on the bases of geometrical law of powder burning for
special treated powders can to lead at significant errors. For that, in this paper
it is presented a new algorithm for the solving of fundamental problem, which is
based on analytical relations from l med method, changed accordingly with the
essential specific features at these powders.. For an extant ballistic system, the
theoretical results, obtained with the aid of both this algorithm and the another
algorithm as well as experimental data are compared.
Keywords: Flegmatised powder, burning rate law, firing phenomenon, ballistic system.
1.
INTRODUCTION
The flegmatised powders are the powders, usually of simple form, treated with
substances which decrease burning rate. As a result, in case of this powders the ballistic
characteristics - the powder force f , the powder gases covolume , the burning rate
characteristic u1 - vary stratum by stratum due to the penetration depth of flegmatising
substance in the thickness of powder grain, fact which determines that the forming process
of gases to be distinct in every stratum. Consequently, the fundamental problem solving of
interior ballistics for this powders which is based on burning geometrical law, proposed by
French ballistic researcher Vieille, can to lead at significant errors in the value evaluation of
main ballistic magnitudes: the powder gases pressure p ; the projectile velocity v ; the
displacement of projectile inside of the barrel l . In order to avoid this errors in the value
evaluation of main ballistic magnitudes in case of using of flegmatised powders it is
modified the analytical relations for this magnitudes, which was obtained in l med method,
taking into account known simplified hypotheses [1, 2, 4]. In this case, besides these simplified
hypotheses, are accepted following hypotheses:
a. All powder elements have same nonhomogenity of physical-chemical composition of
strata. Every stratum has its individual characteristics: thickness 2e1.i ; powder force fi ;
burning rate characteristic u1.i ; gases covolume i ;
b. Every stratum has its personal composition of burning produces, which does not mix
during of the powder burning or adiabatic detente;
756
Regarding the Fundamental Problem Solving of Interior Ballistics in the Case of . . .
Czech Republic
2006
[Content]
c. First period of firing phenomenon is divided on phases, in every phase burns a stratum
which has its individual characteristics; in the preliminary period and in the first phase of I
period the first stratum burns.
The respective relations are established for the powders with simple form, which was
flegmatised.
The notations from this paper are those notations that are used usually in interior
ballistics [1, 2, 4, 5, 6, 7].
2.
THE RELATIONS FOR THE PRELIMINARY PERIOD
The used relations in the preliminary period are usually relations from specialty
literature [1, 2, 4] .
The relation gives the fraction of burned powder at the end of preliminary period
0 =
1 1

f1
p0 pa
+ 1 1
(1)
The relative surface of burned powder is calculated with the aim of relation
0 = 1 + 4 0 .
(2)
The relative thickness of burned powder is calculated using the relation

2 0
0 +1)
z0 = (
(3)
The relation for time at preliminary period end is
t 0 = 2,303
I k .1(11 )
f1
lg p 0 .
(4)
The fictitious length of free volume of loading chamber at the end of preliminary period
is calculated with the aim of relation
l 0 = l 0 [1 (1 1 ) 0 ] .
3.
(5)
THE RELATIONS FOR I PERIOD
3.1
THE I PHASE
The relations for the calculation of main ballistic magnitudes in the I phase of I period
are those known relations from specialty literature [1, 2, 3, 4].
The projectile velocity is calculated with the aim of relation
v=
2
Titica Vasile, Cristian Barbu, Doru Safta
sI k .1
q
x,
(6)
757
Czech Republic
2006
[Content]
in which:
e
I k .1 = u 1 ; x = z z 0 ; x (0, z1.k z 0 ) ; z = ee ; z1.k =

1
1.1
e1.1
e1
; z 0 = e0 ;
1
e
e1 - the half-thickness of powder element;

u1.1 -the characteristic of burning rate of burned powder in I phase.
The relation gives the fraction of burned powder
= 0 + k1 x + x 2 ,
(7)
where
k1 = (1 + 2z 0 ) .
The projectile displacement inside of barrel in this phase it is expressed using the
relation
l = l med ( Z x B1 1) ,
B
(8)
in which:
l med = l 0 [1 (1 1 ) med ] ; med = 12 ( + 0 ) ;

s 2 I 2 k .1
f1 q
B=
; B1 =
B
2
; Z x = 1
) (1 )
b +1
2b
= k1 x ; 1 = 1+2b ; 2 = 12b ; b = 1 + 4 ; =
B
l0 =
W0
s
; =
W0
b 1
2b
B1
k12
0 ;
Relation gives the pressure of powder gases inside of barrel in this phase
p=
f1( 0 + k1x B1x 2 )

,
s (l +l)
(9)
in which
l = l 0 [1 (1 1 ) ] .
The position of maximum pressure and its value are calculated through successive
approximations using the relation
xm =
k1
.
B( + 1)
2
pmax
1
1 + (1 )
f1
(10)
If xm [0, x1.k ] , then the maximum pressure is achieved in this phase, in contrary case
it will achieve in following phases of I period.
758
Czech Republic
2006
[Content]
THE i PHASE
3.2
At the end of phase I can be calculated the values v1.k , p1.k , l 1.k and 1.k for the
independently variable x = x1.k which represents the border between the first phase and the
second phase. The borders between the following phases are noted with
x2.k , x3.k ,, x i.k ,, xn.k , which will represent the values x at the end of every phase.
If these values are expressed with the aid of relative thickness of strata, then it is
obtained
x i .k =
e1.1 + e1.2 +...+ e1.i e0

;
e1
(11)
The projectile velocity is calculated with the aid of relation
v = sq
pdt = sq ( I I 0 ) ,
where:
I=
0 pdt - the impulse of gases pressure at the time t ;
I0 =
t0
pdt - the impulse of gases pressure at the time t 0 .
In the case of simple flegmatised powders, taking into account the supplementary
hypotheses, the I impulse can be expressed in following way:
I = I k .1
e1.1
e1
+ I k .2
e1.2
e1
+ ... + I k .i
ei
.
e1
Using the relations (11), the I impulse can be written as
I = I k .1 z1.k + I k .2 ( x2.k x1.k ) + ... + I k .i ( xi xi 1.k ) .

The impulse I 0 is expressed with the aid of relation
I 0 = I k .1 z 0 .
Taking into account both relations of the impulses I and I 0 , the projectile velocity in
i phase is written
vi = sq I med .i + sq I k .i ( xi xi 1.k ) ,
(12)
where
I med .i = I k .1 x1.k + I k .2 ( x2.k x1.k ) + ... + I k .i 1 ( xi 1.k xi 2.k ) .

The fundamental equation of interior ballistics for flegmatised powder, in i phase, will
be
759
Czech Republic
sp(l + l i ) = f med . i f i + f i i
2006
qvi2
2
[Content]
(13)
where
f med . i = f1 1. k + f 2 ( 2. k 1. k ) + ... + f i (1 i 1. k ) ;
1.k , 2.k ,..., i1.k - the fraction of burned powder at the end of first phase, second
phase,, (i 1) th phase.
The differential equation of projectile displacement inside of barrel has the following
aspect:
dli
l +l i
qvi d vi
f med . i f i + f i i
qv 2 .
i
2
Taking into account the relations of vi and i . versus xi , the differential equation of
projectile displacement inside of barrel becomes
dl i
l + li
( A2.i + Bi xi )dxi
,
Ai + A1.i x i B1.i xi2
(14)
where:
Bi =
s 2 I k .2i
;
fi q
B1.i =
Bi
2
Ai = F 1.i + 0 Bmed .i 2 + B2.i xi 1.k Bi xi21.k

Bmed .i = s
2 2
I med .i
fi q ;
F1.i =
f med .i
fi
2;
1;
A1.i = k1 B 2.i + Bi xi 1.k ;

B2.i = s
I k .i I med .i
fi q ;
A2.i = B 2.i B i xi 1.k .
Taking into account that l = l med .i , by integration of the relation (14) for l i , since
l i1.k until l i and for xi , since xi 1.k until xi , it is obtained
Bi
l i = (l med . i + l i 1. k ) Z x B1. i l med . i ,
(15)
where:
l med .i = l 0 [1 (1 1 )1.k ( 2 1 )( 2.k 1.k ) ...

( i 1 1 )( i 1.k i 2.k ) ( i 1 )
Zx = (x
xi xi .1
i 1.k xi .1
760
A1.i ( bi +1) A2.i B1.i

+
2
Bi
bi A1.i
(x
xi xi .2
i 1.k xi .2
A1.i ( bi 1) A2.i B1.i
2
Bi
bi A1.i
i i 1.k
];
2
xi.1 =
A1.i (1+bi )
;
2 B1.i
xi.2 =
A1.i (1bi )
;
2 B1.i
Czech Republic
bi = 1 +
4 B1.i Ai
A12.i
2006
[Content]
The gases pressure results from the fundamental equation of interior ballistics, which is
written in the form
f i ( Ai + A1.i xi B1.i xi2 )

pi =
,
s (l + l i )
(16)
in which
l =l 0 [1 (1 1 )1.k ... ( i 1 )( i i 1.k ) .

For the determination of the maximum pressure value and its position inside of the
barrel it is used the relation
xi.m =
pi. max ( i 1 )k1
A1.i A 2.i +
fi
pi. max ( i 1 )
]
Bi ( + 1) 211[1 +
fi
which is solved by successive approximation, taking for pi. max in first approximation a
close value of expected maximum pressure.
If xi.m [ xi 1,k , xi.k ] , then the maximum pressure is achieved in this phase, in
contrary case it will achieve in following phases of I period.
4.
RELATIONS FOR THE SECOND PERIOD
The ballistic magnitudes p and v , versus l , in second period of firing phenomenon. are
calculated with the aid of the relations that are presented in [1, 2, 4].
The values of main ballistic magnitudes at the beginning of this period will be the same
values at the end of first period, namely: = k = 1; v = vk ; p = p k ; l = l k ;
l = l 1 = l 0 (1 ) ; f = f med .n .
The powder force f = f med .n is calculated on the bases of relation
f med .n = f11.k + f 2 ( 2.k 1.k ) + ... + f i ( i.k i 1.k ) + f n (1 n ) . (17)

With the aid of presented relations can be establish the way of variation of gases
pressure and projectile velocity versus its displacement inside of barrel.
761
5.
Czech Republic
2006
[Content]
THE FUNDAMENTAL PROBLEM SOLVING VERSUS TIME
Can be establish the time of projectile motion inside of barrel following integration of
relation
dt = 1 dl ,
v
(18)
since 0 until t , for t and since 0 until l , for l .

It is observed that for inferior limit of integral, when l = 0 and v = 0 , the integral is
indefinite, the function 1 v is infinite. In order to avoid this thing, it is considered a space
l that is not zero, but is very small and that corresponds at the time t .
If it is admitted the hypothesis that on the space l the projectile velocity varies linear,
then the time t is calculated with the aim of relation
t =
l = 2l .
vmed
v
(19)
Further, in order to determine the time of projectile motion inside of barrel, it is

integrated the relation (18) using a approximate method for evaluation of surface below the
curve f (l) = 1 v , thus:
n
l
l
t = l 1 dl = ( 1 + 1 ) i + 1 i .
l v
v vi + 1
2
i =1 i
(20)
In this way it is obtained the dependence t = t (l) . On the bases of this dependence and
the established dependencies v = v(l) and p = p(l) , can be draw the curves of variation of
gases pressure and projectile velocity versus time.
6.
RESULTS AND CONCLUSIONS
In the Fig.1 and Fig.2 it is presented the variation of gases pressure versus displacement
and time, obtained with the aid of the interior ballistics software that was elaborated on the
bases of proposed new algorithm, as well as usually algorithm, for an extant gun.
From the analysis of the data that are presented in the diagrams from Fig.1 and Fig.2,
can be concluded that there are some diferences of the fundamental problem solving of
interior ballistics in the case of flegmatised powders with the aid of both algorithms.
From the comparing of the experimental data with the theoretical values obtained in the
case of the proposed new algorithm had resulted that the difference between the muzzle
velocities values is approximately 0,3% and between the maximum pressures values is
approximately 5%.
762
Czech Republic
2006
[Content]
250000
200000
150000
Serie1
100000
Serie2
50000
0
0
10
20
30
Fig.1 The variation of pressure versus displacement

250000
200000
150000
Serie1
100000
Serie2
50000
0
0,000
0,005
0,010
Fig.2 The variation of pressure versus time
REFERENCES
[1]
[2]
E.V. CIURBANOV: Vnutrenniaia Ballistika, Izdatelstvo VAOLKA, Leningrad, 1975.

M.E. SEREBRIAKOV: Internal ballistics of guns and solid propellant rocket engines, vol. 1, 2,
Military Academy Publishing House, Bucharest, 1970.
[3]
I.D. SATNIK: Modifitirovannoi method reseniia osnovoi zadaci vnutrennei ballistiki dlia
zariadov iz porohov s peremennami termodinamiceskimi parametrami, Sbornik trudov
4-oi mejdunarodnoi conferentii, Kiev, 2000, pg. 206-221.
[4]
T. VASILE: Internal ballistics of guns, vol.1, II, Military Academy Publishing House,
Bucharest, 1993, 1996.
T. VASILE: Contributions to the fundamental problem solving of interior ballistics,
PROCEEDINGS of the II-nd International Conference Artillery Barrel Systems, Ammunition,
Means of Artillery Reconnaissance and Fire Control, October, 27-29, 1998, Kiev, pg.440-445.
T. VASILE: Studies and researches regarding the fundamental problem solving of interior
ballistics for weapon with grooved barrel, PROCEEDINGS of III-rd International Armament
Conference on Scientific Aspect of Armament Technology, October, 11-13, 2000, Waplewo,
pg.157-165.
T. VASILE, C. BARBU and D. SAFTA: The Fundamental Problem Solving of Interior
Ballistics Using the Phisical Law of Powder Burning, Academic and Applied Research in
Military Science Journal, Vol. 3, ISSN 1588-8789, 2004, Budapest, pg. 407-414.
[5]
[6]
[7]
763
Czech Republic
2006
[Content]
NUMERICAL DESCRIPTION OF THE EFFECTS IN GAS

GENERATOR
T. Wolszakiewicz*, T. Gawor*, Z. Walenta**
* - Institute of Organic Industrial Chemistry, Annopol 6, 03 236 Warsaw, Poland
** Institute of Fundamental Technological Research, Polish Academy of Science
Swietokrzyska 21, 00-049 Warszawa, Poland
Abstract
Gas generators are designed to produce gases of the proper pressure and temperature
within the precise time. Thus, the authors have undertaken to make numerical
calculations of the pressure and temperature course as well as density changes of the
researched system adopting the direct Monte Carlo method. What is more, the authorial
software written in Fortran language and controlled by Linux system has been used for
the calculations.
Keyword : gas genarator, Monte Carlo simulation
INTRODUCTION
The paper presents the numerical calculations of the physical chemical effects
observed during the gas generators work. Gas generator system works to provide the
constant gas emission at the accurate operation parameters that allows the proper operation
within the power transmission systems work. Until now, the applied calculation methods
have not inquired into the structure of local pressure, temperature and density changes of
gases. Thus, the numerical calculations the authors have undertaken to make are to enhance
the system.
RESULTS
The following simplifications in the calculations have been applied:
-
Monatomic gas has been accepted.
Flat frontage of burning has been accepted. The existence of the expanded fuel
form in the initial step of burning has not been taken into consideration.
The initial number of molecules has been 3.5 million. However, the given
number has risen up to 10.2 million after the ending of the calculations.
Full particle adjustment to chambers walls has been assume.
Molecules have been emitted randomly on the surface of burning while the
speeds have been consistent with Maxwell distribution (that is precisely beyond
the speeds backward to the propellant).
Even though the making of calculations lasted a long time (14 days non-stop), slight
percentage of the whole process has been made. The real time of gas generators work has
been from 12 to 14 seconds, and the presented results have described the first 5 s. Despite
the applied simplifications, the character of the effects and local changes of gradients have
been consistent with the physical nature of the effect. The authors present the initial results
and perform further calculations.
764
200.00
Czech Republic
2006
[Content]
200.00
150.00
150.00
100.00
100.00
50.00
50.00
-80.00 -60.00 -40.00 -20.00
Fig. 1a. Change of pressure after 20 steps
-80.00 -60.00 -40.00 -20.00
Fig. 1b. Change of pressure after 40 steps
Fig. 1a and 1b present the changes of pressure in the gas generators burning chamber.
The greater amount of generated molecules and the increase of local pressure have become
visible after the 40 steps of calculations.
200.00
200.00
150.00
150.00
100.00
100.00
50.00
50.00
-80.00 -60.00 -40.00 -20.00
Fig. 2a Change of density after 40 steps
-80.00 -60.00 -40.00 -20.00
Fig. 2b. Change of density after 60 steps
Tomasz Wolszakiewicz, Tomasz Gawor, Zbygniew Walenta
765
Czech Republic
2006
[Content]
Fig. 2a and 2b present the changes of gas densities that flow in the gas generator. The
applied units have been the relative values. At the initial stage of burning, the density has
been comparable to that on the front of burning. The density has been getting its maximum
values in the contraction section (about 4.3 to 6 of initial value), which is consistent with
physical predictions.
temp. otoczenia)
12.00
200.00
Temperatura (krotno
150.00
100.00
8.00
4.00
50.00
0.00
0.00
50.00
100.00
150.00
200.00
250.00
Przekrj komory
-80.00 -60.00 -40.00 -20.00
Fig. 3 Temperature changes
Fig. 4 Gas temp. changes in the function of burning chamber section
3.00
Zmiana g sto ci
2.00
1.00
0.00
0.00
50.00
100.00
150.00
200.00
250.00
Odlego od czoa palenia
Fig. 5 Gas densities in the function of burning chamber section
766
Czech Republic
2006
[Content]
CONCLUSIONS
1. The losses of energy connected with the flows are insignificant at such selection
of engine dimensions, and their occurrences are related with walls cooling.
2. Too intensive walls cooling answers for the decrease in temperature.
3. There is the subsonic flow in the chamber area whereas the outflow from the
contraction that is at the end of the sleeve reveals the supersonic character.
4. For the sake of the linear pressure maintenance in a time function (in the
chamber and the contraction), the management of the whole engine will depend
on the speed of gas emission and not on the speed of flow, which means that the
form of a jet has a decisive role in the process.
Tomasz Wolszakiewicz, Tomasz Gawor, Zbygniew Walenta
767
768
100
unit
(16)
25
25
12.5
(20)
9 mm
125 unit
mm)
28.5
0.8 mm
mmm
2.5 unit
Czech Republic
2006
Length unit = 0.16 mm

Time unit = 0.04 s
Speed of gas outlet = 9.8 m/s
One cell volume = 0.064 * 10-10 m3
10 molecules has been initially introduced in each cell
80.5
Burning front
Schematic diagram of gas generator calculations area

[Content]
Czech Republic
2006
[Content]
THE MECHANISMS OF THERMAL DECOMPOSITION OF SIMPLE

HYDRONITROGEN COMPOUNDS
Ying XIONG*, Yuanjie SHU1*, Ge ZHOU**, Xinfeng WANG*,
Anming TIAN**, and Ning-Bew WONG***
(* Institute of Chemical Materials, CAEP, 621900Mianyang, China
** Department of Chemistry, Sichuan University, Chendu, China
*** Department of Biology and Chemistry, City University of Hong Kong,
Kowloon, Hong Kong)
Abstract:
Using Ab Initio Molecular Dynamics (AIMD) method and Density Functional Theory
(DFT), the mechanisms of thermal decompositions of five kinds of hydronitrogen
compounds were studied, which are N2H2, N2H4, N3H, N4H2, N4H4. For the ab initio
molecular dynamics studies on the decomposition of the five hydronitrogen compounds,
dissociation trajectories are simulated at different temperatures using the VASP
package. Based on these results, further calculations of minimum points and transition
state structures are performed using Gaussian 03 at B3LYP/6-31G(d,p) level, whereas
CCSD(T)=full/6-311+G(3df,2p) are used for the high accuracy single point calculation.
For N2H4, the main decomposition pathway is N-H bond rupture which is further
verified by rate constant calculation. For N2H4, N3H, N4H2 and N4H4, they decompose
mainly through different N-N bond dissociation. According to the result by B3LYP, the
thermal stability increases by N4H2 < N4H4 < N2H2 < N2H4 < N3H. Nevertheless, at
CCSD(T) level, the order is N4H2 < N4H4 < N3H < N2H2 < N2H4. Both the two
theoretical levels show that the energy barrier of N4H2 dissociation is too low for using
as structure unit of energetic materials. Our work provided some useful information for
further studying the mechanisms of thermal decompositions of various complex high
nitrogen energetic materials theoretically and experimentally..
Keywords:
1.
hydronitrogen compounds, density functional theory (DFT), ab initio

molecular dynamics (AIMD), mechanisms, thermal decomposition
INTRODUCTION
In the past 20 years, the compound containing N atoms has been extensively studied as
possible candidate for high energy density materials (HEDMs), due to the dissociation of the
compound containing N atoms with the release of large amount of energy through the
slightest input of external energy, for example pressure, impact, or heat, which can be used
as propellants and explosives[1-12]. However, as a kind of good candidate for HEDMs, it is
necessary to learn the thermal decomposition mechanism in detail. So, thermolysis behavior
of the compound containing N becomes the focus of recent experimental and theoretical
research [13]. There have been many theoretical studies at various computational levels on the
compound containing nitrogen atoms [14-16]. In all kinds of the compounds containing N
atoms, the hydronitrogen compound received considerable attentions [17,18]. Industrial interest
in organic compounds containing N atoms began with the use of azides for the synthesis of
heterocycles such as triazoles and tetrazoles as well as with them as blowing agents and as
1. E-mail:syjfree@sina.com
1
Ying Xiong, Yuanjie Shu, Ge Zhou, Xinfeng Wang, Anming Tian, . . .
769
Czech Republic
2006
[Content]
functional groups in pharmaceuticals [19]. In this work, some simply hydronitrogen molecules
are selected as the model molecules (see Fig.1). Their thermal decomposition behavior are
studied theoretically. Ab inito molecular dynamics (AIMD) was used to discuss the
decomposition process.The doublet potential energy surfaces had been obtained and the rate
constants were calculated to confirm the mechanism of thermal decomposition. Results of
pyrolysis are compared for the small model compounds which contain N-N, N=N, and azide
group.
N2H2
N2H4
Fig 1.
2.
N3H
N4H2
N4H4
Selected simple hydronitrogen compounds.
The principles of ab initio MD (AIMD) method have been described in [20-24]. The
Vienna Ab Initio Simulation Package (VASP) program was used for the thermal
decomposition trajectory studies since it uses a rather traditional self-consistency scheme to
evaluate the instantaneous electronic groundstate at each MD step. The molecule being
simulated was put in a cubic box with a length of 8 to imitate gas phase condition. The
PerdewWang [25] gradient correction was added to the exchange-correlation functional [26].
For the core region, the optimized Vanderbilt ultrasoft pseudopotentials [27] supplied with the
VASP package [28, 29] were used for N and H atoms. For the trajectory study, a preliminary
equilibrating simulation was first performed with micro canonical ensemble, starting with
the ground-state equilibrium geometry. The time-step used was 0.5 fs, and the length of the
simulation was 5 ps. One hundred geometrical configurations were selected randomly from
the equilibrium results as the starting structures for the trajectory study. The resulted system
was then used as the initial state for the simulations run at higher temprature with canonical
ensemble. Each trajectory lasted 3 ps with the time-step of 0.5 fs (6000 steps). A trajectory
was terminated after 6000 steps if no reaction was observed [30].
The local energy minimum structures predicted in the trajectory study and the
corresponding transitional state structures have been fully optimized by the analytic gradient
techniques using the density functional theory with Becke s three parameter (B3) [31]
exchange functional along with the LeeYangParr (LYP) nonlocal-correlation functional
[32, 33]
(B3LYP). The standard basis set 6-31G(d,p) was used in conjunction with the DFT
method. Vibrational frequencies were also calculated at the B3LYP/6-31G(d,p) level of
theory to take account of the zerro point energy and identify the transition structures and the
local minimum structures. The coupled cluster method CCSD(T)=full/6-311+G(3df,2p))was
used for the single-point energy calculations in order to ensure the reliable energy
predictions. The GAUSSIAN 03 program package [34] was used in the calculations. For the
decomposition pathways with similar energetic barriers, the rate constants were investigated
with Vklab program, where the required energy and potential surface were abtained at
B3LYP/6-31G(d,p) level.For direct dissociation, the VT ensemble was adopted, and for
reaction through TS, the translation state theory (TST) was used.
3.
770
The Mechanisms of Thermal Decomposition of Simple Hydronitrogen . . .
Czech Republic
2006
[Content]
The selected model molecules are N2H2, N2H4, N3H, N4H2, N4H4 (Fig 1). The details of
thermal decomposition pathways of each compound are discussed below.
3.1
N 2H 2
The preliminary equilibrating simulation was performed at 6000K, and the trajectory
study was performed at 6500K. For each 100 trajectories, 71 decomposed at 6500K. Out of
the 71 reactions, 69 followed channel 1: N-H cleavage; 2 followed channel 2: hydrogen
migration. The two different reaction channels observed in the MD trajectories are listed in
Table 1.
Table 1. The decomposition paths of N2H2 observed in the 100 trajectories at
6500K
Path
1
Reaction Observed
N
3
Total
N
N
H H
Freqs
69
2
N
No reaction
100
29
The geometrical parameters of the various transition states and local minima are
depicted in Fig.2 and their energy properties are summarized in Table 2.
1.040
1.021
1.246
1.016
N1.040
N2 H2
1.436
1.016
N 1.021
N 1.385
N 1.385 H N1.217 N 1.050
1.013 N 1.249
N 1.013
H
1.05
1.020
H
H
M1
1.018
1.018 H
1.045
N
H
M4
NH3
M5
1.033
1.030
N 1.453 N 1.044
H
0.743
1.106
1.044
H H
N2
M3
M2
1.184 1.184
N4H2
1.021
H 1.033
N4H4
1.018
N3H
N2H4
1.020
1.035 N 1.489 N 1.035 H

H 1.237 N 1.237
N 1.139N 1.242 N 1.022

H
1.060 N 1.180
1.182
H2
N 1.422 H
N 1.014
1.020
M6
M7
H
N1.052 H
2.113
1.017
N1.017H
1.017
M8
771
H
1.111
N
1.122
1.550
H
1.062
H
TS1
1.341
1.280
Czech Republic
1.457 H 1.079 H
N 1.012
N
1.013
1.033 1.655
TS3
Fig 2.
1.154 N 1.606
H
N 1.022
1.025
TS4
TS5
[Content]
TS2
H 1.603 1.304 1.017

N
N
1.019
H 1.240
H
2006
3.105
1.0490 H
1.017
N1.017
H
1.017
TS6
Bond lengths (in ) of the minimum structures the transition states
Table 2. The energy and the zero-point correction of the minimum structures
and the transition states (in hartree)
N2H2
N2H4
N3H
N4H2
N4H4
M1
M2
M3
N2
NH3
M4
M5
M6
M7
M8
H
H2
TS1
TS2
TS3
TS4
TS5
TS6
EZP
B3LYP/6-31G(d,p) CCSD(T)=full/6-311+G(3df,2p)
0.028265
-110.643011
-110.522547
0.053304
-111.868658
-111.742141
0.021417
-164.785840
-164.609540
0.037854
-220.081037
-219.847144
0.063210
-221.335428
-221.089118
0.026569
-110.607411
-110.482821
-55.878981
-55.814154
0.018971
0.051281
-111.790888
-111.670299
0.005599
-109.524129
-109.419147
0.034441
-56.557769
-56.494011
0.007485
-55.137670
-55.090147
0.009601
-164.138543
-163.959949
0.013261
-110.032726
-109.910763
0.032552
-165.370432
-165.185219
0.044807
-111.705640
-111.591533
0.00000
-0.500273
-0.499810
-1.178539
-1.170823
0.010174
0.006593
-110.013307
-109.895261
0.020697
-110.522739
-110.400769
-111.761264
-111.634122
0.046768
0.030419
-165.368462
-165.184272
-111.747464
-111.609185
0.041514
0.044419
-111.705531
-111.591074
In Fig.3, the calculated energy profiles at B3LYP/6-31G(d,p) and CCSD(T)=full/6311+G(3df,2p) are presented. In channel 1, the energy barriers for the first N-H bond
breaking step are 249.45 kJ/mol and 254.59 kJ/mol respectively. The products, consisting of
N2 and H2, lie significantly below N2H2. The hydrogen migration pathway is shown in
channel 2. A relatively large energy amount of 295.90 KJ/mol and 299.86 kJ/mol is required
772
Czech Republic
2006
[Content]
to transfer H from one nitrogen to another. Accordingly, the direct N-H bond cleavage is the
main decomposition pathway, which is consistent with the result predicted by the ab initio
MD simulation.
Channel 1: The cleavage of the N-H bond
N
N
N
33.48
H 23.19
Channel 2: Hydrogen migration
31.74
64.82
249.45
254.59
N
H
206.89
200.01
295.90
299.86
H N
440.61
422.78
N H
H N
N
Fig 3.
3.2
NH
H
N N
H
N H
The reaction channels of the thermal decomposition of N2H2. Energy

barriers are calculated at B3LYP/6-31G(d,p) and CCSD(T)=full/6311+G(3df,2p) (underlined) levels, corrected by the zero-point energies (in
kJ/mol)
N 2H 4
the 69 reactions, 52 followed channel 1: N-N bond cleavage; 9 followed channel 2: hydrogen
migration; 8 follow channel 3: H2 elimination. In channel 2, N2H4 is easy to tautomerize to
NHNH3, but it can usually recover to N2H4 form. Three different reaction channels observed
in the MD trajectories are listed in Table 3.
5500K
Path
1
Reaction Observed
H
H
N
N
H
N
H
H
H
N
H
N
N
H
N + NH3
H
H N
H
Freqs
52
H
N
4
Total
H
H
H
H
H
N
H H
N
H
No reaction
31
100
The three main reaction pathways and their energy barriers are shown in Fig.4. The
energy barrier of N-N bond split is 250.3 kJ/mol at B3LYP/6-31G(d,p) level and 258.54
kJ/mol at CCSD(T)=full/6-311+G(3df,2p) level in channel 1, which is similar to the energy
barrier of N-N bond split in N2H4. The slightly higher energy is required for proton transfer
5
773
Czech Republic
2006
[Content]
in channel 2 (264.81 kJ/mol at DFT level and 266.44 kJ/mol at CCSD(T) level). However,
the decomposition of the intermediate need much higher energy than that of back to the
reactant. The previously studies [35] carried using MP2 and QCISD(T) show that the
intermediate NHNH3 can exit stably in term of thermodynamics ,which is in agreement with
our result from quantum chemistry calculation, but inconsistent with the conclusion by
AIMD. In order to confirm the dominant decomposition mechanism, we further performed
rate constant calculation for channel 1 and channel 2 with Vklab program.The rate constants
obtained are shown in Table 4.We found the rate constant of N-N cleavage is at least 600
times larger than that of hydrogen migration. Moreover,the reverse rate constant of hydrogen
migration is more than 106 times larger than the forward one.Consequently, channel 2 is not
easy to take place, as observed in the trajectories simulation. In channel 3, the energy barrier
for H2 elimination is the highest (287.24 kJ/mol at B3LYP level and 318.12 kJ/mol at
CCSD(T) level), so it is also a minor way, as predicted by ab initio MD simulation.
Channel 1: The cleavage of the N-H bond
H
Channel 2: Hydrogen migration

H
H
N
250.30
258.54
206.09
189.99
65.93
83.13
H
H
N
264.81
266.44
N
H
NH3 N
-0.74
0.19
19.22
11.80
H N
H
H
N
N
H
Channel 3: H2 elimination
H
H N
H
N
H
113.57
129.25
287.24
318.12
H
H
N
H
H
N
Fig 4.
774
N
H

kJ/mol).
Czech Republic
2006
[Content]
Table 4. The rate constants of N-N cleavage (channel 1) and hydrogen migration
(channel 2)
T(K)
k(s-1)
H migrationforwardTST H migrationreverseTST
1.157E-19
1.058E+06
5.280E-16
7.511E+06
4.532E-13
3.657E+07
1.152E-10
1.350E+08
1.173E-08
4.047E+08
5.901E-07
1.031E+09
1.706E-05
2.312E+09
3.162E-04
4.673E+09
4.083E-03
8.680E+09
3.913E-02
1.503E+10
2.925E-01
2.454E+10
1.773E+00
3.813E+10
8.987E+00
5.676E+10
1.490E+02
1.132E+11
1.553E+03
2.019E+11
1.133E+04
3.302E+11
6.235E+04
5.042E+11
2.740E+05
7.284E+11
N-N cleavageVT
5.465E-13
6.934E-10
2.135E-07
2.339E-05
1.176E-03
3.247E-02
5.583E-01
6.570E+00
5.679E+01
3.804E+02
2.061E+03
9.331E+03
3.627E+04
3.767E+05
2.632E+06
1.356E+07
5.492E+07
1.836E+08
400
450
500
550
600
650
700
750
800
850
900
950
1000
1100
1200
1300
1400
1500
3.3
N 3H
the 51 reactions, 45 followed channel 1: N2-N3 bond cleavage, 6 follow channel 2: N2-H1
bond cleavage. Two different reaction channels observed in the MD trajectories are listed in
Table 5.
Table 5. The decomposition paths of N3H observed in the 100 trajectories at 7000K
Path
1
2
Reaction Observed
4
N
4
N
3
N
1
Freqs
45
N
1
3
Total
No reaction
49
100
The geometry parameters of the reactant and the intermediates are depicted in Fig.2 and
their energy properties are summarized in Table 2. The two main reaction pathways and their
energy barriers are shown in Fig.5. the energy barrier for N2-N3 bond breaking is 303.79
kJ/mol at B3LYP level and 241.32 kJ/mol at CCSD(T) level. In channel 2, the cleavage of
775
Czech Republic
2006
[Content]
N2-H1 need higher energy (354.99 kJ/mol and 362.23 kJ/mol respectively), so it is minor in
the decomposition of N3H, as predicted by using VASP.
Channel 1: The cleavage of the N2-N3 bond. Channel 2: The cleavage of the N-H bond
N
N N N H
303.79
241.32
354.99
362.23
Fig 5.
3.4
The reaction channels of the thermal decomposition of N3H. Energy

kJ/mol).
N 4H 2
study was performed at 1800K. For each 100 trajectories, 93 decomposed at 1800K. The 93
reactions followed the same way at first step: N2-N3 bond cleavage (Table 6). From the
trajectory simulation, we found N2-N3 is very easy to break.
1800K
Path
1
Reaction Observed
4
N
N
N
N
N
H
N N H H
N N
Freqs
57
28
N N
N
H
N
H
4
Total
N
H
N
N
No reaction
100
H
N
H
N
N
H
The first step of decomposition was further examined by well-established ab initio MO

method. The geometry parameters of reactant and the intermediates are depicted in fig.2.
Their energy properties are summarized in Table 2. The results verify the energy barrier for
N2-N3 breaking is quite low (only 11.17kJ/mol at B3LYP/6-31g(d,p) level and 37.51kJ/mol
at CCSD(T)=full/6-311+G(3df,2p) level) (see Fig.6).
776
Czech Republic
H
N
N
2006
[Content]
11.17
37.51
Fig 6.
3.5

kJ/mol).
N 4H 4
study was performed at 3800K. For each 100 trajectories, 98 decomposed at 3800K. The 98
reactions followed the same way at first step: N2-N3 bond cleavage (Table 7).
3800K
Path
1
Reaction Observed
H
N
N
N
N
N
N
H
N
1 H
1 H
N
3
N
N
H
Freqs
96
N
H
4
Total
H
H
N N
H
N
N
H
No reaction
N
H
2
100
We also further examined the first step of decomposition with ab initio MO method. The
geometry parameters of reactant and the intermediates are depicted in fig.2. Their energy
properties are summarized in Table 6. The energy barrier for N2-N3 breaking is
195.15kJ/mol at B3LYP/6-31g(d,p) level and 204.94kJ/mol at CCSD(T)/6-311+G(3df,2p)
level) (see Fig.7).
H
N
N
195.15
204.94
H
N
H
Fig 7.
N
N
H
N
H

kJ/mol).
777
4.
Czech Republic
2006
[Content]
CONCLUSION
The combined ab initio MD and ab intio MO study has revealed the dissociation
mechanisms for some simple hydronitrogen compounds. For N2H2, the main decomposition
pathway is N-H bond rupture, which need 249.45 kJ/mol in energy at B3LYP/6-31G(d,p)
level and 254.59 kJ/mol at CCSD(T)/6-311+G(3df,2p) level. For N2H4, N3H, N4H2 and
N4H4, they decompose mainly through different N-N bond breaking. The energy barriers are
250.30 kJ/mol, 303.79 kJ/mol, 11.17 kJ/mol and 195.15 kJ/mol respectively at B3LYP/631G(d,p) level and 258.54 kJ/mol, 241.32 kJ/mol, 37.51 kJ/mol and 204.94 kJ/mol at
CCSD(T)/6-311+G(3df,2p) level. According to the result by B3LYP, the thermal stability
increases by N4H2 < N4H4 < N2H2 < N2H4 < N3H. Nevertheless, at CCSD(T) level, the order
is N4H2 < N4H4 < N3H < N2H2 < N2H4. Both the two theoretical levels show that the energy
barrier of N4H2 dissociation is too low for using as structure unit of energetic materials. Our
further work is to add N4H2, N4H4, N2H4, N2H2 to heterocyclic nitrogen compounds as
bridges or substitute groups to construct some potential high nitrogen energetic materials
with good properties.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
778
Van der putten NHeijdentijk D and Schenk H. Tetrazole, Cryst. Struct. Commun. ,1974, 3:
321
Lesnikovich A I,Levchik S V.Balabanovich A I,Ivashkevich O A,Gaponik P N. The thermal
decomposition of tetrazoles, Thermochimica Acta,1992,200:427.
Thomas M.Klapke, Peter Mayer, Axel Schulz and Jan J. Weigand, 1,4-Bis-[1-Methyltetrazol5-yl]-1,4-Dimethyl-2-Tetrazene: A Stable, Highly Energetic Hexamer of Diazomethane
(CH2N2)6, Propellants, Explosives, Pyrotechnics 29 (2004), No.6:325-332
Wong M W. Chem. Phys.Lett.,1996,256:391.
Yoshiaki A,Yoshio O,Masamitsu T.Proc. Int.Pyrotech.Semin.,1996,22nd :13-22
Anton Hammerl, T.M.Klaptke,G.Holl , Peter Mayer, and Jan J. Weigand, Synthesis, Structure,
Molecular Orbital Calculations andDecomposition Mechanism for Tetrazolylazide CHN7, its
Phenyl Derivative PhCN7 and Tetrazolylpentazole CHN9, Propellants, Explosives,
Pyrotechnics 30 (2005), No. 1
Kathleen B , Isabella L K. The crystal and molecular structure of 5-aminotetrazole
monohydrate, [ J ] . Acta Crystallography ,1967 ,22 : 308.
Mary M W ,William S M , Ronald A H. The heats of combustion of substituted triazoles and
related high nitrogen compounds, [ J ] . Journal of Physics & Chemstry ,1957 ,61 : 261.
Leonard E B. Electro impact2induced fragmentation of 5-aminotetrazole, [J ] . Journal of
Heterocycle Chemistry , 1970 ,7 (6) : 1223.
Levchik S V , Ivashkevich O A ,Balabanovich A I ,et al. Thermal decomposition of 5aminotetrazoles, [J ] . Thermochimica Acta ,1992 ,207 : 115.
My-Hang V. Huynh, Michael A. Hiskey, Ernest L. Hartline, Polyazido High-Nitrogen
Compounds: Hydrazo and Azo-1,3,5-triazine, Angew. Chem. Int. Ed. 2004, 43, 4924 4928,
D.E. Chavez, M.A. Hiskey, J. Energetic Mater. 17 (1999) 357.
Hiskey M, Chavez D, Naud D. Progress in high-nitrogen chemistry in explosives ,propellants
and pyrotechnics, [A] . Proceedings of 27th International Pyrotechnics Seminar [C] ,July16-21
,USA ,Colorado ,2000.3-14.
Y. Oyumi,T . B. Brill, Thermal Decomposition of Energetic Materials 4. High-Rate,In Situ
Thermolysis of the Four,Six, and Eight Membered,Oxygen-Rich, Gem-Dinitroalkyl Cyclic
Nitramines,TN AZ, DNNC, and HNDZ, Combust. Flame 1985, 62,225.
Czech Republic
2006
[Content]
[15] T. B. Brill,Connecting the Chemical Composition of a Material to its Combustion

Characteristics, Prog. Energy Combust. Sci. 1992, 18,91.
[16] Brian D. Roos and Thomas B. Brill, Thermal Decomposition of Energetic Materials 84:
Pyrolysis of 5-Substituted 1,3,5-Trinitrohexahydropyrimidines, Propellants, Explosives,
Pyrotechnics 28 (2003), No . 2
[17] P. A. S. Smith, Org. React. 1946, 3, 337 349.
[18] J. H. Boyer, F. C. Canter, Chem. Rev. 1954, 54, 1 57.
[19] T. S. Lin, W. H. Prusoff, J. Med. Chem. 1978, 21, 109 112.
[20] R. Car, M. Parrinello, Phys. Rev. Lett. 55 (1985) 247.
[21] M. Parrinello, Solid State Commun. 102 (1997) 107.
[22] D.K. Remler, P.S. Madden, Mol. Phys. 70 (1990) 921.
[23] M.C. Payne, M.P. Teter, D.C. Allan, T.A. Arias, J.D.
[24] M.E. Tuckerman, P.J. Ungar, T. Vonrosenvinge, M.L.Klein, J. Phys. Chem. 100 (1996) 12878.
[25] J.P. Perdew, in: P. Ziesche, H. Eschrig (Eds.), Electronic Structure of Solids 91, Academie
Verlag, Berlin, 1991, p. 11.
[26] J.P. Perdew, A. Zunger, Phys. Rev. B 23 (1981) 5084.
[27] D. Vanderbilt, Phys. Rev. B 41 (1990) 7892.
[28] G. Kresse, J. Hafner, J. Phys. 6 (1994) 8245.
[29] G. Kresse, J. Hafner, Phys. Rev. B 48 (1993) 13115.
[30] Jing Wang, Jiande Gu, Anmin Tian, The mechanisms of the thermal decomposition of 5-nitro1-hydrogen-tetrazole: ab initio MD and quantum chemistry studies, Chemical Physics Letters
351 (2002) 459-468
[31] A.D. Becke, J. Chem. Phys. 98 (1993) 5648
[32] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785
[33] B. Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys.Lett. 157 (1989) 200.
[34] M.J. Frisch et al., GAUSSIAN 98, Revision A.3, Gaussian,Pittsburgh, PA, 1998
[35] Ding Fujiang, Zhang Liangfu, Ab-inito Study on HNNH3, ACTA PHYSICO-CHIMICA
SINICA, Vol 12, No.11, Nov. 1996
11
779
Czech Republic
2006
[Content]
INFLUENCING PARAMETERS ON HE PROJECTILES WITH

NATURAL FRAGMENTATION
Berko Zecevic, Jasmin Terzic, Alan Catovic, Sabina Serdarevic-Kadic
Mechanical Engineering Faculty, Defense Technologies Department, Vilsonovo setaliste 9, 71000
Sarajevo, Bosnia and Herzegovina
Abstract:
Design of the HE warhead is process confronted with series of contradictory
requirements. Influence of warhead design on lethal efficiency is very complex. Lethal
efficiency of HE warhead depends on the form and dimension of the warhead, quantity
and type of explosive, warhead case material, warhead case thickness, fuse type,
explosive train etc. At warheads natural fragmentation, fragments geometry, their mass
and spatial distribution are functions of designed shape of the warhead case (shell),
mechanical properties of case material (tensile strength and yield strength) and
performances of explosive (physical and energetic). It is essential to have a capability to
make warhead performance predictions in the earliest phases of ammunition or
warheads preliminary design.
Keywords:
1.
warhead, natural fragmentation, fragment velocity, fragment size,

geometry of fragments, case material, design, lethality, explosive
charge, fragments spatial distribution.
INTRODUCTION
Artillery projectiles or rocket warheads are usually considered two-dimensional axial

symmetric. Natural fragmentation of projectiles or warheads results in wide range of random
distributions of fragment sizes (masses and geometries). Expansion of warhead case caused
by detonation products of explosive charge brings about a warhead structure being split into
various sized fragments [4].
HE warhead performances depend on its geometrical shape and dimensions, mass of
explosive charge and explosive type, material of warhead case, initiation way and initiation
point position, fuse type, round-to-round variations, etc. Lethal efficiency of HE warhead is
a function of fragments velocity, geometrical shape and mass of natural fragments and their
spatial distribution.
Fragment velocity depends on the explosive mass and warhead metal case mass ratio,
detonation velocity and explosive density.
Fragments spatial distribution around detonated cylindrical warhead is not uniform.
Naturally fragmented cylindrical warhead typically splits into long axially oriented strings.
Splitting effect of warhead case radial fracture depends on toughness, brittleness and
material structure grain size, explosive power (magnitude of the detonation impulse).
Further, these strings are broken up into ultimate fragments in both ways, radial and
longitudinal, during subsequent detonation products expansion, whose fragments mass
distribution can be described approximately by the Mott formula. Detonation products solicit
expansion of the warhead case greater than twice the warhead initial radius. For cylindrical
780
Czech Republic
2006
[Content]
steel warhead cases, initial elastic-plastic expansion of the case occurs when it is extended
from the original volume to about 1,44 times. When the current case volume being risen to
about 2,56 to 3,24 times of the initial warhead volume, the detonation products are released
through cracks and subsequently an expanding detonation products cloud is developed
beyond the fractured warhead case [9].
Fig 1.
Sequences of a HE warhead natural fragmentation

(www.nawcwpns.navy.mil/mov/energet/seg/WB.mov)
Natural fragments spatial distribution (including their geometrical shapes and masses) is
a complex function depending on internal and external geometry of the warhead case
surface, warhead case mechanical properties (tensile strength and yield strength) and
energetic characteristics of the explosive.
Geometrical shape of natural fragments, their mass and spatial distribution are functions
of geometrical forms of internal and external warhead case surfaces, mechanical properties
of warhead case material (tensile strength and yield strength) and performances of explosive
(physical and energetic).
It is essential to have a capability to make warhead performance prediction in the earliest
phases of HE ammunition or warheads preliminary design. Ability for warhead
performances prediction depends on comprehensive data base of warheads natural
fragmentation features, including data on fragment numbers, initial fragment velocities,
warhead case material performances, fragment shape features and spatial fragment
distributions, etc.
Measurements of warhead performances require very complex measuring equipment and
measuring process itself is expensive as well.
Capability of warheads performances and efficiency assessment is based on complexity
of our database of natural fragmentation parameters which should encompass data on
number, mass, initial velocity, fragments shape factor and spatial distribution, material
characteristics of warhead case and explosive charge, etc.
Researches performed in USA were aimed on development of simulation methods for
prediction of fragmentation characteristics of HE ammunition [5]. The recent attainment is
development of CALE computer simulation program, which is able to simulate
fragmentation performances of two and three - dimensional axial symmetric warheads and
HE ammunition.
2.
TEST METHODS FOR HE PROJECTILES NATURAL

FRAGMENTATION ASSESSMENT
Parameters of natural fragmentation process are determined with analytical methods,

experimental researches and numerical modeling methods. Number of fragments, their mass,
geometrical shapes and spatial distribution are determined experimentally, with Pit test
method and Arena test method.
2
781
2.1
Czech Republic
2006
[Content]
Pit test
In Pit test, warhead is detonated in closed space (pitfall), filled with sand (Fig.2.). After
the fragmentation of warhead, fragments are obtained from the sand. Mass and shape of
fragments are determined, and fragments are classified by their mass groups.
Experimental researches performed by
authors were undertaken in order to estimate all
relevant performances of warhead natural
fragmentation (number, mass and fragments
shape) when the material of the warhead case is
changed (three types of steel with different
mechanical properties). Number, mass and
fragments shape of each mass group are
determined using the Pit test, and relative
estimation of spatial fragments efficiency was
done by Arena test.
Prediction of fragments mass distribution is
usually performed by application of Mott
formula, or Held formula. Each of mentioned
formulas has certain limitations. These empirical
formulas are based on experimental data gained
from many fragmentations in Pit and Arena
facilities.
Fig 2.
Pit test facility
2.1.1 Mott formula

The Mott equation has been used for many years for prediction of fragments mass
distribution in naturally fragmented warheads and ammunition, and this method is only one
used in U.S.A. [10]:
)]
N (m ) = M 0 2 M k2 e
m 0 , 5 M k
(1)
where: N(m) is fragments number with a mass greater than m, M0 is the total fragments mass
and Mk is parameter which characterizes fragments mass distribution. Parameter Mk is a
function of warhead case thickness ti, internal diameter of the warhead di and explosive
charge. In the parameter Mk, constant B depends on the explosive charge and casing material.
1
M k = B t i 6 d i3 (1 + t i d i )
(2)
For a mild steel case, the constant B obviously decreases with increasing of detonation
pressure (yielding smaller fragments); but it also decreases with increasing case hardness [8].
2.1.2 Held formula
For prediction of mass distribution of natural fragmented warhead, Held formula with
two parameters and the total mass M0 or the best fit of total mass M0Best as inputs, gives an
excellent description of the experimentally found mass distributions of a natural fragmented
warhead [6]. An improved fit to natural-fragmentation data can be obtained using equation:
M (n ) = M 0 1 e Bn
782
(3)
Czech Republic
2006
[Content]
where; B and are both empirically determined constants, with B = const. d / t and of
order 10-2 and of order 2/3. In Held equation M0 is the total mass of all fragments, M(n) and
n are cumulative fragments mass and cumulative fragments number beginning with the
heaviest fragment. Held found that it was necessary to discard a few of the heaviest
fragments in order to obtain a curve fit to data over the rest of the range. The constants B and
are determined from above equation by mathematical transformation:
[M 0 M (n )] M 0 = e Bn
(4)
and the natural logarithm of the above equation is:
ln[(M 0 M (n )) M 0 ] = B n
(5)
If the logarithm of above equation is performed again, it is possible to determine the

constants B and in the log-log plot. By differentiating of the equation, Held M(n) gave the
approximate mass of the n-th fragment:
m(n ) = dM (n ) dn = M 0 B n 1 e Bn
(6)
From the fragment mass distribution (FMD) log-log diagram, constant B and exponent
with correlation coefficient r2 are obtained. If in the log-log diagram, the straight line does
not fit the measuring data very well, given total mass M0 is not an optimum mass for such
fragments mass distribution. Now, an optimum mass (or best mass) M0Best is calculated:
M 0 Best = M (n ) 1 e Bn
(7)
The constants B and are determined based on original mass of projectile metal body.
The new constants BB and B are determined with the total mass M0Best
M (n ) = M 0 Best 1 e BB n
(8)
This procedure is repeated until a satisfactory correlation coefficient is obtained ( r 2 0,99 ).
2.2
Arena tests
Spatial fragments distributions were determined through Arena tests. At the arena
fragmentation test, fragments concentration density per steradian is measured when
projectile or warhead are put in horizontal position at a certain distance from the ground
(Fig.3).
Fig 3. Arena fragmentation test (measurement of the fragments concentration density per
steradian)
4
783
Czech Republic
2006
[Content]
After the warhead detonation, number of hits and perforations at wooden sectors are
counted and then hits number and perforations number per m2 are calculated for each sector.
From such obtained data, dependency of fragments concentration density per m2 as a
function of sector distance is established. Further, a characteristic distance, at which
fragments concentration density of 1 perforation per m2 is obtained, is determined from the
above function. This distance is called warhead efficiency radius, which means that warhead
with greater efficiency radius has corresponding greater lethal zone.
When it is necessary to compare lethal efficiencies for similar warheads, then a
projectile or a warhead is put on the ground in the vertical nose-down position (Fig.4).
Fig 4. Arena fragmentation test (measurement the fragments concentration density per m2)
2.2.1 Spatial distribution fragments

Fragments concentration density per arena sectors area at certain distance from the
warhead detonation center is given with formula:
d (Ri ) = N pen (Ri ) / A(Ri )
(9)
where Npen is number of perforations at arena sector made of wooden panels, A is the sectors
area and R is radial distance between the warhead explosion center and a certain sector.
Spatial distribution and density of effective fragments is determined experimentally,
with fragmentation of HE projectiles in Arena facility.
Arena consists of k semicircular
sectors, divided on nk segments, where
increment of polar angle (k) is constant.
As a result of conducted experiments,
number of penetrations is determined in
every segment, and for every sector of
Arena. Based on given results it is possible
to determine total number of fragments in
j-th polar zone and for every distance
(radius of sector). After the experiment,
number of fragments that penetrated the
target is obtained (nj, for j = 1-22). As the
surface of every segment (target) is equal,
we can determine the number of fragments
penetrating particular polar zone, with
following equation:
784
Fig 5. Spatial distribution of fragments
N jk = n jk
Czech Republic
2006
S jk
[Content]
(10)
Sk
where:
Njk - Scaled fragment density or total number of fragments in j-th polar zone of k-th sector
njk - Number of fragments (penetrating the targets) in j-th segment of k-th sector
Sk
- Total surface of segment (target)
Sjk - Total surface of j-th polar zone in k-th sector, and it is determined by formula
If we introduce spatial angle (steradian) of j-th polar zone:
j = 2 (cos j cos j +1 )
(11)
Total surface of j-th polar zone in k-th sector is determined by formula

S jk = j Rk = 2Rk (cos j cos j +1 )
2
2.3
(12)
Fragment velocities
Fragment velocities are measured using different techniques as electronic, optical or xray. If these techniques are not available, initial velocity of fragments released from
explosion of a warhead is approximated by the Gurney formula. The simplest expression of
the Gurney formula for symmetrical configurations is:
vGurney = 2 E 1 / (0,5 + M / C )
(13)
where 2 E is Gurney constant, M-metal mass of warhead case and C - mass of explosive
charge [7].
The Gurney constant can be approximated by the simple expression 2 E = 0,338 D ,
where D is the detonation velocity (depending on explosive type and its density).
3.
EXPERIMENTAL RESEARCH
3.1
Influence of warhead case material on natural fragmentation

performances
Experimental researches performed by authors were undertaken in order to estimate all

relevant performances of warhead natural fragmentation for 120mm W1 and 120mm W2
(three types of steel with different mechanical properties).
Author also analyzed experimental data for projectiles 155 mm HE M107 of U.S. type
and Bofors 155 mm HE M54 in order to determine the fragmentation effect for two different
types of projectile body material [2]. Both projectiles are practically identical geometrically.
Number, mass and fragments shape of each mass group are determined using the Pit test,
and relative estimation of spatial fragments efficiency was done in Arena test.
The main point of first research was directed to an influence of warhead metal case
material properties on warhead natural fragmentation performances (number, mass and
fragments geometrical shape). Dimensionless thickness of the warhead shell W1 was
t/d=0,0707, and ratio of metal warhead shell and explosive charge mass was CTNT/m =
6
785
Czech Republic
2006
[Content]
0,565. Dimensionless thickness of the warhead shell W2 was t/d=0,0629, and ratio of metal
warhead shell and explosive charge mass was CTNT/m = 0,653 [12].
Tested warheads cases were made of following steels 45Cr2 (ratio of tensile strength and
yield strength Rm/Rv=1,09), C70D (Rm/Rv=1,59) and steel AB or 9180VP (Rm/Rv=1,41)
marked according to Army standard, former SFRJ.
Fig 6. Warhead 120 mm W1
Fig 7. Warhead 120 mm W2
Four warheads fragmentation tests were performed for each casing material of the
warhead and obtained data were processed according to Mott methodology (Fig.8).
3500
Warhead W1
t/d=0,0687
C/m = 0,586
Explosive TNT
3000
2500
Fragments number > m aver
Fragments number > m aver
3500
Steel 9180VP (JAS)
2000
Steel 45Cr2
1500
Steel C70D
1000
500
0
Warhead W2
t/d=0,063
C/m = 0,653
Explosive TNT
3000
2500
Steel 9180VP (JAS)
2000
Steel 45Cr2
1500
Steel C70D
1000
500
0
2
3
4
5
6
Mean mass of fragments group - m aver , g
2
3
4
5
6
Fig 8. Dependency of fragments number variation greater than m from the mean fragments
mass for warheads W1 and W2
Comparing the two warheads W1 and W2, greatest number of fragment has warhead
W2, made of steel C70D. Warhead W1 with case made of the steel 9180 has greatest
fragments number and minimal fragments number with case stell 45Cr2. Warhead W2 has
minimal number of fragments for case made of steel 45Cr2. Between steels C70D and 9180
there are slightly differences in number of fragments for warhead W2.
Authors noted a strong dependency of fragments number and their mass from the ratio
Rm/Rv. Steels with higher ratios Rm/Rv generated considerable higher fragments number
(steels C70D and 9180). However, it was not possible to conclude from Fig.8 which material
would be the most favorable. Because of that, authors presented their results in a different
way (fig. 9). The mean fragments mass of each particular group was taken as a variable.
Relationship between fragments mass groups number, whose mean masses are less than the
mean mass of the total fragments number, or between fragments mass of each group and the
total warhead fragments mass were established [12].
From above (fig. 9) diagrams, comparing number of fragments and mass participation it
can be more precisely done by optimization of warhead type as function of warhead body
material. In warhead W1 steel 9180 gives greater number of fragments and greater mass
participation comparing to steel C70D for fragments with mean mass smaller than 5 gr,
while for larger fragments mass more favorable steel is C70D. For warhead W2, for
fragments mass smaller than 4 gr most favorable is steel C70D, while for larger fragments
mass (greater from 4gr) most favorable is steel 9180. As a conclusion, steel C70D has better
lethality performances in warheads which are used mainly against infantry, while steel 9180
would have better penetration capability against light armored vehicles.
786
1
0.8
0.6
0.6
0.4
Warhead W1
t/d=0,0687
C/m = 0,586
Explosive TNT
0.2
0
0
Steel
Steel
Steel
Steel
Steel
Steel
9180 - Ni/N
45Cr2 - Ni/N
C70D - Ni/N
9180 - Mi/M
45Cr2 - Mi/M
C70D - Mi/M
2
4
6
8
Mean mass of fragments group - maver, g
0.4
0.2
0
10
[Content]
1
0.8
Ni /N for m i <maver
0.8
2006
Mi /M for m i <maver
Czech Republic
0.8
0.6
0.6
Steel
Steel
Steel
Steel
Steel
Steel
0.4
0.2
0
0
9180 - Ni/N
0.4
45Cr2 - Ni/N
Warhead W2
C70D - Ni/N
t/d=0,063
9180 - Mi/M 0.2
C/m = 0,653
45Cr2 - Mi/M
C70D - Mi/M
Explosive TNT
0
2
4
6
8
10
Mean mass of fragments group - m aver, g
Fig 9. Fragments number or mass participation as a function of mean fragments mass for
warheads W1 and W2
Warheads with case made of the steel with larger ratio of Rm/Rv generates greater
number of fragments but with less mean mass and greater fragments mass participation.
Authors explained this phenomenon as a consequence of warhead case ability to expand into
more considerable volume with thinner case walls under detonation products pressure before
fragmentation (warhead body mass conservation condition). Experimental researches have
shown that initial volume of a warhead can be increased several times before fragmentation
of warhead case material has been occurred (Fig. 1). Authors also found the relationship
ti/t0=(Vi/V0)-0,5 between relative volume rise and relative wall thickness during the warhead
case expansion [12]. At 4 times increase of warhead volume, new corresponding warhead case
thickness decreases at 0,52 of initial case thickness. When the ratio of Rm/Rv rises, the ratio
Vi/V0 increases as well and ti/t0 decreases, what results in greater fragments number but with
less mass.
Second analysis was analysis of natural fragmentation experimental data for projectiles
155 mm HE M107 (steel AISI 1045) and Bofors 155 mm HE M54 (High fragmentation
steel), made in Bofors facilities. Projectiles were geometrically identical, with different types
of projectile body material [2].
Before the tests, the ordinary bursting charges in the shells were removed, and the shells
were filled with cast hexotol corresponding to "Composition B". The composition of this
hexotol was RDX 59.5%, TNT 39.5% and wax 1%. The weight of the explosive was approx.
7 kg. The average weights of the complete shells (without fuzes used for the tests were:
Bofors 155 mm HE shell M54 42.4 kg, US 155 mm HE shell M107 42.6 kg [2].
The shells were exploded statically, in target arrangements A and B of different designs.
The shells were placed in horizontal and vertical positions.
The target arrangement type A (shell in horizontal position) was built up in the form of
an enclosure. Three sides consisted of 2 mm mild steel plate, size 1 x 6 m. The fourth side
was made of 7 mm steel plate, brand SIS 1411 (s = 250 MPa, B = 420 MPa) size 1 x 6 m.
The distance at right angles from the pole on which the shell was placed to the sides was 3
m. The target arrangement type B (shell in vertical position) was made of steel plates with
the dimensions 1 x 2 m, with thicknesses of 2 mm and 7 mm, respectively, and with the
same mechanical properties as those used in target arrangement type A. The plates were
placed on edge to a height of 3 m and a width of 2 m. For fragments efficiency assessment
on unprotected personnel, 19mm chipboards with the dimensions 1,22 x 2,5m was used, at
distances of 6m and 12 m [2].
The targets were placed so that the bottom edges of the lower parts were on the same
horizontal plane. The height at which the shell was placed, in an upright position, was fixed
at 1,50 m, and the line of sight to each target was drawn in [2].
8
787
Czech Republic
2006
[Content]
In Bofors study, the general impression obtained from these comparative static bursting
tests was that the Bofors shell gave considerably better fragmentation. Bofors 155 mm HE
shell M54, gave twice as many perforations as the U.S. 155 mm HE shell M 107. The
appearance of the holes also differed. The 155 mm HE shell M107, gave large, longish
perforations, while the Bofors 155 mm HE shell M54, produced holes of more regular sizes.
This was entirely due to the properties of the materials in the two types of shells.
Authors have determined, using the earlier explained method of fragments spatial
distribution, following:
Target arrangement type A: In first polar zone (-45 to +45) there is no significant
difference in penetration density through targets (thickness 2mm). In polar zone of 45 to
135 (for 2mm target thickness) penetration density is 35% greater, and in polar zone of
225 to 315 (for 7mm target thickness), penetration density is 100% greater. Further more,
in polar zone of 135 to 225 there is 92 % greater penetration density for Bofors 155mm
M54 projectile. General conclusion is that Bofors warhead has about 35% better efficiency
against infantry, and about 100% better efficiency against light armored and mechanized
vehicles.
By analysis of experimental results of projectiles natural fragmentation in B arena
(influence of lateral fragments), on different distances (3 m, 6 m and 12 m) and against
targets of different thicknesses (2mm and 7mm steel plate) authors concluded that efficiency
of Bofors 155mm M54 projectile against armored targets is 1,8 to 2,05 times better at
distance from 3m to 12m, respectively, while efficiency of Bofors 155mm M54 projectile
against infantry is from 2,1 to 1,42 times better at distances 6m and 12m, respectively.
As a general conclusion for both tests, Bofors 155mm M54 projectile with high
fragmentation steel has significantly greater number of fragments, compared to US projectile
155mm M107. Bofors 155mm M54 projectiles efficiency is significantly better against light
armored and mechanized vehicles and that penetration capability (min 80%) doesnt
significantly change with increasing the distance, comparing to US projectile 155mm M107.
Against infantry, that lethal efficiency is minimum 35% greater, comparing to 155mm M107
and what is important to mention, it decreases with increasing the distance. That penetration
capability of fragments doesnt considerably change with increasing the distance, even
though projectile 155mm M54 has considerably greater number of fragments, and smaller
mean mass of fragments, with similar velocity of fragments; this points to fact that important
role plays shape of fragments, during process of penetration through target, specific energy
of fragment.
3.2
Experimental research on influence of explosive charge to natural

fragment size distribution
The point of this research was an estimation of influence of detonation velocity on

natural warhead fragmentation performances (number, mass, geometrical shape and velocity
of fragments).
Experimental fragmentation tests were carried out with four types of HE warheads 120
mm W1 and W2, 128 mm M87 and 152 mm M84.
Warheads cases W1 and W2 were made of the steel C70D, and warhead masses were
kept the same, but only types of explosive charge were varied (TNT and Composition B).
Ratios of explosive charge mass and warhead metal case mass of the warhead W1 was
CTNT/m = 0,565 and CComp B/m = 0,587, and for warhead W2 was CTNT/m = 0,653 and CComp
B/m = 0,667. Dimensionless thickness of the warhead shell W1 was t/d = 0,0707, and for
788
Czech Republic
2006
[Content]
warhead W2 was t/d = 0,0629. Four fragmentation tests were carried out for each type of
explosive charge.
For warheads W1 and W2 fragments were classified by mass groups and experimental
data were processed using Held methodology (Fig.10).
6000
Cumulative fragments mass, g
Cumulative fragments mass, g
6000
5000
4000
Warhead W1
Steel C70D
t/d = 0,0707
3000
2000
Explosive Comp B, loss of

fragments 11,7%, C/m = 0,587
1000
Explosive TNT, loss of fragments

8,4%, C/m = 0,565
5000
4000
Warhead W2
Steel C70D
t/d = 0,0629
3000
2000
Explosive Comp B, loss of

fragments 10,8%, C/m = 0,667
1000
Explosive TNT, loss of fragments

6,1%, C/m = 0,653
1000
2000
3000
4000
Cumulative fragments number
5000
1000
2000
3000
4000
Cumulative fragments number
5000
Fig 10. Cumulative fragments mass dependency on cumulative fragments number for
warheads W1 and W2 [13]
Warheads W1 and W2 with the case made of steel C70D and Comp B produces larger
number of fragments compared to the warheads with TNT explosive charge. At the same
time, in these two warheads with Comp B greater loss of fragment mass has been noticed
(metal part of warhead body has been converted to very small fragments), for W1 11,7 %
(Comp B) and 8,7 % (TNT), comparing to W2 10,8% (Comp B) and 6,1% (TNT).
However, it is not possible to conclude from Fig.10 which warhead has better
fragmentation performances. Because of that, authors presented their results in a different
way (Fig. 11).
0.8
0.6
0.4
Ni/N for Explosive Comp B

Ni/N for Explosive TNT
0.2
Mi/M for Explosive Comp B

Mi/M for Explosive TNT
0
0
2
4
6
8
0.6
0.4
0.2
0
10
0.8
Ni /N for m i <m aver
Warhead W1
Steel C70D
t/d = 0,0707
Mi /M for m i <m aver
0.8
0.8
Warhead W2
Steel C70D
t/d = 0,0629
0.6
0.4
Ni/N for Explosive Comp B

Ni/N for Explosive TNT
0.2
Mi/M for Explosive Comp B

Mi/M for Explosive TNT
0
0
2
4
6
8
0.6
0.4
0.2
0
10
Fig 11. Fragments number contribution, or fragment mass contribution as a function of the
mean fragments mass for warheads W1 and W2 [13]
The mean mass of particular groups was taken as a variable. Correlation between
fragments number of mass groups, whose mean masses are less than mean fragments mass
related to total fragments number, or between fragments mass of these mass groups and total
fragments mass of the warhead were established. The warheads W1 and W2 with explosive
charges of Comp. B generate more fragments but with less mean mass, compared to
warheads W1 and W2 filled with TNT. Warhead W1, with Comp. B has in fragment groups,
with mean mass smaller than 10gr, dominantly larger number of fragments and larger mass
participation of these fragments, while in warhead W2 these differences arent significantly
greater.
Authors explained such results as consequence of more intensive influence of detonation
products generated by more powerful explosive. When more powerful explosive is used,
10
789
Czech Republic
2006
[Content]
initial warhead volume is increased more intensive and warhead case thickness is decreased
(condition of mass conservation).
Since the sophisticated measuring equipment for fragment velocity measurement was
not available to authors, the Gurneys formula was used in order to predict initial fragments
velocities [7]. Initial fragment velocity variation for warheads filled with TNT and
Composition B explosives charge and for particular warhead segments was determined using
methodology described in the Crulls report [3] (fig. 12 and fig. 13).
2000
2000
1800
1800
Velocity of fragments, m/s
From these diagrams of velocity of fragments vs. relative distance of warhead case
segments it can be seen that mean fragment velocity is about 300 m/s larger in warheads
filled with Comp. B, fragments with less mass will have less energy and radius of lethality.
1600
1400
Warhead W1
Explosive Comp B, C/m = 0,587
Explosive TNT, C/m = 0,565
1200
1000
800
600
400
200
0
1600
1400
Warhead W2
1200
1000
800
600
400
200
0
0.2
0.4
0.6
0.8
Relative distance of warhead case segment
0.2
0.4
0.6
0.8
Relative distance of warhead case segment
2000
2000
1800
1800
Fig 12. Variation of fragment velocity as a function of explosive type for warheads W1 and
W2 [13]
1600
1400
Warhead 128mm M87

1200
1000
800
600
400
200
0
0
0.2
0.4
0.6
Relative position
0.8
1600
1400
1200
1000
Warhead 152 mm M84
800

600
400
200
0
0
0.2
0.4
0.6
0.8
Relative position
Fig 13. Variation of fragment velocity as a function of explosive type for warheads 128 mm
M87 and 152 mm M84 with explosive Comp B and TNT
Fragments concentration density per steradian as a function of polar zone for warheads
128 mm M87 with explosive Comp B and TNT is presented in Fig. 14 [11]. Lethality zone for
warheads 152 mm M84 with explosive Comp B and TNT is presented in Fig. 15.
For assessment of spatial fragments distribution and efficiency of warheads 128 mm

M87 (steel C70D) and 152 mm M84 (steel 45Cr2), experiments were conducted in two types
of Arena facilities. In first Arena spatial distribution of fragments is determined, and in
second one radius of lethality zones.
From diagram on fig. 14 it can be seen that greater fragment energy (with smaller mass,
but larger velocity) have fragments from warhead filled with Comp. B, i.e. for fragment
density of 400 frags/steradian width of polar zone for warhead with Comp. B is 35, while
for warhead with TNT it is 20. Warhead filled with TNT had maximal fragment density of
930 frags/steradian, and for the same fragments density, warhead filled with Comp. B has
width of polar zone of 15 and maximal fragment density of 1300 frags/steradian.
790
11
1200
1000
Explosive TNT,
C/m = 0,558
600
[Content]
Warhead 152 mm M84
Steel 45Cr2
t/d = 0,117
Explosive Comp B,
C/m = 0,299
Explosive TNT,
C/m = 0,282
Explosive Comp
B, C/m = 0,598
800
2006
Warhead 128 mm M87

Steel C70D
t/d = 0,071
Density, frags/m 2
Scaled Density Frags / Steradian
1400
Czech Republic
400
5
4
3
2
1
200
0
0
20
40
60
80
100
120
140
160
180
Polar Zone, deg
Fig 14. Fragments concentration density per

steradian as a function of polar zone
for warheads 128 mm M87 with
explosive Comp B and TNT
15
20
25
30
35
40
Radius of lethality zones, m
Fig 15. Lethality zone for warheads 152 mm

M84 with explosive Comp B and
TNT
Diagram on fig. 15 shows that projectile 155mm M84 filled with Comp. B has 33%
larger lethality area.
3.3
Influence of warhead design on natural fragmentation

performances
Experimental researches carried out by the authors were aimed on natural fragmentation
performances estimation (number, mass and fragment form) of the warhead when the
warhead or projectile design shape is changed [14]. Spatial fragments distributions were
determined through arena tests.
Experimental fragmentation tests in order to determine the influence of projectile design
on fragmentation characteristics were carried out with 11 types of HE warheads 105 mm M1
[1]
, 120 mm W2, 120 mm M62P3, 122 mm OF-462, 122 mm M76, 128 mm M63, 128 mm
M87 [11], 152 mm OF-540, 152 mm M84, 155 mm M107 [1] and 155 mm ERFB.
Fig 16. Warheads
Fragments concentration density per steradian (spatial fragments density) for each polar
sector of the arena:
12
791
2000
Warhead 122 mm M76,

Explosive Comp B, C/m =
0,305, t/d = 0,115
1800
1600
2000
Warhead 122 mm OF-462,

Explosive TNT, C/m =
0,225, t/d = 0,135
1400
1200
1000
800
600
400
200
0
0
20
40
60
Czech Republic
80
100
120
Polar Zone, deg
140
160
[Content]
Warhead 128 mm M63,

Explosive TNT, C/m = 0,484,
t/d = 0,078
1800
1600
1400
Warhead 128 mm M87,

t/d = 0,071
1200
1000
800
600
400
200
0
0
180
2000
2006
20
40
60
80
100
Polar Zone, deg
120
140
160
180
155 mm M107, Explosive TNT,

C/m=0,221, t/d = 0,138
C/m=0,222, t/d = 0,138
1800
1600
1400
1200
1000
800
600
400
200
0
0
20
40
60
80
100
120
Polar Zone, deg
140
160
180
Fig 17. Fragments concentration density per steradian as a function of polar zone
Diagram (fig. 17) shows the complexity in determination of design influence on

fragment distribution prediction for different polar zones. Beside outer design (slenderness
of projectile ogive) significant role plays ratio of explosive mass to metal mass C/m, type
and density of explosive charge and relative ratio t/d (equivalent thickness of projectile body
wall). Projectiles with greater slenderness, greater ratio C/m, and smaller ratio t/d have larger
density of fragments per steradian, for the same type of explosive. It is interesting to note
that projectile 105 mm M1, 122mm OF-462 and 155 mm M107 have almost the same ratios
C/m and t/d, and also approximately the same value of maximal density of about 1100
frags/steradian in polar angle of about 100.
20
16
14
12
10
8
6
4
Warhead 120 mm, W2,

t/d = 0,0629
Warhead 120 mm, M62P3,
t/d = 0,116
18
16
Density, frags/m 2
Density, frags/m 2
20

C/m=0,222, t/d = 0,138
18
14
12
10
8
6
4
2
10
15
20
25
30
35
40
10
15
Warhead 122 mm M76,

Explosive Comp B, C/m =
0,305, t/d = 0,115
Explosive TNT, C/m =
0,225, t/d = 0,135
18
Density, frags/m2
16
14
12
10
8
6
4
20
25
30
35
40
Warhead 128 mm M63,

t/d = 0,078
Warhead 128 mm M87,
t/d = 0,071
18
16
Density, frags/m2
20
20
14
12
10
8
6
4
2
0
10
15
20
25
30
35
40
10
15
20
25
30
35
40
Fig 18. Warhead lethality zone
792
13

20
14
12
[Content]
C/m=0,221, t/d = 0,138
155 mm ERFB, Explosive TNT,
C/m = 0,249, t/d = 0,128
18
16
Density, frags/m 2
Density, frags/m 2
16
2006
20
Warhead 152 mm M84,

t/d = 0,117
t/d = 0,116
18
Czech Republic
10
8
6
4
2
14
12
10
8
6
4
2
0
10
15
20
25
30
35
40
10
15
20
25
30
35
40
Fig 19. Warhead lethality zone
Comparative analysis of lethality radius for projectiles with the same caliber and
different design confirm observations mentioned above. Projectile 120 mm W2 has
considerably greater ratio C/m and smaller ratio of t/d comparing to projectile 120mm
M62P3, so the lethality zone of this warhead is 2,8 time greater. Projectile 122 mm M76
(Comp. B) has 44% larger lethality area compared to projectile 122 mm OF-462 (TNT),
because it has greater detonation velocity, greater C/m ratio, greater slenderness of ogive and
smaller ratio of t/d. Warhead 128 mm M87 (TNT) has 50% larger lethality area or 80%
larger lethality area when it is filled with RDX compared to projectile 128 mm M63 (TNT).
2000
1800
1200
Velocity fragments, m/s
1400
1000
800
600
105 mm M1, Explosive TNT, C/m=0,222
400
200
1200
1000
800
600
Warhead 120 mm, W2,

Warhead 120 mm, M62P3,
400
0
0.2
0.4
0.6
Relative position
0.8
2000
1800
1800
2000
1600
1400
1200
1000
800
600
Warhead 122 mm M76, Explosive

Comp B, C/m = 0,305,
400
200
0
0
0.2
0.4
0.6
Relative position
0.8
800
600
200
0
1800
1200
1000
800
Warhead 152 mm M84,
200
0
0
0.2
0.4
0.6
Relative position
Warhead 128 mm M63,

Warhead 128 mm M87,
400
2000
1400
1000
1600
0.8
1200
1800
400
0.4
0.6
Relative position
1400
2000
600
0.2
1600
1400
200
1600
0.2
0.4
0.6
Relative position
0.8

C/m=0,221
155 mm ERFB, Explosive TNT,
C/m = 0,249
1600
1400
1200
1000
800
600
400
200
0
0.8
0.2
0.4
0.6
Relative position
0.8
Fig 20. Fragment velocity as a function of relative position
14
793
Czech Republic
2006
[Content]
Warhead 128 mm M87 has greater C/m ratio and slightly smaller ratio of t/d and greater
slenderness of ogive. Comparing two projectiles 152 mm OF-540 and 152 mm M84 is
interesting because of the influence of hollow base at the projectile (M84) bottom on
fragmentation efficiency. This projectile has greater slenderness of ogive and similar ratios
C/m and t/d, but the influence of hollow base reduces his lethality capabilities.
Projectile 155 mm ERFB has greater ratio C/m and smaller ratio t/d and significantly
greater slenderness of ogive comparing to projectile 155 mm M107, so his lethality zone is
85% greater.
Significant influence on projectile efficiency is made by fragment velocity distribution
along projectile symmetry axis. It is especially important for projectile ogive and his
cylindrical part, because the fragments from these parts have highest influence on spatial
distribution of lateral fragment spread.
Maximum fragment velocities are achieved at warheads having greater explosive charge
mass and metal case mass ratio C/m, which are appeared within central fragments polar zone
(fig. 20).
4.
CONCLUSION
Influence of warhead design on lethal efficiency is very complex. Very important role
play design factors as external and internal geometrical paths, relative warhead case
thickness, explosive charge mass and warhead metal case ratio, type of explosive, etc.
Until recently, any prediction method, which is capable to quantify particular influence
of above parameters on lethal efficiency radius, has not been developed yet.
It was not possible to clearly define the influence of explosive type on natural
fragmentation performances by using Held method.
Mott formula cannot provide a clear influence of warhead metal case variation on
natural fragmentation performances.
Authors proposed a new approach to presentation of experimental results, which enables
selection of an optimal warhead explosive charge. Authors gave also different approach to
presentation of fragmentation experimental data, which enables selection of an optimal
warhead case material [12, 13].
Authors also established relationship between ti/t0 and Vi/V0 and connected it with the
ratio Rm/Rv in order to find how above parameters affect fragment number and their mass.
Warhead with steel case, which has higher ratio Rm/Rv generates simultaneously greater
fragments number but with less mean fragments mass and with greater fragments mass
participation. Authors also concluded that it was necessary a further development of
empirical relationship between Rm/Rv, and Vi/V0 [12].
In attempt to introspect complexity of influence of design, material of body, types of
explosive and methods of conducting the experimental testing, authors made comprehensive
analysis of eleven types of projectiles and warheads. Authors observed significant influence
of ratio C/m (mass of explosive to mass of metal body ratio), ratio t/d (equivalent thickness
of projectile body wall to caliber ratio), projectile ogive slenderness, fragment velocity
distributions along projectile symmetry axis and geometrical shapes of fragments.
Influence of geometrical shapes of fragments is not investigated in detail, and in
following period numerical simulation (drag, center of pressure, center of mass, etc.) of
behavior of fragment should be conducted during his flight through air and his behavior
794
15
Czech Republic
2006
[Content]
during the penetration through target, because of his influence on penetrating capabilities.
Analysis of geometrical shape is conducted only for warhead 128 mm M87 [11], but it is
insufficient for making scientific conclusions. Our activities are directed to creating expert
base for natural fragmentation processes.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
16
ANON: US Army Test and Evaluation Command Test Operations Procedure Fragment
Penetration Test of Armor, Report number TOP 2-2-722, ADA 125824., 1983
BOFORS: Fragmentation effect of 155m HE shells, Results of static test for comparison between
US 155mm HE shell M107 and Bofors 155mm HE shell M54
M. M. CRULL: Prediction of Primary Fragmentation Characteristics of Cased Munitions,
UXO/Countermine Forum 1998, The Joint UXO Coordination Office, 1998
O. GARLAND: The use of Physics-Based Models to Prediction Fragmentation of Ordnance,
Naval Air Warfare Center Aircraft Division, OMB N0. 0704-0188., 1997
V. GOLD, E. BAKER, NG K. and J. HIRLINGER: A Method for Prediction Fragmentation
Characteristics of Natural and Preformed Explosive Fragmentation Munitions, ARWEC-TR01007, US Army Armament Research, Development and Engineering Center, 2001
M. HELD: Fragmentation Warhead, Tactical Missile Warhead, Edited by Carleone J., Progress
in Astronautics and Aeronautics, Volume 155, AIAA, Washington, 1993.
R. KARPP and W. PREDEBON: Calculation of Fragment Velocity from Natural Fragmenting
Munitions, BRL Memorandum Report N0. 2509, USA Ballistic Research Laboratories,
Aberdeen Proving Ground, Maryland, 1975
R. LLOYD: Conventional Warhead System Physics and Engineering Design, Progress in
Astronautics and Aeronautics, Volume 179, AIAA, (1999).
J. PEARSON: A Fragmentation Model Applied to Shear-Control Warheads, Naval Weapons
Center, 1991
A. VICTOR: Warhead performance Calculation for Threat Hazard Assessment, Department of
Defense Explosive Safety Seminar, Las Vegas, (1996).
B. ZECEVIC and D. EROR: Terminal Ballistic of 128 mm M87 Rocket, FTV-270, Sarajevo,
(1986).
B. ZECEVIC, J. TERZIC and A. CATOVIC: Influence of Warhead Case Material on Natural
Fragmentation Performances, Annuals of DAAAM for 2004 & Proceedings of the 15th
International DAAAM Symposium, Viena, 2004
B. ZECEVIC, J. TERZIC and A. CATOVIC: Experimental Research on Influence of Explosive
Charge to Natural Fragment Size Distribution, Annuals of DAAAM for 2004 & Proceedings of
the 15th International DAAAM Symposium, Viena, 2004
B. ZECEVIC, J. TERZIC and A. CATOVIC: Influence of Warhead Design on Natural
Fragmentation Performances, Annuals of DAAAM for 2004 & Proceedings of the 15th
International DAAAM Symposium, Viena, 2004
795
Czech Republic
2006
[Content]
NEW FINDINGS IN THE ELECTRIC SPARK SENSITIVITY OF

NITRAMINES. PART I.
ASPECTS OF MOLECULAR STRUCTURE
Svatopluk Zemana, Zdenk Friedlb, Ji Koc, Vojtch Pelikna and Ji Majzlka
a
Institute of Energetic Materials, University of Pardubice

b
Faculty of Chemistry, Brno University of Technology

CZ-612 00 Brno, Czech Republic
Austin Detonator, Ltd., CZ-755 01 Vsetn, Czech Republic
Abstract
An ESZ instrument of a new, relatively simple construction has been applied to
determination of electric spark sensitivity (EES) of 16 nitramines. Results obtained are
compared with those from measurements by means of older RDAD instrument. Attention was
focused both on the relationships between the EES values from both the instruments and
theoretical B3LYP/6-311+G(d,p) N-NO2 bond dissociation energies, B3LYP/6-31G(d,p)
Mulliken net charges of the nitro group, heats of fusion and 15N NMR chemical shifts of the
nitrogen atoms of the most reactive nitro groups, respectively. It is stated a larger difference
between results of both instruments. It has been found that these relationships for ESZ KTTV
results are strongly affected by molecular structure factors. It is also pointed out that the
dislocations in the crystals should have some relation to electric spark sensitivity.
Keywords: bond dissociation energy, electric spark, sensitivity, nitramines
1.
INTRODUCTION
Recent studies on sensitivity of energetic materials to the electric spark [1-23] have
shown that this sensitivity depends not only on the chemical entity of the material given
[1,4,6,8,9,12,14-16], its granulometry [1-3,12,16] and grain shape [3], mechanical properties
[5,6], temperature [1,13], moisture content [2,4], its thermal [2,6,7,12,17,19] and impact
[7,16,20] reactivities but also on the configuration of electrodes, general structure of the
circuit [2,4] and finally on the discharge time [26]. Consequently, it has not yet been possible
to devise an international standard test specifying this stability [9,10,18]. Irrespective of the
above-mentioned spectrum of effects influencing the said sensitivity mentioned above, a
linear relationship between the tests results obtained with various apparatuses can be found
[9,10,11].
Our Institute in cooperation with the R&D Department of Austin Detonator Ltd.,
(Vsetn, Czech Republic) took part in constructing of an instrument for determination of
electric spark sensitivity [8-10]. With the help of this instrument, marked as RDAD,
measurements were carried out on a large set of polynitro compounds of secondary
explosive type [8-10,14,19-23]. From the results obtained it was possible to make
conclusions on the chemical mechanisms of initiation of these compounds [14,19-23]. The
outputs from the RDAD instrument correlate satisfactory with those obtained by other
methods of determination of the electric spark sensitivity in some recognized laboratories
[10]. However, this instrument is not suitable for determination of the said sensitivity of
primers and pyrotechnics. For this reason, we have developed with a financial support from
796
New Findings in the Electric Spark Sensitivity of Nitramines. Part I. Aspects of . . .
Czech Republic
2006
[Content]
Czech Ministry of Industry and Commerce [24] a new instrument [25-27], marked as ESZ
KTTV [26]. This instrument is suitable for measurements of energetic materials within the
whole scope of technically attractive sensitivities.
The present paper shows difference between outputs of both instruments mentioned
above from the point of view not only numerical values but also molecular structure. These
differences are demonstrated on some nitramines, theirs molecular structure are relatively
simple polynitro compounds, which mechanism of primary homolysis of N-NO2 bonds is
well understood [28-31].
2.
EXPERIMENTAL
2.1.
Principle of the electric spark sensitivity testing
The sensitivity of nitramines is determined here at conditions of undamped discharge

of charged condenser into the sample tested [25,26]. The energy of electric spark, EES,
required for 50 % initiation probability is calculated (up and down method) from the know
capacity C (in F) of the circuit and voltage U (in V) at the condenser by means of the wellknow equation
EES = 0.5 . C . U2
(1)
Beside the working condenser and test chamber with sample, the series resonance
circuit consists of the loss resistor of circuit (whose value varies from 0.27 to 10.0 ) and
total parasitic inductance (ranging from 1 to 1.6 H). In the course of oscillation discharge,
the increase in output in the space between the electrodes of testing spark gap reaches the
values of dP/dt 5.1018 W/s. The period of oscillations (from 0.5 to 20 s) as well as the
parameters of the instrument mentioned above depend considerably on the capacity value
(C), testing voltage (U), and the sum of Joule losses in the test substance.
2.2.
Measurement with instrument RDAD
The schemes of its wiring and its spark gap are presented in papers [8-10]. The
bottom electrode is a steel base in whose centre there is attached a screw as a leak electrode.
Attached to this screw is a plastic cylinder with a cut for fastening a piece of flexible tubing
of 5 mm height and 5 mm diameter serving as a container of the sample tested. The top
electrode is a steel cylinder of conical shape with attached resistance wire protected by a
plastic distance stop, which presses an Umafol lid to the flexible tubing. The capacity of the
capacitors was chosen so as to allow measurements in the voltage interval from 8 to 14 kV.
If the initiation was successful, the next measurement was carried out with a voltage lowered
by 0.2 kV; if it was unsuccessful, the voltage was increased by the same value. We
considered an initiation to be successful if the sample disappeared or the flexible tubing was
torn. Altogether 25 measurements were carried out with each substance and the results were
treated in the sense of eq. 1 by means of the up and down method.
2.3.
Measurements with instrument ESZ KTTV
The instrument is designed for a small-scale sensitivity testing of energetic materials

[25-27]. A high voltage power supply (operating voltage 4-10 kV) and a set of capacitors of
overall capacity in the range from 100 pF to 350 nF produces an electrostatic discharge of
total energy from 10 mJ to 16 J. A sample of 1 mm height is fitted into an isolation tube
Svatopluk Zeman, Ji Ko, Vojtch Pelikn, Ji Majzlk, Zdenk Friedl
797
Czech Republic
2006
[Content]
mounted on the lower cylindrical metal electrode. The upper electrode is equipped with
manual vertically adjustable positioner. The micro container itself is placed in a separate test
box with ventilation. The time behavior of voltage and current at the spark gap is registered
with a scope and than evaluated using a microcomputer to give the effective energy
transmitted to the sample.
Table 1 Summary of the studied nitramines, their code designation and sensitivities to
electric spark EES (RDAD and ESZ KTTV instruments)
EES required for 50 % initiation
probability obtained
RDAD
ESZ KTTV
Nitramines
No.
a
b
Chemical name
Code
designation
(J)
Ref.
(mJ)
N-Methyl-N,2,4,6-tetranitroaniline
TETRYL
5.49
23
574.2 7.6
2,5-Dinitro-2,5-diazahexane
DMEDNA
8.24
23
392.051.7
3,5-Dinitro-3,5-diazaheptane
DNDAH
12.49
23
225.011.3
2,4,6-Trinitro-2,4,6-triazaheptane
ORDX
8.08
23
243.314.4
1,9-Diacetoxy-2,4,6,8-tetranitro2,4,6,8-tetraazanonane
AcAn
13.93
23
206.714.4
1,3-Dinitroimidazolidine
CPX
9.68
23
403.314.4
1,3-Dinitro-1,3-diazetidine
TETROGEN
6.25
23
494.0a
1,4-Dinitropiperazine
DNDC
15.97
23
190.0 8.3
1,3,5-Trinitro-1,3,5-triazinane
RDX
2.49
23
216.415.6
10
1,3,5-Trinitro-1,3,5-triazepane
HOMO
3.70
23
431.319.3
11
1,3,5,7-Tetranitro-1,3,5,7-tetrazocane
HMX
2.89
23
236.410.1
12
1,3,5,7,9-Pentanitro-1,3,5,7,9pentazecane
DECAGEN
2.96
23
276.5a
13
trans-1,4,5,8-Tetranitrodecahydropyrazino[2,3-b]pyrazine
TNAD
5.43
23
520.053.3
14
1,3,3-Trinitroazetidine
TNAZ
8.76
20
78.37.2
15
2,4,6,8,10,12-Hexanitro-2,4,6,8,10,12hexaazaisowurtzitane
HNIW
4.70
20
462.041.4
16
4,10-Dinitro-2,6,8,12-tetraoxa-4,10diazaisowurtzitane
TEX
285.525.3
the value predicted by means of Fig. 1 (here line A for DECAGEN and line B for TETROGEN);
the method of measurement (i.e., according to the section 2.2.) failed in this case;
798
2.4.
Czech Republic
2006
[Content]
The nitramines studied
A survey of the tested nitramines is given in Table 1. All the nitramines were purified
by crystallization in motion, and often by precipitation crystallization (with a good stirring)
from organic solvents. The crystallization was carried out in such a way as to obtain the
product appearing as individual in thin layer chromatograph (Silufol, hexane-acetone eluent
and developing the chromatograms by 2 % diphenylamine in ethanol with UV detection).
However, the granulometry of the substances investigated was not specified.
2.5.
Molecular structure characteristics
In connection with our recent results of initiation reactivity of energetic materials

[15] the electronic charges at nitro groups, N-NO2 bond dissociation energies, heats of fusion
and 15N NMR chemical shifts of NO2 nitrogen atoms were chosen as characteristics of
molecular structure of the nitramines studied (see in Table 2).
The N-NO2 bond dissociation energies, DH(N-N), were calculated at ab initio level
by DFT B3LYP/6-311+G(d,p) method [32]. The calculation by total optimization gave the
equilibrium geometries of isolated molecules in the gas phase at 0 K. The optimized
structures were verified by normal mode frequency analysis to exclude imaginary
frequencies for the remaining 3N - 6 vibrational degrees of freedom. The net charges of the
nitro group, Qnitro, of the nitramines investigated were obtained by means of Mulliken
population analysis of electron densities obtained at ab initio level by DFT B3LYP/631G(d,p) method [32] without frequency analysis. The relevant data are summarized in
Table 2.
3.
DISCUSSION
The analysis of relationship between the EES values obtained by measuring on RDAD
and ESZ KTTV instruments revealed that the relationship is not unequivocal as is illustrated
in Figure 1. Similar trends in EES values obtained from the two instruments are exhibited by
class A of high-energetic cyclic nitramines. On the other hand, class C of linear nitramines
and particularly class B of cyclic nitramines show opposite trends in EES values. DMEDNA
data correlate with the last class probably due to a fact that its molecular skeleton can be
found as a part of skeletons of remaining members of this class. Surprisingly TETRYL data
correlate with data of the A class and TNAZ is unexpectedly more sensitive in measurements
by means of ESZ KTTV instrument.
Large difference between outputs of measurements on DRAD instrument on the one
hand, and those from ESZ KTTV instrument, on the other (both the values see Table 1),
should be due to mutual difference in construction namely spark gaps of each instrument;
scheme of this gap in the first instrument is presented in Ref. 10 and in the second one in
Ref. 26. Large energetic losses in the spark gap of the RDAD instrument might manifest
oneself as a strong thermal effect of the relevant electric discharge, i.e., outputs of this
measurement connected with thermolysis of the given samples. This statement should be
valid for all equipments whose outputs are in good unambiguous correlation with those of
DRAD instrument (see in Ref. 10).
It is know that initiation reactivity of secondary nitramines is connected with N-NO2
bond homolysis (see Ref. 15 and reference therein). It is logical that the weakest bond will
be broken as the first. Therefore, energy of electric spark, EES, should correlate with the
4
799
Czech Republic
2006
[Content]
lowest values of the bond dissociation energy DH(N-N). The verification can be found in
Figs. 2 and 3. While the relationships for outputs from RDAD instrument have a logical
regularity (Fig. 2), the opposed direction of these relationships is beyond expectation typical
for ESZ KTTV measurements (Fig. 3). Some authors are inclined to accept the idea of
conversion of the spark energy into a thermal effect, i.e., thermolytic mechanism of the
initiation by electric spark [3,6,15,17,19]. However, Rat et al. [5] considers this initiation to
be a multidisciplinary problem. Results of this paper give these authors [5] the truth. Namely
facts in Fig. 2 signalize the multidisciplinarity and that the thermal component of electric
discharge not participate in the ESZ KTTV measurements; the reversed relationships in Fig.
2 still wait for an interpretation (the most probably by means of a method of quantum
chemistry simulation).
Table 2: Calculated N-NO2 bond dissociation energies, DH(N-N), Mulliken net charges of
the nitro group, Qnitro, heat of fussion, Hm,tr, and 15N NMR chemical shifts, N, of
NO2 nitrogen atoms
Code
designation
TETRYL
DMEDNA
ORDX
CPX
TETROGEN
DNDC
RDX
HOMO
HMX
DECAGEN
TNAD
TNAZ
HNIW
TEX
Bond dissociation energiesa

Mulliken net chargesb
position
in
molecule
N(CH3)
2,52461,31,31,41,3,51,5- (ax)
3- (eq)
1,3- (eq)
2,4- (ax)
135791,3- (ax)
2,4- (eq)
12,6,8,124,104,10-
DH(NN)
(kJ mol1
)
130.95
176.55
194.86
186.50
186.83
163.96
168.65
188.72
154.84
161.78
177.70
177.16
179.05
168.09
167.16
167.29
140.58
176.05
154.36
175.24
177.06
151.12
169.46
186.79
Qnitro
Heat of
fussionc
Hm,tr
Ref
.
(e)
(kJ mol
1
)
-0.135261
-0.201818
-0.176718
-0.178246
-0.198959
-0.167031
-0.154150
-0.165965
-0.127376
-0.133985
-0.142745
-0.107995
-0.120368
-0.127184
-0.143869
-0.128259
-0.118018
-0.093269
-0.156531
-0.144025
-0.152032
-0.080958
-0.074584
-0.111044
25.850
29.409
34
34
34.031
34
22.570
26.320
33.823
33.012
27.738
15
N NMR chemical shiftsd
position
in
molecule
(ppm)
Ref.
N(CH3)
2,52,64-
-23.1
-27.8
-28.5
-32.0
37
37
34
34
34
34
34
1,31,31,41,3,51,53-
-31.2
-27.8
-26.2
-32.9
-33.0
-34.4
37
37
37
37
32.100
34
1,3,5,7-
-34.7
37
34.930
34
1,3,5,7,9-
-33.2
37
46.400
34
30.310
43.030
36
35
35
20.5
-40.3
-43.4
-33.4
38
36.100
12,6,8,124,104,10-
37
37
37
37
Calculated by B3LYP/6-311+G(d,p) method; b Calculated by B3LYP/6-31G(d,p) method; c Heats of fusion

taken from literature [34-36]; d 15N NMR chemical shifts taken from literature [37,38].
800
Czech Republic
2006
[Content]
Spak energy in mJ - ESZ KTTV instrument
700
TETRYL
600
TNAD
TNAD
500
HOMO
CPX
HNIW
400
DMEDNA
Ay = 104.57x - 31.935
y = -30.283x + 674.01
2
R = 0.9704
R = 0.8966
300
200
TEX
HMX
DNDC
ORDX
RDX
DNDAH
y = -5.7614x + 291.26
2
R = 0.9209
AcAn
100
TNAZ
0
0
10
12
14
16
18
Spark energy in J - RDAD instrument
Fig. 1 Relationship between the outputs from ESZ KTTV and RDAD instruments.
Logarithm of energy of electric spark / J
3.5
y = 0.0526x - 7.1785
2
R = 0.9888
DNDC
TEX
y = 0.0506x - 6.0647
2
R = 0.9908
2.5
CPX
TNAZ
DIGEN
2
y = -0.0006x + 2.2133
2
R = 0.8621
TETRYL
TNAD
HNIW
1.5
ORDX
DMEDNA
TETROGEN
HOMO
1
DECAGEN
HMX
RDX
0.5
0
135
145
155
165
175
185
195
205
-1
Dissociation energy / kJ mol
Fig. 2 Relationship between EES values from RDAD instrument and DH(N-N) energy.
Mulliken net charges of the total nitro group, Qnitro, were used in the study of impact
sensitivities of nitro compounds [33]. Application of this approach to electric spark
sensitivity gave in this study correlations which are present in Figs. 4 and 5. The highest
Qnitro values (i. e., the lowest absolute values) of each nitramine were used in both cases. The
outputs of the RDAD instrument are relatively scattered even though some relationships are
6
801
Czech Republic
2006
[Content]
intimated here (Fig. 4). As a large part of discharge is enacted in the air, also other factors
should have an influence on the nitramines initiation (they might be partly in floatation
during the discharge). The EES values from ESZ KTTV measurements, on the other hand,
gave the relationships (Fig. 5) which are strongly determined by molecular structure (namely
for compounds along the lines with positive slope).
Logarithm of energy of electric spark / mJ
7
y = -0,039x + 12,857
2
R = 0,9502
y = -0,0286x + 10,618
2
R = 0,9645
6,5
TETRYL
TETROGEN
TNAD
HOMO
HNIW
DMEDNA
CPX
DECAGEN
TEX
ORDX
5,5
HMX
RDX
DNDC
5
y = -0,0356x + 10,721
2
R = 0,9549
4,5
TNAZ
4
140
150
160
170
180
Dissociation energy / kJ.mol
190
200
-1
Fig. 3 Relationship between EES values from ESZ KTTV instrument and DH(N-N) energy.
18
DNDC
16
Energy of electric spark / J
14
TEX
12
10
CPX
CPX
DMEDNA
y = -174,93x - 20,522
2
R = 0,9053
TNAZ
ORDX
TETROGEN
y = -57,627x + 0,0293
2
R = 0,9999
TETRYL
TNAD
HNIW
HOMO
RDX
y = 13,854x + 4,2944
2
R = 0,8982
0
-0,22
RDX
HMX
DECAGEN
DIGEN
-0,2
-0,18
-0,16
-0,14
-0,12
-0,1
-0,08
-0,06
-0,04
-0,02
Charge on nitro group / e
Fig. 4 Relationship between EES values from RDAD instrument and net electronic charge of
the nitro group Qnitro.
802
Czech Republic
2006
[Content]
700
y = -5423,9x - 326,54
2
R = 0,9713
Energy of electric spark / mJ
600
TETRYL
TNAD
TETROGEN
500
y = 463,44x + 485,54
2
R = 0,9476
HNIW
HOMO
DMEDNA
400
HOMO
CPX
DMEDNA
300
y = -6147,7x - 846,97
2
R = 0,9922
TEX
HMX
ORDX
HMX
DNDC
200
y = 1095,8x + 365,34
2
R = 0,8942
DECAGEN
RDX
DNDC
100
TNAZ
0
-0,22
-0,2
-0,18
-0,16
-0,14
-0,12
-0,1
-0,08
-0,06
Charge on nitro group / e
Fig. 5 Relationship between EES values from ESZ KTTV instrument and net electronic
charge of the nitro group Qnitro.
18
16
y = 0,7199x - 12,942
2
R = 0,997
DNDC
Energy of electric spark / J
14
TEX
TEX
12
10
CPX
DMEDNA
DIGEN
TNAZ
y = -0,0281x + 9,0479
2
R = 0,9877
DMEDNA
TETRYL
ORDX
y = 0,2185x - 4,7043
2
R = 0,9999
TETROGEN
HNIW
TNAD
HOMO
HMX
y = -0,2156x + 9,7002
2
R = 0,9721
HNIW
RDX
DECAGEN
RDX
y = -1,2206x + 57,207
2
R = 0,9999
0
20
25
30
35
40
Heat of fusion / kJ.mol
45
50
55
-1
Fig. 6 Relationship between EES values from RDAD instrument and heats of fusion.
803
Czech Republic
2006
[Content]
700
y = -45,4x + 1712,7
2
R = 0,9775
y = 22,887x - 533,86
2
R = 0,9951
600
Energy of electric spark / mJ
TETRYL
TNAD
TETROGEN
500
HOMO
CPX
400
HNIW
HOMO
y = 2,6726x + 349,05
2
R = 0,938
HNIW
DMEDNA
CPX
300
DECAGEN
200
TEX
ORDX
HMX
RDX
RDX
DNDC
100
TNAZ
0
20
25
30
35
Heat of fusion / kJ.mol
-1
40
45
50
Figure 7: Relationship between EES values from ESZ KTTV instrument and heats of fusion.
Figs. 6 and 7 document an interrelationship of the EES values and intermolecular
interactions which are represented here by heats of fusion, Hm,tr, of the studied nitramines.
This approach was successfully applied in the study of influence of intermolecular
interaction on the impact sensitivity of polynitro compounds [15,39-41]. The heat of fusion
is defined as a sum of all heats of the polymorphous transitions and of the heat of melting
[34] and represents the work needed for formation of defects in crystal lattice. Fig. 6,
analogous to Fig. 4, documents again complexity of the initiation in the spark gap of the
RDAD instrument.
Fig. 7 reminds of analogous relationships between impact sensitivity (taken as the
first reaction) and the Hm,tr, values [15,39,41] of nitramines. However, these relationships
strongly differentiate cyclic and linear nitramines [15,39,41] while according to the Fig. 7
this differentiation is not evident. It might mean that a role namely of plastic deformations of
crystal and perhaps of dislocations in energetic crystals doesn't need to have such
signification in the initiation by electric spark as in the case of impact or shock. But the
dislocations should be a reason of existence of the relationships in Fig. 7. That means that
results from measurements by means of SEZ KTTV instrument should be more depended on
the shape and size of crystals as in case of RDAD. Satisfactory correlation of the data from
the first mentioned instrument with characteristics of nitramine molecular structure might be
caused by wide spectrum of their granulometry (given by method of samples crystallization)
which can leads to the averaged EES values.
804
Czech Republic
2006
[Content]
20
18
DPT
DNDC
16
AcAn
Spark Energy / J
14
OCPX
y = 1,1439x + 45,214
2
R = 0,9794
12
y = 8,0187x + 281,68
2
R = 0,9896
CPX
10
ORDX
TETROGEN
DIGEN
TNAZ
OHMX
HNIW
y = 0,4859x + 19,278
2
R = 0,9576
DECAGEN
HMX
RDX
2
0
-43
-38
-33
-28
-23
-18
15
N NMR Chemical Shift / ppm
Fig. 8 Relationship between EES values from RDAD instrument and 15N NMR chemical
shifts of nitrogen atoms of the primarily split off nitro groups (taken from [15,41].
700
y = 41.832x + 1662.5
2
R = 0.8606
Energy of the electric spark / mJ
600
TETRYL
y = -8.379x + 152.35
2
R = 0.9731
HNIW
500
TETROGEN
HOMO
CPX
400
DMEDNA
CPX
TEX
300
DECAGEN DECAGEN
ORDX
HMX
HMX
RDX
RDX
200
DNDC
y = -6.2253x + 30.052
2
R = 0.6326
100
TNAZ
0
-45
-40
-35
-30
15
-25
-20
-15
N NMR Chemical shift /ppm
Figure 9: Relationship between EES values from ESZ KTTV instrument and 15N NMR
chemical shifts of nitrogen atoms of the primarily split off nitro groups.
10
805
Czech Republic
2006
[Content]
Completely mutually different are data in the Figs. 8 and 9 which represent
relationships between 15N NMR chemical shifts of nitrogen atom in the most reactive nitro
groups (i e., primarily split off nitro groups) and the EES values. Fig. 8 somewhat reminds the
relationship between these shifts and square of detonation velocities of nitramines [15,41].
On the other hand, relationships in Fig. 9 go near to those in Fig. 5. It seems the outputs from
ESZ KTTV instrument have more close relation to molecular structure as those from RDAD
instrument.
4.
CONCLUSIONS
From the facts given above it possible to come to the conclusion that the mechanisms
of spark energy transfer into the reaction centre of molecule should be different in
measurement on the two alternative instruments (RDAD and ESZ KTTV). The thermal
component of discharge in the RDAD instrument may be one of the reasons of this
difference. This assumption is also supported by the differences (in order of magnitude)
between the EES values obtained from the individual instruments (see Table 1). Further, the
considerable electric energy losses in the area between the upper electrode and the sample
surface in the RDAD system measurements contribute to the difference mentioned above
most likely a large part of discharge is enacted in the air.
Data from ESZ KTTV instrument have close relation to molecular structure of
measured polynitro compounds [16]. In line with knowledge about initiation reactivity of
polynitro compounds [15,41] the EES values correlate with such characteristics of molecular
structure which correspond to the primarily leaving nitro group in the nitramine molecule.
Reversed relationship between the EES values and dissociation energies DH(N-N) (Fig. 3)
waits for a detailed, preferably quantum chemical interpretation so far. The EES values also
correlate with heats of fusion of the studied nitramines, probably due to influence of
dislocations in their crystals on electric spark sensitivity. This assumption can have some
relation to significant influence of the shapes and size of the crystals on electric spark
sensitivity of the energetic compounds in general.
Acknowledgements
The work was carried partly out under the auspices of the Ministry of Education,
Youth & Sports of the Czech Republic as a part of its research project No. MSM
0021627501 and partly under support from the Ministry of Industry & Commerce of the
Czech Republic within the framework of research project No. FC-M2/05.
5.
REFERENCES
[1]
T. E. LARSON, P. DIMAS AND C. E. HANNAFORD, Electrostatic Sensitivity Testing of

Explosives at Los Alamos, Inst. Phys. Conf. Ser. No. 118: Section 2 (Electrostatic 91), IOP
Publishing Ltd., 1991, pp. 107-117.
[2]
M. ROUX, M. AUZANNEAU AND C. BRASSY, Electric Spark and ESD Sensitivity of Reactive
Solids. Part. I: Experimental Results and Reflection Factors for Sensitivity Test
Optimization, Propellants Explos. Pyrotech. 18, 317-324 (1993).
[3]
M. AUZANNEAU, M. ROUX, Electric Spark and ESD Sensitivity of Reactive Solids. Part. II:
Energy Transfer Mechanism and Comprehensive Study on E50, Propellants Explos.
Pyrotech. 20, 99-101 (1995).
806
Czech Republic
2006
[Content]
[4]
S. AMARI, F. HOSOYA, Y. MIZUSHIMA, T. YOSHIDA, Electrostatic Spark Ignitability of

Energetic Materials, [Proc.] 21st Int. Pyrotech. Seminar, Moscow, Sept. 1995, pp. 13-31.
[5]
R. RAT, M. ROUX, J. P. CHAUMAT, Etudes theoriques sur les phenomenes dinitiation des
matieres explosibles, in M. ROUX (Ed.) [Proc.], Recueil des communications Journees
detudes sur la sensibilit des composants et des substances nergt. lelectricite statique,
Aussois, Mai 1996, pp. 147-168.
[6]
J. P. CHAUMAT, Linitiation est elle thermique? in M. ROUX (Ed.) [Proc.], Recueil des
communications Journees detudes sur la sensibilit des composants et des substances
nergt. lelectricite statique, Aussois, Mai 1996, pp. 173-179.
[7]
T. HASEGAWA, E. KAWASHIMA, K. SATOH, T. YOSHIDA, Correlations between Screening

Test Results of Energetic Materials, [Proc.] 22nd Int. Pyrotech. Seminar, Fort Collins,
Colorado, July 1996, pp. 195-207.
[8]
Z. KAMENSK, Electric Spark Sensitivity of Polynitro Compounds, M.Sc. Thesis, Univ.

Pardubice, June 1995.
[9]
S. ZEMAN, Z. KAMENSK, P. VALENTA, J. JAKUBKO, On the Electrostatic Spark Sensitivity

of Some Organic Polynitro Compounds, in M. Roux (Ed.) [Proc.] Recueil des comunications
Journees detudes sur la sensibilit des composants et des substances nergt. lelectricite
statique. Aussois, Mai 1996, pp. 197-206.
[10]
S. ZEMAN, P. VALENTA, V. ZEMA, Z. KAMENSK, Electric Spark Sensitivity of Polynitro

Compounds: A Comparison of some Authors Results, HanNeng CaiLiao 6, 118-122 (1998).
[11]
T. MATSUZAWA, M. ITOH, M. ARAI, S. HATANAKA, A. MIYAHARA, M. TAMURA, Electric

Spark Sensitivity for Materials and Compositions of Fireworks, Kayaku Gakkaishi 55, 39-45
(1994); Chem. Abstr. 121, 13270e (1995).
[12]
M. ROUX, A. TREVINO, M. AUZANNEAU, C. BRASSY, Sensibilite des substances explosives:

Etude de la sensibilite electro-statique dexplosifs polynitres aromatiques, [Proc.] 16th
Annual Conf. ICT, Karlsruhe 1985, pp. 3/1-3/15.
[13]
F. HOSOYA, K. SHIINO, K. ITABASHI, Electric Spark Sensitivity of Heat Resistant

Polynitroaromatic Compounds, Propellants Explos. Pyrotech. 16, 119-122 (1991).
[14]
S. ZEMAN, V. ZEMAN, Z. KAMENSK, Relationship between the Electric Spark Sensitivity

and the NMR Chemical Shifts of some Organic Polynitro Compounds, [Proc.] 28th Int.
Annual Conf. ICT, Karlsruhe, 1997, pp. 66/1 - 66/9.
[15]
S. ZEMAN, A Study of Chemical Micromechanism of the Organic Polynitro Compounds

Initiation, Chapter 2 in: PETER POLITZER, JANE MURRAY (Eds.), Energetic Materials, Part
II, Elsevier, 2003, pp. 25-52.
[16]
J. KO, S. ZEMAN, J. MAJZLK, J. STRNAD, Notice to Determination of the Electric Spark

Sensitivity of Energetic Materials, [Proc.] 5th Seminar New Trends in Research of Energetic
Materials, Univ. Pardubice, April 2002, pp. 110-126.
[17]
D. SKINNER, D. OLSO, A. BLOCK-BOLTEN, Electrostatic Discharge Ignition of Energetic

Materials, Propellants Explos. Pyrotech. 23, 34-42 (1997).
[18]
F. HOSOYA., Y. WADA, K. SHIINO, T. WAINAI, K. ITABASHI, M. TAMURA, T. YOSHIDA,

Synthesis of Heat-Resistant Nitro Compounds and their Estimation of Explosibility, Kogyo
Kayaku 53, 14-21 (1992).
[19]
J. KO, S. ZEMAN, Spark Sensitivity of Polynitro Compounds. Part IV. A Relation to

Thermal Decomposition Patrameters, HanNeng CaiLiao 8, 18-26 (2000).
[20]
J. KO, V. ZEMAN, S. ZEMAN, Spark Sensitivity of Polynitro Compounds. Part V. A

Relationship between Electric Spark and Impact Sensitivities of Energetic Materials,
HanNeng CaiLiao 9, 60-65 (2001).
12
807
Czech Republic
2006
[Content]
[21]
V. ZEMAN, S. ZEMAN, Relationship Between the Electric Spark Sensitivity and Detonation
Velocities of Some Polynitro Compounds,[Proc.] 28th Int. Annual Conference of ICT,
Karlsruhe, June 1997, pp. 67/1-67/10.
[22]
V. ZEMAN, J. KO, S. ZEMAN, Spark Sensitivity of Polynitro Compounds. Part II. A

Correlation with Detonation Velocities of some Polynitro Arenes, HanNeng CaiLiao 7, 127132 (1999).
[23]
V. ZEMAN, J. KO, S. ZEMAN, Spark Sensitivity of Polynitro Compounds. Part III. A

Correlation with Detonation Velocities of some Nitramines, HanNeng CaiLiao 7, 172-175
(1999).
[24]
Special Scientific and Development Activity in the Field of Energetic Materials, Project No.
FC-M2/05, Ministry of Industry and Commerce, Prague, 2000-2002, solver Univ. Pardubice.
[25]
J. STRNAD, J. MAJZLK, Determination of Electrostatic Spark Sensitivity of Energetic

Materials, [Proc.] 4th Seminar New Trends in Research of Energetic Materials, Univ.
Pardubice, April 2001, pp. 303-307.
[26]
D. JARMAN, W. PRINSE, R. BOUMA, Electrostatic Discharge Initiation of CL-20: Effect of

Discharge Time and Spark Energy, [Proc.] 34th Int. Annual Conf. ICT Karlsruhe, June 2003,
pp. 71/1-71/11.
[27]
M. MARTINEZ PACHECO, R. BOUMA, L. KATGERMAN, Electrostatic Discharge Initiation of

Ti+C Mixtures and the Thermite Al+MoO3, [Proc.] 35th Int. Annual Conf. ICT Karlsruhe,
June 2004, pp. 105/1-105/12.
[28] F. I. Dubovitskii, B. L. Korsoonskii, Kinetics of Thermal Decomposition of Nitro

Compounds (Russ.), Usp. Khim. 50, 1828 (1981).
[29] G. M. NAZIN., G. B. MANELIS, Thermal Decomposition of Aliphatic Nitro Compounds
(Russ.), Russ. Chem. Rev. 63, 313 (1994).
[30] M. D. COOK, Thermal decomposition of RDX: A critical review, J. Energet. Mater. 5, 257
(1987).
[31] G. B. MANELIS, G. M. NAZIN, YU. I. RUBTSOV, V. A. STRUNIN, Thermal decomposition and
combustion of explosives and powders (Russ.). Izdat. Nauka, Moscow, 1996.
[32] TITAN v.1.0.8. Wavefunction, Schrdinger, Inc., USA, 2001.
[33]
C. YHANG, Z. SHU, Z. HUANG, X. YHAO, H. DONG, Investigation of Correlation between

Impact Sensitivities and Nitro Group Charges in Nitro Compounds, J. Phys. Chem. B 109,
8978-8982 (2005).
[34]
S. ZEMAN, Some Predictions in the Field of the Physical Thermal Stability of Nitramines,
Thermochim. Acta 302, 11-16 (1997).
[35]
S. ZEMAN, Z. JALOV, Heats of Fusion of Polynitro Derivatives of Polyazawurtzitanes.

Thermochim. Acta 345, 31-38 (2000).
[36]
M. SUESKA, M. RAJI, S. ZEMAN, Z. JALOV, 1,3,3-Trinitroazetidine (TNAZ). Part II:

Study of Thermal Behaviour, J. Energet. Mater. 19, 241-272 (2001).
[37]
S. ZEMAN, Analysis and Prediction of the Arrhenius Parameters of Low-Temperature

Thermolysis of Nitramines by Means of the 15N NMR Spectroscopy, Thermochim. Acta 333,
121-129 (1999).
[38]
Z. JALOV, S. ZEMAN, M. SUESKA, P. VVRA, K. DUDEK, M. RAJI, 1,3,3Trinitroazetidine (TNAZ). Part I: Syntheses and Properties, J. Energet. Mater. 19, 219-239
(2001).
[39]
S. ZEMAN, New Aspects of the Impact Reactivity of Nitramines, Propellants Explos.

Pyrotech. 25, 66-74 (2000).
808
Czech Republic
2006
[Content]
[40]
S. ZEMAN, M. KRUPKA, New Aspects of Impact Reactivity of Polynitro Compounds, Part

III. Impact Sensitivity as a Function of the Intermolecular Interactions, Propellants Explos.
Pyrotech. 28, 301-307 (2003).
[41]
S. ZEMAN, New Aspects of Initiation Reactivities of Energetic Materials Demonstrated on

Nitramines, J. Hazard. Mater. (2006) in press.
14
809
Czech Republic
2006
[Content]
NEW FINDINGS IN THE ELECTRIC SPARK SENSITIVITY

OF NITRAMINES. PART II.
A PROBLEM OF HOT SPOTS
Svatopluk Zeman*, Ji Ko**, Vojtch Pelikn* and Ji Majzlk*

*)Institute of Energetic Materials, University of Pardubice
**)Austin Detonator, Ltd., CZ-755 01 Vsetn, Czech Republic
Abstract
Attention was focused on the influence of grain size upon the electric spark sensitivity of 1,3,5trinitro-1,3,5-triazinane (hexogen) and 1,3,5,7-tetranitro-1,3,5,7-tetrazocane (octogen) and on the
dependence of this sensitivity upon the content of hard and fine admixtures to these compounds. It
was found that decreasing of the sensitivity corresponds with the increase of the grain size of
corresponding crystals; decreasing of a number of the intergrain points (hot spots) in the volume
unit of these nitramines can be reason in this case. It was suggested that dislocations in the crystals
should play some role here. Electric spark sensitivity of both the compounds also will decrease if
hard and fine particles (hot spots) are added. Also this fact should correspond with a number of
intergrain points of the nitramines particles in the corresponding mixtures. It is stated that an
impact component absents in relevant electric discharge.
Keywords: electric spark, grain size, hexogen, octogen, hot spots, sensitivity
1.
INTRODUCTION
Some molecular structural aspects of the spark sensitivity of nitramines were dealt
with in the first part [1] of this series. An assumption was pronounced that significant
influence of shape [2] and size [2-5] of crystals on the electric spark sensitivity might
connect with dislocations in these crystals [1]. The said factors influence a mechanism of
energy transfer of the primary impulse into the reaction centre of the given molecule. At
present there is no unified model of the said mechanism [6].
A probable influence of the above-.mentioned dislocations might have been also
connected with impact component of the electric discharge. However, Stengach confirmed
[7] the absence of this component in his experiment with electric spark sensitivity of lead(II)
azide. This paper deals with an examination ofthe presence or absence of the impact effect in
the electric spark sensitivity measurements. Hexogen (1,3,5-trinitro-1,3,5-triazinane) and
octogen (1,3,5,7-tetranitro-1,3,5,7-tetrazocane) are used as the studied samples in which hot
spots are artificially introduced. Attention is paid also to the influence of the crystal size of
both the nitramines on their electric spark sensitivity.
810
New Findings in the Electric Spark Sensitivity of Nitramines. Part II. . . .
2.
EXPERIMENTAL
2.1.
Apparatus
Czech Republic
2006
[Content]
Corresponding measurements were realized by means of instrument [1,8-10] which

we developed with a financial support from Czech Ministry of Industry and Commerce [11].
Its code designation is ESZ KTTV [1,9,10]. Function of the instrument is described in the
first part of this series [1]. Electric spark sensitivity is expressed as the energy of the electric
spark, EES, required for 50 % initiation probability.
2.2.
Nitramines
The following samples were tested: hexogen; produced by Chemko Strske in seven
granulometric qualities according to MIL standard R-398C (1977)} and octogen of foreign
provenance in five granulometric qualities according to MIL standard H-45444B (1978)}.
The individual classes of these nitamines are characterized in the Table 1 by their average
grain size (the mean value of limit magnitudes in the given class). This Table also presents
the results of measurements. Also a sample of Dyno Nobels hexogen RDX-RS (grain sizes
from 50 up to 630 m) was used.
Table 1:
Electric spark sensitivity of hexogen and octogen versus

averaged values of their grain size (see also Fig. 1)
RDX
Mean value of
grain size in m
600
450
350
250
150
75
30
2.3.
EES
in mJ
1170
1060
900
535
215
137
55
HMX
Mean value of
grain size in m
1000
400
150
40
27
EES
in mJ
875
495
242
152
88
Hot spots
In the nature of the hot spots are meaning fine and hard particles, here crushed
glass and synthetic diamonds dust of Russian provenance [12]. For evaluation of the
influence of the crushed glass on the electric spark sensitivity, the lowest-grain hexogen (see
Table 1) was mixed with this additive (with grain size below 20 m). Analogue influence of
diamonds dust, whose particle size (its agglomerates) does not exceed 5 m, was evaluated
by means of hexogen RDX-RS. Mixtures containing various amounts of these hard particles
were then measured: composition of the mixtures and the results of measurements are
presented in Table 2.
Svatopluk Zeman, Ji Ko, Vojtch Pelikn, Ji Majzlk
811
Table 2:
Czech Republic
2006
[Content]
Electric spark sensitivities of the mixtures of hexogen with

hard particles of additives (see also Fig. 2)
Crushed glass
content in %
wt.
0
5
10
15
EES
in mJ
55
120
180
200
3.
DISCUSSION
3.1.
Effect of granulometry on EES values
Diamonds dust
content in %
wt.
0
5
10
15
20
EES
in mJ
291
430
560
706
830
The relationship between octogens granulometry and its sensitivity to electric spark
has been dealt with recently by Roux et. al. [2, 4]. Our results obtained by measurement on
the ESZ KTTV instrument are given in Fig. 1. In contrast to the results obtained by Roux et.
al. [2, 4], our dependences have opposite trend. The difference between our and Roux et. al.
finding might be due to mutual different construction of the apparatus circuits and spark
gaps in their and our cases. However, we suppose that our finding confirms the statement by
Auzanneau et. al. [4]: hot spots develop at the thinnest part of the solid, i.e. at intergrain
points. The number of these points in volume unit decreases with the increasing grain size,
and that increases the corresponding EES values. However, on the basis of relationships
between the EES values and heats of fusion of nitramines [1] we should suppose that also
dislocations in the crystals have some effect on the initiation of these compounds by spark.
3.4.
Effect of artificially introduced hot spots on electric spark sensitivity
Stengach studied the electric spark sensitivity of lead(II) azide and found out that an
addition of the hard inert admixtures to this azide (up to 10% by wt.) has practically no effect
on its electric spark sensitivity [7] (while the impact sensitivity is increased). We have
chosen a similar strategy in the measurements. The sample measured was hexogen of mean
grain size equal to 30 m with an admixture of crushed glass and hexogen RDX-RS with
admixture of diamonds dust. The results are given in Fig. 2, and are in accordance with
those given by Stengach [7], stating the absence of effect of impact component in the electric
discharge used. The trend shown in Fig. 2 can be interpreted by the introduced glass or
diamond particles separating the hexogen grains from each other. Thereby the number of the
intergrain contact points of hexogen or octogen grains in volume unit is decreased, and
hence the electric spark sensitivity of the respective mixture is decreased, too. This statement
again stands in accordance with the above-mentioned view by Auzanneau et. al. [4].
812
Czech Republic
2006
[Content]
1400
1200
RDX
Spark energy / mJ
1000
HMX
800
600
400
200
0
0
200
400
600
800
1000
1200
Grain size / m
Figure 1: Relationship between the grain size and electric spark sensitivity of hexogen
(RDX) and octogen (HMX).
900
diamond
800
y = -0.0406x + 27.891x + 289.79

2
R = 0.9995
Energy of electric spark /mJ
700
600
500
400
300
crushed glass
200
y = -0.45x + 16.65x + 53.25

2
R = 0.9952
100
0
0
10
15
20
25
Content of added "hot spots" / % wt.
Figure 2: Electric spark sensitivity of mixtures of the hexogen with crushed glass or
diamonds dust.
813
4.
Czech Republic
2006
[Content]
CONCLUSION
Decreasing of the electric spark sensitivity (i. e. increasing of the EES values), when
grains size of nitramines crystals increase, confirms the idea by Auzanneau et. al. [4] about
the mechanism of spark energy transfer into the powdered reactive solid, i. e. in this case to
decrease the number of intergrain points in volume unit. However, also dislocations in the
crystals should have some effect on this type of energetic materials initiation [1].
Artificially introduction fine and hard particles in the crystalline nitramines decreases
the electric spark sensitivity of the resulting mixtures, i. e. the particles behave as
flegmatizing additive. This fact also agrees with the idea by Auzanneau et. al. [4] and
confirms the Stengachs stating [7] about the absence of the effect of the impact component
in the electric discharge at determination of the electric spark sensitivity.
Acknowledgements
The work was carried partly out under the auspices of the Ministry of Education,
Youth & Sports of the Czech Republic as a part of its Research project No. MSM
0021627501 and partly under support from the Ministry of Industry and Commerce of the
Czech Republic within the framework of research project No. FC-M2/05.
5.
REFERENCES
[1]
S. ZEMAN, J. KO, Z. FRIEDL, V. PELIKN, J. MAJZLK, New Findings in the Spark

Sensitivity of Nitramines. Part I. Molecular Structural Aspects. [Proc.] 9th Seminar New
Trends in Research of Energetic Materials, Univ. of Pardubice, April 2006.
[2]
Energy Transfer Mechanism And Comprehensive Study on E50, Propellants, Explos.,
Pyrotech. 20, 99-101 (1995).
[3]
T. E. LARSON, P. DIMAS, C. E. HANNAFORD, Electrostatic Sensitivity Testing of Explosives

at Los Alamos, Inst. Phys. Conf. Ser. No. 118: Section 2 (Electrostatic 91), IOP Publishing
Ltd., 1991, pp. 107-117.
[4]
Eergy Transfer Mechanisms and Comprehensive Study on E50, Propellants, Explos.,
Pyrotech. 20, 96-101(1995).
[5]
M. ROUX, A. TREVINO, M. AUZANNEAU, C. BRASSY, Sensibilite des substances explosives:

Etude de la sensibilite electro-statique dexplosifs polynitres aromatiques, [Proc.] 16th
Annual Conf. ICT, Karlsruhe 1985, pp. 3/1-3/15.
[6]
R. RAT, M. ROUX, J. P. CHAUMAT, Etudes theoriques sur les phenomenes dinitiation des
matieres explosibles, in M. ROUX (Ed.), Recueil des communications Journees detudes sur
la sensibilit des composants et des substances nergt. lelectricite statique. Aussois, Mai
1996, pp. 147-168.
[7]
V. V. STENGACH , Electric Spark Sensitivity of Lead(II) Azide, Fyz. Goreniya i Vzryva 6,

113- 119 (1970).
[8]
J. STRNAD, J. MAJZLK, Determination of Electrostatic Spark Sensitivity of Energetic

Materials, [Proc.] 4th Seminar New Trends in Research of Energetic Materials, University
of Pardubice, April 2001, pp. 303-307.
814
Czech Republic
2006
[Content]
[9]
D. JARMAN, W. PRINSE, R. BOUMA, Electrostatic Discharge Initiation of CL-20: Effect of

Discharge Time and Spark Energy, [Proc.] 34th Int. Annual Conf. ICT Karlsruhe, June 2003,
pp. 71/1-71/11.
[10]
M. MARTINEZ PACHECO, R. BOUMA, L. KATGERMAN, Electrostatic Discharge Initiation of

Ti+C Mixtures and the Thermite Al+MoO3, [Proc.] 35th Int. Annual Conf. ICT Karlsruhe,
June 2004, pp. 105/1-105/12.
[11]
Special Scientific and Development Activity in the Field of Energetic Materials, Project No.
FC-M2/05, Ministry of Industry and Commerce, Prague, 2000-2002, solved by University of
Pardubice.
[12]
Poroshok ultradispersnyi UDA (Ultra Dispersive Powder UDA), Technical Conditions TU

3972-014-07511703-94, Novosibirsks Factory of Synthetic Fibres, Iskitim, Russia, 1994.
815
Czech Republic
2006
[Content]
THEORETICAL STUDY ON THE DECOMPOSITION OF 1,1DIAMINO-2,2-DINITROETHYLENE

Hehou ZONG, Yuanjie SHU
1, Yigang HUANG, Xinfeng WANG
Institute of Chemical Materials China Academy Engineering Physics, Mianyang
Sichuan, P R China 621900
Abstract: We have investigated the unknown unimolecular decomposition pathway of
the explosive molecule diaminodinitroethylene (FOX-7).With the use of ab initio
molecular dynamics simulations, we found there are mainly three decomposition
pathways: C-NO2 bond cleavage, nitro-to-nitrite rearrangement and hydrogen atom
migration from NH2 to NO2 then HONO elimination. And we used the DFT method to
calculate their energy barriers to access the reaction possibilities. We found the energy
barrier of hydrogen atom migration is rather low but its intermediate has high energy
and HONO elimination need comparatively high energy; C-NO2 bond cleavage energy
is 70.13kal/mol by use of B3LYP/6-31G(d, p), while the energy barrier of nitro-to-nitrite
rearrangement is 62.66kcal/mol with the same method and basis set without zero-point
corrections, the discrepancy is small so we have calculated their rate constants to
compare reaction possibility. Importantly we found that the reaction is initiated by CNO2 cleavage.
Keywords:
1.
decomposition mechanism, FOX-7, theoretical study, ab initio

molecular dynamics (AIMD), density functional theory (DFT) ,
kinetics
INTRODUCTION
In recent years there has been a significant effort to develop new candidates which might
lead to improved explosives in terms of performance, safety and other favorable properties.
FOX-7 is an energetic material with high performance approaching that of RDX, but with
low sensitivity comparable to that of TATB. For this reason FOX-7 is an explosive receiving
a good deal of attentions. It has a density magnitude of 1.878g/cm3. Very much has been
studied concerning FOX-7. We were aware that the synthesis had been reported [1] and the
crystal structure had been determined [2, 3]. The activation energy values scatter in papers of
different authors from 50 to 90kcal/mol [4, 5] at temperature range 200-250.
An understanding of the decomposition mechanism, however, also plays a key role in
the evaluation of an explosive. In this investigation of the decomposition mechanism of
FOX-7, we have focused on discovering the most possible initial decomposition pathway of
FOX-7. We have used the ab initio molecular dynamics simulation to investigate the
possible means of decomposition, then to demonstrate them by means of energy barrier and
rate constant calculations.
2.
syjfree@sina.com or syjfree@sohu.com
816
Czech Republic
2006
[Content]
Ab initio molecular dynamics simulations were performed with VASP program

package [6-12]. We have simulated the balance geometry at 2800K at which FOX-7 wound
not decompose but can obtain the easily decomposing geometries. The simulation time-step
was 0.5fs, the total time was 10ps. We have taken the balance geometries every 200 steps
to simulate their decomposition trajectories. A large part of these geometries will
decompose at 3300K after several trials above 2800K. The time-step was also 0.5 fs, and
the total time was 3ps, which is enough for the fast decomposition reaction. We found that
C-NO2 cleavage is the main initial decomposition pathway. Hydrogen migration from
amino to nitro and nitro-to-nitrite were also observed in the simulation trajectories. All
quantum calculations were carried out with Gaussian 98 program package [13]. The Becke
three-parameter hybrid (B3) [14] was used in conjunction with functional Lee-Yang-Parr
(LYP) [15]. For all calculations, a Gaussian-type basis set, 6-31G (d, p) was used. Thus,
geometries of all reactants, products, and transition states were optimized by the use of the
same method. Vibrational frequencies were calculated with the same approximation for
characterization of the nature of stationary points and zero-point vibration energy
corrections. All of the stationary points have been positively identified for minimum energy
with no imaginary frequency. Intrinsic reaction coordinate (IRC) analysis was carried out
for each transition state to make sure that it is the transition structure connecting the desired
reactants and products. For nitro dissociation that having no transition state, we have
carried out minimum energy pathway calculation of it. Rate constants were all calculated
with Vklab program [16]. We have observed the C-NO2 dissociation is the main
decomposition pathway in the ab initio molecular dynamics (AIMD) simulation, but the
bond dissociation energy (BDE) of C-NO2 is higher than the energy barrier of nitro-tonitrite rearrangement and the energy barrier of hydrogen migration is rather low but it is not
the main reaction in the ab initio molecular dynamics simulation. So we carried out rate
constant calculation to make sure which way is the main initial decomposition pathway.
For the reaction of having transition state, we have used Transition State Theory to
calculate their rate constants, while for the C-NO2 dissociation V T was used.
3.

3.1
Ab initio MD simulations with VASP
In the ab initio molecular dynamics simulations, we have found C-NO2 bond

dissociation is the main decomposition pathway and hydrogen migration from amino to the
nitro of the same side then HONO dissociation and nitro-to-nitrite were also observed.
Other initial reaction pathway such as C-C bond dissociation and hydrogen translation from
amino to the other amino were so seldom to observe so we have not taken them into
account. From figure 1 we can see the main initial reaction pathways obviously:
pathway 1:
NO2
H2N
C
H2N
H2N
NO2
C
NO2
H2N
O
N
O
pathway 2:
Hehou Zong, Yuanjie Shu, Yigang Huang, Xinfeng Wang
817

NO2
H2N
NO2
H2N
H2N
C
O
ON
C
H2N
H2N
Czech Republic
2006
H2N
C
H2N
NO2
[Content]
O
N
O
C
NO2
O
C
NO2
C
H2N
pathway 3:
C
H2N
H N
NO2
H2N
N
C
C
H2N
NO2
H O
H N
C
NO2
H2N
N
C
C
NO2
H N
C
H2N
H O
NO2
N O
Fig 1. A scheme of pathways for the main three initial decompositions of FOX-7
3.2
Quantum calculations with Gaussian 98
Quantum mechanical calculations were performed for the three pathways respectively.
The structures and optimized geometrical parameters of all species involved were obtained,
as well as their corresponding absolute energies.
3.2.1. C-NO2 dissociation.
We have calculated the reactant, intermediates and transition state of C-NO2
dissociation reaction. The scheme of the pathway is shown in figure 2. The total energy
and zero-point correction energy were tabulated in table 1.The main geometry
parameters are shown on the molecules. The energetic profile of the pathway can be seen
from figure 3.
3.06
NO2
H2N
1.43
1.34
1.43 C
C
1.34
H2N 1.43
NO
H2N
1.36
NO2
1.39
C
C 1.38
1.37
H2N
P1
NO2
H N
C
H2N
H 0.97
O
1.46
1.29N O
C 1.23
1.45
TS1
P2
H N
1.29
1.38C
H2N
1.45
H
O 0.97
1.43
1.29 N
O
1.24
C
M1
Fig 2. A scheme of unimolecular FOX-7 decomposition via C-NO2 dissociation reaction

pathway.
Table 1. the total energies and zero-point corrections of the reactants, products and
intermediates
code
Sum of electronic and zero- Zero-point
point Energies(Ha)
correction(Ha)
818
R
P1
P2
TS1
M1
point Energies(Ha)
-598.232973
-393.062668
-205.63380
-393.031713
-393.034178
Czech Republic
2006
[Content]
correction(Ha)
0.092451
0.077013
0.008826
0.074121
0.074791
P1+P2
TS1
70.13
1.97
21.24
R
M1
P1
Fig 3. The energetic profile of FOX-7 decomposition via C-NO2 dissociation reaction
pathway. The units are kcal/mol. The energy barriers are uncorrected for ZPVE.
The nitro dissociation from FOX-7 molecule has no transition state which has been
verified by scanning the potential energy surface of C-NO2 bond. It is can be seen from fig
4.
total energy vs C-NO2 bond length
-598.12
-598.14 0
-598.16
total energy
-598.18
-598.2
-598.22
-598.24
-598.26
-598.28
-598.3
-598.32
-598.34
C-NO2 bond length
Fig4. Potential energy surface of C-NO2 (the units are Ha and for energy and bond
length respectively)
The hydrogen migration from nitro to the amino for product P1 may take place through
transition state TS1 and then produce M1. The transition state TS1 has an imaginary
frequency of -307.5. IRC analysis verified it is the transition structure connecting P1 and
M1.The energy barrier for P1 is not high so the product may easily react after nitro
cleavage.
3.2.2. Nitro-to-nitrite rearrangement
We have calculated the reactant, intermediates and transition state of rearrangement
reaction. The scheme of the pathway is shown in figure 5. The total energy and zeropoint correction energy were tabulated in table 2.The main geometry parameters are
4
819
Czech Republic
2006
[Content]
shown on the molecules. The energetic profile of the pathway can be seen from figure 6.
1.16O
N
1.34
O
O 1.23
H2N
N
1.80
C
C 1.53
1.40
1.42 NO
H2N
2
NO2
H2N
C
H2N
NO2
1.58
H2N
C
H2N
M1
TS1
O
1.46 1.25
C 1.46
H N
N 1.22
O1.26 O
H
H2N
H2N
1.50
O
1.23
M2
H2N
C
1.23
H N
1.33 N O
1.25 H
O
1.21
NO
1.36O
C
1.39 1.41
NO2
H N
1.23
N O
H O 1.37
1.03
M4
TS2
M3
H2N
O
C
H N
M5
H O
N O
M6
Fig 5. A scheme of unimolecular FOX-7 decomposition via nitro-to-nitrite rearrangement

pathway
Table 2 The total energies and zero-point corrections of the reactants, products
and intermediates of the reaction of nitro-to-nitrite rearrangement reactions.
code
Sum of electronic and Zero-point
zero-point Energies(Ha)
correction(Ha)
R
-598.232973
0.092451
TS1
-598.137150
0.088415
M1
-598.238702
0.090345
M2
-468.356838
0.081007
M3
-129.883620
0.004536
TS2
-468.353212
0.077232
M4
-468.352168
0.080499
M5
-262.616769
0.054964
M6
-205.679912
0.020379
TS1
M5+M6
62.66
64.94
TS2
M2+M3
1.91
M1
37.99
1.40
4.64
M4
M2
Fig 6. The energetic profile of FOX-7 decomposition via nitro-to-nitrite rearrangement

pathway. The units are kcal/mol. The energy barriers are uncorrected for ZPVE.
The energy barriers, structures and optimized geometries parameters for all the species
are consistent with the values obtained by Polizer[17] and Asta[18]. The transition state TS1
820
Czech Republic
2006
[Content]
has an imaginary frequency of -819.1. The TS2 has an imaginary -1091.2. Vibration and
IRC analysis conformed they are the true transition states.
3.2.3. Hydrogen migration from amino to nitro
We have calculated the reactant, intermediates and transition state of rearrangement
reaction. The scheme of the pathway is shown in figure 7. The total energy and zeropoint correction energy were tabulated in table 3.The main geometry parameters are
shown on the molecules. The energetic profile of the pathway can be seen from figure 8.
H2N
C
NO2
H N
NO2
H2N
0.98
1.22
H2N
2.79 H 0.97
O1.43
O
1.33 N
1.50
H N
1.45
NO2
H2N
TS1
O1.40
O
1.32 N1.23
C 1.44
NO2
M1
H N
C
H2N
NO2
H O
M2
N O
M3
Fig 7. A scheme of unimolecular FOX-7 decomposition via hydrogen migration pathway.

Table 3. The total energies and zero-point corrections of the reactants, products and
intermediates for the reaction of nitro-to-nitrite rearrangement
code
Sum of electronic and zero- Zero-point

point Energies(Ha)
correction(Ha)
-598.232973
0.092451
-598.186743
0.090590
-598.189734
0.091462
-392.424701
0.063264
-205.679912
0.020379
R
TS1
M1
M2
M3
M2+M3
58.32
TS1
2.24
30.18
M1
Fig 8. The energetic profile of FOX-7 decomposition via hydrogen translation from amino
to nitro pathway. The units are kcal/mol. The energy barriers are uncorrected for ZPVE.
The structures and optimized geometrical parameters of all the species involved were
obtained. The TS1 has an imaginary frequency of -365.5.
3.3
6
Reaction rate constant calculation with Vklab program
821
Czech Republic
2006
[Content]
In the molecular dynamics simulations, we have found that C-NO2 bond dissociation is
the main initial decomposition pathway. But the quantum calculation results are not
consistent with it, but C-NO2 bond dissociation energy is only about 7kcal/mol higher than
the energy barrier of nitro-to-nitrite rearrangement reaction. So we calculated these reaction
rate constants to confirm the initial decomposition pathway. The rate constants of three
decomposition pathways are shown in table 4.
Table 4 The reaction rate constants of FOX-7
temperatures
rearrangement
C-NO2
T(K)
cleavage
forward
backward
400
1.388E-18
3.474E-20 4.770E-23
450
9.623E-15
1.770E-16 3.672E-19
500
1.158E-11
1.654E-13 4.733E-16
550
3.877E-09
4.492E-11 1.663E-13
600
4.956E-07
4.825E-09 2.203E-11
650
3.002E-05
2.536E-07 1.379E-09
700
1.006E-03
7.595E-06 4.787E-08
though three pathways at different

Hydrogen migration
forward
6.558E-03
4.234E-01
1.194E+01
1.840E+02
1.802E+03
1.245E+04
6.539E+04
backward
4.587E+11
6.435E+11
8.452E+11
1.058E+12
1.276E+12
1.497E+12
1.718E+12
750
800
850
900
950
1000
2.091E-02
2.940E-01
2.994E+00
2.330E+01
1.448E+02
7.441E+02
1.450E-04
1.919E-03
1.878E-02
1.428E-01
8.787E-01
4.512E+00
1.037E-06
1.531E-05
1.648E-04
1.363E-03
9.034E-03
4.958E-02
2.755E+05
9.708E+05
2.951E+06
7.934E+06
1.923E+07
4.267E+07
1.936E+12
2.151E+12
2.360E+12
2.564E+12
2.762E+12
2.953E+12
1100
1200
1300
1400
1500
1.241E+04
1.282E+05
9.208E+05
4.986E+06
2.153E+07
7.631E+01
8.074E+02
5.953E+03
3.303E+04
1.460E+05
9.402E-01
1.093E+01
8.725E+01
5.180E+02
2.427E+03
1.692E+08
5.337E+08
1.411E+09
3.249E+09
6.695E+09
3.317E+12
3.655E+12
3.968E+12
4.259E+12
4.529E+12
From table 4, we can find that the rate constants of C-NO2 cleavage are approximately
100 times to that of the arrangement reaction at the same temperature. The rates are so high
for hydrogen migration but backward rate is much higher and the total energy barrier from R
to M1 and M2 is over 80kcal/mol, so this pathway is not the dominant initial decomposition
pathway.
4.
CONCLUSION
In this paper the calculations made and the conclusions drawn are based on ab initio
molecular dynamics, DFT methodology and chemical kinetics. Using AIMD simulation, we
observed C-NO2 bond dissociation is the dominant trajectories. Quantum mechanical
calculations were performed for the three main initial decomposition pathways. The energy
barrier of hydrogen migration from amino to nitro is the lowest, but its intermediate has high
energy and its consequent reaction requires comparatively high energy. The BDE of C-NO2
is about 7kcal/mol than the energy barrier of nitro-to-nitrite rearrangement reaction at
B3LYP/6-31G(d, p) level without zero-point correction. The chemical kinetics calculations
indicate that the rate constant of C-NO2 cleavage is higher than that of rearrangement
822
Czech Republic
2006
[Content]
reaction, and the rate constant of hydrogen migration is very high but that of backward
reaction is much higher. So the results observed and calculated can make out that C-NO2
bond cleavage is the most important initial decomposition pathway. Its bond dissociation
energy is 70.13kcal/mol uncorrected for ZPVE at B3LYP/6-31G (d, p) level.
REFERENCES
[1] Nikolai V LatypovJan Bergman. Synthesis and reactions of 1,1-diamino-2,2-dinitroethylene.
Tetrahedron1998(54)11525-11536.
[2] Bemm, U., and stmark, H. (1998), .1,1-Diamino-2,2-dinitroethylene: a novel energetic material
with infinite layers in two dimensions,. Acta Crystallographica C54, 1997-1999.
[3].Gilardi, R Crystallographic data (excluding structure factors) for the DADNE structure have been
deposited with the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2
1EZ,U.K.(deposition number CCDC127539).
[4] stmark H.,Langlet A.,Bergman H.,Wingborg,N.,Welmar U. and Bermm
U.(1998),FOX-7-A new explosive with low sensitivity and high performance
Proc.Eleventh(Int.)Det.Symp. Preprints1821.
[5] V.P.Sinditskii,A.L.Levshenkov,V.Y.Egorshev,v.V.Serushkin
Study on Combustion and Thermal Decomposition of 1,1-Diamino-2,2-dinitroethylene(FOX7)//8th International Seminar EUROPYRO2003,Saint Molo,France,2003.
[6] Car, R.; Parrinello, M. Phys. Rev. Lett., 1985, 55, 2471.
[7] Payne, M. C.; Teter, M. P.; Allan, D. C.; Arias, T. A.; Joannapoulos, J. D. Rev. Mod. Phys.,
1992, 64, 1045.
[8] Kresse, G.; Hanfner, J. Phys. Rev. B, 1993, 47, 558.
[9] Kresse, G.; Hafner, J. Phys. Rev. B, 1994, 49, 14251.
[10] Kresse, G.; Furthmller, J. Comput. Mat. Sci., 1996, 6,15.
[11] Kresse, G.; Furthmller, J. Phys. Rev. B, 1996, 54, 11169.
[12] Vanderbilt, D. Phys. Rev. B, 1990, 41, 7892.
[13]
[14]
[15]
[16]
[17]
Gaussian 98, Revision A.3, M. J. Frisch etal., Gaussian, Inc., Pittsburgh PA, 1998.
Becke, A. D. J. Chem. Phys.1993,98,5648.
Lee, C.; Yang, W. ; Parr, R. G. Phys. Rev. B1988, 37, 785.
Zhang S, Truong T N. VKLab. Version 1.0 ed.; University of Utah, 2001.
Politzer, P., Concha, M. Grice, M., Murray, J., Lane, P., and Habibollazadeh, D.(1998)
.Computational investigation of the structures and relative stabilities of amino/nitro derivatives
of ethylene,. Journal of Molecular Structure (THEOCHEM) 452, 75-83.
[18] Sta Gindulyte and Lou Massa. Proposed Mechanism of 1,1 Diamino Dinitroethylene
Decomposition A Density Functional Theory Study. J. Phys. Chem. A 1999,103,11045-11051.
823
Abstract
Czech Republic
2006
[Content]
POLYMORFIC TRANSFORMATIONS AND THERMAL

DECOMPOSITION OF AMMONIUM NITRATE AT THERMAL
ANALYSIS CONDITIONS
E.N. Egorychev, V.N. Lashkov, A.V. Strikanov, A.A. Tikhonova, V.I. Rybakov
RFNC-VNIIEF, Sarov, Russia
Abstract:
Ammonium nitrate (AN) is the basic component of explosive compositions (EC) in
explosion welding technology. During mix preparation of the explosive composition,
ammonium nitrate is reduced to fine particles in ball mill and then mixed with
trinitrotoluene to obtain a composition with 5%-mass content of trinitrotoluene.
Actually, we are carring out studies to obtain EC without trinitrotoluene to utilize its in
explosion welding technology.
The main requirements to such EC are:
Detonation initiation by standard detonator
Detonation velocity at charge porosity of 50% is not more than 1500 m/s
Critical thickness is not more than 20mm, slight dependence of detonation velocity on
thickness
We propose the following approach of obtaining satisfactory values in the factors:
Grinding should be more fine, with control of the degree of grinding;
Additives of dispersed metal powder are to be used.
Crystalline structure activation by polymorfic transformation.
In this report we present the results of thermal analysis of commercially available AN
and finely dispersed AN. We determined the parameters of polymorfic transformations
(direct and inverse). We also determined effect of aluminum and titanium powder
additives on thermal decomposition process of ammonium nitrate. It was shown that
inverse polymorphic transitions are running without a delay which means, that the
phase structure of trial sample is recovered after cooling.
Partical size of metal additives affects considerabled on the process of thermal
decomposition of ammonium nitrate. For example, coarse titanium powder that has
particle size more than 20 micrometer decreases heat effect of AN thermal
decomposition while dispersed titanium powder (particle size less than 5 micrometer)
increases it.
Valeriy Lashkov, E. Egorychev, A. Strikanov, A. Tikhonova, V. Rybakov
825
Czech Republic
2006
[Content]
Author Index
Adam Martin 613
Adamk Vladislav 451
Alekseev Alexander 293
Aleshkina Ekaterina 464
Aleynikov Aleksey 304
An-Min Tan 356
Annikov Vladimir 464
Anusevicius Zilvinas 285,
689
Apolenis Alexei 464
Arinina Svetlana 473
Armstrong K. 42
Astachov Alexander 480
Astratev Alexandr 377
Backofen Joseph 76
Bachurina Irina 143
Bajpai Rakesh 251
Bakhmatova Eugeniya 90
Baka Sanko 214
Barbu Cristian 756
Bazrafshan Esmaeil 393
Bellamy Anthony 491
Bladek Jan 500
Blagojevic Djordje 513
Boniuk Henryka 279
Bozic Vladica 513
Bratowski M. 742
Buczkowski Daniel 507
Buka Eduard 480
Burov Yurii 98, 526
Bystrzejewski Michal 544
Cartwright Michael 107
Catovic Alan 780
Cenas Narimantas 285, 689
Clment Dominik 540
Cudzilo Stanislaw 500, 544
Cumming Adam 11
Czerski Helen 118
Dabkowski A. 742
Darwich Chaza 551
Dashko Dmitry 377
Dimitrov A. 416
Dobrilovi Mario 581
Duda Maciej 561
Dudek Kamil 608
Egorova Irina 569
Egorshev Valerii 314
826
Egorshin Vyacheslav 143

Egorychev E. 825
Eisner Ale 613
Ek Stefan 160
Eldster Carina 160
Ester Zvonimir 581
Field John 118
Florczak Bogdan 507, 587,
693
Fojtk Tom 613, 618
Frczak Micha 268
Friedl Zdenk 796
Gawor Tomasz 598, 764
Gbel Michael 127, 202
Goede Patrick 160
Gobek Barbara 279
Gromov Alexander 135
Gruzdev Yurii 143
Guillaume Pierre 150
Hadi Jordanov S. 416
Halecky Martin 251
Held Manfred 16
Horkel Jan 668
Huang Yigang 816
Huczko Andrzej 544
Chemezova Ludmila 293
Chernay Anatoliy 602
Chovancov Martina 529
Ilyushin Mikhail 143, 602
Ivanov Victor 293
Ivshin Victor 260
Jalov Zdenk 608
Jankovi Branimir 581
Janovsk Betislav 451
Jeunieau Laurence 150
Jeov Vra 613, 618
Jigatch Alexey 404
Johansson Martin 160
Jones David 42
Karaghiosoff Konstantin
202
Kasperski Johann 279
Katorov Dmitrii 624
Klapoetke Thomas 631
Klaptke Thomas 127,

202, 551, 641
Ko Ji 796, 810
Kondrikov Boris 569
Korolev Vyacheslav 90, 260
Korsze D. 598
Kouba Martin 723
Kovalenko Igor L. 652
Kov Petr 712
Kozak George 657
Kozak Georgii 464, 473
Kruglyakova Ludmila 480
Krupka Miloslav 329, 668
Ksiczak Andrzej 368
Ksiczak Teresa 368
Kucherova Irina 526
Kuprin Alexandr V. 652
Kuskov Michael 404
Kwok Q. 42
Kwon Young-Soon 135
Kyslkov Eva 618
Ladonin A. 624
Laricheva Olga 404
Laritchev Mikhail 404
Lashkov Valeriy 673, 825
Latypov Nikolaj 160
Lefebvre Michel 150
Leipunsky Ilia 404
Lempert David 169
Lewis Andrew 107
Lika Frantiek 608
Lobanova A. 701
Lomax Victoria 491
Lopch Jozef 529
Lukin Alexander 181
Majzlk Ji 796, 810
Maksimowski Pawe 561
Maldutis Evaldas 285
Manelis G. 169
Maranda Andrzej 279,
680, 693
Mareek Pavel 608
Matj Vt 618
Mayer Peter 631
Meyer Peter 641
Milos Marko 513
Mir Carles 202
Miseviciene Lina 285
Mostak Petr 61

Mudrukov Martina
Murray Jane 65
Muani Sanja 214
723
Nefedov Andrew 480

Nechiporenko G. 169
Nemeikaite-Ceniene Ausra
285, 689
Nivinskas Henrikas 285,
689
Nowaczewski Jerzy 680,
693
Oko Peter 529
Odak Mile 581
Ondrejkovi Mat 608
Orzechowski Andrzej 587,
680, 693
Osmonaliev Mirslan 293
stmark Henric 231
Pca Jan 251
Pantel Gabriele 338
Parekunnel T. 42
Paszula Jzef 748
Paunovi P. 416
Pechov Albta 529
Pelikn Vojtch 796, 810
Petrov D. 701
Petukhova Tatyana 260
Pietrzyk Sylwia 500
Piotrowski Tadeusz 268
Pivina Tatyana 90, 260
Polborn Kurt 631, 641
Politzer Peter 65
Popenko E. 701
Popenko Elena 135
Popovski Orce 416
Porollo Aleksey 90
Pospil Miroslav 712
Pourmortazavi Seyed 393
Powaa Dorota 680, 693
Prchal Pavel 719
Proud William 118
Pshechenkov Pavel 404
Czech Republic
Raikova Vlada 657

Read Kate 491
Revenko Vitaliy 480
Rudakov G. 624
Rudolf Karl 540
Rybakov V. 825
Safta Doru 756
Saaciski Tomasz 268,
279, 507, 587
Sarlauskas Jonas 285, 689
Sedoi Valentin 293, 404
Selesovsky Jakub 668
Selezenev Alexander 304
Serdarevic-Kadic Sabina
780
Sergienko A. 701
Sergienko Alexey 135
Serushkin Valery 143
Shandakov V. 701
Shen Zhaowu 423
Shu Yuanjie 769, 816
Shugalei Irina 602
Sinditski Valerii 314
Sinditskii Valery 143
Skldal Jan 723
Skupiski Wincenty 561
Slavkov D. 416
Slezk Miloslav 723
Smileski D. 416
Smirnov Sergei 314
Sokolov E. 701
Stancl Miroslav 61
Stepanov Andrei 377
Stierstorfer Jrg 641
Stojan Petr 730
Strikanov A. 673, 825
Sueska Muhamed 214
Szala Mateusz 544
eleovsk Jakub
toller Ji 736
329
Tasev M. 416
Teipel Ulrich 135
Teodorczyk Andrzej
2006
[Content]
Terzic Jasmin 780

Tian Anming 769
Ticmanis Uldis 338
Tikhonova A. 673, 825
Toftunova Valeria 602
Tomii Zvonimir 581
Trzciski Waldemar A. 748
Tselinsky Igor 143, 602
Vasile Titica 756
Vvra Pavel 712
Ventura Karel 613, 618
Veselova E. 624
Volkova Galina 293
Voshlov Simona 618
Vyalova N. 624
Walenta Zbygniew 764
Wang Xinfeng 769, 816
Weigand Jan 641
Welch Jan 202
Wilker Stephan 338, 500
Witkowski Waldemar 279
Wlech Jan 631
Wolszakiewicz Tomasz
598, 764
Wong Ning-Bew 769
Xin Wang 356
Xin-Ping Long 356
Xiong Ying 769
Yang Zhou 356
Yuan-Jie Shu 356
Yudina-Wahlstrm Larisa
160
Zecevic Berko 780
Zeman Svatopluk 796, 810
Zhao Huiying 423
Zharinov Y. 701
Zhilin V. 624
Zhou Ge 769
Zielenkiewicz Tomasz 368
Zong Hehou 816
742
827
UP 06-19
ISBN 80-7194-849-7

University of Pardubice - 9th International Seminar NTREM (2006)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

University of Pardubice - 9th International Seminar NTREM (2006)

Uploaded by

Copyright:

Available Formats

proceedings of

New Trends in Research

Pardubice, April 1921 2006

University of Pardubice, Faculty of Chemical Technology

proceedings of the 9th Seminar

New Trends in Research

held at the University of Pardubice

intended as a meeting of students, postgraduate students, university teachers,

9th Seminar of the New Trends in Research of Energetic Materials

DSTL, Sevenoaks, U.K.

IEM, Univ. of Pardubice, Czech Republic

New Trends in Research of Energetic Materials

European Govermental Collaboration in Energetics Research The Future

New Diagnostic Techniques in Blast Waves

The Thermal Behaviour of BTAW, a High Nitrogen Fuel

New Trends in Detection of Explosives

Molecular Surface Electrostatic Potentials in the Computational

The Influence of Geometry and Material Properties on an Explosives

Computer Simulation of Thermal Decomposition Mechanism for

Thermal Decomposition of Solid Energetic Materials

Use of Compressed Gas Gun to Investigate Response of Explosives to a

The Relationship between Shock Sensitivity and Morphology in Granular

Guanidinium Nitroformate Salts: Possible New Oxidizers for High

New Trends in Research of Energetic Materials

Preliminary Study of Superfine Metal Powders Effect on the Decomposition

Preparation, Properties and Application of Energetic Complex Perchlorates

Balistic Stability of a Spherical Propellant: Comparison with a Flattened

Synthesis and Characterization of 1,2-Dinitroguanidine (DNG) and its

Energetic Characteristics of Solid Composite Propellants and Ways for

Mechanism of Low-Frequency Non-Acoustic Instability Phenomena From

Stable Salts of Methylated 5-Aminotetrazoles

Applicability of Dynamic Mechanical and Thermal Methods in Investigation

High Energy Density Materials (HEDM): Overview, Theory and Synthetic

Aerobic Biodegradation of Mononitrotoluenes in Different Reactor

Polyfunctional N-nitramines Structure Differentiation as a Basis for

Process Hazard Evaluation for Explosives Manufacturing Using

Determination of the Thermal Resistance of the Emulsion Explosives

New Trends in Research of Energetic Materials

Reduction of Nitroaromatic Explosives by Oxygen-insensitive

Complex Research of Al Nanopowders Produced by the Exploding Wire

Ab Initio MD Computation of the Vibrational Relaxation Time in HE

Unusual Thermal Decomposition of NTO: Is It a Result of Very Strong

The principal approach to the estimation of cookoff temperature using FEM

Low Temperature Activation Energy of Propellants

Theoretical Study of the Tetrazines C-N Heterocyclic Derivatives

Self-aggregation of Thermal Decomposition Products

Study of the Routes for Producing Nitroguanidine(NQ) in Aqueous Nitric

Fuel Dispersion in Air by Detonation Wave

Preliminary Study of the Behavier of Al Nano Powders Water Slurry

Carbon Nanotubes as Support for Co/TiO2 Electrocatalysts for Hydrogen

Sliding Shock Wave Causing Deformation of the Metal Plate Attached to

New Trends in Research of Energetic Materials

Simulation of Coffee Mill Explosion Accident Underatex 137 Directive

Explosion Hazard of Water Gel-like Systems on a Base of High Explosives

Heat of Melting of Nitrocompounds Measuring by Means of the Differential

Characterization and some Properties of 1-Nitroamidino-3-amino-5nitrimino-1,2,4-triazole

Investigation of Stabilizing Properties of Triphenylamine (TPA)

Some Properties of a Composite Propellant Premix

Thermophysical Properties of Thermoplastic Composite Propellants Based

Thermal Decomposition of the trans-1,4,5,8-Tetranitro-1,4,5,8tetraazadecalin

Observation of Changes in Chosen Properties of Plastic Explosives During

Hexanitrostilbene - Density Dependend Sensitivity

Self-sustaining reductive defluorination of (CF)n and characterization of the

Insensitive Environmentally-compatible and Non-toxic Replacements for