C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q.

, 30 (4) 419–427 (2016) 419

A CFD-based Approach to Predict Explosion Overpressure:

A Comparison to Current Methods
C. Díaz-Ovalle,a,* A. López-Molina,b and R. Vázquez-Románc
a
Tecnológico Nacional de México, Instituto Tecnológico de Roque,
Departamento de Ingenierías, Km 8 carretera Celaya-Juventino Rosas,
Celaya, 38110, Gto., MEXICO
b
Universidad Juárez Autónoma de Tabasco, División Académica
Multidisciplinaria de Jalpa de Méndez, Ingeniería en Petroquímica,
carretera Estatal Libre Villahermosa-Comalcalco Km. 27 s/n
Ranchería Rivera Alta, Jalpa de Méndez, 86205, Tab., MEXICO doi: 10.15255/CABEQ.2015.2244
c
Tecnológico Nacional de México, Instituto Tecnológico Original scientific paper
de Celaya, Departamento de Ingeniería Química, Av. Tecnológico y A.G. Received: June 11, 2015
Cubas s/n, Celaya 38010, Gto., MEXICO Accepted: December 5, 2016

A CFD-based approach has been developed in this work to predict the overpressure
produced during an explosion. An adiabatic exothermal reaction allows computing the
explosion energy release. To validate the proposed CFD approach, overpressure predic-
tions based on this methodology are compared with results produced with the TNO-
based method. It is demonstrated that the physics adopted in our model produces satis-
factory predictions in the open area. The CFD simulations were carried out in the ANSYS
CFX tool. The source of energy corresponds to the one produced by a stoichiometric
proportion in reactants without energy generation. The explosion analysis considered that
explosion occurs geometrically as a sequence of control volumes. Thus, the explosion in
a volume is assumed to occur when the maximum pressure is achieved in the previous
control volume. This way, the explosion is propagated and it is shown that it is equivalent
to conventional predicting methods.
Key words:
CFD, explosion, overpressure, TNO method

Introduction the overpressure wave; however, both final and ini-

Explosions represent a hazard frequently found
in the process industry. It consists of a rapid com-
bustion of clouds producing a fast release of energy
and gas expansion. Several explosion accidents
have occurred in processing plants1. Human and
economic losses justify the study of these accidents.
Explosions are complex phenomena, which come
from a sudden and rapid release of energy. This en-
ergy is dissipated in the form of heat radiation,
acoustic waves, and waves of pressure. Explosions
differ in characteristics depending on their source of
generation, i.e. liquid at boiling, reactions out of
control, and partially confined vapour2. The nature
of explosions limits their study, due to: a) low oc-
currence time scale, b) uncertainty in the pressure
wave, and c) uncertainty on the energy released.
This last point has been estimated using first-princi-
ples of thermodynamic equilibrium for the system.
This estimation includes the mechanical energy of
*
Corresponding author: Christian Díaz-Ovalle. E-mail address:
chdiaz@itroque.edu.mx; Phone: +52 461 61 15903 ext 141
tial conditions of the system are typically unknown3.
Thus, the prediction under thermodynamic equilib-
rium has involved experimental data.
A simplified model consists of mathematical
correlations with parameters of occurrences under
real-world scenarios. This kind of model can reli-
ably represent explosions and determine the profile
of overpressure. Among other, the TNT equiva-
lence4, TNO Multi–energy5 and Baker–Strehlow6.
The results of these methods have been used to al-
locate process units and containment systems. Un-
fortunately, these models do not consider the effect
of obstacles in the explosion. Despite this, these
models have been used frequently due to their sim-
plicity. Nowadays, the CFD (Computational Fluid
Dynamics) predicts the phenomenon of explosion
with greater precision accounting for the local envi-
ronment such as confinement, congestion and
blockage issues.
CFD has been applied in various areas of engi-
neering. CFD represents a solution to the transport
equations under their spatial discretization. In pro-

C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure

419–427
420 C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016)
cess safety, CFD has been used to predict gas and
vapor dispersion7, blast from gas, pool fire8 and dust
explosions9–12. The transport equations include an
energy source parameter when modelling explo-
sions. CFD has solved problems of simulation for
explosion under various physical scenarios such as:
a) an obstructed environment equivalent to a porous
system13, b) a confined environment14,15 and c) vent-
ing ducts16. In these works, the results indicated that
the strength of an explosion increases when increas-
ing confinement and turbulence. Other works have
incremented the complexity in the system due to
dispersion scenarios with the allocation of the igni-
tion source17. These approaches described congest-
ed with forced ventilation systems18, open systems19
and diverse geometries20.
CFD packages simulate the hydrodynamics of
fluid in a variety of systems. The Gexcon package
FLACS is becoming a leader in simulation of ex-
plosions. FLACS results have been validated with
rigorous experiments in different scenarios21. This
package predicts dispersions of dense gases under
the k-ε model of turbulence22. FLACS has estimated
potential overpressures in several case studies23 and
accidents24. Explosion calculation by FLACS con-
siders combustion in a stoichiometric gas mix.
During combustion, the fluctuations of species in-
fluence the kinetic energy dissipation due to the use
of the Arrhenius model25. Temporal and spatial ef-
fects of scale re-evaluate the dissipation of energy
under source terms. Turbulence kinetic energy and
momentum equations include these source terms26.
However, this approach requires the continuous
consumption of gas in stoichiometric proportion,
which can be considered as equivalent to a series of
explosions in a time and space of low scale. This
generates the prediction of an explosion as in an
adiabatic system. The method TNO predicts over-
pressures at different levels of congestion, but in
practical situations, the level of congestion cannot
be defined in the majority of cases.
In this work, ANSYS-CFX simulations are car-
ried out without using a term for energy generation.
The explosion analysis is based on a series of con-
trol volumes without congestion, where the explo-
sion in a control volume starts at the time of its
maximum pressure to end up in a geometrical series
of explosions.
The proposed strategy
The explosion produces three phenomena:
combustion, turbulence, and gas expansion. Com-
bustion starts as a flame whose instability increases
the rate of combustion. This causes further expan-
sion of gases and turbulence. Turbulence acceler-
ates the burning, and the amount of fuel is propor-
tional to the expansion of gases. This description
corresponds to a cycle of power generation, which
is characteristic for an explosion27,28.
The explosion occurs when gas concentration
belongs to the interval defined by the lower and up-
per flammability limits. These limit values are spe-
cific for each fuel. In this analysis, the origin of the
explosion is a cloud of gas fuel in dispersion. When
the concentration achieves values inside the flam-
mability limits, the cloud can be ignited. The whole
cloud is then represented through a series of control
volumes. These volumes allow describing the ex-
plosion as a series of consecutive explosions. Gas
cloud data are taken at 7 minutes after having the
release, as suggested for evacuation purposes.
The proposed strategy starts by simulating the
gas dispersion using the code CFX 14.0 of ANSYS.
The gas dispersion during 7 minutes is simulated,
and the size of the cloud estimated. The next step
corresponds to the activation of the adiabatic reac-
tion assuming that the explosion occurs adiabatical-
ly in the called first volume of control. This first
volume corresponds to the one where ignition is
expected to be produced. Then, the explosion is
modelled in a process simulator (ASPEN Plus 7.2
in this case) to estimate the resulting temperature,
pressure, and mass composition in this volume.
These values are plugged into the CFD simulation
to obtain the effect on surrounding volumes. When
conditions for explosion in surrounding control vol-
umes are achieved, explosions are produced, and
the process simulator is used to calculate again the
resulting conditions. The procedure is repeated until
no more explosions can be produced in the whole
system. Figure 1 shows graphically the proposed
methodology.
CFD model
CFD is used in this work to describe two sce-
narios. The first case corresponds to the cloud gen-
eration during the gas release, while the second case
corresponds to modelling the escalation of the first
explosion into radial explosions. Both cases corre-
spond to transient simulation with compressible flu-
ids, and the governing equations are identical for
both scenarios, i.e. conservation of mass, momen-
tum, energy, and species. These equations are given
in Table 1. The main variables are: v, velocity vec-
tor, P, pressure, τ, shear stress tensor, g, gravitation-
al acceleration vector, T, temperature, q, heat flux
vector, Y, mass fraction per specie i. The main cal-
culated properties are: ρ, density, CP, specific heat
capacity, Dij, kinematic diffusivity of components
i-j. The kinematic diffusivity is calculated by:
 80

C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016) 421

aT1.51.5
aT
=
=DDijij , ,ii≠≠ jj
(1) Dispersion
bb++cT
cT Starts

where a, b and c are correlation parameters for the

Values at 7 min
binary system i-j. Values at 7 min
are reading in
are reading in CV
control volume
The numerical solution of the governing equa-
tions is rather complicated. The nonlinearities in the ASPEN
ASPENcomputescomputes
momentum equation are due to the turbulence29. A adiabatic
adiabaticstoichiometric
stechiometric
wall adjacent to the fluid generates turbulence, ratereaction
rate reaction
which is treated under theories of scale and it should
include a model for energy dissipation. Turbulence Values
Valuesofof
control
CV ANSYS CFX solves
are taken
volume are taken Turbulence models
theory has a number of models based on different
assumptions. The κ-ε model contains two equations,
which are appropriate for an open space with low Is necessary NO Analyze
YES
wall effect. The standard κ-ε model has been de- another results
simulation? End
scribed by Launder and Sharma30, and it includes a
variation in the turbulent kinetic energy, κ: Figure 1 Schematic representation of the proposed methodology
F i g . 1 – Schematic representation of the proposed
∂k
+ v ⋅ ∇k=  : ∇v − ε + ∇ ⋅ ( ( v + vT ) ∇k ) (2) methodology
∂t
and the variation of the turbulence dissipation, ε:
strategy of CFD in the ANSYS Workbench. Thus,
∂ε ε ε2  v   the strategy includes: development of the geometry,
v ⋅ ∇ε 1.44 : ∇v − 1.92 + ∇ ⋅   v + T  ∇ε  (3)
+=
∂t k k   1.3   set-up of mesh, definition of the scenario, and nu-
merical solution. The cloud dispersion is represent-
where v is the kinematic viscosity and vT is the tur- ed in an open geometry. The geometry mesh in-
bulent kinematic viscosity that is calculated by: cludes tetrahedral elements (20,084 nodes and
k2 91,383 elements) and hexahedral elements (27,054
vT = 0.09 (4) nodes and 15,265 elements. The mesh metric quali-
ε ty was obtained for skewness, aspect ratio, and or-
The solution of this equation is possible in all thogonal quality. The skewness indicated how close
CFD codes. The following section describes the the cell size approached the optimum cell size in a
scenarios considered in the dispersion conditions circumsphere. A value close to zero is the most de-
and the characteristics of the CFD-based approach. sirable, and the skewness value for the dispersion
zone was 0.274. The aspect ratio is defined as the
relationship between the length of the longest side
Case study and the length of the shortest. Values lower than 40
are considered acceptable, and the aspect ratio val-
The solution of the governing equations quanti- ue for the dispersion zone was 1.957. The orthogo-
fy the gas dispersion and the dissipation of an ex- nal quality indicated the minimum value of the or-
plosion. The solution starts with the conventional thogonal relation between (A, f) and (A, c), where

Ta b l e 1 – Governing equations
Equation Description

∂ρ
= ∇ ⋅ ρv Continuity
∂t
∂
( ρ v ) = −∇ ⋅ ρ vv − ∇P − ∇ ⋅  − ρ g Momentum conservation
∂t

∂
( ρ CPT ) = −∇ ⋅ ( ρ CPTv ) − ∇ ⋅ q −  : ∇v Energy conservation
∂t

∂
( ρYi ) + ∇ ⋅ ( ρYi v ) = −∇ ⋅ J i Specie i conservation
∂t
 81

422 C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016)

82

Figure 2 TopFview ofTop

i g . 2 – theview
meshing for gas
of the meshing fordispersion
gas dispersionscenario
scenario

–5
Figure 3FPropane cloud atcloud
i g . 3 – Propane 7 min,
at 7iso-surface withwith
min, iso-surface mass fraction
mass fractionof 5 ⋅10
of 5 10–5

A is the face normal vector, f is a vector from the
centroid of the cell to the centroid of the face, and c
is a vector from the centroid of the cell to the cen-
troid of the adjacent cell. The most desirable for
these cases is a value close to 1, and the orthogonal
quality value for the dispersion zone was 0.835.
Figure 2 shows a top view of the mesh for the gas
dispersion.
The source of emission is propane, and its fea-
tures are: a) flow of 1 kg s–1, b) temperature of 288
K, and c) discharge through a pipe of 0.0254 m of
diameter at 1 m above floor level. The Peng-Robin-
son equation of state is used to model the gas be-
havior. The atmosphere is standard with the follow-
ing values: a) temperature 298 K, b) pressure
101354 Pa, and c) 0.233 mass fraction for O2 and
the rest being N2. Terrain roughness is considered as
0.0001 m. This scenario does not consider the effect
of wind, while viscosity and thermal conductivity
have the standard values for air. Kinematics diffu-
sivity follows the expression (1), and parameters
have values of: a = 5.826 ⋅ 10–5 m2 s–1, b = 1.8033
K0.5, and c = –45.1061 K1.5. The solution for disper-
sion took increments of time variables until 7 min.
Figure 3 shows the fraction of 5 ⋅ 10–5 for propane
after running 7 min of dispersion, and Figure 4
shows the concentration and pressure profiles.
A total of eleven control volumes, cubes with 2
m/side, were considered. A volume for the neigh-
borhood was also included. This way, control vol-
umes were immersed in the neighborhood. The
mesh for this geometry had 18,550 nodes and
12,338 hexahedron elements. Figure 5 shows the
mesh using two perspectives.
Results and discussion
The dispersion generated a cloud of 80 m in
length. It required the use of 40 control volumes;
however, 11 control volumes were applied to save
C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016) 423

6E-03 6E-03
4.44 4.44
5E-03 5E-03
4.42 4.42
Propnae mass fraction

Propnae mass fraction

Pressure (Pa)

Pressure (Pa)
4E-03 4E-03 4.4 4.4

3E-03 3E-03 4.38 4.38

4.36 4.36
2E-03 2E-03
4.34 4.34
1E-03 1E-03
4.32 4.32
0E+00 0E+00 4.3 4.3
0 100 2010 3020 4030 5040 6050 7060 8070 9080 90 0 100 2010 3020 4030 5040 6050 7060 8070 9080 90
Distance to source
Distance (m)
to source (m) DistanceDistance
to source
to(m)
source (m)
a b
F i g . 4 – Cloud profiles: a) propane mass fraction, b) pressure

a) a

b) b

Figure 5Control
F i g . 5 – Control volume
volume meshing: a) wireframe meshing: a)viewwireframe view, b) ex
view, b) external faces

view
 424 C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016)

500 the mass fraction of fuel as the mixture becomes

480
500 leaner. Figure 7 describes the pressure in each con-
460
480 trol volume before the blast, as well as the maxi-
Temperature (K)

460
440 mum overpressure effect after the blast. These val-
Temperature (K)

440
420 ues result from the aerodynamic behavior of
420
400 combustion gas inside control volumes. These pro-
400
380 files are independent of the cloud pressure profile
380
360 during dispersion, Figure 4b. Both profiles follow
360
340 the tendency of the fraction mass during the disper-
340
320 sion. This feature shows the dependence of the
320 overpressure on the amount of mass.
300
300
0 10 20 30 40 50 60 70 80 90 Figure 8 represents the relationship between
0 10 20 30 40 to50
Distance 60(m)70 80 90
source
Distance to source (m) the pressure before the explosion and the overpres-
F i g . 6 – Temperature profile at explosions sure. This relationship is linear and has greater val-
Figure 66Temperature ues in zones with highest concentration in the cloud.
Figure
180
Temperature profile at explosions
profile at explosions
9 A similar analysis was applied to the temperature
180 9
170 8 before the explosion and temperature of the explo-
170 8 (kPa)
sion. The relationship between both temperatures
180 9
(kPa)
160 7
7 was similar to the relationship of pressures. Figure
(KPa)

160 170 8
blast
Overpressure(KPa)

6
Pressure prior to blast (kPa)

9 shows the relationship between temperature prior

blast

150 160 7 6
150
Overpressure (KPa)

5
to the explosion and temperature of the explosion. A
toto

6
5
Overpressure

140 150
prior

140 5 4 high amount of points kept high temperature values.

prior

140 4
130 4
130 3
Pressure

130
3 3
Pressure

120 120 2
120 Overpressure (kPa)
Overpressure (kPa) 2
2
110 110 Overpressure
Pressure (kPa)(kPa)
Pressure (kPa) 1
110
1 180
100
Pressure (kPa) 0
1
100 0 180
170
100 0 10 20 30 40 50 60 70 80 90 0
170
(kPa)

0 10
30 Distance 20
40 (m) 50 60 70 80 90 160
0 10
30 Distance
40 20 50
(m) 60 70 80 90
(kPa)

Distance
Figure 7 Profiles of pressure prior to blast and(m)
overpressure
160
150
Overpressure

7 – Profiles
F i g7 .Profiles
Figure of pressure
of pressure priorand
prior to blast to blast and overpressure
overpressure 150
140
Overpressure

Figure 7 Profiles of pressure prior to blast and overpressure 140 Zone

Zone ofofgreater
greater
130 Concentration
of greater of
concetration
Zone
the cloud of
130 Concentration
of
120 thethe
cloud cloud
computational time. This amount of control vol- 120
110
umes represented a first step of the simulation. Sim- 110
ulations in stages started in the 3rd control volume, 100
100
whereas control volume 8 was the last stage simula- 0
0
1
1
2
2
3 4
3Pressure
4 5
5
6
6
7
7
8
8
9
9
(kPa)
tion. Data for control volume 9 were placed in con- Pressure (kPa)
trol volume 3 of the next stage. There were a total g . 888 –
FFigure
iFigure Comparative
Comparative results
results for forprior
pressure pressure prior
to blast and to blast and
overpressure
Comparative results for pressure prior to blast and overpressure
of 7 stages to cover the 40 simulated control vol- overpressure
umes. The calculation time was 1500 s by simulat-
ing in a PC with 8 GB of RAM and processor In- 312
tel-i7 (R) 3.9 GHz.
The ASPEN package was used to calculate adi- 310

abatic reaction conditions. The first control volume

Cloud Temperature (K)

308
obtained a temperature of reaction of 460 K. This
value for temperature did not exceed the self-igni- 306
d

tion temperature for propane (766 K). For this rea-

u
clo
th nt

23
of e

304
son, the onset of the reaction was taken as the mo-
n nd
e

Zoneof
Zone ofgreater
greater
tio ce

ment of maximum pressure in the control volume to

ra es

concetration
Concentration of the
et f d

302
reproduce the worst potential effect. This maximum
nc o

ofcloud
the cloud
co one

pressure was achieved in the first volume of control

Z

300

when time equalled 0.0034 s. This time was pro- 298

duced in all simulations. 23
The results of the simulations included tem- 296 23
perature profiles and overpressure. Figure 6 shows 290 310 330 350 370 390 410 430 450 470 490 510
24
the temperature profile, which follows the trend of Explosion Temperature (K) 22 24

the propane fraction mass in the dispersion. This in- F i g . 9 – Comparative results for cloud temperature and 22 ex-
dicates a clear dependence of the temperature on Figure 9 Comparative results for cloud temperature and explosion temperature
plosion temperature
 C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. 93
Biochem. 93
93 93
Eng. Q., 30 (4) 419–427 (2016) 425
180 93 93 9
Ta b l e 2 – Evolution of overpressure in the first five control 93
12
12 93 190 190
volumes 170 93 93 0.034
0.034 s s 8
10
10 0.068 s s 170 170

Pressure prior to blast (kPa)

0.068
7 150
Evolution
on of overpressure
of overpressure
in the first
in the
Over- five
first
control
five control
volumesvolumes 160 88
0.102 s s
0.102 150

Overpressure (KPa)

Overpressure (kPa)
Evolution
on of overpressure
of overpressure
in the first
in thefive
first
control
five control
volumesvolumes

Overpressure (kPa)
Time 0.136 s s
0.136

(kPa)
Z

Pressure(kPa)
c
6 130
c

pressure Z 130
c

Evolution
onOverpressure
of overpressure
of overpressure
(s)in the first
in thefive
first
control
five
XYcontrol
volumesvolumesYZ
c

ZX 150 66 0.170.17
s s
me (s) Overpressure
(Pa) (Pa)
Z

Evolution
onOverpressure
of overpressure
of overpressure
in the first
in thefive
XYfirst
control
five
XY control
volumesvolumes
c c
ZX
c c
Y
(Pa)
c Z c
c c Y 0.204 s s 110 110
5
me (s) Overpressure
(Pa) (Pa)
Z Z
0.204
c c
X X c c

Evolution
on of overpressure
of overpressure
in the first
in the
five
XYfirst
control
five
XY control
volumesvolumes 44
Z c Z c
Y c
YZ Y c X X
YZ 140

Pressure
Y Y 0.238 s
meOverpressure
(s) Overpressure
(Pa) (Pa)
X Z X Z
c c c c
0.238 s 90
Y
YZ Y X Z Xc Z
YZ
c
c c
Y Y
4 90
meOverpressure
(s) Overpressure
(Pa) (Pa)
Z c Z c
XY XY 22
c c
X X Z c Z c
Y
YZ Y X X
YZ
c c
70
XY XY 130
meOverpressure
(s) Overpressure
(Pa) (Pa)
X Z X c Z c Y Y
70
c c

YZ X Z X Z
YZ
c c
Y Y
350
c c
Y Y
XY XY 0
03457969.457969.4
X X
0.0034 57969.4 Y
YZ Y X
Y
X
Y YZ
120 0 50
03457969.457969.4 -2 Overpressure (kPa) 230
03457969.457969.4 -2 30
03457969.457969.4 110 -4 Pressure (kPa) 110
034
57969.457969.4 100
-4
0 5 10 15 20 25
-10
0 0
10
-10 10 20 30
0 5 10
Distance 15
(m) 20 25 0 10 Dista
20
068
53717.78
53717.78 0 10 20 30 Distance
40 50
(m) 60 70 80 90
068
53717.78
53717.78
0.0068 53717.78 a) Distance (m)
068
53717.78
53717.78 a) b)
068
53717.78
53717.78 12 180
Figure a)
7 Profiles
190 of pressure prior to blast and overpressure b)
9
068
53717.78
53717.78 12 0.034 s
0.034
190
0.034 s
0.2380.034
s s
0.068 ss Figure17010 Profiles of pressure: a) oscillation s of pressure, b) peak of ove
10 170
170
0.4420.238 s 8
10

Pressure prior to blast (kPa)

0.068
0.102 ss
102
61265.2
61265.2 8 Figure
150
10 Profiles of pressure: a) oscillation of pressure,
7 b) peak of 0.68 0.442
s s
102
61265.2
61265.2

(kPa)
0.102
0.136 s 160 150
Pressure (kPa)

Overpressure (KPa)
130 0.8840.68
s s

(kPa)
102
61265.2
61265.20.0102 61265.2 6 0.136
0.17 s s
Pressure (kPa)

0.884 s 6
102
61265.2
61265.2 6 4 0.204s s
0.17 150
130
110

Overpressure
102
61265.2
61265.2 0.238 s
0.204 110 5

Overpressure
90
4 2 0.238 s 140
7090 4
136
62070.4
62070.4 2 0
136
62070.4
62070.4 130 5070
3
136
62070.4
62070.4 0 -2
136
62070.4
62070.40.0136 62070.4 120 3050
Overpressure (kPa) 2
136
62070.4
62070.4 -2 -4 110
1030
-1010 Pressure (kPa) 1
1754368.4
54368.4 -4 0 5 10 15 20 25 0 10 20 30 40 50 60 70 80
100 Distance (m) 0
1754368.4
54368.4 0 5
Distance (m)
10 15 20 25
-10
0 10 20 30 40 50 60 70 80
1754368.4
54368.4 Distance (m) 0 10 20 30 Distance
40 50(m) 60 70 80 90
1754368.4
54368.4 a) b) b) Distance (m)
1754368.4
54368.4 0.017 54368.4
a) 1 0 – Profiles of pressure: a) oscillation of pressure, b)
Figure 7 Profiles of pressure prior to blast and overpressure
Figure 10 Profiles of pressure: a) oscillationF of
i g .pressure, b) peak of overpressure
b) peak of overpressure
ome
on
ome Some
Properties
on Some Properties
Properties of
Properties Calcium-dependent
of Calcium-dependent
of Calcium-dependent
of Calcium-dependent
Figure 10 Profiles α-Amylase
α-Amylase
α-Amylase
of α-Amylase
pressure: a) from
from fromBacillus
from
oscillation of Bacillus
Bacillus
Bacillus
pressure, b) peak of overpressure
ome
on
ome
on Some
tilis Properties
subtilis
through
Some Properties
through
Properties of Calcium-dependent
Submerged of Calcium-dependent
Submerged Fermentation
Fermentationα-Amylase
of α-Amylase
Wheat
of from
Wheat
Bran Bacillus
from
Bran Bacillus
Bacillus
Bacillus
tilis
a, subtilisb Properties
through b ofbCalcium-dependent
through
Submerged ofbCalcium-dependent
Submerged Fermentation
Fermentationα-Amylase
of α-Amylase
c Wheat from
of Wheat
Bran from
Bran
ome
on Some
tilis Properties
a,*subtilis
Nadeem,
M. b Properties
throughthrough
Nadeem,
Q. ofbSyed,
Syed,
b Q. Calcium-dependent
Submerged
H. of b Calcium-dependent
Submerged
H.Fermentation
Abdullah c
Fermentation
Abdullah
Shakir,Shakir,
c and α-Amylase
of
J. α-Amylase
Wheat
c Iqbal
and of
J. from
c
Wheat
Qazi Bran Bacillus
from
c
c QaziBran
Iqbal Bacillus
tilis
a, subtilis
through
Nadeem,
* M. b through
Nadeem,
Q. Submerged
Syed,
Q.
This b Submerged
Syed,
H.
set Abdullah
H.
of pointsFermentation
Fermentation
Abdullah
Shakir,
correspond c to theof
Shakir,
and J. Wheat
ofPakistan.
Iqbal
and
carea J. Wheat
Qazi
with Bran
Iqbal
high-Qazi Bran
c
c Conclusions
tilis
a,*subtilis
ment
ment ofthrough
of
Nadeem,
M. through
Biotechnology,
Biotechnology,
Nadeem,
* M.
Submerged
b
Biotechnology,
Nadeem,
b Q.
Biotechnology,
Nadeem,
Submerged
University
Syed,
University
est
Q. Syed, b H.Fermentation
b
University
b Q. bSyed,
H. of Sargodha,
Abdullah
University
of
concentration
Q. Syed,
H. Abdullah
of
AbdullahFermentation
Sargodha,
Sargodha
Shakir,
H.Sargodha,
inof
the
Abdullah
Shakir,
c and J.
Sargodha,
Sargodha
cloud.
Shakir,
ofIqbal
Wheat
Sargodha
and
andSargodha
Shakir,
ofQazi
J.
Iqbal
Wheat
Pakistan. Bran
c
Iqbal
J.c Pakistan.
QaziBran
Iqbalc Qazic
Pakistan.
and J.Qazi
a, b b b b c
hnology
Biotechnology
ment
Biotechnology,
Nadeem,
* M.
hnology
Research
of Biotechnology,
Nadeem,
Biotechnology
Research
Q. Syed,
Research
Center,
University
Q.
Research Syed,
Center,H. Center,
Pakistan
University
of
Abdullah
H.
Center,
Pakistan
Pakistan
Sargodha,
of Council
AbdullahSargodha,
Shakir,
Pakistan
Council
Council
of
Sargodha
Shakir, J.c of
Scientific
of
Sargodha
andScientific
Council
of
Scientific
Pakistan.
Iqbal &
andScientific
J.Qazi
&Iqbal c
Industrial
Pakistan. Qazi
Industrial
c
& Industrial
& Industrial It seems that simplified models of explosion do
ment
Biotechnology,
of Biotechnology, University
The University
of Sargodha,
turbulence of
of Sargodha,
Sargodha
the system Sargodha
Pakistan.
generated Pakistan.
oscilla-
IR)
ment(PCSIR)
Laboratories
hnology
Biotechnology
Biotechnology,
of Laboratories
Research
Biotechnology, Complex,
Research
Center,
University Complex,Ferozpure
Center,
Pakistan
University
of Ferozpure
Pakistan
Sargodha,
of road
Council
Sargodha, Lahore,
road
Council
of
Sargodha Lahore,
Pakistan
Scientific
of
Sargodha Pakistan
Scientific
Pakistan. & 54600.
Industrial
Pakistan. & 54600.
Industrial
IR)(PCSIR)
Laboratories
hnology
Biotechnology Research tions
Laboratories
Complex,
Research in
Center, pressure
Complex,
Center,
Pakistan prior
Ferozpure to
Ferozpure
Pakistan
Councilthe
roadCouncilexplosion.
Lahore,
road
of Lahore, Figure
Pakistan
Scientific
of 10a
Pakistan
Scientific
& 54600.
Industrial 54600.consider the turbulence effect. The package
not
& Industrial
al
echnology
IR) Biotechnology
(PCSIR)
LaboratoriesLaboratory,
Laboratory,
Laboratories
Complex, Department
Complex, Department
Ferozpure ofroad
Zoology,
Ferozpure of Zoology,
Lahore,
road University
Lahore, University
Pakistan of
Pakistanthe
54600. Punjab
of the Punjab
newapproximates
new
hnology
Biotechnology
al
echnology
IR)
Research
Biotechnology
(PCSIR)
Laboratories
Research
Laboratory, Center,
represents
Laboratory,
Laboratories
Complex,
Center,
Pakistan
these
Department
Complex,
Pakistan
Department
Ferozpure
Council
oscillations
ofroad Council
Zoology,
Ferozpure of of
at Scientific
Zoology,
Lahore,
road
of
different Scientific
University
Lahore,
&
times
University
Pakistan
Industrial
of&
of54600.
Pakistan of54600.
Industrial
the Punjab FLACS
the
54600. Punjab
new new this effect under a first-order
Lahore
Pakistan
al
echnology
IR)(PCSIR) Pakistan
Biotechnology
LaboratoriesLaboratory,
Laboratory,
Laboratories
Complex,
simulation. Department
Complex, Department
Ferozpure
The of Zoology,
Ferozpure
oscillationsroad of Zoology,
Lahore,
road
increased University
Lahore, University
Pakistan
in response of
Pakistanthe
54600.
to reaction
Punjab
of the
54600. Punjab
new andnew applies a source of energy. The CFX
Lahore
Pakistan
al
echnology Pakistan
Biotechnology Laboratory,
Laboratory, Department Departmentof Zoology, of Zoology,University University
of the Punjab
of the
toolPunjab
new new
demonstrated the rigorous turbulence effect
galonding
author:
Lahore
Pakistan author:
irfan.biotechnologist@gmail.com
Pakistan irfan.biotechnologist@gmail.com
the generation of combustion , gases.
mirfanashraf@yahoo.com
,This
mirfanashraf@yahoo.com
is equiva-
gechnology
Biotechnology
onding
author:
Lahore
Pakistan
Laboratory,
author: Laboratory,Department
irfan.biotechnologist@gmail.com
Pakistan
Department
irfan.biotechnologist@gmail.com
lent to the production of waves
of Zoology, of, Zoology,
University University
mirfanashraf@yahoo.com
, of the Punjab
mirfanashraf@yahoo.com
in different spots of
of the Punjab
during
new
an
new
explosion. The applied strategy consid-
tonding
gLahore
author:Pakistan
Pakistan author:
irfan.biotechnologist@gmail.com
irfan.biotechnologist@gmail.com , mirfanashraf@yahoo.com
, mirfanashraf@yahoo.com
gtonding
author:author:
irfan.biotechnologist@gmail.com
irfan.biotechnologist@gmail.com , mirfanashraf@yahoo.com
, mirfanashraf@yahoo.com ered simulation of consecutive explosions, assum-
dy
eals
gtdy
ondingreveals
author: the standing of water.
the irfan.biotechnologist@gmail.com
productionproduction ofTable
α-amylase 2through
α-amylase summarizes,through the
Bacillus evolution
subtilisof
Bacillus subtilis
by submerged
by submerged
theauthor:
tealsreveals the irfan.biotechnologist@gmail.com
productionproduction of α -amylase
of α -amylase
through mirfanashraf@yahoo.com
through , mirfanashraf@yahoo.com
Bacillus Bacillus
subtilis subtilis
by submerged
by
ing an submerged
adiabatic reaction without a source of energy.
tion
tdywheat
eals of wheat
bran
reveals
thewheat athe
asbran
production
the overpressure.
substrate.
as
production a substrate.
of Various
α-amylase The
of α-amylase cutting
Various
conditions
through layouts
conditions
Bacillus
through inwere
wereBacillus Table
optimized
subtilis 2 and maximum
optimized
subtilis
by andsubmerged
submerged
by maximum
tion
dywheat of
ealsreveals bran asbran
the production a
the show substrate.
as
production a substrate.
of α-amylase
symmetry, Various Various
conditions
α-amylase
of indicating through conditions
were
Bacillus
through
a spherical optimized
were
Bacillus
subtilis
pressure optimized
pro- and
subtilis maximum
and
by submerged
by
o maximum
submerged
o
on
dyobserved
tion
eals was
wheat theobserved
of wheat
bran
reveals at as
initial
bran at
afile.
production initial
medium
substrate.
as medium
a substrate. pH
Various
α-amylase of pH
7.0,
Various ofincubation
conditions
theαoverpressure
7.0, incubation
conditions temperature
weretemperature
atBacillus optimized
were
Bacillus temperature
subtilis of 35
optimized
and
subtilis oof
C47035
maximum
and with oC with
maximum
on observed
tion was
wheat wheat
bran atthe
of observed asbran
production
initial
a at
of
initial
medium
Peaks
substrate.
as in
a substrate.
of
medium
pH
Various
-amylase
of pH through
7.0,
Various
conditions
through
ofincubation
7.0, incubation
different
conditions
were optimized times
were of by
temperature
of 35
optimized
and
submerged
by
of
C
maximum
and
o
submerged
35
with oC with
maximum
onof
speed
tion 140
wheat ofof
observed
was rpm140
wheat
bran for
observed
atfor
asrpm48
initial
bran
a48at for
hasof48
initial
medium
substrate.
simulation fermentation
hmedium
of pH
a fermentation
substrate. fermentation
Various
existed ofinpH period.
7.0,
Various period.
Peptone,
ofincubation
conditions
different 7.0, Peptone,
incubation
conditions
were
areas ammonium
temperature
of ammonium
weretemperature
optimized
explosion, sulphate
ofsulphate
optimized
and 35 oofsulphate
C450
maximum
and 35and
with and
C with
maximum 0.034 s
of
speed140
onobserved of
rpm140
was observed rpm
at initial for
h of
at initial48
medium h of
mediumfermentation
pH of pH period. period.
7.0,ofincubation Peptone,
7.0, incubation Peptone,
ammonium
temperature ammonium
temperature
of 35 oof withooC with
sulphate
470
C 35
180 and and 9
favored
ontarch
of
speed140
observed
was favored
ofenzyme
rpm140 for
observed
at enzyme
rpm production
48
initial
at for
Figureh of 48
mediumproduction
initial
10b. as
fermentation
h of
medium
pH nitrogen
as
fermentation
of pH nitrogen
period.
7.0, ofand carbon
period.and
Peptone,
incubation
7.0, carbon
source,
Peptone,
incubation ammonium
temperature source,
respectively.
ammonium
temperature
of respectively.
sulphate
35 of The
sulphate
35and
C430with The
and
C with
0.068
0.034 ss
h The explosion temperature presented an sulphate 450
favored
tarch
of
speed140favored
ofenzyme
rpm enzyme
140forrpm production
48 for
h of48 production
ofasfermentation
fermentation nitrogen
asperiod.
nitrogen
andperiod.
carbon
and Peptone,
Peptone, carbon
source,
ammonium source,
respectively.
ammonium respectively.
The
sulphate
170 and The
and 0.102
0.068 ss 8
amylase
eof was
favored purified
was purified
2.03-fold 2.03-fold
by columnby column
(sephadex (sephadex
G-100) G-100)
chromatography
chromatography having having
etarch140favored
speed ofenzyme
rpm140for enzyme
rpm production
48 for
h of
oscillatory 48productionasfermentation
fermentation
hbehavior,
of nitrogen
asFigure
nitrogen
period.and carbon
period.
11. and Peptone,
Peptone,
This carbon
source,
ammonium
behavior source,
respectively.
ammonium
is respectively.
a sulphate The
sulphate
andhaving The
and
430
410 0.136
0.102 ss
Pressure prior to blast (kPa)

amylase
was
favored
tarch purified
was
favored
enzyme purified
2.03-fold
enzyme 2.03-fold
production
–1 by column
production
–1 as by column
nitrogen
as(sephadex
nitrogen
and (sephadex
G-100)
carbon
and G-100)
carbonchromatography
source, chromatography
source,
respectively. having
respectively.
The The 0.17
0.136s s 7
yfavored
activity
of 480.0 of U
480.0
mg U. The
result mg of–1purified
.byThe
the purified
enzyme
combustion enzyme
had
gases molecular
had carbon
movement molecular
mass
andsource,
theof
mass
54.18 of 160
kDa
54.18 askDa as
yamylase
etarch
was purified
was
favored
enzyme purified
2.03-fold
enzyme 2.03-fold
production column
production
as by column
as(sephadex
nitrogen nitrogen
and (sephadex
G-100)
carbon
and G-100)
chromatography
source, chromatography
respectively. having
respectively. having
410The
390 The
(KPa)
(K)

–1
activity
e of
amylase
was 480.0 of U
480.0
purified
was mg U
purified .
2.03-foldmg
The
increase
–1 purified
.
2.03-fold
–1 inbyThe purified
column
the byenzyme
energy column enzyme
(sephadex
releasedhad molecular
(sephadex
by had
G-100)
the molecular
G-100)mass
chromatography
combustion. of
mass
54.18 of
chromatography kDa
54.18
having as kDa
having as 0.204
0.17 s s 6
ed
12
yamylase
wasby
%480.0
activity
e of SDS-PAGE.
12of %U
purified
was SDS-PAGE.
480.0
mg U. The
purified
2.03-foldmg optimum
.byThe
–1purified
2.03-fold optimum
pH
purified
columnbyenzyme
columnand pH
enzyme
(sephadexhadtemperature
and temperature
molecular
(sephadex had
G-100) ofmass
molecular
G-100) the chromatography
of
enzyme
chromatographyof the
mass
54.18 enzyme
ofwas390 8.5
kDa
54.18
having
370 was
as kDa 8.5
having as
(K)

150
Temperature

–1 0.204 s
ed
12 ofby
yactivity%480.0
SDS-PAGE.
12of %U SDS-PAGE.
480.0
mg U . mg
The optimum
purified
. The optimum
pH
purified
enzyme and pH
enzyme
hadtemperature
and temperature
molecular
had of
molecular the
mass of
enzyme
of the
mass
54.18 enzyme
ofwas
kDa
54.18 8.5was
as kDa 8.5
as 5
Overpressure

2+ –1 2+A comparison between the overpressure curves

yC,
ectively.
ed
12 respectively.
by
%
activity
of 480.0 Ca
SDS-PAGE.
12 of %U
480.0
mg played
Ca
SDS-PAGE.
U . mg
2+The an–1purified
played .important
optimum
Thean purified
important
optimum
pHroleand
enzyme inpH
role
enzymethe
had andstability
inhad
temperaturethe stability
of
temperature
molecular of
molecularthe
the
mass enzyme
of
of the
enzyme
of the
mass enzyme
54.18 retaining
370
enzyme
ofwas350
kDa
54.18 retaining
8.5was
as kDa 8.5
as
Temperature

2+ 140
C,
ectively.
ed
12 respectively.
by 12 Ca
% SDS-PAGE. played
Ca
%2+SDS-PAGE. played
2+The
obtained an optimum
important
withTheanCFXimportant
optimum
pHrolecalculated
and andinpH
role
the stability
inwith
otemperature
and the stability
of
of the
otemperature isthe enzyme
of
of the
theenzyme
enzyme retaining
enzyme
was350 8.5retaining
was 8.5
C, TNO nec- 2+ 2+ 4
yactivity
of 87SDS-PAGE.
of
% %87
and % 67and %played
67
atoptimum
pH
%an at
9.5 pH and 9.5 80 and 80
C,and revealing revealing
ofthe Ca thethedependency
Ca 130dependency
330
yC,
ectively.
12
edof
C,
respectively.
by
%87
activity
ectively. %Ca
12of
Ca87
respectively. 2+
played
Ca
SDS-PAGE.
2+
and % 67
played
Ca 2+The
and
essary
2+% an67
at
played
an
here,
important
The
pH
% atimportant
optimum
9.5
important
an
Figure pH pHrole
and
important
12.
and
9.5
role
inand
80
in
role
pH the
role
theC,
Unfortunately,
o 80stability
in
otemperaturethe
revealing
C,
stability
in the
o
stability
of
otemperature the
revealingthe
the
stability
the of the
literature
enzyme
of
Caof the
enzyme
2+
the
enzyme
of the
2+
enzyme
retaining
enzyme
was
2+
dependency
Ca
enzyme
retaining
8.5 was 8.5
dependency
retaining
2+ 330
310 retaining 3
yhe
ely.
C, enzyme
activity
of
ectively.The
87 enzyme
of
%
respectively.
Ca was
87
and %highly
67was
played
Ca and% highly
stable
67
at
played
an pH
% towards
stable
at
9.5
important
an pH andtowards
important surfactants
9.5
role 80
inandoC,
role
the surfactants
80 oand
revealing
C,
stability
in the oxidizing
and
revealing
the
stability
of the oxidizing
Ca agents.
the
enzyme
of the
2+ agents.
Kinetic
dependency
Ca
enzyme Kinetic
dependency
retaining
2+ retaining
yhe
ely.
ofenzyme
activity The
87 of%enzyme
was
87
and% highly
67was
doesand%not highly
stable
67 % towards
atdisclose
pH stable
at
9.5pHa andtowards
surfactants
9.5
detailed 80and oC, surfactants
and
80revealing
description C, ofoxidizing
and
revealing
thethe oxidizing
–1TNO agents.
Cathe Caagents.
Kinetic
–1 dependency
120
–1Kinetic
dependency
310
290 –1
Overpressure (kPa) 2
dicated
yhethat
ely. thethat
ofenzyme
activity The
87 enzyme
the
enzyme
of
% was
87
and% enzyme
exhibit
highly
67was
and highly
stable
%31,3267 K
atexhibit
m and
. pH
% KV
towards
stable
at
9.5 and Vand
value
towards
surfactants of 4.4oand
value
surfactants of
mgrevealing
4.4
mL mg
oxidizing
and and
mL 2+
714.2
oxidizing
Caagents. and 2+
U
714.2
mL–1
Caagents.
Kinetic U mL
Kinetic
dicated
hethat
ely. thethat
enzyme
The enzyme
the
enzyme
was values
enzyme
exhibit
highly
was highlystableK The
exhibit
m and KpH
V
towards
stable mand
curve max
and
mmax
9.5
of 80
value
towards maxC,
Vthe
max CFX
surfactants of80 revealing
value
surfactants
C,
simulation
4.4and of
mg4.4 mL the
mg
oxidizing
and
–1 data
–1 and
mL the
–1 dependency
714.2
oxidizing and
agents.
–1 U dependency
714.2
mL
agents.
Kinetic
110
290
270 U mL
Kinetic
–1
Pressure (kPa) 1
uble
arch
dicated
hethat
ely.
uble
arch
starch
as
enzyme
The aenzyme
thethat
starch
as a
substrate
as
theaaenzyme
enzyme
was
substrate
as
substrate
highly
was
seems respectively.
exhibit
substrate highly
stable
to Krespectively.
exhibit
m and
follow
respectively. K V
towards
stable and
between
respectively.mmax Vlevels
value
towards max of
surfactants value
4.4
surfactants
5 and
and of
mg7.4.4
mL mg
oxidizing
and
This and
mLagents.
714.2
oxidizing
–1indi- –1 mL0 –1
andagents.
U
714.2
Kinetic U
–1 mL–1
Kinetic –1
dicated
that thethat enzyme
the enzyme
exhibitexhibit Km andKV mmax Vmax of
andvalue value
4.4 of mg4.4 mLmg –1 and
mL–1 714.2
and U 714.2
mL–1U mL–1
270
100 5 10 15 20 25 0
ds:
amylase,
uble
arch
dicated
thatα αstarch
-amylase,
as
the purification,
aenzyme
thatsubstrate
as purification,
theaenzyme
substrate characterization,
respectively.
exhibit characterization,
Krespectively.
exhibit andKisVmmax Bacillus
Vmax of
andvalue Bacillus
valuesp.effect
4.4 sp.
sp.mg of 4.4mL mg and
mL714.2 and U 714.2
mL 0 U mL 5
cates that there no congestion between Distance
10 (m)15 20 25
ds:
amylase,
uble
arch -amylase,
as a purification,
starch as a purification,
substrate substrate characterization,
respectively. m characterization,
respectively. Bacillus Bacillus sp. 0 10 20 30 40 50 60 70 80 90
ction
ds:
amylase,
arch
uble α -amylase,
as
starch
a substrate
as a these
purification, levels,
purification,
substrate which agrees
characterization,
respectively. characterization,
respectively. Bacillus
withBacillus
our
sp. numerical
sp. ex- Distance
Distance (m)
(m)
ction
ds: α-amylase,
amylase, purification,
purification,
periment. characterization,
characterization, Bacillus Bacillus
sp. sp. Figure 11 Temperature oscillations in different times 26
ds: α-amylase,
ction
amylase, purification,
purification,
characterization,
characterization, Bacillus Bacillus
sp. sp. F i g . 1 1 – Temperature oscillations in different times
ction Figure
Figure 711 Temperature
Profiles oscillations
of pressure in different
prior to blast times
and overpressure
ction
26
 426 C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016)

0.034 s
180
190 0.034 s 9 8. Vasanth, S., Tauseef, S. M., Abbasi, T., Abbasi, S. A., As-
0.068 s 170 0.238 s
8 sessment of four turbulence models in simulation of large-
170 0.442 s
scale pool fires in the presence of wind using computation-

Pressure prior to blast (kPa)

0.102 s 150 0.68 s 7
(kPa)

0.136 s 160 al fluid dynamics (CFD), Journal of Loss Prevention in the

Overpressure(KPa)

130 0.884 s
0.17 s
0.204 s 150
110
6 Process Industries 26 (2013).
5 doi: https://doi.org/10.1016/j.jlp.2013.04.001
Overpressure

0.238 s 90
140 9. Klippel, A., Schmidt, M., Muecke, O., Krause, U., Dust con-
70 4
130
50
centration measurements during filling of a silo and CFD
3 modeling of filling processes regarding exceeding the low-
30
120
Overpressure (kPa) 2 er explosion limit, Journal of Loss Prevention in the Pro-
10
110 Pressure (kPa) 1 cess Industries 29 (2014) 122.
-10
20 25 0 10 20 30 40 50 60 70 80 doi: https://doi.org/10.1016/j.jlp.2014.02.006
100 Distance (m) 0
10. Di Sarli, V., Russo, P., Sanchirico, R., Di Benedetto, A.,
0 10 20 30 40 50 60 70 80 90 CFD simulations of dust dispersion in the 20L vessel: Ef-
b) Distance (m)
fect of nominal dust concentration, Journal of Loss Preven-
F i g . 1 2 – Comparison of CFX overpressure with results of tion in the Process Industries 27 (2014) 8.
Figure 7 Profiles of pressure prior to blast and overpressure
TNOof#5overpressure
a) oscillation of pressure, b) peak and #7 doi: https://doi.org/10.1016/j.jlp.2013.10.015
11. Murillo, C., Dufaud, O., Bardin-Monnier, N., López, O.,
This strategy has described the relationship between Munoz, F., Perrin, L., Dust explosions: CFD modeling as a
tool to characterize the relevant parameters of the dust dis-
the intensity of the explosion and the amount of persion, Chemical Engineering Science 104 (2013) 103.
mass converted. The predicted behavior indicated doi: https://doi.org/10.1016/j.ces.2013.07.029
by the TNO method fits very well with our ap- 12. Ren, T., Wang, Z., Cooper, G., CFD modelling of ventila-
proach where no congestion areas are considered, tion and dust flow behaviour above an underground bin and
levels 5 and 6 of TNO, as indicated in that method. the design of an innovative dust mitigation system, Tunnel-
This methodology could be extended to describe ling and Underground Space Technology 41 (2014) 241.
explosions in systems with different levels of con- doi: https://doi.org/10.1016/j.tust.2014.01.002
gestion. 13. Fothergill, C. E., Chynoweth, S., Roberts, P., Packwood, A.,
Evaluation of a CFD porous model for calculating ventila-
tion in explosion hazard assessments, Journal of Loss Pre-
Acknowledgements vention in the Process Industries 16 (2003) 341.
doi: https://doi.org/10.1016/S0950-4230(02)00113-4
The authors thank CONACyT for their support 14. Rigas, F., Sklavounos, S., Experimentally validated 3-D
in this work. simulation of shock waves generated by dense explosives
in confined complex geometries, Journal of Hazardous Ma-
terials 121 (2005) 23.
doi: https://doi.org/10.1016/j.jhazmat.2005.01.031
References
15. Rigas, F., Sklavounos, S., Simulation of Coyote series tri-
1. Marsh-McLennan, The 100 largest losses (2012) 1972– als—Part II: A computational approach to ignition and
2011. combustion of flammable vapor clouds, Chemical Engi-
2. Cozzani, V., Salzano, E., The quantitative assessment of neering Science 61 (2006) 1444.
domino effects caused by overpressure: Part I. Probit mod- doi: https://doi.org/10.1016/j.ces.2005.09.005
23
els, Journal of Hazardous Materials 107 (2004) 67. 16. Ferrara, G., Di Benedetto, A., Salzano, E., Russo, G., CFD
doi: https://doi.org/10.1016/j.jhazmat.2003.09.013 analysis of gas explosions vented through relief pipes, Jour-
3. Crowl, D., Calculating the energy of explosion using ther- nal of Hazardous Materials 137 (2006) 654.
modynamic availability, Journal of Loss Prevention in Pro- doi: https://doi.org/10.1016/j.jhazmat.2006.03.037
cess Industry 5 (1992) 109. 17. Middha, P., Hansen, O. R., Grune, J., Kotchourko, A., CFD
doi: https://doi.org/10.1016/0950-4230(92)80007-U calculations of gas leak dispersion and subsequent gas ex-
4. Held, M., TNT-equivalent, Propellants, Explosives, Pyro- plosions: Validation against ignited impinging hydrogen jet
technics 8 (1983) 158. 26 experiments, Journal of Hazardous Materials 179 (2010)
doi: https://doi.org/10.1002/prep.19830080507 84.
5. van den Berg, A. C., The multi-energy method: A frame- doi: https://doi.org/10.1016/j.jhazmat.2010.02.061
work for vapour cloud explosion blast prediction, Journal 18. Qiao, A., Zhang, S., Advanced CFD modeling on vapor dis-
of Hazardous Materials 12 (1985) 1. persion and vapor cloud explosion, Journal of Loss Preven-
doi: https://doi.org/10.1016/0304-3894(85)80022-4 tion in the Process Industries 23 (2010) 843.
6. Baker, Q. A., Doolittle, C. M., Fitzgerald, G. A., Tang, M. doi: https://doi.org/10.1016/j.jlp.2010.06.006
J., Recent developments in the Baker-Strehlow VCE Anal- 19. Tauseef, S. M., Rashtchian, D., Abbasi, T., Abbasi, S. A., A
ysis methodology, Process Safety Progress 17 (1998) 297. method for simulation of vapour cloud explosions based on
doi: https://doi.org/10.1002/prs.680170411 computational fluid dynamics (CFD), Journal of Loss Pre-
7. Cormier, B. R., Qi, R., Yun, G., Zhang, Y., Mannan, M. S., vention in the Process Industries 24 (2011) 638.
Application of computational fluid dynamics for LNG va- doi: https://doi.org/10.1016/j.jlp.2011.05.007
por dispersion modeling: A study of key parameters, Jour- 20. Dadashzadeh, M., Khan, F., Hawboldt, K., Amyotte, P., An
nal of Loss Prevention in the Process Industries 22 (2009) integrated approach for fire and explosion consequence
332. modelling, Fire Safety Journal 61 (2013) 324.
doi: https://doi.org/10.1016/j.jlp.2008.12.004 doi: https://doi.org/10.1016/j.firesaf.2013.09.015
C. Díaz-Ovalle et al., A CFD-based Approach to Predict Explosion Overpressure, Chem. Biochem. Eng. Q., 30 (4) 419–427 (2016) 427
21. Hansen, O. R., Hinze, P., Engel, D., Davis, S., Using com-
putational fluid dynamics (CFD) for blast wave predictions,
Journal of Loss Prevention in the Process Industries 23
(2010) 885.
doi: https://doi.org/10.1016/j.jlp.2010.07.005
22. Hansen, O. R., Gavelli, F., Ichard, M., Davis, S. G., Valida-
tion of FLACS against experimental data sets from the
model evaluation database for LNG vapor dispersion, Jour-
nal of Loss Prevention in the Process Industries 23 (2010)
857.
doi: https://doi.org/10.1016/j.jlp.2010.08.005
23. Gavelli, F., Davis, S. G., Hansen, O. R., Evaluating the po-
tential for overpressures from the ignition of an LNG vapor
cloud during offloading, Journal of Loss Prevention in the
Process Industries 24 (2011) 908.
doi: https://doi.org/10.1016/j.jlp.2011.07.002
24. Bakke, J. R., van Wingerden, K., Hoorelbeke, P., Brewerton,
B., A study on the effect of trees on gas explosions, Journal
of Loss Prevention in the Process Industries 23 (2010) 878.
doi: https://doi.org/10.1016/j.jlp.2010.08.007
25. Hjertager, B. H., Computer simulation of turbulent reactive
gas dynamics, Modeling, Identification and Control 5
(1985) 211.
doi: https://doi.org/10.4173/mic.1984.4.3
26. Popat, N. R., Catlin, C. A., Arntzen, B. J., Lindstedt, R. P.,
Hjertager, B. H., Solberg, T., Saeter, O., Van den Berg, A.
C., Investigations to improve and assess the accuracy of
computational fluid dynamic based explosion models, Jour-
nal of Hazardous Materials 45 (1996) 1.
doi: https://doi.org/10.1016/0304-3894(95)00042-9
27. CCPS, Guidelines for evaluating characteristics of vapor
cloud explosions, flash fires, and BLEVE’s, Center for
chemical process safety of the AIChE, New York, (1994).
28. CPR, Methods for the calculation of physical effects due to
releases of hazardous materials (liquids and gases) “Yellow
book”, CPR 14E, 3rd ed., Ministerie van Verkeer en Water-
staat, Hague, Netherlands, (2005).
29. Wilcox, D. C., Turbulence Modeling for CFD, DCW Indus-
tries Inc., (2006).
30. Launder, B. E., Sharma, B. I., Application of the Ener-
gy-Dissipation model of turbulence to the calculation of
flow near a spinning disc, Letters in Heat and Mass Trans-
fer 1 (1974) 131.
doi: https://doi.org/10.1016/0094-4548(74)90150-7
31. van den Berg, A. C., Lannoy, A., Methods for vapour cloud
explosion blast modelling, Journal of Hazardous Materials
34 (1993) 151.
doi: https://doi.org/10.1016/0304-3894(93)85003-W
32. Mercx, W. P. M., van den Berg, A. C., Hayhurst, C. J., Rob-
ertson, N. J., Moran, K. C., Developments in vapour cloud
explosion blast modeling, Journal of Hazardous Materials
71 (2000) 301.
doi: https://doi.org/10.1016/S0304-3894(99)00085-0