RonNielsen NuclearReactions

.
Nuclear Reaction
Mechanism and Spectroscopy

Ron W Nielsen
(aka Jan Nurzynski)
2011
Environmental Futures Centre, Griffith School of Environmental Science,

Griffith University, Cold Coast Campus, Qld, 4222, Australia
r.nielsen@griffith.edu.au
Ron W Nielsen, Nuclear Reactions: Mechanism and Spectroscopy Page 1
Contents
Preface 4
12 13
1. Neutron Polarization in the C(d,n) N Stripping Reaction 7
2. A Systematic Discontinuity in the Diffraction Structure 16
27
3. Core Excitations in Al 30
40
4. Ca(d,d), (d,d’), and (d,p) Reactions with 12.8 MeV Deuterons 50
5. Spectroscopic Applicability of the (3He,α) Reactions 61
6. The Discrete Radius Ambiguity of the Optical Model Potential 73
27 3 26 3
7. The Al( He,α) Al Reaction at 10 MeV He Energy 79
28
8. Configuration Mixing in the Ground State of Si 85
9. The j - dependence for the 54Cr(d,p)55Cr Reaction 91
10. Tensor Analyzing Powers for Mg and Si Nuclei 99
11. Optical Model Potential for Tritons 115
12. The 54Cr(d,t)53Cr and 67,68Zn(d,t)66,67Zn Reactions at 12 MeV 128
76,78 74,76
13. A Study of the Se(p,t) Se Reactions at Ep = 33 MeV 138
76,78,80,82
14. Single-neutron Transfer Reactions on Se Isotopes
Induced by 33 MeV Protons 147
15. Analysis of Polarization Experiments 159
16. Reorientation Effects in Deuteron Polarization 172
17. Two-step Reaction Mechanism in Deuteron Polarization 189
18. Tensor Analyzing Power T20(00) for the 3He(d,p)4He
Reaction at Deuteron Energies of 0.3 – 36 MeV 201
19. The Maximum Tensor Analyzing Power Ayy = 1 for the
3
He(d,p)4He Reaction 208
20. Search for the Ay = 1 and Ayy = 1 Points in the 6Li(d,α)4He
Reaction 215
21. A Study of a Highly Excited Six-Nucleon System with
Polarized Deuterons 222
22. A Study of the 5.65 MeV 1+ Resonance in 6Li 239
23. A Study of Spin-orbit and Tensor Interaction of Polarized
Deuterons with 60Ni and 90Zr Nuclei 252
24. Global Analysis of Deuteron-nucleus Interaction 262
25. Collective Excitation Effects in the Elastic Scattering 272
97,101
26. A Study of Ru Nuclei 281
27. Spectroscopy of the 53,55,57Mn Isotopes and the Mechanism
of the (4He,p) Reaction 290
28. Gamma De-excitation of 55Mn Following the 55Mn(p,p’γ)55Mn
Reaction 311
29. The 138Ba(7Li,6He)139La and 140Ce(7Li,6He)141Pr Reactions at
52 MeV 319
30. Nuclear Molecular Excitations 329
7 28 40
31. The Interaction of Li with Si and Ca Nuclei 342
24
32. Triaxial Structures in Mg 351
33. Spin Assignments for the 143Pm and 145Eu Isotopes 368
16 24
34. Search for Structures in the O+ Mg Interaction 377
35. Parity-dependent Interaction 384
Appendices
A. Semi-classical descriptions of polarization in stripping
reactions 394
B. The diffraction theory 399
C. The plane-wave theory of inelastic scattering 405
D. The strong coupling theory 409
E. Theories of direct nuclear reactions 413
F. Nuclear spin formalism 426
G. Polarized ion sources 430
H. Selected nuclear spin structures and the extreme values of
the analyzing powers 438
I. Analytic determination of the extreme Ayy points 445
J. Phase shift analysis 447
K. Reorientation effect in Coulomb excitation 451
L. The Faddeev formalism 458
M. Resonating group theory 464
N. The R-matrix theory 466
Preface
This document presents highlights of my research work in nuclear physics carried
out for over 30 years. It includes my pioneering research work in Poland and the
introduction of a study of direct nuclear reactions in Australia. Methods used in the
analysis and interpretations of these reactions were later applied in the study of
heavy-ion induced reactions.
Research objectives
The main objectives of my research were to study nuclear spectroscopy and the
mechanism of nuclear reactions using
• Direct nuclear reactions
• Polarization phenomena in nuclear reactions
• Heavy-ion induced reactions
Research institutes
I have worked in the following research institutes:
• Institute of Nuclear Physics, Cracow, Poland
• Department of Nuclear Physics, Institute of Advanced Studies, Australian
National University, Canberra, ACT, Australia
• Laboratorium für Kernphysik, ETHZ, Zürich, Switzerland
• Schweizerische Institut für Nuklearforschung, Villigen, Switzerland
• Max-Plank Institut für Kernphysik, Heidelberg, Germany
• Institut für Angewandte Kernphysik and Zyklotron-Laboratorium
Kernforschungszentrum Karlsruhe, Karlsruhe, Germany
Accelerators
The research was supported by the following particle accelerators:
• U-120 cyclotron in Poland
• EN tandem electrostatic accelerators in Canberra and Zurich
• 14 UD Pelletron accelerator in Canberra
• Cyclograph in Canberra
• Isochronous cyclotron in Karlsruhe
• Injector cyclotron at Schweizerische Institut für Nuklearforschung in Villigen
Polarized ion sources
The polarized ion sources used in the study of polarization fenomena included
• The atomic beam polarized ion sources
• Lamb-shift polarized ion sources
Data acquisition systems
I have used a broad range of particle detection systems. They were
• Nuclear emulsions
• Proportional counters
• Scintillation counters
• Solid state detectors
• Resistive-wire gas proportional detector
• Detector telescopes
• Magnetic spectrometers
Other data acquisition and processing systems included:
• Hutchinson-Scarrott pulse hight analyser
• 400 channel RIDL analyser
• 512 channel RCL analyser
• PDP computers (PDP-8 and PDP-11)
• IBM 1620 computer
• IBM 1800 computer
• Hewlett-Packard (HP2100A) computers
• VAX online computers (VAX750, VAX1000, VAX2000, VAX3100, VAX3200,
and VAX 4000)
Theoretical frameworks
In addition to the experimental work, I have also carried out theoretical analysis of
my data. This work was supported by internationally shared computer codes, which I
have adapted to available computers and modified whenever necessary as well as
by many computer programs.
Theoretical frameworks used in my research included:
• Optical model
• Diffraction theory
• Plane wave theory
• Distorted wave theory
• Coupled channels formalism
• Phase-shift analysis
• Resonating group theory
• Faddeev formalism
• R-matrix theory
Mainframe computers
The mainframe computers used to support the evaluation and theoretical analysis of
experimental results included:
• IBM 360/50
• UNIVAC-1108, 1100/42, 1100/82
• VAX-780
The following summary is arranged in approximately chronological order. It starts

with my pioneering research in Poland and ends with experiments in the field of
heavy-ion-induced reactions.
1
Neutron Polarization in the 12C(d,n)13N Stripping Reaction
Key features:
1. A challenging and complex experiment involving a detection of low-yield neutrons and
the measurements of neutron polarization without using a polarized ion source but
rather by using the “double scattering” method. (Polarized ion source is a complex
apparatus and was not available for this experiment.)
2. The first experimental work in the field of nuclear reactions in Poland.
3. The first ever published results on the neutron polarization at medium deuteron
energies.1
4. The first attempt to compare experimental results at these energies with the early
theoretical predictions.
5. Large polarization detected in our experiment suggested that this reaction could serve
as a source of polarized fast neutrons.
Abstract: The polarization of neutrons from the stripping reaction 12C(d,n)13N induced by 12.9
MeV deuterons was measured using a “double scattering” method. Measurements were
carried out in the angular range of 150 – 600 (lab). Results of measurements are compared
with the early theoretical predictions. They demonstrate the violation of the theoretical sign
rule and thus challenge the early interpretations of the mechanism of the nucleon polarization.
In the beginning…
Research in nuclear physics in Poland commenced in 1955 in the School of Physics of
the Jagiellonian University, in Cracow. Established in 1364, it is one of the oldest
universities in Europe. Its well-known alumni include Nicolaus Copernicus and Pope
John Paul II.
The University is located in the centre of Crocow, and thus in the old town area. Most of
the walls of the original Cracow were destroyed and were replaced by a public park,
called Planty. The School of Physics is located next to the park but it is also close to the
original Collgium Maius and next to the new Collegium Novum. My office was on the
first floor of the School of Physics, facing a courtyard. In the middle of the office,
incongruously for such a historical surrounding, was an ion source my colleague and I
were constructing. A few other groups and individuals were also carrying out their work
in various rooms and the basement of this old building. Meanwhile, a new Institute of
Nuclear Physics was being constructed on the outskirts of Cracow and eventually we
have all moved to new offices. It was an unbelievable luxury.
The research work at that time was carried out in three major fields: Nuclear Physics;
Physics of Structure of Solids, Liquids and Gases; and Applied Nuclear Physics. By
1965, the 10th anniversary of the commencement of nuclear research in Poland, the
Institute established close links with various research centres around the world. The
invitation I received from the Australian National University, Canberra, Australia, to
1
Our results (at Ed = 12.9 MeV) were published in 1959 (Budzanowski et al. 1959). Two years later,
Haeberli (1961) published results of measurements for low energy deuterons (Ed < 3.6 MeV).
However, the same results were mentioned earlier in a form of a short abstract at a meeting of the
American Physical Society (Haeberli and Rolland 1957).
work in the Department of Nuclear Physics, was the first link with this remote
continent.
The first director of the Institute was Professor Henryk Niewodniczanski,2 who was
amiably called Papa by those who knew him well. After his premature death of heart
attack, the name of the Institute was changed to The Henryk Niewodniczanski Institute
of Nuclear Physics.
Figure 1.1. The map of the Institute of Nuclear Physics in 1965, on the 10th anniversary of the
commencement of nuclear research in Poland. 1 – The location of the U -120 cyclotron where the first
experimental work in nuclear physics in Poland was done and where the first in the world
measurements of neutron polarization at medium deuteron energy were successfully carried out; 2 –
The main building containing other laboratories, library, lecture theatre, workshops and administration
offices; 3 – The Low Temperature Laboratory; 4 – A fountain made of the water from the cyclotron
cooling system; 5 – A building for storing radioactive materials. Shaded areas mark future (at that
time) developments: 6 – A Computer Centre and Library; 7 – Extension for a new target hall of the U -
120 cyclotron; 8 – Proposed heavy ion cyclotron; 9 Workshops. (Reproduced from IFJ 1965.)
The major research facility in the Institute was the U-120 cyclotron, which was
purchased from Russia. However, two experimental halls belonging to the cyclotron
were empty and the necessary equipment had to be either constructed or
purchased.
In 1959 we have successfully completed the first ever experiment in nuclear reactions
in Poland. This was also the first experimental work in the world on the neutron
polarization in deuteron-induced stripping reactions at medium deuteron energies.
Research in nuclear physics was at that time still in its early stages everywhere. Only
a few years earlier, it was discovered that nuclear reactions do not proceed
necessarily via a compound nucleus but can also involve direct transitions. The new
concept of direct nuclear reactions has thus been introduced and we were interested
in studying their mechanism and their application in nuclear spectroscopy. The
relation between the compound nucleus and direct nuclear reaction mechanisms is
illustrated schematically in Figure 1.4.
2
Called after his death “the Rutherford of Poland” for introducing nuclear research in Poland
(Hodgson, P. E. http://www.zwoje-scrolls.com/zwoje36/text04.htm).
Figure 1.2. International connections of the Institute of Nuclear Physics in Cracow in 1965 in the form
of exchange of scientists for longer periods. The link with Australia marks my presence in the
Department of Nuclear Physics of The Australian National University in Canberra. (Reproduced from
IFJ 1965.)
Figure 1.3. Left-hand side: The aerial view of The Henryk Niewodniczanski Institute of Nuclear
Physics, in 2006 (EPPOG 2006). Right-hand side: The view of the main building in 2006. In the
foreground is the pool for the cyclotron cooling system and behind it the lecture theatre located in the
main building (building 2 in Figure 1.1). (Photo credit: P. Zielinski, IFJ 2005.)
Figure 1.4. Schematic illustration showing two types of reactions, direct and compound nucleus
(modified from Hodgson 1971). Reactions can proceed directly, via compound nucleus, or both.
The first experimental evidence for the direct nuclear reactions was provided by
Burrows, Gibson and Rotblat (1950) and by Holt and Young (1950). These authors
observed forward peaking in reactions induced by 8 MeV deuterons, which could not
be explained by the compound nucleus mechanism.
The first theories describing the direct reaction mechanism were proposed by Butler
(1950, 1951), Hubby (1950), Hubby and Newns (1951), and Bhatia et al. (1952). The
first theoretical descriptions of direct nuclear reactions were based on a simple
plane-wave approximation. Later, distorted wave approximation and coupled
channels formalisms were introduced.
Introduction
Early predictions of nucleon polarization in deuteron stripping reactions ware
presented by Newns (1953), Hittmair (1956); Horowitz and Messiah (1953),
Cheston (1954), Sawicki (1957), Newns and Refai (1958), and Sachtler (1959,
1960). Experimentally, the polarization was observed for the first time by Hillman
(1956) for protons in the reaction 12C(d,p)13C leading to the ground state in 13C.
Later proton polarization from the ( d , p ) reaction has been also reported by
other authors who used 9Be, 10B, 12C, 28Si and 40Ca as target nuclei (Bokhari et al.
1958; Hensel and Parkinson 1958; Hird et al. 1959; Jurić and Čirilov 1959; Juveland
and Jentschke 1958).
When we started our experiment there were no published results on neutron
polarization in the ( d , n) reaction. However, in the course of our measurements a
brief abstract appeared about the measurements of Haeberli and Rolland (1957).
Their measurements were for low-energy deuterons of around 2.4-3.6 MeV.
Apart from this brief abstract, no additional information was available about their
results until about four years later (Haeberli 1961). There were no published
results for higher deuteron energies at that time.
Simple theories were used to explain the polarization generated in deuteron
stripping reactions. For reactions at small angles, the sign of the polarization was
linked with the total angular momentum of the transferred nucleon (Newns 1953;
Newns and Rafai 1958; Satchler 1959). Our aim was to study neutron polarization
for the 12C(d,n)13N reaction induced by 12.9 MeV deuterons and compare the
experimental results with the early theoretical predictions.
Experimental arrangement
Measurements of neutron polarization were carried out using the so-called “double
scattering” method (see Figure 1.5). The method consists of two steps, which
involves two targets. The first target is used to produce the polarization and the
second to measure it. In our case, neutron polarization was produced by the
12
C(d,n)13C reaction and analysed using the n-α scattering.
This method of measurements has been described by Wolfenstein (1956; see also
Ohlsen 1972). The differential cross section for the second target (the analyser) is
given by the following relation:
σ (θ ′,φ ) = σ 0 (θ ′)(1 + P1P2 cos φ )
where σ 0 (θ ′,φ ) is the cross section for the scattering of unpolarized neutrons, P1 is
the polarization produced in the 12C(d,n)13N reaction, P2 is the analyzing power of
the n-α scattering, and φ is the azimuthal angle of the polarization analyser. The
r r
polarization P1 is positive if it is in the direction of kd × kn (see Figure 1.5). The
r r r
azimuthal angle φ is defined by the product P1 ⋅ n , where n is the unit vector in the
r r
direction of kn ×k n′ .
Figure 1.5. The “double scattering” (or the two-step) method of polarization measurements used in the
study of neutron polarization for the 12C(d,n)13N reaction induced by 12.9 MeV deuterons. Neutron
polarization is produced in the first reaction and measured in the second reaction.
We can see that

L ≡ σ (θ ′,φ = π ) = σ 0 (θ ′)(1 − P1P2 )
R ≡ σ (θ ′,φ = 0) = σ 0 (θ ′)(1 + P1P2 )

and therefore
R−L
P1P2 =
R+L
Consequently, by measuring the right-left asymmetry in the second step and by
using its known analyzing power P2, one can determine the polarization P1
produced in the first step.
We can also notice that if the detectors are positionedr in the
r up (U) or down (D)
direction with respect to the plane defined by vectors kd and kn we have:
U ≡ σ (θ ′,φ = π / 2) = σ 0 (θ ′)
D ≡ σ (θ ′, φ = 3π / 2) = σ 0 (θ ′)
And consequently
U −D
=0
U+D
Thus by measuring the up and down asymmetries one can check the degree of the
spurious asymmetries created by imperfections in the experimental setup.
In our measurements we have used a beam of 12.9 MeV deuterons from the
120cm Cracow cyclotron. The beam passed through quadrupole magnetic lenses,
was deflected by a deflecting magnet, and focussed on a target situated in the
target hall about 1 2 m from the cyclotron. The experimental setup is shown in
Figure 1.6.
Figure 1.6. The experimental setup used in the measurements of neutron polarization from the
reaction 12C(d,n)13N. 1 – Tungsten ring; 2 – Carbon target; 3 – Paraffin collimator; 4 – Proportional
counter filled with helium; 5 – Hornyak-type scintillation counter; 6 – Lead shilling; 7 – Water
shielding; θ – reaction angle for the 12C(d,n)13N stripping reaction.
The intensity of the beam current on the target was 3-4μA. The target was made of
a water-cooled 0.3 mm thick tungsten foil covered with a layer of about 7mg/cm2
thick carbon. The first angle of the reaction, θlab = 15° (lab) corresponding to θc.m. =
17° was chosen because it corresponded to a relatively large product of the
expected differential cross section and polarization (Middleton, el-Bedawi and Tai
1953; Satchler 1959).
The elastic n-α scattering was used as the polarization analyser. The degree of the
polarization of neutrons scattered elastically from helium was determined by
Seagrave (1953) and Levintov, Miller and Shamshev (1957).
To reduce the background level of fast neutrons, coincidence method was used
during the measurements. Neutrons scattered elastically on helium nuclei in the
proportional counter were detected using a Hornyak-type fast neutron scintillation
counter (Hornyak 1952).3 Coincidences between the recoil helium nuclei and
scattered neutrons were registered for both the right and left detectors.
The proportional counter was filled with the spectroscopically pure helium at the
pressure of 11 atm. The scattering angle of neutrons was 900 in the centre-of-mass
(c.m.) system. The measurements were controlled by a fixed number of counts in
the proportional counter. The background produced by accidental coincidences
3
A fast neutron detector based on silver activated zinc sulphite, ZnS(Ag).
was determined by periodically switching on and off a 25 microsecond delay circuit
between the Hornyak counter and the coincidence system.
Supplementary measurements carried out using a pure tungsten target (i.e. without
carbon) had shown that the background of neutrons from the stripping reaction on
tungsten was lower than 10% of the yield produced by a target containing a layer of
carbon.
We have also carried out measurements of asymmetries in the up and down
direction to check for the possible instrumental contributions to the measured
asymmetries. They were found to be negligible. For instance, for the
measurements at θlab = 150
U −D
= 0.03 ± 0.04
U +D
Results and discussion

Results of our measurements are shown in Table 1.1. Neutron polarization from
the 12C(d,n)13N reaction at Ed = 12.9 MeV increases from around -0.39 at 150 (lab)
to around +0.60 at 600. Of particular interest is the polarization at 150 because it
can be readily compared with the early theoretical predictions.
Due to the unsatisfactory energy resolution of our polarization analyser we were

not able to separate precisely the neutron groups corresponding to different energy
levels in 13N. However, the neutron spectrum was dominated by the transition to
the 3.50/3.55 MeV doublet in 13N. The contribution from the transitions to the
ground state and first excited states were small and could be regarded as
negligible.
Similar strong transition to the 3.50/3.55 MeV doublet was observed over a wide
range of angles in measurements of angular distributions of the differential cross
sections for the 12C(d,n)13C induced by 8.1 MeV deuterons (Middleton, el-Bedawi
and Tai 1953). Likewise, measurements of the differential cross sections of protons
for the mirror reaction 12C(d,p)13C also showed a clearly dominant transition to the
3.7/3.9 MeV doublet in 13C
Table 1.1
Neutron polarization for the reaction 12C(d,n)13C induced by 12.9 MeV deuterons.
Reaction angle Neutron polarization Experimental error

(lab.) (%) (%)
15 -39 11
30 +3 8
45 +25 8
60 +55 20
The doubled at 3.50/3.55 MeV is made of states with spins 3/2- and 5/2+. Both states
are excited by the j = l + 1 / 2 neutron transfer. If the classical and early quantum-
mechanical models of nucleon polarization are correct and if the reaction is
dominated by the deuteron-nucleus interaction then the observed polarization
should be positive at 150 (see the Appendix A). Our results show that the
measured polarization at this angle is negative. This would mean that either the
reaction mechanism is associated with a strong proton-nucleus interaction or that
the observed polarization cannot be described using the proposed models.
The deuteron is a weakly bound particle. The domination of the proton absorption
over deuteron absorption is both physically and theoretically unlikely (see the
Appendix A). Consequently, we can conclude that our results demonstrated a
violation of the sign rule and thus showed that the early theories provided
inadequate description of the polarization mechanism.
A similar demonstration of the violation of the sign rule was shown experimentally
for the proton polarization 5 years after the publication of our results (Boschitz and
Vincent 1964). These authors measured angular distributions of both the differential
cross sections and proton polarization for the 12C(d,p)13C induced by 21 MeV
deuterons and leading to the ground state in 13C. This reaction is associated with
the l = 1 , j = l − 1 / 2 transfer and in compliance with the theoretically claimed sign
rule, the proton polarization within the range of the first stripping maximum should
be negative. In contrast, Boschitz and Vincent (1964) observed positive
polarization.
Classical and early quantum-mechanical theories give useful but oversimplified
description of the polarization in deuteron stripping reactions. The sign of the
polarization at forward angles is an unreliable test of the reaction mechanism. The
sign may depend on the deuteron energy and on the reaction angle within the
range of the first stripping maximum of the differential cross sections.
Our measurements at 600 suggest a violation of another simple classical and
quantum-mechanical rule about the maximum value of the polarization (Appendix
A). For the transition to the observed doublet, the theoretical maximum of the
absolute value of the polarization should be 39%. This value agrees with the
measured value at 150 but is only just within the experimental error for 600.
From the application point of view, the large absolute polarization values observed
at 150 and 600 suggest that the 12C(d,n)13N could serve as a good source of the
polarized fast neutrons.
In conclusion, we have demonstrated for the first time the violation of the
semiclassical sign rule and thus pointed out the limitations of the early theoretical
descriptions polarization phenomena.
Our results have also a historical value. They were the first experimental results in
nuclear physics in Poland, and the first in the world for the neutron polarization at
the medium deuteron energy.
References
Bhatia, A. B., Huang, K., Huby, R., and Newns, H. C. 1952, Phil. Mag. 53:485.
Budzanowski, A., Grotowski, K., Niewodniczanski, H., and Nurzynski, J. 1959, Bulletin
de l’Acedemie Plolonaise des Sciences, Série des sci. math., astr. at phys. 7:583.
Burrows, H. B., Gibson, W. H. and Rotblat, J. 1950, Phys. Rev. 80:1095.
Bokhari, M. S. Cookson, J. A., Hird, B. and Weesakul, B. 1958, Proc. Phys. Soc. 72:88.
Boschitz, E. T. and Vincent, J. S. 1964, NASA TR R-218, National Aeronautics and Space
Administration report, , Washington, D. C., December 1964.
Cheston, W. B. 1954, Phys. Rev. 96:1590.
Butler, S. T. 1950, Phys. Rev. 80:1095.
Butler, S. T. 1951, Proc. Phys. Soc. A208:559.
EPPOG 2006, ‘Hands on Physics’, http://wyp.teilchenphysik.org/inst_pl_krakow_en.htm
Haeberli, W. 1961, Helv. Phys. Acta, Suppl. VI, 157.
Haeberli, W. and Rolland, W. W. Bull, Am. Phys. Soc. 2:234.
Hansel, J. C. and Parkinson, W. C. 1958, Phys. Rev. 110:128.
Hird, B., Cookson, J. A. and Bokhari, M. S. 1959, Comptes Reudus du Con-gres International
de Physique Nucléaire, Paris 7-12 Juillet 1958, p. 470, Dunod, Paris.
Hittmair, O. 1956, Z. Physik 144:449.
Hodgson, P. E. 1971, Nuclear Reactions and Nuclear Structure, Clarendon Press,
Oxford.
Holt, J. R. and Young, C. T. 1950, Proc. Phys. Soc. (London) 78:833.
Hornyak, W. F. 1952, Rev. Sci. Intsr. 23:264.
Horowitz, J. and Messiah, A. M. L. 1953, J. Phys. Radium 14:731.
Hubby, R. 1950, Nature 166:552.
Hubby, R. and Newns, H. C. 1951, Phil. Mag. 42:1442.
IFJ 1965, Institute of Nuclear Physics, Cracow, Wojskowe Zaklady Kartograficzne,
Krakow, Poland
IFJ 2005, ‘The Henryk Niewodniczanski Institute of Nuclear Physics’,
http://www.ifj.edu.pl/intro/?lang=en
Jurić, M. K. and Čirilov, S. D. 1969, Comptes Rendus du Congres International de. Physique
Nucléaire, Paris 7-12 Juillet, 1958, p. 473, Dunod, Paris.
Juveland, A. C. and Jentschke, W. 1958, Phys. Rev. 110:456.
Levintov, I. I., Miller, A. V. and Shamshev, V. N. 1957, Nucl. Phys. 3:221.
McGruer, I. N., Warburton, E. K. and Bender, S. R. 1955, Phys. Rev. 100:235.
Middleton, R., el-Bedewi, P. A. and Tai, C. T. 1953, Proc. Phys. Soc. A66:95.
Newns, H. C. 1953, Proc. Phys. Soc. 66:477.
Newns, H. C. and Refai, M. Y. 1958, Proc. Phys. Soc. 71:627.
Ohlsen, G. G. 1972, Rep. Prog. Phys. 35:717.
Sawicki, J. 1957, Phys. Rev. 106:172.
Satchler, G. R. 1959, Comptes Rendus du Congres International de Physique Nucléaire,
Paris 7-12 Juillet 1958, p. 101, Dunod, Paris.
Satchler, G. R. 1960, Nucl. Phys. 18:110.
Seagrave, J. D. 1953, Phys. Rev. 92:1222.
Tobocman, W. 1956, Tech. Rep. No 29, Case Institute of Technology
Wolfenstein, L. 1956, Ann. Rev. Nucl. Sci. 6:43.
2
A Systematic Discontinuity in the Diffraction Structure
Key features:
1. We have measured angular distributions of the differential cross sections for the
elastic scattering of 12.8 MeV deuterons on Be, C, Mg, and Ca nuclei. We have
observed an irregularity in the diffraction structure for C and Ca.
2. The subsequent compilation of available data in the vicinity of the 12.8 MeV
bombarding energy revealed a systematic discontinuity in the diffraction pattern in the
form of a vanishing maximum as the mass number of the target nucleus increases
beyond a certain value.
3. In order to reconstruct the observed pattern by using the diffraction theory it is
necessary to assume a contribution of two different geometries for scattering in the
forward direction and backward directions.
4. Optical model can reproduce the observed irregularities in the diffraction structure.
They are interpreted as a ‘geometric’ effect.
Abstract: Angular distributions for the elastic scattering of 12.8 MeV deuterons on Be, C, Mg,
and Ca nuclei have been measured at 50 intervals for angles 150 – 1450 (lab) using a counter
telescope. In all cases, the measured distributions show a pronounced diffraction pattern. A
systematic breakdown in the diffraction structure is demonstrated.
Introduction
Measurements of the elastic scattering of 11.8 MeV (Igo, Lorenz, and Schmidt-Rohr
1961) and 13.6 MeV deuterons on carbon (Gofman and Nemec 1961) revealed a
curious and unexpected feature (see Figure 2.1). The angular distribution of the
differential cross sections measured at 11.8 MeV has only one maximum in the
angular range of around 400 – 1100, but the distribution at 13.6 MeV has two. The
two measurements are separated by only about 2 MeV and one would expect a
smooth transition between the two energies without a change in the number of
oscillations. The observed curious discontinuity was puzzling and merited further
investigation.
Figure 2.1. The angular distributions of the differential cross sections for the elastic scattering of 11.8
MeV (Igo, Lorenz, and Schmidt-Rohr 1961) and 13.6 MeV (Gofman and Nemets 1961) deuterons.
The measurements suggested a discontinuity in the diffraction pattern. (One oscillation maximum in
the middle range of scattering angles for the 11.8 MeV deuterons is replaced by two when the
deuteron energy is increased to 13.6 MeV.)
The scattering of nucleons (protons or neutrons) can be relatively easily described
theoretically. However, it has not been immediately obvious whether simular
treatment could be also applied to deuterons. The deuteron is a weakly bound
particle. Its bounding energy is only 2.225 MeV, which is about four times lower than
the bounding energy of tritons (8.482 MeV) or 3He (7.718 MeV). Such a weakly
bound particle can be expected to lead readily to a stripping reaction, where one
component moves away relatively undisturbed while the other is captured by the
target nucleus. Another likely process is a deuteron break-up.
A study of deuteron scattering was giving a convenient insight into the stability of this
weakly bound system. Consequently, the observed anomaly presented itself as a
phenomenon that should be further investigated.
The energy available in our laboratory was 12.8 MeV, which was between the two
energies for which previous measurements were carried out. We have therefore
decided to study the observed anomaly by carrying the measurements of elastic
scattering not only for carbon but also for a few other nuclei, Be, Mg, and Ca.
Experimental procedure
Angular distributions of the differential cross sections for the elastic scattering of 12.8
MeV deuterons were measured using deuterons accelerated in the 120 cm cyclotron
of the Institute of Nuclear Physics in Cracow (Poland). The beam energy of the
120cm cyclotron was estimated to be (12.8 ±0.3) MeV.
The accelerated deuteron beam was directed to the target area by a system of
magnetic quadrupole lenses and by a horizontal deflection magnet. The target was
located about 12 meters from the cyclotron. The beam of deuterons was cut by a
collimating entrance aperture to a spot of 4 mm diameter on the target. The target
chamber with the detecting system is shown in Figure 2.2.
Figure 2.2. The experimental arrangement.
In order to minimize the interference of protons from the (d,p) reaction it was
necessary to have a particle identification system. Our detector system consisted of
a counter telescope, which was made of two scintillation counters. The first counter
with a thin CsI(Tl) or plastic scintillator acted as a dE/dx detector. With the deuteron
energy loss of about 1.5 MeV in the scintillator the energy resolution of the counter
was about 11%.
The energy loss per unit path, dE/dx, can be expressed as (Bethe and Livingstone
1937):
dE 4πe 4 Z 2 ⎧ 2mV 2 ⎫
− = − log (1 − β ) − β 2 ⎬
2
nz ⎨log
dx mυ 2
⎩ I ⎭
where e is the electron charge, m – the mass of the electron, Z – the atomic number
of the detected particle, υ – the velocity of the particle, z – the atomic number of the
stopping material, n – the number of atoms per cubic centimetre in the stoping
material, β – the ratio of the particle velocity to the velocity of light, and I – the
average ionisation energy of the stopping material.
For β << 1 this expression can be significantly simplified:
dE MZ 2 E
− = C1 log C2
dx E M
where M is the mass of the detected particle and C1 and C2 are constants.
The logarithmic term is a slow varying function of energy and thus the above
expression can be further simplified as:
dE MZ 2
= C3
dx E
In general, the discrimination between various types of particles is done by using the
product:
dE
⋅ E = C3 MZ 2
dx
However, in our measurements clean spectra were obtained by setting a gate on the
output of the dE/dx detector, which was run in coincidence with the E – counter made
of a CsI(Tl) scintillator sufficiently thick to stop the entering deuterons.
The energy resolution of the detection system was about 5%. The gated pulses from
the E - counter were fed into a 100-channel Hutchinson-Scarrott amplitude analyser.
In all cases, except for measurements at largest angles, the resolution of the
detection system ensured a good separation of elastically scattered deuterons from
protons. The defining aperture of the counter telescope subtended an angle of 1.2°.
The detecting system was attached to a phosphor-bronze strip making the sidewall
of the scattering chamber. This arrangement allowed for selecting a scattering angle
in the range of 0° - 145° without disturbing the vacuum in the scattering chamber.
The accuracy of the angle settings was about ±0.25°.
The beam current was integrated by means of a current integrator connected to a
beam stopper, which was placed behind the target. In addition, a CsI(Tl) scintillation
monitor, which detected particles at a fixed angle of 30° was used.
The determination of the absolute values of the differential cross-sections was
checked by using the cross-section for the elastic scattering of deuterons from gold.
Separate measurements indicated that 12.8 MeV deuterons scattering from gold
could be described using Rutherford formula for angles of up to about 35° (see
Figure 2.3). The elastic scattering from gold at small angles was measured several
times during the experiment and used in the calculations of the absolute values of
differential cross-sections.
Figure 2.3. The ratio of the measured and Rutherford cross-sections for the elastic scattering of 12.8
deuterons scattered elastically from Au target.
The targets were in the form of thin foils: 0.42 mg/cm2 thick for Au, 7.6 mg/cm2 for
Be, 4.1 mg/cm2 for Mg and 4.6 mg/cm2 for Ca. A polystyrene foil of 2.11 mg/cm2 was
used as a carbon target.

Results
The differential cross-sections were measured in steps of 5° for angles between 15°
and 145° (lab). Results are presented in Figures 2.4 and 2.5. The relative errors take
into account the statistical errors, the inaccuracies in beam integration and errors
associated with angle inaccuracies. The absolute cross-sections for C, Mg, and Ca
were estimated at ±10%, and for Be at ±15%.
A convenient way to see the structure of the angular distributions for the elastic
scattering is to plot them as ratios to the Rutherford cross-sections. The expression
for the Rutherford scattering cross section has the following form (Rutherford 1911):
2
⎛ dσ ⎞ ⎛ zZe 2 ⎞ 1
⎜ ⎟ = ⎜⎜ ⎟⎟
⎝ d Ω ⎠R ⎝ 2 μυ ⎠ sin 4
(θ / 2)
where ze and Ze are the charges of the incident particle and of the target nucleus, μ
is the reduced mass of the interacting particles, υ is the initial velocity of the
projectile, and θ is the centre-of-mass angle of the scattered projectile.
The Rutherford scattering cross section can be expressed in terms of more
convenient variables and in mb/sr. After substituting the relevant variables one gets:
2
⎛ dσ ⎞ ⎛ A ⎞
2
⎛ zZ ⎞ ⎜⎜1 + p ⎟⎟
1
⎜ ⎟ = 1.2959964⎜ ⎟ (in mb/sr)
⎝ dΩ ⎠ R ⎝ E ⎠ ⎝ At ⎠ sin ( θ / 2 )
4
where, E is the laboratory energy of the projectile in MeV, Αp is the atomic mass
number of the projectile and At the atomic mass number of the target nucleus.
Figure 2.4. The measured differential cross sections (open circles) for the elastic scattering 12.8 MeV
deuterons scattered elastically from Be and C nuclei are compared with the Rutherford cross sections
(dashed lines). The solid lines through the data points are to guide the eye.
Figure 2.5. The measured differential cross sections (open circles) for the elastic scattering 12.8 MeV
deuterons scattered elastically from Mg and Ca nuclei are compared with the Rutherford cross
sections (dashed lines). The solid lines through the data points are to guide the eye.
Plots of the measured differential cross-sections divided by the Rutherford cross-

sections (dσ/ dσR) are presented in Figure 2.6. As can be seen, the measured cross-
sections show pronounced diffraction structure. For the lightest elements, Be and C,
the values of the measured cross-sections exceed considerably the values of the
Rutherford cross-sections, especially at backward angles. Both Be and Mg data
display a regularly spaced diffraction pattern. However, the data for C and Ca show
two nearly merged maxima in the middle range of the reaction angles.
Figure 2.6. Angular dependence of the ratios of the measured and Rutherford scattering cross-
sections for Be, C, Mg and Ca.
Comparing with other data

Our measurements for carbon target can now be compared with the previous
measurements (Gofman and Nemets, 1961; Igo, Lorenz, and Schmidt-Rohr, 1961)
mentioned in the Introduction. This is done in Figure 2.7. As can be seen, the
measured distributions display a clear transition from one maximum in the 400-1000
range at 11 MeV, via two nearly merging maxima at 12.8 MeV to two distinct maxima
at 13.6 MeV.
Figure 2.7. Our results at 12.8 MeV are compared with the measurements of Igo, Lorenz, and
Schmidt-Rohr (1961) at 11.8 MeV and Gofman and Nemets (1961) at 13.6 MeV. They show a
gradually emerging maximum around 900 when the projectile energy is increasing.
In view of this now clearly demonstrated smoothly varying pattern for carbon and the
curious irregular pattern for Ca, I have decided to compile and examine all the
available data on deuteron scattering in the vicinity of the energy used in our
experiment. I have also decided that the easiest way to study the variations in the
diffraction pattern is to draw the positions of the maxima and the minima in the
measured angular distributions. Results of my study are shown in Figure 2.8.
Figure 2.8. The positions of maxima and minima in the differential cross-sections as a function of the
mass number and incident deuteron energy: 11 MeV (Takeda 1960); 12. 8 MeV, our results work; 13.
6 MeV for Ti, Fe, Cu, Zr (Gofman and Nemec 1961); 15 MeV (Cindro and Wall 1960); 21. 6 MeV
(Yntema 1959).
The figure shows a smooth dependence of the positions of maxima and minima on
the mass of the target nucleus and on the incident deuteron energy. The positions
are shifted towards smaller angles with the increasing deuteron energy and mass
number, which is consistent with the diffraction interpretation of elastic scattering
(see below). However, Figure 2.8 shows also a systematic discontinuity in the
diffraction pattern: one of the maximum disappears when either the mass number or
incident energy is increased beyond a certain value. Insufficient data at 21.6 MeV did
not allow for a systematic study of changes in the diffraction pattern at this energy.
Following the publication of our results, Tjin a Djie, Udo, and Koerts (1964) published
their results for scattering of 26 MeV deuterons from a range of target nuclei. They
also observed a systematic change in the diffraction structure and a disappearance
of a maximum. Their results are presented in Figure 2.9, which displays similar
pattern as in the compilation presented in Figure 2.8
Figure 2.9. The positions of maxima of angular distributions for the elastic scattering of 26 MeV
deuterons (Tjin a Djie, Udo, and Koerts 1964).
Different types of irregularities
It is important to distinguish between this observed systematic discontinuity and the
apparent random irregularities observed either for elastic scattering or transfer
reactions.
For instance, the disappearance or emergence of diffraction maxima was observed
in 16O(3He,α0)15O reaction in the energy range of 9.8 – 11.2 MeV for the incident 3He
particles (Bray, Nurzynski, and Bourke 1966). This feature is displayed in Figure
2.10. The left-hand side of the figure shows the angular distributions and the right-
hand side the positions of maxima and minima. The patterns are irregular and they
appear to be associated with compound nucleus contributions.
Figure 2.10. Angular distributions (the left-hand side of the figure) and the positions of maxima and
minima (closed and open circles, respectively on the right-hand side of the figure) for the
16
O(3He,α0)15O reaction (Bray, Nurzynski, and Bourke 1966).
Irregularities in the diffraction structure were also observed for the elastic scattering
of 27.0-35.5 MeV α – particles (Mikumo 1961). Resonances in excitation functions
were identified and the observed irregularities were also interpreted as arising from
the compound nucleus contributions.
Another good example of irregularities associated with compound nucleus
mechanism is for the elastic scattering 12C(α,α)12C in the energy range of around 11
– 23 MeV (Atneosen et al. 1964; Carter, Mitchel, and David 1964). Using their
angular distributions I have prepared a plot of the positions of maxima and minima,
which is presented in Figure 2.11.
Excitation functions measured at 165.80 for 10-19 MeV α - particles revealed a
number of resonances clustered around 11.5, 13, 15, and 18.5 MeV (Carter 1962;
Carter, Mitchell, and Davis 1964). Their positions are marked by arrows in Figure
2.11 and as can be seen, they are roughly located where strong irregularities in the
angular distributions are observed.
No excitation function was measured for energies 20-23 MeV. However, if we use
the data at 1800 we can reproduce the excitation function and find a strong maximum
at around 22 MeV. At this energy, which is also indicated by an arrow in Figure 2.11,
a maximum at around 1500 is replaced by a maximum at around 1050 when the
incident energy of α particles is increased.
In contrast, the patterns observed for the elastic scattering of deuterons in the
energy range of 11-15 MeV, as shown in Figure 2.8, are of an entirely different
nature. They are regular and systematic, and they also involve a wide range of target
nuclei. They do not therefore appear to be in any way connected with the formation
of a compound nucleus but must be caused by some other mechanism.
Figure 2.11. Irregularities in the positions of maxima and minima of the elastic scattering angular
distributions, which can be identified as being associated with compound nucleus interaction. The
horizontal arrows show the positions of resonances in the excitation functions. They roughly coincide
with the discontinuities in the diffraction structure of the angular distributions. The plots are based on
the measurements of Atneosen et al. (1964), Carter 1962, and Carter, Mitchell, and Davis (1964).
The impact parameter

If we examine the Figure 2.8 we shall notice that in all three examples where one
maximum disappears, the curves seem to follow two different patterns. The gradient
for the lowest curves appear to be distinctly smaller than the gradient for the upper
curves. The difference appears to be larger than normally expected for the normal
diffraction scattering.
As the mass number for the target nucleus increases, the top curves come quickly
into the region of the lower curves and one of the maximum belonging to the higher
curves cannot be accommodated. This feature is observed as a disappearing
maximum. However, this distinction is less clear for the 26 MeV data, where results
for very light target nuclei (H, D, and He) have been included.
The lower curves belong to a large impact parameter, i.e. when the approaching
deuteron passes the target nucleus at a relatively large distance. The upper curves
belong to deuterons with smaller impact parameters, i.e. when the passing deuteron
feels better the influence of nuclear forces.
The diffraction model analysis
In order to study the mechanism of the observed systematic discontinuities in the
diffraction structure of the angular distributions I have carried out a conceptual
analysis based on the diffraction theory of nuclear interaction (see the Appendix B).
Diffraction theories have been used extensively in the early interpretations of
experimental data (see for instance Dar, 1963, 1964 and Bassichis and Dar 1965)
In its simplest form, the differential cross section for scattering from a black disc can
be expressed as
dσ
2
⎛ J ( x) ⎞
= (ki R02 ) 2 ⎜ 1 ⎟
dΩ ⎝ x ⎠
where k i≡ 1 / D i is the wave number for the incident particles, R0 the nuclear radius,
and J1(x) is the cylindrical Bessel function of the first order.
r r
The variable x = qR 0 , where q = ki − k f is the momentum transfer.
q 2 = ki2 + k 2f − 2ki k f cosθ
where k f is the wave number for the outgoing particles and θ is the scattering angle.
For the elastic scattering ki = k f and therefore x = 2ki R0 sin(θ / 2) .
A graph of the universal function [J1( x ) / x ] is shown in the Appendix B.

2
In order to calculate theoretical locations of the positions of maxima in the angular

distributions we first have to determine the locations of the maxima of the function
[J1( x ) / x]2 . This can be done by calculating the derivative of this function. The
maxima xi of the [J1( x ) / x ] satisfy the equation
2
2
d ⎡ J1( x ) ⎤ d ⎡ J ( x )⎤
⎢ ⎥ =2 ⎢ 1 =0
dx ⎣ x ⎦ dx ⎣ x ⎥⎦
Using the following relation for the Bessel functions
nJ n ( x ) − xJ n′ ( x ) = xJ n +1( x )
where
d
J n′ ( x ) ≡ Jn( x )
dx
we can find that
d ⎡ J 1( x ) ⎤ J (x)
⎢ ⎥ =− 2
dx ⎣ x ⎦ x
Thus, the location of the maxima xi of the [J1( x ) / x ] function are given by the
2
equation
J 2 ( x)
=0
x
The first maximum is at x0 = 0 because
⎡ J ( x )⎤
lim ⎢ 2 =0
x →0
⎣ x ⎥⎦
Other positions coincide with the xi values for which J 2 ( x ) = 0 . They can be
calculated by a linear interpolation of the tabulated values near the J 2 ( x ) = 0 value.
Alternatively, the position of the maxima of the [J1( x ) / x ] function can be calculated
2
by fitting polynomials to points around the maxima of this function.

The xi values, calculated using both methods are listed in Table 2.1. The last number
calculated using the derivative
d ⎡ J1( x ) ⎤
=0
dx ⎢⎣ x ⎥⎦
represents an approximate value because the table of J 2 ( x ) , which I have used
ends at x = 17.5 (Abramowitz and Stegun 1964).
Table 2.1
[ ]2
Positions xi of the maxima of the J1( x ) / x function
Method x0 x1 x2 x3 x4 x5
Derivative 0.00 5.14 8.42 11.62 14.80 18.00

Polynomial 0.00 5.14 8.43 11.62 14.80 17.97
The determined positions of the maxima of the [J1( x ) / x ] function can now be
2
translated into the mass and angle dependence of the positions of the maxima in the
angular distributions. The relevant conversion formula is
⎛ xi ⎞
θ = 2 sin −1 ⎜⎜ ⎟⎟
⎝ 2kR0 ⎠
where xi is the position of an ith maximum of the [J1( x ) / x ] function.
2
The convenient expression for k is:

k = 0.2187 μEc .m . (fm-1)
where μ is the reduced mass
M pMt Ap At
μ= ≈
M p + M t Ap + At
and
Elab Elab
Ec .m . = ≈ (MeV)
1 + M p / M t 1 + Ap / At
Alternatively, using Elab
k = 0.2187 μ Elab / Ap
Figure 2.12 shows the calculated angle-mass dependence of the positions of

maxima in the angular distributions for the elastic scattering of 12.8 MeV deuterons.
The three upper curves were calculated using R0 = r0 A1 / 3 and the lowest two curves
using R0 = r0 A1 / 3 + 4.2 . For both sets, the parameter r0 = 1.7 fm was used.
This conceptual diagram shows that in order to reproduce the patterns observed
experimentally using diffraction model, one has to assume contributions from two
different geometries. Whether such contributions can be physically justified is not
clear.
Figure 2.12. The positions of the maxima in the elastic scattering angular distributions calculated
using the diffraction model. The two lower curves were calculated using R0 = r0 A1 / 3 + 4.2 fm and the
upper curves using R0 = r0 A1 / 3 fm. In both cases, r0 = 1.7 fm has been used.
It is now well known that the optical model formalism can account for the observed
systematic irregularities. They are interpreted as reflecting a ‘geometrical’ effect
(Hodgson 1964, 1966; Wilmore and Hodgson 1964). This approach does not explain
the physics of the observed phenomenon but only shows that by adjusting optical
model parameters one can describe with some degree of success the measured
angular distributions.
We have also analysed our Ca data using optical model (see Chapter 4). As can be
seen in Figure 2.6, the angular distribution for this nucleus shows an irregularity in
the form of two merging maxima. We were able to reproduce the experimental data
by introducing a spin-orbit component in the optical model potential.
Summary and conclusions

Prompted by observed irregularities in the angular distributions for the elastic
scattering of 11.8 and 13.6 MeV deuterons from carbon (Gofman and Nemec 1961;
Igo, Lorenz, and Schmidt-Rohr 1961) we have carried out measurements of the
elastic scattering at 12.8 MeV for Be, C, Mg, and Ca nuclei. Our measurements
demonstrated a smooth transition between 11.8 and 13.6 MeV. We have observed
clear discontinuities in the diffraction structure for C and Ca in the form of two nearly
merging maxima in each case.
My compilation and analysis of existing data in the vicinity of 12.8 MeV revealed a
puzzling systematic discontinuity in the diffraction structure in the form of a vanishing
maximum as the mass of the target number increases beyond a certain value. This
systematic discontinuity cannot be explained by a compound nucleus mechanism
and must have a different physical interpretation.
Following the publication of our results, similar systematic discontinuity was also
reported for 26 MeV deuterons scattered from a wide-range of target nuclei (Tjin a
Djie, Udo, and Koerts 1964).
In an attempt to understand the physics of the observed phenomenon, I have carried
out calculations using a simple black-disc diffraction theory. I have found that in
order to reproduce the observed pattern I had to assume a contribution from two
geometries for the elastic scattering. The two geometries are distinguished by two
radii: R0 = r0 A1 / 3 + 4.2 and R0 = r0 A1 / 3 fm with r0 = 1.7 fm in both cases. Whether such
two contributing geometries can be physically justified is not clear.
Later optical model calculations indicated that the observed systematic
discontinuities can be reproduced by suitably adjusting optical model parameters.
The discontinuities are interpreted as a geometrical effect. However, such optical
model analyses do not explain the physics of the observed phenomenon.
References
Abramowitz M. and Stegun, I. A. 1964, Handbook of Mathematical Functions, United
States Department of Commerce, National Bureau of Standards, Washington,
DC.
Atneosen, R. A., Wilson, H. L., Sampson, M. B., and Miller, D. W. 1964, Phys. Rev.
135:B660.
Bassichis, W. and Dar, A. 1965, Phys. Rev. Lett. 14:648.
Bethe, H. A. and Livingstone, M. S. 1937, Rev. Modern Phys. 9:245.
Bray, K. H., Nurzynski, J., and Bourke, W. P. 1966, Phys. Lett. 21:536.
Carter, E. B. 1962, Ph.D. Thesis, Florida State University (unpublished).
Carter, E. B., Mitchell, G. E., and Davis, R. H. 1964, Phys. Rev. 133:B1421.
Cindro, N., and Wall, N. S. 1960, Phys. Rev., 119:1340.
Dar, A. 1963, Phys. Lett. 7:339.
Dar, A. 1964, Nucl. Phys. 55:305.
Gofman, Yu. V., and Nemets, O. F. 1961, Soviet Phys. JETP 12:1035; Soviet Phys.
JETP 13:333.
Hodgson, P. E. 1964, Compt. Rend. du Congrès International de Physique Nucléaire,
Paris, Vol. I, p. 257.
Hodgson, P. E. 1966, Advances in Physics, 15:329.
Igo, G., Lorenz, W., and Schiidt-Rohr. U. 1961, Phys. Rev. 124:832.
Mikumo, T. 1961, J. Phys. Soc. Japan, 16:1066.
Rutherford, E. 1911, Phil. Mag. 21:669.
Takeda, M. 1960, J. Phys. Soc. Japan, 15:557.
Tjin a Djie, H. R. E., Udo, F., and Koerts, L. A. 1964, Nucl. Phys. 53:625.
Wilmore, D. and Hodgson, P. E. 1964, Nucl. Phys. 55:673.
Yntema, J. L. 1959, Phys. Rev., 113:261.
3
Core Excitations in 27Al
Key features:
1. Nuclear excitations are usually interpreted using either single-particle or collective
models. However, it has been suggested (De Shalit 1961; Lawson and Uretsky 1957)
that some excited states in odd-mass nuclei could be described as a coupling of a
single nucleon or hole with an excited collective state of an even-mass core. This
mode of excitations is now known as core excitations.
2. Originally, I have intended to study the possibility of such excitations by using a pair of
24
Mg and 25Mg nuclei. However, the prohibitive price of enriched 25Mg isotope forced
me to opt for more readily available 27Al and 28Si pair of nuclei. We have measured
elastic and inelastic scattering of 12.8 MeV deuterons from these nuclei.
3. I have carried out an analysis of our experimental results using diffraction theories
(with a sharp and smooth cut off radii) and a plain wave theory. Our results were also
analysed using a strong coupling model.
4. This study resulted in a textbook demonstration4 of the existence of core-excited
states in 27Al.
5. A broad-range magnetic spectrograph we have designed and constructed was used
for the first time in these measurements. Consequently, much of our experimental
work was devoted to testing and bringing it into operation.
Abstract: Angular distributions for the elastic and inelastic scattering of 12.8 MeV deuterons
to the first few excited states of 27Al and 28Si were measured using a broad-range magnetic
spectrograph and nuclear emulsions. The overall resolution of around 150 keV allowed to
distinguish between deuteron groups belonging to the first and the second excited states in
27
Al, which was essential in the study of core excitations. Experimental data have been
analysed using diffraction theories, plane wave theory, and strong coupling model. Results
demonstrated that low-lying levels in 27Al could be interpreted as belonging to a quintuplet of
core-excited states.
Introduction
The low-lying nuclear levels are usually interpreted using single-particle or collective
models. Single-particle excitations are associated with configurations characterised
by their orbital momenta l and total spins j. The remaining nucleons are usually
treated as passive observers.
For instance, in a single particle stripping reaction a single nucleon is seen as being
deposited into a shell-model orbital characterised by l and j. Likewise, a single
particle pickup is interpreted as involving a pickup of a single nucleon characterised
by specific l and j values. Reactions involving more than one nucleon can also be
interpreted as a transfer of single nucleons to or from specific shell-model orbitals.
The collective excitations involve the collective response of all the nucleons in the
nucleus. The nucleus can be described either as a rigid rotating or a softer vibrating
body or by a combination of the two. Collective excitations can be studied
conveniently by inelastic scattering.
At around the mid 1950s, it has been suggested that some excited states in odd-
mass nuclei can belong to a different type of excitations, in which a single particle or
4
See P. E. Hodgson 1971, Nuclear Reactions and Nuclear Structure, Clarendon Press, Oxford, pp.
386-392; and in R. R. Roy and B. P. Nigam 1967, Nuclear Physics: Theory and Experiment, John
Wiley & Sons, Inc., New York, p. 429, 430.
a hole with a specific shell-model l and j configuration is couple to a collective state
of the core made of the remaining nucleons (De Shalit 1961; Lawson and Uretsky
1957). These modes of excitations are now known as core-excitations.
For instance let us consider a single particle with spin j = 5/2+ coupled to a core with
spin Jc = 2+ (see Figure 3.1). The coupling results in a quintuplet of states with spins
ranging from 1/2+ to 5/2+.
Figure 3.1. The core-excited model
The signatures of core-excited states are as follows:

(1) In an odd-mass nucleus, the coupling of a single particle spin j with an excited
even-even mass core spin Jc will results in a multiplet of states with spins I:
Jc − j ≤ I ≤ Jc + j
(2) The centre of gravity of the multiplet should be approximately equal to the energy
of the collective state of the core
∑ (2 I + 1) E I
Ec ≈ I
∑ (2I + 1)
I
where Ec is the energy of the collective state of the core and EI are the excitation
energies of states in the core-excited multiplet in the odd-mass nucleus.
(3) The shape of the inelastic scattering angular distributions of the differential cross
sections for the members of the multiplet and for the excited state of the core should
by approximately the same. This is because the excitation of members of the
multiplet is formed by only one transition: from the ground state to the excited state
of the core.
(4) The absolute values of the differential cross sections [dσ / dΩ]I for the inelastic
scattering to states I in the odd-mass nucleus and the differential cross section
[dσ / dΩ]Jc for inelastic scattering to a core state Jc in the even-mass nucleus should
satisfy the following relation:
⎡ dσ ⎤ (2 I + 1) ⎡ dσ ⎤
⎢⎣ dΩ ⎥⎦ ≈ (2 I + 1) ⎢⎣ dΩ ⎥⎦
I ∑ I
Jc
The sum of the cross sections for the multiplet should be approximately equal to the
cross section of the core
⎡ dσ ⎤ ⎡ dσ ⎤
⎢⎣ dΩ ⎥⎦ ≈ ∑ ⎢⎣ dΩ ⎥⎦
Jc I I
In order to investigate the possibility of the existence of core-excited states I first

intended to use the 24Mg and 25Mg pair of nuclei. The 25Mg nucleus has a single
neutron in the 1d5/2 configuration located over a well-known rotational core made of
24
Mg. It also has a compliment of low-lying states with spins ranging from 1/2+ to 9/2+.
These states presented themselves as good candidates for a core-excitation
multiplet formed by coupling the 1d5/2 single neutron to the first excited 2+ state of the
24
Mg core. Unfortunately, in its natural form, Mg contains only 10% of 25Mg and the
price for enriched isotope was too high.
I have therefore decided to try another combination made of the 27Al and 28Si pair.
The low-lying states in 27Al form almost a mirror image of the states in 25Mg. The
advantage of using this pair is that natural Al contains 100% of 27Al and natural Si
has 92.2% of 28Si. Thus the targets can be made of natural elements, which is both
cheap and convenient. The difference between the two systems is that in 27Al it is a
1d5/2 proton hole, which would be coupled to the 2+ spin of the 28Si core to produce
the familiar core-excited multiplet.
Table 3.1
Similarities between the two pairs of nuclei, Mg – 25Mg and 28Si – 27Al, which could be used to study
24
the core-excited model
Nucleus Ex (MeV) Spin Nucleus Ex (MeV) Spin

24 + 28
Mg 1.3687 2 Si 1.7790 2+
25 5 + 27 5 +
Mg g.s /2 Al g.s. /2
1 + 1 +
0.5850 /2 0.8438 /2
3 + 3 +
0.9747 /2 1.0145 /2
7 + 7 +
1.6118 /2 2.2111 /2
5 + 5 +
1.9646 /2 2.7349 /2
1 + 3 +
2.5638 /2 2.9820 /2
7 + 9 +
2.7377 /2 3.0042 /2
3 + 1 +
2.8015 /2 3.6804 /2
9 + 1 +
3.4052 /2 3.9568 /2
Ex – Excitation energy.
Experimental procedure and results
Measurements were carried out using a beam of 12.8 MeV deuterons from the 120
cm cyclotron in the Institute of Physics in Cracow, Poland. The overall diagram of the
experimental equipment, including the cyclotron, is shown in Figure 3.2.
Figure 3.2. The general layout of the experimental equipment: (C) the 120-cm cyclotron, (Q)
quadrupole lenses, (D) deflection magnet, (T) target area, (S) the broad-range magnetic
spectrometer.
After leaving the cyclotron (C), the accelerated deuterons passed through a system
of two quadrupole lenses (Q) and were deflected by 130 by a deflection magnet (D)
before entering the target area (T).
The scattering chamber used in these measurements was similar to the one
described in Chapter 2. However, certain modifications were introduced, one of them
being a system of slits, which was designed to facilitate a convenient control of
accelerated deuterons and to improve resolution of detected particles (see Figure
3.3). Other modifications included adding a connection to the broad-range
spectrometer. In fact, the whole chamber had to be first removed to install a system
of supporting rails for the spectrometer.
The slit system
An insulated tungsten aperture with the diameter of 6 mm was placed about 800 mm
from the target. Current from this aperture was measured and used in focusing the
deuteron beam on the target. The tungsten aperture was followed by a collimating
system, which contained two 2 mm by 12 mm gold slits and between them three 5
mm diameter antiscattering brass slits. The system collimated the deuteron beam to
less than 0.50 and contributed significantly to improving the effective energy
resolution of the detected particles. Good resolution was important in our experiment
because we aimed at resolving the first two excited states in 27Al, which were
essential in studying the core-excited model. Without taking special precautions,
resolution is low for particles accelerated by a cyclotron.
Figure 3.3. Target chamber and the system of entrance slits: 1. An insulated tungsten aperture used
to guide the deuteron beam to the target. 2. The Au collimating slits. 3. The antiscattering 5 mm
diameter brass slits. 4. The Au beam monitor target. 5. Scintillation counter used as a beam monitor.
6. Light guide. 7. The main target. 8. Faraday cup.
Beam monitoring
The deuteron beam intensity was monitored using beam integrator connected to a
Faraday cup. In addition, deuteron elastic scattering from a thin, 0.4 mg/cm2 Au
target at a fixed angle of 390 was used. The scattered deuterons were detected
using a thin CsI(Tl) crystal located on a light guide and connected to an EMI 6097
photomultiplier located in an antimagnetic shield.
Detection of reaction products
Deuterons scattered from the main target (27Al or 28Si) were directed to a broad
range magnetic spectrometer, which we have designed and constructed, and were
detected using 100 μm thick nuclear emulsions. The chamber was vacuum-sealed
using a thin, Melinex polyester film. Nuclear emulsion plates were placed outside the
spectrometer chamber and thus could be easily replaced. For the inelastic
scattering, two plates 30 mm x 90 mm were used. Elastic scattering was measured
using only one plate, 30 mm x 46 mm at each angle. A view of the detection system
is presented in Figures 3.4 and 3.5.
Determination of the focal plane, solid angle and defocusing coefficient
Our spectrometer was used for the first time in this experiment, and a good part of
our work was to bring it into operation. One of the tasks was to determine its focal
plane. For this purpose we have used a sufficiently large spectrometer chamber with
nuclear plates placed inside. As a target we used a polystyrene foil and the
spectrometer set at 300. With this arrangement we had 7.6 MeV recoil protons and
12.0 elastically scattered deuterons, which we used to locate the focal plane.
Exposures of the plates were taken at various positions and the focal plane was
calculated using results of measurements.
With this information, we then constructed a smaller spectrometer chamber with one
wall at the position of the focal plane. The focal-plane wall was made of a 6 mg/cm2
Melinex window, to allow for the reaction products to pass through and bombard the
nuclear emulsions, which were placed outside the window. This arrangement
allowed for a significant reduction in the time of collecting the data because there
was no need to open the spectrometer chamber to replace the plates between
measurements at various reaction angles. Furthermore, when placed in vacuum, the
emulsions were often peeling off the plates. Consequently, by placing them outside
the vacuum chamber the danger of such damage was eliminated. The focal plane
wall contained an energy scale to assist in correct placing nuclear emulsions plates
depending on the reaction angle.
Figure 3.4. Target chamber and the magnetic spectrometer. 1. Target chamber. 2. Magnetic
spectrometer. 3. Large spectrometer chamber used to determine the focal plane. After determining
the focal plane, the chamber was replaced by a smaller version, which allowed for the nuclear
emulsions to be placed outside the vacuum. 4. Nuclear emulsions holder.
Before using the spectrometer in the measurements of angular distributions we also
had to determine its solid angle and the defocusing coefficient as a function of the
energy of reaction products. For these measurements we have also used a
polystyrene target.
Figure 3.5. Schematic diagram of the detection system.
Beam alignment and angular scale

Other important introductory work included beam alignment of the accelerated
deuterons to ensure that they were delivered at the centre of the main target, and the
determination of the zero position for the angular scale of the spectrometer. To assist
in the beam alignment we have installed a flexible section in the beam line between
the target chamber and the cyclotron. To determine the zero of the angular scale we
have carried out measurements of Rutherford scattering from a thin Au foil for angles
between –300 and +300.
Determination of the accelerated beam energy
Finally, we have also carried out measurements to determine the energy of the
accelerated deuterons. For this purpose we have used the 27Al(d,d)27Al and
27
Al(d,p)28Al reactions and the reaction angle of 900. This combination was
convenient because the Q – value for the (d,p) reaction leading to the ground state
had a well-known value of 5.498 MeV. The measurements included corrections for
the energy loss in the target. We have found that the energy of accelerated
deuterons was Ed = (12.81 ± 0.065) MeV.
Targets and results of measurements
The targets used in the measurements of angular distributions were 1.76 mg/cm2
27
Al self-supporting foil and 2.43 mg/cm2 SiO2 quartz foil. Examples of the energy
spectra are presented in Figure 3.6. The energy resolution was about 150 keV. This
resolution allowed us to resolve the states at 0.84 and 1.01 MeV excitation energies
in 27Al, which were important in the study of core excited model.
Figure 3.6. Examples of particle spectra for the elastic and inelastic scattering of 12.8 MeV deuterons
from 27Al and SiO2 targets.
Figure 3.7. Angular distributions of the inelastic scattering of 12.8 MeV deuterons from 28Si and 27Al
(the left-hand side of the figure). The lines are to guide the eye. Shown also are the angular
distributions for all three excited states in 28Si (the right-hand side of the figure).
Angular distributions of the differential cross sections for the elastic and inelastic
scattering were measured in steps of 50. Results are presented in Figures 3.7-3.12.
Relative errors of the cross-section measurements consisted of statistical errors,
inaccuracies of scanning of nuclear emulsions (estimated at about 2%) and
inaccuracies caused by errors in angle setting. The combined relative errors were
between about 5% and 9% depending on the intensity of observed groups and the
reaction angle. The absolute values of the differential cross sections are estimated to
be accurate to within about 15%.
Preliminary discussion of data
It may not be completely trivial to observe that all the inelastic scattering angular
distributions, relevant to the study of core excitations (Figure 3.7, left-hand side)
exhibit features, which are consistent with any of the simpler models for direct
interaction, i.e. asymmetry, forward peaking, some oscillatory structure with
periodicity consistent with that for the elastic scattering. This is not a trivial point
since 12.8 MeV energy is not so large an energy that it is obvious that direct
processes will dominate.
The angular distributions are also all very similar, which is consistent with the
interpretation that these levels are excited by a direct reaction in a very similar
fashion.
Bearing the above observations in mind, one may attach considerable importance to
the relative cross sections, as they should be measures of similar nuclear matrix
elements between the ground and excited states. It should be noted that while for
27
Al there is a fair difference in relative magnitudes, there is not an order of
magnitude difference in their values. Thus, it appears that there are no overpowering
selection rules inhibiting any of these transitions. In particular, this bears on the often
repeated but never proved tenant that inelastic scattering strongly favours collective
excitations. We must either say (a) that some of these states have a single particle
character but that single particle excitation is not vastly less than collective excitation
or (b) that collective excitation has been spread among all of the states.
It is interesting to observe that there is some correlation between spin I of the
excited state and the maximum of the differential cross section, except for the 5/2+
state, which will be discussed later.
Theoretical analysis
I have analysed our data using both the diffraction (Blair, 1959, 1961; Blair, Sharp,
and Wilets 1962) model of scattering and the plane wave Born approximation
(PWBA) theory (Butler 1951, 1957; Butler, Austern, and Pearson 1958). These
simple theories give much the same information about nuclear structure and reaction
mechanism as their more complex counterparts. However, their advantage is that
because of their relative simplicity they can be used without computers. At the time
when simple desk calculators or slide rules were as popular as currently used
personal computers, the availability of such simple theories was of considerable
importance.
Our data were also analysed using strong-coupling model (Buck, 1963; Buck,
Stamp, and Hodgson 1963; Chase, Wilets, and Edmonds 1958) and computing
facilities in the USA and UK.
The diffraction model
The description of the diffraction model is presented in the Appendix B. I have
carried out calculations using both a sharp and smooth cut-off boundaries for the
target nuclei. Examples of the results of theoretical analysis are presented in Figure
3.8. Fits to all angular distributions for 27Al are of the same quality as the fit displayed
for the 2.21 MeV level.
Table 3.2 contains the list of the maximum values of the experimental cross sections
used in the normalisation of the theoretical cross sections and the resulting matrix
elements M (see the Appendix B). In this table Ex is the excitation energy; Ιπ – spin
and parity of the excited state; σmax – maximum value of the experimental differential
cross section. Μ is the matrix element related to the deformation distance.
The derived matrix elements were used to calculate model-related parameters δ 2
and hω2 / 2C . The listed values have been calculated using the extreme rotational
and vibrational models, respectively. The β 2 parameters were calculated assuming
the radius Rc = 1.2 A1 / 3 fm.
Figure 3.8. The experimental angular distributions (points) of the elastic and inelastic scattering of
12.8 MeV deuterons from 27Al and 28Si are compared with the calculations using the diffraction
models with a sharp and smooth cut-off radius (full and dashed lines, respectively).
Table 3.2
A summary of the results based on the diffraction model analysis
Nucleus Ex Ιπ σmax Μ δ2 β2 hω2 / 2C2

(MeV) (fm2) (fm2) (fm)
28 +
Si 1.78 2 1.35 0.1463 1.36 0.37 0.028
27 1 +
Al 0.84 /2 0.072 0.0083 1.14 0.32 0.153
3 +
1.01 /2 0.123 0.0140 1.49 0.41 0.130
7 +
2.21 /2 0.213 0.0237 1.94 0.54 0.110
5 +
2.73 /2 0.058 0.0064 1.01 0.28 0.040
9 +
3.00 /2 0.210 0.0230 1.92 0.53 0.086
The plane wave (PWBA) theory

The concept of the plain wave theory for the inelastic scattering, as used in my
calculations, is outlined in the Appendix C. The theory is also mentioned in the
Appendix E. The parameters used in fitting the 27Al and 28Si data are listed in Table
3.3.
As explained in the Appendix C, tables of Lubitz (1957) were prepared to assist in
the interpretation of experimental results using the PWBA theory. However, for the
inelastic scattering of 12.8 MeV deuterons from 27Al and 28Si nulcei, the required R0 -
independent parameter y/x was well outside the values used by Lubitz. The required
values are in the vicinity of 4 whereas the values used by Lubitz are terminated at
y/x = 2.
Table 3.3
Parameters used in the plane wave analysis of the 27Al and 28Si data
Nucleus Ex σmax Bf κf q~ y/ x y S
(MeV) (fm2) (MeV) (fm-1) (fm-1)
27
Al 0.78 0.072 17.05 1.23 0.509 4.88 15.65 0.0168
1.01 0.123 16.82 1.22 0.507 4.88 15.60 0.0287
2.21 0.213 15.62 1.18 0.500 4.86 15.32 0.0495
2.73 0.058 15.10 1.16 0.500 4.83 15.19 0.0135
3.00 0.210 14.83 1.15 0.500 4.81 15.13 0.0488
28
Si 1.78 1.350 10.06 0.94 0.495 3.97 12.39 0.2957
Ex – Excitation energy. σmax – maximum value of the experimental differential cross-section.
B f – The projectile binding energy in the outgoing channel. The projectile binding energy in
the incoming channel, Bi = 17.83 MeV for the 27
Al + d and 11.84 MeV for 28
Si + d systems.
Bi and B f were calculated using binding energies compiled by Mattauch, Thiele, and
Webstra (1965).
κ i= 1.26 fm-1 for 27Al and 1.02 fm-1 for 28Si. (For the definition of κi , κ f and other parameters
used in my calculations see the Appendix C.)
xmax = 3.14 for 27Al and 3.11 for 28Si, resulting in R0 = 6.3 fm for both nuclei.
S – The normalisation factor.
Figure 3.9. The calculated plane wave (PWBA) angular distribution is compared with experimental
results for 27Al and 28Si. To display the data on the same graph, the experimental differential cross
sections for 28Si have been divided by a factor of 6.4. The calculated curve shows that the observed
transitions are associated with the l = 2 angular momentum transfer. The figure shows also close
similarity between the experimental angular distributions for the two target nuclei as predicted by the
core-excited model.
However, for y/x > 2, a linear interpolation can be applied and it gives y ≈ 15 for 27Al
and 12 for 28Si. Unfortunately, these values are again outside the range used by
Lubitz, who tabulates theoretical cross sections for y ≤ 6. Consequently, the tables of
Lubitz could not be used and the necessary function W22 ( x , y ) had to be calculated.
I have calculated this function for a series of y values around 15 for 27Al and 12 for
28
Si and I have found that the position of its prominent maximum xmax = 3.14 for 27Al
and 3.11 for 28Si. Using these values and the values of q~ calculated at the positions
of the experimental maxima I have obtained R0 = 6.3 fm for both target nuclei.
The calculated PWBA angular distribution is compared with experimental results in
Figure 3.9. For the purpose of displaying two experimental distributions on the same
graph, the experimental values for 28Si have been divided by 6.4. Similar fits have
been obtained for the angular distributions to other excited states in 27Al. The
calculations confirmed the l = 2 assignment to the observed transitions.
Strong coupling analysis
The experimental data for the elastic and inelastic scattering (Ex = 1.78 MeV) for 28Si
were analysed using the strong coupling optical model (Buck, 1963; Buck, Stamp,
and Hodgson 1963; Chase, Wilets, and Edmonds 1958) and computing facilities at
Oak Ridge National Laboratory, USA and Oxford University, UK. In this model (see
the Appendix D) the elastic and the first inelastic channel are considered explicitly in
the calculations. The remaining non-elastic channels are taken into account through
an appropriate absorbing potential.
Figure 3.10. The experimental data for the elastic and inelastic scattering of 12.8 MeV deuterons by
28
Si are compared with the strong coupling optical model calculations. The dotted line shows the fit to
all the inelastic data and the full line the fit to the inelastic cross-sections at angles less than 50°. The
total reaction cross-sections and cross-sections for excitation of the 2+ state for these two fits are σ R
= 1357 mb, σ in = 34.6 mb and σ R =1345 mb, σ in = 32.7 mb, respectively. Parameter sets are listed
in Table 3.4.
A simple collective model is used to represent the ground and excited states of the
target nucleus. The rotational and vibrational models give almost identical results for
the 0+ → 2+ excitation and the former was used in the present work. The elastic and
inelastic angular distributions may then be calculated from the central optical
potential together with the nuclear deformation parameter β2.
In our analysis, the undeformed optical potential had the form:
U (r ) = VC (r ) − Vf (r ) − iWg (r )
where VC(r) is the Coulomb potential, taken to be that due to a uniformly charged
sphere of radius Rc = r0 A1 / 3 , V and W are the real and imaginary central potential
depths, f (r ) = [1 + exp(r − R / a )] is the Woods-Saxon radial form factor, where the
−1
nuclear radius R = r0 A1 / 3 and a is the surface diffuseness parameter, and g (r ) is its

normalized radial derivative of f (r ) . Spin-orbit forces were not included in this
calculation.
Figure 3.11. Experimental data for the elastic and inelastic scattering of 12.8 MeV deuterons by 27Al
are compared with the distributions generated by the optical model (dashed line) and strong-coupling
model (full lines). The corresponding reaction cross-section is 1221 mb. Parameter sets are listed in
Table 3.4.
The strong coupling model is in principle applicable to the excitation of any finite
number of collective states but at the time of our calculations the only available
computer program was for the 0+ → 2+ excitation of even nuclei, so only the elastic
and first inelastic distributions for silicon were analysed in this way. To make this
analysis, the parameters of the above potential were systematically adjusted to give
the best fit to the elastic cross-section alone, using a parameter search routine of
Maddison (1962). This potential, together with an approximate value of β2, was then
used in the strong coupling calculation to give the elastic and first inelastic cross-
sections.
The elastic distribution differed from that found from the initial fit to the elastic data
partly due to the effect of the strong coupling and partly to the different value of the
absorbing potential. All the parameters were then iterated to optimise the fit to the
elastic and inelastic data, using the strong coupling search routine of Buck (1963).
At first we have found that the value of β2 increased at each iteration up to quite
unphysical values. The results of one of these calculations are shown in Figure 3.10.
The parameters used in the strong-coupling analysis are listed in Table 3.4.
Table 3.4
Parameters used in the strong-coupling analysis
Nucleus Set V W r0 a β2 δ2
(MeV) (MeV) (fm) (fm) (fm)
28
Si A 93.2 17.59 1.3 0.7 0.866 3.42
B 106.2 11.68 1.3 0.7 0.446 1.74
27
Al C 106.2 11.68 1.3 0.7 0.446 1.74
D 61.4 23.33 1.55 0.53 — —
Set C – The parameter β2 = 0.664 when normalised to the angular distributions for the
2.21 MeV state. The corresponding δ2 = 2.59 fm.
Set D – Conventional optical model analysis.
Figure 3.10 shows that the elastic cross-section is quite well fitted, and also the
shape of the main forward peak of the inelastic cross-section but the calculated
inelastic cross-section falls markedly below the experimental values above 50°. This
failure of the model may possibly be due to the neglect of spin-orbit forces. The
effect of this discrepancy is that the search program pushes up the inelastic cross-
section to give the best overall fit to the data, resulting in an unphysically large value
of β2. The effect is particularly marked as the saturation region is soon reached,
where a large increase of β2 gives only a small increase in the inelastic cross-
section.
To avoid this difficulty, the calculations were continued omitting the inelastic data
above 50°. The iteration then converges to a satisfactory fit to the elastic and
inelastic cross-sections, as shown in Figure 3.10, and gives a reasonable value for
the nuclear deformation β2.
The data for the elastic scattering from 27Al were analysed using conventional optical
model. As the shapes of the experimental angular distributions for the inelastic
scattering for 27Al and 28Si are virtually the same, the appropriately scaled theoretical
distribution for 28Si has been used to compare it with the distributions for 27Al.
Results are presented in Figure 3.11. It can be seen that the shape of the theoretical
angular distribution for the inelastic scattering closely resembles the shapes of the
experimental distributions.
Deformation distance
It should be recognised that it is the deformation distance δlm, which enters the
calculations rather than a dimensionless parameter αlm (or β) defined by
R = R0 (1 + ∑lm α lmYlm* )
More appropriately, the nuclear radius is written as

R = R0 + ∑lm δ lmYlm*
Formally, all this says is that

α lm R 0 ≡ δ lm
There is an important consequence of working with δlm: this parameter is believed to
refer to the deformation of the nuclear matter field, while R0 always includes effects
related to the experiment in question. For example, in the diffraction model analysis,
R0 for scattering of α particles or deuterons on 27Al or 28Si is around 6.5 fm. This
large size comes in part from the finite extent of α particles or deuterons, and from
the fact that the cut-off radius used in the analysis corresponds to a radius well out in
the tail of the nuclear potential.5 On the other hand, for an electron scattering
experiment, R0 would be much smaller (less than around 4 fm), since it refers directly
to the electric charge distribution. For a proton scattering experiment, R0 would lie
between these extremes.
For all these causes, however, we would expect the deformation distance
parameters to be essentially the same since they refer to the deformation of the
nuclear field alone. There might turn out to be some differences between δlm values
derived from α or deuteron scattering experiments and δlm describing charge density,
but it is certainly much more appropriate to compare these quantities rather than αlm
parameters derived using different projectiles. These αlm parameters will differ by a
factor of around 2 even though the deformation is the same.
Rotational model for 27Al

Attempts have been made to apply rotational model to 27Al (Almqvist et al 1960;
Bendt and Eidson 1961; Bishop 1960; Lawergren and Ophel 1962a, 1962b; Towle
and Gilboy, 1962; Vanhuyse and Vanpraet 1963). In this model the first band with K
= 5/2 is built on the ground state, the second and third members being the 7/2+ state at
2.21 MeV and 9/2+ at 3.00 MeV. The 3/2+ state at 1.013 MeV and the 5/2+ state at 2.73
MeV are supposed to belong to the K =1/2 rotational band built on the 1/2+ 0.84 MeV
excited level.
5
In the strong-coupling analysis, a relatively small radius used in the optical model potential was
compensated by large deformation parameters giving generally unphysically large values for the
deformation distance.
However, the rotational model has some difficulty with 27Al. This is unexpected
because this model gives satisfactory description of the neighbouring nuclei 24Mg
and 25Mg. An extension of this model to 27Al would seem a natural one. Some of the
difficulties are as follows:
(1) The deformation distance, δ 2 , as derived from my diffraction analysis for the
5 / 2 → 7 / 2 transition is 1.94 fm if one assumes a permanent deformation.6 This
value is larger than expected in this region, which could mean that rotational model
cannot be applied to 27Al. For instance, similar diffraction model analysis of 24Mg
gives δ 2 = 1.43 fm (Blair 1961).
(2) The deformation distance for this transition disagrees with the deformation
distance calculated from electromagnetic transition.
The spectroscopic quadrupole moment Q is the experimental observable, which in
the case of axially deformed nuclei can be related to the intrinsic quadrupole moment
Q0 and thus to the nuclear charge deformation parameter or the deformation
distance.
3K − I 0 (I 0 + 1)
Q= Q
(I 0 + 1)(2 I 0 + 3) 0
The intrinsic quadrupole moment of a deformed ellipsoidal charge distribution can be
expressed as:
eZR02 β (1 + 0.36 β + ...)

3
Q0 =
5π
Thus
3 3
Q0 ≅ eZR02 β = eZR0δ 2
5π 5π
The measured value for Q is 14 -15 efm2 for 27Al (Stone 2001). The radius for the
nuclear charge distribution can be assumed to be Rc = 1.2 A1 / 3 fm, which for 27Al is 3.6
fm. Using this value and the measured value for Q, we can calculate that the
deformation distance δ 2 determined by electromagnetic transition is between 1.11
and 1.19 fm, whereas the distance determined by the inelastic scattering using
diffraction theory is 1.94 fm.
(3) In the strong coupling model, the only other strongly excited state should be a 9/2+
state located somewhere around 4 MeV or higher. However, what we find is in fact
an equally strongly excited level at 3 MeV. It is clear that the rotational band
sequence is not being followed.
(4) The intensity patterns are all in disagreement with the rotational prescription. The
ratio of 9/2 to 7/2 cross sections should be 1/6 to 10/21, whereas the observed
intensities of inelastic scattering are approximately 1 to 1. Thus, the rotational model
predicts that only two states (9/2 and 7/2) are strongly excited but that of these two
states, the 7/2 MeV level should be most prominent, which is in disagreement with
experimental observation. Furthermore, the collective excitations are zero between
6
As discussed earlier (see also Table 3.4), the analysis based on a more complex strong coupling
theory also had a tendency to produce unrealistically large β2 and δ2 parameters.
states belonging to different bands. They may be excited only through changing the
single particle orbitals. In contrast, experimental results show significant excitation of
states allegedly belonging to the K = 1/2 band.
Core-excited model of 27Al

To understand how the core excited model applies to 27Al we have to look again at
its signatures and compare them with experimental results. The concept of core-
exited model for this nucleus is shown in Figure 3.12.
Figure 3.12. The core-excited model of 27Al.
The spins of core-excited states

The coupling of a 1d5/2 single proton hole with the 2+ state should produce a
quintuplet of states with spins 1/2, 3/2, 5/2, 7/2, and 9/2. As can be seen, 27Al contains
the full complement of these expected states.
The centre of gravity
The core-excited states should be located around the energy of the 2+ state of the
28
Si core. (The centre of gravity energy of the core-excited states should be equal to
the excitation energy of the 2+ state.)
The centre of gravity of the quintuplet in 27Al is 2.32 MeV, which is close to the
excitation energy 1.78 MeV of the 2+ state in 28Si. The higher than expected centre of
gravity energy for the quintuplet can be explained by coupling of its 5/2+ member with
the ground state (see below), which has the same spin and parity. The coupling
repels the 5/2+ excited state and thus pushes it to a higher energy.
Shapes of the angular distributions
All relevant inelastic scattering angular distributions should have the same shape.
(The shapes of the angular distributions for the quintuplet of states and for the 2+
state should be the same.) The remarkable feature of our experimental results is that
indeed all angular distributions display similar features (see Figure 3.7).
All inelastic scattering angular distributions have a prominent forward peak located
approximately at the same scattering angle and all have been fitted using l = 2
angular momentum transfer. Examples of theoretical analysis are shown in Figure
3.8 for the diffraction model, in Figure 3.9 for the plane wave theory and in Figure
3.11 for the strong coupling model. Figure 3.11 shows how the scaled down
theoretical curve for 28Si resembles closely the shapes of the inelastic scattering
angular distributions for 27Al.
The 2I+1 dependence
The absolute values of the differential cross sections for the core-excited quintuplet
should be proportional to 2 I + 1 , where I is the spin of a member of the quintuplet.
Figure 3.13. The integrated experimental cross sections (points) are compared with the expected
linear 2I+1 dependence.
In order to study this signature, the experimental cross sections have been
integrated under the first maximum for each member of the quintuplet. Results of the
integration are presented as a function of 2I+1 in Figure 3.13. It can be seen that with
the exception of one point, which belongs to the 5/2+ excited state, the experimental
points follow closely the linear 2I+1 dependence as expected for the core-excited
model.
The cross section, which is outside the linear 2I+1 dependence, is for the scattering
to the 5/2+ level at 2.73 MeV. Its non-compliance with the rule may be understood as
a consequence of coupling to the ground state, which has the same spin and parity.
The coupling reduces the intensity of the inelastic scattering.
The reduction in the intensity of the scattering to the 5/2+ state may be used to esti-
mate the degree of mixing between this state and the ground state. The wave
functions of the two states can be written as (Vervier 1963):
5 5 5 5 5
= A 0 + 1− A2 2
2 1 2 2 2 2
5 5 5 5 5
= A 2 − 1− A2 0
2 2 2 2 2 2
5 5
where and are the wave functions of the ground state and the excited state,
2 1 2 2
28
the quantum numbers J c jI refer to the total angular momentum of the Si core,
27
the d5/2 hole, and the Al, respectively, and A is the mixing parameter.
The coupling reduces the differential cross section to the 5/2+ member of the
quintuplet by a factor of 0.38 ± 0.06, which corresponds to A2 = 0.72 ± 0.03.
Absolute values of the measured cross-sections
As mentioned in the Introduction, the measured cross sections for a core-excited
member of a multiplet should be related to the cross section for the 2+ state of the
core. The sum of measured cross sections for the multiplet should be equal to the
cross section for the excitation of the core.
Table 3.5 lists the measured and calculated cross sections. The measured cross
sections were obtained by integrating experimental results over the first maxima of
the inelastic scattering angular distributions. The cross sections listed in the last
column were calculated using the formula mentioned in the Introduction:
⎡ dσ ⎤ (2 I + 1) ⎡ dσ ⎤
⎢⎣ dΩ ⎥⎦ = (2 I + 1) ⎢⎣ dΩ ⎥⎦
I ∑ J
Jc
Table 3.5
The measured and calculated cross sections for the inelastic scattering of 12.8 MeV deuterons from
27
Al and 28Si nuclei
As can be seen, the measured cross sections agree reasonably well with the
calculated values. The notable exception is again for the 5/2+ excited state in 27Al,
which as mentioned earlier, can be at least partly explained by coupling to the
ground state. When correction is made for the lost intensity, the measured cross
section increases from the listed value of 1.4 mb to 2.3 mb. The sum of the cross
sections for 27Al is then 14.2 mb, which is close to the expected 18.1 mb. The
difference is caused mainly by the lower than expected excitation of the 9/2+state.
The reason for slightly lower cross section for the 9/2+ state is unclear but it might be
due to sharing its intensity with more complex core-coupling schemes, such as
coupling of the proton hole with the 4+ excited state in 28Si. Simple models are hardly
ever expected to result in a perfect representation of nuclear structure and it is
already remarkable that the simple core-coupling model gives such a good
description of the observed experimental features.

Angular distributions of the differential cross sections for the elastic and inelastic
scattering of 12.8 MeV deuterons from 27Al and 28Si have been measured over a
wide range of angles. Experimental results for the inelastic scattering display clear
features of direct reaction mechanism.
Results of measurements were analysed using a wide range of theories: the
diffraction models, the plane wave Born approximation theory, the optical model and
the strong coupling model. As shown by Rost and Austern (1960) the diffraction
models are a good approximation of more complex distorted wave descriptions of
inelastic scattering. The carried out calculations confirmed that the observed angular
distributions can be described assuming a direct reaction mechanism. They have
also resulted in useful structure-related parameters.
A step-by-step comparison of experimental results with the core-excited model
shows that the low-lying states in 27Al can be described by assuming a coupling a
1d5/2 proton hole with the 2+ state of the 28Si core.
References
Almqvist, E., Bromley, D. A., Gove, H. A. and Litherland, A. E. 1960, Nucl. Phys. 19:1.
Bent, R. D. and Eidson, W. W. 1961, Phys. Rev. 122:1514.
Bishop, G. R. 1960, Nucl. Phys. 14:376.
Blair, J. S. 1959, Phys. Rev. 115:928.
Blair, J. S. 1961, private communication.
Blair, J. S., Sharp, D., and Wilets, L. 1962, Phys. Rev. 125:1625.
Buck, B. 1963, Phys. Rev. 130:712.
Buck, B., Stamp, A. P. and Hodgson, P. E. 1963, Phil. Mag. 8:1805.
Butler, S. T. 1951, Proc. Roy. Soc. (London) A208:559.
Butler, S. T., Austern, N., and Pearson, C. 1958, Phys. Rev. 112:1227.
Chase, D. M., Wilets, L. and Edmonds, A. R. 1958, Phys. Rev. 110:1080.
De Shalit, A. 1961, Phys. Rev. 122:1530.
Lawergren, B. T. and Ophel, T. R. 1962a, Phys. Lett. 2:265.
Lawergren, B. T. and Ophel, T. R. 1962b, Proc. Phys. Soc. 79:881.
Lawson, R. D., and Uretsky, J. L. 1957, Phys. Rev. 108:1300.
Lubitz, C. R. 1957, Numerical Table of Butler-Born Approximation Stripping Cross
Sections, Randall Laboratory of Physics, University of Michigan, Ann Arbor.
Maddison, R. N. 1962, Proc. Phys. Soc. 79:264.
Mattauch, J. H. E., Thiele, W., and Webstra, A. H. 1965, Nucl. Phys. 67:1.
Rost, E. and Austern, N. 1960, Phys. Rev. 120:1375.
Stone, N. J. 2001, Tables of Nuclear Magnetic Dipole and Electric Quadrupole
Moments, Oxford Physics, Clarendon Laboratory, Oxford, UK.
Towle, J. H. and Gilboy, W. B. 1962, Nucl. Phys. 39:300.
Vanhuyse, V. J. and Vanpraet, G. J. 1963, Nucl. Phys. 45:603.
Vervier, J. 1963, Nuovo Cim. 28:1412.
4
40
Ca(d,d), (d,d’), and (d,p) Reactions with 12.8 MeV Deuterons
Key features:
1. We have measured angular distributions of the differential cross sections for the
elastic scattering of 12.8 MeV deuterons from 40Ca nuclei using a broad range
magnetic spectrometer. We have also measured the distributions for the inelastic
scattering to the states 3.35 MeV (0+), 3.74 MeV (3-) and 4.49 MeV (5-) in 40Ca, and
for the deuteron stripping reaction to the ground state (7/2-), 1.94 MeV (3/2-), and 2.46
MeV (3/2-) levels of 41Ca
2. Optical model analysis of the elastic scattering employed central potential with
surface absorption. We have found that by including spin-orbit interaction we can
reproduce the diffraction anomaly discussed in Chapter 2.
3. Inelastic scattering angular distributions were analysed using distorted wave Born
approximation and deformed optical model potential. We have found that the
deformation of both the real and imaginary components was necessary to generate
physically meaningful deformation parameters. Our deformation (and deformation
distance) parameters are in good agreement with the values based on a study of 55
MeV proton scattering (Yagi at al. 1964).
4. Analysis of the stripping reaction was also carried out using the distorted wave Born
approximation. We have carried out a detailed study of the local, nonlocal, finite range
and zero range approximations. Final results included nonlocal and finite-range
effects. The derived spectroscopic factors are lower than theoretically expected and
possible reasons for the discrepancy are discussed.
Abstract: Differential cross sections were measured for the 40Ca(d,d), (d,d’), and (d,p)
reactions induced by 12.8 MeV deuterons. Inelastic scattering to the 3.35 MeV (0+) , 3.74 MeV
(3-) and 4.49 MeV (5-) levels, and (d,p) reactions to the ground state (7/2-), 1.94 MeV (3/2-), and
2.46 MeV (3/2-) levels of 41Ca were studied. Optical model analysis of the elastic scattering
was carried out, and potentials obtained were used in the distorted wave analysis of the
inelastic scattering and stripping reactions. Spin-orbit interaction was necessary to fit the
elastic scattering data. The collective model gave a good fit to the inelastic excitations of the
3- and 5- levels. To obtain realistic deformation parameters it was necessary to use a complex
form of nuclear interaction. The effects of finite range and nonlocality were included in the
deuteron stripping analysis and produced reasonable agreement with the observed cross
sections. The spectroscopic factors extracted when the spin-orbit coupling is included in the
deuteron optical model potential are significantly smaller than expected.
Introduction
The direct nuclear reactions have proved to be useful in studies of nuclear structure.
While inelastic scattering of particles yield parameters characterizing the collective
modes of nuclear excitations, the single-particle aspects of nuclear internal motion
can be studied by stripping reactions. Thus inelastic scattering and transfer reactions
give valuable information about collective and single-particle excitations. Theoretical
analysis of these reactions gives also useful insights into the details of the
mechanism of nuclear interactions.
For our study, we have chosen 40Ca as a target nucleus because of its double magic
properties. Both neutron and proton shells are closed, which means that, in its
ground state, 40Ca should can be described as a spherical nucleus. The excited
states of 41Ca should be then described by a simple single-particle model with a
stripped neutron moving around in a spherical potential of a well bound, doubly-
closed core.
In the work described here we have studied elastic and inelastic scattering of 12.8
MeV deuterons and single-neutron stripping reaction. Theoretical analysis of the
data was carried out using optical model and distorted wave Born approximation.

Procedure
Experimental data were obtained using a beam of 12.8-MeV deuterons from the 120-
cm cyclotron of the Institute of Nuclear Physics in Cracow. A broad-range
spectrometer (see Chapter 3) with photographic emulsions as detectors has been
used for the measurements of angular distributions. The spectra of particles emitted
in the reactions were taken in steps of 5° in the angular range of 10° to 110° in the
laboratory system.
The integrated beam passing through the 40Ca target was collected in the Faraday
cup placed behind the target and measured by means of a standard current
integrator. Beam monitoring was also done by using a thin gold foil placed in the
beam at the entrance to the scattering chamber. An additional detector was used to
monitor changes in the target thickness.
The 40Ca targets with a thickness of 2.28 mg/cm2 were prepared by rolling it from
metallic samples, which were placed immediately in the vacuum in order to reduce
the oxygen content.
Figure 4.1. The low-lying states in 40Ca and 41

Ca. States for which angular distributions have been
measured are marked with the asterisks.
We have measured inelastic scattering angular distributions for the states at 3.35
MeV (0+), 3.74 MeV (3-), and 4.49 MeV (5-). We have also repeated measurements
of elastic scattering (see Chapter 2) using a 5.4 mg/cm2 target. Proton angular
distributions for the deuteron stripping reaction 40Ca(d,p)41Ca were measured for
transitions to the ground state and two excited states (1.94 and 2.46 MeV) of 41Ca.
The low-lying energy levels for 40Ca and 41Ca are shown in Figure 4.1. States
excited with substantial intensity and for which angular distributions have been
measured are marked with the asterisks.
Results
The measured differential cross sections for the elastic and inelastic scattering,
together with theoretical curves, are shown in Figures 4.2, 4.4, and 4.6. No
significant excitation of the 3.90 MeV level of 40Ca nucleus by inelastic scattering
was observed.
The relative errors included the statistical errors as well as inaccuracies in the angle
setting and in scanning of the emulsions. The combined relative errors were around
6%. The error in the absolute values of the cross sections was estimated at 15%. It
was associated with inaccuracies in the beam intensity measurements and in the
determination of the target thickness.
The energy resolution of the detecting system (150 keV) assured good separation of
the detected groups of particles, except for the group of protons leading to the 1.94
MeV state of the 41Ca nucleus where a small admixture from the very weak level at
2.01 MeV remained unresolved.
The two sets of data for the elastic scattering shown in Figure 4.2 are for two
different target thicknesses (4.6 and 5.4 mg/cm2) and two different experimental
arrangements. The 4.6 mg/cm2 data were obtained using a counter telescope (see
Chapter 2). The 5.4 mg/cm2 data represent the results of repeated measurements
with the magnetic spectrometer. The differences between the two sets of
measurements are also believed to be associated with slightly different incident
energies used in both experiments. As can be seen in Figure 4.2, the two sets of
measurements agree well for most angles but diverge for angles larger than around
1200 (c.m.) where differential cross sections are expected to depend strongly on the
incident deuteron energies (cf Chapter 2).
Optical-Model Analysis of the Elastic Scattering
For the purpose of the present analysis, the two available sets of elastic-scattering
data were combined and treated as one. The optical potential used assumes a
surface-peaked absorption and has the form:
U (r ) = U c (r ) − Vf (r ) − WD g (r ) − Vs h(r )L ⋅ s
where Uc(r) is the Coulomb potential from a uniformly charged sphere with the radius
rcA1/3,
2
1 ⎛ d ⎞ 1 ⎛ h ⎞ 1⎛ d ⎞ 1
f (r ) = x g (r ) = 4i⎜ ⎟ x ' h(r ) = ⎜⎜ ⎟⎟ ⎜ ⎟ x
e +1 ⎝ dx' ⎠ e + 1 ⎝ mπ c ⎠ r ⎝ dx ⎠ e + 1
r − r0 A1 / 3 r − r0 ' A1 / 3
x= x' =
a a'
We have assumed rc=1.3 fm for the Coulomb potential and we have carried out
parameter search for V around 100 MeV. We have carried out three sets of
calculations, which resulted in three sets of parameters, which are listed in Table 4.1.
The corresponding theoretical calculations are compared with experimental data in
Figure 4.2.
Table 4.1
40
Optical model parameters describing the Ca(d,d) 40Ca elastic scattering at 12.8 MeV deuteron
energy
Set V r0 a WD r0′ a' Vs χ2

(MeV) (fm) (fm) (MeV) (fm) (fm) (MeV)
1 111.8 1.000 0.850 15.30 1.431 0.613 0.00 7.54
2 131.3 0.840 1.013 16.55 1.509 0.598 0.00 6.70
3 91.6 1.164 0.722 8.88 1.374 0.773 9.60 2.96
The χ value measures the quality of fits to the data. It shows that the best fit is for the set 3,
2
which includes the spin-orbit interaction.
40 40
Figure 4.2. Optical model calculations for the Ca(d,d) Ca elastic scattering at 12.8 MeV deuteron
energy for three sets of parameters (see Table 4.1) are compared with experimental data. Open
circles represent experimental results obtained using a 4.6 mg/cm2 target and a counter telescope as
described in Chapter 2. Closed circles are for the current measurements with the 5.4 mg/cm2 target
and the magnetic spectrometer. The two sets of measurements were combined and treated as one in
the automatic search code employed in our optical model analysis.
There are known to be considerable ambiguities in the choice of optical potentials for
deuterons (Halbert 1964; Perey and Perey 1963) and these have been examined in
some detail for 40Ca + d at other energies (Bassel et al. 1964). However, there is
evidence from analysis of (d,p) reactions that the potential required is one with a real
well depth V ~ 100 MeV, and attention was restricted to this potential in the present
work.
Previous studies (Bassel et al. 1964; Halbert 1964; Perey and Perey 1963) of
deuteron scattering from 40Ca have indicated that the imaginary potential extends to
significantly larger radii than the real potential ( r0′ ≈ 1.5 fm while r0 ≈ 1.0 fm). These
earlier results were used as a starting point for the present searches.
The first calculations were made without spin-orbit coupling, i.e. with Vs = 0. It was
known from earlier works that the least well determined parameters are r0 and a, so
initially these parameters were fixed at the value r0 = 1.0 fm and a = 0.85 fm
previously found for the 12 MeV data (Bassel et al. 1964). The optimum values for
the other parameters are given in Table 4.1, denoted as set 1. When r0 and a are
also allowed to vary, set 2 results; the reduction in χ 2 is only about 10%. The
corresponding predicted cross sections are compared to experiment in Figure 4.2.
A considerable improvement was obtained when spin-orbit coupling was introduced,
i.e. when Vs ≠ 0 was assumed. The optimum χ 2 was then reduced by more than a
factor of 2. The improvement is particularly evident at the maximum around 60°. The
parameters are given in Table 4.1 as set 3, and the comparison of theory with
experiment is included in Figure 4.2. The main effect of the introduction of spin-orbit
coupling upon the optimum values of the other parameters is to reduce considerably
the strength of the absorptive potential.
Inelastic Scattering
Calculations were made for the inelastic scattering to the 3.35 MeV (0+, l = 0), 3.74
MeV (3-, l = 3), and 4.49-MeV (5-, l = 5) states in 40Ca, using the collective-model
interaction and the distorted-wave approximation. Both the model and the method
have been discussed in detail elsewhere (Bassel et al. 1962). Suffice it to say that
the inelastic angular distributions are determined once the optical-model parameters
have been found by fitting the elastic scattering. The one adjustable parameter is
then the deformation β l (or equivalently the deformation distance δ l defined as
δ l = β l r0 A1 / 3 ), which is chosen to reproduce the magnitude of the measured cross
section. The predicted cross sections are proportional to β l2 .
All three optical model potentials described in the previous section were used.
However, since the computer code was unable to include spin-orbit effects for spin-1
particles in both channels, set. 3 was used with Vs = 0 .
A subsidiary calculation with potential set 3, but treating the deuteron as though it
had spin 1/2, showed a slight filling in of the minima in the angular distribution. It is
expected that a correct spin-1 calculation would show a similar small effect.
Figure 4.3. The dependence of the calculated angular distributions on whether only real or both real
and imaginary components of the optical model were assumed to be deformed (real or complex
coupling). Potential set 2 of Table 4.1 was used.
Table 4.2
The deformation β l and deformation distance δ l parameters for the low-lying states in 40Ca
l=0 l=3 l=5 l=0 l=3 l=5
Potential (3.35 MeV) (3.74 MeV) (4.49 MeV) (3.35 MeV) (3.74 MeV) (4.49 MeV)
set β0 β3 β5 δ0 δ3 δ5
2 0.08 0.32 0.16 0.23 0.92 0.46
3 0.05 0.30 0.13 0.20 1.19 0.52
The energies of excited states are shown in the parentheses. The deformation distance
parameters δ l are in fm.
Figure 4.4. Angular distributions for the inelastic scattering of 12.8 MeV deuterons by 40Ca.
Theoretical calculations are for the potential sets 2 and 3 as defined by parameters listed in Table 4.1.
The corresponding deformation parameters are listed in Table 4.2.
Two versions of the model were used, one in which only the real part of the optical
potential was deformed (the real coupling labelled as "real") and one in which both
real and imaginary parts were deformed with the same deformation (the complex
coupling labelled as "complex"). Figure 4.3 compares their predictions, for the
potential set 2.
It is seen that the complex coupling produces a much larger cross section (for a
given β l value) than does real coupling. As a result, when normalising to the
experimental distributions, calculations with real coupling would require unacceptably
large values of β l (0.704 for l = 3 and 0.384 for l = 5). This effect has been noticed
before for deuteron scattering (Dickens, Perey, and Satchler 1965)
Figure 4.4 compares the theoretical calculations with the experimental data using
complex coupling for potential sets 2 and 3. The β l and δ l values are listed in Table
4.2. Excitation of the 3- and 5- levels by 55-MeV protons (Yagi et al. 1964) yielded
deformation parameters (and deformation distances), which are in good agreement
with the values derived here. The application of the model to the 0+ excited state is
perhaps questionable. It might be regarded as a monopole ("breathing mode")
vibration, but one would expect the volume integral of the potential well to be
conserved approximately during the oscillations, and this would introduce a volume
interaction term additional to the surface coupling used here. There is only qualitative
agreement between the measurements and the l = 0 predictions shown in Figure
4.4. The transition is quite weak, as evidenced by the small values of β 0 ~ 0.05 or
0.08 required.
Potential sets 1 and 2 give very similar predictions. The angular distributions from
potential set 3 are not very different, but the cross-section magnitudes are
significantly larger. This result is due to the smaller absorptive strength of this
potential (see Table 4.1).
Deuteron Stripping
The application of the distorted-wave method to the 40Ca(d,p)41Ca reaction has been
described and discussed in detail by Lee at al. (1964). For the analysis of the
present data, the deuteron optical potentials described earlier for the elastic
scattering were used, but the proton optical potential and the shell-model potential
into which the neutron is captured were taken to be the same as used by Lee at al.
(1964). Spin-orbit coupling was included for both the neutron and proton. Corrections
for finite range of the n-p interaction were made in the local energy approximation
(Buttle and Goldfarb 1964; Perey and Saxon 1964), which is known to be very
accurate for this reaction (Austern1965; Perey 1963). Although the optical potentials
used were local, the damping of the wave functions in the nuclear interior, which
would arise from using equivalent nonlocal potentials (Hjorth, Saladin, and Satchler
1965) was also calculated in the local energy approximation (Buttle and Goldfarb
1964; Hjorth, Saladin, and Satchler 1965; Perey and Saxon 1964).
The nonlocality range was taken to be β =0.85 fm for the nucleons and β = 0.54 fm
for the deuteron (see the Appendix E). These values assume that all the observed
energy dependence of the optical potentials is due to nonlocality rather than any
intrinsic energy variation; hence, they give an upper limit to these effects.
Nonlocality reduces the contributions to the reaction amplitude from the nuclear
interior by 30-40%, but it also increases the magnitude of the tail of the neutron
bound-state wave function by 13% (for the 1f state) or 11% (for the 2p states). Finite
range then produces an additional 10-20% reduction in the contributions from the
interior.
As a result, the predicted differential cross-section curves with nonlocality and finite
range fall between those calculated in zero-range approximation with local potentials
with and without a radial cutoff in the nuclear surface, which eliminates all the interior
contributions. This is illustrated in Figure 4.5 for the l = 1 and 3 transitions.
Figure 4.5. Theoretical predictions for the l = 1 and l = 3 stripping comparing the effects of nonlocality
(NL) and finite-range (FR) with those from the use of a radial cutoff in a local (L) or zero-range (ZR)
calculation. Potential set 3 was used. The L/ZR calculations with a cutoff radius eliminate completely
the contribution from the interior of the nucleus. The remaining assumptions (L/ZR or NL/FR) only
reduce these contributions.
Calculations were made for all three deuteron potentials of Table 4.1, including the
spin-orbit coupling of set. 3. The results for potential sets 1 and 2 were almost
identical. It may be seen that the calculations with a non-zero cutoff radius predict
significantly lower differential cross sections then the calculations with no cutoff
radius. Thus the non-zero cutoff radius calculations lead to significantly larger values
of the spectroscopic factors.
The experimental data are compared with theoretical calculations in Figure 4.6 for
the potential sets 2 and 3, with nonlocal and finite-range effects included. The fits to
the angular distributions are as good as those obtained at other energies (Hjorth,
Saladin, and Satchler 1965; Lee et al. 1964). In particular, the discrepancy remains
between theory and experiment for the shape of the second maximum for the l = 3
(1f7/2) transition.
The spectroscopic factors obtained are given in Table 4.3. These values are not
directly comparable to those obtained by Lee et al. (1964), because nonlocality
corrections were not included in that work. The effect on the tail of the neutron wave
function alone would reduce all the spectroscopic factors of Lee et al. (1964) by 22%
(for 2p transitions) or 26% (for 1f transitions), but these reductions are partly
compensated for by the additional damping of the contributions from the nuclear
interior.
The spectroscopic factors for the set 3, which resulted in the best fit to the elastic
scattering angular distribution, are significantly smaller than the values expected for
a closed-shell target. (One would expect the ground-state transition to have the
value of unity and the two p3/2 transitions to sum to unity.) A similar discrepancy for
the f7/2 state was noted at deuteron energy of 14.5 MeV (Hjorth, Saladin, and
Satchler 1965).
Figure 4.6. Comparison of the measured cross sections for the 40Ca(d,p)41Ca stripping reaction with
the distorted-wave predictions using potential sets 2 and 3. Nonlocality and finite-range effects are
included, and no cutoff is used.
Table 4.3
Spectroscopic factors for the reaction 40Ca(d,p)41Ca
(Nonlocal and finite range effects are included)
Spectroscopic factors
Ex Configuration Pot. set 2 Pot. set 3
g.s. f7/2 0.95 0.65
1.94 p3/2 0.44 0.34
2.46 p3/2 0.22 0.17
We have measured angular distributions for the elastic and inelastic scattering for
the low-lying states of 40Ca using 12.8 MeV deuterons and a broad range magnetic
spectrometer. We have also measured distributions for the deuteron stripping
reaction leading to low-lying states in 41Ca.
We have analysed elastic scattering angular distribution using optical model potential
with surface absorption. We have found that by adding spin-orbit interaction we can
reproduce the anomaly described in Chapter 2.
To analyse angular distributions for the inelastic scattering we have used deformed
optical model potential with the same parameters as used for the elastic scattering.
We have found that deformation of both the real and imaginary components was
necessary. If the deformation is assumed to be only for the real component, the
resulting deformation parameters are unrealistically high. If both components are
assumed to be deformed, the deformation parameters determined by our analysis
are in good agreement with parameters obtained from a study of 55 MeV proton
scattering (Yagi et al. 1964).
The analysis of the angular distributions for the deuteron stripping reactions was
carried out using distorted wave Born approximation theory. We have studied the
effects of local, nonlocal, zero range and final range approximations. We have also
carried out calculations using a cutoff radius and compared them with other
calculations.
We have found that a cutoff radius lowered significantly the absolute values of the
differential cross sections but had little effect on the shape of the angular
distributions. In the final calculations, we have included both the nonlocal and finite
range effects.
We have found that the derived spectroscopic factors are lower than the theoretically
expected values. The spectroscopic factor for the ground state should be 1 but is
0.95 if set 2 of optical model parameters is used or 0.65 for the set 2. The sum of the
spectroscopic factors for the p3/2 transfers should be also 1 but is 0.66 for the set 2 of
the optical model potential or 0.51 for the set 3.
The conclusions to be drawn about the spectroscopic factors are not clear. The
results may mean that 40Ca does not have very pure closed-shell structure, or they
may merely reflect a poor choice of parameters for the neutron potential well (Lee at
al. 1964). Another uncertainty concerns the nonlocality of the neutron potential well;
we have no direct evidence on this, and we have seen that using a local potential
would increase the spectroscopic factors by approximately 20%.
References
Austern, N. 1965, Phys. Rev. 137:B752.
Bassel, R. H., Drisko, R. M., Satchler, G. R., Lee, L. L., Schiffer, J. P., and Zeidman, B.
1964, Phys. Rev., 136:B960.
Bassel, R. H., Satchler, G. R., Drisko, R. M., and Rost, E. 1962, Phys. Rev., 128:2693.
Buttle, P. J. A. and Goldfarb, L. B. J. 1964, Proc. Phys. Soc. (London) A83:701.
Dickens, J. K., Perey, F. G., and Satchler, G. R. 1965, Nucl. Phys., 73:529.
Halbert, E. C. 1964, Nucl. Phys. 51:353.
Hjorth, S. A., Saladin, J. X., and G. R. Satchler 1965, Phys. Rev. 138:B1425.
Lee, L. L., Schiffer, J. P., Zeidman, B., Satchler, G. R., Drisko, R. M., and Bassel, R. H.
1964, Phys. Rev., 136:B971.
Perey, F. G. 1963, in Proceedings of the Conference on Direct Interactions and Nuclear
Reaction Mechanisms, Padua, 1962, edited by E. Clement and C. Villi, Gordon
and Breach Science Publishers, Inc., New York.
Perey, C. M. and Perey, F. G. 1963, Phys. Rev. 132:755.
Perey, F. G. and Saxon, D. 1964, Phys. Letters 10:107.
Yagi, K., Ejiri, H., Furukawa, M., Ishizaki, Y., Koike, M., Matsuda, K., Nakajima, Y.,
Nonaka, I., Saji, Y., Tanaka, E., and Satchler G. R. 1964, Phys. Letters 10:186.
5
Spectroscopic Applicability of the (3He,α) Reactions
Key features:
1. The aim of this work was to study the applicability of (3He,α) neutron pickup reactions in
nuclear spectroscopy
2. I have measured differential cross sections for the pickup reaction 26Mg(3He,α)25Mg leading to
low-energy states in 25Mg.
3. I have carried out theoretical analysis of experimental results using direct reaction theory and
computer codes, which I have modified and adapted to run at the Australian National
University.
4. Nuclear interaction in both incoming and outgoing channels is described using central optical
model potential with volume absorption.
5. The study shows a clear preference for deep potentials Vτ ≈ 150 MeV (τ = 3He) and
Vα ≈ 200 MeV
6. Shapes of the experimental angular distributions can be well reproduced by using fixed sets
of optical model parameters with the exception of Vα, which should be adjusted around its
discrete values. This procedure results also in extracting reliable relative ratios of the
spectroscopic factors. However, to extract their absolute values it is necessary to keep the
optical model parameters as closely as possible to the values determined from analysis of
elastic data.
7. Spectroscopic factors determined in this study are in excellent agreement with calculations
based on the assumption of rotational model for 25Mg with coupling between rotation and
particle motions.
8. In general, when using (3He,α) reactions, only relative values of the spectroscopic factors are
determined. Finding their absolute values is hindered by significant uncertainties about the
2
zero-range coefficient D0 for this reaction. Using theoretical values for the spectroscopic
factors, I have found that D0 = 16.4 × 10 MeV ·fm for these reactions, which is in excellent
2 4 2 3
agreement with the theoretical value of 17.0 × 10 MeV2·fm3 calculated using Irving-Gunn
4
wave function. This factor can be used to extract absolute values of spectroscopic factors.
9. The single neutron pickup reaction (3He,α) can serve as a useful spectroscopic tool.
However, my study resulted in formulating certain recommendations, which are summarized
in the last section of this Chapter.
Abstract: Differential cross sections for the reaction 26Mg(3He,α)25Mg were measured using 10.2 MeV
3
He particles. They were then analysed using distorted wave Born approximation (DWBA). The zero-
range coefficient D0 for (3He,α) reactions has been determined and has been found to be in good
2
agreement with the theoretical value calculated using Irving-Gunn wave function. This coefficient can
be used to extract absolute values of spectroscopic factors from (3He,α) angular distributions for other
target nuclei. The derived spectroscopic factors have been compared with the spectroscopic factors
calculated using rotational model for 25Mg with and without the coupling of the rotation and particle
motions. Excellent agreement has been obtained between experimentally determined spectroscopic
factors and model calculations. My results show that low-energy states in 25Mg can be described
using rotational model with the coupling between the rotation and particle motion.
Introduction
At the time of this study, my two PhD students were busy with their respective
projects. The work described in this Chapter was an extra task I have undertaken to
investigate the applicability of (3He,α) reactions in nuclear spectroscopy. To this end,
it was necessary to understand the dependence of the extracted spectroscopic
factors on the optical model parameters employed in fitting the reaction angular
distributions. In particular my aim was to see whether this reaction could be used to
determine not only the relative but also absolute values of the spectroscopic factors.
I have chosen 26Mg because for this target nucleus the (3He,α) reaction leads to a
well-known deformed nucleus 25Mg for which spectroscopic factors are known with a
high level of confidence. This should allow for a reliable comparison of the absolute
values of the experimental and theoretical spectroscopic factors, which in turn should
help in a reliable determination of the notorious zero-range coefficient D02 , for which a
wide range of values was previously used.
Experimental arrangement and results

Measurements of (3He,α) angular distributions were carried out using 0.2μA 3He++
beam from the Australian National University tandem accelerator. The experimental
setup is shown in Figure 5.1. The beam from the accelerator entered a 51 cm
reaction chamber through a system of four collimator slits and after passing the 26Mg
target was stopped in a Faraday cup placed at the opposite end of the chamber. The
target consisted of 103 ± 5 μg/cm2 enriched magnesium containing 98.22% of 26Mg
evaporated on approximately 10μg/cm2 carbon backing.7 The thickness of the target
was determined by measuring Coulomb scattering of 4 and 5 MeV α particles at
small angles and comparing them with the Rutherford scattering cross sections.
Figure 5.1. Schematic diagram of the experimental setup.
7
The target was supplied by GTJ Arnison, Special Techniques Group, N69 AWRE, Aldermaston,
Berks, England. In addition to 26Mg, it also contained 0.71% of 25Mg and 0.07% 24Mg. Other impurities
were mainly Ca (0.5%) and Na (0.1%).
The intensity of the incident beam was measured using a standard, Elcor model
A309A, current indicator and integrator connected to the Faraday cup, and it was
also determined independently by detecting elastically scattered 3He particles from a
thin Au foil located at the entrance to the target chamber. Variations in the effective
target thickness were recorded using a target monitor, which registered 3He particles
scattered elastically at a fixed angle of approximately 300 from the 26Mg target.
Reaction products were detected using an array of ORTEC surface barrier
semiconductor detectors. ORTEC Model 260 Time Pickoff and Model 262 Inspector
units were employed to reduce pileup background for angles below 300. Pulses from
movable detectors were fed to the RIDL pulse height analyser and the particle
spectra were punched on computer cards. The overall resolution for the particle
spectra was around 100keV.
Angular distributions of α particles corresponding to the ground state and the 0.584
MeV (Jπ =1/2+), 0.976 MeV (3/2+), 1.611 MeV (7/2+), and 1.962 MeV (5/2+) excited
states in 25Mg were measured in the range of angles between 100 and 1650 (lab).
They are shown in Figure 5.4. The relative errors of the measured cross sections
include the statistical and current integration inaccuracies. The errors are typically
around ±3%. The uncertainties in the absolute values of the cross sections were
estimated at ±6%.
I have carried out theoretical analysis of the data using the optical model code JIB3
(Perey 1963) and the distorted wave Born approximation (DWBA) code DRC (Gibbs
at al. 1964), both of which I have modified and adapted to run on the ANU IBM
360/50 mainframe computer. One of the modifications I have introduced to the DRC
code was to allow for a six-parameter search of the optical model potential both in
the input and output channels. I have also added plotting subroutines to allow for a
quick and easy evaluation of theoretical results.
Optical model potentials for the 26Mg(3He,α)25Mg reaction
The parameters of the optical model potentials in the incident (3He+26Mg)
channel were determined by measuring and analyzing 3He elastic scattering
at incident energies around 10 MeV.
The energy of α particles produced in the investigated 26Mg(3He,α)25Mg
reaction was between 22 MeV for the transition to the ground state and 19 MeV for
the 1.962 MeV excited state. To determine the parameters for the α + 25Mg channel,
I analysed the data of Budzanowski et al. (1964) for the elastic scattering of 24.7
MeV α particles from Mg and Al nuclei.
The optical potential had the following form:
U (r ) = VC (r ) − Vf ( x) − iWg ( x′)
where VC (r ) is the Coulomb potential generated by a uniformly charged sphere,
f ( x) = (1 + e x ) −1 g ( x′) = (1 + e x′ ) −1
r − r0 A1 / 3 r − r0′A1 / 3
x= x′ =
a a′
The resulting sets of parameters for 3He and α particles in the incoming and
outgoing channels, respectively, are listed in Table 5.1.
Table 5.1
Optical model parameters for 3He and α particles a)
a
) Symbol τ is used for 3He.
b
) Subscript i is τ for 3He or α for α particles.
c
) Numbers in italics represent fixed parameters.
The distorter wave analysis

I have carried out the distorted wave Born approximation (DWBA) analysis of the
transfer reaction data assuming that the pickup neutron moved in the spherical
Woods-Saxon potential, V (r ) = Vn f (r ) with the radius R n = rn A1 / 3 . I have used rn = 1.2
fm and an = 0.65 fm. The depth Vn was adjusted to give the binding energy equal to
the separation energy for the relevant states in 25Mg.
In the preliminary analysis I have studied how the shapes and absolute values of
26
Mg(3He,α)25Mg angular distributions depend on the lower cut-off radius and on the
choice of optical model parameters. A sample of the results of the calculations is
shown in Figure 5.2 for the 1.962 MeV excited state.
This preliminary analysis indicated that deep potentials (with Vτ ≈ 150 MeV and Vα ≈
200 MeV for 3He and α particles, respectively) give nearly identical distributions with
and without the lower cut-off radius, in both the shape and absolute values. They
also resemble most closely the experimental distributions (see Figure 5.4).
When fitting transfer reaction cross sections, the parameters derived from elastic
scattering are often adjusted to optimise the fits. To understand this process and to
see how the adjusted parameters relate to the original parameters determined from
analysis of elastic scattering data, I have calculated and examined the χ2 function:
2
1 N ⎡ σ (θ ) − σ th (θi ) ⎤
χ = ∑ ⎢ exp i
2
⎥
N i =1 ⎣⎢ Δσ exp (θ i ) ⎦⎥
where N is the number of the experimental data points, σexp(θi) is the experimental
differential cross section measured at the angle θi, σth(θi) is the corresponding
theoretical cross section, and Δσexp(θi) is the error in σexp(θi). The function is
multidimensional in the space of the optical model parameters. An example of a two-
dimensional χ2 map is shown in Figure 5.3.
26 3 25
Figure 5.2. The dependence of the Mg( He,α) Mg angular distributions on the choice of optical
model parameters in the input and output channels and on the lower cut-off radius Rc.o. . The
distributions were calculated using three sets of optical potentials for 3He (marked here as τ) and α
particles. The parameters (V, W, r0 , r0′ , a and a’) used in each case are shown in the figure. For the
deep potentials, the shapes and the absolute values of the calculated cross sections do not depend
strongly on the value of the lower cut-off radius. The calculated shapes resemble also more closely
the experimental angular distribution (cf Figure 5.4).
Figure 5.3. An example of a two-dimensional map of the χ2 function for the 26Mg(3He,α)25Mg reaction
25 3
leading to the 0.976 MeV state in Mg. The map was constructed using the set 2 for He particles
(see Table 5.1). Parameters optimising the fits to the reaction cross sections are shown as closed
circles. The parameter sets 4-6 (Table 5.1) obtained by fitting elastic (α,α) scattering are shown as
asterisks. The figure shows that the parameters optimising the fits to the reaction distributions are
close to the parameters based on the analysis of elastic scattering. It also shows that the fits to the
angular distributions can be optimised using a constant value for r0 and adjusting only the potential
depth Vα. However, this procedure leads to a strong dependence of the extracted spectroscopic
factors on Vα.
This example shows that optical model parameters determined by fitting elastic
scattering (indicated using the asterisks) are close to parameters that optimize the
fits to reaction differential cross sections (shown as full circles). Thus, the
parameters determined using elastic scattering can be only adjusted slightly to
optimize the fits to the reaction cross sections.
An alternative way of optimizing the fits to reaction cross sections is to keep
geometrical parameters fixed (such as r0 shown in this example of the χ2 map) and to
adjust only the depth of the optical model potential. However, it will be shown later,
that this procedure leads to an undesirably strong dependence of the absolute
values of the differential cross sections on the choice of the optical model
parameters. Thus, it is better to keep the parameters as closely as possible to the
values determined by the analysis of elastic scattering.
An example of the DWBA calculations for the 26Mg(3He,α)25Mg reaction is shown in
Figure 5.4. In these calculations I used parameter set 2 (see Table 5.1) for 3He
particles and set 5 and 6 for α particles but adjusting the depth Vα to optimise the
fits. As expected, only small adjustments were necessary. The figure shows that the
angular distribution for 1.611 MeV state cannot be described by using direct neutron
pickup reaction.
Figure 5.4. Angular distributions for the reaction 26Mg(3He,α)25Mg are compared with the distorted
wave calculations. Parameter set 2 was used for 3He particles (see Table 5.1). For α particles,
parameter sets 5 and 6 were used but the depth Vα, was adjusted to optimise the fits. The adjusted
values are shown for each calculated distribution.
As mentioned in the Introduction, the aim of the DWBA analysis was not only to
reproduce the shapes of angular distributions and thus to determine the transferred
angular momenta but also to extract absolute values of the spectroscopic factors and
compare them with model calculations.
Theoretical angular distribution can be expressed as:
σ th (θ ) = (2 J + 1) D02 S (l , j )σ lj (θ )
where J is the spin of the final state, and S (l , j ) is the spectroscopic factor, l and j are
orbital momentum and spin of the transferred particle.
The angle-dependent part of the theoretical cross sections, σ lj (θ ) , gives information
about the reaction mechanism. It also gives spectroscopic information in the form of
spins and parities of states in the final nucleus.
Spectroscopic factors S (l , j ) give additional information about nuclear structure.
Experimental values of spectroscopic factors are extracted by comparing the
absolute values of the experimental and theoretical cross sections at forward angles,
that is, at angles where the reaction is dominated by direct transfer process.
It is clear that by comparing the absolute values of the calculated and experimental
cross sections at forward angles, only the products P = D02 S (l , j ) can be determined.
These products can be used to calculate the ratios of the spectroscopic factors,
κ = S (l , j ) / S g .s. (l , j ) , where S (l , j ) is the spectroscopic factor for a given excited state
and S g .s. (l , j ) is the spectroscopic factor for the ground state.
The ratios of the spectroscopic factors give useful and often satisfactory information
about nuclear structure but better information is obtained if absolute values of
spectroscopic factors can be extracted.
Relative values of spectroscopic factors
The derived here relative values of the spectroscopic factors are compared with
results obtained by other authors in Table 5.2.
As shown in Figure 5.4, angular distribution for the 1.611 MeV state in 25Mg cannot
be described well using direct reaction mechanism. Consequently, the experimental
ratios of the spectroscopic factors for this state can be ignored.
Table 5.2
Ratios of the spectroscopic factors for the neutron pickup reactions from 26Mg
a
) κ exp are the experimentally determined ratios of the spectroscopic factors S (l , j ) / S g .s. (l , j ) .
b
) Reynolds (1966)
c
) Dehnhard and Yntema (1967)
d
) My results for the 26Mg(3He,α) 25Mg reaction.
e
) κ th(1) and κ th( 2) are the theoretical ratios of the spectroscopic factors calculated using a simple
rotational model and rotational model with coupling between the rotation and particle motion
(Kerman 1956 and Davidson 1965).
Table 5.2 shows a good agreement between the experimental ratios for the 0.584
MeV state. My result for the 1.962 MeV state agree with the results obtained using
(p,d) and (d,t) reactions but disagrees with the result of Dehnhard and Yntema
(1967) who used (3He,α) reaction at 33 MeV. A marked difference between my
results and results obtained by other authors is for the 0.976 MeV state. However all
my results, including the 0.976 MeV state, are in excellent agreement with model
calculations based on rotational model for 25Mg with coupling between the rotation
and particle motion (Kerman 1956 and Davidson 1965).
The normalization coefficient P = D02 S (l , j )
It is always desirable, but not always possible, to determine the absolute values of
spectroscopic factors. However, to find the absolute values one has to know the
numerical value of the zero-range coefficient D02 . Unfortunately, there is a
considerable uncertainty about this coefficient for (3He,α) reactions.
Normally, when fitting reaction angular distributions, optical model parameters
derived from analysis of elastic scattering are used, with a possibly of only minor
adjustments. To understand the problem with determining the D02 coefficient I have
studied the dependence of not only the shapes but also of the absolute values of the
calculated differential cross sections on the optical model parameters. I have found
that both the shapes of the calculated angular distributions can be reproduced well
by varying only one parameter, the depth Vα. However, to calculate reliably not only
the shapes but also the absolute values of the differential cross sections the
parameters should be adjusted as closely as possible around the values determined
from analysis of elastic data.
Results of this preferable procedure are shown in Table 5.3, which lists the values of
the normalization coefficient P = D02 S (l , j ) and the ratios of the spectroscopic factors
κ = S (l , j ) / S g .s. (l , j ) . Optical model parameters for 3He particles are represented by
set 2 in Table 5.1. The parameters for α particles are represented by sets 4, 5, and
6, respectively. However, the potential depth in sets 4 and 6 have been adjusted
slightly to optimise the fits to the reaction angular distributions.
Table 5.3
The parameters P = D S (l , j ) and
2
0 κ = S (l , j ) / S g .s. (l , j ) for the 26Mg(3He,α)25Mg reaction
E Set 4 Set 5 Set 6 Average a)

(MeV) Vα = 160 MeV V α = 220 MeV V α = 300 MeV
r0 = 1.41 fm r0 = 1.34 fm r0 = 1.27 fm
P κ P κ P κ P κ
0.000 36.57 1.000 41.56 1.000 34.91 1.000 37.68 1.000
0.584 1.91 0.052 2.93 0.070 2.69 0.077 2.51 0.066
0.976 4.36 0.119 4.57 0.110 3.62 0.104 4.18 0.111
1.611 1.63 0.045 1.68 0.040 1.41 0.040 1.57 0.042
1.962 3.19 0.087 3.91 0.094 3.16 0.090 3.42 0.090
a 4 2 3
) The last two columns give the average values of P and κ. The listed factors P are in 10 MeV ·fm .
As mentioned earlier, experimental distributions could be fitted well using fixed
parameters for all three sets but adjusting only the depth Vα. This is a significant
departure from the recommended procedure and while it reproduces the shapes of
the distributions it leads to significant dependence of the calculated absolute values
of the differential cross sections on the discrete values of Vα. For instance, the
normalization coefficient P ≡ D02 S (l , j ) for the ground state is 29.9, 44.9, and 54.9 for
Vα = 160, 220, and 270 MeV respectively. However, this type of the calculations
does not alter significantly the relative values of the calculated cross sections.
Thus, the calculations show that as far as the ratios of the spectroscopic factors are
concerned it does not matter which procedure is used. Referring to the example
presented in Figure 5.3, it does not matter whether parameters are adjusted along
the diagonal line and close to the values obtained from elastic scattering marked by
asterisks or along a vertical line for a fixed value or r0. However, if parameters are
kept close to the parameters determined from elastic scattering, the values of
P = D02 S (l , j ) are significantly less dependent on the choice of optical model
parameters, which means that one should be able to determine the absolute values
of the spectroscopic factors with a better accuracy.
Determining the D02 coefficient
The advantage of using the 26Mg(3He,α)25Mg reaction is that model calculations for
25
Mg can be treated with high degree of confidence. The determined average value
of the normalization coefficient P shown in Table 5.3 can then be used to extract the
experimental value of the zero-range coefficient D02 , which in turn can be applied to
other (3He,α) reactions to extract not only the ratios but also the absolute values of
the spectroscopic factors.
P
D02 =
Sth (l , j )
where P is the experimentally determined normalization coefficient and Sth (l , j ) is the
model-calculated spectroscopic factor.
Results of my calculations are compiled in Table 5.4. This table includes not only the
calculations of D02 based on my experimental results but also on the data of
Dehnhard and Yntema (1967) at 33 MeV 3He energy.
In the case of the 10.2 MeV data, the D02 coefficients were calculated by using the
experimentally determined normalization coefficient P = 37.7 × 104 MeV2·fm3 and the
theoretical spectroscopic factors for the ground state in 25Mg. This procedure gives
then the experimental spectroscopic factors for the excited states, which are
compared with the theoretical values.
I have carried out calculations using model-calculated spectroscopic factors
Sth(1) (l , j ) based on a simple rotational model, and Sth( 2 ) (l , j ) based on a model with
coupling between the rotation and particle motion (Kerman 1956 and Davidson
1965).
Table 5.4
Comparison of the experimental spectroscopic factors S (l , j ) with model calculations and
2
determination of the zero-range D0 coefficients
a) Sth(1) (l , j ) and Sth( 2) (l , j ) are the theoretical spectroscopic factors calculated using a
simple rotational model and a model with coupling between the rotation and particle
motion (Kerman 1956 and Davidson 1965), respectively.
b) The D02 coefficients and the corresponding spectroscopic factors Sexp (l , j )

calculated using the experimental data of Dehnhard and Yntema (1967) and the
relevant theoretical spectroscopic factors for the ground state.
c) The D02 coefficients and the corresponding spectroscopic factors Sexp (l , j )

calculated using my experimental data and the relevant theoretical spectroscopic
2 4 2 3
factors for the ground state. The listed D0 values are in units of 10 MeV ·fm . Excellent
agreement is between my experimental results and the theoretical values
of Sth
( 2)
(l , j ) and thus the experimentally determined D02 = 16.4 × 104 MeV2·fm3.
As can be seen by comparing theoretical and experimental values of the
spectroscopic factors my results are in excellent agreement with the rotational model
for 25Mg with coupling between rotation and particle motion. The experimentally
derived D02 = 16.4 × 104 MeV2·fm3 is in good agreement with the theoretical value of
17 × 104 MeV2·fm3 calculated using Irvin-Gunn wave functions (Bassel and Drisko,
1967).
Conclusions
I have measured angular distributions for the 26Mg(3He,α)25Mg reaction leading to
the ground state and first four excited states in 25Mg. The aim was to study the
spectroscopic applicability of (3He,α) reactions. I have selected 26Mg as a target
nucleus because theoretical values of the spectroscopic factors for states in 25Mg are
relatively well known and thus they can be used to determine the elusive D02 factor
for (3He,α) reactions.
Experimental data were analysed using distorted wave theory. With the exception of
the transition to the 1.611 MeV state in 25Mg, all measured angular distributions can
be described well by direct reaction mechanism over a wide range of angles in the
forward direction. Conclusions of this study can be summarised as follows:
1. Spectroscopic factors extracted in the present study are in excellent
agreement with theoretical values calculated using rotational model for low-
energy states in 25Mg with coupling between the rotation and particle motions
(Kerman 1956 and Davidson 1965).
2. The extracted zero-range coefficient for (3He,α) reactions, D02 = 16.4 × 104
MeV2·fm3, is in good agreement with the theoretical value of 17 × 104 MeV2·fm3
calculated using Irvin-Gunn wave functions (Bassel and Drisko, 1967).
3. The single neutron pickup reaction (3He,α) can serve as a useful
spectroscopic tool. However, my study suggests the following
recommendations.
a. Deep optical model potentials with Vτ >100 MeV for 3He particles (τ =
3
He) and Vα >150 MeV for α particles should be used. Deep potentials
give better fits to the angular distributions. They also allow to minimise
the dependence of the shape of the distributions and the absolute
values of the calculated cross sections on the choice of the lower cut-
off radius Rc.o. . In fact, the lower cut-off radius is unnecessary if deep
potentials are used. In addition, deep potentials allow for a better
estimation of spectroscopic factors.
b. The shapes of the calculated angular distributions do not depend
strongly on the optical model parameters for 3He particles. These
parameters can be kept fixed at the values determined from elastic
scattering. Parameters in the exit channel can be adjusted to optimise
the fits but they should be kept as close as possible to the values
determined by fitting elastic scattering angular distributions.
c. The accuracy of the extracted absolute values of the spectroscopic
factors is improved if analysis is carried out for various sets of the
optical model potential and the resulting spectroscopic factors are
averaged.
d. To extract the absolute values of the spectroscopic factors it is
recommended to use the experimentally determined normalization
coefficient D02 = 16.4 × 104 MeV2·fm3. Alternatively, the theoretical value
D02 = 17 × 104 MeV2·fm3 can be used.
References
Bassel, R. H. and Drisko, R. M. 1967, Proc. Symp. On Direct Nuclear Reactions with
3
He, September, Tokyo (I.P.C.R. Cyclotron Report, Supplement 1) p. 13.
Budzanowski, A., Grotowski, K., Micek, S., Niewodniczanski, H., Sliz, J., Strzalkowski,
A., and Wojciechowski, H. 1964, INP Report No. 347; Phys. Lett. 11:74.
Davidson, J. P. 1965, Rev. Mod. Phys. 37:105.
Dehnhard, D. and Yntema, J. L. 1967, Phys. Rev. 160:964.
Gibbs, W. R. Madsen V. A., Miller, J. A., Tobocman, W., Cox, E. C. and Mowry, L.
1964, NASA TN D-2170.
Glendenning, N. K. 2004, Direct Nuclear Reactions, World Scientific Publishing Co. Pte.
Ltd., Singapore.
Hamburger, E. W. and Blair, A. G. 1960, Phys. Rev. 119:777.
Kerman, A. K. 1956, Mat. Fys. Medd. Dan. Vid. Selsk. 30, No. 15
Nilsson, S. G. 1955, Mat. Fys. Medd. Dan. Vid. Selsk. 29, No 16.
Perey, F. G. 1963, Phys. Rev. 131:745.
Reynolds, G. M. 1966, Ph.D. thesis, Univ. of Minnesota, Minneapolis.
Satchler, G. R. 1958, Ann. of Phys. 3:275.
Satchler, G. R. 1983, Direct Nuclear Reactions, Oxford University Press, Oxford.
Tobocman, W. 1961, Theory of Direct Nuclear Reactions, Oxford University Press,
Oxford.
6
The Discrete Radius Ambiguity of the Optical Model
Potential
Key features:
1. This study revealed a hitherto unknown discrete radius ambiguity. It explained why some
analyses resulted in an unusually large radius for the imaginary part of the optical model
potential.
2. I have carried out theoretical analysis of elastic scattering of 3He particles from 16On nuclei
using optical model formalism and a computer code, which I have modified and adapted to
run on an ANU mainframe computer.
Abstract: Angular distributions for the elastic scattering of 3He on 16O measured at six incident
energies between 9.802 and 10.720 MeV were averaged to yield a distribution at 10.25 MeV. The
averaged distribution was then analysed using optical model potential. Calculations were carried out
using surface or volume absorption, four or six parameters for the central part of the optical potential,
and with or without spin-orbit interaction. Detailed investigation of the χ2 function has led to a
discovery of a hitherto unknown discrete radius ambiguity for the real part of the central components
of the optical model.
Introduction
Previous analyses (Yntema, Zeidman, and Bassel 1964; Bray and Nurzynski 1965;
Bray, Nurzynski, and Bourke 1968) of elastic scattering of 3He particles
demonstrated that low-energy data can be fitted satisfactorily using six-parameter,
volume absorption potential with standard geometry. However, a study of
16
O(3He,3He)16O elastic scattering indicated that this geometry is probably
inapplicable in the case of the target nuclei with atomic mass A ≤ 16. Some other
calculations (Erskine, et al. 1965; Fortune et al. 1968; Kellogg and Zurmuhle 1966;
Weller, Robertson and Tilley 1968) for carbon and oxygen targets suggest that an
unusually large radius r0′ >2 has to be used to fit the experimental data. I have,
therefore, decided to look more closely at the 3He + 16O scattering to understand why
a different sets of geometrical parameters are apparently needed for such light
nuclei.
Experiment
Doubly charged 3He particles accelerated to required energies in the Australian
National University tandem accelerator were used to bombard an oxygen gas target.
Angular distributions for the elastic scattering of 3He on 16O were measured at
bombarding energies 9.802, 9.955, 10.210, 10.313, 10.465, 10.720, 11.229, and
11.738 MeV. The quoted energies were calculated at the centre of the oxygen gas
cell. The measurements were carried out in 5o steps between 15o and 165o (lab) 8.
Results of measurements are presented in Figure 6.1. The absolute values of the
differential cross sections are accurate to about ±2.7%. The various component
errors are shown in Table 6.1. The relative errors of the experimental points vary
between 1% and 5%.
8
lab stand for the laboratory system.
Table 6.1
Components of the absolute error
Figure 6.1. The experimental angular distributions for the elastic scattering of 3He particles on 16O.
The lines have been drawn to guide the eye.
Discussion of the experimental results

At the first glance, the overall shape of the elastic angular distributions appears to be
similar at all energies. The most striking features are the large backward angle
increase in the cross section and a large maximum at 125o (c.m.)9, which are
common to all distributions. The third maximum at about 85o appears somewhat
anomalous in that its size fluctuates with energy. At 9.955 MeV it comes close to the
second maximum but is separated at four higher energies. However, at 11.229 and
11.738 MeV the second maximum has almost disappeared.
Figure 6.2 shows the integrated cross sections as a function of energy. The
integration over each angular distribution has been performed between 15o and 165o
(lab). The experimental points follow roughly the E-2 dependence of the Rutherford
scattering but they also show some marked deviations from the calculated curve.
9
c.m. stand for the centre of mass system
Figure 6.2. The integrated (total) cross sections for the elastic scattering of 3He particles from 16
O
nuclei as a function of incident energy. The full line represents the E-2 dependence.
Optical model analysis

Optical potential
The optical model potential employed in this analysis had the following form:
U (r ) = −Vf (r ) − iWk g k (r ) + VS .O. f S .O. (r )s ⋅ l + VC (r )
where
{ [
f (r ) = 1 + exp (r − r0 A1 / 3 ) / a ]}
−1
Index k is S for the volume (Saxon-Woods) absorption or D for the surface

(derivative) absorption.
{ [
g S (r ) = 1 + exp (r − r0′A1 / 3 ) / a′ ]} −1
d
g D ( r ) = 4 a′ g S (r )
dr
[ ]}
2
⎛ h ⎞ 2 d
f S .O. (r ) = ⎜ ⎟ {
1 + exp (r − r0 s A1 / 3 ) / as
−1
⎝ Mc ⎠ r dr
Ze 2 ⎛ r2 ⎞
VC (r ) = ⎜ 3 − ⎟ for r ≤ rC A1 / 3
rC A1 / 3 ⎜⎝ rC2 A1 / 3 ⎟⎠
2Z 2
VC (r ) = for r > rC A1 / 3
r
In my calculations I used rC = 1.4 fm.
Analysis
The experimental angular distribution used in the analysis was obtained by
averaging measurements at six incident energies between 9.801 and 10.720 MeV.
This is a close cluster of experimental distributions and they show similar angular
pattern.
The computation was carried out using an automatic search code (Perey 1963),
which I have modified and adapted to run on the Australian National University IBM-
360/50 computer.
I have carried out extensive calculations using surface or volume absorption, central
potential with four or six parameter, and by including or excluding the spin orbit
interaction. The parameters were determined by minimising the function
2
1 N ⎡ σ (θ ) − σ th (θi ) ⎤
χ = ∑ ⎢ exp i
2
⎥
N i =1 ⎣⎢ Δσ exp (θ i ) ⎦⎥
where N is the number of the experimental data points, σexp(θi) is the experimental
differential cross section measured at the angle θi, σth(θi) is the corresponding
theoretical cross section, and Δσexp(θi) is the error in σexp(θi).
Results of the optical model analysis
A few examples of fits to the experimental angular distribution are shown in Figure
6.3. The figure has been prepared by copying directly from the computer plots
prepared using my plotting routine, which I have added to the optical model code.
The outputs show that the fits do not depend strongly on the number of optical model
parameters and on whether volume or surface absorption is used, at least for
calculations at forward angles.
Figure 6.3. Examples of the optical model calculations for the elastic scattering of 3He particles from
16
O at 10.25 MeV incident energy. The parameters were optimised around V = 150 MeV. In the
signature n1(n2)k, n1 is the number of optical model parameters (both fixed and variable), n2 the
number of fixed parameters, and k is S for the volume absorption potential or D for the surface
absorption.
The discrete radius ambiguity

To understand the role of the optical model parameters I have carried out a detailed
mapping of the χ2 function. This function is multidimensional in the space of optical
model parameters and as such cannot be presented in a graphic form for all
parameters simultaneously. However, the function can be easily studied by plotting
two-dimensional projections. The plots can be further simplified by showing only the
points corresponding to the minimum values of the χ2 function. An example of such
plots is presented in Figure 6.4. The figure shows clearly the presence of the hitherto
unknown discrete radius ambiguity.
Figure 6.4. An example of discrete radius r0 ambiguities. These results show that it is possible to
obtain practically equivalent fits to the elastic scattering data (compare the minimum values of the
χ 2 function) using discrete values of the radius r0 parameter for the real part of the central
component of the optical model potential.
The figure presents various families of parameter sets that optimise the fits to the
elastic scattering of 3He from 16O nuclei. The families are identified using letters A, B,
C, D, and E.
For instance, in the family B, any set of parameters selected for a given r0 along the
lines indicated by the full circles will fit the data well and will correspond to χ2 = 3.5.
Parameters in the family C will produce nearly identical fits (the minimum of χ2
function is only slightly lower) but their sets of parameters corresponds to an entirely
different range of r0 parameters. Thus, for instance, if V = 150 MeV one can fit the
data using any of the two discrete values of r0 , either 1.21 fm or 1.60 fm for the
volume absorption potential. For the surface absorption potential one can have either
r0 = 1.15 or 1.23 fm.
This study explains also why some searches have led to an unusually large radius
r0′ for the imaginary component of the central part of the optical model potential.
Figure 6.4 shows that this large value belongs to the family B. For the family C the
parameter has lower values r0′ . Depending on the initial values of the optical model,
the automatic search will lead to one of the two equivalent minima in χ2 function.

This work describes a detailed study of the optical model for the elastic scattering of
3
He particles from 16O nuclei at 10.25 MeV. The analysis of experimental angular
distribution was carried out using optical model with six or four parameters for the
central part, volume or surface absorption, and with or without spin-orbit interaction.
Excellent fits were obtained over a large range of angles of up to about 1350 in the
centre of mass system.
I have found that angular distributions were insensitive to spin-orbit interactions.
Likewise, experimental distributions were reproduced well using either surface or
volume absorption. Finally, the shapes of the calculated distributions did not depend
strongly on whether six or four parameters were used for the central part of the
optical model potential.
Detailed analysis of the χ2 functions has revealed a hitherto unknown and therefore
unexpected discrete radius ambiguity for the real component of the central part of
the optical potential. Discrete ambiguities in the depth V of the real component of the
central part of the optical potential are well known. However, the discrete radius
ambiguities have not been previously reported. The demonstrated here discrete
radius ambiguity explains why in certain analyses automatic searches lead to
unusually large value of the imaginary component of the central part of the optical
model potential.
References
Bray, K. H. and Nurzynski, J. 1965, Nucl. Phys. 67:417.
Bray, K. H., Nurzynski, J. and Bourke, W. P. 1968, Nucl. Phys. A114:309.
Erskine, J. R., Holland, R. E., Lawson, R. D., MacFarlane, M. H. and Schiffer, J. P.
1965, Phys. Rev. Lett. 14:915.
Fortune, H. T., Gray, T. J., Trost, W. and Fketcher, N. R. 1968, Phys. Rev. 173:1002.
Kellogg, E. M. and Zurmuhle, R. W. 1966, Phys. Rev. 152:890.
Perey, F. G. 1963, Phys. Rev. 131:745.
Weller, H. R., Robertson, N. R. and Tilley, D. R. 1968, Nucl. Phys. A122:529.
Yntema, J. L., Zeidman, B. and Bassel, R. H. 1964, Phys. Lett. 11:302.
7
The Al( He,α) Al Reaction at 10 MeV 3He Energy
27 3 26
Key features:
1. We have carried out the measurements of angular distributions for the 27Al(3He,α)26Al
reaction, and my early version of a computer code for unfolding Gaussian distributions was
successfully applied to analyse the particle spectra.
2. Experimental data were analysed using distorted wave Born approximation theory.
3. Spectroscopic factors extracted in this study show that the low-energy states in 26Al can be
-2
described using a simple jj coupling scheme that consists of (1d5/2) configuration only. Such
+ +
a configuration gives states with spins ranging from 0 to 5 .
4. All distributions were fitted using l = 2 angular momentum. Contrary to the results for the (p,d)
and (d,t) reactions, no evidence for l = 0 was obtained, which is in agreement with the
conclusion of Blair and Wagner (1962) that when more than one l value are allowed, the
(3He,α) reactions show preference for higher values.
Abstract: Angular for 27Al(3He,α)26Al reaction have been measured at 10 MeV incident 3He energy.
They were analysed using distorted wave Born approximation theory. Spectroscopic factors extracted
in this study show that the low-energy states can be described using a simple jj coupling of two
nucleons in 1d5/2 configuration.
Introduction
The 1d-2s shell nuclei have been the subject of considerable experimental and
theoretical interest, and for many of them the strong coupling Nilsson model (Nilsson
1955) has been successfully applied. The model works particularly well at the
beginning of this shell. However, it has been pointed out (Gove 1960; Bar-Touv and
Kelson 1965) that the deformation changes abruptly from prolate to oblate in the
vicinity of the mass number A = 27, and consequently, as discussed in Chapter 3,
the strong coupling description seems to be inadequate in this region. Indeed, as
shown earlier, the low-energy states of the 27Al nucleus can be described well
assuming the excited-core model. The target and the residual nucleus involved in the
reaction discussed in this study are both in this interesting region of the periodic
table.
Angular distributions of α particles from 27Al(3He,α)26Al reaction have been
measured previously (Taylor et al. 1960) at an incident energy of 5.2 MeV. States
below the 2.07/2.08 MeV doublet were studied and the data were found to exhibit l =
2 angular momentum transfers only. The excitation of 1.76 MeV state was
interpreted as a direct transition in disagreement with the results discussed here and
with the results for the (p,d) reaction (Bevington and Anderson 1966).
Measurements for the analogous neutron pickup reaction (d,t) induced by 19 MeV
deuterons have been also reported (Vlasov et al. 1960). These authors measured
angular distributions between about 50 and 300 only and analysed them using a
simple plane-wave theory. The group corresponding to the third excited state
exhibited an l = 0 angular momentum transfer. On the basis of the extracted reduced
widths for this state, the authors concluded that the s-wave admixture in the ground
state of 27Al must be small. Their conclusion was later confirmed by the (p,d) data
(Bevington and Anderson 1966).
Experimental details
The experimental setup was similar to that used in the 26Mg(3He,α)25Mg
measurements (see Chapter 5). Target thickness was 170 ± 6 μg/cm2. Angular
distributions for the reaction 27Al(3He, α)26Al were measured at 10 MeV 3He energy
between 50 and 1650 (lab) in steps of 50. Figure 7.1 shows a typical 27Al(3He,α)26Al
spectrum.
As can be seen, peaks corresponding to the first and second excited states were
only partly resolved. To extract differential cross sections for these states I have
written a program for unfolding Gaussian distributions. The program was optimising
the fits to particle spectra by searching for both the widths and positions of the peaks
around the positions calculated using nuclear kinematics. Spectral analysis was
carried out using an IBM 1620 computer.
Figure 7.1. Typical spectrum of α particles taken at θ = 600 (lab) for the reaction 27Al(3He, α)26Mg at 10
3
MeV of He energy. The continuous line was calculated using my program for unfolding Gaussian
distributions and an IBM 1620 computer.
Angular distributions of α particles were measured for the ground state and for
0.229, 0.418, 1.059, 1.760/1.852, and 2.07/2.08 MeV states in 26Al. Apart from a few
small angles, the yield for the 1.852 MeV state was not sufficiently high to extract
angular distribution. Angular distributions are shown in Figure 7.3. The relative errors
were generally smaller than the size of the displayed data points. The absolute error
was mainly due to uncertainties in the measurements of the target thickness and
was estimated at about ±4%.
Angular distributions shown in Figure 7.3 were analysed using the DRC code (Gibbs
at al. 1964) and the ANU IBM 360/50 computer.
Optical model parameters in the incident, 3He + 27Al, channel were obtained from an
earlier work (Bray, Nurzynski, and Satchler 1965) of 3He scattering at energies
between 5.5 and 10 MeV. A fit to the data at 10 MeV 3He energy is shown in Figure
7.2. The corresponding optical model parameters are listed in Table 7.1.
A simple optical model potential was used containing only the central part:
U (r ) = VC (r ) − Vf (r ) − iWg (r )
where VC(r) is the potential due to a uniformly charged sphere of radius RC =rCA1/3 , V
and W are the real and imaginary well depths, and the form factors f(r) and g(r) are
of the Woods-Saxon type:
{ [
f (r ) = 1 + exp (r − r0 A1 / 3 ) / a ]}
g (r ) = {1 + exp[(r − r ′A
0
1/ 3
) / a′]}
The parameters for the exit, α + 26Al, channel were suggested by the analysis of
elastic scattering by McFadden and Satchler (1966).
Figure 7.2. The experimental (points) and calculated (line) angular distributions for the elastic
scattering of 10 MeV 3He particles from 27Al. The optical model parameters are listed in Table 7.1.
Table 7.1
3
Optical model parameters for He and α particles used in the distorted waves analysis of the
27
Al(3He,α)26Al angular distributions
Theoretical calculations for 27Al(3He,α)26Al angular distributions are compared with

experimental data in Figure 7.3. All the theoretical distributions were fitted assuming
a direct pickup of a neutron with the orbital angular momentum l = 2 moving in a
Woods-Saxon potential, whose depth was adjusted to give the correct binding
energy. The calculations were carried out either with a lower cut-off radius of 1.7 fm
or with full integration. As can be seen in Figure 7.3, calculations without a cut-off
radius produce good enough fits to the angular distributions to allow for extracting
the spectroscopic factors.
The possible l = 0 transition to the 0.418 MeV state was also tried but was found to
be in gross disagreement with the observations. The calculated l = 0 and l = 2 were
also mixed in various proportions but no improvement to the l = 2 fit was obtained.
Figure 7.3.Angular distributions of α particles from the 27Al(3He,α)26Al reaction are compared with
theoretical calculations. The curves were calculated using the distorted waves theory of direct nuclear
reactions. The full curves correspond to a lower cut-off radius of 1.7 fm. Other curves were calculated
without a cut-off radius.
Discussion
The available information about the spins and parities of the low-energy states in 26Al
are summarised in Table 7.2. The energies quoted are those of Endt and van der
Leun (1962). All the spins and parity assignments are from the work of Horvat et al.
(1963).
Two alternative descriptions of the low-energy states in 26Al have been proposed. In
one of them, the strong coupling Nilsson model was applied (Kelson 1964; Picard
and de Pinho 1966; Pyatov 1963; Varshalovich and Peker 1961) whereas the other
used the shell model (Bouten, Elliott, and Pullen 1967; Brennan and Bernstein 1960;
Ferguson 1964). However, the strong coupling model does not seem to be adequate
for 27Al (Ophel and Lawergren 1964.) In fact, the low-lying states in 27Al are
described remarkably well using core-excited model (see Chapter 3).
Assuming the jj coupling for the odd neutron and proton in 26Al, the most likely
relationship is that they are both in the 1d5/2 configuration coupling to the total spin J
= 0+, 1+, 2+, 3+, 4+, or 5+. There is also a possibility that any member of the unpaired
neutron and proton is excited to 2s1/2 or 1d3/2 orbits and that the resulting wave
functions can mix with the (1d5/2)-2 configuration giving the total spin J = 0+, 1+, 2+, 3+,
or 4+. It should be expected, however, that the nucleus in the ground state, coupled
to Jπ = 5+ is in the pure (1d5/2)-2 configuration.
Table 7.2
Available information on the low-energy states in the 26Al nucleus
a
) Taylor at al. 1960
b
) The ratios of the differential cross-sections at the first maximum
c
) Bevington and Anderson 1966
d
) Vlasov, et al. 1960
e
) Ratios of spectroscopic factors calculated from the reduced widths assuming that
the ratio S/Sg.s. for the state 0.229 MeV state is the same as that measured in the
(p,d) reaction listed in column 7.
f
) Our data for 10 MeV 3He projectiles. The last column contains theoretical
spectroscopic factors calculated using a formula of Macfarlane and French (1960).
In the simplest case, therefore, when the configuration mixing is neglected, the
spectroscopic factors for the single-nucleon transfer reaction can be calculated
readily using an explicit formula of Macfarlane and French (1960). The results for the
transition from (1d5/2)-1 to (1d5/2)-2 between 27Al and 26Al are shown in the last column
of Table 7.2. It has been assumed that the total strength of the (1d5/2)-2 wave function
for J = 3 is in the second excited state.
For all states, except the 0.229 MeV state, the agreement with experiment is
exceptionally good. The 0.229 MeV state has a relative spectroscopic factor about
twice as large as calculated using the jj coupling. The data from other experiments
included in the table show similar results. The close agreement between experiment
and the simple theory in the case of the 0.418 MeV and the 2.07/2.08 MeV states
suggests that the assumption that the 0.418 MeV state contains nearly all of the Jπ =
3+, (1d5/2)-2 configuration is substantially correct.
The major difference between our data and the (p,d) and (d,t) experimental results is
that the reported l = 0 transitions for the 0.418 MeV and the 2.07/2.08 MeV states
observed in these reactions were not confirmed by the neutron pickup induced by
3
He particles. This differences may be a further evidence of the effect, first observed
by Blair and Wanger (1962), that when two values of the angular momentum transfer
are allowed, the (3He,α) reactions tend to proceed via the higher value.
Conclusions
Apart from the transitions to the 1.760/1.852 MeV levels, the present 27Al(3He,α)26Al
experimental angular distributions show forward peaking characteristic of a direct
pick-up reaction and consequently they have been analysed using the distorted-wave
direct transfer theory. Using optical-model parameters, which describe the related
elastic scattering results, and l = 2 neutron transfer was found to be most consistent
with the data in all cases. At large angles it is likely that other reaction mechanisms are
contributing to the angular distributions since the cross sections are comparable to
those found for the 1.760 MeV state, which does not display direct transfer features.
The relative spectroscopic factors calculated from a simple jj coupling scheme in
which 27Al and 26Al are considered to consist of (1d5/2)-1 and (1d5/2)-2 configurations
only, are in good agreement with those determined from the distorted-wave analysis.
The predominance of l = 2 transitions over l = 0 transitions when both are possible
and when the latter have been reported for the analogous (p,d) and (d,t) reactions
may be further evidence for the preference of (3He,α) reactions to proceed via the
higher l value.
References
Bar-Touv, J and Kelson, I. 1965, Phys. Rev. 139:B1035.
Bevington, P. R. and Anderson, A. S. 1966, Bull. Am. Phys. Soc. 11:908.
Blair, A. G. and Wanger, H. E. 1962, Phys. Rev. 127:1233.
Bouten, M. C., Elliott, J. P. and Pullen, J. A. 1967, Nucl. Phys. A97:113.
Bray, K. H., Nurzynski, J. and Satchler, G. R. 1965, Nucl. Phys. 67:417.
Brennan, M. H. and Bernstein, A. M. 1960, Phys. Rev. 120:927.
Ferguson, J. M. 1964, Nucl. Phys. 59:97.
1964, NASA TN D-2170.
Gove, H. E. 1960, Proc. Int. Conf. on Nuclear Structure, ed. By D. A. Bromley and E. W.
Voght, University of Toronto Press, Toronto, p. 438.
Horvat, P., Kump, P. and Povh, B. 1963, Nucl. Phys. 45:341.
Endt, C. M. and van der Laun, Nucl. Phys. 34:1.
Kelson, I. 1964, Phys. Rev. 134:B267.
Macfarlane, M. H. and French, J. B. 1960, Rev. Mod. Phys. 32:417.
McFadden, L. and Satchler, G. R. 1965, Nucl. Phys. 84:177.
Niewodniczanski, H., Nurzynski, J., and Strzalkowski, A. 1963, Le Journal de Physique,
24:944; also in IFJ Report No. 263,1963.
Nilsson, S. G. 1955, Mat. Fys. Medd. Dan. Vid. Selsk. 29, No 16.
Ophel, T. R. and Lawergren, B. T. 1964, Nucl. Phys. 52:417.
Picard, J. and de Pinho, A. G. 1966, Nuovo Cim. 41B:239.
Pyatov, N. I. 1963, Izv. Akad. Nauk SSSR (ser. Fiz.) 27:1436.
Taylor, I. J., de Barros, S. Forsyth, P. D., Jaffe, A. A. and Ramavataram, S. 1960, Proc.
Phys. Soc. 75:772.
Varshalovich, D. A. and Peker, R. K. 1961, Izv. Akad. Nauk SSSR (ser. Fiz.) 25:287.
Vlasov, N. A., Kalinin, S. P., Ogloblin, A. A. and Chuev, V. I. 1960, JETP (Sov. Phys.)
10:844.
8
Configuration Mixing in the Ground State of 28Si
Key features:
1. Angular distributions for the elastic scattering of 3He particles from 28Si and for the pickup
reaction 28Si(3He,α)27Si were measured and analysed using optical model and distorted wave
theory.
28
2. Results of this study show that neutron orbits, 1d5/2, 2s1/2, and 1d3/2 in the ground state of Si
are filled to approximately 60%, 30%, and 40%, respectively.
Abstract: Angular distributions for the elastic scattering and for the 28Si(3He,α)27Si were measured
using 12 MeV 3He particles. The data were analysed using optical model and distorted wave theory.
Spectroscopic factors for low-energy states were extracted. Results of our study are compared with
previous investigations of neutron pickup reactions from 28Si. The extracted spectroscopic factors
were used to calculate configuration mixing in the ground state of 28Se.
Introduction
In the absence of residual interaction, 28Si nucleus should have the closed 1d5/2
subshell for protons and neutrons. However, transitions involved in neutron (Jones,
Johnson and Griffiths 1968; Kozub 1968; Swenson, Zurmühle and Fou 1967;
Wildenthal and Glaudemans 1967) and proton (Gove at al. 1968; Wildenthal and
Newman 1968) pickup reactions suggest significant admixture of 2s1/2 and 1d3/2
configurations. This should be expected when a residual interaction is included.
The relative contributions of the three configurations can be determined by studying
spectroscopic factors. Measurements for neutron pickup have been carried out using
(p,d) and (3He,α) reactions (Jones, Johnson and Griffiths 1968; Kozub 1968;
Swenson, Zurmühle and Fou 1967; Wildenthal and Glaudemans 1967) but there are
some disagreements between determined spectroscopic factors for low-energy
states in 27Si. The aim of this study was to carry out new measurements for the
28
Si(3He,α)27Si reaction, use the new and existing data, and try to determine the
configuration mixing in the ground state of 28Si.
Experiment
The beam of 12 MeV 3He++ ions was provided by the Australian National University
tandem accelerator. Experimental equipment and technique were similar to those
described for 26Mg(3He,α)25Mg reaction (Chapter 5).
The silicon targets were produced by vacuum evaporation of natural silicon powder
using a VARIAN e-gun.10 Initially, an attempt was made to produce self-supporting
targets. These, however, were found to have too short a lifetime under beam
bombardment to be of practical use, and consequently a target, which was backed
by a thin carbon foil, was used to record the data reported here. This target also
contained impurities of both oxygen and tantalum. The tantalum was due to the use of
a tantalum dish as a lining inside the e-gun crucible in order to facilitate the silicon
10
Manufactured by Vacuum Products Division, Varian Associates, Palo Alto, California, USA.
evaporation. Due to these impurities and the carbon backing the angular range of the
data for both the elastic scattering and the (3He,α) measurements was restricted
mainly to forward angles.
Figure 8.1 shows two examples of particle spectra. Contaminant groups are shown
by their chemical symbols. Results of measurements for the elastic scattering of 3He
particles from 28Si at 12 MeV are shown in Figure 8.2. Figure 8.3 shows the
distributions for the 28Si(3He,α)27Si reaction.
Figure 8.1. Spectra obtained with 12 MeV 3He particles incident on a carbon-backed, natural silicon
target at (a) 30°(lab) and (b) 60°(lab). The groups labelled 0, 1, 2, etc, correspond to the ground state
and the excited states of 27Si from the reaction 28Si(3He,α)27Al. Those labelled "16O" and "12C" were
produced by the reactions 16O(3He,α0)15O and 12C(3He,α0)12C respectively. The group marked "Ta" is
the elastic peak from Ta(3He,3He)Ta and that marked "El" is from the Si + 3He elastic scattering.
The elastic scattering angular distributions was analysed using a six-parameter
Woods-Saxon optical model potential with volume absorption and the computer code
(Perey 1963). Samples of the best fits using different sets of parameters are
presented in Figure 8.2 and the corresponding optical model parameters are listed in
Table 8.1.
Potentials of Table 8.1 were used in the subsequent DWBA analysis of the
28
Si(3He,α)27Si angular distributions. The parameters for the exit channel were the
same as in the analysis of the 27Al(3He,α)26Al reaction and were based on the elastic
scattering of McFadden and Satchler (1966). The neutron bound states wave
functions were calculated for a real Woods-Saxon potential of radius 1.2A1/3 fm and
diffuseness of 0.65 fm. The depths were adjusted to give the separation energy for
the transferred neutrons. The DWBA calculations were carried out using the DRC
computer code (Gibbs et al. 1964) as described in Chapter 5.
The distribution corresponding to the 2.17 MeV, Jπ = 7/2+ state was measured but
was not analysed because it does not exhibit characteristics of a direct pickup
transition. Direct pickup transition to this state would require a 1f7/2 admixture in the
ground state wave function of 28Si. Such an admixture is not only unlikely but also
evidently absent.
3 28 3
Figure 8.2. Differential cross sections for the elastic scattering of He from Si at 12 MeV He energy.
The curves show optical model calculations. The full curve is for the set 1 (see Table 8.1). The
dashed and the dash-dot curves are for sets 2 and 3, respectively.
Table 8.1
Optical model parameters for the elastic scattering of 12 MeV 3He particles from 28Si
The results of the calculation are compared with experimental data in Figure 8.3. The
three l = 2 angular distributions did not show a preference for any of the three sets
of the 3He parameters and only the set 1 predictions for l = 2 are shown in Figure 8.3.
The l = 0 transfer for the first excited state is fitted best by set 2 of the optical model
potential. All the curves in Figure 8.3 were calculated with a zero cut-off radius. DWBA
calculations with a lower cut-off radius of 3.5 fm were also tried but did not produce
any significant difference in the calculated distributions.
Discussion
The observed direct pickup transitions to the low-lying states of 27Si and the relatively
large absolute cross sections indicate a significant admixture of 1d5/2, 2s1/2, and 1d3/2
orbits in the ground state wave function of 28Si.
The contributions of 1d5/2, 2s1/2, and 1d3/2 orbits in the ground state wave function of
28
Si can be estimated by calculating neutron occupation numbers:
∑S (i )
lj
V j2 = i
,
2 j +1
where Slj(i ) is the spectroscopic factor of the ith state corresponding to the lj
configuration. The sum is over all states corresponding to the lj configuration.
Figure 8.3. The distorted wave calculations are compared with the measured differential cross
sections for the reaction 28Si(3He,α)27Al at 12 MeV incident 3He energy. The full curves were
calculated using set 1 of the optical model parameters for 3He particles (see Table 8.1). Calculations
using sets 2 and 3 are also shown for the state 0.78 MeV. All calculations were done using a zero
cut-off radius.
Spectroscopic factors found in the present experiment as well as those from other
studies of neutron pickup reactions from 28Si are compiled in Table 8.2. Comparing
them with other (3He,α) measurements taken at two adjacent energies it may be
seen that the 12 MeV results agree better with the 15 MeV data than with those
obtained at 10 MeV.
The last column in Table 8.2 contains the average values of the spectroscopic factors. They
add up to 5.83, which is close to the expected number of 6.
Table 8.2
Spectroscopic factors for neutron pickup reactions from 28Si
a
) Wildenthal and Glaudemans 1967
b
) Our results
c
) Swenson, Zurmühle and Fou 1967
d
) Jones, Johnson and Griffiths 1968
e
) Kozub 1968
f
) Unresolved doublet
Table 8.3
Configuration parameters for the ground state of 28Si
Orbit nlj′ nlj V j2

1d5/2 6.00 3.20-3.87 53-65%
2s1/2 0.00 0.53 27%
1d3/2 0.00 1.43-2.10 36-53%
nlj′ – the number of neutrons in a given configuration
assuming no residual interaction.
nlj – the experimentally determined number of neutrons in
a given configuration.
V j2 – the experimentally determined occupation numbers
for a given configuration.
The experimentally determined number of neutrons nlj in a given configuration and the
occupation numbers V j2 are listed is Table 8.3. The range of values for l = 2 configurations are
due to uncertainties in spin assignments for the 4.275 MeV and 6.324 MeV states.
Without residual interaction, the orbit 1d5/2 would contain 6 neutrons and thus would
be 100% full. Our results show that residual interaction causes about 40% of
neutrons outside the closed N = 8 shell to occupy orbits 2s1/2 and 1d3/2.
Angular distributions for the 28Si(3He,α)27Si reaction have been measured at 12 MeV
incident 3He energy. The distributions were analysed using direct DWBA theory.
Spectroscopic factors were extracted for transitions to the low-energy states in 27Si.
Results were compared with other available data for neutron pickup reactions from
28
Si and average values of the spectroscopic factors have been calculated. Using
them, configuration components for the ground state wave function of 28Si have been
determined.
Without residual interaction, the six neutrons outside the 1p shell in the ground state
of 28Si would occupy the lowest 1d5/2 orbit. However, neutron pickup reactions show
that the ground state wave function contains a mixture of three configurations, 1d5/2,
2s1/2, and 1d3/2. The determined here neutron occupation numbers show that the
1d5/2 orbit is only 53-65% full. Neutrons, which are outside the 1p shell occupy also
other two orbits in the 1d-2s shell. They fill in 27% of the 2s1/2 orbit and 36-53% of
the 1d3/2 orbit.
References
1964, NASA TN D-2170.
Gove, H. E., Purser, K. H., Schwartz, J. J., Alford, W. P. and Cline, D. 1968,
Nucl. Phys. A116:369
Jones, G. D., Johnson, R. R. and Griffiths, R. J. 1968, Nucl. Phys. A107:659.
Kozub, R. L., 1968, Phys. Rev. 172:1078.
McFadden, L. and Satchler, G. R. 1966, Nucl. Phys. 84:177
Perey, F. G. 1963, Phys. Rev. 131:745.
Swenson, L. W., Zurmühle, R. W. and Fou, C. M. 1967, Nucl. Phys. A90:232.
Wildenthal, B. H. and Glaudemans, P. W. M. 1967, Nucl. Phys. A92:353.
Wildenthal, B. H. and Newman, E. 1968, Phys. Rev. 167:1027
9
The j - dependence for the 54Cr(d,p)55Cr Reaction
Key features:
1. The aim of this work was to study the j - dependence using the reaction
54
Cr(d,p)55Cr.
2. The low-energy states in 55Cr belong to 2p3/2 and 2p1/2 configurations and we
have observed both j = 3/2 and j = 1/2 transitions, which indicated considerable
configuration mixing.
3. We have found that j - dependence varies with the excitation energy and is
more pronounced for low excited states.
4. We have explained the j - dependence as being due to the real component of
the spin-orbit interaction. The imaginary component has no effect on the j -
dependence.
5. Using the observed j - dependence we have assigned spin values to the ground
state and low-energy excited states in 55Cr.
6. We have extracted spectroscopic factors and calculated occupation
probabilities of the orbits 2p3/2 and 2p1/2. We have found that the two neutrons
outside the closed shell in the ground state of 54Cr do not occupy entirely the
lower 2p3/2 orbit, as it would have been expected on the basis of the
independent particle shell model, but spend a considerable time also in the
2p1/2 orbit. On average, there are 1.48 neutrons in the 2p3/2 orbit and 0.52 in
2p1/2. These orbits are 37% and 26% full, respectively.
7. We have calculated the centre-of-gravity single particle energies for the
configurations 2p3/2 and 2p1/2 and found that the spacing between these two
energies agrees well with the spacing for the neighbouring nuclei.
Abstract: Differential cross sections for six transitions in the reaction 54Cr(d,p)55Cr have
been measured at a deuteron energy of 8 MeV. The data have been described
qualitatively by the distorted wave Born approximation calculations employing spin-orbit
terms in the deuteron and proton optical model potentials to reproduce the observed
dependence on the spin transfer, j. We have found that the observed j - dependence
arises from the deuteron and proton real spin-orbit interactions. Spin assignments have
been made for the ground state and for the 7 low-lying exited states in 55Cr. The
spectroscopic factors have been determined and used to calculate occupation numbers
for the configurations 2p3/2 and 2p1/2 as well as the centre-of-gravity single particle
energies for these configurations.
Introduction
I was working late one night in my office. The door was open as usual and my
next-door neighbour David Baugh, a visiting research fellow, came in to have a
little chat. We talked for a while and then unexpectedly he asked me whether I
could help him with the supervision of his PhD student, David Rosalky. He was
wondering whether I could suggest a suitable research project for him.
It so happened that I had in mind to look into the j - dependence by using 54Cr
as the target nucleus but I was already too busy with other projects. However,
David’s request sounded attractive so I decided to work with him and his
student.
Usually, direct reactions allow only for the determination of the orbital angular
momentum l but do not distinguish between states with different values of the
total angular momentum, j, which belong to the same l. However, it has been
demonstrated (Lee and Schiffer 1964 and 1967) that in some cases involving
(d,p) reactions it is possible to determine not only l but also j values.
For a spin-zero target nucleus, the spin of the residual nucleus J is equal to the
transferred total angular momentum j and the (d,p) angular distributions depend
directly upon J. The j - dependence, as this property is called, provides a
valuable spectroscopic tool for distinguishing between states belonging to the
same l but differing in their total angular momentum. Thus, using the j –
dependence one should be able to distinguish between states with J = l + 1/2
and J = l - 1/2.
I have chosen 54Cr as a target nucleus because it contains two neutrons
outside the closed 1f7/2 shell. Thus, stripping reaction (d,p) should be expected
to lead to both 2p3/2 and 2p1/2 configurations and thus offer a good opportunity
to study the j -dependence. Indeed, earlier measurements (Bock et al. 1965a)
identified eight states corresponding to the l =1 transfers. The single-neutron
stripping reaction should therefore be expected to show clear j - dependence
and thus should allow for determining the j values for low-energy states in the
residual nucleus of 55Cr. I also hoped that this reaction could be used to study
the conditions for the j – dependence and explain why this feature is not always
clear in the (d,p) reactions.

A 94% enriched target of 54Cr, on a thin carbon backing (15-20 μg/cm2), was
bombarded with 8 MeV deuterons from the ANU EN tandem accelerator. The
charged reaction products were detected by two movable 1000-micron silicon
surface barrier detectors mounted 20° apart in a 51 cm scattering chamber.
Experimental set up was similar as that described in Section 5 for the
26
Mg(3He,α)25Mg reaction.
The spectrum from each detector was collected in 2048 channels of an IBM
1800 data acquisition system with a dispersion of about 7 keV/channel. For
measurements at forward angles, beam intensities were reduced to minimize
pulse pile-up effects and dead time corrections, which were kept below 5%. The
best resolution obtained with this arrangement was 23 keV for 11 MeV protons.
A product of the target thickness and solid angle was determined by elastic scat-
tering of 5 MeV alpha particles in the range 20° to 80°. The ratio of the
experimental to Rutherford cross sections was constant within experimental
errors over this range. The alpha-particle spectra resulting from these
measurements revealed the presence of carbon (used as target backing),
oxygen and tantalum impurities in the target. The thickness of 54Cr was found to
be 52 ± 3 μg/cm2.
Figure 9.1 shows proton spectra measured at 500 and 1200 in the laboratory
system. The proton groups arising from the 54Cr(d,p)55Cr reaction are marked
as pi. Angular distributions between 150 and 160° (except for angles where 160
and 12C contaminant proton peaks interfered) were extracted for the six
strongest l =1 transitions and for the elastically scattered deuterons from 54Cr.
Figure 9.1. Proton spectra from the 54Cr(d,p)55Cr stripping reaction at 8 MeV deuteron bombarding
energy measured at 500 and 1200 (lab). Proton spectra are numbered according to Table 8 of
Macgregor and Brown (1966). The excitation energies of states populated by l = 1 transitions are
quoted.
The level at 2.7 MeV excitation energy, which was weakly excited in our work,
consisted of an unresolved doublet, 2.695 and 2.710 MeV, (Macgregor and
Brown 1966), of which at least one component was formed by an l = 1 transition
(Bock et al. 1965a). An accurate angular distribution could be obtained for the
2.905 MeV level since this is strongly excited compared to the partially resolved
2.874 MeV state.
The measured angular distributions are shown in Figure 9.3, the error bars
represent relative errors which include contributions from counting statistics,
background subtraction, the unfolding of partially resolved peaks and errors in
beam monitoring. Average values have been taken of repeated measurements
weighted according to the inverse square of their errors. The scattering angle
could be set with an accuracy of ± 0.20.
Distorted wave analysis

Zero-range distorted-wave Born approximation (DWBA) calculations were
carried out using optical model potential as defined in Chapter 6 for the
16
O(3He,3He)16O scattering. However, in this analysis we have used only
surface absorption potential.
We have tried both real and imaginary potentials for the spin-orbit interaction
but found that the imaginary component had little effect on the calculated
distributions. Thus, we have found that the dependence of the calculated
angular distributions on j arises from deuteron and proton real spin-orbit
potentials. Consequently, we have used only real component for the spin-orbit
interaction.
To understand the dependence of j - dependence on the excitation energy we
have calculated a series of angular distributions for j = 1/2 and j = 3/2 as a
function of excitation energies. These distributions are shown in Figure 9.3. It
can be seen that theoretical calculations produce different shapes of angular
distributions for different j values. However, the most prominent differences
between j = l - 1/2 (the left hand side of the figure) and j = l + 1/2 (the right
hand side) occur at low excitation energies of the residual nucleus (55Cr). As the
excitation energies increases, it is increasingly more difficult to distinguish
between the two j values.
Figure 9.2. The dependence of the calculated angular distributions for j = 1/2 (the left hand side
of the figure) and j = 3/2 (the right hand side) on the excitation energy in 55Cr for the reaction
54
Cr(d,p)55Cr at 8 MeV deuteron energy. The curves are labelled by the excitation energies with
0 being for the ground state. The figure shows that the j – dependence is clear for the low
excitation energies but becomes gradually less pronounced when the excitation energy is
increasing.
Results of our DWBA calculations are compared with experimental distributions
in Figure 9.3. It can be seen that by comparing theoretical and experimental
distributions one can distinguish between different j values and thus one can
assign the spin values to the relevant states in the residual nucleus.
54
Spectroscopic information based on our study of the Cr(d,p)55Cr reaction is
summarized in Tables 9.1 and 9.2.
Spectroscopic factors for stripping reactions are proportional to the degree a
given orbit is empty. Without a residual interaction, orbit 2p1/2 in the ground state
of 54Cr would be empty and orbit 2p3/2 would contain two neutrons. Our results
show that these orbits are 26% and 37% percent full indicating, as expected,
significant residual interaction. Orbit 2p1/2 contains on average 0.52 neutrons
and orbit 2p3/2 1.48 neutrons.
Figure 9.3. Angular distributions of protons from the 54Cr(d,p)55Cr reaction corresponding to the
l = 1 transitions. The curves are the DWBA predictions for j = 1/2 and j = 3/2. This figure shows
clear the j dependence at low excitation energies. At higher excitation energies, the differences
between the two j values are less clear but it is still possible to make a distinction between the
two j values by comparing the theoretical calculations with experimental data.
Table 9.1
54 55
Spectroscopic information based on the study of the reaction Cr(d,p) Cr at 8 MeV deuteron energy
a
Level j Configuration S (l , j ) a) S (l , j ) )
(MeV) j = 3/ 2 j = 1/ 2
0.000 3/2 2p3/2 0.52
0.245 1/2 2p1/2 0.12
0.573 3/2 2p3/2 0.10
1.487 1/2 2p1/2 0.34
2.7 b) (1/2) (2p1/2) 0.06
2.905 1/2 2p1/2 0.21
3.043 (3/2) (2p3/2) 0.01
3.696 (3/2) (2p3/2) 0.01
a
) Spectroscopic factors for the last two states (3.043 and 3.696 MeV) are
from Bock et al. (1965a).
b
) Unresolved doublet 2.695/2.710 MeV.
Table 9.2
Additional spectroscopic information based on the study of the reaction 54Cr(d,p)55Cr at 8 MeV
deuteron energy
Orbit U 2j hj h ′j V j2 nj n′j Ej
2p3/2 0.63 2.52 4.00 0.37 1.48 2.00 0.14

2p1/2 0.74 1.48 0.00 0.26 0.52 0.00 1.79
U = ∑ Si (l , j ) – The experimentally determined vacancy number, i.e. the degree to
2
j
i
which a given neutron orbit in the ground state of 54Cr is empty.
h j = (2 j + 1)U 2j – The experimentally determined average number of neutron holes in a
given configuration in the ground state of 54Cr.
h′j – The number of expected neutron holes in the absence of the residual interaction.
V j2 = 1 − U 2j – The experimentally determined occupation number, i.e. the degree to which
a given neutron orbit in the ground state of 54Cr is full.
n j = (2 j + 1)V j2 – The experimentally determined average number of neutrons in a given
configuration in the ground state of 54Cr.
n′j – The expected number of neutrons in the absence of the residual interaction.
Ej (MeV) – Centre-of-gravity energies for states with the total angular momentum j.
Spectroscopic factors allow also for calculating the center of gravity energies Ej
for states corresponding to configurations 2p1/2 and 2p3/2.
∑ S (l , j ) E
i i
Ej = i
∑ S (l , j )
i
i
where Ei are excitation energies for spins j, and the sum is over all the states
with the same spin j.
Results of calculations are shown in Table 9.2. The spacing between the two
energies is 1.65 MeV, which is in good agreement with the energy spacing in
the neighboring nuclei (see Table 9.3).
Table 9.3
The spacing between the 2p1/2 and 2p3/2 centre-of-gravity energies for nuclei in the Cr region
47 49 51 51 53 55 59 61
Nucleus Ca Ca Cr Cr Cr Cr Ni Ni
a b c d e f
Spacing 1.86 ) 2.03 ) 1.60 ) 1.64 ) 1.68 1.65 ) 1.90 ) 1.20 f)
a
) Belote et al. 1966; b) Kashy et al. 1964; c) Robertshaw et al. 1968; d) Delic, G. and Robson, B.
A. 1969; e) Our result; f) Fulmer et al. 1964.

We have studied the applicability of the j - dependence for the (d,p) reactions in
assigning spin values of states in residual nuclei. We have found that this
feature is indeed a useful tool in distinguishing between j = l + 1/2 and j = l – 1/2
values.
By studying the dependence of the shapes of the angular distributions on the
excitation energy we have also found that j - dependence is clear at low
excitation energies. As the energy of excited states increases the distinction
between j = l + 1/2 and j = l – 1/2 decreases. However, by comparing theoretical
and experimental distributions it is still possible to distinguish between different j
values even for high-excited states.
We have found that j - dependence is related only to the real component of the
spin-orbit interaction. The imaginary component has no influence.
By comparing the DWBA calculations with the experimental angular
distributions we have assigned spins 3/2, 1/2, 3/2, 1/2,and 1/2 to the ground state,
0.245, 0.573, 1.487, and 2.905 MeV in 55Cr, respectively. These states belong
to the 2p3/2 and 2p1/2 configurations.
We have found that neutron orbits 2p3/2 and 2p1/2 in the ground state of 54Cr are
37% and 26% full and thus they contain on average 1.48 and 0.52 neutrons,
respectively. We have calculated the centre-of-gravity single particle energies
for orbits 2p3/2 and 2p1/2 in 55Cr. The energy spacing between them is in good
agreement with the spacing in the neighbouring nuclei.
References
Bock, R., Duhm, H. H., Martin, S., Rüdel R. and Stock, R.1965a, Nucl. Phys.
72:273
Bock, R., Duhm, H. H., Jahr, R., Santo, R., and Stock, R. 1965b, Phys.
Lett. 19:417.
Belote, T. A., Chen, H. Y., Hansen, O. and Rapaport, J. 1966, Phys. Rev.
142:624.
Delic, G. and Robson, B. A. 1969, Nucl. Phys. A134:470
Kashy, E., Sperduto, A., Enge, H. A. and Buechner, W. W. 1964, Phys. Rev.
135:B865.
Lee, L. L. and Schiffer, J. P. 1964,Phys. Rev. 136:B405.
Lee, L. L. and Schiffer, J. P. 1967, Phys. Rev. 154:1097.
Macgregor, A. and Brown, G. 1966, Nucl. Phys. 88:385
Robertshaw, J. E., Mecca, S., Sperduto, A. and Buechner, W. W. 1968, Phys.
Rev. 170:1013
10
Tensor Analyzing Powers for Mg and Si Nuclei
Key features:
1. Unpolarized deuterons accelerated to 7 MeV were polarized in the elastic scattering
from Mg and Si nuclei and the corresponding angular distributions of the tensor
analyzing powers T20 (θ ) , T21 (θ ) , and T22 (θ ) were measured using the 3He(d,p)4He
polarization analyser.
2. Experimental results were analysed using optical model in combination with the
statistical theory of Hauser and Feshbach. The optical model contained not only spin-
orbit but also tensor components.
3. Various shapes of the tensor component were examined.
4. Our calculations show that only the distributions of the tensor analyzing powers are
sensitive to tensor interaction. Vector analyzing power iT11 (not measured but
calculated) is sensitive only to the spin-orbit interaction.
5. The tensor interaction potential has been found to be shallow, attractive, and long
range.
Abstract: Angular distributions for the elastic scattering cross sections and for the three
components of the tensor analyzing powers T20 (θ ) , T21 (θ ) , and T22 (θ ) were measured for
the elastic scattering of 7 MeV deuterons from Mg and Si nuclei. In addition, differential cross
sections were measured at 7 MeV for Mg and 7, 8, 9, 10, and 11 MeV for Si targets. All the
distributions were analysed using optical model formalism with the inclusion of the statistical
theory of Hauser and Feshbach. Angular distributions of the differential cross sections were
reproduced well without spin-dependent interactions. However, the analysis of the tensor
analyzing powers required not only the spin-orbit but also tensor interaction. Only one
component of the tensor interaction, TR, was necessary. Tensor interaction potential was
found to be shallow and attractive, and to have a long-range.
Introduction
It is well known that nuclear potential contains not only central but also spin
dependent components. Measurements of differential cross sections serve as a
useful tool to study the central part of nuclear interaction. However, the best way to
study the spin-dependent components is by carrying out the polarization
measurements. In particular, tensor components of the interaction potential can be
studied by studying tensor analyzing powers.
Interaction of spin 1/2 particles with spinless target nuclei involves only spin-orbit
forces. However, interaction of spin 1 particles, such as deuterons, involves also
tensor dependent forces. It is to study these tensor-dependent forces that we have
carried out measurements of tensor analyzing powers for the elastic scattering of
deuterons from Mg and Si target nuclei.
Angular distributions of the tensor analyzing powers T20 (θ ) , T21 (θ ) , and T22 (θ ) were
measured for the elastic scattering of 7 MeV unpolarized deuterons from Mg and Si
targets. In addition, to support the theoretical analysis of experimentally determined
tensor moments (tensor analyzing powers), measurements of differential cross
sections of the elastically scattered deuterons were carried out at 7 for Mg and 7, 8,
9, 10, and 11 MeV for Si. All the measurements were carried out using deuterons
accelerated in the EN electrostatic tandem accelerator in the Department of Nuclear
Physics of the Australian National University.
The targets
In order to reduce the data collection time, thick targets were used for the
polarization measurements. However, their thickness had to be kept within a
reasonable limit to avoid an excessive energy spread of elastically scattered
deuterons. The thickness of the Mg target was (44.2 ± 4.5) μg/cm2 or (420 ± 46) keV
for 7.0 MeV deuterons. The thickness of Si target was (58.5 ± 6.2) μg/cm2 or (440 ±
49) keV.
The Mg target was prepared by rolling a natural magnesium strip, which was initially
about 0.2 mm thick. This simple method allowed for the production of thin Mg foils
with the required thickness of about 0.025 mm.
However, due to the very brittle nature of silicone, the preparation of the thick Si
targets was more complex (Djaloeis and Nurzynski 1972). Briefly, the idea was to
glue a silicone disk to a glass slide and then to grind the silicone to a suitable
thickness.
The surface of one side of a circular silicon disk with the diameter of about 22 mm
and thickness of about 1.5 mm was smoothed by grinding it with fine-grained Al2O3
powder (grade 500) on a smooth piece of glass. Kerosene was used to wet the
powder. A glass slide located on top of a hot plate was then heated and after the
temperature had reached approximately 80°C a 'Lakeside 70 thermal glue' 11 was
evenly spread on the glass slide. Next, after cleaning in alcohol, the smoothed side of the
disk was firmly attached to the glass slide by means of the thermal glue while the
glass slide was still on the hot plate. To make the surfaces of the slide and the disk
parallel to each other, the disk was pressed firmly against the glass slide while the
glue was still in its molten state.
The glass slide-glue-disk combination was then quickly removed from the hot plate,
placed on an even surface and similar pressure was again applied to the silicone
disk until the combination cooled down and the glue hardened. This procedure was
necessary to avoid the formation of a wedge-shaped finish between the silicone disk
and the glass slide.
Having completed this stage, the other side of the disk was ground down to a
thickness of approximately 0.025 mm by using a succession of grinding powders of
decreasing grain size. Once the required thickness was approximately obtained the
combination was again heated to melt the glue. The target was pushed, with extreme
care, toward the edge of the glass slide until it became free. The target was then
immersed in alcohol for a few hours until all the adhering glue became dissolved,
11
Manufactured by Hugh Courtright and Co., Chicago, Ill., USA.
thus leaving the target free of contaminants. Finally the target was stored in a dry
and clean atmosphere to allow complete evaporation of the alcohol.
This procedure was, however, long and tedious and the rate of success was low.
From the twelve disks used only two acceptable targets were finally produced.
Initially, the thickness of the resulting targets was measured using a Mitutoyo
micrometer, accurate to 2.5 μm. Later, Rutherford scattering of low-energy α
particles was used to find the thickness of the targets more accurately.
Angular distributions for elastic scattering differential cross sections were measured
over a wider range of angles than for the distributions of the tensor analyzing
powers. The measurements of the differential cross sections could be carried out
with thinner targets, which were prepared by evaporation using VARIAN e-gun.
However, again the preparation of thin silicone targets had to be done with special
care.
We have found that the stability of the target depended critically on the distance
between the crucible and the supporting glass slide. With a short distance, the
deposited wafer of Si tended to crack. When the glass slide was placed far from the
crucible, the evaporation time was too long. We have found that the optimal distance
was 8 cm.
The temperature was also critical. Any attempt to speed up the evaporation resulted
in damage to the material already deposited. The best method was a slow heating
up to the melting point of the silicone (14200C) followed directly by a slow continuous
evaporation. The temperature was measured using a Leeds and Northrup optical
pyrometer.
Experimental setup
Schematic diagram of the experimental setup is shown in Figure 10.1. Figures 10.2
and 10.3 show the horizontal and vertical view of the apparatus, respectively.
Figure 10.1. Schematic diagram of the experimental setup used in the measurements of the tensor
analyzing powers in the elastic scattering of deuterons from Mg and Si targets. Deuterons were
polarized by scattering from the first target (Mg or Si) and the corresponding tensor analyzing powers
were measured by using the 3He(d,p)4He polarimeter. The scattered deuterons were decelerated by
Mylar foil to the energy of about 430 keV before using them to induce the resonance reaction
3
He(d,p)4He.
Unpolarized deuterons were scattered from Mg or Si targets. The scattered (and now
polarized) deuterons were decelerated to around 430 keV using Mylar foil. The
decelerated deuterons were directed to the 3He cell to induce the 3He(d,p)4He
reaction leading though a 3/2+ resonance in 5Li (see Figure 10.4) which served as the
polarization analyser. In general, an operating pressure in the 3He cell was 5 atm,
which was sufficiently high to stop up to 800 keV deuterons.
Figure 10.2. The horizontal view of the apparatus. 1. Beam collimator and anti-scattering baffle. 2.
Solid target (Mg or Si). 3. Beam stop. 4. Havar foil. 5. A defining slit. 6. The 3He cell. 7. The CsI crystal
(only three CsI detectors out of six are shown here). 8. Lucite light pipe. 8. Photomultiplier tube.
Figure 10.3. A vertical cross section of the apparatus. 1. Solid target (Mg or Si). 2. Collimator. 3. Gold
target. 4. Scattering chamber. 5. 3He cell. 6. Square CsI crystal. 7. Annular CsI crystals. 8. Lucite light
pipes. 9. Photomultiplier tubes. 10. Slit for the 3He cell.
Figure 10.4. The energy level diagram for 6Li (Lauritsen and Ajzenberg-Selove 1962). The 3/2+
3
resonance at 16.64 MeV is formed by 430 keV deuterons interacting with He. The corresponding
3 4
He(d,p) He reactions serves as a polarization analyser.
Protons from the 3He(d,p)4He resonance reaction were detected using 6 scintillation
counters placed at azimuthal angles of 00, 450, 900, 1350, and 1800. Square CsI
crystals were used for all detectors except for the detector positioned in the direct
line of the incident deuterons (i.e. deuterons scattered from Mg or Si targets). This
detector was equipped with an annular CsI crystal to prevent the detection of protons
coming directly from the target.
Measurements of the tensor analyzing powers could be carried out with only 4
detectors (see below). The extra two detectors (at ϕ = 450 and 1350) were added to
increase data collection efficiency.
The CsI crystals were attached to Lucite light pipes and mounted on 5 cm Dumont
6392 photomultiplier tubes. The pulses from photomultipliers were fed through
Franklin double-delay-line preamplifiers, amplifiers, and directed via analogue-to-
digital converters to an online IBM-computer.
The data collection time was between about 2 and 30 hours depending on the
reaction angle. The collection time for the differential cross sections was significantly
shorter.
The 3He(d,p)4He polarimeter
The differential cross section for the 3He(d,p)4He reaction at the resonant energy of
430 keV can be expressed by the following formula (McIntyre 1965; Welton 1963):
σ (ω , ϕ ) =
⎧⎪ ⎡ 1 3 ⎤ ⎫⎪
σ 0 (ω )⎨1 − f ⎢ (3 cos 2 ω − 1)T20 (θ ) − 3 (sin ω cos ω cos ϕ )T21 (θ ) + (sin 2 ω cos 2ϕ )T22 (θ )⎥ ⎬
⎪⎩ ⎣2 2 2 ⎦ ⎪⎭
where ω is the reaction angle for the 3He(d,p)4He reaction, ϕ is the azimuthal angle
between the first reaction plane (for the elastic scattering) and the second reaction
plane (for the 3He(d,p)4He reaction), and T2 q (θ ) are the tensor analyzing powers for
the first reaction (the elastic scattering). They are definedr
in the right-handed
coordinate system with the z-axis along the incident beam k0 and the y-axis along
r r
the k0 × kd vector as shown if Figure 10.1. The energy dependence of the factor f has
been studied by Brown, Christ, and Rudin (1966) in the energy range of 300-1700
keV.
It is clear that by a proper combination of measurements at various azimuthal angles
ϕ the components of tensor moments T2 q (θ ) can be determined.
The reaction angle for the 3He(d,p)4He analyser for detectors 2-6 was chosen to
correspond to cos2 ω = 1 / 3 ( ω ≈ 54.740 in the centre of mass system or ≈ 52.550 in the
laboratory system). For the detector 1, ω = 0. The equation for the differential cross
section for the 3He(d,p)4He reaction is given by simpler expressions:
For detector 1
⎡ 1 ⎤
σ 1 = σ 0 ⎢1 − f T20 (θ )⎥
⎣ 2 ⎦
For detectors 2-6
⎡ ⎛ 2 1 ⎞⎤
σ i = σ 0 ⎢1 + f ⎜⎜ (cosϕ )T21 (θ ) − (cos 2ϕ )T22 (θ ) ⎟⎟⎥ i = 2,3,4,5,6
⎣⎢ ⎝ 3 3 ⎠⎦⎥
More explicitly, for detectors 2-6, we have the following relations:
⎡ ⎛ 2 1 ⎞⎤
Detector 2 (ϕ = 00): σ 2 = σ 0 ⎢1 + f ⎜⎜ T21 (θ ) − T22 (θ ) ⎟⎟⎥
⎢⎣ ⎝ 3 3 ⎠⎥⎦
⎡ 1 ⎤
Detector 3 (ϕ = 900) σ 3 = σ 0 ⎢1 + f T22 (θ )⎥
⎣ 3 ⎦
⎡ ⎛ 2 1 ⎞⎤
Detector 4 (ϕ = 1800): σ 4 = σ 0 ⎢1 − f ⎜⎜ T21 (θ ) + T22 (θ ) ⎟⎟⎥
⎣⎢ ⎝ 3 3 ⎠⎦⎥
⎡ 1 ⎤
Detector 5 (ϕ = 450): σ 5 = σ 0 ⎢1 + f T21 (θ )⎥
⎣ 3 ⎦
⎡ 1 ⎤
Detector 6 (ϕ = 1350): σ 6 = σ 0 ⎢1 − f T21 (θ )⎥
⎣ 3 ⎦
It follows that the analyzing powers are given by the following relations:
1 ⎛ σ ⎞
T20 (θ ) = 2 ⎜⎜1 − 1 ⎟⎟
f ⎝ σ0 ⎠
1 3 ⎛σ3 − σ4 ⎞
T21 (θ ) = ⎜ ⎟
f 2 2 ⎜⎝ σ 0 ⎟⎠
1 3 ⎛σ5 −σ6 ⎞
T21 (θ ) = ⎜ ⎟
f 2 ⎜⎝ σ 0 ⎟⎠
1 ⎛ σ3 ⎞
T22 (θ ) = 3 ⎜⎜ − 1⎟⎟
f ⎝σ0 ⎠
1 ⎛ σ2 +σ4 ⎞
T22 (θ ) = 3 ⎜⎜1 − ⎟
f ⎝ 2σ 0 ⎟⎠
σ 2 + 2σ 3 + σ 4 + σ 5 + σ 6
σ0 =
6
Figure 10.5. Examples of proton spectra from the reaction 3He(d,p)4He induced by polarized
deuterons following the elastic scattering from Mg target. Similar spectra are also for the Si target.
In practice, the analyzing powers are determined by measuring the ratios of the
number of counts with a given target (Mg or Si in our case) to the number of counts
with gold target for each reaction angle θ. The above expressions are the same
except that now σ i are replaced with the relevant ratios Ri for each detector, and
σ 0 by R0, which is determined by taking an average value for all 5 detectors as shown
in the last equation. This procedure simplifies the procedure of measuring the
analyzing powers. It also eliminates geometrical asymmetries.
A sample of proton spectra for the 3He(d,p)4He reaction detected by the six
scintillation counters is shown in Figure 10.5. This figure shows that even though the
3
He(d,p)4He reaction angle ω is the same for all six detectors the number of counts
depends on the azimuthal angle ϕ , as it should if the incident deuterons are
polarized. The observed azimuthal asymmetries were used to calculate tensor
moments T2 q for deuterons scattered from the first target (Mg or Si).
Results of measurements are shown in Figures 10.7 and 10.10 for the tensor
analyzing powers, and in Figures 10.6, 10.8, and 10.9 for the differential cross
sections, together with results of theoretical calculations.
The theory
It is well known that direct processes induced by low-energy deuterons bombarding
light targe nuclei can be influenced by compound nucleus formation. It has been
therefore decided to include the statistical corrections in these calculations using
statistical theory as developed by Wofenstein (1949) and Hauser and Feshbach
(1952). The theoretical differential cross section σth(θ) for the differential cross
sections can be then written as:
σ th (θ ) = σ OM (θ ) + Rσ HF (θ )
where σ OM (θ ) is the theoretical differential cross section calculated using optical
model, σ HF (θ ) is the statistical, Hauser-Feshbach cross section, and R is the
reduction factor (Hodgson and Wilmore 1967).
Likewise,
⎡ Rσ HF (θ ) ⎤
[T2q ]
(θ ) th = ⎢1 − [ ]
⎥ T (θ ) OM
σ exp (θ ) ⎦⎥ 2 q
⎣⎢
Theoretical analysis of the experimental differential cross sections σ (θ ) and tensor
moments T2 q (θ ) was carried out using optical model potential containing not only
r r
central and spin-orbit ( S ⋅ L ) components but also tensor terms TR and TL:
(S ⋅ r ) 2 2
TR = −
r2 3
1 2
TL = (L ⋅ S) 2 + (L ⋅ S) − L2
2 3
The potential used in the calculations had the following form:
2
⎛ h ⎞ 1 df S
U (r ) = VC (r ) − VfV − iWD g + VS ⎜⎜ ⎟⎟ (S ⋅ L) + VR hRkTR + VL hLkTL
⎝ π ⎠
m c r dr
where VC (r ) is the Coulomb potential, V, WD, VS, VR, VL are the depths of the relevant
components, and f, g and h are the radial form factors. The form factors f and g are
the usual Woods-Saxon and derivative of Woods-Saxon type, respectively.
If fi is defined as
{ [(
f i ≡ f (r , ri , ai ) = 1 + exp r − ri A1 / 3 / ai ) ]} −1
then
fV ≡ f (r , rV , aV )
f S ≡ f (r , rS , aS )
d
g = 4aW f (r , rW , aW )
dr
The form factors h can have various forms. We have tried the following three options:
(i) Derivative of Woods-Saxon (D):
d
hRD = −4aR f (r , rR , aR )
dr
d
hLD = −4aL f (r , rL , aL )
dr
(ii) Thomas (T):

2
⎛ h ⎞ 1 d
h = −⎜⎜
T
R
⎟⎟ f (r , rR , aR )
⎝ mπ c ⎠ r dr
2
⎛ h ⎞ 1 d
hLT = −⎜⎜ ⎟⎟ f (r , rL , aL )
⎝ mπ c ⎠ r dr
(iii) Gaussian (G)
hiG = exp(− xi2 )
with xi = (r − ri A1 / 3 ) / ai and i = R or L.
The quality of fits was examined both visually and by calculating the function:
1⎡ 2 ⎤
χ2 = ⎢ Nσ χσ + ∑ N q χ q ⎥
2 2
N⎣ q =0 ⎦
where
2
N = Nσ + ∑ N q
q =0
2
1 Nσ
⎡σ exp (θ i ) − σ th (θ i ) ⎤
χσ =
2
Nσ
∑ ⎢
i =1 ⎢ Δσ exp (θ i )
⎥
⎥⎦
⎣
[ ] [ ]
1 q ⎡ T2 q (θ i ) exp − T2 q (θ i ) th ⎤
N
2
χ q2 = ∑ ⎢ Δ T (θ )
N q i =1 ⎢ [ ] ⎥
⎥⎦
⎣ 2q i exp
Nσ is the number of experimental points for the differential cross

sections, σ exp (θ i ) measured at angles θi; Nq is the number of experimental points
[T2 q (θ i )]exp ; Δσ exp (θ i ) and Δ[T2 q (θ i )]exp are the experimental uncertainties of σ exp (θ i ) and
[T2 q (θ i )]exp , respectively.
The analysis of the Mg(d,d)Mg data

Calculations for the differential cross sections σ (θ ) for the scattering from Mg target
are shown in Figure 10.6. Similar fits were obtained for the elastic scattering from Si.
It can be seen that theoretical calculations reproduce well the experimental angular
distribution.
Figure 10.6. The differential cross sections σ(θ) for the elastic scattering of 7 MeV deuterons from Mg.
The curve marked HF is the calculated Hauser-Feshbach contribution to the elastic scattering. The
continuous curve going through the data points is the optical model fit, which included the Hauser-
Feshbach contribution. Optical model potential parameters used in this analysis are listed in Table
10.1 as the set σ-1.
We have spent a considerable amount of computation time trying to understand
influence of various optical model parameters on the calculated distributions of the
tensor analyzing powers for both target nuclei. Examples of the calculated curves for
the tensor analyzing powers for Mg are shown in Figures 10.7. The corresponding
potentials are listed in Table 10.1
The calculated shapes for the T20 tensor analyzing power are similar for all
parameter sets. The χ2 value is the lowest for the Gaussian shape (THF-G set; χ2 =
6.82) and the highest for the spin-orbit interaction without tensor components (THF-
S1 set, χ2 = 9.75).
The best fit for the T21 tensor analyzing power is obtained by using the derivative
shape of the TR component of the optical model potential (set THF-D1;χ2 = 6.68).
Adding the TL component does not improve the fit (set THF-DD1; χ2 = 6.85).
Changing the shape to Thomas or Gaussian results in higher χ2 values (10.27 and
17.37 respectively).
Figure 10.7. Tensor analyzing powers for the Mg(d,d)Mg scattering at 7 MeV deuteron energy.
Experimental points are compared with theoretical calculations. The parameter sets are listed in Table
10.1. The curves belonging to the respective potentials used in the calculations are identified for the
T21 (θ ) angular distribution.
Table 10.1
Parameters of the optical model potential used in the analysis of differential cross sections and tensor
analyzing powers for the elastic scattering of 7 MeV deuterons from Mg
The depths V, WD, VR, and VL are in MeV. VS is in Mev·fm2 and the geometrical parameters r and a are
in fm.
Re – the real part of the central potential; Im – the imaginary part; par – parameters; shape – shape of
the tensor parts of the optical model; D – derivative shape; T – Thomas shape; G – Gaussian shape.
Italicised numbers represent fixed parameters during the search.
The σ-1 potential was used to fit only the experimental differential cross sections σ(θ). Potentials
labelled as THF were used to analyse not only the differential cross sections but also the tensor
analyzing powers. They included calculations of the compound nucleus contributions.
The best fit to the T22 tensor analyzing power is by using the spin-orbit interaction
without the tensor components (set THF-S1; χ2 = 19.23). However, this is mainly due
to one experimental point at θc.m. = 40.430. When TR component is included, the best
fit is for the Thomas shape (set THF-T; χ2 = 31.75).
The overall results indicate that tensor component TR in the optical model potential is
necessary, in addition to the spin-orbit component, to describe the observed tensor
analyzing powers. Adding tensor component TL makes practically no difference.
There is also no compelling need for shapes other than the more conventional
derivative shape.
The analysis of the Si(d,d)Si data
Calculations of the differential cross sections for Si are compared with the
experimental results in Figure 10.8 and the corresponding parameters of the optical
model potential are listed in Table 10.2. Similar fits were obtained for V ≈ 100 MeV.
Figure 10.8. Examples of fits to the differential cross sections for the elastic scattering of 7-11 MeV
deuterons from Si. The parameters of the optical model potential are listed in Table 10.2
Table 10.2
Optical model parameters generating theoretical distributions displayed in Figure 10.8
Ed V rV aV WD rW aW
(MeV) (MeV) (fm) (fm) (MeV) (fm) (fm)
7 65.5 1.05 0.877 10.6 1.539 0.758
8 66.3 1.05 1.013 15.7 1.701 0.485
9 62.0 1.05 0.970 15.2 1.638 0.555
10 65.6 1.05 0.932 18.6 1.565 0.484
11 61.8 1.05 0.944 20.0 1.485 0.551
We have also carried out calculations of the differential cross sections by including
the spin-orbit and tensor components. Results are shown in Figure 10.9 for 7 MeV
deuterons for two sets of parameters OM-1 corresponding to V = 70.3 and OM-2 for
V = 116.0 MeV. The parameters are listed in Table 10.3. Surprisingly, the fit for V =
116.0 MeV is worse than for the more shallow potential.
Finally, we have also included Houser-Feshbach corrections. The calculations,
labelled as HF-D are also shown in Figure 10.9 and the corresponding optical model
parameters are listed in Table 10.3
Figure 10.9. Optical model calculations with spin-dependent components for the differential cross
sections of 7 MeV deuterons scattered elastically from Si nuclei. One set of calculations (HF-D)
includes also Hauser-Feshbach corrections. The corresponding optical model parameters are listed in
Table 10.3.
Table 10.3
Optical model parameters for the calculations with the central, spin-orbit, and tensor components
Figure 10.10 compares optical model calculations for the angular distributions of the
tensor analyzing powers with experimental results. Two sets of curves are displayed.
They both belong to the potential set HF-D of Table 10.3 but with or without tensor
interaction. Shown are also calculations for the vector analyzing power iT11 (θ ).
Calculations using the two other sets (OM-1 and OM-2) of Table 10.3 produced
similar results.
Figure 10.10 shows that angular distributions for the tensor analyzing powers are
sensitive to tensor interaction, and thus this interaction is essential in the analysis of
the experimental angular distributions of the tensor analyzing powers. In contrast,
the vector analyzing power is not sensitive to tensor interaction. The calculated
distributions for this component are nearly identical whether tensor interaction is
included or not. Thus while measurements of the vector tensor analyzing powers can
yield information about the spin-orbit interaction, the study of tensor analyzing
powers opens an opportunity to learns about tensor forces.
The TR component used in these calculations had a derivative shape. The
parameters optimising the fits are similar to the parameters used in the analysis of
the Mg(d,d)Mg scattering, except for aR which for Si has a significantly smaller value
than for Mg.
Figure 10.10. Measured angular distributions of the tensor analyzing powers for Si(d,d)Si elastic
scattering are compared with theoretical distributions. Optical model parameters used in the
calculations are listed in Table 10.3 as set HF-D. Similar results, published but not shown here, have
been also obtained for sets OM-1 and OM-2. The fine line was calculated using central and spin orbit
components only. The thicker solid line was calculated by including also the TR component of the
tensor interaction. Calculations for the vector analyzing power iT11 are also shown. As can be seen,
they are insensitive to the tensor interaction.

In this work we have studied nuclear interaction in the deuteron-nucleus system. We
have measured angular distributions of the differential cross sections and of the
three components of the tensor analyzing powers, T20 (θ ) , T21 (θ ) , and T22 (θ ) in the
elastic scattering of 7 MeV deuterons from Mg and Si. A beam of unpolarized
deuterons was polarized by nuclear forces experienced in the elastic scattering and
its tensor analyzing powers were measured using the resonance reaction
3
He(d,p)4He, which served as the polarization analyser. Being a ‘double-scattering’
experiment (cf Chapter 1), the yield of the detected protons was low, and the data
collection time was between 2-30 hours per angle, depending on the scattering angle
from the first target (Mg or Si).
Experimental results were analysed using the optical model procedure in
combination with the Hauser-Feshbach statistical theory. We have found that the
inclusion of the spin-orbit interaction did not result in any significant improvement of
the fits of the differential cross sections. We have also found that a combination of
the central and spin-orbit interactions alone was inadequate in reproducing the
observed tensor analyzing powers angular distributions. Addition of the tensor
interaction TR was necessary to improve the fits to the tensor analyzing powers. The
tensor potential was found to be shallow, attractive and to have a long range. An
inclusion of the second component TL of the tensor interaction turned out to be
unnecessary.
Measurements of vector analyzing power yield useful information about spin-orbit
interaction. Tensor components of nuclear interaction can be studied by measuring
tensor analyzing powers.
References
Brown, L., Christ, H. A. and Rudin, H. 1966, Nucl. Phys. 79:459.
Djaloeis, A. and Nurzynski, J. 1972, Reprot ANU-P/532.
Hauser, W. and Feshbach, H. 1952, Phys. Rev. 87:366.
Hodgson, P. E. and Wilmore, D. 1967, Proc. Phys. Soc. 90:361.
Lauristen, T. and Ajzenberg-Selove, F. 1962, Energy Levels of Light Nuclei: Nuclear
Data Sheet, Printing and Publishing Office, National Academy of Sciences,
Washington 25, DC.
McIntyre, L. C. 1965, PhD Thesis, University of Wisconsin
Welton, T. A. 1963, Fast Neutron Physics Part II, ed. by J.B. Marion and J. L. Fowler,
Interscience Publisher, New York.
Worfenstein, L. 1949, Phys. Rev. 75:1664.
11
Optical Model Potential for Tritons
Key features:
1. I have carried out a detailed examination of the optical model potential for tritons
interacting with a wide-range of target nuclei (40 ≤ A ≤ 207).
2. The analysis shows a preference for the surface absorption potential.
3. The unusually shallow depth of the real central component, V ≈ 25 MeV, for the
volume absorption potential obtained in earlier calculations, has been explained as
being associated with strong attenuation of the radial wave functions inside the
nucleus.
4. With proper care, it is possible to discriminate between nearly equivalent discrete sets
of parameters. Out of the 16 sets found in this analysis only one family was identified
as giving the best fits to the experimental angular distributions using either four- or
six-parameter potentials. This family corresponds to the volume integral JR ≈ 440
MeV·fm or to the real potential depths V ≈ 150 MeV.
3
5. Volume integrals have been found to depend on the mass number of the target
nuclei.
6. Formulae for the mass dependence of the optical model parameters have been
derived. Geometrical parameters depend weakly on the mass number of the target
nucleus and can be fixed.
7. The dependence of the real and imaginary potential depths on the symmetry
parameter ε = ( N − Z ) / A . Clear linear dependence has been found for the four-
and six-parameter potentials. However, the gradient of the relevant functions depends
on the parameterisation of the optical model potential and is most likely dominated by
the dependence on the mass of the target nuclei.
Abstract: Elastic scattering of 20 MeV tritons from target nuclei with A ≥ 40 has been
analysed using four- or six-parameter optical model potentials. A total of 16 parameter
families have been identified and studied. Formulae for the mass dependence have been
derived. Limitations of the conventional optical model are examined. The dependence of the
potential depths on the symmetry parameter ε = ( N − Z ) / A has been investigated.
Introduction
The interaction of tritons with nuclei has not been explored sufficiently well. Acceleration
of tritons is avoided because of the problems with nuclear radiation and consequently
there have not been enough experimental data to support a systematic theoretical
investigation. Tritium half-life is relatively long (12.3 years) and its mobility is high. When
deposited in various places along the beam line (ion source, slits, Faraday cups, etc)
tritium can spread quickly everywhere in the system. The use of tritons as projectiles to
induce nuclear reactions is therefore unwelcome and unpopular.
The most extensive measurements of triton elastic scattering were carried out at 20
MeV incident energy at Los Alamos (Hafele, Flynn, and Blair,1967; Flynn, et al. 1969).
Results of measurements were analysed using Woods-Saxon volume absorption with or
without the addition of a surface-peaked isospin component (Hafele, Flynn, and
Blair,1967; Flynn, et al. 1969; Urone, et al. 1971a).
I have decided to investigate the interaction of tritons with atomic nuclei for three
reasons. First, such information was needed in connection of my study of both (d,t) and
(p,t) reactions. It is well known that the description of reactions involving mass-three
particles depends significantly on the parameterisation of the optical model potential
generating the relevant distorted waves (Baer, et al. 1970; Barnard and Jones 1968a,
1968b) and yet, the interaction in the triton channel was often approximated by using
potentials derived from 3He scattering.
Another reason was the puzzling results in the previous analysis (Hafele, Flynn, and
Blair,1967; Flynn, et al. 1969), which yielded an unusually shallow potential depths of
around 25 MeV for the real central component.
Finally, there was also the question whether there was a preference for any type of a
form factor for the imaginary part of the optical model potential for tritons. Glover and
Jones (1966) carried out an analysis for 12 MeV tritons scattered from light nuclei and
concluded that the quality of fits did not depend on the type of the form factor of the
imaginary component. However, considering the more thoroughly investigated 3He
scattering, and in particular data extending to large angles, it appeared that there was
good evidence in favour of the surface-peaked absorption (Cage, et al. 1972; Chang, et
al. 1973; Fulmer and Hafele 1972; Marchese 1972; Siegel 1971; Urone, et al. 1971a,
1972). Unfortunately, surface absorption has not been studied sufficiently well for tritons.
The analysis
Interaction potential
The main part of this work is a study of the surface absorption interaction. However, I
have also carried out calculations using volume absorption with an aim of trying to
understand why this type of the potential leads to an unusually low value for the
potential depth of the real component.
Thus most of my analysis was done using the following form of the interaction
potential:
d
U (r ) = −Vf (r , r0 , a ) − i 4a′WD f (r , r0′, a′) + VC (r )
dr
where
{ [
f (r , r0 , a ) = 1 + exp (r − r0 A1 / 3 ) / a ]}
−1
For the imaginary component, r0 is replaced by r0′ and a by a′ . VC (r ) is the Coulomb

potential due to a uniformly charged sphere of radius 1.4 A1 / 3 .
For the limited calculations with volume absorption, the form factor
d
4 a′ f (r , r0′, a′)
dr
was replaced by
{ [
f (r , r0′, a′) = 1 + exp (r − r0′A1 / 3 ) / a′ ]} −1
and WD by W.
The calculations were carried out using computer code JIB-3 (Perey 1963), which I
have modified and adapted to run on the ANU UNIVAC-1108 computer.
The fits to the data were optimised by minimising the function χ2 defined as:
2
1 N ⎡σ (θ ) − σ th (θ i ) ⎤
χ = ∑ ⎢ exp i
2
⎥
N i =1 ⎣⎢ Δσ exp (θ i ) ⎦⎥
where N is the number of the experimental data points for a given angular
distribution, σexp(θi) is the experimental differential cross section measured at the
angle θi, σth(θi) is the corresponding theoretical cross section, and Δσexp(θi) is the
error in σexp(θi).
Investigating the unusually shallow potential
The unusually shallow potential ( V ≈ 25MeV ) was reported for the four-parameter
volume-absorption potential, i.e. for the potential with identical geometrical
parameters, r0 = r0′ and a = a′ for the real and imaginary components (Hafele, Flynn,
and Blair,1967; Flynn, et al. 1969.) I have decided to investigate this
parameterisation to try to understand the reasons for such a shallow potential.
Figure 11.1. Example of the grid search for the four-parameter optical model potential with volume
absorption.12 The figure displays the strong continuous ambiguity of the optical model potential. The
plot of the χ function shows that for V ≥ 20 MeV, any value of V gives equivalent fit to the elastic
2
scattering angular distributions as long as remaining parameters follow the indicated lines.
1/ 2
12
J R and r 2 are the volume integral and root-mean-square radius, respectively.
R
First I have carried out an automatic search and I have reproduced the earlier
results. Next, I have carried out a grid search to see whether there are perhaps less
deep minima in the χ2 function, which might have been skipped in the automatic
search. Results of the grid search are shown in Figure 11.1.
The calculations revealed a strong continuous ambiguity extending over a wide
range of values for the real potential depths. The χ2 function displays no clear
minima. Indeed, the function has a constant value for V ≥ 20 MeV but rises sharply
for V ≤ 15 MeV. Any potential depths greater than around 20 MeV will result in
equivalent fits to the experimental angular distributions. This continuous ambiguity
means that there is no way of determining the preferred set of discrete parameters
for the nuclear potential. In this sense, the four-parameter potential gives no useful
information about the triton-nucleus interaction.
These results should be compared with the results based on the identical
procedure using the surface absorption potential (see Figure 11.2).
Figure 11.2. Example of the grid search for the four-parameter optical model potential with the surface
absorption. In contrast with the results presented in Figure 11.1, the volume absorption potential leads
to distinct discrete values of parameter sets corresponding to well-defined minima in the χ2 function.
This figure shows the well-known characteristics of the optical model. Now the χ2
function has a series of clear and distinct minima, which give a well-defined series of
discrete sets of parameters, which optimise the fits to experimental distributions.
It has been shown (Drisko, Satchler and Bassel 1963) that the discrete series of
parameters correspond to a different number of oscillations of partial wave functions
inside the nucleus. I have therefore decided to investigate this aspect and see
whether there are differences between the volume and surface absorption potentials.
Figure 11.3 shows some examples of the calculated wave functions.
The figure shows clear differences in the behaviour of the radial wave functions
inside the nucleus for potentials with either volume or surface absorption. Radial
wave functions for the volume absorption potential are strongly attenuated inside the
nucleus. In contrast, the wave functions for the potential with surface absorption
display clear and only weakly attenuated oscillations.
I have also carried out calculations using six-parameter volume absorption potential.
It can be also seen in Figure 11.3 that for this parameterisation the attenuation of the
wave functions inside the nucleus is much weaker than for the potential with four
parameters. The six-parameter potential leads to the usual discrete sets of
parameters. It is also worth pointing out that the number of oscillations inside the
nucleus depends not just on the discrete set of parameters as pointed out by Drisko,
Satchler and Bassel (1963) but also on the L value of the radial wave function.
Figure 11.3. Examples of the radial wave functions |uL(r)| calculated for sets of parameters
corresponding to V = 115 MeV for the four-parameter potentials with volume or surface absorptions
(the left- and right-hand side of the figure, respectively). The parameter sets are as shown in Figures
11.1 and 11.2. Two examples of radial wave functions for the six-parameter volume absorption
potential are also shown on the left-hand side of the figure.
Fits to the experimental angular distributions

The remaining calculations have been carried out using only surface absorption
potentials. In these calculations I have identified 9 sets of discrete sets of parameters
for the four-parameter potential and 7 sets for the six-parameter potential. They all
give acceptable fits to the experimental angular distributions.
Examples of fits obtained using set 4 (i.e. with V ≈ 150 MeV) for the four-parameter
potential with surface absorption are shown in Figure 11.4. Excellent fits are
obtained when all four parameters are allowed to vary. However, calculations with
fixed geometry give also satisfactory results.
Six parameter potentials give nearly identical results. The quality of fits cannot be
distinguished visually but only by the minima in the χ2 function, which are slightly
deeper than for the four-parameter potentials.
Figure 11.4. Examples of fits to the 20 MeV triton data obtained using the four-parameter optical
model potential with surface absorption. Results of the calculations represented by thicker lines were
obtained by searching for all parameters. Results represented by thinner lines are for calculations with
fixed geometrical parameters r0 = 1.20 fm, and a = 0.755 fm. Both sets of calculations are for V ≈ 150
MeV.
Selecting the best set of parameters
The next step in the analysis was to try to discriminate between various sets of
parameters and see whether the number of sets can be reduced. Visual examination
of the fits was ineffective and examination of individual minima of the χ2 for all the
angular distributions and parameter sets was too tedious. (The total number of
individual minima was 240.) Accordingly, I have averaged the minimum values of the
χ2 function for all target nuclei for each set of parameters. Results are presented in
Figure 11.5. Each family is identified not only by the family number but also by the
approximate value of the volume integral JR calculated using the following expression:
4πVR 3 ⎡ ⎛ πa ⎞ ⎤
2
JR = ⎢1 + ⎜ ⎟ ⎥
3 AAi ⎢⎣ ⎝ R ⎠ ⎥⎦
where Ai is the atomic mass number of the projectile.
Figure 11.5. Plots of the minima of the χ2 function averaged over the atomic mass of the target nuclei
calculated for each set (family) of the optical model parameters. Only sets with no fixed parameters
were used in this calculation. This figure shows that by a careful study of the minima in the χ2 function
it is possible to eliminate many of the possible discrete sets of parameters and select only one or two
sets (families). In my analysis, the preferable families of parameters are 3 and 4 for the 4-parameter
potential, and 4 and 5 for the 6-parameter potential. The best common set is the family 4, which
corresponds to JR ≈ 440 MeV·fm3 or V ≈ 150 MeV.
It can be seen that the number of acceptable families can be limited to two for the four-
parameter potential. The best families are 3 and 4 corresponding JR ≈ 360 and 440
MeV·fm3 or to V ≈ 100 and 150 MeV. For the six-parameter potential, the best sets are
4 and 5, but families 3 and 6 give also acceptable fits. The best common family is the
family 4 corresponding to JR ≈ 440 MeV•fm3 or V ≈ 150 MeV.
The problem of resolving discrete optical model ambiguities for 3He particles has
been discussed in a number of publications. In the analysis of 33 and 53 MeV 3He
scattering on 57Fe (Marchese 1973) the number of acceptable families was reduced to
two, corresponding to JR ≈ 330 and ≈ 440 MeV·fm3. Most analyses suggested the
preference for the JR ≈ 450 MeV · fm3 family but some calculations (Chang, et al. 1973;
Fulmer and Hafele 1972; Fulmer and Hafele 1973a, 1972b) carried out for data collected
at sufficiently high energy and in a wide range of angles revealed a unique family with JR
≈ 330 MeV · fm3. Only one set of parameters was also found in the analysis of 217 MeV
3
He particles scattered from targets ranging from 6Li to 208Pb (Willis et al. 1973), namely a
set corresponding to JR ≈ 255 MeV · fm3.
The analysis of 139 MeV α-particle scattering on 12C yielded a unique set corresponding to
JR ≈ 353 MeV · fm3 (Smith et al. 1973). Similar results were also reported for α-
particle scattering on 24Mg (Duhm 1968; Sinnh et al. 1969; Yang et al. 1973) and on 90Zr
(Paans, Put and Malfliet 1973) isotopes.
In general, results obtained so far seem to show some preference for the JR ≈ 360
MeV · fm3 family of parameters. However, parameter sets I have identified in this work
were also applied in my study of (p,t) reactions on a range of Se isotopes at 33 MeV
proton energy. Significantly better fits to the reaction distributions were obtained using
sets corresponding to JR ≈ 450 MeV · fm3 (i.e. with V ≈ 150 MeV) rather than to JR ≈
350 MeV · fm3 (i.e. with V ≈ 100 MeV). Thus the analysis of our (p,t) results confirmed my
conclusion based on the analysis of the (t,t) elastic scattering as discussed above.
The mass dependence
In fitting nuclear reaction angular distribution it is often desirable to know the
dependence of the optical model parameters on the mass number A of the target nucleus.
Parameters derived here displayed an approximately linear dependence on A. Tables
11.1 and 11.2 show the formulae obtained by fitting linear functions to the
parameters determined from the optical model analysis. The examples are for the
families 3 and 4.
Table 11.1
Mass dependence of the optical model parameters for the four-parameter potential with surface
absorption
Table 11.2
Mass dependence of the optical model parameters of the six-parameter potential with surface
absorption
It can be seen that geometrical parameters depend only weakly on A. Furthermore, in

contrast with the suggestion of Marchese et al. (1972), JR is not constant. For sets with
no fixed parameters, JR decreases at the rate of ≈ 0.4 MeV · fm3 per a.m.u. for the
family 4 and at ≈ 0.1 MeV · fm3 per a.m.u. for the family 3. The 0.5 MeV · fm3 rate was
reported for JR ≈ 330 MeV · fm3 for 3He particles (Fulmer and Hafele 1973a, 1973b).
Figure 11.6 shows the dependence of JR on A for various families with no fixed
parameters. The gradient dJR/dA varies between families. The difference between the
adjacent values of JR is about 80 MeV · fm3 for A ≈ 120 and it decreases with the
increasing mass of the target nucleus.
Figure 11.6. Examples of the dependence of the volume integral, JR, on the atomic mass number A of
the target nuclei. The family index is used to label separate groups of points. The lines are drawn as a
visual aid. The displayed points were calculated for sets with no fixed parameters.
The symmetry term for tritons
It has been long recognised that the depth of the real part of the optical model
potential depends on a symmetry parameter ε = ( N − Z ) / A (Green and Sood 1958; Lane
1958). Lane (1962a, 1962b) has shown that this dependence might be related to an
isospin term in the optical model potential. If the potential is calculated as an average sum
of the two-body forces then the real part of the optical model potential can be expressed
as:
t⋅T
V = V0 + V1
A
where t and T are the isotopic spins of the incident particle and the target nucleus,
respectively.
By averaging this potential over all allowed values of the total isotopic spin the mean
optical model potential can be written in terms of the symmetry parameter ε = ( N − Z ) / A .
For instance, for protons:
1 N −Z
V = V0 − V1
4 A
The early evidence for such dependence was provided by comparing neutron and
proton scattering (Melkanoff 1956; Melkanoff, Nodvik and Saxon 1957) and by shell
model calculations (Green 1956; Ross, Mark and Lawson 1956; Ross, Lawson and
Mark 1956). It has been also suggested that the symmetry term is complex (Satchler
1967).
Attempts have been made to determine the t ⋅ T interaction for mass three projectiles
using elastic scattering (Becchetti and Greenlees 1971; Drisko et al. 1967; Flynn at al.
1969; Hafele et al. 1967; Urone et al. 1972). The common feature of all the previous
analyses was the presence of a volume absorption term in the optical model potential.
The results are inconclusive and do not form a consistent pattern.
Most analyses seem to indicate zero or weak dependence of the real component on
the symmetry parameter, although a strong symmetry term was reported by Becchetti
and Greenlees (1971) for 3He particles.
Comparison of 3He and triton angular distributions resulted in an imaginary symmetry
term similar to that employed in analyses of the (3He, t) reaction (Urone et al. 1971).
The result was, however, obtained by averaging over a wide range of values differing
by as much as a factor of about 30. It is also worth noting that, owing to the possible
complex multistep mechanism (Schaeffer and Bertsch 1972; Schaeffer and
Glendenning 1973; Toyama 1972), the usual DWBA analysis of the reaction (3He, t) can
hardly serve as a point of reference in assessing the magnitude of the symmetry term.
The apparent agreement is therefore most probably accidental.
Considering previous attempts based usually on volume absorption potentials, it was
interesting to see whether surface absorption would produce better results. For
consistency with the previously published studies I have adopted the recommended
method of deriving the symmetry potential from the elastic scattering (Satchler 1969).
In this method potentials corresponding to a fixed average geometry are used and
compared with a linear dependence on the symmetry parameter ε = ( N − Z ) / A . Plots of
V and WD for the two families of parameters are shown in Figure 11.7. The straight lines
represent the least-squares fits of the functions:
V = V0 + εV1
WD = W0 − εW1
The factor 1/4 is now absorbed in V1 and W1. The relevant numerical values of V0, V1, W0,
and W1 are listed in Figure 11.7.
Figure 11.7. The dependence of the real and imaginary potential depths, V and WD, on the symmetry
parameter ε = ( N − Z ) / A for the elastic scattering of 20 MeV tritons. The potential depths
correspond to a fixed averaged geometry r0 = 1.20 fm, a = 0.755 fm for the four-parameter potential and
r0 = 1.23 fm, a = 0.72 fm, r0′ = 1.15 fm, a' = 0.85 fm for the six-parameter potential. The straight lines
are the least-squares fits of the functions V = V0 + εV1 and WD = W0 − εW1 to the relevant points. The
numerical values of (V0 ,V1 ) and (W0 ,W1 ) in MeV are displayed on the left-hand side of each fitted
line, and the family index on the right-hand side. The optical model with surface absorption was used
in these calculations.
The figure shows clear linear dependence of both V and WD on the symmetry term ε.
However, the derived values V1 and W1 do not appear to represent the strengths of the
isospin potential because they depend on the parameterisation of the optical model
potential.
This simple method of determining isospin interaction, though used in the past, have been
criticised by many authors (see for instance Hodgson 1971 and referenced therein). It has
been pointed out that not only the depths of nuclear potentials but also geometrical
parameters may depend on the isospin interaction, and that it is either difficult or
impossible to separate geometrical effects from the purely isospin effects. Isospin
interaction may have different components, which are neglected in this simple linear
representation. The assumption that nuclear radius is proportional to A1 / 3 is also
questionable. Isospin interaction depends critically on the relative distribution of protons
and neutrons, which is neglected in the simple description given by the optical model.
There could be also nuclear structure effects that are not accounted for in these simple
linear representations of the dependence of potential depths on the symmetry parameter
ε.
The average symmetry potentials found in my analysis are V1 = 88.8 MeV and W1=
50.4 MeV.

A detailed study of the optical model description of the elastic scattering of tritons at
20 MeV incident energy has been carried out. In this work I have concentrated on
studying the surface absorption potential. However, I have also carried out some
calculations with volume absorption.
I have shown that the unusually shallow potential reported for the four-parameter
optical model potential with volume absorption is associated with a strong
attenuation of radial wave functions inside the nucleus. Surface absorption potential
gives better description of triton-nucleus interaction.
The analysis yielded nine sets of parameters for the four-parameter potential and
seven for the six-parameter potential, or a total of 16 sets. By comparing the
averaged minima of the χ2 functions it was possible to reduce the number of
acceptable parameter sets to only two for the four-parameter potential and 2-4 for
the six-parameter potential. Only one group of parameters results in the best fits for
both types of the optical potential. This group corresponds to the volume integral JR ≈
440 MeV·fm3 or to V ≈ 150 MeV.
In general, volume integral decreases with the mass number. However, for one group
of parameters JR is almost constant. This group corresponds to JR ≈ 360 MeV·fm3 or to
V ≈ 100 MeV and it gives almost identical fits to the experimental angular distributions
as the group corresponding to V ≈ 150 MeV.
Formulae for the mass dependence of the optical model parameters have been
derived. They show that the dependence of the geometrical parameters on the mass of
the target nucleus is weak and consequently, these parameters can be fixed.
A study of the dependence of V and WD on the symmetry term ε = ( N − Z ) / A has been
studied. In contrast to most previous analyses for mass-3 projectiles, linear dependence
on ε has been demonstrated. However, the gradient of the linear functions depends on
the parameterisation of the optical model. The observed linear dependence on the
symmetry term represents most likely mainly the mass dependence of the optical model
parameters.
References
Baer, H. W., Kraushaar, J. J., Moss, C. E., King, N. S. P. and Green, R. E. L. 1970,
Phys. Rev. Lett. 25:1035.
Barnard, R. W. and Jones, G. D. 1968a, Nucl. Phys. A106:497.
Barnard, R. W. and Jones, G. D. 1968b, Nucl. Phys. A108:641.
Cage, M. E,m Clough, D. L., Cole, A. J., England, J. B. A., Pyle, G. J., Rolph, P. M.,
Watson, L. H. and Worledge, D. H. 1972, Nucl. Phys. A183:449
Chang, H. H. et al, 1973, University of Colorado, Nuclear Physics Laboratory Technical Report
COO-535-693:14
Drisko, R. M., Satchler, G. R. and Bassel, R. H. 1963, Phys. Lett. 5:347.
Drisko et al, R. M. 1967, Proc. Int. Conf. on Nuclear Structure, Tokyo, p. 347.
Duhm, H. H. 1968, Nucl. Phys. A118:563.
Flynn, E. R., Armstrong, D. D., Beery, J. G. and Blair, A. G.1969, Phys. Rev. 182:1113.
Fulmer, C. B. and Hafele, J. C. 1972, Phys. Rev. C5:1969.
Fulmer, C. B. and Hafele, J. C. 1973a, Phys. Rev. C7:631;
Fulmer, C. B. and Hafele, J. C. 1973a, Phys. Rev. C8:172
Glover, R. N. and Jones, A. D. 1966, Nucl. Phys. 81:268.
Green, A. E. S. 1956, Phys. Rev. 102:1325
Green A. E. S. and Sood, P. C. 1958, Phys. Rev. 111:1147
Hafele, J. C., Flynn, E. R. and Blair, A. G. 1967, Phys. Rev. 155:1238.
Lane, A. M. 1958, Proc. Int. Conf. on Nuclear Physics, Paris, p. 32.
Lane, A. M. 1962a, Phys. Rev. Lett. 8:171.
Lane, A. M. 1962b, Nucl. Phys. 35:676.
Marchese, C. J. 1972, Nucl. Phys. A191:627.
Marchese, C. J., Clarke, N. M. and Griffiths, R. J. 1973, Nucl. Phys. A202:421.
Melkanoff, M. A., Moszkowski, S. A., Nodvik, J. and Saxon, D. S. 1956, Phys. Rev.
101:507
Melkanoff, M. A., Nodvik, J. and Saxon, D. S. 1957, Phys. Rev. 106:793
Hodgson, P. E. 1971, Nuclear Reactions and Nuclear Structure, Clarendon Press, Oxford.
Paans, A. M. J., Put, L. W. and Malfliet, R. A. R. L. 1973, Proc. Int. Conf. Nuclear Physics, Munich,
ed. J. de Boer and H. J. Mang (North-Holland, Amsterdam,) p. 340
Perey, F. G. 1963, Phys. Rev. 131:745.
Ross, A. A., Mark, H. and Lawson, R. D. 1956, Phys. Rev. 102:1613.
Ross, A. A. Lawson, R. D. and Mark, H. 1956, Phys. Rev. 104:401.
Satchler, G. R. 1967, Nucl. Phys. A91:75.
Satchler, G. R. 1969, Isospin in Nuclear Physics, ed. D. H. Wilkinson (North-Holland, Amsterdam,)
p. 389.
Schaeffer, R. and Bertsch, G. F. 1972, Phys. Lett. 38B:159.
Schaeffer, R. and Glendenning, N. K. 1973, Nucl. Phys. A207:321.
Siegel, G. R. 1971, Ph.D. thesis, Washington University, St. Louis.
Singh, P. P., Malmin, R. E., High, M. and Devins, D. W. 1969, Phys. Rev. Lett. 23:1124.
Smith, S. M., Tebell, G., Cowley, A. A., Goldberg, D. A., Pugh, H. G., Reichart, W. and
Wall, N. S. 1973, Nucl. Phys. A207:273.
Toyama, M. 1972, Phys. Lett. 38B:147.
Urone, P. P., Put, L. W., Chang, H. H. and Ridley, B. W. 1971a, Nucl. Phys. A163:225.
Urone, P. P., Put, L. W., Ridley, B. W. and Jones, G. D. 1971b, Nucl. Phys. A167:1383.
Urone, P. P., Put, L. W., Chang, H. H. and Ridley, B. W. 1972, Nucl. Phys. A186:344.
Willis, N., Brissaud, I., Le Bornec, Y., Tatischeff, B. and Duhamel, G. 1973, Nucl. Phys.
A204:454.
Yang et al, G. G. 1973, Proc. Int. Conf. Nuclear Physics, Munich, ed. J. de Boer and H. J. Mang
(North-Holland, Amsterdam,) p. 338
12
The 54Cr(d,t)53Cr and 67,68Zn(d,t)66,67Zn Reactions at 12 MeV
Key features:
1. We have measured a total of 35 angular distributions for the neutron pickup reaction
induced by 12 MeV deuterons: 5 for the 54Cr(d,t)53Cr reaction, 13 for 67Zn(d,t)66Zn and
17 for 68Zn(d,t)67Zn.
2. Particle identification was done by using a ΔE-E particle identification technique.
3. We have carried out theoretical analysis of the experimental differential cross sections
using direct reactions theory and a computer code, which I have modified and
adapted to run on an ANU computer.
4. We have demonstrated j-dependence for l = 1 and 3 transitions. However, the
observed j-dependence could not be reproduced theoretically even if spin-orbit
potentials were used in the calculations for deuterons and tritons.
5. We have assigned spins and parities to states in residual nuclei and extracted
spectroscopic factors.
6. We have extracted information about configuration mixing and compared it with
theoretical calculations using pairing theory and a computer code I have written for
this purpose.
Abstract: Angular distributions for the (d,t) reaction on 54Cr and 67,68Zn target nuclei have been
measured using 12 MeV deuterons. Experimental results were analysed theoretically using
direct nuclear reactions theory. Spin assignments have been made to states in residual nuclei
and spectroscopic factors have been extracted. The j-dependence has been observed but it
could not be reproduced theoretically. Configuration mixing for the f-p shell has been studied
using experimental spectroscopic factors. Results are compared with predictions of the
pairing theory.
Introduction
The nuclei 54Cr, 67Zn, and 68Zn have 2, 6, and 10 neutrons outside the closed N = 28
shell. I have chosen these nuclei to extend the study of nuclear spectroscopy in the
f-p shell. Without residual a interaction, the 2p3/2 orbit should be half full in the
ground state of 54Cr; there should be one 1f5/2 neutron hole in 67Zn; and orbits 2p3/2
and 1f5/2 should be full in 68Zn. Residual interaction causes configuration mixing and
one should expect to observe states belonging to configurations 2p3/2, 1f5/2, 2p1/2 and
possibly even to 1g1/2 for all these target nuclei.
The (d,t) neutron pickup reaction has not been explored well enough in this region.
Fulmer and Daehnick (1964, 1965) examined the l =1 transitions in 55Fe, 59Ni, and
63
Ni nuclei using the (d,t) reactions and reported a deep minimum at backward
angles for j = 1/2 similar to that observed in (d,p) reactions. They also observed
some indication of differences at forward angles between two j values for l = 3
transitions in 59Ni but their measurements for this transition did not extend beyond
about 600. If confirmed, this feature could serve as a tool in assigning not only orbital
angular momenta to excited states but also the spin values. There was also no
previous attempt to reproduce theoretically the observed (d,t) j-dependence.
Measurements of angular distributions were carried out using the ANU EN tandem
accelerator. The beam intensity on the target varied between 15 nA and 350 nA. The
experimental arrangement was similar to that used in the 26Mg(3He,α)25Mg
measurements (see Chapter 5). However, in order to separate triton groups from
deuterons and protons and thus to obtain clear triton spectra, ΔE-E detector
telescope assemblies, made of silicon surface barrier detectors, were used. Care
has been taken to select suitable thickness of the ΔE detectors to optimise the mass
resolution. The ΔE detectors were 40-100 μm thick and E detectors 2 μm thick. To
ensure that the ΔE detectors were fully depleted, a bias voltage of at least 10%
higher than required for the total depletion was applied. We have used three such
telescope assemblies in our measurements. The electronics of the particle
identification system is shown in Figure 12.1.
Figure 12.1. The electronics diagram of the particle identification system used in the measurements of
angular distributions for the (d,t) reaction at 12 MeV incident deuteron energy. Preamp – Preamplifier;
Amp – Amplifier; Sumamp – Summing Amplifier; Timing SCA – Timing Single Channel Analyser;
Dalayamp – Delay Amplifier; Particle Ident. – Particle Identifier; Gate – Linear Gate and Slow
Coincidence; Biasedamp – Biased Amplifier; Mixer – Mixer and Routing Unit; ADC – Analogue-to-
Digital Converter; IBM 1800 – IBM 1800 computer.
The output pulses from the particle identifier, representing the mass spectrum, were
fed into a timing single channel analyser (TSCA). The threshold levels of the TSCA
were set so that only the triton pulses were allowed to pass through a linear gate and
slow coincidence unit. Pulses from the linear gates for all three particle telescopes
were fed into a mixer and routing system. The mixed unipolar pulses were then fed
into an analogue-to-digital converter, which was interfaced with the IBM 1800
computer. The total energy pulses were separated according to their associated
routing signals and stored in different memory locations in the computer. A typical
particle spectrum is shown in Figure 12.2.
The targets were about 100 μg/cm2 thick. They were supported by an approximately
30 μg/cm2 carbon backing. Measurements of the angular distributions were carried
out for a wide range of angles, from 12.50 to 1500 in steps of 2.50.
A total of 35 angular distributions have been measured: 5 for the 54Cr(d,t)53Cr
reaction, 13 for 67Zn(d,t)66Zn and 17 for 68Zn(d,t)67Zn. They are presented in Figures
12.4 to 12.8.
54
Figure 12.2. A typical mass spectrum for the interaction of the 12 MeV deuterons with Cr leading to
the (d,d’), (d,p), and (d,t) reactions.
Figure 12.3. An example of a triton spectrum.
Theoretical analysis of the data was carried out using the computer code DWUCK
(Kunz 1966), which I have modified and adapted to run on the Australian National
University UNIVAC 1108 computer. This program allows for the calculation of the
finite range and non-locality corrections (see the Appendix E).
The optical model potential was made of the central part with the surface absorption
(see Chapter 5). Deuteron parameters (i.e. the parameters for the incident channel)
were derived from the mass dependent formulae of Perey and Perey (1963). The
triton potential (for the outgoing channel) was based on my analysis of the 20 MeV
triton scattering (see Chapter 11). The potential depths V and WD were adjusted
using the gradients of dV / dE = -0.15 and dWD / dE = -0.50 determined from 3He
scattering (Chang and Ridley 1971) to match the energy of tritons from the (d,t)
reactions at 12 MeV.
Both deuteron and triton potentials did not contain the spin-orbit interactions. We
have carried out calculations with spin-orbit interaction in the incident and outgoing
channels but its effect on the calculated angular distributions was negligible. In some
cases spin-orbit interaction resulted in making the fits worse.
Theoretical calculations are compared with experimental results in Figures 12.4 to
12.8. The corresponding spectroscopic information is summarised in Tables 12.1 to
12.3. They contain spin and parity assignments and experimentally extracted
spectroscopic factors.
54
Figure 12.4. Angular distributions for the reaction Cr(d,t)53Cr at 12 MeV deuteron energy compared
with theoretical calculations.
67
Figure 12.5. Angular distributions for the reaction Zn(d,t)66Zn at 12 MeV deuteron energy compared
67
Figure 12.6. Angular distributions for higher excited states for the reaction Zn(d,t)66Zn at 12 MeV
deuteron energy compared with theoretical calculations.
68
Figure 12.7. Angular distributions for the reaction Zn(d,t)67Zn at 12 MeV deuteron energy compared
68
Figure 12.8. Angular distributions for higher excited states for the reaction Zn(d,t)67Zn at 12 MeV
deuteron energy compared with theoretical calculations.
Table 12.1
Summary of the results for the single neutron pickup reactions to states in 53Cr
Ex Jπ (d,t) a) (d,t) b) (p,d) c) (3He,α) d)
(MeV) l j S S S S
1 3/2 0.61 0.83 1.10
1/2 0.24 0.22 0.31 0.26
1.006 5/2- 3 5/2 0.54 0.31 0.51 0.49
1.287 7/2- 3 7/2 0.68 0.61 0.70 0.45
1.539 7/2- 3 7/2 2.30 1.80 3.20 3.00
π
Ex – Excitation energy of levels in 53Cr J – Adopted spins and parities.
l – the angular momentum of transferred neutron.
j – the spin of transferred neutron (j = l ± 1/2.) S – spectroscopic factors.
a) Our results. b) Fitz et al. 1967. c) Whitten 1967. d) David et al. 1969.
Table 12.2
Summary of the results for the single neutron pickup reactions to states in 66Zn
Ex Jπ (d,t) (p,d)
(MeV) l S l S
0.000 0+ 3 0.27 3 0.44
1.039 2+ 1 0.10 1 0.10
1.873 2+ 1 0.01 1 0.01
2.450 4+ (1) 0.02 (1) 0.02
2.704 (2,3)- (0) 0.06
2.781 (1,2)+ 1 0.55 1 0.52
2.941 4+ 3 0.22 3 0.47
3.080 1 0.06 1 0.10
3.229 1+ 1 0.12 1 0.15
3.332 (1,2)+ 1 0.10 1 0.14
3.502 1 0.39 1 0.30
3.680 1 0.28 1 0.30
3.791 1+ 1 0.56 1 0.80
Ex – Excitation energy of levels in 53Cr. Jπ– Adopted or possible spins and parities.
l – the angular momentum of transferred neutron. S – spectroscopic factors.
(d,t) – Our results. (p,d) – McIntyre 1966.
Table 12.3
Summary of the results for the single neutron pickup reactions to states in 67Zn
Ex Jπ (d,t) (p,d)
(MeV) l j S l j S
0.000 5/2- 3 5/2 3.87 3 5/2 3.80
0.093 1/2- 1 1/2 0.58 1 1/2 0.40
0.185 3/2- 1 3/2 0.20 1 3/2 0.19
0.394 3/2- 1 3/2 1.90 1 3/2 1.76
0.602 9/2+ 4 9/2 0.88 (4) 0.90
-
0.888 3/2 1 3/2 0.03 (1) 0.04
+
0.978 (5/2 ) (0,2) 0.39 a) (1) 0.06
1.142 1/2- 1 1/2 0.19 1 1/2 0.18
a
1.370 (2,3) 0.02 )
1.444 3/2- 1 3/2 0.05
1.542 (3/2-) (1,2) (3/2) 0.03 b)
+
1.676 1/2 0 1/2 0.32
1.808 (1/2+) (0) (1/2) 0.19
1.842 (3/2-) 1 (3/2) 0.05
2.100 (3) 0.42
2.172 (1/2+) (0) (1/2) 0.10
2.246 (1/2+) (0) (1/2) 0.28
Ex – Excitation energy of levels in 67Zn. Jπ– Probable spins and parities.
l – the angular momentum of transferred neutron.
j – the spin of transferred neutron (j = l ± 1/2.) S – spectroscopic factors.
(d,t) – Our results. (p,d) – McIntyre 1966. a) For l = 2. b) For l = 1.
The j-dependence
I have selected the nucleus 54Cr for our study because it is a good candidate for
studying the j-dependence in the (d,t) reactions. Spins and parities of the low-energy
states in 53Cr are well known and they belong to both l = 1 and 3 orbital angular
momenta.
Figure 12.9. Experimentally observed j - dependence for the 54Cr(d,t)53Cr reaction for l = 1 and 3
transitions. (Compare the differential cross sections at backward angles.)
The ground state and the first excited state have spins 3/2 and 1/2, respectively,
belonging to l = 1, whereas the third and the fourth excited states have spins 5/2 and
7
/2 belonging to l = 3. The fourth excited state also has spin 7/2. Thus, the reaction
54
Cr(d,t)53Cr offers a good opportunity to study the j-dependence for both l = 1 and 3
transitions.
Figure 12.9 compares angular distributions for l = 1 and 3. It can be seen that for l =
1, the last maximum and minimum for the 3/2 angular distribution is shifted to larger
angles for the 1/2 distribution. For l = 3, the distributions corresponding to j = 7/2 have
more pronounced structure at large reaction angles than the distribution
corresponding to j = 5/2. However, these experimentally observed signatures were
not reproduced theoretically even when spin-orbit potentials were used for deuterons
and tritons.
Configuration mixing
Using spectroscopic factors for the pickup reactions one can calculate configuration
mixing in the ground state wave function of the target nucleus caused by a residual
interaction. Experimentally, occupation numbers, V j2 , and the centre-of-gravity
energies, E j , are given by:
∑ S (l , j )
i
V =
2 i
2 j +1
j
∑ S (l , j ) E ( j )
i
(i )
x
Ej = i
∑ S (l , j )
i
i
where S i (l , j ) is the spectroscopic factor for (l , j ) pickup to state i, and E x(i ) ( j ) is the
excitation energy of the ith state of the residual nucleus with the total angular
momentum j.
The sum is over all states belonging to (l , j ) configuration. The sum of all
spectroscopic factors should be equal to the number of neutrons outside the closed
shell.
n = ∑ Si (l , j )
ij
Measurements for the 68Zn(d,t)67Zn reaction show clearly that the ground state of
68
Zn is a mixture of all four configurations, 2p3/2, 1f5/2, 2p1/2, and 1g9/2, located outside
the N = 28 closed neutron shells.
Experimentally determined occupation numbers, V j2 , and centre-of-gravity single
particle energies, E j , are listed in Table 12.4. They are compared with theoretical
values calculated using the pairing theory of Kisslinger and Sorensen (1960, 1963)
and a computer code I have written for this purpose.
According to this theory
1 ⎧⎪ ε j −λ ⎫
⎪
V j2 = ⎨1 − 1/ 2 ⎬
2⎪
⎩ [
(ε j − λ )2 + Δ2 ] ⎪⎭
[
E j = (ε j − λ ) + Δ2
2
]
1/ 2
−Δ
where εj. λ, and Δ are the single particle energies, the chemical potential and the gap
parameter, respectively, all defined by Kisslinger and Sorensen (1960, 1963). The
single particle energies are calculated using their relations. The parameters λ, and Δ
were determined by solving the gap equations of Kisslinger and Sorensen (1960,
1963).
Table 12.4
68
Occupation numbers and centre-of-gravity energies for ground state of Zn
lj εj n jl n′jl V j2 Ej
Exp. Theory Exp. a) Theory

2p3/2 -0.043 2.76 4.00 0.69 0.85 0.44 0.58
1f5/2 0.371 5.23 6.00 0.87 0.79 0.21 0.32
2p1/2 2.742 0.94 0.00 0.47 0.16 0.35 0.54
1g9/2 2.758 1.07 0.00 0.11 0.10 0.60 0.55
j – single particle configuration.
εj – calculated single particle energies.
n jl – The experimentally determined average number of neutrons in the configuration jl.
n′jl – The number of neutrons expected in the absence of the residual interaction.
Exp. – experimental values as derived from the 68Zn(d,t)67Zn reaction at 12 MeV.
Theory – theoretical values calculated using the pairing theory of Kisslinger and
Sorensen (1960, 1963) and my computer code.
Results of our study show that the population of the configuration 1f5/2 is close to the
expected value of 6. However, the population of the 2p1/2 configuration is significantly
higher than that predicted by the pairing theory and the population of 2p3/2 slightly
lower.
About 52% of all neutrons outside the closed shell N = 28 occupy the 1f5/2 orbit,
which is close to the value of 60% that would be expected if there were no residual
interaction. However, the orbit p1/2, which would have been empty without residual
interaction, is now nearly 50% full. Even the remote orbit 1g9/2 has 11% of its allowed
population filled in. Both orbits, 2p1/2 and 1g9/2, are populated mainly at the expense
of the 2p3/2 configuration, which donates 31% of its neutrons to other orbits. The 1f5/2
donates only 13% of its neutrons.
We have measured a total of 35 angular distributions of the differential cross
sections for neutron pickup (d,t) reaction using 12 MeV deuterons and target nuclei
belonging the f-p shell (54Cr, 67,68Zn). We have carried out theoretical analysis,
extracted spectroscopic factors, and assigned spins and parities to the
corresponding states in the final nuclei.
We have observed j - dependence for both l = 1 and 3 orbital angular momenta.
However, the experimentally observed j - dependence could not be reproduced
using the direct reaction theory even when spin-orbit interaction was included in the
calculations in the incident and outgoing channels.
Using the experimentally determined spectroscopic factors we have calculated the
occupations numbers for orbits in the ground state of 68Zn. In the absence of a
residual interaction, neutrons would occupy only configurations 2p3/2 and 1f5/2.
However, we have found that orbits 2p3/2 and 1f5/2 are only 69% and 87% full,
respectively. Neutrons located outside the closed N = 28 shell spend also considered
time in orbits 2p1/2 and 1g9/2. These orbits are 47% and 11% full, respectively.
References
Chang, H. H. and Ridley, B. W. 1971, University of Colorado Report, COO-535-635:55.
David, P., Duhm, H. H., Bock, R. and Stock, R. 1969, Nucl. Phys. A128:47
Fitz, W., Hegar, R., Jahr, R. and Santo., R. 1967, Z. Phys. 202:109.
Fulmer, R. H. and Daehnick, W. W. 1964, Phys. Rev. Lett. 12:455.
Fulmer, R. H. and Daehnick, W. W. 1965, Phys. Rev. B139:579.
Kisslinger, L. S. and Sorensen, R. A. 1960, Kgl. Danske Videskab. Selskab. Mat. Fys.
Medd. 32, No. 9.
Kisslinger, L. S. and Sorensen, R. A. 1963, Rev. Mod. Phys. 35:853.
Kunz, P. D. 1966, University of Colorado Report COO-536-606.
McIntyre, L. C. 1966, Phys. Rev. 152:1013.
Whitten, Jr, C. A. 1967, Phys. Rev. 156:1228.
13
A Study of the 76,78Se(p,t)74,76Se Reactions at Ep = 33 MeV
Key features:
1. We have observed a total of 53 states in 74Se and 76Se (28 states in 74
Se and 25 in
76
Se) and we have assigned excitations energies to all of them.
2. We have found that the (p,t) reaction is strongly selective – it leads mainly to the
ground states and the first excited states in the residual nuclei. Other states are
excited with significantly smaller intensity.
3. We have measured a total of 38 angular distributions of the differential cross sections
for the reactions 76,78Se(p,t)74,76Se. Measurements were carried out using a ΔE-E
particle identification technique.
4. We have analysed the angular distributions using direct reactions theory.
5. We have assigned a total of 35 spin-parity values to states in 74Se and 76Se.
6. We have found that the calculated absolute values of the differential cross sections
depend strongly on the assumed transfer configuration. We conclude that better
description of the ground state wave functions of the target nuclei is needed to yield
more reliable values of the theoretical spectroscopic factors.
Abstract: The 76,78Se(p,t)74,76Se reactions have been studied at a proton energy of 33 MeV
using detector telescopes and particle identification techniques. Twenty-eight states up to the
excitation energy of 4.64 MeV in 74Se and twenty-five states up to 4.43 MeV in 76Se were
observed. Angular distributions were measured for many of these states in the range of 15°-
90° and the results were compared with the distorted waves direct transfer calculations. Many
Jπ assignments were made on the basis of the theoretical analysis of the data and a
comparison of the angular distributions with empirical shapes for transitions to states with
well-known Jπ. Enhancement coefficients were calculated for the simple two-neutron pickup
configurations. Results indicate that the (p,t) reaction is sensitive to assumed configurations in
the ground state wave functions of the target nuclei and that better description of the relevant
wave functions is needed to yield more reliable values for the theoretical spectroscopic
factors.
Introduction
The study of the (p,t) reaction was carried out in conjunction with the study of the
(p,d) reaction on Se isotopes (see Chapter 14). Only slight modification of the
electronic system was necessary to measure angular distribution simultaneously for
both reactions.
The 76Se and 78Se isotopes have 14 and 16 neutrons outside the closed shell N =
28, respectively. In the absence of the residual interaction, these neutrons would fill
in the orbits 2p3/2, 1f5/2, and 2p1/2, and there would be 2 neutrons in the 1g9/2 orbit in
76
Se and 4 in 78Se.
Nuclei with neutrons in the 1g9/2 region have long resisted a complete and reliable
description in terms of any one model. The large number of valence neutrons and
available orbits complicate the calculations in this region. There was also not a great
deal of experimental information available on many of these nuclei at the time of our
study.
One of the aims of our study was to provide information on the low-spin states in 74Se
and 76Se. Investigation of the single-step direct transfer mechanism to reproduce two-
neutron transfer data in this area was also of interest. It was also interesting to see
whether the existing simple assumptions about the configurations in the ground state
wave functions could result in reliable predictions of the absolute values of the differential
cross sections for these reactions.
These reactions served also as a good test of the of the optical model parameterisation
for tritons as discussed in Chapter 11. We have used my sets of parameters successfully
in the analysis of our results for the (d,t) reaction induced by 12 MeV deuterons (Chapter
12). However, we had to scale down the parameters to match the low energy of tritons in
this reaction. The (p,t) measurements were carried out at a significantly higher energy
and thus the parameters I have derived earlier could be tested with a better accuracy.
The experimental method

Measurements were performed using the 33 MeV proton beam from the ANU
cyclograaff facility consisting of the CNI-30 cyclotron injecting 26 MeV H- beam into the
EN electrostatic tandem accelerator. Experimental arrangement was the same as
described for (d,t) measurements (see Chapter 12) but the electronic system was
modified to allow for simultaneous measurements of both the (p,d) and (p,t) reactions
(see Figure 13.1).
Figure 13.1. A diagram of electronic system used in measurements of the (p,d) and (p,t) reactions on
Se isotopes. P Amp – Pre amplifier; Amp – Amplifier; T. SCA – Timing Single Channel Analyser; F.
Coinc – Fast Coincidence unit; LGS – Linear Gate and Stretcher; LS – Logic Shaper and Delay; P. ID
– Particle Identifier; S. Amp. – Summing Amplifier; D. Amp – Delay Amplifier; B. Amp – Biased
Amplifier; Mixer – Mixer and Routing unit; ADC – Analogue to Digital Converter; IBM 1800 – IBM 1800
computer.
Typical mass spectrum for the (d,t) reaction is shown in Figure 13.2. The deuteron
spectra resulting from the (p,d) reaction were recorded simultaneously with the triton
spectra from the (p,t) rection by setting windows on the deuteron and triton peaks in
the mass spectrum. Coincidences were required between each window and the
mixed total energy spectra. The setting of the mass windows, the establishment of
the appropriate coincidence conditions and the subsequent routing of the deuteron
and triton energy spectra for the ΔE-E telescopes into different computer areas were
all achieved by digital means. This was done by the data acquisition program
Routed Window in the IBM 1800 computer.
78
Figure 13.2. A typical mass spectrum for the interaction of protons with Se leading to (p,p’), (p,d),
and (p,t) reactions induced by 33 MeV protons.
The data were stored in the data acquisition area of the IBM 1800 computer and
dumped into the computer buffer after each run. When the buffer was full or the
measurement was finished the spectra were transferred to a demountable disc pack
for storage.
Each surface-barrier detector telescope was made of a 500 μm ΔE counter and a 2000
μm E counter. The energy resolution obtained varied from 60 to 85 keV.
Typical particle spectra for both isotopes are shown in Figures 13.3 and 13.4. Levels
were observed for up to around 4.5 MeV excitation energy. The energy calibrations
were obtained using several well-known low-lying states in 74Se and 76Se. The quoted
excitation energies are accurate to around 10 keV for strongly excited states and up to 25
keV for some weakly excited states at high excitation. We have observed 28 states in
74
Se and 25 in 76Se.
Figure 13.3. An example of a spectrum of tritons from the 76Se(p,t)74Se reaction.
Figure 13.3. An example of a spectrum of tritons from the 78Se(p,t)76Se reaction. These two spectra
show that the (p,t) reaction is dominated by just a few transitions, mainly to the ground state and to
the first excited state.
The targets were 400 μg/cm2 thick and were made by vacuum evaporation from
enriched material onto 30 μg/cm2 carbon backings. Because Se sublimes rapidly under
beam bombardment, a second thin carbon layer was evaporated onto the Se. These
sandwich targets were able to withstand beams of up to 100 nA, being the highest
currents used during the measurements. The target stability was constantly
monitored by a Si(Li) detector at 90°.
Angular distributions for most states were measured from 15° to 90°. The solid angle-
target thickness product, which was required to convert the relative cross sections
into absolute values, was determined from the Rutherford scattering at proton energy
of 4.0 MeV. The errors in the absolute cross-sections are estimated to be typically
around 15 %.
Even though we were able to assign excitation energies to a total of 53 states in 74Se
and 76Se we could measure angular distributions for only 38 of them. These
distributions are presented in Figures 13.4 and 13.5. For the remaining states, triton
peaks could be observed only at certain angles and thus angular distributions could
not be measured.
Theoretical analysis was carried out using direct reactions distorted waves Born
approximation theory and the computer code (Kunz 1966), which I modified and
adapted to run on the ANU UNIVAC 1108 computer. For protons, we have used
parameter sets of Becchetti and Greenlees (1969). Their potentials contained both
volume and surface absorptions. For tritons we have tested parameters from four
sources: Baer et al. (1973); Becchetti and Greenlees (1970); Flynn et al. 1969; and
Nurzynski (1975). We have found that my sets of parameters produced, in general, the
best fits to the experimental angular distributions. Results of the distorted wave analysis
are presented in Figures 13.4 and 13.5.
Figure 13.4. Measured angular distributions of the differential cross sections for the reaction
76
Se(p,t)74Se induced by 33 MeV protons are compared with the theoretical calculations.
Figure 13.5. Measured angular distributions of the differential cross sections for the reaction
78
Se(p,t)76Se induced by 33 MeV protons are compared with the theoretical calculations.
Selection rules for the (p,t) reactions
Let us consider a reaction A(a,b)B involving a transfer of two nucleons. Each
transferred nucleon is described by a set of three quantum numbers, ji , li , and si
(the total angular momentum, orbital angular momentum, and spin), where i = 1 or 2.
The total angular momentum for the transferred pair can be written as:
J = j1 + j2 = L + S
where
L = l1 + l 2
S = s1 + s2
For a pair of nucleons, S = 0 or 1. In particular, for two neutrons, S = 0.
The conservation of the angular momentum gives the following relations between the
angular momentum J A of the target nucleus and J B of the residual nucleus:
JA = JA + L + S = JA + J
which means that
J A − JB ≤ J ≤ J A + JB
For even-even target nuclei JA = 0 and therefore

JB = J
or
J B = L ± 1 for the S = 1 pair of nucleons
JB = L for the S = 0 pair of nucleons
The conservation of parity is fulfilled if
Δπ = (−) L
Table 13.1
Excitation energies and spin-parity assignments to states in 74Se based on the study of the
76
Se(p,t)74Se at 33 MeV proton energy
Ex Jπ Ex Jπ Ex Jπ Ex Jπ Ex Jπ
0.000 0+ 2.101 2.856 (3-) 3.601 (2+) 4.330
0.635 2+ 2.146 2.917 3.719 4.574
+ - -
0.854 0 2.350 3 3.114 3.769 (5 ) 4.628
+ + -
1.269 2 2.482 (2 ) 3.262 3.866 (5 ) 4.782
1.363 4+ 2.569 3.379 (2+) 4.002 (2+)
1.839 2.723 0+ 3.536 (5-) 4.109 (2+)
Ex – Excitation energy (in MeV) based on our measurements.
Jπ – Spin-parity assignments based on our measurements. In cases of insufficient
data at all angles, only excitations energies have been assigned.
Table 13.2
Excitation energies and spin-parity assignments to states in 76Se based on the study of the
78
Se(p,t)76Se at 33 MeV proton energy
Ex Jπ Ex Jπ Ex Jπ Ex Jπ Ex Jπ
0.000 0+ 2.166 0+ 2.670 3.106 3.693 (3-)
0.559 2+ 2.347 2.820 2+ 3.232 (4+) 3.843
1.122 0+ 2.429 3- 2.922 3.306 3.980 (3-)
1.216 2+ 2.511 (2+) 3.017 (2+) 3.458 (4+) 4.181
+ -
2.033 (4 ) 2.614 (3 ) 3.106 3.591 4.425 (4+)
By fitting the angular distributions for the (p,t) reactions one can find the L value for
the transferred pair of neutrons. If an even-even target nucleus is used (as in our
measurements) one can then assign the Jπ values (the spin and parity) to states in
the residual nucleus. Thus for L = 0, 1, 2, 3, 4, 5, etc Jπ = 0+, 1-, 2+, 3-, 4+, 5- etc.,
respectively.
Table 13.1 and 13.2 summarises our assignments of excitation energies (for the
study of particle spectra) and Jπ values (from fitting the angular distributions) to
states in 74Se and 76Se.
The absolute values of the differential cross sections
In general, theoretical calculations reproduce the measured shapes of angular
distributions sufficiently well. However, as we have anticipated, we have experienced
a problem with reproducing the absolute values of the differential cross sections.
The relationship between the experimental and theoretical differential cross sections
is as follows:
3
⎛1 ⎞2
D ⎜ πΔ2 ⎟ εS 2
2
⎛ dσ ⎞ ⎛ dσ ⎞
0
= ⎝2 ⎠
⎜ ⎟ ⎜ ⎟
⎝ dΩ ⎠exp 2L + 1 ⎝ dΩ ⎠th
⎛ dσ ⎞
where ⎜ ⎟ is the experimental differential cross section, D0 is the zero-range
2
⎝ dΩ ⎠exp
coefficient (see the Appendix E), Δ = 1.7 fm is the root mean square radius of the
triton, ε the enhancement factor, S is the theoretical spectroscopic factor, and
⎛ dσ ⎞
⎜ ⎟ is the theoretical cross section as calculated by the DWBA code.
⎝ dΩ ⎠th
We expected a problem with reproducing the absolute values of differential cross
sections because the (p,t) reactions involve a number of unknown quantities. For a
start, to calculate the D02 coefficient one has to know not only the wave function for
tritons but also the interaction potential between proton and the pair of neutrons.
To compare the absolute values of the experimental and theoretical differential cross
sections, one has to calculate spectroscopic factors for the states involved in the
transition of two neutrons. Each of the two neutrons can be picked up from different
orbits in the ground state of the target nuclei. Thus, a series of spectroscopic factors
would have to be calculated using various combinations of configurations for the two
neutrons. Such calculations would be complex but they were even impossible
because the wave functions for the ground states of the target nuclei were unknown.
In our calculations we used information, which was available to us at the time of our
study.
For D02 we used the value of 23.5 MeV2·fm3 determined by Broglia (1972). To
calculate the spectroscopic factors S, we have assumed that the two neutrons are
picked up from the same orbit, i.e. that the transitions occur only between neutron
configurations j n → j n − 2 (n even). Under this assumption, the spectroscopic factors
are given by the following formulae (Bassani, Hinz and Kavaloski 1964):
n 2 j +3−n
S= for L = 0
2 2 j +1
n(n − 2)(2 L + 1)
S= for L = 2, 4, 6, …
(2 j − 1)(2 j + 1)
Using these spectroscopic factors we have tested theoretical predictions of the
absolute values of the differential cross sections. The degree of the reliability of the
theoretical predictions of the absolute values of the differential cross section is given
by the enhancement factor ε, which should be equal to one if the theory gives correct
predictions. Table 13-3 lists examples of the enhancement factors calculated for
states corresponding to unambiguous spin-parity assignments.
Table 13.3
Enhancement coefficients
Nucleus States
01+ a) 21+ b) 0+2 b) 2+2 b) 41+ 03+ b)
74
Se 1.47 4.13 1.25 0.86 0.34 b) 1.12
76 c
Se 1.89 4.47 1.39 2.00 ) 1.03
a
) Assumed transfer: (2 p3 / 2 ) → ( 2 p3 / 2 ) ; ) Assumed transfer: (1 f 5 / 2 ) → (1 f 5 / 2 )
4 2 b 6 4
c
) Assumed transfer: (1g 9 / 2 ) → (1g 9 / 2 )
4 2
It can be seen that in general ε ≠ 1. Furthermore, the values of enhancement factors

depend on the excited state.
The only quantity in the theoretical cross section that depends on the excited state is
the spectroscopic factor. Consequently, the different values of ε indicate that the
assumed transfer configurations must be incorrect.
We have tried various other options for transfer configurations and found that ε
depends strongly on the assumed configurations. For instance, for the transfer to the
ground state ( 01+ ) in 74Se, assuming (2 p1 / 2 ) 2 → (2 p1 / 2 )0 transfer results in ε = 3.65, as
compared with ε = 1.47 for the (2 p3 / 2 ) 4 → (2 p3 / 2 ) 2 transfer. If assumed transfer is
(1 f 5 / 2 )6 → (1 f 5 / 2 ) 4 then ε = 23.56.
Clearly, the absolute values of the calculated differential cross sections depend
strongly on the assumed configurations involved in the transfer of the two neutrons
and better description of the ground state wave functions is needed to yield more
reliable values for the theoretically calculated spectroscopic factors.
Summary
The 76,78Se(p,t)74,76Se reactions were studied at the proton energy of 33 MeV using
particle telescopes to detect the outgoing tritons. Many new states were observed
and their excitation energies were calculated. Many new Jπ assignments were made
for states in the residual nuclei. Angular distributions were measured for most of the
observed states and were compared with distorted wave calculations. Spin-parity
assignments were made from a comparison of the data with both the distorted wave
calculations and the empirical shapes of angular distributions for well-known states.
Enhancement coefficients were calculated for several well-known states using various
options for configurations in the ground state wave functions of the target nuclei. The
calculated absolute values of the differential cross sections were found to depend strongly
on the assumed configurations indicating a need for a better description of the relevant wave
functions.
References
Bassani, G., Hinz, N. M, and Kavaloski,C. D. 1964, Phys. Rev. 136B:1006.
Baer, H. W., Kraushaar, J. J., Moss, C. E., King, N. S. P., Green, R. E. L., Kunz, P. D.
and Rost, E. 1973, Ann. of Phys. 76:437
Becchetti, F. D. and Greenlees, G. W. 1969, Phys. Rev. 182:1190
Becchetti, F. D. and Greenlees, G. W. 1970, Proc. Third Int. Symp. Polarization
Phenomena in Nuclear Reactions, Madison, ed. H. H. Barschall and W.
Haeberli (Univ. of Wisconsin Press) p. 682
Flynn, E. R., Armstrong, D. D., Beery, J. B. and Blair, A. G. 1969, Phys. Rev. 182:1113
Kunnz, P. D. 1966, University of Colorado Report COO-536-606.
Nurzynski, J. 1975, Nucl. Phys. A246:333
14
Single-neutron Transfer Reactions on 76, 78, 80, 82Se Isotopes Induced
by 33 MeV Protons
Key features:
1. The Se isotopes are in an interesting region of the periodic table where orbits outside
the closed N = 28 shell are almost full. Model interpretation of these nuclei is difficult
and there was a need to support theoretical work by experimental investigation.
75,77,79,81
2. We have observed a total of 120 states in Se isotopes and we have assigned
excitation energies to these states.
3. We have measured a total of 88 angular distributions for the single neutron pickup
reactions 76, 78, 80, 82Se(p,d)75,77,79,81Se induced by 33 MeV protons.
4. We have analysed the distributions using the distorted wave theory. We have
determined orbital angular momenta for the relevant states in residual nuclei and
extracted corresponding spectroscopic factors.
5. Using the experimental spectroscopic factors we have calculated the occupation
numbers for orbits in the f-p shell and compared them with the theoretical predictions
using the pairing theory. We have found a good agreement between experimental
and theoretical values.
6. In addition to the expected ln = 1,3, and 4 transitions we have also observed a
number of anomalous ln = 2 transitions. We have carried out calculations using the
Corriolis coupling model. We have found that in general, the model can account for
the presence of the positive parity states belonging to anomalous transitions.
Abstract: Single-neutron transfer reactions on even-even Se nuclei have been studied

using 33 MeV incident protons from the ANU cyclograaff facility. A total of 120 levels have
been observed below the 4 MeV excitation energy in the 75,7 7 , 79,81Se isotopes. Angular
distributions for 88 states were extracted and analysed with the distorted wave theory.
Coriolis coupling calculations have been carried out for low-level spin states in all four
isotopes.
Introduction
The Se isotopes presented an interesting and challenging case for research. The
76,78,80,82
Se nuclei contain a large number of neutrons outside the closed N = 28 shell,
ranging from 14 for 76Se to 20 for 82Se. Without the residual interaction, orbits 2p3/2,
1f5/2, and 2p1/2 would be fully occupied and orbit 1g9/2 would be filling in. With the
residual interactions, orbits 2p3/2 and 1f5/2 are expected to be about 90% full for 76Se,
orbit 2p1/2 about 85% full, and orbit 40% full. Thus, orbits 2p3/2, 1f5/2, and 2p1/2 would
contain, on average less neutrons than expected in the absence of the residual
interaction, but orbit 1g9/2 would contain more.13 With the increasing mass number,
the distribution of neutrons between various orbits would come progressively closer
to the distribution corresponding to the independent particle shell model distribution
13
Without the residual interaction, orbits 2p3/2, 1f5/2, and 2p1/2 would be 100% full and orbit 1g9/2 only
20% full for 76Se.
until for 82Se, orbits 2p3/2, 1f5/2, and 2p1/2 would be nearly 100% full and orbit 1g9/2
nearly 80% full.
The large number of neutrons outside the closed shell N = 28 makes the model
description of Se isotopes difficult. Theoretical descriptions need to be tested by
experimental results and at the time of our study little experimental information was
available.
The only spectroscopic information available from single neutron transfer reactions
to the odd Se nuclei was based on the 76,78,80Se(d,p)77,79,81Se (Lin 1965) and
76
Se(d,t)75Se (Sanderson 1973) reactions. Lin also identified some levels in 77,79,81Se
using the (d,t) reaction but did not measure the corresponding angular distributions.
Single neutron stripping reactions populate vacant neutron configurations.
Consequently, the (d,p) reactions are not expected to have significantly strong
transitions to 2p3/2, 1f5/2, and 2p1/2, orbits, particularly for heavier Se isotopes. One
can, however, expect transitions to the 1g9/2 orbit (l = 4) or to orbits outside the
closed N = 50 shell (l = 0 and 2 transitions). Indeed, Lin observed many l = 2
transitions but only a small number of l = 1, 3, and 4.
On the other hand, the single neutron pickup reactions are expected to have strong
transitions from the well-populated configurations in the target nuclei. Consequently,
the (p,d) reactions should show many l = 1 and 3 transitions. As the 1g9/2 orbits are
filling in with the increasing mass number, the number of l = 4 transitions is expected
to increase for heavier Se isotopes.
Nuclei in the Se mass region cannot be fully described by any known model. It was
originally thought that the level structure of the even Se nuclei should be well
explained by the vibrational model (Scharff and Weneser 1955). However Barrette et
al. (1974) have shown that with the exception of 74Se, the ratios of the electric
quadrupole transitions probabilities, B(E 2) , in the even Se isotopes do not comply
with the simple vibrational description.
Lieder and Draper (1970), McCauley and Draper (1971), Wyckhoff and Draper
(1973) and Nolte et al. (1977) investigated the even-even Ge, Se and Kr nuclei using
heavy ion reactions and found quasi-rotational bands with spins of up to 10+. The
large moment of inertia determined for these bands indicate that the ground states of
these nuclei may be deformed. Coupling a neutron to such a deformed core should
stabilize the deformation and one could therefore expect the odd nuclei in this mass
region to be also deformed.
The anharmonic vibrator model is also able to describe such quasi-rotational bands
as have been shown in this region in a study of heavy ion reactions. Holzwarth and
Lie (1972) and Lie and Holzwarth (1975) used such a model to describe 76Se and
78
Se and obtained good agreement between the calculated and experimental levels
below 2.5 MeV. They also calculated quadrupole moments and B (E 2) values which
agree reasonably well with the experimental results.
Another problem that needs to be solved and explained is the presence of the low-
lying 5/2+ and 7/2+ levels in all odd N, even Z nuclei throughout the 39 < N < 49 region.
They are unlikely to be single particle states since the 2d5/2 and the 1g7/2 orbits
should be filling in at N > 50. Similar 5/2+ and 7/2+ low-lying states are also found in
odd-Z, even-N nuclei in the 1g9/2 mass region. The presence of such states in all of
these nuclei is somewhat surprising and has led to several theoretical attempts to
explain their origins.
The first attempt to explain these states was made by Flowers (1952) who used the
seniority-coupling model. His calculations predict that the 7/2+ level is never the
ground state. This is in conflict with the experimental results, which show, for
example, that 79Se has a ground state spin of 7/2+.
Kisslinger and Sorensen (1960) coupled the 1g9/2 quasi-particle to the neighbouring
2+ one-phonon state (QPC model). However, their calculations could not explain the
ground state spins of 5/2+ and 7/2+ in 75Se and 79Se, respectively. Their calculations
were later improved by Sherwood and Goswami (1966) who included the quasihole-
phonon coupling (EQPC model) and by Goswami and Nalcioglu (1968) who included
the quadrupole-quadrupole interaction. Again these extended calculations failed to
explain fully the presence of 5/2+ and 7/2+ states, which often lie below the 9/2+ state.
Because of the inability of the spherical shell model with appropriate residual
interactions to describe adequately the odd proton nuclei in this region, Scholz and
Malik (1968) extended their successful use of Coriolis coupling model in the f7/2 shell
to some of these nuclei. The model correctly predicted spins and parities for the low-
lying states with the right energy spacing for all the nuclei they studied.
Since the situation is analogous to the low-lying positive parity states that occur in
the 1g9/2 odd neutron nuclei, Coriolis coupling model has been applied with some
success to 75Se by Sanderson (1963) and to a number of odd neutron nuclei in the
1g9/2 mass region (73 < A < 87) by Heller and Friedman (1974, 1975). These authors
obtained spins, parities and level spacings for the low-lying positive parity states that
are in good agreement with the experimental data. Their calculations predict a
prolate deformation in this mass region.
One of the aims of our study was to extend the study of the low-lying 5/2+ and 7/2+
states for Se nuclei. Available experimental evidence suggests that Se nuclei might
be deformed and that such states could be interpreted as arising from Coriolis
coupling. If such is the case then these states should have significant single particle
components and thus should be excited in the (p,d) reaction.
Experimental method
Experimental arrangement was the same as for the (p,t) reactions (see Chapter 13).
Target enrichments were 76Se 86%, 78Se 96%, 80Se 94% and 82Se 97%. Examples
of deuteron spectra are shown in Figure 14.1. Careful checks were made for each
isotope to ensure that identified peaks did not arise from Se impurities. A total of 120
levels have been identified and the corresponding excitation energies have been
assigned to all of them. The quoted energies are accurate to around 10 keV for
strongly excited states and up to 25 keV for weaker states at high excitations.
We have measured a total of 88 angular distributions, 25 for 76Se, 16 for 78Se, 18 for
80
Se, and 29 for 82Se. They are shown in Figures 14.2 – 14.11. They served as a
basis for assigning orbital angular momenta to the respective states in residual
nuclei and to derive spectroscopic factors.
Absolute cross sections are accurate to around 15% and were determined from the
Rutherford scattering of 4 MeV protons. As mentioned in Chapter 13, (p,d) and (p,t)
reactions were measured simultaneously. As the cross sections for the (p,d) reaction
were in general higher than for the two neutron (p,t) transfer, good statistical
accuracy for the (p,t) reaction resulted in excellent accuracy for the (p,d) data.
76,78,80,82
Figure 14.1. Examples of deuteron spectra for the reactions Se(p,d)75,77,79,81Se induced by 33
MeV protons.
The distorted wave analysis
The distorted wave calculations were carried out using the code DWUCK (Kunz 1966)
which included standard finite-range and non-locality corrections. The proton optical-
model parameters were derived from the sets given by Becchetti and Greenlees
(1969). While many sets of deuteron parameters were tried, those that gave the best
fits to the data were obtained from the elastic scattering on Zn at 25.9 MeV (Perey
and Perey 1966). The transferred neutron was assumed to be bound in a Woods-
Saxon well and the customary separation energy prescription was used. Calculated
angular distributions are compared with the experimental results in Figures 14.2 –
14.7.
Figure 14.2. Deuteron angular distributions for the reaction 76Se(p,d)75Se induced by 33 MeV protons
compared with the distorted wave calculations for ln = 1 and ln = 1 + 3 (the left-hand side of the figure)
and for ln = 3 (the right-hand side).
76 75
Figure 14.3. Deuteron angular distributions for the reaction Se(p,d) Se induced by 33 MeV protons
compared with the distorted wave calculations for the even ln values.
compared with the distorted wave calculations for ln = 1 (the left-hand side of the figure) and for ln = 2,
3, and 4 (the right-hand side).
compared with the distorted wave calculations for ln = 1 and 2 (the left-hand side of the figure) and for
ln = 3 and 4 (the right-hand side).
compared with the distorted wave calculations for ln = 1 and ln = 1 + 3 (the left-hand side of the figure)
and for the even ln values (the right-hand side).
compared with the distorted wave calculations for ln = 3, 4, 2+3, and 3+4.
In general the main transfer peak in the ln = 2, 3, 4 angular distributions is well
reproduced by the calculations while for ln = 1 the calculations often tend to over-
emphasize the depth of the first minimum in the data. With the exception of 81Se, the
calculations had difficulty in reproducing the data for all values of ln at larger angles.
Table 14.1
Spectroscopic information for Se from the reaction 76Se(p,d)75Se induced by 33 MeV protons
75
Ex ln S Ex ln S Ex ln S Ex ln S
0 2 0.12 776 3 0.53 1369 (4) 0.15 1913 (1) 0.15
132 4 3.97 859 1 0.15 1403 (3) 0.22 2037 (1+3) 0.04,0.07
287 1 2.34 895 1 0.15 1484 3 0.32 2288 (1+3) 0.04,0.06
427 3 2.15 962 1 0.58 1580 (2) 0.03 2573 1 0.09
583 1 0.13 1050 3 0.64 1673 (1) 0.10

629 2 0.15 1182 3 0.26 1768 (3) 0.25
667 3 0.88 1246 1 0.04 1810 1 0.23
Ex – Excitation energy in keV; ln – Orbital angular momentum; S – Spectroscopic factor
Table 14.2
Spectroscopic information for 77Se from the reaction 78Se(p,d)76Se induced by 33 MeV protons
0 1 0.61 436 3 0.64 1012 1 0.26 1470 1 0.48
178 4 4.39 522 1 1.7 1183 (3) 0.43 1522 (3) 0.81
250 (3) 4.12 680 2 0.17 1238 3 0.93 1717 1 0.45
302 (2) 0.16 820 1 0.45 1366 (1) 0.17 2209 1 0.19
Table 14.3
Spectroscopic information for Se from the reaction 80Se(p,d)79Se induced by 33 MeV protons
79
0 (4) 0.36 720 2 0.45 1385 (3) 0.72 2092 3 2.02
97 (1) 3.82 979 1 5.17 1674 (2) 0.09 2212 (2) 0.36
137 4 9.26 1096 4 1.98 1817 (3) 1.93
365 3 7.19 1258 2 0.14 1964 1 0.72
623 2 0.325 1328 (3) 1.17 2037 1 0.76
Table 14.4
81 82 81
Spectroscopic information for Se from the reaction Se(p,d) Se induced by 33 MeV protons
0 1 1.32 1056 2 0.21 2056 1 0.16 2656 3 0.46
100 (4) 0.19 1109 2 0.05 2150 1 0.74 2763 (2+3) 0.03,0.07
294 4 5.61 1310 2 0.05 2199 2 0.05
470 (1) 0.60 1417 1 2.25 2282 1 0.09 2893 (3+4) 0.23, 0.14
491 (3) 1.74 1628 1 0.05 2325 (2+4) 0.07,0.12 2985 (3+4) 0.21,0.16
624 3 2.37 1753 2 0.16 2531 4 0.35 3087 (2) 0.16
782 4 0.14 1812 4 1.07 2603 1 0.30 3150 (1+3) 0.05,0.07
Table 14.5
Occupation numbers and centre-of-gravity energies for the ground states in Se isotopes
Isotope lj n jl n′jl V j2 Ej
Exp. Theory Exp. Theory

76
Se 2p1/2+2p3/2 6 4.2 0.70 0.75 0.70 0.96
1f5/2 6 5.4 0.90 0.91 0.76 1.01
1g9/2 2 4.2 0.42 0.41 0.18 0.03
78
Se 2p1/2+2p3/2 6 4.3 0.72 0.80 0.85 1.20
1f5/2 6 6.0 1.00 0.94 0.61 1.35
1g9/2 4 4.4 0.44 0.55 0.17 0.01
80
Se 2p1/2+2p3/2 6 5.8 0.96 0.87 0.76 1.48
1f5/2 6 6.0 1.00 0.96 1.00 1.73
1g9/2 6 5.3 0.53 0.70 0.29 0.11
82
Se 2p1/2+2p3/2 6 5.8 0.97 0.92 1.28 1.70
1f5/2 6 5.8 0.96 0.99 1.24 1.99
1g9/2 8 7.6 0.76 0.76 0.76 0.26
j – single particle configuration.
n jl – The experimentally determined average number of neutrons in the configuration jl.
n′jl – The number of neutrons expected in the absence of the residual interaction.
Exp. – experimental values as derived from the (p,d) reactions of Se isotopes induced by 33 MeV
protons.
Theory – theoretical values calculated using the pairing theory of Kisslinger and Sorensen (1960,
1963) and my computer code.
Spectroscopic factors were extracted from the data by comparing the theoretical
calculations with the experimental distributions. A summary of spectroscopic
information derived in this study (excitation energies, orbital angular momenta, and
spectroscopic factors) is given in Tables 14.1-14.4.
As expected, we have observed many ln = 1 and 3 transitions as well as an
increasing number of ln = 4 transitions with the increasing atomic number A of Se
isotopes. All transitions belong to the pickup of neutrons from the 2p1/2, 2p3/2, 1f5/2,
and 1g9/2 orbits.
However, we have also observed a number of anomalous ln = 2 transitions. The
states belonging to these anomalous transitions will be discussed later.
Using the experimentally determined spectroscopic factors we have calculated
neutron occupation numbers for configurations in the f-p shell and compared them
with the theoretical calculations using the pairing theory and the computer code I
have written. (The procedure is described in Chapter 12.) As the spins of states in
the relevant residual nuclei are generally unknown, we had to consider an average
occupation numbers for configurations 2p1/2 and 2p3/2.
V12/ 2 + 2V32/ 2
(V22p )(avth ) =
3
∑ S (li n = 1)
(V )2 (exp)
2 p av = i
6
where (V22p )(avth ) and (V22p )(exp)
av are the theoretical and experimental average occupation
numbers, respectively, for the configurations 2p1/2 and 2p3/2, Si (ln = 1) is the
spectroscopic factor for ln = 1 transition to state i, and the sum is over all ln = 1 states
in a given Se isotope. Results of our calculations are presented in Table 14.5.
To try to account for ln = 2 transitions we have carried out the Coriolis calculations.
Coriolis coupling model was used by Sanderson and Summers-Gill (1976) to
reproduce energy levels for anomalous low-energy positive parity states as well as
various E2 and M1 transition rates in 75Se. In addition, Heller and Friedman (1974,
1975) have reproduced moments and transition rates for several nuclei in this mass
region using the same model. It was therefore worthwhile to extend such calculations
to other Se isotopes.
Coriolis coupling model assumes that the odd neutron moves in the deformed
potential of the rotating, axially symmetric doubly even core. The coupling of this
neutron to the deformed core then gives rise, in the case of Se, to positive parity
energy spectra associated with strong mixing between rotational bands built on
Nilsson states arising from the 1g9/2 orbital.
We have carried out our calculations using computer code of Caseten and Newton
(1968). We have used the Nilsson and pairing-model parameters derived from
experimental information by Sanderson. The deformation parameter β was varied
from -0.3 to 0.3 but no attempt was made to do an exhaustive search on other
parameters.
Table 14.6 compares the calculated level energies and (p,d) spectroscopic factors
for low-lying positive parity states with those obtained from experiment. A positive
deformation of β = 0.275 was found to give the closest agreement with the data for
all isotopes and is the same as determined Sanderson for 75Se.
In general, the model generates sets of low-lying 5/2+, 7/2+ and 9/2+ states whose
excitation energies are in close vicinity of the corresponding experimental values.
However, the calculated spectroscopic factors, while of the correct order of
magnitude, are not always close to the experimental values.
Table 14.6
Results of the Coriolis coupling calculations compared with the experimental data for 75, 77, 79, 81Se
Nucleus Jπ Eexp Eth Sexp Sth Nucleus Jπ Eexp Eth Sexp Sth
75
Se 5/2+ 0 58 0.12 0.44 77
Se 7/2+ 162 162 0.04
7/2+ 112 364 0.03 9/2+ 178 174 5.39 4.00
9/2+ 132 0 3.97 2.87 5/2+ 302 325 0.19 0.30
79 + 81 +
Se 7/2 0 0 0.14 0.05 Se 7/2 100 100 0.19 0.04
9/2+ 137 101 3.56 3.95 9/2+ 294 236 5.61 3.79
+ +
1/2 527 0.12 1/2 622 0.25
+ +
5/2 623 569 0.12 0.27 3/2 782 0.36
5/2+ 1056 1238 0.21 0.22
As discussed in the Introduction, selenium isotopes present an interesting but

difficult case for nuclear structure interpretation. They are in the mass region where a
variety of nuclear models have been tried and have failed to provide a satisfactory
description of the features observed experimentally. Nuclei in this region are soft and
consequently have a rich band structure with a co-existence of states belonging to
various shapes. Multiple band crossing and band mixing is quite common. Both
neutron and proton rotation-aligned bands are present. There is also evidence for a
shape transition at around N = 40.
Theoretical treatment of such nuclei is difficult. As a possible alternative to the
mentioned theoretical interpretation of nuclei in this region, dynamic deformation
theory has been suggested (Kumar et al. 1977, Kumar 1978). It includes shape co-
existence, shape transition and pair fluctuations. The theory has been used to
interpret the structure of the of 70,72,74Ge nuclei (Kumar 1978).

The odd neutron nuclei 75,77,79,81 Se have been studied using the (p,d) reaction at the
proton energy of 33 MeV. Angular distributions for the emitted deuterons have been
measured from 15° to around 60° using particle identification telescopes, and the
experimental results analysed using the distorted wave procedure. From the distorted
wave calculations assignments of ln = 1, 2, 3 and 4 have been made. Spectroscopic
factors were also extracted from the angular distributions. Many new states were
identified.
Using the experimentally determined spectroscopic factors, occupation numbers
have been calculated and compared with the predictions of the pairing theory.
Experimentally determined occupation numbers are in close agreement with the
calculated values.
The presence of the low-energy 5/2+ and 7/2+ states in the Se region has long
attracted theoretical attention. It has been shown here that the existence of these
states can be explained reasonably well by assuming Coriolis coupling model.
References
Barrette, J., Marrette, M., Lamoureux, G., Monara, S. and Markiza, S. 1974, Nucl. Phys.
A235:154.
Becchetti, F. D. and Greenlees, G. W. 1969, Phys. Rev. 182:1190.
Casten, R. and Newton, C. S. 1968, Niels Bohr Institute Computer Centre Report GAP
15-16.
Flowers, B. H. 1952, Proc. Roy. Soc. 215:398.
Goswami, A. and Nalcioglu, O. 1968, Phys. Lett. 30B:397.
Heller, S. L. and Friedman, J. N. 1974,Phys. Rev. C10:1509.
Heller, S. L. and Friedman, J. N. 1975,Phys. Rev. C12:1006
Holzwarth, G. and Lie, S. G. 1972, Z. Phys. 249:332.
Kisslinger, L. S. and Sorensen, R. A. 1960, Kgl. Danske Videskab. Selskab. Mat. Fys.
Medd. 32, No. 9.
Kumar, K., Remaud, B., Auger, P., Vaagen, J. S., Rester, A. C., Foucher, R.
and Hamilton, J. H. 1977, Phys. Rev. C16:1235.
Kumar, K. 1978, J. of Phys. G4:849.
Kunz, P. D. 1966, University of Colorado Report COO-536-606.
Lieder, R. and Draper, J. 1970, Phys. Rev. C2:531.
and 1975, Phys. Rev. 12C:1035.
Lin, E. K. 1965, Phys. Rev. 139:B340.
McCauley, D. G. and Draper, J. E. 1971, Phys. Rev. C4:475
Nolte, E., Kutschera, W., Shida, Y. and Moringa, H. 1977, Phys. Lett. 33B:294.
Sanderson, N. 1973, Nucl. Phys. A216:173.
Sanderson, N. E. and Summers-Gill, R. G. 1976, Nucl. Phys. A261:93
Scharff, G. and Weneser, J. 1955, Phys. Rev. 98:212.
Scholz, W. and Malik, F. B. 1968, Phys. Rev. 176:1355.
Sherwood, A. and Goswami, A. 1966, Nucl. Phys. 89:465
Wyckhoff, W. G. and Daraper, J. E. 1973, Phys. Rev. C8:796.
15
Analysis of Polarization Experiments
Key features:
1. I have written a computer program, which carries out global analysis of particle
spectra and converts them into angular distributions of differential cross sections and
analyzing powers.
2. Programming language: FORTRAN-77.
3. Input parameters have to be defined for only one particle spectrum. The program
employs nuclear kinematics to trace the positions of peaks for all remaining spectra
taken at different angles and performs Gaussian analysis or channel-by-channel
integration, depending on the selected option.
4. Computing time: On the VAX VMS 3.6 machine, the program takes less than 1.5
minutes to fit Gaussian distributions to about 200 particle groups and convert them to
differential cross sections and polarization distributions.
5. Output is in both tabulated and graphic forms. The program contains interactive
options to increase the flexibility of its use and to allow for a prompt and easy
assessment of results.
6. The program can analyse up to 20 particle groups corresponding to different
excitation energies and produce angular distributions of the differential cross sections
and analyzing powers for all of them. Up to 200 particle spectra can be analysed,
each containing up to 2048 channels, obtained using up to 8 detectors. Thus, the
program can analyse up to 32,000 spectral peaks in one session. These imposed
limitations are more than adequate to analyse any experimental results but they can
be easily increased.
Abstract: Program LORNA executes global analysis of particle spectra generated in nuclear
reactions induced by polarized projectiles. Peak positions have to be defined for only one
spectrum. The program employs nuclear kinematics to trace the positions at different
reactions angles. Spectra corresponding to different signs of the source polarization, taken at
various reaction angles by an array of up to 8 detectors and containing up to 20 particle groups
are analysed and converted to angular distributions of differential cross-sections and the
analyzing powers.
Introduction
When I visited the Max-Plank-Institut für Kerphysik to carry out research on deuteron
polarization I was surprised that there was no suitable program available for
calculating angular distributions of differential cross sections and polarizations from
particle spectra obtained in polarization experiments. It was therefore necessary for
me to write a suitable computer code to support my work.
Measurements of the angular distributions of particles from nuclear reactions are car-
ried out using an array of detectors. Results of measurements are usually contained
in a great number of particle spectra taken at various angles. Each spectrum may be
composed of a number of particle groups corresponding to various excitation energies
of residual nuclei. For reactions induced by polarized projectiles a set of particle
spectra has to be taken for each reaction angle with pairs of detectors positioned on
each side of the incident beam of polarized particles and corresponding to different
signs of beam polarization. For instance, for measurements of deuteron polarization
using cyclotrons, a set of eight spectra is required for each reaction angle to
determine all analyzing powers. Information contained in such an assembly of raw
data has to be untangled and presented in the form of angular distributions suitable
for further theoretical analysis.
The aim in writing the computer code LORNA was to combine various stages of such
data reduction calculations into one and easy to execute global analysis of particle
spectra, in which the massive amount of data recorded by particle detectors is
analysed en block and converted to final angular distributions of differential cross
sections and analyzing powers.
My aim was to minimise the number of input parameters and thus make the job
simple and easy for the user. An important part of achieving this aim was to
incorporate nuclear kinematics calculations in the program. I also wanted the user to
be able to interact with the program and to have opportunity of assessing the results
of the calculations quickly and easily. This requirement made it necessary to
incorporate an interactive component in the program and to supply suitable visual
displays. The program also had to produce results in tabulated form, which could be
readily available to further theoretical analysis.
By using the basic information supplied by the user, the program LORNA calculates
kinematics calibration coefficients for each set of spectra. The user is asked to select
one spectrum only and to indicate groups of particles for which the angular distribu-
tions should be calculated by the program. Employing nuclear reaction kinematics
calculations, the program traces automatically the indicated peaks in all spectra,
performs the desired integration, subtracts background, carries out all other neces-
sary arithmetical manipulations (error calculation, averaging, lab to centre-of-mass
(c.m.) conversion, etc.) and produces angular distributions of the differential cross-
sections and analyzing powers in a final form (both numerical and graphic).
Calculations can be carried out either for gas of solid targets. Peak intensities can
be extracted either through the channel-by-channel integration or by fitting
symmetric Gaussian distributions. Calculated differential cross-sections are
expressed in the absolute units (mb/sr) in the laboratory and centre-of-mass
systems of reference. For the elastic scattering, Rutherford scattering calculation is
included in the program to allow for the customary σ(θ)/σR(θ) presentation of the
data.
Procedure for extracting analyzing powers from the particle spectra depends on the
projectile spin S and on a particular experimental configuration employed in the
measurements. Program LORNA has been written for S = 1 particles. Provision has
been made for other spins and suitable modification of the program can be made
without changing the general calculation procedure adopted in the program. Being
written for polarized ions, the program can be also used for reactions induced by
unpolarized particles.
In order to simplify the user’s interaction with the computer, graphic display has
been incorporated in the program both in the input and output stages. However, the
program can be used without graphic display and, therefore, it is adaptable to a
wider range of computer installations.
There are no strict limitations on the number of spectra sets, which can be
analysed by the program, on the size of the spectra, and on the maximum number
of particle detectors used in measurements. The selected limits of 200, 2048 and 8,
respectively, can be changed easily by changing the appropriate parameter defi-
nitions at the beginning of the main part of the program and in some subroutines.
Programs with broad application and thus containing great number of options and
pathways are often difficult to use. My aim in writing program LORNA was to make
it as simple as possible and yet to allow for sufficient flexibility in order to make it
applicable to different experimental environments.
Program arithmetic
Differential cross sections
Expressions for the differential cross sections have the following forms:
For solid targets
JMqNτ
σ (θ ) = k s (in mb/sr)
CΩtf
where
J – Jacobian of transformation from the lab to c.m. system
M – atomic mass of the target material
q – projectile charge state
N – number of counts
τ – dead time correction factor
C – total beam charge in μC
Ω – detector solid angle in sr
t – target thickness in mg/cm2
f – observed fraction of the total charge distribution of the projectile
k s = 2.66018 × 10 −7
For gas targets
JTqNτ sin θ L
σ (θ ) = k g
CGnpf
where
Τ – gas temperature in 0K
θL – laboratory scattering angle
n – gas atomicity14
p – gas pressure in Torr
G – gas factor15 in cm
k g = 1.65986 × 10 −6
14
The number of atoms in a molecule.
15
G00 in Silverstein (1959).
The analyzing powers
Polarization of an assembly of S =1 particles can be described using either Cartesian
or spherical spin moments, which in turn are defined using basic angular momentum
operators. In the reference frame where the z-axis is aligned with the quantization
r
axis s (see Figure 15.1) deuteron polarization is described using two non-zero
quantities pz and pzz. If N+, N0, and N- represent the fractional populations along the
symmetry axis, for the deuteron spin projections up, zero, or down, respectively, then
the asymmetry in the fractional populations relative to the field direction (i.e. along
the symmetry axis) is given by
N+ − N−
pz=
N+ + N− + N0
This quantity is known as the vector polarization.
Asymmetry in the plane perpendicular to the field direction is given by
N+ + N− − 2 N0
pzz =
N+ + N− + N0
which is known as tensor polarization. The quantities pz and pzz are the expectation
values of the spin operators Sz and Szz given by the following matrices
⎡1 0 0 ⎤ ⎡1 0 0⎤
S x = ⎢0 0 0 ⎥ S zz = ⎢⎢0 − 2 0⎥⎥
⎢ ⎥
⎢⎣0 0 − 1⎥⎦ ⎢⎣0 0 1⎥⎦
(see the Appendix F).

The relevant spherical spin operators are defined as
3 1
τ 10 = S x and τ 20 = S zz
2 2
If we write their expectation values as t10 , for the vector polarization, and t20,for the
tensor polarization, still in the reference frame with the z-axis along the quantization
axis, then we have the following relations between the Cartesian and spherical
definitions of polarizations:
3 1
t10 = pz and t20 = pzz
2 2
According to the Madison Convention (Barshal and Haeberlie 1971), measured

polarization is defined in the right-handed
r
frame of references with the z-axis in rthe
direction of the incident momentum k in and the y-axis in the direction of k in × k out ,
r
where k out is the momentum for the outgoing particles. The rotation from the system
defined by the quantization axis to the new coordinate system complicates the
description of the polarization and introduces expressions that depend on β and ϕ
defined in Figure 15.1.
r
Figure 15.1. The relation between the quantization axis s and the right-handed system of coordinates
(xyz) as defined by the Madison Convention.
Using the spherical components of polarization, the differential cross section for
reactions induced by polarized the S = 1 particles is then given by
σ (θ ,ϕ ) = σ 0 (θ )[1 + A + B + C + D ]
with
A = 2 (sin β )(cos ϕ )t10iT11 (θ )
1
B = (3 cos 2 β − 1)t20T20 (θ )
2
3
C= (sin 2β )(sin ϕ )t 20T21 (θ )
2
3
D=− (sin 2 β )(cos 2ϕ )t20T22 (θ )
2
Where iT11 (θ ) , T20 (θ ) , T21 (θ ) , T22 (θ ) are the parameters describing the polarization of
the outgoing particles. They are also known as the analyzing powers. The quantities
t10 and t20 describe the beam polarization (in the symmetry axis system of reference)
produced by the polarized ion source.
To measure the analyzing powers, the required angles β and ϕ are selected using a
spin rotation device such as a Wien filter. The analyzing powers iT11 (θ ) , T20 (θ ) ,
T21 (θ ) , T22 (θ ) can be measured by using a pair of symmetrically positioned detectors
on the left and the right hand sides of the beam, corresponding to angles ϕ and ϕ+π
and by changing the beam polarization in the source between positive and negative.
Four counting rates are recorded for each reaction angle θ: N L+ , N L− , N R+ , N R− . Using
these counting rates one can calculate the ratios:
N L+ − N L− N R+ − N R−
L= and R =
N L+ + N L− N R+ + N R−
These terms can be expressed as:
L=
1
+ 2 (sin β )(cos ϕ )t10iT11 (θ ) + (3 cos 2 β − 1)t 20T20 (θ )
2
3 3
+ (sin 2ε )(sin ϕ )t20T21 (θ ) − (sin 2 β )(cos 2ϕ )t20T22 (θ )
2 2
R=
1
− 2 (sin β )(cos ϕ )t10iT11 (θ ) + (3 cos 2 β − 1)t20T20 (θ )
2
3 3
− (sin 2β )(sin β )t20T21 (θ ) − (sin 2 β )(cos 2ϕ )t20T22 (θ )
2 2
By selecting suitable values for β and ϕ, these left-right measurements can be used
to determine all four analyzing powers iT11 (θ ) , T20 (θ ) , T21 (θ ) , and T22 (θ ) :
1 L−R
iT11 (θ ) = for ( β ,ϕ ) = (900 ,00 )
t10 2 2
1 L+R
T20 (θ ) = for β = 00
t20 2
2 L−R
T21 (θ ) = for ( β ,ϕ ) = (450 ,900 )
t 20 3 2
2 L+R 1
T22 (θ ) = + T20 (θ ) for ( β ,ϕ ) = (900 ,900 )
t20 3 2 6
Measurements with purely vector-polarized beam have to be carried out to
determine iT11 (θ ) . To determine T22 (θ ) one must first determine T20 (θ ) .
The errors associated with the inaccuracies in β and ϕ are given by:
1
ΔiT11 (θ ) = [(Δβ ) 2 + (Δϕ ) 2 ]iT11 (θ )
2
3 3
ΔT20 (θ ) = (Δβ ) 2 T20 (θ ) + (cos 2ϕ )(Δβ ) 2 T22 (θ )
2 2
⎡ 2⎤
ΔT21 (θ ) = ⎢2(Δβ ) + (Δϕ ) ⎥T21 (θ ) + (Δϕ + ΔβΔϕ )iT11 (θ )
2 1 2
⎣ 2 ⎦ 3
⎡1 2⎤
ΔT22 = ⎢ (Δβ ) + 2(Δϕ ) ⎥T22 (θ ) − (Δϕ )2 T20 (θ )
2 3
⎣2 ⎦ 2
It can be seen that measurements of iT11 (θ ) , T20 (θ ) , and T22 (θ ) are insensitive to
small deviations in the angles β and ϕ. The errors depend quadratically on Δβ and
Δϕ and consequently deviations of up to about 40 result in errors less than 1% for
these analyzing powers. This is important, because precise determination of β and ϕ
is difficult. Only ΔT21 (θ ) depends linearly on Δϕ and to determine this component
with an accuracy of 1% the accuracy of Δϕ ≤ 20 is required.
For experiments carried out with cyclotrons, the angle β is fixed by the cyclotron’s
magnetic field, which means that the angle β is determined by the orientation of the
plane containing particle detectors (see below). For the fixed vertical quantization
axis it is convenient to express the differential cross sections in terms of the
Cartesian spin moments:
⎡ 3 1 1 ⎤
σ (θ ,ϕ ) = σ 0 (θ ) ⎢1 + (cosϕ ) pz Ay (θ ) + (sin 2 ϕ ) pzz Axx (θ ) + (cos 2 ϕ ) pzz Ayy (θ )⎥
⎣ 2 2 2 ⎦
where pz and pzz are the source polarizations, as defined earlier; Ay, Axx, and Ayy are
the polarizations (i.e. the analyzing powers) of outgoing particles. The relations
between spherical and Cartesian spin moments is as follows:
2
Ay (θ ) = iT11 (θ )
3
1
Axx (θ ) = 3T22 (θ ) − T20 (θ )
2
1
Ayy (θ ) = − 3T22 (θ ) − T20 (θ )
2
To determine all three analyzing powers, Ay (θ ) , Axx (θ ) , and Ayy (θ ) , one has to carry
out measurements not only in the horizontal plane but also in the vertical plane. A
convenient way to do it is to have a rotating target chamber and to record eight
counting rates for each reaction angle θ by using pairs of detectors in the horizontal
and vertical planes.
First the target chamber is in the horizontal position and measurements are carried
out as before using symmetrically positioned detectors on the left and right hand side
of the beam. The left detector corresponds to ϕ = 00 and the right to ϕ = π. Four sets
of counting rates are collected: N L+ , N L− , N R+ , and N R− .
Having done that, the target chamber is rotated to a vertical position without
changing the reaction angle θ. The detectors are now in the up and down positions
and they correspond to ϕ = ±π/2. Four more counting rates are recorded,
NU+ , NU− , N D+ , and N D− , for the same reaction angle θ. Using this set of eight counting
rates, one can determine the analyzing powers Ay (θ ) , Axx (θ ) , and Ayy (θ ) .
We now have the following relations:

N L+ − N L− 3 1
L= + −
= + p z Ay (θ ) + p zz Ayy (θ )
NL + NL 2 2
N R+ − N R− 3 1
R= + −
= − pz Ay (θ ) + pzz Ayy (θ )
NR + NR 2 2
NU+ − NU− 1
U= = p zz Axx (θ )
NU+ + NU− 2
N D+ − N D− 1
D= = pzz Axx (θ )
N D+ + N D− 2
These relations give:
1
Ay = ( L − R) for ( β ,ϕ ) = (900 ,00 )
3 pz
1
Axx = (U + D ) for ( β ,ϕ ) = (900 ,900 )
pzz
1
Ayy = ( L + R ) for ( β ,ϕ ) = (900 ,00 )
pzz
The errors associated with the vertical position of the reaction chamber are:
Ay (Δϕ )
1
ΔAy =
2
2
ΔAxx = (Axx − Ayy )(Δϕ )
2
ΔAyy = (Ayy − Axx )(Δϕ )

2
Polarimeters
Measurements of the analyzing powers must be accompanied by the measurements
of the beam polarization to determine the quantities t10 and t20 or pz and pzz. This is
done by using a reaction for which analyzing powers are known. The device
measuring beam polarization is known as a polarimeter. The choice of a polarimeter
is dictated by the type and energy of the incident particles and by the so-called figure
of merit, which is the product of the differential cross section and polarization. This
product should preferably be large for the energy region of interest. A well-known
and widely used tensor polarimeter for the medium-energy deuterons is based on
r 4
the He( d ,p) He reaction with the detection of protons at θ = 00. Examples
3
r r
polarimeters for the vector analyzing powers are 4He( d ,d )4He and 12C( d ,d )12C
scattering.
Program LORNA incorporates the polarimeter spectra in the calculations of the
analyzing powers.
Nuclear reactions kinematics

The program uses the non-relativistic nuclear kinematics formulae (see for instance
Marion 1960) to trace the positions of peaks in particle spectra.
The parameters defining nuclear reactions kinematics are shown in Figure 15.2.
Figure 15.2. A diagram defining the quantities used in the nuclear reactions kinematics calculations.
Angles φ and ζ are in the laboratory system while angles θ and ξ in the centre-of-mass system.
The energy of the light product is given by:
[
E3 = ET A + D + 2( AC )
1/ 2
] [
cosθ = ET B cos φ ± (D B − sin 2 φ ) ]
1/ 2 2
where
ET = E1 + Q = E3 + E4
Q = (M 1 + M 2 − M 3 − M 4 )c 2
M 1M 4 E1
A=
(M 1 + M 2 )(M 3 + M 4 ) ET
M 1M 3 E1
B=
(M 1 + M 2 )(M 3 + M 4 ) ET
M 2M 3 ⎛ MQ ⎞
C= ⎜⎜1 + 1 ⎟⎟
(M 1 + M 2 )(M 3 + M 4 ) ⎝ M 2 ET ⎠
M 2M 4 ⎛ MQ ⎞
D= ⎜⎜1 + 1 ⎟⎟
(M 1 + M 2 )(M 3 + M 4 ) ⎝ M 2 ET ⎠
The plus sign in the E3 equation is used if B<D. If B>D then the maximum emission
angle of the light product is given by
φmax = sin −1 ( D / B)1 / 2
The intensity ratio for light products is given by:
σ (θ ) I (θ ) sin φ dφ sin 2 φ ( AC )1 / 2 ( D / B − sin 2 φ )1 / 2

= = = cos(θ − φ ) =
σ (φ ) I (φ ) sin θ dθ sin 2 θ E3 / ET
Input and output parameters
Every effort has been made to make the use of the program simple and easy. Many
default parameters are supplied to assist the user who, however, can overrule them.
The program contains an interactive part to allow for a grater flexibility of the use of
the program and for an easy assessment of results. The program has an option for
either channel-by-channel integration or Gaussian analysis. In both cases, the
necessary parameters have to be defined only for one particle spectrum.
For the channel-by-channel integration, one needs to define channel limits for each
peak. For the Gaussian integration one needs to give approximate peak positions
and the full width at half maximum.
The program uses the user-supplied initial nuclear kinematic parameters to trace the
positions of peaks in all the spectra and perform the necessary calculations.
Background calculations are included in the program.
Examples of the output are presented in Figures 15.3-15.6. They were copied
directly from the computer-generated plots.
Figure 15.3 An example of the differential cross sections σ (θ ) / σ R (θ ) for the reaction
32
r r
S( d , d )32S leading to the ground state in 32
S calculated by the program LORNA from 198
particle spectra in about 1.5 minutes using the VAX VMS 3.6 computer. The angle is in the centre
of mass system. The figure has been copied directly from the computer output. The first label for
the horizontal scale shows the excitation energy. Here, the excitation energy is 0. The label E1 on
a linear scale means that the displayed numbers should be multiplied by 101. (For instance, in
this example, the number 4 on the horizontal scale means 4 × 101.) The label E1 on a logarithmic
scale means that the largest displayed number is 101. The numbers -1, 0, and 1 on the vertical
scale mean 10-1, 100, and 101, respectively.
Figure 15.4 An example of the angular distribution of the vector analyzing power iT11 (θ ) for the
r r
reaction 32S( d , d )32S leading to the ground state in 32S calculated by the program LORNA. The
numbers on the vertical scale should be multiplied by 10-1 as indicated by the label E-1. See also
the caption to Figure 15.3.
r r
Figure 15.5 The differential cross sections σ (θ ) for the reaction 32S( d , d ′ )32S* leading to the
2.23020 MeV excited state in 32S calculated by the program LORNA. The numbers 0, 1, and 2 on
the vertical scale mean 100, 101, and 102 as indicated by the label E2. See also the caption to
Figure 15.3.
Figure 15.6 The angular distribution of the vector analyzing power iT11 (θ ) for the reaction
r r
S( d , d ′ )32S* leading to the 2.23020 MeV excited state in 32S calculated by the program
32
LORNA. The numbers on the vertical scale should be multiplied by 10-1 as indicated by the label
E-1. See also the caption to Figure 15.3.
Output is in both tabulated and graphic forms. Results can be displayed on the
screen or plotted. The program calculates differential cross sections, analyzing
powers, and the errors. In the case of the differential cross section for the elastic
scattering, the program calculated also the Rutherford scattering cross sections. The
tabulated outputs contain both the list of the differential cross sections, σ (θ ) , and the
ratios of the differential cross sections to Rutherford cross sections, σ (θ ) / σ R (θ ) . In
the graphic form the user has a choice to display any of the two.
As mentioned earlier, the program has been written specifically for the S = 1
projectiles. However, the discussed formalism can be appropriately changed for S =
1/2 particles. For larger spins, the program can be extended using for instance
formulae presented by Darden (1971) for S = 3/2 or the general formalism discussed
by Haeberli (1974).
Acknowledgement
I want to thank my wife Lorna for her cheerful encouragement during the
development and testing of this computer code.
References
Barshal, H. H. and Haeberlie, W. (eds) 1971, Madison Convention, Proceedings of the
3rd International Symposium on Polarization Phenomena in Nuclear Physics,
Madison, 1970, University of Wisconsin Press, Madison, WS, p. xxv
Darden, S. E. 1971, Polarization Phenomena in Nuclear Reactions, eds H. H. Barshal
and W. Haeberli, University of Wisconsin Press, Madison, Wisconsin, p. 39.
Haeberli, W. 1974, Nuclear Spectroscopy and Reactions, ed. J. Cerny, Academic
Press, New York, p. 151 and references therein.
Marion, J. B. 1960, Nuclear Data Tables, Part 3, National Academy of Sciences,
National Research Council, Washington, DC.
Silverstein, E. A. 1959, Nucl. Instr. Meth. 4:53.
Schwandt, P. and Haeberlie, W. 1968, Nucl. Phys. A110:585.
16
Reorientation Effects in Deuteron Polarization
Key features:
1. This study represents the first systematic survey of the deuteron polarization with the
aim of investigating reorientation effects in the scattering of polarized deuterons.
2. Experimental part of this study was carried out using polarized deuterons supplied by
the C-LASKA Karlsuhe Lambshift polarized ion source. Polarized deuterons were
accelerated to 52 MeV using the Karlsruhe isochronous cyclotron.
3. Conversion of particle spectra to the angular distributions of the differential cross
sections dσ (θ ) / dΩ and vector analyzing powers iT11 (θ ) was carried out in
Heidlberg using the MPI VAX-780 computer and the computer code LORNA
described in Chapter 15.
4. Theoretical analysis was carried out using the coupled channels formalism and the
computer code ECIS79, which I have modified and adapted to run at the Australian
National University. All the theoretical calculations were done using the ANU UNIVAC
1100/82 computer.
5. The deuteron-nucleus interaction containing an imaginary term of the spin-orbit
potential has been found to be the most suitable in describing the experimental data.
28
6. In the single case of a strong oblate deformation (for Si) experimental distribution of
the vector analyzing power iT11 (θ ) for the first exited state 21 is distinctly different
+
than the distributions for all the remaining isotopes. This result suggests that
measurements of the iT11 (θ ) analyzing power for the 21 states could be used to
+
identify nuclei with strong oblate deformations.

28
7. With the exception of Si, all other nuclei studied here are either prolate or nearly
spherical. Experimental results show that the iT11 (θ ) distributions to the 21 states do
+
not depend strongly on the degree of the quadrupole deformation for all these nuclei.
8. I have carried out theoretical analysis using rotational and vibrational models. With
some small exceptions, they all produced similar shapes for the iT11 (θ ) distributions
+
for the 21 states indicating that reorientation effects are weak at this energy and for
these range of nuclei.
Abstract: The differential cross sections, dσ (θ ) / dΩ , and vector analyzing powers, iT11 (θ ) ,
were measured for the elastic and inelastic scattering of 52 MeV polarized deuterons from 20Ne,
22
Ne, 26Mg, 28Si, 32S, 34S, 36Ar and 40Ar nuclei. Coupled channels analysis was carried out using an
axially symmetric rotational model with either prolate or oblate quadrupole deformations for
each isotope. Calculations using harmonic vibrator model were also carried out. In general,
reorientation effects were found to be weak. A global optical model potential containing an
imaginary spin-orbit component was found to be the most suitable in describing the
experimental data at this energy. Our results indicate that vector analyzing power, iT11 (θ ) ,
can be used to identify nuclei with oblate deformation but it cannot distinguish between
prolate and spherical nuclei.
Introduction
Reorientation process is associated with self-coupling of an excited state and it
involves transitions between various magnetic substates, IM → IM ′ (see the
Appendix K). It has been pointed out that reorientation of electric quadrupole
moment for the 2+ states may play an important role in scattering of protons and
deuterons (Kurepin, Lombard and Raynal 1973; Raynal1976). For deformed nuclei,
the 2+ - 2+ matrix element, which is proportional to the quadrupole moment, Q2 + , may
have a strong influence on the analyzing powers. The effect is energy dependent
and is expected to be stronger at lower projectile energies. However, due to
contributions from indirect processes, such as energy fluctuations and resonance
scattering, interpretation of the low-energy data may be difficult. However, it was
reported that reorientation effects may be noticeable at such high energies as 65
MeV for protons and 56 MeV for deuterons (Hatanaka et al. 1981).
Prior to our investigation, no systematic study of reorientation effects in scattering of
polarized particles has been undertaken. However, from the available results,
confined mainly to energies below 25 MeV, certain patterns could be noticed.
In nearly all cases, the reorientation process was found to affect relatively strongly
the analyzing powers for scattering to the first, 21+ , excited states in even-even nuclei
and to a certain extend also the inelastic scattering differential cross sections.
Scattering to the second, 2+2 , states is influenced less strongly (Clement et al. 1980).
For oscillating analyzing powers, the reorientation process is claimed to shift
oscillations to smaller angles and damp the oscillation amplitude for prolate
quadrupole deformations. For oblate shapes, the effect is the opposite: oscillations
are shifted towards larger angles and their amplitude is increased.
Such behaviour was demonstrated for 24.5 MeV protons scattered from 152Sm
(Kurepin, Lombard and Raynal 1973), 9.4 MeV deuterons scattered from 28Si
(Clement et al. 1978), 20.5 MeV deuterons scattered from 154Sm (Clement et al.
1981), 20 MeV deuterons scattered from 24Mg, 28Si and 54Cr (Clement et al. 1980)
and 18 MeV deuterons scattered from 32S (Clement et al. 1981).
It should be noted that for 32S, reorientation effects were also studied using energy
averaged results corresponding to deuteron incident energy of Ed = 9.7 MeV
(Clement et al. 1978). At this energy, experimental data could not be fitted using
either prolate or oblate shapes. Good agreement between theory and experiment
was, however, obtained assuming the harmonic vibration model. On the other hand,
at Ed = 18 MeV, distinction between oblate and prolate shapes for 32S is clear with a
prolate quadrupole deformation giving a definitely better description of the
experimental results. Thus one set of data describes 32S as a harmonic vibrator while
the other shows it to be deformed with a prolate shape. It is clear, therefore, that
even in cases for which agreement between theory and experiment is good,
discrimination between nuclear shapes and models should be taken with caution.
At very low energies, ≤ 10 MeV, strong dependence on the sign of the quadrupole
deformation parameter, β2, was found not only for the inelastic but also for the elastic
scattering (Clement et al. 1978). At these energies experimental results had to be
averaged over a range of incident energies in order to minimise effects of resonance
scattering.
Less clear reorientation effects were reported for 20.3 MeV protons scattered from
28
Si (Blair et al. 1970), 12.3 MeV deuterons scattered from 54,56,58Fe (Brown et al.
1973), 10 MeV deuterons scattered from 24Mg (Clement et al. 1978) and 15 MeV
deuterons scattered from 56Fe (Raynal 1976). For 20.3 MeV protons and for 10 MeV
and 15 MeV deuterons, coupled channels calculations predict strong dependence of
the analyzing powers on the sign of β2 However, agreement between theory and
experiment is far from satisfactory. At 12.3 MeV deuteron energy, theoretical
predictions for prolate and oblate shapes are shown only for 58Fe. For this nucleus
fits to the elastic scattering favour an oblate deformation while the distributions for
the 21+ inelastic scattering are described better assuming a prolate shape.
At higher energies, a small but clear angular shift was reported for 65 MeV protons
scattered from 24Mg (Hatanaka et al. 1981). The displayed fit to the analyzing powers
for the 21+ state is exceptionally good for a prolate shape. However, fits to other
angular distributions are not shown. A more extensive coupled channels analysis
was carried out later for 65 protons scattered from 24Mg, 28Si and 32S (Kato et al.
1985). Results of this analysis indicated that good fits could be obtained only for the
elastic scattering. Fits to the inelastic scattering, and in particular to the 21+ analyzing
powers, are far from satisfactory. The discrepancy between theory and experiment is
particularly strong for the 21+ state in 24Mg. These calculations put in doubt the earlier
results of Hatanaka et al. (1981).
An attempt to distinguish between prolate and oblate shapes was also made for 65
MeV protons scattered from 28Si (Nakamura at al. 1978). However, the predicted
small angular shift for the two shapes and the poor quality of fits make the results
inconclusive.
For the 56 MeV deuterons, reorientation effects were reported for 24Mg and 28Si
(Hatanaka et al. 1981). At this energy the angular dependence of the vector
analyzing powers iT11 (θ ) for the 21+ states is distinctly different from that observed at
lower energies. The region sensitive to the sign of β2 was found to be confined to
angles between about 20° and 60°. By comparing coupled channels calculations with
the experimental data it was possible to distinguish between prolate and oblate
shapes with more clear discrimination being for 28Si. However, here again,
theoretical predictions for the two signs of β2 are shown only for the iT11 (θ )
distributions for the 21+ states. The aim of our study was to explore more
systematically this higher energy region for the sd-shell nuclei.
Measurements of angular distributions of differential cross sections dσ (θ ) / dΩ and
vector analyzing powers iT11 (θ ) for the elastic and inelastic scattering of 52 MeV
polarized deuterons were carried out using a Lamb-shift polarized ion source
(Bechtold at al. 1976; see also the Appendix G) and the Karlsruhe isochronous
cyclotron. The beam intensity on the target was about 5 nA and the overall energy
resolution of detected deuterons about 270 keV, which was mainly due to the beam
resolution.
The following targets were used in the measurements: 20Ne (99.9% enriched), 22
Ne
(99.7%), 26Mg(99.7%), Si (with 92.2% of 28Si), S (as SH2 with 95% of 32SH2), 34
S(,
containing 89.8% of 34SH2), 36Ar (99.5%) and Ar (containing 99.59% 40Ar).
Measurements of dσ (θ ) / dΩ and iT11 (θ ) were carried out in the range of angles of
about 10°-80° (lab), in steps of 1.5° for up to about 50° and in steps of 3° for larger
angles. The method of measurement is described in Chapter 15. The detector
system consisted of six ΔE-E solid-state counter telescopes placed at symmetric
angles with respect to the beam direction. The detectors were mounted on two
remotely controlled, moveable tables inside a large scattering chamber. The angular
distance between the adjacent detector telescopes on each table was 3° and the
angular resolution of the detector slit systems was ±0.5°. The thickness of the
detectors was 1500 μm or 2000 μm for the ΔE detectors, and 7000 μm (2000 μm +
5000 μm) for the E-detectors. The usual pulse multiplication method was used for the
particle identification (see Chapter 2). During the measurements the beam
polarization was flipped every few minutes according to a preset value of the beam
current integration counts (see Chapter 15).
r
The absolute value of the beam polarization was monitored using the 12C( d , d )12C
elastic scattering at θ = 47°(lab). This angle corresponds to an optimum value of the
figure of merit [iT11 (θ )] [dσ (θ ) / dΩ] with the analyzing power iT11 (47 0 ) = 0.318±0.035
2
being known from the double scattering measurements r (Seibt and Weddigen 1980).
12 12
Using this known value of iT11 (47 ) for the C( d , d ) C the beam polarization pz was
0
calculated by program LORNA (see Chapter 15) during the data reduction and used
to calculate the iT11 (θ ) for the studied reactions.
The beam polarimeter was mounted downstream, outside the main scattering
chamber and it was followed by a Faraday cup, which was used in the
measurements of the integrated charge. The target of the r beam polarimeter
12 12
consisted of a large polyethylene foil. Deuterons from the C( d , d ) C reaction were
detected by two Nal(Tl) detectors placed symmetrically in respect of the deuteron
beam. The thickness of the Nal(Tl) crystals was chosen in such a way as to allow
r for
a clear separation of deuterons from the high energy protons from the 12C( d , p )13C
reaction. In addition, a thin Al foil was mounted in front of each crystal to suppress
the Z ≥ 2 particles. The average beam polarization, pz, during the measurements was
0.46 ±0.05 and its stability was within about 2% over a long period of data collection.
Deuteron spectra from the main detectors and from the beam monitor were stored
on magnetic tapes and were analysed off-line using an MPI VAX-780 computer
system. All data reduction calculations were carried out using my computer code
LORNA (Nurzynski 1985; see also Chapter 15).
As described in Chapter 15, the program performs a global analysis of particle
spectra and converts them to angular distributions of differential cross sections and
analyzing powers for up to 20 particle groups for each target nucleus. Typically about
200 spectra were taken for each target. Program LORNA converted them to angular
distributions in about 1.5 min for each excitation energy. Calculations of the errors of
the experimental data include statistical uncertainties, background subtraction and
beam polarization errors.
All experimental distributions are shown together with theoretical calculations in the
next section. However, as mentioned earlier, the iT11 (θ ) analyzing powers for the
21+ states are expected to show dependence on the sign of β2. It is, therefore,
interesting to compare them separately in one figure. Figure 16.1 shows that, except
for one isotope, 28Si, all iT11 (θ ) distributions for the 21+ states display similar features.
These results will be discussed further later.
Figure 16.1. Vector analyzing power iT11 (θ ) for the first excited states 21 measured using 52 MeV
+
polarized deuterons. The lines are to guide the eye.
Coupled channels formalism
The principles of the coupled channels formalism are described in the Appendix E.
However, it is useful to mention here some points related to the calculations of our
data.
The Schrödinger equation for the interaction between incident and target nucleus
can be written as:
[T − V (r , ξ ) + H (ξ )]ψ (r , ξ ) = Eψ (r , ξ )
where
h2 2
T =− ∇r
2μ
is the kinetic energy operator of the incident particle, V (r , ξ ) is the interaction
potential between the incident particle and target nucleus, H (ξ ) is the intrinsic
Hamiltonian for the target nucleus, and ξ are the nuclear coordinates.
The intrinsic nuclear states φi (ξ ) are the solutions of the equation:
H (ξ ) χ i (ξ ) = ε i χ i (ξ )
where i labels the intrinsic states.
The intrinsic wave functions form a complete orthogonal set and consequently the
total wave function ψ (r , ξ ) can be expressed as a sum of these functions:
ψ (r , ξ ) = ∑ ϕi (r )χ i (ξ )
i
where the sum is over all states of the target nucleus, discrete and continuum.
Substituting it in the original Schrödinger equation we get a set of coupled equations
for all channels:
(T − E − ε i )ϕi (r ) = ∑Vikϕ k (r )
k
where
Vik = ∫ ϕi∗ (ξ )V (r , ξ )ϕ k (ξ )dξ
The interaction potential V (r , ξ ) depends on the way the target nucleus is described.
For example, for a simple axially symmetric quadrupole deformation, the potential
V (r , ξ ) will have the form:
dV
V (r , ξ ) = V (r − R (θ ,φ )) ≈ V (r − R0 ) − β 2 R0Y20 (θ ,φ )
dr
where
R (θ ,φ ) = R0 (1 + β 2Y20 (θ ,φ ))
β 2 is the quadrupole deformation parameter, and Y20 (θ ,φ ) is the spherical harmonic

function.
For a simple vibrational model
⎛ ⎞
R(θ ,φ ) = R0 ⎜1 + ∑ α m∗ Y2m ∗ (θ ,φ ) ⎟
⎝ m ⎠
where α m∗ are the dynamical distortion parameters that create or annihilate vibrational
phonon of angular momentum 2 and z component m. The mean-square deformation
parameter is given by:
β 22 = 0 ∑ α m 0
2
The potential V (r , ξ ) is then given by:

dV
V (r , ξ ) = V (r − R) ≈ V (r − R0 ) − ∑ α m∗ Y2m∗ (θ ,φ ) R0
m dr
In practice, even with coupled channels formalism, only a small number of coupled
equations are used. In the case discussed here, i.e. for the 0+- 2+ excitations, there
are six coupled equations, one for 0+ and five for 2+. The sign of the quadrupole
moment is related to the sign of the 2+- 2+ nuclear matrix element, which is
proportional to the quadrupole moment Q2 + . For spherical nuclei the matrix element
is zero.
The reorientation effect for the inelastic scattering can be studied by performing
three sets of calculations for each target nucleus: two for negative and positive
deformations and one for vibrational model.
I have carried out the coupled channels analysis our data using the well-known
computer code ECIS79 (Raynal, 1972, 1081), which I have modified and adapted to
run at the Australian National University. All the calculations were performed using
the ANU UNIVAC 1100/82 computer.
For each target isotope four distributions, dσ (θ ) / dΩ and iT11 (θ ) , for the ground and
for the first 21+ excited states, were fitted simultaneously. In order to see whether
theoretical fits are sensitive to the sign of the quadrupole deformation, independent
searches of potential parameters were carried out using either positive or negative β2
parameters for each target isotope. In all these calculations an axially symmetric
rotational model containing quadrupole and hexadecapole deformations was
assumed for all nuclei. The central and the spin-orbit components of the deuteron-
nucleus interaction potential were assumed to be described by the same deformation
parameters.
The optical model parameter search was first carried out using as the starting values
the potential parameters derived earlier by Mairle et al. (1980) for the 52 MeV
deuteron elastic scattering. Unfortunately, searches based on any of their four sets
of parameters failed to fit the measured by us angular distributions.
The potential F' of Daehnick, Childs and Vrcelj (1980), containing an imaginary spin-
orbit component, was tried next and was found to give a significantly better
description of the experimental results. These authors carried out an extensive
global optical model analysis of deuteron scattering at energies 11.8, 17, 34, 52, and
79.5 MeV in the mass range of A = 27 - 232. Their potential F' contains five nuclear
components and the usual Coulomb potential for a uniformly charged sphere with
the radius Rc = rc A1 / 3 . The nuclear components are defined by a total of 13
parameters. The potential I used in my calculations was similar to their potential but
the total number of parameters for nuclear components was 15.16
U (r ) = VC (r ) + U 0 (r ) + U s.o. (r )
where VC (r ) is the Coulomb potential, U 0 (r ) and U s.o. (r ) are the central and spin-orbit
components of the optical model potential.
d
U 0 (r ) = −Vf (r , r0 , a0 ) − i 4aDWD f (r , rD , aD ) − WS f (r , rS , aS )
dr
1 ⎡ df (r , rrso , arso ) df (r , riso , aiso ) ⎤
U s.o. (r ) = D 2π ⎢Vso + iWso ⎥⎦S ⋅ L
r⎣ dr dr
where
2
⎛ h ⎞
D π = ⎜⎜
2
⎟⎟ = 2 fm 2
⎝ mπ c ⎠
1 r − ri A1 / 3
f (r , ri , ai ) = with xi = and i = 0, D, S , rso, or iso
1+ e xi
ai
Table 16.1
The original set of parameters for the potential F'
WD (12 + 0.031E)e β
V 88.0 − 0.283E + 0.88ZA1 / 3 WS (12 + 0.031E )(1 − e β )
rV 1.17 rD = rS 1.376 − 0.01 E
aV 0.717 + 0.0012 E a D = aS 0.52 + 0.07 A1 / 3 − 0.04∑i e − μ i
Vso 5.0 Wso 0.37 A1 / 3 − 0.03E

rrso 1.04 riso 0.80
arso 0.60 aiso 0.25

β = −(E / 100) 2 , μi = [( M i − N ) / 2]2 , M i is the magic number 8, 20, 28, 50, 82, 126,
N – neutron number, E – deuteron energy in the laboratory system in MeV. rc = 1.3 fm.
16
Their potential has the same geometric parameters for the surface and volume absorption; hence
the total number of parameters is 13. In my calculations I have relaxed this restriction and assumed
that the geometric parameters can have different values for these two components and thus the total
number of parameters was increased to 15.
In the initial series of calculations I have analysed the experimental results for each
nucleus using either positive or negative quadrupole deformation parameter β2 and
searching on all 15 parameters in groups of up to 10 parameters at a time. Later, in
the final stages of parameter optimisation, I have searched on all 15 parameters
starting with the values determined earlier.
My calculations have shown that all 15 parameters fluctuated around certain
smoothly varying, mass-dependent values. Close examination of all these
parameters suggested that many of them could be fixed at their original values and
that the 15-parameter search could be reduced to a search on only four parameters.
In particular, the analysis involving searches on all 15 parameters indicated that (a)
the depth of the central volume absorption potential, WS, should have a fixed value of
2.07 MeV for all target isotopes; (b) in contrast with the original potential F', the
central surface and volume absorption potentials should have different radius
parameters, and (c) all other parameters, except for V, a0 , WD and rD , could be
assumed to have the values given by Daehnick, Childs and Vrcelj (1980). In
particular, I have found no compelling evidence for altering the parameters of the
spin-orbit components from their original values.
Taking into account the results of the 15 parameters search, the analysis was
repeated by searching on only four parameters, V, a0 , WD and rD . In general, the
resulting fits were found to be similar to those obtained by searching on all 15
parameters. However, the searching procedure was not only considerably easier and
faster but also it eliminated some spurious parameter fluctuations.
The four individually adjusted parameters for each target isotope and for each sign of
the β2 parameter, are shown in Figure 16.2. They were found to be close to their
original values, which are also shown in the same figure. Deformation parameters, β2
and β4, used in the coupled channels calculations are listed in Table 16.2.
Figure 16.2. The four, out of 15, optical model parameters that had to be adjusted to optimise the fits
to the experimental angular distributions of the differential cross sections and vector analyzing powers
for the ground states and first excited states of the sd-shell nuclei. The full circles correspond to the
full lines in Figures 16.3-16.6 for the rotational model with β2 < 0 for 28Si and β2 > 0 for all the
remaining nuclei. The open circles are the parameters for the reversed signs of the deformation
parameters β2. The full lines show the original parameter values of the potential F’ as determined by
Daehnick, Childs and Vrcelj (1980) and as listed in Table 16.1.
Figure 16.3. The experimental results (dots) for the elastic and inelastic scattering of 52 MeV
polarized deuterons from 20Ne and 22Ne are compared with the theoretical calculations. Errors smaller
than the experimental points are not shown. The coupled channels calculations for an axially
symmetric rotational model with prolate or oblate deformations are shown as full and dotted lines,
respectively. The dashed lines correspond to calculations using a harmonic vibration model.
Figure 16.4. Results for 26Mg and 28Si. See the caption to Figure 16.3. However, here the continuous
line for 28Si corresponds to an oblate deformation.
Figure 16.5. Results for 32S and 34S. See the caption to Figure 16.3.
Figure 16.6. Results for 36Ar and 40Ar. See the caption to Figure 16.3.
Results of the four parameter search for each isotope were taken as representing
the best theoretical predictions at this deuteron energy. They are displayed in
Figures 16.3-16.6 for β2 >0 and β2 < 0. With the exception of 28Si, the full and dotted
lines are for the calculations using prolate and oblate deformations, respectively. For
28
Si the representation is reversed because this nucleus has an oblate deformation.
Finally, in order to see to what extent theoretical predictions are model-dependent,
the four-parameter search was also carried out assuming a harmonic vibrational
model for each isotope. Results of the calculations are shown as dashed lines in
Figures 16.3-16.6, and the corresponding coupling parameters β02 are listed in Table
16.2. In these calculations, the final parameters V, a0, WD and rD were found to be
the same as the parameters corresponding to the full lines in Figures 16.3-16.6.
Table 16.2
Parameters β2, β4 and β02 used in the coupled channels analysis of the 52 MeV polarized deuteron
scattering from the sd-shell nuclei
20 22 26 28 32 34 36 40
Ne Ne Mg Si S S Ar Ar
β2 a) +0.50 +0.37 +0.30 -0.34 +0.27 +0.20 +0.18 +0.17
β 4 a) +0.05 +0.05 -0.03 +0.08 -0.20 -0.20 +0.10 +0.10
β 2 b) -0.53 -0.44 -0.35 +0.34 -0.33 -0.24 -0.18 -0.20
b
β4 ) +0.05 +0.05 -0.03 +0.08 -0.20 -0.20 +0.10 +0.10
β0 2 )
c
0.50 0.37 0.30 0.34 0.27 0.20 0.17 0.17
Q2+ d) -23(3) -19(4) -21(2) +17(3) -15(2) +04(3) +11(6) +01(4)
a
) Deformation parameters associated with the full lines in Figures 16.3-16.6. The corresponding optical model
parameters are shown as full circles in Figure 16.2.
b
) Deformation parameters associated with the dotted lines in Figures 16.3-16.6. The corresponding optical
model parameters are shown as open circles in Figure 16.2.
c
) Coupling parameters used for harmonic vibrational model calculations. The resulting theoretical angular
distributions are shown as dashed lines in Figures 16.3-16.6.
d
) Quadruple moments Q2+ in e fm2. The listed values include the uncertainty of the listed values. For
instance, -23(3) means -23±3 e fm2 (Stone 2001).

Measurements of the differential cross sections and analyzing powers were carried
out using a 52 MeV beam of deuteron polarized by the C-LASKA Karlsuhe Lambshift
source and accelerated by the Karlsruhe isochronous cyclotron. Particle spectra were
stored on magnetic tapes and analysed using my computer code LORNA (see Chapter
15) and the VAX-780 computer at the Max-Plank Institute für Kernphysik, Heidelberg.
This has resulted in a total of 32 angular distributions of the differential cross
sections and vector analyzing powers for the elastic and inelastic scattering on 20Ne,
22
Ne, 26Mg, 28Si, 32S, 34S, 36Ar and 40Ar nuclei. Theoretical analysis of our experimental
results was carried using the Australian National University UNIVAC 1100/82
computer and the coupled channels code ECIS79 (Raynal, 1972, 1981), which I
have modified and adapted to run at ANU.
A compilation prepared by Stone (2001) shows that only 28Si nucleus has a relatively
large positive quadrupole moment Q2 + =16±3 efm2. Another nucleus that might have
36
a relatively large positive quadrupole moment is Ar, for which Q2 + = 11±6 efm2 .
However, the experimental error is also large and thus the moment could be close to
zero. The quadrupole moments for 34S and 40Ar are practically zero. Thus, the
heaviest isotopes in this study are either spherical or almost spherical. The
remaining nuclei have relatively large quadrupole moments.
Our experimental survey of the vector analyzing powers for the 21+ states in the sd-
shell nuclei presented in Figure 16.1 shows that the shapes of the iT11 (θ ) distributions
for all nuclei except one (28Si) have similar oscillation pattern. This group contains
both prolate (20Ne, 22Ne, 26Mg and 32S) and nearly spherical (34S, 36Ar, and 40Ar)
nuclei. This survey therefore indicates that the underlying process of deuteron
polarization at 52 MeV helps to identify only nuclei with strong oblate shapes; it does
not allow for a distinction between prolate, spherical or nearly spherical nuclei.
In my coupled channels analysis of our results I have initially used the optical model
parameters found earlier by Mairle et al. (1980) for 52 MeV deuterons. These
parameters served as the starting values but any attempt to optimise the fits to our
data by searching around these values was unsuccessful. I have found that the
global potential F' of Daehnick, Childs and Vrcelj (1980), which contains an
imaginary spin-orbit component produced significantly better results. The slightly
modified version of this potential contains 5 components with the total of 15
parameters.
Extensive search on all 15 parameters indicated that 10 of them could be kept at the
values defined by the original potential F'. The depth of the volume absorption
potential WS had to be lowered to 2.07 MeV and was found to have the same value
for all the investigated isotopes. The search was then reduced to only 4 parameters,
V, a0, WD and rD .
Considering the calculations for the iT11 (θ ) to the 21+ states, as presented in Figures
16.3-16-6, it is clear that the best distinction between prolate and oblate quadrupole
shapes is for 26Mg and 28Si nuclei. For 26Mg, the prolate and oblate deformation
produce clear out-of-phase oscillation for angles around 300-600. For 28Si, clear out-
of-phase results are for angles around 250-450.
Some dependence on the sign of the quadrupole deformation can be also seen for
32
S and 34S nuclei. Calculations for these nuclei show a phase shift at angles around
300 and an out-of-phase feature at around 450. In both cases, the calculated curves
for the negative deformation in this region of angles is shifted to higher angles as
compared with the curves calculated for the positive deformation.
Similar, but less pronounced, features are also observed for 36Ar and 40Ar nuclei.
However, for 20Ne and 22Ne there are no clear differences between calculations for
the opposite signs of quadrupole deformation.
Fits to 20,22Ne are relatively poor and the agreement between the experimental and
theoretical results is only marginally better for prolate shapes. Experimental results
for this pair of isotopes were found to be the most difficult to fit when using the
coupled channels procedure. The easiest to fit were the experimental results for
32,34
S and for 36,40Ar isotopes. Figures 16.5 and 16.6 show that the assumption of
prolate shapes gives only a little better fit for these nuclei.
It is also interesting to compare rotational and vibrational model calculations. As
mentioned earlier, the basic difference between the two models is in the 2+- 2+ matrix
elements, which vanish for the harmonic vibrator. Results presented in Figures 16-3-
16-6 show that, in general, there are no significant differences between the rotational
and the vibrational model calculations. Even for 26Mg and 28Si nuclei, calculations
using prolate shapes are similar to those obtained using harmonic vibrator. This is in
contrast with the results at lower incident energies where for strongly deformed
nuclei, differences between rotational and vibrational model calculations are
generally better pronounced, even in cases when the fits to the experimental data
are poor.
In summary, this study shows that, in general, reorientation effects at 52 MeV
deuteron energy for the sd-shell nuclei appear to be weak. The experimental
iT11 (θ ) distributions for nuclei with positive prolate deformations are found to have
similar features. Only in a single case of a strong oblate deformation, i.e. for 28Si, a
distinctly different experimental iT11 (θ ) distribution for the 21+ state was observed.
Furthermore, for only two isotopes, 26Mg and 28Si, distinctly different
iT11 (θ ) predictions for the two signs of β2 were obtained. For 20Ne and 22Ne, an
assumption of large oblate deformations did not result in altering significantly the
calculated iT11 (θ ) distributions for the 21+ states. However the overall fits to all four
distributions for each of these two isotopes appear to favour the correct sign of β2.
Finally, my analysis has shown that the best description of our experimental data
required a potential with both real and imaginary spin orbit components. The
potential is described by 15 parameters but 10 of them have the same values as
found earlier by Daehnick, Childs and Vrcelj (1980). Of the remaining five
parameters, the depth of the central volume absorption potential had to be lowered
to 2.07 MeV and was constant for all target nuclei. Only four parameters V, a0, WD
and rD had to be adjusted individually, but generally only slightly, to optimise the
theoretical description of the data.
References
Bechtold, V., Friedrich, L., Finken, D., Strassner, G. and Ziegler, P. 1976, Proc. Fourth
Int. Symp. on Polarization Phenomena in Nuclear Reactions, Zürich, 1975, ed. W.
Gruebler and V. Konig (Birk-hauser, Basel) p. 849
Blair, A.G., Glashausser, C., de Swiniarski, R., Goudergues, J., Lombard, R., Mayer, B.,
Thirion, J. and Vaganov, P., 1970, Phys. Rev. C1:444.
Brown, R.C., Debenham, A.A., Griffith, J.A.R., Karban, O., Kocher, D.C. and Roman, S.
1973, Nucl. Phys. A208:589.
Clement, H., Frick, R., Graw, G., Merz, F., Schiemenz, P., Seichert, N. and Hsun, S. T.
1980, Phys. Rev. Lett. 45:599.
Clement, H., Frick, R., Graw, G., Oelrich, I., Scheerer, H.J., Schiemenz, P., Seichert, N.
and Hsun, S. T. 1981, Proc. Fifth Int. Symp. on Polarization Phenomena in
Nuclear Physics, Santa Fe, NM, 1980, ed. G. G. Ohlsen, R.E. Brown, N. Jarmie,
W.W. McNaught and G.M. Hale, AIP Conf. Proc. No. 69 (AIP, New York,) p. 376
Clement, H., Graw, G., Kretschmer, W. and Stack, W. 1978, J. Phys. Soc. Jpn. Suppl.
44:570
Daehnick, W. W., Childs, J. D. and Vrcelj, Z. 1980, Phys. Rev. C21:2253.
Hatanaka, K., Nakamura, M., Imai, K., Noro, T., Shimizu, H., Sakamoto, H., Shirai, J.,
Matsusue, T. and Nisimura, K. 1981, Phys. Rev. Lett. 46:15
Kato, S., Okada, K., Kondo, M., Hosono, K., Saito, T., Matsuoka, N., Hatanaka, K.,
Noro, T., Nagamachi, S., Shimizu, H., Ogino, K., Kadota, Y., Matsuki, S. and
Wakai, M. 1985, Phys. Rev. C31:1616.
Kurepin, A. B., Lombard, R.M. and Raynal, J. 1973, Phys. Lett. 45B:184.
Mairle, G., Knopfle, K. T., Riedesel, H. and Wagner, G. J. 1980, Nucl. Phys. A339:61
Nakamura, M., Sakaguchi, H., Imai, K., Hatanaka, K., Goto, A., Noro, T., Ohtani, F.,
Kobayashi, S., Hosono, K., Kondo, M., Kato, S., Ogino K. and Kadota, Y. 1978, J.
Phys. Soc. Jpn. Suppl. 44:557.
Nurzynski, J., 1985, Comp. Phys. Commun. 36:295.
Raynal, J. 1972, Computing as a language of physics (IAEA, Vienna,) p. 281.
Raynal, J. 1976, Proc. Forth Int. Symp. on Polarization Phenomena in Nuclear
Reactions, Zurich, 1975, ed. W. Grüebler and V. König (Birkhauser, Basel,) p. 271
Raynal, J. 1981, Phys. Rev. C23:2571.
Seibt, E. and Weddigen, C., 1980, Nucl. Instr. Meth. 100:61.
17
Two-step Reaction Mechanism in Deuteron Polarization
Key features:
1. We have measured angular distributions of the differential cross sections σ 0 (θ ) and

analyzing powers iT11 (θ ) , T20 (θ ) , T21 (θ ) and T22 (θ ) for the elastic scattering of 12
MeV polarized deuterons from 76,78,80,82Se isotopes.
2. Our measurements revealed an unusual behaviour of the analyzing powers: the
amplitudes increased with the increasing mass number A of the target nuclei. This is
contrary to the normal behaviour, which is characterized by the decreasing
amplitudes.
3. I have carried out coupled channels analysis of our experimental results and I have
found that this unusual behaviour can be explained as being due to the contributions
+
from an indirect, two-step, elastic scattering via the first 21 excited states in the target
nuclei. Thus, the observed elastic scattering is made of two components: the normal
direct scattering (d,d) and the two-step scattering (d,d’)2+(d’,d).
4. I have also studied other two-step contributions: (d,t)(t,d) and (d,p)(p,d) via the 2p1/2
or 1g9/2 configurations. I have found that their share in the reaction mechanism is
negligible.
5. This study represents the first clear demonstration of the unusual mass-dependence
of deuteron polarization and a demonstration of the importance of second-order
interaction in the elastic scattering.
Abstract: Angular distributions of the differential cross sections σ 0 (θ )

and of iT11 (θ ) ,
r
T20 (θ ) , T21 (θ ) and T22 (θ ) analyzing powers have been measured for the (d , d ) scattering from
76,78,80,82
Se isotopes at 12 MeV. An unusual mass-dependence of the analyzing powers was
observed. Coupled channels analysis explained the experimental results as being due to
+
contributions of the two-step elastic scattering (d,d’)2+(d’,d) via the first 21 excited states in the
target nuclei. Two-step processes (d,t)(t,d) and (d,p)(p,d) via the 2p1/2 or 1g9/2 configurations
have been also investigated but have been found to have negligible influence on the measured
distributions.
Introduction
While visiting the Laboratorium für Kernphysik in Zürich, Switzerland, I proposed a
study of deuteron polarization using Se isotopes. Normally, one should expect a
predictable mass-dependent behaviour of the analyzing powers. However, I thought
that it would be interesting to see whether Se isotopes would reveal some new,
unexpected features.
As discussed in Chapter 14, selenium isotopes present an interesting case where
neutron configurations 2p3/2, 1f5/2, and 2p1/2 are nearly completely occupied and
where virtually only one configuration, 1g9/2, is filling in as the mass of the selenium
isotope increases. Our results for the (p,d) reactions indicated that the neutron
occupation numbers for the 1g9/2 increased continuously from 40% for 76Se to 80%
for 82Se. These results were in close agreement with the calculations based on the
pairing theory.
The idea behind the proposed measurements of deuteron polarization was to see
whether this systematic closing of the f-p shell might be reflected in the angular
distributions of the analyzing powers. Selenium isotopes belong to a very limited
group of nuclei that can be used for such a study.
Experimental method
Angular distributions of the differential cross sections σ 0 (θ ) , vector analyzing power
iT11 (θ ) and all three tensor analyzing powers T20 (θ ) , T21 (θ ) and T22 (θ ) for the elastic
scattering of 12 MeV polarized deuterons from the 76,78,80,82Se isotopes were
measured using the ETHZ atomic beam polarized ion source and EN Van de Graaff
tandem accelerator. The method of polarization measurements was discussed in
Chapter 15.
A diagram of the reaction chamber is shown in Figure 17.1. The beam entered the
chamber from the left through a collimator, passed through the target and was
collected in a Faraday cup, which was equipped with an electrostatic suppressor
electrode. The scattered particles were collimated by rectangular slit system
consisting of identical slits with antiscattering baffles places between the defining
apertures.
Figure 17.1. Cross section of the reaction chamber used in the measurements of deuteron
polarization on Se isotopes. 1 – insulated slits; 2 – collimator tube; 3 – Se target; 4 – suppressor
electrode; 5 – Faraday cup; 6 – turntables; 7 – solid state detectors; 8 – detector slit assembly; 9 –
bearings for rotating the chamber about the beam axis.
The detectors were mounted on two plates that could be rotated independently
around the target. On each turntable there was room for up to seven detectors
placed at positions of 150 apart. With this system angular distributions could be
measured for reaction angles θ angles between 200 and 1600. Each turntable could
be adjusted from outside to the desired position within 0.10. In general, only four
detectors were used on each turntable because the electronic system had been
designed to accommodate a maximum of only eight detectors. The whole chamber
could be also rotated along the beam axis.
The electronic system was relatively simple and is shown in the diagram in Figure
17.2.
Figure 17.2. Block diagram of the electronic system used in measurements of deuteron polarization
for Se isotopes. PA – preamplifier; Ampl. – amplifier; Disc. – discriminator.
Selenium targets were prepared using the method developed at Australian National
University (see Chapter 14). In order to prevent sublimation of selenium under beam
bombardment, targets were coated on both sides with carbon films of approximately
20 μg/cm2. The target thickness was around 600 μg/cm2. However, in order to
reduce the data collection time, two targets of the same isotope were stacked to
produce a combined thickness of approximately 1200 μg/cm2. The isotopic
enrichment of the target material was 86%, 98%, 94%, and 97% for 76Se, 78Se, 80Se,
and 82Se, respectively.
Separation of the atomic substates was achieved in the field of a tapered sextupole
magnet. The source was equipped with a weak-field and two strong-field RF
transitions to produce vector and tensor polarization states of the deuteron beam
(see the Appendix G). The spin direction was adjusted using a Wienfilter and
switching the sign of the beam polarization was done every few seconds. This
method eliminates first-order errors arising from geometrical effects and from
inaccuracies of the required spin direction (see Chapter 15).
Typical beam current from the source was around 120 nA and on the target around
50 nA. The polarization of the deuteron beam was about 87% of the theoretical value
(i.e. pz = ±0.58 or τ 10 = ±0.71).
Experimental results are displayed in Figures 17.4 and 17.5. Preliminary data
reduction was carried out at the time of measurements, and I well remember the
excitement we felt when, point by point, they were gradually revealing a clear new
behaviour. Our results show that contrary to the normal behaviour, the amplitudes of
the analyzing power increase with increasing atomic mass number A of the target
nucleus. The clearest effect is for the vector analyzing powers iT11 (θ ) . However, the
effect is also present for T20 (θ ) and T22 (θ ) at backward angles.
Figure 17.3. The unusual behaviour of the analyzing powers: the increasing amplitudes with the
increasing mass of the target nuclei. Experimental angular distributions of the vector iT11 (θ ) and
tensor T20 (θ ) analyzing powers for the elastic scattering of 12 MeV polarized deuterons from Se
isotopes. The lines are to guide the eye. The horizontal lines are to help to see how the amplitudes of
the analyzing powers increase with the increasing atomic mass number A of the target nuclei.
Figure 17.4. Experimental angular distributions of the T21 (θ ) and T22 (θ ) tensor analyzing powers for
the elastic scattering of 12 MeV polarized deuterons from Se isotopes. The lines are to guide the eye.
The values for the T21 (θ ) component are too small to see the mass-dependence of its amplitudes but
the T22 (θ ) distributions show an increase in the absolute values of the analyzing power at backward
angles with the increasing atomic mass number A.
Figure 17.5. The unusual experimental results for Se isotopes: the absolute values iT11 (θ ) of the
vector analyzing power increase with the increasing atomic mass number A.
Figure 17.6. The normal behaviour of the absolute values iT11 (θ ) of the vector analyzing power. The
amplitudes decrease with the increasing atomic mass number A. This figure is based on our study for
a wider range of target nuclei (Bürgi et al. 1980)
This interesting mass-dependence is also displayed in Figure 17.5, which shows the
absolute amplitudes iT11 (θ ) . As can be seen, there is a clear and systematic
increase of the polarization amplitudes with the increasing mass number A. For
comparison, Figure 17.6 shows the normal and well-known behaviour of the absolute
values of the amplitudes of the vector analyzing iT11 (θ ) characterised by the
decreasing amplitudes.
The interpretation of the unusual behaviour of the analyzing powers for Se isotopes
was not immediately obvious but it suggested a presence of a systematic reaction
mechanism associated with changes in the internal structure of Se isotopes.
It so happened that around that time I visited the University of Colorado where I
had a chance to talk to Peter Kunz who informed me about his calculations for
the elastic scattering of polarized tritons. His calculations indicated an
enhancement of the analyzing powers by second-order processes. This sounded
interesting and I wondered whether our experimental results could be explained
by the presence of such second-order processes.
Peter wrote a Coupled Channel Born Approximation program, CHUCK, which he
kindly shared with me. I have brought his program to Canberra and I have
adapted it to run on the Australian National University UNIVAC 1100/42
computer.
The program allows for an easy calculation of a variety of second-order
processes. However, it can calculate only the differential cross sections and
vector analyzing power iT11 (θ ) . It does not include calculations of higher order of the
analyzing powers. This restriction did not seem to be important because our unusual
experimental results were most prominent for the iT11 (θ ) distributions.
I have used this program in the analysis of our experimental results for Se isotopes,
but first I decided to carry out the conventional optical model calculations.
In all my calculations, I have used standard parameterisation of the optical model
potential as described in Chapter 16 but I have used only three components: the
central real component, the central surface-absorption component, and the real spin-
orbit component. The potentials were described by 9 parameters: three potential
depths parameters (V, WD, and Vs.o.) and six geometrical parameters
( r0 , a0 , rD , aD , rs.o. , and as.o. ).
Optical model analysis revealed that the differential cross sections and
iT11 (θ ) angular distributions could be fitted by adjusting only one parameter, the depth
of the central imaginary component WD. The final values of this parameter were:
17.25, 16.50, 15.50, and 14.50 MeV for 76Se, 78Se, 80Se, and 82Se isotopes,
respectively. All parameters used in the optical model calculations and in the coupled
channels analysis are listed in Table 17.1.
This was an interesting result, because it showed that the contribution of non-elastic
processes in the deuteron-nucleus interaction, as represented by WD, was
decreasing gradually with the increasing atomic mass number A. Thus, by explicitly
including at least some of these non-elastic channels in the calculations one might
expect to describe the observed features of the elastic scattering by using a set of
fixed parameters for all target isotopes.
Having done the standard optical model analysis, I have then carried out an
extensive analysis of contributions from the second order processes (d,d’)(d’,d),
(d,t)(t,d), and (d,p)(p,d). In this notation, the first process describes the elastic
scattering via the inelastic scattering channels. The inelastic scattering (d,d’) is
quickly followed by a return of the target nucleus to its ground state and the outgoing
deuteron is detected as if it were elastically scattered. This intermediate interaction
cannot be detected by observing the energy of the outgoing deuteron but it might
influence the observed angular distributions of the differential cross sections and
analyzing powers.
In the second process, the neutron pickup reaction (d,t) is quickly followed by the
neutron stripping (t,d) and the outgoing deuteron is also detected as if it were
elastically scattered. In the third process the neutron stripping reaction (d,p) is
followed by the neutron pickup (p,d).
In the intermediate inelastic channel, I have considered direct excitations of the first
21+ and 31− levels. In the neutron pickup and stripping channels, I have considered
neutron transfers via the 2p1/2 and 1g9/2 configurations. All these two-step processes
are summarised in Figure 17.7. More complicated second-order processes could be
considered but my calculations indicated that they were unnecessary. In fact, I have
found (see Figure 17.9) that the only second-order process that has detectable
influence on the elastic scattering is the inelastic scattering via the first excited state
21+ .
Measurements of Coulomb excitation for selenium isotopes (Barrette et al.
1974) show that the deformation parameters derived from the B( Eλ ) γ-transition
probabilities for the 21+ and 31− states decrease with the increasing number of
neutrons in the target nuclei (see Figure 17.8). Spectra taken at a few angles for
the inelastically scattered particles indicated relatively strong excitation of these
two states. The observed mass dependence of the iT11 (θ ) distributions was
therefore expected to be associated with the two-step scattering via these
excited states.
Fig. 17.7. Two-step processes considered in my analysis of the elastic scattering of polarized
deuterons from Se isotopes. Coupling strengths between the elastic and inelastic channels are
proportional to the deformation parameters, which were derived from the B ( Eλ ) γ-transition
probabilities as reported by Barrette et al. (1974). For the intermediate reaction channels, all
calculations (except for a study of effects associated with the spectroscopic strength distributions)
were carried out using total spectroscopic strengths for configurations corresponding to states located
at the relevant centre-of-gravity energies.
Figure 17.8. Energy levels diagram for the 76,78,80,82Se isotopes. The pathways for gamma transitions
and the relevant deformation parameters derived from the B ( Eλ ) γ − transition probabilities as
reported by Barrette et al. (1974) are shown in the figure. It should be noted that the values of β2 and
+ −
β3 parameters for the 21 and 31 states (used in my coupled channels calculations) decrease with the
atomic number of the selenium isotope.
In the coupled channels calculations, the coupling strength for the two-step
processes via the intermediate inelastic channels was defined by the defor-
mation parameters β2 and β3 derived directly from the experimental values of
the B( Eλ ) γ-transition probabilities. In the case of the second-order processes
involving transfer reactions I have assumed that the configurations 2p1/2 and
1g9/2 are located in just two states placed at their corresponding centre-or-
gravity energies and that their respective coupling strengths are given by the
sum of the spectroscopic factors. The normalization of the reaction amplitudes
has been done assuming the coefficients D0 (d,p) = -122 MeV · fm3/2 and D0 (d,t)
= -182 MeV · fm3/2 (Schneider, Burch, and Kunz 1976). In the more familiar form
(see the Appendix E), where the square values are used, these coefficients
correspond to D02 = 1.5 × 104 MeV2 · fm3 for the (d,p) reaction and D02 = 3.3 × 104
MeV2 · fm3 for the (d,t).
The reaction form factors have been calculated using standard parameters r0 =
1.25 fm, a = 0.65 fm and the potential depth adjusted to match the binding
energies for the relevant states.
In order to see whether the assumption of a single state at the centre-of-gravity
energy for a given configuration can produce valid results I have carried out also
calculations using experimentally determined distributions of the single-particle
configurations and their respective spectroscopic factors (Barbopoulos et al.
1979). Computation time increased considerably for these calculations but the
final results were the same as produced by assuming that all the strength is
concentrated in just one state for each of the two single-particle configurations.
In the preliminary calculations, I have used various sets of the optical-model
potential parameters for protons, deuterons and tritons. They included both
shallow and deep potentials with the surface or volume absorption, and they were
taken from a variety of sources. I have found that, allowing for small adjustments of
parameters, various combinations of parameter sets produced similar results.
Furthermore, in the presence of the 01+ ↔ 21+ coupling, calculated results were virtually
insensitive to changes in either proton or triton parameters.
Thus in my final analysis, I have used three selected sets of parameters. For protons, I
used parameters derived by Becchetti and Greenlees (1969). For deuterons, I have
used (as a starting values) the parameters I have just determined in my optical analysis
(see above). For tritons, I used the parameters I have determined in my analysis of the
(t,t) scattering (Nurzynski 1975; see also Chapter 11). However, I have added a spin-orbit
component to the triton potential with parameters I received from Peter Kunz. All
parameter sets are listed in the Table 17.1.
Having carried out a detailed analysis of the relative contributions of the two step
processes (inelastic scattering via the 21+ and 31− states and transfer reactions via the
2p1/2 and 1g9/2 configurations) I have found that only one intermediate channel (the
scattering via the first 21+ states) is needed to explain the observed unusual behaviour
of the amplitudes of the vector analyzing powers, iT11 (θ ) , i.e. their gradual increase
with the increasing mass number A of the target nuclei. This is illustrated in Figure
17.9.
Table 17.1
76,78,80,82
r r
Optical model parameters used in the analysis of the Se( d , d ) elastic scattering at 12 MeV
Target Particle V r0 a0 WD rD aD Vs.o. rs.o. as.o.

Isotope (MeV) (fm) (fm) (MeV) (fm) (fm) (MeV) (fm) (fm)
A
Se p 44 1.25 0.650 13.50 1.25 0.470 7.5 1.25 0.47
A
Se t 150 1.23 0.720 29.50 1.15 0.850 2.5 1.20 0.72
76 a
Se ) d 96 1.14 0.825 17.25 1.35 0.820 4.5 0.76 0.40
78 a
Se ) d 96 1.14 0.825 16.50 1.35 0.820 4.5 0.76 0.40
80 a
Se ) d 96 1.14 0.825 15.50 1.35 0.820 4.5 0.76 0.40
82 a
Se ) d 96 1.14 0.825 14.50 1.35 0.820 4.5 0.76 0.40
A b
Se ) d 96 1.14 0.825 13.75 1.35 0.820 4.5 0.76 0.40
A
Se – means that the same set was used for all selenium isotopes.
a
) Parameters used in the conventional optical model analysis, i.e. without considering the
second-order contributions.
b
) Parameters used in the coupled-channels calculations. They included the two-step the
+
(d,d’)2 (d’d) intermediate scattering and explained the unusual experimental observations.
Figure 17.9. A study of the relative contributions of two-step processes in the elastic scattering of
deuterons. All the calculations used the last set of parameters listed in Table 17.1 for deuterons.
Proton and triton parameters are also listed in Table 17.1. The figure shows that only one intermediate
process, i.e. the (d,d’)2+(d’,d) two-step scattering, has a strong influence on the elastic scattering angular
distributions of the iT11 (θ ) vector analysis power.
The first panel of Figure 17.9 shows two sets of calculations. Group 1 is for the direct
(d,d) elastic scattering as carried out using conventional optical model formalism.
Group 2 is for the direct (d,d) elastic scattering combined with the two-step
(d,d’)2+(d’,d) scattering via the first excited states 21+ in the target nuclei. This panel
shows clearly that the two sets of calculations differ considerably. It also shows that
when the two-step process is added to the direct elastic scattering the amplitudes of
the vector analyzing power iT11 (θ ) increase with the increasing mass number of the
target nuclei, which is in agreement with the observed behaviour.
The second panel shows the direct (d,d) scattering and compares it with the
distributions resulting from adding the two-step contributions via the intermediate
pickup (d,t) reaction involving the 2p3/2 or 1g1/2 configurations. Unlike the results
shown in the first panel, these results show that there is hardly any difference
between calculations for the direct scattering and for direct plus indirect scattering. It
is clear that the two-step process via the neutron pickup reactions can be neglected
in the calculations.
The last panel show similar results as displayed in the second panel but this time for
the contributions from the two-step processes via the intermediate (d,p) stripping
reactions. Again, as for the pickup contributions, the two-step elastic scattering via
neutron stripping channel can be also neglected.
Figures 17.10. The direct and two-step elastic scattering. The calculations for the single step (d,d)
+
elastic scattering (dashed lines) and the single-step plus two-step scattering, (d,d) + (d,d’)2 (d’d), via
+
the first excited states 21 in the target nuclei (continuous lines) are compared with the experimental
data (dots). The figure shows that the unusual enhancement of the amplitudes of the vector analyzing
powers iT11 (θ ) can be explained as being due to the contributions from the two-step elastic scattering
+
via the 21 excited states.
In conclusion, therefore, only one type of the intermediate channel needs to be taken
into account to explain the unusual behaviour of the experimentally observed vector
analyzing powers iT11 (θ ) . This channel is the inelastic scattering via the first 21+ states
in the target nuclei.
Final results of my analysis are shown in Figure 17.10. This figure shows two sets of
theoretical distributions compared with the experimental distributions. One set
(dashed lines) shows the results for the single-step (d,d) direct elastic scattering. The
second set (continuous lines) is for the direct elastic scattering (d,d) plus the two-
step (d,d’)2+(d’,d) scattering via the first 21+ states in the target nuclei. This set of
curves fits perfectly the observed angular distributions and shows that if the two-step
process is included, results for all four isotopes can be reproduced theoretically
using a fixed set of the optical model parameters (see Table 17.1).
Figure 17.11. The effect of including or excluding the two-step scattering on the depth WD of the
optical model potential.
The effect of including or excluding the two-step mechanism in the calculations is

also illustrated in Figure 17.11. Without considering the two-step process explicitly in
the calculations, WD depends linearly on β2. The fit to the data gives the following
dependence:
WD = 10.12 + 23.44 β 2 MeV
By including the two-step process, the dependence on β2 is removed, the description
of the observed distributions is simplified, and only one set of optical potential
parameters is required to fit the experimental results for all four selenium isotopes.
The depth of the imaginary part of the central potential assumes then a fixed value of
13.75 MeV.
The imaginary component of the optical model accounts for contributions, which are
not explicitly considered in the analysis of experimental data. Its constant value
means that only one non-elastic scattering channel was important, the one which
has now been included explicitly in the analysis. All other non-elastic channels have
no effect on the observed distributions.
Measurements of the angular distributions of the differential cross sections and
analyzing powers for the elastic scattering of 12 MeV polarized deuterons on
76,78,80,82
Se isotopes revealed an unusual mass dependence. Contrary to the
normal behaviour, where the amplitudes of the analyzing powers decrease with
the increasing atomic mass number A, we have observed that for selenium
isotopes the amplitudes increase with A. This is particularly clear for the
angular distribution of the vector analyzing power iT11 (θ )
I have carried out a theoretical analysis of the differential cross sections
σ 0 (θ ) and vector analyzing powers iT11 (θ ) assuming contributions from two-step
processes via the intermediate inelastic, and single-neutron transfer channels. I
have also carried out the conventional optical model analysis of the data.
Parameters, which were used to fit the data are summarized in Table 17.1.
The coupled channels calculations included two-step scattering via the 21+ and 31−
states as well as the single-particle excitations of the 1g9/2 and 2p1/2 neutron
configurations in the reactions (d,t)(t,d) and (d,p)(p,d). I have found that the two-
step scattering (d,d')2+(d',d) contributes most strongly to the elastic scattering.
The influence of the 31− state is negligible. The second-order scattering
involving single-particle excitations has a noticeable but also relatively weak
contribution, with the neutron pickup process being slightly stronger than the
neutron stripping. However, in the presence of the two-step (d,d')2+(d',d)
scattering all other second-order processes studied here have a negligible
influence on the calculated distributions.
By including the contributions from the two-step process (d,d')2+(d',d), excellent
fits to the σ 0 (θ ) and iT11 (θ ) angular distributions have been obtained for all four
selenium isotopes (see Figure 17.10) using a single, mass-independent set of
the optical model parameters (the last set in Table 17.1). Results of the analysis
show that the unusual mass-dependence of the iT11 (θ ) amplitudes on the mass
number A of the selenium isotopes can be explained as being due to an interplay
between the direct elastic scattering and the two-step scattering (d,d')2+(d',d) via
the first excited states in the target nuclei.
References
Barbopoulos, L. O., Gebbie, D. W., Nurzynski, J., Borsaru, M. and Hollas, C. L. 1979,
Nucl. Phys. A331:502.
Barrette, J., Barrette, M., Lamoureux, G. and Monaro, S. 1974, Nucl. Phys. A235:154.
Bürgi, H.R., Grüebler, W., Nurzynski, J., König, V., Schmelzhach, P. A., Risler, R.,
Jenny, B. and Hardekopf, R. A. 1980,Nucl. Phys. A334:413.
Nurzynski, J. 1975, Nucl. Phys. A246:333.
Schneider, M. J., Burch, J. D. and Kunz, P. O. 1976, Phys. Lett. 63B:129.
18
Tensor Analyzing Power T20(00) for the 3He(d,p)4He Reaction at
Deuteron Energies of 0.3 – 36 MeV
Key feature:
r
1. The 3He (d , p ) 4He reaction is a useful medium for monitoring tensor polarization of
deuteron beams produced by polarized ion sources.
0
2. The energy dependence of the T20 (0 ) tensor analyzing power for this reaction,
which was measured earlier at lower energies, has now been extended to higher
energies of 11 - 36 MeV. Results can be fitted well using either second- or third-
order polynomials. However, third-order polynomials give a better fit.
3. Previously (Schmelzbach, et al. 1976b) it has been shown that a set of three
second-order polynomials has to be used to reproduce the experimental data for
T20 (00 ) at lower energies. In this new study I have shown that to describe the data
over the whole region of energies (0.3 – 36 MeV) a set of five second-order
polynomials have to be used. However, a better description can be obtained by
using just three third-order polynomials.
r
Abstract: Measurements of the T20 (0 ) tensor analyzing power for the 3He (d , p ) 4He
0
reaction induced by the vector-polarized deuterons have been carried out at the incident
deuteron energies 11-36 MeV. The analyzing power has been found to change smoothly from
-0.92 at 11 MeV to -0.34 at 36 MeV. Excellent description of the experimental data over the
whole range of the incident deuteron energies of 0.3 – 36 MeV has been obtained using a set
of three third-order polynomials.
Introduction
r
The reaction 3He (d , p ) 4He induced by polarized deuterons is a useful medium for
monitoring tensor polarization of deuterons produced by polarized ion sources. A
known relationship between the tensor and vector polarizations of polarized ion
sources can be employed in using this reaction also as a polarimeter in
measurements of vector analyzing powers. This reaction has been used successfully
over years in the Laboratorium für Kerphysik, ETH, Zürich, Switzerland.
At the reaction angle θ = 00, all tensor analyzing powers except T20 (00 ) are equal zero
(Grüebler et al. 1971). At this angle, the general expression for the differential cross
section (see Chapter 15) can be expressed by a simple form:
⎡ 1 ⎤
σ (00 ) = σ 0 (00 ) ⎢1 + t20T20 (00 )(2 cos 2 β − 1)⎥
⎣ 2 ⎦
where t20 is the beam polarization, T20 (00 ) is the analyzing power for the reaction
3
r
He (d , p ) 4He at θ = 00, and β the angle between the quantization axis and the
direction of the incident beam (see Chapter 15).
Assuming that the T20 (00 ) values are known, this reaction can be used as a
convenient way to monitor tensor polarization t20 of the incident deuteron beam.
However, one should notice that 3 cos 2 β − 1 = 0 if cos 2 β = 1 / 3 , which corresponds to
β ≈ 54.740 . Consequently, care should be taken to set up the quantization axis in
such a way as to avid coming close to this angle.
In the polarimeter used at ETHZ, the 3He target was located at the end of the
Faraday cup. The 3He cell had a diameter of 12 mm and was filled with 3He at a
pressure of about 5 atm. The entrance window was made of a 6 μm Harvar foil. The
r
Q-value of the 3He (d , p ) 4He reaction is 18.35 MeV so the protons emerging at 00
have a high energy. The end of cell 3He was in the form of a 0.4 mm stainless steel
wall, which was thick enough to stop the deuteron beam but it absorbed only a small
fraction of the energy of the emitted protons.
A surface-barrier detector was places on the beam axis at a distance of 12 cm from
the 3He target centre to measure the emerging protons. The angular resolution of the
detector was ±2.50. The detector was thick enough to stop protons with the energy of
up to 12 MeV. Depending on the energy of accelerated deuterons, aluminium
absorbers of 0.2 – 1 mm thick were placed in front of the proton detectors to lower
the energy of detected protons. The tensor polarization of the beam t20 was
calculated from the ratio of counting rates for the positive and negative polarization
using the known values of T20 (00 ) for this reaction.
The advantage of using this reaction as a tensor polarization polarimeter are:
• The construction of the polarimeter is simple.
• Only one detector at 00 is needed.
• Because of the high Q-value of the reaction, the protons spectrum is clean
and there is no problem with the background.
• The absolute values of the analyzing power T20 (00 ) are high at low energies
and they vary smoothly with the incident deuteron energy.
Measurements of the T20 (00 ) analyzing powers have been carried out earlier at low
deuteron energies by Trainor, Clegg and Lisowski (1974) and Schmetzbach et al.
(1976a). The analyzing power has been calibrated using the reactions
16
O(d,α1)14N* and 4He(d,d)4He.
Extension to higher energies

Due to the successful use of this reaction at tandem energies it was interesting to
extend the measurements of T20 (00 ) to higher energies available from the SIN17
injector cyclotron. Some changes in the experimental setup have been necessary
at these higher energies. A thicker entrance foil was used, which allowed for the
pressure in the gas cell to be increased to 15 atm, which
r helped to compensate for
3 4
lower differential cross sections for the reaction He (d , p ) He at the higher energies
available from the cyclotron. In order to stop the primary beam, different thicknesses
of absorbers were used between the gas cell and proton detector, depending on the
energy of accelerated deuterons. Two 5 mm thick detectors were used in
coincidence to get clean spectra for the high-energy protons.
17
Schweizerische Institut für Nuklearforschung
The tensor polarization of the SIN deuteron beam was determined to be pzz = 0.80 by
comparison with the Los Alamos measurements of d-α scattering at 17 MeV (Ohlsen
et al. 1973). The results of measurements of the T20 (00 ) tensor analyzing powers at
these higher energies are shown in Figure 18.1. They show smooth energy
dependence with the absolute value of T20 (00 ) decreasing with the increasing
deuteron energy indicating that this reaction might have limited application as a
monitor of the deuteron polarization pzz at higher energies.
0
Figure 18.1. Energy dependence of the T20 (0 ) tensor analyzing power for the reaction
3
r
He (d , p ) 4He. The points represent our new measurements, which extend the earlier measurements
to higher energies of 11-36 MeV. The combined statistical and systematic errors are smaller than the
size of the points. The thick line is drawn as a guide for the eye. The thin line is drawn through the
earlier data at lower incident deuteron energies (Schmeltzbach et al. 1976a; Trainor, Clegg, and
Lisowski 1974).
0
Figure 18.2. Results of our measurements of the T20 (0 ) tensor analyzing power for the reaction
3
r
He (d , p ) 4He in the energy range of 11-36 MeV are compared with the calculations using the
second-order (dotted line) and third-order (continuous line) polynomials as listed in Table 18.1. The
third order polynomial reproduces the data remarkably well.
For practical reasons, it is useful to parameterise the experimental data using
simple mathematical formulae. Results at lower energies have been fitted using a
set of three second-order polynomials (Schmltzbach et al. 1976b). It was therefore
interesting to see whether the new set of data obtained now at higher deuteron
energies could be also fitted using the same procedure. Figure 18.2 shows the fits
obtained using the second- or third-order polynomial. Clearly, the third-order
polynomial gives a better fit. The polynomials are listed in Table 18.1. The fit to the
data is strongly sensitive to the highest order component in each polynomial.
Table 18.1
0
The third- and second-order polynomials used to fit the T20 (0 ) data at the incident deuteron energies
Ed = 11-36 MeV (see Figure 18.1)
Polynomials R2
T20 (00 , Ed ) = −1.6477340 + 0.0822609 Ed − 0.0012911Ed2 0.9882675
T20 (00 , Ed ) = −2.2320672 + 0.1700278 E − 0.0052947 E 2 + 0.0000564 E 3 0.9998055
The T20 (00 ) analyzing power for the energy range of 0.3-36 MeV
Having calculated the polynomials in this higher energy range I decided to join the
new data with the previous measurements at lower energy and to see whether the
formulae derived earlier using the second-order polynomials (Schmeltzbach et al.
1976b) could be still used. I had found that in order to represent both the old and
new data using a smooth transitions between various sets of polynomials the data
had to be divided into different sections then used before and a new analysis had
be carried out.
Table 18.2
A set of the second-order polynomials describing the energy dependence of the tensor analyzing
r
power T20 (0 ) for the reaction 3He (d , p ) 4He in the energy range of 0.3-36.2 MeV
0
Energy Range (MeV) Polynomials

0.3 ≤ Ed ≤ 2.5 T20 (00 , Ed ) = −0.7685 + 0.4785Ed − 0.1900 Ed2
2.0 ≤ Ed ≤ 6.0 T20 (00 , Ed ) = +0.3582 − 0.5800 Ed + 0.0517 Ed2
6.0 ≤ Ed ≤ 11.0 T20 (00 , Ed ) = −1.4908 + 0.0247 Ed + 0.0023Ed2
11.0 ≤ Ed ≤ 17.0 T20 (00 , Ed ) = −2.0517 + 0.1309 Ed − 0.002653Ed2
17.0 ≤ Ed ≤ 36.0 T20 (00 , Ed ) = −1.3068 + 0.0562 Ed − 0.000821Ed2
In my analysis I have also included the data at very low energies, i.e. below 2.2
MeV, which were not described mathematically by Schmeltzbach et al. (1976b). I
have found a new set of the second-order polynomial formulae (see Table 18.2)
that reproduce the observed energy dependence of T20 (00 ) over the whole range of
energies of 0.3-36 MeV (see Figure 18.3).
0
Figure 18.3. Results of measurements of the T20 (0 ) tensor analyzing power for the reaction
3
r
He (d , p ) 4He in the energy range of 0.3-36 MeV are compared with a new set of the second-order
polynomials. The continuous line is made of five polynomials listed in Table 18.2.
It is interesting to compare these new calculations with the calculations of

Schmltzbach et al. (1976b) at low energies. This is done in Figure 18.4. Even
though the two sets of calculations describe the experimental data sufficiently well,
the new calculations appear to reproduce the experimental results a little better.
Figure 18.4. The calculations of Schmeltzbach et al. (1976b) are compared with the new calculations
using the formulae listed in Table 18.2.
Results presented in Figure 18.2 suggest that third-order polynomials might give
better fits to the data. Calculations over the whole range of the incident deuteron
energies are shown in Figure 18.5. The experimental data can now be fitted using
just three functions (as compared to five if the second-order polynomials are used).
Furthermore, the resulting theoretical line is now smoother than the line
corresponding to a set of the second-order polynomials (cf Figures 18.3 and 18.5).
The set of the three third-order polynomials describing the experimental data over
the whole range of energies, 0.3 – 36 MeV, is given in Table 18.3.
It should be noted that the parameters for the last two equations in Table 18.2 and
for the last equation in Table 18.3 are different than the relevant parameters in Table
18.1. This is because the energy range for the formulae in Tables 18.2 and 18.3 are
different than the range for the formulae in Table 18.1.
Table 18.3
A set of the third-order polynomials describing the energy dependence of the tensor analyzing power
r
T20 (00 ) for the reaction 3He (d , p ) 4He in the energy range of 0.3-36 MeV
Energy Range Polynomials

(MeV)
0.3 ≤ Ed ≤ 3.5 T20 (00 , Ed ) = −0.8056 + 0.6166 Ed − 0.3221Ed2 + 0.0344 Ed3
3.0 ≤ Ed ≤ 10.0 T20 (00 , Ed ) = +0.5939 − 0.7875Ed + 0.1057 Ed2 − 0.0043Ed3
9.5 ≤ Ed ≤ 36.0 T20 (00 , Ed ) = −2.2706 + 0.1755Ed − 0.0055Ed2 + 0.0000586 Ed3
0
Figure 18.5. Results of the measurements of the T20 (0 ) tensor analyzing power for the reaction
3
r
He (d , p ) 4He in the energy range of 0.3-36 MeV are compared with the third-order polynomial
calculations. The continuous line is made of only three polynomials listed in Table 18.3.
Earlier low-energy measurements of the T20 (00 ) tensor analyzing power for the
r
reaction 3He (d , p ) 4He have been extended to higher incident deuteron energies of
up to 36 MeV. The energy dependence of the T20 (00 ) analyzing power has been
analysed using the second- and third-order polynomials. Experimental results over
the whole range of energies of 0.3 – 36 MeV can be reproduced remarkably well by
a set of only three third-order polynomials. The information suppliedr by these new
set of data and the associated analysis can assist in using the 3He (d , p ) 4He as the
polarization analyser.
References
Grüebler, W., König, V., Ruth, A., Schmelzbach, P. A., White, R. E. and Marmier, P.
1971, Nucl. Phys. A176:631.
Schmelzbach, P. A., Grüebler, W., König, V., Risler, R., Boerma, D. O. and Jenny,
B.1976a, Nucl. Phys. A264:45.
Schmelzbach, P. A., Grüebler, W., König, V., Risler, R., Boerma, D. O. and Jenny, B.
1976b, Proceedings of the Fourth International Symposium on Polarization
Phenomena in Nuclear reactions, 1975, eds W. Grüebler and V. Köning,
Birkhauser Verlag, Basel, p. 899 and references therein.
Ohlsen, G. G., Lovoi, P. A., Salzmann, G. C., Meyer-Berkhout, U., Mitchell, C. K., and
Grüebler, W. 1973, Phys. Rev. C8:1262.
Trainor, T. A., Clegg, T. B. and Lisowski, P. W. 1974, Nucl. Phys. A220:533.
19
The Maximum Tensor Analyzing Power Ayy = 1 for the 3He(d,p)4He
Reaction
Key feature:
1. Points at which the analyzing powers reach their extreme values ( Ay = 1 and
Ayy = 1 ) are of significant interest both experimentally and theoretically.
Experimentally, they can serve as the reference points for the calibration of
polarimeters. Theoretically, they can be used to test special conditions imposed by
such extreme values.
2. We have found that a Ayy = 1 value for the 3He(d,p)4He reaction is located at Ed =
9.28 MeV (lab) and θ = 23.60 (lab).
3. We have found no Ay = ±1 point in the investigated range of energies and angles.

However, our measurements have resulted in high-precision data, which can be used
as reliable reference points in the calibration of polarimeters or in theoretical
analyses.
4. We suggest that the extreme value of Ayy = 1 is associated with a resonance in 5Li.
Abstract: Very precise measurements of the vector and tensor analyzing powers, Ay ( Ed ,θ )
r
and Ayy ( Ed ,θ ) for the 3He (d , p ) 4He reaction have been carried out at the incident deuteron
energies Ed = 8.5-10.5 MeV and at angles θ = 120 – 320 (lab). The aim of this study was to
search for the extreme values Ay = 1 and Ayy = 1 . A maximum Ayy = 1 has been located at
( Ed ,θ ) = (9.28 MeV, 23.60) in the laboratory system of reference. No extreme value of
Ay = ±1 was detected for this reaction.
Introduction
Points where the analyzing powers reach their extreme values are of significant interest
both experimentally and theoretically. For experimentalists, these points can be used in the
absolute calibration of the experimental equipment used to measure beam polarization.
For theorists, they allow to study special conditions for the M-matrix amplitudes (see the
Appendix H).
The location of the extreme values of the analyzing powers have been investigated
earlier for the elastic scattering of spin-1/2 and spin-1 projectiles from spin-0 target
nuclei. For spin-1/2 projectiles, Plattner and Bacher (1971) showed analytically18 that
Ay = ±1 values must occur at three sets of energy-angle coordinates in the nucleon-
α elastic scattering. For spin-1 particles, conditions for the maximum values of the
analyzing power were investigated by Grüebler et al. (1975) for the d-α scattering.
These authors have shown analytically that three Ayy = 1 points should exist in the
18
For an outline of the analytical determination of the extreme values of the analyzing powers see the
Appendix I.
deuteron energy range between 3 and 12 MeV. They have also located them
experimentally.
In all these cases, there is a simple linear relation between just two elements of the
transition matrix M, which connects the spin states of the incoming and outgoing
channels.
However, the question is whether such points can also occur in nuclear transfer
reactions. Seiler (1976) investigated the conditions imposed on the M-matrix elements for
the spin configuration 1 + 1 2 → 1 / 2 + 0 .19 He found that the Ayy = 1 occurs if and only if
two sums of the spin-non-flip amplitudes and the spin-flip amplitudes of M-matix
elements are equal zero.
M 1,1 / 2;1 / 2 + M −1,1 / 2;1 / 2 = 0 ; M 1, −1 / 2;1 / 2 + M −1, −1 / 2;1 / 2 = 0 (1)
In these equations, the indices denote the spin projections in the incoming and outgoing
channels. This situation is different than the condition for the elastic scattering because two
linear relations involving four M-matrix elements have to be satisfied at the same angle and
energy. The existence of such a point cannot be proven analytically in the same way as
the maxima for the elastic scattering.
Seiler (1976) suggested that the extreme values of Ayy = 1 should occur for the
3
r
He (d , p ) 4He reaction at the ( Ed ,θ ) coordinates of around (9.0 MeV, 270), and for
r
the 6Li (d , p ) 4He at around (5.5 MeV, 300) and (9.0 MeV, 900). To locate such
maxima experimentally, precise mapping of the analyzing powers in the ( Ed ,θ )
coordinates is required.
r
In our study we have used the 3He (d , p ) 4He reaction, which was investigated earlier
by Grüebler et al. (1971). In our measurements we have included the energy-angle
mapping not only for the Ayy component but also for Ay because the previous
measurements suggested that a Ay = −1 point might be present in the same region.
The experimental method employed in this study was as described in Chapters 15,
17 and 18. The spin direction was perpendicular to the scattering plane. The exact
spin position was adjusted by the Wienfilter to a precision better than 1° beforerstarting
the measurements. A polarimeter based on detecting protons from the 3He (d , p ) 4He
reaction at 0° was used to measure the beam polarization continuously (see Chapter
18). This reaction has been calibrated in the absolute sense with the precision better than
1%. A polarized deuteron beam with py = 0.3 and pyy = 0.9 was delivered by the ETHZ
atomic beam polarized source and EN tandem accelerator. The defining diaphragms in
front of the detectors were 4 mm wide and 30 mm high and they were located at a
distance of 256 mm from the middle of the gas target. The procedure of determining the
analyzing powers is described extensively in a Chapter 15. This method, which uses
detectors on the left- and right-hand side of the target and the frequent reversal of the sign
of the beam polarization, cancels instrumental asymmetries. The measurements are
not affected by small deviations of the spin direction from the required position.
19
The polarization formalism for the 1 + 1 2 → 1 / 2 + 0 structure is outlined in the Appendix F.
Experimental results
Experimental data
The analyzing powers Ay and Ayy have been measured between 8.5 and 10.5 MeV
deuteron energy in the angular range θlab = 12° - 32°. Results are shown in Figure
19.1. Extra care has been taken to carry out precise measurements, which is essential
in the determination of the extreme values of the analyzing powers. The statistical
errors are smaller than the size of the displayed data points. The curves represent the
fits obtained by using polynomial functions. As discussed below, the polynomial
interpolation of the experimental data was necessary to construct a two-dimensional
map of the experimental data and to locate the Ayy maxima. Since the measurements
show clearly that for these energies there is no maximum for Ay this component was not
fitted with polynomial functions. However, our results represent very accurate data and
consequently they can be used as reference points in calibrations of the vector
polarization of polarized deuteron beams.
Fig. 19.1. Results of measurements of the analyzing powers Ay and Ayy . The statistical errors are
smaller than the displayed data points. The curves are the polynomial fits.
Interpolation of the experimental data
The mapping of the region of measurements was carried out by interpolation
calculations. In this procedure, each angular distribution was fitted by a polynomial the
order of which was determined by the best fit to the data. We have found that in all
cases the best fit was obtained by the third-order polynomials. The polynomial fits,
together with the experimental points, were then used to construct a two-dimensional
map for the Ayy analyzing power. The map is shown in Figure 19.2.
Figure 19.2. Contour plot of the tensor analyzing power Ayy constructed using our experimental
results and the polynomial interpolation method as described in the text.
Corrections and uncertainties

The uncorrected interpolation gives a maximum value of Ayy = 0.981 ± 0.003 . This
value has to be corrected for the finite solid angle used in the experiment. The
quoted error does not include the uncertainties in the determination of the beam polar-
ization. The necessary corrections and uncertainties, which have to be included in the
final calculations of the Ayy maximum are shown in Table 19.1.
Table 19.1
Corrections and uncertainties for the measurements of the Ayy analyzing power
Final results
Taking into account all possible errors of measurements, the final results are given in
Table 19.2. They show the position of the Ayy = 1 point.
Table 19.2
Experimentally determined maximum value of Ayy
Discussion
Our study shows that within rexperimental errors a maximum tensor analyzing power
Ayy = 1 for the reaction 3He (d , p ) 4He has been located. This result can be tested by
measurements of other observables, which must satisfy the expected theoretical
conditions (Seiler 1976) imposed by the relations between the matrix elements, as
given by equation (1) in the Introduction. Besides the vanishing of several polarization
transfer coefficients and polarization correlation coefficients the proton polarization Py for
unpolarized beam and target must be equal to the negative value of the analyzing
power A0, y of a polarized 3He target. Although the relevant experimental data at the
required angle and energy are not available, results in close proximity to the required
condition appear to agree with the theoretical predictions (Brown and Haeberli 1963;
Grüebler, König, and Schmelzbach, 1973; Hardekopf et al. 1973).
An analysis of the polarization transfer coefficients K xx ′ and K zx′ measured by Hardekopf
et al. (1973) at 8 MeV shows vanishing values of these quantities near the critical
angle, as required by the eqs (1). All these data appear to support the result found in
our study.
A further, experimental test of our result would be to investigate the inverse reaction
4
He(p,d)3He, for which the emitted deuterons should have a maximum tensor
polarization pyy = 1 for the corresponding incident proton energy.
Having found the extreme value of Ayy = 1, it is interesting to consider the physical
interpretation of the equations (1). The probability that the sums of two pairs of
complex amplitudes vanish simultaneously accidentally is very low. A possible reason for
the situation described by equations (1) may lie in the particular symmetric structure of
these relations. A study of the d-α scattering (Grüebler et al. 1975) suggests that this
feature might be explained by a resonance in the vicinity of this energy.
The analysis of the 3He(d,p)4He reaction in the energy range of between 2.8 to 11.5
MeV using Legendre polynomials seems to suggest a resonance behaviour (Grüebler at
al. 1971). In this analysis, the tensor analyzing powers T20 , T21 , and T22 show a
resonance-like behaviour around 9 MeV deuteron energy corresponding to an orbital
angular momentum l = 2. Examination of the energy levels of 5Li (Figure 19.3) suggests
also that a resonance around the required excitation energy is possible.
The energy Ed = 9.28 MeV corresponds to the excitation energy of 22.2 MeV. There
is an energy level at 22.06 MeV in 5Li with spin 3/2-. The spins in the entrance and
exit channels are Si = S d + S 3 He = 1 / 2 or 3 / 2 , and S f = S p + S 4 He = 1 / 2 . If we use the
2 S +1
notation LJ for the initial and final channels, then we can have three possible
resonance transitions to this level (see Table 19.3). Thus the expected reaction
mechanism could be associated with the P and/or F-wave capture in the entrance
channel.
Figure 19.3. The energy level diagram for 5Li
Table 19.3
The possible resonant transitions at Ed = 9.28 MeV
Transition Resonant matrix

element (Welton 1963)
(Lf Sf Jπ|R|Li Si Jπ)
2
P3 / 2 →2P3 / 2 (1 1/2 3/2-|R|1 1/2 3/2-)
4
P3 / 2 →2P3 / 2 (1 1/2 3/2-|R|1 3/2 3/2-)
4
F3 / 2 →2P3 / 2 (1 1/2 3/2-|R|3 3/2 3/2-)
2 S i +1 2 S f +1
Transition: ( Li ) J → (Lf )J
References
Brown, R. I. and Haeberli, W. 1963, Phys. Rev. 130:1163.
Goldfarb, L. J. B. 1958, Nucl. Phys. 7:622.
Grüebler, W., König, V., Ruh, A., Schmelzbach, P. A., White, R. W. and Marmier, P.
1971,Nucl. Phys. A176:631.
Grüebler, W., König, V. and Schmelzbach, P. A. 1973, ‘Results of measurements and
analyses of nuclear reactions induced by polarized and unpolarized deuterons’,
Internal Report, ETH Zurich, May 1973.
Grüebler, W., Schmelzbach, P. A., König, V., Risler, R. Jenny, B. and Boerma, D. O.
1975, Nucl. Phys. A242:285.
Hardekopf, R. A., Armstrong, D. D., Grüebler, W., Keaton, P. W. and Meyer-Berkhout,
U. 1973, Phys. Rev. C8:1629.
Plattner, G. R. and Bacher, A. D. 1971, Phys. Lett. 36B:211.
Seiler, F. 1976, Phys. Lett. 61B:144.
Welton, T. A. 1963, in Fast Neutron Physics, Vol. II, eds J. B. Marion
and J. L. Fowler, Interscience, New York, p. 1317.
20
Search for the Ay = 1 and Ayy = 1 Points in the 6Li(d,α)4He Reaction
Key features:
r
1. We have located two Ayy = 1 points for the reaction 6Li (d , α ) 4He, one at Ed = 5.55
MeV (lab) and θ = 24.20 (lab) and one at Ed = 8.80 MeV (lab) and θ = 76.80 (lab)
2. The Ay = 1 points, if present, should occur at the same energies and angles as
r
Ayy = 1 . No such points were observed for the 6Li (d ,α ) 4He reaction indicating that
two additional conditions for the M-matrix elements as required for Ay = 1 are not
satisfied.
3. The suggested reaction mechanism responsible for the extreme values of the tensor
analyzing power, Ayy = 1 , is a resonance in 8Be.
Abstract: A search for the analyzing powers Ay = 1 and Ayy = 1 was carried out
r
experimentally for the Li ( d , α ) He reaction in the energy range of 5.0-6.5 MeV and 8.0-10.0
6 4
MeV. Two Ayy = 1 maxima had been found but no Ay = 1 points, which should occur at the
same energies and angles, were detected. The precise values for the energies and angles of
the Ayy = 1 maxima were determined by the polynomial interpolation of the experimental data.
Introduction
Having successfully located the extreme value Ayy = 1 of the tensor analyzing power
r
for the reaction 3He (d , p ) 4He, we have decided to extend our search to the
6
r
Li (d ,α ) 4He reaction. In our previous study we have established that the extreme
values exist not only in the elastic scattering but also in nuclear reactions induced by
polarized deuterons. It seemed therefore interesting to search for such extreme values
in another transfer reaction. The importance of such points has been discussed in
Chapter 19.
r
Inspection of the analyzing powers for the 6Li (d ,α ) 4He reaction below 12 MeV of the
incident deuteron energy (Seiler et al. 1976; Seiler, Rad, and Conzett 1976) suggested a
possibility of the existence of the Ayy = 1 maxima near Ed = 6 MeV (θc.m. ≈ 350) and Ed
= 9 MeV (θc.m. ≈ 750). A compilation of previous measurements is presented in Figure
20.1. The experimental data are not sufficiently precise but they seem to suggest
that the analyzing power Ayy reaches its extreme values in two places. In our study
we have included a search for Ay = 1 points, which if present should occur at the
same energies and angles as the Ayy = 1 maxima.
Figure 20.1. A compilation (Seiler at al. 1976) of the data for the Ayy tensor analyzing power. The
curves drown through the experimental points are labelled according to the incident deuteron energy.
This figure shows that the tensor analyzing power Ayy is likely to reach its extreme values of 1 at
around 6 and 9 MeV and at the angles of around 300 and 750, respectively. Accurate measurements
were required to confirm or refute this expectation.
The general experimental arrangement and procedure are described in the previous
Chapter. A self-supporting 6Li foil, enriched to 96%, with a thickness of about 300
μg/cm2 was used as the target.
r
Due to the high Q value for the 6Li (d ,α ) 4He reaction (Q = 22.374 MeV) the emitted α
particles have high energy. Thus, by proper adjusting of the sensitive depth of the
surface barrier detectors the background under the α peaks could be reduced to only
a few percent. However, the cross section for this reaction is small (only about 0.3
mb/sr). This caused a significant problem at forward angles where the elastic scattering
cross section is about three orders of magnitudes larger than the transfer reaction
cross section. For this reason electronic pile-up effects could not be prevented in all
cases and had to be considered carefully in the analysis of the data.
The components Ay and Ayy were measured between 5.00 MeV and 6.50 MeV in the
angular range of 20° to 50° in the c.m. system. Data were obtained in the energy steps
of between 100 and 500 keV.
Measurements were also carried out for the incident deuteron energies of between
8.00 and 10.00 MeV in steps of 0.5 MeV. The c.m. angular range in this case was 60° to
105°.
Results of measurements
Our experimental results for the lower energy range are presented in Figure 20.2.
The statistical errors are smaller than the displayed data points. The displayed data
have not yet been corrected for the finite geometry and electronics effects. The solid
lines are polynomial curves fitted to the data in order to obtain the local maxima. The
Ay = 1 point should occur at the same angle and energy as Ayy = 1 (Seiler et al. 1976).
r
It is clear that the Ay component does not reach an extreme value of 1 for the 6Li (d ,α ) 4He
reaction. However, the uncorrected distributions for the Ayy analyzing power come
close to the extreme value.
Figure 20.2. Tensor and vector analyzing powers measured in the energy range of 5.5-6.5 MeV. The
distributions for the tensor analyzing power, Ayy , are shown on the left-hand side and for Ay on the
right-hand side of the figure. The statistical errors are smaller than the size of the points. The curves
for the Ayy component are the polynomial fits.
Figure 20.3. Vector and tensor analyzing powers measured in the energy range of 8.0-10.0 MeV. The
distributions of the tensor analyzing power, Ayy , are shown on the left-hand side and for Ay on the
right-hand side of the figure. The statistical errors are smaller than the size of the data points. The
curves for the Ayy component are the polynomial fits.
Results obtained in the higher energy range are shown in Figure 20.3. The Ay
changes sign at 90°as expected for a reaction with two identical particles in the exit
channel. The curves are again polynomial fits to the data.
Our data in this higher energy range demonstrate that while the vector component does
not reach its extreme values Ay = 1 in the expected region of energies and angles, the
tensor component Ayy approaches the values of 1 at two energies.
Finding the maxima of the Ayy component.

The experimental mapping of the two energy regions was carried out using the
interpolation procedure described in Chapter 19. At first the raw data were fitted with
polynomial curves and the results are displayed in Figures 20.2 and 20.3. Next
various necessary corrections were included.
The local maximum located between 5.0 and 6.5 MeV and at a small angle required a solid-
angle correction of +1.2%. An additional correction was necessary because of the pile-
up effects caused by the much higher counting rate of the elastically scattered
deuterons. An investigation based on simulating the electronic pile-up effects by using
two random pulse generators with pulse heights and pulse rates corresponding to the
scattered deuterons and the alphas from the reaction, indicated a 3.5% counting loss
in the simulated α-peak in the spectrum and created an additional background on
both sides of the peak. In spite of this large loss in the number of counts the
calculated correction to the analyzing powers was only 0.7%.
At higher energies where a broad maximum is around 90° the calculated correction
from the solid angle geometry is only +0.5%. Because of the much smaller elastic
scattering cross section no pile-up loss correction was necessary.
The final absolute calibration at the maxima found at lower and higher energies was
made with a beam calibrated using the analytically proved Ayy = 1 points in d-α
scattering (Grüebler at al. 1975). This final calibration included also the
measurement of the quantities Axx and Azz . The relation Axx + Ayy + Azz = 0 can then be
used as a consistency check of the data. Using our results we get for the sums (0.0030
± 0.0150) and (-0.0018 ± 0.0094) at the lower and higher energies, respectively.
The fits around the interpolated maxima are shown in Figure 20.4. The dots with
error bars are the corrected and absolutely calibrated results of our measurements.
The open circles represent the determined maximum values. The thick solid lines
represent the values extracted from the polynomial mapping of the experimental
results. The shades band represents the statistical errors of the measurements. An
additional uncertainty of 0.007 caused mainly by the background subtraction and the
uncertainty in the correction for pile-up losses must be applied to the results at 5.55 MeV.
This is shown by the dashed curves.
Finally, the uncertainty of the beam polarization had to be also included. The energy of the
accelerator had to be changed for the d-α calibration points and hence a slight
change in the beam optics and a corresponding deviation of the beam polarization
had to be taken into account. The total uncertainty of the calibration procedure was
estimated at no more than 1 %. This uncertainty is shown as dashed horizontal lines in
Figure 20.4. These lines show that due the uncertainly in the absolute value of the beam
polarization the displayed value of 1 on the vertical scale of each figure can be located
anywhere between the two points indicated by the horizontal dashed line and the top frame of
the figure. In numerical terms, the vertical scale in each figure can be shifted down by a
maximum of 1%. The figure shows that within experimental errors we have located two
Ayy = 1 points for the reaction 6Li(d,α)4He in the investigated ranges of energy and
angles.
Figure 20.4. Experimental data and the interpolation curves based on the polynomial mapping. The
shaded areas show statistical uncertainties. The dashed lines show the boundaries of uncertainties
that include background substruction and pilup-up errors. They apply only to the lower energy data.
The dashed horizontal lines show the maximum value of the beam polarization uncertainty caused by
r
changing the energy between the 6Li (d , α ) 4He measurements and d-α calibration. The interpretation
of these lines is that the vertical scale can be shifted by the distance between the horizontal lines and
the top of the frames.
Figure 20.5. The three-dimensional representation of the experimentally determined tensor analyzing
power Ayy . The maximum Ayy = 1 is indicated by an open circle.
As a sample of the polynomial mapping of the experimental data a three dimensional
representation of the experimentally determined Ayy component is show in Figure
20.5. The maximum value of Ayy = 1 is shown as an open circle. The experimentally
determined Ayy = 1 points are shown in Table 20.1.
Table 20.1
r
Experimentally determined location of the Ayy = 1 points for the reaction 6Li ( d , α ) 4He
Summary and discussion

r
The ground state of 6Li has spin 1, so the spin structure for the 6Li (d ,α ) 4He reaction
r
is 1 + 1 → 0 + 0 . The necessary and sufficient condition for the existence of the Ayy = 1
maximum for this reaction is that
M 11;00 + M 1−1;00 = 0
where M JM ; J ′M ′ are the M-matrix elements. The physical meaning of this equation is
that the absolute values of the amplitudes with parallel and antiparallel spins in the
entrance channel should be equal but they should have opposite signs.
The M-matrix components depend on the incident energy Ed and the reaction angle θ.
The same, but only necessary, condition applies also to the Ay = 1 maxima.
Consequently, if Ay = 1 does exist it should occur at the same energy and angle as
r
Ayy = 1 . In our experimental survey of the 6Li (d ,α ) 4He reaction we have located two
Ayy = 1 maxima but no Ay = 1 maxima. Clearly, two other conditions (Seiler et al.
1976), which are also necessary for the presence of Ay = 1 points, are not satisfied.
Our results show that while the shape of the Ayy ( Ed ,θ ) function is smooth and simple
in the vicinity of 5.55 MeV, the behaviour of the same component around the 8.80
MeV is more complex. Near this energy, the angular distributions change from
displaying only one maximum at 90° to two maxima located symmetrically on each
side of 90° (see Figure 20.5). As the Ayy = 1 surface is nearly flat over a relatively
large energy and angular range the location of this maximum is not well determined.
One may speculate that the absolute maximum occurs at the forking point of the local
maxima.
In our previous study we have argued (see Chapter 19) that extreme maxima are
probably associated with resonance interaction. Possible candidates for the resonance
excitation in 8Be, which might be responsible for the observed maxima of the tensor
analyzing powers are given in Table 20.2.
Table 20.2
r
Possible resonance excitations of the compound nucleus 8Be by the 6Li (d , α ) 4He reaction, which
might be responsible for the observed Ayy = 1 maxima
Ed Ex Resonant matrix Transition

element (Welton 1963)
(Lf Sf Jπ|R|Li Si Jπ)
5.55 26.4 (4 0 4+|R|2 2 4+) 5
D 4 →1G 4
8.80 28.9 (6 0 6+|R|4 2 6+) 5
G 6 →1I 6
Ed – incident deuteron energy.
Ex – excitation energy of 8Be.
2 S i +1 2 S f +1
Transition – ( Li ) J → (Lf )J .
Sî = Sˆd + Sˆt (channel spin) is a sum of the deuteron
and 6Li spins.
Li – orbital angular momentum in the incident
channel.
Jπ – spin-parity of the resonant state in the
compound nucleus 8Be ( Jˆ = Sî + Lî = Sˆ f + Lˆ f ).
Sf – spin of α particles.
Lf – orbital angular momentum in the exit channel.
The notation used in this Table is the same as used in Chapter 19. The resonant
excitation of 8Be at these energies was suggested by the analysis of α-α scattering
(Bacher at al. 1972).
References
Bacher, A. D., Resmini, F. G., Conzett, H. E., de Swiniarski, R., Meiner, H. and Ernst,
J. 1972, Phys. Rev. Lett. 29:1331.
Grüebler, W., Schmelzbach, P. A., König, V., Risler, R., Jenny, B. and Boerma, D.
1975, Nucl. Phys. A242:285.
Seiler, F., Rad, F. N., Conzett, H. E. and Roy, R., 1976, Proc. Fourth Int. Symp. on
Polarization Phenomena in Nuclear Reactions, Zürich, eds W. Grüebler and V.
Konig, Birkhauser Verlag, Basel, p.587
Seiler, F., Rad, F. N., and Conzett, H. E., 1976, Proc. Fourth Int. Symp. on Polarization
Phenomena in Nuclear Reactions, Zürich, eds W. Gruebler and V. Konig,
Birkhauser Verlag, Basel, p.897
Welton, T. A. 1963, in Fast Neutron Physics, Vol. II, eds J. B. Marion
and J. L. Fowler, Interscience, New York, p. 1317.
21
A Study of a Highly Excited Six-nucleon System with Polarized
Deuterons
Key feature:
1. This work represents the first precise measurements of the angular distributions and
excitation functions of the vector Ay and tensor Ayy and Axx analyzing powers in the
energy range of 17-43 MeV. A total of 21 angular distributions have been measured
and 14 excitation functions.
2. The data revealed large values of the analyzing powers at backward angles.
3. The extreme value of Ay = 1 at deuteron energy of around 28 MeV and at the

reaction angle of around 1350 (lab) reported earlier by Conzett at al. (1976b) has not
been confirmed. Even though the values for this component are large in this region
they are well below the extreme value of 1.
4. Using our data and the earlier data at lower energy, we have constructed a contour
map of the Ayy analyzing power. The map suggests the existence of five Ayy = 1 points
for energies of up to 50 MeV, one of which is the point at 35 MeV and 1500 (c.m.)
revealed by our measurements. The remaining four points are below 15 MeV.
5. Angular distributions were analysed using the resonating-group theory and the three-
body bake-up Faddeev formalism. The Faddeev formalism gives significantly better
description of our data.
6. We have also carried out the phase-shift analysis of our data including spins J = 1,2, 3
and 4 and l = 0, 1, 2, 3 and 4. No clear resonance behaviour has been observed.
Smaller energy steps would be required to study possible resonance excitations in this
region of energies.
Abstract: Angular distributions of the vector analyzing power Ay and of the tensor
components Ayy and Axx were measured in the energy range of between 17.0 and 42.8 MeV in
steps of 4 to 5 MeV. A possible Ayy = 1 point was found near 35 MeV and 150° (in the centre-
of-mass system). Our results are compared with predictions of the resonating-group formalism
and the three-body Faddeev calculations.
Introduction
Information on the structure of nuclei composed of a few nucleons and on the
reaction mechanism of deuteron induced reactions with such light nuclei have been
greatly enriched by measurements of polarization phenomena and by comparing
experimental results with model calculations. Such microscopic calculations have been
particularly successful for the six-nucleon system using a three-body model (Gammel,
Hill and Thaler, 1960; Shanley, 1969; Chun, Han and Lin 1973); Charnomordic,
Fayard and Lamont 1977; Elbaz, Fayard and Lamont 1978), refined cluster model
(Hackenbroich, Heiss, and Le-Chi-Niem 1974; Hackenbroich 1975) or the resonating
group formalism (Thompson and Tang 1973; Jacobs, Wildermuth and Wurster 1969;
Lemere, Tang, and Thompson 1976). Most of these calculations are, however,
restricted to excitation energies in 6Li, which are below the threshold of the 3He-t break-
up.
An extraordinary agreement with experimental polarization data for the d-α elastic
scattering was obtained by Hackenbroich, Heiss, and Le-Chi-Niem (1974) who
used the refined cluster model This calculation considered not only central and spin-
orbit terms for the nucleon-nucleon interaction but also tensor interactions, and
included coupling between the elastic channel and the 5He-p, 5Li-n and 3He-t
channels.
In general, in the resonating-group formalism, these reaction channels are taken into
account to a first approximation by phenomenological imaginary potentials. In most of
these calculations, the spin-orbit and tensor terms in the nucleon-nucleon potential are
neglected and thus no polarization observables are predicted.
A substantial improvement in theoretical study of the d-α scattering was made by Lemere,
Tang, and Thompson (1976) who did include spin-orbit interaction. Their calculations
yielded sets of phase-shifts for incident deuteron energies of up to 80 MeV. A study
published by Charnomordic, Fayard and Lamont (1977) solved the three-body Faddeev
equations for two nucleons and a structureless α-particle. Their calculations of the d-α
elastic scattering predict angular distributions of the differential cross sections, vector
analyzing power and the three tensor analyzing powers for deuterons with energies of up
to 43 MeV. Unfortunately the validity of these predictions at higher energies could not
be tested because of the lack of suitable data. At the time of our study, precise
experimental results existed only for incident deuteron energies between 1 and 17
MeV (Meiner et al. 1967; Keller and Haeberli 1970; Grüebler et al. 1969; König et al. 1970;
Grüebler et al. 1970; Ohlsen et al.1973; Chang et al. 1973; Hardekopf et al. 1977; Grüebler
et al. 1979)
At higher energies, measurements of only vector analyzing power Ay have been
reported by Conzett et al. (1976a) in the energy range of 15-45 MeV. Unfortunately, we
shall see later that their reported values were grossly overestimated. Their data
indicated a strong possibility of the existence of a Ay = 1 point at 28 MeV and at a
backward angle (Conzett et al. 1976b), which implies that a Ayy = 1 point should also be
present at the same energy and angle.
Using the procedure outlined in the Appendix H, and the M-matrix for the d-α
scattering expressed in terms of the expansion coefficients (Ohlsen, Gammel, and
Keaton 1972):
⎡ A− B − 2 E − ( A + B )⎤
1⎢ ⎥
M = ⎢ − 2D 2C 2D ⎥
2⎢
⎣− ( A + B) 2E A − B ⎥⎦
we can find that

A2 − 2 B + C + D + E
2 2 2 2
Ayy =
3σ 0
and
2 Im(CD* − AE * )
Ay =
3σ 0
where
3σ 0 = A + B + C + D + E
2 2 2 2 2
We can see that Ayy = 1 requires B = 0 . However, Ay = ±1 requires not only B = 0 but
also A = m iE and C = ±iD . It is clear therefore that if Ay = 1 then Ayy = 1 value should
be also present at the same energy and angle because the two components share
the common condition of B = 0 . However, if Ayy = 1 it does not necessarily mean that
Ay = 1 because for such a point to occur two additional conditions have to be satisfied
for the vector analyzing power.
The M-matrix can be also expressed in therms of M ij components, where ij are the
deuteron spin projections in the incident and outgoing channels (see the Appendix I).
Using this form of the M-matrix, we can find that (Grüebler at al. 1975b) Ayy = 1 if
M 11 = − M 1−1 and Ay = ±1 if not only M 11 = − M 1−1 but also M 00 = m i 2 M 01 and
M 10 = m i 2 M 11 .
Unfortunately, the data of Conzett et al. (1976a and 1976b) are not precise enough to
be sure of the existence of the Ay = 1 point.
Extensive phase-shift analysis of the d-α data was carried out for energies of 3-17
MeV (Gürebler et al. 1975a). These results could be compared with the theoretical
phase shifts of Lemere, Tang, and Thompson (1976). However, from the discussion
presented here it is clear that there was a need to extend the study of this few-nucleon
system to higher deuteron energies.
Experimental arrangement and method

Method of measurements
The polarized deuterons were produced by a polarized ion source based on the
atomic beam method (see the Appendix G). Deuterons were accelerated by the
SIN20 injector cyclotron. The spin direction of the polarized beam extracted from the
cyclotron is fixed in the vertical direction by the magnetic field of the accelerator. The
method of measurements of the analyzing powers is fully described in Chapter 15.
However, I will summarise here some points that are relevant to this experiment.
The differential cross section for a scattering or reaction induced by polarized
deuterons can then be written as:
⎡ 3 1 1 ⎤
σ (θ ,ϕ ) = σ 0 (θ ) ⎢1 + (cos ϕ ) p z Ay (θ ) + (sin 2 ϕ ) pzz Axx (θ ) + (cos 2 ϕ ) p zz Ayy (θ )⎥
⎣ 2 2 2 ⎦
where ϕ is the angle between the direction of the spin and the normal to the
scattering plane, and σ 0 (θ ) is the cross section for an unpolarized beam. The quantities
20
pz and pzz are the source parameters describing the vector and tensor polarization of the
deuteron beam, and Ay , Axx and Ayy are the vector and tensor analyzing powers. In the
case discussed here, these are the analyzing powers for the d-α reaction.
As can be seen, the cross section σ (θ ,ϕ ) is independent of the tensor analyzing
power Axz . This quantity is related to the spherical tensor T21 .
Axz = 3T21
To measure this component, the angle β, which is between the spin and the beam
directions, should be 450 (see Chapter 15). For cyclotrons, β is either 00 (for the
horizontal reaction plane) or 900 for the detectors in the vertical plane. Consequently,
this component could not be measured in the present experiment.
The polarized ion source of the SIN injector cyclotron was equipped with three
successive RF transitions (cf Appendix G). Table 21.1 shows the available
configurations of the RF transitions and the associated theoretical maximum values of
the beam polarization. As can be seen, the system can produce pure vector
polarization with opposite signs and mixed vector and tensor polarization, also with
opposite signs. Thus the method described in Chapter 15 can be used to
measure the analyzing powers. This method consists not only in using detectors
located on two sides of the beam direction but also in changing the direction of
the beam polarization.
Table 21.1
Configurations of the RF transitions and the corresponding theoretical values of the beam polarization
Mode WF 2↔6 3↔5 pz pzz

a × × × 0 0
b 9 × × -2/3 0
c × 9 9 +2/3 0
d 9 9 × -1/3 +1
e × × 9 +1/3 -1
f × 9 × +1/3 +1
WF: weak field 1 ↔ 4 transition; 2 ↔ 6 and 3 ↔ 5 are the strong-field
transitions; × means the RF is off; 9means the RF is on.
However, it should be pointed out that this method requires carefully tuned RF
transitions in order to get the same absolute value of the polarization for both signs.
So, for instance, the measurement of Ay can be done with a purely vector polarized beam
switching between modes b and c in Table 21.1. The advantage of this option is that
the value of pz is larger than in any other case. On the other hand, using the mixed
1
vector and tensor polarized beam in which pz = p zz one obtains the same
3
statistical accuracy for the Ay and Ayy components because the factor 3 in the vector
term in the cross section formula cancels the factor 1/3.
In principle one could measure all three analyzing powers simultaneously by placing
detectors in the horizontal plane (β = 0°) for measuring the Ay and Ayy components,
and in the vertical plane (β = 90°) for the Azz component. This method would require a
complicated system of detectors and a more complex design of the gas target.
The same aim can be achieved by rotating a single-plane scattering chamber
around the beam axis and measuring the three analyzing powers in two runs. This has
the advantage that it is easier to place the detectors at extreme forward and
backward angles and to measure more scattering angles simultaneously. In our study,
measurements of the three analyzing powers Ay , Axx and Ayy were taken in two
runs:
(1) Scattering chamber horizontal, β = 0°. With the detectors placed on the left- and
right-hand side of the beam direction, and with the positive and negative beam
polarizations one has four counting rates N L+ , N L− , N R+ and N R− from which one can
calculate the ratios L and R which are independent of the values of the solid angles of
the detectors (cf Chapter 15):
N L+ − N L− 3 1
L= + −
= + pz Ay (θ ) + pzz Ayy (θ )
NL + NL 2 2
N R+ − N R− 3 1
R= + −
= − p z Ay (θ ) + p zz Ayy (θ )
NR + NR 2 2
From these equation one gets
1
Ay (θ ) = ( L − R)
3 pz
1
Ayy (θ ) = ( L + R)
p zz
(2) Scattering chamber vertical, β = 90°. With the detectors located up and down and
with different signs of the beam polarization one has four counting rates NU+ , NU− , N D+ ,
and N D− , which can be used to calculated the ratios U and D:
NU+ − NU− 1
U= = pzz Axx (θ )
NU+ + NU− 2
N D+ − N D− 1
D= = pzz Axx (θ )
N D+ + N D− 2
These ratios give
1
Axx = (U + D )
p zz
To check the consistency of the data collection one can also change between modes
a and b or a and c for the measurement of the vector analyzing power.
Detector arrangement
The 70 cm diameter reaction chamber used in these measurements was virtually the
same as described in Chapter 17. The polarized beam entered a 32 mm diameter gas
4
He target through a collimation system with a final aperture of 4 mm in diameter.
The beam was aligned by a four-way slit system at the entrance of the collimator. A
cylindrical gas target had a 270° exit foil window. The entrance and exit widows were
made of 6 μm Havar foils. The pressure in the gas cell was 1200 torr. The beam was
collected in a Faraday cup, which was equipped with an electrostatic suppressor
electrode.
The detectors were mounted on two movable segments, which could be rotated
independently by remote control. Each segment contained four detectors mounted 15°
apart and 25 cm from the centre of the chamber. For particles emitted at small angles (θlab
< 40°) additional segments fixed to the existing turntables were used with detectors
positioned 50 apart. These smaller segments also contained four detectors on each
side of the beam line. The whole scattering chamber could be rotated around the
beam axis, which allowed for the measurements to be taken in either horizontal or
vertical plane.
The collimators in front of the silicon surface-barrier detectors were 9 mm wide by 38
mm high for the normal turntables and 4.5 mm wide for the small-angle segments. The
heights of the collimators on smaller segments were suitably chosen to maintain the
similar spread in the azimuthal angles as for detectors at larger angles.
For the detection of scattered deuterons, the detector thickness was between 1 and 5
mm and suitable aluminium absorbers were placed in front of them. The recoil α-
particles were stopped in the 0.2 to 1 mm thick detectors. In all cases the bias voltage of
the detectors was adjusted just to stop the detected deuterons or α-particles. In this
way clean spectra were obtained with a background level smaller than 3%.
The beam polarization was monitored continuously for the tensor component using a
3
He(d,p)4He polarimeter at 0° (see Chapter 18). The 3He gas cell was integrated with
the Faraday-cup system, and the high-energy protons emitted at 0° were detected
either by a 5 mm thick silicon surface-barrier detector combined with suitably chosen
aluminium absorber or by a Csl scintillator. The absolute calibration of the polarization
analyser was checked using measurements of the d-α scattering at Los Alamos at 15.0
and 17.0 MeV (Grüebler at al. 1979).
The detector outputs were combined in a multiplexer connected to an analog-to-
digital converter (see Chapter 17). The resulting signals were routed to a PDP-11
computer, which performed an on-line data processing. The computer controlled also
the procedure for switching the sign of the beam polarization. This took place every few
seconds, whenever the beam current integrator reached a preset charge. Although the
deadtime correction was incorporated in the experimental procedure, the beam
intensity (which was up to 80 nA) was adjusted in such a way as to maintain the
deadtime corrections at the level of less than 10%.
Results of measurements
Excitation functions
Absolutely calibrated excitation functions for Ay , Axx and Ayy analyzing powers are
shown in Figures 21.1-21.3. The functions display smooth energy dependence for all
three analyzing powers. It is interesting to notice that in the vicinity of θ = 1500 and for
the incident deuteron energies of 33-38 MeV the component Ayy appears to be reaching
its extreme value of 1. However, contrary to the claim of Conzett at al. (1976b), the
Ay component remains well below the value of 1 in the whole energy range.
Figure 21.1. Excitations functions of the vector analyzing power Ay for the d-α elastic scattering. The
dots are the experimental results. The combined statistical and systematic errors are smaller than the
size of the displayed points. The thick lines are to guide the eye. The thin curves were drawn through
the earlier data taken at low energies (Grüebler at al. 1969; König at al. 1970; Grüebler at al. 1970;
Ohlsen at al. 1973; Chang at al. 1973; Hardekpf at al. 1977; Grüebler at al. 1979). Contrary to the
claim of Conzett at al. (1976b), the maximum value of Ay at 1500 is well below the extreme value of 1.
Figure 21.2. Excitation functions of the tensor analyzing power Ayy for the d-α elastic scattering. See
also the caption to Figure 21.1. These data show that the Ayy component is probably reaching its
0
extreme value of 1 at around 35 MeV and at the reaction angle of around 150 (c.m.).
Figure 21.3. Excitation functions of the tensor analyzing power Axx for the d-α elastic scattering. See
also the caption to Figure 21.1.
Angular distributions
Angular distributions of the analyzing powers Ay , Ayy and Axx for the elastic d-α
scattering were measured at the incident deuteron energies of 17.0, 20.2, 24.0, 28.0,
33.0, 38.2 and 42.8 MeV. The scattered deuterons were detected in the angular range
of between 20° and 80° (lab system) and the recoil α-particles between 10° and 40°. In
this way two overlapping angular ranges of 30°-110° and 100°-160° (c.m. system) were
obtained.
The range of statistical errors was between 0.001 to 0.010 with an average of about
0.005. In addition to statistical errors, a combined random error had to be also
included. This error was caused by such effects as background subtraction,
instrumental asymmetries, angle uncertainty, and by fluctuations of the beam polariza-
tion. By referring to the reproducibility of the data points and by using the overlapping
angular range for the detection of deuterons and α-particles, a value of 0.005 was
estimated for this random error. The lab angles for all the data are accurate to ±0.1°.
The beam energy was determined to about ±100 keV. The energy spread of the
polarized deuteron beam was about 100 keV.
The uncertainty of the absolute values of the measured analyzing powers are given
by uncertainties of the beam polarization. This uncertainty ranges from about 2.0%
at 17 MeV to about 4% at 42.8 MeV.
The distributions are shown in Figures 21.4 – 21.6. The shape of the angular
distributions for all three analyzing powers changes slowly with the increasing energy.
The Ay and Ayy components show large values at backward angles with a marked
maximum around 150°. Between 33.0 MeV and 38.2 MeV, Ayy reaches such high value
that a maximum Ayy = 1 is possible in this region. Even though the values for Axx are not
so large, the deep minimum at around 1000 and the large maximum near 150° could
serve as good reference points for beam polarization analysers.
Figure 21.4. Angular distributions of the vector analyzing power Ay for the d-α elastic scattering
between 17.0 and 42.8 MeV. The statistical errors are smaller than the size of the data points. The
dashed curves are the predictions based on the resonating-group theory (see Lemere, Tang and
Thompson 1976). The solid curves are the predictions based on the Faddeev three-body brake up
formalism (see Elbaz, Fayard and Lamot 1978).
Figure 21.5. Angular distributions of the tensor analyzing power Ayy for the d-α elastic scattering at
deuteron energies of 17.0 - 42.8 MeV. The statistical errors are smaller than the displayed data points. The
dashed curves are the predictions based on the resonating-group theory. The solid curves are predicted
using on the Faddeev formalism.
Figure 21.6. Angular distributions of the tensor analyzing power Axx for the d-α elastic scattering at
deuteron energies of 17.0 - 42.8 MeV. The statistical errors are smaller than the displayed data points. The
dashed curves are the predictions based on the resonating-group theory. The solid curves are predicted
using the Faddeev formalism.
Extreme values of the analyzing powers

Our study presents, for the first time, absolutely calibrated vector and tensor
analyzing powers for the d-α scattering in the energy range of between 17 and 43
MeV. A total of 21 angular distributions and 14 excitation functions have been
measured for the Ay , Ayy and Axx components of the analyzing powers.
Earlier measurements (Conzett et al. 1976a and 1976b) suggested the existence of
a Ay = 1 point at 28.6 MeV. These measurements are shown in Figure 21.7.
In the same figure, I have also presented our data. It is clear that the measurements
of Conzett et al. produced significantly overestimated values of the Ay component.
Our values are much lower and they do not come close to the expected maximum of
Ay = 1 . The displayed curves are the polynomial fits to the data. The best fits were
obtained using a second-order polynomial for our data and a third-order for the data
of Conzett et al.
The figure contains also our data for the Ayy component. Here, our data show a
possible maximum Ayy = 1 at around 35 MeV as indicated by the polynomial fit. The
best fit was obtained using a third-order polynomial.
Figure 21.7 has been constructed by taking measured values at the relevant
maximum in the angular distributions. The figure should be compared with the
measured excitation functions at 1500 as shown in Figures 21.1 and 21.2.
Finally, using our data and the earlier data at lower energy, we have constructed a
contour map of the tensor analyzing power Ayy . This map is shown if Figure 21.8.
Figure 21.7. Energy dependence of the Ay and Ayy analyzing powers at the reaction angle of
approximately 1500 (c.m.) extracted from the measured angular distributions. Our results were
obtained using the distributions presented in Figures 21.4 and 21.5. The Ay values of Conzett at al.
(1976a) were extracted in a similar fashion, i.e. by using the last maximum of their angular
distributions. The displayed lines are the best polynomial fits.
Figure 21.8. The contour map of the tensor analyzing power Ayy constructed using our experimental
data for the d-α elastic scattering and the earlier data obtained at low energies by Grüebler at al. (1975b).
The short straight lines are to guide the eye to the expected Ayy = 1 points. In particular, the map shows a
Ayy = 1 point near 35 MeV and 150° as suggested by the polynomial fit to our data presented in Figure
21.7.
Theoretical calculations
We have carried out theoretical calculations for our results using the resonating group
method (Lemere, Tang and Thompson 1976) and the three-body Faddeev formalism
(Elbaz, Fayard and Lamont 1978). Results of our calculations are presented in
Figures 21.4-21.6. The dashed curves are for the resonating group method and the
solid curves are for the Faddeev formalism.
In the resonating group study, a single-channel approach was chosen. The effects of
the open reaction channels were taken into account by using phenomenological imaginary
potentials. The nucleon-nucleon potential contained a spin-orbit component, which
was necessary to obtain non-zero values for the vector and tensor polarizations. The
exchange Coulomb potential was also taken explicitly into consideration. However,
no tensor interaction was included in the nucleon-nucleon force.
Even though the resonating-group formalism resulted in a fairly satisfactory agreement
with experimental results at energies below 10 MeV (Lemere, Tang and Thompson
1976), at higher energies, as used in our measurements, only the component Ay ,
which depends mostly on the spin-orbit interaction, shows a similar quality of an
agreement. For the components Ayy and Axx the shape of the angular distributions is
reproduced only approximately. This might be due to the absence of tensor forces in the
theoretical calculations.
The figures show also calculations based on the Faddeev formalism. Here, the
agreement with the experimental data is significantly better. In our calculations, we have
used two-body N-N and N-α interactions. In addition, for practical reasons, a limited
number of the two-body partial waves was used. In the N-α system, only S1/2, P1/2 and
P 3/2 partial waves were used and in the N-N system only partial waves with isospin
zero and l ≤ 1 were included. For the latter system, the coupled 3S1-3D, partial waves
with several different parameterisations were taken into account.
The agreement between experimental data and theoretical calculations deteriorates with
the increasing deuteron energy. It is not clear whether it is because of the neglected
higher partial waves or because of the break-up of the α-particle. (The threshold for
the d + α →3He + t break-up is 21.4 MeV.) Our study suggests that more detailed
calculations and perhaps even more refined theories are needed in order to
understand the structure of this highly excited few-nucleon system.
Phase-shift analysis
A new computer search code for the phase shift analysis of spin-one scattering from
spinless targets was used to analyse the data. The calculation sections of this code
were adapted from the older SPINONE program (McIntyre 1965) but we have added
a general-purpose search routine MINUIT (James and Roos 1975) to allow for an
automatic search. In addition, we have also altered the input data section to simplify
the use of the program and to make the input clearer and more flexible.
Since phase shifts were available at 17 MeV (Grüebler et al. 1975a), we have used
them as starting values for the 17.0 and 20.2 MeV searches in our preliminary
analysis. At 17 MeV, the present data and the older Los Alamos data (Ohlsen et. al.
1973) were analysed separately in order to see if the results were compatible. Our
data do not include the Axz component but they cover a wider range of angles. The
earlier data were used as the basis for the normalization of our new measurements,
and consequently no major differences in the phase shifts were expected. This
expectation was confirmed by our calculations.
The remaining analysis was carried out using the phase shift results from lower
energies as the starting values. Gradient searches were performed to minimize the
χ2 function with all phases and mixing parameters varied simultaneously to achieve
the best fit to the data. The searches were carried out in interrupted intervals so that
intermediate results could be studied and the direction of the search monitored. The
χ2 values are listed in Table 21.2. The energy dependence of the real parts of the
phase shift parameters and the total reaction cross sections are presented in Figure
21.9. The corresponding fits to the data are shown in Figures 21.10-21.13.
Table 21.2
2
The χ values for the phase shift analysis
Ed N χ2
(MeV)
17.0 77 1.1
20.2 91 1.8
24.0 92 2.5
28.0 88 3.0
33.0 92 7.0
38.2 112 13.7
42.8 92 10.4
N – The number of data points
Figure 21.9. The energy dependence of the phase-shift parameters for the d-α elastic scattering in the
energy range of 0-43 MeV. The values below 17 MeV are from Güebler et al. (1975a). The higher-
energy values are from our analysis. The energy dependence of the total reaction cross-section σR
calculated using the imaginary parts of phase shift parameters are also shown in the figure.
Figure 21.10. Differential cross-sections for the d-α scattering. The solid lines are the results of our
phase-shift analysis.
Figure 21.11. Angular distributions of the vector analyzing power Ay (points) for the d-α elastic
scattering are compared with the results of our phase-shift analysis (solid lines).
Figure 21.12. Angular distributions of the tensor analyzing power Ayy (points) for d-α elastic scattering
are compared with the results of our phase-shift analysis (solid lines).
Figure 21.13. Angular distributions of the tensor analyzing power Axx (points) for the d-α elastic
scattering are compared with the results of our phase-shift analysis (solid lines).
The energy range of our new data covers the excitation energy of 12.8-30.0 MeV in
the compound 6Li nucleus. No resonances decaying to the d+α system are known in
this range of the excitation energies. Some of the phase shift parameters show
broad structures above 17 MeV, but it is not clear whether they can be associated
with resonance interactions. The most striking change is in the P1 (J =1, l = 1), which
decreases from near zero at 17 MeV to around –1500 at around 43 MeV (see Figure
21.9). It can be also noticed from Figures 21.10-21.13 that the fits to the data
deteriorate at higher energies. This might indicate that higher l - values are needed.
However, our analysis shows that G waves remain small over the whole range of
energies.
The lack of indication for the presence of resonances in this analysis does not
necessarily prove that there is no resonant behaviour in this region of the excitation
energies. To uncover such interactions one would have to carry out measurements
at significantly smaller energy steps. In particular, it would be necessary to study
more closely the energy region around 35 MeV where the expected Ayy = 1 maximum
is located.

We have carried out precise measurements of the angular distributions of the
differential cross sections and analyzing powers ( Ay , Ayy and Axx ) for the d-α
scattering in the energy range of 17-43 MeV. We have also measured excitation
energies at selected reaction angles.
Our results show a possible Ayy = 1 maximum at around 35 MeV at a backward angle
but do not confirm the previously claimed Ay = 1 maximum (Conzett at al. 1976b) at
around 28 MeV. In fact, even though our results show a maximum around this
energy, its value is well below the expected value of 1.
We have analysed our results using the resonating group theory and the Faddeev
formalism. The calculations using the Faddeev formalism follow more closely our
experimental data than the calculations based on the resonating group theory.
We have also carried out a phase shift analysis of our data using a wide range of spin
values, J = 1, 2, 3, and 4 and l = 0, 1, 2, 3, and 4. The analysis does not reveal any
clear resonance behaviour in this higher energy region. Measurements in smaller
energy steps would be necessary to study the possible resonance behaviour. A more
detailed study around 35 MeV would be desirable to investigate the demonstrated
here possibility of the existence of the Ayy = 1 maximum.
References
Charnomordic, B., Fayard, C. and Lamot, G. H., 1977, Phys. Rev. C15:864
Chun, D. S., Han, C. S. and Lin, D. L 1973, Phys. Rev. C7:1329
Chang, C. C., Glavish, H. F., Avida, R. and Boyd, R. N. 1973, Nucl. Phys. A212:189
Conzett, Dahme, H. E., W., Leemann, Ch., MacDonald, J. A. and Meulders, J. P.
1976a, Proc. Fourth Int. Symp. Phen. in Nuclear Reactions, Zürich, eds W.
Grübler and V. König (Birkhauser, Basel) p. 566
Conzett, H. E., Seiler, F., Rad, F. N., Roy, R. and Larimer, R. M. 1976b, Proc. Fourth
Int. Symp. Phen. in Nuclear Reactions, Zürich, eds W. Grübler and V. König
(Birkhauser, Basel) p. 568.
Elbaz, E., Fayard, C. and Lamot, G. H. 1978, “Proc. Int. Conf. Nucl. Structure, Tokyo
1977”, J. Phys. Soc. Japan 44:304
Gammel, J. L., Hill, B. J. and Thaler, R. M. 1960, Phys. Rev. 119:267
Grüebler, W., König, V., Schmelzbach, P. A. and Marmier, P. 1969, Nucl. Phys.
A134:686
Grüebler, W., König, V., Schmelzbach, P. A. and Marmier, P. 1970, Nucl. Phys.
A148:391
1975a, Nucl. Phys. A242:265
1975b, Nucl. Phys. A242:285
Grüebler, W., Brown, R. E., Corell, F. D., Hardekopf, R. A., Jarmie, N. and Ohlsen, G.
G. 1979, Nucl. Phys. A331:61
Hackenbroich, H. H. 1975, Proc. 2nd Int. Conf. on Clustering Phenomena in Nuclei II,
Maryland, eds D. A. Goldberg, J. B. Marion, S. J. Wallace, ORO-4856-26, p. 107
Hackenbroich, H. H., Heiss, P. and Le-Chi-Niem, 1974, Nucl. Phys. A221:461
Hardekopf, R. A., Grüebler, W., Jenny, B., König, V., Risler, R., Bürgi, H. R. and
Nurzynski, J. 1977, Nucl. Phys. A287:237
Jacobs, H., Wildermuth, K. and Wurster, E. 1969, Phys. Lett. 29B:455
James, F. and Roos, M. 1975, Comp. Comm. 10:343.
Keller, L. G. and Haeberli, W. 1970, Nucl. Phys. A156:465
König, V., Grüebler, W., Schmelzbach, P. A. and Marmier, P. 1970, Nucl. Phys.
A148:380
Lemere, M., Tang, J. C. and Thompson, D. R. 1976, Nucl. Phys. A266:1
McIntyre, L. C. 1965, PhD Thesis, University of Wisconsin.
Meiner, H., Baumgartner, E., Darden, S. E., Huber, P. and Plattner, G. R. 1967, Helv.
Phys. Acta 40:483
Ohlsen, G. G., Gammel, J. L., and Keaton, 1972, Phys. Rev. C5:1205.
Ohlsen, G. G., Lovoi, P. A., Salzman, G. C., Meyer-Berkhout, U., Mitchell, C. K. and
Gruebler, W. 1973, Phys. Rev. C8:262
Shanley, R. E. 1969, Phys. Rev. 187:1328
Thompson, D. R. and Tang, Y. C. 1973, Phys. Rev. C8:1649
22
A Study of the 5.65 MeV 1+ Resonance in 6Li
Key features:
1. The aim of this work was to investigate the 1+ resonance at the excitation energy of
5.65 MeV in 6Li.
2. The distinctive feature of our study is the high precision of experimental data.
3. We have measured the differential cross sections and analyzing powers ( iT11 and
tensor T20 , T21 and T22 ) in the energy range of 6.04 to 7.05 MeV in small energy steps. The
angular range of the data was 11° and 165° (c.m.). We have collected a total of 1126
high-precision data points
4. Using our data, we have carried out phase-shift analysis. Our data allowed us to
obtain reliable sets of not only real but also imaginary components of phase-shift and
mixing parameters.
5. Using the results of our phase-shift analysis as input data, we have carried out the R-
matrix analysis. In these calculations we have used a computer code of McIntyre
(1965). However, we have modified it to allow for (for the first time) an automatic
search of R-matrix parameters. We have obtained a close agreement between the R-
matrix results and the phase shift and mixing parameters even for the imaginary
components.
6. Our study yields precise values of the resonance parameters. It also demonstrates
the sensitivity of the data not only to tensor forces but also to the number and the
nature of assumed open channels.
7. Our results show that by carrying our precise measurements in the vicinity of a
resonance one can obtain detailed information not only about the physical parameters
of the resonance but also about the mechanism of nuclear interactions (the strength
and nature of nuclear forces and the contributions of reaction channels).
Abstract: Previous analyses of the d - 4He elastic scattering have established the presence
of a 1+ resonance near 5.7 MeV excitation energy in 6Li. The energy dependence of the s-
wave to d-wave mixing parameters through this resonance gives an indication of the tensor
force contributing to the interaction. In this work, we have made a detailed study of this
parameter by measuring and then analyzing angular distributions of the differential cross
section and all four analyzing powers for the 4He(d,d)4He elastic scattering at seven energies
between 6 and 7 MeV. The phase-shift analysis of these data provides a detailed
parameterization, which can be compared with theoretical calculations and shows for the
first time the necessity for a complex mixing parameter. The R-matrix fits to the phase shifts
establish more precisely the location and the width of this resonance.
Introduction
The deuteron-α elastic scattering is a popular way to investigate the structure of 6Li.
Measurements of the differential cross sections and analyzing powers for the d-α elastic
scattering have been carried out for up to around 50 MeV (Grüebler et al. 1975; Grüebler
et al. 1980; Mclntyre and Haeberli 1967; Ohlsen et al. 1973; Schmelzbach et al. 1972).
Phase-shift analyses of these data show several well-separated resonances, particularly
for energies below 10 MeV. Of a particular interest for the investigation of tensor
forces is the 1+ resonance, which occurs near the excitation energy of 5.7 MeV, and
which corresponds to the deuteron energy of 6 to 7 MeV (lab).21
Phase-shift analyses of experimental results in this region show a mixing between the
s- and d-waves. Such a coupling between orbital angular momenta differing by two units
is usually interpreted as a result of tensor interaction. Unfortunately, in earlier
measurements, the energy steps in this region have been too large to determine the
details of the mixing parameter.
In order to extract more accurate information about this resonance, we have carried out
precise measurements of the differential cross section σ 0 and vector and tensor
analyzing powers iT11 , T20 , T21 and T22 in the energy range of 6.04 to 7.05 MeV and in small
steps of 100 to 200 keV. These independent observables were measured in the angular
range of between 11° and 165° (c.m.). At each energy, about 160 data points were
collected to support the intended detailed phase shift and R-matrix analyses.
Experimental arrangement and method

The measurements were carried out using the ETH22 atomic beam polarized ion source (see
the Appendix G) and the EN tandem Van de Graaff accelerator. The average beam
current on the target was 30 nA.
The polarized deuteron beam entered the gas target, 16 mm in diameter, through a
collimation system with a final aperture of 2 mm in diameter. A gas cell had a 270° exit foil
window. The entrance and exit windows were made of 1.25 μm Ni foils. A 60 x 20
mm oval-shaped gas cell with 6.5 μm Mylar foil was used for the extreme forward
angles. The pressure in the cell was set at 600 Torr for the Ni foil and 200 Torr for the Mylar.
Pressure variations during the measurements were less than 1 %. The energy of the
polarized deuteron beam was determined in the middle of the gas cell to better than 20 keV.
A general description of the scattering chamber is given in Chapter 17. The arrangement of
detectors was the same as described in Chapter 21 for the measurements at higher
deuteron energies at SIN23.
A 3He polarimeter set up at 0° was used to monitor the tensor polarization of the beam.
Details of this polarimeter are described in Chapter 18. The method of measurements is
given in Chapter 15. The essential parts of the procedure are the use of the left and right
detectors at the same scattering angles, adjusting the spin direction of polarized deuterons
by a Wien filter to an optimal position for the measurement of each specific analyzing power
component, and switching the sign of the beam polarization every few seconds. This
method eliminates first-order errors from geometrical effects and from deviations
from the required spin direction.
Results of measurements together with results of our phase-shift analysis are
presented in Figures 22.1-22.5. The errors are smaller than the displayed data points.
21
The difference between the binding energies of 6Li, deuterons, and alpha particles is 1.4743 MeV.
Elab = 1.5Ec.m. for the d-α scattering. Therefore, the excitation of the 5.65 MeV state in 6Li
corresponds to Elab = 6.3 MeV.
22
Laboratorium für Kernphysik, Eidg. Techische Hochschule, Zürich, Switzerland.
23
A total of 1126 data points have been collected. It is interesting to notice that all
observables measured in the energy range of the 1+ resonance do not show any
dramatic energy dependence. Consequently, on the basis of the experimental data
alone, i.e. without carrying out a phase-shift analysis, one would not expect a
resonance. Such behaviour is typical for broad resonances in light nuclei.
Figure 22.1. Angular distributions of the differential cross-sections for the d-α scattering between 6
and 7 MeV. The dots and crosses are larger than the statistical errors. The distributions identified by
large filled-in circles refer to the left-hand side scale; crosses refer to the right-hand side scale. The
curves are the phase-shift fits.
Figure 22.2. The vector analyzing powers iT11 for the d-α scattering between 6 and 7 MeV. The dots
are larger than the statistical errors. The curves are the phase-shift analysis fits.
Figure 22.3. The tensor analyzing powers T20 for the d-α scattering between 6 and 7 MeV. The dots
Figure 22.4. The tensor analyzing powers T21 for the d-α scattering between 6 and 7 MeV. The dots
Figure 22.5 The tensor analyzing powers T22 for the d-α scattering between 6 and 7 MeV. The dots
For the purpose of the phase shift analysis, it is useful to separate the errors of the
measurements into relative and absolute errors. The relative errors describe
uncertainties in the relative positions of data points. The absolute errors (the scale
errors) give information about the uncertainty of the overall normalization of the data.
For the differential cross-sections, relative errors arise from such effects as statistical
uncertainty in the number of counts, variation in the gas target pressure, errors in
background subtraction, beam-current integration, detector position uncertainty, relative
solid angles. The statistical uncertainties from the number of counts were very small. All
relative errors are not larger than 2%.
The scale error of the differential cross-section data is also estimated at 2.0%. This
includes such effects as target gas purity, calibration of the current integrator and target
pressure gauge.
For the measurements of the analyzing-powers, the statistical uncertainties from the
number of counts are also very small, generally less than 0.005. Although most large
systematic errors cancel for the analyzing-power data obtained with symmetric detector
systems and beam polarization reversal, some small false asymmetries remain. The
scale error i.e. the normalization of the analyzing-power measurements arises only
from the calibration of the polarimeter that monitored the beam polarization. This
uncertainty is estimated at 1 %.
The phase-shift analysis

General discussion and procedure
As the experimental data have become more complete, the phase shifts used to
parameterise d-α scattering have become more precise. Early analyses of differ-
ential cross sections were superseded by Mclntyre and Haeberli (1967) who included
tensor polarizations measured by double scattering. With the development of polarized ion
sources, vector analyzing powers and more accurate tensor analyzing powers were added
to the data. The analysis of Schmelzbach et al. (1972) used for the first time a fairly
complete set of measurements of differential cross sections, vector and all three tensor
analyzing powers to determine the d-α phase shifts between 3 and 11.5 MeV. Grüebler et
al.(1975) extended the analysis to 17 MeV and improved its accuracy by adding new
experimental data from Los Alamos (Ohlsen et al. 1973). The data and analysis have
been later extended to higher energies of up to around 40 MeV (Grüebler et al. 1980).
Descriptions of the phase-shift parameterization and the search procedure may be found
in publications of Mclntyre and Haeberli (1967) and Schmeltzbach et al. (1972). Of
particular relevance to our analysis is the use of the Blatt-Biedenharn prescription
(1952) to account for the off-diagonal terms in the scattering matrix, which depend
not only on the eigen phase shifts but also on mixing parameters (see the Appendix J).
These off-diagonal terms represent coupling between angular momenta l differing by
two units. They are therefore a measure of the tensor force in the interaction.
For the 1+ resonance studied here, Mclntyre and Haeberli (1967) showed how the
mixing parameter is affected by the ratio of the s- and d-wave reduced widths. Their
calculations are shown in Figure 22.6.
Figure 22.6. The dependence of the calculated phase shifts δα1 and δ β1 and of the mixing parameter
ε1 on the ratio of the s - wave to d - wave reduced widths. The rations are: (a) 0, (b) 0.001, (c) 0.010 and
(d) 0.035. The other parameters are: the bombarding energy at resonance – 6.2 MeV; the total laboratory
energy at resonance – 7 MeV; and the interaction radius – 4.5 fm.
For the pure d-wave resonance, i.e. with no tensor force, ε 1 is either zero or π/2
(which just exchanges the roles of the two eigen phase shifts δα1 and δ β1 ).24 In this
case, there is a sudden transition from ε 1 = 0 to ε 1 = π/2 as one crosses the
resonance from the lower to higher incident energies. Increasing the ratio of the s-
wave to d-wave reduced widths smoothes out the transition between ε 1 = 0 and ε 1 =
π/2. With an increased accuracy of the experimental data, greater angular range, and
closer energy steps in our measurements, we expected to determine the shape of this
mixing parameter with a greater precision than it was previously possible and thus to
obtain more precise information on the ratio of the wave functions and on the tensor
coupling.
24
The notation used here is ε , where J is the spin of the resonant state. The same applies to δ .
J J
The distinction is also made between the phase shifts with or without mixing. Phase shifts with mixing
are denoted as δ kJ where k = α, β, …, whereas phase shifts without mixing are denoted as δ lJ where
l is the orbital angular momentum of the incoming wave.
Description of the phase-shift formalism
For an outline of the phase-shift formalism see the Appendix J. Here, I am
presenting only selected points relevant to our analysis.
With no mixing between partial waves, each diagonal element of the scattering or
collision matrix U lJ′l can be described by the corresponding phase shift
U lJ′l = e 2iδ l
If one considers only elastic scattering, the phase shifts are real, automatically giving
a unitary collision matrix. Allowing the phase shifts to be complex accounts for
absorption into inelastic channels. This is often represented by the inelastic parameter
ηlJ = e −2 Im δ
J
l
For J = 1+, the scattering matrix with non-zero off-diagonal elements has the following
form (Blatt-Biedenharn 1952):25
⎡ 2 iδ α1 2 iδ 1 1 2 iδ 1 ⎤
⎤ ⎢(cos ε )e + (sin 2 ε 1 )e β (sin 2ε 1 )(e 2iδ α − e β ) ⎥
1
2 1
⎡U 00
1 1
U 02 2
U =⎢ 1 1 ⎥
=⎢ ⎥
⎣U 20 U 22 ⎦ ⎢ 1 sin( 2ε 1 )(e 2iδ α − e 2iδ β ) 2 iδ 1
1
(sin 2 ε 1 )e 2iδ α + (cos 2 ε 1 )e β ⎥
1 1
⎣ 2 ⎦
The unitary condition of the collision matrix containing complex phase-shift parameters
has to be modified. The new condition is satisfied if the imaginary parts of the phase
shifts are positive ( 0 ≤ ηlJ ≤ 1 ). For complex mixing parameters, additional inequalities
that must be satisfied were given by Arvieux (1967). However, Seyler (1969) pointed out
that only one of these relations is sufficient:
2 iδ β
e − e 2iδ α sinh 2 (Im 2ε ) ≤ (1 − η β2 )(1 − ηα2 )
Examination of this inequality leads to the conclusion that the imaginary part of the
mixing parameter can be non-zero only when there is absorption in both of the mixed
channels, i.e. when Imδα and Imδβ > 0. The inequality proposed by Seyler was included in
our phase-shift analysis as a required constraint.
Precision data are needed to determine not only the real but also the imaginary
component of the mixing parameter. As will be shown later, our data yielded
sufficiently high-quality values for this component to include it in our R-matrix
calculations.
Phase-shift results and discussion
The starting phase-shift parameters were taken from the results of Grüebler et al. (1975).
We have also included some earlier data (Schmetzbach et al. 1972) at both ends of our
energy range. The phase-shift analysis was carried out using a modified code of
Mclntyre (1965). The program used in the calculations minimises the χ2 function for the
data consisting of the differential cross sections and of spin-one analyzing powers iT11 ,
T20 , T21 and T22 .
A summary of the quality of fits generated by the phase-shift analysis is shown in
Table 22.1. As can be seen by looking at the χ2 and weighted variance values, the
25
Blatt and Biedenharn use letter S for this matrix (see also the Appendix J).
resulting fits to the old data (Grüebler et al. 1975) were generally poor. In contrast,
the fits to the new data were significantly better. The reason is that the new data are
of much higher quality.
Table 22.1
Summary of the quality of the phase-shift analysis
Ed Number of data χ2 Weighted

points variance
4.81a) 61 2.2 3.4
a
5.32 ) 59 1.5 2.4
a
5.83 ) 65 1.5 2.3
6.04 162 0.67 0.82
6.24 163 0.75 0.91
6.44 161 0.71 0.86
6.66 160 0.80 0.98
6.76 151 0.66 0.82
5.85 161 0.67 0.81
7.05 168 1.18 1.42
a
7.86 ) 83 1.1 1.7
a
) Old data (Grüebler et al. 1975)
Figure 22.7. The real parts of the eigen phase shifts δ α1 , δ β1 and of the mixing parameter ε 1 for the
d - α elastic scattering. The dots and solid triangles are the result of the analysis of the present
data; open circles and triangles are for the calculations based on the data of Schmelzbach et al. (1972).
The scale for ε1 is on the right-hand side. The curves are the single-level R-matrix fits.
Figure 22.8. The imaginary parts of the eigen phase shifts δα1 , δ β1 and of the mixing parameter ε1 for
the d-α elastic scattering. The dots and solid triangles are the result of analysis of the present data;
open circles and triangles for the calculations based on the data of Schmelzbach et al. (1972). The curves
are the single-level R-matrix fits. The total reaction cross sections calculated from the complete set of
phase shifts are shown in the lowest section of the figure.
The good quality of fits to the data points can be also seen in Figures 22.1-22.5. The
only systematic deviation in the fits occurs near the maximum in T22 . All other
observables are well reproduced by the phase shift calculations. The overall χ2 value
for the new data was 0.78 and the weighted variance 0.95.
The phase parameters that are affected by the 1+ resonance under investigation are
shown in Figures 22.7 and 22.8. The curves in these figures are results of the single-level
R-matrix fit to the phases, as discussed later.
The deuteron break-up 5He+p channel is open in the entire range of energies used in the
present experiment. The 5Li+n channel opens at 6.3 MeV and the clear change in
Im δ α1 might be associated with the opening of this channel. The imaginary part of ε1 is
negligible except when the real part passes through 45°. It might be of interest to
mentions that the imaginary component produces a noticeable improvement in the
fits to the data in this region. Above and below the resonance the unitary condition
forces Im ε 1 to be zero because one of the inelastic phases approaches zero at these
energies.
The R-matrix analysis

Level parameters of the 1+ resonance in 6Li can be extracted using the single-level
approximation of the R-matrix theory. As mentioned earlier, the shape of the mixing
parameter depends on the ratio of two partial widths contributing to the resonance, and
using our more precise data we expected to determine this ratio with a higher
precision than before. In addition, we expected a more precise description of the 1+
resonance.
To achieve these goals, we have carried out R-matrix calculations using our phase
shift parameters. All the calculations were done using the single-level R-matrix code
LEV21 (McIntyre 1965), which we have modified by combining it with the general
search routine MINUIT (James and Roos 1975). The modified program performed an
automatic search of R-matrix parameters while fitting the generated earlier phase shifts.
The resulting level parameters are listed in Table 22.2. They are compared with the
parameters obtained in previous analyses (McIntyre and Haeberli 1967;
Schmeltzbach et al. 1972). The calculated values for the phase parameters are shown
as solid lines in Figures 22.7 and 22.8. The parameter errors from our fits were of the
order of one part in the last significant digit for values listed in the Table 22.2. As can be
seen from this table, the ratio of the s-wave to d-wave reduced widths is 0.002, confirming
the very small effect of the tensor interaction in this resonance.
Table 22.2
Single-level parameters of the J = 1+ resonance
All parameters, except for the intercation radius, are in MeV.

a
) – McIntyre and Haeberli (1967); b) – Schmeltzbach et al. (1972)
Even though the imaginary component of the mixing parameter is not well determined in
this search, it is encouraging that the R-matrix calculations give its correct sign and
shape. Our ability to include this component in our study shows the importance of
precise experimental results in this type of analyses.
Comparison with microscopic calculations

The earliest resonating group calculations for the d-α scattering used no spin-orbit or
tensor force in the nucleon-nucleon potential and thus could not predict polarizations.
Lemere, Tang and Thompson (1976) added a spin-orbit component to the potential in their
calculations that extended to around 54 MeV (c.m.). The most complete calculation
below Ec.m. = 8 MeV was made by Hackenbroich et al. (1974) who not only used a
nucleon-nucleon potential containing spin-orbit and tensor terms, but also coupled
several reaction channels in their cluster-model scheme. These authors were able to
predict the mixing between the l = 0, and l = 2 partial waves. In their calculations, the
break-up of the deuteron was assumed to proceed through the 5He-p and 5Li-n
fragmentations. They have also considered the excited states of 5He and 5Li in such
break-up channels.
Some results of Hackenbroich et al. (1974) are shown in Figure 22.9 together with our
values for the ε 1 mixing parameter. The dashed curve presents an interesting case. It
shows that in principle, if one includes all inelastic channels, strong coupling between the l
= 0 and l = 2 partial waves is possible without a tensor forces. Referring to the
calculations presented in Figure 22.6, this strong coupling is demonstrated by the
strong smoothing out of the shape of the energy dependence of the ε 1 mixing
parameter. However, these results appear to be physically meaningless for the studied
resonance because the calculated curve does not even come close to the experimentally
determined values of the mixing parameter.
The fit to the experimentally determined mixing parameters is greatly improved if tensor
force is included in the calculations. However the resulting fit is still far from perfect (see the
full line in Figure 22.9). The best fit is obtained if tensor force is retained but if inelastic
break-up fragmentations (5He*-p and 5Li*-n) are excluded from the calculations. As the
energies used in our measurements are below the thresholds for these inelastic
fragmentations their exclusion is not surprising. What is surprising, however, is that
their inclusion produces such a large effect.
Figure22.9. Comparison of the results of our phase shift analysis (dots) for the J = 1 mixing parameter
with the microscopic calculations of Hackenbroich et al. (1974) (curves). Solid line: calculations with
tensor potential and with all fragmentations. Dashed line: calculations without tensor potential but with
all fragmentations. Dashed-dotted line: calculations with tensor potential but without 5He*-p and 5Li*-n
fragmentations. The lower section of the figure shows our results (dots) for the absolute values of the
1 2
off-diagonal element U 02 of the scattering matrix compared with the calculation of Schütte et al.
(1976) who used tensor potential and all fragmentations.
2
It is also useful to look at the magnitude of the off-diagonal matrix element U 02
1
. This
element gives a direct measure of the mixing between the two partial waves and is
not complicated by the rapid variation of the ε 1 mixing parameter near the phase
crossing region. In the lower part of Figure 22.9 we have plotted this quantity as
generated by our analysis (dots) together with the calculations of Schütte et al. (1976)
who included tensor force and all fragmentations. Again it is clear that the calculated
mixing is too high when coupling to all channels is included.
In conclusion, we have demonstrated that tensor interaction in the d-α scattering at
energies corresponding to the 1+ resonance in 6Li is small but that it still has a
decisive influence on the coupling between the s- and d-waves. We have also
demonstrated that in order to study the details of resonance interactions it is
essential to have precise data for the distributions of the differential cross sections
and analyzing powers. It is by having a full complement of such measurements that
we were able to gain detailed insights into the 1+ resonance in 6Li and into the
accompanying reaction mechanism. Our data can also serve as a reliable basis for
more refined analyses.
References
Arvieux, J. 1967, Nucl. Phys. A102:513.
Blatt, J. M. and Biedenharn, L. C. 1952, Rev. Mod. Phys. 24:258.
Grüebler, W., Schmelzbach, P. A., König, V., and Boerma, D. 1975, Nucl. Phys.
A242:265.
Grüebler, W., König, V., Schmelzbach, Jenny, B., Bürgi, H. R., Hardekopf, R. A.,
Nurzynski, J. and Risler, R. 1980, Nucl. Phys. A334:365.
Hackenbroich, H. H., Heiss, P. and Le-Chi-Niem, 1974, Nucl. Phys. A221:461.
James, F. and Roos, M. 1975, Comp. Comm. 10:343.
Lemere, M., Tang, Y. C. and Thompson, D. R. 1976, Nucl. Phys. A266:1.
McIntyrem L. C. 1965, PhD thesis, University of Wisconsin
Mclntyre, L. C. and Haeberli, W. 1967, Nucl. Phys. A91:382.
Ohlsen, G. G., Lovoi, P. A., Salzman, G. C., Meyer-Berkhout, U., Mitchell, C. K. and
Grüebler, W. 1973, Phys. Rev. C8:1262.
Schmelzbach, P. A., Gruebler, W., König, V. and Marmier, P. 1972, Nucl. Phys.
A184:193.
Schütte, W., Hackenbroich, H. H., Stöwe, H. Heiss, P. and Aulenkamp, H. 1976, Phys.
Lett. 65B:214.
Seyler, R. G. 1969, Nucl. Phys. A124:253.
23
A Study of Spin-orbit and Tensor Interaction of Polarized Deuterons
with 60Ni and 90Zr Nuclei
Key features:
1. We have carried out precise measurements of the differential cross sections σ 0 and
all four analyzing powers iT11 , T20 , T21 and T22 for the elastic scattering of
60
polarized deuterons from Ni and 90Zr nuclei.
2. The angular distributions were measured in the energy range of 9 - 15 MeV (lab) and
for the scattering angles of 400 - 1600 (c.m.). In addition, we have also measured
excitation functions for the differential cross sections and for the T20 tensor analyzing
power.
3. We have analysed our experimental results using optical model potential that contained the
central, spin-orbit, and tensor TR potentials. Each potential had both real and imaginary
components.
4. We have found that five of the optical model parameters could be fixed and that three could be
described using mass-dependent formulae. The remaining 10 parameters were varied to
optimise the fits to the experimental data for individual energies and target nuclei.
5. Our analysis resulted in determining the central, spin-orbit and tensor potentials for
the deuteron-nucleus interaction.
6. In particular, we have found that the imaginary spin-orbit component was necessary to
improve the fits to the data thus confirming the earlier results of Goddard and Heaberli (1977).
Abstract: Angular distributions of the differential cross-sections σ0 and analyzing

r
powers iT11 , T20 , T21 and T22 have been measured for the elastic (d , d ) scattering from
60 90
Ni and Zr over a wide range of scattering angles. The incident deuteron energies
were at 9, 12 and 15 MeV for 60Ni nuclei and 10, 11, 12 and 15 MeV for 90Zr. Excitation
functions for σ 0 and T20 have been also measured at 175°(lab) in the approximate
energy range of 6-13 MeV for both target isotopes. The experimental results have been
analysed using the optical model with the complex central, spin-orbit and tensor TR
potentials. Excellent fits to all experimental angular distributions have been obtained.
The main features of the excitation functions have been also well reproduced. Out of the
total of 18 parameters describing the interaction potential, five could be fixed and three
could be constrained by simple mass-dependent functions. Further evidence for the
presence of an imaginary component of the spin-orbit and tensor potentials is supplied
by the analysis of the present data.
Introduction
The central part of the nuclear interaction, which can be studied using angular
distributions of the differential cross sections, is relatively well known. Some
information on the spin-depended forces can be also obtained by analyzing
differential cross sections but a more reliable way is to measure and analyse the
distributions of analyzing powers. In particular distributions of tensor analyzing
powers can yield information not only about spin-orbit but also about tensor forces.
Spin-orbit forces are relatively well known for deuterons but usually only a real
component is used in analyses of experimental data. Goddard and Haeberli (1978)
found evidence for the presence of the imaginary component. These authors measured
and analysed angular distributions for deuteron scattering from medium weight nuclei in
the energy range of 10-15 MeV. They have found that the fits to the data can be
improved significantly if an imaginary component is included in the spin-orbit
interaction. The depth of the imaginary spin-orbit potential used in their calculations
was about a half of the real component. Both components had the same sign.
Much less is known about the tensor interaction because to study it one needs to
have good quality data for the tensor analyzing powers. Unfortunately, such data are
scarce.
In our study we have carried out precise measurements of the differential cross
sections and of all four analyzing powers iT11 , T20 , T21 and T22 with the aim to study
the details of spin-orbit and tensor interactions.
The experimental method and results

The experimental method and procedure was described in Chapters 15 and 17.
Briefly, the measurements were made using the ETH atomic beam polarized ion source
(see the Appendix G) and the EN tandem electrostatic accelerator. Targets consisted
of approximately 1 mg/cm2 self-supporting foils of enriched isotopes with the
enrichment greater than 98%. To allow for a simultaneous detection of scattered
particles on both sides of the beam, the targets were mounted at 90° to the beam
direction for the forward and the backward angle measurements, and rotated by 45° for
the intermediate angles. Four silicon surface-barrier detectors were mounted on both
sides of the beam line, 7.5° apart and 25 cm from the centre of the chamber.
The differential cross sections were derived directly from the yields obtained in the
measurements of the analyzing powers. The absolute normalization was determined
by measuring Rutherford scattering of 4 MeV deuterons at a number of angles below
60° (lab). An overall normalization factor was included in the optical-model calculation
and was allowed to vary during the search procedure. The normalization of the 12 MeV
data for 60Ni had to be changed by 10%. For all other measurements the change was less
than 3%.
The experimental angular distributions are shown in Figures 23.1-23.3 and the
excitation functions in Figures 23.4-23.7. They are compared with the optical-model
calculations discussed in the next section. Where no error bars are shown the
uncertainty is smaller than the size of the experimental points.
The optical-model analysis

We have carried out optical model analysis using nuclear potential, which included
not only the usual central part but also spin-orbit and tensor interactions. The
parameterization of the optical model potential has been described earlier but it is
convenient to list explicitly the components used in our analysis of 60Ni and 90Zr data.
The components of the optical model potential incorporate the form factor f (r , ri ,a i ) ,
which is defined as:
1
f (r , ri , ai ) =
1 + e xi
where
r − ri A1 / 3
xi =
ai
Parameters ri and ai , together with the depths of the potentials (see below) define
completely each component of nuclear interaction.
The nuclear potential used in our calculations had the following form:
U ( r ) = U 0 ( r ) + U s .o. ( r ) + U T ( r )
where U 0 (r ) is the central part of the optical model potential, U s.o. (r ) the spin-orbit
part, and U T (r ) the tensor part.
d
U 0 (r ) = −Vf (r , r0 , a0 ) − i 4a0′WD f (r , r0′, a0′ )
dr
1⎡ df (r , rs.o. , as.o. ) df (r , rs′.o. , a′s.o. ) ⎤
U s.o. (r ) = D 2π ⎢Vs.o. + iWs.o. ⎥⎦S ⋅ L
r⎣ dr dr
where D 2π is the square of the pion Compton wavelength,
2
⎛ h ⎞
D π = ⎜⎜
2
⎟⎟ = 2 fm 2
⎝ π ⎠
m c
The tensor interaction can be constructed using the following three terms:
2
TR = (S ⋅ r ) 2 −
3
1 2r
TL = (L ⋅ S) 2 + (L ⋅ S) − L2
2 3
2
TP = (S ⋅ r ) 2 − p 2
3
where p is the relative momentum.
Earlier studies (Goddard 1977; Keaton and Armstrong 1973; Stamp 1970) suggested
that the important tensor interaction is represented by the TR term. We have
therefore used only this term in our calculations. The form of the tensor part of our
optical model potential was as suggested by Keaton and Armstrong (1973):
⎧⎪ d ⎡ 1 df (r , rTR , aTR ) ⎤ d ⎡ 1 d ⎛ 4e xTR′ ⎞⎤ ⎫⎪
U T (r ) = −D 2π r ⎨VTR ⎢ + iW ⎢ ⎜ ⎟⎥ ⎬TR
⎥⎦ dr ⎣ r dr ⎜⎝ (1 + e xTR′ ) 2 ⎟⎠⎦ ⎪⎭
TR
⎪⎩ dr ⎣ r dr
where
′ A1 / 3
r − rTR
′ =
xTR
′
aTR
It is convenient to summarize the full complement of the 18 parameters describing
the nuclear interaction as used in our calculations. This summary is presented in
Table 23.1
Table 23.1
Summary of the parameters describing nuclear interaction used in the optical model analysis of the
elastic scattering of polarized deuterons from 60Ni and 90Zr nuclei
Real component Imaginary component

Central potential V , r0 , a0 WD , r0′, a0′
Spin-orbit potential Vs.o. , rs.o. , as.o. Ws.o. , rs′.o. , a′s.o.
Tensor potential VTR , rTR . , aTR ′ . , aTR
WTR , rTR ′
Figure 23.1. Angular distributions of the differential cross sections (left-hand side) and vector
analyzing powers, iT11 , (right-hand side) for the elastic scattering of deuterons from 60Ni and 90Zr
nuclei. Experimental results (points) are compared with the optical-model calculations (solid lines)
generated by the potential parameters listed in Tables 23.2 and 23.3. If not shown, the error bars are
not larger than the experimental points.
Figure 23.2. Angular distributions of the tensor analyzing powers T20 , T21 for the elastic scattering of
60 90
polarized deuterons from Ni and Zr nuclei. See the caption to Figure 23.1.
Figure 23.3. Angular distributions of the tensor analyzing powers T22 for the elastic scattering of
polarized deuterons from 60Ni and 90Zr nuclei. See the caption to Figure 23.1.
r
Figure 23.4. Energy dependence of the differential cross sections for the elastic (d , d ) scattering from
60
Ni measured at 175° are compared with the optical-model calculations. Theoretical curves
correspond to parameters listed in Tables 23.2 and 23.3.
r
Figure 23.5. Energy dependence of the tensor analyzing power T20 for the elastic (d , d ) scattering
from 60Ni. See the caption to Figure 23.4.
r
Figure 23.6. Energy dependence of the differential cross sections for the elastic (d , d ) scattering
from 90Zr measured at 175° are compared with the optical-model calculations. Theoretical curves
r
Figure 23.7. Energy dependence of the tensor analyzing power T20 for the elastic (d , d ) scattering
from 90Zr measured at 175° are compared with the optical-model calculations. Theoretical curves
Table 23.2
Fixed parameters and parameters described by mass-dependent formulae as used in the optical
model analysis of the elastic scattering of polarized deuterons from 60Ni and 90Ze nuclei
Real Component Imaginary Component
Central V * WD *
r0 1.14 r0′ 1.69 − 0.07 A1 / 3

1.20 + 0.85 A−1 / 3
a0 0.8 a0′ 0.20 + 0.14 A1 / 3
1.29 − 2.10 A−1 / 3
Spin-orbit Vs.o. * Ws.o. 2.0
rs.o. * rs′.o. 0.9
as.o. * a′s.o. 0.55
Tensor VTR * WTR *
rTR 1.89 − 0.075 A1 / 3 ′

rTR *
1.29 + 1.16 A−1 / 3

aTR * ′
aTR *
The asterik (*) refers to parameters that could be neither fixed nor described by mass-
dependent formulae. These parameters are listed in Table 23.3. The two alternative mass-
dependent formulae for r0′ , a0′ , and rTR give almost identical values for these parameters
(see Table 23.4).
Table 23.3
Individually adjusted optical model parameters used in the optical model analysis of the elastic
scattering of polarized deuterons from 60Ni and 90Ze nuclei
The final sets of parameters, which yield the best fits to the experimentally measured
differential cross section and the analyzing powers are listed in Tables 23.2 and 23.3.
Parameters which could be fixed or which could be described by simple analytic
formulae are listed in Table 23.2. All the remaining parameters, which had to be
individually adjusted to optimise the fits at the relevant energies and for given target
nuclei are shown in Table 23.3
Results of our optical model calculations are compared with the experimental angular
distributions in Figures 23.1-23.3. As can be seen, there is a generally excellent
agreement between the theoretical calculations and the experimental results.
Calculations for the excitations functions are shown in Figures 23.4-23.7. Here, the
agreement is less satisfactory but in general the main features are reproduced.
The imaginary spin-orbit interaction has a strong influence on the calculations. By
introducing this component, we were able to reduce the χ2 values by an average of
about 40%, mainly by improving the fits to the differential cross sections, which was both
unexpected and surprising. The only observed quantity with an equal or worse χ2 value is
T21 , which is expected to be sensitive mainly to tensor forces, but less to the central or
spin-orbit potentials (Hooton and Johnson 1971).
Table 23.4
Parameters calculated using two alternative mass-dependent formulae listed in Table 23.2
60 90
Ni Zr
r0′ = 1.69 − 0.07 A1 / 3 1.42 1.38
r0′ = 1.20 + 0.85 A−1 / 3 1.42 1.39
a0′ = 0.20 + 0.14 A1 / 3 0.75 0.83
a0′ = 1.29 − 2.10 A−1 / 3 0.75 0.82
rTR = 1.89 − 0.075 A1 / 3 1.60 1.55
rTR = 1.29 + 1.16 A−1 / 3 1.59 1.55
In the present analysis it has been possible to reduce the number of the free parameters
from 18 to 10. The resulting fixed or mass-dependent parameters are listed in Table 23.2.
This table contains five energy- and mass-independent parameters and three mass-
dependent parameters. The two alterative formulae for r0′ , a0′ , and rTR give almost
the same values for these parameters (see Table 23.4)
Our mass-dependent formula for r0′ may be compared with the formulae derived by
Perrin et al. (1977) and Griffith et al. (1977):
r0′ = 1.20 + 0.85 A−1 / 3 (our formula)
r0′ = 1.15 + 0.75 A−1 / 3 (Perrin et al. 1977)
r0′ = 1.25 + 0.85 A−1 / 3 − 0.004 E (Griffith et al. 1977)

The formula of Griffith et al. (1977) shows negligible energy-dependence, in
agreement with our results.
Both the present and previous analyses show that the diffuseness a0′ of the imaginary
central potential is energy independent. Initially this parameter was allowed to vary.
However, when the convergence was achieved, it was found that the resulting values
could be described by a simple linear relation expressed in terms of A1 / 3 or A−1 / 3 .
It is worth mentioning that this parameter influences mainly the normalization of the
differential cross section. This correlation has been discussed Dickens and Perey (1965).
The radius rTR of the real component of the tensor potential decreases slightly with the
increasing energy. The fits, however, have been found to be rather insensitive to this
parameter. The mass-dependent but energy-independent formula for this parameter
gives good representation of this parameter.
The remaining ten parameters were allowed to vary during the final search. As the
fits to the experimental data are sensitive to small changes in the depth V of the central
potential, this parameter was never kept fixed. However, the depth V can be described
by an analytic expression, which gives an excellent overall description of its mass
dependence:
V = 95 + 1.4ZA−1 / 3 − 0.8Ec.m.
where Ec.m. is the centre-of-mass energy in MeV. This relation is in agreement with the
formulae found by Perrin et al. (1977) and Griffith et al. (1977)
The other nine parameters depend on the mass of the target nucleus and on the
incident deuteron energy. Some systematic behaviour can be noticed for the depth
and the radius of the imaginary tensor potential. They both decrease with the
increasing energy. The depth of the imaginary central potential also increases slightly
with the increasing energy but its value is around 13 MeV in agreement with the results of
Perrin et al. (1977).
Summary and conclusion

We have carried out precise measurements of the angular distributions for the five
observables: σ 0 (θ ) , iT11 (θ ) , T20 (θ ) , T21 (θ ) and T22 (θ ) . Our energy range was 9-15
MeV and angular range 400-1600 (c.m.) We have also measured excitations
functions for the differential cross sections,σ0 , and the tensor analyzing power, T20 .
We have carried out optical model analysis of our experimental results using a
six-component optical model potential made of the central part with the surface
absorption; the spin orbit part containing both real and imaginary components;
and the tensor TR part also containing both real and imaginary components. The
potential was described by a total of 18 components. However, we have found
that five of them ( r 0 , a0 , Ws.o. , rs′.o. and a′s.o. ) could be fixed for the two target nuclei
and for the incident deuterons energies. Three additional parameters r0′ , a0′ and
rTR could be described by mass-dependent formulae. We have found two
alternative but equivalent formulae for each of these three parameters
depending either on A1 / 3 or A−1 / 3 .
Our spin-orbit parameters of the imaginary component are similar to those used
by Goddard and Haeberli (1978). These parameters have a significant effect on
the quality of fits to the experimental distributions. Thus we have confirmed the
earlier finding (Goddard and Haeberli 1978) that the imaginary spin-orbit
component plays a significant role in optical model analyses of experimental
data.
Measured tensor analyzing powers are reproduced very accurately by including a
complex tensor TR potential. Its parameters, however, could not be determined very
precisely.
Fits to the T22 tensor analyzing powers get worse with the increasing energy. It has
been suggested earlier that theoretical angular distributions of T22 depend strongly on
the central and the spin-orbit potentials (Hooton and Johnson 1971; Johnson 1977).
Consequently, we suggest that these potentials may have some additional or an
alternative structure. A possible way of improving the fits to the T22 distributions would be
to use different shapes for these potentials or to introduce an l - dependent potential
(Rawitscher 1977).
In general, we have obtained excellent fits to the angular distributions. Our study has
resulted in important information not only about the central nuclear interaction but
also about spin dependent potentials, including tensor interaction.
References
Dickensm J. K. and Perey, F. G. 1965, Phys. Rev. B138:1080
Hooton, D. J. and Johnson, R. C. 1971, Nucl. Phys. A175:583.
Goddard, R. P. 1977, Nucl. Phys. A291:13.
Goddard, R. P. and Haeberli, W. 1978, Phys. Rev. Lett. 40:701.
Griffith, J. A. R., Irshad, M., Karban, O. and Roman, S. 1970, Nucl. Phys.
A146:193.
Johnson, R. C. 1977, Nucl. Phys. A293:92.
Keaton, Jr., P. W. and Armstrong, D. D. 1973, Phys. Rev. C8:1692.
Perrin, G., Nguyen van Sen, Arvieux, J., Darves-Blanc, R., Durand, J. L., Fiore, A.,
Gondra, J. C., Merchez, F. and Perrin, C. 1977, Nucl. Phys. A282:221.
Rawitscher, G. H. 1977, Bull. Am. Phys. Soc. 22:597.
Stamp, A. P. 1970, Nucl. Phys. A159:399.
24
Global Analysis of Deuteron-nucleus Interaction
Key features:
1. This study represents the first and the most extensive study of the deuteron-nucleus
interaction. It includes not only the differential cross sections σ 0 but also the
analyzing powers, iT11 , T20 , T21 and T22 measured and analysed for a wide range of
target nuclei, A = 40 – 90.
2. Experimental results have been analysed using the optical model containing not only the usual
complex central part but also complex spin-orbit and tensor components.
3. In general, we have obtained excellent fits to the experimental data.
7. We have found global description for all 18 optical model parameters. All optical model
parameters can be represented either by fixed values or the values calculated using simple
mass-dependent formulae. Such global description is useful in analyses of transfer reaction
data.
4. We have also found that both components of the central potential depend on the gamma
transition probabilities.
Abstract: Angular distributions of the differential cross-sections σ 0 (θ ) and analyzing powers

iT11 (θ ) , T20 (θ ) , T21 (θ ) and T22 (θ ) for the elastic scattering of 12 MeV polarized deuterons
were measured in the angular range of 20°-175° (lab) using a wide range of spin-zero target
nuclei with mass numbers A = 40 – 90. The data were analysed using optical model potential
with complex central, spin-orbit and tensor terms. With a few exceptions, excellent fits have
been obtained to all measured angular distributions, yielding a set of global optical model
parameters. The depths of the central part of the optical potentials have been found to depend
on the structure of the target nuclei. Contrary to the results for selenium, which exhibit clear
shell-closure effects, the data for N = 28 nuclei do not exhibit any clear shell-closure
correlation. This feature is attributed to interactions with higher configurations in this region.
Irregularities in the optical-model parameters and problems in fitting the experimental results
are discussed. Possible ways of improving theoretical description is to include coupling
between elastic and reaction channels.
1. Introduction
Encouraged by our successful analysis of experimental data for the 60Ni and 90Zr nuclei
(see Chapter 23) we have extended our study of spin-dependent forces to other target
nuclei. To support this study we have carried out measurements of the angular
distributions of the differential cross sections σ 0 (θ ) and of all four analyzing powers
iT11 (θ ) , T20 (θ ) , T21 (θ ) and T22 (θ ) for 46, 48, 50Ti, 52, 54Cr, 54, 56Fe and 58Ni isotopes. In our
global analysis of the data we have also included the previously measured distributions for
60
Ni, 90Zr and 76,78,80,82Se isotopes.
Experiment
The experimental method and procedure have been already fully described in
Chapters 15 and 17. The targets in the present measurements were in the form of
isotopically enriched self-supporting foils (see Table 24.1). The 46Ti and 50Ti targets
contained substantial admixtures of 48Ti. Scaled contributions from this isotope, known
from independent measurements, were subtracted from the measured values.
Some targets were found to have small (0.2%-0.7%) additional impurities of heavy
elements such as W, Pt or Hg. These impurities were determined quantitatively by
an analysis of proton induced X-rays (Bonani et al. 1978). Corrections due to the
heavy-element impurities are only necessary at forward angles where the elastic
peaks for the medium-heavy and the heavy nuclei could not be resolved. For the
analyzing powers, the contributions from heavy elements are insignificant in this region.
Table 24.1
Isotopes used in our study
The absolute normalization of the differential cross section was determined from
measurements of the Rutherford scattering of low-energy deuterons at forward
angles. Corrections for the impurities in the target materials have been taken into
account in the evaluation of the absolute values of the cross sections. The un-
certainty in the absolute normalization varies in the range of 5-10% for various targets.
For the analyzing powers the statistical error was kept below 0.005. The uncertainties
due to the inaccuracy in the scattering angles, the final geometry and the absolute values
of the beam polarization have also been included. All these contributions, however,
were small.
The experimental results together with optical-model calculations are shown in
Figures 24.1 to 24-5.
Figure 24.1 shows nuclei with the same number of protons (Z = 22) but different number
of neutrons (N = 24, 26 and 28). In the simple shell model description, two protons are in
the 1f7/2 orbit and neutrons are filling in the 1f7/2 sub-shell.
Figure 24.2 shows nuclei with the same number of neutrons (N = 28, which fill in the sub-
shell 1f7/2) but with different number of protons (Z = 22, 24 and 26) in the 1f7/2 orbits.
Figure 24.3 shows nuclei with the same number of neutrons (N = 30, which fill in the1f7/2
sub-shell and with two neutrons in the 2p3/2 orbit) and with different number of protons (Z
= 24, 26 and 28) in the 1f7/2 orbits.
Finally, Figures 24.4 and 24.5 shows results for selenium isotopes from our previous
study (see Chapter 17). These isotopes contain a fixed number of protons (Z = 34, with
the last six being outside the closed 1f7/2 sub-shell, i.e. in configurations 2p3/2 and 1f5/2)
and with different number of neutrons (N = 42, 44, 46 and 48) filling in the 1g9/2 sub-shell.
The selenium measurements are the only results that show clearly the influence of
neutron shell-closure in the form of an enhancement of the oscillations of the analyzing
powers for nuclei approaching neutron number N = 50 (see Chapter 17). This is contrary
to the usual mass dependence in which the oscillations decrease in amplitude with the
increasing mass of the target nuclei. As discussed in Chapter 17, the observed
amplitude enhancement effect for these isotopes has been explained as being due to
contributions from two-step reaction mechanism.
The oscillations for the titanium isotopes have approximately equal amplitudes. The
same is true for nuclei with neutron number N = 30. For nuclei with N = 28, the measured
amplitudes clearly decrease with the increasing mass number. Thus shell-closure
effects are much less pronounced in the Z = N = 28 region than for the N = 50 shell.
Figure 24.1. The differential cross section and analyzing powers for the elastic scattering of 12 MeV
polarized deuterons from the 46,48,59Ti isotopes at 12 MeV. These nuclei contain the same number of
protons (Z = 22) but different numbers of neutrons (N = 24, 26, and 28). In the simple shell model
description, two protons are outside the sd-shell and neutrons are filling in the 1f7/2 sub-shell. The full
lines are the results of our optical-model calculations.
Figure 24.2. The differential cross section and analyzing powers for the elastic scattering of 12 MeV
polarized deuterons from the 50Ti, 52Cr, and 54Fe isotopes containing a fixed number of neutrons N =
28, which are closing the 1f7/2 sub-shell, but different numbers of protons (Z = 22, 24 and 26), which
are filling in the 1f7/2 orbits. The full lines are the results of our optical-model calculations.
Figure 24.3. The differential cross section and analyzing powers for elastic scattering of 12 MeV
54 56 58
polarized deuterons from the Cr, Fe, and Ni isotopes containing a fixed number of neutrons (N =
30, with the filled-in 1f7/2 sub-shell and with two neutrons in the 2p3/2 orbit) and different number of
protons (Z = 24, 26 and 28) in the 1f7/2 orbits. The full lines are our optical-model calculations.
Figure 24.4. Differential cross section for the elastic scattering of 12 MeV polarized deuterons from
the 76,78,80,82Se isotopes. These isotopes contain a fixed number of protons (Z = 34, with the last six
being outside the closed 1f7/2 sub-shell, i.e. in configurations 2p3/2 and 1f5/2) and with different number
of neutrons (N = 42, 44, 46 and 48) filling in the 1g9/2 sub-shell. The full lines are our optical-model
calculations.
Figure 24.5. The vector and tensor analyzing powers for the elastic scattering of 12 MeV polarized
deuterons from selenium isotopes. See the caption to Figure 24.4
The lack of shell-closure effects in all target nuclei but selenium isotopes could be
due to an interplay of various effects such as screening by the Coulomb potential,
variation in the nuclear shapes and irregularities in the filling in of the shells. A recent
systematic study (England et al. 1982) of 25 MeV α-particle scattering from A = 51-80
nuclei revealed a breakdown in the shell-closure for 54Fe. Studies of the neutron pick-
up reactions, (p,d) (Suehiro, Finck and Nolen 1979) and (d,t) (England et al. 1980)
confirmed this result indicating the presence of p3/2 configuration in the ground-state
wave function of 54Fe. (The 54Fe nucleus has 28 neutrons and it should have a
closed 1f7/2 shell.) It is likely that similar irregularities in filling in the 1f7/2 sub-shell occur
for other neighbouring nuclei. This could explain why no shell closing effects are
observed for nuclei other than selenium isotopes.
Figure 24.6. The mass dependence of the positions of the measured maxima and minima of the
vector analyzing power iT11 at 12 MeV. The dashed lines are to guide the eye.
In order to see whether the closure of shells can modify the relative phases of the
analyzing powers, the positions of the maxima and minima for the iT11 component
have been plotted against the mass number A in Figure 24.6. It can be seen that even
for selenium isotopes, the positions of the diffraction patterns follow a linear
dependence on A indicating that the phases of the analyzing powers are not affected
by the shell closure.
Theoretical interpretation of the data

Details of the optical-model analysis have been described in Chapter 23. Briefly, the
potential contained complex central, spin-orbit, and tensor parts. There are 18
parameters defining the nuclear potential (see Figures 24.7 and 24.8 and Table
24.1) but only 10 of these were varied to fit the data for each isotope. The remaining
8 parameters were either fixed or adjusted using simple mass-dependent formulae.
Figure 24.7. Optical-model parameters for the central potential. Where no errors bars are shown the
uncertainties in the parameters are smaller than the size of the points. The lines show the mass-
dependent trends. The crosses are the calculated values using the dependence on the B(E 2)
gamma transition probabilities (see the text).
Figure 24.8. Optical-model parameters for the spin-dependent components (spin-orbit on the left-
hand side and tensor on the right-hand side of the figure). The lines show the mass-dependent
trends.
Table 24.1
The global parameters of the optical model potential for the elastic scattering of 12 MeV polarized
deuterons in the mass range A = 40 - 90
Real Component Imaginary Component
Central V V = 84.57 + 0.13 A WD WD = 9.91 + 0.067 A

V = 68.43 + 6.27 A1 / 3 WD = 1.72 + 3.182 A1 / 3
r0 1.14 r0′ 1.51 − 0.002 A
1.69 − 0.072 A1 / 3
a0 0.80 a0′ 0.57 + 0.003 A
0.20 + 0.137 A1 / 3
Spin-orbit Vs.o. Vs.o. = 7.91 − 0.045 A Ws.o. 2.0
Vs.o. = 13.35 − 2.12 A1 / 3

rs.o. 0.75 rs′.o. 0.90
as.o. 0.40 a′s.o. 0.55
Tensor VTR 0.89 WTR 1.45
rTR . 1.85 − 0.003 A ′

rTR 1.06
1.89 − 0.075 A 1/ 3
aTR 0.26 ′
aTR 0.31
Additional formulae: W (r0′) a′ = 12.693 + 0.137 A and W (r0′) a′ = 3.995 + 6.483 A1 / 3 .

2 2
The best agreement between the theory and experiment has been found for the selenium
isotopes. However, it should be noted that for these nuclei the parameters do not
follow the same pattern as those obtained for other target nuclei. The results for 50Ti,
52
Cr and 54Fe corresponding to N = 28 were hard to reproduce. Similar difficulties have
been observed by Goddard and Haeberli (1978) at other deuteron energies.
In Chapter 23, it was shown that out of the eighteen parameters describing the
nuclear potential five ( r 0 , a0 , Ws.o. , rs′.o. and a′s.o. ) could be fixed and three ( r0′ , a0′ and
rTR ) could be expressed in terms of simple mass-dependent formulae. This feature
has been confirmed in the present analysis of the results for a much wider range of
target nuclei. However, we have now also found that all 18 parameters can be
represented by either fixed values or the values calculated using simple mass-
dependent formulae. This global form of parameters are summarised in Table 24.1.
Such global expressions are useful in analyses of transfer reaction data.
Hjorth, Lin and Johnson (1968) and Lohr and Haeberli (1974) suggested that the
imaginary part of the central optical model potential is related to the
B ( E 2) ≡ B( E 2,01+ → 21+ ) γ - transition probabilities. Taking the B(E 2) values (expressed in
fm4) from Stelson and Grodzins (1965) and from Endt and Van der Leun (1978) we have
found that the parameters for the imaginary components of the central potential can
be described closely by the following relation:
WD (r0′) 2 a0′ = 10.9 + 0.56 ZA−1 / 3 + 10.5[B ( E 2)] A−1 (in MeV·fm3)
1/ 2
A similar dependence has been obtained by Hjorth, Lin and Johnson (1968). However,
their coefficient (417) for the [B( E 2)] A−1 term appears to be incorrect. Examination of
1/ 2
their results and their fig. 4 indicates that this coefficient should have a value close to 13,
which would agree well with our results.
The values of WD (r0′) 2 a0′ extracted from the above formula are represented by crosses
in Figure 24.7. As can be seen, these values follow closely the corresponding values
obtained from the optical-model analysis.
During our investigation, it became clear that the depth of the real part of the central
potential, V, is correlated with the depth WD in the sense that large values of WD are
associated with small values of V and vice-versa. Figure 24.7 shows the dependence of V
on A1/3 and as can be seen, the values obtained from the optical-model analyses (dots)
show departures from the straight line. An attempt has been made to reproduce these
changes using a suitable analytic form for V. If the dependence on B(E 2) is ignored,
then the trend is expressed in the form of the dashed line in Figure 24.7. However, if the
dependence on B(E 2) and ZA-1/3 are included explicitly, then the least square analysis
leads to the following relation for V:
V = 92.5 + 2.1ZA−1 / 3 − 1.0 Ec.m. − 6.7[B ( E 2)] A−1
1/ 2
(in MeV)
The results of this relation are shown as crosses in Figure 24.7. This formula provides
a good description of the fluctuations in the values of V for various targets except for
40
Ar where calculated V value is much too small.
The above relations demonstrate clearly a strong correlation between potential depths V
and WD . Using these two relations it is easy to see that the correlation between V and
WD in the mass range of A = 40 - 90 can be expressed conveniently by the following
approximate relation:
V + WD = 100 + 2.5ZA−1 / 3 − Ec.m. (in MeV)
The parameter values for the spin-orbit and the tensor potentials are shown in Figure
24.8. Some parameter values are found to fluctuate with mass number of the target
nucleus. The dashed lines indicate the trends for these parameters. It is possible that the
observed fluctuations arise partly from an attempt to compensate for shape effects which
are inadequately described by the form factors for the central potential.

The work described here represents the most extensive and systematic study of the
elastic scattering of polarized deuterons from medium-heavy target nuclei. This study
demonstrates that suitable sets of optical-model parameters can be found which
describe accurately all five quantities σ 0 , iT11 , T20 , T21 and T22 measured for nuclei in
the mass range of A = 40 – 90.
Of the eighteen parameters used to define nuclear interaction, five were kept fixed during
the search and three were calculated using simple mass-dependent formulae. However,
we have also found that all 18 parameters can be conveniently represented either using
fixed values of the values calculated using simple mass-dependent formulae. Such global
representation of the optical model parameters is particularly useful in analyses of transfer
reaction measurements. It simplifies the process of selecting suitable parameters, which
need to be used in such studies.
Both, the depths V and WD of the central potential were found to be strongly
dependent on the structure of the target nuclei. These parameters can be expressed
conveniently in terms of the B ( E 2,01+ → 21+ ) γ - transition probabilities. We have also
derived a mathematical correlation between V and WD.
We have found that it was difficult to describe the results for nuclei with N or Z near 28. In
particular, the fits for the N = 28 targets, 50Ti, 52Cr and 54Fe, are poorer than for other
investigated nuclei. A similar problem with 52Cr and 54Fe targets has been reported by
Goddard and Haeberli (1978). There is now sufficient evidence indicating that the 1f7/2
neutron shell is not closed for 54Fe (Ν = 28). Thus the poor theoretical description of
the data may be associated with structure irregularities in this mass region, and a
better fit might be obtained by including structure effects, e.g. two-step processes,
explicitly in the calculation as suggested by my analysis of selenium data (see
Chapter 17).
It is possible that the fits could be improved by using different form factors for the central
potential, other than the conventional Wood-Soxon and its derivative. The inclusion
of l - dependent potentials could be also considered. However, additional parameters in
already parameter-rich descriptions appears undesirable. It has been also shown (Kobos
and Mackintosh 1979) that introducing an l - dependent component is equivalent to the
imitation of a coupling between elastic and reaction channels. Such a coupling could be
considered explicitly in the theoretical analysis as described in the next chapter.
References
Bonani, G., Stoller, Ch., Stöckli, M., Suter, M. and Wölfli, W. 1978, Helv. Phys. Acta
51:272.
Endt and Van, P. M. der Leun, C. 1978, Nucl. Phys. A310
England, J.B.A., Baird, S., Newton, D.H., Picazo, T., Pollacco, E.C., Pyle, G.J., Rolph,
P.M., Alabau, J., Casal, E., Garcia, A.1982, Nucl.Phys. A388:573.
England, J. B. A., Ophel, T. R., Johnston, A. and Zeller, A. F. 1980, J.Phys.
G6:1553.
Goddard, R. P. and Haeberli, W. 1978, Phys. Rev. Lett. 40:701.
Hjorth, S. A., Lin, E. K. and Johnson, A. 1968, Nucl. Phys. A116:1.
Kobos, A. M. and Mackintosh, R. S. 1979,J. Phys. G5:97.
Lohr, J. M. and Haeberli, W. 1974, Nucl. Phys. A232:381.
Stelson, P. H. and Grodzins, L. 1965, Nucl. Data A1:21.
Suehiro, T., Finck, J. E. and Nolen, Jr., J. A. 1979, Nucl. Phys. A313:141.
25
Collective Excitation Effects in the Elastic Scattering
Key features:
1. Conventional optical model analyses consider only the single-step (d , d ) elastic
scattering. This has resulted in significant complications when describing the
interaction potential. For instance, in our analysis (see Chapter 24) we had to use
nine components of the optical model potential with the total of 18 parameters.
2. Following my successful study of selenium isotopes (see Chapter 17), I have decided
to extend my coupled-channels calculations to a wider range of nuclei by including
explicitly the two-step (d , d ′) 21 ( d ′, d ) contributions to the elastic scattering. This
+
procedure has now simplified considerably the description of the deuteron-nucleus

interaction.
a. The interaction potential can now be described using only three components
with the total of only 9 parameters.
b. The imaginary spin-orbit component, which has been found essential in our
conventional optical model analysis, is no longer required.
c. Five of the nine parameters describing the deuteron-nucleus interaction can
now be either fixed for all target nuclei or described using simple mass-
dependent formulae, leaving only four parameters that need to be adjusted to
optimise the fits to the experimental angular distributions.
+
d. The dependence of the potential depths on the B ( E 2;01 → 21 ) γ - transition
2
probabilities, which has been found repeatedly in various conventional optical

model analyses, is now eliminated. This dependence, therefore, reflects the
presence of the second-order processes, which normally are not included in
analyses of elastic scattering.
Abstract: The differential cross sections, σ 0 (θ ) and vector analyzing powers, iT11 (θ ) for 12
MeV vector polarized deuterons scattered elastically from 40Ar, 46,48.50Ti, 52,54Cr, 54,56Fe, 58,60Ni,
76,78,80,82
Se and 90Zr were analysed using the coupled-channels formalism, which included the
+
two-step scattering via the first 21 states in the target nuclei. In contrast with the results of the
conventional optical model analysis for the same isotopes, there was now no need to include
the imaginary spin-orbit component in the description of the interaction potential. The
parameterization of the optical model potential has been significantly simplifies and the
dependence on the γ -transition probabilities, B ( E 2;01+ → 212 ) , has been removed.
Introduction
As described in Chapters 23 and 24, in our analysis of polarization data we had to use the
spin-orbit potential with an imaginary component. However, this component influenced
also significantly the parameter values of the central potential. We have suggested, that
the source of this unexpected effect might be due to the generally adopted mathematical
description of the shape of the nuclear potential using the Woods-Saxon function. Such
representation might not be sufficiently accurate. However, there is also another
alternative: the problem might be associated with the assumed simplified reaction
mechanism for the elastic scattering. The mechanism might be more complex than the
simple shape scattering used in conventional optical-model calculations.
Indeed, in Chapter 17 I have shown that the two-step scattering via the 21+ inelastic
scattering channel plays an important role in the elastic scattering. Without considering
this process, the depth of the imaginary component of the central potential has to be
varied to fit the data, which is a crude way of accounting for contributions from indirect
scattering. I have shown that without coupling to the first excited states, the depth WD of
the imaginary component of the optical model potential depends linearly on the
quadrupole deformation parameters β2 derived from the γ -transition probabilities,
B ( E 2;01+ → 212 ) . However, if the two-step scattering via the inelastic channels is
considered explicitly in the calculations the need for adjusting the potential depth WD is
removed and the data can be fitted using a fixed set of the optical model parameters for
all four selenium isotopes.
In an earlier work, Rawitcher (1978) considered a folding model with non-symmetric
break-up processes, which suggested an l - dependent potentials. Unfortunately, this
kind of approach does not seem practical unless the l - dependence can be
predicted by theoretical considerations. The number of adjustable parameters, which
have to be used to describe the elastic scattering of deuterons, is already too large
and to increase their number is undesirable. Furthermore, in the case of proton
scattering, it has been pointed out (Kobos and Mackintosh 1979) that the physical
meaning of the l - dependent potentials is simply to simulate the coupling to reaction
channels. Consequently, straightforward coupled-channel calculations would appear
more appropriate. My successful coupled-channels analysis of selenium data
(Chapter 17) supports this approach.
The observed dependence of the optical model potentials on the electric quadrupole
transition probabilities B ( E 2) = B ( E 2;01+ → 212 ) or equivalently on the quadrupole
deformation parameter β 2 also suggests that it would be better to include explicitly
the coupling to inelastic channels in analyses of elastic scattering. The B(E 2)
dependence becomes particularly clear for measurements carried out using
polarized beams.
A correlation between the B(E 2) values and the parameters of the central imaginary
potential for deuterons was first suggested by Hjorth, Lin, and Johnson (1968). They
measured the differential cross sections for the elastic scattering of 14.5 MeV
deuterons from isotopically enriched targets in the mass range of A = 54 - 124 and
found that the product WD (r0′) 2 a0′ (see the next section) depends linearly on
[ B( E 2)]1 / 2 / A .
Lohr and Haeberli (1974) carried out measurements of angular distributions of the
cross sections and vector analyzing powers for the elastic scattering of vector
polarized deuterons. They had found that the volume integral of the imaginary central
potential followed a reasonably clear linear dependence on the quadrupole
deformation parameter β 2 for deuterons in the energy range of 7-13 MeV and over
the target mass range of 27 - 208. A systematic study of proton elastic scattering on a
wide range of nuclei also indicated a linear dependence of the depth of the surface
absorption potential on β 2 (Fabrici et al. 1980).
Our study in the mass range of A = 40 - 90 suggested that not only the imaginary but
also the real central potential component depend on B(E 2) (see Chapter 24). Thus,
all these results appear to suggest a detectable influence of the inelastic, 21+ ,
channel on the elastic scattering.
It is reasonable to expect that the dependence of the optical model parameters on
the quadrupole properties of the target nuclei could be accounted for by coupling
between the elastic and inelastic channels. Consequently, by following the same
procedure as outlined in Chapter 17 for selenium isotopes it should be possible to
reduce or even to remove the B(E 2) dependence. My aim therefore was to extend
my investigation of the two-step mechanism to a wider range of nuclei and hopefully
to simplify the parameterization of the interaction potential.
The coupled-channels analysis

The differential cross sections, dσ (θ ) / dΩ and vector analyzing powers, iT11 (θ ) for
the elastic scattering of 12 MeV vector polarized deuterons from 40Ar, 46,48.50Ti,
52,54
Cr, 54,56Fe, 58,60Ni, 76,78,80,82Se and 90Zr nuclei have been analysed by considering
both the direct shape scattering (d , d ) and the indirect (d , d ′)21+ (d ′, d ) scattering via
the first excited states 21+ in the target nuclei (see Chapter 17). The analysis was done
using the computer code CHUCK and the Australian National University UNIVAC
1100/82 computer.
Figure 25.1. The experimental angular distributions of the differential cross sections for the elastic
scattering of 12 MeV vector polarized deuterons are compared with the coupled channels
calculations, which include explicitly contributions of the two-step (d , d ′)21+ (d ′, d ) mechanism. The
full lines were calculated using five fixed parameters and searching for the remaining four (V, WD, r0′
and a0′ ). The dotted lines represent the calculations in which only two parameters, V and WD, were
varied to optimise the fits to the angular distributions.
Figure 25.2. The experimental angular distribution of the vector analyzing powers iT11 (θ ) for the
elastic scattering of 12 MeV vector polarized deuterons are compared with the coupled channels
calculations, which included explicitly contributions of the two-step (d , d ′)21+ (d ′, d ) mechanism. See
In my preliminary analysis I have used both real and imaginary spin-orbit

components. I have found that the imaginary component had no significant effect on
improving the fits to the angular distributions. Consequently, in the remaining
calculations I have used a simpler potential:
U ( r ) = U 0 ( r ) + U s .o . ( r )
where
d
U 0 (r ) = −Vf (r , r0 , a0 ) − i 4a0′WD f (r , r0′, a0′ )
dr
1⎡ df (r , rs.o. , as.o. ) ⎤
U s.o. (r ) = D 2π ⎢Vs.o. ⎥⎦S ⋅ L
r⎣ dr
1
f (r , ri , ai ) =
1 + e xi
r − ri A1 / 3
xi =
ai
Table 25.1
Formulae for the potential depth V
V (MeV) χ2
OM V = 92.5 + 2.1ZA−1 / 3 + 6.7[ B( E 2)]1 / 2 A−1 − Ec.m.
CC4 V = 92.7 + 1.9ZA−1 / 3 + 1.1[ B ( E 2)]1 / 2 A−1 − Ec.m. 0.66
V = 93.2 + 1.9ZA−1 / 3 − Ec.m. 0.67
CC2 V = 89.2 + 2.2ZA−1 / 3 + 2.5[ B( E 2)]1 / 2 A−1 − Ec.m. 0.74
V = 90.4 + 2.2ZA−1 / 3 − Ec.m. 0.76

OM – The formula based on the conventional optical model analysis, which
neglects contributions of the two-step scattering.
CC4 – The two optional formulae based on the coupled-channels analysis,
which includes the two-step scattering. In these calculations, five optical
model parameters were fixed and four were searched for. The two
formulae are obtained by fitting the resulting V values with and without the
B(E2) component. Both functions resulted in the equivalent descriptions of
V (cf the χ2 values).
CC2 – As for CC4 but now coupled-channels analysis was carried out
using 7 fixed parameters and two searched for.
χ2 – The parameter, which describes the quality of the fit to the V values.
The B(E2) values are in e2fm4
Table 25.2
Formulae for WD (r0′) 2 a0′
WD (r0′) 2 a0′ (in MeV•fm3) χ2

OM WD (r0′) 2 a0′ = 10.9 + 0.56ZA−1 / 3 + 10.5[ B ( E 2)]1 / 2 A−1
CC4 WD (r0′) 2 a0′ = 11.0 + 0.82ZA−1 / 3 + 3.1[ B( E 2)]1 / 2 A−1 1.1
WD (r0′) 2 a0′ = 12.5 + 0.86ZA−1 / 3 1.2
CC2 WD (r0′) 2 a0′ = 12.7 + 0.68ZA−1 / 3 + 2.2[ B( E 2)]1 / 2 A−1 1.2
WD (r0′) 2 a0′ = 13.7 + 0.68ZA−1 / 3 1.3
Initially I have searched for all nine parameters. However, I have found that five
parameters follow the values we have determined earlier in the conventional optical
model analysis of the data in this mass region. Keeping these parameters fixed at their
appropriate values, I have repeated the analysis of the data searching on only four
parameters. Finally, I have fixed two additional parameters and search for only two.
In the four parameter search, parameters V, WD, r0′ and a0′ were searched for
while the remaining five parameters, r0 , a0 , Vs.o. , rs.o. and as.o. were fixed at the
following values: r0 =1.14 fm, a0 = 0.8 fm, Vs.o. = 6.3 − 0.4 A−1 / 3 MeV, rs.o. = 0.75 fm and
as.o. = 0.4 fm. Theoretical predictions corresponding to this series of calculations
are shown in the form of the full lines in Figures 25.1 and 25.2. As can be seen,
this simplified potential resulted in excellent fits to the angular distributions. There
was no need to complicate it by adding an imaginary spin-orbit component.
In the two-parameter search, I have kept also the r0′ and a0′ parameters fixed at
the values given by the following mass-dependent formulae: r0′ = 1.20 + 0.85 A−1 / 3
and a0′ = 1.29 − 2.10 A−1 / 3 . This set of calculations was reduced to searching only for
V, and WD. Results are shown as dotted lines in Figures 25.1 and 25.2. As
expected for such severe restrictions, the fits to the data were in some cases less
satisfactory than for the four-parameter search, but surprisingly in many cases
they resulted in nearly equivalent representations of the experimental distributions.
Having determined the new sets of parameters based on the coupled-channels
analysis it was interesting to see how they depended on the electric quadrupole
transition probabilities. Using the determined parameters I have carried out the
least-squares analysis assuming the dependence on ZA−1 / 3 and [B( E 2)] A−1 . The
1/ 2
resulting formulae are listed in Tables 25.1 and 25.2 together with the formulae
derived earlier (see Chapter 24) using conventional optical model analysis, i.e.
without considering contributions from the two-step scattering. As indicated by the
χ2 values, the dependence on B(E 2) can be removed if two-step contributions are
included explicitly in the analysis of experimental results.
Discussion and conclusions

The formulae listed in Tables 25.1 and 25.2 show that the coupled channels
calculations affect essentially only the B (E 2) part of the functions. Consequently, to
see the differences between the conventional optical-model and the coupled-
channels calculations it is convenient to separate the B (E 2) dependence from the
ZA-1/3 dependence and write V and WD (r0′) 2 a0′ functions in the following form:
V = V1 + V2
WD (r0′) 2 a0′ = w1 + w2
where
V1 = a1 + a 2 [ B( E 2)]1 / 2 A−1
V2 = a3 ZA−1 / 3 − Ec.m.
w1 = b1 + b2 [ B( E 2)]1 / 2 A−1
w2 = b3 ZA−1 / 3
Figure 25.3. The B (E 2) - dependent component of the potential depth V. This plot shows that by
including the two-step scattering mechanism (d , d ′) 21 ( d ′, d ) explicitly in the calculations, the
+
dependence on the B (E 2) γ - transition probabilities can be eliminated.
Figure 25.4. The B (E 2) -dependent component of WD (r0′) 2 a0′ . See the caption to Figure 25.5.
The plots of the B(E 2) -dependent components of V and WD (r0′) 2 a0′ , i.e. of V1 and
w1, are shown in Figures 25.3 and 25.4. These plots show clearly that when the
two-step mechanism is included explicitly in the calculations, the dependence of
the optical model parameters on B(E 2) can be removed.
In summary, I have found that if both the direct (d , d ) and two-step (d , d ′)21+ (d ′, d )
scattering are considered in the analysis of experimental data, the potential
describing the deuteron-nucleus interaction can be considerably simplified. The
imaginary component of the spin-orbit interaction, which has been found necessary
in the previous analysis using the conventional optical model procedure (see
Chapter 24), is now no longer required. Even though the tensor analyzing powers
were not included in my calculations, it may be recalled that the main effect of the
imaginary spin orbit component was in improving the fits to the distributions of the
differential cross sections. In the present coupled-channels analysis, excellent fits to
these distributions were obtained without this component. Thus, by including the two-
step scattering mechanism, the deuteron-nucleus interaction can be described using
a simple potential containing only three components and a total of only 9
parameters.
I have found that the potential can be simplified even further by fixing 4 of the 9
parameters ( r0 =1.14 fm, a0 = 0.8 fm, MeV, rs.o. = 0.75 fm and as.o. = 0.4 fm) and by
using a simple mass-dependent formula for one ( Vs.o. = 6.3 − 0.4 A−1 / 3 ). Thus, out of
the total of 9 parameters, 5 can be constrained and only 4 need to be individually
adjusted to optimise the fits to the experimental angular distributions.
I have then constrained two additional parameters ( r0′ = 1.20 + 0.85 A−1 / 3 and
a0′ = 1.29 − 2.10 A ) and searched for only two, V and WD. In many cases, the
−1 / 3
resulting fits to the experimental angular distributions were as good as for the four-
parameter search.
Comparing the present coupled channels analysis with the earlier conventional
optical model calculations I have found that parameters r0 , a0 , Vs.o. , rs.o. and as.o.
have the same values in both cases. Parameters r0′ and a0′ also have similar values.
The essential difference between the two analyses is in the values of the potential
depths V and WD and in particular, in their dependence on B(E 2) .
If the two-step (d , d ′)21+ (d ′, d ) elastic scattering process is included explicitly in the

analysis of experimental data the dependence of V and WD on the B(E 2) γ -
transition probabilities can be eliminated. The interpretation of the
B(E 2) dependence observed in earlier conventional analyses appears now to be
clear: it simply reflects the previously unaccounted for effects of higher order
processes in the elastic scattering, which are mainly due to the two-step scattering
via the first 21+ states in the target nuclei.
References
Fabrici, E., Micheletti, S., Pignanelli, M., Resmini, F. G., De Leo, R., D'Erasmo, G. and
Pantaleo, A. 1980, Phys. Rev. C21:844.
Hjorth, S. A., Lin, E. K. and Johnson, A. 1968, Nucl. Phys. A116:1.
Lohr, J. M. and Haeberli, W., 1974, Nucl. Phys. A232:381.
Kobos, A. M. and Mackintosh, R. S. 1979, J., Phys. G5:97.
Rawitscher, G. H. and Mukherjee, S. N. 1978, Phys. Rev. Lett. 40:1486.
26
97,101
A Study of Ru Nuclei
Key features:
97 101
1. This study resulted in extensive spectroscopic information about the Ru and Ru
isotopes.
2. A total of 38 states have been observed, with 30 of them for the first time. Excitation
energies to these states have been determined with an accuracy of ±7 keV.
3. A total of 30 angular distributions have been measured and analysed using the
distorted wave theory of direct nuclear reactions. Spectroscopic factors and orbital
angular momenta have been determined for all of these states. In addition, spins and
parities were assigned to states with l = 0, 4, and 5.
4. The ground state Q0 - value for the 96Ru(d,p)97Ru reaction have been determined for the
first time with high accuracy of ±0.003 MeV. The determined value is 5.886±0.003 MeV
Abstract: The neutron single-particle strength distributions for the nuclei 97Ru and 101Ru have
been investigated using the (d,p) reaction at deuteron energy of 11.5 MeV, with an overall
experimental resolution of approximately 25 keV. Angular distributions of proton groups
leading to sixteen final states in both nuclei were measured in the angular range of 15.00 to
67.50. The measured cross sections are analysed in the framework of the distorted waves Born
approximation to deduce the l - values and spectroscopic factors of the states in the residual
96 97
nuclei. The ground-state Q0 - value for the Ru(d,p) Ru reaction has been determined to a much
improved accuracy.
Introduction
The level structure of the odd ruthenium isotopes has been investigated using γ - and
β - ray spectroscopy techniques (NDS 1973, 1974a-1974c), and has been shown to be
complex. Theoretical calculations using various models for these nuclei have been attempted
and met with varying degrees of success (Goswami and Sherwood1967; Imanishi,
Fujiwara and Nishi 1973; Kisslinger and Sorensen 1963).
The interpretation of level structure in terms of theoretical models is enhanced by

information on the location and distribution of the single-particle strengths among the
levels. The single-particle neutron strength distributions for the heavier isotopes 103Ru
and 105Ru have been investigated via the (d,p) reaction by Fortune et al. (1971).
The objective of our study was to obtain similar and much needed information for
the 97Ru and 101Ru isotopes.
Experimental procedure and the Q-value determination

The 11.5 MeV deuteron beam was produced by the Australian National University EN
Tandem accelerator. The targets consisted of isotopically enriched 96Ru (98%) or
100
Ru (97 %), evaporated using electron beam bombardment onto thin carbon
backings. The targets were approximately 20 - 50 μg/cm2 areal density, as determined
by comparing elastically scattered 4 MeV deuterons at forward angles with the
Rutherford cross section. The absolute cross sections correct to better than 15%.
Two ΔE-E detector telescopes were used to detect the protons emitted in the (d,p)
reaction. These detectors were cooled to approximately -20°C. A target monitor detector
was placed at 90°. All detectors were of the silicon surface barrier type and
manufactured at the local ANU laboratory. Overall experimental energy resolution was
approximately 25 keV. Examples of spectra from the two reactions are shown in Figure
26.1 for the lab angle of 25°.
96,100
Fig. 26.1. Proton spectra for the Ru(d,p)97,101Ru reactions at the incident deuteron energy of 11.5
MeV and lab angle of 25°.
Contaminants identified in the targets are 160, 13C and 12C. Other light contaminants are
also present in small quantities. The presence of all these contaminants prevented the
extraction of yields for some states and angles.
The very thin nature of the ruthenium targets on the thin carbon backings, enabled us to
make a more accurate determination of the ground-state Q0 - value for the 96Ru(d,p)97Ru
reaction. At certain angles, protons from the 13C(d,p)14C ground-state reaction were
observed to have nearly the same energy as those from the 97Ru ground state. The
ground-state Q0 - value for the 96Ru(d,p)97Ru reaction was determined to be
5.886±0.003 MeV, which is within the limits of the previously listed value of 5.816±0.100
MeV (Wapstra and Gove 1971) but now its accuracy has been significantly improved.
The Q0 - value determined for the 100Ru(d,p)101Ru reaction is consistent with the Wapstra
and Gove value of 4.581±0.004 MeV.
The excitation energies for the levels in both isotopes were determined at several angles
using protons from the (d,p) reactions on 16O, 13C and 12C as calibration points. Angular
distributions for fifteen levels in each of 97Ru and 101Ru were measured from 15° to 45°,
and from 52.5° to 67.5° in 5° steps.
The distorted wave analysis and results

Angular distributions for the (d,p) reaction were analysed using the DWBA formalism
and the computer code DWUCK on the ANU UNIVAC 1108 computer. The calculations
included a finite-range correction factor of R = 0.657, as well as corrections for the
non-locality of the optical potentials using non-locality lengths of 0.54 for the deuteron
channel and 0.85 for the proton channel (see the Appendix E).
The distorted waves for the incident and exit channels were calculated using optical
model potentials of the conventional form with a surface absorptive term and a real
Thomas-type spin-orbit term as defied in Chapter 25.
The neutron bound-state wave functions were calculated using the same geometry
as that of the real part of the Woods-Saxon potential for the proton channel in the
distorted wave calculation. The potential also included a Thomas-type spin-orbit term.
The depth of the real potential was adjusted to reproduce the experimentally determined
separation energy of each level. The potential parameters for deuterons, protons and
captured neutrons are listed in Table 26.1.
Table 26.1
Potential parameters used in the distorted wave analysis of the 96,100Ru(d,p)97,101Ru angular distributions
rC – The Coulomb potential used in the calculations is assumed to be caused by a

uniformly charged sphere with the radius of rC = A1/3.
a
) – The depth of the potential is adjusted to give the correct value of the separation
energy for a given energy level.
b
) – Calculated using the energy-dependent formula V = 59.84 − 0.32 E p .
c
) – Calculated using the energy-dependent formula WD = 12.80 − 0.25E p .
d
) – Calculated using the energy-dependent formula V = 60.87 − 0.32 E p .
e
) – Calculated using the energy-dependent formula WD = 13.34 − 0.25E p .
The relationship between the experimental and calculated cross sections is given by
⎛ dσ (θ ) ⎞ ⎛ dσ (θ ) ⎞
⎜ ⎟ = 1.55S (l , j )⎜ ⎟
⎝ dΩ ⎠exp ⎝ dΩ ⎠th
where S (l , j ) is the spectroscopic factor, l is the transferred orbital angular momentum,
j is the transferred total angular momentum, and 1.55 is a zero-range coefficient
calculated using the Hulthén wave function for deuterons (see the Appendix E).
The 96Ru and 100Ru isotopes have 2 and 6 neutrons outside the closed N = 50 shell,
respectively. The shell just above N = 50 is made of 1g7/2, 2d5/2, 2d3/2, 3s1/2, and 1h11/2
configuration, in that order for the undeformed potential. The stripped neutron can be
deposited to any of these orbitals.
As can be seen from Figures 25.2 and 26.3, the majority of transitions for the
96
Ru(d,p)97Ru and 100Ru(d,p)101Ru reactions display, as expected, the l = 0 and 2 angular
distributions but l = 4 and 5 angular distributions are also present. In general the fits are
good for all the measured distributions.
A few weakly excited states were observed in the spectra and while their excitation
energies were extracted, the corresponding complete angular distributions could not be
obtained.
Figure 26.2. The distorted wave fits to the 96Ru(d,p)97Ru angular distribution data. The l - value for
transferred neutrons and excitation energies in MeV are indicated. Where error bars are not shown,
the size of the data point indicates the approximate statistical error in the cross section.
Figure 26.3. The distorted wave fits to the 100Ru(d,p)101Ru angular distribution data. See the caption to
Figure 26.2
Discussion
Spectroscopic information extracted from the 96,100Ru(d,p)97,101Ru reactions is summarized
in Tables 26.2 and 26.3.
The two target nuclei, 96Ru and 100Ru, have the ground-state spin-parity values of 0+.
Thus the orbital angular momenta of the states formed in the (d,p) reaction are uniquely
determined by comparing their measured angular distributions with the distorted
wave calculations.
For l = 2, there are two configuration options available in the shall outside the closed
shell N = 50: 2d3/2 and 2d5/2. The distorted wave analysis does not allow to
distinguish between these two configuration, so unless the spin j is known from
earlier studies, two values are listed in Table 26.2. For other l values, only single
configurations are available, so unique spin j assignment can be made with a high
degree of confidence to the relevant states on the basis of the l values determined
by the distorted wave analysis.
97
Ru
Table 26.2 lists the excitation energies, Ex (in MeV), orbital angular momenta, l, as
determined by the distorted wave analysis, the possible neutron single-particle
configurations, the total angular momenta, j, and the spectroscopic factors S (l , j )
determined using the 96Ru(d,p)97Ru reaction.
Table 26.2
97
Spectroscopic information for Ru obtained using the 96Ru(d,p)97Ru reaction
Ex l Conf. j S (l , j ) Ex l Conf. j S (l , j )
5 +
0.000 2 2d5/2 /2 0.57 2.080 ?
3 + 1/ +
0.189 2d3/2 /2 2.173 0 3s1/2 2 0.06
7 + 1/ +
0.421 4 1h7/2 /2 0.61 2.284 0 3s1/2 2 0.11
3 + 5 + 1/ +
0.527 2 2d3/2, 2d5/2 /2 , /2 0.13 2.350 0 3s1/2 2 0.02
3 + 5 + 3 + 5 +
0.770 2 2d3/2, 2d5/2 /2 , /2 0.05 2 2d3/2, 2d5/2 /2 , /2 0.05
1/ + 1/ +
0.908 0 3s1/2 2 0.58 2.506 0 3s1/2 2 0.07
3 + 5 +
1.477 2 2d3/2, 2d5/2 /2 , /2 0.19 2.605 ?
11 - 1/ +
1.887 5 1h11/2 /2 0.57 2.652 0 3s1/2 2 0.05
3 + 5 + 1/ +
1.929 2 2d3/2, 2d5/2 /2 , /2 0.13 2.702 0 3s1/2 2 0.08
3 + 5 +
2.005 ? 3.030 2 2d3/2, 2d5/2 /2 , /2 0.09
Excitation energies are determined with the accuracy of ±7 keV.
The ground state and the states with excitation energies 0.527, 0.770, 1.477, 1.929 and
3.030 MeV are populated with an l = 2 transfer. The states with excitation energies
0.908, 2.173, 2.284, 2.506, 2.652 and 2.702 MeV are populated with an l = 0 transfer
and are assigned the spin-parity 1/2+. The states with 0.421 and 1.887 MeV excitation
energies are populated with l = 4 and 5 transfers, and are assigned the spins 7/2+
and 11/2-, respectively.
The angular distribution for the state with 2.350 MeV excitation energy could be fitted
using both an l = 0 and 2 transfer, and thus is presumed to be an unresolved doublet.
The states with 0.189, 2.005, 2.080 and 2.605 MeV excitation energies were only weakly
populated, and reliable angular distributions could not be extracted.
The ground state of 97Ru, with spin-parity 5/2+ consistent with the γ-spectroscopy
measurements of Ohya (1974), is strongly excited in the 96Ru(d,p)97Ru reaction and
carries most of the observed l = 2 strength. Its spectroscopic factor is similar to that
for the ground states of the two isotones 93Zr and 95Mo. The first excited state at 0.189
MeV, with known spin-parity 3/2+ was observed at only a few angles and was too weakly
excited to determine the spectroscopic factor, indicating that this state has little d3/2
single-particle component in its wave function. Similar results are known for the first
excited states of 93Zr and 95Mo.
The 0.421 MeV g7/2 state is observed to have an l = 4 stripping pattern and is the only l
= 4 transition located in our study for this isotope. Ohya (1974) suggests the presence
of 7/2+ or 9/2+ at 0.839, 0.879,1.229,1.932, 1.970, 2.186 and 2.755 MeV. In the isotones
93
Zr and 95Mo, the state carrying most of the l = 4 strength is found at 1.477 and 0.768
MeV respectively. The addition of protons is seen to cause a lowering of the energy of
the g7/2 neutron orbital.
The state at 0.908 MeV is strongly excited and carries most of the l = 0 strength. In 93Zr
and 95Mo, the l = 0 strength is concentrated in states at 0.947 and 0.789 MeV,
respectively.
The state at 1.887 MeV excitation energy is the only state exhibiting an l = 5
stripping pattern for this reaction and is presumed to have 1h11/2 configuration. In 93Zr
and 95Mo, the major h11/2 strength is found in the states at 2.040 and 1.949 MeV,
respectively. The effect of the extra protons on the h11/2 neutron orbital appears to be
not as large as on the g7/2 configuration.
101
Ru
100
Spectroscopic information for the Ru(d,p)101Ru reaction is summarised in Table
26.3.
Table 26.3
Spectroscopic information for 101Ru obtained using the 100Ru(d,p)101Ru reaction
Ex l Conf. j S (l , j ) Ex l Conf. j S (l , j )
5 +
0.000 2 2d5/2 /2 0.35 0.714 ?
3 + 3 + 5 +
0.127 2 2d3/2 /2 0.02 0.827 2 2d3/2, 2d5/2 /2 , /2 0.12
1/ +
0.325 0 3s1/2 2 0.65 0.910 ?
3 + 5 + 3 + 5 +
0.408 2 2d3/2, 2d5/2 /2 , /2 0.05 0.976 2 2d3/2, 2d5/2 /2 , /2 0.18
3 + 5 + 1/ +
0.535 2 2d3/2, 2d5/2 /2 , /2 0.26 1.110 0 3s1/2 2 0.10
7 + 3 + 5 +
0.599 4 1g7/2 /2 0.45 1.588 2 2d3/2, 2d5/2 /2 , /2 0.12
1/ + 11 -
0.625 0 3s1/2 2 0.05 1.695 5 1h11/2 /2 0.18
1/ + 3 + 5 +
0.684 0 3s1/2 2 0.02 1.825 2 2d3/2, 2d5/2 /2 , /2 0.04
3 + 5 + 3 + 5 +
2 2d3/2, 2d5/2 /2 , /2 0.04 1.875 2 2d3/2, 2d5/2 /2 , /2 0.12
Excitation energies are determined with the accuracy of ±7 keV.
The ground state and the states with excitation energies of 0.127, 0.408, 0.535, 0.827,
0.976, 1.588, 1.825 and 1.875 MeV are populated with an l = 2 transfer. The states
with excitation energies of 0.325, 0.625 and 1.110 MeV are populated with an l = 0
transfer and are assigned a spin-parity of 1/2+. The states with 0.599 and 1.695 MeV
excitation energies are populated with l = 4 and 5 transfers, respectively.
The angular distribution for the state at 0.684 MeV could be reproduced by assuming
both an l = 0 and an l = 2 transfer, and is presumed to be an unresolved doublet.
Additional states were observed with 0.714 and 0.910 MeV excitation energies but
angular distributions could not be extracted.
The ground state of 101Ru, determined previously (Fuller and Cohen 1969) to have
spin-parity 5/2+ is strongly excited in this reaction. The first excited state of 101Ru is at
0.127 MeV excitation energy and has known spin-parity 3/2+. This state is more strongly
excited than the 0.189 MeV state of 97Ru, indicating a larger d3/2 single-particle component
in its wave function. Its spectroscopic factor is similar to that of the first excited state of
103
Pd.
The state at 0.325 MeV excitation energy is populated strongly and carries most of the
s1/2 single-particle strength. The state carrying most of the s1/2 strength in 103Pd is found
somewhat higher, at 0.500 MeV excitation energy.
The only state exhibiting an l = 4 angular distribution for this isotope was located at
0.599 MeV excitation energy. The γ-ray measurements indicate many l = 4 states, some
with low excitation energies. In particular one with 0.307 MeV excitation energy, which
if present, would be masked by the strong l = 0 state at 0.325 MeV. In 103Pd the strongest
l = 4 state occurs at an excitation energy of 0.245 MeV.
The state with 1.695 MeV excitation energy was the only state exhibiting an l = 5 angular
distribution and is assumed to have 1h11/2 configuration. In 103Pd, the major l = 5
strength is found lower, in the state with 0.787 MeV excitation energy.
Figure 26.4. The distributions of spectroscopic strength for l = 0, 2, 4 and 5 angular momentum
transfers in the 96,100Ru(d, p)97,101Ru reactions.
The spectroscopic strengths for the 96,100Ru(d,p)97,101Ru reactions are plotted against
the excitation energies in Figure 26.4 for various observed l - transfers. The effect of
the four additional neutrons in the 100Ru core is mainly in lowering the position of the l =
0 strength and in compressing the l = 2 strength distribution.
Summary
The ground-state Q0-value for the 96Ru(d,p)97Ru reaction has been determined to much
improved accuracy. Twenty states in 97Ru with excitation energies up to 3.030 MeV,
sixteen not previously observed, and eighteen states in 101Ru with excitation energies up
to 1.875 MeV, fourteen not previously observed, have been identified. Orbital angular
momentum transfer values and spectroscopic factors have been obtained for states in
both 97Ru and 101Ru. In 97Ru six l = 0, six l = 2, one l = 4, one l = 5 and one
admixture of l = 0 and l = 2 have been assigned. In 101Ru three l = 0, nine l = 2, one
l = 4, one l = 5 and one admixture of l = 0 and l = 2 have been assigned.
The distorted wave calculations, using optical model parameters from global
analyses, provided a good description of the measured angular distributions. The
neutron single-particle strength distribution for 97Ru is similar to those obtained for the
isotones 93Zr and 95Mo. The neutron single-particle strength distribution for 101Ru is
similar to that of the isotone 103Pd, while differing markedly from that of 99Mo.
References
Fuller, G. H. and Cohen, V. W. 1969, Nucl. Data Tables A5:433
Goswami, A. and Sherwood, A. I. 1967, Phys. Rev. 161:1232.
Imanishi, N., Fujiwara, I. and Nishi, T. 1973, Nucl. Phys. A205:531.
NDS 1973, Nucl. Data Sheets 10:1, 47.
NDS 1974a, Nucl. Data Sheets 11:449.
NDS 1974b, Nucl. Data Sheets 12:431.
NDS 1974c, Nucl. Data Sheets 13:337.
Ohya, S., 1974, Nucl. Phys. A235:361.
Wapstra, A. H. and Gove, N. B. 1971, Nucl. Data Tables A9.
27
53,55,57
Spectroscopy of the Mn Isotopes and the Mechanism of the
(4He,p) Reaction
Key features:
1. A total of 122 excited states have been identified in the 53,55,57Mn isotopes and the
corresponding excitation energies have been assigned to all of them. Many of the
states have never been observed before, particularly in 57Mn where we have identified
38 new states.
2. We have found that the reaction mechanism depends strongly on the incident 4He
energy. Most of the angular distributions measured using 18 MeV 4He projectiles are
associated with an indirect reaction mechanism. In contrast, angular distributions
measured at 26 MeV show clear direct reaction features.
3. A total of 95 distributions have been measured using 26 MeV 4He projectiles and
were analysed using the distorted wave formalism. We have found that at this energy,
the (4He,p) reaction can be interpreted as a direct transfer of three-nucleon cluster.
4. The J – dependence have been observed for both L = 1 and L = 3 angular momentum
transfer.
5. A total of 46 Jπ - values have been assigned to states in 53,55,57Mn nuclei.
6. We have found that many states, which are weakly excited in single transfer reaction,
are excited strongly in the (4He,p) reaction. New states, which were not previously
observed in the (3He,d) reaction, have been also accessed by the (4He,p) reaction.
Thus, this reaction offers an important alternative way to study nuclear structure.
50,52,54 4 53,55,57 4
Abstract: The Cr( He,p) Mn reactions have been studied at 18 and 26 MeV He
bombarding energies. From the 26 MeV data, angular distributions for 95 levels were obtained,
nearly all of which could be described by the distorted wave procedure assuming a quasi-triton
transfer process. In contrast, at 18 MeV very few angular distributions could be adequately
described using the direct reaction mechanism. The J - dependence was observed for both L = 1
and L = 3 transfers and used to assign Jπ values for many states in 53,55Mn. In 57Mn, 38 new states (in a
total of 57) were observed and Jπ assignments were made for many of them. The (4He,p) reaction
mechanism and nuclear structure are discussed.
Introduction
The use of the (4He,p) or (p,4He) reactions in nuclear spectroscopy has often been
limited by an inadequate knowledge of the reaction mechanism and by the need for
using simplifying procedures in distorted wave analyses of experimental data.
However, these factors are less restrictive in obtaining spectroscopic information if
detailed single-proton transfer measurements are available to the same final states. In
such cases the great attractions of these multi-particle transfer reactions can be more fully
utilized. Use can be made of their selectivity associated with seniority and isospin, of their
ability to access complex configurations, and of their applicability to resolve the Jπ
ambiguity by using the J - dependence (Bucurescu et al. 1972), which is particularly clear
for the L = 1 angular momentum transfer. Our measurements of the 50,52,54Cr(4He,p)
reactions at 18 and 26 MeV were undertaken to study the spectroscopy of the 53,55,57 Mn
isotopes and to examine the (4He,p) reaction mechanism in this energy-mass region.
Both 5 3 Mn and 55 Mn have been studied earlier. Katsanos and Huizenga (1967) used
the (p,p') scattering to study the 55Mn isotope. Tarara et al. (1976) used the 56Fe(p,4He)
reaction at 14-16 MeV to examine the 53Mn isotope. The single-proton structure of 53Mn
has been studied with the (3He,d) reaction by O'Brien et al. (1969). The same reaction
together with the J - dependence for the (7Li,6He) reaction was used by Gunn, Fix, and
Kekelis (1976) to make Jπ assignments. Rapaport et al. (1969) have used the (3He,d)
reaction in a study of 55Mn. The preponderance of measured l p values in these (3He,d)
studies have been for l p = 1 and 3. Hence the J - dependence for the L = 1 transfer in the
(4He,p) reactions can be used to make Jπ assignments. The availability of L = 3
transitions permits also an experimental test for any similar J - dependence for L = 3
transfer. The (p,3He) and (d,4He) reactions at 27 and 16.5 MeV, respectively, have been
used to study low-lying states in 55Mn by Peterson, Pittel and Rudolph (1971) and by
Peterson and Rudolph (1972).
At the commencement of the present work no information was available on the excited
states of 57Mn. However, in the course of our study, Mateja et al. (1976) published their
results for the 54Cr(4He,pγ)57Mn reaction at 15, 21 and 24 MeV, and Mateja et al. (1977)
for the 55Mn(t,p)57Mn reaction at 17.0 MeV. They have made several spins assignments
on the basis of their p-γ angular correlation studies but no proton angular distributions
were reported.
The 50,52,54Cr isotopes have 24 protons and 26, 28, and 30 neutrons, respectively. In
the simple shell model description, the four protons are outside the Z = 20 shell and
occupy the 1f7/2 orbits. The neutrons assume an interesting set of configurations
spaning the N = 28 shell: in 50Cr two neutrons are missing to close the N = 28 shell;
in 52Cr the N = 28 shell is closed; and in 54Cr there are two neutrons outside the N =
28 shell.
Thus in the (4He,p) reaction to low-lying states, the stripped proton may be expected
to be transferred preferentially to the N = 28 shell, which is made of 1f7/2 orbitals. For
the stripped neutrons, the most likely transfer for the low-lying states excited in the
50
Cr(4He,p)53Mn reaction is to the 1f7/2 orbital. However, the participation of the
configurations in the N = 50 shell, i.e. 2p3/2,1f5/2, 2p1/2 and even 1g9/2 are also possible.
For the 52,54Cr(4He,p)55,57Mn reactions, the two stripped neutrons are less likely to be
transferred to the 1f7/2 orbital but rather to any configurations in the N = 50 shell, i.e.
2p3/2,1f5/2, 2p1/2 and 1g9/2.
The measurements were carried out using two particle accelerators. For the 18 MeV
measurements, we used 400 nA 4He beam delivered by the ANU EN tandem accelerator.
Measurements at 26 MeV were carried out using the ANU 14UD pelletron accelerator.
For 18 MeV measurements, chromium targets of around 50 μg/cm2 on gold backings were
produced from enriched material while for 26 MeV self-supporting targets of around 200
μg/cm2 were used. The reaction products were detected using cooled surface-barrier
detector telescopes consisting of two or three detectors depending on the beam energy
and the (4He,p) Q - values. The overall experimental resolution was typically around 32
keV at 18 MeV and varied from 45 to 60 keV at the higher energy with the forward angle
data having the better resolution. Special attention was given to minimizing, as far as
possible, contributions to the resolution from kinematics and target thickness effects. The
consistency of angular distribution shapes for the same known transitions in all isotopes is a
good indicator that, even at the higher excitation energies, there is little contribution from
unresolved components.
Examples of proton spectra are shown in Figures 27.1 and 27.2. Figure 27.1 shows the 26
MeV spectra for the 50,52Cr(4He,p) 5 3, 5 5 Mn reactions while Figure 27.2 shows the proton
spectra for the 54Cr(4He, p)57Mn reaction at both 18 and 26 MeV.
Energy calibrations for the 53,55Mn isotopes were made using the well-known levels up to
around 5 MeV in each isotope. The Q - values were determined with accuracy generally
better than 10 keV.
50 4 53 52 4 55
Figure 27.1. Examples of proton spectra for the reactions Cr( He,p) Mn and Cr( He,p) Mn at 26
MeV.
Figure 27.2. Examples of proton spectra for the 54Cr(4He,p)57Mn reaction at 18 and 26 MeV. The
circled numbers indicate the peaks that either disappear or show considerably reduced intensity at 26
MeV.
At the commencement of the present work no levels were known in 57Mn so that special
care was exercised in determining the 57Mn Q - values. Since 52Cr was the major
contaminant (7%) in the 54Cr targets, the 54,52Cr(4He,p) reactions were run consecutively at
each angle at both energies. In this way the 57Mn spectra were calibrated using 55Mn
levels and contaminants due to 52Cr were removed by subtracting the normalized
52
Cr(4He,p) spectra. Allowance was made from elastic scattering yields for differences in
target thickness. The Q - values were determined from the better resolution data at 18 MeV
and have errors of the same order as the levels in 55Mn.
Angular distributions were in general measured from 15° to 150° at 18 MeV and from 20°
to 90° at 26 MeV. The differential cross sections at both energies were determined by
measuring them relative to known elastic scattering cross sections. For normalization,
a fixed monitor detector was placed at 45° and counted both the elastic events and the
inelastic scattering to the first 2+ state of the target. At 26 MeV the elastic scattering at
each angle was also recorded in the first detector of each telescope simultaneously
with the reaction. The absolute elastic scattering cross sections were determined from a
comparison with Rutherford scattering. The absolute cross section is accurate to around
7%.
Data analysis
The angular distributions for the 50,52,54Cr(4He,p)53,55,57Mn reactions were analysed
using the distorted wave theory and the computer code DWUCK. The calculations
assumed a quasi-triton cluster transfer. The transferred triton was assumed to be
bound in a Woods-Saxon potential and the customary separation energy prescription was
used. A spin-orbit term was tried in the triton well but did not affect the shapes of the
calculated angular distributions.
The optical model parameters used in both channels were taken from the Perey and
Perey compilation (1974). The 4He-particle parameters were determined by measuring
elastic scattering angular distributions and analysing them using the optical model with
volume absorption. An example of fits for the elastic scattering of 4He at 26 MeV is given
in Figure 27.3. The 4He and triton parameters used in the distorted wave analysis are
listed in Table 27.1.
Table 27.1
The 4He and triton parameters used in the distorted wave analysis of the 50,52,54Cr(4He,p)53,55,57Mn
angular distributions at 18 and 26 MeV
a
) Adjusted to match the triton separation energy.
The magnitudes and shapes of the 50,52,54Cr(4He,p)53,55,57Mn angular distributions have

been found sensitive to the triton well geometry. Good fits to almost all angular
distributions at 26 MeV were obtained using r0 = 1.45 fm and a0 = 0.35 fm for tritons.
However, in a few cases, to improve the fits it was necessary to deviate from this
triton geometry. In particular a0 had to be reduced considerably. These exceptional
cases are indicated by the dashed lines in Figures 27.4 - 27.8, which show the
50,52,54
Cr(4He,p)53,55,57Mn angular distributions. These cases were not used in extracting
the reduced strength information. Furthermore, except for the well-known L = 1
transitions characterized by a very distinct J - dependence, spins associated with
calculated distributions for these cases have not been considered as unambiguous
assignments even for states with known l p values.
Figure 27.3. Examples of the optical model fits to the elastic scattering of 4He particles at 26 MeV.
The parameter sets for all these calculations are listed in Table 27.1.
The calculated distributions shown in Figures 27.4 - 27.9 have been normalized to the
experimental data. The errors shown include both the statistical error and an estimate
of the uncertainty introduced by the spectrum fitting and normalization procedures.
The similarities in the shapes of the angular distributions made it in general difficult to
assign definite spin values unless the l p transfer was already known.

In general, the (4He,p) distributions measured at 18 MeV were difficult to describe using
the distorted wave formalism. The distributions, which could be fitted, are displayed in
Figures 27.4 and 27.5. The displayed l p values were supplied by the (3He,d)
measurements. States with large spectroscopic factors in (3He,d) are strongly excited
in the (4He,p) reaction and even at 18 MeV their angular distributions exhibit the
direct mechanism over a wide range of angles of at least for θ < 90°. The notable
examples of this are the strong transitions to the 7/2- and 3/2- states at 0 and 2.413 MeV in
53
Mn and at 0.128 and 2.258 MeV in 55Mn. However, both the 18 and 26 MeV data reveal
that it is not a necessary condition for a state to have a large spectroscopic factor in a
single-proton transfer reaction in order to be strongly excited by the (4He,p) reaction. In
fact, we have found that many states, which are weakly excited in single transfer
reaction, are excited strongly in the (4He,p) reaction. New states, which were not
previously observed in the (3He,d) reaction, have been also accessed by the (4He,p)
reaction.
Figure 27.4. Examples of angular distributions at 18 MeV that could be interpreted using the distorted
wave formalism. The displayed l p values were taken from (3He,d) results of O’Brien et al. (1969) and
Rapaport et al. (1969).
Figure 27.5. More examples of angular distributions at 18 MeV that could be interpreted using the
distorted wave formalism. The figure also shows examples of angular distributions that could not be
described theoretically.
Figure 27.6. Angular distributions for the 50Cr(4He,p)53Mn reaction at 26 MeV. The numerical labels
correspond to the labels used in the proton spectra (see Figure 27.2). The dashed lines are calculated
using different geometrical parameters (mainly a0) from those listed in Table 27.1 for tritons (see the
text). The dashed curves were not used to extract the reduced strengths.
50,52
Figure 27.7. Examples of angular distributions for the Cr(4He,p)53,55Mn reactions at 26 MeV. See
52,54
Figure 27.8. Examples of angular distributions for the Cr(4He,p)55,57Mn reactions at 26 MeV. See
Figure 27.9. Left-hand side: Angular distributions for the 54Cr(4He,p)57Mn reaction at 26 MeV. (See
also the caption to Figure 27.6.) The right-hand side: Examples of the J - dependence for L = 3.
Figure 27.10. Two examples of strong energy dependence of the reaction mechanism for the (4He,p)
reaction. Featureless distributions at 18 MeV display clear direct transfer mechanism at 26 MeV.
At 26 MeV, nearly all the angular distributions could be described well by the
distorted wave theory. These angular distributions are presented in Figures 27.6-27.9.
Our study revealed that the reaction mechanism for (4He,p) reactions depends strongly
on the incident energy of 4He particles. Many featureless distributions measured at 18
MeV displayed clear characteristics of direct mechanism at 26 MeV. Examples are the
distributions for the 2.578 and 2.684 MeV in 53Cr shown in Figures 27.10 for 18 and 26
MeV.
The 26 MeV data also show a clear J – dependence of the shape of the angular
distributions for L = 3 (see Figure 27.9).
Referring to the results for the (3He,d) reactions, l p = 3 transfers have been located
at 0.0, 3.110, 3.670 and 4.067 MeV in 53Mn and at 0.128, 1.883, 3.136 and 3.608 MeV
in 55Mn. The (4He,p) angular distributions for these states can be categorized into two
groups. One group consists of the ground and 4.067 MeV states in 53Mn shown in Figure
27.6 and the 0.128 and 1.883 MeV states in 55Mn shown in Figure 27.7. This group is
characterized by a peak in the cross section at about 35°.
The second group contains the 3.110 and 3.670 MeV states in 53Mn shown in Figure 27.6
and the 3.136 MeV state in 55Mn shown in Figure 27.7. The angular distributions for these
states show less structure than do those of the first group. In particular they do not
exhibit a maximum in the cross section at θ = 35°. All these distributions are grouped
together in Figure 27.9. Based on the distorted wave calculations the first group
consists of Jπ = 7/2- transfers and the second of Jπ = 5/2- transfers.
We have also observed a clear J -dependence for L = 1 transfers. This can be seen by
comparing distributions for Jπ = 1/2- and 3/2- in Figures 27.6 and 27.7. The distribution
for Jπ = 1/2- display well-defined diffraction structure, which is reproduced reasonably
well using the distorted wave theory. In contrast, the differential cross sections for Jπ
= 3/2- decrease virtually smoothly with the increasing reaction angle.
The 50Cr(4He,p)53Mn reaction
The results of the present work are compared with those from previous measure-
ments in Table 27.2. The high-resolution 56Fe(p,4He)53Mn measurements of Tarara
et al. (1976) indicate that there are at least 81 states in 53Mn with the excitation
energies of up to 5.092 MeV, whereas the (4He,p) reaction at 26 MeV excites only
nineteen resolvable groups.
By comparing theoretical and experimental cross sections one can derive the reduced
strength R values for the (4He,p) transitions. The reduced strength is defined as
∑ σ exp (θ )
R= θ
∑θ σ th (θ )
where σ exp (θ ) is the experimental differential cross section at the reaction angle θ and
σ th (θ ) is the cross section calculated theoretically using the distorted wave formalism. The
reduced strengths for the 50Cr(4He,p)53Mn reaction at 26 MeV are shown in Figure 27.11.
Several theoretical studies of the 53Mn nuclear structure have been made (Benson and
Johnstone 1975; Lips and McEllistrem 1970; Malik and Scholz 1966; McCullen,
Bayman, and Zamick 1964; Osnes 1971; Saayman and Irvine 1976; Scholz and
Malik 1967). The most extensive study was by Benson and Johnstone (1975). Their
study included the 1 f7−/32 , 1 f 7−/42 , 2 p3 / 2 , 2 p1 / 2 , and 1 f5 / 2 configurations. They concluded
that the lowest 1p-4h states belong to predominantly neutron excitations.
Table 27.2
53 50 4 53
Spectroscopic information about Mn extracted from our study of the Cr( He,p) Mn reaction at 26
MeV compared with earlier results
a) See Figure 27.1.

b) Excitation energies determined by our measurements.
c) O’Brien et al. (1969) for Ex ≤ 6 MeV; Gunn, Fox, and Kekelis (1976) for Ex > 6
MeV.
d) Distributions fitted using altered geometrical parameters for tritons (see the
text).
e) Compilation of the orbital angular momentum values for the transferred
proton (Armstrong and Blair 1965; Čujec and Szöghy 1969; Gunn, Fox, and
Kekelis 1976; O’Brien et al. 1969).
f) Compilation of previous spin assignments (Auble and Rao 1970; Armstrong,
Blair, and Thomas, 1967; Gunn, Fox, and Kekelis 1976; Schulte, King, and
Taylor (1975); Wiest et al.1971).
g) Our Jπ assignments.
h) Differential cross sections for the 50Cr(4He,p)53Mn at 26 MeV measured at the
most forward angle θ 0 . For most states θ 0 ≈ 20.80 (c.m.).
Figure 27.11. Reduced strengths R for the 50Cr(4He,p)53Mn reaction at 26 MeV. The figure shows also the
assigned Jπ values.
The first five states of 53Mn at 0, 0.376, 1.288, 1.438 and 1.618 MeV consist mainly of
the (1 f 7−/32 ) J configuration with well-known spins of 5/2-, 3/2-, 11/2-, and 9 / 2 - . As such, the
proton seniority26 (νp) of the 1 f 7−/32 components is necessarily νp = 3 for these states.
The excitation of the 1.288 MeV state in both the (3He,d) and (4He,p) reactions might be
associated with a (1 f 7−/42 )0 2 p3 / 2 component present due to the close proximity of the 2 p3 / 2
single-particle state. The angular distribution for this state could not be fitted satisfactorily
with the distorted wave procedure and as a consequence cannot rule out a more
complicated excitation process. However, the same fitting problems were encountered
with all 3/2- distributions in 53Mn, including the distribution for the state at 2.413 MeV,
which can be identified with the 2.5 MeV state predicted by Benson and Johnstone.
According to the same calculations, the 2.578 MeV level, assigned 7 / 2 - , is probably formed
from two-neutron excitation out of the 1 f 7 / 2 , shell. This could explain both the excitation
of this state in the (4He,p) reaction and its non-population in the single-proton transfer
reaction. The calculation also predicts two 1/2- states at 2.56 and 3.4 MeV and a 5/2-
state at 3.1 MeV. These could be identified with the levels observed at 2.684, 3.460 and
3.110 MeV, respectively. The distorted wave calculations have reproduced these angular
distributions satisfactorily.
Spectroscopic information for the 55Mn nucleus is summarized in Table 27.3.
A comparison of proton spectra in Figure 27.1 shows that many more states are excited
in 55Mn than in 53Mn for levels of up to 4.1 MeV excitation energy. In fact there are 19
groups in the 55Mn spectrum up to 4.086 MeV (peak 19) compared to 10 groups in
the 53Mn spectrum. The larger density of states in 55Mn is not surprising considering
that 53Mn has a closed N = 28 neutron core whereas the two added neutrons in 55Mn
26
The number of unpaired fermions.
have access to the f-p orbitals. The prominent gap in the 53Mn spectrum between the
ground state and the 2.413 MeV level (peaks 1A and 2) is broken only by the weakly
excited 1.288 MeV level (peak IB). However, in 55Mn there are five states excited
between the strongly excited 7/2- and 3/2- levels (peaks 1 and 7). In both nuclei, the
low-lying 5/2- states at 0.376 MeV in 53Mn and for the ground state in 55Mn are absent in
the 26 MeV spectra.
Table 27.3
Spectroscopic information about 55Mn extracted from our study of the 52Cr(4He,p)55Mn reaction at 26
MeV compared with earlier results
a) See Figure 27.1.

b) Our work.
c) Katsanos and Huizenga (1967).
d) Could be fitted only by readjusting triton parameters. See the text.
e) Katsanos and Huizenga (1967) and Kocher (1976).
f) Jπ assignments based on our work.
g) θ 0 = 23.40 (c.m.).
h) Suspected doublet (Kocher 1976).
i) Discussed in the text.
j) Assignment based on the (d,n) reaction (Kocher 1976).
An interesting feature of the 55Mn spectrum is the presence of several strongly excited
states above 4.7 MeV (peaks 23 to 27). At all angles measured in our study, the state
at 5.498 MeV (peak 25) dominated the 26 MeV spectra. It is interesting to note that
similar dominant peaks have been observed in 61,63,65,67Cu isotopes following the (4He,p)
reaction at 19.3 MeV bombarding energy (Bucurescu et al. 1972). Convincing Jπ = 9/2+
assignments were made for these states. It can be seen in Figure 27.8 that in our
study, the angular distribution for the 5.498 MeV state can also be well reproduced by
Jπ = 9/2+. A state at the same excitation energy has been previously assigned
(Rapaport et al. 1969) an l p = 3 transfer. This would suggest spin Jπ = 3/2- or 5/2- for
this state. However, in our study no fit to the data could be obtained using such
values unless the triton diffuseness was reduced to 0.1 fm.
Reduced strengths for states in 55Mn excited via 52Cr(4He,p)55Mn reaction at 26 MeV are
displayed in Figure 27.12.
52
Figure 27.12. Reduced strengths R for the reaction Cr(4He,p)55Mn reaction at 26 MeV. The figure
shows also the assigned Jπ values.
The energy level at 1.291 MeV excitation energy (peak 3) is of particular interest. There is a
considerable uncertainty about the nature of this level (see Chapter 28). Our (4He,p)
measurements at 26 MeV indicate that there is a doublet with Jπ = 1/2- and 11/2- at this
excitation energy. As can be seen in Figure 27.7, the angular distribution for the
52
Cr(4He,p)55Mn reaction leading to this state show a pronounced maximum and
minimum at the same angles as observed for other Jπ = 1/2- transitions. We have
calculated two angular distributions for this state using the distorted wave formalism,
one corresponding to Jπ = 1/2- and one for Jπ = 11/2-. We have then applied a least
squares procedure to fit the experimental angular distribution for this state using the two
calculated theoretical shapes. The resulting fit is displayed in Figure 27.7 by a full line.
The absolute values of the theoretical cross-sections at θ c . m . = 23° are 12 and 10
μb/sr for the Jπ = 1/2- and 11/2- components, respectively
The 2.197 MeV state was not populated strongly in single-proton transfer reactions. Our
study shows that angular distribution for the 52Cr(4He,p)55Mn reaction leading to this state
can be fitted using Jπ = 7/2- (see Figures 27.4 and 27.7) In contrast with its weak
excitation in the (3He,d) reaction, its intensity (peak 6) in the (4He,p) reaction is
comparable to the intensity of the 7/2- level at 0.128 MeV (peak 1).
The adopted level scheme (Kocher 1976) shows two states at 2.727 and 2.753 MeV
with spins of 7/2- and (5/2-, 7/2-), respectively. In our study, the peak observed at 2.741
MeV could therefore be regarded as a doublet. However, the (4He,p) angular
distribution has been fitted with just one Jπ value (Jπ = 7/2-). No improvement to the fit
was obtained by using combinations of Jπ = 3/2- and 5/2- or 3/2- and 7/2-.
The level at 2.980 MeV is supposed to have spins 3/2+ or 5/2+ (Kocher 1976; Rapaprt
et al. 1969). We have tried both of them without any success.
Early shell-model calculations (McGrory 1967; Vervier 1966) of the structure of 55Mn
confined protons to the 1 f 7 / 2 shell. This is an inadequate treatment as indicated by the
number of easily reproducible Jπ = 3/2- transitions observed in the earlier (3He,d) data
and in our (4He,p) measurements. Moreover, the existence of a 1/2- level at about
1.291 MeV, which is needed to reproduce our (4He,p) angular distribution and the
data of Peterson, Pittel, and Rudolph (1971) and Peterson and Rodolph (1972) can be
only predicted if proton excitation into the f-p shell is included. However, their 2p-3h plus
3p-4h calculations fail to predict the close proximity of the strong 3/2- state at 2.258
MeV and the 7/2- state at 2.197 MeV. They also predict that much of the 2p1/2 strength is
at 2.46 MeV while the (4He,p) data indicate that the L = 1 transfers around this energy
correspond to Jπ = 3/2-.

The Q - values for states in 57Mn were determined using the 18 and 26 MeV data and the
energy calibrations obtained for the 55Mn spectra, which were taken under the same
experimental as for the 57Mn measurements. From the 18 MeV data the ground-state Q0 -
value for the 54Cr(4He,p)57Mn reaction is
Q0 = -4.302± 0.008 MeV.
This value is in agreement with the value reported by Mateja et al. (1976). These authors
measured proton spectra at a few angles for the 54Cr(4He,p)57Mn reaction. The
discrepancy between the Q0 -value determined from our (4He,p) data and the value
reported by Gove and Wapstra (1972) can be removed if the data of Ward, Pile, and
Kuroda (1969) on the β - decay of 57Mn is reinterpreted by requiring the 83% branching
mode to populate the 136 keV level in 57Fe instead of the 14 keV level.
During the course of our investigation, Mateja et al. (1976) reported on low-lying levels in
57
Mn by using the (4He,p) and (4He,pγ) reactions at 15, 21 and 24 MeV. They have
identified levels in 57Mn for up to 2.234 MeV excitation energy. The spin assignments made
on the basis of the (4He,pγ) measurements are in agreement with our spin assignments.
Of particular interest in the results of Mateja et al. is a pair of states at about 1.06 MeV
separated by 15 keV and assigned spins of ( 1/2-) and ( 9/2-). This feature parallels the
suspected 1/2- and 11/2- doublet at 1.292 MeV in 55Mn.
Table 27.4
57
Spectroscopic information for Mn
a
) See Figure 27.2. b) Our assignments. c) Mateja et al. (1976). d) Mateja et al. (1977). e) Mateja
at al. (1976, 1977); Ward, Pile, and Kuroda (1969). f) θ0 = 23.40, Eα = 26 MeV.
Results of our work for 57Mn are tabulated in Table 27.4 where they are compared with
results of Mateja et al. (1976, 1977). A comparison of the spectra presented in Figure
27.2 with the (4He,p) spectra of Mateja et al. (1976) reveals a discrepancy in the
number of states excited between 1.378 and 2.188 MeV (peaks 6 and 14). The present
work indicates the existence of seven levels in this energy interval while Mateja et al.
(1976) found none. This may be explained partly by the differences in bombarding
energies and partly by statistics. It should be noted, however, that all these additional
states have been excited in their subsequent work (Mateja et al. 1977) using the (t,p)
two-neutron transfer reaction and have been well reproduced by the distorted wave
theory.
In general it was difficult to make unambiguous spin assignments in 57Mn. This was partly
due to the absence of known l p values and partly because of the close similarity between
angular distributions calculated for different values of Jπ.

The 50,52,54Cr(4He,p)53,55,57Mn reactions were studied at 18 and 26 MeV bombarding
energy. Analysis of the proton angular distributions revealed that contributions from
compound nucleus reactions were significant at 18 MeV.
In contrast, at 26 MeV, nearly all the proton groups detected had angular
distributions, which could be described using the distorted wave theory and
assuming a triton cluster transfer mechanism. The angular distributions of 41 proton
groups were obtained for 53Mn, 27 groups for 55Mn and 27 for 57Mn. Nearly all the
calculated Jπ transfers were for L = 1 or L = 3. Using the L = 1 J - dependence for
the (4He,p) reaction, many spin assignments could be made for states whose l p
values had already been determined from previous (3He,d) studies of 53,55Mn. In
addition, an L = 3 J - dependence was also found for (4He,p) in this energy-mass
region and used to assign the relevant Jπ to the observed states.
A summary of energy levels in 53,55,57Mn for which angular distributions have been
measured at 26 MeV incident 4He energy is shown in Figure 27.13. The figure
contains the excitation energies and Jπ assignments based on our study. A summary
of all spectroscopic information is presented in Tables 27.2-27.4.
The 7/2- strength is spread out among more states in 55Mn than in 53Mn, although not
all the 7/2- strength in 55Mn can be associated with the single particle 7/2-
configuration. The addition of two neutrons increases the number of configurations
that the (4He,p) reaction can excite as opposed to the configurations accessible in
single proton transfer. These extra degrees of freedom also allow for an explanation
of the larger number of states seen in 55,57Mn below 2.4 MeV as compared with 53Mn.
The 9/2- and 11/2- states at 0.983 and 1.292 MeV in 55Mn are presumably excited
through
[(πf )
−3
7/2 7/2 (νj1 j2 )J n
]
J
components, where j1, j2 are 2p3/2 or 1f5/2 and Jn = 2 or 4.

The importance of (3p-4h) configurations in the low-lying states of 55Mn is
demonstrated by the L = 1, Jπ = 3/2- state at 1.528 MeV. The (4He,p) reaction at 26
MeV also corroborates the assignment of a 1/2- state, which is nearly degenerate with
the 11/2- state at 1.292 MeV as suggested by Peterson, Pittel, and Rudolph (1971).
Figure 27.13. Excitation energies and Jπ assignments based on our study of the
50,52,54
Cr(4He,p)53,55,57Mn reactions. Only states for which angular distributions at 26 MeV were
measured are shown.
In 57Mn, Q - values for 57 states were determined using both the 18 and 26 MeV
data. The ground state Q0 - value was determined to be -4.302 ± 0.008 MeV, in
agreement with the value reported by Mateja et al. (1976). The states measured in
(4He,p) for excitation energies greater than 2.2 MeV have not been reported
previously.
The simple model of triton cluster transfer accounts very well for the shapes of the
angular distributions. Our measurements can serve as a basis for more refined
calculations to understand the details of reaction mechanism of three-nucleon
transfer and the structure of the Mn isotopes.
We have found that many states, which were not observed in single-nucleon transfer
reactions were accessible via the three-nucleon transfer in the (4He,p) reaction.
Likewise, many states that were weakly excited in single nucleon transfer reactions
were strongly excited in the (4He,p) reaction. The (4He,p) is a useful tool for
uncovering and studying new configurations in states of residual nuclei.
References
Armstrong, D. D. and Blair, A. G. 1965, Phys. Rev. 140:B1226.

Auble, R. L. and Rao, M. N. 1970, Nucl. Data Sheets 3:127.
Bucurescu, D., Ivascu, M., Semenescu, G. and Titirici, M. 1972, Nucl. Phys. A189:577.
Čujec, B. and Szöghy, I. M. 1969, Phys. Rev. 179:1060.
Gove, N. B. and Wapstra, A. H. 1972, Nucl. Data Tables 11:127.
Gunn, G. D. Fox, J. D., and Kekelis, G. J. 1976, Phys. Rev. C13:595.
Katsanos, A. A. and Huizenga, J. R. 1967, Phys. Rev. 159:931.
Kocher, D. C. 1976, Nucl. Data Sheets 18:463.
Lips, K. and McEllistrem, M. T. 1970, Phys. Rev. C1:1009.
Malik, B. F. and Scholz, W. 1966, Phys. Rev. 150:919.
Mateja, J. F., Neal, G. F., Goss, J. D., Chagnon, P. R. and Browne, C. P. 1976, Phys.
Rev. C13:118.
Mateja, J. F., Browne, C. P., Moss, C. E. and McGrory, J. B. 1977, Phys. Rev.
C15:1708.
McCullen, J. D., Bayman, B. F. and Zamick, L. 1964, Phys. Rev. 134:B513.
McGrory, J. B. 1967, Phys. Rev. 160:915.
O'Brien, B. J., Dorenbusch, W. E., Belote, T. A. and Rapaport, J. 1969, Nucl.
Phys. A104:609
Perey C. M. and Perey, F. G. 1974, Atomic Data and Nucl. Data Tables 13:293.
Peterson, R. J., Pittel, S. and Rudolph, H. 1971, Phys. Lett. 37B:278.
Peterson, R. J. and Rudolph, H. 1972, Nucl. Phys. A191:47.
Rapaport, J., Belote, T. A., Dorenbusch, W. E. and Doering, R. R. 1969, Nucl. Phys.
A123:627.
Osnes, E. 1971, Proc. Topical Conf. on the Structure of 1f7/2 Nuclei, 1971Legnaro, ed.
R. A. Ricci, Editrice Compositori, Bologna, p. 79;
Saayman, R. and Irvine, J. M. 1976,J. Phys. G2:309.
Scholz, W. and Malik, F. B. 1967, Phys. Rev. 153:1071.
Schulte, R. L. King, I. D. and Taylor, H. W. 1975, Nucl. Phys. A243:202.
Benson, H. G. and Johnstone, I. P. 1975, Can. J. Phys. 53:1715.
Tarara, R. W., Goss, J. D., Jolivette, P. L., Neal, G. F. and Browne, C. P. 1976, Phys.
Rev. C13:109.
Vervier, J. 1966, Nucl. Phys. 78:497.
Ward, T. E., Pile, P. H., and Kuroda, P. K. 1969, Nucl. Phys. A134:60.
Wiest, J. E., Robertson, C., Gabbard, F. and McEllistrem, M. T. 1971,Phys. Rev.
C4:2061.
28
Gamma De-excitation of 55Mn Following the 55Mn(p,p’γ)55Mn
Reaction
Key features:
55
1. The gamma de-excitation scheme of Mn has been determined for states of up to
3385 keV excitation energy.
2. A total of 45 γ transitions have been identified between 27 states in 55Mn.
3. The enigma of the 1292 keV level has been studied.
e. Our results confirm the γ de-excitation scheme of Hichwa et al. (1973a) and
Hichwa, Lawson, and Chagnon (1973b) around this excitation energy but do
not confirm the expected γ transition (Kulkarni 1976) to the ground state from
the 1292 level.
f. The incident proton energy of 7.975 MeV was chosen to coincide with the
energy used by Katsanos and Huizenga (1967) who reported a doublet of
states at the 1292 keV excitation energy in (p,p’) scattering. Our high-
resolution measurements do not confirm their results.
g. Combining the results of this study, with the results of our earlier study of the
52
Cr(4He,p)55Mn reaction at 18 and 26 MeV and the existing results by other
authors we conclude that there is a doublet of states at the 1292 keV
55
excitation energy in Mn with the separation energy of less than 10 keV.
However, the excitation of its 1/2- member appears to be strongly selective.
Abstract: Gamma de-excitation scheme has been studied using the 55Mn(p,p’γ)55Mn reaction
induced by the 7.975 MeV protons. A total of 45 γ transitions between 27 states in 55Mn,
extending up to the 3385 keV excitation energy, have been observed in both the singles and
p-γ coincidence spectra. The enigmatic 1292 keV level has been studied using both the
gamma and the high-resolution proton spectra.
Introduction
The study of gamma de-excitation of 55Mg was prompted by the enigma of the 1.29
MeV level. A great deal of confusion surrounded the level structure of 55Mn at about
this excitation energy. In a high resolution (p,p') study using 7.975 MeV protons
Katsanos and Huizenga (1967) claimed to have observed a doublet at this excitation
energy.
Peterson, Pittel and Rudolph (1971) studied the 57Fe(p,3He)55Mn and 57Fe(d,4He)
55
Mn reactions at Ep =27 MeV and Ed = 16.5 MeV. They claimed the existence of a
1 -
/2 state unresolved from a state with 11/2- at 1.29 MeV. This study is in agreement
with our study of 52Cr(4He,p)55Mn reaction using 18 and 26 MeV 4He particles (see
Chapter 27).
Using 4He-particles with the energy of 5 to 7 MeV, Kulkarni and Nainan (1974) found
γ-ray yields for a 1293 keV γ-ray which could be fitted by first order Coulomb
excitation theory for a λ = 2 transition. They claimed that they were exciting a state
at 1.293 MeV with 1/2- ≤ Jπ ≤ 9/2-. Observation of γ-ray transitions to the 7/2- and 9/2-
states at 0.128 and 0.983 MeV limited their spin assignments to 5/2- ≤ Jπ ≤ 9/2-. They
also found that the yields for the 307 and 1167 keV γ-rays, which originate from the
1.293 MeV level could be accounted for by Coulomb excitation theory.
Unfortunately a degree of uncertainty about their interpretation Coulomb excitation
data as pointed out later by Kulkarni (1976) who extended Coulomb excitation
measurements of 55Mn to 8 MeV. He observed the same γ-ray spectrum as in the
earlier publication (Kulkarni and Nainan 1974) except for the 307 and 1167 keV γ-
rays, which could be identified with the 304, and 1164 keV γ-rays observed at 8 MeV.
However, he found that the yield for the 1164 keV γ-rays rose too steeply to be
accounted for by a single E2 excitation and thus this γ-rays could not be identified as
the 1167 keV γ-rays of Kulkarni and Nainen (1974). He concluded that there were
two levels separated by 3 keV: a Jπ = 11/2- state at 1.290 MeV and a 1/2- state at
1.293 MeV. The assignment of spin 1/2- for the state at 1.293 MeV was made on the
basis of both the γ-rays yields and a limited γ-rays angular distribution measured at θγ
= 0° and 90°. According to Kulkarni (1976) the 1.293 MeV (1/2-) level de-excites
100% to the ground state.
Unfortunately, a 1293 keV γ-ray finds no confirmation in the publication of Kocher
(1976). In an earlier study, Hichwa et al. (1973a) and Hichwa, Lawson, and Chagnon
(1973b) saw no evidence for the population of a 1/2 level at 1.29 MeV in their (4He,pγ)
measurements at E =10.5 and 11.1 MeV. Their spin assignment agreed with the
previously assigned 11/2 value but they did not observe a 1292 keV γ-ray transition to
the ground state, which according to Kulkarni (1976) should be the only decay
pathway open from the alleged 1/2- state.
The branching ratios for the 1.29 MeV state γ-decays claimed by Kulkarni and
Nainen (1974) and Kulkarni (1976) are in serious disagreement with those of Hichwa
et al. (1973a) and Hichwa, Lawson, and Chagnon (1973b). It is also not clear
whether the 1293 keV γ-ray observed by Kulkarni and Nainen (1974) and Kulkarni
(1976) originates from a level at 1.29 MeV because they did not report particle-
gamma coincidence measurements or observe any γ-rays in coincidence with the
1293 keV γ-ray.
In our study of 52Cr(4He,p)55Mn reactions (see Chapter 27), the data at 26 MeV
showed a pronounced minimum and maximum at the same angles as observed for
other Jπ = 1/2- transitions. A least-squares combination of Jπ = 1/2- and 11/2- produced a
good fit to the measured angular distribution thus indicating a transition to a doublet
state at this excitation energy. The determined intensities of the two components at θ
=23° are 12 μb/sr for the Jπ = 1/2- state and 10 μb/sr for Jπ = 11/2-.
Experimental procedures and results

We have carried out our measurements using a 7.975 MeV proton beam from the
ANU EN tandem accelerator. The energy was chosen to coincide with the energy
used by Katsanos and Huizenga (1967) in their measurements of (p,p’) scattering.
The target contained a minimum of 99% 55Mn, with maximum limits of iron 0.002%,
lead 0.001%, nickel 0.002% and zinc 0.05%. Preliminary measurements were
performed using the 24" double focussing spectrometer. The best resolution attained
was 12 keV. Measurements of 55Mn(p,p’) spectra were made at 30°, 40°, 50° and
140° (lab). In contrast with the claim of Katsanos and Huizenga (1967) the particle
spectra showed no evidence of a doublet at the 1293 keV excitation energy. This
could either mean that the separation of the doublet states is significantly less than
12 keV, thus confirming the claim of Kulkarni (1976) or that one of the components of
the doublet is not excited in (p,p’) scattering.
We have then carried out the 55Mn(p,p'γ) measurements using the same proton
bombarding energy. A diagram of the experimental arrangement is shown in Figure
28.1. The γ-ray detector was made of 62 cm3 Ge(Li) crystal (4.9 cm diameter, 3.3 cm
long). The detector was located at 90° with respect to the beam to minimize Doppler
shift and the total acceptance angle was ±32°. To detect protons, Si surface barrier
detector was used, with a diameter of 1.9 cm. The detector was set at -80°. Magnetic
electron suppression was used and the detector was cooled by Cu strap connected
to cold finger. We have used a strip 55Mn target, 2 mm wide, approximately 150
μg/cm2, on a thin carbon backing.
Figure 28.1. Experimental arrangement for the 55Mn(p,p’γ)55Mn measurements
Figure 28.2. Electronics for the 55Mn(p,p’γ)55Mn measurements. 1. Si surface barrier detector; 2.
Ge(Li) 62 cm3 Seforad γ-detector; 3. ORTEC 125 preamplifier; 4. Seforad Sr 100 preamplifier; 5.
ORTEC 454 timing filter; 6. ORTEC 463 constant fraction discriminator; 7. Nanosecond delay; 8
Canberra 1443 time to amplitude converter; 9. :Logic shaper and delay; 10. Tennelec TC 203 BLR
amplifier; 11. Tenelec TC 205A amplifier; 12. Canberra 1454 linear gate and stretcher.
Singles γ-ray spectra were collected with the target in and out of the beam. The
spectra were calibrated using a 152Eu source located in the same position as the
target. A sample of singles spectrum is shown in Figure 28.3 for γ energies around
the alleged 1292 transition from the 1292 level to the ground state. No 1292 keV γ-
ray was observed.
According to Kulkarni and Nainan (1974) the branch to the ground state of the 1292
keV state should be five times more likely than the branch to the first excited state
which produces an 1165 keV γ-ray. The observed 1280 keV γ-ray has about 1/5 the
intensity of the 1165 keV γ-ray in Figure 28.3. If a 1292 keV γ-ray is present in this
spectrum its intensity is less than 1/5 that of the 1165 keV transition.
A list of γ-rays seen in singles, along with their relative intensities is presented in
Table 28.1. Gamma rays, which were present in the singles spectrum when the
target was removed from the beam, have not been included in the table.
Figure 28.3. A part of the 55Mn(p,p’γ)55Mn γ singles spectrum at Ep = 7.975 MeV around the alleged
1292 keV γ de-excitation to the ground state from the 1292 MeV level in 55Mn. The γ energies are in
keV.
The p-γ coincidence data were recorded event by event on magnetic tape and the
measurements took approximately 45 hours of running time with 2 to 8 nA on the
150 μg/cm2 55Mn target. Figure 28.4 shows the proton projection and an example of
the γ-ray spectrum in coincidence with the 1292 keV proton group. The TAC time
window was set to 200 ns and the true to chance ratio was 20/1 with a 12 ns FWHM
for the TAC peak.
From the γ coincidence spectrum presented in Figure 28,4 it is clear that no γ-ray
was detected at 1292 keV. Our measurements disagree with the conclusion of
Kulkarni (1976) but confirm the decay scheme of Hichwa et al. (1973a) and Hichwa,
Lawson, and Chagnon (1973b) for the 1292 keV level. We have also carried out the
coincidence γ-rays measurements for the other proton peaks but we have seen no
γ transition to the ground state from the 1292 keV level.
The p-γ coincidence scheme obtained from our measurements is shown in Figure
28.5. Strong and firmly assigned transitions are indicated by solid lines. Weak
transition, corresponding to low number of counts, are shown as dashed lines.
The p-γ coincidence scheme obtained from our work is in substantial agreement with
the scheme determined by Hichwa et al. (1973a) and Hichwa, Lawson, and Chagnon
(1973b) for levels below approximately 2800 keV. We have extended the scheme to
3385 excitation energy.
Figure 28.4. The proton spectrum (upper figure) and an example of the p-γ coincidence spectrum
(lower figure). The position of the alleged 1292 transition to the ground state is indicated but is absent
in the spectrum.
Table 28.1
Gamma rays observed in singles measurements for the 55Mn(p,p’γ)55Mn reaction at Ep = 7.975 MeV a)
Eγ ± ΔEγ RI ± ΔRI Eγ ± ΔEγ RI ± ΔRI
(keV) (keV) (keV) (keV)
129.7 .2 301 13 983.7 .5 250 35
158.2 .2 99 11 1165.1 .3 333 49
238.0 .2 147 22 1212.2 .3 567 75
273.3 .2 688 40 1221.9 .4 1622 162
307.4 .2 109 15 1236.9 .5 154 31
385.2 .2 534 44 1280.4 .5 68 20
411.3 .2 224 36 1315.8 .3 5542 512
416.7 .5 8.7 2.5 1326.5 .3 412 64
442.0 .2 221 33 1369.1 .2 1067 116
477.0 .2 3094 214 1378.6 1.3 105 48
482.9 .7 10 3 1407.7 .2 2067 215
532.0 .2 209 29 1419.2 .7 169 51
743.8 .6 65 21 1433.3 .3 958 127
765.1 .2 76 16 1459.8 .4 375 62
803.2 .2 2172 132 1505.2 .8 191 52
810.6 .2 352 38 1527.7 .4 678 103
826.7 .2 400 44 1554.7 .8 280 72
846.0 .2 1014 79 1572.0 .8 356 84
857.7 .2 1000 1620.7 .5 315 68
895.1 .2 252 55 1638.9 .4 531 94
910.5 .8 68 28 1663.4 .5 287 63
930.8 .2 7964 491 1882.4 .5 379 93
962.9 .8 103 30
a
) Known impurities or non-prompt γ-rays have been excluded.
RI – Relative intensity normalized to the 857.7 keV γ-ray.

We have carried out a study of the gamma de-excitation of 55Mn using the
55
Mn(p,p’γ)55Mn reaction at 7.975 MeV. We have observed 47 gamma transitions
between 27 states in 55Mn extending to 3385 keV excitation energy. In addition, we
have measured proton spectra using a 24” double focusing spectrometer. The
incident proton energy was chosen to coincide with the energy used by Katsanos
and Huizenga (1967) in their (p,p’) measurements. In contrast with their claim, our
high-resolution data do not show a doublet of states at 1292 keV excitation energy.
Figure 28.5. The gamma de-excitation scheme of 55Mn determined by our measurements. The
relatively strong and firmly determined γ transitions are indicated by solid lines. The dotted lines show
weak transitions.
Our gamma de-excitation measurements are in good agreement with the results of
Hichwa et al. (1973a) and Hichwa, Lawson, and Chagnon (1973b) around this
excitation energy. However, in contrast with the conclusion of Kulkarni (1976), our
results clearly demonstrate that there is no gamma transition between the 1292 keV
level and the ground state.
Combining our results for the gamma de-excitation, with our previous results on the
52
Cr(4He,p)55Mn reaction at 18 and 26 MeV and with available information based on
research by other authors we conclude that there must be a closely spaced doublet
of states at 1292 keV excitation energy. The energy spacing between the two
members of the doublet appears to be less than 10 keV. The excitation of the 1/2-
member of the doublet appears to be strongly selective and to depend on the
reaction used to access energy levels in 55Mn.
References
Hichwa, B. P., Lawson, J. C., Alexander, L. A. and Chagnon, P. R. 1973a, Nucl. Phys.
A202:364.
Hichwa, B. P., Lawson, J. C. and Chagnon, P. R. 1973b, Nucl. Phys. A215:132.
Katsanos, A. A. and Huizenga, J. R. 1967, Phys. Rev. 159:931.
Kocher, D. C. 1976, Nucl. Data Sheets 18:463.
Kulkarni, R. G. and Nainan, T. D. 1974, Can. J. Phys. 52:1676.
Kulkarni, R. G. 1976, Physica Scripta 13:213.
Peterson, R. J., Pittel, S. and Rudolph, H. 1971, Phys. Lett. 37B:278.
29
The 138Ba(7Li,6He)139La and 140Ce(7Li,6He)141Pr Reactions at 52 MeV
Key features:
4. Spectroscopic applicability of the single-proton (7Li,6He) reaction has been studied.
138 140
5. The target nuclei of Ba and Ce and have been selected to have a clean access
to a wide range of single-proton configurations: 1g7/2, 2d5/2, 2d3/2, 3s1/2, and 1h11/2.
6. A distinct j – dependence has been observed for the l = 2 transfer, which allowed to
distinguish between the 2d5/2 and 2d3/2 configurations.
7. Spin assignments have been made and spectroscopic factors have been extracted to
states in the residual nuclei 139L and 141Pr. They compare well with earlier studies and
thus show that the (7Li,6He) reaction can serve as a useful spectroscopic tool.
Abstract: Angular distributions have been measured for transitions to low-lying states in
139
La and 141Pr populated by the 138Ba(7Li,6He)139La and the 140Ce(7Li,6He)141Pr reactions at
E7Li = 52 MeV. Elastic scattering of 7Li at 52 MeV on 138Ba and 140Ce, and 6Li at 48 MeV on
139
La and at 47 MeV on 141Pr were measured to determine the interaction potentials in the
incident and outgoing channels. Optical-model parameters extracted from fits to the
scattering data were used in a finite-range distorted wave analysis of the measured angular
distributions for levels below 2.40 MeV excitation energy in 139La and 1.65 MeV in 141Pr.
Final-state spins have been assigned to levels in 139 La and 141Pr. The reaction cross
sections exhibit less structure than predicted by the distorted wave calculations, but the
extracted spectroscopic factors are generally in good agreement with light-ion results.
Introduction
Heavy-ion-induced, single-nucleon stripping reactions can be used to extract
spectroscopic information complementary to that obtained from light-ion work.
However, as the mass of the target-projectile system increases, bell-shaped angular
distributions centred at the grazing angle are observed and they have only limited
spectroscopic usefulness. Furthermore, heavy-ion studies employing 16O, 14N and 12C
projectiles have been limited by energy resolution to residual nuclei with large level
spacing.
To explore the spectroscopic applicability of heavy-ion-induced reactions we have
selected a lighter projectile 7Li. Combined with the available good resolution, reactions
induced by this projectile was be expected to provide useful spectroscopic information
for closely spaced states.
The shell model indicates that systems with 50 or 82 nucleons constitute unusually
tightly bound cores. Consequently, by using single-nucleon transfer reactions on
such nuclei one should be able to study conveniently states corresponding to
configurations outside closed cores.
For our study, we have chosen a single-proton transfer reaction (7Li,6He) and we
have selected 138Ba and 140Ce as target nuclei. These two isotopes have an N = 82
neutron core. They also contain 56 and 58 protons, respectively. Thus, 138Ba
contains 6 protons outside the closed Z = 50 proton core, and 140Ce contains 8. The
stripped proton may be expected to be deposited to any of the following orbitals:
1g7/2, 2d5/2, 2d3/2, 3s1/2, and 1h11/2. However, the 1g7/2 orbitals are nearly full in both
isotopes, particularly in 140Ce so it might be expected that most protons will be
transferred to other configurations. Nevertheless, the 1g7/2 should be also
accessible.
Experimental method and results

To have a complete set of data for the intended theoretical analysis we had to
measure not only the differential cross sections for the (7Li,6He) reactions but also
the elastic scattering cross sections in the incident and outgoing channels. However,
since 6He particles cannot be used as projectiles, we have measured the elastic
scattering of 6Li on the relevant target nuclei.
- -
Beams of 6Li and 7Li from a General lonex sputter source were injected into the
Australian National University 14UD Pelletron accelerator. Beam currents of up to
300 nA of 6 Li +3 and 7Li+3 were obtained on the target.
Targets of enriched 138Ba (> 99%) and 140Ce (> 99%) and natural 139La and 141Pr,
comprised of metal on thin carbon backings, proved to be extremely fragile and
many ruptured before they could be removed from the vacuum system in which they were
prepared. Others broke whilst standing in vacuum storage. Fortunately, at least one
target of each material survived both preparation and beam bombardment. However,
the thickness of surviving targets was small, only about 25 μg/cm2. We could have
had acceptable resolution with significantly thicker targets of about 100 -150 μg/cm2.
Reaction data and elastic scattering data were measured with an Enge split-pole
spectrograph using a resistive-wire gas proportional detector (Ophel and Johnston
1978) located in the focal plane. From the energy loss (ΔE) and the position signal
( ∝ Bρ ) of the focal plane detector, a mass identification signal ( M 2 = ( Bρ ) 2 ΔE ) was
obtained. The difference in magnetic rigidity between 6Li3+ and 6He2+ was sufficient to
allow for unambiguous mass identification. Additionally, the high field necessary to
bend the 6He particles onto the detector completely removed the 7Li3+ elastic events
from the detector, allowing high beam currents to be used at forward angles. Fixed
monitor detectors at 15° and 30° were used for normalization between runs.
To obtain the best possible information about the wave functions in the incident and
outgoing channels the following elastic scattering measurements have been carried
out: 138Ba(7Li,7Li)138Ba and 140Ce(7Li,7Li)140Ce at E(7Li) = 52 MeV and
139
La(6Li,6Li)139La at E(6Li) = 48 MeV and 141Pr(6Li,6Li)141Pr at E(6Li) = 47 MeV. The
energies for 6Li projectiles were chosen to correspond to the average outgoing 6He
energy.
Absolute cross sections were obtained by normalizing the forward angle elastic
scattering to the Rutherford cross sections. The error in the absolute normalization is
estimated to be 5% for the elastic scattering, resulting mainly from angle setting and
dead time uncertainties. Based on the reproducibility of the (7Li,6He) data, the
absolute cross sections for the transfer reactions are accurate to ±12%. The relative
errors in the cross sections are shown by the error bars on the individual data points
where these are larger than the plotted points.
Figures 29.1 and 29.2 show spectra for the 138Ba(7Li,6He)139La reaction at θlab = 27°,
and the 140Ce(7Li,6He)141Pr reaction at θlab = 20°, respectively. The resolution is 70 keV
FWHM and little background is evident at these angles. However, at angles forward of 8°
(lab), background from impurities in the target was larger, but it did not prevent
extraction of the data.
The distributions for the elastic scattering and for the (7Li,6He) reaction are shown in
Figures 29.3 – 29.7.
Figure 29.1. A 6He spectrum for the 138Ba(7Li,6He)139La reaction at 270. States in 139
La are labelled
with the appropriate excitation energies.
Figure 29.2. A 6He spectrum for the 140Ce(7Li,6He)141Pr reaction at 200. States in 141
Pr are labelled
The elastic scattering data were analysed using a simple (central) optical model
potential combined with the usual description of the Coulomb interaction:
U (r ) = Vc (r ) − Vf (r , r0 , a0 ) − iWg (r , r0′, a0′ )
where,
{ [(
f (r , r0 , a0 ) = 1 + r − r0 At
1/ 3
)/ a ]}
0
−1
{ [(
g (r , r0′, a0′ ) = 1 + r − r0′At
1/ 3
)/ a′ ]}
0
−1
Vc (r ) the Coulomb potential between a point projectile charge and the field of a
uniformly charged sphere of radius Rc = rc A1 / 3 .
Vc (r ) =
[ (
Z p Z t e 2 3 − r / rc At
1/ 3
)] for r ≤ rc At
1/ 3
1/ 3
2rc At
Z p Zt e2
Vc (r ) = for r > rc At
1/ 3
r
Zp and Zt are the projectile and target charge.
138 7 7 138
Fig. 3. Angular distributions for the Ba( Li, Li) Ba elastic scattering at 52 MeV and
139
La(6Li,6Li)139La at 48 MeV. The solid lines are the optical-model fits to the data.
Figure 29.4. Angular distributions for the 140Ce(7Li,7Li)140Ce elastic scattering at 52 MeV and for
141
Pr(6Li,6Li)141Pr at 47 MeV. The solid lines are the optical-model fits to the data.
Table 29.1
Nuclear and Coulomb potential parameters used to fit the elastic scattering and the (7Li,6He) reaction
angular distributions
The computer code JIB (Perey 1967), which I have earlier modified and adapted to run
at ANU and which I have used in analyses of scattering of light projectiles, was now
used to fit the elastic scattering data for heavy projectiles. The parameters were varied
two at a time until a minimum χ2 was obtained. The experimental angular distributions
and the optical-model fits are shown in Figures 29.3 and 29.4. The extracted
parameters are listed in Table 29.1.
These parameters were then used in the exact finite-range (EFR) distorted wave calcu-
lations using the computer code LOLA (DeVries 1973) for transitions to the strongly
populated states observed in the 138Ba(7Li,6He)139La and 140Ce(7Li,6He)141Pr reactions.
The wave functions of the bound proton were generated with Woods-Saxon
potentials whose depths were adjusted to give the correct binding energies. The
ground-state binding energies are -9.978 MeV for p + 6He, -6.201 MeV for p+138Ba
and - 5.227 MeV for p + 140Ce (ND 1971). The shape parameters were fixed at r0 =
1.25 fm and a0 = 0.65 fm. The experimental angular distributions and the EFR-
distorted wave fits to the data are shown in Figures 29.5 – 29.7. Generally, it is easy
to distinguish angular distributions belonging to different orbital angular momenta.
However, transfer of a proton to orbitals outside the closed shell Z = 50 involves two l
= 2 configurations, 2d3/2 and 2d5/2. Fortunately, there is a clear j - dependence for
these configurations. Calculations for 2d3/2 and 2d5/2 are shown in Figures 29.5 and
29.6 for states at 0.166 MeV, 1.56 MeV, 1.85 MeV and 1.96 MeV in 139La, and in
Figure 29.7 for the ground state in 141Pr. Clearly the data forward of 6° (c.m.) allow
unambiguous distinction between 2d3/2 and 2d5/2 final states.
Spectroscopic factors are extracted using the following relationship (DeVries 1973)
⎛ dσ ⎞ 2J +1 1 ⎛ dσ ⎞
⎜ ⎟ = B
⎝ dΩ ⎠ exp l
∑
l
(2l + 1)W 2 (l1 j1l2 j2 ; l )C 2 S ( 7Li )C 2 S B ⎜
2
⎟
⎝ dΩ ⎠ LOLA
where B refers to the states in the final nuclei, and subscripts 1 and 2 refer to the
projectile and final nuclei. The l and W are the transferred orbital angular momentum
and the appropriate Racah coefficient, respectively. The C 2 S (7Li ) is the overlap of 7Li
with 6He + p in a p3/2 state and was taken from Cohen and Kurath (1967) to be 0.59.
Extracted spectroscopic factors are obtained by normalization in the region of the
grazing angle (25°-29° for 138Ba and 26°-30° for 140Ce).
The absolute spectroscopic factors obtained from the analysis are listed in Tables 29.2
and 29.3, which also show the spectroscopic factors obtained from the (3He,d) reac-
tions (Wildenthal, Newman, and Auble 1971). The errors in the absolute spectroscopic
factors include the uncertainty in the absolute normalization of the experimental data
and statistical errors. Relative spectroscopic factors, normalized to the ground state,
are also listed in Tables 29.2 and 29.3.
138 7 6 139
Figure 29.5. Angular distributions for states populated in the Ba( Li, He) La reaction. The solid and
dashed lines are the EFR-distorted wave calculations normalized to the data. The 1.77 MeV doublet
is shown using spectroscopic strengths obtained from the (3He,d) work of Wildenthal, Newman, and
Auble (1971) (solid line) and spectroscopic strengths obtained from a best fit to the data (dashed line). The
pure s1/2 and d3/2 components are shown below the data.
Figure 29.6. Angular distributions for states populated in the 138Ba(7Li,6He)139La reaction. The solid
and dashed lines are the EFR-distorted wave calculations normalized to the data. The 2.24 MeV data
are shown with the spectroscopic strengths obtained from a best fit to the data using a sum of d3,2 and
g7/2 contributions. The pure d3/2 and g7/2 components are shown below the data.
Figure 29.7. Angular distributions for states populated in the 140Ce(7Li,6He)141Pr reaction. The solid
and dashed lines are the EFR-distorted wave calculations normalized to the data. The 1.60 MeV and
1.65 MeV states could not be resolved, and are shown using spectroscopic strength ratios obtained from the
(3He,d) work of Wildenthal, Newman, and Auble (1971) (solid line) and spectroscopic strengths obtained
from a best fit to the data (dashed line). The pure s1/2 and d3/2 components are shown below the data.
As can be seen from Figures 29.5 – 29.7, the EFR-distorted wave calculations describe
the data well. Angular distributions for transitions to 3s1/2, levels via the l = 1 transfer at
1.21 MeV and 2.31 MeV in 139La, and 1.30 MeV in 141Pr are well reproduced by the
calculations but are slightly out of phase, a problem which has been observed in other
(7Li,6He) reactions (Moore, Camper, and Charlton 1970) where the data and the
calculations are seriously out of phase.
The states of the unresolved doublet at 1.77 MeV in 139La are described as 3s1/2, and
2d3/2 in the (3He,d) reaction (Wildenthal, Newman, and Auble 1971). The solid line in
Figure 29.5 represents the fit to the data obtained by summing the calculated
differential cross sections for the two configurations and using spectroscopic
strengths as derived in the (3He,d) work. The dashed line represents a similar fit but
obtained by allowing the spectroscopic strengths to be adjusted by the least-squares
fitting procedure.
Table 29.2
Spectroscopic information for 139La
a) Wildenthal, Newman, and Auble (1971). b) Our work.

E – Excitation energy. nl j – Single particle configuration. Absolute/Relative – The
absolute and relative values of the spectroscopic factors, respectively.
Table 29.3
Spectroscopic information for 141Pr
See notes to Table 29.2.
States at 1.60 and 1.65 MeV in 141Pr could not be resolved in our experiment. These
states are described as 2d3/2 (1.60 MeV) and 3s1/2 (1.65 MeV) in the (3He,d) reaction
(Wildenthal, Newman, and Auble 1971). Figure 29.7 shows the best fits obtained either
by taking a sum of the calculated distributions with the spectroscopic strengths as
derived in the (3He,d) work for the two configurations (the solid line) or by adjusting the
spectroscopic strengths in the least-squares fitting procedure (the dashed line). If the
calculated distribution for the 3s1/2, transition to the 1.65 MeV state is out of phase with
the experimental distribution, as it is for the l = 1 transitions to the 1.21 and 2.31 MeV
states (see above), this would strongly affect the fitting procedure and hence the
spectroscopic factors. Thus the spectroscopic factors for these states should be
viewed with caution.
The state at 2.24 MeV in 139La is described as a d - state in the (3He,d) reaction
(Wildenthal, Newman, and Auble 1971). However, neither calculations for a 2d3/2 nor a
2d5/2, state gave a satisfactory fit. A level at 2.232 MeV has been observed in (γ,γ’)
work (Moreh and Nof 1970) and assigned tentative spins of either 7/2+ or 11/2-. The
best fit for the studied here (7Li,6He) reaction has been obtained using a sum of 2d3/2
and 1g7/2 (see Figure 29.6).
Summary and conclusion

We have carried out high resolution measurements for the 138Ba(7Li,6He)139La and
140
Ce(7Li,6He)141Pr reactions at 52 MeV incident 7Li energy. In addition, we have also
measured angular distributions for the elastic scattering 138Ba(7Li,7Li)138Ba and
140
Ce(7Li,7Li)140Ce at E(7Li) = 52 MeV and 139La(6Li,6Li)139La at E(6Li) = 48 MeV and
141
Pr(6Li,6Li)141Pr at E(6Li) = 47 MeV.
We have analysed the elastic scattering data using standard optical model potential
and we have then applied the derived parameters in our exact finite range distorted
wave analysis of transfer reaction data. The calculations described the transfer
angular distributions well and only slight phasing problems were encountered for s1/2
states.
The absolute spectroscopic factors obtained in our study are generally larger than
those obtained from the light-ion (3He,d) reactions (Wildenthal, Newman, and Auble
1971), but the relative spectroscopic factors are in good agreement.
Heavy-ion forward-angle j - dependence has been observed and used to assign the
following spins to d - states in 139La; 0.166 (5/2+), 1.56 (5/2+), 1.85 (3/2+), 1.96 (3/2+);
and to the ground state of 141Pr, (5/2+). The spin of the d - state at 1.60 MeV in 141Pr
could not be assigned with certainty because levels at 1.60 MeV and 1.65 MeV were
unresolved. Previous spin assignments for the d - state levels (Lederer and Shirley
1978) are given as 0.166 (5/2+), 1.56 (3/2+), 1.85 (3/2+) and 1.96 (3/2+) in 139La; and 0.0
(5/2+), 1.60 (3/2+) in 141Pr.
The state at 2.40 MeV in 139La was not observed in the (3He,d) work (Wildenthal,
Newman, and Auble 1971) but has been seen in (α,α') scattering (Baker and Tickle
1972) and assigned a negative parity. Our distorted wave calculation with 11/2- spin
for this state and corresponding to the only allowed negative-parity configuration
1h11/2, is shown in Figure 29.6. However, the statistical errors in the data points
prohibit the definite assignment of any spin to this state.
In conclusion, the (7Li,6He) has been shown to be a functional tool which can be
used in determining final-state spins and spectroscopic factors.
References
Baker, F. T. and Tickle, R. 1972, Phys. Rev. C5:182.
Cohen, S. and Kurath, D. 1967, Nucl. Phys. A101:1.
DeVries, R. M., 1973, Phys. Rev. C8:951.
Lederer, C. M. and Shirley, V. S. 1978, Table of isotopes, 7th Edition.
Moore, G. E., Kemper, K. W. and Charlton, L. A. 1975, Phys. Rev. C11:1099.
ND 1971, Nucl. Data A9:305.
Ophel, T. R. and Johnston, A. 1978, Nucl. Instr. 157:461.
Perey, F. G. 1967, Phys. Rev. 131:745.
Wildenthal, B. H., Newman, E. and Auble, R. L. 1971, Phys. Rev. C3:1199.
30
Nuclear Molecular Excitations
Key features:
1. We have measured angular distributions and excitation functions for the

24
Mg(16O,12C)28Si reaction. The aim was to study nuclear molecular excitations.
2. Three new broad resonances have been discovered in the excitation functions for the
ground state and the first excited 21+ state in 28Si.
3. The analysis of our experimental results combined with a compilation of earlier results
shows that the observed resonances can be explained as nuclear molecular
excitations in either incident or exit channels, i.e. as orbiting nuclear states of the
24
Mg+16O or 28Si+12C systems.
24
Abstract: Excitation functions for the reaction Mg(16O,12C)28Si leading to the ground state
+ 28
and the first 21 excited state in Si were measured at 5°(lab) in the energy range 32.4 < Ec.m.
< 48.6 MeV. Although the resonant structure, previously observed at lower energies, becomes
progressively weaker, three new correlated maxima have been observed at Ec.m. = 37.5, 40.2
and 43.5 MeV. Attempts to find a consistent optical-model fit to the elastic scattering in the
entrance channel and an exact finite-range distorted wave fit to the transfer reaction cross
sections in this energy range were unsuccessful. Such a failure is to be expected if strong
coupling between the elastic and inelastic channels in either the initial or final system is
present. By comparing the angular distribution with the Legendre polynomial
distributions, PJ (cos θ ) , spin assignments J = 27, 29 and 31 were made for the three
2
observed resonances. The observed resonant behaviour can be explained as nuclear

molecular excitations.
Introduction
Considerable resonance structure has been observed in heavy-ion reactions. Typical
gross structure (with width ~2 MeV) has been seen in the excitation functions for the
(16O,16O) elastic scattering at energies above the Coulomb barrier (Maher et al.
1969). Optical-model analyses (Gobbi et al. 1973; Maher et al. 1969) of these data led
to either a shallow and weakly absorbing four-parameter potential or a surface-
transparent six-parameter potential in which the imaginary well has smaller radius
and diffuseness parameters than the real well. Such potentials allow the two colliding
ions to retain their individual structure during a grazing collision and give rise to the
possibility of so-called orbiting molecular states. However, because such collisions are
essentially a direct process, the ions soon pass through so that they stay in the
molecular orbit for perhaps only about 1/3 of a revolution. Nevertheless, even such
brief nuclear molecular configuration appears to have significant influence on the
measured excitation functions and angular distributions. Quantum mechanically, such
states correspond to virtual broad shape resonances in the ion-ion potential.
Consequently, they give rise to a broad resonance (gross) structure of the type seen in
the 16O + 16O elastic scattering.
The surface-transparent potentials cause the lower partial waves to be strongly
absorbed while trajectories corresponding to grazing collisions are only weakly
absorbed. These properties lead to two interesting phenomena: (i) typical diffraction
effects associated with the strong absorption and (ii) the "glory" effect arising from
orbiting trajectories in the weakly absorbing surface region.
The diffraction model of Austern and Blair (1965) has been employed (Phillips et al.
1979) to show that diffraction effects can lead to gross structure of the kind observed in the
12
C +12C and 16O + 16O inelastic scattering excitation functions. However, as pointed
out by Friedman, McVoy, and Nemes (1979), the presence of strong internal
absorption does not exclude the possibility of resonance effects. Indeed, in a full
quantum mechanical treatment, the gross structure should arise from both the
diffraction and orbiting resonance effects of the optical potentials involved.
The glory effect arises through interference of a normal backward scattered wave with
one, which has orbited through a negative deflection angle of about 180°. This leads to
large back-angle elastic scattering cross sections which display large oscillatory
character such as is observed in 16O + 28Si elastic scattering at Ec.m. = 35 MeV (Braun-
Muzinger et al. 1977). These large-angle oscillations often follow the square of a
Legendre polynomial, PJ2 (cosθ ) , suggesting a resonating partial wave of the order of
J. The J - values extracted in this way lie close to the grazing angular momenta
predicted by the appropriate surface-transparent optical potential.
Since the discovery of such phenomena, several descriptions involving Regge poles
(Braun-Muzinger et al. 1977), angular momentum dependent absorption potentials
(Chatwin et al. 1970), resonances (Barrette et al. 1978; Malmin et al. 1972) or parity-
dependent optical potentials (Dehnhard, Shlolnik and Franey 1978) have been
proposed. All these approaches, some of which are closely related, are designed to
enhance one or more partial waves close to the grazing angular momentum.
In some heavy-ion reactions, the cross sections are rapidly fluctuating functions of
energy and the question arises whether these are true resonance structure or simply
statistical Ericson fluctuations. In some cases, e.g. for the 12C + 14N system (Hansen
et al. 1974; Olmer et al. 1974), statistical calculations using the Hauser-Feshbach
method (Hauser and Feshbach 1952) give a good description of such fine structure
provided one subtracts out the underlying gross structure. However, in several cases,
e.g. in the 12C +16O system at Ec.m. = 19.7 MeV, there exists a strong correlation
between the excitation functions of the elastic scattering at several angles, suggesting
a resonance. Such resonance has a width of ~ 0.4 MeV and is an example of
intermediate structure.
Correlated structure of this kind with the width of ~ 0.1 MeV was found in the very first
precision heavy-ion measurements (Bromley, Kuehner, and Almqvist 1960) for the 12C +
12
C scattering. In addition to the widths, the spacings of the observed resonances in
this system are too small to be readily described in terms of simple shape
resonances associated with quasi-bound states in a molecular-type potential between
the two ions. Such intermediate structure has been interpreted (Imanishi 1968, 1969;
Michaud and Voght 1969, 1972) in terms of "doorway" states in which the incident
channel couples to another degree of freedom of the resonating system. In particular,
Imanishi (1968, 1969) proposed that the incident elastic channel may be strongly
coupled to a channel in which one of the 12C nuclei is excited to its first 2+ state at 4.4
MeV. This concept was extended by Scheid, Greiner and Lemmer (1970) and Fink,
Scheid and Greiner (1972) in their double resonance mechanism in which a virtual
state in the entrance channel is excited by a grazing partial wave and acts as a
doorway state which feeds quasi-bound states in inelastic channels corresponding to
excitation of collective states of the individual nuclei. In these approaches, the
intermediate structure is described in terms of coupling between the elastic and
inelastic channels and arises naturally in appropriate coupled-channels calculations (Fink,
Scheid and Greiner 1972; Imanishi 1968, 1969; Scheid, Greiner and Lemmer
1970).
In order to predict the energies and spins of possible intermediate structure resonances
arising from such coupling, Abe (1977) and Kondo, Matsuse and Abe (1978) have
suggested a schematic band-crossing model. In this picture, the resonance structure
arises whenever two quasi-rotational bands of states in the composite system,
corresponding to the elastic and inelastic channels, cross each other and the
molecular-type states involved are neither too narrow nor too broad.
If the above intermediate structure occurs in the grazing partial waves, then similar
resonance-like behaviour is to be expected even at forward angles in direct reactions,
which are strongly surface peaked. Indeed, pronounced resonance structure has been
observed by Paul et al. (1978) for the reaction 24Mg(16O,12C)28Si leading to the ground state
and 1.77 MeV 2+ state of 28Si at two forward angles, 0° and 11° (lab), for 23 ≤ Ec.m. ≤ 38
MeV.
The purpose of our work was to extend these results for the 24Mg(16O,12C)28Si reaction
to higher bombarding energies to determine if the strong resonance structure persists
and in this way to study further the nature of these resonances.

Thin targets (~ 100 μg/cm2) were made by evaporating enriched 24Mg (99.92 % 24Mg,
0.06 % 25Mg and 0.02 % 26Mg) onto a thin carbon backing (~ 10 μg/cm2). The targets
were bombarded with a 16O beam from the Australian National University 14 UD
Pelletron accelerator. The reaction products were momentum analysed using an
Enge split-pole magnetic spectrometer and were detected in a multi-electrode focal
plane detector. The particles were identified using the ratio ( Bρ ) 2 / E where Βρ is the
magnetic rigidity and E is the energy of the detected ions.
The data were recorded event by event onto magnetic tapes using a HP-2100 data
acquisition system. The incident beam intensity was recorded using a beam current
integrator. In addition, two solid-state detectors at 15° and 30° were employed as
monitors.
The elastically scattered 16O7+ ions and the most intense group of 12C ions ( 12 C 6+ )
from the transfer reaction have a similar magnetic rigidity. Thus the transfer reaction
measurements at forward angles were carried out using a 20 cm slot in front of the
focal plane detector to stop all groups of elastically scattered 16O projectiles while
allowing the 12C6 + ions to enter the detector. In the energy and angular regions studied
in the present experiment, the 12C6+ group contains 82-98% of the total intensity of the
12
C charge state distribution.
Excitation functions for the reaction 24Mg(16O,12C)28Si were measured at 5° (lab) in the
energy range 54-81 MeV (lab) with a horizontal acceptance of 4.5°(lab). Test
measurements carried out by varying the reaction angle around 5° for various incident
16
O energies indicated that the second maximum in the ground-state angular
distribution was located well within the acceptance angle. Figure 30.1 shows our results
together with earlier data of Paul et al. (1978), which were taken at 0° and 11°(lab) for the
ground-state transition at lower 16O bombarding energies. It can be seen that the resonant
structure becomes progressively weaker at higher energies. Nevertheless, three
additional correlated maxima are evident near Ec.m. = 37.5, 40.2 and 43.5 MeV.
Figure 30.1. Excitation functions (dots) for the reaction 24Mg(16O,12C)28Si leading to the ground state
28
and the first excited state in Si, measured in the range of energies 32.4-48.6 MeV (c.m.) at 5° (lab)
are compared with the excitation functions measured at lower energies (Paul et al. 1978) at 0° (full
line) and 11° (dash-dot line). Results of Paul et al. (1978) are expressed in arbitrary units. The dashed
lines are used to guide the eye. Arrows indicate the energies at which angular distributions were
measured. The vertical scale is for the elastic scattering. The data for the inelastic scattering have
been displaced to show the structure. The cross sections at around 32 MeV (c.m.) are nearly the
same for both excitation functions.
Figure 30.2. Angular distributions for the reaction 24Mg(16O,12C)28Si (0+, g.s.) measured at the
indicated c.m. energies (points) are compared with the distorted wave calculations (full lines).
Figure 30.3. Angular distributions for the reaction 24Mg(16O,12C)28Si (2+, 1.77 MeV) measured at the
indicated c.m. energies (points) are compared with the distorted wave calculations (full lines).
Angular distributions for the reactions were measured at these three energies and at
Ec.m. = 39.0 MeV, corresponding to a trough between two maxima in the excitation
functions. The energies at which angular distributions were measured are indicated
by arrows in Figure 30.1. An acceptance angle of 1° (lab) was employed for these
measurements. The results are shown in Figures 30.2 and 30.3 where diffraction
patterns are clearly evident for both transitions. However, some irregularities are
present. In particular, a prominent distortion of the simple oscillatory structure occurs for
the ground state distribution at 40.2 MeV in the angular range of 25-40°(c.m.). This
irregularity does not occur for the transition to the 2+ state.
The elastic scattering cross sections for the reaction 24Mg(16O,16O)24Mg were
measured at Ec.m. = 37.5 and 40.2 MeV with the whole detector exposed to the
reaction products. The data are shown in Figure 30.4. It can be seen that while the 37.5
MeV measurements exhibit oscillating structure for θc.m. > 40°, the 40.2 MeV data are
relatively structureless in this angular region.
Analysis of the elastic scattering

As pointed out by Siemssen (1977), two schools of thought exist regarding the
description of heavy-ion scattering in terms of the nuclear optical model. The first
argues that as heavy ions are complex loosely bound particles, they must
disintegrate upon impact so that only strongly absorbing potentials are acceptable.
The second school of thought, however, simply attempts to determine empirically
which features can or cannot be consistently described by the optical model. This
second approach has typically led to a range of potentials, which are weakly
absorbing for surface partial waves. Such "surface-transparent" potentials are often
characterized by a smaller geometry for the absorption potential than the real
potential i.e. r0′ < r0 and a0′ < a0 .27
27
Optical model parameters are as defined in Chapter 29.
Previously, elastic scattering for the 16O + 24Mg system has been studied for various
energies ranging from the Coulomb barrier up to Ec.m.. ≈ 43 MeV. In general, analyses
above the Coulomb barrier favour a moderately shallow real potential depth, which
increases linearly with energy (V ≈ 5 + 0.5Ec.m. MeV), and an imaginary potential
strength, which increases quadratically with energy.
Figure 30.4. Angular distributions for the 24Mg('6O,16O)24Mg elastic scattering measured at two
energies (dots) are compared with the optical-model calculations. Parameter sets are listed in Table
30.2.
These and similar optical-model analyses for neighbouring mass systems have led to
parameters, which can be classified broadly into three groups:
(i) Potentials, which have similar real and imaginary radii and diffuseness (i.e.
r0′ ≈ r0 and a0′ ≈ a0 ) and are strongly absorbing (W is relatively large).
(ii) Potentials, which have similar real and imaginary potential geometries but have
weak absorption strengths.
(iii) Potentials which have r0′ < r0 and a0′ < a0 and moderate absorption strength.
Both potentials (ii) and (iii) give rise to surface transparency for the grazing partial
waves. Examples of these three types of potentials is given in Table 30.1.
In our work, the sets A, E, and F of Table 30.1 were taken as representative potentials for
the three types of interactions and were used as starting sets in parameter searches
to fit the 16O + 24Mg elastic scattering angular distributions at Ec.m. = 37.5 and 40.2
MeV. These data were analysed using the optical-model parameter search code
SOPHIE (Delic 1975) to obtain parameter sets for the distorted wave analysis of the
24
Mg(16O,16O)24Mg transfer reaction. The searches were unconstrained and had led to
potentials shown as sets 1-3 and 6 and 8 in Table 30.2.
Table 30.1
Classification of heavy-ion optical-model potentials a)
a
) Potentials are as defined in Chapter 29. Set 1: Braun-Muzinger et al. 1973; Set 2: Tserruya
et al. 1975; Set 3: Siwek-Wilczynska, Wilczynski, and Christensen 1974: Set 4: Siemssen
1971; Set 5: Ball 1975; Set 6:Lemaire et al. 1974a; Set 7: Siemssen et al. 1969.
Table 30.2
Optical-model potentials for the 24Mg(16O,12C)28Si reaction
Sets 1,2, and 3 were obtained using sets A, E, and F of Table 30.1 as starting
values. The same applies to sets 6, 7, and 8. Sets 4 and 9 were obtained using
sets 2 and 7, respectively, as the starting values. Sets 5 and 10 were obtained
using set G of Table 30.1 as starting values. Set 11 gives the best description of
the angular distributions for the 24Mg(16O,12C)28Si reaction as shown in Figures
30.2 and 30.3.
In attempting to fit the 40.2 MeV data with a shallow imaginary well, the automatic search
converged on an unphysically large value for the imaginary diffuseness parameter
a0′ = 1.12 fm. The set 7 shown in Table 30.2 was obtained by requiring that a0′ ≤ 0.6
fm.
The 37.5 MeV data are described equally well by either a strongly absorbing potential
(set 1) or a weakly absorbing surface-transparent interaction (set 2). Both sets of
parameters generate the necessary oscillatory structure for θc,m. > 40° (Figure 30.4).
The fits obtained for the 40.2 MeV data are distinctly poorer in that all the potentials
predict oscillatory structure in the differential cross section for θc,m. > 40°, which unlike
the 37.5 MeV measurements, is not observed at the higher energy.
From other data, such as few nucleon transfer reactions, there is an evidence
(Lamaire et al. 1974b) of a preference for surface transparency of the optical
potential with a0′ < a0 . This was taken into account in the present analysis when using
sets 2 and 7 (Table 30.2) as starting values for further searches in which the value of
a0′ , was constrained to be ≤ 0.3 fm. The resulting potentials are given in Table 30.2 as
sets 4 and 9. The predictions for these sets together with the corresponding strongly
absorbing potentials (sets 1 and 6) are shown in Figure 30.4. It can be seen that both
types of the 16O + 24Mg interaction describe the 37.5 MeV data well while neither
reproduces the structureless 40.2 MeV angular distribution.
In some analyses, very shallow real potentials have been found. This possibility was
investigated by employing the shallow potential set G of Table 30.1 as a starting point
for a search in which the real strength V was constrained to have a value < 20 MeV.
The resulting potentials from these searches are presented as sets 5 and 10 in Table
30.2. These interactions give excellent fits to the data, particularly to the structureless
angular distribution at 40.2 MeV. However, the shallow potential, which describes the
37.5 MeV data, is significantly different from the potential, which fits the 40.2 MeV
measurements. In particular, the absorption strengths are 11.79 and 24.47 MeV,
respectively. Such a rapid increase in the imaginary potential is much larger than one
expects even if a quadratic energy dependence (Siwek-Wilczynska, Wilczynski, and
Christensen 1974) is assumed.
The distorted wave analysis

Exact finite-range (EFR) distorted wave calculations were carried out for the four-
nucleon transfer reaction 24Mg(16O,12C)28Si leading to the ground and 1.77 MeV states
of 28Si using the code LOLA (DeVries 1972). In the calculations, an α-cluster transfer
from 16O to 24Mg nuclei was assumed. The corresponding bound-state wave functions
in the projectile and residual nuclei were calculated in a Woods-Saxon potential with
radius 1.25 A1 / 3 fm and diffuseness 0.65 fm, the depths being adjusted to obtain the
experimental α-particle separation energies in 16O and 28Si.
In our analysis, an attempt was made firstly to describe the reaction data for the
ground-state transition at 40.2 MeV using the optical-model parameters obtained
from the elastic scattering analysis. For simplicity, the same parameters were
employed for both the entrance and exit channels. Such calculations gave a poor
description of the measurements. However, a better fit to the angular distribution was
obtained by adjusting the parameters of set 9 (Table 30.2) although similar attempts
based upon parameter sets 6 and 8 were unsuccessful.
Figure 30.2 shows the best fits to the experimental angular distributions. The
corresponding set of parameters is listed as set 11 in Table 30.2. The theoretical cross
sections were normalized using factors listed in Table 30.3 to bring them in line with the
experimental data. These numbers, which correspond to a product of spectroscopic
factors (see Chapter 29), are expected to be energy independent if the distorted wave
theory is valid. As can be seen from Table 30.3, the normalization factors are not
constant.
The significant energy dependence of the spectroscopic factors, as implied by such
normalization factors, has been previously observed for this reaction at lower
energies (Peng et al. 1976). While it is possible that the observed energy
dependence could be at least partially removed by allowing the optical-model
parameters to be smoothly energy dependent it is unlikely that the over-all shapes of
the angular distributions could be reproduced at the same time. Moreover, the excitation
functions calculated using such a smoothly energy-dependent potential would not
exhibit rapid fluctuations of the type displayed in Figure 30.1.
Figure 30.2 shows that the calculated curves describe qualitatively the strong oscillatory
character at very forward angles and the smoother angular distribution for θc.m. > 25°.
However, parameter set 11 gives a poor description of the elastic scattering data at
40.2 MeV so there is a consistency problem.
Table 30.3
Normalization factors (DeVries 1973) for the 24Mg(16O, 12C)28Si reaction
Figure 30.3 shows the distorted wave results for the 24Mg(16O,12C)28Si reaction
leading to the 1.77 MeV 21+ state in 28Si. These curves were also calculated using the
parameter set 11. As can be seen, in this case, theory and experiment are out of
phase. In order to remove this gross discrepancy, large changes in the optical-model
parameters, away from those that describe the ground-state transition, would be
necessary. In view of this problem and the over-all inconsistency of the optical model and
distorted wave analyses, it was not considered worthwhile to attach much significance
to the normalization factors of Table 30.3 or to attempt any determination of the
alpha-nucleus spectroscopic factors.
Discussion
The angular distributions for the 24Mg(16O,12C)28Si reaction to the ground state of 28Si
at Ec.m.. = 37.5, 40.2 and 43.5 MeV, corresponding to peaks in the excitation function
for θ = 5°(lab), exhibit strong oscillatory structure at forward angles (θc.m. < 25°), which
can be described reasonably well using the distorted wave formalism. However, it is
easy to see that they can be also well described by the squares of Legendre
polynomials, PJ2 (cosθ ) , with J = 27, 29 and 31, respectively (see Figure 30.5).
The well-pronounced resonances in the excitation functions of Figure 30.1 at Ec.m. =
28.2, 31.2 and 34.2 MeV have been similarly described (Paul et al. 1978) using J =
21, 23 and 25, respectively. If these peaks arise from the enhancement of a single
partial wave, one can use such Legendre polynomial comparisons to assign spin J to the
corresponding maxima in the excitation functions. It is interesting to note, however, that
the positions of forward-angle maxima in the angular distribution corresponding to the
trough in the excitation function at Ec.m. = 39.0 MeV can also be well reproduced by a
PJ2 (cos θ ) distribution.
Figure 30.5. Legendre polynomial fits to the forward-angle cross sections of the experimental angular
distributions measured at the indicated c.m. energies.
The above values of J have been deduced assuming that only a single partial wave
dominates the reaction. However, the deviations in Figure 30.5 between the
PJ2 (cos θ ) fits and the data suggests that other partial waves, from either a non-
resonant reaction mechanism or overlapping resonances, also contribute. Indeed, for the
energy range 24 ≤ Ec.m. ≤ 40 MeV, a study of the transfer reaction at backward angles
(Paul et al. 1980) indicates that this is the case. It appears that the Legendre
polynomial fitting procedure allows, at best, a determination of the dominant J - value to
± 1h , and may be less precise in those cases where there is significant partial wave
interference.
For the energy region 37.5 - 43.5 MeV, the values of J extracted from our experiment
by optimal PJ2 (cos θ ) are shown in Figure 30.5 and are about two units greater than the
grazing angular momenta (defined as the J - value of that partial wave which is 50%
absorbed) for both the entrance and exit channels for the potential 11 of Table 30.2. The
spin assignments for the resonant states at lower energies (Paul et al. 1978) have
shown a similar correlation to the grazing angular momentum.
Figure 30.6. A plot of energies EJ for observed resonances against J(J+1). The J assignments made in
our work are indicated by stars, those obtained from previous studies are given as open and closed
circles (see the text). The best fit (straight line) has a band head of 29,8 MeV and an effective
moment of inertia of 7,08 × 10-42 MeV·s2.
A surface-transparent interaction of the type favoured by the distorted wave analysis

discussed earlier allows the two colliding ions to retain their individual structure
during a grazing collision and gives rise to the possibility of orbiting molecular states
in which two ions rotate about the centre of mass of the composite system. To examine
this possibility, we show in Figure 30.6 a compilation of J values for observed
resonances. The compilation is based on the measurements by Clover et al. (1979) for
the 24Mg(16O,16O)24Mg elastic scattering (closed circles), and by Paul et al. (1978, 1980),
Peng et al. (1976), Clover et al. (1979), and Lee et al. (1979) for the 24Mg(16O,12C)28Si and
28
Si(12C,16O)24Mg reactions (open circles). Stars denote the three new resonances we
have observed.
It can be seen that all the resonances lie close to a straight line suggesting a
rotational-like band, which can be described by the following equation:
h2
EJ = J ( J + 1) + E0
2ℑ
where ℑ is the effective moment of inertia of the system and E0 is the band head.
The gradient of the fitted straight line to the points corresponds to an effective moment
of inertia of 7.08 x 10-42 MeV · s2 and the projected J = 0 resonance lies at 29.8 MeV.
These values are in good agreement with the moment of inertia of 7.05 x 10~ 42 MeV · s2
and the energy 31.6 MeV for the lowest resonance, J = 0, when 24Mg and 16O nuclei are
just touching each other with no relative rotational energy as predicted by Cindro and
Počanić (1980). The series of quasi-bound and virtual states (or shape resonances) of a
molecular-like potential between two heavy ions are expected to form such a rotational-
like band.
Summary and Conclusions

We have measured excitation functions for the 24Mg(16O,12C)28Si (g.s., 21+ ) reaction in
the energy range of Ec.m. = 32.4 – 48.6 MeV at θ = 50 (lab). The chosen angle
corresponds to a maximum in the angular distributions for the 24Mg(16O,12C)28Si
reaction leading to the ground state in 28Si. We have observed three correlated
maxima in the excitation functions corresponding to Ec.m. = 37.5, 40.2, and 43.5 MeV.
We have measured angular distributions for the 24Mg(16O,12C)28Si (g.s. and 21+ excited
state) at these resonance energies and at Ec.m. = 39.0 MeV which corresponds to the
trough between the first and the second observed maxima in the excitation function. In
addition, we have measured angular distributions for the elastic scattering, 16O +
24
Mg at Ec.m. = 37.5 and 40.2 MeV.
The optical model analysis of the elastic scattering data yielded the required
parameters, which we used in our distorted wave analysis of the transfer reaction
data. We have found that while the theory reproduces reasonably well the measured
distributions for the 24Mg(16O,12C)28Si reaction leading to the ground state, the
calculated and experimental distributions for the first excited state, 21+ , are out of
phase. However, the theoretical distributions follow closely the angular trend of the
measured cross sections.
We have then analysed the angular distributions for the 24Mg(16O,12C)28Si (g.s.)
using the square of the Legendre polynomials, PJ2 (cos θ ) , and have found that the
three observed resonances located at energies Ec.m.. = 37.5, 40.2, and 43.5 MeV can
be assigned spins J = 27, 29 and 31, respectively. A compilation of earlier data
combined with our new measurements has shown that the resonance structure observed
in the reaction 24Mg(16O,12C)28Si can be described as nuclear molecular excitations of
binary nuclear systems made of interacting heavy ions either in the incident or outgoing
channels.
References
Abe, Y. 1977, in Nuclear Molecular Phenomena, ed. N. Cindro (North-Holland,
Amsterdam) p. 211.
Austern, N. and Blair, J. S. 1965, Ann. Phys. 33:15.
Ball, J. B., Hansen, O., Larsen, J. S., Sinclair, D. and Videbsfik, F. 1975, Nucl.
Phys. A244:341.
Barrette, J., LeVine, M. J., Braun-Munzinger, P., Berkowitz, G. M., Gai, M., Harris, J. W.
and Jachcinski, C. M. 1978, Phys. Rev. Lett. 40:445.
Braun-Munzinger, P., Berkowitz, G. M., Cormier, T. M., Jachcinski, C. M., Harris, J. W.,
Barrette, J. and LeVine, M. J. 1977, Phys. Rev. Lett. 38:944.
Braun-Munzinger, P., Bohne, W., Gelbke, C. K., Grochulski, W., Harney, H. L. and
Oeschler, H. 1973, Phys. Rev. Lett. 31:1423.
Bromley, D. A., Kuehner, J. A. and Almqvist, E. 1960, Phys. Rev. Lett. 4:365.
Chatwin, R. A., Eck, J. S., Robson, D. and Richter, A. 1970, Phys. Rev. C1:795.
Cindro, N. and Počanić, D. 1980, J. Phys. G6:359; 885.
Clover, M. R., Fulton, B. R., Ost, R. and DeVries, R. M. 1979, J. Phys. G5:L63.
Dehnhard, D., Shkolnik, V. and Franey, M. A. 1978, Phys. Rev. Lett. 40:1549.
Delic, G. 1975, Phys. Rev. Lett. 34:1468.
DeVries, R. M. 1972, Phys. Rev. C5:182.
Fink, H. J., Scheid, W. and Greiner, W. 1972, Nucl. Phys. A188:259.
Friedman, W. A., McVoy, K. W. and Nemes, M. C. 1979, Phys. Lett. 87B:179.
Gobbi, A., Wieland, R., Chua, L., Shapira, D. and Bromley, D. A. 1973, Phys. Rev.
C7:30.
Hanson, D. L., Stokstad, R. G., Erb, K. A., Olmer, C. and Bromley, D. A. 1974, Phys.
Rev. C9:929.
Hauser, W. and Feshbach, H. 1952,Phys. Rev. 87:366.
Imanishi, B. 1968,Phys. Lett. 27B:267.
Imanishi, B. 1969, Nucl. Phys. A125:33.
Kondo, Y., Matsuse, T. and Abe, Y. 1978, Prog. Theor. Phys. 59:465; 1009; 1904.
Lee, S. M., Adloff, J. C., Chevallier, P. C., Disdier, D., Rauch, V. and Scheibling,
F. S. 1979, Phys. Rev. Lett. 42:429.
Lemaire, M.C., Mermaz, M. C., Sztark, H. and Cunsolo, A. 1974a, Proc. Conf. on
Reactions between Complex Nuclei, Nashville, vol. 1, eds R. L. Robinson, F. K.
McGowan, J. B. Ball and J. H. Hamilton (North-Holland, Amsterdam) p. 21.
Lemaire, M.C., Mermaz, M. C., Sztark, H. and Cunsolo, A. 1974b, Phys. Rev.
C10:1103.
Maher, J. V., Sachs, M. V., Siemssen, R. H., Weidinger, A. and Bromley, D. A. 1969,
Phys. Rev. 188:1665.
Malmin, R. E., Siemssen, R. H., Sink, D. A. and Singh, P. P. 1972, Phys. Rev.
Lett. 28:1590.
Michaud, G. and Vogt, E. W. 1969, Phys. Lett. 30B:85.
Michaud, G. and Vogt, E. W. 1972, Phys. Rev. C5:350.
Olmer, C., Stokstad, R. G., Hanson, D. L., Erb, K. A., Sachs, M. W. and Bromley,
D. A. 1974, Phys. Rev. C10:1722.
Paul, M., Sanders, S. J., Cseh, J., Geesaman, D. F., Henning, W., Kovar, D. G., Olmer,
C. and Schiffer, J. P. 1978, Phys. Rev. Lett. 40:1310.
Paul, M., Sanders, S. J., Geesaman, D. F., Henning, W., Kovar, D. G., Olmer, C.,
Schiffer, J. P., Barrette, J. and LeVine, M. J. 1980, Phys. Rev. C21:1802.
Peng, J. C., Maher, J. V., Oelert, W., Sink, D. A., Cheng, C. M. and Song, H. S. 1976,
Phillips, R. L., Erb, K. A., Bromley, D. A. and Weneser, J. 1979, Phys. Rev. Lett.
42:566.
Scheid, W., Greiner, W. and Lemmer, R. 1970, Phys. Rev. Lett. 25:176.
Siemssen, R. H. 1971, Proc. Symp. on Heavy-ion Scattering, Argonne National
Laboratory, March, report ANL-7837, p. 145.
Siemssen, R. H. 1977,in Nuclear Molecular Phenomena, ed. N. Cindro (North-
Holland, Amsterdam) p. 79
Siemssen, R. H., Fortune, H. T., Richter, A. and Yntema, J. L. 1969, Proc. Conf.
on Nuclear Reactions Induced by Heavy Ions, Heidelberg, eds R. Bock and
W. R. Hering (North-Holland, Amsterdam) p. 65.
Siwek-Wilczynska, K., Wilczynski, J. and Christensen, P. R. 1974, Nucl. Phys.
A229:461.
Tserruya, I., Bohne, W., Braun-Munzinger, P., Gelbke, C. K., Grochulski, W., Harney,
H. L. and Kuzminski, J. 1975, Nucl. Phys. A242:345.
31
The Interaction of 7Li with 28Si and 40Ca Nuclei
Key features:
1. Interaction of projectiles with 4 < A < 12 is difficult to describe theoretically. New data
were needed to help to investigate and hopefully understand this problem.
2. We have measured angular distributions for the elastic and inelastic scattering from
28
Si and 40Ca. These two target nuclei have been selected because of the differences
in their collective excitations: rotational and vibrational, respectively.
3. We have carried out standard and double folding optical model analysis of the elastic
scattering. Both give similar and satisfactory description of the experimental angular
distributions. We have found that the absorptive potential dominates the interaction of
7
Li with both 28Si and 40Ca nuclei.
4. Both the elastic and inelastic scattering data for the 7Li + 40Ca system can be
7 28
reproduced well using coupled-channels formalism. However, for the Li + Si
interaction, the theory reproduces the trends of the differential cross sections but
cannot fit simultaneously the elastic or inelastic distributions unless different
deformation parameters are used. The problem might be associated with the mutual
excitations of projectile and target nuclei and thus involving different collective
structures, which are not accounted for in the available theoretical formalism.
Abstract: Elastic and inelastic scattering of 45 MeV 7Li from 28Si and 40Ca have been
measured and analysed. The inelastic scattering distributions corresponded to the excitation
of the 1.78 MeV state in 28Si and to the 3.73 and 3.90 states in 40Ca. Double folding model
9
calculations using a realistic effective nucleon-nucleon interaction similar to that used for the Be +
28 9 40
Si and Be + Ca scattering have been carried out for the elastic angular distributions. The real
potential had to be renormalized to yield agreement with the measured cross sections. Coupled
channels calculations using a Woods-Saxon potential were performed in an effort to describe both
the elastic and inelastic angular distributions. The extracted deformation parameters are in
reasonable agreement with those obtained from light and heavier ion scattering from the same
target nuclei. The effect of strong excitation of the 0.48 MeV state in 7Li and of mutual excitation of
target and projectile is considered in a qualitative manner.
Introduction
The region between light (A < 4) and heavy (A > 12) projectiles is challenging
because of the problems encountered in theoretical interpretations of experimental
data. In particular, the double folding model (Satchler 1976; Satchler and Love 1979)
with a realistic nucleon-nucleon interaction, which has been successful in describing
the elastic α and heavy ion scattering, is not as successful in describing the elastic
scattering of 6Li, 7Li, and 9Be from the same target nuclei unless the real double
folding potential is reduced by a factor of about 0.4 to 0.6.
Furthermore, the optical model using either the calculated double folding potential or
one of Woods-Saxon shape, indicates that the effective interaction distance for these
intermediate nuclei is larger than for α and heavy-ions (Balzer at al. 1977; Ungricht
et al. 1979). Perhaps more important is a transition from the dominance of the real
potential to the dominance of the imaginary potential as one proceeds through this
intermediate region between A = 4 and A = 12 projectiles. In particular, the
interaction of 6Li with 28Si suggests weak absorption (DeVries et al. 1977) but a study
of elastic, inelastic, and fusion interactions of 9Be with A = 20 - 60 targets indicates
the dominance of a strong absorption (Zisman et al. 1980).
In order to investigate this transition in the absorption strength, we have measured
the elastic and inelastic scattering cross sections for the excitation of the 1.78 MeV
state in 28Si, the 3.73 and 3.90 MeV states in 40Ca, and the 0.48 MeV state in 7Li
using 7Li projectiles at a bombarding energy of 45 MeV. In addition, we have
measured the cross section for mutual excitation of the 0.48 MeV state in 7Li and the
1.78 MeV state in 28Si. The experimental details are given below.
The measured elastic cross sections were fitted using both the standard optical
model as well as the double folding procedure (Satchler and Love 1979), which was
successful in describing the elastic scattering of α particles, and heavy ions (A > 12).
The inelastic cross sections were fitted using the coupled channels formalism.
Deformation lengths and deformation parameters are extracted from the fitted cross
sections and compared with those obtained from measurements using other projec-
tiles (Hendrie 1973; Thompson and Eck 1977). The effect of the absorptive potential
on the calculated cross sections is investigated and compared to the dominance it
exhibits in the case of 9Be scattering from the same target nuclei (Zisman et al.
1980).
-
The 7Li beam was extracted from a sputter source in the form of 7Li and was
injected into the ANU 14D Pelletron accelerator. The targets consisted of ~200 μ/cm2
self-supporting SiO2 (greater than 99.5% enriched in 28Si) or ~160 μ/cm2 Ca (greater
than 99.8% enriched in 40Ca) evaporated onto 10 μ/cm2 C foils. The scattered 7Li
ions were detected using the Enge split pole magnetic spectrograph and a focal
plane detector, which was operated in the light-ion mode (Ophel and Johnston
1978). The states in 28Si, 40Ca, and 7Li were clearly resolved for up to about 7 MeV
excitation energy by gating on the 7Li mass. A typical spectrum for 7Li + 28Si is shown
in Figure 31.1
Figure 31.1. Typical position spectrum gated on the 7Li mass for the 7Li + 28Si scattering at θ = 30°
(lab) obtained using the Enge split pole spectrometer and the focal plane detector.
The relative normalization was obtained by using a monitor detector placed at a
laboratory angle of 15°. The absolute normalization was carried out by normalizing
the measured cross sections to Rutherford scattering at a bombarding energy of 25
MeV and a laboratory scattering angle of 7.5°. The absolute normalization is
accurate to about 10%. The measured angular distributions are shown in Figures
31.6 – 31.9.
The elastic and inelastic angular distributions fall off quickly with increasing angle
and show strong diffraction-like patterns. In the measured angular range (10° < θc.m.
< 80°), there is no indication of any levelling off or rising of the cross section with
increasing angle as has been observed for 12C + 28Si and 160 + 28Si (Braun-Muzinger
et al. 1977; Clover et al. 1978).
The optical-model analysis
Prior to carrying out the coupled channels calculations, the elastic scattering angular
distributions were fitted using a standard optical model Woods-Saxon potential
defined in Chapter 29.
The initial parameters were chosen to be those of Cramer et al. (1976), which were
used to fit I6O + 28Si scattering over a wide energy range. Only V and W were varied
and the best-fit parameters are listed in Table 31.1. The calculated best-fit cross
sections are shown in Figure 31.2 and 31.3 for 7Li + 28Si and 7Li + 40Ca, respectively.
Table 31.1
Best-fit optical-model parameters
The ratio of the real to the imaginary potential was calculated at r = Dl/2 where Dl/2 is
the distance of closest approach for the Rutherford orbit for which l = L l / 2 . L1/2 is the
value of l at which the transmission coefficient T = 1/2. The Dl/2, L l / 2 , and the ratio of
W/V evaluated at Dl/2 are listed in Table 31.1.
The next step was to refine the calculations by using a double folding model. The
double folding procedure used in our study is the same as that used by Satchler
(1976, 1979) and Satchler and Love (1979) to describe 6Li, 12,13C, 14,15N, and 16,17,18O
scattering from 28Si, 40Ca, and other nuclei in the same mass region.
The real component of the interaction potential may be written as:
r r r r r
U F ( R ) = ∫ ρ (r1) ρ (r2 )VNN ( R − r1 + r2 )dr1dr2
where R is the distance between the centres of the nuclei, ρ1 and ρ 2 are the nucleon
r
distributions in the interacting nuclei, and VNN (r ) is the effective nucleon-nucleon
interaction.
Figure 31.2. The calculated angular distribution for the 7Li + 28Si elastic scattering at E(7Li) = 45 MeV
using the Woods-Saxon optical model potential and parameters of Table 31.1 are compared with our
experimental data.
7 40 7
Figure 31.3. The calculated angular distribution for Li + Ca elastic scattering at E( Li) = 45 MeV
using the Woods-Saxon optical model potential and parameters of Table 31.1 are compared with our
experimental data.
r
The most widely used forms for the VNN (r ) interaction are Reid or Paris (Brandan and
Satchler 1997; Lacombe et al. 1981; Reid 1968; Satchler and Love 1979). In our
calculations we have used the soft-core Reid potential. The method for constructing
the density distributions is described by Satchler and Love (1979). The real folded
potential U F (R ) thus constructed is multiplied by a normalizing factor N, where N is
varied to obtain the best fit.
A Woods-Saxon imaginary term with R′ = r0′( A1p/ 3 + At1 / 3 ) , where p stands for projectile
and t for target, is added to give the total potential. In the our work, r0′ was fixed at
the value of 1.3 fm. The normalization parameter N for the real potential, the ima-
ginary potential strength W, and the diffuseness of the imaginary well a0′ , were then
varied to obtain the best fit. The best fit parameters are given in Table 31.1 and the
best folding model fit calculations for the measured elastic scattering angular
distributions are shown in Figures 31.4 and 31.5. As can be seen from Table 31.1,
the L l / 2 values are slightly different for each potential for both the 7Li + 28Si and 7Li +
40
Ca scattering. The ratios of W/V indicate the dominance of strong absorption at r =
Dl/2 in both cases.
Figure 31.4. Folding model calculation (solid curve) for the 7Li + 28Si elastic scattering at E(7Li) = 45
MeV using the realistic nucleon-nucleon potential. See the text for details. Experimental cross
sections are shown as closed circles.
Figure 31.5. Folding model calculation (solid curve) for the 7Li + 40Ca elastic scattering at E(7Li) = 45
MeV using the realistic nucleon-nucleon potential. See the text for details. Experimental cross
sections are shown as closed circles.
In order to fit the cross sections, it was necessary to renormalise the potential by a
factor of N ≈ 0.58. This result is similar to that obtained for 6Li and 9Be scattering
from the same target nuclei. It has been shown recently that the necessity for this
renormalization is eliminated for the case of 7Li and 9Be scattering if quadrupole ef-
fects (and therefore reorientation) are included in the data analysis (Hinzdo, Kemper,
and Szymakowski 1981). A difficulty with this approach is that it does not eliminate
the necessity for renormalization of the potential for 6Li scattering (Satchler and Love
1979).
Coupled-channels calculations
The coupled channels calculations were performed using the computer code ECIS79
(Raynal 1972, 1981), which I have adapted earlier to run on an ANU computer. The
calculations were carried out using 80 partial waves and assuming that the low
excited states of 28Si can be described using a rotational model while the lowest
states of 40Ca can be described by a vibrational model. Radial integrations were
carried out to 40 fm to account properly for Coulomb excitation. The optical model
parameters obtained from fitting the elastic cross section were utilized except that W
was reduced by about 10% and the quadrupole deformation parameter β2 was
varied to obtain the optimal fit.
The fits produced by this technique were not totally satisfactory. The fit for β2 = -0.l5
and W = 11.69 is shown in Figure 31.6. For these parameters the fit to the elastic
cross section is good but for the inelastic cross section is too low in magnitude and
exhibits too much structure. Increasing the absolute magnitude of the quadrupole
deformation parameter to β2 = -0.25 improves the fit to the inelastic cross section but
causes a deterioration in the elastic fit due to a reduction in the calculated diffraction
structure, especially at back angles. This can be remedied slightly by decreasing W
but this procedure exaggerates the diffraction oscillations. The fit obtained for β2 = -
0.25 and W = 15.00 MeV is shown in Figure 31.7.
Figure 31.6. Angular distributions for the elastic and inelastic scattering (to the 1.78 MeV state in 28Si)
of 45 MeV 7Li projectiles. The elastic scattering cross section is shown as a ratio to the Rutherford by
the upper set of closed circles. The inelastic cross section is given in absolute units and is shown by
the lower set of closed circles. The solid curves are the cross sections calculated using the coupled
channels theory assuming the deformation parameter β2 = -0.15 and W = 11.69 MeV. See the text for
details.
Inelastic scattering to the 0.48 MeV state in 7Li and the mutual excitation of the 0.48
and 1.78 MeV states have not been included in the calculation. These cross sections
are shown in Figure 31.8 and comparison with Figure 31.6 or 31.7 shows that these
neglected cross sections are similar in magnitude to the cross section for exciting the
1.78 MeV state and need to be included explicitly in the coupled channels calculation
if better results are to be expected. Unfortunately, this kind of calculations has been
inaccessible within the available theoretical framework.
Figure 31. 7. Experimental angular distributions are as displayed in Figure 31.6. The calculated
curves are for the deformation parameter β2 = -0.25 and W = 15 MeV. See the Caption to Figure 31.6
and the text for details.
The 7Li + 40Ca elastic and inelastic cross sections were fitted using a second order
vibrational model to describe the low-lying 2+ and 3- states of 40Ca. The 3.73 (3-)
state is assumed to be the excitation of one octupole phonon and the 3.90 (2+) state
is assumed to be the excitation of a single quadrupole phonon.
By reducing W and adjusting β2 and β3 the obtained fits are shown in Figure 31.9.
For this case W = 18.00 MeV, β2 = 0.06, and β3 = 0.15. As can be seen, the fits to
both elastic and inelastic scattering angular distributions are quite satisfactory. For
7
Li + 40Ca the excitation of the 0.48 MeV state in 7Li is also relatively strong but its
exclusion from the calculation does not appear to affect strongly the final fits. The
values of β2 and β3 can be compared with those obtained from α particle scattering
from 40Ca at Eα = 29 MeV by calculating the deformation distance δ l defined as δ l =
β l R , where the potential radii from α + 40
Ca and 7Li + 40
Ca scattering are 4.76 and
7.20 fm, respectively. The deformation distances are compared in Table 31.2. The
agreement for β2R is fair but there is a significant difference for β3R.
Figure 31.8. Measured cross sections for excitation of the 0.48 MeV state of 7Li (upper set of points)
and for mutual excitation of the 0.48 MeV state of 7Li and the 1.78 MeV state of 28Si (lower set of
points) by 7Li + 28Si scattering at E(7Li) = 45 MeV.
Figure 31.9. Angular distributions for the elastic scattering and inelastic scattering (to the 3.73 and
3.90 MeV states in 40Ca) of 45 MeV 7Li projectiles. Elastic scattering shown as a ratio to the
Rutherford by the upper set of points. Inelastic cross sections for the excitation of the 3.73 and 3.90
MeV states are given in absolute units and are indicated by the middle and bottom sets of dots,
respectively. The solid curves are cross sections calculated using the coupled channels formalism and
the deformation parameters β2 = 0.06, and β3 = 0.15.
Table 31.2
Deformation lengths for 40Ca

Angular distributions for the elastic and inelastic scattering cross sections for 7Li
scattering from 28Si and 40Ca targets have been measured and analysed. The low-
lying excited states of both target and projectile are strongly excited and in the case
of 7Li + 28Si scattering mutual excitations of both target and projectile are significant.
Double-folding model calculations of the elastic scattering cross sections yield
potentials, which must be renormalized by a factor of ≈ 0.6 in order to fit adequately
the measured cross sections.
Coupled channels calculations for 7Li + 28Si reproduce the trends of the experimental
data. However, the theory requires different deformation parameters to fit either the
elastic or inelastic distributions. In the case of 7Li + 40Ca scattering, the cross
sections for the ground state, the first excited state (3-) at 3.73 MeV, and the second
excited state (2+) at 3.90 MeV are well described using the coupled channels
formalism. Comparison of the deformation lengths obtained here and those obtained
from α +40Ca scattering are in fair agreement for β2 but not for β3.
Considering the fairly good agreement between the theory and experiment for the 7Li
+ 40Ca interaction the problems encountered with the theoretical interpretation of the
7
Li + 28Si scattering is hard to explain. In both cases, the same projectile is used, so
the argument based on the weakly bound nature of 7Li appear unconvincing. The
problem might be associated with the mutual excitation of 7Li and the target nucleus,
which are not accounted for by the available coupled channels formalism. Even
though the projectile is the same for the 7Li + 28Si and 7Li + 40Ca, the target nucleus
is different. Excited states in 28Si can be described using rotational model, whereas
40
Ca is a spherical nucleus, which exhibits vibrational excitations. The 2+ state in 28Si
has a sizable quadrupole moment of around +(16 ± 3) efm2 (Stone 2001). The mutual
excitations of different collective structures might have a substantial effect on the
coupling to the observed transitions and thus influence the character of the
measured angular distributions.
References
Balzer, R., Hugi, M., Kamys, B., Lang, J., Muller, R., Ungricht, E., Untermahrer, J. and
Jarczyk, L. 1977, Nucl. Phys. A293:518.
Brandan, M. E. and Satchler, G. R. 1997, Phys. Rep. 285:142.
Braun-Munzinger, P., Berkowitz, G. M., Cannier, T. M., Jachcinski, C. M., Harris, J. W.,
Barrette, J. and Levine, M. J. 1977, Phys. Rev. Lett. 38, 944.
Clover, M. R., DeVries, R. M., Ost, R., Rust, N. J. A., Cherry, Jr., R. N. and Gove, H. E.
1978, Phys. Rev. Lett. 40:1008.
Cramer, J. G., DeVries, R. M., Goldberg, D. A., Zisman, M. S., and Maguire, C. F. 1976,
Phys. Rev. C14:2158.
DeVries, R. M., Goldberg, D. A., Watson, J. W., Zisman, M. S. and Cramer, J. G. 1977,
Hendrie, D. L.. 1973, Phys. Rev. Lett. 31:478.
Hnizdo, V., Kemper, K. W., and Szymakowski, J. 1981,Phys. Rev. Lett. 46:590.
Lacombe, M., Loiseau, B., Vinh-Mau, R., Côté, J., Pires, P. and de Tourreil, R. 1981,
Phys. Lett. B 101:139.
Reid, R. V. 1968, Ann. Phys. 50:411.
Raynal, J. 1972, Computing as a language of physics (IAEA, Vienna,) p. 281.
Ophel, T. R. and Johnston, A. 1978, Nucl. Instr. Methods 157:461.
Satchler, G. R. 1976, Nucl. Phys. A279:61.
Satchler, G. R., 1979, Nucl. Phys. A329:233.
Satchler, G. R. and Love, W. G. 1979, Phys. Rep. 55:185.
Thompson, W. J. and Eck, J. S. 1977, Phys. Lett. 67B:151.
Ungricht, E., Balzer, D., Hugi, M., Lang, J., Muller, R., Jarczyk, L. and Kamys, B. 1979,
Zisman, M. S., Cramer, J. G., Goldberg, D. A., Watson, J. W. and DeVries, R. M. 1980,
32
Triaxial Structures in 24Mg
Key features:
1. High-resolution measurements were carried out for the elastic and inelastic scattering
of 16O from 24Mg nuclei using 72.5 MeV (lab) 16O beam. The doublet of states,
4.12/4.24 MeV, essential in a study of triaxial deformation of 24Mg has been resolved
for the first time for heavy ion projectiles.
2. Energy level schemes for 24Mg have been calculated assuming a rigid asymmetric
( γ ≠ 0 ) rotor. We have used both the standard and extended Davydov-Filoppov
model.
3. Inclusion of hexadecapole deformation resulted in a better description of reduced
transition rates.
4. Coupled channels analysis of experimental angular distributions were carried out
using three types of the interaction potential containing the quadrupole and
hexadecapole deformations for both the real and imaginary components. Nearly
prefect fits were obtained for a surface-transparent potential with a moderate real
depth.
5. Deformation distances for the quadrupole and hexadecapole deformations compare
well with the lengths obtained for light projectiles.
6. The effect of hexacontatetrapole component in the interaction potential has been
investigated and found to be negligible.
16 24
Abstract: Angular distributions for the elastic and inelastic scattering of O from Mg,
+ + +
exciting the 2 1.37 MeV, 4 4.12 MeV and 2 4.24 MeV states, have been measured at
1 , 1 , 2,
Ec.m. = 43.5 MeV, the energy at which resonance-like structure has been observed previously
+ +
in the related 24Mg(16O,12C)28Si reaction. The 41 , 22 doublet has been resolved for the first
time for heavy-ion projectiles. The data have been well described by coupled-channels
calculations within the framework of the Davydov-Filippov asymmetric rotor model for the low-
lying states of 24Mg, which has been extended to include a symmetric hexadecapole shape
component. The optical model potential for the 16O + 24Mg interaction was found to have a
moderate real well depth and surface transparency. The shape parameters for the nuclear
potential were determined to be β 2 = 0.25, γ = 22° and β 4
(N ) (N )
= -0.065 and the
corresponding deformation distances are in good agreement with earlier light-ion results. The
inclusion of a negative symmetric hexadecapole component leads to an improved description
of the reduced transition rates. The triaxial structure of 24Mg is discussed.
Introduction
The possible existence of triaxial structures in some 2s-1d shell nuclei has been
discussed since the early 1960's. Of specific interest in this work was the nucleus
24
Mg, which was considered during that early period (Batchelor et al. 1960; Cohen
and Cookson 1962) within the framework of the Davydov-Filippov (1958) model.
In this model the nucleus is considered to be a rigid ellipsoid with quadrupole
deformation rotating about its centre of mass. The conclusions of that early work
were summarized by Robinson and Bent (1968) who pointed out that the Davydov-
Filippov model, when evaluated for γ = 22°, which gives the best description of the
energy separations for the low-lying states of 24Mg, was unable to predict many of
the branching ratios and magnitudes of the reduced transition rates.
Measurements of lifetimes and branching ratios for 24Mg (Branford, McGough, and
Wright 1975) are in closer agreement with the expectations of a Davydov-Filippov
model with γ = 21.5° for the intra-band transitions although some of the cross-band
transitions are still poorly described. Branford, McGough, and Wright (1975) have
pointed out that both sets of transitions are better described with γ = 14°, a value
which unfortunately places the 2 +2 state at 10.8 MeV excitation energy. It is also not
clear whether one should associate the 4+2 model state with the level observed at
6.01 MeV or with that at 8.44 MeV, although energy considerations favour the latter.
The calculations of Aspelund (1982) using the rotation-vibration model of Faessler
and Greiner (1962) indicate non-negligible band-mixing for the 4+ states but give the
level at 8.44 MeV as the third member of the K = 2 band. The existence of two 4+
levels between 6 and 9 MeV excitation energy (in addition to several other states)
contrary to the rigid asymmetric rotor model indicates that this model may not
provide a good description of these higher 4+ states in 24Mg.
A revived interest in the use of an asymmetric rotor model for 24Mg is motivated by
two considerations. First, the increased sophistication of coupled-channels analyses
of elastic and inelastic scattering experiments has prompted extensive use of such
approaches to determine nuclear shape parameters of low-lying levels of light nuclei.
For 24Mg, Hartree-Fock and other calculations (Abgrall et al. 1969; Grammaticos
1975; Kurath 1972) indicate triaxial deformations for the ground (K = 0) and K = 2
rotational bands, and analyses of proton (Eenmaa et al. 1974; Lombard, Escudié
and Soyeur 1978; Lovas et al. 1977) and α - particle (Kokame et al. 1964; Tamura
1965; van den Borg, Harakeh, and Nilsson 1979) inelastic scattering from 24Mg have
been carried out within an asymmetric rotor model framework.
The second reason arises from the observation of considerable resonance-like
structure in excitation functions for heavy-ion reactions, particularly those involving
the nuclei 12C, 16O, 24Mg and 28Si. It has been proposed, in some cases at least, that
the observed resonances, which are too narrow to be readily described in terms of
simple shape resonances associated with quasi-bound states in a molecular-like
potential between the two ions, should be interpreted as intermediate structure which
arises in a "doorway-state" model in which either the initial or final channel couples
to another degree of freedom of the system (Abe, Kondo, and Matsuse 1980; Fink,
Scheid, and Greiner 1972; Imanishi 1968, 1969; Kondo, Matsuse, and Abe 1978;
Michaud and Vogt 1969, 1972; Scheid, Greiner, and Lemmer 1970). In particular, it
has been suggested (Nurzynski et al. 1981) that resonant structure observed in the
24
Mg(16O,12C)28Si reaction may be a consequence of strong couplings between the
elastic and inelastic channels of either the initial or final systems.
The first step in the testing of the initial-system coupling hypothesis is a detailed
coupled-channels analysis of the 24Mg + 16O system. The light-ion work mentioned
above suggests that an asymmetric rotor model is an appropriate starting point for
such analyses. An earlier attempt (Eck et al. 1981) at such an analysis was
hampered by an experimental inability to resolve the 41+ and 2+2 states in 24Mg and
the lack of forward-angle data.
The discussion presented in this chapter is about the measurements of elastic and
inelastic scattering at 43.5 MeV (c.m.) of 16O from 24Mg, with particular emphasis on
resolving the closely-spaced 41+ , 2 +2 doublet near 4.2 MeV. The bombarding energy
was chosen to correspond to a peak in the related 24Mg(16O,12C)28Si excitation
function (see Chapter 30) in order to study the nature of the 24Mg + 16O interaction at
an energy close to one of its possible shape resonances. Since the inelastic
scattering at forward angles is expected to be much less sensitive than the transfer
reaction to a single resonating partial wave, the data should also permit a reasonable
extraction of 24Mg shape parameters.

In this experiment special effort was made to resolve the doublet near 4.2 MeV
excitation energy in 24Mg. For this purpose very thin targets were used and a high-
resolution delay-line focal-plane detector (Leigh and Ophel 1982) was employed to
detect the reaction products at the focal plane of a split-pole Enge spectrometer. The
targets were prepared by vacuum evaporation of a thin (~ 5 μg/cm2) layer of
enriched (99.92%) 24Mg on to a comparable thickness of carbon backing. A beam of
72.5 MeV 16O projectiles was provided by the Australian National University 14 UD
Pelletron accelerator. The incident 16O energy corresponded to one of the
resonances observed earlier (Nurzynski at al. 1981) in a study of the reaction
24
Mg(16O,12C)28Si.
Figure 32.1. Energy spectrum taken at 19° (lab) for the scattering of 16O from 24Mg and contaminating
elements at 72.5 MeV (lab) bombarding energy.
Data were recorded in event-by-event mode on magnetic tapes using a HP-2100

data acquisition system. Measurements of the angular distributions for both elastic
scattering and inelastic scattering to the 21+ (1.37 MeV), 41+ (4.12 MeV) and 2+2 (4.24
MeV) states in 24Mg were carried out in steps of 1° for the angles 4° - 29° (lab). The
horizontal acceptance angle of the magnetic spectrometer was 1°. At the most
forward angles, the elastic and inelastic scattering cross sections were measured
separately. For the inelastic groups good statistics were obtained by blocking the
elastic group and maintaining high beam intensities. Between about 300 and 2000
counts were obtained for each member of the doublet and the overall resolution of
the detector system was maintained at 80 - 100 keV (FWHM) over the data
collection periods, which were as long as 4 hours for some angles.
Events corresponding to 16O + 24Mg scattering were selected from the magnetic
tapes using two-dimensional windows and a typical 16O particle spectrum as shown
in Figure 32.1. The doublet near 4.2 MeV was analysed by fitting skewed Gaussian
distributions obtained from the elastic and the 21+ inelastic peaks. Absolute cross
sections were determined by normalization of the data to Rutherford scattering at
forward angles and the resultant angular distributions are displayed in Figures 32.3,
32.4 and 32.6. For certain angles, the inelastic scattering data were obscured by 12C
and 16O elastic scattering. The main sources of error in the individual points of the
angular distributions are statistical errors and uncertainties arising from inaccuracy of
the peak fitting.
Extended rigid asymmetric rotor model of 24Mg

The rigid asymmetric rotor model of Davydov and Filippov (1958), which describes
only quadrupole deformation, has been extended by Baker (1979) and Barker at al.
(1979) to include hexadecapole deformation. This extended model is employed in
the present work for the analysis of 16O scattering from 24Mg.
It is assumed that the nuclear charge density is uniform inside a radius given in the
intrinsic coordinate system by
R (θ ′,φ ′) =
{ }
R0 1 + β 2 cos γY20′ + 1 / 2 β 2 sin γ (Y22′ + Y2′− 2 ) β 4Y40′ + a42 (Y42′ + Y4′− 2 ) + a44 (Y44′ + Y4′− 4 )
where Yλμ′ = Yλμ (θ ′,φ ′) . The rigid rotor Hamiltonian for such a distribution has been
derived by Baker (1979) assuming that the inertial parameters (B2 and B4) satisfy the
irrotational relation (Strutt 1926) (i.e. B = B2 = 2B4). In our study, for simplicity, we
have considered only symmetric hexadecapole shapes (i.e. a42 = a44 = 0). The
Schrödinger equation for the triaxial system has the form :
⎧ ⎡ 2
⎤
−1
⎫
⎪⎛ 1 2 ⎞ ⎛ ⎞ ⎛ β ⎞ ⎪
⎨⎜ h / Bβ 2 ⎟∑ J k ⎢sin ⎜ γ − πk ⎟ + ⎜⎜ ⎟⎟ (1 − δ k ,3 )⎥ − ε ⎬ψ = 0
3
2 1 2 2 2 5 4
⎪⎩⎝ 4 ⎠ k =1 2 ⎢⎣ ⎝ 3 ⎠ 4 ⎝ β2 ⎠ ⎥⎦ ⎪⎭
where J k are the operators of the projection of the nuclear angular momentum on
the axes of the body-fixed coordinate system. The wave function ψ for the nth state
of spin I and projection M can be conveniently expanded in terms of basis states
IMK , i.e.
I
ψ IM
( n)
= ∑ AIK
(n)
IMK
K =0
even
where
{D }
1/ 2
⎡ 2I + 1 ⎤
IMK = ⎢ + (− 1) DMI − K
I l
⎥
⎣16π (1 + δ K 0 ) ⎦
2 MK
∑A
K =0
( n)
IK
( n′ )
AIK = δ nn′
even
and the eigenvalues and eigenvectors can be obtained by standard techniques

(Eisenberg and Greiner 1975).
To evaluate transition probabilities and moments of the nucleus, the reduced matrix
elements of the electric multipole operators Μ ( Eλ , μ ) are required in space-fixed
coordinates. We have
Μ ( Eλ , μ ) = ∑ Μ′( Eλ , v) Dμλv
v
where Μ′( Eλ , μ ) are the corresponding Eλ spherical tensor operators in the body-
fixed coordinate frame given in terms of a volume integral over the nuclear charge
density ρ by the relation
Μ′( Eλ , μ ) = ∫ Yλ′v r λ ρ (r ,θ ′, φ ′)dτ ′
For the standard Davydov-Filippov model (i.e. β4 = 0), explicit expressions for the
energies of some of the low-lying energy levels and first-order terms for the reduced
E 2 transition probabilities have been given (Davydov and Filipov 1958; Davydov and
Rostovskii 1959). In particular the ratio of the energies of the 2 +2 and 21+ states is a
function of the angle γ only. One has
{
ε (2+2 ) 3 + [9 − 8 sin 2 (3γ )]
1/ 2
}
=
{
ε (21+ ) 3 − [9 − 8 sin 2 (3γ )]1 / 2 }
which gives γ ≈ 22° for 24Mg.
Normalization of the energy scaling h 2 / 4 Bβ 22 to the experimental value of the 21+
state yields spectrum A of Figure 32.2. It is seen that good agreement with
experimental values (Endt and van der Leun 1978) is obtained for the low-lying
levels and that the 4+2 predicted level lies much closer to the 4+ state at 8.44 MeV
rather than that at 6.01 MeV.
It should be noted that at 8.44 MeV excitation energy, a close (4+,1-) doublet has
been identified (Ollerhead et al. 1968). The corresponding B( E 2) and quadrupole
moment of the 21+ state ( Q2 + ) predictions using the Davydov-Filippov expressions
1
(Davydov and Filipov 1958; Davydov and Rostovskii 1959) are presented in the
fourth column of Table 32.1. Comparison with the experimental values (Branford,
McGough, and Wright 1975; Fewell et al. 1979) shows good agreement for the intra-
band transitions and Q2 + . However, some of the cross-band transition rates
1
[ 2 → 2 , 3 → 4 and 4 (6.01 MeV) → 41+ ] are predicted to be too high. Similar

+
2
+
1
+ +
1
+
discrepancies have been obtained in both shell-model (Lombard, Escudié and

Soyeur 1978) and Hartree-Fock (Branford, McGough, and Wright 1975) studies. In
Table 32.1, the decay properties of the 41+ model state are compared with the
available data (Branford, McGough, and Wright 1975; Meyer, Reinecke, and
Reitmann 1972) for both the 6.01 and 8.44 MeV levels.
Figure 32.2. Energy levels for 24Mg predicted by rigid triaxial rotor models are compared with
experimental values (Endt and van der Leun 1978). Result A is for the standard Davydov-Filippov
(1958) model with γ = 22°. Results B and C include a symmetric hexadecapole deformation β4 = -
0.26β2, with γ = 22° and 18°, respectively. In each case, the energy scale is normalized by matching
the energy of the 21+ (1.37 MeV) state.
Figure 32.2 and Table 32.1 also show the results when a symmetric hexadecapole
deformation (β4/β2 = - 0.26) is included. The values of the parameters β2 and β4 for
this ratio were determined by a coupled-channels analysis of the 24Mg(16O,16O') data
discussed in the next section. The resultant level spectrum, assuming γ = 22° (result
B), gives a somewhat poorer separation of the 41+ and 2+2 states. However, the
B( E 2) predictions for the 2+2 → 21+ , 3+ → 41+ and 4+ (6.01 MeV) → 41+ cross-band
transitions are significantly improved. The predictions for these transitions can be
brought into close agreement with experimental values by retaining the non-zero β4
deformation but reducing the value of γ to 18° (last column of Table 32.1). On the
other hand, Figure 32.2 shows (result C) that the corresponding energies for the K =
2 band states now lie far too high relative to those of the ground-state K = 0 band so
the smaller value of γ seems unsatisfactory.
Table 32.1
The values of the γ transition probabilities B ( E 2) and quadrupole moments Q2 + for 24Mg
1
All calculated values assume R0 = 1.25 At1 / 3 fm and β 2 R02 = 7.3 fm2.
Davydov-Filippov – Only quadrupole deformation is considered. β4/β2 = -0.26 – The effect of
including both quadrupole and hexadecapole deformations.
Coupled-channels analysis
Coupled-channels calculations for the elastic and inelastic scattering of 16O from
24
Mg at a bombarding energy of 72.5 MeV (Ec.m. = 43.5 MeV) were performed using
the computer code ECIS79 (Raynal 1972, 1981)28. The 0+ (g.s.), 21+ (1.37 MeV), 41+
(4.12 MeV) and 2+2 (4.24 MeV) states in 24Mg were included in all the calculations
and select computations to obtain the final results were carried out including also the
3+ (5.24 MeV), 6+ (8.11 MeV) and 4+ (8.44 MeV) states. The calculations involving
the full set of six excited states took an order of magnitude longer to perform than
those in which only the lowest three excited states were taken into account. Thus
most of the preliminary calculations were carried out employing the smaller set of
states. All the calculations employed the Coulomb correction technique developed by
Raynal (1980,1981) with matching at a radius of approximately 14 fm. Comparison
with conventional calculations with matching near 40 fm indicated agreement to
better than 2% in the predicted angular distributions for all states included in the
coupling scheme.
The deformed optical potential was of the form
U (r ,θ ,φ ) = U N (r ,θ ,φ ) + VC (r ,θ ,φ )
where
U N (r ,θ ,φ ) = −VfV (r ,θ ,φ ) − iWfW (r ,θ ,φ )
with
fV (r ,θ ,φ ) = [1 + exp{[r − RV (θ ,φ ) )]/ a0 }]
−1
fW (r ,θ ,φ ) = [1 + exp{[r − RW (θ ,φ ) )]/ a0′ }]

−1
28
See Chapter 16.
and the Coulomb potential is generated from the deformed charge
density according to the relation
r
RC (θ ,φ ) dr ′ RC (θ ,φ ) r
VC (r ,θ ,φ ) = Z p Z t e ∫ r r /∫ dr ′
2
r − r′
Here Zpe and Zte denote the charges of the incident and target nuclei, respectively.
The deformed radius for the triaxial rotor under consideration has the form
⎡ ⎤
RV (r ,θ ,φ ) = r0 (A1p/ 3 + At1 / 3 )⎢1 + β 2(V ) cos γY20 + β 2 sin γ (Y22 + Y2 − 2 ) + β 4(V )Y40 ⎥
1 (V )
⎣ 2 ⎦
⎡ ⎤
RW (r ,θ ,φ ) = r0′(A1p/ 3 + At1 / 3 )⎢1 + β 2(W ) cos γY20 + β 2 sin γ (Y22 + Y2 − 2 ) + β 4(W )Y40 ⎥
1 (W )
⎣ 2 ⎦
⎡ ⎤
RC (r ,θ ,φ ) = rC (A1p/ 3 + At1 / 3 )⎢1 + β 2( C ) cos γY20 + β 2 sin γ (Y22 + Y2 − 2 ) + β 4(C )Y40 ⎥
1 (C )
⎣ 2 ⎦
The quantities Ap, and At are the atomic mass numbers of the projectile and the
target nucleus, respectively.
A number of optical potentials have been considered in the present work. These are
presented in Table 32.2 and are discussed in the following section. The angle γ was
set at 22° in accord with the requirement to describe the energy level spectrum for
24
Mg as discussed earlier. For simplicity, the real and imaginary nuclear potential
deformations were constrained to be identical (i.e. β 2(V ) = β 2(W ) = β 2( N ) and β 4(V ) = β 4(W ) =
β 4( N ) ) but otherwise were treated as free parameters. The quadrupole Coulomb
deformation β 2(C ) , was constrained to satisfy the relation β 2( N ) RC2 = 7.3 fm2, a value
24
determined by matching the relevant B( E 2) and Q2 + experimental values for Mg, as
1
discussed in the previous section.

In each case, the β 4( C ) value was determined by requiring that the ratio β 4(C ) / β 2(C ) be
identical to the corresponding ratio, β 4( N ) / β 2( N ) , for the nuclear deformation. It should
be noted that the predicted angular distributions were found to be relatively
insensitive to the inclusion of the β 4(C ) deformation.
Finally, in the coupled-channels calculations, it is necessary to specify the wave
functions for the relevant 24Mg states expressed as an expansion in the standard
IMK basis states (see the previous section). In each case, these wave functions
were determined using γ and β 4 / β 2 values consistent with those used in the
deformed nuclear and Coulomb potentials.
Optical model potential and shape deformation parameters

Previously reported (Nurzynski et al. 1981) optical model analysis of 16O scattering
from 24Mg and studies for neighbouring mass systems have led to three broad
classifications of possible potentials (see Chapter 30):
(i) Deep real potential wells (~ 80 MeV) with strong absorption strength and nearly
the same geometrical parameters ( r0′ ≈ r0 and a0′ ≈ a0 ).
(ii) Moderate real potential wells (~ 30 MeV) with weak absorption strength and
r0′ ≤ r0 , a0′ ≤ a0 .
(iii) Shallow real potential wells with moderate absorption strength and r0′ < r0 and
a0′ < a0 .
Potentials of types (ii) and (iii) give rise to surface transparency. In the work
described in this chapter, potentials in each of these classes have been examined.
Table 32.2
Optical model potentials
Figure 32.3. Angular distributions (dots) for the reactions 24

Mg(160,16O')24Mg* (0+,g.s.; 21+ , 1.37 MeV;
41+ , 4.12 MeV; 2+2 , 4.24 MeV) measured at Ec.m. = 43.5 MeV are compared with coupled-channels
calculations using potential 1 (solid curves) of Table 32.2 and potential 3 (dashed curves). The
deformation parameters are β 2( N ) = 0.30 and β 4( N ) = - 0.065 (for the solid curves) and β 2( N ) = 0.35
and β (N)
4 = - 0.10 (for the dashed curves).
Figure 32.3 shows typical results for a deep real potential well (potential 1) of Table
32.2, which is based upon potential 6 of Nurzynski et al. (1981) with β 2( N ) = 0.30 and
β 4( N ) = - 0.065 (solid curves). The figure contains also results for the shallow real
potential (potential 3 of Table 32.2). In these calculations, represented by dashed
lines in Figure 32.3, the best fits were obtained for the deformation parameters β 2( N ) =
0.35 and β 4( N ) = - 0.10. Potential 3 is similar to that used by Cramer et al. (1976) in
their analysis of 16O scattering from 28Si and by Eck et al. (1979) in analysis of the
24
Mg(160,16O')24Mg* scattering at 67 MeV bombarding energy. It can be seen that
both sets of calculations produce unsatisfactory fits to the experimental angular
distributions.
Figure 32.4 shows the results obtained for potential 2 of Table 32.2 with β 2( N ) = 0.25
and β 4( N ) = - 0.065. For this moderate real potential well, nearly perfect agreement
with the data is obtained for the magnitudes and shapes of the four angular
distributions, although some of the details are not well described. While the whole of
parameter space could not be searched, it seems unlikely that these fits could be
improved significantly.
Figure 32.4. Angular distributions (dots) for the reactions 24

Mg(16O,16O')24Mg* (0+,g.s.; 21+ , 1.37 MeV;
41+ , 4.12 MeV; 2+2 , 4.24 MeV) measured at Ec.m. = 43.5 MeV are compared with coupled-channels
calculations (solid curves) using potential 2 of Table 32.2, β 2( N ) = 0.25 and β 4( N ) = - 0.065.
Some of the discrepancies arise probably from insufficiencies of the extended

asymmetric rotor model employed for the calculations. For example, for simplicity
and consistency, the deformation parameters for the states of the K = 2 band were
taken to be identical with those for the ground-state band although the analysis of
van der Borg et al. (1979) suggests otherwise. Furthermore, the optical model
potentials were constrained to have Woods-Saxon form factors and no parity-
dependent term (Dehnhard, Shkolnik, and Franey 1978).
Potential 2 is similar to potential 11 of Nurzynski et al. (1981, see Chapter 30), which
was found to give a satisfactory description of the ground-state transition for the
24
Mg(16O,12C)28Si transfer reaction. However, it is less surface transparent since the
geometries of the real and imaginary potentials are similar. Attempts to fit the
present data with more surface transparent interactions led to too much structure at
larger angles. It is possible that the use of an explicit J - dependent absorption
potential (Chatwin et al. 1970) rather than a potential with r0′ ≤ r0 , a0′ ≤ a0 which is
supposed to simulate such an effect may lead to an improved description of the data.
These two kinds of surface transparent potentials are not equivalent (Kondo and
Tamura 1982).
Figure 32.5 shows the corresponding results for the 3+ (5,24 MeV), 6+ (8.11 MeV)
and 4+ (8.44 MeV) states. These cross sections are considerably smaller than those
for the lower states. We have found that the inclusion of these three states in the
coupling scheme affected only the cross section for the 2+2 state. This effect arises
from the intra-band coupling of the 3+ and 4+ levels with the 2+2 state and is only
slightly dependent upon whether the 41+ model state is associated with the level at
8.44 MeV or with that at 6.01 MeV. Figure 32.5 also shows the result, which include
a hexacontatetrapole deformation (a β 6( N )Y60 term) in the deformed optical potential.
For β 6( N ) = 0.0267, a value which gives a potential deformation distance
δ 6 ≡ β 6( N ) r0 ( A1p/ 3 + At1 / 3 ) = 0.188 fm, approximately equivalent to that employed by
Lombard, Escudié and Soyeur (1978) for proton scattering by 24Mg (i.e.
δ 6 = β 6( N ) r0 At1 / 3 =0.179 fm), it is seen that the peak cross section for the 6+ state is only
increased by about 25%. The effect of the finite β 6( N ) deformation on the other states
was much smaller.
Figure 32.5. Coupled-channels calculations (solid curves) of angular distributions for the reactions
24
Mg(16O,16O')24Mg* (3+ , 5.24 MeV; 6+ , 8.11 MeV; 4+, 8.44 MeV)'at Ec.m. = 43.5 MeV using potential 2
of Table 32.2, β 2( N ) = 0.25 and β 4( N ) = - 0.065. The dashed curve shows the effect of including a
β (N)
6 = 0.0267 term in the optical potential.
We have also studied the effect of either removing the hexadecapole deformation or
changing its sign. Figure 32.6 exhibits the result for β 4( N ) = 0 (solid curve) and β 4( N ) =
+0.065 (dashed curve), respectively. For β 4( N ) = 0, the fit to the angular distribution
for the 41+ state is less satisfactory both in magnitude and shape than the fit for β 4( N ) =
-0.065 of Figure 32.4, while for β 4( N ) = +0.065 an additional oscillation is introduced
into the calculated cross sections. Thus the negative value of β 4( N ) is strongly
favoured by the present analysis.
Figure 32.6. Coupled-channels calculations of the angular distribution for the reaction
24 +
Mg(16O,16O')24Mg* ( 41 , 4.12 MeV) at Ec.m. = 43.5 MeV using potential 2 of Table 32.2 with β 2( N ) =
0.25. The hexadecapole deformation was assumed to be either β 4( N ) = 0 (solid curve) or
positive, β
(N)
4 = +0.065, (dashed curve).
Table 32.3
24
Potential deformation distances for the asymmetric rotor model for Mg
a
) Eenmaa et al. (1974); b) Lovas et al. (1977); c) Lombard et al. (1978);
d e f
) Tamura (1965); ) van der Borg et al. (1979); ) Eck et al. (1981);
g h
) Our present work; ) Average values;
i
) δ2 = 1.59 fm employed for coupling of the K = 2 band states; δ4 = -0.48
employed for coupling of the g.s. with 4+2 model state; δ6 = 0.179 fm also included.
Table 32.3 displays the various potential deformation distances δ2 and δ4 obtained by
analyses of both light- and heavy-ion inelastic scattering data for 24Mg. This
comparison takes into account (Hendri 1973) the different parameterization of the
potential radii in terms of At1 / 3 and ( A1p/ 3 + At1 / 3 ) for light and heavy ion scattering,
respectively. It is seen that the present results are in good agreement with the light-
ion work. The earlier (16O,16O') analysis (Eck et al. 1981) employed a shallow real
potential well and the larger deformation distances obtained in this case are a further
argument against such an interaction.
Discussion and conclusion

Angular distributions for the elastic and inelastic scattering of 16O from 24Mg, exciting
the 21+ , 1.37 MeV, 41+ , 4.12 MeV and 2+2 , 4.24 MeV states have been measured at
Elab= 72.5 MeV (Ec.m. = 43.5 MeV) bombarding energy. The data have been
described well by the coupled-channels calculations within the framework of the
Davydov-Filippov (1958) asymmetric rotor model for the low-lying states of 24Mg,
which has been extended to include a symmetric hexadecapole shape component.
The analysis included coupling to the higher 3+ (5.24 MeV), 6+ (8.11 MeV) and 4+
(8.44 MeV) states. We have found that for these higher states only the intra-band
couplings between the 2+2 , 3+ and 4+ (8.44 MeV) states are significant.
The optical model potential for the 16O + 24Mg interaction was determined to have a
moderate real well depth confirming the earlier study by Nurzynski et al. (1981).
Furthermore, the potential appears to be sufficiently transparent for it to be
consistent with the possibility of a formation of quasi-molecular states for the 16O +
24
Mg system. The overall success of the extended asymmetric rotor model for 24Mg
indicates the possibility of excited states, additional to those described by a simple
symmetric rotor model as discussed in our earlier work (Nurzynski et al. 1981), being
strongly coupled to the elastic channel.
The shape parameters of the nuclear potential for the 16O + 24Mg system were found
to be β 2( N ) = 0.25, γ = 22° and β 4( N ) = -0.065. The corresponding deformation
distances are in good agreement with earlier light-ion results (see references in
Table 32.3). The negative β 4( N ) deformation parameter is well established by the
analysis of the 41+ angular distribution. The scattering analysis shows little sensitivity
to the parameter γ, although the 24Mg level spectrum constrains the value to be close
to γ = 22°.
The question, which arises as a consequence of the good description of 16O
scattering, obtained here and of previous works using light ions, is whether this
agreement means that 24Mg is a rigid rotor. As Yamakazi (1963) has pointed out, it is
difficult to distinguish whether the nuclear equilibrium shape is axially symmetric
( γ = 0 ) or asymmetric ( γ ≠ 0 ) by consideration of predictions for only the ground-
state K = 0 band and a K = 2 band based upon either γ - vibrations or a fixed γ -
deformation, respectively. The Hartree-Fock calculations of Grammaticos (1975)
show that 24Mg may be soft to vibrations in the γ - direction, depending upon the
effective interaction employed. If the K = 2 band arises from γ - vibrations, the results
of such a model are similar to those obtained for a rigid triaxial nucleus with an
effective intermediate value of γ, which should be considered as a "freezing
approximation" of the γ -vibrations (Lombard, Escudié and Soyeur 1978). In this
picture, the description of the 24Mg states in terms of a rigid asymmetric rotor,
parameterised by β2, γ and β4, is a convenient and relatively realistic way of
modelling the important couplings involved in the analysis of light- or heavy-ion
scattering to the low-lying levels of 24Mg.
References
Abe, Y., Kondo, Y. and Matsuse, T. 1980, Prog. Theor. Phys. Suppl. 68:303.
Abgrall, Y., Baron, G., Caurier, E. and Monsonego, G. 1969, Phys. Lett. 30B:376.
Aspelund, O. 1982, Phys. Lett. 113B:357.
Baker, F. T. 1979, Nucl. Phys. A331:39.
Baker, F. T., Scott, A., Cleary, T. P. Ford, J. L. C., Gross, E. E. and Hensley, D. C.
1979, Nucl. Phys. A321:222.
Batchelor, R., Ferguson, A. J., Gove, H. E. and Litherland, A. E. 1960, Nucl. Phys.
16:38.
Branford, D., McGough, A. C. and Wright, I. F. 1975, Nucl. Phys. A241:349.
Chatwin, R. A., Eck, J. S., Robson, D. and Richter, A. 1970, Phys. Rev. C1:795.
Cohen, A. V. and Cookson, J. A. 1962, Nucl. Phys. 29:604.
Cramer, J. G. DeVries, R. M., Goldberg, D. A., Zisman, M. S. and Maguire, C. F. 1976,
Davydov, A. S. and Filippov, G. F. 1958, Nucl. Phys. 8:237.
Davydov, A. S. and Rostovskii, V. S. 1959, ZhETF (USSR) 36:1788 [trans.: JETP (Sov.
Phys.) 9:1275]
Dehnhard, D., Shkolnik, V. and Franey, M. A. 1978, Phys. Rev. Lett. 40:1549.
Eck, J. S., Nurzynski, J., Ophel, T. R., Clark, P. D., Hebbard, D. F. and Weisser, D. C.
1981, Phys. Rev. C23:2068.
Eenmaa, J., Cole, R. K., Waddell, C. N., Sandhu, H. S. and Dittman, R. R. 1974, Nucl.
Phys. A218:125.
Eisenberg, J. M. and Greiner, W. 1975, Nuclear theory, vol. I (North-Holland,
Amsterdam,)
Endt, P. M. and van der Leun, C. 1978, Nucl. Phys. A310:1.
Faessler, A. and Greiner, W. 1962, Z. Phys. 168:425.
Fewell, M. P., Hinds, S., Kean, D. C. and Zabel, T. H. 1979, Nucl. Phys. A319:214.
Fink, H. J., Scheid, W. and Greiner, W. 1972, Nucl. Phys. A188:259.
Grammaticos, B. 1975, Nucl. Phys. A252:90.
Hendrie, D. L. 1973, Phys. Rev. Lett. 31:478.
Imanishi, B., 1968, Phys. Lett. 27B:267.
Imanishi, B., 1969, Nucl. Phys. A125:33.
Kendo, Y., Matsuse, T. and Abe, Y. 1978, Prog. Theor. Phys. 59:465; 1009; 1904
Kokame, J., Fukunaga, K., Inoue, N. and Nakamura, H. 1964, Phys. Lett. 8:342.
Kondo, Y. and Tamura, T. 1982, Phys. Lett. 109B:171.
Kurath, D. 1972, Phys. Rev. C5:768.
Leigh, J. R. and Ophel, T. R. 1982, Nucl. Instr. Methods 192:615.
Lombard, R. M., Escudie, J. L. and Soyeur, M. 1978, Phys. Rev. C18:42.
Lovas, I., Rogge, M., Schwinn, U., Turek, P., Ingham, D. and Mayer-Boricke, C. 1977,
Meyer, M. A., Reinecke, J. P. L. and Reitmann, D. 1972, Nucl. Phys. A185:625.
Michaud, G. and Vogt, E. W. 1969, Phys. Lett. 30B:85.
Michaud, G. and Vogt, E. W. 1972, Phys. Rev. C5:350.
Nurzynski, J. Ophel, T. R., Clark, P. D., Eck, J. S., Hebbard, D. F., Weisser, D. C.,
Robson, B. A. and Smith, R. 1981, Nucl. Phys. A363:253.
Ollerhead, R. W., Kuehner, J. A., Levesque, R. J. A. and Blackmore, E. W. 1968 Can.
J. Phys. 46:1381.
Raynal, J. 1972, Computing as a Language of Physics (IAEA, Vienna,) p. 281.
Raynal, J. 1980, Saclay Report D Ph-T/24/80, and private communication.
Robinson, S. W. and Bent, R. D. 1968, Phys. Rev. 168:1266.
Scheid, W., Greiner, W. and Lemmer, R. 1970, Phys. Rev. Lett. 25:176.
Strutt, J. W. (Lord Rayleigh) 1926, Theory of sound, vol. II (Macmillan, London,)
Tamura, T. 1965, Nucl. Phys. 73:241.
van der Borg, K., Harakeh, M. N. and Nilsson, B. S. 1979, Nucl. Phys. A325:31.
Yamakazi, T. 1963, Nucl. Phys. 49:1.
33
Spin Assignments for the 143Pm and 145Eu Isotopes
Key features:
1. Using j – dependence, spin assignments have been made for states belonging to the
2d5/2 and 2d3/2 configurations in 143Pm and 145Eu nuclei. Other spin assignments have
been also made using the unique 1g7/2, 3s1/2,and 1h11/2 proton configurations
accessible via the selected target nuclei.
2. Spectroscopic factors have been extracted and compared with the factors determined
using the (3He,d) reaction.
Abstract: Angular distributions have been measured for transitions to low-lying states in
143
Pm and 145Eu populated by the 142Nd(7Li,6He)143Pm and the 144Sm(7Li,6He)145Eu reactions
at E(7Li) = 52 MeV. Elastic scattering of 7Li at 52 MeV on 142Nd and 144Sm, and 6Li at 46 MeV
on 142Nd and at 45 MeV on 144Sm, were measured. Optical-model parameters extracted from
fits to the scattering data were used in the finite-range distorted waves analysis of the angular
distributions for levels below 1.40 MeV excitation energy in 143Pm and 1.84 MeV in 145Eu. The
reaction cross sections forward of 6° (c.m.) allow unambiguous distinction between 2d5/2 and
2d3/2 states. Final-state spins have been assigned to d - states in 143Pm and in 145Eu. Existing
assignments to other levels in both residual nuclei have been confirmed.
Introduction
As discussed in Chapter 29, lithium-induced, single-nucleon, stripping reactions
present a useful tool for extracting spectroscopic information complementary to that
obtained from light-ion work. For most cases the data can be well described by the
exact finite-range (EFR) distorted-waves Born approximation formalism. The
(7Li,6He) reaction involves the transfer of a proton from a p - wave orbit in the
projectile, and not from a predominantly s - state as is the case of light-ion reactions
such as (d,n), (3He,d) and (α,t). The observed j - dependence for such reactions can
be used to distinguish between spins belonging to 2d5/2 and 2d3/2 configurations.
In the work described in this chapter, the 142Nd(7Li,6He)143Pm and
144
Sm(7Li,6He)145Eu reactions have been studied. In addition, the elastic scattering of
7
Li and 6Li on 142Nd and 144Sm has been measured to obtain optical-model
parameters for the distorted waves calculations.
The 142Nd nucleus has 82 neutrons and 60 protons. Its 1g/2d/3s/1h neutron shell is
closed. Its N = 50 proton shell is also closed and the remaining 10 protons are in the
next, 1g/2d/3s/1h, shell. In the simple shell-model description, 8 of these protons
would occupy the 1g7/2 orbitals, and 2 would have a 2d5/2 configuration. However,
considering the residual interaction, there will be a mixture of other configurations.
Nevertheless, the 1g7/2 will be almost full but other configurations (2d5/2, 2d3/2,
3s1/2,and 1h11/2) will be available for the stripped proton. Thus the strongest
transitions could be expected to belong to these almost empty configurations in the
143
Pm nucleus.
The 144Sm nucleus also has 82 neutrons and thus its 1g/2d/3s/1h neutron shell is
also closed. However, this isotope has 62 protons. The presence of the extra two
protons should be expected to reduce the probability for transfers to the 1g7/2 orbitals
but should not affect significantly the transfers to other configurations.
Beams of 6Li- and 7Li- from a General lonex sputter source were injected into the
Australian National University 14UD Pelletron accelerator. Beam currents of up to
300 nA of 6Li3+ and 7Li3+ were obtained on target. Targets of enriched 142Nd (> 96%)
and 144Sm (> 96%), comprised of metal on thin carbon backings, were used. Target
thicknesses were ~ 25 μ/cm2, although thicker targets (~ 100-150 μ/cm2) could be
used for the required resolution of the 6He groups corresponding to observed states
in the residual nuclei.
Reaction data and elastic scattering data were measured with an Enge split-pole
spectrograph using a resistive-wire gas proportional detector (Ophel and Johnston
1978) located at the focal plane. From the energy loss (ΔE) and the position signal
( ∝ Bρ ) of the focal plane detector, a mass identification signal ( M 2 = ( Bρ ) 2 ΔE ) was
obtained as shown in Figure 33.1. The difference in magnetic rigidity between 6Li3+
and 6He2+ is sufficient to allow unambiguous mass identification. In our
measurements for transfer reactions, 6Li3+ ions did not enter the detector and
therefore were not interfering with the collection of data. Additionally, the high field
necessary to place the 6He particles onto the detector removed completely the 7Li3+
elastic events from the detector, allowing high beam currents to be used at forward
angles. The angular acceptance of the spectrograph was 1°. Fixed monitor detectors
at 15° and 30° were used for normalization between runs and to check on the target
deterioration.
Figure 33.1. A mass identification signal (mass squared) for the 7Li + 144Sm reaction at 5°. The 6Li
particles are excluded because their magnetic rigidity is such that they do not enter the detector.
To obtain information on the elastic scattering wave functions, needed in the

distorted waves analysis, we have also measured the 142Nd(7Li,7Li)142Nd and
144
Sm(7Li,7Li)144Sm at E(7Li) = 52 MeV and 142Nd(6Li,6Li)142Nd at E(6Li) = 46 MeV and
144
Sm(6Li,6Li)144Sm at E(6Li) = 45 MeV. The use of 6Li optical parameters to describe
6
He distorted waves has been shown to work well in the analysis of other (7Li,6He)
reactions (see Chapter 29).
Absolute cross sections were obtained by normalizing to the forward-angle elastic
scattering, where the cross section is purely Rutherford. The error in the absolute
normalization is estimated to be 5% for the elastic scattering, resulting mainly from
possible angle setting errors and uncertainties in the dead-time corrections. Based
on the reproducibility of the (7Li, 6He) data, the absolute cross sections of the transfer
reactions are accurate to ± 12%. The relative errors in the cross sections are shown
by the error bars on the individual data points where these are larger than the plotted
points.
Figure 33.2. A 6He spectrum for the 142Nd(7Li,6He)143Pm reaction at 32°. States in 143
Pm are labelled
6 144 7 6 145 145

Figure 33.3. A He spectrum for Sm( Li, He) Eu reaction at 18°. States in Eu are labelled with
the appropriate excitation energies.
Figures 33.2 and 33.3 show spectra of the 142Nd(7Li,6He)143Pm reaction at θlab = 32°,
and the 144Sm(7Li,6He)145Eu reaction at θlab = 18°. The resolution is 70 keV FWHM
and little background is evident at these angles. Unfortunately, the Q - value for the
12
C(7Li,6He)13N reaction is such that the ground-state group obscures the states at
1.76 and 1.84 MeV in 145Eu at angles forward of 10° (lab). The 13C(7Li,6He)14N
reaction was a less serious contaminant. A group, corresponding to the excitation of
the 3.95 MeV level in 14N, prevented the extraction of the cross section for the 1.17
MeV state of 143Pm at 3° lab.
The elastic scattering data were analysed using the standard optical-model potential
as described in Chapter 29. The computer code JIB (Perey 1967)29 was used to fit
the data, starting with the parameters used to fit the 140Ce(7Li,7Li) and 141Pr(6Li,6Li)
elastic scattering data at 52 and 47 MeV, respectively. The parameters were varied
two at a time until a minimum χ2 was obtained. The experimental angular
distributions and the optical-model fits are shown in Figures 33.4 and 33.5. The
extracted parameters are listed in Table 33.1.
Figure 33.4. Angular distribution for the 142Nd(7Li,7Li)142Nd at 52 MeV and 142
Nd(6Li,6Li)142Nd at 46
MeV elastic scattering. The solid lines are the optical-model fits to the data.
Table 33.1
Optical-model parameters
The radii are defined using At1 / 3 i.e. R0 = r0 At1 / 3 , R0′ = r0′At1 / 3 and Rc = rc At1 / 3 .
29
The same program, which I have modified and adapted to run at ANU, and which I have used to
support my study of nuclear reactions induced by light projectiles.
Exact finite-range (EFR) distorted waves calculations using the optical-model
parameters listed in Table 33. 1 were performed with the computer code LOLA
(DeVries 1973) for transitions to the strongly populated states observed in the
142
Nd(7Li,6He)143Pm and 144Sm(7Li,6He)145Eu reactions. Fifty-six partial waves were
used in the calculations and the radial integrations were carried out to a radius of 30
fm in steps of 0.13 fm. The single-particle bound-state wave functions for 7Li, 143Pm
and 145Eu were generated by the code assuming a volume Woods-Saxon form for
the interaction potential with r0 = 1.25 fm and a0 = 0.65 fm, and the spin-orbit factor λ
= 25. No spin-orbit potential was used in the distorted waves. The depths of the
potentials were adjusted so that the binding energy for the transferred proton in 7Li,
143
Pm and 145Eu were equivalent to the correct separation energies. The
experimental angular distributions and the EFR-distorted waves fits to the data are
shown in Figures 33.6 and 33.7.
Figure 33.5. Angular distribution for the 144Sm(7Li,7Li)144Sm at 52 MeV and 144
Sm(6Li,6Li)144Sm at 45
MeV elastic scattering. The solid lines are the optical-model fit to the data.
It has been shown in Chapter 29 that the angular distributions for (7Li,6He) leading to
2d5/2 and 2d3/2 final states can be distinguished at forward angles using j -
dependence. This distinction can be made even though l = 1, 2 and 3 are allowed for
both final states assuming a p3/2 transferred proton, because the Racah coefficient
multiplying the distorted waves cross section weights the transfers differently for the
two states. Thus the l = 1 component is 8 times stronger for a 2d5/2 state than for a
2d3/2 state so that the forward-angle cross section for a 2d5/2 state is larger than for a
2d3/2 state. Calculations for pure 2d5/2 and 2d3/2 are shown in Figure 33.6 for states at
0.0 MeV and 1.40 MeV in 143Pm, and in Figure 33.7 for states at 0.0 MeV, 1.042
MeV, 1.76 MeV, and 1.84 MeV in 145Eu. Clearly the data forward of 6° c.m. allow
unambiguous distinction to be made between 2d5/2 and 2d3/2 final states.
The absolute spectroscopic factor, extracted as described in Chapter 29, are listed in
Table 33.2 and 33.3, which also show spectroscopic factors obtained using (3He,d)
reactions (Wildenthal, Newman, and Auble 1971; Ishimatsu et al. 1970; Newman et
al. 1970). The errors in the absolute values of the spectroscopic factors include the
uncertainty in the absolute normalization of the experimental data and statistical
errors. Uncertainties resulting from the choice of optical-model parameters and from
the use of standard values for the bound-state potential parameters are not included.
Relative spectroscopic factors, normalized to the ground state, are also listed in
Tables 33.2 and 33.3.
Angular distributions for transitions to 3s1/2 levels via the l = 1 transfer at 1.17 MeV in
143
Pm and 0.809 MeV in 145Eu are well reproduced by the calculations but are
slightly out of phase by 1° to 2°, a problem which has been observed in the analysis
described in Chapter 29 and by Morre, Kemper, and Chalton (1975).
Figure 33.6. Angular distributions populated in the 142Nd(7Li,6He)143Pm reaction. The solid and dashed
lines are the EFR-distorted waves calculations normalized to the data.
Table 33.2
Spectroscopic factors for states in 143Pm
a
) Wildenthal et al. (1971); b) Ishimatsu et al. (1970); c) Our work.
144 7 6 145
Figure 33.7. Angular distributions populated in the Sm( Li, He) Eu reaction. The solid and dashed
lines are the EFR-distorted waves calculations normalized to the data.
Table 33.3
Spectroscopic factors for states in 145Eu
a
) Wildenthal et al. (1971); b) Newman et al. (1970); c) Our work.
Distorted waves calculations for transitions to 1g7/2, levels at 0.27 MeV in 143Pm and
0.329 MeV in 145Eu fit the experimental angular distributions extremely well except at
very forward angles where the calculated cross sections are smaller than the experi-
mental cross sections. However, this discrepancy could be due to the large statistical
errors present in the forward-angle data for g7/2 states, since, as expected (see the
Introduction) these states are not strongly populated.
Angular distributions are also shown for transitions to 1h11/2 levels at 0.96 MeV in
143
Pm and at 0.716 MeV in 145Eu. The distorted waves calculation agrees well with
the experimental points for the level at 0.716 MeV in 145Eu, but is out of phase by 3°-
4° for the level at 0.96 MeV in 143Pm.

We have measured angular distributions for the differential cross sections of the
proton transfer reactions 142Nd(7Li,6He)143Pm and 144Sm(7Li,6He)145Eu at 52 MeV
incident 7Li energy. We have also measured the elastic scattering 142Nd(7Li,7Li)142Nd
and 144Sm(7Li,7Li)144Sm at E(7Li) = 52 MeV and 142Nd(6Li,6Li)142Nd at E(6Li) = 46 MeV
and 144Sm(6Li,6Li)144Sm at E(6Li) = 45 MeV.
We have carried out theoretical analysis of the elastic scattering using the
conventional central, spherical optical model with volume absorption. We then used
the determined interaction parameters in the theoretical analysis of proton transfer
reactions using the exact finite-range distorted waves formalism.
The elastic scattering data were well described by the optical model. The distorted
waves calculations generally described the corresponding transfer angular
distributions well, although a slight phasing problem was encountered with s1/2 states
and more significantly with the 1h11/2 state at 0.96 MeV in 143Pm.
The absolute spectroscopic factors obtained here are slightly lower than those
obtained from the light-ion (3He,d) reactions, but the relative spectroscopic factors
are in good agreement.
Heavy-ion forward-angle j - dependence has been used to assign the following spins
to d - states in 143Pm: 0.0 MeV (5/2+), 1.40 MeV (3/2+); and in 145Eu: 0.0 MeV (5/2+),
1.042 MeV (3/2+). Spins could not be assigned to the d - states at 1.76 MeV and 1.84
MeV in 145Eu due to the lack of forward-angle data. Previous spin assignments for
these levels are given in the compilation by Lederer and Shirley (1978) as 0.0 (5/2+)
and 1.40 (3/2+) in 143Pm; and 0.0 (5/2+), 1.042 (3/2+), 1.76 (3/2+) and 1.84 (3/2+) in
145
Eu. The (7Li,6He) single-proton stripping reactions have been confirmed as a
useful spectroscopic tool.
References
Cohen, S. and Kurath, D. 1967, Nucl. Phys. A101:1.
DeVries, R. M. 1973, Phys. Rev. C8:951.
Ishimatsu, T., Ohmura, H., Awaya, T., Nakagawa, T., Orihara, H. and Yagi, K. 1970, J.
Phys. Soc. Jap. 28:291.
Lederer, C. M. and Shirley, V. S. 1978, Table of isotopes, 7th Edition.
Moore, G. E. Kemper, K. W. and Charlton, L. A. 1975, Phys. Rev. C11:1099.
Newman, E., Toth, K. S., Auble, R. L., Gaedke, R. M. and Roche, M. F. 1970, Phys.
Rev. C1:1118.
Ophel, T. R. and Johnston, A. 1978, Nucl. Instr. Methods 157:461.
Perey, F. G. 1967, Phys. Rev. 131:745.
Wildenthal, B. H., Newman, E. and Auble, R. L. 1971, Phys. Rev. C3:1199.
34
Search for Structures in the 16O + 24Mg Interaction
Key features:
1. Our measurements of excitation functions at forward angles for the 24Mg(16O,12C)28Si

α - transfer reaction has led to a discovery of three broad resonances. We have
argued that these resonant structures, together with similar features at lower incident
energies, are associated with nuclear molecular excitations.
2. In order to study further nuclear molecular excitations, we have now carried out
measurements of excitation functions for the 16O + 24Mg elastic and inelastic
scattering over a wide range of the incident 16O energies at a forward angle.
3. We have found no correlation between excitation functions for the 24Mg(16O,12C)28Si
reaction and 16O + 24Mg scattering. Thus the forward angle excitation functions for
16 24
the O + Mg system display no evidence of nuclear molecular excitations.
16 24
4. We have found that the excitation functions for the elastic and inelastic O + Mg
scattering can be well described using coupled channels formalism.
5. Curious irregularities, which cannot be reproduced by coupled channels calculations,
have been observed in the excitation functions corresponding to the 8.11 MeV excited
state in 24Mg and at 8.4 MeV excitation energy, which represents a group of
excitations belonging to both 16O and 24Mg. We argue that these irregularities cannot
be associated with nuclear molecular excitations.
6. Our work shows that the resonances observed for the 24Mg(16O,12C)28Si reaction are
most probably associated with processes in the exit channel.
16 24
Abstract. Excitation functions for the scattering of O from Mg with excitation energies (Ex)
up to 8.4 MeV have been measured for θlab = 19.5° and 33.6 MeV < Ec.m. < 49.2 MeV. Strong
energy dependence is observed for states above the 8 MeV excitation energy. Coupled-
channels calculations, which predict smooth energy variations, give good agreement with the
data for Ex < 8 MeV.
Introduction
The 24Mg(16O,12C)28Si reaction, discussed in Chapter 30, has led to uncovering
interesting and challenging features, described by a number of authors (Nurzynski et
al. 1981; Paul et al. 1978; Sanders et al. 1985). In particular, gross structures (with
widths Γc.m.~1-3 MeV) have been observed in excitation functions at forward angles
in the case of the transition to the 28Si ground state for the energy range 23 MeV ≤
Ec.m. ≤ 53 MeV. These structures have been assigned parities, and in some cases
spins, by analysis of both angular distributions (Nurzynski et al. 1981; Paul et al.
1978) and excitation functions at θc.m. = 0°, 90° and 180° (Paul et al. 1980; Sanders
et al. 1980a, 1985). Furthermore, it has been found that the forward-angle structures
are correlated for a number of 28Si states up to 10 MeV excitation (Sanders et al.
1980b).
In view of these previous investigations, it seemed natural to look for correlations
between the resonant structures at forward angles for the 24Mg(16O,12C)28Si reaction
and the 16O + 24Mg interaction. Unfortunately, most of available data at that time
were for backward angles (Clover et al. 1979; Lee at al. 1979; Paul et al. 1980),
which were difficult to interpret. Structures at these angles are highly fractionated
and show little apparent correlation with the forward-angle α - transfer structures
(Sanders et al. 1985).
An alternative way is to look for correlations in elastic and inelastic scattering data at
forward angles. One set of measurements (Mitting et al. 1974) has been reported for
16
O + 24Mg elastic and inelastic scattering at θlab = 35° and 50° for 15 MeV ≤ Ec.m. ≤
39 MeV which may contain some resonant structure. Fulton et al. (1983) have
measured excitation function for the 24Mg(160,16O')24Mg* (2+ 1.37 MeV) scattering at
20° < θlab < 40° and 25 MeV < Ec.m. < 39 MeV, which showed no resonant structure.
However, it should be noted that their work covered only a part of the energy range
for which structure has been observed for the α - transfer reaction. Furthermore, the
simple band-crossing model predicts (Nurzynski et al. 1981) that the 2+ 1.37 MeV
channel monitored by Fulton et al. should be active at Ec.m. ~18MeV (i.e. ~7 MeV
below the region investigated by Fulton et al.) and hence their measurement may be
insensitive to the type of effect proposed by Nurzynski et al. (1981).
In the study described in this chapter, we have carried out measurements of
excitation functions at θlab = 19.5° for the 16O + 24Mg elastic and inelastic scattering,
which complement the work of Fulton et al. (1983) by extending the data up to Ec.m. ~
50 MeV for excitation energies up to 8.4 MeV. We have also carried out coupled
channels analysis of the data.
Experimental methods and results

In the measurements described here, the experimental arrangement was similar to
that described by Nurzynski et al. (1981, 1982 see also Chapters 30 and 32). A
beam of 16O ions was provided by the Australian National University 14D Pelletron
accelerator. Particles scattered from a thin (~5 μg/cm2) 24Mg (enriched to 99.92%)
target were detected using a multi-element detector in the focal plane of a split-pole
Enge spectrometer. The forward angle of 19.5° (lab) was chosen to be as close as
possible to 5° (lab), the angle selected for the earlier (16O,12C) measurements
(Nurzynski et al. 1981), whilst at the same time minimising the interference caused
by scattering from carbon and oxygen contaminants in the target. The horizontal
acceptance angle of the spectrometer was 1°, which ensured good resolution of
contaminant peaks.
Figure 34.1. The energy spectrum of 16O particles at θlab = 19.5° and Ec.m = 41.4 MeV. Shaded peaks
correspond to the 16O + 24Mg scattering and indicate groups of particles (labelled according to their
excitation energies) for which excitation functions were measured. Scattering from carbon and oxygen
contaminants in the target is also indicated in the spectrum.
The energy resolution of peaks corresponding to scattering from 24Mg was ≤140 keV.
A typical 16O spectrum is shown in Figure 34.1. Oxygen contamination in the target
was small. However, some interference form 16O + 16O (Ex = 6.05/6.13 MeV)
scattering was present in the 8.4 MeV group at Ec.m.<37.2 MeV.
The incident beam intensity was measured using a beam current integrator. A solid-
state detector at 15° was also used as a target monitor. The data were recorded
event by event on to magnetic tapes and care was taken to obtain acceptable
statistics for weakly excited states. The peak intensities for close-lying groups were
extracted by fitting Gaussian distributions.
Table 34.1
The observed particle groups identified by their |Q| - values for the 16O + 24Mg scattering and their
contributing components.
Figure 34.2 shows the measured excitation functions for the scattering of 16O from
24
Mg. The displayed excitation energies (Ex) are for the groups of particles identified
by their |Q| - values in Table 34.1. The error bars include both statistical and
background subtraction errors. The known excited states contributing to these
groups of 16O particles are listed in Table 34.1.
In general, the data display smooth energy dependences in agreement with earlier
lower energy measurements at forward angles for the 1.37 MeV state (Fulton et al.
1983). However, the excitation functions corresponding to Ex = 8.11 and 8.4 MeV
display strong energy-dependent structures. A data point at 42.6 MeV (c.m.) for Ex =
8.11 MeV cannot be shown. At this energy no peak corresponding to Ex = 8.11 MeV
in the spectrum could be distinguished from the background. The observed
intensities for this group were also small at energies < 31 MeV. The vertical lines in
Figure 34.2 indicate the energies at which peaks occur in the forward-angle
24
Mg(16O,12C)28Si(g.s.) yield (Nurzynski et al. 1981, Sanders et al. 1985). As can be
16
seen there is no obvious correlation between the present measurements for the O
+ 24Mg scattering and the results for the 24Mg(16O,12C)28S reaction.
An attempt was made to measure excitation functions for 16O + 24Mg scattering at
other forward angles. However, the yield for the Ex = 8.11 MeV group was small at
these anglers and the contaminant interference was prohibitively high.
Figure 34.2. Measured excitation functions for the scattering of 16O from 24Mg with 0 ≤ Ex ≤ 8.4 MeV
(see Table 34.1) at θlab = 19.5°. The vertical lines indicate the energies at which peaks occur in the
forward-angle excitation function for the 24Mg(16O,I2C)28Si (g.s.) reaction (Nurzynski et al. 1981,
Sanders et al. 1985).
Coupled-channels calculations for the elastic and inelastic scattering of I6O from
24
Mg were performed using the computer code ECIS (Raynal 1972, 1981)30. The 0+
(g.s.), 2+ (1.37 MeV), 4+ (4.12 MeV), 2+ (4.24 MeV), 3+ (5.24 MeV), 6+ (8.11 MeV)
and 4+ (8.44 MeV) states in 24Mg were included in the calculations. The deformed
optical potential employed for each channel was taken to be potential 2 of Nurzynski
at al. (1983, see also Chapter 32), which was determined by fitting angular
30
See Chapter 16.
distributions at Ec.m. = 43.5 MeV for the lowest four levels of 24Mg. The shape
parameters for the nuclear potential were taken to be the same as determined by
Nurzynski et al. (1983) i.e β 2 = 0.25, γ = 22° and β 4 = -0.065 (see Chapter 32). The
wave functions for the 24Mg states were obtained using the rigid asymmetric rotor
model of Davydov and Filippov (1958) as extended by Baker (1979) and Barker et al.
(1979) to include a symmetric hexadecapole deformation, with shape parameters
consistent with those employed in the deformed optical potential. All the calculations
employed the Coulomb correction technique developed by Raynal (1980, 1981) with
matching at a radius of 15.2 fm.
Figure 34.3. Excitation functions for the reaction 24Mg(16O,16O') 24Mg* (Ex = 0, 1.37, 4.2 and 8.11 MeV)
at θlab = 19.5°. Coupled-channels calculations (full lines) are compared with the data. For the
unresolved 4.2 MeV doublet, the full curve is the sum of the calculated contributions from the 4.12
MeV (dotted curve) and the 4.24 MeV (broken curve) states.
Figure 34.3 shows the coupled-channels predictions (full curves) for several
excitation functions compared with the data at θlab = 19.5° and for 36 MeV ≤ Ec.m. ≤
48 MeV. In the case of the unresolved 4.2 MeV doublet, the full curve is the sum of
the calculated contributions from the 4.12 MeV (dotted curve) and the 4.24 MeV
(broken curve) states of 24Mg. It should be noted that these calculations employed
the same potential parameters as given by Nurzynski et al. (1983) with no
modification for changes in the bombarding energy.
As can be seen, coupled channels calculations produce smooth functions with no
indication of any resonant structure. The broad maxima and minima arise purely
from angular effects associated with the oscillatory nature of the angular
distributions. In these calculations, nuclear potential effects are dominant. This can
be inferred from the angular distribution data (Nurzynski et al. 1983) at Ec.m. = 43.5
MeV, which show that at θlab = 19.5° (θc.m. = 32.36°) the Rutherford ratio has fallen to
approximately 0.1.
Figure 34.3 shows that coupled-channels calculations give a good description of
both the magnitudes and the general features of the data for the lowest three
excitation energies. For the 8.11 MeV state, the calculations predict only the general
magnitude of the cross section. However, the observed structure of the data is not
reproduced and, within the present model, the structure does not arise from an
angular effect.
The calculated excitation functions for the 5.24 MeV and 8.44 MeV states of 24Mg
(not shown in Figure 34.3) were also found to have smooth behaviours. In the case
of the 5.24 MeV state, the calculated cross section was found to be <0.06 mb/sr,
which is consistent with the fact that this state was generally not observed. For the
8.44 MeV state, the predicted excitation function lies below the data. This is not
surprising, since the data for the 8.4 MeV excitation function are believed to include
contributions from five unresolved states (see Table 34.1).
Discussion and summary

The anomalous structure observed for the 6+ 8.11 MeV (and possibly for the 8.4
MeV) excitation function is not readily explained. If the structure is interpreted as
arising from shape resonances in the ion-ion potentials, it is necessary for the
potential associated with the 6+ 8.11 MeV channel to be significantly more surface
transparent than that employed in the present coupled-channels calculations in order
to enhance the grazing partial waves in this channel. Such a large difference
between the absorptive strength for the 6+ channel and for channels with lower
excitation energies seems unlikely and artificial. Moreover, in this picture one should
expect to observe strong correlations with the resonant structures found in the α -
transfer reaction at forward angles (Nurzynski et al. 1981; Paul et al. 1978; Sanders
et al. 1985). However, as shown in Figure 34.2, no such correlations are present in
the 16O + 24Mg excitation functions.
An alternative interpretation (Nurzynski et al. 1981) ascribes the forward-angle α -
transfer structures to a fragmentation of the potential shape resonances, arising from
coupling between the elastic and inelastic channels of both the initial and final
systems. Since there is no evidence of a β6 deformation parameter which could alter
the 6+ inelastic scattering cross section (Nurzynski et al. 1983, see also Chapter 32),
this interpretation requires the additional coupling of intermediate 2+ or 4+ states, via
β2 and β4 terms. However, as the data indicate no significant correlated structures in
the excitation functions for these intermediate states, this explanation (which again
requires a more surface-transparent potential for the 6+ inelastic channel) appears
unlikely.
Concerning the gross resonant structures observed in the 24Mg(16O,12C)28Si reaction
at forward angles, it is interesting to note that the fusion cross sections for the 12C +
28
Si system exhibit some resonance-like structures (Racca et al. 1983) while those
for the 16O + 24Mg system show only weak and uncorrelated structures (Racca et al.
1982).
These facts combined with the present work, which shows smooth excitation
functions described well by the coupled-channels calculations, suggest that the
structures seen in the α - transfer reaction may be associated with resonating
processes in the exit 12C + 28Si channel rather than with those in the entrance
channel.
References
Baker, F. T. 1979, Nucl. Phys. A331:39.
Baker, F. T., Scott, A., Cleary, T. P., Ford, J. L. C., Gross, E. E. and Hensley, D. C.
1979, Nucl. Phys. A321:222.
Clover, M. R., Fulton, B. R., Ost, R. and DeVries, R. M. 1979, Phys. G: Nucl. Phys.
5:L63.
Davydov, A. S. and Filippov, G. F. 1958, Nucl. Phys. 8:237.
Fulton, B. R., Lilley, J. S., Cormier, T. M. and Stwertka, P. M. 1983, Phys. Rev.
C27:1811.
Lee, S. M., Adloff, J. C., Chevallier, P., Disdier, D., Rauch, V. and Scheibling, F. 1979,
Mittig, W., Charles, P., Lee, S. M., Badawy, I., Berthier, B., Fernandez, B. and
Gastebois, J. 1974, Nucl. Phys. A233:48.
Nurzynski, J., Ophel, T. R., Clark, P. D., Eck, J. S., Hebbard, D. F., Weisser, D. C.,
Robson, B. A. and Smith, R. 1981, Nucl. Phys. A363:253.
Nurzynski, J., Atwood, C. H., Ophel, T. R., Hebbard, D. F., Robson, B. A. and Smith, R.
1983, Nucl. Phys. A399:259.
Paul, M., Sanders, S. J., Cseh, J., Geesaman, D. F., Henning, W., Kovar, D. G., Olmer,
C. and Schiffer, J. P. 1978, Phys. Rev. Lett. 40:1310.
Paul, M., Sanders, S. J., Geesaman, D. F., Henning, W., Kovar, D. G., Olmer, C.,
Schiffer, J. P., Barrette, J. and LeVine, M. J. 1980, Phys. Rev. C21:1802.
Racca, R. A., Prosser, F. W., Davids, C. N. and Kovar, D. G. 1982, Phys. Rev.
C26:2022.
Racca, R. A., DeYoung, P. A., Kolata, J. J. and Thornburg, R. J. 1983, Phys. Lett.
B129:294.
Raynal, J. 1972, Computing as a Language of Physics (IAEA, Vienna,) p. 281.
Raynal, J. 1980, Saclay Report D Ph-T/24/80, and private communication.
Sanders, S. J., Paul, M., Cseh, J., Geesaman, D. F., Henning, W., Kovar, D. G., Kozub,
R., Olmer, C. and Schiffer ,J. P. 1980a, Phys. Rev. C21:1810.
Sanders, S. J., Olmer, C., Geesaman, D. F., Henning, W., Kovar, D. G., Paul, M. and
Schiffer, J. P. 1980b, Phys. Rev. C22:1914.
Sanders, S. J., Ernst, H., Henning, W., Jachcinski, C., Kovar, D. G., Schiffer, J. P. and
Barrette, J. 1985, Phys. Rev. C31:1775.
35
Parity-dependent Interaction
Key features:
1. We have measured excitation functions for the 16O + 20Ne (g.s.) and 16
O + 20
Ne*
(1.634 MeV) in the energy range of 21.5 MeV < Ec.m. < 31.2 MeV.
2. We have analysed our 16O + 20Ne (g.s.) experimental results and the results of
Schimizu et al. (1982) using an extended optical model potential containing a parity-
dependent interaction.
3. We have found that the interaction parameters determined in previous studies
(Kondo, Robson, and Smith 1985) resulted in an inadequate description of
experimental data.
4. We have found new interaction parameters (potential 3, Table 35.1) which gives
excellent description of our 16O + 20Ne data. The potential contains a negative parity-
dependent component, which means that it is deeper for the even partial waves. The
potential assigns even spin shape resonances J = 16+, 18+, and 20+ at Ec.m. = 22.5,
25.4, and 28.5 MeV, respectively.
5. We have also carried out parity independent calculations but included explicitly
contributions from α – transfer. These calculations reproduce the shape of the
excitation functions reasonably well and thus support the claim (see Chapter 24 and
references in the text) that the parity-dependent interaction simulates coupling to non-
elastic channels.
Abstract: The excitation functions for the 16O + 20Ne elastic scattering at θc.m. = 90° and for
the 16O + 20Ne* (Ex = 1.634 MeV) inelastic scattering corresponding to θc.m. = 90.95° - 91.45°
have been measured over the energy range 21.5 MeV < Ec.m. < 31.2 MeV. The 16O + 20Ne
elastic scattering was analysed within the framework of an extended optical model, in order to
place constraints on spin assignments to resonant states. Excellent description is obtained
with a potential, which is deeper for the even partial waves.
Introduction
Resonant phenomena in heavy-ion interactions have been observed for systems
involving sd - shell nuclei for elastic, inelastic and a variety of transfer channels (see
Chapter 30). One interesting case is the 16O + 20Ne system, for which prominent
structures have been observed in the excitation functions at forward and backward
angles for the elastic scattering, inelastic scattering to the 20Ne( 21+ ) and 20Ne( 41+ )
states, and for the 20Ne(16O,12C)24Mg reaction (Schimizu et al. 1982).
These data have been described using optical model and coupled channel
techniques, the essential feature of which is that the ion-ion potentials employed
include both the J - dependent absorptive term and a real parity-dependent
interaction. It has been argued that the J - dependent absorption arises from the
requirement that angular momentum and energy must be simultaneously conserved
in the open reaction channels (Chatwin et al. 1970), whilst the parity dependence is
to be expected as a consequence of the Pauli principle (Baye, Deenen, and
Salomon 1977).
The parity dependent term has the effect of staggering the shape resonances, which
arise for surface partial waves, to form "doublets" of resonances. The J - dependent
absorptive term is suitably transparent to surface partial waves so that the "doublets"
appear as unresolved gross structures in the predicted 16O + 20Ne elastic scattering
excitation function at backward angles, in accord with experiment.
It should be noted, however, that in determining the values of parameters used in the
potentials for such analyses, it is necessary to make Jπ assignments to peaks
observed in 16O + 20Ne scattering and 20Ne(16O,12C)24Mg reaction excitation-function
data (Schimizu et al. 1982). The assignments made in optical model analyses were
based upon assignments of 18+, 19- and 24+ to observed peaks at Ec.m. = 24.5, 25.4
and 35.5 MeV respectively, these assignments being consistent with values
extracted by Schimizu et al. (1982), using PJ2 (cosθ ) comparisons with measured
angular distributions. Furthermore, to avoid the possibility of having too many
resonances within the energy range of interest, it was assumed that each resonance
of a given spin and parity was that with the lowest energy.
An ambiguity arises, however, when it is acknowledged that a PJ2 (cosθ ) fitting
technique allows an assignment of spin values to ± 1h at very best. For this reason,
optical potentials were also determined using alternative assignments of spins to
observed peaks allowed within such a ± 1h uncertainty. The main consequence of
such a one-unit change in spin assignment is a reversal of the sign of the parity
dependent interaction in the ion-ion potential. Investigations of the available 16O +
20
Ne elastic scattering and 20Ne(160,12C)24Mg (g.s.) data were unable to resolve
conclusively this ambiguity in sign.
Studies by Kondo et al. (1985) re-examined the 20Ne(160,16O)20Ne excitation function
data within the framework of a model, which attributed the parity dependent term in
the optical model calculations to a requirement to simulate the effects of α - transfer
amplitudes. The results of that work supported strongly a choice of sign for the parity
dependent term, which would be consistent with spin assignments of 17-, 18+ and 23-
to the observed peaks at Ec.m. = 24.5, 25.4 and 35.5 MeV respectively, rather than
the 18+, 19- and 24+ assignments made previously.
In order to resolve the above ambiguity, we have undertaken measurements and
analysis of the θc.m. = 90° excitation function for 16O + 20Ne elastic scattering over the
energy range 21.5 MeV < Ec.m. < 31.2 MeV. Such measurements and analyses place
additional constraints on the resonance spins and hence constraints on the parity
dependent part of the 16O + 20Ne potential. Data have been taken also for the
excitation of the 1.634 MeV (2+) state of 20Ne.

A beam of 16O ions from the Australian National University 14UD Pelletron
accelerator was used to bombard natural neon target retained by thin windows in a
small gas cell shown in Figure 35.1.
The design requirements for the gas cell and detector system were an angular
aperture in the horizontal plane of <1° (lab) and an angular positioning accuracy of
0.1° or better at the required laboratory angle of 51.34°. The reaction energy at the
centre of the gas target should be calculable to better than 0.1 MeV (lab) and with a
range of reaction energies limited to a spread of the order of 0.2 MeV (lab). The
detector should be able to resolve elastic scattering by 20Ne from inelastic scattering
leaving 20Ne in its first excited state at 1.634 MeV and from elastic scattering by the
22
Ne component (9.2%) of natural neon. These design requirements were achieved.
Figure 35.1. A horizontal section through the gas target at the beam height, showing the windows, the
slits mounted on a cylindrical plug, anti-scatter apertures, windows and the defined reaction volume.
The incoming 16O beam was collimated to 3 mm vertically by 1 mm horizontally by

two collimators, one 700 mm and the other 106 mm upstream from the centre of the
main target chamber. Optical alignment to a precision of <0.1 mm ensured the co-
linearity of the two collimators and the centre of the main target chamber horizontally
in the incoming beam direction. The alignment precision in the vertical direction was
better than 0.2 mm.
The gas cell was then aligned, using the following procedure: The gas cell body was
mounted centrally in the main target chamber at the correct height. Its beam inlet
and outlet ports were mechanically aligned, by rotation, to the direction of the nearer
collimator and the cell body was clamped. A cylindrical plug carrying the target
length-defining slits and the anti-scatter slits was inserted along the vertical axis of
the gas cell and was mechanically aligned, using the slits themselves, by rotation to
the beam axis now defined by the beam inlet port. These slits, mounted on plane
faces milled parallel to the cylindrical plug axis, had been previously aligned during
assembly. All mechanical alignments were performed with accuracy better than 0.1
mm. The beam collimators ensured that sideways excursions of the beam were
limited to less than ± 0.2 mm with an expected half-intensity at ±0.1 mm. Vertical
beam excursions were limited to ±1 mm by an aperture mounted on the plug carrying
the beam entrance window.
The gas target assembly was completed by the insertion into the target cell body of
plugs carrying the thin windows. These were then covered by anti-scatter apertures.
The 16O beam enters the gas target through a nickel window and leaves through
another nickel window. The ejectiles leave through Mylar windows (210 (μg/cm2).
The energy losses of these windows were determined in a separate experiment,
measuring to an accuracy of 5 keV the loss after transmission of 48.83 MeV 16O ions
through the windows.
Two detectors were provided, for increased count rates and for redundancy
checking. They each consisted of a tantalum entrance aperture, 5x2 mm, a 210
μg/cm2 Mylar entrance window, 28mm length of isobutane gas at 50 Torr (constant
flow) in a grided ionisation chamber and a silicon surface barrier detector for residual
energy measurement. The performance of these detectors was such that ejectiles of
different Z could be readily distinguished and the required overall energy resolution
was attained. They were mounted at ±51.35 ± 0.05° (lab) with the entrance apertures
of 133.1 mm from the centre of the gas target, subtending a solid angle of 0.565 msr
each.
Slits inside the gas target were mounted 5 mm from the target centre with an
aperture of 2 mm. Combined with the detector aperture, these defined a length along
the beam axis of 2.56 mm for which the full detector aperture was visible and a
further ~0.1 mm at each end with a reduced detector aperture. The available angular
range distribution (taking the solid angle into account) was approximately triangular
with 75% of the distribution within ±0.43° (lab) and the remainder within ±0.86°. The
count rates in the individual detectors were influenced by sideways misalignment of
the beam, by differences in the target defining geometries, in the detector solid
angles and in the detector angular settings.
An early run disclosed unequal count rates in the two detectors for Rutherford
scattering from xenon. An optical-mechanical check showed a zero error of 0.25° in
the angular scale used inside the main target chamber, compared with the
established beam axis. Compensation for this zero error then gave detector count
rate comparisons agreeing to better than 1 %.
The evacuated gas target and ballast volume (6.5 litre) were first filled in a
reproducible way to ~2 Torr by expansion of a limited volume of xenon gas at bottle
pressure. The amount of xenon transferred decreased by 0.5% for each fill but the
absolute pressure of xenon was only known to ±10%. The target and ballast volume
were then filled to ~100 Torr total pressure with natural neon. This pressure was
known to 3%. Because of a slow leak, the total pressure was recorded before and
after each run to provide an average pressure from which the reaction energy was
calculated. Intermittently it was necessary to top up with neon gas and at such times
the partial pressure ratio of xenon and neon was recalculated.
It was important that the reaction energy be reliably calculated. A computer program,
which reproduced the known beam energy loss through the nickel entry window, was
used to calculate energy loss and straggling to the reaction volume and followed the
ejectile through to the residual energy detector. The nickel window gave energy
losses from 1.55 to 2.24 MeV over the range of beam energies used. Calculations
for the individual gas fills gave a further loss of 0.5 to 0.6 MeV to the target centre
over the same beam energy range. Calibrations of the residual energy detectors
based on observed pulse height responses and calculated energies were linear from
5.4 to 45.5 MeV, passing through zero as expected. An unsatisfactory calibration
would lead to non-linearity and a non-zero extrapolation. Therefore we are confident
that the first ~0.5 MeV of energy loss to the reaction volume has been calculated
satisfactorily. The same computer program was also useful in identification of the
elastic or inelastic scattering in the cases where only one of these spectrum peaks
was clearly visible. The predicted position was always found to be within the FWHM
range when identification was no problem. Figure 35.2 shows a spectrum from one
of the detectors taken at the lowest beam energy measured, where the resolution
between the particle groups was poorest but the intensity of both groups was high.
This spectrum has had the identification requirement for 16O particles imposed upon
it.
Figure 35.2. A spectrum of 16O ions scattered from the reaction volume and entering the residual
energy detector. Other ion species have been rejected by the ΔE – Eresidual requirement. The
inelastically scattered group leaving 20Ne in its first excited state enters the residual energy detector
with only 5.4 MeV for 16O beam energy of 40.0 MeV.
Figure 35.3. The absolute cross sections extracted for the elastic and inelastic scattering of 16O from
20
Ne. The rectangle widths show the range of energy averaging while the rectangle heights are ±1
standard deviation total error estimates. Relative errors, excluding the contribution from the absolute
xenon pressure, are somewhat smaller.
From spectra such as these, the yields of scattered 16O from 20Ne and xenon were
extracted. Data were only accepted if there were no significant differences between
the estimated backgrounds or between any of the intensities of the three peaks
mentioned above when spectra from the two independent detectors were compared.
The absolute cross sections for elastic and inelastic scattering from 20Ne are
dependent only on the intensities of the 20Ne peaks extracted, the intensity of
Rutherford scattering from the xenon isotopes and the ratio of partial pressures for
xenon and for 20Ne. The calculated mean reaction energy enters through the
calculation of the Rutherford scattering cross section for the xenon. The only
significant contributions to uncertainties in these absolute cross sections come from
the 20Ne peak statistics and from the uncertainty (±10%) in the absolute pressure of
xenon. All other contributions are small compared with these.
Figure 35.3 shows these cross sections as a function of reaction energy as
rectangles. The height of a rectangle shows ±1 standard deviation for the absolute
cross section determinations while the width of a rectangle shows the calculated
range of reaction energies over which the data are averaged.
The measured θc.m. = 90° (Figure 35.3) and θc.m. = 154°± 2° (Schimizu et al. 1982)
excitation functions for elastic scattering of 16O ions from 20Ne have been analysed in
terms of an extended optical model. The 16O + 20Ne potential was assumed to have
the following form:
U (r ) = Vc (r ) + [V ( Ec.m. , L) + iW ( J )] f (r )
where Vc (r ) is the Coulomb potential for a uniform charge distribution of radius R and
f(r) is the usual Woods-Saxon form factor with diffuseness a0 and radius
R = r0 [(16)1 / 3 + (20)1 / 3 ]
In the above, L is the relative orbital angular momentum and J is the total angular
momentum. For the elastic scattering of spinless projectiles J = L.
The depth of the real nuclear potential is given by
V ( Ec.m. , L) = V0 + VE Ec.m. + (−1) L Vπ
and consists of three terms: a constant V0 ; an energy dependent term; and an

explicit parity dependence.
The depth of the imaginary nuclear potential is given by
W ( J ) = W0 {1 + exp[( J − J c ) / Δ ]}
−1
where J c is a cut-off angular momentum and Δ is a diffuseness parameter. For each

energy, J c is parameterised by the expression (Chatwin et al. 1970):
J c = R [(2μ / h 2 )( Ec.m. − Q )]1 / 2
where R and Q represent average values of the radius and the threshold energy for
the predominant non-elastic reactions, respectively, and μ, is the reduced mass of
the system.
Parameter sets used in our theoretical analysis are listed in Table 35.1.
Table 35.1
Optical model parameters
For all potentials r0 = 1.25 fm, a0 = 0.60 fm and Δ = 0.80.
Initial optical model analysis

As a starting point, we have used two potential sets suggested by previous optical
model studies (Kondo, Robson, and Smith 1983) of 16O + 20Ne elastic scattering.
They are listed as potentials 1 and 2 in Table 35.1. Potential 1 predicts 18+, 19- and
24+ resonances at Ec.m. = 24.5, 25.4 and 35.5 MeV, respectively, while potential 2
predicts 17-, 18+ and 23- resonances at these three energies. As can be seen,
potential 1 has a positive value of Vπ, whereas potential 2 has a negative value of Vπ.
Based upon potentials 1 and 2, two types of calculation were performed and
compared with data. Firstly, optical model calculations were carried out using the full
parity dependent potentials 1 and 2. These results were complemented by
calculations of the type reported Kondo et al. (1985) where the parity-independent
parts of potentials 1 and 2 of Table 35.1 (i.e. Vπ set to zero) were used, and the
calculated α - transfer amplitude at angle (π - θ ) was added coherently to the
calculated elastic scattering amplitude at angle θ. Such calculations allow
investigation of whether the parity dependent term in optical model calculations
might arise from a requirement to simulate the effects of α -transfer amplitudes.
Using both approaches to the calculation, predictions were obtained for the
20
Ne(160,16O)20Ne excitation function at θc.m. = 90° and θc.m. = 154°±2°, and com-
parison made with present measurements and with the data reported Schimizu et al.
(1982).
Optical model calculations of the elastic scattering excitation function for potentials 1
(dash curves) and 2 (solid curves) are compared with the measured excitation
functions at θc.m = 154°±2° (panel (a)) and θc.m = 90° (panel (b)) in Figure 34.4. The
short vertical arrows in the figure represent the energies of the relevant shape
resonances in each case. (In panels (a) and (b), the numbers above and below the
arrows are the spins of the resonating partial waves for potentials 1 and 2,
respectively.)
As can be seen, both potentials provide similar predictions of the excitation function
at θc.m = 154° ± 2°, with unresolved gross structures being predicted at the energies
of each "doublet" of shape resonances. The excitation function predictions at θc.m =
90° however, show significant sensitivity to the optical potential used, with a
preference for potential 2.
Figure 35.4. Excitation functions at θc.m = 154°± 2° (left-had side) and 90° (right-hand side) for 16O +
20
Ne elastic scattering. Panels (a) and (b) show the theoretical results for potentials 1 and 2 compared
(Table 35.1) with the data. Panels (c) and (d) show the results of a coherent sum of elastic scattering
and ground state α - transfer amplitudes calculated using the parity-independent parts of potential 2.
Panels (e) and (f) show the results for potential 3. The short vertical arrows indicate the positions of
shape resonances and the respective J = L values for the resonating partial waves.
Transfer reaction contributions

Calculations, which explicitly take into account the ground-state α - transfer
amplitudes in the prediction of the elastic scattering excitation function were also
performed, and are compared with the measured excitation functions at θc.m =
154°±2° (panel (c)) and θc.m = 90° (panel (d)) in Figure 35.4. In these calculations,
we used the parity-independent part of potential 2 of Table 35.1 (i.e. Vπ set to zero).
It should be noted that equivalent calculations using potential 1 were reported
(Kondo et al. 1985) and were rejected as they were unable to describe the θc.m =
154°± 2° excitation function satisfactorily.
Figure 35.4 shows that the transfer contribution model, which predicts the magnitude
and general structure of the data at θc.m =154°±2° (panel (c)), also provides a good
description of the data at θc.m = 90° (panel (d)). This strengthens the conclusions of
the earlier study (Kondo et al. 1985), which attributed the parity dependence used in
the optical model calculations to a requirement to simulate the effects of α - transfer
amplitudes.
Re-examination of the optical model potential
As a final step, we have decided to depart from the restrictions of using the
potentials 1 and 2 of Kondo, Robson, and Smith (1983) and search for a potential
that could give the best fit to our experimental data. This has resulted in potential 3 in
Table 35.1. The real component V ( Ec.m. , L) of this potential is shown in Figure 35.5.
Figure 35.5. The depth of the real component of the potential 3, which gives the best fit to our
excitation function (see panel (f) in Figure 35.4). The graph shows that the potential is deeper for even
L – values and shallower for odd L – values.
The corresponding fit to the θc.m =90° data is shown in panel (f) of Figure 35.4. As
can be seen, potential 3 produces excellent fit to our data.
The calculations for θc.m =154°± 2° are displayed in panel (e). The fit is less
satisfactory but the calculations reproduce the positions of maxima and minima in
the excitation function sufficiently well. This potential gives spins resonances, 16+,
18+, and 20+ at Ec.m. = 22.5, 25.4, and 28.9 MeV.

We have measured excitation function for 16O + 20Ne elastic scattering at θc.m = 90°
over the energy range 21.5 MeV< Ec.m. < 31.2 MeV. We have then carried out an
extensive theoretical analysis of our data and the data of Schimizu et al. (1982) for
θc.m = 154° ± 2° using an extended optical model with the aim of placing additional
constraints on spin assignments to resonant states, and hence constraints on the
parity dependent part of the 16O + 20Ne potential.
We have found that one of the previously obtained potential (potential 2, Table 35.1)
is favoured by the available new data. However, we have also found a new potential
(potential 3, Table 35.1) which gives excellent fits to our θc.m = 90° excitation function
while maintaining a satisfactory description of the excitation function at
θc.m.=154°±2°. Our study suggests even spin shape resonances J = 16+, 18+, and 20+
at energies Ec.m. ≈ 22.5, 25.4 and 28.5 MeV, respectively.
The potential 3, which gives the best fit to our excitation function, and the favoured
potential 2, both have a negative parity-dependent component, which means that the
real potential depth is deeper for even than for odd partial waves (see Figure 35.5).
We have also carried out an analysis without the parity dependent component but
including explicitly the amplitudes for α – transfer reaction. These calculations,
though less satisfactory than the calculations using potential 3, give sufficiently good
representation of the shape and magnitude of the excitation function. Thus, as
suggested by Kondo et al. (1985), the parity dependent interaction appears to
simulate the effects of α -transfer amplitudes.
More generally, Baye, Deenen, and Salomon (1977), interpret parity dependences of
ion-ion potentials as a consequence of the Pauli principle associated with an
exchange of nucleons between two nuclei. They also suggest that the strongest
parity-dependence should be for nuclei with similar masses, which applies well to the
16
O+20Ne system. Thus, as also mentioned earlier in Chapter 24 for light projectiles,
parity dependent interaction appears to simulate coupling to non-elastic channels. To
include such coupling explicitly in the calculations would mean to study a relatively
large number of channels available in heavy-ion interactions.
References
Baye, D., Deenen, J. and Salmon, Y. 1977, Nucl. Phys. A289:511.
Chatwin, R. A., Eck, J. S., Robson D. and Richter, A. 1970, Phys. Rev. C1:795.
Kondo, Y., Robson, B. A. and Smith, R. 1983,Nucl. Phys. A410:289.
Kondo, Y., Robson, B. A., Smith, R. and Wolter, H. H. 1985, Phys. Lett. B162: 39.
Schimizu, J., Yokota, W., Nakagawa, T., Fukuchi, Y., Yamaguchi, H., Sato, M.,
Hanashima, S., Nagashima, Y., Furuno, K., Katori, K. and Kubono, S. 1982,
Phys. Lett. B112:323, and private communication.
Appendix A:
Semi-classical descriptions of polarization in stripping
reactions
This summary presents some early descriptions of nucleon polarization in stripping

r r
reactions. In this discussion, I use the product kin × kout to define the positive direction
of the vector polarization. In the early publications, the positive direction was defined
r r r r r r
using either kin × kout or kout × kin . Thus for instance, Newns (1953) used kout × kin but in
r r
the next related paper, Newns and Rafai (1958) used kin × kout . In the same year,
r r
Hansel and Parkinson (1958) still used kout × kin . Even later, Sitenko (1959) also used
r r
kout × kin . Thus, care must be exercised when reading the early publications.
The left and right directions may be also ambiguous. For instance, Wolfenstein
(1956) defines them by looking upstream.
To adhere to the uniform description, I have corrected the signs in the published
formulae and in the quoted polarization values whenever necessary. The discussion
is for the A(d,p)B reactions but the same formulae apply also to the A(d,n)B
reactions.
A simple, classical interpretation of the nucleon polarization produced in the
deuteron stripping reaction was proposed by Newns (1953). The mechanism is
illustrated in Figure A.1.
Figure A.1. Classical model (Newns 1953) of nucleon polarization in stripping reactions. This early
model assumes that the nucleus is transparent to deuterons but not to protons.
If we assume that the selection rules allow for only j = jn = ln + 1 / 2 transfer then the
r r
orbital momentum ln and the spin sn of the captured neutron must be parallel. For
neutrons captured in the area (i) the direction of the spin is into the plane and in the
area (ii) out of the plane. As the spins of protons and neutrons in deuterons are in
r
the same direction, the direction of the spin s p of protons emitted in the area (i) are
in the into-the-plane direction and those emitted in the area (ii) are in the out-of-the-
plane direction.
If the nucleus is opaque to protons, but transparent to deuterons, then protons
emerging from the area (ii) will be absorbed more readily than protons emerging
from the area (i). The net spin direction of the outgoing protons
r r
will be in the into-
the-plane direction, which is in the opposite direction to kd × k p . Thus if we assume
that the stripping reaction leads to a state in the final nucleus, which corresponds to
the capture of the neutron with j = l + 1 / 2 the resulting polarization of the outgoing
protons will be negative. Likewise, if j = l − 1 / 2 then the polarization of protons will
be positive.
Using this classical model for stripping reactions, Newns (1953) gives the following
formula for the polarization:
2
⎛ ⎞
⎜ ⎟
(l − m)! ⎜ 2 π
m
⎟
∑ m
(l + m)! ⎡ ⎜ ⎤ ⎡ ⎤⎟
⎜ Γ ⎢ (l − m ) + 1⎥ Γ ⎢ (1 − l − m )⎥ ⎟
m≥0 1 1
P=± ⎝ ⎣2 ⎦ ⎣2 ⎦⎠
2
⎡ ⎤
(l − m)! ⎢ 2m π ⎥
3( j + 1 / 2) ∑ ⎢ ⎥
m ≥ 0 (l + m)! ⎢ ⎡ 1 ⎤ ⎡1 ⎤⎥
Γ (l − m ) + 1⎥ Γ ⎢ (1 − l − m )⎥
⎢⎣ ⎢⎣ 2 ⎦ ⎣2 ⎦ ⎥⎦
The positive sing is for j = l − 1 / 2 and the negative sign for j = l + 1 / 2 .

It should be noted that in this classical description, the polarization of the outgoing
nucleons does not depend on the target nucleus but only on the transferred value
of the total angular momentum.
Using this formula, the following values of the polarization can be calculated
(Newns 1953):
Table A.1
The polarization of outgoing protons from the reaction A(d,p)B calculated the semi-classical
description proposed by Newns (1953)
l 0 1 2 3
j = l + 1/ 2 0 -16.67 -13.33 -18.75
j = l − 1/ 2 0 +33.33 +20.00 +25.00
The classical model of Newns (1953) is useful because it gives a simple

explanation of the mechanism of the nucleon polarization in stripping reactions.
However, the weakness of this model is that it neglects the interaction of deuterons
with the target nucleus.
It has been first pointed out by Tobocman (1956) that if one reverses the roles of
the outgoing proton and incoming deuteron, i.e. if one assumes that the nucleus is
opaque to deuterons but transparent to the outgoing protons then the sign of the
polarization will be reversed: it will be positive for j = l + 1 / 2 and negative for
j = l − 1/ 2 .
Indeed, for the strongly interacting deuterons we should reverse the shaded areas
in Figure A.1. In this reversed representation, the region (ii) is darker and region (i)
lighter. For protons, the whole nucleus will be transparent. In this new classical
description, more deuterons will be absorbed in the region (ii) than in (i).
Consequently, more protons will be coming from the region (ii) than from (i) and if
we again assume that j = jn = ln + 1 / 2 we shall find that the net spin of the outgoing
r r
protons will be in the out-of-the-plane direction, i.e. in the direction kd × k p . The
polarization of the outgoing protons will be positive for j = l + 1 / 2 and negative for
j = l − 1 / 2 . The summary of the two models is presented in Table A.2.
Table A.2
Summary of the classical models of the nucleon polarization produced in stripping reaction of
deuterons
Newns (1953) Tobocman (1956)

Crystal ball Transparent to Transparent to protons
deuterons Cloudy for deuterons
Cloudy for protons
Deuterons Pass easily through the Start being absorbed in
ball the area (II)
Stripped protons Emitted uniformly Emitted preferentially in
throughout the volume the area (II)
of the ball
Observed protons Mainly emitted from the Mainly emitted from the
area (I) area (II)
Proton spins for Mainly in the opposite
r r Mainly in the
r direction
r
j = l + 1/ 2 direction to kd × k p of kd × k p
Proton polarization at ± for j = l m 1 / 2 ± for j = l ± 1 / 2

forward angles
In real life, there will be a contribution from the interactions of both, protons and
deuteron, with the target nucleus. However, Satchler (1960) pointed out that for a
given depth of the nuclear potential the interaction of deuterons is about twice as
effective in producing the resulting polarization as the interaction of the outgoing
protons. He has also shown that only if V p≈ 2Vd and k p = kd that the two competing
effects cancel and the polarization is zero. This is hardly ever the case. We know
now that the depths of the potentials are in the opposite direction, i.e. that
V d≈ 2V p so if we want to apply the classical interpretation of the polarization
mechanism we should expect that the observed signs of the polarization will be
opposite to the signs predicted by the original model of Newns (1953).
It should be noted that the polarization depends on the reaction angle so that even
for the same value of the transferred total angular momentum it can change sign
depending on the reaction angle. The classical picture applies to reactions at small
angles. In the more exact quantum mechanical description, the association of the
sign of the polarization with the transferred total angular momentum applies to the
region of the first stripping maximum, which depending on the transferred orbital
angular momentum l will have a peak at different angles but always in the forward
direction. The polarization will change the sign when the stripping amplitude
changes its sign.
Sitenko (1959) refers to the model of Newns but relates the polarization to the
wave functions of deuterons and protons. He gives the following formula:
⎛ mIm 2 ⎞
2 ⎜∑ l ⎟
P=± ⎜ m ⎟ for jn = l m 1 / 2
3(2 jn + 1) ⎜ ∑ I m 2 ⎟
⎜ l ⎟
⎝ m ⎠
where jn is the total spin of the transferred proton. The quantity I lm is given by
r t (k r ) r r
I lm = ∫ ϕ k*p (r ) l n Ylm* (θ ,φ )ϕ k d (r )dr
tl ( k n R )
where
π
tl ( x ) = K l +1 / 2 ( x)
2x
is the modified Bessel function of the second kind (Abramowitz and Stegun 1964).31
If the target nucleus has spin zero, the polarization of the outgoing protons is
directly related to the spin of the relevant energy level in the residual nucleus
J = j . As can be seen, there are some obvious similarities between the formulae of
Newns and Sitenko.
The predicted maximum of the absolute value of the nucleon polarization produced
in the deuteron stripping reactions is 1/3 (Horowitz and Messiah 1953; Newns
1953). This can be easily seen from the following formulae (Huby et al. 1958;
Satchler 1958; see also Glendenning 1963):
1 ml
P= for j = l + 1 / 2
3 l +1
1 ml
P=− for j = l − 1 / 2
3 l
If ml assumes its maximum value of l then
31
Known also as the spherical MacDonald function, which is the name used by Sitenko (1959).
1 l
P= for j = l + 1 / 2
3 l +1
1
P = − for j = l − 1 / 2
3
Even in the early studies of nucleon polarization it became clear that the observed
values were higher than those predicted by the first theoretical calculations (Allas
and Schull 1959. 1962; Bokhari et al. 1958; Hillman 1956; Isoya, Marrone,
Michaletti, and Reber 1962).
If l = 0 then ml = 0 and the polarization P = 0 . For transitions involving l = 0 , P ≠ 0
if spin-orbit interaction is introduced.
Quantum-mechanically, the ml values can be related to the amplitudes Blml for
stripping reactions32 (Horowitz and Messiah 1953):
∑m B l l
ml 2
ml =
ml
∑B
ml
l
ml 2
where
r r 1 r r r r r r r r
1/ 2 ∫
Blml ≡ Blml (kd , k p ) = ψ *p (k p , rp )φlm* (rn )Vnp (r )φd (r )ψ d (kd , R)drn drp
i (2l + 1)
r r r r r r
r = rn − rp , 2 R = rn + rp .
In the absence of the spin-orbit interaction, Blml = Bl− ml and P = 0 if l = 0 . The same
applies to the formula of Sitenko (1959) because I ml = I l− m .
If the spin-orbit interaction is included then the expression for Blml becomes more
complex and there is no simple formula for the predicted polarization.
References
Abramowitz M. and Stegun, I. A. 1964, Handbook of Mathematical Functions, United States
Department of Commerce, National Bureau of Standards, Washington, DC.Allas, R. G.
and Schull, F. B. 1959, Phys. Rev. 116:996.
Allas, R. G. and Schull, F. B. 1962, Phys. Rev. 125:941.
Bokhari, M. S. Cookson, J. A., Hird, B. and Weesakul, B. 1958, Proc. Phys. Soc. 72:88.
Glendenning, N. K. 1963, Ann. Rev. Nucl. Sci. 13:191.
Hillman, P. 1956, Phys. Rev. 104:176.
Horowitz, J. and Messiah, A. M. L. 1953, J. Phys. Radium 14:731.
Huby, R., Rafai, M. Y. and Satchler, G. R. 1958, Nucl. Phys. 9:94.
Isoya, A., Micheletti, M. J., Reber, L. 1962, Phys. Rev. 128:806.
Newns, H. C. 1953, Proc. Phys. Soc. 66:477.
Satchler, G. R. 1958, Nucl. Phys. 8:543.
Sitenko, A. G. 1959, Успехи Физ. Наук 67:377.
32
Blml is the amplitude for the absorption of a neutron with quantum numbers (l , ml ) .
Appendix B
The diffraction theory
The diffraction theory of nuclear scattering was developed by Blair (Blair 1959, 1961)
assuming a sharp cutoff radius, and extended by Blair, Sharp, and Wilets (1962), by
introducing a diffuse surface for target nuclei.
The elastic scattering

The amplitude for the elastic scattering of spinless particles by spherical nuclei can
be expressed in terms of partial waves:33
∑ (2l + 1)(1 − η )P (cosθ )

1
f (θ ) = l l
2iki l
where ki is the wave number for the incoming particle, Pl (cos θ ) is the Legendre
polynomial, and ηl is the amplitude of the scattered l th partial wave.
If the nucleus is represented as a black disk with a sharp cutoff radius then
ηl = 0 for l < L,
ηl = 1 for l > L,
ηl = 1/ 2 for l = L,
where L is the orbital angular momentum or the partial wave determined by the
nuclear radius R0.
In the diffraction theory, the amplitude f (θ ) is calculated in a similar way as in the
optics using the Fraunhofer approximation:
ik rr
f (θ ) ≅ i ∫∫ dA exp(iki r )
2π
The integration is over the area of the projection of the sphere.
This leads to the following expression for the elastic scattering differential cross
section:
⎛ dσ ⎞
2
⎜ ( )
2 2 ⎡ J ( x) ⎤
⎟ = ki R0 ⎢ 1 ⎥
⎝ dΩ ⎠el ⎣ x ⎦
where k i≡ 1 / D i is the wave number for the incident particles, R0 the nuclear radius,
and J1(x) is the cylindrical Bessel function of the first order.
r r r r
The variable x ≡ qR 0 , where q ≡ ki − k f with ki and k f being the wave vectors for the
incoming and outgoing particles, respectively (see Figure B.1).
33
The Coulomb scattering is not included.
r r r
Figure B.1. The diagram showing the relation between ki , k f , q and the reaction angle θ.
From the well-known relation for the triangle shown in Figure B.1 we find that
q 2 = ki2 + k 2f − 2ki k f cosθ
For the elastic scattering ki = k f and since
sin( θ / 2 ) =
1
(1 − cosθ )
2
we find that x = 2ki R0 sin(θ / 2) .
The universal function [J1(x)/x]2 is presented in Figure B.2 for x ≥ 4 , i.e. starting from
the second maximum. To show the structure of this function, the first large maximum
at x = 0 is not displayed.
[ ]2
Figure B.2. The universal function J1( x ) / x calculated using the tabulated values (West 1970) and
the approximate formula (Bronstein and Semedleiev 1957) for the Bessel function J 1( x ). To show
the structure of the universal function, the first large maximum at x = 0 is not displayed.
For sufficiently large values of x, the cylindrical Bessel function of the order of n can
be calculated using the following approximate formula (Bronstein and Semedleiev
1957):
2 ⎛ nπ π ⎞
Jn( x ) ≈ cos⎜ x − − ⎟
πx ⎝ 2 4⎠
It can be seen from Figure B.2 that for x > 6 the approximate formula for J1(x) gives a
good representation of the universal function [J1( x ) / x ] .
2
The inelastic scattering

For the inelastic scattering from deformed nuclei, integration is also carried out over
the projection area of the scattering nucleus. The integration can be reduced to
integration over the projection of a sphere and integration over a projection of the
deformed part. The whole contribution to the inelastic scattering comes from the
integration from the later area.
The black disc formula for the inelastic scattering corresponding to the l = 2
(quadrupole) excitation has the following form (Blair 1959,1961):
dσ ⎡ ⎤⎡ 1 2
(l = 2, I 0 → I ) = (ki R0 )2 ⎢ 1 2l + 1 ∑ IM δ l ,m I 0 M 0 ⎤
⎥ ⎢ J 0 (qR0 ) + J 2 (qR0 )⎥
3 2
2
dΩ ⎣ 2 I 0 + 1 4π M 0 ,M ,m ⎦⎣ 4 4 ⎦
In this formula, I 0 is the spin of the ground state, I spin of the excited state, δ l ,m the
deformation distance, and J 0 and J 2 are the cylindrical Bessel functions.
The deformation distance is defined by the description of nuclear radius R in the
following way:
R(θ ′, φ ′) = R0 + ∑ δ lmYlm* (θ ′, φ ′)
lm
where Ylm are the spherical harmonic functions, and θ ′ and φ ′ are the body fixed
coordinates.
The cross section formula can be rewritten as:
dσ ⎡ ⎤⎡ 1 2 ⎤
(l = 2, I 0 → I ) = (ki R0 )2 ⎢ 1 5 ∑ ⎥ ⎢ J 0 (qR0 ) + J 2 (qR0 )⎥
3 2
2
IM δ 2 I 0 M 0
dΩ ⎣ 2 I 0 + 1 4π M 0 ,M ⎦⎣ 4 4 ⎦
or in even a simpler form as

dσ
(l = 2, I 0 → I ) = (ki R0 )2 ΜF (x )
dΩ
where
1 5 1
∑
2
Μ≡ IM δ 2 I 0 M 0 ≡ Μ′
2 I 0 + 1 4π M 0 ,M 2I0 + 1
and
⎡1 ⎤
F ( x ) ≡ ⎢ J 02 (x ) + J 22 ( x )⎥
3
⎣4 4 ⎦
As in the case of the elastic scattering, which was described by the
[J1( x ) / x]2 function, F (x ) is also a model-independent universal function. The function
is presented in Figure B.3 for 0 ≤ x ≤ 12 . I have calculated it using the J 0 (x ) and
J 2 ( x ) values tabulated by Abramowitz and Stegun (1964).
Figure B.3. The universal shape function F ( x ) used in describing inelastic scattering.
Procedure in theoretical analysis

Theoretical analysis of experimental data for either elastic or inelastic scattering is
simple. The first step consists in determining the radius R0. This is done by using the
formula:
x
R0 = max
q~
where xmax is the position of the second maximum of the universal function
[J1( x ) / x]2 or F (x ) , q~ is the q - value corresponding to the position θ max of the
relevant maximum of the experimental angular distributions:
(
q~ = ki2 + k 2f − 2ki k f cosθ max )
1/ 2
For instance, for the inelastic scattering, the position of the second maximum of the
universal function F (x ) is at xmax =3.25.
Using thus determined R0 one can then express the function [J1( x ) / x ] or F ( x ) as a
2
function of θ and compare it with the experimental angular distribution.
Model-dependent formulae
The matrix element Μ′ is related to nuclear structure. For extreme collective
models, Μ ′ can be used to calculate model-dependent parameters, δ 2 or
(hω2 / 2C2 ) using the following relations (Blair 1961):
Even-even-mass nucleus
(a) Rotational model
δ 22
Μ′ =
4π
where δ 2 is the deformation distance parameter.
(b) Vibrational model (i.e. surface vibrations with phonon energy given by hω2 )
5 ⎛ hω2 ⎞ 2
Μ′ = ⎜⎜ ⎟⎟ Rc
4π ⎝ 2C2 ⎠
where C2 is the surface tension parameter and Rc is a nuclear matter radius,
which is close to the charge radius. It is reasonable to assume that Rc = 1.2 A1 / 3 ,
where A is the atomic mass of the target nucleus.
Odd-mass nucleus
(a) Rotational model
δ 22
Μ′ = (I 0 , l , K ,0 | I , K )
4π
where (I 0 ,l , K ,0 | I , K ) is the Clebsch-Gordan coefficient given by (Condon and
Shortley 1935):
(I 0 , l , K ,0 | I , K ) = K ⎢ 3(I 0 − K + 1)(I 0 + K + 1) ⎥
1/ 2
⎡ ⎤
⎣ I 0 (2 I 0 + 1)(I 0 + 1)(I 0 + 2 )⎦
For instance, if we assume the strong coupling model for 27Al, then for the ground
state band K = 5/2 and (I 0 , l , K ,0 | I , K ) = 0.69 .
(b) Vibrational model
2I + 1 1 ⎛ hω 2 ⎞ 2
Μ′ = ⎜⎜ ⎟⎟ Rc
2 I 0 + 1 4π ⎝ 2C2 ⎠
We can use the above formulae if we really know that these models apply or if we
wish to have a means of parameterising the magnitude of the nuclear matrix
elements, even though the model does not apply. However, when we are just
starting to analyse the data and do not know whether any of these extreme models
can be applied (as in the case discussed in Chapter 3) then it is best to list just the
factors Μ or Μ′.
Extended diffraction theory

One of the biggest defects of the Fraunhofer version (i.e. the black disk with sharp
cut-off radius) of the diffraction model is that the nuclear surface is taken to be sharp.
An extremely simple way of making this region ‘grey’ is to multiply the sharp-edge
cross sections by some factor f (kiθ ) . A typical form factor is the Gaussian,
f (kiθ ) = exp[− (αkiθ )] , where α is some distance characteristic of the surface.
2
The effect of introducing a grey or diffuse region at the nuclear surface is that the
amplitudes of the calculated angular distributions are progressively pushed lower
with the increasing reaction angle θ. This effect has been also simulated by
introducing an obliquity factor cos2 (θ / 2) . However, a far better way of accounting for
the diffuse region of nuclear surface is to parameterise the amplitude ηl of the
scattered partial wave l. Such a smooth transition from a complete to no absorption
has been introduced by Blair, Sharp, and Wilets (1962) by assuming that ηl has the
following form:
ηl = {1 + exp[(L − l )] / Δ}
They introduced two parameters: the cut-off angular momentum L = ki R0 − 1 / 2 and
the ratio Δ / L , which describes the width of the transition region. The location of the
maxima and minima depend on L and relative heights only on Δ / L . The difference
between the sharp cut-off radius and smooth cut-off radius calculations is illustrated
in Chapter 3 for scattering of deuterons from 27Al and 28Si nuclei.
References
Abramowitz, M. and Stegun, I. A. 1964, Handbook of Mathematical Functions, National
Bureau of Standards, Applied Mathematics Series 55, United States Department
of Commerce, Washigton, DC.
Blair, J. S. 1959, Phys. Rev. 115:928.
Blair, J. S. 1961, private communication.
Blair, J. S., Sharp, D., and Wilets, L. 1962, Phys. Rev. 125:1625.
Bronstein, I. N. and Semendleiev, 1957, Спровочник по Математике (Hanbook
of Mathematics), Moskva.
Condon, E. U. and Shortley, S. H. 1935, The Theory of Atomic Spectra, Cambridge
University Press, Cambridge, UK.
West, R. C. (ed.) 1970, CRC Handbook of Tables for Mathematics, 4th edn, Chemical
Rubber Company, Cleveland, Ohio,
Appendix C
The plane-wave theory of inelastic scattering
In its general form, the transition amplitude for the inelastic scattering from state i to
state f is given by (Rost and Austern 1960):
T fi = ϕ f χ (f − ) V ϕi χ (f + )
where, ϕi and ϕ f are the intrinsic wave functions for state i and f, χ i and χ f the wave
function describing relative motion of the projectile in the incident and outgoing
channel, and V is the interaction potential that causes the transition from state i to f
without influencing the wave functions χ i and χ f .
The differential cross section is then given by:
dσ (i → f ) ⎛ μ ⎞ k f
2
∑
2
=⎜ 2 ⎟
T
dΩ ⎝ 2πh ⎠ ki av fi
where μ is the reduced mass of the projectile and ki and kf are the wave numbers in
the incident and outgoing channels.
The wave functions χ i and χ f are solutions of the Schrödinger equation:
⎡ ⎛ h2 ⎞ 2 ⎤
⎢− ⎜⎜ ⎟⎟Δ + U ⎥ χ i(,±f ) = Ei , f χ i(,±f )
⎣ ⎝ 2μ ⎠ ⎦
where U is the optical model potential.
Butler (1950, 1951, 1952, 1957; Butler and Austern 1954; Butler, Austern, and
Pearson 1958) was the first to introduce the concept of direct nuclear interactions. In
his simple model, he ignored the nuclear interaction U in both the incident and
outgoing channels and thus represented the wave functions χ i and χ f by the
undisturbed plane waves. Under this assumption, the transition matrix can be
written as:
r r rr
T fi = ϕ f exp(−ik f rf ) V ϕi exp(iki ri )
r r
where, ki and k f are the initial and final momenta.
The differential cross section for the inelastic scattering takes then a simple form:
dσ
dΩ
[ ]
= S q 2 + K 2 W 2 [ jl (qr ), hl (iKr )]r = R0
−1
r r
where S is an arbitrary normalisation factor, q = ki − k f , K = κi + κ f ,
W [ jl ( qr ),hl ( iKr )] is the Wronskian, jl and hl are the spherical Bessel and Hankel
functions, respectively, and R0 is the nuclear radius.
The momentum transfer q is given by:
(
q = ki + k f − 2ki k f cosθ
2 2
)1/ 2
where θ is the scattering angle.

The quantity K = κ i + κ f is related to the projectile binding energies in the target
nucleus before and after the scattering, i.e. when the target nucleus is in the ground
state and excited state, respectively.
2 μBi
κi =
h
2 μB f
κf =
h
where Bi and B f the binding energies before and after the scattering.
The Wronskian is given by the following expression:

jl hl
⎡ dh dj ⎤
W ( jl ,hl ) ≡ djl dhl = ⎢ jl l − hl l ⎥
⎣ dr dr ⎦ r = R0
dr dr r = R0
It can be shown that

W [ jl ( x), hi (iy )] = Gl ( y ) jl ( x) + xjl −1 ( x)
where x = qR0 , y = (κ i + κ f )R0 and
hl −1 (iy )
Gl ( y ) = −iy
hl (iy )
For l = 2 we have
W [ j2 ( x), h2 (iy )] = G2 ( y ) j2 ( x) + xj1 ( x)
and
y 2 (1 + y )
G2 ( y ) =
3 + 3y + y2
As with the diffraction theory (see the Appendix B), the fitting procedure consists in
aligning the first maximum of the angle-dependant function [in this case the function
W 2 [ j2 ( x), h2 (iy )] ≡ W22 ( x, y ) ] with the first maximum of the experimental distribution.
However, the problem here is that unlike the universal F ( x) function in the diffraction
theory, the function W22 ( x, y ) depends on two variables, each of which depends on
the radius R0 . Thus, the maximum of W22 ( x, y ) will depend not only on the value of
x = qR0 but also on y = (κ i + κ f )R0 .
Fortunately, the dependence on y is weak. I have calculated the W22 ( x, y ) function for
various values of y using the tabulated values for the spherical Bessel functions
(Abramowitz and Stegun 1964). Figure C.1 shows the W22 ( x, y ) functions for y = 14,
15, and 16, and for x = 0 - 10. It can be seen that there is virtually no change in the
location of the first maximum.
I have also calculated the positions of the first maximum for a wide range of y = 4 -
30. Results are shown in Figure C.2. It can be seen that even for such a wide range
of y values the position of the first maximum changes only by a few per cent.
Figure C.1. Squared Wronskian of the second order plotted for three values of y. The functions were
constructed using the tabulated spherical Bessel functions (Abramowitz and Stegun 1965).
Figure C.2. The position xmax of the first maximum of the function W
2
[ j2 ( x), h2 (iy)] ≡ W22 ( x, y) as
the function of y .
The traditional way of fitting experimental data with the plane wave theory is to use
the tabulated cross sections (Lubitz 1957). First one has to calculate the y/x ratio,
which does not depend on R0:
y κi + κ f
= ~
x q
where, as for the diffraction model, q~ is the q - value corresponding to the position
θ max of the first maximum of the experimental angular distributions.
Next, from a plot of the dependence of y/x on x (Lubitz 1957) for a given l one
determines the value of y, which then one can use to find the corresponding
tabulated σ TAB
l
( x, y ) cross sections. Using the tabulated σ TAB
l
( x, y ) cross sections one
finds the position xmax of the first maximum of function W22 ( x, y ) , and use it to
calculate R0:
x
R0 = max
q~
Finally, using the determined R0 one can express W22 ( x, y ) as a function of θ and
compare it with the experimental angular distribution.
Alternatively, one can assume that the radius R0 is given approximately by
R0 ≈ 1.2 A1 / 3 + R p
where A is the mass number of the target nucleus and Rp the radius of the projectile.
Using thus calculated R0 one can calculate y and W22 ( x, y ) . As the position of the first
maximum of W22 ( x, y ) depends weakly on y, one can locate the xmax value and use it
to recalculate the value of R0.
In both of these options, the procedure can be repeated for different values of y until
satisfactory results are obtained.
References
Abramowitz, M. and Stegun, I. A. 1964, Handbook of Mathematical Functions, National
Bureau of Standards, Applied Mathematics Series 55, United States Department
of Commerce, Washigton, DC.
Butler, S. T. 1951, Proc. Roy. Soc. (London) A208:559.
Butler, S. T. and Austern, N. 1954, Phys. Rev. 93:355
Butler, S. T., Austern, N., and Pearson, C. 1958, Phys. Rev. 112:1227.
Lubitz, C. R. 1957, Numerical Table of Butler-Born Approximation Stripping Cross
Sections, Randall Laboratory of Physics, University of Michigan, Ann Arbor.
Rost, E. and Austern, N. 1960, Phys. Rev. 120:1375.
Appendix D
The strong coupling theory
The concept of the strong-coupling theory (as used in the analysis discussed in
Chapter 3) has been described in the following publications: Buck (1963), Buck,
Stamp, and Hodgson (1963), and Chase, Wilets, and Edmonds (1958).
We start with the Schrödinger equation:
[H n (ξ ) + T + V (rr, ξ )]Ψ (rr, ξ ) = EΨ (rr, ξ )
r
where r are the coordinates of the incident particles, ξ – internal coordinates of the
target nucleus, H n (ξ ) – the target nuclear Hamiltonian, T – the kinetic energy
r
operator for the relative motion, V (r , ξ ) – the interaction energy between particle and
r
nucleus, Ψ (r , ξ ) – the complete wave function for the system, and E – the total
energy. The system of coordinates is in the centre-of-mass frame of reference, the
mass is represented by reduced mass, and the relative energies and momenta are
used.
r
To solve this equation we expand Ψ (r , ξ ) in terms of eigenstates of the total angular
momentum J:
r r
Ψ (r , ξ ) = ∑ aJMψ JM (r , ξ )
JM
Here we have to introduce notations for the angular momenta:

I – target spin
l – the relative orbital angular momentum of the partial wave of the incident particles
J – the total angular momentum ( Jˆ = lˆ + Iˆ )
All these quantities are for the entrance channel. For the exit channels we have I ′
and l′ adding also to J.
r
The function ψ JM (r , ξ ) can be expressed as a superposition of elastic and inelastic
scattering components:
r 1 J r 1 r
ψ JM (r , ξ ) = f Il (r )φIlJM (r , ξ ) + ∑ f IJ'l ' (r )φI 'l ' (r , ξ )
r r I 'l '
where f IlJ (r ) are the radial wave functions for the system with the target in spin state
I and relative target-particle angular momentum l. The sum is over all I ′ values but
excluding the ground state I.
By inserting this function in the Schrödinger equation we then get the following set of
differential equations for the radial wave functions:
⎧ h 2 ⎛ d 2 l (l + 1) ⎞ ⎫
⎨− ⎜⎜ 2 − ⎟⎟ + ε I − E ⎬ f IlJ (r ) + ∑VIlJ: I ′l ′ (r ) f I J′l ′ (r ) = 0
⎩ 2μ ⎝ dr
2
r ⎠ ⎭ I ′l ′
⎧ h 2 ⎛ d 2 l ′(l ′ + 1) ⎞ ⎫
⎨− ⎜⎜ 2 − ⎟⎟ + ε I ′ − E ⎬ f I J′l ′ (r ) + VI J′l ′:Il ( r ) f IlJ (r ) + ∑VI J′l ′:I ′′l ′′ (r ) f I J′′l ′′ ( r ) = 0
⎩ 2 μ ⎝ dr
2
r ⎠ ⎭ I ′′l ′′
where μ is the reduced mass, εI is the excitation energy of the target state with spin
Ι, and VIlJ:I ′l ′ ( r ) are the coupling matrix elements. The sum in the second set of
equations is for I ′′ ≠ I and l′′ ≠ l .
The coupling matrix elements are defined as
r r r
VIlJ:I ′l′ (r ) = φIlJM ( r , ξ ) V ( r , ξ ) φIJM
′l ′ ( r , ξ )
As mentioned in Chapter 3, the computer code used in the analysis of our

experimental results was restricted to scattering from even-even nuclei and for only
0 + → 2 + transitions. This assumption simplifies significantly the calculations. In this
case J = l for the ground state and I ′ = I ′′ = 2. The differential equations for the radial
functions take then a simpler form:
⎧ h 2 ⎛ d 2 J ( J + 1) ⎞ ⎫
⎨ ⎜⎜ 2 − ⎟⎟ + E − V0JJ :0 J ( r ) ⎬ f 0JJ ( r ) − ∑ V0JJ :2l′ (r ) f 2Jl′ ( r ) = 0
⎩ 2 μ ⎝ dr
2
r ⎠ ⎭ l′
⎧ h 2 ⎛ d 2 l ′(l ′ + 1) ⎞ ⎫
⎨ ⎜⎜ 2 − ⎟⎟ + E ′ − V2Jl′:2l′ (r )⎬ f 2Jl′ (r ) − V2Jl′:0 J ( r ) f 0JJ (r ) − ∑ V2Jl′:2l′′ (r ) f 2Jl′′ (r ) = 0
⎩ 2μ ⎝ dr
2
r ⎠ ⎭ l ′′≠l ′
where E = E − ε 2 .
In principle, the equations are for an infinite number of partial waves. However, in
practice, only a limited number of partial waves contributes to the reaction and
consequently J values are terminated at a certain Jmax value, which is determined by
examining the solutions of the differential equations (see below).
The coupling matrix elements involve the use of Clebsch-Gordan coefficients, which
limit the values for l′ and l ′′ to three values for each J, namely J - 2, J, and J + 2.
With these restrictions, the lower set of equations is limited to three, making (with the
upper equation) the total of only four coupled equations for each value of J.
The solutions f 0JJ ( r ) , f 2JJ −2 ( r ) , f 2JJ (r ) , and f 2JJ + 2 ( r ) are then expressed in terms of
four scattering matrix elements α 0JJ , β 2JJ − 2 , β 2JJ , and β 2JJ + 2 determined by the
boundary conditions. The scattering matrix elements are directly related to the better
known S - matrix elements, which enter into the expressions of the elastic and
inelastic scattering amplitudes:
1
A0 (θ ) = f c (θ ) + ∑
2ik J
exp(2iω J )(2 J + 1)( S0JJ − 1) PJ (cosθ )
1 ⎛ k′ ⎞ ⎡ ( J ′ − M )!⎤ J ′J 2 M
1/ 2
A (θ ) = ⎜ ⎟
M
∑ exp[i(ω′ J′ + ω J )](2 J + 1) S2JJ ′ (−1) M ⎢ ⎥CM 0 M PJ ′ (cosθ )
⎣ ( J − M )! ⎦
2
ik ⎝ k ⎠ JJ ′
where
θ – the scattering angle
f c (θ ) – the Coulomb scattering amplitude
γ
f c (θ ) = − exp[− 2iγ log sin (θ / 2 )]
2k sin 2 (θ / 2)
μzZe 2
γ=
kh 2
with z and Z being the atomic numbers for the projectile and target nucleus
respectively
ω J = σ J −σ 0 ,
with σ J being the Coulomb phase shifts (Buck, Maddison, and Hodgson 1960)
J ′ = J − 2 , J, J − 2 ,
S0JJ and S 2JJ ′ are the S – matrix elements
As mentioned earlier, the S – matrix elements are related to the four scattering
matrix elements α 0JJ , β 2JJ − 2 , β 2JJ , and β 2JJ + 2 [which in turn are related to the four
solutions f 0JJ ( r ) , f 2JJ −2 ( r ) , f 2JJ ( r ) , and f 2JJ + 2 ( r ) ] of the coupled-channel equations
in the following way:
S 0JJ = 2α 0JJ + 1
S 2JJ − 2 = β 2JJ − 2
S 2JJ = β 2JJ
S 2JJ + 2 = β 2JJ + 2
C MJ ′J02M – the Clebsch-Gordan coefficients in Beidenharn (1952) notation.
PJ (cosθ ) and PJM (cos θ ) – Legendre and associated Legendre polynomials,

respectively.
The four differential equations for the radial wave functions are solved numerically
for various values of J until α 0JJ , β 2JJ − 2 , β 2JJ , and β 2JJ + 2 become negligible. This
determines the maximum value Jmax, which corresponds to the maximum number of
partial waves contributing to the scattering.
Using the theoretically calculated elastic and inelastic scattering amplitudes and the
S – matrix elements one can calculate the quantities, which can be compared with
the experimental results:
The differential cross section for elastic scattering:
⎛ dσ ⎞
⎟ = A0 (θ )
2
⎜
⎝ dΩ ⎠ el
The differential cross section for inelastic scattering:
⎛ dσ ⎞ +
⎟(0 → 2 ) = ∑ A2 (θ )
+ 2
⎜
M
⎝ dΩ ⎠ M
The total absorption cross section:
π
σ A=
k 2 ∑ (2 J + 1)⎛⎜⎝1 − S
J
J 2
0J
⎞⎟
⎠
The total elastic cross section for incident particles:
π
σ EL=
k 2 ∑ (2 J + 1) 1 − S
J
J 2
0J
The total inelastic cross section:

π k′
σ IN(0 + → 2 + ) =
k2 k
∑ (2 J + 1) S
JJ ′
J 2
2J′
The total cross section for particles:

2π
σ T=
k2
∑ (2 J + 1)(1 − Re S )
J
J
0J
In the above expressions J = 0, 1, 2, … Jmax and J ′ = J , J ± 2 .
References
Biedenharn, L. C. 1952, Tables of Racah Coefficients, Oak Ridge National Laboratory
Report, ORNL-1098.
Buck, B. 1963, Phys. Rev. 130:712.
Buck, B., Maddison, R. N., and Hodgson, P. E. 1960, Phil. Mag. 5:1181.
Buck, B., Stamp, A. P. and Hodgson, P. E. 1963, Phil. Mag. 8:1805.
Chase, D. M., Wilets, L. and Edmonds, A. R. 1958, Phys. Rev. 110:1080.
Appendix E
Theories of direct nuclear reactions
Introduction
Theories of direct nuclear reactions have been discussed in numerous publications.
Convenient sources of reference are two books: Glendening (2004) and Satchler
(1983). Presented here are just general concepts of these theories.
Consider a general reaction A(a,b)B, which is illustrated in Figure E.1 for the stripping
reaction.
Figure E.1. The stripping reaction A(a,b)B. The particle x is stripped from a and absorbed by A to form
the particle B.
Information about the reaction mechanism and nuclear structure is contained in the
transition matrix Tba , which in its general form can be written as
Tba = Ψβ V Ψα
where Ψα and Ψβ are the complete wave functions describing the incident and
outgoing channels, α(a,A) and β(b,B), respectively, and V is the interaction potential
responsible for the transition a → b + x .
The wave functions Ψα and Ψβ , describing the entire systems a + A and b + B
respectively are the solutions of the Schrödinger equations:
( H α + Tα + U α − E )ψ α = 0
( H β + Tβ + U β − E )ψ β = 0
where
H α= H a + H A
H β= H b + H B
H a and H b are the internal Hamiltonians for the incident and outgoing particles;
H A and H B are the Hamiltonians for the target and residual nuclei; Tα and Tβ are the
kinetic energy operators in channels α and β; U α and U β are the operators describing
the interaction of the incident particle a with the target nucleus A and of the outgoing
particle b with the residual nucleus B, respectively.
It is impossible to solve the Schrödinger equations for the entire systems a + A and b
+ B so we have to introduce simplifications, which lead to various versions of theories
of direct reactions.
Functions Ψα and Ψβ can be written as products of three functions
Ψα = ψ aψ Aψ aA
Ψβ = ψ bψ Bψ bB
where
ψ a , ψ b , ψ A , and ψ B are the internal wave functions for the particles a, b, A, and B.
These wave functions are solutions of the following Schrödinger equations:
( H a − ε a )ψ a
( H b − ε b )ψ b
( H A − ε A )ψ A
( H B − ε B )ψ B
It is usually assumed that Hk (k = a, b, A, or B) are shell model Hamiltonians of the form
A
H A = ∑ (Ti + U i ) + ∑V ij
i =1 i≠ j
where
h2 2
Ti = − ∇i
2mi
is the kinetic energy operator of the ith particle, U i ≡ U (ri ) a central potential acting on
r r
a single particle, and Vij ≡ V (ri − rj ) an interaction potential between two particles i and
j, known as the residual interaction.
The wave functions ψ aA and ψ bB describe events taking place in the systems a + A and
b + B but do not describe the intrinsic structure of particles a, b, A, and B.
It is convenient to write the transition matrix element in the form:
(−) r (+) r r r
Tαβ = ∫ ∫ψ aA ( ra ) ψ bψ B V ψ aψ A ψ bB (rb )dra drb
r r
where ra and rb are the relative coordinates of (a,A) and (b,B).
The nuclear matrix element ψ bψ B V ψ aψ A , also known as the form factor, represents
integration over all intrinsic coordinates. This form factor contains information about
nuclear structure.
The Plane Wave Born Approximation (PWBA)
The simplest treatment of direct nuclear reactions is to assume that there is no
interaction between a + A and b + B systems (except for a momentary interaction
changing a into b), i.e. that Uα = Uβ = 0 (see the Appendix C). The wave functions
ψ aA and ψ bB can then be written as plane waves
r r
ψ aA= exp(ika ⋅ ra )
r r
ψ bB= exp(ikb ⋅ rb )
r r
where ka and kb are the relative momenta of particles a and b.
The transition matrix has then the form:
r r r r
r r
TPWBA = ∫ e −ik b ⋅ra ψ bψ B V ψ aψ A eik b ⋅rb dra drb
If we assume that particle b is emitted from the same point where particle a is
r r r
absorbed, which makes ra ≈ rb = r , we have for the following simpler expression for
the transition matrix
r r r r
r r r r
TPWBA = ∫ e −ik b ⋅r ψ bψ B V ψ aψ A eik a ⋅r dr = ∫ eiq ⋅r ψ bψ B V ψ aψ A dr
r r r
where q = ka − kb is the momentum transfer, which depends on the reaction angle θ:
[
q = ka2 + kb2 − 2ka kb cosθ ]1/ 2
The form factor ψ bψ B V ψ aψ A may be written as
ψ bψ B V ψ aψ A = ∑ f l (r )Yl m (θ ′, φ ′)
l ,m
where l is the angular momentum transfer involved in the reaction, which is usually
represented by just one value. In this case,
TPWBA ∝ ∫ jl (qr ) f l ( r )r 2 dr
where jl (qR ) is the spherical Bessel function of the order of l.

If we assume that the nucleus is a black object with the radius R then the integration
can be carried out for r ≥ R . If we further assume that the reaction is confined to
nuclear surface, i.e. that r ≈ R and thus neglect integration over r > R then in its
simplest form
TPWBA ∝ jl (qR )
The differential cross sections, which are proportional to the square of the transition
matrix for the reaction A(a,b)B are in its simplest form proportional to the square of the
spherical Bessel function
dσ
∝ jl2 ( qR )
dΩ
This is a convenient formula, which could be used easily in the interpretation of
experimental data. By comparing the jl2 (qR ) function with experimental angular
distributions one can easily determine the l transfer and thus the orbital angular
momentum of nuclear state excited in a single particle stripping reaction or angular
momentum involved in inelastic scattering.
The radius R is a parameter that is obtained by matching the position of a dominant
maximum in the experimental angular distribution with the relevant maximum of the
jl2 (qR ) function (see the Appendices B and C) . This parameter may be interpreted as
a nuclear radius.
The Distorted Wave Theory

Formalism
The next step in improving theoretical description of direct reactions was to introduce
the concept of distorted waves. In this description it is assumed that there is nuclear
interaction between a and A and b and B, that is, that the interaction potentials
U α ≠ 0 and U β ≠ 0 . However, the key assumption is that the interactions in both the
entrance and exit systems are dominated by the elastic scattering. Coupling between
all other channels (inelastic scattering and rearrangement) which result from a
collision of a with A are treated as perturbations and are absorbed in the imaginary
components of the potentials describing the systems a + A and b + B. The transition
matrix can then be written as
rr rr r r
TDW = ∫ψ aA
( − )*
( k a ra ) ψ bψ B V ψ aψ A ψ bB
(+)
(kb rb )dra drb
where ψ aA
(−)
and ψ bB
(+)
are the “distorted waves”. They are just elastic scattering waves
describing the relative motion of a + A and b + B. These functions are generated by
solving Schrödinger equations containing optical model potentials.
In principle, parameters of optical model potentials in the entrance and exit channels
should be determined by fitting angular distributions for the elastic scattering of a from
A and b from B. However, in general, only the distributions for the elastic scattering of a
from A can be measured and fitted using optical model procedure. To determine
optical model parameters for the exit channel, b + B, various alternative methods are
used. If b can be used as a projectile and B as a target and if the energy of b can be
made to match the energy of b in the exit channel of the studied reaction, then optical
model parameters for the exit b + B channel can be determined by measuring the b +
B elastic scattering angular distribution and fitting it using optical model procedure. If
the elastic scattering angular distributions for the exit channels cannot be measured,
then one can use optical model parameters determined for the neighbouring masses
and energies of b and B. In fact, it may be argued that it is better to use the “average”
optical model parameters both in the entrance and exit channels when fitting the
A(a,b)B angular distributions. The “average” parameters are obtained by studying
elastic scattering angular distributions over a sufficiently wide range of targets A and B
for incident particles a and b and for energies, which are in the vicinity of the energies
for the entrance and exit channels of the reaction A(a,b)B.
In its simplest form, the optical model potential contains only two central components,
the real and the imaginary. The interaction potential U(r) describing the relative motion
of particles can be written as:
U (r ) = −Vf ( x) − iWg ( x′) + VC (r )
where V and W are potential depths, f (x ) and g (x′) are the shape functions and VC(r) is
the Coulomb potential. The function f (x ) has the Woods-Saxon form:
r − r0 A1 / 3
f ( x) = (1 + e x ) −1 where x =
a
The function g ( x′) has either the Woods-Saxon form (for the volume absorption):
r − r0′A1 / 3
g ( x′) = (1 + e x′ ) −1 x′ =
a′
or the derivative of the Woods-Saxon shape (for the surface absorption):
g ( x′) = 4a′
d
dr
[
(1 + e x ′ ) −1 ]
The Coulomb potential VC(r) is taken as the potential due to a uniformly charged
sphere of radius RC:
Ze 2
VC (r ) = for r > RC
r
Ze 2 ⎛ r2 ⎞
VC (r ) = ⎜⎜ 3 − 2 ⎟⎟ for r ≤ RC
2 RC ⎝ RC ⎠
where Z is the charge of the target nucleus.
Other terms may be added to the optical model potential. The most common
additional term is the spin-orbit potential:
2
⎛ h ⎞ 1 dh(r , rso , aso )
V (r ) = Vso ⎜⎜ ⎟⎟ L ⋅S
⎝ mπ c ⎠ r dr
where Vso is the spin-orbit potential depth and
1
h( r , rso , aso ) =
1 + exp[(r − rso A1 / 3 ) / aso ]
Finally, more complex tensor components are also included when necessary (see for
instance Chapter 24).
r r
The distorted waves ψ ( ± ) (k , r ) generated by the optical model potentials in the input
r
and output channels take an asymptotic form of a plane wave with momentum k plus
an outgoing (or incoming) spherical scattered wave, which in the absence of Coulomb
interaction has the form
r r rr
e ± ikr
ψ ( ± ) (k , r ) → eik ⋅r + f (θ )
r
If we assume that the distorting potentials for the incoming and outgoing particles
contain spin orbit interaction then the distorted waves become matrices in the spin
space
v r r r
ψ ( ± ) ( k , r )η s , m = ∑ψ m( ±, m) ′ (k , r )η s , m′
m′
where η s, m are spin functions.
To solve the relevant Schrödinger r r

equations, the usual partial wave expansion is
carried out for functions ψ m , m′ (k , r )
(±)
r r 4π M ′ − m′ r
r
ψ m( ±, m) ′ (k , r ) =
kr
∑ i
J ,L
L
2 L + 1χ JLs ( k , r )( LsMm | J M ′)( Ls M ′ − m′m′ | JM ′)YL ( r ) d L
0, M ′ − m ′ ( k )
where d L
0, m are the rotations functions for integer spin (Edmonds 1957).
The radial parts of distorted waves satisfy the following equations:

⎧ d2 L( L + 1) 2μ ⎫
⎨ 2 +k −
2
2
− 2 [U (r )]⎬ χ JLs (k , r ) = 0
⎩ dr r h ⎭
where U ( r ) is the potential describing elastic scattering of a from A or b from B. It
contains the Coulomb interaction potential and the usual optical model potential (with
the spin-orbit interaction).
From here, the description of the theory becomes a little more complicated.
The form factor ψ bψ B V ψ aψ A , which contains information about nuclear structure can
be expressed as a sum containing elements Blsj and f lsj . Elements Blsj measure the
strength of the interaction and f lsj contain the details of reaction model. To simplify the
calculations, delta function is also introduced. It describes zero-range interaction.
Later simple procedure may be used to correct for the final-range interaction. The
expression for the form factor ψ bψ B V ψ aψ A has the following form:
ψ bψ B V ψ aψ A ≡ J B M Bsb mb V J A M Asa ma
= ∑ Blsj ( J A jM A M B − M A | J B M B )(sa smb ma − mb | sa ma )(lsmma − mb | jM A − M B )
lsj
⎛ A ⎞ r
× flsj (rc )δ ⎜ rb − ra ⎟i − lYl m (ra )*
⎝ B ⎠
The angular momenta satisfy the following relations:
ˆj = Jˆ B − Jˆ A sˆ = sˆb − sâ lˆ = ˆj − sˆ
In practice, usually only one set of jls transfer is involved so the calculations are
significantly simplified.
Using the information about the partial waves and the nuclear form factor, we can now
express the transition matrix element in the following form:
4π
TM AM B ;ma mb =
k a kb
∑
lsj
2l + 1Blsj ( J A jM A M B − M A | J B M B ) Slsjmma mb
This expression contains the already familiar Blsj element and a Slsjmma mb , known as the
reaction amplitude. Hidden in the reaction amplitude are the following elements: (a)
the angular dependence expressed as a sum of Legendre polynomials, (b) the model-
dependent radial form factors f lsj , and (c) the distorted partial waves χ JL for the
entrance and exit channels.
Slsjmmamb = ∑ βlsjmm;Labmb PLmb a −m−mb
Lb
where the amplitude β lsj; Lb has the form:

mma mb
β lsjmm;L m
a
b
b
= ∑ ( L s 0m | J m )( L s Mm
J a La J b m
a a a a a b b a − m − mb mb | J b ma − m)
× ( J bjma − mm | J a m a )(2 Lb + 1)( Lbl 00 | La 0)

⎧ Lb sb Jb ⎫
⎪ ⎪
× (2 sa + 1)(2 j + 1)(2 J b + 1)(2 La + 1) ⎨ l s j⎬
⎪L J a ⎪⎭
⎩ a sa
La − Lb −l
× I J a La Jb L bi
lsj
This formula contains the usual Clebsch-Gordan coefficients (….|..) and 9-j symbol {}.
It also contains the radial integrals I which are related to the partial waves χ in the
following way:
CB ∞ ( − ) A
2 ∫0
I Jlsja La J b Lb = χ J b Lb (kb , ra ) f lsj (rc ) χ J( −a L) a (k a , ra )drc
A B
where C is the mass of the core of the form factor.
The differential cross section for the reaction A(a,b)B, which can be either inelastic
scattering or transfer reaction, is expressed in terms of the transition amplitude T as
follows:
dσ (θ )
dΩ
⎛ μ ⎞ υ
2
1
∑ TM M ;m m
2
=⎜ b2 ⎟ b
⎝ 2πh ⎠ υa (2 J A + 1)(2 sa + 1) M AM B ma mb A B a b
2
1 2 J B + 1 1 kb 1
= ∑ ∑ 2l + 1Blsj Slsjmmamb
4π 2 J A + 1 Ea Eb ka 2 sa + 1 ma mb m lsj
where Ea and Eb are the centre of mass energies in the entrance and exit channels,
respectively.
For the inelastic scattering (a = b) the differential cross section takes the form:
dσ (θ ) 2 J B + 1 2l + 1 2
= Blsj σ alsja′ (θ )
dΩ 2J B +1 2 j +1
where
2
1 1 kb 1
σ (θ ) =
lsj
aa ′ ∑ ∑
4π Ea Eb ka 2sa + 1 ma mb m lsj
Slsjmma mb
For transfer reactions ( a ≠ b )
2
dσ (θ ) 2 J B + 1 1 Blsj
= σ ablsj (θ )
dΩ 2 J B + 1 2 j + 1 10 4
2
1 1 kb 104
σ (θ ) =
lsj
ab ∑ ∑ 2l + 1Slsjmma mb
4π Ea Eb ka 2sa + 1 ma mb m lsj
In particular, if we consider a stripping reaction A(a,b)B with a = b + x and B = A +x

then the nuclear form factor can be written in terms of spectroscopic factors and a D0
coefficient.
J B M Bsb mb V J A M Asa ma = ∑ S 1jl/ 2 R jl (rxA )(lsmμ − m | jμ )( sb smb ma − mb | sa ma )
jl
r r
× ( J A jM A M B − M A | J B M B ) D(rxb )Yl m (rxb )*
where Sjl is the spectroscopic factor, R jl = f lsj is the radial form factor for the transfer of
r
x into the target nucleus A, and D (rxb ) is the product of the projectile internal function
and the interaction potential between x and b.
r
D (rbx ) ≡ Vbx (rbx )ψ a (rbx )
Assuming zero-range interaction we have
r r r
D (rbx ) = D0δ (rx − rb )
which gives
D0 = ∫ Vbx ( rbx )ψ a ( rbx ) drbx
The reaction strength factor Blsj is then given by
Blsj = Slj1 / 2 D0
Consequently, the expression for the differential cross section for the stripping
reaction A(a,b)B takes the form:
⎡ dσ (θ ) ⎤ 2 J B + 1 Slsj D02 lsj
⎢⎣ dΩ ⎥⎦ = σ ab (θ )
st 2 J B + 1 2 j + 1 104
The cross section for the pickup reaction is
2
⎡ dσ (θ ) ⎤ ⎛ ka ⎞ 2 sa + 1 2 J A + 1 ⎡ dσ (θ ) ⎤
⎢⎣ dΩ ⎥⎦ = ⎜⎜ k ⎟⎟ 2 s + 1 2 J + 1 ⎢⎣ dΩ ⎥⎦
pk ⎝ b⎠ b B st
The zero-range coefficient D0 can be either estimated experimentally, if spectroscopic

factors are known, or calculated. For instance, for deuterons, if we use the Hulthèn
function for ψ a (rbx ) , i.e.
⎡ αβ (α + β ) ⎤ ⎡ e −αr − e − βr ⎤
ψ d (rnp ) ≡ ψ d (r ) = ⎢ 2 ⎥⎢ ⎥
⎣ 2π (α − β ) ⎦ ⎣ r ⎦
we get
3
8πε 2 ⎛ α + β ⎞
D = 3 ⎜⎜
2
⎟
0
α ⎝ β ⎟⎠
where
α 2h 2
ε=
2μ
is the deuteron binding energy (ε = 2.23 MeV) and μ is the reduced mass (μ = 1/2)
Assuming that β = 7α we can calculate that
D02 = 1.55 × 10 4 MeV2fm3
The difference between plane wave and distorted wave calculations is illustrated in
Figure E.2.
Figure E.2. Plane-wave and distorted-wave calculations are compared with experimental results for
40
Ca(d,p)41Ca reaction (Lee at al. 1964).
Finite-range and nonlocality corrections

A correction for the finite-range interaction for single particle transfer reactions can be
made by multiplying the form factor flsj by one of the following functions
1
WFR = Hulthèn form
1 + A(r )
WFR = e − A(r ) Gaussian from
where
RFR [Eb − Vb ( rb ) + E x − Vx ( rx ) − Ea + Va (ra )]

2 mb mx 2
A( r ) =
h 2 ma
with RFR being the finite range parameter (see Table E.1), and Ea , Eb, Ex, and Va , Vb, Vx
the energies and potentials of particles a, b, and x with respect to the target.
Table E.1
2
Examples of zero-range coefficients D0 and finite range correction factors RFR
Reaction D02 RFR

(d,p) 1.55 0.621
3
( He,d) 4.42 0.770
3
( H,d) 5.06 0.845
4 3
( He, He) 24-46 0.700
4 3
( He, H) 24-46 0.700
The assumption that optical model potential depends only on the distance of two
interacting particles does not always give a satisfactory description of experimental
data. The correction for nonlocality of the interaction potentials in the entrance and exit
channels, and for the transferred particle, may be made by multiplying the form factor
for each particle by the following function:
⎡ β 2 2m ⎤
WNL = ⎢ i 2 i U i (ri )⎥
⎣8 h ⎦
where βi is the non-locality parameter for particle i. For instance, typical values β are.
0.85 fm, 0.54 fm, and 0.25 fm for nucleons, deuterons and tritons, respectively (Bassel
1966).
Coupled channels theory

The coupled channels theory is an extension of the distorted waves formalism. In this
theory, the wave function for the relative motion of interacting particles includes not
only the elastic scattering but also coupling to other processes, which are present in
any nuclear reaction but which are neglected in the distorted wave theory. For
instance, coupling to strong inelastic channels might influence a measured angular
distribution for a transfer reaction.
Coupled channels theory allows also to carry out calculations for the two-step or multi-
step reactions. An example of such processes is discussed in Chapter 17 for the
vector polarization of elastically scattered deuterons from Se isotopes. In that study I
have considered not only the single step (d,d) scattering but also two step processes
(d,d’)(d’,d), (d,p)(p,d), and (d,t)(t,d). I have shown that the experimentally observed
isotopic effects of deuteron polarization can by explained as an interplay between the
direct (d,d) and two-step (d,d’)(d’,d) scattering. Figure E.3 shows a few examples of
one- and two-step excitations.
In this illustration, it is assumed that states j and j′ in the A+1 nucleus have
configurations based on the ground state I of the A nucleus. They therefore can be
excited directly from the ground state I by a single-step process. Such transitions
can be treated using distorted wave theory.
However, the state j ′′ is assumed to have no parentage based on the ground state I
and consequently cannot be excited via a single-step transition. This state can be
accessed via a two-step reaction, which involves the excitation of either a target or
residual nucleus, or both. Transitions of this type have to be treated using coupled
cannels theory. The coupled channels theory allows also for considering two-step
contributions to transitions, which can proceed via a single-step excitation, such as
for the state j′ in Figure E.3. This state can be excited directly from I and indirectly
via the state j.
Figure E.3. Examples of single-step and two-step transitions from the target nucleus A to the residual
nucleus A+1. The double vertical lines indicate strong excitations by inelastic scattering. The state j
can be excited via a single-step transition. The state j ′′ can be excited only via two-step processes,
which involve the excitation of the target and/or residual nucleus. The excitation of the state j′ can
have contributions from a single-step and two-step transitions.
The concept of the coupled channels theory has been discussed in the Appendix D for
the inelastic scattering. However, to see the basic difference between the coupled
channels and distorted waves treatments of direct nuclear reactions it is useful to
compare here the relevant Schrödinger equations for the partial wave functions.
Let us consider a wave function for a channel c in a representation, which couples the
orbital angular momentum of the relative motion lc to the intrinsic spin of the projectile
sc to give jc: ˆjc = lˆc + sˆc .
The momentum jc is then coupled to the target spin Ic to give angular momentum J
with the projection M: Jˆ = ˆjc + Iˆc .
The wave function for all channels can then be written as
(
ΨJM = ∑ χ lcJc jc (kc , rc ) ψ cjclcψ Icc )
JM
cjclc
The ψ cjclc is the wave function for coupling the relative angular momentum and intrinsic
spin, and ψ Icc is the intrinsic wave function for the target. The Schrödinger equations for
the radial wave function χlcJc jc are:
⎧ d2 lc (lc + 1) 2μc ⎫ 2μ
⎨ 2 + kc −
2
2
− 2 [U cc (rc )]⎬ χ cJjclc (kc , rc ) = ∑ 2c′U cc′ (rc′ ) χ cJjclc (kc′ , rc′ )
⎩ c
dr rc h ⎭ c′ h
We can now compare them with the equivalent equations in the distorted waves
theory:
⎧ d2 L( L + 1) 2μ ⎫
⎨ 2 +k −
2
2
− 2 [U (r )]⎬ χ JLs (k , r ) = 0
⎩ dr r h ⎭
We can see that the essential difference is that the zero on the right hand side of the
equations for the distorted waves is now replaces by a sum of terms describing
coupling between various channels.
The diagonal term Ucc is the potential, which describes the relative motion of particles
in channel c. It contains both the Coulomb and optical model potentials. The off-
diagonal term U cc ′ is a coupling potential, which is given by the relation:
U cc′ = ∑ (l s m ν c c
mc mc′ν cν c′ M c M c′
c c | jcγ c )(lc′ sc′mc′ν c′ | jc′γ c′ )
× ( jc I cγ c M c | JM )( jc′ I c′γ c′ M c′ | JM ) i lc Ylcmcφsccφ Jcc M c V i lc′Ylcm′ c′φscc′′φJcc′′M c′
The formalism contains the familiar elements, such as f lsj and Blsj but now they also
have channels indices (e.g. flsjcc ′ and Blsjcc ′ ). These and other quantities enter into the
elements that are used in the calculations of relevant observables, such as the
differential cross sections.
The differential cross section for channel c is given by:
2
dσ c (θ ) 1 1
dΩ
= ∑
2 I c1 + 1 2 sc1 + 1 M c1 M cν c1ν c
f coul (θ ) + ∑ DMl c mc1 cM cν c1ν c Plcmc (θ )
l c mc
where the subscript c1 refers to the initial elastic channel, f coul (θ ) is the Coulomb
scattering amplitude (see Appendix D), DMl cmc cM cν c ν c are the reaction amplitudes, which
1 1
also contain the Coulombs phase shifts, and Plc (θ ) are the Legendre polynomials.mc
References
Bassel, R. H. 1966, Phys. Rev. 149:791.
Edmonds, A. R. 1957, Angular Momentum in Quantum Mechanics, Princeton University,
Princeton, USA.
Glendenning, N. K. 2004, Direct Nuclear Reactions, World Scientific Publishing Co.,
London, UK.
Lee, L. L., Schiffer, J. P., Zeidman, B., Satchler, G. R., Drisko, R. M., and Bassel, R. H.
1964, Phys. Rev. 136:B971.
Satchler, G. R. 1983, Direct Nuclear Reactions, Clarendon Press, Oxford, UK>
Appendix F
Nuclear spin formalism
This Appendix contains a brief outline of the nuclear spin formalism for spin-1/2 and
spin-1 particles. For more details see such publications as Worfenstein (1956),
Gammel, Keaton, and Ohlsen (1970), and Ohlsen (1972).
Spin-1/2 particles
A single spin-1/2 particle can be represented by a Pauli spinor
⎡a ⎤
χ = ⎢ 1⎥
⎣a2 ⎦
The expectation value of an observable corresponding to a Hermitian operator34 O is
given by35
O = χ +Oχ = a1* [ ]⎡O

a2* ⎢ 11
O12 ⎤ ⎡ a1 ⎤
⎥⎢ ⎥
⎣O21 O22 ⎦ ⎣a2 ⎦
= a1 O11 + a1*a2O12 + a2*a1O21 + a2 O22 = TrρO
2 2
where ρ is the density matrix defined as

⎡ a a* a a* ⎤
ρ = ⎢ 1 ⎥[a1* a2* ] = ⎢ 1 1* 1 2* ⎥
⎡a ⎤
⎣ a2 ⎦ ⎣a2 a1 a2 a2 ⎦
For an ensemble of N particles, each element of the density matrix is replaced by the
average value
N
1
ρ jk =
N
∑a
n =1
( n ) ( n )*
j a
k j , k = 1,2
The state of polarization of an ensemble of spin-1/2 particles is specified by the

expectation values of the Pauli spin operators:
p X = σ x = Trρσ x
pY = σ y = Trρσ y
pZ = σ z = Trρσ z
34 +
An operator O is Hermitian if O =O. By definition, if aij are the elements of matrix O and bij
elements of O then bij = a . Consequently, for a Hermitian operator bij = a ij . The average value of
+ *
ji
a Hermitian operator is real.
35
The general formula is O = TrρO / Trρ = TrρO if Trρ = 1 .
where
⎡0 1 ⎤ ⎡0 − i ⎤ ⎡1 0⎤
σx = ⎢ ⎥ ; σ y = ⎢i ⎥ ;σ z = ⎢ ⎥
⎣1 0⎦ ⎣ 0⎦ ⎣0 1 ⎦
are Pauli spin operators.

If we choose the polarization axis along the quantization axis we shall have N+
particles aligned with the quantization axis and N- particles aligned in the opposite
direction. The density matrix is then given by
1 ⎡N+ 0⎤
ρ=
N + + N − ⎢⎣ 0 N − ⎥⎦
and the beam polarization by

N+ − N−
pZ = Trρσ Z =
N+ + N−
with p X = pY = 0 .36
It can be easily seen that Pauli matrices are orthogonal and normalized, i.e. that
Trσ iσ j = 2δ ij
The three Pauli operators and the 2x2 unit matrix form a complete orthogonal set
and consequently, any 2x2 matrix, M, can be expanded in terms of this basic set.
Spin-1 particles
A spin-1 particle is represented by a three-component spinor
⎡ a1 ⎤
χ = ⎢⎢a2 ⎥⎥
⎢⎣ a3 ⎥⎦
The basis spin-1 angular momentum operators are defined as
⎡0 1 0 ⎤ ⎡0 − i 0 ⎤ ⎡1 0 0 ⎤
1 ⎢ ⎥ 1 ⎢
Sx = ⎢ 1 0 1⎥ ; S y = ⎢ i 0 − i ⎥ ; S x = ⎢⎢0 0 0 ⎥⎥
⎥
2 2
⎢⎣0 1 0⎥⎦ ⎢⎣0 i 0 ⎥⎦ ⎢⎣0 0 − 1⎥⎦
These three vector polarization operators and a unit matrix form a set of four
Hermitian operators but nine are needed to span the 3x3 space. It is therefore
necessary to construct additional operators. These additional operators are defined
by the following relations (Goldfarb 1958):
36
I am using the capital letters for the subscripts to distinguish the polarization components in the
symmetry axis frame of reference and the components in the laboratory frame of reference, which will
be introduced later (see the Appendix H).
3
Sij = ( Si S j + S j Si ) − 2δ ij I j , k = x, y , z
2
where
⎡1 0 0⎤
I = ⎢⎢0 1 0⎥⎥
⎢⎣0 0 1⎥⎦
Explicitly, the additional operators are

⎡0 0 − i ⎤ ⎡0 1 0⎤ ⎡0 − i 0 ⎤
3⎢ ⎥ 3 ⎢ ⎥ 3 ⎢
S xy = ⎢0 0 0 ⎥ S xz = ⎢ 1 0 − 1⎥ S yz = ⎢ i 0 i ⎥⎥
2 8 8
⎢⎣ i 0 0 ⎥⎦ ⎢⎣0 − 1 0 ⎥⎦ ⎢⎣0 − i 0⎥⎦
⎡− 1 0 3 ⎤ ⎡ − 1 0 − 3⎤ ⎡1 0 0⎤
S xx = ⎢ 0 2 0 ⎥ S yy = ⎢ 0 2 0 ⎥⎥
1⎢ ⎥ 1⎢
S zz = ⎢⎢0 − 2 0⎥⎥
2 2
⎢⎣ 3 0 − 1⎥⎦ ⎢⎣− 3 0 − 1⎥⎦ ⎢⎣0 0 1⎥⎦
This definition adds six extra operators while only five are required to make a
complete set of nine. We therefore have an over-complete set. However, it can be
seen that
S xx + S yy + S zz = 0
which means that there are only nine independent operators. It can be also checked
that they form an orthogonal set but they are not normalized.
TrSi S j = 0 for i ≠ j
TrSi S j = 2 for i = j
TrSij S kl = 0 for ij ≠ kl
TrSij Sij = 9 / 2
TrSii Sii = 6
We can define a new set of operators Ωi related to the operators Si and Sij that will
also form a complete set and for which
TrΩi Ω j = 3
Explicitly, this new set of nine operators is as follows:

Ω0 = I
3 3 3
Ω1 = Sx Ω2 = Sy Ω3 = Sz ;
2 2 2
2 2 2
Ω4 = S xy Ω5 = S xx Ω 6 = S xx
3 3 3
1 1
Ω7 = ( S xx − S yy ) Ω8 = S zz
6 2
If, as for spin-1/2 particles, we chose the z axis to be along the quantization axis, then
the density matrix is given by
⎡N+ 0 0⎤
ρ=
1 ⎢0 N0 0 ⎥⎥
N+ + N0 + N− ⎢
⎢⎣ 0 0 N − ⎥⎦
Using this matrix, we can calculate that

⎡ N+ 0 0 ⎤ ⎡1 0 0 ⎤ ⎡ N+ 0 0⎤
ρS z =
1 ⎢0 N0 0 ⎥⎥ ⎢⎢0 0 0 ⎥⎥ =
1 ⎢0 0 0 ⎥⎥
N+ + N0 + N− ⎢ N+ + N0 + N− ⎢
⎢⎣ 0 0 N − ⎥⎦ ⎢⎣0 0 − 1⎥⎦ ⎢⎣ 0 0 N − ⎥⎦
Thus
N+ − N−
pZ ≡ S z = TrρS z =
N+ + N0 + N −
Likewise, we can see that

⎡ N+ 0 0 ⎤ ⎡1 0 0⎤ ⎡N+ 0 0⎤
ρS zz =
1 ⎢0 N0 0 ⎥ ⎢0 − 2 0 ⎥⎥ =
⎥ ⎢ 1 ⎢0 − 2N0 0 ⎥⎥
N + + N0 + N− ⎢ N+ + N0 + N− ⎢
⎢⎣ 0 0 N − ⎥⎦ ⎢⎣0 0 − 1⎥⎦ ⎢⎣ 0 0 N − ⎥⎦
and thus
N+ + N− − 2N0
pZZ ≡ S zz = TrρS zz =
N+ + N0 + N−
We can also calculate that
1
p XX ≡ S xx = pYY ≡ S yy = − pZZ
2
and therefore
p XX + pYY + pZZ = 0
In general, the description of the state of polarization of spin-1 beam, which does not
have an axis of symmetry, may require all three vector components and five
independent tensor components. It should be also noted that expectation values of
the various operators are bound by the limits ±1 for the vector polarization, ±3/2 for
pij components with i ≠ j , and +1 to -2 for pii components.
Some basic definitions
Conjugate matrix. If aij are the elements of matrix A ( A = [ aij ] ) then A* = [bij ] is called
a conjugate matrix of A if bij = aij* , where aij* is the complex conjugate of aij .
Transpose matrix. If aij are the elements of matrix A ( A = [ aij ] ) then AT = [bij ] is
called a transpose matrix of A if bij = aji.
The conjugate transpose matrix. If A = [ aij ] then A+ = [bij ] is called its conjugate
transpose if bij = a*ji .
The inverse matrix. A matrix A−1 is an inverse of a square matrix A = [ aij ] if AA−1 = I ,
where I is the unit (the identity) matrix.
The unitary matrix. A square matrix A is called unitary if AA+ = A+ A = I , where I is
the unit (the identity) matrix. For a unitary matrix A−1 = A+ .
The orthogonal matrix. Matrix A is orthogonal if AAT = I .
A unitary orthogonal matrix is made of a set of orthogonal unit vectors. For instance,
in the case we have considered in this Appendix, the matrix U is made of two two-
dimensional vectors: r
a = (cos ε , sin ε )
r
b = ( − sin ε , cos ε )
It is easy to see that these vectors are not only orthogonal but also normalised to
r r r r r r
unity, i.e. that a ⋅ a = b ⋅ b = 1 and a ⋅ b = 0 . Thus matrix U is both orthogonal and
unitary.
References
Gammel, J. L., Keaton, Jr, P. W. and Ohlsen, G. G. 1970, Los Alamos Laboratory
Report, LA-4492-MS.
Appendix G
Polarized ion sources
The two main types of polarized ion sources are the atomic-beam and the Lamb-shift
varieties. Excellent introduction to these experimental devices may be found in an
article by Haeberli (1967) but their components and performance were also
discussed in numerous other publications. Below, I am presenting their simplified
description.
The atomic beam polarized ion sources

Protons
The conventional polarized ion source is based on the Stern-Gerlach separation of
atoms belonging to different hyperfine substates. This can be understood with the
help of Figure G.1 showing the Zeeman effect.
Figure G.1. Breit-Rabi energy level diagram of the 1S1/2 hyperfine states of the hydrogen atom in the
magnetic field. The energy W is in units ΔW and the magnetic field in Bc (the critical magnetic field
associated with the decoupling of the electron and nuclear moments). For the ground state of the
hydrogen atom ΔW = h × 1420.2 MHz (5.8 × 10-6 eV) and Bc = 507 G. It should be noted that for
weak fields ( χ << 1 ), the proton and electron spins cannot be treated as independent quantities and
consequently, the hyperfine states have to be labelled using mF projection.
In the hydrogen atom the electron with spin j = 1/2 (projections mj = ±1/2) couples
with the proton spin I = 1/2 (projections mI = ±1/2) to form spin F, Fˆ = Iˆ + ˆj , which
gives two values of F (F = 0, 1) with projections mF = 0, ±1.
In the absence of the magnetic field, the separation between the two substates F is
ΔW = h × 1420.2 MHz (5.8 × 10-6 eV) (Nagle, Julian, and Zacharias 1947). When the
external magnetic field is applied, the substate F = 1 splits into three components,
which is known as the Zeeman effect.37 The energy split is described by the Breit-
Rabi formula (Ramsey 1956):
ΔW 4mF
W= + g I μ B mF Bc χ + (−1) F +1 1 + χ + χ2
2(2 I + 1) 2I + 1
where g I = -3.04 × 10-3 and g j = 2.002, which are, respectively, the g-factors of the
proton and electron in the units of the Bohr magneton μ B = -0.927 × 10-20 erg/G.
The external magnetic field is measured in the units of the so-called critical field Bc
and expressed in terms of the ratio χ = B / Bc , where
ΔW
Bc =
( g I − g j )μB
which for the hydrogen atom in its ground state is 507 G.
The basic idea, used in most of the atomic beam polarized ion sources is first to
reject states 3 and 4 (see Figure G.1) but accept states 1, and 2. This is done by
following the original idea of the well-known Stern-Gerlach experiment but by using
improved inhomogeneous magnets. In the original Stern-Gerlach experiment, a
dipole magnet was used. In polarized ion sources the performance is improved by
using a sextupole magnet or to a lesser degree by a quadrupole.
In an inhomogeneous magnetic field, the atom will move to the magnetic field that
causes a decrease in the energy of its hyperfine substate. Thus if we look again at
the Figure G.1 we shall see that the atoms in substates 1 and 2 will move to an area
where the magnetic field is weak. In contrast, the atoms with substates 3 and 4 will
move in the direction of strong field.
In a sextupole (or quadrupole) magnet the gradient of the magnetic field is in the
radial direction, i.e. it decreases radially towards the centre. This means that the
atoms with substates 1 and 2 will move towards the centre, whereas the atoms with
substates 3 and 4 will move away from the centre. Thus, they will be removed from
the beam and pumped away.
This beam manipulation results in an atomic beam polarized in the electron spin ( mj
= +1/2) but not in the proton spin (mI = ±1/2). To polarize the atomic beam in the
proton spin we have to change the population of the hyperfine states. It is possible,
for instance to move the atoms in state 2 to state 4 and thus change the combination
of states 1 and 2 to 1 and 4. For these two states mI = +1/2. Such transitions
between states are achieved by applying an RF field.
Having achieved a desired combination of hyperfine substates, the next step is then
to ionise the polarized atoms to make them ready for the acceleration. This is done
in suitably designed ionisers.
There is, however, a slight complication in this relatively simple process of forming
proton-polarized hydrogen beams because the degree of proton polarization for
substates with antiparallel electron and proton spins depends on the external
37
An analogous splitting in an electric field is known as the Stark effect, or rarely as Stark-Lo Sturdo
effect. It was discovered independently in 1913 by German and Italian physicists, Johannes Stark and
Antonio Lo Sturdo. This effect is used in Lamb-shift sources in combination with the Zeeman effect.
magnetic field in the ioniser. This dependence is associated with the precession of
the electron and proton spins in the weak magnetic field and is shown if Figure G.2.
Figure G.2. The dependence of the vector polarization of protons in the hydrogen atom as the
function of the external magnetic field.
We shall recall (see the Appendix F) that vector polarization pz is given by:
N+ − N−
pZ =
N+ + N−
where N± is the number of atoms with proton spin projections mI = ±1/2.
The proton polarization of the atomic beam is an arithmetic average of the
polarization of its various components. So, for instance, if we apply an RF field and
move the atoms from the state 2 to 4, and if we follow this process by ionisation in a
weak magnetic field then we shall have atoms in state 1 with the net polarization pz =
1 and in state 4 with polarization pz ≈ 0. If we assume that we have the same
number of hydrogen atoms in states 1 and 4, then the resulting average vector
polarization for such a beam will be
pz ≈ 0.5 (states 1 and 4 in the weak field)
However, if we ionise the hydrogen atoms in a strong magnetic field then the
average polarization will be
pz ≈ 1.0 (states 1 and 4 in the strong field)
It should be noted that beam polarization can be created in a weak field without the
use of the RF transition. If we look again at the Figure G.2 we shall see that if we use
the hydrogen atoms in substates 1 and 2 (as they emerge from the sextupole
magnet) and if we assume that the population of these two substates is the same
then the net polarization created in the weak magnetic field will be
pz ≈ 0.5 (states 1 and 2 in the weak field)
Such weak-field devices were used in the late 1950s and early 1960s but they were
not successful because of the low beam intensities they produced.
A simplified schematic diagram of a conventional (atomic-beam) polarized ion source
is shown in Figure G.3
Figure G.3. A simplified schematic diagram showing the major components of the atomic beam
polarized ion source. Atomic Beam – The atomic beam formation; 6-pole Magnet – An
inhomogeneous magnet, sextupole or quadrupole. RF Transition – An RF unit to change the
population of the hyperfine substates; Ioniser – A device used to convert the neutral atomic beam with
polarized protons to negatively or positively charged ions; Rotation – A device used to change the
orientation of the polarization axis.
Deuterons
For the deuterium atoms the principles are the same but now I = 1, mI = 0, ±1, F =
1/2, 3/2, and mF = ±1/2, ±3/2. Consequently, we have three, rather than two,
hyperfine substates in each branch. We also have a possibility of creating a beam
with vector and tensor polarization.
The Breit-Rabi diagram for the deuterium atoms is presented in Figure G.4 and the
corresponding vector and tensor polarizations for various hyperfine components in
Figure G.5. In the absence of the external magnetic field ΔW = h × 327.4 MHz
(1.4 × 10-6 eV). The critical magnetic field Bc = 117 G.
Figure G.4. The Breit-Rabi energy level diagram of the deuterium atom in the magnetic field. For the
ground state of the deuterium atom ΔW = h × 327.4 MHz (1.4 × 10-6 eV) and Bc = 507 G.
After leaving the inhomogeneous magnetic field, the beam of the deuterium atoms
will be composed of 1, 2, and 3 substates. Again, the polarization of their nuclei
(deuterons) can be achieved by a selective transfer between the desired substates
with the use of RF fields.
Figure G.5. The dependence of the vector and tensor polarizations of protons in the deuterium atom
as the function of the external magnetic field.
Table G.1
Examples of the RF transitions the resulting beam polarizations
a b c d e f g
Strong 2↔6 3↔5 3↔5 2↔6 2↔6
Field
3↔5
Weak 1↔4 1↔4 1↔4
Field
2↔3 3↔2
5↔6 6↔5
Substates 1+2+3 2+3+4 1+5+6 1+2+5 3+4+6 1+3+6 2+4+5
mI +1,0,-1 0,-1,-1 +1,0,+1 +1,0,0 -1,-1,+1 +1,-1,+1 0,-1,0
pz 0 − 2/3 + 2/3 +1 3 −1 3 +1 3 −1 3
pzz 0 0 0 −1 +1 +1 −1
t10 0 − 23 + 23 +1 6 −1 6 +1 6 −1 6
t20 0 0 0 −1 2 +1 2 +1 2 −1 2
Weak and strong field refer to the magnetic fields used for the RF transitions. The ionisation is
performed in a strong magnetic field. The relation between Cartesian and spherical polarizations are
(see Chapter 15): t10 = (3 / 2)1 / 2 pz and t20 = 2 −1 / 2 p zz .
As in the case of hydrogen beams, polarization can be also created without the use
of RF fields but just by using the states 1,2, and 3 as they emerge from the
inhomogeneous magnet.
We can recall (see the Appendix F) that for deuterons, vector pz and tensor pzz
polarization are given by:
N+ − N−
pZ =
N+ + N0 + N−
N+ + N− − 2 N0
pZZ =
N+ + N0 + N−
where N + , N − , and N 0 are the number of deuterium atoms with their nuclear
(deuteron) spin projections mI = 1, -1, and 0, respectively.
If we now examine the Figure G.5, we shall see that in the weak field:
pz ≈ 1/3 (states 1, 2, and 3 in the weak field)
pzz ≈ -1/3 (states 1, 2, and 3 in the weak field)
In practice combinations of RF transitions are used to create desired deuteron
polarizations. A few examples are listed in Table G.1.
The Lamb-shift polarized ion sources

As described earlier, the key procedure in creating a polarized beam is to select
suitable hyperfine states of the hydrogen or deuterium atoms. The same applies to
the Lamb-shift sources except that the atoms are now in their excited states and the
selection of suitable hyperfine states is done in a different way.
Figure G.6 shows the energy diagram of the hydrogen atom for the principal
quantum number n = 2 for which there are three states 2P1/2, 2P3/2, and 2S1/2. Figure
G.6 shows only the two states, 2P1/2 and 2S1/2. The 2P3/2 state is 10,968 MHz above
the 2P1/2 and normally is not included in the mechanism of creating polarized beams.
The 2P1/2 and 2S1/2 states are separated by 1058 MHz (4.4 × 10-6 eV), known as the
Lamb-shift (Lamb and Retherford 1950). The state 2P1/2 decays rapidly by emitting
intense radiation in a discrete Lyman α line (1216 Å). Its lifetime is only 1.6 × 10-9 s.
In contrast, the 2S1/2 state decays slowly by emitting radiation over a broad
continuum and its lifetime is around 0.14 s.
The basic idea of creating polarized ion beams using a Lamb-shift source is
explained in the caption to Figure G.6. A schematic diagram of a Lamb-shift source
is shown in Figure G.7.
The metastate hydrogen atoms can be produced by capturing electrons to a
metastate orbit. This can be achieved by first producing H+ ions and then passing
them through a charge donor, such as cesium gas. The H(2S) source will produce
not only the metastate hydrogen atoms but also the ground state H(1S) hydrogen as
well as H+ and H- particles. In fact, the cross section for producing the metastate
hydrogen is small so most of the output will be made of the H(1S) atoms. The
charged particles H+ and H- can be removed by magnetic fields before passing the
beam to the spin filter but the background made of the H(1S) hydrogen will remain.
Figure G.6. The energy-level diagram for the n = 2 excited states of the hydrogen atom. The 2P3/2
state, which is 10,968 MHz above the 2P1/2 state and which is not used in the Lamb-shift polarization
mechanism is not shown. The short-lived 2P1/2 and metastable 2S1/2 states are separated by 1058.0
MHz (= 4.4 × 10-6 eV), known as the Lamb-shift. The idea behind the Lamb-shift polarized ion sources
is to remove the components β- and β+ by mixing them with the short-lived components e- and e+
components and then to remove one of the α components by forcing it to decay to one of the e
components. This operation leaves α+ or α- each containing aligned proton spins. The basic idea for
the deuteron polarization is the same but the number of hyperfine levels in each branch is then
increased to three. In this diagram, labels α, β, e, and f refer to different orientations of the electron
spin (mj = +1/2, -1/2, +1/2, and -1/2, respectively) and the subscripts + and - to different orientations of
the proton spin (mI = +1/2 and -1/2, respectively).
Figure G.7. Schematic diagram showing the major components of the Lamb-shift polarized ion
source. H(2S) Source – a source of the metastable hydrogen (of deuterium) atoms; Spin Filter – a
system of fine-tuned static magnetic, electric, and RF fields producing hydrogen (or deuterium) atoms
with aligned nuclear spins; Ioniser – a selective ioniser producing positive or negative hydrogen (or
deuterium) projectiles; Rotation – spin rotation unit.
The spin filter contains all the necessary units for creating a hydrogen beam with
aligned nuclei. As outlined in the caption to Figure G.6, the first step is to remove the
metastable components β from the atomic beam. If the magnetic field in the direction
of the beam is chosen to have a value of 535 or 605 G (see Figure G.8), and if in
addition, a perpendicular electric field is applied then states β are mixed with states e
and are forced to decay to the ground state. This process is described as quenching.
Quenching leaves the substates α, which have only one projection of the electron
spin, mj =+1/2, but two components of the proton spin, mI = ±1/2. As can be seen in
Figure G.8, the substates α+ and α− are separated by about 1600 MHz from the
states e+ and e- if the external magnetic field is 535 or 605 G. Consequently, if in
addition to the external magnetic and electric fields we apply an RF field with the
frequency of around 1600 MHz we shall be able to force the state α+ or α− to decay
leaving one component α+ or α− containing positively or negatively polarized protons.
Figure G.8. Level diagram for the hydrogen atoms in the magnetic field illustrating the mechanism of
producing polarized beams. The diagram for deuterium is similar but with each branch containing
three substates corresponding to mI = 0, ±1. As in Figure G.6, groups labelled α, β, e, and f are for the
electron spin orientations mj = +1/2, -1/2, +1/2, and -1/2, respectively, and the subscripts + and – for
the orientations of the proton spin (mI = +1/2 and -1/2, respectively).
The atomic beam with aligned protons still contains the unpolarized H(1S) atoms. To
remove them, the output of the spin filter is then passed to a selective ioniser. The
charge-exchange medium for the ioniser can be made of argon or other suitable gas.
The ioniser removes the unwanted H(1S) atoms and produces H+ or H- ions (which
are now polarized) to suit the type of the particle accelerator used in the
experimental work.
The previously discusses dependence of the beam polarization on the magnetic field
in the ioniser (Figures G.2 and G.5) also applies to the Lamb-shift source. However,
decoupling of the electron and nuclear spins occurs at lower magnetic fields. The
critical magnetic field for the metastate hydrogen or deuterium atoms is low (Bc =
63.4 G for the hydrogen and 14.6 G for deuterium, as compared to 507 G and 117 G
respectively for the ground state). So, if a strong field ioniser is used, the produced
polarization will be close to +1 or -1 for the α+ or α− substates, respectively.
After passing a selective ioniser, the polarized beam is suitable for the acceleration.
A spin rotation device may be also added to manipulate the direction of the
polarization.
References
Haeberli, W. 1967, Ann. Rev. Nucl. Sci. 17:373.
Lamb, Jr., W. E. and Retherford, R. C. 1950, Phys. Rev. 79:549.
Nagle, D. E., Julian, R. S., and Zacharias J. R. 1947, Phys. Rev. 72:971.
Ramsey, N. F. 1956, Molecular Beams, The Clarendon Press, Oxford.
Appendix H
Selected nuclear spin structures and the extreme values of the
analyzing powers
r
r 1 1
The 1 + → + 0 spin structure38
2 2
r
An example of this spin structure is the reaction the 3He (d , p ) 4He reaction (Chapter
19).
The initial state is made of projectiles with spin-1 and the target nucleus with spin-1/2.
It is therefore a direct product of spin-1 and spin-1/2 space and thus, the spin
function of any initial state can be described in terms of a six-component vector. The
initial spin state can be expressed as
6
χ i = ∑ a jφ j
j =1
where φ j is a product of various combinations of spin functions of spin-1 and spin-1/2

of interacting nuclei. Explicitly
φ1 = χ11χ 1 1 φ2 = χ10 χ 1 1 φ3 = χ1−1χ 1 1 φ4 = χ11χ 1 1 φ5 = χ10 χ 1 1 φ6 = χ1−1χ 1 1
− − −
22 22 22 2 2 2 2 2 2
where
χ1 M are the eigenfunctions of the spin-1 operator Sz and χ 1 are the
M
2
eigenfunctions of the spin-1/2 operator σz.

The final sate consists of spin-1/2 and spin-0 particles, so it can be expressed as
2
χ fi = ∑ a jφ ′j
j =1
φ1′ = χ 1 1 and φ2′ = χ 1 1 .

−
22 2 2
The scattering amplitude represented by matrix M (Wolfenstein 1956) relates the

initial and final states,
b j = ∑ M jk ak j = 1,2 k = 1,...,6
k
or
38
See Gammel, Keaton, and Ohlsen (1970), and Ohlsen (1972)
⎡ a1 ⎤
⎢a ⎥
⎢ 2⎥
⎡ b1 ⎤ ⎡ M 11 M 12 M 13 M 14 M 15 M 16 ⎤ ⎢ a3 ⎥
⎢b ⎥ = ⎢ M ⎢ ⎥
M 26 ⎥⎦ ⎢a4 ⎥
⎣ 2 ⎦ ⎣ 21 M 22 M 23 M 24 M 25
⎢ a5 ⎥
⎢ ⎥
⎢⎣a6 ⎦⎥
The 6x6 and 2x2 density matrices that describe the initial and final states are
defined, respectively, for an ensemble of N particles as
1 N ( n ) ( n )*
( ρi ) jk = ∑ a j ak
N n −1
j , k = 1,...,6
1 N ( n ) ( n )*
( ρ f ) jk = ∑ b j bk
N n −1
j , k = 1,2
As mentioned earlier, the initial and finial channels are related via the scattering
matrix M
χ f = Mχ i
ρ f = Mρi M +
The density matrix for the initial state can be expanded in terms of products of spin-1
and spin-1/2 operators. If we use the notations σ 0 = I (the unit matrix), σ1 = σx, σ2 =
σy, and σ3 = σz , for the Pauli spin operators, and the normalized spin-1 operators Ωi
(see the Appendix F) we can write
1
ρi = ∑ ω j pk Ω jσ k
6 jk
j = 0,1,...,8 k = 0,1,2,3
In this expression, ω j and pk are the expectation values of the operators Ω j (for spin-
1 particles) and σ k (for spin-1/2 particles), respectively, i.e. ω j = Ω j and pk = σ k .
For the spin structure considered here, only spin-1 particles are polarized in the
entrance channel; spin-1/2 particles are not. Consequently, pk = 0 for k=1, 2, 3 and
only p0 = 1 . In this case the expression for the density matrix in the entrance channel
can be simplified:
1 1
ρi = ∑
6 j
ω jΩ j I = ∑ω jΩ j
6 j
The I is the unity matrix and therefore can be suppressed.
Figure H.1. The definition of angles β and ϕ.
If ρi is normalized to unity, the differential cross section for spin-1/2 particles in the
exit channel is given by:
1 1
∑
σ (θ ,ϕ ) = Trρ f =
6 j
ω jTr ( MΩ j IM + ) = ∑ ω jTr ( MΩ j M + )
6 j
v v
where θ is the scattering angle (i.e. the angle between kin and kout ) and ϕ is the angle
of projection of the quantization axis on the x-y plane (see Figure H.1).
For an unpolarized beam
1
ρi = I
6
and
1 1
σ (θ ,ϕ ) ≡ σ 0 (θ ) = Tr ( MIM + ) = Tr ( MM + )
6 6
We can therefore write
∑ ω Tr (MΩ M
j j
+
)
σ (θ ,ϕ ) = σ 0 (θ ) j
TrMM +
The polarization of spin-1/2 particles in the outgoing channel is given by
) ∑
ω Tr ( MΩ M σ ) +
k′
Tr ( ρ f σ k ′ j j
pk ′ ≡ σ k ′ = = j
Trρ f ∑ ω Tr (MΩ M )
j
j j
+
which can be rewritten as
⎡ ∑ ω jTr ( MΩ j M +σ k ′ ) ⎤
⎢ ⎥
pk ′σ (θ ,ϕ ) = σ 0 (θ ) ⎢ j + ⎥
TrMM
⎢⎣ ⎥⎦
Rotation from the spin-symmetry system of reference, where the polarization of spin-
1 particles is described by just two quantities pZ and pZZ, to the projectile helicity
system of reference as shown in Figure H.1, introduces new polarization
components, which are related to angles β and ϕ.
p x = − pZ sin β sin ϕ
p y = pZ sin β cosϕ
pz = pZ cos β
1
pxx = pZZ (3 sin 2 β sin 2 ϕ − 1)
2
1
p yy = pZZ (3 sin 2 β cos 2 ϕ − 1)
2
1
pzz = pZZ (3 cos 2 β − 1)
2
3 2
pxy = − pZZ sin β cos ϕ sin ϕ
2
3
p yz = pZZ sin β cos β cos ϕ
2
3
p xz = − pZZ sin β cos β sin ϕ
2
p zx = pxz and pzy = p yz
It is easy to see that

3 2
pxx − p yy = − pZZ sin β cos 2ϕ
2
In terms of these components
⎡ 3 1 ⎤
σ (θ ,ϕ ) = σ 0 (θ ) ⎢1 + ∑ p j A j (θ ) + ∑ p jk A jk (θ )⎥
⎣ 2 j 3 jk ⎦
⎡ 3 1 ⎤
pl ′σ (θ ,ϕ ) = σ 0 (θ ) ⎢ Pl ′ (θ ) + ∑ p j K lj′ + ∑ p jk K ljk′ ⎥
⎣ 2 j 3 jk ⎦
where j = x, y, z and jk are all combinations of x, y, z.
The observables are the already mentioned polarization pl ′ of the outgoing spin-1/2
particle (induced by polarized incident beam) as well as A j (θ ) , A jk (θ ) , Pl ′ (θ ) , K lj′ (θ ) ,
K ljk′ (θ ) . They are defined by the following expressions.
Analyzing powers:
Tr ( MS j M + )
A j (θ ) ≡
Tr ( MM + )
Tr ( MS jk M + )
A jk (θ ) ≡
Tr ( MM + )
Outgoing polarization for unpolarized incident beam:
Tr ( MM +σ l′ )
Pl& (θ ) ≡
Tr ( MM + )
Polarization transfer coefficients:
Tr ( MS j M +σ l ′ )
K lj′ (θ ) ≡
Tr ( MM + )
Tr ( MS jk M +σ l′ )
K ljk′ (θ ) ≡
Tr ( MM + )
The M-matrix elements are often written in terms of their expansion coefficients. For
instance, the 6x2 M-matrix can be written as
M = Aχ y+ I + Bχ y+σ y + Cχ x+σ x + Dχ x+σ z + Eχ z+σ x + Fχ z+σ z
where A, B, …, F are coefficients, which depend on the nature of the investigated

reaction, I is the unit matrix, χ i are a complete set of spinors in 3x1 space (see
below) and σ i are the 2x2 Pauli spin operators.
⎡− 1⎤ ⎡1⎤ ⎡0⎤
1 ⎢ ⎥ i ⎢ ⎥
χx = ⎢ 0 ⎥ χy = ⎢ 0⎥ χ z = ⎢⎢1⎥⎥
2 2
⎢⎣ 1 ⎥⎦ ⎢⎣1⎥⎦ ⎢⎣0⎥⎦
χ x+ =
1
[− 1 0 1] χ y+ = − i [1 0 1] χ z+ = 1 [0 1 0]
2 2 2
It is straightforward to show that in terms of the amplitudes A, B, …, F, the M-matrix
has the following form:
1 ⎡− iA − D 2F iA + D − B − C 2E −B+C⎤
M= ⎢ ⎥
2 ⎣ B−C 2E B+C iA + D − 2F − iA − D ⎦
In turn, we can also express the analyzing powers Ay and Ayy in terms of these
amplitudes and find conditions when Ay and/or Ayy reach their extreme values. Thus
for instance, if we use the expression
Tr ( MS yy M + )
Ayy =
Tr ( MM + )
we shall find that Ayy = 1 if A=B=0 (cf Seiler at al. 1976a).
In this case, the M-matrix has the form
1 ⎡− D 2F D −C 2E C ⎤
M= ⎢ ⎥
2 ⎣− C 2E C D − 2F − D⎦
Likewise, we can calculate
Tr ( MS y M + )
Ay =
Tr ( MM + )
and find that Ay = ±1 if not only A=B=0 but also C = m iE and D = m iF .
It is therefore clear that if Ay = 1 then Ayy should also have its extreme value of 1 at
the same energy and angle because both have to satisfy the same condition of
A=B=0. However, if Ayy = 1 , Ay may or may not reach its extreme value of 1. If it
does, the maximum should occur at the same energy and angle as Ayy = 1 . However,
if the M-matrix does not satisfy the two additional requirements of C = ±iE and
D = ±iF then Ay will not reach its extreme value of 1 even if Ayy does. The full list of
the observables, which can be used to check experimentally whether the conditions
A=B=0, C = ±iE , and D = ±iF are satisfied is given by Seiler at al. (1976a).
r
The 1 + 1 = 0 + 0 spin structure
r
An example of this spin structure is the 6Li (d ,α ) 4He reaction (Chapter 20).
To find the conditions for the M-matrix elements, which lead to the extreme values of
the Ay and Ayy components of the analyzing powers we follow a similar procedure as
outlined above. In terms of the expansion coefficients, the M-matrix for this system
can be written as (Seiler at al. 1976b):
⎡A − E 2F A+ E ⎤
1⎢ ⎥
M = − ⎢ 2H − 2K − 2H ⎥
2⎢
⎣ A + E − 2F A − E ⎥⎦
If we use this expression and if we carry out the calculations using the general
expressions for Ay and Ayy , i.e.
Tr ( MS y M + ) Tr ( MS yy M + )
Ay = and Ayy =
Tr ( MM + ) Tr ( MM + )
we shall find that Ay = 1 if A = 0 , H = ±iE , and K = ±iF . In contrast, Ayy = 1 if only
one condition is satisfied, i.e. if A = 0 .
References
Gammel, J. L., Keaton, Jr, P. W. and Ohlsen, G. G. 1970, Los Alamos Laboratory
Report, LA-4492-MS.
Seiler, F., Roy, R., Conzett, H. E. and Rad, F. N. 1976a, Proc. of the Fourth Int. Symp.
On Polarization Phenomena in Nucl. Reactions, Zürich, 1975, Birkhauser Verlag,
Basel, p 550.
Seiler, F., Rad, F. N., Conzett, H. E., and Roy, R.1976b, Proc. of the Fourth Int. Symp.
On Polarization Phenomena in Nucl. Reactions, Zürich, 1975, Birkhauser Verlag,
Basel, p 587.
Appendix I
Analytic determination of the extreme Ayy points
As an example of the analytic determination of the locations for the extreme values
of the analyzing powers I use a study of the d-α scattering (Grüebler et al. 1975b).
The M-matrix for the d-α scattering can be written as
⎡ M 11 M 10 M 1−1 ⎤
⎢
M ( Ed ,θ ) = ⎢ M 01 M 00 − M 01 ⎥⎥ (1)
⎢⎣ M 1−1 − M 10 M 11 ⎥⎦
where the subscripts ij are the deuteron spin projections in the incident and outgoing
channels.
The elements of this matrix depend on the energy of the incident deuterons Ed and
on the reaction angle θ. They can be calculated using experimentally determined
phase shifts.
As shown in the Appendix H, tensor analyzing powers Aij can be expressed in
therms of the M-matrix:
Tr ( MSij M + )
Aij = (2)
Tr ( MM + )
The idea of the analytic determination of the extreme values of the analyzing powers
is as follows: (a) use the experimentally determined phase-shifts to calculate the
relevant matrix elements; (b) use the theoretically established relation between the
elements of the M-matrix, which have to be satisfied for the extreme maxima to
occur (see the Appendix H); (c) study the behaviour of the calculated elements to
see at what points (in energy and angle) the theoretically postulated relationship
between the matrix elements is satisfied.
r r
Using the relation (2) and the matrix (1) we can find that for the 1 + 0 → 1 + 0 spin
structure, Ayy = 1 if
f ( Ed ,θ ) = M 11 ( Ed ,θ ) + M 1−1 ( Ed ,θ ) = 0
Step-by-step numerical calculations can be carried out to locate the
f ( Ed ,θ ) = 0 points.
The function f ( Ed ,θ ) = 0 if both its real and imaginary components are equal zero.
Figure I1 shows the plots of Re(M 11 + M 1−1 ) versus Im(M 11 + M 1−1 ) with the matrix
elements M 11 and M1−1 calculated using the phase shift parameters determined by
Grüebler at al.(1975a).
An alternative way is to calculate the Re(M 11 + M 1−1 ) and Im(M11 + M 1−1 ) components
for various energies and angles but now to identify the ( Ed ,θ ) points where
either Re(M 11 + M 1−1 ) = 0 or Im(M 11 + M 1−1 ) = 0 and plot them. Such a plot is shown in
Figure I2. The points where the lines corresponding to Re(M 11 + M 1−1 ) = 0 and
Im(M 11 + M 1−1 ) = 0 cross over are the points where Ayy = 1
Figure I1. The trajectories of the sum of the M11 and M1-1 components of the M-matrix calculated using
the experimentally determined phase shifts for the d-α reaction. This figure shows that for the
incident energy of between 4.30 and 5.30 MeV, the trajectories move twice over the point of origin of
the coordinate axis. This passes indicates that in this region of energies there should be two points, at
around 600 and 1200, where Ayy = 1 .
Figure I2. Contour plot of Re(M11 + M 1−1 ) = 0 and Im(M 11 + M 1−1 ) = 0 . The Ayy = 1 points should
be located where the contour of Re(M11 + M1−1 ) = 0 crosses over the contours of
Im(M 11 + M 1−1 ) = 0 . The figure shows that there should be two Ayy = 1 points close to 5 MeV and
one at around 12.5 MeV.
References
1975a, Nucl. Phys. A242:265.
1975b, Nucl. Phys. A242:285.
Appendix J
Phase shift analysis
The concept of phase shift analysis of experimental data can be easily understood
by considering the elastic scattering. The differential cross section for the elastic
scattering can be written as (Condon and Shortley 1935):
σ (θ ) = f (θ )
2
where f (θ ) is a complex function, which can be written as
( )
∞
f (θ ) = i π D 2 ∑ (2l + 1) 1 − e 2iδ l Yl , 0 (θ ) (1)
l =0
In this expression, D is the deBroglie wavelength divided by 2π,

λ 1
D= =
2π k
Yl , 0 (θ ) are the well known normalized spherical harmonics as defined by Condon and
Shortley (1935), and δ l are phase shift parameters.
It should be noticed that in the expression (1) the only physical quantities, which are
associated with nuclear interaction, are the phase shift parameters δ l . The
remaining quantities are purely mathematical, with the exception of D . However, this
quantity is not related to the details of nuclear interaction but only to the energy of
the incoming particles.
Phase shift analysis consists in fitting experimental angular distributions with the aim
of extracting the experimentally determined parameters δ l .
In a more general case of nuclear reactions, which may or may not include elastic
scattering, the aim is the same. The general idea is to express the scattering matrix
in terms of the phase shift parameters and analyse experimental observables to
determine the best values for these parameters. Phase shift parameters, which now
define the scattering matrix, can then be used to calculate other observables in the
way described in the Appendices H and I. In this brief summary, I follow the original
notation of Blatt and Biedenharn (1952) who use letter S for the scattering matrix.
Let us consider a general process where two particles collide and two emerge:
a+ X =Y +b
We can use labels α, s, and l to identify the channel before the collision. They are
the channel index, channel spin, and channel angular momentum, respectively. The
state after the collision is labelled using α', s', and l′ .
The channel spin s is made of the intrinsic spin i of the incoming particle a and the
spin I of the target nucleus X ( sˆ = iˆ + Iˆ ). For instance, for the n-p scattering,
i = I = 1 / 2 and s has two values, 1 (triplet state scattering) and 0 (singlet state
scattering).
We also define the total angular momentum of the system as Jˆ = sˆ + lˆ . J is
preserved during the collision, i.e. Jˆ = sˆ′ + lˆ′ . For resonance reactions, J is the
angular momentum of the compound nucleus.
Now we introduce the scattering matrix (scattering amplitude or collision matrix),
which connects channels αls and α ′l ′s′ . We write it as
SαJls ;α ′l ′s ′
The formal definition of the scattering matrix is related to the wave function
Ψαs (JM ) in the channel α , s with total angular momentum quantum numbers JM . At
sufficiently large distance r , the function Ψαs (JM ) can be expressed as a
superposition of an incoming spherical wave with the amplitude AαJM
sl and an
sl .
outgoing spherical wave with the amplitude BαJM
1
Ψαs ( JM ) = ℑMJls Φ asuα (r )
rα υα
where υα is the relative velocity,

l s
ℑMJls = ∑ ∑ (lsm m
ml = − l m s = − s
l s | lsJM )Yl , ml (θ ,φ ) χ s , m s
(lsml ms | lsJM ) are Clebsch-Gordan coefficients, Yl , ml are the spherical harmonics,

χ s, m spin function for spin s and z components ms,
s
Φαs = φaφ X
with φa and φ X being the wave function of a and X, uα (r ) is the radial wave function:
⎡ ⎛ 1 ⎞⎤ ⎡ ⎛ 1 ⎞⎤
uα (r ) = AαJM
sl exp ⎢ − i⎜ kα rα − lπ ⎟⎥ − BαJM
sl exp ⎢ + i ⎜ kα rα − lπ ⎟
⎣ ⎝ 2 ⎠⎦ ⎣ ⎝ 2 ⎠⎥⎦
and kα is the channel wave number
The scattering matrix is defined by the following equation:
BαJM′s ′l ′ = ∑ Sα
α , s ,l
J
′s ′l ′;αsl AαJM
sl
The physical interpretation of the scattering matrix is that its elements determine the
flux in each of the exit channels. For a reaction with N channels, S is an
N × N matrix.
For instance, for the n-p scattering, assuming that J = 1 with even parity, we can
have only two combinations of s and l , s = 1 and l = 0, and s = 1 and l = 2, which
give the required J value. Using the notation 2 s +1l J we have two possibilities 3S1 for l
= 0 and 3 D1 for l = 2. The same applies to the exit channel. Consequently we can
have the following channels: 3S1 →3S1 , 3S1 →3D1 , 3 D1 →3S1 , 3 D1 →3D1 . We therefore have
a 2 x 2 scattering matrix. Its elements will give the probability of flux in each of the
four respective channels, i.e. the matrix will describe how probable is each of the
four possible transitions.
The scattering matrix S can be written as
S = U −1 exp(2iΔ)U (2)
where U is an orthogonal (unitary and real) N × N matrix and Δ is a diagonal matrix,

whose elements are the eigen phase shifts.
1
The matrix U is specified by N ( N − 1) real parameters and Δ by N real eigen-phase
2
1
shifts. Together the collision matrix S is specified by N ( N + 1) real parameters.
2
For a two-channel reaction, N = 2 and the total number of independent real
parameters is 3. In this case, the matrices U and Δ have the following forms
⎡ cos ε sin ε ⎤ ⎡δ 0 ⎤
U =⎢ ⎥ and Δ = ⎢ 1 ⎥
⎣− sin ε cos ε ⎦ ⎣ 0 δ2 ⎦
Therefore
⎡cos ε − sin ε ⎤ ⎡e 2iδ1 0 ⎤ ⎡ cos ε sin ε ⎤
S = U −1 exp(2iΔ)U = ⎢ ⎢ ⎥⎢
⎣ sin ε cos ε ⎥⎦ ⎣ 0 e 2iδ 2 ⎦ ⎣− sin ε cos ε ⎥⎦
which gives
⎡ 2 iδ 1 2 iδ 2 1 ⎤
⎢(cos ε )e + (sin ε )e
2 2
(sin 2ε )(e 2iδ 1 − e 2iδ 2 ) ⎥
S=⎢ 2
1 ⎥
⎢ sin(2ε )(e 2iδ 1 − e 2iδ 2 ) (sin 2 ε )e 2iδ 1 + (cos2 ε )e 2iδ 2 ⎥
⎣ 2 ⎦
The three real parameters are the two eigen phase shifts δ1 and δ 2 and the mixing
(coupling) parameter ε. The parameter ε describes mixing between allowed
configurations. If the coupling between the two configurations is zero (ε = 0), the off
diagonal elements of the matrix S vanish and the matrix is then given by
⎡e 2iδ1 0 ⎤
S=⎢ ⎥ 2 iδ 2
⎣ 0 e ⎦
Generally for s =1 and any given J, we have two radial wave functions corresponding
to l = J − 1 and l = J + 1 :
⎧ ⎡ ⎤⎫ ⎧ ⎡ ⎤⎫
u1 (r ) = A1 exp⎨− i ⎢kr − ( J − 1)π ⎥ ⎬ − B1 exp⎨+ i ⎢kr − ( J − 1)π ⎥ ⎬
1 1
⎩ ⎣ 2 ⎦⎭ ⎩ ⎣ 2 ⎦⎭
⎧ ⎡ ⎤⎫ ⎧ ⎡ ⎤⎫
u2 (r ) = A2 exp⎨− i ⎢kr − (J + 1)π ⎥ ⎬ − B2 exp⎨+ i ⎢kr − ( J + 1)π ⎥ ⎬
1 1
⎩ ⎣ 2 ⎦⎭ ⎩ ⎣ 2 ⎦⎭
The relation between the amplitudes A1 and A2 for the incoming waves, and the
amplitudes B1 and B2 for the outgoing waves is given by:
b = Sa
where b is the column vector with components B1 and B2, a is the column vector with
components A1 and A2 and S is the 2 x 2 scattering matrix.
Explicitly,
⎡ 2 iδ 1 2 iδ 2 1 ⎤
⎡ B1 ⎤ ⎢(cos ε )e + (sin ε )e
2 2
(sin 2ε )(e 2iδ 1 − e 2iδ 2 ) ⎥ ⎡ A ⎤
2
⎢B ⎥ = ⎢ 1
1
⎥ ⎢ ⎥
⎣ 2⎦ ⎢ sin(2ε )(e 2iδ 1 − e 2iδ 2 ) (sin 2 ε )e 2iδ 1 + (cos2 ε )e 2iδ 2 ⎥ ⎣ A2 ⎦
⎣ 2 ⎦
In particular, for the example considered earlier, i.e. for J = s = 1, B1 ≡ B(l = 0) ,

B2 ≡ B(l = 2) , A1 ≡ A(l = 0) and A2 ≡ A(l = 2) .
Once the phase shifts are determined from an analysis of a certain set of
experimental data, the scattering matrix S can be used to calculate (predict) other
observables.
References
Blatt, J. M. and Biedenharn, L. C. 1952, Rev. Mod. Phys. 24:258.
Condon, E. U. and Shortley, G. H. 1935, Theory of Atomic Spectra, Cambridge
University Press, Canbridge, England.
Appendix K
Reorientation effect in Coulomb excitation
Coulomb excitation is discussed extensively by Alder et al. (1956) and Alder and
Winther (1975). Reorientation effects were first discussed by De Boer and Eichler
(1968).
First-order perturbation theory

The process of Coulomb excitation is illustrated schematically in Figure K.1.
Figure K.1. Schematic diagram of Coulomb excitation. On the left-hand side, the deformed projectile
is excited in the electric field of the spherical target nucleus. The arrows indicate the acting force.
They cause rotation as shown in the figure. On the right-hand side, one can see the corresponding
first order (direct) excitation of the projectile from state i to f and the subsequent deexcitation by
emission of γ-ray.
The differential cross section for Coulomb excitation is given by:

dσ ⎛ dσ ⎞
=⎜ ⎟ Pf
dΩ ⎝ dΩ ⎠ R
where (dσ / dΩ) R is the Rutherford scattering cross section and Pf is the probability
to excite the final state f from the initial state i .
In the first-order perturbation theory, the electromagnetic interaction between the

projectile and the target nucleus is described by the time-dependent interaction
r
potential V ( r (t )) . The excitation amplitude is given by:
∞
1 iω t r
ai → f = ∫
ih − ∞
e fi f V (r (t )) i dt
where
E f − Ei
ω fi =
h
Ei and E f are the energies of the initial and final states.
The Coulomb excitation probability is then given by:

2
Pf = ai → f
r r
To evaluate the matrix element f V ( r (t )) i , V ( r (t )) is expanded into a multipole
r
series. With the multipole expansion of the electrostatic part of V ( r (t )) , the total
Coulomb excitation cross section for the electric excitation of the order Eλ is given
by
2
⎛ Zt e ⎞ − 2λ + 2
σ Eλ = ⎜ ⎟ a0 B ( Eλ ) f Eλ (ξ )
⎝ hυ ⎠
where υ is the relative velocity at large distances, a0 is half the distance of closest
approach in a head-on collision, B(Eλ) is the reduced transition probability
associated with a radiative transition of multipole order Eλ, f Eλ (ξ ) the total cross
section function (Alder et al. 1956; Alder and Winther 1956), and ξ the adiabaticity
parameter.
π
16π 3 cos(θ / 2)
Y (π / 2,0) ∫ I λμ (θ , ξ )
3 ∑ λμ
2
f Eλ (ξ ) = dΘ
(2λ + 1) μ 0
sin 3 (θ / 2)
where Yλμ (θ , Φ ) are the normalized spherical harmonics and I λμ (θ , ξ ) are given by
∞
[cosh( w) + ε + i (ε 2 − 1)1 / 2 sinh( w)]μ
I λμ (θ , ξ ) = ∫ eiξ (ε sinh( w) + w) × dw
−∞
[ε cosh( w) + 1]λ + μ
The cross section for the excitation by the magnetic field is given by
2
⎛Z e⎞
σ Mλ = ⎜ t ⎟ a0− 2 λ + 2 B ( Mλ ) f Mλ (ξ )
⎝ hc ⎠
with
π
16π 3 2 (λ + 1) − μ
2 2
cos(θ / 2)
f Mλ (ξ ) = ∑ λ +1,μ
Y (π / 2,0 ) ∫ Iλ +1,μ (θ , ξ ) cot 2 (Θ / 2) dθ
(2λ + 1) 2 μ λ2 (2λ + 3) 0
sin 3 (θ / 2)
It might be interesting to point out that the ratio σ Mλ / σ Eλ is proportional to (c / υ ) 2 ,

which means that magnetic excitations are suppressed by this factor when
compared to the electric excitations.
The probabilities for electromagnetic transitions are defined by (Bohr and Mottelson
1969):
1
∑
2 2
B (πλ; I i → I f ) = I f M f Μ (πλμ ) I i M i = I f Μ (πλ ) I i
μMf 2Ii + 1
Symbol π stands for E or M i.e. for the electric or magnetic transitions, I and M in the
wave functions are the total angular momentum and the magnetic quantum numbers
of the initial and final states, Μ (πλμ ) the multipole operator, and I f Μ (πλ ) I i the
reduced matrix element. The quantity μ gives the angular momentum transfer along
the beam direction ( μ = M i − M f ).
In the first order perturbation theory, the cross section is directly proportional to the
reduced transition probability:
σ πλ ∝ B (πλ ; I i → I f )
Three important parameters in the Coulomb excitation are: (1) the Sommerfeld
parameter η, (2) the adiabaticity parameter ξ, and (3) the excitation strength
parameter χ.
The Sommerfeld parameter is the ratio of the half the distance of closest approach,
a0 , to the de Broglie wave length D . As long as the ratio η >1, there is no direct
interaction of the two nuclei and a semiclassical approach for the trajectory is valid.
Usually η >>1.
The adiabaticity parameterξ describes how swift is the interaction. In order to excite
a final state f from the initial state i , the collision time must be shorter or of the
same order of magnitude as the time of the internal motion of the nucleons. For a
sufficiently short time of collision, the adiabaticity parameter is small and excitations
are possible. The excitation probabilities given by the function f Eλ (ξ ) decrease with
the increasing ξ (see Figure K.2). As the adiabaticity parameter approaches the
value of 1, the probability of the system to interact becomes smaller. For ξ >1 f Eλ (ξ )
decreases exponentially with ξ .
Figure K.2. Examples of the f Eλ (ξ ) function for λ = 1, 2, 3, and 4 (Adler and Wither 1975).
The excitation strength parameter χ describes the strength of the interaction for
different multipoles. As the strength increases, higher order perturbation must be
taken into account.
The excitation strength parameter χ for a transition of the order πλ from state i to
f is defined by
16π Z t e (λ − 1)! I f Μ (πλ ) I i

χπλ =
hs (2λ + 1)!! a0λ 2 I i + 1
where s = υ for Eλ excitations and c for Mλ.
Second-order perturbation theory

The first-order perturbation theory is used to describe direct excitations. Multiple
excitations, which include also reorientation effects, are described by the second
order perturbation theory. In this theory, the probability Pf of the excitation of state
f is made of three components:
Pf = Pf(1) + Pf(1, 2) + Pf( 2)
where Pf(1) is the first order excitation probability written earlier as Pf , Pf( 2) is the
second-order term, and Pf(1, 2) is the interference term between the first- and second-
order amplitudes. A schematic diagram comparing the first- and second-order
excitation processes is presented in Figure K.3.
Figure K.3. A schematic diagram comparing the first- and second-order Coulomb excitation
processes. The left-hand side (a) of the figure shows two excitations of states f directly from the
state i . The right-hand side (b) of the figure contains two examples of the second-order processes.
It is assumed that a direct excitation of state f from state i is not allowed but state f can
excited in a two-step process involving the intermediate state z . This type of excitation is described
( 2)
by the second-order excitation probability P f . The excitation of the first excited state z , i.e. the
transition i → z , can be also followed by the reorientation transition z → z . This process is
described by the second-order excitation probability Pf(1, 2) .
The Pf( 2) term describes multiple excitations, which is analogous to multiple
excitations in particle transfer reaction or scattering as discussed in Chapter 17 and
in the Appendix E. This term becomes significant for transitions, which are either
weak or cannot take place in a single-step excitations. They involve not only the
initial and final states i and f but also an intermediate state z .
The Pf(1, 2) term describes reorientation processes. In this case the intermediate state
is identical with the excited state and transitions are between the magnetic
substates.
Explicit expressions for the excitation probabilities from states i = 0 are (Alder and
Winther 1975):
Pf(1) = ∑ χ 0( →
λ) 2
f Rλ2 (θ , ξ f 0 )
λ
∑
I +I f
Pf(1, 2 ) = (2 I z + 1)(2λ + 1)(2λ ′ + 1)(2λ ′′ + 1) (−1) 0
λλ λ ′ ′′I z
⎧λ λ ′ λ ′′⎫
I 0 ⎭ 0→ f 0→ z z → f ∑
(λ ) ( λ′) ( λ ′′ )
×⎨ ⎬χ χ χ Rλμ (θ , ξ f 0 )G( λ ′λ′′) λμ (θ , ξ z 0 , ξ fz )
*
⎩I z If μ
1
Pf( 2 ) = ∑ (2 I z + 1)(2 I z′ + 1)(2λ1 + 1)(2λ1′ + 1)(2λ2 + 1)(2λ2′ + 1)
4 λ1λ1′λ 2 λ 2′ I z I ′z k
⎧ λ1 λ2 k ⎫⎧ λ1′ λ2′ k ⎫ ( λ1 ) ( λ 2 ) ( λ1′ ) ( λ 2′ )
× (2k + 1)⎨ ⎬⎨ ⎬χ 0→ z χ z → f χ 0→ z ′ χ z ′→ f
⎩ I f I 0 I z ⎭⎩ I f I 0 I z ′ ⎭
× ∑ [ R(*λ1λ 2 ) kκ R( λ1′λ2′ ) kκ + G(*λ1λ 2 ) kκ G( λ1′λ2′ ) kκ ]
κ
where Rλ2 (θ , ξ ) , Rλμ (θ , ξ ) , R( λλ′) kκ and G( λλ′) kκ are the relative probabilities for electric
excitations, normalized orbital integrals, real part of second-order orbital integral
tensor, and imaginary part of second-order orbital integral tensor, respectively, as
defined by Alder and Winther (1975).
It is seen that the last two formulae include summation over contributing λ values
and that their contributions depend on the product χ 0 → z χ z → f of the excitation strength
parameters and on the adiabaticity parameters ξ z 0 and ξ fz , all of which involve
intermediate states z .
In a particular case of the 0 + → 2 + excitation (see Figure K.4), which is related to the
study of reorientation effects in deuteron polarization as discussed in Chapter 16,
Alder and Winther (1975) show that
2
Pf ≡ P2 = χ 0( 2→) 2 R22 (θ , ξ )[1 + χ 22→ 2c(θ , s = 1, ξ )]
where c (θ , s, ξ ) is the coefficient of interference between E2 and E2-E2 excitation.
4 π Z pe 1 1 7 Z pe A1p/ 2 E 3 / 2Q2
χ ( 2)
2→2 = 2 Μ ( E 2) 2 = 2 Q2 = 8.474
15 5 hυ a0 2
a0 90 hυ Z p Z t2 (1 + Ap / At )1/ 2
where Q2 is the quadrupole moment in e ⋅ 10 −24 cm2 and E is the bombarding energy in
MeV.
Figure K.4. The reorientation excitations involving the 2+ state.
Figure K.5. The K (θ , ξ ) function.
It is seen that χ 2→ ( 2)
2 is proportional to Q2 . Thus, measurements involving
reorientation process can be used to determine the sign of the quadrupole moment.
Using the properties of the c(θ , s, ξ ) function, the excitation probability P2 can be
written as (Alder and Winther 1975; De Boer and Eichler 1968):
P2 = P2(1) [1 + qK (θ , ξ )]
where P2(1) is the first-order excitation probability,
Ap ΔE 2 Μ ( E 2) 2
q=
Z t (1 + Ap / At )
with
7 5 1
2 Μ ( E 2) 2 = Q2 = Q2
2π 4 0.7579
and
0.5056
K (θ , ξ ) = c(θ , s = 1, ξ )
ξ
The function K (θ , ξ ) is shown in Figure K.5. The dependence of the cross section for
the excitation of 2+ state on the value of the quadrupole moment is shown in Figure
K.6.
Figure K.6. Reorientation effect in Coulomb excitation. Differential cross sections calculated using the
second order perturbation theory.
As in the case discussed in Chapter 16, the effect is weak. However, careful
measurements can be used to distinguish between prolate and oblate deformation.
Our study described in Chapter 16 shows that similar distinction between the two
types of deformations can be made by studying vector polarization of inelastically
scattered deuterons.
References
Alder, K., Bohr, A., Huus, T., Mottelson, B. and Winther, A. 1956, Rev. Mod. Phys.
28:432.
Alder K. and Winther, A. 1956, Mat. Fys. Medd. Dan. Vid. Selsk. 31, No. 1.
Alder K. and Winther, A. 1975, Electromagnetic Excitation: Theory of Coulomb
Excitation with Heavy Ions, North-Holland, Amsterdam.
Bohr, A. and Mottelson, B. R. 1969, Nuclear Structure, vol. 1, W.A. Benjamin, Inc., New
York.
De Boer, J. and Eichler, J. 1968, Advances in Nucl. Phys. 1:1.
Appendix L:
The Faddeev formalism
The Faddeev formalism describes interaction of three-body systems (Faddeev
1961a, 1961b, 1962, 1965; Schmid and Ziegelmann 1974). To understand it, I shall
start with a more familiar two-body formalism.
Two-nucleon scattering and Lippmann-Schwinger equation

The Schrödinger equation for the scattering of a particle with reduced mass μ and
energy E by a potential V can be written as
(E − K ) ψ = V ψ (1)
where
h∇ 2
K =− (1a)
2μ
is the kinetic energy operator.
This equation can be rewritten in a different form as (Lippmann and Schwinger 1950;
Satchler 1983):
ψ = φ + G0V ψ (2)
This new equation is known as the Lippmann-Schwinger equation or an integral form

of the Schrödinger equation.
In the eq. (2), ϕ is the solution of the homogenous equation
(E − K ) ϕ = 0 (3)
and
1
G0 = (4)
E − K + iε
is the free-particle Green operator, i.e. the operator for an undisturbed system. The
parameter ε is introduced to handle the problem of singularity. This is done by
calculating the limit of the relevant quantities when ε → 0.
As emphasised by Satchler (1983), even though the equation (2) looks like a solution
of the Schrödinger equation it is in fact only an alternative but more convenient form
of the equation (1). This can be seen easily by operating (E – K) on the equation (2).
Assuming that iε is infinitesimally small and thus can be neglected we have
( E − K ) ψ = ( E − K ){φ + G0V ψ }
1
(E − K ) ψ = (E − K ) φ + (E − K ) Vψ
E−K
but
(E − K ) ϕ = 0
and therefore
(E − K )ψ = V ψ
which is the original Schrödinger equation (1).

A formal solution of the Schrödinger equation can be written as (Satchler 1983):
ψ = φ + GV φ (5)
where
1
G= (6)
E − H + iε
is the Green function for the Hamiltonian H.

Using operator algebra, the Green function G for two interacting particles with
potential V can be decomposed as
G = G0 + G0VG0 (7)
We can also introduce a transition operator t for the potential scattering. We define it
as
Vψ ≡tφ (8)
It is an operator that causes a transition of the initial free state to the scattering state
by means of a potential.
By multiplying the Lippmann-Schwinger equation by V from left we can easily find
that
t = V + VG0t (9)
It is also easy to see that the t operator can be evaluated iteratively as:
t = V + VG0V + VG0VG0V + VG0VG0VG0V + ... (10)
This can be represented graphically as
In this diagram, each wavy line represents an interaction caused by potential V. In

between the lines, the particle moves freely as indicated by the free propagator G0.
The cross section is directly related to the t matrix, which is the matrix element of the
transition operator in momentum space
dσ 2
∝ p′ t ( E + iε ) p (11)
dΩ
where
p′ 2 p 2
E= =
2μ 2μ
Three-nucleon scattering and the Faddeev equations

The formal description of three-nucleon scattering resembles the description of two-
nucleon scattering even though the system is significantly more complex. Interaction
of a three-body system involves not only two-body but also three-body potentials.
Many forms of three-body potentials have been proposed but their study is difficult
because from experimental and theoretical investigation we already know that their
effects are weak when compared with the effects caused by two-body interaction.
For most observables, theoretical calculations introduce insignificant differences
when three-body forces are included. In this description of the Faddeev formalism I
will therefore assume that the interaction in the three-body system involves only two-
body potentials.
Some of the quantities used in the description of a three-body system are shown in
Figure L.1. In order to handle the Schrödinger equation for the three-body
interaction, Faddeev used the standard Jacobi system of coordinates.
Figure L.1. A three-body system showing the labelling of the three particles, the Jacobi coordinates
and some of the quantities used in respective channels to describe the system.
We have to introduce the following definitions:

r r r r r
r r r r r m j x j + mk xk r mi xi + m j x j + mk xk
ri = x j − xk ; Ri = xi − ; η= (12)
m j + mk mi + m j + mk
r r r r r
r m j kk − mk k j r (m j + mk )ki − mi (k j + kk ) r r r r
pi = ; qi = ; K = ki + k j + k k (13)
m j + mk mi + m j + mk
r r 1 1 1 1 1 1
Vi = V jk ( x j − xk ) ; = + ; = + (14)
μi m j mk M i mi m j + mk
pi2 q2 1
Hi = + i + Vi ; Gi = (15)
2 μi 2 M i E − H i + iε
r r
ri and Ri are as shown if Figure L.1
r
K – the total momentum of three particles
r
pi – the relative momentum of particles j and k
r
qi – the relative momentum of particle i with respect to the centre-of-mass of
particles j and k
Vi – the interaction potential between particles j and k
H i – the channel Hamiltonian
Gi – the channel Green function
The configuration of a bound pair j-k and a free particle i is described by the function
r r r
Φ i = φi (ri )eiqi ⋅ Ri (16)
which is the solution of
H i Φ i = Ei Φ i (17)
where
qi2
Ei = +e
2M i
is the total energy in the centre-of-mass frame and e is the binding energy of φi .
We can also define the channel wave function ψ i as a solution of the Schrödinger
equation
( H i + V j + Vk ) ψ i = E ψ i (18)
which can be rewritten as the Lippmann-Schwinger equation for the three body
system
ψ i = Φ i + Gi (V j + Vk ) ψ i (20)
and which is similar to the equation (2) for the two-body system. It can be shown
(Foldy and Tobocman 1957) that this equation does not have a unique solution.
Glöckle (1970) has shown that ψ i satisfies also the following two homogenous
equations:
ψ i = G j (Vi + Vk ) ψ i (21a)
ψ i = Gk (V j + Vi ) ψ i (21b)
The equations (20), (21a) and (21b) are called Lippmann-Schwinger triad and can be
written as
ψ i = δ il Φ i + Gl (V j + Vk ) ψ i (22)
As in the case of two-body interaction, we can also introduce a transition operator. In

analogy to the t operator for the two-body interaction (see eq. (9)), the transition
operator T for the three-body system satisfies the following relation
T = V + VG0T (23)
where V = V1 + V2 + V3
Faddeev splits eq. (23) into three equations:
3 3
T = (V1 + V2 + V3 ) + (V1 + V2 + V3 )G0T = ∑ (Vi + ViG0T ) = ∑ Ti (24)
i =1 i =1
where
Ti = Vi + G0T (25)
The eq. (25) can be expressed as
Ti = ti + tiG0 (T j +T k) (27)
where by definition
Vi
ti ≡ (28)
1 − ViG0
It is easy to see that eq. (28) is the same as eq. (9) and thus it is a two-body
transition operator in the three-body space.
Equation (27) can be written in a matrix form as
⎡T 1 ⎤ ⎡ t1 ⎤ ⎡ 0 t1 t1 ⎤ ⎡T 1 ⎤
⎢T ⎥ = ⎢t ⎥ + ⎢t 0 t ⎥G ⎢T ⎥ (29)
⎢ 2⎥ ⎢ 2⎥ ⎢ 2 2⎥ 0⎢ 2⎥
⎢⎣T 3 ⎥⎦ ⎢⎣t3 ⎥⎦ ⎢⎣t3 t3 0 ⎥⎦ ⎢⎣T 3 ⎥⎦
As for the two-body system, the T operator can be evaluated by iteration. If we

define
⎡T 1 ⎤ ⎡ t1 ⎤ ⎡ 0 t1 t1 ⎤
T ≡ ⎢T 2 ⎥ , t ≡ ⎢t2 ⎥ , and τ ≡ ⎢⎢t2 0 t2 ⎥⎥
⎢ ⎥ ⎢ ⎥
⎢⎣T 3 ⎥⎦ ⎢⎣t3 ⎥⎦ ⎢⎣t3 t3 0 ⎥⎦
then the matrix equation for T can be written as

T = t + τG0T (30)
If we now use the same iteration procedure as for the eq. (9) then we shall find that
T = t + τG0t + τG0 τG0t + τG0 τG0 τG0t + ... (31)
References
Faddeev, L. D. 1961a, Sov. Phys. JETP 12:1014.
Faddeev, L. D. 1961b, Soviet Phys. Doklady 6:384.
Faddeev, L. D. 1962, Soviet Phys. Doklady 7:600.
Faddeev, L. D. 1965, Mathematical Aspects of the Three-Body Problem in Quantum
Scattering Theory, Daniel Davey and Co., New York.
Foldy, L. L. and Tocoman, W. 1957, Phys. Rev. 105:1099.
Glöckle, W. 1970, Nucl. Phys. A141:620.
Lippmann, B. A. and Schwinger, J. 1950, Phys. Rev. 79:469.
Satchler, G. R. 1983, Direct Nuclear Reactions, Clarendon Press, Oxford.
Schmid, E. W. and Zigelmann, H. 1974, The quantum mechanical three-body problem,
Pergamon Press, Braunschweig.
Appendix M:
Resonating group theory
Resonating group theory (also known as method or model and abbreviated as RGM)
applies to an interaction of two groups of particles. RGM was first introduced by
Wheeler (1937) to describe resonant transfer of a group of electrons from one group
of atoms to another. Later, the method was applied to an interaction of clusters of
nucleons and was improved by the introduction of three useful methods: the Peierls-
Yoccoz projection method, the generating coordinate method, and the Brink alpha-
cluster model (Brink 1965 and Saito 1977).
The basic assumption of the RGM is that at small distances, the interaction between
individual nucleons belonging to two clusters of interacting particles is mainly defined
by the Pauli principle and that the dynamic effects are described by the wave
function of the relative motions of the reacting clusters.
The Schrödinger equation for the interacting groups is
Hψ = Eψ
where
A
1 A
H = ∑ Ti + ∑Vij
i =1 2 i≠ j
with Ti being the kinetic energy operator of the ith particle and Vij the interaction
potential between particle i and j.
The wave function ψ ≡ ψ ( A) for the whole system of A nucleons can be written as
ψ ( A) = ∑ ∑ A [φ
A1 + A2 = A n1 , n 2
n1 ( A1 )φn2 ( A2 ) f (n1 , n2 ) ]
where A is the antisymmetrisation operator responsible for the Pauli principle,
φn1 ( A1 ) and φn2 ( A2 ) are the wave functions for the interacting nuclei, and f (n1 , n2 ) is the
function for the relative motion.
In the algebraic version of RGM (Filippov 1989), the wave function ψ ( A) is expressed
as
ψ ( A) = ∑ ∑C
A1 + A2 = A n1 n 2 n
n1 n 2 n ( A1 A2 ) n1n2 n
where Cn1n2 n ( A1 A2 ) are the Fourier coefficients of the expansion of ψ ( A) in the many-
particle oscillator basis functions n1n2 n antisymmetric under a permutation of
nucleon coordinates.
The coefficients Cn1n2 n ( A1 A2 ) are determined by solving the following algebraic
equations:
∑∑
A1′ A2′ n1′ n 2′ n
A1 A2 ; n1n2 n H − E A1′A2′ ; n1′n2′ n′ Cn1n2 n ( A1 A2 ) = 0
Theoretical treatment may be expanded by including many-body forces and an
interaction of three clusters of nucleons. For more information about mathematical
description of the RGM and its application see for instance Aoki and Horiuchi 1982;
Hofmann (2002), Lemere, Tang, and Thompson (1976), and Thompson and Tang
(1973)
References
Aoki, K. and Horiuchi, H. 1982, Progr. Theor. Phys. 68:1658, 2028, 69:857, 1154.
Brink, D. M. 1965, The Alpha-Particle Model of Light Nuclei, in International School of
Physics “Enrico Fermi”, Course 37, Varenna, Italy, Academic Press, New York.
Filippov, G. F. 1989, Riv. Nuovo Cim. 9:1.
Hoffmann, H. M. 2002, Refined Resonating Group Model and Standard Neutron Cross
Sections, Lecture notes LNS0520001, Workshop on Nuclear Reactions Data and
Nuclear Reactors: Physics, Design and Safety, Trieste.
Lemere, M., Tang, Y. C., and Thompson, D. R. 1976, Nucl. Phys. A266:1.
Saito, S. 1977, Suppl. Progr. Theor. Phys. 62:11.
Thompson, D. R. and Tang, Y. C. 1973, Phys. Rev. C8:1649.
Wheeler, J. A. 1937, Phys. Rev. 52:1083, 1107.
Appendix N:
The R-matrix theory
The R-matrix theory has been applied successfully to nuclear reactions proceeding
via formation of a compound nucleus. In this theory, the space is divided between
internal and external regions. The internal region refers to the space inside the
nucleus. This division is possible because of the short-range of nuclear forces and
because of the Pauli principle.
Schrödinger equations are solved in both regions. To study resonances in the
compound nucleus, the internal wave function is expressed in therms of a complete
set of orthogonal resonant states, which are also solutions of appropriate
Schrödinger equations and which obey a specific boundary condition.
The matching of the internal and external regions is done by calculating and
equating logarithmic derivatives of the radial wave functions at the boundary
between the two regions. The logarithmic derivative of the internal wave function
leads to an expression containing R-function or R-matrix, depending on the
assumptions used in the calculations. The differential cross section is expressed in
terms of the collision function (or matrix). The matching of the logarithmic derivatives
for the external and internal region allows for expressing the collision function (or
matrix) in terms of the R-function (or R-matrix).
Discussion of the R-matrix theory may be found in the publication of Lane and
Thomas (1958). Excellent outline of this theory may be also found in a set of lectures
by Vogt (2004). To outline the basic ideas of R-matrix treatment, I start with the
discussion of the single-channel case.
Single-channel theory
The differential cross section is given by
2
dσ 1
= 2
dΩ 4 k
∑ (2l + 1)(1 − U l ) Pl (cosθ )
l
(1)
and integrated cross section by

dσ π
σ ≡∫ dΩ = 2 ∑ (2l + 1) 1 − U l
2
(2)
dΩ k l
where
U l= e 2iδ l (3)
is the collision functions with δ l being phase shift parameters.

Internal region
The radial wave function ϕl (r ) in the internal region is a solution of the Schrödinger
equation
h 2 d 2ϕl (r )
− + V (r )ϕl (r ) = Eϕl (r ) (4)
2m dr 2
where V (r ) is the nuclear potential.
This function can be expressed as a sum of mutually orthogonal resonant state wave
functions Χ λl , which are the solutions of the Schrödinger equation
h 2 d 2 Χ λl ( r )
− + V (r ) Χ λl (r ) = Eλl Χ λl (r ) (5)
2m dr 2
If we use radius r = al to denote the division between the internal and external
region, then the boundary condition for Χ λl (r ) is given by
⎛ dΧ (r ) ⎞
al ⎜ λl ⎟ = bl Χ λl (r = al ) (6)
⎝ dr ⎠ r = al
where bl is an arbitrary (real) boundary condition number.
The external and internal regions are joined by matching the logarithmic derivatives
of the respective radial wave functions at the nuclear surface ( r = al ) . For the
internal wave function we have39
⎛ ϕ ′(r ) ⎞ 1 + bl Rl
⎜⎜ r ⎟⎟ = (7)
⎝ ϕ (r ) ⎠ r = a l Rl
where ϕ ′(r ) ≡ dϕ (r ) / dr and Rl is the so-called R-function, which is given by
γ λ2l
Rl = ∑ (8)
λ Eλl − E
with
h2
γ λ2l = Χ λ2l (r = al ) (9)
2mal
being the reduced width. In the multi-channel theory, this function is replaced by R-
matrix.
External region
The radial wave function ϕl (r ) in the external region is a solution of the Schrödinger
equation containing Coulomb interaction
d 2ϕ (r ) ⎡ l (l + 1) ⎛ 2m ⎞⎛ Z1Z 2e 2 ⎞⎤
− ⎢ 2 + ⎜ 2 ⎟⎜⎜ − E ⎟⎟⎥ϕ (r ) = 0 (10)
⎝ h ⎠⎝ r
2
dr ⎣ r ⎠⎦
Its solution can be expressed in terms of the regular Fl and irregular Gl solutions. At
large r values, they are given asymptotically by
Fl → sin[kr − η log(2kr ) − (1 / 2)lπ + σ l ] (11a)
Gl → cos[kr − η log(2kr ) − (1 / 2)lπ + σ l ] (11b)
39
Lane and Thomas (1958) assume bl = 0 (see for instance their analogous expression (IV,1.11) for
this quantity).
where
σ l = arg[1 + l + iη ] (12)
is the Coulomb phase shift, and
Z1Z 2e 2
η= (13)
hυ
is the Coulomb parameter.
In terms of the collision matrix U l
ϕl (r ) ∝ I l − U l Ol (14)
where I l and Ol are the incoming and outgoing wave functions, respectively,
I l = (Gl − iFl )eiωl (15a)
Ol = (Gl + iFl )e − iωl (15b)

with
l
ωl = ∑ tan(η / n) (16)
n =1
By calculating the logarithmic derivatives of the external radial wave function and
by matching them with the logarithmic derivatives of the internal functions we can
express the collision function U l in terms of the R-function:
I l (1 − Rl L*l )
Ul = (17)
Ol (1 − Rl Ll )
where Ll contains the logarithmic derivative of Ol
Ol′
Ll ≡ − bl (18)
Ol
The quantity Ll can be also expressed in terms of the penetration factor Pl and shift
functions Sl :
Ll = Sl + iPl − bl (19)
where
kr
Pl = (20)
Fl + Gl2
2
and
Fl′F + Gl′G
Sl = (21)
Fl 2 + Gl2
We can also introduce the scattering phase shifts Ωl ,
Ωl = ωl − tan( Fl / Gl ) (22)
which enter into the relation
Il
= e 2 iΩ l (22)
Ol
The multi-channel R-matrix theory

The simple case described above gives an introduction into the basic ideas of the R-
matrix theory. We can now consider a more complex case where various channels
are involved in the reaction. Following Vogt (2004) this is illustrated in Figure O.1 for
the 8Be as a compound nucleus.
Figure O.1. A schematic diagram of a few contributing channels leading to or resulting from the
formation of the compound nucleus 8Be*. The internal region is limited by the potential V (r ) . The
thick lines at the surface mark the surface S for each channel. (Vogt 2004.)
A channel is defined by quantum numbers

c ≡ (α , l , s, J , M J )
where
sˆ = Iˆ + iˆ
Jˆ = lˆ + sˆ
with Iˆ and iˆ being the intrinsic spins of the two particles in each channel.
Internal region
As before, the internal wave function ψ corresponding to energy E can be
expressed as a sum of the resonant states Χ λ corresponding to energy Eλ ,
ψ = ∑ Cλ Χ λ (24)
λ
satisfying the boundary condition
⎡ dΧ λ (rc ) ⎤
⎢rc ⎥ = bc Χ λ (rc = ac ) (25)
⎣ drc ⎦ r = a
c c
which may be compared with eqn (6).

It can be shown that coefficients Cλ are given by
h2 ∑ γ λ (r ϕ ′ − b ϕ )
c c c c c
Cλ = c
(26)
2mc ac Eλ − E
where γ λc is the reduced width amplitude.
This leads to an expression containing R-matrix
⎛ h2 ⎞ ⎛ h2 ⎞
⎜⎜ ⎟⎟ϕc = ∑ ⎜⎜ ⎟⎟Rcc ′ [rc ′ϕc′′ − bcϕc ′ ] (27)
⎝ 2mc ac ⎠ c ′ ⎝ 2mc ′ ac ′ ⎠
with
γ λc γ λ c ′
Rcc ′ = ∑
λ Eλ − E
which resembles eqn (8). (28)
External region
The external wave function may be expressed as
ψ = ∑ψ cϕc (29)
c
where ψ c are the wave functions containing spin-dependent components and

functions describing internal excitations of the interacting pair in a given channel,
and
1/ 2
⎛1⎞
ϕc = ⎜⎜ ⎟⎟ ( Ac I c − BcOc ) (30)
⎝ υc ⎠
with Ac and Bc being arbitrary coefficients and I c and Oc the incoming and outgoing
functions in a given channel c.
If we define the collision matrix as
Bc ≡ ∑U cc ′ Ac ′ (31)
c′
then
1/ 2
⎛1⎞ ⎡ ⎛ ⎞ ⎤
ϕc = ⎜⎜ ⎟⎟ ⎢ Ac I c − ⎜ ∑U cc ′ Ac ′ ⎟Oc ⎥ (32)
⎝ υc ⎠ ⎣ ⎝ c′ ⎠ ⎦
If we follow the same prescription as in the single-channel theory, i.e. if we match
the logarithmic derivatives for the internal and external functions, we shall be able
to express the collision matrix in therms of the R-matrix:
U cc& = (kc ac )1 / 2 Oc−1 ∑ [1 − RL]c−c1′′ [δ c ′′c ′ − Rc ′′c ′ L*c ′ ]I c ′ (kc ′ ac ′ ) −1 / 2 (33)
c ′′
This formula is analogous to the eqn (17).

Likewise, we shall find that the differential cross section and the angle-integrated
cross sections can be expressed by equations resembling the formulae (1) and (2),
respectively.
The differential cross section is given by:
dσ α ′s ′;αs 1 ∞
dΩ
= 2 ∑ BL (α ′s′ : αs ) PL (cos θ )
kα ( 2 s + 1) L = 0
(34)
where
( −) s ′ − s
BL (α ′s′ : αs ) =
4
∑i l 1−l 2 − L
Z (l1 J1l2 J 2′ sL)i l ′1− l 2′ − L ′ Z (l1′J1l2′ J 2′ s′L)
J 1 J 2 l1l 2 l1′l&2 (35)
[( 1 1
)(
× Re δ αα ′δ l1l1′ δ ss ′ − U α ′s ′l ′ ;αsl J1 δα ′α δ l 2 l 2′ δ s ′s − U α ′s ′l ′ ;αsl J 2
*
2 2
)]
with Z coefficients being the products of Clebsch-Gordan coefficients.
The angle-integrated cross section can be expressed as:
4π
σ α ′s ′;αs = B0 (α ′s′;αs)
(2s + 1)kα2
(36)
π J + s′
∞ J +s 2
=
(2s + 1)kα2
∑∑ ∑
J =0 l = J − s l ′= J − s′
(2 J + 1) δα ′α δ l ′lδ s ′s − U α ′s ′l ′;αsl J
The total cross section has the form:
∑ (2 J + 1)2 Re[1 − Uα ]
π ∞
σ T (αl ) = ′s′l ′;αsl J
(37)
(2 I + 1)(2i + 1)kα2 J =0
References
Lane, A. M. and Thomas, R. G. 1958, Rev. Mod. Phys. 30:257.
Vogt, E. 2004, R-Matrix Theory, Lecture notes, Joint Institute for Nuclear Astrophysics,
Notre Dam University, South Bend, Indiana.

RonNielsen NuclearReactions

Uploaded by

Document Information

Original Title

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

RonNielsen NuclearReactions

Uploaded by

Copyright:

Available Formats

.

Environmental Futures Centre, Griffith School of Environmental Science,

The following summary is arranged in approximately chronological order. It starts

We can see that

R ≡ σ (θ ′,φ = 0) = σ 0 (θ ′)(1 + P1P2 )

Results and discussion

Due to the unsatisfactory energy resolution of our polarization analyser we were

Reaction angle Neutron polarization Experimental error

Figure 2.2. The experimental arrangement.

Results and discussion

Plots of the measured differential cross-sections divided by the Rutherford cross-

Comparing with other data

The impact parameter

q 2 = ki2 + k 2f − 2ki k f cosθ

A graph of the universal function [J1( x ) / x ] is shown in the Appendix B.

In order to calculate theoretical locations of the positions of maxima in the angular

by fitting polynomials to points around the maxima of this function.

Derivative 0.00 5.14 8.42 11.62 14.80 18.00

The convenient expression for k is:

Alternatively, using Elab

Figure 2.12 shows the calculated angle-mass dependence of the positions of

Summary and conclusions

Figure 3.1. The core-excited model

The signatures of core-excited states are as follows:

In order to investigate the possibility of the existence of core-excited states I first

the core-excited model

Nucleus Ex (MeV) Spin Nucleus Ex (MeV) Spin

Figure 3.5. Schematic diagram of the detection system.

Beam alignment and angular scale

Nucleus Ex Ιπ σmax Μ δ2 β2 hω2 / 2C2

The plane wave (PWBA) theory

nuclear radius R = r0 A1 / 3 and a is the surface diffuseness parameter, and g (r ) is its

More appropriately, the nuclear radius is written as

Formally, all this says is that

Rotational model for 27Al

eZR02 β (1 + 0.36 β + ...)

Core-excited model of 27Al

Figure 3.12. The core-excited model of 27Al.

The spins of core-excited states

Summary and conclusions

Experimental procedure and results

Figure 4.1. The low-lying states in 40Ca and 41

Set V r0 a WD r0′ a' Vs χ2

which includes the spin-orbit interaction.

Experimental arrangement and results

Figure 5.1. Schematic diagram of the experimental setup.

where VC (r ) is the Coulomb potential generated by a uniformly charged sphere,

The distorter wave analysis

E Set 4 Set 5 Set 6 Average a)

b) The D02 coefficients and the corresponding spectroscopic factors Sexp (l , j )

c) The D02 coefficients and the corresponding spectroscopic factors Sexp (l , j )

Discussion of the experimental results

Optical model analysis

Index k is S for the volume (Saxon-Woods) absorption or D for the surface

The discrete radius ambiguity

Summary and conclusions

Theoretical calculations for 27Al(3He,α)26Al angular distributions are compared with

Orbit nlj′ nlj V j2

Experimental procedure and results

Distorted wave analysis

2p3/2 0.63 2.52 4.00 0.37 1.48 2.00 0.14

Summary and conclusions

(ii) Thomas (T):