BETA-DECAY STUDIES OF NEUTRON-RICH NUCLIDES

AND THE POSSIBILITY OF AN N = 34 SUBSHELL CLOSURE

By

Sean Nicholas Liddick

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Chemistry

2004
 ABSTRACT

BETA-DECAY STUDIES OF NEUTRON-RICH NUCLIDES

AND THE POSSIBILITY OF AN N = 34 SUBSHELL CLOSURE
By
Sean Nicholas Liddick

An inspection of periodic trends in ionization energies is one manifestation of

electronic shells in atoms. Analogously, trends observed in nuclear masses led to the

development of the nuclear shell model. Areas of increased stability at the traditional
magic nucleon numbers 2, 8, 20, 28, 50, 82, and 126 have been associated with large

energy gaps encountered in the filling of shell-model orbitals. The reordering of levels
within the shell model is an important question in nuclear structure studies as it may
lead to the development of new magic numbers far from stability.

The experimental characterization of the N = 32 subshell closure in Cr, Ti, and

Ca isotopes stimulated shell model calculations, using a new effective interaction

labeled GXPF1 [1], that also indicated the possibility of an N = 34 subshell closure
in the 22 Ti and 20 Ca isotopes. The N = 32 and suspected N = 34 subshell closures

were attributed to a strong proton-neutron monopole migration of the νf5/2 state as

protons are removed from the πf7/2 level. One indication for the subshell closure at
N = 34, predicted by GXPF1 calculations, is an energy of ∼1500 keV for the the first

excited 2+
1 state in
56
Ti, similar to the energy of the 2+
1 state in
54
Ti at the N = 32
subshell closure.

β decay studies have been performed at the National Superconducting Cyclotron

Laboratory on neutron-rich pf-shell nuclides to investigate the possibility of a subshell

closure at N = 34. The nuclides that were studied included 56 Sc, 57 Ti, 58,59 V, and 60 Cr
86
and were produced through the fragmentation of a 140 MeV/nucleon Kr beam on
a 9 Be target. The nuclides were implanted into a Double-sided Si Strip Detector

which was used to detect both implanted ions and decays. Decays were correlated
with implanted ions on an event-by-event basis and γ rays were monitored with 12
detectors from the Segmented Germanium Array. For all nuclides studied, half-lives

and level schemes were deduced.

56
The experimental study focused on the β decay of Sc, to determine the energy

of the 2+
1 state in
56
22 Ti34 . A value of 1129 keV was determined for the energy of the
2+ → 0+ transition in 56
Ti, 400 keV lower than predicted, suggesting the absence
of a shell closure in the 22 Ti isotopes. The ground state Jπ assignments for 56
Sc and
57
Ti along with other nuclides around the N = 32 subshell closure were used to infer
the migration of the neutron f5/2 state with the removal of protons from the f7/2

level using an extreme single-particle model. While the migration of the f5/2 agrees
qualitatively with calculations, the increase in energy of the νf5/2 with the removal

of protons from the f7/2 seems to be overestimated in GXPF1. However, there still
exists the possibility that with the removal of the last two protons from the f7/2 state,
the N = 34 subshell closure develops.
 ACKNOWLEDGMENTS

First, I would like to thank my advisor Paul. Without his assistance this disser-
tation would not have been possible.

I would also like to thank all the members of the beta group for their willingness
to help run the beta decay experiment that this dissertation is based on and for the
numerous discussions on data analysis that followed.

My fellow graduate students at the lab were also invaluable during my time at
the lab. I would be remiss if I did not specifically mention both Bryan and Jeremey.

Both of them listened to the numerous problems that were encountered along the way
to graduation and both have helped put things into perspective when things seemed

overwhelming. Jeremey, I’m sorry to say that there is no personal anecdote anywhere
in this dissertation. Thank you Bryan for always giving me an audience practice talks.
I’m sure that after listening to my Sc-56 talk more than a dozen times you might be

able to write about the beta decay of Sc-56.

Finally I wouldn’t have been able to make it had it not been for the constant sup-

port from my wife (who is extremely excited about being mentioned in this document
and is now jumping up and down) and family. They always had encouraging words,

especially when it seemed that the dissertation was always just a few days from being
finished.

iv
Contents

1 Introduction 1
1.1 Description of Shell Closures . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Reordering of Single Particle States . . . . . . . . . . . . . . . . . . . 5
1.3 Observation of Shell Closures . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 E(2+1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Proposed Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2 Technique 14
2.1 β decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.1.1 β-delayed neutron decay . . . . . . . . . . . . . . . . . . . . . 19
2.2 γ decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.1 Internal Conversion . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3 Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3 Experimental Setup 24
3.1 Isotope Production and Delivery for Experiment NSCL-02004 . . . . 24
3.2 Detector Setup for Experiment NSCL-02004 . . . . . . . . . . . . . . 26
3.3 Beta Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.3.1 Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.3.2 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3.3 Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.4 γ-ray Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4.1 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

4 Experimental Results 50
4.1 56 Sc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.2 58 V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.3 57 Ti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.3.1 57 V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.4 59 V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.4.1 59 Cr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.5 60 Cr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

5 Discussion 93
5.1 Comparison with Shell Model Results . . . . . . . . . . . . . . . . . . 97

v
 5.2 Monopole Migration of νf5/2 Based on Ground-State Spin and Parity
Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
5.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

6 Conclusions and Outlook 107

Bibliography 109

vi
List of Figures

1.1 Atomic and Nuclear Shell Closures . . . . . . . . . . . . . . . . . . . 3

1.2 Nuclear shell structure . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Migration of the 7/2+ state . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 E(2+ ) in Ca isotopes . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 Chart of the nuclides . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.6 Low-energy level structure of N = 29 isotones. . . . . . . . . . . . . . 10
1.7 E(2+ ) for the 24 Cr, 22 Ti, and 20 Ca isotopes. . . . . . . . . . . . . . . . 12

3.1 Layout of the NSCL . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.2 Experiment 02004 particle ID spectrum . . . . . . . . . . . . . . . . . 27
3.3 Schematic of the β counting system . . . . . . . . . . . . . . . . . . . 29
3.4 Experiment 02004 DSSD electronics . . . . . . . . . . . . . . . . . . . 31
3.5 Experiment 02004 PIN electronics . . . . . . . . . . . . . . . . . . . . 33
3.6 DSSD 228 Th calibration . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.7 DSSD 90 Sr calibration . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.8 Distribution of implanted ions across DSSD . . . . . . . . . . . . . . 37
3.9 Implantation multiplicity in the DSSD . . . . . . . . . . . . . . . . . 38
3.10 Decay condition in software . . . . . . . . . . . . . . . . . . . . . . . 38
3.11 β detection efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.12 Correlation in strip number between DSSD and SSSD1 and SSSD2 . 42
3.13 Experiment 02004 SeGA electronics . . . . . . . . . . . . . . . . . . . 43
3.14 Schematic arrangement of Ge detectors . . . . . . . . . . . . . . . . . 44
3.15 Residual plot for first Ge calibration . . . . . . . . . . . . . . . . . . 45
3.16 Residual plots for the individual SeGA detectors . . . . . . . . . . . . 47
3.17 Residual plot for the second Ge calibration . . . . . . . . . . . . . . . 48
3.18 Ge calibration between 900-1100 keV . . . . . . . . . . . . . . . . . . 48
3.19 Detection efficiency for Ge detector array . . . . . . . . . . . . . . . . 49

4.1 Sample 58 V γγ and time-γ matrices . . . . . . . . . . . . . . . . . . . 51

56
4.2 Sc β-delayed γ-ray spectrum . . . . . . . . . . . . . . . . . . . . . . 51
56
4.3 Sc half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
56
4.4 Sc isomeric γ-ray spectrum . . . . . . . . . . . . . . . . . . . . . . . 54
56
4.5 Sc level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
56
4.6 Sc γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . 56
58
4.7 V β-delayed γ-ray spectrum . . . . . . . . . . . . . . . . . . . . . . 58
58
4.8 V βγ spectra with cuts on the correlation time . . . . . . . . . . . . 59
58
4.9 V half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

vii
 58
4.10 V γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . . 61
58
4.11 V level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
57
4.12 Ti β-delayed γ-ray spectrum . . . . . . . . . . . . . . . . . . . . . . 63
57
4.13 Ti βγ spectra with cuts on the correlation time . . . . . . . . . . . 64
57
4.14 Ti half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
57
4.15 Ti half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
57
4.16 Ti γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . 68
57
4.17 V level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
57
4.18 V γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . . 71
59
4.19 V β-delayed γ-ray spectrum . . . . . . . . . . . . . . . . . . . . . . 72
59
4.20 V βγ spectra with cuts on the correlation time . . . . . . . . . . . . 74
59
4.21 V γ-gated half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
59
4.22 V γ-gated half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
59
4.23 V γ-gated half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
59
4.24 V γ-gated half-lives . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
59
4.25 Cr half-life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
59
4.26 V half-life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
59
4.27 V γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . . 83
59
4.28 V γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . . 84
59
4.29 V γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . . 85
59
4.30 Cr level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
59
4.31 Cr γγ coincidence spectra . . . . . . . . . . . . . . . . . . . . . . . 86
59
4.32 Mn level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
60
4.33 Cr β-delayed γ-ray spectrum . . . . . . . . . . . . . . . . . . . . . . 87
60
4.34 Cr βγ spectra with gates on correlation time . . . . . . . . . . . . . 88
59
4.35 Cr decay curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
60
4.36 Cr γγ coincidences . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
60
4.37 Cr level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
60
4.38 Fe level scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

5.1 Spin-orbit splitting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

5.2 Proton-neutron monopole interaction . . . . . . . . . . . . . . . . . . 95
5.3 E(2+ ) for the 24 Cr, 22 Ti, and 20 Ca isotopes. . . . . . . . . . . . . . . . 97
5.4 E(2+ ) for the N = 32 isotones . . . . . . . . . . . . . . . . . . . . . . 98
5.5 Low-energy level structure of 54,56 Sc from shell model calculations using
the GXPF1 interaction. . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.6 Experimental and Theoretical decay scheme for 57 Ti . . . . . . . . . . 102
5.7 Jπ of nuclear ground states near N=32,34 . . . . . . . . . . . . . . . . 103

viii
List of Tables

2.1 β decay selection rules . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.2 γ transition rates and Weisskopf estimates . . . . . . . . . . . . . . . 21

3.1 List of γ-ray energies in keV for the two Ge energy calibrations. . . . 45
56
4.1 Sc β-delayed γ rays. . . . . . . . . . . . . . . . . . . . . . . . . . . 52
58
4.2 V β delayed γ rays . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
57
4.3 Ti β-delayed γ-rays. . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.4 Ratio of 113- and 175- to the 1579-keV transition area . . . . . . . . 70
4.5 Calculated 57 Ti internal conversion coefficients from Ref. [2]. . . . . . 70
59
4.6 V β-delayed γ-rays. . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.7 Increase in intensity of 59 V daughter and grand-daughter γ rays . . . 79
60
4.8 Cr β-delayed γ-rays . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

ix
Chapter 1

Introduction

1.1 Description of Shell Closures

Discontinuities observed in the ionization energies of atomic electrons as a function

of Z, shown in Figure 1.1, have been linked to the filling of electronic shells. After the
filling of an electronic shell, the next valence electron is placed into a higher energy

orbital decreasing the energy needed to remove the last electron from the atom. Thus,
discontinuities in the ionization energies of atoms were correlated with the underlying
“shell structure”. In studies of the bulk properties of nuclei near the valley of stability,

discontinuities in a number of different physical observables involving nuclear masses

also suggested an extra stability of the nuclear system at ’magic’ nucleon numbers of

2,8,20,28,50,82, and 126. For example, consider the one-neutron separation energies
defined by

Sn (A, Z) = BE(A, Z) − BE(A − 1, Z) (1.1)

where BE is the binding energy of a nucleus, A is the mass number and Z is the atomic
number of the nuclide. From equation 1.1 we associate Sn with the energy necessary

to remove a neutron from the nucleus. Similar to the atomic case, just after a magic
number for an isotopic chain, a sudden drop in Sn is observed. The effect of the magic
numbers on the neutron separation energy can be emphasized by taking differences

1
between one-neutron separation energies between neighboring isotopes, defined as

∆Sn = BE(A, Z) − BE(A − 1, Z) − [BE(A + 1, Z) − BE(A, Z)] (1.2)

where again BE is the binding energy, A and Z represent the mass number and atomic
number of the nuclide, respectively. ∆Sn is shown in Figure 1.1 for all even-even

nuclides up to 94 Pu.

Theoretical attempts to reproduce the observed magic numbers gradually led to

the development of the nuclear shell model. In contrast to the atomic case, where the
electronic shell structure arises from the well known Coulomb potential created by the

protons, the potential for the description of nucleons inside the nucleus was poorly
understood and various models were tested in an attempt to explain the behavior.
A few simple potentials were tried for the nuclear shell model, such as the harmonic

oscillator and the square well, but these potentials were not realistic descriptions of
the expected nuclear potential especially since both require the potential go to infinity

at large radii. In spite of the infinite energy of the harmonic oscillator and square well
potentials, both were able to reproduce the shell closures up to 20 and showed that

the closures are associated with a large energy spacing between two different single-
particle orbits. A more realistic potential approaches zero far from the nucleus, a
condition fulfilled by a Woods-Saxon potential of the form:

−V0
V (r) = (1.3)
1 + e(r−R0 )/a

where a is the diffuseness of the nuclear surface, R0 is the mean nuclear radius, r
is the distance from the center of the nucleus, and V0 is the depth of the potential.

While the Woods-Saxon potential is more realistic (V→0 at large r) it was still unable
to reproduce the magic numbers observed above 20. The problem was finally solved

with the introduction of a spin-orbit force that breaks the degeneracy between the
pairs of states with l > 0 [3, 4]. The spin-orbit force results in the reduction in the

2
 30
Ionization energy (keV)

2p
{
20 3p
4p
{
4d 5p

{
5d 6p
3d

{
{

{
{
{ 4f
10

{
Li K
Na Rb Cs Fr
0
0 25 50 75 100
Z
8
20
28 50 82
6
Dsn (MeV)

126
4

0
0 50 100 150
N
Figure 1.1: In the top of the figure, the ionization energy is plotted as a function of the
atomic number. Large drops in the ionization energy occur following the filling of an
atomic shell. The bottom part of the figure shows the differential neutron separation
energies as a function of neutron number. Lines connect series of isotopes. The peaks
located at the magic numbers suggest additional stability that can be associated with
the filling of nuclear shells.

3
 50
g9/2
p1/2
f5/2
p3/2
28
f7/2
20
d3/2
s1/2
d5/2
8
p1/2
p3/2

2
s1/2

Figure 1.2: Nuclear shell structure up to nucleon number 50 for a Woods-Saxon po-
tential with a spin-orbit interaction. Energies are not to scale and states are labeled
by l and j.

energy of the l+1/2 state and reproduces the known shell closures. The shell structure
described by the Woods-Saxon model with a spin-orbit force is shown in Figure 1.2
up to the nucleon number 50 shell closure [5]. For example, from Figure 1.2 the magic

numbers at 28 and 50 are a result of the spin-orbit splitting between the f7/2 − f5/2
and g9/2 − g7/2 , respectively.

Using the known shell closures, shell model calculations can be performed by di-
viding the nucleus into an inert core, containing most of the nucleons coupled to 0+ ,

and a small number of valence particles outside the core. An effective shell model
interaction is developed for the valence nucleons to describe the nuclei in a specific
region. In this manner, the nuclear shell model, schematically displayed in Figure

1.2, has proven to be very successful in the reproduction and prediction of nuclear
properties close to the valley of β stability. It should not be assumed, however, that

the relative energies between levels shown in Figure 1.2 are static. The possible re-
arrangement of single-particle levels is of great interest in shell structure studies and

can result in the development of new magic numbers.

4
1.2 Reordering of Single Particle States

A significant reordering of single-particle states has been observed in nuclei close

to stability in the N = 50-82 shell. The reordering involves the migration of the
proton g7/2 orbit in the Sb isotopes inferred from the shift in energy of the 7/2+
state as neutrons are added to fill the neutron h11/2 , see Figure 1.3 [6]. Notice that

when neutrons are added to the h11/2 state (beginning around N = 66) the energy
of the g7/2 state decreases and is reflected in the energy of the 7/2+ of Sb, which
123
eventually becomes the ground state in 51 Sb72 . The reduction in the energy of the
g7/2 single particle level has been attributed to an attractive proton-neutron monopole

interaction, between the πg9/2 − νh11/2 states, reducing the energy of the g7/2 level
by approximately 2 MeV across the Sb isotopes. This is an additional interaction
on top of the spin-orbit interaction discussed above. While the monopole interaction

is significant between the πg9/2 − νh11/2 levels, as evidenced in the Sb isotopes, it

is strongest when proton and neutron orbitals have similar values of orbital angular

momentum, l(proton) ∼ l(neutron) [7].

1.3 Observation of Shell Closures

In the single-particle picture shown in Figure 1.2, shell closures occur where there is
a large energy difference between consecutive single-particle states. At a shell closure,

a nucleus may be considered a spherical system with limited collective features at

low excitation energies. The low-energy level structure of nuclei at shell closures can
usually be described using single-particle excitations. However, as nuclei progress

away from a shell closure, collective interactions (involving many nucleons at the
same time) become more likely. Collective behavior in the proximity of shell closures,

is associated with the vibration of the nuclear shape. At midshell, the highly collective
system displays features that can be described as the rotation of a statically deformed

nucleus. The increase in collectivity leads to the evolution of the nuclear shape from

5
 1830
h11/2
1605
1558
h11/2 1501
1484
1427 h11/2
h11/2 1419 1366 h
1348
h11/2 1300
1322 h11/2 11/2
h11/2 h h11/2
h11/2 11/2

832 861 815

769
g7/2 g7/2 g 724
7/2
g7/2
g7/2
527
g7/2

270
g7/2 g7/2
d5/2
0 0 0 0 0 0 0 0 0 0 37 0 0 0 0 0 0
d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2 d5/2
-160
g7/2
-332
g7/2
-491
g7/2 -645
g7/2
-798
g7/2 -963

g7/2

50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82
Neutron Number
Figure 1.3: Migration of the 7/2+ state in Sb isotopes as a function of neutron number.
Energies of the different states relative to the d5/2 level were taken from experimental
data. The migration is attributed to the proton-neutron monopole interaction between
πg7/2 and νh11/2 . As neutrons are added to the h11/2 , beginning around N = 66, there
is a steady drop in the energy of the proton g7/2 state.

6
spherical to a static deformation, resulting in either oblate (discus-like) or prolate

(football-like) shapes. The static deformation can be characterized by the quadrupole

deformation parameter, β2 . The sign of β2 describes the shape of the nucleus with

positive values associated with prolate shapes and negative values associated with
oblate shapes. There exist a variety of different probes to measure the quadrupole
deformation of the nucleus. For the current study the energy of the first excited 2+

state, E(2+ +
1 ), in even-even nuclei will be used and, qualitatively, a higher E(21 ) values

for more rigid spherical nuclei is expected.

1.3.1 E(2+
1)

The experimental signature for the possible presence of a shell closure used herein
will be the energy of the first excited 2+ state in even-even nuclei, E(2+
1 ). Globally,

the E(2+
1 ) is a good indication of shell structure when compared systematically across

a range of isotopes. The behavior of an E(2+

1 ) value can be related to the degree of

collectivity in the nuclear system in a relatively simple way . At shell closures, the
nucleons completely fill the single-particle levels below a large energy gap. To create

a 2+ state, a pair of nucleons must be broken and one nucleon promoted to a higher
single-particle state across the energy gap due to the Pauli principle. This costs a
considerable amount of energy, resulting in a high E(2+
1 ) value. Further from a shell

closure, the configuration of the nucleons is more complicated with unfilled orbitals
and the energy of the 2+ state is lowered as a result of collective interactions among

many nucleons and can be describes in terms of a vibration or, if a static deformation
exists, a rotation of the nucleus.

Grodzins [8] developed the following empirical relation between the quadrupole
deformation parameter, β2 , and E(2+ )

1225M eV
E(2+ ) = (1.4)
β22 A7/3

7
 4000

3000

Energy (keV)
2000

1000

0
20 24 28 32
Neutron Number
Figure 1.4: E(2+ ) values in the 20 Ca isotopes as a function of neutron number. The N
= 28 shell closure and N = 32 subshell closure are revealed from a peak in the E(2+ )
values.

where A is the mass number of the nuclide. As a shell closure is approached the
nuclear shape becomes more spherical, β2 decreases, and the E(2+
1 ) increases. As a

progression is made to the nuclei in the middle of a major shell collectivity increases

and leads to a reduction in the E(2+

1 ).

The E(2+
1 ) values for even-even 20 Ca isotopes are shown in Figure 1.4 as a spe-

cific example. Indications of the shell closures at N = 20,28 are observed in Figure
1.4 as peaks in the E(2+
1 ) value. Also, in the neutron-rich Ca isotopes, there might

be a smaller peak located at N = 32 [9]. The reduced height of the peak at N =

32 can be attributed to a subshell closure in which the energy separation between
two consecutive single-particle orbits is significant but not as large as a shell closure.

The existence of the N = 32 subshell closure, and the possibility of additional sub-
shell closures in neighboring neutron-rich nuclei, is an interesting question since the

development of a subshell closure would inhibit the onset of collectivity.

1.4 Proposed Measurement

The present study concentrated on the region of the chart of the nuclides shown in
Figure 1.5. The region is centered above the Z = 20 proton closed shell and the N =

8
 Proton Number
24 Cr56 Cr57 Cr58 Cr59 Cr60 Cr61 Cr62 Cr63 Cr64 Cr65 42
1.4E+17 y
V55 V56 V57 V58 V59 V60 V61 V62 V63

pf7/2
22 Ti54 Ti55 Ti56 Ti57 Ti58 Ti59 Ti60 Ti61 40
Sc53 Sc54 Sc55 Sc56 Sc57 Sc58 Sc59

20 Ca52 Ca53 Ca54 Ca55 Ca56 Ca57 38

32 34 36
npf

Neutron Number

Decay Q-value Range

Q(b-)>0
Q(b-)-S N>0
Naturally Abundant
Figure 1.5: Chart of the nuclides focusing on the region around Z = 20 and N = 32.
Nuclides available for study in this experiment are colored white and outlined with
the dashed gray line. The five nuclides have a decay Q value that is greater than the
neutron separation energy. Limited spectroscopic information was available for the
nuclides in this region before the study presented here.

56 56
32 neutron closed subshell. The β decay of Sc to levels in Ti was investigated to
determine E(2+
1 ) in
56
Ti at N = 34 and the possibility of a neutron subshell closure.
57 58,59 60
In addition, new data on the β decays of Ti, V, and Cr were obtained.

As mentioned in Section 1.2, the migration of the πg7/2 state in the Sb isotopes
was due to the monopole interaction between the πg7/2 and νh11/2 states, which have

similar l values. The strongest proton-neutron monopole interaction occurs between

nucleons in orbitals that are spin-orbit partners and may lead to a significant reorder-

ing of single-particle states far from stability. Due to the large spin-orbit splitting,
this overlap is strongest in neutron-rich nuclei between the π(l + 1/2) and ν(l − 1/2)
single-particle states that lowers the energy of the ν(l − 1/2) level, possibly leading

9
 4
N=29 isotones

Excitation Energy (MeV) 3

34
2

1 32

0
SM Exp Ti
22 24
Cr 26
Fe Exp SM

20
Ca Ni
28

Figure 1.6: Low-energy level structure of N = 29 isotones. The energy of the 5/2+
state drops dramatically with the filling of the πf7/2 level and is attributed to the
proton-neutron monopole interaction between the πf7/2 and νf5/2 states.

to a dramatic reordering of single-particle states [10]. In the pf shell, the spin-orbit

parnters πf7/2 −νf5/2 are ideal candidates for the development of novel shell structure

far from stability. The low-energy structure of neutron-rich nuclei with nucleons occu-
pying the proton f7/2 state and neutron f7/2 , p3/2 , f5/2 , p1/2 orbitals, shown in Figure
1.6, is dramatically affected by the strong and attractive proton-neutron monopole

interaction between the spin-orbit partners πf7/2 and νf5/2 . As protons are added to
the πf7/2 orbital, going from 20 Ca to 28 Ni, the effective single-particle energy of the

νf5/2 orbital decreases significantly relative to the νp3/2 state. The lowering in energy
of the νf5/2 orbital, in combination with the large spin-orbit splitting between the

νp3/2 and νp1/2 levels, is suggestive of a subshell gap at N = 32 for nuclides in which
the πf7/2 orbital is occupied by four or fewer protons.
In addition to the E(2+ ) value for 52
20 Ca, discussed in the previous section, the

masses [11] of the Ca isotopes suggested added stability and provided early evidence
for a subshell closure at N = 32. However, systematic data for isotopes beyond

10
52
Ca are unavailable, and it was believed that the appearance of a subshell gap at

N = 32 was reinforced by the proton shell closure at Z = 20. Experimental evidence

for a potential subshell closure at N = 32 for the 24 Cr isotopes was reported in

Ref. [12], based on the systematic variation of E(2+

1 ) in the even-even Cr isotopes.

The value E(2+

1 ) = 1007 keV of
56
Cr32 is more than 100 keV above that observed in
54,58
the neighboring Cr nuclei.

The systematic variation of E(2+

1 ) along the N = 32 isotonic chain was recently

completed for the πf7/2 nuclides with the determination of E(2+

1 ) = 1495 keV for

54
Ti [13]. The increase in E(2+
1 ) from 24 Cr to 20 Ca with removal of protons from
the f7/2 orbital, parallels the expected growth of the single-particle energy gap at

N = 32 due to the monopole migration of the νf5/2 orbital (see Figure 1.6). The data
50,52,54
available for the higher-spin states in even-even Ti isotopes provide additional
evidence for the N = 32 gap [13]. In 54 Ti, the significant spacing between the 6+
1 state

at 2936 keV and the cluster of levels 8+ + +

1 , 91 , and 101 at ∼ 5800 keV suggests that a

substantial energy gap must be overcome when promoting one of the coupled νp3/2

neutrons to either the νp1/2 or the νf5/2 level.

It was predicted [14] that the continued monopole shift of the νf5/2 orbital may

also lead to the development of a shell closure at N = 34, and that the low-energy
54 48
structure of Ca may be similar to that of Ca, which is doubly-magic. The shell
model calculations by Honma et al. employed a new effective shell model interaction,

designated GXPF1. The GXPF1 calculations have been successful in reproducing the
experimental E(2+
1 ) values in the neutron-rich 20 Ca, 22 Ti and 24 Cr isotopes, as shown

in Figure 1.7, including the systematic increase in E(2+

1 ) at N = 32. The onset of an

N = 34 shell closure was first expected to be evident in the Ti isotopes, where E(2+
1)

56
for Ti34 was predicted to lie at approximately 1500 keV, similar in energy to its
54
neighbor Ti32 , see Figure 1.7.
The current experimental study focused on the measurement of E(2+
1 ) in
56
Ti34 ,
56
populated following the β decay of Sc to test for evidence of a new subshell at

11
 r
4

Figure 1.7: E(2+ ) as a function of neutron number for the 24 Cr, 22 Ti, and 20 Ca isotopes.
The circles connected by lines represent GXPF1 calculations and the black bars are
experimental points. GXPF1 predicts a shell closure at N = 34 in the Ti and Ca
isotopes evidenced by the systematic behavior of the calculated E(2+ 1 ) values.

12
N = 34 for neutron-rich nuclides. Details regarding the study are provided in this

dissertation, and are organized in the following manner. A review of the different
decay paths for a neutron-rich parent nuclide are given in Chapter 2. A description

of the experimental endstation for the detection of implanted ions, β and γ decays is
given in Chapter 3. Also included in Chapter 3 is a description of the calibration of
the both β and γ detectors, the correlation between implanted ions and decays and

the electronic system used during the experiment. The results from the analysis of
56 57 58,59 60
the β decay of Sc, along with the β decays of Ti, V, and Cr, are given in

Chapter 4. A comparison between experimental results and calculated level structures

56 57
is given for Sc and Ti in Chapter 5. The tentative ground state spins and parities

of nuclides around the N = 32 subshell closure are analyzed in an attempt to describe

the monopole migration of the νf5/2 state as a function of proton number in the f7/2
in Chapter 5. The conclusions and outlook are given in Chapter 6.

13
Chapter 2

Technique

The primary purpose of this study was to measure the E(2+

1 ) of
56
Ti to determine the

possible presence of an N = 34 subshell closure as discussed in the previous chapter.

The E(2+
1 ) values for even-even nuclei can be measured using a variety of different

methods including Coulomb excitation [15], direct reactions [16], and deep-inelastic
reactions [13]. For the present study, β decay was used to selectively populate the

low-spin states in an even-even daughter nuclide. A parent nuclide was produced

and its β decay was monitored. The β decay may populate excited nuclear states
in the daughter and these daughter states can then decay through γ-ray emission,

internal conversion, or if energetically allowed, through the emission of a neutron.

The observation of the secondary radiation is used to construct a decay scheme for

the parent nuclide into the different daughter states using absolute γ-ray intensities
and γγ coincidence information.

2.1 β decay

β decay involves the weak-decay process of converting a neutron (proton) into a proton

(neutron) keeping the mass number constant. Concomitantly, an energetic electron (β

particle) and a low mass electrically neutral particle, called a neutrino, are emitted.

14
β decay collectively refers to three distinct decay processes listed below:

A
β− : Z XN →A −
Z+1 YN −1 + β + ν + Qβ (2.1)

β+ : A
Z XN →A +
Z−1 YN +1 + β + ν + Qβ (2.2)
A
Electron Capture : Z XN + e− →A
Z+1 YN −1 + ν + Qβ (2.3)

where e− is an atomic electron, β ± is a beta particle, ν is a neutrino, and ν is an

anti-neutrino. Qβ is the characteristic amount of energy released in a specific β decay

process and represents the difference in mass-energy between the initial and final
states. If Qβ is positive the decay is energetically possible. To the neutron-deficient

side of the valley of stability both β + and electron capture processes occur, though
for a specific nucleus both may not be energetically possible because the emission of

a β + has a threshold of 1.022 MeV.

During an electron capture process, a proton captures one of the atomic electrons
converting a proton into a neutron. Along with the conversion, a mono-energetic

neutrino is emitted. The capture of an atomic electron leaves a vacancy in one of

the lower electronic shells, typically the K shell. The hole in the atomic shell left by

the capture process is unstable and the electrons quickly reconfigure. The reconfig-
uration leads to detectable X-rays and Auger electrons. Though it is probable that

neutrinos have some mass, it is very small [17, 18] and even so the neutrino escapes
undetected from most experimental apparati due to the low weak interaction cross
section; therefore, the identification of electron capture decay must be followed by

tracking the secondary emission of X-rays or Auger electrons.

Whenever, the mass difference between initial and final states exceeds twice the

mass of an electron (1.022 MeV) β + decay becomes possible. During β + decay, a

proton is converted to a neutron, similar to the electron capture process, with the

emission of a neutrino and β + particle. The β + will interact electromagnetically with

the medium after emission gradually losing energy and slowing down. Once the β +

15
particle has reached thermal energies it will quickly combine with an electron and

annihilate, producing two 511-keV γ rays propagating in opposite directions. The β +

decay can be followed both through detecting the energy deposited in a detector by

the β + itself or by detecting the annihilation γ rays.

For the current work, β − decay was the decay mode of interest. β − decay occurs
on the neutron-rich side of stability in which the weak interaction converts a neutron

into a proton with the emission of a β − particle and an anti-neutrino. The decay is
observed by detecting the emission of the β − , as it will interact electromagnetically

with the electrons of the detection material. Since β emission is a three body process,
the total Qβ value for the reaction is equally shared between the daughter nucleus, β −

particle, and neutrino. This results in a β particle energy spectrum that extends from
zero to the Qβ value (minus a small amount of energy for the mass of the neutrino and
the nuclear recoil). For the decays in the present study, the Qβ values are typically

around 10 MeV. With such a large Qβ value, it is difficult to stop the electron in
the detection system, and the possibility of undergoing β delayed neutron emission

appears.
Important information obtained by monitoring the β activity is the rate constant,

λ, of the decaying nuclides from which the half-life can be determined. An expression
for the half-life as a function of λ can be derived using the first order rate law.

d[N ]
= −N λ (2.4)
dt
N
= e−λ∗t (2.5)
N0

where t is time, N0 is the number of nuclei at time zero, and λ is the rate constant.
At a value of N/N0 = 1/2 the half-life is

ln 2
t1/2 = .
λ

16
 β decay is useful as a spectroscopic tool owing to the selectivity of the decay

process. Angular momentum must be conserved during the transition between the
initial and final states in the β decay process leading to a restrictive set of selection

rules. Note that both of the created particles have an intrinsic spin, S=1/2. If the β
particle and neutrino are emitted with zero orbital angular momentum, l = 0 (the
so-called allowed approximation), then only the intrinsic spins of the β particle and

neutrino need to be considered in the transition between initial and final states. If
the spins of the β particle and neutrino are anti-parallel (S=0), the total change

in nuclear spin, ∆J, between initial and final states must be 0 and the β decay is
termed Fermi decay. If the electron and neutrino are emitted with their spins aligned

(S=1), called Gamow-Teller decay, then ∆J = 0,1. In both Fermi and Gamow-Teller
decays, the condition l = 0 means that the parities of the initial and final states
must be the same, following the condition ∆π = (−1)l , where π is the parity of the

system. If the β particle and neutrino are emitted with l 6= 0, the decay is classified
as “forbidden”. Unlike the name suggests, “forbidden” decays can occur, just with

a much smaller probability than allowed decays.. Each value of l results in a higher
order of forbiddenness. The first, second, third,. . . forbidden decays result from the

β particle and neutrino emitted with l = 1, 2, 3, . . . respectively. For each forbidden

decay, both Fermi and Gamow-Teller types can occur, and the parity change between
initial and final states may not be the same. For example, the first forbidden decay

involves l = 1. If the β particle and neutrino are emitted with spins anti-aligned (Fermi
type) then ∆J=0,1. If the spins of the two emitted particles are aligned (Gamow-Teller

type) then ∆J=0,1,2. In both cases the parity must change between initial and final
states because l = 1 in both cases. The probability of undergoing each successive level

of forbidden decay decreases by a factor of about 104 . The selection rules for β decay
are summarized in Table. 2.1
If the β decay branching ratio to a particular daughter state is known then the

17
Table 2.1: β decay selection rules for allowed and forbidden transitions from Ref. [19].
Logft ranges taken from Ref. [5].
Type ∆J ∆π logft
Allowed 0,1 no 3.5-7.5
First Forbidden 0,1,2 yes 5-19
Second Forbidden 1,2,3 no 10-18
Third Forbidden 2,3,4 yes 17-22
Fifth Forbidden 3,4,5 no 22-24

partial rate constant of the β decay transition to that state can be determined from

λi = BRi ∗ λt (2.6)

where λi is the partial rate constant to state i, BRi is the branching ratio to state i,
and λt is the total decay rate. The sum over all partial rate constants, λi , equals the

total rate constant. The partial half-life to state i can be calculated from

partial ln 2
T1/2 = . (2.7)
λi

From the partial half-life a comparative half-life, f t, where t is in seconds, can be

calculated, where f is the “Fermi” function related to the shape of the β spectrum
that arises from the differing atomic numbers of the daughter and the endpoint energy,

allowing a comparison of different β decay transitions. The comparative half-life can

be used to gauge the level of forbiddenness in a given β transition, but should not

be considered absolute. While compilations exist for finding logf values, a purely
empirical equation for the calculation of logf is [20]:

logfβ − = 4.0logE0 + 0.78 + 0.02Z − 0.005(Z − 1)logE0 (2.8)

where E0 is the β-decay energy and Z is the atomic number of the daughter nuclide.
As the value of the comparative half-life can extend over many orders of magni-

tude, the number frequently reported is logft. The logft value associated with different

18
types of decays are shown in Table. 2.1.

2.1.1 β-delayed neutron decay

Neutron decay is the emission of a neutron from the nucleus under study. The equation
for such a decay is:
A
Z XN →A−1
Z XN −1 + n + Qn (2.9)

The Qn value for the reaction is again the mass-energy difference between the initial

and final products. The neutron separation energy is defined as the negative of the
Qn value when the product is left in its ground state. The neutron separation energy
is a positive number for all bound nuclei so that neutron decay from the ground

state is not possible. As neutrons are added to the nucleus the neutron separation
energy decreases. When the separation energy reaches zero it is not possible to bind

an additional neutron to the nucleus, and if a neutron is added, the nucleus will
undergo spontaneous neutron emission. However, in a β − decay with a large Qβ value

it may be possible to populate excited states in the daughter nucleus that are above
the neutron separation energy. Such states could then decay by neutron emission
and this process is called β-delayed neutron emission. Neutron emission occurs via

the strong, or nuclear, interaction and competes very favorably with electromagnetic
A−1
decay. The neutron decay can populate states in the X daughter which may then

undergo electromagnetic decay typical of that nucleus.

2.2 γ decay

Following a β decay, the daughter nucleus may be left in an excited state. The excited
state will generally decay to the ground state through the emission of one or more

γ rays. The emission of γ rays results from a change in either the charge or current
distribution of the nucleus giving rise to an electric or a magnetic moment, respec-

19
tively. Thus, γ ray emission can be classified as electric or magnetic in character. γ

rays, like β particles, carry off angular momentum from the excited nuclear state and
must satisfy the relation

|Ji − Jf | ≤ λ ≤ Ji + Jf (2.10)

where λ is the multipolarity of the transition and Ji and Jf are the spins of the initial
and final states, respectively.

The change in parity, ∆π, between initial and final states depends on the electric
(E) or magnetic (M) character of the transition and the angular momentum involved

as:

∆π(Eλ) = (−1)λ (2.11)

∆π(M λ) = (−1)λ+1 (2.12)

The transition rates, W, of an electric or magnetic γ-ray transition are summarized

in the following pair of equations [19]:

8π(λ + 1) 1 Eγ 2λ+1
W(Eλ) = α~c ( ) B(Eλ) (2.13)
λ[(2λ + 1)!!]2 ~c ~c
~c 2 8π(λ + 1) 1 Eγ 2λ+1
W(M λ) = α~c( ) ( ) B(M λ) (2.14)
2Mp c2 λ[(2λ + 1)!!]2 ~c ~c

where B(Eλ) and B(Mλ) are the reduced electric and magnetic transition probabilities
that contain information on the initial and final nuclear wavefunction, c is the speed

of light, α is the fine structure constant, Mp is the mass of a proton, and ~ is Plank’s
constant divided by 2π.

It is useful to obtain some simple estimates for the values of the reduced transition
probabilities, B(Eλ) and B(Mλ), but this requires a model of the nucleus. Adopting
an extreme independent particle model, and assuming that the γ radiation involves

the transition of one nucleon from an initial to a final single-particle state, the reduced

20
Table 2.2: γ transition rates and Weisskopf single particle estimates for the reduced
transition probabilities from Ref. [19].
λ W(Eλ) W(Mλ)
1 1.02×1014 A2/3 E3γ 3.15×1013 E3γ
2 7.28×107 A4/3 E5γ 2.24×107 A2/3 E3γ
3 1.02×10A2 E7γ 1.04×10A4/3 E3γ
4 1.02×10−5 A8/3 E9γ 3.27×10−6 A2 E3γ
5 1.02×10−12 A10/3 E11γ 7.36×10−13 A8/3 E3γ

transition probabilities can be calculated as:

1 3 2
Bw (Eλ) = ( ) (1.2)2λ A2λ/3 e2 f m2λ (2.15)
4π λ + 3

10 3 2
Bw (M λ) = ( ) (1.2)2λ−2 A(2λ−2)/3 µ2N f m2λ−2 (2.16)
π λ+3

where µN is the nuclear magneton, λ is the multipolarity, and A is the mass number.

Equations 2.15 and 2.16 are called the Weisskopf single-particle estimates for the re-
duced transition probabilities. Table. 2.2 lists the transition rates for different electric

and magnetic multipoles using the Weisskopf single-particle estimate for the reduced
transition probability.
For a given γ transition, it is possible that several multipole radiations may be

emitted. From the estimated transition rates shown in Table 2.2 it can be seen that
the lowest multipole γ transition will dominate the decay. Secondly, it is possible for

transitions to be of mixed type with magnetic and electric transitions, but the only
prominent case is when a mixed M1/E2 transition is possible.

2.2.1 Internal Conversion

A competing process to γ-ray emission that may be significant for low energy transi-
tions or transitions involving a high value of λ is internal conversion. Internal conver-

sion occurs when the electromagnetic fields responsible for the emission of a photon
interact with one of the atomic electrons resulting in its emission from the atom with-

21
out creation of a photon. In contrast to β decay, the electron is not created during the

decay process, but is a preexisting atomic electron. The energy of the emitted electron
will equal the decay energy minus the binding energy of the electron. It is possible to

emit electrons from different atomic shells, and a measurement of the relative prob-
abilities of emitting conversion electrons from different atomic shells is one way to
determine the multipolarity of the transition involved. The probability of internal

conversion will increase if: Z is increased, the energy of the decay is decreased, or the
multipolarity of the transition is increased. The probability of internal conversion for

a particular decay is defined by the internal conversion coefficient, α.

λe
α= (2.17)
λγ

where λ is the rate of decay of each radiation type, defined as 0.693/tpartial

1/2 . The inter-

nal conversion coefficient can vary between 0 and ∞. An important case of internal
conversion occurs in transitions between 0+ states. Since λ = 0 and ∆π = (−1)0 = +1,

these are E0 transitions. In such transitions, γ emission can not occur and the decay
must proceed through internal conversion or pair production.

2.3 Application

β-delayed γ-ray spectroscopy was utilized to investigate evolution of shell structure

around the N = 32 and the suspected N = 34 subshell closure. Parent nuclides were
produced and the β decay into daughter states was observed. From the selectivity of
the β transition, daughter states were populated if they satisfy the relation ∆J = 0,1

with the same parity. Detection of the delayed γ rays enabled the determination of the
excited nuclear states in the daughter, in particular, the energy of the first excited 2+
1

state was determined giving an indication on the possible existence of shell closures in
the region of study. β-delayed neutron emission was detected through the observation

22
of γ rays in the A-1 mass chain with respect to the nuclide being examined. Internal

conversion was not directly observable but needed to be considered, especially for low-
energy transitions as the process will reduce the observed γ-ray intensity. The β decay

to, and the energies of, excited nuclear states in the daughter provided information on
the underlying nuclear shell structure in this region of exotic nuclei. The next chapter
will focus on the method of production for the parent nuclides and the instrumentation

used for the βγ measurements.

23
Chapter 3

Experimental Setup

3.1 Isotope Production and Delivery for Experi-

ment NSCL-02004

Except for the few long-lived natural radioactive elements, such as U and its radioac-
tive daughters, radioactive isotopes must be artificially produced on earth. At the

National Superconducting Cyclotron Laboratory (NSCL), radioactive isotopes are

produced using the projectile fragmentation method. In this method, a very fast pro-
jectile is impinged onto a target and a fraction is converted into one of a myriad of

reaction products. The reaction products, both stable and radioactive, emerge from
the target with large forward momenta and are analyzed to select only the nuclides

of interest.
The primary purpose of the current experiment was to study the β decay of
56 56 86
Sc. Sc was produced using the projectile fragmentation of a primary Kr beam.
86
The Kr beam was extracted from the K1200 at an energy of 140 MeV/nucleon
and was impinged on a 376 mg/cm2 9 Be fragmentation target. The A1900 fragment

separator [21] was used to isolate the interesting reaction products from the resulting
assortment of nuclei produced in the 86 Kr on 9 Be fragmentation reaction. A schematic

layout of the K500 and K1200 superconducting cyclotrons and the A1900 fragment

24
 Figure 3.1: Layout of the K500, K1200 and A1900 components of the NSCL.

separator is shown in Figure 3.1.

A key feature of the present study was the production, separation, and identifica-
tion of each isotope prior to its decay. For this study, the A1900 had to be tuned to
56
optimize the production of Sc. The magnetic rigidity of the first half of the A1900

was set to Bρ1 = 4.239 Tm. At the second intermediate image the beam achieved
its maximum dispersion in the horizontal direction and was passed through a 330

mg/cm2 Al energy-loss wedge and 1% momentum slits allowing for a physical separa-
tion of the transmitted beam at the focal plane (labeled “dispersive plane” in Figure

3.1). A small momentum slit was chosen due to the unavailability of the scintillator
at the dispersive plane, which is necessary for time of flight correction to the particle
identification when a larger momentum acceptance is used. The magnetic rigidity of

the second half of the A1900 was set to Bρ2 = 3.944 Tm. The transmitted particles
were brought back into focus at the final image and then delivered to the experimen-

tal endstation in the N3 vault. The rate of 56 Sc at the endstation was measured to be
0.0045 particles/s pnA. With an average primary beam current between 10 and 15
56 56 57 58,59 60
pnA the rate of Sc was ∼3/min. Along with Sc, Ti, V, and Cr fragments
were also transmitted to the experimental endstation.
The fragments were implanted directly into a Double-Sided Silicon Strip Detector

(DSSD) to be described later in this chapter. Particle identification was achieved

25
through a combination of energy loss signals in silicon detectors at the endstation and

a measurement of fragment time of flight (TOF). The cyclotron rf signal was used as
the TOF stop throughout the experiment. Two different timing arrangements served

to create the start signal for the TOF measurement. The first start signal came from a
thin scintillator in the experimental vault approximately 2 meters upstream from the
DSSD. One third of the way through the experiment a beamline vacuum pump failed,

resulting in the loss of the scintillator and necessitating its removal from the beam
line. A replacement signal, used for the remainder of the experiment, was created from

a silicon PIN detector (PIN1) located only 1m upstream of the DSSD. While using
the timing signal from the PIN detector resulted in a poorer TOF resolution, due

to poorer intrinsic resolution, the particle identification plot was still clean enough
56
to clearly identify all transmitted products and Sc was well separated from other
implanted isotopes. The energy-loss signal used in particle identification was derived

from the same silicon PIN detector. Plotting the energy loss as a function of TOF
results in the particle identification plot shown in Figure 3.2 (using the first timing

arrangement). The five different isotopes implanted at the experimental endstation

again were: 56 Sc (12300), 57 Ti (192100), 58 V (153400), 59 V (539800) and 60 Cr (275100),

where the number in parentheses was the total number of implanted ions for that
56
nuclide used in the analysis presented in the next section. Notice that Sc was <1%
of the secondary beam delivered to the experimental endstation.

3.2 Detector Setup for Experiment NSCL-02004

During experiment number NSCL-02004 two different detector systems were used in

coincidence to monitor implanted ions, β decays and γ-ray transitions. The radioactive
beam was implanted into the DSSD of the β calorimeter setup to be discussed below.

The β calorimeter was also used for β decay detection. The correlation of implanted
nuclei with their subsequent decays was performed in software. β-delayed γ rays were

26
 60 300
59
V Cr
DE (arb. units)

57
Ti
56 200
Sc

58
V 100

0
Time of Flight (arb. units)

59
V
20000
16000 60
Cr
12000 57
Ti 58
V
8000 56
Sc
4000 s )
unit
.
0 rb
(a
Time of Flight (arb. un
its) DE

Figure 3.2: Particle identification using Time of Flight (start from scintillator 2 m
upstream of the DSSD and stop from the cyclotron rf signal) and energy loss in
the first PIN detector. 56 Sc is well separated from the other implanted isotopes even
though it constitutes <1% of the total beam.

27
monitored with 12 HPGe detectors from the Segmented Germanium Array (SeGA)

[22].

3.3 Beta Calorimeter

The low rates for the production of very neutron-rich radioactive beams makes bulk
activity measurements in which the beam is cycled on and off to study β decay

properties intractable [23]. A device for β decay studies using a continuously implanted
beam into an active detector is preferable, eliminating the need for cycling the beam
on and off. Implanting ions into an active detector would also enable the identification

of implanted ions and subsequent decays on an event-by-event basis allowing for the
study of species which constitute a small percentage of the total number of implanted

nuclei. To this end, the NSCL Beta Counting System [24] was developed. A schematic
diagram of the β counting system is shown in Figure 3.3 The center piece of the β

counting system is a 4cm × 4cm × 1482µm Micron Semiconductor Double-Sided

Silicon Strip Detector (DSSD) segmented into 40 1-mm strips on both the front and
back faces that results in a total of 1600 pixels, each of which is an individual detector!

The detector was oriented so that the p-type Si was the front and the n-type Si was
the back of the detector. The DSSD was biased with +215 V on the back of the

detector.
Following downstream from the DSSD were six 5cm × 5cm Single-sided Silicon

Strip Detectors (SSSD) which constitute the β calorimeter. The thicknesses of SSSD1
through SSSD6 were; 990µm, 977µm, 981µm, 975µm, 989µm, and 988µm thick, re-
spectively. Each of the six SSSDs are segmented into sixteen strips on one face and

were mounted behind the DSSD so that the strip orientation alternated between the
x and y directions, see Figure 3.3. The SSSDs were placed with ∼2 mm between the

center of each detector (∼1mm between faces). The DSSD and the first SSSD were
separated by 7mm. Following the six SSSD detectors are two 5cm × 5cm PIN de-

28
Drawing not to scale Beta Calorimeter

PIN1a PIN2a SSSD1 SSSD3 SSSD5 PIN3

1061-16 2007-8 2194-1 2186-5 2194-14 2103-14
PIN2 SSSD2 SSSD4 SSSD6 PIN4
2095-23 2194-12 2186-10 2194-4 2103-12
DSSD
2035-3

21 mm 7 mm 13 mm
9 mm 9 mm 1 mm 1 mm

PIN1
1404-3 Al degrader Scintillator
Beta Counting System
435
mg/cm2

6 mm
Figure 3.3: Schematic of the β counting system and its position relative to other
detectors used during the experiment. Drawing is not to scale and inter-detector
distances are given. Silicon detector serial numbers are given below the name of the
detector that will be used throughout this document.

29
tectors that are 993µm and 998µm thick, respectively, for a total of approximately 8

mm of silicon behind the DSSD, sufficient to stop β particles up to about 4 MeV [25].
The decays in the present experiment have β-decay Q values in the range of 10 MeV.

To stop the high-energy β particles that escape the Si stack an additional 2 inches of
plastic scintillator were placed after PIN4.
Upstream of the DSSD were four PIN detectors designated PIN1, PIN1a, PIN2,

and PIN2a which were 474µm, 488µm, 992µm, and 966µm thick respectively. PIN1
provided both energy loss and TOF signals used in the particle identification discussed

above. The signal from PIN2a was split and input into high- and low-gain amplifiers
to observe β decays and implanted ions, respectively.

The thickness of the upstream Al degrader (435 mg/cm2 ) located between PIN1
and PIN1a was optimized to allow for the implantation of 56 Sc at a depth of ∼600µm
56
in the DSSD in the direction parallel to the beam axis. With Sc implanted at

∼600µm in the DSSD there was an increased probability of recording a β particle ∆E

signal above noise when the β particle was emitted in the direction of the SSSDs.

3.3.1 Electronics

A schematic diagram of the signal processing system is shown in Figure 3.4. The
signals generated in each strip of the DSSD were first sent to a grounding board.

The grounding board paired each signal with a ground line and grouped the signals
into blocks for further processing in commercial 16-channel electronic modules. The

“segmentation” process grouped the channels 1-16, 17-32, and 33-40 together for both
the front or back of the detector. These blocks of channels for the front or back were

then sent into preamplifiers. The challenge of detecting β decays and implantation
events in the DSSD is the large difference in the total energy deposited for the two
events; high energy implantation events deposit ≈ 1’s GeV while low energy decay

events deposit ≈ 100’s keV. To overcome the disparate energy scales, Multi Channel
System (MCS) preamplifiers with both low- and high-gain outputs were used with

30
 Low-gain CAEN 785
ADC

MCS
Grounding
DSSD CPA 16
Board
preamp
Fast
A OR of 16 strips
Pico Sys.
Delay Bit Register
High-gain Shaper/ Disc
Lecroy 4434

Slow
Scaler

CAEN 785
ADC

ADC Gates

A(front)
AND Master Gate Master Scaler
A(back) AND Fan In/Out
Live
Computer Bit Register
Not Busy (Gate)

TDC Start

Figure 3.4: Electronics setup for the DSSD during experiment 02004 along with trigger
conditions. The diagram shown on top represents one block of channels for the DSSD
readout.

the DSSD. The MCS preamplifiers were also used with the SSSDs, where only the
high-gain side of the preamplifier was used. High-energy implantation and low-energy

decay signals were obtained from the low-gain (0.03 V/pC) and high-gain (2 V/pC)
outputs of the preamplifier, respectively. The low-gain signals were then sent directly

to a VME ADC. The high-gain signals were further processed through Pico Systems
shaper/discriminators. Following shaping, the energy signals (high-gain) were sent to
VME ADCs. The complete diagram of the electronics setup for the DSSD, including

the trigger logic, is shown in Figure 3.4. The timing signals of the 40 front channels,
from the Pico Systems module, are OR’ed together, as are the 40 back channels. The

31
OR’ed front and back signals are AND’ed to provide the master gate signal in the

experiment. The master gate signal AND’ed with a computer not-busy signal define
the master live signal. The master live signal triggered computer acquisition to read

the detectors. The 12 Ge detectors were read on every event but could not trigger
the acquisition independent of the DSSD. The DSSD was read in blocks of 16 low-
and high- energy channels. For example, if front strip 10 registered a decay event,

front strips 1 through 16 were read into the analysis from both the high and low gain
signals. The SSSDs were read out two detectors at a time. For example, if strip 1 in

SSSD1 registered an event then strips 1 through 16 were read into the analysis for
both SSSD1 and SSSD2.

A schematic diagram of the PIN electronics are shown in Figure 3.5. A mentioned
before, PIN1 was used as the TOF start signal and the cyclotron rf signal was used
as the TOF stop for particle identification. PIN2a was used as both a fragment and

a β detector.
Implanted ions and decays were “time stamped” on an event-by-event basis for use

in the software correlation. The time signal was derived from two 16-bit EG&G RC014
Real Time Clock (86 Hz) modules operated in parallel. The 86 Hz internal pulser from

Clock1 was downscaled by a factor of eight and counted by Clock1. At the reduced
frequency, Clock1 filled its 16 bit field in 2 s, a resolution of 30.5 µs. When Clock1
reached the end of its range an overflow signal was sent to Clock2, which counted

the number of resets. The two clocks were read out on every event (both implanted
ions and decays) to determine the time stamp. The time stamp was derived from the

equation:
timestamp = clock1 + 65536 ∗ clock2

3.3.2 Calibration
90 228 228
Each Si detector was calibrated with both a Sr β and Th alpha source. A Th
alpha spectrum was accumulated for each strip for gain-matching purposes. The strips

32
 CAEN 785 LeCroy 2551
ADC Scaler

Slow
Bit Register
PIN Tc178 TC 241S TC 455 Fan
1 preamp In/Out Delay CAEN V775
amp CFD
TDC
Fast
Start CAEN 785
TAC
ADC
Stop
Rate Divider (1)

Cyclotron RF

CAEN 785 LeCroy 2551

ADC Scaler
Slow

Bit Register
PIN Tc178 TC 241S TC 455 Fan
1a,2,3,4 preamp In/Out Delay CAEN V775
amp CFD
TDC
Fast

LeCroy 2551
Scaler
Fast
Bit Register
PIN Tc178 TC 241S TC 455 Fan
2a preamp In/Out Delay CAEN V775
amp CFD
TDC
High Gain CAEN 785
ADC
Slow

Low Gain CAEN 785

ADC

Figure 3.5: Electronics setup for PIN detectors during experiment 02004.

33
were gain-matched in software based on the centroid of the lowest energy alpha peak
228
at 5.4 MeV. The total Th spectra created by summing all 40 strips on front and
back of the DSSD are shown in Figure 3.6(a-d). Two different methods were used to

determine the energy of the alpha particle for the spectra, shown in Figure 3.6, and are
labeled as either Esum or Emax . Esum was determined by summing the energy collected
in the DSSD (above threshold) for all 40 strips, shown in Figure 3.6(a,c) for either

the front or back of the DSSD, respectively. Emax was determined by comparing the
energies deposited in the individual 40 strips of the DSSD (above threshold), for either

the front or back of the detector, and only reporting the maximum energy, shown in
Figure 3.6(b,d). Two important observations were made concerning the alpha spectra.

First, the resolution of the front is much better than the back of the DSSD. As a
result, the energy signals from the front of the DSSD were used whenever a DSSD
energy signal was needed. Secondly, in calibration spectra reporting the maximum

energy deposited into one channel, Figure 3.6(b,d), there is significant low-energy
tailing on the individual alpha peaks, especially for the back of the detector. Low

energy tailing in the maximum-energy spectra can result from either the incomplete
collection of charge in the DSSD, or the division of charge between two adjacent

strips. Comparing the two spectra from the back of the DSSD (sum versus maximum
energy) it is seen that most of the low-energy tailing is removed when the energy sum
is used, suggesting charge division between adjacent strips on the back of the detector

is important. The small low-energy tailing is not improved significantly between the
sum and maximum-energy spectra for the front of the DSSD suggesting that its origin

might be incomplete charge collection.

90
Threshold calibrations using a Sr source were also performed. First, hardware

thresholds were set for each strip individually by observing the shaped output of the
90
DSSD on an oscilloscope triggered with a CFD signal. A Sr spectrum was then
collected for each strip and used to determine the software threshold. Representative
90
Sr spectra for the front and back of the DSSD are shown in Figure 3.7. Strips near

34
 60
(a) 60 (b)
40
40
20 20

0
Counts

Counts
0
60 60

40 40

20 20

0 0
DSSD Front (sum) DSSD Front (max)

40 60
(c) (d)
40
20
20
0
Counts

Counts

0
60
40
40
20 20

0 0
DSSD Back (sum) DSSD Back (max)
Figure 3.6: DSSD 228 Th calibration for the front and back of the DSSD. Arrows mark
the 5.4 MeV 228 Th alpha peak that was used for gain-matching. Panels a,c report the
total energy collected in the DSSD determined from the sum of the energy deposited
in each strip. Panels b,d report the maximum energy collected in one channel.

35
 30 30
(a) (b)
20

Counts
20
Counts
10 10

0 0
DSSD Front Strip 17 DSSD Front Strip 27

30 30
(c) (d)
20
Counts

20

Counts
10 10

0 0
DSSD Back Strip 17 DSSD Back Strip 27

Figure 3.7: Representative DSSD 90 Sr calibration for the front and back of the DSSD.
Arrows mark the location of representative software thresholds.

the edge of a readout cable, such as strip 17, were inherently noisier and required an

increased hardware threshold. Additionally, the back of the DSSD was noisier than
the front, supporting the choice of the front strips for DSSD energy information. The

average threshold for the DSSD strips was ∼120 keV.

To take advantage of the high number of pixels in the DSSD, the secondary frag-

ment beam was defocused to illuminate as much of the detector as possible. The
defocusing resulted in the illumination of approximately 2/3 of the DSSD. The im-
plantation distribution in the x and y dimensions of the DSSD for all implantation

events is displayed in Figure 3.8(a). The distribution for 56 Sc events is shown in Figure
3.8(b). The problem encountered early by Prisciandaro [23] using single output pream-

plifiers, where the average implantation multiplicity was six, making the localization
of an implant event difficult, was remedied by using the dual-output preamplifiers
discussed in Section 3.3.1. Using the dual-output preamplifiers the location of an im-

plantation event could be well defined, for example Figure 3.9 shows the implantation
multiplicity for the front and back of the DSSD.

36
 All implants
3000
2500
Counts

2000
1500
1000
500
0
40
35 40
cha 30 25 30
35
nne
l nu 20 15 25 (x)
mbe 20 um ber
15
r (y 10 10 nel n
) 5 5 chan

25 Only 56Sc implants

20
Counts

15
10
5
0
40 40
35 35
30 30
cha 25 20 25
e r (x
)
nne 20 b
l nu 15 15 um
mbe 10 10 nnel n
r (y
) 5 5 cha

Figure 3.8: The radioactive beam is defocused to illuminate approximately 2/3 of the
DSSD and the distribution of implanted ions in the x and y dimension across the face
of the DSSD is shown in (a). The implantation profile is still distributed across the
face of the DSSD when (b) only 56 Sc events are displayed.

37
 1000000
Back
Front
800000 2000

600000
1000
Counts

400000
0
5 6 7 8 9
200000

0
1 2 3 4 5 6 7 8 9

Implant Multiplicity
Figure 3.9: Implantation multiplicity of the DSSD. The inset shows implantation
multiplicities, five through nine, with a higher resolution on the y axis.

front
AND
back
AND “decay”
PIN2a
AND
SSSD1

Figure 3.10: Software condition between the front and back of the DSSD, PIN2a,
SSSD1 which was defined as a decay.

3.3.3 Correlation

Correlations between implanted ions and decays were performed in software on an

event-by-event basis. An implanted ion was defined as a signal above threshold in

PIN1 and a front and back (low-gain) strip of the DSSD, combined with no signal in
the first SSSD. A decay was defined as a signal above threshold in a front and back

strip (high-gain) of the DSSD, combined with either a signal in PIN2a or SSSD1 or,
alternatively, a signal in neither PIN2a nor SSSD1, see Figure 3.10. An event which

fired both PIN2a and SSSD1 was identified as a penetrating light particle which leaves

38
an energy signal in every detector of the β counting system. After defining the condi-

tions necessary for identification of implanted ions and decays the two distinct events
must be correlated. Correlations between implanted ions and decays were performed

in software on an even-by-event basis based on pixel location and timing informa-

tion. Previous analyses of data from the beta counting system had restricted the
decay-implant correlation to events that occur within the same pixel (see for example

Ref. [26]). The average implantation rate in the current experiment over the entire
detector was lower than 10Hz. The low implantation rate allowed for the expansion

of the fragment-β correlation field to the surrounding 8 pixels.

The two parameters defining the decay-implant correlation function are the maxi-

mum correlation time and the minimum time between implanted ions. The maximum
correlation time is defined as the longest time window in which a β decay can be cor-
related with an implanted ion. The longer the maximum correlation time, the greater

the chance of correlating the implanted ion of interest with a background event. In
addition, the maximum correlation time needs to be long enough to encompass a few

half-lives of the nuclide under study in order to extract a reasonable decay curve. The
minimum implantation time is defined as the minimum amount of time between two

successive implantation events before the second implanted ion will be considered in
the correlation. The shorter the minimum implantation time, the greater is the chance
of incorrectly correlating a decay with an implantation.

In the present analysis the maximum correlation time and the minimum time
56 57
between implanted ions were both set to one second for the analysis of Sc, Ti,
58 59 60
V, and V. Due to the longer half-life of Cr, its analysis was conducted with
the maximum correlation time and minimum time between implanted ions set to

five seconds. A decay curve for each isotope was constructed by histogramming the
time difference between the implanted ion of interest and the correlated β decay. An
58
example is shown in Figure 3.11(a) for the decay of V. By integrating the decay
58
curve for the parent isotope the total number of detected β decays from V can be

39
 10000 fragment-b (a) (b) 300

DE (arb. units)
Counts/50ms 58
200
1000 V
bkgd 58 100
58 V
Cr
0
100 Time of Flight (arb. units)

Figure 3.11: The decay curve in (a) is produced by histogramming the time difference
between 58 V implanted ions and the correlated β decays. By fitting the decay curve,
the contribution due to 58 V β decays (shaded region) can be determined. Using the
total number of 58 V decays and the total number of implanted ions, from the parti-
cle ID spectrum shown in (b), the β detection efficiency for the calorimeter can be
determined. The β detection efficiency for all isotopes was ∼30%.

determined. Combined with the total number of implanted ions, from the particle ID

shown in Figure 3.11(b), the β-detection efficiency of the DSSD was calculated. For
all the isotopes studied the β-detection efficiency was ∼30%.

During the course of the present analysis, two methods were used for handling
the correlation of multiple decays within the maximum correlation time, following a

specific implantation event. In the first method, once a decay was correlated with an
implanted ion, the implantation information was deleted (back-to-back decay rejec-
tion). If another decay is detected that would have been within the correlation time

of the deleted implanted ion the decay is ignored. Back-to-back decay rejection led
to the suppression of correlations at longer times in the decay curve that were asso-

ciated with daughter and grand-daughter activities. The effect is not very significant
for those decay chains with daughter half-lives in the second range (such as the decay

chain associated with 58 V). However, the reduction at longer correlation times signif-
56
icantly affected the decay curve of Sc, where the half-lives of all three generations
56 56 56
are fairly short (40 ms, 200 ms, and 216 ms for Sc, Ti, and V, respectively) and

should reasonably be expected to occur within the one second correlation time. Using
56
back-to-back decay rejection resulted in an inability to fit the Sc decay curve with

multiple generations over the one second correlation window, making it difficult to

40
 56
extract the Sc half-life. The second method of correlating multiple decays within a

given correlation time ensured that all decays that occurred after an implanted ion,
up to the maximum correlation time are correlated. Since the implanted ion infor-

mation was never erased, multiple decays can be correlated with the same implanted
ion. The removal of back-to-back decay rejection increased the events at longer cor-
relation times, associated with daughter and grand-daughter decays, and enabled the
56
fitting of the Sc decay curve through all three generations. The back-to-back decay
rejection condition was removed for all analyses presented here.

Using the high segmentation of the DSSD combined with the segmentation of the
SSSDs, particles can be tracked through the β calorimeter by making multiple mea-

surements of their x and y positions as energy is deposited in the different Si detectors.

With this capability, different events in the SSSDs can be identified. Light particles
transmitted to the endstation along with the primary beam follow straight line paths

through the stack. Since there is little deviation in their path as they traverse the
stack, such events will show a high correlation between y positions (or x positions)

measured with different detectors. For example, the strips on the front of the DSSD
were oriented in the y direction as were the strips of SSSD1. A plot of the measured

y position from the DSSD versus the measured y position from SSSD1 is shown in
Figure 3.12(a). Two distinct distributions can be seen. The first distribution is a
highly correlated pattern running across the diagonal which can be associated with

light particles following straight-line trajectories. The second distribution observed in

Figure 3.12(a) originates from β particles. This distribution can be isolated by ap-

plying the aforementioned software condition for identifying β decays, PIN2a, DSSD,
and SSSD1 can not all fire in coincidence, and results in Figure 3.12(b), which is free

from light particles. There is a small correlation between the two different y-position
measurements but due to the isotropic emission of β particles, it is much weaker than
the correlation observed with light particles. Lastly, when the y dimension, from the

front of the DSSD, is plotted against a measurement of the x position, a plot similar

41
 x104 x102
40 40
(a) 3 (b) 7
6
30 30
5
2
4
20 20
3
1 2
10 10
1
0 0
4 8 12 16 4 8 12 16
Front DSSD strips versus Front DSSD strips versus
SSSD1 strips SSSD1 strips
x103 x102
40 40
(c) (d) 4
5
30 4 30
3
20 3 20 2
2
10 10 1
1
0 0
4 8 12 16 4 8 12 16
Front DSSD strips versus Front DSSD strips versus
SSSD2 strips SSSD2 strips

Figure 3.12: Correlation, in strip number, between the front of the DSSD (y direction)
and (a) SSSD1 (y direction) (b) SSSD1 (y direction) with a restriction to only decay
events, (c) SSSD2 (x direction), and (d) SSSD2 (x direction) with a restriction to
only decay events. In (a) the two distinct distributions arise from light particles and
decay events which can be isolated, and are shown in (b). Plotting the x versus y
coordinate in (c,d) recovers a plot similar to the implantation profile.

to the implantation profile should result. Shown in Figure 3.12(c,d) are the plots of
the DSSD front (y position) versus SSSD2 (x position) for all nuclides, and gated on

decays, respectively. As can be seen, there is no correlation observable in the xy plot.

3.4 γ-ray Detection

γ rays were monitored using twelve HPGe detectors from the MSU Segmented Ger-
manium Array (SeGA) [22]. Each Ge crystal has a diameter of 70 mm, a length of

42
 Ortec 572 Ortec AD413
amp 8k ADC
LeCroy 2551
Scaler
Ortec 863 TC 455
Ge preamp Bit Register
TFA CFD

Phillips 7186H
TDC

Figure 3.13: Electronics setup for SeGA during experiment NSCL-02004.

80 mm and is divided into eight segments along its length and four segments along
the cylindrical axis for a total of 32 segments. As the β-decay studies are conducted
with a stopped beam, the high segmentation typically used for the Doppler correction

of γ-ray energies is not needed and only the central contact which provides a total
energy signal of each detector was used in the present study. A schematic diagram of

the Ge electronics are shown in Figure 3.13.

3.4.1 Calibration

The twelve γ-ray detectors were arranged in two separate rings of six detectors each,

with each cylindrical axis parallel to the beam axis, as shown in Figure 3.14. In this
relatively close-packed configuration the energy and efficiency calibration data were

measured with a Standard Reference Material (SRM) source, containing 125 Sb, 154 Eu,
155 56
and Eu, along with a separate Co source.
56
The SRM and Co sources were placed at the DSSD location and the raw Ge
energy spectra were accumulated for all 12 Ge detectors simultaneously. The peak
locations were fitted using the Oak Ridge Display, Analysis and Manipulation Module

(DAMM) [27]. Energy calibrations were obtained from a second order polynomial fit
to the true peak energy versus the channel position in each of the twelve detectors.

43
 Ge12 Ge11
73373 73445
Ge13
73474
Ge04 Ge10
73452 Ge03 73471
73177
Ge05
73446
Ge14 Ge02
Beam Ge09
73375 73486
Ge06 Line 73288
73494
Ge01
73444

Secondary
Beam
Figure 3.14: Schematic arrangement of the twelve HPGe detectors from SeGA around
the beam line containing the β calorimeter. Serial numbers for individual Ge detectors
are given as well.

Two different Ge energy calibrations have been used to analyze the data.

In the first analysis a number of different energies from the standard sources were
used to calibrate the array from 123.1 to 2598.4 keV, shown in Table 3.1. The resid-

uals of the first calibration, defined as the difference between the true energy and
the calibrated energy and shown in Figure 3.15, displayed a pronounced curvature
and deviated from actual values by more than 0.5 keV at an energy of 1 MeV. This

calibration was used in Ref. [28] and resulted in a small mismatch between the fitted
56
energy of two V γ-ray transitions and the results of previous measurements. The
56
energies for Sc γ-ray transitions were also consistently low compared to a comple-
mentary Gammasphere experiment performed on 56 Ti [29]. As a result, a new energy

calibration was constructed.

The second Ge energy calibration used two energy ranges. The first energy range
(low) extended from 123.1 keV through 1596.5 keV and the second energy range

(high) started at an energy of 1274.4 and extended through 3253.4 keV. The γ ray

44
 Table 3.1: List of γ-ray energies in keV for the two Ge energy calibrations.
Source First Calibration Second Calibration
SRM 123.071 123.071 low
SRM 247.93 247.93 low
SRM 591.763 591.763 low
SRM 723.305 723.305 low
56
Co 846.771 −
SRM 873.19 873.19 low
56
Co 1037.84 −
56
Co 1238.282 −
SRM 1274.436 1274.436 low and high
SRM 1596.495 1596.495 low and high
56
Co 1771.351 1771.351 high
56
Co 2015.181 −
56
Co 2034.755 2034.755 high
56
Co 2598.459 2598.459 high
SRM 2614.533 −
56
Co 3201.962 −
56
Co 3253.416 3253.416 high
56
Co 3272.99 −

0.5
Residual (keV)

-0.5

-1
0 1000 2000 3000
Energy (keV)
Figure 3.15: Residual plot for the 12 detector Ge array using the first energy cali-
bration. Similar curvature is also observed in each of the twelve individual detector
residual plots. There is significant deviation from true values, especially at 1 and 2.5
MeV.

45
transitions used in the low and high energy calibrations are listed in Table 3.1. To

evenly weight the fit over the entire energy range considered, an attempt was made to
use evenly distributed energy values, and in the case of a cluster of energies, at 3200

keV for example, only one of the energies was used. The transition between the low-
and high-energy calibrations was arbitrarily chosen to occur at an energy of 1460 keV
and it was checked that the position of the transition did not affect the calibrated

energies. The low- and high-energy calibrations greatly reduced the magnitude of
the residuals. The individual residuals for all 12 detectors are shown in Figure 3.16.

The residuals of the Ge array, composed of all twelve individual detectors, is shown
in Figure 3.17. Using the low- and high-energy calibrations, the residuals have been

reduced dramatically to ±0.15 keV, and the curvature has been minimized. The error
on reported γ ray energies provided in the results section were obtained by adding in
quadrature the error from the fit (0.15 keV), the error from the peak position given

by DAMM, and the error from the random number generator (0.3 keV). The random
number generator was used to correct for the ADC real-to-integer number conversion.

The error from the random number generator was estimated by observing the shift
in peak energy with a different value used as the start seed. An enlarged region of

the calibration spectrum for all twelve detectors is shown in Figure 3.18 from 900-
1100 keV. The two peaks located at 996.3 and 1004.7 are separated almost to the
baseline and the full width at half maximum is 3.6 keV. Both peaks in Figure 3.18,

not included in the energy calibration, are symmetric suggesting that the detectors
are gain-matched correctly.

γ-ray peak efficiency measurements were completed with the SRM source using
γ-ray transitions with known emission rates extending from an energy of 27.4 keV
56
to 1596.4 keV. The absolute emission rates of transitions in Co source used for
efficiency calibration were not known, but by matching the efficiency between the
56 56
846.8 keV Co transition and the 873.2 keV SRM γ ray, the Co source was used

to provide relative efficiencies up to 3273 keV. The logarithm of the efficiencies as a

46
 1 1
Ge 01 Ge 09
0.5 0.5
0 0
-0.5 -0.5
-1 -1
Ge 02 Ge 10
0.5 0.5
0 0
-0.5 -0.5
-1 -1
Ge 03 Ge 11
0.5 0.5
0 0
Residual (keV)

Residual (keV)
-0.5 -0.5
-1 -1
Ge 04 Ge 12
0.5 0.5
0 0
-0.5 -0.5
-1 -1
Ge 05 Ge 13
0.5 0.5
0 0
-0.5 -0.5
-1 -1
Ge 06 Ge 14
0.5 0.5
0 0
-0.5 -0.5
-1 -1
0 1000 2000 3000 0 1000 2000 3000
Energy (keV) Energy (keV)

Figure 3.16: Residual plots for the individual SeGA detectors. The switch between
low- and high-energy calibrations occurred at 1460 keV.

47
 0.15

Residual (keV) 0

-0.15
0 1000 2000 3000
Energy (keV)
Figure 3.17: Residual plot for the sum of all twelve detectors in the second Ge cali-
bration. See text for details.

10000
1004.7
8000

996.3
Counts

6000

4000

2000

0
900 950 1000 1050 1100
Energy (keV)

Figure 3.18: Sum Ge spectrum for all twelve individual Ge detectors between 900-
1100 keV. The two peaks around 1 MeV are cleanly separated and the full width at
half maximum is 3.6 keV.

48
 -0.6
y = 0.3095x5 - 4.8369x4 + 29.823x3
-0.8
- 90.777x2 + 135.91x - 80.651
log efficiency -1

-1.2

-1.4

-1.6

-1.8
1.8 2.2 2.6 3 3.4 3.8
log Energy (keV)
Figure 3.19: Efficiency of the array of 12 HPGe detectors as a function of energy. The
diamonds represent calculations with the MCNP code using a point source geometry
at the DSSD location. The squares represent experimentally determined efficiencies.
A fifth order polynomial relating the energy with efficiency is shown by the solid line.

function of the logarithm of the energy were then fitted with a fifth order polynomial.
The results are shown in Figure 3.19. The array efficiency was also simulated using

the MCNP code and accurately reproduced the experimental data down to 105 keV,
sufficient for the γ rays observed in the present study. The efficiency of the array at
1 MeV was determined to be 5.3% .

49
Chapter 4

Experimental Results

The nuclides implanted into the DSSD during experiment NSCL-02004 included:
56 57 58,59 60 56
21 Sc, 22 Ti, 23 V, and 24 Cr. Of these nuclides, Sc has not previously been studied,
and the other nuclides have been isolated only in small quantities. Thus, the present

results represent huge improvements in basic knowledge of the β-decay properties of

these nuclides. Additionally, the experiment was configured to detect isomeric γ rays

emitted from the implanted nuclides if present. Altogether, a wealth of information

was obtained for each nuclide. For all implanted nuclides a decay curve and β-delayed
γ-ray spectrum were obtained by making a cut on implantation-correlated β decays.

For multiplicity-two γ-ray events (mγ = 2), a γγ matrix was constructed by plotting
the energy of the second detected γ ray against the energy of the first γ ray. An
58
example of the γγ matrix for V is shown in Figure 4.1(a). Gated slices of the γγ
matrix provided a coincidence spectrum for each γ-ray transition. For multiplicity-one
γ rays (mγ = 1), a plot of the the decay curve versus the γ spectrum was created,

from which a γ-gated decay curve for each γ-ray transition was extracted through a
slice perpendicular to the time axis. An example of such a plot is shown in Figure

4.1(b). The only exception to the γ gated decay curve procedure described above was
in the analysis of 56 Sc. To avoid the loss of statistics associated with the restriction of

the γ-ray multiplicity to a value of one, each individual Ge spectrum (mγ ≥ 1) was

50
 4000 1000
(a) (b)
3200 800
Energy (keV)

Time (ms)
2400 600

1600 400

800 200

0 0
0 800 1600 2400 3200 4000 0 800 1600 2400 3200 4000
Energy (keV) Energy (keV)

Figure 4.1: For 58 V, two dimensional plots of (a) the energy of the the second detected
γ ray versus the energy of the first detected γ ray in a multiplicity-two γ-ray event
and (b) the time versus the γ-ray energy for γ multiplicity-one events. Projections of
(a) produced γγ coincidence spectra while projections of (b) onto the y axis produced
γ-gated decay curves.

30
1129
25
Counts/1 keV

20 690
56V
15 592 752 56 1161
10 V
5
0
0 500 1000 1500 2000
Energy (keV)

Figure 4.2: 56 Sc β-delayed γ-ray spectrum from 0-2 MeV for events occurring within
one second after a 56 Sc implanted ion.

56
gated on the Sc transition of interest. The twelve individual gates were summed
56
together to extract the Sc decay curve.

56
4.1 Sc
56
The β-delayed γ-ray spectrum obtained for Sc from 0-2 MeV, shown in Figure 4.2,
56
contains βγ events that occurred within the first second following a Sc implanted

ion. A total of seven new transitions have been identified in this spectrum. As the
56
Ti daughter has no known γ-ray transitions [30] the five remaining transitions have

51
Table 4.1: 56 Sc β-delayed γ rays along with adopted values, absolute intensities and
the initial and final states for those transitions placed in the 56 Ti level scheme.
Eγ ,β-decay Eγ ,Adopted Iabs
γ Initial State Final State
(keV) (keV) (%) (keV) (keV)
592.3 ± 0.5 592.3 ± 0.5 7±2
690.2 ± 0.4 690.1 ± 0.4 19 ± 4 2980 2290
751.5 ± 0.5 751.5 ± 0.5 9±3
1128.2 ± 0.4 1128.5 ± 0.4 48 ± 11 1129 0
1160.0 ± 0.5 1160.5 ± 0.5 21 ± 5 2290 1129

56
been assigned to the β decay of Sc, and are listed in Table 4.1.
The transitions located at 690.2±0.4, 1128.2±0.4, and 1160.0±0.5 keV compared
favorably with the energies of 690.0 ± 0.5, 1128.8 ± 0.5, and 1161.0 ± 0.5 keV found

in deep inelastic work reported in Ref. [29]. The weighted averages of the β-decay
and in-beam results, 690.1 ± 0.4, 1128.5 ± 0.4, and 1160.5 ± 0.5 keV, were adopted
56
as the energies for these three γ-ray transitions in Ti. The most intense transition
in the βγ spectrum is located at 1129 keV and has been tentatively identified as the
2+ +
1 → 01 transition [28]. The γ-ray transition with energy 592.3 ± 0.5 keV falls within

55
the error of the single γ-ray transition previously observed in Ti [31] and could be
due to a β-delayed neutron branch from 56 Sc to levels in 55 Ti. Such a branch is feasible

considering that the 13.7±0.8 MeV Q-value for the 56 Sc β decay is considerably larger
56
than the 5.3 ± 0.3 MeV neutron separation energy in Ti.
56
The decay curve derived from Sc-correlated β decays is shown in Figure 4.3(a).
The decay curves obtained by requiring an additional coincidence with the 1129-, 690-
, or 1161-keV β-delayed γ rays are shown in Figures 4.3(b-d). The half-life deduced

from the 1129-keV γ-gated decay curve is statistically different from those deduced
from the 690- and 1161-keV γ-gated decay curves, suggesting that two β-decaying
56
parent states were populated in the production of Sc. A weighted average of the
half-lives extracted from the 690- and 1161-keV γ-gated decay curves results in a

value of 60 ± 7 ms, which was postulated as the half-life of a high-spin β-decaying

state in the 56 Sc parent. Since the 1129-keV γ ray has been tentatively assigned to the

52
 (a) 56 T1/2(1+)(b) = (b-690)
1000 Sc(1+) T1/2 = 58 ± 9 ms

Counts/20ms
12
Counts/20ms
56 35 ± 5 ms
Sc (6+,7+)
56
8 Sc
100
56
V
4
bkgd 56
Ti bkgd
30 (b-1129) 10
(b-1161)
25
Counts/20ms

T1/2 = 35 ± 5 ms

Counts/20ms
56 8 T1/2 = 64 ± 11 ms
20 Sc
6
15 56
Sc
4
10
5 2
bkgd bkgd
0 500 1000 0 500 1000
Time (ms) Time (ms)

Figure 4.3: Decay curves for 56 Sc showing (a) fragment-β correlations where the data
were fit with an exponential parent decay, an exponential growth and decay for both
daughter and grand-daughter and a linear background, (b) fragment-β correlation
with an additional requirement of a coincident 1129-keV γ ray, (c) fragment-β corre-
lation with an additional requirement of a coincident 690-keV γ ray, (c) fragment-β
correlation with an additional requirement of a coincident 1161-keV γ ray. Decay
curves shown in (b-c) were fitted with an exponential decay and linear background.

2+ +
1 → 01 transition, it appears that a lower-spin
56
Sc state has a shorter half-life, and

that the 1129-keV γ-gated decay curve is then a mixture reflecting both the low- and
high-spin state half-lives. The percentages of β decays due to the low- and high-spin

state are (83 ± 11)% and (20 ± 4)%, respectively, based on absolute γ-ray intensities
56
and the deduced β-feeding to the Ti ground state.
56
The Sc-correlated β-decay curve, Figure 4.3(a), was first fitted using a single
exponential decay component for 56 Sc with daughter and grand-daughter generations.
The resulting half-life is consistent with the previous determination [28]. However,
56
evidence for two β-decaying states in Sc resulted in a re-examination of the overall
fragment-β decay curve. Since the half-life of the high-spin β-decaying state was

extracted from the decay curves gated on the 690- and 1161-keV γ-ray transitions,
the value for the low-spin half-life can be deduced from the total β-decay half-life

53
 25
140

Counts/2 keV
20 188
15

10 587 (D)
59Mn
5

0
30 200 400 600 800 1000

Energy (keV)

Figure 4.4: Isomeric γ-ray spectrum collected within a 20µs time window following a
56
Sc implanted ion.

curve. The decay curve in Figure 4.3(a) was fitted to a function that included the

high-spin state half-life, along with the percentage of β decays attributed to the high-
and low-spin states, the exponential growth and decay of the daughter and grand-

daughter isotopes, 56 Ti [30] and 56 V [26], and a linear background component. In this
way, a half-life of 35 ± 5 ms was deduced for the lower-spin state in 56 Sc. The isotopes
58,59 57 60
V, Ti, and Cr, which have previously measured half-lives, were implanted
56
along with Sc. The decay curves for these four isotopes were also fitted and used to
verify the linear background.

The experiment as configured also permitted the detection of isomeric γ rays that
decay following the implantation of a nucleus into the DSSD. The isomeric γ-ray
56
spectrum collected within a 20-µs time window following a Sc implanted ion is
shown in Figure 4.4. Three transitions were observed at 140 ± 2, 188 ± 2, and 587 ± 2
keV with absolute γ-ray intensities of (1.4 ± 0.1)%, (1.8 ± 0.3)%, and (2.2 ± 0.6)%,

respectively. It is possible that the 587-keV transition is a doublet. Unfortunately,

the errors on the absolute intensities and the lack of coincidence data do not permit
56
placement of any of the three isomeric transitions in the low-energy structure of Sc
at the present time.
56
The proposed decay scheme for levels in Ti populated following the β decay of
56
Sc is shown in Figure 4.5. The β-decay Q value was derived from the mass excess

54
 56
Sc
T1/2 = 35±5 ms T1/2 = 60±7 ms

(1+) 0+x (6+,7+) 0+y

Qb = 13.7±0.8 MeV

]
[7
3
2.
59
592

Sn 0
55
Ti
(1+) (6+,7+)
9]
[1

Ib(%) Ib(%)
1

(6+)
0.

2980
69

- 100
1]
[2
.5
60

(4+) 2290
11

- 0
8]
[4
.5
28

(2+) 1129
11

35±11 -

- 0+ 0
65±13
56
Ti
Figure 4.5: Proposed 56 Ti level scheme populated following the decay of 56 Sc. The
number in brackets following the γ-ray decay energy is the absolute γ-ray intensity.
The Q value was deduced from data in Ref. [32]. Absolute β-decay intensities are not
shown due to the presence of a possible β-delayed neutron branch to states in 55 Ti
(see text).

55
 2 690-keV gate 1129 (a) 2 (c) 690 1129-keV gate

Counts/2 keV

Counts/2 keV
1 1

0 0
2 752-keV gate (b) 2 (d) 1161-keV gate
Counts/2 keV

Counts/2 keV
1 1

0 0
0 500 1000 1500 2000 0 500 1000 1500 2000
Energy (keV) Energy (keV)

Figure 4.6: γ-γ coincidence data for the 690-, 752-, 1129-, and 1161-keV transitions
following a 56 Sc β decay.

of both parent and daughter as compiled in Ref. [33]. Absolute γ-ray intensities were
56
deduced from the number of observed Ti γ rays, the simulated γ-ray efficiency

curve, and the number of 56 Sc implanted ions correlated with β decays. The last term
was derived from a fit of the decay curve in Figure 4.3(a) as described in section

3.3. Details of the γ ray placements and spin and parity assignments were aided by a
complementary experiment carried out at the ATLAS accelerator at Argonne National
Laboratory with the Gammasphere multi-detector array to study the yrast structure

of neutron-rich Ti isotopes as described in Ref. [29]. Note that the placement of the
690-keV γ ray feeding the 1129-keV state in 56 Sc was also confirmed by fragment-βγγ

coincidences (see Figure 4.6).

56
β feedings to levels in Ti from both β -decaying states were deduced from the

absolute γ-ray intensities, and are summarized in Figure 4.5. The apparent β feedings
should be considered upper limits, based on the sizeable β-decay Q-value and the
possible existence of unobserved γ decays below our detection limit of ≈ 6% absolute

intensity and the possibility of a β-delayed neutron branch to 55 Ti. From the intensity
of the γ ray at 592 keV, a lower limit for the β-delayed neutron branch is 7%. This

56
Table 4.2: 58 V β delayed γ rays with absolute γ-ray intensities, initial and final states.
Eγ (keV) Iabs
γ (%) Initial State (keV) Final State (keV)
879.8 ± 0.4 56 ± 3 880 0
1042.4 ± 0.7 2.2 ± 0.9
1056.4 ± 0.4 23 ± 2 1936 880
1381.4 ± 0.4 2.3 ± 0.4
1500.0 ± 0.5 2.1 ± 0.5
1678.7 ± 0.6 1.7 ± 0.7
2217.5 ± 0.4 8.2 ± 0.6 3097 1936
3123.4 ± 0.6 3.1 ± 0.8

value compares reasonably well with the predicted Pn value of 12% from Ref. [34].
Apparent logft values for the observed β-decay branches are not quoted due to the

uncertainties in the ordering of and energy differences between the low- and high-spin
56 56
β-decaying states in Sc. β-decay branching to the Ti ground and first excited 2+

state of 56 Ti led to the tentative assignment of 1+ to the spin and parity of the parent
56
Sc low-spin β-decaying state. Apparent direct feeding to the 6+ , 2980-keV level in
56
Ti from the high-spin 56 Sc β-decaying state limits the spin of the high-spin isomeric

state to values of 5, 6, or 7. Absence of direct β feeding to the 4+

1 state (the absolute

intensities of the 690- and 1161-keV transitions are equivalent within experimental
56
errors) further restricts the spin of the Sc high-spin β-decaying state to J = (6, 7).

58
4.2 V
58
The β-delayed γ-ray spectrum for V in the range 0-3.6 MeV is shown in Figure
4.7(a,b). There are a total of eight γ rays observed in the spectrum, and they are

listed along with their absolute γ-ray intensities in Table 4.2. Of these eight γ rays,
those with energies of 880, 1042, 1056, 1500, and 2218 kev have been previously
observed [26]. The γ ray reported at 1571 keV in Ref. [26] was unobserved in the

current experiment and a revised energy of 1500 keV is assigned to the 1501 keV γ
ray from Ref. [26]. Three new γ rays at 1381, 1679, and 3123 keV all have absolute

57
 500
880 (a)
Counts/1 keV

400
300
200 1056
100 1042
00 400 800 1200
Energy (keV)
40
2218 (b)
Counts/1 keV

30 1500
1381
20 1679
*
3123
10
0
1200 2000 2800 3600
Energy (keV)
Figure 4.7: 58 V β-delayed γ-ray spectrum from (a) 0-1.2 and (b) 1.2-3.6 MeV for
events occurring within one second after a 58 V implanted ion. The γ ray marked with
an asterisk is from the decay of 59 Cr.

58
 250
0 - 1 sec (a)

Counts/1keV
200
150
100
50
0
58 (b)
Cr 4 - 5 sec
15
Counts/1keV

59
V
10 58
Mn
5
0
59 9 - 10 sec (c)
V
Counts/1keV

58 58Mn
8 Cr
880
1056
4

0
0 875 1750 2125 3500
Energy (keV)

Figure 4.8: 58 V β-delayed γ spectra. The maximum correlation time and minimum
time between implantation were both set to 10 s. Gates on subsets of the correlation
time were created and applied to the βγ spectrum. β-delayed γ-ray spectra are shown
with an additional cut on the correlation time extending between (a) 0-1 s, (b) 4-5 s,
and (c) 9-10 s.

intensities at or below 3%, which is lower than the detection limit in Ref. [26]. To

determine whether any of these γ-ray transitions can be attributed to longer lived
species, the correlation time and maximum time between implantation were both

changed to ten seconds, and three different cuts on correlation time were applied
to the βγ spectrum. Time cuts between 0-1 s, 4-5 s, and 9-10 s resulted in the
spectra shown in Figures 4.8(a,b,c). Prominent peaks located in Figures 4.8(b,c) can

be identified with daughter and grand-daughter decays in the A=58 decay chain or
59
with contamination from V, due to an overlap in the particle ID gates.

The measured decay curves for 58 V β decays, along with γ-gated decay curves over

59
 10000 b (a) 60 b-1056 (c)
T1/2 = 191 ± 3 ms T1/2 = 177 ± 4 ms

Counts/50ms
Counts/50ms 58 40
1000 V
58
V
58 bkgd 20
Cr
bkgd
100 0
b-880 (b) b-2218 (d)
T1/2 = 181 ± 3 ms 12 T1/2 = 177 ± 16 ms

Counts/50ms
Counts/50ms

120
8 58
V
60
58 4
40 V
bkgd bkgd
0 0
0 1000 2000 3000 0 1000 2000 3000
Time (ms) Time (ms)
Figure 4.9: Decay curves for 58 V showing (a) fragment-β correlations where the data
were fit with an exponential parent decay, an exponential growth and decay for the
daughter 58 Cr and a linear background, (b) fragment-β correlation with an additional
requirement of a coincident 880-keV γ ray, (c) fragment-β correlation with an addi-
tional requirement of a coincident 1056-keV γ ray, (c) fragment-β correlation with an
additional requirement of a coincident 2218-keV γ ray. Decay curves shown in (b-c)
were fitted with an exponential decay and linear background.

a three second correlation time were extracted, fit, and are shown in Figures 4.9(a-d).
58
The V decay curve shown in Figure 4.9(a) was fit with an exponential decay along
58
with the exponential growth and decay of the daughter nuclide Cr (t1/2 = 7s). The

grand-daughter, 58 Mn (t1/2 = 3s), decay was found not to contribute to the half-life fit.
A linear background was also included in the fit. A value of 191 ± 3 ms was extracted

for the half-life of 58 V in agreement with previous values of 185 ± 10 ms [26], 205 ± 20
ms [35], and 200 ± 20 ms [36]. γ-gated decay curves for the 880-, 1056-, and 2218-

keV γ-ray transitions are shown in Figure 4.9(b-d) and are all in agreement with each
other, suggesting the absence of multiple isomeric β-decaying states in the 58 V parent.
The 880-keV transition has previously been assigned as the 2+ → 0+ transition
58
in Cr [26]. γ-γ coincidences were determined for the 880-, 1056-, and 2218-keV

60
 12
(a) 1056 880-keV gate
8

4 2218

0
Counts/4keV 12 (b) 880
1056-keV gate

0
3 (c) 2218-keV gate
880

0
0 1000 2000 3000 4000
Energy (keV)
Figure 4.10: γ-γ coincidence data for the (a)880-, (b)1056-, and (c)2218-keV transi-
tions following a 58 V β decay.

transitions and are shown in Figure 4.10. From the coincidence data it can be seen
that both the 1056- and 2218-keV transitions feed the 880-keV 2+ state. The ratio

of intensities between the 1056- and 2218-keV transitions observed in the 880-keV
coincidence spectrum agrees with the ratio between the 1056- and 2218-keV intensities

observed in the βγ spectrum suggesting the transitions are in parallel. Furthermore,

based on the intensity of the 880-keV transition observed in the 1056-keV γ-gated
coincidence spectrum (see Figure 4.10(b)), if the 2218-keV γ ray were in coincidence

with the 1056-keV transition, a peak with an area of approximately seven counts
would be present and is clearly absent in Figure 4.10(b). The coincidence spectra
58
lead to the level scheme shown in Figure 4.11. From the number of V β decays
detected, the γ-ray intensity and the SeGA efficiency curve, the absolute intensity

61
 (1+) 0 T
1/2
= 191±3 ms
58
Q = 11.6 ± 0.4
b

]
V

.2
[8
.5
I (%) log ft

17
b
3097

22
8.2±0.6 5.4±0.1

3]
[2
.4
56
1936

10
23±2 5.19±0.09

6]
[5
(2+) 8
9.
880
87

25±3 5.37±0.08

0+ 0
44±3 5.28±0.08
58
Cr
Figure 4.11: Proposed 58 Cr level scheme populated following the decay of 58 V. The
number in brackets following the γ-ray decay energy is the absolute γ ray intensity.
The Q value was deduced from data in Ref. [32]. Observed coincidences are represented
as filled circles. Absolute β-decay intensity and logft values for the β decay to each
state in 58 Cr is shown on the left of the figure.

62
 1200
175 57
V (a)

Counts/1 keV
800
113
400
59
V 57V
0
0 100 200 300 400
Energy (keV)
120
57
V 1579 (b)
Counts/1 keV

80 1861 1923
57 2300
V 1557 2004
57
V 744 57V 2115
40

0
500 1000 1500 2000 2500
Energy (keV)

Figure 4.12: 57 Ti β-delayed γ-ray spectrum for events occurring within one second
after a 57 Ti implanted ion from (a) 0-0.4 and (b) 0.4-2.5 MeV.

of a γ-ray transition can be determined. From the absolute intensity and the level

scheme the apparent β-decay intensity to each state can be calculated along with an
apparent logft value, using a β-decay Q value of 11.6 ± 0.4 MeV [32]. Supposing that

the 58 V β decay proceeds through an allowed transition to both the 2+ and 0+ states
58 58
of Cr as indicated by the logft values, the parent V ground state is most likely a
1+ state.

57
4.3 Ti

The β-delayed γ-ray spectrum for 57 Ti in the region of 0-2.5 MeV is shown in Figures

4.12(a,b). The spectrum includes all γ rays detected within one second after the
57
implantation of a Ti ion. Six γ rays, including two prominent ones located at 268

and 692 keV, have been assigned to the daughter decay of 57 V in a previous work [26].

63
 250
0 - 1 sec (a)

Counts/1keV
200
150
100
50
0
Counts/1keV 59
V 4 - 5 sec (b)
15
* 57V
10 58
V 59 57
Cr Cr
5

0 59
12 V 9 - 10 sec (c)
Counts/1keV

59
Mn
8 57
V 58
V 59Cr
57
4 Cr

0
0 625 1250 1875 2500
Energy (keV)

Figure 4.13: 57 Ti β-delayed γ spectra. The maximum correlation time and minimum
time between implantation were set to 10 s. Gates were created on subsets of the
maximum time and applied to the βγ spectrum. β-delayed γ-ray spectra are shown
with a cut on the correlation time extending between (a) 0-1 s, (b) 4-5 s, and (c) 9-10
s.

To observe decays with longer half-lives, cuts on the correlation time between 0-1 s,

4-5 s, and 9-10 s were again applied to the βγ spectrum and the results are shown in
58,59
Figure 4.13. The presence of γ rays belonging to the decay of V can be explained
57
by an overlap in the particle ID gates with Ti. As there are no γ rays present in
the 4-5 s and 9-10 s cuts that cannot be assigned to known transitions, the remaining
57
γ rays in the βγ spectrum were attributed to the decay of Ti and are listed along

with absolute intensities in Table 4.3. It should be noted that there are no γ rays in
56
Table 4.3 that can be associated with V. Therefore, it does not appear that there

exists a delayed neutron branch which passes through the 2+ state in 56

V.

64
Table 4.3: 57 Ti β-delayed γ-rays with absolute γ-ray intensities, along with initial and
final states for those transitions placed in the level scheme.
Eγ (keV) Iabs
γ (%) Initial State Final State
(keV) (keV)
113.1 ± 0.4 14 ± 1 113 0
174.8 ± 0.4 31 ± 2 175 0
744.0 ± 0.4 2.3 ± 0.4 2475 1732
1557.3 ± 0.5 2.2 ± 0.5 1732 175
1579.4 ± 0.4 16 ± 2 1754 175
1861.5 ± 0.4 14 ± 2 2036 175
1922.9 ± 0.5 2.6 ± 0.5 2036 113
2003.7 ± 0.6 1.8 ± 0.5
2114.6 ± 0.5 0.7 ± 0.3
2300.4 ± 0.4 5.0 ± 0.5 2475 113

The half-life of 57 Ti deduced from a fit of the decay curve shown in Figure 4.14(a) is
98±5 ms. Previous measurements of the 57 Ti half-life include 67±25 ms [37], 180±30
57
ms [36], and 56 ± 20 ms [38]. However all three of these measurements isolated Ti
in only very small quantities. The half-life fit includes the exponential decay of the
57
parent, exponential growth and decay of the daughter, V, and a linear background
57
term. The half-life of the daughter, V, was taken to be 350 ms as reported in
57
Ref. [26]. The γ-gated half-lives from the decay of Ti are shown in Figures 4.14(b-
h) and Figures 4.15(a-b). All γ-gated half-lives were fit with an exponential decay
and linear background component and are consistent, within errors, with the half-

life determined without any γ-ray coincidence, further aiding their assignment to the
57
decay of Ti. Furthermore, based on the consistency of γ-gated half-lives there is no

reason to suspect multiple β-decaying states.

A γγ matrix was created to identify coincidences between the observed γ rays.
57
Coincidence spectra for various Ti γ rays were extracted and are shown in Figure

4.16. Based on the efficiency-corrected intensity ratios between the 113- and 175-keV
peaks in 1579- and 1861-keV γ-gated coincidence spectra, approximately 50% in both

cases, the 113- and 175-keV transitions were placed in parallel. From an inspection of
the 1923-keV coincidence spectrum, the 1923-keV transition has been placed directly

65
 b (a) 20 b-1557 (e)
T1/2 = 98 ± 5 ms
Counts/50ms
T1/2 = 102 ± 25 ms

Counts/50ms
16 57
10000 Ti
57
Ti 12
8
57
V 4
bkgd bkgd
1000
400 0
b-113 (b) b-1579 (f)
80
Counts/50ms

T1/2 = 114 ± 5 ms

Counts/50ms
300 T1/2 = 97 ± 7 ms
57
57 60 Ti
Ti
200
40
100 20
bkgd bkgd
0 0
b-175 (c) 16 b-1732 (g)
600
Counts/50ms

T1/2 = 103 ± 4 ms 57 T1/2 = 81 ± 40 ms

Counts/50ms

Ti
12
400
8
57
200 Ti 4
bkgd bkgd
0 0
b-744 (d) b-1861 (h)
T1/2 = 97 ± 18 ms
Counts/50ms

T1/2 = 89 ± 10 ms
Counts/50ms

20 60
57 40
Ti 57
10 Ti
20
bkgd bkgd
0 0
0 500 1000 0 500 1000
Time (ms) Time (ms)

Figure 4.14: Decay curves for 57 Ti showing (a) fragment-β correlations where the
data were fit with an exponential parent decay, an exponential growth and decay
for the daughter 57 V and a linear background and (b-j) fragment-β correlations with
an additional requirement of a coincident gamma ray with an energy of (b) 113 keV,
(c)175 keV, (d) 744 keV, (e) 1557 keV, (f) 1579 keV, (g) 1732 keV, (h) 1861 keV. Decay
curves shown in (b-h) were fitted with an exponential decay and linear background.

66
 b-1923 (a) b-2300 (b)
57
T1/2 = 84 ± 10 ms Ti T1/2 = 113 ± 22 ms

Counts/50ms
Counts/50ms
20 57 20
Ti

10 10

bkgd bkgd
0 0
0 500 1000 0 500 1000
Time (ms) Time (ms)
Figure 4.15: Decay curves for 57 Ti showing fragment-β correlations with an additional
requirement of a coincident gamma ray with an energy of (a) 1923 keV and (b) 2300
keV. Decay curves shown in (a) and (b) were fitted with an exponential decay and
linear background.

feeding the 113-keV state due to the absence of a 175-keV coincidence peak, which

would be expected to be twice as strong as the 113-keV transition, based on intensity

ratios observed in the 1579- and 1861-keV coincidence spectra. The absence of either

the 113- or 175-keV transition in the 1732-keV gated coincidence spectrum, which
does display a coincidence with 744, establishes the 1732-keV level. The level scheme
constructed from the coincidence data is shown in Figure 4.17. Since the 113- and

175-keV transitions are in parallel there should be a transition of 62 keV between the
175- and 113-keV states in order for both energies to appear in the 1579 and 1861

coincidence spectra. Such a transition would have been below the hardware thresholds
of the SeGA array and, therefore, not observed. Additionally, such a low energy γ ray
would probably be highly converted. The spin and parity assignments of 57 Ti and 57 V

will be discussed in the next chapter in after comparison to theoretical calculations.

The only concern is with the 113- and 175-keV coincidence spectra. The coinci-

dence spectrum for the 175-keV transition displays a peak located at 113 keV and
vice versa. A determination of whether these two γ rays are truly in coincidence was

made by inspecting intensity ratios between different transitions in the two separate
spectra . Supposing the 113- and 175-keV transitions are in series, a gate placed on
175 keV should reveal a one-to-one ratio between the efficiency corrected areas of the

67
 175

Counts/2keV
(a) 175
Counts/2keV

113-keV 1579-keV (e)

6
10
4 1579 1861
5 113
2
0 0
175-keV 1579 (b) 3 744 1732-keV (f)

Counts/2keV
Counts/2keV

9 113
6 2
1861
3 1
0 0
175 744-keV (c) 1861-keV (g)
Counts/2keV

175
1732 Counts/2keV 10
113
1557 5 113

0
175 1557-keV (d) 3 113 1923-keV (h)
Counts/2keV

Counts/4keV Counts/2keV

2
2
1
1
0 0
0 1000 2000 175 2300-keV (i)
Energy (keV) 4 113

0
0 1000 2000
Energy (keV)

Figure 4.16: γ-γ coincidence data for the (a)113-, (b)175-, (c)744-, (d)1557-, (e)1579-,
(f)1732-, (g)1861-, (h)1923-, and (f)2300-keV transitions following a 57 Ti β decay.

68
 (-) 0 T =98±5 ms
1/2
Q = 10.6±0.6
57 b
Ti
I (%) log ft
b

[2 ]
.0 .1
]
.3
44 4 [5
.

]
00

.5 2.6
]
2475

14
23

61 [
7
7.3±0.7 5.0±0.2

[
18 .91
22

57 [1 ]
[2 ]
]
15 2.2 [16
2036

.3 .2
.2
19

16±2 4.8±0.2 17 79.

4
1754
3
15

16±2 4.9±0.2
1.1±0.7 6.0±0.2 1732

4] .]
[1 s
1 ob
11 .7 [ 1]
3. un
61 8 [3
4.

5±3 5.7±0.2 175

17

0 113
54±3 4.7±0.2 0
57
V
Figure 4.17: Proposed 57 V level scheme populated following the decay of 57 Ti. The
number in brackets following the γ-ray decay energy is the absolute γ-ray intensity.
The Q value was deduced from data in Ref. [32]. Observed coincidences are represented
as filled circles. Absolute β decay intensity and apparent logft values for the β decay
to each state in 57 V is shown on the left of the figure.

69
Table 4.4: Ratio between the efficiency corrected areas of the 113- and 175-keV tran-
sitions to the 1579-keV transition. If 113- and 175-keV transitions are in series the
ratio should be 1. Areas are efficiency corrected.
175-keV spectrum 113-keV spectrum
Eγ (keV) Area Eγ Area
113 104 ± 59 175 87 ± 20
1579 490 ± 74 1579 172 ± 25
Ratio = 0.21 ± 0.12 Ratio = 0.5 ± 0.2

57
Table 4.5: Calculated Ti internal conversion coefficients from Ref. [2].
Energy (keV) E1 E2 E3 E4 M1 M2 M3 M4
113 0.022 0.24 2.2 19 0.016 0.16 1.4 13
175 0.0058 0.044 0.28 1.8 0.0054 0.036 0.22 1.5

1579- and 113-keV peaks. What is found is that the 113 keV area is approximately
20% of the expected 1:1 ratio, see Table 4.4. It is unlikely that internal conversion

could make up for the difference in intensities (see Table 4.5). Similarly, if a gate
is placed on the 113 keV transition the 175-keV peak is not in a 1:1 intensity ratio

with the 1579 keV peak. Therefore, the 113 and 175 keV transitions are most likely
in parallel and they appear in coincidence only due to random coincidence.

57
4.3.1 V
57
Through the decay of Ti it was also possible to study the decay of the daughter
57 57
nuclide V. The delayed γ rays of V appear in Figure 4.12 as a daughter activity.

This nuclide has been previously studied [26], and a level scheme was constructed
based on sum energy relationships and the fact that none of the three most intense

transitions were in coincidence. With the greater statistics in the present experiment,
a coincidence spectrum was extracted by gating on the 268-keV transition and is pre-
sented in Figure 4.18. The coincidences between the 268-keV transition and the 424-

and 1314-keV transitions can be seen, confirming the placement of these γ transitions
in the level scheme reported in [26].

70
 6
268-keV

Counts/2keV
4
424 1314
2

0
0 1000 2000
Energy (keV)
57
Figure 4.18: γ-γ coincidence data for the 268-keV transition following a V β decay.

59
4.4 V

The β-delayed γ-ray spectrum for those events observed within one second after a
59
V implanted ion is shown, in the energy range of 0-3 MeV, in Figures 4.19(a,b,c).

Numerous γ-ray transitions are observed in this spectrum and are listed in Table 4.6.
Of the γ rays listed in Table 4.6, the ones at 102 and 208 keV have been observed in
59 59m
previous studies of the β and isomeric decay of V and Cr, respectively [37, 39].
To determine if any of the transitions originate from longer lived species three

different gates on the correlation time were applied to the βγ spectrum between 0-1
s, 4-5 s and 9-10 s, and are shown in Figure 4.20. From the time-gated βγ spec-
trum there are multiple γ-ray transitions that can be assigned to daughter activities.

The 112-, 662-, 1126-, 1238- and 1899-keV transitions were assigned to the decay
59
of Cr. The γ rays at 112- and 1238-keV have been previously observed [40, 41].

The transitions located at 472, 570, 590, and 726 keV can be assigned to the decay of
59
Mn. Further confirmation of the assignments to daughter and grand-daughter decays
were obtained from a comparison between the Ge spectrum taken with, and without,

back-to-back decay rejection. In such a comparison, γ-ray transitions belonging to

the daughter and grand-daughter nuclei in the A = 59 decay chain were identified by

changes in intensity. The changes in intensity with, and without, back-to-back decays
as compared to the 606-keV transition are shown in Table 4.7 for the 472, 662, 1126,

71
 4000
208 (a)

Counts/1keV
3000
102
2000
59
1000 Cr

0
0 100 200 300
Energy (keV)

400
278 59 59 (b)
Mn Cr
Counts/1keV

267 59
606
300 317 Mn 880
784
463 592 59Cr 800 841 1031 59Cr
200 823 1056 1222
372 426 491 708 977
960
100 1158

0
250 500 1000 1500
Energy (keV)

60 1593
59 (c)
1530 Cr
Counts/1keV

40 1681
2090 2199
20 2375 2601

0
1500 2000 2500 3000
Energy (keV)

Figure 4.19: 59 V β-delayed γ rays in the range (a) 0-0.3, (b) 0.25-1.5, and (c) 1.5-3.0
MeV for events occurring within one second after a 59 V implanted ion. Numerous
transitions from daughter and grand-daughter generations can be observed.

72
Table 4.6: 59 V β-delayed γ-rays with absolute γ ray intensities, and initial and final
states for those transitions that were placed into the 59 Cr level scheme.
Energy (keV ) Iabs
γ (%) Initial State (keV) Final State (keV)
102.0 ± 0.4 21 ± 2 310 208
207.8 ± 0.4 41 ± 3 208 0
317.3 ± 0.4 3.0 ± 0.4 525 208
371.7 ± 0.5 1.6 ± 0.3
425.5 ± 0.4 1.7 ± 0.3 1341 915
490.8 ± 0.5 2.3 ± 0.4 800 310
592.4 ± 0.4 4.2 ± 0.3 800 208
606.0 ± 0.4 6.4 ± 0.4 915 310
707.6 ± 0.5 1.1 ± 0.3 915 208
784.1 ± 0.4 1.8 ± 0.3
799.9 ± 0.5 1.1 ± 0.3 800 0
823.2 ± 0.6 1.1 ± 0.3
841.4 ± 0.4 2.7 ± 0.3 1366 525
879.9 ± 0.5 3.0 ± 0.4
959.9 ± 0.4 2.2 ± 0.3
977.2 ± 0.5 1.4 ± 0.2 2509 1532
1030.8 ± 0.4 2.4 ± 0.3 1341 310
1056.0 ± 0.4 2.5 ± 0.3 1366 310
1157.8 ± 0.5 0.8 ± 0.2 1366 208
1206.5 ± 0.6 0.6 ± 0.2
1222.1 ± 0.4 2.5 ± 0.3 1532 310
1529.6 ± 0.5 1.0 ± 0.3
1593.4 ± 0.5 2.2 ± 0.4 2509 915
1680.9 ± 0.5 1.9 ± 0.3
2089.6 ± 0.5 0.9 ± 0.2
2198.7 ± 0.5 0.5 ± 0.2 2509 310
2375.0 ± 0.6 0.8 ± 0.2
2601.3 ± 0.6 1.2 ± 0.2
2812.5 ± 0.5 0.2 ± 0.1

1238,and 1899-keV transitions. The statistics in the Ge spectrum using back-to-back

decay rejection were too low to identify the peaks at 570, 590, 726 and 112 keV.
Since 59 V was produced in copious amounts, it was possible to deduce γ-gated half-

lives for most of the observed γ-ray transitions. These γ-gated half-lives are shown in
59
Figures 4.21,4.22, and 4.23. The V half-life deduced from the γ-gated decay curves

is 97 ± 2 ms.
The γ-gated decay curves shown in Figure 4.24 are those associated with 59 Cr and

73
 2000
0 - 1 sec (a)

Counts/1keV
1500
1000
500
0
208 59Mn 4 - 5 sec (b)
Counts/1keV 40 *
57
Ti 59
Cr
30
57V
20
58V
10 *

0
208 9 - 10 sec (c)
40
Counts/1keV

59Mn
824 59Cr
30
20 57V
58V
10
0
0 500 1000 1500 2000
Energy (keV)

Figure 4.20: 59 V β-delayed γ spectra The maximum correlation time and minimum
time between implantation were set to 10 s. Gates were created on subsets of the
maximum time and applied to the βγ spectrum. β-delayed γ-ray spectra are shown
with a cut on the correlation time extending between (a) 0-1 s, (b) 4-5 s, and (c) 9-10
s. The asterisks in (b) are due to 59 Cr decay.

59
Mn decays. With the exception of the 1238-keV transition, all of the γ rays were
fairly weak and it is unknown why the decay curves in Figure 4.24(a-e,g) do not show

the expected growth of a daughter or grand-daughter at small correlation times.

The γ-gated half-life spectra for the 880- and 1056-keV lines are both slightly

longer than the other γ-gated half-lives. Transitions at 800- and 1056-keV are also
observed following the β decay of 58 V. The difference in the half-lives of the 880- and
58
1056-keV transitions, compared to the other γ gated half-lives, may be due to V
contamination in the particle ID. In an attempt to eliminate possible contamination
from 58 V, a restricted particle ID was placed on the central portion of the 59 V particle

74
 1200 b-102 (a) 80 b-426 (e)
T1/2 = 97 ± 2 ms
Counts/50ms

Counts/50ms
T1/2 = 95 ± 9 ms
60
800
59 40 59
V V
400
20
bkgd bkgd
0 0
b-208 (b) b-463 (f)
80

Counts/50ms
Counts/50ms

2000 T1/2 = 97 ± 2 ms T1/2 = 98 ± 9 ms

59
V 60
59
V
1000 40
20
bkgd bkgd
0 0
200 b-317 (c) 100 b-491 (g)
Counts/50ms
Counts/50ms

160 T1/2 = 104 ± 7 ms T1/2 = 116 ± 11 ms

80 59
V
120 59
V 60
80 40
40 20
bkgd bkgd
0 0
b-372 (d) b-592 (h)
Counts/50ms
Counts/50ms

T1/2 = 112 ± 11 ms 120 T1/2 = 96 ± 7 ms

80
80 59
V
59
V
40
40
bkgd bkgd
0 0
0 500 1000 500 1000
Time (ms) Time (ms)
Figure 4.21: Decay curves for 59 V showing (a-h) fragment-β correlations with an
additional requirement of a coincident gamma ray with an energy of (a) 102 keV, (b)
208 keV, (c) 317 keV, (d) 372 keV, (e) 426 keV, (f) 463 keV, (g) 491 keV, and (h) 592
keV. Decay curves shown in (a-h) were fitted with an exponential decay and linear
background.

75
 b-606 (a) 50 b-823 (e)
Counts/50ms
160

Counts/50ms
T1/2 = 98 ± 6 ms 59
V T = 101 ± 14 ms
40 1/2
120 59
V 30
80
20
40 10 bkgd
bkgd
0 0
100 (f)
b-708 (b) b-841
Counts/50ms

50 59

Counts/50ms
T1/2 = 72 ± 10 ms 80 V T = 101 ± 9 ms
59 1/2
40 V
60
30
40
20
10 bkgd 20
bkgd
0 0
b-784 (c) b-880 (g)
100
Counts/50ms
Counts/50ms

T = 103 ± 14 ms T = 124 ± 15 ms
60 59 1/2 80 1/2
V (58V + 59V)
60 59
40 V
40
20
bkgd 20 bkgd
0 0
50 b-800 (d) b-960 (h)
60
Counts/50ms
Counts/50ms

T = 95 ± 14 ms T = 102 ± 11 ms
40 59 1/2 50 1/2
V 59
30 40 V
30
20
20
10 bkgd 10 bkgd
0 0
0 500 1000 0 500 1000
Time (ms) Time (ms)

Figure 4.22: Decay curves for 59 V showing (a-h) fragment-β correlations with an
additional requirement of a coincident gamma ray with an energy of (a) 606 keV, (b)
708 keV, (c) 784 keV, (d) 800 keV, (e) 823 keV, (f) 841 keV, (g) 880 keV, and (h) 960
keV. Decay curves shown in (a-h) were fitted with an exponential decay and linear
background.

76
 b-977 (a) b-1530 (e)
50 59V 16
T = 110 ± 19 ms T = 96 ± 9 ms

Counts/50ms
40 1/2 1/2
12
30 59
V
8
20
10 bkgd 4
bkgd
0 0
b-1031 (b) b-1593 (f)
60 T = 97 ± 12 ms T = 88 ± 10 ms

Counts/50ms
Counts/50ms

59 1/2 50 59V 1/2

V
40
40
30
20 20
bkgd 10 bkgd
0 0
50 b-1056 (c) 50 b-1681 (g)
T = 118 ± 15 ms T = 87 ± 15 ms
Counts/50ms
Counts/50ms

40 59 1/2 40 1/2
V
(58V + 59V)
30 30 59
V
20 20
10 10
bkgd bkgd
0 00 500 1000
b-1222 (d)
50 Time (ms)
T = 108 ± 14 ms
Counts/50ms

1/2
40
30 59
V
20
10 bkgd
0
0 500 1000
Time (ms)
Figure 4.23: Decay curves for 59 V showing (a-f) fragment-β correlations with an ad-
ditional requirement of a coincident gamma ray with an energy of (a)977, (b) 1031
keV, (c) 1056 keV, (d) 1222 keV, (e) 1530 keV, (f) 1593 keV, and (g) 1681 keV. Decay
curves shown in (a-g) were fitted with an exponential decay and linear background.

77
 40
b-472 (a) b-1126 (e)
Counts/50ms

Counts/50ms
30 12

20 8

10 4

0 0
b-663 (b) 80 b-1238 (f)
Counts/50ms

Counts/50ms
30 60
20 40
10 20
0 0
b-726 (c) 20 b-1899 (g)
Counts/50ms

Counts/50ms

20 16
12
10 8
4
0 0
0 500 1000 0 500 1000
Time (ms) Time (ms)
Figure 4.24: Decay curves showing (a-g) fragment-β correlations with an additional
requirement of a coincident gamma ray with an energy of (a) 472 keV, (b) 663 keV,
(c) 726 keV, (d) 1126 keV, (e) 1238 keV, and (f) 1899 keV.

78
Table 4.7: Ratio of peak intensities between Ge spectra without and with back-to-
back decay rejection relative to the increase observed for the 606-keV transition. γ
rays that can be assigned to daughter or grand-daughter transitions show a dramatic
increase in peak intensity.
Energy (keV ) Ratio Source
59
102.0 ± 0.4 1.01 ± 0.08 V
59
207.8 ± 0.4 1.00 ± 0.06 V
59
592.4 ± 0.4 0.99 ± 0.09 V
59
472.9 ± 0.5 1.6 ± 0.6 Mn
59
662.6 ± 0.4 1.4 ± 0.4 Cr
59
1126.5 ± 0.5 1.6 ± 1.8 Cr
59
1238.0 ± 0.4 1.6 ± 0.2 Cr
59
1899.5 ± 0.5 1.5 ± 0.3 Cr

ID gate. The 880- and 1056-keV transitions are still observed following the decay of
59
V. Additionally, the intensity of the 1056-keV transition is higher than that of the
880-keV transition. The 1056-keV transition was also observed in coincidence with
59 59
V γ rays and thus can conclusively be placed as following the decay of V. The

880-keV transition, not in coincidence with any 59 V γ ray, may result from a possible
β-delayed neutron branch into the 2+
1 state of
58
V. If such a branch exists, the Pn
59
value for V would be ∼3%. This value is a lower limit as it is determined based on
the observed γ-ray intensity of the 2+ → 0+ transition in 58
V.

A fit of the fragment-β decay curve was complicated by the relatively poor knowl-
59 59
edge of the daughter, Cr, half-life. The first measurement of the Cr half-life,
740 ± 240 [40, 41], was obtained from an average of the γ-gated half-lives of 1.0 ± 0.4

s and 0.6 ± 0.3 s, for the 1238- and 112-keV γ-rays, respectively, weighted by rel-
59
ative intensities. The second measurement conducted on Cr resulted in a half-life

of 460 ± 46 ms [42]. The inconsistency of these two results made it difficult to fit
59 59
the half-life of the V parent. In an attempt to determined the half-life of Cr the
1238-keV γ-gated decay curve from 0-5 s was fit and the results are shown in Figures

4.25(a-b). When the decay curve in 4.25(a) was fit with no restriction on the half-life
59 59
of the V parent, the deduced half-life for Cr is 1100 ± 200 ms and the fitted half-

79
 (a) 59V T = 84±19 ms (b) 59V T = 97 ms
59Cr T1/2, = 1117±140 ms 59Cr T1/2, = 1050±90 ms

Counts/100ms
Counts/100ms
100 1/2 100 1/2

b-1238 b-1238
59 59
Cr Cr
10 bkgd 10 bkgd

0 2500 5000 0 2500 5000

Time (ms) Time (ms)
Figure 4.25: Decay curve showing fragment-β correlations with an additional require-
ment of a coincident gamma ray with an energy of 1238 keV. Fit includes the expo-
nential growth and decay of the daughter 59 Cr with a linear background. In (a) the
59
V half-life was a variable in the fit, in (b) it was constrained to 97 ms.

T1/2 = 95 ± 3 ms
10000 59
V
Counts/20ms

1000
bkgd
59
Cr 59
100 Mn

0 500 1000
Time (ms)

Figure 4.26: Decay curve showing fragment-β correlations for events occurring within
one second after a 59 V implanted ion. The decay curve is fit with an exponential
decay of the parent, growth and decay of the daughter and grand-daughter and a
linear background term. The half-life of the daughter was set to 1100 ms deduced
from the γ gated decay curve fit of the 1238-keV transition.

59 59
life for V (84 ± 19 ms) agrees, within error, with most of the V γ-gated half-lives.

The decay curve in 4.25(b) was fit, restricting the parent half-life to 97 ms, taken
59
from a weighted average of all γ gated half-lives; a Cr half-life of 1050 ± 90 ms was
deduced. From a weighted average of the two 1238-keV γ ray fits, the half-life of 59 Cr
59
was adopted to be 1100 ± 100 ms. The adopted Cr half-life was then used to fit the
59
V fragment-β decay curve, shown in Figure 4.26, leading to a half-life of 95 ± 3 ms
59
for V. Taking a weighted average of the 95 ± 3 half-life with the γ-gated half-lives

80
 59
resulted in a value of 97 ± 2 ms, adopted for the the V half-life. Previous values of

the half-life include 130 ± 20 ms [36] and 75 ± 7 ms [37] but both studies suffered from
59
a lack of statistics and used an imprecise value for the Cr half-life.
59
γγ coincidence spectra for transitions in V are shown in Figures 4.27, 4.28, and
4.29. The γ rays with energies of 102 and 208 keV had been identified and placed
into a level scheme previously [37] with the 102- and 208-keV transitions in series.

The series placement of these two γ rays, with the lowest excited state at an energy
of 208 keV, was confirmed using the γγ coincidence spectra for 59 V. In particular, the

γ rays at 317, 592, and 707 are present only in the 208-keV γ-gated spectrum, and
the coincidence spectra gated on the 317-, 592-, and 707-keV transitions only show a

coincidence with the 208-keV transition not with the 102-keV transition, placing the
208-keV state at the bottom of the level scheme. The ordering is in contrast to the
recent findings of Ref. [43] where the 102- and 208-keV transitions are reversed in the

level scheme.
The proposed level scheme for 59 Cr populated through the β decay of 59 V is shown

in Figure 4.30. The logft values and β decay branching ratios are not shown due to the
large number of decays that have not been placed into the level scheme. However, it

should be noted that based on the number of β decays observed and the low intensity
of the γ-ray transitions that have been placed directly feeding the ground state there
is a significant β branch to the ground state of 59 Cr. The previous work performed on
59
the isomeric Cr decay established a tentative 9/2+ excited state at 503 keV which
deexcites through an M2 γ emission [39] of 193 keV, then cascades through two γ-

rays with energies of 102 and 208 keV, respectively [37]. The M2 transition from the
9/2+ 503-keV state suggests a Jπ assignment for the 310-keV level of 5/2− . From the

γγ coincidence spectra obtained by gating on transitions feeding the 310 keV state
the efficiency corrected intensity ratio between the 102- and 208-keV transitions is
approximately 60%. The discrepancy from the expected 100% can be resolved if the

102-transition is an E2 transition (E2 internal conversion coefficient = 0.39). This

81
leads to a 1/2− assignment to the 208-keV level and a ground state of 3/2− . The
59
previous suggestion of a 5/2− ground state for V [37] would lead to a large ground
59 59
state branch in the V→ Cr decay which is observed.

59
4.4.1 Cr
59 59
The intensity of V ions also allowed a study of the Cr β decay. As mentioned
59
above, the half-life for the Cr β decay has been adopted as 1100 ± 100 ms following

the analysis of the 1238-keV γ-gated decay curve shown in Figure 4.25. γ-coincidence
spectra have also been extracted for transitions suspected to be in the daughter, 59 Cr,

and are shown in Figure 4.31. The 662- and 1238-keV transitions were found to be in
coincidence. Additionally, if the 112-keV and 1238-keV transitions were in coincidence
there would be a significant peak at 112 keV in the 1238-keV γγ-coincidence spectrum,

which was unobserved. The statistics were too low to identify possible coincidences
between the 662-keV and either the 112- or 1126-keV transitions. The level scheme
59 59
for Mn derived from the β decay of Cr is given in Figure 4.32.

82
 208 (a) 102 (g)

Counts/2keV
Counts/2keV
200 102-keV gate 208 426-keV gate
8
100 4 606
1530
0 0
208 (b) 208 (h)

Counts/2keV
Counts/2keV

606 102-keV gate 102 463-keV gate

20 4
426
491 1031 823
10 977 1056 1530 2

0 0
(c) 208 (i)

Counts/2keV
Counts/2keV

200 102 208-keV gate 15 491-keV gate

10 102
100
5
0 0
(d) (j)
Counts/2keV
Counts/2keV

40 317 592 208-keV gate 20 208 592-keV gate

491 606 1031
20 977 1056 10
372 841 102
708 1222
*
0 0
208 (e) 30 208 (k)
Counts/2keV
Counts/2keV

317-keV gate 606-keV gate

20
20
102
102
10 10
841 426
0 0
(f) 208 (l)
208
Counts/2keV
Counts/2keV

15 372-keV gate 8 708-keV gate

10 102 4
5
0 0
0 1000 2000 0 1000 2000
Energy (keV) Energy (keV)

Figure 4.27: γ − γ coincidence data following a 59 V β decay for the (a)102-, (b)102-,
(c)208-, (d)208-, (e)317-, (f)372-, (g)426-, (h)463-, (i)491-, (j)592-, (k)606-, and (l)708-
keV transitions.

83
 208 (a) 208 (g)

Counts/2keV
Counts/2keV 2 784-keV gate
6
977-keV gate
102 4 102
1
2
0 0
(b) (h)

Counts/2keV
3 12
Counts/2keV

208
800-keV gate 1031-keV gate
2 102
6
1
0 0
208 (c) 208 (i)

Counts/2keV
6
Counts/2keV

4 463 823-keV gate 1056-keV gate

102 4 102 (58V+59V)
2 1222
880
2
0 0
208 (d) 6 208 (j)
Counts/2keV
Counts/2keV

12 841-keV gate 1222-keV gate

102
8 4

4 317 2
0 0
3 (e) 3 (k)
Counts/2keV
Counts/2keV

1056
880-keV gate 1238-keV gate
662
2 (58V) 2
1 1
0 0
207 (f) 102 (l)
3
Counts/2keV
Counts/2keV

2 208
959-keV gate 1530-keV gate
2
1
1
0 0
0 1000 2000 0 1000 2000
Energy (keV) Energy (keV)

Figure 4.28: γ −γ coincidence data for following a 59 V β decay for the (a)784-, (b)800-
, (c)823-, (d)841-, (e)880-, (f)959-, (g)977-, (h)1031-, (i)1056-, (j)1222-, (k)1238-,
(l)1530-keV transitions. Panel (e) and (i) contain contamination from 58 V due to
a particle ID overlap.

84
 1593-keV gate (a) 1 1681-keV gate (b)

Counts/2keV
Counts/2keV 2

0 0
0 1000 2000 0 1000 2000
Energy (keV) Energy (keV)

59
Figure 4.29: γ-γ coincidence data following a V β decay for the (a) 1593-, (b)1681-
keV transitions..

- 0 T = 97±2 ms
(5/2 ) 1/2
Q = 10.8±0.4 MeV
b
59
V
[1 ]
97 .4 [ ]
7. 2.2
93 0.5

]
.4
15 .7 [

2
98

2509
21

56 [0.8 ]
.5
57 1 [2

10 4 [ ]
]
5

42 .8 [ ]
[1 ]
.7
.

5. 2.4
84 0 [2

]
.

2
22

.7
8
12

1532
.

1.

5
30

1366
11
10

59 [1 ]
6. .1]

1341
4
9. [6.

2. 1]

3. 3]
31 0.8 .2]
60 [1

.
49 4 [4
[2
79 0
6
7.

0 ]
9

915
70

[
3

800
7.

8 21]
1]
[
[4

525
20 0

-
2.

(5/2 ) 310
10

7.

- 208
(1/2 )
-
(3/2 ) 0

59
Cr
Figure 4.30: Proposed 59 Cr level scheme populated following the decay of 59 V. The
number in brackets following the γ-ray decay energy is the absolute γ-ray intensity.
The Q value was deduced from data in Ref. [32]. Observed coincidences are represented
as filled circles.

85
 5
Counts/2keV 662-keV (a) 3 1238-keV (b)

Counts/2keV
4
3 662
2
1238
2
1
1
0 0
0 1000 2000 0 1000 2000
Energy (keV) Energy (keV)

Figure 4.31: γγ coincidence spectra for two of the transitions suspected of belonging to
the decay of 59 Cr. The 1238- and 662-keV transition are in coincidence. A coincidence
between the 1238 and 112 keV transitions, if present, should be observable in the 1238-
keV gated spectrum based on the intensity of the 662 keV peak. Intensities were too
weak to determine coincidences between 662- and either 112- or 1126- keV transitions.

-
(3/2 ) 0 T1/2 = 1100±100 ms
59 Q = 7.6±0.3 MeV
b
Cr
]
2
[2

]
[9
.5

6
99

2.

1900
66
18

00]
]
[1

[6
0

.5
8.
26
3

1238
12

11

0 ]
[1
8
1.

112
11

0
59
Mn
Figure 4.32: Proposed 59 Mn level scheme populated following the decay of 59 Cr. The
number in brackets following the γ-ray decay energy is the relative γ-ray intensity. The
Q value was deduced from data in Ref. [32]. Observed coincidences are represented
as filled circles.

86
 58
V
1200 348 60
410 745 Fe

Counts
800 57
V
59
Mn 758
58
400 V 60Fe 59
Cr 1532
0
0 500 1000 1500 2000
Energy (keV)

Figure 4.33: β-delayed γ-ray spectrum for those decays observed within 5 s of the
implantation of a 60 Cr ion.

Table 4.8: 60 Cr β-delayed γ-rays with absolute γ-ray intensities and initial and final
states for those transitions placed into the 60 Cr level scheme.
E(keV) Iγabs (%) Initial State (keV) Final State (keV)
348.6 ± 0.4 6.4 ± 0.5 349 0
410.1 ± 0.4 5.2 ± 0.4 759 349
746.0 ± 0.6 0.9 ± 0.4
758.2 ± 0.4 5.0 ± 0.5 759 0

60
4.5 Cr
60
The β-delayed γ-ray spectrum for events that occurred within 5 s after a Cr im-
plantation event are shown in Figure 4.33. The correlation time was extended due to
60
the longer half-life of Cr. Observable in Figure 4.33 are numerous lines that belong
to the decay of longer lived species, as seen from the βγ spectra with different cuts
60
on correlation time shown in Figure 4.34. The remaining γ rays, attributed to Cr,
are listed, along with their absolute γ-ray intensities, in Table 4.8.
60
The decay curve for those events that occurred within 5 s of a Cr implantation
is shown in Figure 4.35(a). From the shape of the decay curve below 1000 ms, it is
apparent that the daughter, 60 Mn, has a shorter half-life than 60 Cr. Fitting the decay

curve in Figure 4.35 with a parent exponential decay, an exponential growth and
60
decay of the daughter, and a linear background, a half-life of 490 ± 10 ms for Cr is

deduced. The 60 Cr half-life agrees favorably with previous values of 510 ± 150 ms [42]

87
 80
(a)
Counts/1keV

60
40
20
0
59 (b)
V 60
Fe
Counts/1keV

57 59
Ti Mn
20 58
Cr 58
V
10 59
Cr

0
59 (c)
V 57Ti
Counts/1keV

60
57 Fe
V
10
58
V 59Cr

0
0 500 1000 1500 2000
Energy (keV)

Figure 4.34: 60 Cr β-delayed γ spectra. β delayed γ ray spectrum for those decays
observed within (a)0-1 s, (b) 4-5 s, and (c) 9-10 s following a 60 Cr implanted ion.

88
and 570±60 ms [41]. The γ-gated decay curves are shown in Figures 4.35(b-d) and are

consistent with the 490 ± 10ms half-life extracted from Figure 4.35(a). In the fit, the
half-life of the daughter was considered a parameter, resulting in a value of 280 ± 20
60
ms for the Mn half-life, in contrast to previous measurements in Refs. [41, 44, 45].
A half-life of 1.8 ± 0.1 s was deduced for a 3+ isomeric state in 60
Mn at an energy
of 272 keV [44, 45]. The half-life of the 0+ 60
Mn ground state was first determined to
60
be 51 ± 6 s [41] and it was suggested that the Mn ground state β decay proceeded
directly to the 60 Fe ground state. Later studies of 60 Mn suggest that 0+ is the correct

Jπ assignment for the 60

Mn ground state but, due to contaminations from higher
60
masses, the 51 ± 6 s half-life of Mn is probably incorrect [46].
60
γγ coincidences following Cr decay are shown in Figure 4.36. The level scheme
determined from the γγ coincidence data is shown in Figure 4.37. From the absolute
60
intensities there appears to be a large ground state branch to Mn.
60
The ground state of Cr should be 0+ . Therefore, the decay into 60
Mn should
only populate low spin states and a significant fraction would be expected to proceed

to the 0+ 60
Mn ground state, which is observed. Similarly, β decay from the 60
Mn 0+
ground state would be expected to populate low-spin states in 60 Fe. γ-ray transitions

at 823 and 1150 keV match known transitions in 60 Fe. The 1532-keV transition could
also be a transition between the second excited 0+ at 2358 ± 3 keV and the 823 keV
2+ state in 60
Fe, but such a transition has not been observed in previous β decay
60 60
work. The β decay into levels of Fe from Mn is shown in Figure 4.38 The γ-gated
decay curves for the 823-, 1150-, and 1532-keV transitions are shown in Figures 4.35(e-

g). The decay curves in Figures 4.35(e,f) were fitted with an exponential daughter
60 60
growth and decay, fixing the half-life of Mn at 280ms. The resulting Cr half-life

was 450 ± 40 and 320 ± 30 for the 823- and 1150-keV transitions, respectively. It was
not possible to fit the 1532-keV decay curve.

89
 10000 b (a) 100 (e)
b-823
60 60

Counts/200 ms
Cr T = 490±10 ms 60
Counts/50 ms
Cr T =450±40 ms
60Mn T1/2 = 280±20 ms 60Mn1/2
1/2 T1/2= 280 ms
40

1000 20

0
b-349 (b) b-1150 (f)
100 60

Counts/200 ms
Counts/200 ms

T1/2 = 500 ± 40 ms Cr T =
320±30 ms
80 20 60Mn1/2
T1/2= 280 ms
60
40 10
20
0 0
b-410 (c) b-1532 (g)
60 Counts/200 ms
Counts/200 ms

T1/2 = 590 ± 60 ms 8
40 6
4
20
2
0 0
b-758 (d) 0 1000 2000 3000 4000 5000
Counts/200 ms

50 T1/2 = 430 ± 50 ms
Time (ms)
40
30
20
10
0
0 1000 2000 3000 4000 5000

Time (ms)
Figure 4.35: Decay curves for 60 Cr showing (a) fragment-β correlations where the
data were fitted with an exponential parent decay, an exponential growth and decay
for the daughter 60 Mn and a linear background, (b-d) fragment-β correlations with
an additional requirement of a coincident gamma ray with an energy of (b) 349 keV,
(c)410 keV and (d) 758 keV, (e-g) fragment-β correlations of transitions following the
β decay of 60 Fe with an additional requirement of a coincident γ ray at an energy of
(e) 823 keV, (f) 1150 keV, and (g) 1532 keV. Decay curves shown in (b-d) were fitted
with an exponential decay and linear background. Decay curves shown in (e-f) were
fitted with an exponential growth and decay and linear background.

90
 16
410 (a) 758-keV gate (c)
Counts/2keV

Counts/2keV
12 349-keV gate 1
8
4
0 0
(b) 0 1000 2000
Counts/2keV

349
8 410-keV gate Energy (keV)

0
0 1000 2000
Energy (keV)

Figure 4.36: γγ coincidences for the (a)349-, (b)410-, and (c)758-keV transitions fol-
lowing the β decay of 60 Cr.

(0+) 0
T1/2 = 490 ± 10 ms
60 Qb = 6.7 ± 0.3 MeV
Cr
Ib(%) log ft
0]

2]
.

.
[5

[5
2

1
8.

0.

4]

759
75

41

.
[6

5.0±0.5 4.3±0.1
6
8.
34

349
1.22±0.6 4.4±0.1
(0+) 0
88.6±0.6 4.5±0.2
60
Mn
Figure 4.37: Level scheme deduced following the β decay of 60 Cr into excited levels
in 60 Mn. β-decay branching ratios along with logft values calculated using Qβ from
Ref. [33].

91
 (0+) 0
T1/2 = 280 ± 20 ms
60 Qb = 8.2 ± 0.1 MeV
Mn

4]
[2
.4
32

]1
0+

[4
15
2356

.2
50
0+

11
1974

0 0]
[1
4
3.
2+
82

823

0+ 0
60
Fe
Figure 4.38: Levels in 60 Fe populated following the β decay of 60 Mn. Relative γ-
ray intensities given in brackets following the γ decay energy. Qβ determined from
Ref. [33].

92
Chapter 5

Discussion

At present, complete ab initio calculation of nuclei are only possible for the lightest

nuclei, A ≤ 17. Beyond that the number of shell model orbitals and the amount of
calculations are prohibitively large. One of the bases of shell model calculations is

the presence of shell closures, allowing the computational problem to be divided into
two parts; an inert core containing most of the nucleons coupled to 0+ and a small

number of valence nucleons outside the core constrained to a certain range of orbitals
called the model space. The properties of the nucleus are then determined by the
valence nucleons. Knowledge of the location of shell closures becomes important for

the study of exotic regions of nuclei. As the locations of the traditional shell closures
at 2,8,20,28,50,82, and 126 are based primarily on experiments performed with stable

nuclei their extrapolation to exotic regions can be questioned. With the development
of radioactive nuclear beams it has become possible to search for the appearance or
disappearance of shell closures far from stability and numerous recent results suggest

that shell closures, particularly in the neutron-rich region, may be dynamic [12,47–50].
A variety of different effects may result in so-called shell quenching. The proton-

neutron monopole interaction discussed in the Introduction is such an effect and was
already demonstrated in near-stable nuclei using the proton single-particle states in

Sb (see Figure 1.3). While the monopole interaction is strong between orbits of similar

93
 (a) (b)
48 56
Sc Sc
f5/2 f5/2 f5/2 f5/2
p1/2 p1/2 p1/2 p1/2
p3/2 p3/2 p3/2 p3/2

f7/2 f7/2 f7/2 f7/2

proton neutron proton neutron

Figure 5.1: The large spin-orbit splitting prevents spin-orbit partners from being oc-
cupied simultaneously as in (a) for 48 Sc. Far from stability the spin-orbit partners will
both be occupied, shown in (b) for 56 Sc, leading to strong proton-neutron monopole
interactions.

l the most dramatic reordering of single-particle states should occur when the orbital

angular momentum, l, is the same for both protons and neutrons [7]. Due to the
large spin-orbit splitting, this condition will be satisfied in neutron-rich nuclei far
from stability (see Figure 5.1). Dramatic changes in effective single-particle energies

due to strongly attractive interactions between spin-orbit coupling partners has been
discussed by Otsuka et al. [10] in the context of the disappearing N = 8, 20 magic

numbers for neutron-rich exotic nuclei.

The migration of the neutron effective single-particle energy due to the monopole

interaction as a result of the filling of proton single-particle levels can be calculated

using the equation below [51]:

X
ε̃jν = εjν + hjν jπ |VM |jν jπ iνjπ (5.1)
jπ

where ε̃jν is the effective neutron single-particle energy, εjν is the ’bare’ single-particle
energy, and hjν jπ |VM |jν jπ i is the two body interaction between protons in state jπ

and neutrons in state jν , and νjπ is the occupancy of the proton orbit. The potential,
VM , in equation 5.1 is attractive in nature serving to lower the energy of the neutron

single-particle state and is linearly dependent on the occupancies of the proton orbits.
Evidence of a significant reordering in single-particle states can be seen in Figure 5.2

94
 (b)

(a)

Figure 5.2: The proton-neutron monopole interaction between (a) the νg7/2 -πg9/2 and
(b) νf5/2 -πf7/2 . In both cases the reduction in energy of the neutron level resulting
from the filling of the proton single-particle state is ∼3 MeV.

where a large drop in the energy of the 7/2+ state is observed as protons are added to

the g9/2 single-particle level from Zr to Sn. Information on the spectroscopic factor (S
factor) provides the link between the single-particle state, νg7/2 , and the observed Jπ

of the low-energy level. The S factor describes the fraction of a particular state that is
derived from a specific nuclear configuration and can vary between 0 and 1. The drop

of the 7/2+ level observed in Figure 5.2 is attributed to the monopole interaction
between the πg9/2 -νg7/2 levels. Although the single-particle level density is high in

95
this mass region the drastic change in single-particle energies results in the formation

of a shell gap at N = 56, evidenced by the high 2+

1 energy in
96
Zr [52] compared to
other Zr isotopes, that subsequently is washed out as the πg9/2 state is filled, lowering

the νg7/2 level.

In regions of lower single-particle level density, such as the pf shells, the substantial
reordering of single-particle states due to the proton-neutron monopole interaction

make neutron-rich nuclei near N = 28 ideal candidates for the emergence of new shell
structure. Between the Ca and Ni isotopes, as protons are added to the f7/2 single-

particle state the monopole interaction between πf7/2 and νf5/2 results in a decrease
in the energy of the f5/2 single-particle state, resulting in a reduced energy of the

5/2− state, see Figure 5.2(b). As protons are removed from the f7/2 single-particle
level the neutron f5/2 state increases in energy and could result in the formation of N
= 32 and N = 34 subshell closures in the Ca and Ti isotopes due to the large energy

separation between the p3/2 ,p1/2 , and f5/2 neutron single-particle states.
The low energy structure of the even-even neutron-rich isotopes beyond the N =

28 shell closure have been investigated to follow the evolution of possible new subshell
closures at N = 32,34. As previously mentioned, the systematic variation of the E(2+ )

in even-even nuclei gives an indication of shell closures. Along with the E(2+ ) energies,
the branching ratios of the β decay of the odd-odd parent into the even-even daughter
allows the inference of a tentative spin and parity assignment to the parent ground

state. The ground state Jπ assignments can be used to track the monopole migration
of the νf5/2 level as a function of protons in the f7/2 state which can be compared to

with shell model predictions to infer single-particle level ordering. With the current
experimental data, the E(2+ ) values for the N = 34 22 Ti and 24 Cr isotopes can be

compared with theoretical predictions to gauge the possible presence of an N = 34

subshell closure.

96
 r
4

Figure 5.3: E(2+ ) as a function of neutron number for the 24 Cr, 22 Ti, and 20 Ca isotopes.
The circles connected by lines represent the GXPF1 interaction. Crosses connected
by lines correspond to KB3G interaction and the black bars are experimental points.
GXPF1 does well at reproducing the N = 32 subshell closure in the Cr, Ti, and Ca
isotopes while KB3G does not predict an N = 32 subshell closure in the Cr isotopes.
GXPF1 predicts a shell closure at N = 34 in the Ti and Ca isotopes evidenced by the
peak in E(2+ ) values.

5.1 Comparison with Shell Model Results

Shell model calculations employing the GXPF1 [1,14] and KB3G [53] interactions are
compared to the experimental energies of the first excited 2+ state of the even-even

24 Cr, 22 Ti and 20 Ca isotopes beyond the N = 28 shell closure in Figure 5.3, and were
presented in Ref. [28]. The evolution of the N = 32 subshell gap, as described by

the GXPF1 interaction, arises from the gradual migration of the νf5/2 orbital, due to
the proton-neutron monopole interaction, with the removal of protons from the f7/2
leading to a increasing energy gap between the νf5/2 and νp3/2 orbitals. The widening

97
 3200

E(4+)
2400

Energy (keV)
1600

800 E(2+)
N = 32
0
20 22 24 26 28 30 32
Proton Number (Z)

Figure 5.4: E(2+ ) for the N = 32 isotones showing the rise in energy that is expected
to parallel the increasing energy separation between the neutron f5/2 and p3/2 single-
particle states.

energy gap between the νf5/2 and νp3/2 can be seen as well in the increasing energy

of the E(2+ ) in the N = 32 isotones as protons are removed from the nucleus, see
Figure 5.4. The GXPF1 calculations also do well in reproducing the high-spin states

observed in various neutron-rich Ti isotopes [13]. The GXPF1 and KB3G calculations
diverge concerning the possible subshell closure at N = 34 in the Ti and Ca isotopes.
The GXPF1 interaction predicts that as the last few protons are removed from the

f7/2 state the νf5/2 single-particle state continues to increase in energy resulting in a
significant energy gap between the νp1/2 and νf5/2 orbitals, and a new magic number

at N = 34 for the 22 Ti and 20 Ca isotopes. A first indication of the subshell closure at

N = 34 would be found in the systematic variation in E(2+
1 ) in Ti and Ca isotopes.

Unfortunately, the energy of the 2+ state in 56

Ti, only 1129 keV, does not bear out
this possibility; falling midway between the GXPF1 and KB3G predictions [28].
Shell model calculations using the GXPF1 interaction were carried out and are
56
compared to the β-decay properties of Sc, along the lines of similar investigations
for the neutron-rich V [26] and Ti [30] nuclides, and were presented in Ref. [31].

The calculations were performed using the codes OXBASH [54] and CMICHSM [55].
56
Both high- and low-spin β-decaying states have been observed in Sc. The low-spin
56
state decays primarily to the ground state of Ti, with a measured branching ratio
of (65 ± 13)%. The results of shell model calculations using the GXPF1 interaction

98
 2056 2+ 2053 3+
1891 4+
1745 5+
1681 2+
1676 6+
1509 3+ 1522 3+
1505 5+
1444 4+
1275 4+

1070 1+

543 6+
490 2+

123 3+
0 4+ 0 1+

54Sc 56Sc
54,56
Figure 5.5: Low-energy level structure of Sc from shell model calculations using
the GXPF1 interaction.

99
 56
predict a 42% β-branch for the Sc 1+ decay to the ground state of 56
Ti. Based

on direct feeding to the first 6+ state in the daughter 56

Ti, the spin and parity of
56
the high-spin β-decaying state in Sc has been tentatively assigned as (6+ , 7+ ). The
56
shell model calculations performed predicted a 68% β-decay branching for the Sc
6+ decay to the first excited 6+ state in 56
Ti, and this was the only direct feeding to
56
excited states in Ti below 4 MeV.

The calculated β-decay half-lives for the 1+ and 6+ states were 16 ms (35 ± 5 ms)
and 110 ms (60 ± 7 ms), respectively, where the corresponding experimental numbers

are given in parenthesis. For both calculations, a Qβ value of 13.7 MeV [32] was used,
as the energy separation between and order of the high- and low-spin β-decaying states

is unknown. The calculated half-lives were corrected by a factor of two to account for
the reduction in calculated Gamow-Teller strength for neutron-rich nuclides in this
56
region. The experimental half-lives for the two β-decaying states in Sc fall within a

factor of two of the shell model results.

56
The presence of two β-decaying states in Sc can be described by studying the
54,56
change in level structure between Sc. The calculated low-energy level structures
54,56 54
for odd-odd parents Sc are shown in Figure 5.5. The ground state of Sc is

predicted to have spin and parity 4+ , and is separated by only 123 keV from the
first excited state, which is calculated to have J π = 3+ . A very different low-energy
56 56
structure is expected for Sc. The ground state of Sc is predicted to have spin and

parity 1+ , and a doublet of states with J π = 2+ , 6+ is calculated at an energy of

∼ 500 keV above the ground state. The following discussion focuses on the significant

changes to the low-energy structure and the β-decay properties of these nuclides that
54 56
result from the addition of two neutrons in going from Sc to Sc.

The presence of an E2 isomeric transition with energy 110 keV in 54 Sc [31] provides
an insight into the low-energy structure of the 54 Sc parent. The calculated position of
the first 1+ level in 54 Sc, which has composition (πf7/2 )1 (νp3/2 )4 (νf5/2 )1 , compared to

the ground state doublet, which is most likely (πf7/2 )1 (νp3/2 )4 (νp1/2 )1 , furthers the

100
understanding of the relative position and energy separation of the νf5/2 and νp1/2

single-particle orbitals. If the monopole migration of the νf5/2 is overestimated in the

GXPF1 as suggested in Ref. [28], the 1+ state may reside much closer to the ground

state, and could give rise to the observed E2 isomeric transition.

56 54
Sc differs from Sc by two neutrons, and the unpaired neutron is expected to
occupy the νf5/2 orbital. A 1+ ground state is predicted for 56 Sc from the shell model
54
calculations employing the GXPF1 interaction. Similar to the situation in Sc, the
separation between the νf5/2 and νp1/2 single-particle states is expected to be smaller

than that reflected in the GXPF1 results shown in Figure 5.5. A condensed low-energy
56
spectrum for Sc may produce several isomers, which are observed experimentally.
56
The three isomeric γ rays observed in Sc have not been placed. The low statistics
in the current experiment did not permit the analysis of coincidence relationships
between the three γ rays. Additional data on the time evolution of each of the isomeric
56
γ rays would also help to place these transitions in the level structure of the Sc
parent.
57 57
Calculations of the β decay of Ti to low energy levels in V have also been
57 57
carried out using the GXPF1 interaction and compared with the Ti → V decay.

GXPF1 calculations predicted a ground state Jπ of 5/2− for 57

Ti with a first excited
57
state, at 422 keV, assigned as a 1/2− level. The low energy level structure of V
was also calculated, resulting in a ground state of 7/2− with a 5/2− excited state
57
at an excitation energy of 95 keV. The calculated β decay of Ti, using a 5/2−
57
ground state, to levels in V, presented in Figure 5.6, shows a large ground state

branch of 44% which compares favorably with the measured value of 54±3%. The β
decay calculations also predict a significant amount of feeding to four states above

1500 keV which was also observed in the experimental level scheme. Considering that
57
the lower detection limit for absolute intensities in the Ti β decay was ∼0.7%, the
experimental and calculated level scheme agree quite well and suggest that the ground
57 57
state spin and parities of Ti and V are 5/2− and 7/2− , respectively.

101
 - -
(5/2 ) 0 T =98±5 ms 5/2 0 T =47 ms
1/2 1/2
57Ti Q = 10.6±0.6 MeV 57Ti Q = 10.6 MeV
b b
Ib(%) Ib(%)

2.5 7/2- 2494

11.3 5/2- 2272
7.3±0.7
2475 0.6 3/2- 2251
0.2 7/2- 2061
2036 4.9 5/2- 2040
16±2 3/2-
0 1934
16±2 1754 8.2 7/2- 1871
1.1±0.7 4.7 5/2- 1749
1732
0.1 5/2- 1665

529
0 3/2-

5±3 175 5 5/2- 95

0 113
54±3 (7/2)- 0 44 7/2- 0
57V 57V

experimental theoretical
Figure 5.6: The experimental decay scheme for levels in 57 V populated following the β
decay of 57 Ti are shown on the left hand side of the figure. The theoretical calculations
are shown on the right hand side of the figure and assume a 57 Ti ground state of 5/2− .
There is overall very good agreement between the calculated and observed branching
ratios, especially for the ground state branch

102
 56 1+ 57 58 1+
np1/2 np1/2
V nf5/2 nf5/2
p(f7/2)3 32 32
np3/2 np3/2
55 (5/2-) 56 57 (5/2-)
Ti np1/2 np1/2
p(f7/2)2 nf5/2 nf5/2
32 32
np3/2 np
54 (3,4)+ 55 56 (1)+
nf5/2 nf5/2
Sc
np1/2 np1/2
p(f7/2)1 32 32
np3/2 np3/2
33 34 35

Figure 5.7: Spin and parities of odd-odd and odd-A nuclear ground states near
N=32,34

A tentative 3/2− ground state assignment has been previously put forward for 57 V
57
based on the characteristics of the β decay into levels of Cr [35], most importantly,
57 57
a large ground state β branch of 55%. Later, the V→ Cr ground state branch
was remeasured and resulted in a value of 21% [26]. A comparison of the 57 V β decay
characteristics with GXPF1 shell model calculations assuming a 3/2− , 5/2− , and
57 57
7/2− ground state for V found the 3/2− V ground state assignment was unable
57
to reproduce the large ground state branch to Cr. Calculations instead favored a
57
5/2− or 7/2− assignment to the ground state of V, in agreement with the present
57
tentative 7/2− V ground state.

5.2 Monopole Migration of νf5/2 Based on Ground-

State Spin and Parity Assignments

54,56 55,57 56,58
Using the tentative spin and parity assignments for the Sc, Ti, and V
isotopes, the monopole migration of the νf5/2 state with removal of protons from the
πf7/2 level can be qualitatively described. The spins and parities for the ground states

of Sc, Ti, and V nuclei around the N = 32 subshell closure, together with schematic
neutron single-particle levels, are shown in Figure 5.7. Working from the N = 32

103
subshell closure, the extreme single-particle model was used to infer the proton and

neutron orbitals that couple to produce the observed ground state spins and parities.
Starting in the 21 Sc isotopes, the odd proton is located in the πf7/2 state, as

evidenced by the ground state spins and parities of odd-A Sc isotopes [31]. Coupling
this proton to an odd neutron in the νf5/2 orbital results in a range of spin and parity
of (1-6)+ , while coupling an f7/2 proton to an odd neutron in the νp1/2 state would

lead to spin and parity of 3+ or 4+ . The ground state spin and parity of 54 Sc has been
tentatively identified as (3,4)+ . Nordheim rules [56, 57] argue against the coupling of

a proton in the πf7/2 to a neutron in the νf5/2 level resulting in a (3,4)+ ground state,
suggesting that the 33rd neutron occupies the νp1/2 orbital in agreement with shell

model calculations. The tentative spin and parity of 1+ in 56

Sc in a single-particle
picture is only possible with the placement of the 35th neutron in the νf5/2 level. From
the schematic neutron levels, it is seen that the νf5/2 level is located above the νp1/2

state in the 21 Sc isotopes.

56,58
In the 23 V isotopes, the odd proton is still located in the πf7/2 orbit. Both V

have been assigned 1+ ground states, placing the both the 33rd and 35th neutrons in
the νf5/2 state and suggesting that the νf5/2 orbital is lower than the νp1/2 level in

23 V nuclei.
55
The β decay of the odd-even nucleus Ti has also been studied [30] and, while
the spin and parity have not been firmly established, the complex feeding observed in
55
the β decay of Ti is not indicative of a dominant νp1/2 single-particle configuration
55 55
for the Ti ground state. The results reported on the β decay of Sc into the low-

energy levels of 55 Ti suggest a significant β-decay feeding the ground state of 55 Ti [31].
Assuming J = 7/2 for the 55 Sc parent state, a β-decay branch directly to the daughter
55
ground state also suggests that the ground-state spin of Ti is greater than 1/2,
limiting the J π to (9/2, 7/2, 5/2)− . In an extreme single-particle model, only a 5/2−
assignment is possible by filling the odd neutron into the f5/2 state.
57
The inclusion of the β decay of Ti completes the picture. A tentative spin and

104
parity assignment of 5/2− to the ground state is based on favorable comparisons

to theoretical shell model calculations in the previous section, most importantly the
57
large branch to the ground state of V. In a single-particle picture, to obtain a 5/2−

spin and parity the odd neutron should be placed in the f5/2 single-particle level.
55,57
Thus Ti suggest that the νf5/2 state is lower in energy than the νp1/2 state in
55
agreement with observations in Ti.

While the νf5/2 and νp1/2 single-particle level orderings, inferred from the spin
54,56 56,58
and parity assignments to the ground states of odd-odd Sc and V isotopes,

agree with shell model calculations, the present data on the low-energy levels of the
neutron-rich Sc and Ti isotopes do not support a large energy separation between

the νf5/2 and νp1/2 orbitals predicted using the GXPF1 interaction. As noted in
Ref. [28], a reduction of ≈ 0.8 MeV in the single-particle energy of the νf5/2 state
would account for the absence of an N = 34 shell closure. To preserve calculated

levels in other nuclei, a weaker monopole interaction should be considered to account

for the slower rise in energy of the νf5/2 state with the removal of protons from the

πf7/2 orbit [1] and a location of the first 2+ state in 54

Ca around 4 MeV. However,
the monopole migration of the νf5/2 state is expected to continue with the removal of

the last two protons from the πf7/2 orbital in the 20 Ca isotopes. Further investigation
of the low-energy level structure of neutron-rich Ca nuclei is warranted to determine
if a νf5/2 -νp1/2 energy gap dominates at N = 34, resulting in a shell closure.

5.3 Summary

The β decay of (a large number) of neutron-rich nuclides was studied. The half-

lives, branching ratios, and γ feeding patterns were determined for these nuclides.
56
Among the most interesting, the low-energy levels of neutron-rich Ti was studied

through the β decay of the parent nuclide 56 Sc, respectively. The systematic variation
in E(2+
1 ) for the even-even 22 Ti isotopes reveal a peak at N = 32 indicative of a

105
subshell closure, similar to the increase in E(2+
1 ) at N = 32 observed for the 20 Ca

and 24 Cr isotopes. Evidence for a shell closure at N = 34, predicted for the Ti
isotopes by shell model calculations carried out using the new pf -shell interaction

GXPF1, were not substantiated. Based on β-decay branching ratios, tentative spin
and parity assignments were made for the odd-odd parent nuclei. A survey of the
spin-parity assignments of the ground states of the odd-A and odd-odd nuclides in

this region was used to analyze, in a schematic way, the monopole migration of the
νf5/2 orbital. While there is evidence that the νf5/2 − νp1/2 level ordering changes

between 23 V and 21 Sc, there is no substantial proof that a large energy gap between
the νf5/2 and νp1/2 develops for the 22 Ti isotopes. Support for this conclusion is

found in the systematic variation of E(2+

1 ) for the even-even Ti isotopes and the

54,56
presence of isomeric transitions in Sc, which suggest a more compressed low-
energy spectrum than predicted for these nuclides by the shell model results reported

here using the GXPF1 interaction. It is important to continue to track the progression
of the monopole migration to the 20 Ca isotopes, which have no f7/2 protons. The “full”

monopole shift of the νf5/2 orbital may lead to a large gap in the νf5/2 −νp1/2 effective
single particle energies, producing a shell closure at N = 34 for the Ca isotopes.

106
Chapter 6

Conclusions and Outlook

48
The shell structure of neutron-rich nuclei above Ca has been investigated through

the β decay of odd-odd and odd-A nuclides. Fast fragments from the NSCL facility
were implanted in the DSSD and correlated with subsequent β decays on an event-by-

event basis. Delayed γ rays, observed using twelve detectors from SeGA, have been
56 57 58,59
observed. Half-lives and level structures were determined for Sc, Ti, V and
60
Cr, representing a significant improvement in the knowledge of nuclear properties
in this neutron-rich region. Tentative spin and parity assignments have been given,
at a minimum, for the ground states of the studied nuclides. The possibility of an

N = 34 shell closure was investigated. Shell model calculations, using the GXPF1
effective interaction, predicted that the strong proton-neutron monopole interaction

between the spin-orbit partners πf7/2 -νf5/2 would result in a gradual increase in the
single-particle energy of the νf5/2 level as protons were removed from the πf7/2 .
Moreover, the rise in energy of the neutron f5/2 would lead to the development of an

N = 32 subshell closure and an N = 34 subshell closure in the Ti and Ca isotopes.

While the N = 32 subshell closure has been identified experimentally, the E(2+ ) value
56
for 22 Ti34 was ∼400 keV lower than theoretical predictions, suggesting the absence
of a subshell closure at N = 34 in the Ti isotopes. The spins and parities of the

neighboring odd-odd nuclei were analyzed in an extreme single-particle framework

107
assuming N = 32 is a good subshell closure. The results suggest that the f5/2 has risen

in energy above the p1/2 level as expected but there is not a large energy separation
between the two orbitals to bring about a neutron subshell closure at N = 34. Further
54,56
experiments on the neutron-rich Ca isotopes, in particular Ca are warranted to
determine if the subshell closure is present with the complete removal of protons from
the f7/2 state. However, the study of these very-neutron rich nuclides through β decay

may have to wait for the next generation accelerator facility due to the extremely
low production cross section. It may be possible to investigate these nuclei through

Coulomb excitation or possibly deep-inelastic reactions. The monopole interaction

might have a role in the evolution of shell structure in heavier neutron-rich nuclei. For

example, the filling of the νg9/2 level following the N = 40 subshell closure may alter
the single-particle energy of the πf5/2 and πg9/2 states possibly affecting shell structure
78
in the vicinity of Ni [58]. Additionally, the proton-neutron monopole interaction

between the spin-orbit partners πg9/2 − νg7/2 and πh11/2 − νh9/2 may affect the energy
of the νg7/2 and νh9/2 single-particle states, disturbing the energies of the N = 82,126

shell closures [10].

108
Bibliography

[1] M. Honma, T. Otsuka, B.A. Brown, and T. Mizusaki. Phys. Rev. C, 69:034335,
2004.

[2] http://www.nndc.bnl.gov/. .

[3] M.G. Mayers. Phys. Rev., 75:75, 1949.

[4] O. Haxel, J. Hans, D. Jensen, and H.S. Suess. Phys. Rev., 75:1776, 1949.

[5] K. S. Krane. Introductory Nuclear Physics. John Wiley & Sons, Inc., New York,
1988.

[6] R.A. Meyer, I. Morrison, and W.B. Walters. Proton-neutron interactions in the
a = 100 nuclides. In Exotic Nuclear Spectroscopy. DNCT, American Chemical
Society, 1990.

[7] P. Federman and S. Pittel. Phys. Lett. B, 69:385, 1977.

[8] L. Grodzins. Phys. Lett., 2:88, 1962.

[9] A. Huck, G. Klotz, A. Knipper, C. Miehe, C. Richard-Serre, G. Walter, A. Poves,

H.L. Ravn, and G. Marguier. Phys. Rev. C, 31:2226, 1985.

[10] T. Otsuka, R. Fujimoto, Y. Utsono, B.A. Brown, M. Honma, and T. Mizusaki.

Phys. Rev. Lett., 87:082502, 2001.

[11] X.L. Tu, X.G. Zhou, D.J. Vieira, J.M. Wouters, Z.Y. Zhou, H.L. Seifert, and
V.G. Lind. Z. Phys. A, 337:361, 1990.

[12] J.I. Prisciandaro, P.F. Mantica, B.A. Brown, D.W. Anthony, M.W.Cooper,
A. Garcia, D.E. Groh, A. Komives, P.A. Lofy, A.M. Oros-Peusquens, S.L. Tabor,
and M.Wiedeking. Phys. Lett. B, 510:17, 2001.

[13] R.V.F. Janssens, B. Fornal, P.F. Mantica, B.A. Brown, R. Broda, P. Bhat-
tacharyya, M.P. Carpenter, M. Cinausero, P.J. Daly, A.D. Davies, T. Glas-
macher, Z.W. Grabowski, D.E. Groh, M. Honma, F.G. Kondev, W. Krolas,
T. Lauritsen, S.N. Liddick, S. Lunardi, N. Marginean, T. Mizusaki, D. Sew-
eryniak, H. Schatz, A. Stolz, S.L. Tabor, C.A.Ur, G. Viesti, I. Wiedenhover, and
J. Wrzesinski. Phys. Lett. B, 546:55, 2002.

109
[14] M. Honma, T. Otsuka, B.A. Brown, and T. Mizusaki. Phys. Rev. C, 65:061301R,
2002.

[15] T. Glasmacher. Annu. Rev. Nucl. Part. Sci., 48:1, 1998.

[16] P.G. Hansen and J.A. Tostevin. Annu. Rev. Nucl. Part. Sci., 53:219, 2003.

[17] Q.R. Ahmad and SNO Collaboration. Phys. Rev. Lett., 89:011301, 2002.

[18] Y. Fukuda and Super Kamiokande Collaboration. Phys. Rev. Lett., 81:1562,
1998.

[19] S. S. M. Wong. Introductory Nuclear Physics. John Wiley & Sons, Inc., New
York, 1998.

[20] G. Friedlander, J. W. Kennedy, E. S. Macias, and J. M. Miller. Nuclear and

Radiochemistry, 3rd ed. John Wiley & Sons, Inc., New York, 1981.

[21] D. J. Morrissey, B.M. Sherrill, M. Steiner, A. Stolz, and I. Wiedenhoever. 204:90,

2003.

[22] W. F. Mueller, J.A. Church, T. Glasmacher, D. Gutknecht, G. Hackman, P.G.

Hansen, Z. Hu, K.L. Miller, and P. Quirin. Nucl. Instr. and Meth. A, 466:492,
2001.

[23] J.I. Prisciandaro. Beta Decay Studies of 69 Ni and 58 V: Development of subshell

gaps within the N = 28 - 50 shell. PhD thesis, Michigan State University, 2001.

[24] J.I. Prisciandaro, A.C. Morton, , and P.F. Mantica. Nucl. Instr. and Meth. A,
505:140, 2003.

[25] G. F. Knoll. Radiation Detection and Measurement. John Wiley & Sons, Inc.,
New York, 2000.

[26] P.F. Mantica, A.C. Morton, B.A. Brown, A.D. Davies, T. Glasmacher, D.E.
Groh, S.N. Liddick, D.J. Morrissey, W.F. Mueller, H. Schatz, A. Stolz, S.L. Ta-
bor, M. Honma, M. Horoi, and T. Otsuka. Phys. Rev. C, 67:014311, 2003.

[27] W.T. Milner. Oak Ridge National Laboratory, unpublished.

[28] S. N. Liddick, P.F. Mantica, R.V.F Janssens, R. Broda, B.A. Brown, M.P. Car-
penter, B. Fornal, M. Honma, T. Mizusaki, A.C. Morton, W.F. Mueller, T. Ot-
suka, J. Pavan, A. Stolz, S.L.Tabor, B.E. Tomlin, and M. Wiedeking. Phys. Rev.
Lett., 92:072502, 2004.

[29] B. Fornal, S. Zhu, R.V.F. Janssnes, M. Honma, R. Broda, P.F. Mantica, B.A.
Brown, M.P. Carpenter, P.J. Daly, S.J. Freeman, Z.W. Grabowski, N. Hammond,
F.G. Kondev, W. Królas, T. Lauritsen, S.N. Liddick, C.J. Lister, E.F. Moore,
T. Otsuka, T.Pawlat, D. Seweryniak, B.E. Tomlin, and J. Wrzesiński. Phys. Rev.
C, (in press).

110
[30] P. F. Mantica, B.A. Brown, A.D. Davies, T. Glasmacher, D.E. Groh M. Horoi,
S.N. Liddick, D.J. Morrissey, A.C. Morton, W.F. Mueller, H. Schatz, A. Stolz,
and S.L.Tabor. Phys. Rev. C, 68:044311, 2003.

[31] S. N. Liddick, P.F. Mantica, R. Broda, B.A. Brown, M.P. Carpenter, A.D. Davies,
B. Fornal, T. Glasmacher, D.E. Groh M. Honma, M. Horoi, R.V.F. Janssens,
T. Mizusaki, D.J. Morrissey, A.C. Morton, W.F. Mueller, T. Otsuka, J. Pavan,
H. Schatz, A. Stolz, S.L.Tabor, B.E. Tomlin, and M. Wiedeking. Phys. Rev. C,
(in press).

[32] G. Audi and A. H. Wapstra. Nucl. Phys. A, 595:409, 1995.

[33] G. Audi, A. H. Wapstra, and C. Thibault. Nucl. Phys. A, 729:337, 2003.

[34] P. Möller, J.R. Mix, and K.-L. Kratz. At. Data and Nucl. Data Tables, 66:131,
1997.

[35] O. Sorlin, V. Borrel, S. Grévy, D. Guilemaud-Mueller, A.C. Mueller,

F. Pougheon, W. Böhmer, K.-L. Kratz, T. Mehren, P. Möller, B. Pfeiffer,
T. Rauscher, M.G. Saint-Laurent, R. Anne, M. Lewitowicz, A. Ostrowski,
T. Dörfler, and W.-D. Schmidt-Ott. Nucl. Phys. A, 632:205, 1998.

[36] F. Ameil, M. Bernas, P. Armbruster, S.Czajkowski, Ph. Dessangne, H. Geissel,

E. Hanelt, C. Kozhuharov, C. Mieha, C. Donzaud, A. Grewe, A. Heinz, Z. Janas,
M. de Jong, W. Schwab, and S. Steinhäuser. Eur. Phys. J. A, 1:275, 1998.

[37] O. Sorlin, C. Donzaud, L. Axelsson, M. Belleguic, R. Béraud, C. Borcea, G. Can-

chel, E. Chabanat, J.M. Daugas, A. Emsallem, M. Girod, D. Guilemaud-Mueller,
K.-L. Kratz, S. Leenhardt, M. Lewitowicz, C. Longour, M.J. Lopez, F. de Oliv-
eria Santos, L. Peitzon, B. Pfeiffer, F. Pougheon, M.G. Saint-Laurent, and J.E.
Sauvestre. Nucl. Phys. A, 669:351, 2000.

[38] T. Dörfler, W.-D. Schmidt-Ott, T. Hild, T. Mehren, W. Böhmer, P. Möller,

P. Pfeiffer, T. Rauscher, K.-L. Kratz, O. Sorlin, V. Borrel, S.Grévy,
D. Guillemand-Mueller, A.C. Mueller, F. Pougheon, R. Anne, M. Lewitowicz,
A. Ostrowsky, M. Robinson, and M.G. Saint-Laurent. Phys. Rev. C, 54:2894,
1996.

[39] R. Grzywacz, R. Beraud, C. Borcea, A.Emsallem, M. Glogowski, H. Grawe,

D. Guillemaud-Mueller, M. Hjorth-Jensen, M.Houry, M. Lewitowicz, A.C.
Mueller, A. Nowak, A. Plochocki, M. Pfu:tzner, K. Rykaczewski, M.G. Saint-
Laurent, J.E. Sauvestre, M. Schaefer, O. Sorlin, J. Szerypo, W. Trinder, S. Vi-
teritti, and J. Winfield. Phys. Rev. Lett., 81:766, 1998.

[40] U. Bosch, W.-D. Schmidt-Ott, P. Tidemand-Petersson, E. Runte, W. Hille-

brandt, M. Lechle, F.-K. Thielemann, R. Kirchner, O. Klepper, E. Roeckl,
K. Rykaczewski, D. Schardt, N. Kaffrell, M. Bernas, Ph. Dessagne, and
W.Kurcewicz. Nucl. Phys. A, 164:22, 1985.

111
[41] U. Bosch, W.-D. Schmidt-Ott, E. Runte, P. Tidemand-Petersson, P. Koschel,
F. Meissner, R. Kirchner, O. Klepper, E. Roeckl, K. Rykaczewski, and
D. Schardt. Nucl. Phys. A, 477:89, 1988.

[42] T. Dörfler. Phys. Rev., 120:927, 1996.

[43] S.J. Freeman, R.V.F. Janssens, B.A. Brown, M.P. Carpenter, S.M. Fisher, N.J.
Hammond, M. Honma, T. Lauritsen, C.J. Lister, T.L. Khoo, G. Mukherjee,
D. Seweryniak, J.F. Smith, B.J. Varley, M. Whitehead, and S. Zhu. Phys. Rev.
C, 69:064301, 2004.

[44] E.B. Norman, C.N. Davids, M.J. Murphy, and R.C. Pardo. Phys. Rev. C,
17:2176, 1978.

[45] E. Runte, K.-L. Gippert, W.D. Schmidt-Ott, P. TidemandPertersson, L. Ziegeler,

R. Kirchner, O. Klepper, P.O. Larsson, E. Roeckl, D. Schardt, N. Kaffrell,
P.Peuser, M. Bernas, P. Dessagne, M. Langevin, and K. Rykaczewski. Nucl.
Phys. A, 441:237, 1985.

[46] W.-D. Schmidt-Ott, K.Becker, U. Bosh-Wicke, T. Hild, F. Meissner, R.Kirchner,

E.Roeckl, and K. Rykaczewski. Inst. Phys. Conf. Ser., 132:627, 1992.

[47] H. Simon, D. Aleksandrov, T. Aumann, L. Axelsson, T. Baumann, M.J.G.

Borge, L.V. Chulkov, R. Collatz, J. Cub, W. Dostal, B. Eberlein, Th.W. Elze,
H. Emling, H. Geissel, A. Grunschloss, M. Hellström, J. Holeczek, R. Holzmann,
B. Jonson, J.V. Kratz, G. Kraus, R. Kulessa, Y. Leifels, A. Leistenchneider,
T. Leth, I. Mukha, G. Mu:nzenburg, F. Nickel, T. Nilsson, G. Nyman, B. Pe-
tersen, M. Pfützner, A. Richter, K. Riisager, C. Scheidenberger, G. Schrieder,
W. Schawb, M.H.Smedberg, J. Stroth, A. Surowiec, O. Tengblab, and M.V.
Zhukov. Phys. Rev. Lett., 83:496, 1999.

[48] T. Motobayashi, Y. Ikeda, Y. Ando, K. Ieki, M. Inoue, N. Iwasa, T. Kikuchi,

M. Kurokawa, S. Moriya, S. Ogawa, H. Murakami, S. Shimoura, Y. Yanagisawa,
T. Nakamura, Y. Watanabe, M. Ishihara, T. Teranishi, H. Okuno, and R.F.
Casten. Phys. Lett. B, 346:9, 1995.

[49] D. Guillemaud-Mueller, C. Detraz, M. Langevin, F. Naulin, M. De Saint-Simon,

C. Thibault, F.Touchard, and M. Epherre. Nucl. Phys. A, 426:37, 1984.

[50] A. Ozawa, T. Kobayashi, T. Suzuki, K. Yoshida, and I. Tanihata. Phys. Rev.

Lett., 84:5493, 2000.

[51] K. Heyde, J. Jolie, J. Moreau, J. Ryckebusch, and M. Waroquier. Nucl. Phys.

A, 466:189, 1987.

[52] Table of Isotopes. edited by R.B. Firestone. Wiley, New York, 1996.

[53] A. Poves, J. Sanchex-Solano, E. Caurier, and F. Nowacki. Nucl. Phys. A, 694:157,

2001.

112
[54] B.A. Brown, A. Etchegoyen, and W.D.M. Rae. The computer code oxbash.
MSU-NSCL Report 524, MSU-NSCL, 1998.

[55] M. Horoi, B.A. Brown, and V. Zelevinsky. Phys. Rev. C, 67:034303, 2003.

[56] L. Nordeim. Rev. Mod. Phys., 23:322, 1951.

[57] M.H. Brennan and A.M. Bernstein. Phys. Rev., 120:927, 1960.

[58] A.M. Oros-Peusquens and P.F. Mantica. Nucl. Phys. A, 669:81, 2000.

113