PH YSICAL REVI EW VOLUME O4, NUMBER APRIL 1, i954

The Ground States of Odd-Odd Nuclei

C. SCHWARTZ
Departraer&t of Physics awd Laboratory of ENctear Sctewce, Massachlsetts Inststate of Techmotogy, CarabrQge, trlassachtcsetts
(Received December 23, 1953)

Applications of techniques already developed for the study of atomic spectra to calculations on odd-odd
nuclei give some particularly simple results. The spins of the ground states of nearly all odd-odd nuclei
measured can be accounted for, and the validity of Nordheim s empirical rules is discussed.

INTRODUCTION Here we write the sequence j&"&J& as meaning e&

' 'T was found by Nordheim' that the total spin of J equivalent particles each of spin j& coupled to a re-
~- the ground state of odd-odd nuclei could be given sultant spin Ji, j2"'J2 is similarly read. This form has
by the rules: no special symmetries regarding interchange of protons

J = j „j„I, —
I
if 1 +j +J„+ j„ is even,
and neutrons, which are thus considered as two distinct
particles. Whatever the actual relation between proton
I
j„—
j„I (J&j „+j„, if 1„+jv+1„+j„ is odd, and neutron may be, our approximation is certainly
valid for heavier nuclei where the total isotopic spin is
where j~, j„are the individual spins of the odd proton
not a good quantum number.
and odd neutron involved, and l~, I„ their respective
At this point all the diGerent levels of the total spin
orbital quantum numbers as determined from the shell
model most successful in the study of odd-even nuclei.
J are degenerate. Now introducing an attractive short-
range interaction between particles of group one and
The calculation of de-Shalit' on nuclei with one odd
those of group two, we shall calculate by perturbation
proton and one odd neutron outside closed subshells
has given a theoretical basis for Nordheim's rule in theory the 6rst-order energy shifts and see which state
this restricted case. Other calculations in special cases of total J
is pushed deepest.
have been carried out by Kurath' and by Flowers. 4 Taking the interaction between protons and neutrons
To discuss the general odd-odd nucleus with ni as a sum of two-body static potentials, we have
equivalent particles of one kind and n2 equivalent
particles of the other kind all outside closed subshells,
"
we shall adopt the "odd group model. In this approach
it is assumed that the particles in the two groups, i.e.,
protons and neutrons, first interact among themselves
to give in their lowest states some well-defined resultant
where the indices i and j
count particles in the first
and second groups, respectively. Now, according to the
group angular momenta Ji
and J2. It is expected that
techniques developed by Racah, ' the two-particle inter-
this approach, neglecting all but the ground configur-
action may be decomposed as
ations of the separate groups, should be valid for the
determination of only the very lowest levels of the
V; = —Qsfs(rt r )t,
combined configuration. '
&'& t;&"&. (2)

THE INTERACTION Here t;&s& is a tensor operator with respect to jt (thus

MATRIX
also with respect to Jt and J'), of rank h, which operates
Ke shall start with the zeroth-order wave function of only on the angle and spin coordinates of particle i.
the Mayer-Jensen j
jco'upled inde-pendent particle
scheme with the two groups separately coupled to
t;&" is similarly dedned with respect to particle j, and
well-defined resultant angular momenta. The con6gur-
t;~~'t, &~& is the scalar product of the two tensors.
ation of the n~ particles of the erst group is character- Since both groups contain equivalent particles, the
ized by the quantum numbers ltd, j&, J&, that of the n2 radial part of the matrix element of every term in the
particles of the second group by /2, j2, J2. double sum (1) and (2) will have the same value for a
The wave function for the combined configuration given k, namely
with total spin J
and magnetic quantum number M is
pOO po0
then written:
~t I'I ~s s j's(r trs) rr'rs'«t«s (3)
4 (j,"Jr,js"Js,JM) = Z A(j, "J&Ma)A(j, "J~,)
I I

M].My
X (J M J M J JsJM).
I
where E» is the total radial part of the wave function
characterized by l& arid the principal quantum number
' L. W. Nordheim, Phys. Rev. 78, 294 (1950).
s A. de-Shaht, Phys. Rev. 91, 1479 (1953). of the state in question.
s D. Kurath, Phys. Rev. 87, 218 (1952) and 91, 1430 (1953).
' B. H. Flowers, Proc. Roy. Soc. {London) A212, 248 (1952). ' G. Racah, Phys. Rev. 62, 438 {1942).
 C. SCHWARTZ
If' we now set Eq. (8) reads

—Q, g. (k)
n2
—Q g. (k) (j "J=jllT"'llj"J= j)= (jllT'"'ll j) «»»dd; (9a)
T (k) and T (k) (4)
2j+1—2s
(j"J=jllT"'ll j"J=j) = (jllT'"'ll j).
Eqs. (1), (2), (3) combine to give 2j —1
V12 Zk fk(rr, rs) Tr'"' Ts " ~ for k even, (9b)
For the erst-order energy shifts we take the diagonal except for k = 0. With an interaction of the Wigner type,
matrix element of (5). Using (3) we obtain: i.e., V;, a function of r; — r; only, it is weIl known that
E(jr"'Jrj2 J2,J)= —Zk I'k(jr"'Jtj2 J2 J~ 1,.(k) —(4&) l P', (k) (10)
X Tr(k) Ts(" jr"'Jr, js"'J2,Jm). (6) If we
I I have a spin force of the form
Now since T&& ' operates only on particles of group one V, ;=(r;.(r;V(r; —r;),
and T2( ' only on particles of group two, we can use a then
well-known theorem of Racah' to decompose this 1 (k') = (4~)ly'. (k)&,.(r)
matrix element (6) of Tr(k) T2(k' into the reduced
matrix elements: ((r is a tensor of rank one) is not an irreducible tensor,
but it may be expressed as a sum of irreducible tensors
( —1)"+'~'(j "'Jrll Tr"'ll j "'Jr) (j2"2Jsll Ts"'ll of ranks:
k'=k —1, k, 4+1.
X j2 J2) IV(J)J2J1J2 J~) (7)
where W is a Racah coeQicient. ~ It can be shown" that only terms for which k and
Ef, for the j-j
coupled' configuration we dehne j"J, also k'+0+1 are even numbers, contribute in the
the Q operator and seniority number s in a manner evaluation of the necessary matrix elements.
completely analogous to that done for S coupling, ' I. Thus for a (r (r force the tensors Tr and T2 in (7) are
it is not dBBcult to derive the pertinent equivalent of of odd rank only. Equation (9a) then tells us that
Racah's Eq. (69)III:
E (jr"'Jr=jr, js"'Js=j2, J)=K(j &jsJ)
(j"»IIT'"'ll j"»') = U"»II T"'ll j'»') for o"o force. (12)

j+ 1— —s) —( —1) (e —o) But E.(j&j 2)Jwas just the term calculated by de-
Shalit. ' In the limit of zero range force,
k
(2 22
X (8)
2/+ 1 —22) &.(jr"'Jr=jr, j2"'Js=j2, J)=&.(j&j J)2
except for 0=0.
This equation gives for any irreducible tensor oper- = —-'2Ps(2 j,+1)(2 j2+1) (j;',j,——,' j,j,JO)' I

ator T(~' the matrix element in a con6guration of n 2J+1

particles. with seniority v in terms of the matrix element I (2gr+1)+ ( —1) "+"+s(2gs+1)$2
in the configuration of only v particles. For example, if
we take T&~& to be the quadrupole moment operator
4J(J+1)
Q= (16)r/5)&F22 (the normalized spherical harmonic is —[1+2(—1)"+'rt ~] . (13)
here written as F ') and s happens to be one (J= j),
Eq. (8) tells us that the quadrupole moment for the For Wigner force Tj and T2 are of even rank. Noting'
configuration j"(J=j) is (2j+1— 222)/(2j —1) times that T~' is the scalar e~ and
the quadrupole moment for a single particle in that
state with the same orientation. (jr"'J) = j)IIT)'ll jr"'Jr= jt) = ~r(2 jr+1)',
Now it has been showri by Edmonds and Flowers'
and similarly for T2', and also with
and also by Racah' that for a wide variety of attractive
interactions among like particles the ground state of ( —1)'+ WU, j,j,j„JO)= L(2j,+1)(2js+1)g-&,
the con6guration j"
for n odd has seniority v=1. This
we have Anally
gives rise to Mayer's rule that an odd number of
equivalent protons or neutrons couple to a resultant J &w(jr"'Jt= jr, js"'J2= js, J)
equal to the j of the individual particles. With v=1,
2 jr+1 —222r 2 js+1—2N2
' G. Racah L Talmi, Physics 18, 1097 (1952). &w(jrj2J)
' G. Racah, and
Phys. Rev. 63, 367 (1943). 2 jQ 2j2-1
' A. R. Edmonds and B. H. Flowers, Proc. Roy. Soc. (London)
A214, 515 (1952). 2 j&+1— 2N& 2 j2+1— 2rss
' G. Racah, I.. Forkos jllemorial Volume (Research Cou.ncil of (14)
Israel, Interscience Publishers, New York, Agents, 1952), p. 294. 2j& —1
 GROUN D STATES OF OD D —OD D NUCLE. I

TAnx, n I. -E(j &j sf)/Fo for zero range forces.

1/2 3/2 ~
5/2, 7/2 9/2
W r+
1 —3 1 1 1 —1.67 1 —1.40 1 1 1 —128 1 1022 1 -1/2
1 1 -0.33 1 —0.60 1 1 1 -071 1 -0.78 1 1 1 -0.82
2 -6 2 1.8 1.8 -3 1.72 -2.40 1.72 1.67 1.67
—0.
-2.14 —
-3/2
1.20 0.67
0.40 —1.20
1.20 0.40
0.89
0.3/
0.20
0.37 -10.89 0./6
0.38
0.76
-0.89
0
0.38
0.'/0
0.39
11 0.70,
039 —0.82 +1/2
1/2
1.20 1.20 —0.86 1.29 —1 1.29 1.33 1.33 -1.09 136 115 136 +3/2
3 9 3 2.57 2.57 -4.29 2.38 -3.33 2.38 —5/2
1.63 1.63 1.29 1.05 0.82 0.30 —
—1. 0.59
05 0.82 3/2
0.69 -2.06 0.69 0.62 1.67 0.62 0.62 —1.45 0.62 —1/2
0.91 0.91 0 0.90 -0.14 0.90 0.88 0.88 —0.24 +1/2
0.29 -0.86 0.29
—1.17 0.27 0.27 —0.77 0.27 -0.71 —0 27 . +3/2
1.43 1.43 1.52 —1.28 1.52 1.57 1.57 1.36 +5/2
4 -12 3.33 3.33 —5.56 —7/2
2.102.10 1.84 1.25 0.91 1.25 —5/2
-2.86
0.95 0.95 0.84 0.84 -2.27 -3/2
0.99 0.36 0.91 0.19 0.91 —1/2
. 0.47 —0. 99
1.40 0.4/ 0.43 0.43 -1.24
0.94 0.94
—0.70
-0.26 0.9S -0.35 —0.95 +3/2
0.23
1.63 1.63
0.23
-1.42
0.22
1.71 - 1.
0.22 0.65
51 1.71
+5/2
+7/2

Again, for zero range forces, ' former effect, due to the second term in Eq. (14), and,
in fact, any translation of the energy scale can be
&w(jtjsJ) ignored here since it does not aGect the level spacing
= —&~a(2 jr+1) (2js+ 1) (jtk js —k jt jsJO)' at any mixture. As nj and n2 increase through the 6rst
I

2J+1 halves of their respective shells, the levels are bilinearly

compressed. When either group is exactly half of a
[(2jr+1)+ ( —1)~'~+f~~(2gs+1)]s) filled shell, all the levels (on the Wigner side) are
X (15) degenerate. As one group, . only passes through the
4J(J+1) second half of its shell the levels are inverted and
dilated: n~ holes and n2 particles give exactly the
In Table I are given values of — E(j&jsJ)/Fs com- inverted level structure of n~ particles and n2 particles.
puted from Eqs. (13), (15) with columns headed W
n& holes and n2 holes are again equivalent to n& particles
for Wigner force and 0~ for spin force in states of
and n2 particles.
positive and negative parity, respectively.
Surveying the data on odd-odd nuclei of Nordheim"
DISCUSSION
"
and of Goldhaber and Hill, and the recent note of

Wigner forces alone will not give the desired dis-

tinction between condgurations of various l's, so we
shall investigate the usefulness of a mixture of Wigner i-) i-) (~) (-) (-) i-) i-) (2)
and Bartlett forces. If we take the interaction as

it is interesting to plot against o, the energy levels

characterized by the diferent ways in which J& and J&
can combine to the resultant J.
In Fig. 1 are a typical
set of such graphs for various values of n~ and n~ when
we have chosen j& — 5/2, j&=7/2 and f1+fr is even.
de-Shalit gives a number of such graphs for several
values of j& and j2 when n&= n2= 1. a*a ant asO anE auO a=1 ao as/ a no a*i
From Eqs. (12), (14) we can easily see how these
graphs change as we put more particles in each shell: Fro. 1. Energy levels of con6gurations (5/2)"&(7/2)"~ with posi-
—
tive parity under interaction Vis= t (1-a)+oe| e~)s(rz-rm).
For a pure e~. e2 force, 0, =1, the level structure is Both subgroups have seniority one.
independent of n~ and n2.
for a pure Wigner force, 0, =0, increasing e~ and n~
'o L. W. Nordheim, Revs. Modern Phys. 2$, 322 (1951).
"M. Goldhaber and R. D. Hill, Revs. Modern Phys. 24, 179
both lowers and compresses the level structure. The (1952).
 C. SCHWARTZ
+ m~=02= I. Such a relation is
(y)) (y) (y)) (y)
(~)
Z(ji"'J, js"IJI,J)
z)
(~) (~)
s (~) sl s
I Q (2J1+1)(2J'+1)li/" (Ji'jiJJI, J,J')

FIG. 2. Energy levels of con-

X (ji"'-'(Jt'), jrJ1 jj ji" Ji)'Z(j„jI"J„J'),
6gnrations (7/2) "75/2(7/2}"'7/I
with positive parity
interaction
under
—L(1-88}
j
F.(j i, I"'JI,J)
V82
+/3878 '5783& (r1 —rI). =752 Q (2JI+1) (2J"+1)~'(JI'jIJ'j 1,'JIJ")
J5sl Jr I

X (js" '(JI'), jIJIjIjs"IJ2)'&(jijIJ"),

where J,
(ji"1 '(Ji'), ji, )ji"'Ji) is the fractional par-
entage coefBcient for the con6guration j~"1J~ as com-
puted by Edmonds and Flowers. Of use in checking'
0.*0 . a*1 a o a*i numerical work is the sum rule

"
King and Peaslee, we 6nd that the above considera-
Qz(2J+1)E(ji"'Ji,js"'JI,J)
tions furnish an excellent accounting of the measured = —rsins(2Ji+1) (2JI+1)Fp. (18)
and indicated ground state spins. Except for a few
nuclei with the configurations (3ds/2)2 and (4fr/I),
there were no cases which could not be 6tted to our
" Using Eq. (17) and Table I, we have computed the
energy levels for the configuration (7/2)'5/I(7/2) with
positive parity under the perturbing potential (16).
graphs. However, it is known from the spins and Since (8) still gives the correct relation between particles
magnetic moments of the odd-even nuclei Na23 and and holes for any seniority, we have the several cases
Mn" that these exceptional configurations are probably shown in Fig. 2.
not of seniority one and must be treated separately The spin 4 for Sc4' is well accounted for, and we
(see next section). would predict spin 6 for Mn" and 4 or 5 for V". Recent
In many cases where the spin was known the con- work" gives spin 6 for V", which is best accounted for
figurations could be unambiguously assigned. In cases
by assigning seniority one to the (7/2)' proton con-
where measurements were incomplete, some certainty figuration. "
For Sc~ the indicated spin 2 can be fitted
about the configurations involved allowed prediction of only with an extremely small value of n. The correct
the most probable spin. configuration here is most likely a mixture of (7/2)'5/I
and (7/2)'7/I, this may also be the case in Mn".
EXCEPTIONAL CONFIGURATIONS
For F'p the configuration (si/2)L(dp/2)25/I] would have
In some cases it is desirable to calculate the level J=3 lowest whereas (si/2)L(dp/I)'I/2$ wouldhave J=2
schemes involving configurations of seniority other than lowest. The measurements do not yet indicate which
one. For such work we need a general relation between of these two is correct.
the perturbation energy in the state (ji"'Ji,jssIJI, J) Calculations for the nuclei Na" and V" with both
and the terms already calculated for the simple case neutron and proton configurations having seniority
diGerent from one require performing both sums of
TABLE II. EGective e-p force mixtures indicated by observed spins (17) and have not been carried out. However, for the
latter nucleus, letting only one group have seniority
Nucleus Configurations Spin one would lead to spin 4 or 5, which are the most
Li0 (PI/2) I/O (PI/2) I/I &~ 0.10 probable indicated values.
LP (P3/2) 3/I (PI/2) I/I &~ 0.25
QS (PIlI) 8/2 (PI/I) 3/I ~&0.25
@10 0.10 NCLUSIONS
Na'4
CP'
(Ps+)38/2
(d")
(d8/I)
.. (PI/I) 3/2
(~5/2)'512
(d8/2)33/I
&~

~& 0.42
&~ 0.25
In Table II
are listed a number of nuclei from which
(~")".
3/2
K40 (f7/I}'7/I ~& 0.50 we may infer something about the necessary amount of
K~ (d3/I) 3/2 (f7/I }'7/I ~& 0.17
mixture of the two forces considered. It should be
Sc4e (f7/I) 7/2 (f7/2)55/I ~& 0.30
Co" (fv/2)vv(~ (PI/I) I/I ~& 0.10
remarked that the two-body interaction taken as a
Co60 (f7/2}77/I (P8/2) 3/2 ~& 0.10 perturbation here is not the total neutron-proton force
or
(f5/2) 5/2 ~& 0.50 but is supposed to be the residual of an averaging
Rbs' (PI/O) 3/I (d5/I) 5/I ~& 0.17; ~& 0.50 over all interparticle forces which gives the central
S$122 &~ 0.10
(g7/2) 7/I (1381/I) 11/2 potential well of the shell model. Especially since none
83
Kiknchi, Sirvetzz, and Cohen, Phys. Rev. 88, 142 (1952).
@ Q. . Vf. King and D. C. Pellet!, Phys. Rev. 90, 1001 {1953). '4 A. Hitchcock, Phys. Rev. 87, 664 (1952).
 GROUN D STATES OF OD D —OD D NUCLEI
but a scalar potential survives between outer particles and protons they are equally valid; and the "strong"
and any subgroup coupled to zero total spin, there is rule,
no simple reason to expect the mixture of forces in the
perturbation to be independent of 3, . However, with
J= j
~ I j j
— „I, for n+j „+1,+l„even,
is still good when ei and e2 are both more or both less
the exception of Co" all of the nuclei can be 6tted than half of a filled shell. However, in general, the
with a value of 0. between 6 and 4. situation is more complicated, especially as it is unsafe
Kith regard to Nordheim's rules, we now see that to rely on the constancy of the mixture of forces.
their validity is not as gerieral as was 6rst indicated in The writer wishes to acknowledge the assistance of
the case of one odd proton and one odd neutron. Dr. A. de-Shalit, who instigated this work and gave
Certainly for one less than a filled shell of both neutrons extensive advice throughout its development.

PH YSICAL REVIEW VOLUME 94, NUMBER APRIL 1, j9$4

The Decay Scheme of Zr"t'*

PHILLIP S. MITTELMANt
Renesselaer Polytechnic Institgte, Troy, See Fork
(Received September 8, 1953)

The decay scheme of Zr9' has been studied using the techniques of beta-ray spectroscopy and beta-gamma
angular correlation. The decay of Zr' is found to proceed by three beta gamma cascades. Two of the beta
transitions are allowed and proceed to Nb'5 levels at 722.0 kev and 754.4-kev. The third beta transition
is to a 235-kev level in Nb". From E-conversion coefficient determinations and shell theory the 722-kev
and 754.4-kev levels are both assigned even parity and a spin of 5/2 or 7/2.

'HE radiations originating from the decay of Zr'5 The magnetic 6eld in the spectrometer was measured
have been studied by a number of investigators using a double-coil monitor with a temperature-
with somewhat divergent results. A recent paper by stabilized permanent magnet supplying the reference
Cork et a/. ,' published after the completion of this study, field. The magnetic field could be kept constant to
contains an excellent survey of the previous work. within 1 part in 20, 000 for extended periods of time.
In the present investigation, beta-ray spectrometer The source material used in all of the experiments
studies were made of the Zr" spectra and gamma ray —
consisted of a Zr" Nb" oxalate mixture obtained from
conversion lines. In addition, the angular correlation Oak Ridge National Laboratory. The Oak Ridge
between the Zr' beta and gamma rays in cascade to analysis indicated that the radiochemical purity of the
the ground state of Nb" was measured. source material was better than 99 percent and that
the source initially consisted of 40 percent Zr" and
BETA-RAY SPECTROMETER STUDIES 60 percent Nb".
The beta-ray spectrometer used in this investigation The source for the spectrometer study of the spectrum
was a double-focusing, high-resolution spectrometer shape was prepared by laying down the source material
patterned after that of Kurie, Slack, and Osaba. ' In
in the form of a line 3 mm wide by 1 inch high on a thin
most of the runs reported here it was operated at a formvar 61m on which a 61m of gold had been vacuum
resolution of 0.2 percent. The electron detector used for evaporated. The areal density of the source backing
was less than 125 pg/cm'. The areal density of the
studies of the shape of the beta spectrum was an end-
source material was approximately 50 pg/cms.
window Geiger counter with a 0.6-mg/cm' rubber
hydrochloride window. Later studies of conversion Figure 1 shows the Kurie plot obtained with this
source. It will be noted that the plot has an "allowed"
lines and of the high-energy end of the beta spectrum
shape from the end point of the Nb" spectrum at 160
employed an end-window counter with a 3-mg/cm'
kev out to approximately 360 kev. At the high-energy
mica window.
end of the spectrum the 900-kev branch of the Zr"
t This work was carried out in partial fulfillment of the require- decay is in evidence but the source was too weak to
ments for the Ph. D. degree at Rensselaer Polytechnic Institute.
It *was supported by the U. S. Atomic Energy Commission. make signi6cant measurements on it. The E conversion
This work was reported at the 1953 Washington meeting of line resulting from the 23S-kev isomeric transition in
the American Physical Society. Nb" is also in evidence in this 6gure.
f. Now at Nuclear Development Associates, Inc. , White Plains,
New York. The conversion lines from the gamma ray transitions
'Cork, LeBlanc, Martin, Nester, and Brice, Phys. Rev. 90, in Nbo and Mo ~ were examined with a stronger source.
5'79 (1953).
' Kurie, Osaba, and Slack, Rev. Sci. Instr. 19, 771-6 (1948}. This source was prepared by ruling a line j. mm wide and