Decay of 90Mo

I.E.1

] t

Nuclear Physics

72 (1965) 113--131; (~)

North-Holland Publishing Co., Amsterdam

Not to be reproduced by photoprint or microfilm without written permission from the publisher

DECAY OF 9 ° M o J. A. COOPER, J. M. HOLLANDER, M. I. KALKSTEIN t and J. O. RASMUSSEN Lawrence Radiation Laboratory, University of California, Berkeley, California tt Received 13 April 1965 Abstract: High resolution beta- and gamma-ray spectroscopic studies of the decay of 5.7 h 9°Mo have been carried out, with use of the Berkeley 50 cm ~r~/2 iron-free spectrometer and lithiumdrifted germanium gamma-ray detectors. The gamma-ray spectrum revealed transitions of

42.8, 122.5, 163.0, 202.9, 257.5, 323.0, ~ 420, 445.2, 472.5, ~ 490, 945, 993, 1272, 1389, 1455 and 1463 keV. The internal-conversion-electron spectrum was studied in the energy range 0 to 400 keV. From measurements of L subshell internal-conversion ratios, the multipolarities of the 122 and 257 keV transitions were determined to be E2 and E3, respectively. A partial decay scheme for 9°Mo is proposed that utilizes the data on the 122 and 257 keV transitions together with considerations of the available shell-model states for 9°Nb. This scheme requires the presence of an undetected low-energy ( < 3 keV) transition in " N b . E

RADIOACTIVITY 9°Mo [from Zr(~, xn), Br(Z4N, xn), 88Nb(p, 4n)]; measured Er, It, cc, e/fl+ ratio. 9°Nb deduced levels, Y, zr. Natural targets.

1. Introduction T h e 9 ° M o isotope was first identified by Diamond 2). This activity, which was produced by the p, 4n reaction on 93Nb, was shown to decay with a half-life of 5.7 h. Subsequently, in a study of 9°Mo decay, Mathur and Hyde 2) observed gamma rays of 120 and 250 keV and a positon end point of 1.2 MeV. Mathur and Hyde also noted that the levels in 9°Nb populated by the decay of 9°Mo are isomeric, and they reported values of 24 see and 10 msec, respectively, for the half-lives of the 120 and 250 keV transitions. They reported no other radiations, and spin assignments to the 9°Nb levels were not made. The daughter activity, 14.6 h 9°Nb, the decay of which populates levels in the closed-subshell nucleus 4oZr, 90 has been the subject of many investigations. When the 9°Nb decay data are examined together with the known information on 9°Mo, some interesting inconsistencies appear, as follows: From its decay properties, 9°Nb has been given a ground-state assignment of 8 + or 9 + by Bjornholm, Nielsen and Sheline 3), and this can be considered rather definite. The even nucleus 9°Mo, with spin and parity 0 +, decays with a small logft value (5.3, calculated from the positon end point and half-life with use of the theoretical EC/fl + ratio from Feenberg and * Summer visitor 1963. Present address: Air Force Cambridge Research Laboratories, Bedford, Mass. tt This work was supported by the U.S. Atomic Energy Commission. 113

114

J.A. COOPERet

al.

Trigg 4); this low l o g f t indicates a spin change in the beta transition of 0 or 1, and implies that 9°Mo decays to a state in 9°Nb with spin 0 or 1. Only two transitions have been reported in the decay of 9°Mo, and the above considerations require that the s u m of their multipolarities be at least 7 units. On the basis of its half-life, the 250 keV transition was assigned by Mathur and Hyde as an M3 or E3. The data on the 120 keV transition were ambiguous, with an E3 assignment indicated by the half-life and K / L conversion ratio but an E2 assignment indicated by the total conversion coefficient. Even if these transitions were both octupoles, as suggested by their half-lives, angular momentum could not be conserved if the suggested spin difference (7 to 9 units) between the initially-populated state and the ground state is correct. Either a higher multipolarity is required for one or both transitions, or additonal transitions are required. This work was undertaken in the hope that an investigation of the internal-conversion-electron spectrum with the high-resolution nx/2 iron-free spectrometer s) and of the gamma-ray spectrum with lithium-drifted germanium detectors might resolve the above inconsistencies and provide a clearer picture of the levels in the odd nucleus 9°Nb.

2. Experimental Procedure The internal-conversion-electron spectrum was examined with the Berkeley 50 cm radius ~ / 2 iron-free spectrometer s) and a 180 ° permanent-magnet photographicrecording spectrograph 6). The counter used in the iron-free spectrometer was a Geiger counter with a window aperture, 1 mm by 20 mm, covered with a Formvar film of surface density ~ 50 #g/cm 2. An electrostatic preaccelerator system was used for the study of the very lowenergy region in the iron-flee spectrometer. An exploded view of the elements of the preaccelerator system is shown in fig. 1. This assembly was designed to replace the standard spectrometer source holder which fits into the source chamber from the top. All source elements were electrically insulated and connected to "feed through" electrodes in the top plate of the machined Lucite mounting block. The accelerating electrodes were made from 30/cm mesh nickel grid t which has ~ 90 ~o open area. The 9°Mo activity used for electron spectroscopy was produced by bombarding natural zirconium foils (99.9+ ~ Zr) with 65 MeV alpha particles in the Berkeley 224 cm cyclotron. The Zr foils were dissolved in concentrated hydrofluoric acid, the resulting solution evaporated to dryness, and the activity taken up in 5N H C 1 0.06N HF. This solution was placed on a Dowex-1 anion exchange column of dimensions 7-8 cm long by 0.4 cm in diam. Molybdenum forms anionic complexes that stick tightly to the column while niobium and zirconium pass on through. The Nb and Zr are quantitatively separated by eluting with several column volumes of 5N H C I - O . O 6 N HF. The Mo acitivity was stripped from the column with 1N HC1 and * The grid was made by Buckbee Mears Corp., St. Paul, Minnesota.

DECAY OF '°Mo

115

further purified and concentrated into 2 drops by repeating the above column procedure with a column 8 mm long by 1 mm in diam. T h e 9°Mo source for the iron-free spectrometer was prepared by vacuum sublimation of the dried chloride solution from a tungsten boat at > 2000 ° C through a collimator 1 mm by 10 mm onto an aluminium foil backing of surface density 7 mg/cm 2. The source for the permanent-magnet spectrograph was prepared in a similar manner, except that the activity was sublimed onto a 1 cm length of 0.25 mm platinum wire.

Source deposited on,~ 7 mg/cm2 AI foil on support ring

",," .1

Teflon insulalor Fine mesh metallic grid soldered \ between0.7- mmthick brassring s :-

.." i"

'

L

~

" ~Grid

Brass plate

Teflon insulator Fig. 1. Exploded view of the components of the preaccelerator system.

The 9°Mo source for the gamma-ray studies was also produced and separated by the procedure described above. In addition, 90Mo gamma sources were produced by bombarding natural NaBr with 80 MeV 1aN ions in the Berkeley Heavy-Ion Linear Accelerator (Hilac) and by bombarding Nb metal foils with 47 MeV protons in the Berkeley 224 cm cyclotron. When NaBr was used as the target, the separation of 9°Mo from the NaBr and other activities was made by dissolving the NaBr in 5N H C 1 - 0 . 0 6 N HF. This solution was passed through a Dowex-1 column and eluted with 5N H C 1 - 0 . 0 6 N HF. All impurities pass through the column. The purified Mo was stripped from the column with 1N HC1. When Nb metal foils were used as the target, they were dissolved in a mixture of concentrated H F and HNO3, and the solution evaporated to dryness. The Mo activity was then taken up in 5N HC1-0.06N H F and purified further with a Dowex-1 anion column as described above. The gamma-ray spectrum was studied with a lithium-drifted germanium detector with an active volume of 6 cm z by 9 mm deep. It was maintained at "liquid-nitrogen

116

COOPER e t al.

J.A.

temperature" (-196°C) with use of a 10 1 gravity-feed liquid-nitrogen reservoir of commercial manufacture t. The associated electronics consisted of a low-noise, low-capacity pre-amplifier and biased-amplifier system designed by Goulding

((~)

MO90

MO 90

122.5 keV

gamma

spectrum

Br79'St(NI4, xn) Ge ( L i )

l0 S

(6cm

% c c

M o 90

delector

2 x 9ram

deep) MO 9o 257.5 keV

2

~D ¢z M o 9o

co

Nb90 MO90 ~42 keV [63.0 keV

g

~D

10 4

MO 90

i0 ~

I

t

[

I

I

I

I

I

I

I

~

I

~

k

I

I

IO0

I

1

I

I

t

r

J

i

r

200 Channel

I0 6

i

i ...........

I

I

number

i

:

i

I

,

•

(b),

MO 99

I

500

I

MO

I

90

I

i

i

,70 mr' a

i

~

spectruM,

~r79,8} ,,!,j14 , ~r~'} MO90

257.5 ke¢

Ge :' L i ) d e t e c t o , "

(6 cm 2 x 9 m m deep) 10 5

L1o 90

511

ke/

o_ ~n c O 0

i04

--

'

M 090 445.2 keV

4

Mo

9o

Mo9°

k

4725keV

M°e° 490keV

103 I00

20S' C n , ] r nel

500 number

t T h e liquid nitrogen reservoirs were o b t a i n e d f r o m Linde Oxygen Co., Inc.

400

117

DECAY OF 9°Mo

iO 5

keY

]

I

I

i

I

]

I

I

I

I

I

I

I

i

I

L

I

[

I

I

l

1

I

L

I

I

I

(c) Mo 90

gamma

Br79,

% c_

Ge(LI)

deteclor

(6cm ~10 c,.

spectrum

81 ( NI4, xn) Mo 90

2 xgmm

deep)

4

Ln

Me 9o

g O t.)

A

MogO 992.7 keY i

10 3

[

I

I

I

I

I

I

l

I

I00

I

I

I

I

I

I

J

I

~

~

~

~

I

I

[

~

]

~

I

I

I

I

I

I

I

I

I

I

I

1

I~

I

I

I

I

I

I

I

400

number

I

I

I

I

I

I

i

I

(d)

i

}

I

L

I

I

Mo 90 gamr'/q Br79,8

Uo 9° o

I

300 Char/ne~

~0 5

I

200

.

Ge

.45.2keY

f

(N~4xfl)

(Li)

',6crn

I

I

I

I

I

I

spectrum MO 90

detector

2 x CJmrn

deep)

Nb 90 i0 4

A

,,,o..,o,,

Mo.o

.b.o

g

/\A

10 5

I

]

l

J

I

I

I

I

L

I

I

L

I

I

I

i

[

I00

]~

L

~

I

I

I

I

1

L

I

I

I

I

I

I

200 Channel

IkJ 4~

I

I

Mo,o

I

t

[

I

I

I

I

I

I

i

I

I

L

I

300 number

I

I

I

I

I

I

I

I

I

I

I

{

L

I

I

I

I

t

(el M O 90 g a m m a -MOgO !589 ~,eV MOg° ~ 14~5 Mo,au

Br 79'81

Ge

1463k~V ~'

~

(Li)

( 6 cm 2

,4 z 7 kek,' [ Mo93'~)

.,o'

spectrum

( N 14 , ×n ) Mo 90

detector x 9ram

deep)

Nb 90

.

,.,o ,eV (Z 320 keV single escape)

I

10 2

•

I

L

I

I

I I I00

l

L

I

L

J. I

I

t

• ~.

I 200

I

- -

I

i

ChanneL

L

I

o 4 r

I

I

:

I

I I 300

2182keY

I

i 400

number

Fig, 2. Gamma-ray spectrum of ~°Mo, observed with 6 c m ' x 9 mm deep Ge(Li) detector system.

118

s.A. COOPERe t

aL

and Landis 7,s) and constructed at this Laboratory. Pulse-height analysis of the spectrum was made with a 400-channel analyser *.

3. Experimental Results 3.1. GAMMA-RAY SPECTRUM The gamma-ray spectrum observed with the Ge(Li) detector clearly shows the two prominent gamma rays previously reported by Mathur and Hyde 2), but in addition a number o f low-intensity gamma rays are found. (The gamma rays were shown to have originated from the decay of 9°Mo by their half-lives and their relative intensities, which were independent of the method used to produce the activity.) Examples of our recorded spectra are shown in figs. 2a-e, and the energy and intensity information on the 9°Mo photons is summarized in table 1. (The photon relative TABLE 1 Photons of 9°Mo decay Energy (keV) 42.76 a) +0.08 122.50 a)4-0.1 162.99 a)4-0.1 202.9 +0.5 257.52 a)4-0.15 323.0 4-0.5 ~,420 445.2 4-1 472.5 4-1 ~490 4-2 945.2 4-1.5 992.7 4-2 1272.5 4-2 1389.0 4-2 1455 4-3 1463 :/:3

Relative intensity b) 0.4 84 5.8 6.6 100 8.9 ~ 1 (complex?) 11 2.4 ~ 2 (complex ?) 12 2.4 9.4 5.2 3.6 0.6

a) Energy values determined from conversion-electron spectrum. Calibration of the iron-free spectrometer was made with the K line of the 208.364-0.2 keV transition from X~Ludecay ofref. 2x). b) The relative accuracy of the quoted intensity figures is thought to vary from 10 ~ for the strongest lines to as much as a factor of two for the weakest lines. intensities were determined by making measurements of the areas under the peaks and correcting for the variation of the Ge(Li) photopeak efficiency with energy. The Ge(Li) photopeak efficiency function was experimentally determined with use of a number of isotopes with well-known photon intensities, and this efficiency curve is reproduced in fig. 3.) t The analyser was manufactured by Radiation Instrument Development Laboratories, Inc., Melrose Park, Illinois.

DECAY OF

9°Mo

119

It is interesting to note that although many other gamma rays were observed in addition to the prominent 122.5 and 257.5 keV radiations, none has an intensity greater than 12 ~o of the 257 keY gamma intensity, and therefore unless one or more of the transitions is very highly converted, none can be in 100 ~o cascade with the 122 and 257 keV transitions. q

I

i

i I I I ~

103~

0E] 10 2

\

LU

i01

I

-0

i31

A

LU 1 7 7 m

D

TO ~8Z

%7

Hg 2O3

I> Bi 2o7

i0 °

i11

IOO

I

200 Photon

,

I , ,,,I

500 energy

C>7

I000 (keY)

Fig. 3. Gamma-ray photopeak efficiency curve for the 6 cm=x 9 m m deep Ge(Li) detector system used for the gamma-ray measurements.

3.2. INTERNAL-CONVERSION-ELECTRON

SPECTRUM

The internal-conversion spectrum of 9°Mo (including 93mMo and 9°Nb), taken with the 180° permanent-magnet spectrograph, is partially reproduced in fig. 4. The energy range of this spectrum is from 10 to 400 keV, and it shows dearly the K and L lines from the 122 and 257 keV transitions from 9°Mo decay, the 132 and 142 keV transitions from 9°Nb decay, and the 262 keV transition from 93mMo. In addition, M lines from the 132, 142, 257 and 262 keV transitions and the very weak K lines from the 163 and 203 keV transitions of 9°Mo are also visible. From this spectrum it is obvious that there are no strong, highly converted transitions within the energy range 10-400 keV that have been previously unreported. A scan with the iron-free spectrometer of the 10-100 keV energy range revealed only

120

J.A. COOPER et al. Energy(keV)

Energy (keY) Nb 90

Mo 90 122

K

132

K

12_2

L

142 132

K L

132

M

163

K

142 142

L M

203

K

Mo 93m 257 257 257

K L M

~

262 262

K L

262

M

Fig, 4. Internal-conversion-electron spectrum of 9°Mo (including ~SMo and 9°Nb), recorded with 180 ° permanent-magnet spectrograph.

I

i

8000

MI90 o

'~

I

, I .-,r Volts on

4 2 . 8 keV K line source: J x l O m m 2

I

I| /I /-

[

counter slil: 2x2Omm 2. Volts )off baffle

aperture'.O.l%

6000

E

-.."

/

4000

Q.

15%

/

/

o 2000

i"

!

,! d? 6.68

6,74 Spectrometer

I

,

7.33

I

I

7.38

current (A)

Fig. 5. K line of the 42.76 keV transition from 9°Mo decay measured with 50 cm iron-free JrV'2 spectrometer, with and without 4.69 kV preaccelerator voltage.

D E C A Y OF 9°Mo

121

the low intensity K and L lines from the 42.76 keV transition (figs. 5 and 6) and a large number of lines from 13 to 18 keV of about the same low intensity as the K line from the 42.76 keV transition. This region was not scanned in detail but it is noted that the energies of these lines corespond to the energies expected for the KXY Auger electrons from Zr, Nb and Mo due to the decay of 9°Nb, 9°Mo and 93raMo. 2000

l

IL [

~

l

I

'

Mo 90

• 1 4 2 , 8 key L 9roup I /source: IxlO mm z , / | counter slit:2x2Omm'

1500

.c E 04 I000

500

0

= 8.74

I

J 8.78

Spectrometer

I

i 8.82

current (A)

Fig. 6. L shell conversion line group of the 42.76 keV transition from 9°Mo decay.

3.3. SEARCH FOR LOW-ENERGY TRANSITION The region from about 0 to 10 keV was studied with the preaccelerator, using an accelerating potential of -4.69 kV. The spectrum observed in the initial scan of this region is shown in fig. 7. The region below 0.235 keV was not studied until several half-lives later because of its high intensity. The spectrum observed at that time is shown in fig. 8. These lines could not be identified unambiguously, but from their energies and the fact that they decayed with a complex half-life, it is thought that they are associated with the LXY Auger and LiL~X Coster-Kr6nig transitions. We cannot, however, rule out the presence of a very low-energy conversion line, mixed in with the Auger lines. From this experiment, we can say that there is no intense transition in 9°Mo decay with conversion lines in the region 2.5-10 keV; below 2.5 keV the results are not definitive. 3.4, MULTIPOLARITIESOF THE 122 AND 257 keV TRANSITIONS As pointed out in sect. 1, the multipolarities of the 122 and 257 keV transitions are crucial to the interpretation of the 90Nb level scheme. In an effort to determine these multipolarities unambiguously we have made careful measurements of the L

122

J.A.

COOPER et al.

subshell conversion intensity ratios with the iron-free spectrometer. The L subshell conversion intensity ratios are a very sensitive indicator of the multipolarity of a transition and have often been definitive whenever the L lines could be resolved. 10 4

0.5 [

~

~

I

I

I

I

i

1.0 I

I

i

1.5', I

i

J

I

Electron

energy

szale

..

°\° E

[

i

i

t"'* ~. ~"

•

°. °

2.5keV i

" °'.~

.~ o"~.*.'"'°~'°"

\

c%1 O

\

M090, k,lo 93, N b 90 t© 5

\

L X Y Auger grio~p

o.

o ~"

,

°

..°..'°'°" ~°

b

~

....°'°'°'°'.'°'•.°\

.

"~

i

i 2.0

]

sOurce: Ix 1.9 r r r " 2 COunter s l i t : 2 x 2 0 rrm 2 co,Jnter ~,lndo~', t h i c k r e s s

/

[

= 50~:g

opert ire. 0.05% preaceelerotor ,voltage or

baffle

i02

5.0

3. I

3,2

3

3.4

Spectrometer

3

current

3.6

3 5

3.7

(A)

Fig. 7. Internal-conversion-electron spectrum of ~°Mo measured with 4.69 kV preaccelerator voltage, showing the LXY Auger group. 10 5

0

r . . . .

~

Electron

Li i l X

c E

d.5 i

energy

I r

l

r

1,0 ,

~

r T

1.51 ,

scale

J

P

I

'2.0keV . . . .

I

i~090 ' Mo93 , NbgO

Coster-Kr6nig

source: I~10 mm 2 c o u r * e r slit : l x 2 O m m 2 ' counter w i n d o w thickness 5 0 p-g baffle aperture: 0.05% preaccelerator voltage on

group

\ IO 4¸

L XY Auger

c o

group

\ %

I0 3

2.9

3'.0

i

3 I

i

I

32

Spectrometer

l

3.3

current

3.4

,

3.5

(A)

Fig. 8. Internal-conversion-electron spectrum of 8°Mo measured with 4.69 kV preaccelerator voltage, showing the LILjX Coster-KrOnig group and the LXY Auger group.

However, the L subshell ratios method has not in the past been used for low-Z nuclei because of the difficulty in resolving the lines• In this particular case the L lines were •not completely resolved, but with the accumulation of good counting statistics the

123

DECAY OF °°Mo

complex structure of the line could be analysed so as to characterize the multipolarity of the corresponding transitions. The following method was used to make comparisons between the experimental 104

i

- ~ =0.13%

to d 103

== ,8 IOZ

Nb /

s o u r c e ~ x I0 mm z counter slit:Zx20 m m 2

i

15.92

15.94

15.96

Spectrometer current

11

15.98 (A)

16.000

Fig. 9. K line of the 142 keV transition from 9°Nb decay used as the standard line shape for the °°Mo L subsbell ratios analysis. 10 4

=

i I

/ ///~'/

i i

i

~

if

~

~

1

./." Nb e° 1 132-keY E3 transition \\ L group \ source: 2xlO mm 2 counter slit : 2x2Omm 2 baffle aperture: 0.1%

IC 16.42

~

d 16.44

i

i 16.46

Spectrometer

i

current

i 16.48

i

i 16.50

(A)

Fig. 10. Comparison o f experimental and theoretical L group line shapes for 132 kcV E3 transition from °°Nb decay. Normalization procedure is described in the text.

124

J.A. COOrER et aL 10 4

_

I

i

i

i

i

i

i

-;;;.--..

e~

10 3 .=_ E

u~ source: 2 x l O m m z counter slit: 2 x 2 0 ram2 b a f f l e aperture: 0.1%

o

\\

\\

10 2 ----M2

-M3 ..... M 4

I0

i

15.68

i 15.70

r

15.r72

I

J 15.74

Spectrometer

i

current

] 15.76

15.78

(A )

Fig. l l. Comparison of experimental L group line shape of 122 keV transition of 8°Mo decay with theoretical composite L group lines for M2, M3 and M4 multipolarities.

E

i

L)

2

. . . . // /

!

I Z Z - keV L. g r o u p sourc~ 2xlOmm 2 ,

//

\ \

2

------El

I0

15.68

I

151.70

i

J

15.72

Spectrometer

~-

\\~

I

5E

I .'?4 current

J

I

15.76

15.78

( .~, )

Fig. 12. Comparison of experimental L group line shape of 122 keV transition of S°Mo decay with theoretical composite L group lines for El, E2 and E3 multipolarities.

125

DECAY OF 9°Mo

I0

4

,

,

,

,

~

,

,

•,,, • \\\e

.~

•

IO 3

E

\•

1o 2

- - - - - - M 2

24.19 24,20

24.22

x

24.24

24.26

Spectrometer

24,28

current

\

~

24.30

(A)

Fig. 13. Comparison of experimental L group line shape of 257 keV transition of 8°Mo decay with theoretical composite L group lines for M2, M3 and M4 multipolarities.

JO 4

I

h

~

t

i

l

]

l

I

L

10 ~

c

E

c Io 2

o

baffle aperture:O.l%

.,,'"

\

- - E 3 . . . . . . E4 24.19 24.20

2422,

\

24.24

2426.

Spectrometer

current

24.28

24.30

(A)

Fig. 14. Comparison of experimental L group line shape of 257 k e y transition of °°Mo decay with theoretical composite L group lines for E2, E3 and FA multipolarities.

126

J . A . COOPER et al.

L subshell ratios and the theoretical ratios for various multipolarities: The experimental composite line was plotted on semi-log paper after background subtraction, and compared with a "theoretical line" constructed by using the theoretical L subshell conversion coefficients of Sliv 9). (The individual theoretical L subshell conversion coefficients for the 122 keV transitions were interpolated from log-log plots of Sliv's conversion coefficients as a function of gamma energy; for the 257 keV transition the theoretical values for k = 0.5 were used.) The theoretical composite L line was constructed as follows: First the current position of the L3h line was determined from the current position of the K line and the known K-L~ electron binding energy difference. (Since the difference 1o) in the K and L natural line widths is less than 5 eV, it was assumed that the L line shapes are the same as that of the closest K line. In the case of the 122 keV L group, this was the K line of the 142 keV transition in 9°Zr (fig. 9), and for the 257 keV L group it was the 257 keV K line.) With use of the experimental line shape, the theoretical composite line was constructed with relative intensities of the individual L lines in the composite taken from the theoretical L subshell ratios. For the comparison with experiment, this theoretical composite L line was adjusted along the ordinate axis until the best fit of this line with the experimental data points was obtained. The validity of this method of analysis is shown in fig. 10, where a comparison is made between the theoretically-constructed L group for the known 132 keV E3 transition in 9°Zr and the experimental L group points. The agreement is seen to be excellent. The experimental points for the 122 and 257 keV L lines are compared in a similar manner to the theoretical composite lines constructed for various multipolarities, and the results are shown in figs. 11-14. This analysis determines unambiguously that the multipolarities of the 122 and 257 keV transitions are E2 and E3, respectively. In addition, the K / L ratio obtained for the 122 keV transition, 5.76 and that for the TABLE 2 Experimental and theoretical characteristics of the 122 and 257 keV transitions in 9°Nb Energy

Theoretical b)

Experimental

(keV)

E1 half-life

24 a)

10 -12

E2 2 × 10 -6

E3

E4

M1

M2

2

l0 s

10 -11

5 × 10 -5

M3 80

M4 10 s

(see) 122

K/L

5.76

8.65

5.90

2.90

~ 1

8.29

7.20

4.59

0.56

0.47

3.1

~35

0.116

1.09

7.2

53

1 × 10 -~a)

lO-la

lO-a

5 × 10 -~

10 a

10 -11

2 × 10 -~

0.9

10 6

5.64

8.77

7.50

5.57

3.58

8.56

7.73

~K

~ 0 . 5 ~)

half-life

3.05

(see) 257

K/L

6.45

5.24

a) These values were reported by M a t h u r and Hyde ~); in this paper the 24 sec half-life is assigned to another (undetected) transition. b) Theoretical half-lives are single p r o t o n values corrected for internal conversion, as read f r o m graphs o f W a p s t r a et al. 18). Theoretical conversion coefficients were taken f r o m the tables by Sliv and Band g).

DECAY OF e°Mo

127

257 keV transition, 5.64, agree very well with the corresponding theoretical K/L ratios. Table 2 contains a summary of the experimental and theoretical information on these transitions, other than the L subshell data, which was not obtained in numerical form. 3.5. T H E EC/fl + R A T I O D E T E R M I N A T I O N

The electron capture-to-positon ratio of 90Mo decay was determined by measuring the intensity of the 122 keV gamma ray (100 ~ transition abundance) relative to the 511 keV annihilation radiation. The source was located 20 cm from a Ge(Li) detector, 2 cm 2 x 7 mm deep, and was sandwiched between two Lucite absorbers to localize the creation of the annihilation radiation so that the same solid angle would be subtended I0

:

,

i

,

i

, ii

i.-i < ii

,,, 1 . 0

i= o

Ma93m I

0.I

500

I00 Energy

I000

(keY)

Fig. 15. G a m m a - r a y p h o t o p e a k efficiency curve for the 2 c m 2 × 7 m m deep Ge(Li) detector s y s t e m used for electron capture-to-positon ratio determination.

by the gamma rays and the annihilation radiation. Correction was made for the contribution to the annihilation peak from the daughter 90Nb decay. An experimental photopeak relative efficiency curve (fig. 15) for the Ge(Li) detector used (with a source-to-detector distance of 20 cm) was obtained by using the known equal intensities of the 122 and 257 keV cascade transitions in 9°Mo decay and the 262 and 685 keV cascade transitions in 93rnMO a s standards. The value so obtained for the EC/fl + is 3.0+0.5. This may be compared to the value of 1.9_ 0.5 obtained from the curves of Feenberg and Trigg a). The agreement is not unsatisfactory if one considers that there is possibly some pure electron capture to high-lying states. Ten percent pure electron capture to upper levels (which is

128

s . A . COOPER et al.

reasonable in terms of observed gamma-ray relative intensities) and subsequent cascade through the 122 keV state would give a value of 2.5_+0.5. The value for the EC/fl+(3.0) together with the half-life a) (5.7 h) and the positon end point 2) (1.15 MeV) determine a log f t value of 5.5_+0.3 for the 9°Mo positon decay. This suggests an allowed or first-forbidden (non-unique) decay from the 0 + ground state of 9°Mo to either a 0 + or 1 ± state in 9°Nb. 4. Discussion The definite assignment of E2 for the multipolarity of the 122 keV transition leads to a gross inconsistency because the reported half-life of 24 see for this transition is a factor of 107 longer than that predicted by the single-particle model for an E2 transition of this energy. In addition, there is a problem with the conservation of O+

5.7h

42M°48 EC

I ,r ( 0,1 +) (I +)

~+

t V2 I0 msec

380,0+~

U'2 E3 24sec 1 3 4 - + ) 14--) 16,7+)

~ e E2

(6+1

( 8 , 9 + 1 18+)

14.6 h --

122.5+~

( ~ < 3 keV)

IV22.5 0

Nb 90 41

49

Fig. 16. Partial decay scheme o f 9°Mo, showing postulated low-energy transition. The preferred state assignments are underlined.

angular momentum because the sum of the two multipolarities is at least 2 units less than the spin difference (7 to 9 units) between the state initially populated by the decay of 9°Mo and the ground state of 9°Nb. To explain the very long half-life for the 122 keV E2 transition and to conserve angular momentum, we propose a decay scheme (fig. 16) in which a very low energy (less than 3 keV) transition is postulated as preceding the 122 keV transition. The 24 sec half-life is then assigned to this very low-energy transition and not to the 122 keV transition. Let us consider the arguments bearing on spin assignments of the states in the main gamma cascade. The ground state of 9°Nb decays by more than 90 % to the 8 + state in 9°Zr and less than 0.1 ~ to the 6 + state 3). The logft value 3) (6.0) indicates

DECAY OF °°Mo

129

either an allowed or first-forbidden transition. According to the shell model, the lowest-lying configuration in 9°Nb (41 protons, 49 neutrons) should be the even parity configuration n(g~)lv(g÷) - i. Thus the transition from 9°Nb to the 8 + state in 9°Zr is probably an allowed transition. The above considerations exclude all spins except 8 + or 9 + for the ground state of 9°Nb. The Brennan-Bernstein rule 11) (which is based on experimental information and delta-force calculations by Schwartz 12)) that one less than the maximum spin lies lowest for odd, particle-hole nuclei, suggests that the ground state of 9°Nb has spin 8 and even parity. Other calculations by Kim and Rasmussen 13), which include a tensor force, suggest that the 8 + state lies lowest and that the next lowest state is 6 +. We choose these assignments for the lowest two states. The l o g f t value of 5.5 for the decay of 9°Mo signifies either an allowed or firstforbidden (non-unique) decay from the 0 ÷ ground state of 9°Mo (42 protons, 48 neutrons) to either a 0 ± or 1 ± state in 9°Nb. The proposed low-energy gamma transition is in cascade with the 122 keV and 257 keV transitions, so its total intensity must be approximately equal to that of the 122 and 257 keV transitions. Since neither a gamma ray nor an electron line of intensity comparable to that of the 122 keV transition was observed above 2.5 keV an upper limit of 3 keV can be set for the energy of this transition. Conservation of angular momentum and parity considerations help us restrict the possible multipolarities of the 24 see isomeric transition. If the state receiving the beta decay is 1 ÷ and 90Nb ground state is 8 ÷, an M2 assignment (with possible E3 admixure) is demanded for the 24 sec transition in cascade with the E3 and E2 transitions. If the initial state in the cascade is 1 -, an E2 assignment is indicated. If the initial state is 0±, an octupole assignment is indicated. Experimental evidence 14) on the levels in 91Nb (41 protons, 50 neutrons) shows an energy separation between the p4 and g~ proton orbitals of only about 100 keV, whereas the separation between the p~ and g~ orbitals is about 1.3 MeV. The experimental evidence 15) on the levels in 89Zr (40 protons, 49 neutrons) gives the energy separation between the p~ and g~ neutron orbitals as about 600 keV and that for the p~ somewhat greater. One therefore expects excited states involving the p~ orbitals to lie higher in the 9°Nb spectrum than those arising from the p~ orbitals. The coupling of a p~ proton (neutron) and g~ neutron (proton) can lead to 4- or 5- states, and the Nordheim strong rule 16) and that of de-Shalit and Walecka 17) puts the 4- level lower. We postulate that the next level above the 6 + in the main cascade of 9°Nb is of this character with spin 4 and odd parity. Admittedly, the assignment of 4- on shell-model grounds is the weakest link in our chain of reasoning; from the experimental data alone alternative assignments of 3-, 3 ÷ and 4 + cannot be excluded. The level assigned as 4- is fed by the 257keV E3 transition from the state that receives primary beta decay. The spin of the latter has already been restricted by ft value considerations to values 0 or 1, but of these only the assignment 1+ is consistent

130

J.A. COOPERet al.

with the 4 - assignment. There are several ways to couple the lower-lying g~, p~ and p÷ orbitals to form 1 +, and it is not obvious which configurations will be predominant. From table 2 we see that the E3 transition rate to the 4 - state is five times faster than the single-particle rate, according to the graphs of Wapstra et al. is). Of the 40 different E3 transitions plotted by Goldhaber and Sunyar 19), only 3 are enhanced; thus the unusual speed of this E3 transition suggests that there is considerable configuration mixing in the 1 + and 4 - states of 9°Nb. Descriptions of these states involving collective octupole phonon excitations may be worth exploring. Yttrium-88 (39 protons, 49 neutrons), which also has an unusually fast E3 transition (factor of 10 over single-proton rate) from a 1 ÷ isomeric state to a 4- ground state, has the same proton and neutron orbitals available as 9°Nb. This similarity in E3 transition-rate enhancement lends support to our proposed 1 + and 4 - level assignments. The 24 sec half-life of the proposed very low-energy transition ( < 3 keV) is not unusually short. The gamma-ray transition probability does decrease rapidly with decreasing transition energy but there is a corresponding increase also in the probability of decay by internal conversion. These opposing trends keep the half-life from becoming very long at very low energies. A case in point of this is the 26 min isomeric state in 235U which is < 1 keV above the ground state zo). In conclusion, the data reported here have established unambiguously the multipolarities of the two most prominent transitions in 9°Nb to be E2 and E3. The present experimental and theoretical information strongly suggests a level scheme for 9°Nb like that of fig. 16 in which a very low energy transition, less than 3 keV, from a 4 - state to a 6 + state has been proposed to conserve angular momentum and parity, and to explain the anomalous half-life of the 122 keV E2 transition. Note added in proof'. The existence of a very low energy transition in the decay of 9°Mo has been indirectly verified in a recent experiment by Cooper, Hollander and Rasmussen, in which it was demonstrated that the decay rate of the 24 sec isomer is dependent on its chemical environment. A 3.2 70 decrease in the intensity of the 122 keV E2 gamma ray and a return to its equilibrium intensity with a half-life of about 20 sec was observed following a rapid dissolution, with hot HNO3-HF, of niobium metal foils containing the isomeric state in equilibrium with 9°Mo. The size of the observed effect, an order of magnitude greater than that reported by Bainbridge, Goldhabel and Wilson 22) for the 2 keV transition in Tc 99", indicates that the transition depopulating the 24 sec isomeric state has a very low energy and decays by internal conversion in the outermost electron shells, where changes in chemical state can be expected to affect most strongly the electron density at the nucleus. References

1) R. M. Diamond, Phys. Rcv. 89 (1953) 1149 2) H. Mathur and E. K. Hyde, Phys. Rev. 98 (1955) 79

DECAY OF 9°Mo

131

3) S. Bjernholm, O. B. Nielsen and R, K. Sheline, Phys. Rev. 115 (1959) 1613 4) E. Feenberg and G. Trigg, Revs. Mod. Phys. 22 (1950) 399 5) 3. M. Hollander and R. L. Graham, Lawrence Radiation Laboratory Report UCRL-10624 (1963) 6) W. G. Smith and J. M. Hollander, Phys. Rev. 101 (1956) 746 7) F. S. Goulding and D. A. Landis, National Academy of Sciences, National Research Council, Washington (1963) Publication 1184 8) F. S. Goulding, Lawrence Radiation Laboratory Report UCRL-11302 (Feb. 1964) 9) L. A. Sliv and I. M. Band, Coefficients of internal conversion of gamma radiation (USSR Academy of Sciences, Moscow-Leningrad, 1956), Part I: K shell, Part II: L shell 10) J. S. Geiger, R. L. Graham and J. S. Merritt, Nuclear Physics 48 (1963) 97 11) M. H. Brennan and A. M. Bernstein, Phys. Rev. 120 (1960) 927 12) C. Schwartz, Phys. Rev. 94 (1954) 95 13) J. O. Rasmussen and Y. E. Kim, unpublished results; Izv. Akad. Nauk 29 (1965) 94 14) J. D. Prentice and K. G. McNeill, Phys. Rev. 104 (1956) 706 15) D. Strominger, J. M. Hollander and G. T. Seaborg, Revs. Mod. Phys. 30 (1958) 585 16) L. W. Nordheim, Phys. Rev. 78 (1950) 294 17) A. de-Shalit and J. D. Walecka, Nuclear Physics 22 (1961) 184 18) A. H. Wapstra, G. J. Nijgh and R. Van Lieshout, Nuclear spectroscopy tables (North-Holland Publ. Co., Amsterdam, 1959) 19) M. Goldhaber and A. W. Sunyar, in Alpha-, beta- and gamma-ray spectroscopy, Vol. IIed. by K. Siegbahn (North-Holland Publ. Co., Amsterdam, 1965) p. 941 20) F. Asaro and I. Perlman, Phys. Rev. 107 (1957) 318 21) P. Marmier and F. Boehm, Phys. Rev. 97 (1957) 103 22) K. T. Bainbridge, M. Goldhaber, and E. Wilson, Phys. Rev. 90 (1953) 430