Volume 195, number 2 PHYSICS LETTERS B 3 September 1987

NEW L I M I T S ON THE DOUBLE BETA DECAY HALF-LIVES OF 94Zr, 96Zr, 116Cd AND 1248n

Eric B. N O R M A N and Dawn M. M E E K H O F Nuclear Scwnce Division, Lawrence Berkeley Laboratory, Umverslty of Cahforma, Berkeley, CA 94720, USA

Received 8 April 1987, revised manuscript recmved 16 June 1987

Searches have been made for the double beta decays of 94, 96Zr ' 116Cd ' and 1248n to excated states m their daughter nuclm. No evidence of any of these decays was found, and lower hmlts on the half-hves against such decays have been estabhshed to be 10~8-10 ~9 yr. In addition, the single beta decay half-hfe of 96Zr has been estabhshed to be >~ 1.7 × 10~ s yr

Much of the current interest in the subject of dou- ble beta decay stems from ats relevance to the ques- tions o f lepton number conservation and of the mass o f the neutrino. Neutrinoless double beta decay can result from a finite neutrino mass and/or f rom the presence of right-handed weak currents. The prin- cipal transitions expected from the neutrino mass term are J~ = 0 ~ ~ J ~ = 0 + decays. However, right- handed currents can lead to J~ = 0 + -~J~= 2 + and J~ = 0 +-~J~ = 0 + decays with comparable strengths [ 1 ]. Thus, searches for this type o f decay are impor- tant for a complete understanding of the double beta decay process.

For a number of nuclei, double beta decays to excited states in their daughter nuclei are energeti- cally possible. A signature of this type of double beta- minus decay would be the emission of a gamma ray from the nucleus ( Z + 2 , N - 2 ) in a sample of the nucleus (Z, N). Except for the well-studied [2 -7] case o f 76Ge, there have been relatively few previous searches for double beta decay to excited states [8-12] . We have chosen 94Zr, 96Zr, l l6Cd, and ~24Sn for the present study. The possible decay schemes of these nuclei are shown in fig. 1. All energies, spins, and parities used in the present work are taken from ref. [ 13]. For each of these nuclei, double beta decay to one or more excited states in the respective daugh- ter nucleus is possible. In addition, for 96Zr single beta decay to 96Nb is also energetically allowed.

Visiting student from California Institute of Technology

126

To detect the gamma-ray signature of double beta decay to an excited state, a 110 cm 3 coaxial high- purity germanium detector was used. In order to maximize the amount of material that could be counted with high efficiency, each sample was dis- tributed around the detector. Several 5 × 5 cm plates stacked to yield approximately 1 cm total thickness were mounted directly against the front face of the detector. The rest of each sample, in the form of thin sheets, was wrapped around the cylindrical outer can of the detector to produce a total thmkness o f up to 0.5 cm. This assembly was shielded on all sides with 10-15 cm of low-activity lead. The experimental apparatus was located in the Low Background Counting Facility at Lawrence Berkeley Laboratory. In order to minimize background radiation, this facility was constructed with 1.3-1.6 m thick walls o f low activity concrete and has an air flow of 10 room volumes/h. All of the samples studied were natural isotopic composit ion metals of at least 99% purity obtained from Alpha Products. The masses and counting periods for all of the samples are listed in table 1. Data were accumulated in 1024 channnels using a multichannel analyzer and were recorded for off-line analysis.

Energy and efficmncy calibrations were performed using standard sources. To determine the actual detection efficiencies for our extended and rather thick samples, numerous efficiency measurements were made by placing calibrated point gamma-ray

0370-2693/87/$ 03.50 © Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division)

Volume 195, number 2 PHYSICS LETTERS B 3 September 1987

6 + 0 (a) 94Nb

0 + 0 94Zr ~ 2 0 1 ~ 871

QN3 = 1148 keV 94Mo- o

(b)

°+ 96 o (6 +) o

Q~ 3350 keV : ~'|778

96M0

(c) 1 + 0

0 + 0 116in -t ~ ~ ~, 2- 2225 ' u C d ~ ~ - . . . _ . - ~ - . ~ - : 2 + I I 2112 \ ~-------~0+11 II 2027

X X N o + l I I I / 1757 QI313 = 2809 keV ~ "'---.~2+1{1{,~11294

116Sn

(o)

3 - 0

0 + 0 124Sb 124Sn ~ ( 0 - ) 1056

QI313 = 2278 k e ~ +1[~ 1325 * 603

I o 124To

Fig. 1 Possible decay schemes of (a) 94Zr, (b) 96Zr, (c) l l6Cd, and (d) 1245n.

Table 1 Samples examined m the present study.

Element Mass Isotopic Counting period Photopeak (g) abundance (h) efficiency

(%) (%)

zlrconmm 646 17.4 (94Zr) 2 80 (96Zr)

cadmium 690 7.5 (116Cd )

tan 647 5.64 (I245n)

1565 1.8 1.9

668 1.2

666 2.4

sources at a variety o f locat ions on the front face and sides of the detector. Such calibrations were done first with no mater ia l between the source and the detector and then with several different thicknesses o f each o f the actual samples placed between the source and the detector. Using the d imens ions o f each sample, appropr ia te ly wmghted average pho topeak detect ion efficiencies were then computed. The results obta ined for the f i rs t -exci ted-~ground-state t ransi t ions sought from the samples used in the present s tudy are sum- mar ized in the last co lumn of table 1.

Fo r 94Zr, the only excited state that can be reached via double beta decay is the J ' = 2 + level at 871 keV in 94M0. For 96Zr, several levels could be popula ted

in double beta decay. However , they all decay with large branching rat ios to the J ~ = 2 + level at 778 keV 96Mo. Consequently, a search for this gamma ray alone can be used to establish l imits on the double beta decay rates to all o f these states. Similarly, for l l6Cd and 124Sn, searches for the 1294 keV and 603

keV gamma rays, respectively, can be used to set l im- its on all o f the possable double be ta decay modes.

127

Volume 195, number 2 PHYSICS LETTERS B 3 September 1987

> G) ,4

o

t -

O o

2500 -

2000 -

1500

1000

500

o L ~

I ' 96Z r I 183'2 I (a) 7 7 8 I 911 -

727 766 94Zr

871

/

I . . . . I _ L _ ~ 700 800 900

I I I I ~11115 116, , - , (b)l h1120 L,O 1000 ]j~ 1173 1 2 9 4

" "" U .... H ~ - 4 - ~ 1 L . i qi 12 - tO .q. 6O0 ~ ~ t-

400 0

200- -

I _ I I I i 0 1100 1200 1300

I I I 8000--

>m 6000 -

0

{,':' 4000 -- t - .

511

I(c)

124Sn

6 0 3

o = 583

2000 -

o - L I I L 500 60o 7o0

E (keV)

Fig. 2 Relevant pomons of the gamma spectra observed m counting (a) 912 h with the zlrcomum sample, (b) 668 h with the cadmmm sample, and (c) 666 h wath the tin sample Peaks labeled only by energy are due to long-hved background actavl- taes (such as thorium and uramum) contained an the shaeldmg materml and/or an the detector atself.

The relevant portions of the spectra observed from the zxrcomum, cadmium, and t in samples are shown in fig. 2. In all of these spectra, the only gamma-ray lines observed are due to the decays of long-lived iso-

topes such as 232Th, 235U, and 238U contained in the

shielding material and/or in the detector itself. The expected positions and widths of the various lines from double beta decays were determined from those of known lines that appear in the spectra. The num- ber of counts observed in each such window was determined. In each spectrum, background windows containing the same number of channels as the peak window were chosen above and below each peak. Using a l inear fit to the background, the net number of counts in each peak window was determined. No excess of counts above background was observed at any of the expected peak positions. Taking the meas- ured 1 cr uncertainties as upper limits on the numbers of counts in each peak, we established 68% confi- dence level lower limits on the applicable double beta decay half-lives of 94Zr, 96Zr, l l6Cd, and 1245n. These

results are summarized in table 2. The results obtained m the present study can be

compared to theoretical estimates of neutrinoless double beta decay half-lives. The quant i ty fi/2rl 2 cal- culated using the formula of Rosen and Primakoff [ 14] is listed in table 2 for several transitions. In all of these calculations, the nuclear matrix element was arbitrarily taken to be 0.1. From our experimental results, one can set a limit on the lepton number non- conserving probabili ty q2 of ~ 0.027. While we are

Table 2 Lower hmlts on the half-hves of 94Zr, 96Zr, 116Cd, and t24Sn estabhshed m the present study compared wxth theoretacal estimates

Decay Final state tl/2 (yr) tt/2q 2 (yr) J~, Ex (keY) ( expt. ) ( theor. )

94Zr---*94Mo 2 +, 871 1 3y1019 1 2X1022

96Zr---*96Mo 2 +, 778 2 O× 1018 5.3X 1016 0 +, 1148 1 8X1018 2 + , 1498 1 3×1018 2 +, 1626 1 7X10 TM

4 +, 1628 1.8X1018

ll6Cd--->ll6Sn 2 +, 1294 2.7N1018 40X1017 0 +, 1757 24N10 Is 0 ÷, 2027 2.5X10 ~s 2 +, 2 1 1 2 1.0X1018 2 +, 2225 1.6X10 ~s

t24Sn---~ 124Te 2 +, 603 2 4× 10 TM 2 2>< 1017 2 + , 1325 2.0X10 ~8

(0+), 1656 2.2X 1018

128

Volume 195, number 2 PHYSICS LETTERS B 3 September 1987

not aware o f any previous searches for the double be ta decays o f 94Zr, 96Zr, l l6Cd, or 1248n to excited

states, searches have been made for neutr inoless ground-s ta te - -ground-s ta te double be ta decays o f these nuclei. Zdesenko et al. [ 15 ] have establ ished that for this mode the half-life o f 96Zr is >13× 1019 yr. Win te r [16] showed that the half-life o f ll6Cd against this type of decay is >~lX1017 yr, and McCar thy [ 17 ] set a l imi t on the corresponding half- life o f 124Sn of >~ 1.5× 1017 yr.

Our results can also be compared to those obtained in searches for double beta decays to excited states per formed on other nuclei. In the case of 76Ge, where the sample under study is the detector itself, a num- ber of groups [ 3 -7 ] have set lower l imits on the half- life against neutrmoless decay to the J~ = 2 + level at 559 keV in 76Se that range f rom ( 1 . 6 - 8 ) × 1022 yr.

Bellotti et al. [8] employed essentially the same method as that used in the present s tudy to establish corresponding l imits on the half-lives o f SSNi, 92M0, I°°Mo, 148Nd, and ~5°Nd that range f rom 2X 10 Is yr

to 4 × 10 t9 yr. A g a m m a - g a m m a coincidence tech-

nique was used by N o r m a n and DeFacc io [ 9 ] and by Norman [ 10] to set similar l imits on the half-lives of 5SNi, 96Ru, and l°6Cd that vary from 5 × 1016 yr

to 5 N 10~9yr. Baskov et al. [ 11 ] used a sodium iodide detector to show that the half-hfe o f 13°Te against double beta decay to excited states in I3°Xe is >2 .5M 10 j9 yr. More recently, Barabash et al. [ 12] ut i l ized a xenon-fi l led ioniza t ion chamber to set a l imi t on the half-life o f 136Xe against decay to the J ~ = 2 + level at 819 keV in 136Ba o f 4.9X 102o yr.

The results ob ta ined from the z i rconium sample can also be used to estabhsh a l imi t on the single be ta decay half-life o f 96Zr. The product o f this type of decay is 96Nb which decays with a 23.4 h half-life to excited states in 96Mo that cascade through the J ' = 2 + level at 778 keV. Thus the absence of this peak in our spectrum allows us to establish that the single beta decay half-life of 96Zr is i> 1 .7× 1018 yr.

This is approximate ly a factor of five greater than the l imit es tabl ished by Eichler et al. [ 18 ]. Assuming that the pr incipal single beta decay mode o f 96Zr is

the four th-forbidden t rans i t ion to the J~ = 4 + level at 142 keV exci tat ion energy in 96Nb, we f ind that the l o g f i for this decay is I>22.3.

We wish to thank A.R. Smith for his assistance in the use of the Low Background Counting Facili ty. We also thank K.T. Lesko for assistance with some o f the da ta analysis and for his comments regarding the manuscript . One o f us (D.M.M.) wishes to thank the Monticel lo Founda t ion for suppor t during her visi t to LBL. This work is suppor ted by the Director , Office o f Energy Research, Divis ion o f Nuclear Physics o f the Office o f High Energy and Nuclear Physics o f the US Depar tmen t o f Energy under Con- tract No. DE-AC03-76SF00098.

References

[ 1 ] M. Din, T, Kotam and E Takasugl, Prog. Theor. Phys. Suppl. 83 (1985) 1.

[2] A. Forster et al, Phys. Lett B 138 (1984) 301. [3] J.J. Simpson et al., Phys. Rev Len 53 (1984) 141. [4] E Bellottl et al., Phys. Lett. B 146 (1984) 450. [5] D.O. Caldwell et al, Phys. Rev. D 33 (1986) 2737. [6] H. Ejan et al., Nucl. Phys. A 448 (1986) 271. [7] F.T. Avlgnone III et al., Phys. Rev C 34 (1986) 666. [8] E Bellottl et al., Lett Nuovo Clmento 33 (1982) 273. [9] E.B. Norman and M.A DeFacclo, Phys. Lett. B 148 (1984)

31. [10] E.B. Norman, Phys. Rev. C 31 (1985) 1937 [ 11 ] M.P. Baskov et al., Proc. Intern. Conf. Neutrino '82 Vol. 1,

eds A. Frenkel and L. Jemk (Budapest, 1982) p 202. [ 12] A.S Barabash et al, Proc. 12th Intern. Conf. on Neutrino

physics and astrophysics, eds T. Katagakl and H. Yuta (World Scientific, Singapore, 1986) p 110

[ 13 ] C M. Lederer and V.S Shirley, Table of isotopes, 7th Ed. (Wdey, New York, 1978)

[ 14 ] S.P Rosen and H Pnmakoff, m: Alpha-, beta-, and gamma- ray spectroscopy, ed. K. Slegbahn (North-Holland, Amsterdam, 1965)p 1499.

[15] Yu. G. Zdesenko et al., Izv. Akad. Nauk (SSSR), Ser Flz. 45 (1981) 1856.

[ 16] R.G. Winter, Phys Rev. 99 (1955) 88. [17] J.A McCarthy, Phys. Rev. 90 (1953) 853. [ 18 ] E. Eichler et al., Phys Rev. C 10 (1974 ) 1572.

129