Double beta decay and neutrino mass. The Heidelberg-Moscow experiment

Prog. Part. Nucl. Phys., Vol. 32, pp. 261-280, 1994 Copyright© 1994ElsevierScienceLtd Printedin Great Britain.All fightsreserved 0146-6410/94 $24.00

Pergamon

0146-6410(93)E0022-I

Double Beta Decay and Neutrino Mass. The

Heidelberg-MoscowExperiment H. V. KLAPDOR-KLEINGROTHAUS* Max-Planck-lnstitutflir Kernphysik, Heidelberg, Germany

ABSTRACT: Double beta (13[~)decay probes beyond standard model physics and at present is the most sensitive tool to probe the electron neutrino mass. The experimental and theoretical status of 13~ research is briefly reviewed. The main emphasis is put on the HEIDELBERG-MOSCOW experiment, which is undertaken in the GRAN SASSO laboratory and which at present has the largest sensitivity and gives the sharpest limits. These are from 2924 kg d of data taking with 6 kg of enriched (86%) 76Ge detectors: Ti/2°v(0+-,0+)>1.9×1024 y and r~n°v(0+--~2+)>8.0x1023y with 90% c.l. They correspond to an effective neutrino mass limit of < my >< 1. I eV. For the neutrinoless decay with majoron emission a half life of 1.7 x 1022 y can be excluded leading to an upper limit for the neutrino-majoron coupling of < gvz > < 1.8 • 10-4 (90% c.L). The background around 2 MeV is b --- 0.2 counts/kg y keV for the total array. The experiment opens new perspectives for the investigation of exotic processes like electron decay, solar axions and dark matter. It was, e.g., possible to improve the existing cross section limits for WIMP masses above ~ 150 GeV and to exclude Dirac neutrinos in the range 26 OeV to 4.7 TeV as the dominant component of the dark halo. The new 13[3-technology has found also application in high resolution balloon and satellite gamma-ray astronomy. Concerning the matrix elements for 1~13decay a major step beyond the frequently used QRPA model has been made by applying the Operator Expansion Method (OEM).

KEYWORDS Double betadecay, neutrino mass 1. INTRODUCTION The neutrino is one of the best examples for the merging of the different disciplines of micro- and macrophysics. The neutrino plays, by its nature (Majorana or Dirac particle) and its mass, a key role for the structure of modem particle physics theories (GUTs, SUSYs, SUGRAs .... ) [1-3]. At the same time it is candidate for non-baryonic dark matter in the universe, and the neutrino mass is connected, by the sphaleron effect, to the matter - antimatter asymmetry of the early universe [4] (Fig. 1). In the search for the neutrino mass we can differentiate between different approaches (Fig. 2), which may be

* Speaker of the HEIDELBERG-MOSCOW collaboration

261

262

H.V. Klapdor-Kleingrothaus

classified as terrestrial and extra-terrestrial, among the latter the most spectacular being the investigation of solar neutrinos and neutrinos from supernovae. In principle also the up to now unobserved cosmic neutrino background radiation contains a mass information. Since the neutrino oscillation experiments measure the difference of the (squared) mass eigenvalues of different neutrino flavors ~un2 =[m~-m2l , they primarily yield information on the heavier neutrino flavors and thus are complementary to those non-accelerator experiments like ~l and ~J3 decay measuring directly the electron neutrino mass. The sharpest limits on the

DOUBLE L. u", c^DI BETA DECAY |n"n"u-~""IREACTOR

h

I

[

IPHYS,CSl~TA0~CCA~EXeL~I

/

WHICH [ SYMMETRY• I

' ' G"T

/1 S~ I~0~)~ I

v

Fig. 1: The neutrino and its role in micro- and macrophysics

SEARCH FOR NEUTRINO HASS TERRESTRIAL NON-ACCELERATOR Tritium decay etc. my,<12.5eV mv~=17 keY ??

[

ACCEL.

~,~-decay

(SIN. AReUS) mv~< 0.25 HeY

mv.c<35 MeV

JDouble beta decay , mv,< 1.3eV

EXTRA- TERRESTRIAL SOLAR COSHIC

J v-oscilL

[

Solarv

[ v- background

oscill., MSW

radiat.

(T=T(mv))

Supernovae I flight time [ v- decay l

I

v-oscill. reactors

J

v-decay reactors

Fig. 2: Classification of experiments investigating the neutrino mass

The Heidelberg-Moscow Experiment

263

electron neutrino come at present from non-acceleratorexperiments: tritium decay and double beta decay. Both of these experiments again are complementary: while the effect of a neutrino mass in the ~ decay spectrum of tritium is independent of the nature of the neutrino (Dirac or Majorana particle), double beta decay can measure only an effective Majorana mass of the neutrino (Fig. 3). So only study of both processes together gives the required information on the Dirac and Majorana mass terms in the neutrino mass matrix. Double beta decay can also probe in a very sensitive way the hypothesis of a 17 keV neutrino, if this is a Majorana particle, by its influence on the effective mass observed in ~ decay.

vc_ v

~

v

=

[ c, VL

J

+v R

Dirac-Neutrino

Majorana-Neutrino

Fig. 3: Possible assignment of the experimentally known (in boxes) neutrino states (of one family) in the theoretical description for Dirac and Majorana fields There are several decay modes discussed in J315 decay, the main ones being two neutrino (2v) and neutrinoless (Or) decay (Fig.4)

~zX "->~+2X+2e- +2v, ~x-~+~ x + 2 f .

(1)

(2)

In addition there could occur Ov decay accompanied by emission of one or two majorons (Fig.4) P e

P

llf...I Ill n n (b)

(a)

P Majoron

P

n

P

W

~

e-

vt~X W

(c)

or "~ •

N (a)

Fig. 4: Graphs for 2v (a), 0v Co), 0v X (c) and ~

(d) double beta decay

H. V. Klapdor-Kleingrothaus

264

Ix -~. A

x+2,, +x

(:3)

A

zX--~z+2 X+2e- + X + Z

(4)

Fig. 5 shows the corresponding spectral shapes of the electron sum spectra. The first process is a usual effect of second order in the classical weak interaction, the second requires as a prerequisite a non-vanishing Majorana neutrino mass or a right-handed weak interaction. In the framework of GUTs, ore more general gauge theories, both of these conditions cannot be seen independently, but in any case a non-vanishing mass is required. In left-right symmetric models more exotic processes then (1)-(4) might contribute to the 0v~3

2.0

~

p00vx

1.S 1.0

o: o

i

o2s

o'5

0.75 ' Energy E/Eo

;

Fig. 5: Shapes of the electron sum spectra of different ]3~ decay modes amplitude like exchange of very heavy neutrinos or Higgs particles (see e.g. [3]). 0 v ~ decay allows to deduce contraints on the masses of the heavy neutrinos introduced e.g. by the see-saw mechanism, and on masses of right-handed W-bosons (see, e.g, [5]). In SUSY models modes connected with gangino exchange might dominate 0 v ~ decay [6] (see Fig. 6),

n

~

(a)

(b)

Fig. 6: Graphs contributing to 0 v ~ decay in left-right symmetric models by Higgs exchange (a), and in some SUSY models by gaugino exchange (b) (from [3,6]) There am three possibilities to break BoL in theory [3], which is required to produce a Majorana mass: (1) explicit B-L breaking, i.e. the Lagrangian contains B-L violating expressions; (2), O) spontaneous breaking of a local or global B-L symmetry. The existence of the majomn Z is associated wiht the third possibility. From the different majoron models characterizod by their weak isospin, cormected with different possibilities to generate Majorana mass terms in extensions of the standard model two am ruled out by the recent LEP measurements [7]: the ~plet majoron [8], which would contribute an effect equivalent to two neutrino flavors to the Z° width, and the doublet majoron [3], which should contribute half a neutrino width.

The Heidelberg-Moscow Experiment

265

However, the existence of singiet majorons or a mixture of singlet- and doublet majorons is possible and it might show up in [~[3decay (see e.g. [9]). Investigation of the doublet majoron, which would be associated with a double beta decay mode with emission of two majornns, could allow to make conclusions on the Zino mass [3]. The model of singlet majorons has recently experienced large interest in connection with attempts to construct neutrino mass hierarchies involving a 17 keV neutrino [ 10,11 ]. It is therefore important to search for this decay mode. It has to be noted that if the electron neutrino is a mixed state (mass matrix is not diagonal in the flavor space):

Iv.)=~u.,lv,)

(5)

l

then the measured quantity in double beta decay is an effective mass

(6) 1

which for appropriate CP phases of the mixing coefficients Uei could be smaller than m i for all i. Neutrino oscillation experiments yield some restrictions of this possibility (see [2]. In general not too pathological models seem to yield mve=(mve~ (see [1]). We can differentiate between two classes of direct (non-geochemical [$~ decay experiments (Fig. 7): a) active source experiments (source = detectors) b) passive source experiments. In the first class of experiments the [3~ process usually is identified only on the basis of the distribution of the total energy of the electrons. The second class of experiments yields more complete information on the [~l~ events by measuring time coincidence, tracks and vertices of the electrons and their energy distribution. But

[DOUBLE BETA DECAY EXPERIMENT_S[ [Passive sourceI I Active source I ISel up with detector for I

1

enerLy me~urements

J

IS., .p J

energy detectors

EFFICIENCYsSO'IO0% RESOLUTION ~).e.Z-le~,

" ° + + + + + "

Information:

f(EI÷E2); time

1 I ,.ot ~.o.b: Lim ~

:

i r(Es); rmz); ,EI+E2);I I time; coincidence

1

> 1024 yJ

~ , . = lO21 y ( .

'j

Informatlnq:

J

r(Ea); KE2); t(EI+E2);

J time; coincidence; L.vertex; t(ces o)

1 Jim ~ a 10"2yJ

+ [Lira ~ a lOZZy J

1~-- ]o,,-to- yj Fig. 7: Classification of ~[} decay experiments

I

266

H. V. Klapdor-Kleingrothaus

also time projection chambers (TPCs) using ~15 active counting gas are belonging to the first class - such as the Gotthard 136Xeexperiment It is obvious from Fig. 8, which shows an overview over measured 0v~13 half-life limits and deduced mass limits, that the largest sensitivity for Ov131~decay is obtained at present by active source experiments, in particular 76Ge [12] and 136Xe [13] (only the geochemical experiment with ]2STe reaches a similar limit). The main reason is that large source strengths can be used (simultaneously with high energy resolution), in particular when enriched 1313emitter materials are used.

1313 Half lives 1024 -

r-lov

1022 1020

[q 1018

I

1016 10I~ 1012

I

&6Ca liSNi 7 I10R 11 ',d 1&SN 4~i m '/0Zn 82Se Q4Zr 100No114Cd12&Sn130Te136XellSON

Upper limits for 1.2

[1/ev]

1

0.8 0.6 0.4 0.2 0

48Ca76Ge

82Se 100Mo 116Cd 130Te 136Xe 150Nd

Isotope

Fig. 8: Measured Oval3 half-life limits (a) and deduced mass limits Co)

The Heidelberg-Moscow Experiment

267

Other criteria for the "quality" of a [~[~emitter are: • a small product ~ "(Y/iv)2, i.e. a large matrix element M °v • a QI~I3value beyond the limit of natural radioactivity (2.614 MeV). The future (next 5 years) of [3[~experiments will be dominated by use of enriched detectors, 76Ge playing a particularly favourable role here [12,14], and enriched source material such as 136Xe (see [13]), 100Mo [15], U6Cd [16] (Fig. 9). Such experiments can probe the neutrino mass in the next years down to about 0.1 eV (see Fig. 18).

1026_ 1025 -

?

IoEx' .

C9.

>

NEMO lOOMo~

1okg I

I

v0

Heidelberg-Kiev

Liquid 13eXe ?11eCdWO t 'IP TPC (USA,

,2ok.

I

I

|

I I

1999

'Supercad'

IHeidelberg-Hoscow[

I I

, I

|

I

*

I Milano Z', Caltech130T.02 Neuchetet-..

4 ~"

TPC

I ci.~ [ "~ Xlano

1993 1021

• 76Ge 82Se 100Mo 116Cd 130Te 136Xe 150Nd

48Ca

Fig. 9: Present situation and perspectives of the most promising ~ experiments. Only for the isotopes shown 0vl~[~ half life limits > 1021 y have been obtained. The thick solid lines correspond to the present, 1993, situation, open bars and dashed lines to 'safe' and 'less safe' expectations for 1999 (from [16]).

2. CALCULATION OF [~[~MATRIX ELEMENTS, QRPA AND OEM For deduction of an (effective) neutrino mass (limit) from a measured 0v decay rate (limit) calculation of nuclear matrix elements is required. The half life for 0v[3[~decay is given by

[T.°~m: --.Op]-'--C...<'~,>~ + C,, <,>~ +C.," m e

+C~ <11> ÷C,~<~.> +C,x <.q><~.> me

(7)

me

or, when neglecting the effect of fight-handed weak currents, by

[~(o~' --,op]-' --c...<'~ >~ -_(M~-M,o"o.)~.._.., <,,v: me

m e

(8)

268

H.V. Klapdor-Kleingrothaus

where G 1 denotes the phase space integral. The half-life for Ovz decay is given by [T~,2]-~ = ( ~ - - M°*)2 FO~ (< 8,~ >)~

(9)

Here the neutrino-majoron coupling constant is given by < g,z >=,T=,g,~v,,v,J

(I0)

i,j

and F0vx denotes the phase space. The half life of the 0v~ decay is

[r,,,]-' =<:=-m,)'(Ug.-U~">'F °'=

(11)

The coupling constant fxx here is connected with the Zino mass according to

fzz = 4M z

g2 cosOw

(12)

After the major step of recognizing the importance of g.s. correlations for their calculation [17], in recent years the main groups used the QRPA model for calculation of M °v. It was found that in most cases the uncertainties of the model were tolerable in the case of 0 v ~ decay (see [18] and Fig. 10). A list of the calculated 0v~3~ half-lives for most potential ~-~--emitters is given in Table 1. The problem of QRPA extreme dependence of the calculated rate (for 2v decay) on the choice of the renormalization of the particleparticle force, gpp, seems recently to have been solved by applying the so-called Operator Expansion Method (OEM) [ 19-21 ]. This method does not explicitely use the intermediate energy spectrum. This is the essential advantage over QRPA, since in this way the dependence on the pp force (which affects the distribution of ~3strength in the intermediate nucleus) is drastically reduced. The method allows Coy ignoring many-particle scattering terms) to write the matrix element for 2v decay in the form

10

44

&

1,,i,ii

0v~p Paris Potm~tial

IIIIII[[,:

II[Ir,

A

Fig. 10: The uncertainty of QRPA-calculated 0v~13 half-lives originating from the limited knowledge of the pp-force for the potential Ol~emitters (from [18])

The Heidelberg-Moscow Experiment

Nucleus

T ~ ( m , ) 2 (y . e V 2)

269

T ~ ( n k ) I ( y . eV z) q

TOZn(-)

9.83. lO =s

ta4Xe

1 . 8 9 . 1 0 =s

rage

2.33.1014

laeXe

2 . 2 1 - 10 =4

e°Se

1 . 1 4 . 1 0 =r

14=Ce

2.77- 10 =4

aaSe

6.03- 10 =a

t4eNd

4.36.10ae(D)

SeKr

2.78. l0 ss

t4SNd(')

1.36.10a4(v)

94Zr(S)

3.97. lO ss

tS°Nd

3 . 3 7 . 1 0 ==

mZr(n)

5.30. l0 ss

1|4Sm

1.39. 10 =4

~ M o (m)

1 . 0 5 . 1 0 =r

ZeOGd

8.56. I0 =a

X°°Mo(*)

1.27.10=4

tTOEr

1 . 3 7 . 1 0 =s

t°4Ru(*)

4 . 2 4 . 1 0 =4(D)

t ~eyb

1.36. I0 =4

xx°Pd(*)

1 . 9 8 . 1 0 =4(~)

lee W

6.35. 1024

XI4Cd(B)

5.07. I0 s=

=*SOs

4 . 0 8 . 1 0 =4

XteCd

4.87. I0 as

IOSpt

4 . 7 0 . 1 0 =s

taaSn

1.27. l0 se

=O4Hs

8.22. 10 =4

la4Sn

1 . 3 6 . 1 0 =4

=e=Th

3.07.102e

t =~.t.,e

T.TT. 10 =4

su U

2 . 6 0 . 1 0 =a

t ~>]'e

4 . 8 9 . 1 0 =e

=44ptI

5 . 7 2 . 1 0 =e

|

Table 1: Calculated 0v half-lives for all 1~[3emitters with A > 70 (from [18]).

M~ = (O;~[X MU'=;"=~'[O~'/="o +', \ I'S / 12(~(r)- ~(r))f/o(ij) M# = A2_16(¢o(r)_~)~(r))2 .~ 4(2~)e~(r))- ~)o(r)- ~)~(r))fl=(iJ) A2 - 16(2ve~(r)- v: (r)) 2

(13)

(14)

Thus the matrix element involves the bare nucleon-nucleon interaction without any adjustable parameter. The wave functions of initial and final states we take from QRPA. Figure 11 shows the results for t00Mo and 23su. Calculations for all potential [~ emitters by this method are given in [20]. 0v[3~ matrix elements calculated by OEM are very close to those of QRPA [21]. A list of calculated 0v[3J3 half-lives for most potential [3-[3-emitters is given in Table 1. That the dependence on the pp force is overestimated by QRPA has been shown recently also by comparing QRPA and shell model for 48Ca [22]. It may be noted that the calculation of 23su ~[~ decay by OEM yielding T~; = 0.9x 102=y [20], was performed before the experimental

270

H.V. Klapdor-Kleingrothaus . i , , l , l , , , ,

I

1023

12 \

I

--

118

I

I

I /

.~

,m/j: o,,~ , , , . y O4

ioo '~-o+

/

<.~,>2

.~, 1023 ....

/

I I

~l

\

I

I

-:_m H

\

/

OEM * QRPA

- -

\

I

e'Z~U: 2 v p p decay

\

/

\

/

\ .-

\

102t ~

*J

QRPA

Wgo .....

-0.8

~(~,t2)

I0 ~

...... m m a ~ k (~.,,.te~

.1.2

0

. . . .

(12

O~

'

. . . .

O~

O~

O~

+ ~

1

I

t

0.2

I

I

0.4

I

I

o.e

I

I

o~

I

0n

f

1.0

Fig. 11: The calculated matrix elements M~.~of 2v[~l~ decay of 100Mo (a) and the half-lives of 0v and 2v decay of 238U as function of the strength of the particle-particle force (gpp) in QRPA and OEM (from [20]). result of [23] was known to us. The far-reaching conclusions drawn in that paper seem not to be on stable grounds. Matrix elements for [~+13+, ~+EC and EC/EC decay have been recently calculated [24].

3. STATUS OF THE HEIDELBERG-MOSCOW EXPERIMENT Use of enriched 7('Ge (86%) instead of detectors from natural Ge (containing 7.8% of 7 6 G e ) could explore the half life of 0 v ~ decay up to _> 102s years and correspondingly the neutrino mass down to - 10"1 eV, probing a class of left-right symmetric GUT models with a right-handed Majorana mass term of about 1 TeV, based on SU(2)L O SU(2) R ® U(1) (see [1], [12]) The HEIDELBERG-MOSCOW ~ experiment [12,25,26] makes use of 16.9 kg of 7(,Ge metal enriched to 86%, corresponding to 14.5 kg of the isotope 7~,e. Use o f - 10 kg in form of enriched detectors would correspond in sensitivity to a setup of more than 1.2 tons of natural Ge detectors (see [12]). Up to now one enriched detector of - 1 kg, another of 2.9 kg (the largest ever produced Ge-deteetor at that time) and another three detectors (or crystals) of 2.5, 2.9 and 3.5 kg, respectively, have been produced. The first detector is running in the GRAN SASSO Underground Laboratory in Italy since end of July 1990, the second since September 1991, the third since August 1992 (Fig. 12). Ov -decay: The results of the first 2924 days x kg of measurement are:

~1/~(0+ --->0+)> 1.95x1024 (90% c,l.), (Fig. 13) the corresponding limit for the neutrino mass is m v < 1.1 eV. The HEIDELBERG-MOSCOW experiment thus yields now the most stringent limit for the 0v~l~ half life of 76Ge and the sharpest limit for the neutrino mass from detector experiments (for details see [12,25,26]). On the other hand we seem to observe (Fig. 14a) a systematic excess of events beyond the expected background precisely at the energy of a 0vl~13 (0 + ~ 0+) transition (present significance - 2.4~). This coincides with the observation of a stagnating 0v[~ half-life limit (Fig. 14b) as function of measuring time inspire of decreasing background. These effects require further attention. For their further investigation we have developed [27] a method of pulse shape analysis which will allow rejection of ~ 80% of multiple (non-~13) events. The background level which characterizes the quality of the set-up is 0.29 events/kg year keV (Fig. 13) in an 80 k e y interval around the hypothetical 2038.5 keV 0vl~[]-line for the total spectrum, and 0.23 events/kg year keV at present. As an upper limit for a [ ~ transition to the first excited state of S2Se which should occur at 1479.5 keV, we deduce a half-life limit ~ ( 0 + ~2+)>8.2x1023 (90% c.L), i.e. far beyond the half-life o f 2.5 x 1022 years claimed by [28] (Fig. 13).

The Heidelberg-Moscow Experiment

271

Fig. 12: The 13l~ laboratory of the HEIDELBERG-MOSCOW experiment in the GRAN SASSO near Rome (a) Mounting of the enriched detectors under low-level conditions (b)

Expeoted "-7 ~ov (0+~0+) line

161 12 lO 8

I~l)ov (0+-"2 +) line~ n ! / 'i i

2~ 0

2000 2010 2020 2030 20~0 2050 2060 2070 2080 Energy [keV]

OI 1470

1480

1490

1500 1510 1520 1530 1540 1550 ;'nergy [keV]

Fig. 13: Details of the spectrum of the HEIDELBERG-MOSCOW collaboration in the region of neutrinoless double beta decay to the ground and first excited state of the daughter nucleus after a measuring time of 2924 kg d.

272 C

o u n t $

H. V. Klapdor-Kleingrothaus Y 10z e 8 P $

3S 30

2~ 20

lOZ~ ~ 15 10

o! o

2O

to

60

~°=o

80

20

4o

7~.~ Exposure[mot,y]

60

80

7¢~ ExpOsure[mo(,y]

Fig. 14a: Number of expected background counts (dashed line) and measured total number of counts (solid line) in the 30 energy range of a 0vl~[] (0÷ ~ 0÷) transition of 76Ge, as function of measuring time in the HEIDELBERG-MOSCOW experiment. The expected background is extrapolated from -30 keV ranges left and right from the centroid energy (2038.6 keV). The deviation under a no-line hypothesis between observed and expected counts corresponds to -2.40 (from [31]). Fig. 14b: Deduced 0vl][~ half-life limit for 76Ge as function of measuring time in the HEIDELBERG-MOSCOW experiment (from [31]). 2v -decay: For 2v[3~ decay subtraction of the background contributions from cosmogenically produced

isotopes, natural decay chains, 21°pb etc. from the spectrum measured with the second detector in 615 kgd .v2~~ .-~0+)ffi(1.42+0.03(stat)+0.13(syst))102] y. This is yields [29] the spectrum shown in Fig. 15 and ~,2,~n÷ 2OO

£~rtmMatt~a

..

Ktrie-~ot

tO. " .~.

~

": ..,'.

~

.. .....

i! ~i~ ,-

11.

j

. . . .

5OO

'.Y"-'~,..'

-:.-"~ .~

.

,

.

eoo

.

.

, . . . . I .... 1000 1500 Energy [keV]

.

,

Iooo

.

.

.

.

,

~oo

~'~.-2O00

.

.

.

.

2ooo I

."T"l-i"

I

25OO

Fig. 15: Measured spectrum after 615 kg d (dotted curve, lines cut out) with the second detector, the resulting spectrum after subtraction (see [29]) of the background (solid histogram), and a fitted 2v speetmm with ~ =1.42×I02t y (lower dashed-dotted curve). The dashed curve corresponds to a half-life of 0.9 times 1021 y claimed by [34].

The Heidelberg-Moscow Experiment

273

probably the first undoubtable evidence for this up to now rarest nuclear decay mode. In the derivation of the experimental 2v spectrum of Fig. 15 for the first time a parameter-free background model basing on quantitative identification and localisation of the different radioactive impurities and estimate of their effect by the code GEANT has been applied. For details see [30] (Fig. 17). Table 2 which compares existing observations of 2v[313 decay, demonstrates that concerning signal to background (S:B) and statistical significance (number of events, exposure time and source strength), the HI~JDELBERG-MOSCOW experiment opens a new era in this type of experiments. Ov ~ -decay: For Majoron-accompanied, 0,x, [3[3decay, subtraction of the 2v-spectrum from Fig. 15 leads to the spectrum of Fig. 16 and an upper limit T°/~>l.7xl0 22 y (90% c.1.) and for the neutrino-majoron coupling constant to < g,x >< 1.8x 10"~ (90% c.1.) [32] thus excluding some recently discussed [33] indication of this decay mode. This is the sharpest limit obtained from detector experiments (see Table 3). Table 2: Comparison of all exclusive measurements of 2vj3~ decay. S:B denotes the signal to background ratio in the evaluated energy interval. The percentage denotes the fraction of the total 2v[3[3intensity contained in this interval. The number of events used to derive the half life is also given (from [29]). Isotope

T~7=

Interval

[keY] [10~° y] S2Se UC Irvine [40]

lOOMo Osaka, [41]

1 l+e.t

"--o.s 0 11~+°'°a • ~-o.o=

lOOMo INR [42] teeMo UC Irvine [43] 7eGe

[TEP-Yer.[44]

700 - 1600

62

Expmure

S:B

[mol.y] 1

36

0.150

3.6:1

336

0.136

1:5.5"

0 03 ~+°'°= • --o.m

560 - 1810" 78

645"

0.041

0 11a+o'e~t • ,,-o.em

800 - 2000 71

67

0.018

7.4:1

2284

11.53

1:10

758

2.05

1:10"

225"

0.679

4135

19.27

700 - 1200

9.0 :E 1

7eOe

eNL-USC

1300 - 2000 34

Number of Events [Counts]

11"3-q:'

43 300 - 2100"

90

7eGe [34] PNL-USC-ITEP

9 9+o.?

500 - 1500"

"--0.4

73

7eGe Heid.-Mmc.

14.2 :E 0.3 *ua + 1.3"7't

500 - 1500

73

1.3:1

* Geomet~ceIly e s t i m a t e d ~ o m t h e figures of the corresponding pubficstion.

fFo-A~£y2As a by product of the ~[~-experiment we obtain [35] for the decay mode e- ~ ~,+v, ~/2 > 1.63x1025 y (68% c.L) which is the sharpest laboratory limit existing. In total we hope to have 12 kg of enriched detectors in operation in the Grail Sasso by end of 1994. These would correspond to a non-enriched Ge experiment of at least 1.2 tons. In five years of measurement we could probe the neutrino mass down to ~ 0.1 eV (see Fig. 18).

274

H.V. Klapdor-Kleingrothaus

L~ii~~iI ~

40.

20 0 ¢J . . . . .

I

600

'

~

1000

I

--U

u -

i

i

-u

i

"'U

g

1500

Energy [k,V]

U

|

2~

Fig. 16: Spectrum remaining after subtraction of the 2v spectrum from the solid histogram of Fig. 15, together with a calculated Ovz-spectrum with ~o~ =l.66x10ZZ y (dashed curve). Also shown is the measured spectrum (dotted histogram).

Table 3: Half-life limits for 0vgl~[5decay and the corresponding limits for the neutrino-Majoron coupling constant for several isotopes (from [32]).

Isotope

Experiment

76Ge 76Ge

MPIK-KIAE ITEP UCSB-LBL PN L-USC Cal.PSi-Neu. LBL-MHC-UNM ITEP Cai.-PSI-Neu. UCI INR ITEP Washington Univ.-Tara

76Ge

76Ge

76Gc I°°Mo 136Xe

136Xe 82Se 15°Nd 'SCa ~28T¢=

aGeochemicai experiment.

Ti/2 (10 21 yr) 104(g~r) Rcf. 16.6(90%) 10(68%) 1.4(90%) 6.0 !.0(90%) 0.33(90%) 0.19(68%) 7.2(90%) 1.6(68%) 0.07(68%) 0.72 7700

1.8 2.2 5.8 2.8 6.9 6.2 12.5 2.0 12.5 1.9 5. I 0.3

[441

[461 [451 [47] [48] [49] [50] [51] [52] [53] [541

The Heidelberg-Moscow Experiment

275

10~ imecmJm w i ¢ ~ Ixu$= mumumg tlme 6"~ ~,d

102

11~

~

w , ~ . ~ G, ,,.

~

t00-2700 kmV, 272 count= %,.,

10"1 10"z

~

~ ~i

keY, ~

"~00-27~0

counts "i.

........ ~

.

.

.

500

.

.

'~

.

1000

1500

2000

2500

1000 1500 2000 Energy [keY]

2500

Energy [keV] C 10~ 0 U "102 rl t S 10t 10o

10"11 102 /

. ,, SO0

i .

.

.

.

.

.

Fig. 17: Some of the identified contributions to the observed spectrum from various radioactive impurities according to Monte Carlo calculations (from [30]) 4. APPLICATION OF ~-TECHNOLOGY TO DARK MATTER SEARCH AND 7-LINE ASTROPHYSICS The results of the HEIDELBERG-MOSCOW experiment show for the fast time that the technology of production of enriched HP Ge detectors can be handled. This has promising consequences for dark matter search and galactic V-spectroscopy satellite missions (see [12,26,35-37]. Enriched 76Ge and 73Ge detectors would allow to improve the LEP limits on Dirac- and Majorana WIMPs (weakly interacting massive particles). Use of enriched ~0Ge detectors in the ESA project INTEGRAL (International Gamma Ray Astrophysics Laboratory) would allow to reduce the ~-background dominating in orbit, by an order of magnitude [36]. This has recently been checked in balloon flights [37] (see Figs. 19,20). The INTEGRAL satellite mission aims at the study of a variety of astrophysical questions by high resolution ~/-spectroscopy. The HEIDELBERG-MOSCOW experiment allowed for the first time a search for dark matter with isotopicaUy enriched material. Data taken with a 2.9 kg detector in 166 kgd were used to set limits on spin-

276

H.V. Klapdor-Kleingrothaus

i

10

'

'

' ' ' " 1

,

. . . . . .

,i

,

BACKGROUNDb (keV). kg.y)-I

i

,

, , , , ,

"~

ISOTOPIC ABUNDANCE f A ' ~ \

UCSB-LBL ,.~ 1 t/ 76

,,.,, -

4j.

~

HEIDELBERG -MOSCOW 10-2

........

1

,

........

10 Mt

(kg

I

10 2 .y)

.......

103

Fig. 18: The ranges of neutrino mass which can be probed with Ge detectors of different enrichment in 76Ge, as function of the product detector mass times measuring time (kg y)

Fig. 19: Balloon campaign GRIS in Alice Springs (Australia), May 1992 with one enriched 70Ge detector (see [37]) independently interacting WIMPs [38]. A background level of 0.102 + 0.005 events/kg . d • keV was achieved (average value between 11 and 30 keV), which is better by a factor of 5-7 than in other dedicated dark matter experiments with Ge detectors (Fig. 21). The existing cross section limits for WIMP masses above ~ 150 GeV were improved (Fig. 22) and Dirac neutrinos could be excluded as the dominant component of the dark halo in the mass range 26 GeV to 4.7 TeV.

The Heidelberg-Moscow Experiment

277

1.7

8 1.0

0.3 10 2

10 ~

Energy [keY] 1.7

'

t7

_ T

'

•

'

'

'

I

.

0 r~ 1.0

0

Jttt t | tttt½ beklrmmd to

0.3

Natural

Patio

of

I

lnHehed

Dq~teeter (lll~tGOIq)

t

t

t

,

J,

~

~ , I

Energy [keV]

Fig. 20: Background ratio of enriched (in 70Ge to 95%) to natural detector in the GRIS and HEXAGONE balloon campaigns in Australia, 1992 (see [37])

278

H.V. Klapdor-Kleingrothaus

.4

' ' . . . . . .

'1

. . . . . .

' ' '1

' ' ' ' ' ' '''1

. . . . . . .

"~

>

excluded . recoil spectra (90~. CL) ,,

"~.3 (;0

•

mnmu,~ n ~ l GoV Dtr~c.-nwstrt.~ 4.~ T a r l~tmc-az, utrfn~

o.

'~..2 za0 09 .1

.0 10

15

20

energy

25

30

[keV]

Fig. 21: Low energy spectrum of the second enriched detector corresponding to a measuring time of 165.6 kgd. The background counting rate in the energy range 11-30 keV is 0.102 + 0.005 events/kg d keV. Also shown are the expected recoil spectra of Dirac neutrinos with 26 G-eV (dashed line) and 4.7 TeV (solid line) (from [38]).

10--s)

10-~

10-~

I0--~ 101

10 z

103

104

WIMP mass [GeV] Fig. 22: Exclusion limits for elastic scattering cross sections of WIMPs on 76C~ from the HEIDELBERG-MOSCOW experiment (a) not corrected (b) corrected for loss of coherence, compared with the best result obtained with a natural Ge detector [39]. WIMPs of a given mass with cross sections a beyond the lines are excluded. The expected cross section for Dirac neutrinos with standard weak coupling is shown as a dashed line

(from [38]).

The Heidelberg-Moscow Experiment

279

1000

100

~, 10 1 ¢LU

0.1

0.01

0.001 0

60

120 180 Neutrino Mass (GeV/c2)

240

300

Fig. 23: Exclusion limits for WIMPs with Ge detectors (dashed area) and expected increase of the sensitivity with enriched 73Ge cryo-detectors (curves 99 and 99.9%). Also shown are theoretical expectations for spin-dependently interacting WIMPs from different G U T models (after [55]). Acknowledgement: The author would like to thank Profs. Bellotti, Cabibbo, Maiani and Monacelli for their continuous generous support of the experiment. We also thank the German Ministry of Research and Technology (BMFI") and the Russian Atomic Energy Committee for their substantial support. The author is grateful to Prof. Yu. Zdesenko for fruitful discussions.

REFERENCES: 1. P. Langacknr, in "Neutrinos", Springer Heidelberg, New York, 1988, ed. H.V. Klapdor, p. 71 2. K. Grotz, H.V. Klapdor, "The Weak Interaction in Nuclear-, Particle and Astrophysics", Adam Hilger, Bristol, New York, 1990 3. R.N. Mohapatra, P.B. Pal: Massive Neutrinos in Physics and Astrophysics, World Scientific, Singapore, 1991 4. V. Kuzmin, V. Rubakov, M. Shaposhnikov, Phys. Lett. B185 (1985) 36, M. Fukngita, T. Yanagida, Phys. Rev. D 42 (1990) 1285, G. Gelmini, T. Yanagida, Phys. Lett B294 (1992) 53, B. Campbell et al., Phys. Lett. B 256 (1991) 457 5. J. Suhonen, O. Civitarese, Phys. Left. B 312 0993) 367 6. R.N. Mohapatra, Phys. Rev. D34 (1986) 3457 7. L Steioberger, Phys. Rep. 203 (1991) 345 8. G.B. Gelmioi, M. Roncadelli, Phys. Lett. 99 B (1981) 411 9. Z.B. Berezhiani, A. Yo. Smirnov, J.W.F. Valle, Phys. Left B (1992) 10. S.L. Glashow, Phys. Lett. B256 (1991) 255 l 1. K.S. Babu, R.N. Mohapatra, Phys. Rev. Leu. 67 0991) 1498, D. Choudhory, U. Sarkar, Phys. Rev. Left 68 (1992) 2875 12. H.V. Klapdor-Kleingrothaus, Prec. Int. Symp. on Weak and Electromagnetic Interactions In Nuclei" Dubna, 1992 (WEIN '92), World Scientific, Singapore, p. 201, and Proc. NEWFRINO '92, Nucl. Phys. B (Proc. Suppl.) 31 (1993) 72.

280

H . V . Klapdor-Kieingrothaus

13. H.T. Wong, J. Phys. G. 17 (1991) Suppl. 14. M. K. Moe, preprint UCI-Neutrino 93-1, Interant. 3. Mod. Phys. E, in press 15. H. Ejiri et al., Phys. Lett 258B (1991) 17 16. Yu. Zdesenko, proc. ~eutrinos in Cosmology, Astro, Particle and Nuclear Physics', Erice (1993), Plenum Press 17. K. Grog, H.V. Klapdor, Nucl. Phys. A460 (1986) 395 18. A. Staudt, K. Muto, H. V. Klapdor-Kleingrothaus, Europhys. Lett. 13 (1990) 31 19. X.R. Wu, A. Staudt, H.V. Klapdor-Kleingrothans, C. R. Ching, T. H. Ho, Phys. Lett. 272B (1991) 169 20. M. Hirsch, X.R.Wu, H.V, Klapdor-Kleingrothans, C.R. Ching, T.H. Ho, Phys. Rep. in press (1993) 21. M. Hirsch, X.R.Wu, H.V. Klapdor-Kleingrothaus, C.R. Ching, T.H. Ho, Z. Phys. A 345 (1993) 163 22. K. Muto, E. Bender, H.V. Klapdor-Kleingrothans, Z. Phys. A339 (1991) 435 23. A.L. Turkevich et al. Phys. Rew Lett. 67 (1991)3211 24. M. Hirsch, K. Muto, T. Oda, H.V. Klapdor-Kleingrothans, Z. Phys. A, in press 1993 25. H.V. Klapdor-Kleingrothans, Proc. XIH Int. Conf. on Particles and Nuclei (PANIC) Pemgia (Italy, June 28 - July 2, 1993, in press 26. H.V. Klapdor-Kleingrothaus, Proceed. NATO Advanced Study Institute, Frontier Topics in Nuclear Physics, Predeal, Romania, August 24 - Sept. 4, 1993, Plenum Press (in press) 27. F. Petry et al. Nucl. Instrum. Meth. AA 332, (1993) 107, and F. Petry, these Proceedings 28. J. Busto et al. Nucl. Phys. A513 (1990) 291 29. A. Balysh et al. (HEIDELBERG-MOSCOW collaboration) Phys. Lett. B (1993) in press 30. B. Maier, Dipl. Thesis, Univ. of Heidelberg, 1993, and B. Maier, these Proceedings 31. M. Beck et al. (HEIDELBERG-MOSCOW collaboration), to be published 32. M. Beck et al. (HEIDELBERG-MOSCOW collaboration) Phys. Rev. Lett. 70 (1993) 2853 33. M.K. Moe et al., UCI-Neutrino 92-1, preprint, (1992), and Proc. NEUTRINO '92, Granada, Nuci. Phys. B (ih-oc. Suppl.) 31 (1993) 68 34. F.T. Avignone et al., Phys. Lett 256B (1991) 559 35. A. Balysh et al. (HEIDELBERG-MOSCOW collaboration) Phys. Lett. 298B (1993) 278 36. H.V. Klapdor-Kleingrothans, Proc. Int. Symp. on Gamma Ray Line Astrophysics, Saclay, Dec. 1990, AlP Conf. Proc. 232 (1991) 464 37. N. Oehrels, Nucl. Instrum. Meth. A292 (1990) 505 38. S.D. Barthelmy, L.M. Bartlett, N. Gehrels, M. Leventhal, B.J. Teegarden, J. Tueller, S. Belyaev, V. Lebedev, H. V. KlapdorKleingrothaus, AlP Conf. Prec. 280 (1993) 1166 39. D. Reusser et al., Phys. Lett. 255B (1991) 143 40. S.R. Elliott, A.A. Hahn and M.K. Moe, Phys. Rev. Lett. 59 (1987) 2020 41. H. Ejiri et al., Phys. Lett. B 258 (1991) 17 42. S.I. Vasil'ev, A. A. Klimenko, S. B. Osetrov, A.A. Pomansky and A.A. Smol'nikov, ~ETP Lett. 51 (1990) 622 43. S.R. Elliott, M.K. Mo¢, M.A. Nelson and M.A. Vient, L Phys. G 1"] (1991) $145. zig. A.A.Vasenko, I.V. Kirpicimikov, V.A. Kuanetsov, A.S. Starostin, A.G. Djanyan, V.S. Pogusov, S.P. Shachysisyan and A.G. Tamanyan, Mod. Phys.Lett A5 (1990) 1299 45. H.S. Miley, F.T, Aviguone I]1, R.L. Brodzinski, J.L Collar and J.H. Reeves, Phys. Rev. Lett. 65 (1990) 3092 46. D.O. Caldwell et al. Phys. Rev. Lett 59 (1987) 419 47. P. Fisher et al. Phys. Lett B218 (1989) 257 48. M. Alston-Garnjost et aL Phys. Rev. Lett. 60 (1988) 1928 49. A.S. Barabash et al. Yad. Fiz. 51 (1990) 3 [Soy. L Nucl. Phys. 51 (1991) 1] 50. J.-L. Vuilleumier (private communication) 51. S.R. Elliot, A.A. Hahn and M.K. Moe, Phys. Rev. Lett 59 (1987) 1649 52. A.A. Klimenko, A.A. Pomansky and A.A. Smoinikov, in Proceed. of the 11th Internat. Conf on Neutrino Physics and Astrophysics (World Scientific, Singapore, 1984) 53. A.S. Barabash, Phys. Lett B216 (1989) 257 54. T. Bernatowicz et al., Phys. Rev. Lett 69 (1992) 2341 55. B. Sadoulet, Prec. TAUP 1993, Assergi, Italy, in press