Nuclear Physics A484 (1988) 635-646

North-Holland, Amsterdam
O+ + 2+ NEUTRINOLESS pp DECAY OF 76Ge*
T. TOMODA
Institut fiir Theoretische Physik, Universitiit Tiibingen D-7400 Tiibingen, Fed. Rep. Germany
and
Swiss Institute for Nuclear Research CH-5234 Villigen, Switzerland**
Received 6 January 1988
(Revised 25 March 1988)
Abstract: The neutrinoless pp decay rate of the process 76Ge(0:) + %e(2:) has been calculated taking
into account both the two-nucleon and A-isobar mechanisms. Since there occurs a cancellation in
the contribution of the two-nucleon mechanism, the A mechanism becomes important. Using the
calculated nuclear matrix elements, the upper limits on the effective coupling constants of right-
banded leptonic currents l(A)1 C 2 x 10e3 and l(v)1 < 2 x 10m4 are deduced from recent experimental
data.
1. Introduction
The neutrinoless double beta (Oz@) decay has been attracting considerable
attention in recent years [for reviews see refs. I-)]. Such a decay takes place if the
neutrino is a Majorana particle (i.e. identical to its own antiparticle) under the
conditions (i) that the mass of the neutrino is non-vanishing and/or (ii) that there
are right-handed leptonic currents coupled to nuclear currents. While a O++ O+ Ovpp
decay can occur under either (or both) of the conditions (i) and (ii), a Of+ 2+ Oz@
decay can be caused only by right-handed leptonic currents. Observation of Ov/3/3
decay to 2+ states would thus yield information different from what we obtain by
studying O+ + O+ Ovpp decays.
Several experimental groups 5-1o) h ave tried to find evidence for a O++ 2+ Ov/3/3
decay of 76Ge and reported lower limits on the half-life of the decay. A 0+-+2+
Ov&3 decay can proceed through an exchange of a Majorana neutrino either between
two nucleons (2N mechanism) or between two quarks in the same baryon [A
mechanism ), see fig. 11. In the latter case the decay amplitude is on the one hand
suppressed because of the large mass difference between the nucleon and the A
isobar, but on the other hand enhanced because of the shorter distance which the
virtual neutrino has to travel. Therefore the A mechanism can be as important as
the 2N mechanism in contrast with the case of O++ O+ Ov@ decay, in which the A
mechanism is forbidden ) by the angular momentum selection rule AJ = 1 or 2.
l Supported in part by the Bundesministerium fiir Forschung und Technologie (Fed. Rep. Germany)
under contract number 06 Tii 778/9.
l * Present address.
0375-9474/88/$03.50 @ Elsevier Science Publishers B.V.
(North-Holland Physics Publishing Division)
636 T. Tomoda / Ou/3@ decay
P
e e uud
\
,.
1/
ddd ddu
w c,-v--J
n-
n
Fig. 1. The Ov/3p transitions between nucleon and A isobar.
Calculations for the O++ 2+ decays of 48Ca [refs. ,)I and 76Ge [ref. )I by the 2N
mechanism have been reported previously. In these calculations the initial and final
nuclear states were treated in the shell model, and a weak-coupling approximation
was employed for the case of the 76Ge decay ). As for the A mechanism, there have
been only rough estimates by Primakoff and Rosen l), and by Doi et al. ).
In the present paper we calculate the matrix elements of the Oz@ decay 76Ge(0:) +
76Se(2:) for both the 2N and A mechanisms. In sect. 2 we first give the two-body
transition operators for the 2N mechanism in the closure approximation. Considering
the excitation or deexcitation of A by one-pion or one-rho exchange, we derive
effective two-body operators for Ovpp decay by the A mechanism which act on
nucleon coordinates.
Using a realistic effective NN interaction, we determine in sect. 3 the Hartree-
Fock-Bogoliubov type wave functions for the initial and final nuclear states by
variation after particle-number and angular-momentum projection, and then calcu-
late the O@ transition matrix elements. Comparing the results of the calculation
with the experimental data, we deduce limits on the coupling constants of the
right-handed leptonic currents.
A summary is given in sect. 4.
2. Formalism
2.1. 0++2+ Ov@ DECAY BY THE 2N MECHANISM
We assume the following effective weak interaction hamiltonian density 253*13)
H,(x) = JiG cos &(jLJct + ~j,,Jt+ + Aj,,J,r) + h.c. ,
(1)
where G = 1.16637 x lop5 GeV-*, cos tic = 0.9729 [ref. 14)], and
jE(x) = t(x)Y(l- YS)V&) ,
jK(x> = ~(x)Y(l+ -J~~,R(X), (2)
T. Tomoda / Ovpp decay 637
are the left- and right-handed leptonic currents with the current neutrinos defined by
veL(x) = Z uejNjL,(x) 9
j=l
v&(x) = f VejNjR(X).
(3)
j=l
Here Nj is the eigenstate of the Majorana neutrino mass matrix with the eigenvalue
mj, and f the number of generations. The left- and right-handed nuclear currents
are assumed to be
(4)
where (P[T]~) = 1, gv= 1, gA = 1.254 [ref. )I. Since the recoil terms [C, and D, of
ref. )I in the nuclear currents are not important for Of+ 2+ decays, they are
neglected in the present work. The natural units h = c = 1 are used throughout this
paper.
In the closure approximation the half-life for a O++ 2+ Or+?/3 decay, r$(O++ 2),
can be written in the form 2*3)
r7~/y2(o++2+)1-1= F*(M,(h)-M,(77))2+F*(Mb(77))2, (9
with
(rl)=77 C ucjKj9 (6)
and
MA = i C,+i Mi )
i =l
(74
M, = ~ C,iMi,
i=l
Gb)
M:,= i C&Mi,
(7c)
i=6
where
are nuclear transition matrix elements. The primed summation in eqs. (6) should
be taken over light neutrinos (mj < 100 MeV). We neglect in the present paper
possible contributions from heavy neutrinos, for which a factorization as in eq. (5)
638 T. Tomoda / Ov$p decay
does not hol d. The two-body operators Mi;,, and the coefficients Cni, C,+ and C$
for the 2N mechanism (i = 1,2,. . . ,7) are listed in table 1. In this table and the
following text, r,,, = r, - r,,,, r+,, = r,, + r,,,, a^=a/la/ for any vector a, and
where H(r, A) is given by eq. (B.l) of ref. 13), and
&(E,)-E(O+)+m,+~Q,,(2+),
(10)
with E(0) being the energy of the initial O+ state, (EN) the average energy of the
intermediate excited states, and QPp(2+) the Q-value of the O+ + 2 j3j3 decay.
Fi( i = 1,2) in eq. (5) are the leptonic phase-space integrals
(gAG OS &I4 m2pIp2E,E S(E
32~~ e
2 1
+E
2-2m,- Qpp(2tll dG G. (11)
Here Ej and pj =e are the energies and the asymptotic momenta of the
electrons, and
3
(m,R)2
=~~If-I+Ifi~l+If-I-tl~21*1,
3
fi z(m,R)2
___ Elf-2*12+ lf-212+ lf-2*12+ If-*I21 9 (12)
where R is the nuclear radius and f^ KI etc. the products of electron radial wave
functions defined by eq. (A.4) of ref. r3).
TABLE 1
The two-body operators M,,,, and the coefficients CAi, C,,! and C& for 0*+2+
Ov&3 decay by the 2N mechanism
i
Miw,,, c:;
6 [(un -u,)o[~,,oF+,,]]2T+nm h
&I g.4)
mm
I [(~,-~,)~[i,,9i+,,](~)](~)~~~ -J k!/g,)
The coefficients which are not given here explicitly are zero: CA; = C,, = 0 for i = 6,7, and C$ = 0 far i G 5.
T. Tomoda / Ovpp decay
2.2. 0++2+ OvpP DECAY BY THE A MECHANISM
639
Now let us consider the p/3 decay at the quark level. The operators rir:Minm
with Mi,, of table 1, which are now interpreted as acting on the quark degrees of
freedom, induce (virtual) Oz$3/3 transitions such as A-+p and n + A++ (see fig. 1).
Assuming the non-relativistic constituent quark model 15), where the radial wave
functions of nucleons and A isobars are products of OS states, only the operator of
the type M,,, of table 1
2
A* =; c
Is&j=3
(-3) E 7Ti+[Ui@Uj]%,
( >
(13)
has a non-vanishing contribution. Here we included in the definition of .II%* the
coefficient C,, = C,,, = -3 and the factor (g,/g,)*, taking into account the fact that
the axial vector coupling constant for quarks is one. The reduced matrix elements
of JII, between a nucleon and a A isobar can be easily calculated as*
Here (h) is the radial matrix element of h between the relative OS states
(h) = (OS1 - r; H(r, Sm)lOs> )
(15)
and Sm = md - mN = 293 MeV neglecting the small energy of the electron in the
second argument of H [see eq. (B.l) of ref. )I.
The processes of fig. 1 cause nuclear Ovpp decay via virtual excitation of A isobar
as is shown in fig. 2. The corresponding effective two-body operator acting on the
nth and mth nucleons can be written as
1 1
r;r+,MAnm =-2Jt4~-- VANNN-2( VANNN)t -.&A,
6m 6m
(16)
ee
\
n-
P
I
n>P
---_-_
P
n,P
.-------
n
++
n n n n
Fig. 2. The 0#3 decay processes via virtual excitation of A isobar.
l Doi et al. 3, gave 6 instead of -2&6, where of course the sign depends on the choice of the relative
phase of N and A. The relative phase adopted in the present work is the same as that of ref. ).
640 T. Tomoda / Ov@3 decay
where the factors 2 take into account the fact that there are two neutrons to be
excited into A.
We use the one-boson exchange potential
V
ANNN= c v,dNNN,
n=lr,p
where*
(17)
ANNN = 1 fNAol fNNa
V,
3 4?r
T, * ~m{a% * ~*vao(r)
+ c,*[3(Sn * warn. 3-f%* u,lVd(r)I. (18)
Here r = r,,, and S(T) is the transition spin (isospin) 18) and
V,,(r) = i exp (-mar) ,
V,,(r)=[1+3(m,r)-+3(m,r)-2]V,o(r),
(Cd,
I
(1, 1)
ccd = (2, -1)
(a = T) 9
(a=p).
(19)
In order to write MA,,, in terms of nucleon variables only, it is convenient to
introduce a transition spin (isospin) of rank 2, Y(Y), which is defined by
(Yfi) Mm = #4tm2b4, (20)
and by an analogous equation for 9, just as the transition spin (of rank 1) is defined
by 18)
(SW)
Mm = (;Ml;ml/L).
(21)
In terms of JY and 5, AA, which satisfies eq. (14), can be written as
.&, = -;a E *(h)[sP=.iF+ (st)P=2ypt] .
( >
(22)
Using this and the relations
Y$, * a2=J~[a,&r2](2),
$[3(S, * ;)(a2 * ;) -s, * UJ =&J, * a,[~~,@ G12y2
-~~[[u~OU~]~~~~[~~~OL;~]~~~]~~~, (23)
we obtain the following expression for MA,,, of eq. (16):
M
Afl l Tl
X [c,0&r,ou,](2)voi0
+~~*&b m
. u [~~,OFn*]2-J ~[[unOum]20[~~nm~E,]2]2)Va2].
(24)
* fNNP in the present work corresponds to 6fN, = vGg,,,(m,/2m,)(l +fNNp/gNNp) of ref. 17).
T. Tomoda / Ovpp decay 641
The total decay amplitudes are obtained by adding the matrix element MA of the
operator MA,,,,, to MA and M,,, where the relation between MA and MA,,, is defined
analogously to eq. (8). We include the contribution of MO in eqs. (7a, b) by extending
the summation over i up to i = 13, where
c,i=c,i=6Js
9
*E~NA;~N~~~, _=( (i=8,%10), (25)
P
(i = 11,12,13).
The operators Mi,, and the coefficients Ci for 8 1 I -Z C 13 are listed in table 2.
2.3. CALCULATION OF THE NUCLEAR MATRIX ELEMENTS
We describe the initial O+ and final 2+ nuclear states in terms of the Hartree-Fock-
Bogoliubov type wave functions which are obtained by variation after particle-
number and angular-momentum projection 19*13). The ground-state correlations due
to proton-neutron two-quasiparticle modes *O,*) are absent in this type of wave
functions. Since it has been shown *I) that their effect does not reduce the O++ Of
Oz$p nuclear matrix elements drastically, we expect the same also in the case of
O++ 2+ Ov@ decays. The calculation of the transition matrix elements proceeds in
a similar way to the case of O++ O+ decay described in ref. 13) with straightforward
modifications concerning angular momentum coupling. Also the effect of NN
short-range repulsive correlations are taken into account in the same way as in ref. 13).
TABLE 2
The two-body operators k&,,,,, and the coefficients c, for O++ 2+ Ovpp
decay by the A mechanism
i M
,nm c,
8 [uOa,l(2)V,,
9 u, . um[imOinml(2)vir2
10 C[a,,0a,l~*~Orin~Oi~l~2~l~2~V**
11
l~.@~,PV,,
-$+
:_
12 un . a,[i,, 0 ~,,,12vP*
13 [[unO~m](Z)O[inmOi~m](2)](2)Vp*
3. Results
We take the full 3hw and 4ho harmonic oscillator shells for the description of
76Ge and %e. As a realistic effective NN interaction we use the nuclear matter
G-matrix calculated from the Bonn one-boson-exchange potential ). In order to
apply it to finite nuclei, the Bethe-Goldstone equation is solved with a relatively
small (in absolute value) starting energy of -25 MeV [refs. *,*r)]. The J = 0 com-
ponents of proton-proton and neutron-neutron interactions are made a little stronger
642 T. Tomoda / Ov@ decay
by factors 1.25 and 1.10, respectively, so as to reproduce the experimental pairing
gaps. All other interaction matrix elements are modified by a single factor 1.13 so
that the excitation energy of the 2: state of %e is reproduced. The single-particle
energies of the nuclear hamiltonian are determined in such a way that the calculated
Hat-tree-Fock single-particle energies [see Ei defined by eq. (23) of ref. )I coincide
with the Woods-Saxon single-particle energies.
In order to calculate the nuclear matrix elements A4i for the 2N mechanism, we
first determine the value of A, eq. (lo), on which they depend weakly. Using
(EN)- E(O+) =7.88 MeV [refs. 2,3)], and C&(2+) = C&(0+) -0.5591 MeV=
1.4816 MeV, where C&(0+) = 2.0407 MeV has been taken from ref. 23), we obtain
A = 9.13 MeV. This value of A is used in the present work. The calculated nuclear
matrix elements Mi as well as the products CAiMi, C,iMi and CL,Mi for the 2N
mechanism (1 G i s 7) are given in table 3. We notice first of all that these matrix
elements are much smaller than the typical matrix element MgG [ =0.3 - 0.5 fm-,
ref. 13,21)] for the 0 + O+ Or@ decay. Furthermore, there occurs a strong cancellation
among the contributions of M,, M2 and M4 to Mh (Ci =1,2,4 C*i Mi = 0.15 x 10m3 fm-)
and also between those of M6 and M7 to M&. Consequently Mh and Mh become
much smaller than M,,.
TABLE 3
The calculated nuclear matrix elements M, and the products C,,M,, C&f, and C&M, for the O++ 2+
Oz@ decay of 76Ge by the 2N mechanism
i
Ml CA&f,
i
M,
C&M,
[10e3 fm-1 [10e3 fm-1 [ 10m3 fm-1 [ 10e3 fm-1
1 29.73 9.91 9.91 6 14.60 8.23
2 5.40 -3.60 -3.60 I 8.24 -8.05
3 -0.31 -0.48 -0.48
4 -9.69 -6.16 6.16 sum(6-7) 0.18
5 -0.79 0.77 0.00
sum (l-5) 0.44 12.00
For the calculation of the matrix elements of the A mechanism, we still have to
determine the coupling constants fNNa, fNaa and the expectation value (h) in eq.
(25). We use )
f2=O078 f&%?=456
NNx
4T * 47r .
and the quark model value 16) for the ratio
f NAa fNAp _ T2
-=-_
f ~NNP NNr
J- 25'
which is not far from the experimental value of the ratio for pion. We calculate (h)
of eq. (15) using a harmonic oscillator OS wave function with the oscillator constant
T. Tomoda / Ovpp decay 643
b = 0.8 fm, and obtain
(h) = 0.546 fm- .
(26)
This should be compared with the limit for Sm + 0:
lim (h)=(Os~~~Os)=~~=0.997fm-1.
S??I-rO
(27)
This value is reduced to the above one due to the rather large mass difference of
the nucleon and the A isobar which is comparable to the typical energy of the
neutrino exchanged between quarks, hc/ b = 250 MeV. Table 4 gives the calculated
matrix elements Mi and the products C,tiMi (= C,iMi) for the A mechanism (8 s i s
13). It is noted that Mg (Ml,) which involves the tensor part of the nucleon-A
interaction gives the dominant contribution in the A mechanism due to one-pion
(one-rho) exchange. The total contribution to the amplitude n/r, or M, from
p-exchange is about 15% of that from n-exchange, and has an opposite sign because
of the dominance of the tensor interaction.
Table 5 summarizes the calculated nuclear matrix elements. It is noted that while
the A mechanism gives a minor correction to M,, it gives the dominant contribution
TABLE 4
The calculated nuclear matrix elements M, and the products C,,M, (= C,,,M,) for the O+ + 2+ Oupp
decay of 76Ge by the A mechanism
i
[lo-Zrn-r]
GM,
[lo- fm-1
i
[lO-XZ,
[lO~?$
8 0.40 0.03 11 0.00092 0.008
9 19.88 1.55 12 0.05254 -0.239
10 -0.13 0.02 13 -0.00030 -0.003
sum (ff) 1.60
sum (P)
-0.234
sum (7r+p) 1.37
Those with 8 s is 10 (115 is 13) represent the contributions from excitation of A by one-pion
(one-rho) exchange.
TABLE 5
The calculated nuclear matrix elements for the O+-t2+ Ovpp decay of
76Ge by the 2N and A mechanisms
[lo-Em] [10-%-1,
M:
[ 10e3 fm-1
2N 0.44 12.00 0.18
A 1.37 1.37 0.00
sum 1.81 13.37 0.18
644 T. Tomoda / Ovpp decay
to MA since the pieces of Mh due to the 2N mechanism cancel accidentally. The
matrix elements for the 2N mechanism have been calculated previously by Haxton
and Stephenson ) using weak-coupling shell-model wave functions. They obtained*
M, = 8.27 x low3 fm- and Mb = 2.71 x lo-3 fm- in the present notation. Although
their values for Mk and for the individual contributions to M, are considerably
different from the corresponding results of the present calculation, there is no large
difference in the value of M,, which dominantly determines the rate of Ov/?p decay
caused by nonvanishing (7). As for the matrix element for the A mechanism, the
present calculated value, 1.37 x 10m3 fm-, is about &, of the rough estimate 86x
low3 fm- by Doi et al. ).
The leptonic phase space integrals F, and F2 of eq. (11) are calculated by using
the solutions of the Dirac equation with the nuclear finite size taken into account
[see the method WFl of ref. )I. The calculated values of Fl and F, with
&(2+) = 1.4816 MeV are
F, = 4.689 x lo-l3 y- fm* ,
F2 = 3.260 x lo-l3 y- fm* .
(28)
Substituting these and the nuclear matrix elements of table 5 into eq. (5), and
comparing with the experimental lower limits -O) on $T2(O+ + 2+), we obtain the
upper limits on l(A)1 and l(v)1 listed in table 6. In this analysis we put** (7) = 0
((A) =0) when we calculate the limits on (A) ((7)). The best limits on (h) and (7)
in table 6 are still three and four orders of magnitude, respectively, less stringent
than those deduced from O++= Oc Ov@3 decays in ref. *l):
l(A)] < 4 x lop6 from 7$(0+ + 0) > 5 x 1O23 y for 76Ge [ref. )I ,
and
l(q)1 < 2 x lo- from $2(0f+ 0) > 5 x 1O24 y for *Te [ref. )I
TABLE 6
The limits on (A) and (7) obtained by comparing the present calculation with experimental
lower limits on the half-life, Ty/,(exp), for the O++ 2+ Ovpp decay of %e
l0)l I(s)I
Group Ref.
Cl.5 x 1o-3 t2.0 x 1o-4
12.9 x 1om3 (3.9 x 1oF
13.3 x lo- <4.5x 1o-4
<3.4x 1o-3 ~4.6 x 1O-4
<4.9x 1o-3 (6.6 x 1o-4
(6.4 x 1O-3 ~8.6 x 10m4
>3 x lo=
>8x lo=
>6x lo*
>S.l x 1022
>2.7 x lo=
>1.6x lo*
UCSB-LBL
PNL-USC
Zar.-Bard.-Stras.
Osaka
Milan0
Guelph-Apt.-Que.
l These numbers are known only through the quotation in ref. ). As the authors of ref. *) considered
only the case where 7 # 0 and A = 0, no value for M, was calculated.
* If we do not impose such a condition we get much less stringent limits (l(A)1 i 1.3 x 10-l and
j(T7)1< 1.8 x 1o-2 for r:;; > 3 x 10z3 y) because the allowed region on the (h)-(v) plane for a given lower
limit on T$~ is highly deformed along the line M,(A) - M,(q) =0.
T. Tomoda / Ovpp decay 44.5
Conversely, l(..+)l < 4 X lo+ ([(q)[ < 2 X lo-) gives r$(O++ 2) > 4.1 x lo** y
(?:/,(0+-+2+)>3.ox 103iy).
4. Summary
The Or@@ decay rate of the process 76Ge(0:) + Se(2:) has been calculated taking
into account both the 2N and A mechanisms. The decay can be caused by the
right-handed leptonic current coupled (i) to the right-handed nuclear current with
the coupling constant (h) and/or (ii) to the left-handed nuclear current with the
coupling constant (n). In the former case, the pieces of the relevant nuclear matrix
element Mh due to the 2N mechanism have been shown to cancel considerably,
and consequently the contribution from the A mechanism becomes important. Its
magnitude calculated in the present work is, however, much smaller than the rough
estimate in ref. ). In the case of the coupling proportional to (r]), the contribution
of the 2N mechanism is dominant and that of the A mechanism can be neglected.
The upper limits on (A) and (n) deduced in the present work from the best
experimental bound ) r$:,(O++ 2*) > 3 x 1O23 y are
j(h)/ < 2 x 1o-3 and /(7)>/<2x 10-4.
These limits are much less stringent than those from the O++ 0 Ov&3 decay data.
Finally, it is hoped that the present result should not discourage the experimen-
talists who are making intensive efforts to find 0+2+ Ov&3 decay events.
The author is grateful to Prof. A. Faessler and Prof. M.P. Locher for their kind
hospitality during his stay at the University of Tiibingen, where the present work
was started, and at the Swiss Institute for Nuclear Research, where it was continued
and completed. Thanks are also due to Dr. J. Missimer and Prof. M.P. Lecher for
their critical reading of the manuscript and useful comments.
References
1) H. Primakoff and S.P. Rosen, Ann. Rev. Nucl. Part. Sci. 31 (1981) 145
2) W.C. Haxton and G.J. Stephenson, Jr., Prog. Part. Nucl. Phys. 12 (1984) 409
3) M. Doi, T. Kotani and E. Takasugi, Prog. Tbeor. Phys. Suppl. 83 (1985) I
4) J.D. Vergados, Phys. Reports 133 (1986) 1
5) D.O. Caldwell, R.M. Eisberg, D.M. Crumm, MS. Witherell, F.S. Goulding and AR. Smith, Phys.
Rev. Lett. 59 (1987) 419;
D.O. Caldwell, private communication
6) F.T. Avignone III, R.L. Brodzinski, J.C. Evans, Jr., W.K. Hensley, H.S. Miley and J.H. Reeves,
Phys. Rev. C34 (1986) 666
7) A. Morales, J. Morales, R. Nufiez-Lagos, J. Puimed&, J.A. Villar, D. Dassie, Ph. Hubert, F. Leccia,
P. Mennrath. M.M. Villard, J. Chevallier and B. Haas, Proc. Int. Symp. on Weak and Electromagnetic
Interactions in Nuclei, Heidelberg, July 1986, ed. H.V. Klapdor (Springer, Berlin, 1986) p. 696;
P. Mennrath, private communication
8) H. Ejiri, N. Kamikubota, Y. Nagai, T. Nakamura, K. Okada, T. Shibata, T. Shima, N. Takahashi
and 7. Watanabe, J. of Phys. G13 (1987) 839
646 T. Tomoda / Ovflp decay
9) E. Bellotti, 0. Cremonesi, E. Fiorini, C. Liguori, A. Pullia, P.P. Sverzellati and L. Zanotti, Nuovo
Cim. 95A (1986) 1
10) J.J. Simpson, P. Jagam, J.L. Campbell, H.L. Malm and B.C. Robertson, Phys. Rev. Lett. 53 (1984) 141
11) H. Primakoff and S.P. Rosen, Phys. Rev. 184 (1969) 1925
12) J.D. Vergados, Nucl. Phys. B234 (1984) 213
13) T. Tomoda, A. Faessler, K.W. Schmid and F. Griimmer, Nucl. Phys. A452 (1986) 591
14) Particle Data Group, Phys. Lett. B170 (1986) 1
15) F.E. Close, An introduction to quarks and partons (Academic Press, London, 1979)
16) G.E. Brown and W. Weise, Phys. Reports 22 (1975) 279
17) K. Holinde, Phys. Reports 68 (1981) 121
18) H. Sugawara and F. von Hippel, Phys. Rev. 172 (1968) 1764
19) K.W. Schmid, F. Griimmer and A. Faessler, Nucl. Phys. A431 (1984) 205
20) 0. Civitarese, A. Faessler and T. Tomoda, Phys. Lett. B194 (1987) 11
21) T. Tomoda and A. Faessler, Phys. Lett. B199 (1987) 475
22) E. Hammaren, K.W. Schmid, F. Griimmer, A. Faessler and B. Fladt, Nucl. Phys. A437 (1985) 1
23) R.J. Ellis, B.J. Hall, G.R. Dyck, C.A. Lander, K.S. Sharma, R.C. Barber and H.E. Duckworth, Phys.
Lett. B136 (1984) 146
24) T. Kirsten, E. Heusser, D. Kaether, J. Oehm, E. Pernicka and H. Richter, Proc. Int. Symp. on nuclear
beta decays and neutrino, Osaka, Japan, June 1986, ed. T. Kotani, H. Ejiri and E. Takasugi (World
Scientific, Singapore, 1986) p. 81