

 3 Oct 1994

DFTT 32/94 A Model Independent Analysis of Solar Neutrino Data

S.M. Bilenky\Lambda Joint Institute of Nuclear Research, Dubna, Russia INFN Torino, Via P. Giuria 1, 10125 Torino, Italy Dipartimento di Fisica Teorica, Universit`a di Torino

C. Giuntiy INFN Torino, Via P. Giuria 1, 10125 Torino, Italy Dipartimento di Fisica Teorica, Universit`a di Torino

(July 1994)

Abstract The results of a solar model independent analysis of the existing solar neutrino data under the assumption of MSW transitions in matter or vacuum oscillations (in the cases of *e-*_(*o/ ) and *e-*S mixing) are presented. The analysis was done in two cases. In the first case no assumptions on the value of the total fluxes of neutrinos from all reactions has been made. In the second case some wide boundaries for the values of the neutrino fluxes, that take into account the predictions of different standard solar models, are imposed. It is shown that rather large regions of the parameters \Delta m2 and sin2 2` are excluded by the existing data.

PACS numbers: 96.60.Kx, 14.60.Pq, 14.60.Lm.

Typeset using REVTEX \Lambda E-mail address: BILENKY@TO.INFN.IT. yE-mail address: GIUNTI@TO.INFN.IT.

1

I. INTRODUCTION Solar neutrino experiments are very important for the investigation of neutrino masses and mixing. Due to the large distance between the source and the detector and the small neutrino energies, solar neutrino experiments are sensitive to very small values of the difference of squared neutrino masses \Delta m2 = m22 \Gamma m21 and (if the MSW mechanism is effective) to a wide region of mixing angles `, including the theoretically important region of small `'s.

At present there exist data of four solar neutrino experiments: the radiochemical experiments Homestake [1], Gallex [2] and Sage [3] and the direct counting experiment Kamiokande [4]. The latest experimental data are presented in Table I. The rates predicted by Bahcall and Pinsonneault (BP) [5], Turck-Chi`eze and Lopes (TL) [6] and Castellani, Degl'Innocenti and Fiorentini (CDF) [7] are given in the last columns of Table I. The rates predicted by these and others [8,9] Standard Solar Models (SSM's) are in agreement with each other and exceed the rates observed in all four solar neutrino experiments. One can see that the event rates observed in the Homestake and Kamiokande experiments, in which high energy neutrinos are detected, are less than, for example, the BP predicted rates by 5.5 and 3.1 standard deviations, respectively. The counting rate in the GALLEX experiment in which the small energy pp neutrinos give an important contribution (about 55% according to the SSM) is about 4 standard deviations less than the predicted rate.

It was shown in Refs. [10-16] that all the existing data can be described in the framework of the SSM by the MSW mechanism [17] in the simplest case of mixing of two neutrino types. Two solutions for \Delta m2 and sin2 2` were found: a small mixing angle solution with \Delta m2 ' 5 \Theta 10\Gamma 6 eV2 and sin2 2` ' 8 \Theta 10\Gamma 3 and a large mixing angle solution with \Delta m2 ' 10\Gamma 5 eV2 and sin2 2` ' 0:8. Let us notice that the existing data can also be described by vacuum oscillations [18] if the mixing angle is large and there is a fine tuning between \Delta m2 and the sun-earth distance [13].

Thus, solar neutrino data give some evidence in favour of non-zero neutrino masses and mixing if we assume that the SSM correctly predicts the fluxes of neutrinos from all neutrino producing reactions occurring in the core of the sun. However, as it is well known, the neutrino fluxes predicted by the SSM are subject to many sources of uncertainties, mainly due to a poor knowledge of some input parameters (especially nuclear cross sections and solar opacities). It is clear that any solar model independent information on the neutrino mixing parameters that can be obtained from an analysis of the data of solar neutrino experiments is very important. In this note we present the results of such a model independent analysis.

Our analysis is based on the assumption that the spectra of neutrinos from different solar reactions are known. These spectra are determined by the interactions responsible for the reactions and, as it was shown in Ref. [19], are negligibly affected by the conditions in the interior of the sun. We consider the (constant) values of the total fluxes (from pp, pep, 7Be, 8B, Hep, 13N, 15O, 17F) as unknown parameters.

We will consider resonant MSW transitions in matter and vacuum neutrino oscillations in the case of mixing between two active neutrino states (*e-*_ or *e-*o/ ) and mixing between *e and a sterile neutrino state *S. In both cases the neutrino transition probabilities depend on two parameters: \Delta m2 and sin2 2`. The event rate in any solar neutrino experiment is determined by the total neutrino fluxes and the values of these parameters. We use the total

2

event rates of all four solar neutrino experiments. At fixed values of \Delta m2 and sin2 2` we calculate the O/2 and the corresponding confidence level for all possible values of the total neutrino fluxes. If all the confidence levels are unacceptable, the corresponding point in the \Delta m2-sin2 2` plane is excluded. This analysis allows us to exclude large regions of the parameter space for transitions of *e's into active as well as into sterile states. Let us notice that some other regions in the \Delta m2-sin2 2` plane were excluded in a model independent way by the Kamiokande collaboration [20] by using the data on the recoil electron spectrum in *-e scattering and the data on the search for the day-night effect.

II. THE METHOD OF ANALYSIS The integral event rate in any experiment a (a = HOM (Homestake), KAM (Kamiokande), GAL (GALLEX+SAGE)1) is given by the expression

Na = X

r Y

r a \Phi

r *e ; (2.1)

where the index r runs over all the neutrino sources which are listed in Table II. Here \Phi r*e is the total initial flux of *e from the source r. The initial spectrum of *e's from the source r has the form

OEr*e(E) = Xr*e(E) \Phi r*e ; (2.2)

where the functions Xr*e (E) are normalized by the condition Z Xr*e (E) dE = 1 and are known (see Ref. [5]). In the case of radiochemical experiments only solar *e on the earth are detected. We have

Y ra = Z

Eath oea(E) X

r *e (E) P*e!*e (E) dE ; (2.3)

with a = HOM; GAL. Here oea(E) is the neutrino cross section, P*e!*e (E) is the probability of *e to survive and Eath is the threshold energy. In the case of MSW transitions the *e survival probability is calculated using the formula given in Ref. [21], which is valid for an exponentially decreasing electron density, and is averaged over the production region. In the case of vacuum oscillations the *e survival probability is averaged over the production region and over the varying distance between the sun and the earth (see Ref. [22]).

In the case of the Kamiokande experiment *e as well as *_ (and/or *o/ ) are detected. We have

Y rKAM = Z

EESth oe*

ee(E) X

r *e (E) P*e!*e(E) dE

+ Z

EESth oe*

_e(E) X

r *e (E) P*e!*_ (E) dE ; (2.4)

1In our calculations we use the combined GALLEX-SAGE data: 78 \Sigma 10 SNU.

3

where oe*`e(E) is the cross section of the process *` e ! *` e (` = e; _), EESth =

1 2 iT

th e + qT the (T the + 2me)j (T

th e is the recoil electron kinetic energy threshold) and

P*e!*` (E) is the probability of transition of *e into *`. In our calculation we took into

account the efficiency and the energy resolution of the Kamiokande detector. These quantities were taken from Ref. [4].

In the analysis of solar neutrino data we take into account the luminosity constraint on the total neutrino fluxes. The thermonuclear energy of the pp and CNO cycles in the sun is generated in the transition

4 p + 2 e\Gamma ! 4He + 2 *e : (2.5) If the sun is in a quasi-static state, from Eq.(2.5) it follows that the luminosity of the sun Lfi is connected with the total neutrino fluxes by the well known relation

Lfi 4ss d2 = Xr \Phi

r *e ^ Q2 \Gamma hEi

r* : (2.6)

Here d = 1 AU = 1:496 \Theta 1013 cm is the average sun-earth distance, Lfi = (3:826 \Sigma 0:008) \Theta 1033 erg sec\Gamma 1 [23], Q = 4 mp + 2 me \Gamma m4He = 26:73 MeV is the total energy release in the transition (2.5) and hEir is the average energy of neutrinos from the source r. The luminosity constraint can be written down in the form of Eq.(2.1). We have

NLUM = X

r Y

r LUM \Phi

r *e ; (2.7)

with NLUM = Lfi=4ss d2 = (8:491 \Sigma 0:018) \Theta 1011 MeV cm\Gamma 2 sec\Gamma 1 and Y rLUM = Q=2 \Gamma hEir. The values of hEir and Y rLUM are given in Table II.

Our procedure for the analysis of the solar neutrino data is the following. For fixed values of the parameters \Delta m2 and sin2 2` and fixed values of the neutrino fluxes we calculate the O/2, defined as

O/2 = X

a

(N expa \Gamma Na)2

(\Delta N expa )2 + ` \Delta oeaoe

a N

a'

2 : (2.8)

The relative uncertainties \Delta oea=oea of the neutrino cross sections for the Homestake and Gallium experiments are 0.025 and 0.041, respectively (see Ref. [5]). We estimate the "goodnessof-fit" by calculating the confidence level (CL). Since we do not determine any parameter, the number of degrees of freedom of the O/2 distribution is equal to the number of data points (i.e. four: three neutrino rates and the solar luminosity). If all the confidence levels found for a given value of \Delta m2, sin2 2` and all possible values of the neutrino fluxes2 are

2It is possible to perform this calculation by trying all possible values of the neutrino fluxes. However, in practice it is computationally much more convenient to find the minimum value of the O/2 with respect to the neutrino fluxes. We find this minimum with the Minuit program, which is available in the CERN program library. Let us emphasize that the minimization of the O/2 is only a computational trick which allows us to calculate the maximum value of the CL without having to try all the possible values of the neutrino fluxes with a very lengthy montecarlo.

4

smaller than ff (we choose ff = 0.1, 0.05, 0.01), it means that the corresponding point in the \Delta m2-sin2 2` plane is excluded at 100(1 \Gamma ff)% CL. In this way we obtain the exclusion plots presented in Figs.1-8. Let us notice that for the purpose of determination of the excluded regions in the parameter space our approach is the most conservative: any decrease of the number of degrees of freedom would increase the excluded regions.

For the exclusion plots presented in Figs.3, 4, 7, 8 the only requirement was that all the total neutrino fluxes are positive. Let us call this case A. However, we found that at some values of the parameters \Delta m2 and sin2 2` the minimum of O/2 is achieved with unreasonable values of some neutrino fluxes (e.g. 100 times larger than the corresponding SSM fluxes). Thus we also considered the following case B: the different solar neutrino fluxes are allowed to vary in the interval ,rmin \Phi r*e(BP) ^ \Phi r*e ^ ,rmax \Phi r*e(BP), where \Phi r*e(BP) are the BP-SSM value of the neutrino fluxes and the factors ,rmin and ,rmax are chosen in such a way to include the predictions of the different existing solar models [5-9,24]. The values of these factors are given in Table II. We determined the minimum (maximum) values for the pp, pep, 7Be and Hep fluxes by subtracting (adding) 3 times the range of solar model predictions to the minimum (maximum) predicted flux (notices that this range is larger than the 1oe error given by BP). Since it has been recently suggested [25] that the value of the astrophysical factor S17(0) could be significantly lower than that used in SSM calculations, we let the 8B flux to be arbitrarily small. In the recent Dar and Shaviv calculation [24] the CNO fluxes are very small. In order to cover this possibility we let also the CNO fluxes to be arbitrarily small. We determined the maximum values of the 8B and CNO fluxes by adding 3 times the 1oe error calculated by BP to the BP average value.

The excluded regions of the parameters \Delta m2 and sin2 2` in case B are presented in Figs.5 and 6. As it can be seen from a comparison of these figures with Figs.3 and 4 the excluded regions in case B are larger than in the case where no limitation is imposed on the values of the neutrino fluxes.

Let us emphasize that the limits for the solar neutrino fluxes in case B are chosen arbitrarily. Nevertheless, it is interesting and instructive to investigate how the forbidden regions in the \Delta m2-sin2 2` plane change if some limits on the allowed values of the neutrino fluxes are imposed. The case B illustrates these changes. Let us also stress that the limits on the allowed values of the neutrino fluxes which we imposed in case B are rather wide in comparison with the predictions of the existing solar models.

III. RESULTS AND DISCUSSION A. Vacuum Oscillations (Constant Transition Probability) We discuss here the results of a model independent analysis of solar neutrino data under the assumption that vacuum neutrino oscillations take place and that the oscillating terms in the transition probability vanish due to averaging over the neutrino spectrum, the region of the sun where neutrinos are produced, and so on. In this case the transition probabilities are constant and are determined by the elements of the mixing matrix. In the case of oscillations between n neutrino states the minimum value of the averaged *e survival probability is equal

5

to Pmin*e!*e = 1=n (see Ref. [18]). This minimum corresponds to maximum neutrino mixing. We consider transitions of solar *e into active as well as into sterile states3.

The curves in Fig.1 represent the maximum confidence levels of the fits as functions of the survival probability P*e!*e for transitions of *e into *_ and/or *o/ . It can be seen from Fig.1 that the following values of P*e!*e are excluded at 90% CL:

P*e!*e ! 0:11 and 0:18 ! P*e!*e ! 0:45 (A) ; P*e!*e ! 0:54 and P*e!*e ? 0:89 (B) :

The following values of P*e!*e are excluded at 95% CL:

P*e!*e ! 0:09 (A) ; P*e!*e ! 0:43 and P*e!*e ? 0:95 (B) :

From Fig.1 one can also see that P*e!*e = 1=2 and P*e!*e = 1=3, which correspond to a maximum mixing in the case of two and three oscillating neutrino states, respectively, are excluded at 87% and 93% CL in case A and at 92% and 98% CL in case B.

In Fig.2 the maximum confidence level as a function of the survival probability P*e!*e is presented for the case of transitions of solar *e's into sterile states. From Fig.2 one can see that the following values are excluded at 90% CL:

P*e!*e ! 0:61 (A) ; P*e!*e ! 0:66 and P*e!*e ? 0:86 (B) :

From Fig.2 one can also see that P*e!*e = 1=2 is excluded at 97% and at 98% CL in cases A and B, respectively.

B. MSW Transitions in Matter Here we discuss the results of a model independent analysis of the solar neutrino data in the case of resonant MSW transitions of solar *e's into *_ (or *o/ ) or sterile neutrinos. Let us remind that a MSW resonance takes place if at some point xR the following equation is satisfied

2 p2 GF E Ne(xR) = \Delta m2 cos 2` : (3.1) Here GF is the Fermi constant, E is the neutrino energy and Ne is the electron density. The electron density has a maximum in the center of the sun: N ce ' 100NA cm\Gamma 3, where NA is the Avogadro number. From Eq.(3.1) it follows that MSW transitions can take place if the following condition is satisfied:

3The difference between these two types of transitions is due to the Kamiokande data: *

_ and/or

*o/ give a contribution to *-e scatterings detected in the Kamiokande experiment while sterile

neutrinos *S do not interact with matter.

6

E & \Delta m

2 cos 2`=eV2

1:5 \Theta 10\Gamma 5 MeV : (3.2) Let us consider first MSW transitions of *e's into *_ (or *o/ ). The region in the \Delta m2- sin2 2` plane excluded by the existing solar neutrino data in case A (no limits on the values of the neutrino fluxes) is presented in Fig.3. In this figure we have also plotted (shaded areas) the two allowed regions (90% CL) found by us from an analysis of the data with the BP neutrino fluxes. These allowed regions practically coincide with those found by other authors [10-16]. The excluded region has the triangular shape typical of a strong *e suppression. Since this region extends below \Delta m2 cos 2` ' 3 \Theta 10\Gamma 6 eV2, from Eq.(3.2) it is clear that in this region a MSW resonance takes place for neutrinos from all sources. In this region the value of the flux of pp *e's on the earth is strongly suppressed. This suppression of the flux of pp *e's is in contradiction with the Gallium data. It cannot be counterbalanced by a large initial pp flux \Phi pp*e because \Phi pp*e is limited by the luminosity constraint. The observed Gallium

event rate cannot be due to a high value of \Phi

8B

*e because the

8B neutrino flux is constrained

by the data of the Kamiokande experiment in which both *e and *_ are detected (notice that in most of the excluded region the 8B *e flux on the earth is not suppressed because the corresponding MSW transition is highly non-adiabatic). The observed Gallium event rate cannot either be due to high values of the other neutrino fluxes (pep, 7Be and CNO) because the values of these fluxes are constrained by the data of the Homestake experiment.

The result of the calculation of the excluded regions in the \Delta m2-sin2 2` plane for *e-*S MSW transitions in case A is presented in Fig.4. The allowed region (90% CL) found with the BP neutrino fluxes is also shown in Fig.4 (shaded area). This allowed region practically coincides with that found in Ref. [12]. It can be seen from Fig.4 that there are two excluded regions of parameters in the case of transitions of *e's into sterile neutrinos. The large region is similar to the excluded region in the case of *e ! *_(*o/ ) transitions. The other region is specific for *e ! *S transitions. Since in this region \Delta m2 & 5 \Theta 10\Gamma 5 eV2, from Eq.(3.2) it follows that this region corresponds to a MSW suppression of the flux of high energy 8B *e's. A strong suppression of high energy 8B *e's only is not allowed by the data of the solar neutrino experiments. In fact, this suppression cannot be counterbalanced by a large initial total 8B flux (to fit the Kamiokande data) because it would give too large Clorine and Gallium event rates (in contradiction with the Homestake and GALLEX+SAGE data) due to the unsuppressed low energy part of the 8B neutrino flux.

Now we discuss what happens with the exclusion regions if limits on the solar neutrino fluxes are imposed (case B). The excluded regions for MSW *e ! *_(*o/ ) and *e ! *S transitions in case B are shown in Figs.5 and 6. A comparison of Fig.3 and Fig.5 (Fig.4 and Fig.6) shows that the region of values of the parameters \Delta m2 and sin2 2` that is forbidden by the existing solar neutrino data is strongly increased if we put some limits on the possible values of the neutrino fluxes. Figs.5 and 6 illustrate the fact that even assuming rather wide limits for the values of the solar neutrino fluxes, the existing solar neutrino data strongly restrict the region of possible values of the parameters \Delta m2 and sin2 2`. The triangular excluded region for \Delta m2 . 5 \Theta 10\Gamma 6 eV2 is due to a strong suppression of the flux of low energy pp *e's. The excluded region with 5 \Theta 10\Gamma 6 eV2 . \Delta m2 . 10\Gamma 4 eV2 is due to a large suppression of the flux of 8B *e's. As it can be seen from Figs.5 and 6, this excluded region is larger in the case of *e ! *S transitions than in the case of *e ! *_(*o/ ) transitions. This

7

is connected with the fact that the sterile neutrinos that are produced in *e ! *S transitions do not contribute to the Kamiokande event rate, while the *_(*o/ ) that are produced in *e ! *_(*o/ ) transitions interact with electrons (with a cross section that is about 0.15 of the cross section for *e-e scattering). Let us notice that in this region, corresponding to a large suppression of 8B *e's, it is also impossible to fit the Clorine and Gallium data: the large flux of 7Be neutrinos that is necessary to fit the Homestake data would give an excessive Gallium event rate. In the large (left) part of the excluded region P*e!*e = 1, which is excluded at 98% CL.

C. Vacuum Oscillations In this section we consider vacuum neutrino oscillations in the case in which the oscillating term in the transition probability does not vanish due to averaging. As it is well known, in the case of oscillations between two neutrino states (*e ! *_(*o/ ) or *e ! *S) the *e survival probability is given by

P*e!*e (E; R) = 1 \Gamma 12 sin2 2` 1 \Gamma cos \Delta m

2 R

2E ! ; (3.3) where R is the distance between the point where neutrinos are produced and the earth. In the expression for the observable event rates this probability is integrated over the neutrino energy, over the region of the sun where neutrinos are produced and over the varying distance between the sun and the earth. If the average neutrino oscillation length is comparable with the sun-earth distance and if the mixing angle ` is large, then the cosine term in Eq.(3.3) could give a non-zero contribution to the event rates. The regions of values of the parameters \Delta m2, sin2 2` which allow to describe the experimental data (with the solar neutrino fluxes predicted by the BP-SSM) were given in Ref. [13]. We have found the regions of values of the parameters \Delta m2 and sin2 2` that are excluded for any value of the initial neutrino fluxes (case A).

For the problem of the "survival" of the contribution of the cosine term in Eq.(3.3) the most important integration is the one over energy. Let us write down this term as

cos ss xx

0 ; (3.4)

where x = 1=E and x0 = 2ss=\Delta m2R. Let x be the characteristic length over which the product of the neutrino cross section and the neutrino flux changes. If x0 o/ x the cosine term does not give any contribution to the integral. If x0 AE x the neutrino oscillations are not observable (P*e!*e = 1). Only in the case x0 ' x we can expect that the cosine term gives a non-zero contribution to the event rate. In this case, for the parameter \Delta m2 we have the following estimate:

\Delta m2 ' 2 ss Ed ' 8:3 \Theta 10\Gamma 12 eV2 EMeV ; (3.5) where E is the average neutrino energy. For pp, 7Be and 8B neutrinos we have \Delta m2 ' 2:5 \Theta 10\Gamma 12 eV2, \Delta m2 ' 7:5 \Theta 10\Gamma 12 eV2, \Delta m2 ' 5:8 \Theta 10\Gamma 11 eV2, respectively.

8

Fig.7 shows the regions of the values of the parameters \Delta m2 and sin2 2` that are excluded in a model independent way by the existing experimental data in the case of *e ! *_(*o/ ) vacuum oscillations. The allowed regions (90% CL) of the values of the parameters that was obtained by us with the BP-SSM values of the neutrino fluxes are also shown in Fig.7 (shaded areas). These regions practically coincide with the allowed regions found in Ref. [13]. In the large excluded region with \Delta m2 ' 2:5 \Theta 10\Gamma 12 eV2 the flux of pp neutrinos is strongly suppressed. As we discussed above, this suppression is in contradiction with the data of the Gallium experiments (if the constraints from the other experiments and from the solar luminosity are taken into account). The other excluded regions with \Delta m2 ! 5 \Theta 10\Gamma 11 eV2 are due to an excessive suppression of the fluxes of pp and 7Be neutrinos.

The excluded regions of the values of the parameters \Delta m2 and sin2 2` in the case of *e ! *S vacuum oscillations are shown in Fig.8. As in the case of MSW transitions in matter, the excluded regions due to a strong suppression of the high energy 8B neutrinos are larger in the case of *e ! *S oscillations than in the case of *e ! *_(*o/ ) oscillations. In this case we did not find any allowed region with the SSM-BP neutrino fluxes (in agreement with Ref. [13]).

IV. CONCLUSION We have presented here the results of a model independent analysis of the existing solar neutrino data. We have considered resonant MSW transitions in matter and vacuum oscillations in the case of mixing between two neutrino types (*e-*_(*o/ ) and *e-*S, where *S is a sterile neutrino). We have obtained forbidden regions of the values of the parameters \Delta m2 and sin2 2` without any assumption about the values of the total neutrino fluxes from all sources (taking into account only the luminosity constraint). We have also obtained the forbidden regions in the \Delta m2-sin2 2` plane in a case in which some wide limits, that take into account the predictions of the different solar models, are imposed on the values of the total neutrino fluxes. We have shown that in this model independent approach the existing solar neutrino data allow to forbid rather large regions of values of the parameters \Delta m2 and sin2 2` (expecially when limits on the values of the total neutrino fluxes are imposed).

ACKNOWLEDGMENTS It is a pleasure for us to express our gratitude to G. Conforto, S. Degl'Innocenti, P. Krastev, F. Martelli, S. Petcov and A. Smirnov for very useful discussions. We would like to thank B. Balantekin and W. Haxton for the hospitality at Institute for Nuclear Theory of the University of Washington where part of this work was carried out.

9

REFERENCES [1] R. Davis Jr., Talk presented at the 6th International Workshop on Neutrino Telescopes,

Venezia, March 1994. [2] GALLEX Collaboration, Phys. Lett. B 285, 376 (1992); Phys. Lett. B 314, 445 (1993);

Phys. Lett. B 327, 377 (1994). [3] V.N. Gavrin, Talk presented at the 6th International Workshop on Neutrino Telescopes,

Venezia, March 1994. [4] K. S. Hirata et al., Phys. Rev. Lett. 65, 1297 (1990); Phys. Rev. D 44, 2241 (1991);

Y. Suzuki, Talk presented at the 6th International Workshop on Neutrino Telescopes, Venezia, March 1994. [5] J.N. Bahcall and R. Ulrich, Rev. Mod. Phys. 60, 297 (1988); J.N. Bahcall, Neutrino

Physics and Astrophysics, Cambridge University Press, 1989; J.N. Bahcall and M.H. Pinsonneault, Rev. Mod. Phys. 64, 885 (1992). [6] S. Turck-Chi`eze, S. Cahen, M. Cass'e and C. Doom, Astrophys. J. 335, 415 (1988); S.

Turck-Chi`eze and I. Lopes, Astrophys. J. 408, 347 (1993); S. Turck-Chi`eze et al., Phys. Rep. 230, 57 (1993). [7] V. Castellani, S. Degl'Innocenti and G. Fiorentini, Astronomy & Astrophysics 271, 601

(1993); S. Degl'Innocenti, INFN-FE-07-93. [8] I.J. Sakmann, A.I. Boothroyd and W.A. Fowler, Astrophys. J. 360, 727 (1990). [9] G. Berthomieu et al., Astronomy & Astrophysics 268, 775 (1993). [10] GALLEX Collaboration, Phys. Lett. B 285, 390 (1992). [11] X. Shi, D.N. Schramm and J.N. Bahcall, Phys. Rev. Lett. 69, 717 (1992). [12] S.A. Bludman, N. Hata, D.C. Kennedy and P.G. Langacker, Phys. Rev. D 47, 2220

(1993); N. Hata and P.G. Langacker, Phys. Rev. D 50, 632 (1994). [13] P.I. Krastev and S.T. Petcov, Phys. Lett. B 299, 99 (1993); Phys. Rev. Lett. 72, 1960

(1994). [14] L.M. Krauss, E. Gates and M. White, Phys. Lett. B 299, 94 (1993); Fermilab-Pub94/176-A (hep-ph@xxx.lanl.. [15] G.L. Fogli and E. Lisi, Astropart. Phys. 2, 91 (1994). [16] G. Fiorentini et al., Phys. Rev. D 49, 6298 (1994). [17] L. Wolfenstein, Phys. Rev. D 17 (1978) 2369; S. P. Mikheyev and A. Y. Smirnov, Il

Nuovo Cimento C 9 (1986) 17. [18] S.M. Bilenky and B. Pontecorvo, Phys. Rep. 41 (1978) 225. [19] J.N. Bahcall, Phys. Rev. D 44, 1644 (1991). [20] K. S. Hirata et al., Phys. Rev. Lett. 65, 1301 (1990); Phys. Rev. Lett. 66, 9 (1991). [21] S.T. Petcov, Phys. Lett. B 200, 373 (1988); P.I. Krastev and S.T. Petcov, Phys. Lett.

B 207, 64 (1988); T.K. Kuo and J. Pantaleone, Phys. Rev. D 39, 1930 (1989); Rev. Mod. Phys. 61, 937 (1989). [22] P.I. Krastev and S.T. Petcov, Phys. Lett. B 285, 85 (1992). [23] Review of Particle Properties, Phys. Rev. D 45, Part II (1992). [24] A. Dar and G. Shaviv, Technion-Ph-94-5 . [25] T. Motobayashi et al., Yale-40609-1141.

10

TABLES TABLE I. Data of solar neutrino experiments and rates predicted by BP [5], TL [6] and CDF [7]. N BPKAM is the Kamiokande event rate predicted by BP.

Experiment Event Rate (SNU) SSM Predictions (SNU)

BP TL CDF Kamiokande N expKAM=N BPKAM = 0:51 \Sigma 0:04 \Sigma 0:06 1 \Sigma 0:14 0:8 \Sigma 0:2 0:98

Homestake N expHOM = 2:32 \Sigma 0:23 8:0 \Sigma 1:0 6:4 \Sigma 1:4 7:8

GALLEX N expGAL = 79 \Sigma 10 \Sigma 6 131:5+7\Gamma 6 123 \Sigma 7 131

SAGE N expGAL = 74 \Sigma 19 \Sigma 10

TABLE II. Solar neutrino fluxes (with 1oe errors) predicted by BP. hEi is the average neutrino energy, YLUM = Q=2 \Gamma hEi, where Q = 26:73 MeV, and ,min and ,max determine the limits for the values of the total neutrino fluxes in case B.

Reaction hEi(MeV) YLUM(MeV) \Phi

*e(BP)

(cm\Gamma 2sec\Gamma 1) ,min ,max

pp 0.265 13.10 (6:00 \Sigma 0:004) \Theta 1010 0.93 1.07 pep 1.442 11.92 (1:43 \Sigma 0:02) \Theta 108 0.61 1.29 7Be 0.813 12.55 (4:89 \Sigma 0:29) \Theta 109 0.46 1.40

8B 6.710 6.66 (5:69 \Sigma 0:82) \Theta 106 0 1.43

Hep 9.625 3.74 1:23 \Theta 103 0.90 1.13

13N 0.7067 12.66 (4:92 \Sigma 0:84) \Theta 108 0 1.51

15O 0.9965 12.37 (4:26 \Sigma 0:82) \Theta 108 0 1.58

17F 0.9994 12.37 (5:39 \Sigma 0:86) \Theta 106 0 1.48

11

FIGURES B ne->nu(t)

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

A

_P

ne->ne

Con fide nce Le vel

FIG. 1. Maximum Confidence Level as a function of the survival probability P*

e!*e in the case

of a constant averaged probability for transitions of solar *e's into *_ and/or *o/ . Curve A was

obtained in case A, in which no limits on the values of the neutrino fluxes are assumed. Curve B corresponds to case B, in which some limits on the values of the neutrino fluxes are imposed.

12

B

Pn

e->ne

_

ne->nS 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

A Con fide nce Le vel

FIG. 2. Maximum Confidence Level as a function of the survival probability P*

e!*e in the case

of a constant averaged probability for transitions of solar *e's into sterile neutrinos. Curve A was

obtained in case A, in which no limits on the values of the neutrino fluxes are assumed. Curve B corresponds to case B, in which some limits on the values of the neutrino fluxes are imposed.

13 10-9 10-8 10-7 10-6 10-5 10-4 10-3

10-4 10-3 10-2 10-1 100

sin22q

\Delta m 2 (e

V2 )

99% CL 90% CL 95% CL

FIG. 3. MSW transitions: excluded regions in the sin2 2#-\Delta m2 plane for *e-*_(*o/ ) mixing in case A (no limits on the values of the neutrino fluxes are assumed). The allowed regions found with the BP neutrino fluxes are also shown (shaded areas).

14

10-9 10-8 10-7 10-6 10-5 10-4 10-3

10-4 10-3 10-2 10-1 100

sin22q

\Delta m 2 (e

V2 )

99% CL 90% CL 95% CL

FIG. 4. MSW transitions: excluded regions in the sin2 2#-\Delta m2 plane for *e-*S mixing in case A (no limits on the values of the neutrino fluxes are assumed). The allowed region found with the BP neutrino fluxes is also shown (shaded area).

15

10-9 10-8 10-7 10-6 10-5 10-4 10-3

10-4 10-3 10-2 10-1 100

sin22q

\Delta m 2 (e

V2 )

99% CL 90% CL 95% CL

FIG. 5. MSW transitions: excluded regions in the sin2 2#-\Delta m2 plane for *e-*_(*o/ ) mixing in case B (some limits on the values of the neutrino fluxes are imposed). The allowed regions found with the BP neutrino fluxes are also shown (shaded areas).

16

10-9 10-8 10-7 10-6 10-5 10-4 10-3

10-4 10-3 10-2 10-1 100

sin22q

\Delta m 2 (e

V2 )

99% CL 90% CL 95% CL

FIG. 6. MSW transitions: excluded region in the sin2 2#-\Delta m2 plane for *e-*S mixing in case B (some limits on the values of the neutrino fluxes are imposed). The allowed region found with the BP neutrino fluxes is also shown (shaded area).

17

ne->nu(t) 10-12 10-11 10-10 10-9

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

sin22q

\Delta m 2 (

eV 2 )

99% CL 90% CL 95% CL

FIG. 7. Vacuum oscillations: excluded regions in the sin2 2#-\Delta m2 plane for *e-*_(*o/ ) mixing in case A (no limits on the values of the neutrino fluxes are assumed). The allowed regions found with the BP neutrino fluxes are also shown (shaded areas).

18

ne->nS 10-12 10-11 10-10 10-9

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

sin22q

\Delta m 2 (

eV 2 )

99% CL 90% CL 95% CL

FIG. 8. Vacuum oscillations: excluded regions in the sin2 2#-\Delta m2 plane for *e-*S mixing in case A (no limits on the values of the neutrino fluxes are assumed).

19

