Solar Neutrinos

Davis neutrino detector at Homestake 1970

Since the pioneering work of Bethe and others, it was anticipated that the energy of the Sun was produced in its core by the fusion of hydrogen into helium [Wei37,Bet38,Bet39]. This process is accompanied by the emission of neutrinos, which are the only witnesses of the phenomenon. In 1964, John Bahcall developed his first solar model and calculated that the chlorine-argon reaction would be the most promising method to observe solar neutrinos [Bah64]. Simultaneously, Ray Davis revived his chlorine-argon detector [Dav55]. He proposed to install 390 000 liters of perchlorethylene (C2Cl4) in the Homestake Gold mine (South Dakota), deep underground to avoid cosmic rays interaction. Such a detector would observe few 37Cl to 37Ar conversion per day [Dav64]. His first results, in 1968, were a big surprise: Davis observed less than 0.2 37Ar produced per day, much less than the prediction [Dav68]. This is the start of the solar neutrino problem, which will be solved only in 2001. At the end of the 70’s, the solar neutrino deficit was better quantified: (0.39± 0.06) 37Ar were produced per day translated into 2.1±0.3 solar neutrino units and compared to 7.8±1.5 expected from solar models [Cle81]. The final result has been published in 1998 [Cle98].

Solar neutrino flux

The basic nuclear reaction in the core of the Sun is the fusion of two protons, giving a deuterium, a positron and an electron-neutrino νe. It follows a complicated cycle of nuclear reactions which also produce neutrinos, each with a specific energy spectrum [Fig]. The spectroscopy of solar neutrinos includes: pp-neutrinos, 7Be-neutrinos, pep-neutrinos and 8B neutrinos, with a small amount of CNO-neutrinos. The chlorine experiment of R. Davis was sensitive only to the 7Be- and 8B neutrinos.

After the first chlorine results, many ideas to observe solar neutrinos have been proposed, radiochemical (using gallium, lithium or bromine targets), geochemical (using molybdenum) or direct counting experiments (indium target or elastic interaction on electrons or deuterium). Few of them survived, in spite of beautiful R&D developments.

The first one, a direct counting experiment, was Kamiokande, a large Cerenkov detector (3000 tons of water seen by 1000 PMT’s) designed to observe proton decay. After an improvement of the detector, the solar neutrino detection started in 1986, with a threshold of 7 MeV (i.e. a sensitivity to solar 8B-neutrinos only). Kamiokande observed the scattered electrons from the elastic reaction νe + e⁻  → e⁻ + νe. In 1989, Kamiokande announced to have observed only about 45% of the expected solar neutrinos [Hir89], confirming the solar neutrino problem.

The gallium target was soon considered as very attractive, since it was sensitive to the complete solar neutrino spectrum, including neutrinos from the primary pp fusion reaction. The idea to use the reaction νe +71Ga → e⁻ + 71Ge (followed by the observed 71Ge decay) had been proposed in 1966 [Kuz66], 12 years before the first concrete proposal [Bah78b]. It took again 12 years to see two gallium experiments starting, GALLEX in the Gran Sasso Underground Laboratory (Italy) and SAGE in the Baksan Laboratory (Russia). The first observation of solar pp-neutrinos was announced by GALLEX in June 1992 [Ans92], later confirmed by SAGE [Abd94]. The gallium experiments observed only 60% of the neutrinos predicted by the solar models. The different deficit observed in the different experiments made the solar neutrino problem more and more puzzling.

The solution to the solar neutrino problem (2001-2002)

In the middle of the 90’s, the solar neutrino problem was experimentally established: a deficit of solar νe was observed in the three experiments (chlorine, Kamiokande and gallium). It was reinforced by the measurement provided by the SuperKamiokande experiment in 1998 [Fuk98a], more precise than the Kamiokande one. Three ways to solve the problem were discussed:

  1. the experiments are wrong (but the check of the gallium experiments with artificial neutrino sources makes this unlikely);
  2. the solar models are wrong (but the models developed by Bahcall and other independent groups have been considerably improved and now provide relatively similar and stable results);
  3. the νe produced in the core of the Sun oscillate in νμ or ντ before they reach the Earth.

This exciting third solution became more and more popular, but it had to be proved! At the same period, starting from the observation that the deficit is not uniform, people developed a simple arithmetic mixing the experimental results with the solar model predictions which showed that solar νe coming from the 7Be chain seemed to have completely disappeared. If no serious solar model could accommodate this surprising conclusion, the oscillation mechanism completed by the MSW effect had no difficulty to explain the phenomenon. But the experimental proof was still missing.

In 1985, Herbert Chen pointed out that the reaction on deuterium ν+ d → νx+ p + n (where νx represents any neutrino flavor) would give a direct measurement of the solar neutrino flux, independently of any possible oscillation [Che85]. This was the starting point of the Sudbury Neutrino Observatory (SNO): the use of a large heavy-water Cerenkov detector deep underground in the Sudbury mine (Canada) [Aar87]. It took more than 10 years to build the detector (1000 tons of heavy-water D2O observed by 9500 photomultipliers) which started to take data in November 1999. With a threshold of 6.75 MeV, the detector was sensitive only to 8B solar neutrinos (like SuperKamiokande). The first result, announced in June 2001, was really great: the observed solar νe flux was similar to that observed by SuperKamiokande in normal water (i.e. showing a deficit), but the total flux was similar to the one predicted by solar models (about 5 10cm-2s-1) [Ahm01]. This was confirmed in April 2002 with a direct measurement of the neutron in the reaction ν+ d → νx+ p + n [Ahm02]. The two important results were that solar neutrinos oscillate and that the solar models are correct.

The demonstration that neutrinos oscillate had already been made by the measurement of atmospheric neutrinos in the Superkamiokande experiment in 1998 [Fuk98a], but it concerned oscillation of νμ into ντ. For SNO, the oscillation is between νe and νμ or ντ. The phenomenology of neutrino mixing can be developed and involves three mixing angles θij and two Δm2ij.

Other tracks for solar neutrino detection

Beyond the main experiments, many R&D developments using different targets have been performed to study solar neutrinos. We quote:

  • Lithium: The use of 7Li isotope as a promising target for capturing 7Be and 8B solar neutrinos via the reaction νe+7Li →7Be + ewas first proposed by Zatsepin and Kuzmin and independently by Bahcall . See [Kop09] for the details.
  • Heron: The use of superfluid helium at very low temperature has been proposed by Lanou in 1987 [Lan87]. The developments of this method sensitive to pp neutrinos is reviewed in [Hua07].
  • Iodine: Haxton proposed in 1988 a new radiochemical experiment, using the reaction 127I→127Xe, similar to the chlorine-argon for the threshold and for the method, but with a rate ten times larger [Hax88].
  • Indium: In 1976, Raghavan proposed a low-threshold, direct counting detector for solar pp-neutrinos. It was based on the inverse beta-decay 115In into 115Sn* [Rag76].  This very attractive idea has been developed in many R&D’s but suffered from the intrinsic radioactivity of the indium which prevented to build a real experiment [Rag01].
  • LENS (Ytterbium): In 1997, Raghavan proposed new ideas for real-time spectroscopy of low energy electron neutrinos from the Sun [Rag97]. The target was scintillator doped with different materials. LENS (an R&D using ytterbium) was developed over several years but did not conclude.
  • Bromine: In 1984, Hurst, Davis et al. studied the feasibility of an experiment directly counting the 81Kr in the reaction  81Br(νe,e)81Kr and sensitive mainly to 7Be solar neutrinos [Hur84].
  • Molybdenium: In 1982, Cowan and Haxton proposed a geochemical experiment to determine the 8B  solar neutrino flux, averaged over the past several million years, from the concentration of technetium-98 in molybdenite [Cow82].
  • ICARUS: in 1986, Bahcall, Baldo-Ceolin, Cline and Rubbia proposed a large liquid argon detector, ICARUS (Imaging Cosmic And Rare Underground Signals), which would be sensitive to 8B solar neutrinos [Bah86]. Important R&D’s have been developed. The T600 prototype, a TPC containing 600 tons of liquid argon, has been built at the Gran Sasso laboratory with other objectives that the measurement of solar neutrinos.
  • Hellaz: A high rate solar neutrino detector with energy determination has been proposed in 1992 [Seg92]. This detector sensitive to solar pp-neutrinos was studied until the beginning of the 2000’s [Gor01]
  • Super MUNU.


Neutrinos and the Sun

Almost a century ago, after the first ideas by Eddington in the 20’s, followed by Atkinson and Houtermans [Atk29], physicists assumed that the solar (and stellar) energy was powered by nuclear reactions. The simplest possible reaction, p + p → 2H + e⁺ + n, was first suggested by Carl von Weizsäcker [Wei37] and calculated by Bethe and Critchfield [Bet38]. This fundamental reaction, which emits the primary pp-neutrinos, was the origin of the so-called pp cycle (different nuclear reactions which are other sources of neutrinos) and Bethe developed also the so-called CNO cycle (which plays a minor role in the Sun but is dominant in more massive stars and produces also neutrinos) [Bet39]. Neutrinos then become privileged witnesses of what happens in the core of the Sun. Our knowledge of the Sun has greatly improved since this time, and neutrinos have played a major role in the deep understanding of “how the Sun shines”. The solar neutrino problem, i.e. the deficit of solar neutrinos observed compared to the predictions of solar models, triggered an intense activity on the experimental side as well as on the theoretical side (solar models and neutrino properties). The great discoveries by SNO were that solar neutrinos are as many as predicted by solar models, and that νe are partially transformed into νμ or ντ proving the neutrino oscillation mechanism.

In addition to this achievement, the story of solar neutrinos has had further success. Since 2007, the Borexino experiment at the Gran Sasso laboratory has been able to measure with a reasonable precision all the components of the pp cycle (starting with the pp-neutrinos from the initial pp fusion reaction [Bel14], continuing with the pep, 7Be and 8B neutrinos), and to put a constraining limit on the CNO neutrinos, limit very close to the solar model predictions.

Brief bibliography on the solar neutrino problem and on solar models

A detailed account of the first steps of the solar neutrino detection and the raise of the “solar neutrino problem” is given by two pioneers in [1]. We also mention the interesting work by Pinch who, motivated by the longstanding solar neutrino problem, studied solar neutrinos from a sociological point of view [2]
[1] John N. Bahcall and Raymond Davis Jr., An account of the development of the solar neutrino problem in “Essays in Nuclear Astrophysics”, Cambridge University Press, 1982, edited by C.A. Barnes, D.D. Clayton, D.N. Schramm;
[2] Trevor Pinch. Confronting Nature, The sociology of Solar-Neutrino Detection, Kluwer/Reidel, 1986
The start of the development of solar models (in the frame of solar neutrino production) is the pioneering paper by John Bahcall [3]. It has been followed by the paper [4] which appeared in 1968 simultaneously with the first result of the chlorine experiment.
[3] J.N. Bahcall. Solar neutrinos. I. Theoretical, Phys. Rev. Lett. 12 (1964) 300
[4] J.N. Bahcall, N.A. Bahcall and G. Shaviv, Present status of the theoretical predictions for the 37Cl solar neutrino experiment, Phys. Rev. Lett. 20 (1968) 1209
Later Bahcall developed the so-called ”standard solar model” (SSM) with different collaborators and published regular updates. The main steps are found in references [5], [6] (solar models takes also into account the helioseismological observations), [7,8] (introduction of the helium and heavy element diffusion in the modeling). Among the other solar models developed over the years, we quote the work by S. Turck-Chièze and collaborators [9]. To explain the solar neutrino problem, many so-called non standard models have been studied; as an example, we quote the reference [10].
[5] J.N. Bahcall, W.F. Huebner, S.H. Lubow, P.D. Parker and R.K. Ulrich, Standard Solar Models and the Uncertainties in Predicted Capture Rates of Solar Neutrinos, Rev. Mod. Phys. 54 (1982) 767
[6] J.N. Bahcall and R.K. Ulrich, Solar Models, Neutrino Experiments and Helioseismology, Rev. Mod. Phys. 60 (1988) 297
[7] John N. Bahcall, M.H. Pinsonneault, Standard solar models, with and without helium diffusion, and the solar neutrino problem, Rev. Mod. Phys. 64 (1992) 885
[8] John N. Bahcall, M. H. Pinsonneault, and G. J. Wasserburg, Solar models with helium and heavy-element diffusion, Rev. Mod. Phys. 67 (1995) 781
[9] S. Turck-Chièze et al., The solar interior, Physics Reports 230 (1993) 57
[10] V. Castellani et al., Solar neutrinos : beyond standard solar models, Physics Reports 281 (1997) 309
After the death of John in 2005, the development of the standard solar model has been pursued by C. Pena-Garay, A. Serenelli and others. The most recent issue is [11].
[11] Núria Vinyoleset al., A new generation of standard solar models, arXiv:1611.09867
A complete account of the first 30 years of the solar neutrinos is given in [12].
[12] John Bahcall, Raymond Davis, Peter Parker, Alexei Smirnov, Roger Ulrich ed., Solar neutrinos: the first thirty years, Addison-Wesley (1994).
In 1989, a first interdisciplinary conference on the Sun was organized in Versailles, covering solar modelling, as well as neutrinos, helioseismology (the other experimental source of information on the interior of the Sun), solar dynamo or transport processes [13]. For science historians, it can also be used as a benchmark: what was known in these years.
[13] G. Berthomieu and M. Cribier, editors, “Inside the Sun”, Proc. of the 121stIAU Colloquium, Versailles, France, May 22-26, 1989, Kluwer Academic Publishers (1990).

Further information

During the conference on the History of the Neutrino (Sept. 5-7, 2018 in Paris) the history of Solar Neutrinos was reviewed in two talks :

  • T. Kirsten (MPI Heidelberg, Germany) for the pioneering experiments : here the slides and the video of his talk.
  • A. McDonald (Queen’s University, Canada) for the experiments which brought the solution to the solar neutrino problem : here the slides and the video of his talk.


Aar87G. Aardsma et al. A heavy water detector to resolve the solar neutrino problem Phys. Lett. B 194 (1987) 321
Abd02J.N. Abdurashitov et al., SAGE collaborationSolar neutrino flux measurements by the Soviet-American Gallium Experiment (SAGE) for half the 22 year solar cycleJ.Exp.Theor.Phys. 95 (2002) 181-193, arXiv:astro-ph/0204245
Abd09J.N. Abdurashitov et al. Measurement of the solar neutrino capture rate with gallium metal. III: Results for the 2002--2007 data-taking period arXiv:0901.2200, Phys. Rev. C80 (2009) 015807
Abd94J.N. Abdurashitov et al. Results from SAGE (The Russian-American Gallium solar neutrino experiment)Phys. Lett. B328 (1994) 234
Ahm01Q.R. Ahmad et al. Measurement of the rate n e + d → p + p + e - interactions produced by 8 B solar neutrinos at the Sudbury Neutrino Observatory Phys. Rev. Lett. 87 (2001) 071301
Ahm02Q.R. Ahmad et al. Direct evidence for neutrino flavor transformation from neutral-current interactions in the Sudbury Neutrino Observatory Phys. Rev. Lett. 89 (2002) 011301
Ans92P. Anselmann et al., Gallex collaborationSolar neutrinos observed by GALLEX at Gran SassoPhys. Lett. B285 (1992) 376
Atk29R. d'E. Atkinson and F.G. HoutermansZur Frage der Aufbaumöglichtkeit der Elementen in SternenZ. Phys. 54 (1929) 656
Bah64J.N. Bahcall Solar neutrinos. I. Theoretical Phys. Rev. Lett. 12 (1964) 300
Bah78bJohn N. Bahcall et al.Proposed solar neutrino experiment using Ga-71Phys. Rev. Lett. 40 (1978) 1351
Bah86J.N. Bahcall, M. Baldo-Ceolin, D.B. Cline, C. RubbiaPrediction for a liquid argon solar neutrino detectorPhys. Lett. B178 (1986) 324
Bel14G. Bellini et al. Neutrinos from the primary proton-proton fusion in the Sun Nature 512 (2014) 383
Bet38H.A. Bethe and C.L. Critchfield The formation of deuterons by proton combination Phys. Rev. 54 (1938) 248
Bet39H.A. Bethe Energy production in stars Phys. Rev. 55 (1939) 434
Che85H.H. Chen Direct Approach to Resolve the Solar-Neutrino Problem Phys. Rev. Lett. 55 (1985) 1534
Cle81B.T. Cleveland, R. Davis and J.K. Rowley Solar neutrino experiments and neutrino oscillationsAIP Conference Proc. 72 (1981) 322;
Cle98B.T. Cleveland et al. Measurement of the solar electron neutrino flux with the Homestake chlorine detectorAstrophysical Journal 496 (1998) 505
Cow82G.A. Cowan and W.C. HaxtonSolar neutrino production of technetium-97 and technetium-98Science 216 (1982) 51
Dav55Raymond Davis Jr. Attempt to Detect the Antineutrino from a Nuclear Reactor by the 37 Cl(n,e - ) 37 Ar reaction Phys. Rev. 97 (1955) 766
Dav64R. Davis, Jr. Solar neutrinos. II. Experimental Phys. Rev. Lett. 12 (1964) 303
Fuk96Y. Fukuda et al. Solar neutrino data covering solar cycle 22Phys. Rev. Lett. 77 (1996) 1683
Fuk98aY. Fukuda et al. Measurements of the solar neutrino flux from Super-Kamiokande’s first 300 daysPhys. Rev. Lett. 81 (1998) 1158
Gor01Ph. GorodetzkyThe solar neutrino project HELLAZ: status report on the hardware and on the simulationNucl. Instr. and Methods A471 (2001) 131
Ham98W. Hampel et al., Gallex collaborationFinal results of the 51Cr neutrino source experiment in GALLEXPhys. Lett. B420 (1998) 114
Ham99W. Hampel et al., Gallex collaborationGALLEX solar neutrino observations: results for GALLEX-IVPhys. Lett. B447 (1999) 127
Hax88W.C. HaxtonRadiochemical neutrino detection via 127 I( n e ,e - ) 127 XePhys. Rev. Lett. 60 (1988) 768
Hir89K.S. Hirata et al. Observation of 8 B solar neutrinos in the Kamiokande-II detector Phys. Rev. Lett. 63 (1989) 16
Hua07Y.H.Huang et al. Potential for precision measurement of solar neutrino luminosity by HERONarXiv:0711.4095
Hur84G.S. Hurst et al.Feasibility of a 81 Br( n e ,e - ) 81 Kr solar neutrino experimentPhys. Rev. Lett. 53 (1984) 1116
Kae10F. Kaether, W. Hampel, G. Heusser, J. Kiko, T. KirstenReanalysis of the GALLEX solar neutrino flux and source experimentsPhys. Lett. B685 (2010) 47
Kop09A.V. Kopylov et al. A lithium-beryllium method for the detection of solar neutrinosarXiv:0910.3889
Kuz66V.A. KuzminDetection of solar neutrinos by means of 71 Ga( n ,e - ) 71 Ge reactionSov. Phys. JETP 22 (1966) 1051
Lan87R.E. Lanou, H.J. Maris, G.M. SeidlDetection of solar neutrinos in superfluid heliumPhys. Rev. Lett. 58 (1987) 2498
Rag01R.S. Raghavanpp-solar neutrino spectroscopy; return tof the indium detectorarXiv:hep-ex/0106054
Rag97R.S. RaghavanNew prospects for real-time spectroscopy of low energy electron neutrinos from the SunPhys. Rev. Lett. 78 (1997) 3618
Seg92J. Séguinot, T. Ypsilantis, A. ZichichiA high rate solar neutrino detector with energy determinationReport Collège de France LPC 92-31 (1992)
Vin16N. Vinyoles et al.A new generation of standard solar modelsarXiv:1611.09867
Wei37C.F. von Weizsäcker Über Elementumwandlungen im Innern der Sterne Physikalisches Zeitschrift 38 (1937) 176


1 thought on “Solar Neutrinos”

  1. In august 1984 a solar neutrino conference was held in Lead (South Dakota) the city close to the Homestake gold mine where the Davis’s experiment was running. Participants stayed in Deadwood, an historic city where Bill Hickok, Calamity Jane and Seth Bullock are burried, a city as in western movies.
    During the reception celebrating the 70th Davis’s birthday many stories were told. I remember one : « To go to the experiment, Davis has to use the elevator as the miners. An elevator full of mud, watered permanently to avoid fire, and descending at high speed in a terrible noise. After the first results of the chlorine experiment, Davis looks concerned. Then an old miners told him : Don’t worry Professor, the summer was really bad this year »

Leave a Reply