Ray Davis (left) and John Bahcall, pioneers in netrino astronomy. "For the last 33 years, Ray and I have been interacting more or less continuously," says Bachall.

   With his many contributions to neutrino astrophysics, John N. Bahcall has played a major part in opening a window on the universe. Thirty years ago Ray Davis, then working at Brookhaven National Laboratory and now at the University of Pennsylvania, suggested that it might be possible to detect neutrinos from the sun as a result of Bahcall's calculations on solar nuclear reactions. The unambiguous detection of these neutrinos directly verified the hypothesis that nuclear reactions generate stellar energy.

   A physicist with a deep interest in neutrino production in the universe, Bahcall graduated in physics from the University of California, Berkeley, in 1956. A master's from the University of Chicago followed a year later. For his doctorate, he was supervised by David Layzer of Harvard University, and after a postdoctoral position at Indiana University he moved to Caltech.

   Bahcall's interest in neutrinos was sparked early in his academic career. At Indiana he attended lectures given by Emil Kanapenski on weak interaction theory. As a self-imposed exercise for the student, he calculated some reaction rates. Later, an astronomer pointed him to the paper on cosmic nucleosynthesis by Margaret Burbidge, Geoffrey Burbidge, Willie Fowler, and Fred Hoyle (E.M. Burbidge, et al., Rev. Mod. Phys., 29:547, 1957). When Bahcall read the appendix, he saw that laboratory reaction rates had been used for beta decay in stellar interiors, but from his own calculations on beta decay he realized that the rates in stars would be very different from those in the laboratory.

   Since 1970 Bahcall has been an astrophysicist at the Institute for Advanced Study in Princeton, New Jersey. His interests range much more widely than studies of solar neutrinos, and include the search for dark matter in the universe and models of the galaxy. Most recently he has taken a leading role in applying the Hubble Space Telescope to studies of quasars. At a recent meeting of the American Astronomical Society, both Bahcall and Davis received prizes for their work on solar neutrinos, and Bahcall spoke to Science Watch's Physics correspondent Simon Mitton about his work in neutrino astrophysics.

            Your association with Willie Fowler, who died recently, was a long and fruitful one. How did that arise?

   Bahcall: I wrote a short paper saying that the weak interaction rates being used by astrophysicists could not be correct because the laboratory rates would be changed in stars: ionization and the Pauli principle would play an effect at the high densities inside stars. Willie Fowler was the referee for that, and he was always very generous when there was a new idea. He invited me to the Kellogg Radiation Laboratory at Caltech to continue my work on weak interactions. At the same time Fowler wrote to Ray Davis saying I'd done some interesting calculations relevant to nuclear reactions in the sun. Ray expressed an interest in the rates for electron capture by 7Be, saying that he would love to build a detector for the neutrinos produced by the 7Be. Fowler encouraged me to work on this, and for the last 33 years Ray and I have been interacting more or less continuously.

            Throughout your period of collaboration with Ray Davis, the story of solar neutrino hunting has been one of the observed flux lagging the predicted rate, with the observed rates one-half or less than predicted. Your two most highly cited papers in Reviews of Modern Physics—on the solar models and neutrino rates—have recorded more than 700 citations between them, which is high for physical sciences papers. What have we learned about neutrino astronomy from the research sparked by these papers?

   Bahcall: On a primitive level we've got a very fundamental result: we've confirmed experimentally that the sun shines by burning nuclear material in its core! This resolved a controversy that goes back to the middle of the 19th century. Advances in geology had shown the great age of Earth and raised the question for astronomers of how the sun could generate energy for millions of years. Eddington, in the early part of the 20th century, suggested that nuclear fusion is the source of that energy, but it took another 20 years before Hans Bethe developed fully the theory of nuclear reactions inside the sun. Ray confirmed this theory with his famous chlorine experiment conducted deep underground in Lead, South Dakota, in which a few atoms of 37Ar are produced each month when a 37Cl atom captures a solar neutrino.
   Today, three further experiments have detected neutrinos, and there is no doubt that they come from the sun. In the Kamiokande experiment (Japan), neutrino-electron scattering occurs. The Cerenkov light from scattered electrons shows that the high-energy 8B neutrinos causing the scattering have come from the direction of the sun. Two gallium experiments, GALLEX (Italy) and SAGE (Russia), detect the low-energy neutrinos from the basic proton-proton fusion reaction. These use absorption of a neutrino by 71Ga to produce 71Ge and an electron. The neutrino flux is measured to an accuracy of about 10%. So all four experiments detect solar neutrinos.
   The measured flux is the same as the predictions to within a factor of about 2-4 too little, depending on the experiment, and their energies are exactly what we expect from the theory. The data are specifically in agreement with the idea that the sun shines by burning four protons to form 4He, emitting two neutrinos in the cycle. This is in contrast with the reactions Hans Bethe thought were dominant, in which C, N, and O nuclei catalyze the four protons. So we've learned that reactions among the lightest nuclei alone keep dwarf stars like the sun shining.
   The interest shown in my 1982 and 1988 papers shows what a tremendous achievement it has been for a huge community of physicists, engineers, chemists, and astrophysicists—hundreds of people—to demonstrate what goes on deep inside stars. We've answered the question of how the stars shine and evolve.

            Your solar modeling work tries to reconcile theory and experiment, and in the course of this you have taken an interest in helioseismology, which aims to probe the solar interior by analyzing the oscillations of the sun's outer layers.

   Bahcall: I've written a series of papers in which the goal is to calculate as precisely as possible the conditions in the solar interior. We successfully refined these models so we now get detailed agreement with the helioseismological frequencies, maybe 10,000 of which are known very accurately. So we now know what makes the sun tick much better than we did early on.

            All this progress cannot conceal the fact that astrophysicists still refer to the "solar neutrino problem."

   Bahcall: Actually there are three solar neutrino problems. First, the classic one involving Ray's chlorine experiment, which has existed for two decades, in which there is a discrepancy between predicted and measured fluxes. Second, the water experiment at Kamiokande is apparently measuring the same thing as Ray—rare high-energy 8B solar neutrinos—but at different threshold energies, and they get different answers by nearly a factor of 2. The two experiments are ostensibly measuring the same process if the standard electroweak theory is correct. This "second" solar neutrino problem is independent of most of the uncertainties in astrophysics and nuclear physics. The third problem is that the gallium experiments are inconsistent with the robust predictions of the standard solar model for the flux of 7Be neutrinos. The dilemma is that either the chlorine or water experiment is wrong, and both of the gallium experiments are wrong, or we need new physics.

            This work is part of a long tradition in astrophysics whereby discoveries in the cosmos have informed physical theory. To what extent has the solar neutrino problem contributed to physics?

   Bahcall: Neither Ray nor I had a vision that looking at a beam of neutrinos from an object 1011m away would teach us new physics. The two most popular mechanisms for explaining the solar neutrino problem via new physics are vacuum neutrino oscillations and matter-enhanced neutrino oscillations. Vladimir Gribov and Bruno Pontecorvo suggested that some sort of schizophrenia between the three neutrino types—electron, muon, and tau—on the long trip from the sun might mean that they switched to mainly muon or tau types by the time they got to Earth. That theory attracted a minority of particle physicists for a time.
   For the matter-enhanced oscillations we need a natural extension to the simplest version of standard electroweak theory. According to this explanation, some electron neutrinos are transformed into muon or tau neutrinos as a result of their interaction with electrons in the sun. Non-zero neutrino masses are required for this effect: theory and all experiments are reconciled with an electron neutrino mass of about 0.003 eV.

            Is the end in sight for the classical solar neutrino problem?

   Bahcall: Four new solar neutrino experiments now under construction will soon test the proposition that new physics is needed. The Superkamiokande and the Sudbury Neutrino Observatory should be operational next year, and will have counting rates two orders of magnitude higher than the four pioneering experiments. Another experiment, being developed at CERN in Geneva, will look at the shape of the energy spectrum of 8B solar neutrinos, and this will tell us whether oscillations are taking place.
   My guess is that as a result of these experiments we will get directed in more specific ranges, but I don't think that it will be possible before the end of the century to say that there is a unique particle physics solution to solar neutrino problems. My hunch is that in the next five years or so it will be likely one of the several proposed particle physics solutions will emerge as the selected one. But there is sufficient richness in the imagination of our particle theorist friends that the number of particle physics solutions far exceeds the number of funded experiments! More than half of those scientists presently in the field incline to the matter-induced oscillations, but to focus on the solution with the rigor that is required will take more than the current generation of experiments but we might be lucky.

            A good example of astronomer's luck was the supernova explosion in the Large Magellanic Cloud in 1987, which led to the detection of neutrinos from beyond our galaxy. How do you see neutrino astronomy developing generally?

   Bahcall: One active area with a lot of experiments is the study of atmospheric neutrinos, which has puzzles of its own. There is no doubt that at the very high energies at which cosmic rays come in they are producing neutrino secondaries. That too has interesting physics and will be active in the next decade. Beyond that there is true neutrino astronomy where we're looking for neutrinos in our galaxy and beyond. There are new experiments at the South Pole, under the ocean in Hawaii, and planned for under the sea near Greece, where people will be taking the first steps towards detecting on a regular basis neutrinos from other astronomical systems. We hope this will teach us about astronomical systems that are very different from those we see with photons: neutrinos come from very different regions to photons and they don't have the same difficulties in escaping from stars. I think these experiments are very promising. The first generation of these experiments will be operating in the next two to three years. Although they might not detect neutrinos from outside the solar system, the next generation will have much larger versions of the current experiments and I am hopeful we will then have extragalactic neutrino astronomy.