In the summer of 1965 workers deep within the Homestake Gold Mine, Lead, South Dakota, completed the excavation of a 30 × 60 × 32 ft3 cavern. This excavation, nearly a mile underground, was the first step in an experiment proposed by Ray Davis Jr. and his Brookhaven National Laboratory (BNL) collaborators. The cavern was soon filled by a large tank containing 610 tons—equivalent in volume to ten railway tankers—of the chlorine-bearing cleaning fluid perchloroethylene. The purpose of this detector was to record, for the first time, the neutrinos produced as a by-product of the thermonuclear reactions occurring in the Sun's core. The results of the Davis experiment presented the physics and astrophysics communities with a puzzle that is only now being resolved.
The Standard Solar Model
It is known that the Sun has burned for about 4.6 billion years, sustaining itself against the crushing effects of its own gravity. The energy required to maintain the pressure of solar gases is produced by thermonuclear reactions. Four protons are converted into a helium nucleus (which contains two protons and two neutrons) plus two electrons and two electron neutrinos: with a net release of energy. The series of reactions by which almost all solar helium synthesis occurs is called the pp chain (see Figure 1). Roughly half the hydrogen fuel that was initially in the Sun's core has been converted into helium over the past 4.6 billion years.
The photons that make up ordinary solar radiation scatter repeatedly within the Sun, and take millions of years to diffuse outward from the solar core. Thus, the sunlight that arrives on Earth last scattered near the Sun's outer surface, called the photosphere. In contrast, neutrinos pass through matter almost unaffected—they lack the electromagnetic interactions by which photons scatter, instead having only "weak" interactions. Consequently, the Sun is transparent to neutrinos, which arrive at Earth directly from the core. As they carry, in their number and energy distribution, detailed information about the nuclear reactions by which they were produced, neutrinos allow one to "see" directly into the Sun's center.
Originally, the motivation for solar neutrino measurements was to test the standard theory of
stellar energy generation and evolution, as applied to the Sun. The Standard Solar Model (SSM) postulates that the Sun burns in hydrostatic equilibrium, with the gravitational force balanced at each point within the star by the gas-pressure gradient. Energy is generated by hydrogen burning and transported by radiation (in the Sun's interior) and by convection (outer envelope). The initial composition of the Sun, by mass roughly 75 percent hydrogen and 25 percent helium, with traces (~2%) of heavier elements, is chosen so that today's luminosity is reproduced after 4.6 billion years of evolution.
Three cycles (I, II, and III) make up the pp chain, with each producing a distinctive neutrino spectrum (see Figure 1). The relative importance of these three cycles depends critically on the Sun's central temperature. The Davis experiment was sensitive primarily to the high-energy 8B neutrinos produced in the pp III cycle. As the flux of these neutrinos varies as where Tc is the core temperature, Davis hoped to measure that temperature with an accuracy of a few percent.
The flux of neutrinos at Earth is enormous, with about 65 billion neutrinos passing through each square centimeter each second. Nevertheless, because matter is so transparent to neutrinos, detecting them requires heroic efforts. The clever idea behind the Davis detector was to exploit the reaction to measure the 8B solar neutrinos that, though only 0.01 percent of the total flux, interact more readily because of their higher average energy. Because argon is a noble gas, the few atoms of radioactive 37Ar produced in the Davis detector (about one every 2 days) could be flushed from the large volume of perchloroethylene and counted by observing their sub-sequent decays. As the half life of 37Ar is 35 days, the Davis tank was flushed about once every 2 months. The exotic site, a mile underground, provided a thick rock shield to screen out cosmic rays, which also trigger production of 37Ar. Davis found about one-third the predicted number of 8B neutrinos, a result that many scientists initially attributed to the Sun's core being somewhat cooler than expected.
The mystery deepened some years later with data from new experiments, sensitive to different combinations of the neutrinos from the three pp cycles. An experiment to measure solar neutrino reactions event by event was performed in a detector mounted in the Kamioka mine in Japan. This detector, originally constructed to search for proton decay, consisted of 4,500 tons of ultrapure water, surrounded by a large array of phototubes. Solar neutrinos scatter off electrons in the water, which then emit a cone of Cerenkov radiation that the phototubes record. Two experiments similar to that done by Davis, but using gallium instead of chlorine, were performed in Russia and Italy. Gallium was chosen because the lowest-energy solar neutrinos, produced in the initial p + p reaction of Figure 1, can change 71Ga into 71Ge.
Together, the chlorine, gallium, and Kamioka experiments determined the principal neutrino fluxes produced by the pp chain. Remarkably, the pattern was not compatible with simple adjustments of the SSM, such as a cool solar core: something more interesting was happening.
It had been recognized for many years that the lack of solar neutrinos might have nothing to do with deficiencies in the SSM but might instead reflect a lack of understanding of the properties of neutrinos. In particular, if neutrinos have a small mass—a possibility not envisioned in the current Standard Model of particle physics—a natural explanation could be offered for the observations. Electron neutrinos produced by the nuclear reactions in stars can then transform (or oscillate) into neutrinos of a different flavor, thereby escaping detection on Earth. (The other flavors, muon and tauon neutrinos, are not recorded by the chlorine and gallium detectors and have a probability for interacting in water that is only 15% that of electron neutrinos.) Because solar neutrinos are low in energy and travel a great distance before they are detected on Earth, solar neutrino oscillations can arise for neutrino masses much smaller(e.g., 10-6 eV) than those detectable by any other means. Furthermore, it was shown in 1985 that as neutrinos make their way from the Sun's core to its surface, the probability of oscillation can be greatly enhanced. This phenomenon, known as the MSW or Mikheyev-Smirnov-Wolfenstein effect, can distort the spectrum of solar neutrinos in distinctive ways.
Super-Kamiokande and SNO
The possibility of discovering massive neutrinos stimulated new efforts to measure solar neutrinos. In Japan a much more massive successor to the Kamioka experiment, the 50,000-ton water Cerenkov detector Super-Kamiokande, was built by a collaboration of Japanese and American physicists. This experiment not only sharpened the case for solar neutrino oscillations but also provided direct evidence that oscillations alter another source of neutrinos, those produced by cosmic ray interactions in the atmosphere.
A second detector, the Sudbury Neutrino Observatory (SNO), is similar in design, except that the water in SNO's central vessel is "heavy," with the hydrogen replaced by deuterium. The SNO detector, which was built by physicists from the United States, the United Kingdom, and Canada, is located two kilometers underground, within the Creighton nickel mine in Sudbury, Ontario, Canada. The acrylic vessel containing 1,000 tons of heavy water is surrounded by a shield of 7,000 tons of ordinary water, with events in the entire volume viewed by 9,500 photomultiplier tubes. The great depth all but eliminates cosmic ray backgrounds. In addition, the detector was constructed with extraordinarily pure materials to reduce backgrounds from natural radioactivity: "clean room" conditions had to be maintained in the mine, as the introduction of even a thimbleful of dust in the 10-story-high detector cavity would cause the experiment to fail.
The heavy water allows the experimentalists to measure, in addition to the elastic scattering (ES) of neutrinos off electrons, the following charge-current (CC) and neutral-current (NC) reactions off deuterium: Only electron neutrinos can induce the first (CC) reaction, whereas neutrinos of any type can stimulate the second (NC). The SNO results, announced in April 2002, are shown in Figure 2. Indeed, two-thirds of the solar neutrinos arrive at Earth as muon or tauon neutrinos, not the expected electron neutrinos. Solar neutrinos do oscillate, and neutrinos do have mass.
The implications of the Super-Kamiokande and SNO experiments are startling. The Big Bang filled the universe with a sea of neutrinos: it is now known that the total mass in these neutrinos is at least equal to that in all the visible stars. There is great hope that the pattern of neutrino masses and oscillations emerging from experiments will provide theorists with the clues they need to construct a new Standard Model of particle physics to replace one that is now known to be incomplete.
Bahcall, J. N. Neutrino Astrophysics (Cambridge University Press, Cambridge, UK, 1989).
Super-Kamiokande Collaboration. "Measurements of the Solar Neutrino Flux from Super-Kamiokande's First 300 Days." Physical Review Letters81 , 1158–61 (1998).
SNO Collaboration. "Direct Evidence for Neutrino Flavor Transformation from Neutral-Current Interactions in the Sudbury Neutrino Observatory." Physical Review Letters89 (1), 011301-1-6 (2002).
Wick C. Haxton