Raymond Davis Jr. shared in a long-deserved Nobel Prize in Physics this year. He won it for a lovely experiment which detected subatomic particles coming from the Sun — and which also demonstrated how real-world science makes progress, step by step, via patient and meticulous work.

Davis began his big project in the 1960's. A decade later, at Caltech, I remember a colloquium that he gave for the physics department. He described the phenomena he hoped to observe, the details of his apparatus, and the puzzling results that it had already yielded. Then the fun began, as he and his audience began discussing what those measurements might mean and how they could, or couldn't, be reconciled with previously-known physics and astrophysics.

But begin at the beginning, in the center of the Sun. Every second the Sun radiates ~4*10³³ ergs of energy from its surface, a tiny fraction of which reaches the Earth and is responsible for essentially all life on this planet. (But ignore that for now.) If the Sun is to keep shining at an almost constant rate for billions of years, as it seems to have done, then it somehow needs to generate that same amount of energy. Chemical reactions can't do the job; nor can gravitational contraction. As far as any reasonable stellar models can determine, the only way to make a star like the Sun function is to have a hot, dense core where hydrogen atoms combine slowly to make helium and, along the way, liberate plenty of nuclear energy.

You can't, however, just stick protons (hydrogen nuclei) together and expect to get helium. A stable helium nucleus needs neutrons to bind it. And converting protons to neutrons is a tricky task, not something to be done quickly via strong nuclear forces. The various pathways that stars use to fuse hydrogen involve weak nuclear interactions, and those interactions give off weakly interacting particles called neutrinos. (Technical quibble: they're antineutrinos in this case, but ignore that for now.)

For electromagnetic radiation the center of the sun is quite opaque; a photon can only move a tiny distance before it gets absorbed or scattered, and so it takes millions of years for heat to diffuse to the solar surface. For a neutrino, on the other hand, the Sun is virtually transparent. So a detector that senses neutrinos could see straight to the core of the sun, and thereby measure temperatures and densities and otherwise-unknowable details of events taking place there.

Aye, there's a rub: if the Sun is transparent to a neutrino then so is the Earth, and so is an erstwhile neutrino observatory. The only ways known to detect solar neutrinos involve having lots of matter and lots of time and lots of sensitivity. You also need lots of shielding, to keep cosmic rays and other non-neutrino noise from swamping the signal you're trying to pick up.

So that's where Ray Davis and colleagues began. They knew roughly the number of neutrinos that should be emitted by the Sun. They also knew how energetic the neutrinos should be and how strongly (or rather, weakly) they should react with various types of nuclei. A few hundred tons of chlorine, for instance, would be enough to absorb a few neutrinos every month. The right kind of chlorine atom hit by a neutrino would turn into an argon atom, of a slightly radioactive isotope. And maybe, just maybe, since argon is an almost inert noble gas element, those few argon atoms could be flushed out of the mass of chlorine and counted.

Not easy ... but hey, if it were, then somebody would have done it long ago. Ray Davis was a careful but aggressive experimenter. He got permission from the owners of the Homestake gold mine in Lead, South Dakota, to put a boxcar-sized tank of perchlorethylene in a chamber almost a mile underground. Perchlorethylene is a dry-cleaning fluid that contains lots of chlorine atoms; working in a deep mine provides good protection against stray radiation from the skies. Once their equipment was up and running, every month or so Davis's team would bubble helium through the tank to extract dissolved atoms of argon. Then they would freeze those atoms out, put them into a sensitive radiation detector, and watch for them to decay.

The bottom line? Davis's experiment found solar neutrinos, but only about a third as many as "should" have been seen. Why? That's where, as I said some time ago, the fun began. There are several good possibilities, in three major categories:

• Stellar structure — the Sun isn't emitting as many neutrinos as expected
• Particle physics — something is happening to the neutrinos on the way to the Earth
• Experimental error — the detector isn't picking up the neutrinos the way it should be

Explanation #1 was conceivable but a bit tough to accept. The rates for nuclear reactions that should be going on in the solar core are pretty well measured in terrestrial labs. There's more uncertainty in the conditions at the center of the Sun, so if the temperature could be a few million degrees lower then that might account for the reduced neutrino emission. But then the composition of the core would have to be weird, radically different in heavy elements from what the spectrum of visible light emitted by the solar surface indicates the Sun is made of. Or alternatively, perhaps the Sun's center could temporarily be cooler than the standard models predict because the Sun is pulsating, and we happen to be living during a low-temperature era. But there's no good reason for that either, and strong historical/biological/geological evidence to the contrary.

Explanation #2, that neutrinos were getting lost in transit between the Sun and the Earth, seemed more plausible. Perhaps there are multiple varieties of neutrino, and the standard "electron" neutrinos are turning into undetectable muon neutrinos or other types? Maybe — but for that to happen neutrinos can't be massless, which almost everybody thought at the time was the case. Overall, a definite maybe.

Explanation #3, problems with the apparatus or the data processing, was always a possibility. But Ray Davis, in the best tradition of experimental physics, was a master at checking his work. Among many other tests Davis & Co. introduced tiny amounts of radioactive gases into their apparatus and measured what fraction of them were picked up. They also varied other parameters and confirmed that the equipment was functioning correctly. They compared measurements with detectors of non-solar neutrinos set up near nuclear power reactors. They watched for seasonal variations in the detected neutrino flux and correlated them with the changing distance between the Earth and the Sun, due to the ellipticity of the Earth's orbit. And Davis was always open to suggestions for improvements in his work, open to visiting inspectors, and open in describing the details of his set-up. Thus, error on the part of the experimenter didn't seem likely.

Nobody at that long-ago research conference in the Caltech physics building had a simple answer. So Davis just kept on running his detector, improving its accuracy, and building a solid set of data. In time other experiments came along and corroborated his results using different methods. One of the people who shared the Nobel Prize with Davis this year, Masatoshi Koshiba, set up huge tanks of ultrapure water deep underground, surrounded them with sensitive photon detectors, and picked up faint flashes of light from rare neutrino interactions. His measurements confirmed and extended Ray Davis's. And over the decades other experiments on high-energy particles have come to suggest strongly that Hypothesis #2 is the right one, and that neutrinos can transmute from one form to another during the 8-minute trip between the Sun and the Earth.

Ray Davis's labors are an example of Good Science. You need patience to do it; you also need to know what has gone before, so you can build upon what works and chip away at what doesn't. Above all, you need to be open-minded, and open to sharing new ideas with others.

(see NegativeResults (2 Nov 1999), QuestionAuthority (18 Jan 2000), SoftOutsideCrunchyCenter (1 May 2000), ScienceAndPseudoscience (6 Oct 2001), NoFinalAnswers (11 Mar 2002), HighPrecision (16 July 2002), ScientificRevolutions (16 Aug 2002), ...)

TopicScience - TopicPersonalHistory - 2002-10-13

(correlates: RayDavis, ProofsAndRefutations, AggressiveAggregation, ...)