Copyright New York Academy of Sciences Jan 1996THE YEAR WAS 1952. ABOUT ONCE a month Raymond Davis Jr. would load a fifty-five-gallon drum onto a hand truck and out to a manicured lawn less than fifteen feet from the graphite nuclear reactor at the Brookhaven National Laboratory on Long Island, New York. Davis would deposit the tank--filled with carbon tetrachloride (CCl sub 4 ), an ordinary but toxic cleaning fluid--on the lawn and walk away. There it would rest undisturbed, until Davis returned a few weeks later. He would then reload the drum onto his hand truck, take it back to the chemistry building and run a set of complicated tests on the fluid.
Now, forty-three years later, Ray Davis takes me for a tour of a successor to his fifty-five-gallon drum. I meet him, a soft-spoken, grandfatherly man wearing a hard hat and a miner's lamp, at a vast working gold mine known as Homestake in Lead, South Dakota. To reach the full-scale experiment you board a mine elevator called the cage, a small, five-by-ten-foot space running on vertical wooden railings and hauled up and down a 4,850-foot shaft with steel cables. The cage shakes and rattles vigorously as it drops deeper than three times the height of the Empire State Building, taking at least three minutes to reach the 4,850-foot level.
Once you exit the elevator you walk about a hundred yards along a mine passageway, with train tracks cut into its rock floor. The faint twinkle of train headlights in the distance makes the length of the passage seem unmeasurable, and the soft, deep rumble of ore being loaded into train cars can always be heard in the background. A right turn at the first intersection quickly takes you beyond the air-conditioning vents, and the temperature rises above eighty degrees Fahrenheit. There in the passage is what seems to be a small chemical laboratory, old personal computers lining one wall, arrays of glass test tubes lining another.
To the left is a side passage, angling steeply downward. At its bottom, about a mile below the surface, is the tank chamber, a room the size of a two-story house, filled with a gray tank nearly forty-eight feet long and more than thirty feet in diameter, which holds 100,000 gallons of the dry-cleaning fluid tetrachloroethylene (C sub 2 Cl sub 4 ). Puddles of muddy, rust-colored water lap at the base of the tank. White and red crystal formations have grown along the edges of the tank in the three decades since its construction. Every other month Ray Davis and his associates come down here to squeeze a minute quantity of argon-37, fifteen atoms on average, out of the tank.
IN A SENSE, LITTLE HAS CHANGED FOR Davis since 1952. His quest, now as then, is to capture the neutrino, at once one of the most plentiful elementary particles in the universe and one of the most fleeting. Neutrinos were created in copious quantities in the big bang, and they can be generated today in nuclear reactions and atomic bombs. They have also played a fundamental role in helping particle physicists determine the nature of quarks and the so-called weak force. Moving at or near the speed of light, the minute particles bear no electric charge, and they are not subject to the so-called strong nuclear force, which binds protons, neutrons and the quark constituents of those particles in the atomic nucleus.
In fact, alone among elementary particles, neutrinos "feel" only the weak force, which makes them interact, literally, only weakly with other matter. To appreciate just how weakly, according to calculations made in 1934 by the physicist Hans A. Bethe, now of Cornell University, and the late English theoretical physicist Rudolf E. Peierls, a column of water 1,000 light-years thick would be needed to capture most of the neutrinos generated in the so-called beta decay of radioactive nuclei. Ten thousand billion neutrinos from the sun pass unnoticed through your body every second of the day, and most of them speed on through the rest of the mass of the earth as if they were, as John Updike once put it, "like dustmaids down a drafty hall/Or photons through a sheet of glass."
THUS DETECTING NEUTRINOS AND VERIFYING their existence in the first place was an achievement of the fist rank. On my descent into the mine with Davis last October we both were keenly aware that only days before, the Royal Swedish Academy of Stockholm had announced the winners of the 1995 Nobel Prize in physics, and sharing that prize was the first man to detect the neutrino. It was not Davis. But if Davis felt any pangs of envy, he did not show it; the race he ran four decades ago against Frederick Reines of the University of California at Irvine and the late Clyde L. Cowan Jr. had long since been lost, and it is Reines who now basks in (much deserved) glory.
But Davis, too, has been able to bask, in a fantastic kind of sunlight that is virtually unshaded by the mile of South Dakota stone over our heads. For nearly thirty years he has come down here regularly to monitor his tank for evidence of neutrinos from the sun. Indeed, Davis's huge bottle of cleaning fluid buried deep in the Homestake Gold Mine is a new kind of telescope, a neutrino telescope, capable of "seeing" deep into the sun's interior and, in principle, recording information that would affect theories of cosmological phenomena from supernovas to the big bang. Those data have proved to be surprising, frustrating and immensely invigorating, challenging physicists and astronomers in directions unplanned and unexpected. Davis's data have thrown the theory of solar burning into a decades-long crisis that is still not resolved. And his data are consistent with the finding, reported late last year, that the neutrino may possess a minute mass, not more than one hundred thousandth the mass of the electron. If so, the so-called standard model of elementary particles, which predicts a neutrino mass of zero, has some explaining to do. And if neutrinos have a nonzero mass, they are so plentiful that the mass of all the neutrino "dark matter" in the universe may be at least double the mass of all visible matter, substantially shaping the evolution of the universe and, perhaps, one day helping end its expansion.
THE EXISTENCE OF THE NEUTRINO WAS first postulated in 1930, in a letter written by the Austrian-born American physicist Wolfgang Pauli to the Austrian-born Swedish physicist Lise Meitner and other physicists assembled for a conference in Tuebingen, Germany. In his letter Pauli proposed what was then a merely hypothetical particle to make up for what was believed to be the nonconservation of energy, momentum and "spin," or intrinsic angular momentum, of atomic nuclei in beta decay. But even Pauli was unhappy with that kind of solution; he regarded it as a "desperate way out" of the theoretical difficulties.
One measure of Pauli's dissatisfaction was the seeming impossibility of finding experimental evidence for his newly postulated particle. Even in hindsight today, it strains credulity that any scientist believed it possible to record evidence of the neutrino, given its astonishingly ephemeral nature. But in the 1940s, when the first nuclear reactors were built, physicists realized they had a potential source of neutrinos many orders of magnitude more intense than naturally radioactive sources. That realization is what brought Davis to Brookhaven in 1948. According to the theory of the day, the Brookhaven reactor was generating vast numbers of neutrinos, and Davis had a simple strategy for detecting them: If enough neutrinos of a certain kind were to pass through the chlorine in the cleaning fluid, some of them would collide with some of the chlorine atoms and convert the chlorine to atoms of argon-37. (It takes 10 sup 42 neutrinos to ensure a reasonable chance of a collision with a chlorine atom.) Extracting and measuring the argon-37 atoms would prove the existence of neutrinos and give a measure of their flux.
Davis's first attempt to measure neutrinos by chlorine decay, however, was crude at best. "Not surprisingly, I got a null result," he remembers. For one thing, neutrinos are not the only particles that can transmute chlorine-37 into argon-37. The subatomic debris generated by the cosmic rays bombarding the upper atmosphere from space swamped any neutrino reactions his tank of cleaning fluid might have recorded. Moreover, as Davis had strongly suspected, the fifty-five-gallon drum was simply not large enough. Too few neutrinos would collide with too few chlorine atoms, giving rise to too few argon atoms for Davis to measure.
IN 1954 DAVIS TRIED SOMETHING BIGGER. He enlarged the tank to 1,000 gallons and had it buried about twenty feet down in the same grassy lawn at Brookhaven, this time isolating the tank from all cosmic radiation except for highly penetrating muons. Davis had the staff glass blowers at Brookhaven build him a portable chemistry set that he could wheel to the tank. "I set up my family camping tent around my equipment and did my tests there, right above the tank." Yet even the 1,000-gallon tank wasn't large enough to register any argon-37 transmuted by neutrinos from the sun. In a paper published in 1955 Davis explained that although he was now able to extract and measure as few as seventy argon atoms a day from the tank, he had found none. Solar neutrinos, if they existed at all, had to be generating argon atoms at an even lower rate. Because the highest predicted value for the flux of solar neutrinos would produce no more than a twentieth of an argon atom a day in a 1,000-gallon tank, one critic pooh-poohed Davis's conclusions: "One would not write a scientific paper describing an experiment in which an experimenter stood on a mountain and reached for the moon, and concluded that the moon was more than eight feet from the top of the mountain."
Davis was not disheartened. The race was still on to detect neutrinos from any source, not just from the sun. In 1956 he had two 500-gallon tanks installed at the Savannah River plant in Georgia. Because the tanks were placed beside a more powerful reactor as well as shielded from most cosmic-ray particles by the reactor's concrete building, Davis hoped that now at least a few neutrinos from the reactor would smash against his chlorine and convert it to a measurable amount of argon.
Ironically, at the same moment that Davis was in Savannah, Reines and Cowan were there running their own neutrino experiment. Instead of carbon tetrachloride, their tank was filled with dissolved cadmium chloride and water. According to theory, a small number of neutrinos would interact with a small number of protons in the water, splitting each proton into a positron and a neutron. The positron would almost immediately collide with an electron, and the two particles would disintegrate into two gamma rays, leaving the scene of the collision in opposite directions. The neutron, however, would persist, on average, about five microseconds before it would be captured by the cadmium dissolved in the water. The energy liberated by the capture would appear as further gamma rays. If Reines and Cowan could record the emission of gamma rays with the distinctive five-microsecond delay, they would prove the existence of neutrinos.
Even as the experimenters carried on their friendly competition at Savannah, theorists were rewriting the rules. In 1956 Tsung-Dao Lee of Columbia University in New York and Chen Ning "Frank" Yang of the Institute for Advanced Study in Princeton, New Jersey, proposed--and Chien-Shiung Wu of Columbia University proved--that some subatomic nuclear reactions could be left-handed or right-handed, not symmetrical as had been thought. That work predicted that the Savannah nuclear reactor would give rise only to antineutrinos, the antiparticles to neutrinos. Antineutrinos cannot cause chlorine decay, and so Davis's two 500-gallon tanks had nothing to measure from the reactor. Again he had a null result.
Reines and Cowan, however, were luckier. Their experiment could and did measure the effects of antineutrinos. More than four decades later that experiment led to Reines's Nobel Prize.
UNFAZED BY REINES AND COWAN'S success, Davis had his 1,000-gallon tank moved to a limestone mine in Ohio, where he had it placed more than 2,300 feet below the surface. Now the cleaning fluid was tetrachloroethylene, less toxic than carbon tetrachloride but just as cheap. The tank was shielded from all but the rarest high-energy muons, and the chlorine would therefore react almost exclusively with neutrinos from the sun.
Physicists had long theorized that the sun's basic nuclear processes could be divided into two reactions, the proton-proton chain (pp) and the carbon-nitrogen-oxygen cycle (CNO). Each would generate neutrinos in huge numbers but at differing energies. Until 1958 it was assumed that only the CNO cycle would create neutrinos with an energy high enough to react with chlorine and create argon-37. But because the CNO cycle gives rise to less than 2 percent of the sun's energy, Davis could not build a tank large enough to measure those neutrinos. Again the 1,000-gallon tank produced no argon-37.
In 1958, however, Harry D. Holmgren and R. L. Johnston of the Naval Research Laboratory in Washington, D.C., discovered that the pp chain, producing almost 98 percent of the sun's energy is more complicated than anyone had thought. Among the nuclear reactions in the chain were some, specifically involving beryllium-7 and boron-8, that would give rise to neutrinos energetic enough to react with chlorine. If those reactions were added to the known CNO cycle, a chlorine tank substantially larger than 1,000 gallons but still of practical size would be able to detect the neutrinos from the sun.
Davis was now confident that for $600,000 he could build the world's first "neutrino telescope," one that would surely see the neutrinos pouring out of the sun's core. First he would expand the size of the tank a hundredfold, from 1,000 to 100,000 gallons. Then he would bury the telescope even deeper underground, to further shield it from cosmic rays, reducing the unwanted high-energy muon collisions to approximately one every three weeks.
Many astrophysicists were enthusiastic about the possibility. William A. Fowler of the Kellogg Radiation Laboratory at the California Institute of Technology, one of the physicists who had worked out the details of the W chain and the CNO cycle, threw his complete support behind Davis, helping him get the necessary funds. If the experiment worked, Fowler said, people could for the first time "see" inside the core of a star.
THE ASTROPHYSICIST JOHN N. BAHCALL, then at Caltech, was also keenly interested in the results of Davis's experiment, having dedicated several years of his life to calculating the number of neutrinos generated by the sun in its internal nuclear reactions. According to Bahcall's well-respected estimates, Davis would measure between two and six atoms a day of argon-37 out of a tank containing approximately 2.2 X 10 sup 30 atoms of chlorine. If that prediction was borne out, Bahcall's theory of the nuclear burning processes in the sun would be confirmed.
Understandably, there was skepticism. As Reines noted in 1960, "the probability of a negative result even with detectors of thousands or possibly hundreds of thousands of gallons of CCl sub 4 tends to dissuade experimentalists from making the attempt." Even if Davis's more optimistic prediction--between four and eleven argon-37 atoms a day--turned out to be right, the tank would still capture an ungodly small number of neutrinos.
In the end, however, the project was approved, partly on the basis of Davis's reputation as a skilled and reliable experimenter, partly because of the support of men such as Fowler and Bahcall. With funding from the U.S. Atomic Energy Commission, the world's first neutrino telescope was built between 1965 and 1967 in the Homestake Gold Mine.
AT HOMESTAKE, DAVIS FINALLY struck gold. Argon-37 began to trickle out of the cavernous underground tank, albeit too slowly to fulfill Bahcall's predictions. In its nearly thirty years of operation, Homestake has routinely measured an average rate of approximately one-half an argon-37 atom a day, well below the predicted number. Even now, after refining their solar models significantly, astrophysicists estimate that Davis's tank should measure no fewer than 1.5 argon-37 atoms a day, three times what the tank has found. The Homestake data have also withstood the healthy skepticism of science, having now been confirmed by three other neutrino telescopes, all of them much larger and more sophisticated than Davis's.
The solar neutrino problem has become one of the most celebrated anomalies in all of physics, and it has caused the solar theorists no little inconvenience. If the models of the nuclear reactions of the sun are wrong, almost all cosmological theory for the past thirty years could be misguided. From the predicted age of the universe to the life and death of individual stars, astrophysicists and cosmologists might find their hard-won hypotheses to be as fleeting as the neutrino itself.
Davis's finding that the flux of neutrinos from the sun is lower than expected has also cast doubt on the assumption, made, as I noted earlier, by the standard model, that the neutrino has no mass. But his low results could be accounted for if neutrinos oscillate among the three neutrino species--the so-called electron, muon and tau neutrinos. Such an oscillation would imply that at least one of the species has a nonzero mass. But because his detector would not pick up all three species, Davis's apparent low detection rate could be caused by just such an oscillation. A group of physicists working at the Los Alamos National Laboratory in New Mexico in October published results confirming neutrino oscillations and postulating a neutrino of extremely low, but nonzero, mass. Still, the neutrino-oscillation theory remains controversial: the latest calculated oscillation frequency appears too large to explain the solar neutrino deficit, leaving many physicists unsatisfied.
AT THE CENTER OF THIS STORY STANDS Ray Davis. His simple and elegant experiments have paid off in advances to human knowledge; they have led to new, more complicated neutrino detectors and inspired what seems like a new solar model every week; and they continue to provide answers to some of the most stubborn questions in science. For example, with the detection of neutrinos from the 1987 supernova, astrophysicists have come to speculate that neutrino "telescopes" may reveal the source of mysterious ultra-high-energy cosmic rays.
Even today, almost ten years after the experiment was to have been shut down, Davis is keeping it going. "We have discovered in the last decade that the neutrino count seems to correlate with the sun's sunspot cycle, something that had been considered impossible by the theorists." The engine that churns out neutrinos lies deep within the core of the sun. Physicists had thought that, since such internal reactions take millions of years to migrate to the sun's surface, the flares and sunspots that erupt there would bear no relation to the processes that fuel the sun's engine. The results of Davis's experiments directly contradict that view. "Since we don't yet know if this is merely a statistical anomaly or an actual finding, it seems logical to continue the experiment and see."
A dramatic example of that kind of anomaly took place in June 1991. Davis happened to be at Homestake and had just finished a routine extraction of argon-37 when the sun emitted one of the strongest solar flares in recorded history. The radiation swept across the entire solar system, registering on sensors in the Pioneer 11, Voyager 1 and Voyager 2 spacecrafts, drifting beyond the orbit of Saturn. Davis did a second extraction, and he was astonished to remove approximately twenty atoms of argon-37. That is equivalent to a neutrino capture rate of 3.2 argon-37 atoms a day--six times the usual rate.
THE SIMPLICITY OF DAVIS'S WORK IS astonishing, especially compared with the modern tendency to turn every scientific experiment, from the Hubble telescope to the CERN reactor, into a large-scale corporate endeavor. For example, the second generation of neutrino detectors use gallium in their tanks and depend on chemistry no more complicated than that used at Homestake. As such, they could conceivably be run with as small a staff.
Yet the two gallium experiments, GALLEX in Italy and SAGE in Russia, seem more akin to government bureaucracies than to chemistry experiments. In outlining their first results, the GALLEX experimenters listed fifty-four coauthors from ten research facilities. SAGE was only slightly better, listing twenty-eight coauthors from five institutions. The presence of so many strong-minded individuals can only make a project more complicated and difficult to execute, increasing the time and money necessary for completing the work.
As the federal government continues to tighten its budgetary belt in the coming years, scientists in all fields as well as government officials would be well advised to look closely at the story of Ray Davis and his tank of dry-cleaning fluid. They might find much more to learn than the strange fact that neutrinos may have mass.
Robert Zimmerman is a writer, filmmaker and speleologist living in New York City. His work appears frequently in THE SCIENCES.
My Research
