Physics Nobel Prize Is Awarded to Giacconi, Davis, and Koshiba

This year’s Nobel Prize in Physics goes to three men who have created and peered through new windows on the cosmos. Half the prize is awarded to Riccardo Giacconi, director of Associated Universities Inc, in Washington, DC, “for pioneering contributions to astrophysics, which have led to the discovery of cosmic x-ray sources.” The other half is shared between Raymond Davis Jr, retired from Brookhaven National Laboratory, and Masatoshi Koshiba, retired from the University of Tokyo. They are cited “for pioneering contributions to astrophysics, in particular for the detection of cosmic neutrinos.”

In 1962 Giacconi discovered the first x-ray sources beyond the Solar System, and he headed the effort that led, in 1978, to the launch of the Einstein Observatory, the first x-ray telescope capable of imaging distant sources. A few years later, in the bowels of a South Dakota gold mine, Davis and a handful of collaborators built the first detector capable of measuring the Sun’s output of neutrinos. In the 1980s, Koshiba’s group in Japan built the first detector that could record the directions, energies, and arrival times of individual neutrinos from the Sun, and even from supernovae.

The x-ray sky

Born in Genoa in 1931, Giacconi earned his PhD in 1954 at the University of Milan, working on cosmic rays under Giuseppe Occhialini. “It took me two years to record just 80 protons,” says Giacconi. “So I swore I’d never do low-statistics work again. That’s what started me thinking about how to concentrate and focus the meager flux of x rays one could expect from sources beyond the Solar System.”

X rays cannot be focused by refraction, but they do exhibit total reflection off metal or glass at grazing angles. In 1960, Giacconi and Bruno Rossi published a paper on how one might concentrate x rays by reflecting them off paraboloid surfaces. ¹ Rossi was then board chairman of American Science and Engineering (AS&E), a private research firm near Boston, and Giacconi was its scientific director.

The Giacconi–Rossi design could concentrate x rays, but not focus them. “We soon learned, however, that Hans Wolter in Germany had already done most of the mathematical analysis needed for focusing,” recalls Giacconi. Wolter had calculated that a reflector combining both paraboloid and hyperboloid segments should do the trick. Giacconi tested this geometry with light reflected internally off the surfaces of a plastic mockup. It worked well at optical wavelengths. But for x rays, one would have to polish the reflecting surfaces to smoothness on a scale of angstroms.

X-ray detectors have to fly aboard satellites, rockets, or balloons, because the atmosphere is opaque to x rays. Not until 1973 did Giacconi and coworkers launch the first focusing x-ray telescope. But already in 1962, his group at AS&E provided the first glimpse of x-ray sources beyond the Solar System. The detector was simply three Geiger counters aboard a US Air Force rocket. And what this first look found was quite unexpected. ² It revealed an amazingly bright source, which was later localized to the constellation Scorpius and nicknamed Sco X-1. “We knew it was something truly novel,” says Giacconi, “because its ratio of x-ray to optical brightness was a billion times that of the Sun.” And the brief rocket flight yielded a second important surprise: an unresolved x-ray background, apparently from everywhere in the sky.

We now know that Sco X-1 was the first of many binary stellar x-ray sources to be discovered. The first x-ray orbiter, Uhuru, was launched by NASA in 1970. It was built by Giacconi’s group, which had moved from AS&E to the Harvard–Smithsonian Center for Astrophysics (CfA). Even without a telescope, Uhuru provided convincing evidence that intense stellar x-ray sources like Sco X-1 are powered by the gravitational infall energy of material being accreted from a star onto a compact binary companion—either a neutron star or a black hole.

The x-ray background has now been largely resolved into discrete sources—the active nuclei of very distant galaxies—by NASA’s orbiting Chandra x-ray observatory, launched in 1999 (see Physics Today May 2000, page 18 ). Even before Chandra, much of this background deciphering, especially at x-ray energies below 2 keV, had already been accomplished by the Einstein Observatory and by the German ROSAT orbiter, launched in 1990.

In 1973, five years before the launch of the Einstein Observatory, Giacconi’s group built the first real x-ray telescope. It flew aboard the Skylab orbiter, but it was used only to image the Sun’s x-ray output. X-ray study of the Sun had been pioneered in the late 1940s by Herbert Friedman at the Naval Research Lab, with captured German V-2 rockets.

The conceptual model for the gargantuan x-ray output of active galactic nuclei is provided, in miniature, by the stellar x-ray binaries. Chandra observations have made a compelling case that supermassive black holes at the centers of these young galaxies are continually accreting material, which radiates in x rays as it becomes very hot en route to its doom.

“The Chandra telescope is eight orders of magnitude more sensitive than the detector with which we discovered Sco X-1,” says Giacconi with pride. He led the campaign in the 1970s for its design and construction. When Giacconi left Harvard in 1981 to become director of the Space Telescope Science Institute in Baltimore, the leadership of the Chandra effort at CfA passed to Harvey Tananbaum.

Chandra’s four nested hyperboloid–paraboloid reflectors provide an unprecedented imaging angular resolution of better than 1 arcsecond (see the image above). With this resolution, Chandra can identify a very distant x-ray galaxy with as few as 10 photons recorded by its CCD detector array in a week-long deep exposure. Having found 350 x-ray galaxies in one such deep exposure of a patch of sky no bigger than a quarter moon, Giacconi and coworkers concluded in 2001 that the entire sky offers 10⁸ x-ray galaxies accessible to Chandra. ³ High angular resolution is also essential for identifying the optical and radio counterparts of x-ray sources in crowded fields.

“An important discovery of x-ray astronomy is the intergalactic gas that pervades clusters of galaxies,” says Giacconi. “It’s too hot to be seen at optical wavelengths.” This hot gas accounts for much more of a cluster’s mass than do the all stars in the galaxies. And its distribution serves as a tracer for the even greater nonbaryonic mass—“dark” at all wavelengths—that is presumed to dominate the gravitational binding of galaxy clusters.

The neutrino shortage

Davis was born in 1914 in Washington, DC. He received his PhD in chemistry at Yale University in 1942. Davis’s interest in the then still hypothetical neutrino began when he joined the new Brookhaven National Laboratory in 1948. Two years earlier, Bruno Pontecorvo had suggested that one might prove the existence of the neutrino (v) by looking for the production of radioactive argon-37 in chlorine by the reaction $$v{+}^{37}\text{Cl}{\to }^{37}\text{Ar}+\text{e}.$$ (1) In 1954, Davis mounted such a search with 1000 gallons of carbon tetrachloride in the basement of a nuclear reactor. He found no ³⁷Ar, which is what we would expect in hindsight, knowing that reactors produce only antineutrinos $\left(\overline{v}\right)$ . This null result can be regarded as the first demonstration that the v is different from its antiparticle.

So Davis began thinking about using reaction 1 to detect neutrinos from the Sun. The problem was, however, that the maximum neutrino energy expected from proton–proton fusion, the Sun’s principal energy source, was only 0.42 MeV, hopelessly below the 0.8-MeV threshold for reaction 1. But in 1958, astrophysicist William Fowler came to the rescue by calling Davis’s attention to a new Naval Research Laboratory measurement of the helium fusion reaction ³He + ⁴He → ⁷Be + γ. The cross section for this minor branch of the solar pp cycle turned out to be a thousand times bigger than previously thought. Therefore, Fowler pointed out, the Sun might be producing enough boron-8, through proton capture by beryllium-7, for a chlorine detector to see the high-energy neutrinos (up to 15 MeV) emitted by ⁸B decay.

The next issue was the competition, in the Sun, between two rival ⁷Be capture processes: the desired proton capture, which produces ⁸B and detectable high-energy neutrinos, and the unwanted electron capture, which results in neutrinos of lower energy. The need to evaluate this crucial competition initiated the four-decade-long close collaboration between Davis and theorist John Bahcall, now at the Institute for Advanced Study in Princeton.

In 1962, Bahcall had published an analysis of ⁷Be electron capture at high temperature. So Davis asked him to calculate the abundance of ⁸B neutrinos one could expect from the Sun. “To give an answer, I had to know the temperature profile of the solar core,” recalls Bahcall. So he availed himself of the stellar-evolution computer codes being developed by Fowler’s group at Caltech. “We concluded,” says Bahcall, “that the flux of ⁸B neutrinos would be far too meager to be measured by the largest practicable chlorine detector.

A year later, however, the enterprise was back in business. To estimate the expected solar-neutrino signal, one had to calculate the rates of nuclear processes not only in the Sun, but also in the detector. And Bahcall realized that he had badly underestimated the rate for reaction 1 by considering only the ground-state transition. He now predicted a superallowed transition to an excited “analog” state of ³⁷Ar that would increase the solar-neutrino capture in a 600-ton detector to an acceptable rate of a few events per day.

Soon after that, Brookhaven direc-the first time, to measure directly the total ⁸B neutrino flux from the solar core, regardless of any metamorphosis along the way. The Sudbury Neutrino Observatory in Canada, with heavy water (D₂O) at its heart, can detect a neutrino of any flavor and even, to some extent, distinguish between neutrino flavors (see Physics Today, July 2002, page 13 ). The result has been a triumph for the SSM and a major step on the road to a detailed theory of the very rich neutrino sector of particle physics.