Virtually everything we know about the Sun has been gleaned from the light it emits. Images collected at various wavelengths provide clues to its composition, magnetic field dynamics, subsurface flows, and more. (See the articles in Physics Today by Eugene Parker, June 2000, page 26, and by Gordon Holman, April 2012, page 56.)
To glimpse directly into the Sun’s opaque core, however, one needs to look not for photons but for neutrinos. Both are products of the fusion of protons into helium-4, the multistep process that powers our parent star. But photons scatter in the Sun’s core for tens of thousands of years before escaping. By the time we see them, they retain little history of their origins. Because neutrinos interact weakly with matter, they escape almost immediately. From their flux and energy, one can deduce rates of reactions occurring in the core.
The most energetic neutrinos emanating from the solar core are born of boron-8 decays. With energies of up to 14 MeV, those neutrinos can be identified by the Cherenkov radiation that’s emitted as they scatter off electrons in a water tank or by the inverse beta decays they instigate in radiochemical detectors. (See Physics Today, August 2001, page 13.) All told, however, 8B neutrinos make up just a fraction of a percent of all solar neutrinos. Detecting the rest has proved a formidable task; the lower a neutrino’s energy, the more easily it is drowned out by background noise from naturally occurring radioactivity.
For the past two decades, researchers at the Borexino detector at Italy’s Gran Sasso National Laboratory have been meticulously engineering a low- background environment in which low-energy breeds of solar neutrinos might be detected. In recent years they’ve detected the roughly 1-MeV neutrinos generated during the decay of beryllium-7 and during rare proton-electron-proton (pep) reactions.1 Now they’ve measured the spectral flux of the Sun’s least energetic neutrinos:2 so-called pp neutrinos, which are emitted when two protons combine to form deuterium and have an upper energy limit of around 400 keV. The measurements fill in one of the last and by far biggest pieces of the solar-neutrino puzzle—pp neutrinos account for about 90% of the Sun’s total neutrino flux.
Noise reduction
About 99% of the fusion in the solar core occurs by way of the proton–proton chain of reactions; the rest occurs via the CNO chain, a secondary pathway mediated by carbon, nitrogen, and oxygen. More than 99% of the time, the proton–proton chain is instigated by the reaction that begets pp neutrinos. In essence, that reaction sets the rate of solar energy generation.
Technically, pp neutrinos have been detected before. In the 1990s, two groups—one working at the SAGE detector in the Russian Caucasus and another at the Gallex detector, Borexino’s predecessor at Gran Sasso—detected the elusive neutrinos in vats of liquid gallium. In those experiments, incoming neutrinos having energy greater than about 230 keV can signal their presence by converting gallium atoms to germanium ones; the flux is then inferred by tallying the Ge atoms that accumulate during an experimental run. (See the article by John Bahcall, Frank Calaprice, Arthur McDonald, and Yoji Totsuka, Physics Today, July 1996, page 30.)
“But the gallium measurements are like long-exposure, black-and-white photographs,” explains Josh Klein (University of Pennsylvania), who wasn’t affiliated with the Borexino study. “They don’t provide spectral information.” And without that information, one can’t readily distinguish pp neutrinos from their more energetic cousins.
By contrast, the Borexino group detects neutrinos in real time using the scintillator detector pictured in figure 1. An incoming neutrino that scatters off one of the scintillator’s electrons is heralded by a burst of light, which is detected by photomultiplier tubes. The brightness of the burst gives the neutrino’s energy.
Figure 1. At the Borexino detector, pictured here, solar neutrinos are detected in a 14-m-tall sphere filled with 278 tons of liquid scintillator. Incident neutrinos are heralded by bursts of light emitted as they scatter off the scintillator’s electrons. (Photo courtesy of the Borexino collaboration.)
Figure 1. At the Borexino detector, pictured here, solar neutrinos are detected in a 14-m-tall sphere filled with 278 tons of liquid scintillator. Incident neutrinos are heralded by bursts of light emitted as they scatter off the scintillator’s electrons. (Photo courtesy of the Borexino collaboration.)
The researchers used a suite of measures to protect their detector from ambient radioactivity: The underground detector is shielded from cosmic rays by more than 1 km of overlying rock; the scintillator itself is immersed in a water buffer, to protect it from radioactive elements in the surrounding metal tank and equipment.
At the low energies of pp neutrinos, however, even trace amounts of radionuclides in the liquid scintillator itself can be ruinous. To sufficiently purify the oil-based scintillator, the group used distillation, extraction, and filtration methods borrowed from petroleum engineering.3 “We essentially built a plant that could take fluid out of the detector and recirculate it, and we kept purifying it as we went,” says Frank Calaprice (Princeton University), of the Borexino collaboration. “We’re still pushing the background down.”
Between January 2012 and May 2013, Borexino registered more than 300 000 neutrino scattering events in the 150–600 keV range, as shown in figure 2. A significant share was attributable to naturally occurring radionuclides, but those background contributions could be tightly constrained using statistical models. Likewise, contributions from low-energy 7Be and pep neutrinos could be inferred from those neutrinos’ spectral line shapes, determined in the previous measurements at higher energies. Only the contribution from CNO neutrinos had to be assumed theoretically. But because that flux is relatively small and spectrally flat, the assumption doesn’t introduce much systematic error.
Figure 2. The signal and the noise. The spectral flux of pp solar neutrinos (pp ν), byproducts of proton–proton fusion in the Sun’s core, is obtained from the total neutrino flux (black crosses) by subtracting background contributions from naturally occurring radionuclides (lead-214, polonium-210, carbon-14, bismuth-210, and krypton-85), other solar neutrino breeds (from beryllium-7 decay, proton-electron-proton fusion, and the carbon-nitrogen-oxygen cycle), and pile-up—instances where two successive events, usually from 14C, are detected as a single count. The black curve is a fit to the data. (Adapted from ref. 2.)
Figure 2. The signal and the noise. The spectral flux of pp solar neutrinos (pp ν), byproducts of proton–proton fusion in the Sun’s core, is obtained from the total neutrino flux (black crosses) by subtracting background contributions from naturally occurring radionuclides (lead-214, polonium-210, carbon-14, bismuth-210, and krypton-85), other solar neutrino breeds (from beryllium-7 decay, proton-electron-proton fusion, and the carbon-nitrogen-oxygen cycle), and pile-up—instances where two successive events, usually from 14C, are detected as a single count. The black curve is a fit to the data. (Adapted from ref. 2.)
The pp neutrino spectrum that remains after correcting for the background falls off sharply near 300 keV, as expected. The total integrated flux, (6.6 ± 0.7) × 1014 m−2 s−1, agrees with the value obtained when the standard solar model is constrained by observed solar luminosities.
The composition question
With its latest measurements, the Borexino group has now characterized the spectral fluxes of every breed of neutrino generated in the proton–proton chain. The ultimate promise of the Borexino detector, however, may lie in what it can tell us about the CNO chain. At present, the heavy elements that mediate that pathway are a source of controversy: Optical measurements of their concentrations find values roughly 30% smaller than those obtained from helioseismology.
One tantalizing possibility is that neither technique is in error. Since optical measurements probe the Sun’s surface, and helioseismology probes its interior, “the discrepancy might be telling us that a fundamental assumption of the standard solar model—that the star has a uniform composition—is just wrong,” Calaprice says.
Forthcoming high-precision measurements at Borexino, at the SNO+ detector in Canada, and at the proposed CLEAN and LENS experiments may help settle the debate. Each experiment aims to tally the flux of CNO neutrinos, from which one could infer the abundance of heavy elements in the core. If the Borexino group is to see those neutrinos, however, they’ll need to reduce their background levels by at least a factor of four. Says Calaprice, “That’s our next goal.”
