There are indications that for several scientific areas of study, our current understanding of the particles and forces that govern normal matter is wrong. Many of these areas seem to involve neutrinos, and that’s in part because these particles rarely interact with normal matter, making them incredibly difficult to detect.
But we’ve gradually gotten better at building detectors, which has allowed us to discover that neutrinos have mass (something unaccounted for in the Standard Model) and shift among different identities as they travel. But the process has also revealed some persistent oddities. One oddity is a long-standing excess in one type of neutrino, first described by researchers from Los Alamos back in the 1990s. The same thing was seen at Fermilab in the initial runs of an experiment called MiniBooNE, but neither of them gathered enough data to announce discovery.
Now Fermilab is back with its latest update, using two additional years of MiniBooNE data. The excess is still there, and it has edged even closer to the statistical standards for discovery. If you combine the Fermi and Los Alamos data, we’re already there. It’s looking more and more like another break in the Standard Model, and the possible explanations include an entirely new type of neutrino.
Neutrinos have a number of unusual properties. They’re far and away the lightest particle we’ve discovered, so light that if you give them a reasonably energetic push, their travel can be accurately estimated by assuming they move at the speed of light. They only interact with other matter via the weak force, and they do so only rarely—you need about a light year of lead to have a good chance of stopping one. They also lack a determinate identity. Instead, a neutrino will shuffle among the three known types as they travel in a process called “flavor oscillations.” There are even some indications that neutrinos and antineutrinos are the same particles, differing only in terms of the orientation of their spin.
Tracking neutrinos depends on interactions between a particle and our detector hardware; the rarity of these interactions makes neutrinos difficult to detect. The solution to this problem has generally been to just make lots of detectors. The Daya Bay experiment parks its detectors near some nuclear reactors, which pump out enormous numbers of neutrinos. Fermilab has used part of the accelerator chain from its former centerpiece, the Tevatron accelerator, to smash protons into a stationary target.
This process doesn’t produce neutrinos directly, but it generates large numbers of an unstable particle called a pion. A pion is charged and can be steered into a beam. When a pion decays, it will produce a neutrino that travels in roughly the same direction (positively charged pions produce neutrinos when they decay; negatively charged ones produce antineutrinos). These can be sent to detectors that are hundreds of kilometers away, since the beam will travel through the Earth without many of its neutrinos interacting with anything. At long distances, the neutrinos have enough time to shift to a new identity, allowing us to study flavor oscillations.
But Fermilab also has some detectors right near the source of the neutrinos. This lets them test new detector technology and ensure that the beam sent to distant targets has the properties we expect. And it also provides the chance to determine if flavor oscillations happen on shorter time scales than we expect.
MiniBooNE serves two of these purposes. It is “mini” because researchers were hoping that a successful test of the detector would lead to the building of a full-sized version (BooNE stands for “booster neutrino experiment”). But it has now been gathering data for more than 15 years, giving researchers a strong sense of any extraneous sources of signals in the detector.
Since 2002, the detector has examined the neutrinos produced by more than 1021 protons slamming into their targets, roughly equally divided between selection for neutrinos and antineutrinos. Some of that data had been analyzed earlier, but researchers have now gone back and added two years of additional data, roughly doubling the number of collisions that they’re looking at.
The beam itself is mostly muon neutrinos, but the new data shows a pretty dramatic excess of electron neutrinos. The neutrino-focused experiments saw 381 more events than the 1,578 expected (a 4.5 sigma difference), while antineutrino experiments had 461 extra detections compared to the 1,977 expected, a 4.8 sigma difference. Physics requires a statistical significance of five sigma to declare discovery; MiniBooNE is tantalizingly close and will probably be there within the next two years. But the researchers also combined their analysis with the data produced in Los Alamos and came up with a 6.1 sigma difference. This is clearly at the point where we have to take it seriously.
All of this data means we should try to come up with something to explain why we see so many of these neutrinos. Physicists, being obliging sorts, have come up with at least six explanations. But several of their explanations focus on a proposed additional type of neutrino, called a sterile neutrino.
For reasons we don’t understand, all the non-force-carrying particles we’re aware of come in sets of three. There are three charged leptons (electrons, muons, and taus) and two sets of three quarks. We have identified three types of neutrinos, and some indications from cosmology suggest that’s all we’re going to see. But some researchers have proposed a type of neutrino that doesn’t interact via the weak force, leaving its only interactions to come via gravity. (The lack of interactions led to the “sterile” label.)
The sterile neutrinos can, however, participate in flavor oscillations. This would give an additional path by which muon neutrinos produced in the decays could oscillate into electron neutrinos, having spent a short bit of time as sterile neutrinos in between. While any sterile neutrinos would be missed, the electron neutrinos they oscillate into would be picked up by MiniBooNE, producing the excess we’d observe. This isn’t the only possible explanation, but it’s possibly the cleanest.
The challenge then becomes one of figuring out how to demonstrate this is happening when there’s no possibility of detecting a sterile neutrino directly. It will likely take other signs of additional flavor oscillations—neutrinos disappearing from a beam as they oscillate into sterile ones, for example—to give this idea experimental support. But, long before we get to that level of evidence, we’ll have a pretty good sense that there are behaviors that can’t be explained by the Standard Model, which will be enough to keep physicists happy.