How do you weigh a ghost? If you're a cosmologist, you could use... the Universe. Combine vast cosmological data with info from particle accelerators, and, it turns out, you have a pretty good scale for measuring the mass of a neutrino - also known as the 'ghost particle'.
This is how a team of scientists, for the first time, have set an upper limit on the mass of the lightest of the three different types of neutrino.
Neutrinos are peculiar little things. They are among the most abundant subatomic particles in the Universe, similar to electrons, but without a charge and almost massless. This means they interact very rarely with normal matter; in fact, billions are passing through your body right now.
This is why they've earned the moniker of ghost particles. It also makes them incredibly difficult to detect. We do have some methods for detection - such as Cherenkov neutrino detectors - but these are indirect, catching the effects of the passing neutrinos, rather than the neutrinos themselves.
All that means measuring the near-zero mass of these particles is a particularly tough challenge.
"We used information from a variety of sources including space- and ground-based telescopes observing the first light of the Universe (the cosmic microwave background radiation), exploding stars, the largest 3D map of galaxies in the Universe, particle accelerators, nuclear reactors, and more," said cosmologist Arthur Loureiro of University College London in the UK.
"As neutrinos are abundant but tiny and elusive, we needed every piece of knowledge available to calculate their mass, and our method could be applied to other big questions puzzling cosmologists and particle physicists alike."
We know neutrinos have mass because they come in three types, or 'flavours' - electron, muon, and tau. But it's not as simple as them being three discrete entities; the particles actually oscillate between these flavours (and which flavour has which mass is still a mystery).
"The three flavours can be compared to ice cream where you have one scoop containing strawberry, chocolate and vanilla," Loureiro explained.
"Three flavours are always present, but in different ratios, and the changing ratio - and the weird behaviour of the particle - can only be explained by neutrinos having a mass."
Upper limits have previously been placed on the combined mass of all three neutrinos using cosmological measurements. We can use data about the Universe because the mass of neutrinos can actually modify some of the cosmological observables.
Thus, once we have calculated the neutrino density, we can use that to place an upper limit on the collective effect neutrinos produce.
This time researchers took that a step further, using the data they had collected as a framework in which to mathematically model neutrino mass. This was fed into a powerful supercomputer called Grace to perform the calculations.
"We used more than half a million computing hours to process the data; this is equivalent to almost 60 years on a single processor," said cosmologist Andrei Cuceu from University College London.
But Grace did it: the supercomputer returned a mass for the lightest of the three neutrons of 0.086 electron volts (with a lower limit of zero), or around 1.5 × 10-37 kilograms. The team also calculated a combined mass for the three neutrinos - 0.26 electron volts. Both these results have a confidence interval of 95 percent.
For context, a stationary electron has a mass of 511,000 electron volts, or 9.109 10−31 kilograms.
We still don't have constraints for the other two masses, nor have we mapped those masses to the neutrino flavours. And there is still some uncertainty about the expansion rate of the Universe (which cosmologists are trying to resolve). But it's a pretty amazing step.
"It is impressive that the clustering of galaxies on huge scales can tell us about the mass of the lightest neutrino, a result of fundamental importance to physics," said astronomer Ofer Lahav of University College London.
"This new study demonstrates that we are on the path to actually measuring the neutrino masses with the next generation of large spectroscopic galaxy surveys, such as DESI, Euclid and others."
The research is due to appear in Physical Review Letters and is available on arXiv.