This is an entry in my series where I read physics articles in Scientific American, and pretend that my physics PhD is useful.

A letter

This month, we have an article called “The Darkest Particles“, about sterile neutrinos (sorry, this one’s paywalled). But before I get to the main attraction, I’m excited because we have the first letter from a responding to an article that appeared earlier in this series! Science writing isn’t just about the cutting edge, it’s also about following up with critical discussion and further research. These letters give us a small taste of critical scientific discussion.  Although, many of these letters are written by non-experts, so it’s not quite the same.

This letter responding to “A Cosmic Crisis“, about discrepancies between different measurements of the expansion rate of the universe. The reader asks whether this could be caused by gravitational effects from objects outside the observable universe. The author replies by saying it would be difficult or even impossible to falsify such a theory, so theories instead focus on other things.

Honestly, I don’t agree with this answer. If a theory predicts a discrepancy between two measurements of the expansion rate of the universe, then it is ipso facto falsifiable. I also think that in order to have gravitational effects from stuff outside the observable universe, we’d need a non-uniform universe, which would likely cause asymmetries in the CMBR, which we do not observe. Okay, so there’s something called the “axis of evil”… but at this point I’m out of my depth and would refer you to cosmologists.

The other sterile neutrinos

Back to “The Darkest Particles”. When I first saw that this article was about sterile neutrinos, I thought it was about the other kind of sterile neutrinos.

Within the Standard Model, there are three types (“flavors”) of neutrinos, called the electron neutrino, muon neutrino, and tau neutrino. Additionally, each neutrino has two subtypes, relating to the direction of its spin. Just as an electron can be “spin up” or “spin down”, a neutrino can have its spin in the same direction as its momentum (“right-handed”) or in the opposite direction of its momentum (“left-handed”). Left-handed and right-handed neutrinos are basically mirror images of each other.

But, in a shocking twist, left-handed neutrinos can interact with other particles via the weak force, while right-handed neutrinos cannot. Right handed neutrinos are inert, or “sterile”. They only interact with other particles via gravity. It is very strange, but it turns out the laws of physics do not have mirror symmetry.

How do we know that right-handed neutrinos exist at all? We don’t detect them directly, but theoretically it should be possible to decelerate a neutrino and reverse the direction of its momentum, thus switching it from left-handed to right-handed. Note, this is only possible if neutrinos have mass. Initially, physicists thought neutrinos were massless, but they were proven to have a very small but nonzero mass.

Correction: Neutrinos actually have two kinds of handedness.  First there’s the helicity, which is what I described above.  The other is chirality, which is quantum… math… stuff.  Helicity and chirality always match if neutrinos are massless–however, as I explain below, neutrinos are not massless.  If we were to decelerate a neutrino and reverse the direction of its momentum, we would switch the handedness of the helicity, but the chirality would remain the same.  Unfortunately, sterile neutrinos are the ones that have right-handed chirality, so we may have proven the existence of neutrinos with right-handed helicity, but we have not proven the existence of sterile neutrinos.

How do we know neutrinos have a nonzero mass? It’s a long story. Many decades ago, physicists sought to measure the number of neutrinos coming from the sun. They found a discrepancy between the measured number of neutrinos and the number predicted by solar physics.  Eventually physicists reached the consensus that neutrinos were changing flavor on their way to Earth (and these detectors were mostly sensitive to only one of the three flavors).  The phenomenon of neutrino flavors changing over time is called neutrino oscillation.

The theory of neutrino oscillation is based on a curious fact, that each neutrino flavor does not have one mass, but has a superposition of three different masses. It’s a general rule in quantum mechanics that anything with energy oscillates at a frequency proportional to its energy. Thus, the three masses of a neutrino oscillate at different frequencies, and the flavor changes as the three masses oscillate in and out of alignment.

That’s a lot of background that I knew before even picking up the article. See, I took a course on particle physics when I was an undergrad, and I remember a few things.

The fourth flavor

So when I saw the article was about sterile neutrinos, I thought it was about right-handed neutrinos. It turns out that it is instead about a theoretical fourth flavor of neutrino, one which is much more massive and does not interact via any force except gravity. If true, then this is pretty huge for particle physics. There are a bunch of particles (“leptons”) which come in exactly three flavors. Why three flavors? I don’t know, maybe there’s a rationale in string theory or whatever. Or maybe there are more flavors and they just become extremely difficult to detect. I’m surprised I hadn’t heard of this research before reading this article.

The experiment used to detect the fourth flavor of neutrino is neat. They use the Coherent CAPTAIN-Mills (CCM) detector, which detects neutrinos via coherent scattering. The important thing that distinguishes CCM from other neutrino detectors is that it’s relatively small. So, by moving the detector they can detect changes in neutrino flavor over relatively short distances. Short distances are important, because if this fourth flavor of neutrino is more massive than the others, then the oscillation into the fourth flavor occurs over relatively short distances.

It’s a long article, but it mostly spends time explaining background I already knew, and then the new stuff was fairly straightforward. Personally, I look forward to the day when these researchers overturn the Standard Model and replace it with something even more bizarre and inexplicable.

Post-script: On statistics

There’s an inaccuracy in the introduction of this article, which would bother many a statistics nerd. They claim that there is a 99.999999% chance that something beyond the scope of known physics is occurring. But clearly what they mean to say is that known physics has less than a 0.000001% chance of giving them a result similar to what they observed. Equivalently, they could say that the p-value is less than 0.00000001. Or to use physics lingo, they have a 6-sigma result.

Explaining p-values to a popular audience is always difficult. It’s tempting to say something like “We have 99.999999% confidence” or “The standard model has less than 0.000001% likelihood.” Both of which are technically correct, because “confidence” and “likelihood” have technical meanings that do most of the work. But it’s only accurate on paper, and then it stops being accurate once it’s transferred into the mind of the average reader who doesn’t understand that these words have technical meanings.

What all of this means is that they have demonstrated that their results are not a statistical fluke. However, this does not necessarily mean it’s caused by a fourth flavor of neutrino. The jury’s still out on that one.