This post is part of a series of posts originally written for my blog at Forbes.com that I’m copying to my personal site, so I have a (more) stable (-ish) archive of them. This is the text of the original post, from August 2017, with only one of the images that appeared with it, which is necessary for the explanation.

A bit more than eight months ago, I blogged about an experiment from the ALPHA collaboration at CERN, about an experiment where they did some rudimentary spectroscopy of antihydrogen. Not even a year later, they’re back, with another paper on antimatter spectroscopy, which is cool enough to need another blog post. This time around, they’re measuring the “hyperfine splitting” in antihydrogen (here’s a news story from Physics World), and explaining why this is a particularly valuable thing to study requires a bit of background about how atoms are put together.

These days, everybody learns in grade school that an atom looks a bit like a little solar system, with a positively charged nucleus orbited by negatively charged electrons. Hydrogen is the simplest atom of all, with a nucleus containing a single proton, orbited by a single electron. This picture isn’t quite right, of course, but it’s a good lie-to-children — that is, a simplified conceptual picture you can use at the start to get a handle on things that are a more complicated than the simple version.

One of the biggest ways this “solar system” picture is wrong is that the orbits of the electron are nothing like those of a planet in the actual Solar System. A planet is very much a classical sort of object, so you could speed it up or slow it down by a tiny amount, and it would happily continue orbiting, just a tiny bit closer or farther away from the Sun. An electron in an atom, on the other hand, is very much a quantum-mechanical object, and can only exist in orbits that have very particular energies. You can’t make an arbitrary change in the energy of an electron, you can only change it by certain very specific amounts, and when you do, the energy added or subtracted has to come from or be carried away by a photon of light.

The simplest such model was worked out by Niels Bohr in 1913, and does a nice job of getting the electron energies in hydrogen from just thinking about the electrostatic attraction between the electron and the proton. With the development of quantum mechanics in the 1920s, we got a better handle on what, exactly, the energies of the electron in hydrogen ought to be, and more importantly what effects were being left out that change those energies.

There are a collection of effects that shift the energy of the electron in an atom that sort of get lumped together as “fine structure,” because they take states that have exactly the same energy in the very simplest analysis and split them apart by a small amount. This means that instead of a single frequency of light that can be absorbed to change the electron energy, you have two very similar (different by less than a percent) frequencies. The most interesting of these effects has to do with the “spin” of the electron, which is an inherent property making the electron behave like a tiny magnet. For certain types of orbits, the electron energy shifts up or down a tiny amount due to an interaction between the spin and the orbit. You can think of it fairly loosely in terms of a current loop: from the electron’s perspective, it’s a magnet at rest being orbited by a proton. That orbiting proton creates a magnetic field, and the electron’s energy goes up or down a tiny bit depending on whether the magnet associated with the electron spin is aligned with the field the proton makes or not.

This isn’t the full story, though, because the proton also has a spin, and behaves like a tiny magnet. Which leads to the “hyperfine” effect (so called because the energy splitting it generates is much smaller than the fine structure, and physicists aren’t good with names). If the electron and proton spins are lined up with each other (both up or both down), the energy goes up a bit as the intrinsic magnetic field of the proton raises the electron’s energy; if they’re pointing in opposite directions, the energy goes down a bit. Among other things, this takes the “ground state” of hydrogen and splits it into two states. This ground-state splitting is incredibly important for astronomy, because the universe is full of clouds of cold hydrogen atoms in the ground state that can move back and forth between these two levels by emitting radio waves with a wavelength of around 21 centimeters. Many radio telescopes are specifically designed to look for this “21-centimeter line” in hydrogen, and we’ve learned an enormous amount about the universe by studying this light.

The paper from the ALPHA team that I discussed back in December was measuring the first kind of these energy effects: they looked at the light needed to drive anti-hydrogen atoms from the ground state to the lowest excited state, and that energy is mostly determined by the simple electrostatic interaction between the electron and proton (or positron and the anti-proton). The energy involved is several million times greater than the hyperfine splitting, and the wavelength of the light in question is several million times shorter than the 21-centimeter line, at 121 nanometers.

The current experiment is measuring that several-million-times-smaller hyperfine splitting, using basically the same principle as the earlier measurement: they collect a bunch of antihydrogen in a magnetic trap, hit them with light of the appropriate frequency, and see how many atoms are left. If the frequency of the light they’re hitting the anti-atoms with matches the transition frequency, the resulting state change causes atoms to fall out of their trap and annihilate with ordinary matter in the walls, which they can detect.

The process is complicated by the fact that their trapped atoms are held in a whopping huge magnetic field, which shifts the energy of the electron orbits. They’re rescued by a quirk of atomic structure, though, which is that in the high-field limit, the states split into two groups of two, giving two transition frequencies that differ by exactly the hyperfine splitting. This lets them do something that looks a lot more like ordinary spectroscopy. In the December paper, they fixed the laser at the frequency for ordinary hydrogen and confirmed that it caused losses of antihydrogen. In this paper, they vary the microwave frequency to find the maximum loss for one of the transitions, then increase the frequency by approximately the hyperfine splitting for hydrogen, and repeat the process to find the maximum for the other. The difference between the two frequencies of maximum loss gives the hyperfine splitting in antihydrogen.

Since hydrogen is the simplest atom, and the only one whose properties can be calculated exactly, this hyperfine splitting has been extensively studied, and measured to impressive precision. The exact frequency of light absorbed or emitted when a hydrogen atom switches ground states is 1,420,405,751.773 Hz, plus or minus about 0.001 Hz. The ALPHA team can’t quite match that phenomenal precision, but it’s a very respectable first effort: 1,420,400,000 Hz plus or minus 500,000 Hz.

So, why is this an interesting measurement? Well, for one thing, the precision to which the hyperfine splitting is known makes it an attractive target. It’s also a whole lot easier to work with microwaves than the vacuum ultraviolet lasers needed for the earlier work, which is part of why the December paper involved a fixed laser frequency. (This is not to say that it’s easy in any objective sense, though — one of the issues they face is that one of the two frequencies they use is much harder to get into their very complicated magnetic trap apparatus than the other, so they have a huge disparity in the intensity of the radiation hitting the trap.)

There are also some physics reasons to think the hyperfine splitting might be a good place to look for differences between hydrogen and anti-hydrogen. The hyperfine interaction is between the spin of the proton and the spin of the electron, which means it’s both very weak (since the magnetic field generated by either is tiny) and very short-range (because it’s a dipole interaction, rather than the direct charge-charge interaction). Much of the shift happens thanks to interactions when the electron is inside the nucleus, and that’s exactly the sort of scenario where you’d expect exotic physics to show up.

They have a long way to go before they get to the sort of precision where anybody might expect differences between matter and antimatter to show up. (In the very simplest models of high-energy physics, there’s absolutely no difference, but those models can’t fully explain why the Big Bang created enough extra matter to make, well, us. So there’s got to be some difference somewhere.) This experiment, maybe even more than the December one, is a promising step in the development of high-precision spectroscopy of anti-matter.