Trapped Antihydrogen

ResearchBlogging.orgThe big physics-y news story of the moment is the trapping of antihydrogen by the ALPHA collaboration at CERN. The article itself is paywalled, because this is Nature, but one of the press offices at one of the institutions involved was kind enough to send me an advance version of the article. This seems like something that deserves the ResearchBlogging Q&A treatment, so here we go:

OK, what's the deal with this paper? Well, the ALPHA collaboration is announcing that they have created antihydrogen atoms-- that is, a single antiproton orbited by a single positron-- at low temperatures, and confined them in a magnetic trap for something like 172 ms.

Awesome! When can we blow up the Vatican? Settle down. We're not talking huge quantities of antimatter, here. In 335 runs with their apparatus, they detected all of 38 atoms of antihydrogen. You're not going to be blowing anything up soon.

What's the point of making antimatter if you can't use it to blow stuff up? The point is to understand the laws of physics better. If you can do spectroscopy of anti-atoms, it will tell us a lot about whether antimatter obeys the same laws as ordinary matter, which might provide a clue as to why everything we see seems to be made of ordinary matter. You could also use it to test how antimatter interacts with gravity, which is something we don't currently have any way to test.

OK, fine, it's all about basic physics. So, this paper is the first time people have made antihydrogen? No, the first observation of antihydrogen was back in 1995, followed by another observation in 1997. The first cold antihydrogen was at CERN in 2002, and again the same year.

So what's today's article about? this is the first time that antihydrogen has been magnetically trapped. Which is a big step, because if you want to do spectroscopy of antihydrogen, you need the atoms to stick around for a while so you can interrogate them with lasers.

OK, so how do they make this stuff? They make it basically by sticking a large number of antiprotons and positrons in the same region of space, and waiting a while. If you've got enough antiprotons and positrons, eventually three of them will collide and form an atom.

Three? Don't you only need one positron and one antiproton? That's what you need to make an atom, but if you want the atom to stick around in a bound state, you need a third body to carry off some excess energy and momentum.

OK, so the creation just sort of happens. How do they trap it, if it annihilates when it comes into contact with normal matter? Well, you can't let it come into contact with any normal matter, that's for sure. What they do is they use strong magnetic fields to confine the atoms to a small region of space. Then they use electric fields to push any remaining loose positrons and antiprotons out of the trap, so they just have antihydrogen in the trap.

Wait, how does that work? Well, the positrons and atiprotons, being charged, respond very strongly to electric fields, so they rush toward or away from a big electric field, depending on their charge. Bound atoms of antihydrogen, though, are neutral, and electric fields just make a small shift in their energy levels. A neutral atom will behave like a magnet, though, so you can push them around with magnetic fields much more effectively than you can with electric fields. Thus, you can push the charged particles away, and keep the neutral atoms around.

OK, so how do you know when you've trapped one? Well, you turn off the trap, and let it fly out and collide with normal matter. When it does, it spits out a bunch of high-energy particles, which then get detected by instruments surrounding the trap. They can tell when they got antihydrogen by tracing the particle tracks backward to show that they originated in the trap region.

Yeah, but aren't there a lot of things that can create tracks in the detectors? Yes, but you can rule most of them out by looking at the shape of the tracks. When the particles are coming from inside, and have the right number and character, you can say that it was due to an antihydrogen atom annihilating on the edge of the trap chamber. A cosmic ray particle, or a radioactive decay somewhere else would produce tracks that start outside the chamber.

OK, but what about the leftover antiparticles? Couldn't an annihilation on the edge of the chamber be due to a leftover antiproton that didn't get pushed out? It could, but they can test that by applying different fields and looking at the distribution of counts. The key figure from the paper is this one:


These are two plots of the distribution of the detected antihydrogen atoms in time (vertical axis) and position along the detector (horizontal axis). The large colored symbols represent the positions of the 38 probable atoms they detected, and are the same in both plots.

The haze of grey dots in the upper picture show the locations of detected atoms in 2000-ish simulated experiments, and show that all the points they're calling atoms fall in the right region. The smaller colored blurs in the bottom pictures represent the locations of detection events if they were looking at leftover antiprotons, for three different configurations of applied fields. You'll notice that with two exceptions, none of the detections they're calling antihydrogen atoms fall in the places you would expect antiprotons, which strongly suggests that they're not leftover antiprotons, but real, true, neutral atoms.

So I'm supposed to believe this is antihydrogen just because they simulated what they think ought to happen and the real data sort of look right? Isn't that kind of weak? Yeah, well, welcome to experimental particle physics. Enjoy your stay.

This is a fairly standard method of evaluating data in the particle and nuclear physics communities, and these plots are really nice as this sort of thing goes. They also have some other evidence, namely that when they prepare the antiprotons and positrons at a much higher temperature, where they don't expect much antihydrogen to form, they don't see any. That, again, is strongly suggestive that they have real antihydrogen.

OK, I guess I'll buy that. So, is this good enough to work with for the basic physics tests you mentioned? It's tough to say without knowing more about the details of their experiment, and how they plan to proceed. It's not clear, for example, where you would put the lasers to do spectroscopy of the antihydrogen atoms.

I can tell you this much, though: the ATTA group at Argonne National Lab did spectroscopic measurements of the size of heavy helium nuclei using an apparatus that trapped a single atom at a time, with atoms having a lifetime of about 0.2 s. So if they can make antihydrogen atoms stick around for 0.17s, they can probably do the sort of spectroscopy you would really like to do with these atoms.

So, how long would it take to get enough of this stuff to blow up the Vatican, anyway? A really long time. For an antihydrogen annihilation equal to a tone of TNT, you'd need a few time 1019 atoms. Each cycle of their experiment takes a couple of seconds, so their 38 atoms in 335 attempts is something like 0.05 atoms/s. At that rate, you would need to run the experiment for about 1500 times the current age of the universe to get enough to blow up the Vatican. It's not going to happen any time soon.

A long, long time before you could make a weapon out of this stuff, though, they'll have enough to tell us more about how the universe is put together. And that's the really exciting part of this result.

Andresen, G., Ashkezari, M., Baquero-Ruiz, M., Bertsche, W., Bowe, P., Butler, E., Cesar, C., Chapman, S., Charlton, M., Deller, A., Eriksson, S., Fajans, J., Friesen, T., Fujiwara, M., Gill, D., Gutierrez, A., Hangst, J., Hardy, W., Hayden, M., Humphries, A., Hydomako, R., Jenkins, M., Jonsell, S., Jørgensen, L., Kurchaninov, L., Madsen, N., Menary, S., Nolan, P., Olchanski, K., Olin, A., Povilus, A., Pusa, P., Robicheaux, F., Sarid, E., Nasr, S., Silveira, D., So, C., Storey, J., Thompson, R., van der Werf, D., Wurtele, J., & Yamazaki, Y. (2010). Trapped antihydrogen Nature DOI: 10.1038/nature09610

More like this

Just doing the math⦠(2 * 35 * (1.67e-24 g)) * (c^2) = 1.0506448e-8 joules

So, they could create an 'explosion' of 1.0506448e-8 joules, which is about ((2 * 35 * (1.67e-24 g)) * (c^2)) / (4.18e15 joules) = 2.51350432e-24 Megatons.

They'd need (.15 * (4.18e15 joules)) / ((c^2) * (1.67e-24 g) * 2) = 2.08871732e24 atoms of anti-hydrogen to get 1 Hiroshima. That's
((.15 * (4.18e15 joules)) / ((c^2) * (1.67e-24 g) * 2)) / Avogadro's number = 3.46839629 molesH or (((.15 * (4.18e15 joules)) / ((c^2) * (1.67e-24 g) * 2)) / Avogadro's number) * 1.0079 g = 3.49579662 grams

I think the Vatican City is safe.

A couple more questions in the blowing-things-up vein, if you don't mind.

Various stfnal places talk about confining neutral antimatter by trapping it inside carefully structured matter - one or maybe two atoms of antimatter inside a buckyball, for instance. Would that actually work?

Is there any even vaguely plausible way of manufacturing antimatter in mole quantities?

If you can do spectroscopy of anti-atoms, it will tell us a lot about whether antimatter obeys the same laws as ordinary matter, which might provide a clue as to why everything we see seems to be made of ordinary matter.

I'm all for basic science and doing things just because we can and they're interesting, but unless I'm missing something this is really overstating the scientific interest of anti-atoms. We've tested the hell out of differences between matter and antimatter. The only hints (which aren't statistically significant enough to be called anything more than hints) of anything we don't understand involve b quarks. There's no way that new sources of CP violation would show up in anti-hydrogen but not in the zillions of kaon decays and other places where such things have been looked for.

I think the tests of fundamental symmetries involved have to do with CPT, not CP. If so, CPT is close to being a tautology, so this is low-probability- high-reward situation and so on and so forth...

I don't think we have even scratched the surface of what can be known about anti MATTER. Particles, yes. Matter, no.

1) We don't even know if it falls due to gravity. That is pretty basic. Some hints, yes, but that is all AFAIK.

2) They still haven't measured the most basic spectrum, let along the Lamb shift, in anti-hydrogen.

3) Molecules? Boiling point?

By CCPhysicist (not verified) on 17 Nov 2010 #permalink

Don't be absurd. We know the particles and their interactions. They're the same ones that give the usual spectrum and Lamb shift. There is no possible way the answer would be different. I'm sure it would be a beautiful experimental tour-de-force to measure them, and someone should do it, but let's not pretend we don't know the answer.

Sorry if the phrasing there comes off as aggressive.


They could easily measure the gravity effects. If you assume that within the moment they quench the superconductors you have gravity over magnetic field you should see a clear pattern flying pattern.

Anyhow Figure B suggest that for the background calculus.

Don't be absurd. We know the particles and their interactions. They're the same ones that give the usual spectrum and Lamb shift. There is no possible way the answer would be different. I'm sure it would be a beautiful experimental tour-de-force to measure them, and someone should do it, but let's not pretend we don't know the answer.

I don't think it's as settled as that. I mean, prior to the observation of parity violation, most physicists would've said just as confidently that it couldn't exist, but now we not only know that it does, we know that there has to be more CP violation than we know about.

And there are some funny results in experiments involving antimatter recently-- the positronium scattering experiment, the mysterious MiniBoonE results. None of this is conclusive, but there are things that at least seem odd, which makes the possibility of doing more precise investigations a lot more interesting.

They could easily measure the gravity effects. If you assume that within the moment they quench the superconductors you have gravity over magnetic field you should see a clear pattern flying pattern.

I think the temperature is still much too high for this to work. That is, the center of mass of the collection of particles should fall under the influence of gravity, but the cloud will also expand as the individual particles fly ballistically outward with their random thermal velocities. The expansion will bring the particles into contact with the walls in all directions much faster than the acceleration due to gravity will pull them down to the bottom part of the detector.

They could easily measure the gravity effects.

No, they can't. The gravitational interaction of two positrons is 40-ish orders of magnitude weaker than their electromagnetic interaction. The antiprotons fare a bit better, but not by nearly enough to be detectable. In a large class of experiments, which includes most of AMO/plasma physics/particle physics, the researchers can safely ignore gravity. This experiment is one of them.

By Eric Lund (not verified) on 18 Nov 2010 #permalink

It's not a question: does anti-H have a Lamb shift? Of course it does. The question is: to how many digits is the Lamb shift of anti-H the same as the Lamb shift of H? Which corresponds to the question--to what precision can we test CPT?

(I'm not sure that the Lamb shift is the most sensitive question you can ask this way, but it's at least familiar to people.)

Most of experimental particle physics deals with parameters which (due to the low statistics) can typically measured to about 2 significant digits, maybe 3 or 4 in kaon physics.

Up to some level, we absolutely do know that CPT is conserved, and agreed there is absolutely no way we'll see something larger than that level. But only up to some level. The point is that spectroscopic methods are likely to be capable of pushing well beyond that level--by many orders of magnitude. If CPT violation is a part in 10^18, maybe this is the only way you find it.

(I made up the number 10^18 for illustrative purposes only. I have no idea what the current bounds are, or how far spectroscopy can go beyond them.)

By Andrew Foland (not verified) on 18 Nov 2010 #permalink

I'm still unconvinced. Very few things in particle physics are constrained by spectroscopic measurements on ordinary hydrogen; g-2 of the electron and muon tend to be leading constraints. Spectroscopy on short-lived antihydrogen states is going to be hard to do anywhere near as precisely as spectroscopy on ordinary hydrogen, isn't it? Even if not, I'm not sure how much we could learn. CPT violation would require Lorentz violation, and we already have absurdly precise tests of Lorentz invariance. We also have extremely precise tests of positron-photon interactions from collider experiments. It's just true that we can enumerate every possible operator that extends the Standard Model, and almost all of them are suppressed by scales of multiple TeV. Tests on antihydrogen are going to be inferior to collider constraints at the TeV scale unless they can probe things to precision of order ten decimal places, at least.

One of the biggest arguments for testing CPT on antihydrogen is that since we don't know where CPT might show up, we should test for it in all possible sectors.
Thus, while different measurements might have test CPT down to a different level of precision, it isn't fair to say that one kind of test is 'better' than another -- they're complementary.

For the truly dedicated, I recommend the webpageof Alan Kostelecky, one of the fore-most post-CPT theorists.

You can also check out his paper on possible leading-order signals in antihydrogen spectroscopy

And lastly, you can find a larger review here

By Eoin Butler (not verified) on 21 Nov 2010 #permalink

Do you think its funny to be writing about killing the pope or blowing up the Vatican? Are you STUPID?

Don't you know that there are lots of Islamic radicals out there who would like to do just that (like Mehmet Ali Agca, the attemptor of pope-icide?

Your comments are disgraceful, yes? Why don't we ask how many atoms of anti-matter it would take to kill you? How would you like someone to write about that?

The limitations on hydrogen spectroscopy to test fundamental theory come from how well we know the proton size, and there is some very interesting new work from the Lamb shift of muonic hydrogen on this point. In a comparison of hydrogen with antihydrogen this drops out unless their sizes are different. The precision of hydrogen spectroscopy is 2 10^14, so that's why it's an interesting potential CPT test. Tests at the 10^10 precision are expected within the next 5 years. Actually, the theorectical limit from the lifetime of the hydrogen 2s state is around 10^18.

The most sensitive CPT test usually quoted is a kaon - antikaon comparison. Since kaons are quark-antiquark pairs, an interaction that affects all quark types by the same amount and acts on antiquarks with the opposite sign would cancel in such tests.

The whole structure of quantum field theories hangs on CPT, so precise experimental tests of it in multiple systems is really necessary. CP violation was discovered experimentally, and was a big surprise.

Do you think its funny to be writing about killing the pope or blowing up the Vatican? Are you STUPID?..

jh, apparently you are not a fan of Dan Brown's books; the plot to blow up the Vatican with an antimatter bomb was the implausible plotline of the pulp-fiction which is Angels & Demons. The article is a dig at the stupidity of Dan Brown's premise rather than an invitation to religious fundamentalism.

By Andy Bowes (not verified) on 23 Nov 2010 #permalink