Commanding the Power of Thor...ium: "Wigner Crystals of 229Th for Optical Excitation of the Nuclear Isomer"

ResearchBlogging.orgI have to admit, I'm writing this one up partly because it lets me use the title reference. It's a cool little paper, though, demonstrating the lengths that physicists will go to in pursuit of precision measurements.

I'm just going to pretend I didn't see that dorky post title, and ask what this is about. Well, it's about the trapping and laser cooling of thorium ions. They managed to load thorium ions into an ion trap, and use lasers to lower their temperature into the millikelvin range. At such low temperatures, the ions in the trap "crystallize."

So, they've demonstrated that if you get something cold, it forms a solid? Dude, that's not shocking new physics. There are scare quotes around "crystallize" for a reason. They're not forming a real crystal, in large part because we're talking about triply ionized thorium here, so each has a charge of +3 electron charges. They repel each other pretty strongly, and if they weren't held in a trap, they'd fly apart at high speed rather than forming a solid.

The "Wigner crystal" that forms is a regularly spaced arrangements of these ions, each being more or less stationary, separated from all its neighbors because of the electric force between them.They make these really nifty pictures using light scattered by the ions during the cooling process, so that each ion shows up as a dot of light:

i-c0ea832616e1fc64406037cf8fe670e8-thorium_ions.jpg

Pretty blue dots! That's false color-- the actual light being used is in the infrared, at 984 nm. But yes, it's a pleasing color choice.

Why are there stacks of three pictures? The top picture in each stack shows both isotopes of thorium. For the middle pictures, they block one of the lasers so they only see light from thorium-229, and for the bottom picture, they only see thorium-232. The big picture in part a) is about 200 ions total, with the separation between isotopes having to do with the extra mass. The other pictures show 6, 5, and 4 ions.

Why bother with the tiny little samples? Why not just trap lots and lots of ions? Well, most of the experiments you would like to do with trapped ions work best when you look at a single ion. They trapped 200 ions just to show that they can get lots of them if they want to, but ultimately, they'd like single ions for precision measurements.

You mentioned precision measurements before, too. What's so special about thorium? Well, a little while back, I talked about an experiment using ultra-precise atomic clocks to demonstrate relativistic effects for very small changes in speed or elevation. They're able to see changes in frequency at the level of a few parts in 1016, which lets them measure tiny effects on time.

These clocks are extremely precise and stable, but still subject to some environmental perturbations, because ultimately, they depend on the energy difference between two states of an orbiting electron, and that electron interacts relatively strongly with the outside world.

Yeah, but what other option do you have? Well, there are the nuclei of atoms. The nucleus of an atom is shielded from a lot of the effects of the environment by the fact that there are all those electrons in a diffuse cloud around the outside of the nucleus-- the electron cloud being something like 10,000 times the size of the nucleus. If you have a stray electric or magnetic field, the electron cloud shifts around a bit, which changes the energy of the electron states, but more or less cancels out the field at the nucleus. This makes nuclear states much less sensitive to perturbations, which is why people talk about using them to make quantum computers.

Yeah, but for the computer, you're just using spin-up and spin-down states. I thought you said that the energy separation of those states depends on what's near them, which is exactly what you don't want for a clock. Right. So the spin states of the nucleus aren't a good idea for a clock. An atomic nucleus, though, has structure of its own, with different arrangements of protons and neutrons having slightly different energies. The energies of those states is set by quantum physics, just like the energies of the electronic states used for ordinary atomic clocks. So, you could make a clock based on the energy difference between two different states of the nucleus of an atom.

Yeah, but when nuclei move between states, they absorb or emit gamma rays. Are you really going to make a clock based on gamma rays? Isn't that more of an Incredible Hulk thing than a Thor thing? Well, yes. Also, it's impossible to manipulate gamma rays cleanly enough to use atomic clock techniques with them.

By a happy coincidence, though, thorium-229 has two nuclear states that are separated by a surprisingly small amount of energy-- just 7.6 electron volts. That corresponds to light in the vacuum ultraviolet region of the spectrum-- around 160 nm wavelength-- which is a little difficult to work with, but nowhere near as difficult to handle as gamma rays would be.

So, the point of all this is to make an even better atomic clock using thorium nuclei? Right. Of course, nobody has ever measured these states exactly, so the first step is to collect a bunch of thorium nuclei, and find out what frequency of light they light to absorb. Hence this paper: trapping and cooling thorium ions is the first step toward measuring these nuclear states precisely, and thus providing another sort of time standard.

Isn't this kind of a lot of work to go to for a new clock? And do we really want to be making clocks out of things that are radioactive? Well, the half life of thorium-229 is several thousand years, so it's not like your clock is going to go "poof" all that often, or be dangerously radioactive. But you underestimate the ambition of physicists interested in making better clocks.

Timekeeping aside, though, there are cool things you could do with a thorium nuclear clock. Specifically, you could make one, and compare the rate at which it ticks to the rate at which an electronic clock like the aluminum ion clock from the relativity experiment ticks.

And this would tell you what? It would tell you whether the constants of nature are really constant, or if they change in time. People look for this sort of change in astronomical sources all the time, and also in laboratory experiments based on comparing different types of clocks. Different types of clocks are sensitive to the constants of nature in different ways, so if you set up two clocks and compare them over a long time, you can tell whether the constants are changing-- if the more sensitive of the two speeds up or slows down significantly, you know something's changing.

A thorium nuclear clock would be radically different from an other type of atomic clock, which ought to allow a really precise test in a relatively short time. That wouldn't necessarily require the clock to work well enough to be able to displace cesium or one of the trapped-ion systems people are making now, and it would be a great probe of new physics.

Which, I'm sure you're going to note smugly, would come from experiments in a laboratory and not a billion-dollar accelerator. Well, yeah. Also, because it would be in a laboratory here on earth, it wouldn't be subject to the same uncertainties as astronomy-based tests, which involve so many variable factors it isn't even funny.

Anyway, that's why this is a worthwhile paper (and was written up in Physics): it's pretty cool in its own right, but the real attraction is that it's the first step toward something really awesome: a nuclear-state-based clock, with the potential to test whether the constants of nature are really constant, or if some exotic new physics is involved in making them change.

OK, that is pretty cool. I'm still not going to forgive you the stupid reference in the title, though. Yeah, well, you can't win 'em all...

C. J. Campbell, A. G. Radnaev, & A. Kuzmich (2011). Wigner Crystals of 229Th for Optical Excitation of the Nuclear Isomer Physical Review Letters, 106 DOI: 10.1103/PhysRevLett.106.223001

Categories

More like this

You should write for Angewandte.

That's false color-- the actual light being used is in the infrared, at 984 nm. But yes, it's a pleasing color choice.

How on earth do you separate the spectra of isotopes? Aren't they supposed to have identical chemistry? There's an effect in Hydrogen/Deuterium because the reduced masses differ enough, but is it really possible to see the difference from 3 units out of 230?!

The lasers they use are narrow enough, and the ions are moving slowly enough, that they can select which isotope they want to interact with. The isotope shift is on the order of 10 GHz in frequency units, while the individual resonance lines are only a few MHz wide. 10 GHz is a pretty big frequency difference for laser spectroscopy, so this isn't a terribly difficult meassurement to make.

Thanks.

As a chemist I'm ashamed by my ignorance.

It's pretty cool, but it seems to be that trapping and laser cooling Th 3+ (which has alkali-like structure?) is the easy part of this experiment. The next part - finding the transition - is the hard part. My impression was that the error bars on that "7.6 eV" number were bigger than an eV, and the Th transition was incredibly weak.

As for why Th would be a better clock for time-variation of fundamental constants, I think it has less to do with being a "radically different" kind of clock (since there are much simpler ways of making clocks that are "radically different" than a hyperfine or optical clock). I think it has more to do with the belief that the 7.6 eV splitting comes from an accidental near-cancellation of two much bigger numbers which each depend on different fundamental constants. This gives a huge enhancement in the sensitivity to changes in those constants relative to each other. But that's probably more detail than you wanted to get into in a blog post.

By Anonymous Coward (not verified) on 07 Jun 2011 #permalink

Really "neat", thanks. Since these are ions, their outer electron structures are different than neutral atoms. So, I suppose they could form a different crystal type than the normal state (listed as face-centered cubic) of Thorium. Hard to tell just looking.
BTW I admit to not having time to carefully read up on the story, so "read blah more carefully" is a great answer as applicable.

Also, might as well have a polite QM wrangle (surely rehash of many other discussions, but topical spot to make points) about why it's so easy to just see individual atoms in place, when if they have precise momenta (so cold) it seems their position is rather crisply defined, etc.

A nuclear transition in the uv, how cool is that!

Nice post by the way. I didn't get everything after reading the article and was wondering whether you could clarify. What's this equation for the total search time? You have to irradiate for tens of hours? Is this on resonance, so after having found the resonance (is the transition that improbable)? And once the nuclear transition has been excited, is it the idea to have a single ion with an excited nucleus, so that you can do precision spectroscopy on it?

By youngpostdoc (not verified) on 07 Jun 2011 #permalink

A nuclear transition in the uv, how cool is that!

Nice post by the way. I didn't get everything after reading the article and was wondering whether you could clarify. What's this equation for the total search time? You have to irradiate for tens of hours? Is this on resonance, so after having found the resonance (is the transition that improbable)? And once the nuclear transition has been excited, is it the idea to have a single ion with an excited nucleus, so that you can do precision spectroscopy on it?

By youngpostdoc (not verified) on 07 Jun 2011 #permalink

"It would tell you whether the constants of nature are really constant, or if they change in time."

Wow. That would be something. What do you think would be the implications if the constants of nature changed with time?