What's So Interesting About Single Quantum Systems? Physics Nobel 2012

In which we do a little imaginary Q&A to explain the significance of Tuesday's Nobel Prize to Dave Wineland and Serge Haroche.


I did a quick post Tuesday morning noting that the latest Nobel Prize in Physics was awarded to two big names from my corner of the field. This would've been a great time to drop a long explainer post about what they did and why it's cool, but alas, I have a day job, and the Nobel committee stubbornly refuses to tell me who they're giving the prizes to in advance. Oh, well.

Still, I'm just vain enough to think I can add something a little different than the many other posts that have already been written about this, so I'll have a go at it in the traditional fake Q&A format.

OK, what's this Nobel Prize stuff about? The citation says it's about "measuring and manipulation of individual quantum systems," but why is that interesting? Well, in one way of looking at the world, the big difference between the everyday world that we're used to, and the world of quantum mechanics is that the everyday world contains too much stuff. Everything in the universe is fundamentally quantum-mechanical, but when you paste lots and lots of different quantum systems together, all the cool quantum mechanical effects sort of cancel each other out.

So, while it's true that all objects can only take on discrete states-- for example, the energy must be a round integer multiple of some fundamental energy, with no "in-between" energy values allowed-- for a massive object, those allowed states are so close together that it looks like they have a continuous range of values. All objects can and do exist in superpositions of multiple states at the same time, but for a massive object containing many particles bound together, those superpositions fall apart very quickly. And so on. Quantum effects are there, but they're so ridiculously small that there's no hope of measuring them directly.

So, if you want to see quantum mechanics at work, you need to look at single systems? Right. There are some macroscopic phenomena that can only be fully explained by quantum mechanics-- the conduction of electrons in a solid, for example-- but if you want to see what you might call "textbook" quantum behavior-- discrete states, superposition states, wave nature, etc.-- you really want to look at single isolated quantum systems.

So, these guys got a Nobel for doing the easy problems? No, they got a Nobel for doing the hard work needed to make it possible to study the easy problems. Looking at single quantum systems is only "easy" in an abstract mathematical sense. Actually doing the experiments is hard.

In what way? Well, go get me a single beryllium atom. Here's a pair of tweezers, in case you need them.

Oh. Yeah, I guess that does sound pretty difficult. OK, so what's involved in the process, here? Well, there are two laureates, and they each worked on different things. Which would you like to hear about first?

Ummm... Well, let's start with Wineland, becase looks like he deserves a special Nobel for facial hair. Yeah, as Matthew Francis notes, he'll definitely be played by Sam Elliott in the movie.

The single quantum systems Wineland has worked on, and made it possible for other people to work on, are isolated single atoms. He's best known for studying the behavior of small numbers of trapped ions (which, despite the different name, are just atoms with one electron removed).

How do you trap an ion, anyway? Happily, I have an old blog post with exactly that title (which is the source of the figure above). The short answer is that because the ion has a positive electric charge, you can easily push it around with high-voltage electrodes. The right arrangement of high voltages, switched very rapidly, creates a trap that will confine a single ion, or many ions, to a small region of space, where it can be studied.

And that's enough to let you see the quantum behavior of a single atom? Not quite. If you really want to see quantum behavior, you also need to cool the atoms, and Wineland was one of the very first people to employ laser cooling-- in fact, the origin of the field is usually traced to two papers in the mid-70's, one of them by Wineland and Hans Dehmelt (the other was by Theodore Hänsch and Arthur Schawlow. All four of the authors of those papers now have Nobel prizes for various things).

How do you cool things by blasting them with lasers? The idea is that light carries momentum, which is transferred to an atom when it absorbs that light. So, every time an atom or ion absorbs a photon of light, it gets a kick in the direction the laser was headed. If you're very clever about setting up your laser, you can arrange it so that atoms only absorb light if they're moving in the opposite direction, which means they only slow down. Temperature is, loosely speaking, a measure of the kinetic energy of motion of the atoms, so when you slow them down, you make them cold. Hence, laser cooling.

This actually works? Yes it does, brilliantly. Wineland demonstrated cooling of ions in the late 1970's (PDF paper, courtesy of NIST), and since then has refined the technique to an amazing degree. Trapped-ion experiments routinely cool ions down to the point where you can see their discrete energy states, and use what are called "Raman transitions" to cool them even further-- the ions absorb light of one frequency, and emit light of a slightly higher frequency that causes them to drop down one level in the trap. Using these techniques, they can regularly produce ions trapped in the very lowest energy level allowed by their traps. It's not really meaningful to speak of a "temperature" for a single trapped particle, but this essentially allows them to get arbitrarily close to absolute zero.

So, how do they know when they have one of these things? Do they stick it under a microscope? In a very limited sense, yes. They shine in light that the ions like to absorb, and a short time after absorbing a photon the ion will re-emit a photon in a random direction. This cycle repeats millions of times a second, spraying photons in all directions, and they collect some of this light with a lens, letting them see an ion as a bright spot floating in the center of their trap. They don't resolve its size in any meaningful sense, but they can tell the difference between one ion and none, and pick out multiple ions caught in the same trap. They can even measure the state of a trapped ion, once they know they have one, by how strongly it absorbs and re-emits light.

And what's this good for? All kinds of stuff-- just go to the NIST Time and Frequency Publications Database, plug Wineland's name into the search box, and click a link at random. It's almost certain to be mind-blowing stuff. They've made "Schrödinger's cat" states of single ions (PDF) and up to six ions (PDF). They've done all sorts of work toward quantum computing with ions, and wrote an excellent Scientific American article about it (PDF). And because Wineland's at the National Institute of Standards and Technology, they've applied this to making the world's best frequency standards, including a clock that's so good you can see the effects of relativity on time. It's amazing stuff from start to finish.

OK, so what about Haroche? What did he do, study single neutral atoms? No, Haroche's contribution was the study of the other great weird quantum object: light. Haroche's share of the prize comes from developing techniques for the study of systems containing a single photon of light.

Again, that seems ridiculously easy. If you want a single photon, you just make a really weak light source. Right, but how do you pick that photon out of all the other photons rattling around?

OK, that's a good point. So, Haroche developed single-photon detector technology? Not exactly, no. He pioneered the field of "cavity QED," which allows you to modify the interaction between light and matter to a level where you can see the effects of single photons interacting with single atoms.

OK, what? I thought the interaction between light and matter was fixed by, you know, the laws of physics. Now you tell me it's something you can crank up if you like? Is this Haroche guy from the Q Continuum or something? You're not really modifying the laws of physics, just the local interactions. Very loosely, what you do is arrange it so that the photon sticks around longer, giving it more chances to interact with a given atom?

Oh, so it's a slow light thing? No, but that does connect to the problem of working with single photons. One of the reasons it's difficult to work with single-photon systems is that photons, by definition, move at the speed of light. If you shot a single photon at a single atom, you've got one chance for the two to interact, and a second later, the photon is 186,000 miles away.

Cavity QED changes this situation by putting the atom between two mirrors. Now, a photon sent along the line between the mirrors will bounce back and forth many times, giving it many more chances to interact with the atom. It's still whipping back and forth at the speed of light, but just bouncing around in a small area, and every time it goes by the atom, they can interact. This ends up looking just like you've cranked up the interaction between the light and the atom.

(Note: this is a very rough picture trying to get the concept across. Strictly speaking, you can't track the actual photon going back and forth, and instead speak of a "mode" of the field, which can be occupied by an integer number of photons. The coupling between that mode and an atom inside the cavity is much stronger than the coupling between a mode outside the cavity with the same frequency.)

So, OK, you blast a laser into the space between these mirrors, and see what happens to the atoms? Sort of. Haroche's experiments mostly use microwaves-- not microwave ovens, but photons whose frequency puts them in the microwave region of the spectrum. These are exceptionally low-energy photons, compared to visible light, but they're the same basic phenomenon, in physics terms.

Why use microwaves for this? Well, because billions of dollars have been spent developing technology to produce and manipulate microwaves, starting with the development of radar in World War II, and continuing through modern cellular telecommunications. They're also convenient because their wavelengths are very long-- millimeters or more-- which means the cavities are on a convenient scale, and easier to manufacture. The same sort of work at optical frequencies presents some real technical challenge, because you need extremely high-quality mirrors positioned within tens of nanometers. Cavity QED in the optical domain is a pretty amazing field in its own right, with a lot of pioneering work done by Jeff Kimble at Caltech. It's basically the same sort of thing that Haroche developed with microwaves, though.

So, if it's harder to use visible light, why bother at all? There are drawbacks to the microwave approach, too. If you're working with microwaves, you need to worry about thermal photons-- at room temperature, there are large numbers of microwave photons being absorbed and emitted due to the thermal energy of atoms and so on. So microwave experiments need to be done at very low temperatures, which makes the experiments much more complicated.

It's also a little annoying to do these experiments with atoms in their lowest energy states-- you can find transitions in the microwave regime, but they tend to have very long lifetimes and it's a little tricky to detect which state a given atom is in. So, Haroche's experiments use highly excited "Rydberg" atoms, which have lots of very closely spaced energy levels that interact strongly with microwaves, and techniques for detecting the exact state of the atoms with high precision. But making the Rydberg atoms isn't trivial, either.

Which is why it's worth a Nobel? Exactly. There's a lot of hard work involved in doing the experiments, but Haroche and his group did a brilliant job of it, and have been able to demonstrate all sorts of cool effects using the strong interaction between photons trapped in the cavity and atoms passing through.

Sadly, most of Haroche's work is in paywalled journal articles, but you can get a sense of some of it from the American Institutes of Physics collection of articles, many of which are temporarily free to read in honor of the Nobel announcement. I've also written up a couple of them for the blog, on the quantum Zeno effect applied to cavity fields and the life cycle of microwave photons. The experiments are a technical tour de force, and demonstrate clearly and unambiguously that light is made up of quantum particles, with all the weird properties that involves.

So, these sound like obvious and well-justified prizes. Is there anything people will grumble about, here? You mean, other than the annual griping that the Nobels don't represent the full range of science, and all that? The only real complaint I can imagine is that they only used two of the three possible "slots" for this prize, and could've added a third laureate, most likely Jeff Kimble. If there's going to be any complaining that somebody got left out, Kimble's the most obvious candidate.

But that happens every year. In fact, Wineland is a past subject of such commentary-- when the 1997 Nobel Prize was awarded for laser cooling of neutral atoms, there were a few complaints that Wineland should've been a part of any laser cooling prize, as his work with trapped ions came before the work of the 1997 laureates (who included my Ph.D. thesis advisor Bill Phillips). As in that case, though, there's a good case to be made that the work Haroche did is distinct enough from Kimble's to be a separate prize. And Kimble would be a decent candidate to get a share of some future quantum optics sort of Nobel.

In terms of the new laureates themselves, though, I can't imagine anybody within the quantum optics field complaining that either Wineland or Haroche don't deserve a trip to Stockholm. They're both Names to conjure with in the quantum optics/ AMO physics community, and both have sterling reputations. I know a bit more about Wineland than Haroche, because I'm on the same side of the Atlantic as him, and I don't recall ever hearing anybody say anything negative about him-- he's kind of quiet, but universally highly regarded, and his former students and post-docs have gone on to do great things. The students and post-docs I know who have worked with Haroche also speak extremely highly of him. Both of them have not only done brilliant science, but set a great example helping make AMO physics one of the most collegial subfields of science I know of.

So, you approve? Not that it matters, but yes. I couldn't be happier with this pick for the Nobel. They're richly deserving of the honor and the wider recognition that comes with it.


More like this