What's So Interesting About Ultracold Matter?

The first of the five categories of active research at DAMOP that I described in yesterday's post is "Ultracold Matter." The starting point for this category of research is laser cooling to get a gas of atoms down to microkelvin temperatures (that is, a few millionths of a degree above absolute zero. Evaporative cooling can then be used to bring the atoms down to nanokelvin temperatures, reaching the regime of "quantum degeneracy." This is, very roughly speaking, the point where the quantum wavelength of the atoms becomes comparable to the spacing between atoms in the gas, at which point the atoms become "aware" of one another.

If the atoms in question contain an even number of protons, neutrons, and electrons, then they behave as bosons, and at a particular critical temperature and density, they will suddenly "condense" into a single quantum state (generally the lowest-energy state of the trap holding them), forming a Bose-Einstein Condensate (BEC). If the atoms contain an odd number of protons, neutrons, and electrons, they behave as fermions, and cannot have multiple atoms in the same quantum state, so they form a degenerate Fermi gas, filling all the lowest-energy trap states up to some level determined by the number of atoms. In both cases, the degenerate regime is somewhere in the nanokelvin temperature range, as measured by the average energy of the trapped atoms.

Once you get matter to these ultralow temperatures, what do you do with it? There are a bunch of different things you can do once you have an ultracold gas of atoms, and in fact, this sub-sub-field is the largest single area at DAMOP, in terms of the number of invited sessions.

Probably the most active area of study in quantum degenerate systems draws heavily on the field of condensed matter physics. In some sense, it consists of doing the experiments that condensed matter physicists wish they could do: cold-atoms systems provide a degree of control that is nearly impossible to achieve with real solids and fluids. You can vary all sorts of parameters-- particle density, interparticle interactions, potential structures, etc.-- without needing to do any complicated materials processing. There are several different approaches to condensed matter with ultracold gases:

Ultracold Atoms in Optical Lattices: In condensed matter systems, physicists are generally concerned with the motion of electrons inside of a solid. Most of the time, this solid can be thought of as a fixed periodic lattice of ion cores, with electrons moving around through the lattice. The electrons are attracted to the ions making up the lattice, but not so strongly that they are necessarily bound to a single atom; the exact details of how they move is then determined by the arrangement of the atoms.

In ultracold atom systems, you can create an optical lattice that is a periodic array of points in space that the atoms are attracted to. This system allows you to make cold atoms play the role of electrons, and there are a huge number of groups out there working on different aspects of this problem. The particular session I highlighted in my talk involved In-situ Imaging of Ultracold Atomic Gases, and I've written about a few of these experiments before: Markus Greiner's group at Harvard did a nice investigation of the insulating phase transition, and Immanuel Bloch's group did some similar work watching insulators melt. Bloch gave a very nice talk at DAMOP on a new paper where they manipulate the states of individual atoms in a lattice, allowing them to write arbitrary patterns in the lattice:

They also do a really cool thing where they clean out all but a single line of atoms, and then watch the atoms "tunnel" from one lattice site to another. This is essentially like studying the physics of electron transport by watching individual electrons move from place to place inside a solid, which is pretty incredible when you think about it.

Tuning Interparticle Interactions: The two most important characteristics of electrons, for condensed matter physics, are their spin statistics and their charge. electrons are spin-1/2 particles, which makes them fermions, not bosons, and they are charged particles, so they interact very strongly with their surroundings. The interaction between electrons inside a solid is modified by the presence of the lattice, though, so different solids have different effective electron interactions. Under the right conditions, these interactions can even become effectively attractive, causing the electrons to "pair up" to form composite bosons. This is the key step for superconductivity.

In ultracold atom systems, you can "tune" the interaction between atoms by applying a magnetic field, which allows you to make pairs of ultracold atoms and study the physics of this pairing. If you start with an ultracold gas of fermions, rather than bosons, and manipulate the interactions appropriately, you can create something that looks very much like the superconducting transition for electrons: the atoms pair up, forming either relatively tightly bound molecules or relatively loosely bound "Cooper pairs," and the resulting pairs can form a superfluid.

This was the big topic a few years ago, and while it's no longer the absolute cutting edge, it's still a major area of research, and there was a focus session at DAMOP on this stuff, including some speculation about ways to make topological states in cold-atom systems, thus merging hot topics in cold atoms with hot topics in condensed matter, creating a state of total March Meeting bliss. Unfortunately, that session was at the same time as something else I wanted to see, so I didn't get to it.

Also in this general area is the study of "Efimov states," which I have never actually understood, but which got their own session (again, at the same time as other cool stuff that I understand better). These are related to three-particle interactions, and get a bunch of people very excited. The tunability of interactions in these systems also has some connections to the AdS/CFT business that's all the rage in high-energy theory, but I didn't see anything about that at DAMOP this year.

Magnetism: Another big area in condensed matter physics, shading into thermodynamics, is the study of magnetic systems. One of the standard theoretical tools for this is the "Ising Model," where you imagine a lattice of spins that can be either up or down, and an interaction between neighboring spins that slightly favors them having the same state. In this sort of system, you see the formation of magnetic "domains," small clumps of spins that all have the same state, which is more or less what happens in real magnetic materials. These domains grow and shrink as you change the temperature, and you see dramatic phase transitions at some points. Ising model systems are an incredibly rich system for the study of thermodynamics.

Cold atoms also give you a way to study these kind of systems in detail, because with a clever choice of atoms, you can get large values of spin angular momentum, and thus large magnetic moments for the atoms. If you throw a bunch of these atoms in a lattice, they behave like the spins in an Ising model. There were a bunch of sessions about this at DAMOP, and it will probably be a major topic for a while yet. There's also this recent paper, which I may try to write up soon.

There were several other sorts of cold-matter sessions at DAMOP, which mostly involve using ultracold atoms as sources for something else, typically either precision measurement or quantum information processing. One session on Atom Circuits was basically about ways to build a SQUID (not a squid, a SQUID) using cold atoms. Another session was about efforts to make Ultracold Molecules, which have huge potential applications in things like electron EDM searches. A session with the puzzling title Non-Equilibrium and Cooperativity in Cold Atoms seemed to mostly deal with potential quantum information applications, and I'm not at all sure what the Synthetic Gauge Fields stuff is about. Finally, there was a session about Cold Rydbergs, involving ultracold atoms that are then excited to very high states, and can even form plasmas, which was one of the papers I was on in grad school, and a topic I continue to follow (though, alas, it was opposite the Hot Topics session, so I didn't make it).

The bulk of the activity in ultracold matter is really in the condensed matter/ thermodynamics end of things, though. If you want to understand why it's an exciting field, those are the topics to look at.

Names to Conjure With: If you want to learn a bunch about this stuff quickly, or just want to sound informed, there are a handful of research groups to pay close attention to.

For optical lattice stuff, the two biggest names at the moment are probably Markus Greiner at Harvard and Immanuel Bloch at the Max Planck Institute in Munich. There are also three major institutes to watch: the AMO Physics group at JILA, including Eric Cornell and Deborah Jin; the Center for Ultracold Atoms at MIT/ Harvard, including the aforementioned Markus Greiner and Wolfgang Ketterle, and the new Joint Quantum Institute at NIST/ Maryland, including Bill Phliips, Steve Rolston, Trey Porto, and Ian Spielman. There are lots of other ultracold atom groups out there, but those groups have been involved in just about everything interesting at some point, so if you keep track of what they're up to, you'll have a good idea of what's interesting in ultracold matter.