Tuesday, March 11, 2008

March APS Meeting II

The March Meeting continues. Other topics that seem relatively hot (based on the number of abstracts) compared to previous years include thermoelectrics and ultracold gases and fluids. The latter are really at the border between condensed matter and atomic/molecular/optical physics, and it's interesting to see the merger of the two disciplines. While the ultracold gases provide an exquisitely clean, tunable environment for studying some physics problems, it's increasingly clear to me that they also have some significant restrictions; for example, while optical lattices enable simulations of some model potentials from solid state physics, there doesn't seem to be any nice way to model phonons or the rich variety of real-life crystal structures that can provide so much rich phenomenology.

Anyway, I saw some very pretty talks today. Taking the prize for coolest graphics in a presentation were definitely two talks from an invited session on Kondo physics. The first was by Andreas Heinrich, giving an overview of the IBM Almaden's use of scanning tunneling microscopy to examine magnetic anisotropy and Kondo physics on the single atom level. The second was by Hari Manoharan of Stanford, who showed around three experiments, the most elegant of which involved using STM of magnetic atoms to demonstrate that sometimes it's possible to really extract phase information about superpositions of quantum states. Basically he showed that one could make a designer system (an elliptical corral that confines the Cu(111) surface states) and then use STM spectroscopy based on the Kondo properties of Co atoms on the Cu(111) surface to identify specific superpositions of the eigenstates of that corral.

Another interesting series of talks took place in a session that I organized, where Lindsay Moore of the Goldhaber-Gordon group at Stanford discussed some recent studies of the so-called "0.7 anomaly". In 2d electron gas, it is possible to use gates to create a 1d constriction for a small number of electronic modes. This is called a quantum point contact (QPC). In zero magnetic field, as the point contact is pinched off the conductance of the QPC drops in quantized steps of 2e2/h until it falls to zero. The 0.7 anomaly is the appearance of an extra plateau in the conductance at around 0.7 x 2e2/h. People have been bandying about possible explanations for this feature for a while now, and finding new probes to apply is a popular tactic. The following contributed talk was by Alex Hamilton from UNSW, who had looked at the 0.7 anomaly in 2d hole systems. The holes have strong spin-orbit scattering effects that, through the study of response to applied magnetic fields, allow one to demonstrate convincingly that the 0.7 anomaly clearly has some mechanism related to spin. Nice.


CarlBrannen said...

Hah! Thanks for the 0.7 anomaly information. I have a suspicion that this is just what one would expect when working in a Pauli MUB (mutually unbiased bases) reduced to two dimensions.

The basic idea is to model the electron scattering in density matrix form where one takes into account four possible interactions.

Because it is 2-d, one eliminates spin in the z direction from the MUB, leaving spin +-x and spin +-y. There are four possible beginning states and four possible ending states. It's easy to solve the equations because there is no quantum (or Berry) phase; you need 3-d to get a geometric phase.

If it works out, I'll type something up on my blog maybe tonight.

CarlBrannen said...

Sure enough, one can use density matrix theory to split the single electron scattering to get a conductance near 0.7. Looking at it more carefully, I think it's purely a spin effect and doesn't have anything to do with the fact that the electrons are kept in 2-d. The spins can go in any direction. The blog post is here.

Thanks, this was highly entertaining. That the effect shows up in such a basic, supposedly easily understood situation gives pretty good evidence that its explanation should not require physics piled very high and deep. Surely that's an imitation to amateurs.

okham said...

Maybe it's because I am a theorist but I hold somewhat of a differing view on optical lattices (OLs).
The main thing that I really like about them is precisely the fact that they allow (in principle, at least) one to investigate systematically those interesting physical effects that in real-life materials are often difficult to disentangle from complications related to crystal structures, defects of all sorts, fabrication issues etc. Very often I feel that the "rich variety" is more a source of confusion than else.

Case in point: the Supersolid phase of matter. I would not be surprised in the least if it could be observed more cleanly and easily in OLs before the community ever reaches an agreement on whether Helium-4 is indeed a supersolid.
As for phonons: I am no expert but I would imagine that a model such as the Holstein's (see here,for instance) should be realizable in OLs. Granted, it is a fairly crude representation of "phonons" but again, I tend to regard the simplicity as an asset, more than a liability.

Doug Natelson said...

Okham - I'm all in favor of simplicity, and I get the point of trying to look at real implementations of, e.g., the t-J model. I just wanted to point out that there are some very interesting systems where the coupling to structural degrees of freedom plays a role, and I think that modeling such systems in optical lattices is likely to be very hard. That's ok though. No experimental tool is perfect for every problem.