Tuesday, March 17, 2009

March Meeting, days 1 and 2

Two days of the March Meeting are over, and it's been fun so far. We've really been lucky with the weather, and it looks like tomorrow will also be nice. I'm from Pittsburgh originally, and it's rather surreal to be going to a conference in my old home town. At least I still remember where a bunch of restaurants are, though there's been a lot of turnover. (For those at the meeting, if you don't mind a little bit pricey, I recommend the Fish Market attached to the Westin, as well as Eleven. I also had some great inexpensive Indian food at Spices of India.)

Talks have been fun, though this year I don't seem to have had the time to go to as many sessions outside my immediate interests as in years past. I did catch an all-invited session about the recent exciting work on LaAlO3/SrTiO3 interfaces. It's been known for a few years now that the interface between these two insulating oxides can play host to a 2d electron gas, for reasons that are still under fierce debate. My friend Yuri Suzuki gave a very nice talk explaining their tests trying to get to the bottom of this, and my former Bell Labs colleague Harold Hwang described his group's work on trying to tune the density and type of charge carriers at the interface using techniques similar to modulation doping. There were also talks by the team working on using a conducting AFM to pattern conducting regions of this interface, with an eye toward devices. Neat stuff.

I saw some very nice talks about single-molecule electronic devices. It's a shame that the Division of Materials Physics and Division of Chemical Physics were unable to avoid scheduling their focus topics (on largely experimental measurements, and on theory of molecular devices, respectively) in direct conflict. A couple of invited talks that were particularly compelling were the ones by Christian Schoenenberger, N.J. Tao, and Mark Ratner. Schoenenberger gave a great review of his group's work on mechanical break junctions to look at single molecules in a solution environment, as well as their recent experiments using arrays of linked metal nanoparticles as an electronic testbed. Tao also discussed single-molecule break junction measurements, particularly emphasizing measurements of force and breaking processes as a window on effective junction temperatures. He also talked about some new experiments using asymmetric molecules as rectifiers, and single-molecule measurements of inelastic electron tunneling spectroscopy. Ratner gave a terrific overview of different theoretical approaches to the problem of calculating molecular conduction properties. Most importantly, he did a great job emphasizing when certain methods work well, and why - he gave me a new way to think about some of the physics in these systems. I also saw a number of good, short talks about single-molecule experiments, with a particular bounty coming from the Venkataraman group and collaborators. Fun stuff.


NONE said...

Latha Venkataraman is an interesting and rare example of someone who left academia for consulting and came back later to have pretty good success.

The whole field of single molecule electronics is highly controversial though. A lot of work has been published where people did not pay proper attention to quality/structure/purity of contacts or details mechanisms of tunneling/hopping. Now it's hard to know what to believe, so I remain somewhat skeptical of the entire field, perhaps unfairly so.

Douglas Natelson said...

Well, since I work in the area I can't avoid commenting.... It's well established that some of the early molecular electronics work was sloppy and either poorly reproducible or badly interpreted. It was most definitely overhyped as well. Still, over the last several years there really has been substantial progress, both in making the relevant systems and in understanding their physics. Clearly I have my personal opinions about which results I think are the strongest, but in general there has been a vast improvement. Latha's work is extremely nice - I particularly like how data acquisition and analysis is entirely automated, so that there's no chance of cherry-picking particular pieces of data. There's definitely something to be said for histogramming *everything*.