World's smallest laser cracks open the door to THz CPU race

So you thought 100nm was about as narrow as lasers could get, huh? Well think again brother, because scientists at Norfolk State University have now demonstrated a 44nm 'spaser' that performs a laser's functions by the alternative means of surface plasmons. By using such an unorthodox technique, the researchers have been able to overcome the minimum size limitation to lasers, and they even claim spasers could be made as small as 1nm in diameter. Peeking into the (not too near) future, this could improve magnetic data storage beyond its current physical limits, and even lead to the development of optical computers that "can operate at hundreds of terahertz" -- and here you were, thinking that your brand spanking new Core i7 system with Blu-ray was future-proof.

[(Unverified) Posted Sep 2nd 2009 3:29AM]

While Killeryo2002 is completely right in how awesome optical computing would be, this research has jack squat to do with that. First of all, this 'laser' was a silica ball with a gold center about 44nm across illuminated by a pump (normal size) laser beam. Secondly, their evidence for lasing is very weak. If you read the paper and just glance at figure 4b, you see that they only have 2 data points (out of merely 5 total) which indicate lasing. One hallmark of a laser is that it crosses a threshold where it quickly goes from nothing to a whole lot of something with a small increase in pumping power. You can't really establish the existence of a threshold without more data points.

Secondly, the spectrum of a laser should rapidly narrow as this threshold is crossed. If you look at fig 4a in their paper you can see that although the peak gets higher, the average width is actually pretty constant.

Sorry to be a downer guys, but I work in the optics field and what these guys say they see is poorly supported by the data and could be explained by a lot of other phenomena.

Unless you are at a university, you probably don't have access to the paper, but if you do, "Don't take my word for it" : http://www.nature.com/nature/journal/v460/n7259/pdf/nature08318.pdf