New Nanophotonic Waveguides developed at MIT

SimilarityEngine writes "Physicists at MIT have developed a new kind of nanophotonic waveguide, overcoming 'several long-standing obstacles' to move us one step closer to optical computers. Photonics strives to improve on electronics in terms of space requirements, speed and energy consumption - but until recently, it was not possible to perform well in all three areas simultaneously. In their paper, John Joannopoulos et al. demonstrate how to reliably encode a signal as surface plasmons which side-steps such limitations."

Comments

MIT

[superpulpsicle]

Why does MIT always take full credit? There is at least a dozen organizations that contributed to the pre-research heading to this. Just look at the pdf if you don't believe.

The cool signifigance of this

[tempest69]

Ok, IANAQP (quantum-physicist), so I am going to butcher this horribly. From what I get from the article is that they are able to pipe photons distances less than their wavelength. by the use of surface plasmons http://en.wikipedia.org/wiki/Plasmon this lets them use longer wavelenght light (ie visible spectrum> with very small parts. The use of short wavelength light (x-rays) can break covalent bonds of the chip, causing the chip to fail.

So pretty much this means that optic technology has made a nice stride in catching up to classic electronic technology. But we wont be seeing the Pentuim-Optic and day soon.

Storm

Re:Can somebody please clear this up?

[Hartree]

Also, how does stacking the insulator layers increase the bandwidth?

A common trick to make extremely small solid state devices more broadbanded is to put a number of them close enough together so that they interact a bit.

It's kinda like hooking multiple springs together. Each has its own prefered vibration frequency, but when they're hooked together, the system can vibrate at not only those original frequencies, but also other ones as well.

When you have very large numbers of interacting springs, the ranges of frequencies allowed tend to smear out. This effectively means that the device can handle a whole range of frequencies, not just the original one.

Just as springs have vibrations, these plasmons are vibrations. Just like springs have preferred frequencies to vibrate at, these waveguides have preferred frequencies. You put the waveguides close enough that they interact a bit, and it tends to allow more frequencies. Stack up a whole bunch of the waveguide layers, and you can smear out the response so that it's more broadbanded.

This sort of thing is done in a lot of systems based on vibrations.

(There are more details to it, but that's the general idea, and the best I can do at the moment with a cold making it hard to think. ;)