Tapering Waveguide Captures a Rainbow

SubComdTaco passes along news of researchers in the US who have trapped a rainbow in a tapering waveguide. The research is described (PDF) on the arXiv. "In 2007, Ortwin Hess of the University of Surrey in Guildford, UK, and colleagues proposed a technique to trap light inside a tapering waveguide [made of metamaterials]... The idea is that as the waveguide tapers, the components of the light are made to stop in turn at ever narrower points. That's because any given component of the light cannot pass through an opening that's smaller than its wavelength. This leads to a 'trapped rainbow.' ... Now Vera Smolyaninova of Towson University in Baltimore, Maryland, and colleagues have used a convex lens to create the tapered waveguide and trap a rainbow of light. They coated one side of a 4.5-mm-diameter lens with a gold film..., and laid the lens — gold-side down — on a flat glass slide which was also coated with film of gold. Viewed side-on, the space between the curved lens and the flat slide was a layer of air that narrowed to zero thickness where the lens touched the slide — essentially a tapered waveguide. When they shone a multi-wavelength laser beam at the... gilded waveguide, a trapped rainbow formed inside. This could be seen as a series of colored rings when the lens was viewed from above with a microscope: the visible light leaked through the thin gold film."

Re:What happens when the laser is turned off?

[jandrese]

Obviously if you can see it the rainbow isn't completely contained. You can't capture a rainbow in the manner that you're thinking of because it would require a perfect vacuum (which we can mostly achieve these days), and a perfectly reflective surface (which we cannot). Every time the light bounces off of whatever you have contained it in, it will lose a bit of energy. Since it's traveling at the speed of light, you'll have enormous numbers of bounces per second and they'll quickly sap all of the energy away from the beam.

Re:Also neat because?

[Ambiguous Coward]

(Optical computing was the answer.)

Nothing new here, move along...

[ctrl-alt-canc]

Selective absorption is a well known effect that takes place whenever a wave propagates in a medium where two boundary conditions have to be fulfilled at once. We observe it regularly in our lab while sending acoustic/elastic waves into a pack of slabs of material. The same thing happens with electromagnetic waves, just like Isaac Newton observed a few centuries ago [harvard.edu]. Sending the light in a direction parallel to the lenght rather than perpendicular does not discover anything new. Next post, please...

Newton's Rings

[Guppy]

Sounds like an old high school science experiment. Take two microscope slides (flat pieces of glass) lay one on top of another with a thin shim separating them at one end, illuminate this with a monochromatic light and see the fringes. With white light, the peaks for each wavelength would occur at different locations, resulting in a 'rainbow'.

What you're referring to is known as "Newton's Rings":
http://en.wikipedia.org/wiki/Newton's_rings [wikipedia.org]

Re:What happens when the laser is turned off?

[dexmachina]

As far as the reflection losses go, it's not converted to anything, it's transmitted. That's what the GP meant by "if you can see it the rainbow isn't completely contained". There's no such thing as a perfect reflector, some of the light is always transmitted through. And since we can't attain perfect vacuum, there will also be internal collisions with gaseous molecules, which can either transmit the absorbed energy via heat in colliding with other molecules, or re-transmit it as light, though possibly in a series of longer wavelengths.

As for the solar cells thing... no. That's a completely different situation. The trapped rainbow is a nearly closed system, with no continuous energy input, and the problem is that we can't make it completely closed (and if we could, it's internal entropy would then increase over time so it still couldn't be perfectly stable). Technically, I suppose all the energy can't be sapped since it's exponential decay, but the system energy asymptotically approaches 0 (and once a small enough amount remains, the fact the energy is discrete becomes important). It's about inefficient energy conversion. Far form implying that we could create 100% efficient solar cells, this is why we can't create 100% efficient solar cells.