It must be 10 years since John Pendry at Imperial College London dreamt up the idea of superlenses. Until then, physicists had thought that the resolution of all lenses was limited by a phenomenon called the diffraction limit, which holds that you can’t see anything smaller than about half the wavelength of the illuminating light.
That’s true if you look at the propagating component of light waves. But light also records smaller sub-wavelength details in its evanescent components, which do not propagate. At least not usually. What Pendry showed was that evanescent components can propagate in a material with a negative refractive index, and he pointed out that a thin film of silver ought to have just the right properties.
Since then, the race has been on to build superlenses. In 2005, Nicolas Fang at the University of Illinois at Urbana-Champaign created one that could record details as small as one-sixth of a wavelength. That was a significant improvement over the diffraction limit, but why not better?
It turns out that silver films just a few tens of nanometers thick are extremely difficult to make. On this scale, silver tends to clump into islands, like water on plastic, making the film rather irregular. This dramatically reduces the ability of evanescent waves to propagate.
Now Fang and a few buddies, including Stan Williams at HP Labs, in Palo Alto, CA, have worked out how to make thin silver films smooth. The trick is to grow the silver on a layer of germanium, which forces it to form a smooth thin film.
This new lens is a huge improvement. With a record-breaking resolution of one-twelfth of the wavelength of light, it opens up an entirely new area of imaging when extended to the far field, a feat that can be achieved by gluing a corrugated silver surface on top of the superlens, says the team.
And greater resolution should still be possible: the theoretical limit is one-twentieth of a wavelength.
Fang and co conclude with the dramatic prediction that these superlenses should make it possible to film molecules in action in real time with visible light.
That ought to be one impressive movie.
Ref: arxiv.org/abs/0906.1213: Molecular Scale Imaging with a Smooth Superlens