Perfect image without metamaterials... and a reprieve for silicon chips (w/ Video)

September 29, 2009

(PhysOrg.com) -- Since 2000, John Pendry's work on metamaterials has been at the van guard of efforts to create a perfect image - images with perfect resolution that can stem from light being moved in odd directions to create, among other tricks of the light, the illusion of invisibility.

One exciting development was Pendry's theoretically perfect work on negative refraction, which offered the possibility of lenses that could create images with resolution not possible with conventional lenses. But this proved problematic in practice as the negatively refracting materials so far produced did not live up to their potential - absorbing a certain amount of the light and spoiling the resolution of the perfect image.

In a new research paper published today, Tuesday, 29 September, in New Journal of Physics called 'Perfect imaging without negative refraction', Ulf Leonhardt, Chair of Theoretical Physics at the University of St. Andrew's, has shown that there is another way to create the perfect image.

Inspired by James Clerk Maxwell's findings, first expounded in the 1850s, Leonhardt is reintroducing the idea of a 'fish-eye' lens; a lens that can work in any direction but had not, until now, been modeled to fully account for the wave-like properties of light.

Professor Leonhardt said, "It is the waviness of light that limits the resolution of lenses. Apparently, nobody had tried to calculate the imaging of light waves in Maxwell's fish-eye. The new research proves that the fish-eye has unlimited resolution in principle, and, as it does not need negative refraction, it may also work in practice.

"The theory was inspired by ideas for invisibility where light is bent around objects to make them disappear from view. Here the ideas behind invisibility are applied for perfect imaging."

While the work is only theoretical at present, it will be exciting news to silicon chip manufacturers as the resolution limit of lenses limits the microchip technology needed for making ever faster computers.

While this development will not overcome the problems posed by the physical limits of smaller and smaller chip circuitry, it will give chipmakers freedom to photograph ever smaller, and more compact, structures of billions of tiny transistors on silicon chips to meet the insatiable appetite for faster and smaller computers.

More information: Journal paper: http://stacks.iop.org/NJP/11/093040.

Source: Institute of Physics (news : web)

• guiding_light - Sep 29, 2009
  • Rank: 3.8 / 5 (4)
  http://books.goog...xwell's fish-eye&f=false
  
  The maxwell fish eye is itself a theoretical construct, ridiculous to find a professor take it so seriously.
  • report abuse
  • • Current rank
    • 1
    • 2
    • 3
    • 4
    • 5
• Husky - Sep 29, 2009
  • Rank: 2.3 / 5 (4)
  with nano waveguides, it could become a practical construct as well
  • report abuse
  • • Current rank
    • 1
    • 2
    • 3
    • 4
    • 5
• guiding_light - Sep 29, 2009
  • Rank: 4 / 5 (3)
  therein lies the rub, with a waveguide, the wavelength enters the picture.
  • report abuse
  • • Current rank
    • 1
    • 2
    • 3
    • 4
    • 5
• guiding_light - Sep 29, 2009
  • Rank: 5 / 5 (4)
  equation (26) of the paper (I think it is publicly accessible) shows the image of an extended object becomes quite distorted (x,y)->(-x/(x^2+y^2),-y/(x^2+y^2)). In lithography, we need depth of focus. This doesn't have it.
  • report abuse
  • • Current rank
    • 1
    • 2
    • 3
    • 4
    • 5
• nkalanaga - Sep 29, 2009
  • Rank: 2.3 / 5 (3)
  If the distortion is predictable it may not be a problem. Design the original with the proper pre-distortion, and the imaging distortion will give the desired result.
  
  Obviously this won't work for general photography, where you don't have control of the original object, but it should for chip making.
  • report abuse
  • • Current rank
    • 1
    • 2
    • 3
    • 4
    • 5
• guiding_light - Sep 29, 2009
  • Rank: 4 / 5 (3)
  If you are using this fish eye lens as the final imaging lens, it won't work for lithography for chip making. x-> -1/x is not imaging. You cannot embed a fish eye inside another, it destroys the point-to-point mapping setup. If you are not using this as the final lens, then you are not improving the resolution.
  • report abuse
  • • Current rank
    • 1
    • 2
    • 3
    • 4
    • 5