[[extropy-chat] diffraction limit

[Dan Clemmensen]

scerir wrote:

>Transistors and other features in chips 
>are made using optical lithography,
>where light etches out patterns on 
>a photosensitive substrate on silicon.
>The minimum feature size possible is 
>equal to the wavelength (lambda) of 
>the light used. This is the "diffraction 
>limit".
>  
>
This is not true, and has not been true for at least ten years. The 
current technology uses 193nm light to achieve 90nm features. This is 
done by using phase-shift masks. a simple mask is called a "binary 
mask."any given part of the mask is one of two states: opaque or 
transparent. The pattern on the mask is exactly reproduced on the 
substrate when the light shines through it. A phase-shift mask works 
differently. The mask has a continuum of translucency from opaque to 
transparent, and is works by creating an interference pattern that is 
the desired pattern of the result. creating a phase-shift mask is a 
black art, so a mask set for a complex semiconductor can cost as 
$250,000 or more.

>If lambda is the (de Broglie) wavelenght
>of an individual photon, using N entangled
>(correlated) photons it could be possible
>to imagine to reduce the "diffraction limit" 
>and also the minimum feature size of chips, 
>since, in this case, the wavelenght might 
>be lambda/N.
>
I don't know if you can do this "in bulk." if you must use a single 
beam, you lose the advantage of parallel simultaneous exposure, and the 
exposures then take too long for efficient production. This is why 
electron-beam lithography is not used for production. The wavelength of 
an electron is quite small, so an electron beam can "draw"finer lines 
than a light source can. E-beams are used to make the phase-shift masks. 
The masks are then used to make (many, many) semiconductor devices.

>  
>
>And in fact ....
>http://www.arxiv.org/abs/quant-ph/0312197
>http://www.arxiv.org/abs/quant-ph/0312186
>
>Any killer application in microscopy?
>Interferometry? Astronomy? (Cosmology?).
>
>_______________________________________________
>extropy-chat mailing list
>extropy-chat at lists.extropy.org
>http://lists.extropy.org/mailman/listinfo/extropy-chat
>
>  
>