Nanotechnology Progress

John K Clark (
Mon, 18 Nov 1996 20:51:40 -0800 (PST)


The reason that the magnification of a light microscope is so limited is that
the wavelength of light is large. The obvious solution is to use light of a
shorter wavelength, because the resolution is equal to the wavelength of the
light times the focal length of the lens divided by the aperture of the lens.
The difficulty is that when you go beyond the near ultraviolet there didn't
seem to be any way to make a lens. Aluminum is almost as transparent to hard
(over 14 kev) X rays as glass is to light, but the index of refraction is so
tiny that the focal length of a simple lens would be huge, far too long to be
practical. Heavier elements like Gold have a higher index of refraction, but
they strongly absorb X rays, your lens would be almost completely opaque.

This is also a serious obstacle to using X rays in photolithography to make
very tiny semiconductor devices. If you don't have any optical elements then
you can't optically reduce the size of your image, your mask must be as small
as the very microprocessor you're trying to make, and you must place it right
on top of the silicon, as in a contact sheet in photography. None of this
seemed very workable to most people.

In the November 7 1996 issue of Nature is an article called "A Compound
Refractive Lens For Focusing High Energy X Rays" by Snigirev, Kohn, Snigireva
and Lengeler. They found a easy and cheap way to focus very Hard X Rays,
up to about 40 kev. Their lens is simply an Aluminum block with holes of
radius .3 millimeter drilled into it, it looks like this:

---------------> | o o o o o o o o o |
X Rays | o o o o o o o o o | . <--- Focal Point
---------------> | o o o o o o o o o |
---------------> |-------------------|

Each hole drilled into the Aluminum block is a lens, one lens may give you a
ridiculously long focal length, but 2 will cut that distance in half,
this first device has 30 lenses. The focal length of their compound lens
is R/ 2NI, R is the radius of the holes, N is the number of holes, and I is
the index of refraction in the X ray region of the material. In this lens the
holes are cylinders so the X rays are focused in only 2 dimensions, for 3
you'd need spheres not cylinders. The authors don't seem to think it would be
very difficult to make a device with hundreds, perhaps thousands of such
lenses. The authors also calculate that if they could make the holes in a
parabolic shape they could reduce the focal length by a further factor of 5.
Although they haven't tried it yet they also think the same lens could also
focus something else that was thought to be impossible, a beam of neutrons.

Aluminum was used in this case because it's easy to machine, but boron,
carbon, plastic and even water should work too. It occurs to me that it might
not be too difficult to make an array of small spherical bubbles in water,
they wouldn't stay in place for long but they wouldn't need to, the X rays
from a laser would be in a VERY short burst.

John K Clark

Version: 2.6.i