@S1-A.ARPA,@MIT-MC.ARPA:john%taveis.DEC@decwrl.ARPA (10/01/85)
From: john%taveis.DEC@decwrl.arpa
I recently came across an old idea which might be a good candidate for
microgravity manufacture. It's the Luneberg Lens, invented in the early
sixties by a Professor Luneberg of Berkeley. He described it in a textbook
on mathematical optics, and it's also described (where I first read about it)
in "The Optics of Non-Imaging Concentrators" by Welford and Winston.
For a long time people have been trying to achieve a perfect optical
system, one free from any kind of aberration. By using more lens and mirrors
and more complicated shapes, they've been able to do better and better, but
some kind of distortion is always there. It seems, in fact, that a perfect
system cannot be achieved with a finite number of elements, although this has
not been proven. However, James Clerk Maxwell (the
EE student's bane) came up with a solution in the 1850's, called the Fisheye
Lens. Unfortunately, it needed a medium with a continuously variable index
of refraction (n), and both the object and image had to be immersed in the
medium.
Luneberg expanded on Maxwell's work. He found a scheme where a perfect
image could be produced of an object at infinity, with both the image and
object in air (i.e. n=1). His lens is a sphere with an index of refraction
that varies with the distance from the center of the sphere (r) as
n(r) = (2 - r^2 / a^2) ^1/2 r < 1
= 1 r > 1
where 'a' is a constant.
The varying index was thought to make the lens impractical. However,
n can be changed by doping glass with various impurities, and in fact this
is done regularly in fiber optics. How, though, can this be done for a
sphere instead of a fiber?
By building it in weightlessness. The sphere would float in the middle of a
vacuum chamber. Glass would be deposited on it one layer at a time, with each
layer having the appropriate index. The glass vapor would flow into the
chamber continuously, and its doping would vary continuously. The
weightlessness would give perfect spherical symmetry. Glass deposition is a
standard feature of semiconductor processes; equipment for it is readily
available. Building up a sphere of any size, however, might take some time.
A perfect optical system, though! That could be something big. If anyone
out there is involved in either optics or space industry they might want to
check it out.
John Redford
DEC-Hudson
Fri 27-Sep-1985 20:35
Sat 28-Sep-1985 10:35 rdp@teddy.UUCP (10/03/85)
In article <3725@mordor.UUCP> @S1-A.ARPA,@MIT-MC.ARPA:john%taveis.DEC@decwrl.ARPA writes: > > Luneberg expanded on Maxwell's work. He found a scheme where a perfect >image could be produced of an object at infinity, with both the image and >object in air (i.e. n=1). His lens is a sphere with an index of refraction >that varies with the distance from the center of the sphere (r) as > >n(r) = (2 - r^2 / a^2) ^1/2 r < 1 > = 1 r > 1 > >where 'a' is a constant. > > The varying index was thought to make the lens impractical. However, >n can be changed by doping glass with various impurities, and in fact this >is done regularly in fiber optics. How, though, can this be done for a >sphere instead of a fiber? > > By building it in weightlessness. The sphere would float in the middle of a >vacuum chamber. Glass would be deposited on it one layer at a time, with each >layer having the appropriate index. The glass vapor would flow into the >chamber continuously, and its doping would vary continuously. The >weightlessness would give perfect spherical symmetry. Glass deposition is a >standard feature of semiconductor processes; equipment for it is readily >available. Building up a sphere of any size, however, might take some time. > Several optical manufacturers have been doing research along these lines, with names like Canon and Nikon coming to mind. In fact, variable refractive index lenses have been done for some time, under normal, gravity-laden conditions. The technique involves taking a lens blank and "cooking" it in a silver-salt (silver halide) soup for some time. Appropriate ions diffuse into the glass from the surfaces at a rate determined by, among other things, concentration, glass characteristics, temperature, surface area, and so forth. The technique has been suffiently perfected to make it commercially (albeit expensively) feasable, although I am not specifically aware of actual products that utilize this technique. I myself developed a simple technique for lower the diffraction of so-called "difraction limited" optics, thus raising the resolving power, but I will not digress because a) this is probably the wrong news group, b) It might be something really neat, and patentable (but I don't think so), and c) I don't feel like it right now :-). Dick Pierce
henry@utzoo.UUCP (Henry Spencer) (10/04/85)
> [How can one make a lens with continuously-varying index of refraction?...] > By building it in weightlessness. The sphere would float in the middle of a > vacuum chamber. Glass would be deposited on it one layer at a time, with each > layer having the appropriate index. The glass vapor would flow into the > chamber continuously, and its doping would vary continuously. The > weightlessness would give perfect spherical symmetry... One very serious problem that I can see is crystallization, also known to the glass community as "devitrification". Glass is an amorphous solid, essentially an extremely viscous liquid. But most glass-forming materials will form crystals as well. Laying it down from vapor strikes me as a good way to get a mass of polycrystalline junk rather than smooth glass. My understanding is that the semiconductor people do *not* lay down glass from vapor; they oxidize the silicon surface to produce it. Possibly devitrification can be suppressed by careful control of conditions. This is the sort of detail that is highly proprietary, so it's hard to say much without being an insider. But the problem is serious, perhaps fatal. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry