As this forum seems to be chock full o' people who know a thing or two or three about microwaves, I figure I'd ask the following questions. Please keep in mind that I'm a laser jock by trade and my practical microwave experience is limited to nuking tv dinners for the most part.
I am working on an inhouse project to build a microwave based Chemical Vapor Deposition system. Basically all that is being done is to use microwave ebergy to dissasociate reactant gas at sub atmospheric pressure to coat a substrate with the thin film of your dreams. I have modeled my efforts thusfar after the work done on microwave based diamond cvd as there is alot of effort put into that field. the first reactors used for this kind of work were called NIRIM systems, after the japaneese agency that developed the reactor design - basically a piece of fused sillica stabbed through a waveguide. This doesnt work well for a number of reasons and cant reach high power density.
The next generation systems basically used an antenna to couple microwave radiation into a cavity, part of witch was under reduced pressure. some of the more novel designs used tricks such as elliptical cavities with an antenna at one foci and the product deposition area at the other to increase power density and remove the plasma from the cavity walls (a major source of contamination).
Essentially all of this work has been done using 2450 or 915 microwaves.
To use the words of my favorite tv cheif - i'd like to kick things up a notch or two. I happen to have in my hot little hands a 20kw source at 18 gig - previously I had been playing with a 100kw source at 915, but since i care about power density and not power I think the shorter wavelength will ahve more pros than cons. What i would like to do is take the output of my klystron to a horn or dish and then refocus the the enegy with an off axis parabolic reflector to a point where as much of that 20kw is concentrated into a region ideally no more than a few cm in diameter... A smallish stainless steel reaction vessel will be located at this focal point and ideally the paraolic reflector will be as 'fast' as possible to keep the e field gradient near the focus as steep as possible to avoid the problem of plasma comming into contact with the chamber walls or pressure window.
now... cosidering my desire for optimum efficency in energy transport from the tube to my product, my desire for highest power density possible, the need to keep plasma away from the chamber walls and the philosophy that this is applied research funded on a small busines budget so therefore must be as cheap as relitivly possible - would anyone care to comment on anything that they see that jumps out that I might be ignoring, not thinking of, alternate suggestions, ect??
Lastly, I am boardering on paranoid when it comes to saftey with 2450 and 915 - i know the military has done some non lethal areal denial weapons testing with mm wave sources - 95 gig iirc - what kind of hazards are present with roughly 18 gig? Will it cook ya just the same as the lower frequency stuff?
Many thanks in advance.
If you would prefer to contact me via email please do so at email@example.com