While electronic microscopes are still costly, you can build a scanning-tunneling microscope using nothing but a couple dollars' worth of parts from Radioshack and Wal-Mart. A student at my old high school a few years ago made one for a science fair project; it only created a one-dimensional image (he didn't have time to extend it to full 2D) but it worked.
How it works: a piezoelectric material, such as that found in speakers, can be manipulated precisely enough to move the head of a scanning tunneling microscope. The "head" of the microscope can be made by using a pair of extremely bad Wal-Mart scissors and cutting a piece of wire; this will usually create a very rough cut with a tiny slice of wire that extends beyond the cut, ending at atom thickness.
This wire can then be strapped to the piezoelectric material, and now all you need are some basic electronics to power and run the thing. Of course, this part is a lot of work, but it doesn't require any fancy expensive hardware.
For now. It's a wonderful service and I hope it continues for a long time, but you shouldn't depend on it to save any particular thing, and it's much harder to find things in archive.org than on the open web. I'm still grieving over the loss of "Networked PostScript Printer Installation as Illustrated Through Interpretive Dance".
The STM project is clever and well-known. A traditional electron microscope, on the other hand, is of much more use to the would-be reverse-engineer/DRM fighter. Hence my interest in widening their availability.
Hmm, yeah, I guess you need to be able to electrify your entire sample in order to look at it with an STM. How much current do you need? Could you sputter silver onto the surface of the chip or something?
"A new, advanced model of the electron microscope, much lower in cost, less complex and only 30 inches high, has been developed by RCA and will be placed on the market later this year."
Ha. That link doesn't show you could buy miniature electron microscopes in the 50s. It shows that pre-release hype for vaporware was alive and well in the 50s.
Scanning tunneling microscopes are relatively easy, scanning electron microscopes (with an electron beam) are a lot harder (if you want any kind of serious magnification).
Getting some results appears to be doable but before you get the resolution and repeat accuracy of a commercial model I think you'll be spending the same amount of money. But you'll learn a ton :)
An STM uses a needle that is only a few atoms across to scan the object under inspection, an SEM uses an electron beam to do the same in a high vacuum chamber.
An STM you can build under $100, an SEM would at a minimum cost you several thousand (and much more if you can not do your own machining).
Maybe. A lot of lab supplies are much more expensive than do-it-yourself versions because:
* they aren't mass-produced,
* the people who are employed to make them are highly paid,
* the supply of grad student labor to substitute for them is limited,
* the required quality is high,
* and the costs of defective goods are high.
So high prices don't necessarily mean that Edmund Scientific is colluding with, I don't know, Siemens, to fix prices at above-market rates. It more likely means that it's damned hard to make a profit selling cheap electron microscopes.
In this case, I suspect that the answer is that it creates a market where none existed before, because the price scientists were willing to pay was lower than the price where anyone could make money selling TEMs or SEMs.
I haven't ever written a grant proposal, but I have the impression that if the committee thinks you'll be wasting the grant money, you're less likely to get the grant. I'd be interested to hear about your grantwriting experiences, especially if they're different.
How it works: a piezoelectric material, such as that found in speakers, can be manipulated precisely enough to move the head of a scanning tunneling microscope. The "head" of the microscope can be made by using a pair of extremely bad Wal-Mart scissors and cutting a piece of wire; this will usually create a very rough cut with a tiny slice of wire that extends beyond the cut, ending at atom thickness.
This wire can then be strapped to the piezoelectric material, and now all you need are some basic electronics to power and run the thing. Of course, this part is a lot of work, but it doesn't require any fancy expensive hardware.