Slashdot Mirror
An Electron Microscope For Your Home?
CuteSteveJobs writes "Could microscopy be in for a new golden age? Wired previewed the desktop-sized Hitachi TM-1000 Electron Microscope a while back. Light microscopes can magnify up to 400X (1,000X at lower quality) — just enough to see bacteria as shapes — but this one offers 20X to 10,000X, giving some amazing pictures. Unlike traditional electron microscopes, this one plugs into a domestic power socket and specimens don't need any special preparation; it's point-and-shoot, much like your typical digital camera. So easy a grade-schooler could use it, and earlier this year that's what happened: The kids at Iwanuma Elementary School in Miyagi, Japan got their own electron microscope. At $60,000, you'll have to give up on the BMW, but the hope is with economy of scale (so far 1,000 have sold) and miniaturization, the price will continue to drop. The only bad news? It runs XP."
It could have been worse. It could have run Vista.
Otherwise, an interesting development.
there is no spoon
Really, I'm thinking 60k is "payable" if you're really into the stuff.
I have some knowledge of the microscopy hobby. Also ham radio. Both are similar in that ultra high cost options are available, and similar in that prices don't drop, at least not like prices in the computer/electronics hobby.
It is considered "normal" to buy a $1000 radio or microscope, use it for a few years, sell for about $800, upgrade to the $2000 model, use it a few years, sell for maybe $1800, buy the $3000 model ... repeat for a few decades, next thing you know, "old" people of rather average income are operating $10000 of radio gear, $20000 telescopes, cameras, microscopes, etc.
This is very difficult for computer people to wrap their heads around, since last years video card is merely a paperweight today, etc. And vice versa, good luck convincing a ham radio guy that his five old PC will not sell for even 50% of its new price.
If computer prices were this stable, I'm sure I would easily have a $60K computer system by now.
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
FYI Slashdot, one of this decade's genuine breakthroughs in science has been finally breaking the diffraction limit for visible light microscopy. The results in the past couple years alone have been nothing short of stunning. Specifically the techniques which are capable of doing this are confocal microscopy, near-field scanning microscopy, stimulated emission depletion microscopy, stochastic optical reconstruction microscopy and structured illumination microscopy. All of these techniques use visible light and can image at below the diffraction limit of ~250nm resolution, but most use complicated techniques using lasers etc. to do so. Except that last one, structured illumination. This technology is going to literally revolutionize microscopy and probably biology as a whole in the coming years. It is a very clever technique and produces unbefuckingleivably amazing images. With it, full 3D reconstructions of individual living cells with ~10 nanometer resolution, at frame rates in the several Hz range can be taken using a relatively simple LCD retrofit to a high quality transmission light microscope which is installed between the light source and the stage. Look at some of these movies taken of cell processes using the technique and try to keep your jaw off the floor. While the resolution may be higher, none of this is possible with SEM or TEMs due to the necessity of imaging in vacuo.
- "Hear that?! The percolations are imminent! Cease your ingress!"
Since I seem to oscillate between looking at TEM images and doing confocal microscopy these days, allow me to chime in here.
Apart from resolution, there are two factors that make EM useful in a way that light microscopy techniques never will be. Namely, that they allow you to look at an entire specimen at once and at the same time see things you weren't looking for. The pretty color images confocal images are made by tagging particular structures in the cell with fluorescent molecules. This is done either with fluorochrome labeled antibodies or by expressing proteins fused to fluorescent proteins. Therefore, you have to know what you're looking for in order to make a picture of it. Furthermore due to the overlap in the emission spectra of the fluorescent tracers, you can generally only look at about 3 things simultaneously. So if you want to see all the structures in a cell at once or you're looking for something, like a virus, but don't know which one or don't have an antibody for it, EM is still the tool of choice.