Building a Scanning Tunneling Electron Microscope

[TheMinister]

Hi All,
Long time no see, but a talk at school today has forced me out of hiding.

Nanotech people from Cambridge talking about err nanotech. Fascinating stuff.
Then this guy mentions we can build one of these microscopes ourselves. With atomic resolution. Oh yeaaah....

http://en.wikipedia.org/wiki/Scanning_tunneling_microscope

Here is a site that seems to know what it's doing (the tip production seems the most useful element of this- the rest of our set up may be a little more Heath Robinson).
http://sxm4.uni-muenster.de/stm-en/
Any ideas how I could measure the size of one of these tips? This might be an important stage in proof-of-concept.

My biggest unresolved problem is vibration dampening. The original designers used levitating magnetic fields, and I've seen suggestions of using rubber pads, springs etc. Has anybody done any similar projects to this, or have any helpful ideas?

Costwise this shouldn't be too bad. The piezo actuators are kinda pricy but that should be it moneywise, and I reckon the school will shell out for those if we ask nicely and show a decent proof of concept.

Help/feedback or anything would be greatly appreciated.

 

[Woo482]

You want to Build a Electron Microscope?

 

[cjp]

I worked on something like that at the university. As far as I can remember, the thing was placed on a heavy stone, hanging on the ceiling with some wires or small steel cables. The tip was moved by using piezo-electric devices.

We weren't really successful in getting atom-size images, but we only had 1/2 day time to do our experiment.

 

[TheMinister]

So some sort of mass dampening system then? Sounds like a sensible idea, I'll look into it.

And its not an electron microscope in the usual sense of the word. Those are generally where a beam of electrons passes through or bounces off a sample, and are ridiculously complex. Here, electrons tunnel across the gap as the needle gets closer, so the greater the current the closer the needle is. That lets you build up a topographical picture of the surface. If its accurate enough then you can pick up individual atoms.

 

[Scarecrow]

So some sort of mass dampening system then? Sounds like a sensible idea, I'll look into it.

And its not an electron microscope in the usual sense of the word. Those are generally where a beam of electrons passes through or bounces off a sample, and are ridiculously complex. Here, electrons tunnel across the gap as the needle gets closer, so the greater the current the closer the needle is. That lets you build up a topographical picture of the surface. If its accurate enough then you can pick up individual atoms.

Wow! I never knew how an STEM worked before this! Thanks for explaining that, I was wondering about the details.

 

[TheMinister]

Aha, we have our first problem.

We've got a good method of producing the nanometre tips needed for this microscope, http://sxm4.uni-muenster.de/stm-en/ (click the sidebar link) but one problem- how do we measure how big our tips our once produced?
The obvious way would be to use some highly sensitive kind of STEM microscope but errrrrr..... you see the problem.

Anybody got helpful ideas?

 

[Scarecrow]

Aha, we have our first problem.

We've got a good method of producing the nanometre tips needed for this microscope, http://sxm4.uni-muenster.de/stm-en/ (click the sidebar link) but one problem- how do we measure how big our tips our once produced?
The obvious way would be to use some highly sensitive kind of STEM microscope but errrrrr..... you see the problem.

Anybody got helpful ideas?

Perhaps you could shine a laser pointer at the tip, and then use a lens to spread the beam, and project it on a wall. It might be below visible intensity, but then you could take a tripod stabilized picture of the projection with a really slow shutter speed, and then measure the shadow.

 

[TheMinister]

Ooh thats nice. Although doesn't light have problems interacting with things getting towards 10 nanometers?

 

[cjp]

Ooh thats nice. Although doesn't light have problems interacting with things getting towards 10 nanometers?

Yes it does, that's why you can never see individual atoms with light.

I asked the people here about the same problem (checking whether your tip is sharp enough), and the answer was that you should simply try it in your microscope and see whether you get sharp images.

 

[ijuin]

You can always use the iterative method. First you make a tip that is sharper than a light microscope can resolve, and then try that out--it will give you a sharper image than the light microscope, but not necessarily atomic-scale sharp. Then make a tip that looks perfectly sharp in THAT image, and use the new tip, etc.

 

[Lunar Pilot]

Question

This is cool and everything, but what is the point of being able to see individual atoms?

 

[ijuin]

It's not seeing the atoms themselves that's important--it's seeing how they are linked to each other. You can actually see the links within a crystal or a molecule, which allows you to prove the shape (and hence yet-untested chemical and physical properties) much more directly than techniques like X-ray crystallography.

 

[Lunar Pilot]

Relief

That makes things a lot clearer. But the question is, what exactly would we see if we could actually ''see'' an atom?

 

[Quinn]

Not only does it allow us to see the links between atom within molecules, it has also been used in experiments dealing with the manipulation of individual atoms. Contructing molecule one atom at a time allows for a whole new level of reserch for physics and chemistry.

 

[ijuin]

Well, images of atoms that we have show them as fuzzy balls. Even if we had a device with higher resolution, the fuzziness would not go away, because the electron shell is INHERENTLY fuzzy--quantum uncertainty means that there is no sharply defined edge.