DIY Scanning Electron Microscope

[Ben Krasnow]  has recently completed a home-built scanning electron microscope and has posted a video of it in action on his blog.

The build itself was done quite creatively using many off-the shelf components. We particularly like how long threaded brass rods were used not only for the supports, but also to maintain column alignment and fine-tune the spacing between the various beam focusing components.  A large glass “bell jar” covers the entire apparatus and is sealed to the bottom plate when the air is removed from within by a mechanical vacuum pump.

In order to produce an image, an electron gun similar to one found in a conventional CRT television tube accelerates the electrons with a 5kV potential from the top of the microscope downwards through a long copper column. Along the way the beam is focused and manipulated by electronic lenses in much the same way that light would be handled by conventional optical lenses. Near the base of the main column there are electrostatic deflection plates placed orthogonally in the X and Y directions that allow for precise scanning of the beam across the sample’s surface. When this high-energy electron beam is scanned across the sample, scattering surface electrons are then picked up by a nearby detector consisting of a phosphor screen and photomultiplier – a system that supposedly allows for higher sensitivity than trying to measure the small numbers of electrons directly.

Although the resolution of the first few scans is only around 50uM, this early success clearly shows that the device functions as intended and will provide a great starting point for future refinement with the final goal being resolutions down to the 1uM range.

Despite Ben’s reassurance that the x-rays produced at this energy level  won’t even penetrate the glass chamber, you can be sure that if we ever visit his garage we will definitely be donning some tin foil protection like these guys.

[Thanks kyle]


22 thoughts on “DIY Scanning Electron Microscope

    1. I’ve been following his videos for a while now. He is pretty smart and has done some other equally-as-impressive things. I believe the spark plugs are so that he can introduce an airtight electrical connection into the vacuum chamber. Spark plugs are designed to seal gas-tight, conduct electricity and electrically insulate said conductor from the rest of the world.

  1. It’s definitely good to be a skeptic during the next week or so leading up to the first of April! This guy really knows his stuff though and it isn’t the first time he’s posted some really impressive hacks online.

    The microscope really does look like a prop out of some movie though – not as I would have expected it, but nice!

  2. spark plugs are a perfect insulated and pressure sealed voltage conduit. This is real people and this really is how electron scanning microscopes work. The other crap is obvious fake since any logical person who understands the components know weather or not that the implied use is plausible.

  3. What is the vacuum level in the glass jar during operation? Brass has zinc in it and will evaporate out at low vacuum levels and may provide a bit of extra “stuff” in there. Most systems have stainless components or aluminum components to avoid sublimation, but given the “available” part description, that may be what was available.


    electronic lenses are electric coils or electromagnets that produce a field that will “focus” the beam of electrons that pass from the cathode ray tube to the sample. The field exerts force on the negatively charged electrons passing through it so they deflect in some way. They can converge, diverge or simply be steered to give the raster pattern for the picture.

  4. Nice work! Outgasing will be an issue I hope he does not have trouble with that. It can cause pulling a vacuum to take a very long time, but he won’t need as good a vacuum as a nm resolution SEM would.

    Also, I think those threaded rods would be a pain in the but to adjust. You don’t just change one, but all of them to move a lens along a tiny fraction.

    You could also use conductivity or capacitance to image the sample in an SEM by attaching a lead to the sample and using a very fine A/D. This can be easier than the phosphorus/photomultiplier method, but not as high resolution.

  5. Outgassing is not really a problem with these. You just leave the pump running while you do the scan. You might want to be looking at something which gasses anyway like a biological specimen.

    I think the problems are tightly focussing the electron beam with stray atoms floating about in there and detecting the scattered electrons. I heard it is possible to use secondary emission to amplify the signal by placing a ring of moderately positively charged conductive stuff like graphite around the target and then counting the electrons getting knocked out of that by measuring the current in another anode.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.