50,000,000x Magnification

Today's video is about my favorite microscope ever. I did a lot of work in gradschool on this STEM, or Scanning Transmission Electron Microscope, and today I get to share how it works! Extra thanks to the Materials Department at UCSB for letting me film in the lab!

Admittedly it's old footage, but since it's REAL research, I was waiting for the paper to get published. If anybody's super-curious, here's the doi:
(unfortunately it's not indexed on SciHub yet - I'm trying to figure out how to get the manuscript posted elsewhere for you guys - check back in a couple days if you're curious!)

Music and graphics:

Map of Kikuchi line pairs down to 1/1Å for 300 keV electrons in hexagonal sapphire (Al2O3), with some intersections labeled
P. Fraundorf

The Quantum Realm by The Whole Other
YouTube Audio Library License

Memes:
Shrek (Dreamworks)
Pokemon (Nintendo)

[AlphaPhoenixChannel]

Hey everybody! comments, corrections, and FAQs here:
1) Go check out reddit.com/r/trytryagain, there are a bunch of posts now and it's great!

2) Thanks to Kunal and Aidan for previewing this video and making sure I wasn't too egregiously wrong in my explanations! Kunal wanted to add that for TEM and STEM imaging, the vast majority of the time the sample isn't pointed "at the beam" - normally it's at a wonky angle to produce very specific diffraction effects. This lets us do some cool things, like look at strain fields from dislocations, but I didn't do that in this video. Also because of diffraction effects, by the time you've zoomed all the way into atoms, the "shadow" analogy is actually pretty terrible, and image formation is complicated (hence switching to STEM). Aidan had some comments on STEM-vs-TEM, notably that despite the beam paths being completely different, they actually DO interact with the sample in largely the same way and produce similar contrast when you aren't at atomic resolution. When I said they were mechanistically completely different, that was accurate from the perspective of the mechanics of the microscope being different, but the beam-sample interaction physics is EXTREMELY similar.

2 (TLDR) I swept a lot under the rug in order to focus exclusively on atomic-resolution microscopy - Diffraction and strain contrast microscopy is arguably a lot more important, certainly more common, but also a LOT more complicated.



5) I’ve had a lot of people ask about the drift - yes the sample is rigidly mounted, and yes the beam keeps pointing in the same direction, but when was the last time you tried to hold something so still that it didn’t move by an atom? Thermal expansion and contraction is actually a huge problem - if the sample stage is warmer than the microscope, when you load it in, the arm that holds the sample is going to cool down and contract - super slowly - for a couple hours minimum, and that’s enough to ruin long exposure pictures with crazy rolling shutter artifacts.

6) a few people have mentioned the mask and while I think most of these commenters are just being irritants because they can, it actually brings up a very interesting point about these scopes. To reduce noise, vibration, and changes in temperature, the room that houses this microscope is designed to have almost zero air changeover. This is fantastic for consistent microscope operation, but real bad to leave an airborne virus in a room and have somebody else walk into the same air a few minutes later.

7) I did some math and changed the title from magnification to magnification (eh - same order of magnitude). magnification is very poorly-defined because it technically depends on the viewing screen. There IS a "standard" display size from when all TEM images were published on paper in the same journal format, but I don't remember the size, so I took one of the shots where I was focusing with a field of view of a little over 4 nanometers and assumed it was 8-10 inches tall on the monitor, then rounded nicely. according to the manufacturer's site, the magnification goes up to 230M - again, not sure what the intended display size is. I wonder if they ship computer monitors with the microscope???

8) I'm sure there are more things wrong or fuzzy so complain away and I'll add notes here!

[danieloconnor]

It’s insane that this kind of technology can be used and showcased to potentially millions of people all by one guy.

[thethoughtemporium]

That is an absolutely stunning machine, and an equally stunning capture! Well done!

[keithrog]

I used to think I was pretty smart. Then I realized that someone had to create this machine and that knocked me down a few pegs. This guy did a great job dumbing it down for me and is a natural educator.

[ericlane]

I remember growing up not too long ago where in science classes we were told that an atom has never been photographed and now it’s so exciting that we finally have

[cylosgarage]

What a spectacular video. I recall Ben Krasnow once commenting about how capturing the initial joy and wonder of scientific discovery in a YouTube video is extremely difficult, but you’ve done a wonderful job of that here, as well as giving excellent explanations as to how it all works. Well done

[unusedTV]

Extremely cool to see the entire process. I've seen the output images in papers and science reporting, but it's refreshing to see the entire (cumbersome) process leading up to such results.

[seregaforeverserega]

please never lose your curiosity - such people like you move humanity forward and give us hope in the future 👍

[henry]

Wow... at 70 years old, I'm absolutely and totally off the scale awestruck. Thank you so much for all of the time and effort you did in producing this video and the excellent explanations.

[Zappygunshot]

"I'm going to be looking for a specific thing, and I'm not sure I'll find it, but I'll still learn something" is the most sciencest science thing any scientist has ever scienced.

[atronajs]

Something that always blows my mind is the engineering that goes in to different scientific instrumentation. Those instruments are the culmination of thousands upon thousands of hours of work from many different people. Years of work before you can even start to do your work.

[jeremyday]

I worked for the DOE for a few years and ran an EB (electron beam) welder. Very different but but this video brought back some memories. Also my uncle ran an Electron Microscope and analyzed samples from Ground Zero for asbestos. Unfortunately he passed a few years ago from mesothelioma but he would take us kids to his lab and show us all kinds of cool things. My fav part was playing with liquid Nitrogen lol. Anyways thanks for bringing back such great memories ❤

[theslowmoguys]

This is amazing. Is there not an ability to save all values to “re-find” a specific spot on a specific orientation of the sample or is it way more complicated than that?

[msachin]

the way you overlap the results you see from theoretical explanation to the real world almost seamlessly really should be what every academic institution should strive to do when teaching these concepts. I've always loved being able to see an explanation like this and I strive to demonstrate it whenever I teach as well. It strikes an incredible chord between seeing tangible results to what seem like abstract concepts at first, as well as imparting a true sense of wonder for the universe. If this doesn't get you excited about science then nothing will! Stunning video as always

[furjaden]

To think that 2-3 decades ago, most of the public would never even know about machines like this, let alone see them work their magic.

[mrparakeet]

Its incredibly difficult to wrap my brain around how we can even fathom seeing something so small. Incredible

[timseguine]

I think you are one of the few people who actually shows the entire process when you do a video like this, and not just the flashy parts.

[Cketzalcoatl]

You've taken me back to when I was doing my PhD! Hours upon hours upon hours on an SEM, trying to find particular weld defects or characterising fracture surfaces (I think I got a little triggered when you mentioned "stig", lol).

You really communicated well that elation you get when, after god knows how many hours and how many samples, you find exactly the feature you were looking for, and you know you've gone from something purely theoretical to something with actual evidence behind it. I also got super pumped when that happened!

[backen]

I cant even comprehend how small that sample is, it must be so finnicky to move

[charlieevergreen]

That’s incredibly badass. I knew advanced electron microscopes were around, but showing us the magnetic lenses, the impressive sample aperture, well... Really great job explaining each step with enough detail to be appreciated! It’s incredible engineering, ambitious objectives, and a mind-blowing achievement, and you did a great job conveying a ton of info in a brief time. Thanks for the video, and congratulations on finding the phenomenon! And a paper! Nice!