Gold. Height image. 50 mV, 1 nA, 1Hz line rate.
Gold. Height image. 50 mV, 1 nA, 1Hz line rate.
Gold. Height image. 50 mV, 1 nA, 1Hz line rate.
Sputtered platinum film. Height image. 100 mV, 1 nA, 1Hz line rate.
Sputtered platinum film. Height image. 100 mV, 1 nA, 1Hz line rate.
Sputtered platinum film. Height image. 100 mV, 1 nA, 5Hz line rate.
Sputtered platinum film. Height image. 100 mV, 1 nA, 1Hz line rate.
Sputtered platinum film. Height image. 100 mV, 1 nA, 1Hz line rate.
Gold. Height image. 1000 nm, 50 mV, 1 nA, 1Hz line rate.
Gold. Height image. 600 nm, 50 mV, 1 nA, 1Hz line rate.
Gold. Height image. 300 nm, 50 mV, 1 nA, 1Hz line rate.
Gold. Height image. 100 nm, 50 mV, 1 nA, 5Hz line rate.
Gold. Tunneling current image. 600 nm, 50 mV, 1 nA, 1Hz line rate.
Graphite atoms. Constant current. 50 mV, 1 nA.
Graphite atoms. Constant height. 50 mV.
Graphite atoms. Constant current. 50 mV, 1 nA.
Graphite atoms. Constant height. 50 mV.
A messy, uncleaved HOPG surface. Cleaving the top few layers with scotch tape produces a much cleaner, flatter surface than this. Constant height, 50 mV, 1 nA
Uncleaved HOPG surface. Tunneling current image acquired in constant current mode. 50 mV, 1 nA.
HOPG surface after cleaving, showing what appears to be a moiré pattern at the top right. Those small bumps are too large to be atoms. Tunneling current image acquired in constant current mode.
~1 um scan of a gold surface. If you look closely you can see the atomic layers. Constant current, 50 mV, 1 nA.
Hexagonal lattice structure of HOPG. Tunneling current image, 0.3 nA setpoint, 50 mV bias.
Hexagonal lattice structure of HOPG. Tunneling current image, 0.3 nA setpoint, 50 mV bias.
Hexagonal lattice structure of HOPG. Tunneling current image, 0.3 nA setpoint, 50 mV bias.
HOPG. Constant-current mode, 1 nA, 50 mV bias.
Atomic planes in gold. Constant-current mode, 1 nA, 50 mV bias.
Atomic planes in gold. Constant-current mode, 1 nA, 50 mV bias.
Atomic planes in gold. Constant-current mode, 1 nA, 50 mV bias.
Crystal grains in gold. Constant-current mode, 1 nA, 50 mV bias.
Crystal grains in gold. Tunneling current image, 1 nA setpoint, 50 mV bias.
Atomic planes in gold. Tunneling current image, 1 nA setpoint, 50 mV bias.
Atomic planes in gold. Tunneling current image, 1 nA setpoint, 50 mV bias.
Dan Berard 
This is great work, can there be some sort of scaled parallax effect in two scans to create the translation into a 3D Stereo image pair , so the complete structure can be manipulated and viewed with an Oculus Quest untethered Virtual Reality headset?
The images could be applied as a normal map texture onto a 3D geometric object for zooming and manipulation as a post processing imaging operation, also the true scaled up geometric representation of the surface could be created as well.
2 left and right eye views would be enough to get started, the main optical issue would be the pseudo parallax need to create the stereoscopic view, it would have to be normalized like a version of stereoscopic macro insect phot0graphy but on steroids.
Once a large enough of materials were studied, I would think a 3D virtual reality nanoCAD design package could be created with dimensional at that level.
LikeLike
why not just flood a chamber with nitrogen gas instead UVH ?
LikeLike
Good question! The problem is that the oxygen concentration in the chamber really does need to be extremely small. The nitrogen would have to be so pure that the density of oxygen contaminant is less than the oxygen density in air pumped down to UHV. At atmospheric pressure, there are about 10^25 molecules/m^3. At UHV, this drops to about 10^13 molecules/m^3. This means you’d need nitrogen that’s around 99.9999999999% pure! I’d imagine that trying to flush out/purge all the air from the chamber would also be extremely difficult.
-Dan
LikeLike
Pingback: A home-built scanning tunneling microscope with atomic resolution | Dan Berard
Don’t accidentally split one of those
LikeLike
You are my new Hero man!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
INCREDIBLE GOOD JOOB!!!!!
LikeLike
Would silicon be a suitable material to scan? Computer chips probably need some ethcing away of the passivation layer though.
LikeLike
Silicon oxidizes very rapidly in air, so it can only be STM’ed in UHV.
LikeLike
Dan, First I’d like to thank you for posting such comprehensive work that easily translates to the home/hobby shop tech. My Question is did you also make a High Vacuum Chamber to inspect samples that readily oxidize?
if not, what are some off the top of your head ideas one should look at before beginning on a design concept. can you readily identify any of your components that would fail under vacuum? I see one area of caution in; The adhesives being used cannot contain ensonified gasses. (solution: perhaps curing under vacuum?)
LikeLike
The trouble is, high vacuum isn’t good enough, it really needs to be ultra-high vacuum. There’s an excellent explanation of why that is here: http://faculty.virginia.edu/harrison/STM/tutorials/UHV.html
Higher vacuum levels reduce the time it takes for a surface to oxidize by reducing the number of gas molecules colliding with the surface. In high vacuum, a surface like silicon will still oxidize in a matter of seconds. In UHV, it will last much longer.
UHV typically requires at least 3 pumps: a roughing pump, high vacuum pump (like a turbopump or diffusion pump), and a UHV pump (like an ion pump). Pumping to UHV can be very time consuming, which makes tip and sample changes much more complicated.
As for adhesives, the ones I used (e.g. super glue) aren’t UHV compatible, but there are epoxies available that are. I’ve never worked with UHV before so I’m no expert on this stuff.
LikeLike
On the other hand, if you just wanted to obtain low temperatures (e.g. LN2 cooling) and didn’t care about oxidation, then UHV probably isn’t needed. In that case, you’d just need some level of vacuum for thermal insulation and to prevent condensation/frost forming on the sample.
LikeLike
The most impressive STM data I’ve seen in last 10 years!!
LikeLike
So awesome!
Will you be able to flip out atoms from the lattice, aplying a higher voltage spike? I bet that if you can and draw HaD logo…you will be an epic win of the decade. XD
LikeLike
Good idea! I’ll definitely try to figure out a way to draw patterns on the surface! Might not work with graphite though, since all the atoms on the top layer are covalently bonded to each other, but it’s worth a try.
LikeLike
Here from HAD – one suggestion to damp out high frequency vibration is simply … an inner tube. We’ve done some holography work (very vibration sensitive) using high mass steel plates and placing them on small inner tubes. Gets rid of all but the very low frequency stuff.
LikeLike
Thanks for the suggestion, I’ve heard of that approach being used for STMs before. If I can find a big enough steel plate I’ll give that a try!
LikeLike
A good suggestion. As an undergrad, friends and I built a high quality holographic table on a stone slab atop a truck inner tube with good results.
LikeLike
Dude, that project is awesome, will/do you provide a pcb layout? 🙂
LikeLike
Working on it! I’m planning for it to have digitally controlled gain and automatic coarse approach with a stepper motor.
LikeLike