Great stuff. We need more of these types of science tidbits on HN. And good links by the commenters too. I love Feynman's classic style of "explaining to a 5 year old" on Cold Welding. The reason for this unexpected behavior is that when the atoms in contact are all of the same kind, there is no way for the atoms to “know” that they are in different pieces of copper. When there are other atoms, in the oxides and grea…
Note that this is a big oversimplification- it doesn't apply to gage blocks. Cold welding and similar phenomenon can only truly happen with metallic bonds and very large grain sizes. Non-metallic bonds are structured and require precise orientations and conditions to reform. Metals are still structured -metals have crystal grains- and the microscale structure needs to line up in order to reform. In cold welding, you…
Around a decade ago, I worked in a machine shop. Mill table surfaces are ground flat. Not precision surface flat, but good enough to be a reference for cutting tools. Mill vise clamping surfaces are ground flat. It's normal for an object that's at least cut (not even ground) flat on one side to get hydraulically suctioned via cutting oil to either of these. It's usually easy to slide the part off to the edge of the surface.
Very flat surfaces do this as well, but with a little more encouragement to work out oil from between parts, they'll start to stick in place. The more finely ground and flat each side is, the more pronounced the sticking moment will be. Extremely finely surfaced blocks being wrung together will frustrate you by sticking before you have them lined up the way you want them.
This made me suspicious of the role that surface tension and vacuum played. Hydraulically stuck things are easily separated by a quick blast of compressed air. Wrung blocks aren't as easily separated by a blast of air. I tried to clean blocks as devoid of liquid as possible and wrung them together. They still stick, but it's not as secure, and they come straight apart the moment they're twisted apart.
After this, it seemed that wringing blocks together worked best with a trace of oil. Brief cleaning with a dry rag does leave a trace, and if the surfaces are flat enough, perhaps that trace is enough to fill some microscopic voids between flat-ground surfaces. Twisting blocks together encourages entrapped air to escape, and can shuffle trace oil into voids. The solution I came up with is that the metal of the two blocks does stick together somehow once it's in contact, and the surface tension provided by a trace of oil contributes additional sticking force where the surfaces don't meet.
We didn't have ceramic blocks, though. It might be an interesting additional experiment to try this with mixed materials. Does wringing together ceramic and steel precision surfaces work the same way? Should it? What would that mean?