Superconductivity is one of those places where chemistry and physics cross paths. That’s especially true as people search for higher-temperature materials, because that seems to involve more and more complex synthesis and characterization of the results. Very tiny changes in conditions or starting materials can make for huge differences in the behavior of the final products, and figuring out the atomic-level structures (not just unit cells, but defects, grain/domain sizes, interface effects, and so on) is a major analytical challenge.
A few weeks ago, a preprint showed up claiming room-temperature superconductivity in a mixture of gold and silver nanoparticles. That’s one out of deep left field, for sure, but so were the (by now well established) copper-oxide materials. This latest result really doesn’t seem likely at all, but you can’t quite rule anything out in this area. And the potential impacts of a room-temp superconductor are so gigantic (scientifically, economically, etc.) that even low-probability claims deserve a hearing.
That hearing they have gotten, and so far it doesn’t look good. Here’s Brian Skinner, a post-doc at MIT (and author of this physics blog) with a look at the original data. What he finds is disturbing: in a plot of magnetic susceptibility versus temperature, two completely separate runs of data (taken at different magnetic field strengths) appear to have exactly the same noise pattern. That is, whenever one of them zigs a bit up or down, so does the other. Every time. Perfectly in synch. It looks, in fact, as if someone copy-pasted one of the lines, changed the color of the points, and offset the new line a bit. Actually, it’s even worse than that: it looks like someone did that for just part of the figure – at the higher temperatures, the noise in the two runs are different.
This isn’t possible. Noise is noise, and it’s different every time you take a series of measurements. As Skinner delicately puts it, this behavior “has no obvious theoretical explanation”, but by gosh it has a very obvious practical one that is immediately obvious to even a casual observer. That is, someone copy-pasted one of the lines, changed the color of the points, and offset the new line a bit. If there were doubts about the validity of this report before this, they shrink into nothing compared to the doubts that people have now. Think about it: if you were about to report a world-changing result like a room-temperature superconductor, wouldn’t you want to make sure that everything about the paper was solid? Go over it a time or two? Make sure that a key figure didn’t include an obvious copy/paste that would call into question the veracity of the whole damn thing? The name “Jan Hendrik Schön” comes to mind, and that’s not something anyone wants to hear.
Here’s a Twitter thread from Skinner with some updates since his own note came out. The authors, he says, are sticking by their results, so it’s going to be something to watch. For my part, I certainly hope that an RT superconductor gets discovered. But this isn’t looking good.
Update: OK, this story is getting weirder, and a lot further from what should be happening if this were a real discovery. . .