- cross-posted to:
- technology@beehaw.org
- science@mander.xyz
- cross-posted to:
- technology@beehaw.org
- science@mander.xyz
And some commentary on the paper:
https://www.science.org/content/blog-post/breaking-superconductor-news
You must log in or register to comment.
Big doubt. People have self-published shit like this a lot over the past decade or two. Usually it turns out to be a measurement error if it goes anywhere at all. Post again when this is peer-reviewed.
Edit: A month on, here’s Nature’s autopsy. The big question for me is why this one went viral when most USOs just fade into obscurity.
If would be one hell of an error, if so, to get measurements across several different properties and temperature ranges that were all consistent with superconductivity.
But even if true, we can make lots of things in very small quantities in laboratories that are far too expensive for practical manufacture. That’s where engineering will come in.
If would be one hell of an error, if so, to get measurements across several different properties and temperature ranges that were all consistent with superconductivity.
You’re right, maybe it’s fraud, at least partly.
I understand it’s very hard to measure the conductivity of a microscopic crystal attached to other different crystals, which is why a lot of less-than-solid claims about high-temperature superconductors get made.
Here’s a video of a macroscopic piece of the material magnetically levitating.
Sure, it could be faked, but that would be a bit more than the usual “massaged the numbers a bit to discover a really small effect” stuff.
At first sight, it looks fake:
- A levitating superconductor locks into place onto the magnetic lines, it shouldn’t be bouncing up and down like that while touching the magnet with one side.
- Pyrolytic carbon can levitate on a magnetic pole grid at ambient temperature and pressure, while magnetic poles can be “printed” onto a single magnet slab, giving the illusion of superconductive levitation where there is none.
- At one point in the video the piece of material falls down completely to the magnet, then it bounces back up when pushed to another point, which looks less consistent with superconductive levitation, and more with a prepared trick magnet.
As I said, it could be faked. But that fakery would involve a very deliberate premeditated fraud and when it is found out (as it will be in short order with something this extreme) the three researchers involved in this would be completely ruined. What are they gaining out of this that would be worth it?
Also, if they’re rigging up a complete fake, why would they half-ass it? If your first take has the fake superconductor fall down why not delete that one and do another take where it doesn’t?
I’m not saying this video is proof, I’m saying that this is not some Em-drive situation where the results are so fine and fiddly that it’s only barely peeking through the data and it could be a result of tiny errors and wishing really hard. The comment I was responding to was talking about how difficult it is to measure the properties of microscopic crystals and I was pointing out that this is a big ol’ chunk of stuff being poked around with the tip of a finger and hovering millimeters over a magnet. This is going to be straightforward to verify.
Edit: Found the companion article that’s specifically about the levitating sample.
I see.
Reading the article, it seems to be on one hand promising, but on the other the characteristics of the material seem to be somewhat flimsy, and they seem to have made a few different samples then modified them between tests.
The final result of becoming an ohmic metal at 127C, with at least a couple orders or magnitude jump, seems to be consistent, so that’s something. The behavior below 127C though, looks kind of iffy. Maybe because they tried different manufacturing and processing methods, maybe because of the different breakdowns they describe in the article… which they don’t fully describe reversing, so it gives the impression of being a borderline one-way only superconductor that starts conducting at about 25C, and in a real world application could lead to a cascade effect from there. There seem to be no recovery behavior tests either, which could be understood for an initial paper, but is a pity not to have them.
The measurements they show on the graphs, are for very low voltages and intensities, so that maybe could explain why the piece on video fell down (lost superconductivity due to overheating), then sprung back up (when it cooled down). Or maybe they just blew on it to cool it down enough (which would be interesting on itself). The material structure transitions are somewhat complex, and happen basically all over the range from -75C to about 50C, changing its characteristics.
You’re right, this seems much more solid than the Em-drive case. It needs better reproducibility and better characterisation, but otherwise looks promising for at least some applications.
Welp, here’s Nature delivering the verdict. Pretty much on the schedule I expected, and here’s the update I’d promised myself I’d do.
LK-99’s amazing properties are Cu2S’s amazing properties. It’s a chunk of stuff with embedded copper sulphide. Like I said, it’s hard to verify the properties of one material mixed in with other materials, and in this case they didn’t even try very hard.
Here the microscopic crystals were the impurities not the “superconductor”, but it’s the same issue.
Sadly, yeah, this appears to be a solid case against LK-99.
But this is all I ever wanted. This is not validation for all the negative-nellies who were instantly dumping all over the possibility that LK-99 could have been superconductive before these further tests were done. When these sorts of apparent breakthroughs are made the proper response is to do as these people publishing in Nature did - take it seriously enough to actually check it out. That’s the only way to avoid missing out on the actual breakthroughs.
That behaviour looks pretty normal for a Type-I superconductor, the “locking on” you are referencing is a property of Type-2 Superconductors. For more information search Meißner-Ochsenfeld Effect (ideal diamagnetism, type 1 SC) and flux pinning (type 2)
I have a fair bit of experience with superconductivity and the submitted manuscript on arXiv looks solid, I didnt notice anything suspicious
Edit: the falling down may be because the material is not superconductive throughout, it looks like it falls down, rotates because of the magnetic field (maybe the non superconductive phase is magnetic?) then pops back up because the magnetic field of the magnet is once again strong enough to lift the sample (the Meißner effect dispels the field inside the Superconducter by generating shielding currents just below the surface, thus “mirroring” the field of the permanent magnet)
I reserve my skepticism. Seriously, I’ve personally watched these papers roll in and then disappear for over a decade.
It’s also not promising it’s out of East Asia, judging by the Korean names, which has a bit of a problem with academic dishonesty right now.
What about the video?
I haven’t watched it, but I’m not sure how you’d visually prove it’s actually the material in question, so it could still be a hoax like all the others.
Replication is literally the only evidence I haven’t hardened to after all the error and fraud that’s come through.
I doubt this so much. Especially since in the field of superconductivity, there have been remarkably many academic fakes.
Also I don’t trust papers which were written in MS Word.
I wrote a whole thing and it sounds super condescending. I’ll leave it here but I’ll let you know I only wanted to tell people about Pre-prints and ArXiv as a whole, it was never my intention to disrespect you or any others! I even had to add this first paragraph as I felt bad about it… But here goes:
Most papers I’ve seen are written in MS Word.
In Physics, which I admit is what these people work in, papers written in LaTeX are more common. But still, not most of them are. No clue about Computer Science and stuff, I mostly work in Nanotechnology, Biotechnology and stuff, so mostly physics and bio stuff (…can you tell which of these fields is mine and which I’m only tangentially related to? Hahaha)
After they are written in MS Word or paint, they are submitted for review. If they are approved (probably after a few rounds of revisions) they are submitted to the editors, which turn the paper into something that does not look written in MS Word.
But this is arXiv. It’s a pre-print server. People submit their papers before they go through the whole peer review process. Which means that these papers can have a few very significant mistakes, or even be fraudulent or wrong. That would be my main concern.
Of course, most of these pre-prints are not the final version of any paper - typically people submit them to pre-prints for a few reasons, while the paper is in peer review. Or often the paper has even already been accepted for publication, but they submit the version without any sort of peer review to the pre print server (I’m actually very early in my scientific career and only have one paper as first author, so this is the part I don’t remember as well. I think we submitted it to ArXiv after it was accepted for publication, but before it was published, and we sent the earliest draft we had submitted for review). These reasons are, off the top of my head:
- The paper gets out faster, so people can see their amazing results earlier. Especially if they are worried about being the first to publish, as it’s very common for a reviewer with vested interests to block a paper from being published while they work on a very similar project. It’s not plagiarism per se, as these are loooong projects and typically the reviewer would already have a very mature project that they heavily invested in.
- Most if not all journals are OK with the pre-print being freely available even after the article is published, even if the journal itself has a hefty fee for accessing the paper (or even for allowing your paper to be Open Access, which is typically very expensive for authors…)
- Finally, a smaller reason: this is the paper as the authors intended, before those pesky reviewers got their hands on it. For you and I this is a negative, for authors this can be a positive. Often, reviewers will have their own interests and ask you to change your paper accordingly, most often by citing one of their papers… The peer review process is anonymous, but somehow you can always tell when this happens!
So: don’t worry about the formatting, worry about the content! Let’s wait until this passes - or fails - peer review before accepting it or discarding it. It could be super exciting results! Or a big pile of nothing.
The paper gets out faster, so people can see their amazing results earlier.
That’s precisely the problem of the entire superconductivity-at-room-temperature field and what I’m implying with the “MS Word”-remark. No other field of physics suffers so many retractions and outright data falsifications than superconductivity.
Supcerconductivity at room teperature is a holy grail that was claimed to have been achieved for the last 25 years, and every time it was either unreplicatable or data falsiciation. Extraordinary claims require extraordinary evidence - and this is what every other field seems to understand. The first direct observation of gravitational waves happened in September 2015, but the observing teams took 6 months to verify their findings before going public. The first direct observation of gravitational waves from a neutron star merger was discovered in August 2017, but only made public one month later - even though the event was accompanied by electromagnetic counterparts, and everybody in the field already knew what had happened, because it was impossible to keep secret. Still, their took their time to properly check and present their results.
Now I’m seeing a MS Word paper with a few plots in it, claiming not only the holy grail of superconductivity at room temperature but also claiming to achieve it at ambient (i.e. “normal”) pressures. A double holy grail. That’d be like the holy grail just fucking right off and god themselves outright appearing as an author on the paper.
No sorry, this I just can not believe in the slightest.
You seem to be very knowledgeable about this field so I won’t try to change your mind, as I can tell you’re more familiar than I am. In fact, I also have my own reservations from reading the actual paper. The same authors actually have another paper in the same topic with the same material with a preprint out after this one, and I doubt someone would salami publish results as marvelous as these ones instead of just going for a single massive impact factor journal.
yeah my concern is definitely moreso with the lack of peer review as opposed to using MS Word. LaTeX is definitely common but I’ve seen a fair few written in word. A lot of times they’ll also get converted to whatever format the journal is using, if said journal is “fancy” enough.
So… Nature quite literally states they prefer submitted manuscripts to be in Word… go look up the guide to authors.
Some commentary on the paper:
https://www.science.org/content/blog-post/breaking-superconductor-news
It’s interesting that so many are in a hurry to doubt. Give it a couple of weeks for tries at replication and you’ll know for sure. I suppose the first to call BS wins the “I told you so” award. Whereas, if true, it’s a Nobel.
