Room Temperature Superconductivity?

July 26, 2023 11:48 AM

A South Korean team has published pre-print results on Arxiv. If true, this could be world-changing. Or is it just the next "cold fusion" scam?
Quantum Insider
IFL Science

Of course, not everybody defines "room-temperature" as "127°C ( 261°F)" but it seems that the key here is not requiring thousands of atmospheres of pressure.

My hope is that smarter people than me on Metafilter have seen the same news and have more to say about it.

posted by ZakDaddy (137 comments total)

Derek Lowe seems optimistic.
posted by Orange Pamplemousse at 11:57 AM on July 26, 2023 [12 favorites]

It works **at and below** 127 C.
posted by NotAYakk at 12:21 PM on July 26, 2023 [8 favorites]

I've seen a few reputable people on X xeeting that this is exciting and different from cold fusion in terms of legitimacy.

[God, that's so fuckin' stupid]
posted by fatbird at 12:28 PM on July 26, 2023 [5 favorites]

Its a preprint. In some cases thats fine and interesting but this seems like something we should wait for peer review on.
posted by pwnguin at 12:34 PM on July 26, 2023 [11 favorites]

We need some good news about how technology is going to change the world.
posted by seanmpuckett at 12:35 PM on July 26, 2023 [1 favorite]

Including IFLScience as a source is not a way to engender confidence. Traditionally they lean heavily toward clickbait over facts.
posted by mrgoldenbrown at 12:41 PM on July 26, 2023 [4 favorites]

We should know if this is legit within a week or two. The procedure to make the material is fairly straightforward and doesn't require exotic materials or equipment. I think an earnest lab could replicate it (or not!) quite quickly.
posted by riotnrrd at 12:42 PM on July 26, 2023 [19 favorites]

oh this drops the same day as the Senate UFO hearings huh

interesting
posted by thecaddy at 12:43 PM on July 26, 2023 [17 favorites]

I've lately seen one or two IFLScience articles that have put them in my good books, and some of their writers are solid.

Faint praise I admit.
posted by edd at 12:44 PM on July 26, 2023

What would you do with this material, if it turned out to be real?
posted by mittens at 12:46 PM on July 26, 2023

This is what happened with the last superconductivity hype. Hopefully this is something more interesting, although it's just one single preprint.
posted by Pyrogenesis at 12:48 PM on July 26, 2023 [3 favorites]

Yeah... I'm afraid not.

Very brief read and not my area of expertise. But...

There are a few things that would be needed to convince other scientists, and they're notably missing here. A firm, well-measured critical temperature supported by a nice graph showing the phase transition, and the Meissner effect turning on and off (mentioned but not demonstrated in any reasonable way in the paper) are a bare minimum. Given that these are lacking with a claim of this magnitude makes me think that this is three-sigma evidence of nonsense.

(And IFLS, IMO, largely consists of hyped, semi-plagiarized press releases. YMMV.)
posted by cgs06 at 12:49 PM on July 26, 2023 [13 favorites]

Elaborating a bit... the real important graph for me is (e). Low quality, no error bars, and the density of measurements near the Tc at (around!) 362ish K is low. For a claim like this, with a real history of false starts, it's really far below the level of evidence I'd expect.
posted by cgs06 at 12:55 PM on July 26, 2023 [7 favorites]

[hugging MeFi server] you don't listen to anything they say about ColdFusion, you're valid and I love you
posted by cortex at 12:57 PM on July 26, 2023 [111 favorites]

oh this drops the same day as the Senate UFO hearings huh

Superconducting healthcare pls
posted by credulous at 1:05 PM on July 26, 2023 [5 favorites]

In a 1989 sci-fi book called "The Long Run" (link to full text), the author Daniel Keys Moran has an angry businessman describe a hacker like this:

“Frank Calley,” said Jerry Jackson with a convincing enough display of anger, “is a thief. He lifted fifteen terabytes of hot RAM, a hundred and five thousand Credit Units worth of room- temperature superconductor memory, from mah warehouse in Georgia.” Under the stress of anger, the accent became more audible.

HOT RAM, BABY. ROOM-TEMPERATURE SUPERCONDUCTOR.

The phrase "room-temperature superconductor" from this book has long stuck in my mind as one of the few things in cyberpunk sci-fi that hasn't come true yet.
posted by wenestvedt at 1:08 PM on July 26, 2023 [8 favorites]

Time to rewatch The Saint.
posted by Fizz at 1:11 PM on July 26, 2023 [3 favorites]

In a 1989 sci-fi book called "The Long Run"

Just wanted to say this is exactly where my mind went when I saw "room-temperature semiconductor" :-D
posted by zbaco at 1:16 PM on July 26, 2023 [1 favorite]

[hugging MeFi server] you don't listen to anything they say about ColdFusion, you're valid and I love you

Wait, WHAT
posted by grumpybear69 at 1:19 PM on July 26, 2023 [6 favorites]

This seems really easy to make. Is this typical for superconducting materials?
posted by mr_roboto at 1:19 PM on July 26, 2023

You just need sufficient apatite for conduction
posted by Flashman at 1:22 PM on July 26, 2023 [19 favorites]

I don't know if it's "typical", but YBCO, a relatively high temperature superconductor, is simple enough to make that it was a lab exercise in my college chemistry class.
posted by phooky at 1:25 PM on July 26, 2023 [1 favorite]

Time to rewatch The Saint.

There used to be a website back in the day (and I'm sure someone will name it immediately) that was about imaginary matchups between celebrities. It wasn't Celebrity Deathmatch because it wasn't usually about personal combat. It presented celebrities with particular challenges, and the two guys who did the site would then argue their respective cases as to why one celebrity would beat the other one.

My suggestion, sadly never acted upon, was that they should give Elizabeth Shue and Keanu Reeves identical high tech labs with all the fancy gadgets they could ask for, and see which one comes up with cold fusion first.

(Based on The Saint, and Chain Reaction. Personally, I'm going to have to go with Keanu because his uses a cool sampling keyboard. Which he has in his lab for some reason.)
posted by Naberius at 1:29 PM on July 26, 2023 [2 favorites]

Interesting, but far enough away from my knowledge that I have no thoughts one way or the other on its reproducibility. But health is something I do know about, and the fact that lead is a major component of these compounds is not good. And if these compounds really do perform as advertised, the thought of major industries using huge amounts of lead; probably in underdeveloped countries and Texas, where environmental regulations are somewhere between weak and nonexistent, is concerning.
posted by TedW at 1:30 PM on July 26, 2023 [6 favorites]

probably in underdeveloped countries and Texas

Shots fired! Touche, sir.
posted by Abehammerb Lincoln at 1:33 PM on July 26, 2023 [10 favorites]

the thought of major industries using huge amounts of lead; probably in underdeveloped countries and Texas, where environmental regulations are somewhere between weak and nonexistent, is concerning.

Flint, Michigan: future home of semi-conductor manufacturing?
posted by The Pluto Gangsta at 1:34 PM on July 26, 2023 [1 favorite]

Flint, Michigan: future home of semi-conductor manufacturing?

Booooooooooooooo.
posted by grumpybear69 at 1:38 PM on July 26, 2023 [2 favorites]

[hugging MeFi server] you don't listen to anything they say about ColdFusion, you're valid and I love you

My story about ColdFusion is that I worked with it for 5 years as a developer for a startup in the early 2000s. A year after I left that job, another ColdFusion position came up. But after the first round of interviews, I realised that I couldn't remember anything whatsoever about the ColdFusion language - the syntax, how to do basic tasks like step through a recordset, anything. I found myself poring over programming guides, and it was like I'd never seen the language before. I've never experienced that sense of mind-wipe with any other language.
posted by pipeski at 1:44 PM on July 26, 2023 [24 favorites]

If this works, I'll eagerly await NileRed figuring out how to make it out of latex gloves and Canadian dimes.
posted by clawsoon at 2:00 PM on July 26, 2023 [13 favorites]

If you're going to commit fraud, you do it with something that takes 2 months, exotic materials, and a fiddly procedure to make. According to the paper, this can be made in 48 hours with normal lab equipment and common ingredients.

We'll find out quickly if this is real, and I can't wait!
posted by Urtylug at 2:14 PM on July 26, 2023 [12 favorites]

Are there any contexts other than superconductivity in which "high-temperature" is colder than "room-temperature"?
posted by What is E. T. short for? at 2:35 PM on July 26, 2023 [2 favorites]

A reminder of a story about how frosted glass bulbs came to exist. They were a striped-paint gotcha project, it seemed impossible to make anything more than incredibly fragile, useless frosted bulbs. Then someone accidentally re-dosed a frosted bulb with acid and it wasn't fragile anymore.

Maybe this is a nothingburger, maybe this is a fantastically lucky find by a team that knew what they were looking at and could document it. Which is to say, the usual partnering of luck and perseverance.
posted by Slackermagee at 2:36 PM on July 26, 2023 [5 favorites]

You just need sufficient apatite for conduction

"Apatite for Conduction" is my favorite Guns N Roses album, hands down.
posted by zardoz at 2:40 PM on July 26, 2023 [31 favorites]

I mean, it's not a tiny measurement discrepancy making a leap to "oh gosh it must be fusion!" or the whole "we designed a massless drive by subtracting an approximation from an exact value" nonsense.

Either they made the thing or they didn't. And hopefully it won't be too long before other people try to make it and we find out whether it works or is just a bizarre screwup with the measuring equipment that looked an awful lot like superconductivity. Or I guess it could be a very exciting way to say you're quitting your academic career? Usually hoaxes try to be more mysterious and harder to test.
posted by Zalzidrax at 2:57 PM on July 26, 2023 [4 favorites]

Are there any contexts other than superconductivity in which "high-temperature" is colder than "room-temperature"?

I've seen it used for Bose-Einstein condensates, which typically only form at temperatures very close to 0 K. Even 20 K would be a "high temperature" for a BEC.
posted by mr_roboto at 3:03 PM on July 26, 2023 [2 favorites]

There's a second preprint.
posted by heatherlogan at 3:07 PM on July 26, 2023 [2 favorites]

Naberius: There used to be a website back in the day (and I'm sure someone will name it immediately) that was about imaginary matchups between celebrities

I think that was World-Wide Grudge Match, which did "Doberman vs. Doberman's Weight In Chihuahuas" long before that was a trope.

The archives are in the Wayback Machine, I believe. Some of them were hilarious.
posted by wenestvedt at 3:14 PM on July 26, 2023 [3 favorites]

Let's say this works. What would be the first product to use it?

(MRI machine, something like that?)
posted by clawsoon at 3:22 PM on July 26, 2023 [2 favorites]

"Apatite for Conduction" is my favorite Guns N Roses album, hands down.

I'm pretty sure that's the one where Axl Rose sang for AC/DC.
posted by clawsoon at 3:24 PM on July 26, 2023 [5 favorites]

If you're going to commit fraud, you do it with something that takes 2 months, exotic materials, and a fiddly procedure to make. According to the paper, this can be made in 48 hours with normal lab equipment and common ingredients.

The line between fraud and overenthusiasm is 48 hours, normal lab equipment, and common ingredients.
posted by clawsoon at 3:25 PM on July 26, 2023 [14 favorites]

My first thought was maglevs for everything.
posted by ver at 3:26 PM on July 26, 2023 [1 favorite]

If it could be put to use quickly in transmission lines that would be a game changer. That's a ton of power loss iirc and ot would make building land hungry renewables like solar and wind more capable despite the distance to market
posted by Slackermagee at 3:29 PM on July 26, 2023 [4 favorites]

As described I think it is mostly useless? Its capacity for current is very low.

The point is that once you have something working you can locally optimize to make it less useless. And you can search "nearby" chemically as well.
posted by NotAYakk at 3:35 PM on July 26, 2023 [4 favorites]

Could be pretty groundbreaking, and popping up on pre-print it's a real lose your cool rule stun, too.
posted by snofoam at 3:38 PM on July 26, 2023

Let's say this works. What would be the first product to use it?

If I'm reading the preprints correctly, it looks like they only see superconductivity at relatively low currents, so no good for power transmission. I don't understand superconductor maglev well enough to say whether it would be useful for that, but the authors do report the magnetic susceptibility; that's the relevant figure of merit.

The material is easily superconducting at currents relevant for microelectronics. Potentially huge applications there.
posted by mr_roboto at 3:40 PM on July 26, 2023 [4 favorites]

I've never experienced that sense of mind-wipe with any other language.

This starts to explain a few things about the historical pace of MeFi feature development.
posted by atoxyl at 3:42 PM on July 26, 2023 [5 favorites]

If it could be put to use quickly in transmission lines that would be a game changer. That's a ton of power loss iirc and ot would make building land hungry renewables like solar and wind more capable despite the distance to market

Almost zero of the challenges of the renewable transition have to do with power loss in long distance conductors, and it seems to me extremely unlikely that the power losses of long distance HVDC (3% per 1000km) are big enough to make it viable to use something other than aluminium cables.
posted by ambrosen at 3:50 PM on July 26, 2023 [3 favorites]

I'm still trying to figure out what to do with this graphene I made with scotch tape and pencil lead.
posted by clawsoon at 3:54 PM on July 26, 2023 [12 favorites]

Almost zero of the challenges of the renewable transition have to do with power loss in long distance conductors

How about energy storage then?
posted by dephlogisticated at 4:02 PM on July 26, 2023 [2 favorites]

Short video of playing with a sample with a magnet
posted by heatherlogan at 4:18 PM on July 26, 2023

If you're going to commit fraud,

While there was a high profile case mentioned upthread, I think the real danger is just optimistic evaluation. You really do need an adversarial approach to suss out process errors, mathematical mistakes, etc. The ease of replication is good news but I wouldn't budge the truth meter much for that.
posted by pwnguin at 4:32 PM on July 26, 2023

Lol at those waiting for peer review. Like I need 2-4 random academics half reading a paper and firing off a cranky opinion to judge whether something is true.
posted by MisantropicPainforest at 4:42 PM on July 26, 2023 [3 favorites]

I think most of us are waiting for reproduction. Peer review isn't really relevant here since there are dozens of labs that are capable of performing these experiments within the next few weeks.
posted by mr_roboto at 4:46 PM on July 26, 2023 [6 favorites]

Lol at those waiting for peer review. Like I need 2-4 random academics half reading a paper and firing off a cranky opinion to judge whether something is true.

Glad you're an expert on "x-ray diffraction, EPR, and more) .....and the Meissner effect (expulsion of a magnetic field), sudden resistivity changes at a critical temperature......, current-voltage (I-V) plots at different temperatures and under different magnetic field strengths, etc. "

I'm a chemistry professor and I don't feel appropriately qualified to judge the data and experiments above to a sufficient level of rigor. But hey peer review, expertzzzz dumb and slow lol.
posted by lalochezia at 4:52 PM on July 26, 2023 [11 favorites]

I think the point about the irrelevance of peer review is that it doesn't matter one nanoshit if the paper was properly written, cites appropriate existing work, has correct statistical controls, etc..

Whether or not this material can be reproduced and actually has the claimed properties is immeasurably more important, and those aren't questions peer review is designed to answer.
posted by 0xFCAF at 4:58 PM on July 26, 2023 [10 favorites]

I’m not. But whether 1) this replicates and 2) the scientific community judges this to be real is way more reliable, informative, and quicker than whether it lands in any peer reviewed journal. My belief about whether this is true or not is exactly the same as whether it’s on Arxiv or in some journal.
posted by MisantropicPainforest at 5:00 PM on July 26, 2023 [2 favorites]

The paper could definitely use some cleanup but given how quickly this should be reproducible it seems easier to do that and just reject the paper if the procedure doesn't work.
posted by Tell Me No Lies at 5:11 PM on July 26, 2023

But who is the scientific community and how will I know their judgement? Peer review has many flaws but Arxiv only cures a few of them.

I guess we can check back in a week and see how the replications nobody funds are going?
posted by pwnguin at 5:11 PM on July 26, 2023 [1 favorite]

I wonder if someone could get NSF RAPID funding to do this. Turnaround time on a RAPID proposal can be as quick as a few days; review is completely internal.

It doesn't seem like particularly expensive work, though.
posted by mr_roboto at 5:18 PM on July 26, 2023 [2 favorites]

I dunno, just having a squiz at those graphs there's clearly a phase change going on, but there's nothing there to rule out a collapse from a resistance of about 1Ω to a resistance of about 2.5µΩ which is nice but also not superconduction. They'd need to get some better measurements because there's enough noise in 1b to hide a lot of things.
posted by nickzoic at 5:35 PM on July 26, 2023 [3 favorites]

I guess we can check back in a week and see how the replications nobody funds are going?

I mean, yes? One of the interesting things about this paper is that the material is cheap and easy to make. The equipment exists in many materials science and chemistry labs, and the components used to make it are common enough that they may already be in a lot of lab stocks. I’m pretty confident that there are grad students all over the place trying to reproduce this now.
posted by learning from frequent failure at 5:36 PM on July 26, 2023 [3 favorites]

What would you do with this material, if it turned out to be real?

Two hoverboards at the same time, maaaaan!
posted by loquacious at 5:58 PM on July 26, 2023 [12 favorites]

Two hoverboards at the same time, maaaaan!

So, hoverskates?
posted by Joakim Ziegler at 6:04 PM on July 26, 2023 [6 favorites]

Is tinkering with materials like this (crystalline substrate with regular repeating voids filled with one salt, replaced with a second salt) a common angle of attacking the room temperature super conductance problem?

I guess what I'm getting at is, is this the absolute best we can accomplish today or if people began to look at the structures and think about what other substitutions are possible, are we instead on the very bottom of a very sharp upward curve?
posted by Slackermagee at 7:19 PM on July 26, 2023 [1 favorite]

Interesting, but far enough away from my knowledge that I have no thoughts one way or the other on its reproducibility. But health is something I do know about, and the fact that lead is a major component of these compounds is not good. And if these compounds really do perform as advertised, the thought of major industries using huge amounts of lead; probably in underdeveloped countries and Texas, where environmental regulations are somewhere between weak and nonexistent, is concerning.

I have some bad news to give you about the use of lead in lead acid batteries, in ammunition, and lead as a byproduct of thorium reactors.

If anything, I'd be really glad to see a new demand for lead for something as intrastructural as electric transmission. If it makes ammunition unaffordable especially.
posted by ocschwar at 7:39 PM on July 26, 2023 [3 favorites]

Much like how existing power lines have made aluminum unaffordable.
posted by ryanrs at 7:44 PM on July 26, 2023

Lead in ammunition and in the primer for bullets is one of those things that I think the future will look back at as a reason for our horrible politics.
posted by interogative mood at 7:49 PM on July 26, 2023 [3 favorites]

are we instead on the very bottom of a very sharp upward curve?

100% bottom of the curve. We've had widespread indoor plumbing for less than 100 years, and our understanding of solid state physics is even more recent.

Generally speaking, we've only explored relatively simple mixtures of a few compounds at a time. Even systems as well studied as steel alloys are still making new discoveries and big advances (like better energy-dissipating steels for making cars safer).

To explore an alloy system of N elements is an N-dimensional problem. And we know that minute changes in composition can result in dramatic changes in physical properties (like adding a little carbon to iron). So you want to explore that huge N-dimensional space quite thoroughly. And we haven't. Not even close. It's a big space.

Check out what we're learning about mixtures of 4 or 5 metals in high-entropy alloys.


That's not to say this discovery is real (it probably isn't). But it could be! Human understanding of complex materials is quite weak in 2023.
posted by ryanrs at 8:36 PM on July 26, 2023 [19 favorites]

I have some bad news to give you about the use of lead in lead acid batteries, in ammunition, and lead as a byproduct of thorium reactors.

The U.S. environmental laws for lead have been extremely successful. Remember lead water pipes and leaded gasoline? Or even just lead pencils?

We’ve spent 75 years removing lead from everything we can, but there are a few things we think are valuable enough to keep using it for. It would be an unfortunate step backward if this turned out to be one.
posted by Tell Me No Lies at 9:35 PM on July 26, 2023 [1 favorite]

in fairness, "pencil lead" was never actually a literal thing ("black lead" was just an old name for graphite)
posted by DoctorFedora at 10:21 PM on July 26, 2023 [22 favorites]

I've seen a few reputable people on X xeeting that this is exciting and different from cold fusion in terms of legitimacy.

Having not only been studying chemistry during the Pons and Fleischmann saga, but taking a class from a nuclear chemist of considerable accomplishment at the time who was giving us updates every class, I can guarantee you at the time that reputable people were saying that cold fusion was exciting and quite different from polywater.

It took about six weeks for the consensus to turn firmly against it. (Broadly speaking, physicists and people who loved theory were far more skeptical from the start; chemists and experimentalists tended to have more faith that a pair of chemists were competent enough to work a fucking calorimeter*.)

Anyway, I'm certainly not using that to predict this will fail! I'm definitely rooting for it. Very much wait and see though; I don't feel any need to speculate about the quality of the work.
posted by mark k at 10:28 PM on July 26, 2023 [5 favorites]

Yeah, it's a little whiffy. But if we do get room temperature superconductors, they're surely more likely to come out of Korea than Roswell.
posted by flabdablet at 10:34 PM on July 26, 2023

but if you need an actual Roman-style lead stylus for the scribings
posted by away for regrooving at 12:29 AM on July 27, 2023

Remember lead water pipes

Indeed.
posted by clawsoon at 1:04 AM on July 27, 2023 [1 favorite]

MetaFilter: 2-4 random academics half reading a paper and firing off a cranky opinion
posted by gelfin at 3:52 AM on July 27, 2023 [18 favorites]

in fairness, "pencil lead" was never actually a literal thing ("black lead" was just an old name for graphite)

Wait, those pills the doctor said would “put lead in my pencil” had graphite in them?
posted by GenjiandProust at 4:34 AM on July 27, 2023 [3 favorites]

Please be legit. It would unlock untold trek tech wonders while also upgrading the efficiency of existing tech. Maybe we'll get the cool solarpunk future after all.
posted by neonamber at 7:53 AM on July 27, 2023 [5 favorites]

in fairness, "pencil lead" was never actually a literal thing ("black lead" was just an old name for graphite)

I did not know that. I always thought there was a switch at some point. Interesting.
posted by Tell Me No Lies at 9:31 AM on July 27, 2023

>Remember lead water pipes

Indeed.

I wonder how much that explains about Chicago.
posted by Tell Me No Lies at 9:32 AM on July 27, 2023

So, hoverskates?

You just had to make this weird for me, didn't you?
posted by loquacious at 10:48 AM on July 27, 2023 [3 favorites]

Short video of playing with a sample with a magnet

That looks like plain old paramagnetics. With permanent magnets that large you can get a lot of different non-magnetic materials to move like that through induced eddy currents, including copper, aluminum, gold, liquid oxygen and many more. Basically anything conductive will exhibit some paramagnetic effects from induced eddy currents.
posted by loquacious at 10:52 AM on July 27, 2023 [3 favorites]

Video specifically says the stuff is deposited on a copper plate. So yeah, any copper plate will do that. It's the same as spinning copper tube levitation, or magnetic brakes on roller coasters.

Notice how the video did NOT demonstrate static repulsion. I want to see the plate and the magnets motionless, with static repulsion visible as an angle on the string.

Instead, when the magnet is held still, what we see is a strong damping/braking effect from eddy current losses.
posted by ryanrs at 11:13 AM on July 27, 2023 [4 favorites]

Basically it proves the sample is resistive, ha ha.
posted by ryanrs at 11:16 AM on July 27, 2023 [3 favorites]

@ryanrs exactly my criteria. Exciting claims demand exciting demos. Levitate a sample over a stationary magnet for a few minutes or GTFO.
posted by nickzoic at 6:13 PM on July 27, 2023

Although pyrolytic graphite will also float above strong magnets, so that's not proof. It would be a lot better than that video, though.

IMO the most damning thing about the video is that if this material was real, they'd have been able to make a better video, heh.
posted by ryanrs at 7:09 PM on July 27, 2023 [3 favorites]

Short video of playing with a sample with a magnet

It would be interesting to know more about the video, like why it was posted in January.
posted by Tell Me No Lies at 10:41 PM on July 27, 2023 [2 favorites]

It would be interesting to know more about the video, like why it was posted in January.

Superconductors enable time travel, obvs.
posted by clawsoon at 4:51 AM on July 28, 2023 [2 favorites]

[mind explodes]
posted by Tell Me No Lies at 6:42 AM on July 28, 2023 [1 favorite]

So are we agreed that if this isn't replicated before the thread closes, then it isn't real? There are some aggressive timelines upthread.
posted by ryanrs at 6:02 PM on July 28, 2023

What I’ve been seeing on Twitter has suggested it’s probably not a superconductor at room temperature, but that it is at least potentially an interesting new material
posted by DoctorFedora at 10:05 PM on July 28, 2023

There is some controversy over how this was publicized. Apparently, Young-Wan Kwon is no longer working with the team and publicized the paper without their permission.

Google translation of Korean article.
posted by eye of newt at 12:39 AM on July 29, 2023 [3 favorites]

Posting a paper without your coauthors' permission is extremely bad.

To be honest the various little notes in Korean that had not yet been edited out of the second preprint should have been a bit of a tip-off...
posted by heatherlogan at 10:11 AM on July 29, 2023

Ugh, that's extremely rude to do, especially for something so controversial that can seriously damage your coauthors' reputations.
posted by tavella at 11:09 AM on July 29, 2023

There is some controversy over how this was publicized. Apparently, Young-Wan Kwon is no longer working with the team and publicized the paper without their permission.

This is now very strange. The first preprint was submitted by Young-Wan Kwon on Saturday July 22, and hence would have become public at 8pm US Eastern time on Monday July 24. The authors listed on the first preprint are Sukbae Lee, Ji-Hoon Kim, and Young-Wan Kwon.

The second, longer, preprint was submitted by Hyun-tak Kim a little over two hours later, also on Saturday July 22, before the first preprint became public. The authors listed on the second preprint are Sukbae Lee, Jihoon Kim, Hyun-Tak Kim, Sungyeon Im, SooMin An, and Keun Ho Auh.

It seems pretty obvious that these two submissions were coordinated, because there is no way that an unconnected observer could have known about the first submission without being directly told by the submitter.* The second preprint was submitted by a different person and doesn't even include Young-Wan Kwon as a coauthor. Blaming a "premature posting" on Kwon seems disingenuous at best, given the existence of the second preprint.

*Source: I have submitted papers to the arXiv and know first-hand how it works.
posted by heatherlogan at 2:16 PM on July 29, 2023 [1 favorite]

(The little notes in Korean that appear in a couple of places in the second preprint seem to be Microsoft Word error messages translating as "error! No bookmarks are defined." and probably represent broken links to references. So, nothing particularly interesting.)
posted by heatherlogan at 2:31 PM on July 29, 2023 [1 favorite]

Ugh, sorry, the second preprint is not actually longer than the first.
posted by heatherlogan at 2:34 PM on July 29, 2023

Comprehensive explainer from the Some Weekend Reading blog, current as of today.
posted by heatherlogan at 7:37 PM on July 29, 2023 [6 favorites]

Great explainer, and a dismaying dip into weird personal/professional drama.

But this is good:

Furthermore a group at US Dept of Energy run Argonne National Laboratories is racing to perform a verification in the coming weeks.
posted by ryanrs at 11:28 PM on July 29, 2023

there is no way that an unconnected observer could have known about the first submission without being directly told by the submitter.

Spooky action at a distance?
posted by TedW at 11:34 AM on July 30, 2023 [1 favorite]

The second preprint was submitted by a different person and doesn't even include Young-Wan Kwon as a coauthor. Blaming a "premature posting" on Kwon seems disingenuous at best, given the existence of the second preprint.

Dunno. If Young-Wan Kwon told someone he had posted it some very pissed off colleagues could have posted the new one so that a) the earlier one wouldn't be used for reference, and b) they could put their "proper" author list on.

If this paper pans out it will be one for the history books, so I can see jockeying to get your name on it.
posted by Tell Me No Lies at 4:40 PM on July 30, 2023 [1 favorite]

There's a new paper tonight on the theoretical side, reporting some computations that support this compound as a good candidate for high-T superconductor (or seem to, anyway; I'm no expert, so can't really assess.)

Also tonight there are two papers (1, 2) reporting that they followed the protocol but the product they got isn't a superconductor.
posted by em at 9:31 PM on July 31, 2023 [4 favorites]

I don't know enough physics to parse the competing claims and speculation I'm seeing online, but it seems notable that the primary betting market tracking the viability of LK-99 has spiked from the teens on July 29th to ~45% in the last couple of hours.

edited to add: this thread on forums.spacebattles.com (where else?) has an updating tracker of ongoing professional and amateur replication attempts from around the world. Link is to most recent update; refresh and head for the later posts for new news.
posted by Rhaomi at 9:50 PM on July 31, 2023 [3 favorites]

There's a new paper tonight on the theoretical side, reporting some computations that support this compound as a good candidate for high-T superconductor (or seem to, anyway; I'm no expert, so can't really assess.)

a plain english explanation of that paper (although i too am not an expert, so can't vouch for its accuracy).

another paper that has been described as supporting some of the original authors' claims.

there is also a claimed successful replication from the huazhong university team, which seems to be a retry, not the partial success referenced in the spacebattles roundup (video and brief summary here from the same guy that provided the plain english explanation above). albeit there seem to be some caveats, as noted in that link.
posted by inire at 4:34 AM on August 1, 2023

With absolutely zero expertise I point out that the cautionary voices are suggesting that the properties the simulation found are also the properties of a semiconductor. We don’t really need another semiconductor.
posted by argybarg at 9:04 AM on August 1, 2023

It seems pretty obvious that these two submissions were coordinated, because there is no way that an unconnected observer could have known about the first submission without being directly told by the submitter.

Out of curiosity, would the other people on the author list have received notification?
posted by Tell Me No Lies at 10:32 AM on August 1, 2023

The whole thing is just bananpants.

The Race To Validate is a tweet/X/rollup of the last few days, featuring the greatest ending imaginable:

🇷🇺 Russia: Soil scientist and anime catgirl Iris, stumbles upon the papers and reads them casually on Friday morning. Russian scientific practice when replicating Western papers is to deconstruct the science into abstraction. Not having the same access to equipment, Russian inquiry delves into the why much more deeply, to concentrate on the essence required to produce the outcome. With that starting point, cursing and swearing in inimical Russian style, she recognizes the smokescreen of the method in both papers immediately...

If you're X averse, I put it into a gist here as well. It reads like a movie script, seriously.
posted by jquinby at 12:36 PM on August 1, 2023 [1 favorite]

Could that be because that's exactly what it is?

Ate-a-Pi
@8teAPi
Jul 31
Sources to follow, maybe incomplete, and remember it’s fiction. Having trouble with Twitter app.

posted by flabdablet at 12:49 PM on August 1, 2023

NO I want to believe.
posted by jquinby at 12:52 PM on August 1, 2023 [1 favorite]

I don't know what they mean by "it's fiction". Maybe just because of the writing style? Because a lot of the things described there are certainly not fiction.
posted by mr_roboto at 12:57 PM on August 1, 2023

Great recent synopsis of latest simulations and reproductions from science.org
posted by butterstick at 1:57 PM on August 1, 2023 [3 favorites]

The two analyses that suggest that the physical basis for the potential superconductor is reasonable seem promising?
posted by tavella at 3:00 PM on August 1, 2023

It’s fictionalized, like a movie script. See this tweet.
posted by wemayfreeze at 3:21 PM on August 1, 2023

Derek Lowe has written a second blog post, still optimistic.
posted by automatronic at 4:23 AM on August 2, 2023 [1 favorite]

if it turns out to be real, what are the limits on its application implied by the reported 200 milliamp critical current limit?
posted by snuffleupagus at 10:40 AM on August 2, 2023

If it turns out to be real it points in the direction of various other materials that might work even better. If it works, the likelihood LK-99 on its own will be the room-temp superconductor of choice is practically zero.

But if it works (even in a mediocre way), that means thousands of researchers can focus directly on similar types of structures that sort of self-compress when there are substitutions in the alloy that create superconducting channels.

We've barely BARELY scratched the surface of what material science is capable of simply because the option space is so immense. If this is replicated, we can change from searching a vast room with a flashlight to examining a tiny speck of the wall with a laser.
posted by tclark at 11:17 AM on August 2, 2023 [5 favorites]

There's also a lot you can do under 200 mA. All currents between the circuit elements of a microchip are much smaller than this. And for applications that require higher currents, you could set up superconducting systems in parallel. This might be very practical for energy storage applications.
posted by mr_roboto at 11:47 AM on August 2, 2023 [2 favorites]

The application that I'm most excited about is lecture-demo equipment and TOYS! [Probably want to coat it in epoxy or something because of the lead, though.]
posted by heatherlogan at 1:47 PM on August 2, 2023 [1 favorite]

Looks like the results have been replicated.
posted by Ipsifendus at 3:16 PM on August 2, 2023 [4 favorites]

Probably want to coat it in epoxy or something because of the lead, though.

NO GONNA EAT A BUNCH TO GET SUPERCONDUCTORPOWERS
posted by GCU Sweet and Full of Grace at 3:51 PM on August 2, 2023 [4 favorites]

If something similar could be cooked up for use in a CPU I'd think it would allow for much faster changes in voltage allowing for higher clock speeds.
posted by VTX at 4:26 PM on August 2, 2023

Looks like the results have been replicated.

There's nothing new over the past 24 hours at least. The Huazhong group has been saying they've replicated it for a couple of days now, but that video ain't much; we need some real materials characterization.

I talked to a couple of my Materials Science colleagues about this; both roughly in the field of solid-state materials synthesis. One is completely dismissive, thinks it's complete nonsense. The other is working on a team to reproduce it.
posted by mr_roboto at 5:05 PM on August 2, 2023 [3 favorites]

From the Wikipedia article:

The original published articles do not claim to have seen definitive features of superconductivity, zero resistance and the Meissner effect, but show the material exhibiting strong diamagnetic properties, including a video of sample of the material partially levitating on top of a large magnet,[9] which is correlated with superconductivity.

So … we have a substance exhibiting some degree of diamagnetism. Is that all? Is that enough to warrant this degree of attention? Are we being had yet again?
posted by argybarg at 1:33 PM on August 3, 2023 [1 favorite]

Eiri Sanada is keeping a regularly-updated log of the various replication attempt. They helpfully include info about which accounts have had their personal or institutional affiliations verified.

Several people have now produced grains of material that show diamagnetism (i.e. magnetic repulsion that does not depend on the orientation of the magnet). A group at Huazhong University of Science and Technology published a preprint yesterday about their levitating sample, and Andrew McCalip (an individual who works at a US space manufacturing startup and has been livestreaming his attempt) posted a video of his levitating sample today. A group at Southeast University in Nanjing reported that they have a sample that they measured as superconducting at 110 Kelvin (−163°C), far below “room temperature” but would still make it at least a new high-temperature superconductor.

The samples all appear to have only small regions with the anomalous behaviors, as it seems hard to get a larger uniform piece. There is also probably a high chance of measurement errors in these preliminary reports, since people haven’t had time to check for problems and refine their setups. The odds in betting markets are still 2 to 1 against.
posted by mbrubeck at 9:17 AM on August 4, 2023 [2 favorites]

An article published today in Nature by Dan Garisto discusses the various replication attempts, and also some reasons for skepticism.
posted by mbrubeck at 10:32 AM on August 4, 2023 [1 favorite]

Marius Gundersen
@[e-mail]

@megmac I'm no physicist, so I'm speculating wildly here, but due to the structure of the crystal might it be that it is only superconducting in one dimension, parallel to the crystal structure? If so, might that be why it turns sideways? I don't know, like I said I'm no physicist
Aug 04, 2023, 12:07

Marius is correct! A material that is superconducting in only one direction will align that axis along the ambient magnetic field, and not expel the field the way a two- or three-dimensional superconductor does. (I have a delightful proof of this which is too long to fit in the margin.)

This explains why the "semi-levitating" samples would "stand up" rather than fully levitating...
posted by heatherlogan at 3:48 PM on August 4, 2023 [2 favorites]

video shows purported flux pinning of LK-99 sample (nitter link w/ embedded video & text by some westerner, not the video producer; original apparently from a pseudonymous researcher on a Chinese streaming site per discussion on the orange site)
posted by the antecedent of that pronoun at 7:01 AM on August 5, 2023

If the flux is pinned they ought to be able to move the magnet and the sample would remain motionless relative to it. That would be a bit more convincing than nudging it, at least to me. Although nudging it and having it jump back into one place is pretty much the same thing. I just like dramatic demonstrations.
posted by seanmpuckett at 9:31 AM on August 5, 2023

Investigators seem to be increasingly leaning towards alternate, non-superconducting explanations for the material’s behavior. Justine Calma, The Verge: LK-99 hasn’t turned out to be the miraculous superconductor some people initially claimed it was.

Betting market odds have lengthened to about 10 to 1 against.
posted by mbrubeck at 5:13 PM on August 10, 2023

Hmm. I don't have any direct gossip but it does seem the skepticism is increasing, as per this Science roundup as well.

Materials science is weird, though, it could still turn out the replication problems are actually fabrication problems and they're not getting the correct material state.

If not, having lived through cold fusion real time, this is very similar to the path cold fusion took, including the new theoretical justification researchers put out.
posted by mark k at 7:24 PM on August 10, 2023

Even the “new theoretical justification” was rather misrepresented in the popular discourse. Sineád Griffin, the author of the LBNL paper that was often cited as supporting the superconductor claims, tweeted more than a week ago, “My paper did *not* prove nor give evidence of superconductivity in Cu-apatite.”
posted by mbrubeck at 8:10 PM on August 10, 2023

What's going on with the original Korean team and the samples they synthesized? Have they given them out to anybody else for independent testing? Seems like that would clear up any possible issues with replication errors.
posted by Rhaomi at 8:33 PM on August 10, 2023

The Korean team announced a few days ago that they’ll send samples to the College of William & Mary in Virginia. Two of their collaborators are physics professors there.
posted by mbrubeck at 6:49 AM on August 11, 2023 [1 favorite]

LK-99 isn’t a superconductor — how science sleuths solved the mystery

Thus adjourns the August meeting of the internet science skeptics club. Same time next month?
posted by pwnguin at 1:11 PM on August 16, 2023 [13 favorites]

Unless the original team pops up with some superconducting LK-99 I'd say this is pretty much done with. Pity.
posted by Tell Me No Lies at 8:59 PM on August 16, 2023

Setting aside people who were foolish enough to go buy stock options based on the news after they popped, I feel like this affair was largely handled in a healthy way?

It generated a wave of interest that did some popular education around both the basic science of superconductors and the engineering/economic/environmental potential of the breakthrough, gave us something to feel optimistic about for a minute, and the enthusiasm was tempered with regular reminders to wait for replication such that I don't think the letdown will backfire in terms of popular perceptions of the value of basic research, etc.
posted by snuffleupagus at 6:55 AM on August 17, 2023 [2 favorites]

In summary: Fucking magnets, how do they work?
posted by kaibutsu at 9:46 PM on August 17, 2023 [1 favorite]

I'd say this is pretty much done with.

Metafilter: apatite for destruction
posted by flabdablet at 3:09 AM on August 18, 2023 [1 favorite]