/sci/ - Science & Math

>>15253535
guess /sci/ doesn't give a shit about room temperature superconductors.

>>15253678
It hasn't happened yet and there's so much fraud in the field that nobody knows.

>>15253685
it's happening tomorrow, which means there's the chance we can get our worst shitposters on the scene to ask dumb questions.

>>15253535
>Near Ambient Pressure*
>*
I guess * means ~100GPa
Clickbait title to put butts in seats.
If they go below 90GPa then I'll admit they made a bit of progress.

>>15253678
It's by the same pajeet that produced the previous discredited papers. No one believes him since he never releases his data and what is available is obviously flawed (the same graph appears in two difference papers for two completely difference materials - suss!).

The only valid argument is that over why Nature accepted the paper.

>>15253808
There was also a scandal where Nature retracted the paper from Dias and his team over lack of necessary data. I think arxiv also did some shit (they banned someone else for publishing a critique that also contained some accusations of faking data).
https://www.science.org/content/article/something-seriously-wrong-room-temperature-superconductivity-study-retracted
Apparently they are not observing superconductivity directly and are instead measuring magnetic susceptibility of their extremely small sample in a noisy environment (relative to the signal).

4 Kelvin
That's the warmest you can get

>>15253808
OP here, forgot to post the abstract:
https://meetings.aps.org/Meeting/MAR23/Session/K20.2
they synthesize the material at high pressure and decrease it.

>>15254149
>they synthesize the material at high pressure and decrease it
It doesn't say that.
From reading a 2023 paper https://arxiv.org/ftp/arxiv/papers/2302/2302.08622.pdf
my guess is they synthesize at kinda high pressure (~4GPa) to grow crystals then pressurize the fuck out of it (up to ~ 100GPa) to get superconductivity.

Stop falling for the sales pitch. The vagueness should tip you off that they are overselling.
>near ambient pressure*
>With these material, the dawn of ambient superconductivity and applied technologies has arrived
>High pressure physicists hate him. Learn 1 weird trick to compress farts back into fecal matter to be deposited in the streets discovered by a single ESL slumdog millionaire trying to extract money from the NSF.

In another life, this guy would be writing clickbait emails for phishing scams out of india.

If dubs it's real!

>>15255505
And this post did not have double digits, therefore it is not real. Only a couple of hours till the talk now.

Anyone got a stream?

>>15256714
I cant find a vid of the talk.
There is a vid on twitter about the strict security for this talk specifically.
https://twitter.com/PramodhYapa/status/1633244437034979329
Also someone said they skipped audience questions
https://twitter.com/sgtmeeow/status/1633249026757001216

>>15253535
>Is room temperature and near room pressure superconductivity here or is this another Fleischmann–Pons
No
>Will there be a huge screaming match at this talk tomorrow
Also No, most the hype and excitement around superconductors was in the late 80s and early 90s when the field was new. Everyone who's researching them since is doing so under the knowledge that progress will likely be slow.

>>15253678
You have to understand that this board gets a lot of tourists, crossboarders, schizos, and brainlets who don't know much about science but will talk a lot about it anyway

>>15253535
>at near ambient pressure*
>*
kek

>>15253535
> another Fleischmann–Pons
Explain plz.

ttps://www.sciencenews.org/article/superconductor-room-temperature-scrutiny
It superconducts at 294 K and 10 kilobar. The pressure's pretty high, but this isn't diamond anvil territory anymore. In fact this pressure is less than what is used to make diamonds. And we make those at high temperature too.

>>15257087
Yeah but their nature article got retracted. They're likely just fucking up the experiment. I doubt they shattered the record for room temp superconductivity by cutting a pressure by six orders of magnitude

>>15253535
>is this another Fleischmann–Pons
you might wanna check back in on Fleischmann–Pons... just sayin'

It gets better. Hirsch published a comment ripping into Dias (Room-temperature superconductivity - or not?). Analyzed his data, found the measured data before background subtraction is more easily explained by adding noise to a desired superconducting measurement, and called his data "pathological."

Hirsch gave a talk today stating that there's "ZERO" evidence of room temperature superconductivity. In the same session as Dias. I'm so stoked for the recording bros. Also keep an eye out for a recording of a talk in that session that was cancelled, giving the audience a chance to ask questions that was skipped after Dias' talk.

>>15253535
Room temperature superconductors are here already. Just reduce the temperature of the room to about 80K

>>15257397
Do-ho-ho

>>15256833
>Everyone who's researching them since is doing so under the knowledge that progress will likely be slow.
So so much to unpack in this statement.
It truly activates the almonds

>>15257466
That's just how most scientific discovery occurs, long stretches of incremental progress occasionally interspersed with large breakthroughs. It can also affect interested in a field early superconductor research in the late 80s cause a lot of enthusiasm and got a lot of people looking into it, but after a while and the low hanging fruit was gone only the diehards who are super autistic about super conductors kept at it.

>>15253535
Good shit. Gunna lurk til archived.

>>15253535
77K superconductivity is all you need. Liquid Nitrogen is literally free. You can just take it from the air.

Room temperature or room pressure it doesn't really matter. What matters is understanding.
Superconductivity is more important for what it means that for something as crude as transmitting electrical current. Understanding superconductivity is the first step to true mastery of condensed matter.
It's our version of transmuting base metals to gold. Alchemists wanted to do that, not to counterfeit gold, but because they thought gold was important and they wanted to gain understanding.

>>15257944
Anon... How do you propose to keep 30km of power cables submerged in liquid nitrogen?

>>15257971
Cash.

>>15257971
It's easy in space, just set up your radiators.

>>15258055
They would have to be the size of a building.
>Notice how a heat sink using air isnt as effective as one in a liquid? Add a few zeros....

>>15258057
So? In space volume is free.

>>15258059
Idk...how many atoms are passing by those sinks? Or is the sink so big its whats dissapating the heat, because at the level its gunna need to be...yeah, it might not need it at all.

>>15258059
Look up "how fast a body freezes in space". Thats how efficient its going to be.

Place your bets.
>*
VS
>hype
https://www.youtube.com/watch?v=XhhvOMuLF94

>>15258132
THANK YOU! I looked for it but couldnt find it.

>>15258132
tl:dw
It is claimed Nitrogen doped Lutetium Hydride superconducts at room temperature around 10 kilobar = 1GPa
>big if true

I have no frame of reference for this, just a big fan of material science.

>>15258132
well well... this will be fun to watch

>>15257944
The energy needed to cool nitrogen into a liquid ain't though.

and now it's been published in nature https://www.nature.com/articles/s41586 dash 023-05742-0
It's big if it doesn't get retracted. They appear to have gotten it to superconduct outside of a diamond anvil. At least they did the magnetic susceptibility testing in something that wasn't a diamond anvil. The material also appears to be stable at room pressure, it doesn't evaporate away. The sample preparation section seems a bit light though.

>>15258181
>pajeet
>fun to watch

>>15259597
i wouldn't call 10kbar (~1 giga pascal) an ambient pressure. can you achieve that without diamond anvil cells?

>>15259635
yes, they used an HMD(brand) pressure cell for magnetic susceptibility testing. It's just metal and plastic, no gems. 1 gigapascal pressures aren't too crazy. Such pressures are used to make molecular machines:
https://pubs.rsc.org/en/content/articlelanding/2013/CC/c3cc42201a
And all the experimental petrology(making lava,torturing rocks) people are able to get higher pressures that at hotter temperatures.

The comments section on this blogpost seems to find major flaws with the work. Apparently they never actually measured zero resistance, they just got it down to a fixed level which they then subtracted, declaring that it was "residual instrument resistance"
https://nanoscale.blogspot.com/2023/03/aps-march-meeting-2023-day-2.html
Imagine trusting this guy.

>>15259839
>He doesnt strictly do solo self-found research like I do on Diablo 2.
>Do everything right, still go down with the ship.
lmfao

>>15259597
Yeah I'm thinking this is bullshit.
The split axis scaling on the transition temperature graph is straight-up malicious and their provided dataset doesn't even reach the 298K transition.
The transition step size for the big result also fits within the uncertainties for the lower transitions. Guess it's Dias being Dias again and trying to scam investors

gonna bump to see what other information I can pump out of /sci/. You know what'd be fucking funny? If this works and leads to a superconductor that works at room temperature, but the critical current and field are so fucking low, it's useless.

>>15262852
>requires significant pressure (even if very low for high-pressure experiments)
>crystal is challenging to grow in large quantities
>superconducting transition of 1 (one) milliOhm on top of significant residual resistivity
Even if it isn't a complete fabrication this is a nothingburger and no closer to useful ambient superconductivity than the yttrium-palladium-hydron meme at 187 GPa

the undoped hydride only superconducts at 200 GPa, with Tc=70K https://arxiv.org/abs/2303.05117

I wonder if throwing some nitrogen in the mix could really make such a drastic change? ...

bump, is /sci/ too busy to talk about actual science?

>>15265736
I guess if one is outside of the field it is a little difficult to talk about the in-depth workings of a given paper. In this case the background of the individuals has attracted more discussion. I'll put in a high-effort post, though this is just a compilation of sources that have been mentioned in links posted ITT.

There's a lot of mainstream media attention given to this. Here's articles from NYT and WSJ, so Dias is basically a household name now:
https://www.nytimes.com/2023/03/08/science/room-temperature-superconductor-ranga-dias.html
https://www.wsj.com/articles/superconductor-breakthrough-energy-reddmatter-90dfa165

Science-specialized media have been more cautious in their reporting. They raise no specific issues with this paper but mention Dias' background, with the 2020 Nature retraction being the most famous issue. However, his other work has also met scrutiny.
https://www.quantamagazine.org/room-temperature-superconductor-discovery-meets-with-resistance-20230308
https://physics.aps.org/articles/v16/40

Specifically, two old papers that Dias was on have PubPeer comments from a few weeks back (pic). They make allegations that these papers contain data from Dias' thesis, which is not an issue in itself. However, in one case the data in the paper is noted to be identical to data in his thesis, though labeled as belonging to an entirely different material. In another case there is a measurement which should be the same material (and the same pictures of the sample are given) but the measured data is very different between the thesis and paper. Further on the thesis, it has been alleged that it contains several pages' worth of section plagiarized from another person's thesis.

>>15266198
(continued)

The latest PubPeer comment pertains to Dias' previous work on metallic hydrogen. Dias got a Science paper out of this (https://www.science.org/doi/10.1126/science.aal1579)), though this too has attracted criticism and even claims that they didn't make metallic hydrogen at all (see e.g. the Wikipedia page on metallic hydrogen). That paper is still up however.

An interesting twist is that Dias has a startup called Brilliant Matters, which the quanta article talks about more. Briefly, it is stated they have raised $20 million in funding. Somewhat oddly Dias listed some big-name CEOs as investors who were not in fact investors. He later explained this as being basically a wishlist rather than actual investments. I have also read that Dias has used the startup and related intellectual property claims as justification to be less fortright about his exact procedures than one might expect for a discovery of this magnitude. However, I don't know exactly what is missing in terms of other teams reading the paper being able to reproduce findings.

Lastly, the nanoscale blog comments section now has a link to raw data uploaded by Dias, as well as some fairly direct allegations of data manipulation or, less severely, misinterpretation, including several by the coauthor mentioned in this thread. Of course, the commenters there are just as anonymous as we are, so I don't know if they know what they are talking about. But if someone is an expert in the field they could now access raw data.

>>15266209
>>15266198

I should also point out that one of Dias' most ardent critics, Jorge Hirsch, is not entirely impartial as he has his own unconventional superconductivity theories according to which hydrogen cannot be a superconductor. From what I gather he has a reputation as a rather abrasive personality to put it nicely. The NYT article talks more about this.

Ultimately it comes down to whether or not the results can be replicated.

>>15266232
>>15266198
(continued, deleted and reposted this because I got a name wrong)

The latest PubPeer comment pertains to Dias' previous work on metallic hydrogen. Dias got a Science paper out of this (https://www.science.org/doi/10.1126/science.aal1579)), though this too has attracted criticism and even claims that they didn't make metallic hydrogen at all (see e.g. the Wikipedia page on metallic hydrogen). That paper is still up however.

An interesting twist is that Dias has a startup called Unearthly Materials, which the quanta article talks about more. Briefly, it is stated they have raised $20 million in funding. Somewhat oddly Dias listed some big-name CEOs as investors who were not in fact investors. He later explained this as being basically a wishlist rather than actual investments. I have also read that Dias has used the startup and related intellectual property claims as justification to be less fortright about his exact procedures than one might expect for a discovery of this magnitude. However, I don't know exactly what is missing in terms of other teams reading the paper being able to reproduce findings.

Lastly, the nanoscale blog comments section now has a link to raw data uploaded by Dias, as well as some fairly direct allegations of data manipulation or, less severely, misinterpretation, including several by the coauthor mentioned in this thread. Of course, the commenters there are just as anonymous as we are, so I don't know if they know what they are talking about. But if someone is an expert in the field they could now access raw data.

>>15258055
not even wrong despite what the scienlets say. The very same mechanism (lack thereof) they claim would make this impossible is what would make it trivial. It's hard on earth because of all the conduction/convection from the environment dumping heat into your system, but in space you just need a high albedo side facing the sun and a high emissivity side facing everywhere else.

Another simple question is why doesn't someone like deepmind simply throw a massive amount of computing power to fully explore the hydrides space.

Let's say it's some 100 000 combinations, throw in some doping which makes them ×10, it's not undoable for them to do phase exploration of all of this.

Instead we have to wait for wang et al to do one at a time like in the 90s

>>15266366
Two questions come to mind, specifically how much resources one such calculation takes and how accurate they are. My own limited experience comes from computational modelling of excited states on single molecules. These calculations could take weeks on a dedicated computing cluster. Of course, only a fraction of the cluster was used on my calculations, but even so going up six orders of magnitude as you described would be not feasible to say the least. Even if the work is high-impact, the other work might be as well. These are shared resources and you can't just tell someone to give all their computing facilities to you for weeks.

Moreover, the calculations I performed were a hit-and-miss match with reality. There isn't such a sharp divide between nonsensical results and "true" results, most commonly the calculations would correctly capture certain trends and effects but might wrongly predict certain orderings, couplings and energies in a way that has a large impact on the observed behaviour. You had to constantly link calculations to experimental results to anchor yourself to reality and figure out where the calculations are failing, you cannot solely rely on modelling.


Incidentally, someone on the blogspot is claiming that Dias essentially is doing what you described:
> Dias is basically randomly smashing elements based on the latest computational trends, hoping for more simulations to eventually back him up (and publishig whatever phase transition he gets in the meantime). To be future proof, he masks the exact measured stoichiometries and structures behind "patents pending".

>>15266419
Sure but I mentioned deepmind because they have access to idle google clusters. Whenever one of them isn't busy, you can sneak in stuff "for free" (they're paying for it anyway), it's just a software engineering problem

Similarly for other big tech companies.

Academics couldn't do this because they would have to book a top10 supercomputer for months, and no funding agency would allow that

>>15259597
There's now quite a few comments on the page of the Nature article, which talk about discrepancies in the data. Apparently some interesting background subtractions going on again.

The guy has the same physiognomy as Liz Holms' partner. Notice that everyone connected with him is another Lankan/Indian. And of course no one can actually reproduce anything this lying scammer has to say.

Is there a proposed structure for the material? They have PXRD? Stability of the sample? For claims like this, it should be sent to other labs to reproduce as part of the review process. Calling your buddies who you totally aren't on good terms with or anything to "inspect" the results is not enough in this instance.

>>15257899
>That's just how most scientific discovery occurs, long stretches of incremental progress occasionally interspersed with large breakthroughs.
Where did you get this from, Kuhn, or Prokhor Zakharov?

>>15265736
No, /sci/ is too full of tourists, crossboarders, and schizos to talk about science

>>15253678
>12 mins in a random lecture hall
you tell me.
near ambient pressure doesn't mean room temp though.

>>15269786
It's not just the conference presentation, it's a Nature paper. And it's "near ambient", meaning room temperature and still fairly high pressure but not diamond anvil tier.

>>15269852
>it's a Nature paper
nature is a political propaganda publication just barely described as a scientific. the disguise is only good enough to trick low iqs, everyone else sees right through it

>>15269852
still has to be 'forged' in a diamond anvil though.

>>15269916
Beside the point, you smarmy contrarian chud. Room-temperature superconductivity at 200 times lower pressures than before is massive regardless of where it's published, and you're the one who was bitching about the presentation being too low-profile or whatever.

>>15269991
Diamond requires massive pressures and temperatures to form and is only metastable at ambient conditions but is very useful anyway.