The report was published in Nature on March 8th, 2023:
Dasenbrock-Gammon, N., Snider, E., McBride, R. et al. Evidence of near-ambient superconductivity in a N-doped lutetium hydride. Nature 615, 244–250 (2023). https://doi.org/10.1038/s41586-023-05742-0
This will resolve to yes if an independent replication is published in a reputable scientific journal before 01/01/2025.
For more background, see:
https://www.nature.com/articles/d41586-023-00599-9
06/29/2023 - Clarification on the Meaning of Replication:
I will resolve this market yes if the publication demonstrates superconductivity up to similar standards as in Dasenbrock-Gammon et al., including the Meissner effect.
I will not require an independent synthesis. The publication can use a sample obtained from the Dias Lab. However, the authors must verify that the sample composition is more or less consistent with what is reported in Dasenbrock-Gammon et al.
I apologize for the initial confusion on these issues. Lesson learned.
@craftyvisage are you intending to keep this open if the paper this market is based on is retracted?
@QuantumObserver Yes, the market will stay open in case the publication is reproduced in spite of the retraction.
Next time I'll add a resolves NO upon retraction clause.
Make a new market for other exciting superconductor news:
https://manifold.markets/QuantumObserver/will-the-lk99-room-temp-ambient-pre?r=UXVhbnR1bU9ic2VydmVy
@RobertCousineau Hah! Who would’ve guessed? Thought it was gonna be the usual arguing about shoddy data
Dias Lab PRL paper from 2021 on the superconductivity of MnS2 will likely be retracted.
Seems like there’s good evidence that the data for MnS2 resistivity data from this paper was simply the rescaled Germanium Selenide data from Dias’ dissertation.
@QuantumObserver This pre-print supports the likelihood of replicating the paper. It basically establishes a theoretical basis for why under high pressures meta-stable phases would likely show superconductivity!
I was trying to figure out what metastable really meant here. My experience with metastable molecular states usually meant they relax over time and need to be reinitialized. If that’s the case here it seems like it would be pretty hard to measure a Tc.
Are they states that are effectively “stable” but only under 100 GPa?
Is it possible to deterministically prepare the sample in such a state?
@QuantumObserver Ranga Dias also comments on the paper here and claims that it supports his findings
Another theoretical study regarding the realization of near-ambient superconductivity on Reddmatter. Our experimental observations closely align with Ouyang et al predictions, a SC transition at approximately 34 K, essentially ambient pressure conditions.
I guess the key difference here is that on the one hand we have an experimental claim (by the Dias lab) by a researcher whose word is doubted (& who has a history of retraction in this area) vs a computational claim (in this paper) subject to whatever assumptions (and limitations of those assumptions) that were made in order to run calculations to model things
Looking at the method of the paper in terms of how the numerical calculations were done, it looks highly technical but not very clear to me exactly how to decipher the forest of numerical condensed matter physics terminology.
I gather that you are an expert in this field, do you happen to know the rough approach that Ouyang et al used, and any potential advantages and/or limitations of it?
Hazarding a brief analysis myself, from the opening two lines of the "Method" section from https://arxiv.org/abs/2307.10699:
The plane-wave method, QUANTUM-ESPRESSO [32] pack-
age was adopted in our DFT calculations. The Perdew-
Burke-Ernzerhof formula was selected as the exchange and
correlation functional [33].
Quantum-Espresso seems to point to this: https://github.com/QEF/q-e (docs on gitlab here, capabilities on the main webpage here). Interestingly, it seems to be mainly written in Fortran. The plane-wave method is implemented in one of the cores package of QE it seems, and relates to this; broadly an eigenvalue expansion of solutions to Maxwell's equations under certain assumptions.
Regarding PBE as an exchange correlation functional, there is this answer here (mattermodeling.stackexchange.com) regarding its popularity, emphasis mine.
Here is an introduction to DFT functionals: https://dft.uci.edu/pubs/RCFB08.pdf
In short, the Perdew-Burke-Ernzerhof (PBE) functional is very popular because it is a non-empirical functional with reasonable accuracy over a wide range of systems. While PBE is typically not the most accurate GGA (generalized gradient approximation) functional for a given system, it usually is not too far off either. Empirical functionals offer better accuracy for systems they are parametrized for, e.g. "BLYP has smaller errors for main-group organic molecule energetics", but fail in others.
He just patented it, if that moves anyone's opinion: https://www.science.org/content/article/embattled-physicist-files-patent-unprecedented-ambient-superconductor
Someone on twitter has downloaded and looked at the data from the arXiv article posted below. Looks like this replication is much weaker than presented in the pre-print. I’m planning to download the data myself once the long weekend rolls around.
https://twitter.com/npeterarmitage/status/1672436350233231361?s=20
@QuantumObserver Have discussed the tweets above and the available data with some colleagues who have done similar measurements. I suspect this will turn out to be a failed replication and the “signal” here is bad electrical contact.
this will turn out to be a failed replication and the “signal” here is bad electrical contact.
That’s so sad. Isn’t it hard to get zero resistance on a circuit though?
@AdriaGarrigaAlonso In some cases, if your instruments are not able to source any current (no electrical connection), they will not read any voltage, so you get a fake 0 Ohm signal.
The data in the preprint have what appear to be hallmarks of this flaky contact. You could probably explain it away somehow, but a slam-dunk measurement wouldn’t have any of those weird features.
@QuantumObserver that is indeed a shame if what you say is true. That being said, extraordinary claims require extraordinary evidence, so a standalone measurement by another lab is likely insufficient anyway.
Presumably between now and 2025 things will become clearer as to whether or not the Dias Lab claim retains its current standing.
Certainly superconductivity at the pressures and temperatures they have been claiming would be remarkable, as if I understand it correctly from a layperson's perspective, it would put it in range of certain moderately practical solid state applications (e.g. computer microcircuitry).
I guess we'll have to wait for a few more labs to weigh in (if interest remains to investigate the Dias lab claim) and see if a slam-dunk measurement is forthcoming in the coming 18 months.
... or of course if another lab is able to reliably synthesise the Dias Labs material. Evidently not required for acceptance but that would be amazing - if the results are indeed real.
Not only have they nearly replicated the findings, but this shows that labs are actively working to do so! Likely a full replication occurs in the next 2.5 years.