The articles by Ranga P. Dias (Univ. Rochester, New York, USA) and his group on superconducting hydrides at room temperature are always controversial. His superconducting carbon-sulfur superhydride with Tc ~288 K at 267 GPa was published in Nature in October 2020; but the article was withdrawn (retracted) two years later (LCMF, Oct 16, 2022). is now published in Nature that a nitrogen-doped lutetium hydride is superconducting at 294 K (21 °C) and 1 GPa (10 kbar or ten thousand atmospheres); if the result were credible, it would be a historical milestone in the 21st century. But as with Aesop’s fable of the shepherd and the wolf, no one buys the new result. The fund subtraction method used is the same as in the previous article, although it was one of the reasons for the withdrawal; the figures are from books, so clean and so perfect that they seem unreal. The magnetic susceptibility and specific heat curves have the same problems as those in the previous article; it is impossible not to suspect. How is it possible that the reviewers of Nature Have they allowed an article like this to slip in after the controversy of the previous one? The article above was cited over 600 times (despite the withdrawal), perhaps the editors of Nature they follow Oscar Wilde’s maxim: let them talk about you, even if it is bad.
The crusade towards the holy grail of a superconductor at ambient temperature and pressure conditions began with one at high pressures (170 GPa) in 2015 and goes through one at almost ambient pressure, like the new material. However, anything King Midas touches on hydride superconductivity is surrounded by controversy. We cannot forget the signs of plagiarism in his doctoral thesis, nor the articles that have been withdrawn in high-impact journals due to doubts about his possible manipulation of the data. Dias says that the person accusing him is a troll and that the withdrawal of their articles is unjustified (for this reason they continue to be cited). In fact, we all hope that this is the case and that this new article survives the controversy. The samples studied have a diameter between 70 and 100 μm and a thickness between 10 and 20 μm. Unfortunately, the chemical composition (stoichiometry) of the new material made at very high pressure is not known (it is surprising that the article does not specify the details of the synthesis method): X-ray diffraction points to LuH3-dNe (92.25 %), con LuN1−dHe (7.29 %) y Lu2O3 (0.46%); future studies will have to clarify it. The problem is that with this composition it is impossible to explain the observed superconductivity (the distance between the hydrogens is too great); the current-voltage (IV) curve is perfectly linear (pointing to unconventional superconductivity). The magnetic susceptibility curve (Meissner effect) shows strange (or unexpected) behavior; It is not clear how the reviewers have not required a detailed discussion.
I’d like to ignore all the controversy, cross my fingers and hope that the new article beats all the criticism; Dias will receive the Nobel and the 21st century will change forever and for the better. The article is Nathan Dasenbrock-Gammon, Elliot Snider, …, Ranga P. Dias, “Evidence of near-ambient superconductivity in an N-doped lutetium hydride,” Nature 615: 244-250 (08 Mar 2023), doi: https: //doi.org/10.1038/s41586-023-05742-0; more informative information in ChangQing Jin, David Ceperley, “Hopes raised for room-temperature superconductivity, but doubts remain,” Nature 615: 221-222 (08 Mar 2023), doi: https://doi.org/10.1038/d41586- 023-00599-9; Dan Garisto, “Allegations of Scientific Misconduct Mount as Physicist Makes His Biggest Claim Yet,” APS Physics 16: 40 (09 Mar 2023) [web]; y Sophia Chen, «Muted Response to New Claim of a Room-Temperature Superconductor,» APS Physics 16: 39 (09 Mar 2023) [web].
[PS 15 Mar 2023] At the end I have added two figures that appear in the comments of Douglas Natelson, “APS March Meeting 2023, Day 2,” Nanoscale Views, 07 Mar 2023; I have reproduced the figures with the data of the article and I can confirm that they are correct. [/PS]
This figure (above) shows the superconducting dome: the critical temperature (Tc) as a function of pressure (in kbar, kilobars). Superconductivity is observed in the so-called phase II, in which the sample has a pink color (Pink in the figure, below) between 3 and 30 kbar; At lower pressures the sample is metallic with a brilliant blue color (Blue) in phase I and at higher pressures it has a red color (Red) in phase III, which is not superconducting. Color changes are associated with phase changes (somewhat reminiscent of the retired article in Nature).
As usual in articles on superconducting hydrides, the crystalline structure is not known. Thus it is difficult to confirm using computer simulations that the material is superconducting. This figure shows the putative compact cubic structure (
Magnetic susceptibility is key to demonstrating that a superconducting candidate exhibits the Meissner effect. Rare-earth hydrides are expected to be type I superconductors, exhibiting perfect diamagnetism below a critical magnetic field. The curves obtained are similar to those expected. This figure shows the temperature dependence of the AC susceptibility, denoted h’(T); it has been measured with a technique similar to that used in the Dias article withdrawn from Nature, using the same background subtraction technique (with a cubic polynomial); doubts about this technique were one of the reasons for the withdrawal. These curves are very similar to those that were criticized by Jorge Hirsch (University of California at San Diego), the leader of the trolls according to Dias, in the removed article; I do not understand why the reviewers have not demanded a detailed discussion from the authors. Hirsch published several articles discussing the rapid transition in h’(T) that was observed in the data of the withdrawn article (its origin seemed to be the background subtraction technique). In the new article the transition is not as fast (ΔT ≈ 0.6 K at Tc = 294 K), although the same background subtraction technique is used; still, I think Hirsch’s arguments apply in this case as well and the transition is faster than expected. It will be necessary to be aware of the articles that he publishes in this regard in the coming weeks.
The measurement of the specific heat as a function of temperature allows one to explore the energy of formation of Cooper pairs; The figure shows a characteristic peak of this phenomenon (which is compared in the article with the peak observed for MgB₂) when an alternating current is applied at a certain frequency, which heats the sample. I am not an expert and the result seems convincing to me. However, at the time there were problems repeating the similar results shown in the withdrawn article. We will have to wait for Dias’s competitors, both Eremets and Goncharov, who will surely try to replicate the new results; in the case of the withdrawn article, they complained that Dias did not help them at all (remember that the new article does not provide details on the synthesis technique, which makes replicability very difficult). The synthesis of these materials is often complicated and competitors are often unsuccessful if they are not informed of the details; reviewers for journals such as Nature should require authors to detail the synthesis of new material in great detail. Unfortunately, this is not the case.
In short, an article in Nature that is well worth a Nobel Prize in Physics if its results are independently replicated. Unfortunately, right now there are many doubts about it. But if confirmed, it will be the first of a new family of lutetium hydride-based superconducting materials that promise to drop near ambient pressure (10,000 atmospheres) to ambient pressure (few atmospheres). In your case, this material could have a large number of applications (where intense magnetic fields are required); for example, in magnetic resonance imaging equipment for hospitals. Its impact on 21st century technology promises to be revolutionary. If the results are independently confirmed, of course.
[PS 15 mar 2023] Perhaps it is worth telling more details about what we know about the fund subtraction method. The caption of the figure Extended Data Fig. 15 indicates that a background subtraction based on fitting the polynomial R(T) = Ro + a T² + b T⁵ has been used (although the values of R0, a and b are not specified), to T < 220 K. The CSV file with the original data (archive) is included in the article, so you can easily draw the result in R or Matlab.
This figure shows the original data from the file (left) and the figure that appears in the article (right). The data shows a resistance between 2 and 9 mΩ (milliohms), which adjusted by the quintic polynomial is supposed to give rise to the curves that appear to the right. Subtracting the polynomial (to eliminate “residual resistance” in the observations) shows virtually zero values between 125 and 225 K, and a sharp transition between 225 and 275 K. This is highly suspicious and should have been discussed in detail in the chapter. article (reviewers should have required it). How does the degree of the polynomial used influence? What justifies the choice of the polynomial? In the article it is suggested in passing that it comes from the Ginzburg–Landau theory for superconductivity at zero temperature, but I’m not at all clear where the polynomial comes from. This is all very suspicious. Peer review of this article has been very lax.
Something very similar occurs with the data (XLS file) from which the figure that opens this piece has been drawn (Fig. 2 of the article). The original data for the curve in blue (10 kbar) show noisy values (there are 65534 data in the temperature range between 99.638 and 288.5, with a mean resistance of 7.57 × 10−7standard deviation of 3.51 × 10−7maximum of 3.61 × 10−6and minimum of −6.65 × 10−7). How do you get the blue item figure? Nowhere does it indicate how, but it is obvious that data is missing. For the red curve (16 kbar) the data are those published (65534 data with temperatures between 107.2 and 289.07, with a minimum resistance of −4.79 × 10−7, and maximum of 0.0585 = 58.5 mΩ); the same occurs with the data of the black curve (20 kbar), between 235.33 and 297.68 K with a maximum resistance of 0.0607 = 60.7 mΩ and a minimum of 1.79 × 10−6. Why are there missing data for the blue curve? Why haven’t the reviewers noticed? [/PS]