In 2020, Ranga Dias was an up-and-coming star of the physics world. A researcher on the College of Rochester in New York, Dias achieved widespread recognition for his declare to have found the primary room-temperature superconductor, a fabric that conducts electrical energy with out resistance at ambient temperatures. Dias printed that discovering in a landmark Nature paper1.Practically two years later, that paper was retracted. However not lengthy after, Dias introduced an excellent greater consequence, additionally printed in Nature: one other room-temperature superconductor2. Not like the earlier materials, the most recent one supposedly labored at comparatively modest pressures, elevating the attractive chance of functions akin to superconducting magnets for medical imaging and highly effective laptop chips.Most superconductors function at extraordinarily low temperatures, under 77 kelvin (−196 °C). So reaching superconductivity at room temperature (about 293 Ok, or 20 °C) could be a “outstanding phenomenon”, says Peter Armitage, a condensed-matter researcher at Johns Hopkins College in Baltimore, Maryland.However Dias is now notorious for the scandal that surrounds his work. Nature has since retracted his second paper2 and lots of different analysis teams have tried and failed to duplicate Dias’s superconductivity outcomes. Some researchers say the debacle has brought about critical hurt. The scandal “has broken careers of younger scientists — both within the area, or considering to enter the sphere”, says Paul Canfield, a physicist at Iowa State College in Ames.
Why a blockbuster superconductivity declare met a wall of scepticism
Earlier reporting by The Wall Avenue Journal, Science and Nature’s information workforce has documented allegations that Dias manipulated information, plagiarized substantial parts of his thesis and tried to impede the investigation of one other paper by fabricating information.Three earlier investigations into Dias’s superconductivity work by the College of Rochester didn’t discover proof of misconduct. However final summer season, the college launched a fourth investigation, led by consultants exterior to the college. In August 2023, Dias was stripped of his college students and laboratories. That fourth investigation is now full and, in response to a college spokesperson, the exterior consultants confirmed that there have been “information reliability issues” in Dias’s papers.Now, Nature’s information workforce reveals new particulars about how the scandal unfolded.The information workforce interviewed a number of of Dias’s former graduate college students, who have been co-authors of his superconductivity analysis. The people requested anonymity as a result of they have been involved concerning the detrimental affect on their careers. Nature’s information workforce verified pupil claims with corroborating paperwork; the place it couldn’t achieve this, the information workforce relied on the truth that a number of, unbiased pupil accounts have been in settlement.The information workforce additionally obtained paperwork related to the acceptance of the 2 Nature papers and their subsequent retractions. (Nature’s information and journal groups are editorially unbiased.)The investigation finds recent particulars about how Dias distorted the proof for room-temperature superconductivity — and signifies that he hid data from his college students, manipulated them and shut them out of key steps within the analysis course of. The investigation additionally reveals, for the primary time, what occurred throughout the peer-review course of for Dias’s second Nature paper on superconductivity. Dias didn’t reply to a number of requests for remark.Collectively, the proof raises questions on why the issues in Dias’s lab didn’t immediate stronger motion, and sooner, by his collaborators, by Nature’s journal workforce and by his college.Zero resistanceDias got here to the College of Rochester in 2017, recent from a postdoctoral fellowship at Harvard College in Cambridge, Massachusetts, the place he labored beneath physicist Isaac Silvera. “He’s not solely a really proficient scientist, however he’s an sincere particular person,” Silvera advised Nature’s information workforce.As soon as Dias settled at Rochester, he pursued high-temperature superconductivity. Three years earlier, the sphere had been electrified when researchers in Germany found superconductivity in a type of hydrogen sulfide with the formulation H3S at 203 Ok (−70 °C) and at extraordinarily excessive pressures3. This was a a lot greater temperature than any superconductor had achieved earlier than, which gave researchers hope that room-temperature superconductivity may very well be across the nook.Dias proposed that including carbon to H3S may result in superconductivity at even greater temperatures.
Ranga Dias on the College of Rochester, New York.Credit score: Lauren Petracca/New York Occasions/Redux/eyevine
His former graduate college students say they synthesized samples of carbon, sulfur and hydrogen (CSH), however didn’t take measurements {of electrical} resistance or magnetic susceptibility that confirmed superconductivity. When a superconducting materials is cooled previous a crucial temperature, its electrical resistance drops sharply to zero, and the fabric shows a equally sharp change in its magnetic properties, known as the Meissner impact. College students say they didn’t observe these key indicators of superconductivity in CSH.Due to this, college students say they have been shocked when Dias despatched them a manuscript on 21 July 2020 asserting the invention of room-temperature superconductivity in CSH. E-mails seen by the information workforce present that the scholars had little time to assessment the manuscript: Dias despatched out a draft at 5.13 p.m. and submitted the paper to Nature at 8.26 p.m. the identical night.When the scholars requested Dias concerning the beautiful new information, they are saying, he advised them he had taken all of the resistance and magnetic-susceptibility information earlier than coming to Rochester. The information workforce obtained e-mails that present Dias had been making comparable claims since 2014. Within the e-mails, Dias says he has noticed a sulfur-based superconductor with a temperature above 120 Ok — which is comparatively excessive, however removed from room temperature. The scholars recall that they felt odd about Dias’s clarification however didn’t suspect misconduct on the time. As comparatively inexperienced graduate college students, they are saying, they trusted their adviser.Throughout peer assessment, nonetheless, Dias’s claims about CSH met extra resistance. Nature’s information workforce obtained the experiences of all three referees who reviewed the manuscript. Two of the referees have been involved over a lack of know-how concerning the chemical construction of CSH. After three rounds of assessment, just one referee supported publication.The information workforce confirmed 5 superconductivity specialists these experiences. They shared among the referees’ issues however say it was not unreasonable for the Nature editors to have accepted the paper, given the strongly constructive report from one referee and what was recognized on the time.The paper was printed on 14 October 2020 to fanfare. Dias and a co-author, Ashkan Salamat, a physicist on the College of Nevada, Las Vegas (UNLV), additionally introduced their new enterprise: Unearthly Supplies, a Rochester-based firm established to develop superconductors that function at ambient temperatures and pressures.On the time, college students say, they trusted Dias’s explanations of the place the resistance and magnetic-susceptibility information got here from. Now, nonetheless, they now not imagine the consequence, or Dias’s clarification for the info. “I don’t assume any of the opposite information was collected,” one pupil says.Issues ariseSoon after the CSH paper was printed, Jorge Hirsch, a condensed-matter theorist on the College of California, San Diego, started urgent Dias to launch the uncooked magnetic-susceptibility information, which weren’t included within the paper. Greater than a yr later, Dias and Salamat lastly made the uncooked information public.In January 2022, Hirsch and Dirk van der Marel, a retired professor on the College of Geneva in Switzerland, posted an evaluation of the uncooked information on the preprint server arXiv4. They reported that the info factors have been separated by suspiciously common intervals — every precisely a a number of of 0.16555 nanovolts. Hirsch and van der Marel said that this function was proof of information manipulation.
Dias’s workforce used laser spectroscopy to measure the strain of samples in diamond anvil cells.Credit score: Lauren Petracca/New York Occasions/Redux/eyevine
Dias and Salamat responded in an arXiv preprint, arguing that the voltage intervals have been merely a results of a background subtraction5 (the preprint was subsequently withdrawn by arXiv directors). In high-pressure experiments, the sign of a pattern’s superconductivity — a drop in voltage — might be drowned out by background noise. Researchers typically subtract this background, however the CSH paper didn’t point out the approach.Questions concerning the information prompted Nature’s journal workforce to look additional. In response to the issues from Hirsch and van der Marel, editors at Nature requested 4 new referees to take part in a post-publication assessment of the CSH paper, which, like most peer assessment, was confidential.Now, Nature’s information workforce has obtained the experiences, which present that two of the nameless referees discovered no proof of misconduct. However two different reviewers, whom the information workforce can establish as physicists Brad Ramshaw at Cornell College in Ithaca, New York, and James Hamlin on the College of Florida in Gainesville, discovered critical issues with the paper.Particularly, Hamlin discovered proof that led him to conclude the uncooked information had been altered. Nature utilized an editor’s observe to the CSH paper on 15 February 2022, alerting readers to issues concerning the information.On 4 March 2022, Dias and Salamat despatched a rebuttal to the referees, denying information manipulation. However the rebuttal, seen by the information workforce, doesn’t present an evidence for the problems that Hamlin and Ramshaw discovered within the uncooked magnetic-susceptibility information. “I don’t know of any cheap approach this might come about,” Ramshaw wrote in a 13 March e-mail to Nature’s manuscript workforce in response to the rebuttal. “The only conclusion could be that these information units are all generated by hand and never really measured.”On 27 March 2022, Hamlin despatched Nature’s journal workforce his response to the rebuttal, which proposed an evidence for the odd information: slightly than deriving the printed information from uncooked information, Dias had added noise to the printed information to generate a set of ‘uncooked’ information.To evaluate the proof for information fabrication, Nature’s information workforce final month requested two superconductivity specialists to assessment the post-publication experiences. They stated that Hamlin’s evaluation provides credence to claims of misconduct.In July 2022, utilizing a distinct evaluation, van der Marel and Hirsch independently got here to the identical conclusion and posted their findings on arXiv as an replace to their unique preprint. In it, they state that the uncooked information should have been constructed from the printed data6.
Why superconductor analysis is in a ‘golden age’ — regardless of controversy
In gentle of those issues, Nature began the method of retracting the CSH paper. On 11 August, Nature editors despatched an e-mail to all of the co-authors asking them whether or not they agreed to the retraction. College students who spoke to the information workforce say that they have been stunned by this, as a result of Dias had saved them out of the loop concerning the post-publication assessment course of. They remained unaware of any of the referees’ findings, together with that there was proof for information fabrication.Nature retracted the CSH paper on 26 September 2022, with a discover that states “points undermine confidence within the printed magnetic susceptibility information as an entire, and we’re accordingly retracting the paper”. Karl Ziemelis, Nature’s chief utilized and bodily sciences editor, says the journal’s investigation ceased as quickly because the editors misplaced confidence within the paper, which “did depart different technical issues unresolved”.The retraction doesn’t state what Hamlin and Ramshaw discovered within the post-publication assessment course of instigated by Nature: that the uncooked information have been most likely fabricated. Felicitas Heβelmann, a specialist in retractions on the Humboldt College of Berlin, says misconduct is troublesome to show, so journals typically keep away from laying blame on authors in retractions. “Lots of retractions use very obscure language,” she says.Publicly, Dias continued to insist that CSH was respectable and that the retraction was merely right down to an obscure technical disagreement.As Nature journal editors have been investigating the CSH paper, the College of Rochester performed two investigations into Dias’s work; a separate one adopted the retraction. One of many college’s inquiries was in response to an nameless report, which included among the proof indicating attainable information fabrication that surfaced throughout Nature’s post-publication assessment.The college advised Nature’s information workforce that the three investigations concerning the CSH research didn’t discover proof of misconduct.A spokesperson for Nature says that the journal took the college’s conclusions into consideration throughout its deliberations, however nonetheless determined to retract the paper.The shortage of industry-wide requirements for investigating misconduct leaves it unclear whether or not the accountability to research lands extra on journals or on establishments. Ziemelis says: “Allegations of attainable misconduct are exterior the remit of peer assessment and extra appropriately investigated by the host establishment.”Heβelmann says the accountability to research can “range from case to case”, however that there’s a pattern of extra journals investigating misconduct, no matter institutional motion.Funding businesses may examine alleged misconduct. On this case, Dias has obtained funding from each the US Nationwide Science Basis (NSF) and the Division of Vitality (DoE). The DoE didn’t reply to questions from Nature’s information workforce about Dias’s grant. The NSF declined to say whether or not it’s investigating Dias, but it surely famous that awards might be terminated and suspended in response to an investigation.The scholars who spoke to Nature’s information workforce say that none of them have been interviewed within the three investigations of the CSH work by the college, which they weren’t conscious of on the time. “We have been hoping somebody would come discuss to us,” one pupil says. “It by no means occurred.”A brand new claimBy the time the CSH paper got here beneath scrutiny by Nature journal editors in early 2022, Dias’s graduate college students have been beginning to develop involved. In summer season 2021, Dias had tasked them with investigating a compound of lutetium and hydrogen (LuH), which he thought is perhaps a high-temperature superconductor.They started testing commercially bought samples of LuH and, earlier than lengthy, a pupil measured the resistance dropping to zero at a temperature of round 300 Ok (27 °C). Dias concluded the fabric was a room-temperature superconductor, regardless that there was extraordinarily little proof, a number of college students advised Nature. “Ranga was satisfied,” one pupil says.
Physicist James Hamlin raised issues about information reported by the Rochester group.Credit score: Zach Stovall for Nature
However the measurements have been suffering from systematic errors, which college students say they shared with Dias. “I used to be very, very involved that one of many probes touching the pattern was damaged,” one pupil says. “We may very well be measuring one thing that appears like a superconducting drop, however be fooling ourselves.” Though college students did see resistance drops in a couple of different samples, there was no consistency throughout samples, and even for repeated measurements of a single pattern, they advised Nature’s information workforce.College students have been additionally anxious concerning the accuracy of different measurements. Throughout elemental evaluation of a pattern, they detected hint quantities of nitrogen. Dias concluded that the samples included the factor — and the ensuing paper refers to nitrogen-doped lutetium hydride. However additional evaluation, carried out after the paper was submitted, indicated that nitrogen was not included into the LuH. “Ranga ignored what I used to be saying,” one pupil says.As a result of they weren’t consulted on the CSH paper, the scholars say they wished to ensure they have been included within the technique of writing the LuH paper. Based on the scholars, Dias initially agreed to contain them. “Then, at some point, he sends us an e-mail and says, ‘Right here’s the paper. I’m gonna submit it,’” one pupil says.E-mails seen by Nature’s information workforce corroborate the timeline. Dias despatched out the primary draft of the LuH paper in an e-mail at 2.09 a.m. on 25 April 2022. “Please ship me your feedback by 10.30 AM,” Dias wrote. “I’m submitting it immediately.” The manuscript they obtained didn’t comprise any figures, making it troublesome to evaluate. The scholars satisfied Dias to carry off on submitting till the subsequent day, after they may focus on it in particular person.One pupil was upset sufficient by the assembly that they wrote a memorandum of the occasions 4 days afterwards. The memo provides particulars of how college students raised issues and Dias dismissed them. College students anxious that the draft was deceptive, as a result of it included an outline of methods to synthesize LuH; in actuality, all of the measurements have been taken on commercially purchased samples of LuH. “Ranga responded by stating that it was by no means explicitly talked about that we synthesized the pattern so technically he was not mendacity,” the scholar wrote.The scholars say in addition they raised issues concerning the strain information reported within the draft. “None of these strain factors correspond to something that we really measured,” one pupil says. Based on the memo, Dias dismissed their issues by saying: “Stress is a joke.”College students say that Dias gave them an ultimatum: take away their names, or let him ship the draft. Regardless of their worries, the scholars say that they had no alternative however to acquiesce. “I simply keep in mind being very intimidated,” one pupil says. The scholar says they remorse not talking up extra to Dias. “However it’s scary on the time. What if I do and he makes the remainder of my life depressing?”Dias made some modifications that the scholars requested, however ignored others; the submitted manuscript contained an outline of a synthesis process that had not been used. He despatched the LuH manuscript to Nature that night.Paper problemsAfter Nature printed the LuH paper in March 2023, many scientists have been crucial of the journal’s determination, given the rumours of misconduct surrounding the retracted CSH paper. They wished to know on what foundation Nature had determined to simply accept it. (Within the case of each papers, neither the peer-review experiences nor the referees’ identities have been revealed.) Nature’s information workforce obtained these opinions and might, for the primary time, reveal what occurred throughout the assessment course of for the LuH paper. Nature editors obtained the manuscript in April 2022 (a few month after Nature obtained the CSH post-publication assessment experiences) and despatched it out to 4 referees.
Physicist Brad Ramshaw, along with James Hamlin, investigated information questions surrounding Dias’s superconductivity analysis.Credit score: Kim Modic
All 4 referees agreed that the findings, if true, have been extremely important. However they emphasised warning in accepting the manuscript, due to the extraordinary nature of the claims. Referee 4 wrote that the journal needs to be cautious with such extraordinary claims to keep away from one other “Schön affair”, referring to the in depth information fabrication by German physicist Jan Hendrik Schön, which has turn out to be a cautionary story in physics and led to dozens of papers being retracted, seven of them in Nature. Referees 2 and three additionally expressed concern concerning the outcomes due to the CSH paper, which on the time bore an editor’s observe of concern however had not but been retracted. Referees raised a plethora of points, from an absence of particulars concerning the synthesis process to unexplainable options within the information.Though Dias and Salamat managed to assuage a few of these issues, referees stated the authors’ responses have been “not passable” and the manuscript went via 5 phases of assessment. Ultimately, just one referee stated there was stable proof of superconductivity, and one other gave certified assist for publication. The opposite two referees didn’t voice assist for publication, and certainly one of them remained unhappy with the authors’ responses and wished extra measurements taken.The information workforce requested 5 superconductivity specialists to assessment key data obtainable to Nature journal editors after they have been contemplating the LuH manuscript: the referee experiences for the LuH paper and the experiences indicating information fabrication within the CSH paper. All 5 stated the paperwork raised critical questions concerning the validity of the LuH outcomes and the integrity of the info.“The second paper — from my understanding of timelines — was being thought-about after the Nature editors and numerous the condensed-matter group have been conscious there have been profound issues” with the CSH paper, Canfield says. The specialists additionally pointed to detrimental feedback from among the LuH referees, such because the remark by Referee 1 that “uncooked information doesn’t appear like a function equivalent to superconducting transition”.When requested why Nature thought-about Dias’s LuH paper after being warned of potential misconduct on the earlier paper, Magdalena Skipper, Nature’s editor-in-chief, stated: “Our editorial coverage considers each submission in its personal proper.” The rationale, Skipper explains, is that choices needs to be made on the premise of the scientific high quality, not who the authors are.Many different journals have comparable insurance policies, and pointers from the Committee on Publication Ethics state that peer reviewers ought to “not enable their opinions to be influenced by the origins of a manuscript”. However not all journals say they deal with submissions independently. Van der Marel, who’s the editor-in-chief of Physica C, says that he would take into account previous allegations of misconduct if he have been assessing a brand new paper by the identical creator. “In case you have good causes to doubt the credibility of authors, you aren’t obliged to publish,” he says.Beneath reviewSoon after the LuH paper was printed in March 2023, it got here beneath additional scrutiny. A number of groups of researchers independently tried to duplicate the outcomes. One group, utilizing samples from Dias’s lab, reported electrical resistance measurements that it stated indicated high-temperature superconductivity7. However quite a few different replication makes an attempt discovered no proof of room-temperature superconductivity within the compound.As beforehand reported in Science, Hamlin and Ramshaw despatched Nature a proper letter of concern in Might. Dias and Salamat responded to the problems later that month, however the college students say they weren’t included within the response, and learnt concerning the issues a lot later.A recording of a 6 July 2023 assembly between Dias and his college students, obtained by Nature’s information workforce, exhibits that Dias continued to control the scholars. All through the hour-long assembly, Dias stated he wished to contain the scholars in deciding how the workforce would reply to issues concerning the LuH paper. However he didn’t inform them that he and Salamat had already responded to the technical points raised by Hamlin and Ramshaw.
Considered one of Dias’s college students adjusts a diamond anvil cell, which the workforce utilized in its experiments.Credit score: Lauren Petracca/New York Occasions/Redux/eyevine
The recording additionally reveals how Dias tried to control the Nature assessment, as a result of he believed the method would flip in opposition to him as soon as extra. “We will fake we’re going to cooperate and purchase time for a month or so, after which collect some senior scientists from the group,” Dias says within the recording. Dias explains how he desires to make use of the credibility of senior scientists — or the College of Rochester — to strain Nature and avert a retraction.However Dias’s plans have been thwarted. Later that month, the scholars obtained an e-mail from Nature’s editors that confirmed Dias and Salamat had, in truth, already responded to the issues. The scholars realized that Dias had despatched them a doc with the dates eliminated, apparently to perpetuate the falsehood.On 25 July 2023, the journal initiated a post-publication assessment and requested 4 new referees to evaluate the dispute. The entire referees agreed that there have been critical issues with the info, and that Dias and Salamat didn’t “convincingly deal with” the problems raised by Hamlin and Ramshaw. A spokesperson for Nature says the journal communicated with College of Rochester representatives throughout the post-publication assessment.Individually, Dias’s college students have been starting to mobilize, re-examining the LuH information they have been in a position to entry. The scholars hadn’t finished this earlier than, as a result of, they are saying, Dias produced virtually all the figures and plots in each of the Nature papers.A number of different researchers advised the information workforce that the principal investigator doesn’t usually produce all of the plots. “That’s bizarre,” Canfield says.The scholars say they have been particularly involved concerning the magnetic susceptibility measurements — once more, the uncooked information appeared to have been altered. the actual uncooked information, one pupil says, the fabric doesn’t appear like a superconductor. However when Dias subtracted the background, the scholar says, that “principally flips that curve the other way up and makes it look superconducting as a substitute”.They continued discovering issues. For the resistance measurements, too, the alleged uncooked information didn’t match information really taken within the lab. As an alternative, it had been tweaked to look neater. “Science might be actually messy … a few of these plots simply look too good,” a pupil says.Again to schoolBy this level, some college students have been deeply involved about their careers. “My thesis goes to be stuffed with fabricated information. How am I alleged to graduate on this lab?” one pupil says. “At that time, I used to be considering of both taking a depart of absence, or of dropping out.”Throughout the summer season, Dias started going through different points. Considered one of his papers in Bodily Overview Letters8 — unrelated to room-temperature superconductivity — was being retracted after the journal discovered convincing proof of information fabrication. Across the identical time, Dias was stripped of his college students and the College of Rochester launched a fourth investigation — this time, the scholars say they have been interviewed.
‘A really disturbing image’: one other retraction imminent for controversial physicist
In late August, the scholars determined to request a retraction of the LuH paper and compiled their issues concerning the information and Dias’s behaviour. Earlier than they despatched a letter to Nature, Dias apparently caught wind of it and despatched the scholars a cease-and-desist discover, which the information workforce has seen. However, after consulting a college official who gave them the inexperienced gentle, the scholars despatched their letter to Nature editors, precipitating the retraction course of. Eight out of 11 authors, together with Salamat, signed the letter and the LuH paper was retracted two months later, on 7 November.Based on a number of sources acquainted with the corporate, Salamat left Unearthly Supplies in 2023 and is beneath investigation at UNLV. He didn’t reply to a number of requests for remark, and a spokesperson for UNLV declined to remark publicly on personnel points.The scandal has additionally had an affect on Nature’s journal workforce. “This has been a deeply irritating scenario, and we perceive the power of emotions this has stirred inside the group,” Ziemelis says. “We’re taking a look at this case rigorously to see what classes might be learnt for the longer term.”With the college’s investigation now full, Dias stays at Rochester whereas a separate course of for addressing “personnel actions” proceeds. He has no college students, shouldn’t be instructing any lessons and has misplaced entry to his lab, in response to a number of sources. Dias’s prestigious NSF grant — which has US$333,283 left to pay out till 2026 — is also in jeopardy if the NSF finds cause to terminate it.Dias has not printed any extra papers about LuH, however on X (previously Twitter), he sometimes posts updates concerning the materials. In a 19 January tweet, Dias shared a picture of information, which he stated confirmed the Meissner impact — “definitive proof of superconductivity!”