When it comes to applied quantum mechanics, there are two “holy grails” in the field.
One is building a large scale quantum computer and the other is achieving superconductivity above the freezing point of water, colloquially known as room temperature superconductivity. Superconductors are materials that have no electrical resistance—meaning that electrons can flow through the object unimpeded—but so far physicists have only been able to achieve superconductivity by bringing the materials to incredibly cold temperatures. If superconductivity could be harnessed at room temperature, it would allow for the free transport of energy, wildly faster computers, and incredibly precise sensors. Indeed, it would fundamentally change the world.
In July, Dev Thapa and Anshu Pandey, two well-regarded chemical physicists from the Indian Institute of Science in Bangalore, India, posted a paper to arXiv that claims they managed to achieve “superconductivity at ambient temperature and pressure conditions” using a matrix of gold and silver particles. This announcement understandably shocked the physics community. Not only did Thapa and Pandey claim to have achieved room temperature superconductivity, but they did it using gold and silver, which have never demonstrated superconductivity even at extremely cold temperatures.
Yet as the physics community began to look closer at the data, something didn’t seem right. On Friday, Brian Skinner, a postdoctoral physicist at MIT, posted a comment on Thapa and Pandey’s arXiv paper that noted a strange correlation between two independent measurements.
Image: Twitter/Brian Skinner
Take a look at the green and blue dots in the above graph. They represent the noise measured during two separate experiments run by Thapa and Pandey to test the magnetic susceptibility of their superconducting material. Noise is by definition random, so there shouldn’t be any correlation between the noise measured in one experiment and the noise measured in another experiment. Yet in the graph above, the blue dots are exactly correlated to the green dots, but shifted down a little.
“If you see two measurements, made at different times and under slightly different conditions, and you get the exact same pattern of random variations, that’s very unusual,” Skinner told me in an email. “It’s not clear yet what this repeated noise means. It could be a real and previous unknown natural phenomenon, or it could be an artifact of the measurement process which we also don’t understand. But it’s a sufficiently strange observation that it’s worth paying attention to.”
Over the weekend, Skinner received a reply to his critique from Thapa and Pandey, he said in a tweet. According to Skinner, the authors said that they hadn’t noticed the correlation before, but didn’t back down from their claim that they had observed superconductivity at room temperature. Nevertheless, this remarkable correlation in the noise data had to be explained.
On Saturday, Pratap Raychaudhuri, a physicist at the Tata Institute of Fundamental Research in Mumbai, made a public Facebook post trying to explain how this data could be legit. One possibility raised by Raychaudhuri is that the noise reported by Thapa and Pandey is not noise at all, but is actually a part of the signal being measured that arises from the movement of particles in a magnetic field. As Raychaudhuri explained, it would be possible to reproduce the noise observed in magnetic fields below a certain strength (that is, below 3 Teslas). Below this strength the particles wouldn’t fully detach from one another and would thus retain a “memory” of their initial formation. Thus if the researchers were to apply a magnetic field of a certain strength, turn it off, and then apply it to the same sample of particles again, they’d expect to see the same noise pattern because the particles would retain their initial starting configuration each time.
Yet as Raychaudhuri pointed out, this possible explanation still doesn’t solve all the issues in the paper, which went beyond curious noise correlations.
The most significant anomaly, in this respect, is that the gold and silver nanostructure exhibited a resistive (in other words, electrical) and magnetic transition to a superconductor at the same temperature. According to Raychaudhuri, this would only happen if the test was done on the same sample, which the authors reported was not the case. Raychaudhuri argued that these issues could be easily resolved if the authors would share their test samples with the wider research community, which he said so far they haven’t done.
“The aim of this note is not to defend Thapa and Pandey, but rather merely to present all possible scenarios that the community should think about,” Raychaudhuri wrote in his Facebook post. “For the sake of healthy academic discourse it is of paramount importance that Thapa and Pandey come out openly with their data and their samples.”
When I reached out to Pandey to ask about why he and Thapa aren’t sharing their samples, Pandey told me that they are having their “results validated by independent experts in the respective research fields.”
In the meantime, the superconductivity drama has become even more strange. On Monday, Raychaudhuri received an email that appeared to come from T.V. Ramakrishnan, one of the most prominent physicists in India. According to another public Facebook post made by Raychaudhuri, the email asked him “not to criticize Thapa and Pandey on social media and be patient.” Attached to the email was an email chain purportedly showing Ramakrishnan amicably discussing the research with Pandey. Raychaudhuri responded to Ramakrishnan’s email requesting “not to form opinions based on second hand sources.”
Image: Brian Skinner/Twitter
“What baffles me [is] who would have an interest in crafting a careful and highly credible mail thread and send it through an encrypted email server only to dissuade me from writing on Facebook,” Raychaudhuri wrote on Facebook on Sunday.
Later that day, Skinner received a Facebook request from a profile named Wiles Licher, according to his Twitter post. The profile had zero friends and only one post which reads, “Julius Caesar. The Caesar that did not stop.” Skinner initially thought the account was a troll, but when he looked into it, Licher’s account had been created 16 days before the superconductivity paper was posted to arXiv.
For now, it’s anyone’s guess whether Thapa and Pandey actually managed to achieve superconductivity at room temperature—a discovery that would change the world as we know it.
I’ve reached out to Pandey to ask more about their research and will update this post if I hear back. It’s worth noting, however, that the anomalies in their paper eerily echo those that were at the basis of one of the biggest scandals in modern physics.
In 2001, the prominent German physicist Jan Hendrik Schön published a paper in Nature in which he claimed to have created a molecular-scale transistor. Like Pandey and Thapa’s superconductivity at room temperature, the implications of Schön's paper were huge—it would have marked the end of silicon-based electronics in favor of organic electronics, extending Moore’s law indefinitely and drastically reducing the price of electronic hardware.
When physicists started looking into Schön's data, however, they found that the noise measurements in a number of different experiments were identical. Since noise is random, these measurements should have been different every time. Although Schön initially denied any wrongdoing, further investigation revealed that he had in fact faked the data on several experiments, which led to Schön being stripped of his doctoral degree and dozens of his papers being retracted from leading scientific journals.
A similar fate may await Thapa and Pandey if it turns out their data is fake. On the other hand, it could be an honest mistake or even a major breakthrough in solid state physics. For now, however, the Indian researchers are keeping their samples and data close to their chest while the scientific community awaits some resolution.