Calling Baker Street! IISc duo’s ‘power breakthrough’ has kicked up a Sherlockian mystery

IISc campus, Bengaluru | IISc official website

If proved true, the discovery of the researchers could well be one of the biggest moments of modern science.

Bengaluru: A research paper announcing a breakthrough that has eluded physicists for over a century. A renowned scientist’s impostor, and a mysterious email seeking to silence criticism of the research. Suspicious Facebook friend requests to physicists from purportedly the same ID.

The physics community possibly finds itself in the grips of a controversy centred on a finding that may well be one of the big scientific breakthroughs of the century: Superconductivity at room temperature.

In layperson’s terms, the innovation entails the seamless transmission of electricity from source to destination without any loss of current on the way, a feat currently pulled off by cooling conductors below a certain limit known as the material’s critical temperature in a fairly expensive process.

The use of such materials dramatically improves electricity generation and consumption. Currently, super-cooled superconductors are used in medicine, for example for MRI machines. If and when the costs come down, they are expected to be used everywhere – from magnetic-levitation trains (like Hyperloop), power grids, telecommunication, to household appliances like refrigerators. Simply speaking, a room-temperature superconductor will revolutionise our lives.

So, it created a big buzz in the physics community when two physicists at the Indian Institute of Science (IISc) claimed they had pulled off the feat. Anshu Pandey and his doctoral student Dev Kumar Thapa posted their paper on the website arXiv, a repository for academic research where any scientist can post the results of their experiments for public perusal.

The duo is part of IISc’s Solid State and Structural Chemistry Department. Titled ‘Evidence for Superconductivity at Ambient Temperature and Pressure in Nanostructures’, the paper was a preprint, i.e., released for the benefit of fellow scientists and public before it is peer-reviewed and published in a journal.

Controversy

The paper, however, caught the attention of Brian Skinner, a physicist at the Massachusetts Institute of Technology (MIT), who posted about the paper, its implications, and an alleged anomaly in a still-developing Twitter thread.

1/ Who wants to hear some scientific intrigue?

A few weeks ago, a group of physical chemists posted a paper online announcing the observation of superconductivity at room temperature.

Today I posted a comment pointing out something funny in their data.https://t.co/Uw1wk0vYXW

— Brian Skinner (@gravity_levity) August 10, 2018

“Looking through the paper one evening, I got curious as to why one of their measurements showed lots of random noise at low temperature, but very little noise at high temperature,” he wrote.

When he plotted the noise in the data, he obtained this graph.

Noise is unpredictable. There’s no telling what causes what kind of disturbances, so two different tests at different temperatures and other differing parameters usually cannot produce identical noise output. The chances of that are nearly zero.

However, in the plot above, Skinner noticed the noise outputs at the bottom of the graph, one in green and the other in blue, are almost identical. If the entire green section is moved slightly lower, it would overlay nearly perfectly on the blue pattern.

This finding seems alarming. The only way there could be two identical noise outputs is if there were two identical inputs — all parameters the same. But it is nearly impossible to achieve that level of duplicity in real life.

The immediate conclusion seems to be that either the authors or Skinner made an error, or the data was fabricated.

Skinner’s thread and a comment of his on the paper have on the internet for over five days now, and physicists the world over have subsequently raised questions and requested access to Pandey and Thapa’s data sets. But the duo has not responded so far.

Impersonation

Meanwhile, Skinner exposed a new angle to the affair to the dozens following his Twitter thread for updates.

One of India’s leading theoretical physicists T.V. Ramakrishnan, an expert in superconductivity who is now a professor at Benaras Hindu University, has said it is likely the findings are legitimate.

Meanwhile, yet another leading voice in the field, Pratap Raychaudhuri of the Tata Institute of Fundamental Research (TIFR- Mumbai), took to Facebook to air his concerns about the paper, and asked the authors to respond to fabrication allegations.

He subsequently received an email asking him not to criticise the team, with the sender identifying himself as Ramakrishnan, his senior in the field. Finding the email address ([e-mail]) unusual, Raychaudhuri marked a CC of the mail to Ramakrishnan’s personal ID.

Raychaudhuri soon received a call from Ramakrishnan, who said he never sent the email. It emerged later that the email address in question was also used to correspond with the authors of the paper and several other physicists.

“I was very surprised like everyone else who came to know of the incident,” said Raychaudhuri to ThePrint. “I have not encountered this kind of an incident before.”

The email address has been derived from ProtonMail, the world’s largest encrypted email service, built by CERN and MIT. It does not even log a sender’s IP. As a result, it is next to impossible to trace who sent these mails. Furthermore, as it is based out of Switzerland, it is out of the purview of cyber police in Mumbai, where Raychaudhuri lives.

Adding to the mystery are Facebook friend requests sent to Skinner and Raychaudhuri from a now-defunct account by the name of ‘Wiles Licher’ that was created just before the paper was submitted to arXiv. When Skinner accepted the friend request, he found himself to be the only friend on the list of this profile.

The physicists are also trying to figure out what exactly the pseudonym ‘Wiles Licher’ stands for, and whether it’s an anagram or a code.

Silence in academia

A lot of the issues unfolding could be slotted into their appropriate categories of suspicion if Thapa and Pandey, who have already filed for a patent, responded to the allegations.

However, it’s common for researchers to be reluctant to share data, as a lot of research is seen by scientists as proprietary and researchers live with the constant fear of appropriation and wrongful use of their data. It is a common practice among researchers to wait until a paper is published in a peer-reviewed journal before its authors address concerns or release data sets.

In fact, scientists have even lost out on the Nobel Prize because they shared their data with someone who used it without credit.

In short, the two are under no obligation to respond.

Speaking to ThePrint, Pandey said: “We are having our results validated by independent experts in the respective research fields. This process takes time. Without validation, the synthesis and device fabrication details are speculative and will add to further confusion.”

“We will announce the results of validation in the appropriate forum as soon as possible,” he added.

Signal vs noise

Another question that needs answering is if the green and blue parts of the graph were random noise or actual duplicated noise from an unanticipated source. This leads into a grey area between signal and noise: How do we distinguish between the two?

Normally, signals are the data we look for in an experiment, and noise is the data that isn’t supposed to be produced by our setup, and is thus to be disregarded.

Suppose we are attempting to record audio during a concert. The music we’re recording is the signal. Any screeches produced by microphones is noise. But there are also the background sounds: The audience chatter. All of them show up in the final recording, and it is up to us to look only for the music and disregard the others.

But noise can also be replicated. Suppose the microphone screeches only at certain frequencies of music: Every time, the music hits those notes, the noises come back. A lot of the times, they can be quite similar, almost identical, when plotted.

The phenomenon is not all that uncommon in scientific experiments, when an ambient factor or the experimental setup causes noise. Periodic noise can often look like a signal and be confusing. It is possible that the blue and green noise we see is periodic and thus repeating.

Some physicists also differ in opinions about the sensitivity of the noise itself. Since numbers in these experiments run to precision of several decimal points, it is likely that when they are rounded off to just two or three, the numbers are repeated.

Thus there is a good chance that the data points for 0.1T & 1T look very much alike.

— Vikram Bakaraju (@Vikramovic) August 14, 2018

Scientists are also fully allowed to “massage” numbers to remove data they think is noise, so that the results fit well with expected signals.

All of these questions will be answered when Thapa and Pandey share their data.

Implications

The story can play out in several different ways. It is possible that the team made an error in their original findings, and they rescind their paper.

It may also turn out that the duo engaged in scientific misconduct and fabricated data, an accusation that everyone is cautious to make until further information is known. Such a scenario might lead to stripping away of titles, degrees, and privileges.

As Skinner pointed out in his thread, this is what happened nearly two decades ago when German physicist Jan Hendrik Schön won multiple awards after he seemingly proved that materials incapable of displaying superconductivity could. His results could not be replicated by anyone else, but his seemingly solid proof made him an overnight celebrity in the physicist circle.

Until, that is, someone noticed a discrepancy. Bell Labs, where he worked then, launched a quick investigation and his data was shown to be fraudulent. Not only were Schön’s awards rescinded, his doctoral degree was revoked and he was barred from peer-reviewing papers for eight years.

Even if this happens with the Pandey-Thapa paper, however, it is safe to say that the IISc, or even Indian science itself, will not suffer.

But if everything can be explained and the results replicated, two Bengaluru scientists will go down in history books for one of the biggest path-breaking findings in physics.

This report has been updated with a new comment by one of the research authors.