Zephyrnet Logo

Science will suffer if we fail to preserve academic integrity – Physics World

Date:

Following allegations of scientific misconduct in high-temperature superconductivity, Caitlin Duffy says that if trust in science is to be maintained, researchers must be able to report their concerns

illustration of crumpled paper in the shapes of human heads. One is read and has a whistle in its mouth
Reporting lines How can scientists best report allegations of academic misconduct to the community? (Courtesy: Shutterstock/Lightspring)

High-temperature superconductivity hit the headlines in spectacular fashion this year following a series of results that supposedly reached a “holy grail” of condensed-matter research. We had a “summer of speculation” over the material LK-99 in which a team of researchers in South Korea claimed that a modified form of lead apatite could conduct electricity without resistance at ambient pressure and temperatures. In the end it took other researchers barely a few weeks to establish that LK-99 was not a room-temperature superconductor at ambient pressure after all. But that still leaves the controversial claims made a few months earlier by Ranga Dias at the University of Rochester and colleagues of near-ambient superconductivity in nitrogen-doped lutetium hydride unresolved.

The saga began in October 2020 when Dias and his team published a paper in Nature (586 373), in which they claimed to have discovered superconductivity at a balmy 15 °C in a hydrogen sulphide material under high pressure. Concerns were, however, soon raised over that finding, which led to rebuttals and counter rebuttals in peer-reviewed publications and on the arXiv preprint server that sometimes verged on the unprofessional. There have even been suggestions that “cease and desist” letters were sent by lawyers to arXiv and to rival scientists. Following concerns with the hydrogen sulphide paper, it was retracted in September 2022 by editors at Nature (though Dias and colleagues maintain that they stand by their work).

The fate of the recent lutetium hydride paper, also published in Nature (615 244), is unknown. Strangely, the paper has an unusual ethics declaration that revokes open access to the information required to reproduce the authors’ exact steps. Although the raw data files have been made publicly available from Dias, arXiv has been inundated with reports from researchers who have failed to reproduce the room-temperature superconductivity with their own samples. Some months later, however, a team led by Russell Hemley from the University of Illinois claimed to have observed evidence for near ambient superconductivity in a sample provided by Dias.

Complex situation

Criticism and scepticism are, of course, vital for the progression of science. But to have scientists publicly express their distrust in other scientists harms the very trust we need from the public to fund scientific work. Such criticism also distracts researchers from making legitimate discoveries as they spend their time replicating or analysing what could be bogus results. As an early-career physicist myself, navigating this minefield is complex. So, how can we prove negligence from people in positions of influence who use their status to continue to attract funding for their university or institute? Where do we go if we have concerns about academic integrity? How can we report our concerns to the community when it takes months to pass through peer review?

I recently posed these issues to physicists at the final session of the Gordon Conference on Superconductivity, which was held in Les Diablerets in Switzerland from 30 April to 5 May. My questions, however, were met with what seemed like a wall of stunned silence. Hands eventually rose and people pointed out that there is no good scientific platform to challenge research that one has question marks about. Some researchers have had conversations on Reddit or PubPeer – a website allowing for anonymous comments on scientific papers – but this may not be considered an avenue for serious scientific discourse. Nature has a “comments” section below the HTML version of an article, but comments are seemingly left unaddressed by editors.

A precarious option is to file a formal case of academic misconduct with journal editors or with the institutes where the researchers are based. But this carries enormous risks

As for arXiv, it is a vital source of information, providing rapid publication of non-peer-reviewed preprints. But its focus is simply on providing a platform for new and original research. While comments are not uncommon, the arXiv website is not geared up for open debate or for people to raise concerns – and those who do could end up being barred by the site’s moderators. Of course, it is possible to raise concerns in a formal paper submitted to a peer-reviewed journal but that can be tricky. Peer review can take months (whereas concerns require immediate action) while finding the correct journal to host a rebuttal is not simple. The original journal where contentious work appeared may not want to court controversy, while a rival journal will simply not want to get involved in another publication’s affairs.

An even more precarious option is to file a formal case of academic misconduct with journal editors or with the institutes where the researchers are based. But this carries enormous risks. For a junior researcher still making their name in science, it could cost them their career. For a senior researcher, it could damage a well-established reputation. Journals will, of course, defend their decision to publish highly citable articles while research institutes will want to defend their employees. Turning a blind eye merely lets perpetrators carry on getting funded, protected by their employers.

These issues must be addressed. Editors of journals need to check the credibility of authors before publishing a high-profile result that may get attention from the world’s media. Research institutes need to undertake an unbiased, open and thorough investigation following allegations of misconduct. Scientists, meanwhile, have to keep discussions professional while reporting suspected foul play and not fear repercussions should they do so.

The current situation in hydride superconductivity has no simple solution. It is made more complex given that Dias has submitted a patent on a lutetium hydride material that is apparently a room-temperature superconductor at ambient pressure. Yet we must find something that works if trust in science is to be maintained. Although science can be fraught with tension and competition, we are a single community and we must all collaborate to preserve our standards of research and academic integrity.

spot_img

Latest Intelligence

spot_img