Science Majorana paper earns an expression of concern

Just months after Nature retracted a paper on the “Majorana” particle because other researchers found issues in the work, Science has placed an expression of concern on a different paper that suggested “a relatively easy route to creating and controlling [Majorana zero modes] MZMs in hybrid materials.”

If such particles exist, they could allow Microsoft — which employs some of the researchers involved in the work — to build a quantum computer. But scientists have suggested that the findings of various studies do not suggest the presence of Majorana particles.

The Science paper has been cited 29 times since it was published in 2020, according to Clarivate Analytics’ Web of Science. The EoC reads:

On 27 March, 2020, Science published the Research Article “Flux-induced topological superconductivity in full-shell nanowires” by S. Vaitiekėnas et al. (1). Pursuant to a reader request, the authors released additional data (archived at Zenodo http://doi.org/10.5281/zenodo.4263106) taken in association with the project that led to their paper. Following the release of the additional data, we received a reader concern that the tunneling spectroscopy data published in the original paper are not representative of the entirety of the data released in association with this project. While we await the outcome of a full investigation commenced by the authors’ academic institution (Niels Bohr Institute, University of Copenhagen), we are alerting our readers to this concern.

The authors, at the Center for Quantum Devices and Microsoft Quantum Lab–Copenhagen at the Niels Bohr Institute at the University of Copenhagen, Microsoft Quantum, and Yale, told Retraction Watch that they do not agree with the expression of concern:

We stand behind our paper completely, every graph, data file, interpretation, and conclusion. We refuted all of the concerns that were raised and made all the data that Science’s editor requested available.  

Science editors asked the Niels Bohr Institute to carry out an  independent examination of the paper.  A committee formed by the NBI chair conducted their evaluation and informed Science editors and the authors that the  concerns raised were unfounded, and that  there was no evidence of impropriety in the paper or omission of relevant data. The assessment was  led by Jan Thomsen, Chair of the Niels Bohr  Institute. 

The authors disagree with Science’s decision to publish their EEoC after they knew that the inquiry they requested found no problems with the paper or the handling of data. 

The authors said there is 

no connection between our paper and the retracted 2018 Nature paper. There are no authors in common, and the  experiments were done in different labs.  

The paper is completely valid, supported point by point by all authors, and fully endorsed by the inquiry that Science  requested.

Sergey Frolov and Vincent Mourik, the researchers who flagged potential issues in both papers, tell Retraction Watch:

We are pleased with the Editorial Expression of Concern issued by Science and find it accurate. The core problem that we found in this paper is non-representative data selection, which is the same core problem as in the now retracted Nature paper on Quantized Majorana Conductance from Delft. Other circumstances are different.

In a July 26 letter to Science obtained by Retraction Watch, Jan W. Thomsen, professor and head of department at the Bohr Institute, writes:

In short, our conclusions read as follows:

(a) We find no problems with the paper, nor with the conclusions in the paper, nor with the data supporting the claims of the paper.

(b) We find the complaints of Mr Frolov and Mr Mourik unjustified.

(c) All data connected with the present paper has – according to demand – been transmitted rightfully to third parties. No additional data is left out.

Frolov and Mourik tell Retraction Watch:

We do not believe that the data currently on Zenodo are full data from those experiments. For the electronic transport part, about 50 datasets were shared from around 10 devices, so on average 5 executions of a measurement script per device. In comparison, the IST Austria data from reproduction experiments on the same nanowires were over 500 data sets, so an order of magnitude larger. The IST data were shared promptly, were self-explanatory and did not raise further questions. We insist that the authors and Niels Bohr Institute share full data from these experiments with us, as required of them by the Danish Code of Conduct for Research Integrity.

And Science editor in chief Holden Thorp said yesterday that the letter “describes an internal investigation that we don’t believe is adequate to resolve the questions on the paper.” He told Retraction Watch:

We had a meeting with Dr. Thomsen 10 days ago where he promised to do a thorough, external investigation similar to that done by the Delft administration on the Nature paper that was retracted.  We agree with him that this is required to resolve this.

He added: 

We await a thorough, independent investigation and hope that Dr. Thomsen will provide that soon.

Frolov and Mourik write:

We have not been contacted by the Niels Bohr Institute or the University of Copenhagen to provide testimony to their investigation. We have not seen the committee report and thus do not know what their conclusion is based upon.

Like Retraction Watch? You can make a one-time tax-deductible contribution or a monthly tax-deductible donation to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].