Starting in 2020, possible irregularities were identified in a couple of Majorana-publications from the group of professor Leo Kouwenhoven. It concerns publications from before 2019. Whenever such irregularities are found, it should be investigated how it came to that situation. Between 2020 and 2023, two separate investigations into possible violation(s) of scientific integrity were conducted by the CWI (and in one case also by the LOWI).

The first CWI investigation & LOWI

The first CWI investigation started in April 2020, following an ‘expression of concern’ about the 2018 Quantized Majorana conductance publication. TU Delft and QuTech reported this. The expression of concern arose after two former members from the Kouwenhoven group informed the authors that they had found inconsistencies between the raw measurement data that was made available to them, and the figures that were published in the paper.

On March 8, 2021, the public part of the expert report was disclosed and TU Delft and QuTech reported on this. The expert report was prepared by an external committee of international experts at the request of the CWI. The experts clearly stated in their report that they ‘found no evidence of fabrication: all data in the publication seem to be genuine results of measurements’. Instead, the experts state that the way the results were presented in the manuscript shows that the authors selected data that supported the phenomenon they were looking for, while omitting data that could have raised questions about the conclusions presented. They found no evidence that this may have been intentional.

With the expert report public, the authors retracted the article (also on March 8, 2021), and Nature published a Retraction Note. The authors also published a repository where they clearly displayed the problems with the paper, along with the raw data behind the publication. They also published a rewrite of the publication.

On March 26, 2021, the CWI issued an advice to the TU Delft Executive Board (CvB). Based on that advice, the CvB formulated an initial judgement. Two accused parties then requested the LOWI to issue an advice on the initial judgment of the CvB. The LOWI issued its advice on April 25, 2022.

On June 8, 2022, the CvB announced its judgment based on the recommendations of the CWI and the LOWI. The judgment was as follows: In the 2018 Nature article Quantized Majorana conductance, the first author involved, Dr Hao Zhang, and corresponding author, Leo Kouwenhoven, were partly negligent and partly culpably negligent. There was no violation of scientific integrity. TU Delft and QuTech report on this.

The anonymized case report of the first integrity investigation is published on the UNL website.

The second CWI investigation

The second CWI investigation was announced on February 8, 2022, following a new internal report regarding possible irregularities in a few publications on Majorana research. TU Delft and QuTech reported this.

The CWI started its investigation into the alleged inappropriate handling of data in relation to the publication Epitaxy of advanced nanowire quantum devices (Nature, 2017), on which QuTech researchers are co-authors. Importantly, the authors proactively published a repository where they clearly displayed several problems with the paper already in March 2021, along with the raw data behind the publication and the unpublished data that were part of their research. This publication has now been retracted. On April 19, 2022, Nature published a Retraction Note and QuTech published an update.

The CWI issued its advice on February 21, 2023. After receiving ‘zienswijzen’ (opinions) from the accused, the CvB formulated an initial judgment (July 11, 2023) and a final judgment (November 7, 2023). The final judgement was as follows: In relation to the 2017 publication, Leo Kouwenhoven was not at fault. The CvB judged that the other accused party acted negligently.

The investigation into the other publications was terminated because the initial report for these publications was withdrawn. The internal data investigation and subsequent disclosures by QuTech have ensured that the scientific community is well informed about the data, the selection of data, the processing of the data, how these data were presented in the publications, as well as the irregularities that were found. The CvB concludes, on the advice of the CWI and after the withdrawal of the internal report, that there is no foundation and necessity for further investigations by the CWI into the other publications.

The anonymized case report of the second integrity investigation is published on the UNL website.

Data investigation & disclosure by QuTech

In addition to the CWI investigations regarding the two Nature publications, QuTech conducted a data investigation, to identify possible discrepancies between published figures and raw data in a number of additional publications. Some errors were found in a few publications, that we publicly disclosed through online repositories. For one publication, this led to a correction after the journal editors consulted independent external experts. The errors did not affect the conclusions drawn in this publication.

For transparency, and to allow the scientific community to check the accuracy of the publications themselves, the authors and QuTech have made the data of the additional papers publicly available through online repositories[i]. These repositories contain all the data underlying published figures without processing (the so-called raw data), as well as all raw data collected around those figures. We would like to point out the uniqueness of disclosing such elaborate data sets, as this is not common in the scientific community.

Data management policy and integrity program

Partly as a result of this case, QuTech has drawn up a progressive research data management policy, effective per April 2022, and has started an integrity program. Compared to other research institutions, the data management policy is very extensive: QuTech publishes at least all raw data underlying published figures, as well as the processing scripts.

Majorana research at QuTech remains promising

The Majorana particle forms the basis of one of the four lines of research along which QuTech is pursuing the development of a quantum computer. At QuTech, other building blocks for the quantum computer, based on superconductors, semiconductors and colour centres, are also being researched. These types of quantum bits (also known as qubits) are part of the National Agenda for Quantum Technology.

Despite the setbacks, Majorana states remain promising, as qubits based on Majorana states could potentially be more robust than other types of qubits. PhD students and postdocs at QuTech have not been idle in recent years. They were successful in developing an alternative approach for realising Majorana states (Nature, 2023 & Nature, 2024). This new approach is promising for fundamental quantum research and possibly later as the basis for a new qubit.

Leo Kouwenhoven returns to QuTech as University Professor

After concluding all investigations, TU Delft and QuTech announced on 28 November, 2023, that Leo Kouwenhoven will be re-appointed at TU Delft as University Professor at QuTech, per 1 March 2024. Leo Kouwenhoven is one of QuTech’s founding fathers and has been a full professor at TU Delft since 1999. QuTech is delighted that he has returned, to contribute his extraordinary knowledge and expertise to advance research in pursuit of robust qubits. The research field in which Leo Kouwenhoven operates is of fundamental significance and involves cutting edge technology. In a recent Perspective, he reflects on 10 years of Majorana research.

Relevant links

• The anonymized case report of the first integrity investigation on the UNL website
• The anonymized case report of the second integrity investigation on the UNL website
• The public part of the expert report in the first CWI investigation
• The LOWI advice in relation to the first CWI investigation
• QuTech Data management policy
• QuTech social safety