Retraction Watch

Tracking retractions as a window into the scientific process

Second expression of concern appears for chemistry group under institutional review

with 21 comments

chemsciThe journal Chemical Science has issued an expression of concern over a 2012 article by a pair of Texas researchers whose “unclick reaction” work has been under scrutiny by their institution.

The article, “Homonuclear bond activation using a stable N,N-diamidocarbene,” was written by Kelly M. Wiggins and Christopher W. Bielawski, of UT Austin. It’s the second EoC that we know of for a paper by Wiggins and Bielawski. We covered a previous one, from Science, that appeared in June.

Here’s the notice (pdf):

The Royal Society of Chemistry has been contacted by the corresponding author of this article and the Research Integrity Officer at The University of Texas at Austin regarding concerns of scientific misconduct affecting this article. The Research Integrity Officer has informed us that an investigation to ascertain the validity of the work reported has found that scientific misconduct has taken place.

Chemical Science is publishing this expression of concern in order to alert our readers that we are presently unsure of the reliability of the data reported in the article. We have contacted the Research Integrity Officer to request more details regarding the scientific misconduct determined in their investigation, in order to determine the appropriate course of action.

An expression of concern will continue to be associated with the article until we receive further information from the Research Integrity Officer on this matter.

The paper has been cited 12 times, according to Thomson Scientific’s Web of Knowledge.

Chemical & Engineering News reported last summer that:

Bielawski tells C&EN that a former group member, whom he declined to identify, came forward and admitted to manipulating data in the Science paper. Bielawski says that his lab “successfully repeated the experiments in question and found that the conclusions of the report were unchanged.” He has submitted a correction to Science to address the concerns. Neither Brantley nor Wiggins could be reached for comment.

Update, 8:30 a.m. Eastern, 12/4/14: As a commenter points out, there has been a third Expression of Concern about work by this group, for “Synthesis of poly(ethylene-co-acrylic acid) via a tandem hydrocarboxylation/hydrogenation of poly(butadiene),” a paper in Polymer Chemistry:

The Royal Society of Chemistry has been contacted by the corresponding author of this article and the Research Integrity Officer at The University of Texas at Austin regarding concerns of scientific misconduct affecting this article. The Research Integrity Officer has informed us that an investigation to ascertain the validity of the work reported has found that scientific misconduct has taken place.

Polymer Chemistry is publishing this expression of concern in order to alert our readers that we are presently unsure of the reliability of the data reported in the article. We have contacted the Research Integrity Officer to request more details regarding the scientific misconduct determined in their investigation, in order to determine the appropriate course of action.

An expression of concern will continue to be associated with the article until we receive further information from the Research Integrity Officer on this matter.

Nicola Wise

21st November 2014

Executive Editor, Polymer Chemistry

This paper has been cited just once.

Update 6 p.m. Eastern, 12/11/14: Chemical & Engineering News has an update on the story:

Officials at UT Austin tell C&EN that their investigation into the matter has concluded: “One author of several papers in question told UT officials that he or she—acting alone—had falsified and otherwise misrepresented data or figures in the papers, which led to the finding of scientific misconduct.”

Comments
  • blatnoi December 3, 2014 at 10:03 am

    I thought Bielawski was going to UNIST in Korea… but he’s still in Texas? Maybe he in Korea only for a month a year or something. Anyways, this is a very unlikely reaction and thus an exciting report and that’s why it got into Science.

    This correction does not assuage my concerns as a practicing chemist. It’s intuitively much more likely that the molecule is ripped apart at the ester and they apparently don’t observe that since all these re-zipping experiments are done. I would need to see a report from a different group and the technique used for something, before I believe that this reaction is true again, after the old expression of concern.

    So, one person admitted to making everything up (because it wasn’t working), and then another person in the same lab repeated it and found out it all works? C’mon…. I’m never going to be trying to pull apart azides by ultrasound, that’s for sure. There is only so much time to waste. And by the way, at least the first report before the correction is scientific misconduct and should be retracted, and a new manuscript submitted with the ‘new’ results.

    • excimer December 4, 2014 at 10:01 am

      The reaction isn’t that surprising, considering the other unusual reactivities that have been found using mechanical energy. But I agree, the Science paper should have been retracted in its entirety.

      It’s intuitively much more likely that the molecule is ripped apart at the ester
      For bond heterolysis, yes. However in mechanically-activated molecules, homolysis is much more likely. People are only beginning the explore the energy surfaces of mechanically-activated systems.

      • blatnoi December 4, 2014 at 6:38 pm

        That would be fine, that homolytic cleavage is a strange quality of mechanochemical activity, but do you really expect a triazole to rip apart via a single homolytic event? There are two bonds that have to be broken at least, and possibly concurrently so that the first one doesn’t recombine, and we get back to the part that even for the ester, the homolytic splitting event for a carbon-oxygen bond (or even a carbon-carbon) is much lover in energy so I still expect it to rip apart somewhere else and cross-link terribly.

        I mean, call me old-fashioned and everything, but if these energy surfaces of mechanically activated reactions are so different, then why do you need a copper catalyst at all to do the click reaction? I mean, we don’t live in some alternative universe where the products are still not more than about 20kcal/mol downhill. Shouldn’t you be able to use ultrasound to do the forward reaction very effectively too? Maybe not in this particular system that was in the Science paper (I mean it’s still plausible since you’re putting in energy), but maybe mix a free azide and alkyne and hit them with ultrasound? That would really be a nice report actually; maybe somebody has done it?

  • Conrad Seitz MD December 3, 2014 at 11:20 am

    Yes, it should have been retracted already. As blatnoi says, unless the work is replicated by another group entirely, it’s suspect. Of course “scientific misconduct” might not “affect the results” but once it is shown (by the university’s scientific integrity office no less), the whole thing is suspect.
    Thanks, blatnoi, for trying to explain the chemistry behind this (since it has been thirty years since my last organic chemistry class.)

  • NotClaireFrancis December 4, 2014 at 7:24 am

    Actually it is the third Expression of Concern, as one was also published in the RSC journal Polymer Chemistry on 21st November.

    http://pubs.rsc.org/en/content/articlelanding/2015/py/c4py90087a#!divAbstract

  • Sylvain Bernès December 5, 2014 at 12:22 pm

    Such a bunch of expressions of concerns is alarming for the lab., because an entire portion of the research becomes questionable, and virtually uncitable. The near future for the following high-impact papers is particularly bleak:
    Wiggins et al. (2010). JACS, 132, 3256-7 (http://dx.doi.org/10.1021/ja910716s)
    [are CD spectra in Fig. 1 *really* reproducible?]
    Wiggins et al. (2011). JACS, 133, 7180-9 (http://dx.doi.org/10.1021/ja201135y)
    Wiggins & Bielawski (2012). Angew. Chem. 51, 1640-3 (http://dx.doi.org/10.1002/anie.201107937). Note that this one is a two-authors paper.

    • n/a December 8, 2014 at 9:46 am

      There are plenty of high impact papers from the Bielawski group that do not include that particular author and are based on very real, reproducible results. It is unfair to allow the mistakes of a single individual ruin the reputations of everyone else associated with the group.

      • NotClareFrancis December 8, 2014 at 8:21 pm

        With all due respect n/a, that is precisely what Professor Sylvain said – a significant portion of the work (those articles featuring a particular authoring combination/ordering, as highlighted) are simply viewed with suspicion by the community. Any articles that feature that particular combination of authors, and there are a number of them, will continue to be viewed with a certain degree of suspicion until the Research Integrity Officer reports exactly what has transpired, which I hope will be sooner rather than later, and validates these articles or otherwise. Reputations are hard to gain and easy to lose.

        That said, this is somewhat different to other well known research integrity situations in chemistry because, unless we hear something very unexpected from the Research Integrity Officer, it is not as if other researchers were telling the PI that the chemistry wasn’t working or that the results couldn’t be reproduced.

    • NotClareFrancis December 9, 2014 at 1:28 pm

      I think Professor Sylvain has spotted something in Figure 1, the CD spectra in the first JACS paper listed. (JACS (2010) 132, 3256

      Take the blue CD spectra on the right of the Figure and flip it vertically (i.e turn it upside down), then compare it to the red one on the left…

      • blatnoi December 9, 2014 at 3:09 pm

        Hey, that is true. The CD spectrum is flipped. Although the binol moieties are stereoisomers, I would not expect such 100% ideal agreement between the samples, let alone over time during an experiment (the same is seen in Figure S8, but racemization over time is not shown there). So, if only one isomer racemized and the other one did not (causing the copying and flipping), this would actually be a much cooler paper. Since this was not reported, it’s hard to say what happened.

        Well… somebody probably informed Peter Stang (editor of JACS) by now I suppose. What a mess.

        • NotClareFrancis December 10, 2014 at 11:29 pm

          I am sure that Professor Stang is expecting problematic articles from this group and I would hope that both he and Professor Bielawski are aware of this particular figure. However, until JACS itself publishes an Expression of Concern or similar or the Research Integrity Officer reports I have no way of knowing if that is the case. Who knows how long that could take?

          I am going to make the assumption that the JACS Editorial Board are fully aware of this CD spectrum because Professor Bielawski and the Research Integrity Officer have advised them directly, much like he advised the Editors of the RSC journals, and therefore the JACS Editors do not need another e-mail from me telling them what they already know.

          I don’t think the JACS Editors would be happy with me for mentioning that CD spectrum on a public forum (albeit deep in the comments section) before advising them and I regretted that comment as soon as I posted it. I think e-mailing JACS first would have been the correct course of action, but I wanted a second opinion on the figure and wasn’t thinking clearly at the time.

          There is no set protocol for informing a journal of data concerns. Perhaps journals should appoint a full-time Data Integrity Officer, like Organic Letters has done, who can be e-mailed directly and respond to concerns and advise the Editor accordingly.

          I hope someone can quietly point the comments on this thread out to the necessary people, and I hope they act speedily.

  • Sharon O'Connor December 5, 2014 at 5:12 pm

    If the manipulated data had no effect on the outcome, why was it included at all?

  • lhac December 18, 2014 at 9:03 pm

    As pointed out above, it is difficult to assess the reliability of the work of this group at this time. Is problematic work limited to that where the one specific student was involved?
    It does not help that the senior author has given false testimony as an expert witness (http://floridaip.blogspot.de/2014/07/expert-lied-on-stand-new-trial-right.html)
    The court specifically used the following words: ” … notwithstanding the serious misconduct by Dr. Bielawski….” “… This was a serious and detrimental occurrence. It insulted the jury, violated the integrity of the judicial process, and interfered with the search for the truth that is the hallmark of our trial-by-jury system.”
    (see: http://docs.justia.com/cases/federal/district-courts/florida/flmdce/3:2011cv00819/261721/393)

    • Conrad Seitz MD December 19, 2014 at 2:43 pm

      I read the report from “Florida IP” and it seems that Dr Bielawski lied on the stand for the defense, although it didn’t affect the outcome of the case (yet), and the defense still won. How much was he paid? That’s not the point. Lying in court is bad (my mother told me that.)
      I don’t think anything Bielawski has his name on is probative, especially when you read his statement again, in the main body of the post above: “Bielawski tells C&EN that a former group member, whom he declined to identify, came forward and admitted to manipulating data”… A ghostly “former group member” did it. Really?

    • blatnoi January 3, 2015 at 6:21 am

      I can see how the problem at the trial would occur though. In articles, everyone writes “Bielawski (or any other name of a big professor) performed this experiment…” or “He showed that this reaction proceeds in this fashion…”, when of course, he did nothing of the sort personally. It was his students, and probably the first author to be more precise, and the name of the corresponding author is a stand-in for the laboratory. This sort of attitude produces humorous pictures of famous senior scientists (who have no business being there) of standing in a laboratory in a labcoat borrowed hastily from a postdoc and pipetting.

      So they asked him if he personally performed the XPS and TOF-SIMS and he said ‘Yes, I did”, which is a perfectly fine answer for a senior scientist if you’re writing an article. But of course, we all known that it was performed by a staff scientist or a lab member, and the boss would probably cause a short circuit if they attempted to operate the machinery, but at least they can make sense of the data in its final form. Apparently this is NOT accepted practice for juries and lawyers, and I can see where they are coming from. However, I have a lot of sympathy with Bielawski (without knowing the exact details of the case) on this one.

      • NotClareFrancis January 3, 2015 at 11:28 am

        Yeah, I agree with that possible interpretation. Bielawski could have gotten the meaning of “you” used by scientists and the very precise language used by the legal system mixed up.

        I remember a talk a professor (who shall remain unnamed) gave some years back concerning the synthesis of various nitrogen containing compounds which, looking at the structures, would have been considered potentially highly explosive by any chemist. The professor stated that they were quite stable, and showed that he had all of the fingers on his hands as proof that they were safe to handle. Whether the students actually doing the experiments still had all of their fingers was not addressed.

  • NotClareFrancis January 2, 2015 at 10:53 am

    Happy New Year everyone.

    This group have a fourth expression of concern, this time for the article Journal of Materials Chemistry 2011, 21, 8355-8359. This journal is also published by the Royal Society of Chemistry. It should be noted that a student has already admitted falsifying data (see Chem and Engineering News, Dec 15th)

    http://pubs.rsc.org/en/content/articlelanding/2011/jm/c0jm03619f#!divAbstract

    The journal has been split into three sections, so the EoC was attached next to the supporting information on the Article Landing page on 8th December 2014. This is why it is not as visible as the other three EoC’s.

    http://www.rsc.org/suppdata/jm/c0/c0jm03619f/c0jm03619f2.pdf

    A commenter on another blog (In the Pipeline) pointed out that the data in this article looked very similar to some that had been published in JACS 2010, 132, 16631-16636. I am not qualified to comment on whether the data reported over the two articles is valid and the trends to be expected or otherwise.

  • Sylvain Bernès January 2, 2015 at 7:41 pm
  • NotClareFrancis January 3, 2015 at 10:12 am

    Another expression of concern from Journal of Materials Chemistry, 2010, 20, 5709-5714
    Statement published 21st November 2014 on the article landing page.

    http://pubs.rsc.org/en/content/articlelanding/2010/jm/c0jm00417k#!divAbstract

    Direct link: http://www.rsc.org/suppdata/jm/c0/c0jm00417k/c0jm00417k2.pdf

    I am pleased JACS have acted, I hope that Angewandte do the same very soon.

  • Sylvain Bernès January 8, 2015 at 1:48 pm

    The Angewandte Chemie I mentioned one month ago has now been retracted:
    http://onlinelibrary.wiley.com/doi/10.1002/anie.201410475/full

  • Post a comment

    Threaded commenting powered by interconnect/it code.