The chemistry blogs have been buzzing this week with the story of a paper in the journal Organometallics that may — or may not — contain fabricated data.
But what makes the story a bit juicier — and yes, it’s sad that fabricated data is a bit ho-hum for us — is that one of the authors of the article appears to have been caught in the act of instructing the first author to make up results.
The article, “Synthesis, Structure, and Catalytic Studies of Palladium and Platinum Bis-Sulfoxide Complexes,” appeared last month and came from a group at the Institute for Organic Chemistry at the University of Zurich. The authors were Emma E. Drinkel, Linglin Wu, Anthony Linden and Reto Dorta.
As ChemBark reported earlier this week:
While the paper itself is a straightforward study of palladium and platinum bis-sulfoxide complexes, page 12 of the corresponding Supporting Information file contains what appears to be an editorial note that was inadvertently left in the published document:
Emma, please insert NMR data here! where are they? and for this compound, just make up an elemental analysis… [our italics]
This statement goes beyond a simple embarrassing failure to properly edit the manuscript, as it appears the first author is being instructed to fabricate data. Elemental analyses would be very easy to fabricate, and long-time readers of this blog will recall how fake elemental analyses were pivotal to Bengu Sezen’s campaign of fraud in the work she published from 2002 to 2005 out of Dalibor Sames’ lab at Columbia.
The compound labeled 14 (an acac complex) in the main paper does not appear to correspond to compound 14 in the SI. In fact, the bridged-dichloride compound appears to be listed an as unlabeled intermediate in Scheme 5, which should raise more eyebrows. Did the authors unlist the compound in order to avoid having to provide robust characterization for it?
Chemistry Blog posted a letter from John Gladysz, the editor of the journal, which is a publication of the American Chemical Society:
Wednesday 07 August
Dear Friends of Organometallics,Chemical Abstracts alerted us to the statement you mention,which was overlooked during the peer review process, on Monday 05 August. At that time, the manuscript was pulled from the print publication queue. The author has explained to us that the statement pertains to a compound that was ”downgraded” from something being isolated to a proposed intermediate. Hence, we have left the ASAP manuscript on the web for now. We are requiring that the author submit originals of the microanalysis data before putting the manuscript back in the print publication queue. Many readers have commented that the statement reflects poorly on the moral or ethical character of the author, but the broad “retribution” that some would seek is not our purview. As Editors, our “powers” are limited to appropriate precautionary measures involving future submissions by such authors to Organometallics, the details of which would be confidential (ACS Ethical Guidelines, http://pubs.acs.org/page/policy/ethics/index.html). Our decision to keep the supporting information on the web, at least for the time being, is one of transparency and honesty toward the chemical community. Other stakeholders can contemplate a fuller range of responses. Some unedited opinions from the community are available in the comments section of a blog posting: http://blog.chembark.com/2013/08/06/a-disturbing-note-in-a-recent-si-file/#comments
If you have any criticisms of the actions described above, please do not hesitate to share them with me. Thanks much for being a reader of Organometallics, and best wishes,
John Gladysz
Editor-in‐Chief
We have tried to reach Dorta by email and will update this post if we learn anything.
Reblogged this on Life Science PhD Aventures and commented:
An update to yesterdays post on Organometallics. Editor responds to peer-review oversight and ethics.
It’s possible, just, that the comment was a joke. I’ve written much worse in manuscripts in prep.
Also, it’s refreshing to see fraud in something other than western blots, innit?
I guess it really depends on how long this paper was in the making. Both the corresponding author and the lead author have left the university that this work was performed at, in 2012 and 2011 respectively. Their linkedin profiles show that Emma is in Brazil performing a PDF and the PI is now in western Australia. If the manuscript was being edited in the last year and a half, then this comment is likely a push by the PI for the student to fake data as neither of them were working on the same project anymore.
As embarrassing as this is for Emma Drinkel, at least she didn’t fake the data. That’s how I’m looking at this.
Here she is: http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=H1448379
Emma might be a simbol of resistance in the new world science.
You guys don’t cover chemistry retractions as often as other fields, I’m not feeling the love here. 🙁
There is a very strange retraction at the flagship journal of the society from which Organometallics is from. It’s a very strange and cryptic retraction since it just says that compound 1 was mis-characterized, but in the original paper compound 1 is the catalyst that did all the chemistry, so does that mean that the catalytic results are just made up, that there was an impurity in 1, or that the structure of 1 is just a bit off, but it is still active as a catalyst?
pubs.acs.org/doi/abs/10.1021/ja406816b
I think you should cover this one. Although I am pleased with the current post as well. Finally we will have more chemistry content on this blog. Represent.
If the catalyst isn’t what they state the paper must be retracted as the chemistry isn’t in accordance to the paper. We can probably be pretty sure that whatever catalyst they actually had produced the results…
If the catalyst structure is off a bit because of a mis-interpreted NMR spectrum, then a simple correction should be enough. If they have traces of Pd, then it’s completely different chemistry and it’s carelessness and should be retracted. I still think the retraction note is vague and lacking in information.
I’m not a native speaker, but I always had the impression that “make up” could also mean “compile”, which would be a perfectly reasonable comment if said analysis was missing in the SI. The various dictionaries I consulted seem to agree with my intuition, so I think it is very questionable to raise accusations of fraud at this point.
Indeed.
I don’t think make up data means compile. You can make up a sample for elemental analysis but making up elemental analysis date clearly implies cooking up non existent data.
Re Bernd and Deus: I couldn’t disagree more. Printers might use the phrase “make up” in a specialist sense to mean prepare or compile but in colloquial English, “just make up” unambiguously means to fabricate.
What the author means by this compound being “downgraded” is not at all clear.
I think “unambigously” is wrong. Looking for “just make up” on Google’s discussion search brings up lots of examples where people (native speakers) do not mean “invent”, e.g.:
https://groups.google.com/forum/#!original/uk.rec.models.engineering/U42MwitgATc/K6oC10Ucr1MJ
https://groups.google.com/forum/#!original/rec.games.frp.dnd/p9KxA0Az0PI/Jl3JTWpJaZYJ
https://groups.google.com/forum/#!original/rec.arts.comics.marvel.xbooks/z-ysvxehwbk/OlIN16KIPg4J
https://groups.google.com/forum/#!original/alt.magic/1GVra-7mLBg/kBL0fUi6arQJ
I agree that it’s probably a less common use, but presenting an accusation of scientific fraud based on a non-native speaker’s uncommon use of English is wrong.
I agree Bill. The magic word that eliminates possible ambiguity is “just”, especially when preceded by very specific instructions to “insert nmr data here”.
It will be interesting to see what happens now that the original microanalytical data/reports have been requested. Those analyses will have been done at a separate lab within the University of Zurich, including ETH. They will not mess around with this. I’ll speculate that more will be found wrong with the microanalyses given the statement on page 2 of the Supporting Information that “hygroscopic compounds were corrected for water content”. For example, the microanalysis of compound 7 (SI page 6) is calculated as a solvate with 3 water molecules. However, the x-ray structure of compound 7 shows no water and instead a solvate with 1 molecule of dichloromethane. My impression is that the microanalytical data were adjusted with whatever amount of hypothetical water was required to make them satisfactory. Not cool if true, and consistent with the “just make up” statement.
A correction to my earlier post which I did not word very well: The microanalytical data do not appear to have been adjusted to render them satisfactory. Rather, the elemental composition was adjusted with hypothetical water so as to match the data. There appear to be other examples, not just compound 7, and consistent with a just make it up attitude.
For me, the most dangerous thing is the journal saying “Chemical Abstracts alerted us to the statement you mention,which was overlooked during the peer review process”…. I am sorry, did ANY of the reviewers actually READ the manuscript?!
I mean, sure, scientists trying to deceive is very sad and sometimes life threatening (clinic research I guess?), but really, nobody caught THAT SENTENCE during the peer-reviewed process (which I am assuming it was twice)? That’s the only one job the reviewers have.
And that is why we can’t have nice things in science.
“Supplementary Information” – many reviewers will not read that, and definitely not in detail; they perhaps only look at the SI if they think they need that to understand a part of the paper. The SI is often the stuff you didn’t get to see at all in the old days, before electronic journals were available.
This could well be innocent: One co-author asks the other to insert the data.
The “make up” in the second half is somewhat suspicious, but need not be anything more than a sloppy formulation. Further, it does not affect the first half, which is likely the more important (from what I can tell without more context).
“just make up the data” is plain English for fraud.
“Downgrade” means we made up a lot of stuff, got caught, and are trying to cover our asses.
For the record, an editorial review of this case has been published.
http://pubs.acs.org/doi/abs/10.1021/om401186q
And the Dorta’s paper is now published, along with a looong “correction” running over 2 pages (available free of charge!), which addresses 28 (sic) issues in the original paper and SI.
Article: http://dx.doi.org/10.1021/om4000067
Correction: http://dx.doi.org/10.1021/om4011615
Nothing serious happened. The private joke about making up elemental analyses was just “internal communication among the authors”.