A prominent Stanford University chemistry lab has been forced to retract a paper in the Journal of the American Chemical Society (JACS). According to the retraction notice:
Due to inconsistencies between some of the assigned structures and the experimental data that appear in the paper, the authors retract this publication. We regret very much this unfortunate occurrence.
The Retraction Watch tipster who alerted us to this retraction explained what the original paper reported:
The original paper claimed to show a chemical reaction that would convert rather simple starting materials into complex products. These particular products, or chemicals similar to them, could have interesting medicinal properties. The transformation reported in the paper is similar to others that have been published by the Trost group that use a similar ruthenium catalyst. This report, however, was an extension of these methods that had implications to explain the mechanism of other transformations and possibly lead to a number if other applications. Although it was an interesting extension of prior work, it was also not entirely unbelievable based on previous results.
The retraction notice implies that some of the structures of the products made in this paper are incorrect, but it is not clear how or which ones. It would be interesting to know if any of the conclusions of the paper are correct, since the mechanism they propose for the reaction could be applicable to other reactions. It would also be interesting to know what the actual structures of the products are. Likely, this was an error in interpreting NMR spectra – while this technique (which you mentioned in a recent post) is very useful to organic chemists for determining molecular structure, the results can be misinterpreted. It’s possible this was an honest mistake, as mistakes of this type have been known to happen.
So what happened? Senior author Barry Trost tells Retraction Watch:
The fundamental new reaction being reported did occur but a significant number of examples were not properly documented. There was simply a human error that occurred which one of my group members fortunately caught. We have subsequently revisited the chemistry and have subsequently published a paper reporting the new validated results. This time we had 3 group members validate the results just to make sure. We did discover a new and better catalyst in this review.
Sure enough, that new paper, “Propargyl Alcohols as β-Oxocarbenoid Precursors for the Ruthenium-Catalyzed Cyclopropanation of Unactivated Olefins by Redox Isomerization,” includes this footnote:
An initial report of this reaction using catalyst 1 was retracted because many of the reported examples were subsequently found to be questionable. This unfortunate circumstance led us to a total re-examination and the discovery of a much improved, verified, and reliable procedure, which is the topic of this paper.
We couldn’t ask for a more straightforward example of self-correcting science.
Good example of how things should work.