Porco, Su, Lei, Bardhan and Rychnovsky. ACIEE, 2006, Early View. DOI: 10.1002/anie.200602854.

So the structure has been confirmed. And using a nice total synthesis too. So do we know if La Clair has lied? Not conclusively; someone would have to follow his synthesis exactly, and try to see where the discrepancies creep in [1]. However, it’s very difficult to see any element of honesty in his work, and what’s more annoying is why. Aferall, had he gone after the original target, completed the synthesis and then discovered that the spectral data didn’t match, it would hardly be the first example. This happens often enough, and he could simply have postulated his thoughts on the discrepancy, as the authors of this paper did with their synthesis of pseudo-brevinal. But no, he seems to have lied. What does this mean for us?

1. Science in general takes another knock. Okay, so this isn’t in the same league as cold-fusion or human embryo cloning, but its another lie in a big journal. We all know that authors can at times be somewhat liberal with the truth, as discussed in Dylans article. And then there’s the whole Sames debacle…(again, linking to the Tenderbutton – your guide to lies in chemistry…), but these are quite different to the lead author lying in print.

2. La Clair’s out of a job. Or at least he should be. No-one who publishes a work of fiction in ACIE should be employed in science, even at the manual-titration end of things.

3. Peer review takes a hit. Again. Okay, so perhaps the scale of the problem wasn’t obvious this early, but there were certainly signs of inconsistency:

“The ¹H NMR spectra for this Communication were determined by contract services. The spectra provided in the Supporting Information were collected by N. Voss (Berlin, Germany). The operator added the peak for CDCl₃ to the spectrum of synthetic hexacyclinol (1), however, this was done incorrectly at [~]7.5 ppm and against the request of the author. Additionally, one spectrum was duplicated and a copy of the spectra for natural 5-epi-hexacyclinol was not provided.”

That should certainly have had the review board / editors worried, and the fact it was a “note added in proof” should really have set alarm bells going.

4. Rycnovsky, and others working in the field of NMR prediction using DFT methods (rather than ChemDraws guess-it-and-hope method) get a boost. Clearly, his method, given in the previous paper, had a lot of validity now. It’s clear from the Porco paper that they have a lot of experience in that area, so their proposed (and now confirmed) structure wasn’t pulled out of thin air (they had previously worked on the related structure, panepophenanthirin). But for those working on total syntheses where the targets configuration isn’t well defined – you have another tool.

5. A healty reminder not to believe everything you read. Okay, we know some journals are more trustworthy than others, but ACIEE was near the top of my list, and I’m disappointed. However, Organic Syntheses stays at the top for now for repeating everything… but even that’s not water-tight.

Trust no-one!

Additional: Some other sites with hexacyclinol commentry/debate:

In the Pipeline, The Endless Frontier, The Chem Blog, and of course, Tenderbutton.

In part III (I’ll put that up later this evening [2]), we’ll look at Porco’s actual synthesis.

Even more Additional: it’s made it to C&EN!
[1] Worst Ph D project ever.
[2] Here in industry (for my CASE placement) we have chiral prep HPLC, Companion automated chromatography, loads of RPLC-MS, cryoprobes coming out of our arses, but no bloody ChemDraw. Stuck with damn ISIS-draw, and it Just Looks Wrong. So it’ll have to wait till I get home…