More reaction predictions

19 November 2010 - Computational chemistry

CSI-chemistry-Shellhammer-2010.svg.png And staying on the topic of computational chemistry (see previous topic) Shellhamer et al. (DOI) have run the electrophilic addition of chlorosulfonyl isocyanate (CSI) with a monofluoroalkene through the computer, again confirming the experimental result. CSI cycloaddition is stereospecific as the E-isomer gives the E-lactam and the Z-isomer the Z-lactam which is taken as evidence for a concerted reaction. A dipolar intermediate would produce a cis/trans mixture as a result of freed-up bond rotation (given the opportunity). Are we happy then with the results?. Not really. It would have been nicer if the methodology would have been able to distinguish between concerted and dipolar by tweaking the nature of the alkene substituent (not just fluorine). And something more disturbing: the computer can reproduce a concerted transition state (B in diagram) but (we recommend the Shellhammer team with their honesty) the obvious dipolar intermediate C (electrophilic CSI at carbon) is nowhere to be found and instead the computer coughs up intermediate A featuring a seemingly secondary carbocation. The supplementary info does not provide details on charge distribution so the negative charge at nitrogen is an uneducated guess. Who is going to sort out this mess?