You might have noticed a certain trend in contemporary approaches to mechanistic investigations using computation: the (over)use of water. It is curious that 10 years ago people rarely considered water and now it is the fixture of almost every other paper, seemingly to patch holes in mechanistic arguments.
Jokes aside, molecular-level involvement of adventitious molecules is an interesting topic. Of course, experiment is the “real deal” and I pay attention to studies that reveal strange rate laws that hint at solvent involvement in transition state assemblies. Several days ago I came back from Halifax, Nova Scotia, where I heard a great talk by Joseph Moran of the University of Strasbourg. He described a system that reveals the enabling role of nitromethane in catalysis (http://pubs.acs.org/doi/abs/10.1021/jacs.5b06055). While listening to his talk, I took note of a peculiar paper by Berkessel, which is not something I was familiar with. In this work, there is experimental evidence for the involvement of 3 HFIP molecules in the rate-limiting step of olefin epoxidation. This reminds me of Ryan Hili’s studies of aziridine aldehyde dimers, where we always noted an important role of trifluoroethanol. However, our kinetic work has not allowed us to conclude that something remarkable was happening on a “molecular level”. In the Berkessel case, the kinetics clearly point at the involvement of 3 molecules of HFIP in the transition state. Do let me know if you are aware of other interesting cases that implicate adventitious molecules in rate-limiting steps.
Baylis-Hillman is famously accelerated by water and alcohols (but unfortunately they shift the reaction equilibrium towards lower conversions) and this led to proposal of using 3-hydroxyquinuclidine as a catalyst that is more effective than DABCO. But even 3-hydroxyquinuclidine catalyzed B-H benefits from adding a tiny amount of hexafluoroisopropanol, in about 0.3 mol ratio to the catalyst.
Another notable catalytic effect of hexafluoroisopropanol and trifluoroethanol is on conjugate addition of secondary amines to acrylates, and ring-opening of epoxides with secondary amines. In both cases the catalysis is presumably due to protonation. In those cases one would use TFE or HFI as a solvent, in noncatalytic amounts.
Thanks – these are great cases.