DDCP and its two enantiomeric product channels

Bifurcating PES with a valley-ridge inflection point

It’s been awhile since I’ve done a post… I’ve been in the process of changing home base from San Francisco back to Bristol. But there’s a paper I’ve recently written which I’m excited about.

It’s a computational study of a molecule, DDCP (meso-2,3-difluoro-2,3-dimethyl-diazo-cyclopropane), which was identified by my colleague, Professor Barry Carpenter. For the last year or so, Barry has been playing around with the solution phase dynamics code that I’ve been developing. He chose DDCP because it has a very interesting potential energy surface (PES) topology: it exhibits a so-called ‘valley-ridge inflection point’ – i.e., passage over a transition state that bifurcates on the way down toward products. In this case, the bifurcating path leads to products which are distinct enantiomers. The simulations that we ran suggest that enantioselectivity on such a PES topology can in fact be dynamically guided by transient interactions with the surrounding solvent environment in which the reaction takes place. We ran a series of MD simulations in achiral solvents, and (as you might expect) these showed equal ratios of the product (R) and (S) enantiomers. However, by computationally screening a range of different solvents, we “discovered” a chiral solvent which gives a dynamically-guided enantiomeric product excess of 15.2 ± 2.1%. It appears that interactions with the solvent cage in the early stages of the reaction preferentially guide the system down a specific product channel, and that excess vibrational energy is quickly dissipated to the solvent degrees of freedom. The observation of this solvent effect is exciting: it is approximately an order of magnitude larger than experimentally observed excesses where the conversion of products to enantiomeric products occurs over separate transition states, and it makes an exciting target system for experimental investigation!