Chemistry Seminar: Dr Junming Ho

A universal approach for solvation modelling: are we there yet?

ABSTRACT

Many chemical and biochemical reactions occur in the solution phase, and the solvent can often exert a strong influence on the outcome of a reaction. As such, our ability to accurately predict the rate and equilibrium constants of reactions depend on how well our computational models can describe solute-solvent interactions.

Currently, quantum mechanical (QM) implicit solvent models (e.g. PCM and SMD) are the most popular methods for modelling solvation because of their computational cost, moderate accuracy and ease of use. In this talk, I will highlight the successes, failures and lessons that we have learned from these models over the last 10 years. In particular, I will provide some examples of how we were able to make accurate predictions to guide organic synthesis and the design of new synthetic procedures.

I will also introduce an alternative approach based on explicit solvent molecular mechanics

(MM) simulations which are in principle more robust, but at a significantly higher computational cost. Unfortunately, it is currently unclear if the extra cost necessarily translates to improvements in accuracy, if any.

In this talk, I will present a case study to compare the performance of implicit and explicit solvent models, and highlight a strategy for the development of future solvent models.