Towards Improving The Accuracy of Implicit Solvent Models and Understanding Electrostatic Catalysis in Complex Solvent Environment

Abstract

This thesis develops improved protocols for studying reactions in solution and uses them to explore the possibility of harnessing complex non-standard solvent environments to catalyse chemical reactions. The thesis covers different but related topics: Improving the accuracy of implicit solvent models. Implicit solvent models are simple cost-effective strategies for modelling solvent as a polarizable continuum. However, the accuracy of this approach can be quite variable. Herein, we examine approaches to improving their accuracy through cavity scaling, the choice of theoretical level and the inclusion of explicit solvent molecules. For SMD, we show that the best performance is achieved when cavity scaling is not employed, while for PCM we present a series of electrostatic scale factors that are radii, solvent and ion type dependent. For both families of method, we also highlight the importance choosing an appropriate level of theory, and identify when explicit solvent molecules are required.. Modelling electrostatic catalysis in complex solvent environment. Recent nanoscale experiments have shown that electric fields are capable of catalysing and controlling chemical reactions, but experimental platforms for scaling these effects remain elusive. Herein, two different approaches to addressing this challenge are explored. The first is using the internal electric field of ordered solvents and ionic liquids, the second is using the electric fields that form naturally at the gas-water interface. A multi-scale modelling approach was developed using polarizable force field based molecular dynamic simulation, post-HF, DFT and semi-empirical quantum chemical calculations. We showed that after exposure to an external electric field, ensembles of solvent or ionic liquid molecules become ordered and this ordering can generate an internal electric field, which persists even after the external potential is removed. Experimental collaborators subsequently detected this field as an open-circuit potential that is strong and long-lived. Computationally we showed that this field is enough to lower reaction barriers by as much as 20 kcal mol-1, and we also developed a predictive structure-reactivity model to choose ionic liquids that optimize this field. In the second approach, we harnessed the electric fields of the gas-water interface. A collaborator showed that in the presence of static, inert gas bubbles, the oxidation potential of HO anion/HO radical was dramatically lowered (by more than 0.5V), much more than any subtle concentration effects predicted by the Nernst equation. Further experiments showed that a high unbalanced concentration of HO- ions (as much as 5M) accumulate at the interface. Our multi-scale modelling calculations showed that this reduction in potential was due to the mutual repulsion of the HO- ions and as little as 1M unbalanced excess was enough to explain the experimental results. The work raises opportunities in reducing the cost of electrochemical processes, and points to electrostatic effects contributing to the well-known catalytic effects of "on water" reactions. Works in this thesis are expected to be useful in the future studies of solution-phase pKa, redox potential, electrostatic catalysis and ionic liquids-based electrochemical devices. Show more