Reactive molecular dynamics: from small molecules in gas phase to enzymatic reactions

Abstract

In this thesis, a comprehensive range of applications for the reactive molecular dynamics method Multi-Surface Adiabatic Reactive Molecular Dynamics (MS-ARMD) will be presented. My research investigates molecules calculated in gas phase as well as simulations in solution and in solvated enzymes. The motivation for each simulations will be highlighted, the parametrization of the energy functions explained and comparisons with other reactive Molecular Dynamics (MD) methods will be made. Reactive molecular dynamics simulations allow the study of experimentally non amenable time- and length scales, providing essential insights into the mechanistic details of reactions. Knowledge drawn from such simulations allow i.a. the refinement of computational models and synthetic pathways, and finding new applications. The power of MS-ARMD lies in the calculation of converged reaction rates since thousands of individual trajectories can be run. This allows for the generation of statistically significant ensemble sizes for analysis such as quantitative characterization of final state distributions. This is usually not possible for conventional mixed quantum mechanics/classical mechanics (QM/MM) molecular dynamics or full ab initio molecular dynamics simulations due to the computational cost of the quantum calculations. Simulations in condensed phase allow for direct comparison with experiments, which are often performed in (aqueous) solution or in the presence of a biologically relevant enzyme. Furthermore, those simulations provide additional insight into the mechanism of reactions with chemical and biochemical interest, beyond experimental findings. It was shown that MS-ARMD force fields (FF) parametrized in gas phase can be applied directly, without further modifications, to condensed phase simulations. This allows for quantitatively fitted FF to be used for simulations in solution and enzymes which is an additional advantage of MS-ARMD in comparison to other reactive molecular dynamics methods.