Molecular dynamics simulations of elementary chemical processes in liquid water using combined density functional and molecular mechanics potentials. II. Charge separation processes

Abstract

A new approach to carry out molecular dynamics simulations of chemical reactions in solution using combined density functional theory/molecular mechanics potentials is presented. We focus our attention on the analysis of reactive trajectories, dynamic solvent effects and transmission coefficient rather than on the evaluation of free energy which is another important topic that will be examined elsewhere. In a previous paper we have described the generalities of this hybrid molecular dynamics method and it has been employed to investigate low energy barrier proton transfer process in water. The study of processes with activation energies larger than a few kT requires the use of specific techniques adapted to ''rare events'' simulations. We describe here a method that consists in the simulation of short trajectories starting from an equilibrated transition state in solution, the structure of which has been approximately established. This calculation is particularly efficient when carried out with parallel computers since the study of a reactive process is decomposed in a set of short time trajectories that are completely independent. The procedure is close to that used by other authors in the context of classical molecular dynamics but present the advantage of describing the chemical system with rigorous quantum mechanical calculations. It is illustrated through the study of the first reaction step in electrophilic bromination of ethylene in water. This elementary process is representative of many charge separation reactions for which static and dynamic solvent effects play a fundamental role. © 1997 American Institute of Physics. ͓S0021-9606͑97͒00809-X͔

A QM/MM/continuum model for computations in solution: Comparison with QM/MM molecular dynamics simulations

“…In fact, this is an important advantage of the method since it makes it possible to determine a reaction path in solution, using a more sophisticated approach than standard continuum models. On one hand, the knowledge of the solvated transition structure is fundamental in order to carry out MD simulations of reaction trajectories using the rare event approach 34. On the other hand, the knowledge of the reaction path is a prerequisite to calculate the free energy profile of the process.…”

Hybrid Dynamics Simulation Engine for Metalloproteins

“…In years since the early works, QM/MM methods have undergone a remarkable advancement, making it possible to study reaction pathways in large systems, such as solvated enzymes (23,24). QM methods employed in QM/MM now range from semiempirical (25)(26)(27)(28)(29)(30)(31), to Hartree-Fock and post-Hartree-Fock methods (13,18,32,33), to density functional theory (DFT) (14,16,(34)(35)(36), as well as ab initio molecular dynamics (AIMD) (17,(37)(38)(39)(40) and Car-Parrinello MD (37,41). Car-Parrinello-MD-based algorithms are especially reliable, and the spectacular advances made in their development over recent years have improved both performance and parallelization efficiency (42,43).…”

Reaching multi‐nanosecond timescales in combined QM/MM molecular dynamics simulations through parallel horsetail sampling

“…In the context of QM/MM MD simulations, MMD sampling methods have not been widely used. We reported the first application of the REA technique to study the kinetics of the ethylene bromination process in water solution, a process that displays huge solvation effects. As the REA trajectories are independent, the calculations can be carried out in parallel and this was probably the first application of the QM/MM MD methodology to the study of reactive trajectories using parallel computing (see also).…”