Global optimization strategies for molecules encapsulated in water clusters

Abstract

A discrete solvent method has been developed for a cluster consisting of a finite number of solute and solvent molecules that is combined with a Basin-Hopping (BH) Monte Carlo algorithm to search for the lowest energy or global minimum structure for the solute-solvent system. The BH algorithm is written in the Python programming language and runs in tandem with the General Atomic and Molecular Electronic Structure System (GAMESS) Quantum Chemistry software. Cycles of the BH algorithm was run using a computationally fast Density Functional Theory (DFT) based effective fragment potential (EFP1) method for the water molecules, and then the final lower energy structures were refined with a more accurate ab initio DFT calculation on the complete solute/water cluster. The BH algorithm is applied to the methylammonium ion CH3NH3+(H2O)1-6, zwitterionic methylcarbamic acid CH3NH2+CO2-(H2O)1-8, and zwitterionic glycine NH3+CH2CO2-(H2O)7-8 clusters.