(726c) ­­­­­Development of New Coarse-Grained Water Models Using Particle Swarm Optimization

Research on understanding the structure and dynamics of water molecules in various processes is a subject of considerable interest in the fields ranging from biology, chemistry, to chemical engineering. Coarse-grained (CG) Molecular Dynamics (MD) simulations, conducted with an accurate model, is a powerful method to study the structure and dynamics of a large system up to micro- to millisecond timescales. In the present study, we have developed polarizable and non-polarizable water models using 2:1 mapping scheme (two water molecules are represented by a single interaction center). The interaction parameters between the CG water beads were developed using a systematic two step approach: 1. Global search was conducted using particle swarm optimization (PSO) method and 2. A local minimum was identified using modified gradient descent algorithm. The new CG models could reproduce the experimentally reported thermodynamic, physical, and chemical properties of water with accuracy. These new models are further utilized to study the phase transitions in stimuli-responsive polymers.