(508c) The Quantum Mechanics Based Polarizable Force Field for Simulations of Complex Materials: Application to Water System

Naserifar, S., California Institute of Technology
Goddard, W. A. III, California Institute of Technology
The use of molecular simulations is critical to the development and manufacturing of new generations of materials. However, the accuracy of the developed potentials significantly depends on how the interatomic (i.e. non-bond) interactions are defined. A typical method has been using one or several empirical data to optimize and validate the interatomic potentials for specific systems. The problem with this method is that the obtained force fields are not usually transferable to other systems and fail to reproduce the properties of combined systems. For example, the most popular empirical water models (TIPnP and SPC families) are fitted to reproduce one or more macroscopic properties of water but none reproduce all important properties of the solid and liquid phase.

Recent dramatic developments in quantum mechanics (QM) methods has enabled accurate description of the interatomic interactions via dispersive forces (e.g. Grimme DFT-D). But the QM methods are limited to ~200 atoms and time scales of ~10 picoseconds which is impractical for real applications that require spatial scales of 100 nanometers and beyond (>100 million atoms) and time scales of microseconds and beyond. Therefore, to obtain a force field that could correctly describe the standard properties of a system and be transferable to new materials, we decided to base it fully on best available QM calculations.

Here, we provide an example of a new force field for water based solely on QM calculations with no empirical data. The QM was at high lever CCSD(T) level for all orientations and distances for water dimer plus X3LYP DFT on 19 larger water clusters. This model provides quite excellent agreement with experimental data for solid and liquid phase of water: Tmelt=273.3K (exp=273.15K) and properties at 298K: ΔHvap=10.36 kcal/mol, density = 0.997 g/cc, entropy= 68.4 eu, dielectric constant=76.1, ln Ds (self-diffusion coef) =-10.08 compared to experimental values of 10.52, 0.997, 69.9, 78.4, and -11.24, respectively. We expect this force field to remain accurate as a function of temperature and pressure. We have used this force field to study the properties of water at the surface including surface thickness, water orientation, hydrogen bond distribution, and vibrational frequencies which are experimentally hard to obtain. In addition, we have discovered for the first time the existence of six-coordinated water molecules at the areas close to the surface which are stable over 10 ps time intervals. They could be responsible for some of the complicated water properties of water at the surface.

Being based solely on QM we expect that we can use QM to extend it to ions, proteins, DNA, polymers, and inorganic systems for applications to biomolecular, pharma, electrocatalysis (fuel cells, water splitting) and batteries where interactions with explicit water molecules plays a significant role.