Molecular packing and chemical association in liquid water simulated using ab initio hybrid Monte Carlo and different exchange-correlation functionals

TL;DR

This study investigates how different exchange-correlation functionals affect the molecular packing and chemical association in liquid water through ab initio hybrid Monte Carlo simulations, revealing variations in hydration free energy contributions and system heterogeneities.

Contribution

It provides a comparative analysis of chemical and packing contributions in water simulated with various density functionals at fixed density, highlighting differences in molecular behavior and finite size effects.

Findings

Water with BLYP at 300 K is more tightly bound than with revPBE or at higher temperatures.

Packing contributions scale with volume or surface area depending on the functional and temperature.

Heterogeneities are observed in BLYP 300 K simulations, affecting packing behavior.

Abstract

In the free energy of hydration of a solute, the chemical contribution is given by the free energy required to expel water molecules from the coordination sphere and the packing contribution is given by the free energy required to create the solute-free coordination sphere (the observation volume) in bulk water. With the SPC/E water model as a reference, we examine the chemical and packing contributions in the free energy of water simulated using different electron density functionals. The density is fixed at a value corresponding to that for SPC/E water at a pressure of 1 bar. The chemical contribution shows that water simulated at 300 K with BLYP is somewhat more tightly bound than water simulated at 300 K with the revPBE functional or at 350 K with the BLYP and BLYP-D functionals. The packing contribution for various radii of the observation volume is studied. In the size range where the distribution of water molecules in the observation volume is expected to be Gaussian, the packing contribution is expected to scale with the volume of the observation sphere. Water simulated at 300 K with the revPBE and at 350 K with BLYP-D or BLYP conforms to this expectation, but the results suggest an earlier onset of system size effects in the BLYP 350 K and revPBE 300 K systems than that observed for either BLYP-D 350 K or SPC/E. The implication of this observation for constant pressure simulations is indicated. For water simulated at 300 K with BLYP, in the size range where Gaussian distribution of occupation is expected, we instead find non-Gaussian behavior, and the packing contribution scales with surface area of the observation volume, suggesting the presence of heterogeneities in the system.