Crystal lattice properties fully determine short-range interaction parameters for alkali and halide ions

• Published in: arXiv.org
• Main Authors: Mao, Albert H, Pappu, Rohit V
• Format: Paper Journal Article
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 17.08.2012
• Subjects: Aqueous solutions; Computational fluid dynamics; Computer simulation; Computing time; Crystal lattices; Dependence; Distribution functions; Hydration; Interaction parameters; Mathematical models; Mixing rules; Optimization; Parameterization; Periodic table; Physical simulation; Physics - Chemical Physics; Properties (attributes); Simulation; Solvents; Water distribution; Water engineering
• ISSN: 2331-8422
• DOI: 10.48550/arxiv.1208.3704

Accurate models of alkali and halide ions in aqueous solution are necessary for computer simulations of a broad variety of systems. Previous efforts to develop ion force fields have generally focused on reproducing experimental measurements of aqueous solution properties such as hydration free energies and ion-water distribution functions. This dependency limits transferability of the resulting parameters because of the variety and known limitations of water models. We present a solvent-independent approach to calibrating ion parameters based exclusively on crystal lattice properties. Our procedure relies on minimization of lattice sums to calculate lattice energies and interionic distances instead of equilibrium ensemble simulations of dense fluids. The gain in computational efficiency enables simultaneous optimization of all parameters for Li+, Na+, K+, Rb+, Cs+, F-, Cl-, Br-, and I- subject to constraints that enforce consistency with periodic table trends. We demonstrate the method by presenting lattice-derived parameters for the primitive model and the Lennard-Jones model with Lorentz-Berthelot mixing rules. The resulting parameters successfully reproduce the lattice properties used to derive them and are free from the influence of any water model. To assess the transferability of the Lennard-Jones parameters to aqueous systems, we used them to estimate hydration free energies and found that the results were in quantitative agreement with experimentally measured values. These lattice-derived parameters are applicable in simulations where coupling of ion parameters to a particular solvent model is undesirable. The simplicity and low computational demands of the calibration procedure make it suitable for parametrization of crystallizable ions in a variety of force fields.