How important is the dielectric constant in water modeling? Evaluation of the performance of the TIP4P/$\varepsilon$ force field and its compatibility with the Joung-Cheatham NaCl model

Abstract

Efficient large-scale computer simulations of aqueous solutions require the use of accurate but simple empirical force fields for water. However, the complexity of these systems evidences the difficulties in describing solution properties without due account of polarization. Different strategies to remedy this problem are parametrizing water force fields to the dielectric constant or charge scaling of solvated ions. In this work, we compare results from TIP4P/$\varepsilon$ and OPC models, which are parametrized to predict the dielectric constant, with results from TIP4P/2005, which is closer in spirit to the charge scaling strategy. The performance of the models is rated according to the Vega-Abascal benchmark. Our results show that TIP4P/$\varepsilon$ and TIP4P/2005 perform equally well, with the OPC model lying significantly behind. TIP4P/$\varepsilon$ can predict bulk phase properties (transport properties, thermal expansion coefficients, densities) of both liquid water and ice polymorphs, but also surface tensions, with an accuracy very similar to TIP4P/2005, while performing very well for dielectric constants over a wide range of pressures and temperatures. On the other hand, TIP4P/2005 provides a better description of phase boundaries, including liquid-vapor and freezing transitions. However, the accurate prediction of dielectric constants allows TIP4P/$\varepsilon$ to describe densities of NaCl solutions for models parametrized to their crystal and melt properties only. This is achieved without the need to rescale charges, modify the Lorentz-Berthelot rule or tune the ion's Lennard-Jones parameters. Our findings hinge on the significance of dielectric constants as a target property and show that a robust parametrization can be achieved without invoking the concept of charge scaling.