A Four-Site Molecular Model for Simulations of Liquid Methanol and Water-Methanol Mixtures: MeOH-4P

In this work, we present a new four-site potential for methanol, MeOH-4P, fitted to reproduce the dielectric constant epsilon, the surface tension gamma(s), and the liquid density rho of the pure liquid at T = 298.15 K and p = 1 bar. The partial charges on each site were taken from the OPLS/2016 model with the only difference of putting the negative charge on the fourth site (M) instead of on the 0 atom, as done in four-site water models. The original Lennard-Jones (LJ) parameters of OPLS/2016 for the methyl moiety (Me) were modified for the fitting of rho and gamma(s), whereas the parameters of the TIP4P-FB water model were used for the 0 atom without change. Taking into account the energetic cost of the enhanced dipole relative to the isolated molecule, the results from simulations with this model showed good agreement with experiments for rho, alpha(p), kappa(T), C-p, and Delta Hv-i. Also, the temperature dependence of gamma(s) and epsilon is satisfactory in the interval between 260 and 360 K, and the critical point description is similar to that of OPLS/2016. It is shown that orientational correlations, described by the Kirkwood factor G(k), play a prominent role in the appropriate description of dielectric constants in existing models; unfortunately, the enhancement of the dipole moment produced a low diffusion coefficient D-MeOH; thus, a compromise was required between a good reproduction of e and an acceptable D-MeOH. The use of a fourth site resulted in a significant improvement for water methanol mixtures described with TIP4P-FB and MeOH-4P, respectively, but required the modification of the LJ geometric combination rule to allow a good description of the methanol molar-fraction dependence of rho, epsilon, and methanol (water) diffusion coefficients D-MeOH (D-H2O) and excess volume of mixing Delta V-mix in the entire range of composition. The resulting free energy of hydration Delta G(hyd) shows excellent agreement with experiments in the interval between 280 and 360 K.