Despite considerable efforts over more than two decades, our knowledge of the interactions in electrolyte solutions is not yet satisfactory. Not even one of the most simple and important aqueous solutions, NaCl(aq), escapes this assertion. A requisite for the development of a force field for any water solution is the availability of a good model for water. Despite the fact that TIP4P/2005 seems to fulfill the requirement, little work has been devoted to build a force field based on TIP4P/2005. In this work, we try to fill this gap for NaCl(aq). After unsuccessful attempts to produce accurate predictions for a wide range of properties using unity ionic charges, we decided to follow recent suggestions indicating that the charges should be scaled in the ionic solution. In this way, we have been able to develop a satisfactory non-polarizable force field for NaCl(aq). We evaluate a number of thermodynamic properties of the solution (equation of state, maximum in density, enthalpies of solution, activity coefficients, radial distribution functions, solubility, surface tension, diffusion coefficients, and viscosity). Overall the results for the solution are very good. An important achievement of our model is that it also accounts for the dynamical properties of the solution, a test for which the force fields so far proposed failed. The same is true for the solubility and for the maximum in density where the model describes the experimental results almost quantitatively. The price to pay is that the model is not so good at describing NaCl in the solid phase, although the results for several properties (density and melting temperature) are still acceptable. We conclude that the scaling of the charges improves the overall description of NaCl aqueous solutions when the polarization is not included.

The solubility of NaCl in water is evaluated by using three force field models: Joung-Cheatham for NaCl dissolved in two different water models (SPC/E and TIP4P/2005) and Smith Dang NaCl model in SPC/E water. The methodology based on free-energy calculations [E. Sanz and C. Vega, J. Chem. Phys. 126, 014507 (2007)] and [J. L. Aragones et al., J. Chem. Phys. 136, 244508 (2012)] has been used, except, that all calculations for the NaCl in solution were obtained by using molecular dynamics simulations with the GROMACS package instead of homemade MC programs. We have explored new lower molalities and made longer runs to improve the accuracy of the calculations. Exploring the low molality region allowed us to obtain an analytical expression for the chemical potential of the ions in solution as a function of molality valid for a wider range of molalities, including the infinite dilute case. These new results are in better agreement with recent estimations of the solubility obtained with other methodologies. Besides, two empirical simple rules have been obtained to have a rough estimate of the solubility of a certain model, by analyzing the ionic pairs formation as a function of molality and/or by calculating the difference between the NaCl solid chemical potential and the standard chemical potential of the salt in solution.

The vapor-liquid phase behavior and the critical behavior of the square-well (SW) fluid are investigated as a function of the interaction range, lambdain [1.25, 3], by means of the self-consistent Ornstein-Zernike approximation (SCOZA) and analytical equations of state based on a perturbation theory [A. L. Benavides and F. del Rio, Mol. Phys. 68, 983 (1989); A. Gil-Villegas, F. del Rio, and A. L. Benavides, Fluid Phase Equilib. 119, 97 (1996)]. For this purpose the SCOZA, which has been restricted up to now to a few model systems, has been generalized to hard-core systems with arbitrary interaction potentials requiring a fully numerical solution of an integro-partial differential equation. Both approaches, in general, describe well the liquid-vapor phase diagram of the square-well fluid when compared with simulation data. SCOZA yields very precise predictions for the coexistence curves in the case of long ranged SW interaction (lambda>1.5), and the perturbation theory is able to predict the binodal curves and the saturated pressures, for all interaction ranges considered if one stays away from the critical region. In all cases, the SCOZA gives very good predictions for the critical temperatures and the critical pressures, while the perturbation theory approach tends to slightly overestimate these quantities. Furthermore, we propose analytical expressions for the critical temperatures and pressures as a function of the square-well range.

The fluid phase behavior of colloidal suspensions with short-range attractive interactions is studied by means of Monte Carlo computer simulations and two theoretical approximations, namely, the discrete perturbation theory and the so-called self-consistent Ornstein-Zernike approximation. The suspensions are modeled as hard-core attractive Yukawa (HCAY) and Asakura-Oosawa (AO) fluids. A detailed comparison of the liquid-vapor phase diagrams obtained through different routes is presented. We confirm Noro-Frenkel's extended law of scaling according to which the properties of a short-ranged fluid at a given temperature and density are independent of the detailed form of the interaction, but just depend on the value of the second virial coefficient. By mapping the HCAY and AO fluids onto an equivalent square-well fluid of appropriate range at the critical point we show that the critical temperature as a function of the effective range is independent of the interaction potential, i.e., all curves fall in a master curve. Our findings are corroborated with recent experimental data for lysozyme proteins.

No abstract