Phase diagram of the NaCl–water system from computer simulations

Abstract

NaCl aqueous solutions are ubiquitous. They can crystallize into ice, NaCl, or NaCl · 2H2O depending on the temperature–concentration conditions. These crystallization transitions have important implications in geology, cryopreservation, or atmospheric science. Computer simulations can help understand the crystallization of these solids, which requires a detailed knowledge of the equilibrium phase diagram. We use molecular simulations in which we put at contact the solution with the solid of interest to determine points of the solid–solution coexistence lines. We follow two different approaches, one in which we narrow down the melting temperature for a given concentration and the other in which we equilibrate the concentration for a given temperature, obtaining consistent results. The phase diagram thus calculated for the selected model (TIP4P/2005 for water molecules and Joung–Cheatham for the ions) correctly predicts coexistence between the solution and ice. We were only able to determine NaCl · 2H2O–solution coexistence points at higher temperatures and concentrations than in the experiment, so we could not establish a direct comparison in this case. On the other hand, the model underestimates the concentration of the solution in equilibrium with the NaCl solid. Our results, alongside other literature evidence, seem to indicate that ion–ion interactions are too strong in the model. Our work is a good starting point for the improvement of the potential model and for the study of the nucleation kinetics of the solid phases involved in the phase diagram.

Freezing point depression of salt aqueous solutions using the Madrid-2019 model

“…Consistent with this, in previous work, we found that the coexistence between ice and a NaCl solution obtained with a 9 Å cut-off and LRC is the same (within the statistical uncertainty), as that obtained with a cutoff of 13 Å and switching off the LJ LRC. 90 Second, to make sure that our results are not significantly affected by finite size effects, we repeated the simulation by doubling the size of the ice slab and the solution along the z-axis for the NaCl system at T = 236 K. In this system, an ice slab with 4096 molecules and 3.6 × 3.1 × 11.7 nm 3 dimensions was put in contact with a solution containing 6660 water molecules so that the system dimensions are roughly 3.6 × 3.1 × 30.5 nm 3 . The results obtained for both system sizes are the same within the statistical uncertainty of the simulations.…”

Self-Consistent Implementation of a Solvation Free Energy Framework to Predict the Salt Solubilities of Six Alkali Halides

“…These hydration free energies, along with a series of osmotic, vapor pressures, , activity coefficients, densities, and diffusion coefficient properties, − were employed as the training set for estimating FF parameters over a broad concentration range. The solubility data were also replicated to verify the validity of FFs in two methods: the thermodynamic approach (TA) − and the direct coexistence approach. ,, Although good results and more interesting information for solution structures can be obtained from these simulations, two issues remain: (1) the relation between the solvation free energy and the traditional thermodynamic property of Gibbs formation free energy; (2) the selection of either Marcus’ or Tissandier’s data as the training set for the FF to represent the bulk solution properties.…”