Calculation of the Gibbs free energy of solvation and dissociation of HCl in water via Monte Carlo simulations and continuum solvation models.

The Gibbs free energy of solvation and dissociation of hydrogen chloride in water is calculated through a combined molecular simulation/quantum chemical approach at four temperatures between T = 300 and 450 K. The Gibbs free energy is first decomposed into the sum of two components: the Gibbs free energy of transfer of molecular HCl from the vapor to the aqueous liquid phase and the standard-state Gibbs free energy of acid dissociation of HCl in aqueous solution. The former quantity is calculated using Gibbs ensemble Monte Carlo simulations using either Kohn-Sham density functional theory or a molecular mechanics force field to determine the system's potential energy. The latter Gibbs free energy contribution is computed using a continuum solvation model utilizing either experimental reference data or micro-solvated clusters. The predicted combined solvation and dissociation Gibbs free energies agree very well with available experimental data.

Vapor-liquid coexistence curves for methanol and methane using dispersion-corrected density functional theory.

First principles Monte Carlo simulations in the Gibbs and isobaric-isothermal ensembles were performed to map the vapor-liquid coexistence curves of methanol and methane described by Kohn-Sham density functional theory using the Becke-Lee-Yang-Parr (BLYP) exchange and correlation functionals with the Grimme correction term for dispersive (D2) interactions. The simulations indicate that the BLYP-D2 description with the TZV2P basis set underpredicts the saturated vapor densities and overpredicts the saturated liquid densities and critical and boiling temperatures for both compounds. Although the deviations are quite large, these results present a significant improvement over the BLYP functional without the correction term, which misses the experimental results by a larger extent in the opposite direction. Simulations at one temperature indicate that use of the larger QZV3P basis set may lead to improved saturated vapor densities, but not to significant changes in the liquid density.

Vapor-liquid nucleation of argon: exploration of various intermolecular potentials.

The homogeneous vapor-liquid nucleation of argon has been explored at T=70 and 90 K using classical nucleation theory, semiempirical density functional theory, and Monte Carlo simulations using the aggregation-volume-bias algorithm with umbrella sampling and histogram-reweighting. In contrast with previous simulation studies, which employed only the Lennard-Jones intermolecular potential, the current studies were carried out using various pair potentials including the Lennard-Jones potential, a modified Buckingham exponential-six potential, the Barker-Fisher-Watts pair potential, and a recent ab initio potential developed using the method of effective diameters. It was found that the differences in the free energy of formation of the critical nuclei between the potentials cannot be explained solely in terms of the difference in macroscopic properties of the potentials, which gives a possible reason for the failure of classical nucleation theory.

First principles Monte Carlo simulations of aggregation in the vapor phase of hydrogen fluoride.

The aggregation of hydrogen fluoride vapor is explored through the use of Monte Carlo simulations employing Kohn-Sham density functional theory with the exchange/correlation functional of Becke-Lee-Yang-Parr to describe the molecular interactions. Canonical ensemble simulations sampling the classical phase space were carried out for a system consisting of ten molecules at constant density (2700 A(3)/molecule) and at three different temperatures (T = 310, 350, and 390 K). Aggregation-volume-bias and configurational-bias Monte Carlo approaches (along with pre-sampling with an approximate potential) were employed to increase the sampling efficiency of cluster formation and destruction. A hydrogen-bond analysis shows that about two thirds of the HF molecules are part of small aggregates at 310 K, whereas only about 10% of the molecules are clustered at 390 K. As for other hydrogen-bonding systems, the size distribution exhibits some sensitivity to the criteria used to define a hydrogen bond, but the qualitative features are not affected by these differences. From the temperature dependence of the equilibrium constants, the dimer and trimer aggregation energies (not corrected for nuclear quantum effects) are estimated using a simple distance-based hydrogen-bonding criterion as -13 +/- 3 and -65 +/- 16 kJ mol(-1), respectively, whereas these binding energies are found to be somewhat different for a combined distance-angular criterion with values of -17 +/- 6 and -63 +/- 11 kJ mol(-1), respectively. The strictness of the hydrogen-bonding criterion plays a significant role for the assignment of clusters to linear, cyclic, and branched architectures with the fraction of the latter being drastically reduced for the distance-angular criterion. The average molecular dipole moment increases from 1.85 Debye for isolated molecules to about 2.0 D for dimers to about 2.75 D for larger aggregates, and the H-F bond length shows a concomitant, but smaller increase from about 0.94 to 0.98 A.