Studies of the thermodynamic stability of clathrate hydrates of natural gas (mostly methane) is important in fields such as offshore gas exploitation and energy storage. Two approaches were used to study the effect of unlike dispersion interactions on methane clathrate hydrates: grand canonical Monte Carlo simulations (which yield adsorption data directly and can be used to infer phase equilibria), and estimation of the heat of dissociation coupled with the Clausius–Clapeyron equation (to calculate the phase equilibria, at the expense of providing no information about the adsorption behavior). It was found that the adsorption isotherm parameters change monotonically with respect to unlike dispersion interactions, although a perfect fit to experimentally-derived values may not be possible, at least using the force fields considered in this study. The heat of dissociation changes monotonically due to changes in the unlike dispersion interaction, and a best fit value of the Berthelot correction factor is achieved.

The determination of conditions at which clathrate hydrates are thermodynamically stable is important in applications such as offshore gas exploitation and energy storage. Adsorbed gas molecules occupy different cavity types within the hydrate lattice and this plays a significant role in the thermodynamic stability of clathrate hydrates. The occupancy of cavities in the hydrate lattice can be studied by undertaking Grand Canonical Monte Carlo simulations. Such simulations were performed in this study for methane clathrate hydrate with several force fields. Langmuir-type adsorption isotherms were fitted to the results of the simulations. The use of a single type of adsorption site was validated for methane clathrate hydrate. The adsorption isotherms which were fitted to the results of the simulations were used to compute the clathrate hydrate phase equilibria, which compared favourably with results from the literature.

The aim of this study was to determine the capability and accuracy of Monte Carlo simulations to predict ternary vapor-liquid-liquid equilibrium (VLLE) for some industrial systems. Hence, Gibbs ensemble Monte Carlo simulations in the isobaric-isothermal (NpT) and isochoric-isothermal (NVT) ensembles were performed to determine vapor-liquid-liquid equilibrium state points for three ternary petrochemical mixtures: methane/n-heptane/water (1), n-butane/1-butene/water (2) and n-hexane/ethanol/water (3). Since mixture (1) exhibits a high degree of mutual insolubility amongst its components, and hence has a large three-phase composition region, simulations in the NpT ensemble were successful in yielding three distinct and stable phases at equilibrium. The results were in very good agreement with experimental data at 120kPa, but minor deviations were observed at 2000 kPa. Obtaining three phases for mixture (2) with the NpT ensemble is very difficult since it has an extremely narrow three-phase region at equilibrium, and hence the NVT ensemble was used to simulate this mixture. The simulated results were, once again, in excellent agreement with experimental data. We succeeded in obtaining three-phase equilibrium in the NpT ensemble only after knowing, a priori, the correct three-phase pressure (corresponding to the force fields that were implemented) from NVT simulations. The success of the NVT simulation, compared to NpT, is due to the fact that the total volume can spontaneously partition itself favorably amongst the three boxes and only one intensive variable (T) is fixed, whereas the pressure and the temperature are fixed in an NpT simulation. NpT simulations yielded three distinct phases for mixture (3), but quantitative agreement with experimental data was obtained at very low ethanol concentrations only.