Unlike interactions in simulated methane clathrate hydrates.

Abstract

Clathrate hydrates are an ice-like substance consisting of networks of water molecules, held together by hydrogen bonds, enclosing trapped gas molecules. Natural gas clathrate hydrates (in which the trapped gas species is chiefly methane) are of interest in the field of offshore gas exploitation, where they frequently form blockages in natural gas pipelines. Knowledge of the phase equilibria of methane clathrate hydrate can thus reduce the overall monetary cost of natural gas extraction. Computer simulation of molecular systems is useful to understand fundamental mechanisms, and serves as a complementary method to laboratory experiments in the study of chemical systems. The Lennard-Jones potential is frequently used to describe intermolecular interactions in molecular simulations. Correction factors are often applied to the Lennard-Jones potential, although the effect of these correction factors on the behaviour of simulated molecular systems is not fully understood. This thesis examines the effect of Lennard-Jones correction factors on simulated methane clathrate hydrates using three different computational approaches: lattice distortion theory, grand canonical Monte Carlo simulations (which emulate gas adsorption into the clathrate lattice), and direct estimation of the heat of dissociation coupled with the Clausius-Clapeyron equation. In addition, the use of the results of grand canonical Monte Carlo simulations to infer phase equilibria was demonstrated in this thesis. The application of Lennard-Jones correction factors in lattice distortion calculations was found to not be viable, due to the extreme sensitivity of the perturbation potential (the quantity of interest in this theory) to changes in the values of the correction factors. Unlike interactions were found to weakly influence methane adsorption into the clathrate hydrate crystal, and so the application of correction factors in grand canonical Monte Carlo simulations is demonstrated to be ineffectual. The direct estimation of the heat of dissociation was shown to be viable when matching to calorimetric data, and the inference of phase equilibria by coupling the Clausius-Clapeyron equation with this approach was shown to yield agreeable results.