Molecular-dynamics simulations of methane hydrate dissociation

• Published in: The Journal of Chemical Physics
• Main Authors: English, Niall J., Johnson, J. K., Taylor, Charles E.
• Format: Article in Journal/Newspaper
• Language: English
• Published: AIP Publishing 2005
• Subjects: Methane hydrate
• Online Access: http://dx.doi.org/10.1063/1.2138697; https://pubs.aip.org/aip/jcp/article-pdf/doi/10.1063/1.2138697/13513759/244503_1_online.pdf

Nonequilibrium molecular-dynamics simulations have been carried out at 276.65K and 68bar for the dissolution of spherical methane hydrate crystallites surrounded by a liquid phase. The liquid was composed of pure water or a water-methane mixture ranging in methane composition from 50% to 100% of the corresponding theoretical maximum for the hydrate and ranged in size from about 1600 to 2200 water molecules. Four different crystallites ranging in size from 115 to 230 water molecules were used in the two-phase systems; the nanocrystals were either empty or had a methane occupation from 80% to 100% of the theoretical maximum. The crystal-liquid systems were prepared in two distinct ways, involving constrained melting of a bulk hydrate system or implantation of the crystallite into a separate liquid phase. The breakup rates were very similar for the four different crystal sizes investigated. The method of system preparation was not found to affect the eventual dissociation rates, despite a lag time of approximately 70ps associated with relaxation of the liquid interfacial layer in the constrained melting approach. The dissolution rates were not affected substantially by methane occupation of the hydrate phase in the 80%–100% range. In contrast, empty hydrate clusters were found to break up significantly more quickly. Our simulations indicate that the diffusion of methane molecules to the surrounding liquid layer from the crystal surface appears to be the rate-controlling step in hydrate breakup. Increasing the size of the liquid phase was found to reduce the initial delay in breakup. We have compared breakup rates computed using different long-range electrostatic methods. Use of the Ewald, minimum image, and spherical cut-off techniques led to more rapid dissociation relative to the Lekner method.