RESUMO
Molecular dynamics simulations were employed to study the structure of molecularly thin films of antiagglomerants adsorbed at the interface between sII methane hydrates and a liquid hydrocarbon. The liquid hydrocarbon was composed of dissolved methane and higher-molecular-weight alkane such as n-hexane, n-octane, and n-dodecane. The antiagglomerants considered were surface-active compounds with three hydrophobic tails and a complex hydrophilic head that contains both amide and tertiary ammonium cation groups. The length of the hydrophobic tails and the surface density of the compounds were changed systematically. The results were analyzed in terms of the preferential orientation of the antiagglomerants, density distributions of various molecular compounds, and other molecular-level properties. At low surface densities, the hydrophobic tails do not show preferred orientation, irrespectively of the tail length. At sufficiently high surface densities, our simulations show pronounced differences in the structure of the interfacial film depending on the molecular features and on the type of hydrocarbons present in the system. Some antiagglomerants are found to pack densely at the interface and exclude methane from the interfacial region. Under these conditions, the antiagglomerant film resembles a frozen interface. The hydrophobic tails of the antiagglomerants that show this feature has a length comparable to that of the n-dodecane in the liquid phase. It is possible that the structured interfacial layer is in part responsible for determining the performance of antiagglomerants in flow-assurance applications. The simulation results are compared against experimental data obtained with the rocking cell apparatus. It was found that the antiagglomerants for which our simulations suggest evidence of a frozen interface at sufficiently high surface densities are those that show better performance in rocking cell experiments.
RESUMO
Steered and equilibrium molecular dynamics simulations were employed to study the coalescence of a sI hydrate particle and a water droplet within a hydrocarbon mixture. The size of both the hydrate particle and the water droplet is comparable to that of the aqueous core in reverse micelles. The simulations were repeated in the presence of various quaternary ammonium chloride surfactants. We investigated the effects due to different groups on the quaternary head group (e.g. methyl vs. butyl groups), as well as different hydrophobic tail lengths (e.g. n-hexadecyl vs. n-dodecyl tails) on the surfactants' ability to prevent coalescence. Visual inspection of sequences of simulation snapshots indicates that when the water droplet is not covered by surfactants it is more likely to approach the hydrate particle, penetrate the protective surfactant film, reach the hydrate surface, and coalesce with the hydrate than when surfactants are present on both surfaces. Force-distance profiles obtained from steered molecular dynamics simulations and free energy profiles obtained from umbrella sampling suggest that surfactants with butyl tripods on the quaternary head group and hydrophobic tails with size similar to the solvent molecules can act as effective anti-agglomerants. These results qualitatively agree with macroscopic experimental observations. The simulation results provide additional insights, which could be useful in flow assurance applications: the butyl tripod provides adhesion between surfactants and hydrates; when the length of the surfactant tail is compatible with that of the hydrocarbon in the liquid phase a protective film can form on the hydrate; however, once a molecularly thin chain of water molecules forms through the anti-agglomerant film, connecting the water droplet and the hydrate, water flows to the hydrate and coalescence is inevitable.