NERSCPowering Scientific Discovery for 50 Years

NERSC Initiative for Scientific Exploration (NISE) 2011 Awards

MD Simulations of Liquid Water Films at the Boundary Between Gas Hydrate and Mineral Surfaces

Ian C. Bourg, Lawrence Berkeley National Lab

Associated NERSC Project: Clay Mineral Surface Geochemistry (mp47)
Principal Investigator: Garrison Sposito, Lawrence Berkeley National Lab

NISE Award: 300,000 Hours
Award Date: March 2011

Liquid water films at gas hydrate grain boundaries (between two hydrate grains or between a hydrate grain and a mineral surface) have been hypothesized to strongly influence the kinetics of dissolution and growth in gas hydrate formations. Therefore, by establishing the existence (or absence) of these films and their properties, our simulations will provide important information on the behavior of gas hydrates in ocean sediments (in particular, on the feasibility of sequestering CO2 as gas hydrates in ocean sediments.

The proposed research will investigate the existence and properties of liquid water films at the boundary between a clay mineral surface and a CO2 hydrate clathrate. Such liquid water films (known as “premelted” water films) are known to exist at boundaries between mineral surfaces and ice at temperatures lower than the freezing temperature of ice. Their existence in gas hydrate clathrates has been hypothesized but not yet demonstrated.

In order to determine whether stable liquid water films exist at boundaries between mineral surfaces and gas hydrates, we shall carry out two molecular dynamics (MD) simulations. Our first simulation will consist in placing a CO2 hydrate crystal in contact with a smectite clay mineral surface at a pressure characteristic of gas hydrate formations in ocean sediments (P > 4 MPa [3]) and at a temperature 10 K below the freezing temperature of the gas hydrate. Our second simulation will consist in placing a 6-nm-thick liquid water film (containing dissolved CO2 at the same H2O/CO2 ratio as the gas hydrate) between a CO2 hydrate crystal and a smectite clay mineral surface at the same pressure and temperature as our first simulation. Based on the known behavior of ice near glass and metal surfaces [1], we expect that the equilibrium thickness of the “premelted” liquid water film in these conditions should be on the order of 1.5 to 3 nm. If this is true, the gas hydrate crystal will melt near the clay mineral surface in our first simulation, but will grow towards the clay surface in our second simulation. Previous simulations of gas hydrate growth or dissolution indicate that simulation times of a few hundred nanoseconds will be sufficient to approach the equilibrium thickness of the liquid water film in our simulations.

The proposed research will use our tested MD simulation methodologies for describing water-clay and water-CO2 interactions. We also shall test water models that accurately describe the freezing temperature of liquid water but for which well-tested mineral-water interaction potentials have not yet been developed, such as the TIP5P or TIP4P-ICE water models.