M. J. Torrejón, S. Blázquez, J. Algaba, Mª M. Conde, F. J. Blas
https://doi.org/10.1021/acs.energyfuels.5c01012
In this work, we determine the dissociation temperature of hydrogen (H2) hydrate by computer simulation using two different methods. In both cases, the molecules of water and H2 are modeled using the TIP4P/Ice and a modified version of the Silvera and Goldman models, respectively, and the Berthelot combining rule for the cross water–H2 interactions has been modified. The first method used in this work is the solubility method, which consists of determining the solubility of H2 in an aqueous phase when in contact with the H2 hydrate (H–Lw) phase and when in contact with the pure H2 phase (Lw–LH2) at different temperatures. At a given pressure value, both solubility curves intersect at the temperature (T3) at which the three phases coexist in equilibrium. Following this approach, we determine the dissociation temperature of H2 hydrate at 185 MPa finding a good agreement with the data previously reported in the literature. We also analyze the effect of the multiple occupancy of the D, or small, and H, or large, cages of the sII hydrate structure. We conclude that the T3 value is barely affected by the occupancy of the H2 hydrate at 185 MPa. From the analysis of the solubility curves and performing extra bulk simulations of the three phases involved in the equilibrium, we also determine the driving force for nucleation (ΔμNEC) at 185 MPa as a function of the supercooling degree and the H2 hydrate occupancy. We determine that, thermodynamically, the most favored occupancy of the H2 hydrate consists of 1 H2 molecule in the D cages and 3 in the H cages (i.e., 1–3 occupancy). We also conclude that the double occupancy of the small D cages is not favored because the ΔμNEC values obtained for this occupancy are the most positive ones. The second approach used in this work is the direct coexistence technique using an initial H2 hydrate phase with 1–3 occupancy. We also propose a new modification of the Berthelot combining rule to improve the predictions of the T3 values. Following this method, we determine the T3 at 100, 185, and 300 MPa finding excellent agreement with the experimental data.
