One of the decarbonization initiatives, designed for various industrial sectors that produce greenhouse gases, is based on the use of green hydrogen. Several proposals are already underway to establish a new infrastructure model, consisting of Hubs dedicated to the production, transportation and storage of massive amounts of H2. Depleted gas reservoirs have large scalable storage capacity and are being tested for H2 storage. Fluid modeling in reservoir simulators allows the use of equations of state (Peng-Robinson or SRK) to predict various equilibrium properties and phenomenological correlations for transport properties. These equations perform relatively well in predicting properties for pure hydrocarbons; however, for mixtures, especially involving other classes of molecules, the deviations increase considerably. Accurate estimation of the thermodynamic and transport properties of H2 mixed with other gases found within the storage system is therefore essential for efficient reservoir operation design. In this study, instead of using empirical correlations, we will employ a unified solution based on molecular simulation models to predict the compressibility factor Z and viscosity simultaneously with the same force field (Darkrim and TraPPE). The NPT ensemble with a simulation box with approximately 300 molecules was used to predict the compressibility factor Z of pure components (H2, C1, C2, C3, N2 and CO2) and mixtures (C1-H2, C2-H2, C3-H2, N2-H2, CO2-H2) at different compositions and temperatures. The same force field was extended to predict viscosity using molecular dynamics (Green-Kubo) for the pure components and mixtures. The excellent agreement obtained between the simulated and experimental data eliminates the need to obtain new experimental curves with H2 and other gases, which can be used as cushion gas in the storage system, with evident savings in time and costs. Furthermore, we reparametrize the binary components of the Peng-Robinson equation to allow its use in modeling fluids in the reservoir simulator. The impacts of deviations from the Peng-Robinson equation are also demonstrated.
