Coarse grained force field for the molecular simulation of natural gases and condensates

The atomistically-detailed molecular modelling of petroleum fluids is challenging, amongst other aspects, due to the very diverse multicomponent and asymmetric nature of the mixtures in question. Complicating matters further, the time scales for many important processes can be much larger than the current and foreseeable capacity of modern computers running fully-atomistic models. To overcome these limitations, a coarse grained (CG) model is proposed where some of the less-important degrees of freedom are safely integrated out, leaving as key parameters the average energy levels, the molecular conformations and the range of the Mie intermolecular potentials employed as the basis of the model. The parametrization is performed by using an analytical equation of state of the statistical associating fluid theory (SAFT) family to link the potential parameters to macroscopically observed thermophysical properties. The parameters found through this top-down approach are used directly in molecular dynamics simulations of multi-component multi-phase systems. The procedure is exemplified by calculating the phase envelope of the methane-decane binary and of two synthetic light condensate mixtures. A methodology based on the discrete expansion of a mixture is used to determine the bubble points of these latter mixtures, with an excellent agreement to experimental data. The model presented is entirely predictive and an abridged table of parameters for some fluids of interest is provided.