“Digital Fluid Physics”: Prediction of phase equilibria for several mixtures of CO2 with petroleum fluid systems

Abstract

Enhanced oil recovery (EOR) techniques (miscible/near miscible hydrocarbon/non-hydrocarbon gas injection, chemical flooding such as surfactant/polymer and smart water, and combinations of the three) aim at increasing oil recovery factor over what could be achieved from natural depletion and pressure maintenance methods such as waterflooding. As we embark on more complex EOR processes (e.g.), complex phase equilibria, complicated rock/fluid interaction, and sophisticated transport through porous media will create several key technical challenges that must be addressed.Currently, all of the existing compositional reservoir simulators, such as ECLIPSE, CMG, NEXUS, etc., use semi-empirical cubic equation of state (EoS) to represent the phase equilibria (mainly for the hydrocarbon calculation of equilibrium phase compositions) and phase properties (densities, viscosities, volume fractions, saturations, etc.). These data in combination with rock/fluid interaction data (relative permeability and capillarity pressure) will describe transport through porous media under the assumption of instantaneous equilibrium. Many empirical binary interaction parameters are needed for cubic EoS. In addition, mixing parameters along with the proper mixing rules are also required for multi-component mixtures. Current cubic EOS models have limited extrapolation capabilities beyond measured data. They also fall short in predicting derivative properties (even within the range measurements), transport properties, critical behavior, and the surfactants and polar components' phase behavior.This work presents the application of “Digital Fluid Physics” to determine the phase behavior of petroleum mixtures. Three different petroleum systems are investigated. Molecular simulation shows promising results in comparison with experimental measurements and therefore can reduce the cost and time required for tuning equations of state.