Vapor-liquid equilibrium (VLE) prediction for dimethyl ether (DME) and water system in DME injection process with Peng-Robinson equation of state and composition dependent binary interaction coefficient

Dimethyl ether (DME) is arousing attention as a novel water-soluble additive in oil industry. The accurate prediction of phase behavior of DME and water system is crucial important for DME-based research. An equation of state has been a powerful tool for solving phase equilibrium problem, which is based on an ideal molecular force model with binary interaction coefficients reflecting the deviation of a non-ideal system. However, the current knowledge of interaction coefficient is limited by obtaining from backward by using phase equilibrium data. This leads to the unclear meaning of the interaction coefficients, insufficient response to the molecular force interaction relationship at the molecular level and no ability to predict phase equilibrium properties without primary use of experimental data. To overcome this, in this study a correction by combining an activity coefficient and a linear solvation energy relationship (LSER) has been proposed to address the solubility deviation from an idealized model. This deviation at the micro level indicates the deviation of nonideal molecular forces from the ideal Raoult's law. Based on this novel correction, binary interaction coefficients have been obtained with a more practical molecular interaction relationship, which leads to more consistent results with experimental data from the literature. Moreover, in asymmetric phase systems, concentration-dependent binary interaction coefficients have been proposed based on previous discussion to represent intermolecular forces in non-ideal phase systems. Furthermore, partitioning coefficients, which lead to a potential criterion for solvent selection for heavy oil recovery, have been discussed from the perspective of intermolecular forces. This study provides new insights into non-ideal phase behavior. By combining a LSER and activity coefficients, a quantitative analysis on non-ideal molecular behavior and related complex phase problems becomes possible.