Development of the Wilson Functional Activity Coefficient Model Using High-Quality and Critically Evaluated Phase Equilibria Data

The activity coefficient is a key thermodynamic parameter, particularly in phase equilibria calculations. Different activity coefficient models have been developed to date, including empirical, correlative, predictive, and pure predictive. In this study, a predictive group contribution activity coefficient model based on the Wilson model was developed, which is called the Wilson Functional Activity Coefficient (Wilson-FAC). High-quality and critically evaluated vapor–liquid phase equilibria (VLE) data was used in the optimization procedure to obtain the interaction parameters between the main groups. Different thermodynamic consistency tests, including the Herington test, Van Ness test, point test, infinite dilution test, EOS test, and end point test, were applied to check the quality of the used VLE data. The interaction parameters between the 50 main groups involved in the 137 molecules were optimized using the fminsearch embedded function in MATLAB software. The developed Wilson-FAC model covers solutions with various thermodynamic behaviors, including ideal, moderately positive, and negative deviation from Raoult’s law and azeotropic systems as highly nonideal solutions. Using the obtained interaction parameters from the binary VLE data, the model can accurately predict the phase equilibria of multicomponent mixtures. The model covers the phase equilibrium behavior of all types of intermolecular interactions in binary systems well.