Simultaneous prediction of vapour–liquid and liquid–liquid equilibria (VLE and LLE) of aqueous mixtures with the SAFT-γ group contribution approach

Group contribution (GC) methodologies have been combined with the statistical associating fluid theory (SAFT), to couple the predictive capabilities of GC methods with the accuracy of the SAFT description of complex fluids and fluid mixtures. An example of such an procedure is the SAFT-γ group contribution approach [A. Lymperiadis, C.S. Adjiman, A. Galindo, G. Jackson, J. Chem. Phys., 127 (2007) 234903], which was developed based on the SAFT-VR equation of state and employs a heteronuclear molecular model of fused segments which represent the various chemical groups. In this work we examine the predictive capabilities of the method focusing on the fluid phase behaviour of aqueous mixtures. Within SAFT-γ it is often possible to obtain information for both the like and unlike group interactions from pure component fluid thermophysical data alone. When this is not possible, e.g., in the case of water where the whole molecule represents a distinct chemical “group”, the interactions can be obtained, in the usual activity coefficient GC fashion, from a minimal amount of experimental fluid phase equilibrium data of the appropriate mixtures. The group like and unlike parameters can then be successfully transferred to represent the thermophysical properties and fluid phase behaviour of a wide range of mixtures in a predictive manner. The binary systems investigated in this work are aqueous solutions of alkanes and alkanols. The successful description of the highly non-ideal fluid phase behaviour of systems of this kind with transferable interaction parameters within the SAFT-γ approach and the simultaneous prediction of both vapour–liquid and liquid–liquid equilibria using the same parameters are the key contributions of our work.

► SAFT-based group contribution method.
► Focus on study of phase behaviour of aqueous solutions.
► Interaction parameters from limited experimental data.
► Transferability of interaction parameters.
► Accurate predictions of VLE and LLE for wide range of systems and conditions.