Abstract

Reduction of interfacial tension is of key importance for mobilising residual oil trapped by capillary forces during a secondary waterflood. The Conductor-like Screening Model (COSMO) and its extension COSMO-RS, developed over the last decades, enables prediction of thermodynamic properties of mixtures. Molecular charge distributions and associated charge distribution properties, so-called sigma moments, can be calculated and used as input parameters for various physical models to predict IFT of fluid mixtures.

In this paper we present and compare four different and complementary approaches to predict IFT based on single-molecule properties derived from COSMO-RS theory. The first method is based on predicted liquid-liquid extraction (LLE) phase-specific mole fractions using a formalism suggested by Apostoluk and Szymanowski (1996). The second method relies on a Taylor-like approximation of chemical potentials of mixtures using a realisation of Method of Moments (MoM) as described by Klamt and co-workers. The third method, recently described by Andersson et al., relies on LLE calculations as well as free energies for molecules present at the interfaces in multiphase immiscible systems. The fourth method, the so-called GSM-model recently described by the authors of this paper, relies on non-linear statistical relations between COSMO-derived energy descriptors and IFT.

By conducting a comprehensive comparative analysis we show that each of the four models estimates IFT with significant accuracy, that these models are complementary and that the models should be chosen based on the specific system of interest as well as the present available system information. Moreover, we demonstrate that COSMO-RS theory, when used in combination with physical models, provides a powerful tool for EOR research enabling fast, accurate computational prediction of IFT of multiphase fluid mixtures. Hence the models presented here may be used for systematic laboratory testing of e.g. surfactants and co-solvents for EOR processes as well as for predicting properties of multiphase fluid systems.

You can access this article if you purchase or spend a download.