Calculation of interfacial tension of binary systems containing water and an organic component by group contribution methods

Abstract

Abstract This work proposes a predictive method for estimating liquid-liquid interfacial tension (IFT) by group contribution methods of systems containing water and an organic component. The model for calculating the interfacial tension of liquid–liquid systems was derived using a general thermodynamic framework that was empirically extended by the introduction of an effective interfacial area. The effective interfacial area is defined for each component and takes into account the effects of groups of the other component present at the liquid-liquid interface. In addition, the interfacial molar volume was estimated by a mixing rule, using the molar volume of pure components in the liquid phase, weighted by the equilibrium compositions calculated by the UNIFAC model. The corresponding thermophysical properties were obtained by a group contribution method, in which a temperature-dependent binary parameter was proposed. This parameter is given by the sum of contributions related to each group present in the molecule, with values estimated from an experimental database of interfacial tension of water and different organic compounds. Therefore, the model can be used to predict the interfacial tension of binary mixtures in different temperatures through the knowledge of the structure of the organic component. The values of the model parameters were determined by a parameter estimation procedure based on an extensive database, covering 546 interfacial tension data of liquid-liquid binary systems containing water and an organic component. The model predicts interfacial tension data for binary systems containing organic compounds with different chemical functions in equilibrium with water, including oxygenated organic compounds, which are rarely predicted with good accuracy. The total mean absolute deviation obtained from this method was 1.20 mN⋅m−1, better than other predictive model and also better than some correlative models available in the literature.