(83d) Liquid-Liquid Equilibria in Systems Containing Ionic Liquids Using COSMO-RS

In the field of chemical engineering and life sciences, the improvement of processes is of continuous interest. In the case of separation, e.g. extraction, an improvement could be made by the use of ionic liquids (ILs). ILs are “designer solvents”, thus the properties necessary for an application can be achieved by the systematic combination of anion and cation. Since there are many possible combinations, predictive methods are needed to replace/minimize measurements and to enable an a-priori screening of possible combinations.

One of the most popular and effective methods here is the COSMO-RS model. Based on the efficient combination of quantum mechanics and statistical thermodynamics, COSMO-RS allows for the phase equilibrium predictions based only on the chemical structure of the system components. In the frame of this model, IL can be treated as one compound (molecular form/meta-files) or as an equimolar mixture of the ions as two single compounds (dissociated form/single ions).

In this study, the influence of such treatment on the prediction results is studied for liquid-liquid equilibria (LLE). LLE in binary as well as in ternary systems (miscibility gaps and partition coefficients) containing ILs and water or alkanes, alcohols, ketones are analyzed.

In reality, both dissociation forms of IL can simultaneously be present in the solution. Therefore, an additional method is developed by introducing a degree of dissociation and considering the partial dissociation of IL. Based on the predictions for a series of ILs with different cations and anions, the effect of the solvent polarity and the strength of the ion-ion interaction between anion and cation on the IL’s degree of dissociation are investigated systematically.

In general, the presented method leads to improved prediction quality and guarantees a reliable screening result. In addition, it gives theoretical insight into the IL’s ionic interaction strength.