Prediction of aqueous solubilities of solid carboxylic acids with COSMO-RS

Abstract

The methodology for the prediction of aqueous solubilities of solid organic compounds, and their temperature dependence, based on the Conductor-like Screening Model for Real Solvents (COSMO-RS/COSMOtherm) procedure, is presented and evaluated. The predictive capability of the quantum chemistry based program and the applied methodology was tested on the most common solid carboxylic acids. From the temperature dependence of the solubilities, the mean apparent enthalpies of solution were derived. The results obtained for a set of 27 carboxylic acids, consisting of aromatic carboxylic acids, dicarboxylic acids, as well as hydroxycarboxylic acids, are in good agreement with the experimental solubility data and their dependence with the temperature. The mean apparent enthalpies of solution, although of the same order of magnitude of the experimentally values. seem to be systematically underestimated. The good response of the applied methodology based on CCSMOtherm/CCSMO-RS on the prediction of non-ideal aqueous solubilities demonstrate the capability of this approach to predict and distinguish the solubility trends and magnitude, resulting from the substituent effects on the non-ideality of the aqueous solution.