Solubility prediction of solutes in non-aqueous binary solvent mixtures

The possibility of replacing theoretically computed Abraham parameters with the experimental Abraham parameters in solubility prediction of solutes in non-aqueous binary solvent mixtures using the Jouyban-Acree model was investigated. The solubilities of 90 data sets collected from the literature were predicted using their Abraham parameters, the solvent coefficients and also the solubilities in mono-solvent systems. The accuracy of the predicted solubilities was evaluated by calculating the mean percentage deviation (MPD) and also individual percentage deviations (IPDs). The overall MPD for the analysis using experimental and computed Abraham parameters were the same and was < 14%. A favoured IPD distribution was obtained for these numerical analyses. The data sets investigated in this work were collected at various temperatures and the results confirmed the possibility of solubility prediction in binary solvents at various temperatures. We did explore the possibility of ab initio solubility prediction of solutes in binary mixtures using the calculated solubilities in mono-solvent systems, however, the difference between the predicted and observed values increased to ca. 60% for gas-to-solvent coefficients and ca. 200% for water-to-solvent coefficients, which is too large for many predictive applications.