Experimental study and thermodynamic modeling of the solubility of 4-aminobenzoic acid in binary mixtures of solvents

Selection of an appropriate solvent is of interest in the pharmaceutical industry. In some case, a mixture of solvents has a more solubility than the pure solvents. A predictive model such as the UNIFAC model can be employed to predict the solubility to reduce the cost of tests. However, the accuracy of the UNIFAC model should be evaluated to ensure the results of this model. In this work, the accuracy of the UNIFAC model is studied in predicting the solubility of 4-aminobenzoic acid in the binary mixtures of methanol, ethanol, acetone, and water. First, the UNIFAC model is used to predict the solubility of 4-aminobenzoic acid in the binary mixtures and then the binary mixtures whose solubility are high are selected to measure their solubility of 4-aminobenzoic acid and to compare with the results of the model. The experimental data showed that the solubility of 4-aminobenzoic acid in the binary mixtures of methanol-acetone (where the mole fraction of methanol is 0.65) and ethanol-acetone (where the mole fraction of ethanol is 0.2) is significantly higher than that of the pure solvents. However, the UNIFAC model predicted the maximum solubility of 4-aminobenzoic acid in methanol-acetone mixture where the mole fraction of methanol was 0.38 and, in the ethanol-acetone mixture where the mole fraction of ethanol was 0.3. Thus, the UNIFAC model can be used for an initial estimation of the maximum solubility in the mixtures; however, this model is not precise model for accurate prediction of the solubility of 4-aminobenzoic acid.