Nti-Gyabaah, Joseph. Solubility and activity coefficient of pharmaceutical compounds in liquid organic solvents. Retrieved from https://doi.org/doi:10.7282/T3D21XSC
DescriptionThe production of pharmaceuticals and oligo-sized biochemicals involves liquid solvent selection as a function of solubility, for purification, chemical reaction and formulation. Selecting the optimum solvent for a particular application is of critical importance to developing efficient process; choosing solvents for pharmaceutical processes has been based on experience and empirical description of experimental results. Therefore, rapid and reliable prediction of drug solubility is needed for design and optimization of cost-effective manufacturing processes.
A semi-automated, two-millimeter-scale method, was designed and validated for rapid measurement of equilibrium solubility of crystalline solids in liquid solvents. Solubilities and stability of model compounds such as lovastatin, simvastatin, and artemisinin, were measured in several solvents and exhibited a maximum deviation of 5% compared to literature data. Extrapolation of the solubility in the form of activity coefficient was determined to agree with the experimental data.
Calculation of the ideal solubility of crystalline solid in liquid solvent requires knowledge of the difference in molar heat capacity of the solid and super-cooled liquid solute at the solution temperature. This is a hypothetical parameter and therefore, can not be measured directly, and hence, three assumptions are commonly used in the literature for its estimation. Evaluation of the assumptions revealed some thermodynamic inconsistencies. A new strategy was explored to estimate the difference in molar heat capacity, allowing the experimental solubility data to be fitted to the Two-Liquid-Non-Random (NRTL) activity coefficient equation to obtain the model binary energetic interaction parameters. The binary interaction parameters were successfully used to estimate solubility of the model compounds in mixed solvents.
In free energy perturbation (FEP) methods, mutation of lovastatin to simvastatin in five different solvents were carried out to obtain the free energy differences, and hence the ratio of the activity coefficients at infinite dilution. The FEP calculation reproduced the experimental solubility results quite well, and provides a basis for development of molecular modeling for estimation of pharmaceutical compounds in liquid solvents.