Molecular solubility and solvent properties can be predicted with accuracy by the tools of quantum chemistry and particularly COSMO-RS (COnductor-like Screening MOdel for Real Solvent). However, the semi-empirical Hansen approach is still widely used to investigate molecular and polymer solubilities that are usually represented by spheres containing good solvents and excluding the others in the Hansen 3D-space (δd ; δp ; δh). Here, COSMO-RS is used for the a priori calculation of the solubility sphere of macromolecules used in the coatings industry.
Different numbers of units were chosen for the polymer modeling in order to find the optimal (i.e. economical though representative) chain length. Then, the polymer is virtually mixed with various solvents used in the initial approach of Hansen at infinite dilution. Mixtures are simulated by the solvent screening tool of COSMOtherm, which provides a ranking of relative solubilities. Results are sorted in groups of solubility and a score is given to each group from 1 (best solvents) to 6 (poorest solvents). Scores are computed via HSPiP software to calculate the coordinates of the solubility sphere. The experimental validation is conducted by mixing the solute in a ratio 1:10 with the solvents used in the previous COSMO-RS calculation. Vials are stirred during 24 hours at 25°C. The solubility is then visually evaluated by a score from 1 to 6. All the results are also computed with HSPiP. The comparison of experimental and a priori results is in good agreement in most of cases.
It is possible to roughly build the Hansen solubility sphere of a polymer without any experimental data by using the COSMO-RS approach. This view is easy to read and to use in order to quickly identify appropriate solvents to solubilize coating materials. Particular attention should be paid to the modeling of macromolecules which strongly influences the results accuracy.