Experimental and computational methods for assigning descriptors for the solvation parameter model

The solvation parameter model employs six fixed descriptors (seven for compounds of variable hydrogen-bond basicity) to model the equilibrium properties of neutral compounds in biphasic systems with numerous applications in separation science and distribution systems of environmental and biomedical interest. The six compound descriptors are: E, excess molar refraction; S, dipolarity/polarizability; A, overall hydrogen-bond acidity; B, overall hydrogen-bond basicity; L, gas-hexadecane partition constant at 25°C; and V, McGowan’s characteristic volume. The E descriptor for liquids and V are available by calculation. The other descriptors are experimental properties determined by chromatographic, liquid-liquid partition, and solubility measurements. Contemporary methods of descriptor assignments are reviewed with an emphasis on information published since 2020. The strengths and limitations of gas chromatography, liquid chromatography, and liquid-liquid partition methods as standalone techniques for descriptor assignments are discussed. Multi-technique approaches are superior to the standalone methods and are optimized to minimize the collection of experimental data. The Abraham and WSU-2025 curated descriptor databases are compared in terms of descriptor quality. The WSU-2025 database provides more precise calibration models resulting in improved descriptor quality. Computational methods for descriptor assignments using artificial intelligence approaches are reviewed and demonstrate good predictive ability for simple monofunctional compounds but performance is variable and sometimes poor for more complex molecules.