The Hildebrand solubility parameter (δ) provides a numerical estimate of the degree of interaction between materials, and can be a good indication of solubility. In this work, a small number of physicochemical variables were appropriately selected from a pool of Dragon descriptors and correlated with the Hildebrand thermodynamic parameter of compounds previously studied as organic solvents of buckminsterfullerene (C60), using multiple linear regression and support vector machines. Models were validated using an external set of compounds and the statistical parameters obtained revealed the high prediction performance of all models, especially the one based on nonlinear regression. These findings provide useful information about which solvent and corresponding characteristics are important for solubility studies of e.g. this increasingly useful carbon allotrope.
- Artificial neural networks
- Hildebrand parameter
ASJC Scopus subject areas
- General Chemical Engineering
- General Physics and Astronomy
- Physical and Theoretical Chemistry