An improved design of fluorophilic molecules: prediction of the ln P fluorous partition coefficient, fluorophilicity, using 3D QSAR descriptors and neural networks

• Published in: Journal of fluorine chemistry 2001-03, Vol.108 (1), p.95-109
• Main Authors: Kiss, László E., Kövesdi, István, Rábai, József
• Format: Article
• Language: English
• Subjects: Fluorine; Fluorous partition coefficient; Neural network; Specific fluorophilicity; Structure–activity relationships
• ISSN: 0022-1139; 1873-3328
• DOI: 10.1016/S0022-1139(01)00342-6

A combination of 3D QSAR molecular descriptors and artificial neural networks have been used to predict fluorophilicities, the natural logarithm of the perfluoro(methylcyclohexane)/toluene partition coefficients, for a wide range of partially fluorinated organic compounds. The average error of the predictions was less than twice the 0.2 experimental error. Multiple linear regression proved to be much less efficient. To better characterise the fluorous partition phenomenon, specific fluorophilicity was defined as the product of fluorophilicity and of the ratio of the van der Waals volumes of the expelled fluorous solvent and the entering solute molecules. This dimensionless term correlates well in a compound family with the calculated Hildebrand parameters of the fluorous molecules. The trifluoromethyl group was found highly effective for increasing the fluorous phase affinities of model compounds when used in combination with longer perfluoroalkyl groups.