The separation between solvent polarizability and solvent dipolarity: Revisiting the Kamlet-Abraham-Taft model equation

Quantitative structure-property relationships were established using multiple linear regressions to cor-relate log k values for the heterolysis reaction of 3-bromo-3-ethylpentane using three well-known model equations, the Kamlet-Abraham-Taft (KAT), the Catalan (Cat) and the Laurence (Lau) equations. The a pri-ori exclusion of the hydrogen bond acceptor (HBA) basicity descriptor for all tested models led to similar responses, thus emptying the argument of an alleged use of a non-balanced set of solvents. Concurrently, a new method was devised to split the original Kamlet-Taft p* descriptor into two independent contri-butions which separately quantify the solvent's polarizability (DI) and dipolarity (Dip). A modified version of the KAT equation (mKAT) including both DI and Dip, together with the exclusion of the HBA basicity descriptor was applied to five selected solvent-dependent physicochemical processes. Results showed that the mKAT model provided significant improvements over the original KAT model. Additionally, it was shown that the introduction of DI in the truncated KAT equation could lead to misleading or, at best, inconclusive interpretations of solvent effects. Finally, the performance of mKAT model was compared with Catalan's and Laurence's model equations in terms of statistical results, relevance of descriptors and gas-phase predictions. In general, these three model equations show very similar outcomes although the mKAT model exhibits an overall better performance. (c) 2022 Elsevier B.V. All rights reserved.

Automated summary

Quantitative structure-property relationships were established and a new model equation mKAT was proposed to improve the interpretation of solvent effects in the heterolysis reaction of 3-bromo-3-ethylpentane.