A Simple and Robust Model for Predicting the Reduction Potential of Quinones Family; Electrophilicity Index Effect

Abstract

This paper describes a procedure revealing the relationship between quantum chemical indices and electronic descriptors as independent variables and half-wave potential (E1/2) of quinones family as a dependent variable. In this manner, three different quinone families including; benzoquinone, naphthoquinone and anthraquinone derivatives were subjected to structure-property relationship, quantitatively. Density functional theory (DFT) method at the level of B3LYP employing 6-31+g(d) basis set was performed to complete geometrical optimization. After calculating the electronic descriptors and quantum chemical indices for each compound, the quantitative relationship between the molecular property and the calculated descriptors was obtained by the multiple linear regression (MLR) and the features (or variables) were selected by genetic algorithm. The results showed that the electrophilicity index has significant correlation with reduction potential of quinones family. The accuracy of the proposed GA-MLR model was illustrated using the following evaluation techniques: cross-validation, validation through an external test set, and Y-randomization. This model, demonstrating high statistical qualities (R2Train = 0.983, Q2LOO= 0.979, Q2LGO = 0.976 and FTrain = 1051.100). An excellent agreement between the predictions and the experimentally obtained half-wave potential values for test set data was found (Q2Ext=0.970), with a root mean square error of (RMSE) of 0.057 V.