A Novel QSPR Approach in Modeling Selectivity Coefficients of the Lanthanum-Selective Electrode

This study presents a pioneering application of a novel quantitative structure-property relationship (QSPR) model to predict the selectivity coefficients of a cation-selective electrode. Specifically, the selectivity coefficients of a Lanthanum (La(III)) membrane sensor utilizing 8-amino-N-(2-hydroxybenzylidene) naphthylamine (AIP) as the sensing ligand were efficiently estimated and predicted. To establish the QSPR model, calculated molecular descriptors were employed, considering the limitation of cation descriptors. A new strategy was introduced for descriptor calculation by optimizing the structure of Mn+-AIP and utilizing density functional theory (DFT) with the B3LYP functional and SBKJC basis set. Genetic algorithm (GA) and stepwise techniques were employed for descriptor selection, with the most significant descriptors identified. Following variable selection, multiple linear regression (MLR) was employed to construct linear QSPR models. Comparative analysis revealed that the GA-MLR modeling approach exhibited superior performance compared to the stepwise-MLR method. Furthermore, the predictions generated by the GA-MLR model demonstrated excellent agreement with the experimental values. The proposed strategy outlined in this study has the potential to be extended to other QSPR investigations involving cation-selective electrodes. These findings contribute to the advancement of predictive modeling in the field of cation-selective sensors and offer valuable insights for future research in this area.