QSPR with extended topochemical atom (ETA) indices: Modeling of critical micelle concentration of non-ionic surfactants

Abstract

In this present study, we have developed predictive quantitative structure–property relationship (QSPR) models with extended topochemical atom (ETA) indices for critical micelle concentration (logCMC) values of 54 non-ionic surfactants. The ETA descriptors have been developed by the present authors’ group and these can be easily calculated from 2D representation of chemical structure without requirement of conformational analysis and alignment steps. Different chemometric tools such as stepwise multiple linear regression (MLR), genetic function approximation (GFA) and partial least squares (PLS) were employed in this study for development of the models. The final PLS models were found to be well validated internally, externally and also by the overall validation technique. From the results, it is clear that the best ETA model shows reliable prediction of the logCMC values and the statistical quality of the model is comparable with the corresponding non-ETA model. It is also observed that the use of ETA descriptors along with the non-ETA ones improved the statistical quality of the models. Thus, it can be inferred that the ETA indices encode important chemical information regarding not only the topological attributes, but also the effect of electronegativity, molecular volume, branching, shape parameter and nature of atoms and bonds, etc. Hence, the ETA descriptors can be satisfactorily employed for modeling the CMC values of non-ionic surfactants.