Computational QSPR Model to Predict the Critical Micelle Concentration (CMC) of Classic and Extended Sugar-Based Surfactants.

Abstract

As known, critical micelle concentration is a crucial characteristic in product formulation. In the current work, six molecular structure descriptors were used to create QSPR models that predicted the critical micelle concentration (CMC) of 119 sugar-based surfactants. The analysis of the qualities of descriptors shows that the micellization process is specifically affected by electronic properties (electronegativity and charges), electro-topology, and symmetry of a molecule. Four statistical learning techniques including Multiple linear regression, Partial least square, Artificial neural networks (ANN), and Adaptive neuro-fuzzy inference system to develop the QSPR models. different statistical metrics were employed to evaluate the reliability and robustness of the models. The best result (= 0.803,= 0.856,=0.982, and= 0.006) were obtained for ANN with {6-6-1} architecture. In addition, estimating the CMC of 6 other sugar surfactants based on simulate of the network gave very good results (R = 0.96).  Therefore, these findings suggest that the developed model is appropriate for predicting and correlating CMC value for sugar-based surfactants.