Predictive Modeling of Anionic Surfactant Content in Industrial Liquid Discharges from Soap and Cosmetics Factories

Abstract

The soap and cosmetics industries generate liquid waste laden with anionic surfactants, which pose a significant risk to the environment and aquatic biodiversity if their concentration is not controlled. Analyzing these harmful organic pollutants requires sophisticated equipment and enormous financial resources, making it difficult to determine their levels. The objective of this work is to develop a predictive model to estimate the anionic surfactant content in industrial waste from soap and cosmetics industries based on readily available physicochemical parameters. This study was conducted using thirty-five (35) samples of influents from the soap industries of Abidjan. The samples were divided into two groups, twenty-five (25) were used for the training set and ten (10) for the validation set. The analysis of standard physicochemical parameters such as COD, BOD₅, anionic and physical surfactants (T, pH, EC, EH), is carried out according to AFNOR and Rodier standards. RQSA/RQSP and Multiple Linear Regression (MLR) methods are used for model determination. A predictive model is obtained with R2=0.9346 and pH and redox potential (EH) as predominant descriptors. Analysis of the contribution of the descriptors indicates that pH is the parameter that best explains the change in surfactant concentration. Furthermore, external validation based on Trospha criteria shows that the model has good predictive power. The results will contribute to better regulation of industrial effluents and protection of aquatic resources in the District of Abidjan. It would be relevant to extend this approach to other types of industries for more comprehensive environmental management.