Statistical modeling of decontamination of wastewater containing Carbendazim using photo-Fenton

Abstract

Homogeneous photo-Fenton photocatalysis is one of the most widely used advanced oxidation processes in treating wastewater containing pesticides, where optimizing the catalyst dosage and oxidizing agent, with the response being mineralization in terms of total organic carbon (TOC) or removal of the specific contaminant, is the goal of any experimental design. Experimental development in treatment processes necessitates executing a significant number of experimental conditions that require the use of reagents, energy, and execution time. Therefore, modeling such phenomena emerges as an alternative to these limitations in wastewater treatment. In this research, the influence of factors FeSO4 and H2O2, each at three levels, on the mineralization of a synthetic solution of the insecticide Carbendazim (50 mg/L) in terms of TOC was evaluated using a multiple linear regression model optimized by response surface methodology. The main results established that the best model fit considers the interaction between FeSO4 and H2O2 (X1*X2) and the quadratic terms of each (X21,X22) with p-values < 0.05. The validation of the model using the Leave-One-Out Cross Validation (LOOCV) technique, as well as accuracy and precision through residual analysis and ordinary least squares assumptions, confirms the validity of the conclusions derived from it.

Keywords

• Carbendazim
• Pesticides
• photo-Fenton
• Polynomial regression
• Regression model