Parametric Study and Process Evaluation of Fenton Oxidation: Application of Sequential Response Surface Methodology and Adaptive Neuro-Fuzzy Inference System Computing Technique

The Fenton oxidation is rarely used industrially due to its high operating cost, large chemical consumption, excessive sludge production, and operability only within a narrow pH range. Therefore, there is a need to evaluate the Fenton oxidation to maximize its ability to degrade high-strength dye wastewater at reduced operating cost. Optimization tools are among the most commonly used tool to maximize the degradation of pollutants. The current study aims at evaluating the applicability of response surface methodology (RSM) and adaptive neuro-fuzzy inference system (ANFIS) to optimize the degradation of Remazol brilliant blue through the Fenton oxidation. The effects of four operating parameters including dye concentration, retention time, and mass ratios of Dye:Fe²⁺ and H₂O₂:Fe²⁺ were evaluated by applying RSM. According to the RSM results, color and chemical oxygen demand (COD) removal of 99.9% and 84%, respectively, were obtained at 120?min at the COD value of 795?mg/L, mass ratios of Dye:Fe²⁺?=?16, H₂O₂:Fe²⁺?=?15 and pH?=?3. ANFIS was also used to evaluate the most influential operating parameters on the COD removal based on the RSM results. The ANFIS results showed that the mass ratio of H₂O₂:Fe²⁺ had the most significant contribution to the COD removal. High R² values (≥90%) indicated that the predictions of RSM and ANFIS models for COD removal were acceptable. In conclusion, this study demonstrated that RSM and ANFIS were able to determine the most significant operating parameters and optimum ratios of pollutant:oxidant:catalyst, which reduced the operating cost directly.