Concurrent Optimization of Decolorization, COD Removal, and Their Costs in Response to Photocatalytic Degradation Integrating Experimental Designs, Artificial Immune System, and Empirical Modeling

Abstract

This study aims at the quantification of color and chemical oxygen demand (COD) removal of methylene blue by TiO2-assisted photocatalytic degradation, and associated costs based on various designs of experiment (DOEs). For this purpose, effects of seven explanatory variables of catalyst dose (CD, g×TiO2×L-1), initial dye concentration (IDC, mg×L-1), initial pH (pH), temperature (T, °C), lamp type (LT), aeration (AR, mL×min-1), and reaction time (RT, min) were investigated on the related response variables. Experimental results demonstrated that CD, AR, IDC, and RT were significantly effective on the response variables. Novelty of the study lies in simultaneously optimizing these four responses based on DOEs, and Artificial immune System (AIS) optimization. Taguchi Orthogonal Array (TOA) as the best optimization DOE led to complete decolorization (> 99%) and 95.1% COD removal with UV-C lamp whose costs corresponded to 6.5 and 41.4 USD, respectively, under aeration rate of 20 mL×min-1, initial dye concentration of 10 mg×L-1, TiO2 concentration of 4 g×L-1, and reaction time of 20 min. AIS optimization yielded complete decolorization and 96.4% COD removal under the same TOA experimental conditions, with their corresponding costs of 0.5 and 3.1 USD, respectively. D-optimality and Box-Behnken designs were found as the second best DOEs for decolorization and COD removal, and their costs, respectively.