Optimization of Photo-electrochemical Treatments for the Degradation of Complex Organic Compounds and Prediction of Degradation via Artificial Neural Networks

Advanced oxidation processes (AOP) are known for their efficiency in degrading organic pollutants. In this way, there is a continuous interest in promoting improvements in operating conditions, which can be done through the combination with homogeneous and electrochemical processes. In this work, the degradation of the mixture of dyes applied in the sanitizer industry, acid yellow 36 (AY36) and acid blue 80 (AB80), was evaluated against photo-electrochemical processes, associating individually and combined way UV-A and UV-C radiations. It was verified that the use of UV-C radiation was more efficient, promoting the complete degradation of the chromophore groups and reducing the aromatic groups by 76% when making use of the photo-electro-Fenton system (PEF/FeCl3). The experimental data followed a nonlinear kinetics suitable for pseudo-first-order models. Multilayer perceptron (MLP) artificial neural networks (6-6-3) using the Statistica 8.0 software allowed the modeling of the treatments applied in this work and showed a good prediction of data for the dye mixture. The results of this study show that the use of photo-electrochemical processes is an effective way to treat dyes used in sanitizing industries.