Development of kinetic models for photoassisted electrochemical process using Ti/RuO₂ anode and carbon nanotube-based O₂-diffusion cathode

A coupled photoassisted electrochemical system was utilized for degradation of C.I. Basic Yellow 28 (BY28) as a cationic azomethine dye under recirculation mode. Experiments were carried out by utilizing active titanium/ruthenium oxide (Ti/RuO₂) anode and O₂-diffusion cathode with carbon nanotubes (CNTs). Transmission electron microscopy (TEM) image of the CNTs demonstrated that CNTs had approximately an inner and outer diameter of 5 nm and 19 nm, respectively. Then, the dye degradation kinetics was experimentally examined under various operational parameters including BY28 initial concentration (mg/L), current density (mA/cm²), flow rate (L/h) and pH. Based on the generally accepted intrinsic elementary reactions for photoassisted electrochemical process (PEP), a novel kinetic model was proposed and validated for predicting the k_app. The developed kinetic model explicitly describes the dependency of the k_app on BY28 initial concentration and current density. A good agreement was obtained between the predicted values of k_app and experimental results (correlation coefficient (R²) = 0.996, mean squared error (MSE) = 2.10 × 10^-4 and mean absolute error (MAE) = 1.10 × 10^-2). Finally, in order to profoundly evaluate and compare the accuracy of the suggested intrinsic kinetic model, an empirical kinetic model was also developed as a function of main operational parameters, and an artificial neural network model (ANN) by 3-layer feed-forward back propagation network with topology of 5:9:1. The performance of the mentioned models was compared based on the error functions and analysis of variance (ANOVA). A comparison according to the errors function demonstrated that the experimental data were fitted appropriately by all of the proposed models with an adequate accuracy. Moreover, ANOVA results showed that there is no significant discrepancy among the predicted values of the three proposed models.