Methylene blue degradation using vortex fluidic device under UV irradiation: Comparison of response surface methodology and artificial neural network

Water contamination from industrial dyes is a serious environmental issue. However, the impact of methods to remove these dyes from water through either adsorption or photodegradation is hampered by the requirement of auxiliary materials. Presented is a catalyst-free degradation method for methylene blue (MB) using a Vortex Fluidic Device (VFD) coupled with UV light (254 nm) through the photo-contact electrification mechanism. This allows for either a reduction or degradation of the colour dye depending on the gas atmosphere. The increased efficiency by the VFD is caused by the high mass transfer and mechanical energy inside the liquid thin film. The rotation speed (rpm), time (min), and MB concentration (ppm) were considered independent variables at three levels, and experiments were arranged according to the central composite design. Results were analysed using response surface methodology (RSM) and artificial neural network (ANN). The results of correlation coefficient (R 2 ) and root mean square error (RMSE) for ANN-WOA are 0.996, and 1.328, respectively, and for RSM they are 0.989, and 2.262, respectively, confirming that ANN-WOA was more precise than RSM. For 5 ppm MB solution, being processed at 8500 rpm for 30 minutes in the VFD, 92.52% degradation was observed. A proportional relationship between rotational speed and time to MB degradation was observed whilst the MB concentration was inversely related to its degradation. Therefore, the VFD, as a new, green and low-cost method, can be applied for dye degradation to achieve clean water. • A catalyst-free degradation method for methylene blue (MB) is presented. • A Vortex Fluidic Device (VFD) coupled with UV light was applied for degradation. • The proposed method is based on the photo-contact electrification mechanism. • The effects of rotation speed, time, and MB concentration were considered. • Results analysed using response surface methodology and artificial neural network.