Design and optimisation of a low-cost titanium dioxide-coated stainless steel mesh photocatalytic water treatment reactor

Photocatalysis has been extensively studied in recent years for environmental wastewater treatment applications. Although promising, it has yet to be globally adopted, as it faces many challenges; namely cost, complexity and efficiency. This present work focuses on the optimisation of a bespoke photo-catalytic water treatment reactor. Contrary to other studies, the reactor was exclusively built from inexpensive and readily available consumer market parts, to facilitate a widespread adoption of this water treatment method. Photocatalytic TiO2 was synthesised and immobilised on stainless steel woven mesh in a one-step process, via reactive pulsed DC magnetron sputtering. A two-levels augmented screening design template was used to optimise the performance of the bespoke photocatalytic reactor, consisting of 20 experimental runs. Five independent variables were studied, UV light intensity, number of TiO2-coated mesh layers, coating thickness, water flowrate and initial dye concentration. Methylene blue dye solution was used as a model pollutant and the removal percentage after 5 h was used as a response. A linear regression model was built from the experimental results and revealed that all first-order terms, with the exception of flowrate, were significant contributors to the model pollutant removal. Increasing the coating thickness and the number of TiO2-coated mesh layers did improve the removal rate of methylene blue. These benefits cancelled each other when both variables were at their highest levels, due to a decreased light permeability through the mesh. ANOVA, lack-of-fit, and R-2 analysis confirmed the significance of the linear regression model. Optimised conditions were identified, leading to the removal of more than 90% of the model pollutant after 5 h of UV-A illumination. The calculated pseudo-first-order constant was as high as 14.5 x 10(-5) s(-1), while the quantum yield was estimated to be 4.22 x 10(-6) molecules/photons and the figure of merit was calculated at 1.14. This substrate/catalyst combination proved to be effective at degrading methylene blue, with no evident performance degradation after 10 repeated cycles, equivalent to 360 h of consecutive use. This present work demonstrates that it is possible to build an efficient photocatalytic reactor from inexpensive computer enthusiast parts, combined with a highly scalable and industry friendly photocatalyst pro-duction technique. (C) 2021 Elsevier Ltd. All rights reserved.