Continuous flow photoreduction and validation of Cr(VI) in wastewater using TiO2 nanoparticles: An interplay between catalyst phase and microfluidic parameters

The presence of hexavalent chromium (Cr(VI)) in water exceeding 0.05 mg/L constitutes a threat to both human health and the environment. Hence, the development of efficient methods for the removal and quantification of Cr(VI) in wastewater is of utmost importance. The photocatalytic reduction represents a valuable approach for transforming the toxic Cr(VI) into a non-toxic trivalent form of chromium (Cr(III)). However, this method encounters several challenges, including a slow reaction rate, the need for an acidic environment, high catalyst loading, and limited recovery efficiency. This study introduces an efficient microfluidic platform for the conversion of Cr(VI) in wastewater by integrating various reactor designs. We explored the use of nanosized TiO2 as a photocatalyst in both pure anatase and rutile phases and a combination of anatase/rutile phases to assess their reduction performance. A remarkable conversion efficiency of 95 % was attained by utilizing a serpentine microreactor coated with a photocatalyst in the anatase phase. In contrast to the traditional batch photoreduction method, which reported a conversion efficiency of only 72 %, microreactor technology demonstrates a superior photoreduction performance. The pseudo-first-order kinetics was observed and the rate of reaction was calculated to be 0.199 sec−1. In addition to confirming the photoreduction of Cr(VI), we have also presented a robust yet user-friendly approach for quantifying Cr(VI) by combining paper-based devices with a smartphone-enabled colorimetric technique. Finally, we employed machine learning (ML) algorithms to predict photoreduction efficiencies using a multiple linear regression model to gain an understanding of the impact of various parameters and the root mean square error (RMSE) for the model prediction was calculated to be equal to 0.92 indicating the overall performance of the ML model was excellent.