A plasmonic S-scheme Ag/ZrO2/TCN photocatalyst for enhancing interfacial charge transfer: Insights to machine learning models and mechanism for photodegradation

Designing a plasmonic heterojunction structure is a new approach for boosting the carriers' separation efficiency and expanding the visible light absorption. Therefore, developing plasmonic S-scheme photocatalysts with machine learning (ML) methods is considered a novel and efficient strategy for reducing problems of high cost, time-consuming, and possible instrumental errors in experimental techniques. A plasmonic S-scheme Ag/ZrO 2 /TCN heterojunction photocatalyst was successfully fabricated to degrade organic dyes including methylene blue (MB) and methyl orange (MO) under visible-light conditions. The loading of Ag NPs in the proposed photocatalyst enhanced the light absorption ability via the surface plasmon resonance (SPR) effect and accelerated the charge separation. Furthermore, the junction of TCN and ZrO 2 semiconductors with different fermi levels boosted interfacial charge transfer (IEF). Thus, compared with TCN and Ag/ZrO 2 , Ag/ZrO 2 /TCN heterojunction revealed more remarkable visible-light photocatalytic activity for eliminating organic pollutants . In addition, the ML models including Artificial Neural Networks (ANN) and particle swarm optimization (PSO) were also evaluated to predict and optimize MB and MO photodegradation . Moreover, the biocompatibility of the solution resulting from the photodegradation of organic dyes was considerable through the growth of wheat seeds. This study provides deep insight into the design of plasmonic S-scheme heterojunction materials and also develops intelligent decision-making and optimization of the pollutant removal processes from the environment. We have successfully developed a plasmonic S-scheme Ag/ZrO 2 /TCN heterojunction materials for optimization of the removal of methylene blue and methyl orange from aqueous solutions via incorporating machine learning models.