This study employs a novel integration of Computer Vision (CV) and Deep Learning (DL) algorithms (CNN/RNN-LSTM algorithm) to investigate the acidity and photocatalytic performance of a green nanocatalyst: copper oxide nanoparticles supported on sodalite nanozeolite (nSOD@CuO-NPs). The nanocatalyst, synthesized using Camelia sinensis (green tea) extract, was characterized, revealing the sodalite, tenorite, and cuprite phases and a good compromise between surface area and stability. The NH 3 -TPD analysis indicated a presence of weak-to-moderate acidity of the nanocatalysts and a significant reduction in the density of acid sites upon CuO-NPs loading, decreasing from Ns = 135 µmol NH3 g -1 catalyst (nSOD) to Ns = 58 µmol NH 3 g -1 catalyst (nSOD@CuO-NPs), probably due to occupation/blockage of nanozeolite active sites by the nanoparticles. This modification suggests a plausible transition from BrØnsted-dominant to Lewis-dominant acidity, potentially enhancing oxidation reactions. Exploratory Data Analysis (EDA) of heterogeneous photocatalysis assays demonstrated that nSOD@CuO-NPs achieved 82% removal of a highly concentrated (70 L -1 ) binary mixture of Crystal Violet (CV) and Methylene Blue (MB) dyes after 240 min at an optimal concentration of 0.8 L -1 and neutral pH. The DL/CV model generated a 3D morphology with an acidity heat map, enabling the proposal of a comprehensive photocatalytic degradation mechanism under visible light, highlighting the critical role of Cu-based Lewis acid sites and the generated ROS in the degradation pathway. A subsequent kinetic study on the main byproduct, aniline, showed an 89.6% removal, confirming the promising photocatalytic activity of the nSOD@CuO-NPs for complete wastewater treatment. • A green copper oxide nanoparticles supported on sodalite nanozeolite was synthesized. • The photocatalytic activity of nSOD@CuO-NPs in a bicomponent system was investigated. • Up to 82% degradation was achieved for CV:MB dye mixture (70 L -1 ) after 240 min • nSOD@CuO-NPs’ acid sites distribution was explored by CNN/RNN-LSTM algorithms. • The acidity of nSOD@CuO-NPs nanocatalyst was explored by NH 3 -TPD and CV/DL analysis.
Oviedo et al. (Thu,) studied this question.