A tailored and rapid approach for ozonation catalyst design

In a new study published in Volume 15 of the journal Environmental Science and Ecotechnology, researchers from the Chinese Research Academy of Environment Sciences employed machine learning, specifically the artificial neural network (ANN) model, to predict catalyst performance based on data collected from 52 different catalysts. The ANN model demonstrated a strong correlation and generalization ability, indicating its robustness in predicting catalyst behavior. Additionally, fluorescence spectroscopy, which provides valuable information on the composition and concentration of organics in wastewater, was integrated with the machine learning model. This innovative approach leads to more efficient and effective treatment of refractory organics. Using the Mn/γ-Al₂O₃ catalyst as an example, the researchers successfully screened a range of catalyst formulations using fluorescence spectroscopy. They determined the optimal impregnation concentration and time of Mn(NO₃)₂ for specific wastewater compositions. The ANN model then generated an optimized formulation for the Mn/γ-Al₂O₃ catalyst, resulting in improved catalytic performance. The predicted and experimental values for total organic carbon removal are closely aligned, confirming the effectiveness of the optimized catalyst. The study also identified the synergistic effect of oxidation radicals (•OH and ¹O₂) and the Mn/γ-Al₂O₃ catalyst as the key factors contributing to the improved performance.

Highlights

•The ozonation catalyst formula is tailored for targeted wastewater via modeling.

•Catalyst formulation and influent fluorescence are the input of the ANN model.

•Fluorescence spectra data help guide the range of catalyst formulation parameters.

•The optimized catalyst as output showed a synergic effect of •OH and ¹O₂ in reaction.

This innovative approach offers a rapid and tailored solution for designing ozonation catalysts based on the unique characteristics of wastewater quality. By combining machine learning and fluorescence spectroscopy, researchers can optimize catalyst formulation more efficiently, leading to enhanced treatment of refractory organics in industrial wastewater. Moreover, applying the ANN model combined with fluorescence spectroscopy holds great potential for further advancements in catalyst development, performance prediction, and process simulation in complex wastewater systems. This approach provides a valuable strategy for researchers and practitioners in wastewater treatment, enabling the development of more sustainable and efficient treatment methods.

###

Reference

Title of original paper: A tailored and rapid approach for ozonation catalyst design

DOI: 10.1016/j.ese.2023.100244

Journal: Environmental Science and Ecotechnology  

Environmental Science and Ecotechnology (ISSN 2666-4984) is an international, peer-reviewed, and open-access journal published by Elsevier. The journal publishes significant views and research across the full spectrum of ecology and environmental sciences, such as climate change, sustainability, biodiversity conservation, environment & health, green catalysis/processing for pollution control, and AI-driven environmental engineering. ESE received its first impact factor of 9.371 (partial), according to the Journal Citation Report^TM 2022.