A Novel Machine Learning Model to Predict the Photo-Degradation Performance of Different Photocatalysts on a Variety of Water Contaminants

Abstract

This paper describes an innovative machine learning (ML) model to predict the performance of different metal oxide photocatalysts on a wide range of contaminants. The molecular structures of metal oxide photocatalysts are encoded with a crystal graph convolution neural network (CGCNN). The structure of organic compounds is encoded via digital molecular fingerprints (MF). The encoded features of the photocatalysts and contaminants are input to an artificial neural network (ANN), named as CGCNN-MF-ANN model. The CGCNN-MF-ANN model has achieved a very good prediction of the photocatalytic degradation rate constants by different photocatalysts over a wide range of organic contaminants. The effects of the data training strategy on the ML model performance are compared. The effects of different factors on photocatalytic degradation performance are further evaluated by feature importance analyses. Examples are illustrated on the use of this novel ML model for optimal photocatalyst selection and for assessing other types of photocatalysts for different environmental applications.

Photo-oxidation of Micro- and Nanoplastics: Physical, Chemical, and Biological Effects in Environments

“…Limited studies have considered the relationship between adsorption and photodegradation of OPs in the presence of MNPs. Thus, more research should associate the physicochemical characteristics of OPs such as hydrophobicity, electronegativity, and functional groups with their phototransformation process in the presence of MNPs, potentially utilizing molecular dynamic simulations, and machine learning. , …”

“…Thus, more research should associate the physicochemical characteristics of OPs such as hydrophobicity, electronegativity, and functional groups with their phototransformation process in the presence of MNPs, potentially utilizing molecular dynamic simulations, 232 and machine learning. 233,234 6.4. Pay Attention to the Effect of Photoaged MNPs on the Biotransformation of Pollutants.…”

Physical Modifications and Algorithmic Predictions behind Further Advancing 2D Water Splitting Photocatalyst: An Overview

“…The encoded metal oxide structure through a crystal graph convolution neural network (CGCNN). [116] The organic molecular structure is encoded through digital molecular fingerprints (MF). Finally, the CGCNN, MF, and the light are used as input at an artificial neural network (ANN) to build the CGCNN-MF-ANN model (Fig.…”

Machine learning integrated photocatalysis: progress and challenges

“…Note that it does not cover the work of applying machine learning in the photoelectrocatalysis field [43][44][45][46][47] and photocatalytic degradation. [48][49][50][51] The advances are first presented from a photocatalysis perspective, focusing on three primary processes: light absorption, charge generation and separation, and surface redox reaction. Then, progress in using machine learning to understand the complex structure-photoactivity relationship and gain insights into the governing factors of activity follows.…”