Machine-Learning-Assisted Optimization of a Single-Atom Coordination Environment for Accelerated Fenton Catalysis

Abstract

Machine learning (ML) algorithms will be enablers in revolutionizing traditional methods of materials optimization. Here, we broaden the use of ML to assist the construction of Fenton-like single-atom catalysts (SACs) by developing a methodology including model building, training, and prediction. Our approach can efficiently extract synthesis parameters that exert a substantial influence on Fenton activity and accurately predict the phenol degradation rate k of SACs with a mean error of ±0.018 min −1 . The extended synthesis window with accelerated learning enables the realization that the heating temperatures during SAC synthesis significantly influence the Fe−N coordination number, which ultimately dictates their performance. Through ML-guided optimization, a highly efficient SAC dominated by Fe−N 5 sites with exceptional Fenton activity (k = 0.158 min −1 ) is identified. Our work provides an example for ML-assisted optimization of single-atom coordination environments and illuminates the feasibility of ML in accelerating the development of high-performance catalysts.

Single‐Atom Heterogeneous Catalysts: Human‐ and AI‐Driven Platform for Augmented Designs, Analytics and Reality‐Enabled Manufacturing

“…(Figure 2a,b) The limitations of the augmented integrated design-synthesis-analysis approaches of single-atom catalysts may be resolved by integration of the continuously updated data-driven approach at large scales (Figure 2c). [79][80][81] Along with the recent development of high-throughput computational approach supplemented with advanced artificial neural network, AI has been widely explored in various fields of science. Considering the large amount of powerful software packages and algorithms, we only focus on solidstate SAHCs design applied to heterogeneous catalysis by using ML approaches [82][83][84][85] (e.g., inverse design, highthroughput screening, multi-parameters sampling, and heterobimetallic programming).…”

Corrosion Degree Evaluation of Polymer Anti-Corrosive Oil Well Cement under an Acidic Geological Environment Using an Artificial Neural Network

“…With the development of computer technology, neural network algorithms have been widely applied to complex nonlinear systems based on human brain structures [13,14]. Hao et al and Shiratori et al [15,16] proposed an improved prediction model based on neural networks for simultaneously predicting electricity consumption and coal consumption during cement calcination, which has high accuracy.…”