Machine Learning in Computational Design and Optimization of Disordered Nanoporous Materials

Materials, Journal Year: 2025, Volume and Issue: 18(3), P. 534 - 534

Published: Jan. 24, 2025

This review analyzes the current practices in data-driven characterization, design and optimization of disordered nanoporous materials with pore sizes ranging from angstroms (active carbon polymer membranes for gas separation) to tens nm (aerogels). While machine learning (ML)-based prediction screening crystalline, ordered porous are conducted frequently, porosity receive much less attention, although ML is expected excel field, which rich ill-posed problems, non-linear correlations a large volume experimental results. For micro- mesoporous solids carbons, silica, aerogels, etc.), obstacles mostly related navigation available data transferrable easily interpreted features. The majority published efforts based on obtained same work, datasets often very small. Even limited data, helps discover non-evident serves material production optimization. development comprehensive databases low-level structural sorption characteristics, as well automated synthesis/characterization protocols, seen direction immediate future. paper written language readable by chemist unfamiliar science specifics.

Language: Английский

Machine learning-assisted model for predicting biochar efficiency in colloidal phosphorus immobilisation in agricultural soils

Biochar, Journal Year: 2025, Volume and Issue: 7(1)

Published: March 14, 2025

Language: Английский