Predicting Gas Adsorption without the Knowledge of Pore Structures: A Machine Learning Method Based on Classical Density Functional Theory

Xiangkun Wu,Yu Liu

DOI: https://doi.org/10.1021/acs.jpclett.3c02708

IF: 6.888

2023-11-03

The Journal of Physical Chemistry Letters

Abstract:Predicting gas adsorption from the pore structure is an intuitive and widely used methodology in adsorption. However, in real-world systems, the structural information is usually very complicated and can only be approximately obtained from the characterization data. In this work, we developed a machine learning (ML) method to predict gas adsorption form the raw characterization data of N₂ adsorption. The ML method is modeled by a convolutional neural network and trained by a large number of data that are generated from a classical density functional theory, and the model gives a very accurate prediction of Ar adsorption. Though the training set is limited to modeling slit pores, the model can be applied to three-dimensional structured pores and real-world materials. The good agreements suggest that there is a universal relationship among adsorption isotherms of different adsorbates, which could be captured by the ML model.

chemistry, physical,physics, atomic, molecular & chemical,nanoscience & nanotechnology,materials science, multidisciplinary