Fingerprinting diverse nanoporous materials for optimal hydrogen storage conditions using meta-learning

Yangzesheng Sun; Robert F DeJaco; Zhao Li; Dai Tang; Stephan Glante; David S Sholl; Coray M Colina; Randall Q Snurr; Matthias Thommes; Martin Hartmann; J Ilja Siepmann

doi:10.1126/sciadv.abg3983

Fingerprinting diverse nanoporous materials for optimal hydrogen storage conditions using meta-learning

Sci Adv. 2021 Jul 21;7(30):eabg3983. doi: 10.1126/sciadv.abg3983. Print 2021 Jul.

Authors

Yangzesheng Sun¹, Robert F DeJaco², Zhao Li³, Dai Tang⁴, Stephan Glante⁵, David S Sholl⁴, Coray M Colina⁶, Randall Q Snurr³, Matthias Thommes⁷, Martin Hartmann⁵, J Ilja Siepmann^{8

2}

Affiliations

¹ Department of Chemistry and Chemical Theory Center, University of Minnesota, 207 Pleasant Street SE, Minneapolis, MN 55455-0431, USA.
² Department of Chemical Engineering and Materials Science, University of Minnesota, 412 Washington Avenue SE, Minneapolis, MN 55455-0132, USA.
³ Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, USA.
⁴ School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0100, USA.
⁵ Erlangen Center for Interface Research and Catalysis (ECRC), Friedrich-Alexander-University Erlangen-Nürnberg, Egerlandtstr. 3, 91058 Erlangen, Germany.
⁶ Department of Chemistry, University of Florida, PO Box 117200, Gainesville, FL 32611, USA.
⁷ Department of Chemical and Bioengineering, Institute of Separation Science and Technology, Friedrich-Alexander-University Erlangen-Nürnberg, 91058 Erlangen, Germany.
⁸ Department of Chemistry and Chemical Theory Center, University of Minnesota, 207 Pleasant Street SE, Minneapolis, MN 55455-0431, USA. siepmann@umn.edu.

Abstract

Adsorptive hydrogen storage is a desirable technology for fuel cell vehicles, and efficiently identifying the optimal storage temperature requires modeling hydrogen loading as a continuous function of pressure and temperature. Using data obtained from high-throughput Monte Carlo simulations for zeolites, metal-organic frameworks, and hyper-cross-linked polymers, we develop a meta-learning model that jointly predicts the adsorption loading for multiple materials over wide ranges of pressure and temperature. Meta-learning gives higher accuracy and improved generalization compared to fitting a model separately to each material and allows us to identify the optimal hydrogen storage temperature with the highest working capacity for a given pressure difference. Materials with high optimal temperatures are found in close proximity in the fingerprint space and exhibit high isosteric heats of adsorption. Our method and results provide new guidelines toward the design of hydrogen storage materials and a new route to incorporate machine learning into high-throughput materials discovery.

Copyright © 2021 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial License 4.0 (CC BY-NC).