journal article·2026-01-29

Combining Machine-Learning and Classical Potentials for Grand Canonical Monte Carlo Simulation of Chemisorption

Binquan Luan, Carine Ribeiro dos Santos, Bowen Zheng, Felipe Lopes Oliveira, Rodrigo Neumann Barros Ferreira, Mathias Steiner, Grace X. Gu

Publisher page

Abstract

Chemisorption, characterized by the strong and highly selective chemical bonding formation between a gas molecule and a solid sorbent, plays a critical role in various applications, such as surface functionalization, carbon capture, and hydrogen storage. Sorption processes are typically modeled by means of a Grand Canonical Monte Carlo (GCMC) simulation. However, GCMC methods are limited to physisorption processes described by molecular-mechanics or classical interatomic potentials. Therefore, a computationally efficient GCMC method that accounts for chemisorption is needed. Here, we report an accurate and efficient modeling method for GCMC simulations of chemisorption using hybrid interatomic potentials. Namely, the sorbate–sorbate interaction is described by classical potentials, and the sorbate–sorbent interaction is rendered by a machine-learning interatomic potential (MLIP) trained on simulation data extracted from quantum calculations. To prove the principal, we have applied the method to the simulation of CO 2 chemisorption in the porous metal–organic framework Mg-MOF-74. By comparing the GCMC simulated CO 2 adsorption isotherms and isobars, respectively, with the experimental data, we obtain better agreement compared with other classical and quantum-chemical simulation methods while keeping computational cost at a minimum. With the rise of transferable MLIP, we expect that the method could enable computationally efficient and highly accurate sorption studies in reticular materials.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Metal-Organic Frameworks: Synthesis and Applications · Covalent Organic Framework Applications · Carbon Dioxide Capture Technologies

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

IBM Research

IBM Research, Av. República do Chile, 330, Rio de Janeiro, RJ CEP 20031-170, Brazil

Department of Mechanical Engineering

University of California

Instituto de Química, Universidade Federal do Rio de Janeiro, Av. Athos da Silveira Ramos, 149, CT A-622, Cid. Univ., Rio de Janeiro, RJ 21941-909, Brazil

References (41)

Show source-provided references
1. 10.1021/jz300047n10.1021/jz300047n ↗
2. 10.1021/jz500202x10.1021/jz500202x ↗
3. 10.1039/C8SC00971F10.1039/C8SC00971F ↗
4. 10.1021/ja00051a04010.1021/ja00051a040 ↗
5. 10.1021/j100389a01010.1021/j100389a010 ↗
6. 10.1021/ja00205a00110.1021/ja00205a001 ↗
7. 10.1002/jcc.2003510.1002/jcc.20035 ↗
8. 10.1021/ct400952t10.1021/ct400952t ↗
9. 10.1080/08927022.2013.82922810.1080/08927022.2013.829228 ↗
10. 10.1021/acs.jpcc.8b0863910.1021/acs.jpcc.8b08639 ↗
11. 10.1016/j.jece.2022.10893010.1016/j.jece.2022.108930 ↗
12. 10.1016/j.jmgm.2022.10812410.1016/j.jmgm.2022.108124 ↗
13. 10.1002/cssc.20160154510.1002/cssc.201601545 ↗
14. 10.1002/anie.20150695210.1002/anie.201506952 ↗
15. 10.1002/adfm.20230747810.1002/adfm.202307478 ↗
16. 10.1039/C4SC02064B10.1039/C4SC02064B ↗
17. 10.1016/j.micromeso.2013.06.02310.1016/j.micromeso.2013.06.023 ↗
18. 10.1021/ja803609610.1021/ja8036096 ↗
19. 10.1002/adts.20190013510.1002/adts.201900135 ↗
20. 10.1021/acs.jpcc.6b0749310.1021/acs.jpcc.6b07493 ↗
21. 10.1038/nchem.143210.1038/nchem.1432 ↗
22. 10.1021/jp306252710.1021/jp3062527 ↗
23. 10.1021/acs.jpcc.5b0370010.1021/acs.jpcc.5b03700 ↗
24. 10.1021/acs.jpcc.6b0339310.1021/acs.jpcc.6b03393 ↗
25. 10.1021/jp508249c10.1021/jp508249c ↗
26. 10.1039/C8CC06178E10.1039/C8CC06178E ↗
27. 10.1021/acsnano.2c1110210.1021/acsnano.2c11102 ↗
28. 10.1063/5.019037210.1063/5.0190372 ↗
29. 10.1038/s41524-024-01205-w10.1038/s41524-024-01205-w ↗
30. 10.1039/D1CP05973D10.1039/D1CP05973D ↗
31. 10.26434/chemrxiv-2022-n1g6010.26434/chemrxiv-2022-n1g60 ↗
32. 10.1021/acs.jctc.3c0049510.1021/acs.jctc.3c00495 ↗
33. 10.1039/c3dt50599e10.1039/c3dt50599e ↗
34. 10.1016/j.cpc.2018.03.01610.1016/j.cpc.2018.03.016 ↗
35. 10.1016/j.cpc.2021.10817110.1016/j.cpc.2021.108171 ↗
36. 10.1002/aic.69047071910.1002/aic.690470719 ↗
37. 10.1039/c1ee01720a10.1039/c1ee01720a ↗
38. 10.1021/acs.energyfuels.6b0251010.1021/acs.energyfuels.6b02510 ↗
39. 10.1038/s41467-020-17755-810.1038/s41467-020-17755-8 ↗
40. 10.1021/acsami.1c1622010.1021/acsami.1c16220 ↗
41. 10.1021/acs.chemrev.0c0126610.1021/acs.chemrev.0c01266 ↗