Details
| Original language | English |
|---|---|
| Article number | 2417891 |
| Number of pages | 9 |
| Journal | Advanced functional materials |
| Volume | 35 |
| Issue number | 13 |
| Publication status | Published - 24 Mar 2025 |
Abstract
Nanoporous and chemically-bridged graphene nanosheets span a wide chemical space with a broad set of applications in sensing and electronics. Modeling the structure and dynamics of such nanosheets is challenging, as chemical bond making and breaking as well as non-covalent interactions must be captured accurately and on equal footing. Here it is showed that recent graph-based machine-learned force field (MLFF) SO3krates [J. T. Frank et al., Nat. Commun. 15, 6539 (2024)] is able to reliably model the dynamics and mechanical response for a broad class of nanoporous graphenes when trained on accurate density functional theory data that includes long-range many-body dispersion (MBD) interactions. In contrast, local moment tensor potentials and empirical reactive potentials are much less accurate. It is also found that recent MLFFs trained on solid-state datasets must be used with care, since even empirical potentials occasionally yield more accurate results. These findings highlight the potential of properly-trained graph MLFFs in modeling the properties of whole chemical spaces of complex functional materials.
Keywords
- DFT calculations, machine learning force fields, nanoporous graphenes
ASJC Scopus subject areas
- Materials Science(all)
- Electronic, Optical and Magnetic Materials
- Chemistry(all)
- General Chemistry
- Materials Science(all)
- Biomaterials
- Materials Science(all)
- General Materials Science
- Physics and Astronomy(all)
- Condensed Matter Physics
- Chemistry(all)
- Electrochemistry
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Advanced functional materials, Vol. 35, No. 13, 2417891, 24.03.2025.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Mechanical Properties of Nanoporous Graphenes
T2 - Transferability of Graph Machine-Learned Force Fields Compared to Local and Reactive Potentials
AU - Kabylda, Adil
AU - Mortazavi, Bohayra
AU - Zhuang, Xiaoying
AU - Tkatchenko, Alexandre
N1 - Publisher Copyright: © 2024 The Author(s). Advanced Functional Materials published by Wiley-VCH GmbH.
PY - 2025/3/24
Y1 - 2025/3/24
N2 - Nanoporous and chemically-bridged graphene nanosheets span a wide chemical space with a broad set of applications in sensing and electronics. Modeling the structure and dynamics of such nanosheets is challenging, as chemical bond making and breaking as well as non-covalent interactions must be captured accurately and on equal footing. Here it is showed that recent graph-based machine-learned force field (MLFF) SO3krates [J. T. Frank et al., Nat. Commun. 15, 6539 (2024)] is able to reliably model the dynamics and mechanical response for a broad class of nanoporous graphenes when trained on accurate density functional theory data that includes long-range many-body dispersion (MBD) interactions. In contrast, local moment tensor potentials and empirical reactive potentials are much less accurate. It is also found that recent MLFFs trained on solid-state datasets must be used with care, since even empirical potentials occasionally yield more accurate results. These findings highlight the potential of properly-trained graph MLFFs in modeling the properties of whole chemical spaces of complex functional materials.
AB - Nanoporous and chemically-bridged graphene nanosheets span a wide chemical space with a broad set of applications in sensing and electronics. Modeling the structure and dynamics of such nanosheets is challenging, as chemical bond making and breaking as well as non-covalent interactions must be captured accurately and on equal footing. Here it is showed that recent graph-based machine-learned force field (MLFF) SO3krates [J. T. Frank et al., Nat. Commun. 15, 6539 (2024)] is able to reliably model the dynamics and mechanical response for a broad class of nanoporous graphenes when trained on accurate density functional theory data that includes long-range many-body dispersion (MBD) interactions. In contrast, local moment tensor potentials and empirical reactive potentials are much less accurate. It is also found that recent MLFFs trained on solid-state datasets must be used with care, since even empirical potentials occasionally yield more accurate results. These findings highlight the potential of properly-trained graph MLFFs in modeling the properties of whole chemical spaces of complex functional materials.
KW - DFT calculations
KW - machine learning force fields
KW - nanoporous graphenes
UR - http://www.scopus.com/inward/record.url?scp=85211815192&partnerID=8YFLogxK
U2 - 10.1002/adfm.202417891
DO - 10.1002/adfm.202417891
M3 - Article
AN - SCOPUS:85211815192
VL - 35
JO - Advanced functional materials
JF - Advanced functional materials
SN - 1616-301X
IS - 13
M1 - 2417891
ER -