Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.13725 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917274710966272 |
|---|---|
| author | Edwards, Connor W. Yang, Fengxu Stracke, Konstantin Evans, Jack D. |
| author_facet | Edwards, Connor W. Yang, Fengxu Stracke, Konstantin Evans, Jack D. |
| contents | Grand canonical Monte Carlo (GCMC) simulations are essential for screening metal-organic frameworks (MOFs) for gas adsorption, yet their accuracy is limited by underlying interatomic potentials. Universal machine-learned interatomic potentials (MLIPs), trained on diverse chemical datasets, promise zero-shot prediction without system-specific training. We introduce MLIP-MC, an open-source Python framework to conduct GCMC simulations with MLIPs, and use this framework to benchmark a series of universal models, including MACE-MP-0, ORB-v3, and fairchem ODAC, for CO2 adsorption on ZIF-8, ZIF-4, and Mg-MOF-74. All universal models exhibit systematic biases, consistently over- or underestimating adsorption energetics. Crucially, accuracy depends on training data composition: only models trained on MOF-adsorbate interactions achieve reasonable agreement with a density functional theory derived reference. Errors grow linearly with CO2 uptake, reflecting compounding inaccuracies in adsorbate-adsorbate interactions. Our results demonstrate that current universal MLIPs require finetuning for quantitative adsorption predictions and demonstrate the power of MLIP-MC to rapidly test models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_13725 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MLIP-MC: A Framework for Adsorption Simulations using Machine-Learned Interatomic Potentials Edwards, Connor W. Yang, Fengxu Stracke, Konstantin Evans, Jack D. Materials Science Grand canonical Monte Carlo (GCMC) simulations are essential for screening metal-organic frameworks (MOFs) for gas adsorption, yet their accuracy is limited by underlying interatomic potentials. Universal machine-learned interatomic potentials (MLIPs), trained on diverse chemical datasets, promise zero-shot prediction without system-specific training. We introduce MLIP-MC, an open-source Python framework to conduct GCMC simulations with MLIPs, and use this framework to benchmark a series of universal models, including MACE-MP-0, ORB-v3, and fairchem ODAC, for CO2 adsorption on ZIF-8, ZIF-4, and Mg-MOF-74. All universal models exhibit systematic biases, consistently over- or underestimating adsorption energetics. Crucially, accuracy depends on training data composition: only models trained on MOF-adsorbate interactions achieve reasonable agreement with a density functional theory derived reference. Errors grow linearly with CO2 uptake, reflecting compounding inaccuracies in adsorbate-adsorbate interactions. Our results demonstrate that current universal MLIPs require finetuning for quantitative adsorption predictions and demonstrate the power of MLIP-MC to rapidly test models. |
| title | MLIP-MC: A Framework for Adsorption Simulations using Machine-Learned Interatomic Potentials |
| topic | Materials Science |
| url | https://arxiv.org/abs/2602.13725 |