A Multi-Scale Graph Neural Network for the Prediction of Multi-Component Gas Adsorption

Lujun Li; Haibin Yu

doi:10.1016/j.eng.2025.08.012

Engineering ›› 2025, Vol. 52 ›› Issue (9) :102 -111. DOI: 10.1016/j.eng.2025.08.012

Research

Article

A Multi-Scale Graph Neural Network for the Prediction of Multi-Component Gas Adsorption

Lujun Li ^a^,^b^,^c
, Haibin Yu ^b^,^c^,^*

Author information +

History +

PDF (2349KB)

Abstract

Metal-organic frameworks (MOFs) hold great potential for gas separation and storage, and graph neural networks have proven to be a powerful tool for exploring material structure-property relationships and discovering new materials. Unlike traditional molecular graphs, crystal graphs require consideration of periodic invariance and modes. In addition, MOF structures such as covalent bonds, functional groups, and global structures impact adsorption performance in different ways. However, redundant atomic interactions can disrupt training accuracy, potentially leading to overfitting. In this paper, we propose a multi-scale crystal graph for describing periodic crystal structures, modeling interatomic interactions at different scales while preserving periodicity invariance. We also propose a multi-head attention crystal graph network in multi-scale graphs (MHACGN-MS), which learns structural characteristics by focusing on interatomic interactions at different scales, thereby reducing interference from redundant interactions. Using MOF adsorption for gases as an example, we demonstrate that MHACGN-MS outperforms traditional graph neural networks in predicting multi-component gas adsorption. We also visualize attention scores to validate effective learning and demonstrate the model’s interpretability.

Keywords

Metal-organic frameworks / Multi-head attention score / Graph neural network / Adsorption

Cite this article

Download citation ▾

Lujun Li, Haibin Yu. A Multi-Scale Graph Neural Network for the Prediction of Multi-Component Gas Adsorption. Engineering, 2025, 52(9): 102-111 DOI:10.1016/j.eng.2025.08.012

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Alezi D, Belmabkhout Y, Suyetin M, Bhatt PM, Weselin´ ski LJ, Solovyeva V, et al. MOF crystal chemistry paving the way to gas storage needs: aluminumbased soc-MOF for CH₄, O₂, and CO₂ storage. J Am Chem Soc 2015; 137 (41):13308-18.

[2]	Listyarini RV, Gamper J, Hofer TS. Storage and diffusion of carbon dioxide in the metal organic framework MOF 5—a semi-empirical molecular dynamics study. J Phys Chem B 2023; 127(43):9378-89.

[3]	Zhang L, Song L, Meng LL, Guo YN, Zhu XY, Qin LZ, et al. Anionic Ni-based metal-organic framework with Li(I) cations in the pores for efficient C₂H₂/CO₂ separation. ACS Appl Mater Interfaces 2024; 16(1):847-52.

[4]	Fu XP, Le XY, Xiao YH, Zeng DM, Zhou KA, Huang L, et al. Cucurbituril-shaped Cd18(triazolate) 12 unit-based metal-organic framework exhibiting an C₂H₂/CO₂ separation ability. Inorg Chem 2023; 62(37):15031-8.

[5]	Eskemech A, Chand H, Karmakar A, Krishnan V, Koner RR. Zn-MOF as a single catalyst with dual Lewis acidic and basic reaction sites for CO₂ fixation. Inorg Chem 2024; 63(8):3757-68.

[6]	Wang Q, Wang H, Ren X, Pang R, Zhao X, Zhang L, et al. Synergetic role of thermal catalysis and photocatalysis in CO 2 reduction on Cu2/MoS2. J Phys Chem Lett 2023; 14(38):8421-7.

[7]	Muddassir M, Alarifi A, Abduh NAY, Saeed WS, Karami AM, Afzal M. Multifunctional Zn(II) coordination polymer as highly selective fluorescent sensor and adsorbent for dyes. Int J Mol Sci 2023; 24(10):8512.

[8]	Casanova-Chafer J, Garcia-Aboal R, Llobet E, Atienzar P. Enhanced CO₂ sensing by oxygen plasma-treated perovskite-graphene nanocomposites. ACS Sens 2024; 9(2):830-9.

[9]	Zhang H, Snurr RQ. Computational study of water adsorption in the hydrophobic metal-organic framework ZIF-8: adsorption mechanism and acceleration of the simulations. J Phys Chem C 2017; 121(43):24000-10.

[10]	Chen C, Zhang M, Zhang W, Bai J. Stable amide-functionalized metal-organic framework with highly selective CO₂ adsorption. Inorg Chem 2019; 58 (4):2729-35.

[11]	Gülsoy Z, Sezginel KB, Uzun A, Keskin S, Yıldırım R. Analysis of CH₄ uptake over metal-organic frameworks using data-mining tools. ACS Comb Sci 2019; 21 (4):257-68.

[12]	Fernandez M, Woo TK, Wilmer CE, Snurr RQ. Large-scale quantitative structure-property relationship (QSPR) analysis of methane storage in metal-organic frameworks. J Phys Chem C 2013; 117(15):7681-9.

[13]	Qiao Z, Yan Y, Tang Y, Liang H, Jiang J. Metal-organic frameworks for xylene separation: from computational screening to machine learning. J Phys Chem C 2021; 125(14):7839-48.

[14]	Shi K, Li Z, Anstine DM, Tang D, Colina CM, Sholl DS, et al. Two-dimensional energy histograms as features for machine learning to predict adsorption in diverse nanoporous materials. J Chem Theory Comput 2023; 19(14):4568-83.

[15]	Deringer VL, Bartók AP, Bernstein N, Wilkins DM, Ceriotti M, Csány G. Gaussian process regression for materials and molecules. Chem Rev 2021; 121 (16):10073-141.

[16]	Osaro E, Mukherjee K, Colón YJ. Active learning for adsorption simulations: evaluation, criteria analysis, and recommendations for metal-organic frameworks. Ind Eng Chem Res 2023; 62(33):13009-24.

[17]	Zhang L, Huang Q, Li L, Yan Y, Yuan X, Liang H, et al. Automatic machine learning combined with high-throughput computational screening of hydrophobic metal-organic frameworks for capture of methanol and ethanol from the air. ACS ES T Eng 2024; 4(1):115-27.

[18]	Fernandez M, Barnard AS. Geometrical properties can predict CO2 and N2 adsorption performance of metal-organic frameworks (MOFs) at low pressure. ACS Comb Sci 2016; 18(5):243-52.

[19]	Anderson R, Biong A, Gómez-Gualdrón DA. Adsorption predictions for multiple molecules in MOFs using the same deep learning model. J Chem Theory Comput 2020; 16(2):1271-83.

[20]	Goeminne R, Vanduyfhuys L, Van Speybroeck V, Verstraelen T. DFT-quality adsorption simulations in metal-organic frameworks enabled by machine learning potentials. J Chem Theory Comput 2023; 19(18):6313-25.

[21]	Zhang Z. Automated graph neural networks accelerate the screening of optoelectronic properties of metal-organic frameworks. J Phys Chem Lett 2023; 14(5):1239-45.

[22]	Chen H, Zheng Y, Li J, Li L, Wang X. AI for nanomaterials development in clean energy and carbon capture, utilization and storage (CCUS). ACS Nano 2023; 17 (11):9763-92.

[23]	Ma R, Colón YJ, Luo T. Transfer learning study of gas adsorption in metal- organic frameworks. ACS Appl Mater Interfaces 2020; 12(30):34041-8.

[24]	Park H, Kang Y, Kim J. Enhancing structure-property relationships in porous materials through transfer learning and cross-material few-shot learning. ACS Appl Mater Interfaces 2023; 15(48):56375-85.

[25]	Peng J, Schwalbe-Koda D, Akkiraju K, Xie T, Giordano L, Yu Y, et al. Human- and machine-centred designs of molecules and materials for sustainability and decarbonization. Nat Rev Mater 2022; 7(12):991-1009.

[26]	Xu K, Hu W, Leskovec J. How powerful are graph neural networks? In:Proceedings of the 7th International Conference on Learning Representations;2019 May 6-9; New Orleans, LA, USA. Red Hook: Curran Associates, Inc.; 2019. p. 1-17.

[27]	Kipf TN, Welling M. Semi-supervised classification with graph convolutional networks. In:Proceedings of the 5th International Conference on Learning Representations;2017 Apr 24-26; Toulon, France. Red Hook: Curran Associates, Inc.; 2017. p. 2713-27.

[28]	Velicˇkovic´ P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. Graph attention networks. In:Proceedings of the 6th International Conference on Learning Representations;2018 Apr 30-May 3; Vancouver, BC, Canada. Red Hook: Curran Associates, Inc.; 2018. p. 1-12.

[29]	Xie T, Grossman JC. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties. Phys Rev Lett 2018; 120(14):145301-7.

[30]	Chen C, Ye W, Zuo Y, Zheng C, Ong SP. Graph networks as a universal machine learning framework for molecules and crystals. Chem Mater 2019; 31 (9):3564-72.

[31]	Gilmer J, Schoenholz SS, Riley PF, Vinyals O, Dahl GE. Neural message passing for quantum chemistry. PMLR 2017; 70:1263-72.

[32]

Yan K, Liu Y, Lin Y, Ji S. Periodic graph transformers for crystal material property prediction. In: Koyejo S, Mohamed S, Agarwal A, Belgrave D, Cho K, Oh A, editors. Advances in neural information processing systems 35: 36th Conference on Neural Information Processing Systems; 2022 Nov 28-Dec 9; New Orleans, LA, USA. Red Hook: Curran Associates Inc.; 2022. p. 15066-80.

[33]	Chen P, Jiao R, Liu J, Liu Y, Lu Y. Interpretable graph transformer network for predicting adsorption isotherms of metal-organic frameworks. J Chem Inf Model 2022; 62(22):5446-56.

[34]	Liu X, Wang R, Wang X, Xu D. High-throughput computational screening and machine learning model for accelerated metal-organic frameworks discovery in toluene vapor adsorption. J Phys Chem C 2023; 127(23):11268-82.

[35]	Chung YG, Haldoupis E, Bucior BJ, Haranczyk M, Lee S, Zhang H, et al. Advances, updates, and analytics for the computation-ready, experimental metal- organic framework database:CoRE MOF 2019. J Chem Eng Data 2019; 64 (12):5985-98.

[36]	Thornton AW, Simon CM, Kim J, Kwon O, Deeg KS, Konstas K, et al. Materials genome in action: identifying the performance limits of physical hydrogen storage. Chem Mater 2017; 29(7):2844-54.

[37]	Willems TF, Rycroft CH, Kazi M, Meza JC, Haranczyk M. Algorithms and tools for high-throughput geometry-based analysis of crystalline porous materials. Microporous Mesoporous Mater 2012; 149(1):134-41.

[38]	Do D, Do H. Pore characterization of carbonaceous materials by DFT and GCMC simulations: a review. Adsorpt Sci Technol 2003; 21(5):389-423.

[39]	Dubbeldam D, Calero S, Ellis DE, Snurr RQ. RASPA: molecular simulation software for adsorption and diffusion in flexible nanoporous materials. Mol Simul 2016; 42(2):81-101.

[40]	Rappe AK, Casewit CJ, Colwell KS, Goddard III WAI, Skiff WM. UFF, a full periodic table force field for molecular mechanics and molecular dynamics simulations. J Am Chem Soc 1992; 114(25):10024-35.

[41]	Sumer Z, Keskin S. Ranking of MOF adsorbents for CO₂ separations: a molecular simulation study. Ind Eng Chem Res 2016; 55(39):10404-19.

[42]	Sezginel KB, Uzun A, Keskin S. Multivariable linear models of structural parameters to predict methane uptake in metal-organic frameworks. Chem Eng Sci 2015; 124(14):125-34.

[43]	Basdogan Y, Keskin S. Simulation and modelling of MOFs for hydrogen storage. CrstEngComm 2015; 17(2):261-75.

[44]	Potoff JJ, Siepmann JI. Vapor-liquid equilibria of mixtures containing alkanes, carbon dioxide, and nitrogen. AIChE J 2001; 47(7):1676-82.

[45]	Bobbitt NS, Shi K, Bucior BJ, Chen H, Tracy-Amoroso N, Li Z, et al. MOFX-DB: an online database of computational adsorption data for nanoporous materials. J Chem Eng Data 2023; 68(2):483-98.

[46]	Boyd PG, Chidambaram A, García-Díez E, Ireland CP, Daff TD, Bounds R, et al. Data-driven design of metal-organic frameworks for wet flue gas CO₂ capture. Nature 2019; 576(7786):253-6.