The efficient separation of Li+ from Mg2+ in salt lake brines is a critical bottleneck for sustainable lithium extraction. Crown ether (CE)-modified graphene oxide (GO) membranes offer enhanced ion selectivity via specific host–guest recognition. However, their performance is governed by multiple structural descriptors, including CE content, interlayer spacing, asymmetric charge, and membrane inclination, which are difficult to optimize using molecular dynamics (MD) simulations alone due to high computational cost and limited quantitative predictive capability. To address this challenge, this study integrates MD simulation with machine learning (ML) to construct a high-accuracy proxy model for predicting the Li+/Mg2+ separation performance of CE-functionalized GO membranes. Using Random Forest (RF) and Extreme Gradient Boosting (XGBoost) regression models, we established quantitative relationships between four key structural descriptors and three performance indicators: water flux, Li+ permeability, and Mg2+ retention rate. RF outperforms XGBoost, achieving high test accuracy. Feature importance reveals distinct mechanisms: water flux is governed by interlayer spacing and membrane inclination, Li+ permeability is co-determined by interlayer spacing and CE number. Mg2+ retention depends mainly on CE grafts and non-uniform charge distribution, reflecting synergy between Donnan effect and specific recognition. Moreover, interactive effects among structural parameters are identified, CE number couples with spacing to enhance Li+ permeability, and with non-uniform charge to boost Mg2+ retention, providing quantitative evidence for the proposed separation mechanisms. Multi-objective optimization yields two membrane schemes, the optimally balanced design achieves both high selectivity and competitive flux, showing strong application potential. This study not only overcomes the limitations of conventional MD simulations but also establishes a data-driven framework for the rational design and efficient optimization of high-performance lithium-selective membranes.
- Open Access
- Article
Machine Learning-Driven Prediction of Li+/Mg2+ Separation Performance in Crown Ether-Modified Graphene Oxide Membranes
- Mengmeng Ge 1,
- Chunlei Wei 2,
- Yi Song 1,
- Timing Fang 1,*,
- Xiaomin Liu 1,*
Author Information
Received: 18 May 2026 | Revised: 15 Jun 2026 | Accepted: 16 Jun 2026 | Published: 25 Jun 2026
Abstract
Keywords
lithium-magnesium separation | molecular dynamics simulation | machine learning | structure–performance relationship
References
- 1.
Wei, C.; Wang, Y.; Ding, Z.; et al. A Universal Strategy toward Low-Cost Aqueous Sulfur–Iodine Batteries. Adv. Funct. Mater. 2023, 33, 2212644.
- 2.
Chen, K.; Li, F.; Wei, T.; et al. An Interlayer-Based Positive Charge Compensation Strategy for the Preparation of Highly Selective Mg2+/Li+ Separation Nanofiltration Membranes. J. Membr. Sci. 2023, 684, 121882.
- 3.
Wang, Q.; Song, J.; Gao, X.; et al. Carbon Nanotube Membranes for the Separation of Li+ and Mg2+ Ions: Effect of Functional Groups with Charges. Desalination 2022, 540, 115996.
- 4.
Grosjean, C.; Miranda, P.H.; Perrin, M.; et al. Assessment of World Lithium Resources and Consequences of Their Geographic Distribution on the Expected Development of the Electric Vehicle Industry. Renew. Sust. Energ. Rev. 2012, 16, 1735–1744.
- 5.
Razmjou, A.; Asadnia, M.; Hosseini, E.; et al. Design Principles of Ion Selective Nanostructured Membranes for the Extraction of Lithium Ions. Nat. Commun. 2019, 10, 5793.
- 6.
Geng, H.; Peng, Y.; Qu, L.; et al. Structure Design and Composition Engineering of Carbon-Based Nanomaterials for Lithium Energy Storage. Adv. Energy Mater. 2020, 10, 1903030.
- 7.
Wang, W.; Wang, C.; Huang, R.; et al. Boosting Lithium/Magnesium Separation Performance of Selective Electrodialysis Membranes Regulated by Enamine Reaction. Water Res. 2025, 268, 122729.
- 8.
Zhai, J.; Balogun, A.; Bhattacharjee, S.; et al. Nanofiltration as Pretreatment for Lithium Recovery from Salt Lake Brine. J. Membr. Sci. 2024, 710, 123150.
- 9.
Peng, H.Y.; Lau, S.K.; Yong, W.F. Recent Advances of Thin Film Composite Nanofiltration Membranes for Mg2+/Li+ Separation. Adv. Membr. 2024, 4, 100093.
- 10.
Li, X.; Mo, Y.; Qing, W.; et al. Membrane-Based Technologies for Lithium Recovery from Water Lithium Resources: A Review. J. Membr. Sci. 2019, 591, 117317.
- 11.
Zhang, C.; Mu, Y.; Zhao, S.; et al. Lithium Extraction from Synthetic Brine with High Mg2+/Li+ Ratio Using the Polymer Inclusion Membrane. Desalination 2020, 496, 114710.
- 12.
Zhang, Y.; Wang, L.; Sun, W.; et al. Membrane Technologies for Li+/Mg2+ Separation from Salt-Lake Brines and Seawater: A Comprehensive Review. J. Ind. Eng. Chem. 2020, 81, 7–23.
- 13.
Jie, K.; Onishi, N.; Schott, J.A.; et al. Transforming Porous Organic Cages into Porous Ionic Liquids Via a Supramolecular Complexation Strategy. Angew. Chem. Int. Ed. 2020, 59, 2268–2272.
- 14.
Christy, F.A.; Shrivastav, P.S. Conductometric Studies on Cation-Crown Ether Complexes: A Review. Crit. Rev. Anal. Chem. 2011, 41, 236–269.
- 15.
Ali, M.; Ahmed, I.; Ramirez, P.; et al. Lithium Ion Recognition with Nanofluidic Diodes through Host–Guest Complexation in Confined Geometries. Anal. Chem. 2018, 90, 6820–6826.
- 16.
Ge, M.; Wei, C.; Fang, T.; et al. Molecular Insight into the Separation Mechanism of Crown Ether-Based Channels for Lithium Extraction. Sep. Purif. Technol. 2024, 338, 126415.
- 17.
Ge, M.; Wang, D.; Wei, C.; et al. Insight into the Separation Mechanisms of MXene@PSS Nanochannels for High-Efficiency Lithium Extraction. Sep. Purif. Technol. 2025, 363, 132082.
- 18.
Ge, M.; Wei, C.; Meng, L.; et al. Insight into the Separation Mechanisms of High-Efficiency Lithium Extraction Based on Janus Membrane with Heterogeneous Charge. Chem. Eng. Sci. 2026, 326, 123504.
- 19.
Liang, L.; Lu, D.; Qin, Y.; et al. Machine Learning in Membrane Science: Bridging Materials, Structures, and Performance for Next-Generation Membrane Design. Sep. Purif. Technol. 2025, 369, 133091.
- 20.
Ji, Z.; Guan, H.; Wang, M.; et al. Polymeric Membrane Concentration of Lithium-Magnesium Solution for Sustainable Resource Recovery with Machine Learning. Water Res. 2025, 287, 124438.
- 21.
Cao, J.; Xu, Z.; Wei, M.; et al. New Insights into Li+/Mg2+ Separation by a Cnt Model Membrane Via Coupling High-Throughput Simulations and Machine Learning. J. Membr. Sci. 2026, 738, 124870.
- 22.
Wu, T.; Zhang, J.; Yan, Q.; et al. Machine Learning in the Design and Performance Prediction of Organic Framework Membranes: Methodologies, Applications, and Industrial Prospects. Membranes 2025, 15, 178.
- 23.
Lu, D.; Ma, X.; Lu, J.; et al. Ensemble Machine Learning Reveals Key Structural and Operational Features Governing Ion Selectivity of Polyamide Nanofiltration Membranes. Desalination 2023, 564, 116748.
- 24.
Sun, J.-O.; Hua, T.-W.; Guan, Y.-F.; et al. Predicting and Understanding the Performance of Polyamide Nanofiltration Membrane for Li/Mg Selective Separation Based on Machine Learning. Water Res. 2025, 285, 124140.
- 25.
Bonnet, N.; Marzari, N. Ion Sieving in Two-Dimensional Membranes from First Principles. ACS Nano 2025, 19, 8552–8560.
- 26.
Ignacz, G.; Bader, L.; Beke, A.K.; et al. Machine Learning for the Advancement of Membrane Science and Technology: A Critical Review. J. Membr. Sci. 2025, 713, 123256.
- 27.
Wu, J.; Li, N.; Liu, S.; et al. Graphene Oxide Membranes with a Confined Mass Transfer Effect for Li+/Mg2+ Separation: A Molecular Dynamics Study. Phys. Chem. Chem. Phys. 2022, 24, 26011–26022.
- 28.
Zhang, N.; Yu, H.; Zhang, J.; et al. Pressure-Driven Li+/Mg2+ Selective Permeation through Size-Sieving Nanochannels: The Role of the Second Hydration Shell. Sep. Purif. Technol. 2023, 327, 124818.
- 29.
Plimpton, S. Fast Parallel Algorithms for Short-Range Molecular Dynamics. J. Comput. Phys. 1995, 117, 1–19.
- 30.
Berendsen, H.J.C.; Grigera, J.R.; Straatsma, T.P. The Missing Term in Effective Pair Potentials. J. Phys. Chem. 1987, 91, 6269–6271.
- 31.
Li, P.; Song, L.F.; Merz, K.M., Jr. Systematic Parameterization of Monovalent Ions Employing the Nonbonded Model. J. Chem. Theory Comput. 2015, 11, 1645–1657.
- 32.
Li, P.; Roberts, B.P.; Chakravorty, D.K.; et al. Rational Design of Particle Mesh Ewald Compatible Lennard-Jones Parameters for +2 Metal Cations in Explicit Solvent. J. Chem. Theory Comput. 2013, 9, 2733–2748.
- 33.
NosÉ, S. A Molecular Dynamics Method for Simulations in the Canonical Ensemble. Mol. Phys. 2002, 100, 191–198.
- 34.
Pirbonyeh, M.A.; Shayegan, M.A. Influence of Feature Normalization Methods on Transfer Learning—A Comparison Study. Multimed. Tools Appl. 2025, 84, 48811–48847.
- 35.
Mathivanan, N.M.N.; Xian, E.F.Z.; Xi, D.F.Y.; et al. Impact of Feature Standardization on Heart Disease Prediction: A Comparative Analysis of Logistic Regression and Support Vector Machine Models. Malays. J. Comput. 2025, 10, 2159–2175.
- 36.
Bui, N.T.; Savova, G.K.; Wang, L. In Assessing the Macro and Micro Effects of Random Seeds on Fine-Tuning Large Language Models, Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, Mumbai, India, 20–24 December 2025; The Asian Federation of Natural Language Processing and The Association for Computational Linguistics: Mumbai, India, 2025; pp. 41–46.
- 37.
Zhang, Y.; Han, B.; Song, Q.; et al. Revealing Key Structural and Operating Parameters on Salt/Dye Separation of Loose Nanofiltration Membrane by Ensemble Machine Learning. J. Membr. Sci. 2025, 732, 124274.
- 38.
Gao, K.; Deng, S.; Liu, S.; et al. Interpretable Machine Learning for Predicting Separation Performance of Mofs Membrane and Suppressing Permeability-Micropollutant Rejection Trade-Off. J. Membr. Sci. 2026, 748, 125345.
- 39.
Altman, N.; Krzywinski, M. Ensemble Methods: Bagging and Random Forests. Nat. Methods 2017, 14, 933–934.
- 40.
Schalk, D.; Bischl, B.; Rügamer, D. Accelerated Componentwise Gradient Boosting Using Efficient Data Representation and Momentum-Based Optimization. J. Comput. Graph. Stat. 2023, 32, 631–641.
- 41.
Chicco, D.; Warrens, M.J.; Jurman, G. The Coefficient of Determination R-Squared Is More Informative Than Smape, Mae, Mape, Mse and Rmse in Regression Analysis Evaluation. PeerJ Comput. Sci. 2021, 7, e623.
- 42.
Lundberg, S.M.; Lee, S.-I. A Unified Approach to Interpreting Model Predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems; Curran Associates Inc.: Long Beach, CA, USA, 2017; pp. 4768–4777.
- 43.
Aas, K.; Jullum, M.; Løland, A. Explaining Individual Predictions When Features Are Dependent: More Accurate Approximations to Shapley Values. Artif. Intell. 2021, 298, 103502.
- 44.
Yogarathinam, L.T.; Abba, S.I.; Usman, J.; et al. Interpretable Shap-Based Machine Learning-Assisted Design for Selecting Ultrafiltration Membranes in Protein-Laden Phosphate Wastewater. Clean. Chem. Eng. 2025, 11, 100187.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Suite 4002 Level 4, 447 Collins Street, Melbourne, Victoria 3000, Australia
General Inquiries: info@sciltp.com