Stacked machine learning for accurate and interpretable prediction of MXenes’ work function

Abstract

MXenes, with tunable compositions and rich surface chemistry, enable precise control of electronic, optical, and mechanical properties, making them promising materials in electronics and energy-related applications. In particular, the work function plays a critical role in determining their physicochemical properties. However, the accurate prediction of the work function of MXenes with machine learning (ML) remains challenging due to the lack of robust models with high accuracy and interpretability. To this end, we propose a stacked model and introduce high-quality descriptors constructed via Sure Independence Screening and Sparsifying Operator method to improve the prediction accuracy of the work function of MXenes in this work. The stacked model initially generates predictions from multiple base models, and then employs these predictions as inputs to a meta-model for secondary learning, thereby enhancing both predictive performance and generalization capability. The results show that by integrating the high-quality descriptors, the model’s performance improves significantly, yielding a coefficient of determination of 0.95 and mean absolute error of 0.2, respectively. Last but not least, we demonstrate that MXenes’ work functions are predominantly governed by their surface functional groups, where SHapley Additive exPlanations value analysis quantitatively resolves the structure–property relationship between surface functional groups and the work function of MXenes. Specifically, O terminations can lead to the highest work functions, while OH terminations result in the lowest value (over 50% reduction), and transition metals or C/N elements have a relatively smaller effect. This work achieves an optimal balance between accuracy and interpretability in ML predictions of MXenes’ work functions, providing both fundamental insights and practical tools for materials discovery.

Keywords

Machine learning / MXenes / interpretability / SHAP / SISSO / work function

References

[1] Gogotsi Y.MXenes: two-dimensional building blocks for future materials and devices.ACS Nano2021;15:5775-80

[2] Wu Y.Recent progress of MXene as an energy storage material.Nanoscale Horiz2024;9:215-32

[3] Björk J.Functionalizing MXenes by tailoring surface terminations in different chemical environments.Chem Mater2021;33:9108-18

[4] Ayodhya D.A review of recent progress in 2D MXenes: synthesis, properties, and applications.Diam Relat Mater2023;132:109634

[5] Gouveia JD,Iben Nassar K.First-principles and machine-learning approaches for interpreting and predicting the properties of MXenes.npj 2D Mater Appl2025;9:529

[6] Bhat MY,Prenger K.Frontiers of MXenes-based hybrid materials for energy storage and conversion applications.Adv Compos Hybrid Mater2025;8:1121

[7] Gokul Eswaran, S.; Rashad, M.; Santhana Krishna Kumar, A.; El-Mahdy, A. F. M. A comprehensive review of Mxene-based emerging materials for energy storage applications and future perspectives.Chem Asian J2025;20:e202401181

[8] Worku AK,Ayele DW,Teshager MA.Recent advances in MXene-based materials for high-performance metal-air batteries.Green Chem Lett Rev2024;17:2325983

[9] Ko TY,Murali G.Functionalized MXene ink enables environmentally stable printed electronics.Nat Commun2024;15:3459 PMCID:PMC11043420

[10] Yan J,Li M.High-throughput computational screening of all-MXene metal-semiconductor junctions for Schottky-Barrier-Free contacts with weak fermi-level pinning.Small2023;19:e2303675

[11] Chen J,Li Z,Lu X.Work-function-tunable MXenes electrodes to optimize p-CsCu₂I₃/n-Ca₂Nb_3-xTa_xO₁₀ junction photodetectors for image sensing and logic electronics.Adv Funct Mater2022;32:2201066

[12] Ahn S,Maleski K.A 2D titanium carbide MXene flexible electrode for high-efficiency light-emitting diodes.Adv Mater2020;32:e2000919

[13] Gregoire JM,Haber JA.Combinatorial synthesis for AI-driven materials discovery.Nat Synth2023;2:493-504

[14] Ghalati MK,El-Fallah GMAM,Dong H.Toward learning steelmaking - a review on machine learning for basic oxygen furnace process.Mater Genome Eng Adv2023;1:e6

[15] Lombardo T,El-Bouysidy H.Artificial intelligence applied to battery research: hype or reality?.Chem Rev2022;122:10899-969 PMCID:PMC9227745

[16] Shi Y,Wen J.Interpretable machine learning for stability and electronic structure prediction of Janus III-VI van der Waals heterostructures.Mater Genome Eng Adv2024;2:e76

[17] Schleder GR,Fazzio A.Exploring two-dimensional materials thermodynamic stability via machine learning.ACS Appl Mater Interfaces2020;12:20149-57

[18] Huang M,Zhu H.A comprehensive machine learning strategy for designing high-performance photoanode catalysts.J Mater Chem A2023;11:21619-27

[19] Bai X,Song P.Heterojunction of MXenes and MN₄-graphene: machine learning to accelerate the design of bifunctional oxygen electrocatalysts.J Colloid Interface Sci2024;664:716-25

[20] Tao Q,Li M.Machine learning for perovskite materials design and discovery.npj Comput Mater2021;7:23

[21] Siriwardane EMD,Perera I.Generative design of stable semiconductor materials using deep learning and density functional theory.npj Comput Mater2022;8:164

[22] Roy P,Koh SW,Choksi TS.Predicting the work function of 2D MXenes using machine-learning methods.J Phys Energy2023;5:034005

[23] Prasad Raju, N. V. D. S. S. V.; Devi, P. N. A comparative analysis of machine learning algorithms for big data applications in predictive analytics.Int J Sci Res Manag2024;12:1608-30

[24] Zhong X,Liu S,Hiszpanski A.Explainable machine learning in materials science.npj Comput Mater2022;8:204

[25] Gjerding MN,Rasmussen A.Recent progress of the Computational 2D Materials Database (C2DB).2D Mater2021;8:044002

[26] Haastrup S,Pandey M.The Computational 2D Materials Database: high-throughput modeling and discovery of atomically thin crystals.2D Mater2018;5:042002

[27] Li H,Li F,Tang Q.Investigation of dual atom doped single-layer MoS₂ for electrochemical reduction of carbon dioxide by first-principle calculations and machine-learning.J Mater Inf2023;3:25

[28] Abdullah GMS,Babur M.Boosting-based ensemble machine learning models for predicting unconfined compressive strength of geopolymer stabilized clayey soil.Sci Rep2024;14:2323 PMCID:PMC10822860

[29] Alzamzami F,El Saddik A.Light gradient boosting machine for general sentiment classification on short texts: a comparative evaluation.IEEE Access2020;8:101840-58

[30] Ouyang R,Ahmetcik E,Ghiringhelli LM.SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates.Phys Rev Mater2018;2:083802

[31] Lundberg S. A unified approach to interpreting model predictions. arXiv 2017, arXiv:1705.07874. https://doi.org/10.48550/arXiv.1705.07874. (accessed 1 Aug 2025)

[32] Sun T,Zeng L.Identifying MOFs for electrochemical energy storage via density functional theory and machine learning.npj Comput Mater2025;11:90

[33] Li S.Density functional theory for catalyst development and mechanistic insights.Nat Rev Clean Technol2025;1:602

[34] Chen L,Yao X.High-entropy alloy catalysts: high-throughput and machine learning-driven design.J Mater Inf2022;2:19

[35] Xu M,Miao X.Deep machine learning unravels the structural origin of mid-gap states in chalcogenide glass for high-density memory integration.InfoMat2022;4:e12315

[36] Pearson K.VII. Mathematical contributions to the theory of evolution. - III. Regression, heredity, and panmixia.Philos Trans R Soc Lond A1896;187:253-318

[37] Ma B,Zhao C.An interpretable machine learning strategy for pursuing high piezoelectric coefficients in (K_0.5Na_0.5)NbO₃-based ceramics.npj Comput Mater2023;9:229

[38] Wang T,Ouyang R.Nature of metal-support interaction for metal catalysts on oxide supports.Science2024;386:915-20

[39] Friedman JH.Greedy function approximation: a gradient boosting machine.Ann Statist2001;29:1189-232https://www.jstor.org/stable/2699986. (accessed 1 Aug 2025)

[40] Roy A.Support vector machine in structural reliability analysis: a review.Reliab Eng Syst Saf2023;233:109126

[41] Geurts P,Wehenkel L.Extremely randomized trees.Mach Learn2006;63:3-42

[42] Xu L,Zhu B.Mn-atomic-layered antiphase boundary enhanced ferroelectricity in KNN-based lead-free films.Nat Commun2025;16:5907 PMCID:PMC12215337

[43] Naimi AI.Stacked generalization: an introduction to super learning.Eur J Epidemiol2018;33:459-64 PMCID:PMC6089257

[44] King RD,Taylor CC.Cross-validation is safe to use.Nat Mach Intell2021;3:276

[45] Hanifi S,Zare-Behtash H.Advanced hyperparameter optimization of deep learning models for wind power prediction.Renew Energy2024;221:119700

[46] Yang Z,Zhu C.Accurate and explainable machine learning for the power factors of diamond-like thermoelectric materials.J Materiomics2022;8:633-9

[47] Altman N.Simple linear regression.Nat Methods2015;12:999-1000

[48] Khazaei M,Sasaki T,Liang Y.OH-terminated two-dimensional transition metal carbides and nitrides as ultralow work function materials.Phys Rev B2015;92:075411

[49] Lang PF.Revisiting electronegativity and electronegativity scales.J Chem Educ2025;102:424-9