FIND: a forward–inverse navigation and discovery platform for hydrogen storage alloys powered by data-driven machine learning

Abstract

High-performance solid-state hydrogen storage alloys are among the key factors enabling the widespread application of hydrogen energy. However, current materials still face challenges such as limited hydrogen storage capacity and excessive thermodynamic stability, which urgently need to be addressed. In this work, we constructed a large-scale solid-state hydride database, encompassing over 1,000 alloy systems and more than 6,000 valid data records. By integrating alloying strategies with machine learning (ML) techniques, the Magpie tool was utilized for feature generation, and a multi-objective regression model was developed to simultaneously predict absorption/desorption plateau pressure, enthalpy change, entropy change, and maximum hydrogen storage capacity using various ML algorithms. Furthermore, we achieved the inverse design of solid-state hydrogen storage materials using a variational autoencoder. By integrating the forward prediction and inverse design models, we developed a forward–inverse navigation and discovery platform for hydrogen storage alloys powered by data-driven ML: FIND. The forward module enables rapid prediction of absorption and desorption properties based on alloy composition and testing temperature. Building upon this, an advanced function allows fast prediction for multicomponent systems with flexible molar ratios. Subsequently, the inverse module facilitates the screening of potential alloy candidates based on user-defined target properties. Finally, the predictive models were integrated with a genetic algorithm to optimize alloy compositions within the Mg–Ni–La–Ce and Mg–Ni–La systems. Multiple novel high-performance alloy candidates were identified, providing a powerful tool and methodological foundation for high-throughput screening and intelligent development of hydrogen storage materials.

Keywords

Hydrogen storage alloys / metal hydride / machine learning / genetic algorithm

References

[1] Sakintuna B,Hirscher M.Metal hydride materials for solid hydrogen storage: a review.Int J Hydrogen Energy2007;32:1121-40

[2] Schlapbach L.Hydrogen-storage materials for mobile applications.Nature2001;414:353-8

[3] Zhang J,Qi J.Advanced Mg-based materials for energy storage: fundamental, progresses, challenges and perspectives.Prog Mater Sci2025;148:101381

[4] Liu H,Zhou C,Liu Y.Hydrogen storage properties of Ti-Fe-Zr-Mn-Nb alloys.J Alloys Compd2023;938:168466

[5] Zhang Y,Gao J,Wei X.Effect of Sm content on activation capability and hydrogen storage performances of TiFe alloy.Int J Hydrogen Energy2021;46:24517-30

[6] Dematteis EM,Capurso G,Cuevas F.Fundamental hydrogen storage properties of TiFe-alloy with partial substitution of Fe by Ti and Mn.J Alloys Compd2021;874:159925

[7] Ali W,Li Z.Effects of Cu and Y substitution on hydrogen storage performance of TiFe_0.86Mn_0.1Y_0.1-xCu_x.Int J Hydrogen Energy2017;42:16620-31

[8] Chen Z,Chen L.Development of Ti–Cr–Mn–Fe based alloys with high hydrogen desorption pressures for hybrid hydrogen storage vessel application.Int J Hydrogen Energy2013;38:12803-10

[9] Zepon G,Zlotea C,Champion Y.Thermodynamic modelling of hydrogen-multicomponent alloy systems: calculating pressure-composition-temperature diagrams.Acta Mater2021;215:117070

[10] Lototsky M,Marinin V.Modelling of phase equilibria in metal–hydrogen systems.J Alloys Compd2003;356-7:27-31

[11] Zhou Z,Ge J,Dai Z.Mathematical modeling of the PCT curve of hydrogen storage alloys.Int J Hydrogen Energy1994;19:269-73

[12] Loh SM,Walker GS.Substitutional effect of Ti-based AB₂ hydrogen storage alloys: a density functional theory study.Int J Hydrogen Energy2023;48:13227-35

[13] Xie X,Chen C.First-principles studies in Mg-based hydrogen storage Materials: a review.Energy2020;211:118959

[14] Sun H,Zhao F,Zhang Y.Interactions of Y and Cu on Mg₂Ni type hydrogen storage alloys: a study based on experiments and density functional theory calculation.Int J Hydrogen Energy2020;45:28974-84

[15] Hu J,Jiang M.A DFT study of hydrogen storage in high-entropy alloy TiZrHfScMo.Nanomaterials2019;9:461 PMCID:PMC6474085

[16] Ward L,Choudhary A.A general-purpose machine learning framework for predicting properties of inorganic materials.npj Comput Mater2016;2:16028

[17] Batalović K,Stojić D.Hydrogen storage in martensite Ti–Zr–Ni alloy: a density functional theory study.J Phys Chem C2013;117:26914-20

[18] Yang H,Li Y.Recent advances in kinetic and thermodynamic regulation of magnesium hydride for hydrogen storage.Rare Met2023;42:2906-27

[19] Yadav S,Mittal MK.Electrochemical hydrogen storage: achievements, emerging trends, and perspectives.Int J Energy Res2022;46:16316-35

[20] Rahnama A,Sridhar S.Machine learning based prediction of metal hydrides for hydrogen storage, part I: Prediction of hydrogen weight percent.Int J Hydrogen Energy2019;44:7337-44

[21] Zhou P,Xiao X.Machine learning in solid-state hydrogen storage materials: challenges and perspectives.Adv Mater2025;37:e2413430

[22] Lu Z,Wu Y,Ma T.Prediction and theoretical investigation of dehydrogenation enthalpy of V–Ti–Cr–Fe alloy using machine learning and density functional theory.Int J Hydrogen Energy2024;50:379-89

[23] Suwarno S,Suyuthi A.Machine learning analysis of alloying element effects on hydrogen storage properties of AB₂ metal hydrides.Int J Hydrogen Energy2022;47:11938-47

[24] Dong S,Li J,Wang L.Exploration and design of Mg alloys for hydrogen storage with supervised machine learning.Int J Hydrogen Energy2023;48:38412-24

[25] Kanti PK,Sharma P.Thermal performance enhancement of metal hydride reactor for hydrogen storage with graphene oxide nanofluid: model prediction with machine learning.Int J Hydrogen Energy2024;52:470-84

[26] Salehi K,Atashrouz S.Machine learning assisted predictions for hydrogen storage in metal-organic frameworks.Int J Hydrogen Energy2023;48:33260-75

[27] Jiang H,Li Y.Hierarchical interface engineering for advanced magnesium-based hydrogen storage: synergistic effects of structural design and compositional modification.Chem Sci2025;16:7610-36 PMCID:PMC11995415

[28] Ding Z,Yang H.Tailoring MgH₂ for hydrogen storage through nanoengineering and catalysis.J Magnes Alloys2022;10:2946-67

[29] Athul AS,Buddhiraju VS,Runkana V.Identification of stable intermetallic compounds for hydrogen storage via machine learning.Energy Storage2025;7:e70115

[30] Radhika N,Akhil U,Jen T.High entropy alloys for hydrogen storage applications: a machine learning-based approach.Results Eng2024;23:102780

[31] Halpren E,Chen ZW.Machine learning assisted design of BCC high entropy alloys for room temperature hydrogen storage.Acta Mater2024;270:119841

[32] Dangwal S,Grabowski B.Machine learning to explore high-entropy alloys with desired enthalpy for room-temperature hydrogen storage: prediction of density functional theory and experimental data.Chem Eng J2024;493:152606

[33] Zhou P,Zhu X.Machine learning enabled customization of performance-oriented hydrogen storage materials for fuel cell systems.Energy Storage Mater2023;63:102964

[34] Ahmed A.Predicting hydrogen storage in MOFs via machine learning.Patterns2021;2:100291 PMCID:PMC8276024

[35] Halevy A,Pereira F.The unreasonable effectiveness of data.IEEE Intell Syst2009;24:8-12

[36] Wilson N,Maharana PR,Joshi K.HyStor: an experimental database of hydrogen storage properties for various metal alloy classes.Int J Hydrogen Energy2024;90:460-9

[37] Hattrick-simpers JR,Corgnale C.A simple constrained machine learning model for predicting high-pressure-hydrogen-compressor materials.Mol Syst Des Eng2018;3:509-17

[38] Witman M,Grant DM.Extracting an empirical intermetallic hydride design principle from limited data via interpretable machine learning.J Phys Chem Lett2020;11:40-7

[39] Van’t Hoff JH. Studies in chemical dynamics (in French). 1884. https://www.google.com/books?id=A_VoAAAAcAAJ. (accessed 26 Aug 2025)

[40] Kingma DP. Auto-encoding variational bayes. arXiv 2013, arXiv:1312.6114. https://doi.org/10.48550/arXiv.1312.6114. (accessed 26 Aug 2025)

[41] Sanchez-Lengeling B.Inverse molecular design using machine learning: generative models for matter engineering.Science2018;361:360-5

[42] Gómez-Bombarelli R,Duvenaud D.Automatic chemical design using a data-driven continuous representation of molecules.ACS Cent Sci2018;4:268-76 PMCID:PMC5833007

[43] Lu Z,Wu Y,Xiao W.Predicting hydrogen storage capacity of V–Ti–Cr–Fe alloy via ensemble machine learning.Int J Hydrogen Energy2022;47:34583-93

[44] Rousseeuw PJ.Silhouettes: A graphical aid to the interpretation and validation of cluster analysis.J Comput Appl Math1987;20:53-65

[45] Davies DL.A cluster separation measure.IEEE Trans Pattern Anal Mach Intell1979;1:224-7

[46] Tsukahara M.Hydrogenation properties of vanadium-based alloys with large hydrogen storage capacity.Mater Trans2011;52:68-72

[47] Ge Y.Alloy selections in high-temperature metal hydride heat pump systems for industrial waste heat recovery.Energy Rep2022;8:3649-60

[48] Li YL,Zhai FQ,Qu XH.Hydrogen storage properties of Mg_1.7M_0.3Ni (M = Mg, La, Ce, Nd) hydrogen storage alloys.Adv Mater Res2012;512-5:1503-8

[49] Witman M,Ling S.Data-driven discovery and synthesis of high entropy alloy hydrides with targeted thermodynamic stability.Chem Mater2021;33:4067-76

[50] Verma A,Joshi K.Solid state hydrogen storage: decoding the path through machine learning.Int J Hydrogen Energy2024;50:1518-28

[51] Wang X,Xue X.Suction-cast strategy to enhance hydrogen storage performance of rare earth-based alloys.Int J Hydrogen Energy2025;104:220-7

[52] Araya SS,Liso V.A comprehensive review of PBI-based high temperature PEM fuel cells.Int J Hydrogen Energy2016;41:21310-44