Machine learning-driven morphology identification and classification of high-throughput functional oxide films

Abstract

Functional oxide films offer precise control over diverse properties through tunable physical characteristics and interface effects, with their functionality primarily determined by morphology. However, conventional methods are incapable of obtaining large-scale morphological data and face significant challenges in data identification and classification, which fundamentally limit the rapid assessment of thin film properties and functional screening. Herein, we establish a comprehensive morphological database of oxide films utilizing high-throughput experimental methods and develop a machine learning framework for automated identification and classification of atomic force microscopy data. Using gradient-thickness SrRuO₃ films as a representative example, this framework achieves enhanced performance through hyperparameter optimization and strategic adjustments, ultimately reaching a classification accuracy of 86.67% in independent tests, demonstrating its effectiveness in morphology analysis of functional oxide films. Furthermore, this approach shows significant potential for automated microstructure analysis of complex oxides and is expected to accelerate research on structure-property correlations in functional oxide films.

Keywords

Machine learning / atomic force microscopy / functional oxide films / SrRuO₃ / automated identification and classification

References

[1] Choo S,Liu H,James RD.From oxide epitaxy to freestanding membranes: opportunities and challenges.Sci Adv2024;10:eadq8561 PMCID:PMC11633760

[2] Wang S,Deng C.Giant electric field-induced second harmonic generation in polar skyrmions.Nat Commun2024;15:1374 PMCID:PMC10866987

[3] Yang X,Ning H.Ultralow-pressure-driven polarization switching in ferroelectric membranes.Nat Commun2024;15:9281 PMCID:PMC11519889

[4] Dou L,Ye X.Flexible high-entropy functional ceramics.Nat Commun2025;16:5915 PMCID:PMC12214991

[5] Liu K,Zhu T.Interface-controlled uniaxial in-plane ferroelectricity in Hf_0.5Zr_0.5O₂(100) epitaxial thin films.Nat Commun2025;16:7385 PMCID:PMC12339713

[6] Ma C,Huang W.Sub-nanosecond memristor based on ferroelectric tunnel junction.Nat Commun2020;11:1439 PMCID:PMC7080735

[7] Liu C,Zhang H.Low voltage-driven high-performance thermal switching in antiferroelectric PbZrO₃ thin films.Science2023;382:1265-9

[8] Yang B,Huang H.Engineering relaxors by entropy for high energy storage performance.Nat Energy2023;8:956-64

[9] Guo Y,Lu G.Remarkable flexibility in freestanding single-crystalline antiferroelectric PbZrO₃ membranes.Nat Commun2024;15:4414 PMCID:PMC11116490

[10] Qian J,He L.Topological bubble domain engineering for high strain response.Sci Adv2025;11:eadw8840

[11] Ren Z,Shao J.Ultrahigh-power-density flexible piezoelectric energy harvester based on freestanding ferroelectric oxide thin films.Nat Commun2025;16:3192 PMCID:PMC11968794

[12] Yao Y,Hu Y.Fluctuating local polarization: a generic fingerprint for enhanced piezoelectricity in Pb-based and Pb-free perovskite ferroelectrics.Nat Commun2025;16:7442 PMCID:PMC12343838

[13] Tao A,Chen S.Ferroelectric polarization and magnetic structure at domain walls in a multiferroic film.Nat Commun2024;15:6099 PMCID:PMC11271601

[14] Pan Q,Sha TT.Strain-induced tunable enhancement of piezoelectricity in a novel molecular multiferroic material.Adv Mater2025;37:e2410585

[15] Trier F,Kim J,Vila L.Oxide spin-orbitronics: spin–charge interconversion and topological spin textures.Nat Rev Mater2022;7:258-74

[16] Gao W,Wang Y,Liu J.A review of flexible perovskite oxide ferroelectric films and their application.J Materiomics2020;6:1-16

[17] Xu R,Saremi S.Kinetic control of tunable multi-state switching in ferroelectric thin films.Nat Commun2019;10:1282 PMCID:PMC6427024

[18] Fan Y,Xue Z.Hidden structural phase transition assisted ferroelectric domain orientation engineering in Hf_0.5Zr_0.5O₂ films.Nat Commun2025;16:4232 PMCID:PMC12059023

[19] Hafner J,Richheimer F.Multi-scale characterisation of a ferroelectric polymer reveals the emergence of a morphological phase transition driven by temperature.Nat Commun2021;12:152 PMCID:PMC7794429

[20] Yao S,Wen J.Electrically modulated photothermal force microscopy for revealing molecular conformation changes during polarization switching at the nanoscale.Nat Commun2025;16:6680 PMCID:PMC12280132

[21] Cai J,Xu K,Wei J.Machine learning-driven new material discovery.Nanoscale Adv2020;2:3115-30 PMCID:PMC9419423

[22] Jiang YF,Cai Y.Adaptive ferroelectric memristors with high-throughput BaTiO₃ thin films for neuromorphic computing.Mater Horiz2025;12:6928-37

[23] Koinuma H.Combinatorial solid-state chemistry of inorganic materials.Nat Mater2004;3:429-38

[24] Barber ZH.High throughput thin film materials science.Mater Sci Technol2008;24:757-70

[25] Yuan J,Gao C,Jin K.Recent advances in high-throughput superconductivity research.Supercond Sci Technol2019;32:123001

[26] Wang Z,Yin H.Data-driven materials innovation and applications.Adv Mater2022;34:e2104113

[27] Green ML,Hattrick-Simpers JR.Fulfilling the promise of the materials genome initiative with high-throughput experimental methodologies.Appl Phys Rev2017;4:011105

[28] Zhang Y,Dong Y.High-throughput scanning second-harmonic-generation microscopy for polar materials.Adv Mater2023;35:e2300348

[29] Tang M,Cheng M.High‐throughput screening thickness‐dependent resistive switching in SrTiO₃ thin films for robust electronic synapse.Adv Funct Mater2023;33:2213874

[30] Fang H,Nie F.Giant electroresistance in ferroelectric tunnel junctions via high-throughput designs: toward high-performance neuromorphic computing.ACS Appl Mater Interfaces2024;16:1015-24

[31] Zhang D,Zachman MJ.High‐throughput combinatorial approach expedites the synthesis of a lead‐free relaxor ferroelectric system.InfoMat2024;6:e12561

[32] Xiao R,Li M.Automated high-throughput atomic force microscopy single-cell nanomechanical assay enabled by deep learning-based optical image recognition.Nano Lett2024;24:12323-32

[33] Thomas-Chemin O,Boumehdi Z.Advancing high-throughput cellular atomic force microscopy with automation and artificial intelligence.ACS Nano2025;19:5045-62

[34] Gregoire JM,Miller CE,Suram SK.High-throughput synchrotron X-ray diffraction for combinatorial phase mapping.J Synchrotron Radiat2014;21:1262-8

[35] Maruyama S,Koganezawa T.High-throughput and autonomous grazing incidence X-ray diffraction mapping of organic combinatorial thin-film library driven by machine learning.ACS Comb Sci2020;22:348-55

[36] Ludwig A.Discovery of new materials using combinatorial synthesis and high-throughput characterization of thin-film materials libraries combined with computational methods.npj Comput Mater2019;5:205

[37] Kimmig J,Schubert US.Digital transformation in materials science: a paradigm change in material’s development.Adv Mater2021;33:e2004940 PMCID:PMC11469275

[38] Butler KT,Cartwright H,Walsh A.Machine learning for molecular and materials science.Nature2018;559:547-55

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

[40] Himanen L,Foster AS.Data-driven materials science: status, challenges, and perspectives.Adv Sci2019;6:1900808 PMCID:PMC6839624

[41] Hardian R,Zhang X.Artificial intelligence: the silver bullet for sustainable materials development.Green Chem2020;22:7521-8

[42] Xu X,Yan B.An electrodeposited nano-porous and neural network-like Ln@HOF film for SO₂ gas quantitative detection via fluorescent sensing and machine learning.J Mater Chem A2021;9:26391-400

[43] Miseta T,Vathy-Fogarassy Á.Surpassing early stopping: a novel correlation-based stopping criterion for neural networks.Neurocomputing2024;567:127028