A critical review of the machine learning guided design of metallic glasses for superior glass-forming ability

Abstract

The discovery of novel metallic glasses (MGs) with high glass-forming ability (GFA) has been an important area of active research for years in materials science and engineering. Unfortunately, the traditional approach based on trial-and-error methods is inefficient, time consuming and costly. Therefore, machine learning (ML) has recently drawn significant research interest as an alternative approach for the development of MGs. In this review, we discuss the current progress regarding the ML guided design of MGs from a variety of perspectives, including the GFA database, data representation, ML algorithms and numerical evaluation. Furthermore, we consider the challenges facing this field, including the scarcity and quality of GFA data, the development of physics informed data descriptors, the selection of appropriate algorithms and the necessity for experimental validation. We also briefly discuss possible solutions to tackle these challenges.

Keywords

Alloy design / metallic glasses / machine learning / glass-forming ability / data featurization

References

[1] Klement W,Duwez P.Non-crystalline structure in solidified gold-silicon alloys.Nature1960;187:869-70

[2] Wang WH.The elastic properties, elastic models and elastic perspectives of metallic glasses.Prog Mater Sci2012;57:487-656

[3] Ashby M.Metallic glasses as structural materials.Scr Mater2006;54:321-6

[4] Scully JR,Payer JH.Corrosion and related mechanical properties of bulk metallic glasses.J Mater Res2006;22:302-13

[5] Sharma A.Review of the recent development in metallic glass and its composites.Metals2021;11:1933

[6] Inoue A.Stabilization of metallic supercooled liquid and bulk amorphous alloys.Acta Mater2000;48:279-306

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

[8] Sun YT,Li MZ.Machine learning approach for prediction and understanding of glass-forming ability.J Phys Chem Lett2017;8:3434-9

[9] Dasgupta A,Mack C.Probabilistic assessment of glass forming ability rules for metallic glasses aided by automated analysis of phase diagrams.Sci Rep2019;9:357 PMCID:PMC6344582

[10] Peng L,Zhao M.Determination of glass forming ability of bulk metallic glasses based on machine learning.Comput Mater Sci2021;195:110480

[11] Xiong J,Zhang T.Machine learning prediction of glass-forming ability in bulk metallic glasses.Comput Mater Sci2021;192:110362

[12] Afflerbach BT,Perepezko JH,Szlufarska I.Molecular simulation-derived features for machine learning predictions of metal glass forming ability.Comput Mater Sci2021;199:110728

[13] Keong K,Malinov S.Artificial neural network modelling of crystallization temperatures of the Ni-P based amorphous alloys.Mater Sci Eng A2004;365:212-8

[14] Cai A,Liu Y,Tan J.Artificial neural network modeling of reduced glass transition temperature of glass forming alloys.Appl Phys Lett2008;92:111909

[15] Cai AH,An WK.Prediction of critical cooling rate for glass forming alloys by artificial neural network.Mater Des2013;52:671-6

[16] Deng B.Critical feature space for predicting the glass forming ability of metallic alloys revealed by machine learning.Chem Phys2020;538:110898

[17] Mastropietro DG.Design of Fe-based bulk metallic glasses for maximum amorphous diameter (Dmax) using machine learning models.Comput Mater Sci2021;188:110230

[18] Majid A,Tariq NUH.Modeling glass-forming ability of bulk metallic glasses using computational intelligent techniques.Appl Soft Comput J2015;28:569-78

[19] Li J,Zekiy AO.Correlative study between elastic modulus and glass formation in ZrCuAl(X) amorphous system using a machine learning approach.Appl Phys A2021;127

[20] Samavatian M,Samavatian V.Discovery of novel quaternary bulk metallic glasses using a developed correlation-based neural network approach.Comput Mater Sci2021;186:110025

[21] Xiong J,Shi S.Machine learning prediction of elastic properties and glass-forming ability of bulk metallic glasses.MRS Commun2019;9:576-85

[22] Xiong J,Zhang T.A machine-learning approach to predicting and understanding the properties of amorphous metallic alloys.Mater Des2020;187:108378

[23] Ward L,Stevick J,Aykol M.A machine learning approach for engineering bulk metallic glass alloys.Acta Mater2018;159:102-11

[24] Liu X,He Q.Machine learning-based glass formation prediction in multicomponent alloys.Acta Mater2020;201:182-90

[25] Chen T,Elveny M.Engineering of novel Fe-based bulk metallic glasses using a machine learning-based approach.Arab J Sci Eng2021;46:12417-25

[26] Li Z,Lei S,Liu X.Predicting the glass formation of metallic glasses using machine learning approaches.Comput Mater Sci2021;197:110656

[27] Zhang Y,Sha Z.A two-step fused machine learning approach for the prediction of glass-forming ability of metallic glasses.J Alloys Compd2021;875:160040

[28] Cai A,Liu Y,Tan J.Artificial neural network modeling for undercooled liquid region of glass forming alloys.Comput Mater Sci2010;48:109-14

[29] Jeon J,Kim H.Inverse design of Fe-based bulk metallic glasses using machine learning.Metals2021;11:729

[30] Xiong J,Zhang T.Machine learning of phases and mechanical properties in complex concentrated alloys.J Mater Sci Technol2021;87:133-42

[31] Zhou ZQ,Liu XD.Rational design of chemically complex metallic glasses by hybrid modeling guided machine learning.npj Comput Mater2021;7:138

[32] Zhang Y.A strategy to apply machine learning to small datasets in materials science.npj Comput Mater2018;4:25

[33] Ren F,Williams T.Accelerated discovery of metallic glasses through iteration of machine learning and high-throughput experiments.Sci Adv2018;4:eaaq1566 PMCID:PMC5898831

[34] Ren B,Deng R.A new criterion for predicting the glass-forming ability of alloys based on machine learning.Comput Mater Sci2021;189:110259

[35] Kawazoe Y,Louzguine DV.Phase diagrams and physical properties of nonequilibrium alloys. 1st ed. Berlin Heidelberg: Springer; 1997.

[36] Kawazoe Y.Nonequilibrium phase diagrams of ternary amorphous alloys. Berlin Heidelberg: Springer; 1997.

[37] Lu Z.A new glass-forming ability criterion for bulk metallic glasses.Acta Mater2002;50:3501-12

[38] Lu Z,Liu C.Recent progress in quantifying glass-forming ability of bulk metallic glasses.Intermetallics2007;15:618-24

[39] Long Z,Ding Y,Xie G.A new criterion for predicting the glass-forming ability of bulk metallic glasses.J Alloys Compd2009;475:207-19

[40] Guo S.Phase stability in high entropy alloys: formation of solid-solution phase or amorphous phase.Prog Nat Sci Mater Int2011;21:433-46

[41] Tripathi MK,Ganguly S.Multivariate analysis and classification of bulk metallic glasses using principal component analysis.Comput Mater Sci2015;107:79-87

[42] Chen T,Sajjadifar S.Engineering of new Mg-based glassy compositions by a computational intelligence model.Mater Lett2021;290:129441

[43] Johnson WL,Demetriou MD.Quantifying the origin of metallic glass formation.Nat Commun2016;7:10313 PMCID:PMC4735709

[44] Ghiringhelli LM,Levchenko SV,Scheffler M.Big data of materials science: critical role of the descriptor.Phys Rev Lett2015;114:105503

[45] Mizutani U.Hume-Rothery rules for structurally complex alloy phases.MRS Bull2012;37:169-169

[46] Greer AL.Metallic glasses.Science1995;267:1947-53

[47] Zhou Z,He Q,Li F.Machine learning guided appraisal and exploration of phase design for high entropy alloys.npj Comput Mater2019;5:128

[48] Raschka S.MLxtend: providing machine learning and data science utilities and extensions to Python’s scientific computing stack.JOSS2018;3:638

[49] Palma-mendoza R,de-Marcos L.Distributed ReliefF-based feature selection in Spark.Knowl Inf Syst2018;57:1-20

[50] Feng S,Zhou H,Lu Z.A general and transferable deep learning framework for predicting phase formation in materials.npj Comput Mater2021;7:10

[51] Suryanarayana C,Inoue A.A critical analysis of the glass-forming ability of alloys.J Non Cryst Solids2009;355:355-60

[52] Inoue A.Recent development and application products of bulk glassy alloys.Acta Mater2011;59:2243-67