Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization

We have developed a machine learning model for critical cooling rates for metallic glasses based on computational properties, supporting in-silico screening for desired Rc values and significantly reducing reliance on time-consuming laboratory work. We compare results for features derived from easy-...

Full description

Saved in:
Bibliographic Details
Main Authors: Lane E. Schultz, Benjamin Afflerbach, Paul M. Voyles, Dane Morgan
Format: Article
Language:English
Published: Elsevier 2025-07-01
Series:Journal of Materiomics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S235284782400203X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849338243699441664
author Lane E. Schultz
Benjamin Afflerbach
Paul M. Voyles
Dane Morgan
author_facet Lane E. Schultz
Benjamin Afflerbach
Paul M. Voyles
Dane Morgan
author_sort Lane E. Schultz
collection DOAJ
description We have developed a machine learning model for critical cooling rates for metallic glasses based on computational properties, supporting in-silico screening for desired Rc values and significantly reducing reliance on time-consuming laboratory work. We compare results for features derived from easy-to-compute functions of elemental properties to more complex physically motivated properties using ab initio, machine-learning potentials, and empirical potential molecular dynamics methods. The established approach enables property acquisition across a diverse range of alloys. Analysis of various features for 34 alloys from 20 chemical systems shows that the best model for critical cooling rates was learned from one elemental property-based feature and three simulated features. The elemental property based feature is an ideal entropy value based on alloy stoichiometry. The simulated features were acquired from estimates of energies above the convex hull, changes in heat capacity, and the fraction of icosahedra-like Voronoi polyhedra. Models were assessed through a demanding cross validation test based on repeatedly leaving out full chemical systems as test sets and had an R2 of 0.78 and a mean average error of 0.76 in units of lg(K/s). We demonstrate with Shapley additive explanation analysis that the most impactful features have physically reasonable influence on model predictions. The established methodology can be applied to other high-throughput studies of material properties of diverse compositions.
format Article
id doaj-art-c81bcb8ebd5f4f04974fca0220b0cab8
institution Kabale University
issn 2352-8478
language English
publishDate 2025-07-01
publisher Elsevier
record_format Article
series Journal of Materiomics
spelling doaj-art-c81bcb8ebd5f4f04974fca0220b0cab82025-08-20T03:44:28ZengElsevierJournal of Materiomics2352-84782025-07-0111410096410.1016/j.jmat.2024.100964Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurizationLane E. Schultz0Benjamin Afflerbach1Paul M. Voyles2Dane Morgan3Corresponding author.; University of Wisconsin-Madison, 1500 Engineering Drive, Madison, WI, 53706, USAUniversity of Wisconsin-Madison, 1500 Engineering Drive, Madison, WI, 53706, USAUniversity of Wisconsin-Madison, 1500 Engineering Drive, Madison, WI, 53706, USAUniversity of Wisconsin-Madison, 1500 Engineering Drive, Madison, WI, 53706, USAWe have developed a machine learning model for critical cooling rates for metallic glasses based on computational properties, supporting in-silico screening for desired Rc values and significantly reducing reliance on time-consuming laboratory work. We compare results for features derived from easy-to-compute functions of elemental properties to more complex physically motivated properties using ab initio, machine-learning potentials, and empirical potential molecular dynamics methods. The established approach enables property acquisition across a diverse range of alloys. Analysis of various features for 34 alloys from 20 chemical systems shows that the best model for critical cooling rates was learned from one elemental property-based feature and three simulated features. The elemental property based feature is an ideal entropy value based on alloy stoichiometry. The simulated features were acquired from estimates of energies above the convex hull, changes in heat capacity, and the fraction of icosahedra-like Voronoi polyhedra. Models were assessed through a demanding cross validation test based on repeatedly leaving out full chemical systems as test sets and had an R2 of 0.78 and a mean average error of 0.76 in units of lg(K/s). We demonstrate with Shapley additive explanation analysis that the most impactful features have physically reasonable influence on model predictions. The established methodology can be applied to other high-throughput studies of material properties of diverse compositions.http://www.sciencedirect.com/science/article/pii/S235284782400203XGlassesMetalsAlloyPotentialMachine learningMolecular dynamics
spellingShingle Lane E. Schultz
Benjamin Afflerbach
Paul M. Voyles
Dane Morgan
Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
Journal of Materiomics
Glasses
Metals
Alloy
Potential
Machine learning
Molecular dynamics
title Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
title_full Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
title_fullStr Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
title_full_unstemmed Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
title_short Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
title_sort machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
topic Glasses
Metals
Alloy
Potential
Machine learning
Molecular dynamics
url http://www.sciencedirect.com/science/article/pii/S235284782400203X
work_keys_str_mv AT laneeschultz machinelearningmetallicglasscriticalcoolingratesthroughelementalandmolecularsimulationbasedfeaturization
AT benjaminafflerbach machinelearningmetallicglasscriticalcoolingratesthroughelementalandmolecularsimulationbasedfeaturization
AT paulmvoyles machinelearningmetallicglasscriticalcoolingratesthroughelementalandmolecularsimulationbasedfeaturization
AT danemorgan machinelearningmetallicglasscriticalcoolingratesthroughelementalandmolecularsimulationbasedfeaturization