Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Accelerated discovery of metallic glasses through iteration of machine learning and high-throughput experiments

Journal Article · · Science Advances
 [1];  [2];  [3]; ;  [4];  [5];  [6]
  1. SLAC National Accelerator Lab., Menlo Park, CA (United States). Stanford Synchrotron Radiation Lightsource (SSRL); SLAC
  2. Northwestern Univ., Evanston, IL (United States). Dept. of Materials Science and Engineering; Univ. of Chicago, IL (United States). Computational Institute
  3. Univ. of South Carolina, Columbia, SC (United States). College of Engineering and Computing
  4. Univ. of New South Wales, Sydney, NSW (Australia). School of Materials Science and Engineering
  5. National Inst. of Standards and Technology (NIST), Gaithersburg, MD (United States). Materials for Energy and Sustainable Development Group
  6. SLAC National Accelerator Lab., Menlo Park, CA (United States). Stanford Synchrotron Radiation Lightsource (SSRL)
+ Show Author Affiliations
With more than a hundred elements in the periodic table, a large number of potential new materials exist to address the technological and societal challenges we face today; however, without some guidance, searching through this vast combinatorial space is frustratingly slow and expensive, especially for materials strongly influenced by processing. We train a machine learning (ML) model on previously reported observations, parameters from physiochemical theories, and make it synthesis method–dependent to guide high-throughput (HiTp) experiments to find a new system of metallic glasses in the Co-V-Zr ternary. Experimental observations are in good agreement with the predictions of the model, but there are quantitative discrepancies in the precise compositions predicted. We use these discrepancies to retrain the ML model. The refined model has significantly improved accuracy not only for the Co-V-Zr system but also across all other available validation data. We then use the refined model to guide the discovery of metallic glasses in two additional previously unreported ternaries. Although our approach of iterative use of ML and HiTp experiments has guided us to rapid discovery of three new glass-forming systems, it has also provided us with a quantitatively accurate, synthesis method–sensitive predictor for metallic glasses that improves performance with use and thus promises to greatly accelerate discovery of many new metallic glasses. We believe that this discovery paradigm is applicable to a wider range of materials and should prove equally powerful for other materials and properties that are synthesis path–dependent and that current physiochemical theories find challenging to predict.
Research Organization:
SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States)
Sponsoring Organization:
National Science Foundation (NSF); US Department of Commerce; USDOE Office of Science (SC), Basic Energy Sciences (BES)
Grant/Contract Number:
AC02-76SF00515
OSTI ID:
1421387
Journal Information:
Science Advances, Journal Name: Science Advances Journal Issue: 4 Vol. 4; ISSN 2375-2548
Publisher:
AAASCopyright Statement
Country of Publication:
United States
Language:
English

References (44)

Thermodynamic prediction of bulk metallic glass forming alloys in ternary Zr–Cu–X (X=Ag, Al, Ti, Ga) systems journal October 2011
Technique for high axial shielding factor performance of large-scale, thin, open-ended, cylindrical Metglas magnetic shields journal July 2011
Organic Glasses with Exceptional Thermodynamic and Kinetic Stability journal January 2007
Super Plastic Bulk Metallic Glasses at Room Temperature journal March 2007
Receiver Operating Characteristic Curves and Their Use in Radiology journal October 2003
Spectral descriptors for bulk metallic glasses based on the thermodynamics of competing crystalline phases text January 2016
On-the-fly Data Assessment for High Throughput X-ray Diffraction Measurement text January 2017
Corrosion behaviour of metallic glasses journal September 1981
A Semi-Empirical Model for Tilted-Gun Planar Magnetron Sputtering Accounting for Chimney Shadowing journal December 2014
High-throughput discovery and characterization of multicomponent bulk metallic glass alloys journal November 2016
A combinatorial thin film sputtering approach for synthesizing and characterizing ternary ZrCuAl metallic glasses journal September 2007
Magnetic properties, phase evolution, and microstructure of the Co–Zr–V ribbons journal November 2013
Prediction of high-entropy stabilized solid-solution in multi-component alloys journal February 2012
Bulk metallic glasses journal June 2004
On-the-Fly Data Assessment for High-Throughput X-ray Diffraction Measurements journal May 2017
Data-Driven Review of Thermoelectric Materials: Performance and Resource Considerations journal May 2013
Random Forests journal January 2001
Machine-learning-assisted materials discovery using failed experiments journal May 2016
Spectral descriptors for bulk metallic glasses based on the thermodynamics of competing crystalline phases journal August 2016
A predictive structural model for bulk metallic glasses journal September 2015
A damage-tolerant glass journal January 2011
Ultrastable glasses from in silico vapour deposition journal January 2013
Combinatorial development of bulk metallic glasses journal April 2014
A general-purpose machine learning framework for predicting properties of inorganic materials journal August 2016
miRNALoc: predicting miRNA subcellular localizations based on principal component scores of physico-chemical properties and pseudo compositions of di-nucleotides journal September 2020
Zero‐magnetostriction and low field magnetic properties of Co‐TM‐Zr amorphous alloys (TM = V, Cr, Mo or W) journal March 1981
Sputtering of amorphous Co‐Zr and Co‐Hf films with soft magnetic properties journal April 1982
Reply to ‘‘Comment on ‘Crystallization characteristics of late transition metal‐Zr glasses around the composition M 9 0 Zr 1 0 ’ ’’ [J. Appl. Phys. 5 9 , 2364 (1986)] journal December 1986
Applications of high throughput (combinatorial) methodologies to electronic, magnetic, optical, and energy-related materials journal June 2013
On the origin of multi-component bulk metallic glasses: Atomic size mismatches and de-mixing journal August 2015
Perspective: Codesign for materials science: An optimal learning approach journal April 2016
Substrate’s surface electrons disrupt molecular self-assembly journal November 2017
Aerosols droplets controlled by catching the right waves journal November 2017
Under what conditions can a glass be formed? journal September 1969
Bulk metallic glasses journal August 2001
Thermal stability and electronic properties of amorphous Zr-Co and Zr-Ni alloys journal April 1979
Combinatorial screening for new materials in unconstrained composition space with machine learning journal March 2014
The random subspace method for constructing decision forests journal January 1998
The WEKA data mining software: an update journal November 2009
Precipitation of amorphous SiO2 particles and their properties journal March 2011
High Strength Bulk Amorphous Alloys with Low Critical Cooling Rates (Overview) journal January 1995
Technique for high axial shielding factor performance of large-scale, thin, open-ended, cylindrical Metglas magnetic shields text January 2011
A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials text January 2016
Python-Ternary: Ternary Plots In Python software December 2015

Cited By (53)

Accelerated Search for BaTiO 3 ‐Based Ceramics with Large Energy Storage at Low Fields Using Machine Learning and Experimental Design journal August 2019
A deep-learning technique for phase identification in multiphase inorganic compounds using synthetic XRD powder patterns journal January 2020
Accelerating small-angle scattering experiments on anisotropic samples using kernel density estimation journal February 2019
A Kriging-Based Approach to Autonomous Experimentation with Applications to X-Ray Scattering journal August 2019
Physical descriptor for the Gibbs energy of inorganic crystalline solids and temperature-dependent materials chemistry text January 2018
Recent Advances in Metallic Glass Nanostructures: Synthesis Strategies and Electrocatalytic Applications journal December 2018
Data‐Driven Materials Science: Status, Challenges, and Perspectives journal September 2019
Data‐Driven Materials Science: Status, Challenges, and Perspectives journal November 2019
Functional Applications of Metallic Glasses in Electrocatalysis journal April 2019
Synchrotron Big Data Science journal September 2018
Rapid Discovery of Ferroelectric Photovoltaic Perovskites and Material Descriptors via Machine Learning journal May 2019
Computational Platform for Manufacturing Bulk Metallic Glasses Based on GFA Parameters journal September 2018
Structural Characteristic Length in Metallic Glasses journal January 2020
Physical descriptor for the Gibbs energy of inorganic crystalline solids and temperature-dependent materials chemistry journal October 2018
Unsupervised discovery of solid-state lithium ion conductors journal November 2019
Active learning in materials science with emphasis on adaptive sampling using uncertainties for targeted design journal February 2019
Automated estimation of materials parameter from X-ray absorption and electron energy-loss spectra with similarity measures journal March 2019
Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks journal May 2019
Embedding physics domain knowledge into a Bayesian network enables layer-by-layer process innovation for photovoltaics journal January 2020
Probabilistic Assessment of Glass Forming Ability Rules for Metallic Glasses Aided by Automated Analysis of Phase Diagrams journal January 2019
Can machine learning identify the next high-temperature superconductor? Examining extrapolation performance for materials discovery journal January 2018
Machine learning of optical properties of materials – predicting spectra from images and images from spectra journal January 2019
Enumeration of de novo inorganic complexes for chemical discovery and machine learning journal January 2020
Nanoinformatics, and the big challenges for the science of small things journal January 2019
A quantitative uncertainty metric controls error in neural network-driven chemical discovery journal January 2019
A universal density matrix functional from molecular orbital-based machine learning: Transferability across organic molecules journal April 2019
Screening billions of candidates for solid lithium-ion conductors: A transfer learning approach for small data journal June 2019
Machine-learning-assisted thin-film growth: Bayesian optimization in molecular beam epitaxy of SrRuO 3 thin films journal October 2019
Model-driven design of bioactive glasses: from molecular dynamics through machine learning journal December 2019
Prediction and optimization of epoxy adhesive strength from a small dataset through active learning journal October 2019
Machine Learning to Instruct Single Crystal Growth by Flux Method journal May 2019
From DFT to machine learning: recent approaches to materials science–a review journal May 2019
Group-theoretical high-order rotational invariants for structural representations: Application to linearized machine learning interatomic potential journal June 2019
The random subspace method for constructing decision forests journal January 1998
Formation criterion for binary metal diboride solid solutions established through combinatorial methods journal January 2020
Machine learning in a data-limited regime: Augmenting experiments with synthetic data uncovers order in crumpled sheets journal April 2019
Machine learning for data-driven discovery in solid Earth geoscience journal March 2019
Crystal symmetry determination in electron diffraction using machine learning journal January 2020
Prediction of new iodine-containing apatites using machine learning and density functional theory journal August 2019
Machine learning prediction of elastic properties and glass-forming ability of bulk metallic glasses journal April 2019
Experiment Specification, Capture and Laboratory Automation Technology (ESCALATE): a software pipeline for automated chemical experimentation and data management journal June 2019
Materials science in the artificial intelligence age: high-throughput library generation, machine learning, and a pathway from correlations to the underpinning physics journal July 2019
Data-centric science for materials innovation journal September 2018
Artificial intelligence for materials discovery journal July 2019
Progress toward autonomous experimental systems for alloy development journal April 2019
Review on Quantum Mechanically Guided Design of Ultra-Strong Metallic Glasses text January 2020
Review on Quantum Mechanically Guided Design of Ultra-Strong Metallic Glasses journal April 2020
Machine Learning in a data-limited regime: Augmenting experiments with synthetic data uncovers order in crumpled sheets text January 2018
Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks preprint January 2018
Group-theoretical high-order rotational invariants for structural representations: Application to linearized machine learning interatomic potential text January 2019
Data-driven materials science: status, challenges and perspectives text January 2019
Prediction and optimization of epoxy adhesive strength from a small dataset through active learning text January 2019
Prediction and optimization of epoxy adhesive strength from a small dataset through active learning text January 2019

Similar Records

Discovering exceptionally hard and wear-resistant metallic glasses by combining machine-learning with high throughput experimentation
Journal Article · Mon Jan 10 00:00:00 UTC 2022 · Applied Physics Reviews · OSTI ID:2564495

Related Subjects

36 MATERIALS SCIENCE