Machine learning guided prediction of the yield strength and hardness of multi-principal element alloys - Research article summary

Q: What materials and properties are studied in this article?

The article focuses on Multi-Principal Element Alloys (MPEAs) and predicts room-temperature yield strength and Vickers hardness.

Q: Which feature is identified as most influential for prediction?

Valence electron concentration (VEC) is reported as the key feature dominating both yield strength and hardness.

Q: How accurate are the model predictions against experimental data?

For most cases in the experimental validation set, predicted yield strength and hardness show less than 15% error relative to experimental values.

Aug 3, 2026

Upload by Olivia Brown

Multi-Principal Element Alloys (MPEAs) offer improved yield strength, hardness, and corrosion resistance versus conventional alloys, yet compositional optimization remains difficult. This research implements machine learning models to predict room-temperature yield strength and Vickers hardness of MPEAs while quantifying predictive uncertainty. Results indicate valence electron concentration (VEC) is the dominant feature controlling both yield strength and hardness, with predicted values typically within 15% error on the experimental validation set. The model enables high-throughput screening and down-selection of promising compositions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

Background

Methods

Results

Conclusions

RESEARCH ARTICLE Machine learning guided prediction of the yield strength and hardness of multi-principal element alloys [version 1; peer review: awaiting peer review] Mohammad Fuad Nur Taufique 1, Osman Mamun 1, Ankit Roy 1, Hrishabh Khakurel2, Ganesh Balasubramanian3, Gaoyuan Ouyang4, Jun Cui4,5, Duane D. Johnson 4,5, Ram Devanathan 1 1 Pacific Northwest National Laboratory, Richland, WA, 99354, USA 2 Department of Mathematics, The University of Texas at Arlington, Arlington, TX, 76019, USA 3 Department of Mechanical Engineering and Mechanics, Lehigh University, Bethlehem, PA, 18015, USA 4Ames National Laboratory, United States Department of Energy, Ames, IA, 50011, USA 5 Department of Materials Science and Engineering, Iowa State University, Ames, IA, 50011, USA v1 First published: 09 Oct 2023, 2:9 [https://doi.org/10.12688/materialsopenres.17476.1](https://doi.org/10.12688/materialsopenres.17476.1) [Latest published:](Latest published: 09 Oct 2023)[ 09 Oct 2023](Latest published: 09 Oct 2023), 2:9 [https://doi.org/10.12688/materialsopenres.17476.1](https://doi.org/10.12688/materialsopenres.17476.1) Abstract Background: Multi-Principal Element Alloys (MPEAs) have better properties, such as yield strength, hardness, and corrosion resistance compared to conventional alloys. Compositional optimization is a challenging task to obtain desired properties of MPEAs and machine learning is a potential tool to rapidly accelerate the search and design of new materials. Methods: We have implemented different machine learning models to predict the yield strength and Vickers hardness of MPEAs at room temperature and quantify the uncertainty of the predictions. Results: Our results suggest that valence electron concentration (VEC) is the key feature dominating the yield strength and hardness of MPEAs. Our predicted yield strength and hardness values for the experimental validation set show < 15 % error for most cases with respect to the experimental values. Conclusions: Our machine learning model can serve as a useful tool to screen half a trillion MPEAs and down select promising compositions for useful applications. Keywords Machine Learning, Yield Strength, Hardness, Multi-Principal Element Alloys Open Peer Review Approval Status AWAITING PEER REVIEW Any reports and responses or comments on the article can be found at the end of the article. Corresponding author: Mohammad Fuad Nur Taufique ([mohammadfn.taufique@pnnl.gov](mohammadfn.taufique@pnnl.gov)) Author roles: Taufique MFN: Conceptualization, Data Curation, Formal Analysis, Software, Writing – Review & Editing; Mamun O: Formal Analysis, Software; Roy A: Formal Analysis, Writing – Review & Editing; Khakurel H: Formal Analysis, Methodology, Software; Balasubramanian G: Supervision, Writing – Review & Editing; Ouyang G: Investigation, Resources, Validation; Cui J: Investigation, Resources, Validation; Johnson DD: Investigation, Resources, Supervision, Validation; Devanathan R: Supervision, Writing – Review & Editing Competing interests: No competing interests were disclosed. Grant information: This effort was principally supported by the U.S. Department of Energy's (DOE) Office of Energy Efficiency and Renewable Energy (EERE) under the Advanced Manufacturing Office (Project WBS [2.1.0.19](2.1.0.19)) through Ames Laboratory, which is operated for the U.S. DOE by Iowa State University under contract DE-AC02-07CH11358 . HK was supported in part through the National Science Foundation (NSF) Mathematical Sciences Graduate Internship (MSGI) Program sponsored by the NSF Division of Mathematical Sciences. This program is administered by the Oak Ridge Institute for Science and Education (ORISE) through an interagency agreement between the U.S. Department of Energy (DOE) and NSF. ORISE is managed for DOE by ORAU. This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States

FAQs

What materials and properties are studied in this article?

The article focuses on Multi-Principal Element Alloys (MPEAs) and predicts room-temperature yield strength and Vickers hardness.

Which feature is identified as most influential for prediction?

Valence electron concentration (VEC) is reported as the key feature dominating both yield strength and hardness.

How accurate are the model predictions against experimental data?

For most cases in the experimental validation set, predicted yield strength and hardness show less than 15% error relative to experimental values.

Aug 3, 2026

Upload by Olivia Brown