Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships
Chalcogenide glasses (ChGs) possess various outstanding properties enabling essential applications, such as optical discs, infrared cameras, and thermal imaging systems. Despite their ubiquitous usage, these materials’ composition–property relationships remain poorly understood, impeding the pace of...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IOP Publishing
2023-01-01
|
Series: | JPhys Materials |
Subjects: | |
Online Access: | https://doi.org/10.1088/2515-7639/acc6f2 |
_version_ | 1797844661997404160 |
---|---|
author | Sayam Singla Sajid Mannan Mohd Zaki N M Anoop Krishnan |
author_facet | Sayam Singla Sajid Mannan Mohd Zaki N M Anoop Krishnan |
author_sort | Sayam Singla |
collection | DOAJ |
description | Chalcogenide glasses (ChGs) possess various outstanding properties enabling essential applications, such as optical discs, infrared cameras, and thermal imaging systems. Despite their ubiquitous usage, these materials’ composition–property relationships remain poorly understood, impeding the pace of their discovery. Here, we use a large experimental dataset comprising ∼24 000 glass compositions made of 51 distinct elements from the periodic table to develop machine learning (ML) models for predicting 12 properties, namely, annealing point, bulk modulus, density, Vickers hardness, Littleton point, Young’s modulus, shear modulus, softening point, thermal expansion coefficient, glass transition temperature, liquidus temperature, and refractive index. These models are the largest regarding the compositional space and the number of properties covered for ChGs. Further, we use Shapley additive explanations, a game theory-based algorithm, to explain the properties’ compositional control by quantifying each element’s role toward model predictions. This work provides a powerful tool for interpreting the model’s prediction and designing new ChG compositions with targeted properties. Finally, using the trained ML models, we develop several glass-selection charts that can potentially aid in the rational design of novel ChGs for various applications. |
first_indexed | 2024-04-09T17:25:54Z |
format | Article |
id | doaj.art-5d48e69d08644f0ca4562709eafea276 |
institution | Directory Open Access Journal |
issn | 2515-7639 |
language | English |
last_indexed | 2024-04-09T17:25:54Z |
publishDate | 2023-01-01 |
publisher | IOP Publishing |
record_format | Article |
series | JPhys Materials |
spelling | doaj.art-5d48e69d08644f0ca4562709eafea2762023-04-18T13:49:21ZengIOP PublishingJPhys Materials2515-76392023-01-016202400310.1088/2515-7639/acc6f2Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationshipsSayam Singla0https://orcid.org/0009-0000-5759-8611Sajid Mannan1https://orcid.org/0000-0002-7887-2250Mohd Zaki2https://orcid.org/0000-0002-4551-3470N M Anoop Krishnan3https://orcid.org/0000-0003-1500-4947Department of Textile and Fibre Engineering, Indian Institute of Technology Delhi , Hauz Khas, New Delhi 110016, IndiaDepartment of Civil Engineering, Indian Institute of Technology Delhi , Hauz Khas, New Delhi 110016, IndiaDepartment of Civil Engineering, Indian Institute of Technology Delhi , Hauz Khas, New Delhi 110016, IndiaDepartment of Civil Engineering, Indian Institute of Technology Delhi , Hauz Khas, New Delhi 110016, India; Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi , Hauz Khas, New Delhi 110016, IndiaChalcogenide glasses (ChGs) possess various outstanding properties enabling essential applications, such as optical discs, infrared cameras, and thermal imaging systems. Despite their ubiquitous usage, these materials’ composition–property relationships remain poorly understood, impeding the pace of their discovery. Here, we use a large experimental dataset comprising ∼24 000 glass compositions made of 51 distinct elements from the periodic table to develop machine learning (ML) models for predicting 12 properties, namely, annealing point, bulk modulus, density, Vickers hardness, Littleton point, Young’s modulus, shear modulus, softening point, thermal expansion coefficient, glass transition temperature, liquidus temperature, and refractive index. These models are the largest regarding the compositional space and the number of properties covered for ChGs. Further, we use Shapley additive explanations, a game theory-based algorithm, to explain the properties’ compositional control by quantifying each element’s role toward model predictions. This work provides a powerful tool for interpreting the model’s prediction and designing new ChG compositions with targeted properties. Finally, using the trained ML models, we develop several glass-selection charts that can potentially aid in the rational design of novel ChGs for various applications.https://doi.org/10.1088/2515-7639/acc6f2chalcogenide glassesmachine learningSHAPglass design |
spellingShingle | Sayam Singla Sajid Mannan Mohd Zaki N M Anoop Krishnan Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships JPhys Materials chalcogenide glasses machine learning SHAP glass design |
title | Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships |
title_full | Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships |
title_fullStr | Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships |
title_full_unstemmed | Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships |
title_short | Accelerated design of chalcogenide glasses through interpretable machine learning for composition–property relationships |
title_sort | accelerated design of chalcogenide glasses through interpretable machine learning for composition property relationships |
topic | chalcogenide glasses machine learning SHAP glass design |
url | https://doi.org/10.1088/2515-7639/acc6f2 |
work_keys_str_mv | AT sayamsingla accelerateddesignofchalcogenideglassesthroughinterpretablemachinelearningforcompositionpropertyrelationships AT sajidmannan accelerateddesignofchalcogenideglassesthroughinterpretablemachinelearningforcompositionpropertyrelationships AT mohdzaki accelerateddesignofchalcogenideglassesthroughinterpretablemachinelearningforcompositionpropertyrelationships AT nmanoopkrishnan accelerateddesignofchalcogenideglassesthroughinterpretablemachinelearningforcompositionpropertyrelationships |