Application of machine learning methods for predicting new superhard materials

Mazhnik, Efim; Oganov, Artem R.

doi:10.1063/5.0012055

Cited by 64 publications

(45 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, machine‐learning methods, which can capture such complex connections, have been created to identify new superhard materials based on the elastic moduli. [ 22,23 ] Although these reports both illustrate the speed and accuracy of machine learning for rapid materials screening, these methods still rely on computationally derived, indirect proxies of hardness that are subject to misinterpretation.…”

Section: Figurementioning

confidence: 99%

Finding the Next Superhard Material through Ensemble Learning

et al. 2020

View full text Add to dashboard Cite

An ensemble machine‐learning method is demonstrated to be capable of finding superhard materials by directly predicting the load‐dependent Vickers hardness based only on the chemical composition. A total of 1062 experimentally measured load‐dependent Vickers hardness data are extracted from the literature and used to train a supervised machine‐learning algorithm utilizing boosting, achieving excellent accuracy (R2 = 0.97). This new model is then tested by synthesizing and measuring the load‐dependent hardness of several unreported disilicides and analyzing the predicted hardness of several classic superhard materials. The trained ensemble method is then employed to screen for superhard materials by examining more than 66 000 compounds in crystal structure databases, which show that 68 known materials have a Vickers hardness ≥40 GPa at 0.5 N (applied force) and only 10 exceed this mark at 5 N. The hardness model is then combined with the data‐driven phase diagram generation tool to expand the limited number of reported high hardness compounds. Eleven ternary borocarbide phase spaces are studied, and more than ten thermodynamically favorable compositions with a hardness above 40 GPa (at 0.5 N) are identified, proving this ensemble model's ability to find previously unknown materials with outstanding mechanical properties.

show abstract

Section: Figurementioning

confidence: 99%

Finding the Next Superhard Material through Ensemble Learning

et al. 2020

View full text Add to dashboard Cite

show abstract

“…For the strength model based on dislocation theory 52 , only dislocation defects are considered, and no bond properties are considered; therefore, this model can only be adopted for some metallic materials with less electron localization. For the hardness model based of valence bond theory [18][19][20][21][22][23] , only bond properties are considered, and no dislocation defects are considered; therefore, this model can only be adopted for some covalent materials with high electron localization. The effect factor on strength is not fully considered in the previous two types of strength models.…”

Section: Reconciling Chemical Bonds and Dislocations In The Unified S...mentioning

confidence: 99%

“…The strength (hardness) models based on valence bond theory are mainly for covalent and ionic materials [18][19][20][21][22][23] . Currently, only the properties of chemical bonds are considered in all these models.…”

Section: Introductionmentioning

confidence: 99%

A unified nonempirical strength model

Feng

Sun

Zhang

et al. 2022

Preprint

View full text Add to dashboard Cite

Strength, as an important indicator of structural materials, has always been an important research topic in materials science. Theoretically, building a strength model is a rewarding method to understand the relationship between the mechanical properties and microstructure of materials. Although many strength models can reduplicate experimental values very well, they are empirical models, and their applicability is limited to materials for which empirical parameters have been obtained. Here, a nonempirical strength model is proposed based on the two-dimensional (2D) displacement potential of dislocation slipping, which can be applied to different chemically bonded crystals. Owing to the large electron localization function (ELF), covalent and ionic crystals have a high 2D displacement potential of dislocation slipping, and their dislocation slip mode prefers the kink-pair mode, further exhibiting a high critical resolved shear stress (CRSS). In contrast, metallic crystals with a small ELF have a low 2D displacement potential of dislocation slipping, and their dislocation slip mode is more inclined to the string mode, showing a low CRSS. This work provides new insights into dislocation-slipping configurations that will be useful for the development of new high-performance structural materials.

show abstract

“…More recently, machine learning (ML) has been found to be a viable way to reduce the number of experiments, as well as computations, to accelerate the design process [3][4][5][6][7]. Demand for robust ML models is there, but a big hurdle in their adoption is the limited availability of data, either experimental or computational, that ML models need to be trained on.…”

Section: Introductionmentioning

confidence: 99%

“…For example, Avery et al used AFLOW and a combined ML and evolutionary search method to predict * henrik.levamaki@liu.se new superhard phases in carbon [6]. Mazhnik et al used Materials Project data to train the crystal graph convolutional neural network (CGCNN) [10] model to predict the bulk and shear moduli of ordered compounds using only the structural and chemical information of the system as input [5], obtaining good results. In this paper we extend the approach to qualitatively different class of materials, disordered alloys (as deposited metastable thin films).…”

Section: Introductionmentioning

confidence: 99%

Predicting properties of hard-coating alloys using ab-initio and machine learning methods

Levämäki¹,

Tasnádi²,

Sangiovanni³

et al. 2021

Preprint

View full text Add to dashboard Cite

Accelerated design of novel hard coating materials requires state-of-the-art computational tools, which include data-driven techniques, building databases, and training machine learning models against the databases. In this work, we present a development of a heavily automated highthroughput workflow to build a database of industrially relevant hard coating materials, such as binary and ternary nitrides. We use Vienna Ab initio Simulation package as the density functional theory calculator and the high-throughput toolkit to automate the calculation workflow. We calculate and present results, including the elastic constants, one of the key materials parameter that determines mechanical properties of the coatings, for X1−xYxN binary and ternary nitrides, where X,Y ∈ {Al, Ti, Zr, Hf} and fraction x = 0, 1 4 , 1 2 , 3 4 , 1. We explore ways for ML techniques to support and complement the designed databases. We find that the crystal graph convolutional neural network model trained on Materials Project data for ordered lattices has sufficient prediction accuracy for the disordered nitrides, suggesting that the existing databases provide important data for predicting mechanical properties of qualitatively different type of material systems, in our case hard coating alloys, not included in the original dataset.

show abstract

Application of machine learning methods for predicting new superhard materials

Cited by 64 publications

References 42 publications

Finding the Next Superhard Material through Ensemble Learning

Finding the Next Superhard Material through Ensemble Learning

A unified nonempirical strength model

Predicting properties of hard-coating alloys using ab-initio and machine learning methods

Contact Info

Product

Resources

About