Abstract High Entropy Alloys (HEAs) are composed of more than one principal element and constitute a major paradigm in metals research. The HEA space is vast and an exhaustive exploration is improbable. Therefore, a thorough estimation of the phases present in the HEA is of paramount importance for alloy design. Machine Learning presents a feasible and non-expensive method for predicting possible new HEAs on-the-fly. A deep neural network (DNN) model for the elemental system of: Mn, Ni, Fe, Al, Cr, Nb, and Co is developed using a dataset generated by high-throughput computational thermodynamic calculations using Thermo-Calc. The features list used for the neural network is developed based on literature and freely available databases. A feature significance analysis matches the reported HEAs phase constitution trends on elemental properties and further expands it by providing so far-overlooked features. The final regressor has a coefficient of determination ( r 2 ) greater than 0.96 for identifying the most recurrent phases and the functionality is tested by running optimization tasks that simulate those required in alloy design. The DNN developed constitutes an example of an emulator that can be used in fast, real-time materials discovery/design tasks.
more »
« less
Composition design of high-entropy alloys with deep sets learning
High entropy alloys (HEAs) are an important material class in the development of next-generation structural materials, but the astronomically large composition space cannot be efficiently explored by experiments or first-principles calculations. Machine learning (ML) methods might address this challenge, but ML of HEAs has been hindered by the scarcity of HEA property data. In this work, the EMTO-CPA method was used to generate a large HEA dataset (spanning a composition space of 14 elements) containing 7086 cubic HEA structures with structural properties, 1911 of which have the complete elastic tensor calculated. The elastic property dataset was used to train a ML model with the Deep Sets architecture. The Deep Sets model has better predictive performance and generalizability compared to other ML models. Association rule mining was applied to the model predictions to describe the compositional dependence of HEA elastic properties and to demonstrate the potential for data-driven alloy design.
more » « less- PAR ID:
- 10366474
- Publisher / Repository:
- Nature Publishing Group
- Date Published:
- Journal Name:
- npj Computational Materials
- Volume:
- 8
- Issue:
- 1
- ISSN:
- 2057-3960
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Sluggish diffusion is postulated as an underlying mechanism for many unique properties in high-entropy alloys (HEAs). However, its existence remains a subject of debate. Due to the challenges of exploring the vast composition space, to date most experimental and computational diffusion studies have been limited to equiatomic HEA compositions. To develop a high-throughput approach to study sluggish diffusion in a wide range of non-equiatomic compositions, this work presents an innovative artificial neural network (ANN) based machine learning model that can predict the vacancy migration barriers for arbitrary local atomic configurations in a model FeNiCrCoCu HEA system. Remarkably, the model utilizes the training data exclusively from the equiatomic HEA while it can accurately predict barriers in non-equiatomic HEAs as well as in the quaternary, ternary, and binary sub-systems. The ANN model is implemented as an on-the-fly barrier calculator for kinetic Monte Carlo (KMC) simulations, achieving diffusivities nearly identical to the independent molecular dynamics (MD) simulations but with far higher efficiency. The high-throughput ANN-KMC method is then used to study the diffusion behavior in 1,500 non-equiatomic HEA compositions. It is found that although the sluggish diffusion is not evident in the equiatomic HEA, it does exist in many non-equiatomic compositions. The compositions, complex potential energy landscapes (PEL), and percolation effect of the fastest diffuser (Cu) in these sluggish compositions are analyzed, which could provide valuable insights for the experimental HEA designs.more » « less
-
more » « less
Abstract Graph neural networks (GNNs) have excelled in predictive modeling for both crystals and molecules, owing to the expressiveness of graph representations. High-entropy alloys (HEAs), however, lack chemical long-range order, limiting the applicability of current graph representations. To overcome this challenge, we propose a representation of HEAs as a collection of local environment (LE) graphs. Based on this representation, we introduce the LESets machine learning model, an accurate, interpretable GNN for HEA property prediction. We demonstrate the accuracy of LESets in modeling the mechanical properties of quaternary HEAs. Through analyses and interpretation, we further extract insights into the modeling and design of HEAs. In a broader sense, LESets extends the potential applicability of GNNs to disordered materials with combinatorial complexity formed by diverse constituents and their flexible configurations.
-
In this work, a dataset including structural and mechanical properties of refractory multicomponent alloys was developed by fusing computations of phase diagram (CALPHAD) and density functional theory (DFT). The refractory multicomponent alloys, also named refractory complex concentrated alloys (CCAs) which contain 2–5 types of refractory elements were constructed based on Special Quasi-random Structure (SQS). The phase of alloys was predicted using CALPHAD and the mechanical property of alloys with stable and single body-centered cubic (BCC) at high temperature (over 1,500°C) was investigated using DFT-based simulation. As a result, a dataset with 393 refractory alloys and 12 features, including volume, melting temperature, density, energy, elastic constants, mechanical moduli, and hardness, were produced. To test the capability of the dataset on supporting machine learning (ML) study to investigate the property of CCAs, CALPHAD, and DFT calculations were compared with principal components analysis (PCA) technique and rule of mixture (ROM), respectively. It is demonstrated that the CALPHAD and DFT results are more in line with experimental observations for the alloy phase, structural and mechanical properties. Furthermore, the data were utilized to train a verity of ML models to predict the performance of certain CCAs with advanced mechanical properties, highlighting the usefulness of the dataset for ML technique on CCA property prediction.more » « less
-
more » « less
Abstract Severe distortion is one of the four core effects in single‐phase high‐entropy alloys (HEAs) and contributes significantly to the yield strength. However, the connection between the atomic‐scale lattice distortion and macro‐scale mechanical properties through experimental verification has yet to be fully achieved, owing to two critical challenges: 1) the difficulty in the development of homogeneous single‐phase solid‐solution HEAs and 2) the ambiguity in describing the lattice distortion and related measurements and calculations. A single‐phase body‐centered‐cubic (BCC) refractory HEA, NbTaTiVZr, using thermodynamic modeling coupled with experimental verifications, is developed. Compared to the previously developed single‐phase NbTaTiV HEA, the NbTaTiVZr HEA shows a higher yield strength and comparable plasticity. The increase in yield strength is systematically and quantitatively studied in terms of lattice distortion using a theoretical model, first‐principles calculations, synchrotron X‐ray/neutron diffraction, atom‐probe tomography, and scanning transmission electron microscopy techniques. These results demonstrate that severe lattice distortion is a core factor for developing high strengths in refractory HEAs.
