EventsThe 2nd International Electronic Conference on Metals
Published
This submission belongs to the session S1. Computation, AI, and Machine Learning on Metals of the event The 2nd International Electronic Conference on Metals
Published date
02 May, 2025
Academic Editor
author-avatarChuang Deng
Citation
Arman Hobhaydar, Xiao Wang, Huijun Li, Nam Van Tran, Hongtao Zhu, An Integrated Experimental and Quantum/Atomic Modelling of FeCrV-Based Refractory Medium Entropy Alloy for Nuclear Application, in Proceedings of The 2nd International Electronic Conference on Metals, 5 May–7 May 2025, MDPI: Basel, Switzerland
Share
Email
Facebook
Twitter
LinkedIn

An Integrated Experimental and Quantum/Atomic Modelling of FeCrV-Based Refractory Medium Entropy Alloy for Nuclear Application

Xiao Wang 1
image
Nam Van Tran 2
image
1. School of Mechanical, Materials, Mechatronic and Biomedical Engineering, University of Wollongong in Australia, Australia
2. School of Materials Science and Engineering (MSE), Nanyang Technological University in Singapore, Singapore
Abstract

Introduction

Rising energy demands and environmental challenges necessitate advancements in nuclear reactor materials to endure harsh conditions, including high temperatures, neutron flux, and oxidation. Medium-entropy alloys (MEAs) offer promising properties like enhanced strength, ductility, and radiation resistance. This study focuses on designing FeCrV-based refractory MEAs for nuclear applications using multi-scale modelling and experimental validation, leveraging the potential of DFT and machine learning to optimize performance and simulate large-scale behaviours.

Methods

First-principle calculations were performed using VASP, employing the PAW method and PBE-GGA for accurate modelling. AIMD simulations with a 0.5 fs timestep generated training data under constant and increasing temperatures. Neural network potentials were trained using DeepPot-SE. Large-scale MD simulations applied periodic boundaries and NPT equilibration at 300 K-1023 K. MC/MD methods facilitated atom exchanges, governed by Boltzmann probability, to explore temperature-dependent behaviours and structural evolution.

Results

XRD and TEM microstructure analyses revealed that the RMEA comprises two distinct phases: BCC1, the nominal alloy phase, and BCC2, a vanadium-rich phase. These two phases were also identified in MD/MC simulations conducted using the DNNP, demonstrating the accuracy of the model. The addition of 8% W to FeCr2V significantly enhances its mechanical properties, achieving an ultimate compression strength of 1700 MPa and a Young’s modulus of 255.28 GPa. According to the DFT results, these improvements are attributed to a balanced interaction between metallic and covalent bonding, producing a highly irradiation-resistant material with an average vacancy formation energy of 2.66 eV. Dislocation analysis conducted via MD revealed that increasing the temperature enhances dislocation mobility, which further improves the ductility of the alloy.

Conclusion

In conclusion, a cost-effective FeCr2V-based RMEA with superior mechanical properties, including high strength, ductility, and irradiation resistance, was successfully designed. With properties surpassing HEAs, this alloy demonstrates improved dislocation mobility at elevated temperatures, highlighting its potential for advanced structural applications in extreme environments.

Keywords
Nuclear Material
Medium Entropy Alloys
Density Functional Theory
Molecular Dynamics
Deep Neural Network Potential
The Potential Property-Tailoring Effects of Cryogenic Treatment on Pure Zinc
Study of Crevice Corrosion in X70 pipeline Steel Using a 2D Finite Element Model