Fundamental Investigation of Thermal and Mechanical Properties of High-Entropy Alloys via Composition and Structural Analysis

High-entropy alloys (HEAs), specifically AlCoCrFeNi HEAs, have attracted significant attention for their distinct microstructures and adjustable properties. To enhance the energy conversion efficiency, stability, and mechanical performance of their applications, further refinement or fine-tuning of the thermoelectric, thermodynamics, and mechanical properties of such HEAs is necessary. Effective analysis and design of HEAs require a comprehensive understanding of the effects of various structural parameters, such as composition and structural orders. However, the structural complexities and vast configurational space of HEAs have posed challenges in advancing the required fundamental understanding.

This research aims to develop an effective methodology capable of addressing elemental compositions and structural orders, thereby elucidating the understanding of the effects of various structural parameters on HEA properties. For this study, molecular dynamics (MD) simulations, Monte Carlo (MC) simulations, first-principles calculations, and Machine Learning (ML) techniques were employed, as they enabled the investigation of atomic properties and efficient data processing.

Firstly, this study investigated the influence of aluminum (Al) content on phonon transport and thermoelectric properties in Al_xCoCrFeNi HEAs across a wide temperature range. Low Al concentrations were found to reduce phonon conductivity, increasing thermoelectric performance with minimal temperature dependence. Conversely, electrical conductivity and Seebeck coefficient increased with temperature, particularly at higher Al concentrations, enhancing the thermoelectric performance. Secondly, the effects of short-range order (SRO) on the thermodynamic properties of Al_xCoCrFeNi HEAs were examined, revealing prevalent negative SRO of Al-Fe and positive SRO of Al-Al pairs. SRO impacts the lattice thermal conductivity, bulk modulus, and thermal expansion coefficient, thereby aiding in the design of HEAs with tailored properties via structural tuning. Finally, machine learning models were trained with SRO parameters to predict yield strength of AlCoCrFeNi HEA families, deepening understanding of the SRO-mechanical property relationship and aiding in the design of HEAs for improved mechanical strength.

This study provides insights into the interactions among element concentrations, local atomic order, and physical properties of HEAs, thereby advancing the development of a quantitative theory-guided HEA design strategy for specific applications and energy conservation efforts. It also establishes a foundation for further studies extending into the design and optimization of other HEA systems