Topic: Machine Learning Approach for Design, Development and Application of High Entropy Materials

Since their advent in 2004, high entropy alloys that comprise more than five principal elements have been attracting tremendous research interest worldwide. Unlike traditional alloys based on one or rarely two principal elements, high entropy alloys are well known for their compositional complexity but yet, a large number of them still show an overall simple solid solution structure. This phenomenon is often attributed to a "high mixing entropy" effect, which, in principle, favors random mixing of materials’ building blocks (e.g., atoms, ions, molecules) over chemical ordering or de-mixing at high temperatures. Therefore, it can be conceived that high entropy alloys are likely to be thermodynamically metastable at ambient temperature, and this structural metastability may impart high entropy materials with unusual structural and functional properties. Aside from alloys, this "high entropy" notion was recently extended to intermetallics and ceramics. Nevertheless, as the number of principal elements increases, the total number of possible compositions can quickly rise to an astronomical value, which defies the traditional "trial-and-error" approach that fixates one composition at a time. Therefore, it is already a consensus in this field that machine learning approaches become crucial to the research of high entropy materials, which can greatly accelerate compositional screening, alloy design and development, and even applications by learning from the big data accumulated in the literature over the past decades.

In this Special Issue, we will emphasize the use of machine learning approaches in tackling the challenging issues in the field of high entropy materials. We welcome original contributions as well as topical reviews. The topics that we are going to cover include, but are not limited to, the following:
● Development of high throughput experimental/computational methods for the establishment of high-fidelity databases
● Machine learning-enabled structural characterizations for high entropy materials
● Machine learning-enabled understanding of thermodynamics and kinetics in high entropy materials
● Machine learning guided the design and development of high entropy materials (i.e., compositional design, processing design, microstructural characterization, etc.)
● Machine learning guided applications of high entropy materials (i.e., structural versus functional applications)

31 Mar 2023