In this work, the CuAgZr metallic glasses (MGs) are investigated, a promising material for biomedical applications due to their high strength, corrosion resistance, and antibacterial activity. Using an integrated approach of combinatorial synthesis, high-throughput characterization, and machine learning (ML), the mechanical properties of CuAgZr MGs are efficiently explored. The investigation find that post-deposition oxidation in inter-columnar regions with looser packing causes high oxygen content in Cu-rich regions, significantly affecting the alloys' mechanical behavior. The study also reveals that nanoscale structural features greatly impact plastic yielding and flow in the alloys. ML algorithms are tested, and the multi-layer perceptron algorithm produced satisfactory predictions for the alloys' hardness of untested alloys, providing valuable clues for future research. The work demonstrates the potential of using combinatorial synthesis, high-throughput characterization, and ML techniques to facilitate the development of new MGs with improved strength and economic feasibility.
Keywords: high-throughput characterization; machine learning; material library; mechanical properties; metallic glasses.
© 2023 The Authors. Advanced Science published by Wiley-VCH GmbH.