Symposium CCC: Integrating Experiments, Simulations and Machine Learning to Accelerate Materials Innovation
Bhaskar S. Majumdar , New Mexico Tech, Socorro, New Mexico, United States
High Throughput Experimentation for Exploration of Structural Alloys
Written by Vineet Venugopal
The discovery of new materials has been a result of hard work and serendipity. The need for new alloys and functional materials emphasize accelerated exploration of the material data space to look for interesting candidates. Bhaskar S. Majumdar outlined three strategies for High Throughput Experimentation. The first is to eliminate as many combinations as possible in the shortest period of time. This involves using theoretical models of these materials, with rapidity preferred over accuracy. The second strategy is to implement multiple levels of screening using relevant performance characteristics to further narrow the combination space. The first strategy is composition-focused while the second is microstructure-oriented. These strategies bring down the number of testable materials from the millions to the thousands. Majumdar underscored many relevant testing approaches, such as principal component analysis, CALPHAD, for example, emphasizing that different approaches are required for functional versus structural materials.