Failure, and a Career That is Anything But: An LMD Symposium Honoring J. Wayne Jones: On-Demand Oral Presentations
Sponsored by: TMS Light Metals Division, TMS: Magnesium Committee
Program Organizers: Victoria Miller, University of Florida; Michael Caton, US Air Force Research Laboratory; Nikhilesh Chawla, Purdue University; Trevor Harding, California Polytechnic State University; Paul Krajewski, General Motors Corporation; Tresa Pollock, University of California - Santa Barbara
Monday 8:00 AM
March 14, 2022
Room: Light Metals
Location: On-Demand Room
Redefining Liquid Metal Embrittlement: Utilizing Machine Learning to Unravel a Liquid Metal Enigma
: Justin Norkett1; Cameron Frampton1; Victoria Miller1; 1University of Florida
Liquid metal embrittlement (LME) has been a persistent curiosity of metallurgists for over 100 years. In that time, and despite numerous efforts, no model has been able to reliably predict embrittlement for any arbitrary combination of liquid and solid metals. This is in large part due to the absence of deep mechanistic understanding of LME. The authors recently proposed a new interpretation of the phenomenology of LME as the result and interplay of no less than three distinct mechanisms. In this talk, the iterative improvement of a machine learning model capable of classifying LME by active mechanism will be described. The experimental work and the model refinements have expanded the understanding of LME phenomena; implications for the future of the field are discussed.