Phase-field models for predicting microstructure evolution: Numerical methods, applications to metal alloys and electrochemical systems, and future directions

Dr. Steve DeWitt

Abstract: Phase field models and related diffuse interface approaches are used to study microstructural evolution due to a variety of phenomena, including solidification, precipitation, grain growth, and electrochemical reactions. This presentation will begin with a general introduction to phase field models, followed a brief discussion of my work prior to joining ORNL on modeling microstructure evolution in electrochemical systems and precipitation in magnesium alloys, as well as my current work involving additive manufacturing. Moving from applications to methods, I’ll discuss the numerical aspects of two frameworks for phase field simulations: PRISMS-PF (matrix-free finite element method) and MEUMAPPS (Fourier-spectral method). The talk will conclude with a discussion of promising areas for future work including the integration of phase field simulations with machine learning and reduced-order modeling techniques, mixed-precision calculations, nonlocal phase field formulations, improved nucleation models, and the potential for integration with fluid flow and fracture codes for “hot cracking” during additive manufacturing.

Speaker’s Bio: Steve DeWitt is a computational scientist in the Computational Sciences and Engineering Division at ORNL (Computational Engineering and Energy Sciences group). Before joining ORNL in September 2019 he spent a while at the University of Michigan, where he received a BSE in Engineering Physics, a PhD in Applied Physics, and then held a research faculty position in the Department of Materials Science and Engineering. His research centers on the use of phase field models in the Integrated Computational Materials Engineering (ICME) paradigm, where computational and experimental approaches are combined to predict material properties and behavior across a range of length and time scales. He is also interested in machine learning and surrogate model creation to efficiently use insights from phase field simulations for materials design and multiscale modeling.