Stephen J DeWitt

Computational Scientist: Large-Scale Simulations

Dr. DeWitt’s research lies at the intersection of high-performance computing and the study of microstructure evolution in materials. His current work centers on the development of scalable phase-field simulation codes for heterogeneous computer architectures and the use of these codes to study microstructure evolution in metal alloys during additive manufacturing. Particularly in the context of additive manufacturing, he is interested in multiscale models and how part-scale thermal fields influence the microstructure through the part. He has also applied phase-field and level set methods to investigate microstructure evolution in magnesium alloys and electrochemical systems. To extract the most value from computationally intensive simulations, he leverages techniques from the fields of machine learning, uncertainty quantification, data assimilation, and materials informatics.

Appointments

2022 - Present:  R&D Staff, Computational Sciences and Engineering Division, Oak Ridge National Laboratory

2019 - 2022:  R&D Associate Staff, Computational Sciences and Engineering Division, Oak Ridge National Laboratory

2017 - 2019:  Assistant Research Scientist (Research Faculty), Dept. of Materials Science and Engineering, University of Michigan

2015 - 2017:  Research Investigator (Research Faculty), Dept. of Materials Science and Engineering, University of Michigan

Education

PhD:  Applied Physics, University of Michigan, 2015

MS:  Applied Physics, University of Michigan, 2011

BSE:  Engineering Physics, University of Michigan, 2009

Simulation Software

MEUMAPPS: https://code.ornl.gov/meumapps/meumapps

adamantine: https://github.com/adamantine-sim/adamantine

Thermo4PFM: https://github.com/ORNL/Thermo4PFM

AMPE: https://github.com/LLNL/AMPE

PRISMS-PF: https://prisms-center.github.io/phaseField

Publications