Members Present:

• Faculty
  • Roger French – Director, MDS3 Center
  • Laura Bruckman – Co-Director
• Staff
  • Tariq Shabazz – Program Manager
• Graduate Students
  • Jonah Bachman – Graduate Student
  • Hein Htet Aung – Graduate Student

Key Highlights:

Additional Notes:

We are proud of our team’s participation and look forward to applying insights gained from the event to our ongoing research activities within the MDS3 Center.

This January, over 20 visitors from several Department of Energy – National Nuclear Security Administration (DOE-NNSA) Labs (Livermore, Los Alamos and Sandia) and Production Facilities (KCNSC and Y12) and from the AWE in the UK visited CWRU for the 7th Bi-Annual Winter Meeting for the NNSA’s Materials Data Science for Stockpile Stewardship (MDS^3) – Center of Excellence housed in the Materials Science & Engineering Department.

CWRU MDS3 Researchers presented a variety of topics related to materials data science and on how to make scientific data “AI-ready” to meet the challenges of the new AI era of science. This includes strategies such as using Semantic Web technologies, Ontologies and Knowledge Graphs to make scientific data understandable to people and machines. This has lead to the development of the Materials Data Science Ontology (MDS-Onto, https://cwrusdle.bitbucket.io/) which enables scientists to FAIRify their datasets, analysis and models.  This FAIR data is Findable, Accessible, Interoperable and Reusable, by both people and machines, and is the perfect “fuel” for AI driven scientific advancement.

The meeting also included tours of ThinkBox and other EMSE Labs, a poster reception, and demos of the many code packages and data science tools developed in MDS^3.

National Lab members can view presentation slides and recordings in the research portal.

Abstract:

Understanding materials from the basic building blocks of matter can enable predictive modeling leading to accelerated material certification, engineering design improvement, nimble manufacturing practices, and physical understanding of materials in extremes. Unfortunately, bridging all the length scales has proven challenging, even for simple material systems, and a highly coupled modeling and experimental approach is often required to realize predictive toolsets. The mesoscale microstructure description of a material has proven especially useful for modeling many engineering material properties since defects such as dislocations can be represented as a continuum and the ordering of atoms which form grains in a polycrystal are responsible for effective properties. In addition, the mesoscale can also be easily characterized experimentally. In this talk, a mesoscale model of high purity titanium will be presented with a focus on understanding the link between microstructure and properties (i.e. plastic behavior). A common theme will be how evolving characterization capabilities (e.g. diffraction methods) and computational methods (e.g. Bayesian optimization and optimal exploration of parameter spaces) are changing how experiments and models are integrated.

Bio:

Dan Savage is a staff scientist in the Materials Science and Technology Division at Los Alamos National Laboratory (LANL) where he is part of the neutron and X-ray scattering team. During his recent postdoc at LANL he was a G.T. Seaborg Institute Postdoctoral Fellow, working to predictively understand microstructure evolution in shape memory alloys and understand shock induced phase transformations using femtosecond X-ray diffraction. Prior to joining LANL in the fall of 2020, Dan was an NSF Graduate Fellow and received his PhD in Mechanical Engineering from the University of New Hampshire where his focus was in crystal and continuum descriptions of plasticity and ductile damage. His current research is centered around microstructure-based understanding of materials and developing applications of optimization, uncertainty quantification, and physics-based modeling in diffraction analysis.