| About this Abstract |
| Meeting |
2023 TMS Annual Meeting & Exhibition
|
| Symposium
|
Neutron and X-ray Scattering in Materials Science
|
| Presentation Title |
Scalable Rietveld Refinements of Diffraction |
| Author(s) |
Daniel Savage, Christopher Biwer, Michael McKerns, Cynthia Bolme, Sven Vogel |
| On-Site Speaker (Planned) |
Daniel Savage |
| Abstract Scope |
Diffraction experiments often result in tens to thousands of patterns from which information can be extracted ranging from microstructure to equation of state. The extraction happens by fitting models of the diffraction process which describe the instrument (e.g. sample-to-detector distances and angles), the sample’s crystal structure (e.g. crystallographic space group and lattice parameters), and the microstructure (e.g. phase fractions and texture). Finding the set of parameters that best describes a measurement is hard to do in a robust way without expert intervention and in a way that leverages parallel resources (critical for real-time analysis). We will demonstrate how complex Rietveld of in-situ shocked Titanium can be broken down into a limited decision tree and a set of multilevel optimization schemes. A newly-developed opensource python scripting framework for MAUD Rietveld software (MILK) will be presented with a new python Rietveld optimization package (Spotlight) which together enable the scalable Rietveld optimization strategy. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Characterization, Titanium, Other |