About this Abstract |
| Meeting |
2027 TMS Annual Meeting & Exhibition
|
| Symposium
|
Microstructure-Sensitive Design and Advanced Characterization: An MPMD/SMD Symposium Honoring David T. Fullwood
|
| Presentation Title |
Patterns First: A Retrieval-Augmented, Uncertainty-Aware Platform for High-Fidelity EBSD Acquisition to Enable Microstructure-Sensitive Design |
| Author(s) |
Ronald Guymon, Gage Erickson, Carl Robison, Xavier Zaitzeff, Sterling G. Baird |
| On-Site Speaker (Planned) |
Sterling G. Baird |
| Abstract Scope |
Trustworthy microstructure-sensitive design begins with trustworthy patterns: HR-EBSD elastic-strain and GND-density measurements, and the statistical microstructure descriptors they feed, are only as reliable as the upstream EBSD acquisition. However, that step can be fragile and operator-dependent. On laser-powder-bed-fusion Al–Si, EBSD indexing yield rose from ~40% to ~77% of points above CI = 0.1 once beam current and a custom AlSi10Mg / diamond-cubic-Si phase model were corrected; standardless SEM-EDS separately over-reported trace Mg several-fold. We present CALIBER, an uncertainty-aware platform that recommends and iteratively refines EBSD and complementary EDS/XRF acquisition parameters for a given sample and analysis goal. Recommendations are grounded in retrieval over characterization standards and HR-EBSD literature, electron-beam and interaction-volume simulation, and a growing provenance-tracked dataset of parameter sweeps. The result is a reproducible, transferable route to high-quality patterns, and a labeled dataset for statistical microstructure quantification, that makes downstream HR-EBSD, GND, and MSD analyses defensible. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
High-Temperature Materials, Ceramics, Solidification |