Additive Manufacturing Benchmarks 2022 (AM-Bench 2022): Materials IV: Grain Scale
Program Organizers: Brandon Lane, National Institute of Standards and Technology; Lyle Levine, National Institute of Standards and Technology
Tuesday 3:30 PM
August 16, 2022
Room: Regency Ballroom III & IV
Location: Hyatt Regency Bethesda
Session Chair: Stephen DeWitt, Oak Ridge National Laboratory
3:30 PM Invited
Methods for Collecting High Resolution Large Area EBSD Maps: David Rowenhorst1; Dillon Watring1; 1Naval Research Laboratory
Electron back-scattered diffraction (EBSD) mapping has many advantages over traditional imaging methods for polycrystalline systems, most notably grain orientations and morphologies are clearly delineated. With the advent of faster detectors, it is now reasonable to collect maps that contain many thousands of grain cross-sections. However, the large sample tilts needed for high quality EBSD patterns, lead to less than ideal imaging conditions for areas larger than 500 micron^2. In this presentation, we will show how this can be mitigated by using a tiled or mosaic collection strategy, and will demonstrate an algorithm that accounts for and largely removes the spatial distortions present in the EBSD maps, creating nearly seamless stitches between the EBSD tiles. We will then demonstrate how one can create EBSD maps that span multiple millimeters, while maintaining sub-micron accuracy and how this was implemented in the AM Bench data collection process.
4:00 PM Invited
Modeling of Location-specific Grain Shape and Texture Development in the AMB2018-01 Bridge Specimen: Matthew Rolchigo1; John Coleman1; Gerry Knapp1; Alex Plotkowski1; James Belak2; 1Oak Ridge National Laboratory; 2Lawrence Livermore National Laboratory
EBSD measurements from various locations within the AMB2018-01 specimen show a complex microstructure, consisting of multiple distributions of grain sizes and textures that further vary with location in the part. We apply AdditiveFOAM and ExaCA to model melt pool development and as-solidified grain structure as a function of build height, starting from the baseplate, in L7, L8, and L9. It is found that melt pool geometry differences as a function of location within the scan pattern, particularly between L8 and L7/L9 and between even and odd deposited layers, are a significant driver of grain size and texture development. Model assumptions and predicted microstructure results are validated against in-situ build observations and characterization results. Calibration of uncertain ExaCA parameters such as nucleation density based on experimental results is also discussed. Work supported by ECP (17-SC-20-SC) and performed under the auspices of Oak Ridge National Laboratory (ORNL) under U.S. Government Contract DE-AC05-00OR22725.
4:30 PM Invited
Integrated Monte Carlo Microstructure and Analytical Temperature Simulations of Additive Manufacturing: Brodan Richter1; Joshua Pribe2; Wesley Tayon1; Samuel Hocker1; Saikumar Reddy Yeratapally2; Edward Glaessgen1; 1National Aerospace and Space Administration; 2National Institute of Aerospace
The large process-design space associated with additive manufacturing (AM) renders correlation of microstructure with processing parameters impractical due to the resources needed for sample creation and analysis. The sensitivity of microstructure to minor processing deviations is also unknown, which makes AM part qualification and certification difficult. The development of simulation methods for accurately replicating physical AM builds is critical for addressing the process-design space challenge. This study represents progress in AM microstructure simulation through the integration of analytical temperature solutions and the kinetic Monte Carlo method. The analytical solution speed enhancements and the sensitivity of grain statistics to processing parameters are outlined. The impacts of implementing texture evolution techniques are discussed. Experimentally characterized AM Ti-6Al-4V microstructures are compared with simulated samples in order to test the simulation techniques. This work demonstrates how validated simulation techniques provide a methodology for understanding and addressing the challenge.