76
S. Ghosh et al.
containing annealing twins is shown in Fig. 10b. The ensemble contains 440 3
type twins in 300 parent grains for a total of 740 twins and grains. The 3D assembly
and microstructure reconstruction is performed in the DREAM.3D software [33]
in four sequential steps, viz., (i) slice registration and alignment; (ii) voxel level
cleanup; (iii) feature segmentation; and (iv) artifact removal. Corresponding to the
angular resolution of orientation measures in the EBSD scans, the segmentation
tolerance in the DREAM.3D code is chosen to be 1 ◦ –5 ◦ . Noisy voxels are identified
by thresholding misorientations with neighboring voxels and assigning thresholds
on the minimum acceptable grain size. If these criteria are not met, morphological
dilation of the voxels or small grains into surrounding grains is conducted.
3.2 Statistically Equivalent Virtual Microstructure (SEVM)
Generation from Characterization and Statistical Analysis
The EBSD data in Fig. 10a and the assembled 3D microstructure in Fig. 10b show
that the Ren ´
e88-DT polycrystalline microstructure contains an abundance of large
aspect ratio, annealing 3 twins that develop during material processing. These
twins have a 60 ◦ misorientation angle with respect to the < 111 > crystal
lattice axis, whereas most of the parent grain boundaries are randomly oriented.
Microstructure characterization further reveals that the majority of annealing twin
boundaries are coherent with boundaries formed along planes with {111} Millerindex facet.
The generation of polycrystalline SEVMs proceeds in two stages. Statistics of
the parent grain without twins, that have random grain boundary orientations, are
first generated. This is followed by the incorporation of twins in the parent grains.
The following steps are executed in sequence to generate statistically equivalent
volumes of twinned polycrystalline microstructures.
1. Process the EBSD section data and construct the digitally assembled polycrystalline ensemble including twins;
2. Identify and remove twins from the digitally assembled microstructure to
manifest the parent grains, shown in Fig. 10c;
3. Extract the statistics of parent grains from the EBSD data;
4. Create statistically equivalent virtual parent grain microstructures from the 3D
EBSD data;
5. Extract correlation statistics of twins with respect to parent grains from the EBSD
data in the digitally assembled microstructure;
6. Insert twins in the parent polycrystalline microstructure to match the statistical
correlations.
The DREAM.3D software [33] is employed in steps 1–4 to create the digital
polycrystalline ensembles from EBSD data in Fig. 10b, remove twins in Fig. 10c and
subsequently extract statistics of the parent grains. The statistics of characteristic
features in the twin-free parent grains in Fig. 10c, including probability distributions
S. Ghosh et al.
containing annealing twins is shown in Fig. 10b. The ensemble contains 440 3
type twins in 300 parent grains for a total of 740 twins and grains. The 3D assembly
and microstructure reconstruction is performed in the DREAM.3D software [33]
in four sequential steps, viz., (i) slice registration and alignment; (ii) voxel level
cleanup; (iii) feature segmentation; and (iv) artifact removal. Corresponding to the
angular resolution of orientation measures in the EBSD scans, the segmentation
tolerance in the DREAM.3D code is chosen to be 1 ◦ –5 ◦ . Noisy voxels are identified
by thresholding misorientations with neighboring voxels and assigning thresholds
on the minimum acceptable grain size. If these criteria are not met, morphological
dilation of the voxels or small grains into surrounding grains is conducted.
3.2 Statistically Equivalent Virtual Microstructure (SEVM)
Generation from Characterization and Statistical Analysis
The EBSD data in Fig. 10a and the assembled 3D microstructure in Fig. 10b show
that the Ren ´
e88-DT polycrystalline microstructure contains an abundance of large
aspect ratio, annealing 3 twins that develop during material processing. These
twins have a 60 ◦ misorientation angle with respect to the < 111 > crystal
lattice axis, whereas most of the parent grain boundaries are randomly oriented.
Microstructure characterization further reveals that the majority of annealing twin
boundaries are coherent with boundaries formed along planes with {111} Millerindex facet.
The generation of polycrystalline SEVMs proceeds in two stages. Statistics of
the parent grain without twins, that have random grain boundary orientations, are
first generated. This is followed by the incorporation of twins in the parent grains.
The following steps are executed in sequence to generate statistically equivalent
volumes of twinned polycrystalline microstructures.
1. Process the EBSD section data and construct the digitally assembled polycrystalline ensemble including twins;
2. Identify and remove twins from the digitally assembled microstructure to
manifest the parent grains, shown in Fig. 10c;
3. Extract the statistics of parent grains from the EBSD data;
4. Create statistically equivalent virtual parent grain microstructures from the 3D
EBSD data;
5. Extract correlation statistics of twins with respect to parent grains from the EBSD
data in the digitally assembled microstructure;
6. Insert twins in the parent polycrystalline microstructure to match the statistical
correlations.
The DREAM.3D software [33] is employed in steps 1–4 to create the digital
polycrystalline ensembles from EBSD data in Fig. 10b, remove twins in Fig. 10c and
subsequently extract statistics of the parent grains. The statistics of characteristic
features in the twin-free parent grains in Fig. 10c, including probability distributions