118
D. W. Eastman et al.
Fig. 11 Surface maps of a 50 × 50 μm tested microsample showing: (a) grain orientations using
standard IPF coloring for FCC, (b) Maximum Schmid factors for each grain, (c) the Young’s
modulus along the loading direction for each grain, and (d) a comparison of the surface strain with
the grains that make up the polycrystalline microstructure
The combination of 2D surface strain mapping and microstructural characterization is a powerful tool for benchmarking microstructurally dependent models.
However, subsurface grains can significantly affect observations made at the sample
surface, as shown in microtensile samples of pure Ni [22, 24]. A 3D dataset of
the full sample microstructure would avoid such complications, and using explicit
3D microstructural datasets as input for CPFEM simulations would facilitate direct
one-to-one comparisons and provide a significant step forward.
Although still in their adolescence, laser-based and mechanical serial sectioning
techniques, and emerging nondestructive characterization techniques such as high
energy diffraction microscopy (HEDM), provide a means to digitally capture 3D
volumes that could be utilized to instantiate a simulation and directly compare with
experimental results. Figure 12 presents a roadmap and initial proof-of-concept
results for collecting and using 3D digital microstructures to instantiate CPFEM
simulations. Once collected, the digital microstructural dataset must be properly
segmented in order to properly identify grains and twins. Once identified, each
D. W. Eastman et al.
Fig. 11 Surface maps of a 50 × 50 μm tested microsample showing: (a) grain orientations using
standard IPF coloring for FCC, (b) Maximum Schmid factors for each grain, (c) the Young’s
modulus along the loading direction for each grain, and (d) a comparison of the surface strain with
the grains that make up the polycrystalline microstructure
The combination of 2D surface strain mapping and microstructural characterization is a powerful tool for benchmarking microstructurally dependent models.
However, subsurface grains can significantly affect observations made at the sample
surface, as shown in microtensile samples of pure Ni [22, 24]. A 3D dataset of
the full sample microstructure would avoid such complications, and using explicit
3D microstructural datasets as input for CPFEM simulations would facilitate direct
one-to-one comparisons and provide a significant step forward.
Although still in their adolescence, laser-based and mechanical serial sectioning
techniques, and emerging nondestructive characterization techniques such as high
energy diffraction microscopy (HEDM), provide a means to digitally capture 3D
volumes that could be utilized to instantiate a simulation and directly compare with
experimental results. Figure 12 presents a roadmap and initial proof-of-concept
results for collecting and using 3D digital microstructures to instantiate CPFEM
simulations. Once collected, the digital microstructural dataset must be properly
segmented in order to properly identify grains and twins. Once identified, each