3D Data for Fatigue in Superalloys
5
Fig. 2 The TriBeam microscope. The optics and beamline is contained within the red box on the
right. The electron and focused ion beam are indicated. The femtosecond laser and beamline are
directly aligned into the FIB-SEM via a coupled floating optics table
detector). Previous studies have shown that the damage resulting from femtosecond
laser ablation is limited to dislocation injection in structural materials [53, 54]. To
date, a wide range of materials including metals [55–57], ceramics [50], composites
[58], and semiconductors [59] have been imaged in 3D using the TriBeam. The
stock mechanically driven microscope stages are used to position the sample into
the scanned laser beam (down to 0.5–1 μm slice thickness), or custom attocube
piezoelectric stages can be utilized for slice thicknesses below 1 μm. However, the
reliability and stiffness of the stock microscope stages are superior.
A typical 3D nickel dataset contains several hundred slices, with each slice
requiring 1–100 min for acquisition, depending on the imaging modalities, imaging
resolution, and whether FIB cleanup is required. The femtosecond laser ablation
material removal step (1–3 min) is a very small fraction of the total slice time, which
is typically dominated by the resolution at which EBSD data is gathered and whether
FIB cleanup is performed.
The data in Table 1 shows the slice times that would be required for a hypothetical
collection of a 1 mm 3 volume TriBeam dataset, with and without FIB cleanup
and with 1 μm cubic voxels. Ga+ FIB cleaning requires approximately 1 min per
every 20,000 μm 2 at glancing angles between 3 and 10 ◦ . The time required for
cleanup does not change with glancing angle because the FIB dosage per area
is held constant, resulting in increased dwell times at more glancing FIB beam
angles. Experiments that do not require a FIB cleanup step reduce the total cycle
time significantly, as shown in the last column in Table 1. Materials that do not
require FIB cleanup in order to obtain acceptable quality EBSD patterns generally
5
Fig. 2 The TriBeam microscope. The optics and beamline is contained within the red box on the
right. The electron and focused ion beam are indicated. The femtosecond laser and beamline are
directly aligned into the FIB-SEM via a coupled floating optics table
detector). Previous studies have shown that the damage resulting from femtosecond
laser ablation is limited to dislocation injection in structural materials [53, 54]. To
date, a wide range of materials including metals [55–57], ceramics [50], composites
[58], and semiconductors [59] have been imaged in 3D using the TriBeam. The
stock mechanically driven microscope stages are used to position the sample into
the scanned laser beam (down to 0.5–1 μm slice thickness), or custom attocube
piezoelectric stages can be utilized for slice thicknesses below 1 μm. However, the
reliability and stiffness of the stock microscope stages are superior.
A typical 3D nickel dataset contains several hundred slices, with each slice
requiring 1–100 min for acquisition, depending on the imaging modalities, imaging
resolution, and whether FIB cleanup is required. The femtosecond laser ablation
material removal step (1–3 min) is a very small fraction of the total slice time, which
is typically dominated by the resolution at which EBSD data is gathered and whether
FIB cleanup is performed.
The data in Table 1 shows the slice times that would be required for a hypothetical
collection of a 1 mm 3 volume TriBeam dataset, with and without FIB cleanup
and with 1 μm cubic voxels. Ga+ FIB cleaning requires approximately 1 min per
every 20,000 μm 2 at glancing angles between 3 and 10 ◦ . The time required for
cleanup does not change with glancing angle because the FIB dosage per area
is held constant, resulting in increased dwell times at more glancing FIB beam
angles. Experiments that do not require a FIB cleanup step reduce the total cycle
time significantly, as shown in the last column in Table 1. Materials that do not
require FIB cleanup in order to obtain acceptable quality EBSD patterns generally