94
D. W. Eastman et al.
on microtensile samples of polycrystalline Ni [22, 23]. Using a dual beam FIBSEM system, multiple samples of a width of 21 μm, a thickness of 38 μm, and a
gage length of 80 μm were characterized in 3D and in combination with surface
strain maps collected during testing used to later benchmark a finite element model
using these explicit representations of microstructure [24]. While these examples
demonstrate the ability of the FIB as a sectioning tool, it is clear that sample size
and material limitations due to the material removal rate of the FIB can limit what
can be done despite the nanometer level of precision that can be achieved.
While the traditional Ga FIB has become a widely used tool in microscale
machining and materials research, the more recent development of the Plasma
FIB (PFIB), using Xe rather than Ga as in a traditional FIB, allows for material
removal rates at orders of magnitude higher. The PFIB can potentially even incur
less damage than a traditional FIB, as reported by Kwakman et al., than compared
to Ga FIB machining; the damage layer when using Xe FIB could be reduced by up
to 25% [25]. Xiao et al. demonstrated that micropillars of Al, a material particularly
susceptible to the effects of Ga implantation, machined with a Ga FIB showed a
lower yield strength than micropillars fabricated with Xe FIB [26]. Burnett et al.
have demonstrated the use of PFIB in serial sectioning and 3D characterization of
a 150 × 120 × 80 μm 3 WC-Co sample using 790 slices of 100 nm thickness with
a pixel resolution of 30 nm [27]. The step forward that these two tools provide in
terms of 3D characterization is very promising, and optimization for sectioning at a
smaller length scales could allow for the collection of statistics and 3D datasets for
benchmarking at a rate that was previously inaccessible.
A more recently introduced tool for the collection of 3D microstructural data at
a larger scale is the TriBeam system, which incorporates a femtosecond laser into
a FIB-SEM dual beam platform. The material removal rate of the laser allows for
material ablation two orders of magnitude faster than is possible with traditional
FIB. In addition, the ablation rate of the femtosecond laser is more materials
agnostic [28]. As a result, the TriBeam is an incredibly versatile system in terms
of both speed and scale. One successful application of the TriBeam is in its use by
Stinville et al. to characterize crack nucleation in René 88DT, in which tested fatigue
samples were subsequently sectioned and characterized in the TriBeam to determine
what features in the microstructure led to crack nucleation and propagation [29].
Another promising technique for characterizing material in 3D is the nondestructive technique of High-Energy Diffraction Microscopy (HEDM). HEDM utilizes
diffraction patterns produced by a monochromatic beam from a synchrotron source
to interrogate a volume of material. Two sets of detectors are used to collect
information from the sample. One detector is located only a few mm from the
sample and is used for near-field HEDM, which provides orientation maps with
spatial locations of grains as well as their morphology and misorientation relative
to neighboring grains [30]. The second detector is placed much farther away and
is used for far-field HEDM, which provides grain centroids and elastic strain
tensors [30]. These combined datasets provide a robust representation for the sample
microstructure, as well as subgrain information [31]. Many examples of using
this technique for 3D characterization have been demonstrated [32–37], and its
D. W. Eastman et al.
on microtensile samples of polycrystalline Ni [22, 23]. Using a dual beam FIBSEM system, multiple samples of a width of 21 μm, a thickness of 38 μm, and a
gage length of 80 μm were characterized in 3D and in combination with surface
strain maps collected during testing used to later benchmark a finite element model
using these explicit representations of microstructure [24]. While these examples
demonstrate the ability of the FIB as a sectioning tool, it is clear that sample size
and material limitations due to the material removal rate of the FIB can limit what
can be done despite the nanometer level of precision that can be achieved.
While the traditional Ga FIB has become a widely used tool in microscale
machining and materials research, the more recent development of the Plasma
FIB (PFIB), using Xe rather than Ga as in a traditional FIB, allows for material
removal rates at orders of magnitude higher. The PFIB can potentially even incur
less damage than a traditional FIB, as reported by Kwakman et al., than compared
to Ga FIB machining; the damage layer when using Xe FIB could be reduced by up
to 25% [25]. Xiao et al. demonstrated that micropillars of Al, a material particularly
susceptible to the effects of Ga implantation, machined with a Ga FIB showed a
lower yield strength than micropillars fabricated with Xe FIB [26]. Burnett et al.
have demonstrated the use of PFIB in serial sectioning and 3D characterization of
a 150 × 120 × 80 μm 3 WC-Co sample using 790 slices of 100 nm thickness with
a pixel resolution of 30 nm [27]. The step forward that these two tools provide in
terms of 3D characterization is very promising, and optimization for sectioning at a
smaller length scales could allow for the collection of statistics and 3D datasets for
benchmarking at a rate that was previously inaccessible.
A more recently introduced tool for the collection of 3D microstructural data at
a larger scale is the TriBeam system, which incorporates a femtosecond laser into
a FIB-SEM dual beam platform. The material removal rate of the laser allows for
material ablation two orders of magnitude faster than is possible with traditional
FIB. In addition, the ablation rate of the femtosecond laser is more materials
agnostic [28]. As a result, the TriBeam is an incredibly versatile system in terms
of both speed and scale. One successful application of the TriBeam is in its use by
Stinville et al. to characterize crack nucleation in René 88DT, in which tested fatigue
samples were subsequently sectioned and characterized in the TriBeam to determine
what features in the microstructure led to crack nucleation and propagation [29].
Another promising technique for characterizing material in 3D is the nondestructive technique of High-Energy Diffraction Microscopy (HEDM). HEDM utilizes
diffraction patterns produced by a monochromatic beam from a synchrotron source
to interrogate a volume of material. Two sets of detectors are used to collect
information from the sample. One detector is located only a few mm from the
sample and is used for near-field HEDM, which provides orientation maps with
spatial locations of grains as well as their morphology and misorientation relative
to neighboring grains [30]. The second detector is placed much farther away and
is used for far-field HEDM, which provides grain centroids and elastic strain
tensors [30]. These combined datasets provide a robust representation for the sample
microstructure, as well as subgrain information [31]. Many examples of using
this technique for 3D characterization have been demonstrated [32–37], and its