511 29
to have fully transformed to ferrite. On further cooling, the
ferrite, super-saturated with Ni, then was thought to decom-
pose to the two phase ferrite plus austenite structure.
However, EBSD has shown that this is not be the correct
path for the observed microstructural evolution (Goldstein
and Michael 2006 ).
. Figure 29.24a is a large-area EBSD map acquired by
mapping smaller areas of about 1 × 1 mm and tiling together
220 of these tiles into a large area map that covers 22 × 10 mm
with a 3-μm step size. The sample was mechanically polished
using standard metallographic practice followed by a few
hours of vibratory polishing on colloidal silica. The entire
map shown consists of a more than 25 million individual pix-
els. The general microstructure at a low magnification is
clearly visible in. Fig. 29.24a.. Figure 29.24b shows a band
contrast image of the ferrite and the austenite as an inverse
pole figure map with respect to the sample surface normal.
Note that all of the austenite has the same or very close to the
same orientation, as shown by the austenite all of the same
color in the inverse pole figure map. This is an important
observation as the austenite could not have formed from pre-
cipitation from the ferrite but must be remnants of the origi-
nal large austenite grains found in the parent meteorite body
at elevated temperatures early in the meteorites life. This is
further demonstrated by the pole figures shown in
. Fig. 29.25. The austenite pole figures show that there is only
a single orientation of austenite in the 22 mm × 10 mm area.
The ferrite pole figures are much more complicated and are a
result of the many variants of ferrite that form from a single
orientation of austenite.
There are also regions in. Fig. 29.24a that are very fine
grained and difficult to resolve with the 3-μm step size used.
Further examination of the microstructure showed that these
regions were extremely fine grained and required higher res-
olution than can be achieved with using bulk EBSD. Due to
the small feature size in these areas, TKD is an excellent
method to utilize.. Figure 29.26 is a secondary electron
image of a focused ion beam produced thin sample. Also
shown is a scanning transmission electron image acquired at
30 kV which demonstrates that the sample is sufficiently thin
for the transmission of 30 kV electrons.. Figure 29.27 is the
resulting TKD map and phase information obtained from the
thin sample using an on-axis TKD detector. The step size for
this image was 4 nm. It is now clear from these images that
the fine-grained regions in. Fig. 29.24a consist of regions of
single crystal austenite that can be seen in. Fig. 29.24b but
also regions of ferrite that have begun to decompose during
cooling to the equilibrium austenite plus ferrite that would be
expected. The presence of twinned austenite precipitates is
somewhat surprising, but may be explained by some of the
stress in the sample during transformation.
This example shows how EBSD and TKD may be applied
to complex microstructures and how the use of TKD is
extremely complementary to EBSD. The visualization of theab. Fig. 29.24 a EBSD inverse
pole figure map with respect
to the sample normal direction
that is constructed by tiling 220
separate 1 × 1-mm maps. The
interesting Widmanstatten pat-
tern can be seen in the large
ferrite plates that are running
nearly the length of the image.
This map contains both indexed
austenite and ferrite; although
at this scale only the larger fer-
rite is visible. b This is the same
area as shown in. Fig. 29.23a,
but now we present the ferrite
as a band contrast or a measure
of the pattern sharpness and
the austenite in the colors of
the inverse pole figure map
with respect to the sample
normal direction. The amazing
observation from this EBSD data
is that the austenite has the same
orientation throughout the large
22 × 10-mm area, which leads
to the interpretation that the
austenite is retained from the
original parent body
29.2 · Electron Backscatter Diffraction in the Scanning Electron Microscope