500
29
For EBSD, we tilt the sample at a high angle with respect
to the electron beam. This tilt is important to obtaining the
high quality patterns that are needed for EBSD. As shown in
. Fig. 29.13, the tilt angle has a large effect on the pattern
quality. The best pattern qualities are obtained at higher tilt
angles with some optimal condition between the pattern
quality and the ease of imaging of the highly tilted sample.
. Figure 29.14a shows the backscattered electron energy dis-
tribution that is developed from a highly titled sample (70°
tilt) compared to 0° tilt. Note that the energy is highly peaked
toward the operating voltage of the microscope at 70° tilt
when compared to the same distribution from an electron
beam that is normal to the sample surface. Further, since the
EBSD pattern is clearly visible, they must form from only the
electrons that have lost either no or a small amount of their
energy. The electrons that have lost only a small amount of
their energy are emitted from the sample in regions that are
very close to the original beam foot print on the surface of
the sample. This is shown in. Fig. 29.14b, which is a Monte
Carlo simulation of the distance from the initial beam impact
point that the electrons emerge from the sample surface for
all the backscattered electrons, those that lost up to 2 kV and
those that lost only 0.2 kV. The pattern is formed from these
low loss electrons and thus the resolution is quite good. Thus,
although EBSD involves backscattered electrons, it is a tech-
nique with much higher spatial resolution than standard
backscattered electron imaging. Typically, resolutions of bet-
ter than 0.1 μm can be achieved for a beam voltage of 20 kV
in the transition metals. The electrons that have lost larger
amounts of energy will contribute the background intensity
in the EBSD pattern. These electrons are the main reason
why there is limited contrast in EBSD patterns without some
form of background removal either through division or sub-
traction or some other computational method of removal
(Michael and Goehner 1996 ).
The high tilt angle limits the spatial resolution attainable
due to the elongation of the electron beam foot print on the
sample surface. The resolution parallel to the tilt axis is much
better than the resolution perpendicular to the tilt axis due to
the high sample tilt angles used to acquire EBSD patterns.
The resolution perpendicular to the tilt axis is related to the
resolution parallel to the tilt axis by
L=perp L/para()1cosθ
(29.3)
a
c
b
. Fig. 29.13 These series of EBSD patterns show the effect of sample tilt on the EBSD pattern quality. a 60°, b 50°, and c 40°
Chapter 29 · Characterizing Crystalline Materials in the SEM