Figure 5.6 Same as Figure 5.4b but after the fractured surface of PZN-7%PT crystal sample was polished with SiC papers of different particle sizes... After repeated polishing with care t
Trang 1Figure 5.6 Same as Figure 5.4(b) but after the fractured surface of
PZN-7%PT crystal sample was polished with SiC papers of different particle sizes
Trang 2tweed-like patterns (Figure 5.7c) After repeated polishing with care to remove a
sufficiently thick layer (>40 µm) from the surface of the sample, the surface domain
patterns as shown in Figure 5.7(a) persisted; and so was the presence of the broad
lower 2θ peak in the XRD profile These observations led us to believe that the
specific thin domain layer is related to the polished surface To check for this, a
PZN-4.5%PT sample, initially polished in the [100]pc direction (Figure 5.7a), was
tilted by an angle of 45° and again polished back and forth in the new direction After
the second polishing, the earlier set of domains disappeared and was replaced by a new
set of elongated domains aligned at an angle of about 30° towards the <110>pc
direction of polishing (Figure 5.7b) This confirms the observed domain patterns are
related to the polished surface We further confirmed that optically the polishing
induced surface domains can be reliably detected only if polishing direction was
roughly parallel to a certain crystallographic direction, e.g., [010]pc or [110]pc in our
experiments Otherwise, the domains are not evident and the sample is covered with
areas exhibiting slight but varying birefringence when observed under the crossed
polarizers (Figure 5.8) Despite the above, the lower 2θ peak is always present in the
x-ray profile following any direction of polishing, suggesting that the surface phase is
always present regardless of the direction and/or mode of polishing
This behavior can be explained in the following way If the polishing
Trang 3Figure 5.7 Surface domain patterns of the (001)-cut PZN-4.5%PT crystal
plate (a) after polishing along the [010]pc direction and (b) after repolishing in the [110]pc direction Arrows indicate the direction
of polishing Note the realignment from (a) to (b) (c) The domain patterns in the underlying material revealed by the focusing technique
[100]
[010]
a
c
b
50 µm
Trang 4[100]
[010]
50 µµµm
Figure 5.8 No clear surface domain patterns of (001)-cut PZN-4.5%PT
crystal plate as a result of none crystallographic polishing direction
Trang 5direction changes in a non-coordinated manner, the directions of induced surface
stresses and domain directions are randomly distributed As the size of domains is very
small, they are not visible in the optical microscope In this case, since the surface
layer is composed of randomly oriented submicroscopic domains, the surface layer
does not exhibit a clear birefringent domain pattern In contrast, after directional
polishing the domains (and thus their optical indicatrix) are aligned into macroscopic
regions These regions are birefringent and can be identified under a polarizing
microscope
5.3.1 Thermal stability
The stability of the deformed surface layer was investigated by subjecting the
as-polished samples to different annealing treatments The heating and cooling rates
used were 1.5 ºC/min Figure 5.9 shows the (002) XRD profiles of the differently
annealed PZN-4.5%PT samples and the (002) XRD profile of the as-polished sample
After annealing for 1 h at 300 °C, the position of the lower 2θ peak shifted to 2θ ≈ 43.95° Subsequent annealing at 400 °C for 1 h did not produce significant change to
either the position or intensity of this peak The lower 2θ peak persisted even after
annealing at 600 °C for 5 h and the position shifted further to 2θ ≈ 44.05° (Figure 5.9)
Trang 62θ
as-polished 300oC (1h) 400oC (1h) 600oC (5h)
<002>R
~43.50o
~44.05o
~44.68o
Figure 5.9 (002) XRD profiles of as-polished (solid curve) and differently
annealed (dashed curves) PZN-4.5%PT showing the effects of the
different annealing treatments on the lower 2θ peak Sample
thickness is 1mm