Formation of a Metastable Crystalline Phase During Ion Irradiation of Spinel
- PDF / 963,232 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 69 Downloads / 219 Views
Mat. Res. Soc. Symp. Proc. Vol. 398 @1996 Materials Research Society
-* 0.202 nm -p Octahedral Al
0
0
0
Al
0
Mg
-*-- 0.202 nm.-s.Tetrahedral Mg
Figure 1. The arrangement of cations in the close-packed anion lattice of spinel. RESULTS AND DISCUSSION The experimental electron diffraction patterns from unirradiated spinel and a region irradiated to 2x10 15 Xe++/cm 2 are shown in figures 2 (a) and (b), respectively. The first order reflections are missing in the diffraction pattern from the irradiated layer. The distributions of the damage and implanted ion in this layer were calculated using the TRIM [8] computer code and are shown in figure 3. The thickness of the irradiated layer predicted by TRIM shows reasonable agreement with that determined by cross-sectional transmission electron microscopy (TEM).
Figure 2. [001] electron diffraction patterns from (a) unirradiated MgA1204 (b) MgAI2 04 irradiated with 400 keV Xe++ at 100 K to 2x10 15 ions/cm 2 .
172
5
0.5 ----
0.4
dpa
4
= 0.3
3
I.°U C.)
2
0.2 oA
Q)U
S0.1 0.0
Al. 0
.... 50
. 1 100 150 Depth (nm)
00-. 200
250
Figure 3. Distributions of Xe ions and displacements in spinel for a dose of 2x10
15
ions/cm 2 .
The calculated diffracted intensity distribution from unirradiated normal spinel is shown in figure 4 (a) along with an experimental plot of this distribution from spinel irradiated to 2x 1015 Xe++/cm 2 shown in figure 4 (b). The disappearance of first order fundamental reflections, such as (220), following irradiation indicates a phase transformation. From the symmetry of the diffraction pattern in 2 (b) and similar patterns taken at other orientations, it is clear that the new phase is single crystalline and cubic, and has a lattice spacing that is half that of unirradiated spinel. Since the oxygen sublattice has this spacing, we started with the assumption that the anion sublattice remains intact while the cations exchange sites leading to inversion. However, inversion does not bring about significant changes in the electron diffraction pattern, because the scattering factors for Mg and Al are almost identical. The calculated values of the square of the structure factor are shown in figure 4 (c), as functions of the reciprocal of the interplanar distance, following complete inversion. From this plot, it is clear that cation site exchange on a 0.808 nm spinel lattice cannot explain this transformation. The lattice periodicity has to be halved to explain the extinction of the first order reflections. Therefore, we assumed a 0.404 nrn face-centered cubic oxygen lattice with cations randomly distributed at the interstitial positions. By choosing, different combinations of interstitials we arrived at different model lattices. Figure 4 (d) is a plot of the square of the structure factor for a rocksalt-like structure obtained by placing Mg and Al cations on three-quarters of the octahedral sites. For this model, the allodd reflections are extremely weak, which is inconsistent with our experimental observations. It must be pointed out
Data Loading...
