Cation diffusion along basal dislocations in sapphire Original Research Article

The diffusion behavior of impurity cations was examined in deformed α-Al2O3 single crystals (sapphire) that had a high density of unidirectional basal dislocations. This behavior was examined in the temperature range of 1150–1400 °C by secondary ion mass spectrometry depth profiling techniques. In order to investigate different pipe diffusion behaviors with different diffusing elements, diffusion behaviors of Cr and Ti were studied. Cr was expected to exhibit behavior similar to that of Al because both are isovalent and because of the complete mutual solubility of Cr2O3 and Al2O3 at high temperatures. Ti was selected because it is representative of an element that is hardly soluble and thus is expected to segregate at dislocations. The lattice and pipe diffusion kinetics were best described by the equations image (m2 s–1) and image (m4 s–1) for Ti, respectively and by equations image (m2 s–1) and image (m4 s–1) for Cr, respectively. A drastic decrease in the activation energy for Ti penetration was observed; however, such behavior was not observed for Cr penetration. This fact well explains that the electrical conductivity of deformed sapphire, which results from diffused Ti nanowires along the dislocations of the crystal. The behavior of Cr was similar to that of 18O, examined previously, in that their activation energies for lattice and pipe diffusion were similar. Because Cr diffusion was expected to mimic cation self-diffusion, it was assumed that the activation energy for pipe diffusion was not very low compared to that for bulk diffusion. This assumption is consistent with the fact that the activation energy for grain boundary diffusion in alumina and other ceramics is equal to or higher than that of bulk diffusion. Thus, the absolutely low temperature dependence of Ti pipe diffusion in sapphire may not be because of the enhanced migration of Ti along dislocations but because of other effects, such as the segregation of Ti or Ti clustering at dislocations.