Stresses and Textures in Thin Films – Selected Peculiar Examples
R. Kužel1, M. Novotný2, J. Buršík3
1 Department of Condensed Matter Physics, Faculty
of Mathematics and Physics, Charles University in Prague, Prague, Czech
Republic
2 Institute of Physics, Academy of Sciences of the Czech Republic, Prague, Czech Republic
3 Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic, Rez near Prague, Czech Republic
kuzel@karlov.mff.cuni.cz
Stresses and textures play an important role for the properties and applications of many different thin films. Sometimes they are introduced artificially in order to improve various magnetic, electrical and mechanical properties in other cases they are parasitic and should be avoided. X-ray diffraction is indispensable tool for their characterization and it is more and more applied. Routine fast XRD applications can, however, give misleading conclusions or just substantial characteristics of particular thin films can be easily overlooked. The residual stresses are usually determined from diffraction peak shifts or better from the well-know sin2y method. Preferred orientation can be quickly estimated by the texture indices from the symmetrical q-q scans or if it is stronger by the so-called ω (y) and j scans as cross-sections of the pole figure. Of course, full pole figures and ODF can also be measured and calculated. However, this is not always necessary and possible. In case of very strong textures, small steps must be selected and obtaining of full figure may be time-consuming. For very thin films, glancing-angle parallel beam diffraction must be applied and possibilities of full analysis are limited.
In this contribution, a few peculiar examples of the films with high technological interest are selected. In the studied cubic KTaO3 thin films rather unusual multicomponent textures were discovered when each of the components was quite narrow. These components were not clearly visible at the beginning but they were revealed by gradual texture component stripping in terms of subsequent fast y-scans on several diffraction peaks. Next examples were very thin films (40-80 nm) of hexagonal ZnO deposited on three different substrates – amorphous fused silica (FS), cubic MgO (100) and sapphire Al2O3 (0001). SEM pictures showed that all the films were nanocrystalline and hence 2q scans were taken showing several weak peaks. Symmetrical scans showed strong (000l) film orientation for the first two substrates but nearly no signal for sapphire. FWHMs of Ω-scans gave the values of 10º and 1.2º, for FS and MgO, respectively and j-scans showed fiber texture. However, these scans which must be taken in asymmetric orientation (inclined planes) do not have a good meaning for very narrow textures as it is in the latter case. Therefore, the same scans but for asymmetric (hkil) planes were performed. The scans for h0l peaks showed clear narrow 12 maxima indicating local epitaxy. This number – double with respect to the expected crystallographic symmetry can be explained by domains, in principle two equivalent orientation of hexagonal basal plane on cubic (100). Combined Williamson-Hall plots constructed from different asymmetric reflections taken at different angles y showed the presence of microstrain, in addition to the effect of small crystallite size. Residual stresses could be measured at very narrow range. Therefore maps of several asymmetric peaks were taken. These and the shifts of basal peaks indicated large compressive stress (up to 14 GPa) on MgO substrate that is highly strained, tensile stress on FS and zero stress on sapphire. This last case was very surprising and strange since strong orientations of a few domains was found more by an accident and no simple plane was oriented parallel to the surface as it could be expected.