COMPARISON OF EBSD AND SLEEM IN ULTRA FINEGRAINED MATERIALS CHARACTERISATION

摘要

The information contained in scanning low-energy electron microscopy (SLEEM) images and electron backscatter diffraction (EBSD) maps both stem from orientation dependence of contrast, in the first case, and of diffraction pattern appearance in the latter one. Both tools seem to be suitable for measuring the grain size distribution in ultra fine-grained (UFG) materials, especially when the grain refinement is achieved by severe plastic deformation, which is known to leave the microstructure inhomogeneous in terms of grain size. Since the grain size distribution in UFG materials spans over a few orders of magnitude, it is somewhat difficult to set appropriate step size and dimensions of an EBSD scan to fully characterize the microstructure in a reasonable time. An advantage of SLEEM over EBSD in this case should be the possibility to capture single high-resolution image of a large area, containing both the small and large grains, providing sufficient contrast and ability to be analyzed by conventional image analysis tools. A comparison was made between grain size distribution measurements based on SLEEM images and on EBSD data. The material under investigation was 99.9 % pure copper, processed by equal channel angular pressing (ECAP) method, following the processing route "Bc" in eight steps. The plane of observation was oriented so as to intersect the plane of shear during the last ECAP pass. SLEEM image data were acquired using two different kinds of instrument: one with conventional (high) vacuum system and another capable of ultra-high vacuum operation (UHV), both equipped with cathode lens to decelerate the primary beam. The system used to acquire EBSD data was a thermal emission type SEM with high vacuum system. The results showed that UHV SLEEM is superior to the high vacuum one in achieving better orientation contrast, but the particular UHV instrument suffered from image distortion at wider fields of view, therefore, image stitching would be necessary to cover the same area at the cost of time and inaccuracies at stitches. Based on our experience with long term stability of the particular EBSD system, the time of EBSD scan needed to be 4 hours at longest to avoid excessive drift and beam intensity decay. This limitation dictated the largest field of view of the EBSD map of 30 × 40 μm, resulting in upper equivalent grain diameter limit of approx. 7 μm, when discarding the edge grains. The grain size distributions resulting from image analysis of series of separate SLEEM pictures plotted as area fraction versus equivalent grain diameter were in a good agreement within the dimensional scale covered by both instruments. Grain size distributions calculated from EBSD data agreed with those from SLEEM images when the "grain tolerance angle" parameter was set to 5°.