An EBSD study on mapping of small orientation differences in lattice mismatched heterostructures.

摘要

Electron back scatter diffraction (EBSD) on a Scanning Electron Microscope (SEM) has experienced rapid development in recent years. However, inadequate attention has been paid to the details of the method. Many of the algorithms in current use were chosen because they were the first ones that were found to work, not because they were optimum. The long term aim of this study is mainly concerned with extending EBSD to characterizing threading dislocations in semiconductor materials. In working toward this objective, a systematic study on the EBSD technique was performed.; The possibility of measuring small changes of orientation within grains by EBSD was explored. Conventional orientation maps (using EBSD) index the orientation of each position on the sample separately. This does not give accurate results for small differences of orientation. Therefore, methods of directly measuring small changes in orientation from one EBSD pattern to the others were examined. Previous workers have measured the change of position of zone axes in the EBSD pattern. A comparison between measuring changes of position of zone axes versus measuring the shift of the peaks in the Hough transform from one diffraction pattern to the next suggests that the latter method is superior. More over, it is possible, with a standard EBSD configuration, to measure the shift of the Kikuchi bands to a precision of about 0.1 pixels, which corresponds to a change of orientation of about 0.1 mrad.{09}This method has been successfully applied on a GaN/Sapphire structure. Based on this method, ways to perform high precision orientation mapping are proposed.; We have also shown that: (1) More than one method can successfully correct a sampling artifact, which is associated with the Hough transform; (2) There is an optimum binning ratio; (3) Gaussian filtering provides an alternative to “butterfly convolution”; (4) Better alternatives for mapping image quality than those in current use are available; (5) saving all the original patterns is practical and advantageous; and (6) Conventional usage of Monte-Carlo simulation for estimating the interaction volume of EBSD is flawed.