Orientation-adaptive virtual imaging of defects using EBSD

Abstract

Electron backscatter diffraction (EBSD) is a foundational technique for characterizing crystallographic orientation, phase distribution, and lattice strain. Embedded within EBSD patterns lies latent information on dislocation structures, subtly encoded due to their deviation from perfect crystallinity — a feature often underutilized. Here, a novel framework termed orientation-adaptive virtual apertures (OAVA) is introduced. OAVAs enable the generation of virtual images tied to specific diffraction conditions, allowing the direct visualization of individual dislocations from a single EBSD map. By dynamically aligning virtual apertures in reciprocal space with the local crystallographic orientation, the method enhances contrast from defect-related strain fields, mirroring the principles of diffraction-contrast imaging in TEM, but without sample tilting. The approach capitalizes on the extensive diffraction space captured in a single high-quality EBSD scan, with its effectiveness enhanced by modern direct electron detectors that offer high-sensitivity at low accelerating voltages, reducing interaction volume and improving spatial resolution. We demonstrate that using OAVAs, identical imaging conditions can be applied across a polycrystalline field-of-view, enabling uniform contrast in differently oriented grains. Furthermore, in single-crystal GaN, threading dislocations are consistently resolved. Algorithms for the automated detection of dislocation-induced contrast are presented, advancing defect characterization. By using OAVAs across a wide range of diffraction conditions in GaN, the visibility/invisibility of defects, owing to the anisotropy of the elastic strain field, is assessed and linked to candidate Burgers vectors. Altogether, OAVA offers a new and high-throughput pathway for orientation-specific defect characterization with the potential for automated, large-area defect analysis in single and polycrystalline materials.