Three-Dimensional STEM Imaging of Dislocations

Oveisi, Emad (Oveisi, Emadeddin)
Advisor: Hébert, Cécile

To gain further insight into the fabrication and to improve the mechanical and optoelectronic properties of materials at the nano-scale, new experimental methods that permit thorough but fast and reliable multi-dimension microstructural and defect analyses are needed. The present contribution promotes the use of scanning transmission electron microscopy (STEM) as an advantageous approach compared to conventional TEM for two- and three-dimensional defect analysis in thick metallic specimens. Annular dark-field (ADF) STEM imaging of dislocations reveal a number of characteristic contrast effects that are shown to depend on both angular detection range and the specific position of the dislocation in the foil. In order to comprehend better the mechanisms underlying these contrast features, STEM dislocation contrast was systematically studied as a function of ADF detection angle, and a rationale based on electron channelling and Bloch-wave theories is proposed to account for such complex contrast phenomena. It is demonstrated that the character and position of a dislocation in the foil can be distinguished simply through a comparison between the images acquired at low- and medium-angle annular detection ranges. Due to its inherent projection nature, a single transmission electron microscopy micrograph is not always sufficient to unravel a complex configuration of dislocations; to do so multiple micrographs in the form of tilted stereo pairs or tilt-series are usually acquired. Here a very efficient and rapid method, in which both image acquisition and three-dimensional reconstruction problems are addressed, is developed that provides a reliable insight into the 3D shape of line dislocations. Instead of tilting the specimen, this method uses the convergent illumination of STEM imaging mode to record a stereoscopic pair of micrographs for a single sample tilt. These stereo images, whose effective viewing directions are separated by only a few degrees, are then treated with an in-house algorithm to yield a reliable 3D reconstruction of the dislocation arrangement.


Related material