Tessmer_cmu_0041E_10566.pdf (23.63 MB)

Simulation of S/TEM Images for Discrete Dislocation Dynamics and Molecular Dynamics

Download (23.63 MB)
posted on 11.11.2020, 16:27 by Joseph Tessmer
Dislocations in crystalline materials have a strong impact on mechanical properties. As a result, identification of dislocation type and density has been a core part of materials characterization for decades. Transmission Electron Microscopy (TEM) is a commonly used technique for such characterization experiments. In recent years, modalities such as Scanning Transmission Electron Microscopy (STEM) have been used to produce images
with defect contrast (STEM-DCI). While there is a robust body of work describing defect contrast in conventional TEM, STEM-DCI is a developing field and thus lacks available comparisons in literature for many types and configurations of dislocations. This work combines dynamical electron diffraction simulations with material modeling techniques to produce simulated S/TEM images of dislocation microstructures. These images can be used to identify contrast features in experimentally captured images, as well as to provide a mechanism for the comparison of simulated microstructures with their real world counterparts. Two material models are considered, namely Molecular Dynamics (MD) and Discrete Dislocation Dynamics (DDD). For each of these, an approach to integrating the results with the scattering matrix formalism of dynamical electron diffraction is described. Then, modifications to the standard approach for scattering matrix computations are detailed which allow STEM images to computed in a computationally efficient manner. Next, simulated images for the same microstructure are presented for both the MD and DDD models, along with validation images comparing the results of these newly developed image simulations with those from existing work. Experimental and simulated images of similar dislocation structures are compared qualitatively, then TEM and STEM images simulated from MD simulations of dislocation dissociation in High-Entropy Alloys (HEAs)
are presented, followed by STEM images of a number of complex dislocation configurations simulated with DDD.
Finally, future work is proposed to integrate additional materials models with the STEMDCI framework presented here, as well as to perform a quantitative dislocation analysis experiment using simulated images.




Degree Type



Materials Science and Engineering

Degree Name

  • Doctor of Philosophy (PhD)


Marc De Graef