Annular Dark Field Scanning Transmission Electron Microscopy of Crystals.

New contributions to the annular dark field (ADF) scanning transmission electron microscopy (STEM) imaging theory are developed and discussed. Recently, ADF STEM imaging has produced experimental images with atomic resolution and atomic number contrast (Z-contrast), but the imaging theory is still not well understood. Due to the complicated non-linear nature of the ADF STEM imaging process, simple analysis is inadequate. Therefore, the primary tool in this research is large scale numerical simulation. This thesis describes the development, verification, and exploitation of the ADF STEM simulation. First, a simple imaging theory, valid only in the thin specimen limit, is presented to provide the basis necessary to discuss and analyze the ADF STEM simulation. This theory also introduces the fundamental concepts of Z-contrast, incoherent imaging, point scatterers, and visibility. Then the ADF STEM multislice simulation and the improved frozen phonon (FPh) simulation are derived, implemented, and executed to investigate the details of the imaging process. Implications of electron channeling, dynamical diffraction, and thermal vibrations are discussed. Experimental verification of the FPh simulation accuracy is demonstrated for both coherent diffraction scattering and incoherent thermal diffuse scattering. The FPh simulation is then used to demonstrate that ADF STEM imaging of zone axis crystals follows the simple incoherent imaging model. The incoherent imaging model allows the separation of the effects of the imaging conditions from the specimen structure information in the ADF STEM image. This result is an important advance toward quantitative atomic structure imaging and is verified experimentally. The strong linkage between simple theory, simulation, and experiment is compelling evidence that this simulation technique is correct, and therefore justifies its use to make further predictions without experimental verification. The FPh calculation is used to provide guidelines for experimental image interpretation, including the image dependence on specimen thickness, detector geometry, and thermal vibration amplitude. The fundamental limitations of ADF STEM imaging, in terms of the minimum difference in specimen composition visible at a given resolution, are also explored.