Image contrast in mirror and low energy electron microscopy

2017-01-13T01:24:22Z (GMT) by Kennedy, Shane Michael
We develop several approaches to understand and interpret image contrast in mirror electron microscopy (MEM) and low energy electron microscopy (LEEM), with potential applications to photoemission electron microscopy (PEEM). We treat both the forward problem, of how surface features and properties create image contrast, and the inverse problem, of how we may infer quantitative information about surface features and properties from experimental MEM, LEEM and PEEM images. The thesis begins with the development of the Laplacian imaging theory of MEM, whereby image contrast is understood as the second derivative of the surface topography, blurred slightly to account for the interaction of the electron beam with the electric field above the specimen, rather than the specimen surface itself. This intuitive method includes the effects of lens aberrations and can be rapidly inverted to recover the surface topography from experimental MEM images. For specimen surface variations that are outside the regime of the Laplacian imaging theory and other models, we develop a caustic imaging theory for MEM. This involves solving the electric field above the specimen and tracing a family or envelope of rays through the immersion lens. Where initially adjacent rays cross, caustics are created, and these strong image features may be used to recover three dimensional surface topography. Both the Laplacian imaging theory and the caustic imaging theory are successfully applied to experimental MEM data to obtain the surface topography. As a complement to this ray-based treatment, we then develop a wave optical treatment of LEEM image contrast, adopting the complex transfer function methodology from transmission electron microscopy. This method includes spherical and chromatic aberration, and may be extended to include higher order aberrations for use in aberration corrected LEEM instruments. With knowledge of the complex transfer function, we then apply phase retrieval methods to simulated LEEM images, recovering the electron wave function and surface topography for a series of step terraces. Finally, we consider a wave optical treatment of MEM, investigating the behaviour of the electron wave in the vicinity of the turn around region. This is extended to explore the application of MEM beyond specimen surfaces, and the feasibility of imaging very weak potentials, such as the ponderomotive potential experienced by an electron in a light field.