Abstract
The diffraction contrast caused by large inclusions (of inclusion diameter greater or equal to one extinction distance) is investigated using the modified Bloch wave theory. Analytical equations are derived for the bright- and dark-field intensities in the two-beam case. These equations describe the dependence of the contrast on the phase and amplitude difference between the modified Bloch waves as well as the dependence on the boundary conditions at the electron beam entrance and exit surfaces.
It is found that the contrast can be explained by ‘phase contrast’, originated by the movement of the excitation points of the modified Bloch waves on the dispersion branches denoted as intraband scattering of modified Bloch waves. Interband scattering can be neglected. The phenomenon called ‘contrast fringes’, observable under dynamic imaging conditions when the inclusion is lying near the foil centre, is explained exclusively by an additional phase difference originating from the extended strain field in the matrix. In addition, anomalously wide asymmetrical images (black-white contrast found when the defect is located close to a foil surface) are caused by surface stress relaxations which change the boundary conditions for the electron waves compared with those in the perfect crystal.
