Image Processing in High Resolution Transmission Electron Microscopy of Inorganic Materials

Sammanfattning: High resolution transmission electron microscopy and digital image processing has been used to investigate the structure and properties of inorganic materials. The image processing was performed with the programming language Semper. All routines are listed in the appendix. The local order in the disordered mineral blatterite was investigated, and was found to extend over several unit cells. The compositional gradient across a GaAs/AlGaAs quantum well/wire interface was investigated using chemical lattice imaging and multivariate statistics. The lower interface was found to be gradual, while the upper interface was sharp. The local strain in InP quantum dots, grown on GaInP surfaces, was investigated using a modified cross-correlation technique that can handle large deviations. The measured strains were compared with theoretical calculations by the finite element method. The two methods gave comparable results. The location of the iron in an iron oxide implanted zeolite Y was determined using a combination of electron diffraction and crystallographic image analysis. A three-dimensional reconstruction showed that the iron was located in the sodalite cage of the zeolite framework. The determined atom positions were refined using electron diffraction intensities. The structure of a palladium phosphide, Pd₃P, cluster was determined using crystallographic image analysis. The three-dimensional reconstruction showed the structure to be closely related to the bulk structure of Pd₃P, the cementite type.