Image Deconvolution - An Effective Tool of Crystal Structure and Defect Determination in High-Resolution Electron Microscopy

Image deconvolution is introduced as an effective tool to enhance the determination of crystal structures and defects in high-resolution electron microscopy. The essence is to transform a single image that does not intuitively represent the examined crystal structure into the structure image. The principle and method of image deconvolution together with the related image contrast theory, the pseudo weak phase object approximation (pseudo WPOA), are briefly described. The method has been applied to different types of dislocations, twin boundaries, stacking faults, and one-dimensional incommensurate modulated structures. Results on the semiconducting epilayers Si0.76Ge0.24/Si and 3C-SiC/Si are given in some detail. The results on other compounds including AlSb/GaAs, GaN, Y0.6Na0.4Ba2Cu2.7Zn0.3O7-δ, Ca0.28Ba0.72Nb2O6 and Bi2.31Sr1.69CuO6+δ are briefly summarized. It is also shown how to recognize atoms of Si from C based on the pseudo WPOA, when the defect structures in SiC was determined at the atomic level with a 200 kV LaB6 microscope.