HREM Profile Image Interpretation In MgO Cubes

Abstract

For the most credible interpretation of HREM images, it is desirable to match computed and experimental images against the variation of some parameter, usually either specimen thickness or defocus (and preferably both). MgO smoke forms in perfect crystalline cubes, so that the thickness, T, of a cube in <110> zone axis orientation can accurately be determined by geometry at each co-ordinate X (normal to the wedge edge) in the image. Figure 1 shows such an experimental 90° MgO wedge lattice image recorded in an exact <110> orientation using a JEOL 200CX. Dynamical multislice calculations for the thickness dependence of the inner beams important for imaging are shown (c) scaled to the abscissa of the experimental image (the crystal thickness, T, at any point is twice the X value shown). Analysis of this image by means of simulated images (not shown) shows the thin-crystal “structure” image changing to one containing vertical (220) fringes at A (due to strong second-order (111) x (111) interference), to a re verse-contrast image at C, to one with horizontal (002) fringes at B (from strong (111) x (111) interference), and to a “structure“-like image at D. Matching with simulated images provides a rare possibility for guantitative structure analysis by HREM, since the complex low order structure factors can be adjusted in the dynamical calculation until the contrast reversals and other turning points occur at the observed thicknesses.