Stem Holography

Abstract

Electron optical imaging is a wave optical process. Therefore, to obtain the whole information about the object structure, one must record both amplitude and phase of the electron in the detector plane, e.g. in the Fourier plane or in the image plane. In the general case, an electron image detected only by its intensity is in principle incomplete [1]. This is true in particular in face of the lens aberrations which scramble amplitude and phase hence limit resolution and interpretability of an electron image.Following Gabor’s proposal of electron holography, these problems are being solved in the TEM [2], and in the STEM we try to tackle them in the following way [3] (fig.l): Two coherent waves, produced by means of the Mollenstedt electron biprism, are focused in the object plane. One of them is transmitted through the object and modulated in amplitude a(x,y) and phase φ(x,y) whereas the other one instead goes through empty space. Subsequently they are superimposed and give rise to an interference pattern (spatial frequency Rc) in the detector plane (coordinate u ) as