High resolution images of a reconstructed surface structure on (111) gold platelets: Interpretation and comparison with theoretical models

Reconstruction of the (111) gold surface has been previously observed, and it has been shown that the surface net is compressed in the <110> direction, as compared with the bulk lattice. We present here some high resolution studies made on this surface reconstruction: the problem was to obtain a structure image corresponding mainly to the uppermost plane which forms the surface. For this, we have formed high resolution images by interferences between the so-called forbidden diffracted beams $\text{1}\text{3}$ (4?22): bulk gold, when observed along the <111> direction, is better described as a stacking of hexagonal two-dimensional layers, and the $\text{1}\text{3}$ (4?22) beams occur as diffracted beams from 1 (or2) hexagonal layers; they nearly disappear every 3 slices. If the top layer is reconstructed, the diffracted beams from the first slice will be different and will not cancel with the diffracted beams from the bulk layers. Hence the image formed with these beams will give some information on the reconstructed top layer. Computed images have been obtained for different models of the reconstructed surface. They show that, indeed, images obtained under these conditions are linked with the topography of the surface. They also show the importance of different experimental parameters, in particular the beam divergence.