Diffraction methods provide easy access to the form and dimension of the crystal unit cell much more accurately than scanning probe methods. The diffraction of electrons is particularly suited to investigate the surface of a crystal since the electrons (at low energy 10eV .. 500eV) penetrate only the first few atomic layers. Furthermore, since the electrons experience multiple scattering processes before they leave the crystal again, the diffraction signal contains all information needed to reconstruct the true atomic structure of the crystaline surface. For this, the diffraction pattern has first to be recorded as function of electron energy. Secondly, since direct inversion of the diffraction signal is impossible (loss of scattering phase information), a suitable crystaline model has to be developed and the calculated energy dependence of the diffraction pattern has to be fitted to the experimental data. This is the same as with any other diffraction method this requires a "good guess" for the true surface structure. In our group these come from intuition, theoretical calculations (DFT - Density Functional Theory) and from STM topographies.
Surfaces in ambient conditions are normally covered by an adsorbate layer (water, hydrocarbons, etc.) which either renders the true crystal surface invisible or in some cases even changes the surface structure. Thus in our experiments we aim to prepare clean surface by operating in ultra-high vacuum (UHV) and preparing ultra clean thin films or nanostructures on surfaces. We have two UHV machines dedicated for this task. Other systems like the low-temperature STM (HADES) or the 2PPE system also contain a LEED setup to check cleanliness and structural quality of surface preparations.