Search

The microscopic study of complex systems has reached a high level of accuracy that allows for a deep understanding of their structure, electronic properties, and optical spectra. The theoretical investigation of surfaces is nowadays routinely done within density functional theory, for ground state properties, and, with a larger computational load, within many-body perturbation theory, for excited states properties. In this paper we present and discuss examples of calculations for group IV two-dimensional systems such as a clean silicon surface, a tin–germanium interface, graphene, and graphane, pointing out the importance of a pertinent treatment of many-body effects.

The microscopic structure of the GaP(001) surface is not well understood, in marked contrast to other III–V compounds, in particular GaAs(001) (see Ref. [1] for a recent review). It has been suggested that ion bombardment and annealing of GaP(001) results in a (4 2) reconstructed Ga-rich surface [2, 3] in analogy to GaAs. On the other hand, based on ion scattering spectroscopy and low-energy electron diffraction (LEED) experiments, a (2 4) translational symmetry has been proposed [4, 5]. In solid-source molecular beam epitaxy a (2 4) diffraction pattern shows up under both Ga and P supply [6]. In addition, (4 4) spots have been observed. A slightly different behaviour is found for samples grown by metal-organic vapour phase epitaxy or chemical beam epitaxy as well as for GaP(001) surfaces prepared by thermal desorption of a protective arsenic/phosphorous double layer cap under ultrahigh vacuum conditions [7–9]. After annealing to 690 K the de-capped GaP(001) surfaces show a (2 1)=(2 2)-like LEED pattern. Annealing to higher temperatures (785 K) leads to a (2 4) translational symmetry. The many-step annealing procedure is accompanied by a transition from a more P-rich to a more Ga-rich GaP(001) surface. Previous total-energy (TE) calculations [8, 9] show the stability of several (2 4) reconstructions. For Ga-rich surfaces they are characterized by single Ga–Ga or mixed Ga–P dimers on top of a complete Ga layer. For lower Ga contents one Ga pair is missing in this layer and one (d structure) or two P dimers (b2 structure) terminate the surface. (4 2) reconstructions with Ga or P dimers in the uppermost atomic layer are unstable. The comparison of measured and calculated reflectance anisotropy spectra support the mixed-dimer reconstruction of the Ga-rich GaP(001) surface [8, 9]. This has recently been confirmed by a comparison of calculated and measured scanning-tunneling microscopy (STM) images [10]. The (2 4) mixed-dimer structure agrees with recent measurements of the surface core-level shifts [5]. The structure of the (2 1)=(2 2) reconstruction of the GaP(001) surface remains unknown. Auger electron spectroscopy indicates a P-rich surface. The similarities with

We present a calculation of the electronic and optical properties of the GaAs(110) : H surface performed within the first-principles density functional theory (DFT) in the local density approximation (LDA). The geometry and electronic structure are analyzed and compared with those of the clean surface. The reflectance anisotropy spectrum and the differential reflectivity are then computed. Taking into account the usual underestimation of the LDA gaps with respect to the experimental values, the comparison between our theoretical spectra with available experimental reflectance data shows a satisfactory agreement, allowing for the interpretation of the main structures in terms of electronic transitions.

We discuss the progress in realistic calculations of the optical response of semiconductor surfaces. It concerns numerical developments as well as the better inclusion of many-body effects. We show that the starting point of such calculations, the atomic structure and the surface equilibrium phase, can be now derived with high accuracy. The different steps are described in detail for the computation of the reflectance anisotropy and the exploration of the In-rich InP(001)2 4 surfaces.