Surface x-ray-diAraction measurements are presented that show a reversible (1X2)~(1X1)phase transition of the Ge(001) surface. The variation of the (1X2) superlattice reAection intensity with temperature gives a transition temperature of T, =955+7 K. The data are interpreted as being due to the creation of adatoms and vacancies on the surface with consequent break up of surface dimers. X-ray reflectivity indicates a corresponding loss of height-height correlation across the surface. A simple three-level model is used to describe the reflectivity, and the results are compared with a simple Monte Carlo simulation of the transition.

In an X-ray diffraction study of the c(2 x 8) reconstructed Ge( 11 1) surface we have measured intensity profiles along integer-order rods of Bragg scattering perpendicular to the surface plane. The diffracted intensity profiles are explained in terms of a simple adatom model for the reconstruction. The adatoms are found to occupy T4 sites on a distorted bulk-like substrate. We rule out any model for c(2 x 8) reconstruction that is based on stacking faults in the substrate. For the adatom model we give best-fit values for the positions of the atoms at the surface and compare them with those obtained from total-energy and Keating-energy minimisation calculations.

We have investigated by the use of surface x-ray diffraction the initial strain relaxation of Ge on Si(001) when a ‘‘surfactant’’ layer of Sb [0.7 monolayers (ML)] is present. Due to the high sensitivity of the technique to lateral strain in the overlayer, we have been able to observe directly the onset of strain relaxation at a coverage of 9–10 ML. This strain relief proceeds gradually as a function of coverage, but unlike the case without an Sb surfactant, it was not possible to relax the overlayer fully. No bulklike Ge was seen even up to a coverage of ≊55 ML. Concurrent specular reflectivity measurements also showed that the overlayer formed in a layer-by-layer mode up to the same level. These results differ quite markedly from those obtained without the surfactant layer, where several stages of island formation and strain relaxation are seen. They show dramatically how the modification of surface energies by the presence of a surfactant can affect the morphology of, and strain in, a lattice mismatched overlayer. In addition, they are direct confirmation that the initial strain relaxation mechanism in operation when a surfactant is not present is the formation of coherent Ge islands, rather than that of dislocations.

No abstract

No abstract