The oxidation of Pd(111) leads to an incommensurate surface oxide, which was studied by the use of scanning tunneling microscopy, surface x-ray diffraction, high resolution core level spectroscopy, and density functional calculations. A combination of these methods reveals a two-dimensional structure having no resemblance to bulk oxides of Pd. Our study also demonstrates how the atomic arrangement of a nontrivial incommensurate surface can be solved by molecular dynamics in a case where experimental techniques alone give no solution.

The oxidation of the Rh(111) surface at oxygen pressures from 10(-10) mbar to 0.5 bar and temperatures between 300 and 900 K has been studied on the atomic scale using a multimethod approach of experimental and theoretical techniques. Oxidation starts at the steps, resulting in a trilayer O-Rh-O surface oxide which, although not thermodynamically stable, prevents further oxidation at intermediate pressures. A thick corundum like Rh2O3 bulk oxide is formed only at significantly higher pressures and temperatures.

The oxidation of the Pd(100) surface at oxygen pressures in the 10(-6) to 10(3) mbar range and temperatures up to 1000 K has been studied in situ by surface x-ray diffraction (SXRD). The results provide direct structural information on the phases present in the surface region and on the kinetics of the oxide formation. Depending on the (T,p) environmental conditions, we observe either a thin (sqrt[5]xsqrt[5])R27 degrees surface oxide or the growth of a rough, poorly ordered bulk oxide film of PdO predominantly with (001) orientation. By either comparison to the surface phase diagram from first-principles atomistic thermodynamics or by explicit time-resolved measurements we identify a strong kinetic hindrance to the bulk oxide formation even at temperatures as high as 675 K.

The active phase of Pd during methane oxidation is a long-standing puzzle, which, if solved, could provide routes for design of improved catalysts. Here, density functional theory and in situ surface X-ray diffraction are used to identify and characterize atomic sites yielding high methane conversion. Calculations are performed for methane dissociation over a range of Pd and PdOx surfaces and reveal facile dissociation on either under-coordinated Pd sites in PdO(101) or metallic surfaces. The experiments show unambiguously that high methane conversion requires sufficiently thick PdO(101) films or metallic Pd, in full agreement with the calculations. The established link between high activity and atomic structure enables rational design of improved catalysts.

The Pd (100) Combining high-resolution core-level spectroscopy (HRCLS), scanning tunneling microscopy (STM) and density-functional theory (DFT) calculations we reanalyze the Pd (100)o -O surface oxide phase. We find that the prevalent structural model, a rumpled PdO(001) film suggested by previous low energy electron diffraction (LEED) work (M. Saidy et al., Surf. Sci. 494, L799 (2001)), is incompatible with all three employed methods. Instead, we suggest the two-dimensional film to consist of a strained PdO(101) layer on top of Pd(100). LEED intensity calculations show that this model is compatible with the experimental data of Saidy et al.