No Arabic abstract
The interaction of CO with the Fe3O4(001)-(rt2xrt2)R45{deg} surface was studied using temperature programmed desorption (TPD), scanning tunneling microscopy (STM) and x-ray photoelectron spectroscopy (XPS), the latter both under ultrahigh vacuum (UHV) conditions and in CO pressures up to 1 mbar. In general, the CO-Fe3O4 interaction is found to be weak. The strongest adsorption occurs at surface defects, leading to small TPD peaks at 115 K, 130 K and 190 K. Desorption from the regular surface occurs in two distinct regimes. For coverages up to 2 CO molecules per (rt2xrt2)R45{deg} unit cell, the desorption maximum shows a large shift with increasing coverage, from initially 105 K to 70 K. For coverages between 2 and 4 molecules per (rt2xrt2)R45{deg} unit cell, a much sharper desorption feature emerges at 50 K. Thermodynamic analysis of the TPD data suggests a phase transition from a dilute 2D gas into an ordered overlayer with CO molecules bound to surface Fe3+ sites. XPS data acquired at 45 K in UHV are consistent with physisorption. Some carbon-containing species are observed in the near-ambient-pressure XPS experiments at room temperature, but are attributed to contamination and/or reaction with CO with water from the residual gas. No evidence was found for surface reduction or carburization by CO molecules.
High-quality and impurity-free magnetite surfaces with (sqrt2xsqrt2)R45o reconstruction have been obtained for the Fe3O4(001) epitaxial films deposited on Fe(001). Based on atomically resolved STM images for both negative and positive sample polarity and Density Functional Theory calculations, a model of the magnetite (001) surface terminated with Fe ions forming dimers on the reconstructed (sqrt2xsqrt2)R45o octahedral iron layer is proposed.
Chemisorption of CO on the stepped Cu(211) surface is studied within ab-initio density functional theory (DFT) and scanning tunneling microscopy (STM) imaging as well as manipulation experiments. Theoretically we focus on the experimentally observed ordered (2x1) and (3x1) CO-phases at coverages 1/3, 1/2 and 2/3 monolayer (ML). To obtain also information for isolated CO molecules found randomly distributed at low coverages, we also performed calculations for a hypothetical (3x1) phase with 1/3 ML. The adsorption geometry, the stretching frequencies, the work functions and adsorption energies of the CO molecules in the different phases are presented and compared to experimental data. Initially and up to a coverage of 1/2 ML CO adsorbs upright on the on-top sites at step edge atoms. Determining the most favorable adsorption geometry for the 2/3 ML ordered phase turned out to be nontrivial both from the experimental and the theoretical point of view. Experimentally, both top-bridge and top-top configurations were reported, whereby only the top-top arrangement was firmly established. The calculated adsorption energies and the stretching frequencies favor the top-bridge configuration. The possible existence of both configurations at 2/3 ML is critically discussed on the basis of the presently accessible experimental and theoretical data. In addition, we present observations of STM manipulation experiments and corresponding theoretical results, which show that CO adsorbed on-top of a single Cu-adatom, which is manipulated to a location close to the lower step edge, is stronger bound than CO on-top of a step edge atom.
We present a detailed ab initio study of the electronic structure and magnetic order of an Fe monolayer on the Ir(001) surface covered by adsorbed oxygen and hydrogen. The results are compared to the clean Fe/Ir(001) system, where recent intensive studies indicated a strong tendency towards an antiferromagnetic order and complex magnetic structures. The adsorption of an oxygen overlayer significantly increases interlayer distance between the Fe layer and the Ir substrate, while the effect of hydrogen is much weaker. We show that the adsorption of oxygen (and also of hydrogen) leads to a p(2$times $1) antiferromagnetic order of the Fe moments, which is also supported by an investigation based on a disordered local moment state. Simulated scanning tunneling images using the simple Tersoff-Hamann model hint that the proposed p(2$times $1) antiferromagnetic order could be detected even by non-magnetic tips.
Crystalline Fe3O4/NiO bilayers were grown on MgO(001) substrates using reactive molecular beam epitaxy to investigate their structural properties and their morphology. The film thickness either of the Fe3O4 film or of the NiO film has been varied to shed light on the relaxation of the bilayer system. The surface properties as studied by x-ray photo electron spectroscopy and low energy electron diffraction show clear evidence of stoichiometric well-ordered film surfaces. Based on the kinematic approach x-ray diffraction experiments were completely analyzed. As a result the NiO films grow pseudomorphic in the investigated thickness range (up to 34nm) while the Fe3O4 films relax continuously up to the thickness of 50nm. Although all diffraction data show well developed Laue fringes pointing to oxide films of very homogeneous thickness, the Fe3O4-NiO interface roughens continuously up to 1nm root-mean-square roughness with increasing NiO film thickness while the Fe3O4 surface is very smooth independent on the Fe3O4 film thickness. Finally, the Fe3O4-NiO interface spacing is similar to the interlayer spacing of the oxide films while the NiO-MgO interface is expanded.
First-principles calculations using density functional theory based on norm-conserving pseudopotentials have been performed to investigate the Cs adsorption on the Si(001) surface for 0.5 and 1 ML coverages. We found that the saturation coverage corresponds to 1 ML adsorption with two Cs atoms occupying the double layer model sites. While the 0.5 ML covered surface is of metallic nature, we found that 1 ML of Cs adsorption corresponds to saturation coverage and leads to a semiconducting surface. The results for the electronic behavior and surface work function suggest that adsorption of Cs takes place via polarized covalent bonding.