The structure of oxygen-induced reconstruction on Cu{100}-c(2×2)–Pt surface alloy: the Pt/Cu{100}-(2×2)–O (original) (raw)
Related papers
Oxygen-induced missing-row reconstruction of Cu(001) and Cu(001)-vicinal surfaces
Physical Review B, 1990
We have used x-ray-diffraction analysis to examine the structure of flat and vicinal Cu(001) surfaces under the influence of oxygen chemisorption. The initial electropolished vicinal surface consists of a fairly regular array of steps and terraces that preferentially orient the oxygen-induced reconstruction. Prolonged annealing leads to a hill-and-valley morphology with large (001) facets, upon which the preferential orientation is lost, just as for the flat surface. We find evidence for only one ordered phase as a function of oxygen coverage, which has 2&2X&2 symmetry. Crystallographic analysis of the diffraction data shows this to be a "missing-row" structure with 25% of the Cu sites vacant and large relaxations in the top layer. The oxygen site is not uniquely determined, however, with two distinct possibilities. This Cu(001)/0 structure has a surprising similarity to that proposed for Cu(110)/O. In fact both surfaces can be decomposed into the same basic structural element, which is a Cu-0-Cu chain oriented along bulk [100]directions.
Surface Science, 2002
The geometric structure and compositional profile of a Cu{1 0 0}-cð2 Â 2Þ-Pt surface alloy formed by thermal activation of a monolayer Pt film has been determined by tensor low energy electron diffraction (LEED). A wide range of models have been tested. The favoured model consists of an ordered cð2 Â 2Þ CuPt underlayer below a Cu terminated surface. Models involving a mixed ordered CuPt layer outermost may be definitively ruled out. The average T-matrix approximation (ATA) has been applied allowing variable Pt concentrations to be introduced into both the outermost layer and deeper into the selvedge (layers 3 and 4) in the form of a random substitutionally disordered Cu x Pt 1Àx alloy. The favoured concentration profile corresponds to an almost pure outermost Cu monolayer (H Pt ¼ 10 AE 10 at.%) with Pt concentrations of 20 AE 20 and 30 AE 30 at.% in layers 3 and 4 respectively. Introduction of Pt into the surface layers induces a significant expansion of the selvedge yielding modification of the outermost three interlayer spacings to
On-surface and sub-surface oxygen on ideal and reconstructed Cu(100)
Surface Science, 2005
In order to understand the first steps of the Cu(1 0 0) oxidation we performed first principles calculations for on-surface and sub-surface oxygen on this surface. According to our calculations, the adsorption energies for all on-surface site oxygen atoms increase, whereas the energies of the sub-surface atoms decrease with the increasing oxygen coverage. At coverage 1 ML and higher on the reconstructed surface, structures including both on-and sub-surface atoms are energetically more favourable than structures consisting only of on-surface adsorbates. On the ideal (1 0 0) surface this change can be perceived at coverage 0.75 ML.
DFT study of reconstructed Cu(100) surface with high oxygen coverages
Surface Science, 2008
Adsorption energies and structures of a Cu(1 0 0) surface with an oxygen coverage from 0.5 ML to 1.5 ML have been studied using DFT methods. The calculations include both on-surface and mixed on-and subsurface sites for oxygen. Also several energy barriers for oxygen transfer below the surface have been calculated. Calculations show that mixed structures are energetically more favourable than pure on-surface adsorption at coverages higher than 0.75 ML and barriers for oxygen atoms are low, around 0.5 eV. Experimental results show that during the oxidation process, disordered metal penetrating islands were not seen until after several structural changes, where the oxygen is expected to be on the surface. This is an opposite trend to what we observed in our calculations. An explanation for this discrepancy is that the dissociation of oxygen molecules is very slow on a reconstructed Cu(1 0 0) surface. However, after dissociation, oxygen atoms will diffuse easily below the reconstructed surface and our calculations are in agreement with the experimentally observed disordered islands.
The Journal of Physical Chemistry C, 2013
The electronic structure and stability of Cu(111)-hosted Pt overlayers with and without the presence of atomic oxygen have been studied by means of core-level spectroscopy and density functional theory (DFT). Because of lattice mismatch, Pt(111) overlayers grown on Cu(111) are compressively strained, and hard X-ray photoelectron spectroscopy together with Pt L 3 -edge X-ray absorption spectroscopy (XAS) reveals a pronounced downshift of the Pt d-band owing to the increased overlap of the d-orbitals, an effect also reproduced theoretically. Exposure to oxygen severely alters the surface composition; the O−Cu binding energy largely exceeds that of O−Pt, and DFT calculations predict surface segregation of Cu atoms. Comparing the adsorbate electronic structure for O on unstrained Pt(111) with that of O on Ptmodified Cu(111) using O K-edge XAS and X-ray emission spectroscopy salient differences are observed and calculations show that Cu-segregation to the topmost layer is required to reproduce the measured spectra. It is proposed that O is binding in a hollow site constituted by at least two Cu atoms and that up to 75% of the Pt atoms migrate below the surface.
2009
Atomic oxygen embedment into a Cu(1 0 0) surface is studied by density functional theory calculation and the nudged elastic band method. As the oxygen coverage increases on the unreconstructed surface from 0.25 monolayer (ML) to 0.75 ML, the energy barrier for oxygen embedment decreases and an energetically favorable sub-surface site is found at 0.75 ML coverage. At a fixed oxygen coverage of 0.5 ML, the oxygen embedment energetics vary with the surface morphology and the embedment is found to be more probable for reconstructed structures compared to the bare surface. On the missing-row reconstructed surface, we find that the energy barrier for atomic oxygen embedment is smaller through the missing-row compared to other paths, suggesting a mechanism for the formation of sub-surface oxygen structures that are consistent with a recent experiment. The energy barrier for sub-surface oxygen diffusion is predicted to be less than that for on-surface diffusion.
Density functional study of oxygen on Cu(100) and Cu(110) surfaces
Physical Review B, 2010
Using density-functional theory within the generalized gradient approximation, we investigate the interaction between atomic oxygen and Cu͑100͒ and Cu͑110͒ surfaces. We consider the adsorption of oxygen at various on-surface and subsurface sites of Cu͑100͒ for coverages of 1/8 to 1 monolayers ͑ML͒. We find that oxygen at a coverage of 1/2 ML preferably binds to Cu͑100͒ in a missing-row surface reconstruction, while oxygen adsorption on the nonreconstructed surface is preferred at 1/4 ML coverage consistent with experimental results. For Cu͑110͒, we consider oxygen binding to both nonreconstructed and added-row reconstructions at various coverages. For coverages up to 1/2 ML coverage, the most stable configuration is predicted to be a p͑2 ϫ 1͒ missing-row structure. At higher oxygen exposures, a surface transition to a c͑6 ϫ 2͒ added strand configuration with 2/3 ML oxygen coverage occurs. Through surface Gibbs free energies, taking into account temperature and oxygen partial pressure, we construct ͑p , T͒ surface phase diagrams for O/Cu͑100͒ and O/Cu͑110͒. On both crystal faces, oxygenated surface structures are stable prior to bulk oxidation. We combine our results with equivalent ͑p , T͒ surface free energy data for the O/Cu͑111͒ surface to predict the morphology of copper nanoparticles in an oxygen environment.
Beilstein Journal of Organic Chemistry, 2014
We have investigated the adsorption of perylene-3,4,9,10-tetracarboxylic acid dianhydride (PTCDA) on the clean and on the oxygen pre-covered Cu(100) surface [referred to as (√2 × 2√2)R45° – 2O/Cu(100)] by scanning tunneling microscopy (STM) and low energy electron diffraction (LEED). Our results confirm the (4√2 × 5√2)R45° superstructure of PTCDA/Cu(100) reported by A. Schmidt et al. [J. Phys. Chem. 1995, 99,11770–11779]. However, contrary to Schmidt et al., we have no indication for a dissociation of the PTCDA upon adsorption, and we propose a detailed structure model with two intact PTCDA molecules within the unit cell. Domains of high lateral order are obtained, if the deposition is performed at 400 K. For deposition at room temperature, a significant density of nucleation defects is found pointing to a strong interaction of PTCDA with Cu(100). Quite differently, after preadsorption of oxygen and formation of the (√2 × 2√2)R45° – 2O/Cu(100) superstructure on Cu(100), PTCDA forms ...
Surface Science, 2012
The oxygen-induced reconstruction of Cu(100) surfaces has been studied in great detail. Experimentally, missing row reconstruction occurs with an O coverage of 0.5 monolayer (ML). Several DFT-based calculations of vacancysurface Cu pairs were performed at O coverage between 0 and 0.5 ML. No structure was identified in these calculations as being energetically favoured over the ideal Cu(100) structure. When a Cu atom (0.125 ML) was added to the system, a new energetically favourable missing row and added Cu-O-Cu row structure was found. It has been suggested that this structure is an important intermediate step towards the missing row formation. The transition from the missing and added row structure to the (2√2×√2)R45 o -O reconstruction has the energetic requirement of very good binding sites for the Cu atoms, which can be found on the step boundaries. These results are in excellent agreement with the step edge growth and the formation of strongly anisotropic islands observed by Lahtonen et al. (J. Chem. Phys. 129 (2008) 124703).