Leed study of Fe epitaxially grown at 190°C on Cu(100)

We report a LEED study of iron deposited on Cu(100) at 190°C. Very sharp LEED patterns were obtained for high iron coverages. We observed evidence of an ill defined Fe/Cu interface for one monolayer of iron deposited at 190°C. The stability of the iron overlayers was tested as a function of time for various coverages. For one monolayer the intensity versus energy curve for the (00) beam shows time dependence at 190°C. For four layers, no significant changes were observed in the LEED spectra over a period of one hour. We measured for five layers of iron a top layer expansion of 2.8% relative to the bulk. The interplanar spacing for bulk fcc Fe at 190°C remains equal to the room value. In these experiments the samples were deposited at 190°C and kept at this temperature during the LEED measurements.     

The structure of Cu{100}-p(2 × 6)-2mg-Sn studied by DFT and LEED

Low Energy Electron Diffraction (LEED) and Density Functional Theory (DFT) have been used to analyse the structure of Cu{100}-p(2 × 6)-2mg-Sn at room temperature. In this work we found that the favoured geometry for this 0.33 ML Cu{100}-Sn phase is a combination of an overlayer structure and a surface alloy; two Sn atoms are alloyed in to the first copper layer and the other two Sn atoms adsorb at off symmetry hollow sites. In order to relieve the stress in the alloyed layer, the alloyed Sn atoms are buckled 0.59/0.45 ± 0.2 Å (DFT/LEED) above the centre of mass of the first layer copper atoms.     

Deposition of iron on MgO(100): Reaction of CO and H2

Ultraviolet photoelectron spectroscopy (UPS), electron energy loss spectroscopy (EELS) and surface extended energy loss fine structure (SEELFS) were used to study the deposition of Fe on MgO(100) and to identify the surface compounds formed after reaction of CO/H₂ (1:1). The clean MgO(100) surface was characterized using the above techniques and the effect of argon ion bombardment damage to the surface was investigated. With the deposition of iron, metallic characteristics appear in the photoemission spectrum; the electron energy loss peaks of the MgO(100) substrate diminish in intensity with no significant shifts in loss energies. Fine structure analysis of the oxygen K-edge of the MgO(100) surface with less than 2 monolayers (ML) of iron suggests that the iron atoms bond with the oxygen at the surface of the MgO(100) lattice. For less than 4 ML of iron, the EEL spectra show that the deposited iron is oxidized after reaction of CO/H₂. Higher iron coverages result in carburization of the surface. Carbon deposition was observed with CO for all Fe coverages. Measurement of the fine structure above the carbon K-edge suggests that the types of carbide formed depend on the iron coverage; one carbide has a short CFe distance of 1.78 Å and the other a distance of 2.06 Å (high metal coverage).     

Structure of the Ag on Si(111) 7 × 7 interface by means of surface exafs

Polarization dependent surface extended X-ray absorption fine structure (SEXAFS) measurements are used to determine the structure of the Ag on Si(111)7 × 7 system at the early stages (< 3 monolayers (ML)) of interface formation. At room temperature (RT) Ag is found to initially (< 0.5 ML) chemisorb in the threefold hollow site, approximately 0.7 Å above the outermost Si layer with an average Ag-Si distance of 2.48±0.05 Å. Above monolayer coverage the SEXAFS spectrum is dominated by the Ag-Ag distance indicating Ag island formation on the surface. Upon heating (200 ?T? 600°C) a (√3 ×√3)R30° LEED pattern is observed. At the lowest coverage ( < 0.7 ML) this pattern is determined to arise from Ag atoms which are embedded in the threefold hollows, ~ 0.7 Å below the first and above the second Si layer, with a Ag-Si distance of 2.48 ± 0.04 Å. At higher coverage ($\text{\$}\text{?}$1 ML) Ag clusters are found to grow on this interface with the same Ag-Ag distance as in Ag metal. Our results are discussed in the context of previous experimental and theoretical results.     

Search for ferromagnetism in ultrathin epitaxial films: Cr/Au(100), Cr/Cu(100), and Fe/Cu(100)

The surface magneto-optic Kerr effect (SMOKE) technique was used to search for ferromagnetism in monolayer-range films of Cr and Fe grown on Au(100) and Cu(100). The growth modes were characterized using low energy electron diffraction (LEED) and Auger electron spectroscopy. The fcc structure of Cr could not be stabilized on Cu(100). Ferromagnetism was not observed for the Cr/Au(100) films at temperature above 100 K. Ferromagnetism also was not observed for fcc Fe/Cu(100) grown at room temperature; but for growth at >150°C, a ferromagnetic, metastable state was observed for the top layer of the Fe film, in the absence of bulk ferromagnetism. The ferromagnetic Fe/Au(100) system was used to establish the sensitivity of the approach.     

Structure of CO adsorbed on Ni (100)

The c(2 × 2) configuration of CO chemisorbed on Ni(100) has been examined by the dynamical LEED method of surface structure analysis. Experimental LEED intensity spectra of the (00), ($\text{1}\text{2}\text{1}\text{2}$) (10) and (11) LEED beams measured at 175 K are compared with the corresponding calculated spectra for two different CO potential constructions and a number of trial structures. The best agreement was found for a structure where the CO molecules sit directly above the Ni atoms with vertical spacings between the Ni and C and the C and O layers of 1.80 ± 0.10 A and 0.95 ± 0.10 Å respectively. It is proposed that the CO molecule is tipped over at an angle of 34° ± 10° with respect to the surface normal so that the actual carbon-oxygen bond length is close to the figure 1.15 Å found in Ni(CO)₄.     

Oxygen induced surface segregation of Cu on the Au0.7Cu0.3(100) surface

The oxygen induced surface segregation of Cu on the Au_0.7Cu_0.3(100) surface was investigated by means of LEED and AES techniques. The dissociative adsorption of O₂ did not take place on this clean surface for a long time exposure at least up to 10⁴ L, and so the oxygen was forcibly introduced onto the surface through a pre-deposition of few a layers of Cu and its successive oxidation. The oxygen coverage was controlled by a heat treatment, which leads the system to a thermal equilibrium state. For the clean surface, the segregation of Au was clearly observed and the surface concentration of Au was estimated to be about 86%, greater than the bulk concentration of 70%. At low coverages below 0.16 ML, no remarkable oxygen induced segregation of Cu was observed. But, above 0.2 ML, the surface concentration of Cu was proportional to the oxygen coverage. The (2 × 4) LEED pattern was observed in a wide range of oxygen coverage. The maximum intensity of the (2 × 4) was observed at about 0.45 ML.     

Structure of the laser annealed Si(1 1 1)-(1 × 1) surface

LEED analysis of the laser annealed Si(1 1 1)-(1 × 1) surface shows that a model with a graphite-like top double layer of atoms with a spacing of 2.95±0.02 Å from the second double layer describes the LEED data as well as the Zehner model, but involves large displacements of the atoms normal to the surface as required by ion scattering results. It is suggested that this model provides a natural interpretation of the low energy He atom scattering data for the Si(1 1 1)-(7 × 7) surface.     

Atomic structure of Fe {111}

A clean Fe {111} surface was prepared and studied with LEED (low-energy electron diffraction) and AES (Auger electron spectroscopy). A LEED intensity analysis was carried out with a new computational scheme (THIN) specially designed for short interlayer spacings. The results are, for the fust interlayer spacing, d₁₂ = 0.70 ± 0.03 Å and for the inner potential V₀ = 11.1 ± 1.1 eV, the confidence intervals referring to 95% confidence level. Thus, the Fe {111} surface is contracted 15.4% with respect to the bulk (0.827 Å).     

Fast leed intensity measurements from Ni(100)c(2 × 2)CO

The Ni(100)c(2 × 2)CO surface structure has been investigated by very fast LEED intensity measurements using a computer controlled television method. It turns out that the intensity spectra are strongly influenced by intolerably long measuring times during which the primary electron beam impinges onto the surface. The spectra have been taken within 16 sec at 100 K immediately after termination of the adsorption process for all beams simultaneously. They are compared with other measurements and with Pendrys model calculations for a CO molecule bonded linearly on top of a Ni atom with straight molecular axis normal to the surface. Using the r-factor formalism for theory-experiment comparison the bond length results to be 1.15 ± 0.1 Å for CO and 1.80 ± 0.1 Å for NiC. This is in agreement with the results of other methods and removes some discrepancies with those of earlier LEED experiments.     

Iron oxide thin film growth on Al2O3/NiAl(1 1 0)

The electronic structure and the growth morphology of iron oxide thin films were studied by means of Synchrotron Radiation Photoelectron Spectroscopy (SRPES) and Low Energy Electron Diffraction (LEED). A thin well-ordered alumina film on a NiAl(1 1 0) single crystal surface as a template for iron oxide growth was employed. Two different methods of iron oxide film preparation were applied. In the first attempt, iron deposited at room temperature was subsequently annealed in oxygen. Even though a whole layer of iron was oxidized, an expected long-range order was not achieved. The second attempt was to perform reactive deposition. For this reason iron was evaporated in oxygen ambient at elevated substrate temperature. This method turned out to be more efficient. Diffused but clear LEED patterns of six-fold symmetry indicating hexagonal surface atoms arrangement were observed. From the PES measurements, binding energies for Fe2p for grown iron oxide film were established as well as energy distribution curves for the valence band. Growth curves based on Fe3p core-level peak intensities for iron and iron oxide were plotted identifying type of film growth for both deposition methods. Based upon these results we have found evidence for interdiffusion in the interface between alumina and iron oxide at the early stages of growth. Further deposition led to formation of Fe₃O₄(1 1 1) (magnetite) overlayer. Moreover, the quality of the film could also be improved by long-time annealing at temperatures not exceeding 575 K. Higher annealing temperature caused disappearance of LEED pattern indicating loss of long-range ordering.     

A tilted precursor for CO dissociation on the Fe(100) surface

Near edge X-ray absorption fine structure (NEXAFS) has been used to study the molecular orientation of the α₃ state of CO on the Fe(100) surface. It is found that the molecule is tilted by 45° ± 10° with respect to the surface normal, allowing direct interaction of the oxygen end of the molecule with the iron surface. The C-O bond is found to be elongated by 0.07 ± 0.02 Å in the α₃ state, relative to the other molecularly adsorbed CO states on this surface.     

Radiation damage in single-crystal ice studied by 100 keV proton channeling

A detailed study has been undertaken of the Ni{100} (2 × 2)C structure formed by cracking ethylene on a clean Ni{100} surface. The LEED pattern shows characteristic missing spots which can be attributed to the presence of glide lines and indicate a space group symmetry of p4g. We show that this can be readily interpreted in terms of a distortion of the top nickel layer both parallel and perpendicular to the surface, which accompanies the carbon adsorption. Detailed comparisons of LEED intensity data with dynamical calculations indicate that the top layer nickel atoms are displaced 0.35 ± 0.05 Å parallel to the surface, 0.20 ± 0.05 Å outwards from the surface, and that the carbon atoms are in 4-fold hollows (now distorted) at a spacing of 0.1 ± 0.1 Å from the surface. These conclusions lead to a nickel-carbon nearest neighbour spacing of 1.803 ± 0.015 Å.     

Structural transformations on Cu(110) under molecular iodine action

Interaction of molecular iodine with the Cu(110) surface is investigated by the methods of ultrahigh vacuum scanning tunneling microscopy (STM) and low energy electron diffraction (LEED). It is found that at the coverage θ =0.5 monolayer (ML) iodine forms a simple commensurate c(2×2) lattice. Further exposure of iodine leads to uniaxial compression of the c(2×2) lattice along the 〈110〉 direction of the substrate. The STM data indicate that compression of the iodine layer proceeds through formation of striped domain walls. As the coverage is saturated at θ = 0.63 ML, iodine forms a uniformly compressed quasi-hexagonal structure. Further exposure of iodine on the Cu(110) surface results in growth of a copper iodide film. STM images of thin (7 to 20 Å) CuI films reveal, in addition to atomic modulation, a superstructure with a period of 90 to 100 Å consisting of double stripes. A structure model of the copper iodide surface allowing for CuI lattice contraction and formation of double stripe domain walls is proposed.     

A LEED/AES study of the oxidation of Fe0.84Cr0.16 (100) and (110)

The interaction between single crystalline Fe_0.84Cr_0.16 (100) or (110) and oxygen gas in the pressure range 10^?9 to 10^?7 torr was studied at room temperature and at 800 K, using LEED and AES. The interaction starts with a chromium-oxygen reaction next to the alloy surface, followed by an iron—oxygen reaction outside the chromium-oxygen layer. At 800 K these reactions are connected with redistribution of cations between the interior of the alloy and the surface region, whereas at room temperature only a redistribution of cations within the surface region is observed. Different symmetries and lattice parameters of oxides which grow epitaxially on Fe_0.84Cr_0.16 (100) are compared with the corresponding surface compositions. It is found that the formation of spinel-like oxide layers is favoured by lower values of the Cr/Fe surface ratio.     

Structural models of the reconstructed W{001} surface

A LEED intensity analysis of 5 beams from the low-temperature W{001}c(2×2) structure indicates that the surface reconstruction involves shifts of the surface atoms along 〈110〉 directions within the plane of the surface, as suggested by Debe and King. At temperatures 100–140K the shifts are in the range 0.15–0.3 Å, with the first interlayer spacing 1.48–1.58 Å (bulk value 1.58 Å). Similar analysis of the room-temperature W{001}c(2×2)-H phase indicates: (i) none of the models proposed, which ascribe the c(2×2) structure directly to ordered hydrogen adsorption, can explain the experimental data; (ii) the W{001}c(2×2)-H structure is probably impurity stabilized by H at room temperature in the same W lattice as the low-temperature reconstructed phase.     

The structure of atomic nitrogen adsorbed on Fe(100)

Nitrogen atoms adsorbed on a Fe(100) surface cause the formation of an ordered c(2 × 2) overlayer with coverage 0.5. A structure analysis was performed by comparing experimental LEED I–V spectra with the results of multiple scattering model calculations. The N atoms were found to occupy fourfold hollow sites, with their plane 0.27 Å above the plane of the surface Fe atoms. In addition, nitrogen adsorption causes an expansion of the two topmost Fe layers by 10% (= 0.14 Å). The minimum r-factor for this structure analysis is about 0.2 for a total of 16 beams. The resulting atomic arrangement is similar to that in the (002) plane of bulk Fe₄N, thus supporting the view of a “surface nitride” and providing a consistent picture of the structural and bonding properties of this surface phase.     

Coverage-dependent molecular tilt of carbon monoxide chemisorbed on Pt{110}: A combined LEED and DFT structural analysis

The adsorption of carbon monoxide on the Pt{110} surface at coverages of 0.5 ML and 1.0 ML was investigated using quantitative low-energy electron diffraction (LEED IV) and density-functional theory (DFT). At 0.5 ML CO lifts the reconstruction of the clean surface but does not form an ordered overlayer. At the saturation coverage, 1.0 ML, a well-ordered p(2 × 1) superstructure with glide line symmetry is formed. It was confirmed that the CO molecules adsorb on top of the Pt atoms in the top-most substrate layer with the molecular axes tilted by ± 22° with respect to the surface normal in alternating directions away from the close packed rows of Pt atoms. This is accompanied by significant lateral shifts of 0.55 Å away from the atop sites in the same direction as the tilt. The top-most substrate layer relaxes inwards by ? 4% with respect to the bulk-terminated atom positions, while the consecutive layers only show minor relaxations. Despite the lack of long-range order in the 0.5 ML CO layer it was possible to determine key structural parameters by LEED IV using only the intensities of the integer-order spots. At this coverage CO also adsorbs on atop sites with the molecular axis closer to the surface normal (< 10°). The average substrate relaxations in each layer are similar for both coverages and consistent with DFT calculations performed for a variety of ordered structures with coverages of 1.0 ML and 0.5 ML.     

Surface X-ray diffraction analysis of Fe nanostructured films grown on c(2 × 2)-N/Cu(100)

We report the results of a Surface X-Ray Diffraction (SXRD) study of Fe nanostructured films deposited on c(2 × 2)-N/Cu(100) at room temperature (RT), with Fe coverage Θ_Fe = 0.5 ML and Θ_Fe = 1 ML. The c(2 × 2)-N/Cu(100) surface is an example of self-organised system, that can be used for growth of arrays of metal nano-islands and organic molecules assemblies. We chose two different values of N coverage, Θ_N = 0.3 ML and Θ_N = 0.5 ML, the second value corresponding to N saturation. We monitored the presence of surface diffraction peaks in hk scans and we performed Crystal Truncation Rods (CTR) analysis with ROD fitting programme. In the case of Θ_N = 0.5 ML, i.e. at saturation coverage, the CTR could be fitted with one surface domain with p4gm(2 × 2) symmetry. In the surface cell adopted, N atoms occupy four-fold hollow sites, with Fe (intermixed with Cu) giving rise to a “clock” reconstruction previously observed on iron nitride films obtained by co-deposition and annealing. This result is an indirect confirmation of N surface segregation on top of the Fe films, occurring during the growth at RT. When subsaturation N coverage (Θ_N = 0.3 ML) is used as a substrate for Fe deposition, the best results could be obtained with a model where two surface domains are present: the first one corresponds to a surface cell with Fe sitting in four-fold hollow sites on bare Cu areas, with possible interdiffusion in the second lattice. The second domain is assigned to growth of Fe on the N-covered square islands occurring once the bare Cu areas are fully covered. The SXRD analysis on N-covered surface domains shows that the mechanism of reconstruction and of N segregation on top layer is already active at RT for all N-coverage values.     

Potassium adsorption on Fe(110)

LEED, AES, UPS and XPS were used to study submonolayer coverages of potassium on Fe(110). At room temperature the maximum potassium coverage is characterized by a LEED superstructure. This LEED pattern is interpreted as being due to a hexagonal close-packed K layer on Fe(110), resulting in a maximum atom density of 5.3 × 10¹⁴ cm^?2, i.e.θ k = 0.31. The work function change and the shift of the K(2p) and K(3p) core levels with potassium coverage indicate a charge transfer from potassium to iron at low potassium coverages.