Surface crystallography of W(110) and W(110) p (2×1)-O: The position of W atoms

The position of W atoms in the surface layers of clean W (110) and W (110) p (2 × 1)-O is studied using Constant-Momentum-Transfer Averaging of LEED intensities. It is shown that the clean surface is not relaxed to an uncertainty of < 0.06 Å. Analysis of superstructure beam intensity averages from W(110) p (2 × 1)-O indicates that oxygen does not reconstruct W(110) at these coverages and below 1000 ° K. An upper limit of 0.05 Å can be put on the out-of-plane displacement of W atoms by the oxygen. Substrate beam averages from W (110) p (2 × 1)-O verify the non-reconstruction. The use of CMTA for adsorbed-layer crystallography in general is briefly discussed.     

Diffusion of W on A W(211) plane

The diffusion of W on a (211) plane of a W field emitter has been re-examined by means of the fluctuation autocorrelation method. Diffusion along channels yielded E = 16.8 ± 0.5 kcal, D₀ = (3 ± 1) × 10⁻⁵ cm² s⁻¹. For diffusion across channels E =6.6 kcal, D₀ = 4 × 10⁻⁹cm² s⁻¹ at T < 752 K, and E = 24 kcal, D₀ = 5 × 10⁻⁴ cm² s⁻¹ at T > 752 K. The results for diffusion along channels yield E and D₀ values intermediate between recent results by Wang and Ehrlich [Surf. Sci. 206 (1988) 451] using field ion microscopy (E = 19 kcal, D₀ = 7.7 × 10⁻³ cm² s⁻¹) and Tringides and Gomer [J. Chem. Phys. 84 (1986) 4049], using the same method as the present work but a larger slit (E = 13.3 kcal, D₀ = 7 × 10⁻⁷ cm² s⁻¹). The results for cross channel diffus good agreement with those of Tringides and Gomer below 752 K, where these authors stopped. The new high temperature results suggest that the channel wall exchange mechanism postulated by Tringides and Gomer for cross channel diffusion at low T gives way to diffusion by climbing over the channel walls with higher E but also higher D₀ above 752 K. Possible reasons for the discrepancies between these three sets of results and the absence of cross channel diffusion in the work of Wang and Ehrlich are briefly discussed.     

Comparison between Rydberg surface resonances on W(001) and W(110)

Comparison is made between high-resolution LEED measurements of electronic surface resonances in the (00) beam from W(001) and W(110). The study of W(110) extended to several polar angles of incidence provides new information that helps to explain the data obtained on W(001) at a fixed angle.     

Surface structures of W(110) and W(100) faces by the dynamical LEED approach

By means of an efficient dynamical LEED method, the surface structures of W(110) and W(100) faces are examined. It is found that the W(110) surface maintains the bulk structure, despite the possibility for the top tungsten atoms to settle into sites of higher coordination number. The W(100) face exhibits a possible contraction of the top atomic layer by 0.10 ± 0.10 Å. These results fit into a trend correlating a top-layer contraction with the absence of close-packing in the top layer, i.e. with the roughness of the surface.     

Surface lattice-dynamical approach to the reconstruction of W(100) and W(100): H

We review the surface reconstruction of the bcc (100) transition metal surface, particularly W(100) and some of the results obtained, with the method of the effective surface lattice dynamics.     

The methanation activity of W(110)

The methanation activity of W(110) was measured over a range of reactant partial pressures and temperatures (P_H2 = 1–1000 Torr, P_CO = 0.1–10 Torr, T = 475–820 K). Plotting the results in an Arrhenius fashion yielded a lower apparent activation energy (E_a = 56 kJmol^?1) than previously determined for Ni(100) (E_a = 103 kJmol^?1) with an activity surpassing that of Ni at lower temperatures. The H₂ pressure dependence of the methanation activity was found to be much stronger for W(110) than for Ni(100), the surface becoming increasingly inactive at the lowest H₂ pressures investigated. Auger electron spectroscopy revealed the active catalytic surface to be carbidic in nature.     

Desorption of H2 from W(110) and Fe covered W(110) surfaces

The kinetics of H₂ desorption from H/W(110) and H/Fe₁/W(110) were studied by measuring work function changes Δø vs time at a number of temperatures. Combination with previously determined Δø vs coverage data and differentiation at various fixed coverages gave rate vs T data from which activation energies of desorption could be obtained. E vs coverage results agree well with previously determine ΔH_des results. In the case of H/Fe₁/W(110) this includes a rise from 20 to 30 kcal mol⁻¹ of H₂ at H/Fe = H/W > 0.3. Plots of rate −dθ/dt vs θ (θ being coverage in units of H/W) vary much more steeply than θ² at most coverages for both systems. The θ dependence can be explained almost quantitatively in terms of the variations of ΔH_des and surface entropy S_s with coverage, by assuming that rates of desorption are equal to the equilibrium rates of adsorption. The latter can be formulated thermodynamically, except for a sticking coefficient, s. Values for s(θ, T) can also be obtained and show relatively little temperature dependence.     

Relaxation of the Mo(112) and W(112) surfaces

Relaxation of the Mo(112) and W(112) surfaces has been simulated within DFT in local density approximation. It has been found that the surface relaxation, which can be described as a 14% contraction of the topmost surface layer with a small (0.1%) shift of surface atomic rows, results in a strong decrease of the surface energy with respect to the bulk truncated crystal surfaces (from 0.2 to 0.17 eV/Å2 for the Mo(112) and from 0.36 to 0.33 eV/Å2 for the W(112)). The surface relaxation is accompanied by the redistribution of the surface density of states, associated with the transformations of surface states.     

The ferromagnetic monolayer Fe(110) on W(110)

Ferromagnetic order in the pseudomorphic monolayer Fe(110) on W(110) was analyzed experimentally using Conversion Electron Mössbauer Spectroscopy (CEMS) and Torsion Oscillation Magnetometry (TOM). The monolayer is thermodynamically stable, crystallizes to large monolayer patches at elevated temperatures and therefore forms an excellent approximation to the ideal monolayer structure. It is ferromagnetic below a Curie-temperatureT _c,mono, which is given by (282±3) K for the Ag-coated layer, (290±10) K for coating by Cu, Ag or Au and ≈210 K for the free monolayer. For the Ag-coated monolayer, ground state hyperfine fieldB _hf (0)=(11.9±0.3) T and magnetic moment per atom μ=2.53 μ_B could be determined, in fair agreement with theoretical predictions. Unusual properties of the phase transition are detected by the combination of both experimental techniques. Strong magnetic anisotropies, which are essential for ferromagnetic order, are determined by CEMS.     

Molecular binding states on clean and oxidized surfaces: SiO(g) + W(clean) and SiO(g) + W(oxidized)

The interactions between a molecular beam of SiO(g) and a clean and an oxidized tungsten surface were examined in the surface temperature range 600 to 1700 K by mass spectrometrically determined sticking probabilities, by flash desorption mass spectrometry (FDMS) and by Auger electron spectroscopy (AES). The sticking probability, S, of SiO has been determined as a function of coverage and of surface temperature for the clean and the oxidized tungsten surface. Over the temperature range studied and at zero coverage S = 1.0 and 0.88 for the clean and oxidized tungsten surfaces respectively. The results are consistent with both FDMS and AES. For coverage up to one monolayer there is one major adsorption state of SiO on the clean tungsten surface. FDMS shows that T_m = constant (T_m is the surface temperature at which the desorption rate is maximum) and that desorption from this state is described by a simple first order desorption process with activation energy, E_d = 85.3 kcal mole^?1 and pre-exponential factor, ν = 2.1 × 10¹⁴ sec^?1. AES shows that the 92 eV peak characteristic of silicon dominates. In contrast on the oxidized tungsten surface, T_m shifts to higher temperatures with increasing coverage. The data indicate a first order desorption process with a coverage dependent activation energy. At low coverage (θ ? 0.14) there is an adsorption state with E_d = 120 kcal mole^?1 and ν = 7.6 × 10¹⁹, while at θ = 1.0, E_d = 141 kcal mole^?1. This variation is interpreted as due to complex formation on the surface. AES shows that on oxidized tungsten, in contrast to clean tungsten, the dominant peaks occur at 64 and 78 eV, and these peaks are characteristic of higher oxidation states of silicon. Thus, it is concluded that SiO exists in different binding states on clean and oxidized tungsten surfaces.     

Study of the W(t, p)W reaction at 20 MeV

The ¹⁸²W(t, p)¹⁸⁴W reaction has been studied at 20 MeV. The outgoing protons were detected in an Elbek broad range magnetic spectrograph. Absolute cross sections and angular distributions were measured. Evidence for inelastic effects in the reaction mechanism was observed for the first excited 2⁺ state, the 3⁻ level at 1221 keV and the 5⁺ state at 1295 keV. The 0⁺ level at 1002 keV was populated with ≈ 2% the ground state cross section. A 4⁺ level at 1536 keV was observed with ≈ 50% the ground state cross section. Calculations of the absolute (t, p) cross sections to this and other states with known structure resulted in excellent agreement with the measured values.     

Long jump rates in surface diffusion: W on W(110)

We have reexamined the diffusion of W adatoms on W(110) using the field ion microscope. The diffusivity is in good agreement with previous results and reveals no unusual features, but from observations of the distribution of displacements, corrected for diffusion during temperature transients, we have for the first time been able to measure the temperature dependence of rates for nearest-neighbor and double jumps, as well as for vertical and horizontal transitions. Activation energies and frequency factors for all the long jumps are significantly larger than for single jumps, and a simple model is proposed to account for our observations.     

Noble-metal nanostructures on carburized W(110)

Noble metal nanostructures of Au, Ag and Cu were prepared on two types of carbon-modified W(110) surfaces—R(15 × 12) and R(15 × 3)—and investigated by means of scanning tunneling microscopy. For all deposited metals qualitatively the same behaviour is observed: On the R(15 × 12)-template always isotropic clusters are formed. In contrast, on the R(15 × 3)-substrate the anisotropy of the nanostructures can be tuned from clusters at low temperatures via thin nanowires to thicker nanobars at high deposition temperatures. At intermediate temperatures on the R(15 × 3) the anisotropic Au nanowires arrange themselves into straight lines along domain boundaries induced by deposition of the Au metal. Similarities and differences to Au nanostructures as recently reported by Varykhalov et al. [A. Varykhalov, O. Rader, W. Gudat. Physical Review B 77, 035412 (2008).] are discussed.     

Surface magnetism of iron on W(110)

Ultrathin bcc iron films grown epitaxially on W(110) have been investigated by means of angle and spin resolving photoelectron spectroscopy. The electron spin polarization, spin resolved intensities and corresponding band structure have been experimentally investigated in dependence of the film thickness, exciting photon energy and variation of the photoelectron detection angle. Additionally, photoemission calculations for bulk iron have been performed in the framework of a relativistic one-step formalism. The comparison between measured and calculated spectra turns out to be in very good agreement for different excitation energies as well as for different angles of emission.     

Vibrational modes of oxygen on W(110)

Electron energy loss measurements of the vibrational modes of oxygen on W(110) as a function of coverage up to 0.5 monolayer are presented and analyzed. A single loss at 67 meV is observed initially; with increasing exposure this loss shifts to 72 meV and another loss appears at 47 meV. These data indicate coexistence of two species on the surface with a coverage-dependent conversion. Angular profiles of the specular elastic beam show a dramatic increase in width with initial oxygen coverage; this is possibly due to an oxygen-induced static disordering of the W surface layer.     

Surface vibrations of CO on W(100)

High resolution electron-energy-loss spectroscopy has been used to study the surface vibrations of CO on a W(100) surface at 300 K. For small exposures (β-CO) two losses at ～68 meV and ～78 meV are observed. This vibrational spectrum of β-CO is a clear indication of dissociative adsorption with the carbon and oxygen atoms in fourfold coordination sites each. With further exposure to CO two additional losses at 45 meV and 258 meV are observed, which represent the vibration of undissociated α-CO in upright position on top of a W atom. Furtheron results of coadsorption of H₂/CO and O₂/CO on W(100) are reported.     

Long-wavelength surface phonon spectrum of W(001)

We calculate eigenvectors and local densities of long-wavelength phonon states for a (001) semi-infinite crystal of W, by a microscopic force constant approach. The results are discussed in relation to surface structure (relaxation) and to surface-sensitive spectroscopies.