Search

The geometric configuration and energy of a hydrogen molecule centered between two point-shaped tips of equal charge are calculated with the variational quantum Monte Carlo (QMC) method without the restriction of the Born-Oppenheimer (BO) approximation. The ground-state nuclear distribution, stability, and low-vibrational excitations are found to deviate significantly from the BO treatment based on a potential energy surface obtained with the same QMC accuracy. The quantum-mechanical distribution of the molecular axis direction and the bond length at a subnanometer level is fundamental for understanding nanomechanical dynamics with embedded hydrogen. The cylindrical symmetry of the tip arrangement yields a uniform azimuthal distribution of the molecular axis vector relative to the tip-tip axis. For fixed tip separation, the QMC sampling shows that the polar angle distribution of the molecular axis is centered around the equatorial plane for positive tip charge (transverse alignment) and around the tip-tip direction for negative tip charge (bridge alignment). These deviations from spherical symmetry are magnified as the tip-tip distance decreases. Our results thus show that the molecular orientation in the junction can be controlled by the tip charge and separation, suggesting an application in the field of molecular machines., This work has been supported in part by the Basque Departamento de Educación, Universidades e Investigación, the University of the Basque Country UPV/EHU (Grant No. IT1246-19), and the Spanish Ministerio de Ciencia y Tecnología (Grants No. FIS2016-76471-P and No. FIS2017-83780-P). M.A.V.H. acknowledges financial support from the Collaborative Research Fund of the Research Grants Council of Hong Kong (Grant No. C2014-15G). I.C.T.S. is supported by the HKBU Institute of Creativity, which is sponsored by the Hung Hin Shiu Charitable Foundation.

Green defect emission is commonly observed in ZnO nanostructures. It is frequently attributed to oxygen vacancies and used to evaluate performance and study physical mechanisms in a variety of applications, such as gas sensing and photocatalysis. However, competing hypotheses have been proposed to explain green emission, which raises questions about the role of oxygen vacancies in sensing and photocatalytic processes. The major problem in correct experimental identification of defects in ZnO is the abundance of defects present, while theoretically there are problems with accurate calculation of a defect energy level in the gap. Thus, here we adopted a different approach and studied experimentally and theoretically the interaction of ZnO with different chemical substances (hydrogen and a silane-based molecule). Based on theoretical predictions and experimental results, we can conclude that green emission can likely be assigned to defect complexes, which may contain zinc vacancies.

The adsorption of methylphosphonic acid (MPA, formula CH 3 –PO 3 H 2 ) on ZnO(10-10) surfaces has been investigated by first-principles density-functional total energy calculations. We show that substrate mediated interactions between co-adsorbates can significantly affect the binding energy of MPA on the ZnO surface, which leads to a preferential molecular dimer assembly along the polar [0001] direction (i.e. along the Zn–O dimer direction). We propose that this is caused by a local charge compensation mechanism due to the relaxation of the ZnO surface and suggest that this concept can be applied to other adsorbates on metal oxide surfaces with metal–oxygen dimers.

Temperature-dependent transformations of the adsorption configuration of a molecule containing 2,2′:6′,2″-terpyridine (terpy) end-groups on a flat Cu(1 1 1) surface are studied by a combination of scanning tunneling microscopy (STM) experiments and density functional theory (DFT) calculations. Several quite different adsorption configurations are detected for the terpy end-group, including flat physisorption, distorted chemisorption through N–Cu bonds, and H-dissociative chemisorption through N–Cu and C–Cu bonds. These illustrate and explore various bonding modes of molecules at surfaces, which also imply a greater richness of reaction pathways, both of which are of central importance in a wide variety of heterogeneous catalytic processes on metal catalysts. As deposited on the cold substrate (near 77 K), the molecules are preferably chemisorbed on the surface through N–Cu bonds. With increasing annealing temperature, the molecules are converted to a physisorption configuration at and above 300 K, and above 370 K a small fraction of molecules undergoes dehydrogenation and chemisorb on the surface through N–Cu and C–Cu bonds. The present study demonstrates that the combination of STM measurements and DFT calculations is very effective for probing the atomic details of molecular adsorption configurations on surfaces.

In response to a recent Surface Science Editorial (2010) [1] and accompanying Prospective article [2], we propose that the community of surface scientists and nanoscientists select a common data format for reporting experimental data and atomic coordinates, and have it accepted by relevant journals.

A compound two-dimensional monolayer mixing Mn atoms and 7,7,8,8-tetracyanoquinoclimethane (TCNQ) molecules was synthesized by supramolecular assembly on a Cu(100) surface under ultrahigh-vacuum conditions. The interactions in the Mn(TCNQ)(2) network and in the full system arc analyzed from a molecular orbital perspective and in the light of scanning tunneling microscopy (STM) imaging and simulations. Structural, electronic, and magnetic properties are studied in detail using density functional theory (DFT) calculations. In the absence of Cu and depending on the theoretical method used, the TCNQ species can be formally described as either dianions TCNQ(2-) interacting with Mn4+ cations according to ligand held theory (using GGA calculations) or, alternatively, as radical monoanions interacting with Mn2+ cations in a high spin state (using GGA+U calculations). In the complete system including a Cu substrate, whatever theoretical method is used, TCNQs appear as dianions interacting with both Mn2+ cations in a high spin state (d(5)) and charged copper surface atoms.

This brief overview is dedicated to Professor G. Ertl’s profound influence on shaping the field of surface science. I sketch recent accomplishments toward the determination and understanding of the atomic-scale structure of nanostructures. I argue that, to properly understand nanoscience and develop nanotechnology, there is a dire need to determine many more than the ∼1,000 detailed surface structures known today. Techniques need to be adapted or invented to make this possible. As will be exhibited here, only two techniques are now essentially ready for this task: scanning tunneling microscopy, if interpreted by theory, and, to a lesser extent, x-ray diffraction. I also describe how low-energy electron diffraction is currently being developed for the same purpose.

Surface structural determination by low energy electron diffraction (LEED) requires a fitting procedure between the theoretical and experimental I(V) curves. This fitting procedure is quantified through an R-factor methodology. However, the R-factor space topology presents a large number of local minima. Thus, the task of identifying the global minimum, i.e. the task of finding the correct surface structure, requires a global optimization method that is able to determine the surface structure of complex systems. In this work we present the results of the application of genetic algorithms to three different systems, including performance tests and a comparison with another optimization method previously applied to the LEED problem, simulated annealing. We also present a scaling relationship of the computational effort versus the number of parameters to be fitted for the genetic algorithm method.

Atomic scale surface structure plays an important role indescribing many properties of materials, especially in the case of nanomaterials. One of the most effective techniques for surface structure determination is low-energy electron diffraction (LEED), which can be used in conjunction with optimization to fit simulated LEED intensities to experimental data. This optimization problem has a number of characteristics that make it challenging: it has many local minima, the optimization variables can be either continuous or categorical, the objective function can be discontinuous, there are no exact analytic derivatives (and no derivatives at all for categorical variables), and function evaluations are expensive. In this study, we show how to apply a particular class of optimization methods known as pattern search methods to address these challenges. These methods donot explicitly use derivatives, and are particularly appropriate when categorical variables are present, an important feature that has not been addressed in previous LEED studies. We have found that pattern search methods can produce excellent results, compared to previously used methods, both in terms of performance and locating optimal results.

This overview sketches the historical development of surface science through nanoscience toward nanotechnology. It spans about 50 years of evolution and emphasizes the multidisciplinary nature of this field, as well as its closely-knit interactions with technology, scientific disciplines, phenomena and techniques. A series of examples is given, drawn mainly from work performed in Berkeley, to highlight various stages in the field: surfaces, interfaces, clusters, thin films, superlattices, nanostructures and “bionano”.

Full hemispherical X-ray photoelectron diffraction (XPD) experiments have been performed to investigate at the atomic level ultrathin epitaxial c-axis oriented PbTiO3 (PTO) films grown on Nb-doped SrTiO3 substrates. Comparison between experiment and theory allows us to identify a preferential ferroelectric polarization state in a 60 Å -thick PTO film. Multiple scattering theory based on a cluster-model [ Phys. Rev. B 63, 075404 (2001)] is used to simulate the experiments., This project has been supported by the Swiss National Science Foundation through the National Center of Competence in Research “Materials with Novel Electronic Properties-MaNEP”, and in part by the Office of Science, Materials Sciences Division, of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098.

The surface structures of cubo-octahedral Pt-Mo nanoparticles have been investigated using the Monte Carlo method and modified embedded atom method potentials that we developed for Pt-Mo alloys. The cubo-octahedral Pt-Mo nanoparticles are constructed with disordered fcc configurations, with sizes from 2.5 to 5.0 nm, and with Pt concentrations from 60 to 90 atom %. The equilibrium Pt-Mo nanoparticle configurations were generated through Monte Carlo simulations allowing both atomic displacements and element exchanges at 600 K. We predict that the Pt atoms weakly segregate to the surfaces of such nanoparticles. The Pt concentrations in the surface are calculated to be 5-14 atom % higher than the Pt concentrations of the nanoparticles. Moreover, the Pt atoms preferentially segregate to the facet sites of the surface, while the Pt and Mo atoms tend to alternate along the edges and vertexes of these nanoparticles. We found that decreasing the size or increasing the Pt concentration leads to higher Pt concentrations but fewer Pt-Mo pairs in the Pt-Mo nanoparticle surfaces.

We investigate the extent of diffraction effects (coherent elastic scattering) in electron spectra measured on materials that do not exhibit perfect crystal structure, such as crystals with a high defect density, polycrystalline materials or even amorphous solids. A single scattering cluster (SSC) model was implemented to calculate photoelectron angular distributions for non-ideal crystals. As a first test, results obtained with this code were compared with a well established code for a perfect crystal. The two codes were found to give essentially identical results. Furthermore, SSC calculations of Ni2s photoelectron depth distributions averaged over crystal orientation so as to apply to an amorphous material were compared with Monte-Carlo (MC) simulations of elastic scattering, entirely based on a non-coherent scattering model. The two sets of results were semi-quantitatively consistent.

This paper studies the impact of the properties of the SiN gate insulator on the dc and dynamic performance of AlGaN/GaN Metal Insulator Semiconductor High Electron Mobility Transistors (MIS-HEMTs). We compare the dynamic and transient behaviour of devices with identical epitaxial structure and different gate insulators: RTCVD-SiN (rapid-thermal-chemical-vapour-deposition) and PEALD-SiN (plasma-enhanced-atomic-layer-deposition). We demonstrate the following important results: (i) the gate leakage of devices with PEALD-SiN insulator is three orders of magnitude lower than that of samples with RTCVD-SiN; (ii) the use of PEALD-SiN reduces significantly the transistor threshold voltage hysteresis; (iii) both sets of samples show measurable threshold voltage shift when submitted to forward gate bias. In addition we demonstrate (iv) that the V TH shift is well correlated with the gate forward leakage and bias, for both sets of samples. This result indicates that trapping is induced by the injection of electrons in the gate insulator when a positive bias is applied to the gate; in PEALD SiN devices, the reduction of the gate (forward) leakage results in a significant decrease in V TH shift.

The surface structure of a multilayer LiF(100) thin film grown on Pt(111) from the vapor has been determined by the automated tensor low energy electron diffraction (LEED) method. The final structure, which refined to a Pendry R-factor (RP) of 0.24, had a surface corrugation (D1) of 0.24+-0.04 Angstrom due to the Li+ being displaced towards the bulk, leaving the initially coplanar F - unshifted. A similar intralayer corrugation due to the movement of the Li+ was also observed in the layer immediately under the surface layer, although to a lesser degree: D2=0.07+-0.04 Angstrom. This asymmetric relaxation resulted in the reduction of the first interlayer spacing, d(F2-Li1), to 1.77+-0.0 6 Angstrom from the ideal value of 2.01 Angstrom. The second interlayer spacing, d(Li3-F2), was within error bars of the bulk value, 2.01 Angstrom.

The deposition of homogeneous thin films on porous substrates has been investigated. The thin film deposition of Ta(N) by physical vapor deposition on porous films with different average pore sizes and material compositions has been studied. The continuity of Ta(N) films on top of porous low-k dielectrics is evaluated by means of ellipsometric porosimetry combined with sheet resistance and atomic force microscopy measurements. Interface reactions are analyzed by x-ray photoelectron spectroscopy profiling. It has been observed that the minimal Ta(N) thickness required to obtain a continuous metal layer on top of the porous film depends, on the one hand, on the porosity and pore size and, on the other hand, on the chemical interaction of the thin film with the porous substrate. The sealing of pores is favored by the presence of carbon in the dielectric matrix. This is explained through a mechanism of local enhancement of the degree of crosslinking in the dielectric matrix, catalyzed by Ta.

We use multiple scattering photoelectron diffraction (MSPD) theory to calculate the angular patterns of electrons photoemitted from the K-shell of CO and N 2 gas-phase oriented molecules, as recently measured by several groups. For low (E < 50 eV ) kinetic energies of the photoemitted electron, the electron scattering cannot be adequately represented by spherically symmetric potentials. We thus include nonspherical scattering potentials in our formalism through nondiagonal scattering matrices. We show that intramolecular scattering and interference are responsible for the experimentally measured patterns. This MSPD approach represents a more accurate and versatile method for dealing with such angular distributions as compared to prior calculations of these effects.

New understanding has been reached on competing forces acting to stabilize the polar surfaces of intrinsic ZnO. To compensate an accumulating dipole moment normal to ionic planes, the Madelung electrostatic force and the bonding ability of undercoordinated Zn ions compete to deplete more Zn atoms from the (0001) face and more O atoms from the $(\mathbf{000}\overline{\mathbf{1}})$ face. In this competition, the former mechanism wins because it provides very low energy binding sites for O ions at face-centered-cubic registries on both surfaces. On the Zn-face, a distorted tetrahedral structure is formed, while on the O-face, a vertical Y structure is formed. In both structures, O ions form the topmost atomic plane. The reconstructed polar surfaces containing these novel structures have cleavage energy of 2.36 J/${\mathrm{m}}^{2}$, comparable to that of nonpolar surfaces and in agreement with experimental observation. An earlier structure found on the Zn-face annealed at below 1000 K is stabilized mainly by the Madelung electrostatic force and is a metastable structure of that surface.

Low Energy Electron Diffraction (LEED) is one of the most powerful experimental techniques for surface structure analysis but until now only a trial-and-error approach has been successful. So far, fitting procedures developed to optimize structural and nonstructural parameters—by minimization of the R-factor—have had a fairly small convergence radius, suitable only for local optimization. However, the identification of the global minimum among the several local minima is essential for complex surface structures. Global optimization methods have been applied to LEED structure determination, but they still require starting from structures that are relatively close to the correct one, in order to find the final structure. For complex systems, the number of trial structures and the resulting computation time increase so rapidly that the task of finding the correct model becomes impractical using the present methodologies. In this work we propose a new search method, based on Genetic Algorithms, which is able to determine the correct structural model starting from completely random structures. This method—called here NGA-LEED for Novel Genetic Algorithm for LEED—utilizes bond lengths and symmetry criteria to select reasonable trial structures before performing LEED calculations. This allows a reduction of the parameter space and, consequently of the calculation time, by several orders of magnitude. A refinement of the parameters by least squares fit of simulated annealing is performed only at some intermediate stages and in the final step. The method was successfully tested for two systems, Ag(1 1 1)(4 × 4)-O and Au(1 1 0)-(1 × 2), both in theory versus theory and in theory versus experiment comparisons. Details of the implementation as well as the results for these two systems are presented.

We investigate the atomic structure of the fivefold surface of an icosahedral Al-Cu-Fe alloy, using scanning tunneling microscopy (STM) imaging and a special dynamical low energy-electron diffraction (LEED) method. STM indicates that the step heights adopt (primarily) two values in the ratio of tau, but the spatial distribution of these two values does not follow a Fibonacci sequence, thus breaking the ideal bulk-like quasicrystalline layer stacking order perpendicular to the surface. The appearance of screw dislocations in the STM images is another indication of imperfect quasicrystallinity. On the other hand, the LEED analysis, which was successfully applied to Al-Pd-Mn in a previous study, is equally successful for Al-Cu-Fe. Similar structural features are found for both materials, in particular for interlayer relaxations and surface terminations. Although there is no structural periodicity, there are clear atomic planes in the bulk of the quasicrystal, some of which can be grouped in recurring patterns. The surface tends to form between these grouped layers in both alloys. For Al-Cu-Fe, the step heights measured by STM are consistent with the thicknesses of the grouped layers favored in LEED. These results suggest that the fivefold Al-Cu-Fe surface exhibits a quasicrystalline layering structure, but with stacking defects.

Three forms of electron or x-ray holography `by contrast' are discussed: they all exploit small changes in diffraction conditions to improve image quality and/or extract additional information. Spin-polarized photoelectron holography subtracts spin-down from spin-up holograms so as to image the relative orientations of atomic magnetic moments around an emitter atom. Differential photoelectron holography subtracts holograms taken at slightly different energies so as to overcome the forward-scattering problem that normally degrades the three-dimensional imaging of atoms, particularly for emitter atoms that are part of a bulk substrate environment. Resonant x-ray fluorescence holography also subtracts holograms at slightly different energies, these being chosen above and below an absorption edge of a constituent atom, thus allowing the selective imaging of that type of atom, or what has been referred to as imaging `in true colour'.

We have used COLTRIMS to measure the angular distribution of electrons released from the K-shell of N2 and the carbon K-shell of CO by absorption of one linear polarized photon. For each ionization event which leads to two charged fragments we determine the angle of the photoelectron with respect to the fragment ion momenta. In addition we determine the charge state and energy of the molecular fragments. We find a breakdown of the axial recoil approximation for CO for kinetic energy releases below 10.2 eV, whereas for N2 that approximation is found to be valid for all fragment energies. Furthermore, the photoelectron emission spectrum for N2 is found to be the same for the molecular breakup channels producing N + N + and N + N ++ . (Some figures in this article are in colour only in the electronic version)

We have applied a new differential method of photoelectron holography to both experimental and theoretical multiple-energy photoelectron diffraction data for Cu 3p bulk emission from Cu(001). Dominant forward-scattering effects contained in a normalized hologram χ k have been eliminated by taking the k -difference δχ k along each direction k . Two data sets that are different in k -space sampling ranges and densities have been analyzed to compare atomic images reconstructed from χ and δχ . In both cases, the original scheme has yielded only very weak images of back-scattering atoms and spurious features in the forward-scattering directions. On the other hand, the differential scheme has been found to be successful in suppressing the forward-scattering effects to image not only side-scattering and back-scattering atoms in either case, but also forward-scattering atoms in one of the two cases.

Elastic scattering of low (10–50 eV) kinetic energy electrons from free diatomic molecules is studied using a single-center expansion of the full molecular potential. Dynamic exchange and polarization are included in a local form. The calculated elastic differential scattering cross-sections (DCS) for electron impact on CO and N2 are in good agreement with available experimental data. The importance of using the full molecular potential instead of a two-center potential approach is pointed out. These corrections are small for energies above 50 eV, but they become increasingly important at lower energies. When discussing the angular distributions of elastically-scattered electrons from oriented molecules (like surface adsorbates), we show that these corrections are particularly significant. The results have implications for other electron scattering problems such as those encountered in low-energy photoelectron diffraction from both core and valence levels.

We review the present status of multi-atom resonant photoemission (MARPE), in which the photoelectron intensity from a core level on one atom is influenced by core-level absorption resonances on neighboring atoms of different atomic number. We find that some prior experimental data has been strongly influenced by detector non-linearity and that the effects seen in new corrected data are generally smaller and of different form. The corrected data are found to be well described by an extension of resonant photoemission theory to the interatomic case, provided that interactions beyond the usual second-order Kramers–Heisenberg treatment are included. This microscopic X-ray theory is also found to simplify under certain conditions so as to yield results equivalent to a classical X-ray optical approach, with the latter providing an alternative, although less detailed and general, physical picture of these effects. The present status of experimental results for such MARPE effects is discussed, together with the potential utility of these effects as near-neighbor probes, as well as their implications for X-ray emission and X-ray scattering experiments., This work was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Materials Sciences Division, under Contract No. DE-AC03-76SF00098. Additional support was provided by the Basque Government and by the Fulbright Foundation (J G. de A., Grant No. 22726216-98), as well as the Miller Institute, U.C. Berkeley (E.A.).

We calculated photoemission spectra within the one-step model for the (001) surface of the (AlAs) 2 (GaAs) 2 superlattice structure. The purpose is to discriminate between spectral features which are caused by the surface, and those which are characteristic properties of the superlattice. Direct transitions indicate the opening of band gaps at the Brillouin zone edge, which are characteristic for the modified periodicity of the superlattice. Furthermore, the layer resolved photocurrent shows that one can also identify excitations from AlAs which are hidden below the first two GaAs layers and from As atoms at the boundary between the different semiconductors. Furthermore, emissions from surface states and resonances are also recognized.

The temperature dependence of the first three interlayer distances of the Ag(111) surface was studied by low-energy electron diffraction (LEED) over the temperature range 128K to 723 K. The first three interlayer spacings and the effective Debye temperatures were extracted from the LEED analysis. At the lowest temperature, the first two interlayer spacings are slightly (0.5 percent) contracted. All three interlayer spacings increase with temperature, finally reaching expansions relative to the bulk of about 0.8 percent at the highest temperature studied. The effective surface Debye temperature is lowest for the outermost layer, increasing toward the bulk value for successive layers.

The temperature dependence of the first three interlayer distances of the Ag(111) surface was studied by low-energy electron diffraction (LEED) over the temperature range 128K to 723 K. The first three interlayer spacings and the effective Debye temperatures were extracted from the LEED analysis. At the lowest temperature, the first two interlayer spacings are slightly (0.5 percent) contracted. All three interlayer spacings increase with temperature, finally reaching expansions relative to the bulk of about 0.8 percent at the highest temperature studied. The effective surface Debye temperature is lowest for the outermost layer, increasing toward the bulk value for successive layers.

We have studied the kinetics of the low-pressure adsorption of oxygen on W(110) via time- and chemical-state-resolved photoelectron spectroscopy (PS) and diffraction (PD). Using high-brightness third-generation synchrotron radiation from the Advanced Light Source, together with a new photoelectron spectrometer/diffractometer system, we are able to resolve four distinct chemical states in W 4f spectra (clean surface, bulk, W bound to two O atoms=O2, and W bound to three O atoms=O3) and to measure such spectra in about 20 s each so as to follow the kinetics of oxygen adsorption at 3×10−9 Torr from the clean surface to near saturation. The time-dependent transformations from one state of the surface W atoms to another have been determined at three temperatures of 298, 360, and 593 K. We also find that, for this adsorption pressure on our surface, no long-range-ordered structures are observable in LEED, even though the previously observed ordered structures of p(2×1), p(2×2), and (1×1)×12 are formed at higher pressures of approximately 10−6 Torr. The room-temperature state-resolved PS data are modelled using a simple Monte Carlo approach which assumes no mobility after molecular dissociation, and these calculations are found to describe the experimental data very well. Combining experiment and theory also permits deriving the sticking coefficient as a function of time, yielding results which agree with prior work. Full-solid-angle PD patterns have also been determined at the end of 298 K oxygen exposure for the O2 and O3 W atoms, and these have been analyzed using multiple scattering theory and R-factor analysis. The final local pseudo-threefold hollow geometries for oxygen are found to be very similar to those for a saturated one-monolayer structure of O on W(110) (the (1×1)×12 structure), including a lateral shift of O away from the position corresponding to three equal bond distances, but with some contraction of the OW vertical separation in going from O2 to O3 suggested. This study indicates considerable potential of such time- and state-resolved PS and diffraction for investigating surface reaction kinetics and structure, particularly for the large number of systems that do not exhibit long-range order and in view of future instrumentation improvements that should lead to much shorter data accumulation times and/or higher ambient pressures of measurement.

Attention is drawn to challenges facing the structure determination of compound surfaces of growing complexity. Besides the questions of large unit cells and disorder, and of many independent atoms in the unit cells, one must address the determination of composition, compositional disorder and simultaneously coexisting structures. Of special concern is the issue of stability of complex and compound surfaces, and the resulting concerns about experimental surface preparation and reproducibility.

The structure of Pd (111)+( 3 × 3 ) R30°-CO has been refined using automated tensor LEED, especially to optimize adsorbate-induced relaxations in the substrate. The re-analysis considerably improves the best fit to experiment, while retaining most of the detailed features, compared with the last structural analysis performed on this system by ‘conventional’ quantitative low-energy electron diffraction (LEED). In particular, this study confirms that CO occupies face-centered cubic hollow sites. Relative to the ideal bulk structure, it finds an expansion of the top PdPd interlayer spacing by 0.05±0.04 A, a negligible expansion of the second PdPd interlayer spacing by 0.01±0.06 A, and a possible lateral enlargement by 0.1±0.1 A of the triangle of palladium atoms around the adsorption site. A somewhat small CO interlayer spacing of 1.11±0.04 A is observed, and no unambiguous CO tilt from the surface normal is detected.

It is shown that low-energy electron diffraction (LEED) patterns of the three high-symmetry surfaces (fivefold, threefold and twofold) of icosahedral Al–Pd–Mn are all compatible with quasicrystallinity, under specific conditions of preparation. This conclusion results from comparing symmetries of experimental surface LEED patterns with bulk X-ray diffraction data which are converted to the conditions of the LEED experiment. This conclusion is also based upon an analysis of relative diffraction spot spacings in LEED. Hence, none of the three surfaces exhibits a massive lateral reconstruction, i.e. massive deviation from quasicrystallinity. The LEED pattern of the fivefold surface is distinct from the LEED pattern of the pseudo-tenfold surface of an orthorhombic approximant. We believe that this rules out the possibility that the fivefold surface of the icosahedral quasicrystal reconstructs to an approximant with tenfold or pseudo-10-fold symmetry. The twofold and threefold surfaces facet more readily, indicating qualitatively that they are less stable than the fivefold surface.

The local adsorption structure of CO on Co(0001) has been determined at 160 K using dynamical low-energy electron diffraction (LEED). The CO molecules in the ( 3 × 3 ) R 30°- CO overlayer adopt the on-top site with the CO axis perpendicular to the surface and induce buckling in the top Co layer pulling the Co atom beneath the CO by 0.04±0.04 A outwards. The optimum length for the CO bond is 1.17±0.06 A and that for the CCo distance 1.78±0.06 A. The first-layer CoCo distance is 2.04 A and thus only slightly expanded from the bulk value. The structure obtained is compared with other known CO structures on transition metals.

The structure of Pt(111)+c(4×2)-2CO has been refined using the powerful capabilities of automated tensor LEED, in particular to determine hitherto-unexplored adsorbate-induced relaxations in the substrate. The refinement considerably improves the fit to experiment obtained in a previous structural analysis performed with low-energy electron diffraction (LEED). This study confirms that CO occupies both top and bridge sites. It finds a buckling of the outermost Pt layer, such that the Pt atoms bonded to CO molecules are raised out of the surface by 0.06±0.04 Å relative to the Pt atoms which do not bond to CO. The raised atoms have a spacing of 2.28±0.04 Å to the next Pt layer, expanded from the bulk spacing of 2.2655 Å, while the depressed Pt atoms have a reduced spacing of 2.22±0.04 Å to the next Pt layer. The second Pt layer itself shows minor buckling of 0.03±0.04 Å. No lateral relaxations in the substrate are detected within 0.1 Å. The CO molecules are perpendicular to the surface, with bond lengths of 1.12±0.04 Å and 1.19±0.04 Å in top and bridge sites, respectively. The C–Pd spacings, measured between bonding atoms, are 1.92±0.04 Å and 1.40±0.04 Å in top and bridge sites, respectively.

We have measured the circular dichroism in angular distributions (CDAD) of W 4f emission from a stable one-monolayer structure of oxygen on W(110) that is locally (1×1) in character. Photoelectrons were excited at kinetic energies of approximately 250 eV from the two chemically-shifted peaks present in these spectra (for bulk W and ‘oxide’ W bound to three O atoms) using right and left circularly polarized radiation (RCP and LCP, respectively), as well as linearly polarized radiation (LP) to provide a reference data set. Measurements were made over a large fraction of the solid angle above the surface, using a new beamline and experimental station at the Advanced Light Source. In switching between RCP and LCP, the data are found to show large CDAD effects of up to ±50%, as well as both peak ‘rotation’ effects (as first discussed by Daimon et al. in measurements on Si(100)), and more complex changes that are not so simply described. Multiple scattering diffraction calculations are found to very well reproduce all of the effects seen experimentally, and to represent a more accurate way of dealing with such effects than a prior peak rotation model. The LP data is also found to closely resemble the sum of the RCP and LCP intensities, and this is quantitatively explained. Such CDAD effects are thus expected to be generally seen in all work in solid materials, and important to allow for in studies of magnetic materials, with multiple scattering theory expected to yield a quantitative description of them.

The authors have studied the atomic structure of the clean W(110) surface by means of site-resolved scanned-angle W4f{sub 7/2} photoelectron diffraction (PD) data obtained over nearly the full 2{pi} solid angle above the surface. Prior to the availability of high-brightness sources such as the Advanced Light Source, such large high-resolution data sets were prohibitively time consuming to obtain. The well characterized W(110) system was used as a reference case to check the accuracy of structure determinations from such scanned-angle data via R-factor comparisons of experiment with theoretical multiple scattering calculations. The photoelectron kinetic energy of {approximately}40 eV used was also lower than in many prior PD studies, providing further challenges to theory. The influence of various non-structural theoretical input parameters (e.g., scattering phase shifts, electron inelastic attenuation length, and inner potential) was thus also assessed. A final optimized structure is presented, together with comments on the future applications of this method.

The local atomic geometry of oxygen adsorbed in a stable one monolayer (1×1) structure on W(110) has been determined for the first time from full solid angle site-resolved photoelectron diffraction (PD). This structure also exhibits an additional (1×12) periodicity, but PD allows determination of the local structure within this periodicity. The data were obtained with a new beamline and experimental system at the Advanced Light Source. W 4f 7/2 spectra excited at kinetic energies of approximately 140 eV were resolved into bulk and ‘oxide’ components, with the latter corresponding to the first layer of tungsten atoms bound directly to three oxygen atoms each. Analyzing these data with multiple scattering theory and R-factor analysis shows that the oxygen atoms sit in pseudo-threefold hollow sites at a vertical distance of 0.91±0.05 A above the first tungsten layer, and with a lateral position shifted by 0.1 A along [110] or [110] directions and away from that yielding three equal bond distances. Future applications of this method to systems with large unit cells which make beam-in/beam-out diffraction analyses difficult and/or with a lack of long range order are thus suggested.

Thin films of NaCl were molecularly deposited and subsequently ordered on Pd(100) and Pt(111). These films exhibited the same adsorbate–substrate interaction on Pd(100) and Pt(111) as evidenced by the lone multilayer desorption peak observed in the temperature programmed desorption spectra of NaCl on these two substrates. The ordered structures of the films were produced by exposing a heated substrate to a flux of NaCl emanating from a Knudsen cell and investigated through the use of low-energy electron diffraction (LEED). On Pd(100), the adsorbed multilayer yielded a (1×1) NaCl(100) LEED pattern. The (1×1) pattern was seen as the result of the film's thickness being greater than the sampling depth of the elastically scattered electrons used in the LEED experiment, thus removing the contribution of the substrate's lattice to the observed pattern. For the (1×1) pattern of NaCl(100) on Pd(100), it was found that the NaCl film grew along the [010]-type surface directions of Pd(100) to take advantage of a near 1:1 lattice match. On Pt(111), the ordered NaCl multilayer gave a (1×1) NaCl(100) LEED pattern also, but, unlike Pd(100), one of the overlayer's unit cell vectors lies parallel to that of the substrate's with the other rotated by 90° from the previous one to preserve the geometry of the NaCl(100) lattice. Multiple intensity versus electron energy ( IV ) data sets were gathered for the NaCl(100) multilayer film on both Pd(100) and Pt(111). An analysis of the I – V curves of the multilayer NaCl(100) pattern on both substrates showed that the multilayer has the same structure on both substrates to a depth sampled by the electrons in the experimental energy range. Data from both systems were used in a fully dynamical LEED calculation. In the final structure, which refined down to a Pendry R -factor ( R P ) of 0.16, the largest deviation from the ideally terminated NaCl(100) structure was the movement of the surface Na + towards the bulk, thereby causing a 0.12±0.03 A corrugation of the surface layer. This shift of the surface cations was not followed by a corresponding displacement of the underlying anions, so the first interlayer spacing, d (Cl 2 –Na 1 ), was reduced to 2.72±0.03 A from its bulk value of 2.82 A. Deviations from the bulk were smaller in subsequent layers, e.g. the second interlayer spacing, d (Na 3 –Cl 2 ), was 2.77±0.03 A, and there was essentially no buckling of the second layer.

Low-energy electron diffraction intensity vs voltage (LEED I-V) measurements and analysis are used to determine the multilayer surface relaxation of W(110). Measurements and analysis are presented for both flat and vicinal surfaces, demonstrating that surface roughness leads to only small errors in LEED structure determinations. Flat, clean W(110) exhibits first-(d{sub 12}) and second-(d{sub 23}) layer relaxations of {minus}3.0{plus_minus}1.3{percent} and +0.2{plus_minus}1.3{percent}, respectively, relative to the bulk lattice spacing d{sub 0}=2.237h{Angstrom}. This experimentally determined surface relaxation of W(110) is compatible with a recent combined density-functional theory calculation and LEED study [M. Arnold, G. Hupfauer, P. Bayer, L. Hammer, K. Heinz, B. Kohler, and M. Scheffler, Surf. Sci. {bold 382}, 288 (1997)]. Surface roughness (in the present case, uniform atomic height steps) is found to produce a small apparent increase in the measured value of d{sub 12} when determined using standard (flat surface) LEED I-V methodology. However, for low step densities ({lt}20 atoms/step) the apparent change in d{sub 12} is small compared to other sources of error, so it is unlikely that surface roughness is a significant source of error in LEED structure determinations. {copyright} {ital 1999} {ital The American Physical Society}

A new multi-atom resonant photoemission (MARPE) effect between core levels of neighboring atoms in multi-element samples has been observed. Experimental evidence for the effect and several possible applications are considered. In the metal oxides MnO, Fe 2 O 3 , and La 0.7 Sr 0.3 MnO 3 , we have observed a significant enhancement in the core-level photoelectron peak intensity associated with one element in the sample (e.g. O1s) when the excitation energy is tuned through an energetically deeper absorption edge of a second element (e.g. Mn2p or Fe2p or La3d). The effects observed are up to a 105% increase in peak intensity, and a 29% increase in energy-integrated intensity. A generalization of the theory of single-atom resonant photoemission has also been applied to MARPE, and is found to be in semiquantitative agreement with our experimental results.

Although Gabor originally proposed using electrons in holography to form atomic-resolution images, it is only very recently that photoelectron diffraction data have been discussed as a practical method for achieving this. Such data enable recording both the amplitudes and phases of the scattered waves (relative to a direct reference wave), thus in principle permitting the holographic reconstruction of atomic positions. Photoelectron holography thus holds much promise of at least providing approximate starting structures that can be refined by more conventional comparisons of multiple-scattering calculations with experiment via R-factors. Criteria for optimizing the taking of such data and their holographic photoelectron diffraction patterns have been measured for surface and bulk core-level-shifted W 4f{sub 7/2} photoemission from W(110) on ALS beamline 7.0, yielding a data set of unprecedented size and quality. These data have been compared to multiple scattering theory, and used as a test case, since the W(110) surface is known not to exhibit significant relaxation in its interlayer spacings relative to the bulk. The authors have analyzed these experimental and theoretical results holographically in order to demonstrate the capabilities and limitations of photoelectron holography as a structural probe.

The structure of Rh(111)+(2×2)-3NO has been reanalyzed using automated tensor LEED, in view of new results which strongly suggest different molecular adsorption sites to those previously concluded. The reanalysis confirms and further refines the new model, according to which the NO molecules occupy top sites and both kinds of three-fold hollow sites. These findings are at odds with earlier conclusions from vibrational spectroscopy, which assigned the molecules to top and bridge sites. Together with other metal/NO and metal/CO systems described in the literature, the Rh(111)+(2×2)-3NO phase once again demonstrates that great care has to be taken in adsorption site assignment using the C–O and especially the N–O stretch frequencies.

The possibility of identifying the chemical nature of atomic adsorbates (C, N, O, F and S) on metal surfaces has been explored both theoretically and experimentally. It is shown that, although the tip and the surface may markedly influence the appearance of adatoms in STM images, atomic size and electronegativity are the dominant factors that determine the contrast. In this work, low-temperature (

The Rehr-Albers (RA) separable Green's-function formalism, which is based on an expansion series, has been successful in speeding up multiple-scattering cluster calculations for photoelectron diffraction simulations, particularly in its second-order version. The performance of this formalism is explored here in terms of computational speed, convergence over orders of multiple scattering, over orders of approximation, and over cluster size, by comparison with exact cluster-based formalisms. It is found that the second-order RA approximation [characterized by $(6\ifmmode\times\else\texttimes\fi{}6)$ scattering matrices] is adequate for many situations, particularly if the initial state from which photoemission occurs is of s or p type. For the most general and quantitative applications, higher-order versions of RA may become necessary for d initial states [third-order, i.e., $(10\ifmmode\times\else\texttimes\fi{}10)$ matrices] and f initial states [fourth-order, i.e., $(15\ifmmode\times\else\texttimes\fi{}15)$ matrices]. However, the required RA order decreases as an electron wave proceeds along a multiple-scattering path, and this can be exploited, together with the selective and automated cutoff of weakly contributing matrix elements and paths, to yield computer time savings of at least an order of magnitude with no significant loss of accuracy. Cluster sizes of up to approximately 100 atoms should be sufficient for most problems that require about 5% accuracy in diffracted intensities. Excellent sensitivity to structure is seen in comparisons of second-order theory with variable geometry to exact theory as a fictitious ``experiment.'' Our implementation of the Rehr-Albers formalism thus represents a versatile, quantitative, and efficient method for the accurate simulation of photoelectron diffraction.

Circular dichroism has been observed in the photoelectron diffraction of chemically shifted W $4f$ photoelectrons from the $\mathrm{W}(110)1\ifmmode\times\else\texttimes\fi{}1\ensuremath{-}\mathrm{O}$ surface. The diffraction peaks from first-layer W atoms bonded most directly to oxygen showed an azimuthal rotation similar to those reported in a prior experiment on Si(001). The maximum rotation angles are also in good agreement with those calculated using the formula $\ensuremath{\Delta}\ensuremath{\varphi}=m/kR{\mathrm{sin}}^{2}\ensuremath{\theta}$ derived previously by Daimon et al. [Jpn. J. Appl. Phys., Part 2 32, L1480 (1993)] and assuming $m=4,$ which is the maximum magnetic quantum number of the photoelectron wave function.

The three complementary surface structure probes of X-ray photoelectron diffraction (XPD), scanning tunneling microscopy (STM), and low-energy electron diffraction (LEED) have been combined in a single instrument. This experimental system has been utilized to study the growth of iron oxide films on Pt(111) over the thickness range from 0.75 to 3.00 monolayers (ML). Each film was formed by first depositing an overlayer of pure Fe with a certain coverage in ML (ranging from 0.75 ML to 3.00 ML) and then thermally oxidizing the Fe at a temperature of 980 K and in an oxygen pressure of 4×10−6 Torr. For films up to ∼1 ML in thickness, a bilayer of Fe and O similar to those in bulk FeO parallel to a (111) plane formed. In agreement with a prior STM and LEED study by Galloway et al., we find this bilayer to be an incommensurate oxide film forming a lateral superlattice or Moire structure with short- and long-range periodicities of ∼3.1 and 26.0 A. From the XPD data, in addition, it can be concluded that the topmost oxygen layer is highly relaxed inward by ∼0.6 A as compared to the bulk FeO (111) interplanar spacing, and that the stacking of the topmost O atoms with respect to the underlying Pt is dominated by one of two structurally very similar possibilities. It is furthermore necessary to consider interactions over at least four atomic layers (O, Fe, and the first two Pt layers) to explain this dominance of one stacking type. For thicker iron oxide films from 1.25 to 3.0 ML, the growth mode is essentially Stranski–Krastanov: iron oxide islands form on top of the FeO(111) bilayer mentioned above. For iron oxide films of 3.0 ML thickness, X-ray photoelectron spectroscopy (XPS) yields an Fe 2p3/2 binding energy and an Fe:O stoichiometry consistent with the presence of Fe3O4. XPD data further prove this overlayer to be Fe3O4(111)-magnetite in two almost equally populated domains with a 180° rotation between them. The structural parameters for this Fe3O4 overlayer generally agree with those of a previous LEED study, except that we do not find a terminating partial monolayer of Fe and arrive at a significant difference in the first Fe–O interplanar spacing. Overall, this work demonstrates the considerable benefits to be derived by using this particular set of complementary surface structure probes in such epitaxial growth studies.

In order to understand the unique properties of quasicrystalline surfaces, it is paramount to compare quasicrystalline and crystalline surfaces of similar chemical composition. To that end, the fivefold surface of an AlCuFe quasicrystal substrate was transformed, by ion bombardment and annealing, into a film with the β-Al(Cu 1− x Fe x ) bulk structure. This is a CsCl structure with Fe and Cu atoms randomly occupying the body-center sites, and Al atoms at the corner sites. The film exposes a (110) free surface: its external structure was analyzed by dynamical low-energy electron diffraction in terms of atomic positions and layer-wise composition. The best fit between theory and experiment slightly favors a copper-rich composition in the surface layers ( x =0), although the sensitivity to the value of x is low, because Cu and Fe are very similar electron scatterers. The Al atoms in the surface layer buckle outward by 0.12±0.04 A relative to the Cu or Fe atoms in the same layer, while other relaxations from the bulk structure are small.

Measurements and theoretical calculations are reported for an interatomic multi-atom resonant photoemission (MARPE) effect that permits direct determination of near-neighbor atomic identities (atomic numbers). MARPE occurs when the photon energy is tuned to a core-level absorption edge of an atom neighboring the emitting atom, with the emitting level having a lower binding energy than the resonant level. Large peak-intensity enhancements of 33 to 105 percent and energy-integrated effects of 11 to 29 percent were seen in three metal oxides. MARPE should also be sensitive to bond distance, bonding type, and magnetic order, and be observable via the secondary processes of x-ray fluorescence and Auger decay.

We apply our recent Green's function based STM formalism to study the adsorption of S monomers and clusters on Re(0001). Starting from previous STM experimental data and LEED results, we consider surface relaxations common to the different S structures observed on this system. The role of such relaxations on the final shape of the theoretical STM images is then examined, and we find that the sensitivity of the images to these structural parameters can be considerable in certain cases. The consistency of proposed relaxed models is also confirmed after a theory–experiment comparison of the STM images.