Surface Metal Atoms Research Articles

The role of atomic ensembles in the reactivity of bimetallic electrocatalysts.

Bimetallic electrodes are used in a number of electrochemical processes, but the role of particular arrangements of surface metal atoms (ensembles) has not been studied directly. We have evaluated the electrochemical/catalytic properties of defined atomic ensembles in atomically flat PdAu(111) electrodes with variable surface stoichiometry that were prepared by controlled electrodeposition on Au(111). These properties are derived from infrared spectroscopic and voltammetric data obtained for electrode surfaces for which the concentration and distribution of the respective metal atoms are determined in situ by atomic resolution scanning tunneling microscopy with chemical contrast. Palladium monomers are identified as the smallest ensemble ("critical ensemble") for carbon monoxide adsorption and oxidation, whereas hydrogen adsorption requires at least palladium dimers.

Surface energy and the early stages of oxidation of NiAl(110)

We have studied the (110) surface of NiAl, an ordered alloy of B2 structure, using a plane-wave pseudopotential method. The clean surface and several oxidized surfaces were investigated, with oxygen coverages up to 1.5 ML (1 ML = 1 O-atom per surface metal atom). In order to compare the energies of the oxidized structures, which comprise different numbers of metal and oxygen atoms, one has to take account of the chemical potentials of the Ni, Al and O. To this end, for each system, we have applied simple analytic models to study their surface energy as a function of temperature, alloy stoichiometry (assumed to be near 50–50) and oxygen partial pressure. The calculations predict how, at oxygen pressures just above the threshold for decomposing NiAl, the clean surface should be coated by an alumina layer, with the consequent depletion of Ni near the surface. Varying stoichiometry has the relatively minor effect of shifting the crossovers in oxygen pressure at which different oxidized surfaces become stable.

Reactivity of Site-Isolated Metal Clusters: Propylidyne on γ-Al2O3-Supported Ir4

To contrast the reactivity of supported metal clusters with that of extended metal surfaces, we investigated the reactions of tetrairidium clusters supported on porous gamma-Al2O3 (Ir4/gamma-Al2O3) with propene and with H2. Infrared, 13C NMR, and extended X-ray absorption fine-structure spectroscopy were used to characterize the ligands formed on the clusters. Propene adsorption onto Ir4/gamma-Al2O3 at 298 K gave stable, cluster-bound mu3-propylidyne. Propene adsorbed onto Ir4/gamma-Al2O3 at 138 K reacted at approximately 219 K to form a stable, highly dehydrogenated, cluster-bound hydrocarbon species approximated as CxHy (such as, for example, C3H2 or C2H). H2 reacted with Ir4/gamma-Al2O3 at 298 K, forming ligands (likely hydrides), which prevented subsequent reaction of the clusters with propene to form propylidyne. Propylidyne on Ir4 was stable in helium or H2 as the sample was heated to 523 K, whereupon it reacted with oxygen of the support to give CO. Propylidyne on Ir4 did not undergo isotopic exchange in the presence of D2 at 298 K. In contrast, the literature shows that propylidyne chemisorbed on extended metal surfaces is hydrogenated in the presence of H2 (or D2) and exchanges hydrogen with gaseous D2 at room temperature; in the absence of H2, it decomposes thermally to give hydrocarbon fragments at temperatures much less than 523 K. The striking difference in reactivities of propylidyne on clusters and propylidyne on extended metal surfaces implies the requirement of ensembles of more than the three metal surface atoms bonded to propylidyne in the surface reactions. The results highlight the unique reactivity of small site-isolated metal clusters.

Additional Investigations of a New Kinetic Method To Follow Transition-Metal Nanocluster Formation, Including the Discovery of Heterolytic Hydrogen Activation in Nanocluster Nucleation Reactions

A few years ago we developed a new kinetic method for following transition-metal nanocluster formation in which the resultant nanocluster's catalytic activity was used as a reporter reaction via the pseudoelementary step concept. This method in turn yielded insights into a new, broadly applicable mechanism of nanocluster formation under H2 consisting of (a) slow, continuous nucleation, A → B, followed by (b) fast autocatalytic surface growth, A + B → 2B (A = the nanocluster precursor, [Bu4N]5Na3[(1,5-COD)Ir·P2W15Nb3O62], B = the resultant nanocluster's surface metal atoms), in which the nanocluster behaves as a “living metal polymer”. Herein, this new kinetic method is investigated and tested further: (i) by following the Ir(0)∼300 nanocluster's kinetics of formation more directly via the H2 uptake reaction of the [Bu4N]5Na3[(1,5-COD)Ir·P2W15Nb3O62] precursordoes this also show an autocatalytic H2 uptake curve?; (ii) by seeing if the predicted initially small, then larger (past the induction period) size...

First-Principles Study of the Surface Electronic Structures of Transition Metal Carbides

Surfaces of transition metal carbides (TiC, ZrC, NbC, HfC and TaC(001)-1×1) are investigated using the first-principles molecular dynamics (FPMD) method. By the full structural optimization of the surface, the carbon and transition metal atoms on the top layer move outward and inward, respectively. All the calculated electronic states of surfaces are metallic. A non-linear core correction is considered for pseudopotentials of transition metals and plays an important role in the structural optimization.

A density functional theory study of CO and atomic oxygen chemisorption on Pt(111)

Ab initio total energy calculations within a density functional theory framework have been performed for CO and atomic oxygen chemisorbed on the Pt(111) surface. Optimised geometries and chemisorption energies for CO and O on four high-symmetry sites, namely the top, bridge, fcc hollow and hcp hollow sites, are presented, the coverage in all cases being 0.25 ML. The differences in CO adsorption energies between these sites are found to be small, suggesting that the potential energy surface for CO diffusion across Pt(111) is relatively flat. The 5σ and 2π molecular orbitals of CO are found to contribute to bonding with the metal. Some mixing of the 4σ and 1π molecular orbitals with metal states is also observed. For atomic oxygen, the most stable adsorption site is found to be the fcc hollow site, followed in decreasing order of stability by the hcp hollow and bridge sites, with the top site being the least stable. The differences in chemisorption energies between sites for oxygen are larger than in the case of CO, suggesting a higher barrier to diffusion for atomic oxygen. The co-adsorption of CO and O has also been investigated. Calculated chemisorption energies for CO on an O/fcc-precovered surface show that of the available chemisorption sites, the top site at the oxygen atom's next-nearest neighbour surface metal atom is the most stable, with the other four sites calculated being at least 0.29 eV less stable. The trend of CO site stability in the coadsorption system is explained in terms of a ‘bonding competition’ model.

Periodic Trends in Hydrotreating Catalysis: Thiophene Hydrodesulfurization over Carbon-Supported 4d Transition Metal Sulfides

The kinetics of atmospheric gas-phase thiophene hydrodesulfurization (HDS) over five carbon-supported 4d transition metal sulfide catalysts (Mo, Ru, Rh, Pd, and CoMo) were studied. Reaction orders (thiophene, H2S, and H2), apparent activation energies, and pre-exponential factors were determined. The activity trends for these catalysts follow the well-known volcano-shape curve. The most active catalyst shows the lowest thiophene reaction order, which is taken to imply that a strong interaction between transition metal sulfide (TMS) and thiophene results in a high HDS activity. The kinetic results are interpreted in terms of trends in metal–sulfur bond energy. These trends are counter to commonly held correlations between metal–sulfur bond energy and periodic position of the transition metal. Both Sabatier's principle and the “bond energy model” appear to be inadequate in explaining the observed trends in kinetic parameters. Instead, an alternative proposal is made: the metal–sulfur bond strength at the TMS surface relevant to HDS catalysis depends on the sulfur coordination number of the surface metal atoms. Transition metals (TM) at the left-hand side of the periodic table, i.e., Mo, form stable sulfides, leading to a low sulfur addition energy under reaction conditions. The sulfur addition energy is the energy gained upon addition of a sulfur atom (e.g., in the form of thiophene) to the TMS. Over to the right-hand side of the periodic table, the stability of the TMS decreases due to lower bulk metal–sulfur bond energies. This can result in more coordinative unsaturation of the TM surface atoms and possibly the formation of incompletely sulfided phases with higher sulfur addition energies. At the right-hand side of the periodic table the activity decreases due to weak metal–sulfur interactions, leading to poisoning of the metallic state.

Qualitative Evidence for Adsorbate-Induced Edge Metal Atom Localization: The Characterization of Lifted Backbond Degeneracy in Adsorbed CO through the Model Adsorption of Methyl Isocyanide on Pd/Al2O3

Qualitative synthetic chemistry strategies are used to show that electron-deficient (i.e., unstable) edge Pd atoms can find stability through separation from the surface band structure and strong bonding to a CO adsorbate state that displays lifted backbond degeneracy. This conclusion is drawn from the transmission infrared observation of a thermally activated bonding mode of CO on Pd/Al2O3 that has characteristics inconsistent with conventional atop- or bridge-bonded CO but similarities with an “atop bent” methylisocyanide adsorbate mode, characterized on the same surface, which is known to have lifted backbond degeneracy. Analysis of qualitative observations in both the data and the literature indicates the thermally stable CO state to be atop bound to an edge atom. Since this state is populated over a wide temperature range, the ΔH for its formation must arise from the attached surface metal atom gaining stability in the process even at the expense of CO stability. Published work [J. Phys. Chem.93, 4890 (1989); Phys. Rev. B24, 754 (1981)] indicates that unstable edge atoms are the likely candidates for that behavior. A TLEED study [Chem. Phys. Lett.201, 393 (1993)] that reports the identical CO adsorbate state on Pd(110) and a study reporting the ability of adsorbed CO to reconstruct the Pd(110) surface [Chem. Phys. Lett.167, 391 (1990)semi; Surf. Sci.249, 1 (1991)] are published observations that are consistent with this conclusion.

New model for ion neutralization at surfaces

We investigate the neutralization of low energy He + ions in close collisions with metal surface atoms. In order to describe the neutralization process as completely as possible, we consider Auger neutralization (AN), resonant neutralization (RN) and resonant ionization (RI). Our calculation agrees well with experimental data and shows that in some metals (like Pd) AN is the dominant process, whereas in others (like Al) RN and RI contribute significantly for energies above the threshold for reionization.

Characterization of Model Three-Way Catalysts: III. Infrared Study of the Surface Composition of Platinum–Rhodium Ceria–Alumina Catalysts

The surface composition of model and commercial bimetallic PtRh catalysts supported on ceria–alumina was tentatively determined by FTIR spectroscopy using the successive adsorption of NO at 473 K and CO at 298 K method which was previously applied to the case of PtRh/Al2O3 catalysts. The study of monometallic model catalysts shows that the adsorption of NO at 473 K on rhodium gives an intense band at 1912 cm−1 which is not modified by the presence of ceria and therefore can be used to quantify the number of surface rhodium atoms. However, the chemisorption properties of platinum toward the subsequent adsorption of CO at 298 K are strongly modified and the quantification of the surface platinum atoms could not be directly measured. However, their number could be indirectly obtained by combining the NO adsorption for rhodium and hydrogen volumetric adsorption for the total number of surface metal atoms, assuming that NO adsorption does not induce an important segregation of rhodium for bimetallic particles. Thus, for the fresh bimetallic model catalyst, the same composition was obtained at the surface and in the bulk. After aging at 1273 K, no rhodium was detected at the surface. This absence of surface rhodium atoms was confirmed by the data obtained from the direct adsorption of CO on platinum at room temperature, which gave the same number of surface atoms as hydrogen chemisorption. Similar results have been obtained with two commercial three-way catalysts. In particular, after aging at 1273 K, rhodium was practically not detected on the surface.

Theoretical analysis of hydrogen chemisorption on Pd(111), Re(0001) andPdML/Re(0001),ReML/Pd(111)pseudomorphic overlayers

Gradient-corrected density-functional theory (DFT-GGA) periodic slab calculations have been used to analyze the binding of atomic hydrogen on monometallic Pd(111), Re(0001), and bimetallic ${\mathrm{Pd}}_{\mathrm{ML}}/\mathrm{R}\mathrm{e}(0001)$ [pseudomorphic monolayer of Pd(111) on Re(0001)] and ${\mathrm{Re}}_{\mathrm{ML}}/\mathrm{P}\mathrm{d}(111)$ surfaces. The computed binding energies of atomic hydrogen adsorbed in the fcc hollow site, at 100% surface coverage, on the Pd(111), Re(0001), ${\mathrm{Pd}}_{\mathrm{ML}}/\mathrm{R}\mathrm{e}(0001),$ and ${\mathrm{Re}}_{\mathrm{ML}}/\mathrm{P}\mathrm{d}(111)$ surfaces, are -2.66, -2.82, -2.25, and -2.78 eV, respectively. Formal chemisorption theory was used to correlate the predicted binding energy with the location of the d-band center of the bare metal surfaces, using a model developed by Hammer and N\o{}rskov. The DFT-computed adsorption energies were also analyzed on the basis of the density of states (DOS) at the Fermi level for the clean metal surfaces. The results indicate a clear correlation between the d-band center of the surface metal atoms and the hydrogen chemisorption energy. The further the d-band center is from the Fermi level, the weaker is the chemisorption bond of atomic hydrogen on the surface. Although the DOS at the Fermi level may be related to the location of the d-band, it does not appear to provide an independent parameter for assessing surface reactivity. The weak chemisorption of hydrogen on the ${\mathrm{Pd}}_{\mathrm{ML}}/\mathrm{R}\mathrm{e}(0001)$ surface relates to substantial lowering of the d-band center of Pd, when it is pseudomorphically deposited as a monolayer on a Re substrate.

Methane steam reforming over rhodium promoted Ni/Al2O3 catalysts

The effect of Rh addition upon catalyst characteristics and performance in methane steam reforming was investigated using Rh-promoted Ni/Al2O3 catalysts. The number of reduced metal atoms exposed on the surface increased for the Rh-promoted catalysts. Rh-promoted catalysts showed an increase in CH4 reforming activity; however, constant turnover frequencies for promoted and unpromoted catalysts suggest that the increase in the number of metal surface atoms caused the activity enhancement. Rh also facilitated reduction of Ni/Al2O3.

Dependence of stretching frequency on surface coverage and adsorbate–adsorbate interactions: a density-functional theory approach of CO on Pd (111)

Total energy and stretching frequency calculations ν(C–O) are reported for the chemisorption of CO on a Pd (111) surface as a function of coverage with a periodic density-functional theory approach including generalized gradient approximation. At low [0.33 monolayer (ML)] and high (0.75 ML) coverages, the most stable structures always present CO in threefold fcc and hcp hollow sites, mixed with top sites for the high coverage structure, in agreement with the interpretation of the RAIRS or HREELS spectra. The calculated anharmonic frequencies for these sites (1828–1830 cm −1 at 0.33 ML) and (1893 and 2085 cm −1 at 0.75 ML) are coherent with the observed IR peaks. At medium coverage, 0.5 ML, the most stable model is the one with both fcc and hcp hollow sites. The observed blue shift of the frequency band between 0.33 ML and 0.5 ML is usually interpreted as a change of the chemisorption site from a hollow to a bridge adsorption. Although a direct comparison of the IR peak with the calculated anharmonic values at 0.5 ML for the structures with two hollow sites (1906 cm −1) or with two bridge sites (1937 cm −1) is this time more delicate, the calculations show that the observed frequency shift is fully compatible with the hollow site model and can be attributed to increased coverage and adsorbate–adsorbate interactions. The results show that two main phenomena occur when the coverage increases. First the shift caused by static interactions between adsorbates is ruled by the coverage-dependent backdonation from the surface metallic atoms towards the antibonding 2π molecular orbitals of the chemisorbed CO molecules. The second shift is due to dynamic interactions between adsorbates. The stability of a vibrational configuration taken from a particular mode is ruled by the electrostatic energy cost due to the competitive charge transfer between the adsorbates and the metallic surface.

Inelastic ion collisions involving metal surfaces exposed to oxygen and electron transfer processes

The effect of oxygen adsorption on excited-state production in ion surface collisions, relevant to the production of scattered and recoiled ions is analysed on the basis of recent studies on the modification of resonant electron capture rates. Experiments on negative ion formation show that electron capture is strongly affected by changes in local electronic structure and electrostatic potentials around adsorbate sites, as this had been discussed in the context of promotion and poisoning of reactions at surfaces. A comparative analysis for the case of production of excited states of inert gas ions and recoiled surface metal atoms and ions is given and shows that the results of some detailed recent experimental studies can be qualitatively understood in these terms.

Photodetachment of surface atoms of metals

Photodetachment of surface atoms of metals

A theoretical explanation of the surface diffusion mechanism in metal electrodes in contact with electrolytes

A theoretical explanation for the surface diffusion mechanism observed in columnar structured metal electrodes in contact with electrolytes is given. The potential energy of a surface metal atom on which ions forming part of the supporting electrolyte are adsorbed is described by means of an anharmonic oscillator curve whereas the energy of a surface metal atom liberated from any adsorption interaction is approximated by a harmonic oscillator energy fuction. Geometric arguments allow to define a symmetry factor δ for which experimental values were previously obtained. A qualitative interpretation of the value of δ has been made.

Application of low‐energy noble‐gas ion scattering to the quantitative surface compositional analysis of binary alloys and metal oxides

Low-energy noble-gas ion scattering (LEIS) probes the outermost atomic layer of a material, but a quantitative compositional analysis of this layer is not straightforward. It is demonstrated that quantification by calibration can be done, assuming that ion fractions and shielding effects are the same for the reference sample and sample of interest. These assumptions are critically evaluated and LEIS experiments on binary alloys and metal oxides are presented that can partly verify these assumptions. The LEIS measurements of a Cu–Au alloy and CuO powder as a function of initial energy indicate the absence of matrix effects in the ion fractions after scattering from the metal atoms in these systems. In metal oxides, shielding of surface metal atoms by shadowing/blocking and ion neutralization by neighbouring atoms can significantly influence the quantification of the metal atom concentration and is determined by the local atomic arrangement as illustrated by LEIS experiments of CuO and ZnO samples.

Application of low‐energy noble‐gas ion scattering to the quantitative surface compositional analysis of binary alloys and metal oxides

Low-energy noble-gas ion scattering (LEIS) probes the outermost atomic layer of a material, but a quantitative compositional analysis of this layer is not straightforward. It is demonstrated that quantification by calibration can be done, assuming that ion fractions and shielding effects are the same for the reference sample and sample of interest. These assumptions are critically evaluated and LEIS experiments on binary alloys and metal oxides are presented that can partly verify these assumptions. The LEIS measurements of a Cu–Au alloy and CuO powder as a function of initial energy indicate the absence of matrix effects in the ion fractions after scattering from the metal atoms in these systems. In metal oxides, shielding of surface metal atoms by shadowing/blocking and ion neutralization by neighbouring atoms can significantly influence the quantification of the metal atom concentration and is determined by the local atomic arrangement as illustrated by LEIS experiments of CuO and ZnO samples. © 1998 John Wiley & Sons, Ltd.

Electronic effects and effects of particle morphology in n-hexane conversion over zeolite-supported platinum catalysts

n-Hexane conversion over zeolite-supported platinum catalysts has been studied at 360°C in a packed bed reactor. In the absence of sulfur impurities, the deactivation appears to be due to two types of coke formation. The propensity for terminal hydrogenolysis is primarily determined by the particle morphology, i.e., the type of surface sites exposed. The increase in benzene selectivity correlates with a shift to lower stretching frequencies of the CO absorption bands, indicating that an increase in the electron density at the surface metal atoms results in higher benzene selectivity. The effect of the support in the high activity and aromatization selectivity of a monofunctional platinum catalyst for n-hexane conversion is observed to be twofold, i.e., the stabilization of extremely small metal particles of specific morphology under reaction conditions and metal-support interaction, resulting in an increased electron density over the metal particles.

High pressure catalytic processes studied by infrared-visible sum frequency generation

The recent development of infrared-visible sum frequency generation (SFG), a surface-specific vibrational spectroscopy, has helped bridge the pressure gap between studies of heterogeneous catalysis under high vacuum and atmospheric pressure. This is achieved by in situ monitoring of surface species at high pressure via their SFG vibrational spectra and correlating the results with the simultaneously measured reaction rate using gas chromatography. Examples of systems studied include olefin hydrogenation and carbon monoxide oxidation over the (111) crystalline face of platinum. In these examples, the studies succeed in revealing the molecular details of the surface reactions. Identification of key intermediates and their concentrations has made it possible for the first time to calculate turn over rates per active surface species rather than just per exposed surface metal atom. In all cases, the key intermediate of the reaction is not detectable on the surface in UHV under similar temperatures.