Structural aspects of chemisorption at Cu(110) revealed at the atomic level

We illustrate the impact that scanning tunnelling microscopy (STM) has made on our understanding of chemisorption and catalysis at metal surfaces at the atomic level by considering four examples where information from surface sensitive techniques was also available. The advantages of STM and the limitations of some of the other experimental methods are discussed. (1) Through a combination of STM and X-ray photoelectron spectroscopy (XPS) we have established that a number of distinct oxygen chemisorbed states can exist at a Cu(110) surface. These are metastable and temperature dependent. Furthermore, the presence of chemisorbed sulphur is shown to promote a specific oxygen state – isolated oxygen strings – which are likely to be more chemically reactive than the oxygen overlayer present at Cu(110). In this sense the sulphur is a structural promoter. (2) The oxidation of ammonia under ammonia-rich conditions results in the growth of imide (NH) strings at a Cu(110) surface and this has been followed quantitatively by STM. The reactive surface oxygen state participates in an oxydehydrogenation reaction generating NH-radical species which undergo surface diffusion and result in string growth. (3) Nitric oxide dissociates at Cu(110) to generate a two-phase system of chemisorbed nitrogen and oxygen adatoms. The oxygen is present in a well ordered (2 × 1) structure and the nitrogen in a (3 × 2) structure. The limitations of an earlier LEED study are discussed. (4) Structural aspects of chemisorbed sulphur generated by the dissociative chemisorption of hydrogen sulphide and methyl mercaptan are discussed. In the latter case carbon–sulphur bond cleavage results in the formation of a sulphide overlayer at 450 K with complete removal of carbon as desorbed hydrocarbons. Various sulphur structures have been delineated over a wide temperature range. This revised version was published online in August 2006 with corrections to the Cover Date.     

Surface composition of nickel-gold alloys

Nickel-gold alloy foil surfaces were analyzed with Auger electron spectroscopy. The gold atom fraction, x^s, at a clean surface is higher than that in the bulk, x^b. Thus, for an alloy with x^b = 0.005, the clean surface showed x^s = 0.5. This alloy chemisorbed oxygen at room temperature, after which gold was not detected at the surface. However, oxygen was not chemisorbed at room temperature on alloys with x^s = 0.7 or 0.86 and these values of x^s corresponding to the clean surface did not change after 3 × 10^?4 Torr/sec oxygen exposures. The gold enrichment at the clean surface can be explained quantitatively by a simple thermodynamic argument. The change in x^s due to oxygen is understood qualitatively. The lack of chemisorption of oxygen on surfaces containing as much as 30% nickel suggests electronic effects due to alloying.     

Oxygen-induced surface state on diamond (100)

The electronic structure of oxygenated diamond (100) surface is studied comparatively by experimental photoemission techniques and first principles calculations. Controlled oxygenation of the diamond (100) 2×1 surface at 300°C yields a smooth O:C (100) 1×1 surface with a distinctive emission state at ∼3 eV from the Fermi edge. Oxygenation of the hydrogenated surface at temperatures above 500°C, however, gives rise to extensive etching and roughening of the surface. The experimentally observed emission state at ∼3 eV following O adsorption is assigned to the O-induced surface state. When the oxygenated surface is annealed to 800°C to desorb chemisorbed O, the surface structure changes from 1×1 to 2×1 and another surface state emission at 2.5 eV associated with the clean surface reconstruction can be observed by UPS. This is attributed to the π-bond reconstruction of sub-surface carbon layers following the desorption of first layer CO from the surface. To understand the origin of the O-induced emission state, we calculated the density of states (DOS) of the oxygenated diamond using the first principles linear muffin-tin orbital (LMTO) method with atomic sphere approximation (ASA) based on density functional theory (DFT) and local density approximation (LDA).     

Probing the Influence of Alloyed Sn on Pt(100) Surface Chemistry by CO Chemisorption

The chemisorption properties of the c(2×2) and (3√2×√2)R45° Sn/Pt(100) alloy surfaces, along with the clean Pt(100) surface, were investigated using CO as a probe molecule. Temperature-programmed desorption (TPD) studies revealed a reduction in CO desorption peak temperature, and thus the chemisorption bond energy, by alloying Sn into the Pt(100) surface. A large decrease was observed in the saturation coverage of CO on these alloyed surfaces at 150 K compared to the Pt(100) surface. The initial sticking coefficient of CO was found, however, to be nearly independent of the surface Sn concentration. High-resolution electron energy loss spectroscopy (HREELS) studies showed that CO was only chemisorbed on atop sites on both alloys. Sn incorporation results in isolated Pt atoms at the surface of these two Sn/Pt(100) alloys and eliminates the possibility of CO bonding to multiple Pt centers, i.e., to pure-Pt 2- and 3-fold bridging sites.     

Intermolecular migration of methyl groups at a Cu(211) surface

The exposure of preadsorbed oxygen to monomethylamine or the coadsorption of a 2:1 monomethylamine (CH₃NH₂)/dioxygen mixture at a Cu(211) surface at room temperature results in the formation of a surface species characterised by C(1s) and N(1s) binding energies of 285.2 and 398.2 eV, respectively, with a calculated carbon to nitrogen ratio of 2:1. This species, which we assign to a chemisorbed dimethylamine ((CH₃)₂NH_x(a)), is the only adsorbed product of the reaction and its formation must involve the breaking of a carbon–nitrogen bond and the intermolecular transfer of a methyl group. This revised version was published online in November 2006 with corrections to the Cover Date.     

A study of the interaction of nitric oxide with nickel and oxidized nickel surfaces by X-Ray photoelectron spectroscopy

The adsorption of nitric oxide by nickel and also nickel surfaces preexposed to oxygen has been studied by X-ray photoelectron spectroscopy in the temperature range 80 to 290 °K. The surface reactivity to NO interaction has been controlled by varying the temperature and conditions of preexposure to oxygen. This has enabled three distinct states of NO adsorption to be delineated, a dissociative state and two molecular states. One of the molecular states is weakly adsorbed while the other, more strongly chemisorbed, is the precursor to dissociation. We suggest the latter is adsorbed in the “bent” form while the former is linearly bonded to the surface. N₂O is also observed within the adlayer at 80 °K particularly with nickel surfaces preoxidized at 290 °K. The N₂O is weakly adsorbed. There are strong analogies between the present data and previous studies with copper, iron, and aluminum, metals with very different electronic structures. It should be noted, however, that although in all cases a new surface phase is generated, the inherent reactivities of the clean metals differ only in degree rather than in kind.     

A Combined XPS/STM and TPD Study of the Chemisorption and Reactions of Methyl Mercaptan at a Cu(110) Surface

Despite the structural similarities between methanol and methyl mercaptan (CH₃SH), replacing the oxygen atom by sulfur has a profound effect on their chemistry at copper surfaces. In a combined STM, XPS and TPD study of the reaction of methyl mercaptan with clean and partially oxidised Cu(110) surfaces we have found that unlike methanol, the scission of the SH bond (both in the presence of and in the absence of oxygen) to give adsorbed mercaptide (CH₃S(a)) results in a restructuring of the surface. Low concentrations of adsorbed mercaptide (< 1 × 10¹⁴ cm^-2) result in a severe degradation of the STM image due to the high mobility of the surface adlayer. A stable surface can be regained by increasing the concentration of mercaptide or the presence of another adsorbate such as oxygen. The reconstructed surface is characterised by very narrow terraces (typically 10-15Å wide) orientated mainly in the 1¯10 direction with a zig-zag structure. Higher resolution images of the terraces reveal an atomic scale structure with a c(2 × 2) unit cell, each cell containing (in total) two bright features. The XPS data confirm that mercaptide is present and show that the concentration at which islands of mercaptide become visible in the STM images is approximately 3 × 10¹⁴ cm^-2. At a surface concentration of 5.1 × 10¹⁴ cm^-2 the c(2 × 2) structure is seen to be complete, consistent with the c(2 × 2) unit cell containing two mercaptide species. On heating to 450K the mercaptide dissociates to give chemisorbed sulfur adatoms and methane. The latter implies that the hydrogen formed when the methyl mercaptan dissociates remains chemisorbed at the surface until removed by reaction with the methyl groups. At pre-oxidised surfaces where chemisorbed hydrogen is removed as water, mercaptide decomposition leads to ethane desorption with minor methane and ethane components, the latter indicating that methyl dehydrogenation is possible at the copper surface. STM shows that following the decomposition of the mercaptide adlayer the copper surface regains its original structure of broad terraces (typically 100-200Å wide), although a monolayer of chemisorbed sulfur is now present.Oxygen chemisorption is completely inhibited by mercaptide concentrations of 5 × 10¹⁴ cm^-2 but occurs at lower concentrations. The STM images show that the mercaptide and oxygen adsorbates form separate islands, in contrast to the coadsorption of hydrogen sulfide and oxygen. The influence of oxygen on the thermal desorption of mercaptide is discussed in the light of the structural data.     

A new approach to the mechanism of heterogeneously catalysed reactions: the oxydehydrogenation of ammonia at a Cu(111) surface

The oxydehydrogenation of ammonia at a Cu(111) surface is a highly efficient process at 295 K, with the selectivity sensitive to the dioxygen-ammonia ratio. However, there is no evidence from either XPS or HREELS for surface oxygen being present during the reaction and, in effect, catalysis occurs at a clean Cu(111) surface. The rate of NH _x (a) formation is indistinguishable from the rate of the dissociative chemisorption of oxygen at close to zero coverage suggesting that the reactive oxygen species are the hot transients O-(s). The chemisorbed oxygen overlayer, the O^2-(a)-like species are, by comparison, unreactive. The reaction is, therefore, not characteristic of either Eley-Rideal or Langmuir-Hinshelwood mechanisms but involves the interaction of rapidly diffusing ammonia molecules and hot transient O-(s)-like species. Models for this type of reaction have been discussed previously, while very recent studies by scanning tunnelling microscopy have provided further evidence for such oxygen transients.     

Curing and thermal behavior of diglycidyl ether of bisphenol A in the presence of a mixture of amines

The curing behavior of diglycidyl ether of bisphenol A (DGEBA) was investigated by differential scanning calorimetry with mixtures of silicon‐containing amide–amines and diaminodiphenyl sulfone (DDS). Silicon‐containing amide–amines were prepared by the reaction of 2.5 mol of 4,4′‐diaminodiphenyl ether (E), 4,4′‐diaminodiphenyl methane (M), 3,3′‐diaminodiphenyl sulfone (mS), 4,4′‐diaminodiphenyl sulfone (pS), bis(3‐aminophenyl) methyl phosphine oxide (B), or tris(3‐aminophenyl) phosphine oxide (T) with 1 mol of bis(4‐chlorobenzoyl) dimethyl silane. Mixtures of the amide–amines and DDS at ratios of 0:1, 0.25:0.75, 0.5:0.5, 0.75:0.25, and 1:0 were used to investigate the curing behavior of DGEBA. A single exotherm was observed on curing with a mixture of amide–amine and DDS. This clearly shows that the mixture participated in the cocuring reaction. The peak exotherm temperature depended on the structure and the molar ratio of amide–amines. With all of the amide–amines and DDS, a significant decrease in the kick‐off temperature of the curing exotherm was observed on the incorporation of a 0.25 molar fraction of amide–amines. Thus, with the mixture, the curing temperatures were reduced and were lowest for ether‐containing amide‐amines and highest for methylene‐containing amide–amines. The char yield was almost similar in the samples cured with amide–amines (E, pS, or mS) or DDS. The char yield was higher than for either of the constituents when a mixture was used. A synergistic behavior was observed when a mixture of E, M, mS, or pS and DDS was used, whereas mixture of B or T and DDS showed antigonism in the char yield. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 90: 1739–1747, 2003     

Kinetics and mechanism of reduction of oxide film on smooth platinum electrodes with hydrogen

The kinetics and mechanism of the reaction of chemisorbed oxide (Pt-O) layer on a smooth Pt electrode with H₂ dissolved in 1 M H₂SO₄ solution were investigated under the open circuit condition.It was found that a monolayer of Pt-O on the electrode surface is reduced first at a slow rate which is of second order in chemisorbed oxygen in the range 1 ≧ θ ≥ 0·65 and then at a rapid rate which is proportional to (1 – nθ)² in the range 0·6 ≥ θ > 0, where θ is the fraction of the surface covered by oxygen. The factor n, which was constant for the electrode oxidized under a given condition, was assumed to be the number of Pt sites deactivated by each one of the chemisorbed oxygen atoms. For the transiently formed oxide, n was estimated to be 1·64. It was also observed that all over the coverage range the reaction rate was proportional to the partial pressure of H₂. The variation in reduction rate with the decrease of the coverage was interpreted in terms of change in reduction mechanism from the chemical reaction to the electrochemical reaction.     

Surface structures in ammonia synthesis

Ammonia synthesis is one of the most structure sensitive catalytic reactions. Reaction studies using single crystals showed the open (111) and (211) crystal faces of iron and the and crystal faces of rhenium to be most active, while the close packed iron (110) and rhenium (0001) crystal faces were almost inactive. These studies suggest that seven (C₇) and eight (C₈) metal atom coordinated surface sites, which are available only on the active surfaces, are very active for the dissociation of dinitrogen, the limiting factor in the reaction rate under most experimental circumstances. In this paper, the experimental evidence for the existence of the C₇ sites in iron is reviewed. In addition, the role of potassium in creating a different active site which is less sensitive to the iron surface structure is discussed. Newly developed surface science techniques should permit investigations into the dissociation of dinitrogen at the C₇ sites and how the resulting chemisorbed nitrogen atoms are removed to allow for reaction turnover. Advances in LEED-surface crystallography, allowing detailed determination of relaxation in the clean metal surfaces and adsorbate induced restructuring of the metal surface, reopen the question of the real structure of the active sites in the presence of atomic nitrogen, or atomic nitrogen coadsorbed with potassium and oxygen. Investigation of the dynamics of surface restructuring involving the movements of both the substrate metal atoms and the chemisorbed atoms by surface diffusion becomes feasible by the availability of the high pressure/high temperature STM system built in our laboratory. Studies of the surface structures of the model iron catalysts under dynamic conditions, using 0.1 ms time resolution and atomic spatial resolution under reaction conditions are now possible.     

Theoretical study of halogen-substituted benzene at a Si(111)7×7 surface

The adsorption of halobenzene (for halogens F, Cl, Br, and I) on Si(111)7×7 was investigated using AM1 quantum mechanical calculations. First the 1,4-cyclohexadiene type of chemisorbed structures with two C—Si bonds at C atoms 1 and 4 have been calculated. Generally, the calculated binding energy increases with the size of the halogen atom, in the series F, Cl, Br, and I. The sp² carbon positions are the most favorable for the halogen atom, and for bromobenzene and iodobenzene there is significant additional stabilization for the geometry that allows interaction of the halogen atom with a nearby Si adatom. This stabilization hinders the transfer of the heavier halogen atoms, Br and I, to the surface. Other chemisorbed structures, involving formally divalent halogen atoms in a C-X-Si type of binding, have also been found to correspond to energy minima in the AM1 calculations. Furthermore, it was possible to calculate physisorbed structures for chlorobenzene and bromobenzene once the Si rest atoms were capped with H atoms. Two types of binding structures are suggested for chemisorption in excess of three molecules per half-unit cell: one is a radical structure binding by a single C-Si adatom bond, and the second one is binding by a single C-X-Si adatom interaction. Both of these singly-bonded structures have been calculated to correspond to energy minima with binding energies smaller than the 1,4-cyclohexadiene type of chemisorbed structure.     

Oxygen States at Magnesium and Copper Surfaces Revealed by Scanning Tunneling Microscopy and Surface Reactivity

By a combination of STM and XPS a study of the dynamics of oxygen chemisorption at Mg(0001) at 295?K has revealed oxygen states involving nucleation sites and the development of hexagonal and square lattice structures; the hexagonal structures develop epitaxially with the Mg(0001) surface. There is extensive surface mobility with, at the early stage of chemisorption, oxygen states being observed at steps at the (1×1)-O adlayer and overlapping magnesium atoms. These are active in ammonia and hydrocarbon oxidation whereas at higher oxygen coverage the surface is inactive. The dissociative chemisorption of nitric oxide generates a surface characterized by the hexagonal oxide structure; nitrogen adatoms known to be present from the N(Is) spectra are disordered. The chemisorption of hydrogen chloride at a Cu(110) surface results in a c(2×2)–Cl structure with at high coverage domains 1.8 nm wide. A sub-surface oxygen state at Cu(110), although unreactive to ammonia, undergoes a chemisorption replacement (corrosive chemisorption) reaction with HCl at 295 K.     

Potassium adsorption on hydrogen- and oxygen-terminated diamond(100) surfaces

The adsorption of K at 298 K is a route for the titration of surface groups like O and OH. The experiments were performed on a semi-conducting natural diamond (100) surface cleaned by a 480-W microwave hydrogen plasma at 750°C and a hydrogen pressure of 20 mbar, resulting in a very clean, ordered (2×1) surface. A second type of C(100) surface was prepared by ex situ oxidation, using a mixture of hydrochloric and nitric acids. The adsorption experiments were carried out in a UHV system equipped with facilities for photoelectron spectroscopy (XPS and UPS). K deposition was achieved using a dispenser source from the SAES Getters Company. The K uptake of H-terminated C(100) at room temperature is marginal, with a sticking coefficient of <0.03. The acid-treated C(100) surface shows the presence of a broad O 1s spectrum with different oxygen states (different binding energies) in the XPS region. The sticking coefficient for K adsorption at 298 K on this surface is nearly 1. The amount of K accommodated on the surface correlates with the oxygen coverage and a stoichiometry of nearly K/O=1 is reached for the saturated K coverage. In the O 1s spectrum, the high-energy state (534.2 eV) disappears, whereas the overall O 1s intensity is constant. Due to the K adsorption, the C 1s peak shifts by 0.5 eV to a higher binding energy. UPS He I spectra demonstrate a lowering of the work function by −1.9 eV.     

A reactive oxygen state at a barium promoted Au (100) surface: the oxidation of ethene at cryogenic temperatures

Although Au (100) does not adsorb oxygen at either 295 K or 80 K, a barium modified Au(100) surface is active in oxygen dissociation resulting, through surface diffusion of oxygen adatoms, in the formation of a chemisorbed oxygen adlayer. This oxygen species is inactive for ethene oxidation, as is the oxygen species pre-adsorbed at an Au(100)–Ba surface at 80 K, and the clean Au(100)–Ba surface. However, when molecularly adsorbed ethene present at a Au(100)–Ba surface at 80 K is exposed to dioxygen and warmed to 140 K, surface carbonate is observed. We conclude that a transient oxygen species is the oxidant.     

Molecular Metal Clusters as Catalysts

Abstract

In earlier analyses [1–8] we have established a correlation between metal clusters and metal surfaces with chemisorbed molecules in the specific contexts of (1) the metal frameworks wherein the metal cluster core structures are fragments of cubic and hexagonal close packed or body centered cubic metal bulk structures; (2) ligand stereochemistry where the geometric features of ligands bound to clusters and to metal surfaces are similar; (3) thermodynamic features where the average bond energies for ligand-metal and metal-metal bonds are comparable, for a specific metal, in the metal cluster and the metal surface regime; and (4) mobility of ligands bonded to metal cluster frameworks and to metal surfaces. Nevertheless, there are sharp distinctions between surfaces and clusters. The average coordination numbers for metal-metal interactions and for metal-ligand interactions are distinctly different for clusters and for surfaces: generally, the former are larger for surfaces and the latter are larger for clusters. Additionally, the surface state is typically differentiated from the cluster state in the degree of coordination saturation—the metal atoms in the surface state are typically less coordinately saturated even for the states in which molecules or molecular fragments are chemisorbed at the surface than those metal atoms at the periphery of a molecular metal cluster. In the crucial chemical issue, metal surfaces are far more reactive than metal clusters. Metal surfaces exhibit a wide range and high level of catalytic activity whereas most metal clusters are catalytically inert, at least under modest reaction conditions, Most reported clusters are relatively stable and nonreactive; they are not the products of sophisticated synthesis procedures designed to generate highly reactive metal clusters. They commonly have been the products of reaction mixtures run at forcing conditions and are thermodynamically controlled, not kinetically controlled, products.     

Curing and thermal behavior of epoxy resin in the presence of silicon‐containing amide amines

The article describes the synthesis and characterization of silicon‐containing amide amines obtained by the reaction of bis(4‐chlorobenzoyl)dimethylsilane with 4,4′‐diaminodiphenyl ether, 4,4′‐diaminodiphenyl methane, 4,4′‐diaminodiphenyl sulfone/3,3′‐diaminodiphenyl sulfone, bis(3‐aminophenyl)methyl phosphine oxide, and tris(3‐aminophenyl)phosphine oxide with dimethyl acetamide as a solvent. Structural characterization of amide amines was done with Fourier transform infrared and ¹H‐NMR spectroscopy. We used these aromatic amide amines as curing agents to investigate the effect of structure and molecular size on the curing and thermal behavior of diglycidyl ether of bisphenol A (DGEBA). The curing behavior of DGEBA in the presence of stoichiometric amounts of silicon‐containing aromatic amide amines was investigated by differential canning calorimetry. A broad exothermic transition in the temperature range of 200–300°C was observed in all the samples. The peak exotherm temperature was lowest in the case of phosphorus‐containing amides and was highest in the case of ether‐containing amides. Thermal stability of the isothermally cured resins was evaluated with dynamic thermogravimetry in a nitrogen atmosphere. A significant improvement in the char yield was observed with silicon‐containing amines, and it was highest in case of samples with both silicon and phosphorus as flame‐retarding elements. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 87: 1345–1353, 2003     

Selective Metal Pattern Fabrication Through Micro-Contact or Ink-Jet Printing and Electroless Plating onto Polymer Surfaces Chemically Modified by Plasma Treatments

Simple versatile processes combining plasma treatments, micro-contact printing (µCP) or ink-jet printing (IJP), and electroless deposition (ELD) have been developed to produce micrometer and sub-micrometer scale metal (Ni, Ag) patterns at the surface of polymer substrates. Plasma treatments were mainly used to graft the substrate surfaces with either nitrogen-containing functionalities on which a palladium-based catalyst can be subsequently chemisorbed (case of Ni deposition through a tin-free process in solution) or oxygen-containing functionalities on which a tin-based sensitization agent can be subsequently chemisorbed (case of Ag deposition through a redox reaction). µCP of the catalyst or of self-assembled monolayers (SAMs) as well as ink printing were used to obtain locally active or non-active areas at the polymer surfaces. The metal micro-patterns were characterized using optical microscopy, scanning electron microscopy (SEM), and atomic force microscopy (AFM). Surface chemical characterization was carried out by X-ray photoelectron spectroscopy (XPS).     

Low temperature aminolysis of methyl stearate catalyzed by sodium methoxide

Aminolysis of methyl stearate by both primary and secondary amine catalyzed by sodium methoxide was found to be rapid at 30°C. under anhydrous conditions. With primary amines under optimum conditions (mole ratio to ester: amine, 10; catalyst, 0.12), the minimum reaction times necessary to obtain yields of amide over 90% were:n-butylamine, 30 min.;iso-butyl-, 1 hour; allyl-, 1.8 hr.; benzyl-, 3.2 hr.;sec-butyl-, 16 hr.; ammonia (a heterogenous reaction requiring an optimum triethylamine to ester ratio of 2 ml./g. and a catalyst mole ratio of 0.20) 2 days. Secondary amines reacted rapidly at 30°C. (15 min. to 24 hr. for a 90% yield of amide) when the nitrogen atom was joined into a saturated ring or held at least one methyl group, but very slowly even at 100°C, when the substituent was dialkyl larger than methyl. Uncatalyzed, all reactions were extremely slow.     

Recent studies on diamond surfaces

The surface properties of diamond have been studied by ultra-violet photoemission spectroscopy (UPS), Kelvin probing and low energy electron diffraction (LEED). The atomic level structure of diamond surfaces was determined by LEED intensity vs. energy [I(E)] measurements in combination with Tensor LEED calculations. The LEED analysis of the C(100)-(2×1)-H surface revealed the formation of symmetrical dimers on the top carbon layer. For the C(100)-(1×1)-O surface, quantitative LEED analysis indicated a structural model where oxygen occupied the bridge site on the surface. Systematic investigations were carried out using UPS and a Kelvin probe measurement to reveal the effect of alkali metal fluoride overlayers on the work function of the diamond surfaces. LiF and RbF have been found to act as effective dipole layers to lower the surface work function and induce a negative electron affinity.