Search

In a combined photoelectron-vibrational spectroscopic study of the chemistry of formic acid interaction with A1(111) surfaces the influence of temperature, surface oxidation and coadsorption with water have been investigated. At the clean surface adsorption at low temperatures leads to deprotonation and on warming to room temperature the major species is formate. On the other hand direct interaction with A1(111) at room temperature leads to more extensive bond cleavage including the formation of chemisorbed oxygen, and carbidic carbon. At the preoxidized A1(111) surface deprotonation is facile at 80 K but there is no evidence for further reaction on warning to room temperature. When formic acid is coadsorbed with water at 295 K, HREEL and XP spectra confirm the formation of surface formate together with hydroxyl and chemisorbed oxygen species.

WOS: A1994MT29700010 A facile oxygen-induced chemisorptive replacement reaction occurs when a Bi(0001)-O overlayer is exposed to hydrogen chloride at 298K. The overlayer, which conforms to the stoichiometry BiOCl2, is stable in contrast to analogous chemistry observed earlier with a Pb(110) surface. Evidence for discrete localized states associated with Bi2+ or Bi3+ species is observed from shifts in the Bi(4f) binding energies while the charge distribution within the bismuth-chlorine bond is shown to be sensitive to the presence of oxygen adatoms within the oxy-chloride overlayer.

The experimentally observed kinetics of the coadsorption of dioxygen and ammonia at a Zn(0001) surface have been modelled theoretically. The rate-limiting step involves a precursor dioxygen—ammonia surface complex, formed via ammonia and dioxygen surface diffusion. The system of differential equations describing the model provides a solution that is in good agreement with the experimental results when account is taken of the dependence of the enthalpy of ammonia adsorption on the coverage of chemisorbed oxygen O2–(a). Thermodynamic arguments are shown to account for the observed large increase (ca. 103) in the probability of dioxygen dissociation during coadsorption, compared with that observed with dioxygen alone. Electrostatic interactions within the precursor {NH3–O–2} zinc complex are suggested to provide a low-energy pathway to the observed chemisorbed products.

The application of X-ray photoelectron spectroscopy to the study of the surface chemistry of wool fibres clarifies some aspects of the role of the cystine disulphide bond. The sensitivity of the electron-binding energy of the S(2p) orbitals to the sulphur oxidation state provides information about the extent of surface modification following two different oxidizing treatments using (a) dichloroisocyanuric acid, and (b) hypochlorous acid (‘Kroy’ chlorination). Interaction of a wool surface with a polymer, Hercosett 125, which is the basis of a shrink-resistant treatment, shows that complete coverage of the wool-fibre surface does not occur even though the average thickness of the polymer overlayer is about 1000 A (100 nm).

The interaction of formic acid with polycrystalline copper, nickel and gold surfaces (and with a Cu(100) single crystal) in the temperature range 80‒500K has been studied by photoelectron spectroscopy. Between 80 and 150K formic acid forms multilayers and the ultraviolet photoelectron spectrum (u. p. s.) is rather similar to that of formic acid vapour. There are, however, small but significant differences in the u. p. s. and also in the X-ray induced photoelectron spectrum (X. p. s.). These differences are shown to be a consequence of hydrogen bonding in the condensed phase. Changes in the spectra with increasing temperature are interpreted as reflecting molecular transformation to the formate ion. This is shown to involve the following sequence of molecular events:( a ) hydrogen bond cleavage in the condensed phase, ( b ) reorientation of the molecule from being parallel to being perpendicular to the plane of the surface and ( c ) removal of a proton. These conclusions are based on the observed surface carbon to oxygen ratio, estimated from the O(1s) and C(1s) spectra, O(1s) f. w. h. m. values at different temperatures, the energies of the peaks in the He I and He II spectra and comparison of these with theoretical cal­culations of the energies of the orbitals of the formate ion and the formic acid molecule. Spectroscopic monitoring of the surface species showed that with gold the formate ion decomposes into gaseous products between 180 and 200K; in the case of nickel decomposition occurs slowly in vacuo at 295K but with copper, the surface form ate ion does not break down until above 400K. The following activation energies have been estimated for decom­position of the surface anion: Au (50kJ mol ‒1 ), Ni (90kJ mol ‒1 ) and Cu ( ca . 120kJ mol ‒1 ). The density of states near the Fermi level has been studied and there is no evidence for the participation of an electronic factor in the catalytic decomposition process. In a limited l. e. e. d. study with a Cu(100) surface no evidence for ordered structures of either the adsorbed formic acid molecule or the anion was obtained. We believe this is the first investigation of a heterogeneously catalysed reaction by both u. v. and X-ray electron spectroscopy, and the results are compared with previous infrared and catalytic studies.

The chemisorption of oxygen at Ni(100) and Ni(210) surfaces has been studied by X -ray photoelectron spectroscopy in the temperature range 77-300 K. O(1s) spectra distinguished between the formation of O - and O 2- species while Ni(2p) spectra revealed the formation of Ni 2+ and Ni 3+ states. The binding energies of the latter are 854.7 and 856 eV respectively. These values were established by the generation of difference spectra in the Ni(2p) spectral region and assignments confirmed by studies of bulk nickel oxides of different stoichiometry. There were two distinct régimes of oxygen interaction; for low oxygen coverages, θ < 0.6, there was no evidence for the higher oxidation states (Ni 2+ or Ni 3+ ). For θ > 0.6 the Ni(2p) difference spectra indicated that both states were formed, their relative proportions being dependent on temperature and coverage. It is suggested that the initial stage of oxygen chemisorption involves only the Ni(4s4p) band while the Ni(3d) electrons participate in oxide formation, which is reflected by the emergence of Ni 2+ and Ni 3+ states. Ni (L 2, 3 M 4, 5 M 4, 5 ) Auger spectra and also recent theoretical work support this conclusion. The mixed valence states, Ni 2+ and Ni 3+ , are a characteristic feature not only of oxygen interaction with Ni(210) and Ni(100) at low temperature but also of the surface of bulk nickel oxides. The latter unexpectedly exhibit a high relative proportion of Ni 3+ states. It is suggested that Ni 3+ is formed in a surface redox reaction from Ni 2+ with the simultaneous generation of O - . Angular dependent studies confirm that the Ni 3+ and O - species are at the surface relative to Ni 2+ and O 2- .

The techniques of X-ray and u. v. -photoelectron spectroscopy have been used to investigate the chemisorption of the individual molecules nitric oxide, oxygen, and water, and also of coadsorbed nitric oxide and water, at Zn(0001) surfaces. The core and valence level spectra established that at 77 K, nitric oxide existed in both dissociative and molecular states, oxygen generated both surface O¯ (a) and subsurface O 2- (b) species, while water was only physically adsorbed. At 160 K and above, Zn(0001) surfaces were unreactive to water vapour, while at 295 K only the dissociative state of nitric oxide was present. The coadsorbed water-nitric oxide adlayer was reactive below 200 K. Surface hydroxyls were formed through the activation of molecularly adsorbed water by chemisorbed oxygen generated in the dissociative chemisorption of nitric oxide. These surface hydroxyls participated in a hydrogen transfer reaction leading to hydrogenation of the chemisorbed nitrogen at 273 K followed by desorption. In the absence of water, chemisorbed nitrogen adatoms are by comparison thermally stable. A detailed mechanism, with each surface species spectroscopically identified, is proposed for the hydrogenation of nitric oxide. Analysis of the X-ray induced core level spectra enabled the concentrations of the surface species to be calculated.

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.

The interaction of oxygen, water vapor, and carbon monoxide with cobalt has been studied by combined in situ X-ray and uv photoelectron spectroscopy. Oxygen interaction gave rise to CoO at 295 K at low pressure (5 × 10 −8 Torr) but to Co 3 O 4 at high pressure (0.25 Torr). This was deduced by studies of the bulk oxides CoO and Co 3 O 4 . O(1s) binding energies were not useful in distinguishing between the two oxides but the shake-up satellites associated with the Co(2p) spectra were diagnostic. Co 3 O 4 could be converted to CoO by heating to 700 K in vacuo which is in agreement with thermodynamic calculations. Water vapor was less reactive than O 2 (g) at 295 K and gave rise to a hydroxylated surface. Carbon monoxide adsorbed molecularly at both 80 and 295 K showing both core-level and valence-level spectra characteristic of the associative state. This is in accord with predictions based on the known heat of chemisorption.

The factors relevant to the design of a flow-system for determining the sticking probabilities of gases on film surfaces are discussed critically. The sticking probability of nitrogen on tungsten films is 0.64 at 298°K and 0.94 at 77° and 130°K. These values, which are higher than those observed with tungsten ribbons, are in quantitative agreement with calculations based on a model for film surfaces suggested by Takaishi. The variation of the number of molecules adsorbed as a function of film weight has been analyzed in terms of equations based on a model for film growth and data determined previously with nickel films.

The kinetics of the reaction of nitrogen with sintered, evaporated calcium films has been studied between 23 and 200°C using a constant-pressure technique. The results are consistent with a modified Mott & Cabrera theory of thin film formation, the rate being controlled by the mobility of the calcium ions through the nitride lattice under the influence of the potential difference set up between the metal and chemisorbed nitrogen at the nitride surface. The rate depends directly on the first power of the pressure suggesting a reversible adsorption to form a ‘surface complex’, N - 2 . The linear dependence of activation energy on the inverse thickness of the nitride layer is verified and the critical temperature determined. In addition, an estimation of the fraction of the total number of metal sites per unit area interface from which the metal ion penetrates the nitride layer has been made.

The possibility of analyzing the complex surfaces of reduced metals by comparing their adsorptive properties with the known behaviour of evaporated films has been explored by experiments on nickel powder. Nickel prepared by reduction of the oxide in hydrogen at 450°C for periods up to 12h has been shown by measurement of hydrogen and krypton adsorption at -183°C to have a surface containing up to 37% of clean nickel atoms. Traces of silica introduced during the preparation of the powder possibly account for the rest of the surface. Adsorption of hydrogen at -183°C is effectively complete within ½ min at about 10 -5 mm, and all gas is desorbed by evacuation at 450°C. Oxide is probably not a significant impurity, because the powder does not exhibit the slow adsorption at 20°C and above which is characteristic of incompletely reduced nickel. Such activated adsorption can be restored by the addition of oxygen and eliminated again by further reduction; its stoichiometry suggests that the process consists of hydroxyl formation. Reduction of the sulphide gives nickel of appreciable surface area but completely inert to hydrogen; presumably a stable layer of sulphide is retained. Some thermodynamic and kinetic problems involved in the preparation of metal powders with surfaces of known purity are briefly discussed in the light of these results.

The interaction of hydrogen sulfide and sulfur vapor with tungsten has been studied by field-emission microscopy. Adsorption of hydrogen sulfide at 22 °C begins along the [111] zones and at a pressure of about 5 × 10 −8 mm the adsorption appears to be complete within 4 min. An increase in the average work function of 0.3 volt has been estimated for the process. On heating a surface saturated with hydrogen sulfide, surface migration begins at about 1150 °C, leading to a concentration of sulfur along the [111] zones from which desorption takes place, i.e., the desorption processes are the reverse of the adsorption sequence. The activation energy for desorption is about 80 to 90 kcal/mole and complete removal of sulfur is observed at about 1600 °C. Sulfur adsorption at 10 −7 mm and 22 °C results in a large number of brightly emitting spots which move continually. These are mainly observed around the (111) and (011) planes. A field-induced process resulting in a sudden enlargement of the emitting area and consequent mechanical failure of the tip occurs at 22 °C. This is attributed to the growth of a sulfide whisker. Presorbed hydrogen has an inhibiting effect on this nucleation phenomenon.

The chemisorption of hydrogen, carbon monoxide, and nitrogen has been investigated on molybdenum films. Evidence is obtained for carbon monoxide being bonded differently on a nitrogen-covered surface than on a clean surface. Surface potential and other data gave no indication of any surface complexes involving nitrogen and hydrogen. Presorbed hydrogen and nitrogen had little influence on the krypton monolayer while carbon monoxide reduced it substantially. Presorbed hydrogen was desorbed by adsorbing nitrogen under certain conditions of hydrogen coverage and temperature. The kinetics of hydrogen adatom recombination and subsequent desorption have been investigated and the specific nature of nitrogen chemisorption emphasized. The activation energy of hydrogen desorption (~20 kcal mole −1 ) is close to the heat of hydrogen adsorption at the same coverage. On the basis of a two-dimensional gas model the molecular desorption process is shown to be very inefficient.

It has been established1–3 that oxygen chemisorbed at a metal surface can activate an otherwise unreactive molecule, as in the activation of S–H, N–H, O–H and C–H bonds. We have now extended these studies to more complex systems involving coadsorbed molecules in the hope of throwing some light on the mechanisms of heterogeneous catalysts. We have previously established the mechanisms by which unreactive molecular water, adsorbed on a Zn(0001) surface, is activated when co-adsorbed with nitric oxide4, and have demonstrated N–H activation in ammonia co-adsorbed with nitric oxide at an atomically clean Mg(0001) surface at 295 K (Fig. la and ref. 5). Activation was attributed to surface oxygen generated in the dissociative chemisorption of nitric oxide. Here we establish the nature of the ‘oxygen’ species involved and reveal details of the molecular steps.

Nitric oxide is dissociatively chemisorbed at a Zn(0001)surface at 80 K, both the chemisorbed nitrogen and oxygen are stable up to 470K; in the presence of coadsorbed water, hydrogenation of the adsorbed nitrogen occurs at 200K by hydroxyl species formed as a result of the activation of molecularly adsorbed water by chemisorbed oxygen generated in the dissociative chemisorption of NO.

The formation of a strongly chemisorbed copper carbamate complex, tilted with respect to the surface, is shown to occur during the coadsorption of CO2 and NH3 at a Cu(100) surface at low temperatures; a mechanism is suggested involving the nucleophilic attack of ammonia at the positively polarised carbon atom of a transient CO2–(s) species.

The adsorption and subsequent decomposition of tetramethylsilane Si(CH3)4 on evaporated films of tungsten and iron have been investigated. The results are compared with results previously obtained with neopentane C(CH3)4. Si(CH3)4 is rapidly and irreversibly adsorbed on tungsten at 293 K, and gaseous hydrogen and methane are formed; further H2 and CH4 are desorbed on heating the surface to temperatures up to 420 K. Additional information regarding the adsorbed phase has been obtained from deuterium exchange experiments. Similar experiments carried out with iron show that much less dissociation of the Si(CH3)4 occurs on adsorption and on heating the adsorbed layer; no gaseous products are observed below 320 K. Comparison of both sets of results with those for C(CH3)4 indicates that Si(CH3)4 is more extensively dissociated under all conditions; this is attributed mainly to the weaker C—Si bond. The results suggest that the rate determining step in hydrocracking reactions of hydrocarbons is likely to be hydrocarbon breakdown rather than desorption of products.

Atomic hydrogen has been shown, in contrast to molecular hydrogen, to interact with lead films to an extent corresponding to about five atoms per surface lead atom at 78 K. At 273 K incorporation is substantially faster and more extensive. There is some evidence for the formation at 273 K of a hydride of fixed composition (PbH0·2) which is active in the adsorption of molecular hydrogen at 78 K. The kinetics of the formation and decomposition of this hydride have been investigated; gas-phase deactivation of thermally activated molecular species occurs at hydrogen pressures 10–2 mmHg.

Condensed multilayers of PbEt4 decompose when irradiated with u.v. light of wavelength 254 nm to give gaseous C2 and C4 hydrocarbons and a lead film; adsorbing PbEt4 on a photodeposit of Pb generates a more reactive species which is active in photolysis at longer wavelengths (280–290 nm).

Oxygen chemisorbed on nickel can be incorporated into the bulk, the process is activated and sensitive to the initial oxygen coverage. The activation energy increases from about 2 to ∼15 kcal mole–1as the surface is depleted of chemisorbed oxygen and at 150° the work function is ∼1eV less than the clean metal. The chemical reactivity of surfaces which had incorporated oxygen to carbon monoxide and hydrogen confirmed that the low work function surface was essentially oxide and not metallic. Oxygen interaction involves competition between oxygen chemisorption and incorporation to form an oxide which may be of high or low work function depending on the temperature and the conditions at the surface. These competitive processes and the ability of chemisorbed oxygen to act as an “anchor” thereby restricting the mobility of nickel and inhibiting the formation of the low work function oxide are suggested to account for the variable work function values observed in the Ni+O2system. Our results and conclusions are discussed with relation to recent L.E.E.D. and photoelectric data. The interaction of hydrogen with NiO is complex at 150° two distinct processes were isolated, a fast initial process leading to a decrease in work function of ∼2 eV followed by a somewhat slower process resulting in an increase of work function.

Hydrogen sulphide is chemisorbed dissociatively at –80°C attaining a maximum coverage of θs≃1.6. The hydrogen adatoms at –80°C are metastable with respect to H2(g) for θs values 0.7. Above 0°C, incorporation of sulphur into the metal occurs and the kinetics of dissociation conform to the following equation: –(dPH2S//dt)θs, T=AθsPH2S//(2πmkT)½exp (–Eθs//RT) The condensation coefficient Aθs and activation energy Eθs increase from 10–7 to 101 and from 5 to 20 kcal mole–1 respectively as θs increases from 1.5 to 4.0. A linear relationship exists between log Aθs and Eθs. The slow-step is considered to be dissociation at the sulphide surface and the variation in Aθs is interpreted in terms of the absolute rate theory where the transition state complex gains mobility during sulphidation. The entropy of activation ΔS≠ increased from +5 cal mole–1 deg.–1≃ 1.5 to +30 cal mole–1 deg.–1 at θs≃4.0.

The interaction of carbon monoxide, hydrogen, and oxygen with manganese films has been investigated. Hydrogen is weakly chemisorbed (ΔH≃ 50 kJ mol–1), but both carbon monoxide and oxygen are incorporated into the lattice. The oxide formed is likely to be MnO. Adsorption data suggest that oxygen incorporation occurs even at 78 K and this is confirmed by work function studies; the surface potential of oxygen on manganese at 78 K is ca.–1·6 V. Carbon monoxide is incorporated extensively above 400 K and the kinetics suggest a diffusion-controlled process involving weakly chemisorbed CO. The catalytic oxidation of CO over manganese was first order in CO and only occurred under conditions where oxygen incorporation was slow compared with the catalytic reaction. A surface which had been oxidised to its maximum extent was inactive in catalysis at the temperature of oxidation.

The role of oxygen in determining the surface characteristics of nickel catalysts has been investigated by work function, photoelectric and chemical reactivity studies. The possible limitations of any one particular method are discussed. Surfaces which have been exposed to oxygen (10–4 mm min) at 23° exhibit the characteristics expected of a surface oxide although the chemical reactivity is not that associated with NiO. Oxide structures of low work function can form, and these were shown to have specific adsorption characteristics.The adsorption of carbon monoxide and oxygen and the catalytic oxidation of carbon monoxide by heavily oxidized nickel films were studied. Two distinct states of adsorbed oxygen were recognized. A substantial decrease in work function occurred on adsorbing carbon monoxide, the decrease being greater the higher the temperature. At 170° and 1.5 mm the decrease was ∼0.85 eV the major portion of this arising from a space charge term. The oxide work function remained virtually unchanged during catalysts at 23°. This was due to a low energy state of adsorbed oxygen inhibiting the adsorption of carbon monoxide thereby preventing the development of a spacecharge.

Thermal transpiration data have been determined for xenon and nitrous oxide at high temperature and low pressure. The results are shown to be in accord with the predictions of Miller's equation.