Low CO Coverage Research Articles

Effects of K, O, and H adatoms on the adsorption kinetics of CO on Pt(111)

The sticking coefficient, S, of CO on clean and K-, O- and H-modified Pt(111) surfaces has been measured directly using a simple kinetic uptake method, as a function of CO and modifier coverage and surface temperature. A simple method is also introduced to determine the coverage-dependent desorption energy from these measurements without the need to know the pre-exponential factor or desorption order for those systems where S is only a function of coverage or only weakly dependent on temperature. The initial sticking coefficient of CO on clean Pt(111) is 0.9 and is independent of substrate temperature. The major influence of adsorbed K on the initial sticking coefficient of CO on Pt(111) at 320 K is site blocking, and the magnitude of this effect correlates well with the size of the adsorbed K ion or atom. The shape of S as a function of CO coverage for low K coverages in the ionic regime is similar to that of CO on clean Pt(111). A dramatic change occurs for K coverages near or greater than 0.24 ML, where S passes through a maximum at about case1 3 of the saturation CO coverage and coincident CO and K desorption is observed. This behavior of S, which has not been seen before, can be fit quite adequately by a simple model that incorporates the reduced site blocking effects of K due to the CO-induced transition of K from a metallic neutral to an ionic state and the accompanying size change of K atoms. These results are best described by a coverage-dependent alkali-CO interaction. In additional studies, we find that adsorption of 0.25 ML O on Pt(111) has no effect on S for low CO coverages at 100 K, and only the saturation CO coverage is reduced from 0.59 to 0.31 ML. In contrast, adsorbed H reduces not only the CO saturation coverage but also S even at low CO coverages.

The adsorption and dissociation of carbon monoxide on clean and oxygen-modified molybdenum(110) surfaces

The adsorption and the dissociation of CO on clean and oxygen-modified Mo(110) surfaces have been investigated using high-resolution electron energy loss spectroscopy (EELS) and thermal desorption mass spectrometry (TDS). An inclined CO species with a v(CO) of 1345 cm -1 has been observed as the only adsorbed species on Mo(110) at low CO coverages at 120 K. It can either be converted completely to atomic carbon and oxygen through a dissociation channel, via a stable intermediate with a u(CO) of 1130 cm -1

One-dimensional CO island formation in the coadsorption of H 2 and CO on the steps of Pt(112)

The interaction of coadsorbed CO and H 2 on the steps of Pt(112) [Pt(S)[3(111)×(001)]] was examined by digital electron-stimulated desorption ion angular distribution (ESDIAD), low-energy electron diffraction (LEED) and temperature-programmed desorption (TPD). On the clean surface, CO and H 2 both adsorb preferentially on step sites at low coverages; CO at stop sites on the step edge and H 2, as atoms, in the step 4-fold sites. For pure CO adsorption, an ordered (2× n) structure of step CO is observed at 0.19 ML (atop step sites half filled), and at least two CO structural transformations occur as the step sites fill further. These structural transformations, due to COCO repulsions, involve the production of tilted CO species whose tilt directions are related to the local CO coverage in one-dimensional chains of CO on the step edges. The structural transformations described above for pure CO layers on the step edges can also be produced by coadsorption of hydrogen on the (2× n) step CO structure. The CO structural transformations in the coadsorbed system are driven by H adsorption, due to the formation of immiscible one-dimensional CO and H islands at the steps. Similar results are observed for low coverages of CO adsorbed on the H atom saturated surface.

Site blocking by hydrogen: CO on clean and H-presaturated Fe(100)

Because CO has been observed to facilitate transfer of adsorbed hydrogen to surface-found ethyl groups on Fe(100) and to weaken the FeH bond, we have studied CO binding on hydrogen-presaturated Fe(100) surfaces with electron energy-loss spectroscopy (EELS) to identify differences in CO binding on the clean and hydrogen-saturated surface. The dominant binding site is shown to be quite different for the hydrogen-presaturated surface. We reproduce the CO(α 3) desorption state on the clean surface reported by others, representing CO bound in four-fold hollows, which exhibits an unusually low stretching frequency of 1140 cm −1. On the clean surface this state dominates at low CO coverage and remains as bridging and atop CO states are populated. The hydrogen-presaturated surface clearly blocks binding of CO in this state; the CO(γ) stretching frequency observed is characteristic of bridging and atop binding. Conversion to the CO(α 3) state occurs after heating to 343 K to desorb the hydrogen. These results show (i) that coadsorption of CO onto the hydrogen-saturated Fe(100) surface does not lead to phase separation into the phase expected for CO alone on Fe(100) and (ii) that site conversion of CO with hydrogen removal may contribute to the driving force for hydrogen transfer reactions.

The interaction of CO with Ni(111)-Fe surface alloys.

The interaction of CO with a Ni(111) surface alloyed with different amounts of iron is studied with ellipsometry and AES. For surfaces with an iron mole fraction of X Fe≤0.26 CO adsorbs only molecularly. The isosteric heat of adsorption is 106±5 kJ/mol and is independent of the CO coverage and X Fe. On surfaces with X Fe≥0.39 both dissociatively and molecularly adsorbed CO are observed. The carbon and oxygen coverages after CO dissociation are equal and increase with X Fe. On these surfaces with dissociatively adsorbed CO further reversible adsorption of molecular CO takes place. Small amounts of carbon and oxygen do not affect the isosteric heat of adsorption at low CO coverage. Because no dissociation of CO proceeds on surfaces with X Fe≤0.26, whereas on surfaces with X Fe≥0.39 it does, the dissociation requires an ensemble of iron atoms that has a minimum size of at least two atoms. CO does not cause segregation of iron to the surface.

Coadsorption of CO and methylamine on Ru(001): effect of coadsorbed CO on dissociation paths of methylamine

The effect of coadsorbed CO on dissociation paths of methylamine on Ru(001) has been examined by LEED, TPD and HREELS. It was found that the effect of CO considerably changes as a function of partial CO coverage. For methylamine alone on Ru(001), the C-H and N-H bonds are dissociated simultaneously at 280–320 K. In the presence of low-coverage (~ 0.3) CO molecules, an NH 2 species is formed persisting on the surface up to > 350 K, resulting in a relative enhancement of C-H bond scission. In the presence of high-coverage (0.5–0.68) CO, a well ordered (2 × 2) phase forms and partial dehydrogenation of methylamine takes place forming a species containing a CN double bond as a stable intermediate, followed by complete dissociation at higher temperatures.

Coadsorption of CO and methylamine on Ru(001): effects of coadsorbed CO on dissociation paths of methylamine

The effect of coadsorbed CO on dissociation paths of methylamine on Ru(001) has been examined by LEED, TPD and HREELS. It was found that the effect of CO considerably changes as a function of partial CO coverage. For methylamine alone on Ru(001), the C-H and N-H bonds are dissociated simultaneously at 280–320 K. In the presence of low-coverage (∼ 0.3) CO molecules, an NH 2 species is formed persisting on the surface up to > 350 K, resulting in a relative enhancement of C-H bond scission. In the presence of high-coverage (0.5–0.68) CO, a well ordered (2 × 2) phase forms and partial dehydrogenation of methylamine takes place forming a species containing a C=N double bond as a stabe intermediate, followed by complete dissociation at higher temperatures.

Effect of CO on hydrogen thermal desorption from Fe(100)

The adsorption and coadsorption of H and CO was studied in order to illuminate the enhanced reactivity of H toward olefin hydrogenation induced by coadsorbed CO. The hydrogen saturation coverage on clean Fe(100) is 1.0 ± 0.1 ML, and H2 desorbs in two states, β1 and β2, with activation energies of 17 and 23 kcalmol, respectively. Low coverages of CO adsorbed on hydrogen-presaturated Fe(100) broaden and reduce the peak temperatures of the β1 and β2 desorption states. The dominant effect of coadsorbed CO is to weaken Fe-H bonds, and three new H2 desorption states are induced below 260 K at higher coverages of CO. No displacement of H2 by coadsorption of CO is apparent. The CO saturation coverage is uneffected by preadsorbed hydrogen, but the CO desorption features below 350 K are distinctly different than those for the clean surface. Desorption of CO above 350 K resembles that for CO adsorbed on the H-free surface. When H2 is adsorbed onto a CO-precovered surface at 100 K. the saturation coverage of hydrogen is reduced, and at the CO saturation coverage of 0.58 ML hydrogen adsorption is blocked. H2 desorption from surfaces precovered with a subsaturated adlayer of CO adsorbed at 100 K display only perturbed β1 and β2 states. CO apparently induces low temperature desorption of H2 only when it is adsorbed on top of a locally saturated H adlayer.

The characterization of carbon monoxide electrosorbed on ordered rhodium(100) by in-situ infrared spectroscopy

Absolute surface infrared spectra for carbon monoxide on ordered Rh(100) in 0.1 M HClO 4 are reported as a function of CO coverage, 0.1 ≲ θ ≲ 0.77, and electrode potential, 0̄.25 V to 0.2 V vs. SCE. While vibrational bands corresponding to both terminal and twofold bridging adsorption geometries are observed, unlike platinum low-index faces and Rh(111), the latter bonding configuration predominates, especially at lower coverages. For CO adlayers prepared by direct dosing using a dilute CO solution to form a given coverage, the bridging v co peak. frequency, v co b, increases substantially (up to ca. 100 cm −1) with increasing θ. In contrast, for adlayers formed by electrooxidative removal from an initial saturated CO layer, v co b diminishes little (< 10 cm −1) with decreasing θ. The latter behavior indicates that large CO islands are formed during adsorbate electrooxidation, so that the large θ-induced v co b upshifts due to adsorbate dipole-dipole and surface electronic interactions, observed for dosed layers, remain virtually constant even to low average CO coverages, θ ~ 0.1. Comparisons are made with corresponding θ-dependent vibrational spectra obtained for CO on Rh (100) in ultrahigh vacuum, and the likely reasons for the greater preference for CO bridge bonding in the electrochemical system are discussed.

Coverage-dependent dipole coupling for carbon monoxide adsorbed at ordered platinum(111)-aqueous interfaces: Structural and electrochemical implications

Surface infrared spectra in the terminal C–O stretching region are reported for various 12CO/13CO mixtures adsorbed on an ordered Pt(111) electrode in 0.1 M HClO4 as a function of CO coverage θ (0.05≲θ≲0.6) and electrode potential E. Contributions to the coverage-induced shift in the C–O stretching frequency νtCO arising from dynamic dipole–dipole coupling ΔνD, and from static dipole and ‘‘chemical’’ effects ΔνC, are obtained from the mixed isotope spectra. Similar to the corresponding ultrahigh vacuum (UHV) system, the former provides the largest contribution, ΔνD being 30–35 cm−1 for saturated CO layers (θ≊0.6). Substantial differences in the νtCO -θ (and ΔνD -θ) dependencies were observed for CO adlayers formed by dosing from suitably low solution concentrations (∼2×10−5 M), and from saturated irreversibly adsorbed layers by means of partial electrooxidative stripping. Using the latter ‘‘stripping’’ procedure, only small (≲5 cm−1) decreases in νtCO and ΔνD were observed as θ is decreased over the range 0.1≲θ&amp;lt;0.6, whereas the former ‘‘dosing’’ procedure yielded substantially larger νtCO -θ and ΔνD -θ dependencies, especially in the presence of coadsorbed hydrogen (at −0.25 V vs SCE) rather than coadsorbed water (at 0.1 V). These behavioral differences are interpreted in terms of coverage-dependent CO island formation. For electrooxidative stripping conditions, extensive island formation survives to low CO coverages, providing direct evidence that CO electrooxidation occurs via nucleation and growth at the island edges. For dosing conditions, smaller and more coverage-dependent CO patches are apparently obtained in the presence of coadsorbed hydrogen. These undergo slightly more facile electrooxidation than the CO islands formed at intermediate coverages by prior stripping. The coverage-dependent frequencies and intensities of the bridging CO form (νbCO ≊1800–1850 cm−1) vs the terminal CO form (νtCO ≊2020–2075 cm−1) are also examined in comparison with corresponding data for related UHV systems. In contrast to Pt(111)/CO in UHV, the bridging site is occupied even at low dosed coverages (θ≳0.05): this is also suggestive of the presence of CO islands having high local coverages. Similarities are noted with vibrational spectra obtained for Pt(111)/CO with coadsorbed water and hydrogen in UHV.

Real-time observation of the nucleation and propagation of reaction fronts on surfaces using photoemission electron microscopy

The catalytic oxidation of CO on Pt{100} has been studied in situ using photoemission electron microscopy (PEEM). The difference in work function between CO-covered regions of the surface and O-covered regions provides image contrast in PEEM using a deuterium lamp. The CO and O2 partial pressures were adjusted to give a CO coverage greater than 0.5 monolayer at a crystal temperature below 500 K corresponding to the low rate branch of the oxidation reaction in which CO inhibits the adsorption of oxygen. The transition to the high rate branch (characterized by low CO coverage) was then initiated by decreasing the CO partial pressure. This allows oxygen adsorption at various types of defect sites, forming holes in the CO film. The holes subsequently enlarge as reaction fronts propagate away from the defects at velocities from 0.5 to over 40 μm s−1 depending on temperature, pressure and directions of steps on the surface. The return to the low rate branch, induced by increasing the CO pressure again, does not give rise to propagating reaction fronts but rather to random adsorption of CO throughout the surface.

Fourier-transform-infrared spectroscopy of carbon monoxide physisorbed on highly oriented graphite

Infrared spectra of carbon monoxide adsorbed on highly oriented pyrolytic graphite (HOPG) have been measured for the first time. It turned out from both theoretical considerations and measurements that highest sensitivity is achieved under almost grazing incidence with the light polarized parallel to the plane of incidence. Thus graphite behaves similar to a metal. At low coverage CO gives rise to a single absorption band slightly redshifted with respect to the corresponding gas frequency, the band width at full monolayer coverage being 0.8 cm −1. At higher coverage band splittings and band shifts occur. Isotherms measured by FT-IR-spectroscopy at temperatures below 35 K show significant steps indicating a two-dimensional phase transition. Preliminary results of FT-IR-investigations of nitric oxide adsorbed on HOPG are presented.

A FT-rairs investigation of the nature of the 3-fold bridge-CO species on Ni(111)

The adsorption of CO on Ni(111) in the low CO-coverage regime (θ CO ≲ 0.07 ML) has been investigated using FT-reflection absorption infrared spectroscopy. The FT-RAIRS results reported here provide a clear experimental evidence that at 85 K CO molecules occupy only the 3-fold bridge sites at low CO coverages. The population of the 2-fold bridge sites occurs only at higher CO coverages. In addition, the extended frequency range and the enhanced sensitivity of the FT-RAIR technique allows us to detect three types of 3-fold bridge v(CO) modes observed at 1815–1822, 1804–1808, and 1793–1799 cm −1, respectively. Possible origins of these v(CO) modes have also been postulated.

A FT-RAIRS investigation of the nature of the 3-fold bridge-CO species on Ni(111)

The adsorption of CO on Ni(111) in the low CO-coverage regime (θ CO ≲ 0.07 ML) has been investigated using FT-reflection absorption infrared spectroscopy. The FT-RAIRS results reported here provide a clear experimental evidence that at 85 K CO molecules occupy only the 3-fold bridge sites at low CO coverages. The population of the 2-fold bridge sites occurs only at higher CO coverages. In addition, the extended frequency range and the enhanced sensitivity of the FT-RAIR technique allows us to detect three types of 3-fold bridge v(CO) modes observed at 1815–1822, 1804–1808, and 1793–1799 cm -1, respectively. Possible origins of these v(CO) modes have also been postulated.

Adsorption and dissociation of CO on Fe(100) at low coverage

Abstract Adsorption and dissociation of low coverage CO on Fe(100) has been studied. The combination of high resolution electron energy loss spectroscopy (HREELS) and temperature programmed reaction spectroscopy (TPRS) has been used to monitor the CO dissociation process in detail. Adsorption of CO at low temperature (103 K) results in associative adsorption. The adsorption state is characterized by a CO stretching frequency around 1170 cm −1 . For CO coverage below 0.15 monolayer (ML), where 1 ML is defined as the number of top layer Fe atoms on the unreconstructed (100) surface, all adsorbed CO dissociates on the surface upon heating to high temperature (423 K). With CO coverage higher than 0.15 ML, direct desorption of adsorbed molecules takes place in competition with the dissociation process. Reducing the CO coverage and annealing the adsorbed CO overlayer both cause the CO stretching frequency to decrease. Dissociated CO is characterized by MO and MC vibrational peaks, located at 400 and 520 cm −1 respectively.

Adsorption and dissociation of CO on Fe(100) at low coverage

Adsorption and dissociation of low coverage CO on Fe(100) has been studied. The combination of high resolution electron energy loss spectroscopy (HREELS) and temperature programmed reaction spectroscopy (TPRS) has been used to monitor the CO dissociation process in detail. Adsorption of CO at low temperature (103 K) results in associative adsorption. The adsorption state is characterized by a CO stretching frequency around 1170 cm −1. For CO coverage below 0.15 monolayer (ML), where 1 ML is defined as the number of top layer Fe atoms on the unreconstructed (100) surface, all adsorbed CO dissociates on the surface upon heating to high temperature (423 K). With CO coverage higher than 0.15 ML, direct desorption of adsorbed molecules takes place in competition with the dissociation process. Reducing the CO coverage and annealing the adsorbed CO overlayer both cause the CO stretching frequency to decrease. Dissociated CO is characterized by MO and MC vibrational peaks, located at 400 and 520 cm −1 respectively.

Desorption kinetics on an alkali metal-precovered surface: CO and K on Pt(111)

The desorption kinetics of CO initially coadsorbed with K on Pt(111) at 100 K has been measured as a function of both CO and K coverage (θ) using temperature programmed desorption (TPD). CO and K desorb separately when the initial θCO is low, supporting the theory that CO and K interact via a strong Coulombic attraction and do not form an oxocarbon (salt-like) compound. Although precoverage with ≥0.15 ML K does induce a few percent of the desorbing CO to undergo atomic exchange, the effect is very small in comparison to that observed on other surfaces. The desorption is found to be first order in θCO on both clean and K-precovered surfaces. With low initial θCO the activation energy (E) and preexponential factor (ν) for desorption from the K-free surface are found to be 28.8±0.7 kcal mol−1 and 1013.3±0.2 s−1, respectively, in excellent agreement with previously reported results. As observed with most alkali metal-precovered surfaces, the low coverage CO desorption peak temperature increases monotonically with K coverage. We find that the higher desorption temperatures result from the combined effects of increasing E and simultaneously decreasing ν, with the change in ν causing most of the shift; when low θCO are initially coadsorbed with 0.15 ML K E=31.3±0.6 kcal mol−1 and ν=1011.4±0.2 s−1. For all surfaces studied both E and ν decreased approximately linearly with increasing θCO&amp;gt;0.1 ML. The θCO and θK dependence of the kinetics is generally consistent with the proposed structural model for this surface based on the formation of CO+K islands. Possible mechanisms for the K-induced change in the kinetics are discussed and further investigations are proposed.

CO diffusion on Pt(111) by time-resolved surface infrared spectroscopy

The mobility of CO chemisorbed on Pt(111) has been measured using a technique that synchronizes pulsed molecular-beam dosing with fast scanning infrared interferometry. The method is based on the observation that molecules are generally more tightly bound at defects (e.g., steps) where they exhibit a different internal vibrational frequency to that associated with the same molecule adsorbed on the low-index terrace site. On a stepped Pt(111) surface, a low coverage (θ∼0.006 monolayer) of CO is deposited in a random spatial distribution (in 600 μs). The time evolution of the infrared spectrum is then recorded with ≥5-ms resolution as the CO diffuses to the steps. A simple kinetic model makes it possible to extract the statistical microscopic hopping rates for CO diffusion on the (111) plane over the range of temperature from 95 to 195 K.

Carbon monoxide–oxygen interaction on the Pt(111) surface: An electron stimulated desorption ion angular distribution (ESDIAD) study

CO adsorption on the p(2×2)O–Pt(111) surface was studied by the digital ESDIAD (electron stimulated desorption ion angular distribution) method in combination with TPD, LEED and work function measurements. Three ESD products were detected: CO+, O+ and metastable CO. The ESDIAD patterns of each of these species were measured. The most significant difference in the ESD behavior of chemisorbed CO on the oxygen-covered surface from that of CO adsorbed on clean platinum surface was found at low CO coverages. This indicates that there is no preferential adsorption on the surface sites nonaffected by oxygen. A small tilting of CO was found.

AN ARUPS STUDY OF CO ON Fe(110) SURFACE

The chemisorption of CO on Fe(ll0) surface has been investigated using angle resolved ultraviolet photoemission spectroscopy (ARUPS)The experimental results, compared with the calculation, show that the molecular axis of Co is nearly perpendicular to the iron surface at 150K and low coverage of CO. The angle between CO axis and the surface normal is less than 10 degrees.