Adsorption and Reaction of Methanol over CeOX(100) Thin Films

Article pubs.acs.org/Langmuir

Adsorption and Reaction of Methanol over CeOX(100) Thin Films Peter M. Albrecht and David R. Mullins* Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6201, United States ABSTRACT: Methanol was adsorbed on oxidized and reduced CeOX(100) thin films to probe the active sites and reaction selectivity of these surfaces compared to those of CeOX(111). Roughly twice as much methoxy was formed on oxidized CeO2(100) compared to that formed on CeO2(111). In addition to more methoxy, hydroxyl is also more stable on CeO2(100). Unlike on CeO2(111), however, methanol on CeO2(100) produced CO, CO2, and H2 in addition to water and formaldehyde. The behavior of CeO2(100) is related to its surface structure, which provides greater access to Ce cations and therefore more active adsorption sites and more highly undercoordinated Ce and O. The undercoordinated O may explain the enhanced dehydrogenation activity leading to CO and H2 formation. The reduction of ceria leads to increased methanol uptake on both CeO2 − X(100) and CeO2 − X(111). However, although the uptake doubled on reduced CeO2 − X(111) compared to the oxidized surface, it increased by only 10% on reduced CeO2 − X(100) compared to that on fully oxidized CeO2(100). Reduction of both surfaces leads to a greater production of CO and H2. Reaction on all surfaces progresses rapidly from methoxy to products. There is no spectroscopic evidence of formyl or formate intermediates. On CeOX(100), carbonate is detected that decomposes into CO2 at high temperature.

1. INTRODUCTION

the O have two coordination vacancies, and the bridge site between Ce cations is very accessible through the first O layer. Methanol has been proposed as an ideal probe for characterizing the catalytic behavior of metal oxides.12,13 The molecule can titrate the active surface sites by acting as a Brønsted acid, deprotonating to deposit methoxy on the acidic cations and H on the basic anions. The reaction products then help characterize the substrate as an acid, base or redox catalyst. Strong acids lead to coupling products such as dimethyl ether, bases lead to dehydrogenation with CO and CO2 as the products, and redox catalysts produce formaldehyde. Earlier work on CeO2(111) produced conflicting results. Ferrizz et al. studied methanol on a CeO2(111) single crystal and observed only limited activity on the fully oxidized surface.14 They ascribed this limited activity to defects on the surface. Siokou and Nix, however, observed substantial methoxy formation on CeO2(111) films grown on Cu(111).15 It has been suggested, however, that these films may not have been fully oxidized.14,16 Our previous work showed that substantial methoxy formation occurred on CeO2(111) grown on Ru(0001).16 It was suggested that perhaps the defect concentration was greater on these films than on the CeO2(111) single crystal.1 Subsequent STM studies have indicated, however, that step edges and O vacancies could not account for the methoxy concentration on CeO2(111)/ Ru(0001).5,17 A more recent study by Matolin et al. demonstrated methoxy formation on a highly oxidized CeO2(111) film grown on Cu(111).18 They did not report an absolute coverage, however. As observed by Siokou and Nix,

An important issue in heterogeneous catalysis is understanding and exploiting the structure−function relationships of catalytic materials. A recent review by Vohs provided an excellent overview of the site requirements for the adsorption and reaction of organic oxygenates on metal oxide surfaces.1 The key factors to consider are the accessibility of the metal cations and oxygen anions on the surface and the coordination of the surface atoms compared to the bulk (i.e., the degree of coordinative unsaturation and the influence of surface defects such as step edges and O vacancies on the observed surface activity). Cerium oxide is an important catalytic material in a number of applications.2,3 The low-index faces of cerium oxide, shown in Figure 1, provide a diverse set of structural characteristics that enable us to test the site requirements for different catalytic reactions. CeO2(111) (Figure 1a) terminates with a single layer of O; however, the structure is relatively open, allowing access to the Ce cations in the second layer.4−7 The coordination numbers for Ce and O are reduced by one compared to the bulk (i.e., seven and three, respectively). On CeO2(110) (Figure 1c), Ce and O both lie in the first layer and the stoichiometry is preserved in the layer with twice as much O as Ce. The O has one coordination vacancy as on CeO2(111), but the Ce has two vacancies. The structure of CeO2(100) is more complicated. A simple bulk truncation would produce a surface that is entirely O or Ce. Such a termination is polar, however, and is inherently unstable.8 The instability can be resolved if half of the surface atoms are removed and placed on the opposing face (to preserve stoichiometry). A symmetric model for CeO2(100) is shown in Figure 1b.8−11 In this model, O is on the surface and Ce is in the second layer. Both the Ce and © 2013 American Chemical Society

Received: January 23, 2013 Revised: March 13, 2013 Published: March 13, 2013 4559

dx.doi.org/10.1021/la400295f | Langmuir 2013, 29, 4559−4567

Langmuir

Article

CeO2(111). Reduction increases this baselike behavior, leading to more CO and H2 on both surfaces.

2. EXPERIMENTAL SECTION The experiments were conducted in two different ultrahigh vacuum (UHV) chambers. The temperature programmed desorption (TPD) experiments were performed in a UHV chamber at Oak Ridge National Laboratory (ORNL) using a Hiden HAL/3F 301 mass spectrometer. The temperature was increased at a rate of 2 K/s, and the sample was biased −70 V to prevent electrons generated by the mass spectrometer ionizer from stimulating reactions at the ceria surface. The TPD measurements were made in a line-of-sight geometry in which the mass spectrometer aperture was positioned ca. 2 cm from the sample face. The set of masses that were monitored included mass 2 (H2), mass 16 (CH4), mass 18 (H2O), mass 28 (CO + CH2O), mass 30 (CH2O + CH3OH), mass 31 (CH3OH), mass 44 (CO2), mass 45 (acetic acid and dimethyl ether), and mass 60 (methyl formate). No products were observed at masses 16, 45, and 60. Soft X-ray photoelectron spectroscopy (sXPS), resonant photoemission spectroscopy (RPES), and near-edge X-ray absorption fine structure (NEXAFS) were performed in a UHV end-station located at beamline U12a of the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory. C 1s and O 1s sXPS spectra were recorded using 400 and 600 eV synchrotron radiation, respectively. The instrumental resolution was ca. 0.5 eV. The binding energy was referenced against the Ce 4d X‴ peak at 122.3 eV.16 Instrumental parameters such as the photon mesh sensitivity, the ESA sensitivity, and the photon spot size varied considerably between the CeOX(111) and CeOX(100) experiments that were conducted over a seven year period. To compare results on the two surfaces, photoemission signals from the clean ceria surfaces were used as intensity calibration standards. The O 1s, Ce 4d, and background intensities at kinetic energies slightly greater than the C 1s or O 1s photoemission peaks were used to normalize C 1s and O 1s spectra on the CeOX(111) and CeOX(100) surfaces. RPES at the Ce NIV,V edge is an extremely sensitive method for analyzing the Ce3+/Ce4+ content of the substrate.18,19 RPES consisted of recording the valence band spectrum under three distinct conditions: (1) off-resonance (hν = 115.9 eV), (2) Ce3+ resonance (hν = 122.3 eV), and (3) Ce4+ resonance (hν = 125.7 eV). These three photon energies are slightly different than those employed by Matolin et al. for RPES on cerium oxide.18,19 This is likely the result of differences in the beamline energy calibration. The intensities of these spectra were normalized to the photon flux measured by a gold mesh. NEXAFS was carried out at the C k-edge. The energy resolution was less than 0.5 eV, and the photon energy was calibrated using the dip in the photon flux at 284.7 eV.20 The X-ray absorption was measured using a partial yield electron detector. The high-pass retarding grid was biased −230 V. Insight into the orientation of the surface intermediate was obtained by recording NEXAFS for two different angles of incidence: normal incidence (which aligns the polarization of the light parallel to the surface) and grazing incidence (70° angle of incidence with respect to the surface normal). Thirdorder X-ray excitation at the Ce MIV and MV edges resulted in apparent absorption peaks at photon energies of 295 and 301.5 eV, respectively. The absorption due to only higher-order radiation was determined by collecting spectra with a retarding grid voltage of −310 V (i.e., greater than the first-order photon energy). The background resulting from the higher-order excitation was then subtracted from the NEXAFS spectra. CeO2(100) thin films were grown ex situ at the Center for Nanophase Materials Sciences at ORNL by pulsed laser deposition on 0.05% Nb-doped SrTiO3(100) (Nb-STO).21 The as-grown films were epitaxial and 19 ± 1 nm in thickness. Auger electron spectroscopy and sXPS indicated that the as-grown CeO2(100) surface contained S, C, K, and Cl impurities. S and C were removed by annealing in an oxygen background (1 × 10−6 Torr O2, 800 K, 5−10 min). K and Cl were removed by gentle sputtering (5 × 10−5 Torr Ar or Ne, 300 K, 1 keV, 1 μA, 600 K) with respect to the oxidized surface. At 291.2 eV, the formate/carbonate feature is evident in the 600 and 700 K spectra. The persistence of methoxy at 600 K is consistent with the TPD where the desorption of C-containing products shifts to higher temperatures on the reduced surface (Figure 3a, blue lines). Figure 5a,b shows O 1s spectra for CH3OH on oxidized CeO2(100) and reduced CeO1.67(100), respectively. In each case, we observe three features: (1) the lattice O peak at 530.5−531 eV, (2) a peak originating from methoxy and hydroxyl species between 532 and 533 eV, and (3) a shoulder (more conspicuous for the reduced surface) at higher binding energy originating from molecular CH3OH that is eliminated upon annealing to 300 K. On the oxidized surface (Figure 5a), the lattice O peak loses 55% of its initial intensity after methanol exposure at 190 K. The decrease in intensity results from both the attenuation of the substrate O due to the methoxy overlayer and a conversion of surface O into −OH following the deprotonation of the methanol. A peak that represents the joint contribution of CH3O− and HO− species emerges at 532.2 eV after methanol exposure at 190 K. This feature gradually decreases in intensity as the sample is annealed to 500 K and water and methanol desorb. As in the C 1s spectra, there is a dramatic change between 500 and 600 K as a majority of the products desorb. On the reduced surface (Figure 5b), the variation in the intensity of the lattice O peak is similar to that observed on the oxidized surface. It loses half of its original intensity upon methanol adsorption at 190 K and then progressively recovers with temperature. The lattice O peak is at 530.9−531.1 eV, constituting a shift toward higher binding energy relative to the oxidized surface. The peak arising from CH3O− + HO− at 532.8−533.1 eV loses only 28% of its initial intensity between

observed on CeO2(111). (2) A high-temperature H2O peak is produced on CeO2(100), which was not observed on CeO2(111). (3) There is a suppression of CH2O on reduced CeO1.67(100), whereas on reduced CeO1.67(111) the formaldehyde production increased with respect to the oxidized surface. (4) There is a single H2 peak on reduced CeO1.67(100) instead of the double-peak structure for reduced CeO1.67(111). 3.3. Soft X-ray Photoelectron Spectroscopy (sXPS). Figure 4a,b shows C 1s spectra for methanol on oxidized

Figure 4. C 1s core-level spectra following the adsorption of 5 Langmuirs of methanol at 190 K and annealed as indicated on (a) oxidized and (b) reduced CeOX(100). Each spectrum is from a separate adsorption/annealing experiment.

CeO2(100) and reduced CeO1.67(100), respectively. To minimize X-ray-induced chemistry on the adsorbate, each of the spectra displayed in Figure 4a,b was the result of a separate experiment in which a clean CeO2 − X(100) surface was prepared and 5 Langmuirs of methanol were subsequently dosed at 190 K. Following methanol exposure, the sample was 4563

dx.doi.org/10.1021/la400295f | Langmuir 2013, 29, 4559−4567

Langmuir

Article

Figure 6. Coverage of C (red) and O (black) in the adsorbates following a 5 Langmuir methanol exposure below 200 K on (A) oxidized and (B) reduced CeOX(100) (circles) and CeOX(111) (triangles). The coverages were derived from the integrated intensities in the core-level sXPS spectra as in Figures 4 and 5. On the oxidized surfaces (Figure 1), the Ce density (1 ML) is 7.9 and 6.8 nm−2 on CeO2(111) and CeO2(100), respectively.

the CeO2(111) surface is CH3O−. The absence of −OH on the surface is consistent with the TPD on CeO2(111) where water desorbs at low temperature and there is no water or H2 desorption at higher temperatures (Figure 3b). 16 On CeO2(100), the O coverage is greater than the C coverage because of the existence of both CH3O− and −OH on the surface. This is also consistent with the TPD (Figure 3a) where water desorbs over a broad range from 200 to 450 K. The differences on the reduced CeO 2 − X(100) and CeO2 − X(111) surfaces are more subtle. On both surfaces, the O adsorbate coverage is approximately twice the C coverage, indicating that CH3O− and −OH are present on both surfaces. Although the C coverage doubled on the reduced CeO2 − X(111) surface compared to that on the oxidized surface, the C coverage on the reduced CeO2 − X(100) increased by only 10%. Furthermore, the C coverage shows only a gradual decrease on CeO2 − X(111) between 200 and 600 K whereas the C coverage on CeO2 − X(100) decays more rapidly. This is also consistent with the TPDs (Figure 3a,b), where the methanol desorption decreases significantly over this range on CeO2 − X(111) but remains fairly intense on CeO2 − X(100). 3.4. Near-Edge X-ray Absorption Fine Structure (NEXAFS). Additional information regarding the identity and structure of the surface species produced by methanol on CeO2(100) was obtained from NEXAFS at the C k-edge.

Figure 5. O 1s core-level spectra following the adsorption of 5 Langmuirs of methanol at 190 K and annealed as indicated on (a) oxidized and (b) reduced CeOX(100). Each spectrum is from a separate adsorption/annealing experiment.

190 and 400 K, before diminishing over the range of 500−700 K as the products desorb. The C 1s and O 1s spectra on oxidized and reduced CeOX(100) are similar to those on CeOX(111) in that CH3O− and HO− are the principal adsorbates observed.16 The concentrations of these adsorbates are very different on the two surfaces, however. Figure 6a,b shows for the oxidized and reduced surfaces, respectively, the coverages of C- and Ocontaining adsorbates on CeOX(100) and CeOX(111). The coverages were derived from the integrated C 1s and O 1s intensities and were calibrated using the known coverage of CH3SH on Ru(0001).16 On the oxidized surfaces, Figure 6a, there are two fundamental differences between the two surfaces. First, the C coverage is much greater on CeO2(100) than on CeO2(111). At 300 K, the coverage on CeO2(100) is roughly twice that on CeO2(111). In addition, on CeO2(111) the O coverage and the C coverage are the same over the entire temperature range. This indicates that the only adsorbate on 4564

dx.doi.org/10.1021/la400295f | Langmuir 2013, 29, 4559−4567

Langmuir

Article

that is not found in the spectrum acquired at normal incidence. Based on the peak position (290 eV), we can conclude that the surface intermediate is a carbonate [(CO3)2−]20,32 rather than formate [(HCOO)−], which has its resonance at 288 eV.33,34 Moreover, on the basis of the angular dependence of the NEXAFS at 560 K, we can conclude that the carbonate is lying flat with its molecular plane parallel to the CeO2(100) surface. The CO π* resonance is excited at grazing incidence whereas the CO σ* resonance is excited at normal incidence.

NEXAFS spectra recorded after exposing the substrate to methanol at 190 K and then annealing to 300 K are shown in Figure 7a. Black lines correspond to normal incidence, whereas

4. DISCUSSION Fully oxidized CeO2(100) is clearly more reactive with respect to methanol adsorption and decomposition compared to oxidized CeO2(111). Roughly twice as much methoxy is produced on CeO2(100) compared to that produced on CeO2(111). However, this additional activity leads to less selectivity in product formation. In addition to formaldehyde and methanol, the methoxy decomposition produces CO and H2, which were not observed on CeO2(111). The greater adsorbate stability, which leads to a higher adsorbate concentration, on CeO2(100) can be understood when considering the surface structure shown in Figure 1b. The dissociation of methanol into methoxy and hydroxyl requires acidic sites (Ce cations) and basic sites (O anions). On CeO2(100), the methoxy and the hydroxyl adsorb on identical sites (i.e., bridge sites between two Ce cations). On CeO2(111), methoxy forms on top of a Ce cation whereas hydroxyl forms on an O anion that is already coordinated to three Ce cations. Viewed from a slightly different perspective, the adsorption sites on CeO2(100) each have two coordination vacancies whereas the adsorption sites on CeO2(111) each have a single coordination vacancy. This makes the adsorption sites on CeO2(100) inherently more reactive. Below 600 K, the only surface species formed by methanol are methoxy and hydroxyl. There is no evidence of formate as proposed by Matolin et al.18 and Siokou and Nix.15 The behavior of the hydroxyls is different on CeO2(100) and CeO2(111). On CeO2(111), the hydroxyls react to form water at low temperature, leaving only methoxy on the surface. This is evident from the TPD (Figure 3b) and the O 1s versus C 1s intensity ratio (Figure 6a). On CeO2(100), the hydroxyls are more stable and react to form water over a broad temperature range from 200 to 450 K. The hydroxyl stability and water formation are very consistent with what was observed during water adsorption and recombination on CeO2(100) and CeO2(111).21 It has been proposed that the adsorption sites on CeO2(111) are sufficiently inactive so as to preclude methanol adsorption and decomposition on the fully oxidized surface.1,14 However, our C 1s results (Figure 6a) indicate that the methoxy concentration on CeO2(111) is ca. 4 nm−2 at 300 K. This is ca. 0.5 ML because the Ce4+ density on the CeO2(111) surface is 8 nm−2. On the reduced surface, the methoxy coverage increases to ca. 7 nm−2 at 300 K (Figure 6b). The TPD results are not fully consistent with these C 1s results. As shown in Figure 3b, the C-containing products increase by four to five times when the surface is reduced rather than doubling as indicated by the C 1s data. The source of this discrepancy is not known; however, an estimate of the defect density on the oxidized CeO2(111) surface indicates that the methoxy coverage is significantly greater than the defect density on this surface in either case. RPES data indicate less than 5% Ce3+ for CeO2(111) grown on Cu(111)18 and ca. 2% on CeO2(100)

Figure 7. Near-edge X-ray absorption spectra (NEXAFS) following the adsorption of 5 Langmuirs of methanol on oxidized CeO2(100) at 190 K and then annealing to (A) 300 and (B) 560 K. The black lines are for X-rays impinging at normal incidence, and the red lines are at grazing incidence. The spectra have been offset for clarity.

red lines correspond to grazing incidence (70° from the surface normal). The spectra are fully consistent with those obtained from methoxy on other surfaces.30,31 Note that the C−O σ* resonance centered around 295 eV is virtually nonexistent at normal incidence and fairly intense at grazing incidence. This resonance is excited when the X-ray polarization vector is oriented along the C−O bond axis. Because the polarization vector is orthogonal to the direction of incidence, the angular dependence indicates that the C−O bond is oriented normal to the surface. The NEXAFS spectra acquired after heating the adsorbatecovered sample to 560 K are shown in Figure 7b. Between 500 and 600 K, desorption products are produced and methoxy disappears from the surface as indicated by the C 1s sXPS in Figure 4a. Solely on the basis of its binding energy, the species that is produced at ca. 291 eV could be either formate or carbonate.29 The NEXAFS spectra are a composite of the absorption produced by methoxy still present on the surface plus the higher-binding-energy species. At grazing incidence (red line), a sharp peak is seen at a photon energy of 290 eV 4565

dx.doi.org/10.1021/la400295f | Langmuir 2013, 29, 4559−4567

Langmuir

Article

UHV and high-surface-area CeO2 cubes and octahedra at atmospheric pressure.36 The cubes, which expose CeO2(100) surfaces, adsorb and react much more methanol than do the octahedra, which expose CeO2(111) surfaces. The primary products from both cubes and octahedra are CO and H2. There was very little CH2O observed. However, this may be due to readsorption and reaction as products proceed through the catalyst bed. Formaldehyde may further react to produce CO and H2. Wu et al. were also able to determine the adsorption sites for methoxy on the ceria nanoshapes based on the ν(CO) vibrational frequencies.36 On the octahedra, the methoxy adsorbed exclusively on top of the Ce cations. This is consistent with the CeO2(111) structure shown in Figure 1a where these are the only adsorption sites available on the Ce. However, on the cubes the vibrational spectroscopy indicates nearly equal amounts of methoxy adsorbed in bridge sites and on top of the Ce cations. In the structure shown in Figure 1b, the on-top sites are inaccessible because of the first-layer O. The results on the nanocubes therefore indicate that the structure of CeO2(100) may be more complicated than is depicted in Figure 1b with possibly open domains of Ce without an O overlayer.

(Figure 2a). RPES data for CeO2(111) on Ru(0001) is not available; however, the Ce 3d and Ce 4d XPS spectra are the same for CeO2(100) and CeO2(111). 5% Ce3+ equates to 1.25% O vacancies (fully reduced Ce2O3 has 25% O vacancies) or 0.1 nm−2. STM images indicate that the CeO2(111)/ Ru(0001) surface is composed of CeO2(111) islands or plates greater than 20 nm in diameter.5,17 If we assume circular islands with a nearest-neighbor separation of 0.38 nm between Ce cations at the edges, the edge-site density would be ∼0.5 nm−2. Either of these two estimates is less than the apparent methoxy coverage at 300 K. We also note that our previous results for formic acid adsorbed on oxidized CeO2(111)/Ru(0001) are in substantial agreement with the results obtained on single-crystal CeO2(111), specifically, that a substantial amount of formic acid reacts and adsorbs on the fully oxidized surface.29,35 An analysis of the C 1s sXPS data for formic acid and methanol on CeO2(111) indicates that the formate coverage is only 30% greater than the methoxy coverage, further supporting the conclusion that methanol adsorption on CeO2(111)/Ru(0001) is not solely related to vacancies and defects. The product distribution from methanol decomposition on CeO2(100) and CeO2(111) provides additional information as to the nature of these surfaces as catalytic materials. Methanol has been proposed as a chemical probe to determine whether a metal oxide is a redox catalyst or an acid−base catalyst.12,13 If the methanol produces primarily CH2O as a product, then the substrate is a redox catalyst. If the result is primarily CH3OCH3, then the substrate is acidic, and if the products are CO and CO2, then the substrate is basic. CeO2(111) produces exclusively CH2O and methanol as its C-containing products whereas CeO2(100) produces CO and CO2 in addition to formaldehyde and methanol. The more highly undercoordinated O anions on the surface of CeO2(100) may explain its more basic/nucleophilic nature. The nucleophilic attack on the methyl group leads to dehydrogenation. Similar dehydrogenation leading to CO and H2 formation occurs on reduced CeO2 − X(100) and CeO2 − X(111) surfaces. When this process starts, it apparently occurs rapidly; no partially dehydrogenated intermediates have been identified on either surface. Carbonate, CO32−, is formed on CeO2(100), but this appears to be a fully dehydrogenated, stable intermediate that ultimately leads to CO2. There is substantial qualitative agreement between the TPD results on CeOX(100) shown in Figure 3a and previously reported TPD results by Ferrizz et al. on CeO2(100)/ YSZ(100).14 The same products were observed in both studies with more methanol, formaldehyde, and water produced on the fully oxidized surface, which give way to CO and H2 on a reduced surface. The CO and H2CO desorption temperatures also shift to higher temperatures as the sample is reduced. The desorption temperatures for all products are greater in the previous study compared to the current results. Ferrizz et al. suggested that the progressive desorption temperatures of the C-containing products, T(CH3OH) < T(CH2O) < T(CO), resulted from the successive recombination or dehydrogenation of the surface methoxy. In the current results, CH3OH and CH2O desorb at the same temperature as they do on CeO2(111), suggesting that these products result from the disproportionation of two methoxides. The higher-temperature desorption of CO, H2, CO, and CO2 results from the dehydrogenation reaction. There are some clear similarities between the behavior of methanol on low-surface-area CeO2(100) and CeO2(111) in

5. CONCLUSIONS CeO2(100) is more active than CeO2(111) for methanol adsorption and reaction but perhaps less selective. On CeO2(100), methanol fully dehydrogenates, producing CO, CO2, and H2 in addition to formaldehyde and water. This can be understood in light of the surface structures where CeO2(100) provides greater access to the Ce adsorption sites and where the Ce cations are more undercoordinated on CeO2(100), producing more stable methoxy surface species. The O anions on the surface of CeO2(100) are also more undercoordinated than on CeO2(111). This produces more stable hydroxyls on (100) but also makes the (100) surface more effective in dehydrogenating the methoxy species.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 865-574-2796. Fax: 865576-5235. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was sponsored by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences, U.S. Department of Energy, under contract DEAC05-00OR22725 with Oak Ridge National Laboratory, managed and operated by UT-Battelle, LLC. Use of the National Synchrotron Light Source, Brookhaven National Laboratory, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract no. DE-AC02-98CH10886. Research at Oak Ridge National Laboratory’s Center for Nanophase Materials Sciences was sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy. The submitted manuscript has been authored by a contractor of the U.S. Government under contract no. DE-AC05-00OR22725. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form 4566

dx.doi.org/10.1021/la400295f | Langmuir 2013, 29, 4559−4567

Langmuir

Article

Adsorption on Rh/CeO2(001). J. Vac. Sci. Technol., A 1997, 15, 1647− 1652. (23) Mullins, D. R.; Overbury, S. H.; Huntley, D. R. Electron Spectroscopy of Single Crystal and Polycrystalline Cerium Oxide Surfaces. Surf. Sci. 1998, 409, 307−319. (24) Hagans, P. L.; Dekoven, B. M.; Womack, J. L. A Laser Drilled Aperture for Use in an Ultrahigh-Vacuum Gas Doser. J. Vac. Sci., Technol. A 1989, 7, 3375−3377. (25) Chen, T. L.; Mullins, D. R. Adsorption and Reaction of Acetaldehyde over CeOX(111) Thin Films. J. Phys. Chem. C 2011, 115, 3385−3392. (26) Powell, C. J.; Jablonski, A.; Tilinin, I. S.; Tanuma, S.; Penn, D. R. Surface Sensitivity of Auger-Electron Spectroscopy and X-ray Photoelectron Spectroscopy. J. Electron. Spectrosc. 1999, 98, 1−15. (27) Ko, E. I.; Benziger, J. B.; Madix, R. J. Reactions of Methanol on W(100) and W(100)-(5 × 1)C Surfaces. J. Catal. 1980, 62, 264−274. (28) Weststrate, C. J.; Westerstrom, R.; Lundgren, E.; Mikkelsen, A.; Andersen, J. N. Influence of Oxygen Vacancies on the Properties of Ceria-Supported Gold. J. Phys. Chem. C 2009, 113, 724−728. (29) Senanayake, S. D.; Mullins, D. R. Redox Pathways for HCOOH Decomposition over CeO2 Surfaces. J. Phys. Chem. C 2008, 112, 9744−9752. (30) Lindner, T.; Somers, J.; Bradshaw, A. M.; Kilcoyne, A. L. D.; Woodruff, D. P. A Photoelectron Diffraction and Nexafs Study of the Structure of the Methoxy Species (Ch3o-) on Cu(100). Surf. Sci. 1988, 203, 333−352. (31) Outka, D. A.; Stohr, J.; Madix, R. J.; Rotermund, H. H.; Hermsmeier, B.; Solomon, J. Nexafs Studies of Complex Alcohols and Carboxylic-Acids on the Si(111)(7 × 7) Surface. Surf. Sci. 1987, 185, 53−74. (32) Brandes, J. A.; Wirick, S.; Jacobsen, C. Carbon K-Edge Spectra of Carbonate Minerals. J. Synchrotron Radiat. 2010, 17, 676−682. (33) Gutierrez-Sosa, A.; Martinez-Escolano, P.; Raza, H.; Lindsay, R.; Wincott, P. L.; Thornton, G. Orientation of Carboxylates on TiO2(110). Surf. Sci. 2001, 471, 163−169. (34) Gutierrez-Sosa, A.; Evans, T. M.; Parker, S. C.; Campbell, C. T.; Thornton, G. Geometry of C1−3 Oxygenates on ZnO(0001)-Zn. Surf. Sci. 2002, 497, 239−246. (35) Stubenrauch, J.; Brosha, E.; Vohs, J. M. Reaction of Carboxylic Acids on CeO2(111) and CeO2(100). Catal. Today 1996, 28, 431− 441. (36) Wu, Z. L.; Li, M. J.; Mullins, D. R.; Overbury, S. H. Probing the Surface Sites of CeO2 Nanocrystals with Well-Defined Surface Planes via Methanol Adsorption and Desorption. ACS Catal. 2012, 2, 2224− 2234.

of this contribution, or allow others to do so, for U.S. Government purposes.

■

REFERENCES

(1) Vohs, J. M. Site Requirements for the Adsorption and Reaction of Oxygenates on Metal Oxide Surfaces. Chem. Rev. 2012,10.1021/ cr300328u. (2) Trovarelli, A. Catalysis by Ceria and Related Materials; Imperial College Press: London, 2002. (3) Vivier, L.; Duprez, D. Ceria-Based Solid Catalysts for Organic Chemistry. ChemSusChem 2010, 3, 654−678. (4) Mullins, D. R.; Radulovic, P. V.; Overbury, S. H. Ordered Cerium Oxide Thin Films Grown on Ru(0001) and Ni(111). Surf. Sci. 1999, 429, 186−198. (5) Lu, J. L.; Gao, H. J.; Shaikhutdinov, S.; Freund, H. J. Morphology and Defect Structure of the CeO2(111) Films Grown on Ru(0001) as Studied by Scanning Tunneling Microscopy. Surf. Sci. 2006, 600, 5004−5010. (6) Fukui, K.; Namai, Y.; Iwasawa, Y. Imaging of Surface Oxygen Atoms and Their Defect Structures on CeO2(111) by Noncontact Atomic Force Microscopy. Appl. Surf. Sci. 2002, 188, 252−256. (7) Gritschneder, S.; Reichling, M. Structural Elements of CeO2(111) Surfaces. Nanotechnology 2007, 18, 044024. (8) Conesa, J. C. Computer Modeling of Surfaces and Defects on Cerium Dioxide. Surf. Sci. 1995, 339, 337−352. (9) Norenberg, H.; Harding, J. H. The Surface Structure of CeO(2) (001) Single Crystals Studied by Elevated Temperature STM. Surf. Sci. 2001, 477, 17−24. (10) Herman, G. S. Surface Structure Determination of CeO2(001) by Angle-Resolved Mass Spectroscopy of Recoiled Ions. Phys Rev B 1999, 59, 14899−14902. (11) Molinari, M.; Parker, S. C.; Sayle, D. C.; Islam, M. S. Water Adsorption and Its Effect on the Stability of Low Index Stoichiometric and Reduced Surfaces of Ceria. J. Phys. Chem. C 2012, 116, 7073− 7082. (12) Tatibouet, J. M. Methanol Oxidation as a Catalytic Surface Probe. Appl. Catal., A 1997, 148, 213−252. (13) Badlani, M.; Wachs, I. E. Methanol: A “Smart” Chemical Probe Molecule. Catal. Lett. 2001, 75, 137−149. (14) Ferrizz, R. M.; Wong, G. S.; Egami, T.; Vohs, J. M. Structure Sensitivity of the Reaction of Methanol on Ceria. Langmuir 2001, 17, 2464−2470. (15) Siokou, A.; Nix, R. M. Interaction of Methanol with WellDefined Ceria Surfaces: Reflection/Absorption Infrared Spectroscopy, X-ray Photoelectron Spectroscopy, And Temperature-Programmed Desorption Study. J. Phys. Chem. B 1999, 103, 6984−6997. (16) Mullins, D. R.; Robbins, M. D.; Zhou, J. Adsorption and Reaction of Methanol on Thin-Film Cerium Oxide. Surf. Sci. 2006, 600, 1547−1558. (17) Zhou, J.; Baddorf, A. P.; Mullins, D. R.; Overbury, S. H. Growth and Characterization of Rh and Pd Nanoparticles on Oxidized and Reduced CeOx(111) Thin Films by Scanning Tunneling Microscopy. J. Phys. Chem. C 2008, 112, 9336−9345. (18) Matolin, V.; Libra, J.; Skoda, M.; Tsud, N.; Prince, K. C.; Skala, T. Methanol Adsorption on a CeO2(111)/Cu(111) Thin Film Model Catalyst. Surf. Sci. 2009, 603, 1087−1092. (19) Matolin, V.; Cabala, M.; Chab, V.; Matolinova, I.; Prince, K. C.; Skoda, M.; Sutara, F.; Skala, T.; Veltruska, K. A Resonant Photoelectron Spectroscopy Study of Sn(O-x) Doped CeO2 Catalysts. Surf. Interface Anal. 2008, 40, 225−230. (20) Stohr, J. NEXAFS Spectroscopy; Springer-Verlag: New York, 2003; Vol. 25, p 159. (21) Mullins, D. R.; Albrecht, P. M.; Chen, T. L.; Calaza, F. C.; Biegalski, M. D.; Christen, H. M.; Overbury, S. H. Water Dissociation on CeO2(100) and CeO2(111) Thin Films. J. Phys. Chem. C 2012, 116, 19419−19428. (22) Overbury, S. H.; Huntley, D. R.; Mullins, D. R.; Ailey, K. S.; Radulovic, P. V. Surface Studies of Model Supported Catalysts: NO 4567

dx.doi.org/10.1021/la400295f | Langmuir 2013, 29, 4559−4567