In Situ Characterization of an Inverse - American Chemical Society

3580

J. Phys. Chem. C 2010, 114, 3580–3587

Unraveling the Active Site in Copper-Ceria Systems for the Water-Gas Shift Reaction: In Situ Characterization of an Inverse Powder CeO2-x/CuO-Cu Catalyst Laura Barrio,† Michael Estrella,† Gong Zhou,† Wen Wen,†,| Jonathan C. Hanson,† Ana B. Hungrıá,‡ Aitor Horne´s,§ Marcos Fernańdez-Garcıá,§ Arturo Martıńez-Arias,§ and Jose´ A. Rodriguez*,† Department of Chemistry, BrookhaVen National Laboratory, Upton, New York 11973, Departamento de Ciencia de Materiales, Ingenierıá Metalu´rgica y Quı´mica Inorgańica, Facultad de Ciencias, UniVersidad de Ca´diz, 11510 Puerto Real, Ca´diz, Spain, and Instituto de Cata´lisis y Petroleoquı´mica, CSIC, C/ Marie Curie 2, Campus de Cantoblanco, 28049, Madrid, Spain ReceiVed: October 29, 2009; ReVised Manuscript ReceiVed: December 22, 2009

An inverse powder system composed of CeO2 nanoparticles dispersed over a CuO-Cu matrix is proposed as a novel catalyst for the water-gas shift reaction. This inverse CeO2/CuO-Cu catalyst exhibits a higher activity than standard Cu/CeO2 catalysts. In situ synchrotron characterization techniques were employed to follow the structural changes of CeO2/CuO-Cu under reaction conditions. Time-resolved X-ray diffraction experiments showed the transformation of CuO to metallic Cu via a Cu2O intermediate. Short-order structural changes were followed by pair distribution function analysis and corroborated the results obtained by diffraction. Moreover, X-ray absorption spectroscopy also revealed oxidation state changes from Cu2+ to Cu0 and the partial reduction of CeOx nanoparticles. The activity data obtained by mass spectrometry revealed that hydrogen production starts once the copper has been fully reduced. The strong interaction of ceria and copper boosted the catalytic performance of the sample. The inverse catalyst was active at low temperatures, stable to several reaction runs and to redox cycles. These characteristics are highly valuable for mobile fuel cell applications. The active phases of the inverse CeO2/CuO-Cu catalyst are partially reduced ceria nanoparticles strongly interacting with metallic copper. The nature and structure of the ceria nanoparticles are of critical importance because they are involved in processes related to water dissociation over the catalyst surface. Introduction At present, hydrogen is mainly produced from the reforming of crude oil, coal, natural gas, wood, organic waste, and biomass.1,2 The CO (1-10% content) present in the reformed fuel degrades the performance of the Pt electrode used in fuel cell systems. The water-gas shift (WGS) reaction is a critical process for procuring clean hydrogen. As noted in the following chemical equation

CO + H2O T CO2 + H2

(1)

for each mole of CO removed, a mole of hydrogen is produced. The WGS reaction then allows not only the removal of CO but also an upgrade in fuel cell efficiency by increasing the H2 concentration. For mobile fuel cell applications, conventional WGS catalysts are not suitable and advanced systems are required. Commercial Cu-Zn-based catalysts are pyrophoric, require special activation procedures, and are intolerant to oxidation. Because of the expected exposure to many start-up/shutdown cycles of mobile fuel cell systems, novel WGS catalysts should be tolerant to redox cycles and steam condensation.1,2 Ceria-based nanocata* To whom correspondence should be addressed. E-mail: rodrigez@ bnl.gov. Fax: 1-631-344-5815. † Brookhaven National Laboratory. ‡ Universidad de Ca´diz. § Instituto de Cata´lisis y Petroleoquı´mica, CSIC. | Current address: Experimental Division, Shanghai Synchrotron Radiation Facility, Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Pudong New Area, Shanghai, People’s Republic of China, 201204.

lysts have been investigated extensively in recent years and are expected to be part of the next generation of WGS catalysts.3-7 To obtain low-temperature WGS activity, cerium oxide is usually loaded with reduction promoter metals, such as Rh, Pt, Cu, or Au.3-7 The design and optimization of new catalysts for the WGS reaction is hindered by the complex reaction mechanism and the difficulty of identifying active species. The characterization of the catalysts and reaction intermediates under WGS process conditions is, therefore, important.8,9 The WGS mechanism on Cu-based catalysts has been a question of study for many years.3,7-16 It has been generally accepted that the active phase involves reduced Cu metal sites supported on metal oxides. Nevertheless, the metal oxide support plays an important role in the reaction mechanism. CeO2-supported systems have shown enhanced activity in the WGS reaction with various metals, and this has been ascribed to the excellent oxygen storage capability of CeO2. In this sense, although CeO2 alone does not exhibit significant WGS activity, it appears typically acting as a support, an essential promoter for achieving enhanced WGS activity. Mechanistic DRIFT studies have suggested that CeO2 is able to readily dissociate water,17,18 whereas Cu sites adsorb CO,14 and then, the oxygen is transferred from the support in order to oxidize the CO.19 However, it has been demonstrated that Cu surfaces alone are also active in the WGS reaction.10,14,20 Furthermore, Cu nanoparticles have also shown the ability to dissociate water and have proved to be far more active than the well-ordered surfaces of Cu (111).14 Traditionally, precious metal nanoparticles are dispersed over metal oxide supports, with the conviction that the supports’ main

10.1021/jp910342b  2010 American Chemical Society Published on Web 02/08/2010

Inverse CeO2-x/CuO-Cu Catalyst for the WGS Reaction

J. Phys. Chem. C, Vol. 114, No. 8, 2010 3581

SCHEME 1: Structures of Conventional Cu/CeO2 and Inverse CeO2/Cu Catalysts

role is to stabilize and disperse active sites along its surface. Recently, inverse model catalysts of CeOx nanoparticles supported over Au(111)21 or Cu(111)22 have shown high catalytic performance in the WGS reaction, up to the point that a CeOx/ Cu(111) system is more active than Cu/CeO2(111) and Cu/ ZnO(0001j) systems. In the present work, we move away from model catalysts and investigate the behavior of inverse powder catalysts with relatively high specific surface area involving cerium oxide nanoparticles dispersed over a CuO/Cu support. This approach will allow us to better study the role of CeO2 and Cu in the reaction mechanism. As shown in Scheme 1, the inverse CeO2/CuO-Cu catalyst exposes ceria nanoparticles to the reactants. Defect sites present in the oxide are not covered by metal particles, as occurs in the case of a traditional Cu/ CeO2 catalyst. On the inverse CeO2/CuO-Cu catalyst, the reactants can interact with defect sites of the oxide nanoparticles, metal sites, and the metal-oxide interface. In this article, in situ characterization techniques were employed to correlate the structure of the inverse CeO2/CuO-Cu catalyst with its activity. Time-resolved X-ray diffraction (TR-XRD)8,23 provided unique information on the crystalline phases present in the sample, and subsequent data refinement revealed changes in phase composition, unit cell size, and oxygen vacancies. The atomic pair distribution function (PDF) method24-26 gave short- and long-range structural information. X-ray absorption near-edge structure (XANES) was used to monitor changes in the oxidation state of the catalyst during the WGS reaction. With the inverse catalyst proposed in the present work, the copper species is the major component, and we will exploit this feature to clarify details of the copper-tocerium oxide interactions. Experimental Section A. Catalyst Preparation. The inverse CeO2/CuO catalyst was prepared by employing reverse microemulsions containing n-heptane, Triton-X-100, and n-hexanol as organic solvent, surfactant, and cosurfactant, respectively, in amounts similar to those reported previously.27 The required amount of Cu(NO3)2 was dissolved in distilled water and added to the former in order to form the reverse microemulsion. Simultaneously, another microemulsion of similar characteristics was prepared containing, dissolved in its aqueous phase, the required amount of tetramethyl ammonium hydroxide (TMAH). After 1 h of stirring of the two microemulsions, the TMAH-containing one was added to the Cu-containing one and it was left for the period of 18-24 h in order to complete the precipitation reaction. The resulting microemulsion was then heated gently up to 339 K using a water bath. This microemulsion containing the precipitated copper was then mixed with another one of similar characteristics in which Ce had previously been precipitated, following mixing of Ce(NO3)2- and TMAH-containing microemulsions. This final microemulsion containing both precipitated Cu and Ce components was kept under agitation for the period of 18-24 h. The resulting solid was then separated by centrifugation and decantation, rinsed with methanol, and dried

Figure 1. Rietveld analysis of an XRD pattern (top panel) and PDF analysis (bottom panel) of fresh inverse CeO2/CuO catalyst.

overnight at 100 °C, and the resulting powder was calcined at 773 K for 2 h under air. Nominal Cu/Ce atomic ratios of 7/3 was employed for this catalyst, close to 2.57 detected by X-ray fluorescence elemental analysis. A specific area value of SBET ) 66.9 m2 g-1 was obtained from BET analysis of the corresponding N2 adsorption isotherm. B. Electron Microscopy. High-resolution electron microscopy (HREM), high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images, and X-ray energy-dispersive spectra (XEDS) were recorded on a 200 kV FEI Tecnai F20-G2 TEM/STEM miscroscope equipped with an EDAX r-TEM ultrathin window (UTW) X-ray detector. XEDS analysis was performed in STEM mode, with a probe size of ∼1 nm. Specimens were prepared by depositing particles of the samples to be investigated onto a molybdenum grid supporting a perforated carbon film. Deposition was achieved by dipping the grid directly into the powder of the samples to avoid contact with any solvent. C. In Situ Time-Resolved X-ray Diffraction. In situ timeresolved X-ray diffraction (TR-XRD) experiments were carried out on beamline X7B of the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory.8,9 The sample (∼5 mg) was loaded into a glass cell with a diameter of 1 mm, which was attached to a flow system.28 A small resistance heater was wrapped around the capillary, and the temperature was monitored with a 0.1 mm chromel-alumel thermocouple placed in the capillary near the sample.28 Two-dimensional powder patterns were collected with a Mar345 image plate detector and

3582

J. Phys. Chem. C, Vol. 114, No. 8, 2010

Barrio et al.

Figure 2. HAADF-STEM image of the inverse CeO2/CuO catalyst, showing a CuO particle in (A) and CeO2 nanoparticles in (B).

the powder diffraction rings were integrated using the FIT2D code.29 The instrument parameters (Thompson-Cox-Hastings profile coefficients) were derived from the fit of a LaB6 reference pattern. Rietveld profile refinements were performed with the aid of GSAS software.30 The series of powder patterns was refined by sequential analysis where the starting model is based on the earlier powder pattern. In addition to Rietveld refinements, pair distribution function (PDF) analyses were performed with the PDFgetX231 program to apply corrections to the intensity data and to calculate the resulting pair distribution function. Diffraction patterns were collected over the catalyst during the WGS reaction, reduction in 1% CO/He, and oxidation in 5% O2/He. The WGS reaction was carried out with a stepped temperature program from room temperature to 500 °C, with 3 h soaks at every 100 °C beyond 200 °C amid a 1% CO and 99% He gas mixture flow rate of 10 mL/min and a space velocity of 4000 h-1 (volumetric feed flow rate/catalytic bed volume). This gas mixture passed through a water bubbler (room temperature) before entering the reactor. The relative ratio of water vapor to CO pressures in the gas mixture was 3. D. In Situ Time-Resolved X-ray Absorption. Cu K-edge and Ce LIII-edge in situ XANES spectra were collected at beamline X19A of the NSLS under similar operational conditions as those for the TR-XRD experiments. The same cell was used for the XANES experiments as that for in situ XRD,28 except that the sample was loaded into a Kapton capillary and heated with a hot air blower. The X-ray absorption spectra were taken repeatedly in the “fluorescence-yield mode” using a passivated implanted planar silicon (PIPS) detector cooled with circulating water. The XANES data were then analyzed using the Athena program.32

The relative product concentrations from both TR-XRD and TR-XANES experiments were measured with a 0-100 amu quadruple mass spectrometer (QMS, Stanford Research Systems). A portion of the exit gas flow passed through a leak valve and into the QMS vacuum chamber. QMS signals at mass-tocharge ratios of 2(H2), 4(He), 17(OH), 18(H2O), 28(CO), 32(O2), and 44(CO2) were monitored and recorded during the experiments. Results and Discussion A. Characterization of the Inverse CeO2/CuO Catalyst. We investigated the long- and short-range structure of the CeO2/ CuO catalyst using XRD and the PDF method.35-37 In the top panel of Figure 1, the XRD pattern for the CeO2/CuO catalyst has only diffraction peaks for CeO233 and CuO.34 A Rietveld refinement of this pattern gave a molar composition of 40% CeO2 and 60% CuO. No other phase was detected in the longrange probed with XRD or in the short-range probed by a PDF analysis. The PDF traces were well-fitted by a sum of CeO2 and CuO features (bottom panel of Figure 1). The CeO2/CuO catalyst shows peaks at 1.95 and 3.05 Å, which correspond to the Cu-O and Cu-Cu distances in the CuO phase. The CeO2 phase has a peak at 2.35 Å for the Ce-O distance and a peak at 3.90 Å for the Ce-Ce distance. Using the Scherrer equation35 and the width of the diffraction peaks in Figure 1, we obtained average particle sizes of 8.7 nm for ceria and 29.4 nm for CuO. The inverse configuration of the CeO2/CuO catalyst was demonstrated by means of HAADF-STEM-XEDS and HRTEM analyses. Figure 2 displays a HAADF-STEM image representative for this catalyst. The image illustrates the presence of relatively big CuO particles, as evidenced by the XED spectra taken in zone A, onto which small (in the range of 5-10 nm)


J. Phys. Chem. C, Vol. 114, No. 8, 2010 3583

Figure 3. HRTEM image of the inverse CeO2/CuO catalyst showing CeO2 nanoparticles supported on a CuO particle.

CeO2 particles appear supported, usually forming small aggregates of several nanocrystals, as evidenced by XEDS performed in the zone around point B. The size (within the range of 5-10 nm), shape, and distribution of the CeO2 nanocrystals, forming small aggregates supported on a CuO particle, is further evidenced by HRTEM, as shown in Figure 3. B. Water-Gas Shift Reaction on the Inverse CeO2/CuO Catalyst. In situ measurements using XRD, PDF, and XANES show that the composition of the inverse CeO2/CuO catalyst changes under the reaction conditions of the water-gas shift. Figure 4 shows TR-XRD patterns for the CeO2/CuO inverse catalyst during the water-gas shift reaction at different temperatures. The starting material was composed of CuO and CeO2 phases. As the reaction progressed, some of the CuO was first reduced to Cu2O as an intermediate and finally to metallic Cu0 at 200 °C. The reduction of CuO under a CO/H2O mixture has been reported previously.8,16 The bottom panel in Figure 4 shows the evolution of the copper phases obtained through a Rietveld refinement of the XRD data. During the reduction process, the Cu2O fraction increases before the copper oxides are fully reduced to Cu0. At 200 °C, CuO, Cu2O, and Cu coexist in the catalyst. In the diffraction patterns, the peaks for the ceria nanoparticles did not disappear under reaction conditions, but a shift in their position denoted an expansion in the ceria unit cell from 5.402 to 5.434 Å. This expansion is larger than that found for a thermal expansion of ceria,3,7,16 and it probably reflects the formation of O vacancies and an increase in the concentration of Ce3+.16,36 The bigger size of Ce3+ over Ce4+ leads to a larger lattice parameter.36 Thus, changes in the ceria lattice parameter can be directly correlated to the concentration of oxygen vacancies and Ce3+ cations in the oxide.36,37 Figure 5A displays in situ Cu K-edge XANES spectra collected over the CeO2/CuO sample during the WGS reaction at different temperatures. The XANES features of the starting material were typical of CuO species.38 As the temperature was increased, the CuO became reduced. At 200 °C, which is the onset for Cu2O phase appearance in diffraction (Figure 4), we observed the profile for Cu2O in the Cu K-edge.38 At 250 °C, the reduction of copper was completed and the spectral features

Figure 4. Top: TR-XRD patterns (λ ) 0.3184 Å) collected during the WGS reaction over an inverse CeO2/CuO catalyst. Bottom: Rietveld refinement of mole fraction variation and lattice parameter changes of ceria as a function of temperature during the WGS reaction.

of XANES matched those of the Cu standards. Thus, the XAFS data are consistent with the XRD results and confirmed the formation of fully reduced Cu under WGS reaction conditions. The Ce LIII-edge was also examined by XANES during the WGS reaction. The results are shown in Figure 5B along with spectra for bulk CeO2 (reference for Ce4+) and Ce(NO3) · 6H2O (reference for Ce3+). The changes in the spectral features show that CeO2 was partially reduced under WGS reaction conditions. The amount of oxygen vacancies and Ce3+ cations increased continuously with increasing temperature. This behavior is consistent with the one obtained by a Rietveld analysis of the XRD results, which also pointed to a partial reduction of CeO2 under reaction conditions. It must be noted that the presence of oxygen vacancies in the ceria lattice is an important property of the active phase of this type of system.7 In view of these results, it can apparently play a significant role in the catalytic activity results achieved over the inverse catalyst (vide infra). The local structure of the CeO2/CuO catalyst during the WGS reaction was investigated using the PDF method. Figure 6 shows in situ PDF data for the CeO2/CuO catalyst exposed to a mixture of CO/H2O at different temperatures. The starting material has peaks for CuO at 1.95 (Cu-O distance) and 3.05 Å (Cu-Cu distance). As the temperature was increased, the CuO signals disappeared while a peak due to the Cu-Cu interaction in metallic copper appeared at 2.55 Å. This change accounts for the reduction of the Cu2+ cations to metallic Cu0, as seen in XRD and XANES. The Ce-Ce interatomic distance of ceria remained constant and the changes in the Ce-O distance were

3584


Barrio et al.

Figure 7. Production of CO2 and H2 during the WGS reaction on the inverse CeO2/CuO catalyst. These data were obtained in the same set of experiments that produced the XRD patterns displayed in Figure 4.

Figure 5. (A) Cu K-edge XANES spectra collected under WGS reaction conditions at the indicated temperatures, with a reference Cu0 foil spectrum. (B) Ce LIII-edge XANES collected under WGS conditions at the indicated temperatures, with reference Ce3+ and Ce4+ spectra.

Figure 6. Time-resolved PDF analysis of the inverse CeO2/CuO catalyst under WGS reaction conditions. The analysis was performed using the XRD patterns from Figure 4.

difficult to follow because it overlapped with the Cu-Cu distance of metallic copper. Above 250 °C, there were no additional changes in the local and long-range structures of the catalyst.

Figure 7 shows the product analysis obtained by mass spectrometry at the exit of a microreactor28 that contained the inverse CeO2/CuO catalyst. The activity onset (i.e., simultaneous formation of H2 and CO2) was observed at ∼200 °C, when the CeO2/CuO f CeOx/Cu transformation was taking place. As in the case of CuO/CeO2 and Ce1-xCuxO2 catalysts,16 the generation of reduced states of copper in the inverse catalyst accompanies the production of H2 through the water-gas shift reaction. High catalytic activity is observed at 300, 400, and 500 °C, temperatures at which the catalyst consists of CeOx/Cu. Noteworthy, the activity profile shows neither deactivation of the catalyst during long reaction times (∼4 h) nor any appreciable deactivation effect at a reaction temperature as high as 500 °C. C. Reduction of CeO2/CuO with CO and Reoxidation with O2. To understand better the redox process occurring in the CeO2/CuO catalyst during the water-gas shift reaction, we examined the interaction of CO with the catalyst. For this, the CeO2/CuO sample was exposed to a mixture of 5% CO/95% He and the temperature was ramped from 25 to 500 °C, monitoring changes in the X-ray diffraction pattern. Figure 8 summarizes the results obtained through a Rietveld refinement of the XRD data. A direct CuO f Cu transformation occurred at 80-150 °C, without the formation of Cu2O as an intermediate. When the CuO is being reduced, the expansion of the CeO2 lattice is linear due only to thermal effects. At higher temperatures (∼150 °C), there is a change in the slope of the ceria lattice expansion, indicating an increase in the reduction degree of the ceria species. Under a flow of CO, the reductions of CuO and CeO2 do not appear to correlate (in agreement with previous redox models for catalysts of this type),39 in contrast with the behavior observed under WGS conditions (bottom panel of Figure 4). The presence of water in the reaction mixture delays the reduction of CuO and makes possible the formation of Cu2O as a reaction intermediate. Figure 9 compares the cell dimensions of ceria in CeO2/CuO during temperature-programmed oxidation, reduction, and during the WGS process. The results correspond to a temperature of 500 °C, and the dashed line marks the cell dimension of pure ceria at that temperature. The ceria cell dimensions increased in the following sequence: O2 < WGS < CO. Because the cell parameter accounts directly for the ceria reduction degree,16,36,37


Figure 8. Rietveld refinement of mole fraction variation and lattice parameter changes of CeO2 as a function of temperature during the reduction of the CeO2/CuO inverse catalyst in a mixture of 5% CO/ 95% He. For comparison, we also include the variation in the lattice parameter of ceria due to thermal heating.

Figure 9. Comparison of ceria cell dimensions obtained at 500 °C during temperature-programmed oxidation, CO reduction, and during the WGS reaction. The dashed line denotes the cell dimension of pure ceria at 500 °C.

it can be concluded that the concentration of oxygen vacancies and Ce3+ cations during the WGS is smaller than after reduction in CO and larger than on a fully oxidized CeO2/CuO catalyst. During the WGS, in accordance with previous studies,7 the water molecules probably adsorb and dissociate on oxygen vacancies (Ovac) created upon interaction with CO

CO(gas) + CeO2(surf) f CO2(gas) + CeO2-Ovac(surf) (2) CeO2-Ovac(surf) + H2O(gas) f H2(gas) + CeO2(surf) (3) leading to a reduction in the concentration of Ce3+ cations and the cell dimensions of ceria. Considering the obtained results (Figures 4, 8, and 9), it can be proposed that CO will adsorb

J. Phys. Chem. C, Vol. 114, No. 8, 2010 3585 on the Cu sites and then migrate to reduce the ceria nanoparticles, where the water will dissociate. The CeOx/Cu catalyst could also be regenerated to its oxidized CeO2/CuO state by heating under an oxidizing atmosphere of 5% O2 in helium (see Figure 10). The catalyst regeneration cycle was performed several times, and the process was always reversible. Reaction of O2 with CeOx/Cu led to the removal of Ce3+ centers and a consequent contraction of the ceria cell dimensions. From the metallic copper phase, the catalyst evolved to a CuO phase, passing through the formation of a Cu2O intermediate. Cu K-edge XANES spectra were recoded during the reoxidation and are shown in Figure 10B. Under oxygen flow, the metallic copper is readily oxidized to CuO. The isosbestic points obtained along the spectra further confirm the phase transition of copper species. The XANES results indicate that only CuO was present after reoxidation. Activity data on oxidized and (in situ) reduced samples show no significant changes (see Figure 11), indicating that, under reaction conditions, the activity is governed by reduced species that are already present at 200 °C, as seen in in situ measurements with XRD, XANES, and PDF. The insensitivity of the catalyst performance to redox cycles is a promising characteristic for mobile fuel cell applications because the catalyst would be stable during multiple start-up/shutdown processes. This appears to be an interesting characteristic of the inverse CeO2/CuO catalyst from a practical point of view. Hydrogen production and hence activity always increase with temperature, and once the steady state is reached, the activity remains constant at a fixed temperature. This result is similar to that found for the WGS on CeOx/Cu(111)22 but differs from previous work for the WGS on Cu/CeO2 and Ce1-xCuxO2 systems where a deactivation was observed at high temperatures.7,16 Furthermore, the inverse CeO2/CuO catalyst also exhibits high stability for preferential oxidation of carbon monoxide (CO-PROX).40 D. WGS Activity of CuO, CuO/CeO2, and CeO2/CuO Catalysts. Nanoparticles or extended surfaces of CeO2 are very poor catalysts for the water-gas shift reaction.7,14 Figure 12 compares the catalytic activity of CuO, a conventional CuO/ CeO2 catalyst,7 and the inverse CeO2/CuO catalyst. The use of the same experimental conditions allows for a reliable comparison of reaction rates. Under reaction conditions, the CuO transforms into metallic copper.7,16 This system exhibits the lowest catalytic activity in Figure 12. Once ceria is added to the catalyst mixtures, there is a substantial improvement in the catalytic activity. From experimental studies with single crystals14,22,41 and density functional calculations (DF),42-44 it is known that the most difficult step for the water-gas shift on pure copper systems is the dissociation of water. In contrast, the CeOx/Cu system apparently constitutes a highly efficient bifunctional catalyst in which, according to the results obtained, CO adsorbs on copper and water dissociates on the partially reduced ceria nanoparticles. The XRD results in Figure 9 point to a water T CeOx interaction, as described in eq 3. Furthermore, DF calculations for the dissociation of H2O on a CeOx/Cu(111) surface gave an exothermic ∆E of -0.33 eV and point to an activation energy smaller than 0.35 eV.22 In contrast, the dissociation of water on pure Cu(111) is endothermic (∆E ) 0.2-0.6 eV) with a large activation energy (Ea ) 0.9-1.4 eV).43,44 This highlights the key role played by the CeOx nanoparticles supported on Cu. A recent article discusses the advantages of using an inverse CeOx/Pt system for catalyzing the water-gas shift reaction.45 When optimizing the performance of copper-based WGS catalysts, the major emphasis is usually in controlling the

3586


Barrio et al.

Figure 10. (A) TR-XRD pattern (λ ) 0.3184 Å) and (B) Cu K-edge TR-XANES spectra during reoxidation of the inverse catalyst in a 5% O2/95% He mixture.

Conclusions

Figure 11. Catalyst stability during several WGS reaction runs over the same sample.

Figure 12. Catalytic performance of the inverse catalyst vs reference samples (CuO and Cu/CeO2).

oxidation state and morphology of the copper within the catalytic system.7-15,18 Our results for CeOx/Cu indicate that an optimization of the physical and chemical properties of the oxide component is as important as the optimization of the properties of the metal component. Because ceria is active as a “support”,36,46,47 one must optimize its properties when designing WGS catalysts.

An inverse CeOx/CuO catalyst has been prepared by an inverse microemulsion approach. Characterization of the fresh material by XRD and TEM demonstrated the inverse nature of the sample with nanoparticles of ceria dispersed on a copper oxide support. In situ WGS reaction tests showed that the composition of the inverse CeOx/CuO catalyst changed under reaction conditions. TR-XRD patterns showed the evolution from CuO to metallic copper via a small amount of Cu2O intermediate. Pair distribution function analysis corroborated the XRD results and ruled out the presence of unidentified amorphous phases. Sequential Rietveld refinement showed an expansion of the CeO2 lattice that correlated with the copper reduction. This cell expansion is mainly ascribed to the partial reduction of Ce4+ to Ce3+. XANES spectra for the Cu K-edge region confirmed the reduction to metallic copper under WGS reaction conditions. Analysis of the Ce LIII-edge verified the partial reduction of the ceria nanoparticles to CeO2-x entities. The presence of oxygen vacancies in the ceria lattice is an important property of the active phase because it opens a new reaction path involving the dissociative interaction of water, which enhances the WGS activity on the inverse catalyst. Reduction of the sample under a 5%CO/He flow shows a direct transformation from CuO to Cu0, without formation of the Cu1+ intermediate. Reduction of the sample occurs at a much lower temperature than under WGS reaction conditions. In the absence of water, the reduction of CeOx is no longer correlated with the copper reduction and occurs at higher temperatures. Reoxidation of the sample was achieved by heating up to 500 °C in a 5% O2/He flow. Both XRD and XANES data show the reversibility of the system properties with respect to redox cycles. Furthermore, the sample reduction-oxidation can be conducted several times without any significant loss in activity, proving the excellent redox stability of the system. WGS activity tests for CuO, Cu/CeOx, and the inverse CeOx/ Cu system showed the importance of the presence of CeOx species on a copper surface. The inverse catalyst apparently operates as a bifunctional system in which the CO is adsorbed on Cu sites while water becomes dissociated over CeOx nanoparticles. The close interaction between copper and ceria allows for the oxygen transfer to the CO. Thus, the inverse CeOx/ Cu catalyst is a highly active and stable system with promising characteristics for mobile fuel cell applications.

Inverse CeO2-x/CuO-Cu Catalyst for the WGS Reaction Acknowledgment. N. Marinkovic and S. Khalid are gratefully acknowledged for their help carrying out the XANES experiments.The work at BNL was financed by the U.S. Department of Energy (DOE), Chemical Sciences Division (DEAC02-98CH10086). The National Synchrotron Light Source is supported by the Divisions of Materials and Chemical Sciences of the U.S.-DOE. L.B. acknowledges funding by the FP7 People program under the project Marie Curie IOF-219674. A.H. thanks the MICINN for a FPU grant under which his contribution to this work was made. Work at ICP-CSIC and U.C. was financed by Comunidad de Madrid (ENERCAM S-0505/ENE/000304) and MICINN (CTQ2006-15600/BQU and CTQ2009-14527/ BQU) projects to whom we are grateful. References and Notes (1) Spivey, J. J. Catal. Today 2005, 100, 171–180. (2) Ladebeck, J. R.; Wang, J. P. Catalyst development for water-gas shift In Handbook of Fuel Cells - Fundamentals, Technology, and Applications; Vielstich, W., Lamm, A., Gasteiger, H. A., Eds.; Fuel Cell Technology and Applications, Vol. 3; John Wiley & Sons, Ltd.: Chichester, U.K., 2003. (3) Wang, X. Q.; Rodriguez, J. A.; Hanson, J. C.; Gamarra, D.; Martinez-Arias, A.; Fernandez-Garcia, M. J. Phys. Chem. B 2005, 109, 19595–19603. (4) (a) Meunier, F. C.; Reid, D.; Goguet, A.; Shekhtman, S.; Hardacre, C.; Burch, R.; Deng, W.; Flytzani-Stephanopoulos, M. J. Catal. 2007, 247, 277–287. (b) Fu, Q.; Saltsburg, H.; Flytzani-Stephanopoulos, M. Science 2003, 301, 935–938. (5) Phatak, A. A.; Koryabkina, N.; Rai, S.; Ratts, J. L.; Ruettinger, W.; Farrauto, R. J.; Blau, G. E.; Delgass, W. N.; Ribeiro, F. H. Catal. Today 2007, 123, 224–234. (6) Jacobs, G.; Ricote, S.; Davis, B. H. Appl. Catal., A 2006, 302, 14– 21. (7) Wang, X. Q.; Rodriguez, J. A.; Hanson, J. C.; Gamarra, D.; Martinez-Arias, A.; Fernandez-Garcia, M. J. Phys. Chem. B 2006, 110, 428– 434. (8) Rodriguez, J. A.; Hanson, J. C.; Wen, W.; Wang, X.; Brito, J. L.; Martinez-Arias, A.; Fernandez-Garcia, M. Catal. Today 2009, 145, 188– 194. (9) Clausen, B. S.; Steffensen, G.; Fabius, B.; Villadsen, J.; Feidenhansl, R.; Topsøe, H. J. Catal. 1991, 132, 524. (10) Jacobs, G.; Chenu, E.; Patterson, P. M.; Williams, L.; Sparks, D.; Thomas, G.; Davis, B. H. Appl. Catal., A 2004, 258, 203–214. (11) Koryabkina, N. A.; Phatak, A. A.; Ruettinger, W. F.; Farrauto, R. J.; Ribeiro, F. H. J. Catal. 2003, 217, 233–239. (12) Fukuhara, C.; Ohkura, H.; Gonohe, K.; Igarashi, A. Appl. Catal., A 2005, 279, 195–203. (13) Martinez-Arias, A.; Gamarra, D.; Fernandez-Garcia, M.; Wang, X. Q.; Hanson, J. C.; Rodriguez, J. A. J. Catal. 2006, 240, 1–7. (14) Rodriguez, J. A.; Liu, P.; Hrbek, J.; Evans, J.; Perez, M. Angew. Chem., Int. Ed. 2007, 46, 1329–1332. (15) Ilinich, O.; Ruettinger, W.; Liu, X. S.; Farrauto, R. J. Catal. 2007, 247, 112–118. (16) Wang, X. Q.; Rodriguez, J. A.; Hanson, J. C.; Gamarra, D.; Martinez-Arias, A.; Fernandez-Garcia, M. Top. Catal. 2008, 49, 7. (17) Daly, H.; Ni, J.; Thompsett, D.; Meunier, F. C. J. Catal. 2008, 254, 238–243.

J. Phys. Chem. C, Vol. 114, No. 8, 2010 3587 (18) Meunier, F. C.; Goguet, A.; Hardacre, C.; Burch, R.; Thompsett, D. J. Catal. 2007, 252, 18–22. (19) Jacobs, G.; Patterson, P. M.; Williams, L.; Chenu, E.; Sparks, D.; Thomas, G.; Davis, B. H. Appl. Catal., A 2004, 262, 177–187. (20) Nakamura, J.; Campbell, J. M.; Campbell, C. T. J. Chem. Soc., Faraday Trans. 1990, 86, 2725–2734. (21) Rodriguez, J. A.; Ma, S.; Liu, P.; Hrbek, J.; Evans, J.; Perez, M. Science 2007, 318, 1757–1760. (22) Rodriguez, J. A.; Graciani, J.; Evans, J.; Park, J. B.; Yang, F.; Stacchiola, D.; Senanayake, S. D.; Ma, S.; Pe´rez, M.; Liu, P.; FernańdezSanz, J.; Hrbek, J. Angew. Chem., Int. Ed. 2009, 48, 8047–8050. (23) Norby, P.; Hanson, J. C. Catal. Today 1998, 39, 301. (24) Chupas, P. J.; Qiu, X. Y.; Hanson, J. C.; Lee, P. L.; Grey, C. P.; Billinge, S. J. L. J. Appl. Crystallogr. 2003, 36, 1342–1347. (25) Chupas, P. J.; Chapman, K. W.; Jennings, G.; Lee, P. L.; Grey, C. P. J. Am. Chem. Soc. 2007, 129, 13822. (26) Egami, T.; Billinge, S. J. L. UNDERNEATH THE BRAGG PEAKS: Structural Analysis of Complex Materials; Elsevier Ltd.: Oxford, U.K., 2003. (27) Gamarra, D.; Belver, C.; Fernańdez-Garcıá, M.; Martıńez-Arias, A. J. Am. Chem. Soc. 2007, 129, 12064. (28) Chupas, P.; Chapman, K.; Kurtz, C.; Hanson, J. C.; Lee, P.; Grey, C. P. J. Appl. Crystallogr. 2008, 41, 822. (29) Hammersley, A. P.; Svensson, S. O.; Thompson, A. Nucl. Instrum. Methods Phys. Res., Sect. A 1994, 346, 312. (30) (a) Toby, B. H. J. Appl. Crystallogr. 2001, 34, 210. (b) Larson, A. C.; von Dreele, R. B. GSAS General Structure Analysis System; Report LAUR 86-748; Los Alamos National Laboratory: Los Alamos, NM, 1995. (31) Qiu, X.; Thompson, J. W.; Billinge, S. J. L. J. Appl. Crystallogr. 2004, 37, 678–678. (32) (a) Ravel, B.; Newville, M. J. Synchrotron Radiat. 2005, 12, 537. (b) Newville, M.; Ravel, B.; Haskel, D.; Rehr, J. J.; Stern, E. A.; Yacoby, Y. Physica B 1995, 208-209, 154. (33) JCPDS Powder Diffraction File; International Centre for Diffraction Data: Newtown Square, PA, 1997; PDF No. 04-0593. (34) JCPDS Powder Diffraction File; International Centre for Diffraction Data: Newtown Square, PA, 1997; PDF No. 48-1548. (35) Scherrer, P. Go¨ttinger Nachrichten 1918, 2, 92. (36) Trovarelli, A. Catal. ReV.-Sci. Eng. 1996, 38, 439. (37) (a) Engler, B.; Koberstein, E.; Schubert, P. Appl. Catal. 1989, 48, 71–92. (b) Yao, H. C. Appl. Surf. Sci. 1984, 19, 398–406. (38) Wang, X.; Hanson, J. C.; Frenkel, A. I.; Kim, J.-W.; Rodriguez, J. A. J. Phys. Chem. B 2004, 108, 13667–13673. (39) Martıńez-Arias, A.; Hungrıá, A. B.; Fernańdez-Garcıá, M.; Conesa, J. C.; Munuera, G. J. Phys. Chem. B 2004, 108, 17983. (40) Horne´s, A.; Hungrıá, A. B.; Bera, P.; Lu¨pez Ca´mara, A.; FernańdezGarcıá, M.; Martıńez-Arias, A.; Barrio, L.; Estrella, M.; Zhou, G.; Fonseca, J. J.; Hanson, J. C.; Rodrı´guez, J. A. J. Am. Chem. Soc. 2010, 132, 34. (41) (a) Campbell, C. T.; Koel, B.; Daube, K. A. J. Vac. Sci. Technol., A 1987, 5, 810. (b) Nakamura, J.; Campbell, J. M.; Campbell, C. T. J. Chem. Soc., Faraday Trans. 1990, 86, 2725. (42) Liu, P.; Rodriguez, J. A. J. Chem. Phys. 2007, 126, 164705. (43) Gokhale, A. A.; Dumesic, J. A.; Mavrikakis, M. J. Am. Chem. Soc. 2008, 130, 1402. (44) Fajin, J. L. C.; Illas, F.; Gomes, J. R. B. J. Chem. Phys. 2009, 130, 224702. (45) Yeung, C. M. Y.; Tsang, S. C. J. Phys. Chem. C 2009, 113, 6074. (46) (a) Si, R.; Flytzani-Stephanopoulos, M. Angew. Chem., Int. Ed. 2008, 47, 2884. (b) Li, Y.; Fu, Q.; Flytzani-Stephanopoulos, M. Appl. Catal., B 2000, 27, 179. (47) Hilaire, S.; Wang, X.; Luo, T.; Gorte, R. J.; Wagner, J. Appl. Catal., A 2001, 215, 271.

JP910342B

In Situ Characterization of an Inverse - American Chemical Society

Recommend Documents