Intrinsic Relation between Catalytic Activity of CO Oxidation on Ru

Oct 15, 2012 - *E-mail: [email protected], [email protected]. Cite this:Nano ..... Hong Jiang. Advanced Sustainable Systems 2017 1 (12), 1700...
0 downloads 0 Views 2MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Intrinsic Relation between Catalytic Activity of CO Oxidation on Ru Nanoparticles and Ru Oxides Uncovered with Ambient Pressure XPS Kamran Qadir,† Sang Hoon Joo,‡ Bongjin S. Mun,§,∥ Derek R. Butcher,⊥ J. Russell Renzas,⊥ Funda Aksoy,¶,# Zhi Liu,¶ Gabor A. Somorjai,*,⊥ and Jeong Young Park*,† †

Graduate School of EEWS (WCU), and NanoCentury KI, KAIST, Daejeon 305-701, South Korea School of Nano-Bioscience and Chemical Engineering, KIER-UNIST Advanced Center for Energy, and Low Dimensional Carbon Materials Center, Ulsan National Institute of Science and Technology (UNIST), UNIST-gil 50, Ulsan 689-798, South Korea § Department of Applied Physics, Hanyang University ERICA, Ansan 426-791, South Korea ∥ Ertl Center for Electrochemistry and Catalysis, Gwangju Institute of Science and Technology, Gwangju 500-712, South Korea ⊥ Department of Chemistry, University of California, Berkeley, California 94720, United States ¶ Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States ‡

S Supporting Information *

ABSTRACT: Recent progress in colloidal synthesis of nanoparticles with well-controlled size, shape, and composition, together with development of in situ surface science characterization tools, such as ambient pressure X-ray photoelectron spectroscopy (APXPS), has generated new opportunities to unravel the surface structure of working catalysts. We report an APXPS study of Ru nanoparticles to investigate catalytically active species on Ru nanoparticles under oxidizing, reducing, and CO oxidation reaction conditions. The 2.8 and 6 nm Ru nanoparticle model catalysts were synthesized in the presence of poly(vinyl pyrrolidone) polymer capping agent and deposited onto a flat Si support as twodimensional arrays using the Langmuir−Blodgett deposition technique. Mild oxidative and reductive characteristics indicate the formation of surface oxide on the Ru nanoparticles, the thickness of which is found to be dependent on nanoparticle size. The larger 6 nm Ru nanoparticles were oxidized to a smaller extent than the smaller Ru 2.8 nm nanoparticles within the temperature range of 50−200 °C under reaction conditions, which appears to be correlated with the higher catalytic activity of the bigger nanoparticles. We found that the smaller Ru nanoparticles form bulk RuO2 on their surfaces, causing the lower catalytic activity. As the size of the nanoparticle increases, the core−shell type RuO2 becomes stable. Such in situ observations of Ru nanoparticles are useful in identifying the active state of the catalysts during use and, hence, may allow for rational catalyst designs for practical applications. KEYWORDS: CO oxidation, Ru nanoparticles, oxidation state, catalytic activity, nanoparticle size, ambient-pressure XPS

H

cells.6 Transition metal single crystals and supported nanoparticles, such as Ru, Rh, Pt, Pd, and Au, are extensively studied for this reaction due to their high catalytic activity and stability.2,7−11 These model systems, such as single crystals, have been thoroughly investigated under highly oxidizing conditions and it has been found that an ultrathin surface oxide forms prior to the onset of bulk oxidation.12 This thin surface oxide is reported to be more active under CO oxidation than corresponding metallic surfaces, which has been the subject of numerous experimental and theoretical investigations.13−24 The nature of the active surface, however, for these metals under catalytic CO oxidation still remains unresolved.

eterogeneous catalysis is of pivotal importance because of its relevance in industrial chemical manufacturing. In addition to activity and selectivity, one of the key issues in practical catalysis is the deactivation of catalysts during use. If we are able to understand the molecular factors governing the phenomenon of deactivation, we can design a catalyst with minimal loss of activity and hence avoid its replacement, significantly reducing operational costs. Industrially relevant catalysts are, however, too complex to allow for such an understanding at molecular scale. It is for this reason that model catalytic systems are being used to investigate deactivation.1−4 CO oxidation is an extensively studied heterogeneous reaction since it is crucially important both technologically and scientifically, such as for removal of CO from automobile exhaust streams5 and purification of hydrogen through preferential oxidation in polymer electrolyte membrane fuel © 2012 American Chemical Society

Received: August 17, 2012 Revised: October 2, 2012 Published: October 15, 2012 5761

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

conditions. We show that the smaller Ru nanoparticles form bulk RuO2 on the surface, which is responsible for the lower catalytic activity. Ru3d spectra of 2.8 and 6 nm Ru nanoparticles were acquired during in situ oxidation under 200 mTorr O2 and during in situ reduction under 80 mTorr CO. Figure 1a

For the case of Pd(111), two-dimensional oxides were observed, even at low oxygen pressures (as intermediate phases between an oxygen overlayer and bulk PdO).25 Further scanning tunneling microscopy (STM), surface X-ray diffraction (SXRD), X-ray photoelectron spectroscopy (XPS), and density functional theory (DFT) studies26 suggested the formation of an incommensurate surface oxide of Pd5O4 stoichiometry and almost coplanar geometry that has no resemblance to any bulk oxides of Pd.27 As in the case of Pt(110), the in situ STM studies28 at atmospheric pressures resulted in morphological changes from adsorbate-covered Pd(100) to an oxidic state depending on the CO/O2 ratios. However, unlike the Pt(110) surface, where roughening was induced by reaction, the Pd(100) surface became rough during oxidation. The CO oxidation was accompanied by further surface roughening. Such an oxidized surface showed a considerably higher reactivity.27 For the case of nanoparticles, it has been demonstrated that small palladium nanoparticles7 are more active for CO oxidation than larger particles and single crystals, whereas the opposite is reported for platinum nanoparticles.29 Well-ordered, ultrathin oxide films are reported to form on Rh surfaces. The combination of high-resolution XPS, STM, SXRD, and DFT revealed the self-limited growth of an O−Rh− O trilayer film on Rh(111) at intermediate oxygen pressures.30 Recently, combined SXRD and reactivity (mass spectrometry) studies of the Rh(111) surface provided strong evidence that the trilayer surface oxide is much more active than metallic Rh in low-temperature (∼500 K) CO oxidation, whereas the Rh bulk oxide was not active at all.27,31 For Rh nanoparticles, it has been shown using ambient pressure X-ray photoelectron spectroscopy (APXPS) that smaller 2 nm nanoparticles oxidize to a larger extent than 7 nm nanoparticles during reaction at 150−200 °C. This thick oxide is correlated with a 5-fold increase in turnover frequency for the smaller nanoparticles during catalytic CO oxidation.15 Among these transition metals, Ru exhibits unique catalytic characteristics under CO oxidation. The Ru single crystal surface is catalytically least active under UHV conditions among the late transition metals; however, under high pressure and oxidizing reaction conditions, Ru becomes much more active than the other late transition metals.27,32 It has been reported that under realistic CO oxidation conditions, ultrathin RuO2 (∼1 nm thick) is formed on the Ru(0001) and Ru(101̅0) metal surfaces, which is extremely active for CO oxidation.33 Assmann et al. proposed a core−shell structure for Ru polycrystalline micrometer-scale powders and supported nanoparticles where the shell is an ultrathin active RuO2 covering around a Ru metal core. The stability of such core−shell structures appears to increase with the size of the nanoparticles.1,13,34 Joo et al. reported the trend of increasing catalytic activity under CO oxidation for Ru nanoparticles within the 2−7 nm size range.35 They correlated the size effect of catalytic activity with the stability of the core−shell structure. In this study, we utilize in situ APXPS36 to find an intrinsic relation between the catalytic activity of CO oxidation on Ru nanoparticles and the stability of the Ru oxides. We have investigated 2.8 and 6 nm Ru nanoparticles synthesized using the polyol reduction method. Catalytic oxidation, reduction, and CO oxidation were carried out on Ru nanoparticle arrays and the surface oxidation states were measured and monitored using APXPS in order to understand the relationship between the oxidation states and catalytic activity under realistic

Figure 1. (a) Ru3d spectra showing in situ oxidation (under 200 mTorr O2) and reduction (under 80 mTorr CO) for 6 nm Ru nanoparticles with increasing temperature. (b) Ru3d spectra showing reversibility of oxidation (under 200 mTorr O2) and reduction (under 80 mTorr CO) of 6 nm Ru nanoparticles at 200 °C during in situ oxidation/reduction cycles. Incident photon energy is 440 eV.

represents the state of the 6 nm Ru nanoparticles during such in situ gas conditions between 100 and 200 °C. The incident Xray photon energy is 440 eV in this case. The Ru3d spectra show that during reduction, the Ru (0) peak (from the Ru metallic surface) at the binding energy of 279.8 eV shows a progressive increase, and at 200 °C the metallic surface of the Ru nanoparticles has negligible surface oxide. After the nanoparticles were reduced, we introduced 200 mTorr O2 into the chamber to monitor in situ oxide formation on the nanoparticles. The Ru3d spectra are acquired during oxidation between 100 and 200 °C. Ru (4+) of the Ru oxide (RuO2) peak is marked at the binding energy of 280.7 eV. At 100 °C, the Ru nanoparticles show significant oxide formation, as can be seen from the Ru (4+) oxide peak enhancement in relation to the Ru (0) metal peak. As the temperature increases further, the peak of the Ru (4+) grows even more dominant. At 200 °C, the Ru (4+) peak dominates and a very small peak of Ru (0) remains. Thus, the nanoparticles show significant surface oxidation with small emissions from the underlying metal core. To assess reversibility of the oxide formed on the nanoparticles, the nanoparticles were subjected to cycles of oxidation and reduction. The reaction conditions were as described above. Figure 1b shows Ru3d spectra, each recorded at 200 °C for cyclic oxidation and reduction. Emissions from Ru metal atoms in the underlying core are marked by the Ru (0) peak and emissions from Ru atoms in the surface oxide are marked by the Ru (4+) peak, as mentioned above. During oxidation at 200 °C, Ru is predominantly oxidized and has a very small Ru (0) peak, as is clear from the first oxidation spectra of Ru3d. During reduction, however, the metallic peak of the Ru nanoparticles has negligible oxide, as exhibited by the Ru3d spectra acquired during reduction at 200 °C. Upon reoxidation, the Ru nanoparticles oxidize in a similar manner as before; re-reducing the nanoparticles in CO results in exposure of the metallic surface of the nanoparticles. This shows that the oxide formed on the Ru nanoparticles is reversible. The nanoparticles show progressive growth of the oxide over the 5762

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

ones, which can be confirmed by the comparatively higher surface oxide peak areas of the smaller nanoparticles. During purely oxidizing conditions (under 200 mTorr O2), RuO2 growth was monitored during its in situ evolution. For this purpose, Ru3p spectra were acquired at 100, 150, and 200 °C at an incident photon energy of 650 eV. Figure 3 shows a

metallic surface during net oxidizing conditions and such an oxide is removed from the Ru nanoparticles during net reducing conditions, resulting in metallic facets. Even when the nanoparticles are reduced at higher temperatures, the C1s peak at 285.5 eV still remains. This peak arises due to the PVP capping around the nanoparticles.15,37 The C1s peak overlaps with the Ru3d oxide peak and deconvolusion is not performed using the Ru3d spectra. These carbon peaks are associated with carbonaceous species that were present on the nanoparticle and silicon surfaces. The carbonaceous species result from thermal decomposition of PVP [poly(vinylpyrrolidone)] capping layers during catalytic reactions. It was shown that polyamide− polyene-like structures were formed after heating PVP-capped Pt and Rh nanoparticles above 200 °C.38 Therefore, we suppose that similar polyamide−polyene-like carbonaceous species are present on the nanoparticle surface. We note that the peak area of C1s remains unchanged during the series of reduction and oxidation experiments, indicating that the carbon species present on the nanoparticle surface do not affect the oxidation/reduction behavior of the nanoparticles. In order to monitor the trend of oxidation with increasing nanoparticle size, Ru3p spectra were acquired under oxidation, reduction, and CO oxidation conditions for 2.8 and 6 nm Ru nanoparticle arrays. After peak deconvolusion, Ru3p spectra were fitted using three peaks. The peak at 461.2 eV has been attributed to Ru (0) from metal atoms, while those at 463.2 and 465.9 eV have been assigned to RuO 2 and RuO x , respectively.8,39−42 Figure 2 shows Ru3p spectra under

Figure 3. APXPS spectra of Ru3p of Ru nanoparticles under oxidation (200 mTorr O2) with increasing temperature. (a) Initially, the 6 nm Ru nanoparticles were in a reduced state, exposing the metallic surface (under 80 mTorr CO). Ru3p XPS of 6 nm Ru under 200 mTorr O2 acquired at (b) 100, (c) 150, and (d) 200 °C. (e) Initially, the 2.8 nm Ru nanoparticles were in a reduced state, exposing the metallic surface (under 80 mTorr CO). Ru3p XPS of 6 nm Ru under 200 mTorr O2 acquired at (f) 100, (g) 150, and (h) 200 °C (incident photon energy is 650 eV).

series of Ru3p spectra under such conditions. Initially under 80 mTorr CO, the metallic surface of the Ru nanoparticles is exposed, as observed previously. After introducing 200 mTorr O2 and heating at higher temperatures (100−200 °C), the Ru nanoparticles show oxide formation around the metallic core, and the peaks attributed to RuO2 and RuOx show significant growth with increasing temperature. In particular, there is a notable increase in the oxide formed between 100 and 200 °C. The smaller 2.8 nm Ru nanoparticles show more oxide formation and this is most evident in the Ru3p spectra when both 2.8 and 6 nm Ru nanoparticles are heated under 200 mTorr O2 at 200 °C. This finding is in agreement with intuition that smaller nanoparticles would be subjected to a greater degree of oxidation than larger ones due to their small size and high surface to volume ratio. We believe that a thick surface oxide (bulk RuO2) is formed as the nanoparticle size decreases. Therefore, increasing the nanoparticle size increases the stability of the thin surface oxide, as is shown for the larger 6 nm nanoparticles. Next, we subjected the nanoparticles to the catalytic CO oxidation reaction under net oxidizing conditions using 200 mTorr O2 and 80 mTorr CO (CO/O2: 0.4) to assess the catalytically active oxide species during reaction conditions formed in situ on the nanoparticles. This allows us to correlate the observed trend of increasing catalytic activity with the nature of the oxide formed around the nanoparticles. Previously, it was reported that the catalytic activity of

Figure 2. APXPS spectra of Ru3p of Ru nanoparticles under reducing conditions (80 mTorr CO) with increasing temperature showing spectra of 6 nm Ru nanoparticle acquired at (a) 100, (b) 150, and (c) 200 °C, and those of 2.8 nm Ru nanoparticle acquired at (d) 100, (e) 150, and (f) 200 °C (incident photon energy is 650 eV).

reducing reaction conditions of 80 mTorr CO at 100, 150, and 200 °C at an incident photon energy of 650 eV. At lower reducing temperatures, the Ru3p spectra show clear enhancement of the surface oxide peaks. As the temperature rises, however, the surface oxide is reduced, which exposes the metallic Ru surface on the nanoparticles. This observation is similar to our findings from the Ru3d spectra acquired under the same conditions. It was found that the smaller 2.8 nm nanoparticles underwent more oxidation than the larger 6 nm 5763

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

observed the smaller nanoparticles being more oxidized than the larger ones, particularly at high temperatures. To summarize the observations obtained thus far from the Ru3p spectra, the peak area ratios of RuO2 in relation to emissions from Ru metal are calculated and plotted with increasing temperature under reducing, oxidizing, and CO oxidation reaction conditions. Figure 5 shows such ratios calculated from Ru3p under (a) reducing and (b) oxidizing conditions. When initially reducing

nanoparticles increases with increasing size within the 2−6 nm size range.35 Figure 4 shows APXPS spectra of Ru3p acquired at various temperatures during catalytic CO oxidation. At the lower

Figure 5. Peak area ratios of the RuO2 peak with respect to the Ru metallic peak (Ru4+/Ru0), as calculated from APXPS spectra of Ru3p. (a) Ru nanoparticles under reducing conditions (80 mTorr CO) and (b) under oxidizing conditions (200 mTorr O2). The plot shows a trend of oxidation at various temperatures for the nanoparticles. The smaller nanoparticles exhibit higher ratios due to the high proportion of surface oxide.

Figure 4. APXPS spectra of Ru3p of Ru nanoparticles under CO oxidation (200 mT O2 and 80 mT CO) with increasing temperature, showing spectra of 6 nm Ru nanoparticles acquired at (a) 50, (b) 100, (c) 150, and (d) 200 °C, and spectra of 2.8 nm Ru nanoparticle acquired at (e) 50, (f) 100, (g) 150, and (h) 200 °C (incident photon energy is 650 eV). Ru3p spectra show a thicker oxide formation around the smaller 2.8 nm Ru nanoparticles.

the Ru nanoparticles in 80 mTorr CO at 100 °C, the RuO2/Ru ratio is 0.50 and 0.63 for 6 and 2.8 nm Ru nanoparticles, respectively. The ratios drop to 0.06 and 0.12 for 6 and 2.8 nm Ru nanoparticles, respectively, at the final reduction temperature of 200 °C. This decrease implies our previous observation of nanoparticles being reduced and exposing the metallic surface. On the other hand, the higher ratios for the smaller nanoparticles imply a higher proportion of surface oxide present on the smaller nanoparticles at all temperatures. After reducing the nanoparticles under 80 mTorr CO, the ratios are plotted with increasing temperature under 200 mTorr O2 to monitor the trend of in situ oxide formation for both 2.8 and 6 nm nanoparticles (see Figure 5b). At the oxidation temperature of 100 °C, both 2.8 and 6 nm Ru nanoparticles have ratios of 0.56 and 0.44, respectively. Thus, we see the smaller nanoparticles becoming more oxidized, as discussed above. At the oxidation temperature of 200 °C, there is a significant increase in oxide ratios for both sizes of nanoparticle and the values increase to 1.22 and 0.64 for 2.8 and 6 nm Ru, respectively. This clearly shows that at higher temperatures, the smaller 2.8 nm nanoparticles show a significantly higher degree of oxidation than the 6 nm nanoparticles and, thus, bulk oxide formation takes place as we reduce the nanoparticle size, particularly at elevated temperatures. In order to monitor oxide formation under catalytic CO oxidation and establish its correlation with catalytic activity of the Ru nanoparticles, RuO2/Ru ratios were plotted with increasing temperature using Ru3p spectra acquired in situ during the reaction. Figure 6 shows the resulting trend of oxidation for both 2.8 and 6 nm Ru nanoparticles. Initially, the RuO2/Ru ratios are 0.35 and 0.24 for the 2.8 and 6 nm nanoparticles, respectively, at 50 °C. As the temperature rises, there is an increase in the RuO2/Ru ratios due to growth of the surface oxide on both sizes of nanoparticle. At the higher CO

reaction temperature of 50 °C, both the 2.8 and 6 nm Ru nanoparticles show mild surface oxide formation. Both of the RuO2/Ru4+and RuOx/RuX+ oxide peaks are very small. The oxide, however, grows progressively as the temperature is raised, shown by an increase in the oxide peak areas. In particular, there is significant surface oxidation between 100− 200 °C for both sizes of nanoparticle. Consequently, both RuO2/Ru4+and RuOx/RuX+ oxide peaks grow dominant in the Ru3p spectra. The thickness dependence of the surface oxide on nanoparticle size is also evident from the Ru3p spectra. During the course of CO oxidation over nanoparticles at various temperatures, the smaller 2.8 nm Ru nanoparticles show a higher proportion of surface oxide formation. This is evident by the higher oxide peaks for the smaller nanoparticles. There is a clear distinction observed in the Ru3p spectra at 200 °C during CO oxidation. Ru3p shows more significant oxide formation on the smaller nanoparticles than on the larger ones. Consequently, a greater enhancement of the surface oxide peak is observed for the smaller nanoparticles. Thus, it is clear that under purely oxidizing conditions (as discussed) and under net oxidizing conditions (CO/O2: 0.4), the smaller nanoparticles show greater oxidation than the larger ones. Thus, we confirm bulk oxide (RuO2) formation on the smaller nanoparticles and that the thin surface oxide becomes stable as the nanoparticle size increases. From the Ru3p spectra discussed above, we observed that under reducing conditions the metallic surface of the Ru nanoparticles is exposed, and when subjected to both oxidizing and CO oxidation reaction conditions the nanoparticles are progressively oxidized. The degree of oxidation is dependent on both temperature and nanoparticle size. Increasing the temperature resulted in higher oxidation, as expected. We 5764

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

We deconvoluted the O1s spectra (see Figure 7a) for analysis of the RuO2 peak shoulder. After peak fitting, we observed three distinct peaks for RuO2. The peak at 531.50 eV is attributed to absorbed O/OH. The peaks at 529.5, 528.7, and 528.13 eV have been assigned to bulk O2− of RuO2, bridge O, and on-top O at the surface of the RuO2, respectively. Under net oxidizing reaction conditions during CO oxidation, the RuO2 (110) surface has bridging O atoms and on-top O atoms as the catalytically active O species that form CO2. On-top O atoms are less weakly bound to the RuO2 surface than bridging O atoms by 150 kJ mol−1. It is to be noted that the bridge oxygen feature is well-defined only for the well-ordered Ru oxide structure.1,43,44 Thus, it is suggested that a well-structured Ru oxide is formed around the nanoparticles. One of the key features of APXPS is the observation of gas phase peaks, as can be seen in O1s spectra acquired in situ. Such unique information cannot be seen in UHV-based experiments and allows us to determine the gas phase composition, such as catalytic turnover from the Ru oxide surface indicated by the CO2 gas phase peaks. Such peaks are observed alongside the core level peaks because the incident photon beam irradiates not only the Ru nanoparticle surface, but also part of the gas phase in front of the sample. Gas phase peaks are observed alongside the surface peaks if the partial pressures of the gas phase species are above ∼0.05 Torr.15,43,45 The binding energy shift between the gas phase and surface species is usually large enough to separate these contributions in the spectra. Under reducing conditions, the gas phase CO peak is observed alongside the nanoparticle surface peaks; while under oxidizing conditions, the gas phase CO and O2 peaks are observed, as shown in Figure 7b. After reviewing the O1s spectra under oxidizing conditions, we also assessed O1s spectra taken during catalytic CO oxidation. This in turn not only allows us to determine the catalytic activity of the surface oxide on the nanoparticles due to emissions from the gas phase CO2 product molecules present in the reaction cell but also sheds light on the nature of the oxide itself. Figure 8 shows a series of O1s spectra at various temperatures during catalytic CO oxidation on 6 nm Ru nanoparticles. As before, O1s (a) after peak deconvolusion and (b) before peak deconvolusion are presented. As seen in Figure 8b, there is a negligible Ru oxide peak shoulder at 50 °C and no

Figure 6. Peak area ratios of the RuO2 peak with respect to the Ru metallic Peak (Ru4+/Ru0), as calculated from APXPS spectra of Ru3p for Ru nanoparticles under CO oxidation reaction conditions of 200 mTorr O2 and 80 mTorr CO (CO/O2:0.4).

oxidation temperature of 200 °C, the RuO2/Ru ratios have final values of 1.55 and 1.10 for 2.8 and 6 nm Ru nanoparticles, respectively. These values are much higher than the initial value at 50 °C, which shows an increase in RuO2 formation at elevated temperatures. Moreover, the ratios for the smaller nanoparticles are higher than for the larger Ru nanoparticles due to the high degree of oxidation during the CO oxidation reaction. This is similar to what we observe when the nanoparticles are oxidized under oxygen only. In order to follow in situ evolution of the Ru oxide and to investigate the nature of the oxide, we carefully assessed corresponding O1s spectra acquired together with Ru3d and Ru3p core level spectra with increasing temperature. Figure 7

Figure 7. APXPS spectra of O1s acquired in situ during oxidation of 6 nm Ru nanoparticles (using 200 mTorr O2) at various temperatures. It shows the nanoparticles progressively forming surface oxide, the nature of which is well ordered.

shows a series of O1s spectra both (a) after and (b) before peak fitting at various temperatures under oxidizing conditions of 200 mTorr O2 for 6 nm Ru nanoparticles. Figure 7b shows the state of the Ru nanoparticles as they form surface oxide. Initially, the Ru nanoparticles under 80 mTorr CO at a higher reduction temperature of 200 °C show negligible oxide and O1s shows a very small surface oxide peak. After introducing 200 mTorr O2 and heating to 100 °C, O1s shows a surface oxide peak shoulder, consistent with our observations in the Ru3p and Ru3d spectra and hence signaling the formation of RuO2. As the temperature increases to 200 °C, the surface oxide peak shoulder grows in intensity, showing a higher degree of oxidation at higher temperatures.

Figure 8. APXPS spectra of O1s acquired in situ during CO oxidation of 6 nm Ru nanoparticles (using 80 mTorr CO and 200 mTorr O2) at various temperatures. During CO oxidation, the Ru nanoparticles show progressive, well-ordered surface oxide formation with increasing temperature. 5765

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

nanoparticles with decreasing nanoparticle size is correlated with the formation of the inactive bulk oxide layer on the Ru metallic core. These results are consistent with the previous reports13,17,23,35,46 and show that the stability of core−shell nanoparticles with a thin shell of active oxide increases as the nanoparticle size increases. Conclusion. In this study, uniform size-controlled Ru nanoparticles, 2.8 and 6 nm in size, were synthesized by the polyol method. Catalytic oxidation, reduction, and CO oxidation were carried out on Ru nanoparticle arrays and the surface oxidation states were measured and monitored using APXPS. We found that the smaller Ru nanoparticles form bulk RuO2 on the surface, which is responsible for the reduced catalytic activity. As the size of the nanoparticles increases, the core−shell type RuO2 becomes stable. The observed trend of increasing catalytic activity of Ru nanoparticles under CO oxidation can therefore be linked to the high stability of the active ultrathin surface oxide formed on the bigger nanoparticles. Our study shows a unique way of investigating catalyst surfaces under realistic conditions. Thus, we can closely monitor the changes occurring in real time during each reaction step. This, in turn, will allow for a more rational design of catalysts with increased activity and stability of the active species against deactivation, and allow us to directly apply the knowledge gained to technological catalysts. Methods. Synthesis of Ru Nanoparticles and Preparation of Langmuir−Blodgett Films. Colloidally monodispersed Ru nanoparticles were synthesized using the one-step polyol reduction method. Details of this synthesis are introduced elsewhere.35 Briefly put, the Ru nanoparticles were synthesized using Ru(acac)3 precursor with PVP as the surface capping agent. The ∼2.8 nm Ru nanoparticles were synthesized using ethylene glycol, while the ∼6 nm Ru nanoparticles were synthesized using butanediol solvent by the seeded growth method where ∼3.1 nm Ru nanoparticles were used as the seeds. Monolayers of as-synthesized Ru nanoparticles were formed by dropping the chloroform solution containing the Ru nanoparticles onto the water subphase of a LB trough (611 Nima technology) at room temperature (RT). The surface pressure was monitored with a Wilhelmy plate and was adjusted to zero before spreading the nanoparticles. The resulting layer of nanoparticles on water was compressed at a rate of 30 cm2/ min. Two-dimensional Ru nanoparticle arrays were obtained by lifting the submerged silicon wafer from the water, as shown in Figure S1 (See Supporting Information). Ambient Pressure X-ray Photoelectron Spectroscopy. Xray photoelectron spectroscopy measurements were performed at the Beamline 9.3.2 at the Advanced Light Source (ALS), Lawrence Berkeley National Lab.36 The experimental procedure for each sample consisted of using 200 mTorr O2 and heating the sample up to 200 °C, followed by cooling and evacuating the chamber (ca. 10−8 Torr). CO was then introduced (80 mTorr) and spectra were acquired at 100, 150, and 200 °C, followed by cooling and evacuation as before. Afterward, 200 mTorr O2 and 80 mTorr CO were introduced and spectra were acquired as before. Finally, following the cooling and evacuation procedures as before, 200 mTorr O2 was introduced and heating was repeated and spectra were acquired from RT up to 200 °C. Ru3d, Ru3p, and O1s spectra were acquired in situ at various temperatures under oxidizing conditions (200 mTorr O2), reducing conditions (80 mTorr CO), and CO oxidation reaction conditions (200 mTorr O2/80 mTorr CO). All of the spectra were calibrated using the Fermi

CO2 gas phase peak can be seen. This is confirmation of the fact that the Ru oxide is a catalytically active species under CO oxidation. On the other hand, at the same temperature, there are relatively high intensity CO and O2 gas phase peaks present in the spectra. As the reaction temperature rises, there is a corresponding rise in the RuO2 peak shoulder in O1s, as seen before under O2 only. In parallel to the RuO2 peak, the CO2 gas phase peak also appears showing that RuO2 is a catalytically active species during in situ CO oxidation conditions. Alternatively, the CO/O2 gas phase peaks diminish at 200 °C when compared to 50 °C, due to formation of CO2 molecules. Figure 8a shows O1s after peak deconvolusion. Similarly, the analysis shows three distinct Ru oxide features, as discussed for O1s under oxidizing conditions using 200 mTorr O2. In this case, the nanoparticles are undergoing less oxidation since the oxide formed is consumed during the ongoing reaction. This further validates that the Ru oxide formed is a catalytically reactive species under CO oxidation. O1s shows a small RuO2 peak for the 6 nm Ru nanoparticles, which suggest a thin shell of active oxide around the Ru metal core. The Ru nanoparticles exhibit a trend of increasing catalytic activity with increasing Ru nanoparticle size for CO oxidation under high pressure and oxidizing reaction conditions. Joo et al. reported that 6 nm Ru nanoparticles show an 8-fold higher activity than 2.1 nm Ru nanoparticles at 240 °C. CO oxidation is found to be structure insensitive on Rh and Ru single crystals, which implies that the size dependence of CO oxidation observed for Ru nanoparticles has an origin other than on the single crystal surface. The size dependence of the nanoparticles on CO oxidation is attributed to various molecular scale factors; in particular, an active surface oxide is suggested as the catalytically active species. Assmann et al. investigated catalytically active species and structural deactivation phenomena for Ru catalysts, such as bulk single crystals, micrometer-scale powders, and supported Ru nanoparticles under catalytic CO oxidation.1 They reported that an ultrathin surface oxide layer of RuO2 forms on the Ru metal (∼1−2 nm thick), which is extremely active in oxidizing CO. Through STM investigations, they showed structural deactivation of this oxide layer. Initially, micrometer-scale RuO2 powder with low energy RuO2 (110) and RuO2 (100) facets were exposed to both oxidizing and reducing reaction conditions. The degree of surface oxidation was determined by the CO/O2 reactant feed ratio and the temperature. They found that under net oxidizing conditions, (CO/O2 < 2) inactive RuO2 (100)-(2 × 2) surface facets are formed; while under net reducing conditions, low-activity metallic ruthenium surfaces (Ru(0001)−O) are exposed. To increase the stability of the active RuO2 overlayer, they activated Ru powder catalysts by reduction in H2 at 750 K and mildly reoxidizing the surface. They proposed a core−shell model for this active state of the Ru catalyst, where the most active state is an ultrathin RuO2 (thickness 1−2 nm) layer supported on a metallic Ru core. This active oxide overlayer gradually becomes inactive as the RuO2 layer grows thicker than ∼1−2 nm under oxidizing conditions. Such a thick oxide becomes roughened and gradually looses its catalytic properties.13 We observed that the smaller Ru nanoparticles are subjected to a higher degree of oxidation than the larger ones (see Figure 8). We confirm from our APXPS findings that smaller nanoparticles are forming inactive bulk oxide on their surfaces and, hence, a larger amount of catalytically inactive species is exposed. Thus, the decrease in catalytic activity of the Ru 5766

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

(12) Lundgren, E.; Mikkelsen, A.; Andersen, J. N.; Kresse, G.; Schmid, M.; Varga, P. J. Phys.: Condens. Matter 2006, 18, R481. (13) Assmann, J.; Crihan, D.; Knapp, M.; Lundgren, E.; Loffler, E.; Muhler, M.; Narkhede, V.; Over, H.; Schmid, M.; Seitsonen, A. P.; Varga, P. Angew. Chem., Int. Ed. 2005, 44, 917. (14) Carlsson, P. A.; Zhdanov, V. P.; Skoglundh, M. Phys. Chem. Chem. Phys. 2006, 8, 2703. (15) Grass, M. E.; Zhang, Y. W.; Butcher, D. R.; Park, J. Y.; Li, Y. M.; Bluhm, H.; Bratlie, K. M.; Zhang, T. F.; Somorjai, G. A. Angew. Chem., Int. Ed. 2008, 47, 8893. (16) Gustafson, J.; Westerstrom, R.; Balmes, O.; Resta, A.; van Rijn, R.; Torrelles, X.; Herbschleb, C. T.; Frenken, J. W. M.; Lundgren, E. J. Phys. Chem. C 2010, 114, 4580. (17) He, Y. B.; Knapp, M.; Lundgren, E.; Over, H. J. Phys. Chem. B 2005, 109, 21825. (18) Kim, Y. D.; Over, H.; Krabbes, G.; Ertl, G. Top. Catal. 2001, 14, 95. (19) McClure, S. M.; Goodman, D. W. Chem. Phys. Lett. 2009, 469, 1. (20) Rogal, J.; Reuter, K.; Scheffler, M. Phys. Rev. Lett. 2007, 98, 046101. (21) Vang, R. T.; Laegsgaard, E.; Besenbacher, F. Phys. Chem. Chem. Phys. 2007, 9, 3460. (22) Park, J. Y.; Aliaga, C.; Renzas, J. R.; Lee, H.; Somorjai, G. A. Catal. Lett. 2009, 129, 1. (23) Kim, S.; Qadir, K.; Jin, S.; Reddy, A. S.; Seo, B.; Mun, B. S.; Joo, S. H.; Park, J. Y. Catal. Today 2012, 185, 131. (24) Westerstrom, R.; Wang, J. G.; Ackermann, M. D.; Gustafson, J.; Resta, A.; Mikkelsen, A.; Andersen, J. N.; Lundgren, E.; Balmes, O.; Torrelles, X.; Frenken, J. W. M.; Hammer, B. J. Phys.: Condens. Matter 2008, 20, 184018. (25) Zheng, G.; Altman, E. I. Surf. Sci. 2002, 504, 253. (26) Lundgren, E.; Kresse, G.; Klein, C.; Borg, M.; Andersen, J. N.; De Santis, M.; Gauthier, Y.; Konvicka, C.; Schmid, M.; Varga, P. Phys. Rev. Lett. 2002, 88, 246103. (27) Shaikhutdinov, S.; Freund, H. J. Annu. Rev. Phys. Chem. 2012, 63, 619−633. (28) Hendriksen, B. L. M.; Bobaru, S. C.; Frenken, J. W. M. Surf. Sci. 2004, 552, 229. (29) Herskowitz, M.; Holliday, R.; Cutlip, M. B.; Kenney, C. N. J. Catal. 1982, 74, 408. (30) Gustafson, J.; Mikkelsen, A.; Borg, M.; Lundgren, E.; Kohler, L.; Kresse, G.; Schmid, M.; Varga, P.; Yuhara, J.; Torrelles, X.; Quiros, C.; Andersen, J. N. Phys. Rev. Lett. 2004, 92, 126102. (31) Gustafson, J.; Westerstroem, R.; Mikkelsen, A.; Torrelles, X.; Balmes, O.; Bovet, N.; Andersen, J. N.; Baddeley, C. J.; Lundgren, E. Phys. Rev. B 2008, 78, 045423. (32) Peden, C. H. F.; Goodman, D. W. J. Phys. Chem. 1986, 90, 1360. (33) Over, H.; Kim, Y. D.; Seitsonen, A. P.; Wendt, S.; Lundgren, E.; Schmid, M.; Varga, P.; Morgante, A.; Ertl, G. Science 2000, 287, 1474. (34) Assmann, J.; Narkhede, V.; Khodeir, L.; Loffler, E.; Hinrichsen, O.; Birkner, A.; Over, H.; Muhler, M. J. Phys. Chem. B 2004, 108, 14634. (35) Joo, S. H.; Park, J. Y.; Renzas, J. R.; Butcher, D. R.; Huang, W. Y.; Somorjai, G. A. Nano Lett. 2010, 10, 2709. (36) Grass, M. E.; Karlsson, P. G.; Aksoy, F.; Lundqvist, M.; Wannberg, B.; Mun, B. S.; Hussain, Z.; Liu, Z. Rev. Sci. Instrum. 2010, 81, 053106. (37) Moulder, J. F.; Chastain, J.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook of X-ray photoelectron spectroscopy: a reference book of standard spectra for identification and interpretation of XPS data; Physical Electronics: Chanhassen, MN, 1995. (38) Borodko, Y.; Lee, H. S.; Joo, S. H.; Zhang, Y. W.; Somorjai, G. J. Phys. Chem. C 2010, 114, 1117. (39) Bensebaa, F.; Patrito, N.; Le Page, Y.; L’Ecuyer, P.; Wang, D. S. J. Mater. Chem. 2004, 14, 3378. (40) Gómez de la Fuente, J. L.; Martínez-Huerta, M. V.; Rojas, S.; Hernández-Fernández, P.; Terreros, P.; Fierro, J. L. G.; Peña, M. A. Appl. Catal., B 2009, 88, 505.

edge taken simultaneously and were deconvoluted for various peaks using CasaXPS software (peak fitting software available at http://www.casaxps.com/). Shirley and linear background subtraction were used for the Ru3p and O1s scans, respectively. Because of overlap of the C1s peak with the Ru3d peak, we deconvoluted the Ru3p3/2 spectra instead. In the Ru3p peak, the large spin orbit splitting (22 eV) between the Ru3p3/2 and Ru3p1/2 line makes it possible to resolve the spectrum from only one peak of the spin orbit doublet.41

■

ASSOCIATED CONTENT

* Supporting Information S

SEM images of Ru nanoparticles (Figure S1), and binding energies of the Ru3d, Ru3p, O1s, and C1s APXPS spectra used for peak deconvolusion analysis of Ru nanocatalysts (Figure S2). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected], [email protected]. Present Address #

Department of Physics, Ç ukurova University, Adana 01330, Turkey. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Director, Office of Science, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231, and partly by WCU (World Class University) program (31-2008-000-10055-0) and KRF-2012-009249 through the National Research Foundation and from the Fundamental R&D Program for Core Technology of Materials funded by the Ministry of Knowledge Economy, Republic of Korea. S.H.J. was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2012-0003813) and by TJ Park Junior Faculty Fellowship.

■

REFERENCES

(1) Assmann, J.; Narkhede, V.; Breuer, N. A.; Muhler, M.; Seitsonen, A. P.; Knapp, M.; Crihan, D.; Farkas, A.; Mellau, G.; Over, H. J. Phys.: Condens. Matter 2008, 20, 184017. (2) Over, H.; Muhler, M. Prog. Surf. Sci. 2003, 72, 3. (3) Somorjai, G. A.; Park, J. Y. Top. Catal. 2008, 49, 126. (4) Somorjai, G. A.; York, R. L.; Butcher, D.; Park, J. Y. Phys. Chem. Chem. Phys. 2007, 9, 3500. (5) Kummer, J. T. J. Phys. Chem. 1986, 90, 4747. (6) Carrette, L.; F., K. A.; Stimming, U. ChemPhysChem 2000, 1, 162. (7) Chen, M. S.; Cai, Y.; Yan, Z.; Gath, K. K.; Axnanda, S.; Goodman, D. W. Surf. Sci. 2007, 601, 5326. (8) Ernst, M. A.; Sloof, W. G. Surf. Interface Anal. 2008, 40, 334. (9) Falsig, H.; Hvolbaek, B.; Kristensen, I. S.; Jiang, T.; Bligaard, T.; Christensen, C. H.; Norskov, J. K. Angew. Chem., Int. Ed. 2008, 47, 4835. (10) Somorjai, G. A.; Park, J. Y. Angew. Chem., Int. Ed. 2008, 47, 9212. (11) Song, H.; Rioux, R. M.; Hoefelmeyer, J. D.; Komor, R.; Niesz, K.; Grass, M.; Yang, P. D.; Somorjai, G. A. J. Am. Chem. Soc. 2006, 128, 3027. 5767

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768

Nano Letters

Letter

(41) Yang, J.; Lee, J. Y.; Deivaraj, T. C.; Too, H. P. J. Colloid Interface Sci. 2004, 271, 308. (42) Zawadzki, M.; Okal, J. Mater. Res. Bull. 2008, 43, 3111. (43) Knop-Gericke, A.; Kleimenov, E.; Havecker, M.; Blume, R.; Teschner, D.; Zafeiratos, S.; Schlogl, R.; Bukhtiyarov, V. I.; Kaichev, V. V.; Prosvirin, I. P.; Nizovskii, A. I.; Bluhm, H.; Barinov, A.; Dudin, P.; Kiskinova, M. X-ray Photoelectron Spectroscopy for Investigation of Heterogeneous Catalytic Processes; Elsevier Academic Press, Inc.: San Diego, CA, 2009; Vol. 52, pp 213−272. (44) Kim, Y. D.; Seitsonen, A. P.; Wendt, S.; Wang, J.; Fan, C.; Jacobi, K.; Over, H.; Ertl, G. J. Phys. Chem. B 2001, 105, 3752. (45) Bluhm, H.; Havecker, M.; Knop-Gericke, A.; Kiskinova, M.; Schlogl, R.; Salmeron, M. MRS Bull. 2007, 32, 1022. (46) Wang, J.; Fan, C. Y.; Jacobi, K.; Ertl, G. J. Phys. Chem. B 2002, 106, 3422.

5768

dx.doi.org/10.1021/nl303072d | Nano Lett. 2012, 12, 5761−5768