Oxidation of CO by Nickel Oxide Clusters Revealed by Post Heating

Article pubs.acs.org/JPCA

Oxidation of CO by Nickel Oxide Clusters Revealed by Post Heating Kazuko Sakuma, Ken Miyajima, and Fumitaka Mafuné* Department of Basic Science, School of Arts and Sciences, The University of Tokyo, Komaba, Meguro, Tokyo 153-8902, Japan ABSTRACT: Reactions of nickel oxide cluster ions, NinOn+x+ (n = 4−10, x = −1∼+3), with CO in a He buffer gas were investigated using mass spectrometry. When the cluster ions react at room temperature, a CO molecule tends to attach readily to NinOn+x+ for all of the cluster ions with different stoichiometry, although rate constants of the CO attachment reaction are more or less stoichiometry-dependent. However, CO was found to be released from the cluster ions when the cluster ions were heated up to 523 K after the reaction. This finding is interpreted such that the CO molecule that physisorbs weakly to NinOn+x+ at room temperature desorbs into the gas phase by the post heating. In addition, we found that an oxygen extraction reaction by a CO molecule actually occurs for oxygen-rich clusters such as Ni6O7+ and Ni8O9+. As the oxygen extraction reaction is one order of magnitude slower than physisorption, it lies hidden underneath of the faster physisorption processes in the mass spectrum but is revealed by the elimination of physisorption through the post heating.

■

INTRODUCTION Carbon monoxide, which is generated in incomplete combustion processes, is a pollutant gas. In order to remove CO in the gas phase, purification catalysts comprising noble metal particles,1 such as platinum,2 palladium,3 rhodium,4 and gold5 have been developed and used in a variety of products. However, it is known that these elements are so rare that we need to develop alternative catalysts. Under such circumstances, transition-metal oxides have attracted much attention due to their potential as cost-effective catalysts. For instance, a mixture of MnO2 and CuO, called hopcalite, serves as a CO oxidation catalyst at room temperature.6 The oxide of nickel, which belongs to group 10 elements with platinum and palladium, is known to exhibit high reactivity for oxidation of CO,7 methane,8 and VOCs (volatile organic compounds).9 Recently, a catalyst of cobalt oxide and nickel oxide supported on activated carbon was found to exhibit high catalytic activity and selectivity for CO oxidation at a wide reaction temperature range.10 In this study, nickel oxide acted as an important promoter in the catalyst. It is important to investigate reaction mechanisms to understand the catalytic processes in detail. By studying the reaction of the clusters in the gas phase, we can understand one aspect of the catalytic processes at the atomic and molecular levels. Catalytic properties of the gas-phase clusters have been investigated by many researchers.11−13 Castleman and coworkers have reported the reactivity of small nickel oxide cluster ions with molecules in the gas phase. Focusing the attention on the reactions of nickel clusters and nickel oxide clusters with nitric oxide and O2, they measured the rate constants with different charge states by using fast-flow quadrupole mass spectromety.14−17 For the reaction with NO, it was found that NO2 is formed on anionic clusters, © 2013 American Chemical Society

leading to the loss of one or more nickel-containing species from the clusters, while the reaction of cationic clusters proceeded by addition of NO to the clusters dominantly. More recently, they studied the reaction of NinOm± with CO by CID (collision-induced dissociation) experiments.18 They found that anionic oxide clusters are reactive enough to transfer a single oxygen atom to CO, indicating the formation of CO2. In contrast, small cationic oxide clusters react mainly through the adsorption of CO onto the clusters followed by the loss of either molecular O2 or NiO. Thus, oxygen transfer is a key process for the metal oxide clusters. To understand the oxygen transfer process, we need to know how oxygen atoms bind to nickel and oxygen atoms in the clusters. However, it is still difficult to know experimentally how strongly the oxygen atoms bind to clusters using only conventional mass spectrometry. Hence, we observe the stable structures of the oxide clusters after rupturing the thermally unstable bond. In reality, we performed the experiments under room temperature and at elevated temperature conditions. Heating the clusters in the gas phase can cause a weakly bound molecule to dissociate, enabling us to identify the stable compositions of the clusters. We also investigated reactions between nickel oxide clusters and CO, aiming to find a relation between the oxygen donation capability of nickel oxide clusters and the CO oxidation. The reactions were examined under the multiple collision conditions and with and without post heating after reactions with CO to elucidate the stable compositions of the products. Received: October 17, 2012 Revised: March 25, 2013 Published: March 25, 2013 3260

dx.doi.org/10.1021/jp400850u | J. Phys. Chem. A 2013, 117, 3260−3265

The Journal of Physical Chemistry A

■

Article

EXPERIMENTAL SECTION The reactivity of nickel oxide clusters with CO was investigated using a reflectron-equipped time-of-flight mass spectrometer (see Figure 1).19,20 Briefly, nickel oxide clusters were produced

distinguish impurities such as attached hydrogen atoms that may affect cluster reactivity.

■

RESULTS Figure 2a displays a typical mass spectrum of cationic nickel oxide clusters. Ion peaks assignable to NinOn+ and NinOn+1+ for

Figure 1. Experimental apparatus used in the present study. Cluster ions formed in the cluster source can be heated at the extension tube just before being expanded into vacuum.

in a laser ablation source. A nickel metal rod (Nilaco Co., Ltd., 99.99%) was ablated using the focused second harmonic (532 nm) of a Nd3+:YAG pulse laser at a typical pulse energy of 10 mJ pulse−1 in the presence of oxygen diluted in helium (1%). The clusters were passed through a reaction gas cell (2 mm diameter, 60 mm long) and an extension tube (4 mm diameter, 120 mm long) before expansion into the first vacuum chamber. Then, they were introduced into the differentially pumped second chamber through a skimmer. Reactant CO gas was injected inside of the reaction gas cell. The gas was diluted by He so that the total stagnation pressure inside of the valve was constant at 800 Torr. The typical number density of gas inside of the reaction gas cell was estimated to be ∼1016 molecules cm−3. The residence time of nickel oxide cluster ions in the reaction gas cell was estimated to be ∼70 μs. The temperature of the extension tube was controlled in the range of 298−673 K using a resistive heater and monitored with a thermocouple. Thermal equilibrium of the clusters was achieved by collisions with the He carrier gas well before expansion into the vacuum. In our experimental setup, the reaction of NinOm+ with CO occurs in the reaction cell, which is maintained at room temperature. In fact, we measured the reaction of a vanadium atom with CO at different temperatures in the extension tube and found no significant difference in the reactivity. Hence, the reactions above take place at room temperature. The cluster ions gained kinetic energy of ∼3.5 keV in the acceleration region for the mass analysis. The ions were steered and focused by a set of vertical and horizontal deflectors and an einzel lens. After traveling in a 1 m field-free region, the ions were reversed by the reflectron and detected using a Hamamatsu double-microchannel plate detector. Signals from the detector were amplified with a 350 MHz preamplifier (Stanford SR445A) and digitized using an oscilloscope (LeCroy LT374). Averaged time-of-flight spectra (typically 300 sweeps) were sent to a computer for analysis. The mass resolution m/Δm, exceeds 1500, which is sufficient to

Figure 2. Mass spectrum of cationic nickel oxide clusters, NinOm+, produced by laser ablation of a nickel metal rod in the presence of oxygen-seeded He carrier gas. Panels a and b show the reference spectra without CO in the reactor, and panels c and d show the mass spectra of NinOm+ clusters before (blue) and after (orange) reaction with CO gas. Panels b and d display the mass spectra with post heating at 523 K. Peaks are labeled according to the notation NinOm+ (n,m). Overlapped areas of the ion peaks are filled with black in panels c and d.

n = 4−10 were observed prominently. In addition to these cluster ions, Ni7O6+, Ni7O9+, Ni8O10+, and Ni8O11+ also appear in the mass spectrum. Clusters containing a larger number of oxygen atoms were not produced in our experimental setup, even when the partial pressure of the oxygen gas in the cluster source was increased. In order to elucidate the stability of nickel oxide cluster ions, we observed a mass spectrum of nickel oxide cluster ions after they were heated up to 523 K (see Figure 2b). It is apparent that Ni7O9+ and Ni8O11+ decrease and Ni7O7+ and Ni8O9+ increase in their intensities, indicating that oxygenrich clusters release weakly bound oxygen molecule. Nickel oxide cluster ions, NinOn+x+, were subjected to reactions with 0.3% CO in the reaction gas cell at room temperature. Figure 2c shows a mass spectrum of the reaction products. It is apparently seen that pristine NinOn+x+ clusters decrease in intensity and NinOn+x(CO)+ clusters are formed after the reactions, suggesting that a CO attachment reaction Ni nOn + x + + CO → Ni nOn + x (CO)+ 3261

(1)



Article

takes place. When the number density of CO in the gas cell was increased by 100 times, two or three CO molecules were found to attach to the clusters. Figure 2d shows a mass spectrum of the reaction products that were post heated at the extension tube at 523 K after the reaction cell. Clearly, NinOn+x(CO)+ disappears in the mass spectrum, suggesting that CO in the clusters is so weakly bound to NinOn+x+ that CO is released into the gas phase by the post heating. In order to confirm that only the adsorption reaction proceeds in the concentration that we used, intensity ratios of NinOn+x+ and NinOn+x(CO)+ after the reaction with 0.3% CO are shown in Figure 3 for a different stoichiometry for n and x.

Figure 4. Rate constants, kad, for the adsorption reaction, NinOn+x+ + CO → NinOn+x(CO)+, as a function of composition of NinOn+x+ clusters, (n,n+x). The rate constants do not change significantly on the composition except for Ni7O6+.

Figure 3. Intensity ratio of NinOm+ (n = 6−9) clusters after the reaction with CO gas for the different cluster sizes, (n,m). The intensities of NinOm(CO)+ clusters are normalized to the intensity of the pristine NinOm+ cluster before the reaction. By using 18O2, cluster composition between NinOm+ and NinOm(CO)+ can be assigned with no ambiguity.

Here, the intensities are normalized to the intensity of the pristine NinOn+x+ before the reaction. Summation of the ratios is almost one, indicating that NinOn+x(CO)+ increased in intensity as much as the pristine NinOn+x+ intensity decreased upon the reaction. Hence, dissociation of NinOn+x+ upon reaction is considered to be minimal. The intensities of NinOn+x+ gradually decrease with an increase in the number density of CO gas. We estimated rate constants for reaction 1, k, to evaluate the reactivity. As the number of cluster ions is much smaller than the number of CO molecules in the gas cell, the concentration of CO is considered to remain unchanged during the reaction time. Using a pseudofirst-order approximation, the decay of NinOn+x+ by reaction 1 is given by [Ni nOn + x +] = [Ni nOn + x +]0 exp( −kt0[CO]) +

Figure 5. Mass spectrum of NinOm+ clusters before and after reaction with CO gas at room temperature followed with post heating at 523 K. Overlapped areas of the ion peaks are filled with black in the bottom spectrum.

The abundance of NinOm+ for n = 4, 5, 6, and 8 changed to a greater extent by the reaction with CO. First, oxygen-rich clusters such as NinOn+1+ (n = 6 and 8) decreased and NinOn+ (n = 6 and 8) increased in their intensities after the reaction. As the CO density increased, the spectral changes became more evident. Second, Ni4O5+ and Ni5O6+ decreased in their intensities after the reaction with CO, but the intensities of Ni4O4+ and Ni5O5+ did not increase at all. Instead, Ni4O3(CO)+ and Ni5O4(CO)+ appear in the mass spectrum despite the absence of bare clusters, Ni4O3+ and Ni5O4+.

(2)

+

where [NinOn+x ]0 and [NinOn+x ] represent the number of NinOn+x+ clusters before and after the reaction, respectively, t0 is the reaction time, and [CO] is the number density of CO in the reaction gas cell. It is difficult to determine the number density of CO in the reaction cell; therefore, the reaction of atomic Ni+ with gas-phase CO prepared under identical experimental conditions was observed as a reference. The rate constant for the reaction of Ni+ with CO is known to be 5.6 × 10−13 cm3 s−1,21 so that the product of t0 and the number density of the reaction gas can be obtained. In conjunction with t0, the number densities of reactant gases were estimated to be ∼1016 molecules cm−3. The rate constants thus obtained are shown in Figure 4 for different stoichiometry. Figure 5 shows the mass spectra of NinOm+ (4 ≤ n ≤ 10) before and after the reaction with 1% CO, which were post heated at the extension tube at 523 K after the reaction cell.

■

DISCUSSION Formation of Stoichiometric and Nonstoichiometric Nickel Oxide Clusters. Nickel oxide clusters were prepared by laser ablation of a nickel metal rod in the presence of oxygen gas. As shown in Figure 2a, 1:1 and oxygen-rich clusters, NinOn+, and NinOn+1+ were selectively formed for n = 4, 5, 6, 9, and 10 when a sufficient amount of an oxygen gas was supplied. The formation of 1:1 clusters is consistent with the fact that nickel atoms and oxygen atoms tend to have +2 and −2 charge states in the bulk phase. The mass distribution of nickel oxide cluster ions is totally consistent with work by both the Castleman and Duncan groups.17,22 3262



Article

For n = 7 and 8, clusters with different stoichiometry were formed, Ni7O6−9+ and Ni8O8−11+ (see Figure 2b). As mentioned in the previous section, the intensities of Ni7O9+ and Ni8O11+ decreased when they were heated to 523 K and had totally disappeared at 673 K in our experimental setup. These findings indicate that the oxygen-rich clusters were formed by uptaking molecular oxygen weakly to the core cluster ions, Ni7O7+ and Ni8O9+. CID experiments for small cationic nickel oxide clusters (n = 1 and 2) show that oxygen-rich clusters tend to release molecular oxygen, O2, as a preferred fragmentation pathway.18 Hence, it is likely that cationic nickel oxide clusters bind oxygen preferentially in the molecular form. In contrast, oxygen-deficient clusters, such as NinOn−1+, were not produced by post heating of NinOn+ and NinOn+1+ (n = 4, 5, 6, 9, and 10). The oxygen atoms are considered to bind strongly to the Ni atoms in the clusters. Hence, we are able to speculate that NinOn−1+ is energetically by far less stable than NinOn+ and NinOn+1+. Physisorption of CO at Room Temperature. When cationic nickel oxide clusters, NinOn+x+, were passed into the reaction gas cell filled with 0.3% CO diluted in He gas at room temperature, CO molecules attached to the nickel oxide clusters (see Figure 2c). The question here is, is CO released from the cluster as CO or CO2? For instance, the abundance of Ni6O7+ after the reaction at room temperature was reduced to 90% of that of pristine Ni6O7+, forming Ni6O7(CO)+ (see Figure 3). Then, the abundance of Ni6O7+ was found to mostly recover by the post heating. These processes can be described as Reaction Ni 6O7+ + CO → Ni 6O7 (CO)+

suggests that the binding energy of CO to nickel oxide clusters is very low. Hence, we are able to consider that a CO molecule physisorbs on the nickel oxide clusters. The rate constants of the physisorption do not depend very much on size and stoichiometry except for Ni7O6+. The variation of the rate constants is only within a factor of 3 (see Figure 4). This is probably because physisorption relates to electrostatic interaction between the charged cluster ion and CO. According to the Langevin cross section giving an uppermost cross section relating to charge-induced dipole interaction, the cross section with CO and the rate constant are estimated to be 1.4 × 10−18 m2 and 2.3 × 10−9 cm−3 s−1, respectively. Comparison of the rate constants suggests that one CO molecule attaches to the nickel oxide clusters as a result of 103 collisions. In the nickel oxide clusters, Ni7O6+ behaves very differently. This is the only ion that appears in the mass spectrum containing fewer oxygen atoms than nickel atoms. At room temperature, CO hardly adsorbs onto Ni7O6+; the rate constant of physisorption is more than 3 orders of magnitude smaller than that of the other clusters. Hence, the geometrical and electronic structures of Ni7O6+ need to be studied intensively in future studies. Oxidation of CO by Nickel Oxide Clusters. As shown in Figure 5, oxygen-rich clusters such as Ni6O7+ decrease while stoichiometric clusters Ni6O6+ increase after the reaction with CO. These spectral changes suggest that the following chemical reaction occurs for Ni6O7+. CO oxidation Ni 6O7+ + CO → Ni 6O6+ + CO2

In contrast, Ni6O5+ was not seen in the mass spectrum even at high CO density, so that oxidation of CO by Ni6O6+ does not proceed. Hence, only oxygen-rich Ni6O7+ is able to release one oxygen atom to CO, forming CO2. We estimated the rate constants for eq 5 from the CO density dependence to be 2.2 × 10−13 cm−3 s−1. For n = 8, a similar increase of Ni8O8+ and decrease of Ni8O9+ were observed, indicating that an oxygen-transfer reaction occurs. CO oxidation

(3)

Post heating Δ

Ni 6O7 (CO)+ → Ni6O7+ + CO

(5)

(4)

Hence, we are able to consider that the major reaction pathway is the release of CO, as described by eq 4. Other NinOm+ clusters behave similarly upon reaction with CO and the post heating. The post heating of the clusters revealed that CO molecules bind to the clusters very weakly. Figure 6 shows the abundance of Ni6O7(CO)+ and Ni6O7+ as a function of temperature. As the temperature rises, the abundance of Ni6O7(CO)+ decreases, whereas the abundance of Ni6O7+ gradually increases. The temperature dependence

Ni8O9+ + CO → Ni8O8+ + CO2

(6)

The rate constant for this reaction was estimated to be 1.2 × 10−13 cm−3 s−1. The post heating is considered to be significantly important to observe the mass spectral changes induced by the oxidation reaction. In fact, the rate constant of physisorption of CO to the cationic nickel oxide clusters (kad) is in the range of 10−12 cm−3 s−1, whereas the rate constant of oxidation of CO by the cationic nickel oxide clusters (koxi) is about 10−13 cm−3 s−1. The comparison of the rate constants indicates 1/10 of the physisorption events lead to oxidation of CO, described by eqs 7 and 8.

Figure 6. Ni6O7+ and Ni6O7(CO)+ were formed by the reaction with CO at room temperature followed by post heating. The intensity ratios are plotted as a function of temperature of post heating. The intensity of the Ni6O7(CO)+ cluster is normalized to the intensity of the pristine Ni6O7+ cluster before the reaction.

In the oxidation reaction of CO, it approaches to NinOm+, forming NinOm(CO)+, in which CO is weakly bound to NinOm+ at first (eq 7). Then, it passes beyond the energy barrier between the physisorption and the chemisorption, and then, 3263



Article

CO strongly attaches to NinOm+. Then, an oxygen atom in NinOm+ and the chemisorbed CO bind together to form CO2 after passing another energy barrier of reaction intermediates. In the present study, no ion peaks corresponding to COchemisorbed Ni6O7(CO)+ and Ni8O9(CO)+ appeared in the mass spectrum after the post heating. This finding suggests that the second energy barrier is lower than the first one, so that once the clusters pass the energy barrier between the physisorption and the chemisorption, the oxidation reaction is completed in a barrierless manner. In other words, the ratedetermining step in the oxidation reaction is the activation barrier between physisorption and chemisorption. In contrast to Ni6O7+ and Ni8O9+, Ni4O5+ and Ni5O6+ give different reaction products, Ni4O3(CO)+ and Ni5O4(CO)+. The reactions are described as

Recently, Castleman and co-workers and He and co-workers elucidated that the reactivity of early transition-metal oxide clusters is influenced by the charge and composition.23−25 The reactivity to CO and hydrocarbons depends on whether the clusters contain radical oxygen or not. Theoretical investigation indicated that stoichiometric cationic metal oxide clusters include a radical oxygen (M−O•) center with an elongated M− O bond that is active with CO and unsaturated hydrocarbons. However, the later transition-metal oxide clusters (M = Fe and Co) are likely to behave differently in the radical formation.25 Indeed, their models are not able to explain the fact that the CO oxidation reaction proceeds for NinOn+1+ (n = 6 and 8), NinOn+1− (n = 4 and 5), and Ni7O7−. Furthermore, it is known that the additional valence d electrons makes fine quantum chemical calculations very difficult.25 We hope that our experimental findings will stimulate further experiments and theoretical discussion of the reactivity of later transition-metal oxide clusters.

■

CONCLUSION We examined reactions of nickel oxide cluster ions, NinOn+x+ (n = 4−10, x = −1∼+3), with CO in a He buffer gas at room temperature. When the cluster ions collide with CO molecules, a CO molecule binds to NinOn+x+. As CO was found to be released from the cluster ions when the cluster ions were heated up to 523 K after the reaction, the CO molecule is considered to physisorb weakly to NinOn+x+. The pseudo-firstorder rate constant for the physisorption ranges over 10−12 cm−3 s−1. In addition, we found that the CO oxidation reaction actually occurs for oxygen-rich clusters such as Ni6O7+ and Ni8O9+; the pseudo-first-order rate constants for the CO oxidation were estimated over 10−13 cm−3 s−1. These values were of the same order as anionic clusters such as Ni4O5−, Ni5O6−, and Ni7O7−. As the CO oxidation reaction is one order of magnitude slower than physisorption, it lies hidden underneath of the faster physisorption processes but is revealed by removing the physisorbed species through the post heating.

This propensity is consistent with the previous work by Castleman and his co-workers.18 They studied CID of NiO2−4+ and Ni2O2−4+ with N2 and Xe, finding that these cationic clusters release O2 preferentially. Moreover, attachment of CO releasing O2 was actually observed for NiO3,4+ and Ni2O2−4+. It should be emphasized that attachment of CO was not observed for n ≥ 6. It is likely that the degree of vibrational freedom is large enough to disperse the heat of reaction and suppress the release of the O2 moiety for the case of n ≥ 6. Difference in the CO Reactivity between Cationic and Anionic Clusters. It is known that small anionic nickel oxide clusters, NinOn+1− (n = 1−4) are able to oxidize CO.18 In this relation, we also examined CO oxidation for larger clusters, NinOm− (n = 4−7). In contrast to the result of cationic clusters, no CO adsorbed products were observed for anionic nickel oxide clusters at room temperature. In addition, CO oxidation proceeds only for NinOn+1− (n = 4 and 5) and Ni7O7−, and no other clusters react with CO (figure not shown). The rate constants are 1.4 × 10−13 and 1.3 × 10−13 cm−3 s−1 for Ni5O6− and Ni7O7−, respectively, which are in the same range as those for cationic clusters. Acknowledging that there is strong size dependence of the reactivity, we are able to summarize that oxygen-rich clusters, NinOn+1+/−, can oxidize CO for both anionic and cationic clusters involving NinOn+1+ (n = 6 and 8), NinOn+1− (n = 4 and 5), and Ni7O7−. The reactivity of the oxygen-rich clusters can be discussed in terms of the stability of the clusters. The post heating of the nickel oxide clusters exclusively produces both NinOn+ and NinOn+1+ (n = 4−10 except for n = 7), indicating that NinOn+ and NinOn+1+ are energetically stable and comparable, whereas oxygen-deficient NinOn−1+ is much less stable. Assuming that the formation energies of NinOn+ and NinOn+1+ are the same, the binding energy of an oxygen atom in NinOn+1+ equals the formation energy of an oxygen atom in the gas phase (2.4 eV). On the other hand, the energy difference between CO and CO2 in the gas phase is −2.7 eV. Hence, the CO oxidation reaction by NinOn+1+ forming NinOn+ and CO2 is totally exothermic at 0.3 eV. In contrast, as the formation energy of NinOn−1+ is expected to be much higher than that of NinOn+, the binding energy of an oxygen atom in NinOn+ must be higher than 2.7 eV. Hence, the CO oxidation reaction is not possible in terms of energy.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the Grant-in-Aids for Scientific Research (B) (No. 22350004) from the Ministry of Education, Culture, Sports, Science and Technology, Japan (MEXT) and by the Genesis Research Institute, Inc. for the cluster research. F.M. acknowledges Professor Greg Metha for helpful discussions on the experimental setup.

■

REFERENCES

(1) Santra, A.; Goodman, D. Electrochim. Acta 2002, 47, 3595−3609. (2) Fukuoka, A.; Kimura, J.; Oshio, T.; Sakamoto, Y.; Ichikawa, M. J. Am. Chem. Soc. 2007, 129, 10120−10125. (3) Zhu, H.; Qin, Z.; Shan, W.; Shen, W.; Wang, J. J. Catal. 2004, 225, 267−277. (4) Mullins, D.; Overbury, S. J. Catal. 1999, 188, 340−345. (5) Haruta, M.; Tsubota, S.; Kobayashi, T.; Kageyama, H.; Genet, M.; Delmon, B. J. Catal. 1993, 144, 175−192. (6) Buciuman, F. C.; Patcas, F.; Hahn, T. Chem. Eng. Process. 1999, 38, 563−569. 3264



Article

(7) Raju, V.; Rao, D.; Chakrabarty, D. Ind. J. Chem., Sect. A 1982, 21, 350−352. (8) Bradford, M.; Vannice, M. Appl. Catal., A 1996, 142, 73−96. (9) Stoyanova, M.; Konova, P.; Nikolov, P.; Naydenov, A.; Christoskova, St.; Mehandjiev, D. Chem. Eng. J. 2006, 122, 41−46. (10) Zhou, G.; Xie, H.; Gui, B.; Zhang, G.; Zheng, X. Catal. Commun. 2012, 19, 42−45. (11) Bell, R.; Zemski, K.; Kerns, K.; Deng, H.; Castleman, A. W., Jr. J. Phys. Chem. A 1998, 102, 1733−1742. (12) Zemski, K.; Justes, D.; Castleman, A. W., Jr. J. Phys. Chem. B 2002, 106, 6136−6148. (13) Reveles, J. U.; Johnson, G. E.; Khanna, S. N.; Castleman, A. W., Jr. J. Phys. Chem. C 2010, 114, 5438−5446. (14) Vann, W. D.; Wagner, R. L.; Castleman, A. W., Jr. J. Phys. Chem. A 1998, 102, 1708−1718. (15) Vann, W. D.; Wagner, R. L.; Castleman, A. W., Jr. J. Phys. Chem. A 1998, 102, 8804−8811. (16) Vann, W. D.; Bell, R. C.; Castleman, A. W., Jr. J. Phys. Chem. A 1999, 103, 10846−10850. (17) Vann, W.; Castleman, A. W., Jr. J. Phys. Chem. A 1999, 103, 847−857. (18) Johnson, G. E.; Reilly, N. M.; Castleman, A. W., Jr. Int. J. Mass Spectrom. 2009, 280, 93−100. (19) Miyajima, K.; Himeno, H.; Yamada, A.; Yamamoto, H.; Mafuné, F. J. Phys. Chem. A 2011, 115, 1516−1520. (20) Yamada, A.; Miyajima, K.; Mafuné, F. Phys. Chem. Chem. Phys. 2011, 115, 1516−1520. (21) Jarvis, M. J. Y.; Pisterzi, L. F.; Blagojevic, V.; Koyanagi, G. K.; Bohme, D. K. Int. J. Mass Spectrom. 2003, 227, 161−173. (22) Dibble, C. J.; Akin, S. T.; Ard, S.; Fowler, C. P.; Duncan, M. A. J. Phys. Chem. A 2012, 116, 5398−5404. (23) Johnson, G. E.; Mitrić, R.; Nössler, M.; Tyo, E. C.; BonačićKoutecký, V.; Castleman, A. W., Jr. J. Am. Chem. Soc. 2009, 131, 5460−5470. (24) Ma, J.; Xu, B.; Meng, J.; Wu, X.; Ding, X.; Li, X.; He, S. J. Am. Chem. Soc. 2013, 135, 2991−2998. (25) Zhao, Y.; Wu, X.; Ma, J.; He, S.; Ding, X. Phys. Chem. Chem. Phys. 2011, 13, 1925−1938.

3265


Oxidation of CO by Nickel Oxide Clusters Revealed by Post Heating

Recommend Documents