Article pubs.acs.org/JPCA
Water-Induced Adsorption of Carbon Monoxide and Oxygen on the Gold Dimer Cation Tomonori Ito,† G. Naresh Patwari,‡ Masashi Arakawa,† and Akira Terasaki*,† †
Department of Chemistry, Graduate School of Sciences, Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan Department of Chemistry, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India
‡
ABSTRACT: It is demonstrated, using tandem mass spectrometry and ion trap, that preadsorption of a H2O molecule on the gold dimer cation, Au2+, enhances adsorption of CO and O2 molecules, which is otherwise inert toward these molecules. The rate of adsorption of CO on Au2(H2O)+ was found to be higher by 2 orders of magnitude relative to bare Au2+. The enhancement of the CO adsorption rate is due to the presence of a reaction channel, which cleaves the Au−Au bond, leading to the formation of Au(H2O)(CO)+. Such an observation can be attributed to weakening of the Au−Au bond upon adsorption of a water molecule. Further, it was also observed that preadsorption of H2O leads to dramatic enhancement of O2 adsorption on the Au2+ ion, which can be attributed to the changes in the electron density following water adsorption.
H 2O + CO ⇄ H 2 + CO2
1. INTRODUCTION Ever since the catalytic activity of gold nanoparticles on metaloxide supports were discovered,1 there has been explosive growth, in the number of investigations and consequent reports in the literature, on the ability of gold nanoparticles, both supported and unsupported, to catalyze a very wide range of reactions.2,3 Among these reactions oxidation of CO is considered as a benchmark reaction to evaluate the catalytic activity since its mechanism is well understood. A large number of reports have confirmed that the catalytic activity not only depends on the size of the nanoparticle but also the support, which influences the oxidation states of metals in nanoparticles.4,5 It can be envisaged that the reactivity of metal nanoparticles can be modulated by the size and also the nature of the support. It has been shown that reducing metal oxides (such as TiO2) have much better catalytic activity in comparison to nonreducing metal oxide supports (such as Al2O3).6,7 The extreme view in this regard is that metal in zero oxidation states do not exist on the supported surfaces.8 The catalytic activity may be controlled more dramatically in the cluster regime by taking an advantage of precise control of the number of constituent atoms. For instance, the low temperature oxidation of CO to CO2 in the presence of O2 over gold clusters has size dependence, wherein the clusters of very low diameters (∼0.5 nm) were found to be most effective catalysts with Au8 being the smallest catalytically active size.9 In this regard, investigations of reactions on the small gas phase clusters facilitate understanding the size and consequent electronic effects thereby elucidating the detailed reaction mechanism of catalytic activity. The catalytic properties of size selected metal clusters have been a subject of great interest in many environmentally benign reactions, such as the water gas shift (WGS) reaction © 2014 American Chemical Society
(1)
which is well-known for the industrial production of clean molecular hydrogen (H2). The WGS reaction is exothermic; therefore, low temperature is desired for better yields. However, low temperature leads to lower reaction rates. The competition between yields and the rates necessitates a suitable catalyst for practical utility. Commercially, the WGS reaction is carried over fine particles of Cu metal dispersed over ZnO/Al2O3. Several other metals such as Fe, Ni, Au, Pt, Ru, Ir, and others supported on appropriate metal oxides have been used as catalysts for the WGS reaction under varying conditions. In the recent times either FeCr or CuZn particles have been industrially used, which are not suitable for a mobile catalyst because of their pyrophoricity.10 Because of the ability of gold nanoparticles to catalyze a very wide range of reactions, it can therefore be surmised that a gold-based catalyst would be ideal in terms of both reactivity and safer handling. Catalytic activities of the gold dimer are well examined for CO oxidation by anions11 and methane activation by cations.12 DFT calculations on the mechanistic aspects of WGS reaction show that gold dimer cation is a better catalyst in comparison with neutral gold dimer and gold dimer anion.13 Reactions involving gas phase metal clusters have tremendous utility due to precise control on the size, charge state, and, in the case of multielement clusters, the composition.14 In the case of CO oxidation over supported gold nanoparticles, it has been shown that trace amounts of water in the reactant mixture significantly alters the catalytic activity.15 This observation can Special Issue: A. W. Castleman, Jr. Festschrift Received: January 31, 2014 Revised: March 1, 2014 Published: March 4, 2014 8293
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
be interpreted as the ability of foreign molecules to control the catalytic activity via the coadsorption effects. Such effects have been reported for CO adsorption on gold-cluster anions in the presence of water vapor,16 for O2 adsorption on gold-hydroxy cluster anions,17 and on gold-cluster cations with preadsorbed H2.18 In the present work, the ability of preadsorbed water molecule to alter the adsorption and reactivity of CO and O2 on the gold dimer cation is investigated. Reaction kinetics were measured by employing a linear radio frequency ion trap, where Au2+ and CO (or O2) interacted for a given reaction time. The effect of a water molecule is evaluated by comparing reaction rate constants measured for Au2+ and Au2(H2O)+. Figure 1. Reaction kinetics of Au2+ with CO measured at room temperature. Amounts of the reactant and product ions are plotted as a function of storage time: Au2+ (closed circles), Au(CO)+ (up-pointing open triangles), and Au(CO)2+ (open squares). The partial pressure of CO was 7.2 × 10−3 Pa. Fitting curves are superimposed by solid lines.
2. EXPERIMENTAL SECTION Details of the experimental apparatus have been described elsewhere.19 Briefly, gold cluster cations were generated in an ion source, where magnetron sputtering of a gold plate is followed by aggregation of sputtered atoms/ions through collisions with helium buffer gas cooled by liquid nitrogen. The cluster ions were guided by radio frequency octopole ion guides. A pick-up cell was placed downstream of the cluster ion source for preadsorption of a H2O molecule on the gold cluster cations. The helium gas was bubbled through water and remixed with pure helium gas to control the composition of water vapor and was introduced into the pick-up cell. The produced ions were mass-selected by a quadrupole mass filter (Max-16000, Extrel CMS) and were stored in a linear octopole ion trap, where a reactant gas (either CO or O2) was introduced continuously along with a He buffer gas. The ions were stored for a given storage time t, were extracted from the ion trap for mass-analysis by a second quadrupole mass filter (Max-4000, Extrel CMS), and were detected by a current meter (Picoammeter 6485, Keithley). Partial pressures of the reactant gas and buffer He gas were measured outside the trap by a residual gas analyzer (RGA100, SRS); actual pressures in the trap were evaluated by calibration based on a known reaction.20 The ion trap and the reactant gases were able to be cooled down to 100 K by liquid nitrogen for low-temperature experiments. Reaction kinetics measurement was performed as follows: Mass-selected ions, Au2(H2O)m+ (m = 0, 1), were loaded into the trap for 200 ms. The number of ions stored typically reached 1 × 107 within the loading time. The partial pressure of the He buffer gas was adjusted to be in the order of 0.1 Pa throughout the experiment, while the partial pressure of a reactant gas was adjusted to be in the range of 10−1−10−5 Pa. The ions stored in the trap were thermalized at the temperature of the He gas within a few milliseconds,21 the time scale of which is much shorter than the reaction rate (∼10 s−1), i.e., the ions were thermalized before reaction. The amounts of reactant and product ions were measured as a function of storage time t to evaluate reaction kinetics.
buffer gas, it was observed that Au2+ adsorbs CO followed by cleavage of Au−Au bond, which is in contrast to a previous report that Au2+ does not adsorb any CO molecules in an ion− cyclotron−resonance experiment.22 The observation of AuCO+ and Au(CO)2+ indicates the sensitivity of the present experimental techniques, given the fact that the reaction rates and yields for the reactive CO adsorption are very small, vide infra. Figure 2 shows temporal evolution of a reaction of the hydrated gold dimer cation, Au2(H2O)+, with CO at room
Figure 2. Reaction kinetics of Au2(H2O)+ with CO measured at room temperature. Amounts of the reactant and product ions are plotted as a function of storage time: Au2(H2O)+ (closed circles), Au2(CO)+ (uppointing triangles), Au(CO)2+ (down-pointing open triangles), and Au(H2O)(CO)+ (open diamonds). The preadsorbed H2O is denoted as W in the legend. The partial pressure of CO was 1.1 × 10−4 Pa. Fitting curves are superimposed by solid lines.
temperature. For this reaction the partial pressure of CO was lowered to approximately 10−5 Pa, which is about 2 orders of magnitude lower than the partial pressure used for the reaction of bare Au2+. The reactant Au2(H2O)+ decreased exponentially with the storage time similarly to observations made for the reaction of bare Au2+. The temporal profile of Au2(CO)+ shows growth followed by decay during the course of the reaction, which indicates that it is a reaction intermediate. However, Au(H2O)(CO)+ and Au(CO)2+ show growth over the course of the reaction, which therefore can be considered as final products.
3. RESULTS 3.1. Reaction with CO. Kinetics measurements for Au2+ + CO reaction were performed at room temperature. The concentration of CO was kept in the order of 10−2 Pa. The amounts of AuCO+ and Au(CO)2+ produced by the reaction of Au2+ with CO under multiple collision conditions as a function of storage time are shown in Figure 1. Also shown is the decrease in the reactant Au2+ in the same time scale. In the present experiment performed in the ion trap filled with He 8294
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
3.2. Adsorption of O2. Bare Au2+ did not adsorb O2 even at temperatures down to 100 K. This is consistent with the previous report.23 The nonreactive character of Au2+ suggests that the interaction energy between O2 and Au2+ should be lower than the thermal energy of the system at 100 K. In striking contrast, a water adduct, Au2(H2O)+, exhibited high reactivity against O2. Figure 3 shows reaction kinetics of
depicted in Figures 1−3, and results are listed in Table 1. The methodology for evaluating the rate constants is based on the experimental conditions. The ion trap was maintained at constant partial pressures of He and CO by introducing the gases continuously into the gas cell of the trap. The number density of the CO gas, nCO, was higher than that of the trapped ions at least by 3 orders of magnitude. Further, the steady-state concentration is maintained by continuous flow of CO gas into the gas cell. Therefore, nCO was practically constant over the entire reaction experiments, hence pseudo-first-order reaction. The temporal evolution of the reaction of Au2+ with CO shown in Figure 1 indicates the following reaction steps: Au 2+ + CO → AuCO+ + Au
(2)
AuCO+ + CO → Au(CO)2+
(3)
We obtained pseudo-first-order rate constants k′ for the above two pathways by fitting the data to rate equations with a software called DETMECH.24 On the basis of bimolecular reaction kinetics, an absolute bimolecular rate constant, kII, is derived as kII = k′/nCO. k′ and kII were determined to be 1.91 s−1 and 9.4 × 10−13 cm3 s−1, respectively. Taking into account an ion-induced dipole interaction effect, simple association rates between ions and molecules can be estimated by the Langevin rate,25 kL = 2πe(4πε0)−1(α/μ)1/2, where e is the elementary charge, α is the polarizability of the target gas molecules, and μ is a reduced mass of the ion−neutral pair. Since CO has a large permanent dipole moment, the association rate tends to be underestimated only with the Langevin rate; it is needed to be corrected for ion−dipole interaction by ADO (averaged dipole orientation) theory. The association rate with the ADO theory is expressed as follows: kADO = (2πe/μ1/2)(4πε0)−1[α1/2 + cμD(2/πkBT)] .26 Here, c represents a locking constant (c = 0.196 for CO),27 μD is the dipole moment of CO (= 0.112 D), kB is Boltzmann’s constant, and T is the temperature of the system. The association rate of CO was calculated to be 6.6 × 10−10 cm3 s−1, whereas the experimentally determined reaction rate was 9.4 × 10−13 cm3 s−1. This means only 0.2% of collisions would indeed lead to CO association. In other words, once a CO molecule interacts with Au2+, the CO molecule tends to be released in most of the cases. This behavior is understood as follows: A CO molecule is bound to a metal cation more strongly than other ordinary molecules because of electronic back-donation from the metal atom to CO. This may create an internally hot intermediate adduct, [Au2(CO)+]*, which proceeds to transition states not only of CO release but also
Figure 3. Reaction kinetics of Au2(H2O)+ with O2 measured at 100 K. Amounts of the reactant and product ions are plotted as a function of storage time: Au2(H2O)+ (closed circles), Au2(H2O)(O2)+ (uppointing triangles), Au2(H2O)(O2)2+ (down-pointing open triangles), and Au2(H2O)(O2)(N2)+ (open diamonds). The partial pressure of O2 was 9.1 × 10−2 Pa. Au2(H2O)(O2)(N2)+ was produced due to residual N2 in the trap. Solid lines are the fits to data according to the reaction kinetics.
Au2(H2O)+ with O2 at 100 K, where three product ions, Au2(H2O)(O2)+, Au2(H2O)(O2)2+, and Au2(H2O)(O2)(N2)+, were produced. Au2(H2O)(O2)+ was a reaction intermediate because it disappeared at a long storage time, by adsorbing an additional O2 molecule leading to the formation of Au2(H2O)(O2)2+, which is the final product under the present reaction conditions. Au2(H2O)+ was able to adsorb up to two O2 molecules in the reaction time scale of about 2 s. Additionally, Au2(H2O)(O2)(N2)+ was observed, which is due to residual N2 in the ion trap of which the partial pressure was estimated to be 1.3 × 10−4 Pa.
4. DISCUSSION 4.1. Evaluation of Reaction Rate Constants. Reactions rate constants were determined, based on all the observed species from the results of reaction kinetics measurements
Table 1. Absolute Bimolecular and Termolecular Rate Constants, kII and kIII, of Each Reaction Steps; the Values Are Subject to a Systematic Error of ∼20% kII (cm3 s−1)
reaction step +
9.4 × 10
(2) Au 2 + CO → AuCO + Au +
(3) AuCO + CO → Au(CO)2
kIII (cm6 s−1)
−13
+
7.8 × 10−28
+
(6) Au 2(H 2O)+ + CO → Au(H 2O)(CO)+ + Au
8.8 × 10−11
(7) Au 2(H 2O)+ + CO → Au 2(CO)+ + H 2O
1.1 × 10−10
(8) Au 2(CO)+ + CO → Au(CO)2+ + Au
1.1 × 10−10
(9) Au 2(H 2O)+ + O2 → Au 2(H 2O)(O2 )+
2.0 × 10−25
(10) Au 2(H 2O)(O2 )+ + O2 → Au 2(H 2O)(O2 )2+
5.2 × 10−26
(11) Au 2(H 2O)(O2 )+ + N2 → Au 2(H 2O)(O2 )(N2)+
1.1 × 10−22 8295
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
of Au release, i.e., [Au2+···(CO)]† and [Au(CO)+···Au]†. According to RRKM theory,28 the weakest bond, i.e., the bond between Au2+ and CO in this case, is to be cleaved preferentially in unimolecular dissociation. The observation of the Au(CO)+ fragment, albeit in very small yields, indicates the sensitivity of the present experimental setup. Formation of Au(CO)2+, which is due to further CO adsorption on the product of eq 2 as given in eq 3, requires stabilization through a collision with a He atom of the buffer gas to release excess energy of adsorption. Formation of Au(CO)2+ should therefore follow a termolecular mechanism as follows: Au(CO)+ + CO ⇄ [Au(CO)2+ ]*
(5)
According to eqs 4 and 5, a termolecular rate constant k is expressed as kIII = k′/(nCOnHe), where nCO and nHe represent the densities of the gases in the ion trap. In the present condition, kIII was obtained to be 7.8 × 10−28 cm6 s−1. 4.2. Hydration Effect (1): Au2(H2O)+ + CO. Preadsorption of a H2O molecule influences the rate constants dramatically. On the basis of the reaction-kinetics measurement in Figure 2, reaction pathways of Au2(H2O)+ with CO were identified as follows: +
Au 2(CO)+ + CO → Au(CO)2+ + Au
(6,7)
+
Au 2(H 2O) + O2 → Au 2(H 2O)(O2 )
(11)
5. CONCLUSIONS We have presented reaction kinetics of Au2+ and Au2(H2O)+ with CO and O2. Au2+ was found to be almost unreactive with CO and completely inert with O2. Preadsorption of H2O molecule on Au2+ leads to enhanced (by a factor of 210) adsorption of CO due to the increased rate of Au−Au bond cleavage as well as H2O release to liberate the excess internal energy; the former can be attributed to the weakened Au−Au bond following adsorption of water. Preadsorption of H2O molecule on Au2+ initiates adsorption of up to two molecules of O2. This enhancement and initiation of the reactions by a coadsorbed H2O were attributed to partial electron transfer from the O atom in the H2O to Au2+.
(8)
Adsorption of a CO molecule is always accompanied by cleavage of either Au−Au bond or Au−H2O bond. Bimolecular rate constants were derived for these elementary reactions. The reactivity of Au2(H2O)+, i.e., the decay rate of Au2(H2O)+, was evaluated to be 2.0 × 10−10 cm3 s−1. This bimolecular rate is much higher than that of Au2+ by a factor of 210. Regarding the rate constants of Au atom release from Au2+ and Au2(H2O)+, i.e., the rate constants of eqs 2 and 6, the latter was about 90 times higher than the former. This provides clear evidence of a weakened Au−Au bond due to adsorption of a H2O molecule. Binding energies of Au2+ and Au2 have been evaluated to be 2.4 and 2.3 eV,29 respectively, by collisioninduced dissociation experiment. In addition, density functional theory (DFT) calculation has derived binding energies of Au2 and Au2− to be 2.22 and 1.81 eV,30 respectively. These results imply that the higher the electron density on the gold dimer is charged, the weaker the Au−Au bond. It can therefore be inferred that adsorption of H2O not only aids in release of adsorption energy of CO16 but also leads to partial charge transfer to Au2+, thereby increasing the electron density on the gold dimer, which leads to weakening of the Au−Au bond. This results in increase in CO adsorption on Au2+ through an adsorption channel accompanied by Au−Au bond cleavage as well as liberation of the preadsorbed H2O. 4.3. Hydration Effect (2): Au2(H2O)+ + O2. Adsorption of O2 on Au2(H2O)+ was observed through formation of Au2(H2O)O2+ and Au2(H2O)(O2)2+ as shown in Figure 3. The temporal evolution of intensities of ions revealed the following processes: +
Au 2(H 2O)(O2 )+ + N2 → Au 2(H 2O)(O2 )(N2)+
Bare is completely inert toward O2 molecule, which has been reported earlier.23 However, preadsorption of H2O molecule facilitates adsorption of O2 on Au2+. Here Au2(H2O)(O2)+ was an intermediate product, which can further adsorb one more O2 or N2 molecule. The adsorption of O2 on Au2(H2O)+ can be due to (a) an increase in the number of degrees of freedom of the system, which increase the lifetime of an internally hot complex, and/or (b) partial electron transfer from H2O to O2. The effect of increase in the number of degrees of freedom can be discarded since it is known that even larger gold cluster cations do not adsorb O2, with an exception of Au10+. However, it is well-known that gold cluster anions with even-numbered atoms (odd-electron systems) adsorb O2, which is stabilized by electron donation from the anion to the O2 molecule.23 It has been recently reported that even neutral gold clusters adsorb O2, which once again are stabilized by electron donation from gold cluster to the O2 molecule, similar to the gold cluster anions.31 It can therefore be inferred that adsorption of O2 requires electron density on the gold cluster, which is depleted in cluster cations. However, adsorption of H2O on Au2+ increases the electron density, relative to bare Au2+, thereby increasing its propensity for O2 adsoprtion. The charge modulation observed here is similar to those reported for AunOH− clusters17 and for AunH2+ clusters.18 The present results signify the effect of ubiquitous H2O molecules on the charge modulation.
III
+ ⎧ ⎪→ Au(H 2O)(CO) + Au ⎨ Au 2(H 2O) + CO⎪ + ⎩→ Au 2(CO) + H 2O
(10)
Au2+
(4)
[Au(CO)2+ ]* + He → Au(CO)2+
Au 2(H 2O)(O2 )+ + O2 → Au 2(H 2O)(O2 )2+
■
AUTHOR INFORMATION
Corresponding Author
*(A.T.) E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS T.I. appreciates the Research Fellowship for Young Scientist (Grant DC2) from the Japan Society for Promotion of Science (JSPS). The present study was supported by a Grant-in-Aid for Scientific Research (A) (23245006) from the JSPS and the Special Cluster Research Project of Genesis Research Institute, Inc.
■
REFERENCES
(1) Haruta, M.; Yamada, N.; Kobayashi, T.; Iijima, S. Gold Catalysts Prepared by Coprecipitation for Low-Temperature Oxidation of Hydrogen and of Carbon Monoxide. J. Catal. 1989, 115, 301−309.
(9) 8296
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
(2) Corma, A.; Garcia, H. Supported Gold Nanoparticles as Catalysts for Organic Reactions. Chem. Soc. Rev. 2008, 37, 2096−2126. (3) Stratakis, M.; Garcia, H. Catalysis by Supported Gold Nanoparticles: Beyond Aerobic Oxidative Processes. Chem. Rev. 2012, 112, 4469−4506. (4) Guzman, J.; Gates, B. C. Catalysis by Supported Gold: Correlation between Catalytic Activity for CO Oxidation and Oxidation States of Gold. J. Am. Chem. Soc. 2004, 126, 2672−2673. (5) Herzing, A. A.; Kiely, C. J.; Carley, A. F.; Landon, P.; Hutchings, G. J. Identification of Active Gold Nanoclusters on Iron Oxide Supports for CO Oxidation. Science 2008, 321, 1331−1335. (6) Lin, S. D.; Bollinger, M.; Vannice, M. A. Low Temperature CO Oxidation over Au/TiO2 and Au/SiO2 Catalysts. Catal. Lett. 1993, 17, 245−262. (7) Arrii, S.; Morfin, F.; Renouprez, A. J.; Rousset, J. L. Oxidation of CO on Gold Supported Catalysts Prepared by Laser Vaporization: Direct Evidence of Support Contribution. J. Am. Chem. Soc. 2004, 126, 1199−1205. (8) Vayssilov, G. N.; Gates, B. C.; Rösch, N. Oxidation of Supported Rhodium Clusters by Support Hydroxy Groups. Angew. Chem., Int. Ed. 2003, 42, 1391−1394. (9) Sanchez, A.; Abbet, S.; Heiz, U.; Schneider, W.-D.; Ha1kkinen, H.; Barnett, R. N.; Landman, U. When Gold Is Not Noble: Nanoscale Gold Catalysts. J. Phys. Chem. A 1999, 103, 9573−9578. (10) Ruettinger, W.; Ilinich, O.; Farrauto, R. J. A New Generation of Water as Shift Catalysts for Fuel Cell Applications. J. Power Sources 2003, 118, 61−65. (11) Socaciu, L. D.; Hagen, J.; Bernhardt, T. M.; Wöste, L.; Heiz, U.; Häkkinen, H.; Landman, U. Catalytic CO Oxidation by Free Au2−: Experiment and Theory. J. Am. Chem. Soc. 2003, 125, 10437−10445. (12) Lang, S. M.; Bernhardt, T. M. Methane Activation and Partial Oxidation on Free Gold and Palladium Clusters: Mechanistic Insights into Cooperative and Highly Selective Cluster Catalysis. Faraday Discuss. 2011, 152, 337−351. (13) Wang, Y.; Zhang, D.; Zhu, R.; Zhang, C.; Liu, C. A Density Functional Theory Study of the Water−Gas Shift Reaction Promoted by Neutral, Anionic, and Cationic Gold Dimers. J. Phys. Chem. C 2009, 113, 6215−6220. (14) Popolan, D. M.; Nössler, M.; Mitrić, R.; Bernhardt, T. M.; Bonačić-Koutecký, V. Tuning Cluster Reactivity by Charge State and Composition: Experimental and Theoretical Investigation of CO Binding Energies to AgnAum± (n + m = 3). J. Phys. Chem. A 2011, 115, 951−959. (15) Daté, M.; Haruta, M. Moisture Effect on CO Oxidation over Au/TiO2 Catalyst. J. Catal. 2001, 201, 221−224. (16) Wallace, W. T.; Wrywas, R. B.; Leavitt, A. J.; Wheten, R. L. Adsorption of Carbon Monoxide on Smaller Gold-Cluster Anions in an Atmospheric-Pressure Flow-reactor: Temperature and Humidity Dependence. Phys. Chem. Chem. Phys. 2005, 7, 930−937. (17) Wallace, W. T.; Wrywas, R. B.; Wheten, R. L.; Mitrić, R.; Bonačić-Koutecký, V. Oxygen Adsorption on Hydrated Gold Cluster Anions: Experiment and Theory. J. Am. Chem. Soc. 2003, 125, 8408− 8414. (18) Lang, S. M.; Bernhardt, T. M.; Barnett, R. N.; Yoon, B.; Landman, U. Hydrogen-Promoted Oxygen Activation by Free Gold Cluster Cations. J. Am. Chem. Soc. 2009, 131, 8939−8951. (19) Terasaki, A.; Majima, T.; Kondow, T. Photon-Trap Spectroscopy of Mass-Selected Ions in an Ion Trap: Optical Absorption and Magneto-Optical Effects. J. Chem. Phys. 2007, 127 (231101), 1−4. (20) Ito, T.; Egashira, K.; Tsukiyama, K.; Terasaki, A. Oxidation Processes of Chromium Dimer and Trimer Cations in an Ion Trap. Chem. Phys. Lett. 2012, 538, 19−23. (21) Westergren, J.; Grönbeck, H.; Kim, S.-G.; Tománek, D. Noble Gas Temperature Control of Metal Clusters: A Molecular Dynamics Study. J. Chem. Phys. 1997, 107, 3071−3079. (22) Neumaier, M.; Weigend, F.; Hampe, O.; Kappes, M. M. Binding Energies of CO on Gold Cluster Cations Aun+ (n = 1−65): A Radiative Association Kinetics Study. J. Chem. Phys. 2005, 122 (104702), 1−11.
(23) Cox, D. M.; Brickman, R.; Creegan, K.; Kaldor, A. Gold Clusters: Reactions and Deuterium Uptake. Z. Phys. D 1991, 19, 353− 355. (24) Schumacher, E. DETMECH - Chemical Reaction Kinetics Software; Chemistry Department, University of Bern: Bern, Switzerland, 2009. (25) Gioumousis, G.; Stevenson, D. P. Reactions of Gaseous Molecule Ions with Gaseous Molecules. V. Theory. J. Chem. Phys. 1958, 29, 294−299. (26) Su, T.; Bowers, M. T. Theory of Ion−Polar Molecule Collisions. Comparison with Experimental Charge Transfer Reactions of Rare Gas Ions to Geometric Isomers of Difluorobenzene and Dichloroethylene. J. Chem. Phys. 1973, 58, 3027−3037. (27) Su, T.; Bowers, M. T. Ion−Polar Molecule Collisions: The Effect of Ion Size on Ion−Polar Molecule Rate Constants; the Parametrization of the Average-Dipole-Orientation Theory. Int. J. Mass Spectrom. Ion Processes 1973, 12, 347−356. (28) Levine, R. D. Molecular Reaction Dynamics; Cambridge University Press: Cambridge, U.K., 2005. (29) Becker, S.; Dietrich, G.; Hasse, H.-U.; Klisch, N.; Kluge, H.-J.; Kreisle, D.; Krückeberg, St.; Lindinger, M.; Lützenkirchen, K.; Schweikhard, L.; Weidele, H.; Ziegler, J. Collision Induced Dissociation of Stored Gold Cluster Ions. Z. Phys. D. 1994, 30, 341−348. (30) Häkkinen, H.; Landman, U. Gas-Phase Catalytic Oxidation of CO by Au2−. J. Am. Chem. Soc. 2001, 123, 9704−9705. (31) Woodham, A. P.; Meijer, G.; Fielicke, A. Charge Separation Promoted Activation of Molecular Oxygen by Neutral Gold Clusters. J. Am. Chem. Soc. 2013, 135, 1727−1730.
8297
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
Water-Induced Adsorption of Carbon Monoxide and Oxygen on the Gold Dimer Cation Tomonori Ito,† G. Naresh Patwari,‡ Masashi Arakawa,† and Akira Terasaki*,† †
Department of Chemistry, Graduate School of Sciences, Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan Department of Chemistry, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India
‡
ABSTRACT: It is demonstrated, using tandem mass spectrometry and ion trap, that preadsorption of a H2O molecule on the gold dimer cation, Au2+, enhances adsorption of CO and O2 molecules, which is otherwise inert toward these molecules. The rate of adsorption of CO on Au2(H2O)+ was found to be higher by 2 orders of magnitude relative to bare Au2+. The enhancement of the CO adsorption rate is due to the presence of a reaction channel, which cleaves the Au−Au bond, leading to the formation of Au(H2O)(CO)+. Such an observation can be attributed to weakening of the Au−Au bond upon adsorption of a water molecule. Further, it was also observed that preadsorption of H2O leads to dramatic enhancement of O2 adsorption on the Au2+ ion, which can be attributed to the changes in the electron density following water adsorption.
H 2O + CO ⇄ H 2 + CO2
1. INTRODUCTION Ever since the catalytic activity of gold nanoparticles on metaloxide supports were discovered,1 there has been explosive growth, in the number of investigations and consequent reports in the literature, on the ability of gold nanoparticles, both supported and unsupported, to catalyze a very wide range of reactions.2,3 Among these reactions oxidation of CO is considered as a benchmark reaction to evaluate the catalytic activity since its mechanism is well understood. A large number of reports have confirmed that the catalytic activity not only depends on the size of the nanoparticle but also the support, which influences the oxidation states of metals in nanoparticles.4,5 It can be envisaged that the reactivity of metal nanoparticles can be modulated by the size and also the nature of the support. It has been shown that reducing metal oxides (such as TiO2) have much better catalytic activity in comparison to nonreducing metal oxide supports (such as Al2O3).6,7 The extreme view in this regard is that metal in zero oxidation states do not exist on the supported surfaces.8 The catalytic activity may be controlled more dramatically in the cluster regime by taking an advantage of precise control of the number of constituent atoms. For instance, the low temperature oxidation of CO to CO2 in the presence of O2 over gold clusters has size dependence, wherein the clusters of very low diameters (∼0.5 nm) were found to be most effective catalysts with Au8 being the smallest catalytically active size.9 In this regard, investigations of reactions on the small gas phase clusters facilitate understanding the size and consequent electronic effects thereby elucidating the detailed reaction mechanism of catalytic activity. The catalytic properties of size selected metal clusters have been a subject of great interest in many environmentally benign reactions, such as the water gas shift (WGS) reaction © 2014 American Chemical Society
(1)
which is well-known for the industrial production of clean molecular hydrogen (H2). The WGS reaction is exothermic; therefore, low temperature is desired for better yields. However, low temperature leads to lower reaction rates. The competition between yields and the rates necessitates a suitable catalyst for practical utility. Commercially, the WGS reaction is carried over fine particles of Cu metal dispersed over ZnO/Al2O3. Several other metals such as Fe, Ni, Au, Pt, Ru, Ir, and others supported on appropriate metal oxides have been used as catalysts for the WGS reaction under varying conditions. In the recent times either FeCr or CuZn particles have been industrially used, which are not suitable for a mobile catalyst because of their pyrophoricity.10 Because of the ability of gold nanoparticles to catalyze a very wide range of reactions, it can therefore be surmised that a gold-based catalyst would be ideal in terms of both reactivity and safer handling. Catalytic activities of the gold dimer are well examined for CO oxidation by anions11 and methane activation by cations.12 DFT calculations on the mechanistic aspects of WGS reaction show that gold dimer cation is a better catalyst in comparison with neutral gold dimer and gold dimer anion.13 Reactions involving gas phase metal clusters have tremendous utility due to precise control on the size, charge state, and, in the case of multielement clusters, the composition.14 In the case of CO oxidation over supported gold nanoparticles, it has been shown that trace amounts of water in the reactant mixture significantly alters the catalytic activity.15 This observation can Special Issue: A. W. Castleman, Jr. Festschrift Received: January 31, 2014 Revised: March 1, 2014 Published: March 4, 2014 8293
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
be interpreted as the ability of foreign molecules to control the catalytic activity via the coadsorption effects. Such effects have been reported for CO adsorption on gold-cluster anions in the presence of water vapor,16 for O2 adsorption on gold-hydroxy cluster anions,17 and on gold-cluster cations with preadsorbed H2.18 In the present work, the ability of preadsorbed water molecule to alter the adsorption and reactivity of CO and O2 on the gold dimer cation is investigated. Reaction kinetics were measured by employing a linear radio frequency ion trap, where Au2+ and CO (or O2) interacted for a given reaction time. The effect of a water molecule is evaluated by comparing reaction rate constants measured for Au2+ and Au2(H2O)+. Figure 1. Reaction kinetics of Au2+ with CO measured at room temperature. Amounts of the reactant and product ions are plotted as a function of storage time: Au2+ (closed circles), Au(CO)+ (up-pointing open triangles), and Au(CO)2+ (open squares). The partial pressure of CO was 7.2 × 10−3 Pa. Fitting curves are superimposed by solid lines.
2. EXPERIMENTAL SECTION Details of the experimental apparatus have been described elsewhere.19 Briefly, gold cluster cations were generated in an ion source, where magnetron sputtering of a gold plate is followed by aggregation of sputtered atoms/ions through collisions with helium buffer gas cooled by liquid nitrogen. The cluster ions were guided by radio frequency octopole ion guides. A pick-up cell was placed downstream of the cluster ion source for preadsorption of a H2O molecule on the gold cluster cations. The helium gas was bubbled through water and remixed with pure helium gas to control the composition of water vapor and was introduced into the pick-up cell. The produced ions were mass-selected by a quadrupole mass filter (Max-16000, Extrel CMS) and were stored in a linear octopole ion trap, where a reactant gas (either CO or O2) was introduced continuously along with a He buffer gas. The ions were stored for a given storage time t, were extracted from the ion trap for mass-analysis by a second quadrupole mass filter (Max-4000, Extrel CMS), and were detected by a current meter (Picoammeter 6485, Keithley). Partial pressures of the reactant gas and buffer He gas were measured outside the trap by a residual gas analyzer (RGA100, SRS); actual pressures in the trap were evaluated by calibration based on a known reaction.20 The ion trap and the reactant gases were able to be cooled down to 100 K by liquid nitrogen for low-temperature experiments. Reaction kinetics measurement was performed as follows: Mass-selected ions, Au2(H2O)m+ (m = 0, 1), were loaded into the trap for 200 ms. The number of ions stored typically reached 1 × 107 within the loading time. The partial pressure of the He buffer gas was adjusted to be in the order of 0.1 Pa throughout the experiment, while the partial pressure of a reactant gas was adjusted to be in the range of 10−1−10−5 Pa. The ions stored in the trap were thermalized at the temperature of the He gas within a few milliseconds,21 the time scale of which is much shorter than the reaction rate (∼10 s−1), i.e., the ions were thermalized before reaction. The amounts of reactant and product ions were measured as a function of storage time t to evaluate reaction kinetics.
buffer gas, it was observed that Au2+ adsorbs CO followed by cleavage of Au−Au bond, which is in contrast to a previous report that Au2+ does not adsorb any CO molecules in an ion− cyclotron−resonance experiment.22 The observation of AuCO+ and Au(CO)2+ indicates the sensitivity of the present experimental techniques, given the fact that the reaction rates and yields for the reactive CO adsorption are very small, vide infra. Figure 2 shows temporal evolution of a reaction of the hydrated gold dimer cation, Au2(H2O)+, with CO at room
Figure 2. Reaction kinetics of Au2(H2O)+ with CO measured at room temperature. Amounts of the reactant and product ions are plotted as a function of storage time: Au2(H2O)+ (closed circles), Au2(CO)+ (uppointing triangles), Au(CO)2+ (down-pointing open triangles), and Au(H2O)(CO)+ (open diamonds). The preadsorbed H2O is denoted as W in the legend. The partial pressure of CO was 1.1 × 10−4 Pa. Fitting curves are superimposed by solid lines.
temperature. For this reaction the partial pressure of CO was lowered to approximately 10−5 Pa, which is about 2 orders of magnitude lower than the partial pressure used for the reaction of bare Au2+. The reactant Au2(H2O)+ decreased exponentially with the storage time similarly to observations made for the reaction of bare Au2+. The temporal profile of Au2(CO)+ shows growth followed by decay during the course of the reaction, which indicates that it is a reaction intermediate. However, Au(H2O)(CO)+ and Au(CO)2+ show growth over the course of the reaction, which therefore can be considered as final products.
3. RESULTS 3.1. Reaction with CO. Kinetics measurements for Au2+ + CO reaction were performed at room temperature. The concentration of CO was kept in the order of 10−2 Pa. The amounts of AuCO+ and Au(CO)2+ produced by the reaction of Au2+ with CO under multiple collision conditions as a function of storage time are shown in Figure 1. Also shown is the decrease in the reactant Au2+ in the same time scale. In the present experiment performed in the ion trap filled with He 8294
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
3.2. Adsorption of O2. Bare Au2+ did not adsorb O2 even at temperatures down to 100 K. This is consistent with the previous report.23 The nonreactive character of Au2+ suggests that the interaction energy between O2 and Au2+ should be lower than the thermal energy of the system at 100 K. In striking contrast, a water adduct, Au2(H2O)+, exhibited high reactivity against O2. Figure 3 shows reaction kinetics of
depicted in Figures 1−3, and results are listed in Table 1. The methodology for evaluating the rate constants is based on the experimental conditions. The ion trap was maintained at constant partial pressures of He and CO by introducing the gases continuously into the gas cell of the trap. The number density of the CO gas, nCO, was higher than that of the trapped ions at least by 3 orders of magnitude. Further, the steady-state concentration is maintained by continuous flow of CO gas into the gas cell. Therefore, nCO was practically constant over the entire reaction experiments, hence pseudo-first-order reaction. The temporal evolution of the reaction of Au2+ with CO shown in Figure 1 indicates the following reaction steps: Au 2+ + CO → AuCO+ + Au
(2)
AuCO+ + CO → Au(CO)2+
(3)
We obtained pseudo-first-order rate constants k′ for the above two pathways by fitting the data to rate equations with a software called DETMECH.24 On the basis of bimolecular reaction kinetics, an absolute bimolecular rate constant, kII, is derived as kII = k′/nCO. k′ and kII were determined to be 1.91 s−1 and 9.4 × 10−13 cm3 s−1, respectively. Taking into account an ion-induced dipole interaction effect, simple association rates between ions and molecules can be estimated by the Langevin rate,25 kL = 2πe(4πε0)−1(α/μ)1/2, where e is the elementary charge, α is the polarizability of the target gas molecules, and μ is a reduced mass of the ion−neutral pair. Since CO has a large permanent dipole moment, the association rate tends to be underestimated only with the Langevin rate; it is needed to be corrected for ion−dipole interaction by ADO (averaged dipole orientation) theory. The association rate with the ADO theory is expressed as follows: kADO = (2πe/μ1/2)(4πε0)−1[α1/2 + cμD(2/πkBT)] .26 Here, c represents a locking constant (c = 0.196 for CO),27 μD is the dipole moment of CO (= 0.112 D), kB is Boltzmann’s constant, and T is the temperature of the system. The association rate of CO was calculated to be 6.6 × 10−10 cm3 s−1, whereas the experimentally determined reaction rate was 9.4 × 10−13 cm3 s−1. This means only 0.2% of collisions would indeed lead to CO association. In other words, once a CO molecule interacts with Au2+, the CO molecule tends to be released in most of the cases. This behavior is understood as follows: A CO molecule is bound to a metal cation more strongly than other ordinary molecules because of electronic back-donation from the metal atom to CO. This may create an internally hot intermediate adduct, [Au2(CO)+]*, which proceeds to transition states not only of CO release but also
Figure 3. Reaction kinetics of Au2(H2O)+ with O2 measured at 100 K. Amounts of the reactant and product ions are plotted as a function of storage time: Au2(H2O)+ (closed circles), Au2(H2O)(O2)+ (uppointing triangles), Au2(H2O)(O2)2+ (down-pointing open triangles), and Au2(H2O)(O2)(N2)+ (open diamonds). The partial pressure of O2 was 9.1 × 10−2 Pa. Au2(H2O)(O2)(N2)+ was produced due to residual N2 in the trap. Solid lines are the fits to data according to the reaction kinetics.
Au2(H2O)+ with O2 at 100 K, where three product ions, Au2(H2O)(O2)+, Au2(H2O)(O2)2+, and Au2(H2O)(O2)(N2)+, were produced. Au2(H2O)(O2)+ was a reaction intermediate because it disappeared at a long storage time, by adsorbing an additional O2 molecule leading to the formation of Au2(H2O)(O2)2+, which is the final product under the present reaction conditions. Au2(H2O)+ was able to adsorb up to two O2 molecules in the reaction time scale of about 2 s. Additionally, Au2(H2O)(O2)(N2)+ was observed, which is due to residual N2 in the ion trap of which the partial pressure was estimated to be 1.3 × 10−4 Pa.
4. DISCUSSION 4.1. Evaluation of Reaction Rate Constants. Reactions rate constants were determined, based on all the observed species from the results of reaction kinetics measurements
Table 1. Absolute Bimolecular and Termolecular Rate Constants, kII and kIII, of Each Reaction Steps; the Values Are Subject to a Systematic Error of ∼20% kII (cm3 s−1)
reaction step +
9.4 × 10
(2) Au 2 + CO → AuCO + Au +
(3) AuCO + CO → Au(CO)2
kIII (cm6 s−1)
−13
+
7.8 × 10−28
+
(6) Au 2(H 2O)+ + CO → Au(H 2O)(CO)+ + Au
8.8 × 10−11
(7) Au 2(H 2O)+ + CO → Au 2(CO)+ + H 2O
1.1 × 10−10
(8) Au 2(CO)+ + CO → Au(CO)2+ + Au
1.1 × 10−10
(9) Au 2(H 2O)+ + O2 → Au 2(H 2O)(O2 )+
2.0 × 10−25
(10) Au 2(H 2O)(O2 )+ + O2 → Au 2(H 2O)(O2 )2+
5.2 × 10−26
(11) Au 2(H 2O)(O2 )+ + N2 → Au 2(H 2O)(O2 )(N2)+
1.1 × 10−22 8295
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
of Au release, i.e., [Au2+···(CO)]† and [Au(CO)+···Au]†. According to RRKM theory,28 the weakest bond, i.e., the bond between Au2+ and CO in this case, is to be cleaved preferentially in unimolecular dissociation. The observation of the Au(CO)+ fragment, albeit in very small yields, indicates the sensitivity of the present experimental setup. Formation of Au(CO)2+, which is due to further CO adsorption on the product of eq 2 as given in eq 3, requires stabilization through a collision with a He atom of the buffer gas to release excess energy of adsorption. Formation of Au(CO)2+ should therefore follow a termolecular mechanism as follows: Au(CO)+ + CO ⇄ [Au(CO)2+ ]*
(5)
According to eqs 4 and 5, a termolecular rate constant k is expressed as kIII = k′/(nCOnHe), where nCO and nHe represent the densities of the gases in the ion trap. In the present condition, kIII was obtained to be 7.8 × 10−28 cm6 s−1. 4.2. Hydration Effect (1): Au2(H2O)+ + CO. Preadsorption of a H2O molecule influences the rate constants dramatically. On the basis of the reaction-kinetics measurement in Figure 2, reaction pathways of Au2(H2O)+ with CO were identified as follows: +
Au 2(CO)+ + CO → Au(CO)2+ + Au
(6,7)
+
Au 2(H 2O) + O2 → Au 2(H 2O)(O2 )
(11)
5. CONCLUSIONS We have presented reaction kinetics of Au2+ and Au2(H2O)+ with CO and O2. Au2+ was found to be almost unreactive with CO and completely inert with O2. Preadsorption of H2O molecule on Au2+ leads to enhanced (by a factor of 210) adsorption of CO due to the increased rate of Au−Au bond cleavage as well as H2O release to liberate the excess internal energy; the former can be attributed to the weakened Au−Au bond following adsorption of water. Preadsorption of H2O molecule on Au2+ initiates adsorption of up to two molecules of O2. This enhancement and initiation of the reactions by a coadsorbed H2O were attributed to partial electron transfer from the O atom in the H2O to Au2+.
(8)
Adsorption of a CO molecule is always accompanied by cleavage of either Au−Au bond or Au−H2O bond. Bimolecular rate constants were derived for these elementary reactions. The reactivity of Au2(H2O)+, i.e., the decay rate of Au2(H2O)+, was evaluated to be 2.0 × 10−10 cm3 s−1. This bimolecular rate is much higher than that of Au2+ by a factor of 210. Regarding the rate constants of Au atom release from Au2+ and Au2(H2O)+, i.e., the rate constants of eqs 2 and 6, the latter was about 90 times higher than the former. This provides clear evidence of a weakened Au−Au bond due to adsorption of a H2O molecule. Binding energies of Au2+ and Au2 have been evaluated to be 2.4 and 2.3 eV,29 respectively, by collisioninduced dissociation experiment. In addition, density functional theory (DFT) calculation has derived binding energies of Au2 and Au2− to be 2.22 and 1.81 eV,30 respectively. These results imply that the higher the electron density on the gold dimer is charged, the weaker the Au−Au bond. It can therefore be inferred that adsorption of H2O not only aids in release of adsorption energy of CO16 but also leads to partial charge transfer to Au2+, thereby increasing the electron density on the gold dimer, which leads to weakening of the Au−Au bond. This results in increase in CO adsorption on Au2+ through an adsorption channel accompanied by Au−Au bond cleavage as well as liberation of the preadsorbed H2O. 4.3. Hydration Effect (2): Au2(H2O)+ + O2. Adsorption of O2 on Au2(H2O)+ was observed through formation of Au2(H2O)O2+ and Au2(H2O)(O2)2+ as shown in Figure 3. The temporal evolution of intensities of ions revealed the following processes: +
Au 2(H 2O)(O2 )+ + N2 → Au 2(H 2O)(O2 )(N2)+
Bare is completely inert toward O2 molecule, which has been reported earlier.23 However, preadsorption of H2O molecule facilitates adsorption of O2 on Au2+. Here Au2(H2O)(O2)+ was an intermediate product, which can further adsorb one more O2 or N2 molecule. The adsorption of O2 on Au2(H2O)+ can be due to (a) an increase in the number of degrees of freedom of the system, which increase the lifetime of an internally hot complex, and/or (b) partial electron transfer from H2O to O2. The effect of increase in the number of degrees of freedom can be discarded since it is known that even larger gold cluster cations do not adsorb O2, with an exception of Au10+. However, it is well-known that gold cluster anions with even-numbered atoms (odd-electron systems) adsorb O2, which is stabilized by electron donation from the anion to the O2 molecule.23 It has been recently reported that even neutral gold clusters adsorb O2, which once again are stabilized by electron donation from gold cluster to the O2 molecule, similar to the gold cluster anions.31 It can therefore be inferred that adsorption of O2 requires electron density on the gold cluster, which is depleted in cluster cations. However, adsorption of H2O on Au2+ increases the electron density, relative to bare Au2+, thereby increasing its propensity for O2 adsoprtion. The charge modulation observed here is similar to those reported for AunOH− clusters17 and for AunH2+ clusters.18 The present results signify the effect of ubiquitous H2O molecules on the charge modulation.
III
+ ⎧ ⎪→ Au(H 2O)(CO) + Au ⎨ Au 2(H 2O) + CO⎪ + ⎩→ Au 2(CO) + H 2O
(10)
Au2+
(4)
[Au(CO)2+ ]* + He → Au(CO)2+
Au 2(H 2O)(O2 )+ + O2 → Au 2(H 2O)(O2 )2+
■
AUTHOR INFORMATION
Corresponding Author
*(A.T.) E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS T.I. appreciates the Research Fellowship for Young Scientist (Grant DC2) from the Japan Society for Promotion of Science (JSPS). The present study was supported by a Grant-in-Aid for Scientific Research (A) (23245006) from the JSPS and the Special Cluster Research Project of Genesis Research Institute, Inc.
■
REFERENCES
(1) Haruta, M.; Yamada, N.; Kobayashi, T.; Iijima, S. Gold Catalysts Prepared by Coprecipitation for Low-Temperature Oxidation of Hydrogen and of Carbon Monoxide. J. Catal. 1989, 115, 301−309.
(9) 8296
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
The Journal of Physical Chemistry A
Article
(2) Corma, A.; Garcia, H. Supported Gold Nanoparticles as Catalysts for Organic Reactions. Chem. Soc. Rev. 2008, 37, 2096−2126. (3) Stratakis, M.; Garcia, H. Catalysis by Supported Gold Nanoparticles: Beyond Aerobic Oxidative Processes. Chem. Rev. 2012, 112, 4469−4506. (4) Guzman, J.; Gates, B. C. Catalysis by Supported Gold: Correlation between Catalytic Activity for CO Oxidation and Oxidation States of Gold. J. Am. Chem. Soc. 2004, 126, 2672−2673. (5) Herzing, A. A.; Kiely, C. J.; Carley, A. F.; Landon, P.; Hutchings, G. J. Identification of Active Gold Nanoclusters on Iron Oxide Supports for CO Oxidation. Science 2008, 321, 1331−1335. (6) Lin, S. D.; Bollinger, M.; Vannice, M. A. Low Temperature CO Oxidation over Au/TiO2 and Au/SiO2 Catalysts. Catal. Lett. 1993, 17, 245−262. (7) Arrii, S.; Morfin, F.; Renouprez, A. J.; Rousset, J. L. Oxidation of CO on Gold Supported Catalysts Prepared by Laser Vaporization: Direct Evidence of Support Contribution. J. Am. Chem. Soc. 2004, 126, 1199−1205. (8) Vayssilov, G. N.; Gates, B. C.; Rösch, N. Oxidation of Supported Rhodium Clusters by Support Hydroxy Groups. Angew. Chem., Int. Ed. 2003, 42, 1391−1394. (9) Sanchez, A.; Abbet, S.; Heiz, U.; Schneider, W.-D.; Ha1kkinen, H.; Barnett, R. N.; Landman, U. When Gold Is Not Noble: Nanoscale Gold Catalysts. J. Phys. Chem. A 1999, 103, 9573−9578. (10) Ruettinger, W.; Ilinich, O.; Farrauto, R. J. A New Generation of Water as Shift Catalysts for Fuel Cell Applications. J. Power Sources 2003, 118, 61−65. (11) Socaciu, L. D.; Hagen, J.; Bernhardt, T. M.; Wöste, L.; Heiz, U.; Häkkinen, H.; Landman, U. Catalytic CO Oxidation by Free Au2−: Experiment and Theory. J. Am. Chem. Soc. 2003, 125, 10437−10445. (12) Lang, S. M.; Bernhardt, T. M. Methane Activation and Partial Oxidation on Free Gold and Palladium Clusters: Mechanistic Insights into Cooperative and Highly Selective Cluster Catalysis. Faraday Discuss. 2011, 152, 337−351. (13) Wang, Y.; Zhang, D.; Zhu, R.; Zhang, C.; Liu, C. A Density Functional Theory Study of the Water−Gas Shift Reaction Promoted by Neutral, Anionic, and Cationic Gold Dimers. J. Phys. Chem. C 2009, 113, 6215−6220. (14) Popolan, D. M.; Nössler, M.; Mitrić, R.; Bernhardt, T. M.; Bonačić-Koutecký, V. Tuning Cluster Reactivity by Charge State and Composition: Experimental and Theoretical Investigation of CO Binding Energies to AgnAum± (n + m = 3). J. Phys. Chem. A 2011, 115, 951−959. (15) Daté, M.; Haruta, M. Moisture Effect on CO Oxidation over Au/TiO2 Catalyst. J. Catal. 2001, 201, 221−224. (16) Wallace, W. T.; Wrywas, R. B.; Leavitt, A. J.; Wheten, R. L. Adsorption of Carbon Monoxide on Smaller Gold-Cluster Anions in an Atmospheric-Pressure Flow-reactor: Temperature and Humidity Dependence. Phys. Chem. Chem. Phys. 2005, 7, 930−937. (17) Wallace, W. T.; Wrywas, R. B.; Wheten, R. L.; Mitrić, R.; Bonačić-Koutecký, V. Oxygen Adsorption on Hydrated Gold Cluster Anions: Experiment and Theory. J. Am. Chem. Soc. 2003, 125, 8408− 8414. (18) Lang, S. M.; Bernhardt, T. M.; Barnett, R. N.; Yoon, B.; Landman, U. Hydrogen-Promoted Oxygen Activation by Free Gold Cluster Cations. J. Am. Chem. Soc. 2009, 131, 8939−8951. (19) Terasaki, A.; Majima, T.; Kondow, T. Photon-Trap Spectroscopy of Mass-Selected Ions in an Ion Trap: Optical Absorption and Magneto-Optical Effects. J. Chem. Phys. 2007, 127 (231101), 1−4. (20) Ito, T.; Egashira, K.; Tsukiyama, K.; Terasaki, A. Oxidation Processes of Chromium Dimer and Trimer Cations in an Ion Trap. Chem. Phys. Lett. 2012, 538, 19−23. (21) Westergren, J.; Grönbeck, H.; Kim, S.-G.; Tománek, D. Noble Gas Temperature Control of Metal Clusters: A Molecular Dynamics Study. J. Chem. Phys. 1997, 107, 3071−3079. (22) Neumaier, M.; Weigend, F.; Hampe, O.; Kappes, M. M. Binding Energies of CO on Gold Cluster Cations Aun+ (n = 1−65): A Radiative Association Kinetics Study. J. Chem. Phys. 2005, 122 (104702), 1−11.
(23) Cox, D. M.; Brickman, R.; Creegan, K.; Kaldor, A. Gold Clusters: Reactions and Deuterium Uptake. Z. Phys. D 1991, 19, 353− 355. (24) Schumacher, E. DETMECH - Chemical Reaction Kinetics Software; Chemistry Department, University of Bern: Bern, Switzerland, 2009. (25) Gioumousis, G.; Stevenson, D. P. Reactions of Gaseous Molecule Ions with Gaseous Molecules. V. Theory. J. Chem. Phys. 1958, 29, 294−299. (26) Su, T.; Bowers, M. T. Theory of Ion−Polar Molecule Collisions. Comparison with Experimental Charge Transfer Reactions of Rare Gas Ions to Geometric Isomers of Difluorobenzene and Dichloroethylene. J. Chem. Phys. 1973, 58, 3027−3037. (27) Su, T.; Bowers, M. T. Ion−Polar Molecule Collisions: The Effect of Ion Size on Ion−Polar Molecule Rate Constants; the Parametrization of the Average-Dipole-Orientation Theory. Int. J. Mass Spectrom. Ion Processes 1973, 12, 347−356. (28) Levine, R. D. Molecular Reaction Dynamics; Cambridge University Press: Cambridge, U.K., 2005. (29) Becker, S.; Dietrich, G.; Hasse, H.-U.; Klisch, N.; Kluge, H.-J.; Kreisle, D.; Krückeberg, St.; Lindinger, M.; Lützenkirchen, K.; Schweikhard, L.; Weidele, H.; Ziegler, J. Collision Induced Dissociation of Stored Gold Cluster Ions. Z. Phys. D. 1994, 30, 341−348. (30) Häkkinen, H.; Landman, U. Gas-Phase Catalytic Oxidation of CO by Au2−. J. Am. Chem. Soc. 2001, 123, 9704−9705. (31) Woodham, A. P.; Meijer, G.; Fielicke, A. Charge Separation Promoted Activation of Molecular Oxygen by Neutral Gold Clusters. J. Am. Chem. Soc. 2013, 135, 1727−1730.
8297
dx.doi.org/10.1021/jp501111f | J. Phys. Chem. A 2014, 118, 8293−8297
