High Catalytic Activity of Au Clusters Supported on ZnO Nanosheets

Aug 20, 2014 - Hassan S. Al Qahtani , Rintaro Higuchi , Takayoshi Sasaki , Jason F. Alvino , Gregory F. Metha , Vladimir B. Golovko , Rohul Adnan , Gu...
1 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCC

High Catalytic Activity of Au Clusters Supported on ZnO Nanosheets Na Guo,† Ruifeng Lu,§,† Shuanglong Liu,† Ghim Wei Ho,∥ and Chun Zhang*,†,‡ †

Department Department § Department ∥ Department ‡

of of of of

Physics, National University of Singapore, 2 Science Drive 3, Singapore 117542 Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543 Applied Physics, Nanjing University of Science and Technology, Nanjing, China 210094 Electrical and Computer Engineering, National University of Singapore, Singapore, 117546

S Supporting Information *

ABSTRACT: Catalytic activity of metal clusters supported on a two-dimensional (2d) nanosheet has attracted great interest recently. In this paper, via first-principles calculations, we investigate for the first time Au catalysis on recently proposed 2d oneatom-thick ZnO nanosheets. An Au8 cluster adsorbed on a 2d ZnO sheet (Au8/ZnO) is chosen as a model catalyst, and the CO oxidation is used as the probe reaction for the catalytic activity. When adsorbed on 2d one-atom-thick ZnO, the Au8 cluster is found to adopt a planar structure and exhibit high catalytic activities for the CO oxidation with a low reaction barrier around 0.3 eV. The catalyzed CO oxidation highly prefers the Eley− Rideal (ER) mechanism, and the coadsorption of CO and O2 on the Au8 cluster plays an essential role in the ER type of reaction in which an additional CO approaches O2 in a vacuum. The high catalytic activity (low reaction barrier) is due to the charge transfer from Au8/ZnO to the antibonding 2π* orbital of the O2 molecule that is facilitated by the coadsorption of a CO molecule. These findings provide new opportunities for the future development of ZnO-based catalysts.

■

devices such as field effect transistors,14 spintronic devices,15 light-emitting diods,16 and nanoscale electricity generators17 have been proposed both theoretically and experimentally. A recent experiment demonstrated the evidence of strong interactions between the encapsulated Au cluster and the ZnO outside, which could lead to high catalytic activity.18 Despite the ever-increasing interest in ZnO and also metal nanocatalysis in last several decades, studies pertaining to ZnOsupported metal catalysis still fell short. So far, ZnO has not been considered as a good candidate for substrates of catalysts partly due to the low chemical activity of bulk ZnO. Here, via first-principles calculations, we show the high catalytic activity of Au clusters supported on a 2d one-atom-thick ZnO sheet, suggesting the great potential of the 2d ZnO nanosheets in catalysis chemistry. Interesting effects of the 2d ZnO support on the catalyzed reaction that are different from those of conventional bulk metal oxide substrates are also revealed. We expect these studies to stimulate new experiments along this direction.

INTRODUCTION Metal catalysis on two-dimensional (2d) one-atom-thick substrates has attracted great interest recently. Various 2d substrates such as graphene,1−6 hexagonal BN (h-BN),7 and polymeric metal−phthalocyanine,8,9 have been considered in the literature. Unusual effects of these 2d substrates on the catalyzed chemical reactions that could lead to novel applications were reported. For example, it was shown that a moderate mechanical strain on graphene substrate could greatly enhance the stabilization and catalytic activity of the supported metal clusters, implying a possible way to control reversibly the chemical reactions by mechanical force.1 It was also suggested that on graphene or polymeric metal−phthalocyanine substrate, high-performance single-atom catalysts that are highly desired could be realized.2,8,9 These interesting phenomena indicate the great promise of the 2d one-atom-thick support for the design of novel catalysts. However, the difficulty in large-scale production hinders the industry applications of previously proposed systems. Great research effort along this direction has been paid to the search for a better system (a better combination of metal catalysts and 2d substrates) that is efficient yet cheap and easy to fabricate. In this paper, we report for the first time the first-principles investigation of Au catalysis on 2d one-atom-thick ZnO sheets. Our studies offer a new avenue for the future design of 2d substrate-supported catalysts. ZnO has been regarded as a key material for future technology due to its excellent mechanical, electric, and optical properties as well as its unique ability to form various kinds of nanomaterials,10−13 such as nanowires, nanobelts, nanotubes, and recently 2d nanosheets. Many novel ZnO-based electronic © XXXX American Chemical Society

■

COMPUTATIONAL METHODS

First-principle DFT calculations were performed by utilizing Vienna Ab Initio Simulation Package (VASP) code.19,20 The plane wave basis with the cutoff energy of 450 eV and the generalized gradient approximation (GGA) in the Perdew− Burke−Ernzerh (PBE) format21 were used in all calculations. Received: July 10, 2014 Revised: August 17, 2014

A

dx.doi.org/10.1021/jp506877z | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

energy is defined here as Ead = EAu + EZnO − EAu/ZnO). For the optimized P2 adsorption configuration, the average distance between the Au and the nearest Zn is estimated to be about 2.6 Å. In Figure 2, we plot the isosurface of the charge difference

Relativistic effects were included in the pseudopotential of Au. The structural relaxations were carried out until the Hellmann− Feynman force on each atom is less than 0.01 eV/Å, and the energy convergence criterion was set to 10−5 eV. A 3 × 3 supercell (Figure 1) and the 3 × 3 × 1 k sampling in Brillouin

Figure 2. Isosurface of charge difference for the adsorption of P2 Au8 on ZnO: (a) The top view and (b) the side view. The blue/green color denotes the accumulation/depletion of electrons. After the adsorption, the interface between Au8 and ZnO loses electrons and the Au8 cluster gains electrons. The calculation shows that approximately 0.3 electrons are transferred to the Au8 cluster from the interfacial region.

Figure 1. Atomic structures of three Au8 isomers (3d, P1, and P2) before and after the adsorption on ZnO nanosheet. Color scheme: Au in yellow, oxygen in red, and zinc in gray. After adsorption, P2 is the most stable one.

for the optimized P2 adsorption. From the figure, we can see that after the adsorption, the Au8 cluster is negatively charged and approximately there are 0.3 electrons transferred from the interfacial region to the cluster. We then consider the adsorption of an O2 or CO molecule on P2-Au8/ZnO, which is the first step of investigating the catalytic activity of the system. For the adsorption of a single O2, there are four inequivalent adsorption sites (A−D), as depicted in Figure 3a. We have found that there is no stable

zone were employed in all calculations. Along the direction perpendicular to the ZnO sheet, 15 Å of vacuum is included in the supercell to avoid interaction of periodic images. The reaction barrier of the catalyzed CO oxidation was determined by the climbing-image nudged elastic band (cNEB) method.22−24 Five images and the convergence criteria of 0.01 eV were used in cNEB calculations. The spin polarization is considered in all calculations.

■

RESULTS AND DISCUSSIONS The 2d one-atom-thick ZnO sheet with a honeycomb lattice was first predicted theoretically25 to be an analogue of h-BN where B and N atoms are replaced by Zn and O, respectively. The flat, layered 2d ZnO was then observed in experiments.10,11 Since then, the mechanical, electronic, and magnetic properties of the 2d one-atom-thick ZnO sheets have been intensively studied,26−29 but the catalytic activity of 2d ZnO-based systems has never been discussed. In this paper, via computational modeling and simulation, we investigated the possibility of using 2d ZnO sheets as substrates for metal nanocatalysts. An Au8 cluster (which is an often-used Au nanocatalyst in the literature) supported on a 2d ZnO sheet (Au8/ZnO as shown in Figure 1) was chosen as the model catalyst, and the CO oxidation was used to probe the activity of the catalyst. Note that, in the supercell structure shown in Figure 1, the distance between the Au8 cluster and its nearest image is around 8 Å, which is far enough to avoid interactions between them. All calculations were done under the framework of density functional theory (DFT). Au8 clusters have three different isomers in the gas phase, one three-dimensional (3d) and two planar (P1 and P2) structures, as shown in Figure 1a−c. The predicted geometries of the Au8 clusters (the Au−Au bond lengths and angles) by our calculations agree well with previously reported results.1,2 For the 2d ZnO sheet, our calculations yielded the optimized Zn−O bond length of around 1.9 Å, also in good agreement with literature.19,20 When adsorbed on 2d ZnO, we found that the P2 adsorption configuration (Figure 1c) is the most stable. The adsorption energy of the P2 configuration is about 0.3 eV higher than the P1 and 3d cases (Note that the adsorption

Figure 3. Optimized adsorption configurations for O2 and/or CO: (a) the adsorption of a single O2. A−D represent four possible adsorption sites. The O2 molecule weakly binds on A−C with the adsorption energy of less than 0.1 eV. There is no binding on site D. (b) The most stable adsorption configuration for a single CO molecule. The adsorption energy is around 0.6 eV. (c) The coadsorption of CO and O2. In this case, the adsorption energy of the O2 molecule is about 0.6 eV. The carbon atom is in black.

adsorption of a single O2 in all these sites. In particular, the O2 is only weakly adsorbed on sites A−C with adsorption energies of less than 0.1 eV, and there is no adsorption (negative adsorption energy) of O2 on site D. For the adsorption of a single CO molecule, we also tested all possible sites, and the most stable adsorption configuration is shown in Figure 3b. The adsorption energy for this case is about 0.6 eV. The C−Au bond length is around 2.0 Å, and the C−O bond length after adsorption is about 1.2 Å, slightly longer than the gas-phase value of 1.1 Å. The adsorption energy of the single molecule (O2 or CO) is calculated by Ead = Emolecule + EAu/ZnO − E molecule/Au/ZnO. The aforementioned weak binding of O2 is not desired for the catalyzed CO oxidation, since it is well-known that the ratelimiting step in the reaction is the O−O bond breaking and to break the O−O bond in cases of weak adsorption needs large B

dx.doi.org/10.1021/jp506877z | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

CO oxidation has two steps: (1) CO + O2 → CO2 + O and (2) O + CO → CO2. We simulated both the Langmuir− Hinshelwood (LH) and the Eley−Rideal (ER) types of reactions for the catalyzed CO oxidation. For the coadsorption of CO and O2 on P2-Au8/ZnO, we found that the first step of the LH reaction has a high reaction barrier (>3.0 eV); therefore it will not happen at room temperature. The calculated energy profile along the reaction path for the LH reaction is shown in the Supporting Information (Figure S1). We then focused on the ER type of reaction, in which an additional CO molecule approaches the O2 from a vacuum. It has been found that the reaction barrier of the ER mechanism depends on the initial position and orientation of the approaching CO. We tried several different initial configurations and found that the barrier varies between 0.3 and 0.5 eV. In Figure 5, we showed a

amounts of energy. In order to achieve a stable O2 binding, we examined the coadsorption of O2 and CO and found that the preadsorption of a CO molecule can greatly stabilize the binding of O2. We considered various adsorption configurations. The predicted lowest-energy coadsorption configuration is shown in Figure 3c, in which the CO binds on the same site as in Figure 3b and the O2 takes the site B. After the preadsorption of the CO, the adsorption energy the O2 is estimated to be 0.6 eV. The O−O bond is significantly elongated in this case to 1.44 Å compared with the calculated gas-phase value of 1.24 Å. Note that for the coadsorption case, the adsorption energies of CO and O2 are almost the same, which is helpful to avoid the notorious CO poisoning problem. To understand the effects of CO preadsorption on binding of the O2 molecule, we performed the Bader charge analysis for different adsorption cases (adsorption of single O2 or CO and the coadsorption). In Table 1, we show the change of net Table 1. Net Charge Difference of Various Atoms under Different Adsorption Configurationsa net charge difference at different atoms (in e) atom

CO adsorption

O2 adsorption

CO + O2 Coadsorption

Au1 Au2 Au3 CO O2

−0.05 0.01 0.11 0.03 N/A

0.03 −0.26 −0.17 N/A 0.70

−0.19 −0.20 −0.17 −0.65 1.47

a

The positive/negative value denotes the gain/loss of electrons. N/A = not applicable.

Figure 5. Calculated energy profile of the first step of the ER type of catalyzed CO oxidation. The energy profile is obtained by the cNEB method. The preadsorbed CO molecule greatly enhances the catalytic activity of the system. In this case, the reaction barrier is estimated to be 0.31 eV.

charges on the adsorbed molecules (O2 or CO) and the Au atoms bonded with the molecules after the adsorption. It can be clearly seen from the table that when CO and O2 are coadsorbed, compared with the single-molecule adsorption, the CO molecule loses significant amounts of electrons (∼0.7 e), and the O2 molecule gains nearly the same amount of electrons. The CO-assisted electron transfer to the O2 molecule greatly enhances the binding of O2. In Figure 4, we plotted the

calculated reaction path with the barrier around 0.3 eV. In the initial state (IS), the distance between the C atom of the CO in vacuum and the nearest O atom of the adsorbed O2 (dCO−O2) is 3.0 Å. The transition state (TS) features a peroxo-type O−O− C−O intermediate complex with a dCO−O2 of 1.7 Å. The O−O bond length in TS is about 1.6 Å. The low reaction barrier of 0.3 eV clearly suggests that the Au8/ZnO under study is an excellent catalyst for CO oxidation. In the second step of the reaction, the remaining O atom is removed by another CO molecule that completes the cycle of the reaction. Our calculations gave a low reaction barrier of around 0.27 eV for the second step of the reaction. The calculated reaction path for the second step is shown in Figure S2 in the Supporting Information. It worth mentioning here that CO2 is weakly adsorbed on single-layer ZnO with the adsorption energy of less than 0.2 eV.

Figure 4. Isosurface of charge difference for the coadsorption case. The charge difference is calculated as the difference between the charge distribution for the combined system and the summation of the isolated systems, δρ = ρCO + ρO2 + ρAuB/ZnO − ρCombined: (a) the top view and (b) the side view. The blue/green color denotes the accumulation/depletion of electrons.

■

CONCLUSION In conclusion, we report for the first time the first-principles investigations of Au catalysis on 2d one-atom-thick ZnO nanosheets. An Au8 cluster adsorbed on a 2d ZnO sheet (Au8/ ZnO) is chosen as a model catalyst, and the CO oxidation is used as the probe reaction for the catalytic activity. We found that, when adsorbed on 2d ZnO sheets, the Au8 cluster adopts a planar structure (P2) and exhibit high catalytic activities for the CO oxidation reaction with a low reaction barrier of around 0.3 eV. The catalyzed CO oxidation highly prefers the Eley−Rideal (ER) mechanism, and the coadsorption of CO and O2 on the

isosurface of charge accumulation and depletion after the coadsorption of CO and O2. The electron transfer to the antibonding 2π* orbital of O2 can be clearly seen in the figure. As consequences of such electron transfer, the O−O bond is significantly elongated. The elongation of the O−O bond suggests the activation of the O2 molecule for the CO oxidation. In general, the catalyzed C

dx.doi.org/10.1021/jp506877z | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

Characteristics of ZnO Nanowire Field Effect Transistors by Proton Irradiation. ACS Nano 2010, 4, 811−818. (15) Dai, Z.; Nurbawono, A.; Zhang, A.; Zhou, M.; Feng, Y.; Ho, G.; Zhang, C. C-Doped ZnO Nanowires: Electronic Structures, Magnetic Properties, and a Possible Spintronic Device. J. Chem. Phys. 2011, 134, 104706−5. (16) Lai, E.; kim, W.; Yang, P. Vertical Nanowire Array-Based Light Emitting Diodes. Nano Res. 2008, 1, 123−128. (17) Wang, Z.; Song, J. Piezoelectric Nanogenerators Based on Zinc Oxide Nanowire Arrays. Science 2006, 312, 242−246. (18) Liu, X.; Liu, M.; Luo, Y.; Mou, C.; Lin, S.; Cheng, H.; Chen, J.; Lee, J.; Lin, T. Strong Metal−Support Interactions between Gold Nanoparticles and ZnO Nanorodes in CO Oxidation. J. Am. Chem. Soc. 2012, 134, 10251−8. (19) Kresse, G.; Hafner, J. Abinitio Molecular-Dynamics for LiquidMetals. Phys. Rev. B 1993, 47 (1), 558−561. (20) Kresse, G.; Furthmuller, J. Efficient Iterative Schemes For Ab Initio Total-Energy Calculations Using A Plane-Wave Basis Set. Phys. Rev. B 1996, 54 (16), 11169−11186. (21) Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50, 17953−17979. (22) Sheppard, D.; Xiao, P.; Chemelewski, W.; Johnson, D.; Henkelman, G. A Generalized Solid-State Nudged Elastic Band Method. J. Chem. Phys. 2012, 136, 074103−074110. (23) Sheppard, D.; Xiao, P.; Chemelewski, W.; Johnson, D.; Henkelman, G. A Generalized Solid-State Nudged Elastic Band Method. J. Chem. Phys. 2012, 136, 074103−074110. (24) Henkelman, G.; Uberuaga, B.; Jónsson, H. A Climbing Image Nudged Elastic Band Method for Finding Saddle Points and Minimum Energy Paths. J. Chem. Phys. 2000, 113, 9901−9904. (25) Freeman, C.; Claeyssens, F.; Allan, N. Graphitic Nanofilms as Precursor to Wurtzite Films: Theory. Phys. Rev. Lett. 2006, 96, 066102−4. (26) Demiroglu, I.; Stradi, D.; Illas, F.; Bromley, S. A Theoretical Study of A ZnO Graphene Analogues: Adsorption on Ag(111) and Hydrogen Transport. J. Phys.: Condens. Matter. 2011, 23, 334215− 334220. (27) Topsakal, M.; Cahangirov, S.; Bekaroglu, E.; Ciraci, S. FirstPrinciples Study of Zinc Oxide Honeycomb Structures. Phys. Rev. B 2009, 80 (23), 235119−2351132. (28) Kou, L.; Li, C.; Zhang, Z. H.; Guo, W. Electric-Field and Hydrogen-Passivation-Induced Band Modulations in Armchair ZnO Nanoribbons. J. Phys. Chem. C 2010, 114 (2), 1326−1330. (29) Kou, L.; Li, C.; Zhang, Z. H.; Guo, W. Tuning Magnetism in Zigzag ZnO Nanoribbons by Transverse Electric Fields. ACS Nano 2010, 4, 2124−2128.

Au8 cluster plays an essential role in the ER type of reaction in which an additional CO approaches O2 in a vacuum. These findings provide new opportunities for both the design of heterogeneous metal catalysts and the future applications of ZnO-based materials.

■

ASSOCIATED CONTENT

S Supporting Information *

The reaction path of the Langmuir−Hinshelwood mechanism for the first step of CO oxidation (Figure S1) and the reaction path of the Eley−Rideal mechanism for the second step of CO oxidation (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]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the support from Ministry of Education (Singapore) and NUS academic research funds (Grant Nos. R144-000-325-112 and R-144-000-298-112). Computations were performed at the Graphene Research Centre and the Centre for Computational Science and Engineering at NUS.

■

REFERENCES

(1) Zhou, M.; Zhang, A.; Dai, Z.; Feng, Y.; Zhang, C. StrainEnhanced Stabilization and Catalytic Activity of Metal Nanoclusters on Graphene. J. Phys. Chem. C 2010, 114, 16541−16546. (2) Zhou, M.; Zhang, A.; Dai, Z.; Zhang, C.; Feng, Y. Greatly Enhanced Adsorption and Catalytic Activity of Au and Pt Clusters on Defective Graphene. J. Chem. Phys. 2010, 132, 194704−194705. (3) Lu, Y.; Zhou, M.; Zhang, C.; Feng, Y. Metal-Embedded Graphene: A Possible Catalyst with High Activity. J. Phys. Chem. C 2009, 113, 20156−20160. (4) Zhou, M.; Lu, Y.; Cai, Y.; Zhang, C.; Feng, Y. Adsorption of Gas Molecules on Transition Metal Embedded Graphene: A Search for High-Performance Graphene-Based Catalysts and Gas Sensors. Nanotechnology 2011, 22, 385502−385509. (5) Yang, M.; Zhou, M.; Zhang, A.; Zhang, C. Graphene Oxide: An Ideal Support For Gold Nanocatalysts. J. Phys. Chem. C 2012, 116, 22336−22340. (6) Yin, H.; Tang, H.; Wang, D.; Gao, Y.; Tang, Z. Facile Synthesis of Surfactant-Free Au Cluster/Graphene Hybrids for High-Performance Oxygen Reduction Reaction. ACS Nano 2012, 6, 8288−8297. (7) Gao, M.; Lyalin, A.; Taketsugu, T. CO Oxidation on h-BN Supported Au Atom. J. Chem. Phys. 2013, 138, 034701−034708. (8) Chen, X.; Yan, J.; Jiang, Q. Single Layer of Polymeric Metal− Phthalocyanine Substrate To Realize Single Pt Atom Catalysts with Uniform Distribution. J. Phys. Chem. C 2014, 118, 2122−2128. (9) Li, Y.; Sun, Q. The Superior Catalytic CO Oxidation Capacity of a Cr−Phthalocyanine Porous Sheet. Sci. Rep. 2014, 4, 4098−4105. (10) Weirum, G.; Barcaro, G.; Fortunelli, A.; Weber, F.; Shennach, R.; Surnev, S.; Netzer, F. Growth and Surface Structure of Zinc Oxide Layers on Pd(111) Surface. J. Phys. Chem. C 2010, 114, 15432−15439. (11) Tusche, C.; Meyerheim, H.; Kirschner, J. Observation of Depolarized ZnO(0001) Monolayers: Formation of Unreconstructed Planar Sheets. Phys. Rev. Lett. 2007, 99, 026102−4. (12) Wang, Z. ZnO Nanowire and Nanobelt Platform for Nanotechnology. Mater. Sci. Eng. R. 2009, 64, 33−71. (13) Wang, X.; Song, J.; Wang, Z. Single-Crystal Nanocastles of ZnO. Chem. Phys. Lett. 2006, 424, 86−90. (14) Hong, W.; Jo, G.; Sohn, J.; Park, W.; Choe, M.; Wang, G.; Kahng, Y.; Welland, M.; Lee, T. Tuning of the Electronic D

dx.doi.org/10.1021/jp506877z | J. Phys. Chem. C XXXX, XXX, XXX−XXX