Strong Metal–Support Interactions Achieved by Hydroxide-to-Oxide

Sep 29, 2017 - The strong metal–support interactions (SMSI) are well-known but crucial for preparation of supported metal nanoparticle catalysts, wh...
0 downloads 19 Views 608KB Size
Subscriber access provided by LONDON METROPOLITAN UNIV

Letter

Strong Metal-Support Interactions Achieved by Hydroxideto-Oxide Support Transformation for Preparation of Sinter-Resistant Gold Nanoparticle Catalysts Liang Wang, Jian Zhang, Yihan Zhu, Shaodan Xu, Chengtao Wang, Chaoqun Bian, Xiangju Meng, and Feng-Shou Xiao ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b01947 • Publication Date (Web): 29 Sep 2017 Downloaded from http://pubs.acs.org on September 30, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Strong Metal-Support Interactions Achieved by Hydroxide-to-Oxide Support Transformation for Preparation of Sinter-Resistant Gold Nanoparticle Catalysts Liang Wang,a* Jian Zhang,a Yihan Zhu,b Shaodan Xu,a Chengtao Wang,a Chaoqun Bian,a Xiangju Meng,a and Feng-Shou Xiaoa* a Key

Laboratory of Applied Chemistry of Zhejiang Province, Department of Chemistry, Zhejiang University, Hangzhou 310028, China. b Department of Chemical Engineering, Zhejiang University of Technology, Hang Zhou 310014, China KEYWORDS. Sinter-resistant; Gold nanoparticle; Strong metal-support interaction; Heterogeneous catalysis

ABSTRACT: The strong metal-support interactions (SMSI) are old but crucial for preparation of supported metal nanoparticle catalysts, which generally occur by reduction and oxidation under harsh conditions. Here, we delineate the example of constructing SMSI without reduction and oxidation, where the key is to employ a hydroxide-to-oxide support transformation. The covering of Au nanoparticles by oxides, electronic interaction, and changes in CO adsorption tests of the catalyst are identical to those of the classic SMSI. Owing to the SMSI with oxide barriers on the Au nanoparticles, the supported Au catalysts are sintering-resistant at high temperatures, which benefit long-life reactions, outperforming the conventional supported catalysts.

Metal nanoparticles (NPs) play key roles in catalysis due to the unusual activity and selectivity.1-5 A general route for preparing heterogeneous catalysts is by dispersing the metal NPs on high-surface-area solids to maximize the metal sites for catalysis. However, these catalysts normally suffer from deactivation, which originates from the decrease of catalytically active sites by NP sintering at high temperatures. The regeneration of these catalysts is complex and expensive, particularly for the noble metal catalysts. A scope of methods have been developed to stablize the metal NPs,1,6-19 where a general strategy is constructing physical barriers to separate metal NPs from each other, such as localizing in the channels of nanotube and porous materials,6-10 overcoating with oxides and carbon layers.11-19 Particularly, the strong metal-support interaction (SMSI), which was first described by Tauster and Fung in 1970s and refers to the physical covering/encapsulating of metal NPs by metal oxides,20 has been regarded as a facile route to form physical barriers by high-temperature reduction with hydrogen.20-23 Since the SMSI discovered from reduction,20 the support has been expanded to a series of easily reducible transitionmetal oxides, including TiO2, Ta2O5, Nb2O5, V2O3, and CeO2. The SMSI, which influences the catalytic performances by geometric and/or electronic modification of the metal NPs by oxides, has motivated the development of new catalysts and catalytic mechanism.24,25 Any new insights to SMSI may lead to the development of new catalysts with enhanced performances. The classic route to form SMSI requires hightemperature reduction in H2 to activate the surface of

reducible metal oxide support, which usually causes the aggregation of metal NPs before forming the barriers under the harsh reduction conditions. Additionally, this method only works on limited metal NPs. For example, it failed for Au NPs because of the low work function and surface energy of gold.26,27 Recently, the emergence of high-temperature oxidation in oxygen leads to efficient formation SMSI for supported Au NPs.15,16 It provides a different method for constructing SMSI but only works on hydroxyapatite, phosphate, and ZnO supports. Up to date, it is still lack of efficient method for constructing SMSI in avoidance of oxidative or reduction conditions. Here we report the SMSI between Au NPs and Mg-Al layered double oxides (Au/LDO) occurs without oxidative or reductive atmosphere. The covering of Au NPs by oxides, electronic interaction, and changes in CO adsorption tests in this SMSI are well consistent with those of the classic SMSI. The key to this success is employment of Mg-Al layered double hydroxide as a support for Au NPs as precursors. During calcination in N2, the layered double hydroxide (LDH) is dehydrated into oxide (LDO), where the SMSI occurred. Owing to this SMSI, the Au/LDO is sintering-resistant, exhibiting long catalyst life in CO oxidation and ethanol dehydrogenation at high temperatures, and outperforming the conventional catalysts consisting of Au NPs on LDO by impregnation. This work delineates successful examples of sinter-resistant Au NP catalysts achieved by SMSI. More importantly, the construction of SMSI by hydroxide-to-oxide support transformation on the Au/LDO is universal and can be extended to synthesize other sinter-resistant metal catalysts such as Rh and Pt-based catalysts.

ACS Paragon Plus Environment

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 1. Synthesis and catalytic strategies for achieving sinterresistant Au/LDO catalyst with SMSI.

Our synthesis strategy started from loading the Au NPs on the as-synthesized LDH flake, followed by calcination in N2 to transfer LDH into LDO, and finally Au/LDO samples were obtained with SMSI (Scheme 1). Mg-Al LDH with Mg/Al molar ratio at 2.84 was synthesized by co-precipitation, and then Au NPs were loaded on LDH by impregnation, obtaining Au/LDH with Au loading at 2.2 wt%. The Au/LDH was then calcined under nitrogen flow for 4 h to get layered double oxide supported Au NPs, which are denoted as Au/LDO-x (x stands for calcination temperature). For comparison, the Au NPs were loaded on LDO support by impregnation, followed by calcination, which are denoted as Au/LDO-IP-x.

A

B

C

Page 2 of 6

morphology is still remained for the LDO. TEM characterization gives direct observation of the Au NPs and the LDH/LDO support (Figures S3 and S4). The EDS maps demonstrate that Mg and Al species are uniformly distributed after the calcination. Figures 1A-C show high-angle annular dark-field (HAADF) scanning transmission electron microscopy (STEM) images of Au/LDO calcined at 400, 600 and 700 ° C, which display highly dispersed Au NPs on the LDO support. Significantly, the Au/LDO-400 sample gives Au NP diameter distribution at 0.5-5.0 nm with a mean size at 2.4 nm, which is very similar to those of the as-synthesized Au/LDH with Au NP diameter distribution at 0.5-5.0 nm and a mean size at 2.1 nm (Figure 2A). Thermal treatment at 600 ° C, which is harsh condition for the survival of small-size Au NPs, still leads a rich amount of Au NPs smaller than 2.0 nm as observed from the high resolution STEM images, giving an overall Au NP size distribution at 1.0-5.5 nm with a mean size at 3.1 nm. Even after calcination at 700 ° C, the Au/LDO-700 also gives Au NPs at 2.0-6.5 nm with mean size at 3.9 nm. These results confirm that the LDO effectively stabilize the Au NPs of Au/LDO for enhancing the stabilities at high temperatures. In contrast, the Au NPs of conventional Au/LDO-IP sample clearly sintered under the same calcination treatment (Figures S5 and S6), giving aggregated Au NPs with mean sizes at 6.3 and >9 nm after calcining at 600 and 700 ° C, respectively (Figure 2B). A

D

E

Au/LDH Au/LDO-400 Au/LDO-600 Au/LDO-700

F

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Particle diameters (nm)

G

H

I

Figure 1. Electronic microscopic characterization of Au/LDO samples. (A-C) STEM and (D-I) TEM images of (A, D, G) Au/LDO-400, (B, E, H) Au/LDO-600, and (C, F, I) Au/LDO-700. Inset: Enlarge view.

The Mg-Al LDH and Au/LDH show typical layered structure and good crystallinity with lattice parameters at a=0.305 and c=2.247 nm, as confirmed by the XRD patterns (Figure S1). After calcination at 400-700 ° C in N2, the Au/LDO samples lost the XRD peaks of layered structure due to the removal of anions in LDH, exhibiting XRD patterns of Au/LDO with peaks at 37.8, 43.8 and 63.4° , which are assigned to the (111) incidence of metallic Au, (400) and (440) incidences of Mg-Al spinel, respectively. Notably, the peak of Au(111) is weak and wide, suggesting the high dispersion and small sizes of Au NPs according to Scherrer equation. As observed in low-resolution TEM image (Figure S2), the plate-like

as synthesized Au/LDO-IP Au/LDO-IP-400 Au/LDO-IP-600 Au/LDO-IP-700

B

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Particle diameters (nm) Figure 2. Au NP diameter distribution of Au samples under different treatments.

Figures 1D-F are TEM images of Au/LDO samples, finding that the Au NPs are embedded with the LDO support. Furthermore, the oxide barriers partially covering the Au NPs are clearly observed in the typical highresolution TEM (HR-TEM) images. Furthermore, we acquired a STEM tilt to visualize the 3D observation (Figure S7), where the oxide barriers are always observable before and after tilt for a certain degree. Energy dispersive X-ray spectroscopy (EDS) analysis demonstrates that the Al species is enriched in the oxide barrier compared with the oxide support (Figure S8). The oxide barriers facilitate the stabilization of Au NPs on the LDO

ACS Paragon Plus Environment

Au/LDH

B 12

Relative Absorbance

2106

Au/LDO-300 Au/LDO-400

2109

Au/LDO-500 Au/LDO-600 Au/LDO-700

2250 2200 2150 2100 2050 2000 -1

Wavenumber (cm )

10 8 6 4 2 0

0 200 400 600 o

T ( C)

Figure 3. (A) In situ FTIR spectra of CO adsorbed on various Au samples; (B) Dependence of the intensity of 2101-2109 cm-1 band on calcination temperature.

The cover of metal NPs by oxides is a typical feature of classic SMSI, which generally results in the suppression of the adsorption of CO on metal NPs because of the physical coverage of adsorption sites by the oxide barriers.16 Figure 3A gives the CO-adsorption FTIR spectra of Au-LDH and Au/LDO-300-700 samples. The spectra of Au/LDH exhibit strong bands at 2101 cm-1, assigning to CO adsorbed on metallic Au NPs. Calcination of this sample in N2 leads to a significant reduction of CO adsorbed on Au NPs (Figure 3B). Considering the slight size change of these Au NPs during the calcination at 100-400 °C (mean sizes of 2.2 nm at 100 °C and 2.4 nm at 400 °C), the decreased band intensity is reasonably assigned to the partial cover of Au NPs by the oxides, which is further confirmed by the Au dispersion test (Table S1), in good agreement with those in classic SMSI. 15,20 While further intensity decreased of IR band at 400700 °C calcination might be caused by the combined effect of SMSI and decreased Au NP surface area, because the Au NP diameters were increased in this process (mean sizes of 2.4 nm at 400 °C and 3.9 nm at 700 °C). Additionally, the band at 2101 cm-1 over Au/LDH is shifted after calcination, giving 2106 cm-1 over Au/LDO300 and Au/LDO-400, 2109 cm-1 over Au/LDO-500, Au/LDO-600, and Au/LDO-700. The blue shift of the CO adsorption suggests the electron transfer from Au NPs to the LDO support,28 which is further confirmed by the Au4f XPS spectra (Figure S9). In contrast, the Au/LDO-IP gives very slight decrease in CO-adsorption band intensity before and after calcination at 400 ° C (Figure S10), because of the absence of oxide barriers on

A

o

220 C

o

190 220 C

100 80 60

as-synthesized Au/LDO-IP Au/LDO-400 Au/LDO-600 Au/LDO-700

40 20

B

o

190 C

CO conversion (%)

2101

A

the NP surface. Additionally, band shift was nearly undetectable in the FTIR spectra (Figure S10), due to the weak electron transfer between Au NPs and the LDO support. These results confirm the lack of SMSI on Au/LDO-IP. The combined results of TEM and CO adsorption tests indicate that the SMSI has been constructed between Au and LDO support, leading to the formation of sinterresistant Au/LDO samples. The features of SMSI on Au/LDO, such as covering of Au NPs by oxides, electronic interaction, changes in CO adsorption tests of the catalyst, as well as the reversible electronic interactions under redox recycles (Figure S11), are well agreed with those of the classic description of SMSI on Pt-group NPs on metal oxides and Au NPs on hydroxyapatite.16,20 It is also emphasized that the classic SMSI requires reductive or oxidative conditions, and N2 treatment condition is regarded as ineffective to construct SMSI yet. However, the SMSI on the Au/LDO has been successfully realized by calcination in N2 via hydroxide-to-oxide transformation, which is remarkably different from the classic methods. Also, the sinter-resistant catalysts are still obtained when Au/LDH was calcined in air instead of N 2 (Figure S12), demonstrating that hydroxide-to-oxide transformation is crucial for SMSI, but calcination atmosphere is not important. Moreover, the oxide barriers on the Au NPs of Au/LDO-400 are observed even after reduction by hydrogen (Figure S13), which is different from the classic SMSI over reducible VIII group metal oxides with barriers disappearing and appearing reversibly under redox conditions. This phenomenon might be assigned to the relatively inert feature of Mg-Al oxides compared with the VIII group metal oxides. These results, again, emphasize the significant difference from the classic methods for constructing SMSI.

CH3CHO sel. (%)

support, achieving Au NPs with enhanced stabilities at high temperatures at high temperatures. In contrast, it is difficult to observe the barriers around Au NPs in the HR-TEM image of Au/LDO-IP. Considering that the Au/LDO and Au/LDO-IP samples have the same composition, their differences should be directly resulted from the distinguishable preparation methods. For the preparation of Au/LDO from Au/LDH precursor, it is most likely that the migration of oxide onto the surface of Au NPs occurs during the hydroxide-to-oxide process in nitrogen, due to the reconstruction of the surface metal and oxygen species from hydroxide to oxide form. In contrast, the direct calcination of the Au/LDO-IP at high temperature in nitrogen fails to construct the barriers owing to the lack of hydroxide-to-oxide transformation.

Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

0

5 10 15 20 25 30 35 40 45 50 55 60 Time on stream (h)

100

EtOH Conv.(%)

Page 3 of 6

90

Au/LDO-600 Au/LDO-IP-400 LDO

80

40 30 20

Au/LDO-600 Au/LDO-IP-400 LDO

10 0

0

4 6 8 10 12 14 Time on stream (h) Figure 4. Catalytic data characterizing various Au samples in CO oxidation and EtOH dehydrogenation. (A) Dependences of CO conversion on time in CO oxidation. Reaction conditions: 0.1 g of cata-

ACS Paragon Plus Environment

2

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

lyst, feed gas of 1%CO/19%O2/80%He, GHSV at 90000 ml/gAu/h; (B) Dependences of EtOH conversion and CH3CHO selectivity on time. Reaction conditions: 0.1 g of catalyst, ethanol feed at 2 mL/h, nitrogen at 5 mL/min, 210 °C. The black arrows highlighted the calcination of the catalyst in oxygen at 330 ° C for 2 h in CO oxidation and at 400 ° C for 2 h in ethanol dehydrogenation.

The new methodology to construct the SMSI is only a part of the story, it is expected that the Au/LDO with SMSI can provide stable catalytic performances at high temperatures. Figure 4A shows the data characterizing the performances of Au catalysts in CO oxidation as probing reaction.29,30 In the beginning of the reaction at 190 ° C, the CO conversion of Au/LDO-IP (90%) is higher than those of the Au/LDO samples (65-86%), and the Au/LDO-700 gives the lowest CO conversion at 65%, which is reasonably assigned to the decrease of active Au sites by the oxide barrier coverage. Increasing the reaction temperature to 220 ° C causes an enhancement of CO conversion, giving full conversion of CO over Au/LDO-IP, Au/LDO-400, and Au/LDO-600, and 96% CO conversion over Au/LDO-700. Very interestingly, after reaction at 220 ° C for another 40 h, the decrease of CO conversion over Au/LDO catalysts is within 5%, suggesting their excellent catalyst life. After reaction for 62 h, the Au/LDO-400 catalyst still gives very similar CO conversion to its initial activity (Figure S14-a and b). In contrast, the CO conversion of Au/LDO-IP is sharply decreased to 70% at reaction for 16 h (Figure S14-c and d). TEM measurements demonstrate that the used catalysts, such as Au/LDO-400, still remain the Au NPs with original size range of 0.5-5.0 nm (Figure S15A), whereas the used Au/LDO-IP has obviously aggregated Au NPs with size of 10 nm (Figure S15C). Significantly, calcination of the catalyst in oxygen can fully regenerate the CO conversion to the values of assynthesized catalysts, due to the removal of carbonaceous deposits on the active sites, demonstrating the superior stability of Au/LDO catalysts. On the contrary, the equivalent calcination of the used Au/LDO-IP catalyst caused a further reduction of the activity because of further aggregation of the Au NPs (Figure S15E). Figure 4B shows the dependences of ethanol conversion and acetaldehyde selectivity on time over various catalysts. The LDO support is active for the ethanol dehydrogenation, giving ethanol conversion at 13.1% in the beginning of the reaction. Functionalizing LDO with Au NPs could effectively enhance the ethanol conversion. The Au/LDO-IP-400 gives initial ethanol conversion at 33.0%, while the Au/LDO-600 exhibits the value at as high as 42.2%. Since Au/LDO-IP-400 and Au/LDO-600 have comparable Au NPs diameters and the same LDO support, the distinguishable catalytic performances of Au/LDO-IP-400 and Au/LDO-600 are reasonably attributed to the SMSI on Au/LDO-600, where the oxide barriers covered on Au NP surface might efficiently cooperate with Au to cause enhanced performance in dehydrogenation. Furthermore, when the Au/LDO-600 catalyst continues for 900 min in a flow reactor (Figure 4B), the ethanol conversion decreases periodically because of coke formation (confirmed by the change in catalyst color and thermogravimetry analysis, Figure S16). After calcination in oxygen, the Au/LDO-600 activity was fully regenerated. Very interestingly, even if reaction at 210 ° C for 15 h, the Au/LDO-600 still exhibits EtOH

Page 4 of 6

conversion comparable to that of the as-synthesized catalyst. These results demonstrate the sinter resistance of the Au/LDO-600 catalyst. In contrast, the activity of Au/LDO-IP-400 was greatly lost in a short time (ethanol conversion at 17.2% in 120 min) because of the combined effect of coke formation and Au NP aggregation. Calcination in oxygen failed to regenerate the activity of Au/LDO-IP-400 due to the poor stability of Au NPs. We also tested the catalysts in the non-oxidative dehydrogenation of alcohols in liquid phase (Figure S17), where the Au leaching always occurs on conventionally supported catalysts, resulting in deactivation during the catalytic recycles. In the dehydrogenation of phenethyl and benzyl alcohols, both Au/LDO-IP-400 and Au/LDO600 exhibited high conversion and selectivity in the first reaction run. However, in the recycling tests, the Au/LDO-IP-400 deactivated remarkably in the fourth and second runs in dehydrogenation of phenethyl and benzyl alcohol, respectively. The ICP-OES analysis of the used catalysts confirms that over 13% of the Au species on the Au/LDO-IP-400 are leached in the recycles. In comparison, the conversions of both phenethyl and benzyl alcohol in the recyclable tests are almost unchanged within error over the Au/LDO-600 catalyst. ICP analysis of the used Au/LDO-600 gives the Au leaching smaller than 0.5%. These results confirm Au/LDO-600 shows much better stability than the Au/LDO-IP-400 catalyst. It is worth noting that the SMSI by hydroxide-to-oxide support transformation is not only limited for supported Au NP catalysts, but also can be extended to synthesize supported Pt and Rh catalysts. Figures S18 and S19 show the TEM images of Rh/LDO and Pt/LDO synthesized from hydroxide-to-oxide support transformation. After calcination at 600 ° C for 4 h, the calcined Pt/LDO-600 and Rh/LDO-600 maintain the Pt and Rh NP sizes well. HR-TEM characterizations provides evidence for oxide barriers partially covering on the metal NPs, indicating the universality of hydroxide-to-oxide support transformation for SMSI. In contrast, the conventionally supported Pt NPs by impregnation (Pt/LDO-IP) aggregate into larger ones during the same calcination due to the poor stability (Figure S20). Additionally, this hydroxideto-oxide support transformation can also be used to synthesize Au NPs with oxide barriers on CeO2 support (Figure S21), confirming the generality of this route. The origination of the SMSI in these cases is still unclear yet. In the SMSI in literatures, the treatment conditions (e.g. high-temperature reduction20,23,31,32 or oxidation15,16) all benefit the activation of the solid surface, thus leading to SMSI for minimization of the surface energy (Table S2). In our cases, we propose that the hydroxide-to-oxide transformation activates/reconstructs the support to cause the SMSI for minimizing the surface energy. However, a detailed mechanism still needs further investigation in the future. In summary, we have demonstrated a new route for constructing SMSI between oxides and Au NPs. The covering of Au NPs by oxides, electronic interaction, and changes in CO adsorption tests of the catalyst are similar to those of the classic SMSI. This SMSI route results in a significant enhancement in catalyst stability, providing sintering-resistant Au catalysts in CO oxidation and ethanol dehydrogenation at high temperatures. More importantly, this SMSI is universal and can be extended to

ACS Paragon Plus Environment

Page 5 of 6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

synthesize other sintering-resistant metal (Pt and Rh) catalysts.

ASSOCIATED CONTENT Supporting Information. Experiment details, XPS, XRD, and more TEM data. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author Email: [email protected] (L.W.) Email: [email protected] (F.-S.X.)

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT This work is supported by National Natural Science Foundation of China (91634201, 21403192, and 91645105) and National Key Research and Development Program of China (2017YFB0702803). We thank Xue Dong in Texas Tech University for helpful discussions.

(21) Braunschweig, E. J.; Logan, A. D.; Datye, A. K.; Smith, D. J. J. Catal. 1989, 118, 227-237. (22) Corma, A.; Serna, P.; Concepcion, P.; Calcino, J. J. J. Am. Chem. Soc. 2008, 130, 8748-8753. (23) Matsubu, J. C.; Zhang, S.; DeRita, L.; Marinkovic, N. S.; Chen, J. G.; Graham, G. W.; Pan, X.; Christopher, P. Nat. Chem. 2017, 9, 120-127. (24) Sun, Y. N.; Giordano, L.; Goniakowski, J.; Lewandowski, M.; Qin, Z. H.; Noguera, C.; Shaikhutdinov, S.; Pacchioni, G.; Freund, H. J. Angew. Chem., Int. Ed. 2010, 49, 4418-4421. (25) Sonström, P.; Arndt, D.; Wang, X.; Zielasek, V.; Bäumer, M. Angew. Chem., Int. Ed. 2011, 50, 3888-3891. (26) Tatarchuk, B. J.; Dumesic, J. A. J. Catal. 1981, 70, 335346. (27) Fu, Q.; Wagner, T.; Olliges, S.; Carstanjen, H. D. J. Phys. Chem. B 2005, 109, 944-951. (28) Mihaylov, M.; Knözinger, H.; Hadjiivanov, K.; Gates, B. C. Chem. Ing. Tech. 2007, 79, 795-806. (29) Li, M.-H.; Chen, Y.-W.; Liu, X.; Kuo, J.; Chu, M.; Mou, C.-Y. ACS Catal. 2016, 6, 115-122. (30) Lu, J.; Aydin, C.; Browning, N. D.; Gates, B. C. Angew. Chem., Int. Ed. 2012, 51, 5842-5846. (31) Tauster, S. J. Acc. Chem. Res. 1987, 20, 389-394. (32) Fu, Q.; Wagner, T. Surf. Sci. Rep. 2007, 62, 431-498.

REFERENCES (1) Strayer, M. E.; Binz, J. M.; Tanase, M.; Shahri, S. M. K.; Sharma, R.; Rioux, R. M.; Mallouk, T. E. J. Am. Chem. Soc. 2014, 136, 5687-5696. (2) Fox, E. B.; Velu, S.; Engelhard, M. H.; Chin, Y.-H.; Miller, J. T.; Kropf, J.; Song, C. S. J. Catal. 2008, 260, 358-370. (3) Zhu, Y.; Zhang, B.; Liu, X.; Wang, D.-W.; Su, D. S. Angew. Chem., Int. Ed. 2014, 53, 10673-10677. (4) Yang, S.; Cao, C.; Sun, Y.; Huang, P.; Wei, F.; Song, W. Angew. Chem., Int. Ed. 2015, 54, 2661-2664. (5) Wan, X.; Zhou, C.; Chen, J.; Deng, W.; Zhang, Q.; Yang, Y.; Wang, Y. ACS Catal. 2014, 4, 2175-2185. (6) Liu, L.; Dí az, U.; Arenal, R.; Agostini, G.; Concepción, P.; Corma, A. Nat. Mater. 2016, 16, 132-138. (7) Goel, S.; Zones, S. I.; Iglesia, E. J. Am. Chem. Soc. 2014, 136, 15280-15290. (8) Wang, N.; Sun, Q.; Bai, R.; Li, X.; Guo, G.; Yu, J. J. Am. Chem. Soc. 2016, 138, 7484-7487. (9) Yang, Q.; Xu, Q.; Yu, S.; Jiang, H.-L. Angew. Chem., Int. Ed. 2016, 55, 3685-3689. (10) Pan, X.; Fan, Z.; Chen, W.; Ding, Y.; Luo, H.; Bao, X. Nat. Mater. 2007, 6, 507-511. (11) Lu, J.; Fu, B.; Kung, M. C.; Xiao, G.; Elam, J. W.; Kung, H. H.; Stair, P. C. Science 2012, 335, 1205-1208. (12) Arnal, P. M.; Comotti, M.; Schüth, F. Angew. Chem., Int. Ed. 2006, 45, 8224-8227. (13) Joo, S. H.; Park, J. Y.; Tsung, C.-K.; Yamada, Y.; Yang, P.; Somorjai, G. A. Nat. Mater. 2009, 8, 126-131. (14) Seipenbusch, M.; Binder, A. J. Phys. Chem. C 2009, 113, 20606-20610. (15) Liu, X.; Liu, M.-H.; Luo, Y.-C.; Mou, C.-Y.; Lin, S. D.; Cheng, H.; Chen, J.-M.; Lee, J.-F.; Lin, T.-S. J. Am. Chem. Soc. 2012, 134, 10251-10258. (16) Tang, H.; Wei, J.; Liu, F.; Qiao, B.; Pan, X.; Li, l.; Liu, J.; Wang, J.; Zhang, T. J. Am. Chem. Soc. 2016, 138, 56-59. (17) Ta, N.; Liu, J.; Chenna, S.; Crozier, P. A.; Li, Y.; Chen, A.; Shen, W. J. Am. Chem. Soc. 2012, 134, 20585-20588. (18) Zhan, W.; He, Q.; Liu, X.; Guo, Y.; Wang, Y.; Wang, L.; Guo, Y.; Borisevich, A. Y.; Zhang, J.; Lu, G.; Dai, S. J. Am. Chem. Soc. 2016, 138, 16130-16139. (19) Tang, H.; Liu, F.; Wei, J.; Qiao, B.; Zhao, K.; Su, Y.; Jin, C.; Li, L.; Liu, J.; Wang, J.; Zhang, T. Angew. Chem., Int. Ed. 2016, 55, 10606-10611. (20) Tauster, S. J.; Fung, S. C. Garten, R. L. J. Am. Chem. Soc. 1978, 100, 170-175.

ACS Paragon Plus Environment

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

Page 6 of 6