Chinese Chemical Letters  2018, Vol. 29 Issue (12): 1883-1887   PDF    
Site-and surface species-dependent propylene oxidation with molecular oxygen on gold surface
Guanghui Sun, Yuekang Jin, Zhengming Wang, Hong Xu, Peng Chai, Weixin Huang*     
Hefei National Laboratory for Physical Sciences at the Microscale, CAS Key Laboratory of Materials for Energy Conversion, and Department of Chemical Physics, University of Science and Technology of China, Hefei 230026, China
Abstract: Fundamental understandings of Au-catalyzed oxidation reactions with molecular oxygen are of great importance. Herein we report a successful preparation of molecularly-adsorbed O2 species on a stepped Au(997) single crystal model surface via a near-ambient pressure and low-temperature oxygen exposure and their oxidation reactivity with propylene via temperature programmed desorption spectra and polarization-modulated reflection-absorption infrared spectra. O2 molecularly adsorbs at the (111) step sites of Au(997) surface while C3H6 adsorbs at both (111) step sites and (111) terrace sites. C3H6(a) adsorbed at the (111) step sites is more stable than at the (111) terrace sites and its oxidation reactions are dominant; meanwhile, O2(a) is much more reactive than O(a) and prefers combustion reaction of C3H6(a). C3H6(a) adsorbed at the (111) step sites undergoes combustion reactions with O2(a) to produce CO2 and CO at low temperatures and partial oxidation reactions with O(a) to produce acrolein at high temperatures while C3H6(a) adsorbed at the (111) terrace sites undergoes combustion and partial oxidation reactions with O2(a) to respectively produce CO2/CO and acrolein at low temperatures. These results reveal site-and surface species-dependent reaction behaviors of propylene oxidation with molecular O2 on Au surfaces and provide a complete fundamental understanding of oxidation reactions with molecular O2 on Au surface.
Keywords: Au(997) single crystal     Combustion     Partial oxidation     Acrolein     Reaction mechanism    

Ethylene epoxide and propylene epoxide are important intermediates in chemical industry, and epoxidation of ethylene and propylene with molecular oxygen is the most desirable route to produce corresponding olefin epoxides and have been extensively studied.Silver-catalyzed epoxidation of ethylene with O2 is a successful industrial epoxidation reaction [1], but silver catalysts exhibit a poor catalytic performance in catalyzing propylene epoxidation with O2 [2, 3]. The methyl hydrogen of propylene was found to be facilely abstracted by oxygen species on Ag surfaces, resulting in the combustion reactions [4-9]. Recently, gold-based catalysts have demonstrated very high catalytic selectivity in catalyzing propylene epoxidation with O2 in the presence of H2 [10-19].

Excellent catalytic performance of gold catalysts in oxidation reactions has inspired extensive fundamental studies of reaction mechanisms and active sites on Au single crystal surfaces under ultrahigh vacuum (UHV) conditions [20-26]. Due to the chemical inertness of bulk Au, O2 does not molecularly or dissociatively chemisorb on Au single crystal surfaces under UHV conditions. Therefore, use of an atomic oxygen source [27] or ozone [28], electron or photon irradiation of Au surfaces with physisorbed O2 at 28 K [29], electron bombardment of chemisorbed NO2 [30] and the decomposition of amorphous N2O4 multilayers [31] were used to prepare atomic oxygen species including oxygen adatom and surface oxide on Au single crystal surfaces for the fundamental studies. Propylene partial oxidation with O(a) on various Au single crystal surfaces was studied [32, 33], and acrolein was observed as the partial oxidation product via the allyloxy intermediate. However, it is noteworthy that fundamental studies of Au single crystal surfaces without the involvement of O2 do not provide a complete understanding of Au-catalyzed oxidation reactions.

In this letter, we prepare molecularly-adsorbed O2(a) species on a stepped Au(997) surface by near-ambient pressure exposures at low temperatures and study propylene oxidation with O2(a) using thermal desorption spectroscopy (TDS) and polarization-modulated reflection-absorption infrared spectroscopy (PM-RAIRS). Oxidation pathways of propylene, including combustion to CO2 and H2O and partial oxidation to acrolein, were observed to sensitively depend on the adsorption sites and surface species. Asfar as we know, our result represent the first report on fundamental studies of oxidation reactions on Au single crystal surfaces with O2(a).

Experimental details are described in the Supporting information. Fig. 1 displays a schematic illustration of the Au(997) stepped surface and the corresponding LEED pattern. The LEED pattern exhibits splitting spots forming a hexagonal symmetry, similar to those of Pt(997) and Cu(997) stepped surfaces [34, 35]. The Au(997) stepped surface is composed of close-packed (111) terraces and mono-atomic (111) steps on which the coordination numbers of Au atoms are 9 and 7, respectively. Thus, Au(997) surface is an ideal model surface for comparative studies of reactivity of Au atoms with the same coordination geometry but different coordination numbers. Meanwhile, The {111} facet is the most stable facet of fcc Au metal and abundant on Au nanoparticles. Au atoms on the monatomic (111) steps of Au(997) with a coordination number of 7 can mimic coordination unsaturated Au atoms on the {111} facet of Au nanoparticles.

Download:
Fig. 1. Schematic illustration (top) and LEED pattern (bottom) of a stepped Au(997) single crystal surface.

Fig. 2A shows C3H6 TDS spectra (represented by m/z = 41) following various C3H6 exposures on Au(997) surface at 110 K. C3H6 is the only desorption product. A C3H6 desorption peak appears at ~260 K at the smallest exposure of 0.02 L C3H6. With the C3H6 exposure increasing, this feature grows, broadens and shifts to low desorption temperatures. It saturates with the desorption maximum at 212 K at an exposure of 2 L C3H6. A new C3H6 desorption peak emerges at ~136 K at an exposure of 3 L C3H6, and grows and shifts to ~133 K at an exposure of 5 L C3H6. Based on previous results of propylene adsorption on Au(111) [32, 33], the low-temperature C3H6 desorption feature at ~135 K can be assigned to C3H6(a) adsorbed at the (111) terrace sites of Au (997), and thus the high-temperature C3H6 desorption feature above 200 K can be assigned to C3H6(a) adsorbed at the (111) step sites of Au(997). This is consistent with our previous results that the (111) step sites with a lower coordination number exhibits a stronger interaction with adsorbed molecules than the (111) terrace sites [36-38]. The shift of C3H6 desorption peaks to low temperatures with the growth in desorption peak areas indicates the presence of repulsive interaction within adsorbed C3H6(a) molecules.

Download:
Fig. 2. (A) C3H6 TDS spectra following various exposures of C3H6 on Au(997) at 110 K; (B) PM-RAIRS spectra of Au (997) surface exposed to various amounts of C3H6 at 110 K; (C) PM-RAIRS spectra of Au(997) surface exposed to 5.0 L C3H6 at 110 K followed by annealing at elevated temperatures.

Fig. 2B shows PM-RAIRS spectra following various C3H6 exposures on Au(997) surface at 110 K. No signals can be identified for C3H6 exposures below 0.5 L likely due to the low C3H6(a) coverages beyond the detection limit of our PM-RAIRS spectrometer. A weak vibrational band appears at 1427 cm-1 after an exposure of 0.5 L C3H6. This band grows with an increase of C3H6 exposure to 1 L, and new vibrational bands emerge at 981 and 1632 cm-1. With a further increase of C3H6 exposure to 3 L, the band at 1427 cm-1 saturates, corresponding to the saturating desorption peak of C3H6 adsorbed on the step sites of Au(997) surface in the TDS spectra; meanwhile, both bands at 981 and 1632 cm-1 grow and a new band appears at 1450 cm-1. The bands at 981, 1450 and 1632 cm-1 increase with a further increase of C3H6 exposure to 5 L, corresponding to the growth of desorption peak of C3H6 adsorbed on the terrace sites of Au(997) surface in the TDS spectra.

Fig. 2C shows PM-RAIRS spectra of 5 L C3H6 exposures on Au (997) surface at 110 K followed by annealing elevated temperatures. Upon an annealing at 150 K, the band at 1450 cm-1 varnishes and those at 981 and 1632 cm-1 weaken, corresponding to the desorption of C3H6 adsorbed on the terrace sites of Au(997) surface in the TDS spectra. Upon a further annealing at 230 K, the bands at 1427 and 1632 cm-1 varnish and no bands could be observed, corresponding to the desorption of C3H6 adsorbed on the step sites of Au(997) surface in the TDS spectra. Therefore, comparing the corresponding PM-RAIRS and TDS results, the vibrational bands at 981/1450/1632 cm-1 and 981/1432/1632 cm-1 arise from C3H6(a) adsorbed on the (111) terrace sites of Au(997) and C3H6(a) adsorbed on the (111) step sites of Au(997), respectively. On the basis of vibrational spectra of gas-phase C3H6 and C3H6(a) adsorbed on Au(111) surface [33, 39], the vibrational bands at 981, 1432/1450 and 1632 cm-1 can be assigned to CH2 wag, CH3 asymmetric deformation and C=C stretch vibrations of C3H6(a) adsorbed on Au(997) surface.

Fig. 3 shows TPRS spectra after Au(997) surface was exposed to various amounts of C3H6 and then to 2.0 mbar O2 at 110 K. Following an exposure of clean Au(997) surface to 2.0 mbar O2 at 110 K, a broad desorption trace of O2 centering at ~135 K appears and can be attributed to molecularly-adsorbed O2(a) at the (111) step sites of Au(997). O2 barely chemisorbs on Au(111) surface [29] but it was reported that O2 could molecularly chemisorb on stepped Au(211) and Au(110)-(1 × 2) surfaces at low temperatures and elevated pressures [40]. Meanwhile, desorption traces of CO, CO2 and H2O were also observed due to inevitable background adsorption of CO and CO2 at the (111) step sites and H2O and likley CO oxidation with O2(a) under the employed O2 exposure condition [37, 38]. Comparing with individual adsorptions, both O2 and C3H6 desorption traces weaken following 2.0 mbar O2 exposures on C3H6(a)-covered Au(997) surfaces at 110 K, and the O2 desorption peaks also shifts slightly to low temperatures likely due to the repulsive interaction within co-adsorbed C3H6(a) and O2(a). Meanwhile, the desorption traces of H2O, CO2 and CO (the 135 K feature) increase, and new desorption traces at m/z = 56 corresponding to acrolein (CH2CHCHO) appear but no desorption trace occurs at m/z = 58 corresponding to propylene epoxide and acetone. These results clearly demonstrate that surface reactions occur between co-adsorbded C3H6(a) and O2(a), including the combustion of C3H6(a) to CO2 and H2O and the partial oxidation of C3H6(a) to acrolein.

Download:
Fig. 3. TPRS spectra of (A) O2, (B) C3H6, (C) acrolein, (D) H2O, (E) CO and (F) CO2 from Au(997) surfaces exposed to various amounts of C3H6 and 2.0 mbar O2 at 110 K. The acrolein desorption trace from Au(997) surface exposed to 5 L C3H6 and 0.5 mbar O2 at 110 K is also included.

The desorption traces of H2O, CO2, CO and acrolein vary with the coverage of pre-adsorbed C3H6(a). Following a 1 L C3H6 + 2.0 mbar O2 co-adsorption that only forms C3H6(a) at the (111) steps, the desorption traces of H2O, CO2 and CO (the 135 K feature) all increase while their desorption temperatures are similar to those of molecularly-adsorbed H2O(a), CO2(a) and CO(a), demonstrating desorption-controlled production of H2O, CO2 and CO. This suggests the occurrence of very facile combustion reactions between C3H6(a) and O2(a) both co-adsorbed at the (111) steps producing H2O, CO2 and CO. Meanwhile, a strong acrolein desorption feature appears at ~287 K and is accompanied by weak desorption traces of C3H6 and CO. No molecularly-adsorbed O2(a) exists on Au(997) at such a high temperature, thus O2(a) dissociates into O(a) at the (111) step sites during the heating process, and the partial oxidation between C3H6(a) and O(a) co-adsorbed at the (111) step sites occurs to produce acrolein. They also react to produce minor CO. The simultaneous very weak C3H6 desorption likely comes from O(a)-stabilized C3H6(a) at the (111) step sites. Previous results show that adsorbed acrolein molecules on clean and O(a)-precovered Au surfaces both desorb below 200 K [33]. Thus the acrolein desorption peak at ~287 K is reaction-controlled. Similar acrolein productions were previously observed for propylene reaction with pre-adsorbed O(a) on Au(111) and Au (100) surfaces and accompanied by strong CO2 formation [32, 33]. However, in our case, the reactions between C3H6(a) with O(a) resulting from O2(a) dissociation on Au(997) surface predomi-nantly produce acrolein with very minor CO. This is likely due to the low coverage of O(a) on Au(997) resulting from O2(a) dissociation during the heating process, suppressing the combustion reactions.

Following a 3 L C3H6 + 2.0 mbar O2 co-adsorption that forms saturating C3H6(a) at the (111) step sites and C3H6(a) at the (111) terrace sites, the desorption traces of H2O, CO2 and acrolein slightly increase while that of CO (the 135 K feature) decreases. Therefore, the decrease of CO desorption feature at ~135 K is likely due to the enhanced site-blocking effect of saturating C3H6(a) at the (111) steps on the background CO adsorption. With a further increase of the exposure of pre-adsorbed C3H6 to 5 L and subsequent C3H6(a) at the (111) terrace sites, the acrolein desorption feature at ~287 K does not change. This is reasonable since C3H6(a) at the (111) step sites saturates at an exposure of 3 L C3H6. However, the H2O, CO2 and CO desorption traces slightly increase; moreover, a new minor acrolein desorption peak emerges at ~135 K. These observation indicate ocuurence of minor surface oxidation reactions other than those of C3H6(a) at the (111) step sites following a 5 L C3H6 + 2.0 mbar O2 co-adsorption, which can be reasonably attributed to the reactions between adjacent C3H6(a) at the (111) terrace sites and O2(a) at the (111) step sites. Following a 5 L C3H6 + 0.5 mbar O2 co-adsorption, the minor acrolein desorption peak at ~135 K also appears (Fig. 3C), supporting its production via the partial oxidation of C3H6(a) at the (111) terrace sites; meanwhile, the acrolein desorption at ~287 K via the partial oxidation between C3H6(a) and O(a) co-adsorbed at the (111) step sites weakens relative to that following a 5 L C3H6 + 0.5 mbar O2 co-adsorption due to the decreased O(a) coverage. The formation of acrolein at ~135 K must follow a different mechanism from that at ~287 K. Interestingly, this acrolein production temperature is even lower than the desorption temperature of acrolein adsorbed on Au surfaces. This might imply that its formation should not involve acrolein adsorbed on the surface. The amounts of CO2, CO and acrolein produced by oxidations of C3H6(a) adsorbed at the (111) step sites are much larger than by oxidations of C3H6(a) adsorbed at the (111) terrace sites. This suggests that oxidations of C3H6(a) adsorbed at the (111) with O2(a) are the major surface reactions. The Au(997) surface was characterized by XPS and CO adsorption after the TPRS experiments, and the results suggested that the surface should be stable.

Fig. 4 shows PM-IRRAS spectra of a 5 L C3H6 + 2.0 mbar O2 coadsorption followed by annealing at elevated temperatures. Four vibrational bands at 981, 1427, 1632 and 1648 cm-1 appear at 110 K, and the bands at 981, 1427 and 1632 cm-1 weaken upon annealing at 150 K, and only the band at 1427 cm-1 is visible upon annealing at 180 K, and no bands could be identified upon annealing at 230 K. Comparing that of a 5 L C3H6 adsorption at 110 K (Fig. 2C), the CH3 asymmetric deformation band at 1450 cm-1 disappears while a new band appears at 1648 cm-1. Since the formation of acrolein via the partial oxidation of C3H6(a) at the (111) step sites is reaction-controlled to occur at ~287 K and the formation of acrolein via the partial oxidation of C3H6(a) at the (111) terrace sites desorbs at ~ 135 K, thus the new band at 1648 cm-1 can not be related to adsorbed acrolein species. We assign the band at 1648 cm-1 to the C=C stretch vibrations of C3H6(a) interacting with co-adsorbed oxygen species. Thus the interaction with co-adsorbed oxygen species enhances the stability of C3H6(a).

Download:
Fig. 4. PM-RAIRS spectra of Au(997) surface exposed to 5.0 L C3H6 and 2.0 mbar O2 at 110 K followed by annealing at elevated temperatures.

The above results clearly demonstrate propylene oxidation reactions with molecular O2 on Au(997) single crystal model surface and reveal site- and surface species-dependent reaction behaviors. As schematically illustrated in Fig. 5, O2 adsorbs at the (111) step sites of Au(997) surface at 110 K while C3H6 adsorbs at both (111) step sites and (111) terrace sites. Upon heating, C3H6(a) adsorbed at the (111) step sites undergoes combustion reactions with O2(a) to produce CO2 and CO at 135 K but partial oxidation reactions with O(a) to mainly produce acrolein at 287 K, while C3H6(a) adsorbed at the (111) terrace sites undergoes both combustion and partial oxidation reactions with O2(a) at 135 K to respectively produce CO2/CO and acrolein. C3H6(a) adsorbed at the (111) step sites is more stable than at the (111) terrace sites and its oxidation reactions are dominant; meanwhile, O2(a) is much more reactive than O(a) and prefers combustion reaction of C3H6(a). Our results suggest that the coverage of oxygen species strongly influence the selectivity of propylene oxidation reactions on Au(997) surface. Propylene preferentially undergoes combustion reactions with O2(a) whose coverage is large at low temperatures but the partial oxidation reaction to acrolein with O(a) whose coverage is small at high temperatures. These results not only provide a complete understanding of propylene oxidation with molecular O2 on Au surfaces but also represent the first study of oxidation reactions with molecular O2 on Au single crystal model surfaces and greatly advance fundamental understandings of Au-catalyzed oxidation reactions. However, it is noteworthy that co-adsorbed water on Au(997) surface is inevitable due to the employed O2 exposure method. Meanwhile, water is known to be involved in low-temperature oxidation reactions over Au catalysts [41-43], thus the role of water in the observed propylene oxidation reactions with molecular O2 on Au(997) surface can not be excluded and needs further investigations.

Download:
Fig. 5. Schematic illustration of site- and surface species-dependent propylene oxidation with molecular O2 on a Au(997) surface. The yellow, blue, red, grey and white spheres represent Au, O, C in C3H6(a) at the (111) step sites, C in C3H6(a) at the (111) terrace sites and H atoms, respectively (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

In summary, we have successfully prepared molecularly-adsorbed O2 species on Au(997) surface via a near-ambient pressure and low-temperature oxygen exposure and elucidated reaction mechanisms of propylene oxidation with molecular O2 via TDS and PM-IRRAS spectra. Site- and surface species-dependent reaction behaviors are revealed. C3H6(a) adsorbed at the (111) step sites undergoes low-temperature combustion reactions with O2(a) to produce CO2 and CO and high-temperature partial oxidation reactions with O(a) to produce acrolein while C3H6(a) adsorbed at the (111) terrace sites undergoes low-temperature combustion and partial oxidation reactions with O2(a) to respectively produce CO2/CO and acrolein. These results provide a complete understanding of propylene oxidation with molecular O2 on Au surfaces and open a window for fundamental studies of oxidation reactions with molecular O2 on Au single crystal model surfaces.

Acknowledgments

This work was financially supported by the National Key R & D Program of Ministry of Science and Technology of China (No. 2017YFB0602205), the Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDA09030103), the National Natural Science Foundation of China (Nos. 21525313, 91745202), the Changjiang Scholars Program of Ministry of Education of China and Collaborative Innovation Center of Suzhou Nano Science and Technology.

References
[1]
R.A. van Santen, H.P.C.E. Kuipers, J. Adv. Catal. Sci. Technol. 35 (1987) 265-321.
[2]
N.W. Cant, W.K. Hall, J. Catal. 52 (1978) 81-94. DOI:10.1016/0021-9517(78)90125-2
[3]
M. Akimoto, K. Ichikawa, E. Echigoya, J. Catal. 76 (1982) 333-344. DOI:10.1016/0021-9517(82)90264-0
[4]
M.A. Barteau, R.J. Madix, J. Am. Chem. Soc. 105 (1983) 344-340. DOI:10.1021/ja00341a008
[5]
J.T. Roberts, R.J. Madix, W.W. Crew, J. Catal. 141 (1993) 300-307. DOI:10.1006/jcat.1993.1137
[6]
Z. Hu, H. Nakai, H. Nakatsuji, Surf. Sci. 401 (1998) 371-391. DOI:10.1016/S0039-6028(98)00025-9
[7]
W.X. Huang, J.M. White, Langmuir 18 (2002) 9622-9624. DOI:10.1021/la026089z
[8]
W.X. Huang, J.M. White, Catal. Lett. 84 (2002) 143-146. DOI:10.1023/A:1021471718608
[9]
W.X. Huang, Z.Q. Jiang, J.M. White, Catal. Today 131 (2008) 360-366. DOI:10.1016/j.cattod.2007.10.045
[10]
T. Hayashi, K. Tanaka, M. Haruta, J. Catal. 178 (1998) 566-575. DOI:10.1006/jcat.1998.2157
[11]
T.A. Nijhuis, B.J. Huizinga, M. Makkee, J.A. Moulijin, Ind. Eng. Chem. Res. 38 (1999) 884-891. DOI:10.1021/ie980494x
[12]
T.A. Nijhuis, T. Visser, B.M. Weckhuysen, J. Phys. Chem. B 109 (2005) 19309-19319. DOI:10.1021/jp053173p
[13]
T.A. Nijhuis, T. Visser, B.M. Weckhuysen, Chem. Angew., Int. Ed. 44 (2005) 1115-1118. DOI:10.1002/(ISSN)1521-3773
[14]
B. Chowdhury, J.J. Bravo-Suarez, M. Date, S. Tsubota, M. Haruta, Chem. Angew., Int. Ed. 45 (2006) 412-416. DOI:10.1002/(ISSN)1521-3773
[15]
M. Haruta, M. Date, Appl. Catal. A Gen. 222 (2001) 427-437. DOI:10.1016/S0926-860X(01)00847-X
[16]
J.H. Huang, T. Takei, T. Akita, H. Ohashi, M. Haruta, Appl. Catal. B 95 (2010) 430-438. DOI:10.1016/j.apcatb.2010.01.023
[17]
S.L. Chen, B.S. Zhang, D.S. Su, W.X. Huang, ChemCatChem 7 (2015) 3290-3298. DOI:10.1002/cctc.201500599
[18]
G.J. Hutchings, Chem. Comm. (2008) 1148-1194.
[19]
C.D. Pina, E. Falletta, L. Prati, M. Rossi, Chem. Soc. Rev. 37 (2008) 2077-2095. DOI:10.1039/b707319b
[20]
R. Meyer, C. Lemire, S.K. Shaikhutdinov, H.J. Freund, Gold Bull. 37 (2004) 72-124. DOI:10.1007/BF03215519
[21]
M.S. Chen, D.W. Goodman, Acc. Chem. Res. 39 (2006) 739-746. DOI:10.1021/ar040309d
[22]
B.K. Min, C.M. Friend, Chem. Rev. 107 (2007) 2709-2724. DOI:10.1021/cr050954d
[23]
J.L. Gong, Chem. Rev. 112 (2012) 2987-3054. DOI:10.1021/cr200041p
[24]
W.X. Huang, K. Qian, Z.F. Wu, S.L. Chen, Acta Phys. Chim. Sin. 32 (2016) 48-60.
[25]
W.X. Huang, G.H. Sun, T. Cao, Chem. Soc. Rev. 46 (2017) 1977-2000. DOI:10.1039/C6CS00828C
[26]
W.X. Huang, Sci. Sin. Chim. 48 (2018) 1076-1093. DOI:10.1360/N032018-00033
[27]
N.D.S. Canning, D. Outka, R.J. Madix, Surf. Sci. 141 (1984) 240-254. DOI:10.1016/0039-6028(84)90209-7
[28]
N. Saliba, D.H. Parker, B.E. Koel, Surf. Sci. 410 (1998) 270-282. DOI:10.1016/S0039-6028(98)00309-4
[29]
J.M. Gottfried, K.J. Schmidt, S.L.M. Schroeder, K. Christmann, Surf. Sci. 511 (2002) 65-82. DOI:10.1016/S0039-6028(02)01555-8
[30]
X.Y. Deng, B.K. Min, A. Guloy, C.M. Friend, J. Am. Chem. Soc. 127 (2005) 9267-9270. DOI:10.1021/ja050144j
[31]
Z.F. Wu, Y.S. Ma, Y.L. Zhang, et al., J. Phys. Chem. C 116 (2012) 3608-3617. DOI:10.1021/jp210028y
[32]
K.A. Davis, D.W. Goodman, J. Phys. Chem. B 104 (2000) 8557-8562. DOI:10.1021/jp001699y
[33]
X.Y. Deng, B.K. Min, X.Y. Liu, C.M. Friend, J. Phys. Chem. B 110 (2006) 15982-15987. DOI:10.1021/jp062305r
[34]
E. Hahn, H. Schief, V. Marsico, A. Fricke, K. Kern, Phys. Rev. Lett. 72 (1994) 3378-3381. DOI:10.1103/PhysRevLett.72.3378
[35]
M. Giesen, U. Linke, H. Ibach, Surf. Sci. 389 (1997) 264-271. DOI:10.1016/S0039-6028(97)00423-8
[36]
Z.F. Wu, Y.K. Jin, L.S. Xu, et al., J. Phys. Chem. C 118 (2014) 8397-8405. DOI:10.1021/jp500058c
[37]
Z.F. Wu, Z.Q. Jiang, Y.K. Jin, F. Xiong, W.X. Huang, J. Phys. Chem. C 118 (2014) 26258-26263. DOI:10.1021/jp509551d
[38]
Z.F. Wu, Z.Q. Jiang, Y.K. Jin, et al., Sci. China Chem. 59 (2016) 752-759. DOI:10.1007/s11426-015-5510-y
[39]
J.G. Radziszewski, J.W. Downing, M.S. Gudipati, et al., J. Am. Chem. Soc. 118 (1996) 10275-10284. DOI:10.1021/ja961668+
[40]
J. Kim, E. Samano, B.E. Koel, Surf. Sci. 600 (2006) 4622-4632. DOI:10.1016/j.susc.2006.07.057
[41]
M.S. Ide, R.J. Davis, Acc. Chem. Res. 47 (2014) 825-833. DOI:10.1021/ar4001907
[42]
K. Qian, W.H. Zhang, H.X. Sun, et al., J. Catal. 277 (2011) 95-103. DOI:10.1016/j.jcat.2010.10.016
[43]
J. Saavedra, H.A. Doan, C.J. Pursell, L.C. Grabow, B.D. Chandler, Science 345 (2014) 1599-1602. DOI:10.1126/science.1256018