Non-mercury catalytic acetylene hydrochlorination over Ru catalysts enhanced by carbon nanotubes

Abstract

Ru-based catalysts with different deposition sites were prepared using multiwalled carbon nanotubes as the support and RuCl₃ as the precursor, to study the effects of multiwalled carbon nanotubes on the catalytic performance of Ru catalysts for acetylene hydrochlorination. It has been suggested that Ru catalysts deposited inside the CNTs channels exhibit the optimal catalytic activity with the acetylene conversion of 95.0% and the selectivity to VCM of 99.9% after 10 h on stream under the conditions of 170 °C and GHSV (C₂H₂) of 90 h⁻¹. In combination with characterizations of BET, TEM, XRD, TPR, TPD and XPS, it is illustrated that the CNTs with the inner diameter about 3–7 nm can functionalize as an efficient support with unique electron property to enhance the catalytic performance of Ru-based catalysts for acetylene hydrochlorination.

---

1. Introduction

Acetylene hydrochlorination reaction is an important coal-based industrial process to produce the vinyl chloride monomer (VCM), which is the monomer to manufacture polyvinyl chloride via polymerization.¹ The reaction is carried out industrially using activated carbon-supported mercuric chloride as a catalyst,² which causes serious environmental pollution owing to the high toxicity and volatility of the active mercuric chloride component. Therefore, it is important to explore a reliable and environmental-benign non-mercury catalyst to substitute the poisonous mercuric chloride for acetylene hydrochlorination. Non-mercuric catalysts, involving the main metallic component of Au,^3–5 Pd^6,7 and Ru,^8–10 have been studied extensively, following the pioneer work of Hutchings.¹¹ However, it is still a challenge to develop an efficient non-mercury catalyst with high activity and long-term stability.

Multiwalled carbon nanotubes (CNTs) have a well-defined tubular structure formed by graphene layers with an electron-deficient interior surface and an electron-enriched exterior surface,^12–15 and are considered promising supports to adjust the activity of dispersed metal catalysts.^16–22 For example, Bao and co-workers studied the effect of CNTs (with the inner and outer diameters about 4–8 and 10–20 nm, respectively) on the catalytic performance of Ru nanoparticles for ammonia synthesis reaction and reported that metallic Ru nanoparticles dispersed on the outside of CNTs displayed an about two times higher turnover frequency than those dispersed inside the CNT channels.²³ Ran et al. studied the cellobiose conversion reaction over Ru nanoparticles and reported that the catalytic activity of Ru nanoparticles dispersed inside the CNT channels was higher than that dispersed on the outside of the CNTs (with the inner and outer diameters about 3–6 and 10–20 nm, respectively).²⁴ It has been suggested that the beneficial deposition sites on CNTs for metallic catalysts are greatly associated with the distinct chemical reactions and the diameters of CNTs. Recently, Li et al. reported that polypyrrole-modified multiwalled carbon nanotubes (PPy-MWCNT) can enhance the catalytic activity of Au-based catalysts for acetylene hydrochlorination.²⁵ These results prompted us to study the effects of different Ru deposition sites of CNTs on the acetylene hydrochlorination reaction.

In this article, we adopted multiwalled CNTs as the supports to prepare Ru-based catalysts deposited on the outside of CNTs or inside the channels of CNTs, and assessed the catalytic activity of these two kinds of Ru-based catalysts for acetylene hydrochlorination. BET, TEM, XRD, TPR, TPD and XPS indicated that the Ru-based catalysts deposited inside the channels of the CNTs show high catalytic activity for acetylene hydrochlorination.

2. Experimental

2.1 Materials

Analytical grade RuCl₃·3H₂O (the content of Ru assay 37.0%) was purchased from Xi'an Kaida Chemical, Ltd. (China) and used without any purification. Two kinds of multiwalled CNTs (raw-CNT, raw-CNT-M) were purchased from Chengdu Organic Chemicals Co., LTD, China. The raw materials of multiwalled CNTs were treated by refluxing in concentrated HNO₃ (68.0%) at 140 °C for 14 h, followed by filtration, washing and then desiccation at 60 °C for 12 h, to make the nanotube terminals open and the length of nanotubes cut into of 200–500 nm segments. The as-prepared CNTs were adopted as the supports to prepare the Ru-based catalysts further. As shown in Fig. S1 of ESI,† the support CNT has an inner and outer diameter of 3–7 nm and 8–15 nm, respectively (denoted as CNT), while another support has an inner and outer diameter of 5–10 nm and 20–30 nm, respectively (denoted as CNT-M).

2.2 Catalysts preparation

Adopting the support CNT, Ru-based catalysts deposited inside the channels of CNT (denoted as Ru-in-CNT) or on the outer surface of CNTs (denoted as Ru-out-CNT) were prepared using an improved wet chemistry method.^26–28 For the synthesis of Ru-in-CNT, the CNTs (1.5 g) were dispersed in a 70 mL solution of RuCl₃ in acetone by sonication for 6 h; the mixture was stirred continuously at room temperature to allow the slow evaporation of acetone, followed by heat treatment in a tube furnace at 150 °C for 18 h with an air flow rate of 25 mL min⁻¹. For the synthesis of Ru-out-CNT, the CNTs (1.5 g) were first dispersed in a xylene solution (70 g) by an ultrasonic treatment for 6 h to fill the channels of the CNTs with xylene, then mixed with a RuCl₃ aqueous solution (2 mL) under magnetically stirring at 80 °C. The obtained mixture was also treated in a tube furnace at 150 °C for 18 h with an air flow rate of 25 mL min⁻¹. To make the discussion clear in the next context, the blank support in-CNT was used to indicate the support CNT that experienced the same treatment procedure to prepare the Ru-in-CNT catalysts but without the ruthenium trichloride precursors, whereas the blank support out-CNT was the support CNT that experienced the same treatment to prepare Ru-out-CNT catalysts without the ruthenium trichloride precursors.

In the case of another support CNT-M, we adopted the same procedures to prepare Ru-based catalysts deposited inside the channel of CNT-M (denoted as Ru-in-CNT-M) or on the outer surface of the CNT-M (denoted as Ru-out-CNT-M). The Ru loading of all catalysts was 1 wt% in this study, as confirmed by the atomic absorption spectroscopy.

2.3 Catalyst characterization

N₂ adsorption/desorption experiments were conducted using a Quantachrome NOVA BET 2200e analyzer. The samples were first degassed at 300 °C for 4 h and analyzed via liquid nitrogen adsorption at −196 °C. Transmission electron microscopy (TEM) was performed on a JEM 2100F field emission transmission electron microscope (JEOL, Tokyo, Japan) working at 200 kV in scanning TEM mode (spot size, 0.4 nm). For sample preparation, the samples were first reduced in H₂ at 450 °C for 5 h and dispersed ultrasonically in ethanol, and then some droplets of the suspension was dipped onto a holey carbon-coated copper grid and dried. X-ray powder diffraction (XRD) was performed on a Bruker D8 Focus diffractometer using Cu Kα radiation at λ = 1.54056 Å with a scanning speed of 4° min⁻¹ and a step of 0.02° (2θ) in the range from 20° to 80°. The X-ray photoelectron spectra (XPS) were obtained using a PHI 5000 Versa Probe (ULVAC-PHI Inc., Osaka, Japan) employing monochromatic Al Kα X-rays (hν = 1486.7 eV) under high vacuum conditions. The data was collected at a sample tilt angle of 45°. The binding energies were corrected using the C 1s peak of aliphatic carbon at 284.8 eV as an internal standard. Atomic absorption spectroscopy (AAS) was performed with a Perkin-Elmer 800 atomic absorption spectrometer using an air-acetylene flame. H₂ temperature programmed reduction (H₂-TPR) was performed on a TPDRO 1100 apparatus equipped with a thermal conductivity detector. For each test, 100 mg sample was heated from room temperature to 800 °C at a rate of 10 °C min⁻¹, and flushed with a 20 mL min⁻¹ gas mixture containing 5% H₂ in N₂ gas. Temperature-programmed desorption (TPD) was analyzed by TPDRO 1100 apparatus. The samples were pretreated under hydrogen chloride and acetylene atmosphere at the reactive temperature (170 °C) for 6 h, respectively. High-purity N₂ (50 mL min⁻¹) was then passed through the sample at 100 °C for 30 min. The TPD profiles were recorded for the sample heated from 100 °C to 650 °C at a rate of 10 °C min⁻¹.

2.4 Catalytic performance evaluation

The catalytic performance was investigated using a fixed-bed glass micro-reactor (i.d. of 8 mm). Acetylene (99.9% purity) was passed through silica-gel desiccant to remove the trace impurities, and hydrogen chloride gas (99.9% purity) was dried using 5A molecular sieves. Acetylene (3 mL min⁻¹) and hydrogen chloride (3.3 mL min⁻¹) were introduced into a heated reactor containing catalyst (2 mL) through a mixing vessel via calibrated mass flow controllers, giving a C₂H₂ gas hourly space velocity (GHSV) of 90 h⁻¹ at 170 °C. The microreactor was purged with nitrogen before the reaction to remove water and air. The reactor effluent was passed through an absorption bottle containing a sodium hydroxide solution to remove the unreacted hydrogen chloride. The gas mixture was analyzed using a Beifen GC-3420A gas chromatograph (GC).

3. Results and discussion

3.1 Catalyst characterization

3.1.1 Catalyst texture properties. The BET measurements were taken to investigate the physical structure changes to the CNT caused by the treatment of nitric acid, acetone, or xylene. Table 1 lists the specific surface area, pore volume and pore diameter of the raw material CNT, the CNT support treated by nitric acid, the blank in-CNT support experienced the same treatment procedure to prepare the Ru-in-CNT catalysts but without the ruthenium precursors, and the blank out-CNT support experienced the same treatment procedure to prepare the Ru-out-CNT catalysts without ruthenium precursors, as well as the fresh catalysts of Ru-in-CNT and Ru-out-CNT. The nitric acid treatment increased the specific surface area from 192 to 236 m² g⁻¹, the total pore volume from 0.28 to 0.32 cm³ g⁻¹, and the pore diameter from 3.03 to 3.82 nm. This indicates that further acetone or xylene treatments result in a small decrease in the surface area and pore volume, comparing in-CNT or out-CNT with CNT. The changes in the morphology of these different nanotubes were characterized by TEM, as displayed in Fig. S2.† After the deposition of ruthenium chloride, the surface area of the fresh catalyst Ru-in-CNT (220 m² g⁻¹) decreased by 5% compared to the in-CNT; while the pore volume of Ru-in-CNT (0.29 cm³ g⁻¹) decreased by 3% compared to the in-CNT, which is due to the partial blocking of ruthenium species inside the support CNT. In the case of fresh catalyst Ru-out-CNT, both the surface area and the pore volume are similar to those of the out-CNT.

Table 1 Pore structure parameters of different CNTs and the supported catalysts

Catalyst	SBET (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore diameter (nm)
a The raw multiwalled CNTs were purchased from Chengdu Organic Chemicals Co., LTD, China.b The CNTs treated by refluxing in concentrated nitric acid at 140 °C for 14 h, which are used as the support to prepare the catalysts.c The blank in-CNT support, which is experienced the same treatment procedure to prepare Ru-in-CNT catalysts but the without ruthenium trichloride precursors.d The blank out-CNT support, which experienced the same treatment procedure to prepare Ru-out-CNT catalysts but without ruthenium trichloride precursors.e The data is reported along with the standard deviation.
Raw-CNT^a	192 ± 0.5^e	0.28 ± 0.006	3.03 ± 0.006
CNT^b	236 ± 0.8	0.32 ± 0.003	3.82 ± 0.006
In-CNT^c	232 ± 0.8	0.30 ± 0.005	3.82 ± 0.002
Out-CNT^d	233 ± 0.5	0.31 ± 0.003	3.83 ± 0.003
Ru-out-CNT	232 ± 0.6	0.31 ± 0.005	3.81 ± 0.008
Ru-in-CNT	220 ± 0.8	0.29 ± 0.007	3.80 ± 0.007

---

3.1.2 Dispersion of Ru particles. Fig. 1 shows the XRD patterns of the support CNT, the fresh catalysts Ru-in-CNT and Ru-out-CNT. Apart from the four characteristic diffraction peaks of CNT located at 25.6°, 42.6°, 53.1° and 77.7°, respectively,²⁹ neither the fresh Ru-in-CNT nor Ru-out-CNT showed peaks indicative of the hexagonal close-packed (hcp) metallic Ru phase or anhydrous tetragonal RuO₂, indicating that all Ru particles are very small (with the size lower than 4 nm),³⁰ which is in accord with the TEM images (Fig. 2).

Fig. 1 X-ray diffraction patterns of (a) the support CNT and the fresh catalysts (b) Ru-in-CNT and (c) Ru-out-CNT.

Fig. 2 TEM images and the particle size distributions of the fresh catalysts of Ru-in-CNT (a and c) and Ru-out-CNT (b and d). (Red circles: Ru nanoparticles confined within channels of CNT; blue squares: Ru nanoparticles located on external surfaces of CNT.)

Fig. 2 displays typical TEM images and the particle size distributions of the fresh catalysts Ru-in-CNT and Ru-out-CNT reduced in H₂ at 450 °C for 5 h. For the fresh catalyst Ru-in-CNT, Ru nanoparticles inside the channel of nanotubes have an average size of about 0.95 nm, which is smaller than the inner diameter of CNT (3–7 nm). The percentage of Ru nanoparticles deposited inside the channel of CNT was calculated by counting the locations of 150–200 Ru particles on at least 100 nanotubes. This indicates that more than eighty percent of ruthenium particles have been introduced into the inner cavity of nanotubes (Fig. 2a). For the fresh catalyst Ru-out-CNT, Ru nanoparticles are distributed exclusively on the exterior surface of CNT with an average size about 1.01 nm (Fig. 2b).

3.1.3 Reducibility and adsorption property of Ru-based catalysts. The H₂-TPR profiles were measured to evaluate the reducibility of two kinds of Ru-based catalysts using CNTs as the support. As shown in Fig. 3, the TPR profiles of the fresh catalyst Ru-in-CNT and Ru-out-CNT are distinct from those of the blank support CNT. Two broad peaks in the range of 350–800 °C were observed for all the samples, which were attributed to the reduction of oxygenated groups in the CNT support.^23,31 There is a broad H₂ consumption peak in the temperature range of 100–350 °C for both Ru-in-CNT and Ru-out-CNT catalysts, compared to the profile of the support CNT, which is due to the reduction of ruthenium species involving the ruthenium oxides and ruthenium chloride in the catalysts.^23,32,33 The reduction of ruthenium species takes place around 292 °C for Ru-in-CNT, whereas it occurs at a higher temperatures (311 °C) for Ru-out-CNT. The interaction between the electron-deficient concave surface of the carbon nanotubes and the anionic chlorine in RuCl₃ or the anionic oxygen in RuO₂ could lead to a weakening of the bonding strength of RuCl₃ or RuO₂ and consequently make it easier to reduce ruthenium species inside the channels of the CNT. On the contrary, for Ru-out-CNT, the weak interactions between the electron density-enriched outer surfaces of nanotubes with the anionic chlorine in RuCl₃ or the anionic oxygen in RuO₂ have less influence on the reducibility of ruthenium species.^15,23 Previous studies also reported that ruthenium species inside the CNT channel are easier to reduce compared to the outside ones.²³

Fig. 3 H₂-TPR profiles of the support CNT (a), and the fresh catalysts of (b) Ru-in-CNT and (c) Ru-out-CNT.

TPD experiments were carried out to illustrate the adsorption property of the fresh catalysts Ru-in-CNT and Ru-out-CNT towards hydrogen chloride and acetylene. As shown in Fig. 4, the support CNT showed no obvious adsorption of hydrogen chloride, while the catalysts Ru-in-CNT and Ru-out-CNT showed the obvious desorption peak of hydrogen chloride in the range of 205–295 °C, and the desorption area of hydrogen chloride from Ru-in-CNT is significantly larger than that from Ru-out-CNT. For another reactant acetylene, as shown in Fig. 5, the catalysts Ru-in-CNT and Ru-out-CNT showed the desorption peak of acetylene in the range of 190–430 °C, and the desorption area of acetylene from Ru-in-CNT was also larger than that from Ru-out-CNT. This indicates that the catalyst Ru-in-CNT shows enhanced adsorption of both hydrogen chloride and acetylene, suggesting that the confinement within the channels of the CNTs results in more active metallic sites in Ru-in-CNT and consequently promotes higher catalytic activities for the acetylene hydrochlorination reaction, as mentioned in Section 3.2.

Fig. 4 HCl-TPD profiles of the support CNT (a), and the fresh catalysts of (b) Ru-in-CNT and (c) Ru-out-CNT.

Fig. 5 C₂H₂-TPD profiles of the support CNT (a), and the fresh catalysts of (b) Ru-in-CNT and (c) Ru-out-CNT.

3.1.4 Ruthenium species associated with deposition sites on CNTs. The Ru 3p3/2 XPS spectra of the fresh catalysts Ru-in-CNT and Ru-out-CNT were deconvoluted into five peaks at 461.7 eV, 462.7 eV, 463.5 eV, 464.8 eV and 466.2 eV (Fig. S3†), corresponding to the metallic Ru, Ru/RuO_y, RuCl₃, RuO₂ and RuO_x species, respectively.^34–37 The relative content and binding energy of all the five Ru species are listed in Table 2. This indicates that the dominant species of Ru-in-CNT are RuO₂ (46.6%) followed by Ru/RuO_y (24.4%), RuO_x (14.5%), RuCl₃ (10.9%) and metallic Ru (3.6%), while the major species of Ru-out-CNT include metallic Ru (28.1%), RuO₂ (25.9%) and RuCl₃ (24.5%). The interior surface of CNT is electron-deficient whereas the exterior surface is electron-enriched.^12,13,16 Thus, the ruthenium inside the channel of CNT works like the donor of electrons so that the dominant species are high-valence oxidation states, while the ruthenium deposited on the outer surface of the CNT can easily accept electrons from CNTs to generate a larger amount of metallic Ru. This suggests that the deposition site of the ruthenium precursor plays an important role in affecting the distribution of the valence states of ruthenium species in the catalysts.

Table 2 Binding energy (eV) and relative content (Area%) of ruthenium species in the fresh catalysts Ru-in-CNT and Ru-out-CNT

Catalysts	Ru^0	Ru/RuOy	RuCl3	RuO2	RuOx
	eV (Area%)	eV (Area%)	eV (Area%)	eV (Area%)	eV (Area%)
Ru-in-CNT	461.7 (3.6)	462.7 (24.4)	463.5 (10.9)	464.9 (46.6)	466.2 (14.5)
Ru-out-CNT	461.2 (28.1)	462.3 (13.0)	463.1 (24.5)	464.8 (25.9)	466.0 (8.5)

---

3.2 Catalytic performance for acetylene hydrochlorination

The catalytic activity of different carbon nanotube supports was measured under the conditions of 170 °C and GHSV (C₂H₂) of 90 h⁻¹, as shown in Fig S4.† This indicates that the initial acetylene conversion is as low as 2.5% over raw-CNT, CNT, in-CNT, and out-CNT, and decreases rapidly. Fig. 6a and b show the catalytic performance of Ru-in-CNT and Ru-out-CNT for acetylene hydrochlorination. Over the catalyst Ru-out-CNT, the initial acetylene conversion is 45.2% and decreases to 37.2% after a 10 h reaction. In contrast, over the Ru-in-CNT, the initial acetylene conversion was 99.1% and decreased to 95.0% after 10 h, and the selectivity to VCM maintains 99.9%.

Fig. 6 Catalytic performance of Ru-based catalysts deposited inside the channel of the support CNT (a and b) and CNT-M (c and d), and those deposited in the outer surface of individual support. Reaction conditions: temperature (T) = 170 °C; C₂H₂ gas hourly space velocity (GHSV) = 90 h⁻¹; feed volume ratio V_HCl/V_C2H2 = 1.1.

Adopting another carbon nanotube support CNT-M with the inner diameter of 5–10 nm, the catalytic performance of Ru-in-CNT-M and Ru-out-CNT-M is shown in Fig. 6c and d, under the same reaction conditions. Over the Ru-out-CNT-M catalyst, the acetylene conversion decreased from 42.8% to 24.9% within a 10 h reaction, whereas the selectivity to VCM increased somewhat from the initial 98.8–99.8% at 10 h. Over the Ru-in-CNT-M catalyst, the acetylene conversion decreased from 91.4% to 80.1% within 10 h, whereas the VCM selectivity was maintained at 99.9%.

To disclose the reason that the catalytic performance of Ru catalysts is dependent on the deposition sites of the ruthenium precursors on the supports, the TEM images and the particle size distributions of the fresh catalysts Ru-in-CNT-M and Ru-out-CNT-M were analyzed. As shown in Fig. S5,† for the fresh catalyst, Ru-in-CNT-M, Ru nanoparticles inside the channel of nanotubes have an average size of about 1.66 nm. The percentage of Ru nanoparticles deposited inside the channels of CNT-M was calculated by counting the locations of 200 Ru particles on at least 100 nanotubes. This indicates that more than ninety percent of the ruthenium particles were introduced into the inner cavities of nanotubes. For the fresh catalyst Ru-out-CNT-M, Ru nanoparticles were distributed exclusively on the exterior surface of CNT-M with an average size of about 2.27 nm. Table S1† lists the surface area and pore diameter of the support CNT-M and supported Ru catalysts. After deposition of the ruthenium precursors, the surface area of Ru-in-CNT-M decreased slightly more than that of Ru-out-CNT-M, which is similar to those supported on CNT.

In addition, through deconvolution of the Ru 3p3/2 XPS spectra of the fresh catalysts, Ru-in-CNT-M and Ru-out-CNT-M, the relative content and binding energy of the Ru species are compared with those supported on CNT. As shown in Fig. S5,† there are five peaks at 461.5 eV, 462.7 eV, 463.1 eV, 464.7 eV, and 466.0 eV, due to the metallic Ru, Ru/RuO_y, RuCl₃, RuO₂, and RuO_x species, respectively. The relative content and binding energy of all five Ru species are listed in Table S2.† This suggests that the major species of Ru-in-CNT-M included RuO₂ (35.5%), Ru/RuO_y (24.7%), RuCl₃ (22.5%), metallic Ru (14.3%), and RuO_x (3.0%), while the dominant species of Ru-out-CNT-M are RuCl₃ (70.1%) followed by metallic Ru (13.7%), RuO₂ (10.4%), Ru/RuO_y (3.5%), and RuO_x (2.3%). This indicates that RuO₂ is the most abundant species in both Ru-in-CNT and Ru-in-CNT-M. Previous work suggested that RuO₂ is the important active ingredient for the acetylene hydrochlorination.¹⁰ Therefore, ruthenium catalysts deposited in the channels of the CNTs with an inner diameter of 3–7 nm exhibit the optimal catalytic performance for acetylene hydrochlorination, which is associated with the abundance of RuO₂.

Previous studies reported that phenol, ether, and carbonyl groups on an activated carbon surface are important for improving the catalytic activity of Au-based catalysts.³⁸ It is reasonable to consider that the CNTs that experienced the treatment of nitric acid, acetone, or xylene possess different functional groups on the surfaces, which are probably associated with the catalytic performance of the Ru-based catalysts. The effects of surface functional groups on the Ru-based catalysts will be studied in future work.

4. Conclusions

Ru-based catalysts with different deposition sites were prepared using multiwalled carbon nanotubes as the support and RuCl₃ as the precursor to study the effects of multiwalled carbon nanotubes on the catalytic performance of Ru catalysts for acetylene hydrochlorination. Characterized by BET, TEM, XRD, TPR, TPD and XPS, it is suggested that the Ru catalysts deposited inside the CNTs channels exhibit the optimal catalytic activity with an acetylene conversion of 95.0% and a selectivity to VCM of 99.9% after 10 h on stream under the conditions of 170 °C and GHSV (C₂H₂) of 90 h⁻¹. This indicates that confinement inside CNTs can greatly influence the amount of ruthenium species involved in Ru⁰, Ru/RuO_y, RuCl₃, RuO₂ and RuO_x in the preparation process of the Ru-in-CNT catalyst, and enhance the adsorption of hydrogen chloride and acetylene over the catalyst. The acetylene conversion over these catalysts at 170 °C and 10 h decreases in the order of Ru-in-CNT (95.0%) > Ru-in-CNT-M (80.1%) > Ru-out-CNT (37.2%) > Ru-out-CNT-M (24.9%). The excellent catalytic performance of the Ru-in-CNT catalyst shows that CNTs with an inner diameter about 3–7 nm can functionalize as an efficient support for Ru-based catalysts to enhance the acetylene hydrochlorination reaction.

Acknowledgements

This work was supported by the Special Funds for the Major State Research Program of China (no. 2012CB720302), the 863 Program (no. 2012AA062901), the NSFC (21176174), and the Program for Chang Jiang Scholars and Innovative Research Team in University (no. IRT1161).

References

1. N. Liu, J. Zhang, W. Li and B. Dai, Front. Chem. Sci. Eng., 2011, 5, 514–520.
2. J. B. Agnew and H. S. Shankar, Ind. Eng. Chem. Prod. Res. Dev., 1986, 25, 19–22.
3. M. Conte, C. J. Davies, D. J. Morgan, T. E. Davies, D. J. Elias, A. F. Carley, P. Johnston and G. J. Hutchings, J. Catal., 2013, 297, 128–136.
4. M. Conte, A. F. Carley, G. Attard, A. A. Herzing, C. J. Kiely and G. J. Hutchings, J. Catal., 2008, 257, 190–198.
5. H. Zhang, B. Dai, W. Li, X. Wang, J. Zhang, M. Zhu and J. Gu, J. Catal., 2014, 316, 141–148.
6. T. V. Krasnyakova, I. V. Zhikharev, R. S. Mitchenko, V. I. Burkhovetski, A. M. Korduban, T. V. Kryshchuk and S. A. Mitchenko, J. Catal., 2012, 288, 33–43.
7. S. A. Mitchenko, T. V. Krasnyakova and I. V. Zhikharev, Theor. Exp. Chem., 2010, 46, 32–38.
8. M. Zhu, L. Kang, Y. Su, S. Zhang and B. Dai, Can. J. Chem., 2013, 91, 120–125.
9. J. Zhang, W. Sheng, C. Guo and W. Li, RSC Adv., 2013, 3, 21062–21068.
10. Y. Pu, J. Zhang, L. Yu, Y. Jin and W. Li, Appl. Catal., A, 2014, 488, 28–36.
11. G. J. Hutchings, J. Catal., 1985, 96, 292–295.
12. R. C. Haddon, Science, 1993, 261, 1545–1550.
13. D. Ugarte, A. Châtelain and W. A. de Heer, Science, 1996, 274, 1897–1899.
14. B. Shan and K. Cho, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 73, 081401(R).
15. W. Chen, X. Pan, M. Willinger, D. Su and X. Bao, J. Am. Chem. Soc., 2006, 128, 3136–3137.
16. X. Pan, Z. Fan, W. Chen, Y. Ding, H. Luo and X. Bao, Nat. Mat., 2007, 6, 507–511.
17. W. Chen, Z. Fan, X. Pan and X. Bao, J. Am. Chem. Soc., 2008, 130, 9414–9419.
18. E. Castillejos, P. J. Debouttiere, L. Roiban, A. Solhy, V. Martinez, Y. Kihn, O. Ersen, K. Philippot, B. Chaudret and P. Serp, Angew. Chem., Int. Ed., 2009, 48, 2529–2533.
19. Z. Chen, Z. Guan, M. Li, Q. Yang and C. Li, Angew. Chem., Int. Ed., 2011, 50, 4913–4917.
20. Z. Guan, S. Lu and C. Li, J. Catal., 2014, 311, 1–5.
21. K. Zhou, J. Si, J. Jia, J. Huang, J. Zhou, G. Luo and F. Wei, RSC Adv., 2014, 4, 7766–7769.
22. L. Wang, J. Chen, L. Ge, V. Rudolph and Z. Zhu, RSC Adv., 2013, 3, 12641–12647.
23. S. Guo, X. Pan, H. Gao, Z. Yang, J. Zhao and X. Bao, Chem.–Eur. J., 2010, 16, 5379–5384.
24. M. Ran, Y. Liu, W. Chu and A. Borgna, Catal. Lett., 2013, 143, 1139–1144.
25. X. Li, M. Zhu and B. Dai, Appl. Catal., B, 2014, 142–143, 234–240.
26. E. Dujardin, T. W. Ebbesen, H. Hiura and K. Tanigaki, Science, 1994, 265, 1850–1852.
27. S. C. Tsang, Y. Chen, P. J. F. Harris and M. L. H. Green, Nature, 1994, 372, 159–162.
28. C. Wang, S. Guo, X. Pan, W. Chen and X. Bao, J. Mater. Chem., 2008, 18, 5782–5786.
29. C. Jin, W. Xia, T. C. Nagaiah, J. Guo, X. Chen, M. Bron, W. Schuhmann and M. Muhler, Electrochim. Acta, 2009, 54, 7186–7193.
30. P. Gao, A. Wang, X. Wang and T. Zhang, Catal. Lett., 2008, 125, 289–295.
31. S. Kundu, Y. Wang, W. Xia and M. Muhler, J. Phys. Chem. C, 2008, 112, 16869–16878.
32. J. Kang, S. Zhang, Q. Zhang and Y. Wang, Angew. Chem., Int. Ed., 2009, 48, 2565–2568.
33. A. Guerrero-Ruiz, P. Badenes and I. Rodríguez-Ramos, Appl. Catal., A, 1998, 173, 313–321.
34. B. Folkesson, Acta Chem. Scand., 1973, 27, 287–302.
35. Y. Park, B. Lee, C. Kim, Y. Oh, S. Nam and B. Park, J. Mater. Res., 2009, 24, 2762–2766.
36. B. Yang, Q. Lu, Y. Wang, L. Zhuang, J. Lu, P. Liu, J. Wang and R. Wang, Chem. Mater., 2003, 15, 3552–3557.
37. A. B. Laursen, Y. Y. Gorbanev, F. Cavalca, P. Malacrida, A. Kleiman-Schwarsctein, S. Kegnaes, A. Riisager, I. Chorkendorff and S. Dahl, Appl. Catal., A, 2012, 433–434, 243–250.
38. J. Xu, J. Zhao, J. Xu, T. Zhang, X. Li, X. Di, J. Ni, J. Wang and J. Cen, Ind. Eng. Chem. Res., 2014, 53, 14272–14281.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra12017e

---