Fabrication of Au/CNT hollow fiber membrane for 4-nitrophenol reduction

Abstract

Catalytic membranes have an extensive range of desirable applications in chemical fields. Here, an Au-coated carbon nanotube (Au/CNT) hollow fiber membrane was fabricated by depositing Au nanoparticles on a CNT hollow fiber membrane. The Au/CNT hollow fiber membrane exhibited excellent catalytic activity for the reduction of 4-nitrophenol to 4-aminophenol. The observed improvement in the catalytic activity can be ascribed to the synergistic effect between the Au nanoparticles and CNT membrane. The Au/CNT hollow fiber membrane described in this paper presented high stability in the 4-nitrophenol reduction process. This study offers a new avenue for the exploration of membranes for practical applications.

---

Introduction

4-Nitrophenol (4-NP), which is a highly toxic environmental pollutant, is frequently generated as a byproduct in the manufacture of chemical products such as pharmaceuticals, agrochemicals, and coloring agents.^1–3 The reduction of 4-NP to 4-aminophenol (4-AP) has been proposed because 4-AP can be employed as a dyeing agent, a photographic developer, a corrosion inhibitor in paints, an anticorrosion-lubricating agent in fuels, and an important intermediate of fine chemicals for the production of analgesic and antipyretic drugs.^4–8 The reduction of 4-NP can be easily realized to yield a sole product, 4-AP, at mild temperatures in the presence of NaBH₄ and over a metal catalyst.^9–11 Various metal nanoparticles, such as Au, Ag, Pd, Pt and Ni, have been selected as catalysts.^12–16 Recent attention has focused on Au nanoparticles because the catalytic activity of Au toward the conversion of 4-NP to 4-AP is directly related to particle size, and the catalytic activity of Au is only observed on a nanometric scale.^17,18 However, small-sized Au nanoparticles have a strong tendency to form large aggregates due to their high surface energy, which causes a significant decrease in their original catalytic activities.¹⁹ To prevent aggregation, Au nanoparticles are usually immobilized on different supports, including carbon materials,²⁰ metal oxides²¹ and polymers.²²

Au nanoparticles that are coated on carbon nanotubes (Au/CNTs) have received extensive attention for 4-NP reduction because of their high mechanical stability, reasonable electron conductivity and improved catalytic activity.^23–25 However, the use of Au/CNT composites as catalysts to satisfy practical application requirements remains a challenge because the reusable catalyst particles must be separated from aqueous solution, which involves inconvenient catalyst separation steps. To solve this problem, the approach of immobilizing Au nanoparticles on a CNT membrane has been implemented. Wang et al. reported the fabrication of a hierarchical CNT membrane by growing vertically aligned CNT arrays on stainless steel mesh and demonstrated the feasibility of CNT membrane for supporting Au nanoparticles to eliminate the need for separation.²⁶

Compared with tube-type or plate-type membrane structures, hollow fiber membranes present a relatively higher surface area per volume ratio, which could provide more reaction sites for 4-NP reduction.^27–29 To date, few studies have investigated the fabrication of Au/CNT hollow fiber membranes with catalytic ability. Here, Au/CNT hollow fiber membranes were prepared by coating Au nanoparticles on a CNT hollow fiber membrane, and the performance of the Au/CNT hollow fiber membranes was evaluated for the reduction of 4-NP to 4-AP.

Experimental section

CNT functionalization

CNTs (outer diameters ranging from 60 to 100 nm) were purchased from the Shenzhen Nanotech Port Co. Ltd., China. The CNTs (5 g) were added to 200 mL of a concentrated solution of HNO₃/H₂SO₄ (1 : 3, v/v), and the mixture was subsequently heated to 60 °C for 4 h with stirring. Then, the mixture was diluted with approximately 1500 mL of deionized water and recovered by filtration. Finally, the functionalized CNTs were dried at 50 °C for 14 h.

CNT hollow fiber membrane preparation

The CNT hollow fiber membrane was prepared using the wet-spinning technique and thermal calcination. 1 g of the functionalized CNTs and 0.5 g of polyvinyl butyral were dispersed in 10 g of N-methyl pyrrolidone to form a homogeneous spinning solution with the assistance of sonication. After degassing, the as-prepared spinning solution was squeezed into deionized water using the outer stainless-steel capillary of a spinneret via a constant-flow pump. The length of the CNT hollow fiber membrane can be controlled during wet-spinning. To eliminate the influence of N-methyl pyrrolidone, the as-prepared CNT hollow fiber membranes were submerged in pure water for 24 h and dried overnight at room temperature. The products were calcinated in a flow of argon (40 sccm) at 600 °C for 1 h at a heating rate of 2 °C min⁻¹.

Au/CNT hollow fiber membrane preparation

Chloroauric acid was employed as the precursor for the deposition of Au nanoparticles. One end of the as-prepared CNT hollow fiber membrane was sealed with silicone gel, and the other end was connected to a permeate collector which was combined with a vacuum pump. This CNT hollow fiber membrane was immersed in chloroauric acid solution (0.05 M), and a vacuum of 0.75 bar was applied as a driving force. The deposition process was sustained for 60 s (Au/CNT membrane 1), 180 s (Au/CNT membrane 3) and 300 s (Au/CNT membrane 5), respectively. When the deposition process was completed, the obtained membranes were dried at room temperature.

Characterization

The general morphology of the products was characterized using a scanning electron microscope (SEM; Quanta 200 FEG). The BET surface area and the pore volume of the Au/CNT membranes were determined by the Brunauer–Emmett–Teller method using an automated surface area and pore size analyzer (Quantachrome Autosorb-1 MP). The chemical composition was investigated by transmission electron microscopy (TEM, JEM-2100F, JEOL, Japan) coupled with energy dispersive X-ray (EDX) spectrometry. The crystallinity of the prepared sample was determined by X-ray diffraction (XRD) using a diffractometer with Cu Kα radiation (Shimadzu LabX XRD-6000). The chemical state of the Au/CNT hollow fiber membranes was analyzed by X-ray photoelectron spectroscopy (XPS, ESCALAB250, Thermo-VG Scientific). The Raman spectra were obtained using a Renishaw Micro-Raman System 2000 spectrometer operated with He–Ne laser excitation (wavelength 623.8 nm; laser power 35 mW) with a beam spot size of approximately 2 μm. The pore size of the Au/CNT hollow fiber membranes was verified using a Porometer (Porolux 1000, IB-FT GmbH, Germany).

The porosity (ε) of the Au/CNT hollow fiber membranes was determined by a gravimetric method and calculated using the following formula:²⁸

where m₁ is the mass of dry membrane, m₂ is the mass of wet membrane, l is the effective length, ρ is the water density, d_inner is the inner diameter of the membrane and d_outer is the outer diameter of the membrane. The mechanical properties of the Au/CNT hollow fiber membranes were tested with an electronic tensile machine (Instron 5543) with 50 mm of clamp distance. The pure water flux J of the Au/CNT hollow fiber membranes was calculated with the following formula:²⁹
where V is the water volume penetrating the membrane in a certain time t, S is the total effective area of the membrane and p is the pressure. The adsorbing capacity can be calculated on the basis of the following equations:^30,31

q_e = V(C₀ − C_e)/m

where q_e is the saturating adsorption capacity; V is the volume of the solution; C₀ is the mass concentration of solution before the adsorption; C_e is the mass concentration of solution after adsorption equilibrium; and m is the quality of the Au/CNT hollow fiber membrane.

Catalytic performance

The catalytic reduction of 4-NP to 4-AP was performed in a cylindrical hollow fiber membrane reactor using a peristaltic pump as the driving force, as shown in Fig. 1. The initial concentration of 4-NP was set to 0.08 mM, which was mixed with NaBH₄ (3 mM). During the entire reduction process, the concentration of 4-NP and of its product 4-AP were subsequently determined using a JASCD V-550 UV-vis spectrophotometer. To evaluate the stability of the Au/CNT hollow fiber membrane, a cyclic catalytic reduction experiment was performed.

Fig. 1 Schematic of the catalytic reduction process of 4-NP to 4-AP.

Results and discussion

SEM images

SEM images of the CNT hollow fiber membrane before and after the deposition of Au nanoparticles are presented in Fig. 2. The outer diameter and inner diameter of the prepared hollow fibers are approximately 700 and 390 μm, respectively. As shown in the enlargement in Fig. 2b, the Au nanoparticles were uniformly deposited on the surface of the CNTs without aggregation.

Fig. 2 SEM images of the CNT hollow fiber membrane (a) before and (b) after the deposition of Au nanoparticles.

The quantity of the Au nanoparticles coated on the CNT hollow fiber membrane was also tested at different deposition times. Fig. 3a–c show SEM images of the Au nanoparticles deposited on the CNT membrane for 60, 180 and 300 s. The quantity of Au nanoparticles gradually increased as the deposition time increased from 60 s to 300 s. Fig. 3c also shows that the CNTs were completely covered with the Au nanoparticles when the deposition time was extended to 300 s. Therefore, the duration of the Au nanoparticle deposition was critical for the formation of the Au/CNT hollow fiber membrane. Meanwhile, the size distribution of the Au nanoparticles from Fig. 3b showed that more than 90% of the Au nanoparticles fall in the size range of 9 to 23 nm, and the mean particle diameter was approximately 16 nm (Fig. 3d). In addition, the calculated specific surface area and total pore volume of Au/CNT membrane 3 were 90.3 ± 1.2 m² g⁻¹ and 0.216 ± 0.005 cm³ g⁻¹, respectively.

Fig. 3 SEM images of Au nanoparticles deposited on the CNT hollow fiber membrane for (a) 60 s, (b) 180 s and (c) 300 s; (d) the size distribution of Au nanoparticles on the CNT membrane.

TEM and EDX

A TEM image of an Au/CNT hollow fiber membrane is shown in Fig. 4a. The Au nanoparticle is highly crystalline, with distinct lattice fringes. The lattice spacing of approximately 0.235 nm is consistent with the lattice of the Au (111) plane (JCPDS 04-0784). In addition, it can be seen from Fig. 4a that some irregular lattice fringes appeared in the TEM image (left bottom), and the spacings of these irregular lattice fringes appear to be larger than those of the gold {111} facets. It is believed that these irregular lattice fringes in the TEM image are from amorphous carbon. As shown in Fig. 4b, the EDX analysis indicates that Au is detected in the obtained membranes.

Fig. 4 (a) TEM image and (b) EDX spectrum of the Au/CNT hollow fiber membrane.

XPS analysis

To investigate the surface chemical states of the Au/CNT hollow fiber membranes, XPS spectra of C1s, O1s and Au4f were collected from the prepared samples (Fig. 5). As shown in Fig. 5b and c, the peaks at 284.8 eV and 532.3 eV are ascribed to C1s and O1s, respectively. The XPS spectrum of Au4f from the Au nanoparticles (Fig. 5d) can be assigned to Au4f_7/2 (84.0 eV) and Au4f_5/2 (87.8 eV), respectively, which indicates the reduction of Au³⁺ to Au⁰.^32–34

Fig. 5 XPS spectra of (a) the full scan, (b) C1s, (c) O1s and (d) Au4f of the Au/CNT hollow fiber membrane.

XRD determination

Fig. 6 shows the XRD patterns of the Au/CNT hollow fiber membranes synthesized with different amounts of Au nanoparticles compared with the pure CNT hollow fiber membrane. As shown in Fig. 6, the peak at 26.5° was assigned to the CNTs, and the diffraction peaks at 38.1°, 44.3°, 64.5° and 77.6° correspond to the (111), (200), (220) and (311) crystal planes of Au (JCPDS 4-0783), respectively.³⁵ As the deposition duration of Au increased from 60 to 300 s, the intensity of these four XRD peaks was correspondingly enhanced. Based on the Scherrer equation and the full width at half-maximum of the (101) crystal plane of Au, the average crystal size of Au was estimated to be approximately 13.5 nm, with no significant change in the crystallite size when the Au deposition time was varied.

Fig. 6 XRD patterns of the CNT membrane and the Au/CNT hollow fiber membrane.

Raman investigation

Fig. 7 presents the Raman spectra of Au coated on the CNT hollow fiber membranes with different deposition times. In these spectra, the characteristic peak at 1332 cm⁻¹ is the D-band of the polycrystalline CNT, and the peak observed at 1584 cm⁻¹ is ascribed to the G-band of the graphitized structure.^36,37 Compared with the peaks of the CNT hollow fiber membrane, the sample of the Au/CNT hollow fiber membrane has a similar spectrum but significantly lower absorption intensity. Calculated from Fig. 7, the ratios between the intensities of the D-bands and the intensities of the G-bands (I_D/I_G) are 1.387, 1.236, 1.029 and 0.927 for the pure CNT membrane, Au/CNT membrane 1, Au/CNT membrane 3 and Au/CNT membrane 5, respectively. The I_D/I_G ratio decreases with increasing Au deposition time from 60 to 300 s. These results indicate the presence of Au nanoparticles on the CNT hollow fiber membranes, which is consistent with the XRD results.

Fig. 7 Raman spectra of the CNT membrane and the Au/CNT hollow fiber membranes.

Membrane properties

Table 1 presents the characteristics of the Au/CNT hollow fiber membrane. It can be seen from Table 1 that its mean density, average pore size and porosity are 0.219 g cm⁻³, 289 nm and 86%, respectively. The mechanical properties ranged from 5 to 10 MPa. The pure water flux of the Au/CNT hollow fiber membrane reaches 4700 L (m⁻² h⁻¹), which is 2.5 times higher than that of polyvinylidene fluoride hollow fiber membranes (1900 L m⁻² h⁻¹ bar⁻¹) with similar pore sizes and is also 4 times higher than that of commercial Al₂O₃ ceramic membranes (1200 L m⁻² h⁻¹ bar⁻¹).²⁹ Meanwhile, it is known that the adsorption capability of CNTs is mainly determined by functional groups introduced by oxidation.³⁸ In this work, the CNTs were treated in HNO₃ and H₂SO₄, leading to the introduction of many oxygen-containing functional groups, such as –C=O, –C–OH, and –COOH, to their surfaces. These functional groups attached on the surface of the Au/CNT hollow fiber membrane improve its adsorption capability for 4-NP. As listed in Table 1, an adsorption capacity of 7.5 mg g⁻¹ is obtained for the Au/CNT hollow fiber membrane.

Table 1 General properties of the Au/CNT hollow membrane

Mean density (g cm^−3)	0.219
Mean pore size (nm)	289
Porosity (%)	86 ± 3
Mechanical property (MPa)	5 to 10
Pure water flux (L m^−2 h^−1 bar^−1)	4700 ± 300
Adsorption capacity (mg g^−1)	7.5 ± 1.2

---

Catalytic activity

The Au/CNT hollow fiber membrane, with its high specific surface area, offers more active sites and catalytic reaction centers, which favors enhanced catalytic activity. 4-NP solution exhibits a typical spectral peak with a maximum absorption at 317 nm, whereas the addition of NaBH₄ causes a shift of the peak position to 400 nm due to the formation of 4-NP ions in alkaline conditions.^39,40 As shown in Fig. 8a, the intensity of the UV-vis absorption spectra of 4-NP at 400 nm gradually decreased during the catalytic process, and increasing peaks appeared at 300 nm, which corresponds to the product of 4-AP. To trace the concentration of 4-NP and 4-AP throughout the catalytic reduction process, the dynamic variation of 4-NP and 4-AP is shown in Fig. 8b. The concentration of 4-NP decreased and the corresponding form of 4-AP appeared, followed by an increase in the concentration of this form. Meanwhile, 4-NP disappeared in 15 min and the yield of 4-AP was equivalent to the removal of 4-NP. This indicated that the Au/CNT hollow fiber membrane worked well for the catalysis of 4-NP to 4-AP with high conversion yield in water. Although the catalytic reduction rates were affected by solvent and the reaction rate decreased accordingly with the addition of polar chloroform, Au/CNT hollow fiber membrane catalyzed the reactions with very high selectivity in water. As is known, in addition to 4-nitrophenol, the catalytic reduction of nitroaryl to aminoaryl compounds is traditionally carried out in aqueous phases by hydrogenation with NaBH₄ in the presence of Au nanoparticles. However, it is important to note that, according to a previous report by Zhou and Yang, the selective reduction of 4-nitroacetophenone to 4-aminoacetophenone could not be achieved by NaBH₄ in aqueous media over Au nanoparticles because NaBH₄ releases sodium hydride in water, which reacts with C=O preferentially due to the typical nucleophilic addition of aldehydes and ketones.¹⁵

Fig. 8 (a) Catalytic reduction of 4-NP to 4-AP using the Au/CNT hollow fiber membrane; (b) concentration variation of 4-NP and 4-AP for the catalytic reduction of Au/CNT hollow fiber membrane.

As shown in Fig. 9, the concentration of 4-NP (initial concentration of 0.08 mM) remained nearly unchanged without catalyst, which indicates that the reduction of 4-NP can be neglected in the absence of catalyst. The concentration of 4-NP decreased when the Au/CNT hollow fiber membrane was employed. The catalytic reduction of 4-NP followed pseudo-first-order kinetics by the linear transform ln C₀/C_t = Kt (C₀ is the initial concentration of 4-NP, C_t is the concentration of 4-NP at time t, and K is the kinetic constant). For the given experimental conditions, the kinetic constant of the 4-NP reduction with the sample of Au/CNT membrane 1 was 0.155 min⁻¹. In the presence of Au/CNT membrane 3, the kinetic constant of the sample was enhanced to 0.257 min⁻¹. When a sample of Au/CNT membrane 5 was applied, the kinetic constant of 4-NP reduction increased to 0.277 min⁻¹. In this regard, the catalytic activities of sample Au/CNT membrane 3 and Au/CNT membrane 5 exceeded the catalytic activity of Au/CNT membrane 1 by factors of 1.66 and 1.79, respectively. These results proved that the quantity of Au nanoparticles on the CNTs served a critical role in 4-NP reduction. Compared with Au/CNT membrane 3, the application of Au/CNT membrane 5 did not significantly promote 4-NP reduction. In consideration of economy of the Au precursor, the following catalytic reduction experiments were performed with Au/CNT membrane 3.

Fig. 9 Comparison of the catalytic activity of samples of Au/CNT membrane 1, Au/CNT membrane 3, and Au/CNT membrane 5 for 4-NP reduction.

4-NP reduction mechanism

Based on the experimental results, a plausible mechanism for the catalytic reduction of 4-NP to 4-AP by the Au/CNT hollow fiber membrane was proposed in the presence of BH₄⁻ and catalytic Au nanoparticles (Fig. 10). The process can be illustrated based on the following equations:

3BH₄⁻ + catalyst → 3BH₄⁻ − catalyst (1)

3BH₄⁻ − catalyst + 12H₂O → 24H − catalyst + 3B(OH)₄⁻ (2)

4(4-NP) + 24H − catalyst → 4(4-NP) − catalyst − 24H (3)

4(4-NP) − catalyst − 24H → 4(4-AP) − catalyst + 8H₂O (4)

4(4-AP) − catalyst → 4(4-AP) + catalyst (5)

Fig. 10 Schematic of the 4-NP reduction process on the surface of the Au/CNT hollow fiber membrane.

Eqn (1) and (2) exhibit the adsorption of BH₄⁻ and its reaction with water on the surface of the catalyst in the process of liberating hydrogen radicals from the borohydrite ions. The reactant 4-NP also adsorbs on the Au/CNT hollow fiber membrane (eqn (3)). The 4-NP adsorbed on the surface of Au is reduced to 4-AP by replacing the oxygen in the nitro group to form an amino group (eqn (4)). The 4-AP yield in eqn (5) occurs in water by desorption of the formation product from the surface of the catalyst.

The adsorption of 4-NP on the catalyst surface and the electron transfer mediated by the catalyst surface from BH₄⁻ to 4-NP are important factors for 4-NP reduction.^41,42 Thus, the adsorption ability and the number of active sites significantly influenced the catalytic activity. CNTs, which are rolled into circular bundles with graphite sheets, have become an attractive candidate for the fabrication of nanostructures due to their strong adsorption capacity and large specific surface area. Coating Au nanoparticles on a CNT hollow fiber membrane not only prevents the aggregation of Au nanoparticles but also provides large reactive sites that are exposed in the reaction medium. Relatively strong π–π interactions between 4-NP and the CNTs may provide a high concentration of 4-NP near the Au nanoparticles on the CNT membrane, which promotes highly efficient contact. The CNTs can serve as a highly conductive framework to enhance the electron transfer of the Au nanoparticle surface from BH₄⁻ to 4-NP due to the outstanding carrier mobility and conductivity. Therefore, the excellent catalytic activity of the Au/CNT hollow fiber membrane for 4-AP production is attributed to the attractive structure of the membrane. Strong synergistic interactions between Au nanoparticles and CNTs were observed, which significantly enhance the catalytic activity. In particular, the hollow fiber membranes enable continuous 4-AP production without the necessity for separation of the catalysts.

Stability

The stability of the Au/CNT hollow fiber membrane was also evaluated, and a cyclic catalytic experiment was performed via the reduction of 4-NP (initial concentration 0.08 mM) to 4-AP. As displayed in Fig. 11, the Au/CNT hollow fiber membrane did not exhibit a distinct loss of activity after six cycles; thus, the Au/CNT hollow fiber membrane described in this paper presents high stability for 4-AP production.

Fig. 11 Reusability of the Au/CNT hollow fiber membrane for 4-AP production.

Conclusions

The Au/CNT hollow fiber membrane discussed in this paper demonstrates promising catalytic ability for the reduction of 4-NP to 4-AP. The enhanced catalytic activity of the Au/CNT hollow fiber membrane may be attributed to the synergistic effects between Au nanoparticles and the CNT membrane, in which the CNT membrane served a key role in accelerating the adsorption ability of Au nanoparticles for the catalytic reduction of 4-NP to 4-AP. The Au/CNT hollow fiber membrane displayed superior recyclability, and no significant catalytic activity loss was observed after reuse for six cycles. This study may create a new avenue toward the practical application of Au/CNT hollow fiber membranes as a convenient approach in the field of chemical production.

Conflict of interest

The authors declare no competing financial interest.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (51478075) and the Natural Science Foundation of Liaoning Province of China (2014020149).

References

1. D. Xu, P. Diao, T. Jin, Q. Y. Wu, X. F. Liu, X. Guo, H. Y. Gong, F. Li, M. Xiang and Y. Ronghai, ACS Appl. Mater. Interfaces, 2015, 7, 16738–16749.
2. W. B. Zhang, X. M. Xiao, T. C. An, Z. G. Song, J. M. Fu, G. Y. Sheng and M. C. Cui, J. Chem. Technol. Biotechnol., 2003, 78, 788–794.
3. S. Gazi and R. Ananthakrishnan, Appl. Catal., B, 2011, 105, 317–325.
4. P. Dauthal and M. Mukhopadhyay, Ind. Eng. Chem. Res., 2012, 51, 13014–13020.
5. S. Gu, S. Wunder, Y. Lu, M. Ballauff, R. Fenger, K. Rademann, B. Jaquet and A. Zaccone, J. Phys. Chem. C, 2014, 118, 18618–18625.
6. S. Wunder, F. Polzer, Y. Lu, Y. Mei and M. Ballauff, J. Phys. Chem. C, 2010, 114, 8814–8820.
7. K. L. Wu, X. W. Wei, X. M. Zhou, D. H. Wu, X. W. Liu, Y. Ye and Q. Wang, J. Phys. Chem. C, 2011, 115, 16268–16274.
8. S. Panigrahi, S. Basu, S. Praharaj, S. Pande, S. Jana, A. Pal, S. K. Ghosh and T. Pal, J. Phys. Chem. C, 2007, 111, 4596–4605.
9. P. X. Zhao, X. W. Feng, D. S. Huang, G. Y. Yang and D. Astruc, Coord. Chem. Rev., 2015, 287, 114–136.
10. S. Wunder, Y. Lu, M. Albrecht and M. Ballauff, ACS Catal., 2011, 1, 908–916.
11. W. L. Xu, J. S. Kong, Y. E. Yeh and P. Chen, Nat. Mater., 2008, 7, 992–996.
12. Y. C. Chang and D. H. Chen, J. Hazard. Mater., 2009, 165, 664–669.
13. P. Zhang, C. L. Shao, Z. Y. Zhang, M. Y. Zhang, J. B. Mu, Z. C. Guo and Y. C. Liu, Nanoscale, 2011, 3, 3357–3363.
14. Q. Wang, W. J. Jia, B. C. Liu, A. Dong, X. Gong, C. Y. Li, P. Jing, Y. J. Li, G. R. Xu and J. Zhang, J. Mater. Chem. A, 2013, 1, 12732–12741.
15. J. Zhou and Q. Yang, Chem.–Asian J., 2012, 7, 2045–2050.
16. Y. G. Wu, M. Wen, Q. S. Wu and H. Fang, J. Phys. Chem. C, 2014, 118, 6307–6313.
17. A. S. K. Hashmi and G. J. Hutchings, Angew. Chem., Int. Ed., 2006, 45, 7896–7936.
18. N. R. Shiju and V. V. Guliants, Appl. Catal., A, 2009, 356, 1–17.
19. J. Li, C. Y. Liu and Y. Liu, J. Mater. Chem., 2012, 22, 8426–8430.
20. Z. Zanolli, R. Leghrib, A. Felten, J. J. Pireaux, E. Llobet and J. C. Charlier, ACS Nano, 2011, 5, 4592–4599.
21. S. Ivanova, W. Pitchon, Y. Zimmermann and C. Petit, Appl. Catal., A, 2006, 298, 57–64.
22. F. Shi, Q. H. Zhang, Y. B. Ma, Y. D. He and Y. Q. Deng, J. Am. Chem. Soc., 2005, 127, 4182–4183.
23. X. Z. Wang, J. W. Fu, M. H. Wang, Y. J. Wang, Z. M. Chen, J. N. Zhang, J. F. Chen and Q. Xu, J. Mater. Sci., 2014, 49, 5056–5065.
24. J. Liu, R. H. Liu, H. T. Li, W. G. Kong, H. Huang, Y. Liu and Z. H. Kang, Dalton Trans., 2014, 43, 12982–12988.
25. J. Li, C. Y. Liu and Y. Liu, J. Mater. Chem., 2012, 22, 8426–8430.
26. H. T. Wang, Z. X. Dong and C. Z. Na, ACS Sustainable Chem. Eng., 2013, 1, 746–752.
27. G. L. Wei, X. Quan, S. Chen, X. F. Fan, H. T. Yu and H. M. Zhao, ACS Appl. Mater. Interfaces, 2015, 7, 14620–14627.
28. G. L. Wei, H. T. Yu, X. Quan, S. Chen, H. M. Zhao and X. F. Fan, Environ. Sci. Technol., 2014, 48, 8062–8068.
29. G. L. Wei, S. Chen, X. F. Fan, X. Quan and H. T. Yu, J. Membr. Sci., 2015, 493, 97–105.
30. W. L. Wang, Q. Y. Wu, Z. M. Wang, L. X. Niu, C. Wang, M. C. Sun and H. Y. Hu, J. Environ. Manage., 2015, 162, 326–333.
31. W. Liu, X. Jiang and X. Chen, J. Solid State Chem., 2015, 229, 342–349.
32. K. Ki Woong, S. Wooseok, J. Min Wook, A. K. Min, K. Soon Yeol, M. Sung, L. Jongsun, L. Sun Sook and A. Ki-Seok, Carbon, 2015, 82, 96–102.
33. N. T. Li, S. C. Tang and X. K. Meng, J. Mater. Sci. Technol., 2014, 30, 1071–1077.
34. Y. C. Wan, H. F. Teoh, E. S. Tok and C. H. Sow, J. Appl. Phys., 2015, 1175.
35. H. Wang, X. Quan, H. T. Yu and S. Chen, Carbon, 2008, 46, 1126–1132.
36. Y. H. Wu, P. W. Qiao, T. C. Chong and Z. X. Shen, Adv. Mater., 2002, 14, 64–67.
37. J. J. Wang, M. Y. Zhu, R. A. Outlaw, X. Zhao, D. M. Manos and B. C. Holloway, Carbon, 2004, 42, 2867–2872.
38. Y. H. Li, S. G. Wang, J. Q. Wei, X. F. Zhang, C. L. Xu, Z. K. Luan, D. H. Wu and B. Q. Wei, Chem. Phys. Lett., 2002, 357, 263–266.
39. T. Premkumar, K. Lee and K. E. Geckeler, Nanoscale, 2011, 3, 1482–1484.
40. J. Zeng, Q. Zhang, J. Chen and Y. Xia, Nano Lett., 2010, 10, 30–35.
41. S. Wunder, Y. Lu, M. Albrecht and M. Ballauff, ACS Catal., 2011, 1, 908–916.
42. K. Kuroda, T. Ishida and M. Haruta, J. Mol. Catal. A: Chem., 2009, 298, 7–11.

---

Footnote

† Qi Zhang and Xinfei Fan contributed equally to this work.

---