Effects of tailoring and dehydrated cross-linking on morphology evolution of ordered mesoporous carbons

Abstract

In this study, using a template (Pluronic P123) as an in situ carbon source, ordered mesoporous carbons (OMC) were successfully synthesized through directing carbonization of templates. Two key factors in the morphology evolution of mesoporous carbon materials were introduced, one is the tailoring role of micelles by ethanediol, and the other is the dehydrated cross-linking effect by sulfuric acid. Through the two factors, the morphology of as-synthesized ordered mesoporous carbons can be altered from bulks to rods and even to sheets. Transmission electron microscopy (TEM) showed ordered mesoporous carbon rods (OMCR) and ordered mesoporous carbons sheets (OMCS) possessed regular morphology and good mesopore structure. Nitrogen adsorption–desorption analysis confirmed that both OMCR and OMCS had high surface area and uniform mesopore distribution, OMCR had the highest surface area (919 m² g⁻¹) and OMCS had the biggest pore size (9.11 nm). This study also showed that OMCR and OMCS were good carbon carriers; the size of Pt loading nanoparticles can be decreased to 3 nm and Pt/OMCS especially presents very high Pt dispersion and uniformly smallest Pt nanoparticles (2.49 nm).

---

1. Introduction

Recently, researchers have found that ordered mesoporous carbon (OMC) materials have great advantages in the fields of fuel cells, lithium batteries, hydrogen storage, adsorption, catalytic reaction, and electrochemistry.^1–8 Thus, the designed synthesis of OMC materials with controlled surface properties and structural ordering is important from both a fundamental and an application point of view.⁹

There are two typical methods for the synthesis of OMC, one is the nanocasting strategy from mesoporous silica hard templates and the other is the direct synthesis from block copolymer soft templates. Typical representatives of the two synthetic methods are as follows. For the nanocasting strategy, Ryoo et al.¹⁰ first reported an ordered mesoporous carbon, CMK-1, which was fabricated by the preformed mesoporous silica MCM-48 serving as a hard template, and then filling the carbon precursor such as sucrose into the mesopore system (MCM-48), followed by pyrolysis, carbonization, and template removal to get CMK-1. For the soft template method, Zhao's group reported an organic–organic self-assembly approach. Using this method, highly ordered mesoporous carbons, denoted as FDU-15 and FDU-16, were synthesized through a solvent evaporation induced self-assembly (EISA) method using amphiphilic triblock copolymers (PEO–PPO–PEO) as templates and a soluble low-molecular weight compound, namely, cresol or phenol, and formaldehyde as the precursor.¹¹ Findings from the two approaches indicate that the OMC material obtained from the soft template method is a positive phase structure and the nanocasting method leads to a relatively precise negative replica of the hard template. Compared with the soft template method, in syntheses using the nanocasting method, it is easy to control the morphology and mesopore structure of resultant OMC due to the fixed template structure.

However, using mesoporous silica as a scaffold, it makes the process expensive, complicated, and time-consuming, and consequently unsuitable for large-scale production and industrial applications.⁹ Based on these disadvantages, some more cost-effective methods were also invented by making use of synthesized mesoporous silica/surfactant mesophases as starting materials, followed by the introduction of extra carbon precursors.^12–16 For example, Moriguchi et al.¹⁴ reported a method for the synthesis of OMC materials with 2 nm wormhole-like mesopores by in situ polymerization of divinylbenzene in the hydrophobic phase of an hexagonally-arrayed micelle/silicated nanocomposite. Li et al.¹⁵ reported that by post-treatment of the synthesized P123/silica composite with sulfuric acid (a catalyst for cross-linking of P123), the yield can be improved by employing sucrose as a supplementary carbon source. However, the surface area of the OMC was only 610 m² g⁻¹ and the pore size was also less than 3 nm. Lee et al.¹⁶ also reported that uniform mesoporous carbons can be directly obtained from carbonization of P123/phenol-resin/silica nanocomposites. However, the OMC products obtained exhibited disordered and wormhole-like pore structure. Therefore, the synthesis conditions and compositions still needed to be optimized to obtain OMC with ordered mesopore structure, high surface area, large pore size, and special morphology.

In the present study, we report a low-cost, simple, convenient and green approach for the synthesis of ordered mesoporous carbon materials from the SiO₂/P123/sucrose composite. By investigating the effects of synthesis conditions and compositions, we found some interesting results that the morphology and mesopore size of obtained OMC products can be altered by the co-solvent and co-template addition, and the pre-treatment by sulfuric acid. The morphology of as-synthesized ordered mesoporous carbons can be altered from bulks (OMCB) to rods (OMCR), and even to sheets (OMCS). Moreover, the mesopore size can also be increased to 5.45 nm and even to 9.11 nm.

2. Experimental section

2.1 Preparation of materials

2.1.1 Preparation of SBA-15-DC-OMC. SBA-15 directly converted ordered mesoporous carbons (SBA-15-DC-OMC) were obtained from SBA-15. Typically, SBA-15 was prepared using the tri-block copolymer, EO₂₀PO₇₀EO₂₀ (Pluronic P123, BASF), and tetraethyl orthosilicate (TEOS, 99%).¹⁷ In our previous synthesis of SBA-15,¹⁸ 4.0 g of Pluronic P123 was dissolved in 126 mL of deionized (DI) water and 20 mL of hydrochloric acid (37 wt%), and then 10 mL of TEOS was added and stirred for 20 h at 35 °C. The slurry was hydrothermally treated at 100 °C for 24 h. Subsequently, the product was filtered, dried and the white powder was pyrolysed at 850 °C for 2 h in nitrogen (99.999%). Finally, the silica template was removed in NaOH (10 M) at 120 °C for 4 h, followed by washing with DI water, and the final black powder named SBA-15-DC-OMC.

2.1.2 Preparation of OMCB. Ordered mesoporous carbons bulks (OMCB) were prepared using sucrose as supplementary carbon source. In this synthesis, 4.0 g of Pluronic P123 was dissolved in 160 mL of sulfuric acid (1 M), stirring until the P123 dissolved. Then, 10 mL of TEOS was added and stirred for 20 h at 35 °C. The slurry was hydrothermally treated at 100 °C for 24 h. Subsequently, the white SiO₂/P123 composites were filtered and dried at 100 °C for 6 h in air dry oven. After drying, the colour of SiO₂/P123 powder becomes brown, which is followed by washing with DI water to pH = 7. To encapsulate supplementary carbon into the channel gap, the brown powders were soaked in a crucible with a given solution for 30 min. The solution was composed of 3.0 g of sucrose and 10 mL of DI water. Then, the crucible was placed into an air drying oven to first remove water at 100 °C for 6 h followed by pre-carbonization at 160 °C for 6 h. The products were then pyrolysed at 850 °C for 2 h in nitrogen (99.999%). After silica removal, the final powders were named OMCB.

2.1.3 Preparation of OMCR. Ordered mesoporous carbons rods (OMCR) were prepared using ethanediol as co-solvent and sucrose as supplementary carbon source. The procedure differs by addition of 30 mL of ethanediol before TEOS addition, the remaining process being the same as the synthesis of OMCB. After silica removal, the final powder was named OMCR.

2.1.4 Preparation of OMCS. Ordered mesoporous carbons sheets (OMCS) were prepared using P123 and sucrose as co-template and co-carbon source. In this synthesis, 4.0 g of Pluronic P123 and 1.0 g sucrose were dissolved in 126 mL of DI water and 20 mL of hydrochloric acid (37 wt%). After complete dissolution, 30 mL of ethanediol was poured into the abovementioned solution and stirred for 2 h at 35 °C; subsequently, 10 mL of TEOS was added with further stirring for 20 h at 35 °C. The slurry was hydrothermally treated at 100 °C for 24 h. Subsequently, the SiO₂/P123/sucrose composites were filtered, washed and dried. For the pre-carbonization process, 1.0 g of SiO₂/P123/sucrose mixture was dispersed in 10 mL of DI water and 1 mL of sulfuric acid (98 wt%) added with stirring for 12 h at room temperature, and then heated at 160 °C for 6 h. After pyrolysis at 850 °C for 2 h and silica removal, the final powder was named OMCS.

2.1.5 Loading Pt nanoparticles on OMC materials. Pt/OMC materials were prepared by an organic colloid method and the detailed preparation method has been described in a previous studies.¹⁹ Briefly, hexachloroplatinic acid and sodium citrate were dissolved in ethylene glycol (EG) to form a mixture and stirred for 30 min. Synthesized OMC was then added to the mixture under stirring. The pH of the solution was adjusted to >10 by dropwise addition of 5 wt% KOH in EG solution. The mixture was then placed in a Teflon®-lined autoclave and conditioned at 120 °C for 6 h, followed by filtering, washing and vacuum drying at 60 °C.

2.2 Characterization

Transmission electron microscopy (TEM) images were obtained on a JEM-2100 transmission electron microscope (JEOL, Japan). X-ray diffraction (XRD) patterns were obtained using a TD-3500 powder diffractometer (Tongda, China). Specific surface areas and pore size distributions were determined using Brunauer–Emmett–Teller (BET) nitrogen adsorption–desorption on a Tristar II 3020 gas adsorption analyzer (Micromeritics, USA). Raman measurement was performed on a LabRAM Aramis Raman spectrometer (HJY, France) with a laser wavelength of 632 nm.

2.3 Electrochemical testing

Electrochemical measurements were conducted in a standard three-electrode glass cell on an electrochemical workstation (Ivium, Netherlands) at room temperature, coupled with a rotating disk electrode (RDE) system (Pine Research Instrumentation, USA). A glassy carbon electrode (GCE, diameter of 5 mm and an electrode area of 0.1964 cm²) was used as the working electrode substrate, with an Ag/AgCl/KCl (3 M) electrode and a Pt wire as the reference and counter electrode, respectively. For simplicity, the Ag/AgCl/KCl (3 M) reference electrode is hereafter abbreviated to Ag/AgCl. Before every measurement, the GCE surface was cleaned by ultrasonication in ethanol and polished with an α-Al₂O₃ slurry (50 nm) on a microcloth, followed by rinsing with DI water and drying under an infrared lamp.

A slurry of the active material was prepared by mixing 5.0 mg Pt/OMCS catalyst with 1 mL of an ethanol solution containing Nafion (0.25 wt%) under ultrasonication. Furthermore, 5 μL catalyst slurry was pipetted onto the surface of the GCE, followed by drying under an infrared lamp to form a catalyst film on the GCE substrate. The catalyst Pt loading was approximately 25 μg Pt cm⁻².

ORR measurements were carried out in O₂-saturated 0.1 M HClO₄ electrolyte, using a rotation rate of 1600 rpm and a scan rate of 10 mV s⁻¹. Before the ORR activity test, all the electrodes were pretreated by cycling the potential between 0 and 1.2 V at a sweep rate of 50 mV s⁻¹ for 20 cycles to remove any surface contaminants. For comparison, a commercial Pt/C electrode (20 wt% Pt, Johnson Matthey, UK) was also tested under the same conditions.

3. Results and discussion

3.1 Effect of sucrose addition on directing carbonization

Fig. 1 shows the TEM images of SBA-15 mesoporous silica and different mesoporous carbon materials. As shown in Fig. 1A and B, the typical SBA-15 exhibits bulky morphology with very ordered mesopore structure. When P123 template was not removed, SBA-15-DC-OMC was gained through direct carbonization of P123, the morphology still remained bulky without ordered shape, as shown in Fig. 1C. Unfortunately, only a small quantity of SBA-15-DC-OMC can be collected after silica removal, and also the ordered mesopore structure disappeared and the carbon materials exhibited a high degree of graphite with disordered mesopores, as shown in Fig. 1D. We speculate that the amount of P123 as a single in situ carbon source is not sufficient for OMC formation; it may be due to the cold shrink of P123 molecules during filtering and washing causing the formation of the disordered mesopore structure. Thus, it is crucial to fill another carbon source into the vacant channels. Sucrose, an abundant and low-cost species, is often utilized as a carbon source to synthesize mesoporous carbon. When sucrose is used as supplementary carbon source, OMCB materials exhibit better mesopore structure than SBA-15-DC-OMC and the white pore stripes are faintly visible, as shown in Fig. 1F. This demonstrated that a more regular pore structure of mesoporous carbon materials could be achieved through adding a supplementary carbon source.

Fig. 1 (A and B) TEM images of SBA-15 with bulky morphology; (C and D) SBA-15-DC-OMC; (E and F) OMCB prepared with sucrose as supplementary carbon source.

3.2 Effect of tailoring of micelles by ethanediol

Based on typical synthesis of SBA-15, we can successfully prepare mesoporous carbon materials using P123 as a structure-directing agent and in situ carbon source, but the shape of the synthesized particles is large and bulky. To solve this problem, we tried to use ethanediol as co-solvent to tailor the micelles of P123. From Fig. 2A, it can be obviously observed that the particles of SBA-15 mesoporous silica become rods from bulks through use of ethanediol as a co-solvent; moreover, ordered mesopore structure in the (110) direction can be clearly observed in Fig. 2B and C. Compared with SBA-15 rods, the morphology of OMCB is basically copied from SBA-15 rods precursor, as shown in Fig. 2D. Remarkably, the mesopore channel seems to be piled up by massive carbon nanowires, as shown in Fig. 2E and F. It shows well-ordered hexagonal arrays of carbon nanorods and further confirms that the OMCR sample has a 2D hexagonal (p6mm) meso-structure.²⁰ Therefore, we can observe that the tailoring effect of ethanediol is very crucial to the formation of OMCR materials. The possible formation process of OMCR is shown in Scheme 1. First, the molecules of P123 self-assemble in acidic solution and then regular ordered micelles with rod-like structure are assembled by adding ethanediol as co-solvent. After hydrothermal treatment, the white SiO₂/P123 composites should be filtered and dried. In the process of drying, the sulfuric acid in original solution gradually becomes concentrated sulfuric acid, which makes further dehydrated cross-linking in the P123. The white gaps are formed around P123 molecules after dehydrated cross-linking, so the supplementary carbon of sucrose will be introduced to fill the white gaps. Subsequently, SiO₂/P123/sucrose composites are pyrolysed at 850 °C for carbonization. Finally, ordered mesoporous carbon rods can be obtained after silica removal.

Fig. 2 (A–C) TEM images of SBA-15 with rod morphology; (D–F) OMCR.

Scheme 1 Schematic of OMCR formation process.

3.3 Effect of dehydrated cross-linking by sulfuric acid

From the abovementioned TEM results of OMCB and OMCR, we found that ordered mesoporous carbon materials with uniform mesopore structure could be prepared through using P123 as a single template and using sucrose as supplementary carbon source. However, question remains whether sucrose can be added in advance, and whether sucrose can be used as a co-template and a co-carbon source with P123 together, are remaining questions. To explore this problem, we carried out experiments with P123 and sucrose as co-template. To avoid the dehydration of sucrose in the concentrated sulfuric acid environment, we used HCl as the acid medium, ethanediol as co-solvent, and SiO₂/P123/sucrose composites were obtained by hydrothermal crystallization. In the next process of sulfuric acid treatment, we found a very interesting phenomenon. When the composites were soaked in concentrated sulfuric acid for dehydrated cross-linking of the carbon sources, the brown filtered sample was placed directly in the drying oven at 160 °C for pre-carbonization if not washed. The structure of the resultant sample (OMCS) obtained was shown in Fig. 3. The OMCS exhibits a thin sheet structure and the ordered mesopore structure in the (110) direction can be clearly observed in the sheet, as shown in Fig. 3B. If washed completely, the product exhibits a long rod structure with large defects and not a sheet structure, as shown in Fig. 4. Results show that the presence of sulfuric acid is very important to the formation of sheet carbon materials. The possible formation process is shown in Scheme 2. When P123 and sucrose are used as co-templates, composites of SiO₂/P123/sucrose are formed after the hydrolysis of TEOS. During the pretreatment of concentrated sulfuric acid, dehydrated cross-linking and pre-carbonization will happen between the molecules of P123 and sucrose, which results in the formation of a hollow carbon tube within the area of the white dots. Lastly, through NaOH etching, the hollow carbon tube breaks off with the removal of SiO₂, and then many ordered mesoporous carbon sheets will be formed, as shown in Scheme 2.

Fig. 3 TEM images of OMCS ​in different magnification (A–D).

Fig. 4 TEM images of OMC without pretreatment by sulfuric acid ​in different magnification (A and B).

Scheme 2 The possible formation process of OMCS.

3.4 Physical properties of OMC

Fig. 5 presents the nitrogen adsorption–desorption isotherms and pore size distribution (by the BJH method as inset figures) of different mesoporous carbon materials. OMCR exhibits the highest surface area of 919 m² g⁻¹, followed by OMCB (710 m² g⁻¹), OMCS (442 m² g⁻¹), and the lowest surface area of SBA-15-DC-OMC (318 m² g⁻¹) (Table 1). The total pore volume of OMCR (1.15 cm³ g⁻¹) is larger than that of OMCS (0.56 cm³ g⁻¹); this may be due to OMCR possessing more pore structure than the two-dimensional structure of OMCS. However, the mean pore size of OMCS (9.11 nm) is bigger than that of OMCR (5.45 nm) and this result is also consistent with the results of previous TEM analysis. In addition, we found that OMCS has a multi-level pore channel, mainly concentrated in 5 nm and 14 nm pores, as shown in Fig. 5D. This may be attributed to the dehydrated or peeled effect of carbon materials in the presence of concentrated sulfuric acid.

Fig. 5 Nitrogen adsorption–desorption isotherms (inset: pore size distribution of corresponding samples): (A) SBA-15-DC-OMC; (B) OMCB; (C) OMCR; and (D) OMCS.

Table 1 Calculated porosity parameters of all samples

Samples	SBET/m^2 g^−1	VBJH/cm^3 g^−1	Pore size/nm
SBA-15 (ref. 21)	647	0.87	7.90
SBA-15-DC-OMC	318	0.19	2.69
OMCB	710	1.14	4.37
OMCR	919	1.15	5.45
OMCS	442	0.56	9.11

---

The low angle range XRD patterns of all synthesized mesoporous carbon materials are shown in Fig. 6A. It is observed that the as-synthesized OMCR exhibited three peaks, which can be assigned to diffraction from the (100), (110), and (200) planes of SBA-15,¹⁷ indicating the presence of a long-range ordered structure. However, we can only observe a very weak diffraction peak corresponding to the (110) plane of OMCS. We guess that may be due to the relatively disordered structure in two-dimensional OMCS. This result is also consistent with the observation in previous TEM analysis. The high angle patterns of all samples are presented in Fig. 6B. At 23.6° and 43.5°, two typical diffraction peaks are detected, assigned to C (002) and C (101) planes, respectively.

Fig. 6 XRD patterns of all synthesized mesoporous carbon materials. (A) Low angle diffraction; (B) high angle diffraction.

Fig. 7 shows the Raman spectra of different samples. Peaks at ∼1315 and ∼1588 cm⁻¹ are assigned to the D and G bands of carbon, respectively. In general, the G band arises from the bond stretching of sp² bonded pairs, including C–C, whereas the D band corresponds to sp³ defect sites, including vacancies and hetero atoms. The ratio of the D and G band intensities (I_D/I_G) is commonly considered as an indicator of the extent of defects in carbon materials. As shown in Fig. 7, the peak intensity ratios of the D and G bands (I_D/I_G) are 1.16, 1.42, and 1.27 for OMCB, OMCR, and OMCS, respectively. Although there is no significant change in the positions of D and G bands, OMCS shows a lower I_D/I_G value than OMCR (1.27 vs. 1.42), possibly caused by the two-dimensional structure of OMCS.

Fig. 7 Raman spectra of different samples.

3.5 Investigation of Pt loaded on OMC materials

As we know, porosity is highly desirable to achieve excellent catalytic performance because it facilitates high mass transfer flux and active loading.²² Therefore, porous carbon nanostructures are most suitable for catalyst preparation. We carried out some trials to investigate whether the morphology or mesopore structure of ordered mesoporous carbons will influence loading of metal nanoparticles. When we used our synthesized mesoporous carbons as carbon carriers and Pt as the loading metal, high dispersion Pt-loaded samples were prepared. In the results shown in Fig. 8, Pt/OMCB exhibits a slight aggregation (Fig. 8B) and the particle size of Pt is centered at 6.21 nm (Fig. 8C). Compared with Pt/OMCB, Pt/OMCR shows better Pt dispersion (Fig. 8E), and its particle size is centered at 2.94 nm (Fig. 8F). Satisfyingly, Pt/OMCS presents very high dispersion on the carbon sheets and uniform Pt nanoparticles decreased in size to 2.48 nm (Fig. 8G). It can be observed that the morphology of the carbon carrier has an important influence on the metal loading. With its high ductility, the two-dimensional sheet structure can be observed as an ideal carrier material.

Fig. 8 TEM images of mesoporous carbon materials loaded with 20 wt% Pt and their corresponding particle size of distribution. (A–C) Pt/OMCB; (D–F) Pt/OMCR; (H–G) Pt/OMCS.

To investigate the catalytic performance of Pt loaded on OMC materials, we made a preliminary study of ORR electrochemical activity. We compared our best Pt dispersion sample, 20% Pt/OMCS catalyst, with commercial Pt/C using almost the same Pt loading. Fig. 9 shows the polarization curves of JM 20% Pt/C and 20% Pt/OMCS in O₂-saturated 0.1 M HClO₄ solution, using a rotating disk electrode (RDE) at 1600 rpm and a scan rate of 10 mV s⁻¹, with the current density normalized to the geometric surface area of the electrodes. As shown in Fig. 9, our 20% Pt/OMCS catalyst exhibited higher ORR catalytic activities than JM 20% Pt/C. With almost the same Pt loading, the half-wave potential of 20% Pt/OMCS catalyst surpassed 20 mV, higher than that of the Pt/C catalyst, and the mass activity of 20% Pt/OMCS catalyst is almost 1.5 times higher than that of the JM 20% Pt/C. This demonstrated that our prepared Pt-loaded catalyst has reached the level of a commercial Pt/C catalyst, but there is also a need to optimize the catalyst through detailed investigation in our next study.

Fig. 9 ORR performances of 20% Pt/OMCS and JM 20% Pt/C catalysts and their mass activity (inset figure).

4. Conclusions

In summary, two studies of ordered mesoporous carbons with regular structures (rods and sheets) were synthesized through facile direct carbonization of templates. Using P123 as template, sucrose as supplementary carbon source or co-template, ethanediol as tailoring agent and sulfuric acid as dehydrated cross-linking agent, OMCR and OMCS were successfully synthesized. Both possess high surface area and uniform mesopore structure, the surface area of OMCR is high, up to 919 m² g⁻¹; moreover, OMCS also possesses high surface area relative to similar layer structures of mesoporous carbon materials. This study has confirmed that both OMCR and OMCS are good carbon carriers; the size of loading Pt nanoparticles can be decreased to 3 nm, Pt/OMCS especially presents very high Pt dispersion and the smallest, uniform Pt nanoparticles. Therefore, it can be observed that OMCR and OMCS will be potential candidates for carbon carriers to fabricate noble metal or non-noble metal electrochemical catalysts. More detailed study on this catalysis application will be reported by our group in the near future.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (NSFC Project no. 51302091, 51102099, 11132004), Science and Technology Planning Project of Guangdong Province, China (2013B022000035), and the Fundamental Research Funds for the Central Universities, SCUT.

References

1. L. Li, H. Song and X. Chen, Electrochim. Acta, 2006, 51, 5715–5720.
2. W. Ding, L. Li, K. Xiong, Y. Wang, W. Li, Y. Nie, S. G. Chen, X. Q. Qi and Z. D. Wei, J. Am. Chem. Soc., 2015, 137, 5414–5420.
3. J. Galo, A. A. de, G. J. Soler-Illia, C. Sanchez, B. Lebeau and J. Patarin, Chem. Rev., 2002, 102, 4093–4138.
4. A. Taguchi and F. Schüth, Microporous Mesoporous Mater., 2005, 77, 1–45.
5. Z. Guo, G. Zhu, B. Gao, D. Zhang, G. Tian and Y. Chen, Carbon, 2005, 43, 2344–2351.
6. J. Lee, J. Kim and T. Hyeon, Adv. Mater., 2006, 18, 2073–2094.
7. C. H. You, S. J. Liao, X. C. Qiao, X. Y. Zeng, F. F. Liu, R. P. Rui, H. Y. Song, J. H. Zeng and Y. W. Li, J. Mater. Chem. A, 2014, 2, 12240–12246.
8. A. H. Lu and F. Schüth, Adv. Mater., 2006, 18, 1793–1805.
9. T. Y. Ma, L. Liu and Z. Y. Yuan, Chem. Soc. Rev., 2013, 42, 3977–4003.
10. R. Ryoo, S. H. Joo and S. J. Jun, J. Phys. Chem. B, 1999, 103, 7743–7746.
11. Y. Meng, D. Gu, F. Zhang, Y. Shi, H. Yang, Z. Li, C. Yu, B. Tu and D. Zhao, Angew. Chem., Int. Ed., 2005, 44, 7053–7509.
12. S. B. Yoon, J. Y. Kim and J. Yu, Chem. Commun., 2002, 14, 1536–1537.
13. J. Pang, V. T. John, D. A. Loy, Z. Yang and Y. Lu, Adv. Mater., 2005, 17, 704–707.
14. I. Moriguchi, Y. Koga, R. Matsukura, Y. Teraoka and Y. M. Kodama, Chem. Commun., 2002, 17, 1844–1845.
15. L. Li, H. Song and X. Chen, Microporous Mesoporous Mater., 2006, 94, 9–14.
16. J. Lee, J. Kim, S. Yoon, S. M. Oh and Y. Hyeon, Chem. Mater., 2004, 16, 3323–3330.
17. D. Y. Zhao, J. L. Feng, Q. S. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548–552.
18. K. Wan, G. F. Long, M. Y. Liu, L. Du, Z. X. Liang and P. Tsiakaras, Appl. Catal., B, 2015, 165, 566–571.
19. Y. N. Wu, S. J. Liao, Z. X. Liang, L. J. Yang and R. F. Wang, J. Power Sources, 2009, 194, 805–810.
20. C. C. Ting, H. T. Wu, S. Vetrivel, D. Saikia, Y. C. Pan, G. T. K. Fey and H. M. Kao, Microporous Mesoporous Mater., 2010, 128, 1–11.
21. J. X. Zhang, G. W. Huang, C. Zhang, Q. H. He, C. Huang, X. Yang, H. Y. Song, Z. X. Liang, L. Du and S. J. Liao, Chin. J. Catal., 2013, 34, 1519–1526.
22. R. Wang, T. Zhou, H. Li, H. Wang, H. Feng, J. Goh and S. Ji, J. Power Sources, 2014, 261, 238–244.

---