Simultaneous fabrication of bifunctional Cu(I)/Ce(IV) sites in silica nanopores using a guests-redox strategy

Abstract

Fabrication of bifunctional Cu(I)/Ce(IV) sites in mesoporous silica SBA-15 requires a series of complicated steps, including introduction of the Ce(III) precursor, calcination to generate Ce(IV), further introduction of the Cu(II) precursor, and repeated calcination to generate Cu(I). This traditional method has low efficiency and wastes energy. Here we provide a highly efficient, convenient and green strategy for the fabrication of bifunctional Cu(I)/Ce(IV) sites in mesoporous silica SBA-15, for the first time. The precursors CuCl₂ and CeCl₃ were simultaneously introduced into SBA-15. Based on a guests-redox strategy, both the reduction of CuCl₂ to CuCl and the oxidation of CeCl₃ to CeO₂ could be realized in one calcination step. This strategy avoids the repeated modification process, and guarantees a high Cu(I) yield of ∼59% without generation of toxic gas. In addition, the resultant materials show excellent performance in gas separation of C₂H₄ from C₂H₆.

---

1. Introduction

Because of their versatility, nontoxicity, and low cost, Cu(I) modified porous materials are highly promising for a variety of applications, such as gas separation of olefins/paraffins,^1–3 deep desulfurization/denitrogenation of transportation fuels,^4–7 and catalytic synthesis of dimethyl carbonate.⁸ CuCl is one of the most widely used Cu(I) sources. In general, there are two predominant methods to prepare CuCl functionalized porous materials (denoted CuCl/PM), that is, (i) direct introduction of CuCl and (ii) introduction of CuCl₂ followed by selective reduction. Taking into consideration that CuCl₂ is more chemically stable and cheaper than CuCl commercially, the use of CuCl₂ as a precursor to produce CuCl/PM materials could simplify the preparation process and reduce the cost. For the second method, i.e. the introduction of CuCl₂ followed by selective reduction, the precursor CuCl₂ can decompose to CuCl and Cl₂ gas at a temperature higher than 450 °C under inert atmosphere.^9,10 However, the amount of CuCl₂ converted to CuCl is quite low, and the harmful byproduct Cl₂ is simultaneously formed in the reduction procedure. Hence, the development of an efficient, green method for reduction of CuCl₂ to CuCl is extremely desirable.

It is reported that Ce(IV) sites can disperse into various geometric positions and utilize the surface of the support effectively.^11–13 Moreover, Ce(IV) may be beneficial to the stability and degree of dispersion of the active metal sites. As a result, Ce(IV) sites are widely used as assistant agents in bifunctional materials.^14,15 In general, Ce(III) salts used as precursors are first introduced into the support and calcined, leading to the generation of a host with modified Ce(IV) sites. Another metallic precursor such as a Cu(II) salt is then introduced, followed by the second calcination to generate Cu(I) sites. Evidently, this traditional approach to the fabrication of bifunctional Cu(I)/Ce(IV) sites on the support is rather complicated. Therefore, it is highly desirable to develop a facile approach to generate bifunctional Cu(I)/Ce(IV) sites on the support.

Herein we report for the first time a strategy for simultaneously constructing bifunctional Cu(I)/Ce(IV) sites on SBA-15 mesoporous silica by using a guests-redox interaction (Scheme 1). The precursors, both CuCl₂ and CeCl₃, were introduced into SBA-15 in a solid-state grinding step, followed by calcination at 350 °C under argon atmosphere. Thus, a bifunctional material, named Cu–Ce/SBA-15, was obtained. During the calcination process, CeCl₃ worked as a reducing agent and was oxidized to CeO₂. Meanwhile, CuCl₂ worked as an oxidant and was reduced to CuCl. Besides the Cu and Ce species, the other elements in the samples are unable to take part in the redox reaction. In particular, the Si element derived from the support has no redox capability. This strategy allows the modification of SBA-15 with bifunctional Cu(I)/Ce(IV) sites in one step, which avoids repeated calcination. More importantly, the Cu(I) yield is as high as ∼59% without any harmful Cl₂ generated. The present strategy offers an efficient, energy-saving and green way to fabricate supported Cu(I)/Ce(IV) sites on SBA-15. We also demonstrate that the bifunctional materials exhibit excellent performance in the gas separation of C₂H₄ from C₂H₆. Both the capacity and the selectivity are superior to the performance of materials modified with only Cu(I) sites.

Scheme 1 Promoting the conversion yield of CuCl₂ to CuCl from ∼15% to ∼59% in silica nanopores using the guests-redox strategy.

2. Experimental

2.1 Materials synthesis

SBA-15. The support, mesoporous silica SBA-15, was synthesized according to the reported method.¹⁶ In a typical synthesis, 2 g of pluronic P123 was dissolved in 75 g of aqueous HCl solution (1.6 M). Then, 4.25 g of the silica source tetraethylorthosilicate (TEOS) was added and stirred at 40 °C for 24 h, followed by hydrothermal treatment at 100 °C for 24 h. The as-synthesized sample was recovered by filtration and dried at ambient conditions. After calcination in flowing air at 550 °C for 5 h, pure SBA-15 was obtained.

xCu/SBA-15, yCe/SBA-15 and xCu–yCe/SBA-15. The salt precursors, CuCl₂·2H₂O and/or CeCl₃·7H₂O, were introduced to SBA-15 by solid-state grinding at ambient conditions for 30 min.¹⁷ The mixture was then calcined in flowing argon at 350 °C for 5 h with a heating rate of 2 °C min⁻¹. The obtained samples were denoted xCu–yCe/SBA-15. The content of copper (cerium) introduced was x (y) mmol per gram of SBA-15.

2.2 Materials characterization

X-ray diffraction (XRD) patterns of the materials were recorded on a Bruker D8 Advance diffractometer with Cu Kα radiation in the 2θ ranges from 0.7° to 5° and 5° to 80° at 40 kV and 40 mA.

The N₂ adsorption–desorption isotherms were measured using an ASAP 2020 system at −196 °C. Prior to analysis, the samples were evacuated at 150 °C for 4 h. The Brunauer–Emmett–Teller (BET) surface area was calculated as the relative pressure ranged from 0.04 to 0.20. The total pore volume was derived from the amount adsorbed at the relative pressure of about 0.99. The pore size distributions were calculated by the Barrett–Joyner–Halenda (BJH) method according to the adsorption branches.

X-ray photoelectron spectroscopy (XPS) analysis was conducted on a Physical Electronic PHI-550 spectrometer equipped with an Al Kα X-ray source (hν = 1486.6 eV) and was operated at 10 kV and 35 mA.

TG analysis was performed in an argon flow from room temperature to 800 °C on a thermobalance (STA-490C, NETZSCH).

2.3 Gas adsorption test

All of the gas adsorption isotherms were measured using an intelligent gravimetric analyzer (IGA-100) supplied by Hiden Analytical, Ltd. The microbalance had a long-term stability of ±1 μg with a weighing resolution of 0.2 μg. The sample (<100 ± 1 mg) was outgassed until it reached a constant weight, at a pressure of <10⁻⁶ Pa and a temperature of 150 °C, prior to measurement. The pressure was then increased gradually, over a period of 60 s, to the desired pressure. The adsorption isotherms were recorded at 25 °C.

3. Results and discussion

3.1 Characterization of the modified materials

Fig. 1A shows the low-angle XRD patterns of the samples. The diffraction lines indexed as (100), (110) and (200) can be identified, except for 8Cu–8Ce/SBA-15. The results indicate that the ordered mesostructure could be well maintained by controlling the content of the metal sites within the limitation of 4 mmol g⁻¹ (SBA-15).

Fig. 1 (A) Low-angle and (B) wide-angle patterns of SBA-15 and modified samples.

Excluding 8Cu–8Ce/SBA-15, the N₂ sorption isotherms of the samples are clearly classified as type IV, with an H1 hysteresis loop, which are characteristic of materials with cylindrical mesopores (Fig. 2A). Fig. 2B shows the pore size distributions of the samples. The corresponding surface areas, pore volumes and pore diameters are shown in Table 1. The pore diameter of SBA-15 is 9.2 nm. With the introduction of bifunctional sites, the pore diameters increase slightly to 9.9 nm, then return to 9.2 nm with the increase of the Cu/Ce content. For the samples of 4Cu/SBA-15 and 4Ce/SBA-15, the pore diameters are 8.4 nm. In addition, for the bifunctional samples, the surface areas and pore volumes decrease gradually, from 500 to 142 m² g⁻¹ and from 0.891 to 0.259 cm³ g⁻¹, respectively, with the increase of the Cu/Ce content. The surface area is 360 m² g⁻¹ for the sample of 4Cu/SBA-15, and 250 m² g⁻¹ for the sample of 4Ce/SBA-15. The pore volume is 0.625 cm³ g⁻¹ for the sample of 4Cu/SBA-15, and 0.415 cm³ g⁻¹ for the sample of 4Ce/SBA-15. It is seen that the introduction of metal sites does not destroy the pore characteristics of SBA-15, except for 8Cu–8Ce/SBA-15. The slight changes in the pore characteristics indicate the successful introduction of metal sites.

Fig. 2 (A) N₂ adsorption–desorption isotherms and (B) pore size distributions of SBA-15 and modified samples.

Table 1 Physicochemical properties of different samples

Sample	SBET (m^2 g^−1)	Vp (cm^3 g^−1)	Dp (nm)
SBA-15	658	1.093	9.2
1Cu–1Ce/SBA-15	500	0.891	9.9
2Cu–2Ce/SBA-15	456	0.760	9.9
4Cu–4Ce/SBA-15	310	0.523	9.9
8Cu–8Ce/SBA-15	142	0.259	9.2
4Cu/SBA-15	360	0.625	8.4
4Ce/SBA-15	250	0.415	8.4

---

For the sample of 4Cu–4Ce/SBA-15, periodic mesopores with uniform pore size and wall thickness can be deduced from the white-dark contrast of the TEM image (Fig. S1†). This also confirms that the ordered structure of mesoporous silica is well preserved after the introduction of bimetal sites. Fig. S2 and S3† show the SEM image and EDX spectrum of the sample 4Cu–4Ce/SBA-15. As shown in the EDX spectrum, Cu and Ce elements are both present in the silica nanopores. Further, judging from the elemental mapping images, Ce and Cu elements are well dispersed throughout the silica nanopores.

The wide-angle XRD patterns of all the samples are presented in Fig. 1B. There are peaks at 28.1°, 33.1°, 47.5° and 56.5° derived from CeO₂ (JCPDS no. 34-0394) appearing for the sample 4Ce/SBA-15. For the sample 4Cu/SBA-15, the peaks at 28.4°, 33.1°, 47.5° and 56.5° are features of CuCl (JCPDS no. 18-0439), and the peaks at 32.4° and 39.6° are characteristic of CuCl₂·3Cu(OH)₂ (JCPDS no. 34-0394). This indicates that a large amount of Cu(II) sites have not been successfully reduced. When it comes to the bifunctional sample 4Cu–4Ce/SBA-15, the patterns are completely different. Crystalline phases of CeO₂ and CuCl with the same peak positions coexist. In addition, there is no peak characteristic of Cu(II) sites. This implies that, with the help of the reducing agent, the reduction process from Cu(II) to Cu(I) sites was carried out more thoroughly. In addition, the metal content has an important effect on the guests-redox strategy. For the sample 8Cu–8Ce/SBA-15, a series of disordered peaks emerged. The patterns of samples that were simply ground, without any thermal treatment, are shown in Fig. S4† for reference. It was found that the disordered peaks that resulted from the CuCl₂·2H₂O and CeCl₃·6H₂O precursors decomposed incompletely.

The valence states of Cu species in 4Cu/SBA-15 and 4Cu–4Ce/SBA-15 were further investigated by X-ray photoelectron spectroscopy (XPS). Fig. S6† depicts the regions of the Cu high-resolution spectra. The wide peaks ranging from 930 eV to 938 eV are ascribed to Cu 2p_3/2. The accompanying satellite peaks ranging from 940 to 950 eV are characteristic of Cu 2p_1/2. The existence of satellite peaks indicates that the samples probably contain Cu(II) species. Further analysis of the peaks of Cu 2p_2/3 was conducted. As shown in Fig. 3A and B, the results obtained from curve-fitting present two peaks at 932.9 eV and 935.1 eV, which can be respectively ascribed to Cu 2p_3/2⁺ and Cu 2p_3/2²⁺.^18–21 According to further integral calculations, the Cu(I) yield is 15% for the sample 4Cu/SBA-15 (Table 2), and 59% for the sample 4Cu–4Ce/SBA-15. In order to confirm these vital results, titration experiments were also carried out to measure the Cu(I) yield. The Cu(I) yield is 14% for the sample 4Cu/SBA-15, and 56% for the sample 4Cu–4Ce/SBA-15. It is evident that the results derived from titration correlate well with those from XPS. It is noteworthy that by use of the guests-redox strategy, the Cu(I) yield is dramatically promoted from ∼15% to ∼59%. In addition, the valence states of the Ce species were also measured, and the results are shown in Fig. S5.† As depicted, the Ce(IV) yield is 61% (Table 2) for the sample 4Cu–4Ce/SBA-15, which is close to the Cu(I) yield. This coincidence implies that the guests-redox strategy results in reactions with the same molar ratio.

Fig. 3 XPS peak fitting of Cu 2p_3/2 for the samples (A) 4Cu/SBA-15 and (B) 4Cu–4Ce/SBA-15.

Table 2 Amount of Cu and Ce in different samples

Sample	Amount of Cu^a (mmol g^−1)	Cu(I) content^b (%)		Amount of Ce^e (mmol g^−1)	Ce(IV) content^f (%) from XPS
		From titration^c	From XPS^d
a Amount of Cu introduced per gram of SBA-15.b Calculated from the equation of Cu(I)/[Cu(I) + Cu(II)].c Determined by titration.d Calculated from the XPS results.e Amount of Ce introduced per gram of SBA-15.f Calculated from the XPS results.
4Cu/SBA-15	4	14%	15%	0	—
4Ce/SBA-15	0	—	—	4	—
1Cu–1Ce/SBA-15	1	53%	—	1	—
2Cu–2Ce/SBA-15	2	53%	—	2	—
4Cu–4Ce/SBA-15	4	56%	59%	4	61%
8Cu–8Ce/SBA-15	8	42%	—	8	—

---

3.2 Adsorption performance of the modified materials

The bifunctional materials were also applied to the adsorptive separation of C₂H₄ from C₂H₆. C₂H₄ is an important chemical raw material with various applications. Over the past 70 years, and until now, C₂H₄/C₂H₆ separation has mainly been carried out by cryogenic distillation, a process that is performed at −30 °C and 300 kPa. This represents one of the most energy-intensive processes in the chemical industry.^22,23 Extensive attention has recently been paid to the separation of C₂H₄ and C₂H₆ by π-complexative adsorption, since this process can be performed under mild conditions.^24–27 Among various alternatives, CuCl/PM is a good choice of adsorbent because of its high activity and good reusability. Thus, gas adsorption experiments were conducted to measure the separation performance of the prepared samples, and all of the isotherms were consistent with the double-site Langmuir model (Table S3†). The adsorption isotherms in Fig. 4 show that 4Ce/SBA-15 adsorbed 8.8 mL g⁻¹ of C₂H₄ at 100 kPa, which is slightly higher than the uptake of C₂H₆ (5.9 mL g⁻¹). Under the same conditions, the C₂H₄ and C₂H₆ uptake increased to 16.0 and 9.7 mL g⁻¹ respectively on 4Cu/SBA-15. It is favorable that the uptake of C₂H₄ appreciably increased to 17.1 mL g⁻¹, and the C₂H₆ uptake decreased to 5.7 mL g⁻¹, for the sample 4Cu–4Ce/SBA-15 at 100 kPa. To further evaluate the adsorption behavior, the C₂H₄/C₂H₆ selectivity was predicted using ideal adsorbed solution theory (IAST), and the results are presented in Fig. 5. The C₂H₄/C₂H₆ selectivity of the different adsorbents decreased in the order 4Cu–4Ce/SBA-15 > 4Cu/SBA-15 > 4Ce/SBA-15. In detail, the C₂H₄/C₂H₆ selectivity was 1.7 for 4Ce/SBA-15 at a C₂H₄ molar fraction of 0.2. A selectivity of 5.7 was obtained for 4Cu/SBA-15, and the selectivity was as high as 11.6 for the sample 4Cu–4Ce/SBA-15. For the bifunctional samples with different metal contents, 4Cu–4Ce/SBA-15 exhibited the best performance, superior to 8Cu–8Ce/SBA-15. The potential reasons for the inferior capacity of 8Cu–8Ce/SBA-15 could be incompletely decomposed Cu(II) sites, and partial destruction of the pore structures. It is known that Cu(I) sites have a highly selective adsorption capacity for C₂H₄ molecules, which contain a C=C bond, through π-complexative forces. This gives rise to preferential adsorption of C₂H₄ over C₂H₆ for the CuCl/PM samples. As a result, the better performance in C₂H₄/C₂H₆ separation for the bifunctional materials should be attributed to the higher yield of Cu(I) sites during the guests-redox process.

Fig. 4 Adsorption isotherms of C₂H₄ and C₂H₆ on modified samples.

Fig. 5 C₂H₄/C₂H₆ selectivity profiles of 4Ce/SBA-15, 4Cu/SBA-15, and 4Cu–4Ce/SBA-15 at 100 kPa.

3.3 Proposed mechanism for the guests-redox strategy

To examine the pathway of the guests-redox strategy, the samples that were simply ground, without further thermal treatment, were monitored by TG analysis. The results are depicted in Fig. 6. For the sample 4Cu/SBA-15, the conversion of CuCl₂·2H₂O starts with the removal of adsorbed and crystalline water followed by the conversion of CuCl₂ to CuCl and Cl₂ as shown in eqn (1).

2CuCl₂ → 2CuCl + Cl₂ (1)

Fig. 6 TG curves of samples before calcination.

The weight loss due to CuCl₂ reduction continues from 270 °C to 650 °C. A similar process for CuCl₂·2H₂O decomposition has been reported before.⁸ A distinct process is also observed for the conversion of CeCl₃·6H₂O in the sample Ce/SBA-15. During the thermal treatment, the existence of crystalline or adsorbed hydrates should give rise to hydrolysis reactions, which leads to the formation of cerium oxides.²⁸ As described above, the decomposition of CeCl₃·6H₂O starts with the hydrolysis reaction, then generates CeO₂ from 120 °C to 600 °C. In contrast to the samples above, the bifunctional material 4Cu–4Ce/SBA-15 shows a quite different conversion route. A simultaneous weight loss attributed to the conversion of CuCl₂ to CuCl, and CeCl₃ to CeO₂, can be observed between 120 °C and 700 °C. According to the XPS results, CuCl₂ and CeCl₃ react in the same molar ratio, and the byproduct is probably HCl as shown in eqn (2).

CuCl₂ + CeCl₃ + 2H₂O → CeO₂ + CuCl + 4HCl (2)

It is interesting to note that the conversion of CuCl₂ to CuCl, and CeCl₃ to CeO₂, could be carried out in one calcination step, and HCl becomes the gaseous byproduct; no longer is any toxic Cl₂ generated.

To further verify the feasibility of the guests-redox strategy, the free energy change of eqn (2) was calculated. As shown in eqn (S9),† the free energy change for the process at 623.15 K and 101.325 kPa is below zero. This proves that the process can occur. However, the free energy change for the process under standard conditions of 273.15 K and 101.325 kPa is over zero (eqn (S10)†). This indicates that the process is a thermodynamically disfavored reaction, and could never occur at standard conditions.

3.4 Factors influencing the adsorption performance

It is known that C₂H₄ and C₂H₆ molecules have similar structural formulas, which results in similar molecular sizes. It is reported that the kinetic diameter is 0.4163 nm for C₂H₄, and 0.4443 nm for C₂H₆.²⁹ Therefore, the molecular sizes of C₂H₄ and C₂H₆ are extremely close. The pore diameter of the mesoporous silica SBA-15 is 9.2 nm. Therefore, the pore size of SBA-15 is over 20 times larger than the kinetic diameters of C₂H₄ and C₂H₆. If the pore size of the material were between the sizes of C₂H₄ and C₂H₆, separation of C₂H₄/C₂H₆ could be realized based on the principle of sieving. Evidently, it is impossible to separate C₂H₄/C₂H₆ by use of the pore characteristics of SBA-15, with a pore diameter of 9.2 nm. The adsorption of C₂H₄ and C₂H₆ on pure SBA-15 mainly depends on the van der Waals force. With the introduction of Cu and/or Ce, the pore diameter undergoes a slight change. However, the minimum diameter is still as high as 8.4 nm. It is clear that the diameters of all the samples are still 18 times larger than C₂H₄ and C₂H₆. As a result, separation of C₂H₄/C₂H₆ molecules by use of the sieving principle is still impossible to achieve with the modified samples. Unfortunately, therefore, separation of C₂H₄/C₂H₆ with our materials is not mainly dependent on the pore characteristics.

Since this is so, what could be responsible for the enhanced C₂H₄/C₂H₆ selectivity of our materials? π complexation adsorption is described as one of the most promising methods to separate C₂H₄/C₂H₆ with high selectivity. It is reported that Cu(I) sites can form the π complexation force with the C=C bond in C₂H₄. In contrast, the bond energy between Cu(I) and C₂H₆ arises from the van der Waals force. The π complexation force is evidently stronger than the van der Waals force. Based on such a mechanism, C₂H₄ could be preferentially adsorbed on the Cu(I) sites. Thus, studies have been devoted to the optimization of the π complexation adsorption. Dispersion of Cu(I) on mesoporous silica has been identified as an efficient method. Although mesoporous silica can only form the van der Waals force with C₂H₄ and C₂H₆, the large surface areas generated by the pores of mesoporous silica could be used to disperse the Cu(I) sites. Dispersed Cu(I) sites can come into contact with more C₂H₄ molecules than aggregated sites. In the present work, it is clear that the predominant sites responsible for C₂H₄/C₂H₆ separation are accessible Cu(I) species. The main function of the pore characteristics of the mesoporous silica SBA-15 is to make the Cu(I) species more accessible on the surface. It therefore follows that, on a support of mesoporous silica SBA-15, the more Cu(I) species are present, the better the separation performance. From this detailed analysis of the present work, we conclude that the C₂H₄/C₂H₆ selectivity is mainly dependent on the precise content of Cu(I) in SBA-15.

4. Conclusions

In summary, a guests-redox strategy was developed to prepare bifunctional adsorbents for gas separation of C₂H₄ from C₂H₆. The precursors CuCl₂ and CeCl₃ were simultaneously introduced into the mesoporous silica SBA-15. In one step of thermal treatment, both the reduction of CuCl₂ to CuCl, and the oxidation of CeCl₃ to CeO₂, could be achieved. In contrast to the conventional approach, our strategy avoids the repeated modification process and avoids the generation of toxic gas. More importantly, the yield of Cu(I) is enhanced under the action of the reducing precursor CeCl₃. We also demonstrate that the resultant bifunctional materials are capable of gas separation of C₂H₄ from C₂H₆ with a high selectivity. Our strategy may open up an avenue for the design and fabrication of new bifunctional materials by using the guests-redox strategy.

Acknowledgements

This work was supported by National Natural Science Foundation of China (21576137), Distinguished Youth Foundation of Jiangsu Province (BK20130045), China Postdoctoral Science Foundation (2015M581750), Jiangsu Planned Projects for Postdoctoral Research Funds (1501114B), State Key Laboratory of Materials-Oriented Chemical Engineering (KL15-13).

References

1. G. Chang, Z. Bao, Q. Ren, S. Deng, Z. Zhang, B. Su, H. Xing and Y. Yang, RSC Adv., 2014, 4, 20230–20233.
2. M. S. Jin, S. S. Kim, Y. D. Kim, J. N. Park, J. H. Kim, C. H. Ko, J. N. Kim and J. M. Kim, J. Mater. Chem. A, 2013, 1, 6653–6657.
3. W. J. Jiang, L. B. Sun, Y. Yin, X. L. Song and X. Q. Liu, Sep. Sci. Technol., 2013, 48, 968–976.
4. I. Ahmed and S. H. Jhung, Chem. Eng. J., 2015, 279, 327–334.
5. W. Dai, J. Hu, L. M. Zhou, S. Li, X. Hu and H. Huang, Energy Fuels, 2013, 27, 816–821.
6. N. A. Khan and S. H. Jhung, Angew. Chem., Int. Ed., 2012, 51, 1198–1201.
7. G. S. He, L. B. Sun, X. L. Song, X. Q. Liu, Y. Yin and Y. C. Wang, Energy Fuels, 2011, 25, 3506–3513.
8. Z. Li, R. Y. Wang, H. Y. Zheng and K. C. Xie, Fuel, 2010, 89, 1339–1343.
9. A. J. Rouco, Appl. Catal., A, 1994, 117, 139–149.
10. A. J. Hernandez-Maldonado and R. T. Yang, J. Am. Chem. Soc., 2004, 126, 992–993.
11. L. B. Sun, X. Q. Liu and H. C. Zhou, Chem. Soc. Rev., 2015, 44, 5092–5147.
12. Y. Yin, D. M. Xue, X. Q. Liu, G. Xu, P. Ye, M. Y. Wu and L. B. Sun, Chem. Commun., 2012, 48, 9495–9497.
13. J. H. Shan, X. Q. Liu, L. B. Sun and R. Cui, Energy Fuels, 2008, 22, 3955–3959.
14. H. Yen, Y. Seo, S. Kaliaguine and F. Kleitz, Angew. Chem., Int. Ed., 2012, 51, 12032–12035.
15. H. Song, Y. Chang and H. Song, Adsorption, 2016, 22, 139–150.
16. 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.
17. W.-H. Tian, L.-B. Sun, X.-L. Song, X.-Q. Liu, Y. Yin and G.-S. He, Langmuir, 2010, 26, 17398–17404.
18. J. X. Qin, Z. M. Wang, X. Q. Liu, Y. X. Li and L. B. Sun, J. Mater. Chem. A, 2015, 3, 12247–12251.
19. Y. Yin, P. Tan, X. Q. Liu, J. Zhu and L. B. Sun, J. Mater. Chem. A, 2014, 2, 3399–3406.
20. H. Zhu, M. L. Du, D. L. Yu, Y. Wang, L. N. Wang, M. L. Zou, M. Zhang and Y. Q. Fu, J. Mater. Chem. A, 2013, 1, 919–929.
21. C. X. Xiang, G. M. Kimball, R. L. Grimm, B. S. Brunschwig, H. A. Atwater and N. S. Lewis, Energy Environ. Sci., 2011, 4, 1311–1318.
22. D. M. Ruthven and S. C. Reyes, Microporous Mesoporous Mater., 2007, 104, 59–66.
23. R. T. Yang and E. S. Kikkinides, AIChE J., 1995, 41, 509–517.
24. S. Aguado, G. Bergeret, C. Daniel and D. Farrusseng, J. Am. Chem. Soc., 2012, 134, 14635–14637.
25. A. G. Albesa, M. Rafti, D. S. Rawat, J. Luis Vicente and A. D. Migone, Langmuir, 2012, 28, 1824–1832.
26. E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown and J. R. Long, Science, 2012, 335, 1606–1610.
27. Z. B. Bao, S. Alnemrat, L. Yu, I. Vasiliev, Q. L. Ren, X. Y. Lu and S. G. Deng, Langmuir, 2011, 27, 13554–13562.
28. H. B. Chen, P. Z. Yang, C. Y. Zhou and C. Y. Jiang, J. Alloys Compd., 2008, 449, 172–175.
29. J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38, 1477–1504.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section, supplementary results. See DOI: 10.1039/c6ra14091b

---