ZnCl₂-activated porous carbon spheres with high surface area and superior mesoporous structure as an efficient supercapacitor electrode

Abstract

Thanks to the unique architectural design, nanosized porous carbon materials exhibit better behavior as electrical double-layer capacitors than conventional carbon-base materials. In this work, porous carbon sphere (PCS) materials with superior porosity and uniform nanospherical morphology were successfully prepared by means of a facile chemical activation route. The analysis of pore structure and morphology of the resultant PCS were characterized by X-ray diffraction, Fourier transform infrared spectroscopy, N₂ sorption technology and electron microscope. The results indicated that this PCS material possess remarkable porosity, extremely large surface area (∼2500 m² g⁻¹), large pore volume (1.37 cm³ g⁻¹) and narrow pore distribution (2.73 nm). The well-developed mesoporous structure and high surface area benefited the PCS to exhibit an excellent charge storage capacity with a specific capacitance of 196 F g⁻¹ in 2 M KOH at a current density of 0.5 A g⁻¹ and long-term cycling stability over 1000 cycles. Compared with ordered mesoporous carbon and other porous carbon materials, PCS present an enhanced electrochemical performance, which could be attributed to its high surface area and well-developed mesoporosity, as well as its nanospherical morphology, favoring the ion accumulation on the electrode surface and facilitating fast electrolyte ion transportation.

---

Introduction

Currently, due to high-consumption of global energy and the fears of environmental pollution, scientists are prompting to develop sustainable and efficiently high advanced technologies of converting energy and storing it without damage to the environment.¹ To meet these demands, energy storage device technologies have been a research highlight for several decades. Supercapacitors are energy storage devices that accumulate energy in the form of electrical charge and bridge the gap between the dielectric capacitors and batteries.² Owing to their high power density, simple principle, long cycle life, excellent pulse charge–discharge capability and environmental friendship, supercapacitors have attracted a significant attention.^3,4

Based on the charge-storage mechanism, supercapacitors can be divided into pseudocapacitors and electrical double-layer capacitors (EDLCs).⁵ Pesudocapacitors generate faradaic current through rapid redox reactions, taking place on the surface of electrode materials, and the energy storage of EDLCs arises from the pure electrostatic charge accumulation on the surface of electrodes. In addition, the surface property and surface area play a vital role for the electrochemical performance in both the types of supercapacitors. Porous materials with high surface area and superior porosity have been demonstrated to favor the enhancement of electrochemical property. For example, Yang et al. reported a high specific capacitance of 170 F g⁻¹ in 6 M of KOH electrolyte from a H₂O activated porous carbon materials, which had a significantly large surface area of 1453 m² g⁻¹.⁶ Kong et al. reported the preparation of nickel hydroxide nanoflakes with an inter-connected mesoporous structure that delivered a high specific capacitance of 2055 F g⁻¹.⁷

In spite of their large specific capacitance values, pseudocapacitors suffer relatively low power density and poor cycle stability, which considerably restrict its practical applications.^8,9 In contrast to the pseudocapacitors, EDLCs are characterized by fast charge and discharge rates, high power density and long cycle life.¹⁰ Currently, EDLCs electrodes are mainly composed of various carbonaceous materials with excellent electrical conductivity and porous structure, such as activated carbons, ordered mesoporous carbons, carbide-derived carbons, carbon nanotubes and graphene-based materials.^11–14 Among them, because of the low cost, extensive resources, large surface area, controllable porosity and good electroconductivity, activated carbons are always one of the most investigated electrode materials in the past few decades.¹⁵ Because of the advantage of large surface area and superior porosity, activated carbons are favorable for the rapid transfer of electrolyte ions in electrode and storing large amounts of charges on surface. In general, there are two forms of activation methods for the preparation of activated carbons: physical activation and chemical activation.¹⁶ The usual activating agents in the former method are steam and carbon dioxide,^17–19 whereas KOH, ZnCl₂ and H₃PO₄ are commonly used as chemical activating agents in latter.^20–24 However, in the activated carbons that are developed using physical activation or chemical activation with KOH and H₃PO₄, micropores are dominant and the small pore size would significantly confine the supply of electrolyte ions, therefore resulting in the enhancement of resistance caused by the concentration polarization effect.²⁵ Mesopores are reported to act as a major reservoir for electrolyte ions and facilitate ion transport through the pore network at fast charge–discharge rates.²⁶ Thus, mesoporous carbons are expected to act as potential electrode materials for the application of EDLCs. However, the template methods, the most common strategies, for the synthesis of mesoporous carbons are multi-step and expensive.²⁷ In comparison with other activation routes, the ZnCl₂ activation easily generates a tunable pore structure from microporous to mesoporous.²⁴ In addition, the simultaneous progress of carbonization and activation simplifies the preparation procedure of activated carbons. Therefore, activation with ZnCl₂ is a worthy route to produce porous carbons with well-developed mesoporosity and large surface area for various purposes, in particular for electrochemical applications. To date, porous carbon materials derived from various precursors and chemically activated with ZnCl₂ have been successfully fabricated and reported, which exhibit superior porosity and improved capacitive performances,^{12,16,20,28,29} as summarized in Table 1.

Table 1 The comparison of the structure and performance of PCS with other porous carbon materials

Carbon source	Activating agent	SBET (m^2 g^−1)	Capacitance (F g^−1)	Electrolyte	Ref.
Glucose	ZnCl2	2560	196	2 M KOH	—
Sugar cane bagasse	NaOH	2871	109	1 M H2SO4	23
Fire-wood	KOH	1064	180	0.5 M H2SO4	22
Cow dung	KOH	1984	117	1 M Et4NBF4	21
Phenolic resin	KOH	1202	179	6 M KOH	13
Polyacrylonitrile	ZnCl2	550	92	6 M KOH	28
Peanut shell	ZnCl2	1552	184	6 M KOH	20
Polyaniline	ZnCl2	824	174	6 M KOH	16
Cationic starch	ZnCl2	1674	136	6 M KOH	29
Polyacrylonitrile	ZnCl2	550	140	6 M KOH	12

---

Moreover, the electrochemical properties of porous carbons are also affected by their shapes and morphologies. For instance, it has been demonstrated that porous carbons with nanosized aspect are characterized for shortening the transferring distance of electrolyte ions from bulk solution to electrode surface, therefore diminishing the resistance of ion diffusion.³⁰ In this regard, it is quite attractive and desirable to design and fabricate carbon nanostructures with regular geometry and well-developed mesopore to be EDLCs electrodes. Herein, we successfully synthesized monodisperse porous carbon sphere (PCS) with superior mesoporous structure by a facile ZnCl₂ activation method. This PCS material owns extremely high surface area, large pore volume and narrow pore size distribution of 2.0–4.0 nm, as well as uniform spherical shape with a diameter of ∼500 nm. Because of these structural advantages, PCS has been evaluated as EDLCs electrode in aqueous KOH solution, and it exhibits excellent electrochemical performance. In addition, 96% of the specific capacitance could be retained after 1000 cycles that suggested its superior stability and long-term cycle reusability.

Experimental

Synthesis of materials

Carbon nanospheres (CNS) were obtained by a simple and controllable hydrothermal synthetic method using glucose as the carbon source.^31,32 PCS were prepared by a facile chemical activation route. In a typical experiment, 1 g of CNS was dispersed into 40 mL of aqueous solution containing 4 g of ZnCl₂ and then stirred for 6 h. Subsequently, the solution was kept into an oven at 110 °C for 10 h to completely evaporate the solvent to obtain ZnCl₂-impregnated CNS material, and the material was then activated in a N₂ atmosphere at 800 °C for 1 h. Finally, the activated sample was repeatedly washed with distilled water and HCl solution (0.5 M), and dried under vacuum at 80 °C for 10 h.

For comparison, mesoporous carbon (CMK-3) was used as a reference material. CMK-3 was synthesized using SBA-15 as the hard template and furfuryl alcohol as the carbon source and carbonized at 800 °C under a N₂ atmosphere.

Characterizations

X-ray diffraction (XRD) patterns were monitored by a DX-2700 diffractometer (Dandong Haoyuan Instrument Co. Ltd., China) using Cu Kα radiation (λ = 0.15418 nm) as an X-ray source. Nitrogen adsorption–desorption isotherm measurements were carried out at −196 °C using a micromeritics ASAP 2020 analyzer. Prior to adsorption, the samples were out-gassed at 200 °C for 10 h. The specific surface area (S_BET) was evaluated using the Brunauer–Emmett–Teller (BET) method, and the mesopore volume and micropore volume were calculated according to the Barrett–Joyner–Halenda (BJH) formula and the t-plot method, respectively. The pore size distributions were calculated according to the density-functional-theory (DFT) method. Fourier transform infrared spectroscopy (FTIR) spectra of a sample in KBr pellet were recorded on a Nicolet Avatar 370 spectrometer. The morphology and pore structure were observed using a JEOL JEM-2100 transmission electron microscope (TEM) with an accelerating voltage of 200 KV and a scanning electron microscope (SEM, Hitachi S-4800).

Electrochemical measurements

The products were tested via a conventional three-electrode system in a 2 M KOH electrolyte solution, which were assessed on a CHI660D electrochemical workstation at room temperature. Platinum foil and an Ag/AgCl saturated KCl electrode were used as counter and reference electrodes, respectively. The working electrodes were prepared by mixing active material, acetylene black and polytetrafluoroethylene (PTFE) binder with a weight ratio of 8 : 1 : 1. After coating the above slurries on foamed Ni grids (1 cm × 1 cm), the electrode was dried overnight at 100 °C before pressing it under a pressure of 20 MPa. Cyclic voltammetry curves were obtained in the potential range of −1.0 to 0 V vs. Ag/AgCl by varying the scan rate from 1 to 50 mV s⁻¹. Charge–discharge measurements were galvanostatically done at 0.5–10 A g⁻¹ over a voltage range of −1.0 to 0 V vs. Ag/AgCl. Electrochemical impedance spectroscopy (EIS) was performed in a frequency range of 10 kHz to 0.01 Hz in open circuit voltage with an alternate current amplitude of 5 mV.

Results and discussion

Fig. 1 presents the electron microscope images of CNS and PCS samples, which clearly reveal the morphology and structure of materials. As observed from the SEM images (Fig. 1a and b), the uniform spherical shapes are found in both CNS and PCS samples, which suggest that the activation process with ZnCl₂ maintain the original morphology and the structure of carbon nanospheres. In addition, from the magnified image of CNS (inserted Fig. 1a) it can be proved that the surface of carbon nanospheres is smooth and nonporous. The TEM images of PCS (Fig. 1c and d) also simultaneously demonstrate the morphology of spherical shape, and the discernible pores are directly showed in PCS sample (Fig. 1d).

Fig. 1 SEM images of CNS (a) and PCS (b) samples; TEM images of PCS sample (c and d); the inset of (a) is the magnifying SEM image of CNS.

To further investigate the porosity of PCS material, N₂ sorption technology is conducted (Fig. 2) and the porous properties of the resultant materials are summarized in Table 2. CMK-3 sample exhibits a typical IV curve with a clear hysteresis loop at relative pressure from 0.40 to 0.60, indicating that the uniform mesoporous channels are retained.³³ PCS material shows a transitional isotherm curve varying from type I to IV, and a significant increase of adsorbed N₂ volume below p/p₀ = 0.4 can be observed, which is lower than the leap of adsorbed N₂ volume of CMK-3, suggesting that the PCS has a smaller pore size. Fig. 2b displays the pore size distributions of PCS and CMK-3. It is quite obvious that the pore size of PCS is considerably smaller than that of CMK-3, however, its pore size is still mesoporous (2.73 nm). Moreover, the high N₂ sorption capacity of PCS implies its high surface area and pore volume. From the data in Table 2, it could be found that the superior mesoporous structure is formed, and a high surface area and large pore volume are obtained after activation by ZnCl₂, which are considerably higher than those of CMK-3.

Fig. 2 N₂ adsorption–desorption isotherm (a) and pore size distribution (b) of PCS and CMK-3.

Table 2 Textural parameters of all samples

Sample	SBET^a (m^2 g^−1)	Smeso^b (m^2 g^−1)	Smicro^c (m^2 g^−1)	Vtotal^d (cm^3 g^−1)	Dp^e (nm)
a BET surface area.b Mesopore surface area calculated using the V–t plot method.c Micropore surface area calculated using the V–t plot method.d The total pore volume calculated by single adsorption at p/po = 0.97.e Pore diameter of peak value in Fig. 2b.
CNS	<5	<5	—	—	—
PCS	2560.2	2646.1	—	1.37	2.73
CMK-3	1070.4	1062.9	7.39	1.01	3.88

---

To reveal the influence of activation process on the structure and surface property of carbon spheres, XRD and FTIR measurements were investigated. Fig. 3a displays the XRD patterns of all samples in wide angle region. It clearly shows that all of these samples present a broad diffraction peak at about 23.2°, which is related to the amorphous structure of carbon.³⁴ In comparison with CNS and CMK-3 samples, after the ZnCl₂ activation and further carbonization, a weak peak at 43.5° appears for the pattern of PCS, and this diffraction is attributed to the interlayer condensation of graphite layers.³⁵ This result suggests that PCS possess a considerably higher degree of graphitization than CMK-3, and therefore is endowed with better electrical conductivity that benefits the fast electron transfer during the electrochemical test. A large number of functional groups are present in the FTIR spectrum of CNS sample. The bands centered at 1000–1300 cm⁻¹ are related to the C–OH stretching vibration and –OH bending vibration, which show the existence of large numbers of residual hydroxy groups.³⁶ The bands at 1720 and 1620 cm⁻¹ can be attributed to C=O and C=C stretching vibrations, respectively, and the results support the concept of aromatization of glucose during the hydrothermal treatment.³² The other absorption bands at 1390–1470 cm⁻¹ and 2830–2930 cm⁻¹ correspond to the C–H bending vibration and stretching vibration, respectively. These results suggest that CNS materials should possess abundant oxygen-containing and hydrogen-containing functional groups, which favor the process of activation. However, only the C=C stretching vibration at 1580 cm⁻¹ can be clearly seen for the PCS and CMK-3 samples, and the characteristic adsorption bands of C=O and C–O have disappeared. The generation of such results sufficiently verifies the mechanism of activation that ZnCl₂ is used as a dehydrating agent to extract H and O atoms from the surface of incomplete carbonized framework, resulting in the elimination of surface oxygen-containing and hydrogen-containing functional groups. The absorption band of –OH bending vibration at 1000–1200 cm⁻¹ and 3400 cm⁻¹ could be due to an adsorbed H₂O molecule.

Fig. 3 The XRD patterns (a) of CNS, PCS and CMK-3; FTIR spectra (b) of CNS, PCS and CMK-3.

To evaluate the capacitive performance of PCS material, cyclic voltammetry was carried out in 2 M KOH aqueous solution with a three-electrode system at room temperature, and CMK-3 was used as a reference material. The specific capacitance of the electrode at various scan rates were calculated on the basis of CV curves according to the following equation:

C = ∫IdV/mVν

where I (A) is the response current density, V (V) is the potential, ν (mV s⁻¹) is the potential scan rate, and m (g) is the mass of electroactive material in the electrode. Fig. 4 shows the CV curves of CMK-3 (Fig. 4a) and PCS (Fig. 4b) at various scan rates ranging from 1 to 50 mV s⁻¹, their calculated specific capacitances at various scan rates (Fig. 4c) and a comparison of CV curves at 50 mV s⁻¹ (Fig. 4d). All the CV curves exhibit quasi rectangular shapes in a potential range of −1.0 to 0 V at different scan rates, suggesting an ideal EDLC behavior and a fast diffusion of electrolyte ions into/out of the electrode materials.³⁷ However, it is obvious that the curves are deformed with an increase of the scan rate, which should be attributed to the slow charge–discharge kinetics of the electrode material.¹¹ Therefore, electrolyte ion diffusions are unable to properly access the surface of carbon in such a short time at high scan rates. As shown in Fig. 4d, the specific capacitance of PCS electrode is calculated to be 206 F g⁻¹ at a scan rate of 1 mV s⁻¹, which is considerably higher than that of CMK-3 at the same scan rate (109 F g⁻¹). Moreover, it is worth noting that the specific capacitances of PCS calculated from CV curves at any scan rate are considerably higher than those of CMK-3 material. In addition, the specific capacitance of PCS electrode still retains 79% of its initial capacitance even at a high scan rate of 50 mV s⁻¹, suggesting the fast ion-transport in the pore channel of PCS. The capacitance value of PCS is indeed considerably higher than those of other porous carbon materials reported in the literature (Table 1). Compared with these porous carbons, the enhanced specific capacitance of PCS is attributed to three factors: (1) its mesoporous structure and large surface area that favor ion transfer in the electrode and store a considerable amount of charge; (2) its nanospherical morphology that shortens the diffusion distance of electrolyte ions and therefore accelerates the ion-transport; (3) its higher graphitization degree (see from its XRD pattern) that is advantageous to the fast electron transfer.

Fig. 4 CV curves of CMK-3 (a) and PCS (b) at different scan rates; (c) specific capacitances of CMK-3 and PCS at different scan rates; (d) CV curves of PCS and CMK-3 at the scan rate of 50 mV s⁻¹.

In order to further test the electrochemical performance of PCS, accurate galvanostatic charge/discharge experiments were also conducted at various current densities. The specific capacitances of the electrodes are calculated by the following equation:

C = IΔt/mΔV

where I (A) is current loaded, m (g) is the mass of active material, Δt (s) is the discharge time and ΔV (V) is the range of potential. As displayed in Fig. 5a and b, all the curves show isosceles triangular shapes suggesting an almost ideal EDLC behavior, whereas, the quasi-linear nature for PCS indicates the occurrence of graphitization and enhancement of conductivity.³⁸ Fig. 5c presents the relationship between the specific capacitance and current density. PCS exhibits a specific capacitance as high as 196 F g⁻¹ at a current density of 0.5 A g⁻¹, and a specific capacitance of 171 F g⁻¹ can be obtained even at a higher current density of 10 A g⁻¹, which are much higher than those of CMK-3 at the same current density (103 F g⁻¹ and 81 F g⁻¹, respectively). It is obvious that over the current density range from 0.5 to 10 A g⁻¹, the specific capacitances of PCS is considerably larger than those of CMK-3 at both low and high current density, and even the specific capacitance of PCS is almost twice than that of CMK-3 at the same current density. These results agree well with the results obtained from the CV tests. Based on the structural characterization, it is demonstrated that PCS material possess high effective surface area and superior pore structure which are compatible with aqueous KOH electrolyte, and provide a short diffusion path for electrolyte ions, favoring the high power capability and large capacitance. To further evaluate the capacitive behavior of PCS, electrochemical impedance spectroscopy test was carried out (Fig. 5d). The resulting Nyquist plots of PCS and CMK-3 display the semicircles in the high frequency region, reflecting the existence of charge transfer resistance.³⁹ The equivalent series resistance is 0.8 Ω for PCS, which is lower than that of CMK-3 (1.2 Ω), showing high charging and discharging rate of the PCS electrode. At low frequencies, a nearly vertical line is exhibited that represents the dominance of double layer change-storage.⁴⁰

Fig. 5 Charge–discharge curves of CMK-3 (a) and PCS (b) at different current densities; (c) specific capacitances of CMK-3 and PCS at different current densities; (d) electrochemical impedance spectra of CMK-3 and PCS under the influence of an ac voltage of 5 mV.

To validate the practical application efficiency of PCS electrode material, the cyclic stability was estimated using charge–discharge cycling. As shown in Fig. 6, the specific capacitance of PCS slightly decreases during the first 20 cycles and remains constant for approximate 180 cycles. However, it is interesting that the specific capacitance increases after slightly decreasing during this cyclic charge–discharge process. This variation may be attributed to the effect from activation of the electrode to expose additional surface area.⁴¹ The specific capacitance of PCS still amounts to about 96% of the initial specific capacitance after 1000 cycles, which indicates that PCS electrode possess excellent cyclic stability.

Fig. 6 Cyclic stability of PCS at a current density of 5 A g⁻¹ for 1000 cycles in 2 M KOH electrolyte. The inset is the galvanostatic charge–discharge cycles.

Conclusions

In summary, porous carbon spheres with superior mesoporous structure and uniform morphology were successfully prepared with glucose as the carbon source by a simple chemical activation method. The PCS material possesses extremely large surface area, large pore volume and outstanding mesoporous structure. In addition, Due to the low cost of the carbon precursor and the facile preparation process, the as-prepared porous carbon sphere materials offer attractive aspect and can be adopted for practical application. In addition, it has been demonstrated that the high surface area and superior porous spherical structure of PCS favor the fast transport and diffusion of electrolyte ions, resulting in an excellent capacitive performance and long-term cycling stability. The PCS electrode material displays a high specific capacitance of 196 F g⁻¹ at a current density of 0.5 A g⁻¹, and the specific capacitance reduces only by 13% at a higher current density of 10 A g⁻¹. Moreover, PCS electrode shows a superior cyclic stability and the specific capacitance of PCS decreases only for approximate 4% after 1000 cycles at a current density of 5 A g⁻¹.

Acknowledgements

The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (51372237), 521 talent project of ZSTU, the program for innovative research team of ZSTU (13060052-Y) and the project-sponsored by the Scientific Research Foundation (SRF) for the Returned Overseas Chinese Scholars (ROCS), State Education Ministry (SEM).

References

1. L. Dai, D. W. Chang, J. B. Baek and W. Lu, Small, 2012, 8, 1130–1166.
2. Z. Niu, H. Dong, B. Zhu, J. Li, H. H. Hng, W. Zhou, X. Chen and S. Xie, Adv. Mater., 2013, 25, 1058–1064.
3. M. Winterand and R. J. Brodd, Chem. Rev., 2003, 104, 4245–4269.
4. P. Simon and Y. Gogotsi, Nat. Mater., 2008, 7, 845–854.
5. G. P. Wang, L. Zhang and J. J. Zhang, Chem. Soc. Rev., 2012, 41, 797–828.
6. C. Kim, Y. O. Choi, W. J. Lee and K. S. Yang, Electrochim. Acta, 2004, 50, 883–887.
7. L. B. Kong, W. J. Wu, M. Liu, Y. C. Luo and L. Kang, J. Solid State Electrochem., 2009, 13, 333–337.
8. X. Dong, X. Wang, L. Wang, H. Song, X. Li, L. Wang, M. B. Chan-Park, C. M. Li and P. Chen, Carbon, 2012, 50, 4865–4870.
9. M. J. Zhi, C. C. Xiang, J. T. Li, M. Li and N. Q. Wu, Nanoscale, 2013, 5, 72–88.
10. C. Largeot, C. Portet, J. Chmiola, P. L. Taberna, Y. Gogotsi and P. Simon, J. Am. Chem. Soc., 2008, 130, 2730–2731.
11. V. Subramanian, C. Luo, A. M. Stephan, K. S. Nahm, S. Thomas and B. Wei, J. Phys. Chem. C, 2007, 111, 7527–7531.
12. B. C. Kim, B. T. N. Ngoc, K. S. Yang, M. Kojima, Y. A. Kim, Y. J. Kim, M. Endo and S. C. Yang, Adv. Mater., 2007, 19, 2341–2346.
13. C. Ma, Y. J. Li, J. L. Shi, Y. Song and L. Liu, Chem. Eng. J., 2014, 249, 216–225.
14. X. L. Yan, X. J. Li, Z. F. Yan and S. Komarneni, Appl. Surf. Sci., 2014, 308, 306–310.
15. A. Burke, Electrochim. Acta, 2007, 53, 1083–1091.
16. X. X. Xiang, E. H. Liu, Z. Z. Huang, H. J. Shen, Y. Y. Tian, C. Y. Xiao, J. J. Yang and Z. H. Mao, J. Solid State Electrochem., 2011, 15, 2667–2674.
17. K. S. Xia, Q. M. Gao, J. H. Jiang and J. Hu, Carbon, 2008, 46, 1718–1726.
18. T. Y. Zhang, W. P. Walawender, L. T. Fan, M. Fan, D. Dangaard and R. C. Brown, Chem. Eng. J., 2004, 105, 53–59.
19. M. Valix, W. H. Cheung and K. Zhang, Adsorption, 2008, 14, 711–718.
20. X. J. He, R. C. Li, J. F. Han, M. X. Yu and M. B. Wu, Mater. Lett., 2013, 94, 158–160.
21. D. Bhattacharjya and J. S. Yu, J. Power Sources, 2014, 262, 224–231.
22. K. Konno, Y. Ohba, K. Onoe and T. Yamaguchi, Tanso, 2008, 231, 2–7.
23. F. C. Wu, R. L. Tseng, C. C. Hu and C. C. Wang, J. Power Sources, 2005, 144, 302–309.
24. T. E. Rufford, D. Hulicova-Jurcakova, K. Khosla, Z. H. Zhu and G. Q. Lu, J. Power Sources, 2010, 195, 912–918.
25. H. Kim, M. E. Fortunato, H. Xu, J. H. Bang and K. S. Suslick, J. Phys. Chem. C, 2011, 115, 20481–20486.
26. D. W. Wang, F. Li, M. Liu, G. Q. Lu and H. M. Cheng, Angew. Chem., Int. Ed., 2008, 47, 373–376.
27. D. Bhattacharjya, M. S. Kim, T. S. Bae and J. S. Yu, J. Power Sources, 2013, 244, 799–805.
28. C. Kim, B. T. N. Ngoc, K. S. Yang, M. Kojima, Y. A. Kim, Y. J. Kim, M. Endo and S. C. Yang, Adv. Mater., 2007, 19, 2341–2346.
29. H. Q. Wang, Y. L. Zhong, Q. Y. Li, J. H. Yang and Q. F. Dai, J. Phys. Chem. Solids, 2008, 69, 2420–2425.
30. W. Xiong, M. Liu, L. Gan, Y. Lv, Y. Li, L. Yang, Z. Xu, Z. Hao, H. Liu and L. Chen, J. Power Sources, 2011, 196, 10461–10464.
31. X. M. Sun and Y. D. Li, Angew. Chem., Int. Ed., 2004, 43, 597–601.
32. B. B. Chang, D. X. Guan, Y. L. Tian, Z. C. Yang and X. P. Dong, J. Hazard. Mater., 2013, 262, 256–264.
33. B. B. Chang, J. Fu, Y. L. Tian and X. P. Dong, J. Phys. Chem. C, 2013, 117, 6252–6258.
34. M. Okamura, A. Takagaki, M. Toda, J. N. Kondo, K. Domen, T. Tatsumi, M. Hara and S. Hayashi, Chem. Mater., 2006, 18, 3039–3045.
35. J. P. Paraknowitsch, J. Zhang, D. Su, A. Thomas and M. Antonietti, Adv. Mater., 2010, 22, 87–92.
36. R. Demir-Cakan, N. Baccile, M. Antonietti and M. M. Titirici, Chem. Mater., 2009, 21, 484–490.
37. J. Wang, L. F. Shen, B. Ding, P. Nie, H. F. Deng, H. Dou and X. G. Zhang, RSC Adv., 2014, 4, 7538–7544.
38. H. Zhu, X. L. Wang, X. X. Liu and X. R. Yang, Adv. Mater., 2012, 24, 6524–6529.
39. Q. Wang, J. Yan, Y. B. Wang, T. Wei, M. L. Zhang, X. Y. Jing and Z. J. Fang, Carbon, 2014, 67, 119–127.
40. L. Huang, C. Li and G. Q. Shi, J. Mater. Chem. A, 2014, 2, 968–974.
41. H. Wang, Z. Xu, A. Kohandehghan, Z. Li, K. Cui, X. Tan, T. J. Stephenson, C. K. King'ondu, C. M. B. Holt, B. C. Olsen, J. K. Tak, D. Harfield, A. O. Anyia and D. Mitlin, ACS Nano, 2013, 7, 5131–5141.

---