Porous and single crystalline Co₃O₄ nanospheres for pseudocapacitors with enhanced performance

Abstract

Porous and single-crystalline Co₃O₄ nanospheres have been synthesized successfully on a large scale. It was found that the reaction time can dramatically affect the growth of the Co₃O₄ nanospheres in nanoreactors constructed with AOT⁻ and Co²⁺ ions. The dynamic evolution of the nanostructures produced in the reaction strongly support a stepwise splitting mechanism proposed to explain the formation of the porous and single crystalline nanospheres. The pseudocapacitors made with the porous and single crystalline Co₃O₄ nanospheres exhibit excellent charge-storage performance. The enhanced electrochemical properties of the materials can be attributed to their superstructures with pores and single crystalline features.

---

Introduction

Co₃O₄ is one of the promising materials widely applied to not only pseudocapacitors,¹ but also Li batteries,² sensors³ and catalysts.⁴ It has been demonstrated that the electrode materials for designing advanced energy storage devices have to facilitate both ion diffusion and electron transportation.⁵ In order to improve their energy storage performance, 3D hierarchical Co₃O₄ nanoarchitectures have been investigated extensively in recent years. Among the techniques developed, topotactic transformations of Co-related precursors can retain the 3D hierarchical nanostructures and generate pores throughout the materials, after the precursors are first produced in solution and then annealed at higher temperature.^{1j,k,m,2f–i,3,6} The porous 3D Co₃O₄ materials obtained, however, are generally composed of nanoparticles with polycrystalline features. Only have a few cases of porous Co₃O₄ materials consisting of quasi-single crystalline nanobelts^2c,d,7 and nanowires^2j,8 been reported and applied to Li-batteries so far. Recently, Li and co-workers found that single crystalline Co₃O₄ polyhedra can perform well in Li-storage predominated by their surface microstructures.^2b The charge storage capabilities of most pseudocapacitors made with Co₃O₄ materials are still much lower than their theoretically specific capacitance of 3650 F g⁻¹.

Because the spherical particles could pack closely to form electrodes with high density, nanospheres with porous and single crystalline features are extremely attractive in designing and fabricating devices for energy storage and conversation. Porous and single crystalline TiO₂ nanospheres synthesized with mesoporous templates, for instance, exhibit highly enhanced energy conversation efficiency in constructed solar cells, because of substantially higher conductivity and electron mobility of the materials in comparison with their polycrystalline counter-parts.⁹ It was also found that carbon materials with pores of less than 1 nm in diameter show anomalous increase in storing charges.¹⁰ These progresses strongly suggest that there is still room for improvement in the charge storing performances of pseudocapacitors made with Co₃O₄ materials through especially optimizing their pore and skeleton structures.

The researches in our groups have been mainly focusing on building nanoarchitectures for microdevices including supercapacitors.^1i,11 Herein, we would like to report porous and single crystalline Co₃O₄ nanospheres. The pseudocapacitors made with as-prepared materials exhibit highly enhanced performances attributed to their nanostructures. The dynamic evolution of nanostructures produced in the reactions suggests a stepwise splitting mechanism for the formation of the supercrystals.

Experimental

All of the reagents were analytically pure and used as received without further purification. AOT was purchased from Aldrich-Sigma and the others from Shanghai Chemical Industrial Co. Ltd., Shanghai, China.

Synthesis of Co₃O₄ nanospheres

In a typical synthesis of Co₃O₄ nanospheres, cobalt nitrate solution was first prepared by dissolving cobalt nitrate hexahydrate (Co(NO₃)₂·6H₂O) (0.7225 g, 2.50 mmol) in deionized water (90 mL) under stirring. AOT (3.3342 g, 7.50 mmol) was then dissolved in 1-butanol (10 mL) under vigorous stirring until it became transparent. Subsequently, the cobalt nitrate solution in water was added dropwise with 5 second intervals into AOT solution in 1-butanol under vigorous stirring to obtain pink colloid solution. After stirred for another 30 min at room temperature, concentrated NH₃·H₂O (0.6 mL, 28 wt%) was added dropwise into the mixture to obtain transparent solution with pale blue colour. The mixture was stirred for another 30 min and finally transferred into a Teflon-lined autoclave and maintained at 200 °C for 5 h. After cooled down to room temperature naturally, black precipitates were collected by centrifugation and thoroughly washed with deionized water and ethanol for 5 times each, respectively. Black powders were obtained after dried in an oven at 85 °C for 12 h.

Characterizations

The compositions of the materials were identified by powder X-ray diffraction (XRD) analysis using a D/max 2550v diffractometer with Cu Kα radiation (λ = 1.54056 Å, scan rate: 0.02° s⁻¹ in the range of 5–80°), and thermal analysis employing a SDTQ600 thermal analyzer working at a constant heating rate of 5 °C min⁻¹ in N₂. The specific surface area and pore size of the materials were measured using a Belsorp-Mini adsorption apparatus (Bel Japan Inc). The microstructures of the materials were characterized with field emission scanning electron microscopy (FESEM, JSM-7001F) and transmission electron microscopy (TEM, JEM-2100 operated at 200 kV).

Electrochemical properties

The electrochemical studies were carried out in a three-electrode system with KOH electrolyte solution (6 mol L⁻¹). A piece of nickel mesh with as-prepared Co₃O₄ nanospheres, a platinum wire, and saturated calomel electrode (SCE) were used as working electrode, counter electrode, and reference electrode, respectively. The mixture for fabricating working electrode were composed of active Co₃O₄ materials (80 wt%), conductive material (acetylene black, ATB, 10 wt%) and binder (PTFE, 10 wt%). It was first coated onto the surfaces of nickel foam sheets (1 cm × 1 cm) and then dried at 100 °C for 12 h. The Ni sheets with electrode materials were finally pressed under 10 MPa to obtain working electrodes. The cyclic voltammograms (CV) were measured with a CHI 660D electrochemical workstation. The CV tests were done between 0 and 0.4 V (vs. saturated calomel electrode (SCE)) at scan rates of 5, 10, 20, and 50 mV s⁻¹, respectively. Galvanostatic charge–discharge cycle tests were performed on a CT2001A LAND Cell test system. The specific capacitances of the supercapacitors were evaluated from the charge–discharge tests with the following equation (eqn (1)):

C = IΔt/mΔV (1)

where C is the specific capacitance of the capacitor (F g⁻¹), I is the current of charge–discharge, and Δt is the discharging time period in seconds for the potential change ΔV in volts; m is the mass of the active material loaded onto the working electrode. All of the electrochemical measurements were carried out at room temperature.

Results and discussion

The as-prepared products were first characterized with field emission scanning electron microscope (FESEM) as shown in Fig. 1a–d. The low magnification FESEM image (Fig. 1a) shows that granules with high uniformity and regularity have been produced on a large scale. Fig. 1e illustrates the size distribution of the granules with diameter of 256.7 ± 30.1 nm. The FESEM images of higher magnifications (Fig. 1b–d) of the granules reveal their surfaces covered with nanoparticles. Some of the granules also possess hemispherical geometry: one of their facets is rather flat and smooth. The XRD profile (Fig. 1f) of the materials consists of diffraction peaks corresponding to (111), (220), (311), (222), (400), (422), (511) and (440) of cubic Co₃O₄. The results are consistent well with those in JCPDS card (no. 42-1467). There is no diffraction peak from any other impurity in the XRD profile, which indicates the high purity of the products (deviation of 5%).

Fig. 1 (a–d) FESEM images, (e) size distribution and (f) XRD pattern of Co₃O₄ hemispheres.

The Co₃O₄ nanospheres were further characterized with transmission electron microscope (TEM), high resolution transmission electron microscope (HRTEM) and selected area electron diffraction (SAED). Fig. 2a presents TEM image of Co₃O₄ nanospheres at low magnification. The TEM image (Fig. 2b) of one nanosphere at higher magnification reveals that the particles consist of nanocuboids of ca. 3 nm in diameter. The SAED pattern (Fig. 2c) of the whole sphere as shown in Fig. 2b is composed of well ordered dots of (220) and (220) facets corresponding to [001] band axis, which can directly demonstrate the single crystalline nature of as-prepared Co₃O₄ granules. The HRTEM images (Fig. 2d and e) of the nanosphere performed at its upper and bottom right areas, respectively, are composed of d-spacings of 0.23 and 0.28 nm attributed to (222) and (220) planes of cubic Co₃O₄, respectively. The nanocuboids grow along 〈001〉 and 〈010〉 directions, respectively, and are exposed in {100} facets. In contrast, the HRTEM image (Fig. 2f) collected at upper left area of the nanosphere is composed of d-spacings of 0.28 nm corresponding to (220) plane of cubic Co₃O₄. The results indicate that these nanocuboids orient along 〈001〉 direction and are mainly exposed in {100} planes. More details concerned with the microstructures of porous Co₃O₄ nanospheres with single crystalline nature can be further revealed by performing HRTEM (Fig. S1†) around their edges. The nanocuboids as shown in Fig. 2f form relative flat and smooth facets for the hemispherical structure of the granules, compared to those in Fig. 2d and e. The defects resulting from the pores of 1–3 nm in diameter can be observed in the HRTEM images also. The inset in Fig. 2f shows the adsorption and desorption isotherm and the corresponding BJH pore size distribution plot of the Co₃O₄ nanospheres. Most of the calculated size of the pores is around 12.6 nm in diameter (pore volume: 0.35 cm³ g⁻¹). The Co₃O₄ nanospheres, which consist of nanocuboids mainly exposed in {100} facets with high BET surface area of 146.2 m² g⁻¹, have been synthesized successfully on a large scale.

Fig. 2 (a) Low magnification TEM images of Co₃O₄ hemispheres. (b) TEM image and (c) SAED pattern of one Co₃O₄ hemisphere. HRTEM images taken at (d) bottom right, (e) upper right and (f) upper left areas of the nanosphere as shown in (b), respectively. Inset in (f): physisorption isotherms and pore-size distribution curve of Co₃O₄ hemispheres.

In order to understand the formation of the 3D hierarchical Co₃O₄ hemispheres, the reaction time effect on the composition and microstructure of the materials was investigated carefully. Fig. 3 shows the XRD profiles of the materials produced by varying the reaction time from 1 to 5 h. After reacting for 1 h, the materials obtained only exhibit several broaden peaks of low intensity for their low crystallinity. After prolonging the reaction to 2 h at 200 °C, however, a series of diffraction peaks, especially the strongest diffraction at 9.6° corresponding to d-spacing of 9.2 Å, appear in the XRD profile. It has been verified that the surfactant – sodium bis(2-ethylhexyl) sulfosuccinate (AOT) molecules prefer to self-assemble into lamellar structures together with metal ions, which can act as nanoreactors to confine the growth of nanocrystals.^11d,e The hydrophobic tail of AOT can be calculated to be ca. 7.70 Å in length, based on the C–C bond length of 1.54 Å. The d-spacing of 9.2 Å can be therefore attributed to the lamellar assemblies formed by two layers of AOT molecules. In addition, it can be observed that the diffractions corresponding to (111) and (311) facets of cubic Co₃O₄ appear in the XRD profile of the materials after reacting at 200 °C for 2 h, which indicates that cubic Co₃O₄ has been readily generated in the reaction. The intensities of the diffractions corresponding to both Co₃O₄ and lamellar assemblies further increase and more diffraction peaks attributed to cubic Co₃O₄ appear in the XRD pattern of materials prepared by reacting for 3 h. In contrast to the intensity increase of diffractions aroused from cubic Co₃O₄, the peaks attributed to lamellar assemblies then decrease and finally disappear completely in the XRD profiles of the materials produced by performing the reactions for 4 and 5 h, respectively.

Fig. 3 The XRD profiles of the materials synthesized at 200 °C for 1, 2, 3, 4 and 5 h, respectively.

The reaction time effect on the microstructures of materials was further investigated with TEM (Fig. 4) and SEM (Fig. S2 and 3†). After reacting at 200 °C for 1 h, 2D lamellar assemblies can be obtained as shown in Fig. 4a1 and a2. The SAED pattern (inset in Fig. 4a2) reveals their low crystallinity, which is consistent well with XRD profile. However, 2D nanosheets (Fig. 4b1 and b2) of several hundred nanometers in dimension with clear edges can be produced by reacting for 2 h. The surfaces of the 2D nanosheets (Fig. 4b1 and b2) are much smoother in comparison to those as shown in Fig. 4a1 and a2. It is interesting to found that the SAED pattern inset in Fig. 4b2 consists of well ordered dots, which indicates the single-crystalline nature of the nanosheets. After increasing the reaction time to 3 h, dendric nanocubes consisting of nanocuboids of ca. 9 nm in length and ca. 3 nm in width can be generated on the surfaces of the nanosheet (Fig. 4c1 and S2†). The surfaces of the nanosheets also become ragged in accompanying with the appearance of nanocubes. After further prolonging the reaction time to 4 h, Co₃O₄ granules of ca. 180 nm in diameter (Fig. 4d1, d2 and S3†) are generated on the surfaces of the nanosheets. Compared to the nanodisks with straight edges as shown in Fig. 4b1, the nanosheets change into irregular disks in the system. The nanosheets eventually disappear completely and only Co₃O₄ nanospheres can be observed in the system (Fig. 4e1 and e2), after keeping the reaction at 200 °C for 5 h. The one-pot growth of Co₃O₄ granules in the approach is dramatically different in comparison to the fabrication of 3D Co₃O₄ nanostructures based on topotactic transformations, where Co-related precursors are first produced and then annealed at ca. 400 °C to produce hierarchical nanostructures with polycrystalline feature.¹ⁱ The structures of the materials produced in the reaction further change into dendritic nanoparticles (Fig. S4–7†) with polycrystalline nature and much lower yields, after keeping the reaction at 200 °C for 6 and 11 h, respectively.

Fig. 4 (a1–e1) Low and (a2–e2) high magnification TEM images of the materials produced at 200 °C for 1, 2, 3, 4 and 5 h, respectively. The insets in (a2–e2) are the SAED patterns of as-prepared materials.

Fig. 5a and S8† show the TEM images of a dendritic structure on the surface of a piece of nanosheet. The nanocuboids of the first generation (G1) split to produce multiple branches of the second generation (G2, highlighted with white arrows), and therefore form dendric structure. The HRTEM of the branches is composed of d-spacing of 0.23 nm corresponding to (222) facets. The nanocuboids grow along their 〈001〉 directions and exposed in {100} facets of cubic Co₃O₄. In addition, the dendric nanocube as shown in Fig. 4c2 is further illustrated in Fig. 5c. It can be observed that more than three generations of the nanocuboids crossly stack to form the dendric structure. The HRTEM image (Fig. 5d) of the dendritic nanostructure shows that it consists of nanocuboids with d-spacing of 0.28 nm corresponding to (220) facets of cubic Co₃O₄. The results suggest the stepwise splitting growth of the dendric nanocubes, and further verify the single crystalline nature of the hierarchical superstructure. The nanocuboids grow step-by-step on the {100} facets of cubic Co₃O₄ to generate porous nanospheres through partially sharing their (100), (010) and (001) planes.

Fig. 5 (a) TEM and (b) HRTEM images of Co₃O₄ nanosheets after reacting for 2 h. (c) TEM and (d) HRTEM images of a nano-jenga after reacting for 3 h.

Based on the experimental observations, a stepwise splitting mechanism (Fig. 6) can be proposed to understand the formation of the porous and single crystalline Co₃O₄ nanospheres. AOT and cobalt ions first form 2D lamellar assemblies consisting of AOT double-layer and two Co²⁺ layers to form nanoreactors (S1). The Co²⁺ ions then hydrolyze to form Co₃O₄ to form nanosheets together with AOT assemblies (S2), after reacting at 200 °C for 2 h. After prolonging the reaction time to 3 h, the Co₃O₄ nanosheets start to convert into Co₃O₄ dendric nanocubes consisting of nanocuboids for the Ostwald ripening. Because the nucleation and growth of Co₃O₄ could simultaneously take place at multiple sites on the surfaces of the nanosheets and then form nanocuboids, 3D dendric nanocubes (S3) can be constructed at the expense of Co₃O₄ nanosheets. Porous and single crystalline Co₃O₄ nanospheres exposed in {100} facets can be eventually produced by the stepwise splitting growth of nanocuboids (S4 and S5). During the formation of the Co₃O₄ nanospheres, AOT surfactant may have played three roles critical: (1) the first, constructing lamellar assemblies together with Co²⁺ ions as nanoreactors (eqn (2) and (3)); (2) the second, confining the fast growth of Co₃O₄ nanosheets; (3) the third, directing the growth of nanocuboids for selectively binding to the {100} facets of cubic Co₃O₄ on the surfaces of nanosheets to produce porous and single crystalline nanospheres. Due to Co₃O₄ can also be etched by NH₃ and AOT in the system (eqn (4)), the nanospheres change their morphology into dendritic particles after further reacting for longer time. It is worth to note that the 2D nanosheets also play a role critical in the formation of the Co₃O₄ nanospheres. Because the granules are in situ generated on the surfaces of nanosheets, a flat and smooth facet can be produced on the surfaces of some nanospheres to form hemispheres.

Co²⁺ + 2AOT⁻ → Co(AOT)₂ (2)

Co(AOT)₂ + 6NH₃·H₂O → Co(NH₃)₆(AOT)₂ (3)

3Co(NH₃)₆(AOT)₂ + 3H₂O + 1/2O₂ ⇔ Co₃O₄ + 12NH₃ + 6NH₄⁺ + 6AOT⁻¹ (4)

Fig. 6 Stepwise self-assembly growth of Co₃O₄ hemispheres.

Using as-prepared nanospheres as electrode materials, we further made pseudocapacitors and investigated their charge storage performances. Fig. 7a shows cyclic voltammograms (CVs) of the pseudocapacitors recorded at different scan rates between 5 and 50 mV s⁻¹ in 6 M KOH solution. All of the CV curves show obvious pseudocapacitance features consisting of a pairs of redox peaks. The peak currents of the CV curves increase as the scan rate increases, which suggest the good reversibility of the fast charge–discharge behaviours of the materials.

Fig. 7 (a) CV curves of the electrode made with Co₃O₄ nanospheres in 6.0 M KOH at different scan rates. (b) The charge–discharge curves of the supercapacitors made with Co₃O₄ nanospheres measured at different current densities. (c) Average specific capacitance of the devices at various discharge current densities. (d) Average specific capacitance versus cycle number of supercapacitors at a galvanostatic charge–discharge current density of 4 A g⁻¹. The insets in (d) show the galvanostatic charge–discharge curves of the device at a current density of 4 A g⁻¹.

Fig. 7b shows the charge and discharge curves of the pseudocapacitors made with as-prepared materials, measured at different discharge current densities within the potential window of −0.07 to 0.33 V in 6 M KOH solution. The specific capacitances (Fig. 7c) of porous and single crystalline nanospheres evaluated from the discharge curves were 1350, 1230, 1100 and 1000 F g⁻¹ at current densities of 1, 2, 4, and 8 A g⁻¹, respectively. The pseudocapacitors made with porous and single crystalline Co₃O₄ nanospheres exhibit more than 2 times higher ability in storing charges, compared to the devices (Fig. S9†) fabricated with dendritic nanoparticles (Fig. S4–7†) produced by prolonging the reaction time to more than 6 hours. The supercrystals with pores and single crystalline features also show much higher abilities in storing charges in comparison with single crystalline Co₃O₄ nanospheres grown around MWCNTs,^11h 3D hierarchical Co₃O₄ twin-spheres¹ⁱ and 3D hierarchical Co₃O₄ nanoarchitectures consisting of nanowalls we reported previously.^11g In addition, the devices made with porous and single crystalline Co₃O₄ nanospheres also exhibit much higher specific capacitances than those fabricated with Co₃O₄ nanoparticles,¹² hollow nanowire arrays,^1m nanotubes¹³ and mesoporous monolayer.^1f The electrode materials show charge storage ability comparable to RuO₂ materials, which are of high costs and therefore prohibitive for most applications.^5,14 The porous and single crystalline features of the Co₃O₄ nanospheres could facilitate both high ion permeability and rapid electron transportation in the 3D hierarchical structures similar to porous and single crystalline TiO₂ nanospheres.⁹

While the specific capacitances of the devices also decrease as increasing the charge–discharge current density for the increase of the potential drop for the resistance of the nanoparticles and the relatively insufficient Faradic redox of the active material under higher charge–discharge current densities, the value of 1000 F g⁻¹ at 8 A g⁻¹ of the pseudocapacitors only decreases 25.9% (Fig. 7c) in comparison with its specific capacitance of 1350 F g⁻¹ at 1 A g⁻¹. The low fading rate in the capacitance of materials at high galvanostatic current density indicates that the materials allow for the reaction taking place rapidly at high current densities, which can be attributed to the novel porous and single crystalline structure of as-prepared materials also.

Fig. 7d shows the cycling performance of the pseudocapacitors made with Co₃O₄ nanospheres at a current density of 4 A g⁻¹ within the potential window of −0.07 to 0.33 V. After cycling for first 200 times, specific capacitance of the device first decreased from 1100 to 930 F g⁻¹, which could be resulted from the consumption of electrolyte secondary to irreversible reaction between the electrode materials and electrolyte. The capacitance of the device was stable after the initial drop. It was retained at 84.5% of 1100 F g⁻¹, after cycling for 5000 times. The inset in Fig. 7d presents the charge–discharge curves of the devices showing their high reversibility. The porous and single crystalline Co₃O₄ nanospheres are promising candidates for designing novel pseudocapacitors with excellent performances.

Conclusions

In summary, Co₃O₄ nanospheres with porous and single crystalline nature have been produced successfully on a large scale. It was found that the reaction time can dramatically affect the growth of Co₃O₄ superstructures in the nanoreactors constructed with AOT⁻ and Co²⁺ ions. The experimental observations strongly support a stepwise splitting mechanism proposed to understand the formation of the superstructures. The pseudocapacitors made with the porous and single crystalline Co₃O₄ nanospheres exhibit excellent charge-storage performances. The superior electrochemical properties of the materials can be attributed to their superstructures with pores and single crystalline feature, which may significantly increase the rate of both ion diffusion and electron transportation. The approach described here shows four major advantages: the first, novel porous and single crystalline 3D Co₃O₄ supercrystals mainly exposed in {100} facets can be produced on a large scale; the second, the one-pot approach is much more convenient and efficient in synthesizing 3D nanoarchitectures, compared to the topotactic transformation; the third, the formation mechanism proposed could be applied to fabricate other functional materials; and the fourth, the supercrystals exhibit excellent capability in storing charges for their porous and single crystalline features. By choosing a suitable surfactant such as AOT, 3D hierarchical nanoarchitectures with novel microstructures can be fabricated successfully for energy storage.

Acknowledgements

The authors are grateful to the financial support from the National Natural Science Foundation of China (NSFC. 21071130 and 21371157), Outstanding Scholar Program of Henan Province (114200510012) and Key Program of Henan Province for Science and Technology (132102210424).

Notes and references

1. (a) S. L. Xiong, C. Z. Yuan, M. F. Zhang, B. J. Xi and Y. T. Qian, Chem.–Eur. J., 2009, 15, 5320; (b) C. Z. Yuan, L. Yang, L. R. Hou, L. F. Shen, F. Zhang, D. K. Li and X. G. Zhang, J. Mater. Chem., 2011, 21, 18183; (c) X. Wang, A. Sumboja, E. Khoo, C. Y. Yan and P. S. Lee, J. Phys. Chem. C, 2012, 116, 4930; (d) X. M. Liu, Q. Long, C. H. Jiang, B. B. Zhan, C. Li, S. J. Liu, Q. Zhao, W. Huang and X. C. Dong, Nanoscale, 2013, 5, 6525; (e) H. W. Shim, A. H. Lim, J. C. Kim, E. Jang, S. D. Seo, G. H. Lee, T. D. Kim and D. W. Kim, Sci. Rep., 2013, 3, 2325; (f) X. H. Xia, J. P. Tu, X. L. Wang, C. D. Gu and X. B. Zhao, Chem. Commun., 2011, 47, 5786; (g) C. Z. Yuan, L. Yang, L. R. Hou, L. F. Shen, X. G. Zhang and X. W. Lou, Energy Environ. Sci., 2012, 5, 7883; (h) T. Zhu, J. S. Chen and X. W. Lou, J. Mater. Chem., 2010, 20, 7015; (i) Y. H. Xiao, S. J. Liu, F. Li, A. Q. Zhang, J. H. Zhao, S. M. Fang and D. Z. Jia, Adv. Funct. Mater., 2012, 22, 4052; (j) D. W. Wang, Q. H. Wang and T. M. Wang, Inorg. Chem., 2011, 50, 6482; (k) F. L. Meng, Z. G. Fang, Z. X. Li, W. W. Xu, M. J. Wang, Y. P. Liu, J. Zhang, W. R. Wang, D. Y. Zhao and X. H. Guo, J. Mater. Chem. A, 2013, 1, 7235; (l) R. T. Wang, L. B. Kong, J. W. Lang, X. W. Wang, S. Q. Fan, Y. C. Luo and L. Kang, J. Power Sources, 2012, 217, 358; (m) B. Wang, T. Zhu, H. B. Wu, R. Xu, J. S. Chen and X. W. Lou, Nanoscale, 2012, 4, 2145.
2. (a) S. L. Xiong, J. S. Chen, X. W. Lou and H. C. Zeng, Adv. Funct. Mater., 2012, 22, 861; (b) X. L. Xiao, X. F. Liu, H. Zhao, D. F. Chen, F. Z. Liu, J. H. Xiang, Z. B. Hu and Y. D. Li, Adv. Mater., 2012, 24, 5762; (c) Y. Wang, H. Xia, L. Lu and J. Y. Lin, ACS Nano, 2010, 4, 1425; (d) L. Tian, H. L. Zou, J. X. Fu, X. F. Yang, Y. Wang, H. L. Guo, X. H. Fu, C. L. Liang, M. M. Wu, P. K. Shen and Q. M. Gao, Adv. Funct. Mater., 2010, 20, 617; (e) N. Du, H. Zhang, B. Chen, J. B. Wu, X. Y. Ma, Z. H. Liu, Y. Q. Zhang, D. Yang, X. H. Huang and J. P. Tu, Adv. Mater., 2007, 19, 4505; (f) X. W. Lou, D. Deng, J. Y. Lee, J. Feng and L. A. Archer, Adv. Mater., 2008, 20, 258; (g) J. Y. Wang, N. L. Yang, H. J. Tang, Z. H. Dong, Q. Jin, M. Yang, D. Kisailus, H. J. Zhao, Z. Y. Tang and D. Wang, Angew. Chem., Int. Ed., 2013, 52, 6417; (h) L. Hu, N. Yan, Q. W. Chen, P. Zhang, H. Zhong, X. R. Zheng, Y. Li and X. Y. Hu, Chem.–Eur. J., 2012, 18, 8971; (i) K. T. Nam, D. W. Kim, P. J. Yoo, C. Y. Chiang, N. Meethong, P. T. Hammond, Y. M. Chiang and A. M. Belcher, Science, 2006, 312, 885; (j) Y. G. Li, B. Tan and Y. Y. Wu, Nano Lett., 2008, 8, 265.
3. W. Y. Li, L. N. Xu and J. Chen, Adv. Funct. Mater., 2005, 15, 851.
4. (a) L. H. Hu, Q. Peng and Y. D. Li, J. Am. Chem. Soc., 2008, 130, 16136; (b) X. W. Xie, Y. Li, Z. Q. Liu, M. Haruta and W. J. Shen, Nature, 2009, 458, 746; (c) Y. Y. Liang, Y. G. Li, H. L. Wang, J. G. Zhou, J. Wang, T. Regier and H. J. Dai, Nat. Mater., 2011, 10, 780.
5. (a) J. R. Miller and P. Simon, Science, 2008, 321, 651; (b) P. Simon and Y. Gogotsi, Nat. Mater., 2008, 7, 845.
6. C. C. Li, X. M. Yin, L. B. Chen, Q. H. Li and T. H. Wang, Chem.–Eur. J., 2010, 16, 5215.
7. H. Huang, W. J. Zhu, X. Y. Tao, Y. Xia, Z. Y. Yu, J. W. Fang, Y. P. Gan and W. K. Zhang, ACS Appl. Mater. Interfaces, 2012, 4, 5974.
8. Y. G. Li, B. Tan and Y. Y. Wu, J. Am. Chem. Soc., 2006, 128, 14258.
9. N. N. Edward, J. W. Crossland, V. Sivaram, T. Leijtens, J. A. Alexander-Webber and H. J. Snaith, Nature, 2013, 495, 215.
10. J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon and P. L. Taberna, Science, 2006, 313, 1760.
11. (a) F. Li, M. Zhu, C. G. Liu, W. L. Zhou and J. B. Wiley, J. Am. Chem. Soc., 2006, 128, 13342; (b) F. Li, J. B. He, W. L. L. Zhou and J. B. Wiley, J. Am. Chem. Soc., 2003, 125, 16166; (c) F. Li, X. Badel, J. Linnros and J. B. Wiley, J. Am. Chem. Soc., 2005, 127, 3268; (d) F. Li, Y. Ding, P. X. X. Gao, X. Q. Xin and Z. L. Wang, Angew. Chem., Int. Ed., 2004, 43, 5238; (e) F. Li, F. L. Gong, Y. H. Xiao, A. Q. Zhang, J. H. Zhao, S. M. Fang and D. Z. Jia, ACS Nano, 2013, 7, 10482; (f) Y. H. Xiao, A. Q. Zhang, S. J. Liu, J. H. Zhao, S. M. Fang, D. Z. Jia and F. Li, J. Power Sources, 2012, 219, 140; (g) Y. H. Xiao, S. J. Liu, S. M. Fang, D. Z. Jia, H. Q. Su, W. L. Zhou, J. B. Wiley and F. Li, RSC Adv., 2012, 2, 3496; (h) Y. Xiao, A. Zhang, D. Jia and F. Li, Mater. Res. Soc. Symp. Proc., Boston, MA, 2013, p. 1497.
12. M. B. Zheng, J. Cao, S. T. Liao, J. S. Liu, H. Q. Chen, Y. Zhao, W. J. Dai, G. B. Ji, J. M. Cao and J. Tao, J. Phys. Chem. C, 2009, 113, 3887.
13. G. X. Tong, J. G. Guan and Q. J. Zhang, Adv. Funct. Mater., 2013, 23, 2406.
14. S. W. Lee, B. M. Gallant, H. R. Byon, P. T. Hammond and S. H. Yang, Energy Environ. Sci., 2011, 4, 1972.

---

Footnote

† Electronic supplementary information (ESI) available: TEM, HRTEM and FSEM images. See DOI: 10.1039/c5ra02739j

---