Hollow NiCo₂S₄ nanotube arrays grown on carbon textile as a self-supported electrode for asymmetric supercapacitors

Abstract

Although supercapacitors possess a fast charge/discharge capability, the practical application of supercapacitors is still hindered largely by their low energy density. Improving the electrochemical performance of supercapacitors depends largely on the development of novel electrode materials and hybrid systems. In this work, hollow NiCo₂S₄ nanotube arrays are successfully grown on carbon textile (CT) with robust adhesion through a two-step synthesis, involving the growth of a solid nanowire precursor and subsequent conversion into NiCo₂S₄ nanotubes using a sulfidation process. Using CT-supported NiCo₂S₄ nanotube arrays as the positive electrode and activated carbon as the negative electrode, a high-performance asymmetric supercapacitor with a maximum voltage of 1.6 V has been fabricated, which manifests high energy density (∼40.1 W h kg⁻¹ at 451 W kg⁻¹), high power density (∼4725 W kg⁻¹ at 21 W h kg⁻¹) and excellent cyclability.

---

1. Introduction

Developing high-performance electrochemical energy storage devices has become an urgent task for scientists and engineers worldwide.^1–4 Among various electric energy storage technologies, supercapacitors have attracted a considerable amount of attention due to their high power density, fast charge/discharge capability (in seconds) and exceptionally long cycle life. However, the energy density of commercial supercapacitors (usually less than 10 W h kg⁻¹) is much lower than that of rechargeable batteries, which greatly limits their practical applications.^5–8 Therefore, one great challenge for supercapacitor technology is to improve the energy density without sacrificing the power density and cycle life.

According to the equation E = 1/2CV², the energy density of supercapacitors can be greatly enhanced by maximizing the specific capacitance (C) and/or the cell voltage (V). Recently, designing asymmetric supercapacitors (ASCs) has been regarded as a promising strategy to increase the cell voltage and thus the energy density.^9–14 High-energy density ASCs mostly consist of a battery-type faradaic electrode favoring energy density and a capacitive electrode favoring power density. ASCs can achieve a high output voltage by combining the different potential windows of two electrodes, which can in principle lead to a significant improvement in the energy density. The key for developing high-performance ASCs with high power and energy densities relies on the development of novel nanostructured electrodes with a rational design and compatible electrolyte.

Traditionally, pseudocapacitive transition metal oxide (such as Co₃O₄,¹⁵ NiO¹⁶ and MnO₂)^17,18 materials are used as positive electrode materials for ASCs. However, one main drawback that restricts the practical application of transition metal oxides is their low electrical conductivity, which usually results in a poor rate capability and cycling stability of the cells. Recently, transition metal sulfides such as cobalt sulfides and nickel sulfides have been investigated as a new type of electrode material for supercapacitors.^19,20 It is reported that ternary NiCo₂S₄ possesses a higher electrochemical activity and higher capacitance than corresponding mono-metal sulfides because of the occurrence of more feasible redox reactions.^21–23 More significantly, NiCo₂S₄ exhibits an excellent electrical conductivity that is at least two orders of magnitude higher than that of NiCo₂O₄.^24,25 Several different types of NiCo₂S₄ nanostructures, including nanoparticles,²⁶ nanoplates,^27,28 nanotubes,^21,29 microspheres,²⁴ hollow spheres,³⁰ and nanocomposites with graphene,^26,31,32 have been recently synthesized and their electrochemical performance investigated. In most cases, nanostructured NiCo₂S₄ materials need to be mixed with additives such as a binder and carbon black and further pressed onto a current collector. This electrode-making process not only makes a large portion of the electroactive NiCo₂S₄ surface inaccessible to the electrolyte, but also unavoidably leads to a reduced energy density for the device because of the significant weight of the auxiliary components.³³ This drawback can partly be overcome by directly growing nanostructured electrode materials on a current-collecting substrate.³⁴ For example, NiCo₂S₄ nanostructures have recently been grown on nickel foam.^32,35,36 However, nanostructured electrodes supported on a metallic current collector are quite rigid in nature, which makes the devices less flexible, and more importantly limits the energy density.

In this work, we develop a high-performance ASC using NiCo₂S₄ nanotube arrays (NTAs) grown on CT as the positive electrode and activated carbon (AC) as the negative electrode. NiCo₂S₄ NTAs are successfully grown on CT with robust adhesion using a surfactant-assisted hydrothermal method followed by a sulfidation process. The resultant NiCo₂S₄/CT hybrid film can directly serve as an additive-free positive electrode for the ASC, which shows a high rate capability (1004 F g⁻¹ at 20 A g⁻¹) and an excellent cycling stability. Remarkably, the ASC based on NiCo₂S₄/CT and AC manifests an energy density of 40.1 W h kg⁻¹ at a power density of 451 W kg⁻¹.

2. Experimental section

Growth of NiCo₂S₄ nanotube arrays (NTAs) on carbon textile

Carbon textile (CT) was cleaned through ultrasonication in deionized (DI) water and ethanol for 15 min each, and was then dried in an oven. In a typical process, 5 mmol of CoCl₂·6H₂O, 2.5 mmol of NiCl₂·6H₂O, 2 mmol of hexadecyl trimethyl ammonium bromide, and 9 mmol of urea are dissolved in 50 mL of DI water to form a transparent pink solution. After adding a piece of cleaned carbon cloth (4 cm × 4 cm), the solution was then transferred to a Teflon-lined stainless steel autoclave and kept at 100 °C. After hydrothermal growth, carbon textile covered with NiCo-precursor nanowire arrays (NWAs) was taken out and carefully rinsed several times with deionized water. Then, the NiCo-precursor NWAs grown on CT were immersed in 0.07 M Na₂S solution and kept at 160 °C for 8 h. After cooling down naturally to room temperature, carbon textile covered with NiCo₂S₄ NTAs was taken out and washed with DI water and ethanol several times, then dried at 60 °C for 12 h. NiCo₂S₄ nanotube-assembled urchin-like structures were also prepared in a similar manner without adding a carbon textile substrate.

Materials characterization

The crystal structure of the obtained samples was characterized using X-ray diffraction (XRD) (Bruker D8 advance) with Cu Kα radiation. X-ray photoelectron spectroscopy (XPS) analysis was performed on a Perkin-Elmer PHI 550 spectrometer with Al Kα (1486.6 eV) as the X-ray source. The microstructures were characterized using transmission electron microscopy (TEM) (TEM, FEI, Tecnai-20, USA), high-resolution transmission electron microscopy (HRTEM, JEOL JEM-2010), and field-emission scanning electron microscopy (FESEM, JEOL, JSM-7000). The nitrogen sorption measurement was carried out on an Autosorb 6B at liquid nitrogen temperature.

Electrochemical measurement

The electrochemical measurements were conducted in a three-electrode electrochemical cell with a Pt counter electrode and a saturated calomel reference electrode in 6 M KOH solution. The carbon textile-supported NiCo₂S₄ NTAs directly served as the working electrode (NiCo₂S₄ mass ≈ 1.8 mg cm⁻²) without any ancillary materials. Besides, the mass percentage of NiCo₂S₄ in the final composite is estimated to be 11%. For the electrochemical measurements of NiCo₂S₄ nanotube-assembled urchin-like structures, the working electrode was prepared by mixing the active material, acetylene black and polytetrafluorene-ethylene (PTFE) binder with a weight ratio of 80 : 15 : 5. After coating the above slurry onto Ni foam, the electrode was dried at 60 °C for 10 h before pressing under a pressure of 10 MPa. The specific capacitance is calculated using the following equation:where I is the discharge current, Δt is the discharge time, ΔV is the voltage range and m is the mass of the active material.

Supercapacitor devices

Asymmetric supercapacitors (ASCs) were fabricated by assembling the NiCo₂S₄/CT cathode and AC anode with one piece of cellulose paper as the separator in two-electrode simulation cells. 6 M KOH solution was employed as the electrolyte. The mass ratio of positive electrode to negative electrode was determined according to the well-known charge balance principle (q⁺ = q⁻). In this relationship, the charge stored by each electrode usually depends on the specific capacitance (C), the potential range for the charge/discharge process (ΔV), and the mass of the electrode (m) described by the following equation:

q = C × ΔV × m (2)

In order to obtain q⁺ = q⁻, the mass balancing can be expressed by the following equation (eqn (3)):

C₊ and C₋ are the specific capacitance of the NiCo₂S₄/CT and AC electrodes, respectively. ΔV₊ and ΔV₋ are the voltage range of one scanning segment (V) of the NiCo₂S₄/CT and AC electrodes, respectively.

The energy density (E) and power density (P) of ASCs are calculated according to the following equations:

where I is the discharging current, V_(t) is the discharging voltage excluding the IR drop, dt is the time differential, M is the total mass of the active electrode materials, and Δt is the discharging time. If the discharge voltage profiles are approximately linear, eqn (4) can be easily reduced to E = 1/2CV² as usual.

3. Results and discussion

The growth procedure and resulting electrode architectures are schematically illustrated in Fig. 1 (see the method for the detailed synthesis). Woven by carbon microfibers (Fig. S1, see the ESI†), lightweight CT is a unique substrate for the controlled growth of NiCo₂S₄ NTAs with high flexibility and high conductivity. In the first step, highly ordered Ni–Co precursor nanowire arrays (NWAs) are easily grown on CT under hydrothermal conditions. From the X-ray diffraction (XRD) pattern (Fig. S2a, see the ESI†), it is clear that the precursor can be indexed as the (Ni, Co) (CO₃)_1/2OH·0.11H₂O phase.³⁷ The precursor NWAs grow uniformly on each carbon microfiber to form a large-scale conformal coating with a well-established textile structure (Fig. S3, see the ESI†). After that, the (Ni, Co) (CO₃)_1/2OH precursor is transformed into NiCo₂S₄ as confirmed using XRD analysis (Fig. S2b, see the ESI†) by reacting with Na₂S under hydrothermal conditions. The accompanying structural evolution from nanowire to nanotube can be ascribed to some diffusion effect, which has been extensively used to synthesize various hollow micro/nanostructures.^38,39 In the earlier stage of the hydrothermal reaction process, S²⁻ ions react with metal ions to form a thin layer of Ni–Co sulfides at the surface of the nanowires, which hinders the direct chemical reaction between the outside sulfide ions and inner metal-ion species. Then the inner metal ions spontaneously diffuse to the external surface of the nanowires for further chemical reaction and the growth of NiCo₂S₄. The continuous outward diffusion of metal-ion species results in the generation of void space inside the starting nanowires. In comparison, NiCo₂S₄ nanotube-assembled urchin-like structures will be formed following the same procedure without the addition of a CT substrate (Fig. S4, see the ESI†).

Fig. 1 Schematic illustration of the formation of NiCo₂S₄ NTAs on CT.

The as-prepared NiCo₂S₄/CT composite is characterized using XRD and X-ray photoelectron spectroscopy (XPS). From the XRD pattern (Fig. S5, see the ESI†), except for the reflections from CT, all the other peaks can be indexed to the cubic NiCo₂S₄ phase (JCPDS card no. 43-1477) although the peak intensity is relatively weak. No residues or impurity phases are detected, indicating that the (Ni, Co)(CO₃)_1/2OH precursor is completely converted to the NiCo₂S₄ phase after sulfidation. The XPS survey spectrum (Fig. 2a) indicates the presence of Ni, Co, and S, as well as C. By using a Gaussian fitting method, the Co 2p spectrum (Fig. 2b) is best fitted with two spin–orbit doublets, characteristic of Co²⁺ and Co³⁺, and one shake-up satellite (indicated as “Sat.”). The Ni 2p spectrum (Fig. 2c) is also best fitted with two spin–orbit doublets, characteristic of Ni²⁺ and Ni³⁺, and two shake-up satellites. The S 2p spectrum (Fig. 2d) can be divided into two main peaks and one shake-up satellite. The peak at 163.8 eV, corresponding to S 2p_3/2, is typical of metal–sulfur bonds.²⁴ The peak at 162.1 eV, corresponding to S 2p_1/2, may be attributed to S²⁻ in low coordination at the surface. Detailed analysis of the high-resolution spectra shows that the sample contains Co²⁺, Co³⁺, Ni²⁺, Ni³⁺ and S²⁻, which is in good agreement with the results in the literature for NiCo₂S₄.^24,27,40

Fig. 2 (a) XPS survey spectrum, and high-resolution XPS spectra of (b) Co 2p, (c) Ni 2p and (d) S 2p for the NiCo₂S₄ nanotubes scratched from the CT.

After the sulfidation conversion, the formed NiCo₂S₄/CT composite still keeps the ordered woven structure of the CT substrate (Fig. 3a). The inset in Fig. 3a shows digital photographs of NiCo₂S₄/CT that can be flexed, demonstrating good flexibility. Fig. 3b and c show the enlarged scanning electron microscopy (SEM) images of the NiCo₂S₄/CT composite. The NiCo₂S₄ nanotubes with diameters of 100–150 nm and lengths of 3–6 μm are grown vertically on the carbon microfiber, which completely retains the array structure of the (Ni, Co)(CO₃)_1/2OH precursor. The existence of hollow NiCo₂S₄ nanotubes is further confirmed using transmission electron microscopy (TEM). As shown in Fig. 3d and e, a tubular structure is clearly observed with a wall thickness of about 15 nm. Fig. 3e reveals that a typical NiCo₂S₄ nanotube is actually composed of many small nanoparticles 10–20 nm in size, which is also confirmed by the high-resolution TEM image (Fig. 3f). The observed interplanar spacings are measured to be 0.28 and 0.54 nm, which respectively match well with the (311) and (111) lattice planes of the NiCo₂S₄ phase. As determined using N₂ sorption measurements (Fig. S6, see the ESI†), the NiCo₂S₄ nanotubes possess a Brunauer–Emmett–Teller (BET) specific surface area of about 50.2 m² g⁻¹ with a pore size mostly below 10 nm. The pores may be formed by the stacking of nanoparticles. The hollow NiCo₂S₄ nanotubes provide a larger surface area (both interior and exterior) and enable the facile transport of ions. This will benefit the fast charge transport when used as supercapacitor electrode materials.

Fig. 3 (a, b) Low and (c) high magnification SEM images of the NiCo₂S₄/CT composite, showing NiCo₂S₄ nanotubes conformally grown on the carbon microfibers. (d, e) Low and (f) high magnification TEM images of one NiCo₂S₄ nanotube scratched from the CT.

The electrochemical performance of NiCo₂S₄/CT as an additive-free working electrode was first evaluated using a three-electrode cell configuration, and the results are shown in Fig. 4. Fig. 4a shows the cyclic voltammetry (CV) curves of the self-supported NiCo₂S₄/CT electrode in 6 M KOH aqueous electrolyte at various scan rates ranging from 5 to 50 mV s⁻¹. One oxidation peak and two reduction peaks are visible in all the CV curves. With the increase in sweep rate from 5 to 50 mV s⁻¹, the general shape of all the CV curves evolves insignificantly with some shift in the peak positions. These distinct peaks can be attributed to the reversible faradaic redox processes of Co²⁺/Co³⁺/Co⁴⁺ and Ni²⁺/Ni³⁺ based on the following reactions:^21,22

CoS + OH⁻ ⇄ CoSOH + e⁻

CoSOH + OH⁻ ⇄ CoSO + H₂O + e⁻

NiS + OH⁻ ⇄ NiSOH + e⁻

Fig. 4 (a) CV curves and (b) galvanostatic charge/discharge voltage profiles of the NiCo₂S₄/CT electrode. (c) Specific capacitance as a function of the current density. (d) Cycling performance of the NiCo₂S₄/CT electrode at a current density of 10 A g⁻¹.

Galvanostatic charge/discharge curves of the NiCo₂S₄/CT electrode at various current densities ranging from 1 to 20 A g⁻¹ are shown in Fig. 4b. Consistent with the CV results, the poorly defined voltage plateaus in the charge/discharge curves suggest the presence of some faradaic processes. The calculated specific capacitance from the discharge curves as a function of the current density is plotted in Fig. 4c. Encouragingly, the NiCo₂S₄/CT electrode manifests an excellent pseudocapacitance of 1279, 1262, 1197, 1123 and 1004 F g⁻¹ at current densities of 1, 2, 5, 10 and 20 A g⁻¹, respectively. The carbon textile contributes very little capacitance, which could be neglected (Fig. S7†). This means that ∼78% of the specific capacitance at 1 A g⁻¹ is still retained when the discharge current density is increased to 20 A g⁻¹. Compared to the NiCo₂S₄/CT electrode, the normal NiCo₂S₄ electrode exhibits an inferior electrochemical performance when characterized under similar conditions (Fig. S8, see the ESI†). As shown in Fig. S8a,† the CV curves of the NiCo₂S₄ electrode have been obviously changed. Also shown in Fig. 4c for comparison, the specific capacitance of the NiCo₂S₄ nanotube electrode is only 476 F g⁻¹ at 20 A g⁻¹, corresponding to only 47.4% of the capacitance achieved by the NiCo₂S₄/CT electrode at the same current density. A charge/discharge cycling test at a current density of 10 A g⁻¹ is carried out to examine the cycling stability of the NiCo₂S₄/CT electrode (Fig. 4d). The specific capacitance of the NiCo₂S₄/CT electrode decreases from 1123 to 1028 F g⁻¹, corresponding to a capacitance retention of ∼92% after continuous cycling for 2000 cycles. The above results of high capacitance and excellent rate capability directly reveal the great advantages of the NiCo₂S₄/CT electrode. The above results also indicate that this rationally designed NiCo₂S₄/CT electrode is apparently superior to many other NiCo₂S₄ electrodes,^21,22,24 and mono-metal sulfides^19,20,41,42 (Table S1, see the ESI†).

To further evaluate the NiCo₂S₄/CT electrode for practical applications, an ASC device is fabricated using the NiCo₂S₄/CT electrode as the cathode and activated carbon (AC) as the anode in 6 M KOH aqueous electrolyte, with one piece of cellulose paper as the separator. A series of CV and charge/discharge measurements with varying voltage windows was carried out to determine the optimum operating voltage window of the ASC (Fig. 5a and b). With an increase in the operating potential window to 1.6 V, there is more faradaic reaction observed (a larger current response). Accordingly, the specific capacitance based on the total mass of the active materials in both electrodes increases greatly from 52.5 to 98.4 F g⁻¹ with the increase in the operating potential window from 1.0 to 1.6 V (Fig. 5b). It is true that the energy density of supercapacitors can be remarkably improved by increasing the cell voltage. However, when the system is tested at a voltage of 1.8 V or even higher voltages, serious limitations related to the electrolysis of water are observed. Thus, we choose an operating voltage window of 1.6 V to further evaluate the overall electrochemical performance of this optimized ASC.

Fig. 5 (a) CV curves of the NiCo₂S₄/CT//AC ASC measured at different potential windows (at 10 mV s⁻¹). (b) Specific capacitance of the ASC with the increase in the potential window at a fixed scan rate of 10 mV s⁻¹.

Fig. 6a presents the typical CV curves for the optimized NiCo₂S₄/CT//AC ASC device at various scan rates between 0 and 1.6 V. From the CV curves, it is clear that both the electric double-layer capacitance and pseudocapacitance are significant. There is no obvious distortion in the CV curves even at a high scan rate of 100 mV s⁻¹, indicating the good fast-charge/discharge properties of the device. From the CV curves, the specific capacitance is calculated based on the total mass of the active materials in both electrodes, as shown in Fig. 6b. The specific capacitance is 111.2 F g⁻¹ at a low scan rate of 5 mV s⁻¹. As the scan rate increased from 10 to 20 and 50 mV s⁻¹, the specific capacitance slightly decreased from 98.4 to 91.5 and 72.9 F g⁻¹, respectively. It should be noted that the specific capacitance still retains 59.2 F g⁻¹ even at a high scan rate of 100 mV s⁻¹, indicating the excellent rate capability. The galvanostatic charge/discharge curves of the ASC at various current densities are shown in Fig. 6c. The charge and discharge curves retain good symmetry at a cell voltage as high as 1.6 V, implying that the cell has excellent electrochemical reversibility and capacitive characteristics. Fig. 6d further reveals the outstanding cycle life of the ASC up to 3000 cycles. The specific capacitance retention is ∼89.2% after 3000 cycles at 5 A g⁻¹.

Fig. 6 (a) CV curves of the ASC device at various scan rates from 5 to 100 mV s⁻¹ measured between 0 and 1.6 V. (b) The specific capacitance as a function of the scan rate. (c) Galvanostatic charge/discharge voltage profiles of the ASC device at different current densities from 1 to 20 A g⁻¹. (d) The specific capacitance of the ASC device as a function of current density. (d) Cycling performance of the ASC device at a current density of 5 A g⁻¹.

We have further evaluated the energy density and power density of the NiCo₂S₄/CT//AC ASC based on the total mass of the active materials in both electrodes. From the Ragone plot (Fig. 7), the NiCo₂S₄/CT//AC ASC displays a high energy density of 40.1 W h kg⁻¹ at a power density of 451 W kg⁻¹. Even at a high power density of 4725 W kg⁻¹, the ASC still delivers an energy density of 21 W h kg⁻¹. The superior electrochemical performance of the NiCo₂S₄/CT//AC ASC might be attributed to the following design features. The positive electrode structure allows each NiCo₂S₄ nanotube to have its own electrical contact with the carbon fiber substrate. The abundant mesopores in the nanotube and large open space between neighboring nanotubes allow facile ion diffusion. This ensures the efficient participation of each NiCo₂S₄ nanotube in faradaic electrochemical reactions.

Fig. 7 Ragone plots of the ASC device based on the total mass of the active materials in both electrodes.

4. Conclusion

In summary, we have rationally designed and fabricated an asymmetric supercapacitor using NiCo₂S₄ nanotube arrays on carbon textile as the positive electrode and AC as the negative electrode. The optimal asymmetric supercapacitor device can be reversibly charged and discharged at an operating voltage of 1.6 V in KOH aqueous electrolyte. Importantly, the supercapacitor is able to deliver a high energy density of 40.1 W h kg⁻¹ at a power density of 451 W kg⁻¹ and a high power density of 4725 W kg⁻¹ at an energy density of 21 W h kg⁻¹ (based on the total mass of the active materials in both electrodes). The device can retain ∼89.2% of the capacitance after 3000 continuous charge/discharge cycles. The present encouraging results show that mixed metal sulfides with a high electrochemical activity might be promising for high-performance asymmetric supercapacitors with a high voltage, and high energy and power densities.

Acknowledgements

This work was supported by the National Basic Research Program of China (973 Program) (No. 2014CB239701), National Natural Science Foundation of China (No. 51372116, 51504139), Natural Science Foundation of Jiangsu Province (BK2011030, BK20150739) and the Fundamental Research Funds for the Central Universities (No. NE2016005, NJ20150028).

References

1. J. R. Miller and P. Simon, Science, 2008, 321, 651–652.
2. B. Z. Tian and C. M. Lieber, Abstracts of Papers of the American Chemical Society, 2009, 238.
3. S. Chu and A. Majumdar, Nature, 2012, 488, 294–303.
4. G. H. Yu, L. B. Hu, M. Vosgueritchian, H. L. Wang, X. Xie, J. R. McDonough, X. Cui, Y. Cui and Z. N. Bao, Nano Lett., 2011, 11, 2905–2911.
5. J. S. Huang, B. G. Sumpter and V. Meunier, Angew. Chem., Int. Ed., 2008, 47, 520–524.
6. X. H. Lu, M. H. Yu, G. M. Wang, T. Zhai, S. L. Xie, Y. C. Ling, Y. X. Tong and Y. Li, Adv. Mater., 2013, 25, 267–272.
7. W. Fu, C. Zhao, W. Han, Y. Liu, H. Zhao, Y. Ma and E. Xie, J. Mater. Chem. A, 2015, 3, 10492–10497.
8. W. Chen, C. Xia and H. N. Alshareef, ACS Nano, 2014, 8, 9531–9541.
9. L. F. Chen, Z. H. Huang, H. W. Liang, Q. F. Guan and S. H. Yu, Adv. Mater., 2013, 25, 4746–4752.
10. A. Sumboja, C. Y. Foo, X. Wang and P. S. Lee, Adv. Mater., 2013, 25, 2809–2815.
11. C. L. Long, T. Wei, J. Yan, L. L. Jiang and Z. J. Fan, ACS Nano, 2013, 7, 11325–11332.
12. H. B. Li, M. H. Yu, F. X. Wang, P. Liu, Y. Liang, J. Xiao, C. X. Wang, Y. X. Tong and G. W. Yang, Nat. Commun., 2013, 4, 1894.
13. Z. Tang, C. H. Tang and H. Gong, Adv. Funct. Mater., 2012, 22, 1272–1278.
14. X. F. Wang, B. Liu, R. Liu, Q. F. Wang, X. J. Hou, D. Chen, R. M. Wang and G. Z. Shen, Angew. Chem., Int. Ed., 2014, 53, 1849–1853.
15. C. Zhou, Y. W. Zhang, Y. Y. Li and J. P. Liu, Nano Lett., 2013, 13, 2078–2085.
16. H. W. Wang, H. Yi, X. Chen and X. F. Wang, J. Mater. Chem. A, 2014, 2, 3223–3230.
17. B. G. Choi, M. Yang, W. H. Hong, J. W. Choi and Y. S. Huh, ACS Nano, 2012, 6, 4020–4028.
18. L. B. Hu, W. Chen, X. Xie, N. A. Liu, Y. Yang, H. Wu, Y. Yao, M. Pasta, H. N. Alshareef and Y. Cui, ACS Nano, 2011, 5, 8904–8913.
19. X. H. Xia, C. R. Zhu, J. S. Luo, Z. Y. Zeng, C. Guan, C. F. Ng, H. Zhang and H. J. Fan, Small, 2014, 10, 766–773.
20. T. Zhu, Z. Y. Wang, S. J. Ding, J. S. Chen and X. W. Lou, RSC Adv., 2011, 1, 397–400.
21. H.Z. Wan, J. J. Jiang, J. W. Yu, K. Xu, L. Miao, L. Zhang, H. C. Chen and Y. J. Ruan, CrystEngComm, 2013, 15, 7649–7651.
22. L. Yu, L. Zhang, H. B. Wu and X. W. Lou, Angew. Chem., Int. Ed., 2014, 53, 3711–3714.
23. L. Shen, J. Wang, G. Xu, H. Li, H. Dou and X. Zhang, Adv. Energy Mater., 2015, 5, 1400977.
24. H. C. Chen, J. J. Jiang, L. Zhang, H. Z. Wan, T. Qi and D. D. Xia, Nanoscale, 2013, 5, 8879–8883.
25. J. W. Xiao, L. Wan, S. H. Yang, F. Xiao and S. Wang, Nano Lett., 2014, 14, 831–838.
26. Y. Xiao, D. Su, X. Wang, L. Zhou, S. Wu, F. Li and S. Fang, Electrochim. Acta, 2015, 176, 44–50.
27. J. Pu, F. L. Cui, S. B. Chu, T. L. Wang, E. H. Sheng and Z. H. Wang, ACS Sustainable Chem. Eng., 2014, 2, 809–815.
28. Z. Wu, X. Pu, X. Ji, Y. Zhu, M. Jing, Q. Chen and F. Jiao, Electrochim. Acta, 2015, 174, 238–245.
29. J. Zhang, H. Gao, M. Zhang, Q. Yang and H. Chuo, Appl. Surf. Sci., 2015, 347, 690–695.
30. L. Shen, L. Yu, H. B. Wu, X.-Y. Yu, X. Zhang and X. W. D. Lou, Nat. Commun., 2015, 6, 6694.
31. S. J. Peng, L. L. Li, C. C. Li, H. T. Tan, R. Cai, H. Yu, S. Mhaisalkar, M. Srinivasan, S. Ramakrishna and Q. Y. Yan, Chem. Commun., 2013, 49, 10178–10180.
32. C. Lamiel and J.-J. Shim, Electrochim. Acta, 2015, 161, 351–357.
33. L. Shen, Q. Che, H. Li and X. Zhang, Adv. Funct. Mater., 2014, 24, 2630–2637.
34. L. F. Shen, E. Uchaker, X. G. Zhang and G. Z. Cao, Adv. Mater., 2012, 24, 6502–6506.
35. H. Wang, C. Wang, C. Qing, D. Sun, B. Wang, G. Qu, M. Sun and Y. Tang, Electrochim. Acta, 2015, 174, 1104–1112.
36. H. C. Chen, J. J. Jiang, L. Zhang, D. D. Xia, Y. D. Zhao, D. Q. Guo, T. Qi and H. Z. Wan, J. Power Sources, 2014, 254, 249–257.
37. J. W. Xiao and S. H. Yang, RSC Adv., 2011, 1, 588–595.
38. Y. D. Yin, R. M. Rioux, C. K. Erdonmez, S. Hughes, G. A. Somorjai and A. P. Alivisatos, Science, 2004, 304, 711–714.
39. H. L. Cao, X. F. Qian, C. Wang, X. D. Ma, J. Yin and Z. K. Zhu, J. Am. Chem. Soc., 2005, 127, 16024–16025.
40. J. Pu, T. T. Wang, H. Y. Wang, Y. Tong, C. C. Lu, W. Kong and Z. H. Wang, ChemPlusChem, 2014, 79, 577–583.
41. H. Pang, C. Z. Wei, X. X. Li, G. C. Li, Y. H. Ma, S. J. Li, J. Chen and J. S. Zhang, Sci. Rep., 2014, 4, 3577.
42. J. Yang, X. Duan, W. Guo, D. Li, H. Zhang and W. Zheng, Nano Energy, 2014, 5, 74–81.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra24068a

---