Highly conductive, twistable and bendable polypyrrole–carbon nanotube fiber for efficient supercapacitor electrodes

Abstract

Carbon nanotubes (CNTs) are a promising candidate for flexible and wearable electronic applications. Here, we report a conductive, twistable and bendable CNT composite fiber with core–shell structure for efficient supercapacitor electrode, where the CNT bundles are coated with a thin layer of polypyrrole (PPy) (denoted as PPy@CNT). The electrochemical properties of the fiber electrode depend on the content and distribution of the PPy. The PPy@CNT fiber exhibits a high specific capacitance of 350 F g⁻¹ and high stability in cyclic testing. The outstanding electrochemical performance originates from the high conductivity of the core–shell structured PPy@CNT composite, which makes charge transfer easily between PPy and CNTs. The PPy@CNT fiber is also stable under bending and twisting.

---

1. Introduction

Recently, flexible supercapacitors based on carbon nanomaterials have gained great attention due to their potential applications as a portable power supply for wearable electronics.^1–4 The flexible supercapacitor might be frequently stretched, bent or twisted during practical application. It requires the supercapacitor electrode not only to have high electrochemical performance, such as high specific capacitance, fast charge–discharge capacity and good cyclic stability, but also to possess flexible behavior, for instance being stretchable, compressible, bendable, and even twistable.^5,6 Conductive fibers could serve as such an attractive supercapacitor electrode. Various materials and their mixtures, including carbon fibers (CFs), carbon nanotubes (CNTs), graphene and metal wires, have been used as conductive fiber electrodes.^7–27 Among these materials, CFs have high mechanical properties and low conductivity; metal wires have high conductivity (∼10⁷ S m⁻¹) and low specific surface area; while CNTs and graphene have relatively high conductivity (∼10³ to 10⁵ S m⁻¹), high specific surface area, abundant porous structure, and chemical stability. The specific capacitance of CNT and graphene fiber electrodes are usually higher than those of the CFs,^7–24 while metal wire electrodes have exhibited high rate capacity.^25–27 For a power supply device, highly conductive electrode is always desired for a low internal resistance, so as to reduce internal energy loss.

The pure CNTs behave as faradic double-layered capacitors, suffering from relative low specific capacitance.^27–29 In order to improve the specific capacitance of the CNT electrodes, high pseudocapacitance materials, such as conductive polymers, and metal oxides, were introduced.^30–41 By combining with conductive polymers, the specific capacitance of the electrodes were enhanced significantly, reaching to 200–550 F g⁻¹.^30–39 As one of conventional conducting polymer, polypyrrole (PPy) has been actively investigated in supercapacitors due to its high pseudocapacitance, low cost and ease of fabrication.^6,30–36 The PPy can provide extra capacitance for energy storage; on the other hand, the PPy also increased the internal resistance due to its poor conductivity. It needs to balance the amount of pseudocapacitance and internal resistance of the electrodes by controlling the content and distribution of the PPy. To date, most works on the hybrid CNT fibers were prepared by depositing conductive polymer on the fibers directly, which the polymer distribute mainly on the outer of the CNT fiber.^{12,17–20,33,34} Furthermore, the flexibility of the fiber supercapacitors were usually illustrated by bending the devices to various angles, interweaving or knitting.^{12,16,18,20,24} It needs to investigate the electrochemical stability of the fibers under bending and twisting due to the existence of interface between the PPy and CNTs.

Here, we present a high efficiency fiber supercapacitor using core–shell structured composite of PPy and CNTs (denoted as PPy@CNTs) as electrode. The PPy@CNT fiber is prepared from electrochemically depositing PPy on the macroscopic CNT film followed by fiber spinning. The flexible PPy@CNT fiber has high conductivity, outstanding electrochemical properties, and high stability in bending and twisting tests.

2. Experimental

Preparation of PPy@CNT fiber

Fig. 1a shows a schematic diagram of the preparation of the PPy@CNT fiber. Macroscopic CNT films used in the experiment were prepared and collected continuously with a wheel rotating perpendicular to the gas flow by an improved floating chemical deposition method.⁴⁴ The CNT film comprises mainly single-walled and double-walled CNT bundles with diameter of about 30 nm (see Fig. S1a and S1b†). The CNT film was unwrapped from the wheel and tailored into smaller one with desired dimension. The amount of CNTs used in the sample were about 5 mg. The CNT film was first immersed in 30 wt% H₂O₂ solution for 2 days and in 37 wt% HCl solution for 1 day to remove the amorphous carbon and catalyst particles.

Fig. 1 (a) Schematic diagram of the preparation of PPy@CNT fibers. (b) Optical image of a CNT film placed in between two stainless steel wire meshes. (c) A PPy@CNT fiber with length over 1 meter. (d) Letters of THU made from a long PPy@CNT fiber.

The PPy was then deposited onto the CNT film by a three-electrode electrochemical deposition method, where CNT film, platinum wire and saturated calomel electrode (SCE) were used as working, counter and reference electrodes, respectively. A potential of 0.7 V (vs. SCE) was applied between CNT film and platinum wire. A solution of 0.1 M pyrrole and 0.1 M NaClO₄ was used as electrolyte for PPy deposition. In order to deposit the PPy on the CNTs evenly, the CNT films were supported by two stainless steel wire meshes. After washing in de-ionic water for several times, the PPy@CNT films were spun into fibers in wet state by two low speed motors. Finally, the PPy@CNT fibers were dried in vacuum at 80 °C for overnight. The weight and length of the fibers were measured in dry state. The content of the PPy in the composite fibers was calculated by the weight gain after PPy deposition.

For comparison, we also spun some pure CNT fibers directly by two low speed motors. Some of the pure CNT fibers were further electrochemically deposited with PPy in the same conditions of making PPy@CNT fiber.

Materials characterization

The fibers were characterized using scanning electron microscope (SEM, LEO1530) and transmission electron microscope (TEM, JEOL 2010). The electrical conductivities of the fibers were measured by a two-probe method using a Keithley 2601 Source Meter to record current and voltage.

Electrochemical measurement

The electrochemical properties of the fiber electrodes were measured in 1 M H₂SO₄ aqueous electrolyte by a three-electrode method using SCE and Pt wire as the reference electrode and counter electrode, respectively. Cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) curves were recorded by an electrochemical workstation (CHI660D) at a potential range of −0.2 V to 0.6 V (vs. SCE) with different scanning rates of 20–200 mV s⁻¹ and current densities of 1–50 A g⁻¹, respectively. The electrochemical impedance spectroscopy (EIS) was tested in the frequency range of 100 kHz to 0.01 Hz at an open circuit potential with an AC perturbation of 5 mV.

3. Results and discussion

Fig. 1b shows an optical image of a pure CNT film with dimension of 5.5 cm × 2.5 cm supported by two stainless steel meshes. The longer CNT films can be fold and supported by the meshes for electrochemical deposition. After coating with PPy, the CNT film is spun to fiber. By using this method, we can prepare very long PPy@CNT fibers. Fig. 1c shows a PPy@CNT fiber with length more than 1 m wounding on a spool. The PPy@CNT fibers have diameter in a range of 100–300 μm, depending on the CNTs used for spinning and also on the densification of the fibers. The PPy@CNT fibers are flexible and can be bent and knitted into letters (Fig. 1d).

Fig. 2a shows a low magnification SEM image of the cross-section of a PPy@CNT fiber with PPy deposition time of 600 s. The PPy distribute evenly in the fiber. A side-viewed SEM image (Fig. 2b) shows that there are lots of spirals on the surface of PPy@CNT fiber. The spirals consist of many micro-grooves and micro-ridges which derive both from the chiral angle and the shrinkage of the composite films (see inset of Fig. 2b). The space between two adjacent spirals is about 5–50 μm. Fig. 2c shows a high magnification SEM image of the PPy@CNT. After deposition, the CNT bundles are coated with a PPy layer evenly, which cause the diameter increase significantly. Because the deposition of PPy is prior to the fiber spinning, pyrrole monomer can fully penetrate into the CNT film, which make the PPy distributes evenly in the fiber. The abundant micro-pore within the CNT film are reserved after the PPy deposition, and even after spinning. Fig. 2d shows a high resolution TEM image of the PPy@CNT composite. It is clear that the CNT bundles are coated with a continuous PPy layer. The thickness of the PPy layer depends on the deposition time, which is about 10 nm for the sample prepared by depositing PPy for 600 seconds. The PPy@CNT composite have core–shell structure and good interface between the CNT and PPy.

Fig. 2 (a) Cross-sectional SEM image, (b) low magnification SEM image, inset is a close view of the surface of PPy@CNT, (c) high magnification SEM image, and (d) HRTEM image of a PPy@CNT fiber with 51 wt% PPy. (e) and (f) are side-viewed and cross-sectional SEM images of a PPy/CNT fiber with 53 wt% PPy, respectively.

The content of PPy in the PPy@CNT fiber is calculated by the weight gain after removing the residual solution by washing with de-ionic water several times and drying. We also provide linear mass density by measuring the weight and length of the fiber. The linear mass density of the fibers increase evidently with the deposition time prolonging. Here, we can control the content of PPy in a range of 34 wt% to 70 wt% by tuning the deposition time from 200 s to 800 s (see Table 1).

Table 1 Parameters of the fiber electrodes

Deposition time (s)	Linear density (Tex)	Content of PPy (wt%)	Conductivity (10 kS m^−1)	Specific capacitance (F g^−1)
0 (pure CNTs)	17	0	23.4	17.4
200	26	34	22.3	153.5
400	30	43	21.9	232.1
600	35	51	17.9	350.5
800	57	70	2.4	224.3
600 (PPy/CNT)	36	53	3.2	124.2

---

For comparison, we also prepared some CNT fibers coating with PPy layer according to recent reports,^33,34 where the PPy was electrochemically deposited on the pure CNT fiber (denoted as PPy/CNT fiber) at the same conditions of preparation of PPy@CNT composite. Fig. 2e shows a SEM image of the PPy/CNT fiber with deposition time of 600 s. The CNT fiber is covered by a continuous PPy layer, which is quite different from the PPy@CNT fiber. The PPy layer might crack along the spiral of the CNT fiber after drying. Because it is hard for pyrrole monomer to fully penetrate into the dense CNT fiber, the PPy distributes mainly on the surface of the CNT fiber (Fig. 2f). The PPy layer has a thickness of 20–30 μm. There is an evident interface between the PPy and CNT fiber.

As illustrated in Table 1, the electrical conductivity decreases after the deposition of PPy. The electrical conductivity of the pure CNT fibers in our experiments is about 23.4 kS m⁻¹. It reduces slightly from ∼22.3 kS m⁻¹ to 17.9 kS m⁻¹ when the deposition time prolongs from 200 s to 600 s. But, it drops dramatically to only 2.4 kS m⁻¹ when the deposition time reaches to 800 s. The conductivity of the PPy@CNT fiber with short deposition time (600 s) is much higher than that of the fibers reported very recently, for example hybrid CNTs (5 kS m⁻¹),⁴² graphene (3.5 kS m⁻¹),⁴³ and CNT/graphene (10.2 kS m⁻¹).¹² The electrical conductivity of the PPy/CNT fiber with 53 wt% PPy is only 3.2 kS m⁻¹, which is one order of magnitude lower than that of the PPy@CNT fiber with the similar content of PPy. It indicates that charge transfer within the PPy/CNT fibers is impeded due to the poor conductivity of the PPy.

The PPy have high pseudocapacitance but poor conductivity. Thus, the electrochemical properties of the PPy@CNT fibers are affected by the content of PPy significantly. Fig. 3a shows the cyclic voltammetry (CV) curves of the PPy@CNT fibers with different PPy contents at a scan rate of 100 mV s⁻¹. The CV curves of the maintained good rectangular shape even when the content of PPy is lower than 51 wt%. There are evident pseudocapacitance peaks in the CV curves at high potential, showing the pseudocapacitance in the fibers, especially for the fiber with high content of PPy (70 wt%). As shown in Fig. 3b, the GCD curves show good triangle symmetry, indicating that the electrodes have a high reversibility between charge and discharge processes. Fig. 3c shows plots of the specific capacitance of the PPy@CNT fibers as the current density increases. It is clear that the specific capacitance of the PPy@CNT fiber is influenced by both of the content of PPy and charge–discharge current density. Because of the introduction of high pseudocapacitance from PPy, the specific capacitance is improved from 153 F g⁻¹ to 350 F g⁻¹ when the load of PPy increases from 34 wt% to 51 wt%. But it drops from 350 F g⁻¹ to 224 F g⁻¹ when the content of PPy further increases from 51 wt% to 70 wt%. In fact, the pseudocapacitance in the PPy@CNT fibers increases with the increase of the content of PPy; on the other hand the internal resistance also increases, especially for the fiber with high content of PPy. Because the conductivity decrease dramatically when the content of PPy is higher than 51 wt%, the specific capacitance decreases.

Fig. 3 Supercapacitor characteristics of the PPy@CNT composite fiber with different content of PPy. (a) CV curves. (b) Galvanostatic charge–discharge curves. (c) The specific capacitance as a function of the current density.

Fig. 4 shows the CV and GCD curves of the three type of fibers: pure CNT, PPy/CNT and PPy@CNT. The pure CNT fibers have nearly perfect rectangular shapes in the CV curves (Fig. 4a) and symmetric triangle in the GCD curves (Fig. 4b), exhibiting double-layered supercapacitor characteristics. The specific capacitance is only 17.4 F g⁻¹ at current density of 1 A g⁻¹. As shown in Fig. 4c, the CV curves of the PPy/CNT fiber show evident pseudocapacitor characteristics. In the cyclic charge–discharge curves (Fig. 4d), there is evident potential drop at discharge start (denoted as IR), indicating large equivalent series resistance (ESR) in the PPy/CNT fiber. The specific ESR of the PPy/CNT fiber calculated from the cyclic charge–discharge curves is as high as 41.6 kΩ g⁻¹, which is about 6 times higher than that of the pure CNT fiber (7.3 kΩ g⁻¹). Charge transfer within the PPy/CNT fiber was impeded seriously due to the thick PPy layer and evident interface between PPy layer and CNT fiber, resulting in dramatic increase of the internal resistance.

Fig. 4 CV and charge–discharge curves of three type of fiber electrodes. (a) and (b) are pure CNT fiber, (c) and (d) are PPy/CNT fiber with 53 wt% PPy, (e) and (f) are PPy@CNT fiber with 51 wt% PPy.

Because the PPy distribute within the PPy@CNT fiber evenly, it exhibits the best electrochemical properties among the three type of fibers. Fig. 4e shows the CV curves of the PPy@CNT fibers with 51 wt% PPy at various scanning rate. The CV curves have ideal rectangular shape, even at a high scanning rate of 200 mV s⁻¹. The current density is as high as 40 A g⁻¹ at potential of 0.6 V when the scanning rate is 200 mV s⁻¹, which is almost 10 times higher than that of the pure CNT and PPy/CNT fibers. The specific capacitance are enhanced by coordinated effect of high pseudocapacitance of PPy and low internal resistance of CNT bundles. Because of the core–shell structure and good interface between PPy and CNT bundles, charges transfer within the PPy@CNT electrode easily. The GCD curves also have good symmetrical triangle shapes (see Fig. 4f), representing a highly reversible charge and discharge process. The specific ESR of the PPy@CNT fiber is 7.7 kΩ g⁻¹, which is close to that of the CNT fiber (7.3 kΩ g⁻¹). The PPy@CNT fibers thus have a high specific capacitance of 350 F g⁻¹ at 1 A g⁻¹, which is about 20 times of the pure CNT fiber electrode (17.4 F g⁻¹), and 3 times higher than that of the PPy/CNT fiber (124 F g⁻¹).

Fig. 5a shows the electrochemical impedance spectroscopy (EIS) of the three kinds of fibers. The plots of the pure CNT and PPy@CNT fibers are almost vertical to the real axis, showing very low ESR for such fibers. Fig. 5b shows the plots of the impedance phase angle vs. the frequency of the three kinds of fibers. The PPy@CNT fiber behaves typical capacitive characteristic at low frequency and impedance characteristic at high frequency, which is quite different from that of the PPy/CNT fiber. The frequency at an impedance phase angle of −45° for the PPy@CNT fiber is ∼9 Hz, in comparison with only 0.4 Hz for the pure CNT fiber. The impedance phase angle of the PPy@CNT fiber reaches to −82° at 1 Hz.

Fig. 5 Comparison of three types of fibers. (a) Impedance plots, Z′: real impedance, Z′′: imaginary impedance. (b) Plots of phase angle vs. frequency. (c) Variation of the specific capacitance with current density. (d) Cyclic stability during the long-term charge–discharge process.

Fig. 5c compares the dependence of specific capacitances on current density for the three kind of fibers. The PPy@CNT fiber has the highest specific capacitance at the same current density. The CNT@PPy fiber retains its specific capacitance more than 50% when the GCD current increases from 1 A g⁻¹ to 50 A g⁻¹, contributing from its outstanding charge transfer ability. The high capacitance retention shows a remarkable rate capability for the PPy@CNT fiber supercapacitor electrodes. For comparison, the specific capacitance of the CNT/PPy fiber drops from 124 F g⁻¹ to only 1 F g⁻¹ when the current density increase from 1 A g⁻¹ to 50 A g⁻¹. The energy density and power density of the PPy@CNT fiber capacitor electrodes reach to 31.2 W h kg⁻¹ and 402 W kg⁻¹, respectively, which are comparable to those reported recently.^18,25,44 The PPy@CNT fibers also show good stability that the CNT@PPy fiber electrode retains ∼88% of its initial capacitance after 5000 CV cycles (Fig. 5d), which is better than that of CNT/PPy (76%). We believe that the decrease of capacitance in the cyclic test derives mainly from the degradation of PPy. In the PPy@CNT fiber, the CNT bundles are covered by a thin layer of PPy. The degradation of PPy has little influence on the electrochemical properties. While for the PPy/CNT fiber, the thick layer PPy and evident interface make the degradation rate affects the capacitance significantly. So, the PPy@CNT shows better cyclic performance compared with the PPy/CNT fiber.

Flexibility is one of the important features for the fiber electrode, which enable it to be bent, twisted, spun and woven. Here, we demonstrate the electrochemical stability of the PPy@CNT fibers under bending and twisting. A PPy@CNT fiber is twined around glass rods with different diameters, which make the fiber be bent into different curvatures. Fig. 6a shows the capacitance change (ΔC) relative to initial value (C₀) depending on the bending curvature. The capacitance changes slightly (∼5%) when the fiber is bent with a curvature varying from 2 cm⁻¹ to 20 cm⁻¹, showing high stability under bending. The PPy@CNT fiber is further twisted into a two-ply fiber with different turns. As shown in inset of Fig. 6b, the CV curves almost overlap with each other when the PPy@CNT fiber is twisted into different turns. The fiber is also shaped into spring. The PPy@CNT spring is stretchable and compressible. The two CV curves of the fiber before and after shaping into spring almost coincide with each other. It shows that the PPy@CNT fiber has high electrochemical stability under bending and twisting, which might be used as flexible electrode in the wearable and portable electronics.

Fig. 6 Electrochemical characteristics of the CNT@PPy fiber electrodes under bending and twisting. (a) The capacitance ratio of a fiber under bending in relative to origin value (C₀). Insets are optical images of a fiber winding on glass rods with different diameters. (b) CV curves of a fiber twined into different turns. (c) Optical images of the CNT@PPy spring under tension and compression. (d) CV curves of a fiber before and after making a spring.

4. Conclusions

In summary, we prepared flexible PPy@CNT fiber with core–shell structure by electrochemically depositing PPy onto macroscopic CNT film followed by fiber spinning, where the PPy distributes in the fiber evenly. The PPy@CNT fiber has a high specific capacitance of 350 F g⁻¹ by taking advantage of high conductivity of CNTs and high pseudocapacitance of PPy. The PPy@CNT fiber has low equivalent series resistance, high specific capacitance, high cyclic stability, and flexibility, which suggested that the PPy@CNT fiber is a promising candidate using as supercapacitor electrodes for wearable and portable devices.

Acknowledgements

This work was supported by National Natural Science Foundation of China (51172122) and Tsinghua University Initiative Scientific Research Program (20111080939).

References

1. M. F. El-Kady, V. Strong, S. Dubin and R. B. Kaner, Science, 2012, 335, 1326.
2. J. R. Miller and P. Simon, Science, 2008, 321, 651.
3. P. Simon, Y. Gogotsi and B. Dunn, Science, 2014, 343, 1210.
4. T. Chen and L. M. Dai, J. Mater. Chem. A, 2014, 2, 10756.
5. M. Armand and J. M. Tarascon, Nature, 2008, 451, 652.
6. L. Nyholm, G. Nystrom, A. Mihranyan and M. Stromme, Adv. Mater., 2011, 23, 3751.
7. Y. Fu, X. Cai, H. W. Wu, Z. B. Lv, S. C. Hou, M. Peng, X. Yu and D. C. Zou, Adv. Mater., 2012, 24, 5713.
8. T. L. Viet, H. Kim, A. Ghosh, J. Kim, J. Chang, Q. A. Vu, D. T. Pham, J. H. Lee, S. W. Kim and Y. H. Lee, ACS Nano, 2013, 7, 5940.
9. Y. C. Cao, M. Zhu, P. X. Li, R. Zhang, X. Li, Q. Gong, K. L. Wang, M. L. Zhong, D. Wu, F. Lin and H. W. Zhu, Phys. Chem. Chem. Phys., 2013, 15, 19550.
10. X. F. Wang, B. Liu, R. Liu, Q. Wang, X. Hou, D. Chen, R. Wang and G. Z. Shen, Angew. Chem., Int. Ed., 2014, 53, 1849.
11. X. F. Wang, X. H. Lu, B. Liu, D. Chen, Y. X. Tong and G. Z. Shen, Adv. Mater., 2014, 26, 4763.
12. D. S. Yu, K. Goh, H. Wang, L. Wei, W. C. Jiang, Q. Zhang, L. M. Dai and Y. Chen, Nat. Nanotechnol., 2014, 9, 555.
13. J. Y. Tao, N. S. Liu, W. Z. Ma, L. W. Ding, L. Y. Li, J. Su and Y. H. Gao, Sci. Rep., 2013, 3, 2286.
14. J. Ren, L. Li, C. Chen, X. Chen, Z. Cai, L. Qiu, Y. Wang, X. Zhu and H. S. Peng, Adv. Mater., 2013, 25, 1155.
15. C. Choi, J. A. Lee, A. Y. Choi, Y. T. Kim, X. Lepró, M. D. Lima, R. H. Baughman and S. J. Kim, Adv. Mater., 2014, 26, 2059.
16. K. Wang, H. P. Wu, Y. N. Meng and Z. X. Wei, Small, 2014, 10, 14.
17. S. W. Pan, H. J. Lin, J. Deng, P. N. Chen, X. L. Chen, Z. B. Yang and H. S. Peng, Adv. Energy Mater., 2015, 5, 1401438.
18. X. L. Chen, L. B. Qiu, J. Ren, G. Z. Guan, H. J. Lin, Z. T. Zhang, P. N. Chen, Y. G. Wang and H. S. Peng, Adv. Mater., 2013, 25, 6436.
19. J. Ren, W. Y. Bai, G. Z. Guan, Y. Zhang and H. S. Peng, Adv. Mater., 2013, 25, 5965.
20. L. Kou, T. Q. Huang, B. N. Zheng, Y. Han, X. L. Zhao, K. Gopalsamy, H. Y. Sun and C. Gao, Nat. Commun., 2014, 5, 3754.
21. H. H. Cheng, C. G. Hu, Y. Zhao and L. T. Qu, NPG Asia Mater., 2014, 6, 113.
22. F. L. Qiu, D. Harrison, J. Fyson and D. Southee, Smart Science, 2014, 2, 107.
23. J. Bae, M. K. Song, Y. J. Park, J. M. Kim, M. L. Liu and Z. L. Wang, Angew. Chem., Int. Ed., 2011, 50, 1.
24. J. A. Lee, M. K. Shin, S. H. Kim, H. U. Cho, G. M. Spinks, G. G. Wallace, M. D. Lima, X. Lepró, M. E. Kozlov, R. H. Baughman and S. J. Kim, Nat. Commun., 2013, 4, 1970.
25. D. H. Zhang, M. H. Miao, H. T. Niu and Z. X. Wei, ACS Nano, 2014, 8, 4571.
26. Z. Yu and J. Thomas, Adv. Mater., 2014, 26, 4279.
27. J. H. Chen, W. Z. Li, D. Z. Wang, S. X. Yang, J. G. Wen and Z. F. Ren, Carbon, 2002, 40, 1193.
28. R. Signorelli, D. C. Ku, J. G. Kassakian and J. E. Schindall, Proc. IEEE, 2009, 97, 1837.
29. R. Shah, X. F. Zhang and S. Talapatra, Nanotechnology, 2009, 20, 395202.
30. S. Jagannathan, T. Liu and S. Kumar, Compos. Sci. Technol., 2010, 70, 593.
31. J. Wang, Y. L. Xu, X. Chen and X. F. Sun, Compos. Sci. Technol., 2007, 67, 2981.
32. V. Gupta and N. Miura, Electrochim. Acta, 2006, 52, 1721.
33. C. Z. Meng, C. H. Liu, L. Z. Chen, C. H. Hu and S. S. Fan, Nano Lett., 2010, 10, 4025.
34. Y. Fang, J. Liu, D. J. Yu, J. P. Wicksted, K. Kalkan, C. O. Topal, B. N. Flanders, J. Wu and J. Li, J. Power Sources, 2010, 195, 674.
35. A. Razaq, L. Nyholm, M. Sjodin, M. Stromme and A. Mihranyan, Adv. Energy Mater., 2012, 2, 445.
36. G. Nystrom, A. Razaq, M. Stromme, L. Nyholm and A. Mihranyan, Nano Lett., 2009, 9, 3635.
37. X. Li and I. Zhitomirsky, J. Power Sources, 2013, 221, 49.
38. X. J. Lu, H. Dou, C. Z. Yuan, S. D. Yang, L. Hao, F. Zhang, L. F. Shen, L. J. Zhang and X. G. Zhang, J. Power Sources, 2012, 197, 319.
39. K. Wang, Q. H. Meng, Y. J. Zhang, Z. X. Wei and M. H. Miao, Adv. Mater., 2013, 25, 1494.
40. I. H. Kim, J. H. Kim, Y. H. Lee and K. B. Kim, J. Electrochem. Soc., 2005, 152, A2170.
41. A. L. M. Reddy, M. M. Shaijumon, S. R. Gowda and P. M. Ajayan, J. Mater. Chem. C, 2010, 114, 658.
42. F. Gao, L. Viry, M. Maugey, P. Poulin and N. Mano, Nat. Commun., 2010, 1, 2.
43. H. P. Cong, X. C. Ren, P. Wang and S. H. Yu, Sci. Rep., 2012, 2, 613.
44. S. Q. He, J. Q. Wei, F. M. Guo, R. Q. Xu, C. Li, X. Cui, H. W. Zhu, K. L. Wang and D. H. Wu, J. Mater. Chem. A, 2014, 2, 5898.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra01917f

---