Ultrahigh specific surface area porous carbon nanospheres and its composite with polyaniline: preparation and application for supercapacitors

Abstract

Porous carbon nanospheres (PCNSs) with an ultrahigh specific surface were successfully prepared by carbonizing monodisperse polypyrrole nanospheres and subsequent active treatment by KOH. A hierarchical composite composed of PCNSs and polyaniline (PANI) nanorods arrays (PCNSs/PANI) were further synthesized via in situ polymerization of aniline on PCNSs. The PCNSs displayed an ultrahigh specific surface area (2817.7 m² g⁻¹), abundant mesoporous structures and high specific capacitance (320 F g⁻¹). They delivered a capacitance retention of 100% after 1000 cycles at a current density of 5 A g⁻¹. When composed with PANI nanorods, the specific capacitance of PCNSs/PANI reached up to 584 and 407 F g⁻¹ at 0.2 and 5 A g⁻¹, respectively. The superior performance of PCNSs/PANI was attributed to the hierarchical structures and synergistic effect between PANI and PCNSs. The unique structures and outstanding properties of these materials make them excellent candidates for high-performance supercapacitors.

---

1. Introduction

During the past decade, supercapacitors have been considered as promising energy storage and delivery devices because of their high power density, fast charge capability, and long cycling life.^1–5 According to the charge storage mechanism, supercapacitors can be divided into electrical double-layer capacitors (EDLCs) and redox capacitors (pseudocapacitors).^6–8 Carbon micro/nanospheres have attracted much attention as electrode materials for EDLCs owing to their large packing density and good conductivity.^9–14 In order to enhance their capacitance, much effort has been devoted to introduce nitrogen doping and improve specific surface area (SSA).^15–19 For example, Ferrero et al. reported that a nitrogen doped microporous carbon microspheres yielded a specific capacitance of 310 F g⁻¹ at a current density of 0.1 A g⁻¹.²⁰ Unfortunately, the SSA of carbon nanospheres in the most of literature reported is still relatively low. Hence, it is still necessary to further improve the SSA of carbon nanospheres, and thus improve their specific capacitance for practical applications on EDLCs.

Pseudocapacitance from reversible faradic reactions affords a higher energy storage capacity.^21,22 Among the pseudocapacitance materials, polyaniline (PANI) is the most promising one because of its low cost and facile synthesis.^23,24 However, PANI exhibits poor cycling stability due to the swelling and shrinkage during the charge–discharge process. To overcome this drawback, combining PANI with mesoporous carbon materials has been proved to be an effective method to reinforce the stability of PANI as well as improve its capacitance.^25–27 Li et al. reported that the PANI polymerization on the surface of mesoporous carbon with irregular morphology possessed a high specific capacitance of 747 F g⁻¹ at 0.1 A g⁻¹ which only remained 66% capacitance after 30 cycles.²⁴ A possible reason was that the low specific surface area of mesoporous carbon provided few growth sites for PANI, resulting in the inefficient anchor of PANI on the surface of the carbon substrate. Hence, the introduction of high specific surface carbon nanospheres as a substrate for PANI can be expected to further improve the capacitive performance of PANI. In addition, another factor affecting the performance of PANI is its microstructures.²⁸ Recently, hierarchical nanocomposites of PANI nanorods arrays on the carbon materials, such as graphene oxide, sulfonated graphene nanosheets, and mesoporous carbon etc. have been fabricated to enhance specific capacitance, rate performance and cycling stability, which may ascribed to its providing a relatively short diffusion pathway for electrolyte ions to access the electroactive surface.^29–32

Herein, we report a convenient method to fabricate porous carbon nanospheres (PCNSs) with ultrahigh SSA derived from the monodisperse polypyrrole (PPy) nanospheres. PANI nanorod array with hierarchical structure were grown on the surface of PCNSs via the in situ polymerization of aniline (Scheme 1). Ultrahigh specific surface area and uniformly interconnected stacking hole simultaneously achieved and resulted in the improved specific capacitance and rate performance of PCNSs. As electrode materials for supercapacitors, the as-prepared PCNSs/PANI combining the advantage of both PCNSs and PANI exhibited superior capacitive performance.

Scheme 1 Preparation of the PCNSs and PCNSs/PANI.

2. Experimental

2.1 Fabrication of PCNSs

PPy nanospheres were synthesized according to our previous report.³³ In order to obtain uniformly distributed carbon nanospheres (CNSs), PPy nanospheres were calcined at 750 °C for 2 h under flowing nitrogen. Then, the obtained CNSs were mixed with an aqueous KOH solution in a weight ratio of KOH : carbon = 4 : 1, and dried in an oven at 85 °C to remove the water. Finally, the mixture was heated to 850 °C for 1 h under a nitrogen atmosphere. The products were repeatedly washed with deionized water until the pH value was about 7, followed by drying at 105 °C overnight. The obtained activated material was named as PCNSs. For comparison, PCNSs were activated at 700, 750, 800, and 900 °C with the same process.

2.2 Fabrication of PCNSs/PANI

The mixture of 0.3 g PCNSs and 50 mL ethanol solution (20%) was sonicated until PCNSs were fully dispersed. Then, 0.44 mL aniline monomer and 5 mL concentrated sulfuric acid were successively added into the above mixture and further sonicated for 30 min. After that, the dispersion was moved to an ice bath for stirring 2 h to control the temperature about 0–5 °C. Then, 10 mL ammonium persulfate (APS) aqueous solution was dropwise added into the mixed solution. The molar ratio of aniline to APS was 1 : 1.25. The reaction was carried out for 24 h in the ice bath under stirring. The dark green products were collected by filter and repeatedly rinsed with deionized water, and then dried in an oven at 60 °C for 12 h. The as-prepared composite was named as PCNSs/PANI. For comparison, other composites with different amounts of PANI grown on the PCNSs were prepared in the same process.

2.3 Characterization

The morphology of the products was characterized by scanning electron microscopy (SEM, S-4800) and transmission electron microscopy (TEM, JEM-3010), respectively. The nitrogen sorption experiments were carried out at 77 K using a Micrometrics ASAP 2020 analyzer. The SSA was obtained by the Brunauer–Emmett–Teller (BET) equation and pore-size distribution (PSD) was obtained by Barrett–Joyner–Halenda (BJH) method. Raman spectra were recorded on a Labram-010 microscopic confocal Raman spectrometer with 632.8 nm laser excitation. FT-IR spectra were measured using a Nicolet Avatar 360 FT-IR spectrometer by dispersing samples in KBr pellets. The thermogravimetric (TG) measurements were performed on a diamond TG/DTA thermal analyzer under a nitrogen atmosphere at a heating rate of 10 °C min⁻¹. X-ray diffraction (XRD) data of samples was collected on a SIEMENS D-5000 diffractometer using Cu-Kα radiation. C, N, H and O elemental contents of samples were measured by an elemental analyzer (Vario EL III).

2.4 Electrochemical measurements

A three-electrode cell system was used to carry out electrochemical measurements, such as cyclic voltammetry (CV), galvanostatic charge–discharge (GCD) and electrochemical impedance spectroscopy (EIS). All the electrochemical measurements were carried out on a CHI660D electrochemical workstation. The working electrodes were prepared by coating the mixture consisting of 80% of active materials, 10% of polytetrafluoroethylene binder and 10% of acetylene black on stainless steel (1 cm × 1 cm). A saturated calomel electrode (SCE) and platinum foils was used as the reference and counter electrodes, respectively. In 1.0 M H₂SO₄ electrolyte, the CV and GCD tests were measured with the potential window from −0.2 V to 0.8 V, the scan rates of CV tests were 2, 5, 10, 20, 50, and 100 mV s⁻¹, while the current density of GCD tests were 0.2, 0.5, 1.0, 2.0, 5.0, and 10 A g⁻¹. The EIS measures were carried out in the frequency range from 10 mHz to 10 kHz with an AC amplitude of 5 mV. The gravimetric specific capacitance values of the samples were calculated from GCD data according to the following equation.³⁴where I is the constant discharging current, Δt is the discharge time, ΔV is the voltage window, and m is the mass of active materials (about 3.5 mg) loaded in the working electrode.

3. Results and discussion

3.1 Structure and morphology characterization

The monodisperse PPy nanospheres were used as carbon and nitrogen source for the preparation of PCNSs. As shown in Fig. S1a,† the PPy nanospheres exhibits smooth surfaces, and the diameter is about 120 nm. After carbonization and activation, the obtained PCNSs still retain spherical morphology with relatively narrow particle size distribution (Fig. 1a). Fig. 1b shows that the PCNSs have a remarkably rough surface with diameter about 100 nm. From the TEM image inserted in Fig. 1b, it appears that the surface of the PCNSs is rough, and abundant mesopores are generated on the PCNSs. In addition, mesopores with diameter of 15–30 nm can be formed through the stacking of PCNSs. Fig. 1c shows the SEM image of the PCNSs/PANI. It can be seen that vertically PANI nanorods are obtained on the surface of PCNSs. The high-magnification SEM image (Fig. 1d) indicates that the average diameter of aligned PANI nanorods is around 20–30 nm. The hierarchical composites further confirm that PANI nanorods can be formed on various substrates by dilute polymerization of aniline monomer.³⁵ In contrast, pure PANI without PCNSs shows random stacking nanorods morphology with 50–60 nm diameters (Fig. S1b†).

Fig. 1 (a) and (b) show SEM images of the PCNSs; (c) and (d) display SEM images of the PCNSs/PANI; TEM image of the PCNSs is inserted in (b).

Nitrogen adsorption is a common technique to analyze the porosity of materials. As shown in Fig. 2a, a distinct hysteresis loop can be observed from P/P₀ = 0.55 to 0.98, and a steep rise of the N₂ adsorption occurred at low relative pressures (P/P₀ < 0.1), indicating a combination of Type IV and Type I isotherm for PCNSs.^36,37 The PSD curve inserted in Fig. 2a shows that the PCNSs has bimodal mesopores around 2–4 nm and 15–30 nm, which may correspond to the pores enlarged by KOH etching and piled up by the PCNSs. Moreover, it also exhibits abundant micropores (<2 nm) formed during the activation process. These results are consistent with the structure of PCNSs shown in the TEM image (inserted in Fig. 1b). N₂ adsorption isotherm of the PCNSs/PANI is shown in Fig. 2b. Obviously, the isotherm is still a combination of Type IV and Type I isotherm. However, the nitrogen adsorption capacity of PCNSs/PANI is much lower than that of PCNSs. The micropores with a diameter less than 2 nm and the mesopores around 2–10 nm can be distinguished in the inserted PSD curve. The porous properties of PCNSs and PCNSs/PANI are listed in Table 1. These results show that the SSA of PCNSs is 2817.7 m² g⁻¹ and the total pore volume is as high as 2.06 cm³ g⁻¹. It is much higher than that of other hierarchically porous carbons.^{10,12,18,34,37,38} On the other hand, the SSA and total volume of PCNSs/PANI is only 190 m² g⁻¹ and 0.076 cm³ g⁻¹, respectively. It suggests that aniline molecules were polymerized both in the micro/mesopores channels and on the surfaces of the PCNSs.

Fig. 2 Nitrogen adsorption/desorption isotherm and BJH pore size distribution (inset): (a) PCNSs, and (b) PCNSs/PANI.

Table 1 Porous property of materials

Sample	SSA^a (m^2 g^−1)	Vtotal^b (cm^3 g^−1)	APD^c (nm)
a SSA measured for P/P0 = 0.060–0.20.b Total pore volume measured at P/P0 = 0.985.c Average pore diameter.
CNSs	2817.7	2.06	4.25
PCNSs/PANI	189.9	0.076	3.86

---

Fig. 3a shows the Raman spectra of PCNSs and PCNSs/PANI. Two strong peaks at 1590 cm⁻¹ (G band) and 1330 cm⁻¹ (D band) are observed for PCNSs.^7,14 The integrated intensity ratio of D and G bands (I_D/I_G) is often used to investigate the defect degree of porous carbon.³⁹ The value of I_D/I_G is 3.52, revealing PCNSs a highly porous structure and non-graphitized carbon containing turbostratic graphene sheets.⁴⁰ This result is consistent with the nitrogen adsorption analysis. The Raman spectra of PCNSs/PANI displays peaks at 1592, 1327 and 1162 cm⁻¹ corresponding to C–C stretching of the benzenoid ring, C–N˙⁺ stretching of the bipolaron structure and C–H bending of the quinoid ring.³¹ Moreover, the peak at 521 cm⁻¹ attributed to the vibration of C–N–C torsion is obviously stronger than that of PANI (Fig. S2a†), indicating an increased doping level of PANI. The FT-IR spectrum of the PCNSs and PCNSs/PANI are shown in Fig. 3b. The spectra of the PCNSs displays weak peaks at 3423, 1575 and 1098 cm⁻¹, ascribing to –OH, C=C and C–C–O, implying a small amount of oxygen-containing functional groups on its surface.^25,29,30 The peaks at 1582, 1507, and 1302 cm⁻¹ of PCNSs/PANI are attributed to the C=C stretching vibration in the quinoid and benzenoid rings, and C–N stretching vibration, clearly indicates the formation of PANI on the PCNSs surfaces.³² The peaks at 1041, 690 and 514 cm⁻¹ indicate the PANI had been doped with SO₄²⁻.⁴¹ The XRD pattern of PCNSs (Fig. 3c) shows a broad diffraction peak at around 2θ = 25° corresponding to the (002) layers of the graphite and a weak peak near 2θ = 43° corresponding to the (101) plane of graphene.^34,40 This is in agreement with the Raman spectra. Two broad peaks at 2θ = 19.2 and 24.7° are observed in the PCNSs/PANI (Fig. 3c) and PANI (Fig. S2c†). It shows that the PANI of the composites has a similar structure with pure PANI. As shown in Fig. 3d, the TGA curve of PCNSs displays an obviously weight loss above 600 °C, which may be caused by the etching of the mesoporous walls. The degradation of PCNSs/PANI exhibits a mass loss of 38.2% between 100 and 900 °C, and a mass loss of 74% can be observed from pure PANI (Fig. S2c†). The relatively lower weight loss of PCNSs/PANI should be attributed to a better thermal stability of PANI in the mesoporous channels and PANI nanorods arrays on the PCNSs. The C, H, N, S and O contents were detected by elemental analysis and shown in Table 2. This result confirms that SO₄²⁻ has been doped into PCNSs/PANI.

Fig. 3 (a) Raman spectra, (b) FT-IR spectra, (c) XRD patterns, and (d) TGA curves of PCNSs and PCNSs/PANI.

Table 2 Elemental percentage of samples

	C	H	N	O	S
PCNSs	90.7	2.3	0.4	6.4	0
PCNSs/PANI	78.4	3.6	8.3	7.4	2.3

---

3.2 Electrochemical properties

The CV curves of the PCNSs at various scan rates are shown in Fig. 4a. It clearly displays that the CV curves are quasi-rectangular shaped indicating its ideal EDLCs behavior.^7,8,34,42 Meanwhile, a wide hump can be observed in each CV plot, which is attributed to redox reactions of the nitrogen and oxygen functionalities on the surface of PCNSs.⁴³ With the scan rate decreasing to 2 mV s⁻¹, three couples of redox peaks appear at C1/A1, C2/A2 and C3/A3 (Fig. S3a†). Peaks C1/A1 and C3/A3 can be assigned to redox reactions of the nitrogen functionalities, as proposed by Frackowiak et al. as the following oxidation/reduction reactions:⁴⁴

–C=NH + 2e⁻ + 2H⁺ ↔ –CH–NH₂ (2)

–CH–NHOH + 2e⁻ + 2H⁺ ↔ –C–NH₂ + H₂O (3)

Fig. 4 (a) and (c), (e) CV curves of PCNSs, PCNSs/PANI and PANI operated at 100 to 2 mV s⁻¹; (b) and (d), (f) GCD curves of PCNSs, PCNSs/PANI and PANI operated at 0.2 to 5 A g⁻¹.

Moreover, peaks at C2/A2 are attributed to redox reactions of quinine/hydroquinone groups.

Fig. S3b† shows that the specific capacitance of PCNSs increased with the activated temperature from 700 to 850 °C, and then decreased with the temperature rising to 900 °C. It means that the best activation temperature is 850 °C. The GCD curves of PCNSs electrode at different current densities are exhibited in Fig. 4b. It clearly shows that all the charge and discharge curves are nearly linear and symmetric, indicating that the electrode displays a good capacitive properties and electrochemical reversibility. At the same time, “IR drops” can't be obviously observed in all of the GCD curves. In order to further investigate the contribution of EDLC to the electrode, the discharge curve at 0.2 A g⁻¹ has been divided into two parts according to the slope. The T_E and T_P represent the discharge time of the EDLC and pseudocapacitance, respectively.^34,45 The EDLC calculated from T_E is 212 F g⁻¹, which is about 66.3% of the corresponding total capacitances. The high EDLC is attributing to the high SSA of PCNSs. Moreover, although the N and O content is as low as 0.4% and 6.4%, their pseudocapacitance is still up to 108 F g⁻¹. Fig. 4c shows the CV curves of PCNSs/PANI at different scan rates. At a scan rate of 100 mV s⁻¹, the CV curve exhibits a spindle shape, indicating combination of EDLC and pseudocapacitance. As the scan rate decrease, two couples of redox peaks can be observed in the curves. Peaks C1/A1 and C2/A2 (Fig. S3c†) are assigned to the redox reactions of PANI between leucoemeraldine/emeraldine and emeraldine/pernigraniline transformation.^22,32 In contrast, no obvious redox peak can be observed in the CV curve of the pure PANI synthesized in an ethanol solution of sulfuric acid (Fig. 4e), which may be due to its poor conductivity. The GCD plots of PCNSs/PANI and pure PANI at various current densities are given in Fig. 4d and f. Comparison with pure PANI, PCNSs/PANI has much smaller “IR drops”. The GCD curves of PCNSs/PANI almost maintain symmetric shape, indicating a high coulombic efficiency. At the same time, it can be clearly seen that the discharge curves display EDLC and pseudocapacitance. According to the T_E and T_P of the GCD curves at 0.2 A g⁻¹, the EDLC and pseudocapacitance is 126 and 458 F g⁻¹, respectively. The influences of the different amount of aniline in the composites on the specific capacitance were investigated and the corresponding results are shown in Fig. S3d.† As the concentration of aniline increased from 0.02 to 0.06 M, the specific capacitance of PCNSs/PANI increase to 592 F g⁻¹ at current density of 0.2 A g⁻¹, and then decrease to 398 F g⁻¹ with the concentration of aniline further increased to 0.09 M. However, at the same current density, the pure PANI only possesses 208 F g⁻¹. The enhanced specific capacitance of PCNSs/PANI is mainly due to the increasement of conductivity and hierarchical structures of the composites.

The rate performance of PCNSs, PCNSs/PANI and pure PANI was evaluated by the discharge curves at various current densities (Fig. 5a). It can be seen that the specific capacitance of PCNSs at a current density of 10 A g⁻¹ is 168 F g⁻¹, which is 52.5% of the capacitance at 0.2 A g⁻¹. The good rate performance of PCNSs can be attributed to its abundant micropores, well-develop mesopores and interconnected packing pores. For PCNSs/PANI, the specific capacitance at 10 A g⁻¹ is 352 F g⁻¹, and 60.3% of the initial capacitance at 0.2 A g⁻¹ is maintained. This suggests that both of PCNSs/PANI and PCNSs have good rate behaviour. However, the specific capacitance of pure PANI is only 220 F g⁻¹ at a current density of 0.2 A g⁻¹, and drops to 42 F g⁻¹ when the current density is increased to 10 A g⁻¹. To investigate the cycle stability of the electrodes, GCD measurements were carried out at a current density of 5 A g⁻¹. As shown in Fig. 5b, the PCNSs displays an initial specific capacitance of 204 F g⁻¹ and maintains at 100% capacitance over 1000 cycles, indicating excellent cycling stability as electrode materials for supercapacitors. As for PCNSs/PANI, the capacitance decay occurred mainly in the first 200 cycles, and the capacitance retention still keeps 70% of the initial capacitance (407 F g⁻¹). However, the pure PANI shows a poor initial specific capacitance of 63 F g⁻¹, and maintains only 29% after 1000 cycles. The enhancement in specific capacitance and cycling stability is believed to be due to hierarchical nanostructured PANI providing more electrochemical active sites and the PCNSs matrix facilitating electron transport and electron transfer.

Fig. 5 (a) Specific capacitance at different current densities and (b) cycling stability at 5 A g⁻¹ of PCNSs, PCNSs/PANI and PANI.

As shown in Fig. 6, the Nyquist plots of samples were measured and further fitted by Zsimpwin. All the curves can be divided into a partial semicircle (at high frequency region) and a straight line (at low-frequency region). The two regions are connected by a transition called as “knee frequency”.⁴⁶ The key factors determining the power of electrochemical capacitors are the contact resistance between the electrode and the current collector, the resistance of the electrode materials, and the electrolyte within the porous layer of the electrode, and.⁴⁷ The PCNSs electrode shows a small semicircle at the high-frequency range and a nearly straight line at the low-frequency, indicating a low charge-transfer resistance (R_ct) and good capacitive behavior.²⁸ It is attributed to the uniformly interconnected stacking hole and high specific surface areas. From 459 to 0.55 Hz section of the plot exhibited a Warburg resistance of 3.52 Ω⁻¹ s^0.5. It may be caused by the frequency-dependent ion diffusion resistance of the electrolyte within the porous layer and between the carbon nanospheres. Fig. 6b and c show that the Nyquist plot of PCNSs/PANI possesses a slightly larger semicircle than that of PCNSs, but it is much smaller than that of pure PANI. The R_ct of PCNSs/PANI and pure PANI electrodes are 0.55 and 28.8 Ω, respectively. This reveals that PCNSs can greatly reduce the R_ct of PANI. In the low frequency, the plot of PCNSs/PANI exhibits a shorter and steeper line than pure PANI, which implies a good capacitive behaviour. The corresponding fitting results are shown in Table 3. These results suggest that PCNSs has a high electrical double-layer capacitance and PCNSs/PANI exhibits a large pseudocapacitance.

Fig. 6 Nyquist plots and simulation circuit of (a) PCNSs, (b) PCNSs/PANI, and (c) PANI.

Table 3 The fitting results of simulative EIS

	PCNSs	PCNSs/PANI	PANI
Rs (Ω)	1.616	1.679	1.297
Cdl (F)	2.66 × 10^−2	4.47 × 10^−4
Rct (Ω)	0.195	0.199	43.5
W (Ω^−1 s^0.5)	3.52	2.33	0.0914
CPE (Ω^−1 s^n)	0.682	1.66	0.00041
n	1	1	0.85
Chi squared	2.10 × 10^−4	4.60 × 10^−4	3.56 × 10^−3

---

4. Conclusions

Ultrahigh specific surface area porous PCNSs have been successfully synthesized from PPy nanospheres by carbonization and followed activation with KOH. The PCNSs with a diameter about 100 nm possess a specific surface area of 2817 m² g⁻¹, well-developed mesopores, and interconnected packing pores. The PCNSs electrode exhibits a large specific capacitance (320 F g⁻¹ at 0.2 A g⁻¹), excellent rate capability, and outstanding cycling stability (100% retention after 1000 cycles). The superior capacitive performances of PCNSs are due to the high specific surface area and porous structures. Aligned PANI nanorods have been grown on the surface of PCNSs by the in situ polymerization of aniline in the ethanol solution in the presence of PCNSs. Such a hierarchical structure of PCNSs/PANI can short ion diffusion length and overcome the degradation of PANI, then improve the electrochemical properties of PCNSs/PANI. The electrochemical measures shows that the specific capacitance of PCNSs/PANI reach up to 589 F g⁻¹ at 0.2 A g⁻¹, and still keep 60.3% of its capacitance as current density increasing 50 times (10 A g⁻¹).

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 51273061 and 51473049).

References

1. G. Wang, L. Zhang and J. Zhang, Chem. Soc. Rev., 2012, 41, 797–828.
2. L. L. Zhang and X. Zhao, Chem. Soc. Rev., 2009, 38, 2520–2531.
3. S. Bose, T. Kuila, A. K. Mishra, R. Rajasekar, N. H. Kim and J. H. Lee, J. Mater. Chem., 2012, 22, 767–784.
4. A. Divyashree and G. Hegde, RSC Adv., 2015, 5, 88339–88352.
5. A. Pandolfo and A. Hollenkamp, J. Power Sources, 2006, 157, 11–27.
6. W. Qian, F. Sun, Y. Xu, L. Qiu, C. Liu, S. Wang and F. Yan, Energy Environ. Sci., 2014, 7, 379–386.
7. M. Zhou, F. Pu, Z. Wang and S. Guan, Carbon, 2014, 68, 185–194.
8. J. Han, G. Xu, B. Ding, J. Pan, H. Dou and D. R. MacFarlane, J. Mater. Chem. A, 2014, 2, 5352–5357.
9. Z. Lei, N. Christov, L. L. Zhang and X. S. Zhao, J. Mater. Chem., 2011, 21, 2274–2281.
10. X. Yu, J. Lu, C. Zhan, R. Lv, Q. Liang, Z. H. Huang, W. Shen and F. Kang, Electrochim. Acta, 2015, 182, 908–916.
11. X. Yu, J. G. Wang, Z. H. Huang, W. Shen and F. Kang, Electrochem. Commun., 2013, 36, 66–70.
12. H. Peng, G. Ma, K. Sun, J. Mu, X. Zhou and Z. Lei, RSC Adv., 2015, 5, 12034–12042.
13. Q. Li, R. Jiang, Y. Dou, Z. Wu, T. Huang, D. Feng, J. Yang, A. Yu and D. Zhao, Carbon, 2011, 49, 1248–1257.
14. Y. Fan, X. Yang, B. Zhu, P. F. Liu and H. T. Lu, J. Power Sources, 2014, 268, 584–590.
15. N. P. Wickramaratne, J. Xu, M. Wang, L. Zhu, L. Dai and M. Jaroniec, Chem. Mater., 2014, 26, 2820–2828.
16. W. H. Lee and J. H. Moon, ACS Appl. Mater. Interfaces, 2014, 6, 13968–13976.
17. F. Su, C. K. Poh, J. S. Chen, G. Xu, D. Wang, Q. Li, J. Lin and X. W. Lou, Energy Environ. Sci., 2011, 4, 717–724.
18. J. Jiang, H. Chen, Z. Wang, L. Bao, Y. Qiang, S. Guan and J. Chen, J. Colloid Interface Sci., 2015, 452, 54–61.
19. J. Zhou, J. Lian, L. Hou, J. Zhang, H. Gou, M. Xia, Y. Zhao, T. A. Strobel, L. Tao and F. Gao, Nat. Commun., 2015, 6, 8503–8510.
20. G. Ferrero, A. Fuertes and M. Sevilla, Electrochim. Acta, 2015, 168, 320–329.
21. W. Wu, Y. Li, L. Yang, Y. Ma, D. Pan and Y. Li, Electrochim. Acta, 2014, 139, 117–126.
22. K. Zhang, L. L. Zhang, X. Zhao and J. Wu, Chem. Mater., 2010, 22, 1392–1401.
23. Y. Du, X. Zhao, Z. Huang, Y. Li and Q. Zhang, RSC Adv., 2014, 4, 39087–39094.
24. L. Li, H. Song, Q. Zhang, J. Yao and X. Chen, J. Power Sources, 2009, 187, 268–274.
25. L. L. Zhang, S. Li, J. Zhang, P. Guo, J. Zheng and X. Zhao, Chem. Mater., 2009, 22, 1195–1202.
26. J. Yan, T. Wei, B. Shao, Z. Fan, W. Qian, M. Zhang and F. Wei, Carbon, 2010, 48, 487–493.
27. H. Wang, Q. Hao, X. Yang, L. Lu and X. Wang, Nanoscale, 2010, 2, 2164–2170.
28. Y. Meng, K. Wang, Y. Zhang and Z. Wei, Adv. Mater., 2013, 25, 6985–6990.
29. J. Xu, K. Wang, S.-Z. Zu, B. H. Han and Z. Wei, ACS Nano, 2010, 4, 5019–5026.
30. B. Ma, X. Zhou, H. Bao, X. Li and G. Wang, J. Power Sources, 2012, 215, 36–42.
31. G. Ning, T. Li, J. Yan, C. Xu, T. Wei and Z. Fan, Carbon, 2013, 54, 241–248.
32. Y. G. Wang, H. Q. Li and Y. Y. Xia, Adv. Mater., 2006, 18, 2619–2623.
33. X. Ning, W. Zhong, S. Li and L. Wan, Mater. Lett., 2014, 117, 294–297.
34. L. Wan, J. Wang, L. Xie, Y. Sun and K. Li, ACS Appl. Mater. Interfaces, 2014, 6, 15583–15596.
35. N. R. Chiou and A. J. Epstein, Adv. Mater., 2005, 17, 1679–1683.
36. A. B. Fuertes and M. Sevilla, ACS Appl. Mater. Interfaces, 2015, 7, 4344–4353.
37. Y. S. Yun, M. H. Park, S. J. Hong, M. E. Lee, Y. W. Park and H.-J. Jin, ACS Appl. Mater. Interfaces, 2015, 7, 3684–3690.
38. W. Xing, C. Huang, S. Zhuo, X. Yuan, G. Wang, D. Hulicova-Jurcakova, Z. Yan and G. Lu, Carbon, 2009, 47, 1715–1722.
39. L. Wei, M. Sevilla, A. B. Fuertes, R. Mokaya and G. Yushin, Adv. Funct. Mater., 2012, 22, 827–834.
40. L. Ji, Z. Lin, A. J. Medford and X. Zhang, Carbon, 2009, 47, 3346–3354.
41. Y. Zhu, H. He, M. Wan and L. Jiang, Macromol. Rapid Commun., 2008, 29, 1705–1710.
42. X. Ning, W. Zhong, S. Li, Y. Wang and W. Yang, J. Mater. Chem. A, 2014, 2, 8859–8867.
43. L. Zhao, L. Z. Fan, M. Q. Zhou, H. Guan, S. Qiao, M. Antonietti and M. M. Titirici, Adv. Mater., 2010, 22, 5202–5206.
44. C. Vix-Guterl, E. Frackowiak, K. Jurewicz, M. Friebe, J. Parmentier and F. Béguin, Carbon, 2005, 43, 1293–1302.
45. X. Yang, D. Wu, X. Chen and R. Fu, J. Phys. Chem. C, 2010, 114, 8581–8586.
46. W. Zhang, Z. H. Huang, G. Cao, F. Kang and Y. Yang, J. Power Sources, 2012, 204, 230–235.
47. L. Dai, D. W. Chang, J. B. Baek and W. Lu, Small, 2012, 8, 1130–1166.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra00596a

---