Polypyrrole/cellulose nanofiber aerogel as a supercapacitor electrode material

Abstract

Polypyrrole (PPy)/cellulose nanofiber (CNF) aerogels are prepared using the citric acid-Fe³⁺ (CIT-Fe³⁺) colloid as a precursor of the Fe³⁺ oxidant, which can effectively control the microscopic morphology of the PPy/CNF aerogel due to in situ controlled release of Fe³⁺ oxidant to initiate polymerization of the pyrrole monomer with the common stimulus being pH. The Brunauer–Emmett–Teller (BET) specific surface area of the PPy/CNFs (25%) aerogel is up to 402 m² g⁻¹. All solid-state film supercapacitors are fabricated using the PPy/CNF aerogel film as the electrode material. The maximum specific electrode capacitance is about 215 F g⁻¹. The devices also exhibit excellent cycle stability.

---

1. Introduction

Tremendous energy consumption and climate change have forced human beings to pay much attention to inexpensive and environmentally friendly energy storage devices. Currently, supercapacitors, as an energy storage device, have attracted considerable attention because of their long cycle lifetime, high power capability, environmentally friendliness and high safety.^1–3 Among many promising supercapacitor electrode materials, electronically conducting polymers (ECPs), such as polypyrrole (PPy), have been extensively studied due to their excellent electrical conductivity, good chemical stability, environmental friendliness, low cost and large pseudocapacitance, which could result in higher specific capacitance than carbon-based supercapacitors.^4–6 Generally speaking, the pseudocapacitance of PPy is based on the fast and reversible faradic reaction (or faradic charge transfer reactions) at the vicinity of the electrode surface.^7,8 Therefore, the electrochemical properties of the PPy electrode are strongly affected by the thickness of PPy layers. If the thick PPy layers are used for electrode material of supercapacitors in bulk form, the electrochemical properties would be seriously degraded, because that the internal electrode material cannot be fully used. In addition, the volumetric swelling and shrinkage of PPy is inevitable as a result of repeated insertion/de-insertion of dopant ions during the charge–discharge cyclic process, which can be imparted to the poor cyclic stability of PPy electrodes.⁹ It is very important for improving the electrochemical performance of the PPy-based supercapacitors to elaborately design the nanostructure and microscopic morphology of PPy in bulk form, which can facilitate electrolytes uptake and dopant ion access, improve the electronic and ionic conductivity, enhance mechanical stability and PPy surface area.^10–13

In recent years, cellulose has been successfully applied at electrode of the supercapacitor due to their renewability, wide availability, and porous nature.^14–18 In particular, cellulose nanomaterials (including cellulose nanofibers (CNFs), bacterial cellulose, and so on) have also been widely employed as a porous substrate material of the PPy-based electrode. That's not only because CNFs possess low cost, environmentally friendly, but also because microscopic morphology of the PPy/CNFs aerogel and PPy layers covered on the surface of CNFs can be controlled to a certain degree.^19–22 However, the composites of conducting polymers with CNFs (even including the PPy/CNFs aerogel) are generally prepared by simple mixing CNFs with the monomer followed by addition of the oxidant (such as Fe³⁺). Unfortunately, we found that the formation of the PPy/CNFs hydrogels (hydrogen-bonding or Fe³⁺ crosslinked) is too quick to uniformly mix the CNFs, monomer and Fe³⁺ after adding the Fe³⁺ oxidizing agent by repeating the reported experiments. Therefore, the resulting PPy/CNFs materials, especially the PPy/CNFs aerogels seriously suffer from a lack of effectively control over the hierarchical nanostructure and properties of the PPy-based electrode material, and thereby cannot form uniform PPy/CNFs hydrogel in our view. Thus, it is very important to design an ingenious way to realize the controlled in situ release of Fe³⁺ oxidant to initiate polymerization of pyrrole (Py) monomer and then produce throughout uniform PPy/CNFs hydrogel.

On the other hand, multivalent metal ions (e.g. Fe³⁺) can be controlled release from metal ions-based colloid suspension with common stimuli being pH.²³ Therefore, the Fe³⁺-based colloid suspension can be used as precursor of Fe³⁺ oxidant to effectively control deposition of PPy on the CNFs surface. In addition, the pH of the Fe³⁺-based colloid suspension (used as Fe³⁺ oxidant precursor of the PPy/CNFs hydrogel) should be controlled around 7.0 due to the change of state from CNFs suspension to CNFs hydrogel under low pH condition. However, ferric hydroxide colloid cannot be suspended stably under high pH condition (about pH > 5–6). In order to solve the above described problem, the new Fe³⁺-based colloid should be redesigned so that they can be stably suspended under neutral pH condition. Recently, we discovered that the citric acid-Fe³⁺ (CIT-Fe³⁺) colloid suspension system not only can be stably suspended under neutral or even slightly alkaline pH condition but also can completely release Fe³⁺ to initiate polymerization of pyrrole monomer at pH < 2. Therefore, the CIT-Fe³⁺ colloid is a perfect precursor of the Fe³⁺ oxidant, which can effectively control over the microscopic morphology of the PPy/CNFs aerogel and PPy layers on the CNFs surface.

In this paper, we endeavour to controlled prepare of the PPy/CNFs aerogel using CIT-Fe³⁺ colloid as precursor of Fe³⁺ oxidant. Then, the PPy/CNFs aerogel-based all solid state supercapacitors are also been successfully prepared. The results indicate that the microscopic morphology of the PPy/CNFs aerogel and PPy layers can be effectively controlled by the CIT-Fe³⁺ colloid. Furthermore, the PPy/CNFs aerogel-based all solid-state supercapacitors possess excellent electrochemical properties.

2. Experimental

2.1 Materials

Hardwood bleached kraft pulps in the never-dried wet state with 78% water content. 2,2,6,6-Tetramethylpiperidine-1-oxyl radical (TEMPO), NaBr, sodium hypochlorite solution, pyrrole (Py), citric acid, NaOH, FeCl₃, polyvinyl alcohol (PVA), LiCl and HCl were obtained from commercial suppliers. Fresh pyrrole was used.

2.2 Preparation of the CIT-Fe³⁺ colloid

FeCl₃·6H₂O (2.71 g) was added into 100 mL of deionized water. Then 1.92 g of citric acid was added with continuous stirring until the solution became clear. The pH of the mixture solution was adjusted to neutral or even slightly alkaline pH condition with NaOH.

2.3 Preparation of the PPy/CNFs aerogel

CNFs are prepared according to the methodology reported by Isogai,^24,25 and described in brief as follows: cellulose fibers (2 g) were suspended in 400 mL of water containing TEMPO (0.033 g) and sodium bromide (0.33 g). The oxidation reaction was started by adding the desired amount of the NaClO solution (15 mmol g⁻¹ cellulose). The pH of the reaction solution was maintained to be 10.00 at 10 °C by adding 0.5 M NaOH for 6 h. The oxidized cellulose was thoroughly washed with water by filtration. 2 mg mL⁻¹ oxidized cellulose/water slurries were sonicated for 15 min at power of 300 W in an ice bath. CNFs dispersion was centrifuged at 10 000 rpm for 10 min to remove the unfibrillated cellulose. The transparent CNFs dispersion was stored at 4 °C before used. The detailed characterizations of the CNFs are provided in the ESI (Fig. S1–S3†).

A certain amount of pyrrole (3.4 μL, 11.4 μL, 21.5 μL, 34.8 μL) was added into 20 g of CNFs dispersion (0.25 wt%) (the mass ratios of pyrrole to CNFs are 5 : 95, 15 : 85, 25 : 75, and 35 : 65). The mixture solution was stirred for about 5 minutes, and then a certain amount (about 1.1026 g, 3.6971 g, 6.9834 g, 11.2809 g) of CIT-Fe³⁺ (the molar ratio of pyrrole to CIT-Fe³⁺ is about 2.1) was added with continuous stirring until the mixture solution became homogeneous. The pyrrole/CNFs/CIT-Fe³⁺ suspension was poured into mould, then exposure to the hydrochloric acid vapor for about 12 h to get well formed at about 4 °C. Then the PPy/CNFs hydrogel was repeated rinsed with distilled water (pH 1.0 adjusted by HCl) for about one week to remove inorganic ions, followed by neutral deionized water until the resulting hydrogel reached a neutral pH. The resulting hydrogel was further turned into the alcogel by using alcohol to replace the water within the network of the hydrogel. Finally, the PPy/CNFs alcogel was dried with supercritical CO₂ to obtain the PPy/CNFs aerogel. The formation process of the PPy/CNFs aerogel is shown in Fig. 1.

Fig. 1 Schematic diagram for the controllable preparation of the PPy/CNFs aerogel.

2.4 Preparation of the all solid-state supercapacitors

8.5 g of LiCl was added into 40 mL of deionized water, and then 4 g of polyvinyl alcohol (PVA) was added with continuous stirring at 85 °C until the solution became clear. Two pieces of PPy/CNFs aerogel were compressed into aerogel films under a pressure of about 5 MPa. Then, two pieces of aerogel films (one side of PPy/CNFs aerogel films were coated by silver paste, and the edge of one side glued to the aluminum foil, about 15.5 mm × 10.0 mm) were soaked in the LiCl–PVA gel electrolyte for 60 min and picked out. After that, two electrode films were left in the fume hood at room temperature for about 4 h to vaporize the excess water. Finally, the two electrode films were assembled into all-solid-state supercapacitors (as shown in Fig. S4†).

2.5 Characterizations

Scanning electron micrographs (SEMs) were recorded on a Hitachi S-4800 scanning electron microscope operated at 5.0 kV. Rheological tests were conducted on a Physica MCR 301 Rheometer with 25 mm diameter parallel-plate geometry at 25 °C. Nitrogen sorption measurements were performed with ASAP 2010 (Micromeritics, USA) to obtain pore properties such as the BET-specific surface area. XRD data were collected with an X'Pert Pro MPD (PANalytical, The Netherlands) diffractometer using monochromatic Cu Kα 1 radiation (k = 1.5406 Å) at 40 kV and 40 mA. XPS was performed using an AXIS Ultra spectrometer with a high-performance Al monochromatic source operated at 15 kV. The electrochemical performances of all-solid-state supercapacitors were tested by cyclic voltammetry (CV), galvanostatic charge/discharge, and electrochemical impedance spectroscopy (EIS, on a CHI 660E, CH Instruments, Inc). The potential range for CV measurements and galvanostatic charge/discharge tests was 0–0.8 V. EIS tests was carried out in the frequency range of 10⁵ to 0.01 Hz at a 10 mV amplitude referring to open circuit potential. The transmission electron microscope (TEM) image was obtained on a JEM-200CX.

3. Results and discussion

The precursor of Fe³⁺ oxidant-CIT-Fe³⁺ colloid can be stably suspended under slightly alkaline pH condition (Fig. 2) due to sufficiently negative surface charge (CIT-Fe³⁺ colloid had ζ-potentials from −19 to −24 mV). The transmission electron microscope (TEM) image of the CIT-Fe³⁺ colloid is shown in Fig. 2, revealing that the CIT-Fe³⁺ colloid can be well separated from each other, and showing a particle size in nanoscale range. The PPy/CNFs hydrogels are prepared by acidizing the homogeneous suspension of CNFs, pyrrole monomer, and CIT-Fe³⁺ colloid with hydrochloric acid vapor, which can not only promote the uniform gel formation, but also control Fe³⁺ oxidant release to initiate pyrrole polymerization. The formation of the PPy/CNFs hydrogel is confirmed by rheology measurements (shown in Fig. 3a). It is obvious that each storage modulus (G′) value is always one order of magnitude higher than the respective loss modulus (G′′) value over the entire frequency range measured. This indicates that elastic response is predominant and the PPy/CNFs hydrogels have a permanent network. Furthermore, the G′ value of the PPy/CNFs hydrogels gradually increases with the increase of the content of PPy, which even exhibits a nearly linear relationship (Fig. 3b). This phenomenon indicates that PPy may be gradually uniform coated onto the CNFs (including junctions between CNFs) to form PPy layers with the controlled release Fe³⁺ oxidant from CIT-Fe³⁺ colloid. The new PPy layers coated on the CNFs (especially junctions) may be able to gradually improve the mechanical strength of the PPy/CNFs hydrogel with the increase of the content of PPy.

Fig. 2 Photograph and TEM image of the CIT-Fe³⁺ colloid.

Fig. 3 (a) Dynamic rheological behaviors of the PPy/CNFs hydrogels, and CNFs hydrogels, (b) the linear relation between storage modulus (G′) and PPy content at frequencies of 10 rad s⁻¹.

The PPy/CNFs aerogels are prepared from the PPy/CNFs alcogels by supercritical CO₂ drying (Fig. S5†). The morphology and hierarchical nanostructure of the PPy/CNFs aerogels are characterized by scanning electron microscopy (SEM) observations (Fig. 4a–d). It can be clearly illustrated that the randomly oriented 1D fiber-like structure (composed of CNFs and PPy) could connect with each other and form a well-defined and open interconnected 3D porous network with pore size of several nanometers. However, the morphology of the PPy/CNFs aerogels changes significantly with the increase of the PPy content. The morphology of PPy/CNFs aerogels becomes more and more compact with the increase of PPy content probably due to more and more PPy are coated around 1D fiber-like pore walls. Nonetheless, the 1D fiber-like pore walls can be still effectively preserved even if the PPy content of the PPy/CNFs aerogels is up to 35%. In addition, PPy seems to be always controllable deposited around 1D fiber-like pore walls and ultimately form a homogeneous coating of PPy on the 1D fiber-like pore walls with different PPy content. This wonderful uniform morphology of the PPy/CNFs aerogels strongly indicates that the CIT-Fe³⁺ colloid as the precursor of Fe³⁺ oxidant can effectively control over the microscopic morphology of PPy layers in the PPy/CNFs aerogels.

Fig. 4 SEM images of the (a) PPy/CNFs hydrogels (5%) aerogel, (b) PPy/CNFs hydrogels (15%) aerogel, (c) PPy/CNFs hydrogels (25%) aerogel, and (d) PPy/CNFs hydrogels (35%) aerogel.

Fig. 5 presents the nitrogen adsorption–desorption isotherms and the Barrett–Joyner–Halenda (BJH) pore diameter distribution of the PPy/CNFs (25%) aerogel. Fig. 5a displays that the PPy/CNFs (25%) aerogel exhibits a typical hysteresis (type-IV), which implies that the pore structures of the PPy/CNFs (25%) aerogel are dominated by mesopores. The pore diameter distribution curve (Fig. 5b) clearly shows that much of the pore diameter lies in the 1.3–138.4 nm range with cumulative pore volume of about 1.12 cm³ g⁻¹. Furthermore, a shoulder peak also appears at around 2.1 nm. The Brunauer–Emmett–Teller (BET) specific surface area of the PPy/CNFs (25%) aerogel is about 402 m² g⁻¹, which is much higher than 246 m² g⁻¹ reported by carlsson.²⁶ This unique porous property of the PPy/CNFs (25%) aerogel indicates that the CIT-Fe³⁺ colloid is an effective precursor of Fe³⁺ oxidant, and Fe³⁺ oxidant can be controlled release from the CIT-Fe³⁺ colloid to effectively control over the microscopic morphology of the PPy/CNFs aerogel.

Fig. 5 Nitrogen adsorption/desorption isotherms (a) and BJH pore distribution (b) of the PPy/CNFs hydrogels (25%) aerogel.

X-ray photoelectron spectroscopy (XPS) is used for characterizing the protonation degree of the PPy for the PPy/CNFs (25%, 35%) aerogel (Fig. 6a and S10†). The N 1s spectrum of sample can be deconvoluted to three component peaks: the imine nitrogen (–N=) at 398.2 eV, the amine nitrogen (–NH–) at 399.5 eV, and nitrogen cationic radical (N⁺) at 401.5 eV. The proportion of positively charged nitrogen atoms in the PPy coating layer (PPy/CNFs (25%) aerogel, and PPy/CNFs (35%) aerogel) is found to be about 20.2%, and 20.5% respectively. Fig. 6b shows the X-ray diffraction (XRD) patterns of CNFs, and PPy/CNFs (25%) aerogel. The diffractogram of two samples are very similar to the diffractogram of cellulose I crystalline structure.^27,28 The result indicates that CNFs are well preserved in the PPy/CNFs (25%) aerogel. In addition, these CNFs in the PPy/CNFs aerogels may be uniformly dispersed in the 1D fiber-like pore walls. Therefore, the new kind of PPy/CNFs aerogels could also exhibit an excellent re-swell performance in aqueous electrolytes, which can significantly enhance the wettability of 1D fiber-like pore walls to improve the utilization efficiency of the surface due to the presence of CNFs with excellent hydrophilic.

Fig. 6 (a) N 1s XPS spectrum of the PPy/CNFs (25%) aerogel, (b) XRD of the CNFs, and PPy/CNFs hydrogels (25%) aerogel.

The PPy/CNFs aerogel (25%, and 35%) film-based solid-state supercapacitors are fabricated in a traditional two-electrode stacked configuration by using LiCl–poly(vinyl alcohol) (PVA) gel as solid-state electrolyte and separator (Fig. S6†).^29,30 The capacitance performances of the PPy/CNFs aerogel (25%, and 35%) film-based solid-state supercapacitors (F-SC-25%, F-SC-35%) are assessed by cyclic voltammetry (CV), galvanostatic charge–discharge (GCD), and electrochemical impedance spectroscopy (EIS) at room temperature. The CV curves of F-SC-25% and F-SC-35% at different scan rates (from 5 to 200 mV s⁻¹) with the potential window ranging from 0 to 0.8 V are shown in the Fig. 7a and b, respectively (the electrochemical properties of F-SC-5% and F-SC-15% are shown in Fig. S7 and S8†). These CV curves all exhibit typical symmetry and quasi-rectangular shape, indicating the good capacitive performance of the PPy/CNFs aerogel film-based solid-state supercapacitors. Furthermore, it is obvious that the symmetry and quasi-rectangular shape of F-SC-25% does not change much even at scan rate of 200 mV s⁻¹. However, the CV profile of F-SC-35% has undergone significant distortion at a scan rate of 200 mV s⁻¹. The significantly dissimilar voltammetric behavior for the PPy/CNFs aerogel film-based solid-state supercapacitors with different PPy contents may be explained by the electrolyte ions-transport rate limitations in the electrode. Generally speaking, the thinner uninterrupted layers of PPy coating deposited on the CNFs, the more favourable electronic conductivity and reversibly diffusion of counter electrolyte ions in the PPy layers during the oxidation and reduction processes. Therefore, the F-SC-25% exhibits better capacitance performance than F-SC-35% due to more thinner of PPy coating for F-SC-25% than F-SC-35%. Galvanostatic charging and discharging tests of F-SC-25% and F-SC-35% at differently current densities are conducted within the potential window range from 0 to 0.8 V (Fig. S9† and 7c). The results reveal that all charging and discharging curves show an almost symmetric triangular profile, indicating good capacitive behavior of the PPy/CNFs aerogel-based all solid-state supercapacitors. Furthermore, it is clearly seen that the iR-drop of F-SC-35% at current densities in the range between 0.09 and 1.92 mA cm⁻² was significantly smaller, due to the excellent electrical conductivity of the uninterrupted PPy coating around CNFs. The specific electrode capacitances are calculated from the different current density discharge curves after iR-drop, and the results are shown in the Fig. 7d. The maximum specific electrode capacitance is about 215 F g⁻¹ at current density of 0.19 mA cm⁻². However, at a current density of 0.09 mA cm⁻², the maximum specific electrode capacitance of F-SC-35% is only 203 F g⁻¹. More importantly, it can be seen that the specific electrode capacitance of F-SC-25% is always higher than that of F-SC-35% at all current densities studied. The possible reason is that, with the increase of PPy content using the CIT-Fe³⁺ colloid as precursor of Fe³⁺ oxidant, the microstructure of the PPy/CNFs aerogel becomes more compact, and the PPy coating becomes thicker and more uniform. This phenomenon means that the reversibly diffusion of counter electrolyte ions (the reversibly diffusion of ions from the electrolyte to the PPy coating, and especially the reversibly diffusion of ions in the PPy coating) becomes more difficult. Therefore, the thinner PPy coating is more favorable full realization energy storage properties of PPy layers. Owing to the more compact microstructure and thicker PPy layers, the specific electrode capacitance of F-SC-35% is inferior to F-SC-25% at all current densities studied. Ragone plots further demonstrate the better performance of the F-SC-25% than F-SC-35% (Fig. S11†). The F-SC-25% is able to achieve a energy density of 4.436 W h kg⁻¹ at a power density of 141.89 W kg⁻¹. However, the PPy/CNFs aerogel-based all solid-state supercapacitors showed not good capacitance retention (about 11%, current densities from 0.09 to 2.1 mA cm⁻²), perhaps due to using of LiCl-PVA solid-state gel electrolyte and chloride dopant.³¹

Fig. 7 Typical CV curves of F-SC-25% (a), F-SC-35% (b) device at different scan rates. (c) Typical galvanostatic charge–discharge curves of F-SC-35% at different current densities. (d) The specific capacitance of F-SC-25%, F-SC-35% as a function of the current density.

Electrochemical impedance spectroscopy (EIS) is used to further evaluate the electrochemical behavior of the PPy/CNFs aerogel-based all solid-state supercapacitors (Fig. 8a). The equivalent series resistance values (the intercept of the Nyquist curve on the x-axis) of F-SC-25%, and F-SC-35% are 4.2 and 4.3 Ω, respectively, indicating the good conductivity of the PPy/CNFs aerogel film-based electrode. At medium frequency, the projected length of the Warburg-type line with a slope of about 45° on the x-axis is related to the reversibly diffusion of electrolyte ions in the PPy/CNFs aerogel film-based electrode. The projected length of F-SC-25% is significantly shorter than that of F-SC-35%, indicating the easier reversibly diffusion in the F-SC-25%, probably due to the thinner PPy coating, and the looser porous structure of the PPy/CNFs aerogel (25%). In the low frequency region, the Nyquist plots of F-SC-25%, and F-SC-25% exhibit the vertical shape, showing an ideal capacitance behavior. Furthermore, the electrochemical cycling stability tests of F-SC-25% and F-SC-35% are carried out by using a galvanostatic charge–discharge at a current density of 3.1 mA cm⁻² and 3.2 mA cm⁻² for 2000 cycles (Fig. 8b). Interestingly, the specific capacitance of the F-SC-35% does not decrease but gradually increases with the increase of the cycle number, and increases to 111.7% after 2000 cycles. The highly possible reason is that the PPy coating around CNFs of the PPy/CNFs (35%) aerogels exhibit thicker compared to that of the PPy/CNFs (25%) aerogels. Generally speaking, many internal PPy of the thicker PPy coating is inoperative at first, and the diffusion of electrolyte ions is a slow process in the interior of the PPy coating. However, CNFs have excellent hydrophilicity. Therefore, the CNFs enclosed in the PPy coating can be used as internal aqueous electrolyte nano-reservoirs, which can not only effectively improve the accessible surface area of the PPy coating with aqueous electrolyte but also significantly reduce the ion diffusion distance from aqueous electrolyte nano-reservoirs to the interior of the PPy coating. The long time charging and discharging can let the ions (including Li⁺ and Cl⁻) gradually diffuse into the deeper layer of PPy coating. These ions maybe can not only significantly improve the content of PPy impregnated with an electrolyte solution, but also improve the proportion of positively charged nitrogen atoms in the PPy coating layer, and ultimately effectively reduce the content of inoperative PPy. Therefore, the electrochemical cycling stability of F-SC-35% gradually increases with the increase of the cycle number. Furthermore, the specific capacitance of the F-SC-25% gradually decreases with the increase of the cycle number. However, only a 3.2% fall in the initial specific capacitance is observed after 2000 cycles. The results indicate that the cyclic stability of the CNFs/PPy aerogel electrode can be significantly improved when PPy is combined with CNFs under the controlled conditions.

Fig. 8 (a) Nyquist impedance plots of the F-SC-25% and F-SC-35%. (b) Cycling stability of the F-SC-25% and F-SC-35% over 2000 cycles.

4. Conclusions

In conclusion, the PPy/CNFs aerogel is prepared by using the CIT-Fe³⁺ colloid as precursor of Fe³⁺ oxidant. The microscopic morphology of the PPy/CNFs aerogel and PPy layers on the CNFs surface can be effectively controlled due to in situ controlled release of Fe³⁺ oxidant to initiate polymerization of pyrrole monomer with common stimuli being pH. The BET specific surface area of the PPy/CNFs (25%) aerogel is up to 402 m² g⁻¹. All solid-state film supercapacitors are fabricated by using the PPy/CNFs aerogel film as the electrode material. The maximum specific electrode capacitances of F-SC-25% is 215 F g⁻¹ (0.19 mA cm⁻²). The F-SC devices also exhibit excellent cycle stability.

Acknowledgements

Financial support was kindly supplied by grants from National Natural Science Foundation of China (No. 21501154, and No. 21601162), Important Research Project at the University of Henan Province (16A430031), and Doctoral Research Foundation of Zhengzhou University of Light Industry.

References

1. M. Winter and R. J. Brodd, Chem. Rev., 2004, 104, 4245–4269.
2. A. S. Arico, P. Bruce, B. Scrosati, J. M. Tarascon and W. Van Schalkwijk, Nat. Mater., 2005, 4, 366–377.
3. P. Simon and Y. Gogotsi, Nat. Mater., 2008, 7, 845–854.
4. C. Meng, C. Liu, L. Chen, C. Hu and S. Fan, Nano Lett., 2010, 10, 4025–4031.
5. F. Meng and Y. Ding, Adv. Mater., 2011, 23, 4098–4102.
6. Q. Niu, K. Gao, Z. Lin and W. Wu, Anal. Methods, 2013, 5, 6228–6233.
7. A. Razaq, L. Nyholm, M. Sjodin, M. Stromme and A. Mihranyan, Adv. Energy Mater., 2012, 2, 445–454.
8. L. Nyholm, G. Nystrom, A. Mihranyan and M. Stromme, Adv. Mater., 2011, 23, 3751–3769.
9. E. Frackowiak, V. Khomenko, K. Jurewicz, K. Lota and F. Beguin, J. Power Sources, 2006, 153, 413–418.
10. Z. Yin and Q. Zheng, Adv. Energy Mater., 2012, 2, 179–218.
11. D. Wei, X. Lin, L. Li, S. Shang, M. C.-W. Yuen, G. Yan and X. Yu, Soft Matter, 2013, 9, 2832–2836.
12. H. D. Tran, D. Li and R. B. Kaner, Adv. Mater., 2009, 21, 1487–1499.
13. L. Nyholm, G. Nystrom, A. Mihranyan and M. Stromme, Adv. Mater., 2011, 23, 3751–3769.
14. K. Gao, Z. Shao, J. Li, X. Wang, X. Peng, W. Wang and F. Wang, J. Mater. Chem. A, 2013, 1, 63–67.
15. K. Gao, Z. Shao, X. Wu, X. Wang, Y. Zhang, W. Wang and F. Wang, Nanoscale, 2013, 5, 5307–5311.
16. L.-F. Chen, Z.-H. Huang, H.-W. Liang, H.-L. Gao and S.-H. Yu, Adv. Funct. Mater., 2014, 24, 5104–5111.
17. X. W. Shi, Y. L. Hu, M. Z. Li, Y. W. Y. Duan, Y. X. Wang, L. Y. Chen and L. N. Zhang, Cellulose, 2014, 21, 2337–2347.
18. Q. Niu, K. Gao and Z. Shao, Nanoscale, 2014, 6, 4083–4088.
19. Z. Shi, S. Zang, F. Jiang, L. Huang, D. Lu, Y. Ma and G. Yang, RSC Adv., 2012, 2, 1040–1046.
20. H. Wang, L. Bian, P. Zhou, J. Tang and W. Tang, J. Mater. Chem. A, 2013, 1, 578–584.
21. K. Gao, Z. Shao, X. Wang, Y. Zhang, W. Wang and F. Wang, RSC Adv., 2013, 3, 15058–15064.
22. D. O. Carlsson, G. Nystrom, Q. Zhou, L. A. Berglund, L. Nyholm and M. Stromme, J. Mater. Chem., 2012, 22, 19014–19024.
23. H. Huang, S. Lu, X. Zhang and Z. Shao, Soft Matter, 2012, 8, 4609–4615.
24. A. Isogai, T. Saito and H. Fukuzumi, Nanoscale, 2011, 3, 71–85.
25. T. Saito, Y. Nishiyama, J.-L. Putaux, M. Vignon and A. Isogai, Biomacromolecules, 2006, 7, 1687–1691.
26. D. O. Carlsson, G. Nystrom, Q. Zhou, L. A. Berglund, L. Nyholm and M. Stromme, J. Mater. Chem., 2012, 22, 19014–19024.
27. Y. Okita, T. Saito and A. Isogai, Biomacromolecules, 2010, 11, 1696–1700.
28. R. J. Moon, A. Martini, J. Nairn, J. Simonsen and J. Youngblood, Chem. Soc. Rev., 2011, 40, 3941–3994.
29. Z. Weng, Y. Su, D.-W. Wang, F. Li, J. Du and H.-M. Cheng, Adv. Energy Mater., 2011, 1, 917–922.
30. G. Wang, X. Lu, Y. Ling, T. Zhai, H. Wang, Y. Tong and Y. Li, ACS Nano, 2012, 6, 10296–10302.
31. J. Wang, Y.-L. Xu, X. Chen, X.-F. Du and X.-F. Li, Acta Phys.–Chim. Sin., 2007, 23, 299–304.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23216g

---