Porous nitrogen-doped carbon tubes derived from reed catkins as a high-performance anode for lithium ion batteries

Abstract

Crude protein-containing reed catkins with a natural tubular structure are employed as a carbon precursor to prepare porous nitrogen-doped carbon tubes (PNCTs) by ZnCl₂ activation. The obtained PNCTs, with a large specific surface area of 1751.1 m² g⁻¹ and satisfactory N-content of 4.22 wt%, exhibit a super-high specific capacity of 2016.3 mA h g⁻¹ at 0.02 A g⁻¹, and an excellent cycling stability of 512.4 mA h g⁻¹ after 500 cycles at 1 A g^−1, as an anode material for lithium ion batteries (LIBs).

---

Introduction

In recent years, a host of materials as anodes for lithium ion batteries (LIBs), such as metal oxides and silicon or tin based materials have risen into sight.^1–9 Despite their great potential, in the field of commercial LIB anodes, carbon is in the driver’s seat, for it possesses high electrical conductivity and great cyclic stability as well as low cost.^10–16 However, the low specific capacity of 372 mA h g⁻¹ and limited rate capability of commercial graphite anodes are hardly satisfactory in this era when electronics are widespread and the electrical vehicle market is booming.^17,18

Porous carbon materials derived from biomass are promising and have attracted a lot of attention from researchers, due to not only the large material–electrolyte interface and shorter diffusion pathways which facilitate charge and ion transfer, but also the abundance, easy access, fast regeneration, benignity to the environment and very low cost.^19,20 Biomass like silk,²¹ peanut shells,²² wheat straw,²³ rice husk,²⁴ egg whites,²⁵ ox horns,²⁶ cherry stones,²⁷ etc., has been used as a carbon precursor to obtain porous carbon materials for use as anodes for LIBs with a high capacity and rate capability.

In order to obtain porous carbon materials from biomass, a series of activating agents can be employed such as KOH,²⁸ ZnCl₂,²⁹ H₃PO₄,³⁰ K₂CO₃,³¹ etc. Among these, KOH is the most utilized activating agent which can lead to abundant pores with different pore sizes by reacting with carbon in biomass precursors at a high temperature to etch the carbon frame.²⁸ Nevertheless, the major problems of KOH activation are the serious destruction of the carbon skeleton and the low yield. Compared with KOH, ZnCl₂ is a relatively moderate activating agent used to mainly generate micropores and mesopores in the target carbon products.³² During activation by ZnCl₂, which works as a dehydration reagent,^32,33 the carbon frame of the precursors can be kept, and a high yield can be achieved.^28,34,35 Furthermore, nitrogen doping has been proven to be an effective way to improve the electrochemical performance of porous carbon anodes for LIBs.^21,23,25 The heteroatom of N in the porous carbon anodes can exhibit favorable binding with Li because of its strong electronegativity, and offer more active sites for Li storage by creating defects in the carbon.^25,36–38

Reed is a kind of widespread tallgrass growing beside irrigation ditches, marshes and river levees, and it consists of leaves, catkins, stems and roots. Among these components, reed catkins, which are the flowers of the reed, are greatly attractive for they possess a natural tubular structure. Moreover, unlike reed stems which can be used in papermaking, reed catkins are almost of no use, being cut from the whole reed and left behind to rot. In this paper, we report a facile economic approach to prepare porous nitrogen-doped carbon tubes (PNCTs) using reed catkins as a carbon precursor and ZnCl₂ as an activating reagent (Scheme 1). Reed catkin is a natural carbon tube composed of not only cellulose and lignin, but also some crude protein which is suitable as a native nitrogen source. Moreover, ZnCl₂ as the activating reagent can keep the tubular structure intact during the activating process. The as-prepared PNCTs with good tubular structure, a large specific surface area of 1751.1 m² g⁻¹ and a satisfactory N-content of 4.22 wt%, exhibit a high specific capacity and excellent cycling stability as an anode material for LIBs.

Scheme 1 Schematic illustration of the synthesis procedure of PNCTs.

Experimental

Preparation of PNCTs

As shown in Scheme 1, reed catkins were separated and collected from wild reed, with removal of the seeds. This raw material was dried in an air-circulating oven after having been cleaned by deionized water. Subsequently, the pre-treated sample was immersed in a ZnCl₂ solution for 24 h at a weight ratio of 1 : 3 (the sample to ZnCl₂), then dried at 120 °C for 5 h. Pyrolysis was conducted at 600 °C for 1 h in a tube furnace under argon with a heating rate of 3 °C min⁻¹. The carbonized product was collected and immersed in a 0.2 mol L⁻¹ HCl solution for 30 min before being washed with a 0.2 mol L⁻¹ HCl solution, followed by deionized water to neutralise, in order to remove the residual ZnCl₂ and other impurities. After drying at 80 °C, the final product PNCTs were obtained. As a comparison, the material without ZnCl₂ activation (denoted NCTs) was also prepared through a similar approach but without the immersion in ZnCl₂.

Material characterization

The as-prepared products were characterized using various techniques. X-ray diffraction (XRD) characterization was carried out using a TD-3500 X-ray powder diffractometer (Dandong, China) with Cu Kα radiation (λ = 1.5418 Å) running at 30.0 kV, and 20.0 mA. X-ray photoelectron spectroscopy (XPS) was conducted on a Kratos XSAM 800 spectrometer (Manchester, UK). Raman spectra were recorded on a Confocal LabRAM HR800 spectrometer, HORIBA Jobin Yvon (Paris, France). Field emission scanning electron microscopy (FESEM) and high-resolution FESEM were performed on a Hitachi S4800 scanning electron microscope (Tokyo, Japan) and a Hitachi SU8010 scanning electron microscope (Tokyo, Japan), respectively. Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) images were obtained using a FEI Tecnai G2 20 transmission electron microscope (Hillsboro, OR, USA) operating at 200 kV. The specific surface area and pore size distribution were obtained from the results of nitrogen (N₂) adsorption and desorption using a Micromeritics TriStar II 3020 automatic analyzer (Norcross, GA, USA), using the Brunauer–Emmett–Teller (BET) technique and the density function theory (DFT) model, with the samples degassed at 120 °C for 24 h before adsorption measurement.

Electrochemical measurements

The working electrodes were prepared by mixing the 80 wt% active material, 10 wt% polyvinylidene fluoride (PVDF) and 10 wt% acetylene black (AB), then dispersing them in N-methylpyrrolidone (NMP) to form a uniform slurry. The slurry was then spread onto copper foil and vacuum dried at 100 °C. The working electrodes with a mass loading of around 1.2 mg cm⁻² were assembled into CR2032 coin-type cells with Li metal acting as the counter electrode, Celgard 2300 microporous film as the separator, and 1.0 M LiPF₆ solution in ethylene carbonate (EC) and dimethyl carbonate (DMC) (volumetric ratio EC : DMC = 1 : 1) as the electrolyte in a glovebox (Delix LS800S, Chengdu, China) filled with Ar. The cells were aged for 24 h to make sure that the electrolyte fully permeated the electrodes before measurement. Galvanostatic charge–discharge tests were carried out between 0 and 3 V at room temperature with different current densities on a Neware CT-3008W multichannel battery measurement system (Shenzhen, China). Cyclic voltammetry (CV) measurements were conducted on an Autolab PGSTAT 302 electrochemical workstation (Utrecht, the Netherlands) in the voltage range 0–3 V (vs. Li⁺/Li) at a scan rate of 0.1 mV s⁻¹.

Results and discussion

Reed catkins were chosen as the nitrogen-containing carbon precursor to prepare PNCTs not only because they are widespread and an available waste product, but also for the reason of their natural tubular structure, which makes it possible to obtain porous carbon tubes by chemical activation and carbonization. Characterization and comparison of the carbon tubes derived from reed catkins without ZnCl₂ activation (NCTs) and the porous carbon tubes obtained via ZnCl₂ activation (PNCTs) were conducted. As the SEM images in Fig. 1a and b show, the NCTs display a tubular structure with rough surfaces. In contrast, the PNCTs also exhibit a tubular structure but with smooth surfaces, as shown in Fig. 1c and d, displaying that the surfaces were smoothed during the ZnCl₂ activation process and the wrinkles were removed. The wrinkles on the surfaces of NCTs are believed to be unstable and easy to peel off. During the ZnCl₂ activation, these wrinkles would become loose and porous, making them much easier to fall off. Therefore, it can be inferred that the smooth surfaces of PNCTs are the result of the surface wrinkles being removed. Meanwhile, the similar tubular structure of NCTs and PNCTs illustrates that ZnCl₂ activation can keep the original structure of the reed catkins intact. This tubular structure can offer stable and continuous carbon frameworks which will allow electron transport and improve the electrochemical stability of electrode materials. Fig. 1e and f display the TEM and HRTEM images of the PNCTs. From Fig. 1e, it can be observed that the whole material is uniformly decorated with a great number of nanopores (micropores and small mesopores). As shown in Fig. 1f, the sample is mainly composed of disordered carbon domains and a few ordered domains displaying a low degree of graphitization.

Fig. 1 SEM images (a) and (b) of the NCTs; SEM images (c) and (d), TEM image (e), and HR-TEM image (f) of the PNCTs.

To further determine the porous structure, BET measurements of N₂ adsorption and desorption were employed, as shown in Fig. 2a. The Brunauer–Emmett–Teller (BET) specific surface area of the PNCTs is 1751.1 m² g⁻¹, much larger than that of the NCTs (104.8 m² g⁻¹), which demonstrates that ZnCl₂ activation greatly enlarged the specific surface area. The high specific surface area will contribute to a large electrode–electrolyte interface during the electrode reaction. The pore size distribution of the PNCTs, deduced from desorption data and computed from the isotherm using the density function theory (DFT) model, is displayed in Fig. 2b. It can be seen from the pattern that the pores in the PNCTs are mainly micropores and small mesopores within the size range of 1–2 nm and 2–4 nm, respectively, which is consistent with the TEM image. The high-resolution SEM was used to further observe the whole porous structure of the PNCTs (Fig. S1†). The mesopores can serve as Li⁺ reservoirs and pathways for the electrolyte, while the micropores are suggested to allow charge accommodation which will be essential for high energy storage.^39–41 Therefore, the PNCTs are expected to exhibit a high capacity as LIB anodes.

Fig. 2 (a) Nitrogen adsorption–desorption isotherms for NCTs and PNCTs; (b) PSD for the PNCTs.

The X-ray diffraction (XRD) patterns of both the NCTs and the PNCTs are shown in Fig. 3a. The patterns both show two broad diffraction peaks at approximately 22° and 43°, which correspond to (002) and (100) of the pseudographitic domains. An empirical parameter (R), defined as the ratio of the height of the (002) Bragg peak to the surrounding background, was used to signify the degree of graphitization of the samples.⁴² A schematic description of how R was calculated is shown in Fig. S2.† The R value of 1.94 for the PNCTs is smaller than that of 2.31 for the NCTs, indicating that the PNCTs are of lower graphitization. This result demonstrates that ZnCl₂ activation effectively increased the degree of disorder, and this relatively amorphous structure is beneficial for Li⁺ intercalation–deintercalation and could accommodate more Li ions for LIB applications.^43,44 Raman spectra for the NCTs and PNCTs are shown in Fig. 3b, which both display a wide D-band at ∼1350 cm⁻¹ and a G-band at ∼1590 cm⁻¹. It is known to all that the D-band can be attributed to edges, other defects, and disordered carbon, while the G-band belongs to the vibration of the sp² bonded carbon atoms in a graphite-like 2D hexagonal lattice in carbon materials.⁴⁵ The ratio of the intensity of the D-band and the G-band (I_D/I_G) is a measure of the degree of disorder of a carbon material.^44–46 It can be calculated that the I_D/I_G value of 1.02 for the PNCTs is higher than that of 0.91 for the NCTs, representing a higher degree of disorder, and more edges and defects after ZnCl₂ activation. This could provide more Li⁺ storage sites, allowing enhanced reversible capacity.^44,46

Fig. 3 (a) XRD spectra and (b) Raman spectra of the NCTs and PNCTs.

Combustion elemental analysis was performed to determine the content of hydrogen, carbon and nitrogen. The result shows that the PNCTs possess a hydrogen content of 3.68 wt%, a carbon content of 84.83 wt% and a nitrogen content of 4.37 wt%. The high H/C atomic ratio of 0.52 is beneficial to the reversible specific capacity, for Li atoms can bond in the vicinity of H atoms in the hydrogen-containing carbons.⁴⁷ X-ray photoelectron spectroscopy (XPS) was employed to investigate the nature of the nitrogen species at the surface, shown in Fig. 4. As the full wide-scan spectrum of the PNCTs (Fig. 4a) displays, there exist three strong peaks corresponding to C 1s, O 1s and N 1s. The amounts of nitrogen and oxygen were 4.22 wt% and 7.05 wt% respectively, and the value of the N-content is close to the elemental analysis result. Nitrogen possesses higher electronegativity compared with carbon because of its smaller atomic diameter. Meanwhile, the interaction between Li⁺ and the nitrogen-doped carbon materials is stronger than that between Li⁺ and the carbon materials without N-doping because of the hybridization of the nitrogen lone pair electrons with the π electrons in the carbon. It is believed that the electronegativity and hybridization could make favorable binding sites for Li ion storage.⁴⁸ Furthermore, the existence of N atoms will create lots of defects in the carbon material and produce more active sites for Li⁺ storage.²⁵ The XPS spectrum of C 1s (Fig. 4b) can be divided into four peaks at 284.7, 285.6, 286.0 and 289.1 eV. Among them, the peak at 284.7 eV belongs to the sp² hybridized carbon (C–C), 285.6 eV corresponds to carbon atoms single bonded to nitrogen or oxygen in the form of pyrrolidonic, phenol or ether (C–N/C–O), 286.0 eV is attributed to the pyridinic carbon linked to nitrogen (C=N/C=O) and 289.1 eV is related to O=C–O bonding.^49,50 The corresponding atomic percentages and the binding energies are listed in Table 1. The N 1s spectrum of the PNCTs (Fig. 4c) can be split up into 4 peaks at 398.5, 400.3, 401.1 and 402.6 eV, corresponding to pyridinic N, pyrrolic/pyridonic N, quaternary N and oxidized N, respectively. Among them, according to theoretical calculations, pyridinic N is advantageous to Li⁺ storage which is a critical factor in the enhancement of reversible capacity.⁴⁸ Furthermore, the O 1s spectrum is divided into three peaks at 531.3, 532.4 and 533.6 eV (Fig. 4d), corresponding to the presence of the oxygen functionalities C=O, O–C–O/C–OH and O–C=O, respectively. These oxygen functionalities result in the surface redox reaction between the carbon–oxygen functional groups and Li⁺, providing extra Li⁺ storage sites.⁵⁰ A similar N content has been examined in the NCTs, and some metallic impurities have also been found such as K and Ca (Fig. S3†), which were not detected in the PNCTs and could have been eliminated during the washing process with HCl solution following ZnCl₂ activation. The elimination of these impurities could produce more active sites for Li⁺ storage, thus improving the capacity of the PNCTs.

Fig. 4 The total XPS spectrum (a) and the C 1s (b), the N 1s (c) and the O 1s (d) spectra of PNCTs.

Table 1 Peak assignment of C 1s, N 1s and O 1s for PNCTs

Peak	Binding energy (eV)	Assignment	Fraction of species (%)
C 1s	284.7	C1: sp^2 C–C	56.0
	285.6	C2: C–N/C–O	16.9
	286.0	C3: C=O/C=N	9.2
	289.1	C4: O=C–O	17.9
N 1s	398.5	N-6	22.5
	400.3	N-5	15.3
	401.1	N–Q	41.9
	402.6	N–X	20.3
O 1s	531.3	O1: C=O	25.8
	532.4	O2: O–C–O/C–OH	38.7
	533.6	O3: O–C=O	35.4

---

The electrochemical performances of the PNCTs as an anode material for LIBs were investigated in a half-cell with metallic lithium as the counter electrode. The cyclic voltammogram (CV) profiles of the PNCTs in Fig. 5a show that the PNCTs exhibit a typical CV curve of carbonaceous anode materials with an obvious cathodic peak at 0–1 V during the 1^st cycle and at 0–0.3 V during the 2^nd and 3^rd cycles. The intensity of the peak in the 1^st cycle is much stronger than in the following cycles. These differences are attributed to not only the irreversible consumption of charge that arises from the formation of the solid electrolyte interphase (SEI) layer, but also the loss of some Li in irreversible storage sites within the material.²⁵ For the same reason, the discharge curve of the PNCTs in the 1^st cycle shows a much higher capacity of 4509.6 mA h g⁻¹ at 0.02 A g⁻¹ than that of 2378.5 mA h g⁻¹ in the 2^nd cycle, shown in Fig. 5b. The discharge–charge curves of the PNCTs in Fig. 5b display that the initial coulombic efficiency is 50.7%. The formation of the SEI layer could prevent the electrolyte from suffering continuous decomposition on the carbon electrode and improve electrochemical stability of the materials for it acts as a passivation layer. The initial reversible capacity of the PNCTs is as high as 2284.8 mA h g⁻¹ at 0.02 A g⁻¹, which is retained at 2016.3 mA h g⁻¹ after 10 cycles. As shown in Fig. S4,† the NCT anode delivers an initial reversible capacity of 579.0 mA h g⁻¹ and a stable reversible capacity of 422.1 mA h g⁻¹ at 0.02 A g⁻¹, contrastingly. The higher reversible capacity of the PNCTs can be ascribed to its higher specific surface area and porous structure.

Fig. 5 Cyclic voltammograms of the PNCTs at a scan rate of 0.1 mV s⁻¹ (a), charge–discharge curves of the PNCTs at 0.1 A g⁻¹ (b), charge–discharge capacity versus cycle number of the NCTs and the PNCTs at different rates (c) and cycling performance and corresponding coulombic efficiency of the PNCTs at a current density of 1 A g⁻¹ (d).

Fig. 5c shows the specific capacities of the PNCTs and the NCTs in comparison at various discharge–charge current densities, with 10 cycles at each rate. It is obvious that the PNCT anode demonstrates a much higher reversible capacity than the NCT anode, delivering stable reversible capacities of 2016.3, 1442.1, 1038.7, 825.9, 711.5, 521.6 and 271.3 mA h g⁻¹ at current rates of 0.02, 0.05, 0.1, 0.3, 0.5, 1 and 3 A g⁻¹, respectively, and it can be recovered to 1010.4 mA h g⁻¹ when the discharge–charge current density returns to 0.1 A g⁻¹, indicating that the formed SEI and the electrodes are very stable during cycling. Although the rate capability of the PNCTs is not the most prominent among the biomass-derived porous carbons published by our group,^23,26,51 the PNCT electrode shows an ultra-high reversible capacity of 2016.3 mA h g⁻¹ at the low rate. In order to possess an overall understanding of the recent development of carbon anodes for LIBs, the reversible capacities and rate capabilities of carbon anodes derived from biomass precursors and other precursors reported recently have been summarized and listed in Tables S1 and S2,† respectively. By contrast, the reversible capacity of the PNCTs at the low rate is also remarkable. The super-high capacity may be attributed to: (i) the large specific surface area of 1751.1 m² g⁻¹ and the porous structure containing micropores and small mesopores which can provide a sufficient electrode–electrolyte interface, more paths for penetration and transportation of ions, and Li ion and charge accommodation; (ii) the elimination of impurities which could offer extra sites for Li storage; (iii) the N-doping which brings about strong electronegativity and interaction between Li and the carbon network and creates lots of defects leading to more active sites for Li⁺ storage. Furthermore, the presence of oxygen functionalities in PNCTs is a double-edged sword. On the one hand, the oxygen functionalities result in the surface redox reaction between the carbon–oxygen functional groups and Li⁺; i.e., –C=O + Li⁺ + e ↔ –C–O–Li, and provide extra ion storage sites.^52,53 On the other hand, the oxygen-containing groups will lower electronic conductivity of the PNCT electrode leading to the deterioration of the rate performance. To investigate the effects of oxygen-containing functional groups, the oxygen functionalities on the PNCTs were eliminated by re-calcinating the PNCTs under an Ar–H₂ gas mixture (5 : 1 in volume). The freshly obtained PNCTs possess a very low O-content of 0.82 wt%, detected by XPS. As shown in Fig. S5,† the stable reversible capacities of the freshly obtained PCNTs are 1123.1, 818.4, 625.2, 513.7, 476.3, 384.0 and 233.6 mA h g⁻¹ at current rates of 0.02, 0.05, 0.1, 0.3, 0.5, 1 and 3 A g⁻¹, respectively. When the current density increases from 0.02 A g⁻¹ to 3 A g⁻¹, the reversible capacity of the freshly obtained PNCTs decreases from 1123.1 mA h g⁻¹ to 233.6 mA h g⁻¹ with a capacity loss ratio of 79.2%, lower than that of the PNCTs (86.5%). The reduction of oxygen functionalities brings about the improvement in rate performance but results in the decrease in reversible capacity.

The cycling performance and corresponding coulombic efficiency of the PNCTs for 500 discharge/charge cycles at a current density of 1 A g⁻¹ are shown in Fig. 5d. The reversible capacity still remains at 512.4 mA h g⁻¹ after 500 cycles, with coulombic efficiency over 99% after 10 cycles, which represents that the PNCT anode possesses excellent cycling stability and good reversibility. Similarly, the NCT anode also exhibits satisfactory cycling performance (Fig. S6†). The impressive cycling capability can be ascribed to the tubular structure of the PNCTs and NCTs which endows the electrode material with good thermal stability and chemical stability on the one hand, and the stable SEI film formed on the surface of the PNCTs on the other hand.²¹

Conclusions

In summary, porous nitrogen-doped carbon tubes (PNCTs) were successfully prepared from crude protein-containing reed catkins by ZnCl₂ activation. The obtained PNCTs possess high N and O content of 4.22 wt% and 7.05 wt% respectively as well as a large specific surface area of 1751.1 m² g^−1, and exhibit excellent lithium-storage performance when evaluated as a LIB anode. The key to the successful preparation of PNCTs is the choice of the ZnCl₂ porogen, which is relatively moderate compared with KOH, and can keep the natural tubular structure of reed catkins intact during the activating process. These findings and this preparation method provide the possibility to utilize biomass waste to produce low cost but high performance electrode materials for LIBs.

Acknowledgements

The present work was supported by National Natural Science Foundation of China (No. 21506131 and 21477084) and Science and Technology Support Program of Sichuan Province (No. 2014GZ0095 and 2015RZ0057).

Notes and references

1. G. Q. Zhang, H. B. Wu, H. E. Hoster, M. B. Chan-Park and X. W. Lou, Energy Environ. Sci., 2012, 5, 9453–9456.
2. X. Jia, Z. Chen, A. Suwarnasarn, L. Rice, X. L. Wang, H. S. Sohn, Q. Zhang, B. M. Wu, F. Wei and Y. F. Lu, Energy Environ. Sci., 2012, 5, 6845–6849.
3. S. Boukhalfa, K. Evanoff and G. Yushin, Energy Environ. Sci., 2012, 5, 6872–6879.
4. J. E. Lee, S. H. Yu, D. J. Lee, D. C. Lee, S. I. Han, Y. E. Sung and T. Hyeon, Energy Environ. Sci., 2012, 5, 9528–9533.
5. X. J. Zhu, Y. W. Zhu, S. Murali, M. D. Stoller and R. S. Ruoff, ACS Nano, 2011, 5, 3333–3338.
6. S. H. Ng, J. Wang, D. Wexler, K. Konstantinov, Z. P. Guo and H. K. Liu, Angew. Chem., Int. Ed., 2006, 45, 6896–6899.
7. M. Suzuki, J. Suzuki, K. Sekine and T. Takamura, J. Power Sources, 2005, 146, 452–456.
8. K. N. Zhao, L. Zhang, R. Xia, Y. F. Dong, W. W. Xu, C. J. Niu, L. He, M. Y. Yan, L. B. Qu and L. Q. Mai, Small, 2016, 12, 588–594.
9. D. H. Youn, A. Heller and C. B. Mullins, Chem. Mater., 2016, 28, 1343–1347.
10. B. Scrosati, J. Hassoun and Y. K. Sun, Energy Environ. Sci., 2011, 4, 3287–3295.
11. L. W. Ji, Z. Lin, M. Alcoutlabi and X. W. Zhang, Energy Environ. Sci., 2011, 4, 2682–2699.
12. K. Jost, C. R. Perez, J. K. McDonough, V. Presser, M. Heon, G. Dion and Y. Gogotsi, Energy Environ. Sci., 2011, 4, 5060–5067.
13. C. Liu, F. Li, L. P. Ma and H. M. Cheng, Adv. Mater., 2010, 22, E28–E62.
14. Z.-L. Wang, D. Xu, H.-G. Wang, Z. Wu and X.-B. Zhang, ACS Nano, 2013, 7, 2422–2430.
15. Y. Zhou, Y. Zeng, D. Xu, P. Li, H.-G. Wang, X. Li, Y. Li and Y. Wang, Electrochim. Acta, 2015, 184, 24–31.
16. H.-G. Wang, Y. Wang, Y. Li, Y. Wan and Q. Duan, Carbon, 2015, 82, 116–123.
17. K. Kang, Y. S. Meng, J. Bréger, C. P. Grey and G. Ceder, Science, 2006, 311, 977–980.
18. M. Armand and J. M. Tarascon, Nature, 2008, 451, 652–657.
19. Y. S. Hu, P. Adelhelm, B. M. Smarsly, S. Hore, M. Antonietti and J. Maier, Adv. Funct. Mater., 2007, 17, 1873–1878.
20. L. Qie, W. M. Chen, Z. H. Wang, Q. G. Shao, X. Li, L. X. Yuan, X. L. Hu, W. X. Zhang and Y. H. Huang, Adv. Mater., 2012, 24, 2047–2050.
21. J. Hou, C. Cao, F. Idrees and X. Ma, ACS Nano, 2015, 9, 2556–2564.
22. G. T. K. Fey, D. C. Lee, Y. Y. Lin and T. P. Kumar, Synth. Met., 2003, 139, 71–80.
23. L. Chen, Y. Z. Zhang, C. H. Lin, W. Yang, Y. Meng, Y. Guo, M. L. Li and D. Xiao, J. Mater. Chem. A, 2014, 2, 9684–9690.
24. L. P. Wang, Z. Schnepp and M. M. Titirici, J. Mater. Chem. A, 2013, 1, 5269–5273.
25. Z. Li, Z. Xu, X. Tan, H. Wang, C. M. B. Holt, T. Stephenson, B. C. Olsen and D. Mitlin, Energy Environ. Sci., 2013, 6, 871–878.
26. J. K. Ou, Y. Z. Zhang, L. Chen, Q. Zhao, Y. Meng, Y. Guo and D. Xiao, J. Mater. Chem. A, 2015, 3, 6534–6541.
27. J. C. Arrebola, A. Caballero, L. Hernán, J. Morales, M. Olivares-Marín and V. Gómez-Serrano, J. Electrochem. Soc., 2010, 157, A791–A797.
28. J. C. Wang and S. Kaskel, J. Mater. Chem., 2012, 22, 23710–23725.
29. L. W. Ji and X. W. Zhang, Electrochem. Commun., 2009, 11, 684–687.
30. P. Wang, H. L. He, X. L. Xu and Y. D. Jin, ACS Appl. Mater. Interfaces, 2014, 6, 1563–1568.
31. Y. H. Zhu, X. X. Xiang, E. H. Liu, Y. H. Wu, H. Xie, Z. L. Wu and Y. Y. Tian, Ionics, 2013, 19, 409–414.
32. F. Rodríguez-Reinoso and M. Molina-Sabio, Carbon, 1992, 30, 1111–1118.
33. S. Balci, T. Doǧu and H. Yücel, J. Chem. Technol. Biotechnol., 1994, 60, 419–426.
34. D. Lozano-Castelló, J. M. Calo, D. Cazorla-Amorós and A. Linares-Solano, Carbon, 2007, 45, 2529–2536.
35. E. Raymundo-Piñero, P. Azaïs, T. Cacciaguerra, D. Cazorla-Amorós, A. Linares-Solano and F. Béguin, Carbon, 2005, 43, 786–795.
36. Z. S. Wu, W. Ren, L. Xu, F. Li and H. M. Cheng, ACS Nano, 2011, 5, 5463–5471.
37. D. Hulicovajurcakova, A. M. Puziy, O. I. Poddubnaya, F. Suárezgarcía, J. M. D. Tascón and Q. L. Gao, J. Am. Chem. Soc., 2009, 131, 5026–5027.
38. H.-K. Jeong, M. Jin, E. J. Ra, K. Y. Sheem, G. H. Han, S. Arepalli and Y. H. Lee, ACS Nano, 2010, 4, 1162–1166.
39. D. W. Wang, F. Li, M. Liu, G. Q. Lu and H. M. Cheng, Angew. Chem., 2008, 120, 379–382.
40. H. Liu, M. Q. Jia, N. Sun, B. Cao, R. J. Chen, Q. Z. Zhu, F. Wu, N. Qiao and B. Xu, ACS Appl. Mater. Interfaces, 2015, 7, 27124–27130.
41. S. L. Zhang, F. Yao, L. Yang, F. Z. Zhang and S. L. Xu, Carbon, 2015, 93, 143–150.
42. Y. Liu, J. S. Xue, T. Zheng and J. R. Dahn, Carbon, 1996, 34, 193–200.
43. J. Ding, H. L. Wang, Z. Li, K. Cui, D. Karpuzov, X. H. Tan, A. Kohandehghan and D. Mitlin, Energy Environ. Sci., 2015, 8, 941–955.
44. D. Y. Pan, S. Wang, B. Zhao, M. H. Wu, H. J. Zhang, Y. Wang and Z. Jiao, Chem. Mater., 2009, 21, 3136–3142.
45. A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth and A. K. Geim, Phys. Rev. Lett., 2006, 97, 187401–187404.
46. F. Tuinstra and J. L. Koenig, J. Chem. Phys., 1970, 53, 1126–1130.
47. J. R. Dahn, T. Zheng, Y. Liu and J. S. Xue, Science, 1995, 270, 590–593.
48. C. C. Ma, X. H. Shao and D. P. Cao, J. Mater. Chem., 2012, 22, 8911–8915.
49. U. Zielke, K. J. Hüttinger and W. P. Hoffman, Carbon, 1996, 34, 983–998.
50. W. J. Si, J. Zhou, S. M. Zhang, S. J. Li, W. Xing and S. P. Zhuo, Electrochim. Acta, 2013, 107, 397–405.
51. J. K. Ou, Y. Z. Zhang, L. Chen, H. Y. Yuan and D. Xiao, RSC Adv., 2014, 4, 63784–63791.
52. J. L. Figueiredo, M. F. R. Pereira, M. M. A. Freitas and J. J. M. Órfão, Carbon, 1999, 37, 1379–1389.
53. K. L. Hong, L. Qie, R. Zeng, Z. Q. Yi, W. Zhang, D. Wang, W. Yin, C. Wu, Q. J. Fan, W. X. Zhang and Y. H. Huang, J. Mater. Chem. A, 2014, 2, 12733–12738.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra21620j

‡ These authors contributed equally.

---