In situ grown Nb₄N₅ nanocrystal on nitrogen-doped graphene as a novel anode for lithium ion battery

Abstract

The metal-rich niobium nitride of Nb₄N₅ has higher conductivity than Nb₃N₅ and a higher theoretical specific capacity than NbN. To rationally design a metal-rich anode material, Nb₄N₅ nanocrystals coated by nitrogen-doped graphene (N-G) have been successfully synthesized by a facile in situ ice bathing method with subsequent annealing in NH₃. The use of these as an anode material is reported for the first time. The discharge capacity is 487 mA h g⁻¹ at the current density of 0.1 A g⁻¹ (0.0819 mA cm⁻²) after 200 cycles and the high rate discharge capacity is 125 mA h g⁻¹ at a current density of 5 A g⁻¹ (4.0926 mA cm⁻²). Specially, the discharge capacity is still enhanced after 200 cycles at 0.1 A g⁻¹ (0.0819 mA cm⁻²). The Nb₄N₅/N-G hybrid could be a promising anode material for LIBs with a high rate performance and long cycle life.

---

Introduction

Lithium ion batteries (LIBs) have attracted considerable attention in the field of energy storage and conversion. The performance of LIBs is mainly determined by the properties of active electrode materials which are controlled by composition, structure, morphology and so on.¹ Commercial LIBs usually use graphite as an anode. However, the theoretical capacity of graphite is severely limited to 372 mA h g⁻¹, which limits its application for electrical/hybrid vehicles.^2–4 Generally, many metal oxides have been explored as high capacity anode materials, however, they suffer from low conductivity. Compared to metal oxides, the metal nitrides often show improved conductivity. Therefore, the nitridation of metal oxides is a suitable method to acquire high performance anode materials.

Metal nitrides are emerging as promising electrode materials for high-performance LIBs^5,6 due to their high Li⁺ diffusion, excellent electrical conductivity⁷ and flat potential close to that of lithium metal.⁸ Niobium nitrides are mainly composed of Nb₃N₅, Nb₄N₃, NbN and some other nonstoichiometric niobium nitride compounds. Among them, the metal-rich phases (such as Nb₄N₅ and Nb₅N₆) have also been presented in δ-NbN system, which comprise lots of Nb⁵⁺ ions.^9,10 They are promising candidates for LIBs because of their high-span valency transformation.¹⁰

Herein, we report the synthesis of Nb₄N₅/nitrogen-doped graphene hybrid nanomaterial (Nb₄N₅/N-G) by in situ growth in ice bathing followed by NH₃ annealing. The as-prepared Nb₄N₅/N-doped graphene hybrid is reported for the first time using it as a LIBs anode material, which exhibits excellent electrochemical performance. Besides superior conductivity of Nb₄N₅, the internal defects between graphene layers motivated by the substitution of nitrogen atoms may enhance lithium storage.¹¹ Meanwhile, nitrogen-doped graphene can remit the expansion of Nb₄N₅ availably during discharge process and help forming stable solid electrolyte interface (SEI) layer. Therefore, the Nb₄N₅/nitrogen-doped graphene hybrid can be a novel promising anode for LIBs.

Experimental section

Material

Niobium pentachloride was purchased from Aladdin Chemical Reagent. Anhydrous ethanol (>99.7%) was purchased from Beijing Chemical Works. Ammonium hydroxide was purchased from Beijing Tong Guang Fine Chemicals Company. Graphene oxide was prepared according to the previous reported literature by Hummers¹² from graphite powder (purchased from Aldrich, powder, <20 micron). All used water in the process was deionized.

Material synthesis

Typically, 0.5489 g NbCl₅ was dissolved into 10 mL ethanol to form solution A. 90 g ice was added into 30 mL 3 mg mL⁻¹ GO solution (solution B) by vigorously magnetic stirring. The solution A was added into solution B dropwise. Then, the pH of mixed solution was adjusted to ca. 7 by ammonium hydroxide. After centrifuging, the resulting Nb₂O₅/GO precipitate was washed by water for three times and freeze-dried for 12 h. Finally, the Nb₂O₅/GO was converted to Nb₄N₅/N-G by annealing in NH₃ at 700 °C under the flow rate of 350 mL min⁻¹ for 4 h.

For comparison, the fabrication of Nb₄N₅ was similar to the above.

Physical characterization

The microstructures were studied by scanning electron microscope (SEM) using a ZEISS Merlin Compact Field-emission scanning electron microscope (ZEISS, Germany) and transmission electron microscopy (TEM) used a JEM-2100 electron microscope (JEOL Ltd, Japan) working at 200 kV. The surface chemical composition was studied by X-ray Photoelectron Spectroscopy using Axis Ultra Imaging Photoelectron Spectrometer (Kratos Analytical Ltd, Japan). X-ray diffraction (XRD) was performed on a D8 Focus diffractometer with monochromatized Cu Kα radiation (λ = 1.5418 Å). Raman spectrum was measured using a Renishawi instrument via Raman Microscope with laser excitation at 532 nm. Nitrogen adsorption isotherms were measured using a Micromeritics ASAP 2020 analyzer at 196 °C. The weight of the sample used in the measurement was 100–200 mg. The Brunauer–Emmett–Teller (BET) method was utilized to calculate the specific surface area.

Electrochemical tests

To prepare the working electrodes, the Nb₄N₅/N-G, Nb₄N₅ and N-G were mixed with carbon black (Super P) and polyvinylidene fluoride binder with a ratio of 8 : 1 : 1 in N-methyl-2-pyrrolidone (NMP) to form slurry, respectively. The three slurries were coated onto copper foil. The coated foil was cut into disk electrodes of 14 mm in diameter. Then the electrodes were vacuum-dried overnight at 120 °C. Lithium foil (China Energy Lithium Co., Ltd) was used as the counter and reference electrode. A glass fiber from Whatman was used as separator. The electrolyte was 1 M LiPF₆ in a 50 : 50 (w/w) mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC). Coin Cell (LIR-2016 type) assembly was carried out in a recirculating an Argon glovebox where both the moisture and oxygen contents were below 1 ppm. All the cells were cycled between 0.01 and 3 V (vs. Li/Li⁺) at certain current densities on a LAND CT2001A battery test system. The rate capability was evaluated by varying the current density from 0.1 A g⁻¹ to 5 A g⁻¹. Specific capacities were calculated based on the mass of the each activated materials. Cyclic voltammetry (CV) was obtained on a CHI1000C electrochemistry workstation at the scan rate of 0.3 mV s⁻¹ and the potential from 0.01 to 3 V (vs. Li⁺/Li). The electrochemical impedance spectrum was obtained on the CHI600E with amplitude of 10 mV in the frequency range from 100 kHz to 0.01 Hz.

Results and discussion

Characterization

XRD patterns of Nb₄N₅, N-G and Nb₄N₅/N-G are shown in Fig. 1. The as-synthesized pure N-G displays a broad diffraction peak of (002) at 27.5° which is assigned to wrinkle layer-to-layer distance.¹³ The diffraction peaks of pure Nb₄N₅ at 36.1°, 41.7°, 60.9°, 73.2°, 76.4° are indexed to (211), (310), (312), (213) and (422) crystal lattices, respectively (JCPDS No. 51-1327, I4/m). The Nb₄N₅/N-G diffraction peak curve is higher than pure Nb₄N₅ from 25° to 35° because of the peak overlay of N-G and Nb₄N₅. Then, the Nb₄N₅/N-G hybrid is confirmed by Raman spectroscopy. As shown in Fig. 2b, two broad peaks emerge at 1340 and 1590 cm⁻¹ which are attributed to D and G bands of nitrogen-doped graphene, respectively. The D band strongly associates with the disorder degree of graphene.^14,15 The G band corresponds to the zone center E_2g mode related to phonon vibrations in sp² carbon materials.^16,17 Clearly, the I_D/I_G ratio (1.1 : 1) indicates the presence of a large amount of defects in the graphene layer due to the doping of nitrogen atoms. The Raman spectrum reveals that nitrogen-doped graphene with crystalline structures and many defects are obtained, which are favorable for the electrochemical properties.¹⁸

Fig. 1 (a) XRD patterns of Nb₄N₅, N-G and Nb₄N₅/N-G hybrid; (b) Raman spectrum of Nb₄N₅/N-G hybrid.

Fig. 2 (a) SEM image of Nb₄N₅/N-G hybrid; (b) low-magnification TEM image of Nb₄N₅/N-G hybrid; (c) high-magnification TEM image of Nb₄N₅/N-G hybrid. (Note: △ and ★ stand for N-G and Nb₄N₅, respectively.)

The morphology of Nb₄N₅/N-G hybrid is characterized by SEM and TEM, as shown in Fig. 2. The typical wrinkle structure of N-G is clearly observed in Fig. 2a, which results from thermodynamically spontaneous bending. Meanwhile, the nanocrystals of Nb₄N₅ are tightly captured by the N-G. Fig. 2b further confirms that the nanocrystals of Nb₄N₅ are uniformly dispersed on N-G without severe aggregation because of the protection of N-G. The SAED pattern (Fig. 3c) can further confirm the high crystallinity of Nb₄N₅/N-G. These diffraction rings can be indexed to (312) of Nb₄N₅, (002) of N-G and (220) of Nb₄N₅ which are well in accordance with XRD patterns (Fig. 1a).

Fig. 3 (a) XPS survey. High-resolution Nb₄N₅/N-G XPS spectrum of (b) Nb 3d, (c) C 1s and (d) N 1s. The black scatter is the raw date and the red curve is the sum of different fitting peaks.

X-ray photoelectron spectroscopy (XPS) measurement is performed to study the surface chemical composition and oxidation state. As shown in Fig. 3a, the XPS survey spectrum shows that there are four elements: C, N, Nb and O without other impurity. Nb 3d spectrum involves two distinct doublets and a weak shoulder peak (Fig. 3a). The highest Nb 3d_5/2 binding energy at 207.0 eV can be indexed to the Nb⁵⁺–O due to the oxidation of surface. The lowest binding energy emerges at 203.8 eV which belongs to the Nb 3d_5/2 in Nb₄N₅ and the peak located at 205.0 eV can be indexed to Nb⁵⁺–N.^19–21 The peaks at 206.3, 208.0 and 209.9 eV correspond to Nb 3d_3/2. The C 1s spectrum is displayed in Fig. 3b, which mainly exhibits three conspicuous peaks. The obvious core level peak binding energy at 284.9 eV represents graphite-like sp² C–sp² C, indicating that the most parts of carbon atoms exist in conjugated lattice.¹⁸ The relatively weak peaks binding energy at 285.7 and 286.6 eV can be ascribed to N–sp² C and N–sp³ C, respectively, because of N atoms and the defects.²² The results of N–sp² C and N–sp³ C correspond with the Raman analysis. As shown in Fig. 3d, the broad N 1s XPS peak can be fitted to the five peaks of Nb–N (396.9 eV), pyridinic N (398.5 eV), pyrrolic N (399.8 eV), graphitic N (401.5 eV) and oxygenated N (402.8 eV).

To investigate the content of N-G in Nb₄N₅/N-G, the TG-DSC measurement was carried out. As shown in Fig. 4, the weight increase from ca. 150 to ca. 400 °C because the 1 molecular Nb₄N₅ is oxidated to be 2 molecular Nb₂O₅. After then, the weight decreases from ca. 400 to ca. 800 °C due to the combustion of N-G and the oxidation of Nb₄N₅. The content of N-G is calculated in detail (see in ESI†). According to the analysis and calculation, the content of N-G in Nb₄N₅/N-G is about 16.78%.

Fig. 4 The TG/DSC curves of Nb₄N₅/N-G.

Three kinds of materials are examined by N₂ adsorption–desorption measurements (Fig. S1†). The BET surface area of bare Nb₄N₅ is 44.53 m² g⁻¹ and the N-G is 36.86 m² g⁻¹. After the Nb₄N₅ is anchored on N-G, the dispersity of Nb₄N₅ loading on N-G is enhanced, so that the BET surface area increases into 49.64 m² g⁻¹.

Electrochemical performance

In Nb₄N₅/N-G, Li ion can intercalate first and later on under conversion reaction. During the process, a large number of Li⁺ can be stored at a low potential for lithium ion battery anode material. The reaction of Nb₄N₅ and Li can be speculated as follow:²³

Nb₄N₅ + xLi → Li_xNb₄N₅ (intercalation reaction)

Li_xNb₄N₅ + (15 − x)Li → 4Nb + 5Li₃N (conversion reaction)

The cycle voltammogram (Fig. 5a) presents a stable electrochemical performance of Nb₄N₅/N-G. During the first cathodic process, two apparent cathodic peaks are observed at 1.41 V and 0.50 V (vs. Li/Li⁺). The origin of cathodic peak at 1.41 V can be attributed to the Li ion intercalation into the Nb₄N₅/N-G. The peak at 0.50 V is due to the solid electrolyte interface (SEI) formation. Furthermore, the minor slope changing at 0.15 V can be assigned to a conversion reaction from Li_xNb₄N₅ to Nb and Li₃N. After the 5th cycle, the peak at 1.41 V shifts to a higher potential (1.58 V) due to the difficulty of Li intercalation into residual Nb₄N₅/N-G. Similarly, the peak at 0.15 V shifts to a prominent peak at 0.17 V because of the electrode polarization. During the anodic sweep, the oxidation peak can be attributed to the oxidation reaction, from Nb⁰ to Li_xNb₄N₅ and Li_xNb₄N₅ to Nb₄N₅.

Fig. 5 (a) Cyclic voltammogram of Nb₄N₅/N-G hybrid with a scanning rate of 0.3 mV s⁻¹ ranging from 0.01 to 3 V (vs. Li⁺/Li); (b) Nyquist plots of Nb₄N₅/N-G and Nb₄N₅ at opening circuit voltage after 3 cycles.

To further clarify the difference of Nb₄N₅/N-G and Nb₄N₅ in electrochemical performance, electrochemical impedance spectroscopy (EIS) is carried out at frequencies from 100 kHz to 0.01 Hz to identify the correlation between the electrochemical performance and electrode kinetics. Fig. 5b shows the Nyquist plots for the Nb₄N₅/N-G and Nb₄N₅ after three cycles. The Nyquist plots for Nb₄N₅/N-G and Nb₄N₅ share the similar feature of a high-frequency depressed semicircle and a medium-frequency depressed semicircle followed by a linear tail in the low-frequency region. The intercept on the Z′ axis at the high frequency end is the equivalent series resistance (R_s). The Nb₄N₅ and Nb₄N₅/N-G have a similar intercept on the Z′ axis. The size of the semicircular that includes the medium-frequency response represents charge-transfer resistance (R_ct). Obviously, the diameter of the semicircle for Nb₄N₅/N-G electrode is markedly lower than Nb₄N₅ in the high-frequency region. This phenomenon indicates that Nb₄N₅/N-G possesses lowest contact and charge-transfer impedances so that electron can transport rapidly during the electrochemical lithium extraction and insertion.²⁴ These results confirm that the disorderly wrinkle layer-to-layer N-G (likes huge charge-transfer network) can provide more active sites and greater contact area with Nb₄N₅. The inclined linear line in the low-frequency region stands for the Warburg impedance (Z_w) which relates to lithium diffusion in the solid electrode.²⁵ All the advantages of Nb₄N₅/N-G above result in enormous improvement of the electrochemical performance (specific capacity, rate performance, etc.) compared to bare Nb₄N₅.

Fig. 6a shows galvanostatic charge profiles of Nb₄N₅/N-G electrode within the voltage of 0.01–3 V (vs. Li/Li⁺) at a current density of 0.1 A g⁻¹ (0.0819 mA cm⁻²). The profiles are in good agreement with the CV results. Two conspicuous plateaus (1.5 and 2.2 V vs. Li/Li⁺) are obtained during the discharge while prominent plateaus at around 2.1 V is observed in charge process. The galvanostatic charge/discharge curves of Nb₄N₅/N-G, Nb₄N₅ and N-G at a current of 0.1 A g⁻¹ over a potential range of 0.01–3.0 V (vs. Li/Li⁺) are shown in Fig. 6b and c and S2,† respectively. At the first cycle, the charge/discharge capacities of Nb₄N₅, N-G and Nb₄N₅/N-G are 432.4/899.2, 461.4/771.3 and 663.3/941.8 mA h g⁻¹, respectively. Apparently, the coulombic efficiency gets prodigious increasement from 48.09%, 59.8% to 70.43% which reflects synergistic effect between Nb₄N₅ and N-G. The enhanced initial coulombic efficiency may be ascribed to the catalytic sites on the surface of N-G, related to the decomposition of electrolyte occupied by Nb₄N₅.²⁶ This phenomenon effectively helps the formation of solid electrolyte interface (SEI) and the remission of severe volume expansion, leading a decreased irreversible capacity. Fig. 5a shows that the reversible capacity of Nb₄N₅/N-G retains at 487.3 mA h g⁻¹ after 200 cycles, with no capacity loss from the 20th cycle. Meanwhile, the tendency of charge–discharge curve still maintains enhanced step by step. The metal nitride reveals ordered and disordered phases with significant levels of lithium vacancies.²⁷ Li vacancies are the charge carriers in Li₃N, which is a well-known fast-ion conductor with the highest Li⁺ conductivity.²⁷ The formation of additional vacancies in the nitride phases implies their potential for enhanced lithium ion diffusion. The reason why the Nb₄N₅/N-G has remarkable performance can be interpreted by following: (1) suitable kinetic reaction due to the fast diffusion of Li⁺; (2) tremendous conductive network of N-G.

Fig. 6 (a) Galvanostatic discharge–charge profiles of Nb₄N₅/N-G within the voltage window of 3.00–0.01 V at a current density of 0.1 A g⁻¹ (0.0819 mA cm⁻²); (b) cycle performance and coulombic efficiency of Nb₄N₅/N-G at 0.1 A g⁻¹ (0.0819 mA cm⁻²); (c) cyclic performance and coulombic efficiency of Nb₄N₅ at 0.1 A g⁻¹ (0.0955 mA cm⁻²).

The rate performance of Nb₄N₅/N-G with different current densities is given in Fig. 7a. The incremental rates discharge capacity are 481, 430, 365, 265, 125 mA h g⁻¹ at 0.2 A g⁻¹ (0.1637 mA cm⁻²), 0.5 A g⁻¹ (0.4093 mA cm⁻²), 1 A g⁻¹ (0.8185 mA cm⁻²), 2 A g⁻¹ (1.6370 mA cm⁻²), 5 A g⁻¹ (4.0926 mA cm⁻²), respectively. After high rate charge/discharge, the specific capacity of Nb₄N₅/N-G at the current density of 0.1 A g⁻¹ (0.0819 mA cm⁻²) can recover to the initial value, which confirms the structure stability. The capacity of Nb₄N₅/N-G hybrid at even 5 A g⁻¹ (4.0926 mA cm⁻²) is as high as 125 mA h g⁻¹ while that of Nb₄N₅ (Fig. 7b) is only 70 mA h g⁻¹ at the current density of 1 A g⁻¹ (0.9549 mA cm⁻²). The N-G provides enormous conductive network for Nb₄N₅, which is favorable for charge transport. The Nb₄N₅/N-G hybrid has more active sites because of the metal-rich material and N-G. After discharge, the generated Nb atoms take lower area leading fast Li⁺ diffusion. The rate performance of Nb₄N₅/N-G hybrid anode further confirms the possible application for LIBs.

Fig. 7 (a) Rate performance of Nb₄N₅/N-G at varied current density; (b) rate performance of bare Nb₄N₅ at varied current density.

Conclusions

In conclusion, Nb₄N₅/N-G hybrid is successfully synthesized by a facile in situ ice bathing method with subsequent annealing in NH₃. The conductive Nb₄N₅ nanocrystals are uniformly dispersed on the N-G. The Nb₄N₅ nanocrystals have mainly no band gap which make contributions to lower charge-transfer resistance and electrode polarization so that the cells have the excellent electrochemical performance.²⁸ The specific capacity is 487 mA h g⁻¹ at 0.1 A g⁻¹ (0.0819 mA cm⁻²) which still demonstrates upgrade tendency. The hybrid also exhibits good rate capability compared to the bare Nb₄N₅ with capacity of 125 mA h g⁻¹ at the current density of 5 A g⁻¹ (4.0926 mA cm⁻²). Above all, Nb₄N₅/N-G hybrid can be a novel promising anode material for LIBs with high energy density and long cycle life.

Acknowledgements

This work was financially supported from National Natural Science Foundation of China (Grant no. 61376056, 51125006, 91122034, 51121064, 51222212), National Program of China (Grant No. 2016YFB0901600), and Science and Technology Commission of Shanghai (Grant no. 13JC1405700, 14520722000).

Notes and references

1. X. Xia, D. Chao, Y. Zhang, Z. Shen and H. Fan, Nano Today, 2014, 9, 785–807.
2. G. Cui, L. Gu, L. Zhi, N. Kaskhedikar, P. A. van Aken, K. Müllen and J. Maier, Adv. Mater., 2008, 20, 3079–3083.
3. M. Yoshio, H. Wang, K. Fukuda, T. Umeno, T. Abe and Z. Ogumi, J. Mater. Chem., 2004, 14, 1754–1758.
4. H. Li, Z. Wang, L. Chen and X. Huang, Adv. Mater., 2009, 21, 4593.
5. G. Cui, L. Gu, A. Thomas, L. Fu, P. A. van Aken, M. Antonietti and J. Maier, ChemPhysChem, 2010, 11, 3219–3223.
6. N. Suzuki, R. B. Cervera, T. Ohnishi and K. Takada, J. Power Sources, 2013, 231, 186–189.
7. X. Lu, G. Wang, T. Zhai, M. Yu, S. Xie, Y. Ling, C. Liang, Y. Tong and Y. Li, Nano Lett., 2012, 12, 5376–5381.
8. B. Das, M. Reddy, P. Malar, T. Osipowicz, G. S. Rao and B. Chowdari, Solid State Ionics, 2009, 180, 1061–1068.
9. M. Wen, C. Hu, Q. Meng, Z. Zhao, T. An, Y. Su, W. Yu and W. Zheng, J. Phys. D: Appl. Phys., 2009, 42, 035304.
10. H. Cui, G. Zhu, X. Liu, F. Liu, Y. Xie, C. Yang, T. Lin, H. Gu and F. Huang, Adv. Sci., 2015, 2(12), 1500126.
11. L. Tang, Y. Wang, Y. Li, H. Feng, J. Lu and J. Li, Adv. Funct. Mater., 2009, 19, 2782–2789.
12. W. S. Hummers Jr and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
13. D. Geng, S. Yang, Y. Zhang, J. Yang, J. Liu, R. Li, T.-K. Sham, X. Sun, S. Ye and S. Knights, Appl. Surf. Sci., 2011, 257, 9193–9198.
14. Y. T. Lee, N. S. Kim, S. Y. Bae, J. Park, S.-C. Yu, H. Ryu and H. J. Lee, J. Phys. Chem. B, 2003, 107, 12958–12963.
15. K. Suenaga, M. Yudasaka, C. Colliex and S. Iijima, Chem. Phys. Lett., 2000, 316, 365–372.
16. A. Ferrari, J. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. Novoselov and S. Roth, Phys. Rev. Lett., 2006, 97, 187401.
17. A. Malesevic, R. Vitchev, K. Schouteden, A. Volodin, L. Zhang, G. Van Tendeloo, A. Vanhulsel and C. Van Haesendonck, Nanotechnology, 2008, 19, 305604.
18. H. Wang, C. Zhang, Z. Liu, L. Wang, P. Han, H. Xu, K. Zhang, S. Dong, J. Yao and G. Cui, J. Mater. Chem., 2011, 21, 5430.
19. Y. Ufuktepe, A. Farha, S. Kimura, T. Hajiri, K. Imura, M. Mamun, F. Karadag, A. Elmustafa and H. Elsayed-Ali, Thin Solid Films, 2013, 545, 601–607.
20. Y. Ufuktepe, A. H. Farha, S.-i. Kimura, T. Hajiri, F. Karadağ, M. A. Al Mamun, A. A. Elmustafa, G. Myneni and H. E. Elsayed-Ali, Mater. Chem. Phys., 2013, 141, 393–400.
21. J. Alfonso, J. Buitrago, J. Torres, J. Marco and B. Santos, J. Mater. Sci., 2010, 45, 5528–5533.
22. A. L. M. Reddy, A. Srivastava, S. R. Gowda, H. Gullapalli, M. Dubey and P. M. Ajayan, ACS Nano, 2010, 4, 6337–6342.
23. D. K. Nandi, U. K. Sen, D. Choudhury, S. Mitra and S. K. Sarkar, ACS Appl. Mater. Interfaces, 2014, 6, 6606–6615.
24. C. He, S. Wu, N. Zhao, C. Shi, E. Liu and J. Li, ACS Nano, 2013, 7, 4459–4469.
25. J. Z. Wang, C. Zhong, D. Wexler, N. H. Idris, Z. X. Wang, L. Q. Chen and H. K. Liu, Chem.–Eur. J., 2011, 17, 661–667.
26. I. s. Kim, P. Kumta and G. Blomgren, Electrochem. Solid-State Lett., 2000, 3, 493–496.
27. Z. Stoeva, R. Gomez, A. G. Gordon, M. Allan, D. H. Gregory, G. B. Hix and J. J. Titman, J. Am. Chem. Soc., 2004, 126, 4066–4067.
28. C. Chen, Y. Huang, H. Zhang, X. Wang, G. Li, Y. Wang, L. Jiao and H. Yuan, J. Power Sources, 2015, 278, 693–702.

---

Footnote

† Electronic supplementary information (ESI) available: Fabrication of graphene oxide (GO), TG-DSC analysis and calculation and coulombic efficiency of N-G at 0.1 A g⁻¹ (0.0832 mA cm⁻²). See DOI: 10.1039/c6ra13647h

---