Integrated reduced graphene oxide multilayer/Li composite anode for rechargeable lithium metal batteries

Abstract

Suppressing the growth of dendritic lithium is one of the most critical challenges for the development of Li metal batteries. Herein we report an integrated reduced graphene oxide (rGO) multilayer/Li composite electrode, in which filtration-synthesized free-standing rGO film acts as a conductive support for the strong anchoring of Li metal. When tested as an electrode for rechargeable Li metal batteries, the rGO/Li composite exhibits noticeable enhancement of electrochemical performance with better cycling stability than the unmodified Li. The interconnected rGO layers not only help to suppress the formation of dendritic Li, but also store the dead Li and restrain the uneven surface potential. The proposed electrode design protocol would provide a better insight into the preparation of other high-performance lithium-based batteries.

---

1. Introduction

The earliest commercial products containing rechargeable Li batteries appeared in the 1970s, employing Li metal as the anode. However, metallic Li electrolytes were quickly discarded due to a potential safety hazard. In the 1990s, Li-ion batteries were introduced by Sony Corporation to address the dendrite issues by hosting Li in a graphitic material, but the enhanced battery safety of Li-ion batteries is at a significant cost in energy density. Recently, the high energy density Li metal anode was investigated world-wide again because of its irreplaceable position in the “next generation” rechargeable batteries, such as Li–air and Li–S batteries.^1–7 So to speak, the high energy density Li anode is a key to breaking the bottleneck of the lithium electricity industry. Unfortunately, large-scale application of rechargeable Li metal anodes still faces two major barriers.^8–11 One problem is that metallic Li is too active and it will easily react with electrolyte, resulting in low coulombic efficiency and poor cycle life due to the rapid loss of Li and electrolyte. The other is that the formation of dendritic Li leads to virtually infinite volumetric change and causes short circuit of the batteries. In view of the above issues, extensive researches have demonstrated that the solid electrolyte interphase (SEI) film between the lithium surface and the electrolyte can act as a sufficient passivation layer to suppress the dendrite growth.^12–15 However, the SEI film cannot completely eliminate the formation of dendritic Li because of its weak mechanical strength. In the past decades, tremendous efforts have been devoted to solving these problems, mainly involved two categories. The first strategy is to improve the stability and uniformity of the SEI film by adding additives into the electrolyte.^16–22 Although enhanced results are proven, the mechanical strength of the modified SEI film is still not satisfactory. As a result, upon Li deposition, the surface of Li anode usually cracks arising from volumetric expansion and cause exposure of fresh Li metal to the electrolyte for further reactions.²³ The second approach is to make a pre-formed protective film/layer with high Li ion conductivity, such as carbon films,^24–26 and lithium nitrogenous compounds.^11,27 The protective layer can not only prevent Li metal anodes from contacting with the electrolyte, but also improve the uniformity of SEI film and suppress the growth of dendritic Li. Recently, Zheng et al. adopted a monolayer of interconnected amorphous hollow carbon nanospheres as a protective layer for the Li metal anode and an improved electrochemical performance was demonstrated. It is reported that the aforementioned carbon spheres can isolate the Li metal depositions and facilitate the formation of a stable SEI film.²³ Meanwhile, our previous research demonstrated that chemically stable and mechanically flexible pre-formed interfacial layers are excellent protective coating for Li metal anode.^26,27

As a two-dimensional crystal of sp² conjugated carbon atoms, graphene sheet possesses high Li ion conductivity and excellent mechanical strength (tensile strength 130 GPa),²⁸ which makes it a promising candidate as the protective layer for Li metal anode. In all kinds of graphene structure, flexible, free-standing and paper-like graphene films have aroused great interest because of their special layered structure and good electrical conductivity.^29–31 Up to now, large area flexible graphene films can be easily obtained by simple solution processes such as filtration,^32–35 spray coating,³⁶ spin casting,³⁷ Langmuir–Blodgett assembly,³⁸ and self-assembly at the liquid/air interface.^28,39 Among them, filtration-synthesized graphene film is considered to be one of the most promising graphene paper film due to its easy processing, large surface area and low cost. Currently, there is no report about fabrication of filtration-synthesized graphene/Li composite electrodes and their application for Li-based batteries. In this present work, we report a facile approach for the preparation of integrated rGO multilayer/Li composite electrode. The rGO film prepared by filtration serves as a protective coating on Li metal anodes. The electrochemical performances of the obtained rGO/Li composite are thoroughly characterized as the anode of Li batteries. In addition, the suppressive effect of dendritic lithium is investigated in detail.

2. Experimental

2.1 Material fabrication

Graphene oxide (GO) was prepared by a modified Hummers' method, which was similar to our previous report.⁴⁰ To prepare the flexible graphene films, 20 mg GO powder was dispersed with 100 mL deionized water, and then the mixture was treated by ultrasonication. Afterwards, the homogeneous brown dispersion was filtered through a cellulose acetate membrane to produce flexible GO film. The GO film was exfoliated from the membrane through wetting in alcohol. The lithium foils were pasted on copper sheets with double-side adhesive. Then the rGO film was pressed onto the Li/Cu composite foil under the pressure of 5 MPa. The rGO/Li electrodes were fabricated by punching the graphene modified Li/Cu composite foil into 15 mm wafer. The area with double-side adhesive must be avoided when punching the electrode wafer. The Cu sheet was used as the current collector during the following electrochemical cycling. During the filtration process, the volume of mixture solution has been controlled in 8 mL, 10 mL, 12 mL and 14 mL. When the volume of mixture solution is 8 mL, the film is too thin to peel off from the cellulose acetate membrane. When the volume of mixture solution increases, the EIS performance of Li/rGO electrode declined obviously, as shown in Fig. S1.† The result turns out that the thickness of rGO film should be as thinner as possible, so we choose 10 mL solution to filter the GO film. The flexible GO film was then heated in an oven at 90 °C for 12 h under a hydrazine hydrate atmosphere, to obtain the corresponding rGO film.

2.2 Characterization

The surface and cross-section morphologies were characterized by using a scanning electron microscope (SEM, Hitachi S-4800 equipped with GENENIS 4000 EDAX detector). The structure and composition were characterized by X-ray diffraction (XRD, RigakuD/Max-3B) and Raman spectroscopy (JobinYvon Labor Raman HR-800). The bonded structures of pure Li and r-GO/Li electrodes after cycling were examined by an X-ray photoelectron spectroscopy (XPS) using an ESCAL 220i-XL electron spectrometer, operating with a monochromated Al-K X-ray radiation source in a base pressure of 10⁻⁷ Pa. Before characterization, the modified Li foil was sealed with a polyimide film in order to prevent undesirable reactions in air.

Symmetric lithium-metal coin cells and Li/Cu half-cells (CR2025 type) were assembled in a glove box filled with high-purity argon. The symmetric lithium-metal coin cells were used for research the stability of lithium electrodeposition. Li/Cu half-cells were prepared for electrochemical performance characterization and the dendrite morphology observation. A polypropylene micro-porous film (Celgard 2300) was used as the separators and a solution of 1 M LiPF₆ in ethylene carbonate (EC) and dimethyl carbonate (DMC) (1 : 1 in volume) was employed as the electrolyte. The galvanostatic charge–discharge tests were carried out on LAND battery test system (Wuhan, China). Cyclic voltammogram (CV) tests were carried out using the CHI660C electrochemical workstation in a potential range of −0.4 to 0.8 V (vs. Li/Li⁺) at a scan rate of 1 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) measurements were performed on the CHI660C over a frequency range of 100 kHz to 10 mHz and the amplitude was set to 10 mV. After the electrochemical tests, the working electrodes were removed from the coin cells with the battery removal machine (Kejing, MSK-110D).

3. Results and discussion

The XRD patterns of GO and rGO films are shown in Fig. 1a. The diffraction peak (001) at 11.9° of GO film is a typical peak of GO and the additional broadened peak of rGO film around 16° confirms the conversion of GO to rGO after annealing in Ar as described in the previous literatures.^41,42 The diffraction peak of rGO film at 26° represents the (002) crystal plane of rGO, indicating a significant shrinking in lattice space. The Raman spectrum of GO and rGO films are shown in Fig. 1b. The D-band at around 1350 cm⁻¹ and G-band at around 1590 cm⁻¹ are observed. After annealing, the intensity ratio of D-band to G-band (I_D/I_G) is increased from 0.91 to 1.08, indicating that the GO film has been reduced successfully. The EDS mapping patterns and photographs of GO and rGO films are shown in Fig. S2.† There is about 28 wt% O in GO film while only 8 wt% O remains after annealing. As shown in Fig. S1a,† the free-standing GO film is dark brown. By contrast, the rGO film shown in Fig. S1b† is dark grey with shining metallic luster because of the increase in reflectivity of visible light.²⁸ Therefore, the GO film was successfully reduced into rGO after annealing at 600 °C in Ar.

Fig. 1 (a) XRD patterns of GO and rGO films, (b) Raman spectra of GO and rGO films.

The top surface and cross sectional SEM images of GO, rGO film and rGO/Li electrode are shown in Fig. 2. The surface morphology of GO film is flat with slight fold (Fig. 2a), and the cross-sectional image exhibits a compact layer-by-layer stacking structure (Fig. 2b), with a thickness of about 3 μm. As shown in Fig. 2c, the top surface of rGO film becomes coarse, and the rGO sheets are crumpled with many wrinkles. During the filtering process, the GO sheets tended to aggregate and assemble along the liquid/membrane interface. The nascent GO will then capture the other GO sheets through interlayer van der Waals forces and began to stack.²⁸ As a result, the layer-by-layer film with ordered stacking morphology was formed. After the annealing of GO film in argon atmosphere, the rGO film can be obtained. As indicated by the cross-sectional SEM images (Fig. 2d), the morphology of rGO film maintains well. After pressing, the rGO film can integrated with Li foil closely, due to the good malleability of Li. According to the surface and cross-sectional images of rGO/Li electrode (Fig. 2e and f), the rGO film tiles on the Li foil tightly and the surface of Li is squeezed into the layered clearance of graphene film homogeneously.

Fig. 2 (a and b) Surface and cross-sectional SEM images of GO film; (c and d) surface and cross-sectional images of rGO film; (e and f) surface and cross-sectional images of rGO/Li electrode.

Voltage versus time profiles for pure Li and rGO/Li electrodes in symmetric lithium cell system are shown in Fig. 3. Each half cycle lasts for 3 h at 0.4 mA cm⁻², three-hour interval of strip/plate is chosen to allow enough lithium to transport between the electrodes. Therefore, the voltage change could be observed.^43,44 Fig. 3a is the voltage versus time profiles for rGO/Li–Li symmetric lithium cell, which exhibits a stable voltage change until 600 h (100 cycles). The voltage change during the first 60 h (10 cycles) can be ascribed to the activation reaction. By contrast, the voltage versus time profiles for pure Li–Li symmetric lithium cell up and down violently (Fig. 3b). The haphazard changes in the profiles indicate the poor stability of the cell, which can be ascribed to the dendrites formation on the surface of pure Li electrodes. The good stability of rGO/Li–Li symmetric lithium cell can be attributed to the protective effect of rGO film.

Fig. 3 Voltage versus time profiles for rGO/Li (a) and pure Li (b) electrodes in symmetric lithium cell system with each half-cycle lasts 3 h.

Electrochemical performance of pure Li and rGO/Li electrodes in Li/Cu half-cell system are shown in Fig. 4. In order to characterize the stability of pre-formed SEI film, the impedance performance and equivalent circuit diagram of pure Li and rGO/Li electrodes stored at room temperature for 1 h, 24 h and 48 h are researched. As shown in Fig. 4a, the curve shape of pure Li electrodes stored for different times display an obvious change. The charge-transfer resistance (R_ct) of pure Li electrode stored at room temperature for 1 h, 24 h and 48 h is 238.4 Ω, 112.0 Ω and 207.3 Ω, respectively. The electrolyte has not been fully infiltrating when the battery is just finished, which leading to the R_ct value of Li stored for 1 h electrode is bigger than the other two samples. After that, the decline of impedance performance can be ascribed to the complex reaction between the Li electrodes and the electrolyte. With extending the storage time, there will be more side reaction happened, leading to the non-uniform morphology of SEI layer on the surface of pure Li electrode. By contrast, the Nyquist curve of rGO/Li electrode is more stable as the storage time extends (Fig. 4b). The R_ct value of Li/rGO electrode stored at room temperature for 1 h, 24 h and 48 h is 318.1 Ω, 195.8 Ω and 190.6 Ω, respectively. The rGO modification increases the impedance to some extent, but the R_ct values display a small change when the electrodes have been fully wetting, indicating that the rGO film on the Li surface can prevent the occurrence of side reaction efficiently. The CV curves of pure Li and rGO/Li electrodes are displayed in Fig. 4c. Different from conventional CV test, the electrodeposition of Li starting from −0.09 V in the negative scanning process is an overpotential-driven nucleation/growth process. The positive current above 0 mA represents the dissolution process in the positive scanning process. It is interesting that there are two dissolution peaks for the pure Li, while there is only one single peak for the rGO/Li electrode. According to previous works,^26,27 the dissolution peaks of pure Li electrode at around 0.2 V and 0.6 V represent the dissolution of Li deposits and the stripping of insoluble reduction species, respectively. The insoluble reduction species precipitated on the Cu substrate surface are Li halides, LiOH, Li₂O and ROCO₂Li and so on.²¹ By contrast, the single peak of the rGO/Li electrode reflects the re-oxidation of Li deposits.⁴⁵ Moreover, the peak area of the rGO/Li electrode is much smaller than that of pure Li, indicating fewer side reactions between the modified Li electrode and the electrolyte. The cycling efficiencies of both the electrodes are shown in Fig. 4d. The cycling efficiencies of rGO/Li electrode still remain at 90% after 100 cycles, while those of pure Li electrode decay rapidly. After 50 charge–discharge cycles, the cycling efficiencies of pure Li electrode vanish abruptly, which means the totally broken of pure Li/Cu cell. From what has been discussed above, the rGO film can improve the electrochemical performance efficiently. On one hand, the excellent mechanical flexibility of rGO sheet is helpful to suppress the Li dendrite growth. On the other hand, the layer-by-layer structure of rGO film provides a flexible lithium storage space and can adjust to the volumetric change during cycling.

Fig. 4 Electrochemical performance of pure Li and rGO/Li electrodes in Li/Cu half-cell: Nyquist plots and the equivalent circuit for (a) pure Li and (b) rGO/Li electrodes storage at room temperature for 1 h, 24 h and 48 h; (c) CV curves of pure Li and rGO/Li electrodes; (d) cycling efficiency of pure Li and rGO/Li electrodes.

The surface and cross sectional SEM images of the pure Li and rGO/Li electrodes after 50 cycles are shown in Fig. 5. The top-view images of the pure Li electrode display sharp granule morphology (Fig. 5a), while the surface SEM image of the rGO/Li electrode after 50 cycles shows relatively flat mossy-like morphology (Fig. 5c). Combining with Fig. 3 and 4, it can be suggested that the rGO film can help suppress the formation of dendritic Li efficiently. During the charge–discharge process, Li⁺ shuttled between the Li and the counter electrode, the uneven dissolution of the dendrites leaves Li crystals detached from the Li substrate. The isolated Li crystals become electrochemically “dead” but chemically reactive due to their high surface area.⁴⁶ For the pure Li electrodes, the dead Li will accumulate on the surface of electrode and cause a chain reaction because of its uneven electric potential. As shown in Fig. 5b, the cross sectional SEM image of pure Li electrode displays collapse morphology after 50 cycles. As for the rGO-modified Li electrode, the dead Li can be “stored” in the rGO sheet, and the uneven surface potential can be restrained significantly.

Fig. 5 Surface and cross-sectional SEM images after 50 cycles: (a and b) pure Li electrode; (c and d) rGO/Li electrode.

Fig. 6 shows the C 1s and Li 1s XPS spectra of the pure Li and rGO/Li electrodes after 50 cycles. According to the NIST X-ray Photoelectron Spectroscopy Database, the peaks at 284.6 eV in C 1s XPS spectra can be assigned to the graphite band (G-band, sp² bonding). In addition, the C–O bonding, C=O bonding, –C–OR/–C–OH bonding and Li₂CO₃/C–F bonding are also found at binding energy of 286.9, 288.8, 286.2 and 289.7 eV, respectively (Fig. 6a). While in the Li 1s XPS spectra of the pure Li electrode after 50 cycles (Fig. 6b), there is an unknown peaks at 58.2 eV existed except for the peak of LiO₂, LiF and Li₂CO₃. By contrast, the constituent of the surface of rGO/Li electrode after 50 cycles is not that complicated. There is only sp²-C, –C–OR/–C–OH and Li₂CO₃/C–F can be observed in Fig. 6c and d. The unknown composition of the pure Li electrode can be ascribed to the side reaction between metallic Li and the electrolyte. Related to the CV curves of pure Li and rGO/Li electrodes (Fig. 4c), the side reactions between the Li and the electrolyte are suppressed obviously due to the modification of rGO film.

Fig. 6 XPS spectra of pure Li and rGO/Li electrodes after 50 cycles: (a) C 1s XPS spectra of pure Li electrode; (b) Li 1s XPS spectra of pure Li electrode; (c) C 1s XPS spectra of rGO/Li electrode; (d) Li 1s XPS spectra of rGO/Li electrode.

To investigate the process of the formation of dendritic lithium, the morphologies of the as-prepared electrode after different cycles were measured by SEM. The dendrite morphology cannot be observed from Fig. 7a, showing that the dendrite has not yet formed after 10 cycles. While after 20 cycles (Fig. 7b), there are many granular protuberances can be observed on the surface of rGO/Li electrode. It is suggested that the formation of dendritic Li has already begun. As shown in Fig. 7c, the granular protuberances grow up vertically and present a rugged topography. After 40 to 50 cycles (Fig. 7d and e), these granular dendrites grow up laterally and connect with each other to form a noodle-like dendrite morphology. The Nyquist plots of the rGO/Li electrode after different cycles are shown in Fig. 7f. Both the plots are mainly composed of a small intercept at high frequency, an arc at high to medium frequency and a linear part in low frequency. The arc in high frequency region represents the charge transfer resistance R_ct and the electrical double-layer capacitance. The semicircle radius of rGO/Li electrode grow larger continuously during cycling, indicating the increase in charge transfer resistance, which can be ascribed to the growth of dendritic Li and the formation of passivated SEI layer. There is no obvious change in the straight slope during cycling, suggesting that the conductivity of Li⁺ has not been affected significantly during the dendrite growth process.

Fig. 7 SEM images of rGO/Li electrodes after different cycles: (a) 10 cycles; (b) 20 cycles; (c) 30 cycles; (d) 40 cycles; (e) 50 cycles; and (f) the Nyquist plots of rGO/Li electrodes after different cycles.

4. Conclusions

In summary, we have constructed rGO/Li composite electrode by combining rGO film with Li metal. In view of good binding and synergistic effect between the rGO layer and Li, the rGO/Li electrode displays enhanced electrochemical performances with better cycling stability than the unmodified counterpart. Importantly, the growth of dendritic Li is successfully suppressed in the rGO/Li composite electrode. In addition, the layer-by-layer structure of rGO film can also help “store” the dead Li and restrain the uneven electric potential on the surface of electrode.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (51271167) and the Program for Innovative Research Team in University of Ministry of Education of China (IRT13037).

References

1. P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J. M. Tarascon, Nat. Mater., 2012, 11, 19–29.
2. J. C. Guo, Y. H. Xu and C. S. Wang, Nano Lett., 2014, 11, 4288–4294.
3. H. Kim, H. D. Lim, J. Kim and K. Kang, J. Mater. Chem. A, 2014, 2, 33–47.
4. S. H. Oh, R. Black, E. Pomerantseva, J. H. Lee and L. F. Nazar, Nat. Chem., 2012, 4, 1004–1010.
5. Z. Peng, S. A. Freunberger, Y. Chen and P. G. Bruce, Science, 2012, 337, 563–566.
6. M. A. Lowe, J. Gao and H. D. Abruna, RSC Adv., 2014, 4, 18347–18353.
7. B. Scrosati, J. Hassoun and Y. K. Sun, Energy Environ. Sci., 2011, 4, 3287–3295.
8. F. Ding, W. Xu, X. L. Chen, J. Zhang, Y. Y. Shao, M. H. Engelhard, Y. H. Zhang, T. A. Blake, G. L. Graff, X. J. Liu and J. G. Zhang, J. Phys. Chem. C, 2014, 118, 4043–4049.
9. F. Ding, W. Xu, G. L. Graff, J. Zhang, M. L. Sushko, X. Chen, Y. Shao, M. H. Engelhard, Z. Nie, J. Xiao, X. Liu, P. V. Sushko, J. Liu and J. G. Zhang, J. Am. Chem. Soc., 2013, 135, 4450–4456.
10. Y. Zhang, J. Qian, W. Xu, S. M. Russell, X. Chen, E. Nasybulin, P. Bhattacharya, M. H. Engelhard, D. Mei, R. Cao, F. Ding, A. V. Cresce, K. Xu and J. G. Zhang, Nano Lett., 2014, 14, 6889–6896.
11. M. F. Wu, Z. Y. Wen, Y. Liu, X. Y. Wang and L. Z. Huang, J. Power Sources, 2011, 196, 8091–8097.
12. J. B. Goodenough and Y. Kim, Chem. Mater., 2010, 22, 587–603.
13. A. Lewandowski and A. Świderska-Mocek, J. Power Sources, 2009, 194, 601–609.
14. P. Verma, P. Maire and P. Novák, Electrochim. Acta, 2010, 55, 6332–6341.
15. S. S. Zhang, J. Power Sources, 2006, 162, 1379–1394.
16. D. Aurbach, A. Zaban, Y. Gofer, O. Abramson and M. Benzion, J. Electrochem. Soc., 1995, 142, 687–696.
17. D. Aurbach, A. Zaban, Y. Gofer, Y. E. Ely, I. Weissman, O. Chusid and O. Abramson, J. Power Sources, 1995, 54, 76–84.
18. C. Brissot, M. Rosso, J. N. Chazalviel and S. Lascaud, J. Power Sources, 2001, 94, 212–218.
19. N. S. Choi, Y. M. Lee, K. Y. Cho, D. H. Ko and J. K. Park, Electrochem. Commun., 2004, 6, 1238–1242.
20. F. Croce, G. Appetecchi, L. Persi and B. Scrosati, Nature, 1998, 394, 456–458.
21. K. Kanamura, H. Tamura, S. Shiraishi and Z. I. Takehara, J. Electroanal. Chem., 1995, 394, 49–62.
22. H. Ota, K. Shima, M. Ue and J. I. Yamaki, Electrochim. Acta, 2004, 49, 565–572.
23. G. Zheng, S. W. Lee, Z. Liang, H. W. Lee, K. Yan, H. Yao, H. Wang, W. Li, S. Chu and Y. Cui, Nat. Nanotechnol., 2014, 9, 618–623.
24. A. A. Arie and J. K. Lee, Diamond Relat. Mater., 2011, 20, 403–408.
25. A. A. Arie, O. M. Vovk, J. O. Song, B. W. Cho and J. K. Lee, J. Electroceram., 2008, 23, 248–253.
26. Y. J. Zhang, X. Y. Liu, W. Q. Bai, H. Tang, S. J. Shi, X. L. Wang, C. D. Gu and J. P. Tu, J. Power Sources, 2014, 266, 43–50.
27. Y. J. Zhang, W. Wang, H. Tang, W. Q. Bai, X. Ge, X. L. Wang, C. D. Gu and J. P. Tu, J. Power Sources, 2015, 277, 304–311.
28. C. M. Chen, J. Q. Huang, Q. Zhang, W. Z. Gong, Q. H. Yang, M. Z. Wang and Y. G. Yang, Carbon, 2012, 50, 659–667.
29. C. Jiang, S. Markutsya and V. V. Tsukruk, Adv. Mater., 2004, 16, 157–161.
30. S. F. Pei, J. P. Zhao, J. H. Du, W. C. Ren and H. M. Cheng, Carbon, 2010, 48, 4466–4474.
31. J. Z. Wang, C. Zhong, S. L. Chou and H. K. Liu, Electrochem. Commun., 2010, 12, 1467–1470.
32. Y. X. Xu, H. Bai, G. W. Lu, C. Li and G. Q. Shi, J. Am. Chem. Soc., 2008, 130, 5856–5857.
33. C. M. Chen, Y. G. Yang, Y. F. Wen, Q. H. Yang and M. Z. Wang, New Res. Carbon Mater., 2008, 23, 345–350.
34. D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. Dommett, G. Evmenenko, S. T. Nguyen and R. S. Ruoff, Nature, 2007, 448, 457–460.
35. S. Park, N. Mohanty, J. W. Suk, A. Nagaraja, J. An, R. D. Piner, W. Cai, D. R. Dreyer, V. Berry and R. S. Ruoff, Adv. Mater., 2010, 22, 1736–1740.
36. V. H. Pham, T. V. Cuong, S. H. Hur, E. W. Shin, J. S. Kim, J. S. Chung and E. J. Kim, Carbon, 2010, 48, 1945–1951.
37. Y. F. Xu, G. K. Long, L. Huang, Y. Huang, X. J. Wan, Y. F. Ma and Y. S. Chen, Carbon, 2010, 48, 3308–3311.
38. X. L. Li, G. Y. Zhang, X. D. Bai, X. M. Sun, X. R. Wang, E. G. Wang and H. J. Dai, Nat. Nanotechnol., 2008, 3, 538–542.
39. C. Chen, Q. H. Yang, Y. Yang, W. Lv, Y. Wen, P. X. Hou, M. Wang and H. M. Cheng, Adv. Mater., 2009, 21, 3007–3011.
40. W. S. Hummers Jr and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
41. H. Tang, J. P. Tu, X. Y. Liu, Y. J. Zhang, S. Huang, W. Z. Li, X. L. Wang and C. D. Gu, J. Mater. Chem. A, 2014, 2, 5834–5840.
42. H. Tang, Y. J. Zhang, Q. Q. Xiong, J. D. Cheng, Q. Zhang, X. L. Wang, C. D. Gu and J. P. Tu, Electrochim. Acta, 2015, 156, 86–93.
43. Y. Y. Lu, Z. Y. Tu and L. A. Archer, Nat. Mater., 2014, 13, 961–969.
44. Y. Y. Lu, Z. Y. Tu, J. Shu and L. A. Archer, J. Power Sources, 2015, 279, 413–418.
45. Y. J. Liu, Y. Cheng, T. Ohba, K. Kaneko and H. Kanoh, Int. J. Hydrogen Energy, 2011, 36, 12902–12908.
46. K. Xu, Chem. Rev., 2004, 104, 4303–4418.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra25553h

---