Power-output reduction of graphene oxide and a MnO₂-free Zn/GO primary cell

Abstract

The reduction of graphene oxide (GO) by zinc powder is essentially an electron-input process. Is it possible to output the electron during the reduction for a primary battery? Here, a primary battery with a Zn plate as the anode and a centimeter-sized dried foam film of graphene oxide (GO) as the cathode has been fabricated, and the electron-output has been achieved while a reduced graphene oxidised film was obtained synchronously. The output capacity of the batteries depends on the oxygenated degree and the amount of GO, and the maximum specific capacity is 642 mA h g⁻¹, while the average voltage is up to 0.6 V. Electrochemical impedance spectroscopy (EIS) measurements revealed the decreasing resistance of the Zn/GO cell during the discharging process. An LED lamp can be lit by the battery, indicating its potential for practical applications. In addition, the primary battery is MnO₂-free, and hence is more environmentally benign than the traditional Zn/Carbon dry batteries.

---

Introduction

Since obtaining a single layer structure in 2004,¹ graphene, the thinnest carbon material, has attracted enormous attention for its potential applications in many fields.^2–7 Batteries are the major applied fields, while graphene has been used as materials for electrodes because of its high electrical conductivity and surface area.^8–11 Basically, the preparation of graphene or reduced graphene oxide (rGO) is still a challenge, especially on a large scale.¹² Chemical reduction of graphene oxide (GO) is a major route to prepare rGO, and various kinds of reducing agents or methods have been reported.¹³ Among them, metals, such as Zn, Al, Fe etc., have been used as reducing agents for rGO preparation using GO as the feedstock.^14–18 Mechanism studies revealed that the driving force for the reduction comes from the release of electrons from the metal to the GO sheets while the metal was oxidized. The acceptance of electrons by GO under acidic conditions help in removing the oxygen-containing groups in GO and results in the formation of rGO.¹⁵ Similar processes can also be found in the electrochemical reduction of GO to rGO, where GO nanosheets are directly coated on the surface of electrodes during electron injection, and the process is a typical electrolysis with the consumption of electric energy.^19,20 Is it possible to output power during the reduction of GO?

A primary cell or battery is an ideal device for power-output, and usually a metal works as the anode, for example, a Zn/carbon cell, which has been widely used as a dry battery in daily applications.^21,22 Typically, in this process, metal anode lost electrons while oxidation to form metal ions diffusion into the electrolyte, and the electrons were output by a conducting wire to electrical devices, which were then injected into the cathode to reduce the cathode materials. Similarly, the reduction of GO by metals can be used to design a primary cell, where the metal (Zn plate) works as the anode and GO works as the cathode (Fig. 1a). The power-output can be achieved by using an exterior passage to export electrons while the GO can be reduced at the cathode, playing the role of a depolarizer, and the as formed rGO directly works as the current collector and cathode. In the process, aqueous solutions of NH₄Cl or KOH are used as electrolytes. In addition, in a typical Zn/C cell, MnO₂ with Hg is usually employed as the cathode material, where MnO₂ and Hg are the source of metal pollution during the disposal of waste batteries.^23,24 Therefore, MnO₂-free cells will be attractive because of its environmentally benign properties. Here, in the design of a new Zn/GO cell, GO will replace of MnO₂, and the next generation of MnO₂-free cell is the target.

Fig. 1 Schematic diagram showing the Zn/GO cell (a), photoimages of the liquid film of GO (b), dried foam film of GO (c), and rGO film formed during work as the cathode of the Zn/GO cell (d), and discharge curve of the Zn/GO primary cell with Zinc plate as anode and the dried foam film of GO as the cathode (e) or with GO deposited on nickel foam as anode (f). Both were carried out in KOH electrolyte at 1.5 mA.

Here, a free-standing film of GO was directly employed as the cathode as an antitype of the Zn/GO cell, and a powder-output primary cell has been fabricated.

Experimental materials

Graphite powder, 100 mesh, 99.9% (metals basis) acetylene was purchased from Alfa Aesar. 60% PTFE was purchased from Aladin. Analytical grade NaNO₃, P₂O₅, ZnCl₂, NH₄Cl, KClO₃, KOH, KMnO₄, 98% H₂SO₄, 30% H₂O₂ aqueous solution, zinc plate were purchased from Shanghai Chemical Reagents Company, and were used directly without further purification. Ultra-pure water (18 MΩ) was produced by a Millipore System (Millipore Q, USA). Chlorosulfonic acid (ClSO₃H) was purchased from Graxia Chemical Technology Co., Ltd. Electrolytically precipitated, Manganese(IV) Oxide (88%) was purchased from Alfa Aesar.

Preparation of GO

600 mg graphite powder (100 mesh) was immersed with 6.0 mL chlorosulfonic acid in a Teflon reactor at 150 °C for 12 h, and the mixture was transferred to a 250 mL beaker after the addition of 15.0 mL H₂SO₄ and 1.0 g P₂O₅. Then, 8.0 mL 30% H₂O₂ was slowly added into the acidic mixture. Subsequently, 200.0 mL DI water was added to wash the sample, and the expanded graphite powder was collected by filtration and freeze-drying. The expanded graphite was employed as the raw material for oxidation to prepare graphene oxide; the expanded graphite prepared by the rapid thermal expansion method can also be used as a feedstock. For GO preparation, a modified Brodie method was employed.²⁵ 0.2 g expanded graphite was mixed with 16.0 mL 98% H₂SO₄, 3.0 mL concentrated nitric acid and a specific amount of KClO₃, and the reaction process was continued for 4 days at 30 °C. The amount of KClO₃ was 1.0 g, 1.6 g and 2.2 g for GO-1, GO-3 and GO-4, respectively. Then, the sediments were collected after centrifugation at 7200 rpm for 6 times, and the final GO powders were obtained after the freeze-drying treatment. The thermally treated GO-4 powder at 90 °C for 3 days was named GO-2. GO was also prepared by the modified Hummers method using expanded graphite as the raw material. Typically, 400 mg expanded graphite was added into a 50 mL beaker and 25.0 mL of H₂SO₄ was added under stirring in an ice-bath. Then, a specific amount of KMnO₄ (1.2 g KMnO₄ for the moderate oxidation sample and 2.4 g KMnO₄ for the strong oxidation sample) was added slowly into the beaker under stirring and the temperature of the system was controlled lower than 20 °C. Next, the system was left to react for 1 day at room temperature, and then 100 mL water was slowly added into the system and stirred for another 20 min. 80 mL of hot water at a temperature of 60 °C and 3.0% H₂O₂ aqueous solution were added to reduce residual KMnO₄ until the bubbling disappeared. GO was collected by centrifugation at 7200 rpm for 6 times. The final GO powder was obtained after the freeze-drying treatment.

Preparation of dried GO foam films

First, an aqueous solution of GO at a concentration of 6 mg mL⁻¹ was prepared by dispersing GO in deionized water under mild ultrasonication for 15 min, and 0.1 v% butylamine was added. Then, a cycle of Ag or copper wire was vertically or horizontally dipped into the solution and gently lifted from the solution. A small volume of the aqueous solution of GO was captured by the substrates during the process, and the substrates were then allowed to stand in air for hours to dry. In general, the drying process was carried out at room temperature under normal humidity conditions (∼20%). For a rapid GO foam film fabrication, it can also be dried in warm air at about 60 °C.

Fabrication of Zn-GO primary cell

For the antetype cell, the dried foam film was directly used as the cathode, a zinc plate was used as the anode, and NH₄Cl worked as the electrolyte. Nickel foam was used as the current collector for the GO cathode, and 15 mg GO was deposited on it without any addition. If the GO powder was used as the active electrode material, the cathode was prepared by mixing GO, PTFE and acetylene with an amount ratio of 7 : 2 : 1; carbon felt or foam nickel was employed as the anodic collector while NH₄Cl or KOH aqueous solutions were used as the electrolytes, respectively. A zinc plate (1.5 cm × 2.0 cm × 0.1 cm) was employed as the anode material, and a cellulose filter was employed as the cell membrane. In general, the NH₄Cl electrolyte was composed of 4.0 M NH₄Cl, 1.0 M ZnCl₂, and a small amount of BiCl₃, while the alkaline electrolyte was composed of 5.0 M KOH.

Electrochemical characterization

Cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS) measurements and galvanostatic discharge experiments were performed on a CHI660C potentiostat–galvanostat (CH Instruments Inc.). For EIS measurements, the AC frequency ranged from 0.01 to 10⁵ Hz with an AC amplitude of 5 mV. For CV analysis, 5.0 M KOH was placed in an electrochemical cell as the electrolyte, which included a piece of nickel foam containing 1 mg active material (working electrode), a platinum wire (counter electrode) and an Hg/Hg₂Cl₂ reference electrode. CV tests were carried out at a scan rate of 0.5 mV s⁻¹ over a potential range of 0 to 0.9 V (vs. SCE). To collect rGO after discharge without impurity for experimental analysis, GO was deposited in the nickel foam without additional PTFE and acetylene; the end voltage was 0.6 V and 0.25 V for 0.6 V discharged GO and fully discharged GO, respectively. After discharge, the rGO powder was collected by ultrasonication and washed with HCl (1 M) and water before drying.

Characterization

Wide-angle X-ray diffraction (XRD) analyses were carried out on an X-ray diffractometer (D/MAX-1200, Rigaku Denki Co. Ltd., Japan). XRD patterns with Cu Ka radiation (λ = 1.5406) at 40 kV and 100 mA were recorded in the range of 2θ = 5–65°. A commercial atomic force microscope (AFM, Nanoscope IIIa; Digital Instruments, Santa Barbra, CA) equipped with a J scanner was used to measure the morphologies and thicknesses of the samples. A Si₃N₄ tip (Nanoprobes, Digital Instruments Inc.) was used in the contact mode. Scan rates were between 1.0 and 2.4 Hz. X-ray photoelectron spectroscopy (XPS) were recorded on an Escalab MK II photoelectron spectrometer (VG Scientific Ltd., United Kingdom). Raman spectra were obtained on a RAMAMLOG 6 (Spex, USA) using a 50× objective lens with a 514.5 nm laser excitation. The structures of the films were measured using a Sirion 200 FESEM at an accelerating voltage of 10 kV. Elemental analysis was carried out by a VARIO ELIII analyzer (Elemental Co., Germany). The optical images of film formation were determined by optical microscopy (POM) (LW200-PC, Cewei, Shanghai) using a 40× objective lens at room temperature.

Results and discussion

In our previous study, we found that it is easy to prepare a centimeter-size free-standing ultrathin dried foam film of GO by the novel bobble film technique with a metal frame;²⁶ here, a copper rectangular frame was used to prepare the free-standing film of GO, as shown in Fig. 1b and c. The as-prepared free-standing GO film was transparent with 1.5 cm in width and 3.1 cm in length, and it was directly used as the cathode, while a Zn plate worked as the anode and NH₄Cl as the electrolyte for the Zn/GO cell. Once the cell began to work, the transparent GO film became dark from the brim to the interior of the film, and at the end, the film became completely dark, indicating the efficient reduction of the GO film to rGO. Notably, during the process we did not observe hydrogen bubbles occurring from the cathode, thus demonstrating the effective electron capture ability of GO.

Fig. 1e and f show the discharge performance of a Zn/GO cell using the above-mentioned rGO film as the cathode, and the discharge current density is 1.5 mA. The cell voltage is about 0.3 V at the beginning, decreases slowly to about 0.18 V at 2100 s, and then quickly decreases to zero at a time of 2400 s. Before 2100 s, the cell works normally with the reduction of the GO film by electrons, and at the end no more GO can be reduced by the process, and the cell is fully discharged while a black rGO film (Fig. 1c) was obtained. The discharge reaction of the Zn/GO primary cell can be described as:

Anode: Zn + 4OH⁻ → [Zn(OH)₄]²⁻ + 2e

Cathode: GO + 2e + 2H⁺ → rGO + H₂O

The foam film was limited by its centimeter scale size for the large scale reduction of GO by the discharging process. GO was deposited on nickel foam, which could act as a current collector. Improved conductivity of the working electrode resulted in a higher working voltage of over 0.6 V.

Fig. 2a shows the cyclic voltammograms of MnO₂ and GO prepared by the Hummers or Brodie methods, respectively, and the scan rate is 0.5 mV s⁻¹ using KOH as the electrolyte. For MnO₂, both the reduction and oxidation peak appear, and the reduction peak is at −0.267 V. However, for Brodie and Hummer GO, the reduction peaks appear at −0.76 V and −0.81 V, respectively, while their relative chemical contents are C₂O_1.18H_1.05 and C₂O_1.32H_1.06, indicating the lower working voltage of GO than that of MnO₂.

Fig. 2 Cyclic voltammetry analysis of different cathode materials at 0.5 mV s⁻¹, three electrode system, nickel foam as current collector of working electrode (a); XPS profiles of the GO (b) and rGO (c) after the cell was fully discharged; XRD (d), FT-IR (e), and Raman (f) spectra of GO and rGO formed at 0.6 V or fully discharged.

XPS measurements could provide the direct evidence of the reduction of GO. Fig. 2b and c show the C_1s XPS spectra of both the Brodie GO and the as-formed rGO after the cell is fully discharged. Both the curves were fitted considering the following contributions: C=C (sp²; peak 1), C–C (sp³; peak 2), C–O/C–O–C (hydroxyl and epoxy groups; peak 3), C=O (carbonyl groups; peak 4), O–C=O (carboxyl groups; peak 5).²⁷ Fig. 2b also shows the C_1s peaks of GO, which consists of three main components arising from C–O (hydroxyl and epoxy, 286.9 eV), C=O (carbonyl, 287.8 eV) and C=C/C–C (284.9 eV) species, and a minor component of O=C–O (carboxyl, 289.5 eV) species. After reduction by the fully discharged cell, the oxygen species of C–O (hydroxyl and epoxy, 286.9 eV), C=O (carbonyl, 287.8 eV) and O=C–O (carboxyl, 289.5 eV) are reduced significantly, as shown in Fig. 2c, indicating an efficient deoxidization. The major species remaining were C=C and C–C. The calculated C/O mole ratio increased from 1.64 to 8.4 after reduction.

Fig. 2d shows the X-ray diffraction (XRD) patterns of GO and rGO obtained after 0.6 V discharge and full discharge. For GO, there are two diffraction peaks at 7.2° and 20.3°, which corresponds to a d-spacing of 1.23 and 0.44 nm, respectively. After discharge, the diffraction peaks appear at 23.4° and 24.1° for 0.6 V discharged and fully discharged rGO, respectively, which corresponds to the interlayer spacing of 0.38 nm and 0.37 nm, respectively, which are still higher than that of the pristine graphite (0.34 nm), indicating that the packing density of the rGO sheets became higher than that of GO.

Fig. 2e shows the typical FT-IR spectra of GO and rGO obtained by 0.6 V or full discharge. For GO, the characteristic peaks appear for carbonyl C=O (1726 cm⁻¹), aromatic C=C (1570 cm⁻¹), epoxy C–O (1410 cm⁻¹), and C–O (1070 cm⁻¹).²⁸ After reduction by discharging, the peaks for the epoxy functional group were reduced significantly, and the peaks for the carboxy and hydroxide groups also decreased observably, indicating the efficient reduction of GO during the discharging.

Fig. 2f shows the typical Raman spectra of GO and the rGO obtained by 0.6 V or full discharge. All the spectra indicate the existence of the D, G and 2D bands. For GO, the G band is located at 1595 cm⁻¹, while for the rGO obtained after 0.6 V or full discharge the G band moved to 1588 cm⁻¹, which is close to the value of pristine graphite and confirms the reduction of GO during the discharge. However, the presence of the D band at 1341, 1349 and 1353 cm⁻¹ corresponding to the GO and rGO of 0.6 V and full discharge also predicts the defect of the sample and the size of the in-plane sp2 domain.²⁹ The intensity ratio of the D and G band (I_D/I_G), varies from 0.96 to 1.33 (for both discharged rGO). The result is quite unexpected and apparently contradicts with the idea that a significant reduced D band should be observed after the electrochemical reduction of oxidized graphene. It is believed that this contradiction comes from the amorphous character of GO; for a large distortion of the 6-fold aromatic rings in amorphous carbon, the I_D/I_G ratio decreases instead of increasing upon the oxidization of this highly defected carbon structure. Thus, here, the I_D/I_G ratio cannot be used as a measure of structural disorder and, accordingly, comparisons between the materials of amorphous GO and graphitic rGO are no longer valid. Similar results also have been reported by Stankovich et al. during the chemical reduction process of GO.³⁰

The above-mentioned results reveal that GO can be efficiently reduced when it acts as the cathode in the Zn/GO cell. However, the voltage and capacity are still low, indicating that the amount of GO in the cell is insufficient.

Improved conductivity of the electrode and the amount of GO could increase the discharge voltage before it reaches the ideal potential of the Zn/GO cell. Here, a new Zn/GO cell was fabricated where the cathode was prepared by mixing GO powder, PTFE and acetylene in a mass ratio of 7 : 2 : 1; carbon felt or foam nickel were employed as the anodic collectors while NH₄Cl or KOH aqueous solutions were used as the electrolytes, respectively. A zinc plate (1.5 cm × 2.0 cm × 0.1 cm) was employed as the anode, and a cellulose filter was employed as the cell membrane. A LED lamp was connected by a conductive wire to the anode, while the cathode was not connected, as shown in Fig. 3c. Then, both the electrodes were connected to allow cell discharge, and the LED lamp was lit, as shown in Fig. 3d, indicating that the Zn/GO cell can output enough energy for practical applications. The galvanostatic curves of both GO and MnO₂ for Zn/GO or Zn/MnO₂ were measured. Fig. 3a shows the curve of MnO₂ in NH₄Cl or KOH electrolytes at 1.5 mA cm⁻², and discharge capacities are 237 mA h g⁻¹ and 206 mA h g⁻¹, respectively, when the discharge voltage is up to 0.6 V. For GO, the galvanostatic curves strongly depend on the formula of the GO, as shown in Fig. 3b; for GO-1, the discharge capacity is only about 200 mA h g⁻¹; while for GO-4, the maximum capacity is 642 mA h g⁻¹, and this value can expand to 705 mA h g⁻¹ if 0.5 V is the end voltage. Thus, the chemical composition of GO is a key factor. A series of GO have been tested in NH₄Cl electrolyte, and the results are listed in Table 1. Clearly, high oxygen content of GO favors the big capacity of the Zn/GO cell. To further clarify the relationship between oxidization degree and capacity of GO, a theoretical columbic capacity of 662 mA h g⁻¹ was calculated, according to the component change from GO to rGO, C₂O_1.18H_1.05 for GO as the cathode material, and C₂O_0.344H_0.458 for the discharged rGO. Both were measured by elemental analysis. The working capacity of the corresponding cell was measured to be 642 mA h g⁻¹, which is 92% of the theoretical value. This shows a high relativity between the oxidization degree and the capacity of the GO.

Fig. 3 Galvanostatic discharge curves of MnO₂ in NH₄Cl and KOH aqueous electrolytes (a) or GO with different oxidation degrees in NH₄Cl electrolyte at 1.5 mA cm⁻² (b); photos of a Zn/GO primary cell before (c) and after (d) the LED lamp was lit.

Table 1 Capacity, average working voltage and energy density of cathode materials in primary cells

	MnO2	MnO2	GO-1	GO-2	GO-3	GO-4	GO-ideal
Formula	MnO2	MnO2	C2O0.62H0.66	C2O0.96H0.90	C2O1.18H1.05	C2O1.44H1.13	C2O
Electrolyte	KOH	NH4Cl	NH4Cl	NH4Cl	NH4Cl	NH4Cl	—
Capacity (mA h g^−1)	237	206	216	610	642	505	1340
Voltage (V)	1.15	1.19	0.94	0.76	0.68	0.88	—
E (W h kg^−1)	272	245	204	466	440	445	—

---

What happened to GO during the discharge process? Here, the discharging process of the GO dried foam film was directly observed under a microscope during discharging (Fig. 4a), and the GO dried foam film was made according to our previous method³¹ with silver wire as the conductive frame. At the beginning of the discharge, the transparent GO film turned black at the edge, indicating the reduction of GO to rGO (Fig. 4c). Then, the process continued with the black region diffusing into the center of the film until the entire film became black (Fig. 4d–g), indicating the full discharge of the cell. This process also can be described as a domino process, as shown in Fig. 4b. The graphene oxide sheet near the silver frame captured electrons first and was reduced to rGO. Then, the conductive rGO became a new edge for electron transfer, causing the reduction of graphene oxide sheets until all the connected graphene oxide sheets were reduced. This also made the GO electrode discharging process smooth and displayed better conductivity properties than that of MnO₂, as shown in the EIS studies of cells (Fig. S4†).

Fig. 4 (a) In situ observation of the discharging process of GO dried foam film immersed in NH₄Cl electrolyte as cathode while zinc as anode, magnification at 40. (c–g) Color changes of the GO film during discharge process, scale bar 1 mm. (b) Domino-like reduction of GO nanosheets inside the film during discharging.

Conclusions

In summary, a power-output type reduction of GO has been developed, where the dried foam film of GO was used as the cathode and a zinc plate as the anode. The discharging process can efficiently reduce a GO film to an rGO film, while electrical energy was the output. It was found that the output capacity of the Zn/GO primary cell depends on the oxygen content of the GO sheets. A maximum specific capacity of 642 mA h g⁻¹ has been achieved when the formula of the GO is C₂O_1.18H_1.05, and the average voltage changed from 0.68 to 0.94 V. Carbonyl and epoxy groups are the main groups contributing to the capacity. Electrochemical impedance spectroscopy (EIS) measurements revealed the decreasing resistance of the Zn/GO cell during the discharging process. The reduction of the GO film during discharge can be described as a domino-like process. The Zn/GO primary cell has potential practical applications as a battery, and a LED lamp can be lit here. Moreover, we also can connect cells in series for a higher working voltage and wider applications.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (no. 51373162), the National Basic Research Program of China (No. 2010CB923302 and 2011CB921403), and The USTC Special Grant for Postgraduate Research, Innovation and Practice.

References

1. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
2. X. J. Wan, Y. Huang and Y. S. Chen, Acc. Chem. Res., 2012, 45, 598–607.
3. Y. Zhu, D. K. James and J. M. Tour, Adv. Mater., 2012, 24, 4924–4955.
4. D. A. C. Brownson, D. K. Kampouris and C. E. Banks, Chem. Soc. Rev., 2012, 41, 6944–6976.
5. Y. Chen, B. Zhang, G. Liu, X. D. Zhuang and E. T. Kang, Chem. Soc. Rev., 2012, 41, 4688–4707.
6. X. Huang, X. Y. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 2012, 41, 666–686.
7. K. P. Loh, Q. L. Bao, G. Eda and M. Chhowalla, Nat. Chem., 2010, 2, 1015–1024.
8. L. M. Dai, Acc. Chem. Res., 2013, 46, 31–42.
9. X. Huang, Z. Y. Zeng, Z. X. Fan, J. Q. Liu and H. Zhang, Adv. Mater., 2012, 24, 5979–6004.
10. Y. Q. Sun, Q. O. Wu and G. Q. Shi, Energy Environ. Sci., 2011, 4, 1113–1132.
11. C. H. Xu, B. H. Xu, Y. Gu, Z. G. Xiong, J. Sun and X. S. Zhao, Energy Environ. Sci., 2013, 6, 1388–1414.
12. C. K. Chua and M. Pumera, Chem. Soc. Rev., 2014, 43, 291–312.
13. T. Kuila, A. K. Mishra, P. Khanra, N. H. Kim and J. H. Lee, Nanoscale, 2013, 5, 52–71.
14. N. H. Kim, P. Khanra, T. Kuila, D. Jung and J. H. Lee, J. Mater. Chem. A, 2013, 1, 11320–11328.
15. S. Yang, W. B. Yue, D. Z. Huang, C. F. Chen, H. Lin and X. J. Yang, RSC Adv., 2012, 2, 8827–8832.
16. R. S. Dey, S. Hajra, R. K. Sahu, C. R. Raj and M. K. Panigrahi, Chem. Commun., 2012, 48, 1787–1789.
17. S. Sarkar and D. Basak, Chem. Phys. Lett., 2013, 561, 125–130.
18. X. G. Mei and J. Y. Ouyang, Carbon, 2011, 49, 5389–5397.
19. Y. Y. Shao, J. Wang, M. Engelhard, C. M. Wang and Y. H. Lin, J. Mater. Chem., 2010, 20, 743–748.
20. Z. J. Wang, X. Z. Zhou, J. Zhang, F. Boey and H. Zhang, J. Phys. Chem. C, 2009, 113, 14071–14075.
21. A. A. El Mel, M. Buffiere, C. P. Ewels, L. Molina-Luna, E. Faulques, J. F. Colomer, H. J. Kleebe, S. Konstantinidis, R. Snyders and C. Bittencourt, Carbon, 2014, 66, 442–449.
22. Z. Q. Wang, N. Bramnik, S. Roy, G. Di Benedetto, J. L. Zunino and S. Mitra, J. Power Sources, 2013, 237, 210–214.
23. S. P. Singh, M. Kumari, S. I. Kumari, M. F. Rahman, M. Mahboob and P. Grover, J. Appl. Toxicol., 2013, 33, 1165–1179.
24. H. A. Barrett, A. Ferraro, C. Burnette, A. Meyer and M. P. S. Krekeler, J. Power Sources, 2012, 206, 414–420.
25. B. C. Brodie, Philos. Trans. R. Soc. London, 1859, 149, 249–259.
26. W. F. Chen and L. F. Yan, Adv. Mater., 2012, 24, 6229–6233.
27. S. Hyeon-Jin, K. Ki Kang, A. Benayad, Y. Seon-Mi, P. Hyeon Ki, J. In-Sun, J. Mei Hua, J. Hae-Kyung, K. Jong Min, C. Jae-Young and L. Young Hee, Adv. Funct. Mater., 2009, 1987–1992.
28. S. Park, J. H. An, R. D. Piner, I. Jung, D. X. Yang, A. Velamakanni, S. T. Nguyen and R. S. Ruoff, Chem. Mater., 2008, 20, 6592–6594.
29. J. I. Paredes, S. Villar-Rodil, P. Solis-Fernandez, A. Martinez-Alonso and J. M. D. Tascon, Langmuir, 2009, 25, 5957–5968.
30. S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen and R. S. Ruoff, Carbon, 2007, 45, 1558–1565.
31. W. Chen and L. Yan, Adv. Mater., 2012, 24, 6229–6233.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra06354f

---