The morphology, structure and electrocatalytic ability of graphene prepared with different drying methods

Abstract

Reduction in an aqueous suspension of exfoliated graphene oxide sheets with ammonia water results in the formation of graphene. Here, vacuum drying and freeze drying were used to prepare graphene, and then the morphology and characteristics of the two samples were mainly compared via electron microscopy, spectra analysis, and electrochemical measurements. The changes in the oxygen-containing functional groups, confirmed using Fourier Transform Infrared Spectroscopy (FTIR), showed the successful preparation of graphene oxide and graphene. The characteristic peak at 11° disappeared, and a peak appeared at 24° in the X-ray diffraction (XRD) spectra. This confirmed that the formation of graphene from graphene oxide required a reaction time of 15 h. The results of X-ray photoelectron spectroscopy (XPS) revealed that the freeze-dried graphene contained more nitrogen and less oxygen, indicating a more complete reduction in the freeze dried graphene, this finding echoes the results of the Raman spectra. The value of the zeta potential of graphene prepared via freeze drying was about −32.6 mV, while for the graphene prepared via vacuum drying it was −21.6 mV. The results suggested that an aqueous solution of freeze-dried graphene has a stronger dispersion and stability. The BET surface areas of vacuum-dried graphene and freeze-dried graphene were 473.8 and 558.4 m² g⁻¹, respectively. The peak currents of graphene-modified glass carbon electrode samples prepared via these two methods of drying were 83.95 μA and 59.6 μA, respectively. The nearly vanished semicircle shown from electrochemical impedance spectroscopy highlights that the graphene had excellent electrocatalytic ability. Rotating disk electrode results in systems of [Fe(CN)₆]^3−/4− freeze-dried graphene modified GCEs showed lower electrocatalytic activity, while in a uric acid system it was more active towards the electrooxidation of uric acid. A steeper slope demonstrates better oxygen reduction reaction (ORR) activity, so freeze-dried graphene has more ORR activity. Above all, we found that the Gr2 had a relatively better capacity to promote charge transfer.

---

1. Introduction

Graphene (Gr) is a novel class of carbon-based nanomaterial. It consists of the sp² hybridization of carbon atoms tightly packed into a honeycomb structure.¹ Graphene is a collection of single-atom-thick 2-dimensional carbon atoms with a thickness of about 0.35 nm. It is the thinnest 2D material in the world.^2,3 It is also the basic building block for graphitic materials such as fullerenes, graphite, and carbon nanotubes.^4,5 Recently graphene has attracted intense interest due to its unique nanostructure and intrinsic characteristics including its remarkable optical characterization, mechanical properties, electrical conductivity, thermal conductivity, high surface area and excellent electron transport properties.^6–8 Despite its short history, graphene has recently attracted great attention from the electrochemical community. Several reviews focusing on graphene-based electrochemical applications have also appeared.

Several methods of preparing graphene have been developed since its discovery including mechanical exfoliation, epitaxial growth,⁹ chemical vapor deposition (CVD),^10,11 chemical reduction,¹² electrochemical exfoliation,¹³ electrochemical reduction,^14,15 etc. The oxidation–dispersion–reduction method mainly uses a modified Hummer's method¹⁶ and prepares graphite oxide (GtO). This then undergoes ultrasonic dispersion into graphene oxide solution followed by a reduction reaction. Researchers mostly applied toxic materials such as hydrazine hydrate, sodium borohydride, etc. as reducing agents to reduce GO. Meanwhile, the more toxic dimethyl amide used as a solvent avoids graphene aggregation. To reduce the use of toxic materials, ammonium hydroxide was chosen as a reducing agent for the preparation of graphene. This could not only improve the graphene stability, but also reduce the GO. The water in ammonia water can reduce the reaction and significantly decrease the oxygen content and thus improve the electrical conductivity. Though many studies have applied vacuum drying and freeze drying to prepare Gr, no one has yet compared their difference. Thus, in this work, we report for the first time the effect of different drying methods on Gr's morphology and electrical catalytic activity. The vacuum-dried and freeze-dried graphene is called Gr1 and Gr2, respectively.

2. Experimental

2.1 Materials

High-purity graphite flakes were obtained from Nanjing XFNANO Materials Tech Co., Ltd. as well as potassium hexacyanoferrate (III), potassium hexacyanoferrate (II) trihydrate, and potassium chloride (KCl); uric acid was obtained from Sigma (UA, ≥99). All chemicals were of analytical grade and used without further purification. All aqueous solutions were made with deionized water and purified with a Milli-Q system (Millipore).

2.2 Synthesis of graphite oxide (GtO)

GtO was synthesized from graphite flakes via a modified Hummers' method.^2,17,18 In an ice bath, 46 mL of concentrated H₂SO₄ was mixed with 2.0 g of graphite flakes and 0.8 g of KNO₃ with continuous stirring. Then, 6.0 g of KMnO₄ was slowly added under constant stirring for 30 min. A thick paste was formed and 160 mL of deionized water was added with stirring for 30 min; the temperature was increased to 95 °C. Finally, 100 mL of deionized water was added to terminate the reaction. The color of the suspension turned from brown to bright yellow upon addition of 30 mL of H₂O₂ (30%). The filtration was conducted while the suspension was still warm to avoid the precipitation of the slightly soluble salt of mellitic acid that formed as a side reaction.¹ The filter cake was sequentially washed with HCl (100 mL, 0.1 M) and DI water (50 °C) until it was nearly neutral (pH = 6.5). The HCl and DI water were used to remove metal ions and acid, respectively. After cleaning, the samples were separately dried via vacuum drying (60 °C) and vacuum freeze drying.

2.3 Synthesis of graphene (Gr)

The oxidation product was redissolved in deionized water followed by ultrasonic treatment for 4.5 h to achieve full exfoliation from graphite oxide to graphene oxide. The solution was then centrifuged at 10 000 rpm for 10 min. The upper supernatant was separated carefully from the bottom sediment to yield a yellow-brown transparent GO solution (about 1 mg mL⁻¹). Finally, the prepared GO dispersion was further centrifuged at 5000 rpm to remove any unexfoliated GtO.¹⁶

In this work, we employed 25–28% NH₃·H₂O as the reductive agent because the NH₃·H₂O could improve the stability of the Gr and effectively reduce the GO. This decreased its toxicity. The GO dispersion solution was ultrasonically treated for about 30 min before the pH was tuned to 11 with 25–28% NH₃·H₂O before reacting for 16 h at 95 °C. The mixture was cooled to room temperature and was dialyzed until it was near neutral. Finally, a black Gr dispersion solution was obtained through ultrasonic treatment.

2.4 Characterization of samples

The morphology and microstructure of the samples were observed using a scanning electron microscope (SEM; Japan NTC, JSM-6390LV) and transmission electron microscopy (TEM) was performed using a H-7650 (Japan); the accelerating voltage was 200 kV. The bonding properties of the samples were characterized using Fourier transform infrared spectroscopy (FTIR, Nexus 470). Powdered samples were placed in a KBr pellet with a pressure of 2.0 × 10⁷ Pa. The FTIR spectra were collected from 400 to 4000 cm⁻¹. The zeta potentials of GO before and after reduction with NH₄OH were also measured in triplicate using a Malvern Zetasizer (ZetaSizer 3000HSA). X-ray diffraction spectra of the samples were obtained using an X-ray diffractometer (XRD; Melvin, D8 Advance) equipped with a CuKα (λ = 1.5406 Å) source at a wide-angle range of 0–100° (2θ) and a scan speed of 2° min⁻¹. X-ray photoelectron spectroscopy (XPS) measurements were carried out on XPS apparatus (VG Multilab 2000). Raman spectroscopic analysis was performed on an Invia (Renishaw) Raman spectrometer using a 633 nm excitation laser. The Brunauer–Emmett–Teller (BET) specific surface area was detected from nitrogen gas adsorption–desorption isotherms at 77 K using a V-sorb 2800 TP specific surface area and pore size analyzer. Electrical conductivity measurements of the two graphene samples were taken by pressing the graphene powder into a 12 mm disc (diameter) using a hydraulic press. The conductivity of the graphene samples was determined through the measurement of their resistance using a two point probe method where an average of six different sites was taken. The distance between the two probes was about 10 mm.

2.5 Electrochemical measurements

Before each experiment, the GCE was mechanically polished with 0.05 μm alumina powder on microcloth pads until a mirror-finish surface was obtained. The electrode was instantly rinsed thoroughly with deionized water and outright washed successively with anhydrous alcohol and deionized water in an ultrasonic bath. Prior to modification, the electrode surface was washed with distilled water and dried in a N₂ atmosphere.

To the best of our knowledge, the electrochemistry of rGO has widely been studied.² In the electrochemical detection of Gr, cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were performed using an Autolab PGSTAT204 electrochemical analyzer interfaced to a computer system with the corresponding electrochemical software. The analyzer was assembled with a conventional three-electrode system: a glassy carbon working electrode (GCE; 3 mm diameter), a platinum (Pt) wire counter electrode, and a saturated calomel reference electrode (SCE). All electrochemical experiments were detected in 2 mM [Fe(CN)₆]^3−/4− containing 0.3 M KCl as was similar to our previous method at room temperature.

For rotating disk electrode (RDE) cyclic voltammetry studies, a Rotating Disk Electrode Rotator (RRDE-3A) (BAS, Japan) was applied. A GC disk (3.0 mm diameter) or graphene modified working electrode was used, along with an Ag/AgCl electrode as a reference electrode and Pt counter electrodes employed in the solution of 2 mM [Fe(CN)₆]^3−/4− and 0.1 mM uric acid (0.1 M sodium phosphate buffer, pH 7.5). Oxygen reduction reaction (ORR) voltammograms were recorded with a three-electrode system in an oxygen saturated 0.1 M KOH (pH 13) electrolyte solution. The graphene modified on the glassy carbon RDE was used as a working electrode, the Ag/AgCl electrode was used as a reference electrode and a Pt sheet was used as a counter electrode.

In the above methods, the graphene samples were electrodeposited onto the electrode surface through the potential cycled between 0 and −1.4 V vs. Ag/AgCl with a scan rate of 0.05 V s⁻¹ for 10 cycles in an aqueous solution containing 1 mg mL⁻¹ graphene oxide and 0.05 M Na₂HPO₄.

3. Results and discussion

3.1 Morphology of GO and graphene

The morphology of the graphene samples and their oxides prepared via the two different drying methods was investigated using SEM (Fig. 1A, B, E and F). The GO does not present a thin and layered structure, but does show the interlayer structure (Fig. 1A and E) and partial aggregation such as wrinkling and folding (Fig. 1B and F). The phenomenon is due to the oxygen-containing functional groups inserting into the interlayer spacing to reduce the free energy of the slice layers, which increases the layer spacing. Ultrasound increases the presence of the thin-layered structure and decreases reunion.

Fig. 1 SEM-images of (A) GO1, (B) GO2, (C) Gr1, (D) Gr2 (Gr1: vacuum-dried graphene; Gr2: freeze-dried graphene) (E)–(H) are magnifications of (A)–(D), respectively.

Fig. 1C, D, G and H show that the reduced graphene was not flat and consisted of randomly aggregated thin and crumpled sheets closely associated with each other. These data suggest the presence of individual sheets in our reduced GO materials.² We also found that the freeze-dried graphene has a more porous structure than the vacuum-dried structure, which may lead to different electrocatalytic abilities.

As shown in TEM images (Fig. 2), there is a chaotic and folded structure for the graphene sample. The graphene nanosheets are randomly compact and stacked together, showing uniform laminar morphology like crumpled silk veil waves after reduction with ammonia. This morphology may be attributed to defective structures formed upon exfoliation or the presence of doped nitrogen atoms.¹⁹

Fig. 2 TEM images of (A) GO1, (B) GO2, (C) Gr1, (D) Gr2 (Gr1: vacuum-dried graphene; Gr2: freeze-dried graphene).

3.2 FTIR of the GO and graphene samples

As shown in Fig. 3, we investigated the FTIR spectra of graphite, graphene oxide, graphene, and graphite oxide. Fig. 3A and B show that the infrared absorption peaks of graphite are quite gentle. When the graphite was oxidized we observed several characteristic peaks: an O–H stretching vibration at 3430 cm⁻¹ (broad) that may be caused by residual moisture in GO, C=O (carbonyl and carboxyl) stretching vibrations at 1728 cm⁻¹ and C–O (epoxy and alkoxy) stretching peaks at 1228 cm⁻¹ and 1050 cm⁻¹, respectively.²⁰ By reducing the GO, the intensity of the peak of the oxygen-containing groups obviously decreases.

Fig. 3 Total reflectance Fourier transform infrared spectroscopy (FTIR) of (a) graphite, (b) GO, and (c) Gr produced via (A) vacuum drying and (B) vacuum freeze drying.

3.3 Zeta potential of the GO and graphene samples

The dispersibility and stability of the samples were investigated using zeta potential measurements (Fig. 4). It is well known that the repulsive force between particles increases with the increasing absolute value of the zeta potential. The zeta potential was less than −30 mV, and the repulsion between graphene oxide particles increased. This is sufficient to maintain the stability of the dispersed system.²¹ Due to the presence of negatively charged carboxyl and hydroxyl groups, the zeta potential of the graphene oxide is higher than graphene. The zeta potential showed that the effects of electrostatic repulsion cannot be ignored and contribute to the stability and dispersion of GO in aqueous solution. We discovered that the zeta potential of GO2 (−44.1 mV) was much lower than that of GO1 (−28.9 mV). This showed that GO2 had better dispersion and stability. Similarly, the zeta potential of Gr2 was about −32.6 mV, and that of Gr1 was −21.6 mV; obviously the aqueous solution of Gr2 has better dispersion and stability.

Fig. 4 The zeta potential of GO1, GO2, Gr1, and Gr2.

3.4 XRD of the GO and graphene samples

XRD patterns of flake graphite, GO, and Gr are shown in Fig. 5A. The patterns of natural graphite show an intense peak at 2θ = 26.3° (d-spacing = 0.339 nm). This corresponds to the graphitic structure.³ This shows that the graphite has a highly ordered crystal structure. The preparation of GO through a modified Hummers' method shows that the reflection peak disappears entirely at 2θ = 26.3°. This shifts to a relatively weak peak at 2θ = 11.2°. We used the software MDI 6.0 and the Bragg equation to calculate the interlayer spacing of GO1 and GO2. The layer spacing is about 7.6 Å and 7.9 Å, respectively. This is obviously higher than natural graphite. The increase in distance is due to the intercalation of a number of oxygen-containing groups on the edge of each layer.⁴ These oxygen-containing groups allowed water molecules to enter the graphite oxide layers. This also increase the interlayer spacing.

Fig. 5 (A) X-ray diffraction patterns of (a) natural flake graphite, (b) GO, and (c) Gr. The inset shows the XRD of graphene. (B) Different reduction times of Gr at (a) 6 h, (b) 9 h, (c) 12 h and (d) 15 h.

Fig. 5B shows weak XRD data generated near 23° (d-spacing = 0.38) as well as the complete transformation from GO to Gr after 15 h. The broadened peak revealed that the stacking of graphene was poorly ordered.²² The d-spacing is still slightly higher than that of graphite, suggesting that the presence of oxygen-containing groups and crystallization water leads to a decrease in layer spacing.

3.5 XPS of graphene prepared via different drying methods

X-ray photoelectron spectroscopy (XPS) characterizations were performed to analyze the detailed surface elemental compositions of graphene prepared via different drying methods. The survey spectra (Fig. 6) for both the Gr1 and Gr2 samples showed a predominant narrow C 1s peak at 284.6 eV, an O 1s peak at about 532 eV, and a N 1s peak at 400 eV, which correspond with ref. 19. The surface atomic percentages of the Gr1 sample are 73.02% C, 23.5% O, and 3.47% N, while Gr2 contained 78.13% C, 15.95% O, and 5.91% N. XPS spectra for the two drying treated graphene samples clearly show the incorporation of nitrogen atoms within the graphene sheets, which were maybe mainly introduced via the reduction process with ammonia water. The calculated N/C atomic ratio is 0.048 and 0.07. The results indicate that there are N and O heteroatoms in the composition of graphene²³ and Gr2 has more nitrogen compared with Gr1, which may be related with the electrocatalytic activity.

Fig. 6 XPS spectra of (A) Gr1, and (B) Gr2.

3.6 The Raman spectra of the graphene samples

Raman spectroscopy is the most direct and non-destructive technique to characterize the structure and the layers of graphene. Fig. 7 shows the Raman spectra of GO and Gr, which exhibit two significant peaks at approximately 1330 (D band) and 1586 cm⁻¹ (G band) (Fig. 7A). The G band is related to the graphitic components in the structure, while the D band is associated with structural defects and the partially disordered structures of the sp² domains.^20,24 The I_D/I_G values of the GO samples (GO1, 1.01; GO2, 1.06) were less than those of the Gr samples (Gr1, 1.09; Gr2, 1.11); these results were similar to ref. 2. The values of I_D/I_G between Gr1 and Gr2 are similar, the latter was a little larger, which indicates that the graphitic degree of both of the Gr samples was improved due to the superior reduction and the self-repairing of the graphene layer, the average size of the newly formed sp² domains decreasing and the presence of unrepaired defects that remain after the removal of oxygen moieties.^20,25,26

Fig. 7 Raman spectra of (a) GO and (b) Gr prepared via (A) vacuum drying treatment and (B) the freeze drying method.

3.7 The BET surface area of the graphene samples

Typical N₂ adsorption–desorption isotherms for the two graphene samples, Gr1 and Gr2, are shown in Fig. 8. The surface area measurements indicated that the two graphene samples of Gr1 and Gr2 had a BET (nitrogen, 77 K) surface area of 473.8 and 558.4 m² g⁻¹, respectively, in the solid state.

Fig. 8 The nitrogen adsorption and desorption isotherms of (A) vacuum-dried graphene and (B) freeze-dried graphene.

3.8 Electrical conductivity and catalytic activity of graphene

3.8.1 Electrical conductivity of the graphene samples. Electrical conductivity is perhaps the best indicator of the extent to which graphite oxide has been reduced to graphene. The conductivity of the graphene paper was taken by measuring the resistance between two points on the sample. The resistivity was calculated using the equation: where R is resistance, ρ is resistivity, A is the cross-sectional area of the sample in contact with the electrode, and L is the distance between the electrodes. While the conductivity was then determined from the inverse of the resistivity.^27,28 Thus through the calculation, the average conductivity of Gr1 and Gr2 was approximately 534 and 576 S m⁻¹, respectively, so the conductivity of Gr2 was a little higher than that of Gr1.

3.8.2 The optimization of graphene volume. As shown in Fig. 9, we found that upon the addition of Gr1 to the GCE, the volume is 6 μL and the peak current is at a relative maximum. Upon further addition, the current becomes nearly stable. This may be because of the thicker graphene film. This inhibited the transfer of the electric charge on the interface. Similarly, the optimal addition volume of Gr2 is 10 μL. Besides, at the same addition volume, the peak current of Gr2 is much higher than for Gr1. This indicates that Gr2 has better electrocatalytic activity.

Fig. 9 The optimization of graphene volume modified on the glassy carbon electrode ((a) vacuum-dried graphene and (b) freeze-dried graphene).

3.8.3 The electrocatalytic comparison of Gr prepared via two different drying methods. Graphene has a strong conduction ability, and it can promote electron transfer. This results in an increase in the redox current of [Fe(CN)₆]^3−/4− (Fig. 10A). Also after scan 100, the relative errors of the peak current of the Gr1 modified GCE and Gr2 modified GCE were about 5.7% and 3.4%, which demonstrated that Gr2 had a better duration of the catalyst. For EIS images, the semicircle diameter that equals the charge transfer resistance nearly disappeared. This suggests that there is a decrease in R_ct for the Gr-modified GCE versus the bare electrode (Fig. 10B). Gr2 has a more significant ability to transfer charge than Gr1 according to CV. However, this mechanism has not been shown clearly, it may be associated with the different structure of graphene. Fig. 11 shows the typical CVs of the Gr-modified GCE at different scan rates ranging from 30 to 180 mV s⁻¹ including the relationship between the square of the scan rate and the peak current. We found that the square of the scan rate vs. the peak current has a very good linear relationship for both Gr1 and Gr2. This indicates that the reaction was a surface-controlled process.²⁹

Fig. 10 The comparison of the electric catalytic activity of Gr produced via two different drying methods in 2 mM K₃Fe(CN)₆ containing 0.3 M KCl: (A) CV and (B) EIS ((a) bare GCE, (b) Gr1/GCE, and (c) Gr2/GCE).

Fig. 11 CVs of the graphene modified GCEs at different scan rates (from a to f) of 30, 80, 100, 120, 150 and 180 mV s⁻¹ in 2 mM [Fe(CN)₆]^3−/4− containing 0.3 M KCl and the inserts show the linear relation of scan rate versus peak current at Gr/GCE: (A) Gr1 and (B) Gr2.

As shown in Fig. 12A, the values of the half wave potentials (E_1/2) of the voltammograms for the [Fe(CN)₆]^3−/4− process were different between the rotated Gr2 modified electrode and rotated Gr1 modified GC electrode. The slope of the voltammogram in the potential region near E_1/2 for the Gr2 modified GCE is significantly less steep than the Gr1 modified electrode for the [Fe(CN)₆]^3−/4− process. According to the RDE voltammograms for UA oxidation (Fig. 12B), the shapes of the curves were different between two graphene modified electrodes. Graphene prepared electrochemically from graphene oxide contains some functional groups, such as –COO–, which may interact with and stabilize the unstable quinonoid diimine formed upon the oxidation of uric acid. Thus, a sharp oxidation component is present (Fig. 12B-c). However, the presumed adsorption component was not detected when the Gr1 modified electrode was used, even under the same conditions (Fig. 12B-b). Also the slope of the voltammogram near E_1/2 was steeper with the Gr2 modified electrode compared to the Gr1/GCE. This suggested that the Gr2 modified electrode was more active than the Gr1 modified electrode toward the electrooxidation of UA under RDE conditions.³⁰

Fig. 12 RDE voltammograms obtained using 3.0 mm-diameter GC (a), Gr1 modified (b), and Gr2 modified (c) electrodes with a scan rate of 0.01 V s⁻¹ and a rotation rate of 1000 rpm in aqueous solution containing (A) 2 mM [Fe(CN)₆]^3−/4− (0.1 M KCl), and (B) 0.1 mM uric acid (0.1 M sodium phosphate buffer, pH 7.5).

To further investigate the electrocatalytic performance of the graphene, we conducted ORR experiments. Fig. 13 shows the RDE voltammograms of oxygen reduction on two graphene sample modified electrodes with various rotation speeds in O₂-saturated 0.1 mol L⁻¹ KOH solution. The current density increases when the rotation rate increases from 250 to 2000 rpm, which clearly indicates the sensitivity of the ORR activity on the availability of O₂. A steeper slope demonstrated better ORR activity,³¹ so Gr2 has more ORR activity.

Fig. 13 The ORR current response of the graphene ((A) Gr1 and (B) Gr2) under various rotation rates in O₂ saturated 0.1 M KOH aqueous solution.

4. Conclusions

In summary, Gr nanosheets were successfully produced by chemically reducing GO with ammonia water. The use of ammonia water as a reductant provides a relatively safe and cost-effective method to reduce GO and replace the use of toxic chemicals, and produced completely reduced Gr in about 15 h. The SEM and TEM images showed the morphology of the graphene samples, and the FTIR and XRD data clearly demonstrated that graphene was prepared successfully. The zeta potential experiments suggested that the graphene we prepared was very dispersible. XPS was used to analyze the elements in the graphene samples, and the results revealed that nitrogen was introduced into the graphene. The results of the Raman spectra were similar for the differently dried graphene. The BET results showed Gr2 had a larger surface area. The CV and EIS spectra revealed that the graphene had good electrocatalytic activity. For the RDE results in systems of [Fe(CN)₆]^3−/4−, the Gr1 modified GCE showed lower electrocatalytic activity, while in a UA system Gr2 was more active towards the electrooxidation of UA.

Acknowledgements

The authors gratefully acknowledge the financial support of the National Natural Science Foundation of China (Grant No. 31471602).

References

1. J. I. Paredes, S. Villar-Rodil and A. Martinez-Alonso, Langmuir, 2008, 24, 10560.
2. 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.
3. F. Liu, K. S. Choi, T. J. Park, S. Y. Lee and T. S. Seo, BioChip J., 2011, 5, 123.
4. M. Zhou, Y. Zhai and S. Dong, Anal. Chem., 2009, 81, 5603.
5. 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.
6. I. Jung, M. Pelton, P. Richard, D. A. Dikin, S. Stankovich, S. Watcharotone, M. Hausner and R. S. Ruoff, Nano Lett., 2007, 7, 3569.
7. L. Tang, Y. Wang, Y. Li, H. Feng, J. Lu and J. Li, Adv. Funct. Mater., 2009, 19, 2782.
8. K. Parvez, R. Li, S. R. Puniredd, Y. Hernandez, F. Hinkel, S. Wang, X. Feng and K. Mṻllen, ACS Nano, 2013, 7, 3598.
9. C. Berger, Z. Song, X. Li, X. Wu, N. Brown, C. Naud, D. Mayou, T. Li, J. Hass, A. N. Marchenkov, E. H. Conrad, P. N. First and W. A. Heer, Science, 2006, 312, 1191.
10. H. J. Park, J. Meyer, S. Roth and V. Skákalová, Carbon, 2010, 48, 1088.
11. W. Ren, L. Gao, L. Ma and H. Cheng, New Carbon Mater., 2011, 26, 71.
12. N. Zhao, X. Cheng, J. Yang, M. Yang, S. Zheng and Y. Zhou, J. Phys. Chem. Solids, 2014, 75, 1141.
13. K. Parvez, Z. S. Wu, R. Li, X. Liu, R. Graf, X. Feng and K. Mullen, J. Am. Chem. Soc., 2014, 136, 6083.
14. L. Chen, Y. Tang, K. Wang, C. Liu and S. Luo, Electrochem. Commun., 2011, 13, 133.
15. M. Hilder, B. Winther-Jensen, D. Li, M. Forsyth and D. R. MacFarlane, Phys. Chem. Chem. Phys., 2011, 13, 9187.
16. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
17. P. G. Klemens, Theory of the a-plane thermal conductivity of graphite, J. Wide Bandgap Mater., 2000, 7, 332–339.
18. Y. Zhou, H. Dong, L. Liu, J. Liu and M. Xu, Biosens. Bioelectron., 2014, 60, 231.
19. Z. Sheng, S. Lin, J. Chen, W. Bao, F. Wang and X. Xia, ACS Nano, 2011, 5, 4350.
20. D. N. H. Tran, S. Kabiri and D. Losic, Carbon, 2014, 76, 193.
21. D. Li, M. B. Muller, S. Gilje, R. B. Kaner and G. G. Wallace, Nat. Nanotechnol., 2008, 3, 101.
22. F. T. Johra, J.-W. Lee and W.-G. Jung, J. Ind. Eng. Chem., 2014, 20, 2883.
23. L. Wang, F. Yin and C. Yao, Int. J. Hydrogen Energy, 2014, 39, 15913.
24. X. Chen and B. Chen, Environ. Sci. Technol., 2015, 49, 6181.
25. J. Geng and H. T. Jung, J. Phys. Chem. C, 2010, 114, 8227.
26. H. Kim, A. A. Abdala and C. W. Macosko, Macromolecules, 2010, 43, 6515.
27. H. Guo, X. Wang, Q. Qian, F. Wang and X. Xia, ACS Nano, 2009, 3, 2653.
28. M. Choucair, P. Thordarson and J. A. Stride, Nat. Nanotechnol., 2008, 4, 30.
29. J. Hui, W. Li, Y. Guo, Z. Yang, Y. Wang and C. Yu, Bioprocess Biosyst. Eng., 2014, 37, 461.
30. S. X. Guo, S. F. Zhao, A. M. Bond and J. Zhang, Langmuir, 2012, 28, 5275.
31. Z. Luo, S. Lim, Z. Tian, J. Shang, L. Lai, B. MacDonald, C. Fu, Z. Shen, T. Yu and J. Lin, J. Mater. Chem., 2011, 21, 8038.

---