High-concentration aqueous dispersions of graphene produced by exfoliation of graphite using cellulose nanocrystals

CARBON

7 0 ( 2 0 1 4 ) 1 5 7 –1 6 3

Available at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/carbon

High-concentration aqueous dispersions of graphene produced by exfoliation of graphite using cellulose nanocrystals Pedro M. Carrasco a, Sarah Montes a, Ignacio Garcı´a a, Maryam Borghei b, Hua Jiang b, Ibon Odriozola a, Germa´n Caban˜ero a, Virginia Ruiz a,* a b

New Materials Department, IK4-CIDETEC, Paseo Miramo´n 196, E-20009 San Sebastia´n, Spain Department of Applied Physics, Aalto University, P.O. Box 15100, FI-00076 Aalto, Finland

A R T I C L E I N F O

A B S T R A C T

Article history:

Stable high-concentration aqueous dispersions (>1 mg ml1) of single and few-layer

Received 18 September 2013

graphene flakes were produced by direct exfoliation of graphite using cellulose nanocrys-

Accepted 27 December 2013

tals (CNC). Biodegradable and widely available from renewable sources, CNC have proven

Available online 7 January 2014

to be very efficient graphene stabilizers even at low concentrations (0.2 mg ml1), thus enabling remarkably high graphene/CNC ratios (up to 3.8).  2014 Elsevier Ltd. All rights reserved.

1.

Introduction

Producing processable graphene flakes in large quantities remains an ongoing challenge for large-scale applications. To that end, top-down colloidal approaches such as the exfoliation of powdered graphite in the liquid phase is a very promising route due to its simplicity, utilization of low-cost and readily available graphite, high-throughput potential, possibility of integration with other processes such as blending or casting and no need of transferring processes from the growth substrate [1,2]. The most common route for liquidphase exfoliation (LPE) involving graphite functionalization is graphite oxidation followed by exfoliation to yield graphene oxide, which is subsequently reduced to obtain reduced graphene oxide whose properties differ from pristine graphene. On the other hand, direct exfoliation of unfunctionalized graphite by sonication and good dispersion of resulting graphene sheets require the use of stabilizers that interact noncovalently with graphene and prevent their stacking. Various stabilizers have been proposed both in organic and aqueous media such as ionic and non-ionic surfactants [3–6], polymers [7,8], organic salts [9] and aromatic molecules [10–13]. Direct exfoliation and stabilization in aromatic

solvents [14] or ionic liquids [15] has also been demonstrated yet it suffers from low throughput and poor reproducibility. Despite its capability for large-scale processing, in many cases the resulting dispersions of graphene from LPE are at low concentrations (in the order of 0.1 mg ml1) or require long-lasting sonication times (up to several days). Therefore, considerable effort has been recently focused on developing more efficient exfoliation routes capable of yielding stable high concentration graphene dispersions with minimal amount of stabilizer to preserve graphene properties [16,17]. In this regard, attaining high exfoliation degrees by means of exfoliating agents and stabilizers that are widely available, low-cost, environmentally-friendly and biodegradable would definitely place LPE one step closer towards industrial viability. To this end, we demonstrate here that cellulosic nanomaterials, specifically cellulose nanocrystals (CNC) are very efficient for graphite exfoliation by tip sonication and stabilization of resulting graphene flakes in aqueous dispersions at high concentrations. CNCs can be regarded as very promising graphene stabilizers due to their interesting features such as low density, high surface area, good mechanical properties, biodegradability and availability from renewable

* Corresponding author. E-mail address: [email protected] (V. Ruiz). 0008-6223/$ - see front matter  2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.carbon.2013.12.086

158

CARBON

7 0 ( 2 0 1 4 ) 1 5 7 –1 6 3

resources, which has triggered rising interest during the past decade in these nanocellulose materials especially for nanoreinforcement and barrier property improvement in polymer matrices [18]. Moreover, CNC can be obtained from cellulose, the most abundant biomass material in nature, which can be extracted from natural fibers, among other sources [19]. Furthermore, CNC-assisted LPE of graphite yields graphene flakes decorated with the CNC stabilizer. Such composite systems based on graphene/graphene oxide and cellulose nanomaterials have received tremendous recent interest due to their great promise for different applications such as electrode materials for supercapacitors [20] and composites with enhanced electrical [21], mechanical [22] and thermal insulation [23] properties.

2.

Experimental

2.1.

Materials

Microcrystalline cellulose, sodium dodecyl sulfate and sulfuric acid were purchased from Sigma–Aldrich, Triton X-100 from Panreac and high purity graphite powder (SP-1) was obtained from Bay Carbon Inc.

2.2.

Methods

Scanning electron microscopy was carried out with a Carl Zeiss Ultra Plus field emission scanning electron microscope (FE-SEM) equipped with an energy dispersive X-ray spectrometer (EDXS). Transmission Electron microscopy (TEM) was performed using Tecnai 12 BioTwin with LaB6 gun at 120 kV. High-resolution transmission electron microscopy (HRTEM) and electron diffraction measurements were performed using a double-aberration corrected JEOL 2200FS (JEOL, Japan) microscope equipped with a field emission gun (FEG) operated at 80 kV. A Gatan 4 · 4 k UltraScan 4000 CCD camera was employed for digital recording of HRTEM images. Gatan Digitalmicrograph software was used for camera control and image processing. AFM images of graphene samples on mica substrates were obtained in a tapping mode at room temperature using a scanning probe microscope (Molecular Imagings PicoScan) equipped with a Nanosensors tips/cantilever, a resonance frequency of 330 kHz, and a spring constant of about 42 N/m with a tip nominal radius lower than 7 nm. UV–Vis absorption spectra of graphene dispersions were measured with a UV/VIS/NIR V-570 JASCO spectrophotometer. Raman analysis of graphene dispersions was performed with a Confocal Raman Voyage (BWTEK) spectrometer using a 532 nm excitation laser beam with a power of 5 mWand a 20· objective. For electrical conductivity measurements, graphene powders (collected after water evaporation from the dispersions) were pressed into 1.6 cm diameter disks. Conductivity was determined by a standard four-probe technique at room temperature with a Keithley (model 2400).

3.

Results and discussion

CNC were extracted from microcrystalline cellulose (MCC) by acidic hydrolysis with sulfuric acid (Fig. 1) according to a

protocol reported elsewhere [24]. Typically, MCC powder (20 g) was added to a preheated solution of 64% w/w sulfuric acid (175 ml) and reacted with vigorous stirring at 45 C for 45 min. To quench the reaction, the reaction mixture was diluted with 5 l of deionized water and allowed to settle over night. In order to concentrate the CNC and remove excess acid and water, the suspension was centrifuged at 3500 rpm for 30 min. The precipitate was repeatedly rinsed and centrifuged with deionized water until the supernatant became turbid. Further CNC purification was done by dialysis against deionized water until reaching a pH 5–6 (membrane Spectra/ Por 2, MWCO 12,000–14,000). Finally, the nanocrystals were freeze-dried and stored until use. CNC length and width, estimated by AFM, was 199 ± 40 and 11 ± 2 nm, respectively (Fig. 1). Graphite exfoliation was intentionally carried out applying short sonication times (less than 4 h) as previous work has shown that long sonication times afford higher graphene concentrations but at the expense of graphene quality and flake size [3,25]. Also, different graphite/CNC concentration ratios were investigated in order to maximize graphene/stabilizer concentration ratio in the resulting dispersions. Thus, in a typical procedure, CNC were dispersed in Millipore water at various concentrations between 0.2 and 4 mg ml1 in a sonication bath. Graphene dispersions were prepared by adding graphite (20–100 mg) at 4 mg ml1 and subject to ultrasonication (tip sonicator Hielscher UP400S, 0.5 s pulse, 70% amplitude) in an ice bath to avoid overheating. Unexfoliated graphite was removed by centrifugation at 4500 rpm for different times until no more precipitate was noted (typically 2–3 h). The resulting dark black dispersions are very stable, with negligible levels of sedimentation over periods of months (Fig. 2, inset). The tremendous influence of the experimental procedure (sonication time, centrifugation speed and time, stabilizer concentration, etc) on dispersed graphene concentration and quality, pointed out by several authors, renders difficult

Fig. 1 – Scheme illustrating the extraction of cellulose nanocrystals from cellulose microfibrils and AFM of obtained CNC. (A color version of this figure can be viewed online.)

CARBON

7 0 (2 0 1 4) 1 5 7–16 3

Fig. 2 – UV–Vis spectra of graphene dispersions stabilized by CNC and Triton X-100. Inset: photographs of CNC-stabilized graphene dispersions (0.9 mg ml1) as-produced and 4 months later. (A color version of this figure can be viewed online.)

to benchmark the exfoliation efficiency of CNC against reported results for other aqueous stabilizers. Hence, for comparative purposes, we carried out control experiments under the same exfoliation and centrifugation conditions with a widely used and studied aqueous surfactant, Triton X-100 [4,17]. UV–Vis absorption spectra for Triton X-100 and CNC-stabilized graphene dispersions obtained under identical exfoliation conditions (Fig. 2) were comparable, both featuring a band centered at kmax = 269 nm (that can be attributed to p ! p* transitions of aromatic C–C bonds), in agreement with kmax reported for reduced graphene oxide dispersions [26]. The exfoliation of graphite was further supported by Raman spectroscopy. Fig. 3 compares Raman spectra of starting graphite and graphene produced using both CNC and Triton X-100. Graphite exhibits its characteristic spectrum with a clear G (1581 cm1) and 2D band (2715 cm1) [27].

Fig. 3 – Raman spectra of starting graphite and graphene dispersions stabilized by CNC and Triton X-100.

159

A downshift of the 2D band to lower frequency (23 cm1) was noted in graphene samples together with an intensity enhancement compared to that for graphite, which are indicative of the few-layer nature of CNC-stabilized graphene in the dispersions, in agreement with previous reports [11,12,16,28]. A notable increase of the D band intensity is observed in graphene samples, indicating that the number of defects and edges in the lattice increase during exfoliation. The background intensity is due to the presence of CNC or Triton X-100 on the graphene surface. Graphite exfoliation was carried out using different stabilizer concentrations and starting graphite/stabilizer concentration ratios in order to achieve highly concentrated graphene dispersions with minimal stabilizer concentration, thus preserving the mechanical, thermal and electrical properties of graphene flakes. Table 1 summarizes concentration of graphene flakes and graphene/CNC concentration ratio together with exfoliation yield (expressed as wt.% of the starting graphite) obtained for different CNC concentrations. The concentration of graphene flakes was estimated by weighing the amount of solid dispersed after evaporation of water (taking into account the proportion of graphitic content given by TGA analysis) for known volumes of dispersions. With these concentration values, the extinction coefficient at 660 nm (a660) was estimated from the absorbance using Beer–Lambert law, obtaining values a660 = 1406 ml mg1 m1 in agreement with previous reports for surfactant-stabilized aqueous graphene dispersions [4,6]. Overall, remarkably high graphene flake concentrations (>1 mg ml1) and exfoliation yields were obtained with relatively low CNC concentration, which affords exceptionally high graphene/CNC ratios (1.3–3.8), one order of magnitude larger than typical ratios for conventional stabilizers such as surfactants, polymers and aromatic molecules [16]. The proportion of CNC in produced graphene flakes was estimated from the weight loss in the temperature range 130–575 C measured by TGA. Both graphene concentration and exfoliation yield increase with the amount of CNC (up to a concentration of 2 mg ml1 of CNC) yet at the expense of decreasing graphene/CNC ratio in the dispersion. The higher content of CNC in graphene obtained with large feeding CNC concentrations was also corroborated by EDX measurements, which detected the presence of oxygen from sulfonated CNC. Thus, oxygen content in CNC-stabilized graphene increased from 15 ± 4 to 50 ± 2% wt. for feeding CNC concentrations of 0.2 and 4 mg ml1, respectively. Further increase of stabilizer concentration did not improve the exfoliation yield. Again, as exfoliation efficiency is strongly influenced by sonication time and centrifugation conditions, direct comparison with previously reported results could be meaningless. Therefore, we benchmarked CNC exfoliation and stabilization efficiency against results obtained for Triton X-100 under identical experimental conditions in the same concentration range (0.2–2 mg ml1). With exfoliation yields for Triton X-100 lower than 10% and graphene concentrations between 0.1 and 0.3 mg ml1 regardless of Triton X-100 concentration, CNC afford 4-fold enhancements in exfoliation yield and dispersed graphene concentration. This improved exfoliation and stabilization efficiency for CNC does not entail poorer graphene quality, as both CNC

160

CARBON

7 0 ( 2 0 1 4 ) 1 5 7 –1 6 3

Table 1 – Effect of the stabilizer concentration and starting graphite-to-stabilizer concentration ratio on the resulting concentration of graphene flakes (GF), final GF-to-stabilizer concentration ratio and exfoliation yield. Starting graphite:CNC

Feeding CNC conc. (mg ml1)

GF conc. (mg ml1)

GF conc./CNC conc.ª

Exfoliation yieldb (wt.%)

20:1 10:1 4:1 2:1 1:1

0.2 0.4 1 2 4

0.32 0.59 0.91 1.08 0.99

3.8 2.8 1.3 0.8 0.5

8.6 16.5 24.8 28.1 24.2

a b

CNC concentration in exfoliated GF determined by TGA. Calculated as the percentage of graphene obtained from the starting graphite.

Fig. 4 – Low magnification TEM images of graphene sheets produced at different graphite:CNC concentration ratios: (a) 10:1, (b) 1:1. (c, d) HRTEM images of graphene sheets produced at 10:1 graphite:CNC ratio: (c) edge image of a 5-layer sheet and corresponding FFT pattern (inset), (d) region displaying the hexagonal lattice and corresponding SAED pattern (inset). (A color version of this figure can be viewed online.)

and Triton X-100 stabilized graphene dispersions exhibit very similar amount of defects (ID/IG) in the Raman spectra (Fig. 3). The starting graphite/CNC ratio and consequently CNC content in the resulting CNC-stabilized graphene has a notable influence in the electrical conductivity of the samples, with values ranging from 103 S m1 for the highest graphite:CNC ratio (20:1) down to 101 S m1 for the lowest graphite:CNC ratio (1:1). That is, at low CNC concentrations highly conductive samples can be produced, exhibiting conductivity comparable to that of graphene obtained by surfactant-free LPE in NMP [29] and 10-fold that reported for reduced graphene oxide [30]. In the other limit, graphene produced at larger CNC feeding concentrations contains higher amount of CNC,

which makes it less suitable for applications demanding high conductivity. However, graphene with higher CNC content holds great promise for applications benefitting from the synergistic effect of both nanomaterials such as nanoreinforcement in composites for enhanced mechanical, thermal and barrier properties, highly porous and conducting membranes, paper and aerogels, or electrode material for energy applications. Figs. 4 and 5 show representative FESEM, TEM and HRTEM images of exfoliated samples obtained with different CNC concentrations. Overall, TEM, AFM and FESEM reveal the presence of sheets with lateral dimensions ranging from few hundred nanometers up to several micrometers. In samples

CARBON

7 0 (2 0 1 4) 1 5 7–16 3

161

Fig. 5 – FESEM (a, b) and AFM (c, d) images of graphene flakes produced at 10:1 graphite:CNC concentration ratio. Height profiles taken along the lines are shown in the insets. (e) Statistical analysis of graphene flake thickness from AFM measurements (N = 100 flakes). (A color version of this figure can be viewed online.)

produced at high CNC concentration (4 mg ml1, 1:1 starting graphite:CNC concentration ratio), CNC are clearly visible as small whiskers and dots covering the surface of the graphene flakes (Figs. 4b and 5b). HRTEM images illustrate the atomic ordering of carbon atoms (Fig. 4d), which was further confirmed by the corresponding selected area electron diffraction (SAED) patterns (Fig. 4d, inset). SAED shows the typical sixfold symmetry expected for graphene [31], demonstrating the high crystallinity of graphene sheets. Additionally, the fast Fourier transform (FFT) of topography near an edge of a 5-layer graphene flake (Fig. 4c, inset) also exhibited the hexagonal

spot patterns characteristic of graphene [32]. Thickness distribution of graphene flakes was investigated by AFM (Fig. 5e). Average thickness was 0.9 ± 0.2 nm for 75% of measured flakes, which corresponds to single-layer graphene, with the rest of measured flakes being <3.5 nm thick (fewlayer graphene). In general, flakes with smaller lateral dimensions (few hundred nm) are typically single layer whereas larger flakes (in the lm range) tend to be thicker, consisting of few-layer graphene (Fig. 5c and d). The mechanism of stabilization by CNC most likely involves electrostatic repulsion forces between the sulfate

162

CARBON

7 0 ( 2 0 1 4 ) 1 5 7 –1 6 3

groups in the CNC (38 wt.% determined by TGA) originating from the hydrolysis extraction process from MCC with sulfuric acid. Higher stabilization efficiency has also been noted among a series of pyrene derivatives for those containing sulfate moieties [16]. However, under the same preparative conditions (sonication and centrifugation) a sulfate containing surfactant such as sodium dodecyl sulfate (SDS) exhibits lower exfoliation efficiency than CNCs, with SDS-stabilized graphene concentrations being around one third those achieved for CNCs (Table 1), in agreement with concentrations previously reported for SDS stabilizer [4]. Therefore, the stabilizing influence of sulfate groups in CNC was corroborated by carrying out the exfoliation in presence of CNC with less sulfate content, obtained by neutralizing the sulfate groups according to a previously reported method [33]. The percentage of sulfate groups in CNC was estimated from the weight loss in the temperature range 70–260 C measured by TGA. Indeed, under the same preparative conditions and stabilizer concentration, 20–50% lower exfoliation yields were obtained using as stabilizers CNC with 8 wt.% sulfate content. In order to gain further insight into the exfoliation and stabilization mechanism, we are currently exploring a family of polysaccharides with identical polymer backbone as CNC but different lateral functionalities. Understanding better the influence of the stabilizer structure on its exfoliation efficiency will help us to propose more efficient and high-yield LPE routes.

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

4.

Conclusions

In conclusion, we have demonstrated a low-cost and scalable process to prepare very stable and highly concentrated (>1 mg ml1) aqueous graphene dispersions using low concentrations of cellulose nanocrystals for graphite exfoliation by tip sonication and stabilization of resulting graphene flakes. With a low density, high surface area, good mechanical properties, biodegradability and availability from abundant renewable resources, CNCs can be regarded as very promising graphene stabilizers. Therefore, the proposed methodology, a truly green process, extends the scope for scalable highthroughput liquid-phase production and processing of graphene, which will promote the applications of graphene.

[12]

[13]

[14]

[15]

[16]

Acknowledgments Financial support from the Spanish MICINN (V.R. Ramo´n y Cajal Program), the Basque Government (Etortek Program, nanoIKER project IE11-304) and the European Commission (FP7 Program, ECLIPSE project FP7-NMP-280786) is gratefully acknowledged.

R E F E R E N C E S

[17]

[18]

[19] [20]

[1] Hernandez Y, Nicolosi V, Lotya M, Blighe FM, Sun Z, De S, et al. High-yield production of graphene by liquid-phase exfoliation of graphite. Nat Nanotechnol 2008;3:563–8. [2] Du W, Jiang X, Zhu L. From graphite to graphene: direct liquid-phase exfoliation of graphite to produce single- and

[21]

few-layered pristine graphene. J Mater Chem A 2013;1:10592–606. Lotya M, King PJ, Khan U, De S, Coleman JN. Highconcentration, surfactant-stabilized graphene dispersions. ACS Nano 2010;4:3155–62. Guardia L, Ferna´ndez-Merino M, Paredes J, Solı´s-Ferna´ndez P, Villar-Rodil S, Martı´nez-Alonso A, et al. High-throughput production of pristine graphene in an aqueous dispersion assisted by non-ionic surfactants. Carbon 2011;49:1653–62. Vadukumpully S, Paul J, Valiyaveettil S. Cationic surfactant mediated exfoliation of graphite into graphene flakes. Carbon 2009;47:3288–94. Lotya M, Hernandez Y, King PJ, Smith RJ, Nicolosi V, Karlsson LS, et al. Liquid phase production of graphene by exfoliation of graphite in surfactant/water solutions. J Am Chem Soc 2009;131:3611–20. Wajid AS, Das S, Irin F, Ahmed H, Shelburne JL, Parviz D, et al. Polymer-stabilized graphene dispersions at high concentrations in organic solvents for composite production. Carbon 2012;50:526–34. Bourlinos AB, Georgakilas V, Zboril R, Steriotis TA, Stubos AK, Trapalis C. Aqueous-phase exfoliation of graphite in the presence of polyvinylpyrrolidone for the production of watersoluble graphenes. Solid State Commun 2009;149:2172–6. Du W, Lu J, Sun P, Zhu Y, Jiang X. Organic salt-assisted liquidphase exfoliation of graphite to produce high-quality graphene. Chem Phys Lett 2013;568–569:198–201. Hao R, Qian W, Zhang L, Hou Y. Aqueous dispersions of TCNQ-anion-stabilized graphene sheets. Chem Commun 2008;48:6576–8. Das S, Irin F, Ahmed H, Cortinas AB, Wajid AS, Parviz D, et al. Non-covalent functionalization of pristine few-layer graphene using triphenylene derivatives for conductive poly (vinyl alcohol) composites. Polymer 2012;53:2485–94. Geng J, Kong B-S, Yanga S, Jung H-T. Preparation of graphene relying on porphyrin exfoliation of graphite. Chem Commun 2010;46:5091–3. Zhang M, Parajuli R, Mastrogiovanni D, Dai B, Lo P, Cheung W, et al. Production of graphene sheets by direct dispersion with aromatic healing agents. Small 2010;6:1100–7. Bourlinos AB, Georgakilas V, Zboril R, Steriotis TA, Stubos AK. Liquid-phase exfoliation of graphite towards solubilized graphenes. Small 2009;5:1841–5. Wang X, Fulvio P, Baker G, Veith G, Unocic R, Mahurin S, et al. Direct exfoliation of natural graphite into micrometre size few layers graphene sheets using ionic liquids. Chem Commun 2010;46:4487–9. Parviz D, Sriya Das, Tanvir Ahmed HS, Fahmida Irin, Sanjoy Bhattacharia, Green MJ. Dispersions of non-covalently functionalized graphene with minimal stabilizer. ACS Nano 2012;6:8857–67. Sun Z, Masa J, Liu Z, Schuhmann W, Muhler M. Highly concentrated aqueous dispersions of graphene exfoliated by sodium taurodeoxycholate: dispersion behavior and potential application as a catalyst support for the oxygenreduction reaction. Chem Eur J 2012;18(22):6972–8. Siqueira G, Bras J, Dufresne A. Cellulosic bionanocomposites: a review of preparation, properties and applications. Polymers 2010;2(4):728–65. Eichhorn SJ. Cellulose nanowhiskers: promising materials for advanced applications. Soft Matter 2011;7:303–15. Ouyang W, Sun J, Memon J, Wang C, Geng J, Huang Y. Scalable preparation of three-dimensional porous structures of reduced graphene oxide/cellulose composites and their application in supercapacitors. Carbon 2013;62:501–9. Gao K, Shao Z, Wu X, Wang X, Li J, Zhang Y, et al. Cellulose nanofibers/reduced graphene oxide flexible transparent conductive paper. Carbohydr Polym 2013;97:243–51.

CARBON

7 0 (2 0 1 4) 1 5 7–16 3

[22] Zhang X, Liu X, Zheng W, Zhu J. Regenerated cellulose/ graphene nanocomposite films prepared in DMAC/LiCl solution. Carbohydr Polym 2012;88:26–30. [23] Javadi A, Zheng Q, Payen F, Javadi A, Altin Y, Cai Z, et al. Polyvinyl alcohol-cellulose nanofibrils-graphene oxide hybrid organic aerogels. ACS Appl Mater Interfaces 2013;5:5969–75. [24] Cranston ED, Gray DG. Morphological and optical characterization of polyelectrolyte multilayers incorporating nanocrystalline cellulose. Biomacromolecules 2006;7:2522–30. [25] Li D, Mu¨ller MB, Gilje S, Kaner RB, Wallace GG. Processable aqueous dispersions of graphene nanosheets. Nature Nanotechnol 2008;3:101–5. [26] Khan U, O’Neill A, Lotya M, De S, Coleman JN. Highconcentration solvent exfoliation of graphene. Small 2010;6:864–71. [27] Park K, Kim B, Song S, Kwon J, Kong B, Kang K, et al. Exfoliation of non-oxidized graphene flakes for scalable conductive film. Nano Lett 2012;12(6):2871–6.

163

[28] Ferrari AC. Raman spectroscopy of graphene and graphite: disorder, electron–phonon coupling, doping and nonadiabatic effects. Solid State Commun 2007;143:47–57. [29] Zhang X, Coleman A, Katsonis N, Browne W, van Wees B, Feringa B. Dispersion of graphene in ethanol using a simple solvent exchange method. Chem Commun 2010;46:7539–41. [30] Park S, Ruoff R. Chemical methods for the production of graphenes. Nat Nanotechnol 2009;4:217–24. [31] Meyer J, Geim A, Katsnelson M, Novoselov K, Booth T, Roth S. The structure of suspended graphene sheets. Nature 2007;446:60–3. [32] Campos-Delgado J, Romo-Herrera J, Jia X, Cullen D, Muramatsu H, Kim Y, et al. Bulk production of a new form of sp2 carbon: crystalline graphene nanoribbons. Nano Lett 2008;8(9):2773–8. [33] Wang N, Ding E, Cheng R. Thermal degradation behaviors of spherical cellulose nanocrystals with sulfate groups. Polymer 2007;48:3486–93.