CARBON 4 9 ( 2 0 1 1 ) 2 8 0 9 –2 8 1 6

available at www.sciencedirect.com

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

High-quality few layer graphene produced by electrochemical

intercalation and microwave-assisted expansion of graphite

Gustavo M. Morales a,*, Pablo Schifani a, Gary Ellis b, Carmen Ballesteros c,

Gerardo Martı́nez b, César Barbero a, Horacio J. Salavagione b,*
a
Universidad Nacional de Rı́o Cuarto (UNRC), Departamento de Quı́mica, Ruta 36, Km 601, 5800 Rı́o Cuarto, Argentina
b
Instituto de Ciencia y Tecnologı́a de Polı́meros (ICTP-CSIC), C/Juan de la Cierva 3, 28006 Madrid, Spain
c
Universidad Carlos III de Madrid, Departamento de Fı́sica, Avenida Universidad 30, 28911 Leganés, Spain

A R T I C L E I N F O A B S T R A C T

Article history: Few-layer graphene is synthesized from electrochemically-produced graphite intercalation

Received 28 December 2010 compounds in aqueous perchloric acid. Although anodic intercalation is more efficient in
Accepted 2 March 2011 terms of time, cathodic pre-treatment is preferred to avoid the formation of graphite oxide.
Available online 6 March 2011 The materials are characterized by high resolution transmission electron microscopy and
scanning electron microscopy, UV–visible, infrared and Raman spectroscopy. We demon-
strate that the method, under the experimental conditions used in this work, does not pro-
duce damage to the sp2 carbon lattice. The synthetic approach using electrochemical-
potential control is very promising to obtain, in a controllable manner, graphene with dif-
ferent degrees of oxidation.
 2011 Elsevier Ltd. All rights reserved.

1. Introduction of the properties of graphene still requires the development of

methods to produce large quantities of FLG with low density
Graphene, a one-atom-thick planar sheet of sp2-bonded car- of defects in the crystal lattice [9]. The exfoliation of expand-
bon atoms, is a quasi-2-dimensional (2D) material. The fasci- able graphite (EG) via graphite intercalation compounds (GIC)
nating properties of single-layer graphene (SLG) and few-layer may be one of the most attractive methods proposed because
graphene (FLG) have made it one of the most promising mate- this process, which has been extensively studied in the past,
rials of the first decade of the 21st century [1]. It has been ob- is cheap and scalable [10,11]. Nevertheless, the possibility of
served that graphene presents an ambipolar field effect [2], a its use as a route to obtain FLG without damage to the sp2
quantum Hall effect at room temperature [3–5], very high structure – denominated as high quality graphene, HQG – is
charge carrier mobility that is temperature-independent and still not clear. Stankovich et al. [12] reported numerous failed
translates into ballistic transport [6,7], and very high surface attempts to produce HQG from GIC, including the synthesis
area [8]. Graphene and its derivatives are promising candi- by exfoliation of potassium-intercalated graphite [13].
dates as components for applications in the field of energy- Recently, electrochemical techniques have been employed
storage and energy conversion materials, thermally and elec- in the production of graphene. Guo et al. reported the electro-
trically conductive reinforced nanocomposites, nanoelectron- chemical reduction of exfoliated graphite oxide (GO) in PBS
ics and sensors, among many others. [14] and sodium sulfate [15] solutions. However, this method
However, graphene suffers from a problem which is com- presents the same disadvantages as all synthetic approaches
mon to many novel nanomaterials; the lack of effective meth- where GO is used as starting material; the sp3 defects cannot
ods for large-scale production. Consequently, full exploitation be efficiently transformed back to sp2 as was evident from

* Corresponding authors: Fax: +54 358 4676233 (G.M. Morales); fax: +34 915644853 (H.J. Salavagione).
E-mail addresses: [email protected] (G.M. Morales), [email protected] (H.J. Salavagione).
0008-6223/$ - see front matter  2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbon.2011.03.008
2810 CARBON 4 9 ( 2 0 1 1 ) 2 8 0 9 –2 8 1 6

Raman spectroscopy [14]. Liu et al. synthesized ionic-liquid- The working electrode (WE) was a 1 mm thick laminated
functionalized graphite sheets by applying 15 V between two graphite foil of dimensions 8 mm · 15 mm (SIGRAFLEX, Ger-
graphite rods immersed in a water/ionic–liquid mixture [16], many) , a normal hydrogen electrode (NHE) was employed as
but the nature of the method produced graphene with high de- the reference electrode (RE), and in order to avoid contamina-
fect densities. In addition, Lu et al. reported the synthesis of tion by metals, a piece of large surface area carbon [30] served
fluorescent carbon nanostructures – including graphene – by as the counter electrode (CE). All experiments were conducted
using ionic liquid-assisted electrochemical exfoliation of at room temperature 25 ± 2 C using an AutoLab PGSTAT 12
graphite electrodes, also under oxidative conditions [17]. potentiostat, controlled by GPES 4.9 electrochemical software
The exfoliation of graphite associated to the charging/dis- (Eco-Chemie, Utrecht, The Netherlands). The potentials are
charging process is a well known phenomenon in the field of expressed throughout with respect to the NHE.
lithium-ion batteries using carbonaceous material. In fact, it
is an undesired phenomena occurring during lithium interca- 2.2. Preparation of high quality FLG
lation/de-intercalation, leading to severe battery failure. In
this context, much effort has been invested in engineering The FLG was obtained by an electrochemical treatment fol-
methods to avoid graphite exfoliation [18–20]. Under appro- lowed by microwave-assisted thermal expansion and ultra-
priate experimental conditions, graphite can be electrochem- sonic exfoliation of laminated graphite.
ically oxidized and reduced to give Cþ 
n and Cn lattices which Electrochemical pre-expansion of a graphite electrode: The
can hold anions and cations respectively [21,22]. In addition, WEs were initially characterized by means of cyclic voltam-
when a sufficiently negative (positive) potential is applied to metry (CV), and then subjected to double potential step exper-
a graphite-working electrode in aqueous solution, molecular iments. A potential of 0 V was initially applied for 10 s, so it
hydrogen (O2, CO2 and GO) can be produced simultaneously may be assumed that at zero time no faradaic reactions took
to the intercalation process. To the best of our knowledge, place on the surface of the electrode. A perturbation potential
the electrochemical production and subsequent intercalation was then imposed for the time period t, ranging from 0 to s,
of hydrogen on a graphite electrode coupled with the expan- and then the potential was reset to the original value for
sion and exfoliation assisted by hydrogen gas evolution has 10 s. The value of s was typically 1200 s for cathodic poten-
not been studied to date. Moreover, the electrochemical inter- tials. However, positive potentials lead to electrode exfoliation
calation of anions – in this work, perchlorate – and posterior after ca. 10 min and solids are observed at the bottom of the
exfoliation of graphite in aqueous acid media has been inves- cell. During the perturbations the current was recorded as a
tigated in more detail [23–28]. For example: (a) it has been function of time.
shown that, the perchlorate ion is one of the best intercalat-
ing species at low acid concentrations [23] and (b) the anion 2.2.1. Preparation of EG
intercalation can subsequently damage the sp2 lattice due The second step involves thermal treatment assisted by
to side reactions such as GO and carbon dioxide (CO2) forma- microwave radiation. The electrochemically pre-expanded
tion [24,29]. However, by employing concentrated perchloric electrodes were immediately introduced into a conventional
acid and with a careful selection of the intercalation poten- microwave oven (2.4 GHz) and treated for 5–10 s at a nominal
tial, the contribution of those reactions, and consequently power of 800 W under argon atmosphere.
the creation of defects in the carbon network, can be mini-
mized [26,28]. 2.2.2. Exfoliation of EG
Here we report a simple method to produce FLG that com- Finally, the EG was dispersed into N-methyl-1-pyrrolidone
bines the anodic and cathodic electrochemical intercalation (NMP) by treatment with an ultrasonic finger during five min-
of graphite in aqueous perchloric acid and posterior expan- utes. As has been demonstrated, the ultrasonic treatment is
sion by microwave radiation. This method has several advan- critical since it should be applied during a minimum period
tages: First, strong oxidizing conditions are avoided, thus of time in order to ensure exfoliation but without exceeding
irreversible sp3 defects caused under oxidation are not gener- the time in which irreversible damage is caused to the sp2
ated. Second, complex and expensive organic compounds are network [31].
not necessary. Third, contaminants in the final product – The product was centrifuged at 2000 rpm for 30 min and
hydrogen, protons, perchloric acid, perchlorate and water – the precipitate was discarded to ensure that single or FLG
can be eliminated by simple water-washing and further evap- was obtained.
oration. Fourth, if some degree of functionalization – oxida-
tion – is required, this can be tuned by choosing the 2.3. Characterization of FLG
appropriate potential and time applied during the electro-
chemical treatment. The actual concentration of the FLG solution was calculated
as follows. After centrifugation 50 lL of graphene solution
were transferred to a pre-weighted DSC crucible and dried un-
2. Experimental der vacuum at 50 C for 72 h. The crucible with the carbona-
ceous residue was weighed and the mass of graphene
2.1. Electrochemical measurements calculated by difference.
The EG samples were examined by scanning electron
Electrochemical experiments were performed using a conven- microscopy (SEM) employing a Philips XL30 SEM equipment
tional one-compartment three-electrode electrochemical cell. operating at 25 kV.
 CARBON 4 9 ( 20 1 1 ) 2 8 0 9–28 1 6 2811

High resolution transmission electron microscopy and then either form molecular hydrogen via desorptive
(HRTEM) was carried out in a Philips Tecnai 20F FEG micro- recombination (IV and V), or diffuse into the graphite host lat-
scope operating at 200 kV, equipped with a Scanning Trans- tice [33]
mission Electron Microscopy (STEM) module, with a dark
2H ! H2 ðIIÞ
field high angle annular detector for Z-contrast imaging. Crys-
Cn þ H ! Cn Had ðIIIÞ
tal structure studies were carried out by electron diffraction
Cn Had þ H2 O þ e ! H2 þ OH þ Cn ðIVÞ
pattern simulation, using Fast Fourier Transform (FFT) of
Cn Had þ Cn Had ! H2 þ 2Cn ðVÞ
the HRTEM images. For TEM measurements 5 lL of FLG dis-
persion in NMP were deposited onto the holey copper TEM In a second experiment, the potential of the WE was also
grids by drop-casting and dried under vacuum. stepped to positive potentials where perchlorate ions are
UV/Vis absorption spectra of FLG dissolved in NMP were intercalated into the graphite according to the following reac-
recorded on a Perkin Elmer Lambda 40 spectrophotometer. tion [24]:
The infrared spectra were recorded at 4 cm1 resolution Cn þ A ! Cþ  
nA þe ðVIÞ
using a Perkin Elmer System 2000 FTIR spectrometer coupled
to an i-Series IMAGE Infrared microscope equipped with a Concurrently to the intercalation process, faradaic reactions
cooled HgCaTe detector. The samples were deposited on a such as oxygen evolution and carbon oxidation to form GO
Low-E slide (Keveley Technologies Inc., Chesterland, OH, and CO2 are produced in aqueous acid solutions [21].
USA) and recorded in reflection–absorption mode through a Fig. 1 shows typical chronoamperometric response curves
100 lm aperture. to different potential steps applied to the WEs in 1 M HClO4. In
Confocal Raman measurements were made in the Raman all cases, when a negative (positive) potential step is applied,
Microspectroscopy Laboratory of the Characterization Service the cathodic (anodic) current increases rapidly and then de-
in the Institute of Polymer Science & Technology, CSIC. A Ren- creases for a short period of time to finally increase again
ishaw InVia Reflex Raman system (Renishaw plc., Wotton-un- gradually. The perturbation in the potential applied to the
der-Edge, UK) was used employing a grating spectrometer WE can produce different processes. Let us analyze the catho-
with a Peltier-cooled charge-coupled device (CCD) detector, dic intercalation first. The quick initial increase of the current
coupled to a confocal microscope. All spectra were processed can be ascribed to the double-layer charging of the pristine
using Renishaw WiRE 3.2 software. The Raman scattering was graphite electrode. The subsequent response in current is
excited using an Argon ion laser wavelength of 514.5 nm. The dominated by reaction (I). In this stage of the treatment, the
laser beam was focused on the sample with a 100· micro- current diminishes – clearly visible when 0 ! 1.5 V (Fig. 1,
scope objective (N.A. = 0.85), with a laser power at the sample curve a) and 0 ! +1.7 V (Fig 1, curve e) are applied to the elec-
of 62 mW. trode – according to a Cottrell-like behavior characteristic of
an electrochemical process limited by diffusion. However,
deviations – the current dependence is not linear with t1/2
3. Results and discussion
– become more important as time increases. Therefore pro-
As described before, we produced FLG in three steps: electro- cesses other than faradaic ones should be occurring simulta-
chemical treatment followed by microwave-assisted thermal neously. As discussed above, the molecular hydrogen formed
expansion and ultrasonic exfoliation of laminated graphite. in (I) can diffuse into the graphite producing hydrogen inter-
After the first two steps EG is obtained. The subsequent exfo- calation and consequently expansion of the WE. Indeed, the
liation of this EG gives the final FLG. We carefully studied the
materials produced through the whole process of preparation.
Consider a graphite electrode subjected to a potential step
from a non-faradaic region to a more negative potential. In
the simplest scenario there are two different events that
can contribute to the measured current: (a) charging of the
electric double-layer and (b) faradaic process if the potential
is low enough to produce a reduction on the electrode surface.
Both are strongly dependent on the electrochemically active
area presented by the electrode. In most of the electrochemi-
cal experiments the area of the electrode does not change
throughout the experiment and the current–time response
follows the Cottrell equation [32]. But in an intercalation
experiment, the active area changes dramatically. Consider-
ing the cathodic and anodic intercalation in more detail:
Firstly, the potential of a graphite electrode in contact with
an aqueous acid solution is stepped from 0 V to regions where
atomic hydrogen is produced according to the reaction:
Fig. 1 – Potential-step chronoamperograms of a laminated
H3 Oþ þ e ! H þ H2 O ðIÞ
graphite electrode in 1 M HClO4. The potential was stepped
The atomic hydrogen formed in (I) can recombine producing from 0 to 1.5 V (a), 1 V (b), 0.8 V (c), 0.6 V (d) and
molecular hydrogen (II), or absorb on the carbon surface (III) +1.7 V (e).
2812 CARBON 4 9 ( 2 0 1 1 ) 2 8 0 9 –2 8 1 6

expansion phenomena in the WE can be observed in the

electrochemical cell by the naked eye. The WE expansion in-
creases the electrochemically active area with a simultaneous
rise in the current mainly due to the faradaic reaction (I). This
process becomes more important with time and explains
both the deviation from the Cottrell equation and the final
current rise observed in the chronoamperometric curves. It
should be mentioned that no significant expansion of the
graphite electrode was observed for potentials in the range
of 0 to 0.6 V. The over-potential for the reduction of
hydrogen ions on graphite in acid media obtained from CV
experiments is roughly 0.5 V. Thus it can be concluded that
reaction (I) is necessary for the expansion process, even when
the molecular hydrogen evolution through the Tafel reaction
(V) is the favored step after the formation of atomic hydrogen
in acidic medium [31]. This fact is in good agreement with
previous observations of hydrogen insertion into carbon
Fig. 2 – Raman spectra of (a) laminated graphite and (b) EG
fabric in aqueous H2SO4 [34].
prepared by electrochemical treatment at 1 V with band
Hydrogen evolution plays an important function in the
assignments given.
expansion mechanism. This fact could be explained by a
similar mechanism to the exfoliation of graphite by electro-
chemical decomposition of propylene carbonate [35]. The The morphological characterization of the EG shows the
electrolyte penetrates into graphite openings such as pores typical worm-like shape of the EG with thin layered graphene
and defects. Then, a proton is reduced inside those openings structures (Fig. S2) [12]. Figure S2b indicates the presence of
to hydrogen which can produce hydrogen gas (reactions I and layered graphene structures. All the EGs produced in this
II). The large amount of molecular hydrogen produced in a work display similar features. The stacked graphene lami-
short time induces mechanical stress in the interface of the nates appear wrinkled, which is typical of graphene sheets
carbon/electrolyte solution helping to overcome the van der [38]. These observations suggest that the EG obtained by our
Waals forces between graphene layers. Hydrogen evolution method are similar to those obtained by using other
is more pronounced as the cathodic potential becomes larger. procedures.
In fact, negative potentials greater than 1.7 V result in exfo- This EG has also been characterized by Raman spectros-
liation of graphite particles which are deposited on the bot- copy (Fig. 2). The most important features in the Raman spec-
tom of the cell. In addition, fluctuation in the current tra obtained using a laser excitation of 514.5 nm are the G
observed in the chronoamperograms in Fig. 1a and b are band appearing around 1580 cm1, the second order 2D band
attributed to perturbations of the WE surface produced by at around 2700 cm1 and the disorder-induced D and D 0 bands
the violent hydrogen bubbling. A quantitative interpretation at 1350 cm1 and 1620 cm1, respectively. The second order
of the chronoamperometric curves can be performed using 2D 0 mode can also be observed close to 3250 cm1. The Raman
a model of nucleation and growth plus a moving boundary spectra in Fig. 2 compare the initial graphite and EG prepared
as was recently reported by Levi et al. for the electrochemical using a perturbation potential of 1.0 V. The first-order G-
intercalation of lithium in graphite [36]. While qualitatively mode appears in both samples at 1581 cm1, whereas in the
such model seems to be operative, such analysis is out of D-mode for the EG, FWHM broadening from 35 to 45 cm1
the scope of the present work and will be reported elsewhere. and slight shift in peak frequency from 1350 to 1356 cm1
A similar behavior is observed for positive potential steps, in was observed. Evidence for the D 0 band at around 1622 cm1
agreement with the nucleation and growth mechanism pro- is also present in both spectra. However, the pre-treatment
posed by Bard et al. for the oxidation of graphite [37]. For com- does not appear to have a negative influence on the material
parison purposes, Fig. 1 shows the 0 ! +1.7 V step since integrity. In fact, the ID/IG ratio, which is a measure of the de-
potentials above +2 V produce exfoliation of the graphite fects in the carbon network, remains practically constant dur-
WE in a few minutes, probably due to a combination of GO ing the entire process. From the Raman spectroscopy data, it
formation and CO2 and O2 evolution. can be concluded that the electrochemical treatment and the
The EG obtained after electrochemical intercalation/ thermal expansion are not detrimental to the sp2 structure of
microwave expansion was characterized by SEM and Raman graphite.
spectroscopy. Optical inspection suggests that the volume of FLG have been obtained by ultrasonic treatment of EG in
the EG increases as the potential modulus increases (Support- NMP during five minutes. After centrifugation, well-dispersed
ing Information, Fig. S1). The volume depends on the polarity graphene solutions are obtained (Fig. 3). The concentration of
of the electrode, revealing a higher expansion for positively- graphene produced from negatively polarized electrodes, lies
polarized electrodes (Fig. S1). Nevertheless, in order to pre- in the order of 0.6–0.8 g L1 for samples prepared at 1 V.
vent the formation of oxygenated groups that can lead to However, higher values can be obtained depending on the
sp3 carbon and breaking the sp2 network, the cathodic poten- electrochemical conditions (Table S1). Beyond the experimen-
tials are more appropriate, producing the most favorable re- tal uncertainties and the possibility that some NMP can be
sult when 1 V is applied to the WE. irreversibly adsorbed on the graphene layers, the graphene
 CARBON 4 9 ( 20 1 1 ) 2 8 0 9–28 1 6 2813

Fig. 3 – (a) Stable dispersions of FLG prepared at different applied potentials in NMP (potential values in the figure). The
dispersion of reduced graphite oxide (RGO) in NMP is shown for comparison. (b) and (c) Bright field micrographs of a graphene
flake, in (c) the image is intentionally out of focus to increase the contrast. (d) Z-contrast dark field image, higher intensity
indicates higher thickness. Scale bars: 1 lm.

concentration obtained by our method is somewhat higher the Z-contrast image corresponding to Fig. 3c. Even in the
than the obtained by other procedures (0.01–0.03 g L1) thinner areas of the flakes it was possible to observe the pres-
[39,40] and similar to values recently reported by Coleman ence of several layers of graphene. It should be emphasized
and co-workers (up to 1.2 g L1) employing much longer pro- that, even if they are present in the sample, single layers
cedures such as ultrasonic treatment for more than 400 h are difficult observe in a TEM experiment because of the high
[31]. This level of concentration has only been achieved by tendency to stacking during drying on the TEM grid. Two
reduction of GO whose properties, due to sp3 defects in the examples are shown in Fig. 4. In Fig. 4a, 10 graphite layers
carbon lattice, are very poor compared with defect-free graph- are observed, whereas in Fig. 4b the fast Fourier transform
ene obtained here. In addition, our method produces FLG with (FFT) pattern simulation (inset) indicates the superposition
higher lateral dimensions and better quality (lower degree of of two hexagonal structures rotated 30. Regarding the quality
structural defects) than graphene obtained by exhaustive of the graphene sheets, the majority of the flakes showed
ultrasonic treatment [31], as will be discussed later. intermixing of defaulted and ordered areas [41]. Image filter-
The absorption spectra of the FLG dispersions in NMP are ing of both defaulted and undefaulted areas is shown in
flat and featureless in the visible region (Fig. S3) as expected Fig. 4c.
for quasi two-dimensional materials [39]. In addition, the ob- Even if it were possible, it is not obvious that the defects
served low scattering effect is indicative of a dispersion con- were generated during the expansion–exfoliation procedure.
taining non-aggregated ultrathin nanostructures. Also the In fact, the graphite electrode used as starting material con-
infrared spectrum of FLG, obtained by reflection–absorption tains some structural defects as was observed from the Ra-
mode on a glass slide, shows no features in the carbonyl or man spectrum in Fig. 2. However, it cannot be ignored that
hydroxyl regions (Fig. S4) suggesting that no defects has been the defects may have their origin in different steps of the syn-
created during the preparation procedure. The typical bands thetic (i.e. edge defects produced during the intercalation or
for GO around 3600–3000 cm1 (mOH), 1715 cm1 (mC@O), exfoliation) or the characterization process (i.e. damage
1616 cm1 (bOH) and 1368 cm1 (bOH), are not present and caused by the electron beam during the acquisition of images)
the spectrum of FLG obtained at a polarization of 1 V resem- [42].
bles that of graphite where only the bands of skeletal vibra- Due to the lack in uniformity it is not clear that the inten-
tions of graphitic domains are observed at 1550 cm1 (mC@C) sity of the {2 1 1 0} and {1 1 0 0} reflections can be used in this
and 1025 cm1 (mC–C) (Fig. S4). case as a fingerprint to indicate the presence of graphene
The level of exfoliation, lateral size, thickness and the monolayers. The high resolution image of Fig. 4c is represen-
structure of the material dispersed can be analyzed by TEM tative of the images obtained from the thinner flakes. As a
by dropping a small quantity of the dispersion onto a holey consequence of the non-homogeneous sample, different
carbon grid (Fig. 3). Fig. 3b and c show bright field images of intensities of the spots associated with the structural misori-
the typical flakes observed; the image of Fig. 3c is intention- entations can be observed in the Fourier transform pattern
ally out of focus to show the presence of folded layer and mul- (ftp) diffraction simulation. The thickness of the imaged area
tilayer graphene. The lateral size of the flakes is around a few in Fig. 4c is well below four monolayers, observed at the edge,
micrometers, which is somewhat larger than that reported for as indicated by the difference in intensity obtained at the
exfoliation of graphite in NMP by sonication [31,39]. Fig. 3d is STEM-Z contrast images.
2814 CARBON 4 9 ( 2 0 1 1 ) 2 8 0 9 –2 8 1 6

Fig. 4 – (a) and (b) HRTEM micrographs of stacked flakes: (a) fringes at the edge of a flake, indicating the stacking of 10
monolayers, (b) Image and FFT electron diffraction pattern simulation at the inset, where the superposition of two hexagonal
structures rotated 30 is observed; (c) the fringes indicate a maximum stacking of four monolayers (left). The filtered images
make on the same conditions of two contiguous areas with thickness well below four monolayers are also shown (right): top
correspond to defaulted and bottom to free of defects areas. Scale bars: 5 nm.

In order to obtain more evidence on the average number slides. The ID/IG ratio maintains the values observed prior
of graphene layers and the quality of the samples after exfo- to the exfoliation step, and the low intensity of the D band
liation, Raman spectra of the FLG samples were recorded and the lack of a substantial D 0 band indicate that high qual-
after deposition from solution onto quartz microscope ity FLG is obtained (Fig. 5a), different to that found for graph-

Fig. 5 – (a) Raman spectrum observed from exfoliated 1 V FLG sample deposited on a quartz slide, (b) optical image of a
typical deposit on the quartz slide and false-color map of the zone indicated, and (c) representative spectra from the points
indicated of the 2D region of FLG. Scale bars in b: 2 m.
 CARBON 4 9 ( 20 1 1 ) 2 8 0 9–28 1 6 2815

ene prepared by sonication of graphite over long periods of and Argentine Ministry of Science, Technology and Productive
time [31]. These results are reasonable since the electro- Innovation (FONCYT PICT 04 25521, PAE 04 22771 and PICT 07
chemical pre-treatment diminishes the cohesive forces be- 02214) are grateful acknowledged. H.J. Salavagione thanks
tween graphene laminates, and consequently shorter MICINN by a Ramón y Cajal research position. G.M. Morales
sonication times are required, and less defective materials and C.A. Barbero are permanent research fellows of CONICET.
produced. These results also suggest that the defects viewed TEM work has been carried out at LABMET (Red de Laborato-
in TEM images can be associated to irregularities in the edge rios de la Comunidad de Madrid). The help of M.I. Ortiz with
plane. the TEM is acknowledged.
On the other hand, although the efficiency of intercala-
tion/expansion is higher for positive potentials due to interca-
Appendix A. Supplementary data
lation of more voluminous perchlorate ions (Fig. S1), oxidative
potentials over reasonable times lead to side reactions form- Optical images of the WEs treated at different potentials, UV–
ing GO and CO2 damaging in sp2 network [24,29] . Therefore, visible and FTIR characterization SEM images of EG, and table
cathodic potentials were chosen to preserve the sp2 network of disperssability of FLG as a function of the applied poten-
and the outstanding properties of graphene. To ensure high tials. Supplementary data associated with this article can be
intercalation avoiding the formation of sp3 defects during found, in the online version, at doi:10.1016/j.carbon.
anodic treatment, a lower positive potential or shorter times 2011.03.008.
should be employed. Further study and optimization of the
process is currently underway in our laboratory.
Characteristic spectra of multiple and FLG were observed,
R E F E R E N C E S
as could be determined by the analysis of the second order 2D
peaks arising from the zone boundary phonons [43]. As an
example, Fig. 5b shows an image of a graphene deposit ob-
[1] Geim AK, Novoselov KS. The Rise of Graphene. Nat Mater
served on the quartz slide of the same sample, and a false col- 2007;6(3):183–91.
or image of the baseline to 2680 cm1 intensity, where the [2] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y,
darker squares represent higher intensity. Fig. 5c presents Dubonos SV, et al. Electric field effect in atomically thin
the Raman spectra observed at different points marked on carbon films. Science 2004;306(5696):666–9.
the sample in the high frequency region corresponding to [3] Novoselov KS, Jiang Z, Zhang Y, Morozov SV, Stormer HL,
the 2D modes. Spectra at points A–C correspond to FLG of Zeitler U, et al. Room-temperature quantum Hall effect in
graphene. Science 2007;315(5817):1379.
3–6 layers [44–46] whilst the spectrum at point D coincides
[4] Novoselov KS, McCann E, Morozov SV, Fal’ko VI, Katsnelson
with the thicker area at the centre of the deposit and corre- MI, Zeitler U, et al. Unconventional quantum Hall effect and
sponds more closely to that of graphite. Raman and TEM re- Berry’s phase of 2p bilayer graphene. Nat Phys
sults suggest the formation of polydisperse samples 2006;2(3):177–80.
composed of tri, four, few and multilayer graphene. Obvi- [5] Zhang Y, Tan Y-W, Stormer HL, Kim P. Experimental
ously, the method here reported can be optimized to obtain observation of the quantum Hall effect and Berry’s phase in
graphene. Nature 2005;438(7065):201–4.
less polydisperse products and single layer graphene [47,48].
[6] Novoselov KS, Geim AK, Morozov SV, Jiang D, Katsnelson MI,
However; here our objective was to show that the electro-
Grigorieva IV, et al. Two-dimensional gas of massless Dirac
chemical intercalation of graphite in acid media is an inter- fermions in graphene. Nature 2005;438(7065):197–200.
esting alternative to produce FLG combining large quantity, [7] Bolotin KI, Sikes KJ, Jiang Z, Klima M, Fudenberg G, Hone J,
high quality and short time. et al. Ultrahigh electron mobility in suspended graphene.
Solid State Commun 2008;146(9–10):351–5.
[8] Stoller MD, Park S, Zhu Y, An J, Ruoff RS. Graphene-based
4. Conclusions ultracapacitors. Nano Lett 2008;8(10):3498–502.
[9] Allen MJ, Tung VC, Kaner RB. Honeycomb carbon: a review of
In summary, a new method to produce FLG has been de- graphene. Chem Rev 2010;110(1):132–45.
scribed. The GIC are synthesized in aqueous perchloric acid [10] Chung DDL. Exfoliation of graphite. J Mater Sci
under well-defined electrochemical-potential control. Further 1987;22(12):4190–8.
[11] Chung DDL. Review graphite. J Mater Sci 2002;37(8):1475–89.
thermal expansion and the subsequent exfoliation by ultra-
[12] Stankovich S, Dikin DA, Piner RD, Kohlhaas KA,
sound yield a material composed mainly of FLG. The spectro-
Kleinhammes A, Jia Y, et al. Synthesis of graphene-based
scopic and morphologic characterization of the final material nanosheets via chemical reduction of exfoliated graphite
shows that by selecting the appropriated conditions the syn- oxide. Carbon 2007;45(7):1558–65.
thetic procedure does not produce damage to the sp2 carbon [13] Viculis LM, Mack JJ, Mayer OM, Hahn HT, Kaner RB.
lattice. This work demonstrates the potential of electrochem- Intercalation and exfoliation routes to graphite
istry to produce GIC as precursor to a synthesize FLG with dif- nanoplatelets. J Mater Chem 2005;15(9):974–8.
[14] Guo HL, Wang XF, Qian QY, Wang FB, Xia XH. A green
ferent degrees of chemical modification.
approach to the synthesis of graphene nanosheets. ACS
Nano 2009;3(9):2653–9.
[15] Shao Y, Wang J, Engelhard M, Wang C, Lin Y. Facile and
Acknowledgment
controllable electrochemical reduction of graphene oxide
and its applications. J Mater Chem 2010;20(4):743–8.
Financial support from the Spanish Ministry of Science and [16] Liu N, Luo F, Wu H, Liu Y, Zhang C, Chen J. One-step ionic-
Innovation (MICINN) (MAT2009-09335 and MAT-2007-66181) liquid-assisted electrochemical synthesis of ionic-liquid-
2816 CARBON 4 9 ( 2 0 1 1 ) 2 8 0 9 –2 8 1 6

functionalized graphene sheets directly from graphite. Adv [33] Bockris JOM, Reddy AKN. Modern electrochemistry, vol.
Funct Mater 2008;18(10):1518–25. 2. New York: Plenum Press; 1970.
[17] Lu J, Yang JX, Wang J, Lim A, Wang S, Loh KP. One-pot [34] Jurewicz K, Frackowiak E, Béguin F. Towards the mechanism
synthesis of fluorescent carbon nanoribbons, nanoparticles, of electrochemical hydrogen storage in nanostructured
and graphene by the exfoliation of graphite in ionic liquids. carbon materials. Appl Phys A – Mater 2004;78(7):981–7.
ACS Nano 2009;3(8):2367–75. [35] Winter M, Moeller KC, Besenhard JO. In: Nazri G-A, Pistoa G,
[18] Wu YP, Rahm E, Holze R. Carbon anode materials for lithium editors. Carbonaceous and graphitic anodes in Lithium
ion batteries. J Power Sources 2003;114(2):228–36. batteries: science and technology. Boston, Dordrecht, New
[19] Campana FP, Buqa H, Novák P, Kötz R, Siegenthaler H. In situ York, London: Kluwer Academic Publishers; 2004.
atomic force microscopy study of exfoliation phenomena on [36] Levi MD, Markevich E, Wang C, Aurbach D.
graphite basal planes. Electrochem Commun Chronoamperometric measurements and modeling of
2008;10(10):1590–3. nucleation and growth, and moving boundary stages during
[20] Märkle W, Tran N, Goers D, Spahr ME, Novák P. The influence electrochemical lithiation of graphite electrode. J Electroanal
of electrolyte and graphite type on the PF 6 intercalation Chem 2007;600(1):13–22.
behaviour at high potentials. Carbon 2009;47(11):2727–32. [37] Gewirth AA, Bard AJ. In situ scanning tunneling microscopy
[21] Besenhard JO, Möhwald H, Nickl JJ. Preparation and of the anodic oxidation of highly oriented pyrolytic graphite
characterization of graphite compounds by electrochemical surfaces. J Phys Chem 1988;92(20):5563–6.
techniques. Synth Met 1981;3(3–4):187–94. [38] Meyer JC, Geim AK, Kaysnelson MI, Novoselov KS, Booth TJ,
[22] Noel M, Santhanam R. Electrochemistry of graphite Roth S. The structure of suspended graphene sheets. Nature
intercalation compounds. J. Power Sources 1998;72(1):53–65. 2006;446(7131):60–3.
[23] Beck F, Junge H, Krohn H. Graphite intercalation compounds [39] Hernandez Y, Nicolosi V, Lotya M, Blighe FM, Sun Z, De S,
as positive electrodes in galvanic cells. Electrochim Acta et al. High-yield production of graphene by liquid-phase
1981;26(7):799–809. exfoliation of graphite. Nat Nanotechnol 2008;3(9):563–8.
[24] Alsmeyer DC, McCreery RL. In situ Raman monitoring of [40] Hamilton CE, Lomeda JR, Sun Z, Tour JM, Barron AR. High-
electrochemical graphite intercalation and lattice damage in yield organic dispersions of unfunctionalized graphene.
mild aqueous acids. Anal Chem 1992;64(14):1528–33. Nano Lett 2009;9(10):3460–2.
[25] Goss CA, Brumfield JC, Irene EA, Murray RW. Imaging the [41] Gómez-Navarro C, Meyer JC, Sundaram RS, Chuvilin A,
incipient electrochemical oxidation of highly oriented Kurasch S, Burghard M, et al. Atomic structure of reduced
pyrolytic graphite. Anal Chem 1993;65(10):1378–89. graphene oxide. Nano Lett 2010;10(4):1144–8.
[26] Zhang J, Wang E. STM investigation of HOPG superperiodic [42] Hashimoto A, Suenaga K, Gloter A, Urita K, Iijima S. Direct
features caused by electrochemical pretreatment. J evidence for atomic defects in graphene layers. Nature
Electroanal Chem 1995;399(1–2):83–9. 2004;430(7002):870–3.
[27] Alliata D, Häring P, Haas O, Kötz R, Siegenthaler H. Anion [43] Ferrari AC, Meyer JC, Scardaci V, Casiraghi C, Lazzeri M, Mauri
intercalation into highly oriented pyrolytic graphite studied F, et al. Raman spectrum of graphene and graphene layers.
by electrochemical atomic force microscopy. Electrochem Phys Rev Lett 2006;97(18):187401.
Commun. 1999;1(1):5–9. [44] Ferrari AC. Disorder, electron–phonon coupling, doping and
[28] Schnyder B, Alliata D, Kötz R, Siegenthaler H. Electrochemical nonadiabatic effects. Solid State Commun 2007;143(1–
intercalation of perchlorate ions in HOPG: an SFM/LFM and 2):47–57.
XPS study. Appl Surf Sci 2001;173(3–4):221–32. [45] Dresselhaus MS, Jorio A, Hofmann M, Dresselhaus G, Saito R.
[29] Kötz R, Barbero C, Schnyder B, Haas O. Spectroscopic Perspectives on carbon nanotubes and graphene raman
ellipsometry of carbon electrodes during electrochemical spectroscopy. Nano Lett 2010;10(3):751–8.
activation. Thin Solid Films 1993;233(1–2):69–73. [46] Hao Y, Wang Y, Wang L, Ni Z, Wang Z, Wang R, et al. Probing
[30] Bruno MM, Cotella NG, Miras MC, Koch T, Seidler S, Barbero C. layer number and stacking order of few-layer graphene by
Characterization of monolithic porous carbon prepared from Raman spectroscopy. Small 2010;6(2):195–200.
resorcinol/formaldehyde gels with cationic surfactant. [47] Green AA, Hersam MC. Solution phase production of
Colloid Surf A 2010;358(1–3):13–20. graphene with controlled thickness via density
[31] Khan U, O’Neill A, Lotya M, De S, Coleman JN. High- differentiation. Nano Lett 2009;9(12):4031–6.
concentration solvent exfoliation of graphene. Small [48] Green AA, Hersam MC. Emerging methods for producing
2010;6(7):864–71. monodisperse graphene dispersions. J Phys Chem Lett
[32] Bard AJ, Faulkner LR. Electrochemical methods: 2010;1(2):544–9.
fundamentals and applications. 2nd ed. New York: Wiley;
2001.