Liquid Phase Production of Graphene by Exfoliation of
Graphite in Surfactant/Water Solutions
Mustafa Lotya,† Yenny Hernandez,† Paul J. King,† Ronan J. Smith,†
Valeria Nicolosi,‡ Lisa S. Karlsson,‡ Fiona M. Blighe,† Sukanta De,†,§
Zhiming Wang,† I. T. McGovern,† Georg S. Duesberg,§,| and
Jonathan N. Coleman*,†,§
School of Physics, Trinity College Dublin, Dublin 2, Ireland, Department of Materials,
UniVersity of Oxford, Parks Road, Oxford OX1 3PH, United Kingdom, CRANN, Trinity College
Dublin, Dublin 2, Ireland, and School of Chemistry, Trinity College Dublin, Dublin 2, Ireland
Received September 29, 2008; E-mail: colemaj@tcd.ie
Abstract: We have demonstrated a method to disperse and exfoliate graphite to give graphene suspended
in water-surfactant solutions. Optical characterization of these suspensions allowed the partial optimization
of the dispersion process. Transmission electron microscopy showed the dispersed phase to consist of
small graphitic flakes. More than 40% of these flakes had <5 layers with ∼3% of flakes consisting of
monolayers. Atomic resolution transmission electron microscopy shows the monolayers to be generally
free of defects. The dispersed graphitic flakes are stabilized against reaggregation by Coulomb repulsion
due to the adsorbed surfactant. We use DLVO and Hamaker theory to describe this stabilization. However,
the larger flakes tend to sediment out over ∼6 weeks, leaving only small flakes dispersed. It is possible to
form thin films by vacuum filtration of these dispersions. Raman and IR spectroscopic analysis of these
films suggests the flakes to be largely free of defects and oxides, although X-ray photoelectron spectroscopy
shows evidence of a small oxide population. Individual graphene flakes can be deposited onto mica by
spray coating, allowing statistical analysis of flake size and thickness. Vacuum filtered films are reasonably
conductive and are semitransparent. Further improvements may result in the development of cheap
transparent conductors.
1. Introduction
The discovery of monolayer graphene in 20041 has led to
the demonstration of a host of novel physical properties in this
most exciting of nanomaterials.2 Graphene is generally made
by micromechanical cleavage, a process whereby monolayers
are peeled from graphite crystals. However, this process has
significant disadvantages in terms of yield and throughput. As
such, there has been significant interest in the development of
a large-scale production method for graphene. In the long term,
for many research areas the growth of graphene monolayers3-5
is by far the most desirable route. However, progress has been
slow, and, in any case, this technique will be unsuitable for
certain applications. Thus, in the medium term, the most
promising route is the exfoliation of graphite in the liquid phase
to give graphene-like materials. The most common technique
has been the oxidation and subsequent exfoliation of graphite
to give graphene oxide.6-10 However, this technique suffers
from one significant disadvantage; the oxidation process results
in the formation of structural defects as evidenced by Raman
spectroscopy.6,9 These defects alter the electronic structure of
graphene so much as to render it semiconducting.11 These
defects are virtually impossible to remove completely; even after
annealing at 1100 °C, residual CdO and C-O bonds are
observed by X-ray photoelectron spectroscopy.10 Even relatively
mild chemical treatments, involving soaking in oleum, result
in non-negligible oxidation, which requires annealing at
800 °C to remove.12† School of Physics, Trinity College Dublin.
‡ University of Oxford.
§ CRANN, Trinity College Dublin.
| School of Chemistry, Trinity College Dublin.
(1) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang,
Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Science 2004,
306, 666–669.
(2) Geim, A. K.; Novoselov, K. S. Nat. Mater. 2007, 6, 183–191.
(3) Berger, C.; Song, Z. M.; Li, X. B.; Wu, X. S.; Brown, N.; Naud, C.;
Mayo, D.; Li, T. B.; Hass, J.; Marchenkov, A. N.; Conrad, E. H.;
First, P. N.; de Heer, W. A. Science 2006, 312, 1191–1196.
(4) Ohta, T.; El Gabaly, F.; Bostwick, A.; McChesney, J. L.; Emtsev,
K. V.; Schmid, A. K.; Seyller, T.; Horn, K.; Rotenberg, E. New J.
Phys. 2008, 10.
(5) Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.;
Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H. Nature 2009, 457, 706-
710.
(6) Eda, G.; Fanchini, G.; Chhowalla, M. Nat. Nanotechnol. 2008, 3, 270–
274.
(7) Li, D.; Muller, M. B.; Gilje, S.; Kaner, R. B.; Wallace, G. G. Nat.
Nanotechnol. 2008, 3, 101–105.
(8) Niyogi, S.; Bekyarova, E.; Itkis, M. E.; McWilliams, J. L.; Hamon,
M. A.; Haddon, R. C. J. Am. Chem. Soc. 2006, 128, 7720–7721.
(9) Stankovich, S.; Dikin, D. A.; Piner, R. D.; Kohlhaas, K. A.;
Kleinhammes, A.; Jia, Y.; Wu, Y.; Nguyen, S. T.; Ruoff, R. S. Carbon
2007, 45, 1558–1565.
(10) Becerril, H. A.; Mao, J.; Liu, Z.; Stoltenberg, R. M.; Bao, Z.; Chen,
Y. ACS Nano 2008, 2, 463–470.
(11) Jung, I.; Pelton, M.; Piner, R.; Dikin, D. A.; Stankovich, S.;
Watcharotone, S.; Hausner, M.; Ruoff, R. S. Nano Lett. 2007, 7, 3569–
3575.
Published on Web 02/19/2009
10.1021/ja807449u CCC: $40.75 2009 American Chemical Society J. AM. CHEM. SOC. 2009, 131, 3611–3620 9 3611
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
下划线
Administrator
下划线
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
矩形
Recently, a significant breakthrough was made when two
independent groups showed that graphite could be exfoliated
in certain solvents to give defect-free monolayer graphene.13,14
This phenomenon relies on using particular solvents, such as
N-methyl-pyrrolidone, whose surface energy is so well matched
to that of graphene that exfoliation occurs freely.14 However,
this process is not without its drawbacks. These solvents are
expensive and require special care when handling. In addition,
they tend to have high boiling points, making it difficult to
deposit individual monolayers on surfaces. Unfortunately, the
most useful solvent of all, water, has a surface energy that is
much too high to work on its own as an exfoliant for graphene.
With these factors in mind, it is easy to see what is needed.
We require an alternative, liquid phase process that results in
the exfoliation of graphite to give graphene at reasonably high
yield. The method should be non-oxidative and should not
require high temperature processes or chemical post treatments.
In addition, it should be compatible with safe, user-friendly,
low boiling-point solvents, preferably water.
In this Article, we demonstrate such a method. We disperse
graphite in surfactant-water solutions in a manner similar to
surfactant aided carbon nanotube dispersion.15-20 By transmis-
sion electron microscopy (TEM) analysis, we demonstrate
significant levels of exfoliation including the observation of a
number of graphene monolayers. Atomic resolution TEM shows
the monolayers to be well graphitized and largely defect free.
Raman, IR, and X-ray photoelectron spectroscopies also show
the graphite/graphene to be relatively defect free and only very
slightly oxidized. These dispersions can be vacuum filtered to
make thin conductive films and deposited onto surfaces as
individual flakes.
2. Results and Discussion
2.1. Optimization of Dispersion Conditions. The absorption
coefficient, R, which is related to the absorbance, A, through
the Lambert-Beer law (A ) RCl, where C is the concentration
and l is the path length), is an important parameter in
characterizing any dispersion. To accurately determine R, we
prepared a dispersion (∼400 mL) with initial graphite concen-
tration, CG,i ) 0.1 mg/mL, and surfactant (sodium dodecylben-
zene sulfonate, SDBS) concentration, CSDBS ) 0.5 mg/mL. This
was then centrifuged and decanted, and the absorption spectrum
was measured (inset of Figure 1). As expected for a quasi two-
dimensional material, this spectrum is flat and featureless21
everywhere except below 280 nm where we observe a strong
absorption band, which scaled linearly with SDBS concentration
but was independent of the graphite concentration; we attribute
this band to the SDBS. A precisely measured volume of the
dispersion was filtered under high vacuum onto an alumina
membrane of known mass. The resulting compact but relatively
thick film (∼5 µm) was washed with 1 L of water and dried
overnight in a vacuum oven at room temperature. The mass of
material in the filtered volume of stock dispersion was then
determined using a microbalance. From thermogravimetric
(TGA) analysis (not shown) of the dried film, we found that 64
( 5% of the film was graphitic; the remainder was attributed
to residual surfactant. We are not surprised to find so much
residual surfactant in these films. Their considerable thickness
(∼5 µm) makes it very difficult to wash away the surfactant
during film formation. Knowledge of the mass of graphite in
the film allowed us to determine the final concentration of the
stock dispersion. A sample of the stock dispersion was then
serially diluted with 0.5 mg/mL SDBS solution, allowing the
measurement of the absorbance per unit length (A/l) versus
concentration of graphite (after centrifugation, CG), as shown
in Figure 1. A straight line fit through these points gives the
absorption coefficient at 660 nm of R ) 1390 mL mg-1 m-1 in
reasonable agreement with the value measured for graphite/
graphene in various solvents.14 The non-zero intercept in Figure
1 is attributable to the A/l of residual SDBS in the dispersion
(intercept of A/l ) 0.72 m-1 compares with residual absorbance
of A/l ≈ 0.5 m-1 for SDBS at CSDBS ) 0.5 mg/mL).
Using R for our dispersions, it is possible to determine CG
for all subsequent samples. Thus, the fraction of graphite
material remaining for any sample after centrifugation (CF) can
be calculated from the ratio of dispersed graphite after CF to
that before CF: CG/CG,i. Using this fraction-remaining as a gauge,
the concentrations CG,i and CSDBS could be optimized. Holding
(12) Li, X.; Zhang, G.; Bai, X.; Sun, X.; Wang, X.; Wang, E.; Dai, H.
Nat. Nanotechnol. 2008, 3, 538–542.
(13) Blake, P.; Brimicombe, P. D.; Nair, R. R.; Booth, T. J.; Jiang, D.;
Schedin, F.; Ponomarenko, L. A.; Morozov, S. V.; Gleeson, H. F.;
Hill, E. W.; Geim, A. K.; Novoselov, K. S. Nano Lett. 2008, 8, 1704–
1708.
(14) Hernandez, Y.; et al. Nat. Nanotechnol. 2008, 3, 563–568.
(15) Moore, V. C.; Strano, M. S.; Haroz, E. H.; Hauge, R. H.; Smalley,
R. E.; Schmidt, J.; Talmon, Y. Nano Lett. 2003, 3, 1379–1382.
(16) O’Connell, M. J.; Bachilo, S. M.; Huffman, C. B.; Moore, V. C.;
Strano, M. S.; Haroz, E. H.; Rialon, K. L.; Boul, P. J.; Noon, W. H.;
Kittrell, C.; Ma, J. P.; Hauge, R. H.; Weisman, R. B.; Smalley, R. E.
Science 2002, 297, 593–596.
(17) O’Connell, M. J.; Boul, P.; Ericson, L. M.; Huffman, C.; Wang, Y. H.;
Haroz, E.; Kuper, C.; Tour, J.; Ausman, K. D.; Smalley, R. E. Chem.
Phys. Lett. 2001, 342, 265–271.
(18) Strano, M. S.; Moore, V. C.; Miller, M. K.; Allen, M. J.; Haroz, E. H.;
Kittrell, C.; Hauge, R. H.; Smalley, R. E. J. Nanosci. Nanotechnol.
2003, 3, 81–86.
(19) Bergin, S. D.; Nicolosi, V.; Cathcart, H.; Lotya, M.; Rickard, D.; Sun,
Z. Y.; Blau, W. J.; Coleman, J. N. J. Phys. Chem. C 2008, 112, 972–
977.
(20) Sun, Z.; Nicolosi, V.; Rickard, D.; Bergin, S. D.; Aherne, D.; Coleman,
J. N. J. Phys. Chem. C 2008, 112, 10692–10699. (21) Abergel, D. S. L.; Fal’ko, V. I. Phys. ReV. B 2007, 75.
Figure 1. Absorbance per unit length (λ ) 660 nm) as a function of graphite
concentration (after centrifugation) for an SDBS concentration, CSDBS )
0.5 mg/mL. Graphite concentration before centrifugation was CG,i ) 0.1
mg/mL. NB, the curve does not go through the origin due to the presence
of a residual SDBS absorbance. (Intercept of A/l ) 0.72 m-1 compares
with residual absorbance of A/l ≈ 0.5 m-1 for SDBS at CSDBS ) 0.5 mg/
mL.) Bottom inset: Absorption spectrum for a sample with CSDBS ) 0.5
mg/mL and CG ) 0.0027 mg/mL. The portion below 400 nm is dominated
by the surfactant absorption and has been scaled by a factor of 1/8 for
clarity. The portion above 400 nm is dominated by graphene/graphite with
some residual SDBS absorption. Top inset: Surfactant-stabilized graphite
dispersions (A) before and (B) immediately after centrifugation. Note that
the dispersions are almost transparent due to the low concentration of
graphite.
3612 J. AM. CHEM. SOC. 9 VOL. 131, NO. 10, 2009
A R T I C L E S Lotya et al.
Administrator
矩形
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
下划线
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
下划线
Administrator
高亮
Administrator
高亮
Administrator
下划线
Administrator
矩形
Administrator
高亮
Administrator
高亮
CSDBS constant at a relatively high value of 10 mg/mL, CG was
measured as a function of CG,i (Figure 2). Interestingly, we
observe an empirical relationship of the form: CG ) 0.01�CG,i.
The highest concentration achieved after CF was CG ) 0.05
mg/mL for CG,i ) 14 mg/mL. We have observed concentrations
in the range 0.002 mg/mL < CG < 0.05 mg/mL. We note that
this is very similar to the range of concentrations generally
achieved for surfactant-stabilized nanotube dispersions.22 The
largest fraction remaining was ∼3 wt % at CG,i ) 0.1 mg/mL
(top inset, Figure 2). This graphite concentration was then fixed
and CSDBS varied. Measurement of the fraction remaining
showed a broad peak (lower inset, Figure 2), similar to those
observed for nanotube-surfactant dispersions.19 The graphitic
content was maximized for CSDBS between 0.5 and 1 mg/mL,
concentrations very close to the critical micelle concentration
(CMC), which is ∼0.7 mg/mL for SDBS.23 The falloff in
dispersed graphite below CSDBS ≈ 0.5 mg/mL is reminiscent of
the destabilization of nanotube dispersions as the surfactant
concentration is reduced below the CMC.19,24 With this in mind,
we can hypothesize that the minimum surfactant concentration
required for successful dispersion of graphite is the critical
micelle concentration. If this is the case, the surfactant concen-
tration could possibly be reduced by using alternative surfactants
with lower CMC. In this work, to keep the concentration of
surfactant to a minimum, all subsequent experiments were
performed on standard dispersions with surfactant concentration
close to the CMC: CSDBS ) 0.5 mg/mL (also CG,i ) 0.1 mg/
mL). (NB, the fraction remaining in the experiment described
in Figure 1, was much smaller than would be expected from
the data shown in Figure 2. This is due to the fact that in the
former experiment a much larger volume was used resulting in
less efficient sonication.)
2.2. Evidence of Exfoliation. To further characterize the exact
form of nanocarbons in the dispersions, we conducted a detailed
TEM analysis of our standard dispersion. TEM samples were
prepared by pipetting a few milliliters of this dispersion onto
holey carbon mesh grids (400 mesh). TEM analysis revealed a
large quantity of flakes of different types as shown in Figure 3.
A small quantity of monolayer graphene flakes was observed
(Figure 3A). A larger proportion of flakes were few-layer
graphene, including some bilayers and trilayers as shown in
Figure 3B and C. In addition, a number of rather disordered
flakes with many layers, similar to the one in Figure 3D, were
observed. The disorder suggests that these flakes formed by
reaggregation of smaller flakes. Finally, a very small number
(2) of very large flakes were observed (Figure 3E). It can be
shown that these are graphite by the observation of thin
multilayers protruding from their edges (Figure 3E, inset). Note
that while these large flakes are rare when counted by number,
they will contribute disproportionally by mass. It is possible to
estimate the number of layers per flake for all but the largest
flakes. These data are illustrated in the histogram for the standard
dispersion in Figure 4A (the very large flakes are ignored in
this histogram). These statistics show a reasonable population
of few-layer graphene. For example, ∼43% of flakes had <5
layers. More importantly, ∼3% of the flakes were monolayer
graphene. While this value is considerably smaller than that
observed for graphene/solvent dispersions,14 working in aqueous
systems brings its own advantages. In general, the majority of
these few-layer flakes had lateral dimensions of ∼1 µm. Thicker
(22) Bergin, S. D.; Nicolosi, V.; Streich, P. V.; Giordani, S.; Sun, Z.;
Windle, A. H.; Ryan, P.; Peter, N.; Niraj, P.; Wang, Z.-T. T.; Carpenter,
L.; Blau, W. J.; Boland, J. J.; Hamilton J. P.; Coleman, J. N. AdV.
Mater. 2008, 20, 1876-1881.
(23) Lockwood, N. A.; de Pablo, J. J.; Abbott, N. L. Langmuir 2005, 21,
6805.
(24) McDonald, T. J.; Engtrakul, C.; Jones, M.; Rumbles, G.; Heben, M. J.
J. Phys. Chem. B 2006, 110, 25339–25346.
Figure 2. Graphite concentration after centrifugation (CF) as a function
of starting graphite concentration (CSDBS ) 10 mg/mL). Upper inset: The
same data represented as the fraction of graphite remaining after CF. Lower
inset: Fraction of graphite after centrifugation as a function of SDBS
concentration (CG,i ) 0.1 mg/mL).
Figure 3. Selected TEM images of flakes prepared by surfactant processing.
(A) A monolayer (albeit with a small piece of square debris close to its
left-hand edge). (B) A bilayer. (C) A trilayer. (D) A disordered multilayer.
(E) A very large flake. Inset: A closeup of an edge of a very large flake
showing a small multilayer graphene flake protruding. (F) A monolayer
from a sample prepared by sediment recycling.
J. AM. CHEM. SOC. 9 VOL. 131, NO. 10, 2009 3613
Liquid Phase Production of Graphene A R T I C L E S
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
矩形
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
矩形
Administrator
高亮
Administrator
高亮
Administrator
高亮
Administrator
下划线
Administrator
下划线
flakes, with more than a few graphene layers per flake, were
larger, ranging up to 3 µm in diameter.
The sediment remaining after centrifugation can be recycled
to improve the overall yield of graphene exfoliation. The
sediment was recovered, and fresh (0.5 mg/mL) SDBS solution
was added. This sediment dispersion was then processed in the
same manner as the original dispersion, and TEM analysis was
carried out. In this case, we also observed the presence of
isolated monolayer graphene in about 3% of cases (Figure 3F).
In addition, the flake thickness distribution shifted toward thinner
flakes with large quantities of bilayers and trilayers; 67% of
flakes observed had <5 layers (Figure 4B). Notably, there were
no large flakes with greater than 10 layers observed, indicating
that the reprocessing of recycled sediment gives better exfo-
liation than processing of the original sieved graphite. We
suggest that the second sonication breaks up the already partially
exfoliated chunks of graphite into even smaller pieces from
which exfoliation occurs more easily.
The ability to easily deposit graphene flakes on a TEM grid
allows their detailed characterization using high-resolution TEM
(HRTEM). We can use this to confirm the presence of graphene
monolayers in these surfactant-stabilized dispersions. Shown in
Figure 5A is a HRTEM image of a graphene monolayer similar
to that shown in Figure 3A. Significant nonuniformities can be
seen, suggesting the presence of residual surfactant. The inset
depicts a fast Fourier transform (FFT) of this image. This is
equivalent to an electron diffraction pattern. The {1100} spots
can clearly be seen. However, the {2110} spots are too faint to
see. This intensity difference is the fingerprint of monolayer
gra