www.advmat.de
By Aoneng Cao,* Zhen Liu, Saisai Chu, Minghong Wu, Zhangmei Ye,
Zhengwei Cai, Yanli Chang, Shufeng Wang,* Qihuang Gong, and
Yuanfang Liu*
The assembly of semiconductor nanoparticles, such as quantum
dots (QDs), on matrices has been extensively studied for their
promising optoelectronic applications.[1–7] To enhance the
photocurrent generated by these semiconductor–matrices systems, it is essential to retard the recombination of electron-hole
species in the semiconductors by molecular electron-relay
semiconductor structures or efficient electron-transport matrices,
such as conductive polymer films or carbon nanotubes
(CNTs).[3–7] The superior electrical conductivity and the flexible
atom-thin 2D feature of graphene[8–11] would make it an excellent
electron-transport matrix. However, there is no such graphenebased optoelectronic system reported till now. Herein, we report
the synthesis of a graphene–CdS quantum dot (G-CdS)
nanocomposite that shows promising optoelectronic properties.
A picosecond ultrafast electron transfer process from the excited
CdS QDs to the graphene matrix has been observed by
time-resolved fluorescence spectroscopy.
Currently, the yield of single-layer graphene sheets from
various mass production methods is quite low,[11–21] and the
major product is usually multiple-layer graphene sheets.[12–16] An
even more serious problem is that single-layer sheets of graphene
are not stable in solution and tend to aggregate back to graphite
gradually. We developed a one-step method to synthesize G-CdS
directly from graphene oxide (GO) in dimethyl sulfoxide
(DMSO), as illustrated in Figure 1a. This approach overcomes
the above two problems by synthesizing G-CdS directly from GO
in a facile one-pot reaction, where the reduction of GO and the
deposition of CdS on graphene occur simultaneously. In addition
to the advantage of simplicity and low cost, the high stability of the
[*] Prof. A. Cao, Z. Liu, Prof. M. Wu, Z. Ye, Z. Cai, Y. Chang, Prof. Y. Liu
Institute of Nanochemistry and Nanobiology
Shanghai University, Shanghai, 200444 (P. R. China)
E-mail: ancao@shu.edu.cn
Prof. S. Wang, S. Chu, Prof. Q. Gong
State Key Laboratory for Mesoscopic Physics, School of Physics
Peking University, Beijing, 100871 (P. R. China)
E-mail: wangsf@pku.edu.cn
Prof. Y. Liu
Beijing National Laboratory of Molecular Science
College of Chemistry and Molecular Engineering
Peking University, Beijing, 100871 (P. R. China)
E-mail: yliu@pku.edu.cn
DOI: 10.1002/adma.200901920
Adv. Mater. 2010, 22, 103–106
COMMUNICATION
A Facile One-step Method to Produce Graphene–CdS
Quantum Dot Nanocomposites as Promising
Optoelectronic Materials
single-layer GO in solution (Fig. 2a) guarantees the formation of
single-layer graphene sheets in the final nanocomposite, hence
they possess better structural and optoelectronic properties. Once
the reaction was complete, CdS-decoration helps to prevent not
only the aggregation of the single-layer graphene sheets, but also
the aggregation of CdS QDs. In fact, our G-CdS composite can be
stored in the solid state, and the solid product can be
re-suspended in different solvents by sonication. The stability
of the G-CdS composite against sonication demonstrates the
strong binding between the CdS QDs and the graphene sheets. It
is also worth mentioning that the CdS QDs are directly decorated
on the graphene sheets, and no molecular linkers are needed to
bridge the QDs and the graphene matrices.
In the above reaction, DMSO serves as a solvent and as a source of
sulfur. The reduction mechanism of GO in the above process may be
a result of: 1) thermal reduction, which has been reported for the
reduction of GO in other solvents at high temperature,[14,15] and 2)
the production of the reductant H2S from DMSO at 180 8C. In
addition, we found that GO could also be reduced solvothermally to
graphene in DMSO at 180 8C, but without the addition of
Cd(CH3COO)2 (Fig. 1b). Therefore, this method can be used to
produce graphene without use of the usually employed toxic
hydrazine as the reducing agent. The electrical conductivity of this
DMSO-reduced graphene is comparable to or slightly better than
that of the hydrazine-reduced one, but is still lower than the pristine
graphene (see Supporting Information).[22]
Transmission electron microscopy (TEM) images (Fig. 2b,c)
show that the G-CdS consisted of single-layer 2D graphene sheets
decorated with CdS QDs. Wrinkles of G-CdS, a characteristic
feature of the single-layer graphene sheets, are observed (Fig. 2c).
Both Figure 2b (heavily decorated) and 2c (sparsely decorated)
show that the individual CdS nanoparticles are well separated
from each other and well spread out on the graphene sheets.
There is no apparent aggregation of CdS QDs on the graphene
sheets, nor large areas of the graphene sheets without CdS
decoration. The good distribution of CdS QDs on graphene
sheets guarantees the efficient optoelectronic properties of
G-CdS. In the studies on the carbon nanotube (CNT)–CdS
nanocomposite, however, it is usually difficult to achieve such a
good decoration of CdS QDs on the CNTs, because the size of the
CdS QDs is in the same range as the diameters of the CNTs.[3–6]
The size of the CdS QDs in G-CdS is around 10 nm as shown by
the high-resolution TEM image in Figure 2d. In the X-ray
diffraction pattern of G-CdS (see Supporting Information), there
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
103
COMMUNICATION
www.advmat.de
104
stable CdS–DMSO complex shell on the CdS
QD’s surface when the CdS QDs are synthesized in DMSO.[24] This stable CdS–DMSO
complex shell can stabilize the CdS QDs.[24]
Further evidence comes from FTIR spectra (see
Supporting Information). The representative
FTIR peaks of the oxygen-containing functional
groups of GO, which include the bands at
1055 cm1 (CO stretching vibrations),
1393 cm1 (tertiary COH groups stretching),
and 1726 cm1 (C¼O stretching of COOH
groups situated at edges of the GO sheets),
are absent in the FTIR spectrum of G-CdS
synthesized from GO, which indicates the
reduction of these functional groups. A new
absorption band at 1570 cm1 attributed to the
skeletal vibration of the graphene sheets appears
in the FTIR spectrum of G-CdS.[14] The
reduction of GO in G-CdS synthesized from
GO is also confirmed by the shift of the G-band
to a lower wavenumber in the Raman spectra,
and the decrease of the D/G band ratio[14] (see
Supporting Information). Although there are
still some oxygenated carbons in the G-CdS
composite, our control experiments with pristine graphene show that the oxygenated groups
that remain are not necessary for the binding of
CdS particles to the graphene sheet (see
Supporting Information).
As a result of its efficient electron-transport
property, graphene significantly quenched the
Figure 1. a) Scheme of the one-step synthesis of G-CdS. The CdS QDs are not shown at their fluorescence of the CdS QDs decorated on it,
demonstrating the potential application of
actual size. b) Scheme of the solvothermal reduction of GO to graphene in DMSO. c) Bottle I,
GO suspension in DMSO; bottle II, completion of reaction as G-CdS settles down; bottle III,
G-CdS in the field of optoelectronics. TimeG-CdS after washing with acetone and ethanol; bottle IV, resuspension of G-CdS in ethanol.
resolved fluorescence spectroscopy was
employed to monitor the emission lifetimes of
free CdS QDs and G-CdS. Two temporal
scanning ranges were used. The instrument response functions
are three main peaks at scattering angles of 26.5068, 43.9608, and
(IRFs) are 16 ps for the 2.2 ns scanning range, and 4 ps for the
52.1328, which correspond to the (111), (220), and (311) crystal
160 ps scanning range. Figure 4a shows the fluorescence decay
planes of CdS, respectively. This result shows that the CdS QDs
on the graphene sheet are of a blende structure (JCPDS 10-0454).
curves of the free CdS QDs and G-CdS in the 2.2 ns range. Both
curves can be fitted with three decay components. The slow
X-ray photoelectron spectroscopy (XPS) (Fig. 3a) proves that
component with a time constant of about 2 ns (2.2 ns and 1.9 ns
graphene produced by this solvothermal reduction in DMSO is
for free CdS QDs and G-CdS, respectively) is probably a result of
equivalent to that synthesized by the hydrazine-reducing method.
the surface defects of the CdS particles that trap the conduction
Deconvolution of the C 1s peak of our DMSO-reduced graphene
band electrons and generate a new excited state. Therefore, the
(Fig. 3a) indicates about 73% of non-oxygenated ring C (284.8 eV),
slow component is not affected by the decoration on graphene.
while that of the hydrazine-reduced graphene (Fig. 3b) is about
The faster components have time constants of 0.47 and 0.12 ns
70%.[16,23]
for the free CdS QDs and G-CdS, respectively. The difference in
As a comparison we also used the hydrazine-reduced graphene
these time constants is likely a result of the different surface areas
instead of GO as the starting material to synthesize G-CdS
of the CdS QDs upon decoration of the graphene sheets.
following the same procedure to prove that graphene in the
The ultrafast decay at the picosecond range was shorter than
G-CdS synthesized directly from GO exists in the reduced form.
the instrument response limit of the 2.2 ns scanning experiDeconvolution of the C 1s peak of the XPS of G-CdS from GO
ments, and was further confirmed in the 160 ps range
(Fig. 3c) shows about 55% of non-oxygenated ring C, while that of
experiments with an IRF of 4 ps (Fig. 4b). This component is
G-CdS from the hydrazine-reduced graphene (Fig. 3d) is about
almost negligible for the free CdS QDs, according to its small
51%. These results demonstrate that G-CdS synthesized from GO
amplitude. On the contrary, the ultrafast component with a time
is similar to that synthesized by the same method but from the
constant of about 5 ps is the major component (about 86% in
hydrazine-reduced graphene. The calculated non-oxygenated ring
amplitude) for the G-CdS in the 160 ps scanning range. We
C in G-CdS is lower than that of graphene, because there is a
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2010, 22, 103–106
www.advmat.de
COMMUNICATION
Figure 2. a) AFM image (3 mm 3 mm) shows the single layers of GO.
b) TEM image of a G-CdS sheet with densely coated CdS QDs. c) TEM
image of a G-CdS sheet sparsely coated with CdS QDs, showing natural
wrinkles of a single graphene sheet. d) High-resolution TEM image of CdS
crystals on a graphene sheet.
Figure 4. Time-resolved fluorescence decays (dotted curves) of the free
CdS QDs and G-CdS. a) 2.2 ns scanning range with an IRF of 16 ps and
b) 160 ps scanning range with and IRF of 4 ps. Bold curves are fitted results.
ascribe this ultrafast process to the electron transfer from the
excited CdS to the graphene matrices.
In conclusion, a G-CdS nanocomposite material with good
structural and optoelectronic properties has been successfully
and directly synthesized from GO by a facile one-step reaction.
XPS, FTIR, and Raman measurements evidence that GO has
been simultaneously reduced to graphene during the deposition
of CdS. This simple approach takes advantage of the stable
single-layer property of GO to guarantee the final G-CdS product
in a single-layer form. A picoseconds ultrafast electron transfer
process from the excited CdS to the graphene sheet has been
detected by time-resolved fluorescence spectroscopy, which
demonstrates the potential optoelectronic application of this
new type of graphene-based semiconductor hybrid system. In
comparison with CNTs, the large 2D flexible atom-thin layer of
graphene makes it easier to control the distribution of CdS on the
graphene sheet and fabricate future optoelectronic devices. More
experiments that employ a similar one-step synthesis strategy to
produce G-semiconductor hybrid systems directly from GO using
other semiconductors and solvents are underway.
Experimental
Figure 3. XPS spectra of the C 1s peaks of a) DMSO-reduced graphene,
b) hydrazine-reduced graphene, c) G-CdS synthesized directly from GO,
and d) G-CdS synthesized from hydrazine-reduced graphene. Dashed
curves show the deconvoluted peaks.
Adv. Mater. 2010, 22, 103–106
Natural graphite powder (30 mm, with purity >99.85 wt %) was
purchased from Sinopharm Chemical Reagent co., Ltd, China. GO was
prepared by Hummers method [25] as modified by Kovtyukhova [26]. To
obtain single-layer GO sheets, solid GO was dispersed in water (0.5 g L–1),
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
105
COMMUNICATION
www.advmat.de
and sonicated (40 kHz, 500 W) for 30 min under ambient conditions until
the solution became clear. The dispersion was then centrifuged at
3000 rpm for 10 min to remove any unexfoliated GO. The resultant
homogeneous yellow-brown dispersion was stable for months.
To produce graphene by solvothermal reduction of GO in dimethylsulfoxide (DMSO), GO (40 mg) was dispersed in DMSO (40 mL, 99%,
Sinopharm Chemical Reagent Co., Ltd, China). After vigorous stirring, a
stable suspension was obtained. The suspension was transferred into a
Teflon-lined stainless steel autoclave (50 mL), and reacted at 180 8C for
12 h. Graphene was precipitated out as a black solid in the suspension. For
comparison, GO was also reduced by hydrazine [16].
To prepare G-CdS directly from GO, GO (40 mg) and
Cd(CH3COO)2 2H2O (0.106 g, 98.5%, Sinopharm Chemical Reagent
Co., Ltd, China) were dispersed in DMSO (40 mL). After vigorous stirring,
the solution was transferred into a Teflon-lined stainless steel autoclave
(50 mL) and reacted under 180 8C for 12 h. The obtained solution was then
washed extensively with acetone and then alcohol in a sonication washer to
remove non-reacted reactants and CdS QDs not bound to the graphene
sheet. Finally, the product was centrifuged at 5000 rpm, and dried in a
vacuum drier. For comparison, hydrazine-reduced graphene, instead of
GO, was used to produce G-CdS following exactly the same procedure. As a
control, free CdS QDs were also synthesized in DMSO under the same
reaction conditions without adding GO or graphene [24].
Atomic force microscopy (AFM) images were recorded with a Shimadzu
SPM-9600 (Shimadzu, Japan) in tapping mode. AFM samples were
prepared by drop casting the GO suspension in water onto freshly cleaved
mica surfaces, and dried under room temperature. Low-resolution TEM
images were obtained on a JEM 200CX microscope (JEOL, Japan), using an
accelerating voltage of 120 kV. High-resolution TEM images were taken on
a JEOL JEM-2010F microscope (JEOL, Japan) at an acceleration voltage of
200 kV. The specimens were prepared by drop casting the sample
dispersion onto a carbon-coated 300 mesh copper grid and dried under
room temperature. X-ray powder diffraction patterns were recorded using a
D/MAX-2550 diffractometer (Rigaku, Japan), equipped with a rotating
anode and with a Cu Ka radiation source (l ¼ 1.54178 Å). FTIR spectra
were recorded on a Thermo Nicolet Avatar 370 FT-IR spectrometer
(Thermo Nicolet, USA) with a resolution of 2 cm1, and samples were
dried at 80 8C under vacuum for 24 h prior to fabrication of the KBr pellet.
Raman spectra were recorded on a Renishaw Invia Plus laser Raman
spectrometer (Renishaw, UK), with an excitation laser wavelength of
514.5 nm. The XPS data were determined on an AXIS Ultra instrument
(Kratos, UK) at 293 K. The chamber pressure was kept below 108 torr. A
binding energy of 284.8 eV for the C 1s level was used as an internal
reference. The C 1s peaks were deconvoluted using XPS Peak 4.1.
The time-resolved fluorescence spectra were collected by a C5680
synchroscan streak camera (Hamamatsu, Japan). The sample solutions were
placed in rotating cells to avoid photobleaching. The samples were excited by
a frequency doubled Ti:sapphire laser pulse (Mira 900F, Coherent, USA) at
415 nm. The pulse width was about 120 fs, and the repetition rate was
76 MHz. After the excitation, the sample emissions passed through a
polarizer at a magic angle to the laser polarization for isotropic fluorescence
decay measurement. The fluorescence was then focused into the entrance
slit of the spectrograph, which spatially disperses the spectrum before
entering the streak camera. The temporal scanning ranges were selected as
160 ps and 2.2 ns, with an IRF as fast as 4 and 16 ps, respectively.
Acknowledgements
The authors thank Prof. Haifang Wang, Prof. Xuefeng Guo, and Mr. Lin Gan
for experimental assistance and helpful discussions. This work was
106
supported by the China Natural Science Foundation (Nos. 20673003,
10821062 and 60878019), the China Ministry of Science and Technology
(973 Program No. 2009CB930200 and No. 2006CB705604), Shanghai
Municipal Education Committee (09YZ16), and Shanghai Leading
Academic Disciplines (S30109). Supporting Information is available online
from Wiley InterScience or from the author.
Received: June 8, 2009
Published online: September 3, 2009
[1] S. Banerjee, S. S. Wong, Nano Lett. 2002, 2, 195.
[2] S. Ravindran, S. Chaudhary, B. Colburn, M. Ozkan, C. S. Ozkan, Nano Lett.
2003, 3, 447.
[3] Q. Huang, L. Gao, Nanotechnol. 2004, 15, 1855.
[4] I. Robel, B. A. Bunker, P. V. Kamat, Adv. Mater. 2005, 17, 2458.
[5] L. Sheeney-Haj-Khia, B. Basnar, I. Willner, Angew. Chem. Int. Ed. 2005, 44,
78.
[6] L. Sheeney-Haj-Ichia, J. Wasserman, I. Willner, Adv. Mater. 2002, 14, 1323.
[7] E. Granot, F. Patolsky, I. Willner, J. Phys. Chem. B 2004, 108, 5875.
[8] X. L. Li, G. Y. Zhang, X. D. Bai, X. M. Sun, X. R. Wang, E. Wang, H. J. Dai,
Nat. Nanotechnol. 2008, 3, 538.
[9] A. K. Geim, K. S. Novoselov, Nat. Mater. 2007, 6, 183.
[10] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang,
S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Science 2004, 306, 666.
[11] A. B. Kashuba, Phys. Rev. B. 2008, 78, 085 415.
[12] Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, S. De,
I. T. McGovern, B. Holland, M. Byrne, Y. K. Gun’Ko, J. J. Boland,
P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison,
V. Scardaci, A. C. Ferrari, J. N. Coleman, Nat. Nanotechnol. 2008, 3,
563.
[13] D. Li, M. B. Muller, S. Gilje, R. B. Kaner, G. G. Wallace, Nat. Nanotechnol.
2008, 3, 101.
[14] C. Nethravathi, M. Rajamathi, Carbon 2008, 46, 1994.
[15] X. Fan, W. Peng, Y. Li, X. Li, S. Wang, G. Zhang, F. Zhang, Adv. Mater. 2008,
20, 4490.
[16] S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes,
Y. Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff, Carbon 2007, 45, 1558.
[17] C. Berger, Z. M. Song, X. B. Li, X. S. Wu, N. Brown, C. Naud, D. Mayou,
T. B. Li, J. Hass, A. N. Marchenkov, E. H. Conrad, P. N. First, W. A. de Heer,
Science 2006, 312, 1191.
[18] Y. S. Dedkov, M. Fonin, U. Rudiger, C. Laubschat, Phys. Rev. Lett. 2008, 100,
107 602.
[19] X. Li, X. Wang, L. Zhang, S. Lee, H. Dai, Science 2008, 319, 1229.
[20] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. Katsnelson,
I. V. Grigorieva, S. V. Dubonos, A. A. Firsov, Nature 2005, 438, 197.
[21] P. W. Sutter, J. I. Flege, E. A. Sutter, Nat. Mater. 2008, 7, 406.
[22] C. Gómez-Navarro, R. T. Weitz, A. M. Bittner, M. Scolari, A. Mews,
M. Burghard, K. Kern, Nano Lett. 2007, 7, 3499.
[23] S. Stankovich, R. D. Piner, X. Q. Chen, N. Q. Wu, S. T. Nguyen, R. S. Ruoff,
J. Mater. Chem. 2006, 16, 155.
[24] M. E. Wankhede, S. K. Haram, Chem. Mater. 2003, 15, 1296.
[25] W. S. Hummers, Jr, R. E. Offeman, J. Am. Chem. Soc. 1958, 80, 1339.
[26] N. I. Kovtyukhova, P. J. Ollivier, B. R. Martin, T. E. Mallouk, S. A. Chizhik,
E. V. Buzaneva, A. D. Gorchinskiy, Chem. Mater. 1999, 11, 771.
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2010, 22, 103–106