Journal of Colloid and Interface Science 398 (2013) 161–167

Contents lists available at SciVerse ScienceDirect

Journal of Colloid and Interface Science

www.elsevier.com/locate/jcis

Bismuth oxyiodide–graphene nanocomposites with high visible light

photocatalytic activity
Hong Liu ⇑, Wei-Ran Cao, Yun Su, Zhen Chen, Yong Wang ⇑
Department of Chemical Engineering, School of Environmental and Chemical Engineering, Shanghai University, 99 Shangda Road, Shanghai 200444, PR China

a r t i c l e i n f o a b s t r a c t

Article history: A series of chemically bonded Bismuth oxyiodide (BiOI)–graphene (GR) nanocomposites have been syn-
Received 30 October 2012 thesized by a facile one-step hydrothermal method. Both the reduction in graphene oxide (GO) and the
Accepted 4 February 2013 formation of BiOI nanocrystals were achieved simultaneously during the hydrothermal reaction. The pre-
Available online 24 February 2013
pared materials were characterized by means of powder X-ray diffraction (XRD), Fourier transform infra-
red spectroscopy (FT-IR), Raman spectra, high-resolution transmission electron micrographs (HRTEM),
Keywords: UV–vis diffuse reflectance spectra (DRS), and photoluminescence (PL) emission spectroscopy. The photo-
Photocatalysis
catalytic activities of these BiOI–GR nanocomposites were evaluated by the degradation of methyl
BiOI
Graphene
orange. Under visible irradiation (k > 420 nm), the BiOI–GR photocatalysts were found to exhibit higher
Visible light photocatalytic activities than pure BiOI, and the activity was increased by almost 6 times when loaded
Methyl orange with 2.0 wt% graphene. The enhanced photocatalytic activity can be attributed to more effective charge
transportations and separations arisen from the strong chemical bonding between BiOI and graphene, the
high dye adsorption performance, and the increased light absorption.
Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction
Semiconductor photocatalysis has attracted great interest be-
cause it provides a promising pathway for solving energy supply
and environmental pollution problems [1–3]. Titanium dioxide
(TiO2), as an important semiconductor, has been extensively inves-
tigated in the photocatalytic field, owing to its peculiar chemical
and physical behaviors [4,5]. However, its relatively wide band
gap of 3.2 eV limits its application in the visible light region
(400 nm < k < 750 nm), which accounts for 43% of the incoming so-
lar energy. Moreover, the low separation rate of the photoexcited
electron–hole in TiO2 leads to its limited quantum efficiency. Given
the conditions above, it is critical and promising to exploit the no-
vel nontitania-based photocatalysts with high photocatalytic per-
formances under visible irradiation [6–8].
Recently, as novel ternary oxide semiconductors, bismuth oxy-
halides (BiOX, X = F, Cl, Br, and I) have drawn much attention for
their potential application in photocatalysis due to their uniquely
layered structures with the internal static electric field perpendicu-
lar to each layer, which can induce the effective separation of
photo-generated electron–hole pairs, and therefore achieve a high
photocatalytic performance [9–12]. Among them, Bismuth oxyio-
dide (BiOI), with an estimated band gap of about 1.8 eV, is of
particular importance because it possesses the strongest absorption
⇑ Corresponding authors. Fax: +86 21 66137725.
E-mail addresses: liuhong@shu.edu.cn (H. Liu), yongwang@shu.edu.cn
(Y. Wang).
in the visible light region and the best photocatalytic activity [13].
Unfortunately, the quick recombination of photo-generated charge
carriers still exists in BiOI, and the individually BiOI show limited
photocatalytic efficiency [14]. So, its photocatalytic efficiency must
be further improved in order to be suitable for practical applica-
tions. It was found recently that the photocatalytic activity of BiOI
could be significantly enhanced by iodine self-doping [15] or sur-
face modification by nobel metals, such as Ag [16] and Pt [17]. In
addition, BiOI combinated with other semiconductor to form
heterostructures or nanocomposites, such as BiOCl/BiOI [18],
BiOBr/BiOI [19], BiOI/TiO2 [14], AgI/BiOI [20], BiOI/Bi2O3 [21], and
ZnO/BiOI [22], have also proven an effective way to expand the
photoabsorption range and facilitate the separation of the photoin-
duced carriers.
Various carbon materials, including activated carbon, carbon
black, and graphite, as well as carbon nanotubes (CNTs), are found
to play important roles in heterogeneous catalysis as either cata-
lysts or catalyst supports [23]. Graphene (GR), a new carbon mate-
rial with a monolayer of sp2-hybridized carbon atoms in a dense
honeycomb crystal structure, has generated great interest both in
the fields of fundamental research and industrial applications due
to its good conductivity, high electron mobility, superior chemical
stability, and large surface-to-volume ratio [23–26]. Similar to
CNTs, graphene can also act as an excellent electron-acceptor/trans-
port material. Efforts have been made to combine semiconductor
and graphene to obtain hybrid materials with superior photocata-
lytic performance. For example, many works have reported the en-
hanced photocatalytic activity in the water photocatalytic splitting,

0021-9797/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcis.2013.02.007
162 H. Liu et al. / Journal of Colloid and Interface Science 398 (2013) 161–167

antibacterial application, and the degradation of organic pollutions laser light. UV–vis diffuse reflectance spectra (DRS) of the samples
by GR–TiO2 nanoparticle composite [25,27–29]. Zhu’s group syn- were obtained on an UV–vis spectrophotometer (Hitachi U-3010)
thesized ZnO–GR composite by reducing graphene oxide (GO) using BaSO4 as a reference. The photoluminescence (PL) spectra
coated on the surface of ZnO nanoparticles using hydrazine and were measured using a Hitachi F-7000 fluorescence spectropho-
found that the ZnO–GR composite showed an improved photocata- tometer at room temperature.
lytic efficiency in the degradation of methylene blue [30]. Pan’s
group synthesized CdS/GR composite with visible light response
2.3. Photocatalytic activity test
by a microwave-assisted method and the CdS/GR composites exhib-
ited enhanced photocatalytic performance for the reduction in
Photocatalytic activities of the samples were determined by the
Cr(VI) [31]. Additionally, other graphene-based composites, such
photocatalytic degradation of methyl orange (MO) in an aqueous
as Bi2WO6/GR [32], ZnFe2O4/GR [33], InNbO4/GR [34], and Sr2Ta2-
solution under visible light irradiation. A 500 W Xe lamp with a
O7xNx/GR [35], have also been fabricated recently. However, up
k < 420 nm cutoff filter was used as visible light source. Circulation
to now, no report has discussed the preparation and properties of
of water through an external cooling coil and a ventilating fan was
BiOI combined with graphene composites and their application in
used to prevent any thermal catalytic effects. All experiments were
the degradation of organic dyes.
conducted at room temperature in air. In a typical photocatalytic
Motivated by these facts, we developed a facile one-step hydro-
experiment, 0.05 g of the photocatalyst was added into 50 ml of
thermal method to prepare BiOI–GR nanocomposites for the first
10 mg/L MO solution in a reaction cell with a Pyrex jacket. Prior
time. The prepared samples were characterized, and the photocat-
to irradiation, the suspension was magnetically stirred in the dark
alytic activities under visible light irradiation (k > 420 nm) were
for 1 h to reach an adsorption–desorption equilibrium of the dye
evaluated by the degradation of methyl orange (MO). The result re-
on the catalyst surface. Then, the suspension was exposed to visi-
vealed that all BiOI–GR nanocomposites exhibited much higher
ble light irradiation under magnetic stirring. At given time inter-
photocatalytic activities than the pure BiOI, and 2.0 wt% graphene
vals, about 5 ml suspensions were collected and centrifuged
in the photocatalyst gave the best activity. Furthermore, a mecha-
(13,000 rpm, 15 min) to remove the photocatalyst particles. Then,
nism for photocatalytic reaction in the BiOI–GR system was
the MO concentration of the obtained solution was analyzed by a
proposed.
UV–vis spectrophotometer (Hitachi, U-3310) by checking the
absorbance at 464 nm.
2. Materials and methods
3. Results and discussion
2.1. Preparation of BiOI–GR nanocomposites
3.1. Crystal structure
All the chemicals were of analytical grade from the Sinopharm
Chemical Reagent Co., Ltd. and were used as received without fur-
Fig. 1 shows the XRD patterns of the BiOI–GR nanocomposites
ther purification. Graphene oxide was prepared according to the
and pure BiOI. It is found that the main diffraction peaks of BiOI–
modified Hummers method [36]. The BiOI–GR nanocomposites
GR composites are similar to that of pure BiOI and corresponds
were obtained via a one-step hydrothermal method. In detail,
to tetragonal BiOI (JCPDS 10-0445), which indicates that the pres-
0.97 g Bi(NO3)35H2O was firstly dissolved in 3 ml acetic acid
ence of graphene does not result in the development of new crystal
(HAc). Then, the solution was added slowly to another one contain-
orientations or changes in preferential orientations of BiOI. How-
ing 0.30 g NaI in 40 ml deionized water under magnetically stirring.
ever, the intensity of the diffraction peaks for the composites
Meanwhile, GO of different mass was dispersed in 20 ml ethanol by
shows a little decrease as the content of graphene increased. No
sonication for 1 h to get a homogenous suspension of exfoliated
apparent peaks of graphene oxide are observed in all the BiOI–
graphene oxide. Then, the obtained GO solution was added to the
GR composites. This can be ascribed to the fact that the regular
reacted-mixture gradually. After stirring for 2 h at room tempera-
stacking of GO was destroyed by the reduction process under the
ture, the mixture was transferred into a 100 ml Teflon-lined stain-
hydrothermal conditions and could not be resolved by XRD. Mean-
less steel autoclave and maintained at 120 °C for 6 h then was
while, no characteristic peaks assigned to graphene are observed in
cooled to room temperature naturally. Under such a hydrothermal
condition, the solvent of ethanol–water has the strong power to
reduce GO to GR [27,37]. So, the reduction in GO to GR and the
(012)

deposition of BiOI onto the GR sheet can be simultaneously

(110)

achieved. The resulting precipitate was filtrated, washed thor-

(122)
(014)

(114)
(013)
(004)

(005)

oughly, and vacuum dried at 60 °C overnight. The as-synthesized

(016)
(024)
(220)
(002)

(115)

BiOI–GR sample with 1.0, 2.0, and 3.0 wt% GR was labeled as (a)
BG1.0, BG2.0, and BG3.0, respectively. For comparison, pure BiOI
Intensity (a.u.)

was prepared using the same hydrothermal method without the

addition of GO. (b)

2.2. Characterization (c)

The crystalline phases of prepared samples were identified by

X-ray diffraction (XRD, D/MAX-2550, Cu K, k = 0.15418 nm). The (d)
morphologies and microstructures of the catalysts were evaluated
by high-resolution transmission electron microscopy (HRTEM)
10 20 30 40 50 60 70 80
(JEM-2010F). Fourier transform infrared spectroscopy (FT-IR) anal-
2 Theta (degree)
ysis was carried out on a Nicolet AVATAR 370 spectrophotometer.
Raman spectra were recorded on a microscopic confocal Raman Fig. 1. XRD patterns of BiOI and BiOI–GR composites: (a) BiOI, (b) BG1.0, (c) BG2.0,
spectrometer (Renishaw, INVIA) with an excitation of 514.5 nm and (d) BG3.0.

the composites, which might be due to the low amount of graph-
ene and relatively low diffraction intensity of GR in the composite.
However, the presence of graphene in the BiOI–GR composites can
be easily evidenced by Fourier transform infrared and Raman spec-
troscopy, as discussed later.
3.2. FT-IR spectra
FT-IR spectra of graphite oxide, BiOI, and BiOI–GR composites
are shown in Fig. 2. For GO, the characteristic peaks of oxygen-con-
taining functional groups are revealed by the bands at 1060, 1228,
1396, and 1726 cm1, which correspond to CAOAC stretching
vibrations, the CAOH stretching peak, the OAH deformation of
the CAOH groups, the C@O stretching vibrations of the ACOOH
group [33,38], respectively. The peak at 1631 cm1 can be assigned
to the skeletal vibrations of unoxidized graphitic domains [39]. The
wide band at 3440 cm1 should originate from the absorption of
water or OAH groups [40]. For BiOI–GR composites, the character-
istic features of GO are almost disappeared, revealing that these
oxygen-containing functional groups were almost removed. Addi-
tionally, the new absorption band at 1570 cm1 observed in the
FTIR spectrum of BiOI–GR can be assigned to the skeletal vibration
of graphene sheets [37], indicating the successful reduction in GO
to graphene during the hydrothermal reaction. Since hydrothermal
reduction in GO usually results in incompletely reduced products
[41,42], it is reasonable that some oxygen-containing functional
groups still remain after reduction. The small peak around
1702 cm1 is assigned to C@O stretching of the residual COOH
groups, which is beneficial to improve the dispersion of the com-
posites in water. The above results confirm the reduction in GO
and the presence of graphene in the BiOI–GR composites. Pure BiOI
shows absorption at low frequency (below 1000 cm1), which cor-
responded to the vibration of Bi@O@Bi bonds. However, in the as-
prepared BiOI–GR composites, the absorption below 1000 cm1
shifts toward low wavenumber, indicating an intensive chemical
interaction between graphene and BiOI occurred [25]. This could
be due to the residual carboxylic acid functional groups of graph-
ene oxide interacted with the surface hydroxyl groups of BiOI dur-
ing the hydrothermal reduction and finally formed the chemically
bonded BiOI–GR composites [42].
3.3. Raman spectra
Raman spectroscopy is a useful nondestructive tool to charac-
terize carbonaceous materials, particularly for distinguishing
1726
1631
Transmittance (a.u.)
(a)
3440
17021570
1702 1570
(b)
(c)
1570
(d) 1702
(e)
4000 3500 3000 2500 2000
Wavenumber (cm-1)
Fig. 2. FT-IR spectra of GO, BiOI and BiOI–GR composites: (a) GO, (b) BiOI, (c) BG1.0,
(d) BG2.0, and (e) BG3.0.
H. Liu et al. / Journal of Colloid and Interface Science 398 (2013) 161–167 163
(a)
(b)
(c)
(d)
(e)
Intensity (a.u.)
500 1000 1500 2000 2500 3000
Raman Shift (cm-1)
Fig. 3. Raman spectra of GO, BiOI and BiOI–GR composites: (a) GO, (b) BG3.0, (c)
BG2.0, (d) BG1.0, and (e) BiOI.
ordered and disordered carbon structures. Fig. 3 shows the Raman
spectra of graphene oxide, pure BiOI, and the BiOI–GR composites.
In the Raman spectrum of graphene oxide, two typical bands of
graphene oxide can be found at 1350 (D band) and 1607 cm1
(G band) [43]. The G band is generally assigned to the E2g phonon
of sp2 bonds of carbon atoms [44]. The D band is attributed to local
defects and disorders, particularly the defects located at the edges
of graphene and graphite platelets [45,46]. For the BiOI–GR
composites, all the Raman bands for tetragonal BiOI can be found.
Significantly, the two characteristic peaks at about 1347 cm1
(D band) and 1598 cm1 (G band) for the graphitized structures
are observed in the Raman spectroscopy of BiOI–GR composites.
Also, in comparison with the bands of GO, the D band moves to
1347 cm1, and the G band shifts to 1598 cm1, which matches
the value of pristine graphite (1600 cm1) [28], indicating the
reduction in GO. A further observation indicates that the BiOI–GR
composites show an increased D/G intensity ratio in comparison
with that of pure GO (the ID-band/IG-band ratio of the samples are
shown in Table 1). This change suggests a decrease in the average
size of the in-plane sp2 domains upon reduction in the exfoliated
GO, confirming the existence of graphene sheets in the BiOI–GR
composites [28]. Table 1 also reveals that the D/G intensity ratio
increased with increasing graphene content, which suggested that
samples with thinner carbon shells formed a more ordered carbon
structure during the graphitization process [25].
3.4. Hybrid structures
The hybridization structure of BiOI and graphene was investi-
1396
12281060 gated by HRTEM. The graphene nanosheets and BiOI nanoplates
are clearly observed from Fig. 4a. The graphene sheets are not very
791
494 flat but display intrinsic microscopic wrinkles, and BiOI nanoplates
dispersed on graphene nanosheets. In Fig. 4b, the lattice space in
490 BiOI–GR is about 0.385 and 0.280 nm, corresponding to the
789 (0 0 2) plane of GR and the (1 1 0) plane of tetragonal BiOI, respec-
789 490
tively. The image in Fig. 4b shows the intimate contact between
490 BiOI and graphene. This intimate contact can make the electronic
789 interaction between BiOI and graphene possible and improve the
charge separation and the photocatalytic activity.
Table 1
Degree of graphitization of the samples.
Sample Degree of graphitization ID-band/IG-band
1500 1000 500
GO 0.782
BG1.0 0.944
BG2.0 0.946
BG3.0 0.952
164 H. Liu et al. / Journal of Colloid and Interface Science 398 (2013) 161–167

Fig. 4. HRTEM images of BG2.0.

3.5. UV–vis diffuse reflectance spectra and band gap structures
Fig. 5 reveals the UV–vis diffuse reflectance spectra (DRS) of
BiOI–GR composites and pure BiOI. It is seen that pure BiOI has a
characteristic absorption at about 680 nm, corresponding to the
charge transfer process from the valence band to conduction band.
The deposition of graphene has remarkable influences on the spec-
tra of UV–vis diffuse reflectance. With the increase in graphene
content, the samples show a stronger absorption in both UV and
visible range and an obvious redshift in the band gap transition.
When the content of graphene increasing from 1 to 3%, the absorp-
tion edges (the intercept of the tangent to the X-axis) of BiOI–GR
samples are 730, 750, and 760 nm, respectively, an obvious red-
shift of ca. 50–80 nm compared to bare BiOI. The result indicated
that the narrowing of the band gap of BiOI is associated with the
increasing graphene content. In the earlier publications, a decrease
in band gap energy was observed for GR/TiO2 composite due to the
formation of the TiAOAC bond [47,48]. Thus, this narrowing
should be attributed to the chemical bonding between BiOI and
graphene, that is, the formation of BiAOAC bond. This is in accor-
dance with the results from FTIR spectra.
The band gap energy of a semiconductor can be calculated by
the following formula:
respectively. Among them, n depends on the characteristics of the
transition in a semiconductor, that is, direct transition (n = 1) or
indirect transition (n = 4). For BiOI, the value of n is 4 for the indirect
transition [13]. Therefore, the band gap energy (Eg value) of the
resulting samples can be estimated from a plot ðahmÞ1=2 versus pho-
ton energy (hm). The intercept of the tangent to the X-axis would
give a good approximation of the band gap energy of the samples
(Fig. 6). Their calculated band gap energies are summarized in
Table 2. The band gap of BiOI is evaluated to be 1.72 eV, close to
the values reported in the literature [13,49]. It is found that the
band gap of BiOI–GR samples ranges from 1.54 to 1.50 eV with
the content of graphene increasing from 1% to 3%. These results also
reveal that all the samples have suitable band gaps, which can be
activated by visible light for photocatalytic decomposition of
organic contaminants.
The valance band (VB) of a photocatalyst is an important factor
for the effective photocatalytic decomposition of organic contami-
nants. The valence band edge position of BiOI–GR was estimated in
this work according to the concepts of electronegativity. Herein,
the electronegativity of an atom is the arithmetic mean of the
atomic electron affinity and the first ionization energy [50]. The va-
lence band potentials of a semiconductor at the point of zero
charge can be calculated by the following empirical equation [51]:

ahm ¼ Aðhm  Eg Þn=2 ð1Þ EVB ¼ X  Ee þ 0:5Eg ð2Þ

where a, h, m, Eg, and A are the absorption coefficient, Planck where EVB is the VB edge potential, X is the electronegativity of the
constant, the light frequency, the band gap, and a constant, semiconductor, which is the geometric mean of the electronegativ-
ity of the constituent atoms, Ee is the energy of free electrons on the

1.8
0.9
(d) 1.6 (d)
0.8
(c) 1.4 (c)
0.7
(b) (b)
Absorbance (a.u.)

1.2
0.6 (a) (a)
(αhν)1/2

1.0
0.5

0.4 0.8

0.3 0.6

0.2 0.4

0.1 0.2

0.0 0.0
200 300 400 500 600 700 800 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
Wavelength (nm) Photo energy (eV)

Fig. 5. Diffuse reflectance spectra of BiOI and BiOI–GR composites: (a) BiOI, (b) Fig. 6. Plots of the (ahm)1/2 versus photon energy (hm) for as-synthesized samples:
BG1.0, (c) BG2.0, and (d) BG3.0. (a) BiOI, (b) BG1.0, (c) BG2.0, and (d) BG3.0.
 H. Liu et al. / Journal of Colloid and Interface Science 398 (2013) 161–167 165

Table 2
Band gaps, valence band, and photocatalytic activities of as-prepared samples.

Sample Bandgap Eg (eV) Valence band EVB (eV) Conduction band ECB (eV) Rate constant K (h1) R
BiOI 1.72 2.35 0.63 0.0813 0.9986
BG1.0 1.54 2.26 0.72 0.3554 0.9984
BG2.0 1.52 2.25 0.73 0.5217 0.9998
BG3.0 1.50 2.24 0.74 0.2836 0.9999

hydrogen scale (about 4.5 eV), Eg is the band gap energy of the
semiconductor, the X value of BiOI is ca. 5.99 eV, and ECB can be
determined by ECB = EVB–Eg. The EVB and ECB of the samples are cal-
culated and have been summarized in Table 2. Obviously, all of the
valence band (VB) edge potentials of the photocatalysts are more
positive than the standard redox potentials of OH/OH (1.99 eV),
H2O2 (1.77 eV), and O3 (2.07 eV), indicating that such photocata-
lysts may have much stronger oxidation abilities.
3.6. PL properties
The photocatalytic activity of materials is closely related with
the recombination rate of the photoinduced electrons and holes,
which will decrease the quantum yield. The photoluminescence
(PL) emission spectra can be regarded as an effective approach to
understanding the separation capacity of the photoinduced carri-
ers. The higher the PL intensity is, the less efficient carriers partic-
ipate in the photocatalytic procedure [52]. In Fig. 7, a emitting peak
at around 587 nm is observed for pure BiOI sample when excited
by 390 nm laser, which is enough to promote electronic transitions
from the VB to the CB of BiOI according to the above DRS spectrum
(Fig. 5). Though shapes and peaks positions of BiOI–GR samples are
similar to that of pure BiOI, the emission intensity of the compos-
ites decreases. This indicates the presence of graphene can sup-
press the radiative recombination process, leading to weak
recombination of the e/h+ pairs and high photon efficiency. This
effect is also related to the concentration of graphene, the optimal
graphene content (2%) caused the biggest decrease in the intensity
of the photoluminescence peak.
3.7. Photocatalytic activity
The photocatalytic activities of as-prepared samples were eval-
uated by measuring the degradation of MO in an aqueous solution
under visible light. It is well known that there are three possible
mechanisms for MO photodegradation: a photolysis process, a
dye photosensitization process, and a photocatalytic process. In
order to study the mechanisms, the degradation of MO in the ab-
sence of photocatalyst and presence of P25 under visible light were
also investigated. Fig. 8 shows that the MO photolysis is not obser-
vable in the blank experiment and a slight degradation over P25,
which indicate that MO is very stable and the photolysis as well
as the dye-sensitive process are neglectable. Therefore, in our
experiment, the mechanism for MO degradation is a photocatalytic
process. As shown in Fig. 8, under visible light irradiation, BG1.0,
BG2.0, and BG3.0 could remove 75.7%, 88.0%, and 67.8% of MO in
4 h, respectively, whereas only 27.6% of MO was removed with
pure BiOI. This comparison suggests that the introduction of graph-
ene can efficiently enhance the photocatalytic performance of BiOI.
The photoactivity of the BiOI–GR composites is found to be depen-
dent on the ratio of BiOI and graphene. Obviously, BG2.0 shows the
highest MO removal efficiency (88.0%). More graphene contents
(BG3.0) can be detrimental to the photonic activity.
To quantitatively understand the reaction kinetics of the MO
degradation, the pseudo-first-order model as expressed by Eq. (3)
was used to analyze the photocatalytic degradation data if the ini-
tial concentration of pollutant is low:
lnðC 0 =CÞ ¼ kt ð3Þ
where C0 and C are the concentrations of MO at time 0 (the time to
obtain adsorption–desorption equilibrium) and t, respectively, and
k is the pseudo-first-order rate constant. The rate constants evalu-
ated from the data plotted in Fig. 8 are also summarized in Table 2.
It can be seen that a fairly good correlation to the pseudo-first-order
reaction kinetics (R > 0.99) is found. The rate constant of BG1.0,
BG2.0, and BG3.0 is 0.3554, 0.5217, and 0.2836 h1, respectively,
greatly higher than that of BiOI (0.0813 h1), revealing that BiOI–
GR is a much more effective photocatalyst than bare BiOI due to
the introduction of graphene. Among them, BG2.0 possesses the
highest removal constant, which is about 6 times that of BiOI.
In order to study the influence of formation of heterostructures,
BG2.0 was compared with the mechanically mixed counterpart
sample 2%GR + BiOI on the degradation of MO under visible light
irradiation (Fig. 9). In the mixture of BiOI and graphene, the

1.0

(a)
(b) 0.8

(c)
Relative intensity

(d) 0.6
C/C0

0.4 blank
P25
BiOI
0.2 BG3.0
BG1.0
BG2.0

0.0
560 580 600 620 640 0 1 2 3 4
Wavelength (nm) Time (h)

Fig. 7. PL spectra of the as-synthesized samples: (a) BiOI, (b) BG3.0, (c) BG1.0, (d) Fig. 8. Photocatalytic activities of BiOI and BiOI–GR samples on the degradation of
BG2.0. MO under visible light irradiation.
166 H. Liu et al. / Journal of Colloid and Interface Science 398 (2013) 161–167

chemical bonding was not established. Although the 2%GR + BiOI

sample exhibits better activity than that of pure-phase BiOI, its
photocatalytic activity is much lower than that of the chemically
bonded BiOI–GR, which indicates the chemical bonds play a signif- CB e-
icant role in the photodegradation. The above result suggests the
heterojunction might form between BiOI and GR. The electronic hν
interaction between BiOI and graphene can effectively enhance BiOI
(Visible)
the separation of photogenerated charge and thus enhance photo-
catalytic activity significantly. H2O VB
OH- h+

3.8. Mechanism on enhancement of photocatalytic activity

.OH
It is well known that the generation and separation of the pho-
toinduced electron–hole pairs are the key factors to influence a MO degradation
photocatalytic reaction. If photoinduced e/h+ can be separated
effectively, it is benefited to the photocatalytic activities. Theoret-
ically, graphene sheets with 100% sp2-hybridized carbon atoms .OH .OOH H+ .O2- O2
have a high electrical conductivity in storing and shuttling elec-
trons [23,24]. When graphene is combined with other materials, Scheme 1. Photocatalytic mechanism of BiOI–GR composites.
electrons would flow from one material to the other (from the
higher to lower Fermi level) to align the Fermi energy levels at 
OH which is responsible for the degradation of pollutant. All pho-
the interface of two materials [24,53]. In view of the higher work
tocatalytic reactions are summarized in the following equations:
function of graphene than that of BiOI, electrons will flow from
BiOI into graphene to adjust the Fermi energy levels, leading to hv þ
BiOI ! BiOIðecb  hvb Þ ð4Þ
the formation of a Schottky barrier at the BiOI–graphene interface
[53]. This Schottky barrier can capture electrons from BiOI to
ecb þ GR ! GRðecb Þ ð5Þ
graphene and prevent their back flowing to the BiOI. Thus, in
BiOI–GR, graphene served as an acceptor of the generated electrons
GRðecb Þ þ O2 !  O2 ð6Þ
of BiOI and effectively suppressed the charge recombination, which
has been confirmed by the results of PL (shown in Fig. 7). On the
basis of the above results and analysis, we propose a possible

O2 þ Hþ !  OOH ð7Þ
mechanism to explain the superior performances of BiOI–GR to re-
move MO under visible light irradiation as Scheme 1. Under visible

OOH þ ecb þ Hþ ! H2 O2 ð8Þ
excitation, the electron of BiOI can be promoted from the valence
band to the conduction band, leaving behind a hole in the valence H2 O2 þ ecb !  OH þ OH ð9Þ
band. Then, the electron transfers to graphene. In the photocata-
þ
lytic process, the photogenerated electrons accumulated on the H2 O þ hvb !  OH þ Hþ ð10Þ
surface of graphene had good fluidity and could be transferred to
surface-absorbed oxygen rapidly to form activated  O 2 . The acti-
þ
OH þ hvb !  OH ð11Þ
+
vated  O 
2 further produces OH via a series of reaction with H . This
step is the photoreduction process. On the other hand, holes accu- 
OH þ MO ! products ð12Þ
mulated at the valence band of BiOI could also react with H2O to
give rise to hydroxyl radical OH, which is the photooxidation pro- Adsorption performance of the targeted substance is also very
cess. Both the photoreduction and photooxidation step generate important for the surface reaction. To evaluate the adsorption abil-
ity of BiOI and BiOI–GR, the residue rate of MO after 60 min stirring
in the dark was determined. As shown in Fig. 10, the residue rate of
MO in the solution with BiOI as the adsorbent is 89.8%, whereas
1.0 that of BiOI–2%GR as the adsorbent is 69.2%. It is obviously that
the adsorption ability of the BiOI is enhanced by introduction of
0.8
graphene, which is a prerequisite for good photocatalytic activity.
In addition, the enhancement of the photocatalytic performance
should also be ascribed to the increase in the light absorption
0.6 intensity and range, which has been confirmed from the DRS
C/C0

absorption measurement (see Fig. 5). Therefore, we suggest that

the enhanced photocatalytic activity of the BiOI–GR nanocompos-
0.4
ites is mainly attributed to more effective charge transportations
and separations arisen from the strong chemical bonding between
BiOI
0.2 2% GR+BiOI BiOI and graphene, improved dye adsorption ability, and the in-
BG2.0 creased light absorption.
The amounts of graphene also have important effect on photo-
0.0 catalytic activity. It was found that the optimal loading amount of
0 1 2 3 4
graphene was approximately 2%. However, when the graphene
Time (h)
content is further increased above its optimum value, the photo-
Fig. 9. Comparison of the photocatalytic efficiency of BiOI, BG2.0 and BiOI– catalytic performance deteriorates. This is ascribed to the following
graphene physical mixing. reasons: (1) the excessive GR can act as a kind of recombination
 H. Liu et al. / Journal of Colloid and Interface Science 398 (2013) 161–167
1.0
0.9
C/C0
0.8
0.7
BiOI
BG2.0
0.6
0 20 40 60
Time (min)
Fig. 10. Comparison of the adsorption efficiency of BiOI and BG2.0.
center like excess gold or silver nanoparticles, which has been con-
firmed by the results of PL; (2) GR may absorb some visible light
and thus there exists a light harvesting competition between BiOI
and GR with the increase in GR content [31], which lead to the de-
crease in the photocatalytic performance.
4. Conclusions
In summary, a series of novel BiOI–GR photocatalysts have been
synthesized by a facile one-step hydrothermal method. During the
synthesis, both the reduction in GO and the formation of BiOI
nanocrystals were achieved simultaneously. Under visible irradia-
tion, the BiOI–GR photocatalysts were found to exhibit higher pho-
tocatalytic activities than pure BiOI in the degradation of MO. The
activity was increased by almost 6 times when loaded with 2.0 wt%
graphene. The enhanced photocatalytic activity can be attributed
to more effective charge transportations and separations arisen
from the strong chemical bonding between BiOI and graphene,
the enhanced adsorption performance, and the increased light
absorption. While only MO was investigated in this work, it is be-
lieved that such BiOI–GR composites can display good photocata-
lytic properties for the degradation of other organic pollutants
under visible light irradiation. What is more, the promising tech-
nique presented here can be further used to prepare other semi-
conductor nanocomposites based on graphene nanosheets, which
provides a path to obtain a new class of graphene-based materials
and their applications in a variety of areas, such as photocatalytic
hydrogen evolution, lithium-ion batteries, chemical sensors, and
so forth. The corresponding research is ongoing in our laboratory.
Acknowledgments
The authors gratefully acknowledge Natural Science Foundation
of Shanghai (12ZR1410300), Shanghai Municipal Commission of
Education (11YZ15), Shanghai Leading Academic Disciplines
(S30109), National Natural Science Foundation of China
(11072137), the Shu-Guang Program of Shanghai Municipal Com-
mission of Education (11SG38), Science and Technology Commis-
167
sion of Shanghai Municipality (11JC1403900) and the Program
for Professor of Special Appointment in Shanghai (Eastern Scholar)
for the financial support.
References
[1] Z.G. Zou, J.H. Ye, K. Sayama, H. Arakawa, Nature 414 (2001) 625.
[2] L. Xu, Y.L. Hu, C. Pelligra, C.H. Chen, L. Jin, H. Huang, S. Sithambaram, M.
Aindow, R. Joesten, S.L. Suib, Chem. Mater. 21 (2009) 2875.
[3] H. Choi, A.C. Sofranko, D.D. Dionysiou, Adv. Funct. Mater. 16 (2006) 1067.
[4] M.D. Hernandez-Alonso, F. Fresno, S. Suarez, J.M. Coronado, Energy Environ.
Sci. 2 (2009) 1231.
[5] J. Zhang, Q. Xu, Z.C. Feng, M.J. Li, C. Li, Angew. Chem. Int. Ed. 47 (2008) 1766.
[6] I. Tsuji, H. Kato, A. Kudo, Chem. Mater. 18 (2006) 1969.
[7] J. Sato, N. Saito, Y. Yamada, K. Maeda, T. Takata, J.N. Kondo, M. Hara, H.
Kobayashi, K. Domen, Y. Inoue, J. Am. Chem. Soc. 127 (2005) 4150.
[8] A. Ishikawa, T. Takata, J.N. Kondo, M. Hara, K. Domen, J. Phys. Chem. B 108
(2004) 2637.
[9] X. Xiao, W.D. Zhang, J. Mater. Chem. 20 (2010) 5866.
[10] Z.H. Ai, W. Ho, S. Lee, L.Z. Zhang, Environ. Sci. Technol. 43 (2009) 4143.
[11] W.Y. Su, J. Wang, Y.X. Huang, W.J. Wang, L. Wu, X.X. Wang, P. Liu, Scr. Mater.
62 (2010) 345.
[12] C.H. Wang, C.L. Shao, Y.C. Liu, L.N. Zhang, Scr. Mater. 59 (2008) 332.
[13] X. Zhang, Z.H. Ai, F.L. Jia, L.Z. Zhang, J. Phys. Chem. C 112 (2008) 747.
[14] X. Zhang, L.Z. Zhang, T.F. Xie, D.J. Wang, J. Phys. Chem. C 113 (2009) 7371.
[15] X. Zhang, L.Z. Zhang, J. Phys. Chem. C 114 (2010) 18198.
[16] H. Liu, W.R. Cao, Y. Su, Y. Wang, X.H. Wang, Appl. Catal. B: Environ. 111–112
(2012) 271.
[17] C.L. Yu, J.C. Yu, C.F. Fan, H.R. Wen, S.J. Hu, Mater. Sci. Eng., B 166 (2010) 213.
[18] T.B. Li, G. Chen, C. Zhou, Z.Y. Shen, R.C. Jin, J.X. Sun, Dalton Trans. 40 (2011)
6751.
[19] W.D. Wang, F.Q. Huang, X.P. Lin, J.H. Yang, Catal. Commun. 9 (2008) 8.
[20] H.F. Cheng, B.B. Huang, Y. Dai, X.Y. Qin, X.Y. Zhang, Langmuir 26 (2010) 6618.
[21] Y.Y. Li, J.S. Wang, H.C. Yao, L.Y. Dang, Z.J. Li, Catal. Commun. 12 (2011) 660.
[22] J. Jiang, X. Zhang, P.B. Sun, L.Z. Zhang, J. Phys. Chem. C 115 (2011) 20555.
[23] A.K. Geim, Science 324 (2009) 1530.
[24] S. Stankovich, D.A. Dikin, G.H.B. Dommett, K.M. Kohlhaas, E.J. Zimney, E.A.
Stach, R.D. Piner, S.T. Nguyen, R.S. Ruoff, Nature 442 (2006) 282.
[25] Y.J. Wang, R. Shi, J. Lin, Y.F. Zhu, Appl. Catal. B: Environ. 100 (2010) 179.
[26] C. Xu, X. Wang, J. Zhu, J. Phys. Chem. C 112 (2008) 19841.
[27] H. Zhang, X.J. Lv, Y.M. Li, Y. Wang, J.H. Li, ACS Nano 4 (2010) 380.
[28] Q.J. Xiang, J.G. Yu, M. Jaroniec, Nanoscale 3 (2011) 3670.
[29] Y.Y. Liang, H.L. Wang, H.S. Casalongue, Z. Chen, H.J. Dai, Nano Res. 3 (2010)
701.
[30] T.G. Xu, L.W. Zhang, H.Y. Cheng, Y.F. Zhu, Appl. Catal. B: Environ. 101 (2011)
382.
[31] X.J. Liu, L.K. Pan, T. Lv, G. Zhu, Z. Sun, C.Q. Sun, Chem. Comun. 47 (2011) 11984.
[32] E.P. Gao, W.Z. Wang, M. Shang, J.H. Xu, Phys. Chem. Chem. Phys. 13 (2011)
2887.
[33] Y.S. Fu, X. Wang, Ind. Eng. Chem. Res. 50 (2011) 7210.
[34] X.F. Zhang, X. Quan, S. Chen, H.T. Yu, Appl. Catal. B: Environ. 105 (2011) 237.
[35] A. Mukherji, B. Seger, G.Q. Lu, L.Z. Wang, ACS Nano 5 (2011) 3483.
[36] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339.
[37] C. Nethravathi, M. Rajamathi, Carbon 46 (2008) 1994.
[38] T.F. Yeh, J.M. Syu, C. Cheng, T.H. Chang, H. Teng, Adv. Funct. Mater. 20 (2010)
2255.
[39] Y.X. Xu, H. Bai, G.W. Lu, C. Li, G.Q. Shi, J. Am. Chem. Soc. 130 (2008) 5856.
[40] J.L. Wu, S. Bai, X.P. Shen, L. Jiang, Appl. Surf. Sci. 257 (2010) 747.
[41] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia,
Carbon 45 (2007) 1558.
[42] X.Q. An, J.C. Yu, RSC Adv. 1 (2011) 1426.
[43] K.N. Kudin, B. Ozbas, H.C. Schniepp, R.K. Prudhomme, I.A. Aksay, R. Car, Nano
Lett. 8 (2008) 36.
[44] A.C. Ferrari, J. Robertson, Phys. Rev. B 61 (2000) 14095.
[45] D. Graf, F. Molitor, K. Ensslin, C. Stampfer, Nano Lett. 7 (2007) 238.
[46] F. Tuinstra, J.L. Koenig, J. Chem. Phys. 53 (1970) 1126.
[47] Y.H. Zhang, Z.R. Tang, X.Z. Fu, Y.J. Xu, ACS Nano 4 (2010) 7303.
[48] X.M. Tu, S.L. Luo, G.X. Chen, J.H. Li, Chem. Eur. J. 18 (2012) 14359.
[49] J. Henle, P. Simon, A. Frenzel, S. Scholz, S. Kaskel, Chem. Mater. 19 (2007) 366.
[50] M.A. Butler, D.S. Ginley, J. Electrochem. Soc. 125 (1978) 228.
[51] X.P. Lin, J.C. Xing, W.D. Wang, Z.C. Shan, F.F. Xu, F.Q. Huang, J. Phys. Chem. C
111 (2007) 18288.
[52] M. Iwamoto, H. Furukawa, K. Matsukami, T. Takenaka, S. Kagawa, J. Am. Chem.
Soc. 105 (1983) 3719.
[53] Z.H. Ai, W. Ho, S. Lee, J. Phys. Chem. C 115 (2011) 25330.