Research Article

www.acsami.org

Synthesis of Fullerene−, Carbon Nanotube−, and Graphene−TiO2

Nanocomposite Photocatalysts for Selective Oxidation: A
Comparative Study
Min-Quan Yang, Nan Zhang, and Yi-Jun Xu*
State Key Laboratory Breeding Base of Photocatalysis, College of Chemistry and Chemical Engineering, Fuzhou University, Fuzhou,
350002, P.R. China
*
S Supporting Information

ABSTRACT: A series of TiO2−graphene (GR), −carbon nanotube (CNT),

and −fullerene (C60) nanocomposite photocatalysts with different weight
addition ratios of carbon contents are synthesized via a combination of sol−gel
and hydrothermal methods. Their structures and properties are determined by
the X-ray diffraction (XRD), UV−vis diffuse reflectance spectra (DRS),
transmission electron microscopy (TEM), nitrogen adsorption−desorption,
and photoelectrochemical measurements. Photocatalytic selective oxidation of
benzyl alcohol to benzaldehyde is employed as a model reaction to evaluate the
photocatalytic activity of the TiO2−carbon (GR, CNT, and C60) nano-
composites under visible light irradiation. The results reveal that incorporating
TiO2 with carbon materials can extend the adsorption edge of all the TiO2−
carbon nanocomposites to the visible light region. For TiO2−GR, TiO2−CNT,
and TiO2−C60 nanocomposites, the photocatalytic activities of the composites
with optimum ratios, TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−1.0% C60, are very close to each other along with the
irradiation time. Furthermore, the underlying reaction mechanism for the photocatalytic selective oxidation of benzyl alcohol to
benzaldehyde over TiO2−carbon nanocomposites has been explored using different radical scavenger techniques, suggesting that
TiO2−carbon photocatalysts follow the analogous oxidation mechanism toward selective oxidation of benzyl alcohol. The
addition of different carbon materials has no significant influence on the crystal phase, particle size, and the morphology of TiO2.
Therefore, it can be concluded, at least for nanocomposites of TiO2−carbon (GR, CNT, and C60) obtained by the present
approach, that there is no much difference in essence on affecting the photocatalytic performance of semiconductor TiO2 among
these three different carbon allotropes, GR, CNT, and C60. Our findings point to the importance of a comparative study of
semiconductor−carbon photocatalysts on drawing a relatively objective conclusion rather than separately emphasizing the unique
role of GR and joining the graphene gold rush.
KEYWORDS: TiO2, graphene, fullerene, carbon nanotube, visible light irradiation, selective oxidation

1. INTRODUCTION Researchers have synthesized multifarious, versatile GR−

Graphene (GR), as a new allotrope of carbon, has attracted an semiconductor nanocomposites as photocatalysts for degrada-
enormous amount of interest from both theoretical and tion of pollutants (e.g., dyes, bacteria, and volatile organic
experimental scientists since its discovery in 2004 by Geim pollutants),25−31 selective organic transformations for synthesis
and co-workers.1 Due to its exceptional properties, such as of fine chemicals,32−36 and water splitting to H2.37−44 What is
excellent electron mobility,2,3 theoretically large surface area of notable from the reported literature is that nearly all of the
∼2600 m2/g,4 high thermal conductivity of ∼5000 W m−1 K−1, research works are inclined to highlight that the enhanced
and optical transparency,5,6 GR projects as a rapidly rising star photocatalytic activity of GR−semiconductor nanocomposites
on the horizon of materials science and condensed matter is aroused from the addition of GR having exceptional
physics.7,8 Thus far, GR−based nanocomposites have been properties. But if we dispassionately look back at the
widely explored in a myriad of fields, including biosensors, development history of the carbon family, this situation
nanoelectronics, intercalation materials, drug delivery, catalysis, seems to have ever happened, when zero-dimension fullerene
supercapacitors, and polymer composites.9−22 With regard to (C60) and one-dimensional carbon nanotube (CNT) first
the domain of photocatalysis, GR, the thinnest and the appeared,45,46 which we praised generously with kind words,
strongest material ever known in the universe,9 also catches
the eyes of researchers in this field and, indeed, promotes great Received: December 5, 2012
interest to synthesize GR−semiconductor nanocomposites as Accepted: January 16, 2013
photocatalysts for target applications.14,20,21,23−42 Published: January 16, 2013

© 2013 American Chemical Society 1156 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164
ACS Applied Materials & Interfaces Research Article

too. That is, the abundant C60−semiconductor and CNT− the possibility of deriving unique metastable structures at low
semiconductor nanomaterials with various morphologies as reaction temperatures. Using photocatalytic selective oxidation
photocatalysts have already been reported,47−58 and it has been of benzyl alcohol to benzaldehyde as a model reaction, the
well demonstrated that the addition of fullerene and carbon influences of the carbon types and their contents on
nanotubes is able to improve the photocatalytic performance of photocatalytic activity are discussed. Our results demonstrate
semiconductors, such as TiO2,49−51,54,59 very much similar to the significant influence of preparation methods on the
their allotrope GR. If we compare those C60−semiconductor photocatalytic performance of TiO2−carbon composites, and
and CNT−semiconductor photocatalysts with their counter- GR can not manifest its unique role as compared to its carbon
parts of GR−semiconductor, the following remarks can be allotropes. It is hoped that our research work could promote
easily found. The enhancement of photoactivity for all of the the more objective understanding on the analogy and difference
semiconductor−carbon (C60, CNT, and GR) nanocomposites of these three carbon allotropes, graphene, fullerene, and
is ascribed to the fact that incorporation of carbon contents into carbon nanotube on the rational synthesis and photoactivity
the matrix of semiconductors will increase the adsorptivity, the improvement of semiconductor−carbon composites, instead of
absorption capability in the visible light region, and the life span joining the graphene gold rush.
of photoexcited electron−hole pairs. In particular, most
research works state that C60, CNT, and GR all can act as an 2. EXPERIMENTAL SECTION
electron reservoir to trap photoexcited electrons from semi- 2.1. Preparation. Materials. Graphite powder, nitric acid (HNO3,
conductors, thereby improving the life span of electron−hole 65%), absolute ethanol (C 2 H6 O), benzyl alcohol (C 7 H 8 O),
pairs, which is always regarded as the most important factor benzaldehyde (C7H6O), ammonium oxalate ((NH4)2C2O4·H2O),
contributing to the enhancement of photoactivity of semi- and silver nitrate (AgNO3) are analytical grade; tert-butyl alcohol
conductor−carbon (C 60 , CNT, and GR) nanocompo- (C4H10O), benzoquinone (C6H4O2), and tetrabutyl titanate (Ti
sites.27,32,33,37,47−49,53 Thus, these three carbon allotropes of (OC4H9)4), purity ≥ 98.0%) are chemical pure. All of the above
C60, CNT, and GR in the carbon family seem much similar in chemicals were obtained from Sinopharm Chemical Reagent Co., Ltd.
(Shanghai, China) and used as received without further purification.
the aspect of improving photocatalytic performance of
Benzotrifluoride (BTF, purity > 99%) was supplied by Alfa Aesar
semiconductors. Furthermore, noting that GR is a two- China Co., Ltd. (Tianjin, China). Carbon nanotubes were purchased
dimensional sheet of sp2 hybridized carbon,15,60 its unique from Shenzhen Nanotech Port Co., Ltd., China. High-purity C60
extended honeycomb network can be viewed as a basic building (99.9%) was obtained from Yongxin Chemical Reagent Company
block for other carbon allotropes with different dimension- (Henan, China). Deionized (DI) water used in the synthesis was from
alities, including the wrapped zero-dimension buckyballs local sources.
(fullerene) and the rolled one-dimension carbon nanotubes Synthesis. (a) Synthesis of Graphene Oxide (GO). Graphene oxide
(CNTs).7,10 (GO) was synthesized from natural graphite powder by a modified
Thus, it is natural to raise such fundamental questions as the Hummers’ method, as also reported in our previous research
works.26,32,63
following. Since C60, CNT, and GR have many similar structure (b) Treatment of Fullerenes (C60) and Carbon Nanotubes (CNT).
and electronic properties in common, are they similar in The purification and surface functionalization of C60 and CNTs were
improving the photocatalytic performance of semiconductors carried out before used for nanocomposites. A 50 mg portion of raw
when we use them to assemble carbon−semiconductor C60 was refluxed in 150 mL of concentrated nitric acid at 140 °C for 4
composite photocatalysts? Without a basic comparison study h. Then, the dark brown solid was collected by centrifugation and
between composite photocatalysts of GR−semiconductor, washed with DI water several times until pH = 7. After that, the
CNT−semiconductor, and C60−semiconductor, are we rational product was dried at 60 °C in an oven. The CNTs used here were
to claim that the enhancement of photoactivity of GR− multiwalled carbon nanotubes, which were treated by the same
semiconductor is due to the unique and excellent electron procedure.
(c) Fabrication of TiO2−Carbon (GR, CNT, C60) Nanocomposites.
conductivity of GR which prolongs the life span of photo- The preparation of TiO2−carbon nanocomposite photocatalysts is
excited electron−hole pairs significantly? In other words, do we outlined as follows. The weight addition ratios of carbon are selected
give incomplete or exaggerated information on the contribution as 0.1%, 0.5%, 1%, 5%, 10%, and 20%. A certain amount of carbon
role of GR to enhance the semiconductor photocatalytic materials was sonicated in a mixed solution of 9 mL ethanol and 18
activity, as compared to its carbon allotropes, fullerene, and mL DI water. The ultrasonic time should be long enough to ensure the
carbon nanotube?25,26 thorough dispersion of carbon materials. Then, 1.7 mL tetrabutyl
Bearing these questions in mind, an integrated and titanate (TBOT) was mixed with 9 mL ethanol and added dropwise to
comparison study which is still lacking in this field has been the above solution of carbon materials with magnetic stirring. After
stirring for 3 h, the suspension was transferred into a 50 mL Teflon-
carried out in this work. By taking the mostly studied TiO2
lined autoclave and conducted hydrothermal treatment at 180 °C for
semiconductor as an example, we have synthesized a series of 12 h. The hydrothermal process is able to make the sufficient
TiO2−carbon (C60, CNT, and GR) composite photocatalysts reduction of GO to GR.25,26,32 The precipitates thus obtained were
with different weight addition ratios of carbon contents using centrifuged and washed with DI water until the pH of the supernatant
the same sol−gel approach to guarantee the good interfacial was neutral and followed by a rinse of ethanol. After that, the
contact between TiO2 and carbon ingredients. The sol−gel sediments samples were dried at 60 °C in an electric oven.
processing is one of the most common methods to produce 2.2. Characterization. The phase composition of the samples
nanocomposite photocatalysts, and it allows compositional and were determined by a Bruker D8 Advance X-ray diffractometer (XRD)
microstructural tailoring through controlling the precursor at 40 kV and 40 mA with Ni-filtered Cu Kα radiation in the 2θ range
from 10° to 80° with a scan rate of 0.02° per second. UV−vis diffuse
chemistry and processing conditions;61 the approach makes it reflectance spectra (DRS) were recorded on a Cary-500 UV−vis−NIR
possible to control a number of determining parameters of the spectrometer in which BaSO4 powder was used as the internal
final product such as homogeneity, purity, and microstructure standard. Nitrogen adsorption−desorption isotherms and the
(in particular porosity and surface area).62 Furthermore, the Brunauer−Emmett−Teller (BET) surface areas were collected at 77
sol−gel approach provides excellent chemical homogeneity and K using Micromeritics ASAP2010 equipment. Transmission electron

1157 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164

ACS Applied Materials & Interfaces Research Article

microscopy (TEM) and high-resolution transmission electron

microscopy (HRTEM) images were obtained using a JEOL model
JEM 2010 EX instrument at an accelerating voltage of 200 kV.
Photoelectrochemical measurements were performed in a homemade
three electrode quartz cells with a PAR VMP3Multi potentiostat
apparatus. Pt plate was used as the counter electrode, and the Ag/AgCl
electrode was used as the reference electrode, while the working
electrode was prepared on indium−tin oxide (ITO) conductor glass.
The sample powder (10 mg) was ultrasonicated in 1 mL of anhydrous
ethanol to disperse it evenly to get slurry. The slurry was spread onto
ITO glass whose side part was previously protected using Scotch tape.
The working electrode was dried overnight under ambient conditions.
A copper wire was connected to the side part of the working electrode
using a conductive tape. Uncoated parts of the electrode were isolated
with epoxy resin. The electrolyte was 0.2 M of aqueous Na2SO4
solution (pH = 6.8) without additive. The visible light irradiation
source was a 300 W Xe arc lamp system equipped with a UV cutoff
filter (λ > 400 nm).
2.3. Catalyst Activity. The photocatalytic selective oxidation of
benzyl alcohol was performed in a 10 mL Pyrex glass bottle that
contained a mixture of alcohol (0.1 mmol) and 8 mg catalyst was
dissolved in the solvent of benzotrifluoride (BTF) (1.5
mL).32,33,36,64−68 The BTF was saturated with pure molecular oxygen.
The reason for choosing BTF as solvent is because of its inertness to
oxidation and high solubility for molecular oxygen.69,70 The Pyrex glass
bottle was filled with molecular oxygen at a pressure of 0.1 MPa and
stirred for half an hour to make the catalyst blend evenly in the
solution. Then, the suspensions were irradiated with a 300 W Xe arc
lamp (PLS-SXE 300C, Beijing Perfectlight Co., Ltd.) with a UV-CUT
filter to cut off light of wavelength < 400 nm. After the reaction, the
mixture was centrifuged at 12000 rmp for 10 min to remove the
catalyst particles thoroughly. The supernatant was analyzed with an
Agilent Gas Chromatograph (GC-7820 fitted with a capillary FFAP
analysis column). Controlled photoactivity experiments using different
radical scavengers (tert-butyl alcohol as scavenger for hydroxyl
radicals,71,72 ammonium oxalate as scavenger for photogenerated
holes,73 silver nitrate as scavenger for electrons,74,75 and benzoquinone
as scavenger for superoxide radical species72,76) were performed
similar to the above photocatalytic oxidation of benzyl alcohol except
that the radical scavengers (0.1 mmol) were added to the reaction
system. Conversion of alcohol, yield of aldehyde, and selectivity for
aldehyde were defined as follows.
conversion (%) = [(C0 − Calcohol)/C0] × 100
Figure 1. XRD patterns of TiO2−GR nanocomposites (a) and their
yield (%) = Caldehyde/C0 × 100 analogues TiO2−CNT (b) and TiO2−C60 nanocomposites (c).

selectivity (%) = [Caldehyde/(C0 − Calcohol)] × 100 the one hand, the weight addition ratios of carbon materials in
the nanocomposites are relatively low. On the other hand, the
Where, C0 is the initial concentration of alcohol, and Calcohol and main characteristic peaks of GR at 25.0° and CNT at 26.2° are
Caldehyde are the concentration of benzyl alcohol and benzaldehyde at a probably shadowed by the (101) peak at 25.3° of anatase TiO2,
certain time after the photocatalytic reaction, respectively. which is consistent with the previous reports.26,32 As shown in
Figure 1c, when the weight addition ratio of C60 is low in the
3. RESULTS AND DISCUSSION nanocomposites, there is no obvious diffraction peaks of C60
The XRD patterns of the as-prepared TiO2−carbon composites and the XRD patterns are similar with the analogues GR−TiO2
are shown in Figure 1. It is obvious that all of the TiO2−carbon and TiO2−CNT nanocomposites. However, as the weight
nanocomposites including TiO2−GR nanocomposites and their addition ratios of C60 reach 20%, apparently, two new peaks
analogues TiO2−CNT and TiO2−C60, exhibit similar XRD located at 17.7° and 20.8° are present which can be indexed to
patterns. The diffraction peaks for all samples match well with the (220) and (311) crystal planes of Buckminster full-
the anatase TiO2 (JCPDS No. 21-1272). There are just some erene.49,77
slight differences in the XRD patterns with the different carbon Figure 2 displays the UV−vis diffuse reflectance spectra
material. In comparison with the standard card of anatase TiO2, (DRS) of the as-obtained TiO2−carbon nanocomposites. It can
it is easy to see that the kind of carbon materials and their be seen clearly that the addition of GR, CNT, or C60 all induce
weight addition ratio in the TiO2−carbon nanocomposites have the significant increased light absorption intensity in the visible
no obvious influence on the characteristic peaks of TiO2. In light region. The continuous absorption band in the range of
Figure 1a and b, it can be found that no typical diffraction peaks 400−800 nm is caused by the addition of carbon materials.
of GR and CNT are observed in the corresponding Though the change of the kind of carbon material affects the
nanocomposites, which can be ascribed to two reasons. On shape of the absorption curve, which results from the
1158 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164
ACS Applied Materials & Interfaces Research Article

nanocomposites of TiO2−GR, TiO2−CNT, and TiO2−C60

effectively.
To understand the difference of GR, CNT, and C60 on
enhancing the photocatalytic activity of TiO2, we choose
photocatalytic selective oxidation of benzyl alcohol to
benzaldehyde as a model reaction.32 Taking a view of the
results as summarized in Figure 3a−c, we can see that, for the
three different series of TiO2−carbon nanocomposites, the
optimum ingredient ratios of TiO2−carbon always exist and
differ in the kind of carbon materials. According to their
photocatalytic performance for selective oxidation of benzyl
alcohol to benzaldehyde under visible light irradiation, for
TiO2−GR, TiO2−CNT, and TiO2−C60, the optimum nano-
composites are TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−
1.0% C60, respectively. For these three optimum nano-
composites, the conversions of benzyl alcohol are all close to
each other (ca. 40%), along with the selectivity higher than
95%. The photocatalytic selective oxidation of benzyl alcohol
over the optimized nanocomposites as a function of time under
visible light irradiation are also shown in Figure 3d−f. It is
found that, for TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−
1.0% C60 nanocomposites, there is no significant difference in
the conversion of benzyl alcohol and the selectivity for
benzaldehyde along with the reaction time of 2, 4, 6, and 8
h. The similar phenomenon is also observed for selective
oxidation of other substituted benzylic alcohols, i.e. no
significant difference of photocatalytic performance among
these three optimum TiO2−carbon composite photocatalysts
under visible light irradiation (Supporting Information Figure
S2). In addition, it should be noted that the photoactivity of
optimum TiO2−0.1% GR and TiO2−0.5% CNT is higher than
the optimum TiO2−5% GR and TiO2−5% CNT that were
prepared from the hydrolysis of TiF4 along with hydrothermal
post-treatment process.32 In the previous work of our group,
the photocatalyst of TiO2−GR with intimate interfacial contact
between GR and TiO2 exhibits significantly enhanced photo-
catalytic activities as compared to TiO2−CNT with poor
interfacial contact. It has been proved that the interfacial
contact between carbon and semiconductor is an important
factor, which affects the photocatalytic activities of the carbon−
semiconductor composite photocatalysts.32 Thus, in order to
obtain a relatively objective and rational comparison among the
three different carbon materials on affecting the photoactivity of
Figure 2. UV−vis diffuse reflectance spectra of TiO2−GR nano- semiconductor, similar interfacial contact should be considered.
composites (a) and their analogues TiO2−CNT (b) and TiO2−C60 In the present work, by employing the combination of sol−gel
nanocomposites (c). and hydrothermal method, we have prepared a series of TiO2−
GR, TiO2−CNT, and TiO2−C60 nanocomposites with different
weight addition ratios of carbon contents, which all have a good
dissimilarity in the natural optical properties of different carbon interfacial contact, ensuring that the comparison study among
materials as reflected in the DRS of GR, CNT, and C60 GR, CNT, and C60 is performed in a reasonable framework. In
(Supporting Information Figure S1), the rhythmicity stays the addition, as compared with our previous report, the obvious
same. For all the carbon materials, the light absorption intensity difference in photoactivity clearly implies that the preparation
in the visible light region increases accompanied with the methods play a significant effect on the synergetic interaction
augment of the addition amount of carbon materials. between semiconductor TiO2 and carbon contents and, thus,
Furthermore, a qualitative red shift to higher wavelength is different photocatalytic performance.25,32
observed in the absorption edge of all TiO2−GR, TiO2−CNT, To further obtain the microscopic structure information of
and TiO2−C60 nanocomposites, which can be attributed to the carbon−TiO2 nanocomposites and study the influence of
electronic interactions between GR, CNT, or C60 and the carbon materials on TiO2 morphology, transmission
TiO2.26,48,78 Such an extended optical absorption has also electron microscopy (TEM) analysis has been carried out, as
been observed in previous research works regarding GR−, displayed in Figure 4. It can be seen from Figure 4a and b that
CNT−, and C60−semiconductor nanocomposites.26,28,32,47−50 GR nanosheets and carbon nanotubes are covered with TiO2
Therefore, the introduction of GR, CNT, or C60 into the matrix nanoparticles. The energy dispersive X-ray spectrum (EDX)
of TiO2 is able to promote the visible light response of the was employed to prove the existence of GR in TiO2−0.1% GR.
1159 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164
ACS Applied Materials & Interfaces Research Article

Figure 3. Selective oxidation of benzyl alcohol to benzaldehyde over the nanocomposites of TiO2−GR (a), TiO2−CNT (b), and TiO2−C60 (c) with
different weight addition ratios of GR, CNT, and C60, respectively, under visible light irradiation of 4 h; time-online profiles of conversion, yield, and
selectivity over the optimal TiO2−0.1% GR (d), TiO2−0.5% CNT (e), and TiO2−1.0% C60 (f) nanocomposites.

As displayed in Figure S3 (Supporting Information), the result issue, the transient photocurrent responses of TiO2−0.1% GR,
of EDX gives the signals of C, O, Cu, and Ti elements. Because TiO2−0.5% CNT, and TiO2−1.0% C60 have been investigated
the lacey support film without carbon coating is used, the signal under intermittent visible light illumination with the wave-
of C must come from the GR sheet in the nanocomposites, length range used in the photocatalytic reactions, and the
which confirms the composition of the sample. As for TiO2− results are showed in Figure 5. Because it is well-known that
C60 nanocomposites, it is easy to observe from Figure 4c that a TiO2 has negligible photocurrent under visible light irradi-
coverture layer with amorphous structure covers the surface of ation,32 it is easy to observe that the addition of different carbon
the TiO2 nanoparticles. The thickness of the coverture layer ingredients all can enhance the photocurrent significantly for
was estimated to be 1 nm, close to the size of the C60 molecule TiO2−carbon photocatalysts under visible light irradiation and
(0.71 nm).79 Therefore, it can be estimated that the outer layer the photocurrent rapidly decreases to zero as long as the light is
is C60, which is dispersed on the surface of TiO2 with a switched off. The photocurrent is formed mainly by the
monolayer structure and this is in accordance with the previous diffusion of the photogenerated electrons to the back contact,
wok.47,49 Watching all the TEM images, we can see that there is and meanwhile, the photoinduced holes are taken up by the
no obvious influence of carbon addition on the morphology hole acceptor in the electrolyte.82 Therefore, the enhanced
and particle size of TiO2 nanoparticles, regardless of what kinds photocurrent over TiO2−carbon nanocomposites indicates a
of carbon ingredients was used to combine with TiO2. For the more efficient separation of the photoexcited electron−hole
three optimal ingredient ratios, TiO2 all displays the similar pairs and longer lifetime of the photogenerated charge carriers.
morphology; the particle shape and the size of the TiO2 Moreover, no obvious photocurrent decay is observed. This
nanoparticles in TiO2−carbon composites are all about 10 indicates that the transport of photogenerated electrons to
nm. The selected area electron diffraction (SAED) patterns as carbon materials is markedly effective. The adjacent and stable
displayed in the insets of Figure 4 indicate that the TiO2 in the photocurrent of TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−
nanocomposites possesses the polycrystalline structure, in 1.0% C60 nanocomposites highlight the similar role of GR,
agreement with the result of XRD analysis. In addition, from CNT, and C60 in prolonging the lifetime of photogenerated
the TEM analysis, it can be seen that the carbon ingredients electron−hole pairs. More importantly, there is no significant
and TiO2 have a good interfacial contact for the composites of difference on the ability of three carbon materials on
TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−1.0% C60. lengthening the lifetime of photogenerated electron−hole
It is known that the lifetime of photogenerated electron− pairs of TiO2−carbon nanocomposites.
hole pairs is a key factor determining the photocatalytic activity In addition, electrochemical impedance spectroscopy (EIS)
of carbon−semiconductor nanocomposites.80,81 Since the good Nyquist plots have also been carried out. As shown in Figure 6,
interfacial contact between the carbon materials and TiO2 is the Nyquist plots of TiO2−0.1% GR, TiO2−0.5% CNT, and
observed, is there significant difference in the roles of fullerene, TiO2−1.0% C60 nanocomposites electrode materials cycled in
carbon nanotube, and graphene on lengthening the lifetime of 0.2 M Na2SO4 electrolyte solution all show semicycles at high
photogenerated electron−hole pairs? To address the above frequencies. Since the preparation of the electrodes and
1160 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164
ACS Applied Materials & Interfaces Research Article

Figure 5. Transient photocurrent response of TiO2−0.1% GR, TiO2−

0.5% CNT, and TiO2−1.0% C60 nanocomposites in 0.2 M Na2SO4
aqueous solution without bias versus Ag/AgCl under the irradiation of
visible light.

Figure 6. Electrochemical impedance spectroscopy (EIS) Nyquist

plots of TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−1.0% C60
nanocomposites.

spectra, the high-frequency arc corresponds to the charge

transfer limiting process and can be attributed to the double-
layer capacitance in parallel with the charge transfer resistance
at the contact interface between electrode and electrolyte
solution.84 It can be seen from Figure 6 that the change of
carbon materials leads to little difference in the EIS Nyquist
plots; on the other hand, the charge transfer resistance can be
directly measured as the semicircle diameter. So, the nearly
overlapped plots of TiO2−0.1% GR and TiO2−1.0% C60 mean
a similar separation of photogenerated electron-hole pairs, and
the interfacial charge transfer to the electron donor/electron
acceptor of them are a bit faster than TiO2−0.5% CNT
nanocomposites, which is consistent with the results of
photocurrent test.
In order to explore the influence of GR, CNT, and C60 on
surface area and porosity of TiO2−carbon nanocomposites, and
thus understand the effect of different carbon materials on the
photocatalytic performance, the surface area and porosity of
TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2−1.0% C60 have
been investigated, as displayed in Figure 7. According to the
IUPAC classification,85 it can be seen that all these three
nanocomposites display a type IV isotherm with a typical H3
Figure 4. TEM images of TiO2−0.1% GR (a), TiO2−0.5% CNT (b), hysteresis loop characteristic of mesoporous solids, which is
and TiO2−1.0% C60 (c). confirmed by the corresponding pore size distribution as shown
in the inset of Figure 7. The similar shapes of their hysteresis
electrolyte used are alike, the high-frequency semicircles are loops also indicate the similar pore shapes. The specific
related to the resistance of the electrodes.83 In electrochemical Brunauer−Emmett−Teller (BET) surface area and pore
1161 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164
ACS Applied Materials & Interfaces Research Article

Figure 7. BET adsorption−desorption isotherm of TiO2−0.1% GR, Figure 9. Controlled experiments using different radical scavengers for
TiO2−0.5% CNT, and TiO2−1.0% C60 nanocomposites. (inset) the photocatalytic selective oxidation of benzyl alcohol over TiO2−
Corresponding pore size distribution. carbon nanocomposites in the BTF solvent: reaction with tert-butyl
alcohol (TBA) as the radical scavenger for hydroxyl radicals, reaction
volume are 174 m2/g and 0.35 cm3/g for TiO2−0.1% GR, 166 with ammonium oxalate (AO) as scavenger for photogenerated holes,
m2/g and 0.35 cm3/g for TiO2−0.5% CNT, 165 m2/g, 0.36 reaction with benzoquinone (BQ) as scavenger for superoxide radicals,
cm3/g for TiO2−1.0% C60. It is clear that they are very close to and reaction with AgNO3 as scavenger for photogenerated electrons
each other. This is reasonable because, with such a small under visible light irradiation for 4 h.
amount doping of carbon, the surface area and porosity are
mainly dominated by TiO2 ingredients. In addition, adsorption
experiments in the dark for benzyl alcohol also have been added into the reaction system,71,73 although the conversion of
performed. As displayed in Figure 8, the results suggest that benzyl alcohol has a measurable decrease, a moderate
conversion of benzyl alcohol can still be achieved. The
photocatalytic conversion almost falls by half when the radical
scavenger, benzoquinone (BQ), for superoxide radical species
(O2·−) is added into the reaction system.71,72,76 Besides, the
controlled experiment, using AgNO3 as the radical scavenger
for electrons (e−),71,74,75 shows that the conversion of benzyl
alcohol is significantly declined to about 10%. These results
clearly suggest that the photocatalytic selective oxidation of
benzyl alcohol to benzaldehyde over the as-prepared TiO2−
carbon photocatalysts is intimate with the photogenerated
electron−hole pairs and the superoxide radical species (O2·−).
In other words, photogenerated holes, electrons, and super-
oxide radicals are the primary active species for photocatalytic
selective oxidation of benzyl alcohol. In addition, it should be
noted that the photocatalytic experiments are performed under
Figure 8. Remaining fraction of benzyl alcohol after the adsorption− oxygen-saturated condition and the present molecular oxygen
desorption equilibrium is achieved over TiO2−0.1% GR, TiO2−0.5% (O2) can act as electron-acceptors by which oxygen is activated
CNT, and TiO2−1.0% C60 nanocomposites. and the recombination of electron−hole pairs is inhibited.71,86
Summing up the above discussion, we can propose for the
there is no obvious difference of adsorptivity among the three series of TiO2−carbon (GR, CNT, and C60) photocatalysts that
different TiO2−carbon nanocomposites, and this case is also they follow the analogous tentative reaction mechanism toward
observed for other benzylic alcohols (Supporting Information selective oxidation of benzyl alcohol in the BTF solvent under
Figure S4). visible light irradiation.
To further understand the underlying reaction mechanism The present work suggests that the photocatalytic perform-
for the photocatalytic selective oxidation of benzyl alcohol over ance of TiO2−carbon is significantly affected by the preparation
the as-prepared TiO2−carbon photocatalysts, a series of methods. The difference in preparation methods causes the
controlled experiments with addition of different scavengers different structural composition and synergetic interaction
for the photogenerated radical species have been imposed on between TiO2 and carbon, which thus influences the photo-
the oxidation process.71 As shown in Figure 9, when the catalytic performance of TiO2−carbon composites. Thus,
trapping agent of tert-butyl alcohol (TBA) as the radical although GR is more popular than its forebears (CNT and
scavenger for hydroxyl radicals (·OH) is added to the BTF C60) at present with regard to synthesis and application of
dispersions of the three optimum nanocomposites,71,72 semiconductor−carbon composite photocatalysts, it is still too
compared with the original experiments without the radical early to draw a definitely decisive answer for GR’s unique
scavengers, there is almost no change on the conversion of superiority to other carbon allotropes on improving the
benzyl alcohol. This observation is reasonable because, in the photocatalytic performance of semiconductor. More efforts
BTF solvent, no ·OH radicals are formed.32,33,65−67,71 When should be keenly required to understand the role and
the quencher of ammonium oxalate (AO) for holes (h+) is mechanism of GR on affecting the photocatalytic properties
1162 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164
ACS Applied Materials & Interfaces Research Article

of GR−semiconductor composite photocatalysts, instead of tive Research Team in Universities (PCSIRT0818), Program
joining the GR gold rush.25 for Returned High-Level Overseas Chinese Scholars of Fujian

■ CONCLUSIONS
In summary, we have prepared a series of TiO2−carbon (GR,
Province, and the Project Sponsored by the Scientific Research
Foundation for the Returned Overseas Chinese Scholars, State
Education Ministry, is gratefully acknowledged.

■
CNT, and C60) nanocomposites with different weight addition
ratios of carbon materials by a sol−gel process along with REFERENCES
hydrothermal post-treatment. By using the photocatalytic
(1) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang,
oxidation of benzyl alcohol to benzaldehyde as a testing Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Science 2004, 306,
reaction, we have investigated the photocatalytic performance 666−669.
of the synthesized TiO2−carbon composites. The results have (2) Bolotin, K. I.; Sikes, K. J.; Jiang, Z.; Klima, M.; Fudenberg, G.;
demonstrated that for the TiO2−GR, TiO2−CNT, and TiO2− Hone, J.; Kim, P.; Stormer, H. L. Solid State Commun. 2008, 146, 351−
C60 nanocomposites, the optimal ingredient ratios exist at low 355.
weight additions, and their photoactivities are rather similar. (3) Morozov, S. V.; Novoselov, K. S.; Katsnelson, M. I.; Schedin, F.;
The photocatalytic selective oxidation of benzyl alcohol over Elias, D. C.; Jaszczak, J. A.; Geim, A. K. Phys. Rev. Lett. 2008, 100,
the TiO2−carbon nanocomposites with different carbon 016602.
materials follows the analogous reaction mechanism. In (4) Stoller, M. D.; Park, S.; Zhu, Y.; An, J.; Ruoff, R. S. Nano Lett.
addition, the introduction of carbon materials makes little 2008, 8, 3498−3502.
difference in the crystal phase, particle size, surface area, pore (5) Balandin, A. A.; Ghosh, S.; Bao, W.; Calizo, I.; Teweldebrhan, D.;
volume, and the morphology of TiO2. The addition of GR, Miao, F.; Lau, C. N. Nano Lett. 2008, 8, 902−907.
(6) Li, X.; Zhu, Y.; Cai, W.; Borysiak, M.; Han, B.; Chen, D.; Piner, R.
CNT, and C60 all can induce the increased light absorption
D.; Colombo, L.; Ruoff, R. S. Nano Lett. 2009, 9, 4359−4363.
intensity in visible light region and promote the visible light (7) Geim, A. K.; Novoselov, K. S. Nat. Mater. 2007, 6, 183−191.
response of the nanocomposites of TiO2−GR, TiO2−CNT, (8) Katsnelson, M. I.; Novoselov, K. S. Solid State Commun. 2007,
and TiO2−C60 effectively; they can also promote an efficient 143, 3−13.
separation of the photoexcited electron−hole pairs in a similar (9) Geim, A. K. Science 2009, 324, 1530−1534.
way. Thus, we do not observe the significant difference of GR (10) Allen, M. J.; Tung, V. C.; Kaner, R. B. Chem. Rev. 2009, 110,
on improving the photoactivity of TiO2 as compared to CNT 132−145.
and C60. It is hoped that this work could promote the more (11) Rao, C. N. R.; Sood, A. K.; Subrahmanyam, K. S.; Govindaraj, A.
objective understanding on the analogy and difference of these Angew. Chem., Int. Ed. 2009, 48, 7752−7777.
three carbon allotropes on the rational synthesis and photo- (12) Guo, S.; Dong, S. Chem. Soc. Rev. 2011, 40, 2644−2672.
activity improvement of semiconductor−carbon composites for (13) Bai, H.; Li, C.; Shi, G. Adv. Mater. 2011, 23, 1089−1115.
target applications in heterogeneous photocatalysis. Research in (14) Kamat, P. V. J. Phys. Chem. Lett. 2009, 1, 520−527.
(15) Stankovich, S.; Dikin, D. A.; Dommett, G. H. B.; Kohlhaas, K.
this respect would significantly advance how the remarkable M.; Zimney, E. J.; Stach, E. A.; Piner, R. D.; Nguyen, S. T.; Ruoff, R. S.
properties of GR could best be utilized to design the unique, Nature 2006, 442, 282−286.
more efficient GR-based composite photocatalysts for specific (16) Li, X.; Wang, X.; Zhang, L.; Lee, S.; Dai, H. Science 2008, 319,
applications.

■
1229−1232.
(17) Compton, O. C.; Nguyen, S. T. Small 2010, 6, 711−723.
ASSOCIATED CONTENT (18) Wang, X.; Zhi, L.; Mullen, K. Nano Lett. 2007, 8, 323−327.
*
S Supporting Information (19) Machado, B. F.; Serp, P. Catal. Sci. Technol. 2012, 2, 54−75.
UV−vis diffuse reflectance spectra of GR, CNT, and C60, (20) An, X.; Yu, J. C. RSC Adv. 2011, 1, 1426−1434.
(21) Xiang, Q.; Yu, J.; Jaroniec, M. Chem. Soc. Rev. 2012, 41, 782−
selective oxidation of substituted benzylic alcohols to aldehydes
796.
over the optimal TiO2−0.1% GR, TiO2−0.5% CNT, and (22) Liang, Y. T.; Hersam, M. C. J. Am. Chem. Soc. 2010, 132,
TiO2−1.0% C60 nanocomposites under the irradiation of visible 17661−17663.
light; energy dispersive X-ray (EDX) image of the TiO2−0.1% (23) Ng, Y. H.; Lightcap, I. V.; Goodwin, K.; Matsumura, M.; Kamat,
GR nanocomposite; remaining fraction of substituted benzylic P. V. J. Phys. Chem. Lett. 2010, 1, 2222−2227.
alcohols after the adsorption−desorption equilibrium is (24) Bell, N. J.; Ng, Y. H.; Du, A.; Coster, H.; Smith, S. C.; Amal, R. J.
achieved over TiO2−0.1% GR, TiO2−0.5% CNT, and TiO2− Phys. Chem. C 2011, 115, 6004−6009.
1.0% C60 nanocomposites. This material is available free of (25) Zhang, N.; Zhang, Y.; Xu, Y.-J. Nanoscale. 2012, 4, 5792−5813.
charge via the Internet at http://pubs.acs.org. (26) Zhang, Y.; Tang, Z.-R.; Fu, X.; Xu, Y.-J. ACS Nano 2010, 4,

■
7303−7314.
AUTHOR INFORMATION (27) Chen, C.; Cai, W.; Long, M.; Zhou, B.; Wu, Y.; Wu, D.; Feng, Y.
ACS Nano 2010, 4, 6425−6432.
Corresponding Author (28) Zhao, D.; Sheng, G.; Chen, C.; Wang, X. Appl. Catal., B 2012,
*Tel./fax: +86 591 83779326. E-mail: yjxu@fzu.edu.cn. 111−112, 303−308.
Notes (29) Zhang, J.; Xiong, Z.; Zhao, X. S. J. Mater. Chem. 2011, 21,
The authors declare no competing financial interest. 3634−3640.

■
(30) Akhavan, O.; Ghaderi, E. J. Phys. Chem. C 2009, 113, 20214−
ACKNOWLEDGMENTS 20220.
(31) Ai, Z.; Ho, W.; Lee, S. J. Phys. Chem. C 2011, 115, 25330−
The support by the National Natural Science Foundation of 25337.
China (NSFC) (21173045, 20903023), the Award Program for (32) Zhang, Y.; Tang, Z.-R.; Fu, X.; Xu, Y.-J. ACS Nano 2011, 5,
Minjiang Scholar Professorship, the Natural Science Founda- 7426−7435.
tion of Fujian Province for Distinguished Investigator Grant (33) Zhang, N.; Zhang, Y.; Pan, X.; Fu, X.; Liu, S.; Xu, Y.-J. J. Phys.
(2012J06003), Program for Changjiang Scholars and Innova- Chem. C 2011, 115, 23501−23511.

1163 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164

ACS Applied Materials & Interfaces Research Article

(34) Li, X.-H.; Chen, J.-S.; Wang, X.; Sun, J.; Antonietti, M. J. Am. (71) Zhang, Y.; Zhang, N.; Tang, Z.-R.; Xu, Y.-J. Chem. Sci. 2012, 3,
Chem. Soc. 2011, 133, 8074−8077. 2812−2822.
(35) Liang, Y. T.; Vijayan, B. K.; Gray, K. A.; Hersam, M. C. Nano (72) Raja, P.; Bozzi, A.; Mansilla, H.; Kiwi, J. J. Photochem. Photobiol.,
Lett. 2011, 11, 2865−2870. A 2005, 169, 271−278.
(36) Zhang, N.; Zhang, Y.; Pan, X.; Yang, M.-Q.; Xu, Y.-J. J. Phys. (73) Carp, O.; Huisman, C. L.; Reller, A. Prog. Solid State Chem.
Chem. C 2012, 116, 18023−18031. 2004, 32, 33−177.
(37) Jia, L.; Wang, D.-H.; Huang, Y.-X.; Xu, A.-W.; Yu, H.-Q. J. Phys. (74) Liu, S.; Zhang, N.; Tang, Z.-R.; Xu, Y.-J. ACS Appl. Mater.
Chem. C 2011, 115, 11466−11473. Interfaces 2012, 4, 6378−6385.
(38) Xiang, Q.; Yu, J.; Jaroniec, M. J. Am. Chem. Soc. 2012, 134, (75) Primo, A.; Marino, T.; Corma, A.; Molinari, R.; García, H. J. Am.
6575−6578. Chem. Soc. 2011, 133, 6930−6933.
(39) Ye, A.; Fan, W.; Zhang, Q.; Deng, W.; Wang, Y. Catal. Sci. (76) Stylidi, M.; Kondarides, D. I.; Verykios, X. E. Appl. Catal., B
Technol. 2012, 2, 969−978. 2004, 47, 189−201.
(40) Mukherji, A.; Seger, B.; Lu, G. Q.; Wang, L. ACS Nano 2011, 5, (77) Apostolopoulou, V.; Vakros, J.; Kordulis, C.; Lycourghiotis, A.
3483−3492. Colloids Surf., A 2009, 349, 189−194.
(41) Iwase, A.; Ng, Y. H.; Ishiguro, Y.; Kudo, A.; Amal, R. J. Am. (78) Zhang, Y.; Zhang, N.; Tang, Z.-R.; Xu, Y.-J. ACS Nano 2012, 6,
Chem. Soc. 2011, 133, 11054−11057. 9777−9789.
(42) Ng, Y. H.; Iwase, A.; Kudo, A.; Amal, R. J. Phys. Chem. Lett. (79) Kratschmer, W.; Lamb, L. D.; Fostiropoulos, K.; Huffman, D. R.
2010, 1, 2607−2612. Nature 1990, 347, 354−358.
(43) Fan, W.; Lai, Q.; Zhang, Q.; Wang, Y. J. Phys. Chem. C 2011, (80) Woan, K.; Pyrgiotakis, G.; Sigmund, W. Adv. Mater. 2009, 21,
115, 10694−10701. 2233−2239.
(44) Fan, W.; Zhang, Q.; Wang, Y. Phys. Chem. Chem. Phys. 2012, (81) Leary, R.; Westwood, A. Carbon 2011, 49, 741−772.
DOI: 10.1039/C2CP43524A. (82) Soedergren, S.; Hagfeldt, A.; Olsson, J.; Lindquist, S.-E. J. Phys.
(45) Kroto, H. W.; Heath, J. R.; O’Brien, S. C.; Curl, R. F.; Smalley, Chem. 1994, 98, 5552−5556.
R. E. Nature 1985, 318, 162−163. (83) Wang, D.; Choi, D.; Li, J.; Yang, Z.; Nie, Z.; Kou, R.; Hu, D.;
(46) Iijima, S. Nature 1991, 354, 56−58. Wang, C.; Saraf, L. V.; Zhang, J.; Aksay, I. A.; Liu, J. ACS Nano 2009,
(47) Zhu, S.; Xu, T.; Fu, H.; Zhao, J.; Zhu, Y. Environ. Sci. Technol. 3, 907−914.
2007, 41, 6234−6239. (84) Lu, T.; Zhang, Y.; Li, H.; Pan, L.; Li, Y.; Sun, Z. Electrochim. Acta
(48) Fu, H.; Xu, T.; Zhu, S.; Zhu, Y. Environ. Sci. Technol. 2008, 42, 2010, 55, 4170−4173.
8064−8069. (85) Sing, K. S. W.; Everett, D. H.; Haul, R. A. W.; Moscou, L.;
(49) Yu, J.; Ma, T.; Liu, G.; Cheng, B. Dalton Trans. 2011, 40, 6635− Pierotti, R. A.; Rouquerol, J.; Siemieniewska, T. Pure Appl. Chem.
6644. 1985, 57, 603−619.
(50) Xu, Y.-J.; Zhuang, Y.; Fu, X. J. Phys. Chem. C 2010, 114, 2669− (86) Maldotti, A.; Molinari, A.; Amadelli, R. Chem. Rev. 2002, 102,
2676. 3811−3836.
(51) Lin, J.; Zong, R.; Zhou, M.; Zhu, Y. Appl. Catal., B 2009, 89,
425−431.
(52) Tang, Z.-R.; Li, F.; Zhang, Y.; Fu, X.; Xu, Y.-J. J. Phys. Chem. C
2011, 115, 7880−7886.
(53) Kim, Y. K.; Park, H. Energy Environ. Sci. 2011, 4, 685−694.
(54) Chen, W.; Fan, Z.; Zhang, B.; Ma, G.; Takanabe, K.; Zhang, X.;
Lai, Z. J. Am. Chem. Soc. 2011, 133, 14896−14899.
(55) Peng, T.; Zeng, P.; Ke, D.; Liu, X.; Zhang, X. Energy Fuels 2011,
25, 2203−2210.
(56) Yu, J.; Yang, B.; Cheng, B. Nanoscale 2012, 4, 2670−2677.
(57) Liang, Y. T.; Vijayan, B. K.; Lyandres, O.; Gray, K. A.; Hersam,
M. C. J. Phys. Chem. Lett. 2012, 3, 1760−1765.
(58) Kamat, P. V.; Gevaert, M.; Vinodgopal, K. J. Phys. Chem. B 1997,
101, 4422−4427.
(59) Eder, D.; Windle, A. H. J. Mater. Chem. 2008, 18, 2036−2043.
(60) Meyer, J. C.; Geim, A. K.; Katsnelson, M. I.; Novoselov, K. S.;
Booth, T. J.; Roth, S. Nature 2007, 446, 60−63.
(61) Wang, C.-C.; Ying, J. Y. Chem. Mater. 1999, 11, 3113−3120.
(62) Keshmiri, M.; Mohseni, M.; Troczynski, T. Appl. Catal., B 2004,
53, 209−219.
(63) Hummers, W. S.; Offeman, R. E. J. Am. Chem. Soc. 1958, 80,
1339−1339.
(64) Tang, Z.-R.; Zhang, Y.; Xu, Y.-J. ACS Appl. Mater. Interfaces
2012, 4, 1512−1520.
(65) Zhang, N.; Fu, X.; Xu, Y.-J. J. Mater. Chem. 2011, 21, 8152−
8158.
(66) Zhang, N.; Liu, S.; Fu, X.; Xu, Y.-J. J. Phys. Chem. C 2011, 115,
22901−22909.
(67) Zhang, N.; Liu, S.; Fu, X.; Xu, Y.-J. J. Mater. Chem. 2012, 22,
5042−5052.
(68) Zhang, Y.; Zhang, N.; Tang, Z.-R.; Xu, Y.-J. Phys. Chem. Chem.
Phys. 2012, 14, 9167−9175.
(69) Zhang, M.; Chen, C.; Ma, W.; Zhao, J. Angew. Chem., Int. Ed.
2008, 47, 9730−9733.
(70) Zhang, M.; Wang, Q.; Chen, C.; Zang, L.; Ma, W.; Zhao, J.
Angew. Chem., Int. Ed. 2009, 48, 6081−6084.

1164 dx.doi.org/10.1021/am3029798 | ACS Appl. Mater. Interfaces 2013, 5, 1156−1164