CARBON 5 0 ( 2 0 1 2 ) 2 4 7 2 –2 4 8 1

Available at www.sciencedirect.com

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

TiO2 nanoparticles loaded on graphene/carbon composite

nanofibers by electrospinning for increased photocatalysis

Chang Hyo Kim a, Bo-Hye Kim b,*

, Kap Seung Yang b,c,**

a
Department of Advanced Chemicals & Engineering, Chonnam National University, 300 Yongbong-dong, Gwangju, Republic of Korea
b
Alan G. MacDiarmid Energy Research Institute, Chonnam National University, 300 Yongbong-dong, Gwangju, Republic of Korea
c
Department of Polymer & Fiber System Engineering, Chonnam National University, 300 Yongbong-dong, Gwangju, Republic of Korea

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

Article history:
Received 29 October 2011
Accepted 26 January 2012
Available online 4 February 2012
Graphene/carbon composite nanofibers (CCNFs) with attached TiO2 nanoparticles (TiO2–
CCNF) were prepared, and their photocatalytic degradation ability under visible light irra-
diation was assessed. They were characterized using scanning and transmission electron
microscopy, X-ray diffraction, Raman spectroscopy, X-ray photoelectron spectroscopy,
and ultraviolet–visible diffuse spectroscopy. The results suggest that the presence of graph-
ene embedded in the composite fibers prevents TiO2 particle agglomeration and aids the
uniform dispersion of TiO2 on the fibers. In the photodegradation of methylene blue, a sig-
nificant increase in the reaction rate was observed with TiO2–CCNF materials under visible
light. This increase is due to the high migration efficiency of photoinduced electrons and
the inhibition of charge–carrier recombination due to the electronic interaction between
TiO2 and graphene. The TiO2–CCNF materials could be used for multiple degradation cycles
without a decrease in photocatalytic activity.
 2012 Elsevier Ltd. All rights reserved.

1. Introduction
TiO2 is one of the most promising catalysts because of its supe-
rior photocatalytic performance, easy availability, long-term
stability, and nontoxicity [1,2]. Typically, photoexcited elec-
tron–hole pairs can be generated by irradiation with light with
an energy greater than the band gap energy of TiO2
(Ebg = 3.2 eV for anatase). However, several problems can arise
when TiO2 is applied as a photocatalyst: (1) photogenerated
electron–hole pairs can recombine quickly, which affects the
photocatalytic efficiency [3–7], and (2) TiO2 can only be excited
with ultraviolet (UV) light, which is less than 5% of solar light,
because of its wide band gap [7–9]. These disadvantages of
TiO2 result in a low photocatalytic activity in practical applica-
tion. Therefore, a number of recent studies have focused on the
preparation and modification of TiO2 to narrow its band gap and
enhancethephotocatalyticactivityundervisiblelightradiation[10–30].
Among the carbon nanostructures (e.g., C60, carbon nano-
tubes, and graphenes), graphenes offer new opportunities in
photovoltaic conversion and photocatalysis by the hybrid
structures with a variety of nanomaterials, due to their excel-
lent charge carrier mobility, a large specific surface area, and
good electrical conductivity [31–36]. For example, TiO2 com-
bined with graphene acts as an electron trap, which promotes
electron–hole separation and facilitates interfacial electron
transfer [14,24–26,30]. Furthermore, graphene can help con-
trol the morphology of TiO2 nanoparticles because it controls
nucleation and growth of TiO2 nanoparticles and allows for
optimal chemical interactions and bonding between nanopar-
ticles and graphene.

* Corresponding author at : Alan G. MacDiarmid Energy Research Institute, Chonnam National University, Yong-Bong dong 300, 500-757
Gwangju, Republic of Korea. Tel.: +82 62 530 0774; fax: +82 62 530 1779.
** Corresponding author at : Department of Polymer&Fiber System Engineering, Chonnam National University, Yong-Bong dong 300,
500-757 Gwangju, Republic of Korea. Tel.: +82 62 530 0774; fax: +82 62 530 1779.
E-mail addresses: bohye@chonnam.ac.kr (B.-H. Kim), ksyang@chonnam.ac.kr (K.S. Yang).
0008-6223/$ - see front matter  2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbon.2012.01.069
 CARBON 5 0 ( 20 1 2 ) 2 4 7 2–24 8 1
In this work, carbon nanofibers (CNFs) containing
micropores were used to support TiO2 because CNFs exhibit
a high adsorption capability and adsorption rate due to the
shallow and uniform pore structure. Shallow micropores
open directly to the surface, which results in a large adsorp-
tion capacity and fast adsorption/desorption [37,38]. In addi-
tion, electrospinning has been widely used as a versatile
technique to fabricate various hybrid nanofibers [39,40]. How-
ever, reports have seldom been published on the fabrication
and evaluation of TiO2 nanoparticles loaded on graphene/car-
bon composite nanofibers (TiO2–CCNF) produced by electros-
pinning techniques.
In the present work, TiO2–CCNF catalysts were prepared by
electrospinning to determine if they exhibit improved photo-
catalytic efficiency in the visible spectrum, as illustrated in
Fig. 1. By increasing the photocatalytic activity of the TiO2–
CCNF materials, the current study aims to (i) extend the light
absorption spectrum into the visible region, (ii) reduce
electron/hole pair recombination, and (iii) enhance the
adsorption capability and high adsorption/desorption rate of
organic pollutants, which increases the reaction efficiency
because of the high specific surface area of the CNFs.
2. Experimental
2.1. Materials and methods
The graphenes used in this study were xGNP-C750-grade
materials produced by XG Science, USA. The elementary anal-
ysis of graphene was characterized as 88.68% carbon, 0.79%
hydrogen, 1.11% nitrogen, and 7.65% oxygen using the Mettler
method (Metler-Toledo AG, Switzerland). The graphene with
functional group can be dispersed in organic solvent. The
3 wt.% grpahene (3 wt.% relative to PAN) sample was im-
mersed in dimethylformamide (DMF) and sonicated in a
bath-type sonicator. The PAN (Mw = 150,000, Aldrich Chemical
Co.) was dissolved in the homogeneous graphene-DMF solu-
tion. This mixture was continuously stirred at 60 C until a
homogeneous solution formed, and it was then cooled to
room temperature. This solution was then spun into nanofi-
ber webs using an electrospinning apparatus (NTPS-35K,
Ntsse Co., Korea) operating at 25 kV. Spinning solutions were
fed through a capillary tip (diameter = 0.5 mm) using a syr-
inge (30 ml). The anode of the high voltage power supply
was clamped to a syringe needle tip and the cathode was con-
nected to a metal collector. During electrospinning, the dis-
Fig. 1 – TiO2–CCNF hybrids and its response under visible irradiation.
2473
tance between the tip and collector was 17 cm, and the flow
rate of the spinning solution was 3 ml/h. The electrospun fi-
bers were collected on aluminum foil wrapped on a metal
drum rotating at approximately 300 rpm. The electrospun fi-
ber webs were stabilized in flowing air at 280 C with a heat-
ing rate of 1 C/min for 1 h. A 30 mL ethanol solution
containing 2.04 g of titanium n-butoxide (Ti(OnBu)4, STREM
CHEMICALS) was diluted with 30 mL of toluene. The stabi-
lized nanofibers (SNF) were dipped in the prepared sol solu-
tion, washed sufficiently with ethanol and water, dried in
air, and carbonized at 800 C with a heating rate of 5 C/min
in N2 atmosphere with a flow rate of 200 mL/min. This sample
was identified as TiO2–CCNF. To compare the photocatalytic
activity under visible light irradiation, the TiO2 loaded on
PAN based carbon nanofiber (TiO2–CNF), Graphene/carbon
composite nanofibers (CCNF), and PAN based carbon nanofi-
ber (CNF) samples were synthesized.
2.2. Characterization
The surface structures of the TiO2–CCNF materials were char-
acterized by a Hitachi S-4700 field-emission scanning electron
microscope (FE-SEM) and energy dispersive X-ray spectros-
copy (EDX). The transmission electron micrographs (TEM)
and the selected area electron diffraction (SAED) micrographs
were taken with a TECHNAI-F20 unit (KJ111, Phillips) operat-
ing at 200 kV. The crystalline structure was characterized with
an X-ray diffraction unit (XRD, D-Max-2400 diffractometer)
equipped with graphite monochromatized CuKa radiation
(k = 0.15418 nm). Backscattering Raman measurements were
carried out with a Renishaw in via-Reflex using 633 nm at
room temperature. The chemical state of the surface was
examined by X-ray photoelectron spectroscopy (XPS) using a
VGScientific ESCALAB250 spectrometer equipped with a
monochromatized AlK X-ray source (15 mA, 14 kV). The
surface functionalities of the copolymer were examined by
Fourier Transform Infrared spectroscopy (FT-IR, Nicolet 200
instrument). All the samples were analyzed using the KBr pel-
let technique and scanned in the range from 4000 to 400 cm
1
. Specific surface areas were analyzed by the Brunauer–
Emmett–Teller (BET) method using a surface area analyzer
(Micrometrics, ASPS2020, USA). To observe the optical proper-
ties of the samples, the photoluminescence (PL) emission
spectra (Laser Raman and PL spectrometer, SPEX1403;
325 nm (55 mW) He–Cd Laser) were measured which give
information about the recombination rate of photogenerated
2474 CARBON 5 0 ( 2 0 1 2 ) 2 4 7 2 –2 4 8 1
carriers. The porosity was investigated from the nitrogen
adsorption isotherm at 77 K (ASAP 2020, Micromeritics,
USA). The specific surface area, the mesopore size distribu-
tion, and the micropore size distribution of the samples were
evaluated using the Brunauer–Emmett–Teller (BET) method
and the Barrett–Joyner–Halenda (BJH) method. Photocatalytic
efficiencies were evaluated by decolorizing an aqueous solu-
tion of methylene blue (MB) under irradiation by a fluorescent
lamp. A 0.1 g sample of the photocatalyst powder was dis-
persed in 100 mL of 15 ppm MB (Aldrich) solution by stirring
with a magnetic stirrer. A 13 W fluorescent lamp (FRX 13EX-
D) with light output in the 400–800 nm spectrum range was
used as the visible light source. The MB concentration was
measured every 30 min using a UV–Visible spectrophotome-
ter (Shimadzu 160-A).
3. Results and discussion
Fig. 2 presents SEM images of TiO2–CCNF and TiO2–CNF hy-
brids prepared with and without graphene. The electrospun
PAN based- and PAN/graphene based nanofibers showed a
smooth surface with an average diameter of 180 and
270 nm, respectively, before coating of TiO2. It was observed
that the nanoparticles were uniformly distributed across the
surface of the TiO2–CCNF materials (Fig. 2a) without aggrega-
tion, whereas the TiO2 on pure PAN/based fibers (TiO2–CNF)
consisted of large, aggregated nanoparticles with diameters
in the range of 80–110 nm (Fig. 2b). The graphene, functional-
ized with carboxylic acid groups, is capable of binding to me-
tal oxide particles, such as TiO2, without aggregation effects
and yields small nanparticles [14]. Goncalves et al. [41]. have
shown that oxygen functionalities at the graphene surface
Fig. 2 – SEM images of (a) electrospun PAN based nanofiber, (b) electrospun PAN/graphene based nanofiber, (c) TiO2–CNF, (d)
TiO2–CCNF.
act as reactive sites for the nucleation and growth of gold
nanoparticles. They observed that the nucleation and growth
of these metal nanoparticles were dependent on the density
of oxygen functional groups in the graphene surface sheets
[42]. TiO2 nanoparticles directly grown on graphene appeared
to exhibit strong interactions, which should lead to advanced
hybrid materials for various applications, including
photocatalysis.
TEM images, depicted in Fig. 3a, were collected to obtain
microstructural information for the TiO2–CCNF materials.
The images indicated that TiO2 nanoparticles, with an aver-
age particle size of 10–30 nm, had been successfully grown
on the surface of CNFs. SAED measurements (Fig. 3b) con-
firmed the nanocrystalline nature of the investigated TiO2
samples. Based on the SAED ring pattern shown in Fig. 3b,
the clear lattice fringes indicate that the nanoparticles have
some crystallinity. In addition, the SAED patterns indicate
that the nanoparticles, identified as TiO2 rutile structures, re-
flected from the (1 1 0), (2 1 1), and (1 0 1) lattice planes as in-
dexed in Fig. 3b. The corresponding elemental mapping of
an individual TiO2–CCNF materials provided information on
the distribution of Ti, O, and C atoms in the fiber (Fig. 3c).
Fig. 4 presents the XRD patterns of the TiO2–CCNF and
TiO2/CNF materials. The broad peak located between 20
and 30 was attributed to the (0 0 2) plane of the carbon struc-
ture. The diffraction peaks of TiO2 were also clearly observed
in Fig. 4a and b. The peak locations for TiO2 are cited from the
Joint Committee on Powder Diffraction Standards (JCPDS)
database. The peaks located at 25 Correspond to the (1 0 1)
plane of the anatase phase (JCPDS 21-1272), and the peaks lo-
cated at 28o, 36o, 41o, and 54 Correspond to the (1 1 0), (1 0 1),
(1 1 1), and (2 1 1) planes of the rutile phase (JCPDS 21-1276),
 CARBON 5 0 ( 20 1 2 ) 2 4 7 2–24 8 1 2475

Fig. 3 – (a) TEM image of the TiO2–CCNF hybrids. (b) SAED patterns of the TiO2 nanostructures. (c) Elemental mapping of the
individual TiO2–CCNF hybrids.

Fig. 4 – XRD patterns of the (a) TiO2–CCNF and (b) TiO2/CNF.

respectively [43]. All products were confirmed to be a mixture
of the anatase and rutile phases. The anatase phase content
for all products was calculated from the XRD patterns using
the following equation: Xa = [1 + 1.26(Ir/Ia)] 1, where Xa is the
share of anatase in the mixture and Ir and Ia are the integrated
intensities of the (1 0 1) reflection of anatase and the (1 1 0)
reflection of rutile, respectively. TiO2–CCNF has a 1:1 ratio of
anatase to rutile, whereas the ratio for TiO2–CNF is 1:4. There-
fore, this result indicates that the addition of graphene to TiO2
can effectively suppress the transformation of TiO2 from an
anatase to a rutile-type phase and prevents the growth of
TiO2 particles.
The Raman spectrum (Fig. 5) indicates the presence of TiO2
crystalline nanoparticles and showed three Raman bands:
144(B1g) for anatase and 431(Eg)/612(A1g) for the rutile structure
ranging between 100 and 800 cm 1 [44]. Fig. 5 shows that the
two structural types of TiO2 coexist, which is consistent with
the reported XRD pattern. In addition to different TiO2 modes,
the broad D-band (defect-induced mode) at 1340–1360 cm 1
and G band (E2g2 graphite mode) at 1570–1590 cm 1 were
observed. The intensity ratio of D band to G band (ID/IG) is pro-
posed to be an indication of disorder in TiO2–CCNF, TiO2–CNF,
and CNF, originating from defects associated with vacancies,
grain boundaries, and amorphous carbons. It is seen that
there is little change in the ID/IG of TiO2–CCNF (1.0547) com-
pared with TiO2–CNF (1.0551) and CNF (1.0522). This result
seems to indicate that graphitic (sp2) nature of the carbon in
the photocatalysts remained the same after loading with TiO2.
XPS analysis can provide valuable insight into the surface
structure of the CCNF-supported TiO2 photocatalyst. The Ti2p
photoelectron peaks of TiO2–CCNF materials are presented in
Fig. 6a. The peaks at 465.23 and 459.62 eV represent the Ti2p1/2
2476 CARBON 5 0 ( 2 0 1 2 ) 2 4 7 2 –2 4 8 1
Fig. 5 – Raman spectrum of TiO2–CCNF hybrids.
and Ti2p3/2 of Ti4+, respectively, and the Ti3+ peaks at
461.90 eV (Ti2p1/2) and 458.58 eV (Ti2p3/2) were also observed.
However, in the case of TiO2–CNF materials (Fig. 6b), the peaks
of Ti2p1/2 and Ti2p3/2 of Ti4+ were present, whereas the peaks
of Ti3+ in Ti2O3 or TiO were barely observed. The Ti3+ species
were generated for the TiO2/CCNFs, which might be due to
the carboxyl groups by graphene. Some of these groups com-
bine with titanium through chelation, and then the positions
of the oxygen in TiO2 were occupied by carboxyl groups. The
chelation effect reduce the valence of titanium, whereas in-
crease the content of Ti3+ [46]. Generally, the Ti3+ state on
the TiO2 surface plays a dominant role in the photocatalytic
activity because it can trap photogenerated electrons and
leave behind unpaired charges to promote photoactivity.
Therefore, the increasing Ti3+ density promotes effective seg-
regation of electron and cavity, interface charge transfer as
well as lowered the probability of compounding cavity, and
then increases the photocatalysis performance.
Fig.7 shows the high-resolution XPS spectra of the C1s and
O1s region on the surface of TiO2–CCNF. The deconvolution of
the C1s XPS spectrum (Fig. 7a) revealed three Gaussian curves
Fig. 6 – XPS spectra of the deconvoluted Ti(2p) peaks for the (a)TiO2–CCNF and (b) TiO2–CNF materials.
centered at 285.0, 286.3, 287.4, and 289.1 eV, which can be as-
signed to graphitic carbon (C–C and CHn), alcohol or ether car-
bon (C–OH, C–O–C), carbonyl groups (C@O), and carboxyl or
ester carbon (O@C–O), respectively. The O1s core level spec-
trum (Fig. 7b) was deconvoluted into two peaks, indicating
the presence of C@O or Ti–O bond of TiO2 (530.5 eV) and
C–O in ether (O2 , 532.2 eV) [46]. Atomic ratios of titanium to
oxygen for TiO2–CCNF and TiO2–CNF can be calculated
through the rough stoichiometry using the XPS data. The
Ti:O ratios of TiO2–CCNF and TiO2–CNF were determined to
be 1:2.63 and 1:2.50, respectively. However, it has been sup-
posed that the titanium-oxygen ratio is less than 1:2. There
is an overall increase in the Ti:O ratio, which is consistent
with the increase in the proportion of oxygen in the photocat-
alysts. It means that there may be more oxygen species such
as carboxylic acid group of graphene in the TiO2–CCNF photo-
catalyst surface besides the crystal lattice oxygen [46,47].
The FT-IR spectrum of the TiO2–CCNF was reported in
Fig. 8. We can observe absorption at 3439.08 cm 1 (OH broad-
ened band of either alcoholic or hydroxyl groups), 1612.49/
1454.33 cm 1 (C@C and C–C stretching bands in the aromatic
range, carbonyl), and 1000–1300 cm 1 (C–O stretching band of
ether). In particular, the intensity of SNF is larger than that of
TiO2–CCNF, which represent indirectly that TiO2 nanoparticles
are grown on surface of CCNF with the aid of the oxygen func-
tionalities as reactive sites at the graphene surface. Hence,
the XPS and FT-IR spectra show the presence and decrease
of the oxygen functional groups, such as –C–O–, –C@O, and
–O–C@O, on the TiO2–CCNF catalyst after carbonization.
The photoluminescence (PL) emission spectroscopy has
been widely used to study the transfer behavior of the photo-
generated electrons and holes in semiconductor materials, so
that it can reflect the separation and recombination of photo-
generated carries. [48,49]. The measured PL-emission spectra
of TiO2–CCNF, TiO2–CNF, and CNF in the range of 300–850 nm
are presented in Fig. 9. The spectra are broad (extending from
400 to 700 nm) and centered around 485 nm (2.56 eV) for TiO2–
CCNF and 497 nm (2.50 eV) for TiO2–CNF. The intensity of the
PL-spectra decreased in the following order for the photocat-
alysts: CNF, TiO2–CNF, and TiO2–CCNF; this result indicates
that the reduction of the PL-intensity shows the diminution
 CARBON 5 0 ( 20 1 2 ) 2 4 7 2–24 8 1 2477

Fig. 7 – Deconvolution of (a) C1s, (b) O1s core levels of TiO2–CCNF according to XPS spectra.

photocatalysts [50]. Therefore, the low PL intensity for TiO2–

Fig. 8 – FT-IR spectra of the TiO2–CCNF and SNF.
of the electron/hole pairs recombination process. Generally,
the lower PL intensity suggests the lower the recombination
rate of photogenerated electron–hole pairs, which leads to
the high the photocatalytic activity of semiconductor
CCNF indicates that the photocatalytic activities of TiO2 may
be improved due to the interactions between the excited elec-
tron of TiO2 particles and the graphene.
The photocatalytic activities of the TiO2–CCNF, TiO2–CNF,
CCNF, and CNFs were evaluated by measuring the decolor-
ation rate of MB under visible light irradiation (Fig. 10) at
18 C. The experimental results showed that the MB decom-
position rate in the samples occurred in the following order:
TiO2–CCNF > TiO2–CNF  CCNF > CNF. The concentration re-
duced rapidly over the first 30 min, and then the decomposi-
tion rate slowed down or stopped after the amount of time for
all of the samples. As seen in Fig. 10a, the TiO2–CCNF showed
the best photocatalytic activity under visible light irradiation.
The use of the TiO2–CCNF catalyst yielded nearly a 100%
reduction in the first 30 min of irradiation. The MB decompo-
sition reaction followed pseudo-first-order kinetics, and the
kinetic constants (k, min 1) were calculated and are pre-
sented in the inset in Fig. 10a. The superior photocatalytic
activity of TiO2–CCNF materials was also demonstrated in
the degradation of MB. The average rate constant for the
TiO2–CCNF (k = 0.16 min 1) is much larger than those for
TiO2–CNF (k = 0.012 min 1), CCNF (k = 0.011 min 1), and CNFs

Fig. 9 – Photoluminescence spectra of (a) TiO2–CCNF, (b) TiO2–CNF, (c) CNF.

2478 CARBON 5 0 ( 2 0 1 2 ) 2 4 7 2 –2 4 8 1

Fig. 10 – (a) Photocatalytic degradation of MB monitored as the concentration change versus irradiation time (inset: average
reaction rate constant (min 1) for the photodegradation of MB with free TiO2). (b) Photocatalytic activity of the TiO2–CCNF
hybrids for MB degradation over five degradation cycles.

(k = 0.003 min 1). Additionally, the stability of the TiO2–CCNF
materials was examined by following the degradation of MB
during a five-cycle experiment. After each run, the TiO2–CCNF
photocatalyst was evacuated for 30 min and was reused in the
next run. As shown in Fig. 10b, the photocatalytic degradation
of MB with TiO2–CCNF phtocatalysts under visible light irradi-
ation was consistently effective over five degradation cycles.
In general, the efficiency of the photodegradation by TiO2
depends on the illuminated catalyst surface area in contact
with the solution and on the mass transfer to the catalyst sur-
face [37,38]. Nitrogen adsorption isotherms and pore size dis-
tributions for TiO2–CCNF, TiO2–CNF, and CNF are shown in
Fig. 11. The adsorption isotherms (Fig. 11a) of TiO2–CNF and
CNF show typical type I behavior representing the micropo-
rous adsorption, and the adsorption of nitrogen was nearly
complete at a low relative pressure (P/P0 < 0.1). The adsorption
isotherms of TiO2–CCNF exhibit combined type I and type II
characteristics. Hysteresis at a relative pressure higher than
P/P0 = 0.5 was observed, which was typical type II behavior
of mesoporous adsorption and micropore filling were ob-
served at a low relative pressure of P/P0 < 0.1, which was
typical type I behavior [51]. The pore size distribution in
TiO2–CCNF, TiO2–CNF, and CNF samples based on the BJH
method show a broad distribution of mesopores, with sizes
ranging from 2–50 nm (Fig. 11b). In particular, TiO2–CCNF
sample has a much larger pore volume around 3.8 nm than
the other samples, suggesting that the presence of mesopores
can lead to adsorption of MB molecular easily. Because the
molecular size of MB is 1.36 · 0.47 · 0.24 nm [20], and it can
access pores with diameters larger than 1.5 nm [52]. Further-
more, as shown in Table 1, the pore characteristics of the
photocatalysts decreased in the following order: TiO2–CCNF,
CNF, and TiO2–CNF; this result indicates that the exposed
CNFs surface with high surface area will be functioned as
centers of condensing substrates with a physical adsorption
process. Therefore, rapid decomposition of MB in the

Fig. 11 – (a) Nitrogen adsorption isotherms at 77 K, (b) Differential pore volume of various photocatalysts as a function of the
pore diameter.
 CARBON 5 0 ( 20 1 2 ) 2 4 7 2–24 8 1 2479

color could be due to consequential processes of adsorption

Table 1 – Pore characteristicsa of the photocatalysts used in
this study. into CNF layer and then decomposition on the surface of
TiO2 particle.
Samples BET surface Total pore Average pore As CNFs and graphene of photocatalysts have some
area volume diameter
adsorption capacity for MB, as shown in Fig. 11 and Table 1,
(m2 g 1) (cm3 g 1) (nm)
a comparison between adsorption and photocatalytic
TiO2–CCNF 434 0.23 2.17 degradation of MB was carried out experimentally with and
TiO2–CNF 361 0.15 1.60 without visible light irradiation to evaluate the actual photo-
CCNF 447 0.17 1.58
catalytic activity. Under the same conditions, adsorption of
CNF 405 0.18 1.81
TiO2–CCNF and TiO2–CNF to MB in the dark was also tested
a
Pore parameters were obtained from N2 adsorption. and the results are shown in Fig. 12. When the catalyst is illu-
minated by visible light irradiation, the MB concentration is
much higher decreased than that without visible light irradi-
ation, indicating that MB is rapidly decomposed by the
TiO2–CCNF catalysts photocatalytically with visible light
irradiation.
The results indicate that TiO2–CCNF has a higher efficiency
for the decomposition of MB than the TiO2–CNF, CCNF, and
CNFs. It was confirmed that the introduction of graphene
and CNFs to the TiO2 photocatalyst could increase the decom-
position rate of some organic compounds using a photocata-
lytic process. The reasons for these improvements are
attributed to the following functions, as shown in Fig. 13.

(i) The graphene with carboxylic acid groups is capable of

Fig. 12 – Adsorption behavior of MB on TiO2–CCNF and TiO2–
CNF.
beginning is reasonably supposed to be due to further adsorp-
tion of MB into CNF layer, and the following gradual change in
binding TiO2 without aggregation effects, resulting in
small nanoparticles. As the particle size of TiO2
becomes small, the photogenerated electrons and holes
can easily migrate to reaction sites on the surface,
thereby reducing the recombination probability [13,45].
(ii) In the TiO2–CCNF materials with a mixture of anatase
and rutile structures, the graphene may act as an elec-
tron acceptor (Fig. 13a) for rutile (Ebg = 3.0 eV) and as a
photosensitizer (Fig. 13b) for anatase (Ebg = 3.2 eV),
which prevents the recombination of electron–hole

Fig. 13 – Schematic illustration of photocatalysis with TiO2–CCNF hybrids.

2480 CARBON 5 0 ( 2 0 1 2 ) 2 4 7 2 –2 4 8 1
pairs and promotes photodegradation [13]. This phe-
nomenon may be responsible for extending photocata-
lytic activity into the visible light range.
(iii) A large surface area is necessary for the photocatalytic
degradation of organic compounds because adsorption
of the organic compound is an important process. The
exposed fibers with high-surface-area TiO2–CCNF
materials will function as centers of substrate conden-
sation in the physical adsorption process. There the
peculiar properties of CNFs are expected to elevate
the photocatalytic activities of TiO2.
4. Conclusion
Small TiO2 nanoparticles were successfully deposited on elec-
trospun CCNFs by the sol–gel method, and these composites
were very active photocatalysts in the photodegradation of
MB under visible light irradiation. Furthermore, TiO2–CCNF
materials could be recycled without a decrease in the photo-
catalytic activity. The results suggest that graphene acts as an
electron acceptor and a photosensitizer, which causes an in-
crease in the photodegradation rate and reduces electron–
hole pair recombination. In addition, the CNFs’ high surface
area synergistically improved the photocatalytic activity of
TiO2 by enhancing the physical adsorption of the substrate.
Therefore, TiO2–CCNF materials are expected to be a promis-
ing candidate for photocatalytic processes using sunlight
irradiation.
Acknowledgements
This research was supported by the Basic Science Research
Program through the National Research Foundation of Korea
(NRF) funded by the Ministry of Education, Science and
Technology (No. 2011-0005890) and the Ministry of Education,
Science and Technology (MEST) (K20903002024-11E0100-
04010).
R E F E R E N C E S
[1] Zhang J, Xu Q, Feng Z, Li M, Li C. Importance of the
Relationship between Surface Phases and Photocatalytic
Activity of TiO2. Angew Chem In Ed 2008;47:1766–9.
[2] Ni M, Leung KH, Leung DYC, Sumathy K. A review and recent
developments in photocatalytic water-splitting using TiO2 for
hydrogen production. Renew Sust Energy Rev 2007;11:401–25.
[3] Fox MA, Dulay MT. Heterogeneous photocatalysis. Chem Rev
1993;93:341–57.
[4] Heller A. Chemistry and applications of photocatalytic
oxidation of thin organic films. Acc Chem Rev 1995;28:503–8.
[5] Blake DM, Maness PC, Huang Z, Wolfrum EJ, Huang J, Jacoby
WA. Application of the photocatalytic chemistry of titanium
dioxide to disinfection and the killing of cancer cells. Sep
Purif Methods 1999;28:1–50.
[6] Kamat PV. Photochemistry on nonreactive and reactive
(semiconductor) surfaces. Chem Rev 1993;93:267–300.
[7] Wu T, Liu G, Zhao J, Hidaka H, Serpone N. Evidence for H2O2
generation during the TiO2-assisted photodegradation of
dyes in aqueous dispersions under visible light illumination. J
Phys Chem B 1999;103:4862–7.
[8] Zhao W, Chen C, Li X, Zhao J, Hidaka H, Serpone N.
Photodegradation of Sulorhodamine-B Dye in Platinum as a
Functional Co-Catalyst. J Phys Chem B 2002;106:5022–8.
[9] Zhang Z, Shao C, Zhang L, Li X, Yichun L. Electrospun
nanofibers of V-doped TiO2 with high photocatalytic activity. J
Colloid Interface Sci 2010;351:57–62.
[10] Chen X, Mao SS. Titanium dioxide nanomaterials: synthesis,
properties modifications and applications. Chem Rev
2007;107:2891–5.
[11] Liu G, Wang L, Yang HG, Cheng HM, Lu GQ. Titania-based
photocatalysts-crystal growth, doping and heterostructuring.
J Mater Chem 2010;20:831–43.
[12] Wang W, Serp P, Kalck P, Faria JL. Visible light
photodegradation of phenol on MWNT-TiO2 composite
catalysts prepared by a modified sol–gel method. J Mol Catal
A: Chem 2005;235:194–9.
[13] Leary R, Westwood A. Carbonaceous nanomaterials for the
enhancement of TiO2 photocatalysis. Carbon 2011;49:741–72.
[14] Liang Y, Wang H, Casalongue HS, Chen Z, Dai H. TiO2
nanocrystals grown on graphene as advanced photocatalytic
hybrid materials. Nano Res 2010;3:701–5.
[15] Li Y, Li X, Li J, Yin J. TiO2-coated active carbon composites
with increased photocatalytic activity prepared by a properly
controlled sol–gel method. Mater Lett 2005;59:2659–63.
[16] Kominami H, Kohno M, Matsunaga Y, Kera Y. Thermal
decomposition of titanium alkoxide and silicate ester in
organic solvent: a new method for synthesizing large-
surface-area, silica-modified titanium(iv) oxide of high
thermal stability. J Am Ceram Soc 2001;84:1178–80.
[17] Chen H, Shi J, Zhang W, Ruan M, Yan D. Incorporation of
titanium into the inorganic wall of ordered porous zirconium
oxide via direct synthesis. Chem Mater 2001;13:1035–40.
[18] Chun H, Yizhong W, Hongxiao T. Influence of adsorption on
the photodegradation of various dyes using surface bond-
conjugated TiO2/SiO2 photocatalyst. Appl Catal B Environ
2001;35:95–105.
[19] Takeda N, Torimoto T, Sampath S, Kuwabata S, Yoneyama H.
Effect of inert supports for titanium dioxide loading on
enhancement of photodecomposition rate of gaseous
propionaldehyde. J Phys Chem 1995;99:9986–91.
[20] Ding Z, Lu GQ, Greenfield PF. A kinetic study on
photocatalytic oxidation of phenol in water by silica-
dispersed titania nanoparticles. J Colloid Interface Sci
2000;232:1–9.
[21] Hirano M, Nakahara C, Ota K, Inagaki M. Direct formation of
zirconia-doped titania with stable anatase-type structure by
thermal hydrolysis. J Am Ceram Soc 2002;85:1333–5.
[22] Kim B-H, Nataraj SK, Yang KS, Woo H-G. Synthesis,
characterization, and photocatalytic activity of TiO2/SiO2
nanoparticles loaded on carbon nanofiber web. J Nanosci
Nanotechnol 2010;10:3331–5.
[23] Luo Y, Heng Y, Dai X, Chen W, Li J. Preparation and
photocatalytic ability of highly defective carbon nanotubes. J
Solid State Chem 2009;182:2521–5.
[24] Williams G, Seger B, Kamat PV. TiO2-graphene
nanocomposites. UV-assisted photocatalytic reduction of
graphene oxide. ACS Nano 2008;2:1487–91.
[25] Zhang H, Lv X, Li Y, Wang Y, Li J. P25-graphene composite as a
high performance photocatalyst. ACS Nano 2010;4:380–6.
[26] Jiang G, Lin Z, Chen C, Zhu L, Chang Q, Wang N, et al. TiO2
nanoparticles assembled on graphene oxide nanosheets with
high photocatalytic activity for removal of pollutants. Carbon
2011;49:2693–701.
[27] Tsumura T, Kojitani N, Umemura H, Toyoda M, Inagaki M.
Composites between photoactive anatase-type TiO2 and
adsorptive carbon. Appl Surf Sci 2002;196:429–36.
 CARBON 5 0 ( 20 1 2 ) 2 4 7 2–24 8 1
[28] Modestov AD, Lev O. Photocatalytic oxidation of 2,4-
dichlorophenoxyacetic acid with titania photocatalyst.
Comparison of supported and suspended TiO2. J Photochem
Photobiol A Chem 1998;112:261–70.
[29] Kim CH, Kim B-H, Yang KS. Visible light-induced
photocatalytic activity of Ag-containing TiO2/carbon
nanofibers composites. Synthetic Metals 2011;161:1068–72.
[30] Wenqing F, Qinghua L, Qinghong Z, Ye W.
Nanocomposites of TiO2 and reduced graphene oxide as
efficient photocatalysts for hydrogen evolution. J Phys
Chem C 2011;115:10694–701.
[31] Robel I, Bunker BA, Kamat PV. SWCNT-CdS nanocomposite as
light harvesting assembly. Photoinduced charge transfer
interactions. Adv Mater 2005;17:2458–63.
[32] Oh WC, Zhang FJ, Chen ML. Synthesis and characterization of
V-C60/TiO2 photocatalysts designed for degradation of
methylene blue. J Ind Eng Chem 2010;16:299–304.
[33] Geim AK, Novoselov KS. The rise of graphene. Nat Mater
2007;6:183–91.
[34] Bolotin KI, Sikes KJ, Jiang Z, Klima M, Fudenberg G, Hone J,
et al. Ultrahigh electron mobility in suspended graphene.
Solid State Commun 2008;146:351–5.
[35] Wang HL, Robinson JT, Diankov G, Dai HJ. Nanocrystal growth
on graphene with various degrees of oxidation. J Am Chem
Soc 2010;132:3270–1.
[36] Si YC, Samulski ET. Exfoliated graphene separated by
platinum nanoparticles. Chem Mater 2008;20:6792–7.
[37] Amjad HES, Alan PN, Hafid A-D, Suki P, Neil C, Steven Y.
Deposition of anatase on the surface of activated carbon. Surf
Coat Tech 2004;187:284–92.
[38] Liu J-H, Yang R, Li S-M. Preparation and application of
efficient TiO2/ACFs photocatalyst. J Environ Sci
2006;18:979–81.
[39] Kim SK, Lee SK, Hwang SH, Lim H. Photocatalytic deposition
of silver nanoparticles onto organic/inorganic composite
nanofibers. Macromol Mater Eng 2006;291:1265–70.
[40] Drew C, Liu X, Ziegler D, Wang XY, Bruno FF, Whitten J, et al.
Folding transition of large DNA completely inhibits the action
of a restriction endonuclease as revealed by single-chain
observation. Nano Lett 2003;3:143–6.
[41] Goncalves G, Marques PAAP, Granadeiro CM, Nogueira HIS,
Singh MK, Gracio J. Surface modification of graphene
2481
nanosheets with gold nanoparticles: the role of oxygen
moieties at graphene surface on gold nucleation and growth.
Chem Mater 2009;21:4796–802.
[42] Kamat PV. Graphene-Based Nanoarchitectures. Anchoring
Semiconductor and Metal Nanoparticles on a Two-
Dimensional Carbon Support. J Phys Chem Lett. 2010; 1: 520–
527.
[43] Dai S, Wu Y, Sakai T, Du Z, Sakai H, Abe M. Preparation of
highly crystalline TiO2 nanostructures by acid-assisted
hydrodrothermal treatment of hexagonal-structured
nanocrystalline titania/cetyltrimethyammonium bromide
nanoskeleton. Nanoskeleton Nanoscale Res Lett
2010;5:1829–35.
[44] Zhou Q-f, Wu S-h, Shang Q-q, Zhang J-x. Structure and
Raman Spectroscopy of Nanocrystalline TiO2 Powder Derived
by Sol-Gel Process. Chin Phys Lett. 1997; 14: 306-369.
[45] Kudo A, Miseki Y. Heterogeneous photocatalyst materials for
water splitting. Chem Soc Rev 2009;38:253–78.
[46] Deqing M, DaiQi Y. Surface study of composite photocatalyst
based on plasma modified activated carbon fibers with TiO2.
Surf Coat Tech 2009;203:1154–60.
[47] Steven FD, Nilson CC, Elidiane CR, Mario ABM. Plasma
enhanced chemical vapor deposition of titanium (IV)
ethoxide-oxygen-helium mixtures. Thin Solid Films
2008;516:4940–5.
[48] Tang H, Berger H, Schmid PE, Levy F, Burri G.
Photoluminescence in TiO2 anatase single crystals. Solid
State Commun 1993;87:847–50.
[49] Jin J, Marshal D, Shin J-H, Kim J-C, Getoff N. Surface
properties and photoactivity of TiO2 treated with electron
beam. Radiat Phys Chem 2006;75:583–9.
[50] Xu L, Zhongqing L, Jian Z, Xin Y, Dandan L, Si C, et al.
Characteristics of N-doped TiO2 nanotube arrays by N2-
plasma for visible light-driven photocatalysis. J Alloys Compd
2011;509:9970–6.
[51] Wang D-W, Li F, Gao ML, Lu Q, Cheng H-M. 3D aperiodic
hierarchical porous graphitic carbon material for high-rate
electrochemical capacitive energy storage. Angew Chem Int
Ed 2008;47:373–6.
[52] Yuan R, Guan R, Shen W, Zheng J. Photocatalytic degradation
of methylene blue by a combination of TiO2 and activated
carbon fibers. J Colloid Interface Sci 2005;282:87–91.