Carbon coated nitrogen doped P25 for the photocatalytic removal of organic pollutants under solar and low energy visible light irradiations

Article abstract of DOI:10.1016/j.molcata.2013.11.030

The commercially available TiO₂ Degussa P25 was modified using a simple technique to produce a visible light active carbon (C) coated and nitrogen (N) doped P25. The modified photocatalyst was prepared by mechanical mixing of P25 powder with various amount of urea and followed by thermal heating under semi-closed reactor at different temperatures. The results showed that a layer of carbon graphite was deposited on the surface of P25 and N atoms were incorporated into the lattice of the nano-scaled photocatalyst. It was observed that the UV-Vis spectra of the modified photocatalyst exhibited red shift to the visible region as compared to that of the pristine P25, implying the decreased of band gap energy for the modified photocatalyst. Moreover, the modified photocatalyst was found to yield efficient photocatalytic degradations of the RR4, MB and phenol pollutants under visible light and solar light irradiations.

Full text of DOI:10.1016/j.molcata.2013.11.030

Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Carbon coated nitrogen doped P25 for the photocatalytic removal of
organic pollutants under solar and low energy visible light irradiations
W.I. Nawawi, M.A. Nawi∗
Photocatalysis Laboratory, School of Chemical Sciences, Universiti Sains Malaysia, Pulau Pinang, Malaysia
a r t i c l e i n f o
a b s t r a c t
Article history:
The commercially available TiO2 Degussa P25 was modified using a simple technique to produce a visible
light active carbon (C) coated and nitrogen (N) doped P25. The modified photocatalyst was prepared by
mechanical mixing of P25 powder with various amount of urea and followed by thermal heating under
semi-closed reactor at different temperatures. The results showed that a layer of carbon graphite was
deposited on the surface of P25 and N atoms were incorporated into the lattice of the nano-scaled photo-
catalyst. It was observed that the UV–Vis spectra of the modified photocatalyst exhibited red shift to the
visible region as compared to that of the pristine P25, implying the decreased of band gap energy for the
modified photocatalyst. Moreover, the modified photocatalyst was found to yield efficient photocatalytic
degradations of the RR4, MB and phenol pollutants under visible light and solar light irradiations.
© 2013 Elsevier B.V. All rights reserved.
Received 5 August 2013
Received in revised form
18 November 2013
Accepted 26 November 2013
Available online 6 December 2013
Keywords:
Degussa P25
Urea
Reactive red 4
Methylene blue
Phenol
1. Introduction
photocatalytic activities [10–12]. Chen et al. [10] prepared C, N co-
doped TiO2 using the sol–gel method where the product exhibited
The various advantages of titanium dioxide (TiO2) as a photo-
catalyst for effective photocatalytic removal of organic pollutants
in water effluents have received great attention from many
researchers [1]. Unfortunately, the use of TiO2 technology is limited
by its wide band gap energy (Eg = 3.2 eV) which requires UV light
irradiation for its photocatalytic activity. The UV light of the sun
only accounts for a small fraction (3–5%) of the sun’s light spec-
trum compared to visible light [2]. Thus any attempt of making
TiO2 to absorb visible light will have a profound positive effect
on its photocatalytic applications in environmental remediation.
For this purpose, various efforts have been directed towards the
development of visible-light responsive TiO2 materials.
The modification of TiO2 towards narrowing its band gap allows
the utilization of a wider fraction of visible light for the production
of charge carriers. Incorporation of metal ions like Pt, Au, Ag, Cu,
Fe and nonmetals such as C, N, S, F and P into TiO2 have been
proposed in order to improve its photocatalytic activity [3–13].
Nitrogen and carbon have been widely used for modification of TiO2
due to their capabilities in narrowing the photocatalyst’s band gap
which led to better photo-response and subsequently enhanced
photocatalytic activity [3–5]. Interestingly, simultaneous co-doped
TiO2 with C and N was reported to show a synergistic effect on its
higher photocatalytic activity under visible light as compared to
either C-doped TiO2 or N-doped TiO2. They attributed this phe-
nomenon to the enhanced adsorption of carbonate species on the
surface of the C, N co-doped TiO2 where the visible light photo-
activity was induced by N-doping into the TiO2 lattice. Another
interesting modification of TiO2 involved the preparation of a com-
posite of N doped TiO2–Al2O3. In this work, Li et al. [14] had proven
that N element can be doped into the lattice of TiO2 using urea and
nitrate as the N precursor in the presence of aqueous Al2O3 solution
at the calcination temperature of 300 ◦C. However, they also proved
that C was not detected within the modified catalyst. The product
was observed to be 43.6 times faster than that of pristine P25 TiO2
in the photocatalytic degradation of methyl orange. The high visi-
blelight photocatalytic activity was attributed to synthetic effects
between amorphous Al2O3 and TiO2, low recombination efficiency
of photo-excited electrons and holes, N-doping, and a large specific
surface area.
According to Park et al. [4] C doped has been interpreted either
as an anion that replaces oxygen substitutionally in the lattice
to produce carbides with Ti–C bond, or as a cation that occupy
interstitial lattice site to produce carbonates with Ti–O–C bond.
The presence of C without the detection of Ti–C or Ti–O–C bond-
ing systems normally means that C is coated on TiO2. Wang
et al. [15] produced C-sensitized N-doped TiO2 via a facile sol–gel
method using titanium butoxide as both titanium precursor and
carbon source, and nitric acid as nitrogen source. The incorporated
∗
Corresponding author. Tel.: +6046534031; fax: +6046574854.
E-mail address: masri@usm.my (M.A. Nawi).
1381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.11.030

84
W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
carbonaceous species could serve as photosensitizer, while the
nitrogen doping could lead to the remarkable red shift of absorption
edge of C/N–TiO2. The C/N–TiO2 exhibited the highest pho-
tocatalytic activity for sulfanilamide (SNM) degradation under
irradiation of visible-light-emitting diode (vis-LED). A simple
preparation of N-doped and S, N-doped P25 was done by Her-
rera et al. [16] using a simple mechanical mixing of P25 with urea
as N precursor and thiourea as N, S precursor annealed under
open atmosphere. It was found that S, N doped P25 showed bet-
ter photocatalytic activity than N doped P25 in the photocatalytic
degradation of phenol and E. coli under visible light irradiation.
The main objective of this study is to produce C coated, N doped
P25 by a simple technique which engages the mechanical mixing
of P25 with urea where urea serves as both carbon and nitrogen
precursor. The preparation of C coated, N doped P25 was done
according to the method applied by Hererra et al. [16] with several
modifications whereby the mixture of photocatalyst powder and
urea was heated under semi-closed reactor to produce C coated
and N doped P25. The modified photocatalyst was characterized
by HRTEM, XPS, BET, UV–Vis DRS, PL and its photocatalytic activi-
ties were tested by photocatalytic degradation of cationic RR4 dye,
anionic MB dye and phenol under solar and low energy visible light
fluorescent lamp.
Fig. 1. The experimental setup for preparing C coated N doped P25 using the cal-
cinations process. (a) Semi-closed reactor, (b) sample (c) muffle furnace and (d)
ventilation tubes.
2. Experimental
2.3. Characterization of C coated N doped P25
2.1. Material
The percentage of C and N in TiO2 (P25) was determined by
elemental analyzer CHNS-O (LECO-938) with 0.001% detection
limit. UV–Vis diffuse reflectance spectra were recorded using the
UV/Vis spectrometer model Lambda 35, Perkin Elmer. High res-
olution transmission electron microscopy (HRTEM) analysis was
performed using FEI Tecnai 20 TEM. BET surface areas was obtained
via Micromeritics ASAP 2000 gas adsorption surface analyzer, zeta
potentials for photocatalysts in aqueous solutions were measured
using Zetasizer Nano Series ZS analyzer model Malvern Nano ZS.
Photoluminescent spectrometer model Joblin Yvon HR800 UV was
used to obtain the photoluminescence spectra (PL) of the sam-
ples while X-ray photoelectron spectroscopy (XPS) spectra were
recorded using Omicron Nanotechnology (ELS5000) system using
Al K␣ radiation at a base pressure below 5.5 × 10−9 Torr.
TiO2 Degussa P25 powder was used as the starting material in
the preparation of C coated, N doped P25. Urea from Fluka (chem-
ical formula: NH2CONH2, MW: 60.06 g mol−1) was used as the C
and N precursor. Reactive red 4 (RR4) dye or commonly known as
Cibacron Brilliant Red (Colour Index Number: 18105, chemical for-
mula: C32H23ClN8Na4O14S4, MW: 995.23 g mol−1, ꢀmax: 517 nm)
with 50% dye content was provided by Aldrich Chemical. Methy-
lene blue (MB) dye (c.a. 98%, colour index number: 52015, chemical
formula: C16H18ClN3S.2H2O) was purchased from Unilab while
phenol (99.5%) was obtained from Scharlau. Ultra pure water
(18.2 Mꢁ cm−1) was used to prepare all solutions in this work.
Table 1 provides a summary of organic pollutants models used in
this work.
2.2. Preparation of C coated, N doped P25 samples
2.4. Adsorption study
C coated, N doped P25 samples were prepared by thermal heat-
ing process using a custom made semi-closed reactor placed in a
muffle furnace. The overall experimental set-up is shown in Fig. 1.
For a typical preparation, 3 g of TiO2 (P25) was mixed with various
amount of urea in powder form by mechanical mixing process for
5 min. The mixed powder was then placed into a conical flask in
the semi-closed reactor. The reactor was then placed into a muf-
fle furnace under the heating temperature ranging from 300 to
500 ◦C under normal atmospheric condition for 2 h. The sample
was then taken out from the muffle furnace and cooled down to
room temperature. The products were cleaned by sonicating the
sample in a 0.1 N HCl and centrifuged to isolate the contamination.
The modified sample was finally washed thoroughly using distilled
water under 0.45 ␮m cellulose nitrate membrane filter to yield a
yellowish solid of pure C coated, N doped P25 sample.
For adsorption study, experiments were conducted in the dark.
In each test, 0.024 g of photocatalyst samples were added into a
20 mL solution of 30, 12 and 10 mg L−1 of anionic RR4 dye, cationic
MB dye and phenol solutions, respectively, to form suspensions.
Those suspensions were then poured into a custom made glass cell
with dimension 50 × 10 × 80 (L × B × H) mm and aerated in the dark.
Samples were taken at 15 min interval for up to 1 h, filtered with
0.45 ␮m syringe filter. The decolorization degree of RR4 and MB dye
was determined by using a spectrophotometer HACH DR/2000 at
wavelength 517 and 661 nm, respectively, while a Shimadzu LC-10
ATVP high performance liquid chromatography (HPLC) with a C18
column was used for detecting phenol where its UV detector was
set at 220 nm with 60: 40 methanol–water mixture as the mobile
phase.
The labeling of each sample was based on the amount of urea
added and the heating temperature used. The preparation condi-
tions of C coated, N doped P25 samples are listed in Table 2 (column
1 and 2). A sample labeled as PU1-350 means C coated, N doped P25
prepared from the doping of 1 g of urea and calcined at 350 ◦C (see
Table 2, column 1).
2.5. Photocatalytic study
In each test, 0.024 g of photocatalyst sample was added into
a 20 mL of 30, 12 and 10 mg L−1 of anionic RR4 dye, cationic MB
dye and phenol solutions, respectively, to form suspensions and
were then individually poured into a glass cell and irradiated under

W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
85
Table 1
Summary of organic pollutants.
Chemical structures and organic pollutant
Properties
Chemical formula
Molecular weight (g mol−1
)
Absorption max (nm)
(1)
C32H23ClN8Na4O14S4
995.23
517 nm
Reactive Red 4 dye (RR4)
(2)
C16H18ClN3S.2H2O
319.85
94.11
661 nm
220 nm
Methylene blue dye (MB)
(3)
C6H6O
continuous aeration with a low energy Phillips 45-W compact fluo-
rescent lamp (source no.1) and solar (source no.2) light. In another
experiment, a UV filter Contact L6 was placed right in front of the
45-W fluorescent lamp to filter the UV irradiance leakage thus
providing a visible light source (source no. 3). The UV detector
Radiometer (Solar light co. PMA 2100) connected with a UV-A, UV-B
detector (range from 260 to 400 nm) and PAR Quantum Light Sen-
sor detector (range from 401 to 700 nm) was used to determine
the UV and visible light irradiance of each source, respectively.
The experiments under solar light were carried out on a sunny
day from 11.00 am to 3.00 pm. The experimental set up for the
lamp, solar and visible light irradiation (source no.1, 2 and 3) are
shown in Fig. 2. An aquarium pump model NS 7200 was used
as the aeration source. A direct reading air flowmeter (Gilmont)
was used to determine the aeration rate supply to the reactor.
Samples were taken at a specific time interval until almost com-
plete decolourisation was attained. The degradation degree of RR4
and MB and phenol was determined by using a HACH DR/2000
spectrophotometer and HPLC as described in the previous sec-
tion.
For total organic carbon (TOC) analysis, three 10 mL portions of
each treated solutions were taken and transferred into vials. The
Table 2
Experimental conditions and results of the modified TiO2. Pseudo first order rate constants (k) for the degradation of 30 mg L−1 RR4 anionic dye were obtained under low
energy Phillips 45 watts fluorescent light irradiation with residual UV leakage of 3.5 W m−2
.
TiO2 samples
Temp (◦C)
Urea (g)
C (wt%)
N (wt%)
S.Area (m2 g−1
)
Rate const. (k min−1
)
RR4
MB
phenol
P25
P25
30
0
0
0.00
0.00
0.72
0.75
0.74
0.73
0.72
1.37
1.39
1.37
1.38
1.39
2.03
2.01
2.03
1.95
2.04
2.48
2.42
2.47
2.42
2.44
2.91
2.86
2.89
2.85
2.87
4.13
0.00
0.00
1.98
1.92
1.91
1.84
1.75
2.25
2.24
2.22
2.19
2.05
3.77
3.72
3.61
3.54
3.47
4.88
4.68
4.51
4.38
4.20
5.81
5.67
5.58
5.50
5.31
8.64
49.25
49.21
0.088
0.057
0.038
0.053
0.061
0.044
0.024
0.052
0.158
0.132
0.051
0.050
0.048
0.148
0.128
0.041
0.034
0.047
0.117
0.100
0.034
0.020
0.047
0.117
0.100
0.034
0.020
0.047
0.021
0.019
0.018
0.017
350
300
350
400
450
500
300
350
400
450
500
300
350
400
450
500
300
350
400
450
500
300
350
400
450
500
350
PU0.5–300
PU0.5–350
PU0.5–400
PU0.5–450
PU0.5–500
PU1–300
PU1–350a
PU1–400
PU1–450
PU1–500
PU1.5–300
PU1.5–350
PU1.5–400
PU1.5–450
PU1.5–500
PU2–300
PU2–350
PU2–400
PU2–450
PU2–500
PU2.5–300
PU2.5–350
PU2.5–400
PU2.5–450
PU2.5–500
PU5–350
0.5
0.5
0.5
0.5
0.5
1.0
1.0
1.0
1.0
1.0
1.5
1.5
1.5
1.5
1.5
2.0
2.0
2.0
2.0
2.0
2.5
2.5
2.5
2.5
2.5
5.0
0.021
0.038
0.034
0.031
0.017
0.026
0.024
0.022
34.54
0.028
0.013
0.020
0.011
a
Labeling of samples such as PU1-350 means P25 doped with 1 g Urea heated at 350 ◦C to produce C-coated N-doped TiO2 samples.

86
W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
Fig. 3. XRD patterns of the pristine P25 and C coated N doped P25 samples prepared
using 0.6 g urea dosage at different calcination temperatures.
in its pristine form was also heated at 350 ◦C where no detection of
C and N was observed. As shown in Table 2, the C and N contents of
the product depended mainly on the amount of urea added into the
P25 samples. The selected range of temperatures used for this study
(300 to 500 ◦C) did not significantly affect the C content of the prod-
ucts which was in contrast to other published works where they
found that increasing calcination temperatures (>700 ◦C) decreased
the amount of C [17,18]. The probable reason for this observation
was due to the relatively low temperature usage during the sam-
ple preparation. On the other hand, the amount of N seems to be
slightly reduced with increasing calcination temperatures. Accord-
ing to Sato et al. [19], N content decreased even at temperature as
low as 400 ◦C. They also reported that the decreasing amount of N
was due to the sintering effect at 400 ◦C.
Table 2 also provides the BET surface areas for PU1-350 and P25
samples. As observed by a number of workers [20,21], C coated
TiO2 and N doped P25 normally have smaller BET surface areas as
compared to the pristine P25. The additional C and N elements in
TiO2 should result in the increment of particle size of the C coated N
doped P25. The decreased in surface area was said to be due to the
sintering effect [22]. In line with others [22–24], the surface area of
PU1-350 was 30% smaller compared with pristine P25.
3.1.2. XRD spectra
Fig. 3 shows the XRD patterns for pristine and C coated N doped
P25 at various temperatures. As observed by others [22,23,25], no
phase transformation occurred for the calcined samples even at
500 ◦C. In fact Zhang et al. [23] observed the occurrence of phase
transformation of C coated TiO2 only when the calcining tem-
perature was at 900 ◦C while Hsu et al. [22] observed the phase
transformation for N doped TiO2 to occur at 800 ◦C. In our case, the
crystallinity of C coated, N doped P25 was not affected since the
operation temperature was carried out within the range of 300 to
500 ◦C which was too low for phase transformation of P25 to occur.
Fig. 2. The photocatalytic experimental set up for the degradation of anionic RR4
dye, cationic MB dye and phenol where (a) under a 45-W fluorescent lamp (UV
irradiance leakage of 3.5 W m−2, Visible light: 311 W m−2), (b) under solar irradiation
(UV: 22 W m−2, Visible light: 430 W m−2), and (c) under visible light irradiation (UV:
0.0 W m−2, Visible light: 256 W m−2).
3.1.3. UV–Vis spectra
Fig. 4 shows the UV–Vis diffuse reflectance spectra of C coated N
doped P25 and pristine P25. As expected, the pristine P25 sample
exhibited absorption only in the UV region, whereas the optical
response of C coated N doped P25 had absorption both in the UV
and visible region. The visible light absorption for C coated N doped
P25 is associated with the doped nitrogen which can introduce an
impurity level between the valence and conduction band of TiO2
or narrow the band gap by mixing the N 2p and O 2p states [3,26].
Kubelka–Munk function was used to estimate the band gap energy
of the prepared samples by plotting (␣ hv) 1/2 versus energy of light
(see Figs. 5 and 3b). Results indicated that the band gap energies of
pristine P25 were 3.1 eV which was equal to the band gap energy
amounts of TOC of the treated solutions were then evaluated by a
total carbon analyzer (Shimadzu TOC 5000).
3. Results and discussion
3.1. Characteristics
3.1.1. CHN and BET analyses
C and N content of the modified photocatalyst samples prepared
at various heating temperatures and dosages of urea are summa-
rized in Table 2 (column 4 and 5). For control purposes, a P25 sample

W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
87
strengthened the evidence that C was just coated onto the surface
of TiO2 while N has a chemical bonding with the TiO2 particle.
3.1.5. TEM and HRTEM
Fig. 6(a–d) shows the TEM and HRTEM images for pristine P25
and PU1-350 samples. TEM image confirmed the average parti-
cle size of the C coated N doped P25 to be within the range of
16 nm (Fig. 6b) and the size was more or less the same with the
P25 particles (18 nm) as can be seen in Fig. 6a.
HRTEM was also used to estimate the thickness of C on the
surface of TiO2-P25 particles. The image of Fig. 6c shows that the
particle of P25 was coated with a thin layer of C with the fringe spac-
ing similar to graphite at about 0.34 nm [37–39]. This confirmed the
observed XPS results earlier that indicated the presence of the C
graphite layer. The HRTEM image (Fig. 6d) also shows that the par-
ticle of P25 was completely surrounded with the C graphite layer.
The coated C layer was observed to be about 2.04 nm. The thickness
of the C layer is very important to the photocatalytic process since
the thin layer of C is expected to allow better penetration of light
to the surface of TiO2 and should yield enhanced photocatalytic
activity.
3.1.6. PL spectra
The photoluminescence (PL) spectroscopy was used to study
the fate of electron–hole pairs in semiconductor particles since PL
emission results from the recombination of charge carriers [39,40].
Fig. 7a and b shows the PL spectra of pristine P25 and various
C coated N doped P25 samples at different heating temperatures
using the excitation wavelength of 325 nm. It can be seen that all
samples exhibit an obvious PL signal with a similar shaped curve
at the wavelength range from 420 to 520 nm. The peak at 420 nm
represents the band–band PL signal while the peak at 520 nm can
be due to the excitonic PL signal originated mainly from the surface
oxygen vacancies (OV) and defects of semiconductors [40]. It was
generally accepted that nitrogen doping can form a new electronic
state (N state) just above the valence band, making TiO2 to absorb
visible light [41] and this was further confirmed experimentally by
XPS valence band analysis (VB XPS), and visible light induced PL by
Wu et al. [42].
Fig. 4. UV–Vis diffused reflectance spectra of (a) pristine P25 and C coated N doped
P25 samples and (b) plots of the transformed Kubelka-Munk function vs the energy
of the light absorbed for PU1-350 (i) and pristine P25(ii).
From the density functional calculations (DFT), N-doping
favored the formation of OV, which was experimentally found to
be about 0.8 eV below the bottom of the conduction band [42,43].
The OV of nanosized semiconductor easily captures or bind with the
photoinduced electrons [44] as proven by the electronic spin reso-
nance method [45]. Hence, the larger the OV content, the stronger
the PL intensity. However, during the process of photocatalytic
reaction, OV and defects can become centers to capture photo-
induced electrons, so that the recombination of photo-induced
electrons and holes can be effectively inhibited [39]. Thus, it can be
suggested that OV and defects are in favor of photocatalytic reac-
tions and the stronger the excitonic PL spectrum, the larger the
content of OV or defect, and the higher the photocatalytic activity.
The PL intensity of C coated, N doped P25 using 1 g urea at dif-
ferent temperatures are shown in Fig. 7a. Increasing excitonic PL
intensity was detected when the temperature was increased up to
350 ◦C. However, the PL intensity decreased when the calcination
temperature was increased beyond 350 ◦C. The decreasing PL inten-
sity with increasing temperatures beyond 350 ◦C might be due to
the decreasing doped N content [10] at higher calcination temper-
ature as can be seen in Table 1 (column no 5). Thus the content of
OV formed decreased and less photo-induced electrons would be
trapped by the OV.
for the standard anatase TiO2 [27], while C coated N doped P25
has lower bandgap energies of ca. 2.6 eV. Due to the lower bandgap
energy, this modified TiO2 is expected to be visible light responsive.
3.1.4. XPS spectra
The chemical state and composition of the C coated N doped
P25 at the surface were studied by XPS whose spectrum reveals
the presence of Ti2p, O1s, C1s and N1s (see Fig. 5a–d) with the
binding energy at 455, 526.5, 284.5 and 397.5 eV, respectively. As
shown in Fig. 5a, XPS signals for Ti2p were observed at binding
energies around 462.4 and 456.8 eV which as expected, confirmed
the presence of Ti2p1/2 and Ti2p3/2, respectively [28]. Two minor
peaks at 457.2 and 459.1 eV may be assigned to the presence of
Ti–N–O and Ti–O–N, respectively [14]. The deconvolution of
O1s in Fig. 5b represented Ti–O and O–H bonds at 531.6 and
530.8 eV, respectively, [29,30] while the small peak at 533.4 eV was
assigned as O–N [31]. In Fig. 5c, the deconvolutions of C1s peaks
were observed at 284.5 and 283.9 eV which can be assigned to the
adventitious element from the carbon tape used in the analysis
and C–C bond from the carbon graphite bonding [32–34]. No peak
was found around 282.6 to 282.9 eV that are normally associated
with Ti–C [35] thus the possibility of C substitution doping was not
observed in PU1-350 sample.
PL spectra for C coated N doped P25 samples with different
amount of urea and calcined at 350 ◦C are shown in Fig. 7b. C coated
N doped P25 samples showed higher PL intensity than pristine P25
where the sample with the highest amount of doped N possessing
The binding energy for Ti–N was observed in the peak at
397.5 eV, as can be seen in Fig. 5d [36]. Therefore, XPS spectra

88
W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
Fig. 5. XPS spectra of (a) the survey spectrum, (b) Ti 2p, (c) O 1s and (d) C 1s for the C coated N doped P25, PU1-350.
the highest PL intensity. The excitonic PL was increased by increas-
ing the amount of doped N. However, the excitonic PL for PU1-350
was lower compared with the samples of PU0.5-350. This might
be due to the presence of more C coated on the surface of TiO2 in
PU1-350 than for PU0.5–350. Higher C coating can produce lower
electron-hole recombination since C can function as electron scav-
enger as described by Janus et al. [46]. Therefore, two important
phenomena occurred simultaneously in the C coated N doped P25:
Fig. 6. TEM images of (a) P25 (b) PU1-350 while (c) and (d) are HRTEM images of PU1-350.

W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
89
(1) Doping of N into P25 increases excitonic PL intensity due to
2000
2000
a
increasing OV and defect state of the photocatalyst and (2) coat-
ing of C onto P25 inhibits electron-hole recombination resulting
in the decrease of the the excitonic and band–band PL intensities
[42]. On both occasions, the overall effect is to strongly inhibit the
recombination of charge carriers within the photocatalyst.
PU1-350
P25
1500
1500
PU1-300
PU1-450
1000
1000
3.1.7. Effect of aeration rate
All photocatalytic evaluations in this study were carried out
under continuous aeration. Oxygen was found to be essential for
the semiconductor photocatalytic degradation of organic contam-
inants [47]. Dissolved oxygen is usually employed as an electron
acceptor [48]. The presence of an electron acceptor is essential
in increasing the separation of the photogenerated electrons and
holes, thus improving the efficiency of the photocatalytic process.
In the aerated solution, the oxygen molecule is adsorbed onto the
catalyst surface. The adsorbed oxygen molecules will be reduced to
O2•- by electron that is promoted into the conduction band, leading
to the formation of perhydroxyl radicals. In addition, perhydroxyl
radical will produce hydrogen peroxide which can produce super
hydroxyl radicals. The role of oxygen as an electron acceptor is
summarized by the following equations [49]:
500
500
0
0
250
250
500
500
750
750
1000
1000
Wavelength (nm)
2000
2000
b
PN2.0-350
PU2-350
PN1-350
PU1-350
1500
1500
PN0.5-350
PU0.5-350
P25
P25
PN1-300
PU1-300
PN1-450
PU1-450
1000
1000
-
•-
e + O2 → O2
(1)
(2)
(3)
(4)
(5)
(6)
•-
•
H
+ + O2 → HO2
500
500
HO2• → H2O2 + O2
H2O2 → 2OH•
•-
0
0
250
500
750
1000
250
500
750
1000
-
H2O2 + O2 → •OH + OH + O2
Wavelength (nm)
H2O2 + e → •OH + OH
-
-
Fig. 7. Photoluminescence spectra of (a) pristine P25 and C coated N doped P25 at
different calcination temperatures but similar amount of urea and (b) pristine P25
and C coated N doped P25 prepared at 350 ◦C containing different amount of urea.
Fig. 8 shows the effect of flow rates towards photocatalytic
degradation of RR4 dye using PU1-350 as the photocatalyst. It
was found that 25 mL min−1 was the optimum aeration rate where
the degradation rate of RR4 was almost two times faster than the
system running without aeration (stirring process). Therefore all
photocatalytic and adsorption experiments in this work were car-
ried out under this optimum aeration rate.
Photocatalytic degradation of RR4 dye using PU1-350 in the
presence of nitrogen gas was also carried out in order to provide evi-
dence for the role of O2 in the photocatalytic process. It was found
that the degradation rate of RR4 dye reduced significantly as com-
pared to the experiment ran under aeration. This was due to the
absence of the electron scavengers and the lack of the formation
of superoxides. The role of superoxides in enhancing photocat-
alytic degradation of dyes was proven by running the experiment
under aeration and in the presence of 1,4-benzoquinone, a known
superoxide scavenger. Supplementary Fig. 1 clearly indicates that
1,4-benzoquinone significantly reduced the efficiency of the pho-
tocatalytic degradation of RR4 dye eventhough the solution was
optimally aerated. Apparently superoxides produced via Eq. (1) was
consumed by 1,4-benzoquinone thus making reactions 2–6 ceased
to exist.
calcinations temperature of 300, 450 and 500 ◦C became less effec-
tive. The trend for the K values was similar with the PL intensity in
Fig. 7a where PU1-350 > P25 > PU1-300 > PU1-450. Beyond the cal-
cining temperature of 350 ◦C, the photocatalytic efficiency of the
products and also their respective PL intensity decreased due to
the lesser production of OV as described earlier. A similar trend
was observed by Chen et al. [10] when they synthesized C, N co-
doped TiO2 via a sol–gel method at heating temperatures of 400
and 500 ◦C and applied the product to degrade MB under a visible
light irradiation.
It was also observed that adding too much urea would also
reduce the photocatalytic efficiency of the modified photocatalyst
where the optimum TiO2 to urea ratio was 3:1. The different K
0.17
0.16
0.15
O
O2
3.2. Effect of C and N contents
N
N2
0.14
0.13
0.12
0.11
In this work, anionic RR4 dye was used as the model pollut-
ant for optimizing the preparation of C coated N doped P25. The
pseudo first order rate constants (K) for the degradation of RR4
using the 45-W compact fluorescent lamp without UV filter (source
no 1: 3.5 W m−2 of UV leakage) by various C coated N doped TiO2-
P25 samples under different calcining temperatures is presented
in Fig. 9. The calcination of the catalyst at 350 and 400 ◦C produced
significant photocatalytic activity where K values were higher com-
pared with pristine P25. On the other hand, the catalyst prepared at
0
25
50
75
100
Aeration flow rate (mL/min)
Fig. 8. Effect of aeration flow rates of air and nitrogen gas on the photocatalytic
degradation of 30 mg L−2 RR4 by PU1-350 under a 45-W fluorescent lamp.

90
W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
100
0.17
0.13
0.09
0.05
0.01
300
PU1-350- Phenol-a
PU1-350- Phenol-b
PU1-350- MB-a
PU1-350- MB-b
PU1-350- RR4-a
PU1-350- RR4-b
80
60
40
20
0
350
400
450
500
0
0.5
1
1.5
2
2.5
3
0
15
30
45
60
75
90 105 120 135
Amount of urea (g)
Times (min)
Fig. 9. Plots of the pseudo first order rate constants (K) for the degradation of
30 mg L−1 RR4 under 45-W fluorescent lamp for C coated N doped P25 samples
prepared at different calcination temperatures and different amount of urea.
Fig. 11. Plots of the photocatalytic degradation of pollutants using (a) pre-saturated
PU1-350 with individual pollutant and (b) unsaturated PU1-350 both under 45-W
fluorescent lamp.
shows the removal of pollutants (RR4, MB and phenol) under
pre-saturated PU1-350 and unsaturated PU1-350 samples. The
degradation of RR4, MB and phenol was always better when both
processes occurred simultaneously (adsorption + photocatalytic) as
compared to the isolated photocatalytic process only.
values observed for the RR4 degradation was due to the different
amount of N and C incorporated into the TiO2-P25 crystal lattice.
In one hand, N that was incorporated into the TiO2 lattice could
result in the band gap narrowing and red shift of absorption edge
towards the visible-light region which increased its photocatalytic
activity [36]. However when the N content is higher than the opti-
mum amount, it could form high recombination centers [50,51]
which eventually reduced its photocatalytic activity.
3.3. Effect of different light irradiations
As can be seen in Fig. 10, pristine P25 adsorbed 30 mg L−1 RR4,
12.5 mg L−1 MB and 10 mg L−1 phenol by about 42, 11 and 7%,
respectively, after 1 h of contact time. PU1-350 sample on the other
hand, adsorbed c.a. 59, 11 and 8% of RR4, MB and phenol solutions,
respectively. Pristine TiO2-P25 and PU1-350 has a strong adsorp-
tion for anionic RR4 dye due to the coulombic attraction by the
positively charged pristine P25 and PU1-350. This is proven via the
measurement of zeta potentials of these photocatalysts at ambient
condition (pH 5.5). The zeta potential of pristine P25 was observed
to be 0.79 mV while the zeta potential for PU1-350 was 15.57 mV.
This explains why PU1-350 adsorbed anionic RR4 dye better than
pristine P25 despite of its lower surface area. The adsorption was
less for MB and phenol due to the repulsion that occurs between
the positively charged MB and non-charged phenol (at ambient pH)
against the positively charged catalyst particles.
The 45-W compact fluorescent lamp had a residual UV irra-
diance leakage of 3.5 W m−2. The UV irradiation becomes zero
(0.0 W m−2) when the UV filter was attached to the 45-W fluores-
cent lamp while the visible light reading recorded was 265 W m−2
.
Fig. 12a shows the photocatalytic degradation of RR4 dye by PU1-
350 and pristine P25 under the 45-W fluorescent lamp (UV–Vis).
It was found that PU1-350 took only 30 min to complete the RR4
degradation as compared with pristine P25 that needed longer irra-
diation time to complete the same degradation process. The pseudo
first order rate constant K values for the degradation of RR4 by
PU1-350 was about 1.8 times faster than pristine P25 where the val-
ues were 0.158 and 0.088 min−1, respectively. High photocatalytic
activity of PU1-350 was due to the presence of C as the electron
scavenger that reduced the electron-hole recombination during the
photocatalytic process as previously confirmed by the lowest PL
intensity of PU1-350 sample as well as its ability to utilize visi-
ble light component of the light source. Similar observation was
reported by Janus et al. [46] where the C coated on the TiO2 surface
photoaccelerated the process by acting as the electron scavenger.
The photocatalytic degradation of RR4 under visible light irra-
diation was also displayed in Fig. 12a (UV–filter). Pristine P25
was observed to degrade RR4 dye by about 42% but no further
degradation was observed beyond that level even after prolonged
irradiation. In fact the removal of RR4 dye by pristine TiO2 here was
mainly from the adsorption process where no photocatalytic reac-
tion actually occurred as indicated in Fig. 10. Similar observation
(no photocatalytic degradation) was also observed when pristine
P25 was applied for the degradation of MB (Fig. 12b) and phenol
(Fig. 12c) under visible light irradiation. However, photocatalytic
activity was observed when PU1-350 was irradiated under visible
light irradiation (UV–filter) where it completed the degradation of
RR4 within 45 min of contact time. PU1-350 had also degraded MB
and phenol under visible light irradiation as can be seen in Fig. 12b
and c. These results clearly proved that the band gap energy of PU1-
350 was red shifted due to the mixing of N 2p and O 2p states [3,16]
thus allowing it to function under visible light. The calculation of the
electronic state of N-doped TiO2 by Chamber et al. [52] indicated
that there was virtually no shift of the valence band by mixing the N
2p and O 2p states within the N-doped TiO2. However, they found
that the N impurity level appeared slightly above the valence band
.
During the removal of pollutants under light irradiation, two
processes occurred simultaneously namely adsorptive and photo-
catalytic processes. In order to minimize the effect of the physical
adsorption and to make the photocatalytic process dominant, a
study was also done whereby the samples were first pre-saturated
with the respective pollutants prior to switching on the light for
photocatalysis. In this way, it was assumed that only photocat-
alytic process occurred in the removal of the pollutants. Fig. 11
Fig. 10. Adsorption study of pristine P25 and PU1-350 under RR4, MB and phenol.

W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
91
100
100
80
60
40
20
0
a
c
b
80
60
40
20
0
0
2
4
6
0
20
40
60
Time (h)
Time (min)
100
80
60
40
20
0
PU1-350 (UV-Filter)
PU1-350 (UV-Filter)
P25
(UV-Filter)
PU1-350 (Lamp)
P25 (UV- Filter)
PU1-350 (Lamp)
P25
P25 (Lamp)
(Lamp)
PU1-350 (Solar)
PU1-350 (Solar)
P25
P25 (Solar)
(Solar)
0
1
2
3
Time (h)
Fig. 12. The effect of different light sources on the photocatalytic activities of the pristine P25 and PU1-350 photocatalysts in the degradation of (a) 30 mg L−1 RR4, (b)
12 mg L−1 MB (b) and (c) 10 mg L−1 phenol solution under 45-W fluorescent lamp and (a) 60 mg L−1 RR4, b) 36 mg L−1 MB (b) and (c) 30 mg L−1 phenol solution under solar
irradiation.
(∼0.8 eV). This is in agreement with our XPS results shown in Fig. 5c
where it shows that N was incorporated into TiO2 lattice by replac-
ing an O atom. It is therefore suggested that an isolated N impurity
level above the valence band in PU1-350 may be responsible for the
visible light absorption.
achieved by PU1-350 were much higher than that of the pris-
tine TiO2-P25. For PU1-350, complete mineralization of RR4, MB
and phenol was achieved after 5, 7 and 7 h of irradiation, respec-
tively, using the compact 45 W fluorescent light source. For pristine
TiO2–P25 sample, the remaining TOC values were still significantly
higher at 47, 40 and 34%, respectively, for RR4, MB and phenol under
the same conditions.
Solar light which contains about 5% UV radiation [53] (22 W m−2
in this study) should produce excellent photocatalytic activity of
the samples. As shown in Fig. 12a–c, a complete degradation of 60,
36 and 30 mg L−1 of RR4, MB and phenol by PU1-350 sample under
solar irradiation took only 20, 15 and 15 min, respectively, com-
pared to pristine P25 which took 30, 45 and 30 min using the same
pollutants. It can be summarized that C coated N doped P25 was
faster by 2 times than the pristine TiO2-P25 in the photocatalytic
degradation of RR4, MB and phenol, respectively, under solar irra-
diation. Therefore this study proved that thin C coated N doped P25
has a great potential to be used for solar detoxification of organic
pollutants.
9
RR4 (P25)
RR4 (PU1-350)
8
7
MB (P25)
6
5
4
3
2
1
0
MB (PU1-350)
Phenol (P25)
Phenol (PU1-350)
3.4. Mineralization study
Mineralization is defined as the complete decomposition of
organic compounds into CO2 and H2O. Achieving a complete min-
eralization should therefore be the target of any photocatalytic
processes. The degree of mineralization can be indicated by moni-
toring the total organic carbon (TOC) values of the treated samples.
TOC is considered the most relevant for the determination of
organic pollution. The TOC values of phenol, MB and RR4 samples
after photocatalytic treatment with both pristine P25 and PU1-350
samples are given in Fig. 13. The percentages of mineralization
0
2
4
6
8
10
Time (h)
Fig. 13. Mineralization process of RR4, MB and phenol of pristine P25 and PU1-350
under 45-W fluorescent lamp.

92
W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
Fig. 14. Photocatalytic efficiency of PU1-350, C coated N doped P25 upon recycled applications in the degradation of RR4 dye irradiated under a 45 W fluorescent lamp.
Table 3
than P25 under visible light fluorescent lamp and 2 times better
under solar irradiation. Finally, C coated N doped P25 was very sta-
ble and possessed excellent sustainable photocatalytic activity with
no loss of C and N detected even after 5 times of reuse.
Carbon content of C coated N doped TiO2-P25 upon prolonged irradiations with a
45 W fluorescent lamp in distilled water for 0, 24, 48 and 72 h.
Irradiation times (h)
Carbon content (%)
0
24
48
72
1.39 ( 0.03) 1.35 ( 0.03) 1.35 ( 0.03) 1.32 ( 0.03)
Nitrogen content (%) 2.27 ( 0.03) 2.24 ( 0.03) 2.24 ( 0.03) 2.22 ( 0.03)
Acknowledgements
We would like to thank the Malaysian Ministry of Science
and Technology (MOSTE) for providing generous financial support
under Scifund and FRGS grants to conduct this study and Universiti
Sains Malaysia for providing all the needed facilities.
3.5. The stability of the photocatalyst
For the stability study of the photocatalyst, the suspended
PU1-350 particles in ultrapure water were exposed to the 45 W flu-
orescent lamp irradiation for 24, 48 and 72 h, respectively. Table 3
shows the amount of N and C content of PU1-350 under prolonged
irradiation times. The result shows that the N and C content of the
irradiated.
PU1-350 particles remained constant even up to 72 h of direct
irradiation. Fig. 14 shows the result of the repeated reuse or
recycling of the PU1-350 photocatalyst for the degradation of RR4
dye upon 15 min of irradiation time. It was observed that each
time of recycled application produced 97.6 0.6% removal of RR4
indicating a sustainable photocatalytic efficiency characteristic. In
other words, a strong interaction of C with P25 occurred due to its
strong chemisorption on the surface of TiO2 where it was not easily
removed even after 5 times of repeated usage.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.molcata.
2013.11.030.
References
[1] M.R. Hoffmann, S.T. Martin, W. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995)
69–96.
[2] A. Kudo, K. Omori, H. Kato, J. Am. Chem. Soc. 121 (1999) 11459.
[3] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293 (2001) 269–271.
[4] Y. Park, W. Kim, H. Park, T. Tachikawa, T. Majima, W. Choi, Appl. Catal., B.
Environ. 91 (2009) 355–361.
[5] G.T. Lim, K.H. Kim, J. Park, S-H. Ohk, J-H. Kim, D.L. Cho, J. Ind. Eng. Chem. 16
(2010) 723–727.
[6] C. Han, M. Pelaez, V. Likodimos, A.G. Kontos, P. Falaras, K. O’Shea, D.D. Dionysiou,
Appl.Catal., B: Environ. 107 (2011) 77–87.
4. Conclusion
A visible light active C coated N doped P25 photocatalyst was
successfully prepared via controlled doping of urea into P25 pow-
der under a relatively low temperature heating process. It was
observed that utilization of 3:1 P25 to urea ratio with 350 ◦C calci-
nation temperature produced C coated N doped P25 with optimum
photocatalytic activity. The treatment process however caused a
reduction of BET surface areas of the photocatalyst but did not pro-
duce any phase transformation even at 500 ◦C. From the HRTEM
image, it was observed that a thin C graphite layer of 2.04 nm was
coated onto the surface of P-25 particles. XPS analysis confirms that
the modified photocatalyst possessed N which substitutes the oxy-
gen molecules to produce Ti–N at binding energy of 395.7 eV. The
bandgap energy of the C coated N doped P25 was 2.6 eV as seen from
its UV–Vis DRS spectrum. The enhancement of the photocatalytic
efficiency was postulated to be due to the significant reduction of
the e-h+ recombination, as indicated by the high excitonic PL signal
intensity for the optimum PU1-350 sample. Moreover, the presence
of doped N makes PU1-350 to be active under visible light irradi-
ation. The efficiency of the modified catalyst was 1.8 times better
[7] H. Sun, H. Liu, J. Ma, X. Wang, B. Wang, L. Han, J. Hazard. Mater. 156 (2008)
552–559.
[8] N. Todorova, T. Giannakopoulou, T. Vaimakis, C. Trapalis, Mater. Sci. Eng. B 152
(2008) 50–54.
[9] C. He, D. Shu, M. Su, D. Xia, M.A. Asi, L. Lin, Y. Xiong, Desalination 253 (2010)
88–93.
[10] D. Chen, Z. Jiang, J. Geng, Q. Wang, D. Yang, Ind. Eng. Chem. Res. 46 (2007)
2741–2746.
[11] A. Roguska, A. Kudelski, M. Pisarek, M. Opara, M.J. Czachor, Appl. Surf. Sci. 257
(2011) 8182–8189.
[12] L. Tao, Z. Zhu, F. Li, Colloids Surf., A 425 (2013) 92–98.
[13] T. Morikawa, T. Ohwaki, K. Suzuki, S. Moribe, S.T. Kubota, Appl. Catal., B: Envi-
ron. 83 (2008) 56–62.
[14] F. Li, Y. Zhao, Y. Hao, X. Wang, R. Liu, D. Zhao, D. Chen, J. Hazard. Mater. 239
(2012) 118–127.
[15] P. Wang, T. Zhao, R. Wang, T.T. Lim, Water Res. 45 (2011) 5015–5026.
[16] J.A.R. Herrera, J. Kiwi, C. Pulgarin, J. Photochem. Photobiol., A 205 (2009)
109–115.
[17] T. Tsumura, N. Kojitani, I. Izumi, N. Iwashita, M. Toyoda, M. Inagaki, J. Mater.
Chem. 12 (2002) 1391–1396.
[18] J.H. Park, S. Kim, A.J. Bard, Nano Lett. 6 (2006) 24–28.
[19] S. Sato, R. Nakamura, S. Abe, Appl. Catal., A: Gen. 284 (2005) 131–137.
[20] A. Fujishima, K. Honda, Nature 238 (1972) 37–38.
[21] A. Zaleska, Recent Pat. Eng. 2 (2008) 157–164.

W.I. Nawawi, M.A. Nawi / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 83–93
93
[22] Y.C. Hsu, H.C. Lin, C.W. Lue, Y.T. Liao, C.M. Yang, Appl. Catal., B: Environ. 89
[39] M. Inagaki, Carbon 50 (2012) 3247–3266.
(2009) 309–314.
[40] J.M. Timothy, Ultrason. Sonochem. 10 (2003) 175–179.
[23] J. Zhang, Z. Huang, Y. Xu, F. Kang, New Carbon Mater. 26 (2011) 63–70.
[24] B. Tryba, A.W. Morawski, T. Tsumura, M. Toyoda, M. Inagaki, J. Photochem.
Photobiol., A 167 (2004) 127–135.
[25] H. Zhang, J.F.Banfield, J. Phys. Chem. B 104 (2000) 3481–3487.
[26] R. Nakamura, T. Tanaka, Y. Nakato, J. Phys. B 108 (2004) 10617.
[27] C. Shifu, L. Yunzhang, Chemosphere 67 (2007) 1010–1017.
[28] M. Xing, J. Zhang, F. Chen, Appl. Catal., B: Environ. 89 (2009) 563–569.
[29] Q. Xu, D.V. Wellia, M.A. Sk, K.H. Lim, J.S.C. Loo, D.W. Liao, R. Amal, T.T.Y. Tan, J.
Photochem. Photobiol., A 210 (2010) 181–187.
[30] L. Ge, M. Xu, H. Fang, Thin Solid Films 515 (2007) 3414–3420.
[31] F. Meng, Z.L. Hong, J. Amdt, M. Li, M.J. Zhi, F. Yang, N.Q. Wu, Nano Res. 5 (2012)
213–221.
[32] X. Chen, P. Glans, X. Qiu, S. Dayal, W.D. Jennings, K.E. Smith, C. Burda, J. Guo, J.
Electron. Spectrosc. Relat. Phenom. 162 (2008) 67–73.
[41] S. Livraghi, M.C. Paganini, E. Giamello, A. Selloni, C. DiValentin, G. Pacchioni, J.
Am. Chem. Soc. 128 (2006) 15666–15671.
[42] Z. Wu, F. Dong, W. Zhao, S. Guo, J. Hazard. Mater. 157 (2008) 57–63.
[43] M.A. Henderson, W.S. Epling, C.H.F. Peden, C.L. Perkins, J. Phys. Chem. B 107
(2003) 534–545.
[44] L.Q. Jing, B.F. Xin, F.L. Yuan, L.P. Xue, B.X. Wang, H.G. Fu, J. Phys. Chem. B 110
(2006) 17860–17865.
[45] L.Q. Jing, Z.L. Xu, X.J. Sun, J. Shang, W.M. Cai, Appl. Surf. Sci. 180 (2001) 308–314.
[46] M. Janus, M. Inagaki, B. Tryba, M. Toyoda, A.W. Morawski, Appl. Catal., B: Envi-
ron. 63 (2006) 272–276.
[47] O. Carp, C. L. Huisman, A., Reller, Prog Solid State Chem. 32 (2004) 33-177.
[48] M.N. Chong, B. Jin, C.W.K. Chow, C. Saint, Water Res. 44 (2010) 2997–3027.
[49] K.H. Wang, Y.H. Hsieh, P.W. Chao, C.Y. Cgang, J. Hazard. Mater. 95 (2002)
161–174.
[33] F. Jia, Z. Yao, Z. Jiang, C.X. Li, Catal. Commun. 12 (2011) 497–501.
[34] Q. Xiao, J. Zhang, C. Xiao, Z. Si, X. Tan, Sol. Energy 82 (2008) 706–713.
[35] S.M. Yun, K. Palanivelu, Y.H. Kim, P.H. Kang, Y.S. Lee, J. Ind. Eng. Chem. 14 (2008)
667–671.
[50] G. Liu, X. Wanga, L. Wang, Z. Chen, F. Li, H.M. Gao Qin, J. Cheng, Colloid Interface
Sci. 334 (2009) 171–175.
[51] G. Liu, X. Wang, Z. Chen, H.M. Cheng, G.Q. Lu, J. Colloid Interface Sci. 329 (2009)
331–338.
[36] L. Mi, P. Xu, H. Shen, P.N. Wang, J. Photochem. Photobiol., A 193 (2008)
222–227.
[52] S.A. Chamber, S.H. Cheung, V. Shutthanandan, S. Thevuthsan, M.K. Bowman,
A.G. Joly, Chem. Phys. 339 (2007) 27–35.
[37] L.W. Zhang, H.B. Fu, Y.F. Zhu, Adv. Funct. Mater. 18 (2008) 2180–2189.
[38] Y. Wang, R. Shi, J. Lin, Y. Zhu, Appl. Catal., B: Environ. 100 (2010) 179–183.
[53] R.P. Sinha, M. Klisch, A. Vaishampayan, D.P. Hader, Acta Protozoologica 38
(1999) 273–290.