Preparation of nitrogen doped K2Nb4O11 with high photocatalytic activity for degradation of organic pollutants

Article abstract of DOI:10.1016/j.apcata.2011.04.015

Nitrogen doped K₂Nb₄O₁₁ (K ₂Nb₄O₁₁-N) has been prepared by solid state reaction between K₂Nb₄O₁₁ and urea at 400 °C. K₂Nb₄O₁₁-N has been characterized by XRD, SEM, XPS and UV/vis diffuse reflectance. The photodegradation of various organic pollutants in water by this material, including Orange G (OG), bisphenol A (BPA) and pentachlorophenol (PCP) have been studied at λ > 330 nm and >399 nm. The results show that the photocatalytic activity of K ₂Nb₄O₁₁-N at >399 nm is higher than those of K₂Nb₄O₁₁ and Degussa TiO₂ P25, indicating the activating effect of nitrogen doping. A mechanism for the photodegradation of organic substrates by K₂Nb₄O ₁₁-N is proposed.

Full text of DOI:10.1016/j.apcata.2011.04.015

Applied Catalysis A: General 402 (2011) 23–30
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Preparation of nitrogen doped K Nb O with high photocatalytic activity for
2
4
11
degradation of organic pollutants
Yongfu Qiu , Lei Wanga,b,c, Chi-fai Leunga,b, Guijian Liub,c, Shihe Yang , Tai-Chu Lau
a
d
a,b,∗
a
State Key Laboratory in Marine Pollution and Department of Biology and Chemistry„ City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, China
Advanced Laboratory of Environmental Research and Technology (ALERT), Joint Advanced Research Center, USTC-CityU, Suzhou, Jiangsu 215123, China
CAS Key Laboratory of Crust-Mantle Materials and the Environment, School of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui 230026, China
Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Nitrogen doped K Nb O (K Nb O -N) has been prepared by solid state reaction between K Nb O and
2
4
11
2
4
11
2
4
11
Received 20 February 2011
Received in revised form 4 April 2011
Accepted 11 April 2011
◦
urea at 400 C. K2Nb4O11-N has been characterized by XRD, SEM, XPS and UV/vis diffuse reflectance. The
photodegradation of various organic pollutants in water by this material, including Orange G (OG), bisphe-
nol A (BPA) and pentachlorophenol (PCP) have been studied at ꢀ > 330 nm and >399 nm. The results show
that the photocatalytic activity of K2Nb4O11-N at >399 nm is higher than those of K2Nb4O11 and Degussa
TiO2 P25, indicating the activating effect of nitrogen doping. A mechanism for the photodegradation of
organic substrates by K2Nb4O11-N is proposed.
Available online 16 April 2011
Keywords:
Photodegradation
Niobate
©
2011 Published by Elsevier B.V.
Nitrogen-doped
Organic pollutants
1
. Introduction
photocatalysts for the degradation of organic pollutant has received
little attention until very recently, when Ye reported efficient pho-
todegradation of Rhodamine B under visible-light irradiation using
nitrogen-doped lamellar niobic acid [8]. Although the layered per-
ovskite type niobates and their protonated derivatives show high
photoactivity under visible-light irradiation, these photocatalyts
are generally not very stable because the layered structure is sus-
ceptible to collapse and is easily hydrated after protonation even
under atmospheric conditions [8–11].
Organic pollutants in water have been causing serious environ-
mental problems. Photocatalytic degradation of these pollutants
using sunlight is an attractive solution to this global problem.
Semiconductors have been used as highly active photocatalysts,
such as ZnO [1], TiO2 [2–4], BiWO6 [5], CdS [6], Fe O3 [7], and
2
HNbO3 [8]. These semiconductors, however, can only be excited
by UV light due to their large band gaps. For better utilization
of sunlight and indoor illumination, Nb-based photocatalyts have
recently been studied because the layered perovskite type niobates
can be excited by visible light [8–11]. In 1981, Dion reported the
family of layered perovskite type niobates, generally formulated
as AM Nb O (A = K, Rb, Cs; M = La, Ca, etc.), which show notice-
able photocatalytic activity [9]. Soon after, Yoshimura reported a
layered perovskite type niobate, RbPb Nb O10, that could photo-
We are interested in the use of non-layered niobate salts as
photocatalysts as we anticipate that they would be more robust
than the layered niobates. A suitable candidate is K Nb O11, which
2
4
is constructed from NbO6 octahedra and has a tetragonal tung-
sten bronze (TB) crystal structure with triangle, quadrilateral and
pentagonal tunnels. The pentagonal and quadrilateral tunnels are
occupied by K cations and the triangle tunnels by Nb cations [12]. It
2
3
10
2
3
catalytically generate H2 from water by visible light [10]. Recently,
Wu reported the high photoactivity of K2 La Ti NbxO10 and
their protonated derivatives for water splitting under visible light
has been reported that Cu-doping of K Nb O results in increased
photocatalytic activity for the degradation of acid red G under UV
irradiation [13].
2 4 11
− x
2
3 − x
[
11]. However, the use of the layered perovskite type niobates as
In this paper we attempt to prepare a highly active pho-
tocatalyst by doping K Nb O
with nitrogen (denoted as
K Nb O -N). The efficiency of K Nb O11-N for the photodegra-
2
4
11
2
4
11
2
4
dation of organic pollutants at >399 nm irradiation has been
investigated by using Orange G (OG), bisphenol A (BPA) and
∗ Corresponding author at: State Key Laboratory in Marine Pollution and Depart-
ment of Biology and Chemistry, City University of Hong Kong, Tat Chee Avenue,
Kowloon, Hong Kong, China. Tel.: +852 2788 7811; fax: +852 2788 7811.
E-mail address: bhtchlau@cityu.edu.hk (T.-C. Lau).
pentachlorophenol (PCP) as substrates. OG,
a synthetic azo
dye with formula C16H10N Na O7S (7-hydroxy-8-phenlyazo-1,
2
2
2
0
926-860X/$ – see front matter © 2011 Published by Elsevier B.V.
doi:10.1016/j.apcata.2011.04.015

2
4
Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
3
-naphthalenedisulfonic acid disodium salt), is an endocrine dis-
ruptor [14]. It is a common reagent in molecular biology and is
used in histology in many staining formulations. Bisphenol A is also
a known endocrine disruptor, which has been widely used for the
production of polycarbonate (PC) and epoxy resins used in food
containers [15–17]. Its concentration in wastewaters is increas-
ing worldwide [17]. Another common toxic pollutant in aquatic
environment is pentachlorophenol (PCP). PCP has a relatively long
life-time in water; hence a rapid method for its decomposition is
necessary [18,19]. Our results show that the K Nb O11-N photo-
2
4
catalyst has a much higher photoactivity than pure K Nb O or
2
4
11
Nb O5. It is also more active than Degussa P25 TiO2 at >399 nm.
2
This indicates that the photoactivity of K Nb O has been greatly
2
4
11
enhanced by nitrogen doping.
2
. Experimental
2.1. Synthesis
K Nb O was prepared by heating a mixture of 0.40 g Nb O5
2
4
11
2
◦
and 1.60 K CO at 900 C for 24 h. For nitrogen doping, 1.0 g of
2
3
K Nb O was finely ground with 10.0 g urea, and the mixture was
2
4
11
◦
then heated at 400 C for 4 h to give a yellow product. TiO P25 was
2
Fig. 1. XRD pattern of: (A) K2Nb4O11 (B) K2Nb4O11-N and (C) standard K2Nb4O11
JCPDS 31-1059).
purchased from German Degussa Corporation.
(
2.2. Characterization
The materials were characterized by powder X-ray diffraction
XRD), scanning electron microscopy (SEM), X-ray photoelectron
spectroscopy (XPS), UV/vis diffuse reflectance and photolumi-
nescence spectroscopy (PL). The XRD analysis was performed
on a Rigaku D-max X-ray diffractometer with Cu K␣ irradiation
Agilent 1200 series liquid chromatograph equipped with an Agi-
lent C18 column, a UV detector and a API 2000 mass spectrometer
(
(electrospray ionization). The mass spectrometer was operated at
the anionic mode in the m/z range 50–500 for LC/MS and 50–300
for LC/MS/MS. Aliquots were withdrawn at various time intervals
and centrifuged before analysis.
◦
(
ꢀ = 1.5406 A˚ ) at a scanning speed of 0.025 /s over the 2ꢁ range
◦
of 20–70 . The morphologies were directly examined by a Philips
XL30 environmental scanning electron microscope (ESEM) at an
accelerating voltage of 10 kV. The catalyst surface analysis was done
by X-ray photoelectron spectroscopy using a Leybold Heraeus-
Shengyang SKL-12 electron spectrometer equipped with a VG
CLAM 4 MCD electron energy analyzer, with Al-K␣ as the excitation
source. UV/vis diffuse reflectance was performed on a Perkin Elmer
Lambda 750 UV/vis Spectrophotometer. Photoluminescence (PL)
spectra were measured using a FluoroMax-3 spectrofluorimeter
equipped with a pulsed xenon lamp as light source.
3. Results and discussion
3.1. X-ray diffraction (XRD)
Powder X-ray diffraction (XRD) shows nearly identical patterns
for all the samples prepared. Typical XRD patterns of K Nb O and
2
4
11
K Nb O -N are shown in Fig. 1A and B, respectively. For K Nb O
2
4
11
2
4
11
(Fig. 1A), all the diffraction peaks can be indexed as a tetragonal
tungsten bronze structure with lattice constants of a = 0.126 nm
and c = 0.398 nm, which together with the intensity distribution
2.3. Photocatalytic degradation experiments
are consistent with those of the standard card for the K Nb O
2
4
11
A 200 W xenon arc lamp (Newport, Model 71232) was used
as the light source. OG, BPA or PCP aqueous solution (30 mL,
0 mg/L) and photocatalyst (10 mg) were put in a quartz tube
crystal (Joint Committee on Powder Diffraction Standard, JCPDS 31-
059). When doped with nitrogen, the XRD pattern is unchanged, as
1
2
shown in Fig. 1B, indicating that there is no effect of nitrogen dop-
ing on the crystal structure of K Nb O11, this suggests that doping
reactor (12 mm in diameter and 200 mm in length) and the mix-
ture was sonicated for 5 min to disperse the catalyst in the
aqueous solution. The distance between the liquid surface and
the light source was about 11 cm. Before irradiation, the sus-
pension was stirred in the dark for one hour so as to allow
adsorption–desorption equilibrium to be established. The Infrared
and UV light emitted from the Xe-lamp was filtered by a water
jacket and a composited cutoff filter (Scott AG KV 399). Sam-
ples were collected at regular time intervals and centrifuged
before analysis. The concentrations of substrates were determined
spectrophotometrically using a Shimadzu UV-1700 UV/vis spec-
trophotometer.
2
4
occurs only on the surface.
3.2. SEM
The SEM photographs of K Nb O and K Nb O11-N are shown
2
4
11
2
4
in Fig. 2A and B, respectively. From Fig. 2A, the sizes and shapes
of particles are inhomogeneous and the surface is clean. When
◦
K Nb O was heated with urea at 400 C for 4 h, the surface of the
2
4
11
sample became flock-like, but the sizes and shapes of the particles
were not significantly changed. The SEM and XRD results together
indicate that the morphology and crystal structure of K Nb O
2.4. Analysis of photodegradation intermediates and products
2
4
11
are not affected by nitrogen doping, but the surface of the sam-
ple becomes rough, suggesting that nitrogen doping may have
occurred on the surface of the sample.
The intermediates and products formed during the degradation
of bisphenol A were identified by LC/MS and LC/MS/MS using an

Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
25
Fig. 3. XPS spectrum in the whole energy range. (A) K2Nb4O11 and (B) K2Nb4O11-N.
be observed in K Nb O and K Nb O11-N. However, N can only
2
4
11
2
4
be seen in K Nb O11-N, indicating the successful doping of N onto
2
4
the surface of K Nb O11. The N concentration is calculated to be 3.9
2
4
atom% using the equation,
IN/SN
CN = ꢀ
Ii/Si
i
where CN is the nitrogen concentration, IN and I are the peak inten-
i
sities of nitrogen and other elements, respectively; SN and S are the
i
relative sensitivity factors of nitrogen and other elements, respec-
tively.
To further determine the chemical states of the elements, core
level XPS spectra of Nb, O and N in K Nb O and K Nb O11-N have
Fig. 2. SEM images of: (A) K2Nb4O11 and (B) K2Nb4O11-N.
2
4
11
2
4
been obtained and are shown in Fig. 4. Fig. 4A1 and B1 shows that
the binding energies of Nb 3d5/2 for K Nb O and K Nb O11-N
2
4
11
2
4
3
.3. XPS
The X-ray photoelectron spectra (XPS) of K Nb O and
11
are 206.8 and 206.9 eV, respectively; which are consistent with the
reported values [20]. The chemical shifts of the binding energies
of Nb 3d5/2 in these two materials are small. However, the full
widths at half maximum (FWHM) of the Nb 3d5/2 peaks are dif-
ferent. The FWHM of Nb 3d5/2 for K Nb O and K Nb O11-N are
2
4
K Nb O11-N in wide energy range are shown in Fig. 3. No significant
2
4
contamination, besides carbon, is found in the spectra. The bind-
ing energy was determined by reference to C 1s line at 284.8 eV.
In the whole energy range spectrum the elements K, Nb and O can
2
4
11
2
4
1.7 and 1.9 eV, respectively. The broadening of the Nb 3d5/2 peak
indicates that the electron density on the Nb atoms in K Nb O11-N
2
4
Fig. 4. XPS spectra of the elements in K2Nb4O11 (A) and K2Nb4O11-N (B). 1, 2 and 3 refer to Nb, O and N, respectively.

2
6
Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
Fig. 5. (A) UV–vis diffuse reflectance spectra of K2Nb4O11 and K2Nb4O11-N; (B) The
curve deduced from A according to the equation (˛hꢂ)2 = A (hꢂ − Eg).
Fig. 6. (A) Plot of C/Co (C is the concentration at time t, Co is the initial concentration)
versus time for the photo-degradation of OG using 330 nm cut off filter; (B) Plot
of C/Co versus time for photo-degradation of OG using 399 nm cut off filter; (a)
K2Nb4O11, (b) TiO2 P25 and (c) K2Nb4O11-N.
is higher than that in K Nb O [21]. The main O 1s peak at 530.8
2
4
11
and 530.6 eV in Fig. 4A2 and B2 are assigned to lattice oxygen of
K Nb O and K Nb O11-N, respectively [22–24]. A higher binding
2
4
11
2
4
energy shoulder is found for both samples at about 532.6 eV, this
is assigned to a mixture of surface hydroxyl and carbonate groups
3.5. Photocatalytic degradation of orange G
[
21]. Nitrogen is found only on K Nb O11-N and the core-level N 1s
2 4
The photodegradation of OG with 330 nm cutoff filter using
XPS is shown in Fig. 4B3. The N 1s spectrum is deconvoluted into
two components with peak energies of 398.7 and 400.8 eV, which
are assigned to be (N) or (NO) and (NO) or (NO ) , respectively
K Nb O and K Nb O11-N as photocatalysts was investigated and
2
4
11
2
4
results are shown in Fig. 6A. The degradation of OG was neg-
ligible after 4 h when K Nb O was used as the photocatalyst.
i
O
i
2 O
2
4
11
[
22]. (N) represents N in the interstitial space, (NO)O denotes NO
i
On the other hand, when K Nb O11-N was used, nearly 90% OG
2 4
sitting at the site for the lattice O. Similarly, (NO) and (NO ) des-
ignate the interstitial NO and substitutional NO2 for the lattice O,
respectively.
i
2
O
was degraded after 2 h of irradiation, indicating nitrogen dop-
ing greatly enhances the photocatalytic activity of the K Nb O11.
2
4
Control experiments show that both light and the photocatalyst
are required for the degradation of OG. The degradation of OG
by K Nb O11-N is only slightly less efficient than TiO2 P25 when
2
4
3.4. UV/vis diffuse reflectance
3
30 nm cutoff filter is used. On the other hand, when 399 nm cut-
off filter was used, the photocatalytic activity of K Nb O11-N is
2
4
The light absorption of the samples can be measured by using
higher than that of TiO2 P25. As shown in Fig. 6B, nearly 90% of
OG is degraded over the K Nb O11-N after 12 h of irradiation when
UV/vis diffuse reflectance spectroscopy [25]. Fig. 5 shows the
UV/vis diffuse reflectance spectra of K Nb O and K Nb O11-N.
2
4
2
4
11
2
4
a 399 nm cutoff filter is used, while only 46% is degraded over TiO2
P25.
It is known that the optical absorption coefficient near the band
2
edge follows the equation (˛hꢂ) = A (hꢂ − Eg) for a direct-band
Fig. 7A shows the spectral changes of OG during irradiation using
gap material; where ˛, h, ꢂ, Eg and A are the absorption coeffi-
cient, Planck constant, light frequency, band gap, and a constant,
respectively [25]. This relationship gives the band gaps (Eg) by
extrapolating the straight portion of (˛hꢂ)2 against hꢂ plots to
3
99 nm cutoff filter with K Nb O11-N. The main absorption band
2 4
of OG is at around 478 nm, which decreases with time upon irra-
diation, but ꢀmax remains at around 478 nm, indicating that the
photodegradation does not occur by a dye self-photosensitized
oxidative mechanism [26]. Apart from the peak at 478 nm, the
photodegradation of OG by K Nb O11-N also results in the dis-
the point ˛ = 0, which are 3.27 eV and 3.12 eV for K Nb O and
2
4
11
K Nb O -N, respectively. The absorption edge of K Nb O11-N
2
4
11
2
4
2
4
shifts to longer wavelength compared with that of the undoped
one, indicating that nitrogen-doping has a narrowing effect on the
band gap of the material on the surface [25].
appearance of the peak at 330 nm, indicating that both the OG
chromophores and the aromatic rings have been destroyed [8].
Also the total organic carbon (TOC) value of the solution decreases

Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
27
dark light on
A
a
1.0
0.8
0.6
0.4
0.2
b
c
d
e
4
-4
0
8
12
16
20
Irradiation time (h)
dark light on
a
b
c
B
1
.0
.8
.6
.4
.2
0
d
e
0
0
0
-
30
0
30
60
90
120
150
180
Irradiation time (minutes)
Fig. 7. (A) Spectral changes of OG during irradiation using 399 nm cutoff filter with
K2Nb4O11-N; (B) Plot of (TOC)/(TOC)o [(TOC) is the total organic carbon at time t,
Fig. 8. (A) Plot of C/Co (C is the concentration at time t, Co is the initial concentration)
versus time for photo-degradation of BPA (A) and PCP (B) using a 399 nm cutoff filter
with (a) no catalyst; (b) Nb2O5; (c) K2Nb4O11; (d) Degussa TiO2 P25; (e) K2Nb4O11-N.
(
TOC)o is the initial total organic carbon] versus time for photo-degradation of OG
using a 399 nm cut off filter with K2Nb4O11-N.
by approximately 25% at 90% OG conversion after 12 h of irradia-
tion (see Fig. 7B). We conclude that the OG is mainly degraded to
aliphatic organic compounds and is only partially mineralized to
CO2 and/or CO.
BPA was degraded after 6 h and 80% of PCP was degraded after 2 h.
This photoactivity is higher than that of Degussa TiO2 P25, and is
much higher than that of pure K Nb O and Nb O5. These results
2
4
11
2
confirm that nitrogen doping greatly enhances the photoactivity of
K Nb O11.
To assess the stability of the photocatalyst, a sample of
2
4
K Nb O11-N was aged at atmospheric ambient conditions for six
2
4
The intermediates formed during the degradation of bisphenol A
were identified by LC/MS and LC/MS/MS experiments. Fig.S3 shows
the total ions chromatogram (TIC) from LC/MS for the intermedi-
months and its photocatalytic activity was then tested. As shown in
Fig. S2, the photocatalytic activity of the aged K Nb O11-N sample
2
4
decreases by about 6% compared with the freshly prepared sam-
ple. This shows that that K Nb O11-N photocatalyst is reasonably
ates formed during bisphenol A degradation by K Nb O11-N after
2
4
2
4
4 h. Five major intermediates were identified, as listed in Table 1.
stable when stored under ambient conditions.
In the liquid chromatogram, the peak f is bisphenol A with a m/z
−
of 227. The m/z value for each peak corresponds to the [M–H]
3
.6. Photocatalytic degradation of BPA and PCP
ion that arises from deprotonation of a hydroxyl group. Since a
reverse-phase LC column and a polar mobile phase were used, the
products were eluted in decreasing order of polarity. Each peak was
identified by interpretation of the MS/MS spectra. Similar interme-
diates have also been reported for the degradation of bisphenol
A using Fenton’s reagent, TiO2 photocatalyst and a supported Ru
polypyridyl catalyst [29–31]. The compounds in peak a and b have
the same molecular weight and fragment ions, indicating that they
are closely related isomers. The intermediates d and e are assigned
as the hydroxylated BPA and the corresponding quinine, respec-
tively. The formation of these intermediates is consistent with
It has been suggested by several authors that the photocat-
alytic degradation of dyes under visible light may be induced by
light absorbed by the dye rather than by the catalyst [27,28]. In
order to understand the photoactivity of K Nb O11-N under visi-
ble light, bisphenol A (BPA) and pentachlorophenol (PCP), which
do not absorb in the visible region and are emerging pollutants in
2
4
water, were selected as substrates. For comparison, Nb O5, pure
K Nb O and Degussa TiO2 P25 were also used as photocatalysts
2
2
4
11
and results are shown in Fig. 8. In the absence of photocatalyst, the
concentration of BPA and PCP remained virtually unchanged even
after 20 h of visible light irradiation (399 nm cutoff). Also, BPA and
PCP were not degraded by the photocatalysts in the dark. However,
upon visible-light irradiation in the presence of K Nb O11-N, 90% of
•
an oxidation mechanism that involves hydroxyl radicals ( OH), as
proposed in our earlier report [31]. These phenolic and quinone
derivatives of bisphenol A then undergo oxidative C–C bond cleav-
•
age by OH to give benzophenone, substituted benzophenones and
2
4

2
8
Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
Table 1
List of major intermediates (a–e) and main fragments (m/z) obtained by LC-MS.
Peak
Retention time (min)
m/z (amu)
M. W.
MS/MS fragments (relative abundance)
Proposed structure
HO
H C
3
O
H C
2
O
O
a
1.4
163
164
163(100)
07(64)
1
O
H C
3
OH
H C
2
b
c
1.7
3.1
163
133
164
134
163(100)
07(64)
1
H C
3
OH
O
H C
2
133(100)
17(16)
3(25)
1
9
CH3
HO
HO
HO
OH
OH
CH3
O
d
6.4
257
258
257(100)
2
2
2
42(18)
29(97)
14(50)
CH3
CH3
OH
e
7.0
243
227
244
228
243(100)
2
1
27(60)
33(24)
7
CH3
OH
CH3
f
12.7
227(100)
2
1
12(63)
33(20)
substituted phenols (a–c). The compound in peak a should be more
polar, hence it is assigned as the ortho-quinone derivative, while
that in peak b is assigned as the para-quinone derivative. The mech-
anism for the generation of hydroxyl radicals will be discussed later
suggested that the visible-light response in N-doped TiO2 is due
to the N 2p states isolated above the valence-band maximum of
TiO , rather than to the narrowing of the band gap by mixing the
2
N 2p and O 2p states. Recently, Lee et al. performed first-principles
density-functional calculations to study the electronic properties
of N-doped and C-doped TiO2 [36], and their results is consistent
with Irie’s model. Our results also reveals significantly enhanced
photodegradation efficiency under UV light irradiation compared
to that under visible light (see Fig. 6), again in line with the above
studies.
(
Section 3.7).
3
.7. Proposed mechanism for photodegradation by K Nb O11-N
2 4
Modified semiconductor materials have been receiving great
attention because of their potential applications in solar-driven
photocatalysis. One method for modifying the semiconductor is
non-metal doping, usually with elements that can narrow the band
gap and shift the optical adsorption edge of the photocatalyst to the
visible region [32,33]. Nitrogen is one of the most popular dopants
because of its appropriate ionic radius and the ease of doping into
oxide lattices [34–36]. The mechanism of the photocatalytic activ-
ity of nitrogen doped semiconductor materials has been discussed
On the basis of the above analysis, a band structure for the
K Nb O11-N is proposed, as schematically shown in Fig. S4. In the
2
4
K Nb O11-N photocatalyst, there exists isolated N 2p states above
2
4
the valence-band maximum of K Nb O11, which give rise to the
2
4
strong absorption enhancement in the visible region. Under visi-
ble light irradiation, electron and hole pairs would be generated
between impurity N 2p states and the conduction band of Nb 4d
(Eq. (1)).
[
34–36]. Asahi et al. found that nitrogen atoms substitute the lat-
tice oxygen sites of TiO2 and narrow the band gap by mixing the N
p and O 2p states [34]. However, in studying the decomposition
K Nb O -N + visiblelight → h+ + eCB−
(1)
2
4
11
2
−
of gaseous 2-propanol on N-doped TiO , Irie et al. observed that
The excited electrons eCB in the conduction band would move
to the surface and combine with surface-adsorbed oxygen to pro-
2
the photocatalytic quantum yields are significantly enhanced by
irradiating with UV light than by visible light [35]. Therefore they
−
−
duce O2 superoxide anion radicals. The O2 radicals could then

Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
29
2
0 mg/L PCP by 10 mg of the catalysts is the same (90 min) for four
consecutive runs.
5. Conclusions
In this report, K Nb O11-N has been successfully prepared, fully
2
4
characterized and used for the photodegradation of OG, BPA and
PCP. XRD and SEM show that the crystal structures of K Nb O11-N
2
4
and K Nb O are nearly identical, but the surface morphology has
2
4
11
been changed significantly due to the nitrogen doping. XPS and PL
indicate that the nitrogen doping primarily occurs at the surface of
K Nb O11, while UV/vis diffuse reflectance data reveal that nitro-
2
4
gen doping narrows the band gap of K Nb O on the surface. The
2
4
11
photocatalytic activity of the K Nb O11-N has been evaluated by
2
4
photodegradation of OG, BPA and PCP under visible light irradia-
tion (ꢀ > 399 nm). The results show that the photocatalytic activity
of K Nb O -N is significantly higher than that of pure K Nb O
Fig. 9. Plot of C/Co (C is the concentration at time t, Co is the initial concentration)
versus time for photo-degradation of BPA by K2Nb4O11-N at different pH: (a) pH3;
2
4
11
2
4
11
and Degussa TiO P25, highlighting the importance of nitrogen dop-
2
(
b) pH6; (c) pH10.
ing of K Nb O11. Moreover, this photocatalyst is very stable (at
2
4
least six months at room temperature), its preparation is simple
and highly reproducible, and it is easily separated from the solu-
tion after photocatalysis and can be readily recycled without loss
of activity.
•
react with H O to produce OH radicals (Eq. (2)) [37,38], which is
2
known to be a strong oxidizing species. On the other hand, the reac-
tive holes h would react with adsorbed OH on the catalyst surface
to also form OH radicals (Eqs. (3)–(4)) [37,38].
+
−
•
O2ads + 2H O → OH + OH− + H O2
•
−
•
(2)
(3)
(4)
2
2
Acknowledgments
−
+ H+
H O → OH
2
ads
The work described in this paper was supported by the Univer-
sity Grants Committee (UGC) of Hong Kong Special Administrative
Region (AoE/P-04/04) and the State Key Laboratory in Marine
Pollution (SKLMP). The equipment used in this work was also sup-
ported by a Special Equipment Grant from UGC (SEG CityU02). Dr.
Yongfu Qiu would like to thank the Natural Science Foundation of
Guangdong Province (No. 10451170003004186), the Foundation
for Distinguished Young Talents in Higher Education of Guangdong,
China (No. LYM10125) and the National Basic Research Program of
China (2010CB227306) for support of his research. We also thank
Dr. Suk-Yin Lai and Mr. Fang Ying (Centre for Surface Analysis and
Research, Department of Chemistry, Hong Kong Baptist University,
HKSAR) for XRD and XPS measurements.
OHads + h+ → •OH
−
•
The OH radicals would react with substrates to produce H O
and CO2 via various intermediates.
2
The effects of pH on the photocatalytic degradation of BPA by
K Nb O11-N were also studied (Fig. 9). It was found that photo-
2
4
catalytic activity of K Nb O11-N increases when the solution pH
decreases. This may be explained by the processes shown in Eqs.
5)–(7), which occur in the presence of H .
2
4
+
(
O2ads + H+ → HO2ads
•
−
(5)
(6)
(7)
•
•
HO2ads + H O → OHads + H O
2
2
2ads
HO2ads + eCB− → •OHads + OHads−
Appendix A. Supplementary data
These processes facilitate trapping of the electrons in the con-
duction band of K Nb O11-N and produce OHads. This trapping
•
2
4
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.apcata.2011.04.015.
mechanism slows down the recombination of electron-hole pairs
and allows a more efficient charge separation. Hence, the transfer
of trapped electrons to dissolved oxygen in the solution would be
enhanced and more holes and hydroxyl radicals will be available
for the oxidation of substrate on the catalyst surface as well as in
the solution phase.
References
[
1] M.A. Behnajady, N. Modirshahla, R. Hamzavi, J. Hazard. Mater. 133 (2006)
26–232.
2
[2] R.A. Damodar, S.J. You, H.H. Chou, J. Hazard. Mater. 172 (2009) 1321–1328.
[3] Z.B. Wu, F. Dong, Y. Liu, H.Q. Wang, Catal. Commun. 11 (2009) 82–86.
[4] A. Vijayabalan, K. Selvam, R. Velmurugan, M. Swaminathan, J. Hazard. Mater.
4
. Advantages of K Nb O11-N as photocatalyst
2 4
1
72 (2009) 914–921.
[5] L.W. Zhang, Y.J. Wang, H.Y. Cheng, W.Q. Yao, Y.F. Zhu, Adv. Mater. 21 (2009)
286–1290.
Apart from higher photocatalytic activity, K Nb O11-N is also
2
4
1
more stable than TiO . Although N-doping of TiO2 also improves
2
[
6] H.Y. Zhu, R. Jiang, L. Xiao, Y.H. Chang, Y.J. Guan, X.D. Li, G.M. Zeng, J. Hazard.
Mater. 169 (2009) 933–940.
[7] E. Rodriguez, G. Fernandez, B. Ledesma, P. Alvarez, F.J. Beltran, Appl. Catal. B:
Environ. 92 (2009) 240–249.
8] X. Li, N. Kikugawa, J. Ye, Chem. Eur. J. 15 (2009) 3538.
9] M. Dion, M. Ganne, M. Tournoux, Mat. Res. Bull 16 (1981) 1429–1435.
its photocatalytic activity under visible light, it becomes thermally
less stable [39]. Moreover, when TiO2 is illuminated with UV light,
the contact angle for water decreased to near 0, this phenomenon is
called photo-induced hydrophilic (PIH) effect [40]. This means that
TiO2 nanoparticles readily undergo photo-dissolution and disper-
sion in water and consequently they are difficult to be separated
and reused after photocatalysis. On the other hand, K Nb O11-N
[
[
[10] J. Yoshimura, Y. Ebina, J. Kondo, K. Domen, A. Tanaka, J. Phys. Chem. 97 (1993)
1970–1973.
[
11] J.H. Wu, Y.H. Cheng, J.M. Lin, Y.F. Huang, M.L. Huang, S.C. Hao, J. Phys. Chem. C
11 (2007) 3624–3628.
12] M. Lundberg, M. Sundberg, J. Solid State Chem. 63 (1986) 216–230.
2
4
1
readily settles to the bottom of the solution if left undisturbed, and it
can be recovered and reused after photocatalysis by centrifugation
or simple filtration. Moreover we found that in the photodegrada-
[
[13] G.K. Zhang, X. Zou, J. Gong, F. He, H. Zhang, S. Ouyang, H. Liu, J. Mol. Catal. A:
Chem. 255 (2006) 109–116.
14] B. Muktha, G. Madras, T.N.G. Row, U. Scherf, S. Patil, J. Phys. Chem. B 111 (2007)
7994–7998.
[
tion of PCP, K Nb O11-N can be recycled at least four times without
2
4
any loss of activity, i.e. the time for the degradation of 90% of
[15] P. Mahata, G. Madras, S. Natarajan, Catal. Lett. 115 (2007) 27–32.

3
0
Y. Qiu et al. / Applied Catalysis A: General 402 (2011) 23–30
[
[
16] P. Mahata, G. Madras, S. Natarajan, J. Phys. Chem. B 110 (2006) 13759–13768.
17] S. Mahapatra, G. Madras, T.N. Guru Row, J. Phys. Chem. 111 (2007)
505–6511.
18] M. Sokmen, A. Ozkan, J. Photochem. Photobiol. A- Chem. 147 (2002)
7–81.
19] C. Guillard, H. Lachheb, A. Houas, M. Ksibi, E. Elaloui, J.-M. Herrmann, J. Pho-
tochem. Photobiol. A- Chem. 158 (2003) 27–36.
20] G.K. Zhang, Y.J. Hu, X.M. Ding, J. Zhou, J.W. Xie, J. Solid Sate Chem. 181 (2008)
[28] (a) X.L. Yan, T. Ohno, K. Nishijima, R. Abe, B. Ohtani, Chem. Phys. Lett. 429 (2006)
606–610;
F. Amano, A. Yamakata, K. Nogami, M. Osawa, B. Ohtani, J. Am. Chem. Soc. 130
(2008) 17650–17651.
[29] B. Gözmen, M.A. Oturan, N. Oturan, O. Erbatur, Environ. Sci. Technol. 37 (2003)
3716–3723.
[30] S. Fukahori, H. Ichiura, T. Kitaoka, H. Tanaka, Appl. Catal. B 46 (2003) 453–462.
[31] Z.M. Hu, C.F. Leung, Y.K. Tsang, H.X. Du, H.J. Liang, Y.F. Qiu, T.C. Lau, New J.
Chem., B (2010), doi:10.1039/c0nj00583e.
[32] T. Ohno, T. Tsubota, M. Toyofuku, R. Inaba, Catal. Lett. 98 (2004) 255–258.
[33] J. Sun, L. Qiao, S. Sun, G. Wang, J. Hazar. Mater. 155 (2008) 312–319.
[34] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293 (2001), 296-271.
[35] H. Irie, S. Washizuka, N. Yoshino, K. Hashimoto, Chem. Commun. 11 (2003)
1298–1299.
C
6
[
[
[
[
[
7
2
133–2138.
21] T. Shishido, M. Oku, S. Okada, K. Kudou, J. Ye, T. Sasaki, Y. Watanabe, N. Toyota,
H. Horiuchi, T. Fukuda, J. Alloys Compd. 281 (1998) 196–201.
22] A. Molak, E. Talik, M. Kruczek, M. Paluch, A. Ratuszna, Z. Ujma, Mater. Sci. Eng.,
B 128 (2006) 16–24.
23] S.-Y. Lai, Y. Qiu, S. Wang, J. Catal. 237 (2006) 303–313.
24] R. Asahi, T. Morikawa, Chem. Phys. 339 (2007) 57–63.
25] M.A. Butler, J. Appl. Phys. 48 (1997) 1914–1920.
[
[
[
[
[36] J.Y. Lee, J. Park, J.H. Cho, Appl. Phys. Lett. 87 (2005) 011904.
[37] K. Nagaveni, M.S. Hegde, N. Ravishankar, G.N. Subbanna, G. Madras, Langmuir
20 (2004) 2900–2907.
26] T. Wu, G. Liu, J. Zhao, H. Hidaka, N. Serpone, J. Phys. Chem. B 102 (1998)
5
845–5851.
[38] F. Dong, W.R. Zhao, Z.B. Wu, S. Guo, J. Hazard. Mater. 162 (2009) 763–770.
[39] M. Batzill, E.H. Morales, U. Diebold, Phys. Rev. Lett. 96 (2006) 026103.
[40] R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima, A. Kitamura, M.
Shimohigoshi, T. Watanabe, Nature 388 (1997) 431–432.
[
27] A. Mills, J.S. Wang, J. Photochem. Photobiol. A 127 (1999) 123–134;
M. Mrowetz, W. Balcerski, A.J. Colussi, M.R. Hoffmann, J. Phys. Chem. B 108
(
2004) 17269–17273.