Photolysis and photocatalytic decomposition of sulfamethazine antibiotics in an aqueous solution with TiO2

Article abstract of DOI:10.1039/c6ra13435a

Photo-decomposition of sulfamethazine (SMT) involves photolytic and photo-catalytic reactions, which occur simultaneously. The relative contributions of these two reactions to the overall SMT photo-decompositions by TiO₂ and the intermediates of SMT photo-decompositions were systematically examined with the effects of TiO₂ loading, and the pH and the initial SMT concentrations in the solutions. The apparent rate constants of SMT photo-decomposition reactions, which were well described by the pseudo-first-order kinetic model, ranged from 0.24 to 1.61 h^-1. The overall photo-decomposition efficiencies of 0.072 mM SMT were the highest at pH 5.5 with 0.5 g L^-1 TiO₂ due to the adsorption-induced photocatalytic decomposition of SMT on TiO₂. However, the SMT photolysis occurred more rapidly at pH 10. Two reactive species of holes and hydroxyl radicals concurrently participated in the photocatalytic decomposition of SMT, and the latter dominated the oxidative reactions of SMT on TiO₂. Eight intermediates of SMT photo-decomposition were determined using LC-MS. Their time-dependent distributions indicated that the photo-decomposition of SMT was triggered by hydroxylation on aniline and dimethylpyrimidinyl moieties, followed by the cleavage of the S-N bond of SMT. Our results illustrated that the intermediates with dimethylpyrimidinyl groups possess strong resistance to photo-decomposition and played a determinant role in the photo-decomposition of SMT.

Full text of DOI:10.1039/c6ra13435a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Photolysis and photocatalytic decomposition of
sulfamethazine antibiotics in an aqueous solution
with TiO2†
Cite this: RSC Adv., 2016, 6, 69301
Tai-Wei Tzeng,a Shan-Li Wang,b Chiing-Chang Chen,c Chen-Chung Tan,a
a
d
a
e
f
*
*
Yu-Ting Liu, Tsan-Yao Chen, Yu-Min Tzou, C.C. Chen and J. T. Hung
Photo-decomposition of sulfamethazine (SMT) involves photolytic and photo-catalytic reactions, which
occur simultaneously. The relative contributions of these two reactions to the overall SMT photo-
decompositions by TiO2 and the intermediates of SMT photo-decompositions were systematically
examined with the effects of TiO2 loading, and the pH and the initial SMT concentrations in the
solutions. The apparent rate constants of SMT photo-decomposition reactions, which were well
described by the pseudo-first-order kinetic model, ranged from 0.24 to 1.61 hꢀ1. The overall photo-
decomposition efficiencies of 0.072 mM SMT were the highest at pH 5.5 with 0.5 g Lꢀ1 TiO2 due to the
adsorption-induced photocatalytic decomposition of SMT on TiO2. However, the SMT photolysis
occurred more rapidly at pH 10. Two reactive species of holes and hydroxyl radicals concurrently
participated in the photocatalytic decomposition of SMT, and the latter dominated the oxidative
reactions of SMT on TiO2. Eight intermediates of SMT photo-decomposition were determined using LC-
MS. Their time-dependent distributions indicated that the photo-decomposition of SMT was triggered by
hydroxylation on aniline and dimethylpyrimidinyl moieties, followed by the cleavage of the S–N bond of
SMT. Our results illustrated that the intermediates with dimethylpyrimidinyl groups possess strong
resistance to photo-decomposition and played a determinant role in the photo-decomposition of SMT.
Received 24th May 2016
Accepted 11th July 2016
DOI: 10.1039/c6ra13435a
www.rsc.org/advances
distributed in the environment because of unintentional or
intentional discharges during agricultural activities. For
1. Introduction
instance, sulfamethazine (SMT), a major sulfonamide antibiotic
with a functional group of –SO2NH2–, is commonly used to
control infectious diseases and facilitate animal growth.2,3
Because veterinary drugs cannot be completely metabolized
aer their uptake by animals, up to 90% of SMT is excreted via
feces and urine and ends up in animal manures.4,5 The SMT-
containing manures, either in fresh condition or aer being
composted, are frequently applied as a fertilizer. This unstop-
pable campaign caused by long-term agricultural activities has
led to the widespread presence of SMT in farmlands. Subse-
quently, the SMT may be relocated through surface runoff/
leaching into surface and ground water, drinking waters, or
eventually the entire ecosystems.6,7 In addition, SMT may be
accumulated in soils and sediments or be spread out through
the food chain, leading to the development of antibiotic resis-
tances.8–10 Consequently, a decrease or loss in the treatment
potency of antibiotics against human diseases may result in
a huge disaster in public health. Therefore, it is important and
necessary to eliminate SMT residues in the environment.
SMT exhibits a semi-resistance to biological degradation,
and thus, the traditional sewage treatment systems with
The presence of pharmaceuticals and personal care products
(PPCPs) has received great scientic concern due to their
potential impacts on the normal functionalities of ecosystems.1
Veterinary drugs, e.g., sulfonamide antibiotics, a typical class of
PPCPs, were commonly applied to livestock animals for thera-
peutic purposes, and these compounds have been widely
aDepartment of Soil and Environmental Sciences, National Chung Hsing University,
250 Kuo-Kuang Rd., Taichung, 40227, Taiwan, Republic Of China. E-mail: ymtzou@
dragon.nchu.edu.tw; Fax: +886-4-22855167; Tel: +886-4-22840373 ext. 4206
bDepartment of Agricultural Chemistry, National Taiwan University, Taipei, Taiwan,
Republic Of China
cDepartment of Science Education and Application, National Taichung University of
Education, Taichung, Taiwan, Republic Of China
dDepartment of Engineering and System Science, National Tsing Hua University,
Hsinchu, Taiwan, Republic Of China
eDepartment of Life Science, National Taiwan Normal University, Taipei 116, Taiwan,
ROC
fDepartment of Horticulture and Landscape Architecture, National Taitung Junior
College, Taitung 95045, Taiwan
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c6ra13435a
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 69301–69310 | 69301

View Article Online
RSC Advances
Paper
a biological process may be less effective decreasing the using TiO2 as a photocatalyst under UV-A irradiation. The
concentrations of SMT discharges to a level with low or no authors found that even if the amounts of sulfamethoxazole
environmental risk.11,12 Advanced oxidation techniques (AOPs), adsorbed on the TiO2 surfaces were very low, rapid photo-
such as electro-Fenton,13,14 sonolysis,15 and photo-Fenton oxidization of sulfamethoxazole was still observed on TiO2.
process,16 have been employed to remove the antibiotics or Jiao et al.26 reported that tetracycline antibiotic could proceed
metabolites of SMT. However, the productions of extra wastes, a self-photolytic reaction, which was highly pH-dependent and
e.g., Fe – containing sludge, substantially increase the operation controlled partially the antibiotic concentrations in the envi-
20
cost. Heterogeneous photocatalysis, one of the typical AOPs, ronments.27,28 Although Calza et al.29 and Beltran et al. had
´
using the stable and low toxic catalysts such as TiO2 is addressed the photo-decomposition of sulfonamide antibiotics
a potential cost-effective alternative because the TiO2 catalysts on TiO2, the proportions of surface adsorption and photolytic of
can provide high efficiency of converting photo- to chemical- SMT involved in the photo-induced redox reactions as inu-
energy and promote photo-decomposition of antibiotics17,18 enced by pH remained unknown.
and organic pollutants.19–22 As TiO2 particles are irradiated by
In this study, the photo-decomposition pathways of SMT with
the photons with energy greater than the band gap of the TiO2 were deduced in terms of the formation of intermediates,
semiconductor, the electrons (eꢀ) are ejected from the valence which were analyzed by electrospray ionization mass spectrometry
band (VB) to the conduction band (CB). This electronic move- in both the positive (ESI+) and negative (ESIꢀ) ion modes, and the
ment generates a photovoltage and creates a charge separation, fragment ion patterns were obtained by total ion chromatogram
leading to the productions of an electron deciency or positively (TIC). The differences in the photodecomposition products or
charged hole (h+) in the VB and an enrichment of negative intermediates of SMT identied using these two modes were
charge, eꢀ, in the CB. The redox couples of eꢀ/h+ on the surfaces examined and compared. Because a large number of small mole-
of TiO2 then participate in the redox reactions with water/ cules are easily ionized by the ESIꢀ rather than that by the ESI+,
oxygen or contaminants. The VB hole can act as a strong both modes were simultaneously applied in the study to identify
oxidizer or promote the formation of strong oxidants, e.g., OH the photo-induced products of SMT in the same system.30,31 Effects
radicals23 and the CB electrons can serve as a reductant for the of initial concentrations of SMT, TiO2 loadings, pH, and photo-
redox reactions. The productions of hole (h+) and electron (eꢀ) induced radicals on SMT decompositions were also investigated.
on TiO2 upon absorption of light energy are described in eqn Thus, the major goals of this study were to (1) clarify the contri-
(1), and the hydroxyl radical productions associated with oxygen butions of adsorption and photolytic reactions to the photo-
reductions are expressed in eqn (2) and (3), respectively.
catalytic decomposition of SMT in the presence of TiO2, (2)
characterize the stable reaction intermediates and products, and
(3) propose the dominant photo-decomposition pathways of SMT.
hn
TiO2 eꢀ þ hþ
(1)
!
TiO2(h+) + H2O/OHꢀ / TiO2(h+) ꢀ H2O/OHꢀ
TiO2 + cOH
/
2. Materials and methods
2.1 Chemicals
(2)
(3)
TiO2(eꢀ) + O2 / TiO2 + O2ꢀc
TiO2 (Degussa P25) was purchased from Sigma-Aldrich and
used as received. The crystal form contents, surface area, and
crystalline size of TiO2 Degussa P25 were 75% anatase and 25%
rutile, 50 ꢁ 1.0 m2 gꢀ1, and about 20 nm with non-porous
structures, respectively.32 The SMT antibiotic (C12H14N4O2S)
with a purity of 99% and molecular weight of 278.34 g molꢀ1
was also purchased from Sigma-Aldrich and used without
further purication. The stock solution of 100 mg Lꢀ1 SMT was
prepared by dissolving 0.1 g of SMT in 1 L de-ionized water and
stored in the dark at 4 ꢂC. Because the concentration of SMT was
subject to change with pH, especially at high pH, a NaH2BO3
buffer solution was prepared by dissolving 5.1 g H3BO3 and 2 g
NaOH in 1 L de-ionized water at a pH of 10 to maintain the
solution pH during SMT measurement.
The possible reactions of forming peroxylated intermediates
and degraded products upon decomposition of the organic
compounds (electron donors, Redorg) by OH radicals and
superoxide radicals (O2ꢀc) are described in eqn (4), and (5),
respectively.24
cOH + Redorg / Ox00
(4)
org
O2ꢀc + Redorg / peroxylated intermediates /
degraded products
(5)
On the other hand, if the organic compound exhibits a more
negative redox potential than that of hole, the organic can be
oxidized by the hole, giving a cation radical of the specic
organic (eqn (6)).
2.2 Photocatalytic reactor
TiO2(h+) + S / TiO2 + Sc+ / product
(6)
All experiments, including adsorption, photolysis, and photo-
catalysis, were performed in a chamber with eight UV lamps (20
watts for each lamp) (GL20, Sankyo Denki, Japan) with an UV-A
intensity of 20.8 mW cmꢀ2 detected by a OPAS UV-A spotter (SN:
811013T, USA). A 500 mL quartz double-wall water jacked
reactor, which was connected to a circulating waterbath to
The photocatalytic decompositions of organic pollutants
such as sulfa-containing drugs, including kinetics and mecha-
nisms, had been previously investigated on TiO2. For instance,
Hu et al.25 studied the photo-oxidation of sulfamethoxazole
69302 | RSC Adv., 2016, 6, 69301–69310
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
maintain a constant temperature in the reactor, was placed in using a UV-visible spectrophotometer (Spectro UV-2550, Lab-
the center of the chamber surrounded by the lamps. The lamps omed, USA) at a wavelength of 262 nm.
were switched on for at least 5–10 min to obtain a stable
2.4.2 Dissolved organic carbon and anions. The extent of
emission prior to conducting the experiments of photo- photo-decomposition of SMT was examined by measuring the
reactions. One end of a plastic tube was inserted into the dissolved organic carbon (DOC) using a TOC analyzer (Analy-
reactor, and the sample solution was extracted using a syringe ticjena multi N/C 2100, Germany). The inorganic anions
connected to the other end of the tube. Before each sample contents in the solution, i.e., NO2ꢀ, NO3ꢀ, and SO42ꢀ, were
collection, the tube was pumped and purged several time to quantied by an ion chromatography (883 basic IC plus, Met-
ensure the sample was not contaminated by the residual rohm; Swissmade). An anion exchange column (Metrosep A
suspensions in the tube.
Supp5) was used with 3 mM Na2CO3 and 1 mM NaHCO3 as
+
mobile phase at a ow rate of 0.7 mL minꢀ1
.
The NH4
36,37
concentration was measured using a cation exchange column
(Metrosep C 4) with an eluent containing 0.7 mM dipicolinic
2.3 Experimental procedures
2.3.1 Adsorption. The adsorption of SMT on TiO2 was carried acid and 1.7 mM nitric acid at a ow rate of 0.9 mL minꢀ1
.
out using a batch method in the dark. The initial SMT concen-
2.4.3 Intermediates of SMT photo-decomposition. The full-
trations varying from 0.036 to 0.162 mM were freshly prepared scan UV spectra of a high performance liquid chromatography
from a 100 mg Lꢀ1 stock solution. The solution of each SMT (Waters 1525 Binary HPLC) equipped with a photodiode array
concentration was then added into a 500 mL quartz beaker con- detector (Model code 2998, USA) were used to determine the
taining TiO2 with a dosage of 0.5, 1.0, or 2.0 g Lꢀ1, respectively. The intermediates of SMT photo-decomposition. A C18 reverse
pH values of the suspensions were adjusted to 3, 5.5, and 10 using phase column (Atlantis, Waters) was applied to separate the
1.0 M HCl or NaOH and the suspensions were stirred continuously intermediates in 100 mL ltrate using a gradient method with
for 24 h. Right aer the sample was extracted and ltrated, SMT in two mobile phases at a ow rate of 1.0 mL minꢀ1 38 Briey, the
.
the ltrate was measured colorimetrically (details please see percentage of mobile phase A (de-ionized water) was main-
below). The differences in SMT concentrations before and aer the tained at 95% for the rst 30 min, followed by increasing line-
reactions were attributed to adsorption.
arly the percentage of mobile phase B (acetonitrile) to 100%
2.3.2 Photo-decomposition of SMT. Occurrence of the self- within 20 min and to 5% for the next 10 min. The intermediates
photolysis of SMT was rst evaluated in the absence of TiO2 were identied by the mass spectrometry (Waters 3100, USA)
under UV irradiation at pH 3–10. The photo-decomposition of with an electrospray interface (ESI) which was operated in both
SMT was started aer 0.036, 0.072, or 0.162 mM SMT was positive and negative ionization modes. The full-UV absorbing
reacted with 0.1–2.0 g Lꢀ1 TiO2 at pH 3–10 in the dark for 1 h. spectra of the intermediates were detected by the photodiode
The overall photo-decomposition of SMT in the presence of array detector which was connected behind ESI.
TiO2 involved both photo-catalytic decomposition of SMT
(PCDS) on TiO2 and photolysis of SMT. Thus, the contributions
2.5 Kinetics of SMT decomposition
of PCDS to SMT decomposition could be approximately esti-
mated by subtracting the photolysis of SMT from the overall
SMT photo-decomposition. During the photo-decomposition
processes of SMT, the full-UV scanning spectra, including the
decomposition intermediates of SMT, were obtained by a HPLC
coupled with mass spectra, and the dissolved carbon contents
were determined by a total organic carbon (TOC) analyzer
(details please see below). In addition, the contributions of
hydroxyl radicals (cOH) and hole (h+) to photo-catalytic
decomposition of SMT were also examined by adding two
radical scavengers of isopropanol (ISP) and sodium iodide (NaI)
to capture hydroxyl radicals (cOH) and hole (h+), respectively.24
The Langmuir–Hinshelwood (L–H) kinetic model was applied
to analyze the photo-decomposition of SMT (eqn (7)):39
ꢀ
ꢁ
dC
dt
krKadC0
r0 ¼
ꢀ
¼
(7)
1 þ KadC0
0
where, C0 (mM) and Kad (L mmolꢀ1) represent the initial
concentration of SMT and the reactant adsorption constant of
SMT on TiO2 surfaces, respectively. The kr (mM hꢀ1) is the
intrinsic reaction rate constant related to the catalyst mass and
efficient photon ow.40
When the SMT concentration is very low (i.e. KadC0 ꢃ 1), The
L–H kinetic model can be simplies to a pseudo-rst-order
(PFO) model (eqn (8)).21,25
2.4 Analytical methods
ꢀ
ꢁ
C
ꢀln
¼ kapp ꢄ t
(8)
2.4.1 SMT concentrations. The SMT-containing ltrates
obtained aer the samples were periodically extracted using
C0
a syringe and passed through a 0.2 mm pore-sized cellulose where, C0 (mM) and C (mM) are the initial concentration and
acetate membrane lter (Advantec). Two millimeters of the concentration of SMT at reaction time t, respectively. The
NaH2BO3 buffer solution (pH ¼ 10) were added into a ltrate of kapp (hꢀ1) value, the slope of PFO linear curve represents an
8 mL prior to the determination of SMT concentrations to avoid apparent PFO reaction rate constant.
the pH-dependent changes in SMT species during spectroscopic
The contributions of PCDS on TiO2 to the overall SMT photo-
analyses.33–35 The changes in SMT concentrations during the decomposition were calculated by deducting the amount of
adsorption and photo-decomposition processes were detected SMT self-photolysis from the overall photo-decomposition of
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 69301–69310 | 69303

View Article Online
RSC Advances
Paper
SMT. Thus, the PFO rate constant of PCDS on TiO2 could be solution.44 Therefore, the enhancement of SMT adsorption due
obtained (eqn (9)).
to the changes in SMT species induced by the charged surfaces
of TiO2 was relatively insignicant.
ꢀ
ꢁ
C0 ꢀ ðCUV ꢀ Coverall
Þ
ꢀln
¼ kapp;TiO ꢄ t
(9)
2
C0
3.2 Photolysis and photocatalytic degradation of SMT
where, CUV and Coverall are the concentrations of photolytic and
overall photo-decomposition reactions of SMT at time t,
The occurrences of photolysis of SMT were rst estimated under
UV irradiation in the absence of TiO2. The changes of SMT
concentrations upon photolysis of 0.072 mM SMT at three pH
values, presented in Fig. 2a, showed that photolysis of SMT was
pH-dependent and the maximum photolytic reactions of SMT
occurred at pH 10 with a PFO rate constant of 0.58 hꢀ1 (Fig. 2a,
inserted and Table 1). About 61, 22, and 15% of SMT
respectively. The kapp,TiO (hꢀ1) represents the PFO reaction rate
2
constant of PCDS on TiO2.
3. Results and discussion
3.1 Adsorption of SMT on TiO2
The SMT adsorption on TiO2 was carried out for 24 h in the
absence of light at pH 3, 5.5, and 10. The results showed that
SMT adsorption was low at each tested pH; the highest
adsorption amount of SMT (i.e., 0.54 mmol kgꢀ1) occurred at
pH 5.5 (Fig. 1), attributed to a favorable electrostatic interaction
between SMT and TiO2 at the specic pH. Depending on the
solution pH, SMT can be protonated and deprotonated, forming
charged species of SMT+ and SMTꢀ, respectively, or be con-
verted to form the neutral species of SMT0 (Fig. 1). On the other
hand, the TiO2 adsorbent has a point of zero charge at 6.5,41 and
thus, the surfaces of TiO2 exhibit a pH-dependent charge
+
properties of SMT, forming TiOH2 , TiOH, or TiOꢀ.42 At pH 3
and 10, both SMT and TiO2 bore the same charges, leading to an
unfavorable (repulsive) condition for SMT adsorption. At pH
5.5, the neutral form of SMT0 dominated in the solution, and
thus, the electrostatic repulsion between neutral SMT and
positively charged TiO2 was lower. In addition, the conversion
of SMT0 to relative stable zwitterions (SMTꢁ) in the presence of
positively charged TiO2 (ref. 43) may facilitate SMT adsorption
on TiO2 through the polarized –SO2N– group of SMTꢁ. However,
the SMTꢁ comprises less than 0.2% of SMT0 in an aqueous
Fig. 2 The influences of the solution pHs on an initial concentration
Fig. 1 The pH-dependent speciation of sulfamethazine (SMT) and the 0.072 mM sulfamethazine (a) photolysis and (b) photo-decomposition.
adsorption amount of SMT. Adsorption experiment was conducted in The self-photolytic reaction of sulfamethazine (SMT) in the absence of
a system with an initial SMT concentration of 0.072 mM and 0.5 g Lꢀ1 TiO2. The inset was the simulation results of photolytic and photo-
TiO2.
decomposition on SMT using the PFO model.
69304 | RSC Adv., 2016, 6, 69301–69310
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
0.47, and 0.27 hꢀ1 at pH 5.5, 10, and 3.0, respectively (Table 1).
At pH 10, even if the photolysis of SMT proceeded rapidly with
a kapp of 0.58 hꢀ1, no discernible PCDS were observed (Fig. S1†
and Table 1). Thus, the overall photo-decomposition rate of
SMT was lower at pH 10 than that at pH 5.5 (Table 1) at which
TiO2 exhibited a higher adsorption capacity and photo-catalytic
rate (Fig. S1 and S2†).
Table 1 Pseudo-first-order (PFO) rate constants of sulfamethazine
(SMT) self-photolytic reactions, photo-catalytic decomposition of SMT
(PCDS) on 0.5 g Lꢀ1 TiO2, and the overall photo-decomposition of SMT
at pH 3, 5.5, and 10. The initial SMT concentration was 0.072 mM
Pseudo-rst-order parameters
Overall photo-
Self-photolysis
PCDS on TiO2
decomposition
In the TiO2 system, the SMT photo-decomposition involved
two reactions of SMT self-photolysis and PCDS (i.e., photo-
catalytic decomposition of SMT) on the TiO2 surfaces. While
the contributions of photolytic decomposition of SMT to the
overall SMT photo-decomposition was deducted, the PCDS on
the TiO2 surfaces could be obtained (Fig. S1†). Although
photolytic decomposition of SMT exhibited the highest effi-
ciency at pH 10 (Fig. 2a and S2†), the contributions of PCDS on
TiO2 to the overall SMT photo-decomposition was the lowest
kapp (hꢀ1
)
r2
kapp,TiO (hꢀ1
)
r2
kapp (hꢀ1
)
r2
a
pH
2
3
5.5
10
0.06
0.12
0.58
0.98
0.99
0.98
0.21
0.41
0.025
0.99
0.98
0.83
0.27
0.69
0.47
0.96
0.99
0.99
a
kapp represented the rate constant of pseudo-rst-order (PFO) model.
disappeared at pH of 10, 5.5, and 3, respectively, aer 2 h (Fig. S2†). At pH 10, the low efficiency of PCDS (Fig. S2†) was
reactions under illumination. The results may be attributed to probably attributed to the low adsorption of SMT on TiO2
the differences in the lightꢀsensitivity of the three SMT species (Fig. 1) due to electrostatic repulsion (negative) between these
(i.e., SMT+, SMT0, and SMT ). Baeza and Knappe45 reported that the adsorbent and adsorbate. In contrast, neutral SMT exhibi-
the anionic form of SMT exhibited a higher molar absorption ted a higher affinity to the surfaces of TiO2 at pH 5.5, and thus,
coefficient than the cationic and neutral forms of SMT. There- a signicant amount of SMT adsorption was observed (Fig. S2†).
fore, we presumed that, at pH 10, the dominant anionic form of The reaction rate (kapp,TiO ) of PCDS on TiO2 also exhibited
2
SMT may absorb more photons and lead to a greater photolysis a maximum value at pH 5.5 (Table 1). According to the reports of
of SMT. In addition, Challis et al.46 found that the photo- ref. 48–50, the heterogeneous photocatalytic reaction on the
chemical activity of sulfonamide antibiotics was related to surfaces of TiO2 involved ve steps: (1) generation of the elec-
their pKa values, determined mainly by the numbers of CH3 tron–hole pairs on TiO2 by absorbing light energy higher than
2
groups on the heterocyclic ring of sulfonamide antibiotic. While 3.2 eV; (2) separation of the photo-generated electrons and
being attached, CH3 group could offset the electron withdrawal holes; (3) productions of OH radicals or superoxide radicals
by the two N atoms in the heterocyclic substituents of sulfon- when the holes and electrons migrated to the surfaces of TiO2
amide antibiotics, resulting in a decrease in the proton- and interacted with the adsorbed H2O/hydroxyl groups and
donating ability of NH (or NH2) and pKa value. These authors dissolved oxygen, respectively (eqn (2) and (3)); (4) occurrences
2
further demonstrated that the sulfonamide with a lower pKa
of oxidative reactions in a solution with electron donors by OH
2
exhibited a greater photolytic degradation efficiency of sulfon- radicals (eqn (4)), superoxide radicals (O2ꢀc) (eqn (5)), or photo-
amide at high pH.46 Because the SMT, bearing two CH3 groups generated holes of TiO2 (eqn (6)); and (5) desorption of the
on the heterocyclic ring, has a lower pKa as compared to that of reaction products and reconstruction of the TiO2 surfaces. The
2
sulfonamide, the SMT was expected more susceptive to the higher efficiency for catalytic decomposition of SMT on TiO2 at
photo-energy. Thus, a rapid photolysis of SMT occurred, pH 5.5 could be attributed to the adsorption of SMT on TiO2,
particularly at a higher pH, at which SMTꢀ was the dominant leading to a decrease in the recombination rate and an increase
species of SMT (Fig. 2a). A similar result reported in ref. 47 in the quantum efficiency of TiO2 photocatalytic decomposition
showed that the SMTꢀ exhibited a higher degradation efficiency through the pathway of eqn (6). That is, in the SMT/TiO2 system,
upon exposure to a simulated solar radiation. Because SMT can the adsorption of SMT on TiO2 may be a key factor controlling
proceed a self-photolytic reaction, to evaluate precisely the role the efficiency of electron–hole separation and subsequent SMT
of TiO2 for heterogeneous catalysis of SMT decomposition, photo-decomposition. To further conrm this, the effects of the
photolytic reactions of SMT must be excluded from the overall TiO2 doses on SMT photo-decomposition were evaluated at pH
photo-decomposition of SMT.
5.5 with a xed SMT concentration of 0.072 mM, and the results
Inuences of the solution pH (3, 5.5, and 10), SMT initial were shown in Fig. S3.† We did nd that provisions of sufficient
concentrations (0.036, 0.072, and 0.162 mM), and TiO2 loadings adsorptive sites of TiO2 indeed enhanced the amounts and the
(0.1, 0.5, and 2.0 g Lꢀ1) on photo-decomposition of SMT were reactive rates of SMT photo-decomposition (Fig. S3† and
investigated. In the system with TiO2, approximately 76% inserted). For instance, the PFO rate constant (kapp) of photo-
removal efficiency of SMT was observed aer the system was decomposition of 0.072 mM SMT increased from 0.52 to 1.17
exposed to the light for 2 h at a pH of 5.5. About 67% of SMT was hꢀ1 with increasing in TiO2 doses from 0.1 to 2.0 g Lꢀ1 (Table 2).
photo-decomposed at pH 10 and the decomposition efficiency In addition, the mineralization efficiency of SMT was also
of SMT was the lowest at pH 3 (49%) aer the same reaction higher with a greater suspension density of TiO2, i.e., 2.0 g Lꢀ1
time, i.e., 2 h (Fig. 2b). Indifferent to the reaction pH, the overall (Fig. S4a†).
SMT photo-decomposition followed the PFO model, and the
Effects of initial SMT concentrations on photo-
apparent PFO reaction rate constant (kapp) values were 0.69, decomposition of SMT in the presence of 0.5 g Lꢀ1 TiO2 at pH
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 69301–69310 | 69305

View Article Online
RSC Advances
Paper
Table 2 Pseudo-first-order rate (PFO) rate constants of sulfamethazine (SMT) photo-decomposition in the presence of radical scavengers of
isopropanol (ISP) and sodium iodide (NaI) with 0.036–0.162 mM SMT and 0.1–2.0 g Lꢀ1 TiO2 at pH 5.5
Scavengers molar ratio
(mol molꢀ1
ISPa/SMT
)
PFO rate constant
SMT concentration (mM)
TiO2 loading (g Lꢀ1
)
NaI/SMT
kappb (hꢀ1
)
r2
Initial concentration effect
0.036
0.072
0.162
0.5
0.5
0.5
0
0
0
0
0
0
1.61
0.69
0.24
0.99
0.99
0.99
TiO2 loadings effect
0.072
0.072
0.072
0.1
0.5
2.0
0
0
0
0
0
0
0.52
0.69
1.17
0.98
0.99
0.99
Radical scavengers effect
0.072
0.072
0.072
0.5
0.5
0.5
10
100
0
0
0
100
0.32
0.10
0.36
0.99
0.98
0.94
a
ISP indicated the isopropanol. b kapp represented the rate constant of pseudo-rst-order (PFO) model.
5.5 were shown in Fig. S4b.† Although the photo-decomposition decomposition decreased greatly and closed to that of SMT
rate of SMT decreased with an increase in the initial concen- self-photolysis (Fig. S5†). That is, a complete inhibition of the
trations of SMT (Fig. S4b,† inserted and Table 2), the amount of catalytic oxidation of SMT on TiO2 was observed in a system
photo-induced SMT removal was still higher when a higher with an ISP/SMT molar ratio of 100, and thus, the rate constant
concentration of SMT was added. The photo-decomposition of SMT photolysis (0.12 hꢀ1) was closed to that of TiO2,ISP/
kinetics of SMT could also be well described by PFO model,
system (0.10 hꢀ1) (Table 2). Because ISP is a major OH
SMT¼100
and the kapp values of the overall photo-decomposition of SMT radical quencher, the strong inhibition of SMT photo-
were 1.61, 0.69, and 0.24 hꢀ1 in a system with the initial SMT decomposition with the addition of ISP suggested that OH
concentrations of 0.036, 0.072, and 0.162 mM, respectively radicals may be a dominate oxidant for SMT when TiO2 was
(Table 2). Because, the heterogeneous PCDS on TiO2 dominated present.
SMT photo-decomposition and related greatly on the amount of
According to eqn (2), the holes on the TiO2 were the activated
SMT adsorption at pH 5.5, the lack of sufficient adsorption sites sites of producing OH radicals. Upon OH radical productions,
on TiO2 may inhibit SMT adsorption and subsequent they could be adsorbed on TiO2 or be present in solution.24
decomposition.51
Although SMT could be photo-oxidized directly by the holes, the
To distinguish the contributions of major reactive species of hydrophilic properties of TiO2 surfaces may favor H2O photol-
holes or OH radicals to SMT photo-decomposition, sodium ysis on the electron-deciency sites (i.e., holes), leading to OH
iodide (NaI) and isopropanol (ISP) were used as the radical radical productions. Thus, we presumed that the adsorbed SMT
scavengers of holes and OH radicals, respectively, during the may be more readily oxidized by the vicinity of adsorbed OH
photo-catalytic reaction.52,53 As shown in Fig. S5,† in the absence radicals than that by free OH radicals with a short lifetime.
of any radical scavengers (represented by normal system), Therefore, adsorption of neutral SMT on the surfaces of TiO2
a highest photo-decomposition of SMT was observed with a kapp facilitates the photo-catalytic decomposition of SMT.
value of 0.69 hꢀ1. Aer adding the ISP reagent, the PFO rate
constant (kapp) of SMT photo-decomposition decreased from
0.69 to 0.32 hꢀ1 with a ISP/SMT molar ratio of 10, and the kapp
3.3 Intermediates productions during photo-decomposition
of SMT
value decreased further to 0.10 hꢀ1 when the ISP/SMT molar
ratio was increased to 100 (Table 2). Similarly, the kapp value The intermediates (Int.) of the photo-decomposition of SMT were
decreased from 0.69 to 0.30 hꢀ1 with the addition of NaI (i.e., analyzed by HPLC-MS, and the ESI+ and ESIꢀ fragmentation
NaI/SMT molar ratio of 100). The decrease in the efficiency of patterns of intermediates were obtained from the full-scan UV
SMT photo-decomposition due to the presence of radical scav- spectra (Fig. S6†). The pathways of SMT photo-decomposition
enger indicated that the productions of holes and OH radicals developed at each ESI mode were proposed in Fig. 3. Five inter-
on the TiO2 surfaces upon absorbing light energy were mediates were identied based on the fragmentation patterns of
responsible for SMT removal (Fig. S5†). However, the ISP may be ESI+ which were obtained from the major peaks of chromato-
a
more efficient reagent of inhibiting SMT photo- graph spectra (Fig. S6†). The SMT parent compound was depicted
decomposition than NaI because, upon the addition of ISP at a retention time of 15.565 min with a fragment ion at m/z 279
with a ISP/SMT ratio of 100, the rates of SMT photo- (Fig. S6b†). Two peaks with the retention time of 14.667 and
69306 | RSC Adv., 2016, 6, 69301–69310
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 3 The possible photo-decomposition pathways of sulfamethazine (SMT) obtained by using (a) ESI+ and (b) ESIꢀ detection modes.
14.221 min exhibited the same fragment ions at m/z 295, repre- 311), and Int. 8 (m/z 140), were detected and their concentra-
senting two intermediates of the hydroxylation analogs of SMT tions increased rapidly during the initial 30 min irradiation
(i.e., Int. 1 and Int. 3 in Fig. S6c and d†). The fragment ion of (Fig. 4b). The productions of maximum concentrations of the
intermediate 5 (m/z 311) (Fig. S6e†) depicted at the retention time intermediates followed the order of Int. 1 > Int. 8 > Int. 4 > Int. 3
of 13.483 min was attributed to the attachment of hydroxide > Int. 5 (Fig. 4b). Aer photo-decomposition of SMT for 1 h, the
radical on the benzoic ring and N-heterocyclic moiety of the Int. 1 Int. 1, Int. 3, and Int. 4 decreased signicantly, but at the same
and Int. 3, respectively (Fig. 3a). A fragment ion at m/z 124, i.e., reaction time the Int. 5 and Int. 8 reached their maximum
intermediate 4, derived from the cleavage of the S–N bond of accumulation concentrations.
intermediate 3, was obtained from the peak with a retention time
Similar to the fragmentation patterns of ESI+, four same
of 12.530 min (Fig. S6f†). The intermediate 8, obtained from the intermediates of Int. 1, 3, 4, and 5, with the fragment ions at
retention time of 4.404 min, possessed a fragment ion at m/z 140 m/z 293 (Int. 1 and 3), 122, and 309, respectively, were also
and was the oxidative product of intermediate 5 or intermediated found using ESIꢀ detection mode (Table S1†). The fragment
1 upon interacting with OH radical (Fig. 3a). The intermediate 8 ion with an m/z ratio of 277 found on ESIꢀ represented the
might be also from the reactions of hydroxylation on the N- parent compound of SMT (Fig. S6h†). The fragment ions with
heterocyclic moiety of intermediate 4 aer the S–N bond of the m/z ratios of 156 (Int. 2), 172 (Int. 6), and 108 (Int. 7) were
intermediate 3 was broken (Fig. 3a).
three new intermediates identied by ESIꢀ upon photo-
The time-dependent distributions of the intermediates decomposition of SMT. The fragment ions at m/z 156 (Int. 2)
produced from the photo-decomposition of SMT using ESI+ (Fig. S6i†) was suggested an aniline with a SO2 group derived
mode were shown in Fig. 4. The fragment ion with a m/z ratio of from the cleavage of the S–N bond of Int. 1 (Fig. 3b). The
279 (ESI+), representing the SMT parent compound, decreased intermediate 6 was a fragment ion at m/z 172 (Fig. S6j†), given
rapidly over the irradiation time (Fig. 4a). Five intermediates, from the hydroxylation of the intermediate 2 (m/z 156) or from
i.e., Int. 1 (m/z 295), Int. 3 (m/z 295), Int. 4 (m/z 124), Int. 5 (m/z the cleavages of S–N bond of Int. 3 or Int. 5 (Fig. 3b). The
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 69301–69310 | 69307

View Article Online
RSC Advances
Paper
fragment ion at m/z 108 (Int. 7) (Fig. S6k†) was attributed to
the decomposition of Int. 2 and released subsequently SO/
2ꢀ
SO4
(Fig. 3b). Accordingly, the results of mass analyses
indicates that the hydroxylation of electrophilic OH radicals
on amino-phenol or pyrimidinyl moiety54,55 played an impor-
tant role of proceeding subsequently S–N bond and ring
cleavage during photo-decomposition of SMT involving
TiO2.56 The mass instrument with ESI+ source was more
sensitive to the positively charged groups, and thus, the
intermediates with pyrimidinyl group, such as Int. 4 (m/z ¼
124) and 8 (m/z ¼ 140), were preferentially detected upon
photo-decomposition of SMT (Fig. S6†). Although, both
positive and negative sources of ESI are suitable detection
modes to identify the products of SMT photo-decomposition,
the ESIꢀ source possessed more sensitive response for
detecting the acidic products with a sulfate group, e.g., Int. 2,
and 6. A similar result was also reported by ref. 31, when
analyzed pesticides with carboxyl groups, sulfate, or sulfonate
using ESIꢀ.
Fig. 5 Productions of the specific ions upon photo-decomposition of
0.072 mM sulfamethazine (SMT) on 0.5 g Lꢀ1 TiO2.
Upon photo-decomposition of 0.072 mM SMT, about 0.025
2ꢀ
+
and 0.030 mM of SO4 and NH4 , respectively, were detected
ꢀ
aer 2 h reaction (Fig. 5). However, less than 0.005 mM NO3
were detected throughout the photo-decomposition of SMT,
and the concentrations of nitrite were lower than the detection
limit. As observed previously using a TOC analyzer, about 23.1%
of 0.072 mM SMT would be completely mineralized in a system
with 0.5 g Lꢀ1 TiO2 (Fig. S4a†), and thus, the release of S- and N-
containing products, such as SO42ꢀ, NO3ꢀ, and NH4+, were ex-
pected aer the SMT was completely mineralized, releasing
concurrently the oxidative products of CO2 and H2O.54,57
3.4 Proposed the overall pathway of SMT photo-
decomposition
A comprehensive pathway of decomposition of SMT was shown
in Fig. 6. Indifferent to the detection modes of using ESI+ or
ESIꢀ, the occurrences of hydroxylation on the dimethylpyr-
imidinyl moiety and aniline portion of SMT, i.e., Int. 1 and 3,
respectively, were observed (Fig. 6). Meanwhile, the OH radicals
were attached on the vacant hydroxylation positions of Int. 1 or
3, leading to the productions of Int. 5 (Fig. 6). Formations of the
Int. 2, 4, 6 and 8 indicated that the cleavages of S–N bond may
precede the destruction of the aromatic ring of SMT upon OH
radical attack. According to the chromatographic spectra of
Fig. S6a,† the intermediates 2, 6, and 7 appeared at about 30
min reaction; however, a signicant amount of sulfuric acid was
detected aer 1 h reaction (Fig. 5). The results demonstrated
that the cleavage of S–N bond of SMT occurred before the
oxidative released of sulfate ions. Although Yang et al.51 and
Yang et al.58 report that sulfate ions could absorb photons,
leading to a shortage of photons to activated TiO2 for photo-
decomposing pharmaceuticals, the inhibition of SMT decom-
position was undiscernible in the study with an accumulation of
2ꢀ
SO4 concentrations up to 0.025 mM. Thus, the kinetic reac-
Fig. 4 Time-dependent distributions of (a) 0.072 mM sulfamethazine
(SMT) on 0.5 g Lꢀ1 TiO2 and (b) the major photo-decomposition
products of SMT.
tion of SMT photo-decomposition on TiO2 was not affected by
the reactive products.
69308 | RSC Adv., 2016, 6, 69301–69310
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 6 Proposed the overall pathways of photo-decomposition of sulfamethazine (SMT).
As mentioned previously, the Int. 5 was the hydroxylation increased, promoting the photo-catalytic decomposition of SMT by
products of Int. 1 and 3 on the aniline, and dimethylpyrimidinyl OH radicals. The photo-decomposition efficiency of SMT increased
moiety (Int. 4) may be a precursor of Int. 8 because the Int. 4 with increasing TiO2 loading and SMT concentration. During the
appeared before Int. 8 (Fig. 4). The Int. 8 with the dime- photo-decomposition of SMT, the results of DOC analyses indi-
thylpyrimidinyl moiety seemed a more persistent portion of cated that SMT was not completely converted to CO2, and the
SMT against photo-decompositions because of its rapid accu- presence of the C-containing intermediates were identied by
mulation and slow disappearance (Fig. 4). Since the positive ESI mass spectra using ESI+ and ESIꢀ modes. The occurrences of
mode could not observe fully the intermediates of photo- hydroxylations on the benzoic ring and the pyrimidine ring of SMT
decomposition of SMT, the negative ESI mode was a compen- associated with a direct cleavage the S–N bond on sulfonamide
satory strategy of identifying the aniline portion of SMT. Thus, group were suggested to be the major pathways of SMT photo-
the identications of Int. 2, 6, and 7 using the negative ESI decomposition. Accompanied with SMT photo-decomposition,
mode (Fig. S6†) could help the integral descriptions of the a signicant increase of inorganic ions,ꢀ such as SO42ꢀ, was
pathways of SMT photo-decomposition. Here, we presumed found; however, only trace amount of NO3 and NH4+ was detec-
that the Int. 2, formed by the cleavage of S–N bond of Int. 1, ted. This indicates that N-containing intermediates were more
could be the precursor of Int. 6 and 7 (Fig. 6).
persistent to the photo-decomposition than S-containing products.
Although, based on the changes in the SMT concentrations,
SMT could be completely removed in the presence of 0.5 g Lꢀ1
TiO2 at pH 5.5 under illumination, a complete mineralization of
SMT did not occur because approximately 77% of DOC still
existed aer 2 h reaction (Fig. S4a†). The DOC should be the
open-ring products of dimethylpyrimidinyl and aniline, which
were not capable of being detected in our current system.
Acknowledgements
The work was nancially supported by the Ministry of Science
and Technology, ROC under project no. of NSC 101-2621-M-005-
005, NSC 102-2621-M-005-001, MOST 103-2621-M-005-002, and
MOST 104-2313-B-005-014-MY3, and, in part, by the Ministry of
Education, ROC under the Aim for Top University (ATU) plan.
4. Conclusion
This study examined the kinetics of photo-decomposition of SMT
in the absence and presence of TiO2 at pH 3, 5.5 and 10, and the
detail photo-decomposition pathways of SMT were deduced at pH
5.5. Although photolytic reaction dominated SMT decomposition
at pH 10, the overall photo-decomposition of SMT was enhanced at
pH 5.5 because the SMT adsorption on TiO2 surfaces was greatly
References
1 J. Wang, T. Zhou, J. Mao and X. Wu, J. Environ. Chem. Eng.,
2015, 3, 2393–2400.
2 C. C. Su, L. M. Bellotindos, A. T. Chang and M. C. Lu, J.
Taiwan Inst. Chem. Eng., 2013, 44, 310–316.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 69301–69310 | 69309

View Article Online
RSC Advances
Paper
3 Q. Yang, G. Chen, J. Zhang and H. Li, RSC Adv., 2015, 5, 30 Z. R. Wu, W. Q. Gao, M. A. Phelps, D. Wu, D. D. Miller and
25541–25549. J. T. Dalton, Anal. Chem., 2004, 76, 839–847.
4 L. J. Zhou, G. G. Ying, S. Liu, R. Q. Zhang, H. J. Lai, Z. F. Chen 31 E. M. Thurman, I. Ferrer and D. Barcelo, Anal. Chem., 2001,
and C. G. Pan, Sci. Total Environ., 2013, 444, 183–195. 73, 5441–5449.
5 K.-R. Kim, G. Owens, S.-I. Kwon, K.-H. So, D.-B. Lee and 32 B. Abramovic, S. Kler, D. Sojic, M. Lausevic, T. Radovic and
Y. Ok, Water, Air, Soil Pollut., 2011, 214, 163–174. D. Vione, J. Hazard. Mater., 2011, 198, 123–132.
6 H. Dolliver and S. Gupta, J. Environ. Qual., 2008, 37, 1227– 33 L. L. Ji, W. Chen, S. R. Zheng, Z. Y. Xu and D. Q. Zhu,
1237. Langmuir, 2009, 25, 11608–11613.
7 J. G. Davis, C. C. Truman, S. C. Kim, J. C. Ascough and 34 C. Wang, H. Li, S. H. Liao, D. Zhang, M. Wu, B. Pan and
´
ˇ
´
´
ˇ
´
´
K. Carlson, J. Environ. Qual., 2006, 35, 2250–2260.
8 T. Heberer, Toxicol. Lett., 2002, 131, 5–17.
9 A. L. Batt, D. D. Snow and D. S. Aga, Chemosphere, 2006, 64,
1963–1971.
B. S. Xing, Environ. Pollut., 2014, 194, 203–209.
35 M. X. Xie, W. Chen, Z. Y. Xu, S. R. Zheng and D. Q. Zhu,
Environ. Pollut., 2014, 186, 187–194.
36 G. R. Deshpande, B. Rao and N. Someswararao, Rasayan J.
Chem., 2009, 2, 101–107.
´
10 W. Baran, E. Adamek, J. Ziemianska and A. Sobczak, J.
Hazard. Mater., 2011, 196, 1–15.
37 S. G. Shin, G. Han, J. Lim, C. Lee and S. Hwang, Water Res.,
2010, 44, 4838–4849.
´
´
11 M. N. Abellan, B. Bayarri, J. Gimenez and J. Costa, Appl.
Catal., B, 2007, 74, 233–241.
38 X. Huang, Y. Feng, C. Hu, X. Xiao, D. Yu and X. Zou,
Chemosphere, 2015, 138, 183–189.
´
12 O. Gonzalez, C. Sans and S. Esplugas, J. Hazard. Mater., 2007,
146, 459–464.
39 M. R. Hoffmann, S. T. Martin, W. Y. Choi and
D. W. Bahnemann, Chem. Rev., 1995, 95, 69–96.
40 Y. He, N. B. Sutton, H. H. H. Rijnaarts and
A. A. M. Langenhoff, Appl. Catal., B, 2016, 182, 132–141.
13 D. Mansour, F. Fourcade, I. Soutrel, D. Hauchard,
N. Bellakhal and A. Amrane, C. R. Chim., 2015, 18, 39–44.
14 D. Mansour, F. Fourcade, I. Soutrel, D. Hauchard,
N. Bellakhal and A. Amrane, J. Taiwan Inst. Chem. Eng., 41 D. Friedmann, C. Mendive and D. Bahnemann, Appl. Catal.,
2015, 53, 58–67. B, 2010, 99, 398–406.
15 Z. M. Shaykhi and A. A. L. Zinatizadeh, J. Taiwan Inst. Chem. 42 N. Jaffrezicrenault, P. Pichat, A. Foissy and R. Mercier, J.
Eng., 2014, 45, 1717–1726. Phys. Chem., 1986, 90, 2733–2738.
16 M. Perez-Moya, M. Graells, G. Castells, J. Amigo, E. Ortega, 43 M. Teixido, J. J. Pignatello, J. L. Beltran, M. Granados and
G. Buhigas, L. M. Perez and H. D. Mansilla, Water Res.,
2010, 44, 2533–2540.
17 S. S. Boxi and S. Paria, RSC Adv, 2015, 5, 37657–37668.
18 P. Wang, T. Zhou, R. Wang and T.-T. Lim, Water Res., 2011,
45, 5015–5026.
J. Peccia, Environ. Sci. Technol., 2011, 45, 10020–10027.
44 H. Sakurai and T. Ishimitsu, Talanta, 1980, 27, 293–298.
45 C. Baeza and D. R. U. Knappe, Water Res., 2011, 45, 4531–
4543.
46 J. K. Challis, J. C. Carlson, K. J. Friesen, M. L. Hanson and
C. S. Wong, J. Photochem. Photobiol., A, 2013, 262, 14–21.
47 H. Xiao, G. Liu, Z. Chen and R. Wu, Journal, 2015, 1, 63–67.
19 Z. Lu, Y. Luo, M. He, P. Huo, T. Chen, W. Shi, Y. Yan, J. Pan,
Z. Ma and S. Yang, RSC Adv, 2013, 3, 18373–18382.
´
´
20 F. J. Beltran, A. Aguinaco and J. F. Garcıa-Araya, Water Res., 48 M. A. Fox and M. T. Dulay, Chem. Rev., 1993, 93, 341–357.
2009, 43, 1359–1369. 49 R. M. Alberici and W. F. Jardim, Appl. Catal., B, 1997, 14, 55–68.
21 W. Baran, E. Adamek, A. Sobczak and A. Makowski, Appl. 50 K. Demeestere, J. Dewulf, B. D. Witte and H. Van
Catal., B, 2009, 90, 516–525. Langenhove, Appl. Catal., B, 2005, 60, 93–106.
22 M. N. Abellan, J. Gimenez and S. Esplugas, Catal. Today, 51 H. Yang, G. Li, T. An, Y. Gao and J. Fu, Catal. Today, 2010,
2009, 144, 131–136. 153, 200–207.
23 A. Carabin, P. Drogui and D. Robert, J. Taiwan Inst. Chem. 52 C. S. Guo, J. Xu, S. F. Wang, Y. Zhang, Y. He and X. C. Li,
Eng., 2015, 54, 109–117. Catal. Sci. Technol., 2013, 3, 1603–1611.
24 Y. X. Chen, S. Y. Yang, K. Wang and L. P. Lou, J. Photochem. 53 K. Ishibashi, A. Fujishima, T. Watanabe and K. Hashimoto,
Photobiol., A, 2005, 172, 47–54. J. Photochem. Photobiol., A, 2000, 134, 139–142.
25 L. Hu, P. M. Flanders, P. L. Miller and T. J. Strathmann, 54 S. Fukahori and T. Fujiwara, J. Environ. Manage., 2015, 157,
Water Res., 2007, 41, 2612–2626. 103–110.
26 S. Jiao, S. Zheng, D. Yin, L. Wang and L. Chen, Chemosphere, 55 C. Guo, J. Xu, Y. Zhang and Y. He, RSC Adv, 2012, 2, 4720–
2008, 73, 377–382. 4727.
27 W. Yang, F. Zheng, X. Xue and Y. Lu, J. Colloid Interface Sci., 56 S. Fukahori, H. Ichiura, T. Kitaoka and H. Tanaka, Appl.
2011, 362, 503–509. Catal., B, 2003, 46, 453–462.
28 M. Kahle and C. Stamm, Chemosphere, 2007, 68, 1224–1231. 57 E. Vulliet, C. Emmelin, J.-M. Chovelon, C. Guillard and
´
´
29 P. Calza, C. Medana, M. Pazzi, C. Baiocchi and E. Pelizzetti,
J.-M. Herrmann, Appl. Catal., B, 2002, 38, 127–137.
58 L. Yang, L. E. Yu and M. B. Ray, Water Res., 2008, 42, 3480–
3488.
Appl. Catal., B, 2004, 53, 63–69.
69310 | RSC Adv., 2016, 6, 69301–69310
This journal is © The Royal Society of Chemistry 2016