w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3

Available at www.sciencedirect.com

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

Silver-modified mesoporous TiO2 photocatalyst for water

purification

Zhigang Xiong a, Jizhen Ma a, Wun Jern Ng b, T. David Waite c, X.S. Zhao a,*
a
Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117576,
Singapore
b
Nanyang Environment & Water Research Institute, Nanyang Technological University, Nanyang Avenue, Singapore 639798, Singapore
c
School of Civil and Environmental Engineering, The University of New South Wales, Sydney, NSW 2052, Australia

article info abstract

Article history: Mesoporous anatase (TiO2) was modified with silver (Ag) nanoparticles using a photore-
Received 22 July 2010 duction method. Performance of the resulting TiO2eAg nanocomposites for water purifi-
Received in revised form cation was evaluated using degradation of Rhodamine B (RhB) and disinfection of
5 December 2010 Escherichia coli (E. coli) under ultraviolet (UV) irradiation. The composites with different Ag
Accepted 20 December 2010 loadings were characterized using physical adsorption of nitrogen, X-ray diffraction, X-ray
Available online 24 December 2010 photoelectron spectroscopy and UVeVisible diffuse reflectance spectroscopic techniques.
The results showed that metallic Ag nanoparticles were firmly immobilized on the TiO2
Keywords: surface, which improved electron-hole separation by forming the Schottky barrier at the
Silver nanoparticles TiO2eAg interface. Photocatalytic degradation of RhB and inactivation of E. coli effectively
Mesoporous titanium dioxide occurred in an analogical trend. The deposited Ag slightly decreased adsorption of target
Photocatalysis pollutants, but greatly increased adsorption of molecular oxygen with the latter enhancing
Disinfection production of reactive oxygen species (ROSs) with concomitant increase in contaminant
photodegradation. The optimal Ag loadings for RhB degradation and E. coli disinfection
were 0.25 wt% and 2.0 wt%, respectively. The composite photocatalysts were stable and
could be used repeatedly under UV irradiation.
ª 2011 Elsevier Ltd. All rights reserved.

1. Introduction ultraviolet (UV) irradiation. This has often led to incomplete

The titanium dioxide (TiO2) photocatalyst is gaining growing
research interest (Chen and Mao, 2007; Han et al., 2009;
Hoffmann et al., 1995; Linsebigler et al., 1995; Xu et al., 2007;
Xu and Langford, 1995; Yu et al., 2003). Upon UV light irradia-
tion, TiO2 generates highly reactive oxygen species (ROSs) and
valence holes, which can degrade most recalcitrant pollutants,
including non-biodegradable dyes (Han et al., 2009), phenols
and their derivatives (Xu et al., 2007), and inactivate various
microorganisms, such as bacteria, viruses, and tumor cells
(Sunada et al., 2003b). However, the lifetime of ROSs, such as
O



2 , HO , HOO and H2O2, is relatively short, particularly under
mineralization of organic pollutants (Tachikawa et al., 2007).
To effectively improve the mineralization process, a means is
needed to enhance adsorption of organic pollutants or dis-
solved O2 onto the surface of TiO2 (Kang et al., 2008; Mogyorosi
et al., 2002). However, due to its poor affinity towards organic
pollutants and the low surface area of TiO2 nanoparticles, the
adsorption of organic pollutants on TiO2 surface is relatively
low, resulting in slow photocatalytic degradation rates. To
overcome this shortcoming, TiO2 nanoparticles have been
supported on various porous solids (Hidaka et al., 1990; Kang
et al., 2008; Mogyorosi et al., 2002; Xu and Langford, 1995).
Using advanced porous materials for water treatments has

* Corresponding author. Tel.: þ65 65164727; fax: þ65 67791936.

E-mail address: chezxs@nus.edu.sg (X.S. Zhao).
0043-1354/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2010.12.019
2096 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3
been demonstrated to be a very promising approach to water
purification (Pan et al., 2010; Zhao et al., 2010). The materials
can be synthesized via template (Tian et al., 2002; Yu et al.,
2003) and solegel methods (Pinna and Niederberger, 2008;
Raveendran et al., 2008). Mesoporous TiO2 photocatalysts
have been observed to display an improved performance in the
degradation of chlorophenols and dyes under UV irradiation
(Beyers et al., 2005; Schattka et al., 2002). The porous structure
of mesoporous TiO2 may also provide an adequate environ-
ment for the adsorption of dissolved O2 on the oxygen vacan-
cies (Epling et al., 1998). It is recognized that reduction of the
adsorbed O2 is a rate-limiting step in the photocatalytic
degradation of organic pollutants (Hoffmann et al., 1995;
Sadeghi et al., 1996), and so increase in O2 adsorption on TiO2
surfaces would enhance charge separation, and so generate
more ROSs.
Another factor that greatly restricts the photocatalytic
activities of TiO2 is the low quantum yield of excitons due to the
fast electron-hole (eehþ) recombination. Methods to effec-
tively improve the photocatalytic activity of TiO2 have been
widely explored over the past decade. Direct doping of an
element, optimization of morphology and facets, and adding
additional scavengers are amongst the most effective
approaches (Chen and Mao, 2007; Hoffmann et al., 1995; Xu
et al., 2007; Xu and Langford, 1995). In addition, deposition
of metal nanoparticles with a large work function, such as
Ag (Tran et al., 2006), Pt (Kowalska et al., 2008), and Au (Bannat
et al., 2009; Subramanian et al., 2001; Wang et al., 2008), onto
TiO2 surface has been found to efficiently retard the eehþ
recombination because of the Schottky barrier formed at the
metalesemiconductor interface. Here, the metal nanoparticles
act as a mediator in storing and shuttling photogenerated
electrons from the TiO2 surface to an acceptor. Iliev et al. (Iliev
et al., 2006) demonstrated that deposition of Pt or Ag on the
surface of TiO2 greatly enhanced the photocatalytic degrada-
tion of oxalic acid due to the increased separation of eehþ and
higher rate of O2 reduction. Similarly, Li et al. (2007) homoge-
nously embedded gold particles onto the TiO2 framework,
which significantly improved photocatalytic degradation of
phenol and chromium in solution. Other organic pollutants, like
methanol (Ismail et al., 2009), methylene blue (Wang et al., 2008)
and methyl orange (Arabatzis et al., 2003), could also be more
efficiently degraded by the metaleTiO2 nanocomposite. Reac-
tive oxygen species, and particularly hydroxyl radicals (HO
∙),
produced by the irradiated metaleTiO2 has been considered to
be the dominant species contributing to the degradation of
various organic pollutants (Arabatzis et al., 2003).
The metal-modified TiO2 could also be used for the inacti-
vation of microorganisms. Sunada et al. (Sunada et al., 2003a)
compared the photocatalytic disinfection of Escherichia coli in
pure TiO2 and Cu-loaded TiO2 with the latter displaying an
enhanced bactericidal activity. Silver modification of TiO2 has
been shown to enhance the bactericidal activity of UV-irradi-
ated TiO2 by about seventy folds as a result of both improved
microorganism adsorption to the particle surface and suppres-
sion of eehþ recombination (Zhang et al., 2010). The formed
ROSs was found to lead to perturbation of various cellular
processes and eventually to bacterial death (Chen et al., 2009b).
In the present study, Ag-modified mesoporous anatase
TiO2 was synthesized via a photoreduction method. A non-
biodegradable dye, Rhodamine B (RhB), and a gram-negative
bacterium, E. coli, were used as probes to investigate the
photocatalytic detoxification and disinfection properties of
the composite materials under UV irradiation. The effect of Ag
loadings on catalyst performance and the role of Ag nano-
particles in the photocatalytic degradations were investigated.
2. Materials and methods
2.1. Materials
Titanium isopropoxide, silver nitrate, RhB, ethanol, and
phosphate buffer saline of analytical grade were used without
further purification. Pluronic surfactant poly(ethylene glycol)-
block-poly(propylene glycol)-block-poly(ethylene glycol) (EO20
PO70EO20, P123) was purchased from Aldrich. Deionized water
was used in this work.
2.2. Preparation and modification of mesoporous TiO2
The synthesis of mesoporous anatase TiO2 was similar to that
reported previously (Yu et al., 2003). 6.0 g of P123 and 9.0 mL of
titanium isopropoxide were dissolved in 60 mL of ethanol at
ambient temperature. After the suspension was stirred for 1 h,
30 mL of H2O was added and stirred for another 1 h. After
complete evaporation of ethanol and water, white solids were
obtained. The solids were calcined at 350  C for 5 h in air with
a heating rate of 0.5  C/min. The resultant product was
denoted as m-TiO2.
Modification of mesoporous TiO2 with Ag nanoparticles
was conducted using a photoreduction method (Tran et al.,
2006). 1.0 g of m-TiO2 was suspended in 300 mL of water, fol-
lowed by sonication for 10 min to completely disperse the solid.
Then, AgNO3 solution was added to the suspension. After
stirring for 20 min in the dark, the mixture was irradiated with
a 125 W high-pressure mercury lamp (Philips, Belgium) for
4.0 h. The resulting pale-gray solids were collected, washed
with water and dried at ambient temperature. No Ag ion was
detected in the filtrate indicating that all Ag had been
completely loaded onto the m-TiO2. Different volumes of
AgNO3 solution with a concentration of 6.27  102 M were
used to vary the loading of Ag. The samples thus obtained are
denoted as m-TiO2eAgx%, where x% indicates the weight
percentage of Ag on the composite particles.
2.3. Characterization
X-ray diffraction (XRD) patterns were collected on a Shimadzu
XRD-6000 X-ray diffractometer using Cu Ka irradiation oper-
ated at 40 kV and 30 mA. The average crystallite size was
estimated using the Scherrer equation based on the (101)
reflection of anatase. Adsorptionedesorption isotherms of N2
were measured on a Micromeritics ASAP2020 system at 77 K.
The BrunauereEmmetteTeller surface area (SBET) was calcu-
lated in the relative pressure range of P/P0 ¼ 0.05  0.2. The
pore volume was estimated at P/P0 ¼ 0.97. Diffuse-reflectance
spectra were recorded on a Shimadzu 3100 UVeVis-NIR
spectrometer equipped with an integrating sphere ISR-3100
using BaSO4 as the reference. High-resolution transformation
 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3 2097

electron microscope (HRTEM) images were obtained on a Per-

kineElmer JEM-2010F operated at 200 kV. X-ray photoelectron
spectra (XPS) were collected on an AXIS HIS (Kratos Analytical
Ltd., UK) instrument with an Al Ka irradiation source
(1486.7 eV) operated at 15 kV and 10 mA. Photoluminescence
spectra were collected at room temperature on a PTI lumi-
nescence spectrometer with a solid accessory.

2.4. Photocatalytic degradation of RhB

Photocatalytic degradation of RhB was carried out in a semi-

batch swirl flow monolithic-type photoreactor (Zhou et al.,
2006). Before UV irradiation, a suspension containing 300 mL
of 8.6  103 mM RhB solution and 0.15 g of solid photocatalyst
was sonicated for 5 min and stirred for 20 min in the dark to
allow sorption equilibrium. The mixture was then irradiated
with a 125 W high-pressure mercury lamp. At a given time
interval of irradiation, 5 mL aliquots were withdrawn and
filtered through a 0.45 mm membrane. The concentration of
RhB in the filtrate was analyzed using a Shimadzu 1601 PC
UVeVis spectrophotometer.

2.5. Photocatalytic disinfection of E. coli

The antibacterial experiments were carried out in a sterilized

100 mL glass conical flask containing 50 mL 7.7  108 CFU/mL E.
coli and 0.5 g/L solid photocatalyst under room temperature.
The suspension was irradiated by two 8 W Black light Blue (BLB)
lamps (200 mW/cm2). After a given time interval, 50 mL of the
suspension was sampled, and diluted with sterile phosphate
buffer saline (PBS, 0.03 mol/L, pH 7.2) to yield appropriate
starting concentrations (ensuring cells were well suspended by
vortex mixing before transfer). The suspensions were then
spread on Petri Dishes which had been previously coated with
nutrient agar. After being incubated at 37  C for 18 h, the formed
colonies were counted and used to calculate the amount of
active cells. Each data point was determined three times and
was used as an average. Control experiments without the
photocatalyst were also conducted for comparison.
2.6. Stability of photocatalyst
Stability of the photocatalyst was evaluated as follows: 0.15 g
of the photocatalyst m-TiO2eAg 0.25% was suspended in
300 mL of 8.0  103 mM RhB solution. After a given time
interval of UV irradiation, 5 mL aliquots were withdrawn,
filtered and analyzed on the UVeVis spectrophotometer. After
RhB was completely degraded, the suspension was further
irradiated for 20 min. Then, 15 mL of stock solution (0.16 mM)
of RhB was supplied to restore the initial concentration of RhB
(8.0  103 mM). The suspension was equilibrated again for
20 min in the dark before again being irradiated with UV light.
3. Results and discussion
3.1. Characterization
Fig. 1A shows the adsorptionedesorption isotherms of TiO2 and
Ag-deposited TiO2 photocatalysts. It is seen that all samples
Fig. 1 e (A) Nitrogen adsorption isotherms and (B) pore size
distribution curves of mesoporous TiO2 modified with Ag
nanoparticles of loadings of (a) 0%, (b) 0.1%, (c) 0.25%, (d)
0.5%, (e) 1.0%, (f) 2.0%, and (g) 3.0%.
displayed a type IV isotherm with an H3 hysteresis loop, indi-
cating that mesoporous materials are involved (Sing et al., 1985).
The relatively narrow pore-size distribution curves indicate
a single pore-size system (Fig. 1B). Table 1 summarizes the pore
parameters of the samples. The BET surface area of m-TiO2 was
calculated to be about 166 m2/g, higher than that reported
previously (Yu et al., 2003). Modification with Ag gradually
decreased the surface area with increasing loading of Ag - from
0.1% to 2.0%. Pore volume also decreased - from 0.43 to
0.33 cm3/g on increasing Ag loading from 0.1% to 2.0%. The
above results indicate that Ag had been loaded onto the m-TiO2
sample. As the amount of silver increased, more Ag particles
formed on the surface of mesoporous TiO2. The Ag particles
covered some sorption sites and lowered the pore volume of the
TiO2. However, they did not remarkably affect the pore struc-
ture (Fig. 1B) (Liu et al., 2008). For sample m-TiO2eAg 3.0%, an
obvious decrease in pore size can be seen suggesting that the Ag
particles blocked some of the pores.
Fig. 2 shows the XRD patterns of the samples. All resolved
peaks can be indexed as the (101), (004), (200) and (211) reflec-
tion of pure anatase (JCPDS No. 21-1272) (Chen et al., 2009a; Yu
et al., 2003). The crystal sizes were calculated from the (101)
plane and are given in Table 1. It can be seen that within the Ag
loading range of 0.1%e3.0%, the crystal size of the mesoporous
2098 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3

Table 1 e Physical parameters of TiO2 and Ag-deposited

TiO2.
Crystallite SBET Pore Average
size (nm)a (m2/g) volume pore
(cm3/g)b size (nm)c

m-TiO2 6.7 166 0.43 11.3

m-TiO2eAg 0.1% 7.1 162 0.38 10.8
m-TiO2eAg 0.25% 6.7 157 0.38 11.5
m-TiO2eAg 0.5% 6.9 149 0.38 11.5
m-TiO2eAg 1.0% 6.9 139 0.36 11.6
m-TiO2eAg 2.0% 6.6 128 0.33 11.0
m-TiO2eAg 3.0% 6.6 125 0.29 10.1

a Calculated using Scherrer equation.

b Pore volume at P/P0 ¼ 0.97.
c Estimated from the desorption branches using the BJH model.

anatase was not altered. No reflection peak owing to metallic

Ag can be seen in Fig. 2, presumably due to the small Ag particle Fig. 3 e UVeVisible diffuse reflectance spectra of
size. The TEM image (Fig. S1) confirmed that the size of the Ag mesoporous TiO2 modified with Ag nanoparticles of
particles in sample m-TiO2eAg 3.0% with the highest Ag loadings ranging from 0 (bottom) to 3.0% (top). The label on
loading was smaller than 1.5 nm. The size of the Ag particles in top of each curve is the wavelength of the highest
other samples was too small to be distinguished from the TiO2 absorption.
background using the TEM technique. The Ag 3d XPS spectra of
Ag-modified mesoporous TiO2 samples (Fig. S2) showed two
increased with Ag loading, indicating the deposition of Ag
peaks, at 368 and 374 eV, assigned to Ag 3d5/2 and 3d3/2,
nanoparticles on the TiO2 surface (Linsebigler et al., 1995).
respectively (Guin et al., 2007). No peak due to AgD is seen.
However, the maximum wavelength of the surface plasmon
Considering that no Ag ion was detected in the filtrate after the
absorption varied from one sample to another. As Ag loading
photoreduction process, it thus can be concluded that all AgD
increased from 0.1% to 3.0%, the wavelength of the highest
ions added into the solution had been completely reduced to
surface plasmon absorption gradually shifted from 468 nm to
metallic Ag nanoparticles.
498 nm, accompanied with the broadening of the absorption
The UVeVis diffuse reflectance spectra of all samples are
peaks. It is known that surface plasmon resonance (SPR)
shown in Fig. 3. The m-TiO2 samples exhibited strong
strongly depends on the size and uniformity of the nano-
absorption in the UV region with a sharp absorption edge
particles (He et al., 2002) with the SPR peak becoming broader
located at about 400 nm (Zhang et al., 2009). Upon loading
and shifting to longer wavelengths as the particle size
different amounts of Ag, no change in the absorption edge is
increases, and/or when the particle size distribution becomes
seen. A typical surface plasmon absorption of Ag nano-
broad. The results in Fig. 3 suggest that as Ag loading gradually
particles at around 400e550 nm gradually emerged and
increased, more Ag nanoparticles aggregated on the TiO2
surface. Such an aggregation resulted in a decrease in the pore
volume as evidenced by the data in Table 1.

3.2. Photocatalytic detoxification and disinfection under

UV light irradiation

The photocatalytic properties of the mesoporous anatase TiO2

Fig. 2 e XRD patterns of mesoporous TiO2 before and after
modification with Ag nanoparticles of different loadings.
The standard XRD pattern of anatase is also included for
reference.
and Ag-deposited TiO2 photocatalysts were evaluated by
examination of RhB dye degradation and Gram-negative
bacterium E. coli inactivation under UV irradiation. Before
commencement of light irradiation, the mixtures of RhB or
E. coli solution and solid photocatalyst were stirred for 20 min
in the dark to achieve complete sorption equilibrium.
As shown in Fig. 4A, minor removal of RhB occurred with
catalyst m-TiO2eAg in the absence of UV irradiation. This
removal was due to adsorption of the dye on the solid particle
surface. In the absence of a photocatalyst but under UV irra-
diation, however, a slow degradation was observed. In the
presence of m-TiO2, the concentration of RhB in suspension
decreased rapidly with irradiation time. The data could be
satisfactorily described by the first-order kinetic equation, ln
(C/C0) ¼ kappt, where kapp is the apparent rate constant, C and
 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3
Fig. 4 e Photocatalytic degradation of (A) RhB and (B) E. coli
in different system. The initial concentration of RhB and
E. coli were 8.6 3 10L3 mM and 7.7 3 108 CFU/mL,
respectively.
C0 are the concentrations of RhB at time t ¼ t and t ¼ 0,
respectively. The rate constant of the photocatalytic degrada-
tion with m-TiO2 was calculated to be about 0.044 min1. When
m-TiO2eAg 0.25% was used, the RhB degradation was
enhanced exhibiting a rate constant of 0.069 min1, compa-
rable to 0.073 min1 of P25. The absorption spectra of RhB were
gradually decreased with irradiation time with no shift of the
main absorption peak, suggesting complete cleavage of RhB
chromophores. Although the UV energy is also adequate for
RhB excitation, the photosensitized degradation, as evidenced
by a concomitant wavelength shift of the absorption band to
shorter wavelengths (Sung-Suh et al., 2004; Wu et al., 1998), can
be neglected compared to the photocatalytic degradation.
The antibacterial properties of the catalysts were studied
using E. coli as a model strain. In the presence of UV irradiation
and without a photocatalyst (Fig. 4B) or in the presence of
m-TiO2 without UV light, the number of viable E. coli cells
decreased slowly. With m-TiO2eAg 3% and without UV irradi-
ation, the viable E. coli number decreased sharply in the first
60 min, then became stable. This behavior is different to RhB
degradation and may be partially attributed to the intrinsic
disinfection properties of the deposited Ag nanoparticles (Lv
2099
et al., 2009; Sharma et al., 2009). When the suspension was
irradiated with UV in the presence of m-TiO2eAg 3%, the
inactivation of E. coli was greatly enhanced. Analyzing using
first-order kinetics (Chen et al., 2009c), the total rate constants
decreased in the order m-TiO2eAg 3% (0.102 min1) > m-TiO2
(0.055 min1) > m-TiO2eAg 3% (dark, 0.040 min1) > UV
(0.024 min1) > m-TiO2 (dark, 0.0073 min1). The contribution of
Ag inactivation to the total rate constant could be estimated to
be about 0.033 min1.After eliminating the Ag inactivation
effect, the photocatalytic disinfection effect can be assessed as
exhibiting a net rate constant of k ¼ 0.062 min1, larger than
that of m-TiO2. This trend is similar to that observed for the
photocatalytic degradation of RhB. The time-course of inacti-
vation of E. coli however is substantially more complicated than
the degradation of RhB, possibly as a result of the combined
effects of the antimicrobial activity of Ag itself and the photo-
catalytic disinfection effect of the TiO2eAg composite with the
synergic effect of the two processes leading to a significant
improvement of antibacterial activity (Liu et al., 2008). Silver is
known for its antimicrobial properties with some investigators
proposing that Ag nanoparticles may attach to the surface of
the cell membrane and so disturb permeability and respiration
functions of the cell (Kvitek et al., 2008). Notwithstanding this,
the bactericidal mechanism of Ag nanoparticles remains to be
fully understood (Sharma et al., 2009). The photocatalytic
disinfection process has also been widely investigated. Chen
and coworkers (Chen et al., 2009c) compared the photocatalytic
degradation of formaldehyde and inactivation of E. coli with
TiO2 and concluded that both processes depended on the ROS
production rate. This is also apparent in the present system
since the generation of ROSs is an important phenomenon in
the TiO2eAg suspension. Disinfection performance of the
photocatalyst can be evaluated by the analogy method based
on dye degradation under UV irradiation since the latter is
easier to measure quantitatively.
3.3. Effect of Ag loading
Deposition of Ag nanoparticles on TiO2 surface not only
enhances photocatalytic reactions (Sung-Suh et al., 2004) but
Fig. 5 e Effect of Ag loading on the photocatalytic
degradation of RhB (C) and E. coli (B), and on the
photoluminescent intensity ratio between Ag-deposited
TiO2 and TiO2 (¤).
2100 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3
Fig. 6 e Effect of Ag loading on the adsorption of RhB and
the surface area using O2 as an adsorptive. The initial
concentration of RhB was 8.6 3 10L3 mM.
also provides additional disinfection properties (Liu et al.,
2008), however, both activities greatly depend on the amount
of Ag loaded. As shown in Fig. 5, the photocatalytic degradation
of RhB and disinfection of E. coli were changed by the loading of
Ag nanoparticles with the rate constants of both processes
higher at low loadings and lower at high loadings. The optimal
loadings of Ag were different with about 0.25 wt% found to be
optimal for RhB degradation and about 2.0 wt% optimal for E.
coli disinfection. This is perhaps not surprising given that the
photocatalyst itself exhibits bactericidal properties with the
extent of inhibition (in the dark) increasing with increasing Ag
loading while the degradation of RhB is strictly light depen-
dent. Further discussion of the photocatalytically-mediated
degradation of RhB is provided below. Fig. S3 shows the pho-
toluminescence spectra of TiO2 with different Ag loadings. TiO2
has a strong luminescence spectrum with the luminescence
intensity decreasing progressively on increasing the Ag loading
from 0 to 2.0% (Fig. 5). This effect may be attributed to Schottky
barrier formation at the TiO2emetal interface, indicating
a strong inhibition of eehþ recombination as reported for
other noble metals (Bannat et al., 2009; Sadeghi et al., 1996;
Subramanian et al., 2001; Tada et al., 2009; Xin et al., 2005),
eehþ recombination was decreased at the low loadings and
enhanced at the high loadings with an optimal loading of Ag at
about 0.10%. However, the optimal Ag loading for the
Fig. 7 e Schematic illustration of the effect of Ag on the adsorption of RhB and O2.
photocatalytic degradation of RhB (about 0.25 wt%) was much
larger than that for maximal eehþ separation. The results
suggest that the deposition of Ag nanoparticles on TiO2 might
have other roles other than to simply depress eehþ
recombination.
Considering the deposited Ag particles covered the surface
of TiO2, the adsorption affinity of the solid catalyst towards both
dissolved oxygen and RhB would most likely be altered. As
shown in Fig. 6, upon Ag deposition, RhB adsorption decreased
with increasing Ag loading. Since RhB molecules mainly
adsorbed on TiO2 surface with the diethylamino group (Wu
et al., 1998) or the carboxyl group (Zhao et al., 2008), deposi-
tion of Ag particles on TiO2 would inevitably covered some
sorption sites, resulting in a decrease of the RhB adsorption
(Fig. 7). Control experiments (Fig. S4A) showed that the
adsorption of RhB on m-TiO2eAg 0.25% surface followed
a Langmuir model with the maximum absorption capacity at
about 5.3  103 mmol g1. The degradation rate increased
almost linearly with the extent of adsorption of RhB (Fig. S4B)
and could be adequately described by the LangmuireHinshel-
wood (LeH) equation, r ¼ kLHKCe/(1 D KCe), yielding an LeH rate
constant of kLH, ¼ 7.9  104 mM min1 and Langmuir adsorp-
tion constant of K ¼ 34.3 mM1. The decrease in RhB adsorption
with Ag loading would inevitably be expected to lower the rate
of photocatalytic degradation. Significantly, RhB degradation
was much more rapid in an O2 equilibrated suspension than in
an air saturated solution (Table S1) and was greatly inhibited in
an atmosphere of pure N2. Since photocatalytic degradation
greatly depends on the amount of ROSs, which were produced
by electron transfer between excited TiO2 and the adsorbed O2
(Hoffmann et al., 1995; Sun and Xu, 2009), increasing the
concentration of O2 presumably greatly increases the rate of
production of ROSs, thus enhancing the degradation of RhB. O2
adsorptionedesorption isotherms obtained through solid-gas
phase studies (Fig. S5) indicate that as Ag loading increased
from 0 to 3.0%, the BET surface area gradually increased from
134 to 156 m2/g (Fig. 6) and the pore volume increased by about
11%. Apparently, the increased affinity of the catalyst to O2
could be attributed to the deposition of Ag nanoparticles, which
increased the adsorption of oxygen on the photocatalyst
surface (Fig. 7) (Xin et al., 2005). As such, two factors may be
considered to account for the promotion of RhB degradation.
First, electrons are likely to be transferred from the TiO2 surface
to the Ag particles due to formation of the Schottky barrier at the
interface of TiO2 and Ag (Fig. S3) with the adsorption of O2 on the
 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3
Fig. 8 e Reuse experiment for RhB degradation over
m-TiO2eAg 0.25%. The hollow symbols represent the
concentration of RhB before adsorption.
Ag surface facilitating the electron transfer process. Second,
since the reduced O2 will transform to various ROSs, the
enhancement of O2 adsorption would generate more ROSs in
solution, leading to enhanced RhB degradation. Thus, the
impact of the Ag particles on the photocatalytic activity can be
summarized as follows: (1) the Ag particles created a Schottky
barrier that depressed the eehþ recombination and promoted
the utilization of excitons, (2) the Ag particles decreased the
adsorption of RhB on the catalyst surface, resulting in a decline
of the photoactivity, (3) the Ag particles increased the adsorp-
tion affinity of the catalyst towards dissolved oxygen, leading to
an enhanced production of ROSs and the RhB degradation rate
therefore. However, compared to the decrease in RhB adsorp-
tion with increase in Ag loadings, the incremental increase in
extent of O2 adsorption appeared to play a much more impor-
tant role in the photocatalytic reactions. Thus, the optimal Ag
loading further increased to 0.25 wt% as compared to sample m-
TiO2eAg 0.10% with the lowest eehþ recombination.
3.4. Stability of photocatalyst
The recyclability of the catalyst m-TiOeAg 0.25% is shown in
Fig. 8. Since the photocatalytic inactivation of E. coli follows
a similar trend as RhB degradation (Fig. 4), stability of the
catalyst was evaluated with RhB degradation. It can be seen
that the photocatalyst was recyclable for continuous adsorp-
tion and removal of RhB under UV irradiation, although the
photoactivity and sorption capacity gradually decreased from
one run to another. The results can be attributed to two
reasons. First, any intermediates formed would be expected to
compete for the adsorption sites and reactive oxygen species
thereby diminishing oxidation effectiveness. Second, the
catalyst concentration was decreased from one run to another
because of the removal of 5 mL aliquots for analysis purposes.
It is supported by the results shown in Fig. S6. It is seen that
the rate constants of both RhB degradation and E. coli inacti-
vation increased with the catalyst loading under the other-
wise identical conditions to that of Fig. 4.
2101
4. Conclusions
Mesoporous anatase TiO2 materials modified with Ag nano-
particles displayed significant capacity for the degradation of
RhB and disinfection of E. coli under UV light irradiation. Both
dye degradation and E. coli inactivation increased at low Ag
loadings and decreased at high loadings with optimal Ag load-
ings of about 0.25 and 2.0 wt% for RhB degradation and E. coli
disinfection, respectively. Besides formation of the Schottky
barrier at the TiO2emetal interface, the deposited Ag nano-
particles slightly decreased the extent of adsorption of RhB, but
increased the affinity of the surface to oxygen, which seemed to
play a much more important role in enhancing the photocata-
lytic reactions. After four cycles of reuse, the composite catalyst
still showed significant capacity for dye degradation.
Acknowledgment
This work was supported by Environment and Water Industry
Development Council (EWI) of Singapore under project
number MEWR 651/06/161.
Appendix. Supporting information.
TEM images, Ag 3d XPS spectra, and PL spectra of m-TiO2eAg
nanocomposite; Sorption isotherms of RhB and O2, and the
effect of catalyst loading on the photoactivity.
Appendix. Supplementary data
Supplementary data associated with this article can be found
in the online version at doi: doi:10.1016/j.watres.2010.12.019.
references
Arabatzis, I.M., Stergiopoulos, T., Bernard, M.C., Labou, D.,
Neophytides, S.G., Falaras, P., 2003. Silver-modified titanium
dioxide thin films for efficient photodegradation of methyl
orange. Appl. Catal. B. 42, 187e201.
Bannat, I., Wessels, K., Oekermann, T., Rathousky, J.,
Bahnemann, D., Wark, M., 2009. Improving the photocatalytic
performance of mesoporous titania films by modification with
gold nanostructures. Chem. Mater. 21, 1645e1653.
Beyers, E., Cool, P., Vansant, E.F., 2005. Anatase formation during
the synthesis of mesoporous titania and its photocatalytic
effect. J. Phys. Chem. B. 109, 10081e10086.
Chen, D.H., Huang, F.Z., Cheng, Y.B., Caruso, R.A., 2009a. Mesoporous
anatase TiO2 beads with high surface areas and controllable pore
sizes: a superior candidate for high-performance dye-sensitized
solar cells. Adv. Mater. 21, 2206e2207.
Chen, F.N., Yang, X.D., Wu, Q., 2009b. Photocatalytic oxidation of
Escherischia coli, aspergillus niger, and formaldehyde under
different ultraviolet irradiation conditions. Environ. Sci.
Technol. 43, 4606e4611.
Chen, F.N., Yang, X.D., Xu, F.F., Wu, Q., Zhang, Y.P., 2009c.
Correlation of photocatalytic bactericidal effect and organic
matter degradation of TiO2 part I: observation of phenomena.
Environ. Sci. Technol. 43, 1180e1184.
2102 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3
Chen, X., Mao, S.S., 2007. Titanium dioxide nanomaterials:
synthesis, properties, modifications, and applications. Chem.
Rev. 107, 2891e2959.
Epling, W.S., Peden, C.H.F., Henderson, M.A., Diebold, U., 1998.
Evidence for oxygen adatoms on TiO2(110) resulting from O-2
dissociation at vacancy sites. Surf. Sci. 412e13, 333e343.
Guin, D., Manorama, S.V., Latha, J.N.L., Singh, S., 2007.
Photoreduction of silver on bare and colloidal TiO2
nanoparticles/nanotubes: synthesis, characterization, and
tested for antibacterial outcome. J. Phys. Chem. C 111,
13393e13397.
Han, F., Kambala, V.S.R., Srinivasan, M., Rajarathnam, D.,
Naidu, R., 2009. Tailored titanium dioxide photocatalysts for
the degradation of organic dyes in wastewater treatment:
a review. Appl. Catal. A 359, 25e40.
He, J.H., Ichinose, I., Kunitake, T., Nakao, A., 2002. In situ
synthesis of noble metal nanoparticles in ultrathin TiO2-gel
films by a combination of ion-exchange and reduction
processes. Langmuir 18, 10005e10010.
Hidaka, H., Yamada, S., Suenaga, S., Zhao, J., Serpone, N.,
Pelizzetti, E., 1990. Photodegradation of surfactants. 6.
Complete photocatalytic degradation of anionic, cationic and
nonionic surfactants in aqueous semiconductor dispersions. J.
Mol. Catal. 59, 279e290.
Hoffmann, M.R., Martin, S.T., Choi, W.Y., Bahnemann, D.W., 1995.
Environmental applications of semiconductor photocatalysis.
Chem. Rev. 95, 69e96.
Iliev, V., Tomova, D., Bilyarska, L., Eliyas, A., Petrov, L., 2006.
Photocatalytic properties of TiO2 modified with platinum and
silver nanoparticles in the degradation of oxalic acid in
aqueous solution. Appl. Catal. B. 63, 266e271.
Ismail, A.A., Bahnemann, D.W., Bannat, I., Wark, M., 2009. Gold
nanoparticles on mesoporous interparticle networks of
titanium dioxide nanocrystals for enhanced photonic
efficiencies. J. Phys. Chem. C 113, 7429e7435.
Kang, C., Jing, L., Guo, T., Cui, H., Zhou, J., Fu, H., 2008.
Mesoporous SiO2-modified nanocrystalline TiO2 with high
anatase thermal stability and large surface area as efficient
photocatalyst. J. Phys. Chem. C 113, 1006e1013.
Kowalska, E., Remita, H., Colbeau-Justin, C., Hupka, J., Belloni, J.,
2008. Modification of titanium dioxide with platinum ions and
clusters: application in photocatalysis. J. Phys. Chem. C 112,
1124e1131.
Kvitek, L., Panacek, A., Soukupova, J., Kolar, M., Vecerova, R.,
Prucek, R., Holecova, M., Zboril, R., 2008. Effect of surfactants
and polymers on stability and antibacterial activity of silver
nanoparticles (NPs). J. Phys. Chem. C 112, 5825e5834.
Li, H.X., Bian, Z.F., Zhu, J., Huo, Y.N., Li, H., Lu, Y.F., 2007.
Mesoporous Au/TiO2 nanocomposites with enhanced
photocatalytic activity. J. Am. Chem. Soc. 129, 4538e4539.
Linsebigler, A.L., Lu, G.Q., Yates, J.T., 1995. Photocatalysis on TiO2
surfaces-principles, mechanisms, and selected results. Chem.
Rev. 95, 735e758.
Liu, Y., Wang, X.L., Yang, F., Yang, X.R., 2008. Excellent
antimicrobial properties of mesoporous anatase TiO2 and Ag/
TiO2 composite films. Micropor. Mesopor. Mater. 114, 431e439.
Lv, L., Lu, Y.Q., Ng, W.J., Zhao, X.S., 2009. Bactericidal activity of
silver nanoparticles supported on microporous titanosilicate
ETS-10. Micropor. Mesopor. Mater. 120, 304e309.
Mogyorosi, K., Farkas, A., Dekany, I., Ilisz, I., Dombi, A., 2002. TiO2-
based photocatalytic degradation of 2-chlorophenol adsorbed
on hydrophobic clay. Environ. Sci. Technol. 36, 3618e3624.
Pan, J.H., Dou, H.Q., Xiong, Z.G., Xu, C., Ma, J.Z., Zhao, X.S., 2010.
Porous photocatalysts for advanced water purifications. J.
Mater. Chem. 20, 4512e4528.
Pinna, N., Niederberger, M., 2008. Surfactant-free nonaqueous
synthesis of metal oxide nanostructures. Angew. Chem. Int.
Ed. Engl. 47, 5292e5304.
Raveendran, P., Eswaramoorthy, M., Bindu, U., Chatterjee, M.,
Hakuta, Y., Kawanami, H., Mizukami, F., 2008. Template-free
formation of meso-structured anatase TiO2 with spherical
morphology. J. Phys. Chem. C 112, 20007e20011.
Sadeghi, M., Liu, W., Zhang, T.G., Stavropoulos, P., Levy, B., 1996.
Role of photoinduced charge carrier separation distance in
heterogeneous photocatalysis: oxidative degradation of
CH3OH vapor in contact with Pt/TiO2 and cofumed
TiO2eFe2O3. J. Phys. Chem. 100, 19466e19474.
Schattka, J.H., Shchukin, D.G., Jia, J.G., Antonietti, M., Caruso, R.A.,
2002. Photocatalytic activities of porous titania and titania/
zirconia structures formed by using a polymer gel templating.
Chem. Mater. 14, 5103e5108.
Sharma, V.K., Yngard, R.A., Lin, Y., 2009. Silver nanoparticles:
green synthesis and their antimicrobial activities. Adv. Colloid
Interfac. 145, 83e96.
Sing, K.S.W., Everett, D.H., Haul, R.A.W., Moscou, L.,
Pierotti, R.A., Rouquerol, J., Siemieniewska, T., 1985.
Reporting physisorption data for gas solid systems with
special reference to the determination of surface-area and
porosity (recommendations 1984). Pure Appl. Chem. 57,
603e619.
Subramanian, V., Wolf, E., Kamat, P.V., 2001. Semiconductor-
metal composite nanostructures. To what extent do metal
nanoparticles improve the photocatalytic activity of TiO2
films? J. Phys. Chem. B. 105, 11439e11446.
Sun, Q., Xu, Y.M., 2009. Sensitization of TiO2 with aluminum
phthalocyanine: factors influencing the efficiency for
chlorophenol degradation in water under visible light. J. Phys.
Chem. C 113, 12387e12394.
Sunada, K., Watanabe, T., Hashimoto, K., 2003a. Bactericidal
activity of copper-deposited TiO2 thin film under weak UV
light illumination. Environ. Sci. Technol. 37, 4785e4789.
Sunada, K., Watanabe, T., Hashimoto, K., 2003b. Studies on
photokilling of bacteria on TiO2 thin film. J. Photochem.
Photobiol. A 156, 227e233.
Sung-Suh, H.M., Choi, J.R., Hah, H.J., Koo, S.M., Bae, Y.C., 2004.
Comparison of Ag deposition effects on the photocatalytic
activity of nanoparticulate TiO2 under visible and UV light
irradiation. J. Photochem. Photobiol. A 163, 37e44.
Tachikawa, T., Fujitsuka, M., Majima, T., 2007. Mechanistic
insight into the TiO2 photocatalytic reactions: design of new
photocatalysts. J. Phys. Chem. C 111, 5259e5275.
Tada, H., Kiyonaga, T., Naya, S., 2009. Rational design and
applications of highly efficient reaction systems
photocatalyzed by noble metal nanoparticle-loaded titanium
(IV) dioxide. Chem. Soc. Rev. 38, 1849e1858.
Tian, B.Z., Yang, H.F., Liu, X.Y., Xie, S.H., Yu, C.Z., Fan, J., Tu, B.,
Zhao, D.Y., 2002. Fast preparation of highly ordered
nonsiliceous mesoporous materials via mixed inorganic
precursors. Chem. Commun., 1824e1825.
Tran, H., Scott, J., Chiang, K., Amal, R., 2006. Clarifying the role
of silver deposits on titania for the photocatalytic
mineralisation of organic compounds. J. Photochem.
Photobiol. A 183, 41e52.
Wang, X., Mitchell, D.R.G., Prince, K., Atanacio, A.J., Caruso, R.A.,
2008. Gold nanoparticle incorporation into porous titania
networks using an agarose gel templating technique for
photocatalytic applications. Chem. Mater. 20, 3917e3926.
Wu, T.X., Liu, G.M., Zhao, J.C., Hidaka, H., Serpone, N., 1998.
Photoassisted degradation of dye pollutants. V. Self-
photosensitized oxidative transformation of rhodamine B
under visible light irradiation in aqueous TiO2 dispersions.
J. Phys. Chem. B. 102, 5845e5851.
Xin, B., Jing, L., Ren, Z., Wang, B., Fu, H., 2005. Effects of
simultaneously doped and deposited Ag on the photocatalytic
activity and surface states of TiO2. J. Phys. Chem. B. 109,
2805e2809.
 w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 2 0 9 5 e2 1 0 3
Xu, Y., Lv, K., Xiong, Z., Leng, W., Du, W., Liu, D., Xue, X., 2007.
Rate enhancement and rate inhibition of phenol degradation
over irradiated anatase and rutile TiO2 on the addition of NaF:
new insight into the mechanism. J. Phys. Chem. C 111,
19024e19032.
Xu, Y.M., Langford, C.H., 1995. Enhanced photoactivity of
a titanium(IV) oxide-supported on ZSM5 and Zeolite-A at low-
coverage. J. Phys. Chem. 99, 11501e11507.
Yu, J.C., Zhang, L.Z., Zheng, Z., Zhao, J.C., 2003. Synthesis and
characterization of phosphated mesoporous titanium
dioxide with high photocatalytic activity. Chem. Mater. 15,
2280e2286.
Zhang, D.Q., Li, G.S., Yu, J.C., 2010. Inorganic materials for
photocatalytic water disinfection. J. Mater. Chem. 20,
4529e4536.
2103
Zhang, F.L., Zheng, Y.H., Cao, Y.N., Chen, C.Q., Zhan, Y.Y., Lin, X.
Y., Zheng, Q., Wei, K.M., Zhu, J.F., 2009. Ordered mesoporous
AgeTiO2-KIT-6 heterostructure: synthesis, characterization
and photocatalysis. J. Mater. Chem. 19, 2771e2777.
Zhao, D., Chen, C., Wang, Y., Ma, W., Zhao, J., Rajh, T., Zang, L.,
2008. Enhanced photocatalytic degradation of dye pollutants
under visible irradiation on Al(III)-modified TiO2: structure,
interaction, and interfacial electron transfer. Environ. Sci.
Technol. 42, 308e314.
Zhao, D.Y., Jaroniec, M., Hsiao, B.S., 2010. Editorial for themed
issue on “advanced materials in water treatments”. J. Mater.
Chem. 20, 4476e4477.
Zhou, J.K., Zhang, Y.X., Zhao, X.S., Ray, A.K., 2006.
Photodegradation of benzoic acid over metal-doped TiO2. Ind.
Eng. Chem. Res. 45, 3503e3511.