Journal of Molecular Catalysis A: Chemical 242 (2005) 135–140

Synthesis of photocatalytic nanosized TiO2–Ag particles with

sol–gel method using reduction agent
Man Sig Lee a , Seong-Soo Hong b , Madjid Mohseni a,∗
a Department of Chemical & Biological Engineering, The University of British Columbia, 2216 Main Mall,
Vancouver, BC, Canada V6T 1Z4
b Division of Applied Chemical Engineering, Pukyong National University, 100 Yongdang-dong,

Nam-ku, Pusan 608-739, South Korea

Received 17 January 2005; received in revised form 28 July 2005; accepted 28 July 2005
Available online 6 September 2005

Abstract

TiO2 –Ag nonoparticles were prepared with the sol–gel method using a reduction agent. The physical properties of the prepared particles
were investigated by several characterization techniques. The major phase of all the prepared TiO2 –Ag particles was an anatase structure
regardless of the AgNO3 content. When the AgNO3 content was 2 mmol/mol of TiO2 , the major phase (1 1 1) of silver could be clearly seen.
The crystallite size of the TiO2 particles calcined at 300 ◦ C was 5–6 nm and that of the Ag particles increased from 10 to 15 nm with increasing
AgNO3 content. The high-resolution transmission electron micrographs (HR-TEM) showed that TiO2 –Ag nanoparticles possessed a spherical
morphology with a narrow size distribution. The lattice fringe was 3.5 Å, which corresponds to the lattice spacing of (1 0 1) plane in the
anatase phase. In addition, the presence of Ag in TiO2 –Ag nanoparticle prepared by the sol–gel method using a reduction agent improved the
photodegradation of p-nitrophenol and the photocatalytic activity of TiO2 –Ag increased with the increase in the AgNO3 content.
© 2005 Elsevier B.V. All rights reserved.

Keywords: TiO2 –Ag particles; Sol–gel process; Reduction agent; Photocatalytic degradation of p-nitrophenol

1. Introduction results in a low efficiency of photocatalysis. For the purpose

of overcoming these limitations of TiO2 as a photocatalyst,
Titanium dioxide, among the transition metal oxide semi- numerous studies have been recently performed to enhance
conductors widely used in photocatalysis, is the most suitable electron–hole separation and to extend the absorption range
photocatalyst due to its high photocatalytic efficiency and of TiO2 into the visible range. These studies include doping
chemical and physical stability under the reaction conditions. metal ions into the TiO2 lattice [2], dye photosensitization
However, there are still many challenges to be addressed for on the TiO2 surface [3], and deposition of noble metals [4].
improving the photocatalytic activity of TiO2 . Because the In particular, noble metal-modified semiconductor nanopar-
semiconductor TiO2 has a high band-gap (Eg > 3.2 eV), it is ticles have become the focus of many studies for maximizing
excited only by UV sources with wavelengths shorter than the efficiency of photocatalytic reactions. The noble metals
388 nm in order to inject electrons into the conduction band such as Pt [5] and Au [6] deposited or doped on TiO2 have
and to leave holes in the valence band [1]. Thus, this practi- the high Schottky barriers among the metals and thus, act as
cally limits the use of sunlight or visible light as an irradiation electron traps, facilitating electron–hole separation and pro-
source in photocatalytic reactions on TiO2 . In addition, the moting interfacial electron transfer process [7].
high rate of electron–hole recombination on TiO2 particles Silver is particularly suitable for industrial applications
due to its low cost and easy preparation. The effects of Ag
∗ dopants on the lattice or surface of TiO2 have been examined
Corresponding author. Tel.: +1 604 822 0047; fax: +1 604 822 5407.
E-mail address: [email protected] (M. Mohseni). [8,9]. TiO2 loaded with silver enables the catalyst to perform

1381-1169/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2005.07.038
136 M.S. Lee et al. / Journal of Molecular Catalysis A: Chemical 242 (2005) 135–140

more effectively and shortens the illumination period [10]. 2.2. Characterization of prepared TiO2 –Ag
He et al. [11] have investigated the effect of silver doping on
the microstructure and photocatalytic activity of TiO2 films The major phase of the synthesized particles was analyzed
prepared by the sol–gel method. It is found that suitable silver by an X-ray diffractometer (XRD) (Rigaku D/MAXIIC)
dopant can increase the activity and the mechanism is mainly using Cu K␣ radiation. Crystallite size of the prepared par-
attributed to the change of anatase grain size. ticles was determined from the broadening of the anatase
In this paper, nanosized TiO2 –Ag particles were prepared main peak at 2θ = 25.3◦ by the Scherrer equation [12]. The
by a simple sol–gel method process using a reduction agent. prepared particles were dispersed in ethyl alcohol and then
The physical properties of the prepared particles were inves- sonicated ultrasonically to separate out individual particles
tigated by UV-DRS, X-ray diffractometer (XRD), energy- for the determination of the particle size. The particle size and
dispersive (EDS), and high-resolution transmission electron external morphology of the particles were examined using
micrographs (HR-TEM). The effects of the AgNO3 content an electron microscope (JEOL JEM 2020 200 kV electron
on the physical properties of nanosized TiO2 –Ag particles microscope, Cs = 0.5 mm, point resolution of 0.19 nm) with a
and their impacts on the photocatalytic decomposition of a beryllium window energy-dispersive detector. In addition, the
model compound, p-nitrophenol, were also studied. band-gaps of the prepared particles were obtained by the mea-
surement of absorbance using a UV-DRS instrument (Varian
Cary 100, Varian Inc.).
2. Experimental
2.3. Photocatalytic reaction experiment
2.1. Synthesis of nanosized TiO2 –Ag
A biannular quartz glass photoreactor with the lamp
Nanosized TiO2 –Ag particles were prepared by a sol–gel immersed in the inner part was used for all the photocat-
process involving a reduction agent. Titanium tetraisopropox- alytic experiments (Fig. 1). The batch photoreactor was filled
ide (TTIP, 98% Aldrich) and silver nitrate (AgNO3 , Aldrich) with 500 ml of an aqueous dispersion in which the concen-
were used as precursors of titania and silver, respectively. trations of catalyst and p-nitrophenol were 1.0 and 50 mg/L,
Sodium citrate tribasic dihydrate (C6 H5 Na3 ·2H2 O, Aldrich) respectively. Continuous stirring maintained uniform con-
was used as a reduction agent, and water used in the experi- centration and temperature throughout the reactor. A 500 W
ments was doubly distilled and deionized. high-pressure mercury lamp (Kumkang Co.) was used for
AgNO3 (1.0 and 2.0 mmol, respectively) was mixed with the irradiation of the solution. The circulation of water in the
500 mL of 4.0 mmol/L sodium citrate solution in a sealed quartz glass tube between the reactor and the lamp allowed
four-way flask, and the reaction temperature was raised to cooling of the lamp and maintaining the reactor at the desired
80 ◦ C with continuous stirring. The color of the solution was temperature. A mixture of oxygen and nitrogen (80% nitro-
observed to change from colorless to violet–brown indicating gen and 20% oxygen) was used at a flow rate of 160 mL/min.
the reduction of silver. Then 1 mol titanium tetraisopropoxide Nitrogen was used as a carrier gas and pure oxygen was used
and 0.15 mol HNO3 were added to the solution, and the reac- to supply oxygen for the oxidation reaction. The samples
tion was maintained at 50 ◦ C for 24 h to obtain TiO2 –Ag sol. were immediately centrifuged and the quantitative determi-
The prepared sol was dried at 105 ◦ C for 1 day, and calcined nation of p-nitrophenol was performed by a UV–vis spec-
in the range between 300 and 700 ◦ C for 2 h in air. trophotometer (Shimazu UV-240).

Fig. 1. Schematic diagram of the experimental apparatus for photocatalytic reaction.

 M.S. Lee et al. / Journal of Molecular Catalysis A: Chemical 242 (2005) 135–140 137

3. Results and discussion

3.1. UV-DRS analysis

The nanosized TiO2 and TiO2 –Ag sols synthesized in this

experiment were transparent and stable for several months.
Fig. 2(a) and (b) are the UV–vis spectra of the pure TiO2
sol and Ag sol prepared, respectively, by the sol–gel method.
Fig. 2(c) is the UV–vis spectra of the TiO2 –Ag 2 mmol sol
prepared by the sol–gel method using a reduction agent. From
Fig. 2(c), it can be seen that there is a framework silver species
at 405 nm, indicating that the TiO2 –Ag particles were suc-
cessfully prepared by the sol–gel method using sodium citrate
as the reduction agent.
It is well known that the yield of photogenerated
electron–hole pair depends primarily on the intensity of inci-
dent photons with energy exceeding or equaling the TiO2 Fig. 3. Diffuse reflectance spectra of nanoparticles prepared at different con-
ditions.
band-gap energy. The effect of silver addition on the band
energy of TiO2 was investigated with diffuse reflectance spec-
troscopy (DRS). The DRS of the pure TiO2 and TiO2 –Ag (1 1 1) of silver could be clearly seen, suggesting that the
particles calcined at 300 ◦ C are shown in Fig. 3. Compari- TiO2 –Ag nanoparticles were successfully prepared by this
son of the diffuse reflectance spectra of the pure TiO2 and process involving the reduction agent. The crystallite sizes
TiO2 –Ag indicated that the band energy of TiO2 for silver of the particles were determined from the broadening of the
addition did not change and the band-gap of the particles anatase main peak by the Scherrer equation [12]. The crys-
was 3.1 eV. This suggested that the photocatalytic activity of tallite sizes of all the prepared particles were between 5 and
TiO2 –Ag particles was not due to the change of the band 6 nm (Table 1). As shown in Table 1, the addition of AgNO3
energy of TiO2 . did not significantly affect the crystallite size of the particles.
Fig. 5 shows the XRD patterns of the TiO2 –Ag 2 mmol
3.2. X-ray diffraction analysis particles calcined at different temperatures. It is well known
that calcination improves the crystallinity of the particles
The crystalline phase of the prepared nanoparticles were with the amorphous TiO2 changing into the anatase phase
analyzed by XRD, and the results are shown in Fig. 4. These and then the anatase phase changing into the rutile phase
particles were only calcined at 300 ◦ C for 2 h in air. The with increasing calcination temperature. It is reported that
major phase of all the prepared TiO2 –Ag particles is an the transformation of the anatase phase into the rutile phase
anatase structure regardless of the AgNO3 content. When the occurs at between 450 and 600 ◦ C and that the transforma-
AgNO3 content was 2 mmol/mol of TiO2 , the major phase tion temperature depends on the kind of precursors and the
preparation conditions and properties of the particles [13].

Fig. 2. UV–vis adsorption spectra of nanosol prepared at different condi- Fig. 4. X-ray diffraction patterns of TiO2 –Ag particles prepared at different
tions: (a) TiO2 (b) Ag, and (c) TiO2 –Ag. AgNO3 content: (a) TiO2 –Ag 1 mmol and (b) TiO2 –Ag 2 mmol.
138 M.S. Lee et al. / Journal of Molecular Catalysis A: Chemical 242 (2005) 135–140

Table 1
Characterization of nanoparticles prepared at different conditions
Catalysta XRD TEM Activityd

Structure Crystallite sizeb (nm) Crystallite sizec (nm) k (min−1 ) × 10−2

Pure TiO2 Anatase 6 <5/– 1.8
TiO2 –Ag 1 mmol Anatase–Ag 5 <5/10 2.9
TiO2 –Ag 2 mmol Anatase–Ag 6 <5/15 3.4
a Calcinations temperature = 300 ◦ C.
b Obtained by the Scherrer equation.
c Crystallite size of TiO2 and Ag, respectively.
d Apparent first-order constants (k ) of p-nitrophenol after 60 min irradiation.

As shown in Fig. 5, the particles calcined at 300 ◦ C of crystalline particles within the viewing area. The crystal-
were identified as nanocrystalline anatase with the silver lite size of the prepared pure TiO2 was 5 nm. The crystallite
diffraction peak. With increasing calcination temperature to sizes calculated from XRD analysis were nearly the same as
500 ◦ C, the silver diffraction (1 1 1) disappeared and the par- those obtained from TEM analysis. In addition, the crystallite
ticles were identified as nanocrystalline anatase. Upon fur- size of Ag particles increased from 10 to 15 nm with increas-
ther increasing the temperature to 600 ◦ C, the rutile peak ing AgNO3 content, which agreed closely with the results
appeared. It is thought that the silver species was covered by reported by Liu et al. [14]. Fig. 6(c) shows the lattice fringe
the crystals of TiO2 that increased due to the high calcination of TiO2 –Ag 2 mmol nanoparticles. The lattice fringe, which
temperature. The crystallinity increases with increasing cal- has been used for phase determination show a value of 3.5 Å,
cination temperature because higher ordering in the structure corresponding to the lattice spacing of (1 0 1) plane in the
of titania particles makes X-ray peaks sharper and narrower. anatase phase.
The crystallite sizes of the particles prepared by different cal- Fig. 7 shows the EDS-pattern of TiO2 –Ag (2 mmol) par-
cination temperatures are given in Table 2. It can be seen that ticles in the HR-TEM mode from an area of 50 nm2 . The
the crystallite size of the anatase phase increased from 6 to presence of four X-ray lines associated with O K␣, Ag K␣,
37 nm as the calcination temperatures increased from 300 to Ti K␣, and Ni K␣ is evident. Given that the Ni K␣ line corre-
700 ◦ C. The size of rutile crystallites calcined at 600 ◦ C was sponds to the nickel grid used for TEM analysis, the results
47 nm and increased to 54 nm at 700 ◦ C. indicate that Ti, O and Ag are the constitutive elements of the
nanoparticles prepared by the sol–gel method with different
3.3. TEM analysis AgNO3 contents.

Fig. 6(a) and (b) show the TEM micrographs of the nano- 3.4. Photocatalytic degradation of p-nitrophenol
sized TiO2 –Ag particles prepared with different AgNO3 con-
tents and calcined at 300 ◦ C for 2 h in air. The nanoparticles The photocatalytic degradation of organic pollutants in
were spherical and had narrow size distribution. The crys- water generally follows a Langmuir–Hinshelwood mecha-
tallite size was determined (Table 1) by averaging the size nism [15,16], with the rate being proportional to the coverage
θ:
dC KC
r=− = kθ = k (1)
dt 1 + KC
where k is the true rate constant which is dependent upon
various parameters such as the mass of catalyst, the flux of

Table 2
Characterization of TiO2 –Ag 2 mmol particles calcined at different
temperatures
Calcination XRD Activityb
temperature (◦ C)
Structure Crystallite k (min−1 ) × 10−2
sizea (nm)
300 Anatase–Ag 6 3.4
500 Anatase 15 4.6
600 Anatase/rutile 22/47 2.8
700 Anatase/rutile 37/54 1.6
a Obtained by the Scherrer equation.
Fig. 5. X-ray diffraction patterns of TiO2 –Ag 2 mmol particles calcined at b Apparent first-order constants (k ) of p-nitrophenol after 60 min irradia-
different temperatures. tion.
 M.S. Lee et al. / Journal of Molecular Catalysis A: Chemical 242 (2005) 135–140 139

Fig. 6. TEM images of TiO2 –Ag particles prepared at different AgNO3 content: (a) TiO2 –Ag 1 mmol; (b) TiO2 –Ag 2 mmol, and (c) lattice fringe of TiO2 –Ag
2 mmol.

efficient photons, the coverage in oxygen, etc., K the adsorp- The photocatalytic activities of nanosize TiO2 –Ag and
tion constant, t the time, and C is the concentration of organic TiO2 particles for the decomposition of p-nitrophenol were
pollutant (in this case, p-nitrophenol). For the low initial con- examined and the results are shown in Fig. 8, as well as in
centrations of pollutants, the term KC in the denominator can Tables 1 and 2. The photocatalytic oxidation results were
be neglected with respect to unity and the photocatalytic oxi- compared with those obtained from control experiments,
dation rate approaches first order: involving no photocatalyst particles (data not shown). There
was no significant removal of p-nitrophenol in the system
dC
r=− = kKC = k C (2) when no photocatalyst was used in a control experiment.
dt The apparent first order rate constant for the control test
was more than one order of magnitude smaller than those
where k is the apparent rate constant of the pseudo-first order
obtained with the photocatalyst (presented in Fig. 8 and
kinetics. The integral form, C = f(t) of the rate equation is:
Tables 1 and 2). It is evident that the TiO2 –Ag particles had
C higher photoactivity than the pure TiO2 and that the photo-
ln = −k t (3) catalytic activity increased with the increase in the AgNO3
C0
content. More specifically, the photocatalytic activity of
where C0 is the initial concentration of p-nitrophenol. TiO2 –Ag (1 mmol) nanoparticles was about 60% higher than

Fig. 7. EDS-pattern of TiO2 –Ag 2 mmol particles prepared by sol–gel method using reduction agent.
140 M.S. Lee et al. / Journal of Molecular Catalysis A: Chemical 242 (2005) 135–140

synthesized was transparent and stable for several months.

Comparison of the diffuse reflectance spectra of the pure
TiO2 and TiO2 –Ag indicated that the band energy of TiO2
with silver addition did not change and the band-gap of the
particles was 3.1 eV. The major phase of all the prepared
TiO2 –Ag particles was an anatase structure regardless of the
AgNO3 content. When the AgNO3 content was 2 mmol/mol
of TiO2 , the major phase (1 1 1) of silver could be clearly
seen. The crystallite size of TiO2 particles calcined at 300 ◦ C
was 5–6 nm and that of Ag particles increased from 10 to
15 nm with increasing AgNO3 content. The high-resolution
transmission electron micrographs showed that TiO2 –Ag
nanoparticles possessed a spherical morphology and had
a narrow size distribution and the lattice fringe was 3.5 Å,
which corresponds to the lattice spacing of (1 0 1) plane
Fig. 8. Photodegradation of p-nitrophenol with pure TiO2 and TiO2 –Ag in the anatase phase. In addition, the presence of Ag in
nanoparticles. TiO2 –Ag particle prepared by sol–gel method using a
reduction agent resulted in higher photodegradation of
that of the pure titania (Table 1). Statistical analysis of the p-nitrophenol and the photocatalytic activity of TiO2 –Ag
apparent first order rate constants indicated that the higher increased with increasing AgNO3 content.
activity of TiO2 –Ag photocatalyst was significant at 95%
confidence intervals. From these results, it may be concluded
that the presence of Ag in TiO2 –Ag particles prepared by Acknowledgement
the sol–gel method using a reduction agent has a positive
effect on the photodegradation reaction of p-nitrophenol. Of This work was supported by the Post-doctoral Fellow-
course, it had been expected that Ag particles, deposited on ship Program of Korea Science & Engineering Foundation
the TiO2 surface, as electron–hole separation centers [7,17], (KOSEF).
would have played important roles in the photocatalytic
activities. This hypothesis was based on the fact that the
electron transfer from the TiO2 conduction band to silver References
particles at the interface is thermodynamically possible
[1] C. Kormann, D.W. Bahnemann, M.R. Hoffmann, J. Phys. Chem. 92
because the Fermi level of TiO2 is higher than that of silver
(1988) 5196.
metals [8]. Such phenomenon would be expected to lead to [2] S.D. Mo, L.B. Lin, J. Phys. Chem. Solids 55 (1994) 1309.
the formation of Schottky barrier at metal–semiconductor [3] K. Vinodgopal, D.E. Wynkoop, P.V. Kamat, Environ. Sci. Technol.
contact region, which could improve the charge separation 30 (1996) 1660.
and consequently the photocatalytic activity of TiO2 . [4] V. Vamathevan, R. Amal, D. Beydoun, G. Low, S. McEvoy, J. Pho-
tochem. Photobiol. A 148 (2002) 233.
The effect of calcination temperature on photocatalytic
[5] A. Sclafani, J.-M. Herrmann, J. Photochem. Photobiol. A 113 (1998)
activities of TiO2 –Ag nanoparticles is presented in Table 2. 181.
Anatase-type titanium dioxide generally exhibits a higher [6] V. Subramanian, E. Wolf, P. Kamat, J. Phys. Chem. B 105 (2001)
photocatalytic activity than the other types of titanium diox- 11439.
ide (e.g. rutile) with regards to the decomposition of organic [7] J.-M. Herrmann, J. Disdier, P. Pichat, J. Phys. Chem. 90 (1986)
6028.
pollutants by suppressing the electron–hole recombination
[8] K. Shiba, H. Hinode, M. Wakihara, React. Kinet. Catal. Lett. 64
[18]. As shown in Table 2, the photocatalytic activity of (1998) 281.
the TiO2 –Ag particles calcined at 500 ◦ C was the highest [9] I. Ilisz, Z. Laszlo, A. Dombi, Appl. Catal. A 180 (1999) 25.
and decreased with increasing calcination temperature. In [10] M. Sokmen, A. Ozkan, J. Photochem. Photobiol. A 147 (2002)
the case of titania particles calcined at 600 and 700 ◦ C, 77.
[11] C. He, Y. Yu, X.F. Hu, A. Larbot, Appl. Surf. Sci. 200 (2002) 239.
the anatase and rutile phase were combined and the amor-
[12] B.D. Cullity, Elements of X-Ray Diffraction, second ed., Addison-
phous phase was mixed with the anatase phase, leading to Wesley, Reading, MA, 1978, p. 102.
the reduced photocatalytic activities and decomposition of [13] K.M. Reddy, C.V.G. Reddy, S.V. Manorama, J. Solid State Chem.
p-nitrophenol. 158 (2001) 180.
[14] S.X. Liu, Z.P. Qu, X.W. Han, C.L. Sun, Catal. Today 93–95 (2004)
877.
[15] C.S. Turchi, D.F. Ollis, J. Catal. 122 (1990) 178.
4. Conclusions [16] G. Al-Sayyed, J.C. D’Oliveira, P. Pichat, J. Photochem. Photobiol.
A 58 (1991) 99.
TiO2 –Ag particles were successfully prepared by sol–gel [17] A. Henglein, J. Phys. Chem. 83 (1979) 2209.
process involving a reduction agent. The TiO2 –Ag sol [18] A. Sclafani, L. Palmisano, E. Davi, New J. Chem. 14 (1990) 265.