Journal of Colloid and Interface Science 267 (2003) 127131

www.elsevier.com/locate/jcis

Photocatalysis and hydrophilicity of doped TiO2 thin films

Young Cheol Lee,a Young Pyo Hong,a Ha Yong Lee,a Hoon Kim,a Young Jin Jung,a
Kyung Hyun Ko,a, Hyun Seok Jung,b and Kug Sun Hong b
a Department of Materials Science and Engineering, Ajou University, Suwon 442-749, South Korea
b School of Materials Science and Engineering, Seoul National University, Seoul 151-742, South Korea

Received 6 November 2002; accepted 29 May 2003

Abstract
TiO2 thin films were prepared using the dip-coating method with a polymeric sol including additives such as Al, W, and Al + W to
examine two major properties: photocatalysis and hydrophilicity. W-doped films showed the best photocatalytic efficiency, while Al-doped
film was poorer than undoped samples. However, good hydrophilicity in terms of saturation contact angle and surface conversion rate was
found in Al- and (Al + W)-mixed-doped films. It was found that deep electron-hole traps and high surface acidity of W-doped TiO2 thin
film were the major factors in high photocatalytic efficiency. In addition, low surface acidities of Al- and (Al + W)-doped films provided
better hydrophilicity than W-doped ones. However, the amount of [Ti3+ ] point defects on the surface was another major factor, probably the
most important, in getting the best hydrophilicity. Conclusively, it seemed that many parts of the photocatalysis mechanism depend more on
bulk-related properties than do those of hydrophilicity, which can be defined as an interfacial (surface) or near-surface-restricted process.
2003 Elsevier Inc. All rights reserved.
Keywords: TiO2 ; Photocatalysis; Hydrophilicity; Surface acidity; Electron trap; Ti3+ defect

1. Introduction
Owing to the nature of wide-gap semiconductors titanium
dioxide (TiO2 ) has been known not only as a photocatalyst
decomposing organic compounds [13], but also as a UVsensitive hydrophilic surface conversion agent [46]. Wang
et al. have reported that the transition between the hydrophobic and hydrophilic states could possibly be connected to
photoactive electronic transition across the energy gap, i.e.,
the conversion of Ti4+ sites into Ti3+ on the surface under
UV illumination. Therefore, in terms of UV activation, there
are common features between the photocatalytic mechanism
and hydrophilicity.
However, there has recently been some consensus that
the basic mechanisms of these two phenomena may not be
the same. According to T. Watanabe [7], the existence of
sodium ions in the TiO2 showed very different effects on
these photoinduced reactions, suggesting two different photoinduced defect reaction mechanisms on the surface. The
essential mechanism of photocatalysis could be explained in
terms of bulk property such as charge transfer efficiency of
* Corresponding author.

E-mail address: sputter07@hanmail.net (K.H. Ko).

0021-9797/$ see front matter 2003 Elsevier Inc. All rights reserved.
doi:10.1016/S0021-9797(03)00603-9

wide gap semiconductor. Therefore it seems that the photocatalysis of TiO2 is more dependent on bulk properties,
while the hydrophilicity of TiO2 is an inherently interfacial
property. Hydrophilicity phenomenon is limited to interface
between TiO2 surface(solid) and water(liquid). So far, there
have been many researches and industrial efforts that try to
understand this typical interfacial event, hydrophilicity, as a
form of photocatalytic behavior. Therefore, a new point of
view on hydrophilicity is needed.
To investigate this, in this work, TiO2 films were doped
with Al, W, and (Al + W) in order to make the photoactivity of the surface suitable for either photocatalysis or hydrophilic conversion under UV light.

2. Materials and methods

2.1. Preparation of TiO2 thin film
Transparent TiO2 thin films were prepared by a dipcoating process using a polymeric solution on soda-lime
glass and quartz substrates. The clear and stable polymeric
TiO2 coating solutions were synthesized from a mixture
of titanium tetraisopropoxide (TTIP), water-free ethanol

128

Y.C. Lee et al. / Journal of Colloid and Interface Science 267 (2003) 127131

(EtOH), deionized water, and nitric acid as a catalyst. Aluminum, tungsten, and (Al + W, 1:1) were then added to this
basic TiO2 solution as dopants. Aluminum tri-sec-butoxide
and tungstic acid were used as doping sources. Aluminum
tri-sec-butoxide was soluble in ethanol and tungstic acid
was soluble in aqueous alkali solution. At first, aluminum
tri-sec-butoxide was mixed with ethanol, and then titanium
tetra-isopropoxide (TTIP) was added into this solution. Finally, alkali aqueous solution including tungstic acid was
added into the solution.
Using this polymeric solution, TiO2 thin film was coated
onto soda-lime glass and quartz substrates by a dip-coating
process. For each pass, the TiO2 thin film was baked at
150 C under a pressure of about 1 Torr to prevent peeling
off, and finally annealed at 500 C.
Photocatalytic activity was evaluated by decomposition
of methylene blue, which has a major absorption peak at
664 nm. A TiO2 thin film of area 35 cm2 was loaded in a
10-ppm solution of methylene blue. TiO2 photocatalysis was
preceded by BLB UV light (1.0 mW/cm2) irradiation and
the reaction time was 2 h. The absorption spectra of decomposed methylene blue solution were measured by UVvis
spectroscopy, and photocatalytic efficiency was evaluated
by peak intensity of 664 nm. According to many other reports [811], the photocatalytic activity of TiO2 thin film on
soda-lime glass decreased remarkably because alkali ions
diffused from the glass substrate provided recombination
centers for electronhole pairs. To exclude the effect of in
situ alkali ions of soda-lime glass from those of other intentional dopants, the deposition of thin SiO2 layers has been a
common practice. However, an SiO2 layer was not employed
in this research and a quartz substrate was used instead for
comparative study.
The hydrophilicity of TiO2 thin film was evaluated in
terms of its resistance to conversion from the hydrophilic
to the hydrophobic state and also in terms of its recovery
rate from the hydrophobic to the hydrophilic state by monitoring the change of water contact angle. To accelerate the
surface state conversion, water was sprayed onto a fresh hydrophilic TiO2 thin film. After it was dried for half an hour in
a darkroom, water contact angles were measured. This completed the first cycle and successive cycles were carried out
immediately without refreshing the surface with UV illumination. To measure the recovery rate, surface fully converted
to the hydrophobic state was irradiated by BLB UV light
and changes of contact angles were recorded as a function
of time.
The crystalline phase of dip-coated TiO2 thin films
was measured by an X-ray diffractometer (M18XHF-SRA,
McScience, Japan) with CuK radiation. Optical band
structures of undoped and doped TiO2 thin films were measured by PL spectra using a 325 nm HeCd laser source.
Changes of surface morphology as dopant were observed
using field emission scanning electron microscopy. To measure surface acidity, spherical silica particles were at first
prepared by precipitation according to the Stber et al.

method [17]. The silica spheres were then coated with TiO2
polymeric solution followed by annealing at 500 C for
crystallization. Lewis surface acidities were measured by
titrating 0.5 g of TiO2 -coated silica spheres suspended in
benzene with 0.1 N n-butylamine benzene solution, using
methyl red as an indicator. The concentrations of [Ti3+ ] and
hydroxyl ions on the TiO2 surface, known as major factors
for hydrophilicity, were measured by X-ray photoelectron
spectroscopy (XPS).

3. Results and discussion

Figure 1 shows the X-ray diffraction pattern for crystallized TiO2 thin film after 500 C annealing. The crystalline
phase of all dip-coated TiO2 thin films was anatase. After
some cycles of the conversion resistance test of the TiO2
thin film coated on soda-lime glass, the water contact angles were saturated (Fig. 2a). In the case of undoped and
Al-doped TiO2 thin film, the saturated contact angle was

Fig. 1. XRD patterns of TiO2 thin film: (a) undoped, (b) Al-doped,
(c) W-doped, (d) (Al + W)-doped.

Fig. 2. Change of water contact angle of TiO2 thin film on soda-lime glass:
(a) after wetdry cycle, (b) under irradiation by UV light (after saturation
of water contact angle).

Y.C. Lee et al. / Journal of Colloid and Interface Science 267 (2003) 127131

about 35 , while W-doped TiO2 thin film showed the lowest resistance (highest saturation = 50 ). However, when
Al and W were co-doped, TiO2 thin film had the highest resistance (saturation = 25 ). Figure 2b shows the restoring
of contact angle under UV irradiation. The recovery rates for
undoped, Al-doped, and W-doped TiO2 thin films were relatively sluggish, while (Al + W) co-doped film restored the
original hydrophilic state within approximately 3 h. In the
case of quartz substrate, the water contact angle of postannealed TiO2 thin film was not zero (0 ). However, the trends
in hydrophilic/hydrophobic conversion were similar. It was
thought that alkali ions from substrate did not have any significant influence on the hydrophilic property (Fig. 3).
Photocatalytic activity was evaluated by measuring the
rate of decomposition of methylene blue; the intensity of the
absorption spectra at = 664 nm corresponding to the color

Fig. 3. Change of water contact angle of TiO2 thin film on quartz substrate:
(a) after wetdry cycle, (b) under irradiation by UV light (after saturation
of water contact angle).

Fig. 4. UVvis spectra of methylene blue after photocatalytic reaction using

TiO2 thin film on soda-lime glass.

129

of methylene blue decreased with photocatalytic decomposition by TiO2 thin film [19] (Fig. 4). W- and Al-doped TiO2
thin film showed the highest and lowest photocatalytic efficiency, respectively. In terms of efficiency, the effects of Aland W-doping were reversed from those of hydrophilicity.
Also, (Al + W) co-doped samples showed efficiency similar to that of undoped film. In addition, it was found that the
photocatalytic behavior of TiO2 thin films coated on quartz
substrate was also similar to that of films deposited on sodalime glass (Fig. 5). Therefore, it seems that either the effects
of W and Al could overcome those of alkali ions or the
amount of alkali ions from substrate might be much smaller
than expected.
There can be many factors affecting hydrophilicity and
the photocatalytic process, for example, electronic traps for
electrons or holes, surface acidity, and the concentration of
[Ti3+ ] on surface [1215]. Since tungsten is a heavy transition metal, it might be expected that doped tungsten ions

Fig. 5. UVvis spectra of methylene blue after photocatalytic reaction using

TiO2 thin film on quartz substrate.

Fig. 6. Photoluminescence spectra of TiO2 thin films: (a) undoped,

(b) Al-doped, (c) W-doped, (d) (Al + W)-doped.

130

Y.C. Lee et al. / Journal of Colloid and Interface Science 267 (2003) 127131

would form deep level states somewhere in the band gap

of titanium oxide. The photoluminescence (PL) spectra at
room temperature for undoped and doped TiO2 thin films
are shown in Fig. 6. While all samples showed PL peaks at
corresponding to the band-to-band transition especially, only
W-doped TiO2 thin film showed a additional PL band around
410 nm, possibly related to level (W)-to-band transition. The
intensity of PL could be an indication of a larger concentration of nonradiative traps for electrons and holes. Because
all other films including (Al + W)-doped samples showed
poorer photocatalytic efficiency, but better hydrophilic properties than W-doped films, the existence of electronic traps in
the bulk of films could play a major role in the photocatalytic
process, but not in hydrophilicity. Rather, it seemed that the
hydrophilic property was mostly determined by change of
surface state within a few topmost layers of films. It was
suggested that the isoelectric point would greatly influence
the adsorption of organic substrates onto TiO2 surface [15].
Furthermore, hydrophilicity was created by the formation
of a layer of adsorbed hydroxyl(OH) and multiple layers
of water molecules [5,13]. Therefore, the more acidic the
surface, the fewer hydroxyl ions could be adsorbed [16].
Consequently, if the surface is acidic, hydrophilicity will
be poor due to the lack of a preadsorbed layer of hydroxyl
ions. So, if the isoelectric point, in other words, the surface acidity, is changed by impurity doping, there can be
significant rise and fall in the photocatalytic activity and hydrophilicity of TiO2 thin film. Since, according to Li [16],
the isoelectric point of W-doped TiO2 decreases from that of

the undoped state, it seemed that the photocatalytic activity

of W-doped films can be enhanced due to more acidic surface. In Lewis acidity measurements, the color of methyl red
indicator would be changed from red within pH 4.2 to yellow within pH 6.2. In this work, Al-doped TiO2 thin film
was chosen as a standard zero of surface acidity. Compared
with Al-doped TiO2 , undoped and (Al + W)-doped TiO2 thin
films were titrated at 0.3 and 0.2 ml of n-butylamine, which
indicated 0.06 and 0.04 mmol/g of surface acidity, respectively, while W-doped film had the highest surface acidity
at 0.42 mmol/g. Therefore, it was concluded that the very
high surface acidity of W-doped TiO2 thin films could influence high photocatalytic activity. As mentioned above, if the
hydrophilic state favors low surface acidity, then this might
easily explain how the Al-doped and (Al + W)-doped TiO2
thin films had a lower saturated contact angle than that for
W-doped thin films (Fig. 2a). However, when the discrepancy between the surface acidities and saturation angles of
Al- and (Al + W)-doped films was noticed, it seemed that
surface acidity was an important factor in hydrophilicity, but
not the sole factor.
Figure 7 shows the Ti2p XPS spectra for TiO2 thin
films after irradiation by UV for 3 h. All the samples show
shoulder peaks around 457 eV, which indicates the existence of Ti3+ . The concentration of [Ti3+ ] on TiO2 surface is a major factor influencing hydrophilicity [12,13,18].
The concentrations of [Ti3+ ] in undoped, Al-doped, and
W-doped TiO2 thin films were 58.63%, 46.45%, and 57.19%
(Ti3+ /Ti4+ ), respectively. However, the XPS spectra for

Fig. 7. Ti2p XPS spectra of TiO2 thin films: (a) undoped, (b) Al-doped, (c) W-doped, (d) (Al + W)-doped.

Y.C. Lee et al. / Journal of Colloid and Interface Science 267 (2003) 127131

mixed-doped TiO2 thin film shows the highest concentration

of [Ti3+ ](63.81%). The same trends were observed in both
films coated on soda-lime glass and quartz. Conclusively, it
was considered that the low saturated contact angle and high
recovery rate to the hydrophilic state of mixed-doped TiO2
thin film were caused by a high concentration of [Ti3+ ] on
the surface, as well as low surface acidity.

References
[1]
[2]
[3]
[4]
[5]
[6]
[7]

4. Conclusions
[8]
[9]

Photocatalysis and hydrophilicity are two distinct properties of TiO2 based on the same photoinduced electronic
transition. However, the detailed material properties required
for enhanced efficiency are different from each other. For
enhanced photocatalysis, deep electronic traps and high surface acidity are more vigorously needed to lengthen lifetime
of photoexcited electrons and holes and to ensure better adsorption of organic substances onto a surface. Meanwhile,
low surface acidity and, most of all, a large quantity of [Ti+3 ]
is essential for hydrophilic surface conversion. These are
probably related to the fact that photocatalysis is more likely
to be sensitive to bulk properties, while hydrophilicity is a
near-surface-restricted and limited interfacial phenomenon.

131

[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]

R.M. Roman, N.C. Martinez, Catal. Today 40 (1998) 353.

D.S. Muggli, J.T. McCue, J.L. Falconer, J. Catal. 173 (1998) 470.
U. Stafford, K.A. Gray, P.V. Kamat, J. Catal. 167 (1997) 25.
R. Wang, K. Hashimoto, A. Fujishima, M. Cjikuni, E. Kojima, A. Kitamura, M. Shimohigoshi, T. Watanabe, Nature 388 (1997) 431.
R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, A. Kitamura,
M. Shimohigoshi, T. Watanabe, Adv. Mater. 10 (2) (1998) 135.
J. Yu, X. Zhao, Mater. Res. Bull. 36 (2001) 97.
T. Watanabe, S. Fukayama, M. Miyauchi, A. Fujishima, K. Hashimoto,
J. SolGel Sci. Technol. 19 (2000) 71.
A. Heller, Acc. Chem. Res. 28 (1995) 503.
Y. Kikuchi, K. Sunada, T. Iyoda, K. Hashimoto, A. Fujishima, J. Photochem. Photobiol. A 106 (1997) 51.
T. Watanabe, S. Fukayama, M. Miyauchi, A. Fujishima, K. Hashimoto,
J. SolGel Sci. Technol. 19 (2000) 71.
Y. Paz, A. Heller, J. Mater. Res. 12 (1997) 2759.
H.Y. Lee, Y.H. Park, K.H. Ko, Langmuir 16 (18) (2000) 7289.
N. Sakai, R. Wang, A. Fujishima, T. Watanabe, K. Hashimoto, Langmuir 14 (1998) 5918.
J. Yu, X. Zhao, Q. Zhao, Mater. Chem. Phys. 69 (2001) 25.
K.Y. Song, M.K. Park, Y.T. Kwon, H.W. Lee, W.J. Chung, W.I. Lee,
Chem. Mater. 13 (2001) 2349.
X.Z. Li, F.B. Li, C.L. Yang, W.K. Ge, J. Photochem. Photobiol. A
Chem. 141 (2001) 209.
W. Stber, A. Fink, J. Colloid Interface Sci. 26 (1968) 62.
X.P. Wang, Y. Yu, X.F. Hu, L. Gao, Thin Solid Films 371 (2000) 148.
S. Lakshmi, R. Renganathan, S. Fujita, J. Photochem. Photobiol. A
Chem. 88 (1995) 163.