Full text of DOI:10.1111/j.1151-2916.2003.tb03523.x

J. Am. Ceram. Soc., 86 [9] 1605–608 (2003)
journal
Anatase Sol Prepared from Peroxotitanium Complex Aqueous Solution
Containing Niobium or Vanadium
Hiromichi Ichinose, Masanori Taira, Sachiko Furuta, and Hiroaki Katsuki*
Ceramics Research Laboratory, Arita, Nishimatsuura, Saga 844-0024, Japan
Niobium- or vanadium-doped anatase sols were prepared by
hydrothermal treatment of 0.1 mol/dm3 peroxotitanium com-
plex aqueous solutions dissolving 0–10 mol% niobium or
vanadium at 100°C for 8 h. Niobium-doping caused the
increase of lattice constants of anatase and the shape change of
anatase crystal from spindle-like to cubic-like structure, but no
change of the optical absorbance. Vanadium-doping caused
the decrease of lattice constant of c-axis, the miniaturization of
anatase crystal and the increase of optical absorbance at the
wavelength from 350–700 nm.
concentration of ammonia for the crystallization to anatase was
also determined.
II. Experimental Procedure
Titanium powder (Ͼ98% of purity, Wako Pure Chemical
Industries, Osaka, Japan), niobium powder (99.5% of purity, Wako
Pure Chemical Industries), and vanadium turnings (99.5% of
purity, Wako Pure Chemical Industries) were used as starting
materials. First, 0.01 mol titanium powder was dropped in the
mixed solution of ammonia water (3 mol/dm3) 15 mL and H2O2
solution (ϳ30 mass%) 65 mL, and dissolved completely over
several hours. Next, ammonia concentration in the solution was
I. Introduction
IO2 is useful for solar cells,1 gas sensors,2,3 photocatalysts,4
T
and more. In particular, TiO2 film has been studied extensively
by the sol–gel method5 and solid-state reaction.6 In recent years,
much importance has been given to the studies of niobium- or
7,8
vanadium-doped TiO2 prepared by the coprecipitation method
with organic metals and chlorides as starting materials. On the
other hand, it is well-known that H2O2 forms peroxotitanium
complex with titanium.9–11 It was also reported that anatase sol,
containing well-dispersed anatase crystallite surface modified with
peroxo groups, was obtained by autoclaving peroxotitanium com-
plex aqueous solution or peroxotitanium hydrate suspension at
100°C for 6 h.12–14 The peroxotitanium complex aqueous solution
can be prepared by adding aqueous H2O2 into freshly prepared and
well-washed titanic acid12 or peroxotitanium hydrate.14 The per-
oxotitanium complex aqueous solution and the anatase sol contain
some ammonia as a stabilizer and are neutral or weak base in pH.
If a chemically homogeneous mixture of the peroxotitanium
complex aqueous solution and transition metal ions can be ob-
tained, it may become possible to prepare metal-doped anatase sol
by the hydrothermal reaction at low temperature, and it is expected
that the application will spread as the form of thin films. However,
the homogeneous mixture cannot be performed because insoluble
precipitate is deposited by mixing transition metal ions to the
peroxotitanium complex aqueous solution. This phenomenon is
explained as polymerization by the bridge construction reaction
between peroxotitanium complexes through transition metal ions
in an instant. Because niobium or vanadium are known to form
peroxo-complexes,15 the mixture is expected to mix homoge-
neously with peroxotitanium complex aqueous solution.
Fig. 1. FT-IR spectra of the liquids prepared by heating peroxotitanium
complex aqueous solution at 100°C for 8 h with different ammonia
concentrations.
In this study, the peroxotitanium complex aqueous solution
containing a peroxo-complex of niobium or vanadium was pre-
pared, and then niobium- or vanadium-doped anatase sols were
synthesized by hydrothermal treatment. Moreover, the optimal
Y. Ohya—contributing editor
Fig. 2. XRD patterns of the liquids prepared by heating peroxotitanium
complex aqueous solution at 100°C for 8 h with different ammonia
concentrations.
Manuscript No. 186443. Received February 3, 2003; approved April 1, 2003.
*Member, American Ceramic Society.
1605

1606
Communications of the American Ceramic Society
Vol. 86, No. 9
Fig. 3. TEM photographs of niobium- or vanadium-doped anatase sol: (a) nondoped, (b) Nb 2 mol%, (c) Nb 5 mol%, (d) Nb 10 mol%, (e) V 2 mol%, (f)
V 5 mol%, (g) V 10mol%.
reduced by following methods; adding Hϩ-type cation-exchange
resin (Amberlite IR118, Organo, Japan) to the dissolved titanium
solution while stirring slowly until pH is set to ϳ5, separating the
resin, and carrying out ultrasonic treatment to decompose excess
H2O2 until pH increased to ϳ7. This ion-exchange process was
repeated three to four times, and then a transparent peroxotitanium
complex aqueous solution of 0.1 mol/dm3 was obtained. The
concentration of ammonia was adjusted to 0.0063, 0.014, 0.021,

September 2003
Communications of the American Ceramic Society
1607
mol/dm3, only anatase phase was detected. Therefore, it is neces-
sary to make ammonia concentration into Ͻ0.014 mol/dm3 to
crystallize smoothly to anatase phase. However, when ammonia
concentration was decreased to 0.0063 mol/dm3 or less, the
peroxotitanium complex solution became too high in viscosity to
obtain a stable anatase sol and finally solidified as jelly-like.
Consequently, it turned out that ammonia concentration in the
range of 0.0063–0.014 mol/dm3 is suitable for the preparation of
anatase sol. In the next experiments, ammonia concentration was
performed as 0.01 mol/dm3.
(2) Properties of Niobium- or Vanadium-Doped Anatase Sol
The peroxotitanium complex aqueous solutions dissolving nio-
bium or vanadium crystallized only to anatase phase by heating at
100°C for 8 h. The anatase particles dispersed homogeneously in
the solvent. These translucent anatase sols were stable for a long
time. TEM photographs of anatase particles are shown in Fig. 3. In
the case of nondoping, there were spindle-like and arrowhead-like
anatase crystals. The aspect ratio of anatase particle decreased
gradually with the addition of niobium, and the shape of crystals
turned out to be cubic at 10 mol% niobium. According to electron
diffraction analysis, the square faces of crystals were anatase
(101). The crystal size and the aspect ratio of anatase crystals
decreased gradually with the addition of vanadium. The size was
5–8 nm at 10 mol% vanadium. The specific surface area of
nondoped sol was 165 m2/g, and decreased gradually with the
addition of niobium. However, the specific surface area increased
dramatically up to 210 m2/g even at 2 mol% vanadium, and it
didn’t increase any more with further additions.
Fig. 4. Lattice constants of niobium- or vanadium-doped anatase.
Lattice constants of niobium- or vanadium-doped anatase are
shown in Fig. 4. Although the lattice constants of nondoped
anatase closed to the reference values, they increased gradually
with the addition of niobium. In the case of the addition of
vanadium, the lattice constant of the c-axis decreased with the
addition, but that of the a-axis did not change entirely. UV-visible
light reflective spectra of dried vanadium-doped anatase sols are
shown in Fig. 5. The optical absorption increased with the addition
of vanadium at the wavelength from 350 nm to a visible light
region, but there was no change with the niobium addition. It has
been reported that Nb2O5 forms an interstitial solid solution Ͼ0.6
mol% at 700°C with the increase of lattice constants.17 It is
suggested in this study that niobium formed an interstitial solid
solution in anatase. It was also suggested that small V5ϩ ion (0.054
nm) substituted for the Ti4ϩ ion (0.061 nm), and the lattice
constant was reduced by the size difference. However, the amounts
of substitutions could not be clarified.
Fig. 5. Optical absorbance spectra of dried vanadium-doped anatase sol.
0.028, 0.036, and 0.043 mol/dm3. The translucent anatase sol was
obtained by heating the prepared solutions in an airtight container
at 100°C for 8 h. Niobium- or vanadium-doped anatase sols
(dopant ratio: 0, 2, 5, and 10 mol%) were prepared by the same
method, but the ammonia concentration was 0.01mol/dm3.
The anatase sols were characterized by powder X-ray diffrac-
tometry (XRD; Geigerflex RAD-B, Rigaku, Tokyo, Japan), BET
specific surface meter (Auto-sorb 1, Yuasa Ionics, Osaka, Japan),
FT-IR spectrophotometry (Model FT/IR-5300, Jasco, Tokyo, Ja-
pan), UV-visible spectrophotometry (Model U-3100, Hitachi,
Tokyo, Japan), transmission electron microscopy (TEM; JEM-
2010, JEOL, Tokyo, Japan). UV-visible light reflection spectra
were measured for the films, which were prepared with the powder
and little water so that it might be 2 mg/cm2 on a glass. The
concentration of ammonia was measured by the colorimetry of
indophenol with the light wavelength of 630 nm.
IV. Conclusions
We synthesized peroxotianium complex aqueous solutions con-
taining niobium or vanadium using the metals as starting materials,
and niobium- or vanadium-doped anatase sols were obtained by
hydrothermal treatment at 100°C. Moreover, the optimal concen-
tration of ammonia for the crystallization to anatase phase turned
up. It should be emphasized that the peroxotitanium complex
aqueous solution is highly stable at a pH of ϳ7, and therefore has
potential applications in inexpensive and convenient syntheses of
metal-doped anatase sols for industrially important products.
III. Results and Discussion
(1) Optimal Ammonia Concentration for Crystallization
FT-IR spectra of the liquids prepared by heating peroxotitanium
complex aqueous solution at 100°C for 8 h with different ammonia
concentrations are shown in Fig. 1. In the case of 0.021 mol/dm3
and more of ammonia concentration, the infrared absorption
resulting from free peroxide (O-O, bond order ϭ 1) stretching
References
1B. O’Regan and M. Gra¨tzel, “A Low-Cost, High-Efficiency Solar Cell Based on
Dye-Sensitized Colloidal TiO2 Films,” Nature (London), 353 [24] 737–40 (1991).
2E. Traversa, M. L. Vona, S. Licoccia, M. Sacerdoti, M. C. Carotta, L. Crema, and
G. Martinelli, “Sol-Gel Processed TiO2-Based Nano-Sized Powders for Use in
Thick-Film Gas Sensors for Atmospheric Pollutant Monitoring,” J. Sol-Gel Sci. Tech.,
22 [1/2] 167–79 (2001).
mode was detected at 890 cmϪ1 16
It is suggested that the
.
peroxotitanium complex remained after heating by a stabilizing
effect of ammonia. The broad peaks at 3200–3400 cmϪ1 and a
clear peak at 1630 cmϪ1 are attributed to water molecules. A peak
at 1400 cmϪ1 is due to the stretching vibration of N-H bonds in
ammonia. XRD patterns are shown in Fig. 2. In the case of Ͻ0.014
3H. Tang, K. Prasad, R. Sanjine´s, and F. Le´vy, “TiO2 Anatase Thin Films as Gas
Sensors,” Sens. Actuators B., 26–27, 71–75 (1995).
4G. Irick, “Determination of the Photocatalytic Activities of Titanium Dioxides and
Other White Pigments,” J. Appl. Polym. Sci., 16, 2387–95 (1972).

1608
Communications of the American Ceramic Society
Vol. 86, No. 9
5M. Gra¨tzel, “Sol-Gel Processed TiO2 Films for Photovoltaic Applications,” J.
12H. Ichinose, M. Terasaki, and H. Katsuki, “Synthesis of Peroxo-Modified
Anatase Sol from Peroxo Titanic Acid Solution,” J. Ceram. Soc. Jpn., 104 [8] 715–18
(1996).
Sol-Gel Sci. Tech., 22 [1/2] 7–13 (2001).
6T. Carlson and G. L. Griffin, “Photooxidation of Methanol Using V2O5/TiO2 and
MoO3/TiO2 Surface Oxide Monolayer Catalysts,” J. Phys. Chem., 90 [22] 5896–900
(1986).
13H. Ichinose, A. Kawahara, and H. Katsuki, “Properties of Anatase Films for
Photocatalyst from Peroxotitanic Acid Solution and Peroxo-Modified Anatase Sol,”
J. Ceram. Soc. Jpn., 104 [10] 914–17 (1996).
7J. Kiwi, J. T. Suss, and S. Szapiro, “EPR Spectra of Niobium-Doped TiO2 and
Implications for Water Photocleavage Processes,” Chem. Phys. Lett., 106 [1,2]
135–38 (1984).
14H. Ichinose, M. Terasaki, and H. Katsuki, “Properties Peroxotitanium Acid
Solution and Peroxo-Modified Anatase Sol Derived from Peroxotitanium Hydrate,” J.
Sol-Gel Sci. Tech. 22 [1/2] 33–40 (2001).
8S. T. Martin, C. L. Morrison, and M. R. Hoffmann, “Photochemical Mechanism
of Size-Quantized Vanadium-Doped TiO2 Particles,” J. Phys. Chem. 98 [51]
13695–704 (1994).
15W. P. Griffith, “Studies on Transition-Metal Peroxy-Complexes, Part III,”
J. Chem. Soc., 1964, 5248–53 (1964).
16D. Klissurski, K. Hsdjiivanov, M. Kantchva, and L. Gyurova, “Study of
Peroxide-Modified Titanium Dioxide (Anatase),” J. Chem. Soc. Faraday Trans., 86
[2] 385–88 (1990).
9J. Mu¨hlebach, K. Mu¨ller, and G. Schwarzenbach, “The Peroxo Complexes of
Titanium,” Inorg. Chem., 9 [11] 2381–90 (1970).
10D. Schwarzenbach, “The Structure of a Chelated Dinuclear Peroxytitanium(IV),”
Inorg. Chem., 9 [11] 2391–97 (1970).
17S. Hishita, K. Koumoto, and H. Yanagida, “Inhibition of Anatase-Rutile
Transformation Due to Nb2O5 Addition: The Relation between the Solid Solution
Formation and the Inhibitory Effect of Nb2O5,” Yogyo-Kyokaishi, 91 [9] 399–403
11S. Komarneni, R. Roy, and E. Breval, “A New Method of Making Titania Gels
and Their Microstructure,” J. Am. Ceram. Soc., 68 [2] C41–C42 (1985).
(1983).
Ⅺ