Preparation, Characterization, and Photoactivity of Polycrystalline Nanostructured TiO2 Catalysts

Article abstract of DOI:10.1021/jp0312924

Various preparations of nanostructured TIO₂ starting from Ti(iso-OC₃H₇)₄ or TiCl₄ are reported. The samples were characterized by X-ray diffractometry, specific surface area and porosity determinations, scanning and transmission electron microscopy, and diffuse reflectance spectroscopy. 4-Nitrophenol photodegradation in aqueous medium was employed as a probe reaction to test the photoactivity of the catalysts. The photoactivity of some samples derived from Ti(iso-OC ₃H₇)₄ was found comparable with that of commercial powders. Calcination after the hydrolysis process was necessary to achieve crystallization of the particles before using them as photocatalysts for the reaction studied. The samples deriving from TiCl₄ were the most photoactive among the home-prepared catalysts, and neither filtration nor calcination was needed to obtain a highly photoactive anatase phase.

Full text of DOI:10.1021/jp0312924

J. Phys. Chem. B 2004, 108, 3303-3310
3303
Preparation, Characterization, and Photoactivity of Polycrystalline Nanostructured TiO2
Catalysts
†
†
†
†
Maurizio Addamo, Vincenzo Augugliaro, Agatino Di Paola, Elisa Garc ´ı a-L o´ pez,
†
†
‡,§
,†
Vittorio Loddo, Giuseppe Marc `ı , Raffaele Molinari, Leonardo Palmisano,* and
Mario Schiavello†
Dipartimento di Ingegneria Chimica dei Processi e dei Materiali, UniVersit a` di Palermo, Viale delle Scienze,
9
0128 Palermo, Italy, and Dipartimento di Ingegneria Chimica e dei Materiali, UniVersit a` della Calabria,
Via P. Bucci, 87030 Rende (CS), Italy.
ReceiVed: December 8, 2003
2 3 7 4 4
Various preparations of nanostructured TiO starting from Ti(iso-OC H ) or TiCl are reported. The samples
were characterized by X-ray diffractometry, specific surface area and porosity determinations, scanning and
transmission electron microscopy, and diffuse reflectance spectroscopy. 4-Nitrophenol photodegradation in
aqueous medium was employed as a probe reaction to test the photoactivity of the catalysts. The photoactivity
of some samples derived from Ti(iso-OC
Calcination after the hydrolysis process was necessary to achieve crystallization of the particles before using
them as photocatalysts for the reaction studied. The samples deriving from TiCl were the most photoactive
3 7 4
H ) was found comparable with that of commercial powders.
4
among the home-prepared catalysts, and neither filtration nor calcination was needed to obtain a highly
photoactive anatase phase.
medium,2
0-24
although the existence of stable nanoparticles is
Introduction
known to be not easy, due to their tendency to agglomerate.
Even if in most cases the samples used did not consist of discrete
nanoparticles but only of nanostructured particles, this kind of
photocatalysts is reported to show a high photoactivity compared
Heterogeneous photocatalysis has been found effective for
the achievement of the photooxidation of many organic pollut-
ants present in liquid effluents1-4 or in air as VOCs. This
method offers some advantages compared to the traditional ones
because it is not selective; the reactions usually occur at room
temperature and atmospheric pressure, and UV radiations of
low intensity are needed.
5,6
23-28
with that of many commercial samples.
On the other hand,
the presence of very small particles cannot be excluded when
the photoreactivity experiments are carried out in mixed liquid-
solid systems. The use of submicrometric particles is possible
TiO2 is the most used photocatalyst, due to its (photo)stability
2
9
today by means of membrane photoreactors.
7
and low cost. The preparation methods are reported to influence
In this paper different TiO2 samples, consisting of nanostruc-
tured particles, were prepared by taking into account the up-
significantly morphological, structural, surface physicochemical,
and electronic properties that are related to the extent of
2
7-31
to-date literature
but changing the starting precursors,
8
,9
photoactivity. Unfortunately, near-UV photons (band gaps
equal to 3.2 and 3.0 eV for anatase and rutile phases,
respectively) are needed for the excitation of TiO2. This
drawback can be overcome by using not stoichiometric or doped
TiO2 because a bathochromic shift in the light absorption is
generally observed in those cases. Nevertheless this phenomenon
does not give rise always to an enhancement of the photoactivity
of the material.1
Ti(iso-OC3H7)4 or TiCl4, and/or modifying some operative
parameters such as molar ratio of the reagents, heating and
calcination temperatures, and times.
Bulk and surface characterizations of the powders were
carried out by means of X-ray diffractometry (XRD), determi-
nation of the BET specific surface areas, porosity measurements,
scanning electron microscopy (SEM), transmission electron
microscopy (TEM), and diffuse reflectance spectroscopy (DRS)
techniques. 4-Nitrophenol photodegradation was employed as
a probe reaction in order to test the photoactivity of the home-
prepared samples compared with that of some commercial
samples.
0-13
Recently, many studies have been reported on the preparation
and characterization of various nanostructured semiconductors
because it has been found that the nanoparticles exhibit special
photochemical characteristics.14-19 In particular, the band gap
of the nanoparticles increases with the decrease of the size and
other important properties such as optical and physical absorp-
tion and luminescence emission undergo drastic changes.
TiO2 nanoparticles have been recently employed in photo-
catalysis for photodegradation of noxious species in aqueous
Experimental Section
Catalysts Preparation. Titanium isopropoxide, titanium
tetrachloride, ethanol, and 2-propanol were obtained from Carlo
Erba. All chemicals were used as received. The commercial TiO2
samples were Degussa P25, Merck and Tioxide.
The preparation methods for various nanostructured TiO2
samples were the following:
*
Corresponding author. Tel: +39-091-6567246. Fax: +39-091-
6
567280. E-mail: palmisan@dicpm.unipa.it.
†
Universit a` di Palermo.
Universit a` della Calabria.
Tel: +39-0984-492090. Fax: +39-0984-492058. E-mail: r.molinari@
‡
§
Preparation 1. Titanium hydroxide was prepared by hydroly-
sis of titanium isopropoxide carried out at room temperature
unical.it.
1
0.1021/jp0312924 CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/11/2004

3304 J. Phys. Chem. B, Vol. 108, No. 10, 2004
Addamo et al.
for 1 h in a mixture of water and 2-propanol (molar ratios Ti/
523 K prior to the measurement, using a Sorptomatic 1900 Carlo
Erba Instrument.
2-propanol/H2O equal to 1:22:5; volume (mL) ratios 10:60:3):
XPS Analysis. The X-ray photoelectron spectroscopy analyses
were performed by means of a VG Microtech ESCA 3000
Multilab, equipped with a dual Mg/Al anode. The working
Ti(iso-OC H ) + 4H O f Ti(OH) + 4C H OH
3
7 4
2
4
3
7
The precipitate was then filtered, and different portions were
successively calcined for 24 h at 573, 673, or 773 K. The
following code was used for the samples: P1-TiO2(i-Pr, 573,
73, or 773) where i-Pr indicates 2-propanol and 573, 673, or
73 the calcination temperature in Kelvin.
Preparation 2. A highly dispersed stable colloidal suspension
of titanium hydroxide was prepared by hydrolysis of titanium
isopropoxide carried out at 333 K for 5 h in a mixture of
-6
pressure was around 1 × 10 Pa. The samples were analyzed
as pellets after being ground in a mortar.
Scanning Electron Microscopy. SEM observations were
obtained using a model Philips XL30 ESEM microscope,
operating at 25 kV on specimens upon which a thin layer of
gold had been evaporated. For some samples a drop of
suspension on the stab was directly dried at room temperature
or sprayed and dried after dilution (1:10 v/v).
Transmission Electron Microscopy. The particle size of the
powders has been determined by using a JEOL JEM 12-20
transmission electron microscope (TEM) and the micrographs
were recorded at 120 kV. The samples were prepared by
dispersing the powders in water and depositing a drop of
suspension onto a thin Formavar film supported on a Cu grid.
Diffuse Reflectance Spectroscopy. Visible-ultraviolet spectra
were obtained by diffuse reflectance spectroscopy by using a
Shimadzu UV-2401 PC instrument. BaSO4 was the reference
sample and the spectra were recorded in the range 200-800
nm.
6
7
ethanol, strong acid, and water. The strong acids used were HCl
+
(
37%) or HNO3 (65%), (molar ratios Ti/ethanol/H /H2O equal
to 1:20:0.5:200; volume (mL) ratios 10:40:2:130 using HCl or
0:40:1:130 using HNO3). The suspension was dried at 323 K
and the obtained solid was washed with water. Successively
different fractions were calcined at 573, 673, or 773 K for 24
h. The following codes were used for the samples: P2-TiO2(Cl,
73, 673, or 773) and P2-TiO2(N, 573, 673, or 773), where Cl
and N indicate HCl and HNO3, respectively.
Preparation 3. Titanium hydroxide was prepared by hydroly-
sis of titanium isopropoxide in a mixture of ethanol and water
molar ratios Ti/ethanol/H2O equal to 1:5:200; volume (mL)
ratios 10:11:130) carried out at 333 K for 5 h. The suspension
was then dried at 323 K, and successively different fractions
were calcined for 24 h at 573, 673, or 773 K. The following
code was used for the samples: P3-TiO2(Et, 573, 673, or 773)
where Et indicates ethanol.
1
5
(
PhotoreactiVity Experiments. A Pyrex batch photoreactor of
cylindrical shape containing 0.5 liters of aqueous suspension
was used. The photoreactor was provided with ports in its upper
section for the inlet and outlet of gases, for sampling, and for
pH and temperature measurements. A 125 W medium-pressure
Hg lamp (Helios Italquartz, Italy) was immersed within the
Preparation 4. A sol was prepared by hydrolysis of titanium
tetrachloride at room temperature for 24 h (molar ratios Ti/
water equal to 1:60; volume (mL) ratios 10:100). The resulting
clear solution was boiled for ca. 3 h in order to form a solid
and to eliminate at the same time most of chloride ions as HCl(g).
In some cases, the boiling times were 1, 12, or 24 h. The code
used for these samples was P4-TiO2(ex TiCl4, h) where h
indicates the boiling time. The dispersions containing very small
particles were directly employed for the photoreactivity experi-
ments. Some dispersions were tested after 7, 30, or 90 days of
aging. The characterization of all samples was carried out on
the solids obtained by drying the dispersion at ca. 323 K.
Sample Characterization. X-ray Diffraction. XRD patterns
of the powders were recorded at room temperature by a Philips
powder diffractometer using the Cu KR radiation and a 2θ scan
rate of 2°/min. X-ray diffraction was used for crystal phase
identification and estimation of the crystallite size of each phase
present. The percentage of rutile in the samples was determined
from the respective integrated XRD peak intensities using the
photoreactor and the photon flux emitted by the lamp was Φi
-2
)
13.5 mW cm . It was measured by using a radiometer “UVX
Digital” leaned against the external wall of the photoreactor
containing only pure water. O2 was continuously bubbled into
the suspensions for ca. 0.5 h before switching on the lamp and
throughout the occurrence of the photoreactivity experiments.
The temperature inside the reactor was ca. 300 K. The
quantitative determination of 4-nitrophenol was performed by
measuring its absorption at 315 nm with a spectrophotometer
Shimadzu UV-2401 PC. Nonpurgeable organic carbon (NPOC)
determinations were performed by using a Shimadzu TOC
analyzer 5000-A.
Samples Obtained by Preparations 1-3. The amount of
-
1
catalyst was always 1.0 g L , and the initial 4-nitrophenol
-
1
(
BDH) concentration was ca. 20 mg L . The initial pH of the
suspension was adjusted to 2 by addition of HNO3 (Carlo Erba
RPE), and a milky dispersion was obtained. The photoreactivity
runs lasted 6.0 h including the first half hour during which the
lamp was switched off. Samples of 5 mL volume were
withdrawn from the suspensions every 5, 30, or 60 min, and
the catalyst was separated from the solution by filtration through
32
-1
following equation: ø ) (1+ 0.8IA/IR) , where ø is the weight
fraction of rutile in the powder, and IA and IR are the X-ray
intensities obtained from the areas of the peaks relative to the
0
.10 µm Teflon membranes (Whatman).
101 and 110 reflections of anatase and rutile, respectively, which
Samples Obtained by Preparation 4. To obtain the amount
are the most intense reflections in the diffractograms.
-
1
of catalyst (ca. 1 g L ) employed for the photoreactivity runs,
the necessary volume of suspension was determined by the
weight of the solid obtained drying directly 10 mL of suspen-
sion. An alternative procedure consisted in weighing the
precipitate separated after adjustment of the pH of the suspension
to ca. 7 with 1 M NaOH . In both cases similar values were
found.
The crystallite size was estimated from the Scherrer for-
mula: Φ ) Kλ/(â cos θ) where Φ is the crystallite size, λ is
the wavelength of the X-ray radiation (0.154 nm), K is usually
taken as 0.89, â is the peak width at half-maximum height after
subtraction of the equipment broadening, 2θ ) 25.3° and 2θ )
2
7.4° for TiO2 (anatase) and TiO2 (rutile), respectively.
Specific Surface Area and Porosity. The specific surface areas
were determined in a Flow Sorb 2300 apparatus (Micromeritics)
by using the single-point BET method. Porosity was monitored
from the nitrogen adsorption-desorption isotherms obtained at
the liquid nitrogen temperature on samples degassed for 2 h at
Results and Discussion
Characterization. Titanium isopropoxide and titanium tetra-
chloride have been chosen as precursors for the preparation of

Polycrystalline Nanostructured TiO2 Catalysts
J. Phys. Chem. B, Vol. 108, No. 10, 2004 3305
Figure 1. XRD diffractograms of P2-TiO
73) (b); P2-TiO (Cl, 573) (c);- P2-TiO (N, 773) (d); P2-TiO
e); P2-TiO (N, 573) (f). A ) anatase, R ) rutile.
2
(Cl, 773) (a); P2-TiO
2
(Cl,
6
(
2
2
2
(N, 673)
2
Figure 2. XRD diffractograms of P3-TiO
73) (b); P3-TiO (Et, 573) (c); P1-TiO (i-Pr, 773) (d); P1-TiO
673) (e); P1-TiO (i-Pr, 573) (f); P4-TiO (ex TiCl , 3) (g). A ) anatase,
2
(Et, 773) (a); P3-TiO
2
(Et,
6
2
2
2
(i-Pr,
3
3-40
nanostructured TiO2 catalysts since both Ti(iso-OC3H7)4
2
2
4
and TiCl44
1-47
are largely used to prepare nanophase TiO2
R ) rutile.
particles. The properties of the sol-gel derived catalysts depend
on the preparation conditions including the kind of solvent,
solution preparation sequence, aging temperature, boiling time,
and calcination temperature.
that the phase rutile begins to appear only at 873 K in the
powders obtained by hydrolysis of titanium isopropoxide in a
34
water/ethanol mixture. According to Terabe et al., the presence
of residual unhydrolyzed alkyls prevents the crystallization of
the powders. This effect decreases with increasing the amount
of H2O or HCl addition. The presence of rutile in the sample
P3-TiO2(Et,773) calcined at 773 K is probably due to the initial
large water concentration.
Figures 1 and 2 show the X-ray diffraction patterns of some
selected samples. The diffractograms indicate only the presence
of a badly crystallized anatase phase for all of the samples
obtained by preparations 1-3 and calcined at 573 K. The
samples calcined at 673 K consist of a better crystallized anatase
phase with the exception of P2-TiO2(Cl, 673) for which a
mixture of anatase and rutile phases was found. The last result
indicates a positive influence of the chloride ions on the
crystallization process of the rutile phase. This is in agreement
with the findings of Jung and Park39 who observed rutile peaks
over 673 K for TiO2 particles obtained from titanium isopro-
poxide in the presence of HCl. It is worth noting that also the
hydrothermal treatment of amorphous TiO2 suspensions prepared
from TiCl4 leads to the formation of a small amount of rutile
when HCl is used as a cooperative catalyst for the peptization
Starting by TiCl4, preparation 4 allows us to obtain a
photoactive (see Photoreactivity Experiments) anatase phase,
although badly crystallized, in experimental conditions milder
than those used in hydrothermal processes carried out at higher
23-28, 40-42
pressures in closed vessels.
The phases present in the various samples and their relative
amounts are reported in Table 1 together with the diameter of
crystallites determined by the Sherrer equation indicating that
all the samples tested are nanostructured. The microcrystallites
sizes range from 5 to 40 nm. The size of the particles, as
expected, increases by increasing the calcination temperature.
P3-TiO2(Et, 573) and P2-TiO2(Cl, 773) consist of the smallest
and the biggest particles, respectively.
process.4
1,42
Zhang and Gao found that HCl peptization
48
accelerates the anatase-to-rutile phase transformation of the TiO2
nanoparticles prepared by hydrolysis of metatitanic acid and
reduces the phase transformation temperature.
The size of the particles formed in 2-propanol is slightly
higher than that of the particles obtained in the presence of
ethanol. This effect could be ascribed to the different dielectric
constants of the two solvents. By increasing the molecular
weight of the alcohol, the dielectric constant of the solvent
decreases, lowering the electrostatic barrier against the aggrega-
The samples calcined at 773 K, with the exception of P1-
TiO2(i-Pr, 773), consist of a mixture of anatase and rutile phases
in different amounts, depending on the preparation. The presence
-
of Cl seems to favor the formation of rutile even at this
temperature.
5
0
The kind of solvent used influences the crystallization
behavior, porosity, and microstructure of the TiO2 powders
prepared through hydrolysis and condensation of titanium
alkoxides in aqueous media.3 The effect probably arises from
the different abilities of the various solvents to allow condensa-
tion and aggregation of the particles. An important role is also
tion of the primary particles. As found by Vorkapic and
3
8
Matsoukas, the result is decreased stability, enhanced rate of
reaggregation, and formation of larger particles.
7,49
Nitrogen adsorption-desorption isotherms were used to
ascertain the porosity of the catalysts. The inset of Figure 3
displays a typical example of the isotherms obtained with the
various samples. The curves were similar, irrespective of the
3
1
played by the molar ratio water/precursor. Kim et al. found

3306 J. Phys. Chem. B, Vol. 108, No. 10, 2004
Addamo et al.
TABLE 1: Bulk and Physicochemical Properties of the Samples
b
SSA (m g-1)
c
2
× 10 (mol L-1 s-1
9 d
sample
phasea
Φ (nm)
r
0
)
P1-TiO
P1-TiO
P1-TiO
P2-TiO
P2-TiO
P2-TiO
P2-TiO
P2-TiO
P2-TiO
P3-TiO
P3-TiO
P3-TiO
P4-TiO
P4-TiO
P4-TiO
P4-TiO
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
(i-Pr, 573)
(i-Pr, 673)
(i-Pr, 773)
(N, 573)
(N, 673)
(N, 773)
(Cl, 573)
(Cl, 673)
(Cl, 773)
(Et, 573)
(Et, 673)
(Et, 773)
A (badly cryst.)
A (100%)
A(100%)
A (badly cryst.)
A (100%)
8
124
98
79
104
27
3
126
32
14
150
101
49
105
115
110
100
50
10
19
20
negligible
negligible
negligible
negligible
negligible
15
30
10
11
A(50%), R(50%)
A (badly cryst.)
A(56%), R(44%)
A(22%), R(78%)
A (badly cryst.)
A (100%)
A(60%), R(40%)
A (badly cryst.)
A (badly cryst.)
A (badly cryst.)
A (badly cryst.)
A(80%), R(20%)
A (100%)
15[A], 23[R]
7
20[A], 26[R]
40[A], 33[R]
negligible
5
7
13
20
22
10
35
16
10
50
40
20
15[A], 33[R]
(exTiCl
(exTiCl
(exTiCl
(exTiCl
4
4
4
4
, 1)
, 3)
, 12)
, 24)
8
8
8
8
Degussa P25
Tioxide
25[A], 33[R]
7
10
Merck
A (100%)
60
a
A ) anatase, R ) rutile. b Φ ) diameter of crystallites calculated by the Scherrer equation. c SSA ) BET specific surface area. d
0
r ) initial
reaction rate for 4-nitrophenol degradation.
exhibited by the powders prepared from ethanol. The porosity
of the samples derived from TiCl4 was not determined since
outgassing at 523 K for 2 h could cause irreversible changes in
the texture of these catalysts.
The BET specific surface areas (SSA) of the samples (see
Table 1) decrease by increasing the calcination temperature. The
SSA values of P4-TiO2(ex TiCl4, h), not subjected to calcination,
are similar to those of the other samples derived from prepara-
tions 1-3 calcined at 573 K and appear quite independent of
the boiling time.
The decrease of SSA for the samples obtained by means of
preparation 2 (in the presence of HCl or HNO3) was generally
more significant with respect to the corresponding samples
prepared in the absence of acids. The collapse of the porous
network during the calcination treatment does not seem attribut-
able to a detrimental role played by residual traces of inorganic
species since XPS analysis (with the exception of P2-TiO2(Cl,
5
73) containing ca. 1.3 at. % of Cl) did not reveal appreciable
-
-
amounts of Cl or NO3 on the surface of these samples.
Selected SEM micrographs of P1-TiO2(i-Pr, 573), P3-TiO2(Et,
73), P3-TiO2(Et, 773), and P4-TiO2(ex TiCl4, 3) are shown in
Figure 3. Pore size distribution for the samples obtained according to
6
preparation 1: P3-TiO
2
(Et, 573) (a); P3-TiO
2
(Et, 673) (b); P3-TiO
2
(Et,
Figure 4. The samples present irregular shapes and consist of
aggregates of particles whose sizes appear generally bigger than
those calculated by the Sherrer equation, due probably to the
difficulty to obtain good micrographs at a high resolution. All
the powders have small primary particles and larger secondary
particles. The close-packed agglomeration of the particles causes
probably the formation of mesoporous samples.
7
73) (c). Inset: N adsorption-desorption isotherm of P3-TiO
2
2
(Et, 573).
preparation method. According to the IUPAC classification,51
the isotherms were of type IV and exhibited hysteresis loops
of type H2. Thus all the samples were mesoporous and the pores
51,52
were networked.
The mesoporous structure was retained after
calcination at 673 and 773 K, revealing that the close-packed
agglomeration remained although the surface areas of the
samples decreased (see below) due to the grain growth.
The micrographs reveal a remarkable effect of the calcination
temperature on the size of the aggregates of crystallites. Indeed
the biggest particles (ca. 90-170 nm) are present in the samples
calcined at 773 K (Figure 4c), while the smallest ones (ca. 40-
The pore size distribution was calculated using the Dollimore-
Heal method.53 Figure 3 shows the pore size distribution of the
5
4
0 nm) are exhibited by the samples calcined at 573 K (Figure
a) and by the sample P4-TiO2(ex TiCl4, 3) (Figure 4d) that
samples obtained from ethanol after calcination at different
temperatures. The distribution profiles of all the curves were
almost the same, but the maximum of the pore size distribution
shifted toward larger pore diameters with increasing the calcina-
tion temperature. In particular, the maximum pore size increased
from ca. 7 nm at 573 K to 14 nm at 773 K. The decreasing
pore volume is ascribable to the collapse of the network
was not subjected to calcination. In particular the particles of
the last sample consist of aggregates whose size is more uniform
than that of samples derived from the other preparations.
In Figure 5, SEM micrographs of P4-TiO (ex TiCl , 24) and
2
4
P4-TiO (ex TiCl , 3) with 90 days of aging are reported. It can
2
4
(
smallest pores) during the calcination.
The samples obtained from propanol and calcined at the same
be noted that the primary particles (between 100 and 300 nm)
and the secondary ones (bigger than 2 µm) of the samples
obtained after 24 h of boiling or after 90 days of aging appear
temperatures revealed pore sizes slightly smaller than those

Polycrystalline Nanostructured TiO2 Catalysts
J. Phys. Chem. B, Vol. 108, No. 10, 2004 3307
Figure 4. Selected SEM micrographs of P1-TiO
Magnification: ×100,000.
2 2 2 2 4
(i-Pr, 573) (a); P3-TiO (Et, 673) (b); P3-TiO (Et, 773) (c); P4-TiO (ex TiCl , 3) (d).
Figure 5. Selected SEM micrographs of P4-TiO
a) and (c) ×10,000; (b) and (d) ×100,000.
2 4 2 4
(ex TiCl , 24) (a) and (b); P4-TiO (ex TiCl , 3) after 90 days of aging (c) and (d). Magnification:
(
to be larger than those observed in the micrograph of the sample
obtained after 3 h of boiling, without aging (see Figure 4d).
TEM micrographs of P3-TiO2(Et, 573), P3-TiO2(Et, 673), and
P3-TiO2(Et, 773) samples are shown in Figure 6. These
micrographs indicate that the size of the primary particles of
the samples is quite uniform and in accord with the figures
obtained by means of the Scherrer equation.
Figure 7 shows the diffuse reflectance spectra of some
samples. Spectra (a), (b), and (d) of well crystallized samples
indicate that the band-gap energy increases by increasing the
amount of anatase. The spectrum of the P4-TiO2(ex TiCl4, 3)
sample is similar to that of Degussa P25, although a straight-
forward comparison cannot be done, due to its bad crystallinity.
Photoreactivity Experiments. The apparent kinetics of
disappearance of 4-nitrophenol was followed by determining
the concentration of the substrate at various time intervals.
Figure 8 shows the results of 4-nitrophenol oxidation in
the presence of commercial TiO2 powders and of repre-
sentative samples obtained according to different prepara-
tions. The values of the initial reaction rate (r0) are reported in
Table 1.
The calcination temperature seems a key factor strongly
influencing the photoactivity and, in particular, the highest
values of r0 were obtained with the samples calcined at 673
and 773 K. The photoactivity of the calcined samples increases
with increasing the crystallinity of the particles although the
surface area is reduced. Calcination is used to improve the
crystallinity of the powders since the existence of amorphous
contents has generally detrimental effects on the photocatalytic
2
8
properties of TiO2 nanoparticles, due to the recombination of

3308 J. Phys. Chem. B, Vol. 108, No. 10, 2004
Addamo et al.
Figure 6. Selected TEM micrographs of P3-TiO
2
(Et, 573) (a); P3-
(Et, 773) (c). Magnification: ×250,000. P3-
(Et, 573) (d). Magnification: ×400,000.
TiO
TiO
2
(Et, 673) (b); P3-TiO
2
2
Figure 8. 4-Nitrophenol concentration versus irradiation time: (a) P3-
TiO
2
(Et, 573) ([), P3-TiO
2
(Et, 673) (9), P3-TiO
2
(Et, 773) (2); (b)
(ex TiCl , 3) (b).
Degussa P25 ([), Merck (9), Tioxide (2), P4-TiO
2
4
r0 ) 35 × 10 mol L-1 s-1) was tested also without any
-9
(
adjustment of pH value (in this case the pH was ca. 1). No
substantial difference of the initial observed reaction rate was
found, indicating that the presence of Na deriving from NaOH
+
Figure 7. Diffuse reflectance spectra of selected samples: P2-TiO
73) (a); P3-TiO (Et, 773) (b); P4-TiO (ex TiCl , 3) (c); Degussa P25
d).
2
(Cl,
used to adjust the pH was not detrimental.
7
(
2
2
4
Some tests were carried out by using P4-TiO2(ex TiCl4, 12)
and P4-TiO2(ex TiCl4, 24) samples. The final pH values of the
suspensions were ca. 1.5 and 2, respectively, and for both cases
sedimentation of the catalyst particles was observed by stopping
mixing due, probably, to an increase of their size. The activity
of these photocatalysts decreased by increasing the boiling time
photoexcited electrons and holes at defects located on the surface
and in the bulk of the particles.54
As shown in Table 1, the photoactivity of the rutile/anatase
mixed phase P3-TiO2(Et, 773) was higher than that of the pure
anatase samples. This is in agreement with the results of Bickley
-
9
-1 -1
(r0 ) 16 × 10 mol L
s
for 12 h boiling and r0 ) 10 ×
55
56
10 mol L s-1 for 24 h boiling). The comparison with the
-9
-1
et al. and Fotou and Pratsinis reporting that TiO2 photocata-
lysts containing some rutile are more active than the samples
containing only anatase. Moreover, mixed rutile/anatase nano-
particles prepared from titanium isopropoxide42 or from TiCl4
have been found to be more active than pure anatase.
photoactivity observed by using the sample boiled for 1 h (r0
-
9
-1 -1
) 10 × 10 mol L s ) suggests an optimal boiling time
25
between 2 and 4 h.
P4-TiO2(ex TiCl4, 3) was also tested at pH ) 5 and pH )
10. For both pH values an aggregation of particles was observed
and the following lower values of initial reaction rates were
Both P2 sets did not show appreciable photoactivity inde-
pendent of the calcination temperature, indicating that the
presence of HNO3 or HCl is detrimental. On the contrary, P1-
TiO2(i-Pr, 673), P1-TiO2(i-Pr, 773), P3-TiO2(Et, 673), P3-
TiO2(Et, 773), and P4-TiO2(ex TiCl4, 3) revealed a photo-
catalytic activity quite comparable with that of the three
commercial samples.
P4-TiO2(ex TiCl4, 3) appears to be the most promising
catalyst from an application point of view among the various
home-prepared samples because neither separation and washing
of the solid nor its calcination were needed to obtain a highly
active photocatalyst. It is worth reminding indeed that a
photoactive anatase phase was obtained only by means of
treatment in very mild experimental conditions. Several runs
were carried out by using the suspension containing the
photocatalyst, that was stable at pH ) 2, simply by adding the
model pollutant and afterward transferring the dispersion in the
batch photoreactor. The photoactivity of P4-TiO2(ex TiCl4, 3)
-
9
-1 -1
obtained: r0 ) 10 × 10 mol L
s
for pH ) 5 and r0 ) 9.6
-
9
-1 -1
-9
-1 -1
× 10 mol L
s
for pH ) 10 (r0 ) 35 × 10 mol L
s
for pH ) 2).
Figure 9 shows selected experiments carried out by using
the dispersion of P4-TiO2(ex TiCl4, 3) after 7, 30, or 90 days
of aging. No significant differences were found between fresh
and one-week-aged catalysts, but a decreasing photoactivity was
observed by increasing the times of aging (30 and 90 days).
The decrease of photoactivity observed in both cases (prolonged
boiling or aging) is probably related to the growth of the particles
(see Figure 5), although the influence of other physicochemical
factors cannot be excluded.
Figure 10 shows the decrease of the nonpurgeable organic
carbon (NPOC) with the irradiation time for some representative
runs. All the samples appear to be effective to mineralize
completely 4-nitrophenol although after different times. The

Polycrystalline Nanostructured TiO2 Catalysts
J. Phys. Chem. B, Vol. 108, No. 10, 2004 3309
results, and the technique presented here seems an economical
and fast way for the preparation of a highly active photocatalyst.
Acknowledgment. The authors thank the Ministero
dell′Istruzione, dell′Universit a` e della Ricerca, MIUR, Rome
(Progetto L. 488 of Interuniversity National Consortium “Chem-
istry for the Environment”, INCA) for the financial support of
this work and Dr. V. Marcian o` for the assistance in recording
the TEM micrographs. We are indebted to Dr. N. Liotta and
Dr. A. M. Venezia for the porosity and XPS measurements.
References and Notes
Figure 9. 4-Nitrophenol concentration versus irradiation time for P4-
(
1) Photocatalysis and EnVironment. Trends and Applications; Schia-
vello, M., Ed.; Kluwer Academic Publishers: Dordrecht, 1988.
2) Photocatalysis. Fundamentals and Applications; Pelizzetti, E.,
Serpone, N., Eds.; John Wiley & Sons: New York, 1989.
2 4
TiO (ex TiCl , 3): fresh ([), 7 days aged (9), 30 days aged (b), and
9
0 days aged (2).
(
(
35.
3) Linsebigler, A. L.; Lu, G.; Yates, J. T., Jr. Chem. ReV. 1995, 95,
7
(
4) Herrmann, J. M. HelV. Chim. Acta 2001, 84, 2731.
(5) Peral, J.; Domenech, X.; Ollis, D. F. J. Chem. Technol. Biotechnol.
1
997, 70, 117.
6) Augugliaro, V.; Coluccia, S.; Loddo, V.; Marchese, L.; Martra, G.;
Palmisano, L.; Schiavello, M. Appl. Catal. B: EnViron. 1999, 20, 15.
7) Fujishima, A.; Hashimoto, K.; Watanabe, T. TiO2 Photocatalysis:
Fundamentals and Applications; Bkc: Tokyo, 1999.
8) Sclafani, A.; Palmisano, L.; Schiavello, M. J. Phys. Chem. 1990,
4, 829.
9) Kominami, H.; Kohno, M.; Matsunaga, Y.; Kera, Y. J. Am. Ceram.
Soc. 2001, 84, 1178.
10) Venezia, A. M.; Palmisano, L.; Schiavello, M.; Mart ´ı n, C.; Mart ´ı n,
I.; Rives, V. J. Catal. 1994, 147, 115.
(
(
(
9
(
(
(
11) Litter, M. I. Appl. Catal B: EnViron. 1999, 23, 89.
(12) Yamashita, H.; Harada, M.; Misaka, J.; Takeuchi, M.; Ichihashi,
Figure 10. Nonpurgeable organic carbon versus irradiation time:
Degussa P25 ([), P3-TiO (Et, 773) (9), P1-TiO (i-Pr, 573) (2),
P4TiO (ex TiCl , 3) (b).
Y.; Goto, F.; Ishida, M.; Sasaki, T.; Anpo, M. J. Synchroton Rad. 2001, 8,
69.
13) Di Paola, A.; Marc `ı , G.; Palmisano, L.; Schiavello, M.; Uosaki,
2
2
5
2
4
(
K.; Ikeda, S.; Othani, B. J. Phys. Chem. B 2002, 106, 637.
mineralization rate in the initial reaction steps was generally
low, probably due to the accumulation of organic intermediates.
It can be noticed that by using the P4-TiO2(ex TiCl4, 3) sample
a complete mineralization was achieved after a time comparable
with that of Degussa P25 (ca. 4 h). The use of this nanostruc-
tured catalyst in a hybrid membrane photocatalytic reactor could
be useful because of its very easy “in situ” preparation. Studies
are in progress on the behavior of this submicrometric catalyst
in hybrid photoreactors.
(14) Nozik, A. J. In Photocatalytic Purification and Treatment of Water
and Air; Ollis, D. F., Al-Ekabi, H., Eds.; Elsevier: Amsterdam, 1993; p
9.
3
(
15) Hagfeldt, A.; Gr a¨ tzel, M. Chem. ReV. 1995, 95, 49.
(16) Anpo, M. Catal. SurVeys from Japan 1997, 1, 169.
17) Beydoun, D.; Amal, R.; Low, G.; McEvoy, S. J. Nanoparticle Res.
999, 1, 439.
18) Miyake, M.; Torimoto, T.; Sakata, T.; Mori, H.; Yoneyama, H.
(
1
(
Langmuir 1999, 15, 1503.
(19) Torimoto, T.; Kontani, H.; Shibutani, Y.; Kuwabata, S.; Sakata,
T.; Mori, H.; Yoneyama, H. J. Phys. Chem. B 2001, 105, 6838.
(20) Pal, B.; Sharon, M. J. Mol. Catal. A: Chem. 2000, 160, 453.
(21) Bahnemann, D. W.; Kholuiskaya, S. N.; Dillert, R.; Kulak, A. I.;
Conclusions
Kokorin, A. I. Appl. Catal. B: EnViron. 2002, 36, 161.
22) Skubal, L. R.; Meshkov, N. K.; Rajh, T.; Thurnauer, M. J.
Photochem. Photobiol. A: Chem. 2002, 148, 393.
23) Zhang, Z.; Wang, C. C.; Zakaria, R.; Ying, J. Y. J. Phys. Chem. B
998, 102, 10871.
24) Kominami, H.; Kato, J.-i.; Murakami, S.-Y.; Kera, Y.; Inoue, M.;
Inui, T.; Ohtani, B. J. Mol. Catal. A: Chem. 1999, 144, 165.
25) Zhang, Q. H.; Gao, L.; Guo, J. Appl. Catal. B: EnViron. 2000, 26,
07.
26) Hashimoto, K.; Wasada, K.; Toukai, N.; Kominami, H.; Kera, Y.
(
Nanostructured TiO2 catalysts having various particle sizes
and phase composition were prepared by different preparations.
The differences of photoreactivity exhibited by the various
samples depended on the preparation procedures, the calcination
temperature, and the different values of the specific surface
areas. The samples obtained by hydrolysis of titanium isopro-
poxide in mixtures of water and alcohol and in the absence of
acid (preparation 1 and 3) showed an increasing rate of the
photodegradation reaction of 4-nitrophenol as the calcination
temperature increased. This is attributable to the effect of the
presence of a well-crystallized anatase phase that plays a more
important role than that of the particle size.
(
1
(
(
2
(
J. Photochem. Photobiol. A: Chem. 2000, 136, 103.
(27) Maira, A. J.; Yeung, K. L.; Soria, J.; Coronado, J. M.; Belver, C.;
Lee, C. Y.; Augugliaro, V. Appl. Catal. B: EnViron. 2001, 29, 327.
(28) Kominami, H.; Kumamoto, H.; Kera, Y.; Ohtani, B. Appl. Catal.
B: EnViron. 2001, 30, 329.
(29) Molinari, R.; Palmisano, L.; Drioli, E.; Schiavello, M. J. Membr.
Sci. 2002, 206, 399.
(30) Gao, L.; Zhang, Q. Scripta. Mater. 2001, 44, 1195.
The results indicate that the presence of HCl or HNO3 has
detrimental effects on the photocatalytic properties since the
samples prepared by mixtures ethanol-strong acid-water re-
vealed negligible photoactivitites even though the phases and
the size of the particles were similar to those of the samples
obtained in neutral solutions.
The photocatalytic activity of some powders approaches that
of commercial TiO2 catalysts. In particular, the results of several
tests show that the samples derived from TiCl4 exhibit the best
(
31) Kim, J.; Song, K. C.; Pratsinis, S. E. J. Nanoparticle Res. 2000, 2,
419.
(
(
(
32) Spurr, R. A.; Myers, H. Anal. Chem. 1957, 29, 760.
33) Dagan, G.; Tomkiewicz, M. J. Phys. Chem. 1993, 97, 12651.
34) Terabe, K.; Kato, K.; Miyazaki, H.; Yamaguchi, S.; Imai, A.; Iguchi,
Y. J. Mater. Sci. 1994, 29, 1617.
(35) Choi, W.; Termin, A.; Hoffmann, M. R. J. Phys. Chem. 1994, 98,
1
3669.
(36) Khalil, K. M. S.; Zaki, M. I. Powder Technol. 1997, 92, 233.
37) Wu, K.-T.; Spencer, H. G.; J. Non-Cryst. Solids 1998, 226, 249.
(
(38) Vorkapic, D.; Matsoukas, T. J. Am. Ceram. Soc. 1998, 81, 2815.

3310 J. Phys. Chem. B, Vol. 108, No. 10, 2004
Addamo et al.
(
39) Jung, K. Y.; Park, S. B. J. Photochem. Photobiol. A: Chem. 1999,
27, 117.
40) Burnside, S. D.; Shklover, V.; Barb e´ , C.; Comte, P.; Arendse, F.;
Brooks, K.; Gr a¨ tzel, M. Chem. Mater. 1998, 10, 2419.
41) Yin, H.; Wada, Y.; Kitamura, T.; Kambe, S.; Murasawa, S.; Mori,
H.; Sakata, T.; Yanagida, S. J. Mater. Chem., 2001, 11, 1694.
(48) Zhang, R.; Gao, L. Mater. Res. Bull. 2001, 36, 1957.
(49) Hu, L.; Yoko, T.; Kozuka, H.; Sakka, S. Thin Solid Films 1992,
219, 18.
(50) Moon, Y. T.; Park, H. K.; Kim, D. K.; Kim, C. H. J. Am. Ceram.
Soc. 1995, 78, 2690.
(51) Sing, K. S. W.; Everett, D. H.; Haul, R. A. W.; Moscou, L.; Pierotti,
R. A.; Rouqu e´ rol, J.; Siemienieniewska, T. Pure Appl. Chem. 1985, 57,
603.
1
(
(
(
(
42) Yanagisawa, K.; Ovenstone, J. J. Phys. Chem. B 1999, 103, 7781.
43) Anpo, M.; Shima, T.; Kodama, S.; Kubokawa, Y. J. Phys. Chem.
1
987, 91, 4305.
44) Kormann, C.; Bahnemann, D. W.; Hoffmann, M. R. J. Phys. Chem.
988, 92, 5196.
45) Pelizzetti, E.; Minero, C.; Borgarello, E.; Tinucci, L.; Serpone, N.
Langmuir 1993, 9, 2995.
46) Zhang, Q. H.; Gao, L.; Guo, J. K. Nanostruct. Mater. 1999, 11,
293.
47) Eremenko, B. V.; Bezuglaya, T. N.; Savitskaya, A. N.; Malysheva,
M. L.; Kozlov, I. S.; Bogodist, L. G. Colloid J. 2001, 63, 173.
(52) Gregg, S. J. In Adsorption at the Gas-Solid and Liquid-Solid
Interfaces; Roquerol, J., King, K. S. W., Eds.; Elsevier: Amsterdam, 1982;
pp 153-164.
(
1
(
(53) Dollimore, D.; Heal, G. R. J. Appl. Chem. 1964, 14, 109.
(54) Ohtani, B.; Ogawa, Y.; Nishimoto, S.-i. J. Phys. Chem. B 1197,
101, 3746.
(
1
(55) Bickley, R. I.; Gonz a´ lez-Carre n˜ o, T.; Lees, J. S.; Palmisano, L.;
Tilley, R. J. D. J. Solid State Chem. 1991, 92, 178.
(56) Fotou, G. P.; Pratsinis, S. Chem. Eng. Commun. 1996, 151, 251.
(