Nanosized anatase TiO2 as precursor for preparation of sulfated titania catalysts

Article abstract of DOI:10.1016/S1381-1169(01)00280-1

Nanosized anatase TiO₂ has been prepared by a sol-gel hydrothermal process. The effect of water:alkoxide molar ratio, hydrothermal aging temperature and aging time on the properties of the nanocrystalline anatase TiO₂ was investigated in detail. The specific surface area, sulfur content and surface acidity of sulfated titania catalysts prepared by sulfation of sol-gel hydrothermal derived nanocrystalline TiO₂ were greater than those of the analogous catalyst prepared from ordinary amorphous titania hydrate. The former catalysts displayed higher catalytic activities in isopropanol dehydration, cumene cracking and n-pentane conversion reactions. The influence of the crystalline state and crystalline grain of titania precursors on the properties of sulfated titania catalysts was studied and discussed.

Full text of DOI:10.1016/S1381-1169(01)00280-1

Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
Nanosized anatase TiO2 as precursor for preparation
of sulfated titania catalysts
Zhen Ma, Yinghong Yue, Xingyi Deng, Zi Gao∗
Department of Chemistry, Fudan University, Shanghai 200433, PR China
Received 14 March 2001; accepted 22 June 2001
Abstract
Nanosized anatase TiO2 has been prepared by a sol–gel hydrothermal process. The effect of water:alkoxide molar ratio,
hydrothermal aging temperature and aging time on the properties of the nanocrystalline anatase TiO2 was investigated in
detail. The specific surface area, sulfur content and surface acidity of sulfated titania catalysts prepared by sulfation of
sol–gel hydrothermal derived nanocrystalline TiO2 were greater than those of the analogous catalyst prepared from ordinary
amorphous titania hydrate. The former catalysts displayed higher catalytic activities in isopropanol dehydration, cumene
cracking and n-pentane conversion reactions. The influence of the crystalline state and crystalline grain of titania precursors
on the properties of sulfated titania catalysts was studied and discussed. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Sol–gel hydrothermal process; Nanosized TiO2; Sulfated titania; Solid acid catalyst
1. Introduction
to characterize these materials. It has been reported
in the literature [1,6–10] that amorphous Zr(OH)4 or
Ti(OH)4 should be used as the precursor for sulfa-
tion to ensure high sulfur content and strong acidity,
but a few authors have found that sulfated zirconia
catalysts obtained by H2SO4 impregnation of a crys-
talline ZrO2 [11] or a zirconia aerogel [12] followed
by calcination have identical acidity and activity as
catalysts prepared from amorphous zirconia hydrate.
Among the sulfated metal oxides, sulfated zirconia
is known to possess the strongest acidity and exhibit
superior activity to many acid-catalyzed reactions.
Hence, a great deal of research works in this area
have been concentrated on the study of this type of
material. In comparison to the voluminous literatures
on sulfated zirconia, knowledge of sulfated titania is
limited. However, the latter material has been found
to be efficient for catalyzing isomerization [13–15],
alkylation [8,16–18], Friedel–Crafts acylation [19,20],
esterification [21–23], CFC decomposition [24,25],
Sulfated metal oxides such as sulfated zirconia,
titania, tin oxide and iron oxide are solid strong acids
that have arisen increasing interest in recent years be-
cause of their high thermostability, strong acidity and
unique catalytic activities in many environmentally
benign reactions [1–5]. The method of preparation
significantly affects the catalytic properties of these
materials. Numerous methods have been reported
in the literature in particular for the preparation of
sulfated zirconia. They differ mainly in the type
of oxide precursor, type of sulfating agent, method
of impregnation and calcination temperature. Some-
times the results are controversial because different ex-
perimental techniques or test reactions have been used
∗ Corresponding author. Tel.: +86-21-65642792;
fax: +86-21-65641740.
E-mail address: zigao@fudan.edu.cn (Z. Gao).
1381-1169/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S1381-1169(01)00280-1

98
Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
photocatalytic oxidation [26,27] and selective cat-
alytic reduction of NOx [28]. The increasing appli-
cation of sulfated titania draws our attention to the
study of the influence of preparation parameters on
its properties and catalytic activities.
solution for 30 min, followed by filtering, drying at
110 ◦C overnight and calcining at 500 ◦C for 3 h.
2.2. Characterization
BET surface area of the samples was measured
on a Micromeritics ASAP 2000 equipment using
N2 as the adsorbent. X-ray powder diffraction mea-
surements were performed on a computer-controlled
Rigaku D/MAX-IIA instrument with Cu K␣ radia-
tion at 40 kV and 20 mA with scan speed 8◦/min and
scan range 10–80◦. The crystallite size of various
samples was estimated by applying Scherrer formula
on the (1 0 1) diffraction peaks of anatase TiO2. The
amount of acid sites was measured by NH3-TPD in a
flow-type fixed-bed reactor at ambient pressure. The
0.1 g sample was heated at 500 ◦C for 2 h and cooled
to 120 ◦C in flowing He. At 120 ◦C, the sample was
treated by sufficient pulses of NH3 till adsorption
saturation, followed by a He purge for about 2 h. The
temperature was raised up from 120 to 500 ◦C at a
rate of 10 ◦C/min to desorb NH3, and then was fur-
ther maintained at 500 ◦C for 0.5 h. Sulfur content of
the samples was determined by a chemical method.
Dehydrated Na2CO3 and ZnO (1:4) were used as
fusing agents, and the sulfate was turned into BaSO4
and detected by gravimetric method.
In this work, nanosized anatase TiO2 was prepared
by a sol–gel hydrothermal method [29] and used as the
precursor for preparation of SO42−/TiO2. The physic-
ochemical properties, surface acidity and catalytic
activities of SO42−/TiO2 prepared from this type of
nanosized anatase TiO2 and conventional amorphous
titania hydrate were investigated and compared in
detail.
2. Experimental
2.1. Sample preparation
Anatase TiO2 nanocrystals were prepared by
a combination of conventional sol–gel hydroly-
sis precipitation and hydrothermal processing [29].
In a typical synthesis, 0.1 mol titanium butoxide
(Ti(OC4H9)4) was dissolved in 1.5 mol anhydrous
ethanol, and the butoxide solution was added drop-
wise to a water–ethanol solution containing 1 mol
ethanol with rigorous stirring at room temperature.
The water:alkoxide molar ratio ranged between 5 and
150 in this paper. A white precipitate of hydrous ox-
ide was formed and the mixture was stirred for 3 h.
Then the mother liquor containing the amorphous
precipitate was transferred to a stainless steel auto-
clave and aged at 80–180 ◦C for 1–5 days, followed
by filtering, washing with deionized water and drying
at 110 ◦C overnight. The products were denoted as
SGH-TiO2–r–t–d, where r means the water:alkoxide
molar ratio, t the hydrothermal treatment temperature
(◦C) and d the aging time (days).
For comparison, two other precursors were pre-
pared: (1) amorphous Ti(OH)4 (A-TiO2) was prepared
by dropping aqueous ammonia to TiCl4 dissolved in
HCl till pH = 9–10, aging at room temperature for
24 h, washing, filtering and drying at 110 ◦C overnight;
(2) precrystallized TiO2 (A-TiO2 (500 ◦C)) was pre-
pared by calcining the amorphous Ti(OH)4 powder at
500 ◦C for 3 h.
2.3. Reaction testing
The dehydration of isopropanol was performed in a
pulsed microreactor with a catalyst load of 50 mg. The
catalyst was preheated in H2 flow at 250 ◦C for 3 h. H2
was used as the carrier gas at a flow rate of 30 ml/min.
The amount of isopropanol injected for each test was
5 ␮l and the reaction temperatures were 150, 175 and
200 ◦C.
The activity of the samples toward cumene cracking
reaction was tested in a pulsed microreactor loaded
with 50 mg catalyst. The catalyst was preheated under
N2 atmosphere at 350 ◦C for 3 h. H2 was used as the
carrier gas at a flow rate of 40 ml/min. The amount of
cumene injected for each test was 1 ␮l and the reaction
temperature was 300 ◦C.
n-Pentane conversion reaction was carried out in a
pulsed microreactor under ambient pressure on 0.4 g
catalyst. The catalyst was preheated under N2 atmo-
sphere at 400 ◦C for 3 h. Upon reaction, the carrier gas
Sulfated titania samples were prepared by immers-
ing the above titania precursors in 1 mol/l H2SO4

Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
99
was N2 with a flow rate of 30 ml/min. The amount of
n-pentane injected for each pulse was 0.5 ␮l and the
reaction temperatures were 325 and 350 ◦C.
exceptional, because crystallization was incomplete
after aging at 80 ◦C for 1 and 3 days. For systems
with the water:alkoxide ratio greater than 30, increas-
ing the water:alkoxide ratio had no significant effect
on the BET surface area of the samples. The above
observations are consistent with the results in the
literature [29].
3. Results
3.1. Sol–gel hydrothermal derived TiO2
The BET surface area of SGH-TiO2 decreased as
the aging temperature was raised from 80 to 180 ◦C
and the aging time was prolonged from 1 to 5 days.
However, the effect of aging time was less pronounced
than that of aging temperature. BET surface areas of
the SGH-TiO2 samples prepared with a water:alkoxide
ratio of 50 were in the range of 149–223 m2/g. The
calculated equivalent sizes of TiO2 nanocrystals would
be 6.9–10.3 nm, assuming a monodispersed spherical
morphology.
The variation of BET surface area with calci-
nation temperature for various titania samples was
illustrated in Fig. 2. The surface area of A-TiO2 de-
creased drastically at 300–350 ◦C when the phase
transformation from amorphous titania to anatase
was accomplished. Although the initial surface ar-
eas of SGH-TiO2 samples were smaller than that
of A-TiO2, the reduction in their surface area with
calcination temperature was much slower, show-
ing that nanocrystals of anatase TiO2 prepared by
sol–gel hydrothermal process are thermally more
stable than amorphous TiO2. At calcination tem-
peratures above 350 ◦C, BET surface areas of the
samples were in the order of A-TiO2 < SGH-TiO2–
In the preparation of sulfated titania, sulfation of
the oxide precursor is usually carried out by immers-
ing dried oxide precursor into a dilute H2SO4 solu-
2−
tion allowing adequate adsorption of SO4 onto the
precursor. Then the sulfated precursor is calcined at
a high temperature to produce highly covalent sulfate
structure on the surface, which is considered as the
source of the strong acidity on this type of material.
It was expected that the specific surface area of the
oxide precursor might exert a significant influence on
the properties of the final catalyst. Thus, the effect of
water:alkoxide molar ratio, aging temperature and ag-
ing time on BET surface area of sol–gel hydrothermal
derived TiO2 was investigated in detail and the results
were illustrated in Fig. 1.
The BET surface area of the SGH-TiO2 samples
was sensitive to the water:alkoxide molar ratio as
the water:alkoxide ratio was below 30. For most
of the samples increasing the water:alkoxide ratio
from 5 to 30 led to an increase in the surface area.
Samples SGH-TiO2–5–80–1, SGH-TiO2–5–80–3,
SGH-TiO2–15–80–1 and SGH-TiO2–15–80–3 were
Fig. 1. Effect of preparation parameters (water:alkoxide ratio, hy-
drothermal aging temperature and aging time) on BET surface
area of sol–gel hydrothermal derived TiO2.
Fig. 2. Variation of BET surface area with calcination temperature
for various TiO2 samples.

100
Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
Table 1
Physicochemical properties of various precursors and catalysts
Samples
SBET (m2/g)
Total pore
volume (cm3/g)
Average pore
size (nm)
Crystallite
size (nm)
SO3 (wt.%)
NH3-TPD
acidity (mmol/g)
A-TiO2
367
34.8
169
223
62.9
94.5
101
24.8
120
138
0.26
0.06
0.28
0.29
0.14
0.20
0.12
0.06
0.28
0.23
2.9
7.3
6.7
5.3
8.9
8.7
4.6
8.9
9.4
6.8
–
–
–
–
–
–
–
3.3
0.6
3.2
3.6
–
0.23
–
–
–
A-TiO2(500 ◦C)
50
10
6.5
25
15
12
65
12
10
SGH-TiO2–50–180–1
SGH-TiO2–50–80–1
SGH-TiO2–50–180–1 (500 ◦C)
SGH-TiO2–50–80–1 (500 ◦C)
SO42−/A-TiO2
SO42−/A-TiO2 (500 ◦C)
SO42−/SGH-TiO2–50–180–1
SO42−/SGH-TiO2–50–80–1
–
0.66
0.45
0.72
0.79
50–180–1 < SGH-TiO2–50–80–1. For example, BET
surface areas of A-TiO2, SGH-TiO2–50–180–1 and
SGH-TiO2–50–80–1 calcined at 500 ◦C were 34.8,
62.9 and 94.5 m2/g, respectively (see Table 1).
and 10 nm, respectively, by applying Scherrer formula
on the (1 0 1) diffraction peaks. These results coincide
very well with those calculated from BET surface ar-
eas.
XRD patterns of SGH-TiO2 samples prepared with
a water:alkoxide ratio of 50 were compared with those
of A-TiO2 and A- TiO2 (500 ◦C) in Fig. 3. A-TiO2 pre-
pared by hydrolysis of TiCl4 with ammonia solution
was amorphous in nature and had a high BET surface
area of 367 m2/g. A-TiO2 (500 ◦C) obtained by cal-
cining A-TiO2 at 500 ◦C was crystalline anatase with
intense diffraction peaks, as its BET surface area was
only 34.8 m2/g. All the SGH-TiO2 samples displayed
low and broad diffraction peaks of anatase TiO2. The
crystallite sizes of the SGH-TiO2 samples aged at 80,
120 and 180 ◦C for 1 day were estimated to be 6.5, 7.5
XRD patterns of various TiO2 samples calcined
at 500 ◦C were compared in Fig. 4. No rutile
peaks were detected in all these patterns. The in-
tensity of the anatase peaks decreased in the or-
der of A-TiO2 (500 ◦C) > SGH-TiO2–50–180–1 >
SGH-TiO2–50–80–1, and the estimated crystallite
sizes of those three samples using Scherrer formula
were 50, 25 and 15 nm, respectively. This further con-
firms that the nanocrystals of SGH-TiO2 are thermally
more stable. After calcination at 500 ◦C, the crys-
tallite sizes of SGH-TiO2 are two-fold or three-fold
smaller than that of A-TiO2 (500 ◦C).
Fig. 3. XRD patterns of TiO2 samples prepared in different ways.
Fig. 4. XRD patterns of various TiO2 samples calcined at 500 ◦C.

Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
101
3.2. Sulfated titania catalysts
According to the results of chemical analysis, the
sulfur content of SO42−/SGH-TiO2–50–80–1 was
slightly higher than that of SO42−/SGH-TiO2–50–
180–1 and SO42−/A-TiO2. The sulfur content of
SO42−/A-TiO2 (500 ◦C) was much lower than that
of all the other catalysts due to the limited initial
adsorption of sulfate on the precrystallized anatase
precursor with a low surface area.
Sulfated titania catalysts were prepared using
A-TiO2, A-TiO2 (500 ◦C), SGH-TiO2–50–180–1 and
SGH-TiO2–50–80–1 as precursors and characterized.
The characterization results of the catalysts were
listed and compared with those of the neat precursors
in Table 1.
BET surface areas of SO42−/A-TiO2, SO42−/SGH-
TiO2–50–180–1 and SO42−/SGH-TiO2–50–80–1
calcined at 500 ◦C were lower than those of the
non-calcined precursors but higher than those of the
precursors calcined at 500 ◦C. The XRD character-
istic peaks of anatase TiO2 for these SO42−/TiO2
samples calcined at 500 ◦C were lower and broader
than those for the corresponding precursors calcined
at 500 ◦C, as shown in Fig. 5. These observations are
in support of the generally recognized viewpoint that
sulfation inhibits the sintering of titania and stabilizes
the surface area of sulfated titania catalyst [4]. Data
in Table 1 show that finer anatase crystallites are
found in sulfated titania catalysts than in pure titania
calcined at the same temperature, which accounts for
the smaller reduction in surface area upon calcination
for the sulfated titania catalysts. The measured BET
surface areas of the catalysts were in the order of
SO42−/SGH-TiO2–50–80–1 > SO42−/SGH-TiO2–
The surface acidity of the catalysts was measured by
NH3-TPD method. The NH3-TPD profiles of the sul-
fated catalysts were shown in Fig. 6. There was only
one asymmetric broad peak on the profiles, and no ev-
ident peak temperature could be read out from the pro-
files. The maxima of the NH3 adsorption peaks were
approximately in the range of 350–450 ◦C, indicating
that the catalysts were abundant in medium-strong acid
sites. The total amount of acid sites of the catalysts
followed the order of SO42−/SGH-TiO2–50–80–1
(0.79 mmol/g) > SO42−/SGH-TiO2–50–180–1 (0.72
2−
mmol/g) > SO42−/A-TiO2 (0.66 mmol/g) > SO4
/
A-TiO2 (500 ◦C) (0.45 mmol/g), which was parallel
to the order of their surface area.
3.3. Activity testing
The catalytic behavior of the sulfated titania cata-
lysts towards acid-catalyzed reactions requiring acid
50–180–1
>
SO42−/A-TiO2
ꢀ
SO42−/A-TiO2
(500 ◦C). In general, the pore volume of the samples
is related to their crystallite size. The pore volume is
increased as the anatase crystallite size in the samples
becomes finer and more uniform.
2−
Fig. 6. NH3-TPD profiles of: (a) A-TiO2 (500 ◦C); (b) SO4
/
A-TiO2 (500 ◦C); (c) SO42−/A-TiO2; (d) SO42−/SGH-TiO2–50–
180–1; (e) SO42−/SGH-TiO2–50–80–1.
Fig. 5. XRD patterns of various sulfated titania catalysts.

102
Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
Fig. 7. Conversion of isopropanol as a function of reaction tem-
perature.
Fig. 8. Conversion of cumene as a function of pulse number at
300 ◦C.
SO42−/SGH-TiO2–50–180–1 > SO42−/A-TiO2
>
sites with different acid strengths, such as isopropanol
dehydration, cumene cracking and n-pentane conver-
sion, has been investigated.
SO42−/A-TiO2 (500 ◦C), which was the same as that
of their total acidity.
Isopropanol dehydration is known to be catalyzed
by weak acid sites. The conversion of isopropanol on
various catalysts as a function of reaction temperature
was shown in Fig. 7. Non-sulfated TiO2 was inactive
for the reaction under these reaction temperatures,
but an obvious enhancement in activity was observed
for all the sulfated catalysts. However, there was a
marked difference in activity among the catalysts
prepared from different titania precursors. At all the
three reaction temperatures, the dehydration activity
followed the order of SO42−/SGH-TiO2–50–80–1 >
Cumene converts to benzene and propene on acid
catalysts with medium-strong and strong acid sites.
The conversion of cumene dropped rapidly with in-
creasing pulse number because of coking on the sul-
fated catalysts. The variation of cumene conversion
with pulse number on the catalysts at 300 ◦C was
shown in Fig. 8. The initial activity of the catalysts for
cumene cracking followed the same order as that for
2−
isopropanol dehydration, implying that the SO4
/
SGH-TiO2 catalysts have not only greater total acidity
but also greater medium-strong and strong acidity.
Table 2
Conversion of n-pentane and product distribution on various samples
Catalyst
Product distribution (mol%)
Conversion (%)
C1
C2
C3
i-C4
n-C4
i-C5
n-C5
325 ◦C
0
0
0.4
0.8
0.9
350 ◦C
0
TiO2
0
0.1
3.4
6.4
7.6
0
0.1
17.5
23.2
29.1
0
0
0
0.5
6.4
0
0
8.8
5.6
9.0
0
100
0
0.3
30.2
36.5
53.0
SO42−/A-TiO2 (500 ◦C)
SO42−/A-TiO2
0.1
0.1
0
99.7
69.8
63.4
47.0
SO42−/SGH-TiO2–50–180–1
SO42−/SGH-TiO2–50–80–1
0
TiO2
0
0.4
11.0
16.6
22.5
0
0.7
29.4
38.3
47.8
0
0
0
0
0
0
0
100
0
1.5
47.5
60.1
78.4
SO42−/A-TiO2 (500 ◦C)
SO42−/A-TiO2
0
0.1
0.8
0
0.3
4.9
3.1
4.9
98.5
52.5
39.9
21.6
1.4
2.1
3.2
SO42−/SGH-TiO2–50–180–1
SO42−/SGH-TiO2–50–80–1
0

Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
103
The catalytic performance of the catalysts in
n-pentane conversion was similar to that in cumene
cracking. The conversion of n-pentane also decreased
rapidly with increasing pulse number due to coking
on the catalysts. The initial conversion of n-pentane
and the product distribution for different catalysts at
325 and 350 ◦C were listed in Table 2. The main reac-
tion product was C3. The selectivity to C4 was higher
than that to C2 at 325 ◦C, and au contraire at 350 ◦C.
The sequence of the initial activities of the catalysts in
this reaction was similar to that in cumene cracking,
which again confirms that SO42−/SGH-TiO2 catalysts
have more medium-strong and strong acid sites than
SO42−/A-TiO2 and SO42−/A-TiO2 (500 ◦C) catalysts.
It has been noted that the crystalline grain and
specific surface area of the oxide precursor exert a
significant influence on the properties of the cata-
lyst. Catalysts obtained by sulfating SGH-TiO2 with
smaller crystalline grain and higher surface area
possess greater sulfate content and higher activ-
ity, whereas sulfating A-TiO2 (500 ◦C) with larger
crystalline grain and lower surface area gives unsatis-
factory results although both precursors are nanocrys-
talline anatase TiO2. Since sulfation of the oxide is
2−
carried out by SO4 adsorption onto the precursor,
a smaller crystalline grain and a higher surface area
of the precursor enhance the adsorption and, hence,
increase the sulfate content of the final catalyst, even
if a part of the sulfate is lost during thermal activation
of the catalyst. On the other hand, a larger amount of
sulfate adsorbed in the precursor inhibits the sintering
of the oxide more efficiently upon calcination [4].
These two reasons may explain the positive effect of
the small crystalline grain and high specific surface
area of the SGH-TiO2 precursors on the properties of
the final sulfated oxide catalysts.
4. Discussion
Experimental results in this work demonstrate that
the type of precursor plays an important role in the
preparation of sulfated oxide catalysts. The fact that
the catalysts prepared from sol–gel hydrothermal
derived nanocrystalline anatase TiO2 have greater
surface area and higher catalytic activity than ordi-
nary SO42−/TiO2 catalyst prepared from amorphous
titania hydrate tells that to use amorphous oxide as
precursor is no more a requisite for the preparation
of a good sulfated oxide catalyst. The validity of the
generally realized concept that sulfation of crystal-
lized oxide does not produce strong acidity [4] be-
comes questionable, since both SGH-TiO2–50–180–1
and SGH-TiO2–50–80–1 are nanocrystalline anatase
TiO2 with distinct reflections in their XRD patterns.
Furthermore, it is also interesting to note that not
all nanocrystalline anatase TiO2 are good precursors.
Precrystallized A-TiO2 (500 ◦C) with a crystallite size
of 50 nm obtained by calcination of amorphous tita-
nia belongs to the category of anatase nanocrystals as
well, but the sulfated catalyst prepared using A-TiO2
(500 ◦C) as a precursor displays low surface area and
poor activity. Therefore, it is clear that amorphous
or crystalline, crystalline or nanocrystalline are not
the watersheds between a good and a poor precur-
sor. Careful measurements of the samples at different
preparative stages in the present work suggest that
there are other factors controlling the quality of the
oxide precursors used for the preparation of sulfated
oxide catalysts.
Nevertheless, the initial surface area is not the sole
criterion for an oxide precursor, because the surface
area and activity of SO42−/A-TiO2 catalyst prepared
from amorphous A-TiO2 with a BET surface area as
high as 367 m2/g are appreciably lower than those of
SO42−/SGH-TiO2 catalysts. Such an apparent abnor-
mality may be explained on the basis of thermosta-
bility. After sulfation, the titania precursors must be
calcined at a high temperature (>500 ◦C) to produce
strong acidity. In our previous work, it has been ob-
served that the transition of amorphous TiO2 to crys-
talline anatase TiO2 takes place at 350 and 500 ◦C
for pure TiO2 and sulfated TiO2, respectively [13].
The phase transition is accompanied by a significant
reduction in surface area as shown in Fig. 2. Mean-
while, the nanocrystals of anatase TiO2 in sulfated
SGH-TiO2 samples are more thermally stable. Upon
calcination, there is no phase transition and the pres-
2−
ence of SO4 species stabilizes the surface area, so
the final SO42−/SGH-TiO2 catalysts retain higher sur-
face areas and higher catalytic activities.
The structure of the surface sulfate complex and
the nature (Lewis or Bronsted type) of the strong
acid sites on sulfated oxide catalysts are still a sub-
ject of debate, because technological advances to
date are inadequate for a complete characterization

104
Z. Ma et al. / Journal of Molecular Catalysis A: Chemical 178 (2002) 97–104
of these catalysts. However, information in the lit-
erature [12,30–32] show that the texture of sulfated
zirconia catalysts prepared from amorphous, aerogel
and precrystallized zirconia differs from each other,
but the feature of the active sites on these catalysts is
much the same. This is probably also true for sulfated
titania catalysts prepared from different precursors
according to the results of this work. From the XRD,
NH3-TPD and activity measurements there is no ev-
ident indication that the nature of the surface sulfate
species has been altered due to the change of pre-
cursors. The acidity and the catalytic activities of the
catalysts toward various acid-catalyzed reactions cor-
relate fairly well with their sulfur content and specific
surface area, implying that the variation in texture of
the catalysts rather than in structure or nature of the
active sites accounts for the difference in the catalysts.
[7] F.R. Chen, G. Coudurier, J.F. Joly, J.C. Vedrine, J. Catal. 143
(1993) 616.
[8] C. Guo, S. Yao, J. Cao, Z. Qian, Appl. Catal. A 107 (1994)
229.
[9] R.A. Comelli, C.R. Vera, J.M. Parera, J. Catal. 151 (1995)
96.
[10] C. Morterra, G. Cerrato, F. Pinna, M. Signoretto, J. Catal.
157 (1995) 109.
[11] T. Riemer, D. Spielbauer, M. Hunger, G.A.H. Mekhemer, H.
Knozinger, Chem. Commun. (1994) 1181.
[12] Y.Y. Huang, B.Y. Zhao, Y.C. Xie, Appl. Catal. A 172 (1998)
327.
[13] Z. Gao, J.M. Chen, Chem. J. Chin. Univ. 15 (1994) 873.
[14] Z. Gao, J.M. Chen, W.M. Hua, Y. Tang, Stud. Surf. Sci.
Catal. 90 (1994) 507.
[15] A.K. Dalai, R. Sethuraman, S.P.R. Katikaneni, R.O. Idem,
Ind. Eng. Chem. Res. 37 (1998) 3869.
[16] A. Corma, A. Martinez, C. Martinez, Appl. Catal. A 144
(1996) 249.
[17] A. Hess, E. Kemnitz, Appl. Catal. A 149 (1997) 373.
[18] S.K. Samantaray, T. Mishra, K.M. Parida, J. Mol. Catal. A
156 (2000) 267.
[19] Y.D. Xia, W.M. Hua, Z. Gao, Catal. Lett. 55 (1998) 101.
[20] W.M. Hua, Y.D. Xia, Y.H. Yue, Z. Gao, J. Catal. 196 (2000)
104.
Acknowledgements
This work was supported by the Major State
Basic Research Development Program (Grant no.
2000077507) and the Foundation for University Key
Teacher by the Ministry of Education.
[21] G.Z. Lu, Appl. Catal. A 133 (1995) 11.
[22] J. Lin, Chin. J. Org. Chem. 20 (2000) 805.
[23] J. Lin, H.T. Liu, R.Q. Zhao, Chin. J. Inorg. Chem. 16 (2000)
829.
[24] X.Z. Fu, W.A. Zeltner, Q. Yang, M.A. Anderson, J. Catal.
168 (1997) 482.
[25] Z. Ma, W.M. Hua, Y. Tang, Z. Gao, Chin. J. Chem. 18 (2000)
341.
References
[26] X.Z. Fu, Z.X. Ding, W.Y. Su, D.Z. Li, Chin. J. Catal. 20
(1999) 321.
[27] W.Y. Su, X.Z. Fu, K.M. Wei, Acta Physico-Chimica Sinica
17 (2001) 28.
[1] K. Arata, Adv. Catal. 37 (1990) 165.
[2] B.H. Davis, R.A. Keogh, R. Srinivasan, Catal. Today 20
(1994) 219.
[3] A. Corma, Chem. Rev. 95 (1995) 559.
[4] X. Song, A. Sayari, Catal. Rev. Sci. Eng. 38 (1996) 329.
[5] G.D. Yadav, J.J. Nair, Micropor. Mesopor. Mater. 33 (1999)
1.
[28] S.M. Jung, P. Grange, Catal. Today 59 (2000) 305.
[29] C.C. Wang, J.Y. Ying, Chem. Mater. 11 (1999) 3113.
[30] C. Morterra, G. Cerrato, V. Bolio, Catal. Today 17 (1993)
505.
[6] K. Tanabe, H. Hattori, T. Yamaguchi, Crit. Rev. Surf. Chem.
1 (1990) 1.
[31] C.X. Miao, Z. Gao, Mater. Chem. Phys. 50 (1997) 15.
[32] A.F. Bedilo, K.J. Klalunde, J. Catal. 176 (1998) 448.