Synthesis, characterization and catalytic properties of Pt/CeO2-Al2O3 and Pt/La2O3-Al2O3 sol-gel derived catalysts

Article abstract of DOI:10.1016/S1381-1169(00)00495-7

Platinum supported on La₂O₃-Al₂O₃ and CeO₂-Al₂O₃ were prepared using the sol-gel technique. Specific surface areas of 572, 459, and 418 sq m/g were obtained on the solids. The FT

Full text of DOI:10.1016/S1381-1169(00)00495-7

Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
Synthesis, characterization and catalytic properties of
Pt/CeO2–Al2O3 and Pt/La2O3–Al2O3 sol–gel derived catalysts
A. Vazqueza, T. Lopezb, R. Gomeza,b,∗, X. Bokhimic
a
Instituto Mexicano del Petroleo, Eje Central Lazaro Cardenas No. 152, Mexico 07000, D.F., Mexico
b
Department of Chemistry, Universidad Autonoma Metroplitana-Iztapalapa,
P.O. Box 55-534, Mexico 9340, D.F., Mexico
UNAM Instituto de Fisica, Ciudad Universitaria, Mexico 09000, D.F., Mexico
c
Received 6 April 2000
Abstract
Platinum supported on La2O3–Al2O3 and CeO2–Al2O3 were prepared by the sol–gel method. High specific surface areas
between 572–418 m2/g were obtained. FTIR-pyridine absorption band proved the formation of Lewis sites with comparable
intensity in the supports. The stabilization of the ␥-Al2O3 in the La2O3–Al2O3 support was observed by X-ray diffraction when
the sample was calcined at 1100◦C, while segregation of CeO2 was observed at such temperature in the CeO2–Al2O3 prepara-
tion. The formation of AlV and AlVI coordinated aluminum was observed by MAS-NMR 27Al spectra in La doped-alumina. In
the Ce doped-sample only AlVI coordination was identified. Propene and acetone were the products in the iso-propanol decom-
position over CeO2–Al2O3, meanwhile, for the La2O3–Al2O3 support only propene was obtained. The platinum (0.5 wt.%)
particle size distribution as determined by electron microscopy showed very small particles <1.0 nm. On the other hand, in
the n-heptane dehydrocyclization the activity and selectivity to toluene of the Pt/La2O3–Al2O3 catalyst was found bigger than
those of Pt/CeO2–Al2O3 catalyst. It is proposed that doped-alumina prepared by the sol–gel method has structural defects
which induce modification in the metal activity. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Sol–gel catalysts; Platinum lanthanum–alumina; Platinum cerium–alumina; MAS-NMR sol–gel catalysts; Acidity sol–gel catalysts
prepared by co-gelling aluminum tri-sec-butoxide and
tetrabutyl tin, efficient catalysts for the n-heptane de-
hydrocyclization were obtained [14–17]. Bimetallic
Pd–Co/SiO2 catalysts prepared by the sol–gel method
have been reported as selective catalysts for the hydro-
genation of CO [18,19]. The gas exhaust catalytic con-
verter reactions like carbon monoxide oxidation was
carried out in high metal sintering resistance Pt/TiO2
sol–gel catalysts [20]. The non-selective catalytic re-
duction of NO by CO on Pt/Al2O3 and Rh/Al2O3
and Pd/La2O3–Al2O3 prepared by the sol–gel method
have also been reported [21–23]. Moreover, Rh/ZrO2
catalysts show high activity for the combustion of
1. Introduction
The preparation of metal supported catalysts by the
sol–gel method has been demonstrated to be useful
for a large number of applications [1]. For example,
highly resistant catalysts to sintering or poisoning by
coke and sulfur were prepared by adding ruthenium
[2–4], palladium [5–8] or platinum precursors to sil-
ica gel [9–13]. In reforming Pt–Sn/Al2O3 catalysts
∗ Corresponding author. Tel.: +52-5804-5668;
fax: +52-5804-4666.
E-mail address: gomr@xanum.uam.mx (R. Gomez).
1381-1169/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S1381-1169(00)00495-7

92
A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
methane [24]. The also called one-step preparation
of supported metal catalysts improved the catalytic
properties of the metallic phase [25,26]. Little atten-
tion, however, has been given in the preparation of
metal supported sol–gel catalysts in which modifica-
tions to the support are induced in the preparation
step. Most papers concerning alumina support modi-
fications are related to the alumina mixed oxides. The
synthesis by the sol–gel method of alumina–zirconia,
alumina–magnesia or alumina–silica as supports have
been recently reported [27–29].
Without doubt, alumina is the support most widely
used in metal-supported catalysts and because of its
importance, alumina is the support largely used in
gas exhaust catalytic converters. In the three way cat-
alysts, the alumina contains CeO2 as support addi-
tive. The role of the CeO2 in the reduction of CO
or NO emissions is confined to an additional oxygen
source [30–32]. The effect of ceria in the activity of
the noble metals, however, has not been clearly de-
fined and metal–Ce interactions are given to explain
the increased activity of the metal [33,34]. The prepa-
ration of CeO2–Al2O3 or La2O3–Al2O3 supports cur-
rently is done by the impregnation of the support with
the rare earth precursors. The effect of the lanthanides
on the catalytic properties is then confined to alumina
surface effects [35].
In the present work with the aim to induce tex-
tural and structural effects in alumina, alumina-gel
was prepared by adding to aluminum tri-sec-butoxide
cerium or lanthanum nitrate. The gelled CeO2–Al2O3
and La2O3–Al2O3 supports were then impreg-
nated with hexachloroplatinic acid to obtain the
metal/doped-alumina catalysts (sol–gel method). In
such a way, the La2O3 or CeO2 role is not limited
to surface effects, since Ce or La could be incor-
porated in the alumina network [36–38]. Structural
defects as well as a good thermal stabilization of the
alumina is expected, depending on the gelling con-
ditions and rare earth metal oxide precursors. The
characterization of the obtained sol–gel supports and
platinum supported catalysts were done by means
of FTIR-pyridine absorption spectroscopy, X-ray
diffraction, MAS-NMR spectroscopy, and by the cat-
alytic test in the iso-propanol dehydration (support
acidity), cyclohexane dehydrogenation (metallic dis-
persion), and n-heptane aromatization (bi-functional
reaction).
2. Experimental
2.1. Catalysts preparation
Sol–gel Al2O3 reference support was prepared by
mixing aluminum tri-sec-butoxide (Aldrich 97%),
Al(OC4H9sec)3 (0.207 mol), in 6.5 mol of absolute
ethanol (Merck) at 60◦C. The reactants were main-
tained under constant stirring and refluxing. Distilled
water (2.94 mol) was added to the mixture and the
gelling reaction was completed. Thereafter, the gel
was dried in air at 70◦C for 24 h (fresh sample). For
characterization various thermal treatments were done
at the temperatures given in the text.
Sol–gel La2O3/Al2O3 support were prepared by
refluxing aluminum tri-sec-butoxide (Aldrich 97%),
Al(OC4H9sec)3 (0.207 mol), in 6.5 mol of abso-
lute ethanol (Merck) at 60◦C. Afterwards 2.94 mol
of distilled water containing the appropriated lan-
thanum nitrate concentration was added to the
aluminum alkoxide solution. The gelling reaction
was accomplished after 1 h under reflux. The gel
was then dried in air at 70◦C. For the synthe-
sis of CeO2–Al2O3 support, the procedure used to
prepare La2O3–Al2O3 samples was accomplished
adding in this case cerium nitrate as cerium pre-
cursor. The La2O3 and CeO2 content in alumina
was 5 wt.%.
The Pt/La2O3–Al2O3 and Pt/CeO2–Al2O3 catalysts
were prepared by impregnation of the sol–gel supports
(calcined at 500◦C in air for 2 h) with an aqueous so-
lution containing the desired amount of hexachloro-
platinic acid (K&K) to obtain 0.5 wt.% of Pt. After
impregnation the catalysts were dried in air at 120◦C
for 12 and then reduced in hydrogen flow for 4 h at
500◦C.
2.2. Specific surface area determination
The specific surface area, pore volume and mean
pore diameter were measured with an ASAP 2000
Micromerictics analyzer, the data were obtained
from the nitrogen adsorption isotherm of the 500◦C
calcined supports. The specific area was calculated
from the BET equation and the mean pore size
diameter from the BJH method. Before the mea-
surements the samples were desorbed at 400◦C
for 2 h.

A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
93
2.3. FTIR-pyridine adsorption
the same reactant system described for iso-propanol
decomposition. The reaction conditions were: gas
flow nitrogen, cyclohexane partial pressure 29 Torr,
reaction temperature 200◦C. For the n-heptane dehy-
drocyclization, the catalytic activity test conditions
were: gas flow hydrogen; n-heptane partial pressure
11 Torr; reaction temperature 500◦C. Before activity
evaluation (50 mg), the samples were reactivated in
the glass reactor (2 ml) with gas flow (nitrogen or
hydrogen) for 2 h at 400◦C.
The calcined supports were characterized in situ
with a Nicolet 170 SX spectrometer using a vacuum
cell (1 × 10−6 Torr) in which atmosphere and tem-
perature are controlled. Self support samples were
prepared by pressing the solid until transparency was
obtained. The acid sites were characterized expos-
ing the pellet to pyridine which is introducing in
the cell through a saturator with flowing nitrogen
at room temperature. Excess pyridine was removed
under vacuum and the temperature of the cell was
then raised to 100◦C and after to 200◦C. The pyridine
absorption spectra were obtained at these desorption
temperatures.
3. Results
3.1. La2O3–Al2O3 and CeO2–Al2O3 characterization
Data on the specific surface area, mean pore size
diameter and pore volume are shown in Table 1. The
Al2O3 sample shows a large specific surface area
(572 m2/g) and a pore size diameter of 7.0 nm. In
La2O3 or CeO2, doped-alumina the BET areas were
459 and 418 m2/g and the pore size diameter 5.5
and 6.0, respectively. A BET area diminution around
20–30% for doped-alumina was then obtained. Typi-
cal isotherms and pore size distribution for alumina
and doped-alumina are shown in Figs. 1 and 2,
respectively.
2.4. X-ray diffraction analysis
In samples annealed at different temperatures, the
crystalline or amorphous structure was obtained at
room temperature by using X-ray powder diffraction
(Siemens 500-D). The reflection diffractometer used a
Cu K␣ radiation, and had a graphite monochromator
in the secondary bean. Intensity data were measured
by step scanning in the 2θ ranges between 2 and 110◦
and between 2 and 70◦, with a 2θ step of 0.02◦ and a
measuring time of 1 s per point.
FTIR-pyridine adsorption was used to determine the
type of acid sites on supports calcined at 500◦C. In
the infrared spectrum of pyridine absorption on acid
catalysts the bands observed around 1590, 1490 and
1450 cm−1 are assigned to the different modes of vi-
bration of the pyridine adsorbed on Lewis sites [39].
The absorption bands of Fig. 3 correspond then to
supports showing Lewis acidity. The relative inten-
sity between these bands differs slightly in alumina
or doped-aluminas. Note that the pyridine absorption
band around 1540 cm−1 assigned to Bronsted acid
sites [40] is not observed in the sol–gel alumina and
doped samples.
Amorphous alumina was observed in the X-ray
diffraction pattern when samples were thermally
treated at temperatures of 500◦C. When the samples
were treated at 1000◦C low crystalline ␥-Al2O3 for-
mation is observed. However, in samples annealed at
1100◦C, the formation of highly crystalline ␣-Al2O3
is observed for alumina sol–gel (Fig. 4). For La2O3
doped-alumina, the ␥-Al2O3 crystalline phase remains
2.5. MAS-NMR spectroscopy
Powder samples were analyzed in a Brucker
ASX-300 NMR spectrometer. The study was done
with the 27Al isotope at a frequency of 78.21 MHz
with a spinning speed of 12 kHz. Chemical shifts
3+
were referred to aluminum in Al(H2O)6 ion.
2.6. Catalytic test
iso-Propanol decomposition activity was per-
formed in a flowing microreactor linked with a gas-
chromatograph. iso-Propanol (Aldrich 99.9%) was
contained in a double H saturator embedded in a
recipient with a mixture of ice and water. Saturated
nitrogen with iso-propanol was passed through the
reactor. The partial pressure of the reactant was
22.4 Torr and the reaction temperature 200◦C. The
cyclohexane dehydrogenation was performed using

94
A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
Table 1
Specific surface area, mean pore diameter and pore volume of Al2O3, doped La2O3–Al2O3 and CeO2–Al2O3 sol–gel prepared supports
Supporta
BET area (m2/g)
Pore diameter (nm)
Pore volume (cm3/g)
Al2O3
La2O3–Al2O3
CeO2–Al2O3
572
459
418
7.0
5.3
6.0
0.88
0.61
0.56
a Calcined in air at 500◦C for 4 h.
present even after thermal treatment at 1100◦C. The
X-ray diffraction pattern for the CeO2 doped-alumina
is different, after thermal treatment at 1000◦C, the
␥-Al2O3 phase is still observed. However, if the
sample is calcined at 1100◦C, the transformation to
␣-Al2O3 and cerianite can be observed. The different
behavior shown by the ceria doped-alumina is indica-
tive that possible segregation of ceria occurs in the
sample.
reported between 55 and 70 ppm for tetrahedral alu-
minum AlIV, around 25–30 ppm for a pentacoordi-
nated AlV (or deformed tetrahedral aluminum), and at
0–11 ppm for octahedrally coordinated AlVI [29,41].
Alumina and CeO2–Al2O3 samples show the corre-
sponding peak for octahedrally coordinated aluminum,
while in La2O3–Al2O3 in addition to the octahedral
aluminum a signal assigned to deformed tetrahedral
coordination is observed.
iso-Propanol decomposition gives propene, acetone
and iso-propylether as main products and the reac-
tion is reported as a test reaction to determine the
acidic–basic character of solids [39]. Comparable ac-
tivities were obtained in doped and non-doped alumina
(Table 2). However, the selectivity pattern notably dif-
fers between them. In Al2O3 support propene (87%)
MAS-NMR 27Al spectra of the samples treated at
1100◦C are shown in Fig. 5. 27Al RMN signals are
Fig. 1. Nitrogen adsorption isotherms of Al2O3, La2O3–Al2O3
and CeO2–Al2O3 sol–gel prepared supports.
Fig. 2. Pore size distribution of Al2O3, La2O3–Al2O3 and
CeO2–Al2O3 sol–gel prepared supports.

A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
95
Fig. 5. MAS-NMR spectra of Al2O3, La2O3–Al2O3 and CeO2–
Al2O3 sol–gel prepared supports.
and iso-propylether (13%) were obtained. On the other
hand, in ceria–alumina propene (65%) and acetone
were the products, while for La2O3–Al2O3 alumina
the only product was propene (100%). The total rate is
related to the total acidity of the catalysts while the se-
lectivity depends of its relative strength [39]. These re-
sults are indicative that in alumina and doped-alumina
sol–gel samples, the total acidity is comparable, while
different acid site strength distribution is expected.
Fig. 3. FTIR-pyridine adsorption spectra of Al2O3, La2O3–Al2O3
and CeO2–Al2O3 sol–gel prepared supports.
3.2. Pt/La2O3–Al2O3 and Pt/CeO2–Al2O3
characterization
The metallic phase of the Pt supported on La and Ce
doped-alumina was characterized by determining its
catalytic properties. In Table 3, the activity of the cata-
lysts for the cyclohexane dehydrogenation is reported.
The Pt/Al2O3 reference catalyst showed an initial ac-
tivity twice that shown by the Pt/La2O3–Al2O3 and
Pt/CeO2–Al2O3 catalysts. This constitutes a support
or metallic dispersion effect in activity. In Table 3, the
Table 2
Activity and selectivity for the iso-propanol dehydration on Al2O3,
doped La2O3–Al2O3 and CeO2–Al2O3 sol–gel prepared supports
Supporta
Specific rate
(× 107) (mol/g s)
Selectivity (%)
Propene Acetone iso-Propyl-
ether
Al2O3
La2O3–Al2O3 1.40
CeO2–Al2O3 1.12
1.18
87
100
65
–
–
35
13
–
Fig. 4. X-ray diffraction spectra of Al2O3, La2O3–Al2O3 and
CeO2–Al2O3 sol–gel prepared supports.
a Calcined in air at 500◦C for 4 h.

96
A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
Table 3
Cyclohexane dehydrogenation on Pt/Al2O3, Pt/La2O3–Al2O3 and
Pt/CeO2–Al2O3 sol–gel prepared catalysts
Catalysts
Conversion (%)
kd × 103 mina
Pt/Al2O3
Pt/La2O3
Pt/CeO2
20
10
10
2.3
0.2
0.7
a Constant deactivation.
deactivation constant for the reaction is reported. It
is calculated from the deactivation rate law C0/C =
1 + kdt. The relative deactivation constant was ob-
tained by plotting the reciprocal of conversion as a
function of time. The appropriate use of this equation
to calculate the deactivation constant is reported and
discussed in previous work [2]. A relative deactiva-
tion constant of 2.3 was obtained for the Pt/Al2O3,
while small values of 0.2 and 0.7 were obtained for
the Pt/La2O3–Al2O3 and Pt/CeO2–Al2O3 catalysts,
respectively. These results show that in platinum
doped alumina the deactivation is strongly inhibited
if compared with non-doped alumina.
Fig. 6. Catalysts evolution on function of time of Pt/Al2O3 and
Pt/CeO2–Al2O3 sol–gel prepared supports during the cyclohexane
dehydrogenation.
toluene/C5–C7 trend was also found on the catalysts.
These results show that the Pt/La2O3–Al2O3 catalyst
has the better activity and selectivity to the desired
products on the n-heptane transformation.
4. Discussion
The n-heptane transformation is an activity test
indicative of the performance of catalysts in naftha
reforming process. In Table 4, the activity and the se-
lectivity for the n-heptane transformation in the plati-
num supported catalysts are reported. High activities
(38 and 34%) for the platinum supported in La and
Ce doped-alumina are observed, while the activity
of Pt supported on non-doped alumina is 16%. Cer-
tainly, the support acidity plays an important role in
the n-heptane transformation. The selectivity pattern
for the reaction is reported as a useful value to com-
pare catalyst behaviors. The products in n-heptane
reforming are C1–C4, C5–C7 fractions and toluene. It
can be seen in Table 4, that the toluene/C1–C4 ratio
are 1.08, 1.26 and 2.08 for Pt/Al2O3, Pt/CeO2-Al2O3
and Pt/La2O3–Al2O3, respectively. A similar ratio
The adsorption isotherms of the three samples
shown in Fig. 1 can be classified as similar, i.e. they
each show a hysteresis loop. The lowest BET areas
of the doped-aluminas (459 and 418 m2/g) compared
with the non-doped one (572 m2/g) are certainly due
to the addition of cerium or lanthanum nitrate to the
aluminum alkoxide during gellation. Using sol–gel
synthesis, the incorporation of lanthanum or cerium
into the alumina network can be expected and hence
a diminution of the pore size diameter can occur. In
Fig. 2, a narrow pore size mesoporous distribution is
observed for alumina (mean pore diameter 7.0 nm),
while a broad pore size distribution and mean pore
size diameters of 5.3 and 6.0 nm is obtained for
the La2O3–Al2O3 and CeO2–Al2O3 supports, re-
spectively. The lowest specific surface areas of the
doped-aluminas may be a side effect due of the
diminution of the pore size diameter. However, it is
not possible with only this information to discriminate
between an effect due to the incorporation of the rare
earth oxides into the alumina network or an effect
due to the deposit of La2O3 or CeO2 into the pore.
To discriminate between such effects X-ray diffrac-
tion studies were carried out. The spectra obtained for
Table 4
n-Heptane dehydrocyclization on Pt/Al2O3, Pt/La2O3–Al2O3 and
Pt/CeO2–Al2O3 sol–gel prepared catalysts
Catalysts
Conversion Selectivity (mol%)
(%)
(a)C1–C4 (b)C5–C7 Toluene a/b
Pt/Al2O3
Pt/La2O3–Al2O3 48
Pt/CeO2–Al2O3 34
16
35
24
30
27
26
32
38
50
38
1.08
2.08
1.28

A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
97
alumina and doped-aluminas annealed to 1000◦C cor-
respond to ␥-Al2O3 structure (Fig. 4). When the sam-
ples were treated at 1100◦C, the crystalline structure
of alumina reference was transformed in ␣-alumina,
while the CeO2–Al2O3 shows two crystalline struc-
tures ␣-Al2O3 and a cerianite. On the other hand,
we can see that for La2O3–Al2O3, ␥-Al2O3 phase is
the only structure present. These results suggest that
in preparing the alumina by the sol–gel method, the
␥-Al2O3 phase is stable up to 1000◦C. The stabiliza-
tion of the ␥-Al2O3 phase at such temperature is not
usual, since frequently a mixture of the ␥-Al2O3 and
␣-Al2O3 phases containing important percentage of
the latter is observed when the alumina was prepared
from soluble salts. The formation of the ␣-Al2O3 and
cerianite structures in the CeO2–Al2O3 support after
treatment at 1100◦C is indicative that in this sample a
segregation of ceria is occurring. Ceria stabilizes the
␥-Al2O3 up to 1000◦C, afterwards the ceria goes out
of the alumina structure and then the ␣-Al2O3 phase
is formed. La2O3 segregation phenomena is not ob-
served in the La2O3–Al2O3 sample. The lanthanide
stabilizes the ␥-Al2O3 crystalline phase even after
thermal treatment at 1100◦C [38].
that the lanthanum is responsible for the modification
of the aluminum environment due to its incorporation
and stabilization into the alumina network.
The incorporation of rare earth oxides into alumina
network and its effects in acidity are shown in Fig. 3.
In general, the doped-aluminas show higher intensity
and definition of the Lewis acid sites determined by
pyridine absorption spectra as compared with alumina
support. The effect of cerium and lanthanum oxides
in the developed acidity is probably due to the modi-
fication of the aluminum coordination as was inferred
from FTIR-pyridine adsorption [35]. FTIR-pyridine
adsorption give information about the nature of the
sites (Bronsted or Lewis), but it is difficult to obtain
quantitative results. Information about the site density
and strength can not be obtained.
To confirm the information obtained from the
pyridine adsorption, the iso-propanol decomposition
was carried out (Table 2). Comparable total activi-
ties are reported for alumina and doped-aluminas.
These results compared with those obtained from
pyridine adsorption are then in good agreement. Total
activity is related to the total acidity of the sample
under study [41]. Fortunately, the selectivity pattern
of the iso-propanol decomposition gives additional
information. The main products obtained from the
iso-propanol decomposition are: (i) the dehydration
product propene, which depends of strong acid sites;
(ii) the iso-propylether, which depends of medium
and weak acid sites and (iii) acetone which is a acid–
base reaction, it is formed only if acid and basic sites
coexist [42].
The products obtained on alumina were propene
and iso-propylether, an indication that in alumina there
are a large number of medium and weak acid sites
[43]. In La2O3–Al2O3 support, the only product was
propene. In this solid, a comparable acidity to that of
alumina (same total rate) is obtained, but a different
acid site strength is presumed. Finally, the products
obtained on CeO2–Al2O3 are propene and acetone.
Ceria is considered as a basic oxide, the formation
of acetone is indicative that ceria segregation even
at 500◦C is occurring. The study of the iso-propanol
decomposition shows that when the sol–gel method is
used for the preparation of alumina or doped-alumina,
we can obtain a large variety of solids. The use of
this method of preparation allows us to make supports
or catalysts in which the acidic–basic properties, the
Since in the three samples treated at 500◦C inci-
pient ␥-Al2O3 crystalline phase is observed by X-ray
diffraction, it is proposed that during the gellation of
alumina the rare earth oxides were incorporated in the
alumina network. The amount of the oxide incorpo-
rate is difficult to evaluate. However, the probability
that La2O3 was incorporate in a larger quantity than
cerium is supported from the X-ray diffraction data.
Unfortunately, the X-ray diffraction technique is not
sensitive enough to identify high dispersed La2O3 on
the alumina surface.
To identify La in the alumina network, the study
of the Al environment by MAS-NMR was made. The
spectra of samples treated at 1100◦C are shown in
Fig. 5. The spectrum of the lanthanum doped-alumina
sample shows two peaks one around −3.6 ppm and the
another one centered at ca. 35 ppm. The first peak is as-
signed to aluminum with octahedral symmetry, while
the second one is assigned to aluminum in tetrahedral
deformed environment, for some authors AlV [41]. On
the other hand, in alumina and ceria doped-alumina
samples, only aluminum in octahedral symmetry is
observed. The formation of tetrahedral aluminum in
the La2O3-doped sample is then an additional support

98
A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
alumina phase, the BET specific area as well as the
reactivity can be controlled from the initial conditions
of synthesis.
catalysts) was determined for our preparations. The
n-heptane dehydrocyclization is a reaction which de-
pends of the support acidity and of the metallic surface
area [16]. The results of Table 4 show that the reac-
tion is higher in platinum supported in doped-aluminas
than in the reference Pt/Al2O3 catalyst in spite of its
greater metallic dispersion (Table 3). We must then
consider that the main factor affecting the activity for
the dehydrocyclization reaction is the support acidity.
Form the FTIR spectra shown in Fig. 3, we can
conclude that few differences in acidity can be ob-
served between samples. The indirect acidity deter-
mined by means of the iso-propanol decomposition is
then useful on this case. The rate was comparable for
the three catalysts, however, the higher selectivity to
propene correspond to the Pt/La2O3–Al2O3 (Table 2).
We can expect then for this catalyst a larger selec-
tivity to toluene and selectivities to this product of
50, 38 and 38% for Pt/La2O3–Al2O3, Pt/CeO2–Al2O3
and Pt/Al2O3, respectively, are reported in Table 4.
The hydrocarbon reforming reactions occurring in a
industrial reactor require of a fine equilibrium be-
tween the support acidity and the exposed metallic
surface area. An increase in acidity must produce high
yields in hydrogenolysis products. Nevertheless, if the
metallic phase (metal dispersion) is present in ade-
quate proportion to the acidity, the dehydrocyclization/
hydrogenolysis selectivity ratio should be increased.
The toluene/C1–C4 ratio obtained for the Pt/La2O3–
Al2O3, Pt/CeO2–Al2O3 and Pt/Al2O3 catalysts was
2.08, 1.28 and 1.08, respectively. These results sup-
port the characterization given to the doped-alumina
supports and platinum supported catalysts.
In noble metal-supported catalysts, the metallic
surface area is of great importance to have a com-
plete characterization of the catalysts. Platinum area
determination in glasses has been reported by hydro-
gen chemisorption [9]. In doped-alumina containing
lanthanum or cerium oxides the specific area of Pd,
Rh and Pt has been determined also by hydrogen
chemisorption [23,44,45]. However, due to phenom-
ena like spill-over, low metal content and probable
metal-support interactions, it has been recommended
to determine the metallic area by means of a “struc-
ture insensitive reaction”. Benzene hydrogenation
and cyclohexane dehydrogenation are classified as
structure insensitive reactions and they are reported
as an indirect determination of the accessible metal
area [44,45]. In our catalysts, the activity for the cy-
clohexane dehydrogenation (metal mono-functional
reaction) was used as a comparative parameter for the
determination of the metallic surface area. The results
of Table 3 show that the activity of Pt/A2O3 catalysts
is twice of that shown by the doped-alumina catalysts.
These results suggest that in the reference catalyst the
metal dispersion is higher. The lower metallic surface
area of the doped-catalysts can be due to the La2O3
and CeO2 oxides producing one or more of the ef-
fects described above, i.e. spill-over, metal support
interaction, etc. A comparison between cyclohexane
aromatization activity and gas chemisorption has not
been carried out for our catalysts. In some attempts,
it was difficult to determine the volume of hydrogen
chemisorbed. The H2 uptake was continuous on func-
tion of the H2 pressure. Hence, the extrapolation to the
y-axis to determine the volume adsorbed by the mono-
layer was not accomplished. Therefore, the metal
particle size was then determined by an independent
TEM (transmission electron microscopy) technique.
The particle size distribution obtained by TEM for
Pt/Al2O3, Pt/La2O3–Al2O3 and Pt/CeO2–Al2O3 cat-
alysts were 1.5, 2.0 and 2.3 nm, respectively, values
which can be considered in good agreement with
those obtained by determining the activity test in the
cyclohexane dehydrogenation (Fig. 6).
5. Conclusions
In preparing Al2O3, La2O3–Al2O3 and CeO2–
Al2O3 supports by the sol–gel method specific sur-
face areas of 572, 459 and 418 m2/g were obtained
on the solids. The FTIR-pyridine adsorption spectra
of the supports shows the formation of Lewis sites
with comparable intensity. Nevertheless, from the se-
lectivity pattern of the iso-propanol decomposition,
different acid site strength distribution on the supports
is proposed.
The ability of doped lanthanum and cerium alu-
mina sol–gel supports for the bi-functional n-heptane
dehydrocyclization (one of the reactions in reforming
Stabilization of the ␥-Al2O3 phase was obtained in
samples annealed up to 1100◦C when the support is

A. Vazquez et al. / Journal of Molecular Catalysis A: Chemical 167 (2001) 91–99
99
La2O3 doped-alumina. In CeO2 doped-alumina, the
␥-Al2O3 phase was transformed totally in ␣-Al2O3
and cerianite, suggesting in this case the segregation
of CeO2. The insertion of La in the alumina network
was inferred from the formation of pentacoordinated
(deformed tetrahedral aluminum) AlV and aluminum
with octahedral symmetry.
[16] R. Gomez, V. Bertin, T. Lopez, G. Ferrat, I. Schifter, J. Mol.
Catal. 109 (1996) 55.
[17] K. Balakrishman, R.D. Gonzalez, Langmuir 10 (1994) 2487.
[18] L. Guczi, L. Borko, Zs. Koppany, F. Mizukami, Catal. Lett.
54 (1998) 33.
[19] L. Guczi, Z. Schay, G. Srefler, F. Mizukami, J. Mol. Catal.
A: Chem. 141 (1999) 177.
[20] R. Gomez, T. Lopez, S. Castillo, R.D. Gonzalez, J. Sol–Gel
Sci. Technol. 1 (1994) 205.
Using the cyclohexane dehydrogenation as an indi-
rect metallic surface area determination, the Pt/Al2O3
catalyst showed the higher dispersion. When the cat-
alysts were evaluated in the bi-functional n-heptane
aromatization, the Pt/La2O3–Al2O3 catalyst had the
better selectivity to the formation of toluene.
[21] S. Castillo, M. Moran-Pineda, R. Gomez, T. Lopez, J. Catal.
172 (1997) 263.
[22] S. Castillo, M. Moran-Pineda, V. Molina, R. Gomez, T. Lopez,
Appl. Catal. B: Environ. 15 (1998) 205.
[23] S. Fuentes, N.E. Bogdanchokova, G. Diaz, M. Pedraza, G.C.
Sandoval, Catal. Lett. 47 (1997) 27.
[24] P. Reyes, G. Pecchi, T. Lopez, R. Gomez, J.L.G. Fierro, Appl.
Catal. B: Environ. 17 (1998) L7.
[25] C.K. Lambert, R.D. Gonzalez, Appl. Catal. A: Gen. 172
(1998) 233.
[26] C.K. Lambert, R.D. Gonzalez, Microporous Mater. 12 (1997)
179.
[27] M. Moran, S. Castillo, T. Lopez, R. Gomez, A. Cordero-
Borboa, O. Novaro, Appl. Catal. B: Environ. 21 (1999) 79.
[28] J.A. Wang, A. Morales, X. Bokhimi, O. Novaro, T. Lopez,
R. Gomez, Chem. Mater. 11 (1999) 308.
Acknowledgements
We acknowledge the CONACYT and IMP-FIES by
the support given to this study.
[29] M. May, M. Asomoza, T. Lopez, R. Gomez, Chem. Mater.
9 (1997) 2395.
References
[30] D. Martin, D. Duprez, J. Phys. Chem. 100 (1996) 9429.
[31] C. Serre, F. Garin, G. Belon, G. Maire, J. Catal. 141 (1993) 1.
[32] C. Serre, F. Garin, G. Belon, G. Maire, J. Catal. 141 (1993) 8.
[33] M. Fernandez-Garcia, E. Gomez-Rebollo, A. Guerrero-Ruiz,
J.C. Conesa, J. Soria, J. Catal. 171 (1997) 146.
[34] R.S. Monteiro, F.B. Noronha, L.C. Dieguez, M. Schmal, Appl.
Catal. A: Gen. 131 (1995) 89.
[35] C. Monterra, G. Magnacca, J. Chem. Soc., Faraday Trans. 92
(1996) 5111.
[36] A. Vazquez, T. Lopez, R. Gomez, X. Bokhimi, A. Morales,
O. Novaro, J. Solid State Chem. 128 (1997) 161.
[37] K. Tadanaga, H. Kobayashi, T. Minami, J. Non-Cryst. Solids
225 (1998) 230.
[38] H. Kobayashi, K. Tadanaga, T. Minami, J. Mater. Chem. 8
(1998) 1241.
[39] J.A. Wang, X. Bokhimi, O. Novaro, T. Lopez, F. Tzompantzi,
R. Gomez, J. Navarrete, M.E. Llanos, E. Lopez-Salinas, J.
Mol. Catal. A: Chem. 137 (1999) 239.
[40] J. Ward, Zeolite chemistry and catalysis, in: J.A. Rabo (Ed.),
Proceedings of the ACS 171, American Chemical Society,
Washington, DC, 1976.
[41] J.A. Wang, X Bokhimi, A. Morales, O. Novaro, T. Lopez, R.
Gomez, J. Phys. Chem. B 103 (1999) 299.
[1] T. Lopez, R. Gomez, in: L.C. Klein (Ed.), Sol–Gel Optics,
Kluwer Academic Publishers, Boston, 1994, 345 pp.
[2] T. López, A. Lopez-Gaona, R. Gomez, Langmuir 6 (1990)
1343.
[3] T. Lopez, P. Bosch, M. Asomoza, R. Gomez, J. Catal. 133
(1992) 247.
[4] T. Lopez, L. Herrera, R. Gomez, W. Zou, K. Robinson, R.D.
Gonzalez, J. Catal. 136 (1992) 621.
[5] M. Asomoza, T. Lopez, A. Zamalloa, E. Garciafirgueroa, R.
Gomez, J. Catal. 138 (1992) 463.
[6] T. Lopez, P. Bosch, J. Navarrete, M. Asomoza, R. Gomez, J.
Sol–Gel Sci. Technol. 1 (1994) 193.
[7] G. Carturan, G. Facchin, V. Gottardi, M. Gugliemi, G.
Mavacio, J. Non-Cryst. Solids 48 (1982) 219.
[8] G. Carturan, G. Facchin, G. Cocco, S. Enzo, G. Navacio, J.
Catal. 76 (1982) 405.
[9] G. Cocco, L. Schifin, G. Strukul, G. Carturan, J. Catal. 65
(1980) 348.
[10] G. Carturan, G. Cocco, L. Schifin, G. Strukul, J. Catal. 65
(980) 359.
[11] M. Asomoza, T. Lopez, R. Gomez, R.D. Gonzalez, Catal.
Today 15 (1992) 547.
[12] P. Bosch, T. Lopez, V.H. Lara, R. Gomez, J. Mol. Catal. 80
(1993) 299.
[42] J.A. Wang, X. Bokhimi, O. Novaro, T. Lopez, R. Gomez, J.
Mol. Catal. A: Chem. 145 (1999) 291.
[13] P. Bosch, T. Lopez, M. Asomoza, R. Gomez, M.A. Cauqui,
J.M. Rodriguez-Izquierdo, Langmuir 11 (1995) 4328.
[14] R. Gomez, V. Bertin, M.A. Ramirez, T. Zamudio, P. Bosch,
T. Lopez, J. Non-Cryst. Solids 147/148 (1992) 748.
[15] R. Gomez, V. Bertin, P. Bosch, T. Lopez, P. Del Angel, I.
Schifter, Catal. Lett. 21 (1993) 309.
[43] P. Berteau, B. Delmon, J.L. Dallons, A. Van Gysel, Appl.
Catal. 70 (1991) 307.
[44] F. Fajardie, J.F. Temepere, G. Djega-Mariadassou, G.
Blanchard, J. Catal. 163 (1996) 77.
[45] E. Rogemond, N. Essayem, R. Frety, V. Perichon, M. Primet,
F. Mathis, J. Catal. 166 (1997) 229.