Synthesis of mixed ZrO2-TiO2 oxides by sol-gel: Microstructural characterization and infrared spectroscopy studies of NOx

Article abstract of DOI:10.1016/j.molcata.2007.11.008

Single ZrO₂, TiO₂ and mixed ZrO₂-TiO₂ oxides with various Zr:Ti ratio (90:10, 50:50 and 10:90) were prepared by the sol-gel method. The samples were characterized by DSC, nitrogen adsorption, SEM, XRD, molecular

Full text of DOI:10.1016/j.molcata.2007.11.008

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
Synthesis of mixed ZrO –TiO oxides by sol–gel: Microstructural
2
2
characterization and infrared spectroscopy studies of NOx
a,∗
a
R. P e´ rez-Hern a´ ndez , D. Mendoza-Anaya ,
a
b
M.E. Fern a´ ndez , A. G o´ mez-Cort e´ s
a
Instituto Nacional de Investigaciones Nucleares, Carr. M e´ xico-Toluca S/N La Marquesa, Ocoyoacac,
Edo. de M e´ xico C.P. 52750, Mexico
b
Instituto de F ´ı sica-UNAM, Apdo. Postal 20-364, M e´ xico 01000 D.F., Mexico
Available online 21 November 2007
Abstract
Single ZrO
2
2 2 2
, TiO and mixed ZrO –TiO oxides with various Zr:Ti ratio (90:10, 50:50 and 10:90) were prepared by the sol–gel method. The
samples were characterized by DSC, nitrogen adsorption, SEM, XRD, molecular simulation and DRIFT. According to the ZrO
experimental results reveal that the TiO loading into ZrO
transition from amorphous to crystalline phase except in the ZT50 xerogel. The optimum concentration of titania into mixed ZrO
obtain high surface area (∼230 m /g) and highly acidic surface was in the equimolar composition. An amorphous XRD pattern was observed in the
ZT50 sample, this behaviour was related to the mixed Zr–O–Ti bond formed, which caused an atomic disorder in the structure of the sample and
this result was confirmed by molecular simulation. All the oxides were constituted by microspheroid particles regardless of the Zr/Ti molar ratio
2
/TiO
2
ratio, the
2
2
had effects on shifts of exothermic peaks towards higher temperature, attributed to
2
–TiO oxides to
2
2
x 2 2
and crystalline phase. Different kinds of surface nitrates (NO , x = 2, 3) species were observed on the mixed ZrO –TiO samples; these species are
formed on Lewis sites and the amount of these species on the surface oxide is intimately related to the support acidity, which in turn depends on
−
the specific area of the samples, as well as, on the Zr/Ti ratio and the Zr–O–Ti bond population. Bridged and bidentate NO
on the Zr-rich samples.
3
species are adsorbed
©
2007 Elsevier B.V. All rights reserved.
Keywords: ZrO2–TiO2 mixed oxides; Acidic properties; Mixed Zr–O–Ti bond; Structural and morphological characterization; NOx adsorption
1
. Introduction
more severe processing conditions. For example, several authors
have reported that the addition of less than 10 mol% of silica to
titania results in significant increments in both surface area and
pore volume and delays the crystallization of titania to higher
temperatures [1–5]. These effects have beenrelatedto the silica’s
ability to impede titania’s normal sintering processes through
formation of Ti–O–Si “hetero-linkages” [6]. Solid acid catalysts
have the additional advantage that the nature of the active sites
are known and may be defined by the presence of surface pro-
tons generating Bronsted acidity or co-ordinately unsaturated
cationic centers, i.e. Lewis acid sites [7]. Therefore, much atten-
tion has been focussed on the development of mixed oxides
catalysts [8]. ZrO2–TiO2 system are strong solid acids showing
catalytic activity in different acid catalyzed reactions, such as
isomerization and cracking alkanes, hydration and polymerisa-
tion of alkenes [9,10]. Other important issues for these systems
are related with fast and direct environmental impact, where
catalysts like hazardous liquid acid and base catalysts, such as
HF, H2SO4, HCl, Ca(OH)2, etc. used by several oil refineries
Sol–gel chemistry has been applied to many mixed oxide
systems for preparation of glasses, coatings, and other ceramic
materials. This technique provides the ability to control
molecular-scale mixing and the ability to form the M1–O–M2
bonds (where M1 and M2 denote two different cations). Combi-
nationofdissimilaroxidesalsopresentsthepotentialforcreating
surface acid sites that are not present in either of the pure compo-
nents. The reasons for these mixed oxides are they play a central
role in chemical and petrochemical processing as catalysts and
support for catalytic active species. A major component oxide
can be stabilized against degradation of its textural properties
upon heat treatment by introduction of even smaller amount of
a minor component, thereby allowing the material to be used at
∗
Corresponding author. Tel.: +52 55 53 29 72 39; fax: +52 55 53 29 72 40.
E-mail address: pehr@nuclear.inin.mx (R. P e´ rez-Hern a´ ndez).
1
381-1169/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2007.11.008

R. P e´ rez-Hern a´ ndez et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
201
and chemical industries can be replaced by mixed oxides. The
breakthrough came with the development of solid acid and base
catalysts [11] is gradually gaining importance in the catalytic
field and it is also possible to develop more. On the other hand,
it has been shown that ZrO2–TiO2 system can be satisfactorily
used as catalytic support for the selective catalytic reduction
specimen holder, obtaining the images with the backscattering
electron signal with acceleration voltage of 20 kV. Powder X-
ray diffraction patterns were recorded with a Siemens D-5000
diffractometer using Cu K␣ (λ = 0.15406 nm) all samples were
◦
scanned in the range between 2θ = 15–70 . Diffuse reflectance
infrared fourier transformed spectra (DRIFT) were collected
on a Nicolet-Nexus 670 FT-IR, E.S.P. FT-IR spectrophotometer
(SCR)reactionoftheNOwithhydrocarbons, because, themixed
−
1
oxide displays unusual acid properties and very high-thermal
resistance respecting to its individual ZrO2 and TiO2 compo-
nents [12–16] and selective catalysts by the incorporation of a
transition metal [15,16].
The aim of the current work was to investigate the additive
effect of TiO2 to ZrO2 on the structure, morphology and textural
properties studied by TGA/DSC, XRD, SEM, N2 adsorption,
DRIFT and molecular simulation, the last was used to infer the
formation of Zr–O–Ti bond in the mixed ZrO2–TiO2 oxide.
working at a resolution of 4 cm , equipped with a high tem-
perature environmentally controlled cell (Spectra Tech). About
0.030 g of the powdered oxide was packed in a sample holder
and pretreated in situ in the DRIFT cell under a H2 flow, then the
◦
◦
temperature was increased at a rate of 10 C/min up to 500 C
and kept for 1 h, after that the oxide was cleaned for 30 min
◦
with He flow. The sample was subsequently cooled to 100 C in
He prior to exposure to the mixture of NO (2500 ppm) and 2%
O2 gas mixture, after 15 min of passing the mixture, the sample
was purged with He and cooled at room temperature (RT). The
DRIFT spectra were collected after 15 min at RT collecting 256
scans.
2
. Experimental
2
.1. Single oxides
4. Simulation method
ZrO2 and TiO2 were prepared by sol–gel method from zirco-
nium(IV) propoxide (Fluka) or titanium(IV) propoxide (Fluka)
The model for ZrO2 and TiO2 interface (ZT50) were built
as a precursor in n-propanol (Aldrich) solution and basic catalyst
in the (1 0 1) directions according to the crystalline data of
˚ ˚
the tetragonal unit cell of the ZrO2 (a = 3.570 A, c = 5.16 A)
(
NH4OH) (Fluka) with constant stirring. The mixed ZrO2–TiO2
˚ ˚
and TiO2 (a = 3.78 A, c = 9.51 A). The interface was grown in
oxides (10, 50, 90 mol% of TiO2 in samples) were prepared
as follows: mixtures of the alkoxide precursors were used as
starting material at appropriate composition in n-propanol and
NH4OH with constant stirring. After that, the solution was pro-
cessed at reflux temperature and the water was added drop wise
to the solution at this temperature. The reflux was continued for
˚
˚
[0 1 0] direction with 3 A of thickness for “a” and 10 A for “b”
and “c” parameters in the cell respectively for both oxides, in
order to determine the electronic charge density and stability
of Zr–O–Ti bond. Quantum mechanics calculations were made
2
through the Cerius program. The calculus is based on density
1
h and the resultant mixture was cooled at room temperature and
functional theory (DFT) which uses the local density approxi-
mation (LDA) and generalized gradient approximation (GGA)
to achieve geometrical optimization and determine the charge
density [17]. The simulated XRD pattern of the model struc-
ture after the geometry optimization was acquired by Bragg
aged for 24 h. The residual liquid was removed by decanting and
washed two times with deionised water.
The molar ratio used for the synthesis was propoxide/
alcohol/H2O/NH4OH = 1:4:4:0.33. The xerogel was obtained
after heating in an oven at 100 C for 24 h in static air using a
temperature ramp of 5 C/min. Then the xerogels were calcined
at 100 C for 1 h next the temperature was raised to 500 C and
kept for 5 h. The labelling of different oxides will be referred
as ZrO2, TiO2 and ZTn, where Z = ZrO2, T = TiO2, n = 10, 50,
◦
2
Law Theory with Cerius program [18]. The diffraction pattern
was obtained by simulating a Siemens D-5000 diffractometer
◦
◦
◦
using an X-ray source of copper, with a λ = 0.1540 nm and the
◦
◦
◦
scanned 2θ was from 15 to 80 . The calculations were made in
a supercomputer Origin 2000 of Silicon Graphics with 8 R10000
processors.
9
0 mol% of TiO2 in mixed oxide, respectively.
3
. Characterization
5. Results and discussion
DSC of xerogel was performed in a DSC-10 from TA-
5.1. Thermal analysis (TGA/DSC)
◦
Instruments under N2 atmosphere in the range of 20–600 C at a
◦
heating rate of 10 C/min. Total surface area was calculated by
The TGA curves of the xerogels, (not presented) showed a
similar behaviour in their weight loss. In general, the weight
loss of the samples was about 10% and it was attributed to
the elimination of the physically adsorbed water, physically and
chemically adsorbed alcohol and residual organic material com-
ing from the synthesis. Fig. 1 shows DSC profile corresponding
to the xerogels, all of them have an endothermic peak from 30
the BET method from N2 adsorption by the single-point method
using a 30% N2/He gas mixture, recorded at the temperature
of liquid nitrogen in a RIG-100 multitask characterization unit
◦
from ISRI. Samples were first degassed at 150 C for 1.5 h to
eliminate moisture. The morphology and chemical composition
analysis of the samples were performed in a Philips XL-30 SEM
instrument, fitted with energy dispersive X-ray spectrometer
◦
to 250 C and, it corresponds to the removal of the physically
(
EDS). Before the analysis, samples were fixed on the specimen
adsorbed water and alcohol [15,19,20]. The exothermic peak
at 424 C observed on the zirconia xerogel, was assigned to
◦
holder with an aluminium tape and mounted on an aluminium

2
02
R. P e´ rez-Hern a´ ndez et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
◦
Fig. 2. XRD patterns of ZrO2–TiO2 oxides calcined at 500 C. *: monoclinic,
◦
:
tetragonal phases from zirconia and +: anatase-TiO2.
◦
TiO2) phases were observed at 535 and 561 C, respectively.
(
However, the ZT50 sample did not show an exothermic peak
related to the phase transition, from an amorphous phase to a
◦
crystalline one up to 600 C. These results demonstrated that
the presence of the titania on zirconia in the mixed oxide delays
the formation of ZrTiO4 compound. An exothermic peak was
◦
reported on ZrO2–TiO2 samples at 600–700 C and it was asso-
ciated withthe crystallizationof amorphous material [21,23]. On
the other hand, in the Zr-rich samples an exothermic peak related
with the burning of organic material was not observed, this could
be associated to the particular set of preparation conditions used
in this work.
5.2. Crystalline phases (XRD)
Fig. 2 shows the XRD patterns of the TiO2 and ZrO2 sam-
ples and mixed ZrO2–TiO2 oxides. The XRD pattern of the pure
ZrO2 sample shows a mixture of two crystalline structures; in
this case a large predominance of the tetragonal phase and a
small amount of monoclinic phase of zirconia were present.
While in XRD pattern of the bare TiO2, the crystalline struc-
ture was consistent with the anatase phase. The addition of TiO2
to ZrO2 diminishes the intensity of the diffraction peaks and it
was interpreted as a lose of the crystallinity in the mixed oxides.
ZrO2-rich oxide showed the same crystalline phases identified in
the bare ZrO2. Meanwhile, for the TiO2-rich sample, the anatase
was the unique phase observed. The XRD pattern of the mixed
ZT50 oxide was consistent with an amorphous material, even
Fig. 1. Differential scanning calorimetric curves of ZrO2–TiO2 xerogels.
a change of phase from an amorphous material to crystalline
tetragonal zirconia [15,21,22]. For the TiO2 xerogel, this peak
◦
when the solid was calcined until 600 C during 8 h. This result
◦
was observed at 423 C and corresponds to the crystallization of
is in agreement with DSC results for this oxide, which did not
show an exothermic peak related with the phase transformation
from amorphous to crystalline material. An amorphous material
was reported [15,23] in the ZrO2–TiO2 samples independently
of the route of the synthesis in the mixed oxides. The amorphous
ZT50 sample could be favoured by decreasing the particle size of
the material [15]. However, crystallization of amorphous ZT50
the anatase phase [15,19,20], another peak was observed close to
◦
3
00 C and it could be attributed to the decomposition and burn-
ing of organic groups [15,19,22]. The increase in the titania or
zirconia content on the mixed ZrO2–TiO2 xerogels, cause a shift
on the exothermic peaks towards higher temperature compared
to the single xerogels. These peaks are attributed to the crys-
tallization of the amorphous material. On the ZT10 and ZT90
xerogels the crystallization of the tetragonal (ZrO2) and anatase
◦
oxide to ZrTiO4 phase is reported above 600 C [21,23–25]
regardless the method of synthesis.

R. P e´ rez-Hern a´ ndez et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
203
Fig. 3. (a) Surface of initial model of mixed ZrO2–TiO2 bond; (b) surface model after de relaxation process; (c) pattern slice of electronic charge density from (b);
d) XRD pattern from (b).
(
5
.3. Analysis of the Zr–O–Ti bond by quantum mechanics
They suggested the presence of the mixed Zr–O–Ti links in the
dry and calcined samples up to 600 C in which the solids were
◦
Fig. 3(a–d) shows the molecular simulation results of the
amorphous.
mixed ZT50 oxide. Fig. 3a shows the theoretical model of the
interface with the same Zr–O and Ti–O bonds before the relax-
ation process used to determine if the Zr–O–Ti bond is stable
on the mixed oxides and how this bond affects the crystalline
structure of the samples. After the relaxation process the origi-
nal symmetry of the model was distorted as is shown in Fig. 3b.
Fig. 3c shows the map of the electron charge density (ECD)
of pattern slice from the model after relaxation process. From
this map it can be seen that the ECD is distributed around the
mixed bond, which confirms the existence of the mixed Zr–O–Ti
link. However, the map of ECD showed clearly that the charge
density is concentrated on the Ti–O bond, causing a deficiency
5.4. Textural properties (BET, SEM)
2
The surface area of the two pure oxides is 10 and 29 m /g
for ZrO and TiO , respectively. In the mixed oxides the sur-
2
2
face area was increased proportionally to the titania loading to
zirconia until an equimolar composition was reached in com-
parison with the single oxides. For ZT10 and ZT90 oxides the
2
surface area is 71 and 80 m /g, respectively. For the sample
with equimolar composition the surface area reached a maxi-
2
mum up to 232 m /g. It is clear that the ZrO –TiO samples
2
2
with higher surface area also show the high-temperature tran-
sition phase (Fig. 1) and low crystallinity. Fig. 4 shows the
+
4
of charge in the Zr cation; which could probably act as an
adsorption centre for the catalytic reaction. Thus the generation
of mixed Zr–O–Ti bond, leads to electron transfer from Zr–O
bond to Ti–O link and causes the new distribution of charge on
the mixed link. On the other hand, the XRD pattern of the ZT50
model obtained after the relaxation process was consistent with
an amorphous material (Fig. 3d). Therefore, the atomic disorder
of the mixed Zr–O–Ti bond in the lattice structure is caused by
◦
SEM image corresponding to ZrO and TiO calcined at 500 C.
2
2
It can be seen that both samples are conformed by particles
with spherical tendency and, a zirconia particle size of 0.3 ␮m,
while for TiO it was 0.9 ␮m. A spherical tendency was also
2
observed for mixed oxides, but the particle size found was
between 0.3–0.9 ␮m.
+
4
+4
the incorporation of the Zr or Ti cations into the host matrix
ZrO2 or TiO2) producing an amorphous material. This result
5.5. NO and O co-adsorption over ZrO –TiO
2
2
2
(
confirms that the mixed Zr–O–Ti bonds are responsible for the
lose of the crystallinity and corroborate our experimental XRD
result for the ZT50 oxide. Xu et al. [23] studied EXAFS and
XANES of the ZrTiO4 samples synthesized by different routes.
2 2
In situ DRIFT spectra of the single and mixed ZrO –TiO
oxides exposed to a NO + O gas mixture were recorded at room
2
temperature with the aim to evaluate the influence of the sur-
face acidity on the type and amount of different surface species.

2
04
R. P e´ rez-Hern a´ ndez et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
Fig. 5. DRIFT spectra of NO and O2 co-adsorption at room temperature on
mixed ZrO2–TiO2 oxides.
◦
Fig. 4. SEM micrographs of ZrO2–TiO2 oxide calcined at 500 C. (a) ZrO2 and
b) TiO2.
becomes a broad band in the ZT50 sample, for the TiO2-rich
(
−1
samples this band shift to 1493 cm . Therefore, we can affirm
−
1
that the species adsorbed at 1628 and 1526 cm prefer the zir-
−1
Several authors have reported that the role of the acid sites in
the samples is to promote the NO oxidation step [26–29] during
SCR reaction of NO with hydrocarbons. Fig. 5 shows an infrared
conia entities, while the species at 1582 cm are adsorbed on
titania entities on the ZrO2–TiO2 oxides.
The assignments of absorption bands described above are
consistent with those reported previously on ZrO2 or TiO2 and
are classified as nitrite/nitrate species [14,30–34]. Free nitrite ion
−
1
spectra in the 2000–1000 cm region of the ZrO2–TiO2 oxides.
−
1
In general, the absorption bands in the 1350–1200 cm range
are the same independently of the ZrO2/TiO2 ratio but the inten-
sity is influenced by the percentage of each one of the oxides in
the mixed samples. The bands at 1286 and 1246 cm increase
their intensity when titania is loaded to the zirconia until the
composition reached ZT50; however, the later band is larger
than the former. For the ZT90 oxide these bands are very weak
and again they increase for the single TiO2. It is clear that
−
−1
(NO2 ) has been reported at 1250 (νas(NO2)) cm [35]. Had-
jiivanov [36], Kantcheva [32] and Watson et al. [37] attributed
−
1
−1 −
to bidentate nitrate (NO3 ) species
the band at 1240 cm
formed on ZrO2, Mn/TiO2 and titanium oxide surfaces, respec-
−
1
tively. Whereas, the band near 1286 cm has been attributed to
−
the monodentate nitrates (NO3 ) on TiO2 [32], nitro species
on ZrO2 [36] and νas(NO2) on CuSO4/ZrO2 [33]. For the
−
1
the band at 1246 cm in the titania is weaker than in zirco-
nia.
Pt/ZrO2–TiO2 catalyststhisbandwasattributedtoNOstretching
−1
modes of bidentate nitrate [16]. The bands at 1320–1340 cm
The bands in the 1650–1400 cm 1 range are shifted to lower
−
were assigned to nitro species in the form of M–NO2 [35,38]
wavenumbers as a function of titania loading into zirconia. The
and chelated nitro (νs(NO2)) [33]. In this work the bands at 1286
−
1
−1
positions of the peaks at 1628, 1589 and 1526 cm observed on
and 1246 cm are assigned to nitro and nitrate species adsorbed
−
1
+4
+4
the ZrO2-rich samples are shifted to 1613, 1582 and 1493 cm
,
on Zr or Ti cations from the bare oxides respectively, these
bands increase their intensity as TiO2 is loaded to ZrO2 until an
equimolar composition, as a result of the generation of the mixed
Zr–O–Ti bonds which promote the absorption centres for those
respectively in the TiO2-rich samples. Besides, it is evident that
as TiO2 is loaded to ZrO2 in the mixed oxides, the bands at
−
1
1
628, 1589 and 1526 cm increase their intensity until ZT50
−
1
sample, after this, the intensity diminishes again. The specie
species. The nitrate species observed at 1246 cm is preferably
−
1
+4
adsorbed at 1628 cm on ZrO2 is practically the same until the
adsorbed on Zr cation; because on ZT90 and TiO2 oxides this
−
1
−
composition is reached ZT50. However, the band at 1589 cm
NO 3 is not stable (their intensity is very low).
−
1
−1
shifts to 1582 cm from ZT10 until TiO2 samples. Although
The absorption bands in the 1650–1400 cm range are typ-
−
1
the band at 1526 cm observed on the ZrO2 and ZT10 samples
ical of different kinds of surface nitrate species [30,38,39].

R. P e´ rez-Hern a´ ndez et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
205
Table 1
Possible structures of surface NO3 and NO2 species on ZrO2–TiO2 oxides
TiO surface. Nitrate species can be formed by disproportiona-
tion of NO2 (produced by oxidation of NO with O2) [39,43,44]
2
−
−
Sample
Frequency
Possible assignment
at the Lewis acid–base pairs [30]. Tanabe et al. [45] predict the
Lewis acid for the zirconia-rich mixed oxide; while in the titania-
rich mixed oxide predict the Br o¨ nsted acid. Lahousse et al. [46]
reported for the mixed ZrO2(23%)–TiO2(77%) oxide the exis-
tence of the Br o¨ nsted acidity. These results could explain the
phenomena observed in the spectra of the ZT90 oxide which
clearly indicates that the oxidation of NO by O2 is irrelevant,
for that on this sample the Lewis acidity is extremely low. Yung
et al. [47] characterized by DRIFT technique at the Co/ZrO2
catalyst and found that the bidentate and bridging nitrates still
−1
(
cm
)
+
4
−
−
1628
1589
1526
1346
1286
1246
Zr –NO3 (s-bridged)
+
4
Zr –NO3 (s-bidentate)
+
4
Zr –NO2 (s-bridge-nitro)
ZrO2
+
4
Zr –NO2 (w-nitro)
+
4
Zr –NO2 (s-nitro)
+
4
−
Zr –NO3 (s-bidentate)
1628
1582
1526
1346
1286
Zr 4–NO3 (s-bridged)
+ −
+
4
−
Ti –NO3 (s-bidentate)
+
4
Zr –NO2 (s-bridge-nitro)
ZT10
ZT50
ZT90
+4
+4
◦
remain up to 350 C. Ito et al. [48] observed by DRIFT that the
Zr and Ti –NO2 (w-nitro)
+
4
Zr –NO2 (s-nitro) and
nitrate (NO ) specie is strongly adsorbed into ZrO –TiO sam-
3
2
2
+
4
−
−
−
Ti –NO3 (w-bidentate)
ples, they concluded that these materials could be used as NOx
adsorbent at the range of composition of (n)TiO2–(1−n)ZrO2
between n = 0.25–0.50. Takahashi et al. [49] reported high NOx
+
4
1246
Zr –NO3 (s-bidentate) and
+
4
Ti –NO3 (w-bidentate)
Zr 4–NO3 (s-bridged)
+
−
−
1626
1582
1526
1346
1286
◦
◦
storage at temperatures >500 C < 600 C on the ZrO2–TiO2
samples with 70% and 80% of ZrO2.
+
4
Ti –NO3 (s-bidentate)
+
4
Zr –NO2 (s-bridge)
Zr+4 or Ti+4–NO2 (w-nitro)
The above IR bands observed on the bare ZrO2 and TiO2
+
4
−
−
Zr –NO2 (s-nitro) and NO3
bidentate)
and ZrO2–TiO2 samples are assigned to inorganic NOx (x = 2,
(
3) species formed on Lewis acid sites. The amount of these
species on the surface oxides is intimately related to support
acidity, which in turn depends on the specific area of the sup-
port [14], as well as on the enlargement of the mixed Zr–O–Ti
bond population and the Zr/Ti ratio in the range of ZT(100–n)
between n = 90–10. It could be possible that when the elec-
trons of the oxygen of the Zr–O bond from Zr–O–Ti link is
+
4
−
−
1
246
613
Zr –NO3 (s-bidentate) and
+
4
Ti –NO3 (w-bidentate)
+
4
−
−
−
1
Ti –NO3 (w-bidentate) or
+
4
Ti –NO3 (w-bridged)
+
4
1
1
582
493
Ti –NO3 (w-bidentate)
+4
Ti –NO2 (w-bridge
nitro-nitro)
Zr and Ti+4–NO2 (w-nitro)
+
4
+4
1
1
346
286
strongly attracted by Ti ion, it causes a delocalization of ECD
+
4
Zr –NO2 (w-nitro) and
+4
−
in the Zr ion, (Fig. 3c), consequently NOx bond is formed on
+
4
−
−
−
Ti -NO3 (w-monodentate)
+4
Zr ion. Because of this the selective adsorption of bidentate or
bridged NO3 and nitro species reached a maximum between
628 and 1613, 1286 and 1246 cm bands for ZrO2, ZT10
and ZT50 samples. Table 1, shows the possible NOx (x = 2, 3)
+
4
1
246
613
Zr –NO3 (w-bidentate) and
−
+
4
Ti –NO3 (w-bidentate)
−
1
1
+
4
−
−
−
1
Ti –NO3 (bidentate) or
−
+
4
Ti –NO3 (bridged)
+
4
2 2
species adsorbed on the mixed ZrO –TiO oxides.
1582
1493
1346
1286
1246
Ti –NO3 (bidentate)
TiO2
+
4
Ti –NO2 (bridge nitro-nitro)
+
4
Ti –NO2 (w-nitro)
6. Conclusions
+
4
−
−
Ti –NO3 (s-monodentate)
+
4
Ti –NO3 (w-bidentate)
ZrO2–TiO2 mixed oxides were synthesized by sol–gel
S = strong, w = weak.
method and characterized by DSC, XRD, SEM, N adsorption,
2
molecular simulation and DRIFT. DSC showed a shift of the
exothermic peak towards high temperature, attributed to tran-
sition from the amorphous to crystalline phase except on the
ZT50 xerogel. Surface area of the oxides was modified depend-
ing of the molar ratio of the ZrO2/TiO2. Single and mixed oxides
showed morphologies constituted by microspheroids aggregates
independently of the ZrO2/TiO2 molar ratio and crystalline
structure, indicating a homogeneous kinetics of nucleation of
the microspheroids during the synthesis. XRD results showed
that the crystalline phases of the mixed oxides diminish progres-
sively as the titania is loaded to the zirconia until the equimolar
composition. On Zr-rich oxides the tetragonal and monoclinic
phases were observed, while anatasa phase was identified on
titania-rich oxides. In the mixed oxide with equimolar compo-
sition an amorphous material was obtained. In any case, we
did not observe a segregation of phases on the mixed oxides
indicating a high solubility of TiO2 or ZrO2 in the host oxide.
−
1
The band at 1640–1630 (ν(N O)) cm has been reported as
−
−
−1
−1
bridged NO3 , bidentate NO3 at 1615–1550 (ν(N O)) cm
and bridged nitro (NO2 ) at 1545–1530 (νas(NO2 )) cm
−
−
[
14,33,34,40] on catalysts using ZrO2 as a support. Delahay et
−
1
al. [40] assigned the band at 1590 cm to the nitrate species
adsorbed on ZrO2. Kantcheva [32] has reported formation
−
−1
of bidentate nitrates (NO3 ) at 1615 and 1580 cm , while,
adsorbed NO2 was tentatively assigned at 1610 cm on TiO2
and bridged nitro-nitrite (NO2 ) at 1487 cm on MnOx/TiO2.
−
1
−
−1
−
1
Watson and Ozkan [37] assigned the band at 1618 cm to the
−
1
bridged nitrate and the band at 1580 cm to the bidentate nitrate
specie formed on reduced TiO2. When NO was adsorbed on
−
1
TiO2, bands at 1616 and 1580 cm were observed and assigned
to nitrate species [41]. Chien et al. [42] reported the formation
of bidentate NO3 species at 1607 and 1584 cm adsorbed on
−
−1

2
06
R. P e´ rez-Hern a´ ndez et al. / Journal of Molecular Catalysis A: Chemical 281 (2008) 200–206
Zr–O–Ti mixed bond promotes the lose of crystalline phase on
the ZrO2–TiO2 mixed oxides, which reaches their maximum
for the sample with equimolar composition as has been demon-
strated by molecular simulation, as well as the delocalization
[16] M. Machida, S. Ikeda, D. Kurogi, T. Kijima, Appl. Catal. B: Environ. 35
2001) 107–116.
(
[
17] M.E. Fern a´ ndez, C. Zorrilla-Cangas, R. Garc ´ı a-Garc ´ı a, J.A. Ascencio, J.
Reyes-Gasga, Acta Cryst. B B 59 (2003) 175–181.
[
18] Molecular Simulation Inc., 1997, Cerius2. Molecular and Quantum
Mechanics and Analytical Instruments, San Diego, California.
+
4
of electrons in Zr cation. The modification in the electron
charge density of the Zr–O–Ti link can be acting as adsorp-
[19] J.P. Perdew and Y. Wang, Phys. Rev. B, 45, 13244.
´
−
[
20] I.A. Montoya, T. Viveros, J.M. Domınguez, L.A. Channels, I. Schifter,
Catal. Lett. 15 (1992) 207–217.
tion centres for NOx . On the ZrO2–TiO2 samples different
kinds of the surface nitrates are observed and these are formed
on the Lewis sites. The amount of these species on the surface
oxide is intimately related to the support acidity, which depends
on the specific area of the ZrO2–TiO2. On the mixed Zr–O–Ti
bond population and on Zr/Ti ratio in the range of (n)ZrO2-
[21] Q. Xu, M.A. Anderson, J. Am. Ceram. Soc. 76 (8) (1993) 2093–2097.
[
22] A. Knell, P. Barnicke n˜ , A. Baiker, A. Wokaun, J. Catal. 137(1992)306–321.
[23] J. Xu, C. Lind, A.P. Wilkinson, S. Pattanaik, Chem. Mater. 12 (2000)
3347–3355.
[
24] M. Daturi, A. Cremona, F. Milella, G. Busca, E. Vogna, J. Eur. Cer. Soc.
18 (1998) 1079–1087.
(
100–n)TiO2 when ≤90 n ≥ 50. On ZT10 sample the presence
[
[
[
25] H. Zou, Y.S. Lin, Appl. Catal. A: Gen. 265 (2004) 35–42.
26] Y. Ly, J.N. Armor, J. Catal. 145 (1994) 388–399.
27] J.O. Petunchi, W.K. Hall, Appl. Catal. B 2 (1993) L17–L26.
−
of NOx species is larger than in ZT90 oxide. However, the Zr-
rich samples preferred the adsorption of the bridged NO3 and
−
nitro species.
[28] C.J. Loughran, D.E. Resasco, Appl. Catal. B 7 (1995) 113–126.
[
[
[
29] E. Kikuchi, K. Yogo, Catal. Today 22 (1994) 73–86.
30] K.I. Hadjivanov, Catal. Rev. Sci. Eng. 42 (1/2) (2000) 71–144.
31] Y. Kintaichi, M. Haneda, M. Inaba, H. Hamafa, Catal. Lett. 48 (1997)
Acknowledgements
1
21–127.
Thanks to I.Q. Leticia Carapia for technical support and to
the projects ININ-CA-711 and CONACyT (M e´ xico) J-48924
for financial support.
[
32] M. Kantcheva, J. Catal. 204 (2001) 479–494.
[33] M. Kantcheva, Appl. Catal. B 42 (2003) 89–109.
34] M. Kantcheva, A.S. Vakkasoglu, J. Catal. 223 (2004) 352–363.
35] K. Nakamoto, Infrared an Raman Spectra of Inorganic and Coordination
Compounds, Part B, 5th ed., Wiley, New York, 1997.
[
[
References
[
[
[
[
[
36] K. Hadjiivanov, Catal Lett. 68 (2000) 157–161.
37] J.M. Watson, U.S. Ozkan, J. Catal. 210 (2002) 295–312.
38] S.-J. Huang, A.B. Walters, M.A. Vannice, J. Catal. 192 (2000) 29–47.
39] M. Kantcheva, E.Z. Ciftlikli, J. Phys. Chem. B 106 (2002) 3941.
40] G. Delahay, B. Coq, E. Ensuque, F. Figu e´ ras, Langmuir 13 (1997)
[
[
1] J.B. Johnston, S.T. Miller, E.I. Ko, J. Catal. 150 (1994) 311–320.
2] J.R. Sohn, H.J. Jang, M.Y. Park, E.H. Park, J. Mol. Catal. 93 (1994)
1
49–167.
[
[
3] Z. Liu, J. Tabora, R.J. Davis, J. Catal. 149 (1994) 117–126.
4] C. Ingemar-Odenbrand, J.G.M. Brandin, G. Busca, J. Catal. 135 (1992)
5588–5592.
[
[
41] K. Hadjiivanov, H. Kn o¨ zinger, Phys. Chem. 2 (2002) 2803.
42] S.-H. Chien, M.-C. Kuo, C.-H. Lu, K.-N. Lu, Catal. Today 97 (2004)
5
05–517.
[
[
[
[
[
5] Ji-jian Cheng, Dong-wei Wang, J. Non-Cryst. Solids 100 (1988) 288–291.
6] K. Karmakar, D. Ganculi, Indian J. Technol. 25 (1987) 282.
7] A. Corma, Chem. Rev. 95 (1995) 559–614.
8] H. Tanaka, M. Boulinguiez, M. Vrinat, Catal. Today 29 (1996) 209–213.
9] K. Tanabe, M. Misono, Y. Ono, H. Hattori, New Solids and Bases,
Kodansha–Elsevier, Tokyo, 1989, p. 199.
1
21–127.
[
[
[
[
43] M.M. Kantcheva, V.P. Bushev, K.I. Hadjiivanov, J. Chem. Soc., Faraday
Trans. 88 (1992) 3087–3089.
44] K. Hadjiivanov, V. Bushev, M. Kantcheva, D. Klissurski, Langmuir 10
(
1994) 464–471.
45] K. Tanabe, T. Sumiyoshi, K. Shibata, T. Kiyoura, J. Kitagawa, Bull. Chem.
Soc. Jpn. 47 (5) (1974) 1064–1066.
46] C. Lahousse, A. Aboulayt, F. Maug e´ , J. Bachelier, J.C. Lavalley, J. Mol.
Catal. 84 (1993) 283–297.
47] M.M. Yung, E.M. Holmgreen, U.S. Ozkan, J. Catal. 247 (2007) 356–367.
48] K. Ito, S. Kakino, K. Ikeue, M. Machida, Appl. Catal. B: Environ. 74 (2007)
[
[
[
10] T. Yamaguchi, Appl. Catal. 61 (1990) 1–25.
11] K. Arata, Adv. Catal. 37 (1990) 165.
12] M. Haneda, Y. Kintaichi, M. Inaba, H. Hamada, Bull. Chem. Soc. Jpn. 70
(
1997) 499–508.
[
[
[
[
13] E.P. Reddy, T.C. Rojas, A. Fern a´ ndez, Langmuir 16 (2000) 4217–4221.
14] R. Mariscal, S. Rojas, A. G o´ mez-Cort e´ s, G. D ´ı az, R. P e´ rez, J.L.G. Fierro,
Catal. Today 75 (2002) 385–391.
15] R. P e´ rez-Hern a´ ndez, A. G o´ mez-Cort e´ s, J. Arenas-Alatorre, S. Rojas, R.
Mariscal, J.L.G. Fierro, G. D ´ı az, Catal. Today 107/108 (2005) 149–156.
137–143.
[
49] N. Takahashi, A. Suda, I. Hachisuka, M. Sugiura, H. Sobukawa, H. Shinjoh,
Appl. Catal. B: Environ. 72 (2007) 187–195.
[