Zirconium doped mesoporous silica catalysts for dehydration of glycerol to high added-value products

Article abstract of DOI:10.1016/j.apcata.2012.05.015

A series of zirconium doped mesoporous silica, with different Si/Zr molar ratios, has been synthesized and tested in the gas-phase dehydration of glycerol. The surface characterization of these solids by using NH ₃-TPD and pyridine adsorption coupled to FTIR spectroscopy has revealed the existence of well dispersed acid sites, mainly of Lewis type, associated to Zr(IV) species deficiently coordinated located on the pore walls of the siliceous framework. These acid catalysts are active in the glycerol dehydration, increasing the conversion with the zirconium content until values higher than 90 mol% for a Si/Zr molar ratio of 4, at 325 °C after 5 h of reaction. However, the catalysts suffer deactivation, which is more important when zirconium oxide is incorporated by impregnation of mesoporous MCM-41 silica. The main reaction products were acrolein, acetaldehyde and acetol. Moreover, the catalysts with a SiZr molar ratio higher than 5 are more selective to acetaldehyde. The acrolein yield was, in all cases, lower than 15 mol% after 24 h of TOS, but a pretreatment under a helium flow saturated with water vapour allows reaching an acrolein yield of 28 mol% and ameliorates the stability of catalysts. The selectivity towards acrolein and hydroxyacetone can be explained by considering the influence of the nature of active sites.

Full text of DOI:10.1016/j.apcata.2012.05.015

Applied Catalysis A: General 433–434 (2012) 179–187
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Zirconium doped mesoporous silica catalysts for dehydration of glycerol
to high added-value products
C. García-Sancho, R. Moreno-Tost∗, J. Mérida-Robles, J. Santamaría-González,
A. Jiménez-López, P. Maireles-Torres
Departamento de Química Inorgánica, Cristalografía y Mineralogía (Unidad Asociada al ICP-CSIC), Facultad de Ciencias, Universidad de Málaga, Campus de Teatinos s/n., 29071
Málaga, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of zirconium doped mesoporous silica, with different Si/Zr molar ratios, has been synthesized and
tested in the gas-phase dehydration of glycerol. The surface characterization of these solids by using NH3-
TPD and pyridine adsorption coupled to FTIR spectroscopy has revealed the existence of well dispersed
acid sites, mainly of Lewis type, associated to Zr(IV) species deficiently coordinated located on the pore
walls of the siliceous framework. These acid catalysts are active in the glycerol dehydration, increasing
the conversion with the zirconium content until values higher than 90 mol% for a Si/Zr molar ratio of
4, at 325 ◦C after 5 h of reaction. However, the catalysts suffer deactivation, which is more important
when zirconium oxide is incorporated by impregnation of mesoporous MCM-41 silica. The main reaction
products were acrolein, acetaldehyde and acetol. Moreover, the catalysts with a SiZr molar ratio higher
than 5 are more selective to acetaldehyde. The acrolein yield was, in all cases, lower than 15 mol% after
24 h of TOS, but a pretreatment under a helium flow saturated with water vapour allows reaching an
acrolein yield of 28 mol% and ameliorates the stability of catalysts. The selectivity towards acrolein and
hydroxyacetone can be explained by considering the influence of the nature of active sites.
© 2012 Elsevier B.V. All rights reserved.
Received 28 March 2012
Received in revised form 7 May 2012
Accepted 13 May 2012
Available online 22 May 2012
Keywords:
Mesoporous silica
Glycerol
Zirconium oxide
Acid catalysis
Acrolein
1. Introduction
from the hydrolysis of animal fats and vegetable oils. Due to the
incessant increment of the worldwide production of biodiesel, the
Acrolein (2-propenal) is the simplest unsaturated aldehyde,
with a high synthetic and technical potential due to the conjugation
of the carbonyl group with a vinyl group. Acrolein is widely used
as intermediate in the production of building materials, herbicides
and algaecides, water treatment chemicals, and essential amino
acids like methionine used as supplementary in fodder, especially
for poultry. At low concentrations, it finds also application for pro-
tecting liquid fuels against microorganisms, H2S scavenger and
anti-algae agent.
The current industrial process for acrolein production is by par-
tial catalytic oxidation of propylene. A variety of different catalysts
are available for this reaction, the most efficient catalysts are com-
plex mixed-metal oxides that consist largely of Bi, Mo, Fe, Ni, Co,
and/or K and either P, B, W, or Sb [1]. The principal side reactions
produce acrylic acid, acetaldehyde, acetic acid, carbon monoxide,
and carbon dioxide [2,3]. Since propylene is obtained from crude-
oil resources, it is desired to utilize a lower cost and renewable raw
material for acrolein production, being glycerol a potential candi-
date. Glycerol is a natural by-product of the manufacture of soap
production of glycerol, a by-product of the biodiesel industry, has
notably increased causing a drop in its price. In this way, glycerol
has been turned into an interesting starting raw material for oth-
ers chemical products, where the double dehydration of glycerol to
acrolein is one of the proposed routes for glycerol valorization. This
field of studying has attracted the attention of numerous research
groups and in fact, recently, a critical review of acrolein production
from glycerol has been released [4].
The dehydration reaction of glycerol yielding acrolein is known
since the nineteenth century. Already in 1918, Sabatier et al.
reported the decomposition of glycerol to different products,
including acrolein, in the presence of alumina catalysts [5]. Because
of crude glycerine is found diluted in water, the catalysts must be
active and resistant to the presence of water avoiding a separation
step of the glycerol from the water and therefore reducing the price
of production of acrolein. Various solid acid catalysts have been
tested in the dehydration of glycerol, including zeolites [6–10],
supported heteropolyacids [11–14] and phosphoric acid impreg-
nated activated carbon [15] or Al2O3 and TiO2 [16], niobium oxide
[17,18] and WO3/ZrO2 catalysts [19,20] in a gas phase reaction. Sev-
eral studies have demonstrated that the catalytic activity depends
on the textural, mainly the pore size, and acid properties of the
solid catalysts, as recently reviewed by Pathak et al. [21]. Thus,
∗
Corresponding author. Tel.: +34 952132021; fax: +34 952131870.
E-mail address: rmtost@uma.es (R. Moreno-Tost).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.05.015

180
C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
Tsukuda et al. [13] claimed that catalysts with small pores showed
low catalytic activity, as long as catalysts with a mesopores ranged
between 6 and 10 nm showed a high glycerol conversion and high
selectivity towards acrolein. It seems to be accepted that Brönsted
acid sites of weak-moderate strength catalyse acrolein production
[9,10,17,19], whereas Lewis acid sites and the presence of basic sites
increased the selectivity of hydroxyacetone (acetol) [22–24].
However, the main drawback of most tested acid catalysts con-
cerns the deactivation by coke formation on the catalyst surface
[14,16,17], due to the acid strength and the pore network, although
in some cases the catalytic activity can be recovered by calcining
the catalyst at temperatures above 500 ◦C in flowing air [18]. Corma
et al. [25] reported that glycerol in water can be converted into
acrolein, olefins, and acetaldehyde by using zeolites in a continu-
ous fluidized-bed reactor, where continuous regeneration can be
performed, if needed. The stability against deactivation has been
also reached by doping caesium 12-tungstophosphate salt with
platinum group metals, together with co-feeding hydrogen [26].
Katryniok et al. [4] showed a long-time performance of a catalyst
based on silicotungstic acid supported on SBA-15 grafted with ZrO2.
This effect was interpreted in terms of modification of the interac-
tion between the support and the heteropolyacid which results in
a lower Brönsted acidity. Lauriol-Garbay et al. [27] have synthe-
sized a catalyst based on a mixed Zr–Nb oxide which exhibits a
promising stability (82% glycerol conversion after 177 h on stream).
The key of the stability of these catalysts is the coverage of Lewis
acid sites by polymeric Nb oxide species which were the unse-
lective coke initiator sites. Another technical improvement in the
dehydration of glycerol was proposed by Yan et al. [15], by using
a low-pressure packed bed reactor to favour the evaporation of
glycerol and remove reaction mixture from dew-points. Thus, the
formation of liquids on the catalyst surface was not favoured and,
therefore, the coke formation is decreased.
catalysts. This catalyst was labelled as Si–30Zr. On the other hand,
the SiZr5 catalyst has been treated in a flow of He saturated with
steam during 4 h at 315 ◦C, and the resulting solid was labelled as
SiZr5-315.
2.2. Catalyst characterization
Powder XRD measurements were performed on a Siemens
D5000 automated diffractometer, over
a
2ꢀ range with
Bragg–Brentano geometry using the Cu K␣ radiation and
a
graphite monochromator. Thermogravimetric and differential
thermal analyses (TG-DTA) were performed on a Pyris-Diamond
PerkinElmer apparatus. The temperature was varied from room
temperature up to 1000 ◦C, at a heating rate of 10 ◦C min−1 with a
flux of air of 100 mL min−1, using a mass around 15 mg.
X-ray photoelectron spectroscopy (XPS) studies were per-
formed with
a Physical Electronics PHI 5700 spectrometer
equipped with a hemispherical electron analyzer (model 80-365B)
and a Mg K␣ (1253.6 eV) X-ray source. High-resolution spectra
were recorded at 45◦ take-off-angle by a concentric hemispherical
analyzer operating in the constant pass energy mode at 29.35 eV,
using a 720 mm diameter analysis area. Charge referencing was
done against adventitious carbon (C 1s at 284.8 eV). The pressure in
the analysis chamber was kept lower than 5 × 10−6 Pa. PHI ACCESS
ESCA-V6.0 F software package was used for data acquisition and
analysis. A Shirley-type background was subtracted from the sig-
nals. Recorded spectra were always fitted using Gauss–Lorentz
curves in order to determine more accurately the binding energy
of the different element core levels.
N2 adsorption–desorption isotherms at −196 ◦C were obtained
using an ASAP 2020 model of gas adsorption analyzer from
Micromeritics, Inc. Prior to N2 adsorption, the sample were evac-
uated at 200 ◦C (heating rate 10 ◦C min−1) for 18 h. Pore size
distributions and pore volume were calculated with the BJH
method.
The aim of this research is to test a family of zirconium doped
mesoporous silica catalysts in the dehydration of glycerol to add
value chemicals. These catalysts possess analogous textural prop-
erties but different acidities depending on the degree of zirconium
doping, which effect on their catalytic activity, selectivity and sta-
bility has been evaluated.
Evolved gas analysis at programmed temperature by mass spec-
trometry (EGA-MS) was carried out to the spent catalysts. The
evolved H2O and CO2, arising from the combustion of the carbona-
ceous deposits, were monitored by recording the m/z = 18 and 44
signals, respectively. 80 mg of the sample was loaded in a quartz
microreactor and 50 mL (STP) min−1 of O2 (5 vol.% O2 in He) were
passed through while the sample was heated at 10 ◦C min−1. A
quadrupole mass spectrometer, model BALZERS Prisma QMS 200
controlled by BALTZER QuadstarTM 422 software, was connected
on-line to the reactor outlet for the analysis of gases generated.
FTIR spectra of adsorbed pyridine were recorded on a Shimadzu
8300 FTIR spectrometer at a resolution of 4 cm−1. Each spectrum
was average over 128 scans. Self-supporting wafers of samples with
a weight/surface ratio of about 12 mg cm−2 were placed in a vac-
uum cell with greaseless stopcocks and CaF2 windows. Solids were
evacuated at 350 ◦C and 10−4 Torr overnight, exposed to pyridine
vapours for 15 min at room temperature, and then degassed at
different temperatures. The net FTIR spectra of adsorbed pyridine
were obtained after subtracting the background spectrum of the
solid.
2. Experimental
2.1. Catalysts synthesis
A series of zirconium doped mesoporous MCM-41 silica, with
different Si/Zr molar ratios, has been synthesized by following
the method proposed by Jones et al. and reported in previous
papers [28,29]. Briefly, it was prepared by adding appropriate
amounts of tetraethoxysilane and zirconium n-propoxide to an
ethanol–propanol solution, which was stirred at room temperature
for 15 min, and then added to an aqueous solution of hexade-
cyltrimethylammonium bromide (25 wt.%), previously stirred at
80 ◦C for 30 min. The surfactant/(SiO2 + ZrO2) ratio was 0.5. The pH
was adjusted to 10 by addition of an aqueous solution of tetram-
ethylammonium hydroxide (25 wt.%), and the resulting gel was
stirred at room temperature for 4 days. Solid product was recov-
ered by centrifugation, washed several times with ethanol, dried at
70 ◦C, and then calcined in air at 550 ◦C for 6 h (1 ◦C min−1 heating
rate) in order to remove the surfactant molecules. Three different
Si/Zr molar ratios (4, 5 and 10) were used and the catalysts were
labelled as SiZrx, where x indicates the Si/Zr molar ratio.
2.3. Catalytic reaction
The dehydration of glycerol was performed, at atmospheric
pressure, in a fixed-bed continuous-flow stainless steel reactor
(9.1 mm in diameter, and 230 mm in length), operated in the
down-flow mode. The reaction temperature was measured with
an interior placed thermocouple in direct contact with the cat-
alyst bed. For the activity tests, 0.3 g of catalyst (particle size
0.85–1.00 mm), diluted with SiC to 3 cm3 volume, was used. Cat-
alysts were pre-treated in situ at atmospheric pressure under a
For comparison reasons, a catalyst with the same weight per-
centage of ZrO2 as the SiZr5 catalyst (30 wt.%) has been prepared
by wet impregnation of a pure MCM-41 silica. Then, the catalyst
was dried and calcined in air under the same conditions as SiZrx

C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
181
Table 1
Textural and acid characteristics of SiZrx catalysts.
Catalyst
SBET (m2/g)
Vp (cm3/g)
dp (nm)
TPD-NH3 (␮mol/(g cat))
TPD-NH3 (␮mol/m2)
SiZr4
SiZr5
SiZr10
MCM-Si
Si–30Zr
626
652
685
1000
562
0.627
0.671
0.597
0.824
0.378
3.1
3.3
2.7
2.5
1.8
480
326
262
n.d.
11
0.77
0.50
0.38
n.d.
0.02
n.d., no detectable.
nitrogen flow of 30 mL min−1 at 325 ◦C for 30 min. The glycerol
solution (10 wt.% in water) was supplied by means of a Gilson
307SC piston pump (model 10SC) at 0.05 mL min−1 feed rate in a
N2 flow (30 mL min−1). The evolution of catalytic tests was stud-
ied by collecting liquid samples after 5 and 24 h in a vial cooled in
an isopropanol–liquid N2 trap. These vials were sealed for poste-
rior analysis by gas chromatography. The reaction products were
analysed by means of gas chromatograph (Shimadzu GC-14B),
equipped with a flame ionization detector and a capillary column
(TRB-14).
dehydration to yield acrolein [32–34]. The surface concentration of
acid sites has been evaluated by means of TPD-NH3. It is generally
accepted that the acid strength depends on the ammonia desorp-
tion temperature: weak (100–200 ◦C), medium (200–400 ◦C) and
strong (>400 ◦C) [35]. The absence of distinguishable maxima on the
TPD-NH3 curves (Fig. 3) makes difficult their deconvolution. There-
fore, the acidity of catalysts has been calculated by integrating the
curves in the range between 100 and 550 ◦C (Table 1). Moreover,
the intrinsic acidity of the catalysts or density of acid sites (␮mol
of NH3/m2) is reported in Table 1. These curves are broad and the
complete ammonia desorption is not achieved even at 550 ◦C, evi-
dencing a heterogeneous distribution of acid strengths, although
the weak-medium acid sites are predominant since the majority of
ammonia is desorbed between 100 and 400 ◦C. Thus, SiZrx catalysts
The glycerol conversion (mol%) and the selectivity to the iden-
tified and calibrated products have been calculated as follows:
nGly,in − nGly,out
CGly
=
× 100 and
nGly,in
ni
zi
Sl =
×
× 100
(a)
nGly,in − nGly,out
zGly
500
where nin and nout represent the mol of glycerol at the inlet and
outlet of the reactor. ni are the molar concentration of the product
i at the outlet of the reactor and zi and zGly are the number of atoms
of carbon of the product i and glycerol. The carbon balance (mol%)
was calculated by summing up the unreacted glycerol and the total
quantities of detected and calibrated products.
400
300
200
100
0
3. Results and discussion
3.1. Catalysts characterization
In this paper, a series of zirconium-doped mesoporous silica,
previously used as support of different active phases [30,31], has
been tested in the gas-phase dehydration of glycerol. The meso-
porous character of these solids has been confirmed by the analysis
of the adsorption-desorption isotherms of N2 at −196 ◦C (Fig. 1a).
The isotherms of all synthesized materials are of type IV according
to the IUPAC classification. All of them exhibit the sharp capillary
condensation step at relative pressures of 0.3–0.4. Moreover, it is
observed an unimodal pore size distribution (Fig. 1b), typical of
mesoporous materials. Their high specific surface areas, pore vol-
umes and pore sizes reveal the mesoporous nature of this family
of zirconium doped silica (Table 1). The hexagonal structure of
mesoporous materials (Fig. 2) has been proven by means of the
X-ray diffraction technique because, after calcination at 550 ◦C, all
the materials exhibit in their XRD patterns the intense and broad
low-angle diffraction peak assigned to the (1 0 0) diffraction planes,
except the SiZr4 catalyst which points out its lack of long-range
order. The absence of other well-defined diffraction peaks in the
region 2ꢀ = 3–8◦ can be justified by the low temperature used in
the synthesis of these mesoporous materials, differently from those
reported in most of papers, which are synthesized under hydrother-
mal conditions at higher temperatures. Moreover, XRD patterns in
the high-angle region did not evidence the characteristic reflec-
tions of crystalline ZrO2 phases, thus pointing to the existence of
amorphous or ZrO2 species of small size on the silica walls.
0.0
0.2
0.4
0.6
0.8
1.0
P/P
0
(b)
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
1
2
3
4
5
6
7
8
Pore Diameter (nm)
Fig. 1. (a) Nitrogen adsorption isotherms of SiZrx catalysts: SiZr4 (ꢀ), SiZr5 (ꢀ) and
SiZr10 (ꢁ); and (b) pore size distributions of SiZrx catalysts: SiZr4 (ꢀ), SiZr5 (ꢀ) and
SiZr10 (ꢁ).
It has been reported that the catalyst acidity is one of the
key factors affecting the catalytic performance in the glycerol

182
C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
SiZr10
1.0
b
0.8
0.6
SiZr5
SiZr4
0.4
c
a
0.2
1500 1350 1200 1050 900 750 600
0
1
2
3
4
-1
Wavenumber (cm )
2θ (º)
Fig. 4. FTIR analysis of the SiZrx in the range 600–1500 cm−1
.
Fig. 2. Powder XRD patterns at low angle of SiZrx catalysts: SiZr4 (a), SiZr5 (b) and
SiZr10 (c).
1050–1070 cm−1, which intensity gradually is reduced as the Zr
content increases. The red shift of this band with the increase of zir-
conium content is a demonstration of the replacement of Si atoms
in the framework by Zr atoms [40]. The band at 808 cm−1 is asso-
display a total acidity higher than pure mesoporous silica [36] and
it is noticeable the increase in the acidity with the amount of Zr
incorporated to the framework, being the SiZr4 catalyst the most
acidic among the SiZrx catalysts (Table 1). In general, the intrinsic
acidity of SiZrx catalysts is lower than those reported for zeolites
[34], SAPO catalysts [16], Me–Al2O3–PO4 [37] and Zr-TUD-1 sam-
ples [38] but, on the other hand, the acidity of SiZrx is higher than
that exhibited by a hexagonal ordered mesoporous silica grafted
with 20 wt.% ZrO2 [4] or Al incorporated in the framework [10], or
quite similar to a montmorillonite pillared with ZrO2 [39]. The acid
sites are mainly due to the presence of Zr into the mesoporous net-
work where superficial Zr(IV) species with low coordination exist.
The presence of Zr in the framework of the siliceous mesoporous
material has been proven by means of FTIR of the SiZrx catalysts
(Fig. 4). These catalysts show some characteristics features, such
ciated to the Si
O Si symmetric stretching [41] and finally, the
band at 970 cm−1 is ascribed to the Si (OH) stretching of terminal
silanols [38,29].
One commonly used method to characterize the nature of the
acid sites (Brönsted or Lewis) is the study of the shift of the IR and
Raman vibration bands of basic probe molecules when adsorbed on
the catalyst surface [42]. In this sense, the adsorption of pyridine is
widely used as probe molecule [10,22,43] since different adsorbed
pyridine species can be observed: pyridine bonded by hydrogen
interaction on weakly acidic OH groups, pyridine on deficiently
coordinated centres (Lewis sites) and protonated pyridine on Brön-
sted sites. In this way, the acid sites have been characterized by
adsorption of pyridine after outgassing at different temperatures
(Fig. 5 and Table 2). The concentration of both types of acid sites
has been estimated from the integrated absorption at 1550 and
1450 cm−1 using the extinction coefficients EB = 0.73 cm ␮mol−1
and EL = 1.11 cm ␮mol−1 for Brönsted and Lewis sites, respectively.
The IR spectra of the SiZr5 catalyst, after pyridine adsorption and
outgassing at 100 and 200 ◦C, are shown in Fig. 5. The SiZr10
and SiZr4 catalysts exhibit the same bands after contacting with
pyridine vapours. The SiZr5 catalyst displays bands at 1446 and
1596 cm−1, assigned to pyridine adsorbed on weak Lewis acid sites
or to weakly hydrogen-bonded pyridine [22,44,45]. The band at
1577 cm−1 can be also ascribed to pyridine adsorbed on weak Lewis
acid sites [46]. These bands are almost vanished after outgassing
at 100 ◦C, thus revealing the existence weak Lewis acid sites. The
bands at 1637 and 1544 cm−1 can be assigned to pyridinium ring
vibrations, due to the proton transfer from Brönsted acid sites to
pyridine [34,46]. After outgassing at 100 and 200 ◦C, the bands asso-
ciated to Brönsted sites are still present on the surface of catalysts.
Finally, outgassing at temperatures higher than 200 ◦C provokes the
disappearance of bands associated to pyridine bonded to both Brön-
sted and Lewis centres. It is well-known that pure MCM-41 silica
andZrO2 are Lewis solids withoutBrönstedacidsites[43,44];there-
fore, the Brönsted acid sites in the silica MCM-41 network detected
by the adsorption of pyridine coupled with FTIR are caused by the
as the asymmetric stretching vibration of Si
O
Si bonds found at
SiZr4
SiZr5
SiZr10
500
100
200
300
400
Temperature (ºC)
Fig. 3. TPD-NH3 curves of SiZrx catalysts.

C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
183
Table 2
Concentration of Brönsted and Lewis acid sites of SiZrx catalysts, after evacuation at different temperatures.
Catalyst
Evacuation temperature (◦C)
Brönsteda sites (␮mol/g)
Lewisa sites (␮mol/g)
IB/(IB + IL)
100
200
100
200
100
200
68.5
8.8
36.3
35.9
107.1
32
51.9
7.4
9.2
0
88.4
34.9
0.57
0.54
0.80
1.0
0.54
0.50
SiZr4
SiZr5
SiZr10
100
200
107.8
105.7
50.3
11.8
0.68
0.90
SiZr5-315
a
Concentration of acid sites calculated from the extinction coefficients calculated by Datka et al., J. Catal. 135 (1992) 186–199.
Table 3
furthermore, the binding energies of Zr 3d5/2 (182.7 eV) are higher
than that of ZrO2 (182.2 eV) supporting the incorporation of the
heteroatom in the siliceous framework. The higher binding energy
of the Zr 3d5/2 band of SiZrx catalysts implies a higher ionic char-
acter of the bonding Zr O [45,49] and therefore involving a higher
positive charge over Zr which leads to a stronger Lewis acidity in
these materials.
Surface characterization of SiZrx catalysts determined by XPS analysis.
Catalyst
O1s (eV)
Zr3d5/2 (eV)
Atomic ratio
Si/Zr
9.8
O/(Si + Zr)
531.1 (12.5%)
533.2 (87.5%)
531.3 (8.1%)
533.2 (91.9%)
531.6 (4.8%)
533.2 (95.2%)
182.7
182.8
182.8
2.1
2.2
2.2
SiZr4
SiZr5
SiZr10
13.2
46.7
3.2. Catalytic performance
In the dehydration of glycerol to yield acrolein, ZrO2 has been
used as support of WO3 [19], where its seems to ameliorate the
long-term stability of catalysts [4,12]. Moreover, Chai et al. [32]
have reported an acrolein yield of 7 mol% at 315 ◦C, after 9–10 h
of time-on-stream and glycerol GHSV = 80 h−1 when ZrO2 was
employed.
presence of atoms of Zr replaced the Si ones [45]. The origin of these
Brönsted acid sites has been related to changes in electron density
on Si, due to charge imbalance or difference of electronegativity,
resulting from introduction of Zr atom in the vicinity of the Si OH,
weakens the SiO H bond and generates Brönsted acidity [47].
The superficial enrichment of Zr is corroborated by XPS, since
the surface Si/Zr atomic ratio increases with the Si/Zr molar ratio
used in the synthesis gel (Table 3). The XPS analysis in the region of
O1s signal becomes more asymmetric as the Zr content is higher,
indicating the existence of two different oxygen environments. This
Preliminary experiments were carried out to rule out the contri-
bution to the overall glycerol conversion of silicon carbide, which
has been used as diluents of the catalysts. The results show that SiC
is inert in the whole range of tested temperatures (275–350 ◦C),
hence the glycerol conversion observed must be associated to the
catalysed process. Moreover, at these reaction temperatures, a ther-
mal homogenous glycerol conversion has not been observed.
The analysis of the liquid phase, after 5 and 24 h of time on
stream (TOS) at 325 ◦C, reveals the formation of acrolein, acetalde-
hyde and hydroxyacetone as main products, as well as acetone,
ethanol, methanol, allyl alcohol and propanaldehyde were detected
as minor components. These minor components along others no
analysed have been included in the column labelled as Others
(Table 4). The unidentified products can be possibly formed by
secondary reactions among products or betweenproducts and glyc-
erol, together with gaseous species such as CO and CO2. In spite
of those unidentified products, the carbon balance was usually
above 90 mol%, except for the SiZr4 catalyst [37,43,50]. Full con-
version of glycerol is not accomplished with none of the tested
catalysts within the first 5 h of reaction, although SiZr4 and SiZr5
catalysts lead to glycerol conversion close to 90 mol%, whereas
the glycerol conversion found for the SiZr10 catalyst is lower
(73 mol%), which reveals that glycerol conversion increases with
the total acidity of catalysts [16]. However, after 24 h of TOS, all
the tested catalysts exhibit some degree of deactivation, being less
pronounced in the case of the SiZr4 catalyst which maintains a glyc-
erol conversion of 73 mol%. The SiZr4 catalyst is the most active
and stable, but it is not the most selective to acrolein (Table 4)
since its selectivity to acrolein is 14 mol% after 24 h of TOS, being
the SiZr5 catalyst the most selective to acrolein with a value of
31 mol%. Along with acrolein, acetaldehyde is the major reaction
product, which has been proposed as a decomposition product
of 3-hydroxypropanaldehyde [13,16,51]. On the other hand, the
acetol (hydroxyacetone) is also detected as dehydration product
of glycerol (Table 4). In fact, in the case of the SiZr10 catalyst,
the hydroxyacetone is the major constituent among the reaction
products, reaching up to 24 mol% of selectivity after 24 h of TOS.
O1s band can be deconvoluted into two components: Si
O Zr (at
531.1 eV) and Si Si (at 533.2 eV), the former increasing their
O
intensity with the Zr content at expense of the later, indicating
the replacement of Si by Zr atoms. The presence of segregate ZrO2
could be ruled out by the lack of the O 1s band at 529 eV [48] and
SiZr5-315
R. T.
100ºC
200ºC
SiZr5
R. T.
100ºC
200ºC
1700 1650 1600 1550 1500 1450 1400
Wavenumber (cm-1)
Fig. 5. FTIR analysis of pyridine adsorption on SiZr5 and SiZr5-315 catalysts, after
outgassing at 100 and 200 ◦C.

184
C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
Table 4
Glycerol conversion and product selectivity over SiZrx catalysts, at 325 ◦C.
Catalyst
TOS (h)
Conversion (%)
Product selectivity (%)
Acrolein yield (%)
C balance (%)
Acrolein
Acetaldehyde
Acetol
Allyl alcohol
Others
50
44
24
18
15
5
5
24
5
24
5
24
5
24
5
92
73
89
47
97
80
73
44
51
12
15
14
26
31
40
35
24
19
81
0
19
22
29
37
33
35
28
22
9
13
17
21
14
13
23
30
24
11
46
3
2.7
0
0
0
1.7
8.3
5.2
0
13
10
23
15
38
28
19
11
41
0
57
68
79
91
86
96
93
87
100
99
SiZr4
SiZr5
SiZr5-315
SiZr10
Si–30Zr
10
30
0
24
17
0
37
Experimental conditions: weight of catalyst = 0.3 g diluted with SiC to 3 cm3 volume. Feed composition: 10 wt.% glycerol in water; liquid flow = 0.05 mL min−1; N2
flow = 30 mL min−1; T = 325 ◦C.
The acrolein selectivity after 24 h TOS decreased in the order:
SiZr5 > SiZr10 > SiZr4. This selectivity trend is also observed for
the acetaldehyde which is the main reaction product obtained at
325 ◦C. However, the selectivity to hydroxyacetone reached a min-
imum for SiZr5 catalyst.
taking into account the nature of the acid sites. Thus, the SiZr5 cat-
alyst is the most selective to acrolein due to its higher B/(L + B) acid
sites ratio (Table 2) among the tested catalysts, and the SiZr10 cat-
alyst is the most selective to hydroxyacetone since it possesses the
higher concentration of Lewis acid sites.
These SiZrx catalysts show a reaction product distribution quite
different to that reported in bibliography. Thus, Chai et al. [11,12,32]
reported that bulk ZrO2 showed acrolein selectivity lower than
10 mol% and hydroxyacetone appeared as the main product. Tao
et al. [52] reported that mixed silica zirconia showed acrolein selec-
tivity of 15 mol% for a 20 mol% of ZrO2 and acrolein selectivity of
21 mol% if the molar percentage of ZrO2 was increased up to 75%,
after 10 h of TOS. In this case, the hydroxyacetone was also the
main by-product and acetaldehyde selectivity was as low as 5 mol%.
Stosic et al. [24] reported a commercial zirconia with a glycerol
conversion of 59% after 90 min TOS and an acrolein selectivity of
5 mol% and hydroxyacetone selectivity of 25 mol%. As it is known,
zirconia is a typical Lewis acid solid, so in the dehydration reac-
tion of glycerol, the selectivity is addressed to the formation of
hydroxyacetone instead of acrolein, even though a significant con-
version of glycerol is displayed. Finally, it is noticeable to mention
that the SiZrx catalysts show a better catalytic performance than
those reported in bibliography based on bulk or mixed oxides with
ZrO2 [11,24,52,54].
In general, the mechanisms proposed to explain the large vari-
ety of products coming from this reaction distinguish two main
pathways for glycerol activation. On the one hand, on Lewis acid
sites, the terminal OH of glycerol is activated leading to the forma-
tion of hydroxyacetone, but, on the other hand, on Brönsted acid
sites, the secondary OH protonation occurs evolving towards the
acroleinformation [4,9,17,25,32,43]. Theresults ofpyridineadsorp-
tion have shown the presence of both Brönsted and Lewis acid sites
on the SiZrx catalysts and therefore, the two reaction paths cannot
be ruled out. Thus, the formation of acrolein indicates a mecha-
nism based on protonation of the secondary alcohol of glycerol to
form the unstable 3-hydroxypropanaldehyde, intermediate which
suffers dehydration to form acrolein. The by-product hydroxyace-
tone is formed after activation of glycerol on the Lewis acid sites
present on the surface of the catalysts. Kim et al. [55] have reported
a correlation between the hydroxyacetone yield and the concen-
tration of Lewis acid sites. The hydroxyacetone formation has been
also described by Auneau et al. [56] in the hydrogenolysis of glyc-
erol to 1,2-propanediol by means of a dehydration mechanism on
the Rh(1 1 1) surface. The formation of acetaldehyde could happen
from the unstable 3-hydroxypropanal, via a retro-aldol reaction
[13,32] or from hydroxyacetone as Suprun et al. [16] or Lauriol-
Garbey et al. [50] have reported. Corma et al. [25] also reported
other way for the acetaldehyde formation from an unstable dione,
which decomposes rapidly into CO and acetaldehyde. According
to the pyridine tests, the selectivity pattern could be explained by
Several authors [22,26] have reported that Brönsted sites can
be generated on the catalyst surface from Lewis sites if water is
present, with a concomitant increasing of the selectivity to acrolein.
In order to transform Lewis acid sites into Brönsted ones, the SiZr5
catalyst was pre-treated in a helium flow saturated with steam
water at 315 ◦C for 4 h. This pretreatment of the SiZr5 catalyst leads
to an important increase of the concentration and strength of Brön-
sted acid sites in detrimental of the Lewis acid centres, as evidenced
the FTIR spectrum after outgassing at 200 ◦C (Table 2 and Fig. 5).
This shift in the distribution of acid sites ameliorates the stabil-
ity of catalysts against deactivation, since the glycerol conversion
hardly decreases after 24 h of TOS (Table 4) and the acrolein yield is
increased until 28 mol%. Moreover, the selectivity to acetol is lower
after 5 h of TOS, revealing that a fraction of Lewis acid sites are
converted in Brönsted sites after the steaming treatment, although
after 24 h the selectivity to acetol is higher than that observed for
SiZr5 catalyst.
The evaluation of the effect of reaction temperature on the dehy-
dration of glycerol was examined over the SiZr5 catalyst, after
5 h of TOS (Fig. 6). As it was expected, the glycerol conversion
remarkably increased with the reaction temperature from 275
Conversion
100
80
60
40
20
0
Acrolein
Acetaldehyde
Hydroxyacetone
Others
Acrolein Yield
275
300
325
350
Temperature (ºC)
Fig. 6. Conversion and selectivity of glycerol dehydration over SiZr5 as a function of
the reaction temperature. Experimental conditions: weight of catalyst = 0.3 g diluted
with SiC to 3 cm3 volume. Feed composition: 10 wt.% glycerol in water; Liquid
flow = 0.05 mL min−1; N2 flow = 30 mL min−1
.

C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
185
Table 5
Characterization of spent SiZrx catalysts.
Catalyst (TOS = 24 h)
SBET (m2/g)
Vp (cm3/g)
dp (nm)
Weigth lossa (%)
CHN analysis
C (mol%)
H (mol%)
H/C
SiZr4
SiZr5
SiZr10
389
251
194
0.614
0.279
0.438
5.3
3.5
6.4
16.3
11.4
13.4
0.83
0.64
0.90
1.32
1.31
1.71
1.58
2.05
1.90
a
From TG results.
to 350 ◦C. Thus, Ulgen et al. [20] stated that temperatures lower
than 280 ◦C favoured the intermolecular dehydration, yielding glyc-
erol oligomers instead the intramolecular dehydration leading to
acrolein. These authors claimed that these oligomers are responsi-
ble of the decrease of the acrolein selectivity at low temperatures.
The acrolein yield increases slightly with the reaction temper-
atures, reaching a value of 26 mol% at 325 ◦C. It is noteworthy
the important decrease in the selectivity to Others by increas-
ing the reaction temperature. This catalytic behaviour had been
already observed by Corma et al. [25], who pointed an increase
of coke deposited on the catalysts when the reaction temperature
decreased from 350 until 290 ◦C, while observing a lower pro-
duction of by-products whereas the acrolein production remained
almost unchanged throughout the tested temperature range. On
the other hand, the acetaldehyde selectivity increased with the
temperature, reaching the highest value at 350 ◦C, similarly to that
reported by Kim et al. [34] for H-zeolites. Haider et al. [53] con-
sider that at temperatures higher than 290 ◦C the formation of
acetaldehyde could be preferential via a retro aldol fragmenta-
tion of 3-hydroxypropanal. At the same time, the selectivity to
acetaldehyde is increased at 350 ◦C, the hydroxyacetone selectivity
is decreased and the acrolein selectivity is maintained. Lauriol-
Garbey et al. [50] observed the same behaviour, and they pointed
out that hydroxyacetone could be converted to acetaldehyde and
acetone over acidic sites.At the beginning of this section, it was
mentioned that a key characteristic of zirconia is to ameliorate the
long-term stability of catalysts in the dehydration of glycerol, but
up to date it has not been tested as an active phase itself, i.e. zirco-
nia have been only used as support or stabilizer of an active phase.
Furthermore, we have proved that SiZrx catalysts show better cat-
alytic performance that those tested in bibliography based only
on ZrO2. On the other hand, since the discovery of M41S solids by
Mobil researchers, their excellent textural properties have encour-
aged their use to disperse catalytically active phases. Therefore,
we have prepared a catalyst by dispersing ZrO2 in a pure MCM-41
silica and compared its catalytic activity with that of the SiZr5 cat-
alyst. Thus, mesoporous MCM-41 silica has been impregnated with
30 wt.% ZrO2, an analogous Zr content to that of the SiZr5 catalyst,
and tested in the dehydration of glycerol. The main textural prop-
erties of this catalyst are collected in Table 2. Table 1 shows that the
Si–30Zr catalyst is more selective to acrolein within the first 5 h of
reaction, but acrolein was not detected among the products after
24 h of TOS, whereas the hydroxyacetone selectivity increased up
to 46 mol%. Moreover, Si–30Zr is less resistant to the deactivation
since glycerol conversion drops up to 12 mol% after 24 h of TOS.
Therefore, the incorporation of zirconium into the siliceous net-
work gives rise to acid catalysts with a catalytic performance in
the glycerol dehydration better than that exhibited when ZrO2 is
dispersed on the mesoporous silica walls.
responsible of deactivation. Prior to the characterization, these
spent catalysts have been dried in situ under the carrier gas stream.
A considerable amount of carbonaceous deposits could be inferred
from the dark colour of the spent catalysts. The results of their
textural characterization (Table 5) reveal a noticeable decrease in
the SBET and pore volume values, which is more important for the
SiZr10 catalyst. This catalyst showed the lowest values of pore
diameter and pore volume among the tested catalysts, which could
be explained by the presence of carbonaceous deposits blocking
the porous network. Thus, it is observed that the more acidic cat-
alyst displays the highest weight loss, as measured by TG analysis,
after the catalytic run (Table 5). The H/C ratios of the carbonaceous
deposits formed at 325 ◦C suggest that the deposits are hydrogen
deficient, which can be associated to species highly condensed like
aromatic molecules. The formation of these molecules could be due
to consecutive reactions of glycerol, as well as to side reactions
between products [16].
The thermal stability of carbonaceous deposits has been studied
by means of DTA and EGA-MS analysis. The DTA curves of the spent
catalysts (Fig. 7) show two overlapped exothermic peaks extending
from 150 to 700 ◦C. The shape of these curves point out the presence
of different types of carbonaceous deposits on the surface of the
catalysts The first peak is more intense for all catalysts and the max-
ima are observed at 363 and 383 ◦C for SiZr5 and SiZr10 catalysts,
respectively. This first peak could be ascribed to carbon deposits
easily oxidized [9] like aliphatic molecules [26]. Meanwhile, the
second peak is discernible for the SiZr4 and SiZr10 catalysts but
it appears like a shoulder and at lower temperature for the SiZr5
catalyst. This peak at higher temperature could be ascribed to car-
bon deposits more difficult to burn, such as those associated to
polyaromatic molecules [26]. Among the tested catalysts, the SiZr4
catalyst shows a larger quantity of heat emitted during the burn-
ing of the carbonaceous deposits, pointing out a higher amount of
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
SiZr4
SiZr10
SiZr5
0.0
3.3. Coke characterization
-0.5
0
100 200 300 400 500 600 700 800 900
Taking into account that catalyst deactivation is an impor-
tant drawback for most of catalysts in the glycerol dehydration,
spent catalysts have been characterized after the catalytic test in
order to know the nature of the adsorbed species, which could be
Temperature (ºC)
Fig. 7. DTA analysis of spent catalysts.

186
C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
0.015
0.012
0.009
0.006
0.003
0.000
-0.003
0.165
0.160
0.155
0.150
0.145
0.140
0.135
0.015
0.012
0.009
0.006
0.003
0.000
-0.003
0.165
0.160
0.155
0.150
0.145
0.140
0.135
SiZr10
SiZr4
0
100 200 300 400 500 600 700 800
0
100 200 300 400 500 600 700 800
Temperature (ºC)
Temperature (ºC)
0.015
0.012
0.009
0.006
0.003
0.000
-0.003
0.165
0.160
0.155
0.150
0.145
0.140
0.135
SiZr5
0
100 200 300 400 500 600 700 800
Temperature (ºC)
Fig. 8. EGA-MS analysis of the spent catalysts. Red line: evolved water; Black line: evolved carbon dioxide and blue line: evolved oxygen.
carbonaceous deposits on the surface of this catalyst such as TG
analysis shown.
4. Conclusions
The EGA-MS analysis has been also used to study the carbona-
ceous deposits in the spent catalysts. The catalysts have been
heated in a flow of oxygen (5 vol.% in He) from 25 ◦C to 800 ◦C. Fig. 8
displays the consumption of O2 as well as the evolved CO2 and H2O
during the combustion of carbonaceous deposits formed during the
catalytic tests at 325 ◦C over the SiZr5 catalyst. The evolved H2O
curve shows a peak at low temperature without emission of CO2,
which can be ascribed to the physisorbed water not evacuated on
the catalyst surface after the drying treatment. Between 200 and
700 ◦C, the H2O curve shows two peaks at 376 and 411 ◦C and a
shoulder at 520 ◦C. These peaks are associated to the combustion
of carbonaceous deposits since the evolved H2O is accompanied by
the releasing of CO2. The curve of evolved CO2 emission is fairly
symmetrical in the temperature range from 200 to 700 ◦C; this
fact along with the intensity of CO2 and H2O signals could indicate
the presence of highly hydrogenated carbonaceous species, which
would be of aliphatic nature. This is consistent with the results of
DTA analysis, which showed that the deposits on this catalyst were
less resistant to combustion. However, the composition of the coke
deposited on SiZr10 and SiZr4 catalysts may differ from the SiZr5.
On one hand, the curve of evolved H2O also shows a peak at tem-
peratures lower than 100 ◦C due to physisorbed H2O and a band
extending from 250 to 700 ◦C with a maximum at 430 ◦C and a
shoulder at 560 ◦C. The peak at 430 ◦C is accompanied by the emis-
sion of CO2 with a similar intensity. This peak, as in the case of the
SiZr5 catalyst, is due to aliphatic carbon species, while the shoulder
at 560 ◦C is accompanied by a more intense emission of CO2, indi-
cating that these species are more deficient in hydrogen, perhaps
unsaturated hydrocarbon more resistant to combustion. In conclu-
sion, the DTA and EGA-MS analysis shows that the coke deposited
on the SiZr5 catalyst is more aliphatic than deposited in the other
two catalysts which present more unsaturated species.
A series of zirconium doped mesoporous silica, with different
Si/Zr molar ratios, has been synthesized and tested in the gas-
phase dehydration of glycerol. NH3-TPD and pyridine adsorption
coupled to FTIR spectroscopy has revealed the presence of well dis-
persed acid sites, mainly of Lewis type, associated to Zr(IV) species
deficiently coordinated located on the pore walls of the siliceous
framework. The glycerol conversion increases with the zirconium
content, reaching values higher than 90 mol% for a Si/Zr molar
ratio of 4, at 325 ◦C after 5 h of TOS. However, the catalysts suf-
fer from deactivation, which is more important when zirconium
oxide is incorporated by impregnation of a mesoporous MCM-41
silica. EGA-MS and TG-DTA have demonstrated that the carbona-
ceous deposits are mainly aliphatic molecules, except for the SiZr5
catalysts, where the presence of saturated species is detected. The
acrolein yield is in all cases lower than 23 mol%, but the pretreat-
ment under a saturated-steam helium flow allows reaching an
acrolein yield of 38 mol% and ameliorates the stability of cata-
lysts. The selectivity towards acrolein and hydroxyacetone can be
explained by considering the influence of the nature of active sites.
Acknowledgments
The authors are grateful to the Spanish Ministry of Science and
Innovation (ENE2009-12743-C04-03 project), Junta de Andalucía
(P09-FQM-5070) and FEDER funds for financial support. RMT would
like to thank the Ministry of Science and Innovation for the financial
support under the Program Ramón y Cajal (RYC-2008-03387).
References
[1] Y. Moro-Oka, W. Ueda, Adv. Catal. 40 (1994) 233–273.
[2] C. Huang, W. Guo, X. Yi, W. Weng, H. Wan, Catal. Commun. 8 (2007) 162–166.

C. García-Sancho et al. / Applied Catalysis A: General 433–434 (2012) 179–187
187
[3] F. Ivars, B. Solsona, S. Hernández, J.M. López Nieto, Catal. Today 149 (2010)
260–266.
[30] A. Jiménez-López, E. Rodríguez-Castellón, P. Maireles-Torres, L. Díaz, J. Mérida-
Robles, Appl. Catal. A 218 (2001) 295–306.
[4] B. Katryniok, S. Paul, V. Bellière-Baca, P. Rey, F. Dumeignil, Green Chem. 12
(2010) 2079–2098.
[5] P. Sabatier, G. Gaudion, Compt. Rend. 166 (1918) 1033–1039.
[6] C.J. Zhou, C.J. Huang, W.G. Zhang, H.S. Zhai, H.L. Wu, Z.S. Chao, Stud. Surf. Sci.
Catal. 165 (2007) 527–530.
[7] E. Yoda, A. Ootawa, Appl. Catal. A 360 (2009) 66–70.
[8] Y.T. Kim, K.D. Jung, E.D. Park, Microporous Mesoporous Mater. 131 (2010)
28–36.
[9] C.J. Jia, Y. Liu, W. Schmidt, A.H. Lu, F. Schüth, J. Catal. 269 (2010) 71–79.
[10] A.S. de Oliveira, S.J.S. Vasconcelos, J.R. de Sousa, F.F. de Sousa, J.M. Filho, A.C.
Oliveira, Chem. Eng. J. 168 (2011) 765–774.
[11] S.H. Chai, H.P. Wang, Y. Liang, B.-Q. Xu, Appl. Catal. A 353 (2009) 213–222.
[12] S.H. Chai, H.P. Wang, Y. Liang, B.Q. Xu, Green Chem. 10 (2008) 1087–1093.
[13] E. Tsukuda, S. Sato, R. Takahashi, T. Sodesawa, Catal. Commun. 8 (2007)
1349–1353.
[14] H. Atia, U. Armbruster, A. Martin, J. Catal. 258 (2008) 71–82.
[15] W. Yan, G.J. Suppes, Ind. Eng. Chem. Res. 48 (2009) 3279–3283.
[16] W. Suprun, M. Lutecki, T. Haber, H. Papp, J. Mol. Catal. A 309 (2009) 71–78.
[17] S.H. Chai, H.P. Wang, Y. Liang, B.Q. Xu, J. Catal. 250 (2007) 342–349.
[18] N.R. Shiju, D.R. Brown, K. Wilson, G. Rothenberg, Top. Catal. 53 (2010)
1217–1223.
[31] M.C. Carrión, B.R. Manzano, F.A. Jalón, D. Eliche-Quesada, P. Maireles-Torres, E.
Rodríguez-Castellón, A. Jiménez-López, Green Chem. 7 (2005) 793–799.
[32] S.H. Chai, H.P. Wang, Y. Liang, B.Q. Xu, Green Chem. 9 (2007) 1130–1136.
[33] Q. Liu, Z. Zhang, Y. Du, J. Li, X. Yang, Catal. Lett. 127 (2009) 419–428.
[34] Y.T. Kim, K.-D. Jung, E.D. Park, Appl. Catal. A 393 (2011) 275–287.
[35] P. Berteau, B. Delmon, Catal. Today 5 (1989) 121–137.
[36] C. García-Sancho, R. Moreno-Tost, J.M. Mérida-Robles, J. Santamaría-González,
A. Jiménez-López, P. Maireles-Torres, Appl. Catal. B 108–109 (2011) 161–167.
[37] W. Suprun, M. Lutecki, R. Gläser, H. Papp, J. Mol. Catal. A 342–343 (2011)
91–100.
[38] A. Ramanathan, M.C. Castro Villalobos, S. Cees Kwakernaak, U. Telalovic, Hane-
feld, Chem. Eng. J. 14 (2008) 961–972.
[39] M. Knapczyk, M. Motek, W. Suprun, H. Papp, T. Grybeck, Catal. Today 176 (2011)
331–335.
[40] Q. Zhao, X. Zhou, Y. Li, T. Jiang, H. Yin, C. Li, Appl. Surf. Sci. 255 (2009) 6397–6403.
[41] T.S. Jiang, X.P. Zhou, Y.H. Li, Q. Zhao, H.B. Yin, Inorg. Mater. 47 (2011) 296–302.
[42] G. Busca, Catal. Today 41 (1998) 191–206.
[43] F. Cavani, S. Guidetti, L. Martinelli, M. Piccinini, E. Ghedini, M. Signoretto, Appl.
Catal. B 100 (2011) 197–204.
[44] D. Carriazo, C. Domingo, C. Martin, V. Rives, J. Solid State Chem. 181 (2008)
2046–2057.
[19] A. Ulgen, W. Hoelderich, Catal. Lett. 131 (2009) 122–128.
[20] A. Ulgen, W.F. Hoelderich, Appl. Catal A 400 (2011) 34–38.
[21] K. Pathak, K.M. Reddy, N.N. Bakhshi, A.K. Dalai, Appl. Catal. A 372 (2010)
224–238.
[45] P. Salas, J.A. Wang, H. Armendariz, C. Angeles-Chavez, L.F. Chen, Mater. Chem.
Phys. 114 (2009) 139–144.
[46] J.M.R. Gallo, C. Bisio, G. Gatti, L. Marchese, H.O. Pastore, Langmuir 26 (2010)
5791–5800.
[22] H. Atia, U. Armbruster, A. Martin, Appl. Catal. A 393 (2011) 331–339.
[23] C.L. Lima, S.J.S. Vasconcelos, J.M. Filho, B.C. Neto, M.G.C. Rocha, P. Bargiela, A.C.
Oliveira, Appl. Catal. A 399 (2011) 50–62.
[24] D. Stosic, S. Bennici, J-L. Couturier, J.-L. Dubois, A. Auroux, Catal. Commun. 17
(2012) 23–28.
[25] A. Corma, G.W. Huber, L. Sauvanaud, P. O’Connor, J. Catal. 257 (2008)
163–171.
[26] A. Alhanash, E.F. Kozhevnikova, I.V. Kozhevnikov, Appl. Catal. A 378 (2010)
11–18.
[47] J.A. Anderson, C. Fergusson, I. Rodríguez-Ramos, A. Guerrero-Ruiz, J. Catal. 192
(2000) 344–354.
[48] M.D. Gracia, A.M. Balu, J.M. Campelo, R. Luque, J.M. Marinas, A.A. Romero, Appl.
Catal. A 37 (2009) 85–91.
[49] J. Iglesias, J.M. Melero, L.F. Bautista, G. Morales, R. Sánchez-Vázquez, M.Y.
Andreola, A. Lizarraga-Fernández, Catal. Today 167 (2011) 46–55.
[50] P. Lauriol-Garbey, S. Loridant, V. Bellière-Baca, P. Rey, J.M.M. Millet, Catal. Com-
mun. 16 (2011) 170–174.
[51] F. Wang, J.L. Dubois, W. Ueda, J. Catal. 268 (2009) 260–267.
[52] L.-Z. Tao, S.-H. Chai, Y. Zuo, W.T. Zheng, Y. Liang, B-Q. Xu, Catal. Today 158
(2010) 310–316.
[27] P. Lauriol-Garbay, J.M.M. Millet, S. Loridant, V. Bellière-Baca, P. Rey, J. Catal. 280
(2011) 68–76.
[28] D.J. Jones, J. Jiménez-Jiménez, A. Jiménez-López, P. Maireles-Torres, P.
Olivera-Pastor, E. Rodríguez-Castellón, J. Rozière, Chem. Commun. 5 (1997)
431–432.
[29] E. Rodríguez-Castellón, A. Jiménez-López, P. Maireles-Torres, D.J. Jones, J. Roz-
ière, M. Trombetta, G. Busca, M. Lenarda, L. Storaro, J. Solid State Chem. 175
(2003) 159–169.
[53] M.H. Haider, N.F. Dummer, D. Zhang, P. Miedziak, T.E. Davies, S.H. Taylor, D.J.
Willock, D.W. Knight, D. Cadwick, G.J. Hutchings, J. Catal. 286 (2012) 206–213.
[54] Y.T. Kim, K.D. Jung, E.D. Park, Bull. Korean Chem. Soc. 11 (2010) 3283–3290.
[55] Y.T. Kim, K.D. Jung, E.D. Park, Appl. Catal. B 107 (2011) 177–187.
[56] F. Auneau, C. Michel, F. Delbecq, C. Pinel, P. Sautet, Chem. Eur. J. 17 (2011)
14288–14299.