Zirconium doped MCM-41 supported WO3 solid acid catalysts for the esterification of oleic acid with methanol

Article abstract of DOI:10.1016/j.apcata.2010.03.001

A series of zirconium doped MCM-41 silica supported WO_x solid acid catalysts, with WO₃ loading ranging from 5 to 25 wt%, has been prepared by impregnation with ammonium metatungstate, and subsequent activation at temperatures between 450 and 800 °C. XRD, Raman and UV-vis spectroscopies have allowed us to establish that at temperatures higher than 700 °C occurs the formation of (WO)_x/ZrO₂ nanoparticles on the surface of MCM-41 support, which exhibit acidic properties and are active in the esterification of oleic acid with methanol at 65 °C. WO₃ loadings of 15-20 wt%, after activation at 700 °C, lead to the most active catalysts, with conversion values close to 100%. The catalyst with 15 wt% WO₃ is stable even when the reaction is carried out at 200 °C, reusable at least during four cycles and no leaching of tungsten species in the liquid phase was found.

Full text of DOI:10.1016/j.apcata.2010.03.001

Applied Catalysis A: General 379 (2010) 61–68
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Zirconium doped MCM-41 supported WO3 solid acid catalysts for the
esterification of oleic acid with methanol
I. Jiménez-Morales, J. Santamaría-González, P. Maireles-Torres, A. Jiménez-López∗
Departamento de Química Inorgánica, Cristalografía y Mineralogía (Unidad Asociada al ICP-CSIC), Facultad de Ciencias, Universidad de Málaga,
Campus de Teatinos, 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 MCM-41 silica supported WOx solid acid catalysts, with WO3 loading ranging
from 5 to 25 wt%, has been prepared by impregnation with ammonium metatungstate, and subsequent
activation at temperatures between 450 and 800 ◦C. XRD, Raman and UV–vis spectroscopies have allowed
us to establish that at temperatures higher than 700 ◦C occurs the formation of (WO)x/ZrO2 nanoparticles
on the surface of MCM-41 support, which exhibit acidic properties and are active in the esterification
of oleic acid with methanol at 65 ◦C. WO3 loadings of 15–20 wt%, after activation at 700 ◦C, lead to the
most active catalysts, with conversion values close to 100%. The catalyst with 15 wt% WO3 is stable even
when the reaction is carried out at 200 ◦C, reusable at least during four cycles and no leaching of tungsten
species in the liquid phase was found.
Received 15 December 2009
Received in revised form 24 February 2010
Accepted 1 March 2010
Available online 6 March 2010
Keywords:
Esterification
Oleic acid
Tungstated zirconia
Mesoporous solids
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
preferable, since it offers an easy separation of solid catalysts from
reactants and products by filtration and the possibility of using a
The abusive use of petroleum is increasing air pollution levels
and accelerating global warning due to the increase in atmospheric
CO2. Recently, important efforts are being made in many countries
to develop clean alternative fuels from renewable resources [1,2].
Among the potential biofuels, biodiesel derived from vegetable oil
is revealing as a promising substitute for conventional diesel fuel.
Furthermore, biodiesel does not contain sulphur compounds, thus
leading to zero SOx emissions after combustion. On the other hand,
biodiesel is an oxygenated fuel (more than 10% oxygen), which
could increase the rate of fuel combustion, reducing the produc-
tion of CO and soot. However, the enhanced combustion of biodiesel
increases the temperature in the engine cylinders giving rise to an
increment in the NOx emissions compared to those produced by
petro–diesel.
Biodiesel consists in a mixture of alkyl esters of fatty acids
formed by transesterification of vegetable oils or animal fats
with methanol or ethanol [3,4]. The rate of the transesterification
reaction is faster employing a base catalyst, and for this reason
the industrial process uses, as catalysts, alkaline hydroxides or
methoxides dissolved in methanol, under homogeneous condi-
tions. However, this method implies some problems as the removal
of the base after reaction and biodiesel purification, which generate
an important volume of aqueous wastes. Heterogeneous catalysis is
continuous reactor operation [5–7]. Recently, we have developed
new heterogeneous base catalysts, such as Mg/M (M = Al and Ca) [8],
CaO stabilized in siliceous SBA-15 [9], and the system CaO/ZnO, pre-
pared from calcium zincate dihydrate as precursor, which is very
active [10].
In biodiesel production, the cost of raw materials, mainly triglyc-
erides, represents the majority of the overall production cost.
However, there are large amounts of low cost feedstock, like
non-edible oils, fried waste oils and animal fats which could be con-
verted to biodiesel, thus lowering the production costs. At the same
time, the process of biodiesel production should be rendered more
environmentally friendly. The problem of processing low cost feed-
stocks lies in their large amount of free fatty acids (FFA) that cannot
be converted into biodiesel using a base catalyst [11]. Thus, neu-
tralization of FFAs must be carried out by addition of a base excess,
but this leads to the formation of soaps difficult to separate, hence
requiring a high consumption of water and additional separation
steps. A suitable pre-treatment for this purpose is the previous
esterification of these FFAs by employing hydrochloric or sulphuric
acids as catalysts, but this requires the use of corrosion-resistant
materials and ulterior neutralization before the homogeneous or
heterogeneous transesterification processes in the presence of a
base catalyst, but this process generates large amounts of salts for
subsequent environmental disposal [12].
For this reason, the use of solid acid catalysts offers many advan-
tages; they have especially the ability to simultaneously catalyze
the esterification and transesterification reactions, and many of
∗
Corresponding author. Tel.: +34 952131876; fax: +34 952137534.
E-mail address: ajimenezl@uma.es (A. Jiménez-López).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.03.001

62
I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
them have been already reported in the literature to be active in
both reactions. Thus, Nafion resins maintain high catalytic activi-
ties, although the reaction temperature must be lower than 280 ◦C
in order to preserve resin stability [13–15]. Another solid acid cat-
alyst very active in these reactions is sulphated zirconia, but it
presents leaching of sulphur at even mild temperatures [13,16].
Tungsten oxide-based materials comprise another interesting
class of acid solids, first reported as a strongly acidic system by
Hino and Arata [17]. These authors found catalytic activity for
isomerization of light alkanes with high cracking selectivities a low-
temperature, which depended on WOx surface density, oxidation
temperature and ZrO2 area [18]. The formation of superficial WOx
domains on ZrO2 of intermediate sizes appears to be necessary to
delocalize a charge imbalance of these heteropolytungstate clus-
ters compensated by Brönsted acid sites. The acid properties of this
system are enhanced by addition of small amount of noble metal
(<1 wt% Pt) due to the partial reduction of some W(VI) species, thus
leading to the formation of more Brönsted acid sites in a reactant
environment containing H2 or hydrocarbons [19].
However, the esterification of fatty acids and the transesterifica-
tion of triglycerides using acid catalysts have been scarcely studied.
Furuta et al. reported that soybean oil can be converted to biodiesel
at 250 ◦C, reaching maximum conversion when tungstated zirconia
alumina activated at 800 ◦C is used [20]. These results are in agree-
ment with those reported by López et al., which found 800 ◦C as
optimum calcination temperature [21]. At higher temperatures, the
formation of crystalline WO3 in detrimental of heteropolytunstates
clusters leads to decay in the catalytic activity. However, Ramu et
al. have studied the activity of this family of catalysts in the esterifi-
cation of palmitic acid with methanol and they reached the highest
activities after activation at 400–500 ◦C, where tetragonal zirconia
and amorphous tungsten oxide coexist [22].
On the other hand, the use of MCM solids as catalytic support
has resulted in significant improvements for many reactions when
compared to conventional and commercial catalysts. The family of
mesoporous MCM-41 presents attractive properties to be used as
support, such as a narrow pore size distribution, very high specific
surface areas and high thermal stability [23,24]. Recently, many
catalytic reactions have been successfully studied by using meso-
porous pure or heteroatoms-doped silica as support of diverse
active phases [25–27].
In this paper, we describe the preparation of a series of cata-
lysts based on WOx supported on zirconium doped mesoporous
MCM-41 silica with the aim of obtaining highly dispersed WOx on
the pore walls of this support. The influence of parameters such as
activation temperature and WO3 loading on the acid properties has
been studied; finally, the catalytic behaviour in the esterification of
oleic acid with methanol has been evaluated.
study of the different WO3 loading was carried out after thermal
activation at 700 ◦C. The catalysts were labelled as x-WO3-T, where
x is the weight percentage of WO3 and T the activation temperature.
2.2. Characterization of catalysts
X-ray powder diffraction patterns were obtained by using an
XˇıPert Pro MPD automated diffractometer equipped with a Ge
(1 1 1) primary monochromator (strictly monochromatic Cu K␣1
radiation) and an XˇıCelerator detector. UV–visible spectroscopy
studies were carried out in the diffuse reflectance mode with a
Shimadzu MPC3100 spectrophotometer and BaSO4 as reference.
Raman spectra were recorded on a Raman Senterra (Bruker)
micro-spectrometer equipped with a thermoelectrically cooled
CCD detector. Excitation radiation at 532 nm was used as supplied
from an Argon laser. Raman spectra were performed from powder
samples without any previous treatment.
The textural parameters were evaluated from the nitrogen
adsorption–desorption isotherms at −196 ◦C, as determined by an
automatic ASAP 2020 system from Micromeritics. Prior to nitro-
gen adsorption, samples were evacuated at 400 ◦C (heating rate
10 ◦C min−1) for 1 h. Pore volume was determined with the BJH
method.
Temperature-programmed desorption of ammonia (NH3-TPD)
was carried out to evaluate the total acidity of the catalysts. Cata-
lysts were pre-treated at atmospheric pressure by flowing helium
(35 mL min−1) from room temperature to 550 ◦C, with a heat-
ing rate of 10 ◦C min−1, and maintaining the sample at 550 ◦C for
10 min. Then, samples were cooling until 100 ◦C under a helium
flow and after adsorption of ammonia at this temperature, the NH3-
TPD was performed between 100 and 550 ◦C with a heating rate of
10 ◦C min−1, by using a helium flow and maintained at 550 ◦C for
15 min. The evolved ammonia was analyzed by on-line gas chro-
matography (Shimadzu GC-14A) provided with a TCD.
The catalytic isomerization reaction of 1-butene was performed
in a tubular glass flow microreactor. Samples (70 mg) were pre-
treated for 2 h in a helium flow (30 mL min−1) at 400 ◦C and
the experiments were carried out at this temperature, with a
ꢀ = 67.2 gcat (g1-but)
−1 h and time on stream of 120 min. The 1-
butene and the reaction products were analyzed on-line in a gas
chromatograph (Shimadzu GC-14B) equipped with a wide-bore
KCl/AlCl3 column and provided with a FID detector.
X-ray photoelectron spectra were collected using a Physical
Electronics PHI 5700 spectrometer with non-monochromatic Al K␣
radiation (300 W, 15 kV, 1486.6 eV) with a multi-channel detector.
Spectra of pelletized samples were recorded in the constant pass
energy mode at 29.35 eV, using a 720 ␮m diameter analysis area.
Charge referencing was measured against adventitious carbon (C 1s
at 284.8 eV). A PHI ACCESS ESCA-V6.0 F software package was used
for acquisition and data analysis. A Shirley-type background was
subtracted from the signals. All recorded spectra were always fitted
using Gaussian–Lorentzian curves to more accurately determine
the binding energy of the different element core levels. Because
the W 4f signal partially overlaps the Zr 4p signal from the sup-
port, when analysing the W 4f signal of W-containing catalysts, the
contribution due to Zr 4p signal of the support was considered.
2. Experimental
2.1. Catalysts preparation
The synthesis of a zirconium doped mesoporous silica with a
Si/Zr molar ratio of 5, hereinafter Zr-MCM-41, used as support,
has been previously described [28]. Before impregnation with the
active phase, the support was steamed at 190 ◦C during 4 h to gen-
erate surface OH groups. The incorporation of tungsten species was
performed by incipient wetness impregnation by using ammonium
metatungstate aqueous solutions. The concentration of the pre-
cursor solution was adjusted to achieve WO3 percentages in the
resulting catalyst ranging between 5 and 25 wt%. After impregna-
tion, all materials were dried at 60 ◦C. The influence of the activation
temperature was followed by activation of the sample containing
10 wt% of WO3 at several temperatures (450–800 ◦C) during 2 h. The
2.3. Esterification of oleic acid with methanol
The esterification of oleic acid with methanol was carried out
in liquid phase at atmospheric pressure. In a typical experiment,
3 mL of oleic acid was put in a 200 mL round bottom flask and then
25 mL of dry methanol was added. When the mixture of reactants
was at 65 ◦C, 0.5 g of catalyst was added and the reaction mix-
ture was refluxed under vigorous stirring. Aliquots (ca. 0.75 mL)
were taken at different reaction times. After filtration, 0.15 mL

I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
63
of solution was diluted until 2 mL with 2-propanol/hexane (5:4
v/v). The resulting solution was analyzed by high performance liq-
uid chromatography (HPLC) using a JASCO liquid chromatograph
equipped with quaternary gradient pump (PU-2089), multiwave-
length detector (MD-2015), autosampler (AS-2055), column oven
(co-2065) using a PHENOMENEX LUNA C18 reversed-phase col-
umn (250 mm × 4.6 mm, 5 ␮m). The solvents were filtered through
a 0.45 ␮m filter prior use and degassed with helium. A linear gradi-
ent from 100% methanol to 50% methanol + 50% 2-propanol/hexane
(5:4 v/v) in 15 min was employed. Injection volumes of 15 ␮L and
a flow of rate of 1 mL min−1 were used. The column temperature
was held constant at 40 ◦C.
3. Results and discussion
3.1. Catalyst characterization
Zirconium doped MCM-41 is a mesoporous solid previously
studied by different characterization techniques [28]. Thus, it was
found by EXAFS that zirconium exhibits a coordination number
of 7–8, compatible with a preferential location of these species
at the pore surfaces. We have prepared a series of catalysts
containing different WO3 loading (5–25 wt%) supported on this
mesoporous solid, which were also activated at different tempera-
tures (450–800 ◦C).
As described in Section 2, Zr-MCM-41 was impregnated with
ammonium metatungstate by using the incipient wetness method.
In aqueous solutions of this ammonium salt, the following equilib-
rium takes place [29]:
Fig. 2. XRD patterns of catalysts with different WO3 loadings, after activation at
700 ◦C.
of the mesoporous MCM-41 solid, whereas the second group cor-
responds to samples activated at higher temperatures. These XRD
patterns exhibit broad reflections at 2Â = 29.9◦, 34.4◦, 50.3◦ and
60.0◦, characteristics of tetragonal ZrO2. Since pure Zr-MCM-41 is
thermally stable even after calcination at 900 ◦C, the presence of
these reflections clearly indicates that at high temperatures exists
a strong interaction between (WO)x and the support, leading to the
extraction of ZrO2 from the mesoporous support. However, in this
series of catalysts, the diffraction peaks of WO3 are not observable,
revealing a good dispersion of this phase. After thermal activation
of the catalysts, the mesoporous structure of the support is main-
tained since DRX patterns at low angles show a broad and intense
peak centred at 2Â = 2◦, similar to the pristine support.
W
12O4110− + 21H2O ꢀ 12WO42− + 14H3O+
After addition of the support to this solution, the anchoring of
W(VI) oxo-species, preferentially in the monomeric form, occurs
on the support surface. After calcination, the polymerization of
monomeric units by formation of W–O–W bridges leads to (WO)x
clusters. When these (WO)x clusters are supported on ZrO2 the
resulting system can accommodate protons due to the existence
of the negative charges delocalizated across the extended W–O
network, similarly to that observed for heteropolyacids [30–32].
The X-ray diffractograms at high angle of the catalysts contain-
ing 10 wt% of WO3 after activation at different temperatures (Fig. 1)
point to the existence of two groups of samples; the first one formed
by catalysts activated between 450 and 700 ◦C, whose X-ray pat-
terns only present the typical broad band of amorphous silica walls
Fig.
2 exhibits the diffraction patterns of catalysts after
activation at 700 ◦C. Among these diffractograms, only those cor-
responding to samples containing WO3 loading above 10 wt%
present peaks at 23.1◦, 23.6◦ and 24.6◦, which are associated to
WO3, together with the peaks of tetragonal zirconia. These results
demonstrate that the interaction between (WO)x and superficial
ZrO2 on the pores of Zr-MCM-41 is very strong, especially at
high activation temperatures and for high WO3 loading, leading
to the extraction of this superficial zirconia and the formation
of (WO)x/ZrO2 nanoparticles together with crystalline WO3. This
strong interaction between (WO)x and zirconia has been already
observed by López et al. for a catalyst containing 16 wt% of WO3
after calcination at temperatures above 750 ◦C, where the forma-
tion of crystalline WO3 was found, being the interaction between
(WO)x and the support strong enough to prevent the transfor-
mation of tetragonal zirconia into the monoclinic form [21]. This
behaviour is different to that found for a similar system formed by
WO3 supported on Zr doped SBA-15, where the catalyst is stable
after calcination at high temperatures and no segregation of ZrO2
particles is observed [33].
The presence of pure WO3 particles on the support surface for
catalysts with WO3 loading higher than 10 wt% could be explained
by taking into account that Zr-MCM-41 contains only 21.5 wt% of
ZrO2, and, for WO3 loading of 15 wt% or higher, there is an excess of
active phase in comparison with conventional catalysts prepared
supporting this phase on pure zirconia, where a WO3 loading of
16 wt% already favours the formation of crystalline WO3 at high
activation temperatures [21,32].
UV–vis diffuse reflectance absorption spectra of catalysts con-
taining different WO3 loading and activated at 700 ◦C (Fig. 3) show
a sharp absorption between 210 and 260 nm due to the ligand to
metal charge transfer (O2p–W5d–O2p), being both tetrahedral and
Fig. 1. XRD patterns of catalysts with 10 wt% of WO3, activated at different temper-
atures.

64
I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
species also exhibit Raman bands below 680 cm−1, but overlapping
with those from zirconia. At the same time, bands corresponding to
crystalline WO3 appear at 807, 720 and 273 cm−1, assigned to W–O
stretching, W–O bending and W–O–W deformation modes, respec-
tively [38]. Raman scattering cross section for crystalline WO3 are
much intense than for surface polytungstate species, as a result
those bands dominate the spectra, even being the polytngstate
species the most abundant [39]. These results are in agreement
with those found by XRD and UV–vis spectroscopy where the pres-
ence of diffraction lines of WO3 and red-shifted absorptions were
observed for loadings higher than 15%. However, the Raman spec-
tra reveal that in all the catalysts the formation of crystalline WO3
is detected, but their crystal sizes in catalysts with low WO3 loading
are below the detection limit of the X-ray diffraction technique.
In order to get information about the chemical state of tungsten
in these compounds, XPS spectra of catalysts have been obtained.
The binding energy values corresponding to the Zr 3d, Si 2p and
O 2s appear at 182.5, 103.5 and 532.7 eV, respectively, in accor-
dance with the values previously found by Eliche-Quesada et al. for
this support [40]. All XPS spectra in the W 4f region are similar to
that shown in Fig. 5. The core level spectrum displays two over-
lapped peaks corresponding to W 4f 7/2 and W 4f 5/2 with binding
energies of 36.0 and 37.8 eV, respectively. From these values and
the difference in energy, it can be inferred the exclusive presence
of W(VI), thus precluding the existence of W(V) [41,42]. However,
some authors have found this oxidation state for tungsten, owing to
its partial reduction during the calcination step, possibly by oxygen
losing from the framework [43,44].
On the other hand, superficial W/Zr atomic ratios of catalysts
have been calculated from XPS measurements. The theoretical
W/Zr bulk ratio for the catalysts with 10 wt% of WO3 is 0.20;
however, for the catalysts with this WO3 loading and activated
at different temperatures (Table 1) the surface W/Zr atomic ratios
vary between 0.12 and 0.18. These results reveal a good dispersion
of the active phase, and a superficial enrichment of tungsten which
is favoured at high activation temperatures, due to the formation
of polytungstates or WO3, as can be inferred from XRD, Raman and
UV–vis absorption data. For catalysts with different WO3 loadings
and activated at 700 ◦C, only surface W/Zr atomic ratios are lower
than the theoretical ones for catalysts with 5 and 10 wt% of WO3. For
catalysts with higher loadings, both theoretical and surface W/Zr
atomic ratios match well, indicating that, for high loadings and after
activation at 700 ◦C, the formation of these polytungstates and WO3
is more easily accomplished.
Fig. 3. Diffuse reflectance UV–vis absorption spectra of the catalysts with WO3
loadings of 5–25 wt% after activation at 700 ◦C.
octahedral W(VI) species responsible of these absorptions [34,35].
The mesoporous support also absorbs at 210 nm. On the other hand,
a broad band between 300 and 330 nm appears in all catalysts,
which becomes more intense and shifts to the red with increas-
ing the WO3 loading. This band could be assigned to the formation
of two-dimensional polytungstate on zirconia by W–O–W bonds
between well dispersed (WO)x species. These extended domains of
W–O–W leads to a narrowing of the HOMO–LUMO gap [36]. The tail
of these spectra at 350 nm could correspond to the microcrystalline
WO3.
The catalysts with 10 wt% of WO3, activated at different temper-
atures, show similar spectra, i.e., containing the same absorption at
210 and 260 nm, whereas that at 300–330 nm shifts to low energies.
Therefore, an increasing of the activation temperature and the WO3
loading favour the extraction of ZrO2 from the pore walls and the
formation of (WO)x/ ZrO2 nanoparticles or microcrystalline WO3 on
the support surface. Both effects tend to increase the (WO)x density
and domain size.
Raman spectra of catalysts with different WO3 loading and acti-
vated at 700 ◦C are shown in Fig. 4. In all cases, a Raman feature at
1019 cm−1 is observed which is assigned to the symmetric stretch-
ing mode of terminal W O bonds present in monotungstate and
polytungstates (WOx) species at the surface of ZrO2 [37]. WOx
Fig. 4. Raman spectra of catalysts activated at 700 ◦C: bulk WO3 (a), 5 wt% WO3 (b),
10 wt% WO3 (c), 15 wt% WO3 (d), 20 wt% WO3 (e) and 25 wt% WO3 (f).
Fig. 5. W 4f core level spectrum of the 15-WO3-700 catalyst.

I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
65
Fig. 6. NH3-TPD of catalysts with 10 wt% of WO3, activated at different temperatures (A) and catalysts with different WO3 loadings, activated at 700 C (B).
Table 1
attributed to the presence of nanoparticles of (WO)x/ZrO2 in the
pores and on external surface, which are partially blocking the
access to mesopores. Therefore, the greater is the activation tem-
perature and the WO3 loading the greater the decrease in the
textural parameters. Nevertheless, the average pore diameters
became larger for catalysts activated at 750 and 800 ◦C and for those
having 20–25 wt% of WO3, which could be attributed to the gen-
eration of additional porosity associated to (WO)x/ZrO2 and WO3
particles located on the external surface of support, as previously
observed for other supported phases [45–47].
The acid properties of this family of catalysts were evaluated
by NH3-TPD, and the results obtained for catalysts with 10 wt% of
WO3 and activated at different temperatures are plotted in Fig. 6A.
As can been observed, the total acidity of these catalysts decreases
when the activation temperature rises. The extraction of zirconia
and the subsequent formation of (WO)x/ZrO2 nanoparticles located
on the external surface hinders the access of ammonia molecules
to the pores. On the other hand, this decrease in the surface acidity
is parallel to that of the surface area values. Table 2 also display the
acidity values, expressed as ␮mol of NH3 desorbed per surface unit
(square meter), for this series of catalysts, which are close to one for
all of them. However, the acid properties of catalysts with differ-
ent WO3 loading show a different variation (Fig. 6B). In fact, these
acidity values became higher for catalysts with greater WO3 load-
ing. That means that the (WO)x/ZrO2 nanoparticles, which increase
with the WO3 content, as deduced from XRD, Raman and UV–vis
spectroscopies, could contribute to the total acidity.
W/Zr atomic ratios, obtained from XPS analysis, of catalysts.
Catalyst
W/Zr atomic ratio
5-WO3-700
10-WO3-700
15-WO3-700
20-WO3-700
25-WO3-700
10-WO3-450
10-WO3-550
10-WO3-650
10-WO3-700
10-WO3-750
0.10
0.16
0.32
0.45
0.63
0.12
0.13
0.16
0.16
0.18
The textural parameters of the support and catalysts are com-
piled in Table 2. The nitrogen adsorption isotherms are of Type
IV in the IUPAC classification and similar to that observed in the
case of the support. However, the incorporation of the active phase
progressive decreases the amount of nitrogen adsorbed, and the
specific surface area and pore volume were lower to those cor-
responding to the mesoporous support. This decrease could be
Table 2
Textural properties of WO3 supported on Zr-MCM-41 catalysts.
´
˚
Catalyst
SBET (m2 g−1
)
Vp (cm3 g−1
)
dp (A)
Acidity
(␮mol NH3 des/m2)
Zr-MCM-41
10-WO3-450
10-WO3-550
10-WO3-650
10-WO3-700
10-WO3-750
10-WO3-800
535
442
524
403
355
330
253
0.62
0.56
0.69
0.54
0.49
0.56
0.45
46.4
57.3
52.7
53.6
55.2
67.9
71.1
1.01
1.05
1.13
1.00
0.99
0.93
1.00
By considering that the esterification reaction of an organic acid
with an alcohol via heterogeneous catalysis is primarily catalyzed
by protons, it is interesting to study the behaviour of this family
of acid catalysts in the 1-butene isomerization reaction in order to
gain insights about the strength of their Brönsted acid sites [21]. The
isomerization of 1-butene can occur through a first step consisting
in a double bond shifting, giving cis- and trans-2-butenes, which
easily reaches the thermodynamic equilibrium and needs strong
acid sites with Hammett function of 0.82 > H0 > −4.04 [48]. In a sec-
ond step, methyl migration gives rise to iso-butene or iso-butane.
5-WO3-700
10-WO3-700
15-WO3-700
20-WO3-700
25-WO3-700
414
355
328
254
224
0.59
0.49
0.45
0.39
0.37
57.3
55.2
55.1
61.8
66.7
1.04
0.99
1.24
1.58
1.49

66
I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
Table 3
1-Butene isomerization data obtained at 400 ◦C, after 2 h of time on stream.
Catalyst
Conversion (%)
Cis-2-butene (%)
Trans-2-butene (%)
Iso-butene (%)
5-WO3-700
10-WO3-700
15-WO3-700
20-WO3-700
25-WO3-700
67.4
67.9
77.4
70.5
70.1
40.9
40.9
40.0
40.5
39.8
56.9
54.3
51.8
55.4
53.9
2.2
4.8
8.3
4.2
6.4
10-WO3-800
10-WO3-750
10-WO3-700
10-WO3-550
67.1
68.9
67.9
69.3
42.1
40.9
40.9
40.6
54.8
51.2
54.3
54.1
3.1
7.9
4.8
5.3
This second isomerization requires stronger acid sites (H0 < −6.63)
and the reaction takes place on Brönsted acid sites via carbenium
ion intermediates in a three-step mechanism [49].
The experimental values, obtained after 2 h of time on stream
at 400 ◦C, reveal that all catalysts are active in this reaction, being
more selective toward the formation of double bond isomeriza-
tion products, indicating the presence of Brönsted acid sites of mild
strength (Table 3). The highest conversion and selectivity toward
iso-butene are found for catalysts with WO3 contents of 15 and
20 wt%, thus being the most acid catalysts. These catalysts are also
those presenting high acidity expressed as ␮mol of NH3 desorbed
per square meter (Table 2).
3.2. Esterification of oleic acid with methanol
Fig. 8. Influence of the activation temperature on the conversion for catalysts with
10 wt% of WO3 (reaction conditions: methanol/oleic acid molar ratio = 67, cata-
lyst = 18.7 wt% and T = 65 ◦C, reaction time = 24 h).
This family of acid catalysts has been tested in the esterifica-
tion of oleic acid with methanol at 65 ◦C. The amount of catalyst
used was 18.7 wt% with respect to the weight of oleic acid and
the methanol/oleic acid/molar ratio was 67. The evolution of the
conversion against reaction time is shown in Fig. 7 for catalysts con-
taining different WO3 contents, activated at 700 ◦C. From this plot,
it can be deduced that the reaction rate is low under the selected
experimental conditions, since 24 h are necessary to attain high
conversions. Moreover, it is noteworthy that only catalysts acti-
vated at 700 and 750 ◦C reached high conversion, with values of
67%, after 24 h (Fig. 8). From these results, an activation temper-
ature of 700 ◦C was selected to activate the other catalysts with
different WO3 loading. Conversions close to 100% are only reached
for catalysts with WO3 loading higher than 10 wt% (Fig. 9), after
24 h of reaction time. The selectivity of the reaction toward methyl
ester was always 100%.
That means that at these temperatures occurs the extraction of
superficial ZrO2 with the concomitant formation of (WO)x/ZrO2
nanoparticles, which (WO)x surface density is suitable to gener-
ate Brönsted acid sites able to catalyze the esterification reaction.
However, an activation temperature of 800 ◦C decreases the con-
version, possibly due to the loss of OH groups in the process of
formation of crystalline WO3 which is inactive, according to Barton
et al. [31]. These authors have demonstrated that in WO3 domains
a fraction of the protons are located inside the WO3 crystals, thus
being inaccessible to the reactant molecules.
On the other hand, the conversion depends on the WO3 loading;
in fact, only catalysts containing 15–25 wt% of WO3 are very active.
This fact points out that the formation of (WO)x/ZrO2 nanoparti-
cles, after thermal activation at 700 ◦C, is important when the WO3
loading increases, as deduced from XRD patterns where diffraction
peaks of ZrO2 were observable together with those of crystalline
These results suggest that the catalytic behaviour depends on
both the thermal activation of catalysts and the WO3 loading. Con-
cerning the activation temperature, only catalysts with 10 wt% of
WO3 activated at temperatures higher than 700 ◦C are quite active.
Fig. 7. Evolution of the conversion as a function of time on stream for catalysts with
different WO3 contents, activated at 700 ◦C (reaction conditions: methanol/oleic
acid molar ratio = 67, catalyst = 18.7 wt% and T = 65 ◦C).
Fig. 9. Evolution of the conversion for catalysts with different WO3 loadings and
activated at 700 ◦C (reaction conditions: methanol/oleic acid molar ratio = 67, cata-
lyst = 18.7 wt% and T = 65 ◦C, reaction time = 24 h).

I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
67
WO3. The high conversion observed for these catalysts is expected
by considering the high acidity deduced from NH3-TPD and the
formation of iso-butene in the isomerization of 1-butene, thus
revealing as the most acid catalysts. However, for the catalyst with
25 wt% of WO3 the conversion is lower, possibly due to the presence
of crystalline WO3 phase. It is necessary to consider, as previously
noted, that Zr-MCM-41 contains only 21.5 wt% of ZrO2, and thus
a 25 wt% of WO3 loading represents an important excess and as
a consequence tungsten may be as crystalline WO3, which is not
active.
experiment was repeated four times and in all cases the conver-
sion was maintained close to 97%, thus revealing that catalysts are
very stable against lixiviation and poisoning, even under drastic
experimental conditions.
Therefore, these active catalysts, based on (WO)x/ZrO2 as active
phase, are potential candidates to catalyze the transesterification
reaction of vegetable oils with methanol, and, at high temperatures
(200 ◦C), it could be thought that they could simultaneously favour
both reactions, the esterification of FFAs and the transesterification
of used oils with methanol in a single step, thus suggesting their
potential industrial applications in biodiesel preparation from low
cost feedstocks like used cooking oils.
The catalytic behaviour of our family of catalysts could be
explained by taking into account both the different catalyst prepa-
ration method and the nature of the vegetable oil used, since, for
instance, Ramu et al. [22] studied the esterification of palmitic acid
with methanol. These authors found the higher catalytic activity
after activation of the catalysts at 400–500 ◦C. Thus, the different
behaviour could be attributed to the different nature of the sup-
port employed. The use of the Zr-MCM-41 support requires high
temperatures and high WO3 loadings to generate the (WO)x/ZrO2
nanoparticles, which are active in this esterification reaction. How-
ever, in both works, the maximum TOF attained is very similar
4. Conclusions
The use of Zr-MCM-41 as a support for the preparation of cat-
alysts containing WO3 leads to the segregation of (WO)x/ZrO2
nanoparticles when the catalyst precursors are activated at tem-
peratures higher than 700 ◦C, and especially for high WO3 loading.
These nanoparticles are acid, as deduced from NH3-TPD and the
isomerization reaction of 1-butene, and they are active in the ester-
ification reaction of oleic acid with methanol at 65 ◦C. The catalysts
with WO3 loading of 15 and 20 wt%, after thermal activation at
700 ◦C, are the most active, giving rise to conversions close to 100%.
The catalyst with 15 wt% WO3 is stable even when the reaction is
run at 200 ◦C, reusable at least during four cycles and no leaching
of tungsten species in the liquid phase was found.
−1
and close to 0.8 mmol h
g
cat
−1. Nevertheless, our esterification
results are better than those reported by Kim and co-workers
[50] in the esterification of palmitic acid with methanol (TOF of
−1
0.06 mmol h
g
cat
−1), which can be explained by considering that
these authors accomplished the esterification of oleic acid with
methanol in the presence of an excess of soybean oil, and the
transesterification reaction also takes place. On the other hand, in
general, it is difficult to compare our results with those reported in
literature, because the experimental conditions employed in each
case are different, such as the reaction temperature [51]. Our cata-
lysts exhibit similar catalytic activity than those reported by Furuta
et al. [20] with similar conversion after 20 h, but working in the last
case at reaction temperatures ranging between 175 and 200 ◦C. Rao
et al. prepared a series of catalysts by a surface grafting method
of (WO)x on zirconium phosphate [52], which exhibited excellent
Acknowledgements
The authors are grateful to financial support from the Span-
ish Ministry of Science and Innovation (ENE2009-12743-C04-03
project) and Junta de Andalucía (P06-FQM-01661 and P09-FQM-
5070). IJM would like to thank the Spanish Ministry of Science
and Innovation for a JAE-Predoctoral grant (Research Training, JAE
Programme).
−1
activity and high TOF (21.9 mmol h
g
cat
−1), presenting potential
application in industrial biodiesel production, but this system is
different to a tungstated zirconia, being more comparable to a sup-
ported heteropolyacid.
References
[1] A. Dermirbas, Biodiesel:
Springer, London, 2008.
[2] Z. Helwani, M.R. Othman, N. Aziz, J. Kin, Fuel Process. Technol. 90 (2009)
1502–1514.
[3] H. Fukuda, A. Kondo, H. Noda, J. Biosci. Bioenerg. 92 (2001) 405–416.
[4] F. Ma, M.A. Hanna, Bioresour. Technol. 70 (1999) 1–15.
[5] Y. Ono, T. Baba, Catal. Today 38 (1997) 321–337.
[6] J.H. Clark, D.J. Macquarrie, Chem. Soc. Rev. 25 (1996) 303–310.
[7] H.J. Kim, B.S. Kang, M.J. Kim, D.K. Kim, J.S. Lee, K.Y. Lee, EuropaCat IV, Innsbruck,
Austria, B1, 2003, p. 089.
[8] M.C.G. Alburquerque, J. Santamaría-González, J. Mérida-Robles, R. Moreno-
Tost, E. Rodríguez-Castellón, A. Jiménez-López, D.C.S. Azevedo, C.L. Cavalcante
Jr., P. Maireles-Torres, Appl. Catal. A 347 (2008) 162–168.
[9] M.C.G. Alburquerque, I. Jiménez.-Urbistondo, J. Santamaría-González, J.
Mérida-Robles, R. Moreno-Tost, E. Rodríguez-Castellón, A. Jiménez-López,
D.C.S. Azevedo, C.L. Cavalcante, P. Maireles-Torres, Appl. Catal. A 334 (2008)
35–43.
[10] J.M. Rubio-Caballero, J. Santamaría-González, J. Mérida-Robles, R. Moreno-Tost,
A. Jiménez-López, P. Maireles-Torres, Appl. Catal. B 91 (2009) 339–346.
[11] M. Canakei, J.V. Germen, Trans. Am. Soc. Agric. Eng. 44 (2001) 1429–1436.
[12] A. Corma, H. García, Catal. Today 38 (1997) 257–308.
[13] D.E. López, J.G. Goodwin Jr., D.A. Bruce, E. Lotero, Appl. Catal. A 295 (2005)
97–105.
[14] D.E. López, J.G. Goodwin Jr., D.A. Bruce, J. Catal. 245 (2007) 381–391.
[15] Y. Liu, E. Lotero, Goodwin Jr., J. Catal. 242 (2006) 278–286.
[16] J. Jitputti, B. Kitiyanan, P. Kapteijn, K. Bunyakiat, L. Attanatho, P. Jenvanitpan-
jakul, Chem. Eng. J. 116 (2006) 61–66.
A Realistic Fuel Alternative For Diesel Engines,
Finally, to improve the catalytic activity of these catalysts, espe-
cially to reduce the reaction time, the influence of the catalyst
amount and the reaction temperature on the catalytic performance
have been evaluated. Thus, by increasing the amount of catalyst up
to 37.4 wt%, the conversion is 80% after 15 h of reaction. However,
more significant is the effect of the reaction temperature, since at
200 ◦C, using only 18.7 wt% of catalysts, the conversion is close to
100% after 4 h of reaction.
A key aspect in the preparation of biodiesel is the evaluation of
the degree of lixiviation of the active phase. With this purpose we
have analyzed by ICP the filtered liquids after the reaction of ester-
ification of oleic acid with methanol at 200 ◦C and the amount of W
found in solution is lower than 0.3 ppm, pointing to that lixiviation
is negligible.
On the other hand, we have studied the influence of the water
on the reaction of esterification by adding 5.5 wt% of water to the
reactants at 200 ◦C. The conversion was maintained close to 97%,
pointing to that water is not adsorbed on the active centres and oleic
acid molecules can approach to produce the esterification reaction.
Moreover, the catalyst reutilization is a key parameter to con-
sider the viability of a solid catalyst for biodiesel production, since
it can reduce the overall process costs. In order to evaluate the
reusability of this family of catalysts, the 15-WO3-700 catalyst was
repeatedly used for the esterification of oleic acid with methanol
at 200 ◦C. After a first catalytic run, the catalyst was filtered and
reused without any treatment, such as washing or calcination. This
[17] M. Hino, K. Arata, J. Chem. Soc. Chem. Commun. (1988) 1259.
[18] K. Arata, M. Hino, in: M.J. Philips, M. Ternan (Eds.), Proceedings 9th International
Congress on Catalysis, Calgary Ottawa, 1988, pp. 1727–1730.
[19] E. Iglesia, D.G. Barton, S.L. Soled, S. Miseo, J.E. Baumgartner, W.E. Gates, G.A.
Fuentesand, G.D. Meiztner, Stud. Surf. Sci. Catal. 101 (1996) 533–542.
[20] S. Furuta, H. Matsuhashi, K. Arata, Catal. Commun. 5 (2004) 721–723.
[21] D.E. López, K. Suwannakarn, D.A. Bruce, J.G. Goodwin Jr., J. Catal. 247 (2007)
43–50.

68
I. Jiménez-Morales et al. / Applied Catalysis A: General 379 (2010) 61–68
[22] S. Ramu, N. Lingaiah, B.L.A. Prabhavathi Devi, R.B.N. Prasad, I. Suryanarayana,
P.S. Prasad, Appl. Catal. A 276 (2004) 163–168.
[23] J. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.W.
Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, J. Am.
Chem. Soc. 114 (1992) 10834–10843.
[36] D. Masure, P. Chaquin, C. Louis, M. Che, M. Fournier, J. Catal. 119 (1989) 415–425.
[37] S.S. Chan, I.E. Wachs, L.L. Murrell, L. Wang, W.K. Hall, J. Phys. Chem. 88 (1984)
5831–5835.
[38] K. Ohwada, Spectrochim. Acta A 26 (1970) 1035–1044.
[39] S.S. Chan, I.E. Wachs, L.L. Murrell, J. Catal. 90 (1984) 150–155.
[40] D. Eliche-Quesada, J. Mérida-Robles, P. Maireles-Torres, E. Rodríguez-Castellón,
A. Jiménez-López, Appl. Catal. A 262 (2004) 111–120.
[24] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J. Beck, Nature 359 (1992)
710–712.
[25] D. Eliche-Quesada, J. Mérida-Robles, P. Maireles-Torres, E. Rodríguez-Castellón,
G. Busca, E. Finocchio, A. Jiménez-López, J. Catal. 220 (2003) 457–467.
[26] A. Corma, A. Martínez, V. Martínez-Soria, J.B. Montón, J. Catal. 153 (1995) 25–31.
[27] O.C. Bianchi, M. Campanatti, P. Maireles-Torres, E. Rodríguez-Castellón, A.
Jiménez-López, A. Vaccari, Appl. Catal. A 220 (2001) 105–112.
[28] 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.
[41] J.F. Moulder, W.F. Stickle, P.E. Sool, K.D. Bomber, Handbook of X-Ray Photoelec-
tron Spectroscopy, Perkin-Elmer, Eden Prairie, MN, 1992.
[42] J.R. Sohn, M.Y. Park, J. Ind. Eng. Chem. 4 (1998) 84–93.
[43] M.A. Cortés-Jácome, C. Angeles-Chávez, X. Bokhimi, J.A. Toledo-Antonio, J. Solid
State Chem. 179 (2006) 2663–2673.
[44] F. Di Gregorio, V. Keller, J. Catal. 225 (2004) 45–55.
[45] M. Gómez-Cazalilla, A. Infantes, J. Mérida-Robles, E. Rodríguez-Castellón, A.
Jiménez-López, Energy Fuels 23 (2009) 101–110.
[29] B.H. Jeziorowski, H. Knözinger, J. Phys. Chem. 83 (1979) 1166–1173.
[30] D.G. Barton, S.L. Soled, E. Iglesia, Top. Catal. 6 (1998) 87–99.
[31] D.G. Barton, M. Shtein, R. Wilson, S.L. Soled, E. Iglesia, J. Phys. Chem. 103 (1999)
630–640.
[46] A. Infantes, J. Mérida-Robles, E. Rodríguez-Castellón, J.L.G. Fierro, A. Jiménez-
López, J. Catal. 240 (2006) 258–267.
[47] B. Pawelec, P. Castan˜o, J.M. Arandes, J. Bilbao, S. Thomas, M.A. Pen˜a, J.L.G. Fierro,
Appl. Catal. A 317 (2007) 20–33.
[32] S.L. Soled, G.B.G.B. Mc Vicker, L.L. Murrell, L.G. Sherman, N.C. Dispenziere Jr.,
S.L. Hsu, D. Waldman, J. Catal. 111 (1988) 286–295.
[33] Y.Q. Zhang, S.J. Wang, J.W. Wang, L.L. Lou, C. Zhang, S. Liu, Solid State Sci. 11
(2009) 1412–1418.
[34] A. Iannibello, S. Marengo, P. Tittarelli, G. Morelli, A. Zecchina, J. Chem Soc.
Faraday Trans. I 80 (1984) 2209–2223.
[35] M. Fournier, C. Louis, M. Che, P. Chaquin, D. Masure, J. Catal. 119 (1989) 400–414.
[48] P. Patrono, A. La Ginestra, G. Ramis, G. Busca, Appl. Catal. A 107 (1994) 249–266.
[49] J.P. Damon, B. Delmon, J.M. Monnier, J. Chem. Soc. Faraday Trans. I 73 (1977)
372–380.
[50] Y.M. Park, D.W. Lee, D.K. Kim, J.S. Lee, K.Y. Lee, Catal. Today 131 (2008) 238–243.
[51] H. Kim, B. Kang, D. Kim, J. Lee, K. Lee, Stud. Surf. Sci. Catal. 153 (2004) 201–204.
[52] K.N. Rao, A. Sridhar, A.F. Lee, S.J. Tavener, N.A. Young, K. Wilson, Green Chem.
8 (2008) 790–797.