Surface active sites in alumina-supported MoVNbTeO oxide catalysts

Article abstract of DOI:10.1016/j.cattod.2010.06.015

Methanol oxidation reaction probes the chemical nature of the active sites (formaldehyde forms on surface redox sites and dimethylether, on acidic ones) and has been used to characterize the active sites present on the surface of Mo-V-Nb-O catalysts which has been correlated with the activity for propane oxidation to acrylic acid. The effect of calcination atmosphere and the role of tellurium addition have been discussed as a function of the total coverage. The structure of Mo-V-O mixed phases strongly depends on these parameters, as determined by XRD, XPS and Raman spectroscopy. Such structures determine the catalytic behavior of catalysts during both propane and methanol oxidation. At low and medium coverages, the presence of oxidized dispersed surface vanadium structures (VOx) determines the catalytic behavior; such phases are active for methanol oxidation to formaldehyde. At high coverages, the formation of Al-Mo-O and Mo-V-O crystalline aggregates decreases the number of redox sites, whereas the presence of reduced species in rutile-like and/or MoO₂ increases the redox activity for inert-calcined samples. For samples with high coverage, tellurium addition decreases the activity during methanol oxidation. This is due to a decrease in the population of surface VOx species. The interaction of tellurium with molybdenum and vanadium sites results in mixed phases with intermediate oxidation state that exhibit higher yields to acrylic acid.

Full text of DOI:10.1016/j.cattod.2010.06.015

Catalysis Today 158 (2010) 139–145
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Surface active sites in alumina-supported MoVNbTeO oxide catalysts
R. López-Medinaa,b, H. Golinskab, M. Ziolekb, M.O. Guerrero-Pérezc, M.A. Ban˜aresa,∗
a Catalytic Spectroscopy Laboratory, Instituto de Catálisis y Petroleoquímica, CSIC, Marie Curie 2, E-29049 Madrid, Spain
b Adam Mickiewicz University, Faculty of Chemistry, Grunwaldzka 6, 60-780 Poznan´, Poland
c Departamento de Ingeniería Química. Escuela de Ingenierías., Universidad de Málaga, E-29071 Málaga, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 2 August 2010
Methanol oxidation reaction probes the chemical nature of the active sites (formaldehyde forms on
surface redox sites and dimethylether, on acidic ones) and has been used to characterize the active sites
present on the surface of Mo–V–Nb–O catalysts which has been correlated with the activity for propane
oxidation to acrylic acid. The effect of calcination atmosphere and the role of tellurium addition have
been discussed as a function of the total coverage. The structure of Mo–V–O mixed phases strongly
depends on these parameters, as determined by XRD, XPS and Raman spectroscopy. Such structures
determine the catalytic behavior of catalysts during both propane and methanol oxidation. At low and
medium coverages, the presence of oxidized dispersed surface vanadium structures (VOx) determines the
catalytic behavior; such phases are active for methanol oxidation to formaldehyde. At high coverages, the
formation of Al–Mo–O and Mo–V–O crystalline aggregates decreases the number of redox sites, whereas
the presence of reduced species in rutile-like and/or MoO2 increases the redox activity for inert-calcined
samples. For samples with high coverage, tellurium addition decreases the activity during methanol
oxidation. This is due to a decrease in the population of surface VOx species. The interaction of tellurium
with molybdenum and vanadium sites results in mixed phases with intermediate oxidation state that
exhibit higher yields to acrylic acid.
Keywords:
Methanol oxidation
Propane oxidation
Acrylic acid
Active sites
Mo–V–O catalysts
Surface properties
Nanocrystalline
© 2010 Elsevier B.V. All rights reserved.
1. Introduction and objectives
acrylonitrile (ammoxidation) [18,15,19]. Niobium oxide is an inter-
esting additive since it easily reacts with many other metal oxides
Understanding the factors that determine surface reactivity is
necessary to design new and more effective catalytic processes [1].
Methanol is a well-established probe to determine the number of
surface active sites present in metal oxide catalysts and the dis-
tribution of acid–basic and redox sites [1–5]. Methanol oxidation
over metal oxides produces different reaction products depend-
ing on the surface active site: formaldehyde on surface redox sites,
dimethylether on surface acidic sites, and CO2 on surface basic
sites [6–9]. Combination of acid and redox or basic and redox sites
will produce dimethoxymethane or methyl formate, respectively
[10,11].
Mo–V–O oxides are active and selective in several oxidation
reactions, like the direct conversion of propane into acrylonitrile or
acrylic acid, or the propane ammoxidation to acrylonitrile [12,17].
Upon addition of Te and/or Nb, these oxides can present two main
phases, named M1 and M2 [13–17]. The M1 phase is associated with
the selective activation of propane into acrylic acid (oxidation) and
resulting in phases that are important for the activation of alkanes
[15,17,20–22].
This paper studies several supported Te- and Nb-doped Mo–V–O
catalysts, and the role of total coverage and calcination under dif-
ferent atmospheres on their structural and reactive features. The
catalysts are characterized by XRD, XPS and Raman spectroscopy
whereas the nature of the active sites is analyzed by methanol
oxidation. These catalysts have been tested for propane partial oxi-
dation to acrylic acid.
2. Experimental
The synthesis of alumina-supported MoVNbTeO mixed metal
oxide catalysts were prepared from an aqueous slurry compris-
ing molybdenum (NH4)6Mo7O24·4H2O (Sigma–Aldrich), vanadium
NH4VO3 (Sigma–Aldrich), niobium (ammonium niobium soluble
complex; Niobium Products) and tellurium (telluric acid TeOH6
Sigma–Aldrich) in the appropriate atomic ratios with an surface
atoms density of 4, 8 and 12 MoVNbTe atoms/nm2 (A) denomi-
nated as “A”MoxVyNbwTez using tartaric acid as complexing agent.
The support (15 g ␥-Al2O3 Sasol Puralox SCCa-5/200) was added to
a solution containing the slurry and then was evaporated to dry-
∗
Corresponding author at: Catalytic Spectroscopy Laboratory, Instituto de Catáli-
sis y Petroleoquímica, CSIC; Marie Curie 2, Campus Cantoblanco, E-29049-Madrid,
Spain. Tel.: +34 91 585 4788; fax: +34 91 585 4760.
E-mail address: banares@icp.csic.es (M.A. Ban˜ares).
0920-5861/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2010.06.015

140
R. López-Medina et al. / Catalysis Today 158 (2010) 139–145
ness at 80 ◦C. The product was dried at 110 ◦C overnight in ambient
air. The thermal treatments in air or nitrogen were performed at
a heating rate of 5 ◦C/min up to 600 ◦C for 2 h. The catalysts are
denoted as “–air” and “–inert”, depending of this final heat treat-
ment in air or N2, respectively. For reference purposes, the same
procedure was used, without addition of alumina support, for the
preparation of a MoVNbTe rutile sample. All catalysts were sieved
to 0.5–1 mm.
The partial oxidation of methanol reaction was performed in
a tubular fixed-bed reactor with a coaxial thermocouple pocket
under atmospheric pressure. The reactor was placed in an electri-
cal oven for the experiments. The samples were activated in helium
flow (40 ml/min) at 400 ◦C for 2 h. The catalyst weight was 100 mg
and the reaction feed consisted of 40 ml/min of He/O2/MeOH
(88/8/4 mol%) (mass flow controller). The experiments were car-
ried out in the 250–300 ◦C reaction temperatures range. Heated
lines take reaction products to an on-line gas chromatograph GC
8000 Top equipped with a capillary column of DB-1 fitted to a FID
detector, and Porapak Q and 5A molecular sieves columns, to a TCD
detector.
be fitted into two components. Measurement error of the spectra
was 0.1 eV. Due to X-ray irradiation and vacuum conditions,
transition metal oxides tend to reduce partially. Thus, the average
oxidation state is moderately lower than in reality. However, it
is possible to compare relative population of different oxidation
states.
UV–vis DRS spectra are obtained by taking using a halon white
(PTFE) reflectance standard in an AvaLight DHS spectrometer
equipment with a Deuterium-Halogen light source, 215–2500 nm,
TTL shutter and SR fiber optic. Reflectance data were converted
to the absorption spectra using Kubelka–Munk function F(R). The
thickness of the catalyst bed was 2 mm, and the amount of the cat-
alyst used was around 20 mg. The ␥-Al2O3 support possesses no
absorbance in the region of interest.
3. Results
Table 1 lists the BET areas of the samples; which decrease with
total coverage, as expected. These are similar at low and medium
coverages, regardless of calcination treatments. At high coverages
(12 atoms/nm2), the BET surface area values obtained for the inert-
calcined series are significantly higher that those of the air-calcined
series.
The partial oxidation of propane was carried out at atmo-
spheric pressure in a flow-type fixed-bed microreactor. Typical
reaction conditions were: GHSV = 4800 h−1
;
catalyst amount
200 mg; temperature range 350–450 ◦C; feed gas (vol.%),
propane/O2/H2O/He = 12.5/20.4/15.9/51.2. Reaction products
were analyzed on line with a Varian 3800 gas chromatograph
The XPS binding energies are listed in Table 1. The binding
energies for niobium are characteristic of Nb5+ species [23,24] and
show little changes in the different series. The binding energies val-
ues of the Te 3d5/2 peak are typical of Te6+ species [25,26] with a
minor contribution of lower oxidation state (Te4+). A small amount
of reduced Te4+ species is detected in all the Te-containing sam-
ples, their concentration is higher in the case of inert-calcined
samples. For the molybdenum X-ray photoelectron spectra decon-
volution, the spin–orbit splitting between the Mo 3d5/2 and Mo
3d3/2 signals was fixed to 3.1 eV and the intensity ratio was set
to 0.67. Three different molybdenum species can be identified in
the XPS spectra, characteristic of Mo6+, Mo5+ and Mo4+ species
[27–30]. Reduced Mo4+ species are only detected for inert cal-
cination. The 3d3/2 and 3d5/2 shoulder peaks are at 234.7 and
231.3–231.7 eV, which can be assigned to the Mo5+ state, whose
concentration increases for inert-calcined series. V 2p BE region
typically exhibits two components; the one at higher binding
energies (516.9–517.0 eV) is assigned to V5+ sites [31]; a second
component, at lower binding energy (516.1–516.3) is associated
with V4+ species [32]; and a third component at lower BE values
(515.4 eV) corresponds to reduced V3+ species [24,25]. V4+ popula-
tion, in general, is higher for inert-calcined samples. A rather unique
sample is 12Mo5V4Nb1–inert, which possesses a significant popu-
lation of V4+ species and no V5+ sites. The presence of tellurium has
a clear effect on vanadium sites upon inert calcination, in which a
significant fraction remains as V5+, unlike the Te-free inert-calcined
series.
Fig. 1 shows the XRD patterns, which shape depends on the cal-
cination atmosphere. In the case of air-calcined samples (A and
B), several peaks in the 22–30◦ range, corresponding to Al–Mo–O,
Mo–Nb–O and Mo–V–O mixed phases, are evident at high cov-
erages. The Mo5O14-type structure [33,34] can be identified by
its strongest reflections at 22.1◦ and 24.9◦ that in this case can
overlap with the other signals, thus, the presence of such oxide
cannot be excluded. MoO2 and a rutile Mo–V–Nb–O phase are
detected at high loadings in inert-calcined samples. The rutile
phase exhibit high intensity peaks at 26.6, 36.8, 38.0, 41.6 and
broad peaks at 53.9, 62.0, 67.9, 72.0 and 80.3◦ [35–37]. The main
diffraction peaks of MoO2 [38] and the rutile Mo–V–O diffraction
patterns overlap at 26.6◦, 36.8◦ and 53.9◦. In addition, the peak at
68◦, assigned to the rutile Mo–V–O structure, overlaps with the
diffraction peaks corresponding to the alumina support. Thus, rutile
equipped with
detectors.
a flame ionization and thermal conductivity
The BET surface areas of catalysts were measured using a
Micromeritics ASAP 2000 analyzer by nitrogen adsorption at
liquid nitrogen temperature. The samples were previously out-
gassed under nitrogen flow at 140 ◦C. The XRD patterns were
recorded in the 2ꢀ = 4–90◦ range before and after calcination
treatments in a Seifert 3000P XRD diffractometer using CuK␣
radiation (ꢁ = 0.15418 nm) and a graphite monochromator (40 kV,
30 mA, and scanning rate of 2◦/min for Bragg’s angles). Raman
spectroscopy was performed with a Renishaw System 1000 spec-
trometer equipped with green laser (Spectra Physics, ꢁ = 514 nm,
power 19 mW, 1 mW on sample). For Raman spectra, the fresh cata-
lysts were dehydrated in situ in dry air stream (50 ml/min) at 200 ◦C.
The spectra were obtained under dehydrated conditions (ca. 200 ◦C)
in a hot stage (Linkam TS-1500). The Raman spectra of the cata-
lysts were also obtained under ambient hydrated conditions. In situ
Raman spectroscopy was performed at various temperatures from
50 to 600 ◦C with a ramp of 10 ◦C/min) on the supported MoVNb
oxide catalysts.
X-ray photoelectron spectroscopy (XPS) spectra were
recorded on a VG Escalab 200R spectrometer equipped with
a hemispherical electron analyzer and a MgK␣ (hꢂ = 1254.6 eV,
1 eV = 1.6302 × 10−19 J) X-ray source powered at 120 W. The sam-
ple powders were pressed into small copper supports and then
mounted on a support rod placed in the pre-treatment chamber.
Prior to the analysis, oxide samples were evacuated at 25 ◦C for 2 h
under atmospheric pressure. The X-ray chamber was evacuated
to 1 × 10−6 Pa. As a reference, we used the C 1 s signal of the
adventitious carbon (carbon of any surface adsorbed), which we
fixed at 284.6 eV. The latter procedure involved linear background
subtraction and a curve-fitting using a mixed Gaussian–Lorentzian
function. The quantitative analysis was based on the areas of the
XPS peaks, which was corrected on the atomic sensitivity factor of
the corresponding elements and the area of peaks was determined
after subtraction of background by Shirley method with an error
ꢃ2 minimum. The spectra of Mo 3d level were decomposed into
peak couples with parameters of spin–orbit separation ꢄEp (Mo
3d5/2–Mo 3d3/2) = 3.15 eV, ꢄEp (Te 3d5/2–Te 3d3/2) = 10.4 eV. The
deconvolution of Te 3d5/2 peaks showed that all of the peaks can

R. López-Medina et al. / Catalysis Today 158 (2010) 139–145
141
Mo–V oxide phase can only be identified by the diffraction peak
near 26.4◦, which is visible in samples with the highest cover-
age.
Upon tellurium doping, air-calcined samples (Fig. 1B) show
similar diffraction patterns the same phases than undoped ones
(Fig. 1A, dominated by Al2(MoO4)3), and new peaks at 22.1◦, 28.3◦,
36.2◦, 50.0◦ and 55.0◦ show the presence of hexagonal Te0.33MO3.33
with M = Mo, V and/or Nb. In the case inert-calcined samples
(Fig. 1D), the diffractograms are also dominated by the peaks of the
rutile and MoO2 diffractions, as in the case of samples without Te;
but the peaks of the rutile phase are more intense, and are develop
at lower coverages (Fig. 1D). A peak near 22.1◦ is also detected
in the samples with high coverage that was not detected in the
samples without Te, it may belong to the M1 phase (orthorhombic
Te2M20O57 structure) [39–41], which presents peaks at 6.6◦, 7.9◦,
9.0◦, 22.1◦, 27.3◦ and 45.3◦.
Fig. 2 shows representative Raman spectra of the catalysts.
Air-calcined samples exhibit a strong signal near 990–1030 cm−1
,
characteristic of the stretching modes of Mo O and/or V O bonds,
respectively. The broad band exhibits a maximum neat1000 cm−1
that is characteristic of the Mo O mode. It is difficult to assess
the relative population of these two Mo O and V O sites since
the Raman section of molybdenum oxide species is more intense
than that of vanadium oxide species. A shoulder near 1020 cm−1 is
indicative of the V O mode of surface vanadium species. This sug-
gests the presence of surface molybdenum and vanadium species
[42]. The Raman bands corresponding to ␣-MoO3 crystalline phase
(992, 817, 375, 334, 283, 234, 193 and 148 cm−1) are visible in
the high coverage air-calcined samples (Fig. 2A and B). The Raman
bands due to MoO2 appear at 227, 362, 494, 572 and 742 cm−1 [43]
and are visible in Fig. 2C (12Mo5V4Nb1–inert). The broad Raman
band near 820–840 cm−1 corresponds to a rutile-type structure.
The Raman band near 885 cm−1 is characteristic of Mo5O14-type
structure [44,45]. In the case of Te-containing catalysts, the band at
885–890 cm−1 can be also assigned to the stretching mode of bridg-
ing Mo–O–M (M = Mo, V) bonds of highly distorted Te2M20O57 on
alumina [46–48]. 12Mo5V4Te0.5–inert exhibits Raman bands near
820 and 470 cm−1 (Fig. 2D(c)) which are characteristic of M1 phase
[49].
The UV–vis DR spectra depend on the calcination atmosphere
(Fig. 3). A strong signal is detected between 250 and 400 nm for
air-calcined samples. Such band is assigned to ligand–metal charge
transfer (LMCT) transitions involving primarily the charge transfer
from oxygen ligands to Mo6+ or V5+ metal cations, in addition to
Nb5+ and Te4+[50,51]. This band extends up to 600 nm in the case
of Te-containing samples, this additional absorption is assigned to
partially reduced Mo cations [52–54]. In the case of inert-calcined
samples, a band between 300 and 400 nm suggests the presence
of V5+ species in a different environment. The broad absorption
between 500 and 800 nm is typically associated with V4+ cations
[55].
Fig. 4 shows the yields of main products during methanol
oxidation and, for comparative purposes, the yield to acrylic
acid obtained during propane partial oxidation has also been
included. Dimethylether is the main product at low coverages dur-
ing methanol oxidation, whereas formaldehyde forms at higher
coverages; evidencing a transition from essentially acidic to redox
properties. The methanol conversion on the air-calcined sam-
ples decreases with coverage, probably due to the presence of
Al–Mo–O and Mo–V–O oxidized phases, detected by XRD (Fig. 3).
For tellurium containing samples, methanol conversion decreases
at high coverages, indicating that dispersed oxide species are more
active for methanol activation than Te-containing samples; which
is not the case for propane oxidation, which activity increases
upon Te addition. Such trend is less intense at lower coverages
(Table 2).

142
R. López-Medina et al. / Catalysis Today 158 (2010) 139–145
Fig. 1. X-ray diffraction pattern of (A) (a) 4Mo5V4Nb1–air, (b) 8Mo5V4Nb1–air and (c) 12Mo5V4Nb1–air; (B) (a) 4Mo5V4 Nb0.5Te0.5–air, (b) 8Mo5V4 Nb0.5Te0.5–air and (c)
12Mo5V4 Nb0.5Te0.5–air; (C) (a) 4Mo5V4Nb1–inert, (b) 8Mo5V4Nb1–inert and (c) 12Mo5V4Nb1–inert and (D) (a) ␥-Al2O3, (b) MoVTeNbO–rutile, (c) 4Mo5V4Nb0.5Te0.5–inert,
(d) 8Mo5V4 Nb0.5Te0.5–inert and (e) 12Mo5V4 Nb0.5Te0.5–inert catalysts calcinated for 2 h at 600 ◦C.
4. Discussion
to formaldehyde than the inert-calcined catalysts, indicative that
some redox sites are present on the surface. These samples present
dispersed MOx phases on their surface, which are responsible for
such redox activity. Metal oxide phases are identified by Raman
spectroscopy by the strong signal near 1000 cm−1 (Fig. 2A and B).
In such phases vanadium and molybdenum are present in their
highest oxidation states (Table 1). This redox activity due to sur-
face MOx species, is able to yield formaldehyde, however, these
dispersed metal oxide species are not efficient to transform the
The methanol oxidation results (Fig. 4) indicate that the catalytic
behavior strongly depends on the total Mo + V + Nb + Te cover-
age. At low coverage (4 atoms/nm2) the main product is dimethyl
ether, indicative that the surface of catalysts is dominated by acid
sites, which must belong to the hydroxyl groups from the alumina
support, which are not totally titrated and to the presence of sur-
face niobium species. Air-calcined samples present a higher yield
Table 2
Catalytic activity in partial oxidation of methanol at T = 250 ◦C. These results are after 2 h of reaction.
Conversion
% MeOH
Selectivity %
Catalysts
Formaldehyde HCHO
Methyl formate
HCOOCH3
Dimethyl ether
CH3OCH3
Dimethoxy
methane
CO2
(CH3O)2CH2
4Mo5V4Nb1–air
8Mo5V4Nb1–air
12Mo5V4Nb1–air
4Mo5V4Nb1–inert
77.8
79.4
37.9
59.9
28.5
78.5
58.1
60.4
42.1
59.2
78.2
39.1
44.9
44.4
69.7
26.3
56.0
77.9
46.6
86.0
73.1
24.5
52.9
66.7
7.9
6.7
2.3
16.5
18.4
6.0
15.5
5.3
45.0
46.3
26.1
56.2
25.0
15.7
37.2
8.0
18.0
64.0
29.8
23.8
0.1
0.3
0.1
0.1
0.2
0.1
0.1
0.2
0.6
0.1
0.3
0.2
2.1
2.3
1.8
0.9
0.4
0.3
0.6
0.5
0.5
1.0
0.6
0.2
8Mo5V4Nb1–inert
12Mo5V4Nb1–inert
4Mo5V4Nb0.5Te0.5–air
8Mo5V4Nb0.5Te0.5–air
12Mo5V4Nb0.5Te0.5–air
4Mo5V4Nb0.5Te0.5–inert
8Mo5V4Nb0.5Te0.5–inert
12Mo5V4Nb0.5Te0.5–inert
7.8
10.4
16.4
9.1

R. López-Medina et al. / Catalysis Today 158 (2010) 139–145
143
Fig. 2. Dehydrated Raman spectra of (A) (a) 4Mo5V4Nb1–air, (b) 8Mo5V4Nb1–air and (c) 12Mo5V4Nb1–air; (B) (a) 4Mo5V4Nbo.5Teo.5–air, (b) 8Mo5V4Nb0.5Te0.5–air, (c)
12Mo5V4Nb0.5Te0.5–air; (C) (a) 4Mo5V4Nb1–inert, (b) 8Mo5V4Nb1–inert, (c) 12Mo5V4Nb1–inert and (D) (a) 4Mo5V4Nb0.5Te0.5–inert, (b) 8Mo5V4Nbo.5Teo.5–inert and (c)
12Mo5V4Nb0.5Te0.5–inert at 200 ◦C.
Fig. 3. Diffuse reflectance UV–vis DRS spectra for supported MoVNb(Te)O catalysts calcined in air (A and B) and nitrogen (C and D) for 2 h at 600 ◦C.

144
R. López-Medina et al. / Catalysis Today 158 (2010) 139–145
Fig. 4. Catalytic results (yields of formaldehyde, dimethylformate and dimethylether) obtained in selective oxidation of methanol, reaction conditions: total flow 40 ml/min,
molar ratio He/O2/MeOH = 88/8/4 mol%, 100 mg of catalyst, T = 250 ◦C; and yield to acrylic acid obtained during the propane partial oxidation, reaction conditions: total flow
40 ml/min feed gas (vol.%) C3/O2/H2O/He = 12.5/20.4/15.9/51.2, 200 mg, GHSV = 4800 h−1, T = 400 ◦C.
propane into acrylic acid. It should be noted that the addition of
tellurium to these catalysts increases yield to acrylic acid.
(DME). At this coverage, crystalline aggregates form, evidenced by
XRD (Fig. 1). Inert-calcined Te-free samples are essentially redox
(formaldehyde) with few acidic (DME) sites. The presence of tel-
lurium brings a balance between redox and acidic properties.
Such balance affords higher yield to acrylic acid; however, activ-
ity depends on redox acid balance but also on the catalyst phases.
12Mo5V4Nb1–air hardly yields any acrylic acid, despite exhibit-
ing similar redox/acid properties balance. This catalyst possesses
essentially fully oxidized molybdenum and vanadium species and
mixed phases with high-oxidation states. The inert calcination at
high coverages forms of MoO2 and rutile-like phase, visible by XRD
and Raman spectroscopy. These phases are able to give redox activ-
ity to the catalysts, but such redox sites are not selective to acrylic
acid formation (Fig. 4). The surface of air-calcined samples is dom-
inated by Al–Mo–O and Mo–V–O aggregates, identified by XRD
(Fig. 1), with low BET surface area values (Table 1). These phases
have less redox activity than the dispersed MOx species present in
the sample with lower coverage (8Mo5V4Nb1–air); thus, the activ-
ity during methanol oxidation is lower for the sample with high
coverage (12Mo5V4Nb1–air). The incorporation of tellurium in the
At intermediate loadings (8 atoms/nm2), the surface is close to
monolayer coverage, which is near 9 atoms/nm2 in the case of VOx
species on an alumina support [56]. Thus, most acid sites from alu-
mina must be titrated. The detected acid sites must originate from
supported oxides; their presence clearly decreases upon inert cal-
cinations or upon tellurium addition. Redox sites (formaldehyde)
decrease upon inert calcinations for Te-free samples; this is consis-
tent with the much lower binding energies values for molybdenum
and vanadium species. The presence of tellurium affects this trend,
and the catalyst exhibits much higher redox properties. Inert cal-
cination promotes further entanglement of tellurium ions with
molybdenum and niobium ions, leading to the formation of mixed
phases. This stabilizes intermediate oxidation states that afford a
balance between redox (formaldehyde) and acidic (dimethylether)
properties. This appears to favor propane oxidation to acrylic acid.
At high coverages (12 atoms/nm2), formaldehyde is the main
product for all samples, indicative that the major part of active
sites on the surface of catalysts are redox, with some acidic sites

R. López-Medina et al. / Catalysis Today 158 (2010) 139–145
145
case of samples with high coverages decreases the activity during
methanol oxidation, since the amount of MOx species on the sur-
face decreases. However, the yield to acrylic acid during propane
partial oxidation increases. Thus, tellurium is able to interact with
molybdenum to form crystalline aggregates visible by XRD (Fig. 1),
selective to the acrylic acid formation. At high coverage all the
hydroxyl groups from the alumina support must be covered, thus,
the acidic activity detected in this case must be due to surface acidic
Mo6+ and/or Nb5+ sites.
[12] M.O. Guerrero Pérez, J.N. Al-Saeedi, V.V. Guliants, M.A. Ban˜ares, Appl. Catal. A
260 (2004) 93–99.
[13] M.O. Guerrero Pérez, L.J. Alemany, Appl. Catal. A 341 (2008) 119–126.
[14] V.V. Guliants, R. Bhandari, J.N. Al-Saeedi, V.K. Vasudevan, R.S. Soman, M.O.
Guerrero Pérez, M.A. Ban˜ares, Appl. Catal. A 274 (2004) 123–132.
[15] M.O. Guerrero Pérez, M.C. Herrera, I. Malpartida, M.A. Larrubia, L.J. Alemany,
Catal. Today 133–135 (2008) 919–924.
[16] P. Korovchenko, N.R. Shiju, A.K. Dozier, U.M. Graham, M.O. Guerrero-Pérez, V.V.
Guliants, Top. Catal. 50 (2008) 43–51.
[17] M.O. Guerrero Pérez, M.A. Ban˜ares, Catal. Today 142 (2009) 245–251.
[18] M.O. Guerrero Pérez, M.A. Ban˜ares, Appl. Catal. A 274 (2004) 123–132.
[19] P. Korovchenko, N.R. Shiju, A.K. Dozier, U.M. Graham, M.O. Guerrero-Pérez, V.V.
Gulian, Top. Catal. 50 (2008) 43–51.
[20] B. Kilos, M. Aouine, I. Nowak, M. Ziolek, J.C. Volta, J. Catal. 224 (2004) 314–325.
[21] I. Nowak, M. Ziolek, Chem. Rev. 99 (1999) 3603–3624.
[22] M. Ziolek, Catal. Today 78 (2003) 47–64.
[23] V.V. Atuchin, I.E. Kalabin, V.G. Kesler, N.V. Pervukhina, J. Electron Spectrosc.
Relat. Phenom. 142 (2005) 129–134.
[24] M. Ziolek, I. Nowak, Catal. Today 78 (2003) 543–553.
[25] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photo-
electron Spectroscopy, Perkin-Elmer, USA, 1992.
[26] R. Häggblad, J.B. Wagner, B. Deniau, J.M.M. Millet, J. Holmberg, R.K. Grasselli, S.
Hansen, A. Andersson, Top. Catal. 50 (2008) 52–65.
5. Conclusions
The combined information on reactive properties (methanol
probe) and structural features help understanding structure–
activity relationships. These results indicate that methanol probe
reaction is a very useful tool to characterize the active centers
present on the surface of catalysts, which complement structural
characterization data, such as XRD, XPS, UV–vis and Raman spec-
troscopy.
[27] A.M. Gaffney, S. Chaturvedi, M.B. Clark Jr., S. Han, D. Le, S.A. Rykov, J.G. Chen, J.
Catal. 229 (2005) 12–23.
For the catalysts studied here, combined probe reaction and
structural characterization confirm that propane conversion to
acrylic acid depends on both structural and reactive features. Effi-
cient catalysts possess both, acid and redox sites and the right
kind of structure, such sites are efficient when the catalyst contains
mixed oxide phases in which vanadium and molybdenum possess
intermediate oxidation states. But when the catalyst contains dis-
persed oxides or mixed oxides in high-oxidation state, these not
work for propane to acrylic acid.
[28] J. Holmberg, S. Hansen, R.K. Grasselli, A. Andersson, Top. Catal. 38 (2006) 17–29.
[29] P. DeSanto Jr., D.J. Buttrey, R.K. Grasselli, W.D. Pyrz, C.G. Lugmair, A.F. Volpe Jr.,
T. Vogt, B.H. Toby, Top. Catal. 38 (2006) 31–40.
[30] A.M. Venezia, Catal. Today 77 (2003) 359–370.
[31] F. Ivars, B. Solsona, E. Rodríguez Castellón, J.M. López Nieto, J. Catal. 262 (2009)
35–43.
[32] R. Häggblad, J.B. Wagner, S. Hansen, A. Andersson, J. Catal. 258 (2008) 345–
355.
[33] G.Ya. Popova, T.V. Andrushkevich, G.I. Aleshina, L.M. Plyasova, M.I. Khramov,
Appl. Catal. A 328 (2007) 195–200.
[34] M. Roussel, M. Bouchard, E. Bordes Richard, K. Karim, S. Al-Sayari, Catal. Today
99 (2005) 77–87.
[35] J. Holmberg, R.K. Grasselli, A. Andersson, Top. Catal. 23 (2003) 55–63.
[36] L. Yuan, V.V. Guliants, J. Porous Mater. 16 (2009) 613–622.
[37] N. Haddad, E. Bordes Richard, A. Barama, Catal. Today 142 (2009) 215–219.
[38] M. Roussel, S. Barama, A. Löfberg, S. Al-Sayari, K. Karim, E. Bordes Richard, Catal.
Today 141 (2009) 288–293.
[39] N.R. Shiju, V.V. Guliants, Catal. Commun. 9 (2008) 2253–2256.
[40] N.R. Shiju, X. Liang, A.W. Weimer, C. Liang, S. Dai, V.V. Guliants, J. Am. Chem.
Soc. 130 (2008) 5850–5851.
Acknowledgements
COST action D36, WG No. D36/0006/06, the Polish Ministry of
Science (Grant No. 118/COS/2007/03) and the Spanish Ministry
of Science and Innovation (Grant No. CTQ2008/02461/PPQ) are
acknowledged for the financial support. R.L.M. thanks MAEC-AECID
(Spain) for his pre-doctoral fellowship and COST D36 for the finan-
cial support (STM) during his stay. Authors thank Elizabeth Rojas
García (ICP-CSIC) for her help with catalytic tests and to Olaf Torno
(SASOL Germany GmbH) for providing alumina support.
[41] P. Botella, A. Dejoz, M.C. Abello, M.I. Vázquez, L. Arrúa, J.M. López Nieto, Catal.
Today 142 (2008) 272–277.
[42] M.A. Ban˜ares, I.E. Wachs, J. Raman Spectrosc. 33 (2002) 359–380.
[43] P.A. Spevack, N.S. Mclntyre, J. Phys. Chem. 96 (1992) 9029.
[44] G. Mestl, Top. Catal. 38 (2006) 69–82.
[45] G. Mestl, J. Raman Spectrosc. 33 (2002) 333–347.
[46] S.A. Holmes, J. Al-Saeedi, V.V. Guliants, P. Boolchand, D. Georgiev, U. Hackler, E.
Sobkow, Catal. Today 67 (2001) 403–409.
References
[47] L. Yuan, V.V. Guliants, M.A. Ban˜ares, S.J. Khatib, Top. Catal. 49 (2008) 268–
280.
[1] V.V. Guliants, H.H. Brongersma, A. Knoester, A.M. Gaffney, S. Han, Top. Catal. 38
(2006) 41–50.
[2] M.O. Guerrero-Pérez, T. Kim, M.A. Ban˜ares, I.E. Wachs, J. Phys. Chem. C 112
[48] X. Yang, W. Zhang, R. Feng, W. Ji, C.T. Au, Catal. Lett. 124 (2008) 288–296.
[49] B. Solsona, M.I. Vázquez, F. Ivars, A. Dejoz, P. Concepción, J.M. López Nieto, J.
Catal. 252 (2007) 271–280.
(2008) 16858–16863.
[50] M.O. Guerrero Pérez, M.C. Herrera, I. Malpartida, M.A. Larrubia, L.J. Alemany,
M.A. Ban˜ares, Catal. Today 126 (2007) 177–183.
[51] J.M. Oliver, J.M. López Nieto, P. Botella, Catal. Today 96 (2004) 241–249.
[52] P. Botella, E. García González, B. Solsona, E. Rodríguez Castellón, J.M. González
Calbet, J.M. López Nieto, J. Catal. 265 (2009) 43–53.
[3] J.M. Tatibouët, H. Lauron Pernot, J. Mol. Catal. A 171 (2001) 205–216.
[4] L.E. Briand, W.E. Farneth, I.E. Wachs, Catal. Today 96 (2004) 211.
[5] X. Wang, I.E. Wachs, Catal. Today 96 (2004) 211.
[6] I.E. Wachs, J.M. Jehng, W. Ueda, J. Phys. Chem. B 109 (2005) 2275–2284.
[7] S.K. Korhonen, M.A. Ban˜ares, J.L.G. Fierro, A.O.I. Krause, Catal. Today 126 (2007)
235–247.
[53] P. Botella, P. Concepción, J.M. López Nieto, Y. Moreno, Catal. Today 99 (2005)
51–57.
[8] I.E. Wachs, Y. Chen, J.M. Jehng, L.E. Briand, T. Tanaka, Catal. Today 78 (2003)
13–24.
[54] B. Solsona, J.M. López Nieto, J.M. Oliver, J.P. Gumbau, Catal. Today 91/92 (2004)
247–250.
[9] H. Liu, E. Iglesia, J. Phys. Chem. B 109 (2005) 2155–2163.
[10] J.M. Tatibouët, Appl. Catal. A 148 (1997) 213–252.
[11] S.A.R.K. Deshmukh, M. van Sint Annaland, J.A.M. Kuipers, Appl. Catal. A 289
(2005) 240–255.
[55] P. Botella, A. Dejoz, J.M. López Nieto, P. Concepción, M.I. Vázquez, Appl. Catal.
A 298 (2006) 16–23.
[56] I.E. Wachs, L.E. Briand, J.M. Jehng, L. Burcham, X. Gao, Catal. Today 57 (2000)
323–330.