Influence of Phase Composition of Bulk Tungsten Vanadium Oxides on the Aerobic Transformation of Methanol and Glycerol

Article abstract of DOI:10.1002/ejic.201800059

A series of W–V–O catalysts with different m-WO₃ and h-WO₃ phase contents were hydrothermally synthesized by employing different tungsten, vanadium, and ammonium precursors and characterized by powder XRD, N₂ adsorption, SEM, X-ray energy-dispersive spectroscopy, thermogravimetric analysis, Raman and FTIR spectroscopy, NH₃ temperature programmed desorption, H₂ temperature-programmed reduction, and XPS. Finally, the acid/redox properties were analyzed by using aerobic transformation of methanol as a characterization reaction. A correlation between phase composition as well as acid and redox properties was observed, which were correlated to the catalytic performance of the title materials in a one-pot oxydehydration reaction of glycerol. The hexagonal tungsten bronze (h-WO₃) phase shows a significantly higher concentration of acid sites than monoclinic m-WO₃, so that the acid properties of W–V–O oxides are directly related to the presence of h-WO₃ crystals. The presence of a higher concentration of acid sites in V-containing h-WO₃ crystals is a key factor to achieve high selectivity to both acrolein and acrylic acid during one-pot glycerol oxydehydration. Also, V sites in h-WO₃ show higher selectivity in the consecutive reaction (partial oxidation of acrolein to acrylic acid), while V sites in the m-WO₃ phase fundamentally lead to the formation of carbon oxides.

Full text of DOI:10.1002/ejic.201800059

A Journal of
Accepted Article
Title: Influence of phase composition of bulk tungsten-vanadium oxides
on the one-pot synthesis of acrylic acid from glycerol
Authors: Daniel Delgado, Alessandro Chieregato, Maria Dolores
Soriano, Elena Rodríguez-Aguado, Lidia Ruiz-Rodríguez,
Enrique Rodríguez Castellón, and José M. López Nieto
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201800059
Link to VoR: http://dx.doi.org/10.1002/ejic.201800059

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
Influence of phase composition of bulk tungsten-vanadium
oxides on the aerobic transformation of methanol and glycerol
Daniel Delgado,[a] Alessandro Chieregato,[a,b] M. Dolores Soriano,[a] Elena Rodríguez-Aguado,[c] Lidia
Ruiz-Rodríguez,[a] Enrique Rodríguez-Castellón,[c] and José M. López Nieto*[a]
Abstract: A series of W-V-O catalysts which present different m-
WO3 and h-WO3 phase contents were hydrothermally synthesized
employing different tungsten, vanadium and ammonium precursors.
Characterization of these materials was carried out by several
physicochemical techniques such as powder X-Ray diffraction, N2-
of linear and cyclic alkenes, nitroarenes or unsaturated
organosulfur compounds.[15] Moreover, they have been widely
studied as acid catalysts for different processes,[16, 17] but also as
acid-redox multifunctional materials, due to their ability to
accommodate different kinds of metals and non-metals within
adsorption, SEM, XEDS, TG, Raman, FT-IR spectroscopy, TPD-NH3, the structure. Recently, mixed metal oxides with tetragonal
H2-TPR and XPS. Finally, acid-redox properties were analyzed using
methanol aerobic transformation as a characterization reaction. A
correlation between phase composition as well as acid- and redox-
properties was observed, which were correlated to the catalytic
performance of the titled materials in the one-pot glycerol
oxydehydration reaction. The hexagonal tungsten bronze, h-WO3,
phase shows a concentration of acid sites significantly higher than
that of the monoclinic m-WO3, making acid properties of W-V-O
oxides directly related to the presence of h-WO3 crystals. The
presence of a higher concentration of acid sites in V-containing h-
WO3 crystals is a key factor to achieve high selectivity to both
acrolein and acrylic acid during the one-pot glycerol oxydehydration.
Also, V-sites in h-WO3 present higher selectivity in the consecutive
reaction (acrolein partial oxidation to acrylic acid), while V-sites in m-
WO3 phase fundamentally lead to the formation of carbon oxides.
tungsten bronze (TTB) structure have shown high activity and
selectivity for the gas phase partial oxidation of olefins,[18]
whereas materials presenting hexagonal tungsten bronze (h-
WO3) structures have been proposed as active and selective
catalysts in the aerobic transformation of alcohols, such as the
one-pot oxydehydration of glycerol to acrylic acid.[19, 20] In this
way, a parallelism between the selectivity to acrylic acid from
glycerol and the catalytic behavior of the corresponding catalysts
during the aerobic transformation of methanol was proposed.[21]
From the synthetic point of view, the control of key parameters
of the synthesis, such as pH, organic additives or the precursors
used on the formulation, made it possible to obtain different
phase compositions, particles sizes and crystal shapes. It was
demonstrated that all these features can have an important
influence in their functionality, specifically in their catalytic
performance.[22-25]
Herein, we report on the preparation and characterization of a
series of W-V-O mixed oxides and their catalytic behavior in
both the aerobic transformation of methanol and the
oxydehydration of glycerol. Depending on the W-, V- or
Introduction
Tungsten oxides have received special attention in the last
decades since they present potential uses in a wide variety of
fields such as electrochromic displays, gas sensing,
superconductivity or catalysis.[1-4] Particularly, they have
emerged in the last years as interesting materials for several
catalytic applications. Indeed, they are well-known multi-
functional materials for fuel cells, being employed as
electrocatalysts and supports.[5-8] Oxygen deficient tungsten
oxides-based materials, known as tungsten bronzes, are known
as catalysts for the electrochemical oxidation of C2-C3 alcohols
ammonium-precursors,
different
morphologies,
thermal
stabilities and ratios of monoclinic (m-WO3) to hexagonal (h-
WO3) phases were obtained. The analysis of the selectivity
profiles in the aerobic transformation of methanol sheds light on
the acid-redox properties of the materials, which appear to be
correlated with the m-WO3/h-WO3 phase ratio. In order to
corroborate these findings, catalysts were finally tested in the
glycerol oxydehydration reaction, where both acid and redox
sites are necessary to achieve a relatively high yield to acrylic
acid. It was found that the presence of h-WO3 is pivotal to tune
the acid and redox properties for the selective transformation of
glycerol to acrylic acid.
[9-11]
and hydrogen in acid media.[12] Tungsten oxide bronzes
have also been used in electrocatalysts for oxygen reduction, [13]
photocatalysts [14] and as efficient catalysts in the hydrogenation
Results and Discussion
[a]
D. Delgado, Dr. A. Chieregato, Dr. M. D. Soriano, L. Ruiz
Rodríguez, Prof. J. M. Lopez Nieto
Instituto de Tecnología Química
Universitat Politècnica de València-Consejo Superior de
Investigaciones Científicas
Characterization of W-V-O metal oxides
The main physical and chemical features of W-V-O materials are
summarized in Table 1. V/(W+V) ratios after synthesis are in the
range between 0.12-0.18 for all the studied materials. This
means that almost half of the vanadium added into the synthesis
gel (W/V ratio in the synthesis of 1/0.3) does not incorporate into
the oxides framework after hydrothermal treatment, regardless
Avenida de los Naranjos s/n, 46022 Valencia, Spain
E-mail: jmlopez@itq.upv.es
Dr. A. Chieregato (current affiliation)
R&D Department of Catalysis for Refining and Base Chemicals
Total Research & Technology
Zone Industrielle Feluy C, B-7181 Seneffe, Belgium
E. Rodríguez-Aguado, Prof. E. Rodríguez Castellón
Departamento de Química Inorgánica
[b]
[c]
Universidad de Málaga, 29071 Málaga, Spain
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
Table 1. Characteristics of W-V-O mixed oxides catalysts.
[f]
Sample
Precursors [a]
V/(W+V) atomic ratio
h-WO3 phase content (%)[d]
SBET
Weight loss
at 350 ºC[e]
TPD-NH3
W
V
NH4+-salt
-
Bulk[b]
0.17
0.17
0.16
0.12
0.12
0.17
0.18
Surface[c]
0.05
As-prepared
Heat-treated
(m2 g-1)
20.5
16.1
32.8
13.5
7.2
(%)
(μmol/g)
n.d.
127
241
n.d.
50
AMT-1
AMT-2
TA-1
AMT
AMT
TA
VS
100
n.d.[g]
100
42
100
n.d.[g]
100
31
5.42
5.37
5.65
2.58
1.77
2.91
4.82
VOAC
VS
-
0.16
NH3(aq)/NH4Cl
NH4SO4
NH4Cl
0.07
TA-2
TA
VS
n.d.
TA-3
TA
VS
0.08
46
26
TA-4
TA
VS
NH4CH3CO2
NH4CH3CO2
0.15
41
0
7.9
86
TA-5
TA
VOAC
0.19
60
54
26.6
88
[a] AMT= (NH4)6H2W12O40; TA= H2WO4; VS= VOSO4; VOAC= VO(acac)2. [b] V/(W+V) ratio obtained by X-Ray energy-dispersive spectroscopy (XEDS). [c]
Calculated by X-Ray Photoelectron Spectroscopy (XPS). [d] Calculated from equation: h-WO3 (%)= 100 x I100(h-WO3)/[I022(m-WO3) + I100(h-WO3)]. [e] Obtained by
thermogravimetric analysis. [f] Temperature-programed desorption of ammonia. [g] A pseudocrystalline phase is only observed.
the type of precursor used for the preparation.
Moreover, catalysts presenting higher percentages of bronze
phase also present higher surface areas (Table 1).
Figure 1 shows XRD patterns of both as-prepared (Figure 1A)
and heat-treated materials (Figure 1B). Focusing on as-prepared
catalysts, two pure bronze phases where obtained when
ammonium metatungstate or an ammonia/ammonium chloride
mixture were used as ammonium source (Figure 1A, patterns a
to c). Characteristic diffraction patterns of these phases
correspond to those of an hexagonal tungsten bronze type
B
A
g
g
f
f
structure (h-WO3) (JCPDS: 33-1387) (Figure 1A, patterns a and
26]
c),[19,
and a pseudo-crystalline phase related to the ReO3-
phase, with long-range order just along c-axis (Figure 1A,
pattern b).[20] On the other hand, when using tungstic acid and
other ammonium salts such as ammonium sulfate or ammonium
acetate as precursors, mixtures of h-WO3 and perovskite related
ReO3-type monoclinic WO3 phase (m-WO3) were obtained
(JCPDS: 43-1035) (Figure 1A, patterns d-g). An unknown
crystalline phase was also formed when using ammonium
acetate, tungstic acid and vanadyl acetylacetonate on the
synthesis (Figure 1A, pattern g), which decomposes after heat-
treatment (Figure 1B, pattern g). In this regard, the presence of
meta-tungstate in the synthesis gel, or even the possible in-situ
formation of isopolytungstates during the hydrothermal treatment
could have an important influence in the formation of pure
bronze-type compounds.[18, 27, 28] XRD patterns of heat-treated
materials are presented in Figure 1B. It can be seen that as-
prepared pure h-WO3 phases remain stable after the thermal
treatment (Figure 1B, patterns a and c). However, in the
remaining cases bronze-type phases partially decompose into
m-WO3 (Figure 1B, patterns b, d-e). Relative proportions of h-
WO3 phase in the samples were estimated from the intensities of
(100) and (022) Bragg peaks of h-WO3 and m-WO3 phases,
respectively (Table 1). As-prepared materials present h-WO3
phase percentages in the range 41-100 %, while after heat-
treatment percentages in the range 0-100 % were obtained, due
to bronze phase partial or total decomposition (Table 1).
e
e
d
d
c
c
b
x2
x2
b
a
a
10
20
30
40
50
60
10
20
30
40
50
60
2θ
2θ
Figure 1. XRD patterns of as-prepared (A) and heat-treated (B) W-V-O
oxides: a) AMT-1; b) AMT-2; c) TA-1; d) TA-2; e) TA-3; f) TA-4; g) TA-5. Red:
h-WO3; Blue: m-WO3.
Figure 2 shows SEM micrographs of as-prepared materials.
Pure h-WO3 and pseudocrystalline phases present mainly 1D
morphologies, such as rod and needle-shaped structures
(Figure 2, a-c).
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
Figure 2. SEM micrographs of as-prepared oxides: a) AMT-1; b) AMT-2; c) TA-1; d) TA-2; e) TA-3; f) TA-4; g) TA-5.
channels, favoring phase transitions to m-WO3 type structures.[29,
30] Moreover, it can be noted that after heat treatment block-type
particles undergo decomposition into platelet-shaped ones,
which has been identified with m-WO3 type structure (Figure 3a).
However, 1D morphologies retain their shape after heat-
treatment (Figure 3b). According to XRD patterns, and other
results reported previously in literature, these morphologies can
be attributed to tungsten bronze-type structures.[31-33] Therefore,
heat-treated catalyst, which present a mixture of both tungsten-
bronze and m-WO3 phases, show both 1D structures and
platelet-shaped particles (Figure 3c). Nevertheless, we cannot
exclude the influence of the effective substitution of V for W
within h-WO3 framework, which has also been reported to
increase the thermal stability of the hexagonal tungsten bronze
structure.[21]
Raman spectra of heat-treated samples are depicted in Figure 4.
They show typical Raman profiles for tungsten based oxides at
ca. 709, 808 and 969 cm-1, that can be assigned to W-O-V, W-
O-W and M=O (M: W, V) bond vibrations respectively.[34-36] It is
noteworthy to mention that no bands over 1000 cm-1 are found
on the spectra, which means that no V2O5 was formed after
calcination.[37] Pure monoclinic m-WO3 phase material presents
Raman bands centered at 709, 808 and 983 cm-1 (Figure 4,
spectrum f), while a slight shift to lower frequencies occurs on
the pure hexagonal bronze phase, h-WO3, with corresponding
Raman signals at 683, 800 and 969 cm-1 respectively (Figure 4,
spectrum a). Furthermore, the presence of both h-WO3 and m-
Figure 3. SEM and TEM micrographs of heat-treated materials. a) TA-4; b)
AMT-1; c) TA
WO3 phases gives rise to
a Raman spectrum showing
characteristic bands from both of them (Figure 4, spectra d-e
and g).
The as-prepared materials have been also studied by
Differential Thermogravimetric (DTG) analysis. DTG profiles
show two main weight losses in the temperature ranges already
reported for other tungsten oxide-based materials (Fig. S1,
On the other hand, samples consisting of h-WO3/m-WO3
mixtures present both block-shaped and rod/ needle-shaped
structures (Figure 2, d-g). This fact could have an important
influence in the stability of the tungsten bronze phases since
block-shaped morphologies exposing higher proportion of (001)
faces could facilitate the elimination of NH4+ cations through the
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
supporting information): i) below 200 ºC; and ii) in the range 250-
450 ºC, which have been assigned to the elimination of
physisorbed water and ammonium cations from the channels
39]
respectively.[38,
These weight losses are notably higher in
samples with high proportion of h-WO3 or pseudocrystalline
phase (Table 1), which also present a weight gain around 500
ºC, fact that could be associated to the oxidation of V4+ and/or
W5+ centers to V5+ and W6+, respectively (Figure S1, patterns a-
c). In this way, an additional weight loss is also observed at 400
ºC in sample TA-5, that could be related to the previously
mentioned unknown phase found by XRD (Fig. S1, pattern g).
g
808
983
709
f
e
Figure 5. FTIR spectra of as-prepared (A) and heat-treated (B) catalysts: a)
AMT-1; b) AMT-2; c) TA-1; d) TA-2; e) TA-3; f) TA-4; g) TA-5.
d
c
Acid features of heat-treated materials were studied by means of
temperature programmed desorption of ammonia (Figure 6). It
can be noted that the highest concentration of acid sites is
obtained for high contents of bronze-type phases, either h-WO3
or pseudocrystalline phase (Table 1).
b
800
683
969
a
400
600
800
1000
1200
Wavenumber (cm-1
)
Figure 4. Raman spectra of heat-treated W-V-O catalysts: a) AMT-1; b) AMT-
2; c) TA-1; d) TA-2; e) TA-3; f) TA-4; g) TA-5.
e
d
FTIR spectra of as-prepared and heat-treated materials were
collected in order to study the activation process of the catalysts
(Figure 5). As-prepared materials present three main bands in
the range 3700-3000 cm-1 (Figure 5A). The signal centered at
3430 cm-1 has been attributed to O-H groups stretching
vibrations, while remaining signals appearing at 3215 and 3144
cm-1 are associated to N-H stretching vibrations of the
corresponding ammonium cations located inside the channels.[38,
c
b
a
100
200
300
400
500
Temperature (ºC)
40, 41] However, heat-treated samples show no bands in the NH4
+
region due to the elimination of these ions from the structure
(Figure 5B). The substitution of H+ for ammonium ions within the
channels has been reported to be the origin of the acid centers
in tungsten bronze-type structure-based catalysts.[42]
Figure 6. TPD-NH3 profiles of selected W-V-O materials: a) TA-4; b) TA-3; c)
TA-5: d) AMT-2; e) TA-1.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
Thus, these results suggest that the presence of relatively wider
channels could favor the formation of Brönsted acid sites, by
elimination of NH4+ ions during the activation at high temperature
as reported elsewhere,[42] meanwhile m-WO3 structure would
present a lower concentration of surface acid sites.
Acid/redox features of W-V-O oxides were evaluated taking into
account the selectivity profiles obtained for each material at low
methanol conversion (c.a. 10%) (Figure 8). Hence, selectivity to
DME progressively increases with the amount of h-WO3 phase
in the catalysts (Figure 8), while the opposite trend is observed
in the selectivity to partial oxidation products, i.e. POPs (Figure
8). On the one hand, as m-WO3 concentration increases in the
catalysts, the acid function is progressively lost, for the benefit of
the redox one, achieving the highest selectivity to POPs in the
absence of h-WO3 phase.
Despite this, the influence of other variables, such as surface
area or surface vanadium concentration cannot be ruled out.
Thus, XPS experiments show very different vanadium surface
contents (much lower than that of bulk composition in some
cases) depending on the synthesis procedure (Table 1). The V
2p3/2 core level spectra of all the studied W-V-O mixed oxide
catalysis shows a single peak at 517.1-517.6 eV assigned to the
presence of V5+, except in the case of sample AMT-1 were two
contributions at 517.1 eV (86%) and 515.7 eV (14%) assigned to
V5+ and V4+ are observed, respectively (Figure S2). The W 4f
core level spectra for the studied catalysts are very similar, with
a W 4f7/2 peak centered at 35.9-36.3 eV, assigned to W6+ (Figure
S3).[43]
Interestingly, H2-TPR profiles of phase-pure W-V-O materials, i.e.
presenting monophasic h-WO3-type (AMT-1 sample) and m-
WO3-type (TA-4 sample) structures, display two main reduction
peaks (Figure S4). The peak appearing at a lower temperature
(528 and 597 ºC for h-WO3 and m-WO3-type phase,
respectively) can be attributed to the reduction of V5+ species,
whereas H2-uptakes at higher temperatures (635 and 662 ºC for
h-WO3 and m-WO3-type phases respectively) can be ascribed to
the reduction of W6+ cations. [19] In this sense, the reducibility V5+
and W6+ species within the hexagonal tungsten bronze
framework is higher with respect to those in the m-WO3-type
phase. This fact can have important implications when we deal
with oxidation reactions, as it will be discussed later.
100
POPs
80
60
40
0
20
40
60
80
100
MeOH Conversion (%)
60
40
20
0
DME
0
20
40
60
80
100
MeOH Conversion (%)
Catalytic results in the aerobic transformation of methanol
Figure 7. Selectivity to dimethyl ether (DME) and selectivity to partial oxidation
products (POPs) of W-V-O catalysts: AMT-1 (); AMT-2 (x); TA-1 (); TA-2
(▲); TA-3 (); TA-4 (); TA-5 (). Experimental conditions in text.
Methanol transformation has been widely used as a test reaction,
either for determining acid and/or redox features, or for the
elucidation of surface structure-reactivity relationships of bulk
metal oxide catalysts.[44-46] Indeed, the selective transformation
of methanol in aerobic conditions has been reported to be an
100
80
60
40
20
0
efficient method to estimate acid/redox multifunctional properties
47]
in tungsten bronze-based catalysts.[21,
When a molecule of
methanol reacts with Brönsted acid sites, it essentially leads to
the formation of dimethyl ether (DME). On the other hand, when
the reaction takes place on redox sites (e.g., V4+/5+ sites) it leads
to the formation of partial oxidation products (mainly
formaldehyde, but also other oxygenates products such as
methyl formate or dimethoxymethane, as minorities) and carbon
oxides (COx).[45]
Figure 7 shows the variation of the selectivity to partial oxidation
products (i.e. mainly formaldehyde, and small amount of methyl
formate and dimethoxymethane, grouped as “POPs") and
dimethyl ether (DME) with methanol conversion during the
aerobic transformation of methanol over W-V-O catalysts.
According to these results, there are important differences
depending on both the physico-chemical and structural
characteristics of catalysts.
0
20
40
60
80
100
h-WO3 phase content (%)
Figure 8. Variation of the selectivity to DME (□,Δ) and to partial oxidation
products, POPs (●,▲) vs. the amount of h-WO3 phase in catalysts during the
aerobic transformation of methanol on W-V-O oxides at a methanol conversion
of 10 %. The amount of h-WO3 phase as in Table 1. The percentage of bronze
phase of AMT-2 catalyst (Δ, ▲) has been estimated by considering the
selectivity to DME and POPs during the catalytic experiment.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
On the other hand, the presence of h-WO3 phase leads to
multifunctional acid/redox catalysts, in which its acid character
increases with the concentration of hexagonal tungsten bronze
phase, achieving a maximum DME selectivity of c.a. 50 %. Thus,
a clear relationship exists between the amount of h-WO3 phase
in W-V-O catalysts and the acid/redox properties of the samples,
as shown in Figure 8. Accordingly, the selectivity to POP´s and
DME observed over sample AMT-2 (with a pseudocrystalline
phase) should correspond to a material presenting ca. 88 % of
h-WO3 phase. In fact, XRD pattern shows a slight decomposition
of this phase into m-WO3 after the heat-treatment (Figure 1B,
pattern b).
0.50
0.45
0.40
0.35
0.30
60
50
40
30
20
0
20
40
60
80
100
Catalytic results in the oxydehydration of glycerol
h-WO3 phase content (%)
Tungsten and molybdenum/vanadium-based bronze phases
have been reported to be one of the best catalytic systems for
the production of acrolein (AC) and acrylic acid (AA) from
glycerol.[19-21, 47-51] In this regard, the acid-redox character of the
catalysts is pivotal. Acrolein is obtained by the direct dehydration
of glycerol on acid sites of moderated strength, while acrylic acid
is generated in a consecutive step by partial oxidation of acrolein
on redox sites. In fact, the presence of V4+/5+ and Mo5+/6+ species
within the crystal structure gives rise to the redox function in
these materials (likely structural O2- sites via Mars-van Krevelen
mechanism, as typically observed for redox oxides).[51] Therefore,
the presence of a bifunctional acid-redox catalyst, in which both
functions work at the same temperatures is necessary in order
to obtain acrylic acid in a one-pot reaction system. Figure 9
shows the effect of h-WO3 phase-content (in %) on the
selectivity to both acrolein and acrylic acid (AA+AC) and COx,
and the acrylic acid/acrolein ratio. As it can be seen, the
presence of V-containing h-WO3 crystals promotes the formation
of both acrolein and acrylic acid. In the opposite trend, lower
amounts of h-WO3 phase in the catalyst give rise to the main
formation of carbon oxides (rather than acrylic acid) and heavy-
compounds.[19] This is in line with H2-TPR results, in which V-
containing h-WO3 crystals present V5+ sites with higher
reducibility than those in V-containing m-WO3 crystals (Figure
S4), favoring the oxidation via structural O2- species, which
would give rise to the partial oxidation of acrolein to acrylic acid.
On the other hand, the lower reducibility of V5+ sites in W-V-O
catalysts presenting m-WO3 would favor the oxidation via
Figure 9. Variation of the selectivity to acrylic acid + acrolein () selectivity to
COx () and the acrylic acid / acrolein ratio () as a function of the amount of
h-WO3 phase in catalysts during the glycerol oxidehydration over W-V-O
materials. Reaction conditions: T= 320 ºC; Contact time, W/F, of 81 gCAT
(molGLY)-1. The amount of h-WO3 phase as in Table 1.
h
In this way, adjusting the reaction conditions (mainly
temperature, feed composition and contact time) when dealing
with multifunctional catalysts and multi-step reactions is a key
factor in order to achieve maximum yields to products. Thus,
using 100 % h-WO3-type W-V-O mixed oxide (AMT-1 sample)
at optimum reaction conditions it is possible to attain yields to
acrolein and acrylic acid of 23.6 and 32.5 %, respectively (T=
266ºC, and a rate of formation of acrylic acid per unit mass of
catalyst, STY, of 0.36 gAA h-1 gcat-1) (Table S1).
Conclusions
Several W-V-O catalysts presenting different proportions of h-
WO3 and m-WO3 phases have been synthesized. Selecting the
appropriate tungsten, vanadium and ammonium precursors,
materials with different morphologies and thermal stability can
be obtained. Indeed, block-shaped crystals undergo a higher
thermal decomposition of the initial h-WO3 phase into m-WO3,
while bronze-type phases, presenting 1D-type morphologies
(mainly rod and needle-shaped), show higher stability under the
same thermal treatment. Catalytic tests on the aerobic
transformation of methanol show that pure m-WO3-type phase
presents very low concentration of acid sites, showing mainly
redox sites (initial selectivity to POPs of c.a. 98 %). As h-WO3
phase proportion increases, acid function is progressively
generated, as it can be inferred from the selectivity profiles that
show increasing amounts of DME, reaching approximately 50 %
of initial selectivity to both POPs and DME over catalysts
presenting mainly h-WO3 type phase. Catalytic tests on glycerol
oxydehydration reaction indicate that the absence of acid sites in
m-WO3 type phases has a negative effect in the production of
acrolein and, accordingly, in the production of acrylic acid. The
presence of h-WO3 phase containing both acid and redox sites
working in a cooperative way increases the production of
acrolein in the first step and acrylic acid in the consecutive
adsorbed oxygen species (i.e. O- or O2 ) which would lead to
-
total oxidation of acrolein to carbon oxides.[51] Moreover, the
AA/AC ratio increases with h-WO3 content in the catalysts.
These observations not only mean that vanadium sites in m-
WO3 containing catalysts are not selective in the partial oxidation
of acrolein to acrylic acid, but also that their low acid
characteristics (as it can be inferred from TPD-NH3 and
methanol aerobic transformation results) give rise to low
selectivity to acrolein in the first reaction step, favoring the
formation heavy by-products. [19]
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
samples were cooled to 10 ºC and then exposed to flowing pure
ammonia for 5 min. With the aim to remove the physisorbed ammonia,
samples were cleaned using He flow (35 mL min-1) again. TPD-NH3 was
performed between 100 ºC and 500 ºC with a heating rate of 10 ºC min-1
by using a helium flow and maintained at 500 ºC for 15 min. The
desorbed ammonia was analyzed by online gas chromatography
(Shimadzu GC-14A) provided with a TCD.
reaction. Moreover, acrylic acid/acrolein ratio increases with the
h-WO3 proportion, which means that V-sites into the bronze
structure are selective to the partial oxidation of acrolein to
acrylic acid, whereas redox sites in m-WO3-type phase are less
selective in the partial oxidation of glycerol, producing essentially
carbon oxides (and also heavy-compounds).
X-ray photoelectron spectra were registered using a Physical Electronics
PHI 5700 spectrometer with non-monochromatic Mg-Kα radiation (300 W,
15 kV, and 1256.6 eV) with a multi-channel detector. Spectra of powder
samples were recorded in the constant pass energy mode at 29.35 eV,
Experimental Section
using
a 720 μm diameter analysis area. Charge referencing was
Synthesis of W-V-O oxides
measured against adventitious carbon (C 1s at 284.8 eV). A short
acquisition time of 10 min was first used to examine C 1s, V 2p regions to
avoid photoreduction of vanadium(V) species. 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. Recorded
spectra were always fitted using Gaussian–Lorentzian curves in order to
determine the binding energies of the different element core levels more
accurately.
Temperature programmed reduction experiments (H2-TPR) were carried
out under a N2/H2 flow (10% H2 at a total flow of 50 mL min-1) with 15 mg
of catalyst. The measurements were conducted in the temperature range
100-800 ºC at a heating rate of 10 ºC min-1.
Preparation of W-V-O materials was carried out under hydrothermal
conditions from gels containing different tungsten, vanadium and
ammonium precursors, while maintaining a constant W/V ratio of 1/0.3 in
the synthesis gel (Table 1). Gels were obtained from aqueous solutions
of the corresponding metal- (W- and V-) and ammonium-precursors:
ammonium metatungstate hydrate (≥85 wt% WO3 basis, Sigma–Aldrich),
tungstic acid (99%, Sigma Aldrich), vanadium(IV) oxide sulfate hydrate
(≥97%, Sigma–Aldrich), vanadyl acetylacetonate (99.98%, Sigma–
Aldrich), ammonium chloride (99.998%, Sigma–Aldrich), ammonium
sulfate (99.0 %, Sigma–Aldrich), ammonium acetate (98%, Sigma–
Aldrich) or aqueous ammonia (25 %, Panreac). These materials will be
named as AMT-series or TA-series, depending on the tungsten
precursors, i.e. ammonium metatungstate or tungstic acid, respectively.
The synthesis gels were introduced in Teflon-lined stainless steel
autoclaves, under N2 atmosphere (1 bar), and treated at 175 ºC for 48h.
The resulting solids were then filtered, washed with water, and dried at
100 ºC overnight. Finally, heat-treatment was carried out in two steps: i) a
heat-treatment at 200 ºC for 1 h, in air flow; and ii) a heat-treatment at
600 ºC for 2 h under N2 flow (15 mL min-1 g-1).
Reactivity tests
The aerobic transformation of methanol was carried out in a fixed bed
reactor in the 180-380 ºC temperature range, at atmospheric pressure
and a contact time, W/F, of 6.8 gCAT h (molCH3OH)-1. The feed consisted of
methanol/oxygen/nitrogen with a molar ratio of 6/13/81. Analysis of
reactants and reaction products was performed by means of gas
chromatography using two different chromatographic columns: i)
Molecular sieve 5 Å (3m length); and ii) RT-U-bond, fused Silica PLOT
(30 m, 0.53 i.d.).
Characterization of catalysts
Powder X-Ray diffraction (XRD) patterns were collected in a PANalytical
X’Pert PRO diffractometer equipped with and X´Celerator detector in a
Bragg-Brentano geometry using Kα radiation of copper.
The catalytic tests for the gas phase glycerol oxydehydration were also
conducted in a fixed-bed reactor at atmospheric pressure, with the
following reaction mixture: 2 mol% glycerol, 40 mol% water, 4 mol%
Surface areas were calculated by BET method, from adsorption
oxygen and 54 mol% helium, and a contact time W/F of 81 gCAT
h
isotherms recorded on
a Micromeritics ASAP 2000. Solids were
(molGLY)-1. The effluent stream was bubbled through a condenser, which
was refrigerated at 0-3 ºC. The remaining gaseous stream containing
carbon oxides and oxygen was analyzed by an online gas
chromatograph equipped with: (i) molecular sieve 5 Å (3 m); and (ii)
Porapak Q (3 m). The condensed aqueous solution containing all the
reaction products and the unconverted glycerol was analyzed by gas
chromatography using a Varian 3900 chromatograph equipped with a
100% dimethylpolysiloxane capillary column (100 m x 0.25 mm x 0.5).
degassed previously under vacuum at 400 ºC.
X-Ray energy-dispersive spectroscopy (XEDS) analyses were performed
using an Oxford LINK ISIS system connected to a JEOL 6300 electron
microscope with the SEMQUANT program, which introduces ZAF
correction. The counting time for each analysis was 100s.
Scanning electron microscopy (SEM) micrographs were carried out in a
field emission ZEISS Ultra-55 scanning electron microscope operating at
an accelerating voltage of 2 kV.
Raman spectra were collected with an inVia Renishaw spectrometer
using an exciting wavelength of 514 nm. The laser used was a Renishaw
HPNIR with an approximate power on the samples of about 15 Mw.
Thermogravimetric experiments were carried out in a Mettler-Toledo
TGA/SDTA 851 thermobalance. Approximately 10 mg of sample were
heated at 600 ºC at 10 ºC/min heating rate in a 50 mL/min synthetic air
flow.
Infrared spectra in the 400-4000 cm-1 region were acquired at room
temperature with a Nicolet 205xB spectrophotometer at a spectral
resolution of 1 cm-1 and 128 accumulations per scan.
Temperature programmed desorption of ammonia (TPD-NH3) was
carried out to evaluate the total acidity of the catalysts. Prior to the
adsorption of NH3, 80 mg of each sample were heated from room
temperature to 500ºC (heating rate: 10 ºC min-1) and then maintained at
500ºC during 10 min under flowing He (35 mL min-1). Subsequently,
Acknowledgements
The authors acknowledge the DGICYT in Spain CTQ2015-
68951-C3-1-R and CTQ2015-68951-C3-3-R. Authors from ITQ
also thank Project SEV-2016-0683 for financial support. D.D.
thanks MINECO and Severo Ochoa Excellence Program for his
fellowship (SVP-2014-068669). The group of research of Prof.
Fabrizio Cavani (University of Bologna, Italy) and Consorzio
INSTM (Firenze) are gratefully acknowledged for a PhD grant to
AC. Authors also thank the Electron Microscopy Service of
Universitat Politècnica de València for their support.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
[30] I. M. Szilágyi, J. Madarász, G. Pokol, P. Király, G. Tárkányi, S. Saukko,
J. Mizsei, A. L. Tóth, A. Szabó, K. Varga-Josepovits, Chem. Mater.
2008, 20, 4116-4125.
Keywords: glycerol • acrylic acid • oxydehydration • W–V mixed
oxides • hexagonal tungsten bronze
[31] C. Guo, S. Yin, P. Zhang, M. Yan, K. Adachi, T. Chonan, T. Sato, J.
Mater. Chem. 2010, 20, 8227-8229.
[1]
[2]
J. D. Guo, M. S. Whittingham, Int. J. Mod. Phys. B 1993, 7, 4145-4164.
H. Long, W. Zeng, H. Zhang, J. Mater. Sci.: Mater. Electron. 2015, 26,
4698-4707.
[32] P. Botella, E. García-González, J. M. López Nieto, J. M. González-
Calbet, Solid State Sci. 2005, 7, 507-519.
[3]
[4]
[5]
[6]
N. Haldolaarachchige, Q. Gibson, J. Krizan, R. J. Cava, Phys. Rev. B:
Condens. Matter Mater. Phys. 2014, 89, 104520.
Z.-F. Huang, J. Song, L. Pan, X. Zhang, L. Wang, J.-J. Zou, Adv. Mater.
2015, 27, 5309-5327.
[33] Y. Kong, H. Sun, X. Zhao, B. Gao, W. Fan, Appl. Catal. A 2015, 505,
447-455.
[34] P. Botella, B. Solsona, J. M. Lopez Nieto, P. Concepcion, J. L. Jorda, M.
T. Domenech-Carbo, Catal. Today 2010, 158, 162-169.
[35] W. P. Griffith, P. J. B. Lesniak, J. Chem. Soc. (A) 1969, 1066-1071.
[36] Z. Zheng, B. Yan, J. Zhang, Y. You, C. T. Lim, Z. Shen, T. Yu, Adv.
Mater. 2008, 20, 352-356.
T. Maiyalagan and B. Viswanathan, J. Power Sources 2008, 175, 789-
793.
a) M. F. Weber, A. J. Bevolo, H. R. Shanks, G. C. Danielson, J.
Electrochem. Soc. 1981, 128, 996-1003; b) B. Wickman, M.
Wesselmark, C. Lagergren, G. Lindbergh, Electrochim. Acta 2011, 56,
9496-9503.
[37] C. Sanchez, J. Livage, G. Lucazeau, J. Raman Spectrosc. 1982, 12,
68-72.
[38] I. M. Szilagyi, J. Madarasz, G. Pokol, F. Hange, G. Szalontai, K. Varga-
Josepovits, A. L. Toth, J. Therm. Anal. Calorim. 2009, 97, 11-18.
[39] N. E. Fouad, A. K. H. Nohman, M. A. Mohamed, M. I. Zaki, J. Anal.
Appl. Pyrolysis 2000, 56, 23-31.
[7]
[8]
[9]
K.R. Dey, T. Debnath, C.H. Rüscher, M. Sundberg, A. Hussain, J.
Mater. Sci. 2011, 46, 1388–1395.
Z. Zhang, J. Liu, J. Gu, L. Su, L. Cheng, Energy Environ. Sci. 2014, 7,
2535-2558.
[40] L. Huo, H. Zhao, F. Mauvy, S. Fourcade, C. Labrugere, M. Pouchard,
J.-C. Grenier, Solid State Sci. 2004, 6, 679-688.
M. Murawska, J. A. Cox, K. Miecznikowski, J. Solid State Electrochem.
2014, 18, 3003-3010.
[41] D. Perra, N. Drenchev, K. Chakarova, M. G. Cutrufello, K. Hadjiivanov,
RSC Adv. 2014, 4, 56183-56187.
[10] X. P. Li, X. D. Xiang, H. Y. Yang, X. J. Wang, C. L. Tan, W. S. Li, Fuel
Cells 2013, 13, 314-318.
[42] K. Oshihara, T. Hisano, W. Ueda, Top. Catal. 2001, 15, 153-160.
[43] J. R. Sohn, M. Y. Park, Langmuir 1998, 14, 6140-6145.
[44] I. E. Wachs, K. Routray, ACS Catal. 2012, 2, 1235-1246.
[45] J. M. Tatiboueet, Appl. Catal. A: Gen 1997, 148, 213-252.
[46] M. Badlani, I. Wachs, Catal. Lett. 2001, 75, 137-149.
[47] A. Chieregato, M. D. Soriano, F. Basile, G. Liosi, S. Zamora, P.
Concepcion, F. Cavani, J. M. Lopez Nieto, Appl. Catal. B 2014, 150-
151, 37-46.
[11] P. J. Kulesza, I. S. Pieta, I. A. Rutkowska, A. Wadas, D. Marks, K. Klak,
L. Stobinski, J. A. Cox, Electrochim. Acta 2013, 110, 474-483.
[12] B. Broyde, J. Catal. 1968, 10, 13-18.
[13] G. Li, Ch. Guo, M. Yan, S. Liu, Appl. Catal. B: Environ. 2016, 183, 142-
148.
[14] a) Y. Xi, Z. Chen, V. Gan Wei Kiat, L. Huang, H. Cheng, Phys. Chem.
Chem. Phys. 2015, 17, 9698-9705; b) Y. Liu, S. Shrestha, W. E.
Mustain, ACS Catal. 2012, 2, 456-463.
[48] K. Omata, K. Matsumoto, T. Murayama, W. Ueda, Catal. Today 2016,
259, 205-212.
[15] J. Song, Z.-F. Huang, L. Pan, J.-J. Zou, X. Zhang, L. Wang, ACS Catal.
2015, 5, 6594-6599.
[49] A. Chieregato, F. Basile, P. Concepción, S. Guidetti, G. Liosi, M. D.
Soriano, C. Trevisanut, F. Cavani, J. M. Lopez Nieto, Catal. Today
2012, 197, 58-65.
[16] K. Okumura, S. Ishida, R. Takahata, N. Katada, Catal. Today 2013, 204,
197-203.
[17] C. Yue, X. Zhu, M. Rigutto, E. Hensen, Appl. Catal. B 2015, 163, 370-
381.
[50] Y. S. Yun, K. R. Lee, H. Park, T. Y. Kim, D. Yun, J. W. Han, J. Yi, ACS
Catal. 2015, 5, 82-94.
[18] P. Botella, B. Solsona, E. Garcia-González, J. M. González-Calbet, J.
M. Lopez Nieto, Chem. Commun. 2007, 5040-5042.
[19] M. D. Soriano, P. Concepcion, J. M. Lopez Nieto, F. Cavani, S. Guidetti,
C. Trevisanut, Green Chem. 2011, 13, 2954-2962.
[20] A. Chieregato, M. D. Soriano, E. García-González, G. Puglia, F. Basile,
P. Concepción, C. Bandinelli, J. M. Lopez Nieto, F. Cavani,
ChemSusChem 2015, 8, 398-406.
[51] B. Katryniok, T. Bonnotte, F. Dumeignil, S. Paul in Chemicals and Fuels
from Bio-Based Building Blocks (Eds: F. Cavani, S. Albonetti, F. Basile,
A. Gandini), Wiley-VCH, Weinheim, 2016, pp. 217-243.
[21] M. D. Soriano, A. Chieregato, S. Zamora, F. Basile, F. Cavani, J. M.
Lopez Nieto, Top. Catal. 2016, 59, 178-185.
[22] D. Nagy, D. Nagy, I. M. Szilagyi, X. Fan, RSC Adv. 2016, 6, 33743-
33754.
[23] S. Lin, Y. Guo, X. Li, Y. Liu, Mater. Lett. 2015, 152, 102-104.
[24] B. Miao, W. Zeng, S. Hussain, Q. Mei, S. Xu, H. Zhang, Y. Li, T. Li,
Mater. Lett. 2015, 147, 12-15.
[25] A. C. Marques, L. Santos, M. N. Costa, J. M. Dantas, P. Duarte, A.
Goncalves, R. Martins, C. A. Salgueiro, E. Fortunato, Sci. Rep. 2015, 5,
9910.
[26] A. Magneli, Acta Chem. Scand. 1953, 7, 315-324.
[27] C. D. Vanderpool, M. B. MacInnis, J. C. Patton, US Patent, US3936362,
1976.
[28] M. Sanchez Sanchez, F. Girgsdies, M. Jastak, P. Kube, R. Schlögl, A.
Trunschke, Angew. Chem. Int. Ed. 2012, 51, 7194-7197.
[29] E. Garcia-Gonzalez, M. D. Soriano, E. Urones-Garrote, J. M. Lopez
Nieto, Dalton Trans. 2014, 43, 14644-14652.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800059
European Journal of Inorganic Chemistry
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Text for Table of Contents
D. Delgado, A. Chieregato, M. D.
Soriano, E. Rodríguez-Aguado, L. Ruiz-
Rodríguez, E. Rodríguez Castellón and
J. M. Lopez Nieto*
Page No. – Page No.
((Insert TOC Graphic here: max.
Influence of phase composition of
bulk tungsten-vanadium oxides on
the one-pot synthesis of acrylic acid
from glycerol
width: 5.5 cm; max. height: 5.0 cm))
Layout 2:
FULL PAPER
D. Delgado, A. Chieregato, M. D.
Soriano, E. Rodríguez-Aguado, L. Ruiz-
Rodríguez, E. Rodríguez-Castellón and
J. M. Lopez Nieto*
Acrolein
Acrylic acid
COx
Glycerol
200 nm
200 nm
h-WO3-type W-V-O oxide
(acid-redox sites)
m-WO3-type W-V-O oxide
Page No. – Page No.
(redox sites)
Influence of phase composition of
bulk tungsten-vanadium oxides on
the one-pot synthesis of acrylic acid
from glycerol
+
By selecting the appropriate W-, V-, and NH4 -precursors it is possible to tune the
acid-redox properties in the system W-V-O. These catalytic properties depend on
the content of h-WO3-type (acid/redox sites) and/or m-WO3 (redox sites) phases in
the catalysts. The presence of V-containing h-WO3 type crystals is pivotal to
achieve good yields to acrolein and acrylic acid in the one-pot gas-phase
oxydehydration of glycerol.
This article is protected by copyright. All rights reserved.