Selective oxidation of propane and ethane on diluted Mo-V-Nb-Te mixed-oxide catalysts

Article abstract of DOI:10.1016/j.jcat.2007.09.019

Te-free and Te-containing Mo-V-Nb mixed oxide catalysts were diluted with several metal oxides (SiO₂, γ-Al₂O₃, α-Al₂O₃, Nb₂O₅, or ZrO₂), characterized, and tested in the oxidation of ethane and propane. Bulk and diluted Mo-V-Nb-Te catalysts exhibited high selectivity to ethylene (up to 96%) at ethane conversions < 10 %, whereas the corresponding Te-free catalysts exhibited lower selectivity to ethylene. The selectivity to ethylene decreased with the ethane conversion, with this effect depending strongly on the diluter and the catalyst composition. For propane oxidation, the presence of diluter exerted a negative effect on catalytic performance (decreasing the formation of acrylic acid), and α-Al₂O₃ can be considered only a relatively efficient diluter. The higher or lower interaction between diluter and active-phase precursors, promoting or hindering an unfavorable formation of the active and selective crystalline phase [i.e., Te₂M₂₀O₅₇ (M = Mo, V, and Nb)], determines the catalytic performance of these materials.

Full text of DOI:10.1016/j.jcat.2007.09.019

Journal of Catalysis 252 (2007) 271–280
www.elsevier.com/locate/jcat
Selective oxidation of propane and ethane on diluted Mo–V–Nb–Te
mixed-oxide catalysts
a,b
b
a
b
a
a,∗
B. Solsona , M.I. Vázquez , F. Ivars , A. Dejoz , P. Concepción , J.M. López Nieto
a
Instituto de Tecnología Química, UPV-CSIC, Avda. Naranjos s/n, 46022 Valencia, Spain
b
Departamento de Ingeniería Química, Universidad de Valencia, Dr. Moliner 50, 46100 Valencia, Spain
Received 28 May 2007; revised 6 August 2007; accepted 23 September 2007
Available online 25 October 2007
Abstract
Te-free and Te-containing Mo–V–Nb mixed oxide catalysts were diluted with several metal oxides (SiO , γ -Al O , α-Al O , Nb O , or ZrO ),
2
2
3
2
3
2
5
2
characterized, and tested in the oxidation of ethane and propane. Bulk and diluted Mo–V–Nb–Te catalysts exhibited high selectivity to ethylene
up to 96%) at ethane conversions <10%, whereas the corresponding Te-free catalysts exhibited lower selectivity to ethylene. The selectivity
to ethylene decreased with the ethane conversion, with this effect depending strongly on the diluter and the catalyst composition. For propane
oxidation, the presence of diluter exerted a negative effect on catalytic performance (decreasing the formation of acrylic acid), and α-Al O
(
2
3
can be considered only a relatively efficient diluter. The higher or lower interaction between diluter and active-phase precursors, promoting or
hindering an unfavorable formation of the active and selective crystalline phase [i.e., Te M
performance of these materials.
O
20 57
(M = Mo, V, and Nb)], determines the catalytic
2
©
2007 Elsevier Inc. All rights reserved.
Keywords: Propane oxidation; Ethane oxidation; Acrylic acid; Ethylene; Mo–V–Nb–Te diluted catalysts; Catalyst characterization (FT-Raman, XRD); α-Al O
2
3
1
. Introduction
lene transformed into acrylic acid in two separated reactors,
with acrolein as a reaction intermediate).
Although vanadium oxides supported on different materials
were initially proposed as selective in the ODH of light alka-
nes [1–6], mixed metal oxides seem to be the most promising
catalytic systems for the selective oxidative activation of C2–C3
alkanes [6–10]. In this way, MoVTe(Sb)NbO catalysts are the
most effective catalysts in the (amm)oxidation of propane to
acrylonitrile and acrylic acid [11–13] and for the ODH of
ethane to ethylene [14,15] at reaction temperatures as low as
The oxidative dehydrogenation (ODH) and partial oxidation
of light alkanes are drawing great interest due to the wide avail-
ability of these hydrocarbons. The ODH of ethane over redox
catalysts is an interesting alternative for industrial ethylene pro-
duction, because it has some advantages with respect to steam
cracking or catalytic dehydrogenation [1–6]: (i) It is an exother-
mic reaction; (ii) it does not need a catalyst regeneration step,
because catalyst regeneration occurs in situ by the oxygen feed;
◦
◦
350–425 C. These materials are characterized by the presence
and (iii) relatively low reaction temperatures (350–500 C) are
of an active and selective crystalline phase, that is, the or-
thorhombic Te2M20O57 or (SbO)2M20O56 (M = Mo, V, Nb),
the so-called M1 phase [16–21]. In addition to this, other crys-
talline phases also can be observed, depending on the catalyst
preparation procedure [20].
MoVTe(Sb)NbO mixed metal oxides have low surface area,
which limits their potential industrial applications. The incor-
poration of metal oxides on a support improves the mechanical
properties of the catalysts and generally modifies their catalytic
behavior, which can have an unfavorable effect on catalytic per-
required.
The partial oxidation of propane to acrylic acid also has
many advantages over the current industrial process to obtain
acrylic acid [6–10]. For example, propane is used instead of
the more expensive propylene, and only one step is required, in
contrast to the current two-step industrial process (i.e., propy-
*
Corresponding author. Fax: +34 96 3877809.
E-mail address: jmlopez@itq.upv.es (J.M. López Nieto).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.09.019

272
B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
−
formance in some cases. Holmberg et al. [22] recently reported
that Mo–V–Te–Nb–O catalysts diluted with silica, γ -alumina,
or titania are significantly less efficient than the bulk catalyst in
propane ammoxidation to acrylonitrile.
tive, 50×; spectral resolution, 2.5 cm 1; integration time, 20 s
per spectrum; and 50 scans. All samples were measured without
previous activation. Raman spectra were obtained at different
points in each sample to determine homogeneity and/or hetero-
geneity of the sample. The spatial resolution of each analysis
was about 0.4 µm.
In the present paper we will describe the characterization and
catalytic performance for the selective oxidative transformation
of propane and ethane of Mo–V–Nb–Te mixed oxides diluted
with several metal oxides. We discuss how the interaction be-
tween the diluter and the mixed metal oxide precursors during
the catalyst preparation strongly influences both the nature of
crystalline phases and the catalytic performance. In addition,
we demonstrate that the presence of diluter has a very low in-
fluence on the selectivity to ethylene from ethane, in contrast to
that observed for the selectivity to acrylic acid from propane.
2.3. Catalytic tests
The catalytic experiments were carried out at atmospheric
pressure, in the 340–420 C range, using a fixed-bed quartz
◦
tubular reactor (20 mm i.d., 400 mm long). Catalyst sam-
ples were introduced in the reactor diluted with silicon car-
bide to maintain a constant volume in the catalytic bed. The
feed consisted of a mixture of C3H8/O2/He/H2O (with a mo-
lar ratio of 4/8/58/30), C3H6/O2/He/H2O (with a molar ra-
tio of 1.7/6.8/76.5/15), or C2H6/O2/He (with a molar ratio
of 30/10/60 to 30/30/40). The amount of catalyst was var-
ied from 0.2 to 2 g to obtain several contact times. Reactant
and products were analysed by gas chromatography using two
packed columns: molecular sieve 5 Å (2.5 m) and Porapak Q
(3 m). Blank runs showed no conversion in the temperature
range studied [15,20].
2
. Experimental
2
.1. Catalyst preparation
Bulk Mo–V–Nb–Te mixed metal oxide catalyst was pre-
pared by a slurry method [23]. An aqueous niobium oxalate
solution was added to an aqueous solution of ammonium meta-
vanadate, telluric acid, and ammonium heptamolybdate. The
◦
solution was rotaevaporated, dried at 100 C overnight, and
heat-treated at 600 C in flowing N2 for 2 h.
◦
3. Results and discussion
Diluted Mo–V–Nb–Te catalysts with a 20 wt% of the metal
oxide diluter were prepared similarly as the bulk catalyst. An
aqueous solution containing Mo, V, Te, and Nb salts was rotae-
vaporated until a wet paste was obtained, to which the support
3.1. Catalyst characterization
The characteristics of bulk and diluted catalysts are summa-
rized in Table 1. The XRD patterns of the Te-containing and
Te-free Mo–V–Nb catalysts are shown in Fig. 1. In all cases,
◦
was added. After drying at 100 C overnight, the solid was heat-
◦
treated at 600 C in flowing N2 for 2 h.
2
◦
Commercial SiO2 (Degussa, Aerosil-200, SBET = 186 m
the catalysts show the presence of a peak at 2θ = 21.8–22.0
−1
2
−1
g
(
), α-Al2O3 (Pechiney, SBET = 9 m g ), and γ -Al2O3
characteristic of oxidic bronzes with an interplanar distance of
ca. 4 Å.
2
−1
Sued-Chimie, SBET = 187 m g ) were used. ZrO2 was pre-
pared by hydrolysing ZrOCl2·8H2O with a solution of NH4OH
The XRD patterns of Te-containing catalysts suggest the
presence of the orthorhombic Te2M20O57 (M1 phase) and the
pseudo-orthorhombic Te0.33MO3.33 (M2 phase) as the main
crystalline phases (Fig. 1, patterns a–f), even though they
present different relative intensities for their corresponding
most important reflections. A first estimation on the variation
of the amount of each phase in these catalysts was obtained
based on the variation in intensity of the reflections observed
◦
◦
(
3
at pH 7), drying at 100 C, and calcining in air at 550 C for
2
−1
h (SBET = 38 m g ) [24]. Nb2O5 was obtained by calcina-
2
◦
−1
tion of Nb-oxalate in air at 550 C for 3 h (SBET = 31 m g ).
Bulk and diluted Mo–V–Nb–O catalysts were also prepared
similarly as the corresponding Mo–V–Nb–Te catalysts. In this
case, SiO2, α-Al2O3, γ -Al2O3 were the only metal oxide di-
luters studied. The denomination and characteristics of the cat-
alysts are given in Table 1.
◦
◦
at 2θ = 27.2 and 22.0 (I27.2/I22.0, related to the presence of
◦
the M1 phase) or the reflections observed at 2θ = 28.2 and
◦
2
.2. Catalyst characterization
22.0 (I28.2/I22.0, related to the presence of the M2 phase). The
resulting values are summarized in Table 1. Accordingly, the
presence of the M1 phase decreases with the incorporation of
diluter, especially in the case of ZrO2- or Nb2O5-containing
samples. On the other hand, the formation of M2 phase seems
to be favored in SiO2-, ZrO2-, or Nb2O5-containing samples
with respect to the undiluted catalyst.
Note that besides the M1 and M2 phases and the corre-
sponding crystalline metal oxide diluter (α-Al2O3 and traces
of Nb2O5 or ZrO2), other crystalline phases were detected
as well. Al2(MoO4)3 (JCPDS: 23-764) is clearly observed
in the γ -Al2O3-containing catalyst (Fig. 1A, pattern c), and
(V0.07Mo0.93)5O14 (JCPDS: 31-1437), Nb0.09Mo0.91O2.8
Catalyst surface areas were determined by multipoint N2
adsorption at 77 K, and data were treated in accordance with
the BET method. X-ray diffraction patterns (XRD) were col-
lected using a Philips X’Pert diffractometer equipped with a
graphite monochromator, operating at 40 kV and 30 mA and
using nickel-filtered CuKα radiation (λ = 0.1542 nm).
FT-Raman spectra were recorded with an “in via” Renishaw
spectrometer equipped with an Olympus microscope. The sam-
+
ples were excited by the 514.5 nm line of an Ar laser (Spectra
Physics model 171) with a laser power of 2.5 mW. The follow-
ing spectrometer characteristics were used: microscope objec-

B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
273
Table 1
Characteristics of bulk and diluted (20 wt% of support) MoVTeNbO catalysts
Catalyst
Diluter
Surface area
Mo/V/Nb/Te
atomic ratio
Crystalline phases
(XRD)
I
/I
c
ratio
I
/I
28.2 22.0
ratio
2
7.2 22.0
2
a
b
d
(m /g)
(XRD)
(XRD)
CT-1
–
SiO
γ -Al O
α-Al O
Nb O
ZrO
5
1/0.22/0.13/0.23
1/0.21/0.13/0.22
1/0.26/0.13/0.25
1/0.24/0.13/0.24
1/0.23/0.22/0.23
nd
Te M
O
MO
, Te
MO
3.33
0.29
0.17
0.20
0.24
0.08
0.10
0.66
0.94
0.44
0.57
0.96
0.88
2
20 57
0.33
CT-2Si
CT-3Al
CT-4aAl
CT-5Nb
CT-6Zr
21
24
6
9
7
Te
>Te M
O
20 57
2
0.33
3.33
2
Te M
O
>Te
, Te
MO
, Al (MoO )
, α-Al O
2
3
3
2
20 57
0.33
3.33 2 4 3
Te M
O
MO
2
2
20 57
0.33
3.33 2 3
Te
Te
MO
MO
>Te M O , >M O , TeMo O
>Te M O , M O , TeMo O
2 20 57 5 14 5 16
2
5
0.33
0.33
3.33
3.33
2 20 57 5 14 5 16
2
C-7
–
SiO
γ -Al O
α-Al O
6
31
35
6
1/0.22/0.13/0
1/0.22/0.13/0
1/0.22/0.13/0
1/0.22/0.13/0
M O , Mo
V
V
V
V
O
O
–
–
–
–
5
14
0.67 0.33
2
2
C-8Si
C-9Al
C-10aAl
M O , Mo
2
5
14
0.67 0.33
M O , Mo
O , Al (MoO )
O , α-Al O
2 3
2
3
3
5
14
0.67 0.33 2 2 4 3
M O , Mo
–
–
2
5
14
0.67 0.33
2
a
Atomic ratio obtained by atomic absorption.
b
c
d
Te M
O
, Te
MO
and M O with M = Mo, V and/or Nb.
2
20 57
0.33
3.33
5 14
Estimation of the relative amount of Te M
O
(M1 phase).
(M2 phase).
3.33
2
20 57
Estimation of the relative amount of Te
MO
0.33
Fig. 1. XRD patterns of bulk and diluted MoVNb(Te)O catalysts: (a) CT-1; (b) CT-2Si; (c) CT-3Al; (d) CT-4aAl; (e) CT-5Nb; (f) CT-6Zr; (g) C-7; (h) C-8Si;
(
i) C-9Al; (j) C-10aAl. Symbols: Te M
O
V
(M1 phase) ("); Te
O (1).
2
MO
(M2 phase) (!); TeMo O (+); (V0.07Mo
) O and/or Nb
Mo
O
(2);
2
20 57
0.33
3.33
5
16
0.93 5 14
0.09
0.91 2.8
α-Al O (P); Al (MoO ) (×); Mo
2
3
2
4 3
0.67 0.33
(
JCPDS: 27-0310), and TeMo5O16 (JCPDS: 31-874) can be
874 cm 1 with a broad shoulder toward lower frequencies, in
−
−
1
seen in Nb2O5- or ZrO2-diluted MoVNbTeO catalysts (Fig. 1,
patterns e and f, respectively).
The XRD patterns of bulk and diluted Mo–V–Nb mixed-
oxide catalysts are also shown in Fig. 1 (patterns g–j). These
patterns suggest the presence of (Mo0.93V0.07)5O14 (JCPDS:
the 770–840 cm Raman region, and a weak shoulder toward
−
1
higher frequencies, at around 970 cm . In addition to these,
−
1
bands at lower frequencies (470 and 437 cm ) are also ob-
served. This spectrum is similar to those reported previously
[13,25]. The shoulder at 970 cm can be assigned to stretching
−
1
3
1-1437) and/or Nb0.09Mo0.91O2.8 (JCPDS: 27-0310) in ad-
vibrations of terminal Mo=O and V=O bonds [26], whereas
−
1
dition to Mo0.67V0.33O2 (JCPDS: 30-0849), although with a
different relative intensity for the most important reflections.
Furthermore, the XRD pattern of γ -Al2O3-diluted MoVNbO
catalyst (C-9Al) also suggests the presence of Al2(MoO4)3
the bands at 770–880 cm
appear in a typical regime of
asymmetric Me–O–Me bridge stretching modes, and the low-
−
1
frequency bands (around 470 cm ) appear in a typical regime
of their symmetric Me–O–Me bridge stretching modes.
(
JCPDS: 23-764), whereas the characteristic diffraction peaks
For comparative purposes, the Raman spectra of two bulk
MoVTeNbO samples containing pure M1 or pure M2 phase
also have been included (Fig. 2, spectra b and c), because,
as reported for the XRD pattern of bulk Mo–V–Nb–Te sam-
ple, it shows the presence of Te2M20O57 (M1 phase) and
of α-Al2O3 (JCPDS: 10-0173) remains in the α-Al2O3-diluted
sample (Fig. 1, pattern j).
The Raman spectrum of bulk Mo–V–Nb–Te catalyst, shown
in Fig. 2 (spectrum a), exhibits an intense Raman band at

274
B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
appear to be quite homogeneous, with Raman spectra simi-
lar to those seen for the undiluted catalyst (Fig. 3, spectra a
and b, respectively), although a higher contribution of the bands
−
1
at 796 and 437 cm can be seen in the SiO2-containing sam-
ple. Accordingly, a greater contribution of the M2 phase can
be inferred in these catalysts, as previously suggested by XRD
measurements (see Table 1).
On the other hand, some heterogeneity with the presence
of other oxide phases can be seen in catalysts diluted with γ -
Al2O3, Nb2O5, or ZrO2 (Fig. 3, spectra c–h). Two types of
Raman spectra are observed in these cases. For the γ -Al2O3-
containing catalyst, characteristic bands related to M1 and M2
(Fig. 3, spectrum c) are seen; however, a different Raman spec-
trum characterized by the presence of bands at 974, 818, and
−
1
3
70 cm also can be obtained in some particles (Fig. 3, spec-
trum d). According to the literature, the broad band centered
−
1
at 950 cm has been attributed to a molybdenum–support
interaction in γ -Al2O3-supported molybdenum oxides [27],
−
1
whereas Raman bands at 1006 and 380 cm have been as-
signed to the formation of Al2(MoO4)3 [28]. In our case, the
band at 1006 is overlapped by the strong Raman band centered
Fig. 2. FT-Raman spectra of Mo–V–Nb–Te catalysts: (a) CT-1; (b) pure M1
phase; (c) pure M2 phase.
−
1
at 970 cm
.
The Raman spectra of Nb2O5-containing Mo–V–Nb–Te cat-
Te0.33MO3.33 (M2 phase). The most intense Raman band at
alyst (Fig. 3, spectra e and f) show the presence of the M2 phase
−1
8
74 cm is present on both pure phases, whereas different
−1
(
Raman bands at 796 and 437 cm ). Moreover, these spec-
Me–O–Me stretching modes can be observed in both crystalline
tra suggest some heterogeneity. In this way, the presence of a
−
1
phases. In this way, bands at 796 and 437 cm are observed in
the M2 phase and are shifted to higher frequencies (827 and
−1
broad, intense Raman band at 690 cm indicates particles con-
taining Nb2O5 oxide (Fig. 3, spectrum f) [29].
−
1
4
70 cm ) in the M1 phase.
For the ZrO -containing sample, two types of spectra can be
2
Raman spectra of diluted Mo–V–Nb–Te samples are shown
observed (Fig. 3, spectra g and h). The first type suggests the
in Fig. 3. Both the SiO2- and α-Al2O3-containing catalysts
presence of the M2 phase (with the Raman bands at 796 and
Fig. 3. FT-Raman spectra of diluted MoVTeNbO samples: CT-2Si (a); CT-4aAl (b); CT-3Al (c and d); CT-5Nb (e and f); CT-6Zr (g and h).

B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
275
−
−1
4
37 cm 1) (Fig. 3, spectrum g); the second indicates the pres-
and 393 cm can be seen in all samples. These bands can be
related to a Mo–V–Nb mixed oxide [31–33], as suggested from
the corresponding XRD patterns. Bands at 970, 900, 850, and
ence of ZrMo2O8 (characterized by strong Raman bands at 748,
44, and 1000 cm [30]), together with bands at 909, 863,
45, 760, 715, and 685 cm (Fig. 3, spectrum h). The band at
37–640 cm is related to ZrO2 [30], whereas (as discussed
−1
9
8
6
−
1
−1
710 cm have been attributed to Mo5O14; some of these shift
to 970, 920, 855, 817, and 610 cm for (Mo0.8V0.1W0.1)5O14
−1
−1
later) the rest of these bands correspond to Mo–V–Nb mixed
oxide, similar to those observed on Te-free catalysts. Based on
the XRD results, it has been proposed that the formation of the
M2 crystalline phase is mainly favored over the M1 phase in the
presence of Nb2O5 or ZrO2. The Raman spectra of these cata-
lysts show partial incorporation of molybdenum species to the
metal oxides used as diluters.
[32,33]. Accordingly, M5O14 (M = Mo, V, and/or Nb) are
mainly formed in our Mo–V–Nb catalysts, with the shift of the
bands observed in our case explained by the different concen-
trations of transition metal ions [32].
In addition to these, new bands at 969, 818, and 365 cm
can be seen in the γ -Al2O3-diluted Mo–V–Nb–O catalyst
−1
(Fig. 4, spectrum c), whereas bands at 739, 568, 492, 459, 360,
−
1
The Raman spectra of bulk and diluted Mo–V–Nb catalyst
are shown in Fig. 4. Bands at 980, 909, 863, 845, 715, 685, 463,
344, 226, and 203 cm are observed on the sample diluted
with α-Al2O3 (Fig. 4, spectrum d). The former bands are as-
sociated with the presence of Al2(MoO4)3 [28]; the latter, with
the presence of polymolybdates [27]. However, and in addition
to this, other Mo–V–Nb mixed oxides [i.e., (Mo0.93V0.07)5O14,
Nb0.09Mo0.91O2.8 or Mo0.67V0.33O2] also can be formed, in
agreement with the XRD results. Accordingly, M5O14-type
crystals (M = Mo, V, and Nb), rather than M1-/M2-type crys-
talline phase, can be inferred in the absence of tellurium cations.
3
3
.2. Catalytic results
.2.1. Propane and propylene oxidation
Blank tests in the propane oxidation using the supports alone
(
4
α-Al2O3, γ -Al2O3, Nb2O5, ZrO2, SiO2) were carried out at
◦
00 C and with a contact time of 600 g hmol . In all cases,
cat
−1
C
3
the propane conversion achieved did not reach 1.0%.
Table 2 summarizes the catalytic results obtained in the ox-
◦
idation of propane over bulk and diluted catalysts at 400 C.
Significant differences can be seen among the catalysts tested.
Acrylic acid, propylene, acetic acid, carbon monoxide, and car-
bon dioxide were the main reaction products on Te-containing
catalysts, with acetone and acrolein as minor products (the car-
bon balance of all catalytic tests was 100 ± 3%). However, the
presence of a diluter had a negative influence on the selectivity
to acrylic acid on these catalysts. Also note that both the cat-
Fig. 4. FT-Raman spectra of bulk and diluted Mo–V–Nb catalysts: C-7 (a);
C-8Si (b); C-9Al (c); C-10aAl (d).
Table 2
a
Oxidation of propane on bulk and diluted Mo–V–Nb–(Te) mixed oxide catalysts
b
c
ꢀ
d
Catalyst
W/F
Propane conversion
%)
Selectivity (%)
Acrylic acid
STY
STY
AA
AA
(
Propylene
Acetic acid
CO
8.8
16.1
20.2
10.9
23.6
26.7
CO
2
CT-1
300
600
600
600
600
600
37.0
19.3
31.2
30.2
24.1
11.6
68.0
38.3
23.6
57.2
22.5
8.7
6.8
11.9
6.4
7.2
8.6
3.1
3.7
4.4
3.8
3.5
1.1
13.2
29.7
45.0
20.5
41.6
47.2
59
59
11
11
27
7.9
1.5
CT-2Si
CT-3Al
CT-4aAl
CT-5Nb
CT-6Zr
8.7
8.7
21
6.3
1.2
16.0
C-7
300
300
300
300
15.0
6.9
18.8
6.6
0
0
0
0
12.7
24.8
11.0
25.2
3.5
2.1
2.4
1.7
40.7
35.5
42.1
33.8
42.9
37.1
44.2
38.9
0
0
0
0
0
0
0
0
C-8Si
C-9Al
C-10aAl
a
◦
Catalytic results obtained at a reaction temperature = 400 C and a C /O /He/steam molar ratio of 4/8/58/30.
3
2
b
−1
Contact time, W/F , in g hmol
.
cat
C
3
c
−1 −1
h .
Space time yield, STY , in g kg
AA
AA cat
d
ꢀ
−1
−1
h ).
AA MoVTeNb
STY
is the rate of acrylic acid formation per unit mass of active components (in g kg
AA

276
B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
−
C
3
1
Fig. 5. Variation of the propane conversion with the reaction temperature at a contact time, W/F , of 300g hmol (a) and variation of the sum of the selectivity to
cat
◦
partial oxidation products (POP: propylene, acrylic acid, acrolein, acetone, and acetic acid) with the propane conversion at 380–400 C (b) on undiluted and diluted
Mo–V–Nb–Te catalysts. Symbols: (2) CT-1; (!) CT-2Si; (Q) CT-3Al; (e) CT-4aAl; (+) CT-5Nb; (×) CT-6Zr.
◦
Fig. 6. Variation of the selectivity to acrylic acid (a), propylene (b), CO (c) and CO (d) with the propane conversion, in the 380–400 C temperature range, during
2
the oxidation of propane over undiluted and diluted Mo–V–Nb–Te catalysts. Symbols: (2) CT-1; (!) CT-2Si; (Q) CT-3Al; (e) CT-4aAl; (+) CT-5Nb; (×) CT-6Zr.
alytic activity and selectivity to acrylic acid strongly depended
on the nature of the diluter incorporated.
No acrylic acid was detected over Te-free catalysts (regard-
less of the diluter used), although propylene and acetic acid
were observed in all cases.
Fig. 5a shows the variation of the propane conversion with
the reaction temperature obtained during propane oxidation
on Mo–V–Nb–Te catalysts. Accordingly, the catalytic activity
for Te-containing catalysts decreased as follows: undiluted >
with the propane conversion on Te-containing catalysts. A very
high initial selectivity to POP (ca. 90%) was found for all
catalysts, but the selectivity to POPs decreased with propane
conversion in a different extension, depending on the di-
luter.
Fig. 6 shows the evolution of the selectivity to the main re-
action products (acrylic acid, propylene, CO, and CO2) with
◦
propane conversion at reaction temperatures of 380–400 C on
Mo–V–Nb–Te-containing catalysts. The selectivity to acrylic
acid varied as follows: bulk > α-Al2O3- ꢁ SiO2- > γ -Al2O3-
> Nb2O5- > ZrO2-containing catalyst. This indicates that
α-Al O is preferred as diluter, although this catalyst exhib-
γ -Al O - > α-Al O - > Nb O - > SiO - > ZrO -containing
catalyst.
Fig. 5b plots the variation of the selectivity to partial oxi-
dation products (POP: propylene + acrylic acid + acetic acid)
2
3
2
3
2
5
2
2
2
3

B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
277
Table 3
a
Oxidation of propylene on bulk and diluted Mo–V–Nb–Te–O catalysts
b
c
ꢀ
AA
d
Catalyst
W/F
Propylene
conversion (%)
Selectivity (%)
Acrylic acid
STY
STY
AA
Acetone
Acetic acid
CO
8.4
8.0
15.5
10.7
18.4
14.9
CO
2
CT-1
75
150
150
150
150
150
86.5
74.8
78.9
78.5
65.3
54.9
87.2
78.6
71.9
83.9
68.1
73.7
0.8
1.0
1.1
0.8
1.0
1.2
0
3.5
10.4
9.9
4.3
9.4
724
282
272
316
213
194
724
353
340
395
266
243
CT-2Si
CT-3Al
CT-4aAl
CT-5Nb
CT-6Zr
1.7
1.4
0.2
2.6
1.5
4.3
a
◦
Catalytic results obtained at a reaction temperature = 400 C, and a C
−1
/O /He/steam molar ratio of 1.7/6.8/76.5/15.
2
C H =
3
6
b
Contact time, W/F , in g hmol
.
cat
C =
3
c
−1 −1
h .
Space time yield, STY , in g kg
AA
AA cat
d
ꢀ
−1
−1
h ).
AA MoVTeNb
STY
is the rate of acrylic acid formation per unit mass of active components (in g kg
AA
ited lower acrylic formation than the corresponding undiluted
catalyst.
Table 4
Oxidation of ethane on diluted Mo–V–Nb–Te–O and Mo–V–Nb–O catalysts
a
b
ꢀ
C H
2 4
c
Also note that among POP, propylene was the main primary
reaction product from propane (see Fig. 6b). Meanwhile, sim-
ilar selectivities to propylene were observed among the bulk
and diluted catalysts; the selectivity to CO and CO2 carbon ox-
ides (Figs. 6c and 6d, respectively) exhibits an opposite trend
to that observed for the selectivity to acrylic acid (Fig. 6a).
These findings suggest that the most important difference be-
tween undiluted and diluted Te-containing catalysts is related
to the higher or lower extension of the overoxidation of partial
oxidation products.
Catalyst Ethane con- Selectivity (%)
STY
STY
C H
2
4
version (%)
Ethylene CO
CO
2
CT-1
CT-2Si
CT-3Al
CT-4aAl 29.5
CT-5Nb 18.8
CT-6Zr
35.6
18.0
23.9
93.6
95.4
93.6
95.0
93.2
93.6
4.0
2.8
4.0
3.1
4.3
4.4
2.5 245
1.8 126
2.4 165
1.9 206
2.5 129
245
158
207
258
161
123
14.2
1.9
97.9
C-7
5.5
3.5
7.8
2.7
73.5
65.9
65.3
63.2
19.8
24.9
25.4
6.8
9.2
9.3
29.7
37.1
21.3
46.9
15.8
C-8Si
C-9Al
C-10aAl
17.0
37.5
12.6
To investigate the influence of the presence of diluter in the
oxidation of reaction intermediates, we also studied the oxi-
dation of propylene over bulk and diluted catalysts. Table 3
summarizes some of the catalytic results obtained for propylene
25.7 11.1
a
◦
Catalytic results obtained at a reaction temperature = 400 C, a C /O /He
2
2
−
1
molar ratio of 30/10/60; and a contact time, W/F , of 40 g hmol
.
cat
C
2
b
c
−1 −1
h .
cat
◦
Space time yield, STY
, in g
kg
oxidation at 380 C. As shown, in all of the catalysts studied,
C2H4
C2H4
ꢀ
STY
is the rate of ethylene formation per unit mass of active compo-
the selectivity to acrylic acid remained very high, although the
undiluted sample had the highest selectivity to acrylic acid. It
also can be seen that the selectivity to acrylic acid during the
oxidation of propylene follows the same order as that observed
in the oxidation of propane (bulk > α-Al2O3- ꢁ SiO2- >
γ -Al2O3- > Nb2O5- > ZrO2-containing catalyst), although the
differences are considerably lower in the case of propylene ox-
idation.
C H
2
4
−
1
−1
nents (in gC H h kg
).
2
4
MoVTeNb
3
.2.2. Ethane oxidation
Table 4 summarizes the catalytic results of bulk and diluted
MoV(Te)NbO catalysts obtained during the oxidation of ethane
◦
at 400 C. Ethylene, carbon monoxide, and carbon dioxide were
observed only as reaction products, with a carbon balance of
1
00 ± 2% for all catalytic tests. In this case, all of the cat-
alysts exhibited relatively high initial selectivity to ethylene,
although Te-containing catalysts were more active and selec-
tive in the oxidative dehydrogenation of ethane to ethylene
compared with the corresponding Te-free Mo–V–Nb catalysts.
In addition, and although all Te-containing catalysts presented
similar selectivities to ethylene at low ethane conversions, the
catalysts exhibited important differences in catalytic activity for
ethane oxidation.
Fig. 7. Variation of the ethane conversion with the reaction temperature at a
−1
C
2
contact time, W/F , of 40 g hmol obtained during the ethane oxidation on
cat
undiluted and diluted Mo–V–Nb–Te catalysts. Symbols: (2) CT-1; (!) CT-2Si;
(Q) CT-3Al; (e) CT-4aAl; (+) CT-5Nb; (×) CT-6Zr.
V–Nb–Te catalysts. It can be concluded that the catalytic ac-
tivity for ethane conversion decreased in the following order:
undiluted > α-Al2O3- > γ -Al2O3- > SiO2-, Nb2O5- > ZrO2-
containing catalyst.
Fig. 7 presents the variation of ethane conversion with the
reaction temperature obtained over undiluted and diluted Mo–

278
B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
Fig. 8. Variation of the selectivity to ethylene with the ethane conversion in the
◦
3
80–400 C temperature range obtained during the ethane oxidation on undi-
luted and diluted Mo–V–Nb or Mo–V–Nb–Te catalysts. Symbols: (1) CT-1;
(
!) CT-2Si; (P) CT-3Al; (e) CT-4aAl; (+) CT-5Nb; (×) CT-6Zr; (2) C-7;
(") C-8Si; (Q) C-9Al; (a) C-10aAl.
Fig. 8 shows the evolution of the selectivity to ethylene with
the ethane conversion in the reaction temperature range of 380–
◦
4
00 C. Minimal differences in the selectivity to ethylene can be
observed at low ethane conversions for Mo–V–Nb–Te catalysts
with selectivities to ethylene of ca. 95% at ethane conversions
(
of 10%), whereas Te-free catalysts demonstrate significantly
lower selectivity to ethylene compared with the corresponding
Te-containing catalysts.
Fig. 9. Variation of the propane conversion (a), the rate of acrylic acid formation
ꢀ
per unit mass of active components, STY
the rate of ethylene formation per unit mass of active components, STY
(b), the ethane conversion (c), and
AA
Small differences in the selectivity to ethylene also can
be seen at high ethane conversions over Te-containing cata-
lysts, although both the undiluted and α-Al2O3-diluted cata-
lysts present the highest selectivity to ethylene. Moreover, it
is evident that overoxidation of ethylene is very important for
Te-free catalysts, suggesting that some of the crystalline phases
present in these catalysts favor the overoxidation of ethylene.
Based on the results given in Table 4, we can conclude that
both the space time yield (STYC H ) and the rate of formation
ꢀ
C H
2
4
(d) with the relative presence of the M1 phase (I
/I
ratio determined by
2
7.2 22.0
XRD) in undiluted and diluted Mo–V–Nb–Te catalyst. The catalytic results for
propane and ethane oxidation from Tables 2 and 4, respectively.
quence of the formation of different crystalline phases (which
is favoured by the presence of diluter).
But we observed a different catalytic behavior during ethane
oxidation on our diluted catalysts. The catalytic activity de-
creased following the same trend as for propane oxidation, but
little differences in the selectivity to ethylene (at isoconver-
sion conditions) were found among these catalysts (Fig. 8). For
this reason, if the changes observed in the nature of crystalline
phases in these catalysts were responsible for their catalytic
behavior, then the influence of these crystalline phases on the
selectivity to acrylic acid for propane oxidation should be dif-
ferent than that on the selectivity to ethylene for ethane oxida-
tion.
2
4
ꢀ
of ethylene per unit mass of active components (STY
) were
C2H4
higher for the undiluted catalyst than for most of the diluted cat-
alysts. Only in the α-Al2O3-containing catalysts was it possible
to obtain a catalytic performance similar to that of the undiluted
catalyst.
3
.3. General remarks
The catalytic results of diluted Mo–V–Nb–Te catalysts dur-
ing the propane oxidation indicate that they are less active and
selective in the formation of acrylic acid than the undiluted
catalyst, although this effect depends on the diluent metal ox-
ide incorporated. In this way, the selectivity to acrylic acid for
propane oxidation decreased in the following order: bulk >
α-Al2O3- ꢁ SiO2- > γ -Al2O3- > Nb2O5- > ZrO2-containing
catalyst.
In agreement with the results presented here, Holmberg
et al. [22], studying the ammoxidation of propane and propy-
lene, reported that the dilution of Mo–V–Nb–Te mixed oxides
with γ -Al2O3, SiO2, or TiO2 favored a drastic decrease in both
catalytic activity and selectivity to acrylonitrile, as a conse-
According to our characterization results, the formation of
crystalline phases other than M1 and M2 depends strongly
on the metal oxide diluter. Fig. 9 shows the variation of the
conversion of propane (Fig. 9a) and ethane (Fig. 9c) with the
I
27.2 22.0
ratio obtained from the XRD patterns. For compar-
/I
ison, the figure also compares the variation of the rate of for-
mation of acrylic acid (Fig. 9c) and ethylene (Fig. 9d) per unit
ꢀ
ꢀ
mass of active components (i.e., STY
and STY
in Ta-
AA
C2H4
bles 2 and 4, respectively) with the I27.2/I22.0 ratios obtained
for undiluted and diluted catalysts (Table 1). It can be seen
that the catalytic activity in the oxidation of both ethane and

B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
279
propane, as well as the formation of partial oxidation products
i.e., acrylic acid or ethylene), increased with the amount of
linked to the different selectivities achieved [49]. Thus, a higher
oxidation state of the surface (as occurs in propane oxidation)
probably favors deep oxidation of POPs, resulting in lower se-
lectivity to acrylic acid than that observed during the oxidation
of propylene.
(
M1 phase in the diluted Mo–V–Nb–Te catalysts. These find-
ings confirm that M1 phase is the key factor in the selective
oxidative activation of ethane and propane [15–23], but they do
not completely explain the differences observed in the selectiv-
ity to acrylic acid (Fig. 6a) and to ethylene (Fig. 8) over diluted
Mo–V–Nb–Te catalyst.
Comparison of the catalytic results over Te-free catalysts
demonstrated no acrylic acid from propane in any of these
catalysts (Table 2), whereas relatively high selectivity to eth-
ylene was obtained at low ethane conversions during ethane
oxidation (Table 4). These results suggest that the presence of
crystalline phases other than M1/M2 in diluted catalysts [i.e.,
M5O14 (M = Mo, V, Nb), TeMo5O16, and supported MoOx
species] has different effects on the overoxidation of the cor-
responding partial oxidation products.
Finally, the varying predominance of M1 phase in bulk and
diluted Mo–V–Nb–Te catalysts can be tentatively explained by
the partial interaction between the metal oxide diluter and the
transition metal cations (especially polymolybdate species) in-
volved in formation of the M1 phase during the evaporation
step, changing the effective composition during the synthesis
of diluted catalysts. This is in agreement with the formation
of Al2(MoO4)3 or ZrMo2O8 observed in γ -Al2O3- or ZrO2-
containing catalysts, which supports the interaction between
Mo species and some of the metal oxides used as diluters, lead-
ing to modifications in both the nature of crystalline phases and
the catalytic performance. For this reason, a metal oxide di-
luter with low surface area and/or low capacity is needed to
interact with vanadia and molybdena, as α-Al2O3, to main-
tain the concentration of transition metal cations during syn-
thesis.
SiO2 and α-Al2O3 are known to be less reactive than γ -
Al2O3, ZrO2, or Nb2O5 with molybdenum species in solution.
However, the formation of M2 is also favored in SiO2- or
α-Al2O3-diluted catalysts. This suggests that other factors be-
sides the effective concentration of Mo species (e.g., pH of the
final solution, partial interaction of M1 and diluter during the
calcination step) must play a role in the formation of M1/M2 in
diluted catalysts.
Mo–V–Nb mixed-metal oxides are active but relatively uns-
elective for the ODH of propane [34–36], although acetic acid
also can be observed in M5O14 (M = Mo, V, Nb)-containing
catalysts [35]. Moreover, Mo–V–Nb mixed-metal oxides are
relatively selective in the ODH of ethane [34,37], but M5O14
(
M = Mo, V and/or Nb) demonstrated poor activity in both
ethane and ethylene oxidation under the reaction conditions
used in the present study (reaction temperature below 425 C)
◦
[
38]. Consequently, it has been suggested that the presence
of tellurium in Mo–V-containing catalysts not only facilitates
the formation of M1 phase, but also inhibits the presence of
MoO2 and some not desirable nonstoichiometric mixed oxides
[
15,35], which also are unselective in oxidation reactions [37].
On the other hand, supported MoOx catalysts are active in
the ODH of propane to propylene [39,40] but show low selec-
tivity to POPs during the oxidation of propylene [41]. However,
supported MoOx [42] catalysts exhibit very poor activity in
both ethane and ethylene oxidation. Finally, TeMo5O16 is active
and relatively selective in the propylene oxidation to acrolein
4. Conclusion
The incorporation of a metal oxide, such as SiO2, γ -Al2O3,
α-Al2O3, Nb2O5, or ZrO2, into a MoVTeNbO mixed-metal ox-
ide can improve the physical and mechanical properties of these
types of catalysts by modifying both the nature of crystalline
phases and the catalytic performance in oxidation reactions.
According to our results, a parallelism between the presence
of the M1 phase and the catalytic performance for ethane and
propane conversion can be proposed, confirming that this is the
active phase in the selective oxidative activation of short-chain
alkanes.
Nevertheless, a strong influence of the presence of crys-
talline phases other than M1 and M2 on the selectivity to acrylic
acid has been observed during propane oxidation over diluted
catalysts. Moreover, the presence of these crystalline phases
seems to have only a slight influence on the selectivity to eth-
ylene during the oxidation of ethane over the same catalysts.
This could be related to the higher reactivity of POPs observed
in propane oxidation (i.e., propylene, acrolein) compared with
that of ethylene on the crystalline phases (Mo5O14, TeMo5O16,
or supported MoOx species) detected in diluted catalysts. In
this way, α-Al2O3 seems to be the most appropriate diluter for
achieving good catalytic performance in the partial oxidation of
both ethane and propane.
[
43] but is inactive in ethylene oxidation. Therefore, the pres-
ence of M5O14 (M = Mo, V, Nb), or supported MoOx species
in Te-containing catalysts favors the overoxidation of propylene
and decreased formation of acrylic acid during the oxidation of
propane, and exert only slight influence on the overoxidation of
ethylene.
Nevertheless, we must note that the selectivity to acrylic acid
from propane on diluted catalysts was lower than that from
propylene, and that for the latter, only small differences were
observed for the different catalysts. This finding is in agree-
ment with previous observations during propane ammoxidation
over diluted Mo–V–Nb–Te catalysts [22] and in the oxidation
of propane/propylene over similar mixed-metal oxide catalysts
[
44–47]. The higher selectivity to acrylic acid from the oxida-
tion of olefin compared with that from the corresponding alkane
likely can be related to the differences in the oxidation state of
the surface of the catalysts depending on the substrate used,
as has been suggested for the oxidation of butenes/n-butane
[
48] or pentenes/n-pentane [1] over VPO catalysts. It is known
that the catalyst surface is more reduced during the oxidation
of olefins than during the oxidation of alkanes, which must be

280
B. Solsona et al. / Journal of Catalysis 252 (2007) 271–280
Acknowledgments
[23] J.M. Oliver, J.M. López Nieto, P. Botella, A. Mifsud, Appl. Catal. A
Gen. 257 (2004) 67.
[
24] A. Corma, V. Fornés, M.I. Juan-Rajadell, J.M. López Nieto, Appl. Catal.
Financial support was provided by the DGICYT of Spain
project CTQ2006-09358/BQU). B.S. thanks the MEC (Ramón
y Cajal Programme).
A 116 (1994) 151.
(
[25] I.E. Wachs, J.-M. Jehng, W. Ueda, J. Phys. Chem. B 109 (2005) 2275.
[26] H. Knözinger, H. Jezlorowski, J. Phys. Chem. 82 (1978) 2002.
[27] H. Hu, I.E. Wachs, S.R. Bare, J. Phys. Chem. 99 (1995) 10897.
[28] G. Mestl, T.K.K. Srinavasan, Catal. Rev.-Sci. Eng. 40 (1998) 451.
[29] Y.S. Jin, A. Auroux, J.C. Vedrine, J. Chem. Soc. Faraday Trans. 1 85
References
(
1989) 4179.
[
[
[
[
[
[
[
1] F. Cavani, F. Trifiró, Catal. Today 51 (1999) 561.
2] H.H. Kung, M.C. Kung, Appl. Catal. A 157 (1997) 105.
3] T. Blasco, J.M. López Nieto, Appl. Catal. A 157 (1997) 117.
4] M. Bañares, Catal. Today 51 (1999) 319.
5] H.X. Dai, C.T. Au, Curr. Top. Catal. 3 (2002) 33.
6] J.M. López Nieto, Top. Catal. 41 (2006) 3.
[
[
30] Z. Liu, Y. Chen, J. Catal. 177 (1998) 314.
31] M. Roussel, M. Bouchard, E. Bordes-Richard, K. Karim, S. Al-Sayari,
Catal. Today 99 (2005) 77.
[
[
[
32] G. Mestl, J. Raman Spectrosc. 33 (2002) 333.
33] M. Dieterle, G. Mestl, Phys. Chem. Chem. Phys. 4 (2002) 822.
34] A. Adesina, N.W. Cant, A. Saberi-Moghaddam, C.H.L. Szeto, D.L. Trim,
J. Chem. Technol. Biotechnol. 72 (1998) 19.
7] R.K. Grasselli, J.D. Burrington, D.J. Buttrey, P. DeSanto, C.G. Lugmair,
A.F. Volpe, T. Weingand, Top. Catal. 23 (2003) 5.
[
35] P. Concepción, P. Botella, J.M. López Nieto, Appl. Catal. A 278 (2004)
[8] T. Ushikubo, Catal. Today 78 (2003) 43.
4
5.
[9] D. Vitry, J.L. Dubois, W. Ueda, J. Mol. Catal. A Chem. 220 (2004) 67.
[
[
36] Z. Zhao, X. Gao, I.E. Wachs, J. Phys. Chem. B 107 (2003) 6333.
37] M. Mezourki, B. Taouk, L. Tessier, E. Bordes, P. Courtine, Stud. Surf. Sci.
Catal. 75 (1993) 753.
38] J.M. López Nieto, P. Botella, P. Concepción, A. Dejoz, M.I. Vázquez,
Catal. Today 91–92 (2004) 247.
39] K. Chen, S. Xie, A.T. Bell, E. Iglesia, J. Catal. 198 (2001) 232, and refer-
ences therein.
40] E. Heracleous, M. Machli, A.A. Lemonidou, I.A. Vasalos, J. Mol. Catal.
A 232 (2005) 29.
[
[
10] M.M. Lin, Appl. Catal. A 207 (2001) 1.
11] T. Ushikubo, H. Nakamura, Y. Koyasu, S. Wajiki, US Patent 5,380,933
(1995); EP 0,608,838 B1 (1997).
[
[
[
[
[
[
[
[
[
[
[
[
12] T. Ushikubo, K. Oshima, A. Kayo, T. Umezawa, K. Kiyono, I. Sawaki,
EP529853 A2 (1993); assigned to Mitsubishi.
13] T. Ushikubo, K. Oshima, A. Kayou, M. Hatano, Stud. Surf. Sci. Catal. 112
(
1997) 473.
14] J.M. López Nieto, P. Botella, M.I. Vázquez, A. Dejoz, Chem. Commun.
2002) 1906.
15] P. Botella, E. García-González, A. Dejoz, J.M. López Nieto, M.I. Vázquez,
J. González-Calbet, J. Catal. 225 (2004) 428.
16] J.M.M. Millet, H. Roussel, A. Pigamo, J.L. Dubois, J.C. Jumas, Appl.
Catal. A Gen. 232 (2002) 77.
17] J.M.M. Millet, M. Baca, A. Pigamo, D. Vitry, W. Ueda, J.L. Dubois, Appl.
Catal. A Gen. 244 (2003) 359.
18] P. DeSanto, D.J. Buttrey, R.K. Grasselli, C.G. Lugmair, A.F. Volpe, B.H.
Toby, T. Vogt, Z. Kristallogr. 219 (2004) 152.
19] H. Tsuji, K. Oshima, Y. Koyasu, Chem. Mater. 15 (2003) 2112.
20] P. Botella, E. García-González, J.M. López Nieto, J.M. González-Calbet,
Solid State Sci. 7 (2005) 507.
(
41] C. Martin, I. Martin, C. Mendizábal, V. Rives, Stud. Surf. Sci. Catal. 75
(1993) 1987.
42] A. Chistodoulakis, E. Heracleous, A.A. Lemonidou, S. Boghosian, J. Ca-
tal. 242 (2006) 16.
43] P. Botella, J.M. López Nieto, B. Solsona, J. Mol. Catal. A 184 (2002) 335.
44] M. Lin, T.B. Desai, F.W. Kaiser, P.D. Klugherz, Catal. Today 61 (2000)
[
[
2
23.
[
45] P. Botella, J.M. López Nieto, B. Solsona, A. Mifsud, F. Márquez, J. Ca-
tal. 209 (2002) 445.
46] E.K. Novakova, J.C. Vedrine, E.G. Derouane, J. Catal. 211 (2002) 235.
47] W. Ueda, D. Vitry, T. Katou, Catal. Today 96 (2004) 235.
[
[
[
[
[21] H. Murayama, D. Vitry, W. Ueda, G. Fuchs, M. Anne, J.L. Dubois, Appl.
Catal. A 318 (2007) 137.
[48] G. Centi, F. Trifiró, J.R. Ebner, V.M. Franchetti, Chem. Rev. 88 (1988) 55,
and references therein.
[22] J. Holmberg, R. Häggblad, A. Andersson, J. Catal. 243 (2006) 350.
[49] G. Centi, G. Fornasari, F. Trifió, J. Catal. 89 (1984) 44.