Synthesis and characterization of MoVTeCeO catalysts and their catalytic performance for selective oxidation of isobutane and isobutylene

Article abstract of DOI:10.1016/j.jcat.2007.07.028

A series of Mo-based catalysts (MoVCeO, MoTeCeO, MoVTeO, and MoVTeCeO mixed metal oxides) were prepared and tested for the oxidation of isobutane and isobutylene. Characterization results (XRD, FTIR, TPR, XPS, and BET) showed that their structure and properties depend on the composition of the calcined samples. Catalytic tests showed that molybdenum may be the key element for the activation of isobutane, whereas the selective oxidation of isobutylene to methacrolein (MAL) might proceed mainly on the surface of the TeMo-containing crystalline phase. In addition, Te⁴⁺ may favor the formation of methacrylic acid (MAA). The incorporation of Ce ions may modify the redox process of the catalysts and thus influence the catalytic behaviors for the selective oxidation of isobutane. Thus, the selectivities and yields of MAL and MAA can be increased over these specially designed catalysts.

Full text of DOI:10.1016/j.jcat.2007.07.028

Journal of Catalysis 251 (2007) 354–362
www.elsevier.com/locate/jcat
Synthesis and characterization of MoVTeCeO catalysts and their catalytic
performance for selective oxidation of isobutane and isobutylene
Jingqi Guan, Shujie Wu, Hongsu Wang, Shubo Jing, Guojia Wang, Kaiji Zhen, Qiubin Kan ∗
College of Chemistry, Jilin University, Changchun 130023, PR China
Received 21 June 2007; revised 28 July 2007; accepted 31 July 2007
Available online 18 September 2007
Abstract
A series of Mo-based catalysts (MoVCeO, MoTeCeO, MoVTeO, and MoVTeCeO mixed metal oxides) were prepared and tested for the
oxidation of isobutane and isobutylene. Characterization results (XRD, FTIR, TPR, XPS, and BET) showed that their structure and properties
depend on the composition of the calcined samples. Catalytic tests showed that molybdenum may be the key element for the activation of isobutane,
whereas the selective oxidation of isobutylene to methacrolein (MAL) might proceed mainly on the surface of the TeMo-containing crystalline
4+
phase. In addition, Te may favor the formation of methacrylic acid (MAA). The incorporation of Ce ions may modify the redox process of the
catalysts and thus influence the catalytic behaviors for the selective oxidation of isobutane. Thus, the selectivities and yields of MAL and MAA
can be increased over these specially designed catalysts.
© 2007 Elsevier Inc. All rights reserved.
Keywords: Isobutane; Isobutylene; Selective oxidative; Methacrolein; Methacrylic acid
1. Introduction
tion of alkane [16]. In addition, another important aspect of this
reaction is that alkane-rich conditions have been widely used
as feed materials [17]. However, under these conditions, the
isobutane conversion is inevitably very low (in general, ꢀ25%);
therefore, it is necessary to recycle the unconverted reactants.
In addition, the mixed metal oxides are very promising cat-
alysts for the selective oxidation of isobutane to isobutylene
and MAL [6–12]. Among these, Ce–Zr mixed oxides have been
used for the selective oxidation of isobutane to isobutylene, and
their catalytic performance has been reported to be relatively
good [18]. The oxidative dehydrogenation reaction is improved
due to the redox properties of CeO2 and its ability to store and
release oxygen under oxidation and reduction conditions, re-
spectively.
But the most effective mixed-oxide catalysts for isobutane
oxidation to MAL are Mo–V–Te–Sb–O catalysts, as reported
by Guan et al. [11]. It has been suggested that in the Mo–V–
Te–Sb–O system, the pseudohexagonal so-called “M1 phase”
(Te0.33MO3; M = Mo, V, and Sb) might favor selectivity to
MAL. These authors also have reported that adding a small
amount of antimony to the MoVTeP-based system produces a
relative increase in the selectivity to MAL but decreased selec-
tivity to COx (CO and CO2) [12]. The incorporation of Sb can
The one-step selective oxidation of isobutane to methacro-
lein (MAL) and methacrylic acid (MAA) is an attractive alter-
native to the conventional multistep catalytic oxidation reaction
of acetone with hydrogen cyanide due to its simplicity, inexpen-
sive raw materials, and negligible environmental impact. Thus
reaction has been extensively studied for a long time worldwide
[1–5]. Many catalytic systems have been investigated for the se-
lective catalytic oxidation of isobutane [1–12].
Up to now, the most promising catalysts for the selective
oxidation of isobutane to MAA are heteropolyacids and their
polyoxometalate derivatives with Keggin structures [13–15]. It
has been suggested that Mo-containing species should be the
active sites [14], whereas the addition of some transition metals
(e.g., V, Cu, Fe) can enhance the catalytic performance by pro-
moting catalyst reoxidation. However, a problem encountered is
the low thermal stability of the Keggin-type polyoxometalates,
which begin to decompose at the temperature for the activa-
*
Corresponding author.
E-mail address: qkan@mail.jlu.edu.cn (Q. Kan).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.07.028

J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
355
modify the active and selective sites for the transformation of
isobutane to isobutylene and MAL, and slow the consecutive
total combustion reactions of MAL to COx.
In this paper, we report on the introduction of Ce ions, which
have a strong capability to transport oxygen to the MoVTe-
based catalysts to facilitate the migration of lattice oxygen in
the lattice. TPR results have shown that the reducibility of cat-
alysts has been modified. Catalytic data indicate that incorpo-
rating a small amount of Ce increases the selectivities to MAL
and MAA.
diluted with 0.5 g of SiC particles to prevent temperature gradi-
ents and hot spots in the reactor. Under our reaction conditions,
the homogeneous reaction can be neglected. Carbon balance
was always >95%.
The feed was controlled by a mass flow controller, and water
was fed by a mini-pump. The catalytic reaction condition was
as follows: molar ratio of the feed gas i-C4H10:O2:N2:H2O
= 1:2.5:2:2, i-C4H8:O2:N2:H2O = 1:6:6:6. The experiments
were carried out at 420 ◦C. Gaseous products were analyzed by
an online gas chromatograph with a GDX-501 column and a
flame ionization detector (FID; Shimadzu GC-8A). Other prod-
ucts were analyzed by a gas chromatograph with a GDX-501
column containing an 8% polyethylene glycol sebacate/2% se-
bacic acid and an FID (Shimadzu GC-8A). Isobutylene (i-C=4 ),
methacrolein (MAL), methacrylic acid (MAA), COx (CO,
CO2), propylene (C=), and acetic acid (HAC) were the main
products. Small am3ounts of propane, acetone, acrolein, and
acrylic acid also were detected.
2. Experimental
2.1. Catalyst preparation
The MoVTeCeO catalysts were prepared as follows. First,
ammonium metavanadate was dissolved in water at 80 ◦C under
stirring, and then ammonium heptamolybdate and telluric acid
were added into the metavanadate solution. Half an hour later,
a different amount of cerous nitrate was added. The mixture was
further stirred at 80 ◦C to remove volatiles. The resultant solid
was dried at 110 ◦C overnight, and finally calcined in a flow
of nitrogen at 600 ◦C for 2 h to obtain the final metal oxides.
For comparison, MoVO, MoVTeO, MoVCeO, and MoTeCeO
catalysts were also prepared under similar conditions.
3. Results and discussion
3.1. XRD studies
XRD patterns of the prepared catalysts are shown in Fig. 1.
For the Te-free sample MoV0.3Ce0.2, the peaks at 2θ =
22.1◦, 23.7◦, 26.2◦, 29.7◦, and 31.7◦ can be assigned to
(V0.07Mo0.93)5O14 [JCPDS 31-1437], whereas peaks at 2θ =
14.7◦, 22.6◦, 24.6◦, 29.1◦, 32.9◦, 39.1◦, and 39.7◦ can be at-
tributed to Ce2(MoO4)2(Mo2O7) [JCPDS 76-1040]. Moreover,
peaks at 2θ = 28.5◦, 33.1◦, 47.5◦, and 59.1◦ can be assigned
to CeO2 [JCPDS 81-0792], and peaks at 2θ = 12.8◦, 23.7◦,
25.9◦, 27.3◦, 33.7◦, 39.1◦, 39.7◦, and 49.3◦ can be attributed
to MoO3 [JCPDS 35-0609]. In contrast, the XRD pattern of
the V-free sample MoTe0.23Ce0.2 indicates the presence of
MoO3, Ce2(MoO4)2(Mo2O7), and TeMo5O16 (2θ = 21.8◦,
24.6◦, 26.6◦, 27.3◦, 30.5◦, 31.0◦, and 34.4◦) [JCPDS 88-0407].
2.2. Catalyst characterization
The specific surface areas of the catalysts were measured
based on the adsorption isotherms of N2 at −196 ◦C using the
BET method (Micromeritics ASAP2010). Powder X-ray dif-
fraction (XRD) patterns were collected using a Shimadzu XRD-
6000 scanning at 4◦/min with CuKα radiation (40 kV, 30 mA).
The infrared spectra (IR) of various samples were recorded at
room temperature using a NICOLET Impact 410 spectrometer.
X-ray photoelectron spectra (XPS) were recorded on a VG
ESCA LAB MK-II X-ray electron spectrometer using AlKα
radiation (1486.6 eV, 10.1 kV). The spectra were referenced
with respect to the C 1s line at 284.7 eV. The measurement
error of the spectra was 0.2 eV.
Meanwhile, the XRD pattern of the Ce-free sample MoV0.3
-
Te0.23 indicates the presence of MoO3, (V0.07Mo0.93)5O14, Te-
H2-temperature programmed reduction (TPR) experiments
were carried out in a flow reactor system, in which 10 mg of cat-
alyst was charged each run into a U-shaped quartz microreactor
(4 mm i.d.). After purging with Ar gas from 50 to 300 ◦C at a
ramp rate of 10 ◦C/min, holding at 300 ◦C for 30 min, and cool-
ing to 100 ◦C, the sample was reduced in a 5% H2/Ar stream
(25 ml/min). The reduction temperature was raised uniformly
from 100 to 800 ◦C at a ramp of 10 ◦C/min. H2 consumption
was measured by a thermal conductivity detector (TCD).
2.3. Catalytic tests
The reaction was performed in a stainless steel tubular fixed
bed reactor (16 mm i.d., 400 mm long) under atmospheric pres-
sure. Each catalytic test was carried out using 1.0 g of catalyst,
which was granulated into particles of 20–30 mesh size and
◦
Fig. 1. XRD patterns of samples calcined at 600 C in N : (Q) MoO , (P)
2
3
CeO , (") Ce (MoO ) (Mo O ), (!) (V
Mo
) O , (2) TeMo O ,
7
2
2
4 2
2
0.07
0.93 5 14 5 16
(1) CeMo O , (a) TeMO (TeVMoO and/or TeVCeMoO) crystalline phase.
2
8

356
J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
Mo5O16, and a new TeMO (M = Mo and V) crystalline phase
(2θ = 22.1◦, 28.3◦, 36.4◦, 44.6◦, and 50.2◦), which could cor-
respond to the hexagonal phase M1 (Te0.33MO3; M = Mo, V,
and Nb), as proposed by Aouine et al. [19–22]. When cerium
was added into the MoVTe-based system at a Ce/Mo ratio of
0.1, the peaks corresponding to TeMo5O16 disappeared and
the intensities of the peaks corresponding to MoO3 increased.
However, a further increase of cerium content to a Ce/Mo ra-
tio of 0.2 caused the appearance of peaks at 2θ = 22.6◦, 25.9◦,
27.5◦, 30.3◦, and 36.4◦, suggesting the presence of CeMo2O8
[JCPDS 32-0194], and the appearance of peaks at 2θ = 22.6◦,
23.4◦, 25.9◦, 27.5◦, 30.3◦, and 36.4◦, which can be attributed
to Ce2(MoO4)2(Mo2O7) [JCPDS 76-1040]. However, the peaks
corresponding to (V0.07Mo0.93)5O14 disappeared. When cerium
content was further increased to a Ce/Mo ratio of 0.3, a new
phase CeO2 could be observed, whereas the peaks correspond-
ing to MoO3 disappeared.
In addition, the XRD patterns of MoV0.1Te0.23Ce0.2 (Fig. 1e)
were relatively different from those of other MoVTeCeO sam-
ples, such as MoV0.3Te0.23Ce0.2 (Fig. 1c), because the peak
at 2θ = 28.5◦ corresponding to TeMO crystalline phase was
shifted to 28.2◦ and its intensity decreased remarkably. This is
because the decreased V content in the system may relatively
reduce the content of the TeMO phase. Botella et al. also re-
ported similar results, demonstrating that the new TeMO phase
increased with increasing V content [20].
Similarly, for the V-free catalyst MoTe0.23Ce0.2 (Fig. 2f), it
can be seen that the bands at 993, 870, and 820 cm−1 are associ-
ated with the presence of MoO3, whereas the bands at 852 cm−1
are probably due to antisymmetric vibrations of Mo–O–Me (Me
= Mo and Ce) bridging bonds. Moreover, the bands at 923,
820, 699, 646, and 585 cm−1 with a shoulder at 765 cm−1 can
be attributed to TeMo5O16 [20]. However, Ce species could be
partially incorporated into the TeMo5O16 structure and change
the positions of the Mo–O bands (bands at 913, 800, 725,
and 500 cm−1) and Te–O bands (bands at 765 and 630 cm−1
[20].
)
In the same way, the FTIR spectrum of Ce-free sample
MoV0.3Te0.23 (Fig. 2a) shows a band at 993 cm−1, which
should be attributed to symmetric stretching vibration of
Mo=O bonds in MoO3, whereas the band detected at 845 cm−1
is due to antisymmetric vibrations of Mo–O–Me (Me = Mo
and V) bridging bonds. Furthermore, the bands at 923, 724, and
585 cm−1 should be related to TeMo5O16 phase. All of these
bands may have shifted from their original positions due to the
incorporation of Ce atoms into the system. The incorporated
Ce may have interacted with Mo atoms and thereby formed
new Mo–O–Ce bands, which in return affected the growth
of other Mo–O species. As a result, when cerium was added
into the MoVTe-based system, the intensity of the band at 923
cm−1 corresponding to Mo–O bond in the TeMo5O16 phase
decreased dramatically. But the intensities of the bands corre-
sponding to Mo–O–Ce increased gradually, especially of the
band at 759 cm−1. In addition, when the Ce/Mo ratio reached
0.3, the band at 993 cm−1 disappeared.
3.2. FTIR studies
Combined with the XRD results, these findings lead us to
The FTIR spectra of the prepared catalysts are shown in
Fig. 2. For the Te-free sample MoV0.3Ce0.2 (Fig. 2g), the bands
at 1012, 907, and 585 cm−1 are probably related to V=O
groups and V–O–Me (Me = Mo, V, and Ce) bonds [11,20],
whereas the bands at 993, 878, and 820 cm−1 can be attributed
to MoO3. In addition, bands at 500–900 cm−1 (e.g., 845, 791,
759, 733, 631, and 557 cm−1) are due to antisymmetric vibra-
tions of Mo–O–Me bridging bonds [23–25].
conclude that no MoO3 crystallite exists in the MoV0.3Te0.23
-
Ce0.3 catalyst. The infrared spectra along with the XRD patterns
indicate that in both cases, we have an intimate mixture of sev-
eral phases, some of which are XRD-amorphous.
3.3. Specific surface area determination and XPS studies
Table 1 shows the specific surface area and XPS data of
the catalysts. The catalysts’ specific surface areas depended on
their composition. In general, the initial incorporation of cerium
into the Mo–V–Te-based system decreased the specific surface
area, which began to increase slowly with increasing cerium
content. Moreover, the V and Te content also affected the cata-
lysts’ surface area. The catalysts’ surface composition strongly
depended on the nominated amount of each element. No vana-
dium ions were detected on the catalyst surface when relatively
low amounts of cerium were added into the MoV0.3Te0.23Cex
system. Only when relatively high content of cerium (Ce/Mo =
0.3) were added could vanadium ions be detected by XPS. In
addition, the tellurium content increased when the cerium was
initially added into the catalysts, whereas it decreased with fur-
ther increases in cerium content. However, the cerium content
on the surface of the MoV0.3Te0.23Ce0.2 catalyst was the lowest
among the samples tested.
To gain deeper insight into the surface constitution and prop-
erties of these catalysts, we investigated their binding energies
(Mo 3d5/2, Te 3d5/2, V 2p3/2, and Ce 3d5/2). The corresponding
◦
Fig. 2. FTIR spectra of samples calcined at 600 C in N .
2

J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
357
Table 1
◦
BET surface area and XPS data of samples calcined at 600 C in N
2
Sample
S
Surface
composition
Binding energy (eV)
Surface oxidation states
BET
2
−1
(m g
)
6+
5+
5+ 4+
6+
4+
4+
3+
Mo/V/Te/Ce
Mo 3d
V 2p
Te 3d
Ce 3d
5/2
Mo /Mo
V
/V
Te /Te
Ce /Ce
5/2
3/2
5/2
MoV Te
3.0
1.7
2.1
2.6
0.8
1.2
3.4
1/0.06/0.13/0
1/0/0.17/0.08
1/0/0.11/0.03
1/0.1/0.08/0.1
1/0.06/0.41/0.08 232.8
1/0/0.23/0.06
1/0.22/0/0.09
232.9
232.9
232.9
232.8
516.5
–
–
516.5
517.9
–
576.2
576.2
576.3
576.1
577.2
577.1
–
–
0.62/0.38
0.63/0.37
0.86/0.14
0.95/0.05
0.65/0.35
0.83/0.17
0.25/0.75
0/0
0/0
0.59/0.41
0.75/0.25
–
0.23/0.77
0.23/0.77
0/1
0.12/0.88
0.84/0.16
0.86/0.14
–
–
0.3 0.23
MoV Te
Ce
Ce
Ce
Ce
881.6
882.8
882.6
885.1
884.8
0.68/0.32
0.61/0.39
0.56/0.44
0/1
0.19/0.81
0.38/0.62
0.3 0.23 0.1
MoV Te
0.3 0.23 0.2
MoV Te
0.3 0.23 0.3
MoV Te
0.1 0.23 0.2
MoTe
Ce
0.23 0.2
232.8
232.8
MoV Ce
517.1
883.8–886.5 0.9/0.1
1/0
0.3 0.2
Fig. 3. X-ray photoelectron spectra corresponding to the main transitions (Mo 3d , Te 3d , V 2p , and Ce 3d ) of calcined catalysts: (a) MoV Te ,
0.3 0.23
5/2
5/2
3/2
5/2
(b) MoV Te
Ce , (c) MoV Te
Ce , (d) MoV Te
Ce , (e) MoV Te
Ce , (f) MoTe
Ce , (g) MoV Ce
.
0.3 0.23 0.1
0.3 0.23 0.2
0.3 0.23 0.3
0.1 0.23 0.2
0.23 0.2
0.3 0.2
spectra are shown in Fig. 3, they are composed and integrated
with the results shown in Table 1. The Mo 3d5/2 peak of the cat-
alysts can be fitted mainly into two components at 231.7 and
232.8 eV, which can be related to Mo5+ and Mo6+, respec-
tively [26,27]. However, the binding energy of Mo ions in a 4+
oxidation state (typical value of 230.9 eV in MoO2) is not ob-
served [28]. As shown in Fig. 3 and Table 1, the Mo6+/Mo5+
surface atomic ratios increased with increasing Ce content, sug-
gesting that the coordinatively unsaturated Mo5+ was pronely
oxidized to Mo6+ by Ce4+
.
The V 2p3/2 peak of catalysts could be fitted into two com-
ponents at 516.2 and 517.3 eV, which can be related to V4+ and
V5+ species, respectively [29]. A greater amount of V5+ was
present on the Ce-containing catalyst surface than on the Ce-
free catalyst surface. In addition, the presence of V4+ species
in Te-containing samples, which was not observed in Te-free

358
J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
sample, suggests that the presence of Te can favor the reduction
of V5+
free sample MoV0.3Te0.23 (Fig. 4g) displays three peaks at 652,
710, and 746 ◦C. According to previous work [33–35], the weak
reduction peak observed at 652 ◦C may be attributed to a step-
wise reduction of V2O5 → 1/3V6O13 and a complex reduction
reaction of MoO3 → MoO2 and/or TeO2 → Te. The peak at
.
The Te 3d5/2 peak of catalysts could be fitted into two com-
ponents at 576.2 and 577.3 eV, which can be related to Te4+ and
Te6+ species, respectively [30]. However, Te4+ cations were
present mainly on the surface of the MoVTeCeO catalysts with
relatively high V content (e.g., V/Mo = 0.3), whereas Te6+
species were seen mainly on the surface of catalysts with rela-
tively low V content (e.g., V/Mo = 0.1). However, Te0 (binding
energy of 573.0 eV) was not observed.
The Ce 3d5/2 peaks of the catalysts could be fitted into two
components at 881.8 and 884.8 eV, which can be related to
Ce4+ and Ce3+ ions, respectively [31,32]. Table 1 and Fig. 3
show that Ce3+ ions existed mainly in MoVTeCeO catalysts
with relatively low V content (e.g., V/Mo = 0.1), whereas
Ce4+ ions were present mainly in the catalysts with relatively
high V content (e.g., V/Mo = 0.3). However, Ce3+ ions ex-
isted mainly in the Te-free catalyst with relatively high V
content (e.g., V/Mo = 0.3). In addition, Ce4+/Ce3+ surface
atomic ratios decreased slightly with increasing Ce content in
the MoV0.3Te0.23Cex catalyst system, suggesting that the re-
duction of Ce4+ may have been relatively restrained by the high
V content.
710 ◦C could be related to a stepwise reduction of V6O13
→
6VO2, whereas the peak at 746 ◦C could be assigned to a step-
wise reduction of MoO2 → Mo. When cerium was added into
the catalysts at a Ce/Mo ratio of 0.1 or 0.2, for only one peak
each appeared for Mo, V, and/or Te in the catalyst. Moreover,
the value of Tmax for the main peak shifted to around 682 ◦C.
Compared with the previous reports [33,34] and our earlier ex-
perimental results [35], this peak may not be attributed to the
reduction of single Mo, V, Te, or Ce ions (Figs. 4a and 4b). The
reduced species may be a crystalline phase (e.g., M1 phase).
Consequently, this indicates that in the presence of cerium,
there are synergetic effects among the Mo, V, Te, and Ce ions
in the reduction procedure. A further increase in cerium con-
tent to a Ce/Mo ratio of 0.3 caused the appearance of a peak at
around 751 ◦C, suggesting the reduction of MoO2 → Mo.
For the low V-containing or V-free samples (Figs. 4d and 4e),
a first TPR peak appeared at about 601 ◦C, considered to be due
to the reduction of the surface-capping oxygen of ceria. A sim-
ilar result was reported by Yasyerli et al. [36] in a study on H2S
oxidation over a Ce–V catalyst. However, with increasing V
content from a V/Mo ratio of 0.1 to 0.3, this peak disappeared
and a second peak at around 692 ◦C shifted to 684 ◦C for the
MoV0.3Te0.23Ce0.1 sample. These results indicate that the rela-
tive vanadium content in the catalysts can properly modify the
catalysts’ redox capability.
3.4. H2-TPR studies
The H2-temperature programmed reduction (H2-TPR) pro-
files and amount of H2 consumption of MoV0.3Ce0.2, MoTe0.23
-
Ce0.2, MoV0.1Te0.23Ce0.2, and MoV0.3Te0.23Cex (x = 0.1−0.3)
are shown in Fig. 4 and Table 2. The H2-TPR profile of the Ce-
To estimate the H2 consumption, the area of TPR profile was
integrated (Table 2). The H2 consumption was found to increase
in the sequence MoV0.3Te0.23Ce0.2 < MoV0.3Te0.23Ce0.1
MoV0.1Te0.23Ce0.2 < MoTe0.23Ce0.2 < MoV0.3Te0.23Ce0.3
<
<
MoV0.3Ce0.2 < MoV0.3Te0.23. Combining these findings with
the XPS data (Table 1) demonstrates that the Mo6+/V5+/Ce4+
content on the surface of the MoVTeCeO catalysts varies
widely; thus, determining which element contributes most to
the reduction is difficult. In other words, the results demon-
strate that the reducibility of Mo–V–Te–Ce–O catalysts may not
be determined by single element in a high-oxidation state (e.g.,
Mo6+, V5+, or Ce4+) on the catalyst surface. Such behavior
seems to be related to changes in the nature of the crystalline
phases as well as in the oxidation state of each element.
3.5. Catalytic properties
The catalytic results obtained for the oxidation of isobutane
over the catalysts at 420 ◦C, given in Table 3, indicate that all
◦
Fig. 4. The H -TPR profiles of samples calcined at 600 C in N .
2
2
Table 2
◦
The H -TPR results of samples calcined at 600 C in N
2
2
Sample
MoV Te
MoV Ce
MoTe
Ce
MoV Te
Ce
MoV Te
Ce
MoV Te
Ce
MoV Te Ce
0.3 0.23 0.3
0.3 0.23
0.3 0.2
0.23
0.2
0.1 0.23 0.2
0.3 0.23 0.1
0.3 0.23 0.2
◦
Temperature ( C)
Weight (mg)
652/710/746
10
672/700/739
10
601/692/751
10
601/690/751
10
684
10
687
10
687/751
10
H
consumption (µmol) 13.4/34.5/26.2 25.1/12.3/10.7 5.6/22.8/0.4
3.6/20.0/4.1
26.6
23.3
33.8/6.1
2

J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
359
Table 3
Catalytic data of the catalysts for isobutane oxidation at 420 C
◦
a
Sample
Conversion (%)
Selectivity (%)
=
i-C
4
=
b
i-C H
MAL
MAA
CO
C
HAC
Others
x
4
10
3
MoV Te
18.8
19.5
20.2
22.7
21.6
23.4
24.5
1
14
10
5
15
17
2
15
24
33
20
21
20
4
11
15
20
10
7
58
34
28
51
44
43
73
6
6
4
4
7
7
5
8
4
2
8
4
5
6
1
3
3
2
2
2
1
0.3 0.23
MoV Te
Ce
Ce
Ce
Ce
0.3 0.23 0.1
MoV Te
0.3 0.23 0.2
MoV Te
0.3 0.23 0.3
MoV Te
0.1 0.23 0.2
MoTe
Ce
0.23 0.2
6
9
MoV Ce
0.3 0.2
a
−1
−1
cat
Reaction condition: GHSV = 3600 ml h
g
, P = 101 kPa.
b
Acetone, acrolein, acrylic acid, etc.
Fig. 5. Variation in the selectivities to the main products of the partial oxidation with the isobutylene conversion obtained during the oxidation of isobutylene at
◦
420 C over (a) the MoV Te
catalyst, (b) the MoTe
Ce catalyst, (c) the MoV Ce catalyst, (d) the MoV Te
Ce catalyst. Symbols: (Q) MAL,
0.3 0.23
0.23 0.2
0.3 0.2
0.3 0.23 0.2
(a) MAA, (") CO , (!) MAL + MAA.
x
of the Mo–Te-containing catalysts were active and selective for
the partial oxidation of isobutane to MAL. Moreover, the se-
lectivity to MAL and MAA was improved considerably by the
addition of small amounts of cerium. For example, the selec-
tivity to MAL and MAA reached 24 and 15%, respectively, at
Ce/Mo ratios as low as 0.1. The MAL and MAA selectivities in-
creased continuously with increasing cerium content, reaching
a maximum (33% MAL selectivity and 20% MAA selectivity)
at a Ce/Mo ratio of 0.2. However, the MAL and MAA selectiv-
ities decreased with further increases in cerium content.
In addition, clearly the highest selectivity to isobutylene and
the lowest selectivity to MAA could be obtained over V-free

360
J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
◦
Fig. 6. Variation in the selectivities to the main products of the partial oxidation with the isobutane conversion obtained during the oxidation of isobutane at 420 C
over (a) the MoV Te catalyst, (b) the MoTe Ce catalyst, (c) the MoV Ce catalyst, (d) the MoV Te Ce catalyst. Symbols: (2) isobutylene,
0.3 0.23
0.23 0.2
0.3 0.2
0.3 0.23 0.2
(Q) MAL, (a) MAA, (") CO , (1) isobutylene + MAL + MAA.
x
catalyst MoTe0.23Ce0.2. Thus, it can be concluded that the in-
corporation of vanadium into the Mo–Te–Ce-based system had
a positive effect on the transformation of isobutylene to MAA
in the selective oxidation of isobutane.
ucts have nonzero intercepts, whereas secondary and high-order
products have zero intercepts [37]. Therefore, in the case of
selective oxidation of isobutylene over MoV0.3Te0.23Cex cata-
lysts (Fig. 5), clearly the dominant primary product was MAL,
because MAL selectivity increased gradually with decreasing
isobutylene conversion.
In contrast, the catalytic results given in Table 3 show that
the selectivities to MAA over MoTe0.23Ce0.2 and MoV0.1Te0.23
-
Ce0.2 catalysts, in which Te ions are mainly in a +6 oxidation
state on the catalyst surface (Table 1), were lower than those
obtained over the Te4+-containing MoV0.3Te0.23Ce0.2 catalyst.
It can be suggested that the presence of Te4+ ions in the cat-
alysts favors the formation of MAA. In this way, these results
demonstrate that the V–Te4+-containing phase (e.g., M1 phase)
was active and selective in the selective oxidation of isobutane
to MAA.
Fig. 5 shows the variation in selectivities to the main prod-
ucts of the reaction with the isobutylene conversion obtained
during the oxidation of isobutylene over the four Mo-based cat-
alysts. It is well known that the primary products of a given re-
action can be discriminated from high-order products by extrap-
olating product selectivity to zero conversion. Primary prod-
In the V-free catalyst MoTe0.23Ce0.2 (Fig. 5d), MAL selec-
tivity was strongly affected by the reaction gas hourly space
velocity. Only when the gas hourly space velocity remained at
an optimum value could maximal MAL selectivity and yield be
obtained. Thus, it can be concluded that the incorporation of a
small amount of V may have a positive effect on the adsorption
and activation of isobutylene, which in return improves the se-
lectivity to MAL. However, over the V-containing catalysts, the
selectivity to MAA increased and the selectivity to MAL de-
creased with increasing isobutylene conversion, indicating that
MAA may be formed by a consecutive reaction of MAL. More-
over, the addition of Ce into the MoV0.3Te0.23-based catalyst
enhanced MAL selectivity at relatively high isobutylene con-
version. Considering that the Mo6+/Mo5+ surface atomic ratio

J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
361
In analogy with the oxidation of propane to acrylic acid
[38–42], we propose the following reaction pathway for the oxi-
dation of isobutane (Fig. 7). From isobutane, we can envision at
least three possible reaction pathways leading to various partial
oxidation products as described in Fig. 7. In path a, iso-butanol
is formed by insertion of one oxygen atom into the methyl
C–H bond, which dehydrates to isobutylene or is oxidized to
MAL and MAA through iso-butyric aldehyde and iso-butyric
acid, respectively. Path b is a direct oxidative dehydrogenation
from isobutane to isobutylene, followed by oxidation to MAA
through MAL. In path c, tert-butanol is formed by insertion of
one oxygen atom into the methine C–H bond, which can dehy-
drate to isobutylene or further oxidize to acetone. As a matter
of fact, in the oxidation of isobutane, iso-butanol is obtained in
substantial quantities, especially in the case of reaction at rela-
tively low temperature (360–380 ◦C). This result suggests that
path a is very possible. However, we are unable to determine
whether pathway c or an iso-butanol intermediate truly exist.
Selective oxidation reactions of alkanes on mixed-oxide cat-
alysts often show similar reaction kinetics, in which the acti-
vation of alkanes is generally a rate-determining step. In the
heteropolyacid system, Mo is the active site for the selective
oxidation of isobutane to MAA [14]. According to our results
given in Table 3, Figs. 5 and 6, the V-free catalysts also showed
relatively high activity for the reaction compared with the V-
containing catalysts. In particular, the relatively low Ce-content
V-containing catalysts exhibited good catalytic performance,
although no V species were detected on the catalyst surface.
Therefore, Mo appears to be a key element in isobutane activa-
tion. In other words, the active crystalline phase for the transfor-
mation of isobutane to isobutylene should be a Mo-containing
phase or a mixture of several phases. However, the effect of V
on the isobutane activation cannot be excluded.
Fig. 7. Reaction scheme for the oxidation of isobutane on MoVTeCeO catalysts.
in the MoV0.3Te0.23Ce0.2 catalyst was higher than that in the
MoV0.3Te0.23 catalyst, it has been suggested that Mo6+ species
should be involved in the selective O insertion of olefinic inter-
mediate [21,26].
Fig. 6 shows the variation in selectivities to the main reac-
tion products with the isobutane conversion obtained during
the oxidation of isobutane over the four Mo-based catalysts.
The product distributions were relatively different, depending
strongly on the chemical composition of the catalysts. In gen-
eral, CO and CO2 were the main products. Isobutylene, MAL,
and MAA also were main products over the MoVTeCeO cat-
alyst. Of the four catalysts studied, the V-free MoTe0.23Ce0.2
catalyst (Fig. 6b) exhibited the highest selectivity to isobutylene
and the lowest selectivity to MAA. But when vanadium was
added into the Mo–Te-based system (Fig. 6a), the selectivity
to isobutylene decreased significantly, whereas the selectivity
to MAA increased only relatively. These findings suggest that
part of the isobutylene was transformed to MAA over a V-
containing phase. Moreover, the incorporation of Ce into the
Mo–V–Te-based system also improved the selectivity to MAA
in the selective oxidation of isobutane, although it had almost
no effect on the selective oxidation of isobutylene. This indi-
cates that the incorporation of Ce can have a positive effect on
both activity and selectivity.
It is generally accepted that the presence of TeMo-containing
crystalline phase may be a key factor in the selective oxidation
of propene to acrolein over MoVTe-based catalysts. Our ex-
perimental results lead to a similar conclusion, that isobutylene
oxidation to MAL occurs over a TeMo-containing phase. It has
been reported that in the selective oxidation of propane over
Mo–V–Sb–Nb–O catalyst, the presence of (MxMo1−x)5O14
(M = V or Nb) favors the formation of acrylic and acetic acid
[42,43]. This proposal is not supported by our experimental
results, however. As mentioned above, we found that the Te4+
-
containing phase (e.g., M1 phase) appears to be a determining
factor for the selective oxidation of isobutane to MAA.
According to our experimental results, the role of V is re-
lated to the formation of active and/or selective crystalline
phases, such as M1, which has been reported to be active for
the propene oxidation to acrolein and acrylic acid [20,21] and
was found to be active for isobutylene oxidation to MAL in the
present study. In addition, the formation of the redox cycle V4+
+ Mo6+ ꢁ V5+ + Mo5+ may improve the redox capability of
Mo species.
It has been reported that Ce has a strong capability for stor-
ing oxygen, which can improve the mobility of lattice oxygen
in the catalysts. Moreover, Ce can join the redox cycle Ce3+
+ Mo6+ ꢁ Ce4+ + Mo5+ [44]. In addition, the role of Ce is
In addition, it can be concluded that the selectivity to
MAA in the selective oxidation of isobutane over MoV0.3Te0.23
(Fig. 6a) and MoV0.3Te0.23Ce0.2 (Fig. 6d) catalysts is gener-
ally higher than that in the selective oxidation of isobutylene
(Figs. 5a and 5d). Based on the foregoing discussion, we can
propose that MAA was formed not only from MAL, but also
directly from isobutane or an intermediate. A reaction scheme
for the selective oxidation of isobutane is given in Fig. 7.

362
J. Guan et al. / Journal of Catalysis 251 (2007) 354–362
related to the formation of active TeMO crystalline phases. It
should be noted that the addition of small amounts of Ce to the
MoVTe-based catalyst modified this catalyst’s reduction to im-
prove the title reaction. Due to these factors, we conclude that
the MoVTe-based catalysts with an optimum Ce content may
exhibit relatively good activity and selectivity for the selective
oxidation of isobutane. In summary, the presence of all four
elements—Mo, V, Te, and Ce—is necessary to achieve high se-
lectivities in MAL and MAA.
[12] J. Guan, S. Wu, M. Jia, J. Huang, S. Jing, H. Xu, Z. Wang, W. Zhu,
H. Xing, H. Wang, Q. Kan, Catal. Commun. 8 (2007) 1219.
[13] C. Marchal-Roch, N. Laronze, N. Guillou, A. Tézé, G. Hervé, Appl. Catal.
A 199 (2000) 33.
[14] G.P. Schindler, T. Ui, K. Nagai, Appl. Catal. A 206 (2001) 183.
[15] S. Gao, J.B. Moffat, Appl. Catal. A 229 (2002) 245.
[16] F. Cavani, F. Trifirò, Appl. Catal. A 88 (1992) 115.
[17] E. Etienne, F. Cavani, R. Mezzogori, F. Trifirò, G. Calestani, L. Gengem-
bre, M. Guelton, Appl. Catal. A 256 (2003) 275.
[18] C. Leitenburg, A. Trovarelli, J. Llorca, F. Cavani, G. Bini, Appl. Catal.
A 139 (1996) 161.
[19] M. Aouine, J.L. Dubois, J.M.M. Millet, Chem. Commun. (2001) 1180.
[20] P. Botella, J.M. López Nieto, B. Solsona, A. Mifsud, F. Márquez,
J. Catal. 209 (2002) 445.
[21] R.K. Grasselli, J.D. Burrington, D.J. Buttrey, P. DeSanto Jr., C.G. Lug-
mair, A.F. Volpe Jr., T. Weingand, Top. Catal. 23 (2003) 5.
[22] T. Ushikubo, K. Oshima, A. Kayou, M. Hatano, Stud. Surf. Sci. Catal. 112
(1997) 473.
[23] I.E. Wachs, Catal. Today 27 (1996) 437.
[24] W.X. Kuang, Y.N. Fan, K.W. Yao, Y. Che, J. Solid State Chem. 140 (1998)
354.
4. Conclusion
The addition of Ce to a MoVTe-based system modified its
physicochemical properties (e.g., crystalline phases, reducibil-
ity, surface composition, and surroundings of Mo), affecting
the system’s catalytic performance for the selective oxidation
of isobutane to MAL and MAA. Under our reaction condi-
tions, the MoV0.3Te0.23Ce0.2 catalyst achieved the best MAL
and MAA selectivity (53%) and yield (10.7%) at 420 ◦C.
[25] J.C.J. Bart, F. Cariati, A. Sgamelloti, Inorg. Chim. Acta 36 (1979) 105.
[26] P. Botella, P. Concepción, J.M. López Nieto, Y. Moreno, Catal. Today 99
(2005) 51.
Acknowledgments
[27] Y. Zhu, W. Lu, H. Li, H. Wan, J. Catal. 246 (2007) 382.
[28] J.W. Robinson, Handbook of Spectroscopy, vol. 1, CRC Press, Boca Ra-
ton, FL, 1974, p. 615.
[29] M. Demeter, M. Newman, W. Reichelt, Surf. Sci. 454–456 (2000) 41.
[30] H. Hayashi, N. Shigemoto, S. Sugiyama, N. Masaoka, K. Saitoh, Catal.
Lett. 19 (1993) 273.
This work was supported by the National Basic Research
Program of China (2004CB217804) and the National Natural
Science Foundation of China (20673046).
References
[31] A. Laachir, V. Perrichon, A. Badri, J. Chem. Soc. Faraday Trans. 87 (1991)
1601.
[1] M. Misono, N. Nojiri, Appl. Catal. 64 (1990) 1.
[2] R.H. Schwaar, Methacrylic Acid and Esters, Process Economics Program
Report No. 11D, SRI International, 1993.
[3] W. Li, W. Ueda, Catal. Lett. 46 (1997) 261.
[4] J.S. Min, N. Mizuno, Catal. Today 66 (2001) 47.
[5] J.H. Holles, C.J. Dillon, J.A. Labinger, M.E. Davis, J. Catal. 218 (2003)
42.
[6] J.S. Paul, J. Urschey, P.A. Jacobs, W.F. Maier, F. Verpoort, J. Catal. 220
(2003) 136.
[7] T. Shishido, A. Inoue, T. Konishi, I. Matsuura, K. Takehira, Catal. Lett. 68
(2000) 215.
[8] H. Liu, E.M. Gaigneaux, H. Imoto, T. Shido, Y. Iwasawa, Appl. Catal.
A 202 (2000) 251.
[9] T. Inoue, K. Asakura, W. Li, S.T. Oyama, Y. Iwasawa, Appl. Catal. A 165
(1997) 183.
[10] V. Cortés Corberán, M. Jia, J. El-Haskouri, R.X. Valenzuela, D. Beltrán-
Porter, P. Amorós, Catal. Today 91–92 (2004) 127.
[11] J. Guan, M. Jia, S. Jing, Z. Wang, L. Xing, H. Xu, Q. Kan, Catal. Lett. 108
(2006) 125.
[32] F.L. Normand, L. Hilaire, K. Kili, J. Phys. Chem. 92 (1988) 2561.
[33] I.L. Botto, C.I. Cabello, H.J. Thomas, Mater. Chem. Phys. 47 (1997) 37.
[34] L. Chen, J. Liang, H. Lin, W. Weng, H. Wan, J.C. Védrine, Appl. Catal.
A 293 (2005) 49.
[35] J. Guan, G. Wang, S. Jing, Z. Wang, Q. Kan, Polish J. Chem. 80 (2006)
1397.
[36] S. Yasyerli, G. Dogu, T. Dogu, Catal. Today 117 (2006) 271.
[37] J.N. Michaels, D.L. Stern, R.K. Grasselli, Catal. Lett. 42 (1996) 135.
[38] M. Ai, Catal. Today 13 (1992) 679.
[39] M.M. Bettahar, G. Costentin, L. Savary, J.C. Lavalley, Appl. Catal. A
Gen. 145 (1996) 1.
[40] M. Lin, T.B. Desai, F.W. Kaiser, P.D. Klugherz, Catal. Today 61 (2000)
223.
[41] M.M. Lin, Appl. Catal. A Gen. 207 (2001) 1.
[42] J.C. Védrine, E.K. Novakova, E.G. Derouane, Catal. Today 81 (2003) 247.
[43] E.K. Novakova, E.G. Derouane, J.C. Védrine, Catal. Lett. 83 (2002) 177.
[44] X. Zhang, H. Wan, W. Weng, X. Yi, J. Mol. Catal. A Chem. 200 (2003)
291.