Performance of WOx-added Mo-V-Te-Nb-O catalysts in the partial oxidation of propane to acrylic acid

Article abstract of DOI:10.1016/j.apcata.2010.02.002

The performance of Mo_1.0V_0.3Te_0.23Nb_0.23O_x catalyst in the partial oxidation of propane to acrylic acid was improved when the catalyst was modified with an optimal amount of tungsten oxide (WO_x), which was present largely in an amorphous phase. The added WO_x increased the catalyst activity by promoting the dehydration of the oxygenate intermediates in the oxidation process, which eventually accelerated the production of propylene. An increase in the dehydration rates on the W-added catalysts was verified by separate experiments using 1- and 2-propanol as the reactants. The selectivity for the production of acrylic acid was also improved for the W-added catalysts because the dehydration step was promoted to a greater extent than the dehydrogenation step and the number of acidic sites in the catalysts was decreased. The reaction results obtained using the W-added catalysts can be explained based on the reaction paths involved in the propane-oxidation process and on the surface properties of the catalysts analyzed using X-ray diffraction (XRD), infrared (IR) spectroscopy, and the temperature-programmed desorption of ammonia (NH₃-TPD).

Full text of DOI:10.1016/j.apcata.2010.02.002

Applied Catalysis A: General 378 (2010) 76–82
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Performance of WO -added Mo–V–Te–Nb–O catalysts in the partial oxidation of
x
propane to acrylic acid
Bu Young Jo, Sang Seop Kum, Sang Heup Moon *
School of Chemical & Biological Engineering and Institute of Chemical Processes, Seoul National University, Seoul 151-744, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Article history:
The performance of Mo1.0V0.3Te0.23Nb0.23O catalyst in the partial oxidation of propane to acrylic acid
x
Received 5 October 2009
Received in revised form 20 January 2010
Accepted 1 February 2010
Available online 6 February 2010
was improved when the catalyst was modified with an optimal amount of tungsten oxide (WO
x
), which
was present largely in an amorphous phase. The added WO increased the catalyst activity by promoting
x
the dehydration of the oxygenate intermediates in the oxidation process, which eventually accelerated
the production of propylene. An increase in the dehydration rates on the W-added catalysts was verified
by separate experiments using 1- and 2-propanol as the reactants. The selectivity for the production of
acrylic acid was also improved for the W-added catalysts because the dehydration step was promoted to
a greater extent than the dehydrogenation step and the number of acidic sites in the catalysts was
decreased. The reaction results obtained using the W-added catalysts can be explained based on the
reaction paths involved in the propane-oxidation process and on the surface properties of the catalysts
analyzed using X-ray diffraction (XRD), infrared (IR) spectroscopy, and the temperature-programmed
Keywords:
Propane
Oxidation
Acrylic acid
Tungsten oxide
Dehydration
Acidity
desorption of ammonia (NH -TPD).
3
ß 2010 Elsevier B.V. All rights reserved.
1
. Introduction
Acrylic acid, which is a raw material for synthesizing acrylate
were very active for propylene oxidation [6–8], they were not so
active for propane oxidation [9,10].
The most effective catalysts for propane oxidation to acrylic
acid were Mo–V–Te–Nb–O types reported by Ushikubo et al. [3]
and Lin and Linsen [11], who observed significantly higher
activities of the above catalysts compared with the cases of other
ones. The catalyst performance could be further improved by
substituting Te of Mo–V–Te–Nb–O with Sb [12]. A similar study
was made independently by Takahashi et al. [13]. However, the
yields of acrylic acid obtained using these catalysts were reported
to remain below 48% [1], which makes the propane-oxidation
process still less competitive with the current process based on
propylene oxidation.
Because the rate-determining step (RDS) of propane oxidation
is the formation of propylene from propane, one of strategies to
improve the catalyst is to promote it with a proper component that
is expected to accelerate the conversion of propane to propylene.
Propylene is obtained as an intermediate in propane oxidation
either by a single step, the oxidative dehydrogenation of propane,
or by two steps that involve the initial oxidation of propane to 1- or
2-propanol and the subsequent dehydration of propanol to
propylene [12,14,15].
polymers for use as plastics, adhesives and paints, is commercially
produced by the partial oxidation of propylene [1,2]. However, an
alternative process is under investigation to produce acrylic acid
directly from propane, which is cheaper than propylene. A key to
the success of the new process is a catalyst that effectively
activates the relatively stable propane and allows the production of
acrylic acid at high yields. A few catalysts have been proposed for
the purpose, but they still need further improvement prior to
commercialization [1,3,4].
The application of multi-component metal oxide (MMO)
catalysts for the oxidation of propane to acrylic acid started in
the 1990s, although Mo–V–Nb mixed oxide that formed the basis
of the catalysts was proposed for the oxidation of ethane to ethene
and acetic acid in the 1970s [5]. The Mo–V–Nb oxide catalyst also
activated propane at 300 8C but the products were only acetic acid,
acetaldehyde and carbon oxides. Also studied was Mo–V–Te oxide
catalyst, which achieved high propane conversions but again
produced no acrylic acid. Although mixed oxide catalysts lacking V
According to a previous report, VO
amorphous tungsten oxide (WO ) shows an improved activity for
the oxidative dehydrogenation of propane to propylene [16]. WO
is reported to be an active catalyst for the dehydration of alcohols
x 2 3
/Al O promoted with
x
*
Corresponding author. Tel.: +82 2 880 7409; fax: +82 2 875 6697.
x
E-mail address: shmoon@surf.snu.ac.kr (S.H. Moon).
0926-860X/$ – see front matter ß 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.02.002

B.Y. Jo et al. / Applied Catalysis A: General 378 (2010) 76–82
77
such as methanol and 2-propanol [17,18]. In this study, we
attempted to increase the activity of Mo–V–Te–Nb oxide, a known
catalyst for the partial oxidation of propane to acrylic acid [19–21],
separation of hydrocarbons, and
a
60/80 CARBOXEN 1000
(Supelco) column for the separation of nitrogen, oxygen, CO and
CO . The conversion of propane was calculated based on carbon
2
by promoting it with different amounts of WO
x
.
balance and the selectivity for acrylic acid from the amounts of
produced acrylic acid compared with those of converted propane.
The prepared catalysts were tested for propane oxidation and
also for the oxidation of 1- and 2-propanol. The latter compounds
were used as the model reactants because they were obtained as
the intermediates of propane oxidation. The reaction results were
correlated with the surface properties of the catalysts, which were
analyzed by X-ray diffraction (XRD), infrared spectroscopy (IR),
2.2.2. 1- or 2-propanol oxidation
The oxidation of 1- or 2-propanol was carried out at 200 or
175 8C using a reactant stream containing 1-propanol (or 2-
propanol), oxygen, nitrogen, and water in the molar fractions of
0.08, 0.21, 0.33, and 0.38, respectively. The reason for carrying out
propanol oxidation at lower temperatures than in the case of
propane oxidation was because propanol oxidation proceeded very
rapidly at 350–400 8C such that differences in the rates among
sample catalysts could not be observed. The amounts of the
catalyst and the space velocity of the reactant stream were the
same as for propane oxidation.
and the temperature-programmed desorption of ammonia (NH
TPD).
3
-
2
. Experimental
.1. Catalyst preparation
A model catalyst with the nominal composition of Mo/V/Te/
2
Nb = 1.0/0.3/0.23/0.23, based on the amounts of the individual
components that were added to the preparation solution, was
prepared by a co-precipitation method [14,19–21]. The precursors
of metal oxides were hexaammonium heptamolybdate tetrahy-
2.3. Catalyst characterization
2.3.1. XRD and ICP-AES
The catalysts were analyzed by X-ray diffraction (XRD: MAC
Science Co., M18XHF-SRA). The crystalline phases of the catalyst
components were identified based on the XRD-peak assignments
made in a previous study [22]. The amounts of metal species in the
catalyst were estimated by inductively coupled plasma-atomic
emission spectroscopy (ICP-AES: Perkin-Elmer (USA), Optima
4300DV).
drate ((NH
telluric acid (H
NNbO
ÁxH
For catalyst preparation, 1.18 g of the Mo precursor, 0.23 g of
the V precursor and 0.35 g of the Te precursor were diluted in
4
)
6
Mo
7
O
24.4H
TeO ), and ammonium niobate oxalate hydrate
2
O).
2 4 3
O), ammonium metavanadate (NH VO ),
2
6
(C
4
H
4
9
3
5
0 cm of distilled water maintained at 35 8C. Another solution was
3
prepared by dissolving 0.24 g of the Nb precursor in 4 cm of
distilled water. Two solutions were mixed and stirred at 300 rpm
in a round flask at 35 8C for 3 h. Nitric acid was added to the mixed
solution to control the pH to 3.1 and the resulting solution was
stirred for another 1 h. The solution was dried in a rotary
evaporator at 70 mbar and 60 8C for 50 min, and additionally
dried in an air-circulating oven at 120 8C overnight. Dried catalyst
was ground in a pulverizer (Retsch, MM 301) shaking at 20 Hz for
3
2.3.2. FT-IR and NH -TPD
The infrared (IR) spectra of the catalysts were recorded on an
FT-IR (Midac, model M2000) instrument, which was equipped
with an MCT detector and a diffuse reflectance accessory (Pike
Technologies Inc., DiffusIR Accessory). As the catalyst powders
were weighed in the same amount, 0.07 g, and placed in the DRS
cell of a fixed dimension, 5 mm  5 mm  5 mm, the IR peaks were
obtained with reproducible intensity. The amounts of acidic sites
on the catalyst surface were estimated from the results of the
2
4
min. The catalyst was heated to 200 8C in air at a ramping rate of
8C/min and finally calcined in N at 600 8C for 2 h, prior to
2
impregnation with WO
The precursor of W, ammonium tungstate pentahydrate
(NH O), was added to the model catalyst in
different amounts, 0.0005–0.4 wt% based on W metal, by an
impregnation method. The prepared catalysts were dried at 80 8C
for 1 h and subsequently at 120 8C overnight in an air-circulating
oven before they were ground into powders. The powder catalyst
x
.
temperature-programmed desorption of ammonia (NH
obtained using a catalyst analyzer (Bel-Cat, Bel Japan Inc.). For
the NH -TPD, the catalysts were pre-treated in He at 110 8C for 1 h
and then cooled to 20 8C, prior to adsorption with NH at
atmospheric pressure. After purging the reactor with He at room
temperature for 1 h, NH -TPD peaks were recorded while the
temperature was raised from 20 to 650 8C at a rate of 10 8C/min.
The effluent gas was analyzed for NH with a TCD.
3
-TPD)
(
4
)
10
W
12
O
41Á5H
2
3
3
3
was calcined in N
2
while the temperature was raised from 20 to
3
5
00 8C at a rate of 5 8C/min and finally maintained at 500 8C for 2 h.
2
.3.3. O
The oxygen mobilities of the catalysts were evaluated based on
the results of the temperature-programmed desorption of oxygen
(O -TPD) obtained using a TCD. For the O2-TPD, the catalysts were
pre-treated in argon at 150 8C for 1 h and then cooled to 20 8C prior
to the desorption of O at atmospheric pressure. The same amount
of the sample catalysts, 0.1 g, was used for the experiments. O
2
-TPD
2
2
.2. Reaction
.2.1. Propane oxidation
2
Reaction tests were made in a stainless-steel tubular reactor
(ID = 4.5 mm) at temperatures between 350 and 400 8C using a
2
reactant stream containing propane, oxygen, nitrogen, and water
in the molar fractions of 0.08, 0.21, 0.33, and 0.38. The catalysts
2
-
TPD peaks were recorded while the temperature was raised from
used in the reaction were powders of 180–250
m
m in size, which
20 to 650 8C at a rate of 10 8C/min. The effluent gas was analyzed
were obtained by meshing. Diluent of the catalyst was not used.
The reactor contained 0.1 g of the catalyst, and the space velocity of
the reactant stream was 1000 h . A gas line connecting the reactor
2
for O with a TCD.
À1
3. Results
with a gas chromatograph was maintained at 180 8C to prevent the
products from possible condensation or polymerization during
flow in the line. The gas chromatograph (GC) was equipped with a
flame ionization detector (FID) for analyzing hydrocarbons and a
thermal conductivity detector (TCD) for analyzing nitrogen,
oxygen, and carbon oxides. Two parallel capillary columns, CP-
3.1. Composition and crystallinity of prepared catalysts
The ICP-AES results indicated that the model catalyst, which
was prepared with a nominal composition of Mo/V/Te/Nb = 1.0/
0.3/0.23/0.23, contained individual components with atomic ratios
of Mo/V/Te/Nb = 1.0/0.26/0.21/0.24, which were not affected by
2 3
Al O /KCl (Varian) and SOLGEL WAX (SGE), were used for the

7
8
B.Y. Jo et al. / Applied Catalysis A: General 378 (2010) 76–82
Table 1
The W contents and surface areas of the sample catalysts.
3
a
3
b
c
2
W loading (wt%)
W/Mo (Â10 )
W/Mo (Â10 )
Surface area (m /g)
0
0
0
0
0
0
.01
.02
.05
.1
0.3
0.5
0.3
0.7
1.1
2.3
4.6
9.2
5.3
5.1
5.3
5.2
5.2
5.3
1.3
2.6
.2
5.1
.4
10.6
a
b
c
The nominal atomic ratios of W to Mo used in the preparation step.
The atomic ratios of W to Mo measured by ICP-AES.
2
Surface area of model catalyst: 5.2 m /g.
the W addition. The atomic ratios of W to Mo in the prepared
catalysts were lower than the nominal ratios used in the
preparation step, but the two sets of values changed in parallel
with each other, as shown in Table 1. The W/Mo ratios for the
catalysts containing W in the amounts lower than 0.005 wt% were
not obtained because the W amounts were below the detection
limit of ICP-AES. According to the XRD patterns shown in Fig. 1, the
model catalyst contained the crystallite phases of Mo5Àx(V/
Fig. 2. Performance of sample catalysts in propane oxidation. Reaction condition:
catalyst = 0.1 g, temperature = 350–400 8C, propane:oxygen:nitrogen:water =
0.08:0.21:0.33:0.38 (molar fractions). *The nominal atomic ratios of W to Mo
Nb)
x
O14 (2
u
= 22.1, 23.38, 25.28, 29.78, 31.58, 32.48 and 33.58),
used in the preparation step.
TeM
3
O
10 (2
u
= 22.18, 28.28, 36.28, 45.28 and 50.08)
,
and Te
2
M
20
O
57
ꢀ
(2u
= 22.18, 26.28, 26.88, 27.38, 29.28, and 35.4 ), where M in the
equations represented either Mo, V or Nb [22–24].
The XRD result remained the same and the peaks representing
3
the crystalline phase of WO were not observed in the W-added
catalysts, probably because the amounts of added W were small,
with W/Mo ratios less than 0.01.
3.2. Propane oxidation
Fig. 2 shows changes in the conversion of propane obtained at
different reaction temperatures with the amounts of added W,
which are presented in a log scale to cover the wide range of the W
amounts. The conversions were enhanced by the W addition and
À4
showed a maximum when the W/Mo ratio was between 2.6 Â 10
À4
and 5.1 Â 10 for all reaction temperatures.
The selectivity for acrylic acid decreased with an increase in the
conversion, as shown in Fig. 3a, because acrylic acid was obtained
as an intermediate in the consecutive oxidation of propane that
eventually led to the products of higher oxidation. When W was
added to the catalyst, the selectivity–conversion curves moved
upward, indicating an increase in the selectivity at the same
conversion. Similar to the conversions shown in Fig. 2, the
selectivity was at the maximum when the W loading was in the
Fig. 3. (a) Changes in the selectivity for the production of acrylic acid in propane
oxidation. The nominal W/Mo ratios used in the preparation step: (&) 0, (&)
À5
À5
À4
À4
À4
1
.3 Â 10 , (*) 2.6 Â 10 , (^) 1.3 Â 10 , (") 2.6 Â 10 , (b) 5.1 Â 10 , (*)
À3 À3 À3 À2
1
.3 Â 10 , (^) 2.6 Â 10 , (~) 5.1 Â 10 and (!) 1.1 Â 10 . (b) The acrylic-acid
selectivity obtained at a fixed 45% conversion versus the W/Mo ratio.
proper range, as demonstrated in Fig. 3b in the case of a 45%
conversion. The selectivity increased with an initial increase in the
À4
W/Mo ratio up to about 1.3 Â 10 and remained constant until it
Fig. 1. Crystallite phases of the model catalyst: (*) Mo5Àx(V/Nb)
and (&) Te 57 (M = Mo, V, Nb).
x 1 3 10
O14, (*) Te M O
À3
2
M
20
O
started to drop at ratios higher than 1.3 Â 10
.

B.Y. Jo et al. / Applied Catalysis A: General 378 (2010) 76–82
79
Table 2
Product distribution in propane oxidation at a fixed conversion, 45%.
3
W/Mo (Â10 )
Products (%) (propane conversion = 45%)
Acrylic
acid
Propylene
Acetic
acid
Acetone
CO
CO
2
Model catalyst
60.5
63.8
64.2
67.7
67.6
67.8
67.9
65.5
64.2
64.6
4.9
5.3
5.2
6.5
6.9
6.7
6.4
6.1
5.6
5.5
4.2
4.5
4.8
5.3
5.1
5.2
5.2
4.8
4.3
4.6
0.34
0.21
0.13
0.19
0.16
0.12
0.14
0.20
0.17
0.25
16.3
13.9
14.0
10.8
11.4
10.9
11.3
12.9
14.3
13.8
13.8
12.3
11.7
9.5
0
0
0
0
0
1
2
5
1
.013
.026
.13
.26
.51
.3
8.8
9.3
9.1
.6
10.5
11.4
11.2
.1
0.6
The product distribution in propane oxidation is provided in
Table 2. The fractional amounts of propylene and acetic acid
change with the W/Mo ratio, showing a similar trend as in the case
Fig. 5. Performance of sample catalysts in 2-propanol oxidation: (&) conversion of
2-propanol, (*) selectivity for the production of propylene and (b) selectivity for
the production of acetone.
2
of acrylic acid, while the amounts of CO and CO show an opposite
trend. The amounts of acetone are insignificant compared with the
other products.
À4
5
.1 Â 10 and leveled off at higher W loading. Similar to the case
3.3. Oxidation of 1- or 2-propanol
of the conversion, the selectivities for the production of propylene
and acetone also increased with an increase in the W/Mo ratios up
À4
As mentioned above, propylene is produced either by a single
to 5.1 Â 10
and leveled off, or slightly decreased, at higher
step, the oxidative dehydrogenation of propane, or by two steps
that consist of the initial oxidation of propane to propanols and the
subsequent dehydration of the propanols. In this study, we tested
the W-added catalysts for the conversion of propanols.
Fig. 4 shows the conversions and the amounts of three major
products obtained in 1-propanol oxidation carried out at 200 8C
using the W-added catalysts. According to previous studies
loading. On the other hand, the selectivity for propionic acid
showed a trend nearly opposite that of propylene. The above
results indicate that the enhanced catalytic activities of the W-
added catalysts for 1-propanol oxidation were obtained largely
because the dehydration step, instead of the dehydrogenation step,
was promoted by the W addition.
The added W species also promoted the catalytic activity for the
oxidation of 2-propanol, as shown in Fig. 5. Among two major
products of the process, propylene was produced in increased
[
12,14,15], 1-propanol is converted via two routes: the dehydro-
genation of 1-propanol to produce propionaldehyde, which is
further oxidized to propionic acid, and the dehydration of 1-
propanol to produce propylene, which is further oxidized to
acetone, acetic acid, acrolein, and acrylic acid [12,14,15]. In the
present study, propionic acid, propylene, and acetone were
obtained as the major products. Propionaldehyde and carbon
oxides were also produced but their amounts were small, changing
in less than 1% for different catalysts. The reason why neither
acrolein nor acrylic acid was observed in this study was because
the reaction temperature, 200 8C, was too low to allow further
oxidation of propylene to the products.
amounts nearly in parallel with the amounts of added WO
x
up to
À4
the W/Mo ratios of 2.6 Â 10 and reached a saturation level at
higher W loading. On the other hand, acetone was produced
showing a trend nearly opposite to that of propylene. In 2-propanol
oxidation, propylene is produced by the dehydration of 2-
propanol, similar to the case of 1-propanol oxidation, but acetone
is produced by the direct dehydrogenation of 2-propanol, unlike
the case of 1-propanol oxidation. Carbon oxides were obtained in
small amounts similar to the case of 1-propanol oxidation.
Accordingly, the results of Fig. 5 indicate the same effect of added
Fig. 4 illustrates how the conversion of 1-propanol significantly
increased with an increase in the W/Mo ratios up to about
WO
x
as for 1-propanol oxidation: the promotion of the dehydra-
tion step instead of the dehydrogenation step.
3.4. Catalyst characterization
Fig. 6a shows the IR spectra of the catalysts containing different
À1
amounts of added W. Peaks appearing in the 900–1000 cm range
(
Region I) are assigned to the symmetric stretching vibration of
À1
Mo 55 O and those in the 700–900 cm (Region II) range to the
asymmetric vibration of Mo–O–M (M = Mo, Te, Nb) [24]. A new
À1
peak observed at 975 cm in the W-added catalysts is ascribed to
the stretching vibration of W 55 O in the amorphous phase of WO
x
À1
[22,25,26]. As the peak at 845 cm , representing the crystalline
3
phase of WO [25], was observed at low intensities for all catalysts,
it can be concluded that the added W species were present largely
as an amorphous phase instead of a crystalline one, at least when
À4
the W/Mo ratio was lower than about 2.6 Â 10 . The amounts of
À1
amorphous WO
x
, represented by a peak at 975 cm , increased and
eventually leveled off with an increase in the W loading, as shown
in Fig. 6b.
Fig. 4. Performance of sample catalysts in 1-propanol oxidation: (&) conversion of
Fig. 7 shows that the amounts of the acidic sites, which were
1
-propanol, (") selectivity for the production of propionic acid, (*) selectivity for
the production of propylene and (b) selectivity for the production of acetone.
3
estimated from the intensity of the NH -TPD peak, monotonically

8
0
B.Y. Jo et al. / Applied Catalysis A: General 378 (2010) 76–82
À4
Fig. 8. (a) O
2
-TPD from sample catalysts: (a) model catalyst, (b) W 1.3 Â 10 , (c) W
À4
À3 À3 À2
5
.1 Â 10 , (d) W 1.3 Â 10 , (e) W 2.6 Â 10 and (f) W 1.1 Â 10 . (b) Changes in
À5
Fig. 6. (a) IR spectra of the catalysts: (a) model catalyst, (b) W 2.6 Â 10 *, (c) W
the amounts of desorbed O
of catalysts.
2
, represented by TPD-peak areas, with the W/Mo ratios
À4
À3
À3
À5
2
.6 Â 10 , (d) W 1.3 Â 10 and (e) W 5.1 Â 10 . *W 2.6 Â 10 denotes the
À5
catalyst modified with W at a nominal W/Mo ratio of 2.6 Â 10 . (b) Changes in the
À1
intensity of IR peak at 975 cm with the W/Mo ratios of catalysts.
decreased with an increase in the W loading. The observed trend
was not attributed to a decrease in the catalyst surface area, which
in fact remained nearly constant in the range of the W loading used
in this study.
2
As shown in Fig. 8a, two peaks were observed in O -TPD, which
were shifted to lower temperatures with an increase in the W/Mo
À4
ratios up to 5.1 Â 10 . Similar to the trends observed with the
conversion of propane and the selectivity for the production of
acrylic acid, the oxygen mobility, which was represented by the
shift of the O
2
-TPD peaks, increased with an increase in the W/Mo
À4
ratios up to 5.1 Â 10 and decreased at higher ratios. The amounts
of the relatively weak lattice oxygen, represented by the TPD-peak
areas, increased with the W loading, as shown in the insert of
a
Fig. 8b. The activation energies (E ) for the desorption of the lattice
oxygen in the catalysts, which were estimated from the shifts of
the TPD peaks with changes in the heating rates (Fig. 9a and b),
À4
were 23 and 17 kcal/mol for the model and W 5.1 Â 10 catalysts,
respectively.
4. Discussions
4.1. Reaction pathways in propane oxidation
Fig. 10 describes reactions involved in the partial oxidation of
propane [12,14,15], which can be grouped into three paths. Path I
comprises the conversion of propane to propylene, which is further
oxidized to acrylic acid via acrolein. Path II involves the oxidation
of propane to 1-propanol, which is subsequently converted to
Fig. 7. Changes in the acidity of the catalyst with the W/Mo ratios of catalysts.

B.Y. Jo et al. / Applied Catalysis A: General 378 (2010) 76–82
81
of propylene in Paths II and III, which involve the oxidation of
propane to propanols and the subsequent dehydration of the
alcohols to propylene. An apparent difference between the two
cases is that the reaction intermediates produced in Path I remain
on the catalyst as the surface species while those in Paths II and III
are desorbed as molecular propanols. Accordingly, it can be
suggested from the above consideration that the conversion of
propane to propylene, which is the rate-determining step (RDS) of
the propane-oxidation process, can be accelerated by the promo-
tion of either or both of the two steps: the partial oxidation of
propane to oxygenates, including propanols, and the dehydration
of the oxygenates to propylene.
4.2. Oxidation of 1- or 2- propanol
The results of propanol oxidation, shown in Figs. 4 and 5,
indicated that the added W species promoted the dehydration step
to a greater extent than either the dehydrogenation or the
oxidation step and consequently increased the amounts of
produced propylene until the W/Mo ratio was about (2.6–
À4
5
.1) Â 10 . These results agree with previous reports that
amorphous WO
17,18]. The IR observation of the catalysts, shown in Fig. 6a,
verified that the added W species were present largely as
x
was active for the dehydration of alcohols
[
amorphous WO
x
when the W/Mo ratio was lower than about
À4
2.6 Â 10
. At higher W loading, the propanol conversions
remained nearly constant, or slightly decreased, because parts of
the added W species were in a crystalline phase, which was
inactive for the dehydration and simply covered the catalyst
surface.
The selectivity for acetone production increased with the W
loading in 1-propanol oxidation but decreased in 2-propanol
oxidation. In fact, the opposite trend in acetone production
between the two cases is consistent with the fact that the added W
species were more active for dehydration than for dehydrogena-
tion. In 1-propanol oxidation, acetone was produced by the
oxidation of propylene that had been obtained by the dehydration
of 1-propanol. Accordingly, the production of acetone should
change in parallel with that of propylene, which was promoted on
the W-added catalysts. On the other hand, in 2-propanol oxidation,
acetone was produced by the direct dehydrogenation of 2-
propanol, which was competing with the dehydration step for a
common reactant, 2-propanol. Accordingly, the promotion of the
latter step by the W addition would suppress the former step to
produce smaller amounts of acetone.
Fig. 9. (a) O
2
-TPD from model catalyst obtained at different heating rates. (b) O
2
-
À4
TPD from W 5.1 Â 10 catalyst obtained at different heating rates.
4.3. Activity for propane oxidation
Fig. 10. Reaction pathways in propane oxidation [12,14,15].
As discussed in Section 4.1, the catalytic activity for propane
oxidation can be increased by promoting either or both of the two
steps that constitute the RDS of the process: the partial oxidation of
propane to oxygenates and the dehydration of the oxygenates to
propylene.
Although the promotion of the initial oxidation step leading to
oxygenates by the W addition was not confirmed experimentally
in the present study, the trend can be anticipated from a previous
propylene by dehydration or to propionaldehyde and propionic
acid by further oxidation. Path III is similar to Path II except that
propane is oxidized to 2-propanol, which is either dehydrated to
propylene or further oxidized to acetone and acetic acid. Additional
steps to produce 2-propanol by propylene hydration and acetone
by propylene oxidation were also proposed. Between two paths
involving propanols, Path II is much slower than Path III and
therefore the products of Path II are obtained in smaller amounts
study that was made with WO
oxidative dehydrogenation of propane [16]. It was reported that
the added WO species, which themselves were inactive for
x 2 5 2 3
-added V O /Al O catalysts for the
[
14]. When the process is carried out at high temperatures, the
x
oxygenate products are further oxidized to carbon oxides, thus
lowering the selectivity for acrylic acid production.
propane oxidation, increased the activity by providing additional
sites for the weak adsorption of propane, which was eventually
In Path I, propylene is produced by the oxidative dehydrogena-
tion of propane, which proceeds by the initial oxidation of propane
to the oxygenate species on the catalyst surface, followed by the
dehydration of the surface oxygenates to propylene. In fact, this
mechanism is similar to that of the steps leading to the production
2 5
transferred to, and oxidized on, the V O surface.
The significant promotion of the dehydration step in the W-
added catalysts was verified in this study by the results of propanol
oxidation. Accordingly, the observed increase in the activity of the
W-added catalysts can be explained by the fact that the added W

8
2
B.Y. Jo et al. / Applied Catalysis A: General 378 (2010) 76–82
species promoted the two steps involved in the RDS of propane
oxidation.
extent than the dehydrogenation step, and because the amounts of
the acidic sites were decreased.
The activity for propane oxidation was also promoted by the
increased oxygen mobility in the catalyst [27,28], in addition to the
5. Conclusions
influence of the above-mentioned factors. The O
2
-TPD results
(
Fig. 9a) demonstrated that the activation energy for the
The WO
partial oxidation of propane compared with the unmodified
catalyst (Mo1.0 ), because the amorphous WO
promoted the steps involved in the RDS of propane oxidation: the
partial oxidation of propane to oxygenates and the dehydration of
the oxygenates to propylene. The enhanced oxygen mobility in the
W-added catalyst also contributed to propane oxidation.
x
-modified catalysts showed higher activity for the
desorption of the lattice oxygen was lower for the W-added
catalyst than for the unmodified one. This result was obtained
because the oxygen mobility increased more in the presence of the
structurally random, amorphous WO
of the crystalline phase [29].
V
0.3Te0.23Nb0.23
O
x
x
x
species than in the presence
A decrease in activity at W/Mo ratios higher than about (2.6–
À4
5
.1) Â 10 , observed in Fig. 2, is believed to be due to the coverage
x
The added WO species also improved acrylic-acid selectivity,
of the catalyst active sites, which are responsible for the conversion
of propane to propylene in Path I (known as the RDS of propane
oxidation), with the excessive amounts of the added W species.
because the species promoted the dehydration of propanols, which
were produced as the intermediates of propane oxidation, and
additionally decreased the amounts of acidic sites in the catalyst to
suppress side reactions leading to undesired oxygenate products.
4.4. Selectivity for acrylic acid
Acknowledgements
Fig. 3 indicated that the selectivity for acrylic acid changed with
the amounts of added WO
x
, showing a maximum at the W/Mo
This work was supported by Brain Korea 21 (BK 21) project,
National Research Laboratory (NRL) program, LG Chemical Co., and
the Center for Ultramicrochemical Process Systems (CUPS).
À4
À3
ratios of 1.3 Â 10 to 1.3 Â 10 , which was a trend similar to the
one observed for the conversions of propane (Fig. 2). The similar
trend in the dependence of the selectivity and the activity of the
catalysts on the W loading suggests that the two parameters were
affected by a common factor that was modified by the W addition.
It is believed that the promotion of the dehydration step on the W-
added catalysts was responsible for the characteristic changes of
the two parameters.
The propane conversions increased when the dehydration step
in the RDS was promoted, as discussed above. The selectivity was
determined by the relative rates of individual steps in Paths I and
III, instead of Path II that proceeded at low rates. The promotion of
Path I by the W addition, due to the increased rates of dehydration
in the RDS, likely contributed to the increased selectivity. The
selectivity was also affected by the relative rates of the dehydration
and the dehydrogenation of 2-propanol in Path III. Because the
added W species promoted the dehydration step to greater extents
than the dehydrogenation step, as discussed above (Section 4.2),
selectivity was improved by the W addition.
References
[
[
1] M.M. Lin, Appl. Catal. A 207 (2001) 1–16.
2] M. Ai, T. Ikawa, J. Catal. 40 (1994) 203–211.
[3] T. Ushikubo, H. Nakamura, Y. Koyasu, S. Wajiki, US Patent 5,380,933 (1995).
[
[
[
4] R.K. Grasselli, Catal. Today 49 (1999) 141–153.
5] E. Thorsteinson, T. Wilson, F. Young, P. Kasai, J. Catal. 52 (1978) 116–132.
6] T. Ohara, M. Ueshima, I. Yanagisawa, JP Patent 47-42241B (1972).
[7] N. Kurata, T. Matsumoto, T. Ohara, K. Oda, JP Patent 42-9805B (1967).
[8] I. Nagai, I. Yanagisawa, M. Ninomiya, T. Ohara, JP Patent 58-17172B (1983).
[9] G. Blanchard, G. Ferre, EP Patent 609,122-A1 (1994).
[10] C. Mazzocchia, E. Tempesti, R. Anouchinsky, A. Kaddouri, FR Patent 2,693,384
(1994).
[11] M. Lin, M. Linsen, EP Patent 962,253,A2 (1999).
[
[
[
12] M. Baca, A. Picamo, J.L. Dubois, J.M.M. Millet, Catal. Commun. 62 (2005) 215–220.
13] M. Takahashi, S. To, S. Hirose, JP Patent 98,118,491 (1998).
14] M. Lin, T.B. Desai, F.W. Kaiser, P.D. Klugherz, Catal. Today 61 (2000) 223–229.
[15] P. Concepci o´ n, P. Botella, J.M. L o´ pez Nieto, Appl. Catal. A 278 (2004) 45–56.
[
[
[
16] B. Mitra, I.E. Wachs, G. Deo, J. Catal. 240 (2006) 151–159.
17] C.D. Baertsch, K.T. Komala, Y.-H. Chua, E. Iglesia, J. Catal. 205 (2002) 44–57.
18] D. Kulkarni, I.E. Wachs, Appl. Catal. A 237 (2002) 121–137.
The selectivity would be lowered by the additional steps that
converted propylene to either 2-propanol or acetone (Fig. 10),
which are known to be promoted on the acidic sites of the catalysts
[19] M. Ai, J. Catal. 101 (1986) 389–395.
[
[
[
20] A. Kaddouria, C. Mazzocchiaa, E. Tempesti, Appl. Catal. A 180 (1999) 271–275.
21] L. Luo, J.A. Labinger, M.E. Davis, J. Catal. 200 (2001) 222–231.
22] J.M. Oliver, J.M. L o´ pez Nieto, P. Botella, A. Mifsud, Appl. Catal. A 257 (2004) 67–76.
[
12,15]. Acidic sites are also detrimental to selectivity because they
promote the complete oxidation of products. However, the results
of NH -TPD (Fig. 7) indicated that the W addition decreased the
[23] P. Botella, E. Garc ı´ a-Gonz a´ lez, J.M. L o´ pez Nietoa, J.M. Gonz a´ lez-Calbet, Solid State
Sci. 7 (2005) 507–519.
[
[
24] J.M. L o´ pez Nieto, P. Botella, B. Solsona, J.M. Oliver, Catal. Today 81 (2003) 87–94.
3
25] M. Goti c´ , M. Ivanda, S. Popovi c´ , S. Musi c´ , Mater. Sci. Eng. B 77 (2000) 193–201.
amounts of acidic sites in the catalyst, thus contributing to an
increase in the selectivity. Based on the above arguments, it can be
concluded that the selectivity was improved on the W-added
catalysts, because the dehydration step was promoted to a greater
[26] M. Deepa, M. Kar, S.A. Agnihotry, Thin Solid Films 468 (2004) 32–42.
[27] W. Kuang, Y. Fan, Y. Chen, J. Colloid Interface Sci. 215 (1999) 364–369.
[
28] M. Alifanti, J. Kirchnerova, B. Delmon, D. Klvana, Appl. Catal. A 262 (2004) 167–
76.
29] H. Hosono, J. Non-Cryst. Solids 352 (2006) 851–858.
1
[