Chemistry Letters Vol.32, No.11 (2003)
1029
Table 1. Propane oxidation over Mo–V–(Te)–O catalystsa
2
ꢃ1
Selectivity/%c
Catalysts
SBET/m ꢂg
Bulk composition
Reaction
ꢁ
C3H8 conv./%
(
Mo/V/Te)b
1/0.34/0
—
temp./ C
AA
5.7
6.0
6.4
3.2
1.1
49.3
PEN
6.8
6.0
19.9
14.7
30.8
10.4
AcA
19.0
21.1
24.9
29.5
23.9
15.1
Ace
0.4
0.5
1.1
0.9
1.2
2.7
CO2
29.1
28.5
19.1
18.8
15.0
10.2
CO
39.0
37.9
28.6
32.9
28.0
12.3
Cat-A-1
Cat-A-2
Cat-A-3
Cat-B-1
Cat-B-2
6.1
5.9
9.3
10.3
7.6
5.8
350
345
346
349
343
354
24.0
26.0
10.8
12.4
5.9
—
1/0.29/0
—
1/0.44/0.1
Cat-C-1
25.6
a
ꢃ1
b
Catalyst weight = 0.5 g;feed flow rate = 20 mL ꢂmin , feed composition (mol%), C3H8:O2:H2O:N2 = 6.5:10.0:45.0:38.5. Bulk composition was deter-
c
mined by ICP analysis. AA = acrylic acid, PEN = propene, AcA = acetic acid, Ace = acetone.
2
ꢃ1
ored solid with one broad diffraction peak at the low angle re-
gion instead of the characteristic three peaks (Figure 2e), though
the same diffraction peaks at 22 and 45 were observed. The ob-
was smaller than that of Cat-B-1 (10.3 m ꢂg ). It is clear from
this result that the orthorhombic structure has intrinsically high
catalytic oxidation activity. This conclusion is supported by
the following results: The activity of Cat-A-3, in which the or-
thorhombic structure has been decomposed by increasing calci-
ꢁ
tained material was abbreviated as Cat-B. Its bulk composition
was determined to be Mo1V0:29Ox (Table 1) which was interest-
ingly quite resembled to that of the orthorhombic material.
Based on the observed broad XRD pattern and elemental analy-
sis, we speculate a disorder in the arrangement of Mo and V giv-
ing the broad peak at the low angle region. We have already
ꢁ
nation temperature (500 to 600 C), was less than half of that of
mono-phasic orthorhombic Mo–V–O (Cat-A-1). Moreover, Cat-
B-2 in which the main phase was (V0:07Mo0:93)5O14 type struc-
ture showed very poor activity.
In the case of Mo–V–Te–O catalyst, air calcination before
N2 calcination can improve the selectivity to acrylic acid from
propane as described before. In the case of Mo–V–O catalyst,
however, this air calcination hardly influenced the catalytic per-
formance (Cat-A-2) as well as the XRD patterns. Thus, air cal-
cination is an unnecessary treatment in the case of orthorhombic
1
0
reported the similar disordered structure previously. The main
structural difference between Cat-A and Cat-B is, therefore,
whether the slab plane is ordered or not.
Thermal stabilities of the above materials were tested by
XRD and the results are shown in Figure 2. Tellurium-contain-
ing catalyst (Mo6V3Te1Ox, abbreviated as Cat-C), which was ef-
6
ꢁ
fective for selective oxidation from propane to acrylic acid, was
also tested for comparison. The preparative procedure was
Mo–V–O and N2 calcination at 500 C is enough to attain the
catalyst performance.
5
,6
reported previously. Cat-A and Cat-B were heat-treated in
The most striking point in the present work is that the ortho-
rhombic Mo–V–O (Cat-A-1) and the orthorhombic Mo–V–Te–
O (Cat-C-1) was greatly different in the catalytic selectivity.
The propane conversion over Cat-A-1 was about the same of that
of Cat-C-1, but as for the selectivity to acrylic acid, Cat-A-1 was
only a tenth part of Cat-C-1. Accordingly, it is perceived that the
effect of Te is not only thermal stabilization, but also the im-
provement of the selectivity to acrylic acid from propane.
The catalytic function of orthorhombic Mo–V–Te–O is now
clear;intrinsic catalytic oxidation activity comes from Mo and V
oxides but these elements have to be organized in a ordered
structure, and the selectivity derives from the third element, like
Te or Sb, which has to be located in the above ordered structure
without disturbing the structure. In this respect the element ar-
rangement shown in Figure 1 seems to be one of the most suit-
able structures.
ꢃ1
ꢁ
N2 stream (50 mLꢂmin ) at 500 C (Cat-A-1 and Cat-B-1) or
ꢁ
600 C (Cat-A-3 and Cat-B-2) for 2 h. Since Cat-C is normally
calcined in the condition (in air at 280 C for 2 h and in N2 at
600 C for 2 h, Cat-C-1) for achieving high selectivity to acrylic
acid, Cat-A was also calcined in the similar condition (in air at
ꢁ
ꢁ
ꢁ
ꢁ
2
80 C for 2 h and in N2 at 500 C for 2 h) for comparison
Cat-A-2).
As can be seen in Figure 2, Cat-A and Cat-B was thermally
(
ꢁ
stable in N2 up to 500 C and the stability (Cat-A-2) was not ap-
preciably affected by air treatment unless the heat-treatment
ꢁ
ꢁ
temperature in N2 exceeds 500 C. When the temperature of cal-
cination increased up to 600 C, many other phases, such as
V0:95Mo0:97O5 [JCPDS 77-0649], (V0:07Mo0:93)5O14 [JCPDS
31-1437], and MoO3 [JCPDS 76-1003] (for only Cat-B-2), were
detected in the both cases (Figure 2d for Cat-A-3 and Figure 2g
for Cat-B-2). However, it should be noted that the orthorhombic
phase still partially remained in Cat-A-3 but the broad peaks
were completely disappeared in Cat-B-2. On the other hand,
Cat-C maintained the mono-phasic state even after the heat-
References
1
a) T. Ushikubo, H. Nakamura, Y. Koyasu, S. Wajiki, U. S. Patent 5380933
1995). b) W. Ueda, K. Oshihara, D. Vitry, T. Hisano, and Y. Kayashima,
Catal. Surv. Jpn., 6, 33 (2002). c) M. Lin, T. Desai, F. Kaiser, and P. Klugherz,
Catal. Today, 61, 223 (2000). d) L. Luo, J. A. Labinger, and M. E. Davis, J.
Catal., 200, 222 (2001). e) P. Botella, J. M. Lopez-Nieto, and B. Solsona,
Catal. Lett., 78, 383 (2002).
(
ꢁ
treatment at 600 C (Figure 2i for Cat-C-1). These results indi-
cate that the thermal stability of the orthorhombic Mo–V–O is
slightly poor compared with Mo–V–Te–O though both catalysts
have the same orthorhoimbic structure, implying that third ele-
ment Te must stabilize the orthorhombic structure of the Mo–
V–O catalyst.
Based on the above structural results, we discuss the origin
of alkane oxidation activity and selectivity of the Mo–V–Te–O
catalysts. Catalytic activities of the catalysts prepared in the
present work for the propane oxidation are shown in Table 1.
The reaction conditions were shown in the footnotes. The pro-
pane conversion over the orthorhombic Mo–V–O catalyst
2
H. Watanabe and Y. Koyasu, 4th WCOC, Potsdam, recent report (2001),
Vol. 16.
3
4
5
6
7
8
H. Tsuji and Y. Koyasu, J. Am. Chem. Soc., 124, 5608 (2002).
W. Ueda and K. Oshihara, Appl. Catal., A, 200, 135 (2000).
D. Vitry, Y. Morikawa, J. L. Dubois, and W. Ueda, Top. Catal., 23, 47 (2003).
K. Oshihara, T. Hisano, and W. Ueda, Top. Catal., 15, 153 (2001).
D. Vitry, Y. Morikawa, J. L. Dubois, and W. Ueda, Appl. Catal., A, in press.
J. M. M. Millet, M. Baca, A. Pigamo, D. Vitry, and W. Ueda, Appl. Catal., A,
244, 359 (2003).
9
a) L. Kihlborg, Acta Chem. Scand., 21, 2495 (1967). b) H. A. Eick and L.
Kihlborg, Acta Chem. Scand., 20, 722 (1966). c) Y. T. Hu and P. K. Davis,
J. Solid State Chem., 105, 489 (1993). d) L. Kihlborg, Acta Chem. Scand.,
2
3, 1834 (1969). e) T. Ekstrom, and M. Nygren, Acta Chem. Scand., 26,
(
(
Cat-A-1) was twice of that over disordered Mo–V–O catalyst
Cat-B-1), despite the surface area of Cat-A-1 (6.1 m ꢂg
1827 (1972).
2
ꢃ1
)
10 W. Ueda, N. F. Chen, and K. Oshihara, Chem. Commum., 1999, 517.
Published on the web (Advance View) October 13, 2003;DOI 10.1246/cl.2003.1028