1
82
JIE YU et al.
Therefore, in the present work, the synergy effect reached the set point and stabilized, oxygen regulated
between α-Bi Mo O and MoO was uncovered. by a flowmeter was fed in for one hour, then deionized
2
3
12
3
Besides, by taking into consideration process integrity, water was introduced into the system using a pulseless
the catalytic behaviors of three phases of bismuth pump for one hour, after that, the picoline in water was
molybdates as well as the performances of titania sup- used in place of pure water. Crude products were col-
ported α-Bi Mo O and MoO catalysts in oxidation lected and analyzed using an Agilent 6820 GC instru-
2
3
12
3
ment. The conversion of picolines and the selectivity
to pyridylaldehydes were calculated based on their
peak areas in gas chromatograph (external standard
method).
reactions of picolines were also investigated in detail.
EXPERIMENTAL
Catalyst Preparation. Catalysts were prepared
using the co-precipitation method. Commercially
available chemical reagents were used without any fur-
RESULTS AND DISCUSSION
ther purification. (NH ) Mo O · 4H O was dissolved
4
6
7
24
2
Our initial efforts then focused on screening of
in deionized water and heated to 70°C with stirring,
then a solution of 1M Bi(NO ) · 5H O in 2M HNO
three bismuth molybdates, namely, Bi Mo O (α),
2
3
12
3
3
2
3
Bi Mo O (β), and Bi MoO (γ), by using 2-picoline
2 2 9 2 6
was dropwise added. Subsequently, a certain amount
as the probe molecule. Catalytic test results were sum-
marized in Table 1. As shown, all the three phases can
be readily prepared at 600°C (Entry 1–3 and 6–8),
while at 550°C (Entry 5 and 10), the β phase was
unstable, and thus decomposed into α and γ phases
of TiO (anatase) was added, the resulting slurry was
2
stirred for 3 h and water was removed by a rotary evap-
orator, the residue was then extruded with the size of
2
mm in diameter and 2.5 mm in length, dried at 80°C
for 2 h, and calcined at 550°C for 10 h.
[
23], irrespective of whether loaded into titania or not.
Catalyst Characterization. The BET surface areas It is obvious that catalysts with pure α or β phase
were determined by nitrogen adsorption at 77K exhibited similar reactivity, both better than that of the
employing a Micromeritics TriStar II 3020 instru- γ phase (e.g. Entry 6, 7 to Entry 8), and the existence
ment.
of the γ phase in the catalyst caused a decline in con-
version of 2-picoline and selectivity to 2-pyridylalde-
hyde (Entry 5 and 10). In addition, for α and β phases,
the conversion nearly doubled when supported on
titania, while the selectivity only slightly changed with
a difference not exceeding 7%, exemplified in the case
of MB32/550 (α, conv. 15.5%, sel. 87.3%) and
MB32T550 (α, conv. 29.5%, sel. 81.3%).
X-ray diffraction (XRD) patterns were collected on
a Rigaku D/max-2550pc diffractometer using Cu Kα
radiation (40 kV, 250 mA) at diffraction angles (2θ)
from 20° to 90°.
Temperature-programmed desorption of ammonia
(
NH -TPD) was determined by a PX200 adsorption
3
instrument. The sample (100 mg) was pretreated at
5
00°C for 1 h in helium atmosphere, cooled to 40°C,
Based on the above, the α phase supported on tita-
filled with 4% NH /He mixed gas for half an hour, and nia (MB32T550) was subsequently used for further
3
–1
research on account of its stability as well as the supe-
rior catalytic performance over others. The reaction
temperature was an important parameter in terms of
the reactivity. With other conditions remaining
unchanged as in Table 1, when the reaction was car-
ried out at 290°C, an increase of the conversion to
then switched to pure helium (30 mL min ) for 1 h to
remove weakly adsorbed ammonia, and finally heated
–1
–1
to 500°C at 10°C min in pure He (30 mL min ).
Temperature-programmed reduction (TPR) was
carried out on a Micromeritics AutoChem II 2920
instrument, and the samples of the catalyst (100 mg)
were crushed into 60 meshes and preheated under
argon atmosphere at 400°C for 1 h, then cooled down
4
7
5.2% followed by a decrease of the selectivity to
0.3% was observed, further increasing the tempera-
ture to 310°C provided an improvement of the conver-
sion to 57.2% but also led to a downturn of the selec-
to 80°C and reduced with a 10% H /Ar mixed gas
2
–1
(
1
30 mL min ) from 80 to 800°C at a heating rate of
tivity to 61.0%. In terms of atom economy and as well
–
1
5°C min .
as product yield, a temperature of 290°C was assumed
Temperature-programmed desorption of oxygen as a reasonable and expedient value for follow-up
(
O -TPD) was performed with a GC1690 instrument, experiments.
2
and the samples of the catalyst (100 mg) were treated
In the previous research, in order to pursue
better catalytic performance, we had found
–1
with a 5% O /Ar mixed gas (30 mL min ) at 300°C for
2
1
2
h, then cooled down to 50°C, and heated to 900°C at that by introducing an extra amount of MoO into the
3
–1 –1
0°C min in pure He (40 mL min ).
Catalytic Tests. The catalytic tests were carried out cantly improved. Therefore, five catalyst samples were
in a fixed-bed stainless tube reactor with 10 mm ID of prepared (Table 2) and catalytic tests were carried out
α-Bi Mo O /TiO catalyst, the reactivity was signifi-
2 3 12 2
6
00 mm length under atmospheric pressure. The tem- accordingly (Table 3) [21]. It was shown that the con-
perature in the catalyst zone was kept constant and version for MB32T and MT each with only one active
measured using a thermocouple, after the temperature component was much lower than that for the other
PETROLEUM CHEMISTRY
Vol. 60
No. 2
2020