Ammoxidation of 2-picoline catalyzed by modified V2O 5/TiO2

Article abstract of DOI:10.1007/s00706-014-1196-7

A vanadium-titanium catalyst was used in the preparation of aromatic nitriles by ammoxidation in a fixed-bed reactor. The catalyst prepared by the mixing method showed high conversion (91 %) and selectivity (95 %) in the ammoxidation of 2-picoline. The effect of varied conditions on the ammoxidation of 2-picoline was also explored, including LHSV, H₂O/NH₃ molar ratio, reaction temperature, and the catalyst calcination temperature. This catalyst was also used in the ammoxidation of 3-picoline, 4-picoline, and toluene derivatives.

Full text of DOI:10.1007/s00706-014-1196-7

Monatsh Chem
DOI 10.1007/s00706-014-1196-7
ORIGINAL PAPER
Ammoxidation of 2-picoline catalyzed by modified V O /TiO
2
2
5
Jiong-bin Pan
Jia-long Lan
•
Jia-min Huang
•
•
Jia-xu Li
Chao Qian Xin-zhi Chen
•
Zeng-yang Jiang
•
•
Received: 2 November 2013 / Accepted: 24 February 2014
Ó Springer-Verlag Wien 2014
Abstract A vanadium-titanium catalyst was used in the
preparation of aromatic nitriles by ammoxidation in a
fixed-bed reactor. The catalyst prepared by the mixing
method showed high conversion (91 %) and selectivity
intermediate; then this intermediate could be reacted with
KCN to form the corresponding cyanopyridine. However,
this method showed poor yield and used toxic cyanide.
Another approach is dehydration of primary amides,
although some amides on thermolysis showed some ten-
dency for dehydration to nitriles [9, 10]; the formed water
can hydrolyze another molecule of amide to acid. Besides,
ammoxidation is a significant reaction for producing aro-
matic nitriles [11, 12]. In this approach, the by-product is
only water, and the yield of product is higher than in the
(
95 %) in the ammoxidation of 2-picoline. The effect of
varied conditions on the ammoxidation of 2-picoline was
also explored, including LHSV, H O/NH3 molar ratio,
2
reaction temperature, and the catalyst calcination temper-
ature. This catalyst was also used in the ammoxidation of
3
-picoline, 4-picoline, and toluene derivatives.
above two methods. In addition, NH used in this method is
3
Keywords Catalysts Á Aromatic nitriles Á
friendlier to the environment than the cyanide used in
cyanation. Many previous researches on the ammoxidation
of picoline isomers have been reported. Conversions of
Fix-bed reactor Á Calcination temperature Á Mixing method
3-picoline and 4-picoline were similarly high as well as the
Introduction
corresponding nitrile yields [1, 4, 13, 14]. While in case of
using 2-picoline as material, the yield of 2-cyanopyridine
was low, and pyridine was the main by-product. Until now,
there have been no reports on the ammoxidation of 2-pic-
oline giving a high yield of the corresponding nitrile.
A V-containing catalyst such as VTi, VMo, VSb, VSn,
or VCr oxides is a good catalytic system for ammoxidation
[4, 15–18]. The catalytic properties are largely influenced
by the composition, character of support, and method of
preparation [18–22]. Narayana et al. [14] reported on the
influence of V O content on the V O /TiO catalyst for
The aromatic nitriles are important reactants in the fine
chemical sector [1–3]. In addition, they can be converted to
commercially interesting and valuable intermediates for the
synthesis of several pharmaceuticals and dyestuffs as well
as pesticides and final products [4].
At present, there are a few methods available for the
preparation of aromatic nitriles. One of them is cyanation
[
5–8]. The pyridine nitrogen was activated by using dini-
trogen pentoxide or nitric acid to give a corresponding
2
5
2
5
2
ammoxidation of 3-picoline. Their study revealed that the
catalysts show increasing conversion of 3-picoline up to
J. Pan Á J. Huang Á J. Li Á Z. Jiang Á J. Lan Á C. Qian Á X. Chen
Department of Chemical and Biological Engineering, Zhejiang
University, Hangzhou, People’s Republic of China
e-mail: chemtec@163.com
3
3
.4 mol% V O content with close to 100 % selectivity of
2 5
-cyanopyridine. Rao et al. [13] studied the ammoxidation
of 3-picoline and 4-picoline over ceria-titania catalysts.
Their research revealed anatase-supported ceria catalysts
exhibited higher ammoxidation activities (68.1 % yield)
than rutile-supported ones (37.5 % yield). Besides, the
method of preparation is a significant factor of the catalytic
J. Pan (&) Á J. Huang Á C. Qian Á X. Chen
State Key Laboratory of Chemical Engineering, Zhejiang
University, Hangzhou, People’s Republic of China
e-mail: 3090103113@zju.edu.cn
123

J. Pan et al.
properties. Most of the vanadium-containing catalysts used
in the ammoxidation were prepared by the impregnation
method, but the mixing method had been rarely used.
Martin et al. [23, 24] reported on the use of vanadium
phosphate catalysts for the ammoxidation of 2-picoline,
and V–P–O catalyst showed the highest 2-cyanopyridine
yield of only 57 %. Verma [25] studied the ammoxidation
of 2-picoline on V–Sb catalysts with a 2-cyanopyridine
yield of 65 %. Almost all of the ammoxidation catalysts
showed poor yield of 2-cyanopyridine.
2-picoline at the same condition as the reaction using the
V–O–Ti-a catalyst. As the results show in Fig. 2, the V–O–
Ti catalytic activity decreased with increasing calcination
temperature, but the catalytic selectivity of V–O–Ti first
increased and then slightly decreased. The 2-picoline
reacting in the presence of V–O–Ti-c catalyst gave the
highest 2-cyanopyridine yield (86 %). The results reveal
that the vanadium–titanium catalysts prepared by the
mixed method could efficiently catalyze the ammoxidation
of 2-picoline, and the catalytic selectivity could be
improved by properly increasing the catalyst calcination
temperature.
In this article, we show that the modified vanadium-
titanium catalyst prepared by the mixing method can act as
an efficient catalyst for the ammoxidation of 2-picoline.
This catalyst also showed good activity and selectivity in
the ammoxidation of 3-picoline, 4-picoline, and toluene
derivatives. Furthermore, the influence of varied conditions
on the ammoxidation of 2-picoline was also studied.
Next, the V–O–Ti-c catalyst was used in the ammoxi-
dation of 3-picoline, 4-picoline, and toluene derivatives.
-
1
The reactions were carried out at 370 °C and 0.18 h
LHSV (2-picoline). As shown in Table 1, 3-picoline,
4-picoline, and several kinds of structurally diverse toluene
derivatives could be converted into the corresponding
nitriles. Both 3-picoline and 4-picoline gave high conver-
sions and high selectivities. In general, good conversions of
toluene derivatives and high selectivities of products were
obtained. In the case of toluene (entry 4), the conversion of
substrate and the selectivity of benzonitrile were consid-
erably high. Xylenes (entry 5–7) gave tolunitriles in high
conversions and moderate to high selectivities. In the case
of o-xylene, the selectivity of o-tolunitrile was high
because of hardly any ammoxidation of o-tolunitrile owing
to the space steric effect of the ortho methyl group. In the
ammoxidation of m-xylene and p-xylene, m-phthalodinit-
rile and terephthalonitrile were formed, respectively, as the
main products because of the easy ammoxidation of the m-
tolunitrile and p-tolunitrile. The electronic effect of sub-
stituent was not significant in the ammoxidation of toluene
derivatives. The conversion of 4-chlorotoluene and the
selectivity of 4-chlorobenzonitrile were high. When the
substrate was 4-methylanisole, the result was slightly
worse than the reaction using 4-chlorotoluene. These
results show that the V–O–Ti-c catalyst also had a good
activity and selectivity in the ammoxidation of 3-picoline,
4-picoline, and toluene derivatives.
Results and discussion
Initially, the ammoxidation of 2-picoline was carried out in
the presence of vanadium-titanium catalyst prepared by the
mixing method. This catalyst was calcined at 500 °C and
marked as V–O–Ti-a. The reaction was carried out at
-
1
3
70 °C and 0.18 h
LHSV (2-picoline); the 2-pico-
line:NH :H O:air molar ratio was 1:6:3.3:22. The V–O–Ti-
3
2
a catalyst showed 97 % conversion of 2-picoline but only
0 % selectivity of 2-cyanopyridine. According to the
3
mechanism of 3-picoline ammoxidation, which was
reported by Rao et al. [26], we can guess the possible
routes of 2-picoline ammoxidation as Fig. 1 shows. As the
corresponding amide and a large number of pyridines were
observed in the product, route 3 in Fig. 1 is recommended.
The 2-picoline was probably first oxidized to the corre-
sponding carboxylic acid, and the corresponding
carboxylic acid decarboxylated to pyridine easily. This
reaction behavior was probably because of the involvement
of the lone electron pair of the nitrogen ring in the
chemisorption process of the 2-picoline molecule, and the
methyl group is too near to the surface. Hence, the C–C
bond is more easily ruptured [1], so we tried to reduce the
catalytic activity to increase the selectivity of 2-cyano-
pyridine. In the preparation of the catalyst, a high catalyst
calcination temperature could reduce the specific surface
area and surface activity. Hence, we guess the selectivity of
To determine how the varied reaction conditions affect
the ammoxidation of 2-picoline, several related experi-
ments were done in the model reaction, and the results are
depicted in Table 2. The effect of liquid hourly space
velocity (2-picoline) on the ammoxidation of 2-picoline
was studied. The conversion and selectivity of product are
highly influenced by the LHSV, because it determines the
contact time between the reactant and the catalyst. The
conversion of 2-picoline increased from 79 to 99 % with
2
-cyanopyridine could be increased by the higher catalyst
calcination temperature. To verify whether the catalytic
selectivity would be improved by the increase in calcina-
tion temperature, V–O–Ti-b, V–O–Ti-c, and V–O–Ti-d
were prepared by the V–O–Ti-a catalysts being calcined at
-
1
the decrease of LHSV from 0.3 to 0.15 h (entry 1–3). In
addition, the effect of reaction temperature on the am-
moxidation of 2-picoline was studied. On the one hand,
low temperature led to a low conversion of 2-picoline. On
600, 700, and 800 °C, respectively. These three catalysts
were used as the catalysts in the ammoxidation of
1
23

Ammoxidation of 2-picoline catalyzed
Fig. 1 The possible routes of ammoxidation. [O] represents lattice oxygen
Table 1 Ammoxidation of picoline isomers and toluene derivatives
Entry Substrate Product Conversion/% Yield/%
1
2
3
4
5
6
7
8
9
2-Picoline
3-Picoline
4-Picoline
2-Cyanopyridine 91
3-Cyanopyridine 92
4-Cyanopyridine 95
86
88
92
91
77
97
77
91
88
a
a
C
6
H
5
CH
(CH
3
C
6
H
5
CN
(CN)
-C
(CN)
CN
93
96
m-C
6
H
4
3
)
2
m-C
o-CH
p-C
p-Cl-C
p-CH O-
CN
6
H
4
2
o-C
p-C
6
H
4
(CH
(CH
3
)
2
3
6
H
4
CN 98
H
6 4
3
)
2
6
H
4
2
90
95
92
p-Cl-C
p-CH O-
CH
6
H
4
CH
3
6
H
4
3
3
C
H
6 4
3
6
C H
4
Reaction temperature 370 °C; feed: substrate ? H
2
O ? ammo-
-
1
nia ? air; LHSV = 0.18 h ; substrate:NH :H O:air = 1:6:3.3:22
3 2
Fig. 2 The results of ammoxidation of 2-picoline in the presence of
V–O–Ti-a, V–O–Ti-b, V–O–Ti-c, and V–O–Ti-d
a
Reaction temperature 380 °C
the other hand, too high temperature gave a low selectivity
of 2-cyanopyridine (entries 2, 4, 5) due to the separation of
the methyl group with the increase of temperature. The
influence of the mole ratio of 2-picoline to ammonia and
after catalyst preparation are shown in Fig. 3, and the
crystallite diameter was estimated by the Scherrer formula
the mole ratio of 2-picoline to H O were also explored
2
(D = crystallite
diameters = Kk/Bcosh,
K = 0.89,
(
entries 2, 6, 7). The decrease in the mole ratio of 2-pic-
oline to ammonia and the increase in the mole ratio of
-picoline to H O both raised the selectivity of 2-cyano-
k = 0.154 nm, B = FWHM, h = diffraction angle). The
BET surface area and the pore size were determined by
nitrogen adsorption at 77 K (ASAP 2020; Micromeritics
Co., Ltd., USA). The crystallite diameter of the catalyst,
BET surface area, and pore size results are shown in
Table 3. The catalyst characterization results reveal that
the vanadium–titanium catalyst calcined at higher tem-
perature obtained a larger crystallite diameter of the
catalyst, smaller surface area, and larger pore size, which
cause the decrease of the catalyst activity but increased the
catalyst selectivity to some extent. The BET surface areas
of V–O–Ti-a, V–O–Ti-b, V–O–Ti-c, and V–O–Ti-d
2
2
pyridine. According to route 3 in Fig. 1, the increased
amount of ammonia could restrain the generation of pyri-
dine, and the decreased amount of H O could promote the
2
yield of 2-cyanopyridine, consistent with the results.
Catalyst characterization
The XRD analyses (D/max-rA; Rigaku Co., Ltd., Japan) of
V–O–Ti-a, V–O–Ti-b, V–O–Ti-c, and V–O–Ti-d catalysts
123

J. Pan et al.
Table 2 The effect of varied conditions on the ammoxidation of 2-picoline
-
1
a
3
a
Entry
T/°C
LHSV/h
2-Picoline:NH
2-Picoline:H
2
O
Conversion/%
Selectivity/%
1
2
3
4
5
6
7
370
370
370
350
390
370
370
0.15
0.18
0.3
1:6
1:6
1:6
1:6
1:6
1:3
1:6
1:3.3
1:3.3
1:3.3
1:3.3
1:3.3
1:3.3
1:7
99
91
79
71
98
85
90
78
95
96
89
60
84
75
0.18
0.18
0.18
0.18
3
-1
The reactions were carried out at 150 cm s air in the presence of V–O–Ti-c catalyst
a
Mole ratio
Add water
extrusion
NH
TiO
4
VO
3
cylindrical
catalyst
dried for
1h at 60°C
mixture
2
6
7
8
00°C
00°C
00°C
V-O-Ti-b
V-O-Ti-c
V-O-Ti-d
calcined
for 5h at
calcined
for 2h at
250°C
V-O-Ti-a
5
00°C
Fig. 4 The process of preparing the V–O–Ti catalyst by the mixing
method
Experimental
Fig. 3 XRD patterns of V–O–Ti-a, V–O–Ti-b, V–O–Ti-c, and V–O–
Ti-d catalyst
Catalyst preparation
The V–O–Ti catalyst was prepared by using the mixing
method (Fig. 4). Solid TiO (150 g) and 20 g NH VO
3
2
3
4
Table 3 BET surface area, pore size, and crystallite diameter of the
catalyst
crystals were added to 50 cm distilled water. The mixture
was formed into a cylindrical catalyst of 7 mm length and
2
BET surface area/m g
-1
Catalysts
Pore size/nm
D/nm
4
mm diameter. The cylindrical catalyst was dried for 1 h
V–O–Ti-a
V–O–Ti-b
V–O–Ti-c
V–O–Ti-d
0.9
0.8
0.7
0.7
14.3
15.5
17.0
17.7
55
62
68
72
at 60 °C, then calcined for 2 h at 250 °C and for 5 h at
500 °C (V–O–Ti-a) in a muffle furnace. The V–O–Ti-b, V–
O–Ti-c, and V–O–Ti-d catalysts were prepared by the
calcination of V–O–Ti-a catalysts for 4 h at 600, 700, and
8
00 °C, respectively. The V O loading of the V–O–Ti
2
5
catalyst was 8.3 mol%. Compared to the impregnation
method [14], this mixing method is simpler for preparation
of V O /TiO catalyst, and the V–O–Ti catalyst had
2 -1
catalysts are from 0.7 to 0.9 m g , but Narayana et al.
2
14] observed an 18.6 m g
-1
[
BET surface area in a
2
5
2
7
.2 mol% V O /TiO catalyst. These two different results
2
comparative catalytic activity.
5
2
are due to the different methods of preparing the catalyst.
The catalyst prepared by the impregnation method had a
larger BET surface area because of the supporter. How-
ever, the surface area of the catalyst prepared by the mixing
method was mainly obtained from the porous structure that
was generated during the decomposition of ammonium
vanadate.
Reaction procedure
The synthesis was carried out in a stainless-steel fixed-bed
reactor. The reactants were fed from the top of the reactor.
Three calibrated flow meters were used for ammonia, air,
and the mixture of picoline and water. The reactor was
1
23

Ammoxidation of 2-picoline catalyzed
T
Acknowledgments The authors are grateful for the financial sup-
port from the Natural Science Foundation of China (21376213), the
Research Fund for the Doctoral Program of Higher Education of
China (20120101110062), and the Low Carbon Fatty Amine Engi-
neering Research Center of Zhejiang Province (2012E10033).
Vaporization
P
Catalyst
T
References
Heating jacket
2
Picoline+H O
1
. Martin A, Kalevaru VN, Lucke B, Bruckner A (2008) Appl Catal
A Gen 335:196
NH
3
T
2
3
4
5
. Ushijima S, Moriyama K, Togo H (2011) Tetrahedron 67:958
. Manaka A, Sato M (2005) Synth Commun 35:761
. Martin A, Lucke B (2000) Catal Today 57:61
. Katritzky AR, Scriven EFV, Majumder S, Tu HB, Vakulenko
AV, Akhmedov NG, Murugan R (2005) Synthesis 993
. Fife WK (1983) J Org Chem 48:1375
. Fife WK, Boyer BD (1984) Heterocycles 22:1121
. Vorbruggen H, Krolikiewicz K (1983) Synthesis 316
. Calderwood DJ, Davies RV, Rafferty P, Twigger HL, Whelan
HM (1997) Tetrahedron Lett 38:1241
Air
Condenser
Product
Discharge
6
7
8
9
Fig. 5 The fixed-bed reactor
1
1
0. Katritzky AR, Lapucha AR, Siskin M (1990) Energy Fuels 4:555
1. Grasselli R, Burrington J, Dicosimo R, Friedrich M, Suresh D
(
1988) Stud Surf Sci Catal 41:317
surrounded by a heating jacket to control the temperature
of the fixed bed. The mixture was reacted in a 40-cm
1
2. Rizayev R, Mamedov E, Vislovskii V, Sheinin V (1992) Appl
Catal A Gen 83:103
catalyst bed, and the effective radius of the reactor was
3
5 mm. The amount of catalysts packed was 260 cm . The
13. Rao PK, Rao KSR, Masthan SK, Narayana KV, Rajiah T, Rao
VV (1997) Appl Catal A Gen 163:123
1
1
4. Narayana KV, Venugopal A, Rao KSR, Rao VV, Masthan SK,
Rao PK (1997) Appl Catal A Gen 150:269
5. Mehandru SP, Anderson AB, Ross PN (1986) J Catal 100:210
products were cooled using a condenser. The figure of the
fixed-bed reactor is shown in Fig. 5.
1
The product analysis was checked by GC (Agilent
16. Bond GC (1997) Appl Catal A Gen 157:91
1
7. Chary KVR, Kishan G, Lakshmi K, Ramesh K (2000) Langmuir
6:7192
1
790F and SE-30 column; Nan-jing Xuxi Apparatus Co.,
1
Ltd., China) and GC-MS (TRACEGC2000/TRACEMS;
Thermo Quest, USA). The conversion and selectivity were
determined by an external standard method, the steps of
which were as follows:
1
1
8. Jiang F, Xu W, Niu L, Xiao GM (2013) Catal Lett 143:1200
9. Chary KVR, Kumar CP, Reddy KR, Bhaskar T, Rajiah T (2002)
Catal Commun 3:7
0. Roy SK, Dutta P, Nandi LN, Yadav SN, Mondal TK, Ray SC,
Mitra S, Samuel P (2004) J Mol Catal A Chem 223:211
1. Kumar CP, Reddy KR, Rao VV, Chary KVR (2002) Green Chem
2
2
2
1
.
First, seven standard samples of different concentra-
tions were prepared and then analyzed by GC at the
same injection.
4
:513
2. Srinivas N, Kishan MR, Kulkarni SJ, Raghavan KV (2000)
Microporous Mesoporous Mater 39:125
2
.
Second, a scatter diagram with a regression line was
made, for which the abscissa was the concentration of
the standard sample and the ordinate was the peak
area.
23. Martin A, Lucke B, Seeboth H, Ladwig G (1989) Appl Catal
9:205
4
4. Martin A, Lucke B (1996) Catal Today 32:279
2
2
5. Verma PK, Agarwal A (2003) Production of heteroaromatic
nitriles by oxidative ammonolysis using vanadium and antimony
oxide-based catalysts. World patent WO 2,003,022,819, Mar 20,
3
.
Finally, the reaction mixture was analyzed by GC at
the same injection volume as the standard samples, and
then the expected product’s concentration was
obtained.
2
003; Chem Abstr 138:239701
2
6. Ramachandra Rao R, Kulkarni S, Subrahmanyam M, Rama Rao
A (1996) Zeolites 16:254
123