Amination of allyl alcohol to propionitrile over a Zn30Cr 4.5/γ-Al2O3 bimetallic catalyst via coupled dehydrogenation-hydrogenation reactions

Article abstract of DOI:10.1016/j.apcata.2013.07.026

A Zn₃₀Cr_4.5/γ-Al₂O₃ bimetallic catalyst that can perform coupled dehydrogenation and hydrogenation reactions was prepared for the amination of allyl alcohol to propionitrile. During the catalysis, the hydrogen derived from the dehydrogenation of the alcohol and imine acted as an in situ source for the hydrogenation of the carbon-carbon double bond. The catalyst exhibited good performance for the reaction at atmospheric ammonia pressure. The parameters that affect the catalyst performance were studied thoroughly, and an optimized process for synthesizing propionitrile from allyl alcohol and ammonia over the catalyst was obtained. Under the optimized conditions, the propionitrile yield was greater than 65%. The characterization results indicated that the dehydrogenation reaction mainly occurred on the Lewis acid sites and revealed that ZnAl ₂O₄ is the active species for the coupled dehydrogenation-hydrogenation reactions. Chromium doping of the γ-Al ₂O₃-supported zinc catalyst Zn₃₀/γ- Al₂O₃ resulted in a decrease in the size of the ZnAl ₂O₄ crystallites, which was favorable for the dehydrogenation-hydrogenation reactions. The characterization results also revealed that the catalyst deactivation was due to carbon deposition on the catalyst during the catalytic run. The catalyst could be reactivated by blowing air into the reactor at a high temperature.

Full text of DOI:10.1016/j.apcata.2013.07.026

Applied Catalysis A: General 467 (2013) 154–162
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Amination of allyl alcohol to propionitrile over a Zn Cr /␥-Al O
3
3
0
4.5
2
bimetallic catalyst via coupled dehydrogenation–hydrogenation
reactions
∗
Yuecheng Zhang, Tianyu Wei, Yanjie Pian, Jiquan Zhao
School of Chemical Engineering and Technology, Hebei University of Technology, Tianjin 300130, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
2 3
A Zn30Cr4.5/␥-Al O bimetallic catalyst that can perform coupled dehydrogenation and hydrogenation
Received 11 April 2013
Received in revised form 5 July 2013
Accepted 11 July 2013
reactions was prepared for the amination of allyl alcohol to propionitrile. During the catalysis, the
hydrogen derived from the dehydrogenation of the alcohol and imine acted as an in situ source for the
hydrogenation of the carbon-carbon double bond. The catalyst exhibited good performance for the reac-
tion at atmospheric ammonia pressure. The parameters that affect the catalyst performance were studied
thoroughly, and an optimized process for synthesizing propionitrile from allyl alcohol and ammonia over
the catalyst was obtained. Under the optimized conditions, the propionitrile yield was greater than 65%.
The characterization results indicated that the dehydrogenation reaction mainly occurred on the Lewis
acid sites and revealed that ZnAl2O4 is the active species for the coupled dehydrogenation–hydrogenation
reactions. Chromium doping of the ␥-Al2O3-supported zinc catalyst Zn30/␥-Al2O3 resulted in a decrease
in the size of the ZnAl2O4 crystallites, which was favorable for the dehydrogenation–hydrogenation
reactions. The characterization results also revealed that the catalyst deactivation was due to carbon
deposition on the catalyst during the catalytic run. The catalyst could be reactivated by blowing air into
the reactor at a high temperature.
Available online 20 July 2013
Keywords:
Allyl alcohol
Amination
Propionitrile
Bimetallic catalyst
Zn30Cr4.5/␥-Al2O3
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
to acetaldehyde. Then two molecules of acetaldehyde undergo
aldol condensation on the acid sites of the catalyst to give
crotonaldehyde. After the formation of crotonaldehyde, there
are two possible pathways (A and B) to produce butyronitrile
from crotonaldehyde. In path A, crotonaldehyde condenses with
ammonia to generate the intermediate imine. Then the imine
is dehydrogenated to give crotonitrile and hydrogen. Lastly, the
carbon-carbon double bond in the crotonitrile molecule is hydro-
genated with the hydrogen generated from the dehydrogenation
of the imine and alcohol to give butyronitrile. In path B, the hydro-
genation of crotonaldehyde occurs first to give butyroaldehyde.
Then butyroaldehyde condenses with ammonia, and dehydrogena-
tion follows to give butyronitrile. The results indicated that the
hydrogen generated in situ can serve as a hydrogen source for
the hydrogenation of unsaturated compounds in the transforma-
tion of crotonaldehyde to butyronitrile. The interesting results
motivated us to find an effective catalyst with the capacity
to perform dehydrogenation–hydrogenation reactions using the
hydrogen generated in situ. The amination of allyl alcohol to pro-
pionitrile was used as a model reaction, and a series of catalysts
were tested. Interestingly, the dehydrogenation–hydrogenation
coupling was successful; yet, 3-methyl pyridine was obtained
In organic synthesis, nitriles are important intermediates that
can easily be converted into a variety of compounds [1]. Low-
molecular-weight aliphatic nitriles, such as acetonitrile, are used
as solvents. Aliphatic nitriles can be obtained from the amination-
dehydrogenation of alcohols [2,3]. In this process, two moles of
hydrogen are released, as shown in Eq. (1):
RCH OH + NH → RCN + H O + 2H
(1)
2
3
2
2
Recently, we successfully synthesized acetonitrile through the
amination-dehydrogenation of ethanol over a CoNi/␥-Al O3 cat-
2
alyst [4–6]. Analysis of the results showed that pyridine bases
and butyronitrile were formed in addition to acetonitrile. Pyri-
dine bases are common by-products in the synthesis of acetonitrile
from ethanol [7] or acetaldehyde [8]. However, the detection
of butyronitrile with a selectivity as high as 7.6% under certain
conditions was unexpected. We proposed that butyronitrile is
formed as shown in Scheme 1[5]. First, ethanol is dehydrogenated
∗ Corresponding author. Tel.: +86 22 60202926; fax: +86 22 60202926.
E-mail address: zhaojq@hebut.edu.cn (J. Zhao).
as
results.
a major by-product. Herein, we report the preliminary
0
926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.07.026

Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
155
with a diameter of 20 mm and weight of 50 mg. Prior to analysis,
◦
all samples were pretreated at 400 C for 1 h under high vacuum
−
5
◦
conditions (5 × 10 Pa), followed by cooling to 200 C. Then pyri-
dine was adsorbed at this temperature for 15 min. The physisorbed
◦
pyridine was removed by evacuation at 200 C under high vac-
uum conditions (5 × 10 Pa) for 1 h. Then the infrared spectra were
−
5
recorded.
2.3. Catalytic tests
The catalytic amination of allyl alcohol to propionitrile was per-
formed in a fixed-bed reactor with an internal diameter of 20 mm
and length of 1000 mm. The fixed-bed reactor was placed in an elec-
tric furnace consisting of four heating zones equipped with four
temperature controllers. In the experiments, 15 ml of the prepared
solid catalyst was charged into the center of the reactor to avoid
any temperature gradients, while the top and bottom of the reac-
tor were filled with an inert packing material (Pyrex glass chips).
The thermocouple was inserted in the center of the catalyst bed to
monitor the temperature. The catalyst was reduced with hydrogen
◦
at 360 C before the reaction. The controlled flows of ammonia and
Scheme 1. Pathway for the formation of butyronitrile from ethanol.
allyl alcohol were passed separately into a vaporizer in which the
two reactants were preheated, vaporized and mixed perfectly. Then
the mixed gas was fed into the catalyst bed. The reaction tempera-
ture was varied from 360 C to 480 C in approximately 30 C steps.
The products were collected in a condenser, and the tail gas of the
reactor was fed successively into flasks filled with water to recover
the low boiling point compounds and unreacted ammonia. The liq-
uid products were separated in a gas–liquid separator and analyzed
by a gas chromatograph equipped with a 30 m PEG-20M capillary
column and thermal conductivity detector (TCD). H2 was used as
2
. Experimental
◦
◦
◦
2
.1. Catalyst preparation
All the chemicals were reagent grade. The zeolite and calcium
bentonite were purchased from Nankai University Catalyst Plant,
Tianjin, China. ␥-Al O , ZrO and SiO were purchased from Tianjin
2
3
2
2
Research and Design Institute of Chemical Industry, Tianjin, China.
The supported metal oxide catalysts were prepared by kneading a
the carrier gas. The exhaust (H , NH , C H , etc.) was analyzed using
2
3
2
4
the same gas chromatograph with N2 as the carrier gas. All data
were measured after the catalyst was on stream for 4 h. The selec-
tivity was calculated using the peak area internal standard method.
The components of the mixture were identified with a HP5971 MS
equipped with a 30 m PEG-20M capillary column.
mixture of ␥-Al O3 with an aqueous solution of the corresponding
2
transition metal nitrates. These mixtures were then dried and cal-
◦
cined at 550 C for 6 h. For example, the Zn30Cr4.5/␥-Al O3 catalyst
2
(
the subscripts represent the zinc and chromium contents in the
catalyst) was prepared as follows: 25.11 g of Zn(NO ) ·6H O and
3
2
2
6
1
.34 g of Cr(NO ) ·6H O were dissolved in 5 g of water, and then
3
2
2
2.0 g of ␥-Al O was added to the solution. The mixture was then
3. Results and discussion
2
3
kneaded for 2 h in a kneader, and the resulting kneaded material
was processed in an extruder to obtain extrudates with a diameter
3.1. Catalyst selection
◦
of 2 mm and length of 2.5 mm. The extrudates were dried at 110 C
◦
for 6 h and then calcined at 550 C for 6 h to yield the catalyst.
The purpose of this work was to design a catalyst that cat-
alyzes not only the dehydrogenation of alcohols and imines but also
the hydrogenation of carbon-carbon double bonds with the hydro-
gen generated in situ from the dehydrogenation. It is well known
that many supported transition metals can perform hydrogenation
catalysis and some transition metals, such as zinc, nickel, etc., can
catalyze the dehydrogenation of imines to nitriles in the amination
of alcohols to nitriles [9,10]. Our previous studies showed that some
Zn-based catalysts exhibit excellent performance in the dehydro-
genation of imines in the amination of ethanol and styrene oxide to
nitriles [4,5,9]. Initially, several catalysts containing 20% zinc were
prepared on ␥-Al O , H-ZSM-5, MCM-41, ZrO and calcium ben-
2
.2. Catalyst characterization
To determine the metal content, the sample was first dis-
solved in concentrated HNO3 and HF. Then the metal content of
the solution was analyzed using a T.J.A. ICP-9000 (N + M) ICP-AES
instrument. The XRD patterns of the samples were recorded with
a Rigaku D/max 2500 X-ray diffractometer using Cu K␣ radiation
◦
(
40 kV, 150 mA) in the range of 2Â = 10–90 . X-ray photoelectron
spectroscopy (XPS) was performed using a PHI 1600 spectrom-
eter and Mg K␣ X-ray excitation source. Transmission electron
microscopy (TEM) was performed using a JEOL 100CX-II instru-
ment. The specimens for TEM analysis were prepared by ultrasonic
dispersion in n-butanol and subsequent evaporation of a drop of
the resultant suspension on a lacey carbon/Cu grid. The surface
area, total pore volume and pore size distribution of the samples
were measured by nitrogen adsorption at 77 K using a Micromeri-
tics ASAP 2020 surface area and porosity analyzer. The IR spectra of
adsorbed pyridine were recorded on a Thermo Nicliet Nicolet Nexus
2
3
2
tonite. Their catalytic performances in the reaction of interest were
tested, and the results are presented in Table 1. It was found that the
allyl alcohol conversions over the catalysts supported on ␥-Al O3
2
and H-ZSM-5 were higher than 90% under atmospheric ammonia
pressure at 693 K. However, the allyl alcohol conversions were very
low in the cases of the ZrO and calcium bentonite supports. The dif-
2
ferent activities of the catalysts might be due to the difference in the
acidity of the support materials. GC–MS analysis indicated that the
main components in the volatile products were propionitrile and
3-picoline. In addition, acetonitrile, isobutyronitrile and pyridine
bases, including pyridine and 3,5-lutidine, were also detected.
4
70 spectrometer equipped with a heatable and evacuatable IR cell
with CaF2 windows connected to a gas dosing-evacuation system.
The powdered samples were pressed into self-supporting wafers

1
56
Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
Table 1
Performance of the zinc catalysts with different supports.
Catalyst
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3,5-Luti
FD
Zn30/SiO2
Zn30/calcium bentonite
Zn30/ZrO2
Zn30/␥-Al2O3
Zn30/H-ZSM-5
82.7
36.6
50.3
94.4
97.5
18.9
0
36.2
46.5
21.0
14.7
28.2
0
8.9
26.0
3.9
6.5
6.8
2.1
4.2
2.5
11.3
0
2.4
2.1
5.7
0
4.7
2.8
3.5
0.7
5.2
0
2.2
1.4
0.7
0.6
1.3
1.6
0.3
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
◦
−1
.
Reaction conditions: reaction temperature 420 C, ammonia/allyl alcohol molar ratio 3:1, atmospheric pressure, GHSV: 200 h
Table 2
Effect of the zinc content on the catalytic performance.
Catalyst
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3, 5-Luti
FD
␥
-Al2O3
98.6
99.0
99.0
94.4
92.6
4.5
20.5
23.5
46.5
6.9
17.5
10.6
9.4
8.9
2.4
0
1.1
0.5
2.5
2.4
1.7
0.7
1.3
2.6
2.8
0.05
6.2
3.0
4.0
2.2
0.7
1.4
0.4
0.7
1.6
1.8
Zn10/␥-Al2O3
Zn20/␥-Al2O3
Zn30/␥-Al2O3
Zn40/␥-Al2O3
1.0
0.6
2.1
0.4
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
◦
−1
.
Reaction conditions: reaction temperature 420 C, ammonia/allyl alcohol molar ratio 3:1, atmospheric pressure, GHSV 200 h
Of all the catalysts, the one on the ␥-Al O3 support
catalyst exhibited better performance than the Zn30/␥-Al O3 cat-
2
2
had the highest conversion and selectivity in the coupled
dehydrogenation–hydrogenation reactions. Therefore, the effect of
the zinc content on the catalysis was studied in more detail to opti-
mize the catalytic performance. A series of catalysts with different
zinc contents was prepared and tested. The catalytic test results
are given in Table 2. It can be seen that zinc is necessary for the
transformation of allyl alcohol to propionitrile. The yield of the
desired propionitrile product increased with increasing zinc con-
tent, reaching a maximum of 46.5% when the zinc content was 30%.
Then it decreased sharply as the zinc content was further increased.
When the zinc content was 40%, the propionitrile yield was only
alyst. Specifically, doping Zn30/␥-Al O3 with Cr increased the
propionitrile yield from 46.5% to 61.4%.
2
3
.2. Performance of the Zn30Cr4.5/ꢀ-Al O catalyst under
2 3
different reaction conditions
The influence of the reaction temperature on the performance of
the Zn30Cr4.5/␥-Al O catalyst was investigated in the temperature
2
3
range of 633–753 K under atmospheric ammonia pressure, which
was maintained with a molar ratio of ammonia to allyl alcohol of
−1
3
:1 and GHSV of 200 h . The results shown in Table 4 indicated
6
.9%. The yields of 3-picoline and 3,5-lutidine decreased monoton-
ically as the zinc content was increased.
The Zn30/␥-Al O catalyst still did not give a satisfactory propi-
that the conversion of allyl alcohol increased with an increase in
temperature. Meanwhile, the propionitrile yield increased with an
increase in the temperature from 633 to 723 K and reached a maxi-
mum of 65.2% at 723 K. Above that temperature, the yield decreased
with increasing temperature. It was also found that the yields of
the acetonitrile and isobutyronitrile by-products increased with
an increase in temperature. In contrast, the selectivity toward 3-
picoline and other pyridine bases decreased with an increase in
temperature. In principle, the formation of propionitrile competes
with that of pyridines. Overall, high temperatures favored the for-
mation of nitriles, but low temperatures favored the formation of
pyridine bases in the experimental temperature range.
2
3
onitrile yield. To improve the catalytic performance, Zn30/␥-Al O3
2
was doped with other metals, including Fe, Cu, Co, Ni and Cr, to
prepare the corresponding bimetallic catalysts. The content of the
second metal in all the bimetallic catalysts was 4.5%. The bimetallic
catalysts were used in the amination of allyl alcohol to propionitrile,
and the results are listed in Table 3.
It was found that the doping metals exhibited completely dif-
ferent effects on the catalyst. Co and Ni had no obvious effect on
the catalytic performance, while Fe and Cu doping, especially the
latter, led to a decrease in the propionitrile yield. However, the
The influence of the molar ratio of ammonia to allyl alcohol on
the reaction was investigated under atmospheric ammonia pres-
3
-picoline yield increased to some degree when the catalyst was
doped with any of these four metals. Cr had a completely differ-
ent effect on the catalyst than the other four metals. The Cr-doped
−1
sure and GHSV of 200 h at 723 K. The results are presented in
Table 4. As shown in the table, the propionitrile yield increased with
Table 3
Effect of the doped metals on the performance of the zinc-based catalyst.
Catalyst
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3,5-Luti
FD
Zn30/␥-Al2O3
94.4
98.0
97.5
99.6
99.3
99.0
46.5
43.1
49.9
8.0
49.4
61.4
8.9
14.0
11.8
17.5
11.6
8.5
2.1
2.2
4.6
2.0
3.4
3.6
2.4
1.1
0.8
0.9
0.5
1.9
2.8
2.0
2.8
1.7
2.2
3.2
2.2
0.7
2.1
1.4
0.5
0.08
1.6
0.5
1.5
1.4
0.1
0.2
Zn30Fe4.5/␥-Al2O3
Zn30Co4.5/␥-Al2O3
Zn30Cu4.5/␥-Al2O3
Zn30Ni4.5/␥-Al2O3
Zn30Cr4.5/␥-Al2O3
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
−
1
Reaction conditions: reaction temperature 693 K, ammonia/allyl alcohol molar ratio 3:1, atmospheric pressure, GHSV 200 h
.

Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
157
Table 4
Effect of the temperature on the performance of the Zn30Cr4.5/␥-Al2O3 catalyst.
Temperature (K)
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3,5-Luti
FD
6
6
6
7
7
33
63
93
23
53
91.3
94.5
99.0
99.0
100
37.4
46.3
61.4
65.2
62.4
13.7
10. 9
8.6
7. 6
5.2
2.9
5.8
3.7
6.9
6.7
4.0
2.5
2.0
0.5
0.4
2.7
2.6
3.3
4.4
4.2
0.8
0.7
0.1
0.5
0.0
0.8
0.8
0.3
1.4
1.8
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
−
1
Reaction conditions: ammonia/allyl alcohol molar ratio 3:1, atmospheric pressure, GHSV 200 h
.
Table 5
Effect of the molar ratio of ammonia to alcohol on the performance of the Zn30Cr4.5/␥-Al2O3 catalyst.
Ammonia/alcohol
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3,5-Luti
FD
1
:1
1
1
97.2
99.1
100
50.2
65.2
58.4
57.7
7.2
7. 6
7.4
2.4
6.0
6.9
8.1
9.7
0.5
0.4
0.3
0.3
5.0
4.4
2.3
0.5
0.4
0.5
0.3
0.2
0.04
1.4
0.3
:3
:6
:12
1
100
0.7
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
◦
−1
.
Reaction conditions: reaction temperature 420 C, atmospheric pressure, GHSV: 200 h
an increase in the molar ratio of ammonia to allyl alcohol from 1:1
to 3:1 and reached a maximum of 65.2% at a ratio of 3:1. The yield
then decreased gradually with further increases in the molar ratio
of ammonia to allyl alcohol. It was also found that the acetonitrile
by-product yield increased with an increase in the molar ratio of
ammonia to allyl alcohol and acetonitrile became the second major
component of the reaction mixture when the molar ratio of ammo-
nia to allyl alcohol was higher than 3:1. The 3-picoline yield did
not change significantly when the molar ratio of ammonia to allyl
alcohol increased from 1:1 to 6:1. In general, an excess of ammo-
nia did not favor the desired reaction, possibly because the excess
ammonia molecules adsorbed on the acid centers of the catalyst,
weakening the acidity of the catalyst and consequently impeding
the condensation of aldehydes with ammonia to form imines, a key
step in the formation of nitriles as shown in Scheme 2 (Table 5).
Generally, GHSV is another very important parameter that must
be considered for a catalytic reaction occurring in a fixed-bed reac-
tor. Therefore, the influence of GHSV on the catalytic performance
was evaluated under atmospheric ammonia pressure and a molar
ratio of ammonia to allyl alcohol of 3:1 at 693 K. The results shown
in Table 6 indicated that the product yields did not change sig-
catalyst was on stream. Then both the conversion and selectivity
decreased gradually; the conversion and selectivity dropped to 88.3
and 52.06%, respectively, after the catalyst was on stream for 48 h.
The catalyst deactivation over time on stream was due to the forma-
tion of carbonaceous deposits inside the pores during the catalytic
run, which was confirmed by the TEM-EDX analysis and will be
described in detail below. The regeneration of the catalyst was per-
formed by continuously blowing air (STP 250 ml min ) into the
fixed-bed reactor at 723 K for 3 h on-line, and the reaction was run
once again using the regenerated catalyst. It was observed that both
the conversion of allyl alcohol and the selectivity toward propioni-
trile obtained with the regenerated catalyst were similar to those
obtained with the fresh catalyst. The results indicated that the car-
bonaceous deposits were removed, allowing the active sites to be
exposed to the reactants and, therefore, the performance of the
catalyst to be recovered after air regeneration.
−
1
3.3. Pathways for product generation
The composition of the products was analyzed by GC–MS. The
analysis results revealed that the collected product mixture was
composed of propionitrile, 3-picoline, acetonitrile, isobutyronitrile,
pyridine, 3,5-lutidine and formaldehyde. No amine was detected in
the mixture. Based on the analysis results, organic reaction princi-
ples [10] and related literature [11–14], the possible pathways for
the generation of the products are given in Scheme 2.
−1
−1
nificantly as GHSV was varied from 100 h to 300 h . However,
−1
when GHSV was increased above 400 h , the conversion of allyl
alcohol and propionitrile yield decreased considerably due to the
short residence time of the reactants on the catalyst.
The lifetime and regeneration of Zn30Cr4.5/␥-Al O on-line were
2
3
investigated, and the results are shown in Table 7. The conversion
of allyl alcohol was higher than 99%, and the selectivity toward
propionitrile was approximately 65% within the first 12 h that the
As shown in Scheme 2, allyl alcohol is first dehydrogenated
to acrolein, a key intermediate in the reaction. After the for-
mation of acrolein, there are two possible pathways that yield
Table 6
Effect of GHSV on the catalytic performance.
GHSV (h 1
−
)
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3,5-Luti
FD
1
2
3
4
5
6
00
00
00
00
00
00
100
100
99.5
93.9
91.2
89.6
66.0
65.8
66.4
58.3
52.0
45.9
7.5
7.3
6.1
7.8
7.1
3.4
7.2
9.0
8.2
5.8
8.4
3.4
0.3
0.5
0.4
0.5
0.1
0.5
3.3
4.1
4.5
3.2
1.4
1.8
0.6
0.4
0.5
0.4
0.5
0.1
0.1
0.2
0.1
0.2
0.2
0.3
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
Reaction conditions: reaction temperature 693 K, ammonia/allyl alcohol molar ratio 3:1, atmospheric pressure.

1
58
Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
Scheme 2. Pathways for the product formation from the amination of allyl alcohol.
propionitrile similar to those for the formation of butyronitrile
in the ethanol amination shown in Scheme 1. One is the con-
densation of acrolein with ammonia to give the intermediate
imine, which is then dehydrogenated to generate acrylonitrile and
hydrogen. Lastly, the carbon-carbon double bond in the acryloni-
trile molecule is hydrogenated with the hydrogen generated from
the dehydrogenation of the imine and alcohol to produce propi-
onitrile. The other pathway involves the hydrogenation of acrolein
Table 7
Catalytic performance for different times on stream.
Time on steam (h)
Conversion (%)
Selectivity (%)
PN
3-Pico
AN
Pyr
iso-BN
3,5-Luti
FD
4
99.0
100
98.3
93.3
88.3
99.7
65.2
65.5
57.8
57.0
52.1
65.2
7. 6
6.5
8. 8
12.0
10.8
9.1
6.9
6.5
7.5
3.1
3.3
4.6
0.5
0.6
0.6
0.9
1.1
1.0
4.4
5.7
3.5
3.7
1.5
4.3
0.5
0.9
0.9
0.9
0.7
0.5
1.4
0.9
0.3
0.5
0.7
1.0
1
2
3
4
1
2
4
6
8
2
a
PN: propionitrile; 3-Pico: 3-picoline; AN: acetonitrile; Pyr: pyridine; iso-BN: iso-butyronitrile; 3,5-Luti: 3,5-lutidine; FD: formaldehyde.
◦
−1
.
Reaction conditions: reaction temperature 420 C, ammonia/allyl alcohol molar ratio 3:1, atmospheric pressure, GHSV 200 h
a
Regenerated catalyst.

Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
159
Table 8
with the generated hydrogen to give propionaldehyde. Then the
propionaldehyde condenses with ammonia and is dehydrogenated
to give propionitrile. For the formation of acetonitrile, acrolein is
first hydrated to generate 3-hydroxy propionaldehyde. Then 3-
hydroxy propionaldehyde undergoes inverse aldol condensation to
yield acetaldehyde and formaldehyde. The amination of acetalde-
hyde to give acetonitrile is described elsewhere [5]. The previously
produced propionaldehyde can condense with formaldehyde to
generate 2-methyl acrolein, which is hydrogenated to 2-methyl
propionaldehyde. Then 2-methyl propionaldehyde reacts with
ammonia to give isobutyronitrile, which is similar to the reaction
for the formation of acetonitrile. The formation of pyridine bases
occurs as described in the literature [11–13] and is not discussed
here.
Textural properties of Zn30/␥-Al2O3 and fresh, used and regenerated Zn30Cr4.5/␥-
Al2O3.
SBETa (m2
g
−1
)
Vb (cm3
g
−1
)
dpc (nm)
Catalyst
Zn30/␥-Al O
Zn30Cr4.5/␥-Al2O3 (fresh)
Zn30Cr4.5/␥-Al2O3 (used)
2
3
99.61
99.86
54.16
97.77
0.316
0.268
0.129
0.239
12.68
8.89
9.51
9.76
Zn30Cr4.5/␥-Al2O3 (regenerated)
a
BET surface area.
BJH cumulative desorption pore volume.
Mean pore diameter = 4 V/SBET.
b
c
Cr doping of Zn30/␥-Al O3 decreased the size of the ZnAl O4 crys-
tallites. These results were confirmed by the TEM images (Fig. 2).
2
2
ZnAl O4 particles with diameters of 2–5 nm and 5–15 nm were
2
3.4. Catalyst characterization
observed as dark spots in the TEM images of Zn30Cr4.5/␥-Al2O3
and Zn30/␥-Al O , respectively. These particles were dispersed on
2
3
To elucidate the active species and deactivation mechanism of
the ␥-Al2O3 surface. Thus, both the XRD and TEM data supported
the conclusion that the average ZnAl2O4 particle size was smaller
on Zn30Cr4.5/␥-Al2O3 than on Zn30/␥-Al2O3, suggesting that Cr
doping can suppress the growth of ZnAl2O4 grains. A similar phe-
nomenon in which Cr doping of supported ZnO decreased the ZnO
grain size was reported in the literature [21]. The XRD and TEM
characterization in combination with the catalytic test results indi-
cated that the small ZnAl2O4 particle size was favorable for the
reaction. Meanwhile, the diffraction patterns of the used and regen-
erated Zn30Cr4.5/␥-Al2O3 samples did not change at all compared
to those of the fresh sample, which indicated that the ZnAl2O4
active species was stable during the catalysis and regeneration pro-
cedure. Therefore, the deactivation of the catalyst was not due to
the decomposition or transformation of the ZnAl2O4 active species.
Comparison of the TEM images of the fresh, used and regenerated
Zn30Cr4.5/␥-Al2O3 samples showed that some substances covered
the surface of the used samples. After regeneration by on-line com-
bustion, the substances disappeared. Therefore, the deactivation of
the catalyst can be attributed to the carbonaceous deposits gen-
erated by the chemical adsorption of alkaline substances during
the catalytic run. The EDX analysis supported this conclusion with
measured carbon contents of 23.6% and 46.6% at points 1 and 2,
respectively, in Fig. 2e.
the catalyst, the catalyst was characterized by XRD, XPS, TEM-
EDX and N2 adsorption–desorption, and IR spectra of adsorbed
pyridine were collected. Fig. 1 shows the diffraction patterns
of the fresh, used and regenerated Zn30Cr4.5/␥-Al O3 samples,
2
Zn30Cr4.5/␥-Al O and ␥-Al O . All the patterns except for that of
2
3
2
3
␥
[
-Al O demonstrated the presence of a ZnAl O4 crystalline phase
2
3
2
15–17]. No other crystalline phase was observed.
As reported in the literature [18–20], the heterogeneous
ZnAl O catalyst is active in many reactions, such as dehydration,
2
4
hydrogenation, dehydrogenation, dehydrogenative condensation
of normal alcohols, methylation of phenolic compounds and
N-alkylation of 2-hydroxypyridine with methanol. It can there-
fore be concluded that ZnAl O4 is the active species for the
2
dehydrogenation–hydrogenation of the intermediate acrolein
imine to propionitrile. It was observed that the ZnAl O4 peaks
2
of the fresh Zn30Cr4.5/␥-Al O3 sample were weaker and broader
2
than those of the Zn30/␥-Al O3 sample. In general, the full width
2
at half maximum of the XRD peak is related to the particle size
of crystal materials. The width increases as the size of the crystal-
lites decreases. The average diameters of the ZnAl O4 crystallites
2
in Zn30/␥-Al O and Zn30Cr4.5/␥-Al O3 calculated using the Scher-
2
3
2
rer formula were 5.2 nm and 7.1 nm, respectively. Therefore, the
To strengthen this assumption, the porous structure of the cat-
alyst was further analyzed with nitrogen adsorption experiments.
Table 8 presents the specific surface areas and the range of the pore
structural parameters of Zn30/␥-Al O and fresh, used and regener-
2
3
ated Zn30Cr4.5/␥-Al O . As expected, fresh Zn30Cr4.5/␥-Al O had a
2
3
2
3
slightly lower surface area than Zn30/␥-Al O due to the Cr doping.
2
3
After the Zn30Cr4.5/␥-Al O3 catalyst was on stream for 48 h, the
2
2
−1
2
−1
surface area decreased from 99.15 m g to 54.16 m g , corre-
sponding to a decrease in the pore volume from 0.276 cm g
to 0.129 cm g . The nitrogen adsorption and EDX results both
indicated that carbon was deposited in the catalyst pores. After
3
−1
3
−1
2
−1
to
regeneration, the surface area increased from 54.16 m g
2
−
1
3
−
1
9
0
7.77 m g , and the pore volume increased from 0.129 cm g to
3
−1
.239 cm g , which indicated that the textural properties of the
catalyst, and thus the catalytic performance, were largely recov-
ered.
The XPS technique is much more sensitive than XRD for the
analysis of surface oxides. Therefore, the surface compositions of
the fresh, used and regenerated samples were determined by XPS.
Figs. 3 and 4 show the analysis results. The results revealed that
the Zn 2p3/2 binding energies of Zn30/␥-Al O3 and Zn30Cr4.5/␥-
2
Al O were 1022.16 and 1021.96 eV, respectively, indicating that Zn
2
3
was present as ZnAl O4 in the catalysts [22]. The Zn 2p3/2 binding
2
energies of the used and regenerated Zn30Cr4.5/␥-Al O3 samples
2
Fig. 1. XRD patterns of the catalyst samples. (a) Zn30/␥-Al2O3; (b) Zn30Cr4.5/␥-Al2O3
were similar to that of the fresh one, indicating that the catalyst
was stable during the catalytic run as also revealed by the XRD
(
Fresh); (c) Zn30Cr4.5/␥-Al2O3 (Used); (d) Zn30Cr4.5/␥-Al2O3 (Regenerated); (e) ␥-
Al2O3.

1
60
Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
Fig. 2. TEM micrographs of the catalyst samples and EDX spectra at points 1 and 2. (a) Zn30/␥-Al2O3; (b) Zn30Cr4.5/␥-Al2O3 (Fresh); (c) Zn30Cr4.5/␥-Al2O3 (Used); (d)
Zn30Cr4.5/␥-Al2O3 (Regenerated); (e) EDX spectra at points 1 and 2.
analysis. In addition, the XPS spectra of the fresh, used and regen-
Zn30Cr4.5/␥-Al O3 catalysts. The results are shown in Fig. 5. A
2
−
1
erated Zn30Cr4.5/␥-Al O3 catalyst samples exhibited peaks near
peak at 1450 cm
ascribed to pyridine adsorbed on Lewis acid
2
5
77 eV corresponding to the Cr 2p3/2 binding energy. This bind-
sites [24–26] appeared in the spectra of Zn30/␥-Al O3 and the
2
3+
ing energy revealed the presence of Cr , most likely in the form of
ZnCr O4 [23].
fresh Zn30Cr4.5/␥-Al O3 catalyst. After the Zn30Cr4.5/␥-Al O3 cat-
2
2
−
1
alyst was used, the peak at 1450 cm nearly disappeared, but it
reappeared after regeneration. It therefore was concluded that the
intermediate condensation of acrolein or propionaldehyde with
2
The IR spectra of adsorbed pyridine were collected to determine
the acidity of the Zn30/␥-Al O3 and fresh, used and regenerated
2

Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
161
ammonia to form imines occurred on the Lewis acid sites of the cat-
alyst. Obviously, an excess of ammonia over allyl alcohol impeded
the reaction due to the adsorption of ammonia molecules on the
−
1
acid sites. The disappearance of the peak at 1450 cm
in the
spectrum of the used sample indicated that alkaline carbonaceous
deposits were adsorbed on the Lewis acid sites and blocked the
ZnAl O active species, which led to the deactivation of the catalyst.
2
4
After regeneration, the carbonaceous deposits were removed, and
the acid sites and ZnAl O active species were exposed to reactants
2
4
again. Therefore, the activity of the catalyst was recovered.
4. Conclusions
A Zn30Cr4.5/␥-Al O3 bimetallic catalyst was prepared and
2
showed good performance for the amination of allyl alcohol to
propionitrile. During the catalysis, the hydrogen generated dur-
ing the dehydrogenation of the alcohol and imine acted as an
in situ source for the hydrogenation of the carbon-carbon double
bond. The parameters that affect the catalyst performance were
studied thoroughly, and an optimized process for synthesizing pro-
pionitrile from allyl alcohol and ammonia over the catalyst was
obtained. Under the optimized conditions, the propionitrile yield
was greater than 65%. The characterization results indicated that
the dehydrogenation reaction mainly occurred on the Lewis acid
Fig. 3. Zn 2p and Cr 2p XPS spectra of the samples. (a) Zn30/␥-Al2O3; (b) Zn30Cr4.5/␥-
Al2O3 (Fresh); (c) Zn30Cr4.5/␥-Al2O3 (Used); (d) Zn30Cr4.5/␥-Al2O3 (Regenerated).
sites and revealed that ZnAl O4 is the active species for the cou-
2
pled dehydrogenation-hydrogenation reactions. Chromium doping
of the ␥-Al O -supported zinc catalyst Zn /␥-Al O3 decreased
2
3
30
2
the size of the ZnAl O4 crystallites, which was favorable for the
2
dehydrogenation–hydrogenation reactions. The characterization
results also revealed that the catalyst deactivation was caused by
carbon deposition on the catalyst during the catalytic run. The cat-
alyst could be reactivated by blowing air into the reactor at a high
temperature.
Acknowledgements
We thank the financial support of the National Natural Science
Foundation of China (Grant no. 20976034), the Natural Science
Foundation for Young Scientists of Hebei Province, China (Grant no.
B2009000009) and the Foundation for Innovative Talents of Hebei
Province, China (Grant no. CPRC012).
Fig. 4. High-resolution XPS spectra of Zn30Cr4.5/␥-Al2O3. (a) Fresh; (b) Used; (c)
Regenerated.
References
[
[
1] S. Iida, H. Togo, Tetrahedron 63 (2007) 8274–8281.
2] Y. Hu, J. Cao, J. Deng, B. Cui, M. Tan, J. Li, H. Zhang, React. Kinet. Catal. Lett. 106
(
2012) 127–139.
[
[
3] C. Feng, F. Guo, Y. Zhang, J. Zhao, Fine Chemicals (in Chinese) 27 (2010) 567–578.
4] Y. Zhang, Y. Zhang, C. Feng, C. Qiu, Y. Wen, J. Zhao, Catal. Commun. 10 (2009)
1
454–1458.
5] C. Feng, Y. Zhang, Y. Zhang, Y. Wen, J. Zhao, Catal. Lett. 141 (2011)
68–177.
6] D. Zhang, Y. Zhang, Y. Wen, K. Hou, J. Zhao, Chem. Eng. Res. Des. 89 (2011)
147–2152.
7] S.J. Kulkarni, R.R. Rao, M. Subrahmanyam, A.V.R. Rao, Appl. Catal. A 113 (1994)
–7.
8] K. Rajiv, J.P. Narahar, C.G. Moreshwar, N.P. Suresh, A. Ashutshot, V.P. Kumar,
S.K. Samir, WO2005000816 (2005).
9] Y. Zhang, W. Xu, J. Zhao, RSC Adv. 2 (2012) 6590–6598.
[
[
[
[
[
1
2
1
[
[
10] F. Carey, R. Sundberg, Adv. Org. Chem., Springer US (2007) 63–214.
11] A.J. Castro, D.K. Brain, H.D. Fisher, R.K. Fuller, J. Org. Chem. 19 (1954)
1444–1448.
[
[
[
[
12] J.R. Calvin, R.D. Davis, C.H. McAteer, Appl. Catal. A 285 (2005) 1–23.
13] Y. Higashio, T. Shoji, Appl. Catal. A 260 (2004) 251–259.
14] J. Barrault, Y. Pouilloux, Catal. Today 37 (1997) 137–153.
15] G. Lakshminarayana, L. Wondraczek, J. Solid State Chem. 184 (2011)
1931–1938.
[
[
[
16] X. Duan, D. Yuan, Z. Sun, H. Sun, D. Xu, M. Lv, J. Cryst. Growth 252 (2003) 4–8.
17] S. Farhadi, S. Panahandehjoo, Appl Catal. A 382 (2010) 293–302.
Fig. 5. IR spectra of pyridine adsorbed on the catalysts. (a) Zn30/␥-Al2O3; (b)
Zn30Cr4.5/␥-Al2O3 (Fresh); (c) Zn30Cr4.5/␥-Al2O3 (Used); (d) Zn30Cr4.5/␥-Al2O3
18] H. Grabowska, M. Zawadzki, L. Syper, Appl. Catal.
26–232.
A 314 (2006)
(
Regenerated).
2

1
62
Y. Zhang et al. / Applied Catalysis A: General 467 (2013) 154–162
[
19] J. Wrzyszcz, M. Zawadzki, J. Trawczy n´ ski, H. Grabowska, W. Mi s´ ta, Appl. Catal.
A 210 (2001) 263–269.
[23] F. Simard, U.A. Sedran, J. Sepúlveda, N.S. Fígoli, H.I. de Lasa, Appl. Catal. A 125
(1995) 81–98.
[
20] H. Grabowska, W. Mi s´ ta, J. Trawczy n´ ski, J. Wrzyszcz, M. Zawadzki, Appl. Catal.
[24] C.E. Volckmar, M. Bron, U. Bentrup, A. Martin, P. Claus, J. Catal. 261 (2009) 1–8.
[25] R. Barthos, F. Solymosi, J. Catal. 235 (2005) 60–68.
[26] E. Modrogan, M.H. Valkenberg, W.F. Hoelderich, J. Catal. 261 (2009)
177–187.
A 220 (2001) 207–213.
21] H.H. Hng, P.L. Chan, Ceram. Int. 35 (2009) 409–413.
22] Y. Wu, J. Du, K.-L. Choy, L.L. Hench, J. Guo, Thin Solid Films 472 (2005) 150–156.
[
[