Heterogeneous catalytic synthesis of quinoline compounds from aniline and C1-C4 alcohols over zeolite-based catalysts

Article abstract of DOI:10.1039/c7ra08442k

The synthesis of quinolines from aniline and a C₁-C₄ alcohol was conducted under gas-phase reaction conditions over a series of zeolite-based catalysts. The texture and acid properties of catalysts were characterized by XRD, FT-IR, BET and NH₃-TPD techniques. It was found that the total yield of quinolines was positively related to the relative content of Lewis acid sites of the catalyst. Among others, the ZnCl₂/Ni-USY-acid catalyst possessed the best performance. Over this catalyst, the reactions of aniline and most of the alcohols provided a 42.3-79.7% total yield of quinolones under mild conditions, however, those of aniline and methanol, ethanol and iso-propanol predominantly led to N-alkylanilines. Furthermore, the reaction pathways for synthesizing quinolines via aniline reacting with polyhydric alcohols or monohydric alcohols was proposed in our work.

Full text of DOI:10.1039/c7ra08442k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Heterogeneous catalytic synthesis of quinoline
compounds from aniline and C –C alcohols over
1
4
Cite this: RSC Adv., 2017, 7, 48275
zeolite-based catalysts†
Chen Huang, An Li and Zi-Sheng Chao *ab
a
a
The synthesis of quinolines from aniline and a C
conditions over a series of zeolite-based catalysts. The texture and acid properties of catalysts were
characterized by XRD, FT-IR, BET and NH -TPD techniques. It was found that the total yield of
quinolines was positively related to the relative content of Lewis acid sites of the catalyst. Among others,
the ZnCl /Ni-USY-acid catalyst possessed the best performance. Over this catalyst, the reactions of
1 4
–C alcohol was conducted under gas-phase reaction
3
2
Received 31st July 2017
Accepted 9th October 2017
aniline and most of the alcohols provided a 42.3–79.7% total yield of quinolones under mild conditions,
however, those of aniline and methanol, ethanol and iso-propanol predominantly led to N-alkylanilines.
Furthermore, the reaction pathways for synthesizing quinolines via aniline reacting with polyhydric
alcohols or monohydric alcohols was proposed in our work.
DOI: 10.1039/c7ra08442k
rsc.li/rsc-advances
to synthesize quinolines. However, these raw materials are
usually toxic, expensive, and particularly, prone to polymeriza-
Introduction
22,24–26
Quinolines (quinoline and substituted quinolines) are a type of tion during reaction.
Relative to carbonyl compounds,
important heterocyclic compound, which can be widely used as small molecular alcohols are hard to polymerize, much more
pharmaceuticals, fungicides, herbicides, corrosion inhibitors eco-friendly and cheaper, since they can be alternatively man-
1
–8
and functional chemicals.
involving liquid phase reaction, such as, Skraup,
Various traditional routes ufactured from the fermentation of biomass materials besides
9,10
Doebner- the chemical synthesis from fossil-based raw materials. This
and Ptzinger enables small molecular alcohols to be potential substitutes of
11,12
13,14
15,16
Miller,
Friedl ¨a nder,
Combes,
17
methods, have been developed and utilized in the synthesis of carbonyl compounds in the synthesis of quinolines. In fact, few
quinolines. However, these routes all suffered from a few papers dealt with the application of alcohol as feedstock to
2
0,21,23
drawbacks, e.g., anhydrous reaction condition, volatile organic synthesizing quinolines via gas phase reaction route.
For
21
solvent, expensive or toxic feed, corrosive and/or costly homo- examples, Campanati et al. obtained a 41% yield of 2-methyl-8-
geneous catalyst, prolonged reaction time, tedious work-up ethylquinoline from the reaction of ethylene glycol and 2-eth-
9
–13,15,18,19
ꢀ
procedure and difficult recycling of the catalyst.
contrast, gas phase reaction route for the synthesis of quino- Reddy et al. reported a 65% yield of quinoline via the reaction
In ylaniline at 330 C over K10 montmorillonite catalyst, while
23
20–25
ꢀ
lines employs an appropriate heterogeneous catalyst,
which of glycerol and aniline at 425 C over mixed metal oxides cata-
can solve effectively the above issues associated with liquid lyst in the presence of air. However, such kind of attempts has
phase reaction route, and thus, it is receiving more and more been very limited, excepting the above-mentioned reports in the
attentions.
literature. Therefore, an exploration on the utilization of various
In the gas phase reaction route, up till now, small molecular types of alcohols as feedstock in reacting with aniline to
aldehydes or carbonyl compounds containing active hydrogen generate quinolines is of signicant importance.
on a-carbon, such as, acetaldehyde, propionaldehyde and
The catalysts for the synthesis of quinolines via gas phase
butyraldehyde, were most frequently employed as the raw reaction route reported in the literature included mainly metal
materials, due to their high activity in the reaction with aniline halide clusters, K10, amorphous silica–alumina and mixed
20,22–25
metal oxides.
These catalysts usually presented low to
middle activities for the generation of quinolines but low
College of Chemistry and Chemical Engineering, Hunan University, Changsha, catalytic stabilities, due to their poor acid performance and
a
4
10082, China. E-mail: zschao@yahoo.com; Fax: +86-731-88713257; Tel: +86-731-
textural properties. Zeolites represent an important class of
porous materials that have found numerous applications in
catalysis, especially in the pyrolysis of petroleum to gasoline,
because of their unique and tunable acid performance, large
stability of framework and well-dened microporous structure.
8
8713257
b
College of Materials Science and Engineering, Changsha University of Science and
Technology, Changsha, 410110, China
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c7ra08442k
1
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 48275–48285 | 48275

View Article Online
RSC Advances
Paper
Y-Type zeolite was employed as catalyst in the synthesis of
substituted quinolines, showing an appreciable catalytic
24
activity. However, zeolite catalysts usually had to face an issue
of transport limitation caused by the slow diffusion rates of
27,28
reactant and bulky product in microporous.
In this respect,
the issue is more serious, since quinolines are relatively bulky,
relative to gasoline molecules. To reduce the diffusion limita-
tion, mesostructured ultra-stable Y zeolite (USY) was more
29–31
frequently employed as catalyst in a series of reactions,
due
to their modied acid performance and large average pore size
and therefore high stability and catalytic activity. Mesoporous
2
9,31–34
USY zeolite could be prepared by acid or alkali treatment,
30,32,35,36
steam dealumination,
or in situ synthesis with meso-
37
porous structure-directing agent. Among others, the acid-
treated method of USY zeolite was considered as the most
simple and efficient one, and thus, accepted by many
researchers. However, to the best of our knowledge, no report
on the employment of acid-treated USY zeolite as catalyst for the
synthesis of substituted quinolines had been documented in
the literature.
In this paper, we reported for the rst time the systematical
studies on the synthesis of quinolines from aniline and a C –C
1 4
Fig. 1 XRD patterns for the unmodified and modified USY zeolite
catalysts: (a) USY; (b) USY-acid; (c) Ni-USY-acid; (d) ZnCl
2
/Ni-USY-acid
ꢀ
alcohols over modied USY catalysts via gas phase route. It was (the inset is the enlargement of the patterns for 2q ¼ 9.5–12.5 ).
claried that the transition metal-modied mesoporous USY
could be used as an efficient catalyst for the reaction of various
alcohols and aniline to quinolines, exhibiting not only a high
catalytic activity but also a good catalyst stability. Furthermore,
the related reaction mechanisms, dependent on the type of
alcohols, were also proposed in this paper.
Fig. 2 shows FT-IR the spectra of various catalysts. The bands
at ca. 1188, 1064 and 824 cm over the USY catalyst are char-
ꢁ
1
acteristic of TO tetrahedron units, which can be assigned to the
4
external asymmetric, internal asymmetric and external
symmetric stretching vibrations of Si–O–T linkages for TO
4
(T ¼
39–41
Si and/or Al) tetrahedral.
These bands are known to shi
Results and discussion
Fig. 1 shows the XRD patterns for various catalysts. One can see
that all the catalysts exhibit the characteristic diffraction peaks
for the USY zeolite, and the Ni-USY-acid and ZnCl
2
/Ni-USY-acid
catalysts do not shown the diffraction peaks for Ni and ZnCl
2
species. The intensities of diffraction peaks for the USY-acid, Ni-
USY-acid and ZnCl /Ni-USY-acid catalysts are all lower than for
2
the parent USY. It shows that the original structure of USY
zeolite is retained, though suffering from a destroy to some
extent, aer acid treatment, and Ni and ZnCl species possess
2
very small dimension and have been highly dispersed on USY
zeolite. From the enlargement of diffraction peaks, e.g., those
ꢀ
ꢀ
ranging from 9.5 to 12.5 , one can nd that the diffraction
peaks shi slightly towards higher angel over the USY-acid
catalyst but obviously towards lower angel over both the
Ni-USY-acid and ZnCl /Ni-USY-acid catalysts, relative to the
2
parent USY zeolite. As is well known, the difference in the
diameters of framework T atoms causes theoretically the
expanse or contraction of zeolite framework and thus the whole
38
shi of characteristic diffraction peaks of zeolite, while the
2
+
2+
diameters of the 4-coordinated ions Zn (74 pm), Ni (69 pm),
3
+
4+
Al (53 pm) and Si (40 pm) decrease sequentially. It is
therefore deduced that the partial dealumination occurs during
acid treatment of USY zeolite and a proportion of Ni and Zn
ions are inserted into the framework during the Ni ion exchange
29
Fig. 2 FT-IR spectra for the unmodified and modified USY zeolite
catalysts: (a) USY; (b) USY-acid; (c) Ni-USY-acid; (d) ZnCl
acid.
2
/Ni-USY-
2
and ZnCl loading over USY-acid zeolite.
48276 | RSC Adv., 2017, 7, 48275–48285
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
Fig. 3 shows the N
2
adsorption–desorption isotherms for
various catalysts, and the corresponding textural properties are
summarized in Table 1. One can see from Fig. 3 that, for all the
catalysts, the isotherms rise steeply at very low relative pressure,
showing the presence of micropores. In addition, the hysteresis
loop at P/P
acid and ZnCl
0
> 0.4 is obviously present for the USY-acid, Ni-USY-
/USY-acid catalysts but very unclear for the USY
2
catalyst. It shows that mesopores are generated via the acid
treatment of USY zeolite, of which the pore sizes range from 1.5
to 6.0 nm and possess a most possible value at ca. 2.0 nm (see
the inset in Fig. 3). From Table 1, one can nd that, compared to
the USY catalyst, the USY-acid catalyst possesses a larger mes-
opore volume (Vmeso), total volume (Vtotal) and external surface
area (Sext) but smaller micropore volume (Vmic), micropore
surface area (Smic), micropore size (Dmic) and specic surface
area (SBET). Besides, mesopores with an average size of 2.0 nm
are determined for the USY-acid catalyst but not for the USY
catalyst. It indicates that the generation of mesopores is a result
44,45
of the partial dealumination of zeolite by acid treatment,
and also, a proportion of extracted Al species may have occupied
the micropores of zeolite. Compared to the USY-acid, the pore
volumes (Vmeso, Vmic and Vtotal), surface areas (Sext, Smic and
Fig.
2
3 N -adsorption/desorption isotherms and BJH pore-size
distribution curves for the unmodified and modified USY zeolite
catalysts: (a) USY; (b) USY-acid; (c) Ni-USY-acid; (d) ZnCl
acid.
2
/Ni-USY-
S
BET) and pore size (Dmic and Dmes) all decrease sequentially
over the Ni-USY-acid and ZnCl /Ni-USY-acid. This can be due to
the fact that, during the Ni ion exchange and ZnCl loading,
towards higher wavenumbers with increasing the Si/Al molar a proportion of Ni and Zn ions enter into the pores and
2
2
2
+
2+
ꢁ
1
2+
2+
ratio of zeolite. Besides, the band at 455 cm over the USY another proportion of Ni and Zn ions are incorporated into
catalyst is attributed to the internal T–O bending vibration of the framework of zeolite.
4
2
TO
4
tetrahedron, and this band is nearly unaffected by the
Fig. 4 shows the NH
3
-TPD proles for various catalysts. The
) and integral area (A ) of the
Si–O–T stretching vibrations at 1188, 1064 and 824 cm shi peaks, which correspond respectively to the strength and
Si/Al molar ratio of zeolite. Compared to the USY catalyst, the temperatures at the maximum (T
i
i
ꢁ1
ꢁ1
respectively to 1192, 1068 and 825 cm over the USY-acid concentration of acid sites, are summarized in Table 2. The T
catalyst, indicating that the partial dealumination of USY
and T peaks are present over all the catalysts, and they have
zeolite occurs during the its acid treatment. However, the above a maximum at ca. 141–147 C, 265–280 C and 330 C, respec-
Si–O–T stretching vibrations shi respectively to 1178, 1059 and tively. The T peak is attributed to the weak adsorption of
/Ni-USY-acid catalysts. ammonia molecule over the surface terminal silanol (Si–OH)
1
,
T
3
4
ꢀ
ꢀ
ꢀ
1
ꢁ1
8
20 cm , over the Ni-USY-acid and ZnCl
2
43
46
Szostak et al. reported that the incorporation of gallium group, while the T
4
peak can be assigned to the surface
replaced the framework aluminum of zeolite and in turn led to Br ¨o nsted acid site, i.e., bridged hydroxyl group (Si–OH–Al), in
47
the signicant shi of Si–O–T stretching vibrations towards Y-type zeolite. It is known that the extra-framework Al species
lower wavenumber, due to the change in the reduced mass of behave as Lewis acid site and possess a weaker strength than
47
the Si–O–T harmonic oscillator. Accordingly, the above varia- the bridged hydroxyl group in Y-type zeolite. Accordingly, the
tions in the Si–O–T stretching vibrations can be due to the
peak can be attributed to the Lewis acid site associated with
incorporations of a proportion of Ni and/or Zn ions into the the extra-framework Al specie. Over various catalysts, the area of
T
3
framework of USY-acid zeolite. The above results are consistent
with those derived from XRD characterization.
T
4
peak exhibits an order USY > USY-acid > Ni-USY-acid z
ZnCl /Ni-USY-acid and that of T displays an inverse order,
2
3
Table 1 Textural properties for the parent and modified USY zeolite catalysts
a
2
ꢁ1
a
2
ꢁ1
a
2
ꢁ1
b
3
ꢁ1
b
3
ꢁ1
b
3
ꢁ1
c
c
Catalyst
S
BET (m g
)
S
mic (m g
)
S
ext (m g
)
V
mic (cm g
)
V
meso (cm g
)
V
total (cm g
)
D
micro (nm)
D
meso (nm)
USY
USY-acid
Ni-USY-acid
704
664
631
618
536
512
505
86
0.21
0.19
0.18
0.17
0.09
0.19
0.18
0.16
0.30
0.38
0.36
0.33
0.74
0.73
0.73
0.70
—
128
119
104
2.08
2.04
2.01
2
ZnCl /Ni-USY-acid 609
a
b
S
BET, Sext and Smic: specic surface area, external surface area and micropore surface area, respectively, and SBET ¼ Sext + Smic
.
V
total, Vmic and
c
V
meso: total pore volume, micropore volume and mesopore volume, respectively.
D
micro and Dmeso: micropore pore size and average mesopore
pore size, respectively.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 48275–48285 | 48277

View Article Online
RSC Advances
Paper
ꢀ
2
ca. 185 C has been identied over the ZnCl /Ni-USY-acid
catalyst but not over the other catalysts, and this peak can be
assigned to the new Lewis acid site related to the –O–Zn–Cl
groups. It is observed that the desorption temperature for the T2
peak is lower than that for the T
3
peak, and also, the desorption
/Ni-USY-acid
temperature for the T peak is lower over the ZnCl
3
2
and Ni-USY-acid catalysts than over the USY-acid and USY
catalysts. This can be due to the facts that the Ni ion exchanged
over the zeolite can behave as a kind of Lewis acid site, and thus,
it contributes to the T peak, besides the extra-framework Al
3
species over the ZnCl /Ni-USY-acid and Ni-USY-acid catalysts.
2
2
+
2+
3+
Also, the electronegativity for the Zn , Ni and Al increases
sequentially, and accordingly, the strength for the correspond-
ing Lewis acid sites increases in order. Besides the above T
1
–T
4
ꢀ
peaks, a new peak T
present over the ZnCl
5
with its maximum at ca. 508 C is also
2
/Ni-USY-acid, Ni-USY-acid and USY-acid
catalysts but almost absent over the USY catalyst. It was re-
ported that the partial removal of the Al atoms in the second-
neighbor positions increased signicantly the strengths of
49
50
Bronsted acid sites in the Y-typed zeolite. Liu et al. found that
the strength of acid sites in the micropores or small cages was
stronger than that in the mesopores or larger cavities over the
3
Fig. 4 NH -TPD profiles for the unmodified and modified USY zeolite
catalysts: (a) USY; (b) USY-acid; (c) Ni-USY-acid; (d) ZnCl
acid.
2
/Ni-USY-
51
dealuminated USY zeolite. Besides, Fritz et al. and Beyerlein
52
et al. found that the strength of the bridged Br ¨o nsted acid sites
Si–OH–Al groups) were increased obviously due to the induc-
tive effect of extra-framework aluminum species. Accordingly,
(
while the area of T
1
peak has an order USY z USY-acid >
Ni-USY-acid > ZnCl /Ni-USY-acid. The result can be due to the
2
the peak T can be attributed to the strong Bronsted acid sites,
fact that, rstly, the acid treatment and Ni ion exchange both
result in the partial dealumination of zeolite, and this decreases
the concentration of Si–OH–Al groups but increases that of
extra-framework Al species over the USY-acid and Ni-USY-acid
catalysts, relative to the USY catalyst; secondly, the Ni ion
exchange leads to the replacement of hydrogen ion in the
surface Si–OH–Al, and also, to some extent, that in the surface
Si–OH group, and this decreases both the concentrations of the
two kinds of hydroxyl groups over the Ni-USY-acid catalyst,
relative to the USY-acid catalyst; thirdly, it was reported that
5
which are generated due to the removal of some adjacent
framework Al species, and/or the inducement of the normal
Bronsted acid sites, particularly those in the micropores, by the
extra-framework Al species. Compared that over the USY-acid
5 2
catalyst, the areas of peak T over the Ni-USY-acid and ZnCl /
Ni-USY-acid catalysts decrease, particularly for the latter one.
This may be due to the partial blockage of the micropores in
zeolite by a proportion of Ni and Zn species, which can be evi-
denced by the N -physisorption determination. From Table 2,
2
one can also see that the ratio for the total concentration of
Lewis acid sites to the total Bronsted acid sites over various
catalysts, i.e., (A + A )/(A + A ), exhibits an order ZnCl /Ni-USY-
2 3 4 5 2
ZnCl could react with the surface hydroxyl groups of Mont-K 10
2
clay to form the Lewis acid sites (–O–Zn–Cl) by the thermal
48
activation. Accordingly, the surface hydroxyl groups (Si–OH
and Si–OH–Al) over the Ni-USY-acid catalyst can react with
acid > Ni-USY-acid > USY-acid > USY.
Table 3 shows the results for the reaction of aniline and
n-propanol over various catalysts. One can see that 2-ethyl-3-
methyl-quinoline (2E-3MQ) over all the catalysts and 2,3-
dimethylquinoline (DMQ), 2-ethylquinoline (2-EQ) and
ZnCl
groups, and also, would generates the new Lewis acid site (–O–
Zn–Cl) over the ZnCl /Ni-USY-acid catalyst, relative to the Ni-
USY-acid catalyst. In fact, a new peak T with its maximum at
2
, and this decreases the concentrations of surface hydroxyl
2
2
3
Table 2 NH -TPD results over the parent and modified USY zeolite catalysts
a
ꢀ
b
ꢁ1
T
T
i
( C) and A (mmol g ) for various desorption peaks
i
Catalyst
1
A
1
T
2
A
2
T
3
A
3
T
4
A
4
T
5
A
5
2 3 4 5
(A + A )/(A + A )
USY
USY-acid
Ni-USY-acid
147
143
141
144
2.62
2.61
2.36
1.54
—
—
—
185
—
—
—
1.77
280
280
265
265
0.08
0.17
0.20
0.21
330
330
330
330
1.30
0.40
0.16
0.05
—
—
0.06
0.16
0.25
508
508
508
0.67
0.65
0.05
ZnCl
2
/Ni-USY-acid
19.80
a
ꢀ
b
T
i
refers to the temperature ( C) at the maximum of desorption peak i.
A
i
refers to the integral area of desorption peak i, and it also means the
ꢁ1
concentration (mmol g ) of acid sites corresponding to the desorption peak i.
48278 | RSC Adv., 2017, 7, 48275–48285
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
a
Table 3 Results for the reactions of aniline and n-propanol over various catalysts
c
S
i
(mol%)
b
f
g
h
i
j
k
l
d
e
Entry Catalyst
Carrier gas c (mol%) 2E-3MQ 2,3-DMQ 2-EQ 2-MQ NPA DNPA Others
S
QS (mol%)
Y
QS (mol%)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
SiO
Ni/SiO
ZSM-5
Fe-ZSM-5
Ni-ZSM-5
HY
Fe-Y
Ni-Y
USY
Ni-USY
USY-acid
Zn-USY-acid
Cu-USY-acid
Co-USY-acid
Fe-USY-acid
Ni-USY-acid
2
–Al
2
O
3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
51.3
55.4
42.5
45.1
44.4
41.3
48.6
50.3
55.3
58.4
58.1
60.2
63.3
61.4
62.3
68.6
97.6
86.5
7.6
8.4
9.7
0
0
0
0
0
0
0
0
0
0
1.1
1.1
1.4
2.9
1.5
2.1
2.3
1.4
1.3
0.2
0.1
0
0
0
0
0
0
1.3
1.4
3.4
2.1
2.1
1.3
1.9
2.0
1.5
0.7
0
85.3
86.2
71.4
74.5
72.3
63.6
62.7
65.6
43.6
38.6
41.5 3.1
43.1 3.3
45.0 2.1
41.4 2.5
38.6 2.6
40.1 1.9
9.3 3.1
0
0
0
0
0
0
0
0
0
0
7.1
5.4
18.9
13
14.6
14.7
9.6
7.6
8.4
9.7
3.9
4.7
4.1
5.6
5.8
2
–Al
2
O
3
11.2
11.3
19.4
24.2
23.1
29.1
33.5
35.1
36.1
34.2
36.7
42.8
42.5
62.1
54.6
1.3
1.8
2.3
2.2
2.3
8.3
9.5
9.7
9.3
10.6
11.4
11.3
11.6
19.4
13.1
12.5
13.1
21.7
27.7
27.9
41.9
46.5
49.8
48.2
48.8
52.4
57.0
56.1
81.7
67.8
9.0
13.5
14.0
23.2
27.2
28.9
29.0
30.9
32.2
35.5
38.5
79.7
58.6
6.5
14.5
14.9
5.6
5.4
4.1
3.7
1.8
1.9
5.9
0
1
2
3
4
5
6
7
8
ZnCl
ZnCl
2
/Ni-USY-acid
2
/Ni-USY-acid Air
0
21.3 4.9
6.0
a
ꢁ1
ꢀ
b
c
Reaction condition: LHSV (aniline) ¼ 0.4 h ; molar ratio of propanol/aniline ¼ 3; reaction temperature ¼ 410 C. c: conversion of aniline.
i
S :
d
e
f
g
selectivity to component i.
SQS: total selectivity to quinolines. YQS: total yield of quinolines. 2E-3MQ: 2-ethyl-3-methylquinoline. 2,3-DMQ: 2,3-
i j k l
h
dimethylquinoline. 2-EQ: 2-ethylquinoline. 2-MQ: 2-methyl quinoline. NPA: N-propylaniline. DNPA: N,N-dipropylaniline. Others: 2-
propylaniline, 4-propylaniline and 3-methylindole.
3-methylquinoline (3-MQ) over some catalysts are generated. formation is favorable in the larger pores than in the smaller
These products have an order 2E-3MQ [ DMQ [ 2-EQ pores. Moreover, zeolites are known to possess a narrow
ꢂ 3-MQ. Besides, the byproducts, including predominantly distribution of pore size and thus favor the selective formation
N-propylaniline (NPA) and small amounts of 4-propylaniline, of quinolines through cyclisation–aromatization, relative to the
2 2 3
2-propylaniline, N,N-dipropylaniline, and 3-methyl indole, are amorphous SiO –Al O that has a broad pore size distribution.
also identied in the product mixtures. Over the unmodied One may have noticed that the selectivity to NPA as the
catalysts (entries 1, 3, 6, 9 and 11), the conversion of aniline predominant byproduct over the unmodied catalyst has an
exhibits an order USY-acid > USY > SiO /Al O > HZSM-5 > HY. order USY-acid < USY < HY < HZSM-5 < SiO –Al O , being just
2
2
3
2
2 3
This order is found to be approximately reverse to that for the reverse to that for the total selectivity to quinolines. It hints that
concentrations of Br ¨o nsted acid sites over SiO /Al O , HZSM-5 the generation of quinolines can be a result of the further
2
2 3
ꢁ1
and HY is usually in the range of ca. 0.01–0.03 mmol g
.1–0.3 mmol g and 1–3 mmol g , respectively; because of (entries 2, 4, 5, 7, 8, 10 and 12–16), particularly those modied
,
conversion of NPA. Over the mono-metal-modied catalysts
ꢁ1
ꢁ1
0
dealumination, the concentration of Br ¨o nsted acid sites over by Ni, the conversion of aniline as well as the total selectivity to
HY, USY and USY-acid decreases sequentially. Therefore, the quinolines and total yield of quinolines are all increased, rela-
lower concentration of Br ¨o nsted acid sites is favorable to the tive to the corresponding unmodied catalysts (entries 1, 3, 6, 9
conversion of aniline, probably due to the smaller tendency for and 11). Moreover, the total yield of quinolines over the modi-
the deactivation of basic aniline by the protons of Br ¨o nsted acid ed catalysts exhibits the same order as that for the corre-
sites. However, USY catalyst is also known to possess a higher sponding unmodied catalysts. It indicates that the
concentration of Br ¨o nsted acid sites than SiO
/Al O catalyst, modication of metal has presented a positive effect on the
2 2 3
while the former catalyst exhibits a larger conversion of aniline catalytic performance, and the promotion effect of metal
for the synthesis of quinolines than the later one. This is most display an order Ni > Fe > Co > Cu > Zn. The Ni-USY-acid catalyst
probably because USY catalyst usually possesses a larger (entry 16) provides a 38.5% total yield of quinolines with 68.6%
2
ꢁ1
specic surface area (>700
(
m
g
)
than SiO
2
/Al
2
O
3
conversion of aniline and 56.1% total selectivity to quinolines.
2
ꢁ1
<350 m g ), enabling more active sites, e.g., Lewis acid sites, Over the ZnCl further modied Ni-USY-acid catalyst, i.e., ZnCl /
2
2
to be exposed for the activation and conversion of aniline. Both Ni-USY-acid (entry 17), the total yield of quinolines is found to
the total selectivity to quinolines (SQS) and total yield of quin- increase signicantly, providing an as high as 97.6% conversion
olines (YQS) have an order USY-acid > USY > HY > HZSM-5 > of aniline with 79.7% total selectivity to quinolines. Besides,
SiO
zeolites employed in this work have an order USY-acid > USY > selectivity to NPA (9.3%) decreases obviously over the ZnCl
HY > HZSM-5, while quinolines possess a larger molecule size USY-acid catalyst. As is well known, transition metal ions
than the byproducts as mentioned above and therefore their behave as Lewis acids, while ZnCl as a Lewis acid catalyst has
2
–Al
2
O
3
. This can be due to the fact that the pores size of the compared to that (40.1%) over the Ni-USY-acid catalyst, the
2
/Ni-
2
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 48275–48285 | 48279

View Article Online
RSC Advances
Paper
exhibited good performances in many reactions like Friedel– active sites for the generation of quinolines, too large amounts
50,51,53
Cras alkylation.
Therefore, the larger Lewis acidity is of ZnCl2 may block the channels of zeolite catalyst so as to
highly desired for the catalyst in the synthesis of quinolines inhibit the diffusion of reactants and products. Fig. 5(B) shows
from aniline and propanol, and ZnCl promotes not only the the effect of temperature, which indicates that, with increasing
2
N-alkylation of aniline and n-propanol to generate NPA but also the temperature, the conversion increases all through, however,
ꢀ
the further conversion of NPA to quinolines. Aer the carrier both the selectivity and yield exhibit a maximum at 410 C. This
gas H
decreased obviously over the ZnCl
cf. entries 17 and 18). It is known that the formation of quin- propanol over the catalyst, too higher temperatures may also
olines from aniline and a carbonyl compound involves the lead to some side reactions, e.g., the cracking of reactants and
2
is replaced with air, the total yield of quinolines is occurs owing to the fact that, though the increase in tempera-
2
/Ni-USY-acid catalyst ture promotes the activation and conversion of aniline and
(
22
cyclization-dehydrogenation step. Accordingly, the presence of products and the formation of carbon deposition. Fig. 5(C)
low valent metal species, acting as both Lewis acid sites and shows the effect of the molar ratio of propanol/aniline, which
dehydrogenating active centers, is vital for a catalyst to provide indicates that, with increasing the molar ratio of propanol/
high performance in the synthesis of quinolines from aniline aniline from 1 to 3, both the conversion and yield increase
and alcohol. The above results indicate that the ZnCl
acid exhibits the highest catalytic performance among all the the molar ratio of propanol/aniline, the conversion is almost
examined catalysts. unchanged, and both the selectivity and yield are decreased.
2
/Ni-USY- and the selectivity decreases. However, with further increasing
Fig. 5 illustrates the effects of various conditions on the This is due to the fact that, as mentioned from the discussion of
conversion of aniline as well as the total selectivity to quinolines Table 1, the generation of quinolines can be a result of the
and the total yield of quinolines for the reaction of aniline and further conversion of NPA. Therefore, the increase in the molar
propanol over the ZnCl
carrier gas. Fig. 5(A) shows the effect of ZnCl
catalyst. One can see that, with the increase of ZnCl
the conversion, selectivity as well as yield all increase obviously, 2-propylaniline and 4-propylaniline, competing with the
achieving their maxima at 10 wt% ZnCl loading, and then formation of quinolines. Fig. 5(D) shows the effect of LHSV
decrease. While the loading of ZnCl introduces more Lewis (aniline). It can be seen that with increasing the LHSV, the
2
/Ni-USY-acid catalyst, using H
loading in the NPA to N,N-dipropylaniline (DNPA) but also the direct reaction
loading, between aniline and propanol to other byproducts such as
2
as ratio of propanol/aniline may lead to not only the conversion of
2
2
2
2
Fig. 5 Effect of reaction conditions on the synthesis of quinolines from aniline and propanol over the ZnCl
2 2
/Ni-USY-acid catalyst. (A) ZnCl
loading; (B) reaction temperature; (C) molar ratio of propanol/aniline; (D) LHSV of aniline (the reaction is conducted under the conditions of
ꢀ
ꢁ1
temperature ¼ 410 C, molar ratio propanol/aniline ¼ 3, LHSV ¼ 0.4 h and molar ratio of propanol/aniline ¼ 3, one among which is varied for
A)–(D) while the others remaining unchanged).
(
48280 | RSC Adv., 2017, 7, 48275–48285
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
conversion decreases all through and both the selectivity and
ꢁ
1
ꢁ1
yield exhibit a maximum, which appear at 0.6 h and 0.4 h
,
respectively. This is because that the increase in LHSV decreases
not only the contact time and thus the conversion of aniline
over catalyst, but also the possibility of side reactions due to the
fast diffusion of products away from the catalyst. However, a too
higher LHSV would also lead to an insufficient adsorption and
activation of aniline the thus its transformation into quinolines
over the catalyst.
From Fig. 5, it is obtained that the optimized reaction
conditions for the generation of quinolines from aniline and
propanol over the ZnCl /Ni-USY-acid catalyst are: ZnCl loading
2
2
ꢀ
ꢁ1
¼
10 wt%; carrier gas ¼ H ; temperature ¼ 410 C; molar ratio
2
propanol/aniline ¼ 3; and aniline LHSV ¼ 0.4 h . Under those
optimized reaction conditions, the conversion of aniline and
total yield of quinolines are respectively 97.6 mol% and
Fig. 6 Lifespan of ZnCl
2
/Ni-USY-acid catalyst for the reaction of
7
9.7 mol% at 81.7 mol% total selectivity to quinolines. This
aniline and propanol to quinolones (reaction condition: carrier gas ¼
provides a 60.6 mol% (54.3 wt%) GC yield of 2-ethyl-3-
methylquinoline, according to Table 3.
Fig. 1S† shows the GC-MS chromatogram of the products
mixture received from the reaction of aniline and propanol
ꢁ1
H
2
; molar ratio of n-propanol/aniline ¼ 3; LHSV (aniline) ¼ 0.4 h
;
ꢀ
temperature ¼ 410 C).
under the above optimized conditions. It is identied that high as 86.3–88.7%. For the reaction of aniline and polyhydric
-ethyl-3-methylquinoline has been generated as the main alcohols or C –C monohydric alcohols, quinolines are gener-
2
3
4
product. Aer separation and purication, the structure of the ated as the main products, of which the total yield of quinolines
recovered 2-ethyl-3-methylquinoline is further conrmed by is in the range of 42.3–79.7%, while N-alkylaniline is formed as
1
13
both H NMR(Fig. 2S†) and C NMR (Fig. 3S†) characteriza- the main byproduct. However, exception to the above regularity
1
tions ( H NMR: d ¼ 8.03 (d, 1H, J ¼ 8.4 Hz), 7.84 (s, 1H), 7.70 occurs for the reaction of aniline and iso-propanol, which leads
(
d, 1H, J ¼ 8.0 Hz), 7.61 (dd, 1H, J ¼ 8.0, 7.2 Hz), 7.44 (dd, 1H, J ¼ to the generation of NAA (56.5% yield) as main product and
7
7
1
.6, 7.2 Hz), 3.02 (q, 2H, J ¼ 7.6 Hz), 2.50 (s, 3H), 1.37 (t, 3H, J ¼ quinolines (1.2% total yield) as minor ones. This probably
13
.6 Hz); C NMR: d ¼ 163.30, 146.67, 135.73, 129.40, 128.52, because the steric effect of isopropyl group is disadvantageous
28.29, 127.33, 126.70, 125.61, 29.51, 19.10, 13.08). Besides, the to the dimerization of N-iso-propylaniline, which is the key step
purity of 2-ethyl-3-methylquinoline is determined as 98.6% for the cyclization. It is identied that the most predominant
1
from the H NMR characterization and the recovery yield of quinolines generated by employing various alcohols are as
2
-ethyl-3-methylquinoline aer separation and purication is follows: 2-ethyl-3-methyl-quinoline (n-propanol), 2-propyl-3-
determined as 90.2%, which corresponds to a 44.0 wt% sepa- ethyl-quinoline (n-butanol), 2-methyl-quinoline (ethylene
rated yield (see ESI†).
glycol), 2-ethyl-3-methyl-quinoline (1,2-propanediol) and quin-
Fig. 6 shows the conversion of aniline as well as the total oline (glycerol). This hints that, while quinoline is formed via
selectivity to quinolines and the total yield of quinolines as the reaction between equimolecular aniline and glycerol,
functions of reaction time over the ZnCl
2
/Ni-USY-acid catalyst alkylquinolines are formed via the reaction between one
under the above optimized conditions. One can see that, within molecular aniline and two molecular other proper C –C alco-
2
4
3
0 h of reaction, the average level of the conversion of aniline, hols. The above results also indicate the reaction of aniline and
selectivity to quinolines and yield of quinolines remains at ca. n-propanol, among all other alcohols, has provided the highest
3%, 82% and 76%, respectively. For the reaction time longer total yield of quinolines (79.7%).
than 30 h, the catalytic performance turns to decrease. This
From the above results for the catalytic performance evalu-
occurs probably due to either the formation of carbon deposi- ation and catalyst characterization, it is obtained that both the
tion over the catalyst or the gradual loss of active ZnCl from the total yield of quinolines from the reaction of aniline and
9
2
catalyst. In any way, the residual conversion of aniline, selec- n-propanol to quinolines and the relative content of Lewis acid
tivity to quinolines and yield of quinolines still keep a level of sites exhibit an order ZnCl /Ni-USY-acid > Ni-USY-acid > USY-
2
77%, 65% and 50%, respectively for the reaction time of 65 h. acid > USY, but the porosity of catalyst displays a different
The above result shows that the ZnCl /Ni-USY-acid possesses an order. Accordingly, the catalytic performance is positively
2
appreciably good stability of catalytic performance.
related to the relative content of Lewis acid sites of catalyst.
Besides, for methanol, ethanol or iso-propanol as reactant,
Table 4 shows the results for the reactions of aniline and
C
1
–C
4
alcohols (methanol, ethanol, n-propanol, iso-propanol, NAA is generated as the predominant product, almost without
n-butanol, ethylene glycol, 1,2-propanediol and glycerol) over the generation of quinolines. For C –C polyhydric alcohols or
2
4
the ZnCl
2
/Ni-USY-acid catalyst. One can see that, for the reac- C –C monohydric alcohols as reactant, the quinolines con-
3 4
tions of aniline and C
1
2
–C monohydric alcohols, no quinolines structed theoretically by one molecular aniline and two
but N-alkylaniline (NAA) is generated, and the yield of NAA is as
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 48275–48285 | 48281

View Article Online
RSC Advances
Paper
a
Table 4 Results for the reactions of aniline and C
1
–C
4
2
alcohols over the ZnCl /Ni-USY-acid catalyst
Yield of quinolones (mol%)
b
Indoles yield NAA yield
c
Reactant
(mol%)
(mol%)
Others Total yields
Methanol
Ethanol
n-Propanol
Iso-propanol
n-Butanol
0
0
1.2
1.1
0
88.7
86.3
12.1
56.5
12.4
6.6
0
0
0
0
0
0
6.0
39.7
0
0
0
0
18.9
0
2.8
3.7
5.4
0
0
0
60.6
0
0
0
0
0
52.6
0
0
0
0
0
1.2
2.6
2.2
2.1
0
0
0.2
1.2
0
56.1
1.3
0.5
79.7
1.2
56.6
71.8
44.7
42.3
0
Ethylene glycol 9.7
,2-Propanediol 3.5
9.4
29.8
0
1
8.0
5.4
0
0
Glycerol
0.1
a
ꢁ1
ꢀ
b
Reaction condition: LHSV (aniline) ¼ 0.4 h ; molar ratio of propanol/aniline ¼ 3; reaction temperature ¼ 410 C; carrier gas ¼ H
2
.
NAA: N-
c
alkylaniline, including NPA and DNPA. Others: other substituted quinolines, such as, 2-ethylquinoline, 2-methyl-3-ethylquinoline.
molecular alcohols are generated as the predominant products, Particularly, glycerol can be dehydrated directly into acrolein at
23
with NAA being generated as the most abundant byproduct. high yield. In addition, the stable Schiff base is usually
58
Exception to the above regularity is the case for glycerol as prepared from the reaction of aniline and aldehyde/ketone.
reactant, in which the predominant product quinoline is theo-
retically constructed by equimolecular aniline glycerol.
Basing on the above facts, it is proposed that the reaction of
aniline and alcohol to quinolines may follow the mechanisms
2 4
It has been reported that the mechanism for the synthesis of as described by Scheme 1. For a C –C polyhydric alcohol as
quinolines from aniline may involve either unsaturated alde- reactant (Scheme 1A), it is rst dehydrated into the corre-
20,23,30
54,55
hyde/ketone
or Schiff base/Schiff base dimer
as the key sponding aldehyde, which is further converted into unsaturated
intermediate. As is well known, a C –C monohydric alcohols aldehyde via Aldol reaction. Then, the unsaturated aldehyde
1
4
very hard to be transformed into the corresponding aldehyde/ reacts with aniline via Michael addition, followed by cyclization
ketone but easily dehydrated into the corresponding alkene and dehydrogenation-aromatization, and this nally generates
5
6–58
over solid acid catalyst,
monohydric alcohol over molecular sieve catalyst has consti- a C
tuted the most important route for the synthesis of N-alkyl the corresponding Schiff base. Then, the Schiff base is sub-
and the reaction of amine and quinolines. It is also possible that the aldehyde obtained from
2
–C polyhydric alcohol rst reacts with aniline to generate
4
58
compound. Contrastively, it is feasible to convert a C –C4 jected to dimerization, cyclization and dehydrogenation–
2
polyhydric alcohol into the corresponding aldehyde/ketone over aromatization, and this nally generates quinolines. Specially,
solid acid catalyst, while aldehyde/ketone can be further con- for glycerol as reactant, it is rst directly dehydrated into acro-
59
verted into unsaturated aldehyde/ketone via Aldol reaction.
lein, being an unsaturated aldehyde, and then subjected to the
Scheme 1 Reaction pathways for the synthesis of quinolines via the reactions of aniline and polyhydric alcohols (A) or monohydric alcohols (B).
48282 | RSC Adv., 2017, 7, 48275–48285
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
cyclization and dehydrogenation–aromatization, generating of various transition metals via ion-exchange method. In the
quinoline. For a C –C alcohol as reactant (Scheme 1B), it rst typical procedure, USY-acid was rst dispersed into aqueous
1
4
reacts with aniline to generate the corresponding N-alkylani- solution of a transition metal nitrate salt, according to a weight
ꢀ
line, which is subsequently dehydrogenated into Schiff base. ratio liquid/solid ¼ 10/1, and then reuxing treated at 90 C for
Then, the Schiff base is subjected to dimerize, cyclization and 6 h under strong stirring. Aer that, the suspension was l-
dehydrogenation–aromatization, and this nally generates trated, washed thoroughly with deionized water and dried at
ꢀ
ꢀ
quinolines. It should be addressed that the dimerization of 100 C, followed by calcination at 550 C for 4 h. The resultant
ꢁ
Schiff base involves the generation of carbanion Ph ¼ N–R , powder was repeatedly subjected to the above procedure for
which is impossible for R ¼ methyl and very unstable for R ¼ three times, with the fresh salt solution being employed for
ethyl but relatively stable for larger alkyls. Therefore, only each time. The thus-prepared catalyst was denoted as M-USY-
N-alkylanilines are generated from the reactions of aniline and acid (M ¼ Fe, Co, Ni, Cu and Zn, respectively).
methanol and ethanol, while quinolines can be predominantly
Ion exchange of Y, USY and ZSM-5. As comparisons, a series
formed from the reactions of aniline and n-propanol or of M-Y, M-USY and M-ZSM-5 catalysts (M ¼ Fe or Ni) were also
n-butanol. Besides, there is a large steric effect for the dimer- prepared by the same ion-exchange method as described above
ization of N-iso-propylaniline, and this accounts for the very low for M-USY-acid.
selectivity to quinolines but high selectivity to N-alkylaniline
Wet impregnation of USY. The above-prepared Ni-USY-acid
from the reaction between aniline and iso-propanol. In the was rst impregnated with a calculated amount of 0.1 M
above mechanism for the reaction of aniline and alcohol, the ZnCl aqueous solutions at room temperature for 12 h under
2
ꢀ
presence of Lewis acid sites promotes the adsorption activation vigorous stirring. Then the slurry was dried at 120 C overnight,
ꢀ
of aniline and alcohols and the subsequent various steps, while followed by calcination at 500 C for 4 h in air. The thus-
the presence Bronsted acid sites may cause a few negative prepared catalyst was denoted as ZnCl
2
/M-USY-acid, and it
effects, e.g., the formation of inactive protonated aniline and contained 5–40 wt% of ZnCl
2
. Unless otherwise mentioned
ꢁ
the inhibition to the formation of carbanion Ph ¼ N–R . specially, the content of content of ZnCl
2
in the ZnCl
2
/M-USY-
Therefore, the employment of the catalyst with high concen- acid catalyst was 10 wt%.
tration ratio of Lewis acid sites to Bronsted acid sites is largely
Deposition–precipitation of SiO
2
–Al
2
O
3
.
A
calculated
favorable to the synthesis of quinolines from aniline and amount of Ni(NO ) and an excess amount of urea were rst
3
2
alcohol, and this has been evidenced by the above results for dissolved in distilled water at the room temperature. Then,
catalyst characterization and catalytic performance evaluation. a certain amount of SiO –Al O was dispersed into the above
2
2 3
solution under stirring. Aer that, the resultant slurry was
ꢀ
heated to 90 C and reuxed for 3 h under strongly stirring.
Experimental
Chemicals
Finally, the slurry was ltrated room temperature, dried at
ꢀ
ꢀ
120 C overnight, and followed by calcination at 550 C for 4 h.
The thus-prepared catalyst was denoted as Ni/SiO
contained 10 wt% of Ni in oxide form.
2 2 3
–Al O , and it
All organic reagents for the synthesis of quinolines and inor-
ganic salts for the preparation of catalysts were purchased from
commercial companies and had purities higher than 99%.
These chemicals were employed as received, without further
purication.
Characterization of catalyst
X-ray diffraction spectroscopy (XRD) was performed with
a Bruker D8-Advance X-ray diffractometer. The operation
Zeolites USY (Si/Al ¼ 10) and HY (Si/Al ¼ 6) as well as
amorphous SiO
2
–Al
2
O
3
(Si/Al ¼ 6–9) catalysts were all received
˚
conditions were as follows: Cu target Ka ray (l ¼ 1.54187 A);
from SINOPEC Catalyst Changling Division, and HZSM-5 (Si/Al
scanning voltage 40 kV, scanning current 40 mA; scanning
¼
50) were purchased from Shanghai Fuxu Zeolite Company.
ꢀ
ꢀ
speed 0.2 s, scanning step 0.02 .
These catalysts were rst calcined at 550 C for 5 h, and then,
they were either used directly as catalysts or further subjected to
the procedures for the preparation of modied catalysts as
described in the subsequent section.
Fourier transform infrared (FT-IR) spectroscopy was recor-
ded on a Varian 3100 spectrometer equipped with a GTGS
detector. The catalyst was rst mixed with KBr (1 : 100) by
thoroughly grounding, and then, the mixture was pressed into
the sample holder and mounted in the cavity of the spectrom-
eter. Each spectrum was obtained by scanning for 32 times at
Preparation of modied catalyst
ꢁ
1
Acid treatment of USY. The commercial zeolite USY were rst a resolution of 2 cm
.
dispersed in a 0.5 M hydrochloric acid aqueous solution,
2
N -physisorption was performed at liquid nitrogen temper-
according to a weight ratio liquid/solid ¼ 10/1, and then ature using a QuantaChrome Autosorb-1 instrument. Before the
ꢀ
reuxing treated at 60 C for 30 min under strong stirring. Aer adsorption measurement, the specimen degassed for 16 h at
ꢀ
that, the suspension was ltrated, washed thoroughly with a temperature of 300 C under a residual pressure lower than 4
ꢀ
ꢁ4
deionized water and dried at 100 C. The thus-prepared catalyst ꢃ 10 Pa in the degas port of the adsorption apparatus. The
was denoted as USY-acid. specic surface area was calculated by using the multipoint BET
Ion exchange of USY-acid. The above-prepared USY-acid equation. The pore volume and area was calculated by using the
powder was further modied respectively by the nitrate salts t-plot micropore analysis method. The pore size and its
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 48275–48285 | 48283

View Article Online
RSC Advances
Paper
distribution of microporous were obtained from the N
2
-
Selectivity to component i (%):
desorption branch employing the SF method, and those of
N
i
S
i
ð%Þ ¼ X ꢃ 100
mesoporous were determined from the N -desorption branch
2
N
i
by the BJH method.
3
Ammonia temperature programmed desorption (NH -
TPD) was determined on a Micromeritics AutoChem II 2920
instrument equipped with a thermal conductivity detector
Yield of component i (%):
QS (%) ¼ c ꢃ SQS ꢃ 100,
(
TCD). The catalyst specimen was rst degassed in a ow of
Y
ꢁ1
ꢀ
helium with a ow rate of 50 mL min at 400 C for 30 min,
followed by cooling to 100 C. Then, NH3 was repeatedly
ꢀ
where Naniline in feed and Naniline in products mixture referred to the
moles of aniline in the feed and in the products mixture, N to
the moles of component i in the products mixture and SQS to the
total selectivity to quinolines, respectively.
pulse-injected until a saturation adsorption over the spec-
i
imen had been achieved. Aer that, NH3 was desorbed by
ꢀ
heating the specimen from 100 to 800 C at a rate of
ꢀ
1
0 C min. During the adsorption and desorption of NH
3
, the
helium ow was retained and its ow rate was maintained
Conclusions
ꢁ
1
constant at 60 mL min
.
The metal contents of catalysts were determined over The novel route for the synthesis of quinolines from the reac-
a Varian 240AA atomic absorption spectrometer. The opera- tion of aniline and a C –C alcohol under gas-phase reaction
1
4
tion conditions were as follows: sample aspiration rate ¼ condition over heterogeneous catalyst had been developed in
ꢁ
1
3

.0 mL min ; lamp current ¼ 3.0 mA; slit width ¼ 0.2 nm; air this work. It was identied that, for the reaction of aniline and
ꢁ1
ow rate ¼ 0.8 mL min
and acetylene ow rate ¼ propanol, the USY zeolite-based catalysts possessed obviously
higher performances, relative to other catalysts. Over various
ꢁ
1
5
.0 mL min .
modied USY catalysts, both the total yield of quinolines and
the concentration ratio for Lewis acid sites to Bronsted acid
Catalytic performance evaluation
sites exhibited an order USY < USY-acid < Ni-USY-acid < ZnCl
2
/
Ni-USY-acid. Under the optimized conditions over the ZnCl /Ni-
2
Catalytic performance test was carried out in a xed-bed tubular
quartz reactor (i.d. 6 mm, total length 400 mm) at atmospheric
pressure. The catalyst (1.0 g) was loaded in the middle of
reactor, with the upper space of catalyst bed in the reactor being
USY-acid catalyst, the reaction of aniline and most of alcohols
provided a 42.3–79.7% total yield of quinolines, however, the
reaction of aniline and methanol, ethanol or iso-propanol
predominantly generated N-alkylanilines. It was proposed that
the mechanism for the formation of quinolines from aniline
and alcohol involved mainly Schiff base as the key intermediate,
while that involved unsaturated aldehyde also could not be
excluded.

lled with quartz granulates (20 meshes). The mixture obtained
by dissolving aniline into a C –C alcohols (methanol, ethanol,
1
4
n-propanol, iso-propanol, butanol, ethylene glycol, 1,2-pro-
panediol and glycol) at a certain molar ratio (aniline/alcohol ¼
1/1–1/5) was employed as feedstock. Firstly, the catalyst was in
situ pre-treated at certain reaction temperature ranging from
ꢀ
3
10 to 430 C for 1 h in a ow of carrier gas (H
2
or air; 10
Conflicts of interest
ꢁ1
mL min ). Then, the feedstock was pumped into a preheater to
be vaporized at 200 C, at a liquid hourly space velocity (LHSV)
ꢀ
There are no conicts of interest.
ꢁ
1
ranging from 0.2–0.8 h . Then, the feedstock was introduced
in to the reactor at a pre-set temperature, with the carrier gas
being kept. Aer the reaction had been run stably for 2 h, the
products were collected by a condenser which was placed in an
ice-water bath.
Acknowledgements
We are grateful to the nancial supports from the project
#
(
21376068 for National Natural Science Foundation of China
NSFC), Program for Lotus Scholar in Hunan Province, and also,
Program for New Century Excellent Talents in University.
The composition of products mixture was analysed by
a Varian CP-3800/Saturn 2200 gas chromatography-mass spec-
trometry (GC-MS). Two CP8944 capillary columns (VF-5, 30 m ꢃ
0
.25 mm ꢃ 0.25 mm) were respectively connected to mass
detector and ame ionization detector (FID) for the quantitative References
and qualitative identications.
1
2
3
4
D. G. Markees, V. C. Dewey and G. W. Kidder, J. Med. Chem.,
970, 13, 324–326.
A. A. Alhaider, M. A. Abdelkader and E. J. Lien, J. Med. Chem.,
985, 28, 1394–1398.
Y. Morimoto, F. Matsuda and H. Shirahama, Synlett, 1991,
202–203.
Basing on the converted aniline, the conversion of aniline
c), selectivity to component i (S ) and yield of quinolines (YQS)
were respectively calculated as follows:
Conversion of aniline (%):
1
(
i
1
N
aniline in feed ꢁ Naniline in products mixture
c ð%Þ ¼
ꢃ 100
N
aniline in feed
M. Isobe, T. Nishikawa, N. Yamamoto, T. Tsukiyama, A. Ino
and T. Okita, J. Heterocycl. Chem., 1992, 29, 619–625.
48284 | RSC Adv., 2017, 7, 48275–48285
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
5
6
7
8
V. V. Kouznetsov, L. Y. V. M ´e ndez and C. M. M. G ´o mez, Curr. 33 J. L. Agudelo, B. Mezari, E. J. M. Hensen, S. A. Giraldo and
Org. Chem., 2005, 9, 141–161. L. J. Hoyos, Appl. Catal., A, 2014, 488, 219–230.
S. Eswaran, A. V. Adhikari, I. H. Chowdhury, N. K. Pal and 34 D. Verboekend, G. Vil ´e and J. P ´e rez-Ram ´ı rez, Adv. Funct.
K. D. Thomas, Eur. J. Med. Chem., 2010, 45, 3374–3383. Mater., 2012, 22, 916–928.
S. Gogoi, K. Shekarrao, A. Duarah, T. C. Bora, S. Gogoi and 35 G. Tonetto, M. L. Ferreira and H. de Lasa, J. Mol. Catal. A:
R. C. Boruah, Steroids, 2012, 77, 1438–1445. Chem., 2004, 216, 83–99.
V. V. Kouznetsov, C. M. Melendez Gomez, M. G. Derita, 36 D. D. Guo, B. J. Shen, Y. C. Qin, J. X. Sun, Q. X. Guo, S. Y. Ren,
L. Svetaz, E. Del Olmo and S. A. Zacchino, Bioorg. Med.
Chem., 2012, 20, 6506–6512.
Z. H. Skraup, Monatsh. Chem., 1880, 1, 316–318.
0 R. H. F. Manske and M. Kulka, Org. React., 1953, 7, 59–93.
1 O. Doebner and W. V. Miller, Ber. Dtsch. Chem. Ges., 1881, 14, 38 C. W. Luo, C. Huang, A. Li, W. J. Yi, X. Y. Feng, Z. J. Xu and
X. H. Gao, X. M. Pang, B. J. Wang, H. J. Zhao and H. H. Liu,
Microporous Mesoporous Mater., 2015, 211, 192–199.
37 A. Ishihara, K. Inui, T. Hashimoto and H. Nasu, J. Catal.,
2012, 29, 581–590.
9
1
1
2
812–2817.
Z. S. Chao, Ind. Eng. Chem. Res., 2016, 55, 893–911.
39 I. Othman, R. M. Mohamed, I. A. Ibrahim and
M. M. Mohamed, Appl. Catal., A, 2006, 299, 95–102.
40 L. Shirazi, E. Jamshidi and M. R. Ghasemi, Cryst. Res.
Technol., 2008, 43, 1300–1306.
1
1
2 W. P. Utermohlen, J. Org. Chem., 1943, 8, 544–549.
3 S. Concilio, P. M. Pster, N. Tirelli, C. Kocher and
U. W. Suter, Macromolecules, 2001, 34, 3607–3614.
1
4 G. Chelucci, I. Manca and G. A. Pinna, Tetrahedron Lett.,
2
005, 46, 767–770.
41 A. Vimont, F. Thibault-Starzyk and J. C. Lavalley, J. Phys.
Chem. B, 2000, 104, 286–291.
1
1
5 J. L. Born, J. Org. Chem., 1972, 37, 3952–3953.
6 S. A. Yamashkin, L. G. Yudin and A. N. Kost, Chem. 42 C. S. Blackwell, J. Phys. Chem., 1979, 83, 3251–3257.
Heterocycl. Compd., 1992, 28, 845–855. 43 R. Szostak and T. L. Thomas, J. Catal., 1986, 101, 549–552.
7 H. Zhu, R. F. Yang, L. H. Yun and J. Li, Chin. Chem. Lett., 44 A. Corma, Chem. Rev., 1997, 97, 2373–2420.
1
1
1
2
2
2
2
2
010, 21, 35–38.
8 K. Taguchi, S. Sakaguchi and Y. Ishii, Tetrahedron Lett., 2005,
6, 4539–4542.
9 Y. Tsuji, K. T. Huh and Y. Watanabe, J. Org. Chem., 1987, 52,
673–1680.
0 M. Campanati, A. Vaccari and O. Piccolo, Catal. Today, 2000,
0, 289–295.
1 M. Campanati, P. Savini, A. Tagliani and A. Vaccari, Catal. 49 J. R. Sohn, S. J. DeCanio, P. O. Fritz and J. H. Lunsford, J.
Lett., 1997, 47, 247–250. Phys. Chem., 1986, 90, 4847–4851.
2 R. Brosius, D. Gammon, F. Vanlaar, E. Vansteen, B. Sels and 50 J. Liu, D. Yin, D. Yin, Z. Fu, Q. Li and G. Lu, J. Mol. Catal. A:
P. Jacobs, J. Catal., 2006, 239, 362–368. Chem., 2004, 209, 171–177.
3 B. M. Reddy and I. Ganesh, J. Mol. Catal. A: Chem., 2000, 151, 51 P. O. Fritz and J. H. Lunsford, J. Catal., 1989, 118, 85–98.
45 M. A. Kuehne, H. H. Kung and J. T. Miller, J. Catal., 1997,
171, 293–304.
4
46 N. Y. Topsøe, K. Pedersen and E. G. Derouane, J. Catal., 1981,
70, 41–52.
1
47 B. Christian, P. Christoph, L. Michael, D. Frieder and
B. Dieter, Microporous Mesoporous Mater., 2017, 241, 1–10.
48 J. Massam and D. R. Brown, Catal. Lett., 1995, 35, 335–343.
6
289–293.
52 R. A. Beyerlein, G. B. McVicker, L. N. Yacullo and
2
2
4 C. McAteer, R. D. Sr and J. Calvin, US Pat., 5700942, 1997.
J. J. Ziemiak, J. Phys. Chem., 1988, 92, 1967–1970.
5 S. Kamiguchi, I. Takahashi, H. Kurokawa, H. Miura and 53 V. R. Choudhary, S. K. Jana and N. S. Patil, Catal. Lett., 2001,
T. Chihara, Appl. Catal., A, 2006, 309, 70–75. 76, 235–239.
6 J. R. Calvin, R. D. Davis and C. H. McAteer, Appl. Catal., A, 54 C. S. Cho, J. S. Kim, B. H. Oh, T.-J. Kim, S. C. Shim and
2
2
005, 285, 1–23.
N. S. Yoon, Tetrahedron, 2000, 56, 7747–7750.
55 Y. Watanabe, Y. Tsuji, Y. Ohsugi and J. Shida, Bull. Chem.
Soc. Jpn., 1983, 56, 2452–2457.
2
2
2
7 W. Schirmer, Phys. Chem., 1994, 186, 269–270.
8 D. M. Ruthven, Stud. Surf. Sci. Catal., 1995, 97, 223–234.
9 Z. Yan, D. Ma, J. Zhuang, X. Liu, X. Liu, X. Han, X. Bao, 56 D. E. Bryant and W. L. Kranich, J. Catal., 1967, 8, 8–13.
F. Chang, L. Xu and Z. Liu, J. Mol. Catal. A: Chem., 2003, 57 R. Le Van Mao, P. Levesque, G. McLaughlin and L. H. Dao,
1
94, 153–167.
0 L. Zhang, W. Q. Fu, Q. P. Ke, S. Zhang, H. L. Jin, J. B. Hu, 58 I. Takahara, M. Saito, M. Inaba and K. Murata, Catal. Lett.,
S. Wang and T. D. Tang, Appl. Catal., A, 2012, 433, 251–257. 2005, 105, 249–252.
1 L. P. Zhou, H. T. Zhao, L. T. Cui, Y. Q. Bai, J. J. Bian, T. L. Lu, 59 S. Ito, K. Tanuma, K. Matsuda, A. Hayashi, H. Komai,
Appl. Catal., 1987, 34, 163–179.
3
3
3
Y. L. Su and X. M. Yang, Catal. Commun., 2015, 71, 74–78.
2 A. H. Janssen, A. J. Koster and K. P. de Jong, J. Phys. Chem. B,
Y. Kubota and M. Asami, Tetrahedron, 2014, 70, 8498–8504.
2002, 106, 11905–11909.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 48275–48285 | 48285