Vapor-phase synthesis of 3-methylindole from glycerol and aniline over zeolites-supported Cu-based catalysts

Article abstract of DOI:10.1016/j.molcata.2013.06.015

Zeolites-supported Cu-based catalysts for the vapor-phase synthesis of 3-methylindole from glycerol and aniline were investigated. The catalysts were characterized by N₂ adsorption, XRD, NH₃-TPD and TG-DTA techniques. The results indicated that Cu/NaY catalyst exhibited higher activity and selectivity than Cu/5A, Cu/HZSM-5 or Cu/MCM-41 due to better dispersion of copper particles and larger amount of weak acid sites on the catalyst. The addition of Na₂O to Cu/NaY not only improved the selectivity of the catalyst efficiently, but also inhibited the deactivation of the catalyst. Over Cu/NaY-Na₂O, the selectivity of 3-methylindole reached 72.3% and the yield of the aim product was up to 50.4%. The characterizations revealed that Na₂O promoter could decrease the amount of middle-strong acid sites of the catalyst remarkably as well as prevent from the formation of the coke. Furthermore, a probable catalytic mechanism for the synthesis of 3-methylindole from glycerol and aniline was proposed.

Full text of DOI:10.1016/j.molcata.2013.06.015

Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Vapor-phase synthesis of 3-methylindole from glycerol and aniline
over zeolites-supported Cu-based catalysts
∗
Yanxi Cui, Xiaoshuang Zhou, Qi Sun, Lei Shi
Institute of Chemistry for Functionalized Materials, Liaoning Normal University, Dalian 116029, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Zeolites-supported Cu-based catalysts for the vapor-phase synthesis of 3-methylindole from glycerol
and aniline were investigated. The catalysts were characterized by N2 adsorption, XRD, NH3-TPD and
TG–DTA techniques. The results indicated that Cu/NaY catalyst exhibited higher activity and selectivity
than Cu/5A, Cu/HZSM-5 or Cu/MCM-41 due to better dispersion of copper particles and larger amount
of weak acid sites on the catalyst. The addition of Na2O to Cu/NaY not only improved the selectivity
of the catalyst efficiently, but also inhibited the deactivation of the catalyst. Over Cu/NaY–Na2O, the
selectivity of 3-methylindole reached 72.3% and the yield of the aim product was up to 50.4%. The char-
acterizations revealed that Na2O promoter could decrease the amount of middle-strong acid sites of the
catalyst remarkably as well as prevent from the formation of the coke. Furthermore, a probable catalytic
mechanism for the synthesis of 3-methylindole from glycerol and aniline was proposed.
Received 13 April 2013
Accepted 21 June 2013
Available online 1 July 2013
Keywords:
3
-Methylindole
Zeoliteds-supported copper-based catalysts
NaY
Sodium oxide
Glycerol
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
can give 3-methylindol in a good yield [23,24], however, some
drawbacks exist, such as the use of complicated and expensive
reagents or catalysts, a large number of toxic solvents and heavy
pollution [25–27], which has limited the further development of
3-methylindole. Thus, increasing interest has been focused on the
vapor-phase synthesis of 3-methylindole. Subrahmanyam and co-
workers [28] carried out the reaction from indole and methanol
(molar ratio of indole/methanol was 1:6) over CeHY zeolite cata-
lyst and a 33% yield of 3-methylindole was obtained. Vaccari and
co-workers [29] conducted the synthesis of 3-methylindole from
Recently, much attention has been given to the use of glycerol
for glycerol is the main by-product (about 10 wt%) in biodiesel pro-
duction [1–5]. Finding out new outlets to make good use of the
surplus glycerol agrees with the concept of green chemistry in
terms of the utilization of renewable resources and can acceler-
ate the development of biodiesel. Up to now, many efforts have
been put toward the conversion of glycerol and many value-added
chemicals are obtained such as propanediols, acrolein, glyceric acid,
methyl glycerate, solketal and ethylene glycol [6–11].
1,2-propanediol and aniline over ZrO /SiO2 catalyst, but the yield
2
3
-Methylindole, an important nitrogen-containing heterocyclic
of 3-methylindole was only 12%. Recently, our group developed
a new approach for the vapor-phase synthesis of 3-methylindole
with feeding glycerol and aniline as reactants (Scheme 1). Over
Cu/SiO –Al O , the yield of 3-methylindole reached 40% [30]. Since
chemical, can be widely applied in the fields of industry, agriculture
and medicine [12–14]. For example, it can be used in perfumes and
in the manufacture of herbicides, fungicides, dyes, antihyperten-
sive, anti-diuretic, muscular relaxant and anticancer medicaments
2
2
3
glycerol, comes from biomass, is not only cheap and abundant, but
also a green feedstock, this synthesis, from the economic and envi-
ronmental viewpoints, can be considered a most promising route.
In the last few decades, zeolites have attracted increasing
interest in the catalytic synthesis of fine chemicals because of
their special features such as the acidity, the inherent capability
of product selectivity, commercially available, non-corrosive and
eco-friendly nature, etc [31–34]. The objective of this work is to
research zeolites (NaY, 5A, HZSM-5 and MCM-41)-supported cop-
per catalysts for the vapor phase synthesis of 3-methylindole from
[
15,16]. Recently, 3-methylindole was found other interesting uses,
such as, affecting the production of serotonin, hemolyzing bovine
erythrocytes, activating transcription of cytochrome P450 enzymes
in hepatocytes and a broad bacteriostatic effect on some gram-
negative enterobacteria [17–20].
For many years, the synthesis of 3-methylindole has been an
area of focus for the chemists engaged in fine chemicals synthesis
and numerous methods have been developed [21,22]. Among them,
liquid-phase method is often considered conventional because it
glycerol and aniline. Na O promoter was introduced into the opti-
2
mal zeolite-supported copper catalyst to increase the selectivity of
3-methylindole. BET, XRD, NH3-TPD and TG–DTA techniques were
used to reveal the relationship between the structure of catalysts
∗ Corresponding author. Tel.: +86 411 82158329; fax: +86 411 82158309.
E-mail addresses: shilei515@dl.cn, shilei515@hotmail.com (L. Shi).
1
381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.06.015

Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
239
CH3
2.3. Catalyst characterization
H C OH
2
Cat.
HC OH
BET surface areas and pore volumes of the catalysts were
determined by pure nitrogen adsorption using an AUTOSORB-1MP
apparatus. Before each measurement, the sample was outgassed
overnight at 200 C under 10 Torr. Calculation of pore sizes fol-
lowed the method of Barret–Joyner–Halenda (BJH) according to the
desorption portion of the isotherms.
+
3H O
2
+
NH2
N
H
H C OH
2
◦
−3
Scheme 1. Vapor-phase synthesis of 3-methylindole from glycerol and aniline.
X-ray diffraction (XRD) patterns were recorded by using a D8
Advance diffraction meter with Cu K␣ source. MCM-41 sample
and their catalytic performances. Moreover, a probable catalytic
mechanism for the synthesis of 3-methylindole was suggested.
◦
◦
was scanned in the 2ꢀ range of 0.5–10 in steps of 1 /min, while,
◦
zeolites-supported Cu-based catalysts were scanned from 30 to
◦
◦
8
0
at a rate of 0.5 /min. The voltage and current were 40 kV and
2
. Experimental
4
0 mA, respectively. Average copper crystallite size was calculated
from peak broadening using Scherrer equation:
2.1. Materials
Kꢁ
D =
The zeolites of NaY and 5A were obtained from Shanghai Hengye
ˇ cos ꢀ
Molecular Sieve Co. Ltd. (China). HZSM-5 zeolite was obtained
from Nankai Chemical Co. Ltd. (China). MCM-41 was prepared as
described in Ref. [35] and the XRD pattern of MCM-41 sample is
depicted in Fig. 1. A dominant (1 0 0) peak centered at about 2 was
appeared, as well as the additional weak (1 1 0) and (2 0 0) peaks
were also obviously observed, corresponding to a highly ordered
where K is the shape factor (0.89 for the present instrument), ꢁ
is the wavelength of X-ray, typically 1.54 A˚ (Cu K␣), ˇ is the line
broadening at half the maximum intensity (FWHM) in radians, and
◦
ꢀ
is the Bragg angle, D is the crystallite size.
The temperature programmed desorption of ammonia (NH3-
TPD) was carried out in a quartz reactor with inside diameter of
2
D-hexagonal mesoporous silica framework.
6
5
mm and length of 350 mm. A 150 mg sample was pretreated at
00 C for 1 h in a flow of ultrapure helium gas (35 mL/min) to
◦
remove water and other contaminants from the catalyst and cooled
2
.2. Catalyst preparation
◦
◦
down to 100 C, then saturated with ammonia gas at 100 C. After
◦
the sample was purged with helium gas (35 mL/min) at 100 C for
The zeolite (NaY, 5A, HZSM-5 or MCM-41)–supported cop-
2
h to remove physisorbed ammonia, TPD was carried out from 100
per catalyst was prepared by equal volume impregnation. After
impregnating the zeolite support (20–40 mesh) in copper nitrate
aqueous solution for 15 h at room temperature, the sample was
dried at 120 C for 4 h, then, calcined at 500 C for 4 h to obtain
the catalyst precursor. Cu/NaY–Na O catalyst was prepared by
sequential impregnation. Na O was first loaded on the surface of
NaY, followed by copper and the process was the same as zeolite-
supported copper catalyst described above.
Prior to activity test, 3 mL catalyst precursor was in situ reduced
in a gas mixture composed of N2 (30 mL/min) and H2 (30 mL/min)
at 250 C for 2 h, then cooled to 220 C in 0.5 h and kept at the
temperature for another 0.5 h. Copper loading was 5.4 wt%.
◦
◦
to 700 C with a temperature ramp of 10 C/min.
Thermogravimetric and differential thermal analysis (TG–DTA)
was carried out with a Diamond 6300 apparatus to characterize
the coke on the catalyst. About 7.8 mg of the used catalyst (all the
used catalysts were proceeded the same reaction time of 4 h) was
◦
◦
2
2
placed in the pan. The sample was first heated in ultrapure nitrogen
◦
(
1
20 mL/min) from room temperature to 300 C at a heating rate of
◦
◦
0 C/min to remove physisorbed species, then cooled to 50 C and
◦
◦
heated in air (20 mL/min) to 800 C at a ramping rate of 10 C/min.
◦
◦
2.4. Catalyst testing
The catalytic reaction was carried out in a fix-bed continuous
flow glass reactor with an inside diameter of 12 mm under atmo-
spheric pressure. The reactor was placed vertically inside a tubular
furnace and heated electrically. The reactor temperature was mon-
itored by a thermocouple with its tip located at the catalyst bed
and connected to a temperature controller. The solution of reac-
tants (molar ratio of aniline/glycerol = 3:1) was pumped through
the preheater where it was vaporized and the reactants were car-
ried into the reactor by flowing H2 (20 mL/min), steam (6 mL/min)
(
100)
◦
and N2 (54 mL/min). The reaction was carried out at 220 C. The
total space velocity (SV) of raw materials and liquid hourly space
−
1
−1
and 0.4 h , respec-
velocity (LHSV) of reactants were 1700 h
tively.
2.5. Product analysis
The products were analyzed on a gas chromatograph with a
(
110)
(200)
mass spectrometer (Agilent GC–MS 6890N-5975B) equipped with
a HP-5 capillary column (0.25 mm, 30 m). The process was carried
◦
◦
out in ultrapure helium gas from 90 C to 230 C at a heating rate of
1
2
3
4
5
6
7
8
9
10
◦
5 C/min and kept at the temperature for 30 min.
The quantitative analyses of the reactants and products were
carried out on a SP-6890A gas chromatograph equipped with a SE-
54 capillary column (0.32 mm, 30 m). 1-Hexyl alcohol was used as
2
θ ( )
Fig. 1. XRD pattern of MCM-41 sample.

2
40
Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
Table 1
The activity and selectivity of the copper catalysts supported on different zeolites.
Catalyst
Glycerol conversion (%)
Product selectivity (%)
3-MI
N-MA
N-EA
2,5-DMPP
2,3-DMI
Others
Cu/NaY
Cu/5A
Cu/HZSM-5
Cu/MCM-41
72.0
65.3
60.1
53.6
56.1
32.2
30.0
24.3
3.1
1.4
–
5.5
44.2
46.5
15.8
19.1
6.8
6.9
–
–
–
16.2
15.4
16.6
32.3
2.8
12.2
12.6
◦
−1
−1
Reaction conditions: 220 C, aniline/glycerol = 3:1 molar ratio, SV = 1700 h , LHSV = 0.4 h , H2 = 20 mL/min, steam = 6 mL/min, N2 = 54 mL/min. Copper loading: 5.4 wt%. The
results were taken at the second hour.
Table 2
an internal standard. The temperature programmed process was
Specific surface area and pore volume of the copper catalysts supported on different
zeolites.
the same as GC–MS analysis with ultrapure nitrogen as the carrier
gas.
The conversion of glycerol, the yield and selectivity of the prod-
uct were calculated based on the following formulas:
2
3
Catalyst
Specific surface Area (m /g)
Pore volume (cm /g)
Cu/NaY
Cu/5A
Cu/HZSM-5
Cu/MCM-41
527
453
580
801
0.423
0.250
0.156
0.820
moles of glycerol converted
moles of initial glycerol
Conversion of glycerol(%) =
× 100%
with moles of glycerol converted = moles of initial glycerol − moles
of final glycerol.
5
A, HZSM-5 or MCM-41 zeolite, it can be obtained that the spe-
cific surface area or pore volume was not the crucial factor for the
vapor-phase synthesis of 3-methylindole.
It is known that metal copper is the active component for the
vapor-phase synthesis of 3-methylindole from glycerol and ani-
line who plays a crucial role in the production of 3-methylindole
moles of product formed
moles of initial glycerol
Yield of product(%) =
× 100%
yield of product
conversion of glycerol
Selectivity of product(%) =
× 100%
[
30]. The better the copper particles was dispersed, the higher the
activity was. Fig. 2 illustrates the XRD patterns of zeolites (NaY,
A, HZSM-5 and MCM-41)-supported copper catalysts. Two small
5
3
. Results and discussion
◦
◦
diffraction peaks of copper at 2ꢀ of 43.3 and 50.5 , assigned to
the (1 1 1) and (2 0 0) reflections, were observed on Cu/NaY [40].
On 5A, HZSM-5 or MCM-41-supported copper catalyst, the inten-
sity of copper diffraction peaks was much stronger than that on
3
.1. The synthesis of 3-methylindole over different zeolites
supported copper catalysts
◦
Cu/NaY, meanwhile, another diffraction peak at 74.7 , assigned to
3
.1.1. The activity and selectivity of different zeolites supported
the (2 2 0) reflection of copper, was observed and the intensity of
copper diffraction peaks increased in the order of Cu/5A, Cu/HZSM-
copper catalysts
In the reaction of glycerol and aniline over zeolite (NaY,
A, HZSM-5 or MCM-41)-supported copper catalyst, except
5
and Cu/MCM-41.
5
Based on the half peak width of (1 1 1) copper crystal phase, the
the aim product of 3-methylindole (3-MI), many by-products
were produced. Among them, N-methylaniline (N-MA), N-
ethylaniline (N-EA), 2,5-dimethyl-1-phenyl-pyrrole (2,5-DMPP)
and 2,3-dimethylindole (2,3-DMI) were the major by-products,
in addition, minor amounts of the other by-products were also
detected, such as ethylic acid, N-isopropylaniline, N-propylaniline,
propylene glycol, indole, 3-ketoneindole, N-isopropylindole and
N,N -dimethyl-N,N -diphenyl-1,2-ethanediamine. The reaction
pathway of glycerol and aniline to form the by-products is given
in Scheme 2 [3,29,36–39].
Table 1 is the activity and selectivity of the copper catalyst sup-
ported on NaY, 5A, HZSM-5 or MCM-41. It can be seen that Cu/NaY
exhibited higher activity and selectivity, which 72.0% glycerol con-
version with 56.1% selectivity of 3-methylindole was obtained.
Over Cu/5A or Cu/HZSM-5, N-ethylaniline was the major prod-
uct and the selectivity of 3-methylindole was only 32.2 or 30.0%.
Over Cu/MCM-41, larger amounts of undesired by-products were
produced, which the selectivity of 3-methylindole was less than
average sizes of copper particles on Cu/NaY, Cu/5A, Cu/HZSM-5 and
Cu/MCM-41 were calculated using the Scherrer Equation and the
results were 18, 33, 39 and 49 nm, respectively. The average size of
Cou (111)
ꢀ
ꢀ
Cou (200)
Cou (220)
(
4)
(
3)
(
2)
2
5%.
3
.1.2. Catalyst characterization
The specific surface area and pore volume on zeolite (NaY, 5A,
(1)
HZSM-5 or MCM-41)-supported copper catalyst are presented in
Table 2. It is clearly seen that Cu/MCM-41 possessed the largest spe-
2
3
30
40
50
60
70
80
cific surface area (801 m /g) and pore volume (0.820 cm /g), Cu/5A
2
displayed the lowest specific surface area (453 m /g) and Cu/HZSM-
2
( )
3
5
showed the lowest pore volume (0.156 cm /g). Considering the
activity and selectivity of the copper catalyst supported on NaY,
Fig. 2. XRD patterns of Cu/NaY (1), Cu/5A (2), Cu/HZSM-5 (3) and Cu/MCM-41 (4).

Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
241
O
aq
HO
O
O
O +
O
OH
NH2
-H2O
H2
N
H
2O
O
-
H
O
OH
H2
O
2
OH
OH
H
-H2
-
OH
NH2
H2O
H2
OH
N
H
O
-
O
2
H
-
Cat.
HO
OH
N
OH
NH2
H2O
N
H
H
2
N
H
N
CH OH
3
CH OH
2
-
,
2CH OH
3
CH2OH
HO
HO
NH2
H2O
+
C H OH
N
H
2
5
-
O
H 2
-
H 2
OH
CH OH
2
NH2
-H2
2CH OH
3
,
CH OH
-2H O
-2H2O
2
2
,
-
N
H
N
H
H
2
O
-
H
2
O
N
O
H
OH
-
H2O
N
H
HO
OH
NH2
O
-
4H2O
HO
OH
O
2H2
2
-H O
N
Scheme 2. The reaction pathway of glycerol and aniline to form the by-products over zeolite-supported copper catalyst.
copper particles on Cu/NaY was smallest compared with the other
catalysts, suggested a better dispersion of the copper particles on
the catalyst, while, on Cu/MCM-41, the average size of copper par-
ticles was the biggest although the catalyst possessed the largest
specific surface area. This should be the reason that the interaction
between copper particles and MCM-41 support was weak, which
made the copper particles gather together easily, as a result, big
copper particles could be formed on MCM-41 [41]. The interaction
between copper and NaY, however, was relatively strong, which
brought about the copper particles dispersed well on the support.
From Fig. 2, Tables 2 and 3 it can be obtained that there was no cor-
relation between the dispersion of copper particles and the specific
surface area or pore volume. Taking the results of Tables 1 and 3
into account, it can be concluded that Cu/NaY with smaller copper
particles was favorable to the synthesis of 3-methylindole.
decrease in the order of Cu/NaY, Cu/5A, Cu/HZSM-5 and Cu/MCM-
41, while, the areas of the middle-strong acid sites decreased in the
following sequence of Cu/NaY, Cu/MCM-41, Cu/5A and Cu/HZSM-
5. Taking into account the results of NH -TPD and Table 1 it
3
can be found that the more the weak acid sites, the higher the
selectivity of 3-methylindole. Moreover, it seemed to be conduc-
(
4)
(3)
Fig. 3 shows the acidity and acid strength distribution of
zeolites-supported copper catalysts. More than two desorption
peaks can be seen on the catalysts. The peaks at the lower tem-
(
2)
◦
perature (<200 C) were assigned to the weak acid sites, while the
◦
others at the higher temperature (200–400 C) were corresponded
to the middle-strong acid sites [42]. The areas of the weak acid
sites on the zeolites-supported copper catalysts were observed to
(
1)
Table 3
Average copper particle size of the copper catalysts supported on different zeolites.
1
00 150 200 250 300 350 400 450 500
Temperature (ºC)
Catalyst
Average copper particle size (nm)
Cu/NaY
Cu/5A
Cu/HZSM-5
Cu/MCM-41
18
33
39
49
Fig. 3. NH3-TPD profiles of Cu/NaY (1), Cu/5A (2), Cu/HZSM-5 (3) and Cu/MCM-41
4).
(

2
42
Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
Glycerol
-Methylindole
N-Methylaniline
N-Ethylaniline
Glycerol
1
00
100
3
3-Methylindole
A
N-Methylaniline
N-Ethylaniline
2,5-Dimethyl-1-phenyl-pyrrole
B
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
90
80
70
60
50
40
30
20
10
0
2,5-Dimethyl-1-phenyl-pyrrole
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Time on stream (h)
Time on stream (h)
Fig. 4. Glycerol conversion and the selectivity to different products versus time on stream over Cu/NaY (A) or Cu/NaY-Na2O (B).
tive to the formation of N-ethylaniline when the amount of the
middle-strong acid sites was very small.
somewhat with the increase of Na O amount, while, the selectivity
of N-methylaniline or 2,5-dimethyl-1-phenyl-pyrrole decreased.
2
For the vapor-phase synthesis of 3-methylindole from glycerol
and aniline, it is known that weak acid sites on Cu-based catalysts
are favorable to the production of 3-methylindole because they can
facilitate the adsorption of reactants on the surface of the cata-
lyst. Middle-strong acid sites, however, take disadvantage of the
reaction. Because 3-methylindole possesses weak alkalinity, it is
difficult for 3-methylindole to desorb from the middle-strong acid
sites, as a result, the further reaction could easily happen and bring
about the decrease of 3-methylindole selectivity [30]. There were
larger number of the weak acid sites and better dispersion of copper
particles on Cu/NaY, therefore the catalyst exhibited higher activity
and selectivity than Cu/5A, Cu/HZSM-5 or Cu/MCM-41.
Fig.
4 shows the activity and selectivity of Cu/NaY or
Cu/NaY–Na O as a function of reaction time with Na O amount
2
2
of 0.6 mmol/g. Over Cu/NaY, glycerol conversion decreased with
time on stream rapidly, while, the selectivity of 3-methylindole
increased during the first 2 h, then dropped. The changing trend
to the selectivity of 2,5-dimethyl-1-phenyl-pyrrole was the same
as 3-methylindole. The selectivity of N-methylaniline increased a
little with time on stream, while, the selectivity of N-ethylaniline
decreased slightly. A similar phenomenon was seen over Cu/NaY-
Na O, but the selectivity of 3-methylindole were much higher
2
than that over Cu/NaY and the decline greatly slowed down
within the same period, indicating that Na O promoter had a
2
substantial improvement not only in increasing the selectivity of
3
-methylindole but also in decelerating the deactivation of the cat-
alyst.
Fig. 5 shows the XRD patterns of NaY, the fresh or the used sam-
3
.2. The effect of Na O promoter on the performance of the
2
Cu/NaY catalyst
ples of Cu/NaY and Cu/NaY–Na O. Compared with the XRD pattern
of the bare NaY support, no obvious change for the crystal diffrac-
tion peaks of NaY was observed on the fresh or the used Cu/NaY
From Fig. 3 it is known that there were many middle-strong
acid sites on Cu/NaY, which can bring about some side reactions
caused by 3-methylindole. In order to decrease the middle-strong
2
and Cu/NaY–Na O, suggested that the crystal structure of NaY was
acid sites on Cu/NaY, Na O was introduced into Cu/NaY and Table 4
2
2
exhibits the effect of Na O amount on the activity and selec-
2
tivity of Cu/NaY–Na O catalyst. As expected, the selectivity of
2
3
-methylindole increased remarkably after adding Na O to Cu/NaY
2
although the conversion of glycerol decreased slightly. And with
Cu (111)
o
the increase of Na O amount, the selectivity of 3-methylindole
2
increased first, then, decreased. When the amount of Na O was
2
Cu (200)
0
.60 mmol/g, the selectivity of 3-methylindole reached 72.3%,
o
Cu (220)
o
which the yield of 3-methylindole was up to 50.4%. From Fig. 3
it can also be found that the selectivity of N-ethylaniline increased
(
5)
Table 4
Effect of Na2O amount on the activity and selectivity of Cu/NaY–Na2O catalyst.
(4)
3)
(2)
1)
Na2O Amount
mmol/g)
Glycerol
conversion
Product selectivity (%)
(
(
(%)
3-MI N-MA N-EA 2,5-DMPP Others
0.0
0.2
0.4
0.6
0.8
1.0
72.0
71.1
70.5
69.7
68.9
67.8
56.1
60.3
65.5
72.3
65.2
60.5
3.1
2.9
2.7
2.5
2.2
2.0
5.5
6.3
7.6
8.7
9.8
19.1
17.6
16.0
15.3
13.5
11.4
16.2
12.9
8.2
1.2
9.3
(
3
0
40
50
60
70
80
10.9
13.2
2
◦
−1
,
Reaction conditions: 220 C, aniline/glycerol = 3:1 molar ratio, SV = 1700 h
−
1
LHSV = 0.4 h , H2 = 20 mL/min, steam = 6 mL/min, N2 = 54 mL/min. Copper loading:
.4 wt%. The results were taken at the second hour.
Fig. 5. XRD patterns of the fresh samples of NaY (1), Cu/NaY (2), Cu/NaY-Na2O (3)
5
and the used samples of Cu/NaY (4) and Cu/NaY-Na2O (5).

Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
243
7
7
7
7
7
.8
.6
.4
.2
.0
(4)
(3)
6.8
6
6
6
6
.6
.4
.2
.0
(2)
(
2)
(
1)
(1)
1
00
150
200
250
300
350
400
450
0
100 200 300 400 500 600 700 800
Temperature (ºC)
Temperature (ºC)
Fig. 6. NH3-TPD profiles of NaY (1), NaY-Na2O (2), Cu/NaY (3) and Cu/NaY-Na2O (4).
Fig. 7. TG curves of the used samples of Cu/NaY (1) and Cu/NaY-Na2O (2).
decreased the amount of the middle-strong acid sites clearly, the
selectivity of 3-methylindole increased greatly.
remained well on the catalysts. On the fresh Cu/NaY, only two cop-
per diffraction peaks can be seen, while, on the fresh Cu/NaY-Na O,
2
TG–DTA analysis was used to investigate the coke formation of
three copper diffraction peaks were observed and the intensity
was stronger than that on the fresh Cu/NaY catalyst. On the used
NaY-supported copper catalysts, the intensity of copper diffraction
peaks was much stronger than that on the fresh catalysts and the
intensity of copper diffraction peaks was almost the same strong.
Table 5 presents the average sizes of copper particles on the
the used Cu/NaY and the catalyst modified by Na O. The results
2
are shown in Figs. 7 and 8. From Fig. 7 the obvious mass loss was
observed on the catalysts, indicated that there existed the forma-
tion of coke during the reaction. On the used Cu/NaY–Na O, the
2
mass loss was much less than that on the used Cu/NaY, revealed
that Na O could inhibit the formation of the coke effectively. From
2
fresh or the used Cu/NaY and Cu/NaY–Na O based on the Scherrer
2
Figs. 6 and 7, it can easily be found that the amount of the coke was
relevant to the amount of acid sites on the catalyst. With fewer acid
sites, there was less coke.
Equation. It can be seen that the average size of copper particles on
the fresh Cu/NaY–Na O was bigger than that on the fresh Cu/NaY,
2
indicated that the addition of Na O to Cu/NaY did not promote the
2
In Fig. 8, two exothermic peaks were observed, indicated that
dispersion of copper particles on the support. Compared with the
fresh catalysts, the average sizes of copper particles on the used Cu-
based catalysts became much bigger, suggested that there existed
the sintering of copper particles during the reaction. On the used
◦
there existed two types of coke. The exothermic peak at 340 C was
assigned to the removal of easily oxidized carbon, while, another
◦
exothermic peak at 484 C was the oxidation of regular or graphitic
carbon [43]. After adding Na O to Cu/NaY, the amount of the easily
2
Cu/NaY–Na O, the average size of copper particles was the same as
2
oxidized carbon increased, while, the amount of regular or graphitic
that on the used Cu/NaY, revealed that Na O did not aggravate the
2
sintering of copper particles.
Fig. 6 shows the NH -TPD profiles of NaY, NaY–Na O, Cu/NaY
3
2
and Cu/NaY–Na O. On NaY or NaY–Na O, there existed a lot of
2
2
(
2)
1)
weak acid sites, while, few or no middle-strong acid sites was
observed. On NaY supported copper catalyst, the amount of the
weak acid sites decreased slightly, however, large amount of the
◦
middle-strong acid sites appeared at about 360 C. After adding
Na O to Cu/NaY, the amount of the weak acid sites or the middle-
2
(
strong acid sites decreased and the amount of the middle-strong
acid sites decreased more remarkably.
From the experiments it was known that NaY or NaY–Na O
2
exhibited very low activity and selectivity with the yield of
3
-methylindole less than 8%, which revealed again that metal
copper is the active component for the vapor-phase synthesis
of 3-methylindole from glycerol and aniline [30]. Because Na O
2
Table 5
Average copper particle size of Cu/NaY and Cu/NaY-Na2O.
Catalyst
Average copper particle size (nm)
Fresh sample
0
100 200 300 400 500 600 700 800
Temperrature (ºC)
Used sample
Cu/NaY
Cu/NaY-Na2O
18
23
41
41
Fig. 8. DTA profiles of the used samples of Cu/NaY (1) and Cu/NaY-Na2O (2).

2
44
Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
H
- H O
2
CH2
OH
C
HO
HO
OH
HO
OH2
HO
OH
D
OH
A
OH
B
O
HO
- H
HO
HO
H
OH
O
OH
OH
H
E
F
G
H
N
OH
OH
OH
H
2
NH2
N
N
-
H O
2
OH
O
I
J
K
OH2
N
N
-
H O
2
- H
H
N
N
H
L
M
N
O
Scheme 3. A probable mechanism for the synthesis of 3-methylindole from glycerol and aniline over zeolites-supported Cu-based catalysts.
carbon decreased. Compared with the used Cu/NaY, there was no
obvious shift for the temperature of the exothermic peaks on the
4. Conclusions
used Cu/NaY–Na O, indicated that the texture of the coke was not
The vapor-phase synthesis of 3-methylindole from glycerol and
aniline over zeolites-supported copper-based catalysts was stud-
ied. Cu/NaY catalyst, in terms of both the conversion of glycerol and
the selectivity to 3-methylindole, was found to be better than 5A,
HZSM-5 or MCM-41–supported copper catalyst, which the yield of
2
changed after adding Na O to Cu/NaY.
2
3.3. Probable catalytic mechanism for the vapor-phase synthesis
of 3-methylindole
3
-methylindole reached 40.3%. The higher activity and selectivity
of Cu/NaY stemmed from its better dispersion of copper particles
It has been reported that glycerol can adsorb on acid sites
and be converted to different products through the intermediate
of acetol [7,44–46]. In order to shed light on the catalytic syn-
thesis mechanism of 3-methylindole from glycerol and aniline
over zeolites-supported Cu-based catalysts, the reaction products
of glycerol with aniline were carefully analyzed by GC–MS. In
addition, the conversion of glycerol over the catalysts was also
investigated and acetol was found as the major product. Taking
into account the experimental results and the relevant literatures
and larger amount of the weak acid sites. The addition of Na O to
2
Cu/NaY not only increased the selectivity of the catalyst remark-
ably, which 72.3% selectivity with 50.4% yield for 3-methylindole
was achieved, but also decelerated the deactivation of the catalyst,
because Na O could decrease the amount of middle-strong acid
2
sites of the catalyst remarkably and inhibit the formation of the
coke during the reaction. Adding Na O to Cu/NaY did not change
2
the texture of the coke. Based on the results of the experiments and
the relevant literatures, a possible mechanism for the synthesis of
[
46–50], a probable mechanism for the catalytic synthesis of 3-
3
-methylindole from glycerol and aniline over zeolites–supported
methylindole was hypothesized (Scheme 3).
Cu-based catalysts was proposed, in which acetol was the interme-
diate to produce 3-methylindole.
Glycerol was firstly treated by protonation and dehydration to
form the carbocation (C). C, then, underwent rearrangement to give
the intermediate of D or E. E was subsequently converted to acetol
Acknowledgement
(
F) by losing a hydrogen proton [7,45–47], and F was readily trans-
formed to 2-hydroxy-1-propanal (H) via enol-keto tautomerism.
The amine cation (I) was formed through the reaction of H with
aniline, and I underwent the migration of hydrogen proton and
dehydrated to form the imine (K) [46,47]. A hydrogen proton was
soon transferred to K to form the intermediate (L), and L was finally
converted to the aim product of 3-methylindole by dehydration and
eliminating a hydrogen proton.
It should be noted especially that copper particles with high dis-
persion played an important role in the synthesis of 3-methylindole
from glycerol and aniline because their empty orbits could receive
the electrons from glycerol or aniline, which was favorable for the
adsorption of reactants on the surface of catalyst. More importantly,
Cu had a good catalytic activity to the cleavage of C O bond [47,51]
which promoted the formation of C, K or M. In the reaction, there-
fore, copper particles and weak acid sites acted together to promote
the synthesis of 3-methylindole.
We gratefully acknowledge the financial support provided by
the National Natural Science Foundation of China (No. 21173110).
References
[1] T. Miyazawa, Y. Kusunoki, K. Kunimori, K. Tomishige, J. Catal. 240 (2006)
2
13–221.
2] A. Behr, J. Eilting, K. Irawadi, J. Leschinski, F. Lindner, Green Chem. 10 (2008)
3–30.
[3] I. Gandarias, P.L. Arias, J. Requies, M.B. Güemez, J.L.G. Fierro, Appl. Catal. B 97
2010) 248–256.
[
1
(
[
[
4] W. Suprun, M. Lutecki, T. Haber, H. Papp, J. Mol. Catal. A 309 (2009) 71–78.
5] J. George, Y. Patel, S.M. Pillai, P. Munshi, J. Mol. Catal. A 304 (2009) 1–7.
[6] R.B. Mane, C.V. Rode, Green Chem. 14 (2012) 2780–2789.
[7] F. Cavani, S. Guidetti, L. Marinelli, M. Piccinini, E. Ghedini, M. Signoretto, Appl.
Catal. B 100 (2010) 197–204.
[
8] D. Liang, J. Gao, H. Sun, P. Chen, Z.Y. Hou, X.M. Zheng, Appl. Catal. B 106 (2011)
23–432.
4

Y. Cui et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 238–245
245
[
9] R.K.P. Purushothaman, J. Haveren, D.S. Es, I. Melián-Cabrera, H.J. Heeres, Green
Chem. 14 (2012) 2031–2037.
[32] M.A. Alotaibi, E.F. Kozhevnikova, I.V. Kozhevnikov, J. Catal. 293 (2012) 141–144.
[33] J.K. Nam, M.J. Choi, D.H. Cho, J.K. Suh, S.B. Kim, J. Mol. Catal. A 370 (2013) 7–13.
[34] T. Subramanian, M. Kumarraja, K. Pitchumani, J. Mol. Catal. A 363–364 (2012)
115–121.
[35] L.Y. Chen, T. Horiuchi, T. Mori, K. Maeda, J. Phys. Chem. B 103 (1999) 1216–
1222.
[
[
10] L. Li, T.I. Korányi, B.F. Sels, P.P. Pescarmona, Green Chem. 14 (2012) 1611–1619.
11] L.J. Zhu, P.J. Guo, X.W. Chu, S.R. Yan, M.H. Qiao, K.N. Fan, X.X. Zhang, B.N. Zong,
Green Chem. 10 (2008) 1323–1330.
[
[
[
[
12] S. Kamijo, Y. Yamamoto, J. Org. Chem. 68 (2003) 4764–4771.
13] D.G. Hulcoop, M. Lautens, Org. Lett. 9 (2007) 1761–1764.
[36] M.Á. Aramendía, V. Borau, C. Jiménez, J.M. Marinas, F.J. Romero, Appl. Catal. A
183 (1999) 73–80.
[37] S. Narayanan, A. Sultana, Appl. Catal. A 167 (1998) 103–111.
[38] J. Liu, C.M. Li, X.M. Huo, L. Shi, Q. Sun, Chin. J. Catal. 29 (2008) 159–162.
[39] L.C.A. Oliveira, M.F. Portilho, A.C. Silva, H.A. Taroco, P.P. Souza, Appl. Catal. B
117–118 (2012) 29–35.
[40] D.H. Ji, W.C. Zhu, Z.L. Wang, G.J. Wang, Catal. Commun. 8 (2007) 1891–1895.
[41] Y.J. Huang, H.P. Wang, J.F. Lee, Chemosphere 50 (2003) 1035–1041.
[42] D.F. Jin, Z.Y. Hou, L.W. Zhang, X.M. Zheng, Catal. Today 131 (2008) 378–384.
[43] N.A. Hermes, M.A. Lansarin, O.W. Perez-Lopez, Catal. Lett. 141 (2011)
1018–1025.
[44] Y. Kusunoki, T. Miyazawa, K. Kunimori, K. Tomishige, Catal. Commun. 6 (2005)
645–649.
[45] Z.L. Yuan, L.N. Wang, J.H. Wang, S.X. Xia, P. Chen, Z.Y. Hou, X.M. Zheng, Appl.
Catal. B 101 (2011) 431–440.
[46] S.H. Zhu, Y.L. Zhu, S.L. Hao, H.Y. Zheng, T. Mo, Y.W. Li, Green Chem. 14 (2012)
2607–2616.
[47] M.A. Dasari, P.P. Kiatsimkul, W.R. Sutterlin, G.J. Suppes, Appl. Catal. A 281 (2005)
225–231.
14] C.S. Xie, Y.H. Zhang, Z.D. Huang, P.Z. Xu, J. Org. Chem. 72 (2007) 5431–5434.
15] M. Howe-Grant, Kirk-Othmer Encyclopedia of Chemical Technology, 4th ed.,
New York : John Wiley & Sons, 1995, p. 161.
[
[
[
16] M. Howe-Grant, Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed,
New York : John Wiley & Sons, 1989, p. 213.
17] E. Doran, F.W. Whittington, J.D. Wood, J.D. McGivan, Chem. Biol. Interact. 140
(
2002) 81–92.
18] G. Chen, R.A. Cue, K. Lundstrom, J.D. Wood, O. Doran, Drug Metab. Dispos. 36
2008) 56–60.
(
[
[
[
[
[
[
[
[
[
[
19] J.M. Weems, G.S. Yost, Chem. Res. Toxicol. 23 (2010) 696–704.
20] B. Deslandes, C. Gariépy, A. Houde, Livest. Prod. Sci. 71 (2001) 193–200.
21] M. Bandini, A. Eichholzer, Angew. Chem. Int. Ed. 48 (2009) 9608–9644.
22] G.R. Humphrey, J.T. Kuethe, Chem. Rev. 106 (2006) 2875–2911.
23] C.A. Simoneau, A.M. Strohl, B. Ganem, Tetrahedron Lett. 48 (2007) 1809–1811.
24] M.A. Fox, D.A. Chandler, C. Lee, J. Org. Chem. 56 (1991) 3246–3255.
25] B. Robinson, Chem. Rev. 63 (1963) 373–401.
26] P. Magnus, L.S. Mitchell, Tetrahedron Lett. 39 (1998) 4595–4598.
27] C.S. Cho, J.H. Kim, T.J. Kim, S.C. Shim, Tetrahedron 57 (2001) 3321–3329.
28] D.V. Gopal, B. Srinivas, V. Durgakumari, M. Subrahmanyam, Appl. Catal. A 224
[48] M. Campanati, A. Vaccari, O. Piccolo, Catal. Today 60 (2000) 289–295.
[49] A. Corma, G.W. Huber, L. Sauvanaud, P. O’Connor, J. Catal. 257 (2008) 163–171.
[50] J.J. Li, Name Reactions A Collection of Detailed Reaction Mechanisms, 3rd ed.,
Springer, Berlin, Heidelberg, 2006, pp. 545.
[51] Z.W. Huang, F. Cui, H.X. Kang, J. Chen, X.Z. Zhang, C.G. Xia, Chem. Mater. 20
(2008) 5090–5099.
(
2002) 121–128.
[29] M. Campanati, S. Franceschini, O. Piccolo, A. Vaccari, J. Catal. 232 (2005) 1–9.
[30] W. Sun, D.Y. Liu, H.Y. Zhu, L. Shi, Q. Sun, Catal. Commun. 12 (2010) 147–150.
[31] M. Karthik, C.J. Magesh, P.T. Preumal, M. Palanichamy, B. Arabindoo, V. Muruge-
san, Appl. Catal. A 286 (2005) 137–141.