Influence of Al3+ insertion in the stannosilicate MFI framework on the catalytic performance in vapor phase aniline N-methylation

Article abstract of DOI:10.1016/j.apcata.2011.05.019

Silica-based Sn-MFI, Al-MFI and Al-Sn-MFI molecular sieves were synthesized by hydrothermal crystallization of gels having molar compositions SiO ₂:xSnO₂:yAl₂O₃:0.23 (TPA) ₂O:35H₂O, where x ranges from 1/0 to 1/200 and y from 1/0 to 1/400. Keeping molar Si/(Al + Sn) = 50, the amount of tin and aluminum in hydrogel was varied (Sn:Al molar ratios = 1:0, 1:0.33, 1:1, 1:3, 0:1) to investigate the synergy between Lewis and Br?nsted acid sites in acid catalyzed aniline N-methylation reaction. Catalyst characterization was done by Powder X-ray diffraction, DRUV-vis spectroscopy, temperature programmed ammonia desorption (TPAD) and FTIR spectroscopy. An increase in the Al³⁺ insertion in Sn-MFI framework resulted in the increase in the stronger acid sites. Al-Sn-MFI showed higher aniline conversion than their monometallic counterparts. A sample having B/L ratio in between 0.67 and 0.91 was found to be optimum for maximizing the NMA yield, indicating the existence synergistic properties of Al-Sn-MFI. Upon process parameter optimization, the optimum sample M50 [Si/Al = 50.2, Si/Sn = 93.8, Si/(Al + Sn) = 43.2] showed the maximum aniline conversion (67%) and NMA selectivity (81%) at reaction temperature = 220 °C, weight hourly space velocity (WHSV) = 3 h^-1, molar ratio (aniline to methanol) = 1:8 and TOS = 4 h.

Full text of DOI:10.1016/j.apcata.2011.05.019

Applied Catalysis A: General 401 (2011) 182–188
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
3+
Influence of Al insertion in the stannosilicate MFI framework on the catalytic
performance in vapor phase aniline N-methylation
∗
∗
P.S. Niphadkar, P.N. Joshi , S.S. Deshpande, V.V. Bokade
Catalysis Division, National Chemical Laboratory (CSIR), Pune 411008, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Silica-based Sn-MFI, Al-MFI and Al-Sn-MFI molecular sieves were synthesized by hydrothermal crys-
tallization of gels having molar compositions SiO2:xSnO2:yAl2O3:0.23 (TPA)2O:35H2O, where x ranges
from 1/0 to 1/200 and y from 1/0 to 1/400. Keeping molar Si/(Al + Sn) = 50, the amount of tin and alu-
minum in hydrogel was varied (Sn:Al molar ratios = 1:0, 1:0.33, 1:1, 1:3, 0:1) to investigate the synergy
between Lewis and Brønsted acid sites in acid catalyzed aniline N-methylation reaction. Catalyst char-
acterization was done by Powder X-ray diffraction, DRUV–vis spectroscopy, temperature programmed
Received 11 January 2011
Received in revised form 2 May 2011
Accepted 14 May 2011
Available online 20 May 2011
Keywords:
Al-Sn-MFI
Brønsted and Lewis acid
Aniline
Methanol
3
+
ammonia desorption (TPAD) and FTIR spectroscopy. An increase in the Al insertion in Sn-MFI framework
resulted in the increase in the stronger acid sites. Al-Sn-MFI showed higher aniline conversion than their
monometallic counterparts. A sample having B/L ratio in between 0.67 and 0.91 was found to be optimum
for maximizing the NMA yield, indicating the existence synergistic properties of Al-Sn-MFI. Upon process
parameter optimization, the optimum sample M50 [Si/Al = 50.2, Si/Sn = 93.8, Si/(Al + Sn) = 43.2] showed
N-methylaniline
◦
the maximum aniline conversion (67%) and NMA selectivity (81%) at reaction temperature = 220 C,
−1
weight hourly space velocity (WHSV) = 3 h , molar ratio (aniline to methanol) = 1:8 and TOS = 4 h.
2011 Elsevier B.V. All rights reserved.
©
1
. Introduction
rials (Al-MCM-41) [24]. Beside reaction parameters, the factors
contributing significantly to the performance and product distri-
bution of the acid-catalyzed organic transformation reaction are
the type, number and the strength of acid sites in the solid cata-
lyst. An influence of Lewis and Brønsted acid sites on conversion
of aniline to alkylated aniline products has already been reported
[17,25–27]. The use of strong Brønsted acid sites leads to increase
the aniline conversion with more formation of byproducts such as
N-dimethyl aniline (NNDMA), C-alkylated products and the coke.
On other hand, the use of Lewis acid sites leads to increase the
selectivity of N-alkylated products at the cost of aniline conver-
sion. The simultaneous increase in both the Lewis and Brønsted
acid sites has been reported [28] to improve the aniline conversion
over vanadia-impregnated montmorillonite K-10. However, so far,
there is no report on systematic studies on the investigating the
synergy between Lewis and Brønsted acid sites to maximize the
aniline conversion as well as N-methyl aniline selectivity.
N-alkylation of aniline is commercially important reaction as
the alkylated products are valuable materials for the synthe-
sis of dyes, rubbers, explosives, herbicides and pharmaceuticals
[
1–10]. Due to excellent antioxidant and antidetonant proper-
ties, N-methylaniline (NMA) has identified as potential additive
for improving the quality and performance of commercial prod-
ucts such as gasoline, rubbers, polymers, latexes, etc. [11–13]. The
conventional methods for the bulk production of NMA involve liq-
uid phase alkylation using mineral acids and Friedel-Crafts type
catalysts, which lead to the serious disadvantages such as cor-
rosion, pollution and formation of non-recyclable by-products.
Hence, solid acids as catalyst have attracted much attention for
the development of environmental benign process for the synthe-
sis of NMA. In view of this, vapor phase alkylation of aniline with
methanol was studied over variety of solid catalysts such as metal
oxides (␥-Al O , MgO, spinel Fe O -ZnO structure) [14–16], zeo-
Because of the Brønsted/Lewis acid synergy, a remarkable
increase in the acid strength of dealuminated HY zeolite is reported
[29]. Similarly, mixed Al-Zr-TUD-1 catalyst demonstrated the syn-
ergy for Brønsted acid catalyzed Prins cyclisation reaction [30].
An incorporation of Zn or Fe into Al-MCM-41 was also reported
2
3
2
3
lites (X,Y,L and ZSM-5) [17–19], aluminophosphate [20], clay [21],
vanadia-impregnated silica/clay [22,23] and mesoporous mate-
[
31] for the generation of strong Brønsted acid sites, which lead to
∗ Corresponding authors. Tel.: +91 20 25902458; fax: +91 20 25902634.
E-mail addresses: pn.joshi@ncl.res.in (P.N. Joshi), vv.bokade@ncl.res.in
V.V. Bokade).
increase the toluene conversion in toluene isopropylation. The cre-
ation/enhancement of new redox and acidic sites in multi-metallic
(
0
926-860X/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.05.019

P.S. Niphadkar et al. / Applied Catalysis A: General 401 (2011) 182–188
183
molecular sieve were observed as compared to monometallic coun-
terpart [32–39]. A combination of Brønsted and Lewis acid sites on
the catalyst surface was reported [40] to favor the aniline alkyla-
tion reaction. Moreover, in Al-Zr-beta catalyst, it is reported [41]
that, presumably Zr4+ may strengthen the Brønsted acid sites and
hence influence the isomerization of m-xylene. The information
on prerequisites and eco-friendly considerations has triggered our
interest in undertaking the research work on the synergy between
Lewis and Brønsted acid sites for maximizing the aniline conversion
and NMA selectivity.
cold water. The solid was recovered by filtration, then washed thor-
◦
oughly with deionised water, dried at 120 C and finally calcined at
◦
550 C for 12 h.
2.2. Characterization
The phase identification, degree of crystallization and purity
were determined by XRD. The XRD patterns were recorded on X-
ray diffractometer (Rigaku Miniflex, Japan) using Cu-K␣ radiations.
To elucidate the nature of tin species, a DRUV–vis spectrum was
recorded with a PerkinElmer Lambda 6 spectrometer (Lambda 650)
using BaSO4 as a reference.
Recently, we have reported [42] that, Sn-MFI, a Lewis acid
solid catalyst, has exhibited promising results in aniline alkyla-
tion with methanol. In view of maximizing the aniline conversion
and NMA selectivity, we have investigated the synergetic effect of
Lewis and Brønsted acid sites keeping total number of acid sites
identical in Sn-MFI catalyst. The samples containing both Lewis
and Brønsted acid sites were prepared by isomorphous substitu-
The overall acidity and acid strength associated with sites were
measured by TPAD using a Micromeritics AutoChem (2910, USA)
equipped with thermal conductivity detector. Prior to the measure-
◦
3
−1
ments, sample was dehydrated at 400 C in He (30 cm min ) for
1
◦
h. The temperature was then decreased to 50 C and then NH3
3
+
4+
tion of Al in place of Sn in Sn-MFI framework. The percentage
substitution of Al3+ was varied from 0 to 100. The samples were
characterized by various characterization techniques such as pow-
der X-ray diffraction (XRD), DRUV–vis spectroscopy, temperature
programmed ammonia desorption (TPAD), and FTIR spectroscopy.
Catalysts with different compositions were screened for aniline N-
alkylation with methanol. The optimization of process parameters
using optimum catalyst was performed to maximize the aniline
conversion and NMA selectivity.
was allowed to adsorb by exposing sample to a gas stream contain-
ing 10% NH3 in He for 1 h. It was then flushed with He for another
3
−1
1
h. The NH3 desorption was carried out in He flow (30 cm min
)
◦
by increasing the temperature up to 450 C with a heating rate of
0 C min
The nature of acid sites (Brønsted and/or Lewis) in the catalyst
◦
−1
.
1
was elucidated by ex situ pyridine-FTIR. After activating a cata-
◦
lyst powder sample at 400 C for 2 h, it was cooled down to room
temperature under high vacuum. It was then exposed to pyridine
vapors for 2 h. The physisorbed pyridine was driven off by acti-
◦
2
. Experimental
vating at 150 C for 2 h under high vacuum. The FTIR spectra were
recorded on a Shimadzu (Model-820 PC) spectrophotometer under
DRIFT (Diffuse Reflectance Infrared Fourier Transform) mode.
The chemicals used were tetraethyl ortho silicate (TEOS, 98%
purity, Aldrich), tetra propyl ammonium hydroxide (aq. 20% TPAOH
◦
Low temperature (−196 C) nitrogen adsorption and desorp-
soln., V. P. Chemical, Pune), SnCl ·5H O (98%, Loba), aluminum
tion isotherms were obtained using NOVA-1200(Quanta chrome,
4
2
◦
isopropoxide (98% Acros), distilled water, aniline (99.5%), and
methanol (99.9% Rankem).
USA). The calcined sample was degassed at 300 C for 10 h prior
to measurements. The specific surface area was calculated using
BET method. The chemical composition of the samples was estab-
lished by following conventional elemental analysis methods and
provided in Table 1.
2
.1. Catalyst synthesis
Keeping overall molar Si/(Al + Sn) ratio of 50 fixed, catalysts
with varying molar Sn:Al ratios (ca. 1:0, 1:0.33, 1:1, 1:3, 0:1) were
hydrothermally synthesized from gels having molar compositions
2.3. Aniline alkylation
Vapor phase aniline alkylation with methanol was carried out
of SiO :xSnO :yAl O :0.23 (TPA) O:35H O, where x ranges from
2
2
2
3
2
2
in a down flow tubular SS316 (40 cm length × 2 cm internal diam-
eter) reactor. The reactor was heated to the desired temperature
with an electrical tubular furnace having digital temperature con-
troller cum indicator. About 4 g of the catalyst (20–40 mesh size)
was sandwiched at the centre by inert ceramic beads. Before start of
1
/0 to 1/200 and y from 1/0 to 1/400. The well crystalline Sn-
MFI phases with no amorphous or impurity phase/s was obtained
by establishing kinetics of crystallization. Upon calcination, the
final products were designated as M100, M75, M50, M25 and M0,
where the numbers indicate the percentage of Sn moles in total
−
1
the reaction, the catalyst was activated in a N2 flow (50 ml min
)
(
Sn + Al) moles present in the initial hydrogel. The initial hydrogel
◦
at 400 C for 2 h. After catalyst activation, the catalyst tempera-
ture was reduced to desired reaction temperature in nitrogen flow.
A mixture of aniline and alcohol was fed at a required rate. The
product was collected at a regular interval of time. The liquid mass
balance for all the reactions was observed to be 98 ± 2%.
was prepared by mixing (1) a solution containing required amount
of aluminum isopropoxide dissolved into required amount of aq.
TPAOH, and (2) a solution prepared by addition of aqueous solution
of SnCl ·5H O to an appropriate amount of TEOS under constant
4
2
stirring. Then, the stirring was continued further for 1 h. A clear gel
thus obtained was then transferred to an autoclave and placed in
The feed and products were analyzed by Gas Chromatography
(HP 5890A) using a flame ionization detector and Carbowax column
◦
an oven for crystallization at 160 C for desired period (up to 40 h).
The autoclave was then removed from the oven and quenched with
(30 m length × 0.3 mm diameter) with hydrogen as a carrier gas.
Table 1
Unit cell composition, BET surface area and acidity data of all the samples.
2
Sample designation
Unit cell composition (dry basis)
BET Surface area (m /g)
mmol of NH3 desorbed /g
Total
Weak
Rel. strong
M100
M75
M50
M25
M0
(SiO2)93.96 (SnO2)2.04
431
429
415
405
390
0.53
0.55
0.59
0.65
0.73
0.53
0.35
0.38
0.42
0.45
–
(SiO2)94.02 (SnO2)1.46 (AlO2)0.52
(SiO2)93.83 (SnO2)1.00 (AlO2)1.17
(SiO2)93.8 (SnO2)0.48 (AlO2)1.72
(SiO2)93.4 (AlO2)2.6
0.19
0.21
0.23
0.28

1
84
P.S. Niphadkar et al. / Applied Catalysis A: General 401 (2011) 182–188
3. Results and discussion
3.1. Characterization
(M 0 )
The powder XRD patterns of M100, M75, M50, M25 and M0
in their as-prepared form are depicted in Fig. 1. All samples have
shown well crystalline nature and their XRD data match well with
that of reported MFI phase [40]. No any contributions due to amor-
phous phase or other crystalline phase/s (impurity) were observed.
The DRUV–vis spectra of M100, M75, M50 and M25 were
(
M
2 5 )
5 0 )
recorded for the verification of the framework occupancy of Sn4+
.
All these spectra are presented in Fig. 2. Each sample has shown
an absorption at ∼220 nm which has been assigned to charge tran-
sitions (CT) from O2 to Sn in tetrahedral coordination [42–44].
Moreover, the intensity of this band was found to decrease with the
decrease in the tin content from M100 to M25. All spectra showed
no bands typical of polymeric Sn–O–Sn type species at 280 nm
−
4+
(
M
[
42–44] indicating the absence of extra-framework occluded tin
(
M
7 5 )
oxide. The identical spectral features in all these samples suggest
the occupancy of Sn4+ species in MFI framework is in tetrahedral
coordinated irrespective of presence of Al in variable proportion.
The chemical composition expressed as unit cell composition of
all the samples is provided in Table 1. The Al and Sn was found to
be absent in the sample M100 and M0, respectively. The Al content
in the samples showed the trend as M0 > M25 > M50 > M75 > M100
which is in close agreement with their initial gel molar composi-
tions. Similar but reverse trend was observed in case of Sn content.
The BET surface area of all the samples is tabulated in Table 1 and
(M
1 0 0 )
40
10
20
30
50
2
θ
2
they are in the range of 390–431 m /g, which agreed well with
Fig. 1. Powder XRD patterns of M100, M75, M50, M25 and M0 samples in their as
prepared form.
those of zeolites with the MFI structures.
An understanding of type, concentration and strength of acid
sites in solid acids is one of the bare necessities to understand
their synergetic effect on catalytic efficiency and behavior. Hence,
all the samples were subjected to TPAD evaluation. Fig. 3 exhibits
M100
M75
0.1 abs
M50
M75
200
250
300
350
400
450
500 200
250
300
350
400
450
500
Wavelength (nm)
Wavelength (nm)
Fig. 2. Diffuse reflectance UV–vis spectra of M100, M75, M50 and M25.

P.S. Niphadkar et al. / Applied Catalysis A: General 401 (2011) 182–188
185
M75
M100
1
00
200
300
400
500
100
200
300
400
500
o
Temperature oC
Temperature C
M50
M25
1
00
200
300
400
500
100
200
300
400
500
o
Temperature o
Temperature C
C
M0
1
00
200
300
400
500
o
Temperature C
Fig. 3. Deconvoluted temperature programmed ammonia desorption profiles.
deconvoluted TPAD profiles of all samples. The deconvolution was
performed using origin 8 software.
with the increase in the Al content with simultaneous decrease in
Sn content.
−
1
Upon deconvolution, M100 showed two desorption peaks. The
The bands appearing at 1545 and/or 1455 cm
in pyridine-
◦
maximum of first peak appeared at 111 C and second peak max-
FTIR are assigned to pyridine coordinated to Brønsted and Lewis
acid site, respectively [45]. In present studies, the nature of the
acid sites in all the samples was characterized by ex situ FTIR
spectroscopy with chemisorbed pyridine. The pyridine-FTIR spec-
tra of all the samples are illustrated in Fig. 4. The M100 sample
◦
imum was at 209 C. Except M100, all other samples exhibited
three peaks in their respective deconvoluted patterns appearing
◦
◦
◦
at 109 ± 7 C, 189 ± 4 C and 381 ± 21 C. First and second peaks in
all the samples can be assigned to desorption of physisorbed NH3
−
1
and to desorption of NH from weak acid sites, respectively [42]. All
showed two bands, one at 1455 cm for Lewis acid site and another
3
−
1
the samples, except M100, showed the existence of 3rd peak which
can be assigned to relatively stronger acid sites. The area under this
peak was found to increase with the increase in Al concentration.
Thus, the origin of the relatively stronger acid sites in these sam-
at 1490 cm . The sample M0 also exhibited two bands, one at
−
1
1545 cm due to Brønsted-bound pyridine (pyridinium ion) and
another at 1490 cm . The common band at 1490 cm in both
−
1
−1
M100 and M0 might be due to the contribution of Lewis and/or
ples might be the framework occupancy of Al3 . Furthermore, the
peak maximum was found to shift to higher temperature with the
increase of Al. Thus, it seems that, with the increase in Al concen-
tration, not only the population but also strength associated with
these sites also increases. On the other hand, with the decrease
in the Sn content, the only decrease in the area of the 2nd peak
was observed in M75 as compared to M100. Surprisingly, with the
further decrease in Sn content, the area of this peak was found to
increase. Probably, higher degree of Al insertion in the framework
might be responsible for the creation of stronger as well as weaker
acid sites. It is evident from Table 1 that, the total acidity increases
+
Brønsted acid sites. The bands at 1545,1455 and 1490 cm due to
−1
Bronsted, Lewis and contribution of Brønsted and Lewis acid site,
respectively were appeared in spectra of samples M75, M50 and
−
1
M25. Moreover, an intensity of band at 1445 cm was found to
decrease with the decrease in Sn concentration. Similar trend was
−
1
observed with the Al content in case of a band at 1545 cm . Based
−
1
on the intensities of the bands appearing at 1545 and 1455 cm
,
the Al-Sn-MFI samples viz. M75, M50 and M25 have shown the
Brønsted/Lewis acidity ratio as 0.67, 0.91 and 1.27, respectively.
Thus, a range of materials were successfully prepared for investigat-
ing the existence of synergy between Lewis and Brønsted acid sites

1
86
P.S. Niphadkar et al. / Applied Catalysis A: General 401 (2011) 182–188
N-methylation of aniline reaction was performed under optimized
process parameters established in our earlier N-methylation of
aniline study [42]. Fig. 5a shows time on stream (TOS) data for
M0
◦
methylation of aniline at temperature = 240 C, weight hourly space
−
1
velocity (WHSV) = 3 h and 1:6 molar ratio of aniline to methanol.
TOS study indicated that, the aniline conversion increases with
increase in TOS up to 2 h and then decreases with further increase
of TOS. Steady state was apparently reached from 4th hour. More-
over, aniline conversion with M100 and M0 is found to be lower
than other samples (M75, M50 and M25) having combination of
Sn and Al. Thus, lower aniline conversion over sample M100 or M0
than sample M75, M50 and M25 suggest that presence of Sn and
Al combination in a sample has distinct influence on aniline con-
version. Data obtained at 4th hour on all catalysts is selected for
comparison and presented in Fig. 5b.
M25
M50
M75
It is evident from Fig. 5b that, the aniline conversion increases
from M100 to M50, reaches to maximum at M50 and then further
decreases from M50 to M0. In samples M100, M75 and M50, Al per
unit cell increases from 0 to 1.17 (Table 1). As a result, the total
acidity increases from 0.53 to 0.59 mmol of NH /g (Table 1). There-
3
M100
fore, the aniline conversion increases from M100 to M50, reaches to
maximum at M50. It is interesting to note that, with further increase
in Al per unit cell (samples M25 and M0), decrease in aniline con-
version was observed. Thus, in spite of having higher total acidity,
M25 and M0 have shown lower aniline conversion. Probably, an
increase in the stronger acid sites might be responsible for drop
in the aniline conversion. Similar observations were reported by
Narayanan et al. [46] on alkylation of aniline over H-ZSM-5 with
SiO /Al O molar ratio from 30 to 280.
1
400
1450
1500
1550
1600
1650
1700
-
1
Wavenumber cm
Fig. 4. Pyridine FTIR spectra of M100, M75, M50, M25 and M0.
2
2
3
Fig. 5b also exhibits the increases of NNDMA formation at the
cost of NMA with the increase of Al per unit cell in sample M100 to
M0 and the formation of other products (toluidine, N-methyl tolui-
dine and N-dimethyl touidine, in the range of 2–6%). Woo et al. [25]
suggested that, the weak acidity helps to N-alkylation; medium
acidity helps to NN-dialkylation and strong acidity to C-alkylation.
The high selectivity to NMA with M100 is because of the pres-
ence of a number of weak acid sites and increasing selectivity to
NNDMA and others in M75–M0 samples is due to increasing trend
of relatively strong acid site (medium acid site) with increasing Al.
Maximum NMA yield over M75 (49.9%) and M50 (49.7%) was
found, while other samples showed lower NMA yield. An optimum
number of weak (0.35–0.38 mmol of NH3 desorbed/g) and relative
strong (0.19–0.21 mmol of NH3 desorbed/g) acid sites in M75 and
M50 were found to maximize the NMA yield. In these samples, the
weak acid sites are originated from framework Sn and Al, whereas;
in Al-Sn-MFI in acid catalyzed aniline N-methylation reaction to
maximize the aniline conversion and N-methyl aniline selectivity.
3.2. Aniline alkylation
Aniline alkylation is an acid catalyzed reaction. Beside reac-
tion parameters, the characteristics of catalyst, in particular, the
type, concentration and strength of acid sites in the catalyst are
the main factors, which influence the activity and selectivity on
N-alkylation of aniline. Using catalysts with different acidic proper-
ties, the synergetic effect of Brønsted/Lewis acid sites and strength
on the catalytic performance was investigated systematically.
3.2.1. Catalytic performance
In order to improve the activity and to investigate an influ-
ence of Al insertion in stannosilicate MFI structure, vapor phase
90
80
70
60
50
40
30
20
10
0
a
b
Aniline conv.
NMA sel.
NNDMA sel.
NMA yield
8
0
70
60
50
40
M100
M75
M50
M25
M0
30
20
10
0
1
2
3
4
5
M100
M75
M50
M25
M0
Time h
Catalyst used
Fig. 5. Aniline methylation over samples M100, M75, M50, M25 and M0 (a) Aniline conversion with time on stream, and (b) comparative data at stable 4th hour.

P.S. Niphadkar et al. / Applied Catalysis A: General 401 (2011) 182–188
187
90
80
70
60
50
40
30
20
10
90
80
70
60
50
40
30
20
10
Anilineconversion
NMASelectivity
NNDMASelectivity
OtherSelectivity
AnilineConversion
NMASelectivity
NNDMASelectivity
Other
0
1
.0
1.5
2.0
2.5
3.0
3.5
4.0
0
-1
2
20
230
240
250
260
WSHV h
o
Temperature C
Fig. 8. Effect of WHSV on the aniline conversion and selectivity over M50 (temper-
ature: 220 C and mole ratio: 1.8).
◦
Fig. 6. Effect of temperature on the aniline conversion and selectivity over M50
mole ratio aniline to methanol: 1:6, and WHSV: 3h ).
−1
(
products such as NN dimethylaniline and others due to the fur-
ther alkylation of NMA to NNDMA and others [17]. Thus, in the
present studies, the optimum reaction temperature with respect
to highest aniline conversion (65%) and selectivity of NMA (80%)
the relative stronger acid sites are induced by framework Al alone
Fig. 4). Generally, Al in zeolite framework induces Brønsted, while
(
Lewis acid site imparted by tetrahedral substitution of Sn in the
MFI framework. In conclusion, combination of Brønsted and Lewis
acid sites with 0.67–0.91 B/L ratio in the samples M75 and M50 are
found to be optimum for maximum NMA yield. Similar influence
of Brønsted and Lewis acid sites combination (synergy between
Brønsted and Lewis acid site) in Al-Zr-TDU-1 catalyst was reported
by Telalovi c´ et al. [30] for Prins cyclisation reaction.
◦
was identified as 220 C.
3.2.2.2. Effect of aniline to methanol molar ratio. At optimum
reaction temperature, different reactions were carried out with dif-
ferent molar ratios of aniline to methanol viz. 1:4, 1:6, 1:8 and 1:10
−
1
at atmospheric pressure and with WHSV of 3 h . Fig. 7 illustrates
the influence of aniline to methanol molar ratio on aniline con-
version and product distribution. It was observed that, an increase
in aniline to methanol molar ratio, the aniline conversion was
increased from 48% to 69%. However, the NMA selectivity was
decreased from 87% to 77% with increase in NNDMA selectivity
from 12% to 22%. An increase in methanol concentration enhanced
the rate of forward reaction leading to increase in the aniline con-
version. At high conversion level, the NMA formed further reacts
with available methanol to yield more NNDMA. The molar ratio
of 1:8 was found to be the optimum molar ratio, which gave maxi-
mum aniline conversion (67%) with 81% NMA selectivity. The study
was further extended to optimize the contact time to maximize the
NMA yield.
To optimize the reaction, the further studies on optimization of
process parameters were carried out on optimum catalyst M50.
3
3
.2.2. Process parameters optimization
.2.2.1. Effect of temperature. The effect of temperature on aniline
conversion and product selectivity over M50 sample was carried
out, keeping other process parameters identical. The results are
◦
tabulated in Fig. 6. As temperature increases from 220 to 260 C,
although, the aniline conversion was found to increase; there was
a decrease in the selectivity towards N-methyl aniline. The drop
in NMA selectivity is associated with the formation of undesired
90
80
70
60
50
40
30
20
10
0
3
.2.2.3. Effect of weight hourly space velocity. To optimize yet
another reaction parameter, in particular WHSV, reactions with
variable WHSV (from 1 to 4 h ) were carried out at optimum reac-
tion temperature (220 C) and with aniline to methanol molar ratio
−
1
◦
of 1:8. The product distribution at 4th hour for all these WHSVs are
presented as Fig. 8. The aniline conversion was found to decrease
with the increase in WHSV. The maximum aniline conversion (72%)
was obtained at WHSV = 1 h . In case of WHSV = 4 h , the aniline
conversion was the least (44%), but the NMA selectivity was max-
imum (84%). There was an inverse proportion between the aniline
conversion and NMA selectivity. The higher contact time (lower
WHSV) increases NMA alkylation rate over aniline with the increase
AnilineConversion
NMASelectivity
−
1
−1
NNDMASelectivity
OtherSelectivity
−
1
in aniline conversion. Considering the NMA yield, WHSV = 3 h
was found to be optimum which gave the maximum aniline con-
version (67%) and NMA selectivity (81%) in the present studies.
1:4
1:6
1:8
1:10
3
.2.3. Catalyst stability
In order to test the stability of the optimum catalyst and to exam-
Aniline : Methanol molar ratio
ine the dependence of stability on the synergy between Lewis and
Brønsted acid sites, M50 and M25 catalysts were subjected to the
Fig. 7. Effect of mol ratio on the aniline conversion and selectivity over M50 (tem-
perature: 220 C, and WHSV: 3h ).
◦
−1

1
88
P.S. Niphadkar et al. / Applied Catalysis A: General 401 (2011) 182–188
100
M50 indicating the dependence of stability on a synergy between
Lewis and Brønsted acid sites in Al-Sn-MFI.
9
0
0
0
0
0
0
0
0
0
8
7
6
5
4
3
2
1
Acknowledgments
P.S.N. is grateful to Director, National Chemical Laboratory,
Pune, India for his support and permission to work for Ph.D.
References
[
1] S. Narayanan, V. Durgakumari, A.S. Rao, Appl. Catal. A 111 (1994) 133–142.
Aniline conversion over M50
Aniline conversion over M25
NMA selectivity over M50
NMA selectivity over M25
[2] J. Santhanalakshmi, T. Raja, Appl. Catal. A 147 (1996) 69–80.
[
3] S. Sunwanprasop, T. Nhujak, S. Roengsumran, A. Petsom, Ind. Eng. Chem. Res.
3 (2004) 4973–4978.
4] S.R. Stauffer, J.F. Hartwig, J. Am. Chem. Soc. 125 (2003) 6977–6985.
4
[
[5] J.S. Kim, O.J. Shon, J.A. Rim, S.K. Kim, J. Yoon, J. Org. Chem. 67 (2002) 2348–2351.
[
[
6] M. Nehate, V.V. Bokade, Appl. Clay Sci. 44 (2009) 255–258.
7] F.M. Baustista, J.M. Campelo, G.D. Luna, J.M. Marinas, A.A. Romero, M.R. Urbano,
J. Catal. 172 (1997) 103–109.
0
0
2
4
6
8
10
[8] F.M. Baustista, J.M. Campelo, G.D. Luna, J.M. Marinas, A.A. Romero, Appl. Catal.
A 166 (1998) 39–45.
Time h
[
9] L.J. Garces, V.D. Makwana, B. Hincapie, A. Sacco, S.L. Suib, J. Catal. 217 (2003)
07–116.
1
Fig. 9. Stability test for M50 and M25 catalysts under optimum reaction conditions
[
10] I.I. Ivanova, E.B. Pomakhina, A.I. Rebrov, M. Hunger, Y.G. Kolyagin, J. Weitkamp,
J. Catal. 203 (2001) 375–381.
◦ −1
temperature: 220 C, mole ratio: 1.8, and WHSV: 3h ).
(
[
[
[
11] V.D. Stytsenko, T.D. Huu, V.A. Vinokurov, Kinet. Catal. 46 (2005) 376–379.
12] German patent 3942413, 1991.
13] V.D. Stysenko, V.A. Vinokurov, Scientific Conf. of Russian State University of Oil
and Gas on the Catalytic Synthesis of Blending Fuels and Antioxidants, Mascow,
1999, p. 151.
stability test under optimum reaction conditions. The alkylation of
◦
−1
aniline with methanol was carried out at 220 C with WHSV = 3 h
using feed containing aniline and methanol in the molar ratio of
:8. The results with regard to product distribution as a function
[14] A. Ko, C. Yang, W. Zhu, H. Lin, Appl. Catal. A 134 (1996) 53–66.
[15] M. Vijayaraj, C.S. Gopinath, J. Catal. 241 (2006) 83–95.
[16] K. Sreekumara, T. Rajab, B.P. Kiranb, S. Sugunana, B.S. Rao, Appl. Catal. A 182
(1999) 327–336.
1
of reaction time up to 10 h are shown in Fig. 9. M50 catalyst was
found to be stable up to 10 h and has shown the stable aniline
conversion and NMA selectivity of the order 36–67% and 69–86%,
respectively. As compared to M50, M25 has shown less stabil-
ity, which has reflected in the drop in conversion from 8th hour.
This phenomenon may be attributed to the higher Brønsted/Lewis
acidity ratio of M25 as compared to M50. Thus, it seems that,
there exist a synergy between Lewis and Brønsted acid sites in
Al-Sn-MFI in aniline N-methylation reaction and catalyst stability
as well.
[17] Y.K. Park, K. Park, S. Woo, Catal. Lett. 26 (1994) 169–180.
[18] C. Yang, N. He, Q. Xu, J. Incl. Phenom. Macro. 35 (1999) 123–130.
[19] B.L. Su, D. Barthomeuf, Appl. Catal. A 124 (1995) 73–80.
[20] S.P. Elangovan, C. Kannan, B. Arabindoo, V. Murugesn, Appl. Catal. A 174 (1998)
213–219.
[21] G.D. Yadav, N.S. Doshi, J. Mol. Catal. A 194 (2003) 195–209.
[22] S. Narayanan, K. Deshpande, J. Mol. Catal. A 104 (1995) L109–113.
[23] S. Narayanan, B.P. Prasad, J. Mol. Catal. A 96 (1995) 57–64.
[24] R. Luque, J.M. Campelo, D. Luna, J.M. Marinas, A.A. Romero, J. Mol. Catal. A 269
2007) 190–196.
25] S.I. Woo, J.K. Lee, B.S. Hong, Y.K. Park, Y.S. Uh, Stud. Surf. Sci. Catal. 49B (1989)
095–1103.
26] P.Y. Chen, M.C. Chen, H.Y. Chu, N.S. Chang, T.K. Chuang, Stud. Surf. Sci. Catal. 28
1986) 739–786.
(
[
[
1
(
[
[
[
27] J.M. Campelo, F. Lafont, J.M. Marinas, Zeolites 15 (1995) 97–103.
28] S. Narayanan, K. Deshpande, Appl. Catal. A 199 (2000) 1–3.
29] S. Li, A. Zheng, Y. Su, H. Zhang, L. Chen, J. Yang, C. Ye, F. Deng, J. Am. Chem. Soc.
129 (2007) 11161–11171.
4
. Conclusion
Different degree of isomorphous substitution of Al3+ in the
[
30] S. Telalovi c´ , J.F. Ng, R. Maheswari, A. Ramanathan, G.K. Chuah, U. Hanefeld,
stannosilicate framework with MFI topology was achieved by
hydrothermal crystallization. These catalysts were synthesized by
hydrothermal crystallization of gels having molar compositions
SiO :xSnO :yAl O :0.23 (TPA) O:35H O, where x ranges from 0 to
Chem. Commun. (2008) 4631–4633.
[
31] R. Savidha, A. Pandurangan, M. Palaichamy, V. Murugesan, Catal. Lett. 91 (2003)
49–61.
[32] V. Parvulescu, C. Anastasescu, B.L. Su, J. Mol. Catal. A: Chem. 211 (2004)
43–148.
33] V. Parvulescu, C. Anastasescu, C. Constantin, B.L. Su, Catal. Today 78 (2003)
77–485;
V. Parvulescu, Cr. Tablet, C. Anastasescu, B.L. Su, Catal. Today 93–95 (2004)
07–313.
2
2
2
3
2
2
1
2
1
00 and y from 0 to 400. Although, Sn:Al ratio was varied (ca. 1:0,
:0.33, 1:1, 1:3, 0:1), the overall molar Si/(Al + Sn) ratio was kept
[
4
constant as 50. It was evident from the TPAD results that, the frame-
work tin species create the weak acid sites whereas insertion of Al
lead to the creation of both the weaker as well as relatively stronger
acid sites. With an increase in Al3 insertion in Sn-MFI molecular
sieves, there is an increase in the Brønsted/Lewis acidity ratio as
well as the total acidity. Based on Brønsted/Lewis acidity ratio, the
Al-Sn-MFI samples followed the trend as M75 < M50 < M25. These
samples showed higher aniline conversion than the samples con-
taining either Al or Sn alone. Interestingly, an increase in stronger
acid sites with an increase in framework Al3 leads to decrease in
aniline conversion. Thus, synergy between Brønsted and Lewis acid
sites in Al-Sn-MFI catalysts was demonstrated to exist in aniline
N-methylation reaction. Upon process parameter optimization, Al-
Sn-MFI with Si/Al = 50.2, Si/Sn = 93.8, Si/(Al + Sn) = 43.2 showed the
maximum aniline conversion (67%) and NMA selectivity (81%) at
3
[
[
34] V. Parvulescu, C. Constantin, B.L. Su, J. Mol. Catal. A: Chem. 202 (2003) 171–178.
35] R. Savidha, A. Pandurangan, Appl. Catal. A: Gen. 276 (2004) 39–50.
+
[36] S. Vetrivel, A. Pandurangan, J. Mol. Catal. A: Chem. 217 (2004) 165–174.
[
37] M. Selvaraj, B.R. Min, Y.G. Shul, T.G. Lee, Microporous Mesoporous Mater. 74
2004) 143–155.
38] V. Parvulescu, B.L. Su, Stud. Surf. Sci. Catal. 143 (2000) 575–584.
(
[
[39] V. Parvulescu, C. Anastasescu, B.L. Su, J. Mol. Catal. A: Chem. 198 (2003)
49–261.
2
[
[
40] S. Narayanan, K. Deshpande, J. Mol. Catal. A: Chem. 104 (1995) 109–113.
41] B. Rakshe, V. Ramaswamy, A.V. Ramaswamy, J. Catal. 188 (1999) 252–260.
+
[42] P.S. Niphadkar, P.N. Joshi, H.R. Gurav, S.S. Deshpande, V.V. Bokade, Catal. Lett.
33 (2009) 175–184.
1
[
43] K. Chaudhari, T.K. Das, P.R. Rajmohanan, K. Lazar, S. Sivasanker, A.J. Chandwad-
kar, J. Catal. 183 (1999) 281–291.
[44] P.S. Niphadkar, A.C. Garade, R.K. Jha, C.V. Rode, P.N. Joshi, Microporous Meso-
porous Mater. 136 (2010) 115–125.
[
45] P.S. Niphadkar, K.R. Patil, P.N. Joshi, Microporous Mesoporous Mater. (2010),
doi:10.1016/j.micromeso.2010.10.36.
◦
−1
reaction temperature = 220 C, WHSV = 3 h , molar ratio (aniline to
methanol) = 1:8 and TOS = 4 h. On account of higher Brønsted/Lewis
acidity ratio, M25 (1.27) was found to be less stable as compared to
[
46] S. Narayanan, V. Durga Kumari, A. Sudhakar Rao, Appl. Catal. 111 (1994)
133–142.