Comparison of the catalytic performance of the metal substituted cage type mesoporous silica catalysts in the alkylation of naphthalene

Article abstract of DOI:10.1016/j.apcata.2010.01.022

Alkylation of naphthalene with propylene to diisopropylnaphthalenes (DIPN) for the use as a high-quality solvent was carried out over mesoporous AlSBA-1, GaSBA-1 and FeSBA-1 catalysts. The AlSBA-1 and GaSBA-1 catalysts were very active in alkylation while the FeSBA-1 samples, although initially active, deactivated quickly. Activity of the SBA-1 catalysts increased with the amount of Al, Ga or Fe incorporation into the silica framework. Regardless of the alkylation activity, all SBA-1 catalysts showed rather low isomerization activity and as a result low 2,6-DIPN selectivity was observed. The AlSBA-1 and GaSBA-1 catalysts proved to be promising for DIPN solvent synthesis due to their high alkylation activity and stability together with low 2,6-DIPN selectivity.

Full text of DOI:10.1016/j.apcata.2010.01.022

Applied Catalysis A: General 377 (2010) 76–82
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Comparison of the catalytic performance of the metal substituted cage type
mesoporous silica catalysts in the alkylation of naphthalene
b
c
Robert Brzozowski a, , Ajayan Vinu , Barbara Gil
*
a Industrial Chemistry Research Institute, Rydygiera 8, 01-793 Warsaw, Poland
b International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan
c Faculty of Chemistry, Jagiellonian University, Ingardena 3, PL-30-060 Krakow, Poland
A R T I C L E I N F O
A B S T R A C T
Article history:
Alkylation of naphthalene with propylene to diisopropylnaphthalenes (DIPN) for the use as a high-
quality solvent was carried out over mesoporous AlSBA-1, GaSBA-1 and FeSBA-1 catalysts. The AlSBA-1
and GaSBA-1 catalysts were very active in alkylation while the FeSBA-1 samples, although initially
active, deactivated quickly. Activity of the SBA-1 catalysts increased with the amount of Al, Ga or Fe
incorporation into the silica framework. Regardless of the alkylation activity, all SBA-1 catalysts showed
rather low isomerization activity and as a result low 2,6-DIPN selectivity was observed. The AlSBA-1 and
GaSBA-1 catalysts proved to be promising for DIPN solvent synthesis due to their high alkylation activity
and stability together with low 2,6-DIPN selectivity.
Received 14 July 2009
Received in revised form 3 January 2010
Accepted 12 January 2010
Available online 21 January 2010
Keywords:
Alkylation
Naphthalene
ß 2010 Elsevier B.V. All rights reserved.
Diisopropylnaphthalene
SBA-1 mesoporous material
1. Introduction
by the wide-pore zeolites and shape selectivity in the naphthalene
alkylation is expected leading to a high b,b-selectivity. Among the
Diisopropylnaphthalene (DIPN) isomeric mixture is used as a
high-quality solvent for copying materials and in many other
applications [1–3]. In Europe the DIPN solvent is produced by
Ruetgers Kureha Solvents GmbH (Duisburg, Germany) and is
known under the commercial name ‘‘Ruetasolv DI’’. It consists of
DIPN isomeric mixture (more than 98%), however, to guarantee
liquid state of the solvent at low temperature, content of DIPN
isomers with high solidification point, such as 2,6-DIPN (m.p.
68.5 8C [1]) should be limited.
On the other hand, the 2,6-DIPN isolated from the isomeric
mixture can be used as a raw material for the production of
advanced polymers. DIPN product is generally prepared by the
alkylation of naphthalene or isopropylnaphthalene (IPN) with
propylene or other alkylating agents using acidic catalysts.
However, depending on DIPN product destination, alkylation
process may require the catalysts with completely opposite
properties.
wide-pore zeolites, high silica mordenite was found to be the best
candidate for the synthesis of 2,6-DIPN selectively [4–13]. Owing
to its optimal pore size and favourable one-dimensional (1D) pore
system, mordenite catalysts promoted a 2,6-DIPN selectivity as
high as 70% and the 2,6-DIPN/2,7-DIPN ratio in the range of 2–3 [7–
9]. Additional enhancement in shape selectivity of mordenites was
achieved by deactivation of the external surface of zeolite crystals
[10,11]. A high 2,6-DIPN selectivity and a high ratio of 2,6-DIPN/
2,7-DIPN were also reported for disproportionation of IPN over
mordenite [14].
It is also possible to obtain DIPN products rich in isomers other
than the 2,6-DIPN by selecting the proper catalyst and reaction
conditions. For instance, Sugi at al. [13] have recently reported a
significantly higher 2,7-DIPN selectivity than that of 2,6-DIPN over
CIT-5 catalyst in the alkylation of naphthalene. A high 2,7-DIPN
selectivity (up to 67%) was also observed in disproportionation of
IPN over HY and H-beta zeolites [14]. It has also been reported that
a high yield of DIPN isomers with both alkyl substituents attached
to one naphthalene ring, i.e. 1,3-DIPN and 1,4-DIPN can be
obtained over zeolite catalysts under the optimized reaction
conditions favouring the pore mouth selectivity [15]. For example,
alkylation of naphthalene with propylene over the zeolite Y
modified with calcium and rare earth metals at 200 8C gave DIPN
product containing 44% 1,4-DIPN and 12% 1,3-DIPN.
Pore size of wide-pore zeolites is in the range between the
critical diameter of slim
b
,
b
-DIPN (2,6- and 2,7-DIPN) isomers and
-isomers
that of bulky -DIPN (1,3-, 1,6- and 1,7-DIPN) and a,a
a
,
b
(1,4- and 1,5-DIPN). Therefore, DIPN isomers can be differentiated
* Corresponding author. Tel.: +48 22 568 20 20; fax: +48 22 568 21 74.
E-mail addresses: robert.brzozowski@ichp.pl (R. Brzozowski),
vinu.ajayan@nims.go.jp (A. Vinu).
The synthesis of DIPN mixture for use as a high-quality solvent
requires active and stable catalysts but producing a DIPN mixture
0926-860X/$ – see front matter ß 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.01.022

R. Brzozowski et al. / Applied Catalysis A: General 377 (2010) 76–82
77
poor in 2,6-isomer, the compound disadvantageous due to its high
melting point. Mesoporous materials are the adequate catalysts for
this purpose, owing to their ordered structure, wide pores and mild
acidity [16]. Even bulky substrates and products can easily diffuse
inside the pores. Also bulky precursors of carbonaceous deposits
can be easily removed from the pores by a liquid product stream.
A high and durable activity of Al–MCM-41 catalysts in alkylation
of 2-IPN and naphthalene with isopropanol has been reported [17–
19]. MCM-41 possesses linear arrays of 1D mesopores which do not
have any connectivity between the pores whereas mesoporous
MCM-48 exhibits three-dimensional (3D) pore system with much
higher surface area that is easily accessible to the bulky reactant
molecules. This feature makes the MCM-48 material superior
catalystoverthe MCM-41. Ithas beenreportedearlier thatAl–MCM-
48 showed much higher catalytic activity than the Al–MCM-41 in
the alkylation of naphthalene using isopropanol as the alkylating
agent [20,21]. We also confirmed a high catalytic activity of Al–
MCM-48 catalysts in the alkylation of naphthalene with propylene
which afforded a low 2,6-DIPN selectivity [22].
The mesoporous SBA-1 material has an ordered 3D pore
structure with large cages A (ca. 4 nm) and slightly smaller cages B
(ca. 3.3 nm) accessible through wide windows (>2 nm) [23–25],
thus it is expected to be a promising catalyst for alkylation of
naphthalene to DIPN.
The successful incorporation of Al species into SBA-1 frame-
work was confirmed by catalytic activity of AlSBA-1 catalysts in the
reaction of n-decane conversion [24]. On the other hand, the
catalytic activity of FeSBA-1 and GaSBA-1 samples was observed in
the reaction of alkylation of phenol with tert-butanol [25,26]. We
also confirmed catalytic activity of GaSBA-1 samples in alkylation
of naphthalene with propylene [27].
The properties of SBA-1 were investigated by adsorption of
pyridine and NO used as probe molecules followed by FTIR
spectroscopy. All samples in the form of self-supporting wafers
were activated under vacuum (10À4 mbar) at 400 8C prior to the
adsorption of probe molecules. The adsorption temperatures were
170 8C for pyridine (POCh Gliwice, analytical grade, dried over 3A
molecular sieve), and ambient temperature for NO (Linde Gas
Polska, 99.95% purified by freeze-and-thaw procedure prior to
each adsorption). For quantitative measurements pyridine was
first adsorbed at the excess at 170 8C then desorbed at the same
temperature for 20 min. These conditions were sufficient for the
removal of the physisorbed species only, leaving intact pyridine
molecules bonded to Brønsted and Lewis acid sites.
IR spectra were recorded with a Bruker Tensor 27 spectrometer
equipped with an MCT detector and working with the spectral
resolution of 2 cmÀ1. Spectra presented in this work are normal-
ized to the standard 10 mg pellet (density 3.2 mg cmÀ2).
Alkylation experiments were carried out in a down flow reactor
in the temperature range 120–300 8C, under a high pressure
(3 MPa), a naphthalene/propylene mole ratio of 1.5 and a feed flow
20 g gÀ1 hÀ1. Inlet temperature was changed starting from the
lowest set point and reaction conditions at each set point were
stabilised for 1 h at each temperature. So the higher the test
temperature the longer was time on stream (TOS). Nevertheless,
preliminary deactivation tests at 150 8C revealed that the most
active catalyst samples did not show propylene conversion
decrease for TOS longer than 8 h. Because Fe-containing samples
were less active and deactivated quicker than Al- and GaSBA-1
catalysts, we started experiments from 150 8C.
More details on alkylation procedure are described in [22].
Alkylation products were analyzed by a GC equipped in FID
detector, using a 60-m-long HP-INNOWax capillary column,
according to the method described elsewhere [28]. Product
samples were also qualitatively confirmed by GC–MS.
In this paper we report on the results of the catalytic activity of
the SBA-1 catalysts containing various amounts of Al, Ga or Fe in
the isopropylation of naphthalene with propylene as an alkylating
agent in order to effectively produce DIPN mixture with low 2,6-
DIPN isomer content.
3. Results and discussion
2. Experimental
Mesoporous SBA-1 with different metal atoms viz. Ga, Al, and Fe
in the framework was prepared using cetyltriethylammonium
bromide as a structure directing agent. The textural parameters
and elemental composition of the all the materials prepared with
different nSi/nMe ratios are listed in Table 1. As can be seen in
Table 1, the amount of metal atom in the framework increases with
increasing amount of the metal species in the initial gel. However,
the nSi/nMe ratio of the all the metal substituted SBA-1 materials is
much higher than the ratio in the respective synthesis gels. This
could be mainly due to a high solubility of the metal source in the
acidic medium. Fig. 1 shows the powder XRD patterns of the
representative metal substituted SBA-1 samples. All the samples
displayed several peaks at lower angles, which can be indexed to
Al, Ga or Fe atoms were incorporated into the silica framework of
SBA-1 catalysts by direct synthesis route in a highly acidic medium
according to the methods described by Vinu et al. [24–26]. Values in
parentheses in the name of catalysts denote the nSi/nMe ratio, where
Me refers to Al, Ga or Fe, respectively. The properties of the catalysts
were confirmed by XRD, N2 sorption, MAS-NMR, IR and EPR (Fe3+
)
spectroscopy. Some properties of mesoporous SBA-1 catalysts are
shown in Table 1. The amorphous aluminosilicate ASA(6) (nSi/nAl
ratio = 6) and the mesoporous Al–MCM-48(26) catalyst (nSi/nAl
ratio = 26) were also used for the comparison studies. More details
on Al–MCM-48(26) properties can also be found in [21].
Table 1
Properties of SBA-1 samples substituted with various metals.
Sample
nSi/nMe (in gel)
nSi/nMe (AAS)
ABET [m2 gÀ1
]
dp [nm]
Vp [cm3 gÀ1
]
Brønsted centres per g of catalyst
Lewis centres per g of catalyst
AlSBA-1(39)
AlSBA-1(64)
AlSBA-1(180)
2
3.3
10
39
64
1324
1350
1341
1.9
2.0
2.1
0.64
0.67
0.69
52
20
7
79
59
17
180
GaSBA-1(17)
GaSBA-1(40)
GaSBA-1(90)
10
20
40
16.9
39.6
90.3
1260
1365
1410
2.49
2.44
2.42
0.67
0.70
0.75
65
59
26
795
461
249
FeSBA-1(90)
40
67
90
1350
1390
2.4
2.4
0.71
0.70
20
52
74
FeSBA-1(120)
120
113
AlMCM-48(26)
ASA(6)
25
–
26
6
1245
323
2.1
–
0.56
0.32
65
–
147
–
Where nSi/nMe = molar ratio of Si to Al, Ga or Fe, respectively; ABET = BET surface area; dp = pore diameter; Vp = pore volume.

78
R. Brzozowski et al. / Applied Catalysis A: General 377 (2010) 76–82
Fig. 1. Powder XRD patterns of metal substituted SBA-1: (a) AlSBA-1(39), (b) GaSBA-
1(40), and (c) FeSBA-1(36).
the (1 1 1), (2 0 0), and (2 2 0) reflections of the face-centred-cubic
Fm3m symmetry. These results indicate that the materials possess
a 3D well-ordered cage type porous structure and the structural
order of the materials is retained even after the metal incorpo-
ration. The detailed characterization of the materials with different
metal atoms and their contents can be found in our earlier reports
[24–26].
Acidity measurement can shed some light on the actual location
and status of metallic sites because of two reasons. First, if metal is
located at the surface of the SBA-1 walls, i.e. in the tetrahedral
environment, it should create Si–O–Me bridge, this way becoming
a source of Brønsted acidity. Metal-containing extraframework
species can be detected as Lewis sites. The number of acidic sites of
both types is usually lower than the metal content measured by
chemical analysis because probe molecules cannot react with
metallic species buried within the walls and with the part of the
oligomeric species [29]. At the same time, such hidden metallic
species would be also inaccessible to reactant molecules [30]. For
quantification of acid centres pyridine is the best probe for both
Brønsted and Lewis acid sites, as absorption coefficients of arising
1545 and 1450 cmÀ1 bands are independent of the type of zeolite,
its composition or morphology [31].
Fig. 2. IR spectra of NO adsorbed on FeSBA-1(90) (a) and FeSBA-1(120) (b). Spectra
normalized to the standard 10 mg sample. The background spectra of activated
samples were subtracted.
FeSBA-1 series. Alkylation is the reaction proceeding mainly on the
Brønsted acid centres via carbocationic transition state, therefore
the concentration of Brønsted sites is thought to be the primary
factor influencing the conversion in the reaction of naphthalene
isopropylation. As it was already mentioned, the concentration of
acid sites for the FeSBA-1 samples was not consistent with the nSi/
nFe ratio determined by the chemical analysis. Iron in zeolites and
mesoporous materials is usually easily reduced from Fe3+ to Fe2+
species, and also oligomerizes readily. Therefore, the presence of
different forms of iron may be the reason of higher acidity of the
FeSBA-1 sample with lower iron loading.
To verify this hypothesis, NO was adsorbed in both samples,
because the frequency of the NO maxima depend not only on the
oxidation states of specific iron species but also on their location in
geometrically different environment. In this experiment, NO was
adsorbed in the excess and then the gaseous phase together with
weakly bonded species were removed by a short evacuation. As it
can be seen in Fig. 2, the IR bands arising from the NO adsorption
The results of pyridine adsorption experiments are summarized
in Table 1. Generally, the higher was the nSi/nMe ratio, determined
by chemical analysis, the higher was the concentration of both
Brønsted and Lewis centres in investigated samples except for the
are different for FeSBA-1 samples. The band at 1820 cmÀ1
characteristic of the Fe2+ trinitrosyls (Fig. 2, spectrum a), is
,
Table 2
Results of alkylation of naphthalene with propylene over AlSBA-1 catalysts.
Temperature [8C]
Conversion [%]
Selectivity to
IPN [%]
2-IPN in IPN [%]
DIPN isomers distribution
DIPN [%]
2,6-DIPN
2,7-DIPN
a,
b-Isomers
a,a-Isomers
AlSBA-1(39)
120
83.9
94.9
96.8
96.5
95.3
68.4
70.6
63.8
60.0
67.9
19.5
18.0
18.7
18.6
21.7
27.3
30.5
40.6
75.9
90.2
4.5
5.6
3.0
4.4
41.5
45.7
57.5
45.7
26.1
50.4
44.3
27.5
5.0
150
200
7.9
6.0
250
25.8
37.0
21.9
36.3
300
0.6
AlSBA-1(64)
120
77.4
78.7
96.2
98.9
97.1
82.6
78.5
77.2
66.2
83.0
12.3
15.8
15.1
19.3
12.8
26.1
28.5
40.3
72.9
90.1
4.3
4.5
2.3
2.2
44.6
45.1
57.4
50.3
26.3
48.9
48.3
29.6
6.1
150
200
7.5
5.5
250
23.3
36.7
18.4
35.6
300
0.0
AlSBA-1(180)
120
150
200
250
300
52.6
55.9
97.4
97.3
97.5
81.0
84.1
73.8
67.3
76.3
13.0
11.9
20.7
17.7
16.8
26.5
26.4
33.9
54.7
88.6
5.1
3.3
2.2
1.8
43.2
43.0
49.8
62.8
29.6
49.6
51.9
40.6
14.6
1.4
5.4
3.4
12.8
36.7
9.9
32.3

R. Brzozowski et al. / Applied Catalysis A: General 377 (2010) 76–82
79
Table 3
Results of alkylation of naphthalene with propylene over GaSBA-1 catalysts.
Temperature [8C]
Conversion [%]
Selectivity to
IPN [%]
2-IPN in IPN [%]
DIPN isomers distribution
DIPN [%]
2,6-DIPN
2,7-DIPN
a,b
-Isomers
a,a-Isomers
GaSBA-1(17)
120
97.0
98.3
98.0
98.1
97.9
74.2
69.9
62.7
62.1
69.8
18.2
22.8
24.5
28.7
24.1
26.0
31.3
51.7
85.1
89.7
4.2
4.7
2.9
3.4
42.4
51.0
55.7
31.2
26.1
49.6
40.0
10.8
2.0
150
200
16.7
33.7
36.1
15.2
32.2
36.0
250
300
0.9
GaSBA-1(40)
120
95.6
98.7
97.5
97.6
96.1
75.9
80.2
78.8
72.2
74.1
17.7
13.6
13.9
14.1
15.6
26.5
29.6
42.9
82.2
91.3
4.1
4.9
2.4
3.2
41.7
46.6
59.9
42.9
24.3
51.0
43.6
25.2
2.9
150
200
7.8
6.0
250
28.4
38.8
24.4
36.9
300
0.0
GaSBA-1(90)
120
70.4
74.7
98.7
97.2
97.6
78.9
78.2
75.7
74.5
83.6
15.5
14.2
17.4
17.3
12.7
25.7
26.7
34.8
73.9
90.6
3.8
3.3
2.2
2.7
42.3
43.9
52.3
52.2
25.9
50.5
50.1
38.2
6.6
150
200
5.3
3.4
250
21.8
37.2
17.8
35.9
300
0.5
dominating the spectrum of FeSBA-1(90). It was suggested by
Feretti at al. [32] that the band of Fe2+(NO)3 at 1810 cmÀ1 is a
measure of iron dispersion—if three NO molecules can be bonded
onto one centre, therefore this particular centre have to be isolated
from other iron species and exposed at the surface. Such iron can
be at the same time a single Lewis centre. Therefore in FeSBA-
1(90), the significant part of iron, introduced during synthesis is
present in the extraframework, reduced and isolated form of Fe2+
cations.
In the spectrum of the FeSBA-1(120), the most intensive is the
band of NO bonded onto Fe3+ oligomeric species [32]. Lower Fe
loading prevented the reduction of most of the species. Even
though they are present as extraframework centres, they are still
trivalent. Other intensive band, present after NO adsorption on this
sample is the band at 1585 cmÀ1, arising from negatively charged
form of NO, the most probably NO3À, bonded via one of its oxygen
atoms [33]. At the same time, the presence of the band at
2224 cmÀ1 (not shown in Fig. 2) is pointing at the presence of N2O,
suggesting that NO underwent disproportionation when contacted
with this sample. Summarising, it is evident that iron was
differently distributed, being more dispersed and reduced in
FeSBA-1(120) sample, which explains its higher acidity.
required temperatures higher than 150 8C. Lower conversion over
FeSBA-1(90), despite higher concentration of the acid centres, may
be explained by the lower acid strength of iron-based Brønsted
centres [34]. Moreover, FeSBA-1 catalysts visibly deactivated and,
therefore, in the tested temperature range, propylene conversion
lower than 42% was observed (Table 4). Initially, propylene
conversion increased with temperature ramp to 250 8C but further
temperature elevation did not cause any increase in the conver-
sion. For both FeSBA-1 samples higher propylene conversion was
observed at 250 8C than at 300 8C. This maximum in propylene
conversion plot was caused by overlapping two effects: the activity
increase with temperature ramp and the rapid catalyst deactiva-
tion with time on stream. The experiments were performed
starting from the lowest temperature and the reaction tempera-
ture was increased step by step, therefore, the higher the test
temperature the longer was time on stream. The origin of the rapid
deactivation of FeSBA-1 catalysts is not clear at present and would
require additional investigation.
Over the most active AlSBA-1 and GaSBA-1 samples, i.e. those
with the highest metal content and higher acidity, a high
conversion was achieved at 120 8C (Fig. 3). The propylene
conversion increased with temperature and at temperature higher
than 200 8C, practically total propylene was consumed in the
reaction carried out over all GaSBA-1 and AlSBA-1 catalyst samples
at our test conditions.
In the tested range of nSi/nMe ratio, if the higher was the Al, Ga or
Fe content in the SBA-1 catalyst and higher Brønsted acidity, the
higher activity was observed. For GaSBA-1 and AlSBA-1 catalysts it
is discernible at temperature lower than 200 8C, when the increase
Results of the isopropylation of naphthalene over AlSBA-1,
GaSBA-1 and FeSBA-1 mesoporous catalysts with different
amounts of Al, Ga, and Fe incorporated into the silica framework
are collected in Tables 2–4, respectively.
AlSBA-1 and GaSBA-1 catalysts were more active than FeSBA-1
samples. Both GaSBA-1 and AlSBA-1 catalysts were active even at
the temperatures as low as 120 8C, whereas the FeSBA-1 samples
Table 4
Results of alkylation of naphthalene with propylene over FeSBA-1 catalysts.
Temperature [8C]
Conversion [%]
Selectivity to
IPN [%]
2-IPN in IPN [%]
DIPN isomers distribution
DIPN [%]
2,6-DIPN
2,7-DIPN
a,b
-Isomers
a,a-Isomers
FeSBA-1(90)
150
5.7
37.4
42.0
18.5
99.3
89.0
83.8
79.9
0.7
8.7
27.0
28.5
33.3
36.8
–
–
–
–
200
4.1
6.7
–
1.8
2.7
–
44.1
47.4
–
50.0
43.2
–
250
11.0
0.0
300
FeSBA-1(120)
150
200
250
300
4.9
13.8
28.9
19.0
85.2
84.0
84.8
79.1
0.0
4.2
29.8
29.2
32.3
35.8
–
–
–
–
0.0
6.0
6.1
0.0
3.6
7.1
47.5
46.4
44.9
52.5
44.0
41.9
10.7
6.6

80
R. Brzozowski et al. / Applied Catalysis A: General 377 (2010) 76–82
samples of mesoporous SBA-1 catalysts with the results obtained
over the amorphous aluminosilicate ASA(6) and the AlMCM-
48(26).
GaSBA-1 and AlSBA-1 were found to be more active than the
ASA(6). Propylene conversion at 150 8C in the case of GaSBA-1(40),
AlSBA-1(39) and GaSBA-1(17) mesoporous catalysts was in the
range of 95–99%, whereas ASA(6) afforded only 86%.
Regardless of nSi/nMe ratio, all GaSBA-1 and AlSBA-1 samples
were more active than the AlMCM-48(26) catalyst and propylene
conversion greater than 56% was observed at 150 8C in the former
case, whereas only 15% in the latter catalyst. Under similar
conditions other AlMCM-48 samples, i.e. with higher nSi/nMe ratio,
revealed even lower propylene conversion than that of the AlMCM-
48(26) [22].
When comparing acidity of the catalysts (Table 1), the lower
alkylation activity of the AlMCM-48(26) is surprising. The AlMCM-
48(26) sample possessed more Brønsted and Lewis acid sites than
AlSBA-1 catalysts what is in accordance with quantities of the
incorporated Al atoms. Moreover, the concentration of Brønsted
sites in the AlMCM-48(26) was not lower than those of GaSBA-1
samples. Therefore, pore structure of SBA-1 type seems to be
superior in alkylation of naphthalene to that of MCM-48.
It is a well-known fact that naphthalene nucleus is initially
Fig. 3. Temperature relationship of propylene conversion for various catalysts.
Symbols: (~) AlSBA-1(39); (^) AlSBA-1(180); (&) GaSBA-1(40); (^) GaSBA-1(90);
(&) FeSBA-1(90); (Â) ASA(6) (dashed line); (*) AlMCM-48(26) (dashed line).
in propylene conversion with Ga or Al content can be observed (see
Tables 2 and 3). Fig. 4 illustrates the influence of the nSi/nMe ratio in
Al–, Ga– and Fe–SBA-1 catalysts on the propylene conversion
achieved in naphthalene alkylation carried out at 150 8C and
confirms the increase in catalyst activity with a larger amount of
metal incorporation. Similarly, a propylene conversion of 84% was
achieved over the AlSBA-1(39) (52 H+/g) catalyst, whereas 77%
over AlSBA-1(64) (20 H+/g) and only 53% over AlSBA-1(180) (7 H+/
g) (Table 2) at the reaction temperature of 120 8C. As shown in
Table 3, GaSBA-1(17) and GaSBA-1(40), with similar number of
Brønsted acid centres, exhibited similar propylene conversion (96–
97%) at 120 8C, but the conversion of only 70% was observed over
the GaSBA-1(90) catalyst with significantly lower number
Brønsted acid centres. Such a relationship of activity is reasonable,
because quantity of active sites on the SBA-1 catalyst surface is
expected to increase with a larger amount of trivalent metal
species incorporated into the silica framework.
alkylated at
-alkylnaphthalenes isomerize to more thermodynamically stable
-alkylnaphthalenes. Therefore, products of naphthalene mono-
a-positions, that are kinetically preferred, and next the
a
b
isopropylation obtained at mild conditions and a short contact
time are rich in 1-IPN but those achieved under severe reaction
conditions and prolonged time became abundant in 2-IPN.
Similarly, DIPN product obtained at kinetic conditions is rich in
a,a- and a,b-isomers, whereas that with composition close to
thermodynamic equilibrium is rich in 2,6-DIPN and 2,7-DIPN
isomers [35,36]. Since 2,6-DIPN is undesirable isomer in DIPN
mixture applied as a high-quality solvent, a selective alkylation
catalyst with a low isomerization activity is profitable.
A comparison of propylene conversion achieved on FeSBA-1(90)
sample with that of GaSBA-1(90) and even with activity observed
on AlSBA-1(180), i.e. the catalyst with a smaller trivalent metal
content, demonstrates clearly that the incorporation of Fe was the
least effective for the alkylation of naphthalene, the most probably
due to lower acid strength of the iron-based acid centres. It is
worth to mention that intensity of the 1545 cmÀ1 band of
protonated pyridine was so small for AlSBA-1 and FeSBA-1
samples that the usual way of measuring the acid strength by
pyridine thermodesorption and comparison of the intensities of
this band for different desorption temperatures would have been
laden with too great error to be trustworthy.
Fig. 5 shows a temperature relationship of 2-IPN content in IPN
products obtained by naphthalene isopropylation over various
catalysts. As expected, the 2-IPN content increased in IPN isomeric
mixture with increasing the reaction temperature for each tested
catalyst. Regardless of the alkylation activity, all SBA-1 catalysts
showed a relatively low isomerization activity. As compared with
amorphous aluminosilicate, SBA-1 catalysts gave a lower 2-IPN
selectivity in alkylation. This is also true even for the samples with
the highest alkylation activity (AlSBA-1(39), GaSBA-1(40) and
GaSBA-1(17)). For example, more than 80% of 2-IPN in IPN was
detected if alkylation was carried out over ASA(6) catalyst at
200 8C, whereas the products obtained on SBA-1 catalysts
contained only 29–52% of 2-IPN (Fig. 5 and Tables 2–4).
Fig. 3 compares temperature relationship of propylene conver-
sion in the alkylation of naphthalene carried out over selected
Fig. 5. Temperature relationship of 2-IPN distribution in IPN obtained on various
catalysts. Symbols: (~) AlSBA-1(39); (^) AlSBA-1(180); (~) GaSBA-1(17); (&)
GaSBA-1(40); (Â) ASA(6) (dashed line); (*) AlMCM-48(26) (dashed line).
Fig. 4. Propylene conversion obtained at 150 8C as a function of nSi/nMe ratio in
various SBA-1 catalysts. Symbols: (&) GaSBA-1; (~) AlSBA-1; (^) FeSBA-1.

R. Brzozowski et al. / Applied Catalysis A: General 377 (2010) 76–82
81
28 and 24%, respectively, whereas those obtained over AlSBA-1(39)
were 26 and 22%, respectively.
The unique properties of AlSBA-1 and GaSBA-1 catalysts of a
relatively high alkylation activity along with low isomerization
activity make them very useful in the synthesis of DIPN high-
quality solvent. They give not only a product poor in 2,6-DIPN but
also a relatively low 2,6-DIPN content be obtained in a wide range
of operating temperature that is important if a catalyst deactiva-
tion is compensated by the process temperature increase. For
instance, if a DIPN product with 2,6-DIPN content lower than 15% is
required, then alkylation over the ASA(6) catalyst can be carried
out at temperature not higher than 170 8C (cf. Fig. 6), whereas over
the GaSBA-1(90), AlMCM-48(26) and AlSBA-1(180) in the temper-
ature ranges up to 230, 250 and even 270 8C, respectively.
Fig. 6. Temperature relationship of 2,6-DIPN content in DIPN products obtained on
various catalysts. Symbols: (~) AlSBA-1(39); (^) AlSBA-1(180); (~) GaSBA-1(17);
(&) GaSBA-1(40); (^) GaSBA-1(90); (Â) ASA(6) (dashed line); (*) AlMCM-48(26)
(dashed line).
4. Conclusions
Alkylation of naphthalene with propylene to diisopropyl-
naphthalenes (DIPN) over mesoporous AlSBA-1, GaSBA-1 and
FeSBA-1 catalysts was investigated. GaSBA-1 catalysts were slightly
more active than AlSBA-1 and FeSBA-1 revealed the lowest activity
mainly due to rapid deactivation. Moreover, AlSBA-1 and GaSBA-1
proved to be more active alkylation catalysts than amorphous
aluminosilicate and mesoporous AlMCM-48 catalyst. On the other
hand, SBA-1 catalysts revealed a relatively low isomerization
activity, therefore, selectivities of both 2-IPN in monoalkylation
and 2,6-DIPN and 2,7-DIPN in dialkylation were lower than those
observed over other alkylation catalysts. Consequently, DIPN
isomeric mixture was poor in 2,6-isomer. In the tested range of
nSi/nMe ratio, if the higher was the Al, Ga or Fe content in the SBA-1
catalyst the higher activity both in alkylation and isomerization was
observed. The high alkylation activity and a low 2,6-DIPN selectivity
could make the GaSBA-1 and AlSBA-1 mesoporous materials
promising catalysts for the synthesis of DIPN solvent.
Similarly as in alkylation, the GaSBA-1 catalysts were slightly
more active in isomerization of IPN than the AlSBA-1 catalysts. For
instance, IPN obtained at 200 8C on GaSBA-1(40) and AlSBA-1(39)
catalysts contained 43% and 41% of 2-IPN, respectively. The 2-IPN
content amounted to 82 and 76% for GaSBA-1(40) and AlSBA-1(39),
respectively, at the reaction temperature of 250 8C. Isomerization
requires stronger acid centres than alkylation, the presence of the
Lewis acid sites, much more numerous for GaSBA-1 may have
inductive effect on the acid strength of the Brønsted centres [37].
The mesoporous AlMCM-48(26) sample revealed a lower
isomerization activity than GaSBA-1 and AlSBA-1 catalysts,
especially at a high temperature, but its alkylation activity was
also lower than that of SBA-1 catalysts. Mesoporous FeSBA-1
catalysts were the least active samples both in alkylation and
isomerization. In the same way as in alkylation, the isomerization
activity increased with increasing the metal content in AlSBA-1 or
GaSBA-1 catalysts.
Acknowledgements
For isomerization of diisopropylnaphthalenes a relationship
similar as in the case of IPN was observed and a relatively small
quantities of 2,6-DIPN were formed. Content of 2,6-DIPN and 2,7-
DIPN in DIPN mixture increases with the increase of alkylation
temperature for all tested catalysts (Tables 2–4). The relationship
of 2,6-DIPN content in DIPN shown in Fig. 6 clearly indicates a
higher isomerization activity of the ASA(6) than those of
mesoporous SBA-1 and MCM-48 catalysts. The 2,6-DIPN isomer
comprised 34% of DIPN mixture obtained in alkylation of
naphthalene with propylene at 200 8C over the ASA(6) but only
5–8% over SBA-1 samples, except for the most active GaSBA-1(17)
catalyst over which 2,6-DIPN content approached 17%.
This work was financially supported by the Ministry of
Education, Culture, Sports, Science and Technology (MEXT) under
the Strategic Program for Building an Asian Science and
Technology Community Scheme and World Premier International
Research Center (WPI) Initiative on Materials Nanoarchitectonics,
MEXT, Japan. IR measurements were supported by the grant from
the Ministry of Science and Higher Education, Warsaw (project no.
N N204 1987 33).
References
Analogously to monoalkylation, the higher was the Al, Ga or Fe
content in the SBA-1 catalysts the higher 2,6-DIPN content in DIPN
mixture was observed (Fig. 6 and Tables 2–4). This result indicates
that, although lower than that of alkylation, the isomerization
activity of catalysts also rose with amount of the incorporated
metal. Nevertheless, the active centres containing Ga seem to
slightly be more effective in isomerization of DIPN isomers to 2,6-
DIPN and 2,7-DIPN than the Al centres (compare results obtained
at temperatures higher than 200 8C), possibly due to inductive
effect of Lewis sites on the strength of Brønsted centres. Isomeric
composition of DIPN products obtained in alkylation at the
reaction temperatures on or lower than 200 8C over GaSBA-
1(40) and AlSBA-1(39) was practically the same (Tables 2 and 3). At
a higher temperature DIPN mixture achieved over GaSBA-1(40)
[1] H.-G. Franck, J.W. Stadelhofer, Industrial Aromatic Chemistry. Raw Materials.
Processes. Products, Springer-Verlag, Berlin–Heidelberg–New York–London–
Paris–Tokyo, 1987.
[2] J.W. Stadelhofer, R.B. Zellerhoff, Chem. Ind. (Lond.) (1989) 208–211.
[3] Internet: http://www.r-k-s.de/pages/ruetas.htm.
[4] A. Katayama, M. Toba, G. Takeuchi, F. Mizukami, S. Niwa, S. Mitamura, J. Chem.
Soc., Chem. Commun. (1991) 39–40.
[5] Y. Sugi, M. Toba, Catal. Today 19 (1994) 187–212.
[6] J.A. Horsley, J.D. Fellmann, E.G. Derouane, C.M. Freeman, J. Catal. 147 (1994) 231–
240.
[7] Ch. Song, C. R. Acad. Sci. Paris, Serie IIc, Chimie/Chemistry 3 (2000) 477–496.
[8] E. Kikuchi, K. Sawada, M. Maeda, T. Matsuda, Stud. Surf. Sci. Catal. 90 (1994) 391–
396.
[9] J.-H. Kim, Y. Sugi, T. Matsuzaki, T. Hanaoka, Y. Kubota, X. Tu, M. Matsumoto,
Micropor. Mater. 5 (1995) 113–121.
[10] J.-H. Kim, Y. Sugi, T. Matsuzaki, T. Hanaoka, Y. Kubota, X. Tu, M. Matsumoto, S.
Nakata, A. Kato, G. Seo, C. Pak, Appl. Catal. A 131 (1995) 15–32.
[11] T. Matsuda, N. Takahashi, E. Kikuchi, in: C. Song, J.M. Garces, Y. Sugi (Eds.), Shape-
Selective Catalysis. Chemicals Synthesis and Hydrocarbon Processing, ACS Symp.
Ser. 738, ACS, Washington, DC, 1999, p. 282.
contained slightly more b,b-isomers than that obtained on AlSBA-
1(39). For instance, the DIPN product obtained at 250 8C over
GaSBA-1(40) exhibited the isomers of 2,6-DIPN, and 2,7-DIPN to be
[12] M.G. Cutrufello, I. Ferino, R. Monaci, E. Rombi, V. Solinas, P. Magnoux, M. Guisnet,
Appl. Catal. A 241 (2003) 91–111.

82
R. Brzozowski et al. / Applied Catalysis A: General 377 (2010) 76–82
[13] Y. Sugi, H. Maekawa, Y. Hasegawa, H. Naiki, K. Komura, Y. Kubota, Catal. Today 132
(2008) 27–37.
[14] R. Brzozowski, W. Skupin´ ski, J. Catal. 220 (2003) 13–22.
[15] R. Brzozowski, W. Skupin´ ski, J. Catal. 210 (2002) 313–318.
[25] A. Vinu, T. Krithiga, V. Murugesan, M. Hartmann, Adv. Mater. 16 (2004) 1817–1821.
[26] P. Srinivasu, A. Vinu, Chem. Eur. J. 14 (2008) 3553–3561.
[27] R. Brzozowski, A. Vinu, Top. Catal. 52 (2009) 1001–1004.
[28] R. Brzozowski, W. Skupin´ ski, M.H. Jamro´z, M. Skarz˙ yn´ ski, H. Otwinowska, J.
Chromatogr. A 946 (2002) 221–227.
ˇ
[16] J. Cejka, B. Wichterlova´, Catal. Rev. 44 (2002) 375–421.
ˇ
ˇ
ˇ
[29] J. De˘de˘cek, N. Zilkova, J. Cejka, Micropor. Mesopor. Mater. 44–45 (2001) 259–266.
[17] J.-H. Kim, M. Tanabe, M. Niwa, Micropor. Mater. 10 (1997) 85–93.
[18] G. Kamalakar, S.J. Kulkarni, K.V. Raghavan, S. Unnikrishnan, A.B. Halgeri, J. Mol.
Catal. A: Chem. 149 (1999) 283–288.
[19] R. Maheswari, K. Shanthi, T. Sivakumar, S. Narayanan, Appl. Catal. A 245 (2003)
221–230.
[20] S.B. Pu, J.B. Kim, M. Seno, T. Inui, Micropor. Mater. 10 (1997) 25–33.
[21] T. Krithiga, A. Vinu, K. Ariga, B. Arabindoo, M. Palanichamy, V. Murugesan, J. Mol.
Catal. A: Chem. 237 (2005) 238–245.
ˇ
´
[30] G. Kosova, S. Ernst, M. Hartmann, J. Cejka, Eur. J. Inorg. Chem. (2005) 1154–1161.
[31] B. Gil, in: J. Cejka, J. Pere´z-Pariente, W.J. Roth (Eds.), Zeolites: From Model
Materials to Industrial Catalysts, Research Signpost, 2008, pp. 173–206.
[32] A.M. Ferretti, C. Oliva, L. Forni, G. Berlier, A. Zecchina, C. Lamberti, J. Catal. 208
(2002) 83–88.
[33] K. Hadjiivanov, Catal. Rev. Sci. Eng. 42 (1&2) (2000) 71–144.
[34] B. Gil, A. Adamski, Micropor. Mesopor. Mater. 127 (2010) 82–89.
´
´
[22] R. Brzozowski, A. Vinu, T. Mori, Catal. Commun. 8 (2007) 1681–1683.
[23] A.E. Garcia-Bennett, S. Williamson, P.A. Wright, I.J. Shannon, J. Mater. Chem. 12
(2002) 3533–3540.
[24] M. Hartmann, A. Vinu, S.P. Elangovan, V. Murugesan, W. Bohlman, Chem. Com-
mun. (2002) 1238–1239.
[35] R. Brzozowski, J.C. Dobrowolski, M.H. Jamroz, W. Skupinski, J. Mol. Catal. A 170
(2001) 95–99.
[36] G. Tasi, F. Mizukami, I. Palinko, M. Toba, A. Kukovecz, J. Phys. Chem. A 105 (2001)
6513–6518.
[37] C. Miradatos, D. Barthomeuf, J. Chem. Soc., Chem. Commun. 2 (1981) 39.