Isopropylation of naphthalene by isopropyl alcohol over USY catalyst: An investigation in the high-pressure fixed-bed flow reactor

Article abstract of DOI:10.1016/S0021-9517(03)00212-4

Catalytic performances of USY, H-mordenite, dealuminated H-mordenite, and H-MCM-22 zeolite catalysts in the isopropylation of naphthalene by isopropyl alcohol with decalin or cyclohexane as a solvent were compared in a high-pressure fixed-bed flow reactor. For the USY catalyst, reaction conditions, such as reaction temperature and pressure, reactant ratio and space velocity, and solvent concentration and type, were controlled to investigate in detail the effect of reaction conditions on the catalytic activity. Over H-mordenite, it was found that 2,6-diisopropylnaphthalene (2,6-DIPN) could be selectively synthesized with a 2,6-/2,7-DIPN ratio of 2.46, and dealumination could enhance not only the selectivity of 2,6-DIPN, with a 2,6-/2,7-DIPN ratio of 2.67, but also the conversion of naphthalene, which was 27.4%, three times as high as that over the unmodified one at 6 h of reaction time on stream. However, neither the H-mordenite or the dealuminated one were catalytically stable and the selectivity of DIPN was at a very low level of less than 12%. In contrast, over the USY catalyst, a high and stable conversion of about 90%, a high selectivity of DIPN of more than 40%, and a considerable 2,6-/2,7-DIPN ratio of 1.46 could be achieved by adjusting the reaction conditions, although no shape selectivity was observed on USY. On the other hand, only a low 2,6-/2,7-DIPN ratio of 0.47 with a low conversion of about 30% was revealed over H-MCM-22, which indicates that the reaction takes place on the external surface of this zeolite. An attempt has been made to explain the catalytic activity, selectivity, and stability in relation to the zeolite structures, product properties, and reaction conditions.

Full text of DOI:10.1016/S0021-9517(03)00212-4

Journal of Catalysis 220 (2003) 265–272
www.elsevier.com/locate/jcat
Isopropylation of naphthalene by isopropyl alcohol over USY catalyst:
an investigation in the high-pressure fixed-bed flow reactor
Jun Wang,1 Jung-Nam Park, Yong-Ki Park, and Chul Wee Lee ∗
Advanced Chemical Technology Division, Korea Research Institute of Chemical Technology (KRICT), PO Box 107, Yusung, Daejeon 305-600, South Korea
Received 15 October 2002; revised 17 December 2002; accepted 29 April 2003
Abstract
Catalytic performances of USY, H-mordenite, dealuminated H-mordenite, and H-MCM-22 zeolite catalysts in the isopropylation of naph-
thalene by isopropyl alcohol with decalin or cyclohexane as a solvent were compared in a high-pressure fixed-bed flow reactor. For the
USY catalyst, reaction conditions, such as reaction temperature and pressure, reactant ratio and space velocity, and solvent concentration
and type, were controlled to investigate in detail the effect of reaction conditions on the catalytic activity. Over H-mordenite, it was found
that 2,6-diisopropylnaphthalene (2,6-DIPN) could be selectively synthesized with a 2,6-/2,7-DIPN ratio of 2.46, and dealumination could
enhance not only the selectivity of 2,6-DIPN, with a 2,6-/2,7-DIPN ratio of 2.67, but also the conversion of naphthalene, which was 27.4%,
three times as high as that over the unmodified one at 6 h of reaction time on stream. However, neither the H-mordenite or the dealuminated
one were catalytically stable and the selectivity of DIPN was at a very low level of less than 12%. In contrast, over the USY catalyst, a
high and stable conversion of about 90%, a high selectivity of DIPN of more than 40%, and a considerable 2,6-/2,7-DIPN ratio of 1.46
could be achieved by adjusting the reaction conditions, although no shape selectivity was observed on USY. On the other hand, only a low
2,6-/2,7-DIPN ratio of 0.47 with a low conversion of about 30% was revealed over H-MCM-22, which indicates that the reaction takes place
on the external surface of this zeolite. An attempt has been made to explain the catalytic activity, selectivity, and stability in relation to the
zeolite structures, product properties, and reaction conditions.
2003 Published by Elsevier Inc.
Keywords: Isopropylation of naphthalene; Diisopropylnaphthalene; USY; Mordenite; MCM-22; Zeolite; Alkylation
1. Introduction
uct mixture due to the placement of alkyl groups at adjacent
nucleus positions [1]. Because of the difficulty of completely
separating these isomers, especially β, β-isomers [2], the se-
lective production of 2,6-DAN with a high 2,6-/2,7-DAN
ratio by alkylation of naphthalene is highly desired. How-
ever, selective synthesis of 2,6-DAN is difficult by alkyla-
tion of naphthalene with conventional Friedel–Crafts cata-
lysts [3,4] or over silica–alumina catalyst [5,6], which lead
to low β, β-selectivity and equal amounts of 2,6- and 2,7-
isomers. Therefore, the use of shape-selective properties of
zeolites in the alkylation of naphthalene should be helpful in
the synthesis of 2,6-DAN, although the average molecular
dimensions of the β, β-isomers are very similar.
Various alkylating agents, such as methanol [7–9], etha-
nol [10], isopropyl alcohol [11–16], propene [11,17–19],
isopropyl bromide [20], tert-butyl alcohol [21–23], cyclo-
hexyl bromide [24,25], and cyclohexene [24–26], have been
evaluated in this type of reaction over various zeolite cat-
alysts. Among these studies, isopropylation of naphtha-
Alkylation of naphthalene is of industrial importance for
producing 2,6-dialkylnaphthalenes (2,6-DAN), the precur-
sor of naphthalene-2,6-dicarboxylic acid used in the synthe-
sis of advanced aromatic polymer materials such as poly-
ethylene naphthalate, polybutylene naphthalate, and ther-
motropic liquid crystalline polymers. Further expansion in
the utilization of these materials depends on the development
of an efficient method for preparation of 2,6-DAN.
There are as many as 10 possible DAN including α, α-
(1,4-; 1,5-; 1,8-), α, β-(1,6-; 1,7-; 1,3-; 1,2-), and β, β-(2,6-;
2,7-; 2,3-) disubstituted isomers, and in fact, 1,2-, 1,8-, and
2,3-isomers are not frequently detected in the reaction prod-
*
Corresponding author.
E-mail address: chulwee@pado.krict.re.kr (C.W. Lee).
On a study leave from the Department of Chemical Engineering, Nan-
1
jing University of Technology, Nanjing 210009, China.
0021-9517/$ – see front matter
2003 Published by Elsevier Inc.
doi:10.1016/S0021-9517(03)00212-4

266
J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
lene has been attracting much attention for both practical
and fundamental reasons. On the one hand, the isopropyl
group is more sterically hindered than the methyl group,
implying a high possibility of the selective synthesis of 2,6-
diisopropylnaphthalene (2,6-DIPN) in the suitable intracrys-
talline channel of zeolite. On the other hand, 2,6-DIPN is
more easily oxidized into naphthalene-2,6-dicarboxylic acid
as compared with 2,6-dimethylnaphthalene, and has more
atomic economy in the oxidation process as compared with
2,6-di-tert-butylnaphthaleneor 2,6-dicyclohexylnaphthalene.
Indeed, Katayama et al. first found a higher selectivity of H-
mordenite for 2,6-DIPN than for H-Y, H-L, and H-ZSM-5,
respectively [11]. After that, dealuminated mordenite was
revealed to be more active and selective for 2,6-DIPN than
the unmodified one [12,17,18,27–30]. By contrast, some lit-
erature demonstrated that H-Y is also a promising catalyst
in terms of activity, β, β-selectivity, and/or 2,6-/2,7-DIPN
ratio in both liquid and gas-phase reactions [13,20,31,32].
However, most of the previous studies have been carried
out in an autoclave. Since the serious drawback of the auto-
clave reactor is the difficulty of evaluating the catalyst stabil-
ity, up to now, only a few papers have investigated the stabil-
ity of the catalyst in this reaction. For example, Inui and co-
workers [14] carried out the isopropylation of naphthalene
by isopropyl alcohol using the atmospheric pressure flow
fixed-bed reactor over Y, Beta, mordenite, MCM-41, and
MCM-48 catalysts, respectively, and deactivation was found
on all catalysts within 3 h of reaction time on stream, with
Y zeolite being comparatively stable. Colon et al. [15,16]
performed this reaction over H-Y and H-Beta catalysts in a
flow reactor under high pressure, but they used a high re-
action temperature of 350 ◦C, and a high number of side
reactions occurred. Very recently, Kamalakar et al. [23] at-
tempted the vapor-phase tert-butylation of naphthalene over
cerium-modified Y zeolites and revealed a clear decrease of
conversion of naphthalene with time on stream.
at 550 ◦C for 8 h. The above procedure was repeated once
more. The obtained sample was designated as H-MD. The
SiO2/Al2O3 ratio of 49 for H-MD was obtained by the ICP-
AES analysis. Na-MCM-22 (SiO2/Al2O3 = 30) was syn-
thesized according to previous literature [33]. H-MCM-22
was prepared by the repeated ion exchange of Na-MCM-22
with an aqueous solution of ammonium nitrate, followed by
drying and calcination.
All experiments were carried out in a high-pressure
fixed-bed down-flow microreactor with an inner diameter
of 10 mm. The reactant mixture of naphthalene, isopropyl
alcohol, and decalin or cyclohexane was injected by a high-
pressure pump into the tubular stainless-steel reactor, the
middle stage of which was charged with 1.0 g of catalyst
in the form of granules (20–40 mesh). The catalyst was ini-
tially pretreated in situ in N2 flow at 550 ◦C for 6 h before
reactants were injected into the reactor. The typical reaction
conditions were as follows: 250 ◦C reaction temperature,
3.0 MPa reaction pressure, 5.3 h−1 weight hourly space ve-
locity (WHSV, total liquid feed), 6 h reaction time on stream
(TOS), and 1:2:6 molar ratio of naphthalene, isopropyl al-
cohol, and decalin (or cyclohexane) in liquid feed. Also, to
keep the system at a steady high pressure, 20 ml/min of N2
was introduced cocurrently with the reaction feed into the
reactor at the reactor head. Under these reaction conditions,
the reaction could take place in the liquid phase. For USY
catalysts, the reaction conditions were altered systematically
to test the effect of reaction conditions. The reactor effluent
was condensed in the sampler at −10 ◦C by cold ethanol
and sampled hourly. Reactants and products were analyzed
by GC (HP 5890 II) using FID as the detector furnished
with a 60 m × 0.25 mm × 0.25 µm SE54 (J&W Scientific)
capillary column. The analytical conditions was as follows:
flow rate of He (carrier gas) in the column of 0.78 ml/min;
split ratio of 80; sample injection volume of 0.1 µl; detector
temperature of 300 ◦C; and injection temperature of 250 ◦C.
The column oven temperature program started with an initial
temperature of 40 ◦C for 2 min, followed by a temperature
increase rate of 4 ◦C/min and a final temperature of 280 ◦C
for 30 min. The conversion and selectivity were calculated
as follows. Conv. = (Mn0 −Mn)/Mn0 ×100%, where Conv.
is conversion of naphthalene, Mn0 is the molar percentage
of naphthalene before reaction, and Mn is the molar percent-
age of naphthalene after reaction. Sel. = M1/Mt × 100%,
where Sel. is the selectivity of IPN (monoisopropylnaph-
thalene), DIPN, or PIPN (polyisopropylnaphthalene), M1 is
the molar percentage of IPN, DIPN, or PIPN, and Mt is the
sum of molar percentage of the above three kinds isopropy-
In this work, we have investigated, in detail, the catalytic
activity, selectivity, and stability of a commercially available
USY (ultrastable H-Y zeolite) in the isopropylation of naph-
thalene by isopropyl alcohol using a high-pressure fixed-bed
flow reactor. In order to understand the catalytic behavior
of the USY catalyst, a comparative investigation is also per-
formed with H-mordenite and H-MCM-22 zeolite catalysts.
2. Experimental
USY (CBV780, SiO2/Al2O3 = 80) and H-M (H-morde-
nite, CP814, SiO2/Al2O3 = 25) were from Zeolyst. Naph-
thalene was from Junsei Chem. Co. Ltd., Japan; isopropyl
alcohol from Duksan Pure Chem. Co. Ltd., Korea; decalin
from Kanto Chem. Co. Inc., Japan; and cyclohexane from
Aldrich. All chemicals were guaranteed reagent grade and
used as purchased. The dealuminated H-mordenite was pre-
pared by treatment of H-M in a 2 M HCl aqueous solution
at 80 ◦C for 2 h, followed by calcination in a muffle furnace
lated products. β, β-Selectivity = (M2,6-DIPN +M2,7-DIPN
+
M
2,3-DIPN)/MtDIPN × 100%, where M2,6-DIPN, M2,7-DIPN
,
and M2,3-DIPN are the molar percentages of 2,6-DIPN, 2,7-
DIPN, and 2,3-DIPN, respectively, and MtDIPN is the sum
of the molar percentage of all the DIPN isomers. 2,6-/2,7-
DIPN stands for the molar ratio of 2,6-DIPN to 2,7-DIPN,
and 2-/1-IPN for 2-IPN to 1-IPN. Since only a negligi-
ble amount of 2,3-DIPN was observed in this work, β, β-

J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
267
selectivity could be understood as the selectivity of 2,6-
DIPN and 2,7-DIPN.
Table 2
Catalytic activity and selectivity of different zeolites in the isopropylation
of naphthalene by isopropyl alcohol under typical reaction conditions
a
Catalyst
USY
89.9
H-M
9.7
H-MD
27.4
H-MCM-22
31.3
3. Results and discussion
Conv. (mol%)
Sel. (mol%)
IPN
DIPN
PIPN
2-/1-IPN
2,6-/2,7-DIPN
β,β-Selectivity (mol%)
3.1. Conversion, selectivity, and stability of various zeolite
catalysts
31.9
40.1
28.0
1.25
1.28
52.8
93.8
6.2
0
1.61
2.46
100
88.5
11.1
0.4
2.36
2.67
83.1
93.1
6.7
0.2
1.30
0.98
55.3
Table 1 compares the product selectivity over various ze-
olite catalysts at a similar conversion level in the range of
17.3–22.6%. It is observed that IPN is the major product on
all catalysts, while less than 1% of PIPN is detected on H-M,
H-MD, or H-MCM-22 catalysts, and the DIPN and β, β-
selectivity on USY is the highest among the four different
catalysts. Moreover, H-MD gives both the highest 2,6-/2,7-
DIPN, and 2-/1-IPN ratios and β, β-selectivity while those
on H-MCM-22 are very low. Although a very high β, β-
selectivity of 85% is observed on USY, 2,6-/2,7-DIPN is
rather low at 0.34. When the conversion and selectivity are
compared under the same reaction conditions, as shown in
Table 2, it is indicated that the naphthalene conversion de-
creases in the order of USY > H-MCM-22 > H-MD >
H-M, and DIPN becomes the major product only on USY.
Furthermore, H-MD still possesses the highest 2,6-/2,7-
DIPN, 2-/1-IPN, and β, β-selectivity, while those on H-
MCM-22 are still at a low level of 0.98. However, with a
very high conversion of 89.9%, USY exhibits a much higher
2,6-/2,7-DIPN ratio of 1.28 than that in Table 1 of 0.34, and
at the same time, β, β-selectivity drops to a comparatively
low level of 52.8%. These results imply that for the USY
catalyst the reaction conditions could impose a significant
effect on the product distribution.
a
◦
Reaction conditions: reaction temperature = 250 C, reaction pres-
−1
sure = 3.0 MPa, WHSV = 5.3 h (total liquid feed), TOS = 6 h, naph-
thalene:isopropyl alcohol:decalin = 1:2:6 (molar).
Fig. 1. Catalytic activity and stability of four different zeolite catalysts in the
isopropylation of naphthalene by isopropyl alcohol under typical reaction
◦
conditions (reaction temperature = 250 C, reaction pressure = 3.0 MPa,
−1
WHSV = 5.3 h
(total liquid feed), TOS = 6 h, naphthalene:isopropyl
alcohol:decalin = 1:2:6 (molar)).
Catalytic activity and stability of four different zeolites
under typical reaction conditions are plotted and compared
in Fig. 1. It is obvious that USY is a very stable and active
catalyst with conversion of about 90%, and by contrast, H-M
deactivates to a conversion of lower than 10% after 6 h TOS.
Through dealumination, the activity of H-MD increases sub-
stantially; however, it still exhibits deactivation with reaction
time on stream. For H-MCM-22, the activity is rather stable,
but at a low conversion of 20–30%.
Table 1
2,6-DIPN and 2,7-DIPN are the slimmest among the 10
DIPN isomers, with the molecular dimension of the for-
mer (0.661 × 0.661 × 1.423 nm) being only slightly smaller
than the latter (0.662 × 0.726 × 1.376 nm) [34], and thus,
it should be difficult to obtain a high 2,6-/2,7-DIPN ratio
even with shape-selective zeolite catalysts. As shown in Ta-
bles 1 and 2, whatever the conversion, a 2,6-/2,7-DIPN ratio
higher than 2.0 can always be observed on the two employed
mordenite catalysts, indicating the shape-selective formation
of 2,6-DIPN. This observation is in a good agreement with
the previous results [11,12,17,18]. It is known that morden-
ite consists of a unique unidimensional pore channel with the
12-membered ring (12MR) window of 0.65 × 0.70 nm [32].
Considering the more linear structure of 2,6-DIPN than that
of 2,7-DIPN, it is reasonable to deduce that in the unidimen-
sional tunnels of mordenite, the attack of isopropyl alcohol
to 2-IPN is more preferable at position 6 of 2-IPN than that
Comparison of catalytic selectivity of various zeolites at a conversion of
about 20% in the isopropylation of naphthalene by isopropyl alcohol
a
Catalyst
b
c
d
c
USY
H-M
H-MD
22.6
H-MCM-22
21.1
Conv. (mol%)
Sel. (mol%)
IPN
DIPN
PIPN
2-/1-IPN
2,6-/2,7-DIPN
β,β-Selectivity (mol%)
18.5
17.3
65.4
23.0
11.7
0.36
0.34
85.0
86.4
13.6
0
1.61
2.02
78.5
84.3
14.9
0.8
3.55
3.12
86.3
91.3
8.1
0.6
0.86
0.47
34.9
a
−1
Reaction conditions: reaction pressure = 3.0 MPa, WHSV = 5.3 h
(total liquid feed), naphthalene:isopropyl alcohol:decalin = 1:2:6 (molar),
others are indicated below.
b
c
d
◦
Reaction temperature = 200 C, TOS = 1 h.
◦
Reaction temperature = 250 C, TOS = 2 h.
◦
Reaction temperature = 230 C, TOS = 2 h.

268
J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
at position 7, resulting in the selective formation of 2,6-
DIPN. This proposal is suggested by molecular graphics and
molecular mechanics studies [35], which pointed out the ex-
istence of significant energy barriers when the 2,7-isomer
is diffusing in mordenite channels while the diffusion of
the 2,6-isomer is unimpeded, and very recently reconfirmed
by the first principal quantum chemical method [34], which
revealed that the 2,6-isomer could diffuse through the mor-
denite main channels the easiest among the possible isomers.
This shape-selective effect of mordenite also causes the very
low selectivity of PIPN (Tables 1 and 2). Despite the exper-
imental and theoretical evidence for the shape selectivity of
mordenite in this reaction, and considering the very slight
difference in their molecular size, one still tends to suspect
that this size effect cannot sufficiently account for the fact
that the ratio of 2,6-/2,7-DIPN is higher than 2.0. How-
ever, based on computational analysis using MOPAC, Song
et al. [36] put forward a hypothesis termed as restricted elec-
tronic transition-state selectivity effected by frontier electron
density, which demonstrates that the carbon at position 6
of 2-IPN has a higher frontier electron density than that
of position 7 and thus higher reactivity toward electrophilic
substitution, suggesting that the formation of 2,6-DIPN is
more electronically favored than that of 2,7-DIPN. As a con-
sequence, the high 2,6-/2,7-DIPN ratio on mordenite must
arise from the combined effect of the conventional shape se-
lectivity and the specific restricted electronic transition-state
selectivity.
Accompanied with high selectivity, severe deactivation
and thus low activity are found for H-M. Due to its uni-
dimensional pore structure, the deactivation of H-M seems
to be accelerated by the formation of polynuclear cracking-
coking products at the major channel mouth, blocking access
for the reactant to the internal sites of the channel in the
mordenite zeolite. Dealumination of H-M not only reduces
the number of acid sites but also modifies the pore distri-
bution, resulting in an increase of mesopore structure [17].
It could therefore be considered that low acid density might
lead to the slow deactivation of the catalyst. Meanwhile, in-
creased mesoporosity could introduce more reactants into or
facilitate the products desorbing out of the mordenite ma-
jor channels. Thus, over H-MD, more contributions from
the internal channel surface could be expected for the over-
all product distribution, as compared with unmodified H-M.
This could be a possible interpretation for the higher activ-
ity, stability, and 2,6-/2,7-DIPN ratio over H-MD than those
over H-M (Tables 1 and 2 and Fig. 1).
is unexpected on H-MCM-22 is the observation of much
lower PIPN selectivity, which is very similar to those on the
two different mordenite catalysts. It was found recently [38]
that H-MCM-22 possesses a high concentration of strong
acid sites; however, they locate mostly inside the supercages
and narrow channels, and are inaccessible to the reactant.
On the other hand, only some medium-strength acid sites
are located on the external surface of this zeolite, and these
medium strength acid sites are supposed to contribute to the
formation of IPN, and not PIPN [23,31], resulting in the low
selectivity of PIPN.
In contrast to the low and unstable activity and low se-
lectivity for DIPN over mordenite catalysts, USY exhibits
a high conversion of 90% up to the employed 6 h TOS
with 40% of DIPN a main product. Moreover, under suit-
able reaction conditions, a considerable 2,6-/2,7-DIPN ratio
of 1.28 is achieved, which is higher than that in the equilib-
rium of DIPN isomer mixtures (about 1) [1]. Y zeolite has a
three-dimensional 12MR channel (0.74 nm) with large inter-
+
val cavities (1.3 nm). When USY is prepared from (NH4
,
Na)-Y zeolite by hydrothermal treatment [39,40], a large
amount of mesopores could be generated with the decrease
of acid density and the increase of acid strength. As a conse-
quence, the high and stable activity of USY in Fig. 1 could
be ascribed to its easily accessible pore channel and low acid
density. On the other hand, the strong acidity of USY might
associate with the comparatively high selectivity of PIPN
(28.7%) in Table 2 [23,31]. It is also noteworthy in Table 2
that the β, β-selectivity of 52.8% over USY is lower than the
thermodynamic equilibrium value (> 70%) [1]. This obser-
vation possibly arises from the shape selectivity that occurs
in the entrance to the pores of large-pore zeolites, which
causes relatively high selectivity for 1,3- and 1,4-DIPN, and
simultaneously decreases the β, β-selectivity [41].
Now that the high and stable activity, as well as the
high 2,6-/2,7-DIPN ratio, can be achieved over USY, and
moreover, catalytic activities can be altered to a great extent
by changing reaction conditions, further investigation under
various reaction conditions is then focused on the USY cat-
alyst as below.
3.2. Effect of the reaction temperature
Fig. 2 shows the activity and selectivity of USY as a func-
tion of the reaction temperature during the isopropylation
of naphthalene. It is observed in Fig. 2A that the conver-
sion increases rapidly with the reaction temperature up to
230 ◦C, and it goes up very slowly beyond 230 ◦C, showing
that the conversion reaches to 93% at 300 ◦C. At the same
time, IPN decreases, and DIPN and PIPN increase drasti-
cally with temperatures up to 230 ◦C, and at higher tem-
peratures, this trend becomes slow. When the temperature
is as high as 300 ◦C, a small amount of cracking of solvent
is observed. Moreover, it is found in Fig. 2B that 2-/1-IPN
increases very quickly with reaction temperature, while 2,6-
/2,7-DIPN possesses a maximum point at 250 ◦C.
MCM-22 is a medium-sized pore zeolite with a supercage
(0.71 × 0.71 × 1.82 nm) and narrow pore window (10MR,
0.52 nm) [37]. Theoretically, the narrow pore windows of
MCM-22 make it impossible for the reactant to diffuse into
and the product to desorb out of the zeolite channels, and
therefore, the reaction likely occurs at the acid sites on the
external surface. This is the reason why H-MCM-22 shows
a low β, β-selectivity and 2,6-/2,7-DIPN ratio and is sta-
ble with low conversion (Tables 1 and 2). However, what

J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
269
Fig. 2. Catalytic performances of the USY catalyst in the isopropylation of naphthalene by isopropyl alcohol as a function of the reaction temperature (reaction
−1
conditions: reaction pressure = 3.0 MPa, WHSV = 5.3 h (total liquid feed), TOS = 6 h, naphthalene:isopropyl alcohol:decalin = 1:2:6 (molar)).
Fig. 3. Catalytic performances of the USY catalyst in the isopropylation of naphthalene by isopropyl alcohol as a function of WHSV (reaction conditions:
◦
reaction temperature = 250 C, reaction pressure = 3.0 MPa, TOS = 6 h, naphthalene:isopropyl alcohol:decalin = 1:2:6 (molar)).
It is known that this reaction is a consecutive reac-
tion [15]. The high reaction temperature would facilitate the
further alkylation of IPN by isopropyl alcohol to DIPN, fol-
lowed by PIPN, which is observed in Fig. 2A. On the other
hand, the α-position of naphthalene is more reactive and
the α-substituted products are thermodynamicallyless stable
than the β-position. At a low reaction temperature, the kinet-
ically controlled α-isomer is formed initially. It is rearranged
into the thermodynamically favored β-isomer at high re-
action temperatures. This is the reason why the 2-/1-IPN
ratio is much higher at high reaction temperatures. However,
this high 2-/1-IPN ratio does not necessarily indicate the
shape selectivity of the USY zeolite here, although 2-IPN
is slimmer than 1-IPN. Furthermore, the comparatively high
ratio of 2,6-/2,7-DIPN (> 1) is also found at high reaction
temperatures. Taking into account conversion, selectivity of
DIPN, 2-/1-IPN, and 2,6-/2,7-DIPN ratios, 250 ◦C can be
suggested as the optimal reaction temperature.
3.3. Effect of the reactant space velocity
Fig. 3 illustrates the effect of weight hourly space veloc-
ity on the catalytic activity of USY. Fig. 3A indicates that
the catalytic activity decreases very slowly when WHSV is
less than 12.5 h−1, and the decreasing rate quickens at a high
WHSV. Since the alkylation reaction is a consecutive reac-
tion, when WHSV becomes high, only a small part of IPN
has enough reaction time to be further alkylated into DIPN
and TIPN subsequently, indicating the high selectivity for
IPN at high WHSV. Regarding the conversion and selectiv-
ity for DIPN, it is reasonable to choose the low WHSV for
this reaction. Fig. 3B reveals that both 2-/1-IPN and 2,6-/
2,7-DIPN decreases with the enhancement of WHSV. It is
suggested that the high WHSV allows first the formation of
thermodynamically favorable 1-IPN, and the low WHSV fa-
cilitates the rearrangement of 1-IPN into 2-IPN. Based on
the above understandings, a WHSV of 5–6 h−1 is preferred

270
J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
Table 3
Catalytic performance of USY in the isopropylation of naphthalene by iso-
a
propyl alcohol as a function of solvent concentration
Molar ratio of naphthalene,
isopropyl alcohol, and decalin
1:2:6
1:2:10
1:2:6
1:2:10
◦
Reaction temperature ( C)
200
200
250
250
Conv. (mol%)
Sel. (mol%)
IPN
DIPN
PIPN
2-/1-IPN
2,6-/2,7-DIPN
β,β-Selectivity (mol%)
12.6
34.9
89.9
93.5
93.4
5.8
0.8
0.2
0.25
83.3
76.2
18.1
5.7
0.28
0.3
31.9
40.1
28.0
1.25
1.28
52.8
22.6
43.4
34.0
1.48
1.46
58.0
83.8
a
−1
Fig. 4. Catalytic activity and stability of USY (hollow symbol) and dealu-
minated mordenite (black symbol) catalysts in the isopropylation of naph-
thalene by isopropyl alcohol as a function of the reaction pressure (reaction
conditions: reaction temperature = 250 C, WHSV = 5.3 h (total liquid
feed), TOS = 6 h, naphthalene:isopropyl alcohol:decalin = 1:2:6 (molar)).
Reaction conditions: reaction pressure = 3.0 MPa, WHSV = 5.3 h
(total liquid feed), TOS = 6 h.
◦
−1
Table 4
Comparison of catalytic performance of USY in the isopropylation of naph-
thalene by isopropyl alcohol with decalin and cyclohexane as the solvent
a
for achieving a high conversion and high 2-IPN and 2,6-
DIPN selectivities, respectively.
Solvent
Decalin Cyclohexane Decalin Cyclohexane
◦
Reaction temperature ( C) 200
200
30.2
250
93.5
250
92.9
3.4. Effect of the reaction pressure
Conv. (mol%)
Sel. (mol%)
IPN
DIPN
PIPN
34.9
The effect of reaction pressure on catalytic activity and
stability under typical reaction conditions is displayed in
Fig. 4. It can be seen that at pressures higher than 1.0 MPa,
USY exhibits high and stable activity with a conversion of
around 90%. At atmospheric pressure, a little deactivation
appears, with a conversion of 85% at 6 h TOS. By con-
trast, for H-MD catalyst it is demonstrated in Fig. 4 that both
catalytic activity and stability decrease with the decrease of
reaction pressure. Although H-MD is not stable at any of
the employed reaction pressures, its stability is improved re-
markably at high pressure such as 3.0 MPa. For this reaction,
some previous literature [14,31] has found the deactivation
of Y and other catalysts in the gas-phase flow fixed-bed re-
actor. Here, the high pressure could assure homogeneity of
the reactants, in either the liquid or the supercritical phase,
and it is suggested that the liquid-phase reaction allows sup-
pression of catalyst deactivation by dissolving some of the
initially formed heavy products (the precursor of the coke)
with the solvent. The role of solvent has not been evalu-
ated, in this reaction, with this effect of reaction pressure
before; however, it has been reported in benzene alkylation
reactions [42,43].
76.2
18.1
5.7
84.6
13.7
1.7
0.28
0.26
81.3
22.6
43.4
34.0
1.48
1.46
58.0
25.9
42.2
31.8
3.38
1.50
61.2
2-/1-IPN
0.28
0.3
2,6-/2,7-DIPN
β,β-Selectivity (mol%)
83.8
a
−1
Reaction conditions: reaction pressure = 3.0 MPa, WHSV = 5.3 h
(total liquid feed), TOS = 6 h, naphthalene:isopropyl alcohol:decalin (or
cyclohexane) = 1:2:10 (molar).
apparently increase of PIPN is found simultaneously. When
the concentration of solvent is raised, the relative concen-
trations of the two reactants decrease, which should cause
the decrease of conversion. On the other hand, in case of the
same WHSV value (total liquid feed) for all reactions, the
space velocities of the two reactants are enhanced, and thus,
as an overall result, an improvement of activity is given in
Table 3, especially at a reaction temperature of 200 ◦C.
The comparison of decalin and cyclohexane as the re-
action solvent over USY catalyst is presented in Table 4.
No significant difference in reactivity can be observed be-
tween the two different kinds of solvents, except that the
β, β-selectivity and 2,6-/2,7-DIPN ratio increase in a very
small amount.
3.5. Effect of the solvent concentration and type
3.6. Effect of the reactant ratio
Table 3 compares the reaction results on USY with 1:2:6
and 1:2:10 molar ratios of naphthalene, isopropyl alcohol,
and decalin. At both high and low reaction temperatures, an
enhancement of conversion of naphthalene is observed when
the concentration of solvent in the reactant mixture is raised,
while slight improvements of 2,6-/2,7-DIPN and 2-/1-IPN
ratios and β, β-selectivity are also observed. However, an
The molar ratio of naphthalene and isopropyl alcohol is
changed to test the impact of reactant ratio upon the re-
activity of USY catalyst, as shown in Table 5. It is found
that both activity and selectivity are influenced drastically
by the change of reactant ratio. With the increase of iso-

J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
271
Table 5
and high selectivity for DIPN (> 40%) can be produced.
Moreover, both conversion and selectivity can be altered by
adjusting the reaction parameters, such as reaction tempera-
ture, pressure, reactant ratio, space velocity, and solvent con-
centration and type. By optimizing the reaction conditions,
it is also possible to achieve a considerable 2,6-/2,7-DIPN
ratio of 1.5 (Table 4), which is clearly higher than that in
the equilibrium DIPN isomer mixtures containing an equal
amount of 2,6- and 2,7-DIPN.
Catalytic performance of USY in the isopropylation of naphthalene by iso-
propyl alcohol as a function of reactant ratio
a
Molar ratio of naphthalene,
isopropyl alcohol, and decalin
1:1:10
66.7
1:2:10
93.5
1:4:10
100
Conv. (mol%)
Sel. (mol%)
IPN
DIPN
PIPN
2-/1-IPN
2,6-/2,7-DIPN
β,β-Selectivity (mol%)
44.6
40.4
15.1
2.95
1.31
70.3
22.6
43.4
34.0
1.48
1.46
58.0
5.7
29.9
64.4
1.26
0.93
33.5
Acknowledgments
a
◦
Reaction conditions: reaction temperature = 250 C, reaction pressure
Dr. Jun Wang acknowledges the financial support of the
Foreign Scientist Invitation Program sponsored by the Korea
Institute of S&T Evaluation and Planning (KISTEP). This
work was supported by grant No. R01-2003-000-10069-0
from the Basic Research Program of the Korea Science and
Engineering Foundation.
−1
= 3.0 MPa, WHSV = 5.3 h (total liquid feed), TOS = 6 h.
propyl alcohol in the reactant mixture, the conversion and
PIPN selectivity rapidly increase, and a decrease in IPN
and β, β-selectivity and 2-/1-IPN ratio can been observed.
Moreover, the highest value of DIPN selectivity (43.4%) and
2,6-/2,7-DIPN ratio (1.46) can be achieved with the molar
ratio of naphthalene and isopropyl alcohol being 1:2. How-
ever, at a 1:1 naphthalene and isopropyl alcohol molar ratio,
a comparatively high β, β-selectivity (70.3%) and consid-
erable 2,6-/2,7-DIPN ratio (1.31) can be obtained, even if
its conversion is not so high (66.7%). The increase of con-
version of naphthalene and selectivity for PIPN at the low
naphthalene/isopropyl alcohol ratio are easily explained by
the consecutive alkylation of IPN by isopropyl alcohol into
DIPN and then into PIPN, which is facilitated by the high
concentration of isopropyl alcohol in the reaction mixture.
At the same time, the initially produced α-IPN, which is
more thermodynamically preferred than β-IPN, tends to be
further alkylated into α, β- and/or α, α-DIPN isomers with
excess alkylating reagent in the reaction mixture before there
is enough time to be transferred into β-IPN through isomer-
ization. This might account for the low β, β-selectivity at the
low reactant ratio in Table 5.
References
[1] R. Brzozowski, J.C. Dobrowolski, M.H. Jamroz, W. Skupinski, J. Mol.
Catal. A 170 (2001) 95.
[2] S. Pu, Y. Tanaka, T. Inui, Sep. Technol. 6 (1996) 189.
[3] G.A. Olah, J.A. Olah, J. Am. Chem. Soc. 98 (1976) 1839.
[4] G.A. Olah US patent 4,288,646, 1979.
[5] W.M. Kutz, B.B. Corson, J. Am. Chem. Soc. 67 (1945) 1312.
[6] Teijin Co Ltd., Japan patent 74,489,49, 1974.
[7] D. Fraenkel, M. Cherniavsky, B. Ittah, M. Ievy, J. Catal. 101 (1986)
273.
[8] T. Inui, S. Pu, J. Kugai, Appl. Catal. A 146 (1996) 285.
[9] J.N. Park, J. Wang, C.W. Lee, S.E. Park, Bull. Korean Chem. Soc. 23
(2002) 1011.
[10] G. Kamalakar, M. Ramakrishna Prasad, S.J. Kulkarni, S. Narayanan,
K.V. Raghavan, Micropor. Meosopor. Mater. 38 (2000) 135.
[11] A. Katayama, M. Toba, G. Takeuchi, F. Mizukami, S. Niwa, S. Mita-
mura, J. Chem. Soc., Chem. Commun. (1991) 39.
[12] C. Song, S. Kirby, Micropor. Mater. 2 (1994) 467.
[13] S. Chu, Y. Chen, Appl. Catal. A 123 (1995) 51.
[14] S. Pu, J. Kim, M. Seno, T. Inui, Micropor. Mater. 10 (1997) 25.
[15] G. Colon, I. Ferino, E. Rombi, E. Selli, L. Forni, P. Magnoux, M. Guis-
net, Appl. Catal. A 168 (1998) 81.
4. Conclusions
[16] I. Ferino, R. Monaci, E. Rombi, V. Solinas, P. Magnoux, M. Guisnet,
Appl. Catal. A 183 (1999) 303.
In a high-pressure fixed-bed flow reactor, selective syn-
thesis of 2,6-DIPN is achieved over both unmodified and
dealuminated H-mordenite, with the selectivity of 2,6-DIPN
almost independent of the catalyst activity. By dealumina-
tion, the conversion and selectivity of 2,6-DIPN are im-
proved remarkably, with 2,6-/2,7-DIPN being around 3.
However, the mordenite catalyst shows significant deactiva-
tion even at high reaction pressures, accompanied with the
disadvantage of very low selectivity of DIPN. For H-MCM-
22 catalysts, isopropylation can only take place at its acid
sites on the external surface, leading to a stable but low ac-
tivity, and very low 2,6-/2,7-DIPN ratio.
[17] J.H. Kim, Y. Sugi, T. Matsuzaki, T. Hanaoka, Y. Kubota, X. Tu, M.
Matsumoto, Micropor. Mater. 5 (1995) 113.
[18] A.D. Schmitz, C. Song, Catal. Lett. 40 (1996) 59.
[19] R. Brzozowski, W. Tecza, Appl. Catal. A 166 (1998) 21.
[20] P. Moreau, A. Finiels, P. Geneste, J. Solofo, J. Catal. 136 (1992) 487.
[21] Z. Liu, P. Moreau, F. Fajula, Chem. Commun. (1996) 2653.
[22] K. Smith, S.D. Roberts, Catal. Today 60 (2000) 227.
[23] G. Kamalakar, M. Ramakrishna Prasad, S.J. Kulkarni, K.V. Raghavan,
Micropor. Meosopor. Mater. 52 (2002) 151.
[24] P. Moreau, A. Finiels, P. Geneste, F. Moreau, J. Solofo, J. Org.
Chem. 57 (1992) 5040.
[25] D. Mravec, M. Michvocik, M. Hronec, P. Moreau, A. Finiels, P. Gen-
este, Catal. Lett. 38 (1996) 267.
[26] P. Moreau, A. Finiels, P. Geneste, F. Moreau, J. Solofo, Stud. Surf. Sci.
Catal. 83 (1993) 575.
[27] Y. Sugi, M. Toba, Catal. Today 19 (1994) 319.
Although no shape selectivity for 2,6-DIPN is found over
USY catalysts, very high and stable conversion (> 90%)

272
J. Wang et al. / Journal of Catalysis 220 (2003) 265–272
[28] Y. Sugi, J.H. Kim, T. Matsuzaki, T. Hanaoka, Y. Kubota, X. Tu, M.
Matsumoto, Stud. Surf. Sci. Catal. 84 (1994) 1837.
[29] 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.
[36] C. Song, X. Ma, A.D. Schmitz, H.H. Schobert, Appl. Catal. A 182
(1999) 175.
[37] M.A. Camblor, C. Corell, A. Corma, M.J. Diaz-Cabanas, S. Ni-
colopoulos, J.M. Gonzalez-Calbet, M. Vallet-Regi, Chem. Mater. 8
(1996) 2415.
[30] A.D. Schmitz, C. Song, Catal. Today 31 (1996) 19.
[31] G. Kamalakar, S.J. Kulkarni, K.V. Raghavan, S. Unnikrishnan, A.B.
Halgeri, J. Mol. Catal. A 149 (1999) 283.
[32] P. Moreau, C. He, Z. Liu, F. Fajula, J. Mol. Catal. A 168 (2001) 105.
[33] A. Corma, C. Corell, J. Perez-Pariente, Zeolites 15 (1995) 2.
[34] G. Tasi, I. Palinko, F. Mizukami, React. Kinet. Catal. Lett. 74 (2001)
317.
[38] D. Meloni, S. Laforge, D. Martin, M. Guisnet, E. Rombi, V. Solinas,
Appl. Catal. A 215 (2001) 55.
[39] M.A. Makarova, J. Dwyer, J. Phys. Chem. 97 (1993) 6337.
[40] A. Aroux, Y. Ben Taarit, Thermochim. Acta 122 (1987) 63.
[41] R. Brzozowski, W. Skupinski, J. Catal. 210 (2002) 313.
[42] W. Liang, Y. Jin, Z. Yu, Z. Wang, B. Han, Zeolites 17 (1996) 297.
[43] X.D. Yuan, J.N. Park, J. Wang, C.W. Lee, S.E. Park, Korean J. Chem.
Eng. 19 (2002) 697.
[35] J.A. Horsley, J.D. Fellmann, E.G. Derouane, M.C. Freeman, J. Catal.
147 (1994) 231.