Selective skeletal isomerization of 1-butene over FER-type zeolites derived from PLS-3 lamellar precursors

Article abstract of DOI:10.1016/j.apcata.2013.01.024

Layered aluminosilicates of PLS-3 were hydrothermally synthesized and employed as the catalysts for the skeletal isomerization of 1-butene. PLS-3 lamellar precursors were successfully prepared at different Si/Al ratios (40-∞) using layered silicate H-kanemite as silica source and tetramethylammonium hydroxide as structure directing agent. A direct calcination caused an interlayer dehydration condensation, converting the precursors into 3-dimensional FER structure with an intersecting micropore system of 8 × 10-membered rings. The physicochemical properties of the Al-PLS-3 materials thus prepared were characterized by various techniques. Their catalytic properties in the skeletal isomerization of 1-butene has been investigated and compared with conventional ferrierite. Al-PLS-3 synthesized at Si/Al = 50 exhibited the same catalytic reactivity as commercial ferrierite. However, possessing smaller crystal size, Al-PLS-3 possessed a much longer catalytic duration against coke formation.

Full text of DOI:10.1016/j.apcata.2013.01.024

Applied Catalysis A: General 455 (2013) 107–113
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Selective skeletal isomerization of 1-butene over FER-type zeolites
derived from PLS-3 lamellar precursors
Boting Yang, Jin-gang Jiang, Hao Xu, Yuming Liu, Honggen Peng, Peng Wu∗
Shanghai Key Lab of Green Chemistry and Chemical Processes, Department of Chemistry, East China Normal University, North Zhongshan Rd. 3663,
Shanghai 200062, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Layered aluminosilicates of PLS-3 were hydrothermally synthesized and employed as the catalysts for
the skeletal isomerization of 1-butene. PLS-3 lamellar precursors were successfully prepared at differ-
ent Si/Al ratios (40–∞) using layered silicate H-kanemite as silica source and tetramethylammonium
hydroxide as structure directing agent. A direct calcination caused an interlayer dehydration condensa-
tion, converting the precursors into 3-dimensional FER structure with an intersecting micropore system
of 8 × 10-membered rings. The physicochemical properties of the Al-PLS-3 materials thus prepared were
characterized by various techniques. Their catalytic properties in the skeletal isomerization of 1-butene
has been investigated and compared with conventional ferrierite. Al-PLS-3 synthesized at Si/Al = 50 exhib-
ited the same catalytic reactivity as commercial ferrierite. However, possessing smaller crystal size,
Al-PLS-3 possessed a much longer catalytic duration against coke formation.
Received 20 October 2012
Received in revised form 22 January 2013
Accepted 23 January 2013
Available online 7 February 2013
Keywords:
Lamellar zeolites
PLS-3
FER
Hydrothermal synthesis
1-Butene isomerization
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
density by increasing the Si/Al ratio [9,12,15–18]. By decreasing
the Al content, undesirable reactions like dimerization and crack-
Skeletal isomerization of 1-butene into isobutene has long
been of industrial and academic research interests as isobutene
is widely used in the reaction with methanol to produce methyl
tert-butyl ether (MTBE), which is an important additive of refor-
mulated gasoline for enhancing the octane number [1–3]. Many
researches have been carried out to investigate the catalytic per-
formance of 1-butene isomerization over medium pore zeolites
with shape-selective properties [4–12]. It seems that the proton
form ferrierite zeolite with the FER topology, which contains a two-
dimensional pore system consisting of 10-membered ring (MR)
channels intersected by 8-MR channels, is much more effective
than other zeolites in terms of selective formation of isobutene.
It is capable of giving an isobutene yield close to thermodynamic
equilibrium value.
Many researches have dealt with the reason for the outstanding
catalytic performance of FER in 1-butene isomerization. Although
the intersected channels of 10-MR ↔ 8-MR endow the FER zeolite
with unique abilities for skeletal isomerization, the isobutene selec-
tivity and catalytic stability are not always guaranteed good unless
the carbonaceous deposits are formed [13–15]. To explore the cat-
alytic processes with enhanced selectivity, the acidic properties of
less selective materials were modified, e.g. reducing the acid site
ing are suppressed effectively under lower acid site density and
comparable weak acid conditions, and then better initial selectiv-
ity and longer catalyst lifetime are achieved. Furthermore, there is
also report giving the viewpoint that smaller crystals are benefit
for isobutene selectivity, for they can afford more 10-MR entrance
mouths per unit weight [19]. Thus, direct synthesis of high silica FER
zeolites simultaneously with nanosized particles if possible is of
practical significance in terms of exploring more effective processes
for 1-butene isomerization.
The FER zeolites are usually prepared in an inorganic system
with the coexistence of alkali cation such as sodium and potassium
[20]. This conventional synthesis generally requires relatively high
crystallization temperatures. Meanwhile, it is difficult to exchange
the potassium ions in the zeolites synthesized from this system,
which directly restricts their applications as solid-acid catalysts and
selective adsorbents. Recently, by using seeding technique, Xiao
et al. have succeeded in the organotemplate-free synthesis of FER
using CDO-structure zeolite building units that are obtained from
the calcined layered silicate RUB-36 [21]. On the other hand, the
hydrothermal synthesis of FER zeolites using organic compounds as
the structure-directing agents (SDAs) has a long history dating back
to 1977. Up to date, the organic SDAs employed for the synthesis
of FER are over 30 kinds, most of which are nitrogen-containing
ammoniums or amines [22–24].
Apart from the traditional ferrierite zeolites that already
possess the 3-dimensional (3D) FER topology, the FER lamellar
∗
Corresponding author. Tel.: +86 21 6223 2292; fax: +86 21 6223 2292.
E-mail address: pwu@chem.ecnu.edu.cn (P. Wu).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.01.024

108
B. Yang et al. / Applied Catalysis A: General 455 (2013) 107–113
˚
precursors have also been synthesized. Marler et al. reported
that a PREFER aluminosilicate of 2D layered structure could be
synthesized in fluoride media in the presence of 4-amino-2,2,6,6-
tetramethylpiperidine as SDA [25]. This precursor is composed
of the layered stacking of the FER sheets that are intercalated
by the SDA species. Upon calcination, interlayer dehydroxylation
and condensation occur to cause a structural transformation to
3D FER topology. In the process, the layer structure-related [2 0 0]
and [4 0 0] diffractions are shifted to higher angles because of a
decrease in interlayer spacing. PREFER possesses a plate-like mor-
phology with a relatively large crystal size of ca. 20 ␮m. Lately, a
dual templating strategy using two SDAs with different molecular
dimensions has also been developed to synthesize the FER-type
layered materials which have similar structure to PREFER but a
different interlayer distance [26–28]. Nevertheless, these lamellar
precursors have a relatively large crystal size that is unsuitable for
1-butene isomerization.
Recently, a new lamellar precursor with the structure similar to
PREFER, i.e. PLS-3, has been reported [29]. It is worth mentioning
that the crystal size of PLS-3 is of nanometer scale (50–150 nm).
Al-containing material, Al-PLS-3, was also patented [30], but the
detailed synthesis conditions for the crystallization at different
Si/Al ratio still need further investigation, and its catalytic prop-
erties are not yet mentioned.
(ꢀ = 1.5405 A). Scanning electron microscopy (SEM) was performed
on a Hitachi S-4800 microscope. Al content was determined by
inductively coupled plasma emission spectrometry (ICP) on a
Thermo IRIS Intrepid II XSP atomic emission spectrometer. The IR
spectra were collected on Nicolet Nexus 670 FT-IR spectrometer
in absorbance mode at a spectral resolution of 4 cm−1. The sample
was pressed into a self-supported wafer with 4.8 mg cm−2 thick-
ness, which was set in a quartz cell sealed with CaF2 windows and
connected to a vacuum system. After evacuated at 723 K for 2 h,
pyridine adsorption was carried out by exposing the pretreated
wafer to a pyridine vapor at 298 K for 0.5 h. The adsorbed pyridine
was evacuated successively at different temperatures (423–723 K)
for 1 h. The spectra were collected at room temperature. 27Al solid-
state MAS NMR spectra were recorded on a VARIAN VNMRS-400WB
spectrometer under one pulse condition. The spectra were recorded
at a frequency of 104.18 MHz, a spinning rate of 10.0 kHz, and a
recycling delay of 4 s. KAl(SO4)2·12H2O was used as the reference
for chemical shift. Nitrogen gas adsorption measurements were
carried out at 77 K on a BEL-MAX gas/vapor adsorption instrument.
The samples were evacuated at 573 K for at least 6 h before adsorp-
tion.
2.3. Skeletal isomerization of 1-butene
In this study, we have optimized the synthesis conditions for
Al-PLS-3 in particular the crystallization process depending on the
Si/Al ratio. The physicochemical properties of the products were
well characterized with various techniques. Their catalytic per-
formances were further investigated in the skeletal isomerization
of 1-butene by comparing to commercial ferrierite. Characteristics
of small crystal size and high Si/Al ratio, Al-PLS-3 proved to be a
more stable isomerization catalyst with a prolonged lifetime.
The isomerization of 1-butene was carried out on a continuous
flow microreactor under atmosphere pressure. In a typical run, 0.2 g
catalyst in proton form was put into a quartz tube with inner diame-
ter of 11 mm, where it was activated at 673 K for 1 h in nitrogen flow.
Then, the mixed gas of 1-butene (99.9%) and nitrogen were fed into
the reactor to start the isomerization at 673 K. The partial pressures
of 1-butene and nitrogen were controlled with mass flowmeter to
be 0.1 and 0.9 bar, respectively. The weight hourly space velocity
(WHSV) was 5.2 h−1 with respect to 1-butene. Reaction products
were analyzed on-line by a gas chromatograph equipped with an
Al2O3/S column (50 m × 0.53 mm × 0.25 ␮m). Since there is a ther-
modynamic equilibrium for the butene isomers at isomerization
temperature even in the absence of any catalyst, the linear butenes
were thus considered as the reactant when calculating the conver-
sion and product distribution.
2. Experimental
2.1. Material synthesis
The lamellar precursors of aluminosilicate Al-PLS-3 were syn-
thesized through a solid-state conversion of protonated kanemite
using tetraethylammonium hydroxide as the SDA according to
literature procedures [29,30]. In a typical synthesis, H-kanemite,
Al(NO3)3·18H2O, NaOH and distilled water were dissolved in
TEAOH (25 wt.% aqueous solution) under magnetic stirring. The
molar composition of the starting mixture was 1.0 SiO2:1/x
Al2O3:0.2 TEA+:y NaOH:6.5 H2O, where x and y were varied in
the range of 20–∞ and 0–0.36, respectively. The synthetic gel was
charged into a stainless autoclave equipped with a Teflon liner and
heated at 443 K for 4–24 h under static conditions. The solid product
was collected by filtration, washed with distilled water, and dried
overnight in an oven at 353 K. The material with the PLS-3 lamel-
lar structure was calcined in air at 823 K for 10 h to remove the
organic species occluded, giving rise to 3D FER structure. The cal-
cined Al-PLS-3 samples were then converted into ammonium form
by stirring in 1 M NH4Cl solution at a solid-to-liquid ratio of 1:50 for
2 h. The ion-exchange was repeated for three times. The samples
were converted into proton form by calcination at 773 K for 5 h.
For control experiment, a commercial FER sample (Si/Al = 8.4;
Na, K form) was provided by Tosoh, and Al-Beta was synthesized
in the above-mentioned system at a higher Na/Si ratio of 0.24.
They were similarly converted into proton form by ammonium
ion-exchange and calcination.
3. Results and discussion
3.1. Hydrothermal synthesis of PLS-3 precursors
Fig. 1 shows the XRD patterns of the samples that were synthe-
sized from the kanemite/TEA+ system at a fixed Si/Al ratio of 50
while changing the Na/Si ratio. According to the literature [29], the
PLS-3 phase is easily crystallized for a pure silica system. However,
the addition of heteroatom Al into the gels made the nucleation and
crystallization complicated. The results indicated that the sodium
ion content was an important parameter determining the phase
purity. The product phase emerged depended greatly on the gel
Na/Si ratio. Increasing the Na/Si ratio from 0 to 0.36, amorphous
silica, Al-PLS-3, Beta and MOR phases were obtained, respectively
(Fig. 1a–d). This implied that this synthetic system was less stable
and controllable in terms of phase discrimination. This outcome
differed slightly from the patent that claimed the pure Al-PLS-3
materials were crystallized readily not strictly related to the Na/Si
ratio [30]. Different from rigid 3D zeolite structure, Al-PLS-3 pos-
sessed a modifiable and changeable layered structure. The layered
structure-related [2 0 0] diffraction shifted from 7.5◦ to 9.4◦ after
the calcination at 823 K for 6 h (Fig. 1b and bꢀ). This is simply a result
of removal of occluded organic species accompanied by condensa-
tion of hydroxyl groups between the layers, inducing a topotactic
transformation from precursor to 3D FER topology as shown in
2.2. Characterization methods
The X-ray powder diffraction (XRD) patterns were measured
on a Rigaku Ultima IV X-ray diffractometer using Cu-K␣ radiation

B. Yang et al. / Applied Catalysis A: General 455 (2013) 107–113
109
1.5
d'
d
3610
3600
1.0
c'
c
d
c
b
0.5
200
200
b'
a
b
0.0
3800
3700
3600
3500
3400
a
Wavenumber (cm-1)
5
10
15
20
25
30
35
2Theta /degree
Fig. 3. IR spectra in the hydroxyl stretching vibration region of (a) Al-PLS-3-50-5h;
(b) Al-PLS-3-50; (c) Al-PLS-3-50-12h; and (d) Al-Beta-50.
Fig. 1. XRD patterns of the samples as-synthesized at Na/Si ratio of (a) 0; (b) 0.04;
(c) 0.24; and (d) 0.36 as well as corresponding calcined ones (bꢀ–dꢀ). Other crystal-
lization conditions: gel molar composition, 1.0SiO2: 0.01Al2O3: 0.2TEA+: 6.5H2O;
temperature, 443 K; time, 5 h.
crystallinity of the FER phase reached 85%, giving rise to a sample
denoted as Al-PLS-3-50-5h (Fig. S4). When the crystallization time
was prolonged to 6 h, the FER phase was fully crystallized, reach-
ing a maximum crystallinity. Meanwhile, a small amount of Beta
phase interfused in the product as evidenced by the XRD reflec-
tion around 7.5◦. The crystallinity of Beta increased and that of FER
decreased with further prolonging the crystallization time to 10 h.
It seems that the Beta zeolite was formed at consumption of a part
of the FER zeolite. Nevertheless, this kind of crystal transformation
occurred to a limited extension of ca. 20%, as the system became
almost steady during the crystallization for 10–20 h. Usually, the
phase transformation in hydrothermal synthesis takes place from
a metastable zeolite to a stable phase completely. However, the
present system was capable of giving a final product dominated by
FER topology at a relatively high Si/Al ratio of 50 (Figs. S3 and S4).
After optimizing the crystallization conditions, we synthesized
the FER samples with different Al contents, Si/Al ratio of 40–100
(Fig. S5). To increase phase purity and crystallinity, they were crys-
tallized by changing the heating time in the range of 5–24 h. It
turned out that well crystallized FER zeolites were obtained at
Si/Al ratio of 50–100, whereas a fraction of Beta phase co-existed
at lower Si/Al ratios. Taking Al-PLS-3-40 as an example, it became
more inclined to form the Beta phase, resulting in a FER zeolite with
lower crystallinity (Fig. S5a). At Si/Al ratio of 100, the product was
of pure FER phase even after the crystallization was carried out for
a long time of 24 h. This result indicated again that a low Al content
was suitable for synthesizing pure Al-PLS-3.
Fig. 2. The crystallization rate in this synthetic system was rela-
tively rapid. Generally, highly crystalline materials were obtained
within 5 h.
Fig. S1 shows the XRD patterns of the samples synthesized at
different Si/Al ratios (20–∞) at 443 K for 5 h before and after calci-
nation. The Na/Si ratio was fixed at 0.04, an appropriate value for
crystallizing a pure Al-PLS-3 phase as indicated above. The prod-
uct obtained at Si/Al ratio of 20 exhibited the reflections due to
pure phase Beta topology (Fig. S1a). Increasing the Si/Al ratio to
40, the reflections due to the FER phase emerged together with the
Beta topology (Fig. S1b and c). While the Si/Al ratio was increased
over 50, pure FER zeolites were crystallized free of Beta phase
(Fig. S1d–g). Based upon a systematical investigation into the syn-
thesis parameters, a schematic diagram for the crystalline products
from the H-kanemite/TEAOH system was achieved (Fig. S2). Since
TEAOH is a typical SDA for synthesizing Beta zeolites [31], it is
reasonable that the crystallization conditions should be carefully
optimized in order to obtain pure FER zeolites. We found that this
purpose could be realized by adjusting the gel Si/Al ratio over 50.
To further explore the synthesis process for Al-PLS-3, we inves-
tigated the time course of crystallization at fixed Si/Al ratio of 50. At
initial stage, only amorphous products were obtained, whereas the
FER phase appeared at 3.5 h (Fig. S3). Fig. S4 shows the change of rel-
ative crystallinity of Beta to FER with the crystallization time. Fully
crystallized Al-PLS-3-50-6h with a pure FER phase was assumed
to be of 100% crystallinity. The relative crystallinity of the samples
synthesized for a different period time was then calculated by com-
paring their sum intensity of XRD diffractions of the FER or Beta
phase to that of Al-PLS-3-50-6h. After crystallization for 5 h, the
3.2. Physicochemical properties of Al-PLS-3 zeolites
IR and 27Al NMR spectroscopies were applied to study the inser-
tion procedure of Al ions into the framework. Fig. 3 shows the
Fig. 2. Schemiatic illustration for topotactic transformation of Al-PLS-3 lamellar precursor to 3D FER-type zeolite through calcination.

110
B. Yang et al. / Applied Catalysis A: General 455 (2013) 107–113
Table 1
Physicochemical properties and catalytic performances in 1-butene isomerization of various FER zeolites.
a
b
No.
Sample
Si/Alc
SBET (m2 g−1
)
Vtotal (cm3 g−1
)
Conv.d (%)
Sel.d (%)
C2=
C3
C3=
n-C4
i-C=4
C5+
1
2
3
4
5
6
7
Si-PLS-3
∞
97
74
47
39
45
8.3
379
476
477
441
306
354
401
0.27
0.28
0.29
0.29
0.23
0.30
0.29
–
–
–
0.03
0.0
–
–
–
–
Al-PLS-3-100
Al-PLS-3-80
Al-PLS-3-50
Al-PLS-3-40
Al-PLS-3-50-5h
FER-Tosoh
19.8
22.2
40.1
23.0
20.6
8.9
0.5
0.4
0.5
0.3
0.4
0.7
3.1
2.0
2.6
3.5
3.0
9.8
1.7
1.8
1.6
1.8
2.0
1.5
87.0
88.6
91.3
90.4
89.8
77.8
5.2
5.6
2.6
3.0
2.7
9.0
0.0
0.02
0.02
0.02
a
Specific surface area given by N2 adsorption at 77 K.
Calculated from the adsorption capacity at P/P0 = 0.9.
Determined by ICP.
b
c
d
1-Butene isomerization: WHSV, 5.2 h−1; temperature, 673 K; TOS, 360 min.
IR spectra in the hydroxyl stretching vibration region for the Al-
PLS-3-50-5h, Al-PLS-3-50 and Al-PLS-3-50-12h samples that were
synthesized by the crystallization for 5 h, 6 h and 12 h along with Al-
Beta-50. Sample Al-PLS-3-50-5h showed a main band at 3745 cm−1
(Fig. 3a), which is attributed to terminal isolated silanols. How-
ever, it exhibited no obvious in the region of structural Si(OH)Al
groups. On the other hand, Al-PLS-3-50, the sample crystallized
for one hour longer than Al-PLS-3-50-5h, exhibited a new band at
3600 cm−1 (Fig. 3b), assigned to the structural Si(OH)Al groups in
the framework of FER zeolites [32]. This implies that more tetrahe-
drally coordinated framework Al ions were achieved by prolonging
crystallization time. These two samples had very similar Si/Al ratios
as evidenced by ICP analysis (Table 1, Nos. 4 and 6). It is deduced
that the FER structure construction did not synchronize with the
insertion of Al ions into the framework. The former took place
more rapidly, whereas the latter had a delay during crystalliza-
tion. The Al ions probably adhered to the crystallites first as extra
framework species, and then they were gradually inserted into the
framework. 27Al NMR spectra provided solid evidences to this pos-
tulation. The spectrum of Al-PLS-3-50-5h showed two resonances
at 0 ppm and 54 ppm, representing the tetrahedral Al and octahe-
dral extra framework Al, respectively [33]. In particular, the 54 ppm
resonance was relatively broad, implying the Al species were of
less symmetric (Fig. 4a). However, the Al-PLS-3-50 sample showed
a single resonance at 54 ppm (Fig. 4b). This resonance was much
sharper and more symmetric, suggesting the Al ions occupied the
framework position to have a better coordination state.
Beta zeolite, Al-Beta-50, synthesized in this system but at a higher
Na/Si ratio of 0.24. Al-Beta-50 showed the stretching vibration for
the structural Si(OH)Al groups centered at 3610 cm−1 (Fig. 3d). The
position is fully in agreement with literature [34]. Thus, the struc-
ture difference leads to 10 cm−1 difference in structural Si(OH)Al
groups between BEA and FER. The 3600 cm−1 band is considered to
be contributed mainly by the tetrahedral Al in Al-PLS-3-50 but not
by Beta phase. The spectrum of Al-PLS-3-50-12h was also given for
comparison. As a result of longer crystallization, it contained more
Beta phase than Al-PLS-3-50. It then showed peak at 3607 cm−1
between 3600 cm−1 and 3610 cm−1 (Fig. 3c), indicating that struc-
tural Si(OH)Al groups of this material were generated from both the
FER and Beta phases. As a result, it is necessary to control precisely
the crystallization process of Al-PLS-3 in order to achieve the FER
solid acid catalysts not only with a high crystallinity and purity but
also containing the framework Al showing Brønsted acidity.
Fig. 5 shows the IR spectra in the hydroxyl stretching vibration
region for the Al-PLS-3 samples synthesized at different Si/Al ratios.
Silicalite Si-PLS-3 reasonably showed on the terminal silanol band
at 3745 cm−1 but no peak at 3600 cm−1 (Fig. 5a). The 3600 cm−1
band due to the framework Al increased in intensity with increas-
ing Al content except for Al-PLS-3-40 with a lower crystallinity
(Fig. 5b–e). The amount of Brønsted acid sites then would increase
with increasing Al content. The 27Al NMR spectra confirmed that
these samples were almost free of the extra framework Al species,
but contained predominately the tetrahedral ones independent of
the Si/Al ratios (Fig. S6).
One may suspect the occurrence of the 3600 cm−1 band is sim-
ply due to the Beta phase contaminated in the FER product. To
rule out this possibility, we compared the spectra with that pure
The catalytic properties of aluminosilicates are closely related
to their Brønsted and Lewis acidity. Pyridine was then employed
as a probe molecule to provide detailed information about the
amount and strength of Brønsted and Lewis acid sites in Al-PLS-3
1.2
1.0
0.8
3600
0.6
e
b
a
d
c
b
a
0.4
0.2
0.0
3800
3600
3400
Wavenumber (cm-1)
100
50
0
-50
Chemical Shift (ppm)
Fig. 5. IR spectra in the hydroxyl stretching vibration region of (a) Si-PLS-3; (b)
Al-PLS-3-100; (c) Al-PLS-3-80; (d) Al-PLS-3-50; and (e) Al-PLS-3-40.
Fig. 4. 27Al MAS NMR spectra of (a) Al-PLS-3-50-5h and (b) Al-PLS-3-50.

B. Yang et al. / Applied Catalysis A: General 455 (2013) 107–113
111
materials did not depend on the Si/Al ratios (40–∞) and the calci-
nation or not. They were all rod-like materials with a similar crystal
size.
1490
1440
1455
1.0
0.5
0.0
The textural properties of Al-PLS-3 samples with different Si/Al
ratios are given in Table 1 together with other physicochemi-
cal properties. The specific surface area was measured by the
Brunauer–Emmett–Teller (BET) method, whereas the total volume
was obtained from the isotherms at P/P0 ≤ 0.9 because a remarkable
increase in the adsorbed volume was observed around P/P0 > 0.9.
This phenomenon suggested the presence of macropores due to
interparticle condensation and adsorption of nitrogen for small
crystal Al-PLS-3. This finding was similar to the adsorption prop-
erties of Si-PLS-3 reported previously [29]. The SBET and Vtotal of
Al-PLS-3-40 and Al-PLS-3-50-5h were smaller than other samples,
which was mainly because of a lower crystallinity as evidenced by
XRD and IR investigations. The bulk Si/Al ratios of Al-PLS-3 given by
ICP were close to the gel compositions, indicating that the conden-
sation of silicon source and aluminum source took place uniformly
during crystallization. The adsorption data of FER-Tosoh were also
measured for comparison (Table 1, No. 7). Its SBET and Vtotal were at
the same level as Al-PLS-3.
1599
1540
a
b
c
d
1650 1600 1550 1500 1450 1400
Wavenumber (cm-1)
Fig. 6. IR spectra of Al-PLS-3-50 after pyridine adsorption and desorption at (a)
423 K; (b) 523 K; (c) 623 K; and (d) 723 K.
samples. The IR spectra of adsorbed pyridine in the range of
pyridine ring-stretching modes were measured after desorption
at various temperatures. As shown in Fig. 6, the bands at 1599
and 1440 cm−1 associated with hydrogen-bonded and physically
adsorbed pyridine decreased in intensity with a rising desorption
temperature [35–37]. On the other hand, the bands at 1540 cm−1
corresponding to Brønsted acid sites and at 1455 cm−1 correspond-
ing to Lewis acid sites together with the bands at 1490 cm−1
characteristic of both Brønsted and Lewis acid sites were more
evacuation temperature resistant and remained even after desorp-
tion at 723 K. These bands are apparently correlated to the different
vibration modes of the pyridine-rings adsorbed on the Al species
and can be taken as the evidence for the presence of Brønsted and
Lewis acid sites in Al-PLS-3-50.
The pyridine adsorption was further carried out on the Al-PLS-
3 samples with various Al contents. Fig. S7 compares the spectra
of the samples after removing weakly adsorbed pyridine by evac-
uation at 423 K. The bands at 1540, 1490, and 1455 cm−1 were
absent for Si-PLS-3, but increased in intensity with increasing Al
content except for Al-PLS-3-40 with a lower crystallinity. Thus indi-
cated that with the introduction of tetrahedrally coordinated Al in
Al-PLS-3 the amount of Brønsted and Lewis acid sites increased
gradually.
Fig. 7 shows the representative scanning electron micrograph
(SEM) of Al-PLS-3 using Al-PLS-3-50 as an example. The morphol-
ogy of Al-PLS-3-50 was rod-like with a very small crystal length of
50–150 nm (Fig. 7a), which was similar to Si-PLS-3 reported in liter-
ature [29]. The image indicates that there was no amorphous phase
or obvious Beta crystals, confirming that it was a highly crystalline
material with high phase purity. Fig. 7b shows the SEM image of
commercial FER-Tosoh. It showed a plate-like morphology and a
larger crystal size of 0.5–1 ␮m, which was much different from Al-
PLS-3–50. As shown in Fig. S10, the morphology of the Al-PLS-3
3.3. Catalytic properties Al-PLS-3 in 1-butene isomerization
The catalytic properties of these Al-PLS-3 samples were inves-
tigated in the 1-butene skeletal isomerization reaction. Catalytic
tests of various Al-PLS-3 catalysts revealed a complex time on
stream (TOS) diagram during the isomerization at atmospheric
pressure, 673 K and a weight hourly space velocity (WHSV) of
5.2 h−1 (Fig. S8). The catalytic activity and product distribution of
Al-PLS-3 at TOS = 360 min are compared in Table 1, including the
results obtained with FER-Tosoh. At the beginning of the reaction,
TOS = 30 min, all the samples showed high conversions of ca. 70%,
but very low selectivities for desirable product isobutene. The main
products on this stage were propylene and pentene which occu-
pied over 50% yield. This phenomenon was similar to the results
reported before, which is assumed to be caused by the bimolecular
mechanism on fresh FER zeolite [14,18,38–43]. At initial stage, the
1-butene molecules experienced dimerization and isomerization to
form octenes, and then the cracking of octenes to the final products.
A decrease in the activity and a significant increase in selectivity for
isobutene can be observed with increasing time on stream. Many
researches supported this view point that a tremendous increase
of isobutene selectivity was caused by the monomolecular reac-
tion occurred on isolated Brønsted acid sites of aged FER zeolites
[40,41,44–46]. The coke deposition in the partially deactive FER
restricted the steric space available for bulky intermediates around
the acid sites and reduced the density of acid sites, afterwards
suppressed the bimolecular processes then made the monomolec-
ular mechanism to be the prevailing one [47]. The conversions
decreased quickly in the first 2 h, and simultaneously the selectivity
to isobutene rose from about 20% up to 85% for all these samples.
Fig. 7. SEM images of (a) Al-PLS-3-50 and (b) FER-Tosoh after calcination.

112
B. Yang et al. / Applied Catalysis A: General 455 (2013) 107–113
50
100
80
60
40
20
0
a
Equilibrium
isobutene sel.
40
30
20
10
0
Al-PLS-3-50
conv.
FER-Tosoh
isobutene yield
0
100 200 300 400 500 600
TOS (min)
0
20
40
60
80
100
1-Butene conv. (%)
100
80
60
40
20
0
b
Fig. 9. Isobutene yield as function of 1-butene conversion obtained for Al-PLS-3-50
isobutene sel.
and FER-Tosoh at 673 K.
Thus, the textural properties were measured for Al-PLS-3-50 and
FER-Tosoh after being used in the 1-butene isomerization for 10 h
(Table S1). In comparison to the fresh catalysts, the used ones both
decreased greatly their surface area and pore volume. Nevertheless,
Al-PLS-3-50 still possessed a surface area of 170 m2 g−1 and pore
volume of 0.14 cm3 g−1after catalytic run, while the used FER-Tosoh
showed a surface area of only 38 m2 g−1. These results were in
accordance with above-mentioned deduction. Furthermore, these
catalytic behaviors were similar to the literature results on high
Si/Al ratio FER reported by Martínez et al. [47], suggesting that a
monomolecular mechanism was probably involved in the isomer-
ization.
conv.
isobutene yield
0
100 200 300 400 500 600
TOS (min)
Fig. 8. The time course for the skeletal isomerization of 1-butene to isobutene over
(a) Al-PLS-3-50 and (b) FER-Tosoh. Reaction conditions: WHSV, 5.2 h−1; tempera-
ture, 673 K; N2/1-butene molar ratio, 9:1.
4. Conclusions
Layered aluminosilicates Al-PLS-3 with different Al contents are
prepared by hydrothermal synthesis at Si/Al ratios of 40–100. The
calcination caused a structural transformation to 3D FER zeolites.
The amount of sodium ions is confirmed to be an important factor
to govern the crystallization of pure Al-PLS-3 phase. High sodium
content in gels easily leads to the formation Beta and MOR zeo-
lites. The insertion of Al ions into the framework of Al-PLS-3 does
not take place in initial stage, but is achieved after the finishing
up of the FER framework construction. The Al ions adhere to Al-
PLS-3 in extra framework form firstly and then further crystallize
into the framework. During this process a fraction of Beta mixed
crystal will form, but it neither make contributions to the acidities
of these materials nor influence their activities. All the Al-PLS-3
samples give excellent catalytic performance for 1-butene isomer-
ization and among them Al-PLS-3-50 synthesized at Si/Al ratio of
50 is highly active and selective to isobutene (over 90%), giving an
isobutene yield about 35%.
Afterwards, the decrease of conversion and increase of selectivity
both slowed down. As shown in Table 1, the isobutene selectivity of
all Al-PLS-3 samples reached about 90% at TOS = 360 min excepted
for Al-PLS-3-40 with too less Brønsted acid sites. The TOS depend-
ence of 1-butenes conversion, selectivity and yield of isobutene
for Al-PLS-3-50 and FER-Tosoh are depicted in Fig. 8. On the first
stage of the reaction, FER-Tosoh gave a higher conversion for 1-
butene than Al-PLS-3-50 since it had a higher Al content and more
Al acid sites. Both catalysts were capable of giving an isobutene
selectivity over 90% with prolonging time on stream. However,
FER-Tosoh deactivated obviously more rapidly than Al-PLS-3-50.
The FER zeolites originated from Al-PLS-3 lamellar served as more
stabile catalysts for 1-butene isomerization.
Fig. 9 shows the dependence of isobutene yield on 1-butene
conversion. Al-PLS-3-50 and FER-Tosoh both exhibited a correla-
tion close to the indicated thermodynamic equilibrium. In terms of
efficient production of isobutene, the former seemed to be more
selective for isobutene. It is worth mentioning that the isobutene
selectivity increased faster with TOS for Al-PLS-3-50 than for FER-
Tosoh (Fig. 8). After the reaction run for TOS of 600 min, the coke
formed on Al-PLS-3-50 accounted for 5.3 wt.%, but reached 9.8 wt.%
on FER-Tosoh as given by TG analysis (Fig. S9). As mentioned above,
the coke deposition play a crucial role in the isobutene selectiv-
ity increasing process, as it can reduce the density of acid sites
and suppress the bimolecular reaction. Thus, a faster increase in
selectivity for Al-PLS-3-50 can be understood easily as its acid site
density was already low. And for the higher stability of Al-PLS-
3-50 than FER-Tosoh, this may be contributed by less acid site
change and micoropore volume loss in the former catalyst, because
the carbonaceous deposition accumulated in its pore was lower.
Acknowledgements
The authors gratefully acknowledge the financial supports from
the National Science Foundation of China (20925310, U1162102),
Ministry of Science and Technology (2012BAE05B02), Shanghai
Municipal Education Commission (11CXY20) and Shanghai Leading
Academic Discipline Project (B409).
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.apcata.
2013.01.024.

B. Yang et al. / Applied Catalysis A: General 455 (2013) 107–113
113
References
[24] E.W. Volyocsik, US Patent 4584286 (1986).
[25] L. Schreyeck, P. Caullet, J.C. Mougenel, J.L. Guth, B. Marler, Microporous. Mater.
6 (1996) 259–271.
[26] A.B. Pinar, L.G. Hortigüela, J.P. Pariente, Chem. Mater. 19 (2007) 5617–5626.
[27] R. García, L.G. Hortigüela, I. Díaz, E. Sastre, J.P. Pariente, Chem. Mater. 20 (2008)
1099–1107.
[28] R. García, L.G. Hortigüela, T. Blasco, J.P. Pariente, Microporous. Mesoporous.
Mater. 132 (2010) 375–383.
[29] T. Ikeda, S. Kayamori, F.J. Mizukami, J. Mater. Chem. 19 (2009) 5518–5525.
[30] T. Ikeda, S. Kayamori, T. Hanaoka, F. Mizukami, JP Patent 120629A (2008).
[31] M.A. Camblor, A. Corma, S. Valencia, J. Mater. Chem. 8 (1998) 2137–2145.
[32] M. Trombetta, G. Busca, S. Rossini, V. Piccoli, U. Cornaro, A. Guercio, et al., J.
Catal. 179 (1998) 581–596.
[33] Y.R. Leshkov, M. Moliner, M.E. Davis, J. Phys. Chem. C 115 (2011) 1096–1102.
[34] Y. Oumi, R. Mizuno, K. Azuma, S. Nawata, T. Fukushima, T. Uozumi, et al., Micro-
porous. Mesoporous. Mater. 49 (2001) 103–109.
[35] B. Buzzoni, S. Bordiga, G. Ricchiardi, C. Lamberti, A. Zecchina, G. Bellussi, Lang-
muir 12 (1996) 930–940.
[36] C. Paze, S. Bordiga, C. Lamberti, M. Salvalaggio, A. Zecchina, G. Bellussi, J. Phys.
Chem. B 101 (1997) 4740–4751.
[37] F. Bonino, A. Damin, S. Bordiga, C. Lamberti, A. Zecchina, Langmuir 19 (2003)
2155–2161.
[38] M. Guisnet, P. Andy, N.S. Gnep, C. Travers, E. Benazzi, J. Chem. Soc., Chem.
Commun. 16 (1995) 1685–1686.
[39] M. Guisnet, P. Andy, N.S. Gnep, C. Travers, E. Benazzi, Proc. Int. Zeolite Congr.,
Microporous. Mater., Stud. Surf. Sci. Catal. 105 (1996) 1365–1372.
[40] P. Meriaudeau, R. Bacaud, L. Ngoc Hung, A.T. Vu, J. Mol. Catal. A: Chem. 110
(1996) L177–L179.
[41] J. Houzˇvicˇka, V. Ponec, Ind. Eng. Chem. Res. 36 (1997) 1424–1430.
[42] W. Blanka, Z. Nada, U. Elena, C. Jirˇí, S. Priit, P. Cristine, et al., Appl. Catal. A 182
(1999) 297–308.
[43] B. de Ménorval, P. Ayrault, N.S. Gnep, M. Guisnet, J. Catal. 230 (2005) 38–51.
[44] L. Domokos, L. Lefferts, K. Seshan, J.A. Lercher, J. Catal. 197 (2001) 68–80.
[45] M. Kangas, T. Salmi, D.Y. Murzin, Ind. Eng. Chem. Res. 47 (2008) 5413–5426.
[46] W. Chularat, N. Somkiat, B. Bundet, P. Piboon, L. Jumras, J. Phys. Chem. C 116
(2012) 5654–5663.
[1] G. Pecci, T. Floris, Hydrocarbon Process 56 (1977) 98–102.
[2] J. Szabo, J. Perrotey, G. Szabo, J.C. Duchet, D. Cornet, J. Mol. Catal. A: Chem. 67
(1991) 79–90.
[3] J. Houzˇvicˇka, V. Ponec, Catal. Rev. Sci. Eng. 39 (1997) 319–344.
[4] W.Q. Xu, Y.G. Yin, S.L. Suib, C.L. Oyoung, J. Catal. 150 (1994) 34–45.
[5] P.B. GrandvaUet, K.P. de Jong, H.H. Mooiweer, A.G.T.G. Kortbeek, B. Kraushaar-
Czarnetzki, Eur. Patent 501577 (1992).
[6] H.H. Mooiweer, K.P. de Jong, B. Kraushaar-Czarnetzki, W.H.J. Stork, B.C.H.
Krutzen, Stud. Surf. Sci. Catal. 84 (1994) 2327–2334.
[7] M.W. Simon, S.L. Suib, C.L. Oyoung, J. Catal. 147 (1994) 484–493.
[8] W.Q. Xu, Y.G. Yin, S.L. Suib, C.L. Oyoung, J. Phys. Chem. 99 (1995) 758–765.
[9] W.Q. Xu, Y.G. Yin, S.L. Suib, J.C. Edwards, C.L. Oyoung, J. Phys. Chem. 99 (1995)
9443–9451.
[10] C.L. Oyoung, R.J. Pellet, D.G. Casey, J.R. Ugolini, R.A. Sawicki, J. Catal. 151 (1995)
467–469.
[11] L.H. Gielgens, I.H.E. Veenstra, V. Ponec, M.J. Haanepen, J.H.C. van Hooff, Catal.
Lett. 32 (1995) 195–203.
[12] G. Seo, H.S. Jeong, S.B. Hong, Y.S. Uh, Catal. Lett. 36 (1996) 249–253.
[13] R.J. Pellet, D.G. Casey, H.M. Huang, R.V. Kessler, E.J. Kuhlman, C.L. Oyoung, et al.,
J. Catal. 157 (1995) 423–435.
[14] M. Guisnet, P. Andy, N.S. Gnep, E. Benazzi, C. Travers, J. Catal. 158 (1996)
551–560.
[15] P. Andy, N.S. Gnep, M. Guisnet, E. Benazzi, C. Travers, J. Catal. 173 (1998)
322–332.
[16] M.A. Asensi, A. Corma, A. Martínez, J. Catal. 158 (1996) 561–569.
[17] W.Q. Xu, Y.G. Yin, S.L. Suib, J.C. Edwards, Ch.L. Oyoung, J. Catal. 163 (1996)
232–244.
[18] M. Guisnet, P. Andy, N.S. Gnep, C. Travers, E. Benazzi, Ind. Eng. Chem. Res. 37
(1998) 300–302.
[19] S.H. Lee, C.H. Shin, S.B. Hong, J. Catal. 223 (2004) 200–211.
[20] J. Arika, H. Miyazaki, K. Igawa, K. Itabashi, US Patent 4650654 (1987).
[21] H.Y. Zhang, Q. Guo, L.M. Ren, C.G. Yang, L.F. Zhu, X.J. Meng, et al., J. Mater. Chem.
21 (2011) 9494–9497.
[22] C.J. Plank, E.R. Rosinski, M.K. Bubin, US Patent 46016245 (1977).
[23] G.Q. Guo, Y.C. Long, Adv. Chem. Ind. (in Chinese) 6 (2000) 16–19.
[47] M.A. Asensi, A. Martínez, Appl. Catal. A 183 (1999) 155–165.