A new catalyst platform: Zeolite Beta from template-free synthesis

Article abstract of DOI:10.1039/c3cy00073g

Structural analysis and catalytic testing revealed that zeolite Beta from template-free synthesis introduces new possibilities in catalysis, as a result of its unprecedentedly high density of active sites with exceptional stability and distinctively order

Full text of DOI:10.1039/c3cy00073g

Catalysis
Science & Technology
View Article Online
View Journal
PAPER
A new catalyst platform: zeolite Beta from
template-free synthesis†
Cite this: DOI: 10.1039/c3cy00073g
Bilge Yilmaz,*a Ulrich Muller,b Mathias Feyen,b Stefan Maurer,b Haiyan Zhang,c
¨
Xiangju Meng,d Feng-Shou Xiao,d Xinhe Bao,e Weiping Zhang,f Hiroyuki Imai,g
Toshiyuki Yokoi,h Takashi Tatsumi,h Hermann Gies,i Trees De Baerdemaekerj and
Dirk De Vosj
Structural analysis and catalytic testing revealed that zeolite Beta from template-free synthesis
introduces new possibilities in catalysis, as a result of its unprecedentedly high density of active sites
with exceptional stability and distinctively ordered nature. Highly active and selective catalysts were
obtained either by using it in the Al-rich form (e.g. alkylation) or after post-synthesis treatments (e.g.
acylation). Such versatility made possible by this novel synthesis route constitutes a new toolbox for
catalysis.
Received 8th December 2012,
Accepted 30th April 2013
DOI: 10.1039/c3cy00073g
www.rsc.org/catalysis
to be applicable to some other zeolite types as well.5 Sub-
sequently, SDA-free synthesis of zeolite Beta was also explored
Introduction
by Okubo, Mintova and their teams.6 Related to these efforts,
another interesting field of study is the solvent-free synthesis of
zeolites, where there have been recent reports on a number of
structures including zeolite Beta.7
As a well-established family of nanoporous catalysts, zeolites
are of vital importance for the chemical and petrochemical
industries. Zeolite Beta, with its intriguing framework architec-
ture (*BEA) and 3-dimensional pore system, displays superior
performance in refinery applications, environmental catalysis
and a variety of organic reactions because of its vastly accessible
pore volume, high adsorption capacity, strong acid sites and
shape/size selectivity.1 Zeolite Beta for industrial utilization is
typically synthesized using tetraethylammonium hydroxide
(TEAOH) as the template or structure directing agent (SDA).2
Various higher cost organic templates were also explored.3
Recently, Xiao and co-workers reported a template-free (TF)
synthesis methodology for zeolite Beta,4 which was later shown
A vast majority of the synthetic zeolites that offer promise as
industrial catalysts require organic SDAs for their synthesis. If
they end up being commercialized, they are classified as
‘‘specialty’’ zeolites from an industrial point of view, which
translates into a prohibitive cost structure for many applica-
tions. For industrial production, SDA-free synthesis means
significant savings in raw material, energy and capital invest-
ment. It completely avoids the energy-intensive high temper-
ature calcination, thereby eliminating the formation of harmful
combustion gases. In addition, the SDA-free route allows the
utilization of simpler low-pressure vessels, with significant cost
savings. Therefore, even zeolites previously considered to be too
expensive can now be employed in petrochemical applications.
Moreover, the SDA-free synthesis has a major influence on
the product characteristics. Preliminary analyses demonstrated
that Beta zeolites from SDA-free synthesis exhibit a significantly
larger crystal size and an unprecedentedly high aluminum
content compared to the conventional templated zeolite
Beta.5,6 Here, we report detailed structural investigations and
catalytic tests using SDA-free Beta, not only in the as-synthesized
form but also after controlled modifications. As will be demon-
strated, the SDA-free Beta as such has striking activity and selectivity
characteristics, because of its significantly higher density of active
sites and uniquely ordered nature. Additionally, the material can be
a BASF Corporation, Iselin, NJ 08830, USA. E-mail: bilge.yilmaz@basf.com
b BASF SE, Ludwigshafen, Germany
c College of Chemistry, Jilin University, Changchun, China
d Department of Chemistry, Zhejiang University, Hangzhou, China
e State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Dalian,
China
f State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian,
China
g Department of Chemical and Environmental Engineering, University of Kitakyushu,
Fukuoka, Japan
h Chemical Resources Laboratory, Tokyo Institute of Technology, Yokohama, Japan
i Institute of Geology, Mineralogy and Geophysics, Ruhr-University Bochum,
Bochum, Germany
j Centre for Surface Chemistry and Catalysis, K. U. Leuven, Heverlee, Belgium
† Electronic supplementary information (ESI) available: Additional structural,
characterization and catalytic testing data. See DOI: 10.1039/c3cy00073g
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
modified in a controlled way to emulate the properties of the with a 4 M NH4NO3 solution at 100 1C for 2 hours followed by
more expensive templated Beta with a higher Si/Al ratio. Even ammonia removal via heat treatment.
though templated and SDA-free Beta share the *BEA framework,
Commercial zeolite CP811 was obtained in the H+-form from
their growth mechanisms, physicochemical characteristics and Zeolyst. Commercial zeolite CP814C (NH4+-form, Zeolyst) was
resulting structure–property relationships are markedly diverse. In converted to the H+-form by calcination at 450 1C for 5 hours.
this contribution, the superior catalytic performance of SDA-free
Characterization
Beta, in the as-synthesized and modified forms, is demonstrated
in alkylations and acylations.
High-resolution X-ray powder diffraction data for samples to be
compared were collected at room temperature with a Siemens
D5000 diffractometer operating in quasi Debye–Scherrer mode
using both Mo-Ka and Cu-Ka radiation with the sample in a
Experimental
SDA-free synthesis of zeolite Beta
capillary sample holder. 29Si MAS NMR spectra were recorded
1
The zeolitic products with the *BEA framework were synthesized at 79.5 MHz with H decoupling at a spinning rate of 6 kHz,
starting from aluminosilicate synthesis mixtures at temperatures 1000 scans, and 4 s recycle delay. The chemical shifts were
between 120–140 1C in the presence of previously calcined referenced to tetramethylsilane. 27Al 3Q MAS NMR experiments
zeolite Beta seeds. A wide variety of zeolite Beta materials were were performed on a Varian Infinityplus-400 spectrometer
explored as seeds and it was shown that zeolite Beta with varying using a three-pulse sequence incorporating a Z-filter at a
characteristics can be utilized for this purpose. In this contri- spinning speed of 12 Hz with a 4 mm probe. The spectra were
bution, characterization and catalytic testing of the products acquired at 104.2 MHz, and the chemical shifts were referenced
from SDA-free syntheses utilizing the calcined CP814C zeolite to a 1.0 wt% Al(NO3)3 aqueous solution. An rf field of 227 kHz
Beta (Zeolyst) seeds are reported. The powder XRD pattern for was used for the creation (0 Q - 3 Q) and the first conversion
the seed crystals used in this investigation is provided in Fig. S1 (+3 Q - 0 Q) pulse. An rf field of 17 kHz was used for the last
(ESI†). The role of seed crystals in SDA-free synthesis has been conversion step (0 Q - Æ1 Q), which was the central transition
discussed in detail elsewhere.5,6 Fumed silica, sodium aluminate selective soft 901 pulse. A two-dimensional (2D) Fourier trans-
(Al2O3 > 41.0%), and sodium hydroxide (NaOH > 96%) were used formation followed by a shearing transformation gave a pure
as the ingredients for the synthesis mixture. In a typical synth- absorption mode 2D contour plot. The second order quadrupolar
esis experiment the following recipe was used:
effect (SOQE) and isotropic chemical shift (diso) values were calcu-
(1) 0.07 g NaAlO2 was dissolved in 7.56 mL distilled water, lated according to the procedures in literature.
followed by addition of 0.312 g of NaOH;
Temperature programmed desorption of ammonia (NH3-
(2) After stirring for 30 minutes, 0.72 g of fumed silica was TPD) experiments were performed using an automated chemi-
added to the solution obtained in step (1);
sorption analysis unit with a thermal conductivity detector
(3) After stirring for 10 minutes, 0.036 g of zeolite Beta seeds (TCD). This setup was also equipped with an on-line mass
were introduced into the synthesis mixture, followed by stirring spectroscopy (MS) system for continuous analysis of the
for 3 minutes;
desorbed species.
(4) The resulting gel with a molar ratio of SiO2 : 0.025Al2O3 :
Infrared spectra of self-supporting wafers of the Beta zeolite
0.36Na2O : 35Á3H2O was transferred into a 15 mL Teflon-lined samples (H+-form) with a thickness of 10–20 mg cmÀ2 were
stainless-steel autoclave to crystallize at 120 1C for 120 hours in measured on a Nicolet 6700 FT-IR spectrometer. Spectra were
an oven;
taken at 150 1C after heating under vacuum to 450 1C (heating
(5) After filtration at room temperature and drying at 80 1C, rate 5 1C minÀ1) and evacuating for 1 hour.
solid products were recovered.
Bulk elemental analysis of the samples was performed using
To convert the as-synthesized samples from the Na+-form a Varian Vista-PRO inductively coupled plasma optical emission
to the H+-form, they were subjected three times to an ion- spectrometer (ICP-OES). Field emission scanning electron micro-
exchange treatment with 0.5 M NH4NO3 at 80 1C overnight scopy (FE-SEM) was performed using a JEOL JSM-7400F unit to
followed by washing with distilled water, drying and heat determine the crystal size.
treatment at 450 1C for 5 hours (heating rate 1 1C minÀ1).
Catalytic reactions
Steam-dealumination of the as-synthesized samples was
performed after ion-exchange with NH4NO3 followed by Alkylation of benzene with ethylene was performed in a Parr
repeated steam treatment. Before each steam treatment the autoclave with a capacity of 100 mL, loaded under nitrogen with
sample was treated with an aqueous NH4NO3 solution at reflux 40 mL benzene (99.7%, Sigma-Aldrich) and 62.5 mg catalyst,
followed by washing and drying. For the steam treatment, the which had been dried overnight in an oven at 150 1C. The reactor
sample was hydrated and heated in a covered vessel to 600 1C was flushed and saturated with ethylene (99.95%, Air Liquide) at
for 2 hours.
a pressure of 5 bar for 30 minutes and pressurized with 50 bar
After steam treatment, part of the dealuminated sample was nitrogen and heated to 150 1C. The stirring speed was set at
further treated with acid using a 1 M HNO3 solution at 80 1C for 1000 rpm. Samples were taken periodically and analyzed on a
3 hours followed by washing and drying at 80 1C. Another part Shimadzu 2010 GC equipped with a CP-Sil-5CB 60 m column and
of the dealuminated sample was ion-exchanged three times an FID detector and by GC/MS. The ethylbenzene or diethylbenzene
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
selectivity was determined by calculating the molar ratio of ethyl-
benzene or diethylbenzene to the sum of all reaction products
derived from ethylene.
Acylation of anisole with acetic anhydride was performed in
a round-bottom flask with a capacity of 10 mL. 50 mg of the
catalyst, which was dried overnight in an oven at 100 1C, was
added to a mixture of 50 mmol anisole (99%, Acros Organics)
and 5 mmol acetic anhydride (98%, VWR) in the flask. The flask
containing the catalyst and the reactants was sealed, and then
heated in an oil bath at 60 1C for 5 hours under stirring with a
magnetic stirrer. Analysis of the reaction products was per-
formed using a Shimadzu 2014GC equipped with a DB-5 50 m
column and an FID detector.
Results and discussion
Fig. 1 XRD patterns (Mo-Ka radiation) for zeolite Beta from templated synthesis
Morphological analysis of zeolite Beta products from SDA-free (a, CP814), SDA-free synthesis (b) and a simulation assuming 44% polymorph
A content (c).
synthesis shows significantly larger crystal size in the 200–
700 nm range, and correspondingly more prominent textural
features compared to the templated zeolite Beta, which typi-
significantly improved crystallinity. The higher degree of order for
cally displays particle sizes between 20 and 100 nm.5 SDA-free
TF-Beta might also explain the unusually high hydrothermal
synthesis also influences the overall porosity of zeolite Beta: the
stability of this material as previously observed.5 Typically, zeolites
as-synthesized product has significantly higher micropore
with such high aluminum content are much less stable compared
surface area and larger median pore size with respect to the
to the high-silica zeolites.
templated zeolite Beta after calcination. The considerably lower
In order to confirm the higher degree of structural order in
Si/Al ratio (4.5–5) and the higher (hydro)thermal stability of
TF-Beta, quantitative 29Si MAS NMR measurements were also
zeolite Beta from SDA-free synthesis are other notable devia-
performed and the results for products from conventional and
tions from templated zeolite Beta.4,5 The underlying reasons for
SDA-free syntheses were compared. After deconvolution and
these apparent differences for zeolitic materials with the same
integration of the spectra (Fig. 2), the full width at half max-
framework topology must be traced back to structural differ-
imum (FWHM) in Hz of various Si environments and the Si/Al
ences as a result of their distinct formation mechanisms and
ratios in the samples were obtained. The results are listed in
building blocks. In order to develop an understanding of this
Table S1 in the ESI.† As illustrated in Fig. 2 and Table S1 (ESI†),
phenomenon, and to potentially use these findings to tailor-
the FWHM of each of the Si species as determined from 29Si
make catalysts for specific reactions, we have first undertaken a
MAS NMR spectra is narrower on TF-Beta than on conventional
targeted structural investigation of the products from SDA-free
templated zeolite Beta. This finding confirms the more ordered
synthesis.
nature of the TF-Beta zeolite.
Analysis of the powder XRD-pattern of zeolite Beta from
SDA-free synthesis (TF-Beta) reveals two significant features
(Fig. 1). It is immediately obvious that the stacking disorder
of the layered building block is very similar to that of the
mineral Tschernichite and also to that of templated zeolite
Beta.8 Simulations of the stacking disorder using DIFFaX show
that the simulated powder XRD-pattern gave the best fit with
an intergrowth of 44% of the A-polymorph and 56% of the
B-polymorph. Therefore, TF-Beta benefits from the interconnected
12MR system of the silicate skeleton, without the structural
features the ordered end-members would have (Fig. S2, ESI†).
Inspecting the powder XRD-pattern more closely and focusing on
those reflections which are not broadened by the structural
disorder, e.g. (008) at B12.31 2y Mo-Ka radiation (Fig. S3, ESI†),
it is obvious that the inherent linewidth is much smaller than for
the templated zeolite Beta material. In Fig. S3 (ESI†), the full width
at half maximum (FWHM) of (008) of TF-Beta was determined to
be 0.0611 2y as compared to 0.0831 2y for templated zeolite Beta, a
Fig. 2 29Si MAS NMR spectra with the deconvoluted peaks for zeolite Beta from
templated synthesis (a, CP814, H+-form) and SDA-free synthesis (b, H+-form).
CP814 material from Zeolyst. This implies that the material from
SDA-free synthesis possesses fewer structural defects, yielding a
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
The extraordinarily high aluminum content of TF-Beta is
Catalysis Science & Technology
definitely one of the most striking features of this material. It is
the key characteristic in defining the catalytic performance,
ion-exchange capacity and framework polarity of this zeolite. In
addition, the potential for further functionalization via isomor-
phous substitution may also depend on the framework alumi-
num atoms. Therefore, understanding the nature of aluminum
in TF-Beta and comparison to the templated system are of
critical importance.
Two-dimensional 27Al 3Q MAS NMR spectra were used to
remove the anisotropic line broadening, allowing identification
of species with similar isotropic chemical shifts but different
quadrupolar coupling constants. As shown in Fig. 3a for the
templated Beta (CP814, Si/Al = 19, H+-form), the overlapping peaks
observed in the 27Al MAS spectrum at 54 ppm are clearly resolved
in the F2 field of the 3Q MAS NMR spectrum. These are assigned to
tetrahedrally coordinated aluminum at different T-sites. Mean-
while, in the F1 field two distinct framework aluminum species,
designated as Al(IV)a and Al(IV)b with SOQE of 1.9 and 1.7 MHz,
respectively, can be seen in the tetrahedral region. A third one,
Al(VI)a, is also observed in the octahedral environment at ca. 0 ppm
with SOQE of 0.75 MHz. By contrast, for the aluminum rich
TF-Beta (Si/Al = 4.6), after ammonium exchange and decomposi-
tion of NH4 to protons (H+-form, Fig. 3b), there is only one
+
framework aluminum signal with SOQE of 2.3 MHz in the projec-
tion of the isotropical-chemical-shift field (F1 field) with only very
minor formation of 6-coordinated aluminum in non-lattice posi-
tions. This means that this large amount of aluminum occupies
only a restricted number of T-sites. The high Al content will affect
the overall strain and the geometries in the structure; therefore,
the different Al-sites in the H-form of TF-Beta cannot be further
resolved or assigned.
In order to confirm the structural differences observed
above between TF-Beta and conventional templated zeolite
Beta, another commercial zeolite Beta sample (CP811, H+-form)
was also investigated following the same procedure. Two-
dimensional 27Al 3Q MAS NMR spectrum for this commercial
zeolite Beta sample with slightly higher Al content (Si/Al = 12) is
presented in Fig. S4 (ESI†). For this sample, in the F1 field two
distinct framework aluminum species, designated as Al(IV)a and
Al(IV)b with SOQE of 1.7 and 1.6 MHz, respectively, can be
detected in the tetrahedral region. A third one, Al(VI), is also
observed in the octahedral environment at ca. 0 ppm with
Fig. 3 Two-dimensional 27Al 3Q MAS NMR spectra of templated zeolite Beta
(a, CP814, H+-form) and TF-Beta (b, H+-form). The corresponding 27Al MAS NMR
spectrum is given on the top of the 3Q MAS plot. The F1 projection is the pure
isotropic spectrum.
SOQE of 1.2 MHz. These findings provide another comparison from the maximum of the second NH3 desorption peak, is the
with the conventional zeolite Beta products from templated synth- same for as-prepared TF-Beta as for the templated Beta. After
eses and confirm the distinct nature of the Al-sites in TF-Beta.
steam-dealumination, the amount of desorbed NH3 decreases
The acidity of TF-Beta was investigated using NH3 temper- and after subsequent acid treatment, the first desorption peak
ature-programmed desorption. Fig. 4 shows the NH3-TPD pro- becomes smaller than the second. This indicates that controlled
files of two templated Beta zeolites (CP811 with Si/Al of 12 and modification allows us to vary the number of the acid sites over a
CP814 with Si/Al of 19) and TF-Beta. NH3-TPD profiles are also broad range, while keeping the strength of the strongest acid
given for these samples after steam-dealumination and further sites at least at the same high level. The sharper TPD peaks for
NH4+ ion-exchange or acid treatment. All Beta zeolites show two TF-Beta derived samples in comparison with the templated
peaks in the TPD profile. TF-Beta shows a much larger amount material are in excellent agreement with more uniform and
of desorbed NH3 than the templated Beta zeolite, in agreement well-defined acid sites. 27Al MAS NMR spectrum of the steam-
with its larger aluminum content and number of acid sites. dealuminated TF-Beta (Fig. S5, ESI†) exhibited a broad peak at
Remarkably, the strength of the strongest acid sites, as evaluated around 30 ppm together with peaks attributed to tetrahedrally
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
benzene with ethylene and the acylation of anisole with acetic
anhydride were selected here as model cases for respectively
TF-Beta as such, or TF-Beta after controlled modification. In
Fig. 6 the benzene conversion in the alkylation of benzene
with ethylene at 150 1C is given for the commercial Beta
zeolites CP811 (Si/Al = 12) and CP814 (Si/Al = 19) and for
TF-Beta (Si/Al = 4.6), at a benzene : ethylene ratio of 8. Full
ethylene conversion is reached for all catalysts after 24 hours.
TF-Beta shows a higher initial activity than the commercial
catalysts in this alkylation reaction. This shows that in spite of
its high aluminum content, the acid sites of TF-Beta are of
sufficient strength and number to catalyze this demanding
ethylation reaction. At all measured conversions, the selectivity
to ethylbenzene (Fig. 6) is distinctly higher for TF-Beta than for
the various templated Beta zeolites. At increasing conversions,
this is accompanied by a lower selectivity to diethylbenzene
(Fig. S6, ESI†). This higher selectivity towards the monoalkylated
Fig. 4 NH3-TPD profiles of Beta zeolites (Si/Al ratios between brackets).
and octahedrally coordinated Al species. The broad peak at
around 30 ppm was no longer observed on the samples that
were obtained by treating the steam-dealuminated zeolite with
either HNO3 aq. or NH4NO3 aq. (Fig. S5, ESI†). Moreover,
octahedrally coordinated Al species were also removed by acid
treatment of the steam-dealuminated TF-Beta, leaving aluminum
only in tetrahedral framework positions.
Fig. 5 shows the O–H stretching vibration domains in the IR
spectra of protonic forms of a templated Beta zeolite (CP811)
and of TF-Beta. For CP811 the hydroxyl vibrations at 3742 cmÀ1
and 3735 cmÀ1 are dominant. These correspond to the non-
acidic silanol groups at the outer surface of the zeolite crystals
and at structural defects inside the crystals, respectively.9
TF-Beta has a much lower concentration of such hydroxyl groups,
in agreement with its significantly higher degree of crystalline
order and larger crystal size. For TF-Beta the bridging hydroxyl
groups at 3607 cmÀ1 are much more prominent compared to the
other Beta zeolites, in line with the high concentration of well-
defined Brønsted acidic sites. A vibration of minor intensity is
observed for TF-Beta at 3660 cmÀ1, corresponding to hydroxylated
monomeric or polymeric extra-framework aluminum species,
which is in line with the NMR characteristics of the TF-Beta after
conversion to the H+-form (Fig. 3b).
The distinct structural features of TF-Beta also translate into
unique catalytic properties for this material. The alkylation of
Fig. 6 Benzene ethylation (150 1C): (top) conversion vs. time (hours), and
(bottom) selectivity vs. conversion for templated Beta zeolites CP814 (’) and
CP811 (E), and for TF-Beta (m).
Fig. 5 Infrared spectra of template zeolite Beta (a, CP811) and TF-Beta (b)
at 150 1C.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 1 Catalytic activity of zeolite Beta materials originating from TF-Beta in
comparison to those originating from templated Beta for the acylation of anisole
with acetic anhydridea
leading to fast catalyst deactivation. On the one hand, deal-
umination of TF-Beta leads to a less polar material that is more
resistant to deactivation which explains the higher activity of
the dealuminated TF-Beta samples. On the other hand, deal-
umination also decreases the number of acid sites, as can be
seen from the NH3-TPD profiles (Fig. 4). This explains why even
though the acid-treated dealuminated TF-Beta has the highest
TON, its overall p-MAP yield is lower than that of the Al-richer
Sample
Si/Alb TONc p-MAP yieldd (%)
TF-Beta
Steam-dealuminated TF-Beta
4.6
7.1
—
11
54
15
26
30
35
0
38
48
54
22
38
36
Acid-treated dealuminated TF-Beta 31
+
NH4 exch., dealuminated TF-Beta
9.9
CP811
CP814
Acid-treated CP814
12
19
26
+
NH4 ion-exchanged dealuminated TF-Beta, the latter showing
the best combination of acid site strength, concentration and
framework polarity.
a
Reaction conditions: 60 1C, 50 mg catalyst in H+ form, 50 mmol anisole,
5 mmol acetic anhydride. Measured by ICP. TON in mol (mol Al)À1
.
b
c
d
After 5 hours of reaction.
Conclusions
product is readily explained by the higher framework polarity of
the aluminum-rich TF-Beta, which favors adsorption of benzene
over that of ethylbenzene and therefore facilitates the ethylben-
zene desorption.
In summary, SDA-free synthesis of zeolite Beta is not only
economically attractive, but also results in zeolitic products
with superior catalytic performance. The as-synthesized material
possesses a high density of active sites with exceptional stability
and distinctively ordered nature, useful in e.g. ethylation of
benzene; after dealumination and/or other post-synthesis treat-
ments, catalysts with varying Si/Al ratios, suitable e.g. for acylation
of anisole, are obtained. The ability to manipulate the framework
aluminum content in a very broad range, while maintaining
structural integrity, proves that TF-Beta zeolites constitute a
powerful toolbox for designing new acid catalysts.
This work was performed under the framework of the
INCOE (International Network of Centers of Excellence) project
coordinated by BASF. T.D.B. acknowledges F.W.O.-Vlaanderen
(Research Foundation – Flanders) for a doctoral fellowship. The
authors thank Prof. A. Atıf Akın of Rutgers University for his
help with the cover page artwork.
As a second reaction to demonstrate the potential of cata-
lysts derived from TF-Beta, the Friedel–Crafts acylation of
anisole with acetic anhydride was performed on TF-Beta and
on commercial Beta zeolites CP811 and CP814. Results are
summarized in Table 1. The H+-form of TF-Beta (Si/Al = 4.6)
showed no activity for the acylation of anisole even after 24 h.
However, other Beta samples derived from TF-Beta through
various post-synthesis treatments were highly productive for
p-methoxyacetophenone (p-MAP, selectivity > 99%). p-MAP
yields based on the initial amount of acetic anhydride are
compared in Table 1. While dealumination of TF-Beta by steam
treatment decreased the aluminum content (Si/Al = 7.1), the
catalytic activity was dramatically improved and this steam-
dealuminated TF-Beta showed a much higher conversion
(p-MAP yield = 38%) than the templated zeolite Beta with the
lowest Si/Al ratio (Si/Al = 12, p-MAP yield = 22%) after 5 hours of
Notes and references
reaction. Acid treatment of the steam-dealuminated TF-Beta 1 (a) G. Bellussi, G. Pazzuconi, C. Perego, G. Girotti and
further enhanced the catalytic activity (48% conversion after
5 hours), although the aluminum content of the zeolite was
noticeably decreased (Si/Al = 31). Meanwhile, NH4 ion-
G. Terzoni, J. Catal., 1995, 157, 227–234; (b) R. A. Innes,
S. I. Zones and G. J. Nacamuli, US 4891458, Chevron
Research Company, 1990; (c) M. Spagnol, L. Gilbert,
E. Benazzi and C. Marcilly, US 5817878, Rhone-Poulenc
Chimie, 1998; (d) A. Corma, L. T. Nemeth, M. Renz and
S. Valencia, Nature, 2001, 412, 423–425; (e) T. F. Degnan
and P. J. Angevine, US 6652735, ExxonMobil Research and
Engineering Company, 2003; ( f ) L. Wang, US 7169291, UOP
LLC, 2007; (g) H. Ohtsuka, T. Tabata, O. Okada, L. M.
F. Sabatino and G. Bellussi, Catal. Lett., 1997, 44, 265–270;
(h) B. Coq, M. Mauvezin, G. Delahay, J. B. Butet and S. Kieger,
Appl. Catal., B, 2000, 27, 193–198.
+
exchange treatment of the steam-dealuminated TF-Beta followed
by calcination (Si/Al = 9.9) led to an even higher catalytic activity
(54% at 5 hours). The turnover number (TON) at 30 minutes for
each catalyst was estimated based on the aluminum content in
the catalyst and the amount of consumed acetic anhydride. Acid-
treated dealuminated TF-Beta showed a remarkably high TON
of 54 in comparison to the templated zeolite Beta samples with
Si/Al of 12 (TON = 26) and 19 (TON = 30). For comparison of this
highly active TF-Beta sample (Si/Al = 31) with the templated
zeolite Beta at a similar Si/Al ratio, the commercial zeolite Beta 2 R. L. Wadlinger, G. T. Kerr and E. J. Rosinski, US 3308069,
sample with the higher Al content (CP814) was also subjected to 1967.
the acid treatment. The resulting catalyst had a comparable Si/Al 3 (a) J. C. van der Waal, M. S. Rigutto and H. van Bekkum,
ratio (Si/Al = 26), yet the p-MAP yield was significantly lower
(p-MAP yield = 36%) compared to the acid-treated dealuminated
TF-Beta.
The low activity of the aluminum-rich TF-Beta in this reac-
tion can be explained by its high framework polarity. This
renders the desorption of the more polar p-MAP more difficult,
J. Chem. Soc., Chem. Commun., 1994, 1241–1242;
(b) P. Caullet, J. Hazm, J. Guth, J. Joly, J. Lynch and
F. Raatz, Zeolites, 1992, 12, 240–250; (c) M. K. Rubin, EP
0159847, Mobil Oil Corporation, 1989; (d) J. Van der Waal,
P. Kooyman, J. Jansen and H. Van Bekkum, Microporous
Mesoporous Mater., 1998, 25, 43–57; (e) S. Zones,
c
Catal. Sci. Technol.
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
Y. Nakagawa, L. T. Yuen and T. Harris, J. Am. Chem. Soc.,
1996, 118, 7558–7567; ( f ) M. A. Camblor, L. A. Villaescusa
2011, 115, 744–750; (c) G. Majano, L. Delmotte, V. Valtchev
and S. Mintova, Chem. Mater., 2009, 21, 4184–4191.
7 (a) R. E. Morris and S. L. James, Angew. Chem., Int. Ed., 2013,
52, 2163–2165; (b) L. Ren, Q. Wu, C. Yang, L. Zhu, C. Li,
P. Zhang, H. Zhang, X. Meng and F.-S. Xiao, J. Am. Chem. Soc.,
2012, 134, 15173.
´
˜
and M. J. Dıaz-Cabanas, Top. Catal., 1999, 9, 59–76.
4 B. Xie, J. Song, L. Ren, Y. Ji, J. Li and F. S. Xiao, Chem. Mater.,
2008, 20, 4533–4535.
5 B. Xie, H. Zhang, C. Yang, S. Liu, L. Ren, L. Zhang, X. Meng,
B. Yilmaz, U. Muller and F.-S. Xiao, Chem. Commun., 2011, 47, 8 R. Szostak, K. P. Lillerud and M. Stocker, J. Catal., 1994, 148,
3945–1947. 91–99.
6 (a) Y. Kamimura, W. Chaikittisilp, K. Itabashi, A. Shimojima 9 (a) A. Simon-Masseron, J. Marques, J. Lopes, F. R. Ribeiro,
and T. Okubo, Chem.–Asian J., 2010, 5, 2182–2191;
(b) Y. Kamimura, S. Tanahashi, K. Itabashi, A. Sugawara,
T. Wakihara, A. Shimojima and T. Okubo, J. Phys. Chem. C,
I. Gener and M. Guisnet, Appl. Catal., A, 2007, 316, 75–82;
(b) I. Kiricsi, C. Flego, G. Pazzuconi, W. O. J. Parker, R. Millini,
C. Perego and G. Bellussi, J. Phys. Chem., 1994, 98, 4627–4634.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol.