Synthesis and catalytic behavior of ferrierite zeolite nanoneedles

Article abstract of DOI:10.1021/cs400025s

The proton form of nanosized, needlelike ferrierite zeolite, which was synthesized using choline and Na⁺ cations as structure-directing agents, was found to be much more efficient for the skeletal isomerization of 1-butene to isobutene than the corresponding cation form of conventional, submicrometric ferrierite with a platelike shape, mainly because of the considerably lower density of strong acid sites, but as well as a result of the higher density of 10-ring pore mouths.

Full text of DOI:10.1021/cs400025s

Letter
pubs.acs.org/acscatalysis
Synthesis and Catalytic Behavior of Ferrierite Zeolite Nanoneedles
†
,§
†,§
†
‡
‡
†
́
Yoorim Lee, Min Bum Park, Pyung Soon Kim, Aurelie Vicente, Christian Fernandez, In-Sik Nam,
†
,
and Suk Bong Hong *
†Department of Chemical Engineering and School of Environmental Science and Engineering, POSTECH, Pohang 790-784, Korea
‡Laboratoire Catalyse et Spectrochimie, ENSICAEN, Universite
de CAEN, CNRS, 6 Boulevard Marechal Juin, 14050 CAEN, France
S Supporting Information
́
́
*
ABSTRACT: The proton form of nanosized, needlelike
ferrierite zeolite, which was synthesized using choline and
+
Na cations as structure-directing agents, was found to be
much more efficient for the skeletal isomerization of 1-butene
to isobutene than the corresponding cation form of conven-
tional, submicrometric ferrierite with a platelike shape, mainly
because of the considerably lower density of strong acid sites,
but as well as a result of the higher density of 10-ring pore
mouths.
KEYWORDS: ferrierite zeolite nanoneedles, synthesis, characterization, catalytic properties
+
eolites are crystalline, microporous aluminosilicates whose
applications as catalysts have led to an innovative shift in
Our interest in Ch as an organic SDA began with the
elegant work by Lewis and co-workers on the charge density
mismatch synthesis of the large-pore zeolites UZM-4 (BPH)
and UZM-22 (MEI) in the presence of this small organic cation
Z
the petroleum and petrochemical industries due to the extreme
uniformity in size and shape of their channels and cavities that
1
,2
+
2+
14
enables shape-selective catalysis.
However, microporosity
together with Li , Sr , or both. Tetramethylammonium
+
+
often brings about severe intracrystalline diffusion limitations,
(TMA ) and tetraethylammonium (TEA ) ions, the two most
studied organic SDAs in the synthesis of zeolites and molecular
sieves, are slightly smaller and larger than Ch , respectively.
lowering the accessibility and molecular transport of reactive
3
+
molecules to/from the active sites confined. One way to solve
this problem is to shorten the diffusion path length in zeolite
micropores. Over the past decade, in fact, much attention has
been devoted to the reduction of zeolite crystal size from the
micrometric to the nanometric range. Consequently, more than
Until now, at least 26 zeolitic materials with different
framework topologies have been synthesized using one of
+
+
13
TMA and TEA , with or without inorganic cations present.
However, they include neither UZM-4 nor UZM-22, implying
+
10 different structure types of nanocrystalline zeolites and
that Ch , although flexible due to its 2-hydroxyethyl chain,
phosphate-based molecular sieves have been successfully
synthesized.
could be very selective for a particular zeolite structure once the
right synthesis conditions are found. This stimulated us to
investigate the influence of inorganic synthesis variables on the
phase selectivity of the crystallization under the optimized
conditions (ChOH/Si = 0.8 and Si/Al = 5) for UZM-4
4
−6
Ferrierite (framework type FER) is a medium-pore, high-
silica zeolite that contains a two-dimensional (2D) pore system
consisting of 10-ring (4.2 × 5.4 Å) channels intersected by 8-
7
14
ring (3.5 × 4.8 Å) channels. This zeolite is well-known for the
formation. As shown in Table 1, we were able to crystallize
exceptional selectivity in the skeletal isomerization of n-butenes
four different zeolitic phases, depending on the Al content of
the synthesis mixtures and the concentration or type (or both)
of alkali metal ions employed as a crystallization SDA.
8
−12
to isobutene.
Although ferrierite is a rare natural zeolite, it
can also be synthesized using a variety of different organic
1
3
When the crystallization was carried out under rotation (60
structure-directing agents (SDAs) in the laboratory. However,
there is little known on the synthesis of nanosized ferrierite
crystals. Here, we describe the synthesis of ferrierite nano-
rpm) at 150 °C for 14 days, for example, ZSM-34, an
15
intergrowth of offretite (OFF) and erionite (ERI), was the
product formed from a sodium aluminosilicate solution with
Na/Si = 0.2 and Si/Al = 5. In addition, an increase in the Si/Al
ratio of this synthesis solution to 20 resulted in the
crystallization of ferrierite. When decreasing its Na/Si ratio to
+
needles with a Si/Al ratio of 9.9 using choline (Ch , (2-
hydroxyethyl)trimethylammonium), one of the cheapest
+
alkylammonium ions, and Na as SDAs. We also report that
the proton form (H-ferrierite) of this nanometric zeolite phase
is much more selective and recyclable for the skeletal
isomerization of 1-butene to isobutene than that of conven-
tional, submicrometric ferrierite with a similar Si/Al ratio (8.9)
but a different platelike crystal morphology.
Received: January 11, 2013
Revised: February 26, 2013
Published: February 27, 2013
©
2013 American Chemical Society
617
dx.doi.org/10.1021/cs400025s | ACS Catal. 2013, 3, 617−621

ACS Catalysis
Table 1. Representative Synthesis Conditions and Results
Letter
a
Si/
Al
Si/
Al
b
b
run
1
M
Li+
product
run
11
M
Li+
product
c
5
amorphous
20 ferrierite +
RUB-10
2
3
4
5
Na+
K+
5
5
5
5
amorphous
ZSM-34
c
12
13
14
15
Na+
K+
20 ferrierite
20 UZM-12
20 amorphous
20 amorphous
+
+
c
c
Rb
sodalite + U
Rb
+
+
Cs
PST-8 +
Cs
amorphous
+
U
6
7
Li+
10 ferrierite +
RUB-10
16
17
Li+
∞
∞
amorphous
PST-8
c
Na+
Na+
Figure 1. (a) Powder XRD pattern and (b) SEM and (c) TEM images
of ferrierite nanoneedles synthesized in this work.
10 ferrierite +
sodalite
8
9
K+
Rb+
10 UZM-12 + U
18
19
K+
Rb+
∞
∞
PST-8 + U
PST-8 + U
10 ferrierite +
UZM-12
crystallites with a platelike morphology. However, the needle-
22
like crystallites have also been reported in the literature.
Cs+
Cs+
c
+
10
10 ferrierite +
sodalite
20
∞
amorphous
There is one patent shown in the literature on the Ch -
mediated synthesis of a ferrierite-type zeolite designated ZSM-
a
The oxide composition of the synthesis mixture is 4ChOH·MCl·xA-
3
8, all the X-ray peaks of which are as broad as those observed
+
+
+
+
+
l O ·5SiO ·150H O, where M is Li , Na , K , Rb , or Cs and x is 0,
2
0
3
2
2
for ferrierite zeolite described here, apart from the existence of
.125, 0.25, or 0.5, respectively. All the syntheses were performed
two small impurity peaks around 2θ = 11.1 and 17.8° in its
under rotation (60 rpm) at 150 °C for 14 days unless otherwise stated.
23
b
powder XRD pattern. According to this patent, ZSM-38 can
be obtained when a sodium aluminosilicate gel with Si/Al = 8.3
containing choline chloride (ChCl/Si = 0.16) and NaOH as
SDAs is heated at 99 °C for 70 days or longer. Although the
synthesis conditions of ZSM-38 are notably different from
those of our ferrierite nanoneedles, most of its physicochemical
properties, including the crystal morphology and size, are still
unknown.
The product appearing first is the major phase, and U indicates the
c
unknown phase. The material obtained after 28 days of heating.
0
.1 (not shown in Table 1), however, we always obtained a
small amount of RUB-10 (RUT) as an impurity phase, as well
as ferrierite. On the other hand, the synthesis using an Al-free
synthesis mixture with Na/Si = 0.2 yielded a layered silicate
denoted PST-8. It is worth noting that while PST-8 is
characterized by a powder X-ray diffraction (XRD) pattern
that is quite similar to that of any of as-made ERS-12, PLS-1,
1
13
+
H− C CP MAS NMR spectroscopy evidence that the Ch
ions are occluded mainly intact in ferrierite nanoneedles
(
Supporting Information Figure S3), and a combination of
elemental and thermal analyses reveals that they have the unit
cell composition |Ch Na OH (H O) | [Al Si O ].
16−19
MCM-65, UZM-13, etc.,
which are known to consist of
ferrierite layers, its calcination in air at 550 °C led to a
transformation into the CDO framework type material with
poor crystallinity (Supporting Information Figure S1). This
suggests that the ferrierite layers of PST-8 are stacked in a fairly
less-ordered manner compared with those of layered materials
with ferrierite layers described above. A structural investigation
of PST-8 using synchrotron powder XRD data is in progress,
and the results will be given elsewhere. Table 1 also shows that
the replacement of NaCl with the equivalent amount of KCl
under conditions where the formation of ferrierite proved to be
reproducible gave UZM-12 (ERI), a high-silica analog of zeolite
3.0
0.6
0.3
2
5.3
3.3 32.7 72
Thus, there is an enrichment of Al in the product (Si/Al =
.9) with respect to the synthesis mixture (Si/Al = 20). To
9
further investigate the nanocrystalline nature of our ferrierite,
we performed N sorption measurements on its as-made and
2
proton forms. Here, we outgassed as-made ferrierite at 150 °C
under vacuum to a residual pressure of 10− Torr for 2 h before
measurement to minimize the decomposition of the occluded
5
+
Ch ions. As shown in Figure 2, the isotherms of both materials
are characterized by a hysteresis loop at high relative pressures,
typical of mesoporosity: their external surface (or mesoporous)
2
−1
2
0,21
areas were determined to be ∼150 and 240 m g , respectively,
erionite.
It thus appears that the oxide composition range
yielding pure ferrierite in the presence of ChOH is quite
narrow.
Figure 1 shows the powder XRD pattern of the as-made
ferrierite obtained from run 12 in Table 1. We note here that all
the X-ray reflections are much broader than those observed for
common ferrierite zeolite crystallized as the platelike
morphology of a submicrometric scale (Supporting Information
Figure S2), although their positions and relative intensities are
in good agreement with each other, with the lack of a
noticeable amorphous background. This suggests its nano-
crystalline nature that can be further evidenced by the scanning
and transition electron microscope images. As shown in Figure
1
, our ferrierite appears as very small, heavily overlapped
Figure 2. Pore size distribution curves for the as-made (left) and
proton (right) forms of ferrierite nanoneedles calculated using the BJH
needles ∼10 nm in diameter and 100 nm in length, the lattice
fringes of which are practically destroyed within a few seconds
in the incident electron beam of 200 kV. In general, both
natural and synthetic versions of ferrierite zeolite consist of
formalism from the N desorption branch isotherm. The inset shows
2
the N2 adsorption−desorption isotherms of the corresponding
materials.
6
18
dx.doi.org/10.1021/cs400025s | ACS Catal. 2013, 3, 617−621

ACS Catalysis
Letter
Figure 3. 1-Butene conversion (left), selectivity to isobutene (middle), and propene plus pentenes (right) as a function of time on stream in 1-
butene skeletal isomerization at 400 °C and 10.1 kPa 1-butene pressure over H-FER(n) (■) and H-FER(c) (□).
Figure 4. 27Al MAS NMR spectra (left), IR spectra (middle) in the OH stretching region, and NH TPD curves (right) of H-FER(n) (bottom) and
3
H-FER(c) (top).
suggesting significant interparticle adsorption. Barrett−Joyner−
Halenda (BJH) analyses indicate that they have an average pore
size of ∼10 nm. From now on, for catalytic comparison, we will
refer to our nanocrystalline ferrierite and a commercially
available, submicrocrystalline ferrierite with a Si/Al ratio of 8.9
and a platelike shape of ∼0.7 μm in base and 0.1 μm in
thickness (Supporting Information Figure S2) as FER(n) and
FER(c), respectively.
as well as that of the parent NaK form of FER(c), exhibits only
one resonance around 50 ppm (Supporting Information Figure
S4), typical of tetrahedral Al in the zeolite framework. As
27
shown in Figure 4, however, the Al MAS NMR spectra of H-
FER(n) and H-FER(c) are characterized by an additional
resonance around 0 ppm. Because the relative intensity (∼29 vs
14%) of this 27Al resonance is considerably stronger in the
spectrum of H-FER(n) than in that of H-FER(c), it is clear that
the amount of Al atoms extracted from the zeolite framework
during the preparation of their proton form is much higher for
the nanocrystalline zeolite, although there is no detectable
decrease in X-ray peak intensity. When combining with data
from elemental analysis, the framework Si/Al ratios of H-
FER(n) and H-FER(c) were determined to be 13.9 and 10.3,
respectively. Quite similar ratios (13.6 and 10.1) were also
Figure 3 shows 1-butene conversion and selectivity to
isobutene and major byproducts as a function of time on stream
(TOS) in the skeletal isomerization of 1-butene over H-FER(n)
and H-FER(c) measured at 400 °C and 10.1 kPa 1-butene in
the feed. It is well established that the rate of coke formation
within the zeolite pores during this isomerization, which has a
strong impact on the catalytic performance of zeolites, differs
8
−12
29
notably according to their acidic properties.
To minimize
derived from the deconvolution of their Si MAS NMR spectra
its effects on the isobutene selectivity, therefore, we regarded
the results obtained at 5 min on stream as the intrinsic activities
of these two H-FER catalysts with different morphologies and
sizes. As shown in Figure 3, no large differences in the initial 1-
butene conversion are observed. However, notice that the initial
isobutene selectivity (75%) of nanocrystalline H-FER(n) is
much higher than that (40%) of submicrocrystalline H-FER(c).
A similar trend can also be observed over the period of TOS
studied, although the extent of selectivity increase with TOS
was greater for the latter catalyst, making the selectivity
difference (∼30%) slightly smaller.
(Supporting Information Figure S5).
The point given above can be further supported by the IR
and NH temperature-programmed desorption (TPD) results
3
−1
(Figure 4). The intensity of the IR band around 3600 cm due
to the hydroxyl groups of Brønsted acid sites was found to be
much weaker for H-FER(n) than for H-FER(c). In addition,
deconvolution of the NH TPD curves shows that the total area
3
of NH desorption (i.e., the density of acid sites) of H-FER(n)
3
is about 40% that of H-FER(c). In particular, the area of the
high-temperature desorption peak around 500 °C due to strong
acid sites, which induce undesired side reactions such as
dimerization followed by cracking into byproducts such as
Although much controversy remains on the reaction
mechanism by which isobutene is produced over H-FER in
the steady state, there is a general acceptance that the initial
isobutene formation is dominated by the nonselective
bimolecular mechanism, that is, dimerization followed by
12
propene and pentenes during 1-butene skeletal isomerization,
from the former catalyst was found to be only one-fourth that
from the latter one. This explains why H-FER(n) gives a higher
initial isobutene selectivity than H-FER(c), together with a
fairly low selectivity to byproducts (Figure 3).
1
2
27
cracking. The Al MAS NMR spectrum of as-made FER(n),
6
19
dx.doi.org/10.1021/cs400025s | ACS Catal. 2013, 3, 617−621

ACS Catalysis
Letter
On the other hand, the spatial distribution of Al atoms (i.e.,
Brønsted acid sites) in the zeolite framework can be altered by
the size and charge distribution of organic SDAs em-
Thus, if the pseudomonomolecular mechanism is prevailing
with prolonged TOS, the number of highly selective sites for
isobutene formation, that is, alkylaromatic tertiary carbenium
1
9,24,25
ployed.
sites in H-FER(n) and H-FER(c), we collected their Al
QMAS NMR spectra (Supporting Information Figure S6).
To examine differences in the location of acid
ions on the coke molecules located near the zeolite pore
27
11,29
mouths,
should then be larger for H-FER(c) with a higher
3
coke content. As shown in Figure 3, however, a considerably
higher isobutene selectivity after some TOS is observed for H-
FER(n) with a lower coke content. As previously proposed by
Although H-FER(n) has a higher amount of Al in the
extraframework environment than H-FER(c), no significant
2
7
28
differences in the shape of the tetrahedral Al resonance were
found. This suggests that notable differences in their acidic
properties have no relation with the framework Al distribution.
our group, therefore, it appears that 1-butene skeletal
isomerization over aged H-FER takes place over the Brønsted
acid sites located at or near its 10-ring pore mouths, although
this monomolecular mechanism cannot avoid the energetically
unfavorable ring-opening of a cyclopropyl cation. We also note
that H-FER(n) shows better recyclability in this reaction than
H-FER(c). As can be found in Supporting Information Figure
S10, the extent of decrease in 1-butene conversion and
isobutene selectivity at 12 h on stream during the four
successive catalytic cycles carried out at 400 °C for 24 h was
lower for the former catalyst. This is not unexpected because
nanocrystallinity leads to shorter diffusion path in zeolite
micropores and is thus of great advantage to regenerate used
zeolite catalysts upon calcination in flowing air at elevated
temperatures.
26
A recent study by Kim et al. has shown that the acid sites on
the external surface of nanocrystalline ZSM-5 zeolites, whose
density should be inversely proportional to the crystal size, are
weaker in acid strength than those in their intracrystalline
space. It thus appears that nanocrystallinity could also exert a
great influence on the acidity of H-FER zeolites. However, the
details are unclear at this time.
The 3QMAS NMR results also suggest that a quite similar
consideration could be made for the dealumination behavior of
zeolites. However, the actual situation is not so simple, because
H-NU-88 (Si/Al ∼ 13), a nanocrystalline material with a
crystallite size of 20−30 nm that could be an intergrowth of
several hypothetical polymorphs in the beta family of zeolites,
has no detectable extraframework Al species, even after heating
In summary, we have successfully synthesized ferrierite
zeolite nanoneedles with a Si/Al ratio of 9.9 using choline and
2
7
+
at 800 °C. This led us to suspect that the structural feature of
zeolites may have a greater effect on the extent of deal-
umination than their nanocrystallinity.
Na ions as SDAs. We also found that the proton form of this
nanocrystalline zeolite is a considerably more efficient catalyst
for the skeletal isomerization of 1-butene to isobutene than the
corresponding cation form of conventional, submicrometric H-
ferrierite with a similar bulk Si/Al ratio (8.9) but a different
platelike crystal morphology. The overall results of our work
suggest that monomolecular, pore mouth shape catalysis over
the Brønsted acid sites located at or near the 10-ring pore
mouths is the origin of the remarkable isobutene selectivity of
H-ferrierite observed only after some time on stream.
The 10-ring channels of the FER framework structure are
reported to be perpendicular to the (001) surface of the
platelike FER crystals that are thus parallel to their basal lines
1
2,28
(Supporting Information Figure S7).
Assuming that they
are along with the long axis of needlelike FER(n) crystallites
with 10 nm in mean diameter and 100 nm in mean length, the
number of the 10-ring channels at their boundary (i.e., 10-ring
pore mouths) was calculated to be about 9 times larger than
that of the analogous mouths of the equivalent weight of
platelike H-FER(c) crystals with ∼0.7 μm base and 0.1 μm
thickness. This ratio is more than twice as large as that (4) of
ASSOCIATED CONTENT
Supporting Information
■
*
S
1
13
Details of experimental procedures, powder XRD, H− C CP,
2
−1
27
29
27
their external surface areas (240 vs 60 m g ) derived from the
t-plot method, which appears to originate from the heavily
overlapped nature of our FER nanoneedles, as well as from the
assumption of cylindrical geometry. However, it is clear that the
number of Brønsted acid sites at or near the 10-ring pore
mouths should be much larger on H-FER(n) than on H-
FER(c), although the former zeolite has a somewhat higher
framework Si/Al ratio (∼14 vs 10).
Al, and Si MAS and Al 3QMAS NMR, crystal morphology
illustrations, TGA/DTA, GC-MS, and results of recyclability
test. This material is available free of charge via the Internet at
http://pubs.acs.org/.
AUTHOR INFORMATION
Corresponding Author
E-mail: sbhong@postech.ac.kr.
■
*
According to the knowledge accumulated thus far, there are
two major types of prevailing reaction mechanisms in the
selective formation of isobutene over H-FER zeolite after some
TOS: (i) the pseudomonomolecular pathway involving an
alkylaromatic tertiary carbenium ion and (ii) the monomo-
Author Contributions
§
These authors contributed equally.
Notes
The authors declare no competing financial interest.
12
lecular pathway involving a primary carbenium ion.
A
ACKNOWLEDGMENTS
combination of thermogravimetric and differential thermal
analyses (TGA/DTA) reveals that the amount of (5.5 vs 9.1 wt
■
This work was supported by the National Research Foundation
(2012R1A3A2048833, 2012K1A3A4A07030457, and 2012-
0008674) of Korea and by the Pierre Hubert Curien STAR
program (2012 Project No. 27816PB) of France.
%
) of coke deposited during the 1-butene skeletal isomerization
at 400 °C for 24 h is considerably smaller in H-FER(n) than in
H-FER(c) (Supporting Information Figure S8), which can be
further supported by ex situ gas chromatography−mass
spectroscopy (Supporting Information Figure S9). Despite
notable differences in their acidity, in addition, the main
components (bi- and tricyclic aromatic species) of their coke
deposits were found to be essentially identical with each other.
REFERENCES
1) Degnan, T. F. J. Catal. 2003, 216, 32.
(2) Camblor, M. A.; Hong, S. B. In Porous Materials; Bruce, D. W.,
■
(
Walton, R. L., O’Hare, D., Eds.; Wiley: Chichester, 2011; p 265.
6
20
dx.doi.org/10.1021/cs400025s | ACS Catal. 2013, 3, 617−621

ACS Catalysis
Letter
(
3) Per
́
ez-Ramírez, J.; Christensen, C. H.; Egeblad, K.; Christensen,
C. H.; Groen, J. C. Chem. Soc. Rev. 2008, 37, 2530.
(
4) Tosheva, L.; Valtchev, V. P. Chem. Mater. 2005, 17, 2494.
(
5) Ng, E.-P.; Chateigner, D.; Bein, T.; Valtchev, V.; Mintova, S.
Science 2012, 335, 70.
6) Zhang, X.; Liu, D.; Xu, D.; Asahina, S.; Cychosz, K. A.; Agrawal,
(
K. V.; Al Wahedi, Y.; Bhan, A.; Al Hashimi, S.; Terasaki, O.;
Thommes, M.; Tsapatsis, M. Science 2012, 336, 1684.
(
7) Baerlocher, Ch.; McCusker, L. B. Database of Zeolite Structures;
http://www.iza-structure.org/databases/ (Accessed March 2, 2013)..
8) Mooiweer, H. H.; de Jong, K. P.; Kraushaar-Czarnetzki, B.;
Storkk, W. H. J.; Krutzen, B. C. H. Stud. Surf. Sci. Catal. 1994, 84,
327.
9) O’Young, C.-L.; Pellet, R. J.; Casey, D. G.; Ugolini, J. R.; Sawicki,
R. A. J. Catal. 1995, 151, 467.
10) Seo, G.; Jeong, H. S.; Jang, D.-L.; Cho, D. L.; Hong, S. B. Catal.
Lett. 1996, 41, 189.
11) Guisnet, M.; Andy, P.; Gnep, N. S.; Travers, C.; Benazzi, E. J.
Chem. Soc., Chem. Commun. 1995, 1685.
12) van Donk, S.; Bitter, J. H.; de Jong, K. P. Appl. Catal., A 2001,
12, 97.
13) Szostak, R. Handbook of Molecular Sieves; Van Nostrand
Reinhold: New York, 1992.
14) Miller, M. A.; Moscoso, J. G.; Koster, S. C.; Gatter, M. G.;
Lewis, G. J. Stud. Surf. Sci. Catal. 2007, 170A, 347.
15) Rubin, M. K.; Rosinski, E. J.; Plank, C. J. U.S. Patent 4,086,186,
978.
16) Ikeda, T.; Akiyama, Y.; Oumi, Y.; Kawai, A.; Mizukami, F.
Angew. Chem., Int. Ed. 2004, 43, 4892.
17) Millini, R.; Carluccio, L. C.; Carati, A.; Bellussi, G.; Perego, C.;
Cruciani, G.; Zanardi, S. Microporous Mesoporous Mater. 2004, 74, 59.
(
2
(
(
(
(
2
(
(
(
1
(
(
(
18) Dorset, D. L.; Kennedy, G. J. J. Phys. Chem. B 2004, 108, 15216.
(
19) Knight, L. M.; Miller, M. A.; Koster, S. C.; Gatter, M. G.; Benin,
A. I.; Wills, R. R.; Lewis, G. J.; Broach, R. W. Stud. Surf. Sci. Catal.
007, 170A, 338.
20) Miller, M. A.; Lewis, G. J.; Moscoso, J. G.; Koster, S.; Modica,
F.; Gatter, M. G.; Nemeth, L. T. Stud. Surf. Sci. Catal. 2007, 170A, 487.
21) Lee, J. H.; Park, M. B.; Lee, J. K.; Min, H.-K.; Song, M. K.;
Hong, S. B. J. Am. Chem. Soc. 2010, 132, 12971.
22) Pinar, A. B.; Gomez-Hortiguela, L.; Perez-Pariente, J. Chem.
Mater. 2007, 19, 5617.
23) Plank, C. J.; Rosinski, E. J.; Rubin, M. K. U.S. Patent 4,046,859,
977.
2
(
(
(
(
1
(
(
(
24) Shantz, D. F.; Lobo, R. F. J. Phys. Chem. B 1998, 102, 2339.
25) Shantz, D. F.; Lobo, R. F. Chem. Mater. 1998, 10, 4015.
26) Kim, K.; Ryoo, R.; Jang, H.-D.; Choi, M. J. Catal. 2012, 288,
1
(
15.
27) Lee, S.-H.; Lee, D.-K.; Shin, C.-H.; Park, Y.-K.; Wright, P. A.;
Lee, W. M.; Hong, S. B. J. Catal. 2003, 215, 151.
(
(
28) Lee, S.-H.; Shin, C.-H.; Hong, S. B. J. Catal. 2004, 223, 200.
29) Andy, P.; Gnep, N. S.; Guisnet, M.; Benazzi, E.; Travers, C. J.
Catal. 1998, 173, 322.
6
21
dx.doi.org/10.1021/cs400025s | ACS Catal. 2013, 3, 617−621