Post-synthesized zirconium-containing Beta zeolite in Meerwein-Ponndorf-Verley reduction: Pros and cons

Article abstract of DOI:10.1016/j.apcata.2015.01.001

Zr-Beta zeolite was prepared by a two-step post-synthesis method involving dealumination of Al-Beta followed by wet impregnation with Zr(NO₃)₄. Compared with Zr-Beta formed under fluoride-mediated hydrothermal conditions, the post-synthesized samples had smaller particle size and stronger Lewis acidity. The materials were tested as catalysts for Meerwein-Ponndorf-Verley reduction. In the reduction of 4-tert-butylcyclohexanone, it exhibited the same excellent stereoselectivity toward cis-4-tert-butylcyclohexanol (>99%) as the HF-synthesized Zr-Beta, but had a lower TOF. Because of the higher density of zirconium sites and the nanosized crystallites, it was a more effective catalyst for the MPV reduction of 1,4-cyclohexanedione, bulky aldehydes and aromatic ketones. However, it is more susceptible to poisoning by water adsorption because of its hydrophilic nature. The easily scalable synthesis method allows a faster preparation of metal-substituted Lewis acid zeolites, although differences in textural and chemical properties should be taken into consideration when the material is applied as a catalyst.

Full text of DOI:10.1016/j.apcata.2015.01.001

Applied Catalysis A: General 493 (2015) 112–120
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Post-synthesized zirconium-containing Beta zeolite in
Meerwein–Ponndorf–Verley reduction: Pros and cons
Jie Wanga,1, Kazu Okumura , Stephan Jaenicke , Gaik-Khuan Chuah
b
a
a,∗
a
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Kent Ridge, Singapore 117543, Singapore
Faculty of Engineering, Department of Applied Chemistry, Kogakuin University, 2665-1 Nakano-machi, Hachioji City, Tokyo 192-0015, Japan
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 October 2014
Received in revised form
Zr-Beta zeolite was prepared by a two-step post-synthesis method involving dealumination of Al-Beta
followed by wet impregnation with Zr(NO3)4. Compared with Zr-Beta formed under fluoride-mediated
hydrothermal conditions, the post-synthesized samples had smaller particle size and stronger Lewis
acidity. The materials were tested as catalysts for Meerwein–Ponndorf–Verley reduction. In the
reduction of 4-tert-butylcyclohexanone, it exhibited the same excellent stereoselectivity toward cis-4-
tert-butylcyclohexanol (>99%) as the HF-synthesized Zr-Beta, but had a lower TOF. Because of the higher
density of zirconium sites and the nanosized crystallites, it was a more effective catalyst for the MPV
reduction of 1,4-cyclohexanedione, bulky aldehydes and aromatic ketones. However, it is more suscep-
tible to poisoning by water adsorption because of its hydrophilic nature. The easily scalable synthesis
method allows a faster preparation of metal-substituted Lewis acid zeolites, although differences in
textural and chemical properties should be taken into consideration when the material is applied as a
catalyst.
2
2 December 2014
Accepted 4 January 2015
Available online 10 January 2015
Keywords:
Zirconium-beta
Zeolite
Post-synthesis
Lewis acidity
Meerwein–Ponndorf–Verley reduction
©
2015 Elsevier B.V. All rights reserved.
1
. Introduction
example, Sn-Beta or Zr-Beta zeolites are highly efficient for a
wide range of organic reactions, including Baeyer-Villiger oxidation
[21], intramolecular carbonyl-ene reaction [22–24], etherification
[25], and Meerwein–Ponndorf–Verley (MPV) reduction [26–29].
More recently, transition metal-containing Beta zeolites caught
the attention of researchers for biomass conversion owing to
the superiority in activating carbonyl groups which are abundant
in biomass-related molecules. These zeolites are used instead of
enzymes to isomerize [30–35] or epimerize [36,37] sugars via
intramolecular hydride shift in a MPVO redox cycle. Sn-Beta and
Zr-Beta are also efficient catalysts for other biomass-related reac-
tions, such as the conversion of sugar to lactic acid derivatives
[38–40], furfural and levulinic acid to ␥-valerolactone [41,42] and
levoglucosenone to chiral ␥-lactones [43].
Zeolites with variable microporous topology, strong acidity and
high stability are widely used as heterogeneous catalysts in the
petrochemical industry and organic syntheses [1–5]. Lewis acid
zeolites are of particular interest in fine chemical synthesis and
biomass conversion [6–9]. In most cases, the Lewis acidity is caused
by trivalent Al . However, Lewis acid zeolites can also be obtained
by the incorporation of tetravalent heteroatoms (e.g., Ti , Zr
3
+
4+
4+
and Sn4 ) into the pure-silica zeolitic framework. A milestone in
Lewis acid zeolites was the discovery of the titanosilicate TS-1,
a titanium-containing zeolite with MFI topology [3]. It has been
successfully commercialized in processes like propylene epoxida-
tion, hydroxylation of phenol, and cyclohexanone ammoximation
+
[
10–14]. Tetravalent heteroatoms have also been incorporated
Transition metal cations have larger radii than Si4+. A fluoride
medium instead of the conventional alkali medium is therefore nec-
essary for the crystallization of these Lewis acid zeolites [17]. In
a fluoride-mediated synthesis, the reaction takes place at neutral
pH to avoid precipitation of the metal precursor as a metal oxide,
which would make it unavailable for incorporation into the frame-
work. However, due to the lower basicity, nucleation and crystal
growth are delayed, resulting in long synthesis times (>20 days) and
large crystals (>1 ␮m). Another drawback is the limited amount of
substitution (Si/M > 75) [27].
into the framework of MWW and BEA type zeolites resulting in
enhanced and sometimes unique catalytic activity [15–20]. For
∗ Corresponding author. Tel.: +65 65162839; fax: +65 67791691.
E-mail address: chmcgk@nus.edu.sg (G.-K. Chuah).
Present address: Johnson Matthey (Shanghai) Catalyst Co., Ltd., 586 Dong Xing
1
Road, Songjiang, Shanghai 201613, PR China.
http://dx.doi.org/10.1016/j.apcata.2015.01.001
0
926-860X/© 2015 Elsevier B.V. All rights reserved.

J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
113
An alternative strategy to obtaining metal-substituted Beta zeo-
lites is by post-synthesis of easily synthesized Al-Beta. Due to the
excellent stability of the BEA framework, the aluminum can be
removed via acid treatment whilst preserving the microporous
topology. Dzwigaj and co-workers [44–48] found that by wet
impregnation, a variety of transition-metal ions (e.g., V, Co, Ni, Fe,
Cr) can be incorporated into the vacant tetrahedral sites of dealu-
minated Beta zeolite. In addition, chemical vapor deposition, solid
state ion-exchange and grafting in isopropanol were attempted for
the synthesis of Sn-Beta via post-synthetic routes [49–51]. These
alternative syntheses make it possible to achieve a higher metal
content in the framework while avoiding lengthy hydrothermal
treatment. Furthermore, smaller crystals are obtained, which is
beneficial for good mass transfer and diffusion.
Si/Zr ratio of 10) and Zr(OH) were prepared as reported elsewhere
[42,53].
4
2.2. Characterization
Surface area and porosity of the samples were determined by
nitrogen adsorption (Micromeritics Tristar 3000). Prior to each
◦
measurement, each sample was thoroughly degassed at 300 C
under a nitrogen flow for 4 h. The crystalline phase was deter-
mined by powder X-ray diffraction (Siemens D5005 equipped with
Cu anode and variable slits). The diffractograms were recorded at a
◦
step size of 0.02 and a dwell time of 1 s. SEM images were obtained
on a JEOL JSM-6701F field emission scanning electron microscope
using 5 kV electron beam. Elemental composition of the catalysts
was determined by inductively coupled plasma-atomic emission
spectroscopy (ICP-AES) after dissolving the sample in a 1 mL of HF
and diluting to 10 mL with deionized water. The diffuse reflectance
spectra (DRS) were taken using a Shimadzu UV-2450 UV-Visible
spectrophotometer. 29Si MAS NMR spectra were measured on a
Bruker DRX-400 widebore solid state spectrometer operating at a
resonance frequency of 79.49 MHz with a spinning rate of 8 kHz,
pulse length of 3 ␮s and a recycle delay of 20 s. 4 mm zirconia
rotors were used and the 29Si chemical shifts are reported rel-
ative to a TMS standard. Zr K edge (18.0 keV) XAFS data were
collected at the BL01B1 station of the Japan Synchrotron Radia-
tion Research Institute (JASRI). A Si(1 1 1) single crystal was used to
obtain a monochromatic X-ray beam. The measurement was car-
ried out in the quick mode. Ion chambers filled with N2 (75%)/Ar
(25%) and Ar (100%) were used to determine I0 and I, respectively.
The samples were pressed into self-supporting wafers. The data
analysis was performed using the REX2000 Ver. 2.0.4 program
These successful examples inspired us to develop zirconium-
containing Beta zeolites with varying Si/Zr from 12.5 to 150 using
the two-step post-synthesis method (Scheme 1). Zr-Beta with
8
wt% Zr (Si/Zr ∼37) had been prepared by Hermans and collabora-
tors using solid state impregnation with zirconium ethoxide [52].
Here, we apply the wet impregnation of dealuminated zeolite sim-
ilar to that described by Dzwigaj et al. [44]. The procedure uses an
aqueous metal salt, and the homogeneous solution ensures good
reproducibility from batch to batch. The influence of the zirconium
salt (ZrOCl2 or Zr(NO ) ) for successful incorporation of zirconium
3
4
into the zeolitic framework was investigated. The activity of the
materials was compared with that of HF-synthesized zeolites, using
the Meerwein–Ponndorf–Verley reduction of carbonyl compounds
as test reaction. When fluoride anions are employed as mineraliz-
ing agent, zeolites formed are hydrophobic with low internal defect
density, whereas zeolites synthesized using hydroxide anions are
hydrophilic and contain vacancy defects in the framework. The
extent to which these defects affect the performance of the zeo-
lites is of interest. Furthermore, the ability to increase the density
of metal active sites is exciting as it is expected to have a positive
effect on the rate of reactions.
3
(Rigaku). Fourier transformations of k ꢀ(k) data were performed in
−
1
the k range of 30–150 nm . Infrared spectra of the samples were
−
1
recorded in the range of 4000–400 cm
using a Bio-Rad Excal-
−
1
ibur FT-IR spectrometer with a resolution of 1 cm . Samples were
pressed into self-supporting discs (10–20 mg) and pre-treated in
an evacuated (<1 mbar) quartz cell with CaF2 window at 450 C for
◦
6
h before measurements. To determine the Lewis and Brønsted
2
. Experimental
acidity, the pyridine adsorption test was used. The samples were
pretreated at 350 C for 2 h before exposure to pyridine vapour
◦
2.1. Materials preparation
◦
(22 mbar) for 15 min, followed by re-evacuated at 100–300 C for
Commercial Beta zeolite (Zeolyst CP814E, Si/Al of 12.5) was
1 h before measuring the IR spectra. The acidic nature of the sam-
ples was also quantified by temperature programmed desorption
◦
dealuminated by treatment in nitric acid solution (12 M) at 80 C for
−1
◦
2
0 h (10 mL g ). After filtration, the dealuminated powder (deAl␤)
of NH . Samples were pretreated at 600 C for 2 h in a flow of helium
3
−
1
◦
was thoroughly rinsed with water and dried overnight in an oven at
(50 mL min ). After cooling to 100 C, NH3 gas was introduced for
15 min. The sample was flushed with helium for another 2 h before
heating at 10 C min . The desorption of NH3 was monitored by a
quadrupole mass spectrometer (Balzers Prisma 200). Thermogravi-
metric analyses to determine the water content were performed on
◦
8
0 C. Next, the dealuminated zeolite was suspended in water and
◦
−1
the calculated amount of zirconium salt (Zr(NO ) or ZrOCl ·8H O)
3
4
2
2
was added. The slurry was heated under stirring to remove the
◦
◦
water, dried overnight at 80 C in an oven and calcined at 600 C for
h. The samples are denoted as Zr␤-NO -x or Zr␤-Cl-x depending
◦
8
a Dupont SDT 2960. The sample was first kept at 100 C for 0.5 h to
remove physically absorbed water, and then heated to 800 C at
3
◦
on the zirconium precursor used and x = Si/Zr ratio of 12.5, 75 and
50.
The hydrothermal synthesis of Zr-Beta zeolite with Si/Zr ratio
◦
−1
.
1
10 C min
of 100 (Zr␤-F-100) in fluoride medium was carried out as previ-
ously reported [27]. Briefly, 10.42 g TEOS was mixed with 10.31 g
tetraethyl-ammonium hydroxide (TEAOH, 40 wt% solution) and
hydrolyzed under stirring. After 2 h, 1.55 g of an aqueous solution
containing the required amount of ZrOCl ·8H O was added drop-
2.3. Catalytic tests
◦
Prior to use, the samples were pre-activated at 500 C for
1 h. Typically, for the MPV reduction, 1 mmol substrate and
5 mL of 2-propanol (65 mmol) were placed in a two-necked
25 mL round bottomed flask equipped with a septum port and
a reflux condenser. The flask was heated to reflux tempera-
2
2
wise. The mixture was stirred for another 7–8 h until the ethanol
formed upon hydrolysis of TEOS was completely evaporated.
Finally, 1.215 mL of HF (40% solution) and 0.105 g pure silica zeolite
Beta seeds in 1 g of water were sequentially added. The crystalliza-
◦
ture (82 C) in an oil bath and the catalyst (100 mg) was added
to the reaction mixture (0 min). The tested substrates include
1,4-cyclohexanedione (Sigma–Aldrich), acetophenone (Lancaster),
4-chloroacetophenone (Merck), benzylacetone (Sigma–Aldrich),
cinnamaldehyde (Fluka), ␣-amyl cinnamaldehyde (Sigma–Aldrich)
and citral (Fluka, cis + trans). For the MPV of 1,4-cyclohexandedione
tion was carried out in a Teflon-lined stainless steel autoclave at
◦
1
40 C for 20 days. The solid product obtained was filtered, washed
◦
◦
with deionized water, dried at 100 C and calcined at 580 C for 10 h.
For comparison of catalytic activity, mesoporous Zr-SBA-15 (with

1
14
J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
Scheme 1. Two-step post-synthetic strategy in the preparation of Zr-Beta.
and 4-tert-butylcyclohexanone, 2 and 5 mmol of the substrate,
respectively, 5 mL of 2-propanol and 50 mg catalyst were used.
Aliquots were taken at different times to follow the progress
of reaction. The samples were analyzed by an Agilent HP 6890
gas chromatograph equipped with a HP-5 capillary column
(
250 ␮m × 0.25 ␮m × 30 m) and FID detector. The identity of the
products was verified either by comparing the retention times
with authentic samples or by GC–MS analyses. Mass balances were
checked by calibration of the peaks with authentic samples and the
error was within 3%.
3
. Results and discussion
3.1. Physicochemical properties
In the first step, aluminum was removed from commercial
Beta zeolite, ␤12.5 (Si/Al 12.5) by treatment with nitric acid to
Si/Al > 1300. The surface area and microporosity were retained,
indicating that the dealumination did not damage the crystallinity
of the samples (Table 1). The samples were wet impregnated
with zirconium nitrate or zirconyl chloride and calcined. Analy-
sis by inductively coupled plasma-atomic emission spectroscopy
Fig. 2. X-ray diffractograms of (a) Zr␤-F-100, (b) ␤-12.5, (c) deAl␤, (d) Zr␤-NO3-12.5
and (e) Zr␤-Cl-12.5.
(
ICP-AES) showed that the zirconium content in the samples was
The X-ray diffractogram of Zr␤-F-100 (Fig. 2) shows sharp
diffraction peaks of well-crystallized zeolite Beta, in agreement
with the large crystallites observed in the electron micrographs.
Zeolite beta comprises two closely related and highly intergrown
polymorphs, and the crystals have a high planar stacking fault den-
sity [54]. As a result, the X-ray diffractogram contains narrow as
well as broad reflections. The shoulder at the higher angle side of
close to the expected values (Table 1). Compared to the dealu-
minated zeolite, deAl␤, the surface area of the post-synthesized
Zr-Beta samples was lower especially for those with high zir-
conium content (Si/Zr 12.5). Although the total pore volume of
3
−1
0
.94–1.28 cm g was relatively large, the micropore volume con-
3
−1
stituted only 0.16–0.17 cm g . The bulk of the pore volume can
be attributed to the interparticle space. From N sorption measure-
◦
the ∼8 peak indicates a slight preference for one of the two poly-
2
ments (Fig. S1), these pores were in the range of 20–50 nm, which
agrees with the size of the particles observed in SEM (Fig. 1). The
zeolite particles had a similar size as the dealuminated precursor
zeolite, and no significant change occurred upon impregnation with
the zirconium salts. In contrast, the Zr-Beta prepared by hydro-
thermal synthesis using HF, Zr␤-F-100, had a total pore volume of
only 0.27 cm g with the bulk of it being due to micropores. The
zeolite particles formed are much larger, in the order of 1–2 ␮m.
This is due to the long crystallization time required for the fluoride-
mediated synthesis.
morphs. In comparison, the commercial Beta zeolite gave broader
peaks of lower intensity, indicative of smaller crystallites. The crys-
tallite size remained essentially unchanged after dealumination
and subsequent impregnation with Zr(NO ) . However, the posi-
3
4
◦
tion of the (3 0 2) peak at 2ꢁ ∼ 22.6 shifted slightly in response to
the changes of composition. Dealumination of ␤-12.5 resulted in a
3
−1
decrease in the interplanar distance, d, for the (3 0 2) planes from
3.950 A˚ to 3.906 A˚ , indicating a contraction of the BEA matrix. Fol-
lowing the incorporation of zirconium into the zeolite framework,
the d(302)-spacing in Zr␤-NO -12.5 increased again to 3.919 A˚ . Even
3
Fig. 1. SEM images of (a) Zr␤-F-100 and (b) Zr␤-NO3-12.5.

J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
115
Table 1
Physicochemical properties of the samples.
−1
Vmicro (cm3
−1
Vtotal (cm3
−1
g )
Samples
-12.5
Si/Ala
Si/Zra
SBET (m2
g
)
g
)
␤
12
–
–
12
12
73
537
521
464
475
495
502
474
0.17
0.17
0.17
0.16
0.16
0.17
0.20
1.21
1.14
0.94
0.91
1.23
1.28
0.27
deAl␤
>1300
>1300
>1300
>1300
>1300
–
Zr␤-Cl-12.5
Zr␤-NO3-12.5
Zr␤-NO3-75
Zr␤-NO3-150
Zr␤-F-100
157
107
a
Determined by ICP-AES.
(
(
d)
c)
(
(
b)
a)
180
280
380
480
580
Wavelength (nm)
Fig. 3. UV/vis diffuse reflectance spectra of (a) deAl␤, (b) Zr␤-NO3-12.5, (c) Zr␤-Cl-
2.5 and (d) ZrO2.
1
with the highest zirconium content (Si/Zr 12.5), no diffraction lines
corresponding to bulk ZrO particles were observed. The absence of
2
bulk ZrO2 was confirmed by the lack of a sharp absorption edge at
∼
240 nm in the UV–vis diffuse reflectance spectrum (DRS) (Fig. 3).
The existence of very small ZrO crystallites is difficult to detect
2
by XRD and DRS. Therefore, the dispersion of zirconium was fur-
ther characterized by extended X-ray absorption fine structure
(
EXAFS). The EXAFS spectrum of Zr␤-Cl-12.5 showed a rapid oscil-
lation on the Zr K edge. The Fourier transform of the EXAFS data
shows an obvious peak at around 3.4 A˚ (Fig. 4), which corresponds
to the Zr
the Zr
O Zr bond distance [55]. In contrast, for Zr␤-NO -12.5,
Zr oscillation was smaller, indicating less ZrO2 aggrega-
3
O
tion in samples synthesized using Zr(NO3)4 as precursor than with
ZrOCl ·H O. Furthermore, the peak at 3.0 A˚ in the Fourier transform
2
2
spectrum of Zr␤-NO -12.5 has a higher amplitude than in Zr␤-Cl-
3
Fig. 4. EXAFS data of Zr K edge and Fourier transforms for (a) Zr␤-Cl-12.5, (b) Zr␤-
NO3-12.5 and (c) Zr␤-F-100.
1
2.5. This peak is assigned to the Zr O Si bond. Hence, the choice
of the zirconium source, Zr(NO3)4 or ZrOCl , used for wet impreg-
2
nation appears to affect the dispersion of zirconium. For Zr␤-F-100,
29Si MAS NMR was used to study the change in the chemical
environment before and after zirconium incorporation (Fig. 6). The
spectrum of the dealuminated sample shows three peaks at ca.
synthesized under fluoride-mediated hydrothermal conditions, the
Zr
O
Zr peak at 3.4 A˚ is very small, pointing to the highest dis-
persion of zirconium amongst the samples. This is not unexpected
because the zirconium content in this sample is much lower than
in the post-synthesized samples.
The incorporation of zirconium into the framework was also fol-
lowed by FT-IR (Fig. 5). The commercial Al-Beta zeolite exhibits a
−
104, −112 and −114 ppm. The broad peak at −104 ppm corre-
3
sponds to Si atoms in a Q (Si(OH)(OSi) ) environment, while the
3
4
latter two peaks are due to Si(OSi) (Q ) at different crystallographic
sites [56,57]. The Q environment is derived from structural defects
4
3
−
1
−1
of connectivity (Si-O- or Si-OH groups). The incorporation of zirco-
nium into the vacant T-atom sites consumes the terminal Si-OH
groups, leading to a reduction in the intensity of the Q3 peak at ca.
sharp band at 3745 cm and a weak band at ca. 3780 m , indica-
tive of terminal Si-OH and Al-OH groups, respectively [51]. After
−
1
dealumination, the band at 3780 cm disappeared, indicating the
absence of Al-OH groups. Instead, the opening of silanol nests
and the creation of vacant tetrahedral T-sites resulted in a broad
band at around 3500 cm . Following the zirconium incorporation,
the broad band disappeared, consistent with the consumption of
silanol nests.
−
102 ppm. Furthermore, the Zr␤-NO -12.5 sample has three broad
Q resonances at ca. −112, −113 and −116 ppm. These overlapping
3
4
−
1
resonances are caused by Si(OSi) and Si(OSi) OZr at different crys-
4 3
tallographic sites, similar to that of Zr-Beta prepared in the fluoride
medium [27].

1
16
J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
(
a)
(
a)
(
(
b)
c)
o
1
00 C
o
2
00 C
o
3
00 C
4000
3800
3600
3400
3200
3000
-1
Wavenumber (cm )
1700
1600
1500
1400
-1
Wavenumber (cm )
Fig. 5. FT-IR spectra of (a) ␤-12.5, (b) deAl␤ and (c) Zr␤-NO3-12.5 samples.
(
b)
-
112
-
114
-
104
-
112
-
113
(
a)
-
116
o
1
00 C
-
102
(
b)
o
2
00 C
o
3
00 C
-
100
-105
-110
-115
-120
-125
-130
Chemical shift (ppm)
1700
1650
1600
1550
1500
-
1450
1400
1
Fig. 6. Solid state 29Si MAS NMR of (a) deAl␤ and (b) Zr␤-NO3-12.5.
Wavenumber (cm )
Fig. 8. Infrared spectra of (a) Zr␤-NO3-12.5 and (b) Zr␤-F-100 after pyridine adsorp-
3.2. Acidity measurements
tion and subsequent desorption at the indicated temperatures.
The acidic properties were studied using NH -TPD (Fig. 7) and
3
◦
temperature range, from 130 to 480 C, showing that these materi-
pyridine FTIR (Fig. 8). The desorption of NH3 from Zr␤-F-100
als contained sites with low as well as high acid strength. Although
physically adsorbed NH3 might contribute to desorption below
00 C, this was minimized by flushing the sample with helium for
h prior to the measurements. Silanol nests that are not taken up
by zirconium or partially hydrolyzed framework zirconium could
add to the wide range of acidity in the post-synthesized samples.
The acidity of Sn-Beta is well studied, and the weak Brønsted acid-
ity has been attributed to residual unoccupied silanol nests [58].
The Sn exists as nonhydrolyzed Sn(-O-Si-)4 or partially hydrolyzed
◦
◦
occurred between 220 and 380 C with a maximum at ∼300 C.
From the amount of desorbed NH , the density of acid sites was
calculated to be 45 ␮mol g . Desorption of NH3 from the post-
3
◦
2
2
−
1
synthesized Zr-Beta zeolites (Si/Zr 12.5–150) occurred over a wider
(
-Si-O-) Sn-OH, and the former was shown to be a weaker Lewis
3
acid than the latter [58,59]. The density of acid sites in the post-
synthesis modified material was higher than for the HF-synthesized
zeolite (Fig. 7, inset). The concentration of acid sites increased with
−
1
the amount of Zr incorporated, from 100 ␮mol g
with Si/Zr = 150 to 347 ␮mol g
for a sample
−
1
for that with Si/Zr = 12.5. Even
at a Si/Zr of 150, the post-synthesized Zr␤-NO -150 contained
3
about twice as many acid sites as Zr␤-F-100. The pyridine IR spec-
tra of the post-synthesized Zr␤-NO -12.5 sample show absorption
3
bands indicative of Lewis and Brønsted acidity even after heat-
◦
ing to 300 C (Fig. 8). Pyridine coordinatively bound to Lewis acid
sites have vibrational modes at ∼1447–1460, 1488–1503, 1580 and
−
1
1
600–1633 cm [60,61]. The adsorption of pyridine at Brønsted
acid site leads to the formation of the pyridinium ion which has
Fig. 7. NH3-TPD profiles of (a) Zr␤-NO3-12.5, (b) Zr␤-NO3-75, (c) Zr␤-NO3-150 and
−
1
(
d) Zr␤-F-100.
vibrational modes at 1485–1500, 1530–1550 and 1630–1640 cm
.

J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
117
Table 2
Catalytic activity for MPV reduction of 4-tert-butylcyclohexanone.
Entry
Catalyst
Conv.
cis Sel.
(%)
Rate
(mmol g
TOFZr
a
−1 −1
)b
−1
b
(
%)
h
(h
)
1
2
3
4
5
6
7
8
9
deAl␤
0
–
>99
>99
>99
>99
>99
18
17
–
>99
–
–
522
20.2
49.6
283
353
–
–
–
–
c
Zr␤-F-100
Zr␤-Cl-12.5
Zr␤-NO3-12.5
Zr␤-NO3-75
Zr␤-NO3-150
Zr(OH)4
ZrSBA-15
Zr␤-NO3-12.5
Zr␤-NO3-12.5
98
31
94
97
71
62
24
0
87.2
23.5
62.6
62.0
39.2
–
–
–
–
d
e
f
1
0
83
a
Reaction conditions: 5 mmol 4-tert-butylcyclohexanone, 5 mL 2-propanol,
◦
5
0 mg catalyst, 82 C, 6 h.
b
Based on the conversion of initial 30 min.
After 2 h.
After 20 h.
c
d
e
Sample hydrated in air.
Sample re-activated by calcination at 500 C for 1 h.
f
◦
Scheme 2. Illustration of the transition states leading to cis- and trans-4-tert-
butylcyclohexanol over Zr-Beta.
In particular, the band at 1530–1550 cm 1 can be unambigu-
ously assigned to Brønsted acidity as it does not overlap with any
bands arising from coordinatively bonded pyridine. For Zr␤-F-100,
−
reaction is proposed to occur via a six-membered transition state
formed by the chemisorption of the sec-alcohol and the ketone at
the Lewis acid site (Scheme 2). Hydrogen from the C H bond of the
alcohol is transferred to the carbonyl carbon of the acceptor. When
the active Lewis acid site is located inside the zeolitic framework,
the transition state leading to the cis-alcohol is preferred as it is
more aligned with the BEA channel than that for the trans-alcohol
[27,62]. The size of the transition state to the trans-alcohol with an
axially oriented form is around 0.65–0.70 nm, which is difficult to
accommodate in the BEA channels (ca. 0.66 nm × 0.67 nm). How-
ever, mesoporous materials such as Zr(OH)4 or Zr-SBA-15, where
pore constraints are not an issue, give the trans-alcohol as the main
product (Table 2, entries 7 and 8). Hence, the high stereoselectiv-
ity to cis-4-tert-butylcyclohexanol indicates that the zirconium in
the post synthesized Zr-Beta has to be well dispersed within the
zeolite framework. The heterogeneous nature of the catalyst was
confirmed by removal of the catalyst after 2 h when the conversion
had reached 58% (Fig. S4). No increase in conversion was observed
in the residual solution after removal of the catalyst.
−1
the 1530–1550 cm
band was very weak, and after heating to
00 C or above, only the absorption bands due to coordinatively
bonded pyridine were observed. This indicates that the fluoride-
synthesized zeolite contained predominantly Lewis acid sites. For
comparison, the pyridine IR spectra of dealuminated zeolite, deAl␤,
◦
2
−
1
show absorption bands centered at 1445 and 1595 cm together
−
1
with a very weak absorption in the 1485–1500 cm region (Fig.
S3). These bands are indicative of pyridine hydrogen-bonded to
surface hydroxyl groups and/or bonded to very weak Lewis acid
sites. As deAl␤ is highly siliceous, it lacks strong Lewis acidity.
3
.3. MPV reduction of 4-tert-butylcyclohexanone
To evaluate the catalytic performance of the post-synthesized
Zr-Beta samples, their activity for the Meerwein–Ponndorf–Verley
MPV) reduction of 4-tert-butylcyclohexanone was investi-
gated. We have previously reported that Zr-Beta synthesized
using HF under hydrothermal conditions showed excellent
activity and stereoselectivity for this reaction, with cis-4-tert-
butylcyclohexanol as the only detectable product [27]. Indeed,
an initial turnover frequency based on zirconium sites (TOFZr) of
(
3.4. Scope of MPV reduction
−
1
5
22 h was obtained over Zr␤-F-100 (Table 2). Zirconium ions in
We conducted the MPV reduction of other substrates to
compare the performance of the post-synthesized and hydrother-
mally synthesized Zr-Beta zeolites. The MPV reduction of
1,4-cyclohexanedione gives initially 4-hydroxycyclohexanone,
which can be further reduced to 1,4-cyclohexanediol. The
the zeolitic framework are the active sites because dealuminated
pure silica Beta zeolite showed no activity at all. The TOFZr was
lower for post-synthesized Zr-Beta than for Zr-Beta synthesized
hydrothermally using HF. For Zr-Beta prepared using ZrOCl ·8H O,
2
2
−
1
the initial TOFZr was only 20.2 h . This zeolite had a much higher
zirconium content than the fluoride-synthesized sample, Si/Zr 12.5
vs 100 (Table 2, entry 3). The lower activity per site may be due to
a lower dispersion of the zirconium species, as deduced from the
EXAFS results. Post-synthesis with Zr(NO3)4 as zirconium source
resulted in more active samples, and conversions of 71 to 97%
were obtained after 6 h. The rate of reaction decreased with the
diol is potentially useful as a penetrant in dermatological
formulations [63]. Over Zr␤-F-100 as catalyst, the conversion of 1,4-
cyclohexanedione required 18 h to reach >97% (Fig. 9a). The yield
of 4-hydroxycyclohexanone reached a maximum of 56% after 8 h.
However, the subsequent reduction of 4-hydroxycyclohexanone
to 1,4-cyclohexanediol was rate limiting with the rate being 13.5
times lower than the first step. Hence, even after 18 h, the reac-
tion mixture still contained about 30% 4-hydroxycyclohexanone.
The reduction of 4-hydroxycyclohexanone to 1,4-cyclohexanediol
−
1
reduction of zirconium content. However, the TOFZr was 353 h
in Zr␤-NO -150 and only 49.6 h for Zr␤-NO -12.5. This decrease
−1
3
3
in TOF shows that a higher Zr loading leads to a less effective use of
each zirconium site. A similar trend had been reported for Sn-Beta
where a higher Sn-loading did not help to boost the rate of sugar
isomerization [51].
Irrespective of the mode of synthesis, all the zeolites
showed excellent stereoselectivity (>99%) toward cis-4-tert-
was faster over the post-synthesized Zr␤-NO -100 and after 18 h,
a higher yield of 1,4-cyclohexanediol was obtained. As both
3
Zr␤-F-100 and Zr␤-NO -100 have the same zirconium content, the
3
higher reaction rate may be due to the smaller particle size of the
latter which would facilitate diffusion to active sites within the pore
channels. Furthermore, the concentration of 1,4-cyclohexanedione
and its intermediate in the pore channels depends on the hydropho-
bic/hydrophilic balance at the surface of the two catalysts. In a more
butylcyclohexanol. This has been attributed to
a pore size
constraint effect on the transition state of the reaction. The MPV

1
18
J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
hydrophilic environment, the concentration of the polar interme-
diate, 4-hydroxycyclohexanone, will be higher, enhancing its rate
of reduction to the diol.
the more bulky amyl cinnamaldehyde as effectively, giving only
33% conversion after 24 h. The use of Zr␤-NO -12.5 is clearly advan-
3
tageous for this substrate as the conversion was almost doubled,
62%, after a similar time. Similarly, a higher conversion of citral was
With Zr␤-NO -12.5, where the zirconium content is higher,
3
the conversion of 1,4-cyclohexanedione reached >97% after only
obtained over Zr␤-NO -12.5 than with Zr␤-F-100 (Table 3, entry
3
6
4
h (Fig. 9b). The maximum yield of 4-hydroxycyclohexanone,
6%, was obtained after only 2 h, and complete reduction to 1,4-
6). However, the selectivity to geraniol and nerol was reduced due
to dehydration and was slightly lower for the more acidic Zr␤-
cyclohexanediol occurred after 18 h. Although each site may be
less active than in Zr␤-F-100, the much greater number of sites
increases the probability of reaction, thus giving a higher overall
reaction rate.
NO -12.5. For these bulky substrates, post-synthesized zeolites
3
are efficient catalysts because their small particle size, 20–50 nm,
translates into shorter diffusion paths for the molecule to reach the
active sites in the pore channels.
Next, we investigated the MPV reduction of aromatic ketones
which are harder to reduce than aliphatic or cyclic ketones due
to steric hindrance at the C O posed by the rigid benzene ring. In
such a case, a higher density of active sites in the post-synthesized
Zr-Beta might be useful to increase the chances of encounters. The
reduction of acetophenone and 4-chloro-acetophenone using Zr␤-
F-100 requires a longer time than 4-tert-butylcyclohexanone even
with a higher catalyst loading (Table 3, entries 1 & 2). After 18 h,
the conversion of acetophenone and 4-chloro-acetophenone was
3.5. Hydrophilicity of post-synthesized Zr-Beta
The above results show that the post-synthesized Zr␤-NO3-
12.5 with a higher zirconium content is more active than the
hydrothermally-prepared Zr␤-F-100 for the MPV reduction of
1,4-cyclohexanedione, aromatic ketones and bulky aldehydes.
However, we observed that the post-synthesized sample, when
left in the open for some time, became totally inactive for the
MPV reduction of 4-tert-butylcyclohexanone (Table 2, entry 9).
This is in contrast to the robust and easy-to-handle nature of the
HF-synthesized Zr-Beta. For the post-synthesized zeolite, recalci-
6
1 and 65%, respectively. In comparison, with Zr␤-NO -12.5 as
3
the catalyst, the initial rates were at least two times higher than
for Zr␤-F-100 and after 18 h, the conversion of acetophenone and
◦
4
-chloro-acetophenone was 86 and 98%, respectively. In benzyl
nation at 500 C for 1 h was necessary in order to fully regenerate
acetone, the aromatic ring is at the ␤-position and the molecule
reacts faster than the acetophenones. Over the post-synthesized
its activity (Table 2, entry 10). It is known that samples pre-
pared by post-synthesis and hydrothermal synthesis differ in
their surface hydrophobicity [64,65]. The post-synthesized zeolites
have a higher density of silanol groups than zeolites synthesized
hydrothermally in fluoride. The water adsorption capacity was
examined by thermogravimetric analysis after drying the samples
Zr␤-NO -12.5, full conversion was reached after 6 h compared to
3
only 56% for Zr␤F-100 (Table 3, entry 3). These results show that
a higher Zr density is helpful in increasing the rate of reaction for
these substrates. Furthermore, the smaller particle size of the post-
synthesized sample allows an easier access to the Zr sites located
in the pore channels.
◦
at 100 C for 30 min. The Zr␤-F-100 showed a weight loss of only
◦
2.2% from 100 to 800 C while Zr␤-NO -12.5 had a much higher
3
We have previously found that Zr␤-F-100 showed very good
activity and chemoselectivity in the reduction of cinnamaldehyde
to cinnamyl alcohol [29]. However, this catalyst could not reduce
weight loss of 7.6% (Table 4 and Fig. S5). The weight loss is due
to the removal of adsorbed water at low temperatures (<200 C)
and water from surface OH-groups at high temperatures. These
◦
Table 3
MPV reduction of various substrates.
−
TOFZr (h )a
−1
Entry
1
Substrate
Catalyst
Time (h)
18
Yield (%)
61
Rate (mmol g 1h−1)a
Zr␤-F-100
0.23
1.3
O
Zr␤-NO3-12.5
Zr␤-F-100
18
18
86
65
0.52
0.58
0.5
3.4
2
O
Zr␤-NO3-12.5
18
98
1.35
1.2
Cl
3
4
Zr␤-F-100
6
56
1.64
9.8
O
Zr␤-NO3-12.5
Zr␤-F-100
6
3
100
98
3.40
4.16
3.0
24.9
O
Zr␤-NO3-12.5
Zr␤-F-100
3
24
88
33
3.92
0.15
3.4
0.9
5
6
O
Zr␤-NO3-12.5
Zr␤-F-100
24
4
62
67
0.27
2.25
0.2
13.4
O
Zr␤-NO3-12.5
4
80
2.94
2.6
◦
Reaction conditions: 1 mmol substrate, 5 mL 2-propanol, 100 mg catalyst, 82 C.
a
Based on the conversion over the initial 30 min or 2 h (entries 1, 2 and 5).

J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
119
100
NMR, XRD and DRS indicates that the zirconium species are highly
dispersed within the framework, but EXAFS revealed the existence
(
a)
of some Zr
O Zr bonds in Zr-Beta with Si/Zr loading of 12.5. The
80
60
40
20
0
post-synthesized samples have a higher density of acid sites than
Zr-Beta synthesized hydrothermally in fluoride medium. In the
MPV reduction of 4-tert-butylcyclohexanone, the post-synthesized
Zr-Beta catalysts exhibit a similarly high stereoselectivity toward
cis-4-tert-butylcyclohexanol, >99%, which supports the assumption
that the zirconium sites are located within the zeolitic framework.
Compared to Zr␤-F-100, the post-synthesized Zr␤-NO -12.5 had
3
higher activity for the MPV reduction of 1,4-cyclohexanedione, aro-
matic ketones and bulky aldehydes, which can be attributed to the
higher zirconium substitution level and the smaller particle size.
TGA measurements revealed a higher adsorption of water at the
surface of the post-synthesized Zr-Beta than on the hydrophobic
HF-synthesized zeolite. Hence, it is necessary to activate the sam-
ples prior to catalytic testing.
0
5
10
15
20
Time (h)
Acknowledgement
100
(
b)
This work is financially supported by Ministry of Education,
Singapore, ARC Tier 1 grants R-143-000-476-112 and R-143-000-
550-112.
80
60
40
20
0
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.
2
015.01.001.
References
[
[
[
[
1] A. Corma, Chem. Rev. 97 (1997) 2373–2420.
2] M.E. Davis, Nature 417 (2002) 813–821.
3] C.S. Cundy, P.A. Cox, Chem. Rev. 103 (2003) 663–702.
4] R. Xu, W. Pang, J. Yu, Q. Huo, J. Chen, Chemistry of Zeolites and Related
Porous Materials: Synthesis and Structure, John Wiley & Sons, Singapore,
0
5
10
15
20
2
009.
Time (h)
[
5] H. van Bekkum, H.W. Kouwenhoven, Zeolite Manual for the Organic Chemist,
Mijnbestseller, Netherlands, 2012.
6] A. Corma, H. Garcia, Chem. Rev. 103 (2003) 4307–4366.
7] E. Taarning, C.M. Osmundsen, X.B. Yang, B. Voss, S.I. Andersen, C.H. Christensen,
Energy Environ. Sci. 4 (2011) 793–804.
8] M. Dusselier, P. van Wouwe, A. Dewaele, E. Makshina, B.F. Sels, Energy Environ.
Sci. 6 (2013) 1415–1442.
9] M. Moliner, Dalton Trans. 43 (2014) 4197–4208.
10] M. Taramasso, G. Perego, B. Notari, US Patent 4,410,501, 1983.
11] M.G. Clerici, P. Ingallina, US Patent 5,221,795, 1993.
12] R.A. Sheldon, M. Wallau, I.W. Arends, U. Schuchardt, Acc. Chem. Res. 31 (1998)
485–493.
13] G. Paparatto, A. Forlin, P. Tegon, EP Patent 1,072,599, 2004.
14] F. Cavani, J.H. Teles, ChemSusChem 2 (2009) 508–534.
Fig. 9. MPV reduction of 1,4-cyclohexanedione over (a) Zr␤-F-100 and Zr␤-NO3-
[
[
1
00 (dashed lines) and (b) Zr␤-NO3-12.5. (•, ꢀ) 1,4-cyclohexanedione; (ꢀ, ꢁ) 4-
hydroxycyclohexanone; (ꢂ,ꢁ) 1,4-cyclohexanediol.
[
[
Table 4
Weight loss of Zr␤-NO3-12.5 and Zr␤-F-100.
[
[
[
Sample
Weight loss (%)
◦
◦
I (100–200 C)
II (200–800 C)
[
[
Zr␤-NO3-12.5
Zr␤-F-100
4.6
1.1
2.8
1.1
[15] A. Corma, M. Camblor, P. Esteve, A. Martinez, J. PerezPariente, J. Catal. 145 (1994)
51–158.
1
[
[
[
16] T. Tatsumi, N. Jappar, J. Phys. Chem. B 102 (1998) 7126–7131.
17] S.V. Valencia, A.C. Canos, US Patent 5,968,473A, 1999.
18] P. Wu, T. Tatsumi, T. Komatsu, T. Yashima, Chem. Lett. 29 (2000)
results show that the post-synthesized zeolite easily binds molec-
ular water, while the Beta zeolite synthesized in fluoride media is
more water-resistant. Being more susceptible to water poisoning, it
is necessary to thoroughly dry the post-synthesized zeolites before
use. Furthermore, their catalytic performance will be affected in
reactions where water is produced [64,66] in contrast to the water
tolerance shown by the Zr-Beta hydrothermally synthesized in a
fluoride medium [27].
774–775.
[19] A. Corma, L.T. Nemeth, M. Renz, S. Valencia, Nature 412 (2001) 423–425.
[
[
20] Y. Zhu, G. Chuah, S. Jaenicke, Chem. Commun. (2003) 2734–2735.
21] M. Renz, T. Blasco, A. Corma, V. Fornés, R. Jensen, L. Nemeth, Chem. Eur. J. 8
(
2002) 4708–4717.
[22] A. Corma, M. Renz, Chem. Commun. (2004) 550–551.
[
[
[
23] Y. Zhu, Y.T. Nie, S. Jaenicke, G.-K. Chuah, J. Catal. 229 (2005) 404–413.
24] Y.T. Nie, S. Jaenicke, G.-K. Chuah, Chem. Eur. J. 15 (2009) 1991–1999.
25] A. Corma, M. Renz, Angew. Chem. Int. Ed. 46 (2007) 298–300;
A. Corma, M. Renz, Angew. Chem. 119 (2007) 302–304.
[
[
26] A. Corma, M.E. Domine, L. Nemeth, S. Valencia, J. Am. Chem. Soc. 124 (2002)
4
. Conclusion
3194–3195.
27] Y. Zhu, G. Chuah, S. Jaenicke, J. Catal. 227 (2004) 1–10.
Zirconium can be incorporated into the framework of dealu-
[28] M. Boronat, A. Corma, M. Renz, J. Phys. Chem. B 110 (2006) 21168–21174.
[
[
29] Y. Zhu, G.-K. Chuah, S. Jaenicke, J. Catal. 241 (2006) 25–33.
30] Y. Román-Leshkov, M. Moliner, J.A. Labinger, M.E. Davis, Angew. Chem. Int. Ed.
minated Beta zeolite to form Lewis acid catalysts. The two-step
post-synthesis method shortens the long synthesis time of >20
days to only 3 days, and avoids the use of HF required in the con-
ventional synthesis of Zr-Beta. Characterization by FT-IR, 29Si MAS
49 (2010) 8954–8957;
Y. Román-Leshkov, M. Moliner, J.A. Labinger, M.E. Davis, Angew. Chem. 122
(2010) 9138–9141.

1
20
J. Wang et al. / Applied Catalysis A: General 493 (2015) 112–120
[
31] M. Moliner, Y. Román-Leshkov, M.E. Davis, Proc. Natl. Acad. Sci. U. S. A. 107
[49] P. Li, G. Liu, H. Wu, Y. Liu, J.-G. Jiang, P. Wu, J. Phys. Chem. C 115 (2011)
3663–3670.
(
2010) 6164–6168.
[
[
32] C.M. Lew, N. Rajabbeigi, M. Tsapatsis, Ind. Eng. Chem. Res. 51 (2012) 5364–5366.
33] R. Bermejo-Deval, R.S. Assary, E. Nikolla, M. Moliner, Y. Román-Leshkov, S.-J.
Hwang, A. Palsdottir, D. Silverman, R.F. Lobo, L.A. Curtiss, Proc. Natl. Acad. Sci.
U. S. A. 109 (2012) 9727–9732.
[50] C. Hammond, S. Conrad, I. Hermans, Angew. Chem. Int. Ed. 51 (2012)
11736–11739;
C. Hammond, S. Conrad, I. Hermans, Angew. Chem. 124 (2012) 11906–11909.
[51] J. Dijkmans, D. Gabriëls, M. Dusselier, F. de Clippel, P. Vanelderen, K. Houthoofd,
A. Malfliet, Y. Pontikes, B.F. Sels, Green Chem. 15 (2013) 2777–2785.
[52] P. Wolf, C. Hammond, S. Conrad, I. Hermans, Dalton Trans. 43 (2014)
4514–4519.
[53] D.M. Do, S. Jaenicke, G.-K. Chuah, Catal. Sci. Technol. 2 (2012) 1417–1424.
[54] J.M. Newsam, M.M.J. Treacy, W.T. Koetser, C.B. de Gruyter, Proc. R. Soc. Lond. A
420 (1988) 375.
[55] K. Okumura, Y. Iwasawa, J. Catal. 164 (1996) 440–448.
[56] E.B. Lami, F. Fajula, D. Anglerot, T. des Courieres, Microporous Mater. 1 (1993)
237–245.
[57] M. Trejda, M. Ziolek, Y. Millot, K. Chalupka, M. Che, S. Dzwigaj, J. Catal. 281
(2011) 169–176.
[
34] S. Saravanamurugan, M. Paniagua, J.A. Melero, A. Riisager, J. Am. Chem. Soc. 135
(
2013) 5246–5249.
[
[
35] R. Gounder, M.E. Davis, ACS Catal. 3 (2013) 1469–1476.
36] W.R. Gunther, Y. Wang, Y. Ji, V.K. Michaelis, S.T. Hunt, R.G. Griffin, Y. Román-
Leshkov, Nat. Commun. 3 (2012) 1109.
37] R. Bermejo-Deval, R. Gounder, M.E. Davis, ACS Catal. 2 (2012) 2705–2713.
38] E. Taarning, S. Saravanamurugan, M.S. Holm, J. Xiong, R.M. West, C.H. Chris-
tensen, ChemSusChem 2 (2009) 625–627.
39] M.S. Holm, S. Saravanamurugan, E. Taarning, Science 328 (2010) 602–605.
40] M.S. Holm, Y.J. Pagán-Torres, S. Saravanamurugan, A. Riisager, J.A. Dumesic, E.
Taarning, Green Chem. 14 (2012) 702–706.
[
[
[
[
[
41] L. Bui, H. Luo, W.R. Gunther, Y. Román-Leshkov, Angew. Chem. Int. Ed. 52 (2013)
[58] M. Boronat, P. Concepción, A. Corma, M. Renz, S. Valencia, J. Catal. 234 (2005)
111–118.
[59] R. Bermejo-Deval, M. Orazov, R. Gounder, S.-J. Hwang, M.E. Davis, ACS Catal. 4
(2014) 2288–2297.
8
022–8025;
L. Bui, H. Luo, W.R. Gunther, Y. Román-Leshkov, Angew. Chem. 125 (2013)
180–8183.
8
[
[
[
42] J. Wang, S. Jaenicke, G.-K. Chuah, RSC Adv. 4 (2014) 13481–13489.
43] A. Corma, C. Paris, M. Moliner, Green Chem. 15 (2013) 2101–2109.
44] S. Dzwigaj, M. Peltre, P. Massiani, A. Davidson, M. Che, T. Sen, S. Sivasanker,
Chem. Commun. (1998) 87–88.
45] S. Dzwigaj, M. Che, J. Phys. Chem. B 110 (2006) 12490–12493.
46] S. Dzwigaj, T. Shishido, J. Phys. Chem. C 112 (2008) 5803–5809.
47] S. Dzwigaj, J.-P. Nogier, M. Che, M. Saito, T. Hosokawa, E. Thouverez, M. Mat-
[60] E.P. Parry, J. Catal. 2 (1963) 371–379.
[61] J.A. Boscoboinik, X. Yu, E. Emmez, B. Yang, S. Shaikhutdinvo, F.D. Fischer, J. Sauer,
H.-J. Freund, J. Phys. Chem. C 117 (2013) 13547–13556.
[62] E. Creyghton, S. Ganeshie, R. Downing, H. van Bekkum, J. Mol. Catal. A 115 (1997)
457–472.
[63] N. Li, W. Jia, Y. Zhang, F. Tan, J. Zhang, Int. J. Pharm. 415 (2011) 169–174.
[64] R. Gounder, M.E. Davis, AIChE J. 59 (2013) 3349–3358.
[65] Z. Qin, L. Lakiss, L. Tosheva, J.P. Gilson, A. Vicente, C. Fernandez, V. Valtchev,
Adv. Funct. Mater. 24 (2014) 257–264.
[
[
[
suoka, M. Anpo, Catal. Commun. 19 (2012) 17–20.
48] A. S´ r e˛ bowata, R. Baran, D. Łomot, D. Lisovytskiy, T. Onfroy, S. Dzwigaj, Appl.
[
Catal. B 147 (2014) 208–220.
[66] R. Gounder, M.E. Davis, J. Catal. 308 (2013) 176–188.