Preparation and characterization of mesoporous Cs2HPW 12O40 salt, active in transformation of m-xylene

Article abstract of DOI:10.1016/j.apcata.2012.09.052

The samples of Cs₂HPW₁₂O₄₀ salt were precipitated with CsCl, CsBr or CsI reagent as well as with commonly used Cs₂CO₃. The use of cesium halides resulted in the Cs ₂HPW₁₂O₄₀ samples of mesoporous structure composed of relatively loosely aggregated primary particles. It was observed that the type of halogen ion influenced textural properties of the Cs ₂HPW₁₂O₄₀ samples. As the atomic size of halogen ion increased (from Cl to I), the specific surface area and microporosity decreased. The so-obtained samples exhibited textural and morphological features similar to those of Cs_2.5H _0.5PW₁₂O₄₀ salt. In the transformation of m-xylene, the pore-size sensitive reaction, the catalytic activity of the Cs₂HPW₁₂O₄₀ samples prepared with CsBr and CsI reagents was about two-fold higher than that of Cs_2.5H _0.5PW₁₂O₄₀ salt. All these samples exhibited similar strength of acid sites. Therefore, high catalytic activity of the samples prepared with CsBr and CsI could be ascribed to their open pore structure, which allowed the accessibility of almost all active sites for m-xylene molecules.

Full text of DOI:10.1016/j.apcata.2012.09.052

Applied Catalysis A: General 450 (2013) 19–27
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Preparation and characterization of mesoporous Cs HPW O salt,
2
12 40
active in transformation of m-xylene
a,∗
a
b
a
a
L. Matachowski , A. Drelinkiewicz , R. Rachwalik , M. Zimowska , D. Mucha ,
M. Ruggiero-Mikołajczyka
a
Jerzy Haber Institute of Catalysis and Surface Chemistry Polish Academy of Sciences, Niezapominajek 8, 30-239 Kraków, Poland
Institute of Organic Chemistry and Technology, Faculty of Engineering and Chemical Technology, Cracow University of Technology, Warszawska 24,
b
31-155 Kraków, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 August 2012
Received in revised form
The samples of Cs HPW12O40 salt were precipitated with CsCl, CsBr or CsI reagent as well as with com-
monly used Cs2CO3. The use of cesium halides resulted in the Cs2HPW12O40 samples of mesoporous
2
structure composed of relatively loosely aggregated primary particles. It was observed that the type of
halogen ion influenced textural properties of the Cs2HPW12O40 samples. As the atomic size of halogen ion
increased (from Cl to I), the specific surface area and microporosity decreased. The so-obtained samples
exhibited textural and morphological features similar to those of Cs2.5H0.5PW12O40 salt. In the transfor-
mation of m-xylene, the pore-size sensitive reaction, the catalytic activity of the Cs2HPW12O40 samples
prepared with CsBr and CsI reagents was about two-fold higher than that of Cs2.5H0.5PW12O40 salt. All
these samples exhibited similar strength of acid sites. Therefore, high catalytic activity of the samples
prepared with CsBr and CsI could be ascribed to their open pore structure, which allowed the accessibility
of almost all active sites for m-xylene molecules.
2
8 September 2012
Accepted 29 September 2012
Available online 3 November 2012
Keywords:
Cesium heteropolysalts
Cesium halides
Mesoporous structure
Catalytic activity
m-Xylene transformation
© 2012 Elsevier B.V. All rights reserved.
1
. Introduction
the other hand, the specific surface area of Cs HPW12O40 sam-
2
ples prepared using cesium chloride or nitrate was reported to
2
2
The cesium salts of 12-tungstophosphoric acid have been widely
studied since over 30 years [1–5]. Among them Cs2.5H0.5PW12O40
be distinctly higher, ranging from 43 m /g [11] to 72 m /g [12].
This explained high catalytic activity of the latter samples in the
n-butane isomerization [12,13]. Similar effect was observed in our
and Cs HPW12O40 salts have attracted most researchers’ attention.
2
They both have structures built of Keggin units, which differ in the
fraction of protons replaced by the cesium cations. Till now, the
studies were mostly devoted to the Cs2.5H0.5PW12O40 salt because
of its outstanding activity in a number of acid centers catalyzed
reactions [6,7]. This was a result of Cs2.5H0.5PW12O40 high specific
surface area due to its micro- and mesoporous structure allowing
highly effective utilization of protons. However, the stoichiometry
recent studies [14]. Textural features of Cs HPW12O40 salt were
2
improved, when instead of Cs CO3 reactant, the CsCl was applied
2
as precipitating agent. The so-obtained Cs HPW12O40 salt had
2
2
specific surface area of about 78 m /g and partially mesoporous
structure (average pore diameter of about 3.5 nm) [14]. Results pre-
sented recently [14] allowed us to postulate that the composition
of supernatant solution upon Cs HPW12O40 precipitation, played
2
of the Cs HPW12O40 salt also could suggest its possible high activity
in acid centers catalyzed reactions.
an essential role in the process of primary crystallites aggregation.
It seemed to be very probable that Cl /HCl existing in colloidal
2
−
The Cs HPW12O40 salt could be precipitated from an aqueous
solution strongly influenced this process. The adsorption of chlo-
2
solution of H PW12O40 using different reagents, such as Cs CO ,
ride ions on the surface of primary particles of Cs HPW12O40 salt,
3
2
3
2
CsCl or CsNO . However, the application of Cs CO as the precipi-
tating agent yielded material of extremely low specific surface area.
facilitated the formation of mesoporous structure [14]. The influ-
ence of the type of an anion (nitrate, chloride or carbonate) in the
precipitating reagent on the specific surface area of ammonium
salt of tungstophosphoric acid was also observed by Lapham and
Moffat [15]. Furthermore, specific surface area, pore volume and
pore width were observed to be cation-dependent (Cs , K , Rb )
during precipitation of the M2.1H0.9PW12O40 salts by the appro-
priate carbonates [16]. Although, according to these authors this
effect is very interesting from the viewpoint of porosity control
3
2
3
Several researchers reported, that such prepared Cs HPW12O40 salt
2
had specific surface area (determined from adsorption–desorption
2
isotherm of nitrogen), which did not exceed 1 m /g [8–10]. On
+
+
+
∗ Corresponding author. Tel.: +48 126395186; fax: +48 124251923.
E-mail address: ncmatach@cyf-kr.edu.pl (L. Matachowski).
0
926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.09.052

2
0
L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
of these materials, no explanation concerning the role of cation
has been reported. According to Okamoto et al. [17] and Inumaru
the cesium cations in [1/4, 1/4, 3/4] position, was used. This fol-
lowed Brown et al. [1] who postulated that each cesium cation must
replace one H5O2+ ion. All crystal planes were taken into account
in the Rietveld refinement because in this “ab-initio” method the
diffractogram is fitted using the structure of the phase. The crystal-
lite sizes were calculated using the LVol-IB method.
[
18] the mechanism of the formation and growth of polyoxomet-
+
+
+
alates M PW12O40 (where M = Cs , Ag , NH ) aggregates is still
3
4
unclear and needs further examination. The course of nanocrystals
formation was suggested to be a self-organization process con-
sisting of three subsequent steps; formation of nanocrystallites,
self-assembly process to form aggregates and growth of particles
by the attachment of other nanocrystallites. The course of such pro-
cesses was observed to be strongly affected by the composition of
2.2.3. Surface area and porosity measurements
The specific surface areas of the Cs HPW12O40 samples as well
2
as Cs2.5H0.5PW12O40 salt were calculated using the BET method.
The nitrogen adsorption–desorption isotherms were obtained at
77 K using an Autosorb-1 (Quantachrome equipment). Prior to the
measurements, the samples were preheated and degassed under
vacuum at 473 K for 2 h. The micropores area of studied samples
was calculated by t-micropore analysis. The pore size distribution
in studied samples was obtained using BJH method.
+
+
solution and in particular by the nature of countercations (Cs , Ag ,
+
NH4 ) i.e. variables determining the dissolution/reprecipitation
processes.
Due to the interesting effects described above, our recently per-
formed studies [14] were extended to the use of CsBr and CsI
as precipitating agents for the preparation of Cs HPW12O salt.
2
40
The influence of ions existing in the solution during precipita-
tion process on the physicochemical and catalytic properties of the
2
.2.4. Scanning electron microscopy and X-ray energy dispersive
spectroscopy
The studies of the morphology of cesium samples were carried
Cs HPW12O40 salt was also studied. The ‘as synthesized’ samples
2
were characterized by BET, FT-IR, XRD and SEM techniques. Their
thermal stability (after annealing at 573 K for 2 h) was determined
with BET, FT-IR and XRD methods. The catalytic activity of the
out by means of the JEOL JSM - 7500F Field Emission Scanning
Electron Microscope with the EDXS (energy dispersive X-ray spec-
troscopy) detection system of characteristic X-ray radiation. The
K575X Turbo Sputter Coater was used for coating the specimens
with chromium.
Cs HPW12O40 samples was tested in the pore-sensitive transfor-
2
mation of m-xylene. Catalytic performance of so-obtained samples
was compared with those of Cs HPW12O40 samples prepared using
2
Cs CO3 and CsCl as well as with that of Cs2.5H0.5PW12O40 salt pre-
2
cipitated using cesium carbonate.
2.2.5. Thermal stability
The ‘as synthesized’ Cs (Cl), Cs (Br) and Cs (I) samples were
2
2
2
2
. Experimental
annealed at 573 K for 2 h in the air atmosphere. The annealed sam-
ples are denoted as Cs (Cl)-573 K, Cs (Br)-573 K and Cs (I)-573 K,
2
2
2
2.1. Preparation procedure
respectively. Their structures as well as textural properties were
investigated with FT-IR, XRD and BET methods.
A
commercially available 12-tungstophosphoric acid
(
H PW12O40·24H O, HPW, Sigma–Aldrich) and Cs CO (Aldrich),
3
2
2
3
2
.2.6. Determination of acid sites strength
Acid sites strength was determined with Hammett indicators
CsCl (POCh S.A), CsBr (Fluka) and CsI (Aldrich) were used to prepare
the Cs HPW12O40 samples. In standard procedure the stoichio-
2
by means of procedure used by other authors [21,22]. Approxi-
mately 25 mg of sample was shaken with 1 cm3 of a solution of
Hammett indicators diluted in methanol and left to equilibrate
for 2 h after which no further color changes were observed. The
color of the sample was then noted. The following Hammett
indicators were used: 4-(dimethylamino)azobenzene (pKa = +3.3),
metric quantities of cesium carbonate (0.4 M) or other cesium
salts (CsCl, CsBr, CsI) were added to the aqueous solutions of
HPW (0.1 M). Then, the obtained colloidal solutions of precipitated
samples were slowly evaporated to dryness overnight in the oven
at 313 K. The Cs HPW12O40 samples precipitated with Cs CO ,
CsCl, CsBr and CsI were denoted as Cs (C), Cs (Cl), Cs (Br) and
Cs (I), respectively. For the comparison, the Cs2.5H0.5PW12O40 salt
was also precipitated using cesium carbonate (denoted in the text
as Cs2.5(C)).
2
2
3
2
2
2
4
-phenylazodiphenylamine (pKa = +1.5), crystal violet (pKa = +0.8),
2
dicinnamalacetone (pKa = −3), anthraquinone (pKa = −8.2), 4-
nitrotoluene (pKa = −11.35), 3-nitrotoluene (pKa = −11.99),
1
-chloro-4-nitrobenzene (pKa = −12.70), 1-chloro-3-nitrobenzene
(
pKa = −13.16), 1.4-dinitrotoluene (pKa = −13.75). The acid
2
.2. Characterization of the catalysts
strength was quoted as being stronger than the weakest indicator
which exhibited a color change but weaker than the strongest
indicator that gave no change. The measurements of acid sites
strengths were performed for the ‘as synthesized’ Cs (Cl), Cs (Br)
2
.2.1. FT-IR spectroscopy
The FT-IR spectra were recorded using Nicolet 380 FT-IR
2
2
spectrometer with standard KBr pellets technique. Prior to mea-
surements, the samples were dried at 363 K in order to remove
crystallization water.
and Cs (I) as well as for the annealed Cs (Br)-573 K and Cs (I)-573 K
samples.
2
2
2
2.3. Catalytic experiments
2.2.2. XRD diffraction patterns and Rietveld refinement
The X-ray diffraction patterns obtained at ambient tempera-
ture were recorded in the range of 2ꢀ = 5–90 , step size 0.02 and
2 s/step using the Siemens D5005 diffractometer (CuK␣ radia-
The transformation of m-xylene was performed in a tubu-
◦
◦
lar down-flow stainless steel reactor with 100 mg of catalyst
(200–315 ␮m fraction) and 2 cm3 of SiC chips (>400 ␮m fraction).
The catalysts were activated in helium flow at 573 K for 1 h. The
tests were carried out in temperature range of 498–548 K in flow-
ing helium and in the pulse mode, in order to avoid an excess of
coke formation.
The analysis of substrate and products was performed by a
Hewlett-Packard 6890 gas chromatograph connected on-line to the
microreactor system, using a packed column (3 m, 2.0 mm i.d.) with
1
tion, 40 kV, 40 mA) equipped with the secondary beam graphite
monochromator. In order to remove the crystallization water,
before measurements the samples were heated overnight at 363 K.
Full-matrix Rietveld refinement performed by TOPAS program was
applied to calculate lattice parameters a and crystallite sizes of
cesium samples [19,20]. As a starting model, H PW12O40·6H O
3
2
+
(
space group Pn-3m (2 2 4)) modified by replacing the H5O2 by

L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
21
Cs (I)-573K
2
Cs (I)-573K
2
Cs (Br)-573K
2
Cs (Br)-573K
2
'
as synthesized' Cs (I)
2
'
as synthesized' Cs (I)
2
'
as synthesized' Cs (Br)
2
'
as synthesized' Cs (Br)
2
4
000
3000
2000
1000
10 20 30 40 50 60 70 80 90
-
1
2Θ
Wavenumber, cm
Fig. 2. XRD diffraction patterns of ‘as synthesized’ and annealed Cs2(Br) and Cs2(I)
samples.
Fig. 1. FT-IR spectra of ‘as synthesized’ and annealed Cs2(Br) and Cs2(I) samples.
8
% Bentone-34, 6% didecyl phthalate and 1% silicon oil A on Chro-
mosorb W (60–80 mesh) and a TCD detector. The response factors
of the detector were established using the appropriate aromatic
hydrocarbons (reagent grade) and their mixtures as the standards.
All reaction parameters were kept constant during the catalytic
tests to maintain maximum reproducibility. Carbon and mass bal-
ances were kept below ± 5%.
of bands characteristic for Keggin anion does not change and their
intensities are almost the same as those observed in the spectra of
the ‘as synthesized’ samples.
The XRD diffraction patterns registered for ‘as synthesized’
Cs (Br) and Cs (I) samples (Fig. 2) are in good agreement with
2
2
this obtained for the C HPW12O40 salt, precipitated with a com-
2
mon procedure using Cs CO [27,28], as well as with this of Cs (Cl)
2
3
2
reported in our previous paper [14]. Moreover, diffractions patterns
recorded for the annealed Cs (Br)-573 K and Cs (I)-573 K samples
3
. Results and discussion
2
2
are almost identical as those of the ‘as synthesized’ samples (Fig. 2).
It can be also underlined that no reflections originating from other
forms of heteropoly acid are observed. This demonstrates that the
∼
All samples were precipitated at pH = 1 because this factor
strongly influences the formation of various heteropolyacid struc-
tures [23]. The structure of the ‘as synthesized’ Cs (Br) and Cs (I)
crystal structure of studied Cs (Br) and Cs (I) samples is stable up
2
2
2
2
samples was verified by FT-IR and XRD methods. The FT-IR spectra
displayed in Fig. 1 confirmed that Keggin structure remained intact
in prepared samples.
to 573 K. Similar data was reported by Okuhara et al. [27] for the
Cs (C) sample evacuated at 573 K. Details concerning structural
2
parameters of both investigated samples are presented in Table 1.
The lattice parameters a are the same for the ‘as synthesized’
Cs (Br) and Cs (I) samples and they agree with the values of param-
On all FT-IR spectra the set of bands located at 1079, 985, 888
−1
and at ca. 800 cm can be seen (Fig. 1). They are characteristic for
Keggin unit and originate from the stretching vibrations of P Oi,
2
2
eter a reported in our previous paper for Cs (Cl) and Cs (C) samples
2
2
W
Ot, W Oc W or W Oe W, where oxygen within Keggin anions
(1.1879 nm and 1.1882 nm, respectively) [14]. Moreover, these val-
ues are also in good agreement with the other literature data
[27,28]. As seen in Table 1, the size of crystallites is similar in
the ‘as synthesized’ Cs (Br) and Cs (I) samples, 13.8 and 14.5 nm,
is denoted as O (internal), Ot (terminal), Oc (corner-sharing) and Oe
i
−
1
(
edge-sharing), respectively. The strong band located at 1079 cm
3−
associated with the presence of the [PW12O40
]
anion is clearly
2
2
visible in the spectra of both Cs (Br) and Cs (I) samples, whereas
respectively. However, they are smaller than those of the Cs (C)
2
2
2
no bands originating from other heteropolyacid structures appears.
and Cs (Cl) samples, 20 and 25 nm, respectively [14]. Although the
2
−
1
The band composed of two superimposed maxima at 993 cm and
crystallite sizes vary to some extent, there is no clear relationship
between the type of used precipitating agent and calculated size of
crystallites.
9
85 cm can be assigned to the W O group interacting with Cs+
−
+
1
and H O ions, respectively. This splitting is characteristic for acidic
3
cesium salts [24,25] and was previously also observed in the spectra
of K, Ag and Tl salts of HPW [26]. Taking into account previously
A thermal treatment of the Cs (Br) and Cs (I) samples does
not induce essential changes in the lattice parameter a (Table 1)
2
2
reported FT-IR spectra for Cs (C) and Cs (Cl) samples [14], it can be
similarly as in the Cs (Cl) sample annealed at 573 K [14]. Alike,
2
2
2
stressed that for all studied samples of the Cs HPW12O40 salt, the
other authors did not observe changes for the Cs (C) sample ther-
2
2
Keggin structure is preserved independently of used precipitating
agent.
mally treated under vacuum (573 K [27]) and (473 K [28]). On the
other hand, after calcination of the Cs (Br) and Cs (I) samples,
2
2
The FT-IR spectra were also recorded for the Cs (Br)-573 K
their crystallite size increases to about 17 nm in both Cs (Br)-
2
2
and Cs (I)-573 K samples, which were obtained after annealing
573 K and Cs (I)-573 K samples (Table 1). It can be suggested that
2
2
at 573 K for 2 h, i.e. at the temperature used for their thermal
pre-treatment prior to catalytic tests. As seen in Fig. 1 the positions
the growth of crystallite sizes is caused by recrystallization pro-
cess during annealing of the samples up to 573 K. Thus, it can be

2
2
L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
Table 1
Structural parameters of ‘as synthesized’ and annealed Cs2(Br) and Cs2(I) samples.
Sample
Lattice parameter a [nm]
Crystallites size [nm]
Relative intensity (%) of reflections in
proportion to 2ꢁ = 26.01 (2 2 2)
◦
◦
◦
2
ꢀ = 10.54 (0 1 1)
2ꢀ = 30.12 (0 0 4)
Cs2(Br) ‘as synthesized’
Cs2(Br)-573 K
Cs2(I) ‘as synthesized’
Cs2(I)-573 K
1.1867
1.1852
1.1862
1.1852
13.8
17.1
14.5
17.0
49.3
51.3
45.3
47.5
31.3
29.9
31.1
30.2
concluded that during recrystallization process, the smaller crys-
tallites are enlarged, what causes a decrease of a specific surface
area of the annealed samples. Similar recrystallization process was
isotherm obtained for the Cs (Cl) sample exhibited similar shape
to the above discussed samples [14].
2
As seen in Fig. 3, the isotherms obtained for the annealed
also observed in the case of Cs (Cl) sample annealed at 573 K [14].
Cs (Br)-573 K and Cs (I)-573 K samples still exhibit hysteresis
2 2
2
Table 1 shows the examples of calculated relative intensities of
loops, characteristic for mesoporous structures. Thus, despite of
the recrystallization process, which causes the increase in crystal-
lite sizes, both thermally treated samples still possess mesoporous
structure.
◦
◦
the reflections at 2ꢂ = 10.54 (0 1 1) and at 2ꢂ = 30.12 (0 0 4), taking
◦
into account the intensity of the strongest reflection at 2ꢂ = 26.01
(
2 2 2) as the reference (100%). Slight differences between the ‘as
synthesized’ samples and those after annealing at 573 K for 2 h
are observed, what confirms the thermal stability of the crystalline
structure in the investigated samples.
The textural characteristics of the ‘as synthesized’ and
annealed Cs HPW12O40 samples were calculated from the nitro-
2
gen adsorption–desorption isotherms and collected in Table 2. For
The nitrogen adsorption–desorption isotherms of the “as syn-
thesized” Cs (Br) and Cs (I) samples are displayed in Fig. 3. They
the clarity of discussion, some data obtained for Cs (C) and Cs (Cl)
samples and presented in our previous paper [14] are also given.
2 2
2
2
are quite different from that obtained for nonporous Cs (C) sample
The specific surface area of the Cs (C) sample is as low as ca.
2
2
2
but are similar to the isotherm observed for mesoporous Cs2.5(C)
salt [25]. All these isotherms exhibit hysteresis loops and accord-
ing to IUPAC classification are IV-H3 type, which is characteristic of
disordered, lamellar pore structure with ‘slit & wedge’ shape pores.
They can be observed mainly in the case of very wide capillaries
having narrow openings and in the case of an interstice between
the parallel plates. The H3 type isotherm was also previously
observed for crystalline metal oxide aggregates of sheet-like shape
1 m /g, whilst those of the Cs (Br) and Cs (I) samples are much
2
2
2
2
larger, 48.5 m /g and 39.5 m /g respectively. The surface area of
78.2 m /g for the Cs (Cl) sample, presented in our previous paper
2
2
[14], is the highest among all three samples prepared with cesium
halides. All of these surface areas are dramatically higher in com-
parison to that for the Cs (C) sample. It can be also observed that
2
the specific surface area, micropores and mesopores areas decrease
when going from Cl to I (Table 2) whilst the contribution of meso-
pores area in the specific surface area increases in the following
[
29]. It should be stressed that the nitrogen adsorption–desorption
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
60
50
40
30
20
10
0
Cs (I)
Cs (Br)
2
2
Cs (I)-573K
2
Cs (Br)-573K
2
0
,0
0,2
0,4
0,6
0,8
1,0
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure, p/po
Relative pressure, p/po
1
1
1
40
20
00
40
30
20
10
0
Cs (Cl)
2
Cs (Cl)+HBr
8
6
4
2
0
0
0
0
0
2
Cs (C)+HBr
2
Cs (C)
2
0
,0
0,2
0,4
0,6
0,8
1,0
0,0
0,2
0,4
0,6
0,8
1,0
Relative pressure, p/po
Relative pressure, p/po
Fig. 3. The nitrogen adsorption–desorption isotherms obtained for ‘as synthesized’ and annealed Cs2(Br) and Cs2(I) samples. The isotherms for Cs2(Cl) and [Cs2(Cl) + HBr]
samples as well as for the Cs2(C) and [Cs2(C) + HBr] samples are also presented.

L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
23
Table 2
Sorption properties of Cs2HPW12O40 salt precipitated with different reagents.
Sample
Ionic radius of
halogena [nm]
Surface area
[m /g]
Micropores
area** [m /g]
Micropores
volume** [cc/g]
Mesopores
area [m /g]
Mesopores area/specific
surface area [%]
2
2
2
Cs2(C)b
1
78.2
4.2
48.5
24.9
39.3
30.0
73.1
39.8
–
–
–
45.0
2.9
42.1
18.6
36.1
23.7
64.9
23.0
–
Cs2(Cl)b
0.167
0.182
0.206
33.2
1.3
6.4
6.3
3.2
6.3
8.2
16.8
0.017
0.006
0.005
0.004
0.003
0.004
0.010
0.015
57.5
69.0
86.8
74.7
91.9
79.0
88.8
57.8
Cs2(Cl)-573 K
Cs2(Br)
Cs2(Br)-573 K
Cs2(I)
Cs2(I)-573 K
[
[
Cs2(Cl) + HBr]
Cs2(Cl) + HBr]-573 K
The values denoted (**) were calculated by t-micropore analysis.
a
Data from Ref. [30].
Data from Ref. [14].
b
order Cl < Br < I. A similar dependence is observed in the case of the
annealed samples (Table 2). This is important from catalytic point
of view.
salt. All the results also clearly show that the type of halogen ion
plays an important role in textural properties of the Cs HPW12O40
2
salt. To confirm the influence of the type of halogen ions on the
The pore size distribution diagrams (Fig. 4) demonstrate that
Cs (Br) and Cs (I) samples exhibit mesoporous structure. In both
formation of mesoporous structure of the Cs HPW12O40 salt, addi-
tional experiments were performed. To the colloidal solution of
2
2
2
samples apart from the mesopores of 3.8 nm in diameter, also the
wider ones of diameter within 6–10 nm significantly contribute
to their porous structure. It is also interesting that the pore size
of 3.8 nm observed in the Cs (Br) and Cs (I) samples is similar to
Cs (C) and Cs (Cl) samples, the HBr acid (at the stoichiometric
amount) was added and the obtained colloidal suspension was
further treated by a common procedure.
2 2
The nitrogen adsorption–desorption isotherm of the
2
2
pore size of 3.6 nm existing in the structure of Cs2.5(C) salt. Accord-
ing to Okuhara et al. [27] the structure of Cs2.5H0.5PW12O40 salt is
consisted of the mesopores with voids and necks. The size of the
latter is about 3 nm in width. Therefore, the formation of mesopores
[Cs (C) + HBr] final sample significantly differs from the start-
2
ing nonporous Cs (C) salt (Fig. 3). Moreover, the isotherm of
2
[Cs (C) + HBr] sample is similar to those obtained for the samples
2
precipitated with CsBr or CsI reactants with characteristic loops
derived from the mesoporous structures. Thus, it can be stressed
of such shape cannot be excluded in the samples of Cs HPW12O40
2
salt precipitated by cesium halides. However, it was previously
that a simple addition of HBr to Cs (C) salt can entirely change its
2
described that in the Cs (Cl) sample the micropores (ca. 1.6 nm)
and mesopores (6.2 nm) exist [14].
compact structure into mesoporous one. This structural modifica-
tion is also reflected in the increase of specific surface area from
2
2
2
Above analysis confirms that the application of the CsBr or CsI
1 m /g for ‘as synthesized’ Cs (C) salt to 34.8 m /g observed for the
2
reagent as the precipitating agents results in the Cs HPW12O40
[Cs (C) + HBr] sample.
2
2
samples with mesoporous structure, similar to that of the Cs2.5(C)
In the case of the Cs (Cl) and [Cs (Cl) + HBr] samples the nitrogen
2
2
adsorption–desorption isotherms obtained for both samples are
almost identical (Fig. 3). The specific surface area of [Cs (Cl) + HBr]
2
2
sample is close to that of the Cs (Cl) salt, 73.1 and 78.2 m /g, respec-
2
tively. However, interesting changes of the textural properties are
3
.6 nm
observed after addition of HBr to the colloidal solution of the Cs (Cl)
2
3
.8 nm
Cs (C)
salt (Table 2). The significant decrease of micropores area from
2
.5
2
3
3.2 to 8.2 m /g accompanied by the increase of mesopores area
2
[
Cs (C) + HBr]
from 45.0 to 64.9 m /g can be noticed. It means that the pres-
2
ence of bromide ions in the colloidal solution of the Cs (Cl) salt
2
Cs (I)-573K
strongly influences the porosity of the obtained sample. It is also
confirmed by the ratio of mesopores area/surface area calculated
2
for the [Cs (Cl) + HBr] sample, which is higher than that for par-
2
ent Cs (Cl) salt, 88.8 and 57.5%, respectively. On the other hand,
this ratio obtained for the former is almost the same as that for the
2
Cs (I)
2
Cs (Br) salt, 86.8% (Table 2).
2
Cs (Br)-573K
2
Interesting results were obtained after annealing of
[
Cs (Cl) + HBr] sample at 573 K, evidenced by the decrease of
2
Cs (Br)
2
2
surface area from 73.1 to 39.8 m /g. Moreover, the micropores
area increased from 8.2 to 16.8 m /g whilst the mesopores area
decreased from 64.9 to 23 m /g (Table 2). This change of porosity
2
Cs (Cl)-573K
2
2
1
.6 nm
can be caused by the removing of bromide ions from the sample
during annealing. It should be underlined that the treatment of
sample before BET measurements i.e. 473 K for 2 h in vacuum
can cause also partial removal of bromide ions. Especially, it can
influence the porosity of ‘as synthesized’ sample. Despite that it
can be concluded that the presence of HBr in the colloidal solution
6
.2 nm
Cs (Cl)
2
5
10
15
20
Pore diameter, nm
of Cs (Cl) salt influences the formation of final sample and in this
2
way its textural properties.
The ICP-AES method was applied to determine the contents of
halogen in studied samples. The Br/Cs and I/Cs atomic ratios in ‘as
Fig. 4. The pore size distribution diagrams for ‘as synthesized’ and annealed Cs2(Cl),
Cs2(Br), Cs2(I) samples. The diagrams of [Cs2(C) + HBr] and Cs2.5(C) are also pre-
sented.

2
4
L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
−
synthesized’ Cs (Br) and Cs (I) samples were close to the stoichio-
metric ratio being near to 1. After annealing the samples at 353 and
be expected that interactions of Cl ions with the surface protons
2
2
−
of primary particles could be the strongest whereas those of I ions
573 K both atomic ratios decreased dramatically to about 0.046 and
0.012, respectively. It demonstrates that already during heating at
353 K almost all bromide or iodide ions were removed. These results
the weakest. It was previously showed that the addition of chloride
ions to the Cs (C) sample strongly influenced its structure [14]. The
2
presence of chloride ions in the colloidal solution of [Cs (C) + HCl] or
2
clearly confirm the previous conclusion that halogen ions play a
vital role in the formation of porous structure of studied samples
during their synthesis.
Cs (Cl) samples caused the formation of larger colloidal particles in
2
both cases [14]. This could be a result of interactions of chloride ions
with primary particles of these samples before aggregation process
or penetration of chloride ions into already formed colloidal parti-
The morphology of Cs (Br), Cs (I), [Cs (C) + HBr] and Cs2.5(C)
2
2
2
samples was studied by scanning electron microscopy technique
and obtained micrographs are displayed in Fig. 5. All samples, which
are characterized by mesoporous structure, form spherical par-
cles of the Cs (C) sample. Similar phenomenon was observed after
2
addition of HBr to the Cs (C) sample.
2
However, the addition of HBr to the colloidal solution of Cs (Cl)
2
ticles consisting of much smaller primary particles. The Cs (Cl)
sample also exhibits similar morphology, showing loosely aggre-
gated almost spherical particles composed of small nanocrystallites
sample practically did not change the specific surface area whereas
some changes in the porosity appeared, as was described above
(Table 2). Because the energy of H Cl bond is much higher than that
of H Br bond the exchange of chloride ions by bromide ions can be
2
[
14]. Hence, the morphology of all samples prepared with cesium
halides differ entirely from that of the Cs (C) sample but is similar
excluded. It seems that the results obtained for the [Cs (Cl) + HBr]
2
2
to that of the Cs2.5(C) salt [14,27]. According to the literature data
sample can indicate the co-absorption of bromide ions, when they
are present in the colloidal solution. This could be a result of
the penetration of bromide ions into colloidal particles. Such pro-
[
27] and our previous results, a highly compact structure of the
Cs (C) sample possesses smooth surface and any primary particles
2
cannot be noticed. However, the Cs (C) salt after addition of HBr
cess could change the porosity of [Cs (Cl) + HBr] sample, which is
2
2
can completely change its morphology because of the formation of
sphericalparticlescomposedof aggregatedprimaryparticleswhich
reflected by the increase of mesopores area accompanied by the
decrease of micropores area. However, this phenomenon seems to
be more complicated and its detailed explanation requires further
studies.
are observed in [Cs (C) + HBr] sample (Fig. 5). It can be supposed
2
−
that after the precipitation process of the Cs (C) sample, the Br
2
ions can penetrate and decompose its compact structure and after
self-organization process, can form mesoporous structure consist-
ing of spherical particles. This shows how halides ions can influence
Thus, it seems to be obvious that halide ions play an important
role in the aggregation of primary particles influencing the forma-
tion of mesoporous structure of the studied samples. On the other
hand, taking into account the ionic radius of halogens (Table 2), it
the formation of mesoporous structure of Cs HPW12O40 salt.
2
−
Our previous studies [14] showed that Cl ions were easily
is easy to explain how they influence the porous structure of the
−
removed from the ‘as synthesized’ Cs (Cl) sample. It was demon-
Cs HPW12O40 salt. The presence of the smallest Cl ions causes that
2
2
strated that this sample was ‘free’ of chloride ions after heating at
the structure with the highest contribution of micropores is formed.
−
−
4
53 K. The Cl atoms were also not detected by EDXS-SEM technique.
When Br or I ions interact with primary particles in Cs (Br) and
2
This suggested that vacuum pre-treatment of the sample prior to
microscopic studies caused the release of chloride ions. In the case
Cs (I) samples, the aggregated structure of the highest contribution
of mesopores is created (Table 2).
2
of Cs (Br) and Cs (I) samples expected values of Br and I atoms
After calcination of the Cs (Cl), Cs (Br) and Cs (I) samples at
2
2
2 2 2
were equal to about 5 and 8 wt%, respectively. However, their con-
tent detected by EDXS measurements was much lower. It has been
assumed that Br and I ions stayed to some extent in the structures
of both samples because their higher atomic masses compared to
chloride ions. This is consistent with the ICP-AES results presented
above which showed much higher content of halogens in the ‘as
synthesized’ samples and practically lack of halogens after their
annealing at 573 K.
573 K for 2 h, their specific surface area decreases and also some
changes in the pore distribution diagrams can be observed (Fig. 4).
Thermal treatment induces the most dramatic changes in the struc-
ture of the Cs (Cl) sample. Strong reduction in surface area (from
2
2
78.2 to 4.2 m /g) is accompanied by remarkable changes in the
porous structure. As Fig. 4 shows, wide mesopores with the max-
imum of about 6.2 nm in diameter vanish almost completely after
calcination, whereas some new narrower mesopores of 3.8 nm in
The role of halogen ions could be explained using a model
described by Okuhara et al. [27] which suggested that the pore
diameter are created. For the annealed Cs (Br) and Cs (I) samples,
2 2
the micropores structure (volume and area) practically does not
change and mesopores of 3.8 nm in diameter are preserved. The
contribution of the larger mesopores, of diameter within 6–10 nm
structure of CsxH3 PW12O40 salt precipitated with Cs CO3 was
−x
2
greatly dependent on the Cs/Keggin anion ratio. The final tex-
ture of these samples was formed via slow aggregation of primary
crystallites during heating and evaporation of the solution. The
crystalline sample of the Cs2.5H0.5PW12O40 was consisted of loosely
aggregated fine crystallites forming both micro- and mesopores.
In contrast, the Cs HPW12O40 salt (prepared with Cs CO ) was
range, decreases to some extent especially in the case of Cs (Br)
2
sample. For the latter sample, ca. two-fold decrease in surface
2
area appears (from 48.5 to 24.9 m /g). The changes are defini-
tively less pronounced for the Cs (I) sample as its surface decreases
2
2
only slightly from 39.3 to 30.0 m /g. The recrystallization process
2
2
3
consisted of dense aggregates of fine crystallites, what resulted in
a lack of porosity.
According to our previous results, at the beginning of acidic
cesium salts precipitation in water solution, primary particles are
formed and their surfaces are enriched by the protons [31,32].
resulting in the formation of larger crystallites (Table 1) may be
responsible for the observed structural changes. Alike, the meso-
pores of 3.6 nm in diameter are present in the Cs2.5H0.5PW12O40
salt (Fig. 4). As previously described, the mesoporous structure of
the Cs2.5(C) is stable under heating at 573 K in vacuum [27]. Data
presented by Zhang et al. [34] demonstrated that the mesoporous
When the Cs HPW12O40 salt is precipitated by Cs CO gaseous CO
2
2
3
2
is liberated from the colloidal solution, whilst when CsCl, CsBr or
CsI are used as the reagents, the HCl, HBr or HI may stay in the solu-
tion influencing the aggregation of primary particles. Therefore, the
structure of CsxH3 PW12O40 samples with x ≥ 2.2 did not change
−x
even after catalytic reaction performed at 653 K.
At this point, a question may be raised as to why the mesopores
of 3.6–3.8 nm in diameter are resistant for annealing at 573 K in
−
−
−
interaction of Cl , Br or I ions with surface protons of primary
particles should be taken into consideration. When ionic radius
Cs (Br), Cs (I) as well as in Cs2.5(C) samples. It was reported that in
2
2
−
−
increases from Cl to I (Table 2), the energy of H Cl > H Br > H
I
the structure of the Cs2.5H0.5PW12O40 salt the closest packed aggre-
gates formed the voids and necks, the latter with the size of about
bond decreases (432, 366 and 298 kJ/mol, respectively [33]). It can

L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
25
Fig. 5. The representative SEM images of prepared samples of Cs2HPW12O40 and Cs2.5H0.5PW12O40 salts (magnification of 100,000).
3
nm [27]. In turn, here in cesium halide-derived Cs HPW12O40
the Cs2(Br) and Cs2(I) samples, their mesoporous structures are par-
2
samples the mesopores of ca. 3.8 nm as well as the wider meso-
pores of 6–10 nm are observed (Fig. 4). The former may be related to
the closest packed aggregates similarly to what is proposed for the
Cs2.5H0.5PW12O40 salt. These mesopores are resistant for anneal-
ing at 573 K. The latter can be ascribed to loosely sticking primary
particles, which form thermally unstable structure. In the case of
tially preserved and both catalysts are active under used conditions
of catalytic test. These samples possess the pore structure charac-
terized by the predominant pore diameters of 3.8 nm (Fig. 4), which
is high enough to accommodate the m-xylene molecule of 0.64 nm
in size. Alike, the Cs2.5(C) salt with the mesoporous structure con-
sisting of pore of 3.6 nm in diameter is also active in investigated
reaction.
Cs (Cl) sample, the formation of the closest packed aggregates after
2
heating at 573 K probably is the reason that mesopores with 3.8 nm
in diameter are created.
The catalytic activity of studied cesium salts was tested in the
transformation of m-xylene which is catalyzed by acid centers
6
5
60
(
Scheme 1). Therefore, the acid sites strength of studied samples
5
5
4
5
0
5
was determined with Hammet indicators by means of procedure
used by other authors [21,22]. In both Cs (Br) and Cs (I) samples
2
2
the strength of acid centers was determined as −13.2 < H < −13.8,
0
what corresponds with the data reported by Okuhara et al. for
Cs (C) and Cs2.5(C) samples [21]. Similar acid sites strength was
2
determined also for the ‘as synthesized’ Cs (Cl) as well as for the
40
2
annealed Cs (Br)-573 K and Cs (I)-573 K samples.
2
2
3
3
5
0
Transformation of m-xylene is frequently studied as a model
reaction for zeolites and molecular sieves e.g. catalysts of various
porosity characteristics [35]. Thus, the catalytic performance of the
Cs (Cl), Cs (Br) and Cs (I) samples was compared with those of
2
2
2
25
20
the Cs HPW12O40 and Cs2.5H0.5PW12O40 salts prepared by a com-
2
2
mon procedure using Cs CO and exhibiting surface area of 1 m /g
and 136.7 m /g, respectively. The obtained conversion of m-xylene
against reaction time is plotted in Fig. 6. As described in Section
2
3
2
Cs (Br)
1
1
5
0
5
2
2
2
, before the catalytic test the samples were activated at 573 K for
h.
Cs (I)
2
Cs (C)
2.5
The obtained data show that Cs (C) sample of nonporous struc-
2
ture is inactive. The Cs (Cl) sample exhibits very low activity and
500
510
520
530
540
550
2
the conversion of m-xylene attains 3% only at reaction tempera-
Temperature, K
ture as high as 573 K. This low activity may be ascribed to very low
2
specific surface area of the Cs (Cl)-573 K sample (4.2 m /g, Table 2)
2 2
Fig. 6. Comparison of catalytic activity of Cs (Br), Cs (I) and Cs2.5(C) samples. The
2
due to highly compact structure. In turn, after thermal activation of
conversion of m-xylene as a function of reaction temperature.

2
6
L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
Scheme 1. Simplified reaction scheme of m-xylene transformation.
As Fig. 6 demonstrates within the whole temperatures range, the
Cs (Br) and Cs (I) are much more active catalysts than the Cs2.5(C)
salt, which is generally considered to be excellent ‘superacid’ cat-
that of disproportionation was determined to be 0.5 on all cesium
salts, being of similar order to that on HY zeolite [37]. The ratio
of 1,2,3- to 1,3,5-trimethylbenzenes was equal to 0.29 (at 548 K),
being close to the thermodynamic equilibrium value of 0.33. This is
an analogous relation as that observed in the presence of faujasite
and Linde type SK-45 zeolite [35]. It also shows that the available
void volumes in the structure of Cs (Br), Cs (I) and Cs2.5(C) cata-
2
2
alyst. At reaction temperature of 498 K the activity of Cs (Br) and
2
Cs (I) is about 3 times higher than that of Cs2.5(C) salt. At higher
2
temperatures, 523 K and 548 K, this ratio attains about 2.1 and 1.6,
respectively. Consequently, the activities of Cs (Br) and Cs (I) sam-
2
2
2
2
ples are roughly estimated to be about two-fold higher than that
of the Cs2.5(C) salt. As the mass of catalysts in the reactor is prac-
tically the same, this clearly demonstrates two-fold higher activity
lysts are sufficient to accommodate the trimethyldiphenylmethane
an intermediate complex of the disproportionation reaction. Thus,
similar selectivity pattern on zeolites and on Cs (Br), Cs (I) as well
2
2
of the Cs HPW12O40 salt, synthesized with CsBr or CsI reagent, per
as Cs2.5(C) catalysts clearly shows that the openings in mesopores
existing in studied cesium salts could not be narrower than those in
the above mentioned zeolites (window sizes of 0.735 and 0.750 nm,
respectively) [38,39].
2
1
g of catalyst than that of the Cs2.5H0.5PW12O40 salt. Taking into
account comparable strength of acid sites in the studied catalysts,
the obtained almost quantitative activity relation may be explained
by a two-fold higher number of active centers (protons) in the
Cs HPW12O40 compared to that in the Cs2.5H0.5PW12O40 salt. It
2
4. Conclusions
should be also underlined that the catalytic activity of [Cs (C) + HBr]
2
sample is similar to that of Cs (Br) salt.
The samples of Cs HPW12O40 salt were precipitated with CsCl,
2
2
In the transformation of m-xylene three main reactions pathway
can be generally observed: isomerization to para- and orto-
xylene, disproportionation into toluene and trimethylbenzenes
and dealkylation to form benzene and toluene [36]. The products
of all these reactions were observed in the presence of studied
cesium salts. The ratio of p/o-xylene isomers, the selectivity of
isomerization versus disproportionation and the distribution of
trimethylbenzenes isomers are the experimental values commonly
considered to be indications of diffusion restrictions (pore-size and
pore-shape limitations) being a result of pore structure in the cat-
alyst [37].
CsBr, CsI and with a common procedure using Cs CO . The use
of cesium halides as precipitating agent causes the formation
of loosely aggregated, partially mesoporous structures of the
2
3
Cs HPW12O40 samples. The type of halogen ions influences the
2
contribution of mesopores in the Cs HPW12O40 samples, which
2
increases when the atomic size of halogen grows in the order
Cl < Br < I. It is also suggested that interaction of halogen ions with
the surface protons of primary particles, before their aggregation
affects the formation of mesoporous structure. The textural and
morphological features of so-obtained samples are similar to those
of the Cs2.5H0.5PW12O40 salt.
The p/o-xylene ratio is predicted by the thermodynamic equi-
librium to be 1.05 (at 623 K) [37]. The higher values reflect
the difference in diffusion of these compounds, due to pore-
size limitations. m-Xylene disproportionation being a bimolecular
reaction involving bulky trimethyldiphenylmethane intermediates
occurs only when an intermediate molecule can be accommo-
dated into the pores, as in the case of zeolites with supercages of
Thermal treatment of cesium halides-derived Cs HPW12O40
2
samples at 573 K for 2 h changes their mesoporous structures to
some extent only. After calcination the wider mesopores (6–10 nm)
vanish almost completely. In Cs (Br) and Cs (I) samples the meso-
2
2
pores of 3.8 nm in diameter are still present whereas in the case of
the Cs (Cl) sample they appeared. Thus, the texture of the Cs (Br)
2
2
and Cs (I) samples and their thermal stability are quite similar to
2
0
.71 nm × 1.84 nm in size. In the presence of all three Cs (Br), Cs (I)
the Cs2.5(C) salt, which is an active catalyst in many reactions.
2
2
and Cs2 (C) samples the p-/o-xylene ratio is ca. 1.0 what clearly
Catalytic properties of all Cs HPW12O40 samples were inves-
.5
2
indicates a lack of diffusion restrictions in all studied cesium salts.
The selectivities of m-xylene transformation by isomerization,
disproportionation and dealkylation pathways were determined to
be 28.5–30.8%, 57.5% and 14.0–11.7% (at 548 K), respectively. All of
them are of similar order as the values obtained in the presence of
other catalysts when no “reagents-molecule size” restrictions were
operative, e.g. faujasite [35]. The selectivity of isomerization versus
tigated in the pore-size sensitive transformation of m-xylene and
compared with that of the Cs2.5(C) salt. The Cs (C) and Cs (Cl)
samples were inactive because of their low specific surface areas.
2
2
However, the change of inactive Cs (C) sample into active cata-
2
lyst by simple modification of preparation procedure i.e. by the
addition of HBr to its colloidal solution is possible. In contrast
Cs (Br) and Cs (I) samples as well as Cs2.5(C) salt were active
2
2

L. Matachowski et al. / Applied Catalysis A: General 450 (2013) 19–27
27
catalysts, however, the two former samples exhibited two-fold
higher activity than the latter. Similar strength of acid sites in all
active catalysts allows to conclude that the number of protons
[19] Bruker AXS (2003): TOPAS V2.1; General Profile and Structure Analysis Software
for Powder Diffraction Data. – User’s Manual, Bruker AXS, Karlsruhe, Germany.
[
[
20] R.W. Cheary, A.A. Coelho, J. Appl. Crystallogr. 25 (1992) 109–121.
21] T. Okuhara, T. Nishimura, H. Watanabe, M. Misono, J. Mol. Catal. 74 (1992)
247–256.
in the Cs HPW12O40 and Cs2.5H0.5PW12O40 salts is a key factor
2
[22] A.K. Singh, S.D. Fernando, Energy Fuels 22 (2008) 2067–2069.
[23] Z. Zhu, R. Tain, C. Rhodes, Can. J. Chem. 81 (2003) 1044–1050.
[24] N. Essayem, A. Holmquist, P.Y. Gayraud, J.C. Vedrine, Y.B. Taarit, J. Catal. 197
(2001) 273–280.
influencing their catalytic activity in transformation of m-xylene.
References
[
[
25] S. Choi, Y. Wang, Z. Nie, J. Liu, C.H.F. Peden, Catal. Today 55 (2000) 117–124.
26] M. Todorovi cˇ , I. Holclajtner-Antunovi cˇ , U. Mio cˇ , D. Bajuk-Bogdanovi cˇ , Mater.
Sci. Forum 55 (2007) 207–212.
[1] G.M. Brown, M.R. Noe-Spirlet, W.R. Busing, H.A. Levy, Acta Crystallogr. B33
(
1977) 1038–1046.
[
27] T. Okuhara, H. Watanabe, T. Nishimura, K. Inumaru, M. Misono, Chem. Mater.
[
[
[
[
[
[
[
[
2] M. Misono, Catal. Rev. Sci. Eng. 29 (2/3) (1987) 269–321.
3] Y. Izumi, Catal. Today 33 (1997) 371–409.
4] I.V. Kozhevnikov, Chem. Rev. 98 (1998) 171–198.
5] T. Okuhara, Chem. Rev. 102 (2002) 3641–3666.
1
2 (2000) 2230–2238.
28] J.A. Dias, E. Caliman, S.C.L. Dias, Microporous Mesoporous Mater. 76 (2004)
21–232.
[
2
[
[
[
29] K. Kaneko, J. Membr. Sci. 94 (1994) 59–89.
6] M. Misono, Chem. Commun. (2001) 1141–1152.
30] R.D. Shannon, Acta Crystallogr. A 32 (1976) 751–767.
31] A. Zi e˛ ba, A. Drelinkiewicz, P. Chmielarz, L. Matachowski, J. Stejskal, Appl. Catal.
A: Gen. 387 (2010) 13–15.
7] J. Kaur, I.V. Kozhevnikov, Chem. Commun. (2002) 2508–2509.
8] T. Okuhara, N. Mizuno, M. Misono, Adv. Catal. 41 (1996) 113–252.
9] J. Bonardet, J. Fraissard, G.B. McGarvey, J.B. Moffat, J. Catal. 151 (1995)
[
32] L. Matachowski, A. Zi e˛ ba, M. Zembala, A. Drelinkiewicz, Catal. Lett. 133 (2009)
1
47–154.
49–62.
[10] G. Koyano, K. Ueno, M. Misono, Appl. Catal. A 181 (1999) 267–275.
[11] P.Y. Gayraud, N. Essayem, J.C. Vedrine, Catal. Lett. 56 (1998) 35–41.
[12] N. Essayem, S. Kieger, G. Coudurier, J.C. Vedrine, Stud. Surf. Sci. Catal. 101A
[
33] http://www.webchem.net/data/BondEnthalpies.htm
[
34] J. Zhang, M. Sun, C. Cao, Q. Zhang, Y. Wang, H. Wan, Appl. Catal. A: Gen. 380
(
2010) 87–94.
35] J.A. Martens, J. Perez-Pariente, E. Sastre, A. Corma, P.A. Jacobs, Appl. Catal. 45
1988) 85–101.
(
1996) 591–600.
[
[
[
13] B.B. Bardin, R.J. Davis, Top. Catal. 6 (1998) 77–86.
(
14] L. Matachowski, A. Drelinkiewicz, E. Lalik, D. Mucha, B. Gil, Z. Bro z˙ ek-Mucha, Z.
Olejniczak, Microporous Mesoporous Mater. 144 (1–3) (2011) 46–56.
15] D. Lapham, J.B. Moffat, Langmuir 7 (1991) 2273–2278.
[
[
36] S. Laforge, D. Martin, J.L. Paillaud, M. Guisnet, J. Catal. 220 (2003) 92–103.
37] P. Lourenco, M.F. Ribeiro, F. Ramoa Ribeiro, J. Rocha, Z. Gabelica, Appl. Catal. A
[
[
[
148 (1996) 167–180.
16] T. Okuhara, Appl. Catal. A 256 (2003) 213–224.
17] K. Okamoto, S. Uchida, T. Ito, N. Mizuno, J. Am. Chem. Soc. 129 (2007)
[
[
38] Ch. Baerlocher, W.M. Meier, D.H. Olson, Atlas of Zeolite Framework Types, fifth
ed., Elsevier, Amsterdam, 2001.
39] http://www.iza-structure.org/databases/
7
378–7384.
[18] K. Inumaru, Catal. Surv. Asia 10 (2006) 151–160.