Synthesis and catalytic properties of titanium containing extra-large pore zeolite CIT-5

Article abstract of DOI:10.1016/j.cattod.2014.01.003

Titanium containing extra-large pore zeolite CIT-5 (IZA code: CFI) was successfully prepared by direct synthesis using Cab-O-Sil M5 and titanium(IV) butoxide in the presence of LiOH. N-Methylsparteinium hydroxide was used as a structure directing agent. The product crystallized into thin plate crystals with an approximate size of 20 × 5 × 0.2 μm. The lowest achieved Si/Ti ratio was 23. The necessary duration of hydrothermal synthesis increased with increasing concentration of Ti in the reaction mixture from 11 days (Si/Ti = 63 in product) to 17 days (Si/Ti = 36) at 155 C. Prepared Ti-CFI samples exhibit BET surface areas in the range of 308-346 m²/g and micropore volumes of 0.094-0.097 cm³/g. CFI possesses extra-large pores (14-ring, 7.2 × 7.5 A?) accessible for bulky molecules, therefore, Ti-CFI is a useful catalyst for epoxidation of double bonds in bulky molecules used in perfumery and pharmacy. The Ti-CFI samples proved to be catalytically active in epoxidation of 1-octene, cyclooctene, α-pinene, and norbornene with hydrogen peroxide as oxidation agent.

Full text of DOI:10.1016/j.cattod.2014.01.003

Catalysis Today 227 (2014) 80–86
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Synthesis and catalytic properties of titanium containing extra-large
pore zeolite CIT-5
∗
ˇ
Jan Prˇech , Martin Kubu˚ , Jirˇí Cejka
Jaroslav Heyrovsky´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, v.v.i., Dolejsˇkova 3, CZ-182 23 Prague 8, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 July 2013
Received in revised form
19 December 2013
Accepted 2 January 2014
Available online 12 February 2014
Titanium containing extra-large pore zeolite CIT-5 (IZA code: CFI) was successfully prepared by direct
synthesis using Cab-O-Sil M5 and titanium(IV) butoxide in the presence of LiOH. N-Methylsparteinium
hydroxide was used as a structure directing agent. The product crystallized into thin plate crystals with
an approximate size of 20 × 5 × 0.2 ␮m. The lowest achieved Si/Ti ratio was 23. The necessary duration
of hydrothermal synthesis increased with increasing concentration of Ti in the reaction mixture from 11
days (Si/Ti = 63 in product) to 17 days (Si/Ti = 36) at 155 ◦C. Prepared Ti-CFI samples exhibit BET surface
areas in the range of 308–346 m2/g and micropore volumes of 0.094–0.097 cm3/g. CFI possesses extra-
Keywords:
˚
large pores (14-ring, 7.2 × 7.5 A) accessible for bulky molecules, therefore, Ti-CFI is a useful catalyst for
Zeolite CIT-5
Titanosilicate
Epoxidation
Catalysis
epoxidation of double bonds in bulky molecules used in perfumery and pharmacy. The Ti-CFI samples
proved to be catalytically active in epoxidation of 1-octene, cyclooctene, ␣-pinene, and norbornene with
hydrogen peroxide as oxidation agent.
Norbornene
␣-Pinene
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
fragrance ingredients trans-carveol and campholeic aldehyde [11]).
Van der Waal reported synthesis of zeolite Ti-BEA and its catalytic
Zeolites represent a rich family of porous crystalline alumi-
nosilicates successfully applied as catalysts, adsorbents and ion
exchangers [1,2]. The pore size of zeolites is defined by the num-
ber of atoms in the pore entrances. The larger pore, the larger
molecules can access zeolite inner volume and diffuse to the active
sites located mainly inside the pores. Zeolites having more than
12 tetrahedrally coordinated silicon or aluminium atoms in the
entrance ring are called extra-large pore zeolites [3,4]. In contrast
to aluminosilicate zeolites, a considerable breakthrough was the
discovery of zeolite TS-1, having the MFI structure, by Taramasso
et al. [5]. The TS-1 is widely used as catalyst in selective oxidations
(epoxidations) with hydrogen peroxide providing low waste and
being environmentally friendly [6]. Nowadays, the TS-1 is industri-
ally applied by ENI in propylenoxide process [7]. Although the TS-1
is excellent epoxidation catalyst for small linear olefins, the access
of larger molecules to the active sites is strongly restricted due to its
narrow (10-ring) pores. The increase in accessibility of active sites
in titanosilicates with larger pores [8,9] or layered Ti-containing
materials was investigated [10] to prepare epoxides from bulky
alkenes, cycloalkenes and terpenes. They are requested particularly
in pharmacy and perfumery (e.g. ␣-pinene oxide is a precursor of
behaviour in epoxidation of 1-octene [8] and subsequently of nor-
bornene, camphene, and limonene [12]. Wu et al. reported that
Ti-MWW is more active in epoxidation of linear alkenes than TS-
1 and Ti-BEA [13]. Corma and co-workers prepared a delaminated
MWW zeolite and introduced Ti into the structure to obtain Ti-
ITQ-2 material combining the advantages of Ti-MWW structure
and easy accessibility of the external surface in delaminated mate-
rial [14]. However, the Ti-ITQ-2 exhibits high activities only with
organic hydroperoxides as oxidants in the absence of water. Tat-
sumi et al. synthesized Ti-YNU-1 material (interlayer expanded
MWW topology) [15], which is active even with aqueous hydrogen
peroxide. Another promising material is specially designed TS-1 by
Ryoo et al. prepared in the form of nanosheets [10]. This material
exhibits activity similar to bulk TS-1 in epoxidation of 1-hexene,
however, its activity is an order of magnitude higher in epoxidation
of cyclooctene.
While several large-pore titanosilicates (e.g. Ti-BEA, Ti-MOR, Ti-
ITQ-7) were synthesized, the only extra-large pore titanosilicate
Ti-DON was reported up-to-now [16].
CIT-5 zeolite (IZA code: CFI) is an extra-large pore zeolite with
one-dimensional 14-ring channel system. This zeolite shows good
thermal stability. Previously it was prepared as pure silica or iso-
morphously substituted with boron, gallium and aluminium [17].
˚
The CFI zeolite has slightly elliptic pores with diameter 7.2 × 7.5 A.
∗
Corresponding author. Tel.: +420 266 053 856.
E-mail address: jan.prech@jh-inst.cas.cz (J. Prˇech).
The synthesis of the CFI is more advantageous in comparison
0920-5861/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2014.01.003

J. Prˇech et al. / Catalysis Today 227 (2014) 80–86
81
with Ti-DON requiring a quaternary ammonium salt as a template
2.2. Synthesis of TS-1
instead of a cobalt organometallic complex for DON zeolite [16].
In this contribution, we report the first synthesis of titanosil-
icate form of the zeolite CIT-5 (Ti-CFI) and a characterization
of this material. High thermal stability of the CFI structure as
well as easy incorporation of Ti into the framework enables its
application as a catalyst for epoxidation of bulky olefins and
terpenes. The results of catalytic testing in epoxidation with hydro-
gen peroxide are also presented as an additional characterization
technique with the aim to demonstrate the usefulness of our
approach.
Titanosilicate TS-1 was prepared from a gel with initial Si/Ti
ratio 40 according to the procedure described in reference [19]
using tetraethyl orthotitanate (Aldrich, technical grade), tetraethyl
orthosilicate (Aldrich, 98%), and tetrapropylammonium hydroxide
(Aldrich, 20% in water) as an SDA.
2.3. Characterization
X-ray powder diffraction patterns (XRD) were collected using a
Bruker AXS D8 Advance diffractometer equipped with a graphite
monochromator and a position sensitive detector Våntec-1 using
CuK␣ radiation in Bragg–Brentano geometry. Data were collected
in continuous mode over the 2Â range of 5–50◦ with scan speed of
0.0610◦/s.
2. Experimental
2.1. Synthesis of CFI
The size and shape of zeolite crystals were examined by scan-
ning electron microscopy (SEM) on a JEOL, JSM-5500LV microscope.
The images were collected with acceleration voltage of 20 kV.
Energy dispersive X-ray spectroscopy (EDX) analyses were per-
formed using Hitachi S-4800 field emission scanning electron
microscope at 25 kV with Noran EDX system. Also some SEM images
were collected using the Hitachi system at 5 kV.
Argon sorption isotherms were measured at liquid argon tem-
perature (−186 ◦C) with Micromeritics ASAP 2020 volumetric
instrument. To attain sufficient accuracy in the accumulation of
the adsorption data, the ASAP 2020 was equipped with pressure
transducers covering the 133 Pa, 1.33 kPa and 133 kPa ranges. Prior
to the sorption measurements, individual zeolites were outgassed
under turbomolecular pump vacuum at 300 ◦C for 6 h.
BET area was evaluated using adsorption data in the range
of a relative pressure from p/p0 = 0.055 to p/p0 = 0.22. The t-plot
method [20] was applied to determine the volume of micropores
(Vmicro). The adsorbed amount of argon at p/p0 = 0.99 reflects the
total adsorption capacity (Vtotal).
UV–vis absorption spectra were collected for calcined samples
using a PerkinElmer Lambda 950 Spectrometer with a 5 mm or
2 mm quartz tube and a large 8 × 16 mm slit. The data were col-
lected in the wavelength range of 190–500 nm.
Titanium(IV) butoxide (Aldrich, 97%), lithium hydrox-
ide hydrate (Fluka, 99%), Cab-O-Sil M-5 (Havel Composites,
Czech Republic), (−)-spartein sulfate pentahydrate (SAFC,
99%), and methyl iodide (Fluka, 99%) were used as pur-
chased.
The N(16)-methylsparteinium hydroxide, used as a structure
directing agent (SDA), was prepared according to the procedure
described in reference [18]. N(16)-Methylsparteinium iodide was
prepared by a reaction of (−)-spartein and methyl iodide in ace-
tonitrile at room temperature for 72 h. The resulting iodine salt
was ion exchanged into hydroxide form using strongly basic anion
exchange resin AG1-X8 (Biorad, OH− form). Properly ion exchanged
SDA solution exhibited pH 14.0 and no opacity after acidification
with 1 M HNO3 to pH 1 and addition of 1 volume of 0.06 M AgNO3
solution.
The synthesis of the CFI was based on the procedure described
in reference [18]. N(16)-Methylsparteinium hydroxide (0.3 M aque-
ous solution, N-MeSpa OH) was mixed with demineralised water
and lithium hydroxide hydrate in a Teflon-lined autoclave. A solu-
tion of titanium(IV) butoxide in 1-butanol (25%, TBOTi) was added
dropwise to the synthesis mixture under vigorous stirring for
2 min. The resulting mixture was homogenized at room temper-
ature for 30 min. Finally, Cab-O-Sil M-5 was added and the mixture
was stirred for another 30 min. In some cases the mixture was
seeded with earlier prepared “as-synthetized” CFI crystals. The
milky homogeneous mixture, with molar composition 0–2.0 TBOTi,
10 N-MeSpa OH, 5 LiOH, 50 SiO2 and 2500 H2O, was closed in an
autoclave and heated under agitation to 155–170 ◦C for 6–17 days.
Lithium hydroxide serves as an accelerator for the crystallization
of the CFI phase. On the other hand Li+ does not compete with
N-methylsparteinium in templating the structure due to its small
ionic radius [18].
Chemical composition of the Ti-CFI (60), (40) and (25) was
determined also by X-ray fluorescence analysis (XRF) with a spec-
trometer Philips PW 1404 using an analytical program UniQuant.
The samples were mixed with dentacryl as a binder and pressed on
the surface of cellulose pellets.
2.4. Catalytic reactions
The catalytic activity of Ti-zeolites was tested in epoxidation
of 1-octene (Aldrich, 98%), cyclooctene (Aldrich, 99%), ␣-pinene
(Acros organics, 98%) and norbornene (Aldrich, 99%) with hydrogen
peroxide (Aldrich, 35 wt.% aqueous solution) as oxidant in ace-
tonitrile (Fisher chemical, HPLC grade). Mesitylene (Sigma–Aldrich,
99%) or 1,3-diisopropylbenzene (Fluka, 95%) were used as internal
standards. The activity was compared with standard TS-1 zeolite.
The reactions were carried out at 50 or 60 ◦C in a following way. The
catalyst was activated by heating at 450 ◦C for 90 min and cooled
down in a desiccator. The catalyst (50 mg) was added to 6 ml of
acetonitrile followed by 300 mg of the alkene and 150 mg of the
internal standard. The mixture was heated in a flask under Dim-
roth condenser to the reaction temperature and the reaction was
started by addition of 0.5 mol equivalent of H2O2 (based on alkene).
Samples of the reaction mixture were taken in regular intervals,
centrifuged and analysed using an Agilent 6850 GC system with
20 m long DB-5 column, an autosampler and a FID or a MS detector.
Helium was used as a carrier gas.
The solid product was collected by filtration, washed out with
demineralised water and dried overnight at 65 ◦C. Calcination of
the samples was carried out in a stream of air at 570 ◦C for
8 h.
To remove extra-framework titanium species from the CFI sam-
ples, some samples were treated with nitric acid solution prior
to the calcination. 30–90 ml (depending on Si/Ti ratio) of 2.0 M
HNO3 solution was used per 1 g of dry as-synthetized material. The
mixture was heated at 100 ◦C for 16 h. After the given time, the
solid material was filtered off, washed with water and calcined as
described above.
In order to remove Li+ anions, some samples were ion-
exchanged with ammonium nitrate. Calcined samples were treated
four-times with 1.0 M NH4NO3 solution (Lach-Ner, 99%) for 4 h
at room temperature using 100 ml of the solution per 1 g of the
sample. The samples were re-calcined at 540 ◦C for 6 h before
use.

82
J. Prˇech et al. / Catalysis Today 227 (2014) 80–86
Table 1
Description of the synthesis of Ti-CFI samples.
Sample
Si/Ti (RM)
Temperature (◦C)
Yield (%)
Crystal size (␮m)
Si/Ti (EDX)
Time (day)
Structurea
Ti-CFI (100)
Ti-CFI (60)
Ti-CFI (40)
Ti-CFI (25)
Ti-CFI (10)
Ti-AFI (a)
TS-1
100
60
40
25
10
80
40
170
155
155
160
160
160
175
85
79
79
98
–
10–40
10–20
10–20
10–20
–
106
63c
36c
23c
–
6
11
17
11b
20b
14
6
CFI + Am.
CFI
CFI
CFI
Am.
AFI
75
85
1–2.5
0.2
106
38
MFI
a
Am. stands for amorphous material.
The synthesis was seeded with Ti-CFI (40).
Si/Ti molar ratio determined by XRF.
b
c
In order to perform a leaching test, a reaction mixture was pre-
N(16)-Methylsparteinium bromide can template also the syn-
thesis of zeolite AFI [21]. Under certain crystallization conditions,
AFI can crystallize together with CFI as observed by Barrett et al.
[22]. The presence of particular anions (Br− vs. OH−) [21] or cations
(Na+, K+ vs. Li+) [17] seems to control the crystallization route and
the purity of the final zeolite product. We observed the same effect
in the case of Ti-CFI. When the SDA was used without ion-exchange
only AFI phase was obtained after 14 days of the crystallization
at 160 ◦C (sample Ti-AFI (a)). Lok et al. invented the synthesis of
titanium containing aluminophosphate with AFI topology (TAPO-
5) [23] but the synthesis of titanosilicate with AFI structure was not
reported until now. Our results suggest that the preparation of AFI
titanosilicate with higher Ti content should be also possible.
There are 3 characteristic diffraction lines of the CFI structure at
low 2Â angles: 6.91◦, 7.32◦ and 13.9◦ followed by a group of 5 peaks
between 18◦ and 22◦ (Fig. 1).
pared as described above and it was left to react at 60 ◦C for 3 h.
After the given time the catalyst was filtered off and the mixture
was stirred on at 60 ◦C and samples were taken in 1 h intervals.
Epoxidation of norbonene in methanol was carried out using the
same conditions as in [12] with 4-times smaller amount of reagents.
471 mg (5 mmol) of norbornene was dissolved in 7.6 ml (6.0 g) of
methanol. 235 mg of mesitylene was added as an internal standard
followed by 25 mg of activated catalyst. The mixture was heated to
60 ◦C and the reaction was started by addition of 2.5 mmol of H2O2.
3. Results and discussion
3.1. Synthesis
Results of syntheses of different Ti-CFI samples are summarized
in Table 1. Well-crystalline CFI samples (Fig. 1) were obtained with
Si/Ti ratios ranging from 106 to 23 in the product. Kubota et al.
[18] reported synthesis of Al-CFI and pure silica CFI under static
conditions with quartz tubes inside the autoclave but the attempts
to use this method for Ti-CFI were not successful providing either
amorphous material or cristobalite. After initial experiments rep-
resented by Ti-CFI (100) sample, the synthesis temperature was
lowered from 170 ◦C to 155 ◦C in order to obtain more uniform
crystal size distribution (see Fig. 2). As expected, this caused a pro-
longation of the synthesis time. However, comparing the syntheses
Ti-CFI (60) and Ti-CFI (40) we conclude that also increasing amount
of titanium in the synthesis mixture increases the synthesis time
(from 11 days of Ti-CFI (60), to 17 days of Ti-CFI (40)). Both seeding
of the reaction mixture or a slight increase in the synthesis temper-
ature significantly shortened the synthesis time. As a result, Ti-CFI
(25) sample crystallized in only 11 days.
The intensity of the first two diffraction lines depends on the
presence of the structure-directing agent in the pores and on the
orientation of the CFI crystals having the morphology of thin plates
(Fig. 2). Kubota et al. reported that some diffraction lines may prac-
tically disappear if there is any preferred orientation of CFI crystals
[18]. This may lead to more complex identification of the CFI phase.
No preferential orientation was observed for Ti-CFI. The calcina-
tion of samples result always in a strong increase in the intensity of
diffraction lines at 2Â = 6.91◦ and 7.32◦ as well as an appearance of
two additional weak diffraction lines at 2Â = 12.2◦ and 12.9◦. These
two lines are also characteristic in the CFI pattern (Fig. 3).
A study of Si/Ti molar ratios in the reaction mixtures and in the
crystalline products revealed that the Si/Ti ratio in the crystalline
product practically follows that one in the reaction mixture. The
lowest Si/Ti ratio achieved is 23, however not all the titanium is
in the framework positions (see the discussion of UV/Vis results
below). Syntheses with initial Si/Ti ratio 10 provided an amorphous
material.
Scanning electron micrographs (SEM) show that the zeolite Ti-
CFI crystallizes into thin plate crystals with the largest dimension
up to approximately 20 ␮m (Fig. 2), sometimes forming aggregates.
This morphology is the same as for aluminosilicate, pure siliceous,
or borosilicate forms of the CFI [17,18]. The thickness of the crystals
is about 0.2 ␮m, which is about 100 unit cells.
Textural properties of synthesized samples are summarized in
Table 2. The data are consistent with those reported for CFI previ-
ously [17,18]. The micropore volume might be taken as a parameter
Table 2
Textural properties of CFI titanosilicates; the initial Si/Ti ratio is in brackets.
Sample
BET (m2/g)
Vmicro (cm3/g)
Vtotal (cm3/g)
Ti-CFI (100)
Ti-CFI (60)
Ti-CFI (40)
Ti-CFI (25)
TS-1
272
308
339
346
445
0.068
0.094
0.098
0.097
0.127
0.524
0.293
0.256
0.235
0.375
Fig. 1. XRD patterns of as-synthetized CFI and AFI samples. From the top: Ti-CFI
(100), Ti-CFI (60), Ti-CFI (40), Ti-CFI (25), Ti-AFI (a).

J. Prˇech et al. / Catalysis Today 227 (2014) 80–86
83
Fig. 2. SEM images of Ti-CFI (100) (top left), Ti–CFI (60) (top right), Ti-CFI (40) (bottom left) and Ti-CFI (25) (bottom right).
characterizing the quality of the sample. Well crystalline samples
(Ti-CFI (60), (40) and (25)) exhibit micropore volume of 0.094, 0.098
and 0.097 cm3/g respectively. CFI samples with Vmicro < 0.09 cm3/g
contain some amorphous material (see Fig. 2, top left image). The
uptake of argon at higher relative pressures is caused by interpar-
ticle condensation among individual flat crystals. Hysteresis loops
result from the cavitation effect that manifest itself during desorp-
tion (Fig. 4).
It is generally assumed that single framework “TiO4” species are
the active sites in epoxidation reactions [24]. On the other hand,
extra-framework Ti species may direct the formation of stable
dense TiO2 phase after calcination, affecting negatively the selectiv-
ity for oxidation reactions [25]. The coordination state of Ti species
was studied using UV-vis spectroscopy. As observed in Fig. 5, all
3 well crystalline samples (Ti-CFI (60), (40) and (25)) have similar
spectra with 2 bands centred at 220 and 260 nm. Zecchina et al.
ascribed these bands to tetrahedrally coordinated “TiO4” species
and hexacoordinated extra-framework “TiO6” species respectively
[26]. We conclude that part of titanium is in isolated extra-
framework positions.
The extra-framework Ti species might be removed by treatment
with boiling HNO3 prior to the calcination of the sample [25]. We
examined this procedure using 2 samples: Ti-CFI (60) and Ti-CFI
(25). Fig. 5 shows that nearly all extra-framework Ti from Ti-CFI (60)
sample was successfully removed. On the other hand, in the spec-
trum of Ti-CFI (25), a new band appeared centred at 330 nm. This
band indicates the formation of dense TiO2 (Fig. 5B) [26]. When the
amount of HNO3 used for the treatment of Ti-CFI (25) was increased
to 60 g of 2 M HNO3 per 1 g of the zeolite, the only band at 220 nm
was observed as in case of Ti-CFI (60). We conclude that high excess
of the acid is necessary to dissolve all extra-framework Ti. Another
significant disadvantage of this HNO3 treatment is the loss of some
tetrahedral titanium indicated by the decrease in intensity of the
band at 220 nm.
All the spectra of calcined samples were collected and the
absence of adsorption bands above 300 nm shows that the for-
mation of dense TiO2 phase does not occur during calcination of
untreated samples. Furthermore Guo et al. reported that not only
framework “TiO4” species but also isolated extra-framework “TiO6”
Fig. 3. XRD patterns of calcined CFI samples. From the top: Ti-CFI (100), Ti-CFI (60),
Ti-CFI (40), Ti-CFI (25).

84
J. Prˇech et al. / Catalysis Today 227 (2014) 80–86
Fig. 4. Argon sorption isotherms of Ti CFI (100) (left black), Ti-CFI (60) (left gray), Ti-CFI (40) (right black) and Ti-CFI (25) (right gray); adsorption branch is in full dots,
desorption is in empty dots; gray isotherms are moved +25 cm3/g for clarity.
species (with absorption band at 270 nm) are active in the epoxida-
tion reaction [27]. Taking into account these findings together with
the loss of some framework titanium mentioned above, we used
untreated samples in catalytic experiments.
entries 8, 9, 14, 15); however, the solvent effect on the reaction was
not systematically studied.
For 1-octene, the performance of the TS-1 is better (56% conver-
sion, 91% selective to the epoxide after 3 h) compared with Ti-CFI
(conversion 26%, 46% selective to the epoxide after 3 h) as the lin-
ear molecule can easily access TS-1 pores, which entrances are over
the whole external surface of the crystal. In contrast to TS-1, only
a small fraction of Ti-CFI external surface contains the entrances to
the channel system (the channels of CFI are parallel to the longest
dimension of the crystals [17]).
3.2. Catalysis
The Ti-CFI catalysts were tested in the epoxidation of 1-octene,
cyclooctene, ␣-pinene and norbornene and compared with TS-1
zeolite. Results of the catalytic tests are summarized in Table 3. In
order to suppress possible intramolecular rearrangements of either
substrates or products, the reactions were carried out in acetonitrile
[12], which is weakly basic in contrast to alcoholic solvents. Sev-
eral examples of reactions in methanol are also presented (Table 3,
When using cyclooctene as a substrate, the Ti-CFI catalysts pro-
vided higher conversions (21%) than TS-1 (14%) after 3 h as well
as higher yield of the epoxide (7.1%, 5.3% contra 1.0% for Ti-CFI
(40), Ti-CFI (100) and TS-1 after 3 h, respectively). The active sites in
Ti-CFI are more easily accessible for bulkier substrate than the cat-
alytic centers in medium-pore TS-1. The selectivity to the epoxide
(defined as amount of epoxide among all observed oxygenate prod-
ucts) increased from 77 up to 99% when Ti-CFI was ion-exchanged
with ammonium nitrate (entry 6) probably due to the removal
of residual Li+ cations from the material [28]. Cyclooctenone
and hydroxy-cyclooctene were observed as main oxidation by-
products in all the reactions. Our results indicate that the material
should be ion-exchanged before catalytic tests. The effect of solvent
was not studied systematically, however, in cyclooctene epoxida-
tion we tested acetonitrile as well as methanol under otherwise
the same reaction conditions over ion-exchanged catalysts Ti-CFI
(40) and Ti-CFI (60). Conversion of cyclooctene in acetonitrile over
ion-exchanged Ti-CFI (40) was 4.8% after 3 h while in methanol
the conversion was only 3.2%. Selectivity to cyclooctene oxide was
practically 100%. In the case of ion-exchanged catalyst Ti-CFI (60),
the conversion of cyclooctene was 10.5% in acetonitrile and 2.4% in
methanol after 3 h. Again, the selectivity to the epoxide was higher
than 99%. It can be concluded that cyclooctene epoxidation in ace-
tonitrile proceeds with a higher reaction rate than in methanol
while the selectivity to the epoxide over both Ti-CFI catalysts is
close to 100%.
␣-Pinene underwent the epoxidation over Ti-CFI catalyst
although the selectivity to epoxide was low (38% after 4 h) in com-
parison to cyclooctene and norbornene (selectivity to the epoxide
75–99%) at the conversion of 20–30%. It is in accordance with the
results obtained by van der Waal et al. in epoxidation reactions
of other terpenes like ␣-terpineol or limonene over Ti-BEA zeo-
lite. Terpenes easily undergo intramolecular rearrangements [12].
It was evidenced in our case by observation of a variety of side
products including carbonyl and hydroxyl compounds. In contrast,
TS-1 completely failed to catalyze the epoxidation of ␣-pinene as
expected. The ␣-pinene is a bulky rigid molecule, not able to enter
the pores of TS-1.
Fig. 5. UV-vis spectra of calcined and HNO3 treated CFI samples and TS-1.

J. Prˇech et al. / Catalysis Today 227 (2014) 80–86
85
Table 3
Oxidation of various alkenes with H2O2 over Ti-CFI and TS-1 catalysts.
Entry
Substrate
Catalyst
Si/Ti
t (◦C)
Time (h)
X(alkene) (%)c
Selectivity to
Epoxide in
oxygenates (%)d
products (%)e
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1-Octene
1-Octene
Ti-CFI(40)
TS-1
Ti-CFI(40)
Ti-CFI(100)
TS-1
36
38
36
50
50
50
50
50
60
60
60
60
50
50
60
60
60
60
3
3
3
3
3
3
3
3
3
4
4
4
4
1
1
26
56
21
21
3
29
34
25
7.2
54
20
66
38
5.8
No epoxide
21
8.6
5.9
1.8
46
91
75
77
99
99
99
99
99
38
Cyclooctene
Cyclooctene
Cyclooctene
Cyclooctene
Cyclooctene
Cyclooctenea
Cyclooctenea
␣-Pinene
106
38
36
63
36
63
63
38
63
38
23
36
14
Ti-CFI(40)b
Ti-CFI(60)b
Ti-CFI(40)b
Ti-CFI(60)b
Ti-CFI(60)
TS-1
4.8
10.5
3.2
2.4
24
␣-Pinene
Norbornene
norbornene
Norbornenef
Norbornenef
Ti-CFI(60)b
TS-1
29
56
16
7
95
90
82
81
Ti-CFI(25)b
Ti-CFI(40)
a
Reaction performed in methanol.
Ion-exchanged catalyst.
Conversion of alkene is defined as overall amount of consumed alkene.
Selectivity to oxygenates is defined as [oxygenate products]/[consumed alkene].
Amount of the epoxide among all observed oxygenate products.
Reaction performed in methanol under conditions used in [12].
b
c
d
e
f
Norbornene was oxidized to 2,3-epoxynorbornane with both
The incorporation of titanium was confirmed by EDX and UV/Vis
spectroscopy. The length of the hydrothermal synthesis increased
while the Si/Ti molar ratio decreased in the reaction mixture. Text-
ural properties of the samples determined from argon adsorption
at −186 ◦C are consistent with those reported previously.
Ti-CFI and TS-1 catalysts. The best result was obtained using ion-
exchanged Ti-CFI (60) catalyst (29% conversion; 21% selective to
oxygenates; selectivity to oxygenates oxidation is defined as [oxy-
genate products]/[consumed alkene]; 95% selective to the epoxide
after 4 h). To compare the performance of the catalyst with previ-
ously published data for Ti-BEA and Ti-MCM-41, two reactions were
carried out in methanol under the same conditions as in reference
[12]. The conversions of norbornene in 1 h over ion-exchanged Ti-
CFI (25) and Ti-CFI (40) were 16% and 7% respectively. Van der Waal
et al. achieved conversion of 13.3% (Ti-BEA) and 6.7% (Ti-MCM-41).
For both Ti-CFI (40) and ion-exchanged Ti-CFI (25) the selectivity
to oxygenates in 1 h is lower than that reported in [12] (1.8% and
5.9%, respectively, contra 67% Ti-BEA; 65% Ti-MCM-41). However,
the Ti-CFI catalysts are more selective to the epoxide than Ti-BEA
(81–82% contra 56%). Main oxygenate byproduct was 3-hydroxy-
norbornan-2-on.
The leaching test was done with ion exchanged Ti-CFI (60) cat-
alyst. The reaction mixture containing cyclooctene as a substrate
was left to react at 60 ◦C for 3 h. Then the catalyst was filtered off
and the mixture was stirred at 60 ◦C and samples were taken in 1 h
intervals. No further formation of cyclooctene oxide was observed
for additional 4 h. This result evidences that the reaction does not
proceed in the homogeneous phase.
Prepared CFI titanosilicates are active catalysts for epoxidation
of C C double bond with hydrogen peroxide as an oxidant. For
bulky substrates like cyclooctene, ␣-pinene, and norbornene, the
performance exceeds conventional TS-1 zeolite demonstrating that
the preparation of large pore titanosilicates is a useful approach.
Acknowledgements
The authors thank to Dr. L. Brabec for SEM images, M. Bousˇa for
SEM-EDX measurements and Dr. G. Sádovská for UV-vis measure-
ments. The authors acknowledge the Czech Science Foundation for
the support of this research (P106/12/G015).
References
ˇ
[1] J. Cejka, A. Corma, S.I. Zones (Eds.), Zeolites and Catalysis: Synthesis, Reactions
and Applications, Wiley-VCH, Weinheim, 2010.
ˇ
[2] J. Cejka, G. Centi, J. Perez-Pariente, W.J. Roth, Catal. Today 179 (2012) 2–15.
[3] M.E. Davis, Chem. Eur. J. 3 (1997) 1745–1750.
[4] L.B. McCusker, Ch. Baerlocher, Zeolite structures, in: J. Cejka, H. van Bekkum
The epoxides are stable under the reaction conditions while
diols-products of epoxide hydrolysis–were not observed. The sta-
bility of the epoxides under the reaction conditions was studied
in an experiment, where cyclooctene oxide and ␣-pinene oxide,
respectively, were used instead of substrate at 60 ◦C under the same
conditions (including the presence of hydrogen peroxide) as the
catalytic experiments. No drop in concentration of the epoxides
was observed during 5 h.
ˇ
(Eds.), Studies in Surface Science and Catalysis, 157, Elsevier, Amsterdam, 2005,
pp. 41–64.
[5] M. Tamarasso, G. Perego, B. Notari, US Pat 4410501, (1983).
[6] C. Perego, A. Carati, P. Ingallina, M.A. Mantegazza, G. Bellussi, Appl. Catal. A:
Gen. 221 (2001) 63–72.
[7] <http://www.eni.com/en IT/attachments/azienda/attivita-
strategie/petrolchimica/licensing/TS1-flyer-lug09.pdf/>, 2013 (26.06.13).
[8] J.C. van der Waal, H. van Bekkum, J. Mol. Catal. A: Chem. 124 (1997) 137–146.
[9] M-J. Diaz-Cabanas, L.A. Villaescusa, M.A. Camblor, Chem. Commun. 76 (2000)
1–762.
[10] K. Na, Ch. Jo, J. Kim, W. Ahn, R. Ryoo, ACS Catal. 1 (2011) 901–907.
[11] N. Kumar, P. Mäki-Arvela, S.F. Diaz, A. Aho, Y. Demidova, J. Linden, A. Shepid-
chenko, M. Tenhu, J. Salonen, P. Laukkanen, A. Lashkul, J. Dahl, I. Sinev, A-R.
Leino, K. Kordas, T. Salmi, D.Y. Murzin, Top. Catal. 56 (2013) 696–713.
[12] J.C. van der Waal, M.S. Rigutto, H. van Bekkum, Appl. Catal. A: Gen. 167 (1998)
331–342.
[13] P. Wu, T. Tatsumi, T. Komatsu, T. Yashima, J. Catal. 202 (2000) 245.
[14] A. Corma, U. Diaz, V. Fornes, J.L. Jorda, M. Domine, F. Rey, Chem. Commun. 779
(1999).
4. Conclusions
The first synthesis of titanosilicate with CFI topology is reported.
The Ti-CFI was synthetized using N(16)-methylsparteinium
hydroxide as the template and titanium(IV) butoxide and Cab-O-
Sil M-5 as the source of titanium and silicon, respectively, in the
presence of Li cations. The lowest Si/Ti ratio in product, obtained
using our synthesis protocol, was 23.
[15] W. Fan, P. Wu, S. Namba, T. Tatsumi, Angew. Chem. Int. Ed. 116 (2004) 238–242.
[16] K.J. Balkus, A.G. Gabrielov, S.I. Zones, The synthesis of UTD-1, Ti-UTD-1 and
Ti-UTD-8 using CP*2CoOH as a structure directing agent, in: L. Bonneviot, S.

86
J. Prˇech et al. / Catalysis Today 227 (2014) 80–86
Kaliaguine (Eds.), Studies in Surface Science and Catalysis, 97, Elsevier, Amster-
dam, 1995, pp. 519–525.
[17] M. Yoshikawa, P. Wagner, M. Lovallo, K. Tsuji, T. Takewaki, C-Y. Chen, L.W.
Beck, C. Jones, M. Tsapatsis, S.I. Zones, M.E. Davis, J. Phys. Chem. B 102 (1998)
7139–7147.
[18] Y. Kubota, S. Tawada, K. Nakagawa, C. Naitoh, N. Sugimoto, Y. Fukushima, T.-A.
Hanaoka, Y. Imada, Y. Sugi, Micropor. Mesopor. Mat. 37 (2000) 291–301.
[19] M. Taramasso, G. Perego, B. Notari, in: H. Robson (Ed.), Verified Syntheses of
Zeolitic Materials, Elsevier, Amsterdam, 2001, p. 207.
[23] B.M. Lok, M.B. Kristoffersen, E.M. Flanigen, EP 121232, (1984).
[24] F. Bonito, A. Damin, G. Ricchiardi, M. Ricci, G. Spano, R. D’Aloisio, A.
Zecchina, C. Lamberti, C. Prestipino, S. Bordiga, J. Phys. Chem. B 108 (2004)
3573–3583.
[25] M. Moliner, A. Corma, Chem. Mater. 24 (2012) 4371–4375.
[26] A. Zecchina, G. Spoto, S. Bordiga, A. Ferrero, G. Petrini, G. Leofanti, M. Padovan,
Framework and Extraframework Ti in Titanium-Silicalite: Investigation by
Means of Physical Methods, in: P.A. Jacobs, N.I. Jaeger, L. Kubelková, B.
Wichterlová (Eds.), Studies in Surface Science and Catalysis, 69, Elsevier,
Amsterdam, 1991, pp. 251–258.
[20] B.C. Lippens, J.H. de Boer, J. Catal. 4 (1965) 319.
[21] R.F. Lobo, M.E. Davis, Microporous Mater. 3 (1994) 61–69.
[22] P.A. Barrett, M.J. Diaz-Cabanas, M.A. Camblor, R.H. Jones, J. Chem. Soc. Faraday
Trans. 94 (1998) 2475–2481.
[27] Q. Guo, K. Sun, Z. Feng, G. Li, M. Guo, F. Fan, C. Li, Chem. Eur. J. 18 (2012)
13854–13860.
[28] C.B. Khouw, M.E. Davis, J. Catal. 151 (1995) 77–86.