Microporous and Mesoporous Materials 44±45 (2001) 435±444
www.elsevier.nl/locate/micromeso
An example of mesostructured zeolitic material: UL-TS-1
D. Trong On a, D. Lutic b, S. Kaliaguine a,*
a
b
Department of Chemical Engineering, Laval University, Ste Foy, Que., Canada, G1K 7P4
Faculty of Chemistry, ``Al.I.Cuza'' University of Jassy, Bd. Carol I Nr. 11, Jassy, Romania
Received 16 March 2000; accepted 24 August 2000
Abstract
A new approach for the synthesis of titanium containing mesostructured zeolitic materials, UL-TS-1, that involves
the use of surfactant containing amorphous mesoporous titania±silica as precursor phase is reported. UL-TS-1 was
synthesized in the solid state by heating TPAOH-impregnated precursor materials at 120°C for several days. Various
techniques including XRD, N2 adsorption, UV±visible, FTIR, TEM and 29 Si MAS NMR were used to monitor the
physico-chemical properties of these materials as a function of crystallization time. The XRD di€ractograms show
broad peaks, which match those of ZSM-5 and grow in intensity as the crystallization time is increased. The increase in
the percentage of crystallinity is also correlated with the corresponding variations in micropore and mesopore volumes,
BET and BJH surface areas. The results indicate that the mesopore walls consist of zeolite nanocrystals. Depending on
crystallization time, a range of materials from totally amorphous up to 80% crystalline is observed, while some of
mesopores are preserved. The FTIR and UV±visible results also show that all titanium ions are essentially incorporated
into the UL-TS-1 framework. Ó 2001 Elsevier Science B.V. All rights reserved.
Keywords: Mesostructured zeolites; UL-zeolites; UL-TS-1; Titanium silicalite; Oxyfunctionalization
1. Introduction
Zeolites and related crystalline molecular sieves
are widely used as catalysts in industry, as they
possess catalytically active sites, as well as uniformly sized and shaped micropores, that allow for
their use as shaped selective catalysts. However,
due to the pore size constraints, the unique catalytic properties of zeolites are limited to reactant
[1].
molecules having kinetic diameters below 10 A
The new family of M41S mesoporous molecular
overcomes the
sieves with pore sizes of 10±300 A,
*
Corresponding author. Tel.: +1-418-656-2708; fax: +1-418656-3810.
E-mail address: kaliagui@gch.ulaval.ca (S. Kaliaguine).
limitation of microporous zeolites and allows the
di€usion of larger molecules. These materials are
however amorphous solids. The mesoporous
silica±alumina are consequently much less strongly
acidic and do not exhibit the spectacular catalytic
properties of acidic zeolites. Moreover, their hydrothermal stability is low and as a consequence
their industrial use as catalysts is very limited so
far [2±5].
Very recently, we reported a general methodology for the production of a new type of materials with bimodal pore structures, overcoming
the limitation of both zeolites and mesoporous
molecular sieves. These new hierarchically mesostructured zeolitic materials, designated as ULzeolites, are obtained by a secondary templated
1387-1811/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved.
PII: S 1 3 8 7 - 1 8 1 1 ( 0 1 ) 0 0 2 1 8 - 9
436
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
crystallization of zeolites starting from the amorphous mesoporous molecular sieves of corresponding elemental composition [6,7]. UL-zeolites
are considered of great potential interest for applications in catalysis and separation, due to easier
transport of guest molecules through the mesopores and shorter di€usion pathways in the zeolitic
walls. Furthermore, UL-zeolites are expected to be
stable in many catalytic applications where M41S
materials would not be applicable.
Titanium silicalite-1 (TS-1) is a remarkable
catalyst for the oxidation of a variety of organic
molecules having kinetic diameters below 6 A,
using hydrogen peroxide as the oxidant [8±10]. Incorporation of titanium into the framework of the
new hierarchically mesostructured zeolitic materials by direct synthesis is expected to lead to catalytic oxidation reactions with improved eciencies,
because these materials combine the bene®ts of
both pore size regimes. In this study, we describe an
example of hierarchically mesostructured titanium±silicalite materials, UL-TS-1. The synthesis
procedures as well as the physico-chemical characterization of these materials using a combination
of XRD, N2 adsorption, FTIR, TEM, 29 Si MAS
NMR techniques will be reported.
2. Experimental section
2.1. Synthesis
UL-TS-1 materials were obtained by a secondary templated crystallization of zeolites starting from the amorphous mesoporous materials of
corresponding elemental composition [6]. It is of
special concern that the walls of the precursor
amorphous mesoporous materials should be as
thick as possible. Refs. [11,12] provide methods for
the preparation of mesoporous molecular sieves
that are useful in this context. A series of UL-TS-1
(atomic ratio Ti/Si ˆ 0:5% and 1.5%) was
prepared using poly(alkylene oxide) triblock copolymer HO(CH2 CH2 O)20 (CH2 CH(CH3 )O)70 (CH2 CH2 O)20 H (designated as EO20 PO70 EO20 , Pluronic
P-123, BASF) and tetrapropylammonium hydroxide (TPAOH) as surfactant and template, respectively, according to the method described in
Ref. [6]. The synthesis of UL-TS-1 consists of two
steps: (i) preparation of the amorphous mesoporous precursor followed by (ii) transformation
of the amorphous mesopore walls into crystalline
walls. A typical synthesis procedure was as follows: (step 1) synthesis procedures for mesoporous
silica reported in Refs. [11,12]. Route I: 10 g of
EO20 PO70 EO20 was dissolved in 100 g of ethanol
(EtOH). To this solution, 0.10 mol of SiCl4 was
added followed by an appropriate amount of tetrapropyl orthotitanate, (TPOT) with vigorous stirring. The mixture is kept stirred for 12 h at room
temperature, then heating is started at 40°C in
order to accelerate hydrolysis and evaporate the
ethyl alcohol, during which the inorganic precursors hydrolyze and polymerize into a network. The
surfactant containing mesoporous solid products
were recovered, air dried at room temperature and
®nally at 60°C for 24 h. Route II: An amorphous
mesoporous titanium containing silica Ti/Si ˆ
1:5%† was synthesized using tetraethyl orthosilicate, TEOS, as silicon source in a strong acidic
medium (2 M HCl solution), according to Ref.
[11]. (step 2) The surfactant containing mesoporous precursor was ®rst dried under vacuum at
60°C for 24 h. Then 20 g of the dried mesoporous
precursor was impregnated with 40 g of a 10 wt.%
solution of TPAOH (free from inorganic alkali).
After aging at room temperature for 12 h, the solid
was heated at 60°C for 24 h in order to eliminate
water, and left over night at room temperature
before being dried under vacuum for about 24 h at
room temperature. Finally, the solid was transferred into a Te¯on lined autoclave and heated at
120°C for several days. It is considered that the
quantity of water adsorbed on the solid plays an
important role in the crystallization. Therefore, the
partly crystalline solid was further crystallized at
the same temperature (120°C) for a given time
after introducing a small amount of water. Because the solid state crystallization continues in the
presence of this small amount of water, the above
process permits to control the crystallinity and the
mesopore size of the solid materials. The products
were washed with distilled water, dried in air at
80°C and ®nally calcined at 500°C for 6 h to remove the organics (heated from room temperature to 500°C with a heating rate of 1°C/min).
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
437
2.2. Instrumentation
3. Results
Powder X-ray di€raction (XRD) patterns of
the materials were recorded on a Philips X-ray
di€ractometer (PW 1010 generator and PW 1050
computer-assisted goniometer) using nickel-®l radiation, 0.025° step
tered CuKa k ˆ 1:5406 A†
size and a 1 s step time. Nitrogen adsorption and
desorption isotherms at 196°C were established
using an Omnisorp-100 apparatus. Prior to the
experiments, the samples were treated in vacuum
at 200°C for 5 h. The speci®c surface area, SBET ,
was determined from the linear part of the BET
equation P =P0 ˆ 0:05 0:3†. The calculation of
the mesopore size distribution was performed
using the desorption branch of the N2 adsorption/
desorption isotherms and the Barrett±Joyner±
Halenda (BJH) formula [13]. The mesopore surface area, SBJH and mesopore volume, VBJH were
obtained from the pore size distribution curves.
The average mesopore diameter, DBJH , was calculated as 4VBJH =SBJH . Although its accuracy is
limited [14±16], the BJH method, which is still
universally utilized in the mesoporous molecular
sieves (MMS) literature yields results that may
easily be compared with the current literature.
High-resolution transmission electron microscope
(TEM) images were obtained on a JEOL 200 CX
transmission electron microscope operated at 120
kV. Samples were embedded in a polymeric resin
and were cut into sections as thin as 20 nm with
an ultramicrotome. They were then deposited on
holey carbon copper grid before TEM observation. FTIR spectra were recorded on pellets of
the dehydrated samples diluted in KBr using a
Biorad FTS-60 spectrometer. Di€use re¯ectance
UV±visible spectra were collected on a Perkin±
Elmer Lambda 5 spectrophotometer interfaced
with an IBM computer using mesoporous silica
as the reference; the Kubelka±Munk function,
F R1 †, was applied. 29 Si MAS NMR measurements were performed at room temperature on a
Bruker ASX 300 spectrometer at a resonance
frequency of 59.71 MHz. 29 Si MAS NMR spectra were obtained with 30° pulse lengths of 1.75
ls, 180 s recycle delay and chemical shifts
were determined relative to tetramethylsilane, Si(CH3 )4 .
Two series of materials were synthesized either
by route I using SiCl4 or by route II with TEOS as
the silicon sources. They are respectively designated as I-[x]UL-TS1[y] and II-[x]UL-TS1[y],
where x and y are the crystallization time in days
and percentage Ti/Si, atomic ratio Ti/Si in the gel
which is essentially the same as the atomic ratio in
the ®nal products.
3.1. Nitrogen adsorption
Figs. 1 and 2 show nitrogen adsorption/desorption isotherms from the calcined samples after
®ve days of crystallization at 120°C, I-[5]ULTS1[1.5] and II-[5]UL-TS1[1.5]. The isotherms
exhibit a typical type IV, as de®ned by IUPAC
[18]. At low relative P =P0 pressure, a steep rise in
uptake, followed by a ¯at curve, corresponds to
®lling of micropores with nitrogen. A sharp in¯ection at higher pressures (e.g. in P =P0 range from
0.7 to 0.9) is characteristic of capillary condensation. The P =P0 position of the in¯ection point is
clearly related to a diameter in the mesopore range
and the sharpness of these steps indicates the
uniformity of the pore size distribution. All calcined materials give typical type IV adsorption/
desorption isotherms with a H1 hysteresis loop and
Fig. 1. Adsorption and desorption isotherm of nitrogen at 77 K
on the calcined sample (prepared from SiCl4 in ethanol) after 5
days of crystallization, I-[5]UL-TS1[1.5]. The insert shows the
BJH pore radius distribution calculated from the desorption
branch of the isotherm.
438
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
modi®ed to obtain more accurate pore size [14±
16]. Because we are principally interested in
changes in the pore size distribution, the BJH value can be used for this purpose. Average BJH
values of the pore radius are given in Table 1. The
pore size distribution is more clearly shown in Fig.
3. A signi®cant increase in the pore diameter and a
broader pore size distribution were observed, as a
function of crystallization time. This indicates
some modi®cation of the tubular channels of these
materials during crystallization.
The total speci®c surface area SBET of both series of samples is reported in Table 1. As the
crystallization time is increased, SBET varies from
820 to 580 m2 /g for I-[x]UL-TS1[1.5] and from 790
to 520 m2 /g for II-[x]UL-TS1[1.5]. The mesopore
surface area SBJH of the same materials varies from
645 to 180 m2 /g and 710 to 145 m2 /g respectively.
Simultaneously, the micropore volume increases
from 0.045 to 0.159 cm3 /g and from 0.025 to 0.149
cm3 /g for the same series of samples. The mesopore volume and radius are reported on Table 1.
Fig. 2. Adsorption and desorption isotherm of nitrogen at 77 K
on the calcined sample (prepared from TEOS in strong acidic
media) after 5 days of crystallization, II-[5]UL-TS1[1.5]. The
insert shows the BJH pore radius distribution calculated from
the desorption branch of the isotherm.
steep rises at low relative P =P0 pressure indicating
the presence of both mesopores and micropores in
UL-TS-1, even in the calcined [0]UL-TS-1 sample.
With increasing crystallization time, the UL-TS-1
materials give isotherms with similar in¯ection but
with reduced sharpness and a shift toward higher
P =P0 values over a larger P =P0 range (not shown).
The BJH pore radius distribution can be calculated from the Kelvin equation and has been
widely used for mesoporous materials. Recent
works suggest that the Kelvin equation should be
3.2. X-ray di€raction
Small-angle XRD of UL-TS-1 were not systematically obtained since low enough angle data
cannot be collected with our XRD instrument.
However, reliable information about the e€ect of
Table 1
Physico-chemical properties of the UL-TS1 materials
Materials
Crystallization time
(days)
SBET
(m2 /g)
SBJH
(m2 /g)
Micropore
volume (cm3 /g)
Mesopore
volume (cm3 /g)
Pore radius
(nm)
Crystallinity
(%)
I-[0]UL-TS1[0.5]
I-[5]UL-TS1[0.5]
0
5
760
620
680
450
0.038
0.095
1.62
1.72
4.2
9.6
±
18
I-[0]UL-TS1[1.5]
I-[3]UL-TS1[1.5]
I-[5]UL-TS1[1.5]
I-[8]UL-TS1[1.5]
I-[10]UL-TS1[1.5]
0
3
5
8
10
820
710
680
655
580
645
495
405
310
180
0.045
0.082
0.123
0.145
0.159
1.45
1.55
1.75
1.95
1.65
4.4
6.3
9.2
11.2
13.2
±
±
12
45
65
II-[0]UL-TS1[1.5]
II-[5]UL-TS1[1.5]
II-[8]UL-T 1[1.5]
II-[10]UL-TS1[1.5]
0
5
8
10
790
730
705
520
710
440
380
145
0.025
0.103
0.103
0.149
1.18
1.40
1.36
0.85
3.3
4.8
6.2
6.4
±
18
54
80
I-[x]UL-T 1[y] where I: route I for the synthesis of precursor materials; x and y: crystallization time in days and percentage of Ti/Si,
respectively; II: route II for the synthesis of precursor materials.
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
439
Fig. 3. (A) BJH pore radius distribution curves for the I-UL-TS1[1.5] sample at various times of crystallization: (a) 3 days, (b) 5 days
and (c) 10 days and (B) BJH pore radius distribution curves for the II-UL-TS1[1.5] sample at various times of crystallization: (a) 0
days, (b) 5 days and (c) 10 days.
crystallization conditions on the mesopore structure of UL-TS1 has been obtained from nitrogen
adsorption experiments and from TEM images.
The crystalline phase in UL-TS-1 upon crystallization was however characterized by wide-angle
XRD di€ractograms, as shown in Fig. 4A and 4B
for the samples I-UL-TS1[1.5] and II-UL-TS1[1.5],
respectively. The mesoporous precursor with
amorphous walls (only the broad feature of
amorphous phase appears, Fig. 4A(a) and 4B(a))
provides a starting material from which nanocrystalline domains can nucleate within the walls.
The XRD di€ractograms of the calcined ULTS1[1.5] samples in Fig. 4A(b) and 4B(b) show
broad peaks, which match those of ZSM5. These
peaks grow in intensity as the crystallization time
Fig. 4. (A) XRD patterns of the calcined I-UL-TS1[1.5] sample (prepared from SiCl4 in ethanol) after various times of crystallization:
(a) 0 days, (b) 8 days and (c) 10 days and (B) XRD patterns of the calcined II-UL-TS1[1.5] sample (prepared from TEOS in strong
acidic media) after various times of crystallization: (a) 0 days, (b) 8 days and (c) 10 days.
440
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
is increased (Fig. 4A(c) and 4B(c)). The relative
increase in the intensity of the characteristic triplet
in the 2h range 21.5±25.5° are shown in Table 1.
Considering TS-1 as 100% crystalline [17], 80%
crystallinity has reached after 10 days of crystallization at 120°C following synthesis route II. The
XRD spectra observed after di€erent times of
crystallization are quite similar for both series of IUL-TS1[1.5] and II-UL-TS1[1.5] prepared by the
two di€erent routes. These data indicate that the
initially amorphous walls of the two mesoporous
materials are progressively transformed into crystalline nanoparticles.
3.3. Transmission electron microscopy
The pore structure of mesoporous materials is
directly visible by transmission electron microscopy. The mesoporous precursor prepared from
SiCl4 (route I) appears to be of a uniform pore size
with a highly disordered pore structure. This is
reminiscent of MSU-1 and KIT-1 mesoporous
materials [19,20], which have wormhole-like pore
frameworks (Fig. 5A). In contrast, a well-ordered
pore structure (Fig. 5B) was observed for the
precursor sample prepared from TEOS (route II).
Fig. 6 shows the TEM images of the samples after
10 days of crystallization, I-[10]UL-TS1[1.5] and
II-[10]UL-TS1[1.5]. Fig. 6A shows that after
crystallization in the presence of TPAOH, the
mesopores of the MSU type precursor (Fig. 5A)
retain their size and morphology. The wormhole
pore lattice is however still present and microdomains of the order of 10 nm are observed. It is
interesting to note that the size of MFI microdomains calculated using the Scherrer formula from
the line broadening in the XRD spectrum of this
sample is 20 nm, which is consistent with these
observations. The speci®c surface calculated for
an average TS-1 particle diameter of 20 nm is 200
m2 /g which matches the SBJH value calculated from
BET data (Table 1). From Fig. 6B it is seen that
the hexagonal structure of the precursor SBA
phase (Fig. 5B) is transformed after the crystallization step. The diameter of the regularly arranged
pores is signi®cantly enlarged from 5 to 10 nm
Fig. 5. TEM images of the calcined mesoporous titania±silica precursor Ti/Si ˆ 1:5%† (A) prepared from SiCl4 (route I) and (B) from
TEOS (route II).
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
441
Fig. 6. TEM images of the calcined samples after 10 days of crystallization in the presence of TPAOH, (A) I-[10]UL-TS1[1.5] and (B)
II-[10]UL-TS1[1.5].
which matches the measured BJH pore diameter of
6.6 and 12.8 nm, respectively (see Table 1). Fig. 6B
also shows nanoparticles of TS-1 having grown to
10 nm size and being slightly agglomerated. The
pore walls themselves show a discontinuous
structure compared to the precursor suggesting
that nucleation of TS-1 begins in these walls.
tetrahedral coordination. No band at 330 nm characteristic of octahedral extra-framework titanium
was observed [21,22]. This suggests that all titanium is essentially incorporated in the UL-TS1
framework.
3.4. UV±visible di€use re¯ectance spectroscopy
Fig. 8 shows the FTIR spectra of a series of II[x]UL-TS1[1.5] samples with various times of
crystallization. The pure SBA silica sample exhibits spectroscopic features similar to those of
amorphous mesoporous silica, a broad bands at
985 cm 1 assigned to silanol groups on the wall
surface is present (Fig. 8a). However, for the II[0]UL-TS1[1.5] sample, a band at 965 cm 1 which
is characteristic of framework titanium is shown
and no band at 550 cm 1 was observed. The band
at 985 cm 1 disappears progressively, while the
bands at 550 and 965 cm 1 develop with increasing
crystallization time. The corresponding FTIR
spectra for the series I-[x]UL-TS1[1.5] are similar
UV±visible spectroscopy has been extensively
used to characterize the nature and coordination
of titanium ions in titanium substituted molecular
sieves. The ultraviolet absorption wavelength of
titanium is sensitive to its coordination and to
TiO2 particle size. Fig. 7 shows UV±visible spectra
of two series of UL-TS1[1.5] samples prepared by
route I and route II, with di€erent crystallization
times. Only a single intense large band at 230 nm
was observed with all samples. This band was
attributed to ligand-to-metal charge transfer
associated with isolated Ti4‡ framework sites in
3.5. FTIR
442
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
Fig. 7. (A) Di€use re¯ectance UV±Visible spectra of the calcined samples prepared from SiCl4 after various times of crystallization: (a)
0 days, (b) 5 days and (c) 10 days and (B) Di€use re¯ectance UV±Visible spectra of the calcined samples prepared from TEOS after
various times of crystallization: (a) 0 days, (b) 5 days and (c) 10 days.
to those in Fig. 8 and they are not reported here
for sake of brevity. Several researchers have assigned the 550 cm 1 band to the asymmetric
stretching mode of the ®ve-membered ring present
in ZSM-5 which should therefore be an indication
of the presence of the MFI structure of TS-1
[23,24]. Splitting of this lattice-sensitive band into
a doublet has been observed in nanophase silicalite
[25]. The FTIR spectra of the samples in Fig. 8
show the doublet band at 561=547 cm 1 and the
band at 965 cm 1 , which are characteristic of
nanocrystals and titanium framework, respectively
[26].
3.6.
29
Si MAS NMR
Fig. 9 shows the 29 Si MAS NMR spectra of the
amorphous mesoporous Ti-material, I-[0]ULTS1[1.5] and of the sample obtained after 10 days
of crystallization, I-[10]UL-TS1[1.5]. The mesoporous titania±silica exhibits a 29 Si MAS NMR
spectrum typical of amorphous materials; two
main resonances at 114 and 104 ppm, and a
very weak peak at
98 ppm correspond to
Si(OSi)4 (Q4 ), Si(OSi)3 (Q3 ) and Si(OSi)2 (Q2 ) silicate species, respectively. The ratio of the relative
peak areas of the deconvoluted peaks, Q4 /Q3 , was
1.8. This ratio was comparable with other calcined
amorphous mesoporous silicas [9]. Upon crystallization for 10 days in the presence of the TPA‡
structure-directing agent, the 29 Si MAS NMR
spectrum showed the main resonance (Q4 ) at 114
ppm along with the only weak resonance (Q3 ) at
104 ppm from surface hydroxyl groups and the
resonance (Q2 ) at 98 ppm had disappeared. The
29
Si MAS NMR spectra of series II-[x]ULTS1[1.5] show the same trends and are therefore
not reported here. The increase in the intensity of
the Q4 resonance and concomitant decrease in the
intensity of the Q3 resonance re¯ect the crystallization process and the transformation of the hydrophilic surface into a hydrophobic one.
4. Discussion
The original aspect of this work consists in
using the highly dispersed amorphous material of
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
443
Fig. 9. 29 Si MAS-NMR spectra of: (a) the calcined precursor
material and (b) the calcined sample after 10 days of crystallization, I-[10]UL-TS1[1.5].
Fig. 8. FTIR spectra of: (a) the calcined SBA silica sample and
the II-[x]UL-TS1[1.5] sample after various times of crystallization, (b) 0 days, (c) 5 days, (d) 8 days and (e) 10 days.
a mesoporous molecular sieve as a precursor phase
to generate unusual mesostructured zeolitic materials (UL-zeolites) by secondary crystallization.
The controlled size of the pore walls is taken advantage of, in controlling the crystal size of the
zeolite product. In this process, zeolite nanoparticles are grown from the mesopore walls and at the
same time the end product retains some of the
mesopore surface of the precursor. As a consequence a rather high mesoporous surface is constituted of the zeolite particles so that this zeolite
surface is accessible to large molecules. It is thus
expected that UL-TS-1 will allow the typical oxidation reactions catalyzed by TS-1 but for large
molecules. Actually, such e€ects are also expected
with all UL-zeolites which should allow to extend
catalysis to large molecules, not converted in usual
zeolites due to spatial restrictions. Another important structural feature of UL-TS-1 is the small
size of the TS-1 nanoparticles which limits the
internal di€usion path length within the zeolite
microporous network. These characteristics of
UL-TS-1 are also of interest in both catalysis and
adsorption/desorption of small molecules which
are expected to be faster than in usual TS-1.
The data reported in this work indicate that the
pore and wall geometry are strongly a€ected by the
secondary crystallization process. As shown in
both Table 1 and Fig. 10, the mesopore volume
and pore radius increase with crystallinity until
35%. It is believed that in these conditions, the
pore walls shrink due to the change in density
associated with the formation of the crystalline
silicalite phase. At higher crystallinity, the mesopore volume begins to decrease even though the
mesopore size is little a€ected. This corresponds to
the migration of the material constituting the walls
which contributes to the extra-walls growth of
silicalite nanoparticles (see Fig. 6B). This process
results in an important damage to the initial regularity of the mesopore network but a signi®cant
fraction of the initial mesopore surface is preserved
in the end product.
444
D. Trong On et al. / Microporous and Mesoporous Materials 44±45 (2001) 435±444
References
Fig. 10. Evolution of the mesopore volume and average radius
as a function of the percentage of crystallinity: (d) I-[x]ULTS1[1.5] and ( ) II-[x]UL-TS1[1.5].
5. Conclusions
Meso-structured titanium±silicalite materials,
UL-TS-1, were synthesized by a secondary templated crystallization of zeolites starting from the
amorphous titanium containing mesoporous molecular sieves. The results revealed that UL-TS-1
contain both micro an d mesopores and all titanium ions are essentially incorporated in the ULTS-1 framework. The mesopore provides an easy
access to the external surface of the zeolite nanoparticles. Moreover, the relatively short di€usion
pathways through the thin walls are expected to
improve reaction eciency and minimize channel
blocking. Detailed studies of the sorption behavior
and catalytic properties are currently underway.
Acknowledgements
The authors wish to thank Dr. S.M.J. Zaidi and
Mr. D. Poisson for help in recording NMR and
UV±visible data. This work is partly ®nanced by
SiliCycle Inc and FCAR.
[1] D.W. Breck, Zeolite Molecular Sieves: Structure, Chemistry and Use, Wiley, London, 1974.
[2] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S.
Beck, Nature 359 (1992) 710±712.
[3] J.S. Beck, et al., J. Am. Chem. Soc. 114 (1992) 10834.
[4] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson,
B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548.
[5] A. Corma, Chem. Rev. 97 (1997) 2413.
[6] D. Trong On, S. Kaliaguine, US patent pending, November 1999.
[7] D. Trong On, S. Kaliaguine, unpublished results.
[8] M. Taramasso, G. Perego, B. Notari, US patent 4 410 501,
1983.
[9] A.C. Voegtlin, A. Matijasic, J. Patarin, C. Sauerland, Y,
Grillet, L. Huve, Micropor. Mater. 10 (1997) 137.
[10] J.E. Gallot, S. Kaliaguine, Can. J. Chem. Eng. 76 (1998)
833.
[11] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson,
B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548.
[12] P. Yang, D. Zhao, D.I. Margolese, B.F. Chmelka, G.D.
Stucky, Nature 396 (1998) 152.
[13] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem.
Soc. 73 (1951) 373.
[14] P.I. Ravikovitch, D. Wei, W.T. Chueh, G.L. Haller, A.V.
Neimark, J. Phys. Chem. B 101 (1997) 3671.
[15] W.W. Luken Jr., P. Schmidt-Winkel, D. Zhao, J. Feng,
G.D. Stucky, Langmuir 15 (1999) 5403.
[16] M. Kruk, M. Jaroniec, Langmuir 15 (1999) 5279.
[17] D. Trong On, M.P. Kapoor, E. Thibault, J.E. Gallot, G.
Lemay, S. Kaliaguine, Micropor. Mesopor. Mater. 20
(1998) 107.
[18] K.S.W. Sing, D.H. Everett, R.A. Haul, L. Moscow, R.A.
Pierotti, J. Rouquerol, T. Siemieniewska, Pure Appl.
Chem. 57 (1985) 603.
[19] S.A. Bagshaw, E. Prouzet, T.J. Pinnavaia, Science 269
(1995) 1242.
[20] R. Ryoo, J.M. Kim, C.H. Ko, C.H. Shin, J. Phys. Chem.
100 (1996) 17718.
[21] D. Trong On, L. Le Loc, L. Bonneviot, J. Chem. Soc.,
Chem. Commun. (1996) 299.
[22] D. Trong On, M.P. Kapoor, S. Kaliaguine, J. Chem. Soc.,
Chem. Commun. (1996) 1161.
[23] P.A. Jacobs, E.G. Derouane, J. Weitkamp, J. Chem. Soc.,
Chem. Commun. (1981) 591.
[24] B.J. Shoeman, in: H. Chon, S.-K. Ihm, Y.S. Uh (Eds.),
Progress in Zeolite and Microporous Materials, Stud. Surf.
Sci. Catal., vol. 105, 1997, p. 647.
[25] R. Ravishankar, C. Kirschhock, B.J. Schoeman, P.
Vanoppen, P.J. Grobet, S. Storck, W.F. Maier, J.A.
Martens, F.C. De Schryver, P.A. Jacobs, J. Phys. Chem.
B 102 (1998) 2633.
[26] D. Trong On, S. Kaliaguine, L. Bonneviot, J. Catal. 157
(1995) 235.