Confinement effects at the external surface of delaminated zeolites (ITQ-2): An inorganic mimic of cyclodextrins

Article abstract of DOI:10.1021/jp0210531

Novel delaminated ITQ-2 zeolite has a remarkably large accessible external surface area (a??800 m² g^-1) and reduced microporosity (0.009 cm^-3 g^-1) and has attracted interest as an alternative to conventional zeolites or mesoporous MCM-41 (Corma, A.; Forneì?s, V.; Pergher, S. B.; Maesen, T. L. M.; Buglass, J. G. Nature 1998, 396, 353-356). ?±,??-Diphenyl polyenes have been used as molecular probes to check the ability of ITQ-2 zeolite to generate organic radical cations. Of these probes, only t,t-1,4-diphenyl-1,3-butadiene (DPB) is transformed into a persistent reaction intermediate upon adsorption on ITQ-2. The process occurs in the cup -like cavities open to the exterior since selective silylation of the cups inhibits completely the generation of this reaction intermediate. Detection of 1-phenylnaphthalene as reaction product, EPR spectroscopy, and alternative laser flash photolysis generation strongly support 1-phenylnaphthalene radical cation as the intermediate formed after the adsorption of DPB onto ITQ-2. This contrasts with the behavior of conventional zeolites ZSM-5 and mordenite in which DPB^ì?+ is the only species formed and demonstrates the uniqueness of the behavior of ITQ-2 as result of its unprecedented topology.

Full text of DOI:10.1021/jp0210531

J. Phys. Chem. B 2003, 107, 1135-1141
1135
Confinement Effects at the External Surface of Delaminated Zeolites (ITQ-2): An Inorganic
Mimic of Cyclodextrins
Mar´ıa S. Galletero, Avelino Corma,* Bele´n Ferrer, Vicente Forne´s, and Hermenegildo Garc´ıa*
Instituto de Tecnolog´ıa Qu´ımica CSIC UPV, UniVersidad Polite´cnica de Valencia,
AVda de los Naranjos s/n, 46022-Valencia, Spain
ReceiVed: April 25, 2002; In Final Form: September 5, 2002
Novel delaminated ITQ-2 zeolite has a remarkably large accessible external surface area (∼800 m2 g-1) and
reduced microporosity (0.009 cm-3 g-1) and has attracted interest as an alternative to conventional zeolites
or mesoporous MCM-41 (Corma, A.; Forne´s, V.; Pergher, S. B.; Maesen, T. L. M.; Buglass, J. G. Nature
1998, 396, 353-356). R,ω-Diphenyl polyenes have been used as molecular probes to check the ability of
ITQ-2 zeolite to generate organic radical cations. Of these probes, only t,t-1,4-diphenyl-1,3-butadiene (DPB)
is transformed into a persistent reaction intermediate upon adsorption on ITQ-2. The process occurs in the
“cup”-like cavities open to the exterior since selective silylation of the cups inhibits completely the generation
of this reaction intermediate. Detection of 1-phenylnaphthalene as reaction product, EPR spectroscopy, and
alternative laser flash photolysis generation strongly support 1-phenylnaphthalene radical cation as the
intermediate formed after the adsorption of DPB onto ITQ-2. This contrasts with the behavior of conventional
zeolites ZSM-5 and mordenite in which DPB•+ is the only species formed and demonstrates the uniqueness
of the behavior of ITQ-2 as result of its unprecedented topology.
Introduction
SCHEME 1
R,ω-Diphenyl polyenes have been used as probe molecules
to demonstrate the ability of zeolites to generate a high
population of persistent organic radical cations.1,2 These mol-
ecules present the advantage of having similar kinetic diameter
and molecular shape, varying exclusively the chain length and
the oxidation potential, which gradually decreases as the number
of CdC increases. It was found that these R,ω-diphenyl
polyenes may form indefinitely persistent radical cations when
they are incorporated within the channels of medium pore size
ZSM-5 zeolite.3,4 It was proposed that the terminal phenyl rings,
tightly fitting with the zeolite framework, protect the inner
polyene bridge bearing the unpaired electron (Scheme 1). The
role of phenyl rings acting as stoppers in the channels of ZSM-
5, as originally proposed, was also expanded to other types of
organic reactive intermediates such as R,ω-diphenyl allyl
cations.5 In addition to ZSM-5, the fate of R,ω-diphenyl
polyenes on other conventional zeolites such as Y, Beta, and
mordenite has also been studied in an attempt to establish a
relationship between the zeolite structure and the ability to
generate persistent organic radical cations.1,3,4,6
In recent years, many novel zeolite structures have been
synthesized. It is therefore of interest to study the behavior of
well-documented probes with novel zeolitic materials in order
to assess whether new topological effects, not previously found
in conventional zeolites, can be observed. In fact, given the
importance of zeolites from an economical point of view, the
synthesis of novel structured porous materials constitutes a very
active area of research.7-10 In this regard, the preparation of
mesoporous aluminosilicates of the MCM-41 type was a
significant breakthrough because these materials combine a very
large surface area and a regular pore size, the latter being
controlled during the synthesis (20-100 nm).11 The ability of
mesoporous MCM-4112,13 to generate a stationary population
of organic radical cations is generally very low, something that
agrees with the previous tight-fit paradigm to rationalize the
persistence of radical cations. Given the large dimensions of
the channels of MCM-41, most organic radical cations will
experience a loose fit and will not be protected from nucleophilic
attacks.
In this context, a new type of zeolitic material described as
delaminated zeolites has been recently reported.14 One example
of such materials is the zeolite ITQ-2 that was prepared from
the layered precursor of MCM-22 by delamination of the
swollen precursor followed by calcination (Scheme 2). The
ITQ-2 solid is formed by an array of disordered zeolitic layers.
These layers are essentially those found in the MCM-22
precursor and they encompases two different topological units
consisting of open cups of ∼7 Å of diameter and ∼8 Å depth
located at the upper and lower planes of the layers and
independent, not connected, sinusoidal 10-membered ring (10
MR) intralayer channels. Figure 1 shows the structure of the
ITQ-2 layers wherein these two morphological components have
been highlighted.
Given that the topology and geometrical features of ITQ-2
are not encountered in any conventional zeolite, it is of interest
to address whether the very large external surface (∼800 m2
g-1) of these materials behaves as a geometrically nonsensitive
* Corresponding authors.
10.1021/jp0210531 CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/09/2003

1136 J. Phys. Chem. B, Vol. 107, No. 5, 2003
Galletero et al.
Figure 1. Structure of the ITQ-2 layers showing the two morphological components (cups and channels) defined by the framework.
SCHEME 2
surface or whether, on the contrary, molecules can experience
a well-structured external surface with open cups wherein
molecular confinement may occur. If the latter case, it should
be possible to generate a significant population of persistent
organic radical cations, even though these radical cations would
be not located deep inside the zeolite particle but in the shallow
layer of the external surface. The aim of this research is to
establish whether the structured external surface area of ITQ-
2, being easily accessible like that of MCM-41, shares or not
with zeolites the characteristic behavior observed in the confined
spaces of zeolites.
available.3,4,6 For the present study we have used as probes R,ω-
diphenyl polyenes from 1 to 4 CdC double bonds.
These molecules are expected to diffuse freely into the open
cups of ITQ-2, but on the contrary it is anticipated that diffusion
into the 10 MR sinousoidal pores present in the layer of this
material should be considerably more impeded due to the
continuous direction change of the channels.15
To address this point we have selected R,ω-diphenyl polyenes
that constitute a homogeneous series of probes wherein the
molecular properties change regularly and for which sufficient
information about their behavior in conventional zeolites is
Experimental Section
R,ω-Diphenyl polyenes, 1-PNH and TP+ (as a BF4- salt) were
commercial samples and used as received. ZSM-5 and ITQ-2

Inorganic Mimic of Cyclodextrins
J. Phys. Chem. B, Vol. 107, No. 5, 2003 1137
TABLE 1: Relevant Physicochemical Parameters of the Porous Aluminosilicates Used in the Present Work
BET surface
area (m2/g)
crystal size
external surface
area (m2/g)
micropore volume
(cm3/g)
topological
characteristics
zeolite
ZSM-5
Na-mordenite
MCM-22
ITQ-2
Al/MCM-41
Si/Al ratio
(µm)
≈40
6.5
50
58
13
386
356
400
806
874
0.5-1
0.1-0.5
≈1
46.26
20.23
∼150
750
0.111
0.165
0.121
0.009
0.46
microporosity internal surface
microporosity internal surface
microporosity internal surface
no microporosity external surface
mesoporosity internal surface
<0.5
0.1-0.3
∼150
were prepared by hydrothermal sol-gel crystallization according
to the detailed procedures reported in the literature.14,32 The
crystal structure of the synthesized aluminosilicates was con-
firmed by powder X-ray diffraction. Silylation of ITQ-2 was
carried out by stirring a suspension of the dehydrated layered
ITQ-2 zeolite (1 g) with hexamethyldisilazane (0.5 g) in toluene
(25 mL) at reflux temperature for 2 h. After this time, the solid
was filtered, exhaustively washed with CH2Cl2 and dried.
Adsorption of R,ω-diphenyl polyenes on the solids was carried
out by stirring magnetically, at reflux temperature for 2 h, a
suspension of the organic compound (10 mg) in isooctane (5
mL) in the presence of the corresponding solid (180 mg) that
had been previously dehydrated by heating at 500 °C overnight.
The solid was filtered, washed with fresh solvent (5 mL), dried
under vacuum and stored in closed vials at the ambient. 1-PNH
was identified by comparison with a commercial sample. Cis
isomer of DPN was obtained by 250 W mercury lamp irradiation
through Pyrex of a 10-2 M solution of DPB in hexane for 3 h.
Diffuse reflectance UV-Vis spectra were recorded using a Cary
5 spectrophotometer adapted with a “praying mantis” diffuse
reflectance accessory and using pristine ZSM-5 as reference.
FTIR spectra of pyridine and 2,6-di-tert-butylpyridine adsorption
on ITQ-2 before and after silylation were recorded in a Nicolet
710 FT spectrophotometer using greaseless quartz cells having
CaF2 windows and a stopcock valve that allowed them to be
sealed under high vacuum. Pyridine and 2,6-di-tert-butylpyridine
vapors were admitted at room temperature at 10-1 Pa onto
dehydrated silylated and unsilylated ITQ-2 samples placed in
the IR cell. The cell was evacuated at 250 °C under 10-2 Pa
for 1 h and the intensity of the IR spectra corresponding to the
residual retained pyridine compared for the two samples. EPR
spectra were recorded at -200 °C on a Brucker ER200D
spectrometer working at the X band (9.65 GHz) and using DPPH
(g ) 2.0036) as reference. Laser flash photolysis experiments
were carried out using for excitation the third (355 nm, e 20
mJ×pulse-1) harmonic of a Surelite Nd:YAG laser (pulse e
10 ns). The signal from the monochromator/photomultiplier
detection system were captured by a Tektronix MDSA digitizer
and transferred to computer that controlled the experiment and
provides suitable processing and data storage capabilities.
Fundamentals and details of similar time-resolved diffuse
reflectance laser setup has been published elsewhere.34
Figure 2. Diffuse reflectance UV-vis spectra (plotted as the Kubelka-
Munk function F(R)) of stilbene (a) t,t-1,6-diphenyl-1,3,5-hexatriene
(b) and all trans-1,8-diphenyl-1,3,5,7-octatetraene (c) adsorbed on ITQ-
2.
from isooctane solutions onto ITQ-2 does not give the charac-
teristic optical spectra reported for the radical cations of these
compounds in ZSM-5 and other zeolites (see Figure 2). We
notice that in the case of t,t,t-1,6-diphenyl-1,3,5-hexatriene
although the diffuse reflectance UV-Vis spectrum rules out
the presence of the corresponding radical cation (λmax 597 nm
shoulder 540, 766, and 802 nm), some extra bands in addition
to those of the starting material can be observed. Something
similar happens for all trans-1,8-diphenyl-1,3,5,7-octatetraene.
On the other hand, we have been able to reproduce the results
reported for these three R,ω-diphenyl polyenes in conventional
acidic ZSM-5 in which the corresponding radical cations are
formed.1
Results and Discussion
Compared to conventional zeolites, the delaminated ITQ-2
exhibits a much larger surface area (predominantly external)
with a reduced microporosity. In this regard, the ITQ-2 zeolite
is similar to MCM-41 in terms of having a large accessible
surface, but the zeolite ITQ-2 layers are crystalline and they
contain smaller cavities rather than mesoporous channels. Table
1 summarizes the main physicochemical and textural parameters
of the samples used in the present work. In this table the major
differences of ITQ-2 compared with conventional zeolites and
mesoporous MCM-41 have been remarked.
Incorporation of t-1,2-diphenylethylene, t,t,t-1,6-diphenyl-
1,3,5-hexatriene and all trans-1,8-diphenyl-1,3,5,7-octatetraene
In sharp contrast, adsorption of t,t-1,4-diphenyl-1,3-butadiene
(DPB) gives a well defined optical spectrum different totally

1138 J. Phys. Chem. B, Vol. 107, No. 5, 2003
Galletero et al.
Figure 3. Diffuse reflectance UV-vis spectra of DPB@ITQ-2 sample (recorded periodically every week since its preparation). Inset: transient
spectrum of 1-PNH (b). The spectrum has been recorded 1 µs after the 355 nm laser flash.
that could give some hints about the intermediate. In the case
here considered, analysis of the solvent after incorporation of
DPB and exhaustive solid-liquid extraction of the sample
allowed us to observe the presence of 1-phenylnaphthalene (1-
PNH) as the only detectable product. In an attempt to confirm
that 1-PNH is the immediate precursor of the observed species
present in ITQ-2, 1-PNH was adsorbed from a CH2Cl2 onto
thermally dehydrated ITQ-2. However, the solid did not become
colored and diffuse reflectance UV-vis spectroscopy did not
reveal the presence of the 550-700 nm absorption band
characteristic of the species generated from DPB. However,
when interpreting this negative result it has to be taken into
account that the sites/locations accessible for less bulky DPB
are not the same as those for 1-PNH which has a larger kinetic
diameter and more restricted diffusion.
Two reaction mechanisms (acid catalyzed and electron-
transfer mediated) can be envisioned to account the formation
of 1-PNH. Thus, to provide evidence about the origin of 1-PNH
Figure 4. Diffuse reflectance UV-vis spectrum (plotted as the
Kubelka-Munk function) of DPB@ZSM-5.
from DPB, two sets of experiments were carried out in solution.
In one of them, an acetonitrile solution of DPB was stirred at
room temperature in the presence of catalytic amounts of
concentrated sulfuric acid. Under these conditions, cis isomers
were formed together with lesser amounts of 1-PNH. This result
indicates that DPBH+ generated from DPB by protonation is
able to undergo cyclization to form the dihydronaphthalene
skeleton (Scheme 3). It has been reported that adsorption of
DPB onto CaY gives 3,4-dihydro-1-phenylnaphthalene together
with two other byproducts.16 Aromatization of this dihydro
precursor (compound not observed in the experiments either in
ITQ-2 or in sulfuric acid) would be achieved either by oxidation
or (most likely in view of the presence of carbocation
intermediate) by hydride transfer. Conversion of 3,4-dihydro-
1-phenylnaphthalene to 1-PNH through a hydride transfer
mechanism has been observed in CaY accompanied with
1-phenyltetraline.17
from that of the starting material. This spectrum is characterized
by a broad absorption band from 550 to 700 nm. The intensity
of this absorption band decreases over the time elapsed after
sample preparation until a final stationary spectrum is obtained
after three months (Figure 3). It is noteworthy that this optical
spectrum does not correspond to the DPB•+ radical cation or to
DPBH+, the only reaction intermediates previously reported in
the incorporation of DPB in acidic zeolites.1,2 Adsorption of
DPB in HY does not give rise to the same transient because
the diffuse reflectance spectrum of the DPB@HY is completely
different (continuous absorption without resolved peaks). For
the sake of comparison, the diffuse reflectance UV-vis
spectrum recorded by us for the DPB@ZSM-5 sample under
the same conditions, corresponding to DPB•+ radical cation, is
shown in Figure 4.
First we have investigated the nature of this unique interme-
diate responsible for the spectrum shown in Figure 3. When
addressing the nature of the adsorbate a good clue is to analyze
the reaction mixture after DPB adsorption, looking for products
In agreement with these reported results for CaY, our own
study adsorbing DPB onto HY in isooctane at 100 °C leads to

Inorganic Mimic of Cyclodextrins
J. Phys. Chem. B, Vol. 107, No. 5, 2003 1139
SCHEME 3
To further exclude DPB•+ as the reaction intermediate leading
to the formation of 1-PNH, laser flash photolysis of DPB
adsorbed onto NaY was carried out. NaY being an inert zeolite
devoid of acidity does not promote any reaction of DPB. Laser
excitation (308 nm) should generate some DPB•+ through
photoionization, a general phenomenon known to occur on the
polar environment of zeolite pores.28-30 However, even under
these conditions, no evidence for the formation of 1-PNH could
be obtained and DPB remained unaltered.
the complete disappearance of the starting material and the
formation of one isomer compatible with dihydronaphthalene
(60%) together with equal amounts of 1-PNH (21%) and a
hydrogenated derivative (19%) tentatively assigned as 1,4-
diphenyl-1-butene based on the MS. Minor amounts of a
complex mixture of tert-butylated products arising from the acid-
catalyzed reaction of DPB with isooctane were also observed.
This product distribution is also in agreement with Scheme 3
if the hydride acceptor (C+) were DPB-H+, whereby 1,4-
diphenyl-1-butene could be formed. Interestingly, the HY zeolite
did not become colored and, according to the diffuse reflectance
UV-vis spectra of the HY zeolite, did not exhibit the 550-700
nm band observed for the case of ITQ-2. Therefore, cyclization
of DPB is a general phenomenon that occurs either in solution
or in zeolite Y. However, it must be noted that what is unique
about cyclodextrin-like ITQ-2 cups is the formation of a
unprecedented intermediate as a persistent (month) species.
On the other hand, the possibility that DPB•+ radical cation
could also intervene in the formation of 1-PNH was studied by
photochemically generating DPB•+ using 2,4,6-triphenylpyry-
lium cation (TP+) as photosensitizer under aerobic conditions
(eq 1).18
Laser flash photolysis studies reveal that upon excitation of
TP+ and intersystem crossing to the TP+ 3 triplet, this excited
state is quenched by DPB through an electron transfer mech-
anism. However, despite the evidences of DPB•+ radical cation
generation in the TP+-sensitized photolysis, the only photopro-
cess observed in a preparative irradiation was cis-trans isomer-
ization, and 1-PNH could not be detected in the reaction mixture.
Our observations are consistent with previous reports on the
chemical behavior of DPB•+ in which only cis-trans isomer-
ization was observed19-27 and rule out the intermediacy of
DPB•+ on the formation of 1-PNH.
Given that 1-PNH is the only product observed and that the
diffuse reflectance UV-vis spectrum of the DPB@ITQ-2 does
not probably correspond to any DPB intermediate, we consid-
ered the possibility that the spectrum shown in Figure 3
corresponds to a 1-PNH derived intermediate. As a matter of
fact, EPR measurements of DPB@ZSM-5 and DPB@ITQ-2
samples (Figure 5) showed a similar spin density. The relative
area of the ZSM-5/ITQ-2 EPR signals obtained by double
integration was 3.47. Based on previous quantification of the
spin count for organic radical cation inside ZSM-5, or estimation
that the population of radical cations for the DPB@ZSM-5 and
DPB@ITQ-2 is in the order of 1019 spin × g-1, that result
corresponds to about 0.5 mg of radical cation per g of zeolite.
The comparable amounts of a high population of radical species
measured for both samples is compatible with the presence of
organic radical ions. Incidentally noteworthy is the fact that the
number of spins per gram for ZSM-5 is very similar to ITQ-2,
even though the topology of the latter is characterized by the
large accessible surface area, compared to the medium pore size
channels characteristic of ZSM-5.
The above EPR data suggest that the species present in the
DPB@ITQ-2 sample could be the radical cation of 1-PNH.
Thus, after formation of 1-PNH from DPB within the ITQ-2
cups, an electron transfer from 1-PNH to the acid sites of the
zeolite would generate 1-PNH•+. Electrochemical measurements
in acetonitrile give an oxidation potential for 1-PNH of 1.55 V
vs SCE that is comparable to that of the parent naphthalene.
The optical spectrum of 1-PNH•+ has not been previously
reported. To record the authentic optical spectrum of 1-PNH•+
that could be compared with the diffuse reflectance UV-vis

1140 J. Phys. Chem. B, Vol. 107, No. 5, 2003
Galletero et al.
1-PNH•+) compared to conventional ZSM-5 and mordenite
(generation of DPB•+). In fact, our observation of 1-PNH•+
radical cation generation constitutes one of the clearest examples
of a topologically controlled transformation due to the differ-
ences in the geometry of the zeolites. It is worth mentioning
that ITQ-2 shares with classical zeolites the ability to generate
a persistent population of organic radical cations (Figure 5).
The question now becomes where these organic radical cations
are formed. DPB is adsorbed on the straight 10 MR channels
of ZSM-5. However, in the case of ITQ-2, the 10 MR channels
are sinusoidal. It is, therefore, expected that the diffusion of
DPB through the 10 MR sinousoidal channel must be severely
restricted.15 Our hypothesis is that 1-PNH•+ is located at the
open cups of 7-Å diameter and 8-Å depth at the upper and lower
planes of the layers. If this was so, we should expect that in the
case of the nondelaminated MCM-22 zeolite, which has the
higher amount of 10 MR sinusoidal channels but a much lesser
amount of open cups than ITQ-2 (see Scheme 2), a lower
number of 1-PNH•+ radical cations would be generated. Indeed,
while the same optical spectrum is observed for ITQ-2 and
MCM-22, the intensity of the signal is much higher in the
former, in agreement with our hypothesis on the formation of
1-PNH•+ within the open external cups and not within the
sinusoidal 10 MR pores.
Figure 5. EPR spectra of DPB@ITQ-2 and DPB@ZSM-5 recorded
under the same conditions three months after sample preparation. For
the sake of clarity, the signal of DPB@ZSM-5 has been divided by 2.
Given the structure of ITQ-2 layers in which silanol groups
are preferentially located at the rims of the cups (see Figure 1),
silylation should impede the access to these cups. Thus, if the
species responsible for the UV-vis spectrum of Figure 3 would
be located in the main planes of the layers or within the cups,
silylation should have a profound influence on the process since
the access of the cups would be thwarted. On the other hand, if
incorporation of DPB would occur preferentially into the 10
MR of ITQ-2 (similar dimensions as ZSM-5 but sinusoidal)
then silylation of the cups should not play any significant role
in the behavior of the ITQ-2. Clearly, silylation should affect a
process occurring at the external surface, but not those taking
place within the interior of the pores.
Figure 6. Transient decay of TP+ excited state monitored at 470 nm
generated upon 355 nm laser excitation of a N2-purged acetonitrile
solution of TP+BF4- (10-4 M): without 1-PNH (b) and after addition
of 0.027 M (O) and 0.24 M (4) of a solution of PNH in acetonitrile.
spectrum of the sample DPB@ITQ-2, laser flash photolysis
generation of 1-PNH•+ radical cation was undertaken. The
procedure followed consists of the electron transfer quenching
of TP+ 3 formed by 355 nm laser excitation by 1-PNH. Control
experiments reveal that 1-PNH does not give rise to any transient
using 355 nm laser at the concentrations used in the TP+
quenching. Thus, upon laser excitation TP+ 3 is selectively
formed. This TP+ triplet is quenched by 1-PNH (quenching rate
constant 1.3 106 M-1 s-1) (Figure 6), and at sufficiently high
concentration of 1-PNH, the transient spectrum of the TP+ triplet
is replaced by that of the 1-PNH•+ cation radical (Figure 3 inset).
The transient spectrum of 1-PNH•+ cation radical obtained
by photoinduced electron transfer is very similar to those
previously reported in the literature for related 1-vinyl-
substituted naphthalenes.31 Importantly, although the resolution
of the laser flash photolysis setup is lower than that of
conventional UV-vis spectrophotometers, this transient spec-
trum is compatible with the diffuse reflectance UV-vis of
DPB@ITQ-2, thus supporting the assignment of the species
generated as 1-PNH•+. Although it is a well-known fact, it is
remarkable that PNH•+ lifetime in solution is on the order of
µs while inside the cyclodextrin-like cups of ITQ-2 it lives for
months!
The degree of silylation was followed by monitoring in the
IR the peak corresponding to the silanol groups. This absorption
band experiences a significant decrease (about 60% of the initial
band, indicating that 7 silanols out of 12 per cup have been
silanized) after treatment with haxamethyldisilazane. To confirm
that the presence of trimethylsilyl groups effectively impedes
the access to the cups but not to the micropores, IR spectra of
pyridine and 2,6-di-tert-butylpyridine adsorbed on a sample of
ITQ-2 before and after silylation were recorded.
The comparison indicates that while the interaction of 2,6-
di-tert-butylpyridine (too large to enter the 10 MR micropores
and proving exclusively the acid sites on the external surface)14
with the acid sites disappears, the amount of pyridinium formed
on ITQ-2 before and after silylation only decreases in a minor
percentage. This result indicates that pyridine can probe the
channels of ITQ-2 equally well before and after silylation.
The diffuse reflectance UV-vis spectrum of a sample of
silylated ITQ-2 after incorporation of DBP was totally different
than that shown in Figure 3 and corresponds to unaltered DPB
A point of the highest importance in the present report is
how to justify the different behavior of ITQ-2 (generation of

Inorganic Mimic of Cyclodextrins
J. Phys. Chem. B, Vol. 107, No. 5, 2003 1141
its geometry due to a well structured external surface formed
by open cups. This effect, which is clearly observed for DPB,
is an example showing that the potential energy well ac-
companying the cups of ITQ-2 is able to generate and stabilize
organic radical cations, even though the species is not deep into
a channel or full pore, in a manner similar to that of the
conventional medium-pore zeolite. Thus, as consequence of its
special topology, the reactivity of ITQ-2 is specific and different
from those of conventional zeolites.
Acknowledgment. Financial support by the Spanish CICYT
(H.G., MAT 2000, 1768-C02-01) is gratefully acknowledged.
B.F. thanks the Spanish Ministry of Education for a scholarship.
We thank Dr. Carlos Go´mez-Garc´ıa (Chemistry Department,
Universitat de Valencia) for recording the EPR spectra and Dr.
Germa´n Sastre (ITQ, Universidad Polite´cnica de Valencia) for
molecular modeling.
Figure 7. Diffuse reflectance UV-vis spectrum (plotted as the
Kubelka-Munk function) of DPB@ITQ-2sil.
References and Notes
(1) Liu, X.; Iu, K.-K.; Thomas, J. K.; He, H.; Klinowski, J. J. Am.
Chem. Soc. 1994, 116, 11811-11818.
(2) Ramamurthy, V.; Caspar, J. V.; Corbin, D. R. J. Am. Chem. Soc.
1991, 113, 594-600.
(3) Ramamurthy, V. Chimia 1992, 46, 359-376.
(4) Ramamurthy, V.; Eaton, D. F.; Caspar, J. V. Acc. Chem. Res. 1992,
25, 299-306.
(5) Cano, M. L.; Forne´s, V.; Garc´ıa, H.; Miranda, M. A.; Pe´rez-Prieto,
J. J. Chem. Soc., Chem. Commun. 1995, 2477-2478.
(6) Pitchumani, K.; Joy, A.; Prevost, N.; Ramamurthy, V. Chem.
Commun. 1997, 127-128.
(7) Martens, J.; Jacobs, P. A. AdV. Funct. Mater. 2001, 11, 337-338.
(8) Davis, M. E. Acc. Chem. Res. 1993, 26, 111-115.
(9) Progress in Zeolite and Microporous Materials; Chon, H.; Ihm,
S.-K.; Sun Uh, Y., Eds.; Elsevier: Amsterdam, 1997; Vol. 105.
(10) Corma, A.; Navarro, M. T.; Rey, F.; Valencia, S. Chem. Commun.
2001, 16, 1486-1487.
(11) Corma, A. Chem. ReV. 1997, 97, 2373-2419.
(12) Kevan, L. J. Chem. Soc., Faraday Trans. 1998, 94.
(13) Konovalova, T. A.; Gao, Y.; Schad, R.; Kispert, L. D. J. Phys.
Chem B 2001, 105, 7459-7464.
(14) Corma, A.; Forne´s, V.; Pergher, S. B.; Maesen, T. L. M.; Buglass,
J. G. Nature 1998, 396, 353-356.
(15) Sastre, G.; Catlow, C. R. A.; Chica, A.; Corma, A. J. Phys. Chem.
B 2000, 104, 416-422.
(16) Pitchumani, K.; Joy, A.; Prevost, N.; Ramamurthy, V. J. Chem.
Soc., Chem. Commun. 1997, 127-128.
(17) Pitchumani, K.; Lakshminarasimhan, P. H.; Turner, G.; Bakker,
M. G.; Ramamurthy, V. Tetrahedron Lett. 1997, 38, 371-374.
(18) Miranda, M. A.; Garc´ıa, H. Chem. ReV. 1994, 94, 1063.
(19) Ramamurthy, V.; Corbin, D. R.; Kumar, C. V.; Turro, N. J.
Tetrahedron Lett. 1990, 31, 47-50.
Figure 8. Molecular modeling showing the best docking of 1-PNH
inside the cups of ITQ-2.
(Figure 7). This result firmly establishes that the ability of ITQ-2
to form the intermediate is totally suppressed by silylation. This
fact together with the behavior of MCM-22 zeolite indicate that
generation of 1-PNH•+ occurs exclusively on the external surface
of the particle.
(20) Wakamatsu, K.; Takahashi, Y.; Kibuchi, K.; Miyashi, T. Tetrahe-
dron Lett. 1994, 35, 5681-5684.
(21) Wakamatsu, K.; Takahashi, Y.; Kikuchi, K.; Miyashi, T. J. Chem.
Soc., Perkin Trans. 2 1996, 2, 2105-2109.
(22) Hug, S. J.; Yee, W. A.; Kliger, D. S. Chem. Phys. Lett. 1990, 168,
385-390.
Taking into account all the evidences reported, we propose
that DPBH+ is initially formed in the open cups of ITQ-2 as in
conventional zeolites through an acid catalysis mechanism.1 But
the larger conformational mobility and particularly the possibility
to form cis isomers permits its cyclization toward the dihy-
dronaphthalene skeleton.17 It is this cis-trans isomerization and
the subsequent intramolecular cyclization that is precluded in
the channels of ZSM-5 and mordenite due to spatial restrictions.
Once 1-PNH is formed, generation of its radical cation and its
stabilization would occur because the shape of this molecule
fits into the cups, thus undergoing the same protection as other
radical cations. Figure 8 provides a molecular modeling showing
the best docking of 1-PNH inside the cups of ITQ-2.
(23) Dudev, T.; Kamisuki, T.; Akamatsu, N.; Hirose, C. J. Phys. Chem.
1991, 95, 4999-5002.
(24) Oberle´, J.; Abraham, E.; Ivanov, A.; Jonusauskas, G.; Rullie`re, C.
J. Phys. Chem. 1996, 100, 10179-10186.
(25) Wang, Z.; McGimpsey, W. G. J. Phys. Chem. 1993, 97, 3324-3327.
(26) Ramamurthy, V.; Caspar, J. V.; Corbin, D. R.; Schlyer, B. D.; Maki,
A. H. J. Phys. Chem. 1990, 94, 3391-3393.
(27) Gessner, F.; Scaiano, J. C. J. Photochem. Photobiol. A: Chem.
1992, 67, 91-100.
(28) Bae, J. Y.; Ranjit, K. T.; Luan, Z.; Krishna, R. M.; Kevan, L. J.
Phys. Chem. B 2000, 104, 9661.
(29) Krishna, R. M.; Prakash, A. M.; Kevan, L. J. Phys. Chem. B 2000,
104, 1796.
(30) Sung-Suh, H. M.; Luan, Z.; Kevan, L. J. Phys. Chem. B 1997,
101, 10455.
(31) Chikai, Y.; Yamamoto, Y.; Hayashi, K. Bull. Chem. Soc. Jpn. 1988,
61, 2281-2285.
In summary, by using R,ω-diphenyl polyenes as probes we
have found that novel delaminated ITQ-2 zeolites exhibit a
unique behavior, shaping the radical cation that best fits within
(32) Argauer, R. J.; Landlolt, G. R. U. S. Patent 3,702,886, 1982.
(33) Mohtat, N.; Cozens, F. L.; Scaiano, J. C. J. Phys. Chem. B 1998,
102, 7557-7562.