'Ship-in-a-bottle' synthesis and photochromism of spiropyrans encapsulated within zeolite Y supercages

Article abstract of DOI:10.1016/S0040-4020(00)00515-9

The ship-in-a-bottle synthesis of three spiropyrans inside the NaY supercages was carried out by the condensation of 2-methylene-1,3,3-trimethylindoline with the appropriate substituted benzaldehyde. After the reaction, spiropyrans were detected in the supernatant and the zeolites became coloured. Combustion chemical and thermogravimetric analyses confirmed the presence of organic material within the solid. UV-Vis and Raman spectroscopies are compatible with the formation of the merocyanine form of the spiropyran photochromic system. Upon visible irradiation the solids bleach, and in some cases the coloration is regained upon standing in the dark for long periods. This abnormal photochromic behaviour is the reverse of that observed for the spiropyran/merocyanine system in ethanol and illustrates the possibility of zeolites as media to alter the molecular properties of incorporated guests. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00515-9

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 6951±6956
`Ship-in-a-Bottle' Synthesis and Photochromism of Spiropyrans
Encapsulated within Zeolite Y Supercages
a
Isabel Casades, Steven Constantine, David Cardin, Hermenegildo Garcõa, Andrew Gilbert
b
b
a,
b
*
Â
a
Â
and Francisco Marquez
a
Â
Â
Â
Instituto de Tecnologõa Quõmica CSIC-UPV, Universidad Politecnica de Valencia, Apartado 22012, 46071 Valencia, Spain
bDepartment of Chemistry, University of Reading, Whiteknights, P.O. Box 224, Reading RG6 6AD, UK
Received 26 December 1999; revised 26 February 2000; accepted 28 February 2000
AbstractÐThe ship-in-a-bottle synthesis of three spiropyrans inside the NaY supercages was carried out by the condensation of
2-methylene-1,3,3-trimethylindoline with the appropriate substituted benzaldehyde. After the reaction, spiropyrans were detected in the
supernatant and the zeolites became coloured. Combustion chemical and thermogravimetric analyses con®rmed the presence of organic
material within the solid. UV±Vis and Raman spectroscopies are compatible with the formation of the merocyanine form of the spiropyran
photochromic system. Upon visible irradiation the solids bleach, and in some cases the coloration is regained upon standing in the dark for
long periods. This abnormal photochromic behaviour is the reverse of that observed for the spiropyran/merocyanine system in ethanol and
illustrates the possibility of zeolites as media to alter the molecular properties of incorporated guests. q 2000 Elsevier Science Ltd. All rights
reserved.
Introduction
SP±H (RˆH) does not exhibit photochromism upon
irradiation in ethanol since the corresponding MC±H (lack-
ing a 6-NO2 group) is much less stable and reverts more
quickly to the colourless SP±H. Therefore, upon irradiation
there is not a suf®ciently high build-up of MC±H to surpass
a threshold concentration. The in¯uence of hydrogen bond-
ing and complexation with transition metals in the SP$MC
interconversion, though not new, has attracted some interest
in recent years.4±6 It is known that minor variations on the
substituents and their location as well as on the medium has
a profound effect on the photochromic properties of the SP/
MC system.
Spiropyrans (SP) are the best studied photochromic
compounds.1±3 Their photochromism is due to the different
UV±Vis spectra of the colourless, closed SP form and the
intensely coloured, open merocyanine (MC) form and the
fact that their interconversion is triggered by UV light
(Scheme 1). Given the very different molecular properties
of the spiro SP and the zwitterionic MC forms in terms of
polarity and their tendency to form hydrogen bonds, the
thermodynamic equilibrium between SP and MC and
their interconversion rate is highly dependent upon the
nature of any substituents (R) and the properties of the
medium. Thus, an adequate combination that leads to
convenient photochromic properties in solution requires a
nitro group at the 6 position of the chromene ring, with
ethanol as the solvent. It is worth noting that the parent
Incorporation of organic species inside the channels and
cavities of microporous zeolites has proved to be a general
methodology to gain control over both their molecular and
photochemical properties.7 The special properties of the
Scheme 1. Photochromic equilibrium between colourless SP and coloured MC.
Keywords: photochromism; spiropyrans; zeolite.
* Corresponding author. Tel.: 134-96-387-7807; fax: 134-96-387-7809; e-mail: hgarcia@qim.upv.es
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00515-9

6952
I. Casades et al. / Tetrahedron 56 (2000) 6951±6956
obtained from preformed commercial spiropyrans and
polymerising alkyl- or phenyl-triethoxysilane has been
studied.14,15 Herein, we have also found that the polarity
of the reaction cavity provided by the zeolite host favours
the coloured MC as the predominant form, but a certain
reversible photochromism can still be observed, although
the SP$MC system operates in the reverse manner as
that occurring in EtOH.
Results and Discussion
Due to the frozen orthogonal con®guration of the indolic
and chromenic moieties, SP is a fairly rigid molecule lack-
ing in conformational freedom, whilst MC is considerably
more ¯exible due to the permitted rotation through the
simple C±C bond connecting the indolic and phenolate
substructures (Scheme 1). However since pure MC cannot
be obtained in solution, its incorporation inside Y zeolites is
not feasible. On the other hand, molecular modelling at the
semiempirical level predicts that even though SP can be
accommodated inside the spherical supercavities of the Y
zeolite (,1.4 nm diameter), its molecular dimensions are
too large to allow its incorporation from an organic solution
towards the interior of the pores since SP cannot cross the
smaller cage windows (0.74 nm diameter). Fig. 1 shows a
visualisation of the best docking of 6-NO2 substituted SP
(SP±NO2) within the supercages of zeolite Y. As it can be
seen there, the nitro group is partially penetrating through
one of the cavity windows, indicating that this molecule
occupies a little more space than that available in a single
cavity. From this molecular model it can be concluded that
in order to prepare SP-incorporating zeolite Y supercages, it
is necessary to proceed using `ship-in-a-bottle' syntheses,
wherein smaller precursors that are able to diffuse freely
through the pores, react inside the cage and lead, ultimately,
to the formation of the target SP. We adopted such a
methodology, stoichiometrically condensing 2-methylene-
1,3,3-trimethylindoline with the appropriately substituted
salicylaldehyde in CH2Cl2 under re¯ux conditions (Scheme
2). This is a general reaction that is commonly used to
prepare SP in the absence of zeolites. Table 1 contains a
Figure 1. Molecular modeling based on semiempirical calculations show-
ing the best docking of SP±NO2 inside the zeolite Y supercages. No
signi®cant van der Waals unfavorable overlap of guest±host atomic radii
is predicted.
rigid internal voids of the zeolites in terms of polarity and
the presence of charge-balancing cations, makes these
crystalline aluminosilicates suitable as solid matrices to
stabilise positively charged as well as zwitterionic species,
increasing markedly their lifetimes compared to those
measured in common organic solvents.8,9
In the present work, we report on the preparation and
characterisation of three SPs inside the supercages of Y
zeolite. For comparison, MCM-41 and amorphous silica±
alumina has also been included in this study. In a related
precedent, SP was adsorbed in the interlamellar region of
layered clays and it was observed that the SP$MC equi-
librium is irreversibly shifted towards the more polar
MC;10±13 as a result, the supramolecular system losses the
photochromic behaviour of SP. In other reports, the photo-
chromism of sulfonated spiropyrans-organic modi®ed
silica nanocomposites (the position of the sulfonic group
and the presence of other substituents not speci®ed)
Scheme 2. `Ship-in-a-bottle' synthesis of SP incorporated within zeolite Y.
Table 1. Main analytical and spectroscopic data of the samples prepared in the present work
Sample
Substituent R
Carbon content (%)
C/N Ratio
lmax(nm)
MC±H/NaY
MC±Cl/NaY
MC± NO2/NaY
SP±H/NaY-aa
MC±H/SiO2±Al2O3
MC± NO2/MCM-41b
H
Cl
NO2
H
H
NO2
1.59
1.13
2.62
.0.2
.0.2
2.8
15.4
14.7
8.69
±
555 (sh 525)
570 (sh 530)
394, 530
±
±
381, 538
±
a Prepared by adsorbing pre-formed SP±H from CH2Cl2 on NaY and subsequently submitting the solid to exhaustive solid±liquid extraction with EtOH.
b
Ê
Prepared by adsorbing pure SP±NO2 on an all-silica MCM-41 (32 A pore size).

I. Casades et al. / Tetrahedron 56 (2000) 6951±6956
6953
combustion/degradation of the adsorbed organic material.
The fact that this second process takes place at high
temperatures is commonly observed when the organic
species is strongly adsorbed within the solid (Fig. 2).
However, in spite of the presence of SP in the supernatant
solution after the synthesis indicated in Scheme 2, the spec-
troscopic data of the zeolite samples do not totally agree
with the presence of SP, rather these spectra are indicative
of the MC form being the predominant species retained in
the solid. Thus, diffuse-re¯ectance UV±Vis spectra of the
solids (Fig. 3, left) each exhibit a band in the visible region
characteristic either of 2-methylene-1,3,3-trimethylindoline
or the MC forms.16 Comparison of the FT-Raman spectra of
pure indoline specimens with those recorded for the zeolite
powders after the synthesis conclusively rules out the
presence of detectable amounts of indoline in the solids.16
It was observed that co-adsorbed water plays a role in the
relative intensities of the absorption bands appearing in the
diffuse-re¯ectance UV±Vis spectra. In this regard, mild
baking of the samples (1108C for 5 h) increases the absor-
bance of the short-wavelength maximum relative to the
long-wavelength band. Comparison of the FT-Raman of
pure SP crystals with that of the zeolites after the ship-in-
a-bottle synthesis (Fig. 3, left) clearly indicates that while all
the peaks corresponding to the SP form are present, the
spectra contain other extra vibration bands. Based on the
formation of SP in the organic phase during the synthesis
and on the diffuse re¯ectance UV±Vis spectra, we attribute
these extra vibrational peaks present in the FT-Raman
spectra to the MC forms encapsulated within NaY in equi-
librium with their spiro forms. In order to support this
assignment, adsorption of pure SP±NO2 on an all-silica
MCM-41 was carried out. It was observed that adsorption
of SP±NO2 also produces the same changes in the diffuse-
re¯ectance UV±Vis (Table 1) and Raman spectra as those
observed when comparing the spectra of pure SP±NO2
crystals with those of the NaY powders after the ship-in-
a-bottle syntheses. We attempted measurements also by
FT-IR spectroscopy, but the much more intense absorption
bands of the Si±O and Al±O vibrations of the zeolite lattice
in the IR compared to Raman rendered these spectra much
less informative. The herein proposed stabilisation of the
MC form over the SP within zeolites agrees with the
reported precedent on the incorporation of spiropyrans in
the interlamellar region of clays.10±13
Figure 2. Thermogravimetric analysis (TG, left axis) and differential scan-
ning calorimetry (DSC, right axis) of MC±NO2/NaY.
list of the samples that have been prepared and a summary
of their main analytical and spectroscopic properties.
GC analyses of the supernatant solutions after the reactions
con®rmed the formation of the corresponding SPs, in addi-
tion to lesser amounts of unreacted starting materials and
indoline derived products 1,3,3-trimethyl-2-indolinone and
1,2,3,3-tetramethylindoline. In view of these results showing
the formation of SP in the CH2Cl2 solutions, it is not
unreasonable to expect an additional fraction of SP also to
have been formed inside the zeolite Y supercages, and which
would remain trapped in the solid after the synthesis. Thus,
combustion chemical analyses (C and N) of the zeolites after
exhaustive extraction with ethanol and complete removal of
the solvent in vacuo revealed the presence of a signi®cant
amount of organic material (Table 1).15 Considering that all
of the organic material corresponds to SP the elemental
analyses give a maximum occupancy estimation of one fourth
of the zeolite cavities, which is a reasonable value when
compared with related `ship-in-a-bottle' syntheses.7
Thermogravimetric±differential scanning calorimetry
studies of the spiropyran containing NaY samples give an
estimation of the adsorbed organic content that also agrees
with the data of chemical analyses. The thermogravimetric
pro®les of the samples show common features. Thus, after
an initial endothermic loss of weight of approximately
20 wt% at temperatures up to 2008C, associated with the
desorption of co-adsorbed water, there is a second loss of
weight between 300 and 7008C (around 2 wt%) correspond-
ing to an exothermic process, and attributed to the partial
Figure 3. Left: Diffuse-re¯ectance UV±Vis spectra plotted as the inverse of the re¯ectance (1/R) of the NaY samples after spiropyran `ship-in-a-bottle'
syntheses and exhaustive solid±liquid extraction. Right: FT-Raman spectrum of MC±NO2/NaY (a) compared to that of pure nitrospiropyran crystals (b)
showing that spectrum (a) contains all the peaks of spectrum (b) plus some extra bands attributable to the MC±NO2 form.

6954
I. Casades et al. / Tetrahedron 56 (2000) 6951±6956
experiment, pure SP was adsorbed onto NaY (supposedly
on its external surface) and the resulting material then
submitted to solid±liquid extraction. Again, neither the
diffuse-re¯ectance UV±Vis spectrum corresponded to that
of MC recorded using the ship-in-a-bottle procedure nor
after exhaustive extraction was any organic material left
in the solid (see Table 1, entry SP±H/NaY-a and the corre-
sponding footnote). This result suggests that the adsorption
procedure leaves SP deposited on the less-polar external
surface of the particles from where it can readily be
desorbed. As predicted by molecular modelling, no
penetration of the SP into the pores seems to have occurred.
Figure 4. Percentage nitrogen atoms of the total number of atoms,
measured by XPS at different depths through the zeolite particle for the
MC±NO2/NaY sample. The line shows the N atom percentage calculated
from the elemental analysis.
To provide direct experimental evidence for the distribution
of the organic material within the zeolite, a sample of
SP±NO2 was analysed by X-ray photoelectron spectro-
scopy (XPS) coupled with Ar1-ion sputtering. XPS gives
a chemical analysis of the outermost exposed layers of the
zeolite grains, and by combining this surface analytical
technique with fast-ion bombardment (which results in
partial destruction of the external layers), it is possible to
map out the chemical composition from the external surface
of the zeolite to several nanometers towards the interior of
the particle. Fig. 4 shows the experimental percentage of
nitrogen atoms against the total number of atoms monitored
as a function of the depth from the exterior of the particle.
These data clearly prove that the nitrogen-containing
organic material is spread out through the bulk of the
particle with its abundance increasing with increasing
depth. Furthermore, the organic material in the external
surface has a nitrogen content below the average of the
whole particle obtained from the combustion chemical
analyses data.
One point of concern when dealing with organic species-
zeolite supramolecular systems is to be able to prove that the
organic material is in fact hosted inside the micropores of
the zeolite and not exclusively deposited on the external
surfaces of the particles. Therefore we have performed
both indirect assays and direct measurements.
Thus, the synthetic procedure depicted in Scheme 2 was
carried out in the presence of amorphous silica±alumina
instead of NaY: silica±alumina has a chemical composition
like that of zeolites, but because it is not a microporous
solid, only external surfaces are available for the guest. As
expected, the organic reaction took place equally well in the
presence of SiO2±Al2O3 as for NaY (as assessed by the
presence of SP in the CH2Cl2 supernatant). However, two
notable differences were observed: (i) the predominant form
in this case appears to be SP, based on the diffuse-
re¯ectance UV±Vis spectrum of the solid; and (ii)
essentially all of the organic material is removed by
exhaustive solid±liquid extraction with ethanol (see Table
1, entry MC±H/SiO2±Al2O3). The lower polarity of the
external surface of silica±alumina compared to that of the
zeolite interior would account for the shift in the relative
stability between the SP and MC forms, whilst since all of
the organic material is recoverable, the implication is that
the structure of the zeolite plays a role in retaining a signi®-
cant fraction of the organic material. Likewise, in another
All the coloured MC±NaY powders undergo a signi®cant
bleaching upon visible light irradiation, the behaviour being
the reverse of that observed for the corresponding SPs in
ethanol. Thus, irradiation (l.450 nm) produces a progres-
sive decolouration of the solid and the course of the
photochemical transformation is conveniently followed by
monitoring the decrease in the intensity of the visible
band (550 nm) of the samples. Once the intensity of the
visible band reaches the minimum, dark storage of the
solid produces over long time periods a remarkable,
though incomplete recovery of the original intensity. The
Figure 5. Left: Variation of the inverse of the re¯ectance (R) measured at 555 nm as a function of time during consecutive irradiation-dark storage cycles for
the MC±H/NaY sample. The arrows indicate when the irradiation is on and off and the time elapsed in each cycle. Right: Diffuse re¯ectance spectra of the
same sample at 10 min interval during irradiation.

I. Casades et al. / Tetrahedron 56 (2000) 6951±6956
6955
percentage of the intensity recovery and the required time is
extremely dependent upon the nature of the substituent R
present in the chromenic moiety. Attempts to observe an
increase in the rate of the SP$MC interconversion by
illuminating with UV light were unsuccessful. A precedent
in solution of reverse photochromic behaviour of a series of
6-nitrospiropyrans containing two additional electron-with-
drawing substituents has been reported.17
This cycling [(i) irradiation (l.450 nm) in the visible with
bleaching of the visible band, and (ii) recovery of the visible
band in the dark] was repeated several times, with a certain
fatigue of the system, in terms of a less profound decolour-
ation, being apparent after the second cycle (Fig. 5). It is
noteworthy that the photochromic response of the MC±H/
SP±H pair (which is not observed in ethanol) is the most
convenient when the compounds are incorporated within
NaY. Thus, it is remarkable that under identical experimen-
tal conditions the irradiation time necessary to observe a
response for the MC±NO2/SP±NO2 pair is much longer.
As in solution, this fact probably re¯ects the in¯uence of the
nitro group stabilising the MC form relative to the SP form.
Given the polarity of the zeolite cages MC±NO2 becomes
too stable to readily cyclise with high quantum yield to the
SP form. The photochromic response of MC±H±NaY
during some consecutive cycles is presented in Fig. 5.
Figure 7. Emission (EM, lexc 520 nm) and excitation (EX, monitoring at
610 nm) spectra of MC±NO2/NaY solid.
excitation spectra of the three MC±R/NaY samples. Thus,
the excitation spectrum of the MC forms coincided with its
UV±Vis absorption spectra and is the mirror image of the
emission (Fig. 7).
In conclusion, the reaction of 2-methylene-1,3,3-trimethyl-
indoline with substituted salicylaldehydes in the presence of
dehydrated NaY leads to the formation of the spiropyrans
(SPs) in the organic phase and the encapsulation inside the
zeolite cages of some organic material that corresponds, at
least in part, to the merocyanine (MC) form of these
spiropyrans. The polarity of the zeolite pores seems to be
responsible for the change in the relative stability between
the SP and the MC forms that favours the latter. These MC-
containing zeolitic materials exhibit a certain photochromic
behaviour that is the reverse of that found for such organic
species (SP/MC) when in EtOH solution, i.e. bleaching of
the 550 nm band upon irradiation and the recovery of this
visible band upon standing in the dark. Although further
work is obviously necessary to optimise the ship-in-a-bottle
synthesis, maximising the loading and improving the photo-
chromic response of these materials probably by using more
suited zeolites, the present work shows the opportunities
that zeolites offer as solid hosts to incorporate light switch-
able molecules to prepare advanced materials with adequate
optical properties.
Since the photochemical transformation of MC into SP and
the subsequent thermal reversion back to MC upon standing
on the dark takes place within minutes or hours, it is also
possible to follow the course of these processes with
conventional FT-Raman spectroscopy. In agreement with
the variations in the diffuse-re¯ectance study, the FT-
Raman spectra of the samples revealed a decrease in some
vibrational bands of the original spectrum and the appear-
ance of new absorptions, some of them compatible with the
partial formation of SP. As an example, Fig. 6 shows the
case of MC±NO2/NaY, wherein the peaks increasing in
intensity upon irradiation have been labelled with asterisks.
Besides characterisation by FT-Raman spectroscopy, the
fact that MC is the stable form of the spiropyran system
allows us to determine other photochemical properties. In
particular, we have been able to record the emission and
Experimental
Combustion chemical analyses were carried out using a
Fisons EA-1108 analyser. Thermogravimetry±differential
scanning calorimetry studies were conducted using a
Netzsch STA 409 thermobalance under air stream and
kaolin as inert standard. Diffuse-re¯ectance UV±Vis
spectra were recorded using a Cary 5G spectrophotometer
adapted with a `praying mantis' accessory for solid samples,
using BaSO4 as the standard. Molecular modelling was
carried out at the semiempirical level using the Insight II
package, running on a Silicon Graphics workstation. FT-IR
spectra were recorded at ambient temperature using a
Nicolet FT-710 spectrophotometer using sealed cells ®tted
with CaF2 windows. The zeolite wafers (10 mg) were
compressed at 1 Ton cm22 and outgassed under 1022 Pa at
1008C for 1 h. FT-Raman spectra were recorded using a
Bio-Rad FT-Raman II spectrophotometer, with the
Figure 6. FT-Raman spectra of the MC±NO2/NaY sample after irradiation
with l.450 nm. The original spectra corresponds to that shown in Fig. 3,
left for the same sample and the most characteristic bands that have grown
upon irradiation are marked with asterisk.

6956
I. Casades et al. / Tetrahedron 56 (2000) 6951±6956
1064 mm line of a Nd:YAG laser for excitation and a liquid
nitrogen-cooled Ge detector. The laser power at the sample
was approximately 100 mW and the spectral resolution
4 cm21. The number of scans varied from 400 to 1000.
The Raman spectra were corrected for instrumental
response using a white light reference spectrum. XPS
measurements were carried out at ambient temperature
with a concentric hemispherical analyser operated in the
constant pass energy mode (50 eV). A Mg Ka X-ray source
(hnˆ1253.6 eV) was used. The analysis chamber was main-
tained at ca. 5£1029 Torr during acquisition of the XPS
spectra. Charging effects were calibrated by the C(1s) line
at 284.6 eV. Sputtering of the samples to obtain the atomic
pro®le of the samples was carried out inside the chamber
using an Ar1 ion gun. The response of each atom was
corrected by the corresponding response factor.
Acknowledgements
Financial support of the European Union (Brite BE97-4455)
and the Spanish DGICYT (grant MAT97-1060-CO2) is
gratefully acknowledged as it is a scholarship for I. C.
Â
from the Universidad Politecnica Valencia.
References
1. Bertelson, R. C.; Brown, G. H. Photochromism, Wiley: New
York, 1971.
2. Durr, H.; Bouas-Laurent, T. H. Photochromism: Molecules and
Systems, Elsevier: Amsterdam, 1990.
3. Maeda, K. J. Synth. Org. Chem. Jpn 1991, 49, 554±565.
4. Winkler, J. D.; Bowwen, C. M.; Michelet, V. J. Am. Chem. Soc.
1998, 120, 3237±3242.
Zeolite NaY (P.Q. Industries), silica±alumina (BASF,
25 wt%),2-methylene-1,3,3-trimethylindoline and the three
p-substituted benzaldehydes (Aldrich) were commercial
samples and used as received. All-silica MCM-41 was
prepared as reported.18 Pure specimens of the three spiro-
pyrans were obtained by double recrystallisation of the
solids, prepared by the stoichiometric condensation of
2-methylene-1,3,3-trimethylindoline (200 mg) and the
corresponding benzaldehyde in ethanol (20 ml) at room
temperature.
5. Inouye, M.; Akamatsu, K.; Nakazumi, H. J. Am. Chem. Soc.
1997, 119, 9160±9165.
6. Suzuki, T.; Lin, F. -T.; Priyadashy, S.; Weber, S. G. Chem.
Commun. 1998, 22685±22686.
Â
7. Scaiano, J. C.; Garcõa, H. Acc. Chem. Res. 1999, 32, 783±793.
Â
8. Corma, A.; Garcõa, H. Top. Catal. 1998, 6, 127±140.
Â
9. Cano, M. L.; Cozens, F. L.; Esteves, M. A.; Marquez, F.;
Â
Garcõa, H. J. Org. Chem. 1997, 62, 7121±7127.
10. Tomioka, H.; Itoh, T. J. Chem. Soc., Chem. Commun. 1991,
532±533.
The `ship-in-a-bottle' syntheses of the NaY-incorporated
SPs was carried out by heating a stirred suspension of
dehydrated (5008C, overnight) NaY (1 g) zeolite in an equi-
molar mixture of indoline (30 mg) and the corresponding
benzaldehyde in CH2Cl2 (25 ml) under re¯ux conditions for
seven days. At the end of the reaction, the solid was ®ltered
and the supernatant CH2Cl2 analysed by GC (Hewlett±
Packard chromatograph provided with a 25 m capillary
column of crosslinked 5% phenylmethylsilicone) and
GC±MS (Varian using the same type of capillary column
as for the GC). Characterisation of the corresponding SP in
the supernatant was carried out by comparison of the
retention time and MS spectra with authentic specimens.
The coloured zeolite was then submitted to exhaustive
solid±liquid extraction with a micro-Soxhlet apparatus
with EtOH as the solvent. Irradiation of the solid powders
was carried out using an Oriel photoreactor adapted with a
420±630 nm dichroic ®lter (Oriel part#66219). Emission
spectroscopy was recorder in a Edinburg FL-900 spectro-
¯uorimeter using a front face attachment for solids. The
samples were N2 purged.
11. Takagi, K.; Kurematsu, T.; Sawaki, Y. J. Chem. Soc., Perkin
Trans 2 1995, 1667±1671.
12. Seki, T.; Ichimura, K. J. Chem. Soc., Chem. Commun. 1987,
1187±1189.
13. Hori, T.; Tagaya, H.; Nagaoka, T.; Kadokawa, J.; Chiba, K.
Appl. Surf. Sci. 1997, 121, 530±533.
14. Tagaya, H.; Nagaoka, T.; Kuwahara, T.; Karasu, M.;
Kadokawa, J.; Chiba, K. Microp. Mesop. Mater. 1998, 21, 395±
402.
15. In the cases of salicylaldehyde and 4-chlorosalicyladehyde,
nitrogen analyses selectively traces indoline derived products
including spiropyrans adsorbed in the solid.
16. Indoline is an intensely coloured compound with a band in the
visible region, at 550 nm. This lmax is very close to that of MCs.
The FT-Raman sample was prepared by the adsorption of the
indoline onto NaY in EtOH solution.
17. Barachevskii, V. A.; Strokach, Y. P.; Valova, T. M.; Arsenov,
V. D.; Gorelik, A. M. Zh. Nauchn. Prikl. Fotogr. 1997, 42, 14±21;
cf. Chem. Abstr. 1997, 128, 173979.
 Â
18. Corma, A.; Fornes, V.; Garcõa, H.; Miranda, M. A.; Primo, J.;
Sabater, M. J. J. Am. Chem. Soc. 1994, 116, 9767±9768.