1,3,5-Triaryl-2-penten-1,5-dione anchored to insoluble supports as heterogeneous chromogenic chemosensor

Article abstract of DOI:10.1016/j.tet.2004.06.090

N-Allyl substituted 1,5-diphenyl-3-(4-N-methylaminophenyl)-2-penten-1,5- dione (1a) has been immobilized on a polystyrene backbone, on the surface of mercaptopropyl-functionalized silica or inside the cavities of zeolite NaY. These solids either in suspen

Full text of DOI:10.1016/j.tet.2004.06.090

Tetrahedron 60 (2004) 8257–8263
1
,3,5-Triaryl-2-penten-1,5-dione anchored to insoluble supports
as heterogeneous chromogenic chemosensor
*
*
Mercedes Alvaro, Carmela Aprile, Avelino Corma, Vicente Fornes, Hermenegildo Garcia
and Encarna Peris
Departamento de Quimica, Instituto de Tecnologia Quimica CSIC-UPV, Universidad Politecnica de Valencia, 46022 Valencia, Spain
Received 5 February 2004; revised 4 June 2004; accepted 16 June 2004
Available online 21 July 2004
Abstract—N-Allyl substituted 1,5-diphenyl-3-(4-N-methylaminophenyl)-2-penten-1,5-dione (1a) has been immobilized on a polystyrene
backbone, on the surface of mercaptopropyl-functionalized silica or inside the cavities of zeolite NaY. These solids either in suspension or in
3
films act as chemosensors of Fe and other strong Lewis acid metal ions such as Cu and Pb in buffered water or ethanol. Br o¨ nsted
C
2C
2C
3C
acids in low pH aqueous solutions also produce the response of the sensor. For sensing of Fe , depending on the loading of 1a (typically
K2
K4
from 2.5 to 0.5 wt%) the solids can test from 10 to 10 M aqueous solutions. The time response can vary from tens of minutes to below a
minute depending on hydrophilic/hydrophobic nature of the support and also on the 1a loading. The solid sensor was reused up to 10 times by
regenerating after every use the initial form with NaAcO treatment.
q 2004 Elsevier Ltd. All rights reserved.
1
. Introduction
the affinity between the sensor and the analyte and leading
to a reduced sensitivity of the anchored chemosensor. Also
the accessibility of the sensor to the analyte may be reduced
when the molecule is immobilized on the internal pores of a
porous particle. All these factors may lead to a decrease on
the sensitivity, increasing the detection threshold and the
response time. These changes are negative in the use of the
solids as sensors. On the other hand, reversibility and
reusability is another issue of considerable importance for
solids sensors that are never considered in solution.
A strategy to convert a soluble chemosensor into a
heterogeneous sensing system could be anchoring cova-
lently an appropriate derivative of the sensor molecule onto
1
,2
an insoluble support. Analogous approach is commonly
used in heterogeneous catalysis to immobilize active
3
–5
homogeneous catalysts.
Immobilization of the sensor
allows to design systems for automatic continuous sensing.
For this purpose organic polymers or inorganic oxides have
6
,7
been frequently used. In contrasts, no much attention has
been paid to the use of inorganic oxides as supports in spite
that they offer the advantage of a large surface area and the
possibility to have structured particles with micro- or meso-
porosity, while organic polymers have an easy processa-
Recently there has been a large interest in the use of
2-penten-1,5-dione derivatives of the type 1 as colorimetric
sensors for a series of cations, anions and diacids.
1
0
2
10
Given the insolubility of these molecules in water, most of
the analytic tests have been performed in organic solvents
such as CH Cl , or water/organic solvent mixtures what
3
,5,8,9
bility.
2
2
may constitute a limitation of the homogeneous phase
sensing methodology.
When anchoring a sensor molecule onto a solid support
some new factors, different from homogeneous sensing that
require to be properly addressed, may arise influencing the
performance of the sensor in solution. Most of differences
between solution and surface-anchored sensor may derive
from the interaction of the sensor with the solid surface.
This interaction could play an unfavourable role reducing
Keywords: Chemosensors; Pyrylium dye; Covalent functionalization.
*
Corresponding authors. Tel.: C34963877807; fax: C34963877809
H.G.).; e-mail: hgarcia@qim.upv.es
(
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.06.090

8258
M. Alvaro et al. / Tetrahedron 60 (2004) 8257–8263
Given the generality of these pentendiones 1 as sensors and
in order to develop a sensitive device for continuous
operation, it would be of interest to support this molecule
type on a series of solid supports and determine the ability of
the resulting solids to respond as insoluble chemosensors.
Herein we describe our results on the covalent anchoring of
pentendione derivative, 1a, on three insoluble supports.
We will show our results towards the development of
convenient solid chemosensors based on cyclizable penten-
dione, as well as limitations in response time, sensitivity and
reusability and some ways to circumvent or minimize these
general problems encountered in heterogeneous sensors.
To immobilize covalently the sensor to the solids we
followed a flexible strategy based on the use of a common
pentendione derivative containing an allylic functionality
attached to the N-atom (1a). This compound was obtained
by reacting N-methyl-N-allylaniline with 2,6-diphenyl-
pyrylium ion followed by the opening of the heterocyclic
ring. Scheme 2 indicates the reaction sequence followed to
prepare the N-allyl functionalized pentendione 1a.
The anchoring of this intermediate on a polystyrene
backbone was simply accomplished through AIBN-initiated
radical co-polymerization of 1a and styrene (Scheme 3).
The ratio styrene/1a was purposely high, so the polymer
contains approximately between 0.5 and 2 wt% of chemo-
sensor depending on the 1a content used in the polymeri-
zation. The resulting 1a-PS polymer is soluble in hot
toluene, but insoluble in water and alcohols.
2
. Results and discussion
2
.1. Synthesis of the solid chemosensors
Among the solid supports selected, we tried to cover organic
and inorganic supports. One of these supports is a
polystyrene backbone to which some pentendione mol-
ecules 1a were introduced during the polymerization step.
The advantage of this polymer is that while it is insoluble in
water, it can be dissolved in hot toluene so films of this
polystyrene-bound sensor (1a-PS) can be cast on appro-
priate substrates. The other two supports are inorganic
oxides namely silica and zeolite Y. In the case of the silica
the pentendione 1a can be bonded to the external surface of
Scheme 3. Co-polymerization of compound 1a and styrene to form 1a-PS.
In the case of covalent anchoring of the allyl derivative 1a
on silica we followed a route consisting on the prior
modification of the silica by introducing terminal mercapto
groups by silylation of the silanol groups. The silica was
modified by reaction with 3-mercaptopropyltrimethoxy-
silane in toluene. Scheme 4 shows the procedure used to
anchor the chemosensor 1a onto the functionalized silica.
the amorphous particles (1a-SiO ). In contrast NaY zeolite
2
has a crystalline structure containing almost spherical
˚
cavities of 13 A of diameter interconnected tetrahedrally
˚
through 12 oxygen ring windows of 7.4 A in which the
pentendione 1a can be accommodated and immobilized
inside the solid (1a@NaY). Scheme 1 shows the structure of
the zeolite used in this work.
Subsequently, modified silica support containing terminal
mercapto group (SiO -SH) was reacted with the N-allyl-N-
2
methyl derivative of pentendione 1a, which undergoes a
radical-chain thiol addition using AIBN as radical initiator
as shown in Scheme 4.
For the solid sensor in which pentendione 1a is mechani-
cally immobilized (as opposed to covalent anchoring) inside
the cavities of porous NaY zeolite (1a@NaY), the N-allyl
group does not play any role but derivative 1a was also used
for sake of synthetic economy. The major factor contri-
buting to support the chemosensor within NaY is the high
adsorption capacity of zeolites to include an organic guest in
the micropore volume (Scheme 5).
Scheme 1. Pictorial representation of the three supports on which
compound 1a has been immobilized.
Scheme 2. Synthetic route for the preparation of 1a.

M. Alvaro et al. / Tetrahedron 60 (2004) 8257–8263
8259
spectroscopy to characterize spectrophotometrically the
visual changes occurring during the sensing, UV/Vis
spectroscopy served also to assess the absence or presence
of some pyrylium ion 2a formed spontaneously during
manipulation of 1a in the anchoring procedure according to
Eq. (1). This is the reverse of the heterocycle opening shown
in Scheme 1.
ð1Þ
In UV spectroscopy the presence of the heterocyclic
pyrylium ion can easily be revealed. This characterization
2
Scheme 4. Preparation procedure of sensor 1a-SiO .
Figure 1. Transmission optical spectrum of pentendione 1a (spectrum a)
K4
and pyrylium ion 2a (spectrum b) recorded for 10 M solution in
dichloromethane.
Scheme 5. Adsorption of compound 1a into the cages of zeolite NaY to
form 1a@NaY.
Encapsulation of 1,3,5-triphenyl-2-penten-1,5-dione, analo-
gous to the substrate 1a used here, has been successfully
achieved by Miranda, Braun et al. through a novel
1
1
encapsulation strategy. They reported that 2,4,6-tri-
phenylpyrylium ion in water was completely adsorbed in
zeolite NaY up to a loading of 25% in weight through the
open 1,3,5-triphenyl-2-pentendione. In our case the penten-
dione 1a was simply adsorbed into dehydrated NaY zeolite
by stirring a suspension of the solid and the organic
compound in dichloromethane at reflux temperature.
2
.2. Characterization of the solid sensors
All the solids have common spectroscopic features arising
from the presence of pentendione 1a. Optical spectroscopy
is particularly relevant since the colorimetric sensing relies
on changes in the visible optical spectrum. Depending on
whether the sensor can be dissolved (1a-PS) or is an
Figure 2. Diffuse reflectance UV/Vis spectra (plotted as the Kubelka–
Munk function of the reflectance, F(R)) of 1a covalently bonded to silica
insoluble opaque powder (1a-SiO or 1a@NaY), trans-
2
mission or diffuse reflectance modes were used to record the
optical spectrum. In addition of the interest in optical
(
spectrum a) and polystyrene (spectrum b) or absorbed on NaY
(spectrum c).

8260
M. Alvaro et al. / Tetrahedron 60 (2004) 8257–8263
of 2a is based on the presence, or absence, of a specific peak,
at lmaxZ560 nm, that is responsible for the intense red color
of the closed heterocyclic cation 2a. This band is absent in
the pentendione 1a, that presents a yellow color and has
characteristic absorption band at 410 nm. Figure 1 shows
the UV/Vis spectra of the pentendione 1a and pyrylium 2a
ion recorded in dichloromethane solution, while Figure 2
shows some spectra recorded for the solids containing
pentendione 1a.
with the electron deficient pyrylium ring. In the case of the
solid sensors the predominant absorption bands are
obviously due to the polymeric or to the inorganic supports,
but sufficient expansion of the carbonyl and aromatic region
allows to detect weak broad bands that are attributable to the
low amounts of the pentendione 1a anchored to the support.
2
.3. Use of solid sensors
As aspected in view of the behaviour of pentendione 1a
reported in solution, all the solids change the color from
Figure 2 establishes the absence of the band at 560 nm,
indicating that the immobilization procedure has not
produced the cyclization of the pentendione 1a. This is
also in agreement with the visual yellow color of all the
supports after the immobilization procedure. The presence
or absence of this band due to pyrylium ion 2a will become
more relevant later when discussing the chemosensing
properties of these solids and the possibility of successful
regeneration upon reuse. In some cases, for inorganic
supports SiO2 and NaY, we observed an extensive
cyclization during the anchoring or the adsorption pro-
cedure as evidenced by the red color of the substrate and by
the appearance of the pyrylium band in the visible region.
According to Eq. (1) we believe that pyrylium 2a arises
from the acid catalysed cyclization of 1a during the
adsorption. To avoid or minimize the presence of acid
sites on the inorganic supports, these solids were neutralized
by contacting them with basic solutions.
1,2
K2
yellow to red upon contacting with 10 M aqueous
solutions of Fe(NO ) , Cu(NO ) or Pb(NO ) . In contrast,
3
3
3 2
2C
aqueous solutions of Al , Mg and Ca or anions (Br ,
3 2
3C
2C
K
2K
HPO4 ) did not give a positive test at this concentration. In
accordance to Eq. (1), it is expected that the heterogeneous
sensor systems based on the interconversion of 1a to 2a
would respond to any Lewis acid cations with sufficient
strength to provoke the cyclization of the pentendione
through coordination with the N atom in the para position of
the 4-aryl ring. In principle this cyclization can also be
carried out by Bronsted acids and in fact aqueous solutions
of HNO (pHZ1) are also able to produce the response of
3
the solid sensors. For this reason, most of the experiments in
aqueous media were carried out in buffered solution at pHZ
5 using HEPES. At this pH a blank control showed that
cyclization of 1a to 2a does not occur for reasonable long
periods of time.
IR spectroscopy is also a very convenient technique to
distinguish between the pentendione 1a and the pyrylium
Besides experiments in suspension performed stirring a
solution of Lewis acid cations and the solid sensor, an
alternative way of performing the tests was to place the
solids as thin films on an inert substrate. For this set-up we
covered a glass slide with a thin layer of each of the three
water-insoluble solid sensors and assays were carried out by
simply dropping a few microliters of the solutions of
different cations on the sensor-covered glass.
ion 2a. While the former shows three bands at 1676, and
1
K1
642 cm corresponding to the vibration of the carbonyl
groups and the C]C double bond present in the dione, the
most remarkable spectral feature of the pyrylium ion 2a is
the absence of these bands and the presence of a very intense
K1
peak at 1588 and 1573 cm
stretching mode of the C]O bond present in the pyrylium
corresponding to the
C
ring. Figure 3 shows the IR of the two compounds.
3C
Upon addition of aqueous solution of Fe (pHZ5) the
change in the color was progressive in time. In the case of
silica and zeolite Y at the lowest loading of 1a employed
(5 mg per g of support) the red coloration was evident in a
few minutes and became complete after 30 min. According
Figure 3. Carbonyl and aromatic region of the FT-IR spectra of compounds
1a (a) and 2a (b) recorded at room temperature on silicon wafers.
1
2
Previously we have reported that the parent 2,4,6-
C
triphenylpyrylium ion has an intense C]O band at
1620 cm
for the N,N-dialkyl derivative 2a can be attributed to the
electron donor effect of the nitrogen atom being conjugated
Figure 4. Photograph of a glass plate in which films of 1a-PS (upper row)
and 1a@NaY (lower row) have been placed as squares. The image was
K1
K1
and the large shift (about 30 cm ) observed
K2
recorded 1 h after dropping 50 ml of distilled water (a) or 10 M aqueous
3
solution of: Fe (b); Mg (c); Al (d); Br (e); HPO4 (f).
C
2C
3C
K
K

M. Alvaro et al. / Tetrahedron 60 (2004) 8257–8263
8261
to the more hydrophobic nature of polystyrene backbone,
when pentadione 1a was covalently anchored on poly-
styrene the response of 1a-PS was even significantly slower,
the red color developing over 1 h and becoming complete in
The response of the glass coated sensors at 0.5 wt% loading
K3
of 1a was also visually observable for 10 M concen-
tration of Fe , while for more diluted solutions not
relevant response was observed.
3
C
5
–6 h. Figure 4 shows a photograph of the sensing
experiments in which the differences between positive or
negative responses can be seen for some analytes.
As commented above, the detection threshold can be
lowered, the visual response reaching 10 M when the
K4
concentration of 1a is increased to 25 mg per gram of
support. This loading also gives much faster response than
lower contents of 1a. However, high contents of 1a give
problems in terms of reusability.
In fact one of the major problems of the solid supports is the
contact time required to observe the response. Obviously, in
contrast to the solids sensors, for solution experiments the
change in the color is almost instantaneous. As commented
above this contact time depends on the hydrophilicity/
hydrophobicity of the support. In this regard, the response
time of 1a-PS is dramatically reduced from hours up to less
than 1 min when Fe(NO ) is dissolved in ethanol instead of
2.4. Reuse of the solid sensors
In fact when working on solid chemosensors, one point of
critical importance is the possibility of reuse by regener-
ation of the initial sensor form after the testing. In the case
considered here and according to Scheme 2, it should be
possible to reopen the heterocyclic pyrylium ion 2a to
3
3
water. Notably for these ethanolic solutions the response of
a@NaY requires longer time than for 1a-PS. We also
1
observed that the response time becomes shorter when
solids sensors with more than 0.5 wt% contents of 1a are
used. Also the intensity of the color increase with loading of
pentendione 1a by treatment with an aqueous base solution.
As expected, upon addition of some drops of 10
K1
M
1
a. Thus, by controlling the loading of 1a on the solid one
NaOAc onto the film of sensor the red color gradually
turned into the yellow and the optical spectrum changed
from the one corresponding to pyrylium to that of the
pentendione. Figure 6 shows the spectra corresponding to
can on one hand shorten the response time and on the other
to decrease the detection threshold of the film. The problem
of using solids with high 1a loading is the reusability as it
will be commented above.
K2
K1
three consecutive 10 M Fe(NO ) /10
3
NaOAc cycles
3
using 1a@NaY as heterogeneous sensor.
It was of interest to determine by optical spectroscopy
whether or not the change in the color is accompanied by a
partial chemical conversion between the pentendione and
the pyrylium form as indicated in Eq. (1) in agreement with
the chemosensing operation of the sensing system. We took
the absorption bands at 440 and 550 nm as specific of the
open form 1a and heterocyclic form 2a of the sensor system,
respectively. We found that the red color, characteristic of
the visual positive test as shown in Figure 4, corresponds to
a true chemical change from 1a to 2a. Figure 5 shows a UV/
K2
Vis spectrum 30 min after addition of 10 M solution of
Fe(NO ) in which the band at 550 nm is observed. This
3
3
support the chemical interconversion between the open and
the cyclic form.
Figure 6. Diffuse reflectance UV/Vis spectra of the same sample of
a@NaY (loading 0.5 wt%) submitted, respectively, to the following
1
treatments: (a) 10 M Fe , 30 min; (b) 10 M NaAcO, 30 min; (c) 10
K2
3C
K1
K
2
3C K1
M Fe , 30 min; (d) 10 M NaAcO, 30 min.
As it can be seen in this figure, UV/Vis spectroscopy reveals
that the 1a@NaY system undergoes a certain fatigue and the
reversibility of the conversion between the open and the
closed form is not complete after a few cycles as evidenced
by the incomplete disappearance of the 560 nm band
characteristic of 2a. The higher the loading of 1a, the lesser
the reversibility of the system. It is worth commenting that
1a@NaY is in fact the less reversible of the three insoluble
solid sensors films since it is expected that Fe
C
Figure 5. Diffuse reflectance UV/Vis spectrum of a 1a@NaY solid 30 min
K2
3
C
will
gradually exchange the Na of the zeolite upon extensive
3 3
after contacting with 1 ml of a 10 M aqueous solution of Fe(NO ) . The
visual appearance of the solid is red. For the initial diffuse reflectance
spectrum of the sample see Figure 1.
reuse. In this regard 1a-SiO or 1a-PS are more reusable
2

8262
M. Alvaro et al. / Tetrahedron 60 (2004) 8257–8263
mixture was stirred for 1 h at 0 8C under N atmosphere and
2
allyl bromide (604.9 mg, 5 mmol) was added. The suspen-
sion was warmed at room temperature and magnetically
stirred for 24 h. Then, the reaction was washed with a 1 M
solution of HCl and extracted with CH Cl . The organic
2
2
layer was washed with NaOH, dried with Na SO and
2
4
evaporated under vacuum. The compound (35%) was
obtained as yellow oil after chromatography using a mixture
1
of hexane/diethyl ether (10/1) as eluent. H NMR (CDCl )
3
d: 2.98 (s, 3H, CH ), 3.94 (d, JZ4.8 Hz, 2H, CH N), 5.14–
3
2
5
7
1
.22 (m, 2H, CHCH ), 5.77–5.90 (m, 1H, CHCH ), 7.22–
2 2
1
.28 (5H, ArH). C NMR (CDCl ) d: 36.6, 55.9, 113.0,
3
3
16.7, 117.0, 129.7, 134.4, 150.1.
Figure 7. Photograph of two dichloromethane solutions (3 ml) of 1a-PS
50 mg in the first run) that has to be submitted to 10 consecutive test-
K3
(
recovery cycles. Left: upon addition of 500 ml of an 2.5!10 M ethanolic
4.1.2. Synthesis of compound 2a. N-Allyl-N-methylaniline
(160.3 mg, 1.1 mmol) was dissolved in DMF (3 ml) and the
2,6-diphenylpyrylium perclorate (724.4 mg, 2.2 mmol),
K3
solution of Mg(OAc)
3 3
solution of Fe(NO ) .
2
. Right: after addition of 50 ml of an 2.5!10
M
1
4,15
synthetized as described in the literature,
was added.
systems. For these two solids at 0.5 wt% loading the
recycling of the sensor and the regeneration was performed
at least 10 times without remarkable differences in the visual
behaviour of the chemosensor. Figure 7 shows a photograph
The mixture was stirred at reflux temperature for 3 h, then
cooled at room temperature and stirred for 20 h. The solvent
was removed under reduced pressure and the resulting
brownish, red oil used for the next step without purification.
K3
of a negative 2.5!10 M ethanolic solution of Mg(OAc)2
K3
and a positive 2.5!10
test using 1a-PS after 10 consecutive reuses regenerating
M ethanolic solution of Fe(NO3)
3
4.1.3. Synthesis of compound 1a. Sodium acetate
(183.3 mg, 2.23 mmol) was dissolved in a mixture of
water (0.6 ml), methanol (1.9 ml) and acetone (3.8 ml). The
K2
every time the pentendione form by treatment with 10
NaOAc aqueous solution.
crude compound 2a was added, the mixture was stirred for
1
5 h at room temperature and for 10 h without stirring.
3
. Conclusions
The solvent was evaporated under reduced pressure and the
residue was purified by flash chromatography using diethyl
ether as eluent yielding pentendione 1a (55%) as yellow oil.
Covalently binding or adsorption of pentendione 1a to
several inorganic and organic supports is a viable strategy to
transform this compound soluble in organic solvent into an
insoluble solid sensor for Fe and other strong Lewis acids
and cations. In aqueous solution, the recoverable sensors
also respond to Br o¨ nsted acids.
1
H NMR (CDCl ) d: 3.01 (s, 3H, CH ), 3.92 (d, JZ4.5 Hz,
3
3
3
C
2
CHCH ), 5.63–5.96 (m, 1H, CHCH ), 7.14–7.98 (m, 14H,
H, CH N), 4.91 (s, 2H, CH COPh), 5.08–5.24 (m, 2H,
2 2
2
2
1
ArH). C NMR (CDCl ) d: 36.6, 55.9, 112.4, 116.8, 126.7,
3
3
1
1
27.0, 128.1, 128.6, 128.9, 129.2, 129.4, 129.6, 133.5,
35.7. HPLC-MS (electrospray) 396.2 (MCH ). Combus-
C
The three major problems are the increase in the response
time, the lesser sensitivity and the reusability of the solids.
The time of response and sensitivity are clearly modulated
by the hydrophilicity of the support and the loading of the
dye. Solid sensors containing 2.5 wt% of compound 1a may
tion chemical analysis: Exp (%): C 80.40, H 6.50, N 3.09;
calculated for C H NO (%): C 82.00, H 6.37, N 3.54.
2
7
25
2
4
.2. Preparation of the solid sensors
show the response time below minutes and can detect below
1
K3
0
M concentration. Reusability relies on the reversi-
Compound 1a-PS. Styrene (1 g) was dissolved in toluene
15 ml), a solution of 1a (10 mg) in toluene (5 ml) and
bility of the heterocyclic ring opening under basic
conditions to regenerate the open pentendione form. The
solid support plays a role in the system by hydrophobic/
hydrophilic interaction with the solvent and also by
promoting undesirable spontaneous cyclization of penten-
dione 1a. Silica and zeolite are more adequate for sensing in
water, while polystyrene gives shorter response time in
ethanol. We are expanding this strategy to the development
of other chromogenic heterogeneous sensors.
(
AIBN in catalytic amount were sequentially added. The
mixture was stirred at reflux temperature under N2
atmosphere for 3 h, then cooled at room temperature. The
solvent was removed under reduced pressure yielding 1a-PS
as yellow solid.
Compound 1a@NaY. The zeolite NaY (Aldrich, 1 g) was
dehydrated by calcination at 500 8C for 6 h, then cooled
under vacuum and suspended in CH Cl (15 ml). A solution
2
2
of 1a (10 mg) in CH Cl (5 ml) was slowly added and the
2
2
4
. Experimental
.1. Synthesis of organic compounds
.1.1. Synthesis of N-allyl-N-methylaniline.. N-Methyl-
mixture was stirred at 40 8C for 3 h. The solid 1a@NaY was
obtained as yellow powder after filtration and extensive
washings with CH Cl .
4
4
2
2
1
3
Compound 1a-SiO . Silica (BASF, 4 g) was dried at 300 8C
2
aniline (535.5 mg, 5 mmol) was dissolved in dry THF
(
under vacuum for 6 h and suspended in dry toluene (40 ml).
3-Mercaptopropyl trimethoxysilane (4 ml) was slowly
t
10 ml) and BuOK (561.4 mg, 5 mmol) was added. The

M. Alvaro et al. / Tetrahedron 60 (2004) 8257–8263
8263
added and the mixture was stirred for 48 h at 110 8C. The
resultant solid was filtered, dried and washed with CH Cl
thanks the Fundation Jose Royo for a postdoctoral
scholarship.
2
2
in a Soxhlet apparatus to remove the unreacted silane. The
modified SiO -SH (1 g) was then suspended in toluene
2
(
15 ml) and NaHCO (200 mg) was added to neutralize the
3
acidity. The suspension was stirred for 20 min then a
solution of 1a (10 mg) in toluene (5 ml) and AIBN, in
catalytic amount were sequentially added. The reaction was
stirred at reflux temperature for 24 h, then filtered and
washed with CH Cl $yielding 1a-SiO as yellow solid.
References and notes
1
. Prasanna de Silva, A.; Nimal gunaratne, H. Q.; Gunnlaugsson,
T.; Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice,
T. E. Chem. Rev. 1997, 97, 1515–1566.
2
2
2
2
3
. Martinez-Manez, R.; Sancenon, F. Chem. Rev. 2003, 103,
4
4
.3. General sensing experiments.
419–4476.
. De Vos, D. E.; Dams, M.; Sels, B. F.; Jacobs, P. A. Chem. Rev.
2002, 102, 3615–3640.
Suspension test. The solid sensors 1a-SiO or 1a@NaY
2
(
100 mg) were suspended in an aqueous solution at pH 6.7
buffered with HEPES). Some drops of the nitrate or sodium
4
5
6
. Corma, A.; Garcia, H. Chem. Rev. 2003, 103, 4307–4365.
. Wight, A. P.; Davis, M. E. Chem. Rev. 2002, 102, 3589–3613.
. McCleskey, S. C.; Griffin, M. J.; Schneider, S. E.; McDevitt,
J. T.; Anslyn, E. V. J. Am. Chem. Soc. 2003, 125, 1114–1115.
. McCleskey, S. C.; Schneider, S. E.; Anslyn, E. V.; McDevitt,
J. T. Abstracts of Papers, 222nd ACS National Meeting,
Chicago, IL, United States, August 26–30, 2001 2001. ORGN-
(
salts of the ions (from 10
K2
K4
to 10 M in water) Fe
3C
,
Al , Mg , Br , HPO were then added to each sus-
3
C
2C
K
K
4
pension and then the mixture was magnetically stirred.
7
Films. 1a-PS 1a-SiO and 1a@NaY was dissolved in hot
2
toluene (1 ml) and some drops of this solution were cast on a
glass previously covered with scotch tape to define squares
5
56.
. Corma, A. Chem. Rev. 1997, 97, 2373–2419.
8
2
1 cm ). Alternatively, 1a-SiO or 1a@NaY were dispersed
(
in acetylacetone/water (1:10) and the paste cast onto the
2
9. Corma, A.; Galletero, M. S.; Garcia, H.; Palomares, E.; Rey, F.
Chem. Commun. 2002, 1100–1101.
glass slide. After drying the films, few ml of the nitrate or
sodium salts (from 10 to 10 M in water or ethanol) of
1
0. Martinez-Manez, R.; Soto, J.; Lloris, J. M.; Pardo, T. Trends
Inorg. Chem. 1998, 5, 183–203.
11. Amat, A. M.; Arques, A.; Bossmann, S. H.; Braun, A. M.; Gob,
K2
K4
3
the ions Fe , Al , Mg , Br , HPO were then dropped
C
3C
2C
K
K
4
over the film.
S.; Miranda, M. A. Angew. Chem., Int. Ed. 2003, 42, 1653.
2. Corma, A.; Forn e´ s, V.; Garc ´ı a, H.; Miranda, M. A.; Primo, J.;
1
Sabater, M. J. J. Am. Chem. Soc. 1994, 116, 2276.
13. Kimura, M.; Futamata, M.; Shibata, K.; Tamaru, Y. Chem.
Commun. 2003, 234–235.
Acknowledgements
Financial support by the Spanish Ministry of Science and
Technology (MAT 2003-1226) and Generalidad Valenciana
14. Stetter, H.; Reischl, A. Chem. Ber. 1960, 93, 1253–1256.
15. Van Allan, J. A.; Chang, S. C. J. Heterocycl. Chem. 1974, 11,
1065–1070.
(
grupos03-020) is gratefully acknowledged. Carmela Aprile