PH-responsive bridged silsesquioxane

Article abstract of DOI:10.1021/cm103327y

A pH-responsive bridged silsesquioxane was obtained via the sol-gel process from a bis-silylated triazine derivative bearing a triple DAD (donor-acceptor-donor) moiety which was strongly hydrogen-bonded to the three ADA (acceptor-donor-acceptor) sites of

Full text of DOI:10.1021/cm103327y

ARTICLE
pubs.acs.org/cm
pH-Responsive Bridged Silsesquioxane
Laurent Fertier, Christophe Thꢀeron, Carole Carcel,* Philippe Trens, and Michel Wong Chi Man*
ICG Montpellier (UMR 5253) CNRS-UMII-ENSCM-UMI, 8 rue de l’ꢀecole normale, 34296 Montpellier CEDEX 5, France
S Supporting Information
b
ABSTRACT: A pH-responsive bridged silsesquioxane was ob-
tained via the solꢀgel process from a bis-silylated triazine
derivative bearing a triple DAD (donorꢀacceptorꢀdonor) moi-
ety which was strongly hydrogen-bonded to the three ADA
(acceptorꢀdonorꢀacceptor) sites of cyanuric acid. The latter
template molecule was easily removed from the resulting hybrid
material upon acidic treatment, which yielded the template-free
bridged silsesquioxane with preserved morphology and unaltered
particle size as demonstrated by SEM. As expected, the surface
area of the acid-treated hybrid was significantly higher than that of
the untreated material, as demonstrated by nitrogen adsorption
measurements. The absence of cyanuric acid in this material was
confirmed by solid state NMR and by IR. It was shown that a mild acidic medium (pH = 5.5) is sufficient to release the cyanuric acid
template, thus providing a novel route to new hybrid materials with potential applications as efficient carriers for pH-responsive
delivery systems.
KEYWORDS: bridged silsesquioxanes, molecular recognition, pH-responsive hybrid silica, imprinting
’ INTRODUCTION
afford mesoporous silica.20 The pore sizes and distribution were
dependent on both the structure of the sacrificial ethynyl templates
and the reaction conditions employed.21
In recent years, significant effort has been focused toward the
synthesis of porous materials with potential applications such as
for drug-delivery carriers1ꢀ3 and for catalysis.4 These materials
consist of either organic or inorganic polymers,5,6 among which
silica meets most of the requirements as a good support: chemical
inertness, thermal and mechanical stability, and also the possibi-
lity of tailoring the size and the shape of the pores and cavities
inside the materials.7ꢀ9 Mesoporous materials were obtained using
surfactants as structuring agents and from this surfactant-me-
diated route, organic groups may be covalently anchored in two
ways: (1) grafting a functional alkoxysilane on a preformed me-
soporous silica (MCM- or SBA-type); and (2) co-condensation
of the functional alkoxysilane with TEOS in the presence of a
suitable surfactant.10ꢀ12 Following these two methods, many
forms of functional mesoporous hybrids have been synthesized
and exploited for many applications.13 For example, mechanized
nanomachines have been prepared for on-demand release of
cargo molecules (including drugs, dyes, biomolecules, etc.)14 as
well as easily recoverable and recyclable homogeneous catalysts
supported on hybrid porous silicates which have applications in a
variety of chemical processes.15 Hybrid mesoporous silica can
also be prepared via the self-assembly of organopolysilane pre-
cursors bearing a long alkylene chain. In this case, self-structuring
occurs to form a purely hybrid silica without added TEOS and
without any surfactant.16,17 Complementary to these are the per-
iodic mesoporous organosilicates (PMO) obtained from bridged
silsesquioxane precursors, which are prepared with surfactants.18,19
In the series of bridged silsesquioxanes, labile SiꢀC covalently
bonded ethynyl units were softly cleaved by a chemical route to
Porous materials can also be obtained by molecular imprinting
techniques which are used to create solid materials containing
chemical functionalities that are spatially localized and fixed by
interactions with the imprint molecules during the synthesis process.
Subsequent removal of these template molecules leaves preformed
sites for the recognition of small molecules, hence affording
materials which may be suitable for applications in separation
chemistry.22 Indeed, protein-imprinted silica prepared via cova-
lent imino group bonding showed enhanced affinity for the tem-
plate protein compared to the corresponding imprinted silica
obtained by entrapment of the protein with no interaction of the
latter with the silica surface.23 Using a chiral cationic surfactant in
conjunction with solꢀgel processing, chiral imprinted silica-
based thin films were obtained, which exhibited good selectivity
toward chiral alcohols.24 Imprinted polysilsesquioxanes have also
been obtained by the co-condensation of a bridged organosilane
(BTSE) precursor with N-[3-(trimethoxysilyl)propyl]ethylene-
diamine (TMS(en)) in the presence of several cations (CuII, ZnII,
and NiII). The removal of these cations created cavities within
the hybrids which exhibited selective rebinding characteristics.25
More recently we described imprinted hybrid silicas (IHS), in
which the covalently bonded molecular recognition sites in the
inorganic matrix were templated with the complementary imprint
Received: November 19, 2010
Revised:
March 16, 2011
Published: March 31, 2011
r
2011 American Chemical Society
2100
dx.doi.org/10.1021/cm103327y Chem. Mater. 2011, 23, 2100–2106
|

Chemistry of Materials
ARTICLE
using a Bruker FT-AM 400 spectrometer using cross-polarization and
magic-angle spinning techniques (CP-MAS). The high-resolution mass
spectra (Q-TOF ESþ) were measured on a JEOL MS-DX 300 mass
spectrometer. Discontinuous nitrogen sorption isotherms were mea-
sured at 77.15 K using a Micromeritics 2010 apparatus. Specific surface
areas were derived using the BET transform of the sorption isotherms,
taking 0.162 nm2 as the cross section area for nitrogen.31 No micro-
porosity could be detected using the t-plot method, taking Aerosil200 as
reference.32 TGA analyses were performed on a TA Instruments ATG
Q50 from 25° to 500 °C, with a ramp rate of 10 °C/min.
Synthesis. Compound 1 (2-N,N-bis(2-CBz-Aminoethyl)amino-4,6-
dichloro-1,3,5-triazine). 2,4,6-Trichloro-1,3,5-triazine (0.478 g, 2.59 mmol)
was dissolved in THF (20 mL) and the resulting solution cooled in a water/
ice bath before adding diisopropylethylamine (0.68 mL, 3.88 mmol). Com-
pound A (0.978 g, 2.59 mmol), dissolved in THF (20 mL), was then added
dropwise under magnetic stirring which was continued for an additional 2 h
after addition at 0 °C. After warming to room temperature, THF was re-
moved under reduced pressure and the residue was partitioned between
CH2Cl2 and 1 N HCl. The organic layer was washed with H2O and dried
over MgSO4. The solvent was removed under vacuum and the resulting solid
was purified by column chromatography (SiO2; CH2Cl2/MeOH 10/0.25)
to afford compound 1 as a white solid in 83% yield. MP: 118ꢀ119 °C. 1H
NMR (CDCl3, 400 MHz): 3.42 (t, 4H); 3.69 (t, 4H); 5.06 (s, 4H); 7.33 (m,
10H). 13C NMR (CDCl3, 400 MHz): 38.99, 48.64, 66.85,
128.20ꢀ128.34ꢀ136.25, 156.56, 165.62, 170. HRMS (ESIþ): calcd for
C23H25N6O4Cl2 (M þ H), 519.1314; found, 519.1306.
Figure 1. Imprinted hybrid silica preparation.
molecule.26 The removal of the latter organic imprint is expected
to leave specific receptor sites with tightly controlled dimensions
and shapes under mild processing conditions (Figure 1). The
formation of a complex involving multiple hydrogen bonds be-
tween the initially confined template and the matrix enables this
system to be controlled by an external stimulus, namely the pH,
since an acidic environment causes the rupture of the hydrogen
bonds, and associated release of the template molecule. This was
demonstrated with a monosilylated precursor, (triethoxysilyl)-
propylcyanurate having two ADA recognition faces which can
easily complex melamine.26 A hybrid silica material was thus ob-
tained by co-condensation of TEOS with the silyl derivative
hydrogen-bonded to pure melamine as template to imprint the
material. In this case, the melamine was completely removed under
strong acidic treatment (pH = 2). It was shown that melamine
could be subsequently reintroduced quantitatively, after pretreat-
ing the template-free hybrid with triethylamine, although the re-
sulting materials exhibited some textural modifications. Bridged
silsesquioxanes27,28 are more suitable than the hybrid silicas ob-
tained by the co-condensation method, as the exclusion of TEOS
enables more organic species to be tethered to the rigid frame-
work, as well as a higher loading of the template molecule in the
resulting hybrid silica.
Compound 2 (2-N,N-bis(2-CBz-Aminoethyl)amino-4,6-bis(N-pro-
pylamino)-1,3,5-triazine). Propylamine (1.52 mL, 18.45 mmol) and com-
pound 1 (0.96 g, 1.85 mmol) were refluxed in THF (30 mL) overnight.
After concentration under reduced pressure, the resulting oil was dissolved
in CH2Cl2. The organic layer was washed with H2O and dried over MgSO4.
The solvent was removed under vacuum to give quantitatively compound 2
as a colorless oil. 1H NMR (CDCl3, 400 MHz): 0.87 (t, 6H); 1.49 (m, 4H);
3.26 (t, 4H); 3.39 (t, 4H); 3.66 (t, 4H); 4.74 (s, 2H); 5.06 (s, 4H); 6.31 (s,
2H); 7.31 (m, 10H). 13C NMR (CDCl3, 400 MHz): 11.0, 22.5, 40.4, 42.0,
46.1, 65.9, 127.5, 128, 136.4, 156.3, 165.1. HRMS (ESIþ): calcd for
C29H41N8O4 (M þ H), 565.3251; found, 565.3255.
Compound 3 (2-N,N-bis(2-Aminoethyl)amino-4,6-bis(N-propyla-
mino)-1,3,5-triazine). To a solution of compound 2 (0.919 g, 1.63 mmol)
in EtOH (40 mL) was added cyclohexene (4 mL) and Pd/C (0.23 g, 1/4
equiv per weight). The mixture was refluxed for 30 min and filtered through
Celite. EtOH was removed under reduced pressure and the resulting oil was
purified by column chromatography (SiO2; CH2Cl2/MeOH 8/2 and then
CH2Cl2/MeOH/NH4OH 8/2/0.33). Compound 3was obtained as an off-
white solid in 81% yield. MP: 69ꢀ71 °C. 1H NMR (CDCl3, 400 MHz):
0.89 (t, 6H); 1.50 (m, 4H); 2.88 (t, 4H); 3.24 (t, 4H); 3.56 (t, 4H); 4.90 (s,
2H). 13C NMR (CDCl3, 400 MHz): 11.4, 23.0, 40.5, 42.4, 50.0, 165.9.
HRMS (ESIþ): calcd for C13H29N8 (M þ H), 297.2515; found, 297.2513.
Compound 4. In a Schlenk tube, compound 3 (1.35 g, 4.55 mmol),
anhydrous CH2Cl2 (30 mL), triethylamine (1.4 mL, 10.01 mmol) and
3-isocyanatopropyltriethoxysilane (2.5 mL, 10.01 mmol) were stirred
overnight under nitrogen. After removal of CH2Cl2 under reduced
pressure, the resulting white solid was washed with dry pentane (3 times,
50 mL). Compound 4 was obtained in 70% yield as a white solid. MP:
The objective of this work is to use molecular imprinting29,30
via weak H-bonds to create bridged silsesquioxanes containing
chemical functionalities that can release the imprint (or template)
molecules under mild acidic activation conditions. Herein, we re-
port the synthesis of a pH-responsive bridged silsesquioxane ob-
tained from a bis-silylated triazine derivative, bearing only one
DAD (donorꢀacceptorꢀdonor) face and two hydrolyzable trialk-
oxysilylgroups, associated with cyanuric acid through H-bonding.
Removal of the cyanuric acid under strong acidic condition was
studied and the possibility of using this bridged silsesquioxane as
a potential carrier to release cyanuric acid at 37 °C at moderate
pH was also examined.
’ EXPERIMENTAL SECTION
1
decomposition. H NMR (CDCl3, 400 MHz): 0.58 (t, 4H); 0.91 (t,
General Methods. The synthesis of the silylated molecular pre-
cursor was carried out in an inert atmosphere and by using vacuum line
and standard Schlenk techniques. The reagents were used as received.
The solvents were dried using MB SPS-800 apparatus. The melting points
were determined on B€uchi Melting Point B-540 and are uncorrected.
The FTIR data were obtained on a Perkin-Elmer FT-IR system Spectrum
6H); 1.18 (t, 18H); 1.54 (m, 8H); 3.06 (q, 4H); 3.27(m, 4H); 3.41 (m,
4H); 3.60 (m, 4H); 3.78 (q, 12H). 13C NMR (CDCl3, 400 MHz): 0.9,
7.5, 11.4, 18.2, 23.0, 23.6, 39.8, 42.5, 42.9, 58.3, 158.8, 165.6. HRMS
(ESIþ): calcd for C33H71N10O8Si2 (M þ H), 791.4995; found, 791.5006.
Hybrid M1. Compound 4 (0.63 g, 0.699 mmol) and cyanuric acid (34
mg, 0.263 mmol) were completely dissolved in DMSO (2.654 mL) at
50 °C. After 1 h, water (172 μL, 9.556 mmol) and NH4F (64 mL, 0.016
mmol, 0.25 M solution) were added. Spontaneously a white gel was
formed and was left under static conditions at room temperature for
BX spectrophotometer equipped with GladiaATR. Liquid 1H and 13
C
NMR spectra were recorded on a Bruker AC-400 spectrometer at room
temperature with CDCl3 as the solvent and tetramethylsilane (TMS) as
an internal reference. Solid state 13C and 29Si NMR spectra were obtained
2101
dx.doi.org/10.1021/cm103327y |Chem. Mater. 2011, 23, 2100–2106

Chemistry of Materials
ARTICLE
Scheme 1. Synthesis of Precursor 4
3 days. After successive washing with water (50 mL), EtOH (100 mL),
and acetone (50 mL), a white power was obtained (400 mg). 13C CP
MAS solid state NMR: 11. 9, 18.8, 25.2, 43.9, 58.7, 160.4, 164.5. 29Si CP
MAS solid state NMR: ꢀ66.7, ꢀ58.1, ꢀ46.4.
step. For convenience (easy handling and purification of the non
moisture-sensitive intermediates 1ꢀ3), silylation was chosen as
the last step of the synthesis pathway.
Preparation of the Imprinted Bridged Silsesquioxane.
Three hybrid materials were obtained for our study: M0, M1
and M2 (Figures 2 and 3).
Hybrid M0. The same procedure was applied to compound 4 with-
out the cyanuric acid, yielding M0 as a white powder. In this case, the gel
was formed only after cooling to room temperature. 13C CP MAS solid state
NMR: 12.1, 24.0, 43.2, 160.4, 166.7. 29Si CP MAS solid state NMR: ꢀ
67.0, ꢀ58.4.
To design the imprinted hybrid material M1, the bis-silylated
precursor 4 (one DAD face) was complexed with cyanuric acid
AC (three ADA faces) in hot DMSO (Figure 2). Due to the
sensitivity of the amino and carbonyl groups of the precursors to
protonation and basic conditions, respectively, a nucleophilic
catalyst (fluoride anion) was used to ensure the assembly of the
two organic units via H-bonding in the resulting hybrid material.
The two organic compounds were dissolved in hot DMSO and
hydrolysisꢀcondensation was initiated by adding water and
ammonium fluoride as a catalyst. A gel was formed spontaneously
and was aged under static conditions at room temperature for
3 days to afford, after filtration, washing, and drying, M1 as a
white powder. M2 was obtained from M1 by treating the latter
with boiling acidic ethanol solution (10ꢀ2 M HCl in EtOH) for
24 h to remove cyanuric acid (Figure 2).
For comparison, a blank experiment M0 was carried out under
the same conditions without using cyanuric acid as template
(Figure 3). In this case, gel formation was also observed, but only
after cooling the reaction mixture to ambient temperature.
Scanning Electron Microscopy (SEM) Studies. SEM images
of M0 revealed a spherical morphology with a heterogeneous
distribution of particle sizes (Figure 4A) ranging from 1 to 70 μm. In
addition, nanoparticles (150 and 300 nm) were also observed
(Figure 4B). In contrast, SEM images of M1 (Figure 4C) exhibit
a completely different morphology. Indeed, nanometer-size and
fiber-like structures, entangled together to form a macroporous
network are observed (width 100 nm). This provides clear
evidence that the templating and imprinted cyanuric molecules
significantly influence the shape of the resulting material and
reflects the strong H-bonding interaction between the two
complementary fragments (DADꢀADA, respectively, from tria-
zineꢀcyanuric acid). Interestingly, these shapes are well pre-
served in M2 (Figure 4D), after complete removal of cyanuric
Hybrid M2. Material M1 (816 mg) was suspended in EtOH (164 mL)
and HCl (139 μL, 10ꢀ2 M) and heated under reflux with stirring for
24 h. The material was filtered and washed several times with ethanol,
yielding a white powder. The latter was then treated with aqueous Et3N
(85 mL of water, 8.4 mL of Et3N) and stirred overnight to neutralize the
protonated NꢀH so as to recover the original DAD center of the triazine
fragment. After filtration and washing with EtOH, M2 was obtained as a
white powder.13C CP MAS solid state NMR: 11. 9, 18.8, 23. 8, 43.2, 58.7,
160.6, 166.7. 29Si CP MAS solid state NMR: ꢀ66.3, ꢀ61.7.
’ RESULTS AND DISCUSSION
Synthesis of the bis-Silylated Triazine Precursor 4. Pre-
cursor 4 was synthesized in four steps (Scheme 1) from cyanuric
chloride via stepwise substitution of the three chloro substituents
by either a primary or a secondary amine.33 In the first step, cyanuric
chloride was reacted with 1.0 equiv of A bearing two CBz pro-
tecting groups34 at 0 °C in the presence of N,N-diisopropylethy-
lamine (DIPEA) in THF, which yielded 1 in 83% yield. In a
second step the N,N0,N00,N000-tetraalkyltriazine derivative 2 was
then obtained in quantitative yield by reacting 1 with an excess of
propylamine at 90 °C in THF. Subsequent CBz deprotection was
realized using cyclohexene with Pd/C in ethanol under reflux for
30 min, to give compound 3 in 81% yield. The two deprotected
primary amino groups of compound 3 were then reacted, under
inert atmosphere, with 3-isocyanatopropyltriethoxysilane (2 equiv)
to give the targeted silylated precursor 4 in 70% yield. The CBz
groups thus protect the two primary amines in the first step of the
synthesis, and hence selectively direct substitution at the sec-
ondary amine. Subsequent deprotection of the primary amines
would then allow silylation with an isocyanate in the subsequent
2102
dx.doi.org/10.1021/cm103327y |Chem. Mater. 2011, 23, 2100–2106

Chemistry of Materials
ARTICLE
Figure 2. Preparation of M1 and M2.
Figure 3. Preparation of M0.
acid by strong pH treatment (pH = 2) which demonstrates the
robustness of the material.
Thermal Analyses. The TGA analysis of M0 is consistent with
the decomposition of the material in a single step commencing at
around 250 °C (Figure 5A). At lower temperature, in the range
100ꢀ200 °C, loosely bound water molecules are desorbed.
At 300 °C, the material decomposes in a very narrow tempera-
ture range, consistent with the synthesis route of the material
which was prepared from a simple bis-silylated precursor without
any templating agent or mineralizing species such as TEOS.
From the weight loss observed at 500 °C, it can be deduced that
M0 is composed of ∼75% of organics.
Figure 4. SEM images of (A) and (B) M0, (C) M1 and (D) M2.
extent in both cases (∼ 55%). However, the weight loss is not as
steep in the case of M0 which suggests that cyanuric acid moieties
destabilize the bis-silylated condensed precursor.
Figure 5C illustrates the decomposition of M2. The TGA curve
obtained with this material is very similar to the one found in the
case of M0, which is consistent with the hypothesis that the
cyanuric acid has been removed in M2. More explicitly, it can be
deduced that the acidic conditions employed are efficient enough
to remove the cyanuric acid moieties from the structure.
These experiments are consistent with the textural character-
ization of the materials showing that the synthesis of the cyanuric
acid/bis-silylated precursor has been successfully achieved. More
interestingly, these data also show that despite host/guest interac-
tion strong enough to retain the guest within the hybrid material,
an acidic treatment is adequate to remove the cyanuric moieties
without breaking down the structure, allowing for structured
cavities to be created within the framework of the material.
The TGA curve associated with M1 in which cyanuric acid is
presentis morecomplex, and several weight lossescan be observed
between 100 and 300 °C (Figure 5B). This indicates that the
material is heterogeneous, confirming the presence of cyanuric
acid in its framework. The weight loss at low temperature
(100ꢀ150 °C) exhibits a trend similar to that observed for M0,
corresponding to the loss of water (3.5ꢀ4.5%). However, in the
region 200ꢀ250 °C, there is a clear difference with a significant
weight lossin the case of M1. Thisisattributed to to the removalof
the loosely H-bonded cyanuric acid moieties and is consistent with
the cyanuric acid/3 bis-silylated precursor molar ratio.
The main weight loss corresponding to the decomposition and
the removal of the bis-silylated condensed precursor has the same
2103
dx.doi.org/10.1021/cm103327y |Chem. Mater. 2011, 23, 2100–2106

Chemistry of Materials
ARTICLE
Figure 5. TGA analysis of (A) M0, (B) M1, and (C) M2.
Figure 6. Discontinuous sorption isotherms of nitrogen for M0 to M2.
Textural Properties of the Different Materials. Figure 6
shows the nitrogen adsorption isotherms of the various samples.
Essentially no sorption was evident in the case of M0 (squares)
made from the precipitation of hybrid silica precursors. In terms
of textural properties, this can be attributed to the presence of a
non porous material made of large aggregated and condensed
particles of the bis-silylated precursor (M0). This sorption isotherm
confirms that a template is required to improve the textural
properties of the condensed bis-silylated precursor. If the tem-
plate is present during the preparation of the material, a different
adsorption isotherm is found (triangles, M1). In this case, some
N2 sorption is evident, although the extent of sorption is still
limited indicating that the majority of cavities are filled by the
template and N2 is sorbed mainly on the external surface of the
particles. These can be seen by SEM (Figure 4C). Indeed, it is
important to note that the as-synthesized M1 probably does not
sorb any nitrogen within the internal cavities, which implies that
the structure of the cyanuric acid/bis-silylated precursor is relatively
dense. In contrast, the sorption isotherms for M2 (Figure 6) are
consistent with significant uptake of N2; this, together with the
relatively steep slope in the isotherm at zero N2 coverage and a
Figure 7. 29Si solid NMR spectra of M0ꢀM2.
that this increase in specific surface area is mostly due to the
removal of the template moieties. It can therefore be concluded
that an appropriate treatment in acidic conditions (pH 2) generates
M2 in which cyanuric acid has been removed, as confirmed by
TGA (and NMR and IR studies, as discussed below).
Solid State NMR. The extent of condensation was examined
by 29Si solid state NMR (Figure 7). In the case of M0, an intense
T3 signal is observed indicating a highly condensed material. The
spectrum of M1 exhibits T2 and T3 peaks with similar intensity
in addition to a weaker T1 peak consistent with a less condensed
material. After the acid treatments of M1 to remove cyanuric acid
and generate M2, the intensity of the T3 signal greatly increased,
while the T2 signal decreased significantly and the T1 signal was
extinguished. Thus the acidic treatment increased the extent of
condensation when M1 is transformed to M2. In addition, all the
29Si solid NMR spectra (M0, M1, and M2) showed no Q signal,
confirming that the organic components remained covalently bound
to the silica network.
specific surface area of around 45 m2 gꢀ1 confirm that the
3
The 13C solid state NMR spectra of the three materials confirmed
the presence of the bis-silylated triazine fragment (Figure 8).
Indeed, in all cases, the spectra exhibited a peak at around 165
ppm characteristic of the triazine CArꢀNH, another peak at
around 160 ppm assigned to the CdO carbon of the urea groups,
textural properties of the material have changed. From the SEM
images (Figure 4D), it can be seen that the acid treatment left the
particle size essentially unchanged, which suggests that the in-
crease in specific surface area cannot be related to a decrease in
the size of the particles between M1 and M2. It is therefore clear
2104
dx.doi.org/10.1021/cm103327y |Chem. Mater. 2011, 23, 2100–2106

Chemistry of Materials
ARTICLE
Figure 8. 13C solid NMR spectra of M0ꢀM2.
and three additional peaks associated with the methylene carbons
at 12 (CꢀSi), 24 (CH2), and 44 (CꢀN) ppm. The peak at
160 ppm is also characteristic of the CdO group of the template
molecule (cyanuric acid) which is present only in M1. Closer
inspection in the range of 150ꢀ170 ppm (see inset to Figure 8)
indicates two main features:
• a slight shift of the CAr-NH peak from 167 to 164 ppm is
observed in the spectrum of M1. This upfield shift is related
to the presence of the H-bonding interaction of the anchored
triazine unit with the cyanuric acid (absent in M0 and M2).
• the relative intensities of the peaks at 140 to 190 ppm in the
spectra of M0 (green) and M2 (red) are similar, with the
CArꢀNH at 167 ppm being more intense than the CureadO
in both cases. In contrast, the corresponding peaks exhibit
comparable intensities in M1 (blue) resulting from the
presence of the additional CdO group of the template
cyanuric acid. The similarities of the peaks of M0 and M2
and the downfield shift of the CArꢀNH peak to 167 ppm in
M2 at exactly the same chemical shift found in M0 indicate
the absence of cyanuric acid in both materials.
Figure 9. Solid FTIR spectroscopic analysis of compound 4 and
materials M0ꢀM2.
Solid FTIR Spectroscopic Analyses. The FTIR spectrum of
the precursor 4exhibits two absorption bands at 984 and 1092 cmꢀ1
characteristic of SiꢀOꢀC2H5 vibrations (Figure 9). The inten-
sity of these bands either decreases (1092 cmꢀ1) or completely
disappears (984 cmꢀ1) during hydrolysis, yielding two broad
bands (centered at 1100 and 1020 cmꢀ1) attributable to siloxane
bridges (SiꢀOꢀSi) in the spectra of the hybrid materials (M0,
M1, and M2).35,36 Interestingly, the band at 1100 cmꢀ1 is more
intense for M0 indicating a higher degree of condensation28 in
this material as observed by solid state 29Si NMR. Moreover, a
sharp peak at 810 cmꢀ1 assigned to a vibration of the triazine
ring37 is also observed in every spectrum, which further confirms
the presence of the triazine fragment in the precursor 4 and the
materials M0, M1, and M2.
fragments. This is the case in the FTIR spectrum of M1, which
exhibits a distinct mode at 1720 cmꢀ1 corresponding to the
CdO vibration of the templating cyanuric acid. This band is
absent in the FTIR spectrum of M2 obtained from M1 after
treating the latter with HCl (pH = 2). As in our earlier report,24
such treatment under strongly acidic conditions enables the com-
plete removal of cyanuric acid and suggests that the hydrogen
bonding interactions between the triazine derivative and the
cyanuric acid fragments are suppressed in the material, facilitat-
ing the release of the cyanuric acid. It should be noted that the
treatment of M1 with boiling EtOH at neutral pH is not efficient
enough to break these H-bonding interactions to remove cya-
nuric acid. Indeed, no traces of cyanuric acid could be detected in
the aliquot even after 24 h. We thus decided to perform further
studies at milder pH. Hence treatment of M1 at 37 °C in an
Infrared spectroscopy is an appropriate tool to inspect the
presence/absence of the template molecule provided that at least
one of its vibrational modes is distinct from those of the anchored
2105
dx.doi.org/10.1021/cm103327y |Chem. Mater. 2011, 23, 2100–2106

Chemistry of Materials
ARTICLE
acetate buffer solution (pH 5.5) was studied. Interestingly, after
24 h stirring, no vibration band is observed at 1720 cmꢀ1 in the
corresponding infrared spectrum which confirms that this mod-
erate pH treatment is efficient enough to promote the release of
the cyanuric acid.
(8) Huo, Q.; Margolese, D. I.; Ciesla, U.; Feng, P.; Gier, T. E.; Sieger,
P.; Leon, R.; Petroff, P. M.; Sch€uth, F.; Stucky, G. D. Nature 1994,
368, 317–321.
(9) Bagshaw, S. A.; Prouzet, E.; Pinnavaia, T. J. Science 1995, 269
1242–1244.
(10) Burkett, S. L.; Sims, S. D.; Mann, S. Chem. Commun. 1996
1367–1368.
’ CONCLUSION
(11) Macquarrie, D. J. Chem. Commun. 1996, 1961.
(12) Lim, M. H.; Blanford, C. F.; Stein, A. J. Am. Chem. Soc. 1997,
119, 4090.
(13) Sanchez, C.; Julian, B.; Belleville, P.; Popall, M. J. Mater. Chem.
2005, 15, 3559–3592.
(14) Coti, K. K.; Belowich, M. E.; Liong, M.; Ambrogio, M. W.; Lau,
Y. A.; Khatib, H. A.; Zink, J. I.; Khashab, N. M.; Stoddart, J. F. Nanoscale
2009, 1, 16–39.
(15) Zamboulis, A.; Moitra, N.; Moreau, J. J. E.; Catto€en, X.;
Wong Chi Man, M. J. Mater. Chem. 2010, 20, 9322–9338.
(16) Shimojima, A.; Kuroda, K. Chem. Rec. 2006, 6, 53–63.
(17) Carlos, L. D.; de Zea Bermudez, V.; Amaral, V. S.; Nunes, S. C.;
Silva, N. J. O.; Ferreira, R. A. S.; Rocha, J.; Santilli, C. V.; Ostrovskii, D.
Adv. Mater. 2007, 19, 341–348.
(18) Inagaki, S.; Guan, S.; Oshuna, T.; Terasaki, O. Nature 2002,
416, 304–307.
(19) Asefa, T.; MacLachlan, M. J.; Coombs, N.; Ozin, G. A. Nature
1999, 402, 867.
(20) Chevalier, P. M.; Corriu, R. J. P.; Moreau, J. J. E.; Wong Chi
Man, M. J. Sol-Gel Sci. Technol. 1997, 8, 603–607.
(21) Chevalier, P. M.; Corriu, R. J. P.; Delord, P.; Moreau, J. J. E.;
Wong Chi Man, M. New J. Chem. 1998, 22, 423–433.
(22) Marx, S.; Avnir, D. Acc. Chem. Res. 2007, 40, 768–776.
(23) Glad, M.; Norrll€ow, O.; Sellergren, B.; Siegbahn, N.; Mosbach,
K. J. Chromatogr. A 1985, 3437, 11–23.
The first example of a pH-responsive bridged silsesquioxane
M1 obtained from a bis-silylated triazine derivative with an ADA
face is reported. This was achieved by the solꢀgel hydrolysisꢀ
condensation of the precursor in the presence of cyanuric acid
H-bonded through the three DAD faces which produced materi-
als with fiber-like morphologies. In contrast, only spherical
objects were formed when the bis-silylated precursor was hydro-
lyzed without cyanuric acid, showing the critical role of the
template molecule in modulating the structure of the hybrid
material. The template could be easily removed from M1 upon
acid treatment without altering the textural integrity of the material.
Furthermore, the functional organic fragments remain covalently
bonded to the silica network demonstrating the chemical and
thermal stability of the hybrid materials. It was shown that a mild
acidic medium (pH = 5.5) is effective in releasing the template
from M1. This positive and encouraging result is stimulating for
further studies in the development of such hybrid materials mainly
for examining and creating new efficient carriers for delivery systems.
’ ASSOCIATED CONTENT
S
Supporting Information. Figures S1 and S2. This material
b
(24) Fireman-Shoresh, S.; Popov, I.; Avnir, D.; Marx, S. J. Am. Chem.
Soc. 2005, 127, 2650–2655.
is available free of charge via the Internet at http://pubs.acs.org.
(25) Burleigh, M. C.; Dai, S.; Hagaman, E. W.; Lin, J. S. Chem. Mater.
2001, 13, 2537–2546.
’ AUTHOR INFORMATION
(26) Arrachart, G.; Carcel, C.; Trens, P.; Moreau, J. J. E.;
Wong Chi Man, M. Chem. Eur. J. 2009, 15, 6279–6288.
(27) Shea, K. J.; Loy, D. A.; Webster, O. W. J. Am. Chem. Soc. 1992,
114, 6700–6710.
Corresponding Author
*E-mail: carole.carcel@enscm.fr (C.C.) and michel.wong-chi-man@
enscm.fr (M.W.C.M.).
(28) Corriu, R. J. P.; Moreau, J. J. E.; Thꢀepot, P.; Wong Chi Man, M.
Chem. Mater. 1992, 4, 1217–1224.
(29) Moreau, J. J. E.; Pichon, B. P.; Arrachart, G.; Wong Chi Man,
M.; Bied, C. New J. Chem. 2005, 29, 653–658.
’ ACKNOWLEDGMENT
We acknowledge FAME, Ecole Nationale Supꢀerieure de Chimie
de Montpellier, and CNRS for financial support. We thank D. Cot
(IEM Montpellier) and P. Dieudonnꢀe (LCVN Montpellier) for
their contribution in the SEM and PXRD measurements, and also
J. Bartlett (University of Western Sydney) for fruitful discussions
and careful reading of the manuscript. C.T. gratefully acknowledges
the “Ministꢁere de l’Enseignement Supꢀerieur et de la Recherche” for
a PhD grant.
(30) Arnal-Herault, C.; Barboiu, M.; Petit, E.; Michau, M.;
Van der Lee, A. New J. Chem. 2005, 29, 1535–1539.
(31) Rouquerol, F.; Rouquerol, J.; Sing, K. In Adsorption by Powders
and Porous Solids; Academic Press: New York, 1999.
(32) Trens, P.; Russell, M. L.; Spjuth, L.; Hudson, M. J.; Liljenzin,
J. O. Ind. Eng. Chem. Res. 2002, 41, 5220–5225.
(33) Arduini, M.; Crego-Calama, M.; Timmerman, P.; Reinhoudt,
D. N. J. Org. Chem. 2003, 68, 1097–1106.
(34) Pittelkow, M.; Lewinsky, R.; Christensen, J. B. Synthesis 2002,
15, 2195–2202.
’ REFERENCES
(1) Farokhzad, O. C.; Langer, R. Adv. Drug Delivery Rev. 2006, 58
1456–1459.
(2) Caruso, F.; Caruso, R. A.; Moehwald, H. Science 1998, 282
1111–1114.
(3) Fatnassi, M.; Tournꢀe-Pꢀeteilh, C.; Cacciaguerra, T.; Dieudonnꢀe,
P.; Devoisselle, J.-M.; Alonso, B. New J. Chem. 2010, 34, 607–610.
(4) Wight, A. P.; Davis, M. E. Chem. Rev. 2002, 102, 3589–3614.
(5) Wang, Y.; Bansal, V.; Zelikin, N.; Caruso, F. Nanoletters 2008,
8, 1741–1745.
(35) Bantignies, J.-L.; Vellutini, L.; Maurin, D.; Hermet, P.;
Dieudonnꢀe, P.; Wong Chi Man, M.; Bartlett, J. R.; Bied, C.; Sauvajol,
J.-L.; Moreau, J. J. E. J. Phys. Chem. B 2006, 110, 15797–15802.
(36) Innocenzi, P. J. Non-Cryst. Solids 2003, 316, 310–319.
(37) Kimizuka, N.; Kawasaki, T.; Hirata, K.; Kunitake, T. J. Am.
Chem. Soc. 1998, 120, 4094–4104.
(6) Son, S. J.; Reichel, J.; He, B.; Schuchman, H.; Lee, S. B. J. Am.
Chem. Soc. 2005, 127, 7317–7317.
(7) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartulli, J. C.; Beck,
J. S. Nature 1992, 359, 710–712.
2106
dx.doi.org/10.1021/cm103327y |Chem. Mater. 2011, 23, 2100–2106