Molecularly imprinted mesoporous silica particles showing a rapid kinetic binding

Article abstract of DOI:10.1039/c003173a

We used periodic mesoporous silica particles for molecular imprinting. The imprinted silica particles showed fast kinetic binding for the template due to their nanosized wall thickness and high surface area.

Full text of DOI:10.1039/c003173a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Molecularly imprinted mesoporous silica particles showing a rapid kinetic
bindingw
Byung Mun Jung, Min Soo Kim, Woo Jin Kim and Ji Young Chang*
Received 15th February 2010, Accepted 19th March 2010
First published as an Advance Article on the web 14th April 2010
DOI: 10.1039/c003173a
We used periodic mesoporous silica particles for molecular
imprinting. The imprinted silica particles showed fast kinetic
binding for the template due to their nanosized wall thickness
and high surface area.
surface of
group.11,12
a silica material with a functional organic
Herein, we report molecularly imprinted periodic meso-
porous silica particles. We chose diethylstilbestrol (DES)
having two terminal phenol groups on both sides as the
template molecule. DES is a synthetic non-steroidal estrogen,
which is known for its ability to treat prostate cancer, but which
Molecular imprinting has been considered a practical way
to produce materials with molecular recognition properties
because of its ease of use and low cost in comparison with the
complicated processes of biological antigen–antibody systems.1–6
In the molecular imprinting process, functional monomers are
complexed with a template molecule through covalent or non-
covalent bonding and then polymerized with a cross-linking
agent. The removal of the template from the polymerized
matrix generates the binding sites that are compatible to the
template. The additional advantages of molecularly imprinted
materials, as compared to biological receptors, include their
mechanical and chemical stability and wide range of operating
conditions. However, they suffer from some drawbacks in
certain applications, such as the heterogeneous distribution
of the binding sites, poor site accessibility and low capacity
and selectivity. The accessibility problem becomes particularly
serious when the imprinted polymers are prepared as bulk
materials. In recent years, considerable efforts have been made
to overcome this problem by adapting a nanofabrication
technique. For example, core–shell structured particles or
nanocapsules with rebinding sites in the outer shell or in the
thin wall have shown improved accessibility and capacity.7–10
Mesoporous silica materials have promising structural
properties for an imprinting matrix, as their rigid structures
are highly suitable for the formation of a delicate recognition
site. In addition, their high pore volume and nanosized pore
wall thickness guarantee the generation of recognition sites
near to the surface and thereby offer high accessibility for a
target molecule. This is especially important for a highly
crosslinked matrix because it prevents a molecule from moving
freely. In general, mesoporous silica materials are prepared by
the gelation of a tetraalkyl orthosilicate with a structure-
directing surfactant and the periodic pore structures are
obtained under controlled reaction conditions. The inner wall
of the pores can be further functionalized by the co-condensation
of a tetraalkyl orthosilicate and a functional organosilica
precursor or by the subsequent modification of the pore
is suspected of being an endocrine disrupting chemical.13
A
template–monomer complex was prepared, where the template
molecule was bonded to two reactive silyl groups by means of
a thermally reversible urethane bond. The template–monomer
complex was used for the sol–gel reaction together with
tetraethyl orthosilicate and a structure-directing surfactant.
As the thermally reversible urethane bond was dissociated at
high temperatures, we were able to extract the template
molecules easily by simple heating. Two-point binding sites
were expected to form between the pores of the silica matrix,
which would show a rapid kinetic uptake.
A triethoxysilane–template complex (DES-Si) was prepared
according to Scheme 1. DES was reacted with 3-(triethoxysilyl)-
propyl isocyanate in THF in the presence of dibutyltin dilaurate
as a catalyst. In the complex, the template was connected to
the silyl group via a thermally reversible urethane bond. We
have previously shown that the urethane bond formed between
an isocyanate and a phenol was stable at room temperature,
but reversible cleavage occurred at elevated temperatures.14–16
The structure of the DES-Si was confirmed by 1H and 13C
NMR spectroscopy, FT-IR spectroscopy and elemental
analysis.
The DES-imprinted mesoporous silica particles were
prepared according to Scheme 2. First, cetyltrimethylammonium
bromide (CTAB) was dissolved in an aqueous base and then
the silica source [tetraethyl orthosilicate (TEOS) and DES-Si]
was added to the solution. The mixture was stirred at 60 1C
to give white precipitates. The surfactant was removed by
washing with aqueous HCl. The control silica was also syn-
thesized in a similar manner, but (3-aminopropyl)triethoxy-
silane was used instead of DES-Si.
Department of Materials Science and Engineering,
College of Engineering, Seoul National University,
Seoul 151-744, Korea. E-mail: jichang@snu.ac.kr;
Fax: +82-2-885-1748; Tel: +82-2-880-7190
w Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/c003173a
Scheme 1 Synthesis of DES-Si.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 3699–3701 | 3699

View Article Online
Fig. 2 (a) X-Ray diffractogram and (b) N2 adsorption–desorption
isotherms of the mesoporous silica particles (inset: pore size
distribution).
1040 m2
g
ꢁ1. The Barret–Joyner–Halenda (BJH) pore size
distribution function calculated from the adsorption branch of
the isotherms [Fig. 2a (inset)] showed uniform mesopores with
an average diameter of 3.06 nm, which was close to that
observed in the TEM image. The pore volume was estimated
Scheme 2 Schematic procedure for synthesizing the molecularly
to be 1.20 cm3 gꢁ1
.
imprinted mesoporous silica particles.
To extract the template molecules, the mesoporous silica
particles were refluxed in a mixture of 1,4-dioxane and
water. In the process of extraction, isocyanate groups were
regenerated by dissociation of urethane bonds and immediately
reacted with water to yield amino groups.16 The extraction
of the template from the silica particles was confirmed by
solid-state 13C NMR spectroscopy. In Fig. 3, the peaks at 6, 18
and 40 ppm were assigned to the propyl carbons bonded to Si.
The peaks for the phenyl carbons of DES moieties appeared
at around 140 ppm. After extraction, these peaks almost
disappeared, indicating that most of the template molecules
were removed.
The ability of the imprinted silica to recognize the template
(DES) was investigated. In the rebinding test, the imprinted
silica or the control silica was added to a solution of DES in
THF at various concentrations. After incubating for 2 h at
room temperature, the silica particles were isolated by filtra-
tion. The amount of the template absorbed by the polymer
was determined by measuring the residual analyte in the
filtrate by reverse phase high performance liquid chromato-
graphy (HPLC).
Fig. 1 (a) SEM and (b) TEM images of the mesoporous silica
particles.
The structure of the mesoporous silica particles was charac-
terized by scanning electron microscopy (SEM), transmission
electron microscopy (TEM), X-ray diffraction (XRD) and
solid-state 13C NMR spectroscopy. Visual images of the particles
were obtained by SEM and TEM (Fig. 1). The SEM image
shows spherical silica particles with an average diameter of
about 100 nm. The TEM image reveals a densely packed,
hexagonal pore structure with an average pore diameter of
about 3 nm and a distance between the pores of about 5 nm.
The hexagonally ordered structure of the mesoporous silica
particles was also confirmed by XRD (Fig. 2a). The XRD
pattern showed three reflections corresponding to d spacings
of 4.64, 2.61 and 2.27 nm, which were indexed as (100), (110)
and (200) reflections, respectively, of a hexagonal lattice with a
lattice parameter of a = 5.36 nm. This value was close to that
observed in the TEM image. The wall thickness between the
pores was estimated to be about 2.3 nm based on the lattice
parameter and the average pore size. Since the length of DES
calculated by an MM2 method was about 1.4 nm, most of the
template molecules were very likely embedded in the silica
matrix between the pores.
Fig. 4 shows the amount of the template bound to the
imprinted silica and to the control silica according to the
sample concentration. The imprinted silica exhibited much
higher recognition ability than the control silica. The superior
The porosity of the silica particles was investigated by
a
nitrogen adsorption–desorption experiment (Fig. 2b),
which showed type IV curves indicating that the pore sizes
were in the mesopore range. The Brunauer–Emmett–Teller
(BET) surface area obtained from the nitrogen isotherms was
Fig. 3 Solid-state 13C NMR spectra of the mesoporous silica
particles: (a) before and (b) after the extraction of DES. The asterisks
(*) denote the peaks of ethanol.
ꢀc
This journal is The Royal Society of Chemistry 2010
3700 | Chem. Commun., 2010, 46, 3699–3701

View Article Online
This result was attributed to the generation of binding sites
between the pores of the periodic mesoporous silica, which
helped the template to access the binding sites easily.
In summary, we demonstrated the possibility of using
periodic mesoporous silica particles for molecular imprinting.
As expected, the imprinted mesoporous silica particles showed
fast kinetic binding for the template due to their thin wall
thickness and high surface area. The thermally reversible bond
was used in the preparation of the template–silica monomer
complex, which facilitated the removal of the template from
the silica matrix by simple heating. We believe that this
approach will allow us to further modify the functionality of
the binding sites and thus improve the selectivity for the
specific target molecule.
Fig. 4 (a) Binding selectivity of the imprinted silica and the control
silica. (b) Kinetic binding profile of DES to the imprinted silica and the
control silica.
Table 1 Partition coefficients and imprinting factors (IFs) for the
imprinted silica and the control silica
This research was supported by a grant from the Funda-
mental R&D Program for Core Technology of Materials
funded by the Ministry of Knowledge Economy, Republic
of Korea.
DES
4,40-Biphenol
Bisphenol-A
Hydroquinone
Ki
Kc
IF
1.40
0.33
4.23
0.08
0.05
1.57
0.34
0.11
3.02
0.05
0.08
0.61
Notes and references
rebinding ability of the imprinted silica, compared to the
control silica, proved that the target molecules were not simply
adsorbed at the surface, but were trapped in the cavities
through hydrogen bonding.
1 G. Wulff, Chem. Rev., 2002, 102, 1.
2 K. Haupt and K. Mosbach, Chem. Rev., 2000, 100, 2495.
3 K. Haupt, Anal. Chem., 2003, 75, 376.
4 R. Ouyang, J. Lei and H. Ju, Chem. Commun., 2008, 5761.
5 K. Yoshimatsu, L. Ye, P. Stenlund and I. S. Chronakis, Chem.
Commun., 2008, 2022.
6 H.-Y. Lee and B. S. Kim, Biosens. Bioelectron., 2009, 25, 587.
7 W. H. Zhou, C. H. Lu, X. C. Guo, F. R. Chen, H. H. Yang and
X. R. Wang, J. Mater. Chem., 2010, 20, 880.
8 X. Shen, L. Zhu, J. Li and H. Tang, Chem. Commun., 2007, 1163.
9 T. H. Kim, C. D. Ki, H. Cho, T. Chang and J. Y. Chang,
Macromolecules, 2005, 38, 6423.
10 C. He, Y. Long, J. Pan, K. Li and F. Liu, J. Mater. Chem., 2008,
18, 2849.
11 D. A. Loy and K. J. Shea, Chem. Rev., 1995, 95, 1431.
12 T. Tani, N. Mizoshita and S. Inagaki, J. Mater. Chem., 2009, 19,
4451.
13 T. Kitamura, S. Nishimura, K. Sasahara, M. Yoshida, J. Ando,
M. Takahashi, T. Shirai and A. Maekawa, Cancer Lett., 1999, 141,
219.
14 J. D. Bass, S. L. Anderson and A. Katz, Angew. Chem., Int. Ed.,
2003, 42, 5219.
We also investigated the specific recognition ability of the
imprinted silica for its structural analogues (4,40-biphenol,
bisphenol-A and hydroquinone). The rebinding test was carried
out in the same manner as described above. The imprinted
silica showed the highest recognition ability for DES. From
the results in Fig. 4, a partition coefficient K could be obtained
by the following equation: K = (moles of the analyte bound to
particles/mass of particles)/(moles of the analyte remaining
in the solution/mass of the solution). Also calculated was
the imprinting factor (IF), which was the ratio of partition
coefficients of the imprinted and the control particles (i.e.,
IF = Ki/Kc). The K and IF values are summarized in Table 1.
The higher partition coefficients of DES for the imprinted
silica indicate a higher affinity of the imprinted silica toward
DES than for its analogues.
15 C. D. Ki, C. Oh, S. G. Oh and J. Y. Chang, J. Am. Chem. Soc.,
2002, 124, 14838.
16 K. Lee, C. D. Ki, H. Kim and J. Y. Chang, Macromolecules, 2004,
37, 5544.
17 N. M. Brunkan and M. R. Gagne, J. Am. Chem. Soc., 2000, 122,
6217.
18 T. Kobayashi, T. Fukaya, M. Abe and N. Fujii, Langmuir, 2002,
18, 2866.
The kinetic uptake of DES by the imprinted silica was
measured in order to evaluate the binding site accessibility
(Fig. 4b). An equilibrium amount of 95% was achieved within
5 min after the addition of DES and complete equilibrium
reached within 10 min. This recognition time was much
faster than that of typical molecularly imprinted materials.16–18
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 3699–3701 | 3701