Вы находитесь на странице: 1из 6

Biosensors and Bioelectronics 25 (2010) 17421747

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journal homepage: www.elsevier.com/locate/bios

Clean synthesis of molecular recognition polymeric materials with

chiral sensing capability using supercritical fluid technology.
Application as HPLC stationary phases
Mara Soares da Silva a , Eva R. Vo a , Mrcio Temtem a , Lus Mafra b , Jorge Caldeira a,c ,
Ana Aguiar-Ricardo a , Teresa Casimiro a,

REQUIMTE, Departamento de Qumica, Faculdade de Cincias e Tecnologia, FCT, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal
Universidade de Aveiro, Departamento de Qumica, CICECO, Campus Santiago, P-3810-193 Aveiro, Portugal
Centro de Investigaco Interdisciplinar Egas Moniz, Instituto Superior de Cincias da Sade Egas Moniz, 2825-511 Caparica, Portugal

a r t i c l e

i n f o

Article history:
Received 31 August 2009
Received in revised form
10 November 2009
Accepted 18 December 2009
Available online 28 December 2009
Supercritical fluid technology
Molecular imprinting
Enantiomeric separation

a b s t r a c t
Molecularly imprinted polymers (MIPs) of poly(ethylene glycol dimethacrylate) and poly(Nisopropylacrylamide-co-ethylene glycol dimethacrylate) were synthesized for the first time in
supercritical carbon dioxide (scCO2 ), using Boc-l-tryptophan as template. Supercritical fluid technology
provides a clean and one-step synthetic route for the preparation of affinity polymeric materials with
sensing capability for specific molecules. The polymeric materials were tested as stationary HPLC phases
for the enantiomeric separation of l- and d-tryptophan. HPLC results prove that the synthesized MIPs are
able to recognize the template molecule towards its enantiomer which opens up potential applications
in chromatographic chiral separation.
2009 Elsevier B.V. All rights reserved.

1. Introduction
Molecular recognition technology is a promising alternative
way to create highly specific sites for a molecule, within a polymer,
via imprinting polymerization. Molecular imprinting technique
uses the functionality of the target molecule (template), to assemble its own recognition site by forming specific interactions with the
matrix (Alexander et al., 2006). In the molecular imprinting process,
the functional group of the monomer interacts with the template
molecule, in the presence of a porogen and a cross-linker agent that
freezes the complex within a porous polymer matrix (Mosbach and
Haupt, 1998). The imprinted sites formed, chemical and physically
complementary to the template molecule are then made accessible
by dissociation of the complex, by template removal.
Regarding the interactions between monomer and template,
three different molecular imprinting approaches can be considered,
covalent (Wulff et al., 1986), non-covalent (Duarte et al., 2006; Wei
and Mizaikoff, 2007; Zhang et al., 2006; Yin et al., 2005) and semi-

Corresponding author. Tel.: +351 212948385; fax: +351 212948385.

E-mail address: teresa.casimiro@dq.fct.unl.pt (T. Casimiro).
URL: http://www.dq.fct.unl.pt/scf/ (T. Casimiro).
0956-5663/$ see front matter 2009 Elsevier B.V. All rights reserved.

covalent (Sellergren and Andersson, 1990). The easy and cheap

preparation, allied with a high specificity, turned out non-covalent
imprinting as the most widely used approach. By means of the
non-covalent route, the monomertemplate complex is stabilized
through numerous non-covalent interactions such as hydrogen
bonds, ion-paring and dipoledipole interactions.
These robust materials have demonstrated high affinity and
specificity, very good thermal and chemical stability, repeated
operations without loss of activity, high mechanical strength, durability to heat and pressure, and applicability in harsh chemical
media. These properties have led to a significant increase in applications in the last decade such as in synthesis and catalysis (Alexander
et al., 2003), extraction (Baggiani et al., 2007), chromatography
(Wei and Mizaikoff, 2007) and sensors (Guan et al., 2008). Their high
selectivity, in particular, has propelled the development of MIPs
as chromatographic stationary phases. MIPs have shown several
advantages over other chiral matrixes (Maier and Lindner, 2007).
They have proven to be more stable towards thermal, mechanical and chemical stress, which makes them perfect for analysis in
aggressive environments (Svenson and Nicholls, 2001).
A recent review details different synthesis approaches
(Alexander et al., 2006). The most used is bulk synthesis, which
yields a hard block polymer, followed by grinding and sieving.

M. Soares da Silva et al. / Biosensors and Bioelectronics 25 (2010) 17421747

Besides time-consuming, particles obtained are irregular and most

interaction sites are destroyed reducing the recognition capacity
(Qiao et al., 2006). Also slow interaction kinetics and heterogeneous nature of the binding sites with respect to geometry and
accessibility have been reported (Alexander et al., 2003). Alternative methods have been proposed, such as in situ, multi-step
swelling and suspension polymerizations (Wei and Mizaikoff,
2007). However they still show limitations such as heterogeneity,
complicated procedures, large use of template, phase partitioning
and use of organic solvents. Much more has to be done so that these
MIPs can be widely accepted in industry concerning specificity,
reproducibility and sustainability.
The increasing importance and world wide applications of these
high affinity materials, the need to improve their efficiency and
turn the processes greener eliminating or drastically reducing
the use of volatile organic compounds (VOCs), are major driving forces of this work. Conventional imprinting polymerizations
involve excessive use of hazardous organic solvents. In most cases
MIPs for enantiomers of amino acid derivatives have been prepared
in organic solvents and used with organic mobile or in aqueous
phases in HPLC (Haginaka and Kagawa, 2004). It is difficult to prepare MIPs in water since the high concentration of water molecules
destroys the polar interactions between the functional monomer
and the template, thus organic solvents are used (Yu and Mosbach,
1998). Due to environmental and human safety concerns, chemists
and chemical engineers are currently seeking new and cleaner
methods for polymer processing. Supercritical fluids are considered interesting alternative to most traditional solvents because of
their physical and chemical properties. Supercritical carbon dioxide
(scCO2 ), in particular, possesses innumerous properties that made
it emerge as the most extensively studied supercritical fluid for
polymerization reactions. It is inexpensive, has a low critical point,
and is a gas under ambient conditions (Soares da Silva et al., 2007;
Barroso et al., 2009).
Recently we have synthesized for the first time a molecular
recognition polymeric material using supercritical fluid technology. Micron-sized spherical particles of poly(diethylene glycol
dimethacrylate), PDEGDMA MIP, were prepared in scCO2 using salicylic acid and acetylsalicylic acids as model templates, for potential
applications in controlled drug release devices (Duarte et al., 2006).
Previously we had optimized the synthesis of this highly crosslinked material in scCO2 (Casimiro et al., 2005). This work was
followed by the work of Ye et al. (2006) who synthesized MIPs of
methacrylic acid and divinylbenzene using propanolol as template.
This new clean synthetic approach for materials with molecular recognition capacity brings notorious advantages compared
to conventional methods such as more controlled morphology,
high porosity and specific area, being obtained as dry, pure, sterile
flowing powders, without making use of organic solvents VOCs, or
intensive purification and drying steps. The molecular recognition
will be achieved by the self-assembly of the functional monomers
and the template, for which the high affinity is wanted, in a highly
cross-linked polymer using scCO2 as porogenic agent.
In the last years scCO2 has already proved to be an excellent medium for the synthesis and processing of polymers due
to its high density, high diffusivity and low viscosity (Cooper,
2000; Giles et al., 2001). It is well known that the polymerization
conditions can highly influence the performance of the molecularly imprinted polymers (Piletska et al., 2009). Temperature,
solvent polarity, pressure, monomer concentration and interactions solvent-monomers-template are known to have practical
effects on selectivity, which are associated to the stability of the
self-assembly complexes during polymerization, which in turn
influences the recognition process. The high diffusivity and low viscosity of scCO2 will decrease the mass transfer limitations found in
conventional synthesis.


The high affinity of molecular recognition-based materials

makes them a versatile tool in separation technology. MIPs can
replace expensive chiral columns, avoid chiral mobile phase additives, derivatisation with chiral reagents, etc. (Maier and Lindner,
The aim of this work was to address the possibility of developing
such a high affinity material using this clean technology, able to
distinguish very similar molecules such as enantiomeric species,
which could have a real impact in chromatography.
Here we report the preparation of a HPLC column in a two step
process: first the benign synthesis of a MIP with specificity for Bocl-tryptophan, using a clean technology, followed by its loading into
a blank HPLC column in order to be tested as stationary phase. Polymers were characterized by scanning electron microscopy, SEM,
adsorption of N2 according to the BET method and by 13 C CPMAS
NMR. Boc-l-tryptophan was used as a model enantiomer as a proof
of concept. The separation of l- and d-enantiomers could have
application for instance in the identification and quantification of
amino acids in biological samples (Zhao and Liu, 2001).
Herein we open up a new way of synthesizing chiral polymeric
stationary phases for chromatography, with no use of organic solvents in the polymerization reaction. The molecular recognition
capability of the developed materials was investigated concerning the influence of functional monomer addition, sample load and
2. Experimental
2.1. Materials
Ethylene glycol dimethacrylate (EGDMA, 98% purity) as crosslinker, N-isopropylacrylamide (NIPA, 97% purity) as functional
monomer, Boc-l-tryptophan (Boc-l-trp, 99% purity) as template
molecule, Boc-d-tryptophan (Boc-d-trp, 98% purity) and azobis(isobutyronitrile) (AIBN, 98% purity) were purchased from
SigmaAldrich. HPLC grade acetonitrile from Scharlau was used.
Carbon dioxide was obtained from Air Liquide with purity better
than 99.998%. All chemicals were used without further purification.
2.2. MIP and NIP synthesis in scCO2
Polymerization reactions in scCO2 were carried out as described
elsewhere (Casimiro et al., 2005) in a 33 ml stainless steel highpressure cell equipped with two aligned sapphire windows and a
Teflon coated magnetic stir bar inside. In a typical procedure to
produce MIPs, cross-linker (3.43 g), monomer (12 wt% with respect
to the amount of cross-linker, if included), initiator (2 wt%) and
template (1 wt%) were loaded into the high-pressure cell. The cell
was immersed in a thermostatted water bath with 0.01 C of
stability and temperature control was made with a RTD probe
contacting the cell, connected to a Hart Scientific PID controller.
Temperature was set to 65 C, the optimal AIBN initiation temperature, and after purging with nitrogen, CO2 was introduced up
to 21 MPa. This pressure was used since at these conditions an
initial single homogeneous phase is assured, with all reactants completely dissolved in the supercritical phase, as it can be observed
through the sapphire windows. This is a crucial factor in order to
have a completely homogeneous polymer initiation and formation. The reaction proceeded for 24 h. At the end of the reaction,
the resulting polymer was slowly washed with fresh high-pressure
CO2 in order to clean the remaining residues of template and
unreacted monomer. For the production of NIPs the same procedure was followed except no template was added. Two MIPs were
produced: MIP1 corresponding to imprinted PEGDMA and MIP2
corresponding to imprinted PEGDMA-co-PNIPA. NIP1 and NIP2 are
the respective non-imprinted materials.


M. Soares da Silva et al. / Biosensors and Bioelectronics 25 (2010) 17421747

2.3. HPLC analysis

The synthesized polymers were slurry packed into a blank column (Supelco; 150 mm 4.6 mm) using a mixture of Millipore
water and acetonitrile (8%, v/v). Chromatographic analyses were
carried out using a Merck L-7100 HPLC pump equipped with an
L-7400 UV detector, D-7000 computer interface and a XT Maraton autosampler. UV detection was made at 275 nm. Prior to the
experiments column was thoroughly washed with acetonitrile with
1% (v/v) of acetic acid in order to further eliminate any unreacted
material, until a stable baseline was achieved. Experiments were
carried out at 25 and 65 C using a column oven (XT Maraton). The
mobile phase used was hydro-organic with 8% acetonitrile by volume at a flow rate of 0.5 ml min1 . 0.254 mM sample were injected
for analysis using loop volume of 5 ml. Acetone was used as void
marker, to determine the column dead time. MIP2 was packed in a
longer column (Supelco; 250 mm 4.6 mm) in order to increase the
enantiomeric resolution. This column was also tested in a Knauer
equipment, with IR detection Smartline 2300 and pump Smartline
2.4. Polymer characterization
Polymers were morphologically characterized using scanning
electron microscopy (SEM) in a Hitachi S-2400 instrument, with
an accelerating voltage set to 15 kV. Samples were mounted on
aluminium stubs using carbon tape and were gold coated.
Specific surface area and pore diameter of the polymeric particles were determined by adsorption of N2 according to the BET
method. An accelerated surface area and porosimetry system (ASAP
2010 Micromeritics) was used under nitrogen flow.
13 C CPMAS NMR spectra were recorded on a Bruker AVANCE
400 (DSX) WB spectrometer (9.4 T), using a double-resonance 4 mm
VTN probe, with a Larmor frequency of 100.62 MHz. The samples were packed into a 4 mm diameter ZrO2 cylindrical rotor and
spun at MAS rates of 7 and 10 kHz. The 13 C CPMAS spectrum was
recorded using a RAMP-CP shape (10050% amplitude); rf field
strengths between 50 and 70 kHz for Hartman-Hahn condition calibration for 13 C and 1 H channels; number of scans = 24 K; recycle
delay of 5 s; contact time = 1.5 ms; 1 H decoupling was employed
using the Small Phase INcremental ALternation (SPINAL64) multiple pulse scheme with 64 steps (Fung et al., 2000), during the
13 C acquisition. A pulse length of 4.5 ms (ca. 165 pulses) was used
for the basic unit of the SPINAL64 scheme, while the 1 H radiofrequency field strength (1 /2) employed was 105 kHz. Chemical
shifts were quoted in parts per million (ppm) from solid glycine.

Table 1
Physical characteristics (surface area, average pore diameter and specific pore volume) of both MIPs and NIPs, obtained by multipoint BET method (type II).

BET surface area

(m2 /g)

Pore volume
(cm3 /g)

Pore diameter





The physical characteristics of both MIPs and NIPs concerning surface area, average pore diameter and specific pore volume,
obtained by multipoint BET method (type II) are shown in Table 1.
The surface areas obtained were very similar with those
obtained for PNIPA synthesized in scCO2 (Temtem et al., 2007),
mainly due to the known porogenic ability of CO2 . MIP1 and NIP1
show a slight increase on the surface area compared to MIP2 and
NIP2 which could be explained by the fact that the primary particles in this case are smaller, thus, providing a material with higher
porosity comprising micron-sized agglomerates of nano-primary
particles, as it can be observed on SEM images. This morphology
is consistent with other reported precipitation polymerizations in
scCO2 without using stabilizers (Cooper et al., 1999; Casimiro et
al., 2005). The small differences observed, on the pore volume and
diameter for the MIPs and NIPs could be explained by the interference of the template molecule in the nucleation process and/or the
presence of template residues within the final polymer.
Fig. 1 depicts the 13 C CPMAS NMR spectra of imprinted and nonimprinted polymers. The strong similarities in the 13 C chemical
shifts and relative intensities between MIP2 and NIP2 co-polymers,

3. Results and discussion

3.1. Characterization of the synthesized polymers
The molecular imprinted and non-imprinted polymers were
obtained as dry, free-flowing powders in high yields (99%, determined gravimetrically). The synthesized polymeric material is
obtained ready to be slurry packed which can be considered one
major advantage over traditional MIPs formation where materials
have to be grounded and sieved prior to use, leading to product losses that sometimes reach more than 50% (Baggiani et al.,
2005). The pressure-temperature conditions of the reaction assure
an initial homogeneous system, as it can be seen through the
sapphire windows. SEM images revealed that the imprinted and
non-imprinted polymers show similar morphology, aggregates of
smooth surfaced discrete nanoparticles (see Fig. S3, supplementary
information). It seems that by introducing a small amount of
template during the polymerization step, the morphology of the
polymers is not greatly affected.

Fig. 1. 13 C CPMAS NMR spectra and peak assignment. (a) MIP1, (b) MIP2, (c) NIP1
and (d) NIP2. Asterisks depict spinning sidebands.

M. Soares da Silva et al. / Biosensors and Bioelectronics 25 (2010) 17421747

and MIP1 and NIP1 polymers, suggest that imprinted and nonimprinted materials are chemically equivalent. Being aware of the
insensitive character of NMR to detect very diluted species, the
NMR data indicates that the Boc-l-trp template is not visible on
the 13 C spectra after washing-out the drug template from MIP1
and MIP2 polymers. This observation suggests a successful polymer cleaning upon washing (if not completely removed, at least
only a very low template concentration remains).
Structural differences between the co-polymer (MIP2/NIP2) and
the polymer (MIP1/NIP1) may be observed by the presence of a faint
resonance at ca. 42 ppm on MIP2/NIP2 due to the additional contribution of the aCH2 groups on NIPAs ligands, which have slightly
more shielded chemical shifts. Moreover, the additional peak intensity at ca. 22 ppm, observed on MIP2/NIP2 (Fig. 1b and d), with
respect to same chemical shift region on the MIP1/NIP1 polymers
(compare Fig. 1a and c), is originated from NIPAs terminal methylic
carbons, which are only present on MIP2/NIP2 polymers. A detailed
peak assignment of the relevant 13 C resonances is also shown in
Fig. 1.
The presence of weak resonances appearing at ca. 125, 137 ppm
(ab unsaturated carbons from NIPA and EGDMA) and ca. 167 ppm
(carbonyl groups from NIPA and EGDMA) shows that a relatively
small amount of unreacted double bonds exists on the polymers, which is in accordance with other works found in literature
(Skogsberg et al., 2007).
3.2. Retention factor and enantioselectivity
The molecular recognition and affinity of the polymers for the
template enantiomer was gauged by their ability to selectively
retain this molecule. Retention or capacity factors, k , of the prepared MIPs were evaluated according to Eq. (1).
k =

tr t0


where tr is the retention time of the amino acid enantiomer and t0

is the retention time of the void marker on the column, acetone.
The enantioseparation factor was calculated using the following




Resolution of the racemic mixture was calculated from Eq. (3),

where WD and WL are the baseline widths at 4.4% of peak height.

2(tL tD )


In order to evaluate the cross-linker capability of creating

specific bonds with the imprinted molecule, a PEGDMA and
PEGDMA-co-PNIPA MIPs were synthesized and tested as chromatographic stationary phases.
MIP1 stationary phase was injected with 1 mM amino acid solutions of Boc-l-trp and Boc-d-trp. The HPLC peaks obtained are broad
and show some asymmetry, which could be explained by the heterogeneity of the binding sites within the matrix (chromatogram
showed on Fig. S4, supplementary information). It is reported
in literature that this heterogeneity is the major contributor to
broad and asymmetric peaks in chromatography (Sellergren and
Shea, 1995). MIP1 shows some molecular recognition for Boc-dtrp, although it is evident that higher retention factors are obtained
for the template molecule (tD
= 19 min compared to tL
= 25 min).
This means that the imprinted stationary phase has non-specific
sites (recognizes both enantiomers) but shows higher specificity
for Boc-l-trp (better recognition of the template enantiomer).
Fig. 2 shows the chromatograms of 0.25 mM amino acid solutions of Boc-l-trp and Boc-d-trp injected separately on MIP2 and


Fig. 2. Chromatograms of MIP2 and NIP2 injected with 0.25 mM solutions of each
tryptophan enantiomer at 25 C. Boc-l-trp (dotted line) and Boc-d-trp (solid line).

NIP2 columns. The ability of a MIP stationary phase to retain the

template molecule compared to the retention at the corresponding non-imprinted column is a commonly accepted indicator for
the selectivity of the produced MIP (OMahony et al., 2006). As it
can be seen the injected amino acid derivatives do not show any
retention on the non-imprinted stationary phase NIP2. In opposition MIP2 show affinity for both molecules, although presenting
a much higher retention factor for the Boc-l-trp, the template
molecule present during polymerization, than for its enantiomer.
Also its peak is broader meaning that there are more interactions
of this molecule with the stationary phase, retarding its elution
Chromatographic experiments were performed at two different
column temperatures 25 and 65 C.
Fig. 3 shows the capacity factors of Boc-l-trp and Boc-d-trp for
both imprinted polymers, MIP1 and MIP2, for the range of sample
loads studied, as calculated by Eq. (1). As it can be seen higher capacity factors can be obtained at 25 C than at 65 C, for both matrixes
and both enantiomers and for all the range of compositions. We
suggest that this could be due to the increasing solvent power of the
mobile phase with temperature, destabilizing the hydrogen bonds
between the eluted species and the stationary phase material and
leading to lower retention times at 65 C. This temperature dependency of the capacity factor is also typically observed in literature
(Haginaka and Kagawa, 2004).
is more pronounced for
Also the difference between kL and kD
MIP2 than for MIP1 as it can be seen comparing Fig. 3c and d, to
Fig. 3a and b, meaning that MIP2 is more selective for the template
enantiomer. MIP2 has less affinity to Boc-d-trp than MIP1, while it
shows higher retention times for Boc-l-trp. This can be explained
by the introduction of NIPA as functional monomer. The introduction of an amide functional group increases the available hydrogen
bonds with the template, which establishes stronger interaction
with the stationary phase in both polar and apolar solvents. This is
in accordance with literature (Cong and Mosbach, 1997). For example, Boc-l-trp showed a higher retention factor on MIP2 at 25 C
and sample load of 0.5 mM (k = 5.03), while the corresponding MIP
without amide functional monomer showed a much lower capacity
factor (k = 3.06) at the same conditions.
increase for loads up to 1 mM,
For MIP1 at 25 C, kL and kD
remaining constant for higher sample loads. For MIP2 at the same
temperature, both capacity factors are almost constant, only a slight
increasing trend is observed up to 4 mM. At 65 C the capacity
factors of MIP and MIP2 for both enantiomers have a notorious
increase for injection loadings up to 1 mM, followed by a less
pronounced increase. Around 3 mM the retention factors tend to

M. Soares da Silva et al. / Biosensors and Bioelectronics 25 (2010) 17421747


Fig. 3. Effect of sample load and temperature on the capacity factors of Boc-l-trp (N) and Boc-d-trp (d). MIP1: (a) 25 C, (b) 65 C and MIP2: (c) 25 C and (d) 65 C.

approach meaning that the stationary phase looses selectivity due

to the saturation of the column.
The capacity factor cannot be directly related to enantiomeric
separation since besides the specific interactions between the template and MIP, there is also the binding site competition of the
enantiomeric species when injecting the racemic solution. The
racemic mixture of Boc-trp was injected on the 150 mm 4.6 mm
columns loaded with MIP1 and MIP2, but no resolution of the peaks
were obtained despite the different capacity factors obtained for
each enantiomer. This can be explained by the competition for
the specific and non-specific sites within the matrix. By increasing the length of the column we were expecting to increase the
enantiomeric resolution. Gilar et al. (2004) described an increase
of the column efficiency with the increase of the column length.
Fig. 4(a) and (b) shows the enantioseparation chromatograms
for MIP2 by injecting the racemate at 65 and 25 C respectively
using a longer column (250 mm 4.6 mm). As it can be seen there
was a significant improvement in the resolution of the enantiomers
by increasing the length of the column, since both species can be
now clearly distinguished. Table 2 presents the retention, enantioselectivity and resolution for Boc-trp enantiomers on MIP2 at different temperatures, when injecting the racemic mixture. As it can
be seen promising enantiomeric separation factors and resolutions
were obtained for both temperatures. By decreasing the column
temperature the stationary phase increased its affinity for the template enantiomer, with an evident increase of the capacity factor.
These results would be quite suitable for instance, for movingbed chromatography, where full separation of pure enantiomers is
Table 2
Retention, enantioselectivity and resolution factors of Boc-trp enantiomers on MIP2
at different temperatures.

25 C
65 C







Fig. 4. Enantioseparation on MIP2 of a 1 mM tryptophan racemic mixture at: (a)

65 C and (b) 25 C.

M. Soares da Silva et al. / Biosensors and Bioelectronics 25 (2010) 17421747

obtained even when the resolution of the two peaks is not excellent
(Juza et al., 2000).
Typically MIPs show better selectivity/rebinding results when
analysed in the same solvent used in their preparation. It has been
reported that changing solvents can affect the integrity of the binding sites weakening the molecular recognition of the template (Yu
and Mosbach, 1998). Since the polymers were synthesized in scCO2
current work is being developed in order to increase resolution by
testing these materials in supercritical fluid chromatography.
4. Conclusions
Herein we were able to prepare polymeric materials with chiral recognition capability by molecular imprinting in scCO2 . This
technology showed to be a promising greener alternative to
conventional techniques in the preparation of chromatographic
columns for enantioseparation. We proved that the synthesized
polymers demonstrate high affinity to the template molecule introduced in the polymerization step, being able to recognize and
differentiate molecules as similar as enantiomeric species by HPLC.
We foresee that a wide range of areas such as sensors, separation,
catalysis, etc., could benefit from the clean development of these
high affinity materials, especially when purity and morphology of
the molecular sensing materials is a key issue.
The authors would like to thank Fundaco para a Cincia e
Tecnologia (FCT-Lisbon) for financial support through projects
PTDC/QUI/66086/2006 and PTDC/CTM/70513/2006, and doctoral
grant SFRH/BD/31085/2006 (M.S.S.), FEDER, FSE and POCTI. The
Portuguese NMR Network is acknowledged for granting access to
the NMR equipment.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bios.2009.12.023.


Alexander, C., Davidson, L., Hayes, W., 2003. Tetrahedron 59, 20252057.
Alexander, C., Andersson, H.S., Andersson, L.I., Ansell, R.J., Kirsch, N., Nicholls, I.A.,
OMahony, J., Whitcombe, M.J., 2006. J. Mol. Recogn. 19, 106180.
Baggiani, C., Anfossi, L., Baravalle, P., Anfossi, L., Tozzi, C., 2005. Anal. Chim. Acta 542,
Baggiani, C., Anfossi, L., Giovannoli, C., 2007. Anal. Chim. Acta 59, 2939.
Barroso, T., Temtem, M., Casimiro, T., Aguiar-Ricardo, A., 2009. J. Supercrit. Fluids 51,
Casimiro, T., Banet-Osuna, A.M., Nunes da Ponte, M., Aguiar-Ricardo, A., 2005. Eur.
Polym. J. 41, 19471953.
Cong, Y., Mosbach, K., 1997. J. Org. Chem 62, 40574064.
Cooper, A.I., Hems, W.P., Holmes, A.B., 1999. Macromolecules 32, 21562166.
Cooper, A.I., 2000. J. Mater. Chem. 10, 207234.
Duarte, A.R.C., Casimiro, T., Aguiar-Ricardo, A., Simplcio, A.L., Duarte, C.M.M., 2006.
J. Supercrit. Fluids 39, 102106.
Fung, B.M., Khitrin, A.K., Ermolaev, K., 2000. J. Magn. Reson. 142, 97101.
Gilar, M., Daly, A.E., Kele, M., Neue, U.D., Gebler, J.C., 2004. J. Chromatogr. A 1061,
Giles, M.R., Griffiths, R.M.T., Aguiar-Ricardo, A., Silva, M.R.C.G., Howdle, S.M., 2001.
Macromolecules 34, 2025.
Guan, G., Liu, B., Wang, Z., 2008. Sensors 8, 82918320.
Haginaka, J., Kagawa, C., 2004. Anal. Bioanal. Chem. 378, 19071912.
Juza, M., Mazzotti, M., Morbidelli, M., 2000. Trends Biotechnol. 18, 108118.
Maier, N.M., Lindner, W., 2007. Anal. Bioanal. Chem. 389, 377397.
Mosbach, K., Haupt, K., 1998. J. Mol. Recogn. 11, 6268.
OMahony, J., Molinelli, A., Nolan, K., Smyth, M.R., Mizaikoff, B., 2006. Biosens. Bioelectron. 21, 13831392.
Piletska, E.V., Guerreiro, A.R., Whitcombe, M.J., Piletsky, S.A., 2009. Macromolecules
42, 49214928.
Qiao, F., Sun, H., Yan, H., Row, K.H., 2006. Chromatographia 64, 625634.
Sellergren, B., Andersson, L., 1990. J. Org. Chem. 55, 33813383.
Sellergren, B., Shea, H.J., 1995. J. Chromatogr. A 690, 2939.
Skogsberg, U., Meyer, C., Rehbein, J., Fischer, G., Schauff, S., Welsch, N., Albert, K.,
Hall, A.J., Sellergren, B., 2007. Polymer 48, 229238.
Soares da Silva, M., Temtem, M., Henriques, S., Casimiro, T., Aguiar-Ricardo, A., 2007.
J. Chem. Eng. Data 52, 19701974.
Svenson, J., Nicholls, I.A, 2001. Anal. Chim. Acta 435, 1924.
Temtem, M., Casimiro, T., Mano, J.F., Aguiar-Ricardo, A., 2007. Green Chem. 9, 75
Wei, S., Mizaikoff, B., 2007. J. Sep. Sci. 30, 17941805.
Wulff, G., Heide, B., Helfmeier, G., 1986. J. Am. Chem. Soc. 108, 10891091.
Ye, L., Yoshimatsu, K., Kolodziej, D., Francisco, J.D.C., Dey, E.S., 2006. J. Appl. Polym.
Sci. 102, 28632867.
Yin, J., Yang, G., Chen, Y., 2005. J. Chromatogr. A 1090, 6875.
Yu, C., Mosbach, K., 1998. J. Mol. Recogn. 11, 6974.
Zhang, H., Ye, L., Mosbach, K., 2006. J. Mol. Recogn. 19, 248259.
Zhao, S., Liu, Y., 2001. Electrophoresis 22, 27692774.