Fluorescent imprinted polymer sensors for chiral amines

Article abstract of DOI:10.1039/b816733h

Molecularly imprinted polymer (MIP) based fluorescent sensors require suitable fluorescent moieties which respond to the binding event with significant fluorescence changes. Two novel polymerisable coumarins: 6-styrylcoumarin-4-carboxylic acid (SCC) and 6

Full text of DOI:10.1039/b816733h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Fluorescent imprinted polymer sensors for chiral amines†
T. Hien Nguyen‡ and Richard J. Ansell*
Received 24th September 2008, Accepted 7th January 2009
First published as an Advance Article on the web 11th February 2009
DOI: 10.1039/b816733h
Molecularly imprinted polymer (MIP) based fluorescent sensors require suitable fluorescent moieties
which respond to the binding event with significant fluorescence changes. Two novel polymerisable
coumarins: 6-styrylcoumarin-4-carboxylic acid (SCC) and 6-vinylcoumarin-4-carboxylic acid (VCC)
have been designed and synthesised. These functional monomers allow for the preparation of
fluorescent sensors of chiral amines, an important class of pharmaceutical compounds. MIPs were
prepared with SCC and VCC, using (-)-ephedrine as a template and ethylene glycol dimethacrylate as a
cross-linker. In MeCN, the polymers exhibited a decrease of fluorescence in response to amines, with
some selectivity for the template over its enantiomer (+)-ephedrine and other structural analogues.
Interestingly the response of SCC to (-)-ephedrine in the MIP occurs in the opposite direction to the
change when recognition occurs in solution. The control polymers (NIPs) exhibited a lesser response to
(-)-ephedrine, and no resolving power, suggesting that imprinting has been successful and selective
recognition sites exist in the MIPs. Recognition in aqueous buffers at different pHs has also been
investigated.
demonstrated in comparison to structurally related compounds,
Introduction
not towards enantiomers. Indeed, the only true evidence for
selectivity in MIPs arising from specific recognition sites is
chiral selectivity because, otherwise, differences in the response
of fluorescent MIPs may be due to various factors such as
different quenching abilities or acidity/basicity of the analyte and
competitors.
Molecular imprinting has been demonstrated over the last three
decades as a versatile technique for the preparation of molecular
receptors capable of the selective recognition of given target
molecules. The approach is based on the self-assembly of a
template molecule with polymerisable monomers possessing func-
tional group(s) interacting with the template.1 After polymeri-
sation, the template is removed, leaving vacant recognition sites
which are complementary in shape and functional groups to the
original template. Molecularly imprinted polymers (MIPs) provide
an exciting alternative to biological receptors as recognition
elements in chemical sensors.2 Sensors based on electrochemical,3
mass-sensitive,4 optical5 and surface plasmon resonance spectro-
scopic (SPR)6 transduction combined with MIPs as the recogni-
tion element have all been fabricated.
Fluorescence is a promising detection method for MIP sen-
sor systems due to its high sensitivity and non-destructive
nature.7,8 The combination of fluorescence and molecular im-
printing techniques for the sensing of analytes has been based
either on the use of fluorescent9/fluorescent-labeled analytes10
or, in case analytes are non-fluorescent, on the formation of
fluorescent complexes between added reagents and functional
groups on the polymers. Several examples of MIPs containing
fluorophores and exhibiting fluorescence modulation on analyte
binding have been reported for analytes including cAMP,8 D-
fructose,11cyclododecylidene pyridine-2-carboxamidrazone,12 and
cyclobarbital13 among others.14 However, selectivity was mostly
Therefore, this work aimed to develop fluorescent sensors
capable of demonstrating enantioselectivity. Novel fluorescent
functional monomers suitable for the preparation of amine
selective MIPs were designed and synthesised. The fluorescent
monomers should interact as strongly as possible with the
templates, ideally on a stoichiometric basis, since any randomly
arranged fluorophores outside recognition sites in the resulting
MIPs may reduce the sensitivity and/or interact with competing
ligands, resulting in the MIPs showing false positive responses.
(-)-Ephedrine was used as the model template for this study
because bulk MIPs based on methacrylic acid (MAA), imprinted
with (-)-ephedrine, have been shown to exhibit very good recog-
nition properties in the HPLC mode.15,16 A bridged hydrogen
bond interaction of the carboxylic acid with both the amine
lone pair and the hydroxyl hydrogen of ephedrine, in the organic
solvent used to prepare the MIP, was postulated. Like many
other chiral amines, (-)-ephedrine is an important drug and it
offers the opportunity for chiral selectivity to be demonstrated by
comparing the sensor response towards (-)-ephedrine with that for
the opposite enantiomer, (+)-ephedrine. The imprinting strategy
is illustrated in Fig. 1.
School of Chemistry, University of Leeds, Leeds, UK LS2 9JT. E-mail:
chmrja@leeds.ac.uk; Fax: +44 (0)113 343 6565; Tel: +44 (0)113 343 6415
Results and discussion
† Electronic supplementary information (ESI) available: Details of the
application of the MapleTM model for producing association constants.
See DOI: 10.1039/b816733h
‡ Current address: School of Engineering and Mathematical Sciences, City
University, London, EC1V 0HB, UK
Synthesis of fluorescent monomers
To be applied successfully in molecular imprinting as a fluorescent
functional monomer, a fluorophore must include a polymerisable
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1211–1220 | 1211
©

View Article Online
Fig. 1 The preparation of an (-)-ephedrine-sensing MIP which exhibits
fluorescence changes upon template binding.
group, interact with the template covalently or non-covalently,
exhibit fluorescence changes upon binding, and have a rigid
structure, minimizing conformational flexibility around binding
sites and maintaining the specific orientation of the functional
group(s) originally positioned by the imprinting process.
Two novel polymerisable coumarins 3 and 5 suitable for
molecular imprinting were prepared in three steps starting from
4-bromophenol as outlined in Scheme 1. The reaction of phenols
with dimethyl acetylene dicarboxylate is a versatile approach for
the synthesis of coumarin-4-carboxylates. 1 was prepared similarly
to the method reported by Hekmatshoar et al.17 The synthesis of 2
was achieved from 1 in a good yield via a Suzuki coupling18,19 with
vinylphenylboronic acid using K2CO3 in dioxane as a base/solvent
mixture. 4 was prepared from 1 via a Stille reaction,18,20 following
the method described by Farina and Krishnan.21 The reaction
initially gave a mixture of starting material 1 and product 4 which
were inseparable by normal or reverse phase chromatography.
However, this problem was overcome by performing a Suzuki
reaction on the inseparable mixture with 3-amino-benzeneboronic
acid that gave 6 containing an amino group which has a high
affinity to silica gel and thus was easily removed from 4. The
synthesis of 3 and 5 was completed by hydrolyzing esters 2 and 4
under mild basic conditions.
Scheme 1 Preparat◦ion of fluorescent monomers. (a) C2(COOMe)2, Ph3P,
◦
=
CH2Cl2, -5 C to 55 C, 100 h, 52%; (b) CH2 CHC6H4B(OH)2, Pd(PPh3)4,
K2CO3, dioxane, 80 ◦C, 48 h, 42%: (c) 1 M NaOH, EtOH/THF (2.5:1),
◦
40 C, 12 h, 84%; (d) Bu3SnCH CH2, Ph3As, Pd2dba3, THF, 70 ◦C,
=
125 h, 38%; (e) 1 M NaOH, EtOH/THF (2.5:1), 40 ◦C, 14 h, 79%;
(f) NH2C6H4B(OH)2·H2O, Pd(PPh3)4, K2CO3, dioxane, 80 ◦C, 50 h.
Absorption and fluorescence behaviours of fluorophores
The absorption spectra of VCC (5) and SCC (3) in MeCN show
two main bands, at about 280 and 340–345 nm (Fig. 2). The shorter
wavelength bands with higher absorbances can be attributed to
the So→S2 transition whereas the longer-wavelength bands with
weaker absorbances may be attributed to the So→S1 transition.
These absorption patterns of VCC and SCC are fairly similar
to those of other 6-substituted coumarins as well as the parent
coumarins reported in the literature,22,23 whose p systems in the
excited states are less conjugated than e.g. 7-substituted ones.
Emission spectra for each compound recorded in the same solvent
using excitation at the absorbance maxima include only one band
in the 510–530 nm region (Fig. 2). Although both compounds
absorb more at 280 nm than at longer wavelengths, the emission
Fig. 2 Absorption and emission spectra of VCC (solid lines) and SCC
(dotted lines) (10-5 M) in MeCN. Emission spectra were recorded with
lex = 340 nm for VCC and 345 nm for SCC, using the same slit widths.
lex = 340 nm than it is with lex = 280 nm This may be because
not all molecules excited to the S2 state using lex = 280 nm fall to
the S1 state, from which fluorescence occurs, during the internal
conversion process, energy may be lost in other ways. Therefore,
excitation wavelengths of 340 nm and 345 nm were used in all
experiments for VCC and SCC, respectively.
of VCC is actually higher with lex = 345 nm than it is with lex
=
Interestingly, although the absorbances of the monomers at the
excitation wavelengths used are roughly the same, VCC is more
280 nm, and the fluorescence of SCC is only slightly lower with
1212 | Org. Biomol. Chem., 2009, 7, 1211–1220
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
fluorescent than SCC, and its emission maximum is observed at
a shorter wavelength. Coumarin molecules have close lying p,p*
and n,p* states which are easily perturbed by changes in solvents,
substituents and other factors,22,24 and hence the S1 state probably
has both p,p* and n,p* character. It is proposed that in VCC a p,p*
state may lie below the n,p* state and in SCC vice versa, as shown
in Fig. 3. This could account for the less intense fluorescence of
SCC, n←p* transitions being generally less intense than p←p*
transitions because they characteristically have longer lifetimes
which enhance intersystem crossing.25 The red-shifted fluorescence
of SCC with respect to VCC may be due to a greater degree of
charge separation in the excited state of SCC, characteristic of
n,p* states, which would therefore be more likely to be stabilised
by solvent prior to fluorescence.
Fig. 3 Possible energy diagrams for VCC and SCC. So: ground state,
S*: excited single state, E: energy, F: fluorescence.
Adding ephedrine to VCC or SCC in MeCN affected the
fluorescence of both compounds (Fig. 4). The fluorescence of
VCC was quenched and shifted to a shorter wavelength whilst
that of SCC increased and shifted to a shorter wavelength. The
data in Fig. 4 can be converted to a plot of emission at a single
wavelength, but it is non-trivial to fit this to an association
model because neither reagent is in excess so the data cannot
be linearised. Nonetheless, we have used the software package
MapleTM to fit the data manually to a 1:1 association (see ESI†).
This gives an extremely high association constant of ~106 M-1.
Even given the strong association previously noted for ephedrine
with MAA in MeCN16 this is clearly unrealistic. There are a
number of complicating factors, including self-association of the
monomer and self-quenching of the fluorophore which mean the
true association constant is likely to be less than this. A full series of
NMR titrations would be required to arrive at an accurate value.
Similar although less pronounced changes were observed when
acetic acid was added. This behaviour must be due to interactions
of the added amine (or carboxylic acid) with the carboxylic acid
group on VCC and SCC, as shown in Fig. 5, which was confirmed
when the corresponding esters (4 and 2) of VCC and SCC did not
respond to the same addition (data not shown). The large change
in fluorescence intensity observed for VCC and SCC upon adding
ephedrine, which approximately saturates after the addition of
just one equivalent of ephedrine, suggests that the interactions
with ephedrine are very strong and roughly on a 1:1 basis. The
shifts to lower lem suggest a destabilisation of the excited state: for
VCC it may gain more n,p* character (causing quenching) and for
SCC more p,p* character (causing the increase in emission).
Fig. 4 Fluorescence spectra of a) VCC (50 mM, lex = 340 mm) and b)
SCC (50 mM, lex = 345 nm) in MeCN with the addition of (-)-ephedrine.
I/I0max is the emission intensity in cps divided by the maximum intensity
for the fluorophore alone, in the absence of added ephedrine. Different slit
widths were used for a) and b) to optimise the signal.
Fig. 5 Hydrogen bonding between a) fluorophore–ephedrine, b) fluo-
rophore–carboxylic acid or fluorophore–fluorophore.
Fluorescence studies of polymers in MeCN
Imprinted polymers and their controls were prepared in
MeCN. These were prepared as monolithic MIPs and ground
into small particles, since these are easier to produce than
spherical beads whichcan be made by precipitation polymerisation
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1211–1220 | 1213
©

View Article Online
and since a minimal amount of solvent is required. To make a
dispersion polymer, excess solvent is required and the equilibrium
will disfavour complex formation, hence dispersion MIPs usually
exhibit a lower density of good binding sites. In previous studies
using fluorescence26 ground monolithic MIP particles have been
used successfully. There is no problem with scattering from the
irregular-shaped particles because with the coumarin used here
the Stokes shift is sufficient that emission can be measured
well above the scattered wavelengths. The molar amount of
fluorescent monomers used was fixed at 1:40 of the cross-
linker for all polymers. Additional ‘inert’ monomer, MMA,
was included so that the MIP should not be too rigid (MIPs
with >90% cross-linker tend to show impaired performance).27
More fluorescent monomers would be expected to give more
recognition sites but too high a concentration of fluorophore
could also result in fluorescence quenching by the inner filter
effect. A template:fluorescent monomer ratio of 1:1 or higher
was employed to avoid ‘false positives’ caused by randomly
incorporated fluorophores.
Fig. 6 Emission spectra* in MeCN of PV-1:1 (lower continuous line),
CV-1:1 (upper continuous line), PS-1:1 (lower broken line), CS-1:1 (upper
broken line), PV-1:1-MAA (upper dotted line) and CV-1:1-MAA (lower
dotted line) with excitation at 340 nm for PVs, CVs and at 345 nm for
PS, CS (same slit widths used in all cases). Polymer concentrations are as
discussed. * The discontinuities on all spectra at 450 nm were due to an
artefact of the instrument.
It was also interesting to see whether the inclusion of MAA
would affect the recognition properties of the MIPs. MAA
can increase the hydrophilicity of the polymers and hence may
improve recognition in aqueous buffers. In addition, it may yield
higher-order complexes in the prepolymerisation mixture which
are expected to give stronger and more selective binding sites,15
but should not out-compete the fluorescent monomers for the
amine template since its pKa is higher than that for coumarin-
4-carboxylic acid (4.58 compared to 2.5228). The PV and PS
polymers differed in the type of fluorescent monomer used (i.e.
either VCC or SCC). Different PV polymers differed in the amount
of template and the amount as well as the type of co-monomer
(MAA or MMA). Control polymers were prepared under identical
conditions, without the addition of the template (-)-ephedrine.
The polymer concentrations used in the fluorescence experiments
were fixed at 0.443 mg mL-1 for PV/CV-1:1 and PV/CV-1:8,
0.437 mg mL-1 for PV/CV-1:1-MAA and PV/CV-1:8-MAA and
0.446 mg mL-1 for PS/CS-1:1. These amounts of polymers were
calculated to give the same fluorophore concentration of 50 mM,
assuming that the polymerisation yield was the same for all
polymers and equal to 100% and the template was completely
removed from the MIPs.
Fig. 6 shows emission spectra of different polymers in MeCN. It
can be seen that both fluorophores emit at lower wavelengths
in the polymers than in their free forms, though PS-1:1 and
CS-1:1 emit at higher lem than PV-1:1 and CV-1:1, consistent
with the higher lem for SCC than VCC. The blue shift could
be because when fixed in a rigid polymer network the excited
states of both fluorophores undergo less stabilisation from solvent
rearrangement. However, PV-1:1 and CV-1:1 exhibit less intense
fluorescence than PS-1:1 and CS-1:1, despite VCC being more
fluorescent than SCC. Possibly the p,p* state of VCC is destabilised
due to the vinyl group undergoing addition, causing loss of
conjugation with the coumarin ring. The higher fluorescence
yields observed for CV-1:1 and CS-1:1 compared to PV-1:1 and
PS-1:1 are most likely because of residual ephedrine left in the
imprinted polymers (ephedrine causes quenching as shown below).
The fluorescence yield of PV-1:1-MAA may be lower than that
of the corresponding NIP because the fluorophore in the NIP
is self-associated or associated with MAA, and the interaction
with a carboxylic acid group decreases the fluorescence of VCC as
discussed previously.
Initially, the fluorescence responses of all polymers to
(-)-ephedrine in comparison to the opposite enantiomer,
(+)-ephedrine, were studied in MeCN, the same solvent used for
MIP preparation, since re-binding has often been demonstrated
to be most selective in the same solvent as that in which the MIP
was prepared.29 Each measurement was performed in duplicate
or triplicate and data presented are average values. Although the
low density of MeCN (0.782) might present a problem as it allows
polymer particles to settle, by shaking the cuvette vigorously before
each measurement the problem can be overcome. No significant
settling out was observed on the timescale of the measurements
and the results were found to be reproducible.
The addition of (-/+)-ephedrine resulted in a decrease in
fluorescence of all polymers. However, the degree of quenching
was different for each polymer and different for the (-)-enantiomer
and (+)-enantiomer in some cases (Fig. 7). Interestingly, SCC
polymers responded to ephedrine in the opposite direction to the
free fluorophore. This might be because an inversion of the lowest-
lying n,p* and p,p* states of the SCC molecule occurred when
SCC was incorporated in the polymers, resulting in the lowest
excited state of SCC being more p,p* in character and hence the
fluorophore behaving more similarly to VCC. The data in Fig. 7
might be interpreted to show that the association of ephedrine
to the polymers is weaker than to the monomer itself (shown in
Fig. 4). Application of the MapleTM model for association to these
data (ESI†) yields apparent association constants for PV and PS
of ~105 M-1. The poor fits demonstrate that the situation is more
complex than a simple 1:1 model with fluorescence dependent only
on the number of empty and number of occupied sites.
To demonstrate imprinting effects, it is desirable that a difference
between enantiomers or at least a difference between MIPs and
NIPs is achieved. All MIPs, except PV-1:1-MAA, showed a greater
decrease in fluorescence upon ephedrine addition than did the
corresponding controls, suggesting that the analyte bound to the
MIPs more strongly than to the NIPs. Moreover, PV-1:1 and
PS-1:1 exhibit a small discrimination between (-)-ephedrine and
1214 | Org. Biomol. Chem., 2009, 7, 1211–1220
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Fig. 7 Fluorescence responses of a) PV-1:1 and CV-1:1, b) PS-1:1 and CS-1:1, c) PV-1:8 and CV-1:8, d) PV-1:1-MAA and CV-1:1-MAA, e) PV-1:8-MAA
and CV-1:8-MAA to (-/+)-ephedrine in MeCN with excitation and polymer concentrations as in Fig. 6. MIP + (+)-ephedrine grey; MIP + (-)-ephedrine
black; NIP + (+)-ephedrine grey hatched; NIP + (-)-ephedrine black hatched. I/I0MIP is the emission intensity in cps divided by the intensity for the
imprinted polymer alone, in the absence of added ephedrine. Data recorded at 454 nm except b) recorded at 505 nm. Different slit widths were used for
a), b), c), d) and e) to optimise the signal. Error bars represent standard deviations in the calculated values.
(+)-ephedrine while the respective control polymers CV-1:1 and
CS-1:1 show no resolving power, indicating that the MIPs contain
chiral recognition sites. At high concentrations of ephedrine, no
chiral selectivity and no further fluorescence quenching are seen
due to the saturation of all available binding sites.
A slight selectivity for (-)-ephedrine over (+)-ephedrine is also
observed for PV-1:8 at the low analyte concentration of 25 mM
(Fig. 7c). PV-1:8 was made with a large excess of template, such
that residual template in the MIP may account for its much
lower initial fluorescence than CV-1:8 and its limited response
to added analyte. PV-1:8-MAA responds more strongly than
CV-1:8-MAA, but chiral selectivity is not apparent (Fig. 7d).
Similar results are obtained with PV-1:1-MAA and CV-1:1-MAA
(Fig. 7e), suggesting MAA does little to enhance the selectivity
of the fluorophore-based binding sites, whether the ratio of
template:fluorophore:MAA is 1:1:7 or 8:1:7.
The highest enantioselectivity, the greatest difference in fluo-
rescence response between MIP and NIP and also the biggest
decrease of fluorescence upon the addition of ephedrine (up to 59%
at 1000 mM (-)-ephe) were observed for PV-1:1. The superiority of
PV-1:1 over PV-1:8 may indicate that the interaction between the
fluorescence monomer and ephedrine is very strong so that a 1:1 ra-
tio of template:monomer is sufficient to complex all monomer and
none is left non-specifically incorporated throughout the polymer.
The lower response of PS-1:1 and CS-1:1 to ephedrine compared
to PV-1:1 and CV-1:1 may be due simply to the photophysical
properties of the fluorophore as discussed previously and does not
necessarily suggest that binding of ephedrine to these polymers is
weaker.
Since PV-1:1 and PS-1:1 were the most promising MIPs,
their selectivity towards different ephedrine-related amines was
further investigated (Fig. 8). The responses of PV-1:1, PS-1:1 and
the corresponding NIPs to the different amines were recorded
at 25 mM and 50 mM where the greatest difference between
(-)-ephedrine and (+)-ephedrine had been observed (Fig. 9). It can
be seen that both MIPs respond more to all amines investigated
than do the NIPs, suggesting that there are more binding sites
available in the MIPs than in the NIPs (since some of the fluo-
rophore is self-associated during preparation of the NIPs). More-
over, the MIPs respond less to any of these compounds than to
the template, (-)-ephedrine whereas the NIPs respond similarly
to (-)-ephedrine, (+)-ephedrine and (+)-pseudoephedrine and less
to the other compounds. This once again indicates successful
imprinting and selective recognition sites in the MIPs.
The fluorescence responses of the NIPs to different amines can
be explained in terms of pKa values. Since there are no specific
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1211–1220 | 1215
©

View Article Online
Fig. 8 Structures and pKa values of different amines used for fluorescence studies.
binding sites in the NIPs, they respond to compounds with higher
pKa values more strongly than to less basic ones. The basicities
of (-)-ephedrine, (+)-ephedrine and (-)-pseudoephedrine are
similar, so these compounds bound similarly to the NIPs and to
a greater degree than (-/+)-norephedrine and (-)-HEBA whose
pKa values are lower. This trend was also seen partly with the
MIPs. However, due to the presence of binding sites selective for
the template, the MIPs bound most strongly to (-)-ephedrine
even though the pKa for pseudoephedrine is slightly higher than
that for (-)- ephedrine (9.74 compared with 9.5628).
Fluorescence studies of polymers in aqueous buffer
It was also interesting to see if selective responses of the MIPs could
be observed in buffered aqueous solutions, resembling the condi-
tions in which biological recognition occurs. The fluorescence of
PV-1:1, PS-1:1 and the corresponding controls was investigated
over a range of pH. 25 mM citrate buffer at various pHs ranging
from 2 to 6 was employed. A small amount of surfactant Triton
X-100 (0.5% (w/v)) was added to reduce non-specific binding
but also to enable the dispersion of the polymers. As preliminary
experiments suggested that polymers showed little response to
ephedrine in aqueous buffer, a high analyte concentration of
1000 mM was used. Fig. 10 shows the response of PV-1:1, CV-1:1
and PS-1:1, CS-1:1 to 1000 mM (+/-)-ephedrine at different pHs.
It can be seen that the polymers show a decrease in fluorescence
with increasing pH, and the decrease is greater for the MIPs than
for the NIPs. A decrease is also observed on adding ephedrine,
which is more apparent at pH 5 and 6, and slightly bigger for the
MIPs. However, selectivity for (-)-ephedrine over (+)-ephedrine is
not apparent. The results suggest firstly that the protonated form
of the fluorophore is most fluorescent, and its pKa in the polymer is
higher than the value in free solution of 2.52.28 Since ephedrine in
aqueous solution is protonated below pH 9.3828 it interacts better
with the polymer at pH 5–6 than at pH 2–3. The interaction is
an ion-exchange process with ephedrine replacing the protons and
buffer cations (sodium)—such that the number of sites occupied
only by protons, and hence the fluorescence, falls.
Fig. 9 Fluorescence responses of a) PV-1:1, b) CV-1:1, c) PS-1:1,
d) CS-1:1 to different amines in MeCN. Polymer alone white; poly-
mer + (-)-ephedrine black; polymer + (+)-ephedrine grey; polymer
+ (-)-norephedrine black hatched; polymer + (+)-norephedrine grey
hatched; polymer + (+)-pseudoephedrine dotted; polymer + (-)-HEBA
horizontal hatched. I/I0MIP is the emission intensity in cps divided by
the intensity for the MIP alone, in the absence of added amine. Same slit
widths used for a) and b) (with emission measured at 454 nm) and for c)
and d) (at 504 nm).
1216 | Org. Biomol. Chem., 2009, 7, 1211–1220
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
MAA or MMA was used as a co-monomer. The polymers differed
in the type of fluorescent monomer used (i.e. either VCC or SCC),
the amount of template and the amount as well as the type of co-
monomer (MAA or MMA). All polymers showed a quenching of
fluorescence with the addition of ephedrine. It was found that the
MIPs prepared with a 1:1 ratio of template:fluorescent functional
monomer and without the addition of MAA demonstrated the
best recognition properties. These MIPs exhibited a decrease
of fluorescence in response to amines, with some selectivity for
the template over its enantiomer, (+)-ephedrine. The control
polymers (NIPs) exhibited a lesser response to (-)-ephedrine,
and no resolving power, suggesting that imprinting has been
successful and selective recognition sites exist in the MIPs. At
high concentrations of ephedrine, no chiral selectivity was seen
with the MIPs due to the saturation of binding sites. Selectivity to
structural analogues was also demonstrated in MeCN. However,
little response to ephedrine was observed in aqueous buffer.
The fluorescent monomers developed in this study are poten-
tially suited for use in the preparation of fluorescent sensors for
other templates containing amine functionalities.
Experimental section
Materials
Ethylene glycol dimethacrylate (EDMA), methacrylic acid
(MAA), methylmethacrylate (MMA), (1R,2S)-(-)-ephedrine,
(1S,2R)-(+)-ephedrine hemihydrate, (+/-)-norephedrine, (+)-
pseudoephedrine, 4-bromophenol, tetrakis-(triphenylphosphine)
palladium(0), 4-vinylphenylboronic acid, dimethyl acetylenedi-
carboxylate, triphenylarsine, triphenylphosphine, 4-dimethyl-
aminopyridine were from Aldrich (Dorset, UK). Tris(dibenzy-
lideneacetone)dipalladium (0) was from Strem Chemicals.
3-aminobenzenboronic acid monohydrate was from Avocado.
6-Bromocoumarin-3-carboxylic acid and tri-n-butyl(vinyl)tin
were from Alfa Aesar. 2,2¢-Azobis-(2-methylpropionitrile)
(AIBN), (S)-(-)-N-(2-hydroxyethyl)-methylbenzylamine (HEBA)
were from Acros Organics (Geel, Belgium). All solvents used were
of HPLC grade. Monomers were distilled under reduced pressure
prior to use to remove inhibitors. Ephedrine isomers were dried
under vacuum over phosphorous pentoxide for 24 h prior to use.
The optical purity of each isomer was checked by polarimetry
(t = 25 ◦C): (-)-ephedrine [a]D = -43.47◦ (c = 0.5, 1M HCl);
(+)-ephedrine [a]D = -43.30◦ (c= 0.5, 1 M HCl).
Fig. 10 Fluorescence responses of a) PV-1:1, b) CV-1:1, c) PS-1:1, d)
CS-1:1 to 1000 mM (+/-)-ephedrine at different pHs (25mM citrate +
0.5% Triton X-100). Polymer alone white; polymer + (-)-ephedrine black;
polymer + (+)-ephedrine grey. I/I0MIP, pH2 is the emission intensity in
cps divided by the intensity for the MIP alone, in the absence of added
amine, at pH2. Same slit widths used for a) and b) (with emission measured
at 454 nm) and for c) and d) (at 504 nm).
Conclusions
Novel coumarin-based fluorescent functional monomers contain-
ing a carboxylic acid functionality, 6-vinyl-coumarin-4-carboxylic
acid (VCC) and 6-vinylphenyl-coumarin-4-carboxylic acid (SCC),
have been synthesised. These functional monomers allow for the
preparation of fluorescent sensors of amines. Their fluorescence
behaviours have been investigated in MeCN. Both monomers
exhibited a significant degree of fluorescence intensity change, but
in opposite directions, upon ephedrine addition. VCC decreased
fluorescence whereas SCC showed a fluorescence enhancement
in response to the binding event. The monomers were found
to respond also to carboxylic acids in the same manner as to
ephedrine but to a lesser extent. The photophysical properties of
the fluorophores have been postulated to explain these phenom-
ena. Although photobleaching has not been studied, coumarins
are frequently used as laser dyes for single-molecule spectroscopy
so are not expected to suffer from excessive photobleaching under
the conditions used here.
Instrumentation
1H and 13C NMR spectra were recorded on Bruker DPX300
and Bruker Avance 500 spectrometers. Mass spectra were run
by positive ion Electrospray (ES) mode on a LCT Micromass
(TOF) or a Bruker Daltonics (MicroTOF) mass spectrometers.
IR spectra were recorded on a PerkinElmer Spectrum One
FT-IR spectrophotometer and were run neat. Optical rotations
were recorded on an AA-1100 automatic polarimeter manufac-
tured by Optical Activity Ltd. Melting points were recorded on a
Reichert melting point apparatus and were uncorrected. Elemental
analyses were carried out at the Microanalytical Laboratory,
Department of Chemistry, University of Leeds. UV measurements
were carried out on a Perkin-Elmer Lambda 900 UV-Vis-nIR
Different polymers were prepared with VCC and SCC using
(-)-ephedrine as a template and EDMA as a cross-linker. Either
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1211–1220 | 1217
©

View Article Online
=
=
spectrophotometer. Fluorescence spectra were measured on a
Jobin-Yvon-Horiba Fluorolog-3 Model FL3–21 spectrofluorom-
eter system with DataMax for WindowsTM software.
(CC6), 138.0 (C6), 137.6 (CCH CH2), 136.6 (CH CH2), 131.5
(C7), 127.7 (aromatic C.), 127.3 (aromatic C.), 125.4 (C5), 120.3
=
(C3), 118.0 (C8), 116.5 (C10), 114.8 (CH CH2), 53.6 (OCH3); MS
(ES+): Calcd. m/z = 307.0970 (C19H15O4). Found m/z = 307.0962
(M + H+); Elem. Anal. Calcd. for C19H14O4 (306.10): C 74.55, H
4.61. Found: C 74.50, H 4.80.
Synthesis of fluorophores
Methyl 6-bromocoumarin-4-carboxylate (1). 1 was prepared
similarly to the method reported by Hekmatshoar et al.17 (the
compound was not properly purified or fully characterised by
those authors). A solution of 4-bromophenol (3.46 g, 20 mmol)
and triphenylphosphine (5.25 g, 20 mmol) in dichloromethane
6-Vinylphenylcoumarin-4-carboxylic acid (3).
2 (1.06 g,
3.44 mmol) was dissolved in THF (22 mL)-EtOH (56 mL), and
1M aqueous NaOH (14 mL) was added. The reaction mixture
was heated to 40 ◦C for 12 h. The organic solvents were removed
in vacuo and water (50 mL) was added to the remaining aqueous
solution. The solution was then washed with CH2Cl2 (100 mL),
ethylacetate (100 mL), filtered to remove any insoluble material
and acidified with concentrated aqueous HCl. The mixture was
extracted with ethyl acetate. The organic extracts were washed with
H2O and saturated aqueous NaCl, dried over MgSO4, filtered, and
concentrated in vacuo. The obtained crude solid was recrystallised
from ethyl acetate to give 3 (844 mg, 84%) as an yellow solid,
◦
(80 mL) was stirred at -5 C in an ice–salt mixture under argon
and treated dropwise with a solution of dimethyl acetylenedi-
carboxylate (2.46 mL, 20 mmol) in CH2Cl2 (40 mL). After 1 h,
the reaction mixture was refluxed for 100 h. The solvent was
then removed under reduced pressure to give a red–brown solid
which was recrystallised from ethanol to give a crude mixture
of the desired product and triphenylphosphine. This solid was
purified by flash chromatography on silica gel using CH2Cl2-petrol
(95:5, v/v) as eluent to afford the coumarin ester 1 (2.9 g, 52%) as
an off-white solid that was used in the next step. A sample of 1 was
further purified by recrystallisation from ethanol to give 1 as fine
needles, mp 117–119 ◦C [lit.17 mp 129–131 ◦C (ethanol)]; IR (neat)
mp 211–212 ◦C; IR (neat) nmax (cm-1) 3814, 3150–2700 (COOH),
1791, 1732(C O), 1522, 1428, 1364, 1215; H-NMR (500 MHz,
1
=
DMSO-d6) d(ppm): 8.47 (d, 1H, H5, J5,7 = 2.2 Hz), 7.98 (dd,
1H, H7, J7,5 = 2.2 Hz, J7,8 = 8.6 Hz), 7.67 (d, 2H, aromatic H
J = 8.3 Hz), 7.62 (d, 2H, aromatic H J = 8.3 Hz), 7.56 (d, 1H,
-1
=
n
max (cm ) 3439, 3075, 2906, 2772, 2330, 1740–1732 (C O), 1598,
1
=
H8, J8,7 = 8.6 Hz), 6.93 (s, 1H, H3), 6.81 (dd, 1H, CH CH2, J =
1554, 1426, 1264, 1182; H-NMR (300 MHz, CDCl3) d(ppm):
8.49 (d, 1H, H5, J5,7 = 2.3 Hz), 7.67 (dd, 1H, H7, J7,5 = 2.3 Hz,
J7,8 = 8.8 Hz), 7.25 (d, 1H, H8 J8,7 = 8.8 Hz), 7.01 (s, 1H, H3),
4.02 (s, 3H, CH3); 13C-NMR (CDCl3) d(ppm): 164.1 (COOMe),
159.7 (C2), 153.5 (C9), 141.3 (C4), 135.7 (C7), 129.9 (C5), 121.1
(C3), 120.9 (C8), 119.2 (C6), 118.2 (C10), 53.8 (OCH3); MS (ES+):
Calcd. m/z = 282.9606 (C11H8O479Br), 284.9585 (C11H8O481Br).
Found m/z = 282.9595, 284.9581 (M + H+); Elem. Anal. Calcd.
for C11H7O4Br (283.075): C 46.67, H 2.49, Br 28.23. Found: C
46.70, H 2.65, Br 28.05.
=
11.0 Hz, J = 17.7 Hz), 5.92 (d, 1H, CH CHaHb J = 17.7 Hz), 5.33
(d, 1H, CH CHaHb, J = 11.0 Hz); 13C-NMR (DMSO-d6) d(ppm):
=
165.6 (COOH), 159.8 (C2), 153.6 (C9), 144.0 (C4), 138.6 (CC6),
=
=
137.0 (CCH CH2), 136.5 (C6), 136.4 (CH CH2), 131.1 (C7), 127.3
(aromatic C.), 124.8 (C5), 119.0 (C3), 117.8 (C8), 116.5 (C10),
+
=
115.1 (CH CH2); MS (ES ): Calcd. m/z = 293.0814 (C18H13O4).
Found m/z = 293.0821 (M + H+); Elem. Anal. Calcd. for C18H12O4
(292.1): C 74.01, H 4.15. Found: C 74.25, H 4.20.
Methyl 6-vinylcoumarin-4-carboxylate (4).
1
(1.86 g,
Methyl 6-vinylphenylcoumarin-4-carboxylate (2). A mixture of
1 (3.11 g, 11 mmol), 4-vinylphenylboronic acid (2.44 g, 16.5 mmol,
1.5 mol equiv.), potassium carbonate (5.7 g, 41.25 mmol, 3.75 mol
equiv.) and dioxane (80 mL) was stirred at room temperature
under argon for 0.5 h. Tetrakis(triphenylphosphine)palladium(0)
(636 mg, 0.55 mmol, 5 mol%) was added. The reaction was heated
to 80 ◦C and left at reflux in the dark for 48 h. After cooling to room
temperature, dioxane was removed and the resulting residue was
redissolved in CH2Cl2 (100 mL), washed with 1M aqueous HCl
(2 ¥ 100 mL) and saturated aqueous NaCl (100 mL). The organic
phase was dried over MgSO4, filtered, concentrated in vacuo to
give the crude product as an orange semicrystalline oil. The crude
product was purified by flash chromatography on silica gel with
CH2Cl2-EtOAc (95:5, v/v) as eluent to give 2 as a yellow solid
which was further purified by recrystallisation from ethanol to
6.58 mmol) was dissolved in dry THF (40 mL) and the
solution was treated with triphenylarsine (403 mg, 0.2 equiv.)
and tris(dibenzylideneacetone)dipalladium(0) (Pd2dba3) (151 mg,
0.05 equiv. of Pd) under argon. After stirring at r.t for 10 min,
tributylvinyltin (2.4 mL, 8.24 mmol) was added. The solution was
◦
heated to 70 C and left under reflux in the dark for 125 h. The
reaction was allowed to cool to r.t. Saturated aqueous sodium
fluoride (20 mL) and H2O (10 mL) were added and left to stir
for 1 h. The mixture was filtered and extracted with CH2Cl2. The
organic extracts were washed 2 times with saturated aqueous
NaCl (50 mL each), dried over MgSO4, filtered, and concentrated
in vacuo. The resulting dark brown residue was chromatographed
on silica gel using CH2Cl2-EtOAc (95:5, v/v) as eluent to give a
yellow solid (951 mg). 1H NMR analysis of this solid showed that
it was a mixture of 4 and the starting material 1, in a 4:1 ratio.
The above mixture (951 mg, containing 0.79 mmol of 1)
was dissolved in dioxane (14 mL). 3-aminobenzeneboronic acid
monohydrate (183 mg, 1.185 mmol, 1.5 equiv.) and K2CO3
(409 mg, 2.96 mmol, 3.75 equiv.) were added and the mix-
ture was stirred at room temperature under argon for 0.5 h.
Tetrakis(triphenylphosphine)palladium(0) (Pd(PPh3)4) (636 mg,
0.55 mmol, 5 mol%) was added. The reaction was heated to 80 ◦C
under argon and left at reflux in the dark for 50 h. After cooling to
room temperature, dioxane was removed and the resulting residue
redissolved in CH2Cl2 (25 mL), washed with 1M aqueous HCl
afford 2 (1.4 g, 42%) as yellow fine crystals, mp 91–92 ◦C; IR (neat)
-1
=
n
max (cm ) 3435, 3073, 2905, 2455, 2312, 2108, 1916, 1742 (C O),
1
1614, 1567, 1429, 1360, 1303, 1244, 1193; H-NMR (300 MHz,
CDCl3) d(ppm): 8.51 (d, 1H, H5, J5,7 = 2.2 Hz), 7.80 (dd, 1H, H7,
J7,5 = 2.2 Hz, J7,8 = 8.6 Hz), 7.57 (d, 2H, aromatic H. J = 8.4 Hz),
7.51 (d, 2H, aromatic H J = 8.4 Hz), 7.43 (d, 1H, H8, J8,7
=
=
8.6 Hz), 6.99 (s, 1H, H3), 6.77 (dd, 1H, CH CH2, J = 10.9 Hz,
=
J = 17.6 Hz), 5.81 (d, 1H, CH CHaHb J = 17.6 Hz), 5.31 (d, 1H,
CH CHaHb,J = 10.9 Hz), 4.02 (s, 3H, OCH3); 13C-NMR (CDCl3)
=
d(ppm): 164.6 (COOMe), 160.3 (C2), 154.0 (C9), 142.6 (C4), 139.2
1218 | Org. Biomol. Chem., 2009, 7, 1211–1220
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Composition of (-)-ephedrine MIPs and corresponding NIPs
Polymer
(-)-ephedrine/mmol
Fluorescent monomer type/mmol
EDMA/mmol
MMA/mmol
MAA/mmol
MeCN/mL
PV-1:1
CV-1:1
PV-1:1-MAA
CV-1:1-MAA
PV-1:8
0.2
–
0.2
–
1.6
–
1.6
–
VCC
VCC
VCC
VCC
VCC
VCC
VCC
VCC
SCC
SCC
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
1.4
1.4
–
–
–
1.4
1.4
–
2.21
2.21
2.17
2.17
2.21
2.21
2.17
2.17
2.21
2.21
–
1.4
1.4
–
–
1.4
1.4
CV-1:8
–
PV-1:8-MAA
CV-1:8-MAA
PS-1:1
1.4
1.4
–
0.2
–
CS-1:1
–
Note that CV-1:1 is identical to CV-1:8 and CV-1:1-MAA is identical to CV-1:8-MAA but were prepared on different days.
(3 ¥ 25 mL) and saturated aqueous NaCl (2¥ 25 mL). The organic
phase was dried over MgSO4, filtered and concentrated in vacuo
to give the crude product as an orange semicrystalline oil. The
crude product was purified by flash chromatography on silica gel
with CH2Cl2-EtOAc (95:5, v/v) as eluent to give 4 as a yellow
solid which was further purified by recrystallisation from ethanol
to afford 4 (574 mg, 38%) as yellow crystals (fine needles), mp 106–
(ES+): Calcd. m/z = 217.0501(C12H9O4). Found m/z = 217.0505
(M + H+); Elem. Anal. Calcd. for C12H8O4 (216.05): C 66.71, H
3.73. Found: C 66.95, H 3.88
Preparation of polymers
The compositions of the different polymers are given in
Table 1. For the imprinted polymers PS and PVs (except PV-1:8-
MAA which was prepared by a slightly different procedure) §,
(-)-ephedrine, coumarin, MMA and MAA were weighed into
borosilicate glass vials and dissolved in MeCN. EDMA was added
followed by azobis(isobutyronitrile) (AIBN). The vials were placed
in a sonicating water bath until the AIBN was fully dissolved, then
purged thoroughly with argon for about 2 min before being tightly
capped and sealed. Polymerisation was carried out at 60 ◦C in the
dark for approximately 17 h and at 80 ◦C for a further 2 h. The hard
bulk polymers were then hand ground with a mortar and pestle
until fine particles (< 50 mm by optical microscope) were obtained.
The polymer particles were washed to remove the template (where
applicable) by repeated incubation in MeOH-AcOH (8:2, v/v)
(45 mL solvent each), centrifugation and re-suspension (11 ¥ 2 h
incubations), followed by the same procedure with MeOH alone
(4 ¥ 2 h incubations) and finally on a sintered filter with MeOH
(250 mL). The filtrate was taken for UV measurement and it was
confirmed that no species was detected from the recovered solution
with a UV spectrophotometer. After washing, polymer particles
were dried under vacuum for at least 24 h before doing fluorescence
measurements.
108 ◦C; IR (neat) nmax (cm-1) 3437, 3074, 2963, 2885, 2326, 1916,
1
=
1747(C O), 1629, 1569, 1428, 1400, 1378, 1274, 1249, 1197; H-
NMR (500 MHz, CDCl3) d(ppm): 8.27 (d, 1H, H5, J5,7 = 2.0 Hz),
7.66 (dd, 1H, H7, J7,5 = 2.0 Hz, J7,8 = 8.6 Hz), 7.33 (d, 1H,
=
H8, J8,7 = 8.6 Hz), 6.96 (s, 1H, H3), 6.76 (dd, 1H, CH CH2, J =
=
10.9 Hz, J = 17.6 Hz), 5.79 (d, 1H, CH CHaHb, J = 17.6 Hz), 5.34
(d, 1H, CH CHaHb, J = 10.9 Hz), 4.02 (s, 3H, OCH3); 13C-NMR
=
(CDCl3) d(ppm): 164.6 (COOMe), 160.3 (C2), 154.1 (C9), 142.5
=
(C4), 135.7 (CH CH2), 134.9 (C6), 130.2(C7), 125.1 (C5), 120.1
=
(C3), 117.8 (C8), 116.2 (C10), 115.6 (CH CH2), 53.6 (OCH3); MS
(ES+): Calcd. m/z = 231.0657 (C13H11O4). Found m/z = 231.0660
(M + H+); Elem. Anal. Calcd. for C13H10O4 (230.07): C 67.86, H
4.38. Found: C 67.75, H 4.55.
6-vinylcoumarin-4-carboxylic acid (5). 5 was prepared in a
manner similar to the preparation of 3. 4 (250 mg, 1.086 mmol) was
dissolved in THF (7 mL)-EtOH (18 mL), and 1M aqueous NaOH
(4 mL) was added. The reaction mixture was heated to 40 ◦C for
14 h under argon in the dark. The organic solvents were removed
in vacuo and water (25 mL) was added to the remaining aqueous
solution. The solution was then washed with CH2Cl2 (2 ¥ 20 mL)
and ethyl acetate (2 ¥ 20 mL), filtered to remove any insoluble
material and acidified with concentrated HCl. The mixture was
extracted with ethyl acetate. The organic extracts were washed
with H2O and saturated NaCl, dried over MgSO4, filtered, and
concentrated in vacuo to give fairly pure 5 as a yellow solid which
was further purified by recrystallisation from chloroform to give
Control polymers CS and CVs were prepared under identical
conditions, with the same composition as PS and PVs, respectively
but without the addition of the template (-)-ephedrine.
Fluorescence measurements in MeCN
Emission spectra of monomers and polymers were recorded using
a front-faced (FF) setup with excitation at 340 nm (VCC) and
5 (185 mg, 79%) as a yellow solid, mp 174–176 ◦C; IR (neat) nmax
-1
=
(cm ) 3811, 3150–2700 (COOH), 1921,1785, 1733(C O), 1668,
1567, 1443, 1370, 1305, 1233, 1190; 1H-NMR (500 MHz, DMSO-
§ Excess ephedrine and carboxylic acids caused precipitation immediately
after mixing, due to complex formation.15 Therefore, PV-1:8-MAA was
prepared by a slightly different procedure. (-)-Ephedrine, coumarin,
EDMA and AIBN were dissolved in MeCN in a glass vial. The vial was
degassed and purged with argon as described above. Before capping and
sealing, MAA was added very quickly and the solution was bubbled with
argon for a further 20 s. The solution was polymerised in the same manner
as the others. By following this procedure, the mixture could be purged of
oxygen more easily.
d6) d(ppm): 8.19 (d, 1H, H5, J5,7 = 1.8Hz), 7.86 (dd, 1H, H7, J7,5
=
1.9 Hz, J7,8 = 8.6 Hz), 7.46 (d, 1H, H8, J8,7 = 8.6 Hz), 6.87 (s, 1H,
=
H3), 6.84 (dd, 1H, CH CH2, J = 11Hz, J = 17.6Hz), 5.87 (d, 1H,
=
=
CH CHaHb, J = 17.6 Hz), 5.35 (d, 1H, CH CHaHb, J = 11.0 Hz);
13C-NMR (DMSO-d6) d(ppm): 165.6 (COOH), 159.8 (C2), 153.7
=
(C9), 144.3 (C4), 135.7 (CH CH2), 134.0 (C6), 129.9(C7), 124.9
=
(C5), 118.5 (C3), 117.6 (C8), 116.2 (C10), 115.6 (CH CH2); MS
This journal is The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1211–1220 | 1219
©

View Article Online
345 nm (SCC). A 0.8 mL (0.2 ¥ 1 ¥ 4 cm) quartz cuvette was used
for rebinding experiments and a 4 mL (1 ¥ 1 ¥ 4 cm) quartz cuvette
was used for all other measurements. Slit widths were adjusted
to give emission below the saturating limit (1000000 cps) of the
detector, and then the same settings were used for a complete set
of experiments.
2 C. Alexander, H. S. Andersson, L. I. Andersson, R. J. Ansell, N. Kirsch,
I. A. Nicholls, J. O’Mahony and M. J. Whitcombe, Journal of Molecular
Recognition, 2006, 19, 106.
3 M. C. Blanco-Lopez, M. J. Lobo-Castan˜o´n, A. J. Miranda-Ordieres
and P. Tun˜o´n-Blanco, Trends in Analytical Chemistry, 2004, 23, 36; D.
Kriz and K. Mosbach, Analytica Chimica Acta, 1995, 300, 71.
4 K. Haupt, K. Noworyta and W. Kutner, Analytical Communications,
1999, 36, 391; S. Stanley, C. J. Percival, T. Morel, A. Braithwaite, M. I.
Newton, G. McHale and W. Hayes, Sensors and Actuators, B: Chemical
Sensors and Materials, 2003, 89, 103.
5 R. Levi, S. McNiven, S. A. Pilesky, S. H. Cheong, K. Yano and I.
Karube, Analytical Chemistry, 1997, 69, 2017; O. Y. F. Henry, D. C.
Cullen and S. A. Pilesky, Analytical and Bioanalytical Chemistry, 2005,
382, 947.
6 E. P. C. Lai, A. Fafara, V. A. VanderNoot, M. Kono and B. Polsky,
Canadian Journal of Chemistry, 1998, 76, 265.
7 A. Prasanna, de Silva, H. Q. Nimal Gunaratne, T. Gunnlaugsson,
A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chemical
Reviews, 1997, 97, 1515.
8 P. Turkewitsch, B. Wandelt, G. D. Darling and W. S. Powell, Analytical
Chemistry, 1998, 70, 2025.
9 F. L. Dickert, M. Tortschanoff, W. E. Bulst and G. Fischerauer,
Analytical Chemistry, 1999, 71, 4559; D. Kriz, O. Ramstro¨m, A.
Svensson and K. Mosbach, Analytical Chemistry, 1995, 67, 2142; F. L.
Dickert, H. Besenbock and M. Tortschanoff, Advanced Materials, 1998,
10, 149; J. Matsui, H. Kubo and T. Takeuchi, Analytical Chemistry,
2000, 72, 3286; J. L. Sua´rez-Rodr´ıguez and M. E. D´ıaz-Garc´ıa,
Analytica Chimica Acta, 2000, 405, 67.
(-)-Ephedrine (18.5 mg, 0.112 mmol) or (+)-ephedrine hemihy-
drate (19.5 mg, 0.112 mmol) (dried over P2O5) was dissolved in
1
700 mL MeCN-d3 (+0.3% TMS as internal standard). H NMR
spectroscopy was carried out to ensure that the concentrations
of each enantiomeric solution were equal, by comparing the
integrals of the ephedrine protons with that due to TMS. The
samples were adjusted by adding more solvent, if required, until
the concentrations were ≤ 2% of each other. The solutions were
then diluted 1 in 64 and then 1 in 10 with MeCN to give the final
stock solutions of 2.5 mM and 0.25 mM. (+/-)-Norephedrine
and (+)-pseudoephedrine stock solutions were prepared in the
same manner except that the concentrations were not checked by
1H NMR spectroscopy.
PV-1:1, CV-1:1, PV-1:8, CV-1:8 (4.43 mg/mL, containing
0.5 mmol coumarin/mL), PS-1:1, CS-1:1 (4.46 mg/mL) and
PV-1:1-MAA, CV-1:1-MAA, PV-1:8-MAA, CV-1:8-MAA (4.37
mg/mL) stock solutions were prepared in the same solvent.
Various amounts of ephedrine stock solution and 120 mL of
polymer stock solution were added to microcentrifuge tubes or
small glass vials and the samples were made up to a total volume of
1.2 mL with MeCN. Each concentration was prepared in duplicate
or triplicate. After incubating in the dark on a shaker for at least
4 h, the contents of the vials were transferred into a 0.8 mL cuvette
and fluorescence spectra were acquired. The cuvette was shaken
vigorously before each measurement.
10 J. L. Suarez-Rodriguez and M. E. Diaz-Garcia, Biosensors and Bioelec-
tronics, 2001, 16, 955.
11 W. Wang, S. Gao and B. Wang, Organic Letters, 1999, 1, 1209; S. Gao,
W. Wang and B. Wang, Bioorganic Chemistry, 2001, 29, 308.
12 D. L. Rathbone, D. Su, Y. Wang and D. C. Billington, Tetrahedron
Letters, 2000, 41, 123.
13 T. Takeuchi, T. Mukawa and H. Shinmori, The Chemical Record, 2005,
5, 263; H. Kubo, N. Yoshioka and T. Takeuchi, Organic Letters, 2005,
7, 359.
14 Y. Liao, W. Wang and B. Wang, Bioorganic Chemistry, 1999, 27, 463;
D. L. Rathbone, A. Ali, P. Antonaki and S. Cheek, Biosensors and
Bioelectronics, 2005, 20, 2353; T. Takeuchi, T. Mukawa, J. Matsui, M.
Higashi and K. D. Shimizu, Analytical Chemistry, 2001, 73, 3869; J.
Matsui, M. Higashi and T. Takeuchi, Journal of the American Chemical
Society, 2000, 122, 5218; N. Lavignac, C. J. Allender and K. R. Brain,
Tetrahedron Letters, 2004, 45, 3625.
Fluorescence measurements in aqueous buffers
(-)-Ephedrine (18.51 mg, 0.112 mmol) or (+)-ephedrine hemihy-
drate (19.52 mg, 0.112 mmol) was dissolved in 700 mL MeOH-d4
15 R. J. Ansell and K. Kuah, Analyst, 2005, 130, 179.
16 R. J. Ansell, D. Wang and K. L. Kuah, Analyst, 2008, 133, 1673.
17 R. Hekmatshoar, Y. S. Beheshtiha, M. M. Heravi and K. Asadollah,
Phosphorous, Sulfur and Silicon, 2002, 177, 703.
1
(+0.3% TMS). H NMR spectroscopy was carried out to ensure
that the concentrations of each enantiomeric solution were equal,
by comparing the integrals of the ephedrine protons with that due
to TMS. The solutions were then diluted 3 in 8 with MeOH to
give the final stock solutions of 60 mM. Polymer stock solutions
of the same concentrations as those in MeCN were prepared in
buffer. Samples were made up similarly to those in MeCN except
that MeCN was replaced with buffer. It was also ensured that the
amount of MeOH in each sample was the same by adding more
pure MeOH and less buffer if required.
18 M. Beller and C. Bolm, eds., Transition metals for organic synthesis,
Wiley-VCH, Weinheim, 1998.
19 F. Diederich and P. J. Stang, eds., Metal-catalyzed cross-coupling
reactions, Wiley-VCH, Weinheim, 1998; N. Miyaura and A. Suzuki,
Chemical Reviews, 1995, 95, 2457.
20 V. Farina, V. Krishnamurthy and W. J. Scott, The Stille reaction, John
Wiley & Sons, New York, 1998.
21 V. Farina and B. Krishnan, Journal of the American Chemical Society,
1991, 113, 9585.
22 S. S. d. Melo, R. S. Becker and A. L. Macanita, Journal of Physical
Chemistry, 1994, 98, 6054.
23 X. Yu, D. Scheller, O. Rademacher and T. Wolff, Journal of Organic
Chemistry, 2003, 68, 7386; C. Kim, A. Trajkovska, J. U. Wallace and
S. H. Chen, Macromolecules, 2006, 39, 3817.
Mixtures were incubated and fluorescence measurements per-
formed as described for samples in MeCN.
24 S. S. Rathi, M. K. Machwe, K. Gopalakrishanan and V. V. S. Murthi,
Current Science, 1972, 41, 326; J.-J. Aaron, M. Buna, C. Parkanyi, M. S.
Antorious, A. Time and L. Cisse, Journal of Fluorescence, 1995, 5, 337.
25 B. Valeur, Molecular Fluorescence: Principles and Applicatons, Wiley-
VCH, 2002, D. M. Hercules, Fluorescence and Phosphorescence Analy-
sis, Interscience Publishers, New York, 1966.
Acknowledgements
THN acknowledges an ORS award from Universities UK.
26 C. E. Hunt and R. J. Ansell, Analyst, 2006, 131, 678.
27 B. Sellergren, Makromol. Chem., 1989, 190, 2703.
28 SciFinder Scholar, American Chemical Society, 2006 edn., 2006.
29 L. I. Andersson, Analytical Chemistry, 1996, 68, 111.
Notes and references
1 G. Wulff, Angewandte Chemie, International Edition, 1995, 34, 1812; K.
Haupt and K. Mosbach, Chemical Reviews, 2000, 100, 2495.
1220 | Org. Biomol. Chem., 2009, 7, 1211–1220
This journal is
The Royal Society of Chemistry 2009
©