Microchim Acta

DOI 10.1007/s00604-014-1230-9

ORIGINAL PAPER

Magnetic molecularly imprinted polymer nanoparticles

for the solid-phase extraction of paracetamol from plasma
samples, followed its determination by HPLC
Saman Azodi-Deilami & Alireza Hassani Najafabadi &
Ebadullah Asadi & Majid Abdouss & Davood Kordestani

Received: 5 January 2014 / Accepted: 7 March 2014

# Springer-Verlag Wien 2014

Abstract We are presenting magnetic molecularly imprinted Introduction

polymer nanoparticles (m-MIPs) for solid-phase extraction and
sample clean-up of paracetamol. The m-MIPs were prepared Biological fluids such as blood, plasma or urine are referred to
from magnetite (Fe3O4) as the magnetic component, paraceta- as complex matrices and normally contain a number of en-
mol as the template, methacrylic acid as a functional monomer, dogenous components. These matrix components may inter-
and 2-(methacrylamido) ethyl methacrylate as a cross-linker. fere with and may adversely affect the subsequent separation
The m-MIPs were then characterized by transmission electron and identification of analyte(s) of interest if they are not
microscopy, FT-IR spectroscopy, X-ray diffraction and vibrat- removed. Therefore, two-thirds of the total analysis time in
ing sample magnetometry. The m-MIPs were applied to the chromatographic methods is typically due to the sample prep-
extraction of paracetamol from human blood plasma samples. aration steps and these are often the major source of error in
Following its elution from the column loaded with the m-MIPs the overall analytical process.
with an acetonitrile-buffer (9:1) mixture, it was submitted to A few types of extraction method are used routinely in
HPLC analysis. Paracetamol can be quantified by this method analytical toxicology such as Liquid-liquid Extraction (LLE)
in the 1 μg L−1 to 300 μg L−1 concentration range. The limit of [1], Liquid-liquid Micro-Extraction (LLE) [2], Solid Phase
detection and limit of quantification in plasma samples are 0.17 Extraction (SPE) [3], Solid-Phase Micro-Extraction (SPME)
and 0.4 μg L−1. The preconcentration factor of the m-MIPs is [4], and Molecularly Imprinted Solid Phase Extraction
40. The HPLC method shows good precision (4.5 % at (MISPE) [5, 6]. Simplicity and versatility of SPE rather than
50 μg L−1 levels) and recoveries (between 83 and 91 %) from LLE for intensive labor and solvent consuming, has been
spiked plasma samples. made it most convenient method at biological fluids such as
drug extractions. MISPE is currently a growing field in clean
up techniques for the analysis of biological samples.
Keywords Molecularly imprinted polymer nanoparticles . SPE is a simple, fast, portable sampling and sample prep-
Magnetic solid phase extraction . Human plasma samples . aration technique using small volumes of solvent and is
Cross-link agent . HPLC solvent-free in several cases. SPE is a very common type of
sample extraction and clean up technique for biological
Electronic supplementary material The online version of this article fluids in analytical chemistry and a well-accepted sam-
(doi:10.1007/s00604-014-1230-9) contains supplementary material, ple preparation technique in the analytical chemistry
which is available to authorized users.
community [3–6].
S. Azodi-Deilami (*) : A. H. Najafabadi : E. Asadi : A Molecularly Imprinted Polymer (MIP) is a synthetic
M. Abdouss (*)
polymer bearing on its surface the molecular imprint of a
Department of Chemistry, Amirkabir University of Technology, P. O.
Box: 15875–4413, Tehran, Iran specific target molecule (the template). Because of its imprint-
e-mail: [email protected] ing according to the size, shape and functional groups of a
e-mail: [email protected] template molecule, MIP acts as an artificial specific receptor.
In fact, the trace analysis of bio-fluids and other complex
D. Kordestani
Faculty of Chemistry, Department of Organic Chemistry, Razi matrices are usually based on selective sample pre-treatment
University, Kermanshah, Iran [7]. However, when a selective extraction from a complex
 S. Azodi-Deilami et al.

sample has to be performed, the typical SPE adsorbents lack Experimental

sufficient selectivity. Therefore, seeking more selective SPE
adsorbents is important. The SPE approaches using highly Materials
selective adsorbents such as MIPs have the potential to over-
come this problem [8]. Besides the high selectivity of MIPs, Iron (III) chloride hexahydrate (FeCl3.6H2O), iron (II) sul-
their simplicity of synthesis and less strict operation condi- phate heptahydrate (FeSO4.7H2O), ammonia solution (28 %,
tions compared with immune sorbents have led to their out- w/w) for synthesis of Fe3O4 nanoparticles, ethanol, tetraethyl
standing widespread applications [9]. The main advantage of orthosilicate (TEOS) and methacryloxypropyl
molecular imprinting is the possibility of preparing a selective trimethoxysilane (MPS) were obtained from Merck (http://
sorbent pre-determined for a particular substance or a group of www.merckmillipore.com, Darmstadt, Germany). For
structurally related substances. The polymer has a permanent preparation of the cross linker, methacrylic acid (MAA), thi-
memory of the template and is capable of selectively onyl chloride (SOCl2) and ethanol amine were obtained from
rebinding the template or structurally similar molecules such Merck (Darmstadt, Germany), Acros organics (http://www.
as the metabolites. acros.com, New Jersey, US) and Sigma companies (http://
The application of MIPs as sorbents in SPE has been found www.sigmaaldrich.com, US). Paracetamol and other drugs
very useful for selective extraction for the reason that MIPs obtained from the Ministry of Health and Medical Education
offer higher selectivity than conventional SPE. MIPs also (Tehran, Iran). 2,2-Azobisisobutyronitrile (AIBN) and sol-
reduce the influence of the matrix on the resulting chromato- vents were also obtained from Merck.
gram and high sample enrichment factors are achieved. The
MIPs are extensively employed in SPE and they found to be
quite effective and selective adsorbents [10]. Instruments and chromatographic conditions
Currently, magnetic nanoparticles have found numerous
1
biological and environmental applications owing to their H NMR and 13C NMR spectra were recorded in a Bruker
unique magnetic properties enabling them to be used by the 250 (250 MHz). Sample size was characterized by trans-
magnetic field in order to separate the target from the samples mission electron microscopy (TEM, Philips CM30) oper-
[11]. Thus, it can be anticipated that the cooperation of the ating at an acceleration voltage of 150 kV. The samples
magnetic nanoparticles with MIPs will offer a new approach for TEM measurements were prepared by placing one
for separation applications and extend employing of MIPs. In drop of the sample on copper grids coated with carbon.
addition, the combination of magnetically susceptible proper- The structure of the powder samples was characterized by
ties of magnetic nanoparticles and the advantage of MIP not X-ray powder diffraction (XRD, INEL EQuinox 3000,
only affords the selectivity for the target molecule but also http://www.inel.fr/, France). The XRD patterns were
provides the ability of one-step separation. The magnetic taken from 10° to 80° (2θvalue) using a Cu Kα radiation
molecularly imprinted polymers (m-MIPs) were used for sep- source (λ=1.542 Å) and resolution below 0.1°. All FT-IR
aration of dibenzothiophene [12]. The extraction of melamine spectroscopic measurements were performed on a Bruker
from milk followed by liquid chromatography-tandem mass EQuinox 55 (http://www.bruker.com/, Germany). The
spectrometry was performed by m-MIPs [13]. The extraction magnetic measurements were carried out in a vibrating
of chloramphenicol from honey using m-MIPs was also car- sample magnetometer (VSM, BHV-55, Riken Denshi,
ried out [14]. Moreover, the solid phase extractions of ceph- h t t p : / / w w w. r i k e n d e n s h i . c o . j p / , J a p a n ) a t r o o m
alosporin from milk [15], synephrine from methanol-water temperature. The concentrations of template in the
media [16], emodin from kiwi fruit root [17], five magnetic polymers were detected on a UV-vis JEVWAY
sulphonylurea herbicides [18], organophosphorus pesticides 6505 spectrophotometer (http://www.jenway.com/, US).
in fruit samples [19] and epinephrine in real samples [20] have The particle size and size distribution of polymer particles
been conducted. were measured by dynamic light scattering (DLS, Malvern
Herein, we successfully synthesized 2-(methacrylamido) Zetasizer ZS, http://www.malvern.com, Malvern, UK). The
ethyl methacrylate that was used as a new cross linker to polymer particles were suspended in acetonitrile and
prepare magnetic molecularly imprinted nanoparticles for measured at a fixed scattering angle of 90°. The HPLC
screening of paracetamol in human plasma using combination system consist a Waters 515 pump, a 486Waters UV/Vis
of solid phase extraction and HPLC method. All magnetic detector, a model 7725i Rehodyne injector with a 20 μL
nanoparticles were characterized by TEM, XRD, FT-IR and sample loop. Chromatographic separation was achieved on
VSM analyses. The resulting magnetic molecularly imprinted an ACE 5 μm, C18 column of 4.6 mm×150 mm column.
polymers (m-MIPs) could be quickly collected and regenerat- HPLC data was acquired and processed using a PC and
ed by an external magnetic field without tedious centrifuga- Millennium 2010 Chromatogram Manager software (Version
tion or filtration. 2.1 Waters).
Extraction of paracetamol by magnetic molecularly imprinted polymers in plasma

MISPE conditions 141.2, RHNCO(CH3)C=CH2; 135.7, RCO2(CH3)C=CH2;

126.1, RCO2(CH3)C=CH2; 121.0, RHNCO(CH3)C=CH2;
m-MIP and m-NIP columns were prepared by packing 64.0, OCH2; 39.7, NHCH2; 19.2, RHNCO(CH3)C=CH2;
100 mg of the polymers into 2 mL empty cartridge. The 18.7, RCO2(CH3)C=CH2.
cartridge was conditioned sequentially with 1 mL of metha-
nol, 1 mL of deionized water and 1 mL ammonium phosphate Synthesis and modification of Fe3O4 MNPs
(20 mmol L−1) at pH 3.0. Extraction experiments involved
loading the column with 5 mL, 50 μg L−1 paracetamol Iron oxide magnetic nanoparticles (MNPs) were prepared
(pH 9.0) at a flow rate of 1 mL min−1. After loading, the according to a reported procedure by Yang et al. [22]. Briefly,
column was washed with 1 mL mixture of acetone:acetonitrile FeCl3.6H2O (5.4 g, 20 mmol) was dissolved in 140 mL ultra-
(3:1v/v) and then 1 mL dicholomethan. After washes, SPE pure water, followed by adding FeSO4.7H2O (2.7 g, 10 mmol)
cartridge was dried thoroughly by vacuum application. Final- in N2 atmosphere. After being absolutely dissolved, 11 mL of
ly, the elution was performed by passing 3 × 1 mL of analytical grade ammonia solution (28 %, w/w) was quickly
acetonitrile:phosphate buffer (90:10, pH 9.0) through the car- added into the solution under rapid mechanical stirring
tridge. All fractions were evaporated to dryness at 20 °C under (900 rpm), and then the mixture was heated up to 60 °C, while
a stream of nitrogen and finally recovered in 1 mL of mobile vigorously stirred by a mechanical stirrer for 1 hr under N2
phase. Then, 25 μL of each sample was injected onto the atmosphere. Finally, the precipitates were collected using an
analytical column of HPLC. external magnet and rinsed with ultra-pure water, and dried to
powder in a vacuum.
Synthesis procedures Then, the MNPs (1 g) were dispersed in 50 mL of distilled
water by the ultrasonic treatment for 20 min to form a ferro
Synthesis of 2-(methacrylamido) ethyl methacrylate as a cross fluid. 2 mL of the ferro fluid was first diluted with water
linker (40 mL), and then the resultant suspension and 5 mL of
NH3.H2O were poured into 140 mL of ethanol with vigorous
Methacryloyl chloride was synthesized by reacting stirring at 40 °C. Finally, under continuous mechanical stir-
methacrylic acid with thionyl chloride. Thionyl chloride ring, 1 mL of TEOS diluted in ethanol (20 mL) was added
(78 mmol, 5.7 mL) was added dropwise to a mixture of dropwise to this dispersion. The resulting dispersion was
methacrylic acid (60 mmol, 5 mL) along with 3 mL of dichlo- continuously stirred mechanically for 14 h at room tempera-
romethane, over a period of 1 h and the container was sealed ture. The magnetic Fe3O4@SiO2 nanoparticles were collected
under N2 atmosphere. After complete addition of thionyl by magnetic separation and washed with ethanol and deion-
chloride, the reaction mixture was maintained at 45 °C for ized water in sequence.
12 h with continuous stirring. Pure methacryloyl chloride was Finally, Fe3O4@SiO2 MNPs were modified by MPS.
distilled out at 55 °C. Briefly, 250 mg of Fe3O4@SiO2 MNPs was dispersed in
Afterwards, based on a procedure by Spivak et al. [21], a 50 mL of anhydrous toluene containing 3 mL of MPS, and
solution of methacryloyl chloride (81.6 mmol, 8 mL) in di- the mixture was allowed to react at 70ºC for 12 h under
chloromethane (20 mL) was added dropwise to a mixture of nitrogen. After magnetic separation, washing by water and
ethanolamine (40 mmol, 2.35 mL), triethylamine (81.6 mmol, drying in vacuum, the products (Fe3O4@SiO2-CH=CH2)
11.3 mL) in dichloromethane (25 mL), in an ice bath for 1 h. were obtained.
After the addition was complete, the reaction mixture was
stirred at 40 °C for a further 24 h. Synthesis of magnetic MIP nanoparticles
The reaction mixture was filtered out and the precipitate
(Et3NHCl) discarded. The filtrate was washed with 0.5 M The schematic diagram for the preparation of m-MIPs is
NaHCO3 (3×15 mL) and 0.5 M Sodium Citrate (3×15 mL). presented in Scheme 1. A solution of 4 mmol MAA as a
The organic phase was extracted and evaporated under vacu- monomer and 1 mmol paracetamol as a template in 30 mL
um. The pale yellow oily product was reached in a 55 % yield. acetonitrile/chloroform (5:1, v/v) was shaken at room temper-
IR (cm−1): 3,300, 1,720, 1,665, 1,631, 1,450, 1,333, 1,158, ature for 5 h to form a template-monomer complex. After-
1,085 and 940. 1H NMR (CDCl3, 250 MHz) δ/ppm 7.0 (1H, wards, 0.5 g of double-bond-functionalized Fe3O4 magnetic
bs), NH; 5.91 (1H, m), RO2C(CH3)C=CHaHb; 5.73 (1H, t), nanoparticles that were dispersed in acetonitrile/chloroform
RHNCOC(CH3)C=CHcHd; 5.56 (1H, t), (5:1, v/v), 10 mmol 2-(methacrylamido) ethyl methacrylate
RO2C(CH3)C=CHaHb; 5.40 (1H, t), and 0.1 mmol AIBN were added to the solution and the
RHNCOC(CH3)C=CHcHd; 4.26 (2H, t) OCH2; 3.63 (2H, mixture was sonicated in an ultrasonic bath. After sonication
quad) NHCH2; 1.9 (3H, s) CH3; 1.8 (3H, s) CH3. 13C NMR (5 min), it was purged with N2 and the flask was sealed under
(CDCl3, 62.5 MHz) δ/ppm 169.1, RNHCOR; 168.1, RCO2R; N2 atmosphere. The magnetic MIPs were synthesized in a
 S. Azodi-Deilami et al.

Scheme 1 Schematic illustration of the synthesis of magnetic MIPs

100 mL flask equipped with a mechanical stirrer in an oil bath necessary to have some treatments with plasma before extrac-
kept at 60 °C for 20 h. The resultant polymeric nanoparticles tion with m-MIP particles. So, the plasma samples was diluted
were collected by the external magnetic field and eluted by a with 25 mM ammonium acetate (pH 9.0), then centrifuge
mixture solvent of methanol and acetic acid (9:1, v/v) for 30 min at 6,000 rpm to remove excess of proteins. Then the
several times to extract the template molecules until the eluent supernatant was filtered through a cellulose acetate filter
was free from paracetamol as detected by UV-vis spectrome- (0.2 μm pore size, Advantec MFS Inc., CA, USA). The filtrate
try (at 257 nm). The obtained polymers were finally rinsed was collected in glass containers and stored at −20 °C until the
with ethanol to remove the remaining acetic acid and dried in analysis was performed. 2 mL of the filtered supernatant was
the vacuum for 24 h before use. For a comparison, magnetic collected to be directly percolated through the m-MIP or the
non-molecularly imprinted polymers (m-NIPs) were prepared m-NIP.
in the absence of the template during the polymerization
process and treated in an identical manner.
Results and discussion
Binding experiment
As illustrated in Scheme 1, the synthesis of the m-MIPs is a
For measuring of template binding, 50 mg of polymer nano- multistep procedure, which involves synthesis of Fe3O4
particles were added to 10 mL paracetamol solution (pH 7.0) MNPs, silica-shell deposition (Fe3O4@SiO2), modification
of various initial concentrations in a conical centrifugation of Fe3O4@SiO2 with MPS, preparation of a cross linker
tube and sealed. The mixtures were thermo stated at 25 °C (2-(methacrylamido) ethyl methacrylate) and m-MIPs. At
for 24 h under continuous stirring and then the polymer first, Fe3O4 MNPs were prepared by the co-precipitation
nanoparticles were collected by the external magnetic field. method. Then, the surface of Fe3O4 MNPs was coated with
Then, the residual concentration of paracetamolin solution silica by the TEOS. The SiO2 shell provided a biocompatible
was established using HPLC-UV at 257 nm. The quantity and hydrophilic surface, and prevented oxidation of Fe3O4.
(Q) of the template bound to m-MIP or m-NIP was calculated Furthermore, silanol groups were beneficial to chemical mod-
according to the following equation [23]: ification on the surface of Fe3O4@SiO2. Thus, double bonds
were introduced onto Fe3O4@SiO2 using MPS to ensure tight
Q ¼ ½ðC0 −Ct Þ  VŠ=W
growth of the imprinted layer. We have prepared
2-(methacrylamido) ethyl methacrylate by from the reaction
of ethanol amine with methacryloyl chloride in a 1:2 mole
Where C0 and Ct (mg L−1) are the initial concentration and ratio in dichloromethane. Finally, m-MIPs were synthesized
the residual concentration of aspirin, respectively; V (L) is the using modified Fe3O4 MNPs, MAA, cross linker, paracetamol
initial volume of the solution, and W (g) is the weight of the and AIBN in acetonitrile/chloroform (5:1v/v). The resultant
m-MIPs or m-NIPs. polymers were used to study the in vitro controlled release of
drug and human plasma assay.
Extraction procedure for human plasma samples
Optimal MIP formulation
Drug-free human plasma was obtained from the Iranian blood
transfusion service (Tehran, Iran) and stored at −20 °C until Various amounts of monomer and cross link agent were used
use after gentle thawing. Due to possibility of protein-binding for optimization of different formulations of MIPs with en-
for paracetamol and reducing the recoveries processes, it is hanced molecular recognition capabilities. Generally, for
Extraction of paracetamol by magnetic molecularly imprinted polymers in plasma

Fig. 1 FT-IR spectra of Fe3O4

(a); Fe3O4@SiO2 (b);
Fe3O4@SiO2–CH=CH2 (c) and
m-MIPs (d)

preparation of enhance specific polymers and formation of magnetically to an external magnetic field and this response
MIP recognition sites, several molar ratios of the functional vanished upon removal of the magnetic field. The saturation
monomer MAA and cross link agent to template were used in magnetizations of Fe3O4, Fe3O4@SiO2−CH=CH2 and m-
this study. As Table S1 (ESM) depicts, the optimum ratio of
functional monomer and cross link agent to template for the
specific rebinding of paracetamol was 3:10:1 (m-MIP 7),
which had the best extraction of 89 %. The extraction of the
corresponding NIP was low as 16 %.

Characterization of magnetic MIPs

FT-IR

The products of Fe 3 O 4 , Fe 3 O 4 @SiO 2 , Fe 3 O 4 @SiO 2 –

CH=CH2, and m-MIPs, were investigated by FT-IR spectros-
copy (Fig. 1). The absorption band at 573 cm−1 is attributed to
the Fe-O stretching bond for Fe3O4 nanoparticles. The peaks
of Si-O-Si group at 1,100 cm−1 and −OH group at about
1,623 cm−1 and 3,400 cm−1 indicated the formation of
Fe3O4@SiO 2. Also, we can conclude that the peak at
1,621 cm−1 in Fe3O4@SiO2−CH=CH2 was caused by C=C
stretching vibration and the peak at 1,100 cm −1 in
Fe3O4@SiO2−CH=CH2 was evidently stronger than that in
Fe3O4@SiO2, both showed that the surface of Fe3O4@SiO2
has been successfully modified with the functional vinyl
group. The FT-IR spectra of m-MIPs and m-NIPs were almost
the same, because after removing the template molecule, the
chemical composition of the m-MIPs was similar to that of the
m-NIPs. A very intense band of C=O stretching is seen at
about 1,720 cm−1 and C−H group of methyl at 2,990 cm−1
indicating that MIP layers have been successfully formed on
the surface of Fe3O4@SiO2−CH=CH2.

VSM

The magnetic properties of the Fe 3 O 4 , Fe 3 O 4 @SiO 2 ,

Fe3O4@SiO2−CH=CH2 and magnetic MIP nanoparticles were
characterized by VSM at room temperature, and the results are
displayed in Fig. S2 (Electronic Supplementary Material, Fig. 2 TEM images of Fe3O4 nanoparticles (a); Fe3O4@SiO2@MPS
ESM). This feature illustrated that the materials respond (b); m-MIPs (c)
 S. Azodi-Deilami et al.

Fig. 3 Effect of pH on rebinding

efficiency of paracetamol. 50 mg
of the polymers; sample volume:
5 mL; paracetamol concentration:
50 mM; room temperature (mean
± S.D., n=3)

MIPs were 61.36, 4.58 and 2.7 emu g−1, respectively. The magnetic MIPs. This result suggested that the Fe3O4 nanopar-
magnetic MIP achieved lower saturation magnetization value ticles were indeed incorporated into MIPs and the structure of
than magnetic nanoparticles. This was expected because the Fe3O4 nanoparticles was not changed during the polymeriza-
polymeric coating has effectively shielded the magnetite. How- tion process.
ever, the magnetic MIPs possess enough magnetic response to
meet the need of magnetic separation within a short time. TEM

XRD The morphological structure of Fe 3 O 4 nanoparticles,

Fe3O4@SiO2−CH=CH2 and m-MIPs were detected by TEM
The X-ray powder diffraction patterns of Fe 3 O 4 , as given in Fig. 2. It can be observed that Fe3O4@SiO2@MPS
Fe3O4@SiO2, Fe3O4@SiO2−CH=CH2, and polymer-coated cores were surrounded by polymer layers.
Fe3O4 nanoparticles are indicated in Fig. S3 (ESM). Five
characteristic peaks for Fe3O4 marked by their indices (220), EDX
(311), (400), (422), (511) and (440) were observed for all the
samples to reveal pure Fe3O4 in the resultant particles [24]. Figure S4 (ESM) shows the patterns of EDX analysis to
The peak positions of the Fe3O4 nanoparticles are all consis- confirm the existence of magnetic imprinted polymers. In
tent with the lower diffraction intensity before and after the Fig. S4(a), Fe and O signals are related to the elemental
polymerization, and it is found that the peaks of magnetic composition of Fe3O4 nanoparticles. Besides peaks of O and
MIPs are obviously broadened due to lower Fe3O4 content in Fe, the presence of Si and C peaks in Fig. S4(b) clearly

Fig. 4 Effect of paracetamol

concentration on the adsorption
capacity on m-MIP and m-NIP
nanoparticles at pH 9.0
Extraction of paracetamol by magnetic molecularly imprinted polymers in plasma

Table 1 Distribution ratio (KD) and selectivity coefficient (α) values for Table 2 Assay of paracetamol in human plasma by SPE-HPLC
magnetic imprinted and non-imprinted polymers procedure

Compound m-MIP m-NIP Sample Spiked value (μg L−1) Recovery% ± SDa

KD α KD α MIP NIP

Paracetamol 900 22 Human serum 5 83±1.1 11±1.0

Bromhexine 51 17.65 24 0.92 10 90±1.2 13±1.7
Aspirin 85 10.59 20 1.10 25 87±1.3 10±1.4
Meloxicam 25 36.00 19 1.16 50 91±1.6 10±1.5
Diphenhydramine 37 24.32 21 1.05 a
Average of three determinations

indicates the modification of Fe3O4 nanoparticles surface with Lesser effects were observed at lower pH values and which
MPS. Higher percentage of C and lower percentage of Si and may have been attributed to the protonation of the function-
Fe in Fig. S4(c) validates successfully synthesis of magnetic al group of paracetamol and consequence breakdown of the
molecularly imprinted polymer nanoparticles. hydrogen bonds at the pH<7.0. Finally, as the recovery rate
of paracetamol flattens at around the pH 9.0, therefore, this
Effect of pH on drug loading pH value was chosen for later SPE experiments.

The effect of pH on the sorption of paracetamol was exam- Adsorption capacity of polymers and preconcentration factor
ined by varying the pH of solutions from 4.0 to 10.0. Several
batch experiments were performed by equilibrating 50 mg One of the important factors we studied was the capacity of a
of the imprinted particles with 5 mL of the solutions con- sorbent to quantitatively remove a specific amount of template
taining 0.05 mM of paracetamol under the desired levels of from the solution. Once the system has come to equilibrium,
pH. The results for different polymers (Fig. 3) displayed that the amount of free template in the solution is measured to
pH have great effects on loading. The binding behavior of determine the amount of adsorbed template. In the measure-
paracetamol was not greatly affected at pH>7. Although, ment of adsorption capacity of m-MIP and m-NIP absorbents,

Fig. 5 HPLC chromatograms

obtained after percolation of 2 mL
human serum spiked with
50.0 μgL−1 of paracetamol with a
cleanup step comprising the a m-
MIP and b m-NIP monitored at
257 nm; conditions: column ACE
5 μm, C18 4.6 mm×250 mm at +
40 °C, eluent acetonitrile:
phosphate buffer (0.01 mol L−1,
pH 3.0) 90:10 at flow rate of
1.0 mL min−1
 S. Azodi-Deilami et al.

Table 3 Figures of merit of comparable methods for determination of paracetamol

Method Analytical range Detection limit Comments Reference

HPLC 1.5–200 μg mL−1 50 ng mL−1 Simple, selective and sensitive [25]

Flow-through UV 0.5–0.8 μg mL−1 0.022 μg mL−1 A simple device based on a suitable active solid [26]
optosensing device support
HPCE 1–100 μM 0.5 μM A modification of CE combining the advantages [27]
of CE and chromatography for separation
from plasma proteins
HPTLC 120–360 μg mL−1 1.8 μg mL−1 On silica gel 60F254 HPTLC plates with [28]
toluene-acetone-methanol (8:1:1, v/v/v) as
mobile phase; detection at 263 nm
MEKC 260–520 μg mL−1 0.5 μg mL−1 Uncoated fused silica capillary for the separation [29]
at an applied voltage of 20 kV. A phosphate
run buffer (pH 9, 0.05 M) containing 0.05 M
sodium dodecyl sulphate was used for
analysis.
Flow solid-phase UV 3–30 μg mL−1 0.104 μg mL−1 A continuous and simple multiparameter sensor [30]
spectrophotometry for the sequential determination of
salicylamide and paracetamol by solid-phase
UV spectrophotometry
HPLC and SFC 0.3–20 μg mL−1 0.1–0.2 μg mL−1 Using packed column SFC employing internal [31]
standard method. The analytes were
monitored spectrophotometrically at 225 nm.
SFC method was compared to an HPLC assay
IC 0.5–7 mg L−1 0.06 mg L−1 A simple technique for the measurement of [32]
paracetamol and its degradation product
FTIR spectrometry 2–10 mg L−1 0.34 mg L−1 FTIR technique for the simultaneous [33]
determination of ibuprofen and paracetamol.
Quantification was carried out by measuring
the absorbances using the baseline for
measurement correction.
CE 0.5–100 μM 0.25 μM Developed CE method using paracetamol [34]
glucuronide as a novel probe for human ß-
glucuronidase activity
MIP-BAW sensor 0.005–0.1 μM 0.005 μM A piezoelectric quartz crystal modified with MIP [35]
to obtain selectivity for a bulk acoustic wave
sensor. Stable, selective and facile
m-MIP and SPE 1–300 μg L−1 0.17 μg L−1 Magnetic molecularly imprinted polymer This work
coupled with HPLC nanoparticles for solid-phase extraction and
sample clean-up of paracetamol simply and
precisely. Avoiding time consuming
separation operation and reducing the loss of
both liquid and solid during the separation
process. Having good precision and
recoveries.

HPCE high performance capillary electrophoresis, HPTLC high performance thin layer chromatography, MEKC micellar electrokinetic chromatogra-
phy, SFC supercritical fluid chromatography, IC ion chromatography, BAW bulk acoustic wave

50 mg samples of the absorbents were added to 100 mL pH 9.0. As all the accessible specific cavities of the m-MIP
paracetamol solutions at concentrations of 10–500 mg L−1. particles are saturated, the retention of the analyte is mainly
The suspensions were mechanically shaken at room tempera- due to non-specific interactions which can be identical for m-
ture, followed by centrifuging and removal of absorbents. The MIP and m-NIP polymers. In order to obtain preconcentration
remaining paracetamol in the supernatant was measured by factor, the effect of sample concentration on the sorption
HPLC-UV. The adsorption isotherm, which is the number of behavior of analyte on m-MIPs was investigated. Amount of
milligram adsorbed per gram of adsorbent (Q) versus the paracetamol in supernatant solution determined by HPLC-
equilibrium concentration of paracetamol, is shown in UV. The sorption of drug was quantitatively recovered at the
Fig. 4. According to these results, the maximum amount of range of 10–600 mg L−1. At the higher concentration of
paracetamol that can be absorbed by m-MIP is 84 mg g−1 at 400 mg L−1, the extractions for analytes were not quantitative.
Extraction of paracetamol by magnetic molecularly imprinted polymers in plasma

The preconcentration factor of the m-MIPs is calculated by the Conclusions

ratio of the highest sample concentration for analyte
(400 mg L−1) and the lowest concentration (10 mg L−1). The In this study, magnetic molecularly imprinted nanoparticles
preconcentration factor was 40. were prepared using a novel cross-link agent.
Superparamagnetic Fe3O4 particles were synthesized and the
surfaces of magnetic Fe3O4 particles were successfully mod-
Study of MIP selectivity ified and functionalized in order to disperse in organic sol-
vents and take part in the reaction of molecular imprinted
In order to evaluate the selectivity of the synthesized m-MIP, polymerization. Then, m-MIPs were prepared by using para-
Paracetamol, and several drugs with the ability to form hydro- cetamol as the template, MAA as the functional monomer,
gen bonds with the m-MIP nanoparticles were selected in this 2-(methacrylamido) ethyl methacrylate as the cross linker and
section. Their molecular structures are shown in Fig. S1 double-bond-functionalized Fe3O4 nanoparticles as the sta-
(ESM). The initial concentration of drugs (5 mL, 0.05 mM) tionary phase by the molecular imprinting technique. The
was extracted by 50 mg of m-MIPs and m-NIPs at pH 9.0. The obtained m-MIPs were characterized by TEM, FT-IR, XRD,
distribution ratio (KD) was calculated using the equation: and VSM analyses. By this method, paracetamol can be

quantified in the 1 μg L−1 to 300 μg L−1 concentration range.
C i −C f V
KD ¼ The recoveries for the spiked human plasma samples were
Cf m
between 83 and 91 % in 5–50 μgL−1 respectively. It could be
Where V, Ci, Cf, and m represent the volume of the solution concluded that the technique has great potential in developing
(mL), drug concentration before and after adsorption (mM) selective extraction method for other compounds. Compari-
and mass of the polymer, respectively. The selectivity coeffi- son of the results of the determination of paracetamol obtained
cient (α) is defined as: by several methods is shown in Table 3. Compared with other
techniques, the proposed method has impressive results and
K D ðparacetamolÞ the detection limit is much lower than other works.
a¼
K D ðforeigncompound Þ
The results are listed in Table 1. The higher KD obtained for
paracetamol strongly confirmed a higher selectivity of m-
MIPs this template and demonstrated the possibility of m- References
MIP as a selective adsorbent for the extraction of paracetamol
from complex matrixes such as biological fluids. 1. Zhong Z, Li G, Zhong X, Luo Z, Zhu B (2013) Ultrasound-assisted
low-density solvent dispersive liquid-liquid extraction for the deter-
mination of alkanolamines and alkylamines in cosmetics with ion
Real sample analysis chromatography. Talanta 115:518–525
2. Rezaee M, Mashayekhi HA, Saleh A, Abdollahzadeh Y, Naeeni MH,
Fattahi N (2013) Determination of abamectin in citrus fruits using
To demonstrate the potential of MIP for the sample clean-up, SPE combined with dispersive liquid–liquid microextraction and
the MIP was applied to the purification of spiked paracetamol HPLC-UV detection. J Sep Sci 36:2629–2634
in human plasma. Aqueous media was employed for the 3. Negreira N, López de Alda M, Barceló D (2013) On-line solid phase
loading solution, and the wash procedure was assessed for extraction–liquid chromatography–tandem mass spectrometry for the
determination of 17 cytostatics and metabolites in waste, surface and
obtaining maximum recovery of the analytes using acetone. ground water samples. J Chromatogr A 1280:64–74
The chromatograms obtained for plasma samples are shown in 4. Rajabi AA, Yamini Y, Faraji M, Seidi S (2013) Solid-phase
Fig. 5. This efficient method allowed cleaner extracts to be microextraction based on cetyltrimethylammonium bromide-coated
obtained and interfering peaks arising from the complex bio- magnetic nanoparticles for determination of antidepressants from
biological fluids. Med Chem Res 22:1570–1577
logical matrices to be suppressed. Results from the HPLC 5. Azodi-Deilami S, Abdouss M, Hasani AR (2010) Preparation and
analyses in Tables 2 showed that the MIP extraction of para- utilization of a molecularly imprinted polymer for solid phase extrac-
cetamol for plasma samples is linear in the ranges of 1– tion of tramadol. Cent Eur J Chem 8:861–869
300 μg L−1 with good precision (4.5 % for 50.0 μg L−1) and 6. Abdouss M, Azodi-Deilami S, Asadi E, Shariatinia Z (2012)
Synthesis of molecularly imprinted polymer as a sorbent for solid
recoveries (between 83 and 91). Typical chromatograms (pre- phase extraction of citalopram from human serum and urine. J Mater
sented in Fig. 5) reveal that the MIP can be used for the sample Sci Mater Med 23:1543–1552
clean up and when MIP sorbent was used, a board peak in 7. Wulf G (2013) Fourty years of molecular imprinting in synthetic
chromatogram was omitted. The limit of detection (LOD) and polymers: origin, features and perspectives. Microchim Acta 180:
1359–1370
limit of quantification (LOQ) for paracetamol in plasma sam- 8. Zhu S, Gan N, Pan D, LI Y, Yang T, Hu F, Cao Y, Wu D (2013)
ples were 0.17 and 0.4 μg L−1. The number of replicate for Extraction of tributyltin by magnetic molecularly imprinted poly-
experiment was been three times. mers. Microchim Acta 180:545–553
 S. Azodi-Deilami et al.

9. Duan YP, Dai CM, Zhang YL, Chen L (2013) Selective trace enrich- 23. Jiang J, Song K, Chen Z, Zhou Q, Tang Y, Gu F, Zou X, Xu Z (2011)
ment of acidic pharmaceuticals in real water and sediment samples Novel molecularly imprinted microsphere using a single chiral mono-
based on solid-phase extraction using multi-templates molecularly mer and chirality-matching (S)-ketoprofen Template. J Chromatogr
imprinted polymers. Anal Chim Acta 758:93–100 A 1218:3763–3770
10. Junping W, Mingfei P, Guozhen F, Shuo W (2009) Preparation of a 24. Sun H, Hong J, Meng F, Gong P, Yu J, Xue Y, Zhao S, Xu D, Dong L,
novel molecularly imprinted polymer by a sol–gel process for on-line Yao S (2006) Novel core-shell structure polyacrylamide-coated mag-
solid-phase extraction coupled with high performance liquid chro- netic nanoparticles synthesized via photochemical polymerization.
matography to detect trace enrofloxacin in fish and chicken samples. Surf Coat Technol 201:250–254
Microchim Acta 166:295–302 25. Jensen LS, Valentine J, Milane RW, Evans AM (2004) The quanti-
11. Valdés MG, Valdés González AC, García Calzón JA, Díaz-García fication of paracetamol, paracetamol glucuronide and paracetamol
ME (2009) Analytical nanotechnology for food analysis. Microchim sulphate in plasma and urine using a single high-performance liquid
Acta 166:1–19 chromatography assay. J Pharm Biomed Anal 34:585–593
12. Hui L, Xu W, Wang N, Ma X, Niu D, Jiang B, Liu L, Huang W, Yang 26. Ayora- Cañada MJ, Pascual-Reguera MI, Ruiz Medina A, Fernández
W, Zhou Z (2012) Synthesis of magnetic molecularly imprinted de Córdova ML, Molina Díaz A (2000) Fast determination of para-
polymer particles for selective adsorption and separation of cetamol by using a very simple photometric flow-through sensing
dibenzothiophene. Microchim Acta 179:123–130 device. J Pharm Biomed Anal 22:59–66
13. He D, Zhang X, Gao B, Wang L, Zhao Q, Chen H, Wang H, Zhao C 27. Kunkel A, Günter S, Wätzig H (1997) Quantitation of acetaminophen
(2014) Preparation of magnetic molecularly imprinted polymer for and salicylic acid in plasma using capillary electrophoresis without
the extraction of melamine from milk followed by liquid sample pretreatment improvement of precision. J Chromatogr A 768:
chromatography-tandem mass spectrometry. Food Control 36:36–41 125–133
14. Chen L, Li B (2013) Magnetic molecularly imprinted polymer ex- 28. Alkayer M, Vallon JJ, Pegon Y, Bichon C (1981) Dosage direct du
traction of chloramphenicol from honey. Food Chem 141:23–28 paracetamol dans les milieux biologiques par polarographie
15. Quesada-Molina C, Claude B, Garcia-Campana A, Olmo-Iruela M, sinusoidale. Anal Chim Acta 124:113–119
Morin P (2012) Convenient solid phase extraction of cephalosporins in 29. Haque A, Stewart JT (1999) Determination of Acetaminophen,
milk using a molecularly imprinted polymer. Food Chem 135:775–779 Caffeine, and butalbital in a commercial tablet dosage from by
16. Fan JP, Tian ZY, Tong S, Zhang XH, Xie YL, Xu R, Qin Y, Li L, Zhu micellar electrokinetic chromatography. J Liq Chromatogr Relat 22:
JH, Ouyang XK (2013) A novel molecularly imprinted polymer of the 2159–2166
specific ionic liquid monomer for selective separation of synephrine 30. Ruiz Medina A, Fernández de Córdova ML, Molina Díaz A (1999) A
from methanol–water media. Food Chem 141:3578–3585 very simple resolution of the mixture paracetamol and salicylamide
17. Yang X, Zhang Z, Li J, Chen X, Zhang M, Luo L, Yao S (2014) by flow injection–solid phase spectrophotometry. Anal Chim Acta
Novel molecularly imprinted polymers with carbon nanotube as 394:149–158
matrix for selective solid-phase extraction of emodin from kiwi fruit 31. Patil ST, Sundaresan M, Bhoir IC, Bhagwat AM (1999) Packed
root. Food Chem 145:687–693 column supercritical fluid chromatographic separation and estimation
18. Tang K, Gu X, Luo Q, Chen S, Wu L, Xiong J (2014) Preparation of of acetaminophen, diclofenac sodium and methocarbamol in phar-
molecularly imprinted polymer for use as SPE adsorbent for the maceutical dosage forms. Talanta 47:3–10
simultaneous determination of five sulphonylurea herbicides by 32. Pérez JL, Bello MA (1999) Determination of paracetamol in dosage
HPLC. Food Chem 150:106–112 forms by non-suppressed ion chromatography. Talanta 48:1199–
19. Sanagi MM, Salleh S, Aini W, Ibrahim W, Naim AA, Hermawan D, 1202
Miskam M, Hussain I, Aboul-Enein HY (2013) Molecularly imprinted 33. Tantishaiyakul V, Phadoongsombut N, Kamaung S, Wongwisansri S,
polymer solid-phase extraction for the analysis of organophosphorus Mathurod P (1999) Fourier transform infrared spectrometric deter-
pesticides in fruit samples. J Food Compos Anal 32:155–161 mination of paracetamol and ibuprofen in tablets. Pharmazie 54:111–
20. Prasad BB, Srivastava A, Prasad A, Tiwari MP (2014) Molecularly 114
imprinted micro solid-phase extraction technique coupled with com- 34. Bohnenstengel F, Kroemer HK, Sperker B (1999) In vitro cleavage of
plementary molecularly imprinted polymer-sensor for ultra-trace paracetamol glucuronide by human liver and kidney β-
analysis of epinephrine in real samples. Colloids Surf B 113:69–76 glucuronidase: determination of paracetamol by capillary electropho-
21. Sibrian-Vazquez M, Spivak DA (2004) Molecular imprinting made resis. J Chromatogr B 721:295–299
easy. J Am Chem Soc 126:7827–7833 35. Tan Y, Zhou Z, Wang P, Nie L, Yao S (2001) A study of a bio-
22. Yang D, Hu J, Fu S (2009) Controlled Synthesis of Magnetite−Silica mimetic recognition material for the BAW sensor by molecular
Nanocomposites via a Seeded Sol−gel Approach. J Phys Chem 113: imprinting and its application for the determination of paracetamol
7646–7651 in the human serum and urine. Talanta 55:337–347