Synthesis of molecular imprinted polymer nanoparticles followed by application of response surface methodology for optimization of metribuzin extraction from urine samples

Abstract

A molecular imprinted polymer was prepared with precipitation polymerization technique and applied as a sorbent for selective extraction and enrichment of metribuzin herbicide prior to high performance liquid chromatography. Optimization of critical variables affecting the efficiency of molecularly imprinted solid-phase extraction (MISPE), such as sorbent mass, sample pH and flow rate of sample, volume, concentration, and flow rate of elution solvent was done by employing central composite design (CCD) of the response surface methodology. Two separate models were developed for the adsorption and recycling steps. The analysis of variance (ANOVA) demonstrated that, experimental data were excellently fitted to the proposed response models. The optimum operating conditions were: a sorbent mass of 25 mg, sample pH 6.19, sample flow rate of 2.15 mL/min, and a 5 mL portion of methanol/acetic acid with 92.7:7.3 (v/v) ratio and flow rate of 2.1 mL/min for the extraction process. Under the optimized conditions, the linear range was obtained from 20 to 120 µg/L (R2 = 0.999) and the lowest detectable concentration (LOD) and the lowest quantitative concentration (LOQ) were calculated as 5.75 and 19.86 µg/L, respectively. Finally, the designed MISPE method was successfully applied to determine trace amount of metribuzin in real samples. The diluted urine samples were spiked with metribuzin at 4 levels and extracted with recoveries ranging from 93.82 to 97.84% and the relative standard deviation (RSD) less than 4.8%.

Full text

1 23
Chemical Papers
ISSN 2585-7290
Chem. Pap.
DOI 10.1007/s11696-018-0546-z
Synthesis of molecular imprinted polymer
nanoparticles followed by application
of response surface methodology for
optimization of metribuzin extraction from
urine samples
Omid Reza Heravizadeh, Monireh
Khadem, Ramin Nabizadeh & Seyed
Jamaleddin Shahtaheri

1 23
Your article is protected by copyright and
all rights are held exclusively by Institute of
Chemistry, Slovak Academy of Sciences. This
e-offprint is for personal use only and shall not
be self-archived in electronic repositories. If
you wish to self-archive your article, please
use the accepted manuscript version for
posting on your own website. You may
further deposit the accepted manuscript
version in any repository, provided it is only
made publicly available 12 months after
official publication or later and provided
acknowledgement is given to the original
source of publication and a link is inserted
to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at link.springer.com”.

Vol.:(0123456789)
1 3
Chemical Papers
https://doi.org/10.1007/s11696-018-0546-z
ORIGINAL PAPER
Synthesis ofmolecular imprinted polymer nanoparticles followed
byapplication ofresponse surface methodology foroptimization
ofmetribuzin extraction fromurine samples
OmidRezaHeravizadeh1· MonirehKhadem1· RaminNabizadeh2· SeyedJamaleddinShahtaheri1
Received: 15 February 2018 / Accepted: 25 June 2018
© Institute of Chemistry, Slovak Academy of Sciences 2018
Abstract
A molecular imprinted polymer was prepared with precipitation polymerization technique and applied as a sorbent for
selective extraction and enrichment of metribuzin herbicide prior to high performance liquid chromatography. Optimiza-
tion of critical variables affecting the efficiency of molecularly imprinted solid-phase extraction (MISPE), such as sorbent
mass, sample pH and flow rate of sample, volume, concentration, and flow rate of elution solvent was done by employing
central composite design (CCD) of the response surface methodology. Two separate models were developed for the adsorp-
tion and recycling steps. The analysis of variance (ANOVA) demonstrated that, experimental data were excellently fitted to
the proposed response models. The optimum operating conditions were: a sorbent mass of 25mg, sample pH 6.19, sample
flow rate of 2.15mL/min, and a 5mL portion of methanol/acetic acid with 92.7:7.3 (v/v) ratio and flow rate of 2.1mL/min
for the extraction process. Under the optimized conditions, the linear range was obtained from 20 to 120µg/L (R2 = 0.999)
and the lowest detectable concentration (LOD) and the lowest quantitative concentration (LOQ) were calculated as 5.75
and 19.86µg/L, respectively. Finally, the designed MISPE method was successfully applied to determine trace amount of
metribuzin in real samples. The diluted urine samples were spiked with metribuzin at 4 levels and extracted with recoveries
ranging from 93.82 to 97.84% and the relative standard deviation (RSD) less than 4.8%.
Keywords Metribuzin· Molecular imprinted polymer· Central composite design· High performance liquid
chromatography· Solid phase extraction· Herbicide
Introduction
Pesticides as a group of toxic materials are used worldwide
in agricultural production to control pests and diseases as
well as maintain the product quality. Due to the biological
activities, pesticides may pose direct and indirect risks to
non-target individuals including human. There are serious
concerns about acute and chronic disorders resulting from
exposure to these compounds take place directly via the pro-
cess of their production, transportation, and consumption in
agricultural and household activities or indirectly through
their residuals in water and food (Khadem etal. 2016,
2017; Matsumura 1985; Mostafalou and Abdollahi 2013;
Rahiminejad etal. 2009; Zuskin etal. 2008).
Metribuzin is one of the triazinone herbicides used to
protect a number of agricultural products against board
leaves grasses (Janíková etal. 2016). It has been placed in
the moderately toxic group according to the classification of
pesticides recommended by the World Health Organization
(WHO) (WHO 2010). In addition to persistence in the soil,
it is also observed in the surface and ground waters due to
the high solubility in the aquatic environments (Kumar etal.
2013). Some evidence-based animal and human studies have
reported that the exposure to metribuzin may be related to
occurrence of DNA damage and genotoxic effects, changes
in the cellular function, lymphoid malignancies, metabolic
changes, and increased oxidative stress as well as changes
in the function of immune system, liver, and thyroid (Bleeke
etal. 1985; Calderón-Segura etal. 2007; Chiali etal. 2013;
* Seyed Jamaleddin Shahtaheri
shahtaheri@sina.tums.ac.ir
1 Department ofOccupational Health Engineering, School
ofPublic Health, Tehran University ofMedical Sciences,
Tehran, Iran
2 Department ofEnvironmental Health Engineering, School
ofPublic Health, Tehran University ofMedical Sciences,
Tehran, Iran
Author's personal copy

Chemical Papers
1 3
Delancey etal. 2009; Löser and Kimmerle 1972; Medjdoub
etal. 2011; Porter etal. 1993; Štěpánová etal. 2012).
Considering the above-mentioned facts, it is necessary
to determine the level of pesticides in different environmen-
tal and biological samples for assessment of the occurred
exposure or contamination. In this regard, there are some
traditional methods to measure the concentration of pes-
ticides including metribuzin, such as gas/liquid chroma-
tography (Gao etal. 2010; Li etal. 2013; Papadakis and
Papadopoulou-Mourkidou 2002; Xie etal. 2017), micellar
electrokinetic chromatography (Huertas-Pérez etal. 2006),
fourier transform infrared (Khanmohammadi etal. 2008),
and gas/liquid chromatography mass spectroscopy (Bichon
etal. 2006; Djozan and Ebrahimi 2008; Djozan etal. 2009;
Henriksen etal. 2002; Tian etal. 2014).
However, even assuming availability of such equipment
in common laboratories, due to some problems such as the
presence of interferences in the sample matrices and insuf-
ficient sensitivity of current detectors to diagnose the trace
amounts of some analytes, development of more selective
and sensitive sample preparation and analytical techniques
can promise effective strategy for their detection in trace
and ultra-trace levels (Alizadeh 2009; Lambropoulou and
Albanis 2007; Omidi etal. 2014a, b; Tankiewicz etal. 2010).
Considering the drawbacks of traditional sample prepara-
tion techniques, growing tendencies are generating toward
new approaches such as solid phase extraction (SPE) (Liu
etal. 2004; Valsamaki etal. 2007) and solid phase micro-
extraction (SPME) (Ghavidel etal. 2014; Sánchez-Ortega
etal. 2005), enabling automated handling of matrix and
subsequent retrieval of analyte. Also, many efforts have
been made to improve the selectivity of above mentioned
methods. Particularly, molecularly imprinted polymers
(MIPs) as new adsorbents with specific functions have been
synthesized and applied based on the immune system and
antibody mechanism (Pichon and Chapuis-Hugon 2008;
Rahiminezhad etal. 2010).
Molecular imprinted polymers are considered as specific
adsorbents with high stability and resistance in different
conditions, making them suitable for sample preparation
of lipophilic compounds such as polychlorinated biphenyls
(PCBs) and pesticides (Koohpaei etal. 2008; Muldoon and
Stanker 1995). Using a three dimensional polymeric network
imprinted in the same size and shape as the template mol-
ecule (target analyte), enables MIP to specifically adsorb and
extract desired compound from different samples (Koohpaei
etal. 2009; Turiel and Martín-Esteban 2010).
Following the primary introduction of MIPs, several
studies were performed in order to improve the functional
properties of these adsorbents in both synthesis and appli-
cation steps. Due to the disadvantages of traditional bulk
polymerization such as producing irregular particles and the
heterogeneous distribution of binding sites after grinding
and sieving process which result in poor site accessibility for
template molecule and lower specific surface area, alterna-
tive synthetic strategies such as suspension, multi-step swell-
ing, insitu, and precipitation polymerization were designed
and developed in order to control the size, shape, and poros-
ity of the particles in the polymerization process (Omidi
etal. 2014b; Yan and Row 2006). Among the mentioned
techniques, precipitation polymerization is the most con-
venient method for synthesizing monodisperse MIP particles
without the use of any surfactant or stabilizer (Xia etal.
2017). In recent years, the critical parameters of precipita-
tion polymerization have been investigated for producing a
nanostructure adsorbent. MIP nanoparticles obtain higher
surface area-to-volume ratios leading to the enhancement of
the binding capacity and recovery of analyte by using less
amount of adsorbent (Abouzarzadeh etal. 2012).
It seems that the combination of benefits of MIP nanopar-
ticles and the optimization of operational factors affecting
the molecular recognition properties is a logical approach
followed in many of current studies in order to improve sen-
sitivity and accuracy of the biological and environmental
sample preparation techniques. Several methods have been
reported for the analysis of metribuzin in different sam-
ples, Zhang etal. used molecularly imprinted polymer in
an on-line system for determination of metribuzin in soil
sample (Zhang etal. 2009). The detection limit and RSD
of the method were reported 8.3 × 10−4mg/kg and 3.2%,
respectively. In another attempt for quantitative analysis of
metribuzin in soil samples, Jia etal. prepared an electro-
chemical sensor via electro-polymerization of L-Norvaline
in which the detection limit and reproducibility of method
were reported 2.14 × 10−3μg/mL and 3.2%, respectively (Jia
etal. 2016). Breton etal. used the combination of biosen-
sor with cyanazine-imprinted polymer for the measurement
of triazines herbicides including metribuzin (Breton etal.
2006). The proposed method allowed sensitive detection
of photosynthesis-inhibiting herbicides in water samples
at the level required by European Union regulations. Also,
considering the drawbacks of traditional one-variable-at-
a-time optimization method such as failure to investigate
the interaction of variables and ignorance of their simulta-
neous effects on extraction efficiency, using chemometric
approaches such as response surface methodology (RSM)
could be an ideal opportunity for modeling and predicting
real conditions in the optimization process in a n-dimen-
sional matrix. Successful application of a factorial design
for evaluation and optimization of different parameters affect
the detection of metribuzin through a spectrophotometric
method reported by Shah etal. (2009). Under the optimized
condition, the recovery of analyte from real samples of pota-
toes was achieved in the range of 86.0% ± 0.9 to 91.7% ± 0.2.
The present study aimed to application of the molecular
imprinted polymer nanoparticles properties in combination
Author's personal copy

Chemical Papers
1 3
with the response surface methodology for selective and sen-
sitive extraction of the herbicide metribuzin from biological
samples (urine).
Experimental
Chemicals
Metribuzin standard was purchased from Sigma-Aldrich
(Germany) and methacrylic acid (MAA), ethylene glycol
dimethacrylate (EGDMA), and 2, 2 azobisisobutyronitrile
(AIBN) were obtained from Merck (Germany). All analyti-
cal grade solvents, sodium hydroxide, and buffer solutions
were purchased from Merck (Germany). Double-distilled
water was provided by Purite purification system.
Apparatus andanalytical conditions
All samples were analyzed using high performance liquid
chromatography system (Knauer, Germany)) equipped with
a K-1001 series high-pressure pump, a K-2006 ultraviolet
detector and a VS injection valve with a 20µL loop. Ana-
lytes were separated by a reverse phase Prontosil 120-5-C18
column (150 × 4.6mm, Germany), using isocratic metha-
nol/water mobile phase composition (70:30; v/v) at the flow
rate of 1mLmin−1 and finally detected by UV wavelength
set at 290nm. During the MIP synthesis and MISPE pro-
cedure, some other equipment were also applied including
microsampler (Socorex, Germany), digital scale (Sartorius,
Germany), ultrasonic bath (Sono, Swiss), magnetic stirrer
hotplate (Chiltern, USA), nitrogen supply system, digi-
tal thermometer (TP3001, China), reactor heater system
(Memmert, Germany), Soxhlet extractor (Duran, Germany),
digital pH meter (Metrohm, Switzerland), vacuum manifold
(Tajhizteb, Iran) equipped with a vacuum pump. The mor-
phological characteristics of the MIP nanoparticles were
investigated using field emission scanning electron micros-
copy (FE-SEM, Tescan, MIRA II, Czech Republic). Analy-
sis of chemical structure and binding properties of imprinted
and non-imprinted polymer nanoparticles were performed
by Fourier transform infrared spectrometer (FT-IR, Nicolet,
Magna-IR 550, USA).
Preparation ofthemolecular imprinted polymer (MIP)
The molecularly imprinted polymer nanoparticles were syn-
thesized using the template molecule, functional monomer,
and cross linker in the ratio of 1:4:20, respectively. First,
1mmol (214.2mg) of metribuzin (template) was mixed in
70mL of toluene with 4mmol (337µL) methacrylic acid
(functional monomer). After 40min, 20mmol (3.77mL)
EGDMA (cross-linker) and 40mg of AIBN (initiator) were
added to the reaction vessel. In order to remove oxygen,
the solution was purged with nitrogen for 7min and then
sealed under that atmosphere. The polymerization process
was performed in an oil bath at 55°C under stirring rate of
80rpm, for 18h.
After polymerization process, template molecules were
removed from the polymer structure by a soxhlet appara-
tus containing methanol/acetic acid (90:10, v/v) for 44h.
Finally, the particles were dried in an oven at 50°C. The
corresponding non-imprinted polymer (NIP), as the control
one, was prepared in the same manner without adding tem-
plate molecule.
Optimization ofMISPE procedure byexperimental
design approach
In order to optimize factors strongly affecting the adsorption
and desorption of metribuzin, six parameter including sorb-
ent mass, sample pH and flow rate of sample, volume and
flow rate of elution solvent, and also the amount of acid in
the composition of elution solvent were selected based on
preliminary studies and experiments.
The classical optimization methods, such as one variable
at a time, are time-consuming and they do not consider the
interaction of the constituent parameters. Due to these draw-
backs, the response surface statistical method was chosen to
design the optimization experiments and modulate the process
of adsorption and recycling of metribuzin by MIP. Accord-
ingly, the selected operational range of variables can be seen
in Table1. A full factorial (2n) central composite design
Table 1 Operational range of
input variables for experimental
design
MISPE phase Input variables Unit Symbol Levels
Lower Upper
Adsorption Sample pH – x13 11
Sorbent mass mg x25 25
Sample flow rate mL/min x31 3
Desorption Elution solvent volume mL x11 5
Amount of acid in elution solvent %, v/v x21 10
Elution flow rate mL/min x31 3
Author's personal copy

Chemical Papers
1 3
(CCD) of orthogonal type with 9 central and 6 axial points
was established via R statistical software package wherein the
rotatability of design was ensured with α = 1.682. Based on
the proposed design, the effective variables were separately
investigated in two set of 29-run experiments for adsorption
and recycling of metribuzin in the MISPE procedure.
Based on the designed experiments, 1–10mg of MIP (X1
factor in Table2) was packed into a SPE cartridge (length of
65mm and i.d., 5mm) between two frits. The prepared MISPE
column was conditioned consecutively with 1mL of methanol
and double-distilled water at a flow rate of 1mL/min. Then
30mL of 5mg/L metribuzin solution was passed through the
cartridge at the specified pH and flow rate followed by inject-
ing to the HPLC for analysis. Eventually, the adsorption effi-
ciency of MIP was calculated by the following equation, where
Ca and Cb are the concentrations of metribuzin in the solution
before and after adsorption, respectively.
To optimize the extraction of metribuzin from the MISPE
column, experiments were performed in 29 runs designed by
R software. Solutions with defined concentration were loaded
onto the column considering the adsorption operational fac-
tors, and then a washing step was done using 2mL of deion-
ized water. Afterward, 1–5mL of methanol containing 1–10%
acetic acid (accurately mentioned as X1 and X2 factors in
Table3) was passed through the cartridge as elution solvent
and it was analyzed by HPLC. The percentage of extracted
metribuzin was calculated through the following equation:
Adsorption efficiency (%)=(Ca−Cb∕Ca)×100
Extraction efficiency(%)=(Cc∕Cd)×100
where Cc and Cd are the metribuzin concentrations in load-
ing solution and eluent, respectively.
Preparation ofsolutions
The stock solution of 1000mg/L metribuzin was prepared in
acetonitrile and kept at −18°C. The required standard solu-
tions were prepared daily by diluting different amounts of
stock solution with double-distilled water. In order to inves-
tigate the ability of designed MISPE procedure for extraction
of metribuzin from real samples, 2mL of urine samples
taken from unexposed persons was diluted in the ratio of
1:5 and spiked with different concentrations of metribuzin
after pH adjustment.
Results anddiscussion
Characterization ofmolecular imprinted
nanoparticles
The morphology and particle size of synthesized MIP were
surveyed using field emission scanning electron micros-
copy (FE-SEM). As it can be seen in Fig.1, spherical par-
ticles with diameter ranging from 15.44 to 28.67nm were
achieved using the precipitation polymerization technique.
Furthermore, to investigate the chemical structure of the
synthesized polymer and also ensure the effective template
removal from MIP structure, FT-IR spectroscopic analysis
of unleached and leached metribuzin imprinted polymer was
done. According to Fig.2a, b, good correlation was obtained
between frequencies of leached and unleached polymers,
Table 2 Central composite
design matrix and the observed
response in adsorption step
SD standard deviation for n = 3
Run X1X2X3Retention (%)
(mean ± SD)
Run X1X2X3Retention (%)
(mean ± SD)
1 7 15 2 83.87 ± 0.17 16 4.62 9.05 2.6 67.54 ± 0.69
2 4.62 9.05 1.4 77.22 ± 0.95 17 9.38 9.05 1.4 63.25 ± 0.73
3 7 15 2 80.91 ± 0.78 18 7 15 3 68.01 ± 0.36
4 9.38 20.95 2.6 75.65 ± 0.01 19 7 15 2 81.69 ± 0.19
5 7 15 2 82.68 ± 0.46 20 7 15 2 79.91 ± 0.10
6 7 15 2 80.75 ± 0.47 21 11 15 2 58.41 ± 0.77
7 7 15 2 84.52 ± 0.41 22 7 15 2 82.69 ± 0.42
8 9.38 20.95 1.4 84.72 ± 0.61 23 7 25 2 97.11 ± 0.09
9 4.62 20.95 2.6 96.46 ± 0.06 24 7 5 2 53.16 ± 0.06
10 7 15 2 82.25 ± 1.03 25 7 15 2 82.99 ± 0.45
11 9.38 9.05 2.6 44.57 ± 1.05 26 7 15 1 86.7 ± 0.03
12 7 15 2 80.95 ± 0.36 27 7 15 2 81.92 ± 0.42
13 7 15 2 81.17 ± 0.65 28 7 15 2 81.89 ± 0.13
14 4.62 20.95 1.4 99.71 ± 0.23 29 3 15 2 89.54 ± 0.25
15 7 15 2 81.54 ± 0.20 – – – – –
Author's personal copy

Chemical Papers
1 3
indicating the similarity of their structures. The displace-
ment of carboxylic acid O–H stretch vibration in the leached
MIP (2957cm−1) toward the lower frequencies can prove
the presence of interaction between hydrogen and template
molecules in unleached polymer (2989cm−1). The IR spec-
tra of the unleached MIP illustrates that the absorptions may
be attributed to C=O (1731cm−1) and C–O (1260cm−1)
from EGDMA and MMA. Totally, comparison of occurred
displacements of peaks in leached and unleached polymers
confirms the interaction between functional monomer and
template molecule. The peaks in the fingerprint region
(beyond 1500cm−1) are complicated to interpret because of
the overlapping many bands.
Optimization oftheMISPE
The obtained results for the metribuzin adsorption and recy-
cling efficiency through the MISPE procedure have been
shown in Tables2 and 3 for all 58 CCD runs. In order to
assess the fitting of experimental results with the RS models,
the main, quadratic, and interaction relationships between
output responses and input factors were investigated by
developing a second-order polynomial model based on the
following empirical equation:
where
̂
R
denotes the predicted response of the process, xi
refers to the coded levels of the factors (independent or con-
trol variables), b0, bi, bii, bij are the regression coefficients,
and
𝜀
is the statistical error.
The adequacy of the reduced quadratic models (where
the non-significant terms were removed) was checked by
the statistical estimators obtained from analysis of vari-
ance (ANOVA) (Tables4 and 5). In this regard, F value
indicator, a measure of the variance of data, was consid-
ered to diagnose the statistical significance of the model.
Based on the results presented in Tables4 and 5, the
�
R
=b0+
n
∑
i=1
bixi+
n
∑
i=1
biix2
i+
n
∑
i
<
j
bijxixj+
𝜀
Table 3 Central composite
design matrix and the observed
response in desorption step
SD standard deviation for n = 3
Run X1X2X3Recovery (%)
(mean ± SD)
Run X1X2X3Recovery (%)
(mean ± SD)
1 1.8 8.17 2.59 94.90 ± 2.83 16 4.19 8.17 2.59 98.03 ± 1.21
2 3 5.5 2 88.20 ± 0.31 17 1.8 2.82 1.4 79.28 ± 1.85
3 3 5.5 2 88.44 ± 0.21 18 3 5.5 2 89.57 ± 0.44
4 4.19 8.17 1.4 89.06 ± 0.01 19 3 5.5 2 89.74 ± 0.31
5 3 5.5 2 88.59 ± 0.61 20 3 5.5 2 89.99 ± 0.62
6 1.8 8.17 1.4 87.80 ± 0.02 21 3 10 2 95.47 ± 2.85
7 4.19 2.8 1.4 85.99 ± 1.33 22 3 5.5 1 84.33 ± 0.66
8 3 5.5 2 88.70 ± 0.58 23 1 5.5 2 87.76 ± 0.42
9 3 5.5 2 88.88 ± 1.08 24 3 5.5 2 90.29 ± 1.25
10 3 5.5 2 88.94 ± 0.39 25 3 1 2 82.29 ± 2.72
11 3 5.5 2 89.10 ± 1.70 26 3 5.5 2 90.55 ± 0.44
12 3 5.5 2 89.28 ± 0.76 27 3 5.5 2 90.44 ± 0.30
13 3 5.5 2 87.89 ± 1.11 28 5 5.5 2 90.60 ± 0.43
14 4.19 2.82 2.59 89.98 ± 0.92 29 3 5.5 3 93.44 ± 2.76
15 1.8 2.82 2.59 86.34 ± 1.00 –––––
Fig. 1 FE-SEM image of prepared MIP nanoparticles
Author's personal copy

Chemical Papers
1 3
reported F values for selected models in the adsorption
and recycling steps considerably depart from unity and
represent high levels, indicating a reliable model predic-
tion from the empirical data. On the other hand, the low p
Fig. 2 FT-IR spectra of washed
(a) and unwashed (b) MIP
nanoparticles
Table 4 Analysis of variance (ANOVA) for the reduced response sur-
face quadratic model (adsorption step)
df degree of freedom
Model term df Sum of squares Mean square F value P value
x11 1145.8 1145.8 842.23 <0.0001
x21 2316.7 2316.7 1702.9 <0.0001
x31 380.6 380.6 279.77 <0.0001
x1
21 152.1 152.1 111.77 <0.0001
x2
21 94 94 69.12 <0.0001
x3
21 34.8 34.8 25.56 <0.0001
x1:x31 27.5 27.5 20.18 <0.0001
x2:x31 32.2 32.2 23.64 <0.0001
Residuals 20 27.2 1.4 – –
Lack of fit 6 6.1 1.02 0.6813 0.667649
Multiple R2 = 0.9935; Adjusted R2 = 0.991
Table 5 Analysis of variance (ANOVA) for the reduced response sur-
face quadratic model (desorption step)
df degree of freedom
Model term df Sum of squares Mean square F value P value
x11 27.95 27.95 34.062 <0.0001
x21 185.82 185.82 226.432 <0.0001
x31 131.8 131.8 160.605 <0.0001
x1:x21 4.5 4.5 5.483 0.0282
x2:x31 3.19 3.19 3.893 0.0606
Residuals 23 18.87 0.82 – –
Lack of fit 9 9.2 1.022 1.4778 0.24712
Multiple R2 = 0.9493; Adjusted R2 = 0.9383
Author's personal copy

Chemical Papers
1 3
values in both phases represent the statistical validity of
the models to predict the response.
The quality of fit for the polynomial model equation is
expressed by the ANOVA coefficient of determination (R2
and adjusted R2). R2, a measure of variation in the response,
can be explained by predictors in the model. According to
the results, a desirable R2 value, close to 1, was obtained in
both adsorption and recycling models. Furthermore, the pre-
dicted R2 values were in agreement with the adjusted coef-
ficient of determination
(
R
2
adj)
which indicates that factors
affecting the efficiency of the MISPE procedure have been
selected correctly. Eventually, since the P value for lack of
fit in the ANOVA tables is higher than 0.05, the model
seems to be acceptable for data at the selected confidence
level.
In the next step, using the Multivariate Regression
method, the following empirical models in terms of coded
factors and based on the significant regression coefficients
(P < 0.05) in the reduced quadratic models were developed
for the prediction of metribuzin adsorption and recycling
efficiency through the MISPE procedure:Adsorption step:
Desorption step:
where
̂
Y
denotes the predicted metribuzin adsorption/recy-
cling efficiency (%) and x1, x2 and x3 are the coded values of
the independent variables.
The behavior of the response surface models was pre-
sented graphically by means of two-dimensional contour
plots where the interaction effects of the operational vari-
ables have been mapped against the response factors. Sample
pH is one of the most important parameters strongly affect-
ing the adsorption of analytes by MIPs, so the effect of its
interactive relation with the other two independent variables
on the adsorption yield is illustrated in Fig.3a and b. As it
can be seen, decreasing sample pH to lower levels enhances
the adsorption of metribuzin by MIP. Regarding the nature
of hydrogen bonding, it is expected that, the efficiency of
the adsorption process to be improved in a neutral to acidic
range. Under the basic condition, the hydrolysis reaction
causes to reduce the ability of sorbent’s functional groups
for effective bonding with the analyte of interest.
Figure3a and c represents the interactive effect of sorbent
mass with sample pH and flow rate. As shown in Fig.3a,
increasing sorbent mass from 5 to 25mg in acidic condition
enhances the adsorption of metribuzin up to maximum. It
is also clear from Fig.3c that, more than 95% metribuzin
̂
Y
=39.82 +5.2x1+2.93x2+7.8x3−0.46x2
1−0.06x
2
2
−3.99x2
3
−1.29x
1
x
3
+0.56x
2
x
3
,
̂
Y
=67.98 +2.48x
1
+1.28x
2
+3.04x
3
−0.23x
1
x
3
+0.39x
2
x
3,
adsorption can be achieved up to 25mg of sorbent mass
even at higher sample flow rates, which can be attributed to
the increase of specific binding sites and adsorption capac-
ity in higher amounts of adsorbent. It is worth mentioning
that, synthesis of the MIP in nano-scaled particles provides
much more specific binding sites per mass unit of adsorbent,
resulting in the higher adsorption and recycling efficiencies.
The effect of the sample flow rate was also investigated in
the present study. As outlined in Fig.3b and c, the adsorp-
tion of metribuzin particularly occurred at the lower level
of the predicted range (1–3ml/min) for sample flow rate in
correlation with the other operational variables. Therefore,
in accordance with some other studies (Wang etal. 2015;
Zarejousheghani etal. 2014), it seems that reducing the sam-
ple flow rate to an appropriate value can improve the analyte
absorbance in the MISPE process.
As expressed previously, a combinatorial effect of vari-
ables including volume and flow rate of elution solvent as
well as the amount of acid in the solvent composition on the
extraction of metribuzin was studied in the desorption step.
According to Fig.3d–f, higher level of each operational vari-
able in desorption step with interaction to the other variables
represents positive impact on the extraction yield. Also, the
greater impact of volume and composition of elution sol-
vent (the amount of acid added as an extractor modifier) in
comparison with the elution flow rate on the desorption of
metribuzin was observed in Fig.3e and f. The maximum
extraction yield is achievable unexpectedly in higher elution
flow rate when the other two variables are applied at upper
levels of the predicted ranges.
Finally, based on the obtained regressional models, the
optimum values of selected parameters were calculated in
Solver add-in of Microsoft Excel 2013 software with respect
to the primary optimization ranges (Table1) and considering
the desired efficiency. The optimum operating conditions
were determined as follows: sorbent mass of 25mg, sample
pH 6.19, sample flow rate of 2.15mL/min, as well as a 5mL
portion of methanol/acetic acid with ratio of 92.7:7.3 (v/v)
and flow rate of 2.1mL/min for the extraction process.
Comparison oftheprepared MIP andNIP
In order to prove the superiority of prepared MIP in rebind-
ing process in comparison with corresponding NIP, the
adsorption behavior of MIP and NIP for metribuzin was
investigated by applying 20mL of 5mg/L metribuzin solu-
tion for the adsorption procedure under optimized condi-
tion and calculating the adsorption capacity. The results
of three replicate measurements indicated that there was a
significant difference between the adsorption efficiencies of
MIP (93.34 ± 0.29%) and NIP (32.23 ± 3.12%) toward the
Author's personal copy

Chemical Papers
1 3
Author's personal copy

Chemical Papers
1 3
metribuzin, confirming the formation of selective adsorption
sites in the MIP structure.
Eect ofinterferences
The selectivity of synthesized MIP for metribuzin herbicide
was examined through an assessment of its function in the
presence of other pesticides. In this regard, different concen-
trations of permethrin, diazinon, and malathion were added
to 50µg/L metribuzin solution. Subsequently, the adsorp-
tion and recycling processes of each sample were performed
under the optimum conditions. The obtained results show
an extraction efficiency of 94.60 ± 1.84 and 92.99 ± 1.90%
(n = 3) for metribuzin in the presence of 50- and 100-fold of
mentioned pesticides, respectively, indicating non-signifi-
cant effect of matrix interferences on sensitive and selective
performance of prepared MIP nanoparticles.
Method validation
Under the optimized condition, calibration curves were
plotted in the range 20–120 and 50–110µg/L for water
and urine samples, respectively. A linear relationship was
observed between metribuzin concentration and chromato-
grams peaks area (y = 0.157x + 0.173; R2 = 0.999 (for water),
y = 0.161x − 0.553; R2 = 0.9943 (for urine)). The limits of
detection (LOD, water samples: 5.75 and urine samples:
12.48μg/L) as well as the limits of quantification (LOQ,
water samples: 19.18 and urine samples: 41.61μg/L) were
also calculated based on the following equations:
where
Sb
is the standard deviation of the blank and m is the
slope of the calibration graph.
Furthermore, in order to assess the reproducibility of the
proposed method, six consecutive experiments were per-
formed for a single metribuzin concentration (100μg/L)
during 1day. The relative standard deviation (RSD) of
3.58% was obtained which shows the satisfactory perfor-
mance of the optimized MISPE method for determination
of metribuzin.
The comparison of the LOD and RSD values obtained by
the proposed method to other reported methods for pretreat-
ment and determination of metribuzin is given in Table6.
As it can be seen, the sensitivity provided by the MISPE-
HPLC procedure is clearly higher than other detection tech-
niques and the use of nano-scaled MIP coupled with liquid
chromatography separation features resulted in a very low
detection limit.
Extraction ofmetribuzin fromspiked urine
To evaluate the applicability of the proposed MISPE tech-
nique for analysis of complex real samples containing
different interferences, various amounts of metribuzin
LOD
=
3S
b
m
LOQ =
10S
b
m
Fig. 3 Response surface contour plots for metribuzin adsorption and
desoption yield as a function of mutual interaction between: a sample
pH and sorbent mass; b sample pH and sample flow rate; c sorbent
mass and sample flow rate; d elution solvent volume and the amount
of acid in elution solvent; e elution solvent volume and elution flow
rate; f the amount of acid in elution solvent and elution flow rate
◂
Table 6 Comparison of the proposed method with the other methods for determination of metribuzin and its degradations
Studied samples Method LOD (µg/L) Reproducibility (%) Ref.
Commercial formulations
and potato
Spectrophotometric method 660 Not reported Shah etal. (2009)
Body fluids (whole blood
and urine)
Solid-phase microextraction
and capillary gas chroma-
tography
0.4-2.0 (urine sample)
5.6–18 (whole blood)
10.3 (urine sample)
14.2 (whole blood)
Kumazawa etal. (2000)
Degradation products of
metribuzin in water and
soil
Pressurized liquid extrac-
tion and capillary zone
electrophoresis
10, 10 and 20 for deami-
nometribuzin, deam-
inodiketometri-buzin
and diketometribuzin,
respectively
2.5, 1 and 3.2 for deami-
nometribuzin, deam-
inodiketometri-buzin
and diketometribuzin,
respectively
Quesada-Molina etal.
(2007)
River and irrigation water Solid-phase extraction and
differential pulse adsorp-
tive stripping voltammetry
(DPAdSV)
0.27 2.53 and 3.66 for 2 and
6µgL−1 metribuzin,
respectively
Skopalová etal. (2001)
Urine sample Molecularly imprinted
solid-phase extraction
(MISPE) and high perfor-
mance liquid chromatog-
raphy
12.48 3.2 This study
Author's personal copy

Chemical Papers
1 3
were spiked in 10mL of diluted urine samples. Then,
the optimized MISPE method was applied to analyze the
metribuzin. As it can be seen in Table7, the recovery of
metribuzin from spiked samples with different concen-
trations was sufficiently acceptable, demonstrating the
reliable, selective, and accurate performance of synthe-
sized MIP to clean up the complex samples, especially
biological ones.
Conclusion
In this study, MIP nanoparticles were prepared by precipita-
tion polymerization technique and they successfully applied
as the SPE sorbent for selective extraction of metribuzin
herbicide prior to its determination by high-performance
liquid chromatography. The parameters affecting the effi-
ciency of MISPE were separately optimized in adsorption
and recycling steps by response surface modeling approach.
Development of analytical methods is promising to detect
the given analytes in complex matrices. In this respect, the
use of MIPs, as advanced adsorbents, can greatly improve
the selectivity and sensitivity of the method. In conclu-
sion, the proposed MISPE-HPLC method can potentially
applied for detection of metribuzin in real samples with no
special sample pretreatment steps. It is worth mentioning
that, experimental design and modern modeling approach
can be successfully applied to optimize the experimental
conditions, to estimate any interaction between the factors,
and to obtain more satisfactory results compared to one-at-
a-time approach.
Acknowledgements This research has been supported by Tehran Uni-
versity of Medical Sciences grant (Project no. 32438). The authors
acknowledge the University and also the laboratory personnel of occu-
pational health department for all valuable supports.
Funding The authors received no specific funding for this work.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
Human and animal rights statement This article does not contain any
studies with human participants or animals performed by any of the
authors.
References
Abouzarzadeh A, Forouzani M, Jahanshahi M, Bahramifar N (2012)
Synthesis and evaluation of uniformly sized nalidixic acid–
imprinted nanospheres based on precipitation polymerization
method for analytical and biomedical applications. J Mol Recognit
25:404–413. https ://doi.org/10.1002/jmr.2201
Alizadeh T (2009) High selective parathion voltammetric sensor devel-
opment by using an acrylic based molecularly imprinted polymer-
carbon paste electrode. Electroanalysis 21:1490–1498. https ://doi.
org/10.1002/elan.20080 4541
Bichon E, Dupuis M, Le Bizec B, André F (2006) LC–ESI-MS/MS
determination of phenylurea and triazine herbicides and their
dealkylated degradation products in oysters. J Chromatogr B
838:96–106. https ://doi.org/10.1016/j.jchro mb.2006.04.019
Bleeke MS, Smith MT, Casida JE (1985) Metabolism and toxicity of
metribuzin in mouse liver. Pestic Biochem Physiol 23:123–130.
https ://doi.org/10.1016/0048-3575(85)90086 -0
Breton F, Euzet P, Piletsky S, Giardi M, Rouillon R (2006) Integration
of photosynthetic biosensor with molecularly imprinted polymer-
based solid phase extraction cartridge. Anal Chim Acta 569:50–
57. https ://doi.org/10.1016/j.aca.2006.03.086
Calderón-Segura ME etal (2007) Metabolic activation of herbicide
products by Vicia faba detected in human peripheral lymphocytes
using alkaline single cell gel electrophoresis. Toxicol InVitro
21:1143–1154. https ://doi.org/10.1016/j.tiv.2007.03.002
Chiali F, Merzouk H, Merzouk S, Medjdoub A, Narce M (2013)
Chronic low level metribuzin exposure induces metabolic
alterations in rats. Pest Biochem Physiol 106:38–44. https ://doi.
org/10.1016/j.pestb p.2013.03.011
Delancey JOL, Alavanja MC, Coble J, Blair A, Hoppin JA, Austin
HD, Freeman LEB (2009) Occupational exposure to metribu-
zin and the incidence of cancer in the agricultural health study.
Ann Epidemiol 19:388–395. https ://doi.org/10.1016/j.annep
idem.2008.12.018
Djozan D, Ebrahimi B (2008) Preparation of new solid phase micro
extraction fiber on the basis of atrazine-molecular imprinted poly-
mer: application for GC and GC/MS screening of triazine her-
bicides in water, rice and onion. Anal Chim Acta 616:152–159.
https ://doi.org/10.1016/j.aca.2008.04.037
Djozan D, Mahkam M, Ebrahimi B (2009) Preparation and binding
study of solid-phase microextraction fiber on the basis of ametryn-
imprinted polymer: application to the selective extraction of per-
sistent triazine herbicides in tap water, rice, maize and onion. J
Chromatogr A 1216:2211–2219. https ://doi.org/10.1016/j.chrom
a.2008.12.101
Gao S etal (2010) Determination of phenylurea and triazine herbicides
in milk by microwave assisted ionic liquid microextraction high-
performance liquid chromatography. Talanta 82:1371–1377. https
://doi.org/10.1016/j.talan ta.2010.07.002
Ghavidel F, Shahtaheri SJ, Jazani RK, Torabbeigi M, Froushani AR,
Khadem M (2014) Optimization of solid phase microextraction
procedure followed by gas chromatography with electron capture
Table 7 Recovery of metribuzin from spiked urine samples under
optimal condition
RSD relative standard deviation
Real Sample Concentration of
analyte (ppb)
Relative recovery (%) RSD (%)
(n = 3)
Urine 50 93.82 ± 4.08 4.34
70 97.18 ± 4.18 4.3
90 94.35 ± 4.53 4.8
110 97.84 ± 2.68 2.73
Author's personal copy

Chemical Papers
1 3
detector for pesticides butachlor and chlorpyrifos. Am J Anal
Chem 5:535. https ://doi.org/10.4236/ajac.2014.59061
Henriksen T, Svensmark B, Juhler RK (2002) Analysis of metribu-
zin and transformation products in soil by pressurized liquid
extraction and liquid chromatographic–tandem mass spectrom-
etry. J Chromatogr A 957:79–87. https ://doi.org/10.1016/S0021
-9673(01)01453 -4
Huertas-Pérez JF, del Olmo Iruela M, García-Campaña AM, González-
Casado A, Sánchez-Navarro A (2006) Determination of the herbi-
cide metribuzin and its major conversion products in soil by micel-
lar electrokinetic chromatography. J Chromatogr A 1102:280–286.
https ://doi.org/10.1016/j.chrom a.2005.10.052
Janíková L, Šelešovská R, Rogozinská M, Tomášková M, Chýlková J
(2016) Sensitive voltammetric method for determination of her-
bicide metribuzin using silver solid amalgam electrode. Monatsh
Chem Chem Mon 147:219–229. https ://doi.org/10.1007/s0070
6-015-1555-z
Jia D, Wang L, Gao Y, Zou L, Ye B (2016) Electrochemical behavior
of metribuzin based on L-Norvaline modified electrode and its
sensitive determination. J Electroanal Chem 764:56–63. https ://
doi.org/10.1016/j.jelec hem.2016.01.016
Khadem M, Faridbod F, Norouzi P, Foroushani AR, Ganjali MR,
Shahtaheri SJ (2016) Biomimetic electrochemical sensor based
on molecularly imprinted polymer for dicloran pesticide deter-
mination in biological and environmental samples. J Iran Chem
Soc 13:2077–2084. https ://doi.org/10.1007/s1373 8-016-0925-8
Khadem M, Faridbod F, Norouzi P, Rahimi Foroushani A, Ganjali MR,
Shahtaheri SJ, Yarahmadi R (2017) Modification of carbon paste
electrode based on molecularly imprinted polymer for electro-
chemical determination of diazinon in biological and environmen-
tal samples. Electroanalysis 29:708–715. https ://doi.org/10.1002/
elan.20160 0293
Khanmohammadi M, Armenta S, Garrigues S, de la Guardia M (2008)
Mid-and near-infrared determination of metribuzin in agrochem-
icals. Vib Spectrosc 46:82–88. https ://doi.org/10.1016/j.vibsp
ec.2007.10.005
Koohpaei A, Shahtaheri S, Ganjali M, Forushani AR, Golbabaei F
(2008) Application of multivariate analysis to the screening of
molecularly imprinted polymers (MIPs) for ametryn. Talanta
75:978–986. https ://doi.org/10.1016/j.talan ta.2007.12.046
Koohpaei A, Shahtaheri S, Ganjali M, Forushani AR, Golbabaei F
(2009) Optimization of solid-phase extraction using developed
modern sorbent for trace determination of ametryn in environ-
mental matrices. J Hazard Mater 170:1247–1255. https ://doi.
org/10.1016/j.jhazm at.2009.05.125
Kumar Y, Singh N, Singh S (2013) Removal of atrazine, metribuzin,
metolachlor and alachlor by granular carbon. J Environ Anal Toxi-
col 3:1–5. https ://doi.org/10.4172/2161-0525.10001 96
Kumazawa T etal (2000) Determination of triazine herbicides in
human body fluids by solid-phase microextraction and capillary
gas chromatography. Chromatographia 52:195–199. https ://doi.
org/10.1007/BF024 90456
Lambropoulou DA, Albanis TA (2007) Methods of sample prepara-
tion for determination of pesticide residues in food matrices by
chromatography–mass spectrometry-based techniques: a review.
Anal Bioanal Chem 389:1663–1683. https ://doi.org/10.1007/
s0021 6-007-1348-2
Li P, Yang X, Miao H, Zhao Y, Liu W, Wu Y (2013) Simultaneous
determination of 19 triazine pesticides and degradation products
in processed cereal samples from Chinese total diet study by iso-
tope dilution–high performance liquid chromatography–linear ion
trap mass spectrometry. Anal Chim Acta 781:63–71. https ://doi.
org/10.1016/j.aca.2013.04.027
Liu R, Zhou J, Wilding A (2004) Simultaneous determination of endo-
crine disrupting phenolic compounds and steroids in water by
solid-phase extraction–gas chromatography–mass spectrometry.
J Chromatogr A 1022:179–189. https ://doi.org/10.1016/j.chrom
a.2003.09.035
Löser E, Kimmerle G (1972) Acute and subchronic toxicity of Sencor
active ingredient. Pflanzenschutz Nachrichten 25:186–209
Matsumura F (1985) Metabolism of insecticides by animals and plants.
In: Matsumura F (ed) Toxicology of insecticides. Springer, Bos-
ton, pp 203–298. https ://doi.org/10.1007/978-1-4613-2491-1_5
Medjdoub A, Merzouk S, Merzouk H, Chiali F, Narce M (2011) Effects
of Mancozeb and Metribuzin on invitro proliferative responses
and oxidative stress of human and rat spleen lymphocytes stimu-
lated by mitogens. Pest Biochem Physiol 101:27–33. https ://doi.
org/10.1016/j.pestb p.2011.06.002
Mostafalou S, Abdollahi M (2013) Pesticides and human chronic
diseases: evidences, mechanisms, and perspectives. Toxi-
col Appl Pharmacol 268:157–177. https ://doi.org/10.1016/j.
taap.2013.01.02
Muldoon MT, Stanker LH (1995) Polymer synthesis and characteriza-
tion of a molecularly imprinted sorbent assay for atrazine. J Agric
Food Chem 43:1424–1427. https ://doi.org/10.1021/jf000 54a00 2
Omidi F, Behbahani M, Abandansari HS, Sedighi A, Shahtaheri SJ
(2014a) Application of molecular imprinted polymer nanopar-
ticles as a selective solid phase extraction for preconcentration
and trace determination of 2, 4-dichlorophenoxyacetic acid in the
human urine and different water samples. J Environ Health Sci
Eng 12:137. https ://doi.org/10.1186/s4020 1-014-0137-z
Omidi F, Behbahani M, Samadi S, Sedighi A, Shahtaheri SJ (2014b)
Coupling of molecular imprinted polymer nanoparticles by high
performance liquid chromatography as an efficient technique for
sensitive and selective trace determination of 4-chloro-2-meth-
ylphenoxy acetic acid in complex matrices. Iran J Public Health
43:645
Papadakis EN, Papadopoulou-Mourkidou E (2002) Determination of
metribuzin and major conversion products in soils by microwave-
assisted water extraction followed by liquid chromatographic
analysis of extracts. J Chromatogr A 962:9–20. https ://doi.
org/10.1016/S0021 -9673(02)00537 -X
Pichon V, Chapuis-Hugon F (2008) Role of molecularly imprinted pol-
ymers for selective determination of environmental pollutants—
a review. Anal Chim Acta 622:48–61. https ://doi.org/10.1016/j.
aca.2008.05.057
Porter W, Green S, Debbink N, Carlson I (1993) Groundwater pes-
ticides: interactive effects of low concentrations of carbamates
aldicarb and methomyl and the triazine metribuzin on thyroxine
and somatotropin levels in white rats. J Toxicol Environ Health
40:15–34. https ://doi.org/10.1080/15287 39930 95317 73
Quesada-Molina C, García-Campaña AM, del Olmo-Iruela L, del
Olmo M (2007) Large volume sample stacking in capillary zone
electrophoresis for the monitoring of the degradation products of
metribuzin in environmental samples. J Chromatogr A 1164:320–
328. https ://doi.org/10.1016/j.chrom a.2007.06.076
Rahiminejad M, Shahtaheri S, Ganjali M, Forushani A, Golbabaei
F (2009) Molecularly imprinted solid phase extraction for trace
analysis of diazinon in drinking water. J Environ Health Sci Eng
6:97–106. http://hdl.handl e.net/1807/62219
Rahiminezhad M, Shahtaheri S, Ganjali M, Koohpaei A, Forushani
AR, Golbabaei F (2010) An experimental investigation of the
molecularly imprinted polymers as tailor-made sorbents of
diazinon. J Anal Chem 65:694–698. https ://doi.org/10.1134/
S1061 93481 00700 63
Sánchez-Ortega A, Sampedro M, Unceta N, Goicolea M, Barrio R
(2005) Solid-phase microextraction coupled with high perfor-
mance liquid chromatography using on-line diode-array and elec-
trochemical detection for the determination of fenitrothion and its
main metabolites in environmental water samples. J Chromatogr
A 1094:70–76. https ://doi.org/10.1016/j.chrom a.2005.07.089
Author's personal copy

Chemical Papers
1 3
Shah J, Jan MR, Ara B, Mohammad M (2009) Extractive spectropho-
tometric method for determination of metribuzin herbicide and
application of factorial design in optimization of various factors.
J Hazard Mater 164:918–922. https ://doi.org/10.1016/j.jhazm
at.2008.08.100
Skopalová J, Lemr K, Kotouček M, Čáp L, Barták P (2001) Electro-
chemical behavior and voltammetric determination of the herbi-
cide metribuzin at mercury electrodes. Fresenius J Anal Chem
370:963–969. https ://doi.org/10.1007/s0021 60100 845
Štěpánová S, Doleželová P, Plhalová L, Prokeš M, Maršálek P, Škorič
M, Svobodová Z (2012) The effects of metribuzin on early life
stages of common carp (Cyprinus carpio). Pestic Biochem Physiol
103:152–158. https ://doi.org/10.1016/j.pestb p.2012.04.013
Tankiewicz M, Fenik J, Biziuk M (2010) Determination of organo-
phosphorus and organonitrogen pesticides in water samples TrAC.
Trends Anal Chem 29:1050–1063. https ://doi.org/10.1016/j.
trac.2010.05.008
Tian M, Cheng R, Ye J, Liu X, Jia Q (2014) Preparation and evalu-
ation of ionic liquid-calixarene solid-phase microextraction
fibres for the determination of triazines in fruit and vegetable
samples. Food Chem 145:28–33. https ://doi.org/10.1016/j.foodc
hem.2013.08.029
Turiel E, Martín-Esteban A (2010) Molecularly imprinted polymers for
sample preparation: a review. Anal Chim Acta 668:87–99. https ://
doi.org/10.1016/j.aca.2010.04.019
Valsamaki VI, Sakkas VA, Albanis TA (2007) Determination of the
pesticides considered as endocrine-disrupting compounds (EDCs)
by solid-phase extraction followed by gas chromatography with
electron capture and mass spectrometric detection. J Sep Sci
30:1936–1946. https ://doi.org/10.1002/jssc.20060 0487
Wang Q, Zhang X, Xu Z, Gao H (2015) Simultaneous determination
of three trace organophosphorus pesticide residues in vegetables
using molecularly imprinted solid-phase extraction coupled with
high-performance liquid chromatography. Food Anal Methods
8:2044–2051. https ://doi.org/10.1007/s1216 1-014-0086-2
WHO (2010) The WHO recommended classification of pesticides by
hazard and guidelines to classification 2009. http://who.int/ipcs/
publi catio ns/pesti cides _hazar d_2009.pdf
Xia Q, Yun Y, Li Q, Huang Z, Liang Z (2017) Preparation and char-
acterization of monodisperse molecularly imprinted polymer
microspheres by precipitation polymerization for kaempferol.
Des Monomers Polym 20:201–209. https ://doi.org/10.1080/15685
551.2016.12391 74
Xie Y-L, Zhao Z-D, Zhang X-L, Tang L-L, Zhang Y, Zhang C-H
(2017) Simultaneous analysis of herbicide metribuzin and its
transformation products in tomato using QuEChERS-based gas
chromatography coupled to a triple quadrupole mass analyzer.
Microchem J 133:468–473. https ://doi.org/10.1016/j.micro
c.2017.04.025
Yan H, Row KH (2006) Characteristic and synthetic approach of
molecularly imprinted polymer. Int J Mol Sci 7:155–178. https ://
doi.org/10.3390/i7050 155
Zarejousheghani M, Fiedler P, Möder M, Borsdorf H (2014) Selec-
tive mixed-bed solid phase extraction of atrazine herbicide
from environmental water samples using molecularly imprinted
polymer. Talanta 129:132–138. https ://doi.org/10.1016/j.talan
ta.2014.05.034
Zhang S, Yang G, Zheng Z, Chen Y (2009) On-line preconcentration
and analysis of metribuzin residues in corn fields by use of a
molecularly imprinted polymer. Chromatographia 69:615–619.
https ://doi.org/10.1365/s1033 7-008-0862-5
Zuskin E, Mustajbegovic J, Schachter EN, Kern J, Deckovic-Vukres
V, Trosic I, Chiarelli A (2008) Respiratory function in pesticide
workers. Int J Occup Environ Med 50:1299–1305. https ://doi.
org/10.1097/JOM.0b013 e3181 845f6 c
Author's personal copy