Molecularly imprinted polymer based electrochemical sensor for the determination of the anthelmintic drug oxfendazole

Journal of Electroanalytical Chemistry 729 (2014) 135–141

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry

journal homepage: www.elsevier .com/locate / je lechem

Molecularly imprinted polymer based electrochemical sensorfor the determination of the anthelmintic drug oxfendazole

http://dx.doi.org/10.1016/j.jelechem.2014.07.0211572-6657/� 2014 Published by Elsevier B.V.

⇑ Corresponding author. Tel./fax: +20 002 057 2403868.E-mail address: [email protected] (A.-E. Radi).

Abd-Elgawad Radi ⇑, Abd-Elrahman El-Naggar, Hossam M. NassefDepartment of Chemistry, Faculty of Science, Dumyat University, 34517 Dumyat, Egypt

a r t i c l e i n f o a b s t r a c t

Article history:Received 3 April 2014Received in revised form 13 June 2014Accepted 21 July 2014Available online 30 July 2014

Keywords:Molecularly imprinted polymersPolypyrroleOxfendazoleSensorsScreen printed carbon electrodeVoltammetry

A molecularly imprinted polymer (MIP) based electrochemical sensor for the determination of the anthel-mintic drug oxfendazole (OFZ) was developed. The polypyrrole (PPy) was electrochemically synthesizedonto a screen printed carbon electrode (SPCE) surface. The determination of OFZ was accomplished onMIP-SPCE using differential pulse (DP) and square wave (SW) voltammetry in phosphate buffer solutionpH 3.8. The MIP exhibited a good selectivity for OFZ with respect to the other benzimidazoles veterinarydrugs. The applicability of the method was tested with milk samples, spiked with OFZ at 30.0–250.0 lg kg�1 with detection limits of 10.0 and 8.0 lg/kg and the mean recoveries obtained were101.8 ± 3.8% and 100.8 ± 4.2%, at spiking concentration of 50.0 lg/kg for DPV and SWV, respectively.

� 2014 Published by Elsevier B.V.

1. Introduction

Oxfendazole OFZ (methyl N-[6-(benzenesulfinyl)-1H-1,3-ben-zodiazol-2-yl] carbamate; Scheme 1) is widely used as anthelmin-tic for the control of gastrointestinal and lung nematodes inlivestock. Moreover, OFZ is the active metabolite of fenbendazoleand febantel [1] and the efficacy of these two latter anthelminticsis due partly to the formation of OFZ in the animal’s body. The useof OFZ as an orally effective anthelmintic in dairy cattle may resultin its presence in milk at low levels. To ascertain the safe use of OFZin food-producing animals, a sensitive analytical method capableof measuring the drug at low level in milk is required. The analysisof benzimidazoles BZDs in animal tissues and biological fluids bycolorimetric [2] fluorometric [3], high-performance liquid chro-matographic (HPLC) [4,5], and radioimmunoassay methods [6]has been reported. The colorimetric and fluorometric methods lackthe specificity and sensitivity needed for assay in milk. HPLC and radioimmunoassay are intended for use in plasma, utilizing mini-mum sample cleanup. These methods are not applicable to theanalysis of trace drug levels in milk, because of the high fat contentof raw milk. An HPLC method for the analysis of benzimidazolecompounds in animal tissues and milk based on ion-exchangechromatography was also reported [7]. The sensitivity achieved

was 50.0 lg/kg in cow milk and 100.0 lg/kg in cattle tissues. Themethodologies for the determination of BZDs in biological matriceshave been discussed in terms of sample handling, analysis, resi-dues included and sensitivity [8]. It was concluded that the meth-odology for determination of benzimidazole residues in foods ofanimal origin needs improvement due the difficult challenge faced,owing to the extensive metabolism of these molecules.

The introduction of MIP materials in the area of electrochemicalsensors is emerging fast as a popular tool owing to the growinginterest in achieving selective analysis of the target molecules indifferent fields such as clinical diagnostics, environmental control,food analysis and drug screening. MIPs are synthetic polymers ableto selectively recognize a template molecule in an easy and a rapidway. The synthetic procedure is cheap and MIPs are stable underharsh conditions of pH and temperature. Basically, MIPs areprepared by the polymerisation of a suitable monomer and across-linker agent in the presence of a template molecule. Afterpolymerisation, the template is removed from the polymericmatrix leaving cavities complementary in size and shape to thetemplate, which should be capable of specific rebinding of the ana-lyte. The electrosynthesis of MIPs has been shown to be a versatileapproach in the choice of functional monomers, as shown by thewide range of molecules successfully used to this end. One of themost widely used polymers in electrochemical imprinting is poly-pyrrole (PPy). PPy is often used in the design of electrochemicalbiosensors. The Ppy layers could be formed by several differentmethods: chemical polymerization initiated by oxidators such as

NH

OO

HN

N

SO

CH3

Scheme 1.

136 A.-E. Radi et al. / Journal of Electroanalytical Chemistry 729 (2014) 135–141

FeCl3 or H2O2 [9], enzymatic polymerization [10] and electrochem-ical polymerization [11,12]. Different electrochemical methodswere developed in order to form a wide variety of PPy layers; aremainly based on potentiostatic [13], galvanostatic [13,14], orpotentiodynamic [11,12] techniques. The potentiodynamic tech-nique seems the most suitable for the formation of stabile conduct-ing polymer layer [12,15] since they allow to increase theconcentration of Py monomer at pre-electrode environment duringthe period when the electrode potential decreases below thepotential, which is required for the initiation of the polymerization.Moreover, the potentiodynamic methods allow significant enrich-ment of PPy layer by entrapped biological compound (e.g. enzyme,antibody, single stranded DNA, etc.) [16–20]. In such way the con-sumption of expensive biomaterials could be reduced. The poten-tiodynamic methods based on both potential cycling andpotential pulses are mostly applied for the formation of PPy layers[21]. The efficiency of electrochemical polymerization reaction andcharacteristics of formed PPy layers depend on setting of parame-ters such as potential sweep rate and vertex potentials. Anothercontrollable and more reliable is another potentiodynamicmethod, which is based on rectangular potential pulses with fixedpotential values [11]. In this method several discreet potentiallevels, which are the most suitable for polymerization and forrelaxation of electrochemical system between polymerizationsteps, are applied. During the relaxation-step concentrations ofpyrrole monomer and materials, which become entrapped withinPpy layer, are restored at pre-electrode environment. A range ofelectrochemical sensors based on molecularly imprinted PPy havealready been used for the determination of bovine leukemia virusglycoproteins [22], benzimidazole [16], bovine hemoglobin [23],paracetamol [19], caffeine [24], epinephrine [25], ascorbic acid[26], and 2,4-dichlorophenoxy acetic acid [27]. Screen-printingpermits miniaturization of electrochemical sensors by integratingthe reference, auxiliary, and working electrodes on the same chip.SPEs are readily combined with simple, inexpensive, and portableelectrochemical instrumentation to make them suitable for on-sitedetermination in real time.

So far no electrochemical methodology has been considered forthe determination of OFZ. Considering the advantages of the elec-trochemical method and the molecular imprinting technique, thekey idea of the present work is to construct an electrochemical sen-sor with high sensitivity and excellent selectivity for OFZ detection.The electrochemical behavior of the MIP for OFZ recognition hasbeen investigated in details. The detection of OFZ was accom-plished at MIP-SPCE using DP and SW voltammetry. Furthermore,the MIP-electrode has been successfully used to detect OFZ in milk.

2. Experimental

2.1. Apparatus

Electrochemical measurements were performed with Bio-logicSAS Electrochemical Analyzer, Model SP50, controlled by EC-Labexpress Version 5.52 software (Bio-logic SAS, France). Home-madethree printed electrodes, a working screen-printed carbon

electrode (3.1 mm diameter) printed from a carbon-based ink(Electrodag 421, Acheson); a silver pseudo-reference electrodemade from a silver-based ink (Electrodag 477 SS, Acheson) andthe auxiliary electrode from a carbon ink, were used. All pH-metricmeasurements were made on a CG 808 digital pH-meter with glasscombination electrode (Schott Gerate, Germany). The electrodewas calibrated with commercially available buffer referencesolutions.

2.2. Reagents and chemicals

The benzimidazole BZDs anthelmintics (oxfendazole albenda-zole, mebendazole, and thiabendazole) were from Sigma (St. Louis,USA). The structures of the BZDs studied are shown in Scheme 2.The standard stock solutions were prepared prepared by dissolvingin methanol and stored at 4 �C. The working solutions were pre-pared by diluting the stock solution with the appropriate 0.20 Mphosphate buffer solution. Pyrrole (Py) (Sigma–Aldrich) wasreagent grade quality and was used as received. The preparationof the aqueous solutions was carried out using ultra pure qualityof water. The phosphate buffer solutions were prepared fromNaH2PO4, Na2HPO4 and H3PO4, with distilled water. Other chemi-cals used were of analytical grade, used without furtherpurification.

2.3. Procedure

An aliquot of 200 lL of polymerization solution of NaClO4, Mpyrrole and OFZ was dropped onto the exposed area of the strip.The MIP was obtained by the electrodeposition on the surface ofthe clean SPCE using CV in the potential range between �0.6 and1.2 V during two cycle (m = 100 mV s�1). The extraction of OFZ tem-plate molecule was performed electrochemically by cyclingbetween �0.6 and 1.0 V in 0.2 M phosphate buffer pH 3.8, for tencycles until all OFZ molecules were stripped from the imprintedPPy film. For comparison, a control NIP-electrode was preparedby the same procedure, only without the addition of template mol-ecule in the polymerization process. An aliquot of 200 lL of thesupporting electrolyte solution and sample containing OFZ wasadded to cover the electrodes, and the voltammograms initiatedin the positive direction were recorded. The anodic potential sweepwas achieved under different operational parameters. All measure-ments were carried out at room temperature.

Milk samples were spiked with OFZ at range of concentrationsfrom 30.0 to 250.0 lg kg�1. After vortexing for 45 s, the mixturewas centrifuged for 10 min at 5000 rpm to remove milk proteinresidues and the supernatant was taken carefully. Appropriate vol-umes of this supernatant were transferred into a volumetric flaskand diluted up to the volume with 0.2 M phosphate buffer solutionpH 3.8. The voltammetric measurements for these solutions werecarried out as described above. Quantifications were performedusing the related calibration equations.

3. Results and discussions

3.1. Electrochemical behavior of OFZ at SPCE

No electrochemical data was found in the literature concerningthe redox behavior of OFZ neither at the solid nor the mercury elec-trode. In the first step, OFZ was subjected to CV, DPV, and SWVstudies with the aim of a detailed characterization of its electro-chemical oxidation behavior on the SPCE. CV for 5.0 lM OFZ inphosphate buffer solution pH 2.0 is shown in Fig. 1, the scanningwas started at 0.2 V in the anodic direction. By reversing at+1.20 V, no reduction signal corresponding to the anodic response

NH

OON

HN

SH3C

CH3Albendazole

NH

OON

HN

O CH3

MebendazoleS

N

HN

Nthiabendazole

Scheme 2.

0.2 0.4 0.6 0.8 1.0 1.2

0

3

6

9

12

E/ V vs. Ag ref. electrode

i/µA

0 2 4 6 8 100

1

2

3

4

5

v 1/2

/(mVs-1)1/2

i/µA

Fig. 1. CV for 5.0 lM OFZ solution (solid line) in 0.2 M phosphate buffer solution pH2.0 at a SPCE, dotted line blank buffer solution at m100 mV s�1. Inset: ip–m1/2 plot.

A.-E. Radi et al. / Journal of Electroanalytical Chemistry 729 (2014) 135–141 137

was observed in the cathodic branch. Scan rate studies over a rangeof 10–100 mV s�1 were carried out to assess whether the processeswere under diffusion or adsorption control. A linear response wasobserved with the square root of the scan rate (m1/2) with a slopeof 0.51, very close to the theoretical value of 0.50, which isexpected for an ideal reaction of solution species [28]. The peakpotentials shifted to the anodic direction when the scan rate wasincreased. The plot of Ep vs. log m was linear with a correlationcoefficient of r = 0.992. This behavior is consistent with the EC nat-ure of the reaction in which the electrode reaction is coupled withan irreversible follow-up chemical step [29,30]. Tafel analysis of

0.0

0.5

1.0

1.5

2.0

2.5

i/µA

A

0.2 0.4 0.6 0.8 1.0

E/ V vs. Ag ref. electrode

Fig. 2. (A) DPV and (B) SWV of 5.0 lM OFZ in 0.2 M p

voltammograms from the oxidation of 5.0 lM OFZ in PBS pH 2.0(m = 5 mV s�1) was conducted. The an value of the anodic reactioncorresponding to the voltammetric oxidation peak was obtainedusing Tafel plot (logi vs. Ep). A value of 0.27 was obtained. Theexchange current density was io = 1.298 � 10�14 lA/cm2. Thesevalues, together with the absence of cathodic response withinthe potential window in cyclic voltammetry (Fig. 1) confirmedthe irreversibility of the oxidation process.

The effect of pH on peak intensity and peak potential were stud-ied for 5.0 lM OFZ in 0.2 M phosphate buffer solutions using DPVand SWV techniques (Fig. 2). Both graphs from DPV and SWV werefound to be similar at SPCE. In all instances, OFZ undergoes onemain irreversible oxidation process, which shifted towards less-positive potentials as the pH was increased. A slope (Ep/pH)of � 51.4 mV/pH unit over a pH range from 2.0 to 10.6 wasobtained. This slope is close to the Nernstian value of 59.2 mVfor a 2-electron, 2-proton process in the rate-determining step[31–33]. The study of the influence of pH on peak currents was alsocarried out to determine the pH value for the maximum signal withthe best peak shape. The optimal was obtained in 0.2 M phosphatebuffer solution pH 2.0 supporting electrolyte. This supporting elec-trolyte was chosen for further work. Taking into account all thestudies performed so far, we may assume that the oxidation pro-cess is located on the sulfinyl group [34]. The anodic oxidation ofthe sulfinyl group to sulfone seems to be irreversible and involvingloss of 2 e� and 2 H+. A tentative reaction mechanism proposed forthe electrooxidation of OFZ is presented in Scheme 3.

3.2. Determination of OFZ at SPCE using DPV and SWV

Fig. 3 shows DP and SW voltammograms at SPCE for differentconcentrations of OFZ in 0.2 M phosphate buffer solution pH 2.0.Using DPV and SWV, linear calibration plots (r = 0.996, and 0.992,

0.2 0.4 0.6 0.8 1.0

0

1

2

3

4

E/ V vs. Ag ref. electrode

B pH 2.00pH 3.77pH 4.70pH 6.00pH 7.00pH 8.00pH 9.10pH 10.60i/ µ

A

hosphate buffer solution at different pHs at SPCE.

-2e,-2H+H2ONH

OO

HN

N

SO

CH3

NH

OO

HN

N

SO

CH3

O

Scheme 3.

0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.0

0.2

0.4

0.6

E/ V vs. Ag ref. electrode

A

i/µA

B

i/µA

0.0 0.3 0.6 0.90.0

0.1

0.2

0.3

[OFZ]/µM

i/µA

0.0 0.3 0.6 0.90.0

0.1

0.2

[OFZ]/µM

i/µA

0.4 0.6 0.8 1.0

E/ V vs. Ag ref. electrode

Fig. 3. (A) DP and (B) SWV at SPCE for different concentrations of OFZ in 0.2 M phosphate buffer solution pH 2.0. Insets: calibration curves.

138 A.-E. Radi et al. / Journal of Electroanalytical Chemistry 729 (2014) 135–141

respectively) for OFZ were obtained (Fig. 3) in the 0.08–0.80 lMrange, with a slope value of 0.30 ± 0.01lA/lM and 0.28 ±0.01 lA/lM, and an intercept of 0.0028 ± 0.0004 lA and 0.0017 ±0.0006 lA, respectively. The limit of detection was calculatedaccording to the 3r/s criterion [35], where s is the slope of the cal-ibration plot and r is estimated as the standard deviation (n = 5) ofthe ip measurements. The obtained values were 0.03 and 0.02 lM.The generally poor selectivity of voltammetric techniques can poseproblems in the analysis of biological samples, which contain oxi-dizable substances; which can be oxidized at a potential near tothat of OFZ at most solid electrodes, resulting in an overlapping

0.4 0.6 0.8 1.0 1.2

0

2

4

6

8

E/ V vs. Ag ref. electrode

i/µA

b

a

Fig. 4. CVs of MIP-SPCE electrode (a), and NIP-SPCE electrode (b) in phosphatebuffer solution (0.2 M, pH 3.8) as supporting electrolyte. m = 100 mV s�1.

voltammetric response. Hence in this study the imprinting step iscrucial in the development of a sensor with high selectivity.

3.3. Electrochemical behavior of OFZ at MIP and NIP electrodes

In order to confirm whether the OFZ molecules had beenembedded in the MIP film, CVs of MIP and NIP electrodes wererecorded in 0.2 M phosphate buffer solution pH 3.8 at a potentialrange from 0.2 to 1.20 V (Fig. 4). Clear oxidation peak appears atthe MIP electrode and no peak for the NIP electrode. Since the elec-trochemical measurement was carried out in OFZ free solution, theoxidation peak was entirely due to the OFZ embedded in the MIPfilm. The OFZ molecules diffuse towards the surface of the SPCEduring the electropolymerization process and trapped into thepolymer matrix. The creation of the molecular imprints is favoredby the diffusion of the electroactive template, generating a farhigher number of recognition sites during the electrodepositionof the polymer.

3.4. Association between the functional monomer and the templatemolecule

The functional monomer is selected for its ability to providespecific points of recognition for the template within the polymermatrix, usually by hydrogen bonding [36]. This is a prerequisite forthe formation of a strong complex between the template and func-tional monomer prior to polymerization. The Py was chosen forthis end. The OFZ molecules are trapped in the polymer matrixduring the electrodeposition of the polymer, as a result of the abil-ity of these molecules to interact with the Py units. The schematicrepresentation of imprinting and removal of OFZ from imprintedPPy film was shown in Scheme 4. OFZ has different functional

extraction of OFZ

rebinding

NH

OO

HN

N

SO

HN

n

NH

NH HN

HN

HN

HNn

NH

n

HN

n

NH

NH HN

HN

HN

HNn

NH

n

n

n

n

n

n

nn

n

n

Scheme 4.

A.-E. Radi et al. / Journal of Electroanalytical Chemistry 729 (2014) 135–141 139

groups which could be involved in hydrogen-bonding formationwith the NAH group of Py units. Chain branching and cross linkingin PPy generate a 3D matrix with niches containing the templateOFZ. The recognition of OFZ molecule is based on shape selectionand positioning of the functional groups [37].

The successive CVs of the MIP electrode recorded in 0.2 M phos-phate buffer solution pH 3.8 showed a significant decrease of thepeak current following an increase in number of cycles; no oxida-tion peak for OFZ can be detected after 10 cycles, which indicatedthat the template in the MIP film was efficiently removed, resultingin the formation of sites in the polymer film that can rebind to thetemplate molecules. The affinity of the MIP electrode was alsocharacterized by DPV and SWV. Fig. 5 shows typical DPV andSWV for 5.0 lM OFZ in PBS at MIP-SPCE (a), SPCE (b), andNIP-SPCE. Comparing the signals, the MIP-SPCE gives the highestcurrent value. This may be explained by the fact that the NIP hasno cavities to bind OFZ, but the MIP film has a stronger affinityto OFZ molecules.

3.5. Optimization of MIP fabrication

The thickness of the film, made with molecularly imprinted PPy,is an important factor affecting the film’s recognition ability and itcould easily be adjusted by controlling the concentration of Py dur-ing the electropolymerization process. If the imprinted polymermembranes are too thick, template molecules situated at the

0.4 0.6 0.8 1.0

0

1

2

3

E/ V vs. Ag ref. electrode

c

a

b

A

i/µA

Fig. 5. (A) DPV and (B) SWV for 5.0 lM OFZ in PBS 0.2 M

central area of the polymer membranes cannot completely beremoved from polymer matrix. The monomer concentrationshould be proportional to the thickness of the deposit and amountof imprinted molecule in the polymeric matrix. To investigate theeffect of monomer concentration to the electrochemical responseof OFZ, the MIP-SPCEs were prepared in solutions of constant con-centration of OFZ (250 lM) and varying pyrrole concentrations inthe range of 0.005–0.050 M. The current response reached maxi-mum with the concentration of 0.014 M, and then decreased withfurther increasing the concentration of Py monomer. Thus, 0.014 MPy monomer was chosen for the electrochemical polymerization toobtain the highest sensitivity for the determination of OFZ.

The number of cycles applied to the preparation of MIP-SPCEduring the electropolymerization was found to affect the sensitiv-ity and linearity of the sensor. A series of experiments in whichelectrodes were fabricated with different numbers of CV cycleswere carried out to get the optimal CV cycles of electropolymeriza-tion. The current response reached maximum with 2 cycles, andthen decreased with further increasing cycle number. The exces-sive number of cycle can lead to more extensive electropolymer-ization, and therefore to the formation of thicker sensing filmwith less accessible imprinted sites. The highest current differencefor OFZ at the MIP-SPCE was obtained by applying 2 cycles in theelectropolymerization.

Moreover, the concentration of the template molecule appliedin the electrosynthesis could also be an important polymerization

0.4 0.6 0.8 1.0

0

2

4

E/ V vs. Ag ref. electrode

c

a

b

B

i/Aµ

, pH 3.8 at MIP-SPCE (a), SPCE (b) and NIP-SPCE (c).

0.4 0.6 0.8 1.0

0.0

0.3

0.6

0.9

0.4 0.6 0.8 1.0

0.0

0.5

1.0

1.5

2.0

E/ V vs. Ag ref. electrode

i/µA

A

E/ V vs. Ag ref. electrode

i/µA

B

0.0 0.2 0.40.0

0.2

0.4

[OFZ] (µM)

i/µA

0.0 0.2 0.40.0

0.2

0.4

0.6

[OFZ] (µM)

i/µA

Fig. 6. (A) DPV and (B) SWV at MIP-SPCE for increasing OFZ concentrations, in 0.2 M phosphate buffer solution pH 3.8. Inset: calibration curves.

0.4 0.6 0.8 1.0

0.0

0.3

0.6

0.9

1.2

1.5

0.0

0.5

1.0

1.5

2.0

2.5

E/ V vs. Ag ref. electrode

A

i/µA

B

i/µA

0 100 2000.0

0.2

0.4

0.6

0.8

[OFZ]/(µg/kg)

i/ µA

0 100 2000.0

0.3

0.6

0.9

1.2

[OFZ]/(µg/kg)

i/ µA

0.4 0.6 0.8 1.0

E/ V vs. Ag ref. electrode

Fig. 7. (A) DPV and (B) SWV at MIP-SPCE for milk spiked with (30–550 lg/kg) OFZ in 0.2 M phosphate buffer solution pH 3.8. Insets: calibration curves.

140 A.-E. Radi et al. / Journal of Electroanalytical Chemistry 729 (2014) 135–141

condition. As the concentrations of template molecule applied tothe polymerization process increases in a range from 50 M to500 lM, the amount of template entrapped in the matrix alsoincreases until it reaches equilibrium at about 250 lM and the cur-rent intensity that corresponds to the oxidation of the entrappedtemplate tends to become stable as no further quantity of templatecan be incorporated into the polymeric chain.

The concentration of NaClO4 supporting electrolyte was animportant parameter for MIP fabrication. Its influence was investi-gated over the range from 0.01 M to 0.50 M. The best results wereobtained employing 0.10 M M NaClO4, further increase in the con-centration did not improve the analytical signal. All the optimizedparameters were then used for further work.

3.6. Selectivity of the MIP electrode

After removing the template molecules, the molecularlyimprinted polymer is expected to present a high selectivity for

the imprinted molecules. In order to confirm the selectivity ofthe MIP, the DP and SW voltammograms were obtained in solu-tions containing 0.25 lM OFZ and other BZDs with analogousstructure to OFZ: ABZ, MBZ and TBZ at concentration levelsbetween 0.25 and 25.00 lM. The MIP exhibited significant selectiv-ity for OFZ and the other BZDs did not affect the anodic current ofOFZ, at the concentrations tested. The high selectivity can beexplained by the size and arrangement of functional group of spe-cific cavities matched with OFZ in the imprinted film. BZDs includ-ing OFZ: ABZ, MBZ and TBZ have a bicyclic ring system in theirstructures in which benzene is fused to the 4 and 5 positions of theheterocycle (imidazole). They are benzimidazole carbamates, ana-logues with benzenesulfinyl; 5-(propylthio; 5-benzoyl-at position5(6)- of the BZD nucleus. Such substituents could prevent the suit-able settlement of interferents molecules in the cavities of MIP. It isnoteworthy that, MBZ would fit into the polymer in exactly the sameway. It might still not interfere, because it would not show a volam-metric signal because it contains a C@O instead of a S@O group.

Table 1Determination of OFZ in milk by HPLC and the voltammetric methods (n = 5).

HPLC [7] DPV SWV

Linearity rang (lg/kg) 5.0–50.0 30.0–250.0 30.0–250.0Slope 51.160 0.003 (lA/(lg/kg)) 0.004 (lA/(lg/kg))Correlation coefficient 0.999 0.998 0.998LOQ (lg/kg) 1.5 10.0 8.0Amount added (lg/kg) 50.0 50.0 50.0Mean (lg/g) 50.3 50.9 50.4Recovery (%) 100.6 101.8 100.8RSD (%) 4.8 3.8 4.2t-Test of significance ttab. = 2.447 0.5602 1.1421F-test of significance Ftab. = 4.107 2.067 1.8452

A.-E. Radi et al. / Journal of Electroanalytical Chemistry 729 (2014) 135–141 141

3.7. Linearity, detection limit, reusability and reproducibility of MIP-electrode

SWV allows the use of faster scan rates than DPV. This is one ofthe main advantages of SWV over DPV. The DPV and SWVresponses to different concentrations of OFZ at MIP-electrode areshown in Fig. 6. Good linear relations are obtained within a con-centration range from 0.05 to 0.50 lM. The linear regression equa-tions are: ip (lA) = 0.8414c (lM) + 0.0003 and ip (lA) = 1.1857c(lM) + 0.0003, with a correlation coefficient of 0.994 and 0.997,and a lower detection limit of 0.025 and 0.015 lM for DPV andSWV respectively. In addition, the reusability of MIP-electrodewas investigated in the presence of 0.25 lM OFZ eight times usingthe same electrode following retention/removal processes. A rela-tive standard deviation (RSD) of 2.3% was obtained. Similarly, thefabrication reproducibility was also estimated by eight differentMIP-modified electrodes. The RSD obtained was 3.7%. It is foundthat when the MIP surface is regenerated after electrode modifica-tion or measurements by a thorough template extraction; could bereused so many times. The storage stability of the MIP electrodewas quite good. Practically no changes in OXF binding wereobserved after storage of the dry MIP electrode at room tempera-ture for 1 month.

3.8. Voltammetric determination of OFZ in milk samples

The MIP-electrode was utilized for the determination of traceOFZ in milk samples according to the experimental proceduredescribed in determination of OFZ in milk samples. None of thesereal samples from the supermarket had neither DPV nor SWVresponses when analyzed using the MIP-electrode. The DPV andSWV obtained from the samples spiked with OFZ from 30.0 to250.0 lg kg�1 showed appropriate shapes to be used for quantifi-cation of the analyte (Fig. 7A and B). The results obtained for milksamples analyzed are summarized in Table 1. The confidence inter-vals were calculated for a significance level of 0.05. As can beobserved, mean recoveries of 101.11 (RSD = 3.8%) and 101.08(RSD = 4.2%) were obtained for spiking concentration of 50.0 lg/kg. These results demonstrate that the methodology developed issuitable for the analysis of OFZ in complex samples such as milk.

4. Conclusion

A reliable method for the analysis of the anthelmintic drug OFZin a complex foodstuff sample such as milk was achieved by cou-

pling MIP-SPCE with DPV or SWV. The developed method exhib-ited good analytical performance in terms of sensitivity,reproducibility and recovery. The data obtained supports the beliefthat the method, combining molecular imprinting and electro-chemistry, is a potential route to the creation of cost-effective,miniaturized sensors for the detection of species in foodanalysis.

Conflict of interest

The authors report that there are no conflicts of interest.

References

[1] P. Delatour, M. Tiberghien, F. Garnier, E. Benoit, Am. J. Vet. Res. 46 (1985)1399–1402.

[2] N.C. Sangster, R.K. Prichard, E. Lacey, J. Parasit. (1985) 645–651.[3] W.R. Baeyens, F.A. Fattah, P.D. Moerloose, Anal. Lett. 18 (1985) 2143–2154.[4] W.J. Blanchflower, A. Cannavan, D.G. Kennedy, Analyst 119 (1994) 1325–1328.[5] J. Bogan, S. Marriner, J. Pharm. Sci. 69 (1980) 422–423.[6] C. Nerenberg, I. Tsina, S. Matin, J. Assoc. Off. Anal. Chem. 65 (1982) 635–639.[7] I. Tsina, S. Matin, J. Pharm. Sci. 70 (1981) 858–860.[8] M. Danaher, H. De Ruyck, S.R. Crooks, G. Dowling, M. O’Keeffe, J. Chromatogr. B

845 (2007) 1–37.[9] K. Leonavicius, A. Ramanaviciene, A. Ramanavicius, Langmuir 27 (2011)

10970–10976.[10] A. Ramanaviciene, A. Kausaite-Minkstimiene, Y. Oztekin, G. Carac, J. Voronovic,

N. German, A. Ramanavicius, Microchim. Acta 175 (2011) 79–86.[11] W. Schuhmann, C. Kranz, H. Wohlschläger, J. Strohmeier, Biosens. Bioelectron.

12 (1997) 1157–1167.[12] A. Ramanaviciene, A. Ramanavicius, A. Finkelsteinas, J. Chem. Educ. 83 (2006)

1212.[13] S.K. Mondal, K.R. Prasad, N. Munichandraiah, Synth. Met. 148 (2005) 275–286.[14] Y.-M. Uang, T.-C. Chou, Biosens. Bioelectron. 19 (2003) 141–147.[15] J.C. Soares, A. Brisolari, V.d.C. Rodrigues, E.A. Sanches, D. Goncalves, React.

Funct. Polym. 72 (2012) 148–152.[16] A. Nezhadali, L. Mehri, R. Shadmehri, Sens. Actuators B: Chem. 171 (2012)

1125–1131.[17] G. Díaz-Díaz, D. Antuña-Jiménez, M. Carmen Blanco-López, M. Jesús Lobo-

Castañón, A.J. Miranda-Ordieres, P. Tuñón-Blanco, Trends Anal. Chem. 33(2012) 68–80.

[18] B.S. Ebarvia, S. Cabanilla, F. Sevilla III, Talanta 66 (2005) 145–152.[19] L. Özcan, Y. S�ahin, Sens. Actuators B: Chem. 127 (2007) 362–369.[20] C.-C. Lin, Y.-M. Chu, H.-C. Chang, Sens. Actuators B: Chem. 187 (2013) 533–

539.[21] H. Karami, M.G. Asadi, M. Mansoori, Electrochim. Acta 61 (2012) 154–164.[22] A. Ramanaviciene, A. Ramanavicius, Biosens. Bioelectron. 20 (2004) 1076–

1082.[23] X. Kan, Z. Xing, A. Zhu, Z. Zhao, G. Xu, C. Li, H. Zhou, Sens. Actuators B: Chem.

168 (2012) 395–401.[24] A. Vinjamuri, S.C. Burris, D. Dahl, ECS Trans. 13 (2008) 9–20.[25] X. Kan, H. Zhou, C. Li, A. Zhu, Z. Xing, Z. Zhao, Electrochim. Acta 63 (2012) 69–

75.[26] L. Özcan, M. Sahin, Y. Sahin, Sensors 8 (2008) 5792–5805.[27] C. Xie, S. Gao, Q. Guo, K. Xu, Microchim. Acta 169 (2010) 145–152.[28] E. Laviron, L. Roullier, C. Degrand, J. Electroanal. Chem. Interfacial Electrochem.

112 (1980) 11–23.[29] R.N. Goyal, V.K. Gupta, M. Oyama, N. Bachheti, Electrochem. Commun. 8

(2006) 65–70.[30] R.N. Goyal, N. Jain, V. Gurnani, Monatsh. Chem. 132 (2001) 575–585.[31] J. Wang, Electroanalytical Techniques in Clinical Chemistry and Laboratory

Medicine, VCH, New York, 1988.[32] P.T. Kissinger, W.R. Heineman, Laboratory Techniques in Electroanalytical

Chemistry, CRC Press, 1996.[33] T.J. O’Shea, D. Leech, M.R. Smyth, J.G. Vos, Talanta 39 (1992) 443–447.[34] W. Krause, J. Chromatogr. B Biomed. Appl. 222 (1981) 71–79.[35] J.C.M.J.N. Miller, Statistics and chemometrics for analytical chemistry, Prentice

Hall, 2005.[36] C. Malitesta, E. Mazzotta, R.A. Picca, A. Poma, I. Chianella, S.A. Piletsky, Anal.

Bioanal. Chem. 402 (2012) 1827–1846.[37] G.G. Wallace, P.R. Teasdale, G.M. Spinks, L.A. Kane-Maguire, Conductive

Electroactive Polymers: Intelligent Polymer Systems, CRC Press, 2008.