Materials Science and Engineering C 37 (2014) 264–270

Contents lists available at ScienceDirect

Materials Science and Engineering C

j ourna l homepage: www.e lsev ie r .com/ locate /msec

Facile stripping voltammetric determination of haloperidol using a highperformance magnetite/carbon nanotube paste electrode inpharmaceutical and biological samples

Hasan Bagheri a,⁎, Abbas Afkhami b, Yunes Panahi a, Hosein Khoshsafar b, Ali Shirzadmehr b

a Chemical Injuries Research Center, Baqiyatallah University of Medical Sciences, Tehran, Iranb Faculty of Chemistry, Bu-Ali Sina University, Hamedan, Iran

⁎ Corresponding author. Tel./fax: +98 218 284 251.E-mail address: h.bagheri@srbiau.ac.ir (H. Bagheri).

0928-4931/$ – see front matter © 2014 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.msec.2014.01.023

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 November 2013Received in revised form 29 December 2013Accepted 8 January 2014Available online 18 January 2014

Keywords:Electrochemical sensorCarbon paste electrodesVoltammetryHaloperidolMulti-walled carbon nanotubesMagnetic multi-walled carbon nanotubes

Multi-walled carbon nanotubes decorated with Fe3O4 nanoparticles were prepared to construct a novel sensorfor the determination of haloperidol (Hp) by voltammetricmethods. Themorphology and properties of electrodesurface were characterized by scanning electron microscopy (SEM) and electrochemical impedance spectrosco-py. Thismodified sensorwas used as a selective electrochemical sensor for the determination of trace amounts ofHp. The peak currents of differential pulse and square wave voltammograms of Hp increased linearly with itsconcentration in the ranges of 1.2 × 10−3–0.52 and 6.5 × 10−4–0.52 μmol L−1, respectively. The detection limitsfor Hpwere 7.02× 10−4 and 1.33 × 10−4 μmol L−1 for differential pulse and squarewave voltammetricmethods,respectively. The results show that the combination of multi-walled carbon nanotubes and Fe3O4 nanoparticlescauses a dramatic enhancement in the sensitivity of Hp quantification. This sensorwas successfully applied to de-termineHp in pharmaceutical samples and biologicalfluids. The fabricated electrode showedexcellent reproduc-ibility, repeatability and stability.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Hp is an antipsychotic drug most widely used in the treatment ofschizophrenia and acute psychotic states and delirium. It is also usedin the control of the symptoms of acute psychosis, such as drug-induced psychosis, psilocybin, amphetamines, ketamine, phencyclidineand psychosis associatedwith high fever ormetabolic disease [1–3]. Ac-curate measurements of antipsychotic drug levels will also facilitatemore precise estimates of dopamine D2 receptor occupancy, a factorthat has been suggested to serve as a cross-species guideline for deriv-ing clinically relevant antipsychotic doses for animal models [4]. Sincethe antipsychotic drugs are very active, they are usually administeredat low daily dosages [5]. At steady state these doses result in plasmalevels in the low ng mL−1 range. In order to quantify the levels of Hp,an analytical method with high sensitivity is required [5]. The analyticalmethods for the determination of Hp and its metabolites in biologicalsamples so far reported include gas chromatography (GC), GC–massspectrometry (MS), high-performance liquid chromatography (HPLC),HPLC–MS, capillary electrophoresis. etc. [2,5–8]. These methods usuallyinvolve time-consuming multi-step pretreatment including liquid–liquid extraction and/or solid-phase extraction to remove impuritiescontained in plasma or serum for the detection of these species. Also,

ghts reserved.

these methods are expensive and not suitable for in-situ analysis dueto the ponderous and complicated instruments [5–8]. That is why, thedevelopment of simple, accurate, selective, fast and cheap method forthe determination of Hp in biologicalfluids and pharmaceutical samplesappeared at present to be an attracted case.

On the contrary, the electrochemical methods as alternative of thesetechniques have been accepted as the efficient methods to determinedrugs due to their excellent sensitivity, short analysis time, portability,low cost and low maintenance [9,10]. Voltammetric methods usingmercury electrodes have suggested for the determination of Hp[11–13]. However, in order to avoid the environmental pollution andhealth hazard caused by traditional mercury electrodes, and some in-conveniences complicating the use of voltammetry by them, variousmercury-free electrodes (carbonaceous electrodes) have been devel-oped over the past years [14,15].

Adams first introduced carbon paste as the electrode material [16].Carbon paste, which is a mixture of carbon powder and a suitable liquidbinder, represents the most flexible substrate for a wide variety ofchemical and biological modifications. Application of chemically modi-fied electrodes (CMEs) in the electrochemical determinations hasbeen widely considered as a sensitive and selective analytical methodfor the determination of trace amounts of important species [16–19].The operation mechanism of such chemically modified carbon pasteelectrodes (CMCPEs) depends on the properties of the modifier usedto improve selectivity and sensitivity towards the target species.When the experimental condition is optimized, this approach

265H. Bagheri et al. / Materials Science and Engineering C 37 (2014) 264–270

eliminates the problem of interferences by other species in the sample[16–19].

The key point to obtain a good and reliable CME is the kind of mate-rials that constitute the modified electrode. The attractive properties ofnanomaterial have paved the way for the fabrication of a wide range ofmodified CPEs that exhibit improved analytical capacities. Multi-walled carbon nanotubes (MWCNTs) as a type of highly conductivenanomaterials, provide excellent performance in enhancing electro-chemical reactivity, promoting electron-transfer reactions and alleviat-ing surface fouling [20–22]. The functionalization of MWCNTs hasresulted in various nanocompositematerialswithmany newproperties.Nanocomposite materials have been increasingly important in electro-chemical researches [17–20]. Nanoparticle hybrid materials madefrom inorganic solids and MWCNTs are highly promising for applica-tions in nanoscale devices and nanoelectronics. Furthermore, uniformlydispersed nanoparticles onMWCNT surfaces yield highly effective sens-ing layer for CMEs. Fe3O4 nanoparticles (Fe3O4NPs) have attracted an in-creasing interest in construction of sensors and biosensors becauseof their good biocompatibility, strong super paramagnetic property,low toxicity, easy preparation and high adsorption ability. Moreover,Fe3O4NPs exhibit high surface area and low mass transfer resistance[23,24]. Therefore, the decoration of MWCNTs with Fe3O4 magneticnanoparticles can provide excellent electrochemical platforms for Hpanalysis due to the combination of the enlarged active surface area,strong adsorptive capability of the nanomaterial and their specific inter-action ability.

To the best of our knowledge, no study has reported the determina-tion of Hp using carbon paste electrodes. Theworks published in the lit-erature report the determination of Hpwithmercuric and glassy carbonelectrodes (Table 1). So, in this study we described the preparation andsuitability of Fe3O4NPs/MWCNTs/CPE as a new electrode for the deter-mination ofHp. The usefulness of themodified electrodewith improvedsensitivity, selectivity and accuracy has been demonstrated for the ap-plications in the pharmaceutical samples and biological fluids.

2. Experimental

2.1. Chemicals and reagents

All chemicals and reagents used in thisworkwere of analytical gradeand used as received without further purification. Haloperidol (UnitedStates Pharmacopeia (USP) Reference Standard) was purchased fromFluka. MWCNTs were purchased from Nanostructured & AmorphousMaterials (Houston, TX, USA). Graphite powder (with a particle sizeless than 1 μm) and paraffin oil were purchased from Aldrich. Allaqueous solutions were prepared with doubly distilled deionizedwater (DDW). Britton–Robinson (B–R) universal buffers (pH 4.5–8.5)were prepared in DDW and were tested as supporting electrolytes.

2.2. Apparatus

All electrochemical experiments including cyclic voltammetry (CV)and differential pulse voltammetry (DPV) and square wave voltamme-try (SWV)were performed using aMetrohm instrument (Swiss),Model797 VA Computrace. A conventional three-electrode system was used

Table 1Comparison of some figures of merit of suggested electrodes for the determination of Hp.

Electrode Method Linear ra

Hanging mercury drop electrode (HMDE) DPV 1–100HMDE SWV 1 × 10−

HMDE SWV 2.49 × 1Glassy carbon DPV 5.3 × 10MWCNT-modified glassy carbon CV 1.0 × 10Fe3O4NPs/MWCNTs/CPE SWV

DPV6.5 × 101.2 × 10

with a carbon-paste working electrode (unmodified ormodified), a sat-urated Ag/AgCl reference electrode and a Pt wire as the counter elec-trode (Azar electrode, Iran). All potentials reported are vs. Ag/AgCl. ApH-meter, Model 713 (Metrohm, Swiss), was used to determine pHvalues of the solutions. The structure of the prepared electrodes wascharacterized by a scanning electron microscope (SEM-EDX, XL30 andPhilips Netherland). The HPLC was used (Agilent 1100, USA).

2.3. Preparation of Fe3O4NPs/MWCNTs

Chemical pretreatment is a simple and effective way to introduceoxygen groups (COOH, C_O and OH) onto the MWCNT surface for im-proving their properties. They were purified and functionalized as de-scribed elsewhere [17,25]. Briefly, a 500 mg of the MWCNTs washeated in an oven under a temperature of 400 °C in nitrogen atmo-sphere for 2 h. Then MWCNTs were dispersed in 50 mL of 6.0 M HClfor 2 h under ultrasonic agitation under the nitrogen atmosphere; fil-tered on a Watman 42 filter paper and washed with DDW until thepH of the solution was neutral. These MWCNTs were dried under theIR lamp. Open-end MWCNTs with hydrophilic surfaces were thusobtained.

Iron oxide–MWCNT composite was prepared according to a previ-ously reported method with some modification [26]. For this purpose,50.0 mg of functionalized MWCNTs was added into 30 mL of distilledwater by ultrasonic irradiation for 20 min. Then 40.0 mg of FeCl3·6H2Owas added under stirring. After the mixture was stirred vigorously for30 min under N2 atmosphere, 60.0 mg of FeCl2·4H2O was added andkept stirring underN2 atmosphere for 30min. 4mL of concentrated am-monia solution diluted with 20mL of distilled water was added into themixture drop by drop. Then the solution was filtered on a 0.65 μm filtermembrane and washed with distilled water repeatedly. The resultingproduct was dried overnight at 100 °C in an oven under nitrogen at-mosphere. The sample was then annealed at 250 °C for 1 h and 650 °Cfor 2 h in a steam of nitrogen.

2.4. Electrode preparation

CPE was prepared by hand-mixing graphite powder and paraffinoil with 70/30 (w/w) in a mortar with a pestle. A portion of thecomposite mixture was packed firmly with a piston-driven carbonpaste electrode holder. For preparing a modified electrode, graphitepowder and 20.0 mg portion of the Fe3O4NPs/MWCNT were dispersedin 10 mL dimethylformamide (DMF) and homogenized ultrasonicallyfor 10 min to increase the reproducibility after each electrode surfacepolishing. The mixture was stirred by a magnetic stirrer till the solventevaporated completely. Then, paraffin oil was added to the mixture.This mixture was mixed in a mortar for at least 20 min to produce theuniform paste. The paste was carefully packed into the body tip toavoid possible air gaps, which often enhance the electrode resistance.When needed, the surface of carbon paste was mechanically renewedby extruding ca. 0.5 mm of carbon paste out of the electrode holderand smoothed with a paper. Usually, this simple operation was madebefore starting a new set of experiments.

nge (μmol L−1) Detection limit (μmol L−1) Ref.

2.66 [11]3–1.5 3.83 × 10−4 [12]0−1–3.38 4.75 × 10−3 [13]−3–2.6 × 10−2 1.3 × 10−3 [14]−1–25.0 8.0 × 10−3 [15]−4–0.52−3–0.52

1.33 × 10−4

7.02 × 10−4This work

266 H. Bagheri et al. / Materials Science and Engineering C 37 (2014) 264–270

2.5. Preparation of real samples

Human blood samples were obtained from volunteers who hadtaken Hp tablets. This study was conducted by the Ethics Committeeand written informed consents were obtained from all volunteers.Blood samplings (6mL)were performed 12–14 h after last night dosingof Hp. The serum sample was centrifuged and then after filtering, dilut-ed with B–R buffer solution (pH 7.5) without any further treatment.

Urine samples were stored in a refrigerator immediately after theircollection. 10 mL of the sample was centrifuged for 20 min at2000 rpm. The supernatant was filtered using a 0.45 μm filter andthen diluted 5-times with B–R buffer solution (pH 7.5). The solutionwas transferred into the voltammetric cell to be analyzed without anyfurther pretreatment. Standard additionmethodwas used for the deter-mination of Hp.

Toprepare tablet solution, ten tablets of Hp, labeled 2.5mgper tablet(Minoo Group, Iran) were completely grinded and homogenized. Then,10 mg of the tablet powder was accurately weighed and dissolved in100 mL water by ultrasonication. After mixing completely, the mixturewas filtered on an ordinary filter paper, and 10 mL of each sample wassubsequently transferred into a 100-mL volumetric flask and dilutedto the mark with DDW. Then, 1.0 mL of the solution was transferredinto the voltammetric cell for analysis using the standard additionmethod.

Injection ampule samples were investigated under optimum condi-tions of experiment without any pretreatment.

3. Results and discussion

3.1. Surface characterization of Fe3O4NPs/MWCNTs/CPE

The surface morphology of the prepared electrodes plays a keyrole in the selective and sensitive detection of target analyte. Fig. 1shows the SEM images of Fe3O4NPs/MWCNTs/CPE and CPE. As canbe seen at the surface of a CPE (Fig. 1a), the irregularly shaped flakesof graphite powder were present and isolated from each other. AfterFe3O4NPs/MWCNTs were added to carbon paste, it can be seen thatthey were distributed on the surface of electrode with special three-

Fig. 1. SEM images from the surface of CPE (a) and Fe3O4NPs/MWCNTs/CPE

dimensional structure (Fig. 1b and c), indicating that the CPE was suc-cessfully modified by Fe3O4NPs/MWCNTs. Also, Fig. 1d shows a typicalSEM–EDS elemental analysis of Fe3O4/MWCNTs. As expected, the sig-nals for C, O and Fe elements fromMWCNTs and Fe3O4NPs are detectedin the EDS result.

3.2. Electrochemical impedance spectroscopy (EIS) studies

EIS is also a powerful technique to study the interface properties ofmodified electrode surface. If the Nyquist plot of impedance spectra in-cludes a semicircle portion and a linear portion, the semicircle portion athigher frequencies corresponded to the electron transfer limited pro-cess and the linear portion at lower frequencies corresponded to the dif-fusion process [27].

Fig. 2 exhibits the Nyquist plots of the CPE, MWCNTs/CPE andFe3O4NPs/MWCNTs/CPE. The equivalent circuit compatible withthe Nyquist diagram recorded for CPE and Fe3O4NPs/MWCNTs/CPEis depicted in Fig. 2d. In this circuit, Rs, C and Rct represent solutionresistance, a capacitance for the double-layer and electron transferresistance, respectively. W is a finite-length Warburg short-circuitterm coupled to Rct, which accounts for the Nernstian diffusion. Rct atthe electrode surface is equal to the semicircle diameter, which can beused to describe the interface properties of the electrode. The semicircleof Fe3O4NPs/MWCNTs/CPE was smaller than that of CPE and MWCNTs/CPE. The estimated Rct for CPE, MWCNTs/CPE and Fe3O4NPs/MWCNTs/CPE are 5.1, 3.5 and 3.1 kΩ, respectively. These results indicate thatthe presence of Fe3O4NPs/MWCNTs is effective in enhancing the rateof electron transfer. The results also indicate that the Fe3O4NPs/MWCNTs firmly modified the electrode surface. The improved perfor-mance of the Fe3O4NPs/MWCNTs/CPE which employs the nano-materials could be due to the facilitation of the electron transferprocess on the electrode surface.

3.3. Electrochemical behavior of Hp and modification effect

Fig. 3 illustrates the cyclic voltammetric responses of 5.0 μmol L−1

Hp in the B–R buffer (pH = 7.5) with the scan rate of 25 mV s−1 onthe surface of the bare CPE, MWCNTs/CPE and Fe3O4NPs/MWCNTs/

(b and c). EDS spectrum of Fe3O4NPs/MWCNTs/CPE on Au substrate.

Fig. 2.Nyquist plots for different electrodes in Fe(CN)63−/4− containing B–R buffer (pH= 7.5): (a) unmodified CPE, (b) MWCNTs/CPE and (c) Fe3O4NPs/MWCNTs/CPE. Equivalent circuitfor the system (d).

267H. Bagheri et al. / Materials Science and Engineering C 37 (2014) 264–270

CPE. Cyclic voltammogram at the bare CPE, did not exhibit anyvoltammetric peak either without prior preconcentration or followingpreconcentration under open circuit conditions (Fig. 3a). However, adistinct anodic peak with good peak current for Hp is observed at theMWCNTs/CPE and Fe3O4NPs/MWCNTs/CPE (Figs. 3c and d). As can beseen in Fig. 3c and d, the anodic peak potential of Hp is about 0.52 Vvs. Ag/AgCl. Results of many previous works showed that the uniqueproperties of MWCNTs, such as high specific surface area, subtle elec-tronic properties, strong adsorption ability and presence of the func-tional groups exhibit excellent properties for the MWCNT-modifiedelectrodes [17–23]. A comparison of oxidation peak currents for Hp atthe MWCNTs/CPE (Fig. 3c) and Fe3O4NPs/MWCNTs/CPE (Fig. 3d) indi-cates an enhancement of nearly 4 μA (almost two times) in the anodicpeak current at Fe3O4NPs/MWCNTs/CPE, which is due to the increasein its surface area as compared with MWCNTs/CPE. Similarly, the cyclicvoltammogramof Fe3O4NPs/MWCNTs/CPE is recorded in the absence ofHp (Fig. 3b). The results obtained clearly show that the combination ofMWCNTs and Fe3O4NPs definitely improves the sensitivity of modifiedelectrode to Hp. The higher peak current at the Fe3O4NPs/MWCNTs/CPE with respect to other electrodes, indicates that Fe3O4NPs can effec-tively adsorb Hp on the electrode surface. In fact, Fe3O4NPs immobilizedat theMWCNT/CPE surfacewith their large surface area increase the ad-sorptive sites, resulting in a significant increase in the oxidation current.

The effect of sweep rate (υ) on peak potential (Ep) and peak current(ip) of 5.0 μmol L−1 Hp in B–R buffer (pH 7.5) at different sweep ratesranging from 10 to 100 mV s−1 was studied at the Fe3O4NPs/MWCNTs/CPE (Fig. S1, Supplementary file). The ip was directly

Fig. 3. Cyclic voltammograms of 5.0 μmol L−1 Hp with a scan rate of 25 mV s−1 in B–Rbuffer (pH 7.5) on the surface of various electrodes. Curves (a) CPE and (c) MWCNTs/CPE in the presence of analyte, (b) Fe3O4NPs/MWCNTs/CPE in the absence of analyteand (d) Fe3O4NPs/MWCNTs/CPE in the presence of analyte.

proportional to the scan rate (υ) with a correlation coefficient (r) of0.999 (Fig. S1), which suggested an adsorption-controlled process onthe surface of the modified electrode through the porous Fe3O4NPs/MWCNT composite [28].

3.4. Optimization of experimental conditions

In order to develop a voltammetric methodology for determiningthe Hp, we selected the differential pulse and square wave modes,since the peaks were sharper and better-defined at lower concentra-tions of Hp than those obtained by CV,with a lower background current.Also, for obtaining the optimum experimental conditions, some vari-ables affecting the peak current including supporting electrolyte, pH,deposition time and amount of modifier were studied.

A series of supporting electrolytes were tested (B–R, phosphate andacetate buffer solutions). Both the peak height and the peak shape weretaken into consideration when choosing the supporting electrolyte. Ofthese, B–R buffer solution gave the best response. The electrochemistryof Hpmolecule is generally pH dependent. Thus, its electrochemical be-havior was studied at different pHs using CV. It was observed that theanodic peak potentials of Hp using Fe3O4NPs/MWCNTs/CPE shift toless positive values with increasing pH (suggesting participation of H+

in the oxidation process) (Fig. S2a). It shows potential-pH diagramsconstructed by plotting the anodic potential values as the function ofpH. As can be seen, the slope is 0.054 V/pH for Ep, indicating that the sys-tem obeys the Nernst equation for an equal electron and proton in theelectrooxidation of Hp at the surface of the modified electrode [28].The effect of buffer pH on the electrochemical responses of Hp at themodified electrode was investigated in the pH range from 4.5 to 8.5.The peak current increased until it attained the maximum at pH 7.5,then it decreased. As the pH increases, the responses are enhanceddue to the deprotonation. According to the literature [29], the pKa

value of Hp is 8.1. In general, when the pH of the solution was muchlower than the pKa, owing to the strong protonation, the oxidationmight become difficult and responses were lower. It was concludedthat the anodic peak current increased as pH rises from 4.5 to 7.5(Fig. S2b). However, further increase in the pH of the buffer solution de-creases slowly the peak current. When the pH of solution exceeded thepKa or increased further, solubility of Hp decreased gradually and itsprecipitation occurred, which could affect the accumulation processand resulted in the decrease in analytical signal. The best analyticalresponse for Hp is observed in solutions with pH 7.5, which can bepresented as a figure of merit for the prepared sensor in practicalapplications.

Accumulation step is usually a simple and effective way to enhancethe sensitivity of the determination. Based on the adsorption of Hp at

268 H. Bagheri et al. / Materials Science and Engineering C 37 (2014) 264–270

the Fe3O4NPs/MWCNTs/CPE, the effect of accumulation time underopen circuit was investigated in a buffered solution with pH 7.5. The ef-fect of accumulation time on the peak current in the range from 30 to300 swas investigated. The peak current increased rapidlywith increas-ing accumulation time in the first 150 s and then enhanced slowly. Thisbehavior may be attributed to the complete coverage of the modifiedelectrode surfacewith theHp. Taking account of sensitivity and efficiencyand also response repeatability, the accumulation timewas set at 240 s inthe following experiments.

The effect of the amount of Fe3O4NPs/MWCNT within the carbonnanotube paste electrode was evaluated. The peak current intensityincreased with the increasing of the amount of modifier. At 20.0 mg ofFe3O4NPs/MWCNT in CPE, the largest peak current was obtained.

To improve the sensitivity of the determination of Hp further, the in-fluences of parameters of DPV and SWV on the measurements of Hpwere studied and for analytical applications, the following parameterswere employed: DPV pulse amplitude, 50 mV; pulse width, 0.04 s;SWV pulse amplitude, 20 mV; and frequency, 55 Hz.

3.5. Analytical performances

The CPE modified with Fe3O4NPs/MWCNTs was used for the assess-ment of the analytical utility of the proposedmethod as a means for theanalysis of trace amounts of Hp. The DPV and SWV methods using the

Fig. 4.DP (a) and SW(b) voltammograms of different concentrations of Hp in 0.1mol L−1 B–R (of Hp concentration in the ranges of 1.2 × 10−3–0.52 μmol L−1 and 6.5 × 10−4–0.52 μmol L−1

suggested modified electrode were applied as a method featuring highsensitivity, very low detection limit and good selectivity in the presenceof other compounds in B–R buffer solution with pH 7.5. Quantitativeevaluation is based on the linear correlation between the peak currentand concentration. Linear calibration curves were obtained for Hp inthe range of 1.2 × 10−3–0.52 μmol L−1 for DPV with a regression equa-tion of Ip (μA) = 44.659 C (μmol L−1) + 0.031 and R2 = 0.9983(Fig. 4a) and in the range of 6.5 × 10−4–0.52 μmol L−1 for SWV, withthe regression equation, Ip (μA) = 48.417 C (μmol L−1) + 0.029and R2 = 0.9982 (Fig. 4b). The detection limits (based on S/N = 3)of the procedures were found to be 1.33 × 10−4 and for SWV and7.02 × 10−4 μmol L−1 for DPV. The obtained values by this methodare considerably better than values reported by other modifiedelectrodes.

3.6. Interference studies

Interference studies were carried out with several chemical sub-stances prior to the application of the proposed method for the assayof Hp in urine, serum and pharmaceutical samples. The influence of var-ious substances as potential interference compounds on the determina-tion of 0.1 μmol L−1 Hp under the optimum conditions was studied.Tolerance limit was defined as themaximum concentration of the inter-fering substance that caused an error less than 4% for determination of

pH 7.5) and under optimum conditions. Insets show the plots of peak current as a function, respectively.

Table 2Determination of Hp (μmol L−1) in the biological samples under optimum conditions(N = 5).

Technique Sample Hp[added]

Hp[found]

Recovery%

RSD%

HPLCmethod [30]

DPV Urine 0.00 0.76 – 1.40 0.742.00 2.80 102.0 0.85 2.72

SWV 0.00 0.79 – 1.422.00 2.81 101.0 0.83

DPV Blood serum 0.00 0.12 – 1.95 0.132.00 2.10 99.0 1.36 2.12

SWV 0.00 0.11 – 1.812.00 2.10 99.5 0.75

DPV Urine 0.00 0.18 – 1.79 0.201.00 1.17 99.0 1.22 1.18

SWV 0.00 0.18 – 1.801.00 1.18 100.6 1.20

DPV Blood serum 0.00 0.07 – 2.03 0.081.00 1.05 98.0 1.35 1.07

SWV 0.00 0.08 – 2.001.00 1.07 99.0 1.31

269H. Bagheri et al. / Materials Science and Engineering C 37 (2014) 264–270

Hp. The results are given in Table S1, which shows that the peak currentof Hp is not affected by all conventional cations, anions, and organicsubstances.

Also, the interference effects of 60.0 μmol L−1 methyl hydroxy-benzoate (methylparaben) and propyl hydroxybenzoate (propyl-paraben) as preservatives in oral and injectable solutions of Hp weretested on the voltammetric response of 0.1 μmol L−1 Hp. No changesin response currents of Hp were observed in the presence ofmethylparaben and propylparaben or their mixtures. Furthermore,the interference effects of 4-chlorophenyl-4-hydroxypiperidine andthe product of the reduction of Hp as major metabolites of Hp were in-vestigated. Under the experimental conditions, they had no influenceon the signal of Hp in their mixture. When compared to the reportedmethods by Tuzhi et al. [14] and Huang et al. [15], the results reportedhere demonstrate a very high selectivity to Hp. Therefore, in thisstudy it was proved that this method can be successfully applied forthe determination of Hp in the presence of the other interference com-pounds in the clinical preparations.

3.7. Reproducibility and stability of the modified electrode

The reproducibility about the preparation procedure of Fe3O4NPs/MWCNTs/CPE was evaluated by determination of 0.1 μmol L−1 Hpsolution using four independently prepared modified electrodes basedon the same fabrication procedure and the resulted RSD was 3.17%(average of three determinations on each electrode). These results con-firmed the good reproducibility of prepared electrode in both the prep-aration procedure and the voltammetric determinations. The RSD of thepeak currents of 0.1 μmol L−1 Hp for 5 repeated determinations withthe same modified electrode was also 1.93%. Furthermore, with usingthe Fe3O4NPs/MWCNTs/CPE daily while storing the modified electrodeunder ambient conditions, the electrode retained 95.8% of its initialpeak current response to 0.1 μmol L−1 Hp after a period of one month,indicating long-term stability of the electrode composite.

Table 3Determination of Hp in commercial tablets and injection ampule under optimum conditions (N

Technique Sample Hp [added] H

DPV Tablet (2.5 mg/tablet) 0.00 2.2.00 4.

SWV Tablet (2.5 mg/tablet) 0.00 2.2.00 4.

DPV Injection Hp (5 mg mL−1) 0.00 5.2.00 7.

SWV Injection Hp (5 mg mL−1) 0.00 5.2.00 7.

3.8. Analytical applications

The applicability of the prepared modified electrode was tested bydetermination of Hp in urine, serum, injection and tablet samples. Theresults presented in Tables 2 and 3 indicated that themodified electroderetained its efficiency for the determination of Hp in various sampleswith satisfactory results. Despite the presence of complex matrix inthe studied real samples, they had no interference effects on the recov-ery percentage obtained by the suggested methods. Besides, the recov-ery studies of the spiked Hp in human blood serum and urine samplesshowed average values in the range from 97.5 to 102.0% (Tables 2and 3), suggesting the successive applicability of the proposed strategyfor the clinical applications. In addition, HPLCmethod [30] was used forthe analysis of the samples, to confirm the accuracy of the proposedmethod.

4. Conclusion

Fe3O4 nanoparticles were synthesized and thenMWCNTs were dec-orated with them, using a simple method. We showed that MWCNTsdecorated with Fe3O4 nanoparticles could be used as a suitablemodifierin electrochemical method to determine Hp in various samples. Thisnanostructured electrode is highly sensitive and can detect Hp at wellbelow the level reported by previous papers (Table 1). Despite thepresence of complex matrixes in the pharmaceutical samples and bio-logical fluids, the percentages of recovery, obtained by SWV and DPwere satisfactory. The suggested sensor offers good precision and accu-racy, and is especially suitable for routine studies of Hp in different sam-ples, or even better similar data observed by the use of HPLC.

Acknowledgments

The authors wish to thank the Researches and Technology Council,Baqiyatallah University of Medical Sciences, Tehran, for valuable finan-cial and facility supports.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.msec.2014.01.023.

References

[1] S. Mennickent, L. Pino, M. Vega, C.G. Godoy, M. Diego, Quantitative determination ofhaloperidol in tablets by high performance thin-layer chromatography, J. Sep. Sci. 30(2007) 772–777.

[2] H. Ebrahimzadeh, Z. Dehghani, A.A. Asgharinezhad, N. Shekari, K. Molaei, Determi-nation of haloperidol in biological samples using molecular imprinted polymernanoparticles followed by HPLC-DAD detection, Int. J. Pharm. 453 (2013) 601–609.

[3] A.J. Giannini, N.A. Underwood, M. Condon, Acute ketamine intoxication treated byhaloperidol: a preliminary study, Am. J. Ther. 7 (2000) 389–392.

[4] S. Kapur, S.C. VanderSpek, B.A. Brownlee, J.N. Nobrega, Antipsychotic dosing in pre-clinical models is often unrepresentative of the clinical condition: a suggested solu-tion based on in vivo occupancy, J. Pharmacol. Exp. Ther. 305 (2003) 625–631.

[5] G.D. Zhang, A.V. Terry Jr., M.G. Bartlett, Sensitive liquid chromatography/tandemmass spectrometry method for the simultaneous determination of olanzapine, ris-peridone, 9-hydroxyrisperidone, clozapine, haloperidol and ziprasidone in ratbrain tissue, J. Chromatogr. B 858 (2007) 276–281.

= 5).

p [found] Recovery % RSD % HPLC method [30]

43 – 0.93 2.4238 97.5 0.77 4.4045 – 1.4242 98.5 0.8303 – 0.88 5.0505 101.0 0.75 7.0802 – 0.7304 101.0 0.73

270 H. Bagheri et al. / Materials Science and Engineering C 37 (2014) 264–270

[6] T. Arinobu, H. Hattori, M. Iwai, A. Ishii, T. Kumazawa, O. Suzuki, H. Seno, Liquidchromatographic–mass spectrometric determination of haloperidol and its metabo-lites in human plasma and urine, J. Chromatogr. B 776 (2002) 107–113.

[7] N. Yasui-Furukori, Y. Inoue, M. Chiba, T. Tateishi, Simultaneous determination of hal-operidol and bromperidol and their reduced metabolites by liquid–liquid extractionand automated column-switching high-performance liquid chromatography,J. Chromatogr. B 805 (2004) 175–180.

[8] R. Driouich, T. Takayanagi, M. Oshima, S. Motomizu, Separation and determinationof haloperidol, parabens and some of their degradation products bymicellar electro-kinetic chromatography, J. Chromatogr. A 903 (2000) 271–278.

[9] S.A. Özkan, B. Uslu, H.Y. Aboul-Enein, Analysis of pharmaceuticals and biologicalfluids using modern electroanalytical techniques, Crit. Rev. Anal. Chem. 33 (2003)155–181.

[10] B. Uslu, S.A. Ozkan, Electroanalytical methods for the determination of pharmaceu-ticals: a review of recent trends and developments, Anal. Lett. 44 (2011)2644–2702.

[11] J.C. Vire, M. Fischer, G.J. Patriarche, G.D. Christian, Electrochemical behaviour ofsome neuroleptics: haloperidol and its derivatives, Talanta 28 (1981) 313–317.

[12] H.S. El-Desoky, M.M. Ghoneim, Assay of the anti-psychotic drug haloperidol in bulkform pharmaceutical formulation and biological fluids using square-wave adsorp-tive stripping voltammetry at a mercury electrode, J. Pharm. Biomed. Anal. 38(2005) 543–550.

[13] F.W.P. Ribeiro, J.E.S. Soares, H. Becker, D. De Souza, P. de Lima-Neto, A.N. Correia,Electrochemical mechanism and kinetics studies of haloperidol and its assay incommercial formulations, Electrochim. Acta 56 (2011) 2036–2044.

[14] P. Tuzhi, Y. Zhongping, L. Rongshan, Voltammetricmeasurement of haloperidol follow-ing adsorptive accumulation at glassy-carbon electrodes, Talanta 38 (1991) 741–745.

[15] F. Huang, Y.Y. Peng, G.Y. Jin, S. Zhang, J.L. Kong, Sensitive detection of haloperidoland hydroxyzine at multi-walled carbon nanotubes-modified glassy carbon elec-trodes, Sensors 8 (2010) 1879–1889.

[16] P.D. Schumacher, K.A. Fitzgerald, J.O. Schenk, S.B. Clark, Preconcentration off-elements from aqueous solution utilizing a modified carbon paste electrode,Anal. Chem. 83 (2011) 1388–1393.

[17] H. Bagheri, A. Afkhami, H. Khoshsafar, M. Rezaei, A. Shirzadmehr, Simultaneous elec-trochemical determination of heavy metals using a triphenylphosphine/MWCNTscomposite carbon ionic liquid electrode, Sensors Actuators B 186 (2013) 451–460.

[18] A. Afkhami, H. Bagheri, A. Shirzadmehr, H. Khoshsafar, P. Hashemi, A potentiometricsensor for Cd2+ based on carbon nanotube paste electrode constructed from roomtemperature ionic liquid, ionophore and silica nanoparticles, Electroanalysis 24(2012) 2176–2185.

[19] A.A. Ensafi, A.R. Allafchian, B. Rezaei, Multiwall carbon nanotubes decorated withFeCr2O4, a new selective electrochemical sensor for amoxicillin determination,J. Nanopart. Res. 14 (2012) 1244–1245.

[20] A. Afkhami, H. Bagheri, H. Khoshsafar, M. Saber-Tehrani, M. Tabatabaee, A.Shirzadmehr, Simultaneous trace-levels determination of Hg(II) and Pb(II) ions invarious samples using amodified carbon paste electrode based onmulti-walled car-bon nanotubes and a new synthesized Schiff base, Anal. Chim. Acta. 746 (2012)98–106.

[21] H. Bagheri, A. Afkhami, M. Saber-Tehrani, A. Shirzadmehr, S. Waqif Husain, H.Khoshsafar, M. Tabatabaee, Novel sensor fabrication for the determination ofnanomolar concentrations of Ce3+ in aqueous solutions, Anal. Methods 4 (2012)1753–1758.

[22] A. Afkhami, H. Khoshsafar, H. Bagheri, T. Madrakian, Construction of a carbon ionicliquid paste electrode based on multi-walled carbon nanotubes-synthesized Schiffbase composite for trace electrochemical detection of cadmium, Mater. Sci. Eng. C35 (2014) 8–14.

[23] S. Shahrokhian, S. Rastgar, M.K. Amini, M. Adeli, Fabrication of a modified electrodebased on Fe3O4NPs/MWCNT nanocomposite: application to simultaneous determi-nation of guanine and adenine in DNA, Bioelectrochemistry 86 (2012) 78–86.

[24] H. Yin, Y. Zhou, Q. Ma, Sh. Ai, Q. Chen, L. Zhu, Electrocatalytic oxidation behavior ofguanosine at graphene, chitosan and Fe3O4 nanoparticles modified glassy carbonelectrode and its determination, Talanta 82 (2010) 1193–1199.

[25] F. Valentini, A. Amine, S. Orlandocci, M.L. Terranova, G. Palleschi, Carbon nano-tubepurification: preparation and characterization of carbon nanotube paste electrodes,Anal. Chem. 75 (2003) 5413–5421.

[26] H. Huang, X.M. Liu, X.Y. Zhang, W.I. Liu, X.G. Su, Z.Q. Zhang, Fabrication of newmag-netic nanoparticles (Fe3O4) grafted multiwall carbon nanotubes and heterocycliccompound modified electrode for electrochemical sensor, Electroanalysis 22(2010) 433–438.

[27] A.A. Ensafi, H. Karimi-Maleh, S. Mallakpour, A new strategy for the selective deter-mination of glutathione in the presence of nicotinamide adenine dinucleotide(NADH) using a novel modified carbon nanotube paste electrode, Colloids Surf. B104 (2013) 186–193.

[28] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications,2nd ed. Wiley, New York, 2001.

[29] A. Bouklouze, M. Elbouzekraoui, Y. Cherrah, M. Hassar, J.M. Kauffmann, Potentio-metric sensor for hydroxyzine determination, Electroanalysis 14 (2002) 1369–1374.

[30] H. Trabelsi, S. Bouabdallah, K. Bouzouita, F. Safta, Determination and degradationstudy of haloperidol by high performance liquid chromatography, J. Pharm. Biomed.Anal. 29 (2002) 649–657.