An electrochemical sensor for warfarin determination based on covalent immobilization of quantum dots onto carboxylated multiwalled carbon nanotubes and chitosan composite film modified electrode

Materials Science and Engineering C 57 (2015) 77–87

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

An electrochemical sensor for warfarin determination based on covalent immobilization of quantum dots onto carboxylated multiwalled carbon nanotubes and chitosan composite film modified electrode Mohammad Bagher Gholivand ⁎, Leila Mohammadi-Behzad Department of Analytical Chemistry, Faculty of Chemistry, Razi University, Kermanshah, Iran

a r t i c l e

i n f o

Article history: Received 23 January 2015 Received in revised form 23 May 2015 Accepted 10 July 2015 Available online 20 July 2015 Keywords: Warfarin CdS-quantum dots Multiwalled carbon nanotubes Chitosan Modified glassy carbon electrode

a b s t r a c t A method is described for the construction of a novel electrochemical warfarin sensor based on covalent immobilization of CdS-quantum dots (CdS-QDs) onto carboxylated multiwalled carbon nanotubes/chitosan (CS) composite film on the surface of a glassy carbon electrode. The CdS-QDs/CS/MWCNTs were characterized by field-emission scanning electron microscopy (FE-SEM), transmission electron microscopy (TEM), Fourier transform infra-red (FTIR) spectroscopy, XRD analysis and electrochemical impedance spectroscopy (EIS). The sensor showed optimum anodic stripping response within 90 s at an accumulation potential of 0.75 V. The modified electrode was used to detect the concentration of warfarin with a wide linear range of 0.05–80 μM and a detection limit (S/N = 3) of 8.5 nM. The proposed sensor has good storage stability, repeatability and reproducibility and was successfully applied for the determination of warfarin in real samples such as urine, serum and milk. © 2015 Published by Elsevier B.V.

1. Introduction Warfarin [3-(α-acetonylbenzyl)-4-hydroxycoumarin, WAR] is a coumarin derivative, and widely used as an oral anticoagulant in various cardiovascular and cerebrovascular disorders such as venous thromboembolism, pulmonary embolism, atrial fibrillation, valvular heart disease and coronary heart diseases [1–3]. WAR is a drug with a narrow therapeutic index and warrants careful monitoring of the patient. In order to ensure the effectiveness and safety of oral anticoagulants, the dose must be adjusted accurately and frequently. Therefore, the therapeutic window of warfarin is very narrow. Exceeding the therapeutic window of this drug causes unwanted bleedings [4]. WAR is a weakly acidic drug (pKa = 5.19) with an enolic group [5] and it was known to inhibit tumor spread, and to stimulate granulocytes, lymphocytes and macrophages [6]. Therapeutic concentration of WAR is about 2.0–5.0 μg ml−1 and has a long half-life (20–60 h) [7]. A number of methods have been used for the determination of WAR including high performance liquid chromatography [8–15], liquid chromatography [16,17], capillary electrophoresis [18], phosphorescence [19], fluorescence [20], cloud point extraction [21], and spectrofluorimetric [22]. Electrochemical techniques are alternative methods for the WAR determination because they are simple, fast, sensitive and low cost. The electrochemical detection of WAR

⁎ Corresponding author. E-mail address: [email protected] (M.B. Gholivand).

http://dx.doi.org/10.1016/j.msec.2015.07.020 0928-4931/© 2015 Published by Elsevier B.V.

has been reported using multiwall carbon nanotubes/molecular imprinting polymer [23] and hanging mercury drop [24] as the working electrodes. It is known that the modification of conventional electrodes has attracted much attention in the last 2 decades because it provides powerful means to bring new qualities to the electrode surface which was exploited for electrochemical purposes [25–31]. Chemically modified electrodes can be obtained by attaching a suitable electron mediator on electrode surface and can be applied in various fields including electroanalysis and electrocatalysis [32–43]. Among the wide range of electrode modifiers, carbon nanotubes have attracted the attention of electrochemists because of their advantageous features such as excellent long term stability, high conductivity, resistance to surface fouling, providing large surface area and their ability to promote electrontransfer process [44–46]. Chitosan (CS) has been gradually used for constructing sensors due to its attractive properties that include excellent film-forming ability, high permeability, good adhesion, nontoxicity, cheapness and a susceptibility to chemical modification. It also facilitates the electron transfer after its swelling in the reaction mixture due to its hydrophilic nature [47,48]. Besides, chitosan can be used as a dispersant to form a stable CNT–chitosan composite which can form a stable film on the electrode surface. Quantum dots (QDs) are semiconductor nanocrystals that possess a size-tunable optical and electronic properties, which have been used in several areas, including catalysis, coatings, textiles, data storage, biotechnology, health care, biomedical, pharmaceutical industries and

78

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

most recently, in bioanalytical chemistry [49]. The surface modification of QDs can change their optical, chemical, electrochemical and photocatalytic properties [50]. Therefore, QDs modified with different functional groups on the surface could provide a new chemistry for the application in electroanalytical chemistry [51]. One of these modifiers is cysteine, which its capped QD has been used as a fluorescent probe for copper ions [52]. Thus, it seems that cysteine can also be used as a proper surface modifier in the electrochemical field. To a better platform for nanoscale sensing, the capped QDs can be attached to a suitable nanoparticle such as MWCNTs [53]. The development of practical strategies for assembling QDs onto MWCNT surface is an area of considerable interests which its combination exhibits synergistic effects towards target analysis. In the present work, the advantages of using CdS-QD/CS/MWCNT/ GC electrode combined with differential pulse stripping voltammetry (DPSV) are presented for the analytical determination of WAR. In order to enhance the electrode sensitivity and electronic transmission, multiwall carbon nanotubes (MWCNTs) immobilized on the surface of GCE. The L-cysteine capped CdS quantum dots (CdS-QDs) for creating a nanostructured platform and chitosan as the stabilizing agent to prevent the MWCNT aggregation are used as further modifiers. However, due to the synergistic effects of the modifiers, the combination of MWCNTs with chitosan and CdS-QDs improved the electron transport activity of the composite film as catalyst for analyte analysis. On the other hand the composite film facilitated the electron transfer more than those of MWCNTs, chitosan or CdS QDs alone [48,54–56]. The behavior of the modified electrode and its application for WAR electroanalysis were investigated by cyclic voltammetry. The prepared modified electrode was successfully applied for voltammetric determination of low level of WAR in real samples. 2. Experimental 2.1. Chemicals Carboxylated multiwall carbon nanotubes with purity 95% (10 nm diameters) and 1–2 μm length were obtained from DropSens (Llanera, Spain). Warfarin sodium, N-hydroxysuccinimide (NHS), 1-ethyl-3-(3dimethylaminopropyl) carbodiimide hydrochloride (EDC), chitosan (95% deacetylation) and cysteine were purchased from Sigma-Aldrich (St. Louis, USA). All reagents were of analytical-reagent grade and used without further purification. Double distilled water was used thoroughly. 2.2. Apparatus Electrochemical experiments were performed via using a μAutolab III (Eco Chemie B.V.) potentiostat/galvanostat by NOVA 1.8 software. A conventional three-electrode cell was used with a saturated Ag/AgCl as reference electrode, a Pt wire as counter electrode and a modified glassy carbon (1.8 mm diameter) as working electrode. The cell was a one-compartment cell with an internal volume of 10 ml. All experiments were typically conducted at room temperature. JENWAY pH-meter (model 3345) was also applied for pH measurements. To obtain information about the morphology of the electrode surface, transmission electron microscopy (TEM) (Zeiss EM 900), scanning electron microscopy (SEM) (Tscan Company, Czech Republic), X-ray diffractometer (XRD) (X'Pert Pro MPD, PANalytical, Netherlands) and Fourier transform infra-red (FTIR) spectroscopy (ALPHA model of Bruker, Germany) were used. 2.3. Synthesis of CdS-quantum dots by seed assistant technique CdS-quantum dots (QDs) have been synthesized in accordance with the literature published before [57]. The functionalized cysteine capped CdS quantum dots were facilely prepared and described as follows: a

1.0 mM portion of cysteine dissolved in 100 ml of deionized water and then purged with pure nitrogen for 60 min under magnetic stirring. The pH value of the solution was adjusted between 8.5 and 9.0 using 0.5 M Tris solution. Subsequently, 0.50 mM portion of Cd(NO3)2 was dropped slowly into the above solution and reacted for 30 min resulting in a molar ratio cysteine:Cd of 2:1. Finally, 0.50 mM portion of S2− (from a Na2S solution) dissolved in 10 ml water was dropped slowly into the vortex of the solution to reach a molar ratio of S:Cd of 1:1. At this point, the mixture solution is colorless. The seed solution was injected into this mixture solution under strong magnetic stirring, which was obtained by directly mixing 2 ml of 10− 4 M Cd2 + solution and the same amount of S2− solution. All steps were performed under magnetic stirring. The bright yellow-green colloid obtained after the reaction solution was sealed, incubated for 2 h at 47 °C bath water, and then flushed with N2 for 30 min to remove most of the unreacted sulfide after stirring for 30 min. The colloid solution was stored at room temperature without any precipitation during several months. 2.4. Preparation of the modified electrode The procedure of the fabrication of the sensor is illustrated in Scheme 1. To prepare a modified electrode, glassy carbon electrode was polished with emery paper followed by alumina (0.05 μm) and then thoroughly washed with double distilled water. To remove the adsorbed particles the electrode was further cleaned in an ultrasound bath. 5 μl of functionalized MWCNT suspension in DMF (1 mg ml−1) was cast on the surface of GCE and dried in air to form a MWCNT film on the electrode surface. Then, chitosan (0.5% in acetic acid, 200 μl) was added to 10 ml of 1 M KCl and electrodeposited onto the MWCNT/GC electrode through cyclic voltammetry by applying 20 successive deposition cycles at − 0.15 to 0.20 V at a scan rate of 20 mV s−1. Then the resulting electrode was immersed into the solution containing CdS-QDs (pH 7.0) in the presence of 0.1 mM EDC and NHS for 12 h to form CdS-QDs/CS/MWCNTs/GCE. During this process the cysteine capped, CdS-QDs will covalently bond to the carboxylic groups of the functionalized MWCNTs and amine groups of CS. Finally, the modified electrode was washed thoroughly with double distilled water to remove the unbound materials and then after washing it was dried in air and kept at room temperature for further use. 2.5. Characterization of modified electrode by FTIR To record FTIR spectra of the modified electrode at different stages of its construction, the deposited material was scrapped off the GC electrodes, grinded with dry potassium bromide (KBr) and this powder mixture was then pressed in a mechanical press to form a translucent pellet through which the beam of the spectrometer can pass. Then this pellet was kept in the socket of FTIR spectrophotometer and its spectrum was recorded. Band intensities in IR spectrum were expressed as transmittance (T). 2.6. Preparation of real samples Serum, urine and breast milk samples were collected from a 36 year old patient volunteer after using a tablet containing 5 mg warfarin (from APOTEX Co.). 0.5 ml methanol, as serum protein denaturation and precipitating agent, was added to 1 ml of the serum sample. After vortexing for 40 s, the precipitated protein was separated out by centrifugation for 4 min at 10,000 rpm. The clear supernatant layer was filtered through a 0.45 μm Milli-pore filter to produce a protein-free human serum. A 200 μl of this solution was transferred into a 10.0 ml volumetric flasks containing phosphate buffer (pH = 2) and different amounts of standard solution of warfarin. After adjusting the volume of solutions, their warfarin contents were determined using an optimized proposed procedure.

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

79

Scheme 1. Schematic representation of the stepwise fabrication process of CdS-QD/CS/MWCNT/GC electrode.

The urine sample was filtered through a 0.45 μm Milli-pore filter. The drug content of 1.5 ml of filtrated urine was extracted into 0.7 ml ethyl acetate. The extract was evaporated and the residual was dissolved in 0.5 ml ethanol. A 200 μl of this solution was transferred into a 10.0 ml volumetric flasks containing phosphate buffer (pH = 2) and different amounts of standard solution of warfarin. After adjusting the volume of the solutions, their warfarin contents were determined using an optimized proposed procedure. A 200 μl of milk human sample of a patient volunteer was transferred into a 10.0 ml volumetric flasks containing phosphate buffer (pH = 2) and different amounts of standard solution of warfarin and stirred for 5.0 min. After adjusting the volume of solutions, their warfarin contents were determined using an optimized proposed procedure.

2.7. General method for electrochemical measurements A 10 ml volume of phosphate buffer solution (PBS) with pH 2.0 was transferred into the voltammetric cell and deaerated with high-purity nitrogen for 5 min. An accumulation potential of 0.75 V was applied to the electrode for 90 s while the solution was stirred at 600 rpm. The stirring was stopped for a period of 10 s (equilibration time), and then the potential was scanned from 0.7 to 1.15 V using DPV and its voltammogram for background correction. Then, an appropriate volume of the sample solution was added to the voltammetric cell and the same procedure was used for recording its voltammogram. After each experiment, the electrode was washed with buffer solution (PBS buffer, pH 2.0).

3. Results and discussion 3.1. Characterization of MWCNT, CS/MWCNT and CdS-QD/CS/MWCNT nanocomposites X-ray power diffraction analysis method was employed to investigate the formation of composite film. The presence of MWCNTs, CS, and CdS-QDs in the composite was confirmed by their characteristic peaks (Fig. 1). The peak at 2θ = 25.6 can be ascribed to the reflection of the MWCNTs [58]. The chitosan characteristic peaks are at 2θ = 10.4 and 19.9 [59]. In Fig. 1, the chitosan/MWCNT pattern shows the characteristic peaks of chitosan with slight shifts at 2θ = 8.4 and 20.6. It is illustrated that chitosan has been decorated onto the surface of MWCNTs. Meanwhile, the diffraction peaks at 2θ = 26.3, 28.2, 43.7 and 47.5 were attributed to the hexagonal CdS phase [60], which revealed the attachment of QDs in the chitosan/MWCNT composite. TEM images of the CS/MWCNTs, CdS-QDs and CdS-QDs/CS/MWCNTs as another morphological property indicator are shown in Fig. 2. As shown, the CS was coated uniformly on the sidewalls of the carbon nanotube (Fig. 2A) and the L-cysteine capped CdS has a relatively uniform spherical shape with serration around the spheres and average particle diameter of 30–40 nm (Fig. 2B). The conjugation of the CS/ MWCNTs with QDs has resulted to the modification of the nanotubes and CS surface. Conjugation of CS/MWCNTs with QDs is fairly evident (Fig. 2C). The surface morphology of the MWCNT/GC and CS/MWCNT/GC electrodes was investigated by scanning electron microscopy (SEM) (Fig. 3). The SEM image of the MWCNT/GC electrode showed a net structure without aggregation that was evenly coated on the surface

80

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

Fig. 1. X-ray diffraction patterns of CdS-QD/CS/MWCNT composite.

Fig. 2. TEM images of CS/MWCNTs (A), CdS-QDs (B) and CdS-QDs/CS/MWCNTs (C).

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

81

Fig. 3. SEM images of (A and C) MWCNT/GC, (B and D) CS/MWCNT/GC and (E) CdS-QD/CS/MWCNT/GC electrodes.

of the electrodes (A and C). After electrochemical deposition of CS film on the surface of the MWCNT/GC electrode, its surface changed, indicating the existence of CS at the MWCNT surfaces (B and D). Moreover, on the CS/MWCNT/GCE surface, the CS film attached on the CNTs is very smooth and homogeneous, which may enhance the interaction between the modified electrode and the WAR, and further improve the sensitivity and stability of the CS/MWCNTs/GCE. Further modification of the resulted electrode with CdS-QDs, the SEM image was again altered and showed a film with relatively homogeneous distribution of globular structure (Fig. 3E), indicating the linkage of L-cysteine capped CdS-QD to MWCNT surface. Electrochemical impedance spectroscopy (EIS) can provide useful information on the impedance changes of the electrode surface during the fabrication process. The electron-transfer resistance (Ret) at the electrode surface is equal to the semicircle diameter of the Nyquist plots and can be used to describe the interface properties of the electrode. Fig. 4 shows the typical Nyquist plots recorded for bare GCE (curve a), MWCNTs/GCE (curve b), CS/MWCNTs/GCE (curve c) and CdS-QDs/CS/MWCNTs/GCE (curve d). The electron transfer resistance (Ret) of the bare GCE was 450 Ω (Fig. 4a) while, after modification of

the GCE with MWCNTs, it reduced to 54 Ω (curve b). The increase in the Ret value of the CS/MWCNTs/GCE (200 Ω) is an evidence for electrodeposition of CS on the MWCNTs/GCE (curve c). A further increase in the electron transfer resistance of the CS/MWCNT/GC electrode to 280 Ω after modification by CdS-QDs suggested that the semiconductor film was successfully immobilized on the electrode surface (Fig. 4d). This observed increase can be attributed to the electrostatic repulsion between [Fe(CN)6]3−/4− redox probe and carboxyl moiety of cysteine capped CdS-QDs. 3.2. FTIR spectra Fig. 5 shows the FTIR spectra of the MWCNT/GC electrode (curve a), MWCNT/CS/GC electrode (curve b) and CdS-QD/MWCNT/CS/GC electrode (curve c). Curve a shows the FTIR spectrum of the MWCNT/GC electrode, revealing several significant peaks. The peak at the 1434 cm− 1 corresponds to the stretching mode of the C_C double bond that forms the framework of the carbon nanotube sidewall. Also the peaks at 1706 and 1110 cm− 1 apparently correspond to the stretching modes of the carboxylic acid groups. FTIR spectra after

82

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

along with N\\H deformation mode and 1110 cm−1 to the stretching vibration mode of the hydroxyl group. The FTIR spectrum of the CdSQD/MWCNT/CS/GC electrode (curve c) shows the appearance of additional bands at 1530 cm−1 assigned to the carbonyl stretch, indicating the CdS-QD binding. 3.3. Electrochemical behavior of warfarin

Fig. 4. Nyquist plots of (a) bare GC, (b) MWCNT/GC, (c) CS/MWCNT/GC and (d) CdS-QD/ CS/MWCNT/GC electrodes in 0.1 M KCl containing 1 mM K3[Fe(CN)6]. EIS conditions: initial potential, 0.15 V; amplitude voltage, 10 mV; and frequency range, 100 kHz to 0.1 Hz.

modification of MWCNTs/GCE with chitosan (MWCNTs/CS/GCE) were accompanied with new peaks (curve c) such as 1710 cm−1 which is assigned to C\\O stretching, 1640 cm−1 is attributed to C\\O stretching

Fig. 6 demonstrates a comparison of cyclic voltammograms (CVs) of 5 μM warfarin in phosphate buffer solution (0.1 M, pH = 2) at the bare GC (curve a), MWCNT/GC (curve b), CS/MWCNT/GC (curve c) and CdS-QD/CS/MWCNT/GC (curve d) electrodes. As can be seen, response at the surface of bare GCE is rather poor and only a weak oxidation peak is observed at 1.05 V (curve a). In order to efficiently use the great properties of MWCNTs, MWCNTs were immobilized on the GCE surface to study the electrochemical responses of WAR. As shown, the anodic current response increases noticeably at the MWCNT/GC (curve b) which shows the significant role of MWCNTs. The CV of CS/ MWCNT/GC (curve c) exhibited higher current than MWCNT/GCE, which shows that CS/MWCNT composite film has a large effective surface area than MWCNTs/GCE. Furthermore change of the resulted electrode (CS/MWCNTs/GCE) with CdS-QDs enhances the response of the electrode about 27 times on that obtained at the bare CGE (curve d) which can be related to the increased active surface area of the modified electrode and accumulation of analyte on the surface of the modified electrode. During the cyclic voltammetric studies, no

Fig. 5. FTIR spectra of MWCNT/GC (a), CS/MWCNT/GC (b) and QD/CS/MWCNT/GC electrodes (c).

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

83

electron transfer rate constant (ks) and transfer coefficient (α) can be determined by the variation of the peak potential with scan rate according to the following equation: EP ¼ E0 þ ð2:303RT=αnFÞ logðRTks Þ þ ð2:303RT=αnFÞ logv:

ð1Þ

The value of αn can be easily calculated from the slope of peak potentials versus the logarithm of the scan rate. The slope was about 0.0451. Using Eq. (1) and the two electrons transferred for warfarin [62], the transfer coefficient (α) of 0.65 was obtained. The value of ks can be determined from the intercept of peak potentials versus the logarithm of the scan rate. The value of E° in Eq. (1) can be obtained from the intercept of the E versus ʋ curve by extrapolating the vertical axis at ʋ = 0 [63]. In this system the intercept for Ep versus log ʋ plot was 0.946 and E° was obtained to be 0.85 V. By introducing these parameter values in Eq. (1), the electron transfer rate constant of 105.4 s−1 was obtained. 3.4. Effect of pH

Fig. 6. The CVs of 5 μM warfarin solution at the (a) bare GCE, (b) MWCNTs/GCE, (c) CS/ MWCNTs/GCE and (d) QDs/CS/MWCNTs/GCE (curve d) in PBS (0.1 M, pH = 2.0) at scan rate = 100 mV s−1.

counterpart cathodic peak observed was accompanied by oxidation of WAR at the surface of all tested electrodes during the reverse scan, which suggests a totally irreversible behavior for the electrode process. Fig. 7A shows the cyclic voltammograms (CVs) of WAR at the QD/CS/ MWCNT/GC electrode in PBS buffer (pH = 2) at different potential sweep rates. As shown not only the peak current of WAR increases with scan rate raising but also its peak potential shifts towards negative values. Furthermore, the anodic peak currents varied linearly with the scan rate in the range of 10–500 mV s−1, suggesting that the WAR oxidation follows an adsorption controlled process (Fig. 7B). For an irreversible electrode process, according to Laviron's theory [61] the

A

The influence of buffer pH on the response of 5 μM WAR in 0.1 M phosphate buffer solution was investigated in the pH range from 2 to 8. As shown in Fig. 8, the anodic peak current decreased with pH increasing from 2 to 8. In addition, it was observed that the anodic peak potential of WAR shifted negatively with increasing pH, suggesting that H+ participates in the electrooxidation process. The linear relationship between anodic peak potential at the CdS-QDs/CS/MWCNTs/GCE and pH can be described by the following equation:   EPA ðmVÞ ¼ −49pH þ 1110:5 R2 ¼ 0:9994 :

ð2Þ

Regarding the slope value of −49 mV per pH unit, which is close to the theoretical slope (−59 mV per pH unit), it will be concluded that an equal number of electrons and protons is involved in the electrooxidation of WAR at the surface of the CdS-QDs/CS/MWCNTs/GCE. Thus according to the above results the oxidation mechanism of WAR is suggested in Scheme 2.

B

Fig. 7. The CVs of 5 μM warfarin solution at the surface of QDs/CS/MWCNTs/GCE at various scan rates (ʋ, 10–500 mV s−1). Inset: Plot of Ip versus ʋ; in PBS at pH 2.

84

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

Fig. 8. The effect of pH on the peak current of 5 μM warfarin at the surface of QDs/CS/MWCNTs/GCE; scan rate = 100 mV s−1. Inset: Dependence of the peak potential (EPA) with pH solution.

Similar mechanism was reported for electrooxidation of a drug that has a similar structure to WAR [62]. 3.5. Optimization of the experimental parameters Accumulation potential is an important parameter for stripping techniques and has a non-negligible influence on the sensitivity of determination. The effect of accumulation potential on the redox peak currents of 5 μM WAR was examined over the potential range of 0.3 to 1.0 V. The redox peak currents increased up to 0.75 V, and then decreased. Thus, an accumulation potential of 0.75 V was chosen for subsequent use. The influence of accumulation time on the redox peak currents was also investigated. Variations of the accumulated time showed that the peak currents of WAR increased gradually with increasing accumulation times from 0 s to 90 s and then leveled off, presumably due to saturation of the electrode surface. So, 90 s was chosen as the optimal accumulation time. Afterward, the effects of the scan rate on the redox peak currents and stirring rate during the preconcentration period were also tested in the range 10 to 150 mV s − 1 and 50 to 1000 rpm, respectively. The best results were obtained when 80 mV s− 1 and 600 rpm were selected as optimal for the scan rate and stirring rate, respectively. 3.6. Analytical performances Under the optimum conditions described above the peak current of WAR was used for plotting the calibration using differential pulse stripping voltammetry (DPSV). The foregoing method was applied with an accumulation potential (Eac) of 0.75 V and an accumulation time of 90 s. Fig. 9 shows the DPVs obtained at different concentrations of WAR on the CdS-QDs/CS/MWCNTs/GCE. As can be seen, the anodic current increased with WAR concentration. The inset of Fig. 9 shows the corresponding calibration curve, demonstrating that in the range of 0.05 to 80 μM, the anodic peak current has a good linear relationship

O

O

H

O

H 2H+

C OH

CH2COCH3

O

2e-

OH O

Scheme 2. The mechanism oxidation of warfarin.

C CH2COCH3

with WAR concentration. The limit of detection (LOD) and limit of quantification (LOQ) were calculated using the relation ks/m [62], where k = 3 for LOD and 10 for LOQ, s represents the standard deviation of the peak currents of the blank (n = 10), and m represents the slope of the calibration curve for WAR. The LOD and LOQ values were found to be 8.5 and 28 nM, respectively, indicating high sensitivity of the proposed method. Detection limit and linear range of the proposed electrode were compared with other previously reported methods, and the results were summarized in Table 1. As it can be seen, the proposed method has a wider linear range than those reported for WAR determination [11–23].

3.7. Amperometric detection of WAR Amperometric method was used to examine the sensitivity of QD/ CS/MWCNT/GC electrode towards the WAR detection. Fig. S1 shows the amperometric i–t response of the rotated modified electrode (2000 rpm) with successive injection of 10 nM of WAR solution at an applied potential of 1.05 V versus the reference electrode. As shown, during the successive addition of WAR a well-defined response was observed. The amperometric current was increased linearly with increasing concentration of WAR from 10 to 190 nM. The calibration plot over the concentration range of 10–190 nM has a correlation coefficient of 0.998 and a detection limit of 3 nM at signal to noise ratio of 3.

3.8. Repeatability, reproducibility and stability of modified electrode The repeatability of QDs/CS/MWCNTs/GCE was investigated by the measurement of the response to the PBS buffer containing 5 μM WAR. The relative standard deviation (RSD) of the oxidation peak currents by 10 successive measurements was 2.33%. Furthermore, under the same and independent conditions, it was found that the oxidation current of 5 μM WAR almost remained the same by four electrodes with an RSD of 2.26%, indicating good reproducibility. In addition, the longterm stability of the CdS-QDs/CS/MWCNTs/GCE was tested after being stored in dry conditions at room temperature for 16 days. The peak current decreased less than 5% of its original response detected for the same WAR solution (5 μM), indicating the excellent stability of the modified electrode. These experimental results revealed the good stability, repeatability and reproducibility of the proposed modified electrode for WAR monitoring.

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

85

Fig. 9. Effect of different concentrations of warfarin (from bottom to top: 0.05–80 μM) on the response of the modified electrode and its calibration curve in PBS at pH 2 under optimal conditions: Eac = 0.75 V; tac = 90 s; and scan rate = 80 mV s−1.

3.9. Interference studies The influence of various substances as interfering compounds on the WAR detection was studied under optimum conditions. Some common substances in pharmaceutical and/or in biological fluids were tested on the determination of the 5 μM WAR. Tolerance limit was defined as the maximum concentration of the interfering substance that caused an error less than ± 5% for the WAR detection. 2− The results showed that 500-fold of Ca2 +, Mg 2 +, K+ , NH+ 4 , SO 4 , ; 200-fold of glucose, sucrose, fructose, sorbitol, citric Cl−, and NO− 3 acid, uric acid, starch and gelatin; and 100-fold of histidine, valine, glycine, alanine, L-tyrosine, L-tryptophan, and ascorbic acid did not affect the selectivity (Fig. S2). These results showed that the selectivity of the method is acceptable and it is suitable for the analysis of WAR in real samples with different matrices. 3.10. Real sample analysis To prove the applicability of the proposed method for the analysis of real sample, the blood serum, urine and milk human samples were

analyzed. The WAR content of these samples determined by CdS-QDs/ CS/MWCNTs/GCE is using a standard addition method. The obtained results are summarized in Table 2. As it is obvious, the recovery of WAR was found between 99.0% and 101.7%. This means that the proposed procedure should be applicable to the analysis of real samples with different matrices.

4. Conclusions This work demonstrates that the combination of different nanomaterials with different functions would enlarge the range of application of the chemically modified electrodes in electroanalytical chemistry. The voltammetric response of the modified glassy carbon electrode to WAR was significantly enhanced by the MWCNTs, CS and CdS-QDs. Under optimum conditions, the CdS-QDs/CS/MWCNTs/GCE showed excellent selectivity and sensitivity towards WAR. This electrode presents the advantages of easy fabrication, reasonable stability and reproducibility, low DL and high sensitivity. In comparison to the reported results [11–23] it has wider linear range. Finally, the modified electrode could

Table 1 Analytical parameters for WAR detection by several methods. Methods

Linear range (μM)

LOD (μM)

Conditions

Real sample

References

HF-LPMEa HPLCb LCc CEd Phosphorescence Fluorescence Cloud point extraction Spectrofluorimetric Voltammetry Voltammetry Voltammetry

0.05–1.8 0.05–0.4 0.2–2.0 0.3–80 1–13 5–500 0.003–1.0 0.8–17 0.0001–0.002 0.005–0.4 0.05–80

0.016 0.01 0.08 0.16 0.25 2 0.00033 0.02 0.00008 0.00065 0.0085

Aqueous solution Acetic acid–ammonium acetate (pH = 4.5) PBS (pH = 3.0) PBS (pH = 2.5) – CH3COOH + NH3 (pH = 6.5) Na3PO4 (pH = 7.0) HTAC surfactant (pH = 7) PBS (pH = 6.5) BR buffer (pH = 5) PBS buffer (pH = 2.0)

Human plasma Human plasma Human plasma Human serum Human plasma, water Soil Urine samples Water samples Human serum Human serum, urine Human serum, urine, milk

[11] [13] [15] [17] [18] [19] [20] [21] [22] [23] This work

a b c d

Hollow fiber liquid phase microextraction. High performance liquid chromatography. Liquid chromatography. Capillary electrophoresis.

86

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

Table 2 Determination of warfarin in real samples (n = 5). Sample

Added (nM)

Found (nM)

Recovery (%)

RSD (%)

Urine

– 500 1000 – 500 1000 – 500 1000

306 820 1825 276 770 1795 116 610 1600

– 101.7 101.0 – 99.2 101.0 – 99.0 99.0

2.43 2.93 2.76 2.13 2.94 2.98 2.28 2.37 2.63

Serum

Milk

be applied for the determination of WAR in real samples with satisfactory results. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.07.020. Acknowledgments The authors gratefully acknowledge the support of this work by the Razi University Research for financial support and Bakhtar Biochemistry Company for drug support. References [1] R.A. Harrington, R.C. Becker, M. Ezekowitz, Antithrombotic therapy for coronary artery disease: the seventh ACCP conference on antithrombotic and thrombolytic therapy, Chest 126 (2004) 513–548. [2] D.N. Salem, P.T. O'Gara, C. Madias, S.G. Pauker, Valvular and structural heart disease: American college of chest physicians evidence-based clinical practice guidelines (8th edition), Chest 133 (2008) 593–629. [3] D.E. Singer, G.W. Albers, J.E. Dalen, A.S. Go, J.L. Halperin, W.J. Manning, Antithrombotic therapy in atrial fibrillation: the seventh ACCP conference on antithrombotic and thrombolytic therapy, Chest 126 (2004) 429–456. [4] S. Sun, M. Wang, L. Su, J. Li, H. Li, D. Gu, Study on warfarin plasma concentration and its correlation with international normalized ratio, J. Pharm. Biomed. Anal. 42 (2006) 218–222. [5] Y. Ishihama, Y. Oda, N. Asakaw, Review Report for the Active Substance Warfarin, http://ec.europa.eu/food/plant/protection/evaluation/existactive/warfarin en. pdf2005. [6] R.D. Thornes, D.W. Edlow, S. Wood Jr., Inhibition of locomotion in cancer cells in vivo by anticoagulant therapy. 1. Effect of sodium warfarin on v2 cancer cells, granulocytes, lymphocytes and macrophages in rabbits, Johns Hopkins Med. J. 123 (1968) 305–316. [7] A. Walfisch, G. Koren, The warfarin window in pregnancy: the importance of halflife, J. Obstet. Gynaecol. Can. 10 (2010) 988–989. [8] O.Y. Alshogran, A.J. Ocque, J. Zhao, B.W. Day, F.A. Leblond, V. Pichette, T.D. Nolin, Determination of warfarin alcohols by ultra-high performance liquid chromatography–tandem mass spectrometry: application to in vitro enzyme kinetic studies, J. Chromatogr. B 944 (2014) 63–68. [9] V.K. Boppana, W.H. Schaefer, M.J. Cyronak, High-performance liquid–chromatographic determination of warfarin enantiomers in plasma with automated on-line sample enrichment, J. Biochem. Biophys. Methods 54 (2002) 315–326. [10] A. Osman, K. Arbring, T.L. Lindahl, A new high-performance liquid chromatographic method for determination of warfarin enantiomers, J. Chromatogr. B 826 (2005) 75–80. [11] J. Malakova, P. Pavek, L. Svecova, I. Jokesova, P. Zivny, V. Palicka, New highperformance liquid chromatography method for the determination of (R)-warfarin and (S)-warfarin using chiral separation on a glycopeptide-based stationary phase, J. Chromatogr. B 877 (2009) 3226–3230. [12] M. Hadjmohammadi, H. Ghambari, Three-phase hollow fiber liquid phase microextraction of warfarin from human plasma and its determination by highperformance liquid chromatography, J. Pharm. Biomed. Anal. 61 (2012) 44–49. [13] T. Lomonaco, S. Ghimenti, I. Piga, M. Onor, B. Melai, R. Fuoco, F.D. Francesco, Determination of total and unbound warfarin and warfarin alcohols inhuman plasma by high performance liquid chromatography with fluorescence detection, J. Chromatogr. A 1314 (2013) 54–62. [14] M.-c. Jin, X.-h. Chen, Hen.-p. Chen, Determination of warfarin and coumatetralyl in serum by HPLC coupled with electrospray ionization mass spectrometry, J. Liq. Chromatogr. Relat. Technol. 29 (2006) 2641–2650. [15] D.K. Kumar, D.G. Shewade, S. Parasuraman, S. Rajan, J. Balachander, B.V. Sai Chandran, Ch. Adithan, Estimation of plasma levels of warfarin and 7-hydroxy warfarin by high performance liquid chromatography in patients receiving warfarin therapy, J. Young Pharm. 5 (2013) 13–17. [16] I. Locatelli, V. Kmetec, A. Mrhar, I. Grabnar, Determination of warfarin enantiomers and hydroxylated metabolites in human blood plasma by liquid chromatography with achiral and chiral separation, J. Chromatogr. B 818 (2005) 191–198.

[17] W. Naidong, P.R. Ring, C. Midtlien, X. Jiang, Development and validation of a sensitive and robust LC–tandem MS method for the analysis of warfarin enantiomers in human plasma, J. Pharm. Biomed. Anal. 25 (2001) 219–226. [18] W.-P. Yau, E. Chan, Chiral CE separation of warfarin in albumin containing samples, J. Pharm. Biomed. Anal. 28 (2002) 107–123. [19] L.F. Capitàn-Vallvey, M.K.A. Deheidel, R. Avidad, Determination of warfarin in waters and human plasma by solid-phase room-temperature transmitted phosphorescence, Arch. Environ. Contam. Toxicol. 37 (1999) 1–6. [20] T.D. Smirnova, N.V. Nevryueva, S.N. Shtykov, V.I. Kochubei, D.A. Zhemerichkin, Determination of warfarin by sensitized fluorescence using organized media, J. Electroanal. Chem. 64 (2009) 1142–1147. [21] Zh. Chang, H.-T. Yan, Cloud point extraction–fluorimetric combined methodology for the determination of trace warfarin based on the sensitization effect of supramolecule, J. Lumin. 132 (2012) 811–817. [22] G. Armas, M. Miró, J.M. Estela, V. Cerdà, Multisyringe flow injection spectrofluorimetric determination of warfarin at trace levels with on-line solid-phase preconcentration, Anal. Chim. Acta 467 (2002) 13–23. [23] B. Rezaei, O. Rahmanian, A.A. Ensafi, An electrochemical sensor based on multiwall carbon nanotubes and molecular imprinting strategy for warfarin recognition and determination, Sens. Actuators, B 196 (2014) 539–545. [24] M.M. Ghoneim, A. Tawfik, Assay of anti-coagulant drug warfarin sodium in pharmaceutical formulation and human biological fluids by square-wave adsorptive cathodic stripping voltammetry, Anal. Chim. Acta 511 (2004) 63–69. [25] V.K. Gupta, R.N. Goyal, R.A. Sharma, Novel PVC membrane based alizarin sensor and its application; determination of vanadium, zirconium and molybdenum, Int. J. Electrochem. Sci. 4 (2009) 156–172. [26] V.K. Gupta, A.K. Jain, L.P. Singh, Upendra Khurana, Porphyrins as carrier in PVC based membrane potentiometric sensors for nickel(II), Anal. Chim. Acta 355 (1997) 33–41. [27] V.K. Gupta, M. Al Khayat, A.K. Singh, M.K. Pal, Nano level detection of Cd(II) using poly(vinyl chloride) based membranes of Schiff bases, Anal. Chim. Acta 634 (2009) 36–43. [28] V.K. Gupta, S. Chandra, Heinrich Lang, A highly selective mercury electrode based on a diamine donor ligand, Talanta 66 (2005) 575–580. [29] V.K. Gupta, R. Jain, K. Radhapyari, N. Jadon, Sh. Agarwal, Voltammetric techniques for the assay of pharmaceuticals—a review, Anal. Biochem. 408 (2011) 179–196. [30] V.K. Gupta, A.K. Jain, P. Kumar, PVC-based membranes of N, N′-dibenzyl-1,4,10,13tetraoxa-7,16-diazacyclooctadecane as Pb(II)-selective sensor, Sens. Actuators, B 120 (2006) 259–265. [31] R.N. Goyal, V.K. Gupta, S. Chatterjee, A sensitive voltammetric sensor for determination of synthetic corticosteroid triamcinolone, abused for doping, Biosens. Bioelectron. 24 (2009) 3562–3568. [32] V.K. Gupta, S. Chandra, R. Mangla, Dicyclohexano-18-crown-6 as active material in PVC matrix membrane for the fabrication of cadmium selective potentiometric sensor, Electrochim. Acta 47 (2002) 1579–1586. [33] V.K. Gupta, R. Prasad, A. Kumar, Preparation of ethambutol–copper(II) complex and fabrication of PVC based membrane potentiometric sensor for copper, Talanta 60 (2003) 149–160. [34] V.K. Gupta, A.K. Jain, G. Maheshwari, Heinrich Lang, Z. Ishtaiwi, Copper(II)-selective potentiometric sensors based on porphyrins in PVC matrix, Sens. Actuators, B 117 (2006) 99–106. [35] R.N. Goyal, V.K. Gupta, N. Bachheti, Fullerene-C60-modified electrode as a sensitive voltammetric sensor for detection of nandrolone—an anabolic steroid used in doping, Anal. Chim. Acta 597 (2007) 82–89. [36] R.N. Goyal, V.K. Gupta, S. Chatterjee, Fullerene-C60-modified edge plane pyrolytic graphite electrode for the determination of dexamethasone in pharmaceutical formulations and human biological fluids, Biosens. Bioelectron. 24 (2009) 1649–1654. [37] R.N. Goyal, V.K. Gupta, S. Chatterjee, Electrochemical oxidation of 2′,3′dideoxyadenosine at pyrolytic graphite electrode, Electrochim. Acta 53 (2008) 5354–5360. [38] V.K. Gupta, R. Prasad, P. Kumar, R. Mangla, New nickel(II) selective potentiometric sensor based on 5,7,12,14-tetramethyldibenzotetraazaannulene in a poly(vinyl chloride) matrix, Anal. Chim. Acta 420 (2000) 19–27. [39] R. Jain, V.K. Gupta, N. Jadon, K. Radhapyari, Voltammetric determination of cefixime in pharmaceuticals and biological fluids, Anal. Biochem. 407 (2010) 79–88. [40] V.K. Gupta, A.K. Jain, P. Kumar, S. Agarwal, G. Maheshwari, Chromium(III)-selective sensor based on tri-o-thymotide in PVC matrix, Sens. Actuators, B 113 (2006) 182–186. [41] R.N. Goyal, V.K. Gupta, S. Chatterjee, Simultaneous determination of adenosine and inosine using single-wall carbon nanotubes modified pyrolytic graphite electrode, Talanta 76 (2008) 662–668. [42] V.K. Gupta, A.K. Singh, S. Mehtab, B. Gupta, A cobalt(II)-selective PVC membrane based on a Schiff base complex of N, N′-bis(salicylidene)-3,4-diaminotoluene, Anal. Chim. Acta 566 (2006) 5–10. [43] V.K. Gupta, A.K. Singh, M. Al Khayat, Barkha Gupta, Neutral carriers based polymeric membrane electrodes for selective determination of mercury (II), Anal. Chim. Acta 590 (2007) 81–90. [44] R.N. Goyal, D. Kaur, S.P. Singh, A.K. Pandey, Effect of graphite and metallic impurities of C60 fullerene on determination of salbutamol in biological fluids, Talanta 75 (2008) 63–69. [45] C. Karuwan, A. Wisitsoraat, T. Maturos, D. Phokharatkul, A. Sappat, K. Jaruwongrungsee, T. Lomas, A. Tuantranont, Flow injection based microfluidic device with carbon nanotube electrode for rapid salbutamol detection, Talanta 79 (2009) 995–1000. [46] L. Lijun, Y. Laibo, C. Hao, C. Qifeng, W. Fengmin, C. Tian, Z. Xiaoyong, K. Hongxing, W. Jianling, The determination of salbutamol sulfat based on a flow‐injection coupling

M.B. Gholivand, L. Mohammadi-Behzad / Materials Science and Engineering C 57 (2015) 77–87

[47]

[48]

[49] [50]

[51]

[52] [53]

[54]

[55]

irreversible biamperometry at poly(aminosulfonic acid)‐modified glassy carbon electrode, Anal. Lett. 40 (2007) 3290–3308. S.J. Ling, R. Yuan, Y.Q. Chai, T.T. Zhang, Study on immunosensor based on gold nanoparticles/chitosan and MnO2 nanoparticles composites membrane/Prussian blue modified gold electrode, Bioprocess Biosyst. Eng. 32 (2009) 407–414. B. Batra, C.S. Pundir, An amperometric glutamate biosensor based on immobilization of glutamate oxidase onto carboxylated multiwalled carbon nanotubes/gold nanoparticles/chitosan composite film modified Au electrode, Biosens. Bioelectron. 47 (2013) 496–501. J. Drbohlavova, V. Adam, R. Kizek, J. Hubalek, Quantum dots—characterization, preparation and usage in biological systems, Int. J. Mol. Sci. 10 (2009) 656–673. Y. Lai, Y. Yu, P. Zhong, J. Wu, Z. Long, C. Liang, Development of novel quantum dots as fluorescent sensors for application in highly sensitive spectrofluorimetric determination of Cu2+, Anal. Lett. 39 (2006) 1201–1209. M.J. Giz, B. Duong, N.J. Tao, In situ STM study of self-assembled mercaptopropionic acid monolayers for electrochemical detection of dopamine, J. Electroanal. Chem. 465 (1999) 72–79. M. Koneswaran, R. Narayanaswamy, L-Cysteine-capped ZnS quantum dots based fluorescence sensor for Cu2+ ion, Sensors Actuators B 139 (2009) 104–109. B.B. Prasad, A. Prasad, M.P. Tiwari, Quantum dots-multi-walled carbon nanotubes nanoconjugate-modified pencil graphite electrode for ultratrace analysis of hemoglobin in dilute human blood samples, Talanta 109 (2013) 52–60. G. Jie, P. Liu, L. Wang, Sh. Zhang, Electrochemiluminescence immunosensor based on nanocomposite film of CdS quantum dots–carbon nanotubes combined with gold nanoparticles–chitosan, Electrochem. Commun. 12 (2010) 22–26. D. Du, W. Chen, W. Zhang, D. Liu, H. Li, Y. Lin, Covalent coupling of organophosphorus hydrolase loaded quantum dots to carbon nanotube/Au nanocomposite for enhanced detection of methyl parathion, Biosens. Bioelectron. 25 (2010) 1370–1375.

87

[56] N. Zhanga, J. Sunb, D. Jiangb, T. Fengb, Q. Lia, Anchoring zinc oxide quantum dots on functionalized multi-walled carbon nanotubes by covalent coupling, Carbon 47 (2009) 1214–1219. [57] J. Chen, A. Zheng, Y. Gao, C. He, G. Wu, Y. Chen, X. Kai, C. Zhu, Functionalized CdS quantum dots-based luminescence probe for detection of heavy and transition metal ions in aqueous solution, Spectrochim. Acta A 69 (2008) 1044–1052. [58] P. Mahanandia, P. Vishwakarma, K. Nanda, V. Prasad, S. Subramanyam, S. Dev, Multiwall carbon nanotubes from pyrolysis of tetrahydrofuran, Mater. Res. Bull. 41 (2006) 2311–2317. [59] A. Webster, P.O. Osifo, H.W.J.P. Neomagus, D.M. Grant, A comparison of glycans and polyglycans using solid-state NMR and X-ray powder diffraction, Solid State Nucl. Magn. 30 (2006) 150–161. [60] Y. Kang, D. Kim, Enhanced optical sensing by carbon nanotube functionalized with CdS particles, Sensors Actuators B 125 (2006) 114–117. [61] E. Laviron, J. Electroanal. Chem. 101 (1979) 19–28. [62] A.A.J. Torriero, E. Tonn, L. Sereno, J. Raba, Electrooxidation mechanism of nonsteroidal anti-inflammatory drug piroxicam at glassy carbon electrode, J. Electroanal. Chem. 588 (2006) 218–225; The United States Pharmacopoeia, The National Formularly, USP 24, NF 19, USP443 Convention Inc., 12601, MD, 2000. 2151. [63] P. Schmuki, Pits and pores II: formation, properties, and significance for advanced materials, Proceedings of the International Symposium, The Electrochemical Society, USA 2001, pp. 352–463.