Application of magnetic lamotrigine-imprinted polymer nanoparticles as an electrochemical sensor for trace determination of lamotrigine in biological samples

Abstract

In the present work, a differently shaped electrode based on carbon paste was designed using a magnet at the back of a carbon paste (made a magnet containing a carbon paste electrode) for determination of lamotrigine, using magnetic molecularly imprinted polymer (Magnetic-MIP) nanoparticles. Magnetic-MIPs have been synthesized for the preconcentration and selective trace determination of lamotrigine (LTG) in urine and plasma samples. The obtained sorbent was characterized using X-ray diffraction (XRD), scanning electron microscopy (SEM), infrared absorption spectroscopy (IR) and thermogravimetric analysis (TG). Separation of the sorbent from the sample solution was simply achieved by an electrode via a magnetic field from the magnet containing electrode. Determination of the extracted drug was performed by the differential pulse voltammetry (DPV) technique. Under the optimum extraction conditions, the method detection limits were 4.7 pM (based on a S/N of 3) for urine samples and 5.9 pM for plasma samples. Two linear dynamic ranges from 0.01–1.0 nM and 1.0–200 nM were obtained for LTG in biological samples.

---

1. Introduction

One of the drugs which is used as an anticonvulsant and for treating epilepsy is lamotrigine (3,5-diamino-6-(2,3-dichlorophenyl)-1,2,4-triazine). Lamotrigine (LTG) is a third generation anticonvulsant drug used in the treatment of epilepsy.¹ It exhibits excellent oral bioavailability with first-order linear pharmacokinetics and has a mean plasma half-life of approximately 24 h. Lamotrigine is a lipophilic weak base with plasma protein binding of ∼55%.² It gets extensively metabolized in the human liver via hepatic glucuronidation by uridine 5′-diphosphate-glucuronosyl transferase (UGT1A4).³ At levels of 10–15 μg mL⁻¹ (39–59 mmol L⁻¹), 24% and above 20 μg mL⁻¹ (78 mmol L⁻¹), 59% of patients showed signs of toxicity while others tolerated levels above 20 μg mL⁻¹.⁴ It is increasingly used in the management of partial and generalized epilepsies. The most famous side effect of LTG is a syndrome called Stevens–Johnson. Skin rashes, which are characterized by painful blistering of the skin and mucous membranes, are the other side effects of lamotrigine.⁵ Lamotrigine and its metabolites in pharmaceutical products and biological fluids typically have been monitored by gas chromatography with a nitrogen phosphorus detector,⁶ radioimmuno assay,⁷ high performance liquid chromatography (HPLC),^8–16 liquid chromatography tandem mass spectrometry,¹⁷ capillary electrophoresis,^18,19 immunofluorometric assay,²⁰ molecular imprinting technology,^21,22 and voltammetric determination.^23–26 Owing to the dangerous side effect of LTG, the pharmaceutical quantity control of LTG is vital. So, development of a sensitive and versatile analytical method is needed for its determination. Due to high efficiency, accuracy, sensitivity, simplicity, and low cost, use of electrochemical techniques in pharmaceutical analysis has attracted more attention. Nowadays, the ability of molecularly imprinted polymers (MIPs) to selectively recognize and bind the template structure, in the presence of closely related chemical species, has made them of interest for use in the quantitative discrimination of target molecules prevalent in real world samples.^27–30 Participation of a magnetic component in the molecularly imprinted polymer (MIP) can build a controllable rebinding process and allow magnetic separation in a convenient way.³¹ The magnetic MIPs can make the separation process simple and fast; because, it can be easily collected by an external magnetic field without additional centrifugation or filtration. Studies about the preparation of magnetic molecularly imprinted polymers (MMIPs) have been reported by many researchers.^32–36 The advantages of magnetic nanoparticles in the electrochemical sensors were previously presented by Gooding et al.^37,38 However, the MIPs prepared by traditional methods suffer from some disadvantages in certain applications, such as the heterogeneous distribution of the binding sites, poor site accessibility to the target molecules and low-efficiency of mass transfer.^39,40 An ideal solution for these problems is the development of a surface molecular imprinting technique. The MIPs prepared by this method have been confirmed to possess much more effective recognition sites and be much faster for mass transfer.^41,42 To achieve surface imprinting, the simplest method is imprinting molecules at the surface of some solid to obtain the core–shell structural MIPs.^43,44 Of various solid support materials, the Fe₃O₄@SiO₂ nanoparticle is an excellent candidate as the supporting material for preparation of the core–shell structural MIPs because of its reliable chemical stability, compatibility, reactivity with various coupling agents and the inherent properties of the magnetic element.^45,46 Recently, a molecularly imprinted polymer on the surface of a core–shell Fe₃O₄@SiO₂ nanoparticle (Fe₃O₄@SiO₂-MIP) modified electrode has been used for electrochemical sensor applications.⁴⁷ However, to the best of our knowledge, the determination of LTG using a Fe₃O₄@SiO₂-MIP modified magnet containing a carbon paste electrode has not been reported.

In this study the LTG-imprinted magnetic nanoparticles were synthesized and then utilized for selective extraction of LTG molecules from a given sample. The magnetic MIP was then separated from the solution and accumulated onto a carbon paste electrode via a magnetic field, provided via a permanent magnet, situated behind the CP electrode surface. The amount of LTG in biological samples was measured by a differential pulse voltammetry (DPV) technique. The procedure is based on the reduction of LTG after its selective extraction in the carbon paste electrode. The measurements with Magnetic-MIP were carried out based on a four-step methodology, namely analyte extraction, separation of the sorbent by a magnetic field onto the electrode, electrode washing, and electrochemical measurement.

2. Material and methods

2.1. Reagents

Methacrylic acid with high purity was purchased from Merck (Darmstadt, Germany). Ethylene glycol dimethacrylate (EGDMA) was obtained from Fluka (Buchs SG, Switzerland). 2,2′-Azobisisobutyronitrile (AIBN) was obtained from Acros Organics (New Jersey, USA). NaOH, HCl, HNO₃, acetic acid (HOAC), sodium acetate, FeCl₃·6H₂O, FeCl₂·4H₂O, tetraorthosilicate, methacryloxy propyltrimethoxysilane, toluene, acetonitrile, ethanol and methanol were purchased from Merck (Darmstadt, Germany). The acetate buffer solutions (0.2 mol L⁻¹) were prepared with CH₃COONa and CH₃COOH. All the other reagents used were of analytical grade and purchased from Merck (Darmstadt, Germany, http://www.merck.de). Ultrapure water was prepared using a Milli-Q system from Millipore (Bedford, MA, USA). Lamotrigine (LTG) tablets were from the Sobhan Daru pharmaceutical company (Tehran, Iran). A stock solution of LTG (0.01 M) was prepared in methanol, stored in a refrigerator between 4 and 8 °C, and protected from light until use. Working solutions of lamotrigine (ranging from 0.01 nM to 200 μM) were prepared by serial dilution of the stock solution.

2.2. Apparatus

All electrochemical measurements were performed with a Palm-Sens (EN 50081-2) potentiostat. A personal computer was used for data storage and processing. The three electrode electrochemical cell used was equipped with a carbon paste electrode with a magnet as a working electrode, a saturated calomel electrode (SCE) as a reference electrode and a platinum electrode as an auxiliary electrode (Azar Electrode Co, Iran). All potentials in the text were reported versus this reference electrode. The pH measurement was performed with a Metrohm model 691 pH/mV meter. The Heidolph heater stirrer model MR 3001 (Germany) was employed for heating and stirring of the solutions. Fourier transform infrared (FT-IR) spectra (4000–200 cm⁻¹) in KBr were recorded using a Bruker IFS66/S FT-IR spectrometer. High angle X-ray diffraction patterns were obtained on a Philips-PW 17C diffractometer with Cu K_α radiation. Scanning electron microscopy (SEM) was performed by gently distributing the powder sample on the stainless steel stubs, using a SEM (KYKY, EM3200) instrument. The thermal properties of synthesized polymers were determined using a BAHR-Thermoanalyse GmbH (Germany) with employing, heating and cooling rates of 10 °C min⁻¹ and using a condenser as the coolant.

2.3. Preparation of Magnetic-MIP and Magnetic-NIP

2.3.1. Synthesis of Fe₃O₄@SiO₂. The description about the synthesis of Fe₃O₄@SiO₂ is presented in the ESI section.†

2.3.2. Synthesis of Fe₃O₄@SiO₂@MPTS. The description about the synthesis of Fe₃O₄@SiO₂@MPTS is presented in the ESI section.†

2.3.3. Preparation of the magnetic molecular imprinted polymer. The magnetic molecularly imprinted polymer (MMIP) was prepared by a surface molecular imprinting technique. Briefly, LTG (0.1 mmol) as the template and methacrylic acid (0.5 mmol) as the functional monomer were dissolved in 50 mL methanol : acetonitrile (50 : 50 v/v). Then, 0.5 g of Fe₃O₄@SiO₂@MPTS nanoparticles was added to the mixture, and stirred for 2 h. Subsequently, ethylene glycol dimethacrylate (3.0 mmol) and 2,2′-azobisisobutyronitrile (75 mg) were added into the system and the mixture was degassed in an ultrasonic bath for 15 min. Then, reaction vassal was flashed with nitrogen gas for 10 min to remove oxygen. The polymerization was performed at 60 °C with nitrogen protection for 24 h. The MMIPs were collected by a magnet, and washed by a mixture of methanol/TFA/HOAC (96 : 2 : 2, (v/v/v%)) to remove the templates and then washed by methanol. Finally, the modified magnetic nanoparticles were dried in the vacuum. The magnetic non-imprinted polymer (MNIP) was prepared by the same method as MMIPs without the addition of the template.

2.4. Preparation of the working electrode

The magnet containing a carbon paste electrode (MC-CPE) was prepared by mixing 268 mg of the graphite powder with 132 mg of paraffin oil. Then, obtained materials were packed into the end of a 5.0 cm high polyethylene tube (i.d. 10 mm), with a copper wire as the inner electrical contact and a magnet with a power of 4.0 tesla (o.d. 10 mm, 2 mm height) was embedded at a depth of 4 mm from the electrode surface. The copper wire would pass through the hole of the magnet, and then it was coated with the carbon paste. Appropriate packing was achieved by pushing the electrode surface against a sheet of paper. Smoothing of the electrode surface was made by hand-polishing on a paper. After pretreatment and when the extraction was done, the magnetic molecular imprinted nanoparticles were firmly magnetically attached to the MCPE surface by magnetic forces. The structure of the MC-CPE is shown in Fig. 1.

Fig. 1 Magnet contain carbon paste electrode procedure based on magnetic molecularly imprinted polymer for determination of lamotrigine.

2.5. Electrochemical measurements

Electrochemical measurements of the prepared electrode were performed by using DPV techniques. For recording the DP voltammograms, the potential was scanned from −1.1 to −0.6 V, using a sweep rate of 80 mV s⁻¹ for DPV respectively. For the analysis of LTG a four step procedure was carried out. At first, 10.0 mg of the Magnetic-MIP was added to 20 mL of the sample (pH 5.5) and it was shaken for 5 minutes (Fig. 1a). At the second step, the Magnetic-MIPs were firmly magnetically attached to the electrode surface by magnetic forces (Fig. 1b) and then the electrode was rinsed. This stage was completed within 2 min. In the third step, the electrode was then removed from the first solution and rinsed by stirring it for 30 seconds in distilled water to remove the residual sample (including the unbound LTG). During the rinsing, no nanoparticles were observed leaving the surface, owing to the strong applied magnetic field. At the final step the electrode was entered in the voltammetric cell and an electrochemical procedure was carried out and a signal was detected (acetate buffer (0.025 M, pH = 5.5)).

2.6. Real sample preparation

Tablets of LTG (nominal 25 mg per tablet) were purchased from Sobhan Daru pharmaceutical company (Tehran, Iran). Five tablets of LTG were weighed accurately and finely powdered and mixed. A quantity of the powder equivalent to the average weight of one tablet was transferred into a 100.0 mL calibrated flask and 50.0 mL of methanol was added. The content of the flask was sonicated for 15.0 min and then filled up to volume with methanol. The solution was centrifuged for 10.0 min at 5000.0 rpm. Appropriate volumes of the clear supernatant were transferred into the volumetric cell and the clear solution was diluted to 20 mL and pH was adjusted to 5.5 using NaOH or HNO₃ for further analysis.

The solution was analyzed according to the procedure for voltammetry analysis and the quantity of LTG per tablet was determined using the calibration curve method. Plasma samples stored at around −20 °C were thawed on the day of extraction at room temperature followed by vortexing to ensure homogeneity. A plasma sample (6.0 mL) was spiked with LTG at the desired concentration levels and then was acidified with 100.0 μL of hydrochloric acid (37.0%) to disturb the LTG protein binding. Then, 150.0 μL of trichloroacetic acid (TCA) was added to denature the proteins. These processes eventually led to precipitation of proteins. Subsequently, the sample was centrifuged at 5000.0 rpm for 10.0 min. An amount of 4.0 mL of the supernatant was transferred to the voltammetric cell and diluted 5.0 times (pH 5.5). Then the solution was analyzed according to the procedure for voltammetry analysis. Urine samples (2.0 mL) stored at around 4 °C were thawed on the day of extraction at room temperature and diluted 10 times (pH 5.5).

3. Results and discussion

3.1. Characterization of the magnetic molecular imprinted polymer

The information about characterization of the synthesized Magnetic-MIP is presented in the ESI.† The morphology of the synthesized materials was assessed by SEM and micrographs are shown in Fig. 2. As shown, the particle sizes of Fe₃O₄, Fe₃O₄@SiO₂@MPTS, and Fe₃O₄@SiO₂@MPTS@MIP are about 25 (±2), 35 (±5), and 73 (±10) nm, respectively.

Fig. 2 SEM images of (A) Fe₃O₄, (B) Fe₃O₄@SiO₂@MPTS, and (C) Fe₃O₄@SiO₂@MPTS@MIP.

3.2. Optimization of experimental conditions

3.2.1. pH effect. pH of the extraction solution was checked and its effect on the LTG extraction by the Magnetic-MIP was studied. For this purpose, the prepared Magnetic-MIP was added into the solutions, containing 1.0 nM of LTG, with various pH values where they were incubated for 5 min at a constant stirring rate. After that, the working electrode was exposed to the Magnetic-MIP which usurped the target drug. The working electrode was immersed in the sample cell for 2 min and after adsorbing the Magnetic-MIP by the electrode via magnetic force, it entered the washing solution for 30 seconds. At the final step the electrochemical procedure was carried out in acetate buffer (0.025 M, pH = 5.5) and a signal was detected (Fig. 4S (ESI†)). It was found that the optimum pH value was 5.5 (Fig. 4S (ESI†)). Below the optimum pH value (pH of 5.5) which is close to its pK_a of LTG (5.7), the target drug is in the protonated form. As you know the extraction of LTG by the synthesized MMIP is according to the hydrogen bonding between active sites of the MIP and target molecules. Therefore, the mentioned interaction can be received in the neutral form of the drug (pH of 5.5). The obtained data show that the maximum interaction of the drug with active sites of the MMIP was achieved at the pH value of 5.5.

3.2.2. The effect of shaking time. The effect of shaking time on I_p in DPV for 1.0 nM LTG was investigated by addition of 10.0 mg of the Magnetic-MIP into (shaken) 20 mL of the sample solution. As shown in the obtained data (Fig. 5S†), the peak current increased for up to 5 min of extraction; however, with longer periods of preconcentration time there was no significant change in the peak current. Hence, for all subsequent measurements, a preconcentration time (shaking time) of 5 min was employed.

3.2.3. Optimization of accumulation time to adsorption of the Magnetic-MIP to the electrode, via magnetic force. After extraction of LTG by the Magnetic-MIP, the MC-CPE was immersed into the sample solution for different periods of time, because the Magnetic-MIPs were adsorbed via magnetic force on the surface of the carbon paste electrode. After that the MC-CPE was inserted into a voltammetric cell and the DPV curves were recorded. The peak current increased up to 2 min and after that no significant change was accrued in the peak current. Therefore in subsequent measurements, 2 min was employed as an accumulation time.

3.3. Comparison of the Magnetic-MIP with the Magnetic-NIP electrode and the evaluation of the selectivity of the Magnetic-MIP

The MC-CPE was incubated in a LTG solution containing the Magnetic-MIP. Then, the electrode was inserted in the electrochemical cell. Subsequently, the DPV technique was applied for the determination of LTG. The same experiment was carried out in the case of the Magnetic-NIP. In another experiment, the Magnetic-MIP and Magnetic-NIP electrodes were inserted in a washing solution for 30 seconds, after being removed from the LTG solution. The obtained voltammetric signals of the two mentioned electrodes are shown in Fig. 6S (ESI†). It can be seen that the signal obtained for the Magnetic-MIP electrode is noticeably higher than that of the Magnetic-NIP electrode, indicating the existence and proper functioning of the selective cavities in the MIP, created in the polymerization step. Washing of the electrodes, after removing them from the analyte solution, decreased both electrode signals indicating the removal of weakly adsorbed molecules on the Magnetic-MIP or Magnetic-NIP surface sites. However, the decrease of current in the Magnetic-NIP signal is higher than that in the Magnetic-MIP electrode. This can be related to the fact that most of the adsorbed LTG molecules are located on the selective sites of the Magnetic-MIP that are attracted strongly and specifically to the Magnetic-MIP, while the sites responsible for LTG adsorption in the Magnetic-NIP are surface adsorption sites of a weak and non-selective nature. The effect of washing time on the response of the electrode was also studied. It was found that the difference between the response of MIP-CPE and NIP-CPE increases till 30 seconds and afterwards the aimed signal reached a partly steady state. Thus 30 seconds was chosen as optimal washing time.

The selectivity of the proposed method for analysis of LTG was evaluated by analysis of LTG and/or some LTG similar compounds such as 2,4,6-triamino-pyrimidine, 2,6-diamine pyridine and 3-amino-1,2,4-triazine. As can be seen in Table 1S (ESI†), the proposed Magnetic-MIP is selective for LTG whereas the responses of all other tested compounds except LTG were small. The response of the CP electrode for analysis of LTG compared with the response for analysis of LTG by the Magnetic-MIP procedure is presented in Table 1. These results proved the selectivity of the Magnetic-MIP procedure for analysis of LTG.

Table 1 Calibration equation of real samples spiked in buffered solution of pH 5.5

Sample	Regression equation	R ^2	DLR^a (nM)	MDL^b (pM)	RSD^c (%)
a DLR: dynamic linear range. b LOD: limit of detection. c RSD: relative standard deviation (for 1 nM of LTG).
Water	I (μA) = −1.3441C (nM) − 0.2997	0.9963	0.01–1.00	3.0	5.1
	I (μA) = −0.1356C (nM) − 0.2145	0.9947	1.00–200	—	—
Plasma	I (μA) = −1.2941C (nM) − 0.1725	0.9812	0.01–1.00	5.9	6.7
	I (μA) = −0.1228C (nM) − 0.1487	0.9882	1.00–200	—	—
Urine	I (μA) = −1.3165C (nM) − 0.2823	0.9836	0.01–1.00	4.7	5.9
	I (μA) = −0.1298C (nM) − 0.1954	0.9847	1.00–200	—	—

---

3.4. Analytical characterizations

A calibration graph was constructed under the optimum conditions; the DPV reduction peak currents of various concentrations of LTG at the MC-CPE were recorded. As shown in Fig. 3, the peak current was proportional to the concentration of LTG in the range from 0.01–1 nM and 1–200 nM.

Fig. 3 DPV of the sensor after incubation in different concentrations of LTG. Linear calibration graphs of the Magnetic-MIP on the surface of the MC-CPE at higher and lower LTG (inset) concentrations under optimized experimental conditions.

The calibration equations for detection of LTG with the proposed sensor were as follows:

I_p (μA) = −1.34(±0.04)C (nM) − 0.30(±0.02), R² = 0.9963 (for 0.01–1 nM)

I_p (μA) = −0.14(±0.05)C (nM) − 0.21(±0.04), R² = 0.9947 (for 1–200 nM).

The method detection limit (MDL), defined as MDL = 3 S/N, was found to be 3.0 pM. The repeatability of the electrode in the determination of LTG was evaluated by performing seven determinations with the same standard solutions of LTG using the same electrode. The relative standard deviation (RSD) for the response of the electrode toward a 1 nM LTG solution was 5.1%.

The reproducibility of the response of the electrode was also studied. Five electrodes were prepared from different batches and were evaluated by performing the determination of a 1 nM LTG solution. The results show that the reproducibility of the sensor for the determination of LTG is acceptable (5.7%).

The extraction efficiency of the MMIP after 24 weeks for extraction and electrochemical detection of LTG by the proposed magnetic sorbent was checked and the obtained data showed that the prepared sorbet is stable over time. The obtained recovery by the proposed method for analysis of LTG showed that the sorbent is stable for analysis of the drug after 24 weeks (Table 2S†). The regression equation, LODs and RSD for the responses of the electrode in water, plasma and urine samples are summarized in Table 1.

3.5. Real sample analysis

In order to evaluate the accuracy of the proposed method and to investigate matrix effects, spiked human urine, plasma samples, and tablets were tested. The recoveries of LTG from the real and spiked samples varied in the range of 97.0–105%. The results clearly indicate the suitability of the new method for determination of LTGs at trace levels in biological samples (Table 2).

Table 2 Recovery results of real samples spiked in buffered solution of pH 5.5

Real sample	Added	Found	Recovery (%)	RSD (%)
Urine	—	N. D.	—	—
	1.00 nM	0.97 nM	97.0	5.9
	100.00 nM	98.70 nM	98.7	6.1
Plasma	—	N. D.	—	—
	1.00 nM	1.05 nM	105.0	6.7
	100.00 nM	99.65 nM	99.6	6.2

---

Tablet sample	Labelled amount (mg)	Found (mg)	Recovery (%)	RSD (%)
Sobhan Daru pharmaceutical company (25 mg)	25.00	25.12	100.50	5.2

---

3.6. Comparison of figures of merit of the proposed method for analysis of LTG with previously reported LTG voltammetric sensors

Table 3 compares the type of electrode, technique used, linear dynamic range (LDR), limit of detection (LOD), and relative standard deviation of the proposed sensor for LTG determination with other voltammetric sensors previously reported in the literature. As is obvious from these results, the LOD of the proposed MC-CPE is superior to most of the other reported sensors and the LDR of the proposed sensor is appropriate.^23–26

Table 3 Comparison of the reported sensor for LTG determination with previously published articles for determination of LTG

Electrode	Technique	Dynamic range/nM	LOD/nM	RSD (%)	Reference
a PGE: pyrolytic graphite electrode. b CASV: cyclic adsorptive striping voltammetry. c HMDE: hanging mercury drop electrode. d DPAdSV: differential pulse adsorptive stripping voltammetry. e MCSPE: modified carbon screen-printed electrode.
PGE^a	CASV^b	1000–100 000	80	5.53	23
HMDE^c	DPAdSV^d	4–120	4.68	5.21	25
MCSPE^e	DPAdSV	330–1500	372	2.58	26
MIP-CP	DPV	0.8–25 & 25–400	0.21	4.2	24
MC-CPE	DPV	0.01–1 & 1–200	0.003	5.3	This work

---

4. Conclusions

This paper has demonstrated that the MMIP can be applied for trace level determinations of LTG. The recommended method possesses a low detection limit and good linear range. It was proven that the MMIP was capable of extracting LTG from solution much more than the MNIP, indicating the presence of highly efficient LTG selective cavities on the magnetic particle surface. Due to high selectivity for LTG shown by this Magnetic-IIP, the method has been successfully employed for the determination of LTG in biological samples. Although the method has a multiple step procedure, with relatively long incubation and washing times, the proposed sensor has several advantages such as low cost and easy preparation and has fine characteristics such as selectivity, sensitivity and reproducibility.

References

1. R. G. Morris, A. B. Black, A. L. Harris, A. B. Batty and B. C. Sallustio, Br. J. Clin. Pharmacol., 1998, 46, 547–551.
2. S. Ghatol, V. Vithlani, S. Gurule, A. Khuroo, T. Monif and P. Partani, J. Pharm. Anal., 2013, 3(2), 75–83.
3. A. F. Cohen, G. S. Land, D. D. Breimer, W. C. Yuen, C. Winton and A. W. Peck, Clin. Pharmacol. Ther., 1987, 42, 535–541.
4. M. Sondergaard Khinchi, K. A. Nielsen, M. Dahl and P. Wolf, Seizure, 2008, 17, 391–395.
5. R. H. Chew, R. E. Hales and S. C. Yudofsky, What Your Patient Needs to Know about Psychiatric Medications, American Psychiatric Pub. Inc, 2nd edn, 2009, pp. 201–205.
6. M. Watelle, P. Demedts, F. Franck, P. P. DeDeyn, A. Wauters and H. Neels, Ther. Drug Monit., 1997, 19, 460–464.
7. R. A. Biddlecombe, K. L. Dean, C. D. Smith and S. C. Jeal, J. Pharm. Biomed. Anal., 1990, 8, 691–694.
8. N. R. Barbosa and A. F. Mídio, J. Chromatogr. B: Biomed. Sci. Appl., 2000, 741, 289–293.
9. M. M. Castel-Branco, A. M. Almeida, A. C. Falcão, T. A. Macedo, M. M. Caramona and F. G. Lopez, J. Chromatogr. B: Biomed. Sci. Appl., 2001, 755, 119–127.
10. S. Bompadre, A. Tagliabracci, M. Battino and R. Giorgetti, J. Chromatogr. B: Biomed. Sci. Appl., 2008, 863, 177–180.
11. C. L. Cheng, C. H. Chou and O. Y. Pu Hu, J. Chromatogr. B: Biomed. Sci. Appl., 2005, 817, 199–206.
12. T. A. C. Vermeij and P. M. Edelbroek, J. Chromatogr. B: Biomed. Sci. Appl., 2007, 857, 40–46.
13. J. Emami, N. Ghassami and F. Ahmadi, J. Pharm. Biomed. Anal., 2006, 40, 999–1005.
14. M. Contin, M. Balboni, E. Callegati, C. Candela, F. Albani, R. Riva and A. Baruzzi, J. Chromatogr. B: Biomed. Sci. Appl., 2005, 828, 113–117.
15. K. M. Matar, P. J. Nicholls, S. A. Bawazir, M. I. Al-Hassan and A. Tekle, J. Pharm. Biomed. Anal., 1998, 17, 525–531.
16. M. Torra, M. Rodamilans, S. Arroyo and J. Corbella, Ther. Drug Monit., 2000, 22, 621–625.
17. S. Ghatol, V. Vithlani, S. Gurule, A. Khuroo, T. Monif and P. Partani, J. Pharm. Biomed. Anal., 2013, 3(2), 75–83.
18. J. Zheng, M. W. Jann, Y. Y. Hon and S. A. Shamsi, Electrophoresis, 2004, 25, 2033–2043.
19. R. Theurillat, M. Kuhn and W. Thormann, J. Chromatogr. A, 2002, 979, 353–368.
20. J. M. Sailstad and J. W. A. Findlay, Ther. Drug Monit., 1991, 13, 433–442.
21. H. Bagheri Sadeghi, S. Ahmad Ebrahimi, A. Tamaddon, F. Bozorgvar, H. Afifinia, N. Almasian and S. Mollaei, Electroanalysis, 2011, 23, 2716–2723.
22. S. A. Mohajeri and S. A. Ebrahimi, J. Sep. Sci., 2008, 31, 3595–3602.
23. R. S. Saberi and S. Shahrokhian, Bioelectrochemistry, 2012, 84, 38–43.
24. M. B. Gholivand, G. Malekzadeh and M. Torkashvand, J. Electroanal. Chem., 2013, 692, 9–16.
25. M. Encarnación Burgoa Calvo, O. Domínguez Renedo and M. Julia Arcos Martínez, Talanta, 2007, 74, 59–64.
26. M. Encarnación Burgoa Calvo, O. Domínguez Renedo and M. Julia Arcos Martínez, Anal. Chim. Acta, 2005, 549, 74–80.
27. H. Chen, J. Kong, D. Yuan and G. Fu, Biosens. Bioelectron., 2014, 53, 5–11.
28. M. Kalate Bojdi, M. H. Mashhadizadeh, M. Behbahani, A. Farahani, S. S. Hosseini Davarani and A. Bagheri, Electrochim. Acta, 2014, 136, 59–65.
29. G. Yang, F. Zhao and B. Zeng, Biosens. Bioelectron., 2014, 53, 447–452.
30. B. B. Prasad, D. Jauhari and M. Prasad Tiwari, Biosens. Bioelectron., 2014, 59, 81–88.
31. Y. Zhang, R. Liu, Y. Hu and G. Li, Anal. Chem., 2009, 81, 967–976.
32. R. J. Ansell and K. Mosbach, Analyst, 1998, 123, 1611–1616.
33. S. Lu, G. Cheng and X. Pang, J. Appl. Polym. Sci., 2003, 89, 3790–3796.
34. S. Lu, G. Cheng, H. Zhang and X. Pang, J. Appl. Polym. Sci., 2006, 99, 3241–3250.
35. L. Chen, J. Liu, Q. Zeng, H. Wang, A. Yu, H. Zhang and L. Ding, J. Chromatogr. A, 2009, 1216, 3710–3719.
36. C. J. Tan and Y. W. Tong, Anal. Chem., 2007, 79, 299–306.
37. K. Chuah, L. M. H. Lai, I. Y. Goon, S. G. Parker, R. Amal and J. J. Gooding, Chem. Commun., 2012, 48, 3503–3505.
38. L. M. H. Lai, I. Y. Goon, M. Lim, D. B. Hibbert, R. Amal and J. J. Gooding, J. Electroanal. Chem., 2011, 656, 130–135.
39. X. Wang, L. Y. Wang, X. W. He, Y. K. Zhang and L. X. Chen, Talanta, 2009, 78, 327–332.
40. B. J. Gao, J. H. Lu, Z. P. Chen and J. F. Guo, Polymer, 2009, 50, 3275–3284.
41. Y. Liang, L. Gu, X. Q. Liu, Q. Y. Yang, H. Kajiura, Y. M. Li, T. S. Zhou and G. Y. Shi, Chem.–Eur. J., 2011, 17, 5989–5997.
42. G. Z. Fang, J. Tan and X. P. Yan, Anal. Chem., 2005, 77, 1734–1739.
43. C. H. Lu, W. H. Zhou, B. Han, H. H. Yang, X. Chen and X. R. Wang, Anal. Chem., 2007, 79, 5457–5461.
44. H. F. Li, C. G. Xie, S. Q. Li and K. Xu, Colloids Surf., B, 2012, 89, 175–181.
45. X. L. Zhao, Y. L. Shi, T. Wang, Y. Q. Cai and G. B. Jiang, J. Chromatogr. A, 2008, 1188, 140–147.
46. J. J. Feng, Z. H. Li, Y. F. Li, A. J. Wang and P. P. Zhang, Microchim. Acta, 2012, 176, 201–208.
47. Y. F. Hu, J. X. Li, Z. H. Zhang, H. B. Zhang, L. J. Luo and S. Z. Yao, Anal. Chim. Acta, 2011, 698, 61–68.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra02096h

---