Fully optimized new sensitive electrochemical sensing platform … 2020. 7. 20. · quality control and a routine determination of EZG. 1. Introduction Ezogabine (EZG), ethyl...

Contents lists available at ScienceDirect

Microchemical Journal

journal homepage: www.elsevier.com/locate/microc

Fully optimized new sensitive electrochemical sensing platform for theselective determination of antiepileptic drug ezogabine

Mona A. Mohameda,b,⁎, Ahmed S. Fayedc, Maha A. Hegazyc, Nahla N. Salamaa, Enas E. Abbasa

a Pharmaceutical Chemistry Department, National Organization for Drug Control and Research, Pyramids Ave., P.O. Box 29, Giza, Egyptb Institute of Electronics, Microelectronics and Nanotechnology (IEMN, UMR CNRS 8520), Université Lille 1, Cité Scientifique Avenue Poincaré−BP60069, 59652Villeneuve d'Ascq, Francec Cairo University, Faculty of Pharmacy, Analytical Chemistry Department, Kasr Elaini St., P.O. Box 11562, Cairo, Egypt

A R T I C L E I N F O

Keywords:EzogabineGraphene oxideIonic liquidModified sensorsFactorial designResponse surface

A B S T R A C T

The electrochemical oxidation of Ezogabine (EZG) based on Graphene Oxide Nanosheets (GO) and Ionic Liquids(IL)-modified carbon paste electrodes (CPE) has been studied using different techniques. GO and IL have asynergetic effect giving rise to highly improved electrochemical responses and provide an advantageous platformfor the basis of an electrochemical sensor with excellent performance. Screening for the best experimentalconditions was performed by fractional factorial design, while optimization studies were performed with the aidof central composite design. The optimum conditions were located by the use of Derringer's desirability function.The sensing of EZG via square wave voltammetry (SWV) is found to exhibit two wide linear dynamic ranges of9.89×10−7–1.92× 10−5 M and 1.92× 10−5–2.0×10−4 M at pH 2.0. The limits of detection and quantifi-cation were calculated to be 9.86× 10−8 and 3.28× 10−7 M, respectively. The suggested sensor has been usedsuccessfully for EZG determination in human plasma as real samples. Satisfactory recoveries of analyte fromthese samples are demonstrated indicating that the suggested sensor is highly suitable for clinical analysis,quality control and a routine determination of EZG.

1. Introduction

Ezogabine (EZG), ethyl N‑[2‑amino‑4‑[(4‑fluorophenyl)methyla-mino]phenyl]carbamate, was recently approved by the US Food andDrug and the European Medicines Agency for adjunctive treatment ofpartial-onset seizures in adults [1]. It appears to work by a uniquemechanism of action compared with other currently available anti-epileptic drugs (AEDs) [2]. Considering the importance of EZG as im-portant drug on the human health and its complementary mechanism ofaction, it is important to develop a simple, fast, sensitive, and accurateelectrochemical method for determination of this drug.

EZG has been quantified using several analytical including spec-trophotometry [3,4], fluorimetry [3] and RP-HPLC methods [4–12].The requirement of clinical and pharmacokinetic studies encourages thedevelopment of simple, inexpensive and sensitive analytical methodsfor the determination of pharmaceuticals in its different presentationforms.

Electrochemical techniques own several advantages and are there-fore attractive choices for pharmaceutical analysis. Electrochemicalmethods of analysis have shown to be accurate, sensitive and cost-

effective, which in turns can be used in clinical analysis, quality controland the routine determination of drugs in pharmaceutical formulationsas an alternative to tedious chromatographic techniques [13–16]. Thedirect electrochemical oxidation of pharmaceutical compounds at bareelectrode occurs at high overpotentials. A decrease in the overpotentialalong with increase in the analytical signal/current can be achievedwith chemically modified electrodes [17,18]. Hence, we aim to developa modified electrochemical sensing platform for sensitive determinationof EZG utilizing the benefits of GO along with IL.

In attempts to further enhance the electro-analytical determinationof the EZG, modification of the electrochemical platform with nano-materials has a potential [19]. Graphene-based materials are anotherclass of modifiers that promote effective electronic transfer. They areextensively used in electroanalytical chemistry due to their intrinsicfavorable properties, such as high electrical conductivity, large specificsurface area and good biocompatibility [20]. An important derivative ofgraphene is graphene oxide (GO) which has a similarly good perfor-mance to graphene owing to its high adsorption capacity, large surfacearea and good biocompatibility. In particular, GO contains a largenumber of hydrophilic functional groups, such as eOH, eCOOH and

https://doi.org/10.1016/j.microc.2018.08.062Received 21 May 2018; Received in revised form 29 August 2018; Accepted 29 August 2018

⁎ Corresponding author at: Pharmaceutical Chemistry Department, National Organization for Drug Control and Research, Pyramids Ave., P.O. Box 29, Giza, Egypt.E-mail address: [email protected] (M.A. Mohamed).

Microchemical Journal 144 (2019) 130–138

Available online 30 August 20180026-265X/ © 2018 Published by Elsevier B.V.

T

http://www.sciencedirect.com/science/journal/0026265Xhttps://www.elsevier.com/locate/microchttps://doi.org/10.1016/j.microc.2018.08.062https://doi.org/10.1016/j.microc.2018.08.062mailto:[email protected]://doi.org/10.1016/j.microc.2018.08.062http://crossmark.crossref.org/dialog/?doi=10.1016/j.microc.2018.08.062&domain=pdf

epoxides on the basal plane and the sheet edge [21,22], resulting ingood hydrophilicity that makes it easily dispersed in solvents with long-term stability. Based on these advantages, GO-based electrochemicalsensors have been developed for the sensitive determination of variousbiological molecules [23].

Room temperature ionic liquids (IL) display various unique prop-erties and great application potential in several areas, so they haveconcerned many researchers' interest due to their excellent propertieslike high ionic conductivity, tunable viscosity, and wide electro-chemical windows [24]. The combination of GO-IL has particular ad-vantages, like strong electrical conductivity, a large specific surfacearea, good dispersibility, and stability [25]. So, combined the ad-vantage of ionic liquid and GO nanosheets can be improve the electricalconductivity of unmodified electrodes for trace analysis of electroactivepharmaceutical compounds such as EZG [13,20]. Furthermore, gra-phene-IL based electrochemical sensors and biosensors have also beenwidely reported in literature for direct electron transfer of various redoxenzymes and the detection of various types of compounds, such asdopamine [25,26], H2O2 detection [27], theophylline [28], glucose[29], and mercury [30,31].

To our best knowledge, no electrochemical method was reported forthe determination of EZG. So our study aims to determine EZG usingsimple, accurate, precise and sensitive voltammetric method via themodification of the electrochemical sensor with GO nano-sheets and ILnamely 1‑n‑hexyl‑3‑methylimidazolium hexafluorophosphate. Themorphological and electrochemical characterization of the electrodematerial has been carried out by using various techniques, such asscanning electron microscopy (SEM), CV, electrochemical impedancespectroscopy (EIS) and chronoamperometry (CA). The developed vol-tammetric method based on nanocomposite shows good sensitivity andselectivity for determination of EZG in pharmaceutical formulation andplasma.

2. Experimental

2.1. Instrumentation

The Bio-logic SP 150 electrochemical workstation was used foranalyzing the electrochemical properties of the samples. A one com-partment cell with a three electrode set-up was connected to the elec-trochemical workstation through a C3-stand from BAS (USA). A pla-tinum wire from BAS (USA) was employed as the auxiliary electrode.All the cell potentials were measured with respect to an Ag/AgCl (3.0 MNaCl) reference electrode from BAS (USA). A Cyberscan 500 digital(EUTECH Instruments, USA) pH-meter with a glass combination elec-trode served to carry out pH measurements. All the electrochemicalexperiments were performed at an ambient temperature of 25 °C.Impedance spectra were recorded over a frequency range from100mHz to 100 kHz. Scanning electron microscopy (SEM) measure-ments were carried out using a JSM-6700F scanning electron micro-scope (Japan Electro Company). Fourier Transform infrared (FT-IR)spectra were recorded on an IR-Affinity-1 Fourier Transform InfraredSpectrophotometer (Shimadzu, Japan).The crystalline phases were de-tected and identified using an X-ray diffractometer (XRD) on an X'PertPRO MRD with a copper source at a scan rate (2θ) of 1 °s−1. Sigma Plot10.0 was used for all statistical data. Energy-dispersive X-ray spectro-scopy (EDX) measurements were carried out using a Zeiss Ultra 60 fieldemission scanning electron microscope. Multivariate screening, opti-mization and statistical interpretation were carried out using Design-Expert ® software version 10, Stat-Ease Inc., Minneapolis, USA.

2.2. Pure and synthetic ally prepared samples

Ezogabine was kindly supplied by MEDIZEN Pharmaceuticals (B. N.OP-EZ/04/15/001). Its purity was found to be 99.46 ± 1.38% ac-cording to manufacturer HPLC method. The tablets were synthetically

prepared by mixing 500mg ezogabine, with different excipients;2380mg avicel, 60mg explotab, 30mg magnesium stearate and 30mgaerosol and compressed.

2.3. Chemicals and reagents

Potassium permanganate, hydrogen peroxide solution, sodium ni-trate and graphite powder (particle size< 50 μm) were obtained fromAldrich. Paraffin oil (Merck Co. Germany) was used as the pasting li-quid for the preparation of the paste electrodes. 1‑n‑hex-yl‑3‑methylimidazolium hexafluorophosphate was supplied by AlfaAesar (Karlsruhe, Germany).

Fresh human plasma was obtained from the blood bank (VACSERA,Cairo, Egypt).

Britton-Robinson buffer (B-R buffer) was prepared by mixing dif-ferent volumes of 0.04M H3PO4, 0.04M acetic acid and 0.04M boricacid with the appropriate amount of 0.2 M NaOH to obtain the desiredpH 2.0–10.0. Phosphate buffer (sodium dihydrogen phosphate anddisodium monohydrogen phosphate plus sodium hydroxide, 0.1M)solutions (PBS) with different pH values were used. All solutions wereprepared from analytical grade chemicals and sterilized Milli-Q deio-nized water was used.

2.4. Standard and working solutions

A stock solution of EZG (1.0× 10−2 M) was prepared daily bydissolving 303.3 mg in methanol HPLC-grade in a 100-mL volumetricflask. Working solutions were prepared by appropriate dilution of stockstandard solutions with methanol to obtain a solution in the con-centration range of 1.0× 10−3 to 1.0× 10−5 M. These solutions werestable for about one week at 4 °C.

2.5. Preparation of different unmodified and modified electrodes

Unmodified carbon paste electrode (CPE) was fabricated by mixinggraphite powder (0.5 g) with paraffin oil (0.3 mL) in a pestle andmortar. The carbon paste was packed into the hole of an electrode bodyand scraping off the excess against a conventional paper and polishingthe electrode on a smooth paper to obtain a shiny appearance.Graphene oxide (GO) was prepared by the Hummer's method [32]. Themodified sensor was prepared using a mixture of (970.0, 980.0, and960.0 mg) graphite powder, (10.0, 10.0, and 30.0 mg) of GO and (20.0,10.0, and 10.0 mg) of ionic liquid (1‑n‑hexyl‑3‑methylimidazoliumhexafluorophosphate). The mixture was carefully homogenized for45min using a mortar and pestle. Afterwards, 0.3 mL of paraffin oil wasadded to obtain a paste. This paste was packed into the hole of theelectrode body and smoothed on a filter paper until shiny appearance.A new surface was obtained by pushing excess of the paste out of thesyringe and polished with weighing paper. The obtained modifiedelectrode was named as GO-IL/CPE. Other modified CPE were preparedfor the comparative studies. Accordingly, an IL/CPE or GO/CPE weremade using 10.0 mg of the ionic liquid or GO, respectively, and990.0 mg of graphite and a bare CPE was made using 1000.0 mg ofgraphite. The same amount of paraffin oil was used for the preparationof the partially modified CPEs.

2.6. Experimental design

Two levels randomized experimental design was applied as ascreening step, it includes four factors; buffer type (Britton Robinsonand Phosphate buffers), pulse width, scan rate and composite content.Then, a central composite- response surface design (CCD-RS) was usedto examine the influence of these four factors on the current. Thescreening was performed in order to select the range for each factor.The selected ranges of the variables were; pH (A) 2.0–4.0, scan ratemVs−1 (B) 10–30 and three different preparations of modified sensors

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

131

(C) coded 1, 2 and 3. A total of 20 experiments including six centralpoints were conducted in a randomized order.

2.7. Recommended experimental procedure

Before any voltammetric measurements were made, the modifiedelectrode GO-IL/CPE was cycled between −250.0 and 850.0 mV with ascan rate of 100mV s−1 in 4.0×10−2 M B-R buffer solution of pH 2.0several times until a reproducible response was achieved. Followingthis, the modified GO-IL/CPE electrode was transferred into anothercell containing 4.0× 10−2 M B-R buffer of pH 2.0 and the properamount of EZG. Cyclic voltammograms (CV) were recorded between−250.0 and 850.0 mV with a scan rate of 100mV s−1.

2.8. Recommended procedure for the calibration curve

Aliquots equivalent to, 9.89×10−7–2.00× 10−4 M of EZG weretransferred separately into a series of 5-mL volumetric flasks usingmicropipette and the volume was completed to the mark with B-Rbuffer pH 2.0. The solution was transferred to the electrolytic cell thensquare wave voltammogram (SWV) was recorded. In this study, a scanrate of 20mVs−1 was chosen because at this value the sensitivity wasrelatively high and the voltammetric curves are well-shaped with arelatively narrow peak width. To obtain relatively high and narrowpeaks, the values of 40mV, 30ms and 20mV s−1 were finally chosenfor pulse amplitude, pulse width and scan rate, respectively.Determination of EZG by SWV analytical procedure was validated ac-cording to the International Conference on Harmonization (ICH)guidelines [33].

2.9. Analysis of real samples

Aliquots equivalent to 1.50×10−7, 35.00×10−6, 6.25×10−5,and 1.00×10−4 M of EZG were transferred separately into a series of5-mL volumetric flasks spiked with 20 μL of fresh plasma samples usingmicro pipette and the volume was completed to the mark with B-Rbuffer pH 2.0. The solution was transferred into the electrochemical cellto be analyzed without any further pretreatment.

2.10. Application to the pharmaceutical formulation

Ten synthetically prepared tablets (50mg EZG/tablet) wereweighed and ground. Accurately weighed amount about 303.3mg wasdissolved in 50mL methanol HPLC-grade in 100-mL volumetric flasksand sonicated for 15min. Then the volume was completed with thesame solvent. The solution was then filtered through 0.45 μm filterpaper. The obtained solution was contained 1.0×10−2 M EZO. Theprocedure described under “Recommended procedure for calibrationcurve” was then followed.

3. Results and discussion

3.1. Surface characterization and electrochemical characterization of thefabricated sensors

Fig. 1 displays typical SEM images of the fabricated electrodes. Thesurface of the CPE is dominated with uniform and smooth shapedgraphite flakes and separated layers (Fig. 1A). Fig. 1B shows the for-mation of fairly homogeneous GO structures with an average length of400 nm. The crystal structure of the synthesized materials was in-vestigated by XRD. The XRD pattern of GO showed a typical (002) GOpeak at 2θ=10°, Fig. 1D. On the other hand, graphite flakes exhibit astrong and sharp peak at 26.26°, indicating highly ordered structure[20,34]. The FT-IR spectrum of the GO confirms the successful oxida-tion of graphite, Fig. 1E.

Also, the broad and wide peaks appeared at 3439 which is referred

to eOH group and 1719 cm−1 is attributed to the carboxyl group [35],confirming the presence of different types of oxygen functionalities inGO [13,36]. While the absorption band at 1560 cm−1 can be ascribed tobenzene rings [37], the sharp intense peak at 1419 cm−1 can be at-tributed to CO-carboxylic [22].

By comparing the energy-dispersive X-ray (EDX) spectra of un-modified graphite CPE, and GO, it is obvious that the existence of smallpeak density for oxygen in case of bare graphite CPE. On the other handin case of GO, the oxygen peak rises to about 65%, indicating completeoxidation of graphite into GO, Fig. 1E and F.

The electroactive area of the electrode was obtained by the cyclicvoltammetric method using 5.0 mM K3Fe(CN)6 in 0.1 M KCl at differentscan rates, using the Randles-Ševčík [38]. The calculated areas were0.055, 0.077, 0.109 and 0.145 cm2 for CPE, GO/CPE, IL/CPE, and GO-IL/CPE, respectively.

In order to elucidate the electrochemical characterization of GO-IL/CPE, its CVs were compared with CPE, GO/CPE, and IL/CPE CVs in asolution containing 5.0 mM [Fe(CN)6]3−/4− as the electrochemicalprobe, Fig. 2A. The redox peaks for GO-IL/CPE were 2.7 and 1.7, and1.3 times higher than those of CPE, GO/CPE, and IL/CPE, respectively.This could be attributed to the more catalytic activity of GO and ILwhen they were together.

Electrochemical impedance spectroscopy (EIS) was used to obtain aconvenient determination of both mass-transport and kinetic para-meters as well as the charge transfer coefficient. Fig. 2B depicts the EISresponse for CPE, GO/CPE, IL/CPE and GO-IL/CPE. EIS measurementswere performed in 5.0 mM K3Fe(CN)6 (1:1) solution in 0.1M KCl.

As can be seen, a semi-circle was verified at the region of highfrequencies and, as is commonly known, the diameter of this semi-circleis equivalent to the charge transfer resistance across the electrode in-terface. Thus, as a preliminary qualitative analysis of the Nyquist plots,the decrease of the semi-circle diameter for the modified CPEs with GO,IL or GO-IL is observed, indicating an improvement of the charge-transfer in these electrodes compared to the unmodified CPE.

The simulated circuit model (inset of Fig. 2B) was selected to fit theacquired impedance data for the electron transfer process (modelled asa resistance to charge transfer, RCT) and the diffusion process (modelledas the Warburg-type impedance, W). The RCT and W were both inparallel with the interfacial capacitance (C). By fitting the data, RCT wasestimated to be 5549.12Ω at the bare CPE, while it decreased to1988.95 and 480.05Ω for GO/CPE and IL/CPE, respectively, and thendecreased to 148.60Ω for GO-IL/CPE, indicating the significantly lowerelectron-transfer resistance of GO-IL/CPE compared with other elec-trodes. The results are consistent with the CV results, demonstratingthat the GO-IL/CPE composite might provide higher electron conduc-tion pathways due to the synergistic effects of its constituents.

Fig. 2C shows typical cyclic voltammograms of the bare CPE, GO/CPE, IL/CPE and GO-IL/CPE towards the electrochemical sensing ofEZG. In the absence of EZG, no measurable anodic or cathodic peakcurrent was observed indicating that the GO-IL/CPE sensor has noelectro-activity in the selected potential range. Regarding the CV, thegood peaks with maximum current were observed at GO-IL/CPE, whichimplies that the reaction is more facile at the modified electrode.

In the case of the bare/unmodified CPE, the anodic peak current,which corresponds to the electrochemical oxidation of EZG, was30.25 μA at 530.0 mV. As depicted in Fig. 2C, the magnitude of thevoltammetric response of the sensor increases due to the presence of GOwhere the anodic peak appears at 491mV with a current value of 38 μA.Thus, GO potentially acts as a promoter/mediator to increase the rate ofelectron transfer due to its catalytic capability. The voltammetric signalis further enhanced through the incorporation of IL, as is evident fromFig. 2C, as the anodic peak current increased to a value of 65.5 μA at472.0 mV. The above results indicate that GO-IL/CPE had better elec-trochemical activity.

The effect of solution pH on the electrochemical oxidation of EZGwas next explored using cyclic voltammetry in the range of 2.0–9.0.

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

132

Fig. 3 shows the peak potentials were shifted towards less positivepotentials with an increase in pH from 2.0 to 9.0 which is attributed tothe protonation–deprotonation properties of EZG [39]. As can be de-picted from Fig. 3, the peak current was decreased rapidly with the riseof pH. Which substantiate the choice of pH 2.0 during multivariateexperimental optimization discussed above.

Analysis of the voltammetric responses in terms of the anodic peakpotential, Ep, shows a linear relationship with the pH of the buffer so-lution according to the following equation:

= − =E (V) 0.545 0.0294 pH; R 0.9988.p 2

As the pH is shifted towards more alkaline conditions, the cyclic

voltammetric profiles change distinctly. Returning to Fig. 3, the line ofbest fit (53.4 mV per pH unit at 25 °C) obeys the Nernst equation in-dicating that the electrochemical process involves one electron and oneproton transfer action. The slope, 29.4mV pH−1, which is close to theNernstian value of −29.5mV for a two-electron, one-proton process.

The suggested schematic representation for the oxidation of EZG atGO-IL/CPE is shown in Scheme S1.

Next, the effect of scan rate on EZG oxidation and redox peakscurrent were investigated using cyclic voltammetry to understand themechanism of the electrode reaction. The influence of scan rate (v) onthe electrochemical response using GO-IL/CPE sensing platforms re-corded in 1.0 mM EZG was studied, and the results are shown in Fig. S1.

Fig. 1. SEM images of (A) CPE, (B) GO, (C) XRD of graphite and GO, and (D) IR spectra of GO, (E) EDX of graphite, and (F) EDX of GO.

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

133

As shown in Fig. S1A, both the oxidation and reduction currentsincreased gradually with the increase of the scan rate. The redox peakcurrents showed a linear relationship with the scan rate (Fig. S1B),demonstrating an adsorption-controlled process. The two linear re-gression equations were calculated as: Ipa= 0.61v+9.00(R2=0.9945) and Ipc=−0.17v− 6.88 (R2=0.9991).

Also, plots of logarithm of peak current versus the logarithm of scanrate were constructed; giving straight lines with slopes of 0.77 and 0.60for anodic and cathodic currents, respectively (Fig. S1C). The values ofthe slopes are near to the theoretical value of 1.0 [40], which is ex-pected for an adsorption-controlled redox system, confirming theabove-mentioned result. Furthermore, with the increase of the scanrate, the oxidation peak potential (Epa) shifted towards the positivedirection, meanwhile, the reduction peak potential (Epc) moved nega-tively. Accordingly, the peak-to-peak separation (ΔEp) slightly in-creased. This result reveals that the electrochemical process is quasi-reversible [41].

Furthermore, the chronoamperometric measurements of EZG wereperformed at a constant applied direct current (DC) potential at GO-IL/CPE in pH 2.0 buffer solution with the current–time profiles obtained bysetting the working electrode potential at +500mV vs. Ag/AgCl forvarious concentrations of EZG, Fig. S2. The inset S2A shows the plots ofcurrents, sampled at fixed time as a function of EZG concentration,added to the blank solution at different times after the application ofthe potential step. For an electroactive material with the diffusioncoefficient (D), the current corresponding to the electrochemical reac-tion (under diffusion control) is described by Cottrell's law [40]:

⎜ ⎟= ∗⎛⎝

⎞⎠

I t nFAC Dt

( )π

The level of the Cottrell current, measured for 25 s, increased byincreasing EZG concentration. The plot of I versus t−1/2, showed astraight line (inset S2B) and from its slope, the value of D could beobtained. The diffusion coefficient of EZG was calculated to be1.29×10−6 cm2 s−1. Optimization of experiments is a valuable pro-cedure to determine the effects of factors depending on definite

responses [42]. So, central composite response surface design was ap-plied (CCD-RS) for the optimization of the experimental conditions forSWV. It constitutes an excellent approach due to possibility of in-vestigating a high number of variables at different levels with only alimited number of experiments. The variables presented in Table 1 werechosen based on our screening two-level factorial design. The response(current) values ranged from 15 to 90. A design matrix was generatedbetween factors and responses by response surface regression analysis.ANOVA was generated for reduced surface designed models exhibitedthat curvature is significant for the three factors studied. Since valuesare< 0.05, the reduced quadratic model was advised. Model termswith values> 0.05 were eliminated by backward elimination toachieve more realistic models [43].

In the present study, the predicted and adjusted R values for eachresponse were in a good agreement [44] which revealed that the ex-perimental data is in good fit with the second order polynomial equa-tions. All the reduced models value 4 is desirable) [45] andtherefore indicates adequate model discrimination. The coefficient ofvariation (% C.V.) was found to be

using GO-IL/CPE in B-R buffer pH 2.0 solution containing various in-dividual concentrations of EZG. The calibration range was establishedthrough consideration of the necessary practical range, according to EZGconcentration present in the pharmaceutical product, to give accurate,precise and linear results, Fig. 5. The plot of the peak current vs. EZGconcentration showed two linear segments with different slopes for theEZG concentrations: for 9.89×10−7–1.92×10−5M, the regressionequation was I(μA)=0.574C+4.00 (R2=0.9986), while for1.92×10−5–2.00×10−4M, the regression equation was I(μA)=0.216C+11.09 (R2=0.9991). The decrease in sensitivity(slope) of the second linear segment is likely due to kinetic limitations[46].

Based on the first linear fitting equation of EZG, the limits of de-tection and quantification were calculated to be 9.86×10−8 and3.28×10−7 M, respectively, Table 2.

Intra-day and inter-day (n= 5) relative standard deviations ofsamples concentrations of EZG were calculated and listed in Table S1.The stability of the modified electrode has been investigated where the

peak current did not change upon storage in air for 10 days. Themodified electrode retained 99.02% of its initial response up to1month.

3.2. Analysis of EZG in real samples

The applicability of GO-IL/CPE to the determination of EZG inhuman plasma was examined using SWV. Concentrations were mea-sured by applying the calibration plot using the standard additionmethod. The results are shown in Table S2. The recovery values show agood accuracy of the suggested sensor. Referring to the experimentalresults, it is very clear that this sensor has an excessive potential for thedetermination of trace amounts of these compounds in biological fluids.As well as, dosage forms (tablet, containing 50mg EZG) were analyzedusing the same parameters of SWV experiment. The recovery experi-ment was presented in Table S2, which show that the GO-IL/CPE ishighly sensitive for detecting low and higher concentrations of EZG inreal samples.

/µ

-60

-40

-20

0

20

40

60

-0.2 0.00.2 0.4

0.6 0.81.0 1.2

1.4

0

2

4

6

8

10

IA

E/V(vs Ag/AgCl)

pH

pH 2pH 3pH 4pH 5pH 6pH 7pH 8pH 9

pH

0 2 4 6 8 10

E/V

(vs

Ag/

AgC

l)

0.2

0.3

0.4

0.5

0.6Ep(V) = 0.545-0.0294 pH

(R2

= 0.9988)

Fig. 3. Cyclic voltammetric responses of 1.0 mM EZG at different pH values using GO-IL/CPE sensing platforms. Scan rate: 100mV s−1. The inset shows a plot of theanodic peak potential of EZG versus pH.

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

135

3.3. Interference studies

Many substances, which can possibly interfere with the electro-analytical determination of EZG, were determined using the re-commended conditions. Several main interferences were selected,which may be found in the pharmaceutical preparations and other

Table 1Central composite design consists of experiments for the study of three ex-perimental factors with the results.

Std Run Factor 1 Factor 2 Factor 3 Response

A: pH B: scan rate C: modified sensors Current

14 1 3 20 3 3816 2 3 20 2 281 3 2 10 1 4911 4 3 10 2 175 5 2 10 3 558 6 4 30 3 196 7 4 10 3 154 8 4 30 1 3215 9 3 20 2 312 10 4 10 1 1918 11 3 20 2 307 12 2 30 3 5313 13 3 20 1 469 14 2 20 2 9017 15 3 20 2 2920 16 3 20 2 3010 17 4 20 2 183 18 2 30 1 8519 19 3 20 2 2912 20 3 30 2 53

Design-Expert® Software

1/Sqrt(current1)

Lambda

Current = -0.5

Best = -0.3

Low C.I. = -0.98

High C.I. = 0.35

Recommend transform:

Inverse Sqrt

(Lambda = -0.5)

Lambda

Ln(R

esid

ualS

S)

Box-Cox Plot for Power Transforms

6

7

8

9

10

11

-3 -2 -1 0 1 2 3

7.03418

Fig. 4. Box Cox plot for power transform of the response.

E/V(vs Ag/AgCl)

0.0 0.2 0.4 0.6 0.8 1.0 1.2Ip

/ A

0

20

40

60

[EZG]/ M

0 50 100 150 200 250

Ip/

A

0

10

20

30

40

50

60

70

Fig. 5. SWV of EZG at GO-IL/CPE in B–R buffer pH 2.0 at a scan rate of 0.1 V s−1

solution containing different concentrations. Insets: the plot of the peak currentas a function of EZG in concentration range 9.89×10−7–1.92×10−5M and1.92×10−5–2.00×10−4M.

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

136

interfering ions. By adding these materials to a fixed EZG concentrationof 1.0× 10−5 M in a pH 2.0 B-R buffer, the current of EZG was ex-amined. From the selected materials, there was no interference ob-tained that could affect the electro-analytical sensing of EZG and thetolerance limit was less than± 4% for the selected interference sub-stance. The results showed that 100-fold of sucrose, glucose, fructose,lactose, and citric acid, 200-fold of potassium sorbate, stearic acid, talcpowder, magnesium stearate, povidone, microcrystalline cellulose,starch, cellulose, colloidal silicon dioxide, croscarmellose sodium,Cu2+, Mn+2, Zn+2, Mg+2, Fe2+, Fe3+, Ca2+ and uric acid did notaffect the selectivity. Also, the results depicted that 150-fold ofNO3−,SO42−, K+, Na+ and Cl−; 250-fold for some metals including:Cd2+, Co2+, Cr3+, Pb2+, Cu2+, Ni2+, and Hg2+ did not interfere withEZG. Results obtained by the proposed SWV method was statisticallycompared with those obtained by applying the manufactureringmethod in both drug substance and drug product, where no significantdifference was obtained regarding both accuracy and precision,Table S3.

4. Conclusion

In this study, we have reported for the first time, the electro-analytical sensing of EZG using GO and IL composite modified elec-trode. The fabricated GO-IL/CPE associated to the excellent electro-catalytic activity of the GO and IL showed high selectivity andsensitivity for the determination of EZG with low detection limits andwide concentration ranges. Electrochemical sensing coupled with che-mometric tools can provide a complete profile of a separation processand is considered highly powerful, providing useful information offactors' interactions. The obtained models demonstrate a significantinfluence of the studied factors. Optimal conditions were obtained byapplying Derringer's desirability function based on defining a globaldesirability according to the specified constraints. In addition to thegood properties of high sensitivity, accuracy, and reproducibility, thiscould provide a simple and suitable method for the quantitative de-termination of low level of EZG for clinical laboratories.

Acknowledgment

The authors acknowledge funding from The Science and TechnologyDevelopment Fund (STDF), Institut Français d'Egypte (IFE) (ID: 30636)and Dr. Nageh K. Allam (Director of the Energy Materials Laboratory(EML), The American University in Cairo (AUC)) for the support of thisresearch.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.microc.2018.08.062.

References

[1] M.J. Gunthorpe, C.H. Large, R. Sankar, The mechanism of action of retigabine(ezogabine), a first-in-class K+ channel opener for the treatment of epilepsy,Epilepsia 53 (2012) 412–424.

[2] J.J. Łuszczki, Third-generation antiepileptic drugs: mechanisms of action, phar-macokinetics and interactions, Pharmacol. Rep. 61 (2009) 197–216.

[3] F. Ibrahim, A. El-Yazbi, M. Wagih, M. Barary, Sensitive inexpensive spectro-photometric and spectrofluorimetric analysis of ezogabine, levetiracetam and to-piramate in tablet formulations using Hantzsch condensation reaction,Spectrochim. Acta A Mol. Biomol. Spectrosc. 184 (2017) 47–60.

[4] N. Tamilselvi, A. Rajasekaran, Stability-indicating RP-HPLC method for the de-termination of ezogabine and identification of its degradation products, J. Pharm.Sci. Res. 8 (2016) 19–23.

[5] B. Lakshmi, K. Saraswathi, T. Reddy, Rp-hplc method development and validationfor the analysis of ezogabine in pharmaceutical dosage forms, Int. J. A Ps BMS 1(2012) 7–14.

[6] P. Satyanarayana, A.S. Madhavi, Validated RP-HPLC method for the estimation ofezogabine in tablet dosage form, Int. J. Res. Pharmaceut. Biomed. Sci. 3 (2012)1261–1265.

[7] X. Wang, H. Zhou, J. Zheng, C. Huang, W. Liu, L. Yu, S. Zeng, Identification andcharacterization of four process-related impurities in retigabine, J. Pharm. Biomed.Anal. 71 (2012) 148–151.

[8] D. Zhang, X. Song, J. Su, Isolation, identification and structure elucidation of twonovel process-related impurities of retigabine, J. Pharm. Biomed. Anal. 99 (2014)22–27.

[9] M. Douša, J. Srbek, S. Rádl, J. Černý, O. Klecán, J. Havlíček, M. Tkadlecová,T. Pekárek, P. Gibala, L. Nováková, Identification, characterization, synthesis andHPLC quantification of new process-related impurities and degradation products inretigabine, J. Pharm. Biomed. Anal. 94 (2014) 71–76.

[10] W. Bu, M. Nguyen, C. Xu, C.-C. Lin, L.-T. Yeh, V. Borges, Determination of N-acetylretigabine in dog plasma by LC/MS/MS following off-line μElution 96-well solidphase extraction, J. Chromatogr. B 852 (2007) 465–472.

[11] X. Wu, F. Shao, C. Tao, J. Li, Development and validation of a stability indicatingHPLC method for the determination of retigabine and its related substances in drugsubstances, J. Chin. Pharm. Sci. 24 (2015) 241–249.

[12] P. Ravisankar, C. Lokapavani, C. Devadasu, G. Rao, An improved RP-HPLC methodfor the quantitative determination and validation of retigabine in bulk and phar-maceutical formulation, Int. J. Res. Pharm. Sci. 4 (2014) 21–26.

[13] M.A. Mohamed, S.A. Atty, H.A. Merey, T.A. Fattah, C.W. Foster, C.E. Banks,Titanium nanoparticles (TiO2)/graphene oxide nanosheets (GO): an electro-chemical sensing platform for the sensitive and simultaneous determination ofbenzocaine in the presence of antipyrine, Analyst 142 (2017) 3674–3679.

[14] M.A. Mohamed, A.S. Saad, S.H. Koshek, M.R. Elghobashy, Smart electrochemicalsensing platform for the simultaneous determination of psychotic disorders drugsisopropamide iodide and trifluoperazine hydrochloride, New J. Chem. 42 (2018)9911–9919.

[15] M.A. Mohamed, F.M. El-Badawy, H.S. El-Desoky, M.M. Ghoneim, Magnetic cobaltferrite nanoparticles CoFe2O4 platform as an efficient sensor for trace determinationof Cu(II) in water samples and different food products, New J. Chem. 41 (2017)11138–11147.

[16] M.A. Mohamed, N.S. Abdelwahab, C.E. Banks, Electroanalytical sensing of the an-timicrobial drug linezolid utilising an electrochemical sensing platform based upona multiwalled carbon nanotubes/bromocresol green modified carbon paste elec-trode, Anal. Methods 8 (2016) 4345–4353.

[17] M.A. Mohamed, A.M. Yehia, C.E. Banks, N.K. Allam, Novel MWCNTs/grapheneoxide/pyrogallol composite with enhanced sensitivity for biosensing applications,Biosens. Bioelectron. 89 (2017) 1034–1041.

[18] M.A. Mohamed, S.A. Atty, A.M. Yehia, C.W. Foster, C.E. Banks, N.K. Allam,Electrochemical determination of the serotonin reuptake inhibitor, dapoxetine,using cesium–gold nanoparticles, ACS Omega 2 (2017) 6628–6635.

[19] M.A. Mohamed, D.M. El-Gendy, N. Ahmed, C.E. Banks, N.K. Allam, 3D spongygraphene-modified screen-printed sensors for the voltammetric determination ofthe narcotic drug codeine, Biosens. Bioelectron. 101 (2018) 90–95.

[20] M.A. Mohammed, A.K. Attia, H.M. Elwy, Electrochemical sensor based on multi-walled carbon nanotube, alizarine red S and chitosan for simultaneous determi-nation of oxomemazine hydrochloride, paracetamol and guaifenesin,Electroanalysis 29 (2017) 506–513.

[21] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu,S.T. Nguyen, R.S. Ruoff, Synthesis of graphene-based nanosheets via chemical re-duction of exfoliated graphite oxide, Carbon 45 (2007) 1558–1565.

[22] M.A. Mohamed, S.A. Atty, N.N. Salama, C.E. Banks, Highly selective sensing plat-form utilizing graphene oxide and multiwalled carbon nanotubes for the sensitivedetermination of tramadol in the presence of co-formulated drugs, Electroanalysis29 (2017) 1038–1048.

[23] Y. Wang, Y. Li, L. Tang, J. Lu, J. Li, Application of graphene-modified electrode forselective detection of dopamine, Electrochem. Commun. 11 (2009) 889–892.

[24] M. Bijad, H. Karimi-Maleh, M.A. Khalilzadeh, Application of ZnO/CNTs nano-composite ionic liquid paste electrode as a sensitive voltammetric sensor for

Table 2Analytical parameters for the determination of EZG in B-R buffer pH 2.0 at GO-IL/CPE using proposed SWV method (n=3).

Analytical parameter SWV method

1st linear segment 2nd linear segment

Linear concentrationrange (M)

9.89× 10−7–1.92× 10−5 1.92× 10−5–2.00× 10−4

Intercept (μA) 4.00 11.09Slope (μA μM−1) 0.574 0.216Standard deviation of

intercept (μA)0.07 0.06

Standard deviation ofslope (μA μM−1)

0.09 0.05

Coefficient ofdetermination(R2)

0.9986 0.9991

LOD (M) 9.86× 10−8 –LOQ (M) 3.28× 10−7 –

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

137

https://doi.org/10.1016/j.microc.2018.08.062https://doi.org/10.1016/j.microc.2018.08.062http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0005http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0005http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0005http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0010http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0010http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0015http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0015http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0015http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0015http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0020http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0020http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0020http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0025http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0025http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0025http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0030http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0030http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0030http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0035http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0035http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0035http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0040http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0040http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0040http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0045http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0045http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0045http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0045http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0050http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0050http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0050http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0055http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0055http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0055http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0060http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0060http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0060http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0065http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0065http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0065http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0065http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0070http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0070http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0070http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0070http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0075http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0075http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0075http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0075http://refhub.elsevier.com/S0026-265X(18)30621-0/rf7869http://refhub.elsevier.com/S0026-265X(18)30621-0/rf7869http://refhub.elsevier.com/S0026-265X(18)30621-0/rf7869http://refhub.elsevier.com/S0026-265X(18)30621-0/rf7869http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0085http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0085http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0085http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0090http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0090http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0090http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0095http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0095http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0095http://refhub.elsevier.com/S0026-265X(18)30621-0/rf9773http://refhub.elsevier.com/S0026-265X(18)30621-0/rf9773http://refhub.elsevier.com/S0026-265X(18)30621-0/rf9773http://refhub.elsevier.com/S0026-265X(18)30621-0/rf9773http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0105http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0105http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0105http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0110http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0110http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0110http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0110http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0115http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0115http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0120http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0120

determination of ascorbic acid in food samples, Food Anal. Methods 6 (2013)1639–1647.

[25] X. Zhuang, D. Chen, S. Zhang, F. Luan, L. Chen, Reduced graphene oxide functio-nalized with a CoS2/ionic liquid composite and decorated with gold nanoparticlesfor voltammetric sensing of dopamine, Microchim. Acta 185 (2018) 166.

[26] X. Zhuang, H. Wang, T. He, L. Chen, Enhanced voltammetric determination ofdopamine using a glassy carbon electrode modified with ionic liquid-functionalizedgraphene and carbon dots, Microchim. Acta 183 (2016) 3177–3182.

[27] D. Chen, X. Zhuang, J. Zhai, Y. Zheng, H. Lu, L. Chen, Preparation of highly sen-sitive Pt nanoparticles-carbon quantum dots/ionic liquid functionalized grapheneoxide nanocomposites and application for H2O2 detection, Sensors Actuators BChem. 255 (2018) 1500–1506.

[28] X. Zhuang, D. Chen, S. Wang, H. Liu, L. Chen, Manganese dioxide nanosheet-de-corated ionic liquid-functionalized graphene for electrochemical theophylline bio-sensing, Sensors Actuators B Chem. 251 (2017) 185–191.

[29] X. Zhuang, C. Tian, F. Luan, X. Wu, L. Chen, One-step electrochemical fabrication ofa nickel oxide nanoparticle/polyaniline nanowire/graphene oxide hybrid on aglassy carbon electrode for use as a non-enzymatic glucose biosensor, RSC Adv. 6(2016) 92541–92546.

[30] N. Zhou, H. Chen, J. Li, L. Chen, Highly sensitive and selective voltammetric de-tection of mercury (II) using an ITO electrode modified with 5-methyl-2-thiouracil,graphene oxide and gold nanoparticles, Microchim. Acta 180 (2013) 493–499.

[31] N. Zhou, J. Li, H. Chen, C. Liao, L. Chen, A functional graphene oxide-ionic liquidcomposites–gold nanoparticle sensing platform for ultrasensitive electrochemicaldetection of Hg2+, Analyst 138 (2013) 1091–1097.

[32] W.S. Hummers Jr., R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc.80 (6) (1958) 1339-1339.

[33] Validation of Analytical Procedure-Methodology (Q2R1), International Conferenceon Harmonization (ICH) of Technical Requirements for Registration ofPharmaceuticals for Human Use, Geneva, Switzerland, 2005.

[34] S. Bykkam, K. Rao, C. Chakra, T. Thunugunta, Synthesis and characterization ofgraphene oxide and its antimicrobial activity against Klebsiella and Staphylococcus,

Int. J. Adv. Biotechnol. Res. 4 (2013) 142.[35] L. Tang, Y. Wang, Y. Li, H. Feng, J. Lu, J. Li, Preparation, structure, and electro-

chemical properties of reduced graphene sheet films, Adv. Funct. Mater. 19 (2009)2782–2789.

[36] T. Zhang, D. Zhang, M. Shen, A low-cost method for preliminary separation of re-duced graphene oxide nanosheets, Mater. Lett. 63 (2009) 2051–2054.

[37] Y. Si, E.T. Samulski, Synthesis of water soluble graphene, Nano Lett. 8 (2008)1679–1682.

[38] R.G. Compton, C.E. Banks, Understanding voltammetry, World Scientific, 2007.[39] G. Ziyatdinova, E. Ziganshina, H. Budnikov, Electrochemical reduction and quan-

tification of menadione in sodium dodecyl sulfate micellar media, J. Solid StateElectrochem. 17 (2013) 2679–2685.

[40] A.J. Bard, L.R. Faulkner, J. Leddy, C.G. Zoski, Electrochemical Methods:Fundamentals and Applications, Wiley New York, 1980.

[41] S. Hu, Y. Wang, X. Wang, L. Xu, J. Xiang, W. Sun, Electrochemical detection ofhydroquinone with a gold nanoparticle and graphene modified carbon ionic liquidelectrode, Sensors Actuators B Chem. 168 (2012) 27–33.

[42] M.A. Mohamed, M.M. Hasan, I.H. Abdullah, A.M. Abdellah, A.M. Yehia, N. Ahmed,W. Abbas, N.K. Allam, Smart bi-metallic perovskite nanofibers as selective andreusable sensors of nano-level concentrations of non-steroidal anti-inflammatorydrugs, Talanta 185 (2018) 344–351.

[43] J. Parajo, J. Alonso, M. Lage, D. Vazquez, Empirical modeling of eucalyptus woodprocessing, Bioprocess Eng. 8 (1992) 129–136.

[44] T. Lundstedt, E. Seifert, L. Abramo, B. Thelin, Å. Nyström, J. Pettersen, R. Bergman,Experimental design and optimization, Chemom. Intell. Lab. Syst. 42 (1998) 3–40.

[45] Q.K. Beg, V. Sahai, R. Gupta, Statistical media optimization and alkaline proteaseproduction from Bacillus mojavensis in a bioreactor, Process Biochem. 39 (2003)203–209.

[46] V. K Gupta, A. Nayak, S. Agarwal, B. Singhal, Recent advances on potentiometricmembrane sensors for pharmaceutical analysis, Comb. Chem. High ThroughputScreen. 14 (2011) 284–302.

M.A. Mohamed et al. Microchemical Journal 144 (2019) 130–138

138

http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0120http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0120http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0125http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0125http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0125http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0130http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0130http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0130http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0135http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0135http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0135http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0135http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0140http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0140http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0140http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0145http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0145http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0145http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0145http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0150http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0150http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0150http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0155http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0155http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0155http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0160http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0160http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0165http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0165http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0165http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0170http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0170http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0170http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0175http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0175http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0175http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0180http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0180http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0185http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0185http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0190http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0195http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0195http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0195http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0200http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0200http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0205http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0205http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0205http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0210http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0210http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0210http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0210http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0215http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0215http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0220http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0220http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0225http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0225http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0225http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0230http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0230http://refhub.elsevier.com/S0026-265X(18)30621-0/rf0230

Fully optimized new sensitive electrochemical sensing platform for the selective determination of antiepileptic drug ezogabineIntroductionExperimentalInstrumentationPure and synthetic ally prepared samplesChemicals and reagentsStandard and working solutionsPreparation of different unmodified and modified electrodesExperimental designRecommended experimental procedureRecommended procedure for the calibration curveAnalysis of real samplesApplication to the pharmaceutical formulation

Results and discussionSurface characterization and electrochemical characterization of the fabricated sensorsAnalysis of EZG in real samplesInterference studies

ConclusionAcknowledgmentSupplementary dataReferences