Electrochemical Determination of Paracetamol Using Fe 3 O 4 / Reduced Graphene-Oxide-Based Electrode

The synthesis ofmagnetic iron oxide/reduced graphene oxide (Fe3O4/rGO) and its application to the electrochemical determination of paracetamol using Fe3O4/rGO modified electrode were demonstrated. The obtained materials were characterized by means of X-ray diffraction (XRD), nitrogen adsorption/desorption isotherms, X-ray photoelectron spectroscopy (XPS), transmission electron microscope (TEM), Fourier transform infrared spectroscopy (FTIR), and magnetic measurement. The results showed that Fe3O4/rGO composite exhibited high specific surface area, and its morphology consists of very fine spherical particles of Fe3O4 in nanoscales. Fe3O4/rGO was used as an electrode modifier for the determination of paracetamol by differential pulseanodic stripping voltammetry (DP-ASV).The preparation of Fe3O4/rGO-based electrode and some factors affecting voltammetric responses were investigated. The results showed that Fe3O4/rGO is a potential electrode modifier for paracetamol detection by DP-ASV with a low limit of detection. The interfering effect of uric acid, ascorbic acid, and dopamine on the current response of paracetamol has been reported. The repeatability, reproducibility, linear range, and limit of detection were also addressed. The proposed method could be applied to the real samples with satisfactory results.


Introduction
Paracetamol (N-acetyl-p-aminophenol, acetaminophen) (henceforth PRC) is widely used as an active ingredient in pharmaceutical preparations as it is not considered to be carcinogenic at therapeutic doses [1].PRC is available in different dosage forms: tablets, capsules, drops, elixirs, and suspensions [2].A limited use of PRC does not cause any harmful side effects.However, overdosing and the chronic use of PRC produce toxic metabolite accumulation that will cause kidney and liver damage [1,3].Therefore, a sensitive, accurate, fast, and simple analytical method for estimating PRC in pharmaceutical preparations and human plasma is needed.
Numerous methods have been employed for detecting PRC, alone and in mixtures, in biological samples and formulations, such as HPLC (high-performance liquid chromatography) [4,5], liquid chromatography-mass spectrometry (LC-MS) [2,6,7], spectrofluorimetry [8], electrochemical techniques [9,10], chemiluminescence [11,12].However, these methods suffer from several disadvantages such as being time-consuming, high cost, and a requirement for sample pretreatment.In some cases, low sensitivity and selectivity limited their application.Anodic stripping voltammetric techniques (ASV) have been recognized as effective techniques for inorganic and organic compounds analysis because of numerous advantages such as faster analysis, higher selectivity and sensitivity, low cost, low detection limit, and a possibility of performing analysis in situ [13,14].Differential pulse-anodic stripping voltammetry (DP-ASV), considered as one kind of ASV methods, has been used for estimating several organic compounds because of its remarkably high sensitivity.Glassy carbon electrodes chemically modified with porous materials such as mesoporous materials [15] and metal organic frameworks (MOFs) [16] have received considerable attention for ASV because they present significant improvements in terms of fast response, high selectivity, low detection limit, and renewability.The DP-ASV method has been widely employed to detect PRC in pharmaceutical preparations.The porous-material-modified electrodes, for instance, graphene-based electrodes [17], multiwalled carbon nanotube modified basal plane pyrolytic graphite electrode [18], and nanoparticles bismuth oxide modified glassy carbon electrode (GCE) [19], have been often used for electrochemical studies of PRC because of their unique properties such as high surface areas and many active sites.
Superparamagnetic iron oxides nanoparticles (NPs) hold a lot of promise for applications in electrochemical fields because of the unusual structure, excellent adsorption, catalytic properties, and inherent electrical conductivity [9,20].Reduced graphene oxide (rGO) is an important product of reducing graphene oxide (GO) or graphite oxide (GrO) using strong reductants such as hydrazine [21] and NaBH 4 [22], through high-temperature treatment [23], UV-assisted photocatalysis [24], or low-temperature annealing reduction [25,26].GO prepared from graphite by means of strong oxidizers [27] to form graphite oxide (GrO), followed by dispersing the resulting GrO in an appropriate solution for exfoliating into few layers, is well-known as a modified Hummers' method [27].GrO or GO contains myriad oxide functionalities (predominantly alcohols and epoxides) but retains a stacked structure similar to graphite.However, rGO is rather different from GrO or GO in terms of structure.The rGO is exfoliated into monolayers or few-layered stacks.The surface functionality (particularly in basic media) greatly weakens the layerlayer interactions, due to its hydrophilicity [28].rGO has received tremendous attention because of its high electrical conductivity, active surface, large surface area per volume, and excellent electrocatalytic properties [29,30].rGO can act as an excellent performance carrier, while NPs can be highly dispersed on its surface, and the charge transfer at the interface of these hybrid materials can provide a synergistic effect to bring about properties that are different from those of each individual component.Then, Fe 3 O 4 /rGO composites provide a potential application in electrochemistry [9].With the advantages of the magnetism and the conductivity of the Fe 3 O 4 /rGO composites, the nanocomposites could be easily adhered to the electrode surface to achieve the direct redox reactions and electrocatalytic behaviors of analytes adsorbed on the modified surface.In fact, Fe 3 O 4 /rGO was employed to produce FePc@Fe 3 O 4 /reduced graphene oxide nanocomposites as biomimetic catalysts for organic peroxide sensing [31], to modify GCE to design a high-performance electrochemical biosensing platform [32], and to load on the magnetic glassy carbon electrode (MGCE) to form a biosensor to test glucose [9].To the best of our knowledge, there is no report based on using Fe 3 O 4 /rGO modified electrodes for the determination of PRC.
In the present paper, magnetic Fe 3 O 4 /rGO composite was prepared by a facile solvothermal method.The obtained Fe 3 O 4 /rGO was used as an electrode modifier to prepare a Fe 3 O 4 graphene-modified glassy carbon electrode (Fe 3 O 4 /rGO/GCE), which can be used for sensitive detection of PRC in pharmaceutical products with the DP-ASV method.2.2.Methods.X-ray diffraction measurement was recorded on a Bruker D8-Advance X-ray diffractometer (Germany) with CuK  radiation ( = 0.1514 nm).Morphology was observed by means of transmission electron microscopy (TEM) using TEM Jeol JEM-2100F (Japan).FTIR analyses were conducted with a Nicolet Nexus 470 FTIR spectrometer (Thermo Nicolet, USA).X-ray photoelectron spectroscopy (XPS) was studied using a Shimadzu Kratos AXISULTRA DLD spectrometer.Peak fitting was conducted by means of the CASA XPS software.The analysis of iron element was conducted by atomic absorption spectrometry (AAS) using AA6800 Shimadzu (Japan).A CPA-HH5 Computerized Polarography Analyzer (Vietnam) was used for voltammetry experiments.All measurements were performed in the cell with three electrodes: a GCE with a diameter of 2.8 ± 0.1 mm used for formatting the modified electrode as working electrode, an Ag/AgCl/3M KCl as a reference electrode, and a platinum wire as an auxiliary electrode.All measurements were carried out at ambient temperature.

Experimental
The HPLC (high-performance liquid chromatography) method was utilized to analyze PRC for the sake of comparison.Chromatographic determinations were performed in a Shimadzu 2030 HPLC system.The chromatographic conditions were HiQ sil C18 (250 mm × 4.6 mm), detector wavelength 225 nm, (water : methanol : acetic acid = 69 : 28 : 3) and mobile phase at ambient temperature.

Synthesis of GO
, rGO, and Fe 3 O 4 /rGO.Graphite oxide (GrO) was prepared using the Hummers method [27] according to our previous paper [33].For the synthesis of rGO, the as-prepared GrO (0.1 g) was exfoliated by ultrasonication in 100 mL of distilled water for 1 h to prepare an aqueous suspension of GO.Ascorbic acid (0.15 g) was added slowly to the GO suspension and stirred for 8 hours at 50 ∘ C. The product (rGO) was collected by centrifugation and washed several times with ethanol and dried at 80 ∘ C in a vacuum oven for 5 h.
For the synthesis of Fe 3 O 4 /rGO, a stable aqueous suspension of rGO was obtained after the ultrasonication of 0.025 g of rGO in 50 mL of distilled water for 1 h.This suspension was bubbled with nitrogen for 15 minutes and adjusted to pH 11-12 using NH 3 .Then, FeCl 2 ⋅4H 2 O (0.25 g) was added to the rGO suspension and stirred for 16 hours at ambient temperature.The final product (Fe 3 O 4 /rGO) was washed several times with ethanol and distilled water and dried at 80 ∘ C in a vacuum oven for 5 h.

Electrochemical Measurements.
Prior to modification, the GCE was polished with 0.02 m alumina powder to a mirror-like surface followed by sonication for about two minutes in double distilled water and dried at ambient temperature.Fe 3 O 4 /rGO or rGO material was first dispersed by means of ultrasonication in an analytical solvent (1.0 mg/mL) for 1 hour to get a suspension.Then, this suspension was mixed with the Nafion solution (1.25% in ethanol) in the ratio of 1 : 4 by volume.After that, 5 L of the aliquot was cast onto the GCE.The solvent was evaporated using a dryer.After the modification, the electrode was washed with double distilled water and then left to dry in the air before the electrochemical studies were carried out.The modified electrodes were denoted as Fe 3 O 4 /rGO/Naf-GCE or rGO/Naf-GCE.
Cyclic voltammograms (henceforth CVs) were recorded from −0.50 V to +0.90 V (forward potential scan) and then from +0.90 mV to −0.5 V (reverse potential scan) at a scan rate of 0.10 V⋅s -1 .DP-ASV voltammograms were recorded from −0.5 V to +0.9 V at a scan rate of 0.02 V⋅s -1 .DP-ASV voltammograms of the blank solution (a solution of 0.1 M ABS pH 6 and double distilled water) were similarly recorded before each measurement.Panadol Extra (Sanofi-Synthelabo Company, Vietnam), Tiffy Dey (Thai Nakorn Patana Company Ltd., Vietnam), and pms-Mexcold (Imexpharm Corporation in technological cooperation with Pharmascience Inc., Canada) were employed in this study.The ten tablets of each analyzed pharmaceutical preparation were exactly weighed and finely ground in a mortar.An adequate amount of the powders was weighed and transferred to a 100-mL calibrated flask, which was completed to volume with the 0.1 M ABS buffer solution pH 6.The standard addition method was used for determining the pharmaceutical formulations.characteristic diffraction peak at the 2 angle of around 11.3 ∘ of GrO was observed indicating that oxygen-containing functional groups on graphite sheets were formed, that is, the formation of GrO [34].Two broad peaks at around 26 ∘ and 44 ∘ could be attributed to the initial graphite indicating that the oxidation of graphite was incomplete.In fact, a similar result is also obtained in reported papers [35,36].After reduced by ascorbic acid, the material does not show the peak at 11.3 ∘ , and the weak and broad reflection peak at 25.8 ∘ corresponds to the relative short-range order structures in disordered stacked rGO [37], which indicates the successful reduction of GrO.In order to study the morphologies of rGO and Fe 3 O 4 /rGO, TEM images were recorded (Figure 2).The TEM images of rGO show a stacked and crumpled morphology due to the exfoliation and restacking process [37] (Figure 2(a)).It is difficult to obtain monodisperse Fe 3 O 4 particles due to their inherent magnetism.However, the presence of rGO can prevent the aggregation of Fe 3 O 4 nanoparticles.The highly dispersed Fe 3 O 4 particles with small uniform single particles over the rGO sheet with an average particle size of 10-15 nm were observed (Figure 2(b)).The amount of iron analyzed by means of AAS was 36.5% compared to the theoretical amount of 57.5%.This may be due to the fact that iron has not been completely incorporated into rGO.

Results and Discussion
The existence of oxygen-containing functional groups was confirmed by FTIR (Figure 3).Upon the oxidation of graphite to GrO, the strong absorption band at 3414 cm -1 is attributed to ] OH .The vibration bands located at 1724 cm -1 and 1616 cm -1 can be assigned to ] C=O of carbonyl and ] C=C of aromatic rings.The absorption bands at 1226 cm -1 and 1056 cm -1 belong to ] C-O of epoxide and alkoxy, respectively [34,38].The vibrational band at 1400 cm -1 can be assigned to ] COO .The presence of the oxidized groups (Figure 3 indicates the successful oxidation of graphite to graphite oxide.In addition, the adsorption band of oxygen-containing functional groups was not observed (Figure 3(b)) indicating the reduction of GrO to form rGO. Furthermore, a new absorption band observed at 1558 cm -1 (Figure 3(c)) may belong to the skeletal vibration of graphene sheets [34].The successful introduction of magnetic oxides onto rGO is confirmed by the sharp peak around 580 cm -1 assigned to Fe-O of Fe 3 O 4 [39,40].
The chemical state of the elements of Fe 3 O 4 /rGO is further confirmed by XPS.The XPS spectrum (Figure 4(a)) shows photoelectron lines implying the presence of three elements at a binding energy of 284, 530, and 720 eV corresponding to carbon (C 1s ), oxygen (O 1s ), and iron (Fe 2p), respectively.The high-resolution XPS for C 1s of rGO is shown in Figure 4(b).The deconvoluted C 1s spectrum implies that rGO consists of functional groups such as sp 3 (C-C, 282.2 eV), hydroxyl and epoxy (C=O, 284.2 eV), and carbonyl (C=O, 286.03 eV) [35].The calculation of the atomic contributions shows the number of C-O and C=O groups on Fe 3 O 4 /rGO to be 43.8% and 16.0%, respectively, while maintaining the amount of C-C groups at 40.2%.The percentage of oxygen contained in the functional groups is rather high compared with that of other reports [41,42], indicating that rGO was only partially reduced.Since XRD patterns of magnetite Fe 3 O 4 and maghemite Fe 2 O 3 are very similar [38], XPS Fe2p core level spectrum has been conducted to confirm magnetite (Figure 4(c)).A peak at 726.6 eV is attributed to 2p1/2 (Fe(III)) and the satellite peaks at 719.35 eV and 733.5 eV are characteristic for maghemite.The broad peaks of 2p3/2 Fe(II) at 713.1 eV and 2p1/2 Fe(III) at 724.75 eV are typical for magnetite [35].Although nitrogen was bubbled during the synthesis process, a part of Fe(II) was oxidized by available oxidant agents (diluted oxygen, organic species in the oxidation form).Then, the mixture of Fe 3 O 4 and Fe 2 O 3 was obtained.The formation of Fe 3 O 4 /rGO and Fe 2 O 3 /rGO may be illustrated according to the following reactions: The magnetic properties of Fe 3 O 4 /rGO were determined using VSM measurements at ambient temperature (Figure 5).The magnetization hysteresis loop has an S-like shape, and the saturation magnetization is 34 emu⋅g −1 .This value is lower than that of the pure nanomagnetic Fe 3 O 4 [43] but compatible with that of Fe 3 O 4 /rGO reported in [38,44,45].This low saturation magnetization of Fe 3 O 4 /rGO may be due to the relatively lower density of magnetic components in the Fe 3 O 4 /rGO nanocomposites.The magnetic coercivity was nearly zero indicating that there is no remaining magnetization upon the removal of the external magnetic field.Therefore, the superparamagnetic behavior of nanocomposite Fe 3 O 4 /rGO was confirmed.
The porous properties of GrO, rGO, and Fe 3 O 4 /rGO were investigated using the nitrogen adsorption-desorption isotherms (Figure 6).The isotherm curves belong to typical type IV according to IUPAC classification.The presence of the hysteresis loop at a high relative pressure region indicates the mesopore which was attributed to the void /rGO is compatible with that of some reports [38,44] but higher than that of other reports [47,48].
The hydrophobic polyaromatic sheets of unoxidized benzene rings of rGO exhibit hydrophobic interaction and - stacks towards the organic molecules [49], while binding properties of iron oxides provide high reactivity.The Fe 3 O 4 /rGO composite with the combination of these properties is expected to be a potential electrode modifier for the determination of the organic compounds.7(a).As ethanol was used to disperse Fe 3 O 4 /rGO, the intensity of  ap (anodic peak current) is 2.2 and 1.2 times higher than those in water and DMF, respectively.This indicates that ethanol is favorable for dispersing Fe 3 O 4 /rGO because it provides the highest  ap with the lowest RSD ip (relative standard deviation) equal to 0.9.Therefore, ethanol was selected as a dispersion solvent for further experiments.

Electrochemical Performance of Fe
The effect of pH on the voltammetric response of PRC was studied using CVs in the range of pH 4.8 to 9.8.The purpose of this study is to evaluate the ratio of electrons and protons participating in the voltammetric oxidation of PRC and also the optimum pH.The B-RBS buffer was used to adjust pH.Since pK a of PRC in solution is 9.5 [18,50], the form of PRC exists as negative species in the solution above 9.5 and positive species below this pH.On the other hand, pH pzc (point of zero charges) of Fe 3 O 4 /rGO is approximately 6.3 [33].Therefore, the Fe 3 O 4 /rGO surface will be charged positively in the solution below pH pzc and negatively above pH pzc .The dependence of  ap on pH can be explained as follows: at pH lower than pH pzc , the peak current signal is low due to the positively charged surface repulsing the positive species of PRC.With the increase of pH, the positive charge of the surface decreases, and the electrostatic attraction results in a higher PRC stripping current.At pH higher than pH pzc , the peak current signal is also low due to the negatively charged surface repulsing the negative species of PRC.The signal intensity is maximal at pH 6.Therefore, pH 6 was selected as the optimal pH.The relationship between pH and the anodic peak potential,  ap , is shown in the inset of Figure 7(b).As can be seen, the anodic peak shifts to more negative potentials with increasing pH from 4.8 to 9.8 indicating that protons associate directly with PRC oxidation.The linear regression equation can be expressed as  ap = 0.659 − 0.053 ⋅ pH;  = −0.997; < 0.001, (2) where  ap is the anodic peak potential,  is the correlation coefficient, and  is the significant value.
The slope of regression is (−0.053V/pH) in excellent agreement with the potential shift predicted by the Nernst equation of −0.059 V/pH at 298 K.This indicates that the number of electrons and protons transferred in the redox reaction of PRC are equal.
Several buffers of pH 6 including ABS, CBS, PBS, UBS, and B-RBS were studied in order to evaluate the relevant buffer for the voltammetric response of PRC. Figure 7(c) shows the cyclic voltammograms recorded with different buffers.As can be seen from the figure, the ABS buffer with the lowest RSD (0.1%) provides the highest peak of the current signal.Then, ABS was selected as the buffer for further experiments.
Introducing Fe 3 O 4 /rGO on the surface of GCE considerably improved the electrochemical behaviors of PRC indicating that Fe 3 O 4 /rGO could play a critical role in the voltammetric response of PRC.Different amounts of Fe 3 O 4 /rGO suspension were investigated.The results showed that the peak current of PRC increased with increasing the volume of Fe 3 O 4 /rGO suspension introduced on the surface of GCE up to 5 L (1 g⋅mL -1 ) (Figure 7(d)).It can be explained that as the Fe 3 O 4 /rGO amount increased, the accumulation efficiency of PRC on the modified GCE also increased, resulting in the peak current enhancement.However, a further increase of the Fe 3 O 4 /rGO amount caused a decrease of the anodic peak current of PRC.This can arise from the larger film thickness causing the increase of resistance of the modifier film against the electron transfer for PRC.As a result,  = 5 L of Fe 3 O 4 /rGO suspension was chosen as an optimal amount.

Electrochemical Behavior of PRC on Modified Electrodes.
The electrochemical experiments on the GCE modified with and without rGO or Fe 3 O 4 /rGO were conducted by CVs in order to confirm its electrochemical behavior of the modified GCE for the detection of PRC (Figure 8).The Nafion was introduced to improve the adhesion of Fe 3 O 4 /rGO and conductivity of the modified electrode.As can be seen from Figure 8, the current response on the bare GCE exhibited the broad peaks.After modifying the GCE by Nafion, rGO, or Fe 3 O 4 /rGO, the current responses on the modified electrodes provided the defined peaks, but that on the Fe 3 O 4 /rGO/Naf-GCE exhibited the well-defined stripping signal with the highest intensity.
The intensities of  ap (anodic peak current) and  cp (cathodic peak current) on Fe 3 O 4 /rGO/Naf-GCE are 1.7-and 1-fold, respectively, compared with that on the rGO/Naf-GCE.PRC exhibits a pair of well-defined redox waves on the Fe 3 O 4 /rGO modified GCE with  ap = 0.37 V and  cp = 0.25 V.The favorable signal-promoting effect of Fe 3 O 4 /rGO indicated that it could enhance the electron transfer rate of PRC and had a good electrocatalytic activity for the redox reaction of PRC.The favorable formed - interaction between the PRC molecules and the rGO sheets of sp 2bonded carbon atoms with strengthening adsorption [51] were possibly responsible for the accumulation of PRC.In addition, the coordination of the nitrogen atoms in PRC with the Fe(II) and Fe(III) ions in Fe 3 O 4 /rGO also contributed to the attracting of the analytes to the electrode surface.The combinations of these effects led to a greater amount of PRC accumulation on the surface of Fe 3 O 4 /rGO/Naf-GCE, greatly improving the PRC voltammetric signal.

Factors Affecting the Stripping Voltammetric Signals
(1) Effect of Scan Rate.Important information about electrochemical mechanism can usually be obtained from the relationship between the voltammetric responses and the scan rate.Therefore, the effects of scan rate on the peak potential (anodic peak potential ( ap ), cathodic peak potential ( cp )) and peak current ( ap and  cp ) were investigated using the CV method (Figure 9(a)).As seen in Figure 9(a),  p increased with an increase in the scan rate.Such dependence of  p on the scan rate indicated that the homogeneous electron transfer in PRC oxidation is irreversible [52].Peak current ( ap and  cp ) in Figure 9(b) increased with an increase in the scan rate within 0.2-0.5 V⋅s -1 suggesting that the electron transfer reaction is involved with a surface-confined process [53].
On the other hand, the plot of the linear regression line for ln  p and ln V can provide the information about diffusion-or adsorption-controlled process.The slope near 1 is attributed to the adsorption-controlled electrode process and that near 0.5 is said to be a diffusion-controlled process [14].Both linear relationships with high relative coefficients ( ≤ 0.001) were obtained (Figure 9(c)) and expressed by the following equations: ln  ap = 1.70 + 0.56 ⋅ ln (V) ;  = 0.995,  < 0.001, ln  cp = 0.36 + 0.60 ⋅ ln (V) ;  = 0.987,  < 0.001.
Its slopes of 0.56 and 0.60 are close to 0.5, again asserting that the redox process of PRC on the modified electrode was controlled by the diffusion.
The electron transfer coefficient () and the electron transfer rate constant (  ) were calculated based on the Laviron equation [54].
From ( 5), ( 6), (7), and (8), the electron transfer coefficient and electron transfer rate constant can be calculated as 0.595 and 0.315 s -1 , respectively.On the other hand, combining ( 5) and ( 7), the value of (1-) ⋅  can be calculated as 0.93.Therefore, the average value of the number of electron transferred () was 2.3.Consequently, assuming  = 2 for PRC, the oxidation mechanisms for PRC involve two electrons and two protons and are likely to be in agreement with those reported in other papers [55,56].The oxidation of PRC leads to a quinine type moiety as proposed earlier [57] (Scheme 1).
An approximate evaluation of the amount of adsorbed PRC on modified GCE (surface coverage of the electrode) can be obtained by using the method proposed by Sharp et al. [58].The  ap is a function of the surface concentration of electroactive species and calculated according to the following equation: where  is the number of electrons involved in the reaction,  is the surface geometrical area (0.062 cm 2 ), and Γ (mol⋅cm −2 ) is the surface coverage.From the slope of the anodic peak currents ( ap ) versus the scan rate, the surface coverage was calculated as 4.1 × 10 -6 mol⋅cm -2 for PRC on the surface of Fe 3 O 4 /rGO/Naf-GCE.
The effective surface area of the modified electrode could be obtained based on the current-potential characteristics.According to Bard and Faulkner's equation [52,59], the relation between current and potential of the oxidation irreversible process is described by the following equation: where  ap is peak current (A),  p is the peak potential (V);  is an active surface area (cm 2 );  is the bulk concentration of PRC (3 × 10 -7 mol -1 ⋅cm 3 ); and other parameters were defined above.
Taking natural logarithm of (10), one obtained the dependence below: The plot of linear regression between ln  p and  p was established (Figure 9(e)).The active surface area of the modified electrode obtained from the intercept of the linear regression line was 5.6 cm 2 .The geometry surface area of electrode is 0.065 cm 2 (diameter of electrode = 2.8 mm).The effective surface area was 86 times larger compared with the geometry surface area of the electrode.The increase of surface of the modified electrode is due to the presence of Fe 3 O 4 /rGO.The larger effective surface area results in more active sites and leads to a higher signal to noise ratio.
(2) Effect of Pulse Amplitude (Δ), Accumulation Potential, and Time.Pulse amplitudes significantly affect the voltammetric stripping signal of the analytes.The pulse amplitude in the range of 0.04 V to 0.20 V was studied using the DP-ASV method.The results showed that  p increased linearly and  p shifted negatively as pulse amplitudes increased (Figure 10(a)).As Δ was larger than 0.1 V, the peak width of the current peak tends to be wide, reducing the peak resolution and hence the selectivity.The pulse amplitude of 0.10 V that provided a symmetric peak with low RSD (1.8%) was selected for the next experiments (inset of Figure 10(a)).
The influence of accumulation potential and time on the  p of PRC at Fe 3 O 4 /rGO was also investigated using the DP-ASV method in the range of −0.1 V to 0.5 V (Figures 10(b) and 10(c)).The  ap increased as the potential shifted positively from −0.1 V to 0 mV and reached a maximum at 0 V; then it decreased as the potentials shifted positively from 0 V to 0.5 V.A potential of 0.0 V was selected as the optimal accumulation potential.The accumulation time also had an effect on the peak current.The  p increased slightly with time, peaked at 75 s, and then decreased afterwards.Therefore, 75 s was used as the accumulation time (Figure 10(c)).
(3) Effect of Interferents.The effect of possible interferents was investigated by adding some compounds to a solution containing 1 × 10 -4 mol⋅L -1 PRC in the 0.1 M ABS (pH = 6).Ascorbic acid (AA), uric acid (UA), and caffeine (CF) are commonly present in pharmaceutical samples, and they were tested as interferents at different molar ratios of the referent to PRC (0.5/1; 1/1 and 2/1 for AA and UA, and 5/1; 10/1 and 20/1 for CF). p in the presence of the interferents was recorded and compared with  p without interferent.The relative error (Re) did not exceed 5%, indicating that the Fe 3 O 4 /rGO/Naf-GCE can be employed for the determination of PRC in the presence of AA, UA, and CF (Table 1).
The dependence of the anodic stripping current ( ap ) for PRC on its concentration was conducted using DP-ASV (Figure 11(a)).The ( ap ) of PRC versus its concentrations exhibited a nonlinear response in the concentration range of 2-150 M (the insert of Figure 11(a)).From the plot of  p versus added concentrations of PRC, two ranges were obtained (Figure 11(b)).The first linear range between 2 M and 10 M and the second range between 10 M and 150 M were described in the following equations: The break in the calibration curve of PRC probably reflects the formation of a submonolayer in the first range of calibration and the formation of a monolayer in the second range as proposed by Kachoosangi et al. [18].The limit of detection (LOD) calculated in the first range of the PRC concentrations (2-10 M) was 0.72 × 10 −6 M. The obtained LOD in this work was comparable with that in other references, in which the electrode was modified with dipyrromethene-Cu(II) monolayers, PAY/nano-TiO 2 ,  Cu(II)-conducting polymer complex, single-walled carbon nanotubes, and multiwalled carbon nanotube [17,18,61] (Table 2).It could be noticed that LOD of PRC from the proposed method was lower than or comparable with those results based on modified electrodes in previous papers.The Fe 3 O 4 /rGO modified electrode performed better than some of the electrodes based on PAY/nano-TiO 2 , Cu(II)conducting polymer complex, and so on but failed to some others.Overall, the Fe 3 O 4 /rGO was proved to be an effective electrode modifier for the detection of PRC.

Determination of PRC in Real
Samples.The proposed method was used to determine three kinds of PRC commercial tablets, namely, Panadol Extra (Sanofi-Synthelabo Company, Vietnam), Tiffy Dey (Thai Nakorn Patana Company Ltd, Vietnam), and pms-Mexcold (Imexpharm Corporation in technological cooperation with Pharmascience Inc., Canada).The tablets were ground to powder and dissolved in distilled water.The PRC concentration was analyzed using the DP-ASV method.The recoveries of the tests performed using DP-ASV were in the range from 95.5% to 107.0% for all the three samples.This result suggested that the determination of PRC applying the DP-ASV with Fe 3 O 4 /rGO/Naf-GCE possessed an acceptable error [62].Table 3 also presents the values of the amounts of PRC determined using the HPLC method for the sake of comparison.The paired samples -test was applied to estimate the data statistically.
With the significant level of  = 0.05, the amounts of PRC analyzed with the proposed method are in good agreement with the content of PRC provided in the labels ( (2) = −1.066, = 0.395 > 0.05).No significant differences were observed between the values determined with the proposed method and the HPLC method for the amount of PRC in the tablets ( (2) = −0.851, = 0.851 > 0.05).These results indicate that the electrode developed in this work is relevant for detecting PRC in commercial tablets.

Conclusions
Fe 3 O 4 /rGO was synthesized using a facile one-step process.
The magnetic iron oxide with very fine spherical particles in nanoscales was highly dispersed on the rGO sheets.Fe 3 O 4 /rGO possessed superparamagnetic properties at room temperature and the saturation magnetization approaches 34 emu⋅g -1 .The Fe 3 O 4 /rGO-based electrode has been developed, and it exhibits an excellent electrocatalytic activity towards the reduction and oxidation of PRC.Fe 3 O 4 /rGObased GCE promotes the sensitivity of the determination of PRC with a low detection limit (0.72 × 10 -6 M).The analytical response was unaffected by the presence of caffeine, ascorbic acid, and uric acid commonly found in commercially available pharmaceutical tables.We also demonstrated the application of Fe 3 O 4 /rGO for the determination of PRC in pharmaceutical preparations with satisfactory results.
N 4 .pH 6 of buffers was adjusted with 1 M KOH or 1 M HCl.Britton-Robinson buffer solutions (B-RBS) in the range of pH from 4.8 to 9.8 were prepared from 0.5 M H 3 BO 3 , 0.5 M H 3 PO 4 , and 0.5 M CH 3 COOH.The desired pH 6 of buffer was adjusted with 1 M KOH or 1 M HCl.

3. 1 .
Characterization of Materials.XRD measurements were used to study the phase of the obtained samples.

Figure 7 (
b) shows the CVs recorded at different pH for the 2.10 -4 M PRC solution.The anodic peak current increases with increasing pH from 4.8 to 6.A further increase in the pH leads to a decline of the current (inset of Figure 7(b)).

Figure 9 :Scheme 1 :
Figure 9: (a) CVs for 3 ⋅ 10 -4 M PRC on Fe 3 O 4 /rGO/Naf-GCE in 0.1 M ABS pH 6 in scan rates (V) from 0.02 V⋅s −1 to 0.5 V⋅s −1 ; (b) dependence of anodic peak current ( ap ) and cathodic peak current ( cp ) on V 1/2 ; (c) plot of ln  ap and ln  cp against ln V; (d) linear regression between anodic peak potential ( ap ) and cathode peak potential ( cp ) versus ln V; (e) plot of ln  p versus of  p .

Figure 11 :
Figure 11: (a) DP-ASV curves of PRC on its concentration in the range of 2 M to 150 M in 0.1 M ABS buffer pH 6; the inset presents the plot of  ap versus  PRC ; (b) double-linear regression of  ap versus concentrations; the inset shows the linear regression line of the first segment ( acc = 75 s, scan rate = 0.01 V⋅s -1 ,  acc = 0 V, Δ = 0.1 V).

Table 2 :
Comparison of the Fe 3 O 4 /rGO modified GCE with the reported methods for the determination of paracetamol.

Table 3 :
Results obtained in the determination of paracetamol in pharmaceutical formulations using the DP-ASV and HPLC.Average of 3 measurements.Re(%) = 100 × (DP-ASV value − HPLC value)/HPLC value. *