Differential Electroanalysis of Dopamine in the Presence of a Large Excess of Ascorbic Acid at a Nickel Oxide Nanoparticle-Modified Glassy Carbon Electrode

Electrochemical determination of dopamine (DA) in the presence of a large excess of ascorbic acid (AA) in their coexistence at a nickel oxide nanoparticle-modi ﬁ ed preoxidized glassy carbon electrode (GC ox /nano-NiO x ) is achieved. The GC ox /nano-NiO x electrode is prepared by electrodeposition of nickel nanoparticles (nano-Ni) onto an electrochemically activated glassy carbon (GC) electrode, and the thus prepared nano-Ni were subjected to electrochemical oxidation in alkaline medium for the formation of nickel oxide (NiO x ). Modi ﬁ ed electrodes were electrochemically and morphologically characterized. The e ﬀ ect of loading level of nickel was investigated by changing the number of potential cycles for the deposition of nano-Ni, i.e., 1, 2, 5, and 10 potential cycles, in the potential range from 0 to -1.0 V vs. SCE. Also, the experimental and instrumental parameters were optimized. Experimental results showed that the modi ﬁ ed electrode di ﬀ erentiates well the oxidation peaks of DA and AA enabling the electrochemical determination of DA in the presence of a large excess of AA. Remarkably, it is found that the oxidation current of DA is 2 times larger than that of AA even the concentration of AA is about 5 times larger than that of DA. The LOD and LOQ of DA were calculated and were found to equal 0.69 and 2.3mM, respectively. This o ﬀ ers the advantage of simple and selective detection of DA free of the interference of AA in real samples.


Introduction
Dopamine (DA) is critical for medical treatment and clinical analysis. DA has long been of interest to neuroscientists and chemists because of its role as an essential neurotransmitter in the functionality of the central nervous, renal, hormonal, and cardiovascular systems [1]. Low level of DA in the mean central nervous system is the main cause of many neurological diseases (e.g., schizophrenia and Parkinson's disease) [2,3]. A major problem for the detection of DA is the coexistence of many interfering compounds in biological systems. The most critical interferent of DA (typical concentration in vivo is 10 -9 -10 -5 mol L -1 )i sA A( t y p i c a lc o n c e n t r a t i o ni nv i v oi s1 0 -4 mol L -1 ), which is a cofactor of multiple enzymes [4,5]. AA is a vital vita-mininthedietofhumans,andithasbeenusedfortheprevention and treatment of common cold, mental illness, and other diseases [6]. To quantify DA coexisting with a large excess of AA, the major challenge is to achieve both high selectivity and sensitivity of the detection technique.
Many analytical techniques have been reported for the determination of DA and AA such as chromatographic methods, mass spectroscopy [7], spectrophotometry [8], and chemiluminescence [9]. However, these methods encounter some drawbacks such as long analysis time, high cost, and the requirement for sample modification. These disadvantages probably make them unsuitable for routine analysis. Electrochemical methods offer several merits such as simplicity, rapidity, high sensitivity and specificity, and ease of fabrication [10].
The simultaneous electrochemical detection of AA and DA faced several challenges as their peaks overlap due to their close redox potentials [11,12]. AA is usually present in vivo at concentrations 100 to 1000 times larger than those of DA, resulting in an overlap of electrochemical responses and electrode fouling by oxidation products [10]. Thus, it is usually difficult to separate the responses of AA and DA at bare electrodes [13] such as GC, gold, and platinum electrodes. In recent years, several strategies to separate the electrochemical response of the two species while retaining the sensitivities have been introduced [14][15][16][17][18][19][20][21][22][23]. It includes the modification of gold and GC electrodes with either an organic or inorganic modifier. Modification with an organic modifier relies on the possible differentiation of the two redox peaks via controlling the electrode surface. For instance, controlling the surface charge could separate the two peaks based on the large difference in the dissociation constants of the two species; DA is positively charged (pK a =8:9) whereas AA is negatively charged (pK a =4:2) [14,15] at a physiological pH. In this correspondence, Nafion- [16,24,25], clay- [26], and polymeric film- [27][28][29] modified electrodes have been reported to remove the interference of AA with DA [30]. However, the reported selectivity enhancement is not accompanied by a sensitivity improvement of DA detection [31,32]. Accordingly, another strategy was developed by modification of an electrode with an electroactive material exhibiting preferential catalytic activity towards DA, such as ruthenium oxide pyrochlore [33], nickel hexacyanoferrate [34], carbon nanotube films (CNTs) [31] and metal nanoparticles [33][34][35].
On the other hand, metal nanoparticles, compared with their counterpart, are excellent electrode materials for potential applications in electroanalysis due to their unique structure and good electrical conductivity [36]. Several nanoparticles (NPs) have been used in this correspondence such as gold [37], silver [38], and platinum [39] as well as metal oxides such as ZnO [40], MnO 2 [41], CuO [42] and NiO [43].
The purpose of this work is to investigate the electroanalysis of DA in the presence of overly abundant AA at the GC electrode modified with nano-NiO x by cyclic and square wave voltammetries. Experimental conditions were optimized for achieving the largest peak separation for the oxidation potential of DA and AA along with the largest possible sensitivity. It has been reported that nickel oxide electrocatalyzes the oxidation of hydroxyl compounds [44]. Thus, it is expected that nano-NiO x will help in the differentiation of responses of DA and AA (hydroxyl compounds). Moreover, oxidized glassy carbon (GC ox ) has been reported to separate electrochemical responses of both species based on the generation of various groups. Some of these groups are anionic groups, and thus, it can repel anionic analytes, as AA is under pH of 7, and attract cationic analytes, as DA is under pH of 7 [45]. Thus, the deposition of nano-NiO x at GC ox is promising for the selective analysis of the two species in their coexistence.

Materials and Reagents.
All chemicals were used as received without further purification. AA and DA were pur-chased from Sigma Chemical Co. (USA). Stock solutions of DA and AA were prepared prior to experiments by dissolving a suitable amount of reagents in deionized water. Appropriate concentrations of analytes are prepared by dilution.

Apparatus and Electrodes.
Electrochemical measurements have been performed on Gamry Instruments (potentiostat/galvanostat/ZRA model Reference 600™). Voltammetric experiments were realized using a conventional electrochemical cell with three electrodes, where the modified glassy carbon electrode (GC ox /nano-NiO x ) was used as the working electrode and the saturated calomel electrode (SCE) as the reference electrode. The auxiliary electrode was a platinum wire. The GC electrode (1 mm in diameter) was polished to get a mirror shine using polishing papers of two grades (3000 and 2000) and then cleaned thoroughly with deionized water.
2.3. Fabrication of the GC ox /Nano-NiO x Electrode. GC was oxidized in 0.5 M H 2 SO 4 by potential cycling from -0.2 to 2 V at a scan rate of 100 mV s -1 . This oxidized GC was denoted as GC ox throughout. GC modification with NiO x was achieved as follows: nickel was electrodeposited from a Watts bath (0:02 M NiSO 4 ·6H 2 O+ 0:03 M NiCl 2 ·6H 2 O+ 0:03 M H 3 BO 3 ) of pH 5.8 [46] by cycling the potential in the range from 0 to -1.0 V vs. SCE at a scan rate of 50 mV s -1 and at 25°C. Then, the thus modified electrode was subjected to several potential cycles in 0.5 M NaOH for oxidizing the deposited nickel (Scheme 1).

Instrumentation.
Morphological characterization of the bare and modified GC electrodes was investigated by scanning electron microscopy (SEM, JSM-7600F Field Emissions Scanning Electron Microscope). Chemical composition was probed by Energy-Dispersive X-ray (EDX, Inca Oxford Instrument), and crystallographic orientation was examined using X-ray diffraction (XRD, D8 Discover Bruker using 2-theta from 5 to 90 degrees with 2 degrees per minute scan speed).

Results and Discussion
3.1. Surface Characterization. Typical EDX, SEM, and XRD micrographs obtained on the surface of the bare GC, GC ox , and GC ox /nano-NiO x electrodes are presented in Figures 1-3, respectively. The elemental analysis of the GC ox /NiO x electrode is probed by EDX ( Figure 1). The extracted data from this figure are shown in Table 1. The GC ox (Figure 1(b)) exhibits a significant increase in the oxygen signal compared with the bare GC (Figure 1(a)). This observation is clear evidence for the oxidation of GC as previously reported [47][48][49] after being subjected to electrooxidation. In Figures 1(a) and 1(d), the nickel signal appeared as evidence for the deposition of Ni on GC ox . Also, in Figure 1 3 Journal of Sensors formation of a granular overlayer of the deposited Ni nanoparticles by small density and size and diverse geometric structure. Upon oxidation of Ni, the nano-NiO x electrodeposited on GC ox becomes denser and larger, as shown in (Figure 2(d)).
X-ray diffraction (XRD) is a powerful nondestructive technique for characterizing crystalline materials. The XRD patterns recorded from the bare GC, GC ox ,G C ox /Ni, and GC ox /NiO x are shown in Figure 3. The sharp peak located at around 19°corresponds to the (002) diffraction of the carbon substrate [50]. In pattern d, diffraction peaks at 37.77°, 43.43°, 63.51°, 76.24°, and 79.45°correspond to the (1 1 1), (2 0 0), (2 2 0), (3 1 1), and (2 2 2) planes, respectively, and characteristics of cubic crystalline NiO x [51,52] are   Journal of Sensors shown indicating the successful fabrication of nano-NiO xmodified electrode. Figure 4 shows a series of cyclic voltammograms obtained from bare GC (curve 1), GC ox (curve 2), GC ox /nano-Ni (curve 3), and GC ox /nano-NiO x (curve 4) electrodes in 0.1 M NaOH solution at a scan rate of 100 mV s -1 . The cyclic voltammograms at bare GC (curve 1) show no redox peak current in the potential region examined, but the GC ox electrode (curve 2) shows an increase in the current due to the fact that the surface area of the GC electrode increases as a result of oxidation, consistent with the SEM shown in Figure 2(b). In curve 3 obtained from GC ox /nano-Ni, the Ni/NiO x redox couple is revealed; the redox couple is represented by the following equation:

Cyclic Voltammetric Studies.
Oxidation peak at 0.55 V corresponds to the oxidation of nickel to Ni 2+ , and subsequently Ni(OH) 2 is oxidized to NiOOH, as represented in Eq. 2 [53] Ni OH However, in this case, the response is obscured by the large background current of GC ox . In curve 4, the redox response obtained after oxidation in NaOH is enhanced; the oxidation peak is negatively shifted and revealed at 0.47 V, along with the increase in oxygen evolution, which started at 0.6 V [53]. The oxidation peak is coupled with a redox one appearing at ca. 0.3 V. The electroactive areas of GC ox /Ni and GC ox /NiO x electrodes were estimated from the charge consumed in the oxidation peak taking into consideration the fact that the formation of one monolayer of Ni(OH) 2 consumes 514 μCcm 2 [54]. The electroactive areas of GC ox /nano-Ni and GC ox /nano-NiO x electrodes were estimated to be 2.1 and 1.6 cm 2 , respectively.
The modified electrodes were examined for the electrochemical behaviour of DA and AA, and data are shown in Figure 5, i.e., at bare GC (1, 2), GC ox (3), GC ox /nano-Ni (4) and GC ox /nano-NiO x (5) electrodes in PBs (pH 6.0) (curve 1) containing 0:25 mM DA + 1:0mMAA (2)(3)(4)(5). At bare GC, the electrochemical responses of DA and AA are overlapping, as well as at GC ox , albeit the response to DA is enhanced and the one of AA is diminished. At GC ox /nano-Ni (curve 4) and GC ox /nano-NiO x (curve 5), the oxidation peaks of DA and AA are differentiated with the larger response obtained at the latter. It has been reported that NiO x rather than metallic nickel is the appropriate electrocatalyst for the oxidation of hydroxyl compounds [44].
The effect of loading of nano-NiO x on the electrochemical response of DA and AA is shown in Figure 6 in which CVs obtained from GC ox /nano-NiO x electrodes, prepared by different potential cycles, in PBs (pH = 6:0) containing 0:25 mM DA + 1:0mMAA are shown. The effect on both separation of the peak potential of DA and AA and the sensitivity is not that large. The largest separation of peak potentials and the large current response for DA along with a smaller response for AA are obtained from GC ox /nano-NiO x prepared by two potential cycles. This helps in the simultaneous analysis of the two species, knowing that AA is always in larger concentration than DA in real samples [4,5]. Hence, the loading prepared by two potential cycles will be used hereafter as the optimum one.
The effect of pH on the electrochemical response of DA and AA at the GC ox /nano-NiO x electrode in phosphate buffer solutions of different pH at a scan rate of 100 mV s -1 was examined. Oxidation of DA (Equations (3)-(5)) and AA (Equation (6)) involves the release of two protons; thus, it is expected to be dependent on the pH of the electrolyte. According to literature [55], it is proposed that dopamine oxidation leads to the formation of dopamine quinone (B), which subsequently undergoes an intramolecular cyclization to form leucodopaminechrome (C). The leucodopaminechrome is further oxidized to form dopaminechrome (D).
The anodic peak potentials were strongly dependent on pH (Figure 7), suggesting the participation of H + in the  Figure 7, the effect of pH is significant on both the peak potential and peak current of DA. Regarding AA, the significant effect is on its peak potential but not peak current. The optimum pH shows the largest separation of peak potential of DA, and the current response is obtained at pH 6 (curve 3). Therefore, PBs of pH 6.0 were selected for all subsequent electrochemical studies of DA and AA. At pH larger than 6, the two peaks of DA and AA oxidation overlap. The linear regression equations are given by The slopes of the linear variations of peak potentials with pH were -51 mV/pH and -41 mV/pH for DA and AA, respectively. The correlation coefficient in the case of AA is smaller than one, probably because in the case of AA (pKa = 4:2)a t pH larger than 4, the oxidation is little bit retarded [56]; i.e., cyclic voltammetric curves in Figure 7 are for AA while it is protonated and nonprotonated. Also, it has been reported that the response to the electrode fouling occurs due to oxidation products of DA [10]. In the case of DA (pKa1 = 8:9 and pKa 2=10:6), it is nonprotonated in the studied pH range [14,15,57], so the electrochemical behaviour shows ideal pH dependence, and the correlation coefficient in this case equals almost one. Negative slopes showed that deprotonation was involved in the oxidation process, which was facilitated at   Journal of Sensors higher pH values. Near Nernstian slopes indicated the equal numbers of electrons and protons in the charge-transfer step at the surface of the modified electrode [58].

Electroanalysis of DA and AA
(1) Selectivity. The selectivity of the GC ox /nano-NiO x electrode for the determination of DA and AA was examined by measur-ing the CVs of 1 mM AA in the presence of different concentrations of DA. As it can be seen in Figure 8, the oxidation peak current of AA is constant irrespective of the concentration of DA. Thus, it can be utilized in selective determination of DA in the presence of AA under the present conditions.
(2) Effect of Scan Rate. The effect of scan rate on the peak currents of DA and AA oxidation is shown in Figure 9.    (1), two (2), five (3) and ten (4) potential cycles. Scan rate: 100 mV s -1 .

Journal of Sensors
Relations between the scan rate and the peak current for DA and AA are shown in Figure 10.
A good linear relationship between the peak current of AA and square root of scan rate (Figure 10(a)) indicates the diffusion process. For DA, the relation (i pa − ν 1/2 ) is not linear. The relation (i pa − ν) for DA (data are not shown) gives a linear relation indicating a surfaceconfined process.
Also, the linear response of the anodic peak current versus scan rate (log i pa − log ν plot) is illustrated in Figure 10(b) in which the slope of log i pa versus log ν is calculated to be 0.83 (R 2 =0:992) for DA and 0.52 (R 2 =0:997) for AA. A slope value close to 0.5 indicates a diffusion-controlled reaction, and a value equals 1 indicates an adsorption reaction. A value between 0.5 and 1.0 indicates a mixing diffusion and adsorption reaction [59], but diffusion control is more dominant. In the present case, the slope of DA 0.83 is very close to the theoretical value of 1 for adsorptioncontrolled processes. Also, the slope of AA 0.52 indicates that oxidation of AA is a diffusion-controlled process.

Square Wave Voltammetry.
A square wave voltammetric (SWV) method is utilized for the selective determination of DA and AA. SWV offers excellent discrimination against double-layer charging current, and accordingly it has high sensitivity [60]. Under the optimum conditions of instrumental parameters: step potential (2 mV), square wave   Figure 11. The oxidation peaks of DA and AA are clearly observed at about 0.2 and 0.02 V, respectively. The linear calibration curves for both species were obtained in PBs (pH 6.0) over a wide concentration range and shown as an inset. Analytical parameters for simultaneous determination of DA and AA at the GC ox /nano-NiO x electrode are summarized in Table 2. As can be seen, the sensitivity (slope) of DA is very large compared with that of AA, even the concentration of AA is larger. The high sensitivity is due to the preconcentration of the DA cations at the electrode. The linear relation with correlation coefficients around 0.980 and 0.987 for DA and AA, respectively, and a limit of detection of 0.69 μM for DA and 18 μM for AA are obtained. The LOD and LOQ were calculated from the calibration curve using the equations LOD = 3 S/N and LOQ = 10 S/N, where S is the standard deviation and N is the slope of the regression line. The comparison of this method with other reported electrochemical methods is given in Table 3. It can be seen that the GC ox /NiO x exhibited a wider linear range with lower detection limit in comparison with other relevant reported results. 9 Journal of Sensors in the oxidation peak potentials for DA and AA at the GC ox /nano-NiO x electrode is large enough for the separation and simultaneous determination of DA in the presence of high levels of AA in a mixture. The SWV response of DA and AA in their coexistence is shown in Figure 12. The oxidation peak currents of these species increase with their concentrations, indicating the stable and efficient electrocatalytic activity of the modified electrode. Calibra-tion plots extracted from this figure are shown as an inset. For both species, a calibration curve with two slopes was obtained. The correlation coefficient values in all cases are larger than 0.97.

Study of Interference.
The influences of some possible species on the determination of DA and AA at the GC ox /nano-NiO x -modified electrode are examined. To investigate the selectivity of the proposed method, various interfering chemicals including amino acids (glycine and tryptophan), urea, oxalate, and catechol were added to the sample solution containing 0.25 mM DA and 1 mM AA at optimum conditions. As can be seen in Figure 13, i pa in the absence (reference) of interferences and in the presence of glycine, urea, potassium oxalate, and tryptophan is almost the same, and catechol exhibits the increased catalytic current of DA compared to other substances. This could be attributed to its similar structure to that of DA, which is a type of catecholamine, 3,4-dihydroxyphenyl (catechol) [66], indicating that this method for detecting DA and AA is of high selectivity.     Table 4. As shown in this table, the recovery is overall between 97% and 106% which is acceptable in this case. The proposed method results were also compared to those of the UV-visible method to confirm that there are no significant differences between these two methods, suggesting that the present method exhibits high selectivity and sensitivity to detect DA and AA. Also, both DA-HCl and vitamin C were mixed and studied by adding different concentrations of vitamin C and DA-HCl (Table 5). The results again proved that the proposed method is accurate and the recovery is in the range of 97.9% and 104%, confirming the suitability of the present method for the analysis of DA in real samples containing a large excess of AA.

12
Journal of Sensors

Conclusions
A simple and rapid electrochemical method to detect DA in the presence of a large excess of AA at the GC ox /nano-NiO x electrode is investigated. The modified electrode shows excellent electroanalytical activity towards the oxidation of DA in the presence of excess AA in PBs (pH 6.0) enabling the electrochemical determination of DA and AA using SWV in their simultaneous coexistence. The SWV of DA in the presence of excess AA indicates that the detection limit of DA on the GC ox /nano-NiO x can be 0.69 μM, which is much lower than that of AA (18.49 μM). It thus appears that the GC ox /nano-NiO x acts as an excellent electrode for detecting DA in an AA-dominating real sample.

Data Availability
The generated or analysed data used to support the findings of this study are included within the article.

Conflicts of Interest
The authors declare that they have no conflicts of interest.