Electrochemical Sensor for Determination of Various Phenolic Compounds in Wine Samples Using Fe3O4 Nanoparticles Modified Carbon Paste Electrode

Phenolic compounds contain classes of flavonoids and non-flavonoids, which occur naturally as secondary metabolites in plants. These compounds, when consumed in food substances, improve human health because of their antioxidant properties against oxidative damage diseases. In this study, an electrochemical sensor was developed using a carbon paste electrode (CPE) modified with Fe3O4 nanoparticles (MCPE) for the electrosensitive determination of sinapic acid, syringic acid, and rutin. The characterization techniques adapted for CPE, MCPE electrodes, and the solution interface were cyclic voltammetry (CV), differential pulse voltammetry (DPV), and electrochemical impedance spectroscopy (EIS). Scan rate and pH were the parameters subjected to optimization studies for the determination of phenolic compounds. The incorporation of Fe3O4 nanoparticles to the CPE as a sensor showed excellent sensitivity, selectivity, repeatability, reproducibility, stability, and low preparation cost. The limits of detection (LOD) obtained were 2.2 × 10−7 M for sinapic acid, 2.6 × 10−7 M for syringic acid, and 0.8 × 10−7 M for rutin, respectively. The fabricated electrochemical sensor was applied to determine phenolic compounds in real samples of red and white wine.


Introduction
Phenolic compounds contain a broad class of flavonoids and non-flavonoids phenols, which occur naturally as secondary metabolites throughout the plant kingdom. They spread widely into several taxonomic groups and play structural and protective functions in plants [1,2]. Flavonoids and phenolic acids contain at least one aromatic ring with one or more hydroxyl groups attached to them. They have a wide range of structures and can be classified based on the number and arrangement of carbon atoms. Flavonoids are classified into flavonols, flavones, flavan-3-ols, anthocyanidins, flavanones, isoflavones, and others, while the non-flavonoids are classified into phenolic acids, hydroxycinnamic acids, hydroxybenzoic acids, stilbenes, and other. These phenolic acids are commonly conjugated to sugars and other organic acids [3]. These phenolic compounds are used in food processing due to their properties associated with color, flavors, preservatives, and antioxidants that improve human health [4]. Phenolic compounds have a ubiquitous presence of different proportions in plant-based foods. Daily consumption of products such as fruits, wines, vegetables, grains, teas, spices, and coffees improves human health. They improve human health through their radical scavenging activities, which provide an anticancer effect against atherosclerosis, inflammatory diseases, and other oxidative electrode to enhance detection limit, provide large electroactive surface area, catalytic effect, high electromagnetic activity, attractive electron transport, sensitivity, and chemical stability [26]. The Fe 3 O 4 nanoparticles also offer a conductivity effect, making it suitable for enhancing the electron transfer between analytes and electrodes. Fe 3 O 4 nanoparticles have significant application areas in biomaterials, bioseparation, biomedical and bioengineering, and food analysis [27].
This research aims to study the electrochemical behavior of various phenolic compounds ( Figure 1) by fabricating an electrochemical sensor using carbon paste electrodes modified with Fe 3 O 4 nanoparticles. This study is the first report on using Fe 3 O 4 nanoparticles to modify carbon paste electrodes for the electrochemical determination of sinapic acid, syringic acid, and rutin based on our careful check of works reported on the detection of these phenolic compounds. CV, DPV, and EIS analyses were performed as characterization studies for CPE, MCPE, and the solution interface. Scan rate and pH studies were performed as optimization studies. A rapid validation test was carried out using gold screen-printed electrodes, and the result was compared to CPE and MCPE. The electrochemical sensor was applied in real samples of red and white wine to determine the presence of phenolic compounds.
Micromachines 2021, 12, x FOR PEER REVIEW 3 of 18 surface area, catalytic effect, and mass transport [24,25]. Fe3O4 nanoparticle belongs to the class of nanoparticles, and they are used for modifying electrodes because of their excellent electrochemical properties [25]. They are used to modify the working electrode to enhance detection limit, provide large electroactive surface area, catalytic effect, high electromagnetic activity, attractive electron transport, sensitivity, and chemical stability [26]. The Fe3O4 nanoparticles also offer a conductivity effect, making it suitable for enhancing the electron transfer between analytes and electrodes. Fe3O4 nanoparticles have significant application areas in biomaterials, bioseparation, biomedical and bioengineering, and food analysis [27]. This research aims to study the electrochemical behavior of various phenolic compounds ( Figure 1) by fabricating an electrochemical sensor using carbon paste electrodes modified with Fe3O4 nanoparticles. This study is the first report on using Fe3O4 nanoparticles to modify carbon paste electrodes for the electrochemical determination of sinapic acid, syringic acid, and rutin based on our careful check of works reported on the detection of these phenolic compounds. CV, DPV, and EIS analyses were performed as characterization studies for CPE, MCPE, and the solution interface. Scan rate and pH studies were performed as optimization studies. A rapid validation test was carried out using gold screen-printed electrodes, and the result was compared to CPE and MCPE. The electrochemical sensor was applied in real samples of red and white wine to determine the presence of phenolic compounds.

Chemicals and Reagents
The powdered phenolic compounds (sinapic acid, syringic acid, and rutin), paraffin oil, and carbon powder were all procured commercially from Sigma-Aldrich, Istanbul, Turkey. Fe3O4 nanoparticles powder was purchased commercially from Sigma-Aldrich, Istanbul, Turkey. They have a particle size of 50-100 nm, scanning electron microscopy (SEM) for surface characterization; Brunauer-Emmett-Teller (BET) surface area analysis is 6-8 m 2 /g, melting point of 1538 °C , titration by Na2S2O3, % of Fe is 71.5%, an appearance of black color, powder form and spherical shape, the density of 4.8-5.1 g/mL at 25 °C , a bulk density of 0.84 g/mL, the purity determined using trace metal analysis is 97% (≤35,000.0 ppm), a quality level of 100, Inductively coupled plasma (ICP) major analysis confirms iron component. All reagents were of analytical standards and used as obtained ( Figure 2). The pH value of the acetate buffer solutions (ABS) used for the study is 0.5 M, ABS pH 4.8. The stock solutions of the phenolic compounds were prepared with ultrapure water at a concentration of 1000 ppm (4.5 × 10 −3 M for sinapic, 5.1 × 10 −3 M for syringic, and 1.6 × 10 −3 M for rutin). The stock solution was then diluted into standard concentrations of 200 ppm (0.9 × 10 −3 M for sinapic acid, 1.0 × 10 −3 M for syringic acid, and 0.3 × 10 −3 M for rutin), which were used as working solutions. The wine samples used for the analysis were commercial brands of wine (Angola kavaklidere-dry white wine-and dikmen kavaklidere-dry red wine) purchased from a market.

Chemicals and Reagents
The powdered phenolic compounds (sinapic acid, syringic acid, and rutin), paraffin oil, and carbon powder were all procured commercially from Sigma-Aldrich, Istanbul, Turkey. Fe 3 O 4 nanoparticles powder was purchased commercially from Sigma-Aldrich, Istanbul, Turkey. They have a particle size of 50-100 nm, scanning electron microscopy (SEM) for surface characterization; Brunauer-Emmett-Teller (BET) surface area analysis is 6-8 m 2 /g, melting point of 1538 • C, titration by Na 2 S 2 O 3 , % of Fe is 71.5%, an appearance of black color, powder form and spherical shape, the density of 4.8-5.1 g/mL at 25 • C, a bulk density of 0.84 g/mL, the purity determined using trace metal analysis is 97% (≤35,000.0 ppm), a quality level of 100, Inductively coupled plasma (ICP) major analysis confirms iron component. All reagents were of analytical standards and used as obtained ( Figure 2). The pH value of the acetate buffer solutions (ABS) used for the study is 0.5 M, ABS pH 4.8. The stock solutions of the phenolic compounds were prepared with ultrapure water at a concentration of 1000 ppm (4.5 × 10 −3 M for sinapic, 5.1 × 10 −3 M for syringic, and 1.6 × 10 −3 M for rutin). The stock solution was then diluted into standard concentrations of 200 ppm (0.9 × 10 −3 M for sinapic acid, 1.0 × 10 −3 M for syringic acid, and 0.3 × 10 −3 M for rutin), which were used as working solutions. The wine samples used for the analysis were commercial brands of wine (Angola kavaklidere-dry white wine-and dikmen kavaklidere-dry red wine) purchased from a market. Micromachines 2021, 12, x FOR PEER REVIEW 5 of 18

Electrochemical Behavior of the Phenolic Compounds at CPE and MCPE
The electrochemical behaviors of the selected phenolic compounds at the CPE and MCPE surface were studied using CV and DPV analysis. The differential pulse voltammograms ( Figure 3) and the inset cyclic voltammograms show the electrochemical behavior of the analytes at the surface of electrodes. The phenolic compounds show visible peak currents higher in MCPE and lower in CPE [21,22,28]. These electrochemical behaviors of the analytes observed from the peak currents, visibly shifting, could be suggested to be a result of the nanoparticles at the electrode surface increasing the current signal as a result of the catalytic effect of the nanoparticles, thus making the current signals to increase in the modified electrodes more than the unmodified electrodes [23].
The anodic peak potentials (Epa) and cathodic peak potentials (Epc) observed from the CV analyses of the phenolic compounds (inset Figure 3) showed positions of oxidation and reduction potentials of the analytes. Sinapic acid presented positions of one Epa, while syringic acid and rutin presented two positions of Epa and Epc; the reduction peaks showed low current peak heights that are observed on the reverse scan. This behavior suggests that the oxidation reaction's product undergoes a further chemical reaction for syringic acid and rutin or is not reduced at the carbon paste electrode for sinapic acid [21]. As the peak current is higher in MCPE than CPE, the shoulders of the peaks observed from MCPE are also broader than the CPE. This behavior can be suggested to be a result of increased electroactive surface area by incorporating the Fe3O4 nanoparticles [23], which is similar to the results obtained from the determination of the electroactive surface area of the electrodes (Figures 4 and S2). If the modified electrode functioned as an electrocatalyst or the reaction was electro-catalyzed, there would have been a reduction in the peak potential, which suggests a reaction that is faster with a less overpotential [23]. The voltammetric behavior of the phenolic compounds is shown to agree with the chemical reaction proposed globally for phenolic group oxidation in aromatic compounds [23][24][25]. The peak heights of the anodic peak current (Ipa) and cathodic peak currents (Ipc) of the analytes at CPE are lower compared to that of the MCPE, suggesting that the activity which occurred at the surface of the CPE is poor and less than the MCPE [21]. The presence of Fe3O4 nanoparticles in the MCPE supports the transfer of electrons, enhances the current response, and can support the adsorption of the analyte and its enrichment onto the surface of the electrode, thereby promoting the oxidation process [29]. The peaks obtained through DPV are

Instrumentation and Methods
All the voltammetric measurements were performed using the potentiostat-galvanostat (AUTOLAB-PGSTAT204, Metrohm, Utrecht, The Netherlands) and operated with Nova 2.1.2 software. The potentiostat-galvanostat was connected to a three-electrode system cell, a carbon paste electrode, and Fe 3 O 4 nanoparticles modified carbon paste electrode as working electrodes. Ag/AgCl/3 M KCl was used as a reference electrode and a platinum wire as an auxiliary electrode in a 10 mL cell containing 0.5 M ABS pH 4.8 as supporting electrolyte. The pH measurements were all carried out with an edge H12002 pH meter (Hanna Instruments, Woonsocket, RI, USA).

Fabrication of Bare CPE and MCPE
The ratios of carbon powder to paraffin oil (binder) were compared for best results using the ratios 70:30 and 60:40 (wt/wt%), respectively, and 60:40 ratio was taken as the optimized proportion for the study. The carbon paste mixture, as the control (CPE), was prepared by hand, mixing 60 mg of carbon powder with 40 mg of paraffin oil to obtain a homogeneous mixture of 60 vs. 40 mg. The carbon paste mixture prepared with Fe 3 O 4 nanoparticles (MCPE) contained 60.0 mg carbon powder, 30.0 mg of paraffin oil, and 10.0 mg of Fe 3 O 4 powder in a ratio of 60:30:10 (wt/wt/wt%) [23]. The homogenous pastes were packed to fill two different 4 mm diameter cavity of Teflon tubes, one for the CPE and the other for the MCPE. A copper wire for conductivity was connected to the end of the electrodes (Teflon tubes). The surfaces of the electrodes were polished by smoothening them with a smooth paper to obtain a smooth and crack-free surface. After each analysis, new electrode surfaces were prepared by inserting the paste into the Teflon tubes and their surfaces polished. This process was repeated ( Figure 2) throughout the experiments before each new measurement.

Voltammetric Measurements
The voltammetric techniques used for the study were CV and DPV. The measurements of these analytes at the electrode surfaces were carried out in ABS 0.5 M, pH 4.8. The scan rate study was done using CV, by varying the applied scan rates at a range of 0.03, 0.06, 0.09, 0.12, 0. 15 6 ], which was used as an electrochemical redox probe in 0.1 M KCl. The same condition was used for the scan rate study by varying the scan rates from 0.1, 0.12, 0.14, 0.16, 0.18, and 0.2 V/s, employing the Randles-Sevcik equation.

Preparation and Detection Procedure of Real Samples (Red and White Wine)
DPV technique was used to analyze phenolic compounds' content in spiked samples of the red and white wine samples. The voltammograms produced were recorded using a method of standard addition of serial dilutions of known volumes and concentrations of the phenolic compounds (sinapic acid, syringic acid, and rutin). A volume of 1 mL of the wine samples only was inserted into a 10 mL beaker and was completed with ABS of (0.5 M, pH 4.8) to a volume of 10 mL. Aliquots of the standard phenolic compounds from (0.03 × 10 −3 -0.05 × 10 −3 M) were then added to the 10 mL beakers having 1 mL of the wine samples and completed to 10 mL with ABS. After which, they were stirred for two minutes with a magnetic stirrer. Measurements from the DPV analysis were recorded from each beaker that contains the wine and aliquots of the standard phenolic compounds [23].

Electrochemical Behavior of the Phenolic Compounds at CPE and MCPE
The electrochemical behaviors of the selected phenolic compounds at the CPE and MCPE surface were studied using CV and DPV analysis. The differential pulse voltammograms ( Figure 3) and the inset cyclic voltammograms show the electrochemical behavior of the analytes at the surface of electrodes. The phenolic compounds show visible peak currents higher in MCPE and lower in CPE [21,22,28]. These electrochemical behaviors of the analytes observed from the peak currents, visibly shifting, could be suggested to be a result of the nanoparticles at the electrode surface increasing the current signal as a result of the catalytic effect of the nanoparticles, thus making the current signals to increase in the modified electrodes more than the unmodified electrodes [23].
The anodic peak potentials (E pa ) and cathodic peak potentials (E pc ) observed from the CV analyses of the phenolic compounds (inset Figure 3) showed positions of oxidation and reduction potentials of the analytes. Sinapic acid presented positions of one E pa , while syringic acid and rutin presented two positions of E pa and E pc ; the reduction peaks showed low current peak heights that are observed on the reverse scan. This behavior suggests that the oxidation reaction's product undergoes a further chemical reaction for syringic acid and rutin or is not reduced at the carbon paste electrode for sinapic acid [21]. As the peak current is higher in MCPE than CPE, the shoulders of the peaks observed from MCPE are also broader than the CPE. This behavior can be suggested to be a result of increased electroactive surface area by incorporating the Fe 3 O 4 nanoparticles [23], which is similar to the results obtained from the determination of the electroactive surface area of the electrodes (Figure 4 and Figure S2). If the modified electrode functioned as an electrocatalyst or the reaction was electro-catalyzed, there would have been a reduction in the peak potential, which suggests a reaction that is faster with a less overpotential [23]. The voltammetric behavior of the phenolic compounds is shown to agree with the chemical reaction proposed globally for phenolic group oxidation in aromatic compounds [23][24][25]. The peak heights of the anodic peak current (I pa ) and cathodic peak currents (I pc ) of the analytes at CPE are lower compared to that of the MCPE, suggesting that the activity which occurred at the surface of the CPE is poor and less than the MCPE [21]. The presence of Fe 3 O 4 nanoparticles in the MCPE supports the transfer of electrons, enhances the current response, and can support the adsorption of the analyte and its enrichment onto the surface of the electrode, thereby promoting the oxidation process [29]. The peaks obtained through DPV are shown to be better defined and have high sensitivity to low concentration of analytes and lower background current when compared to the results obtained using CV. shown to be better defined and have high sensitivity to low concentration of analytes and lower background current when compared to the results obtained using CV. The hydroxy groups of the phenols are oxidized through the transfer of two electrons, which form a quinone group after the liberation of 2H + . The phenolic compounds with one anodic peak indicate an electrochemical behavior, which suggests an oxidation reaction that leads to the formation of a stable quinone group, which is reduced on the reversed scan ( Figure S1A). This is also similar to the hydroxy group's oxidation of other phenolic compounds at their ortho position [30]. However, the phenolic compounds with two anodic peaks indicate the formation of a semiquinone radical in the first step. The second peak corresponds to the oxidation of the semiquinone to the quinone group. The Fe3O4 nanoparticles provide stability for the complete oxidation of the phenolic compounds [21,31] ( Figure S1B).

Evaluation of the Electroactive Surface Area
The electroactive surface area of the electrodes CPE and MCPE were determined using CV in 1 mM K4[Fe(CN)6], which was used as an electrochemical redox probe in 0.1 M KCl. MCPE displayed an enhancement of the current response ( Figure 4), which indicates that the CPE's electrochemical active sites were increased on surface modification by the  Figure S2B) show that the oxidation and reduction potentials were shifted to more positive and more negative potentials, respectively, with a linear increase of the redox peak current as the scan rate is enhanced from 0.1 to 0.2 V/s. The plot of Ipa versus υ 1/2 ( Figure S2C,D) shows linearity with an R 2 value of 0.9963 for CPE and 0.9830 for MCPE. The electrodes' electroactive surface area was estimated according to the slope of Ipa versus υ 1/2 for a known concentration of K4Fe(CN)6 using the Randles-Sevcik equation [32].
Ipa = 2.69 × 10 5 n 3/2 ACoD 1/2 υ 1/2 (1) Ipa: indicates anodic peak current (A), n: the number of electrons exchanged during the redox process, which is presumed to be equal to one, A: surface area of the electrode (cm 2 ), Co: concentration of the redox probe (mol cm −3 ), D: diffusion coefficient assumed to be equal to 6.23 × 10 -6 cm 2 s -1 , and from the slopes of Ipa-υ 1/2 relation, the microscopic electroactive surface area was calculated to be MCPE (0.043 cm 2 ) in comparison with the CPE (0.015 cm 2 ). The results show that the presence of Fe3O4 nanoparticles increased the active surface area of the electrode.

Effect of pH on the Phenolic Compounds Oxidation at CPE and MCPE
The effect of pH of the buffer solution on the current response of phenolic compounds oxidation at CPE and MCPE was studied using CV to observe their electrochemical behaviors ( Figure 5). The pH of the different buffer solutions affected the oxidation activity of the phenolic compounds on the surface of CPE and MCPE, thereby causing changes to the electrochemical behavior of the phenolic compounds. This effect can be The hydroxy groups of the phenols are oxidized through the transfer of two electrons, which form a quinone group after the liberation of 2H + . The phenolic compounds with one anodic peak indicate an electrochemical behavior, which suggests an oxidation reaction that leads to the formation of a stable quinone group, which is reduced on the reversed scan ( Figure S1A). This is also similar to the hydroxy group's oxidation of other phenolic compounds at their ortho position [30]. However, the phenolic compounds with two anodic peaks indicate the formation of a semiquinone radical in the first step. The second peak corresponds to the oxidation of the semiquinone to the quinone group. The Fe 3 O 4 nanoparticles provide stability for the complete oxidation of the phenolic compounds [21,31] ( Figure S1B).

Evaluation of the Electroactive Surface Area
The electroactive surface area of the electrodes CPE and MCPE were determined using CV in 1 mM K 4 [Fe(CN) 6 ], which was used as an electrochemical redox probe in 0.1 M KCl. MCPE displayed an enhancement of the current response (Figure 4), which indicates that the CPE's electrochemical active sites were increased on surface modification by the Fe 3 O 4 nanoparticles. The MCPE presented a larger current response (I pa = 15.20 µA) in comparison to the CPE current response (I pa = 8.64 µA); this can be attributed to the electrocatalytic activity and enhancement of the modified surface area. The cyclic voltammograms of CPE ( Figure S2A) and MCPE ( Figure S2B) show that the oxidation and reduction potentials were shifted to more positive and more negative potentials, respectively, with a linear increase of the redox peak current as the scan rate is enhanced from 0.1 to 0.2 V/s. The plot of Ipa versus υ 1/2 ( Figure S2C,D) shows linearity with an R 2 value of 0.9963 for CPE and 0.9830 for MCPE. The electrodes' electroactive surface area was estimated according to the slope of Ipa versus υ 1/2 for a known concentration of K 4 Fe(CN) 6 using the Randles-Sevcik equation [32].
I pa : indicates anodic peak current (A), n: the number of electrons exchanged during the redox process, which is presumed to be equal to one, A: surface area of the electrode (cm 2 ), C o : concentration of the redox probe (mol cm −3 ), D: diffusion coefficient assumed to be equal to 6.23 × 10 -6 cm 2 s -1 , and from the slopes of I pa -υ 1/2 relation, the microscopic electroactive surface area was calculated to be MCPE (0.043 cm 2 ) in comparison with the CPE (0.015 cm 2 ). The results show that the presence of Fe 3 O 4 nanoparticles increased the active surface area of the electrode.

Effect of pH on the Phenolic Compounds Oxidation at CPE and MCPE
The effect of pH of the buffer solution on the current response of phenolic compounds oxidation at CPE and MCPE was studied using CV to observe their electrochemical behaviors ( Figure 5). The pH of the different buffer solutions affected the oxidation activity of the phenolic compounds on the surface of CPE and MCPE, thereby causing changes to the electrochemical behavior of the phenolic compounds. This effect can be seen ( Figure 5), as the anodic peak currents and potentials of the phenolic compounds on the CPE and MCPE showed a progressive decrease with increasing pH values from 2.6 to 9.2 [33]. The cyclic voltammograms of the phenolic compounds showed a clear pH dependence of their electrochemical behavior at the surface of the electrodes, as the increase in the pH of the ABS gradually lead to a decrease of the anodic peak current. As the buffer solution's pH increases, there is a gradual negative shift of the peak potentials, which shows a linear relationship between the pH values and Epa [33]. The relationship between the anodic peak potentials Epa and the pH values were studied, and the plots produced showed a linear regression relationship with an equation having values for the phenolic compounds presented in Table 1 and Figure S3. All of the values produced from the slope of Epa/pH of the regression line are compared to the Nernstian value of 59 mV/pH and 29.5 mV/pH, which shows the number of electron and protons involved in oxidation/reduction reaction for two-electron/two-proton process and two- As the buffer solution's pH increases, there is a gradual negative shift of the peak potentials, which shows a linear relationship between the pH values and E pa [33]. The relationship between the anodic peak potentials E pa and the pH values were studied, and the plots produced showed a linear regression relationship with an equation having values for the phenolic compounds presented in Table 1 and Figure S3. All of the values produced from the slope of E pa /pH of the regression line are compared to the Nernstian value of 59 mV/pH and 29.5 mV/pH, which shows the number of electron and protons involved in oxidation/reduction reaction for two-electron/two-proton process and two-electron/oneproton process [30,[34][35][36] The electrochemical behavior could further be explained by the fact that at low pH value, the concentration of the analytes protonated form oxidized is high and increases with decreasing the pH [37]. This supports the ease of oxidation reaction and enhances mass transport at the surface of the electrode. When the pH is increased, the current begins to gradually decrease, which could be suggested to be a result of the decrease of the protonated form concentration [37].

Effect of Scan Rate on the Phenolic Compounds Oxidation at CPE and MCPE
The influence of scan rate on the electro-oxidation behavior of the phenolic compounds at CPE and MCPE surface was demonstrated using CV ( Figure 6). The voltammograms show an increase in the peak current signals with increased applied scan rates. The values measured for the peak current were used for plotting linear equation of peak current Ip versus square root of scan rate ν 1/2 , which indicated a typical diffusion-controlled reaction ( Table 2 and Figure S4). Another plot of the peak currents Ip versus scan rate (ν) both for anodic and cathodic peak currents using same experimental conditions was performed and yielded a straight line (Table 2 and Figure S5) which is typical for adsorption controlled. As the scan rate applied increases, the peak currents for anodic and cathodic also increase linearly, indicating a quasi-reversible oxidation reaction [38]. Table 2. Linear regression equations show the dependence of redox peak current Ip on the square root of scan rate ν 1/2 (V/s) 1/2 for controlled diffusion and dependence of redox peak current Ip on scan rate ν (V/s) for controlled adsorption for phenolic compounds at bare CPE and MCPE with their slopes and R-square values. values measured for the peak current were used for plotting linear equation of peak current Ip versus square root of scan rate ν 1/2 , which indicated a typical diffusion-controlled reaction (Table 2 and Figure S4). Another plot of the peak currents Ip versus scan rate (ν) both for anodic and cathodic peak currents using same experimental conditions was performed and yielded a straight line (Table 2 and Figure S5) which is typical for adsorption controlled. As the scan rate applied increases, the peak currents for anodic and cathodic also increase linearly, indicating a quasi-reversible oxidation reaction [38].  The logarithm of anodic peak current and logarithm of scan rate (log Ipa versus log ν) approach was used to confirm whether the electrochemical reaction at the electrode surface is diffusion or adsorption controlled (Table 3 and Figure S6) using linear relationship plots. The values of the slopes obtained were close to 0.5, which is attributed to electrochemical reactions that are diffusion-controlled [39][40][41]. Table 3. Linear regression equation showing the logarithm of anodic peak current and logarithm of The logarithm of anodic peak current and logarithm of scan rate (log I pa versus log ν) approach was used to confirm whether the electrochemical reaction at the electrode surface is diffusion or adsorption controlled (Table 3 and Figure S6) using linear relationship plots. The values of the slopes obtained were close to 0.5, which is attributed to electrochemical reactions that are diffusion-controlled [39][40][41].

Characterization of CPE and MCPE Using EIS
The EIS was used to study the difference in the behavior of the CPE and the MCPE. This method is an effective tool used to study the electrode/solution interface properties and how charge transfer occurs between the redox solution/electrode interface. Both electrodes were measured in the redox solution of [Fe(CN) 6 ] 3−/4− (5 mM) containing 1 M KNO 3 , using the frequency range of 100 kHz-0.1 Hz, to evaluate the charge transfer resistance (Rct) of electrodes which corresponds to the Randles equivalent circuit (Figure 7 inset). Rs represents solution resistance, Rct is charge transfer resistance, Cdl is double-layer capacitance, and W is Warburg impedance. The Nyquist plots (Figure 7) demonstrate the semicircles of CPE and MCPE.

Characterization of CPE and MCPE Using EIS
The EIS was used to study the difference in the behavior of the CPE and the MCPE. This method is an effective tool used to study the electrode/solution interface properties and how charge transfer occurs between the redox solution/electrode interface. Both electrodes were measured in the redox solution of [Fe(CN)6] 3−/4− (5 mM) containing 1 M KNO3, using the frequency range of 100 kHz-0.1 Hz, to evaluate the charge transfer resistance (Rct) of electrodes which corresponds to the Randles equivalent circuit (Figure 7 inset). Rs represents solution resistance, Rct is charge transfer resistance, Cdl is double-layer capacitance, and W is Warburg impedance. The Nyquist plots (Figure 7) demonstrate the semicircles of CPE and MCPE.
The Rct values for CPE and MCPE were 15.32 kΩ and 6.84 kΩ, respectively. The Rct value for CPE that is the largest, indicates a very slow electron transfer rate between the redox solution and the electrode interface. The Rct value offered at MCPE implies fast charge transfer. The results suggest that the nanoparticles' presence can facilitate electron transfer between the electrode surface and the redox solution, thereby increasing electroconductibility. Hence, Fe3O4 nanoparticle was very efficient for developing an electrochemical sensor for the analysis of phenolic compounds [23,42,43].

Application of Gold Screen-Printed Electrode for Rapid Validation Test of Phenolic Compounds Using Cyclic Voltammetry
The gold screen-printed electrode was applied in the detection of rutin and sinapic acid using cyclic voltammetry technique in 0.5 mol L −1 ABS with a pH value of 4.8, at a scan rate of 0.2 V/s in a reversible potential sweep range of −0.4 to +1.0 V. This analysis was performed as a rapid test for the detection of these phenolic compounds and to compare the results with the CPE and MCPE because of its reproducibility, sensitivity, accuracy, and avoidance of preparation and cleaning process. The voltammograms of the rutin and sinapic acid on the electrode surface ( Figure 8) indicate an overlay of the gold screenprinted electrode, CPE and MCPE. The sensitivity of the gold screen-printed electrode to the concentration of the phenolic compounds is compared to CPE and MCPE used, using the current density (J) = Current Intensity (A)/Cross-sectional Area (cm 2 ). The surface area was taken using π r 2 and divided by the value of current response for sinapic and rutin from the three electrodes used. For sinapic acid, the current density obtained is 0.1012 × 10 −3 A/cm 2 for gold screen-printed electrode, 0.1541 × 10 −3 A/cm 2 for CPE and 0.2466 × 10 −3 A/cm 2 for MCPE, for rutin the current density obtained is 0.0499 × 10 −3 A/cm 2 for gold The Rct values for CPE and MCPE were 15.32 kΩ and 6.84 kΩ, respectively. The Rct value for CPE that is the largest, indicates a very slow electron transfer rate between the redox solution and the electrode interface. The Rct value offered at MCPE implies fast charge transfer. The results suggest that the nanoparticles' presence can facilitate electron transfer between the electrode surface and the redox solution, thereby increasing electro-conductibility. Hence, Fe 3 O 4 nanoparticle was very efficient for developing an electrochemical sensor for the analysis of phenolic compounds [23,42,43].

Application of Gold Screen-Printed Electrode for Rapid Validation Test of Phenolic Compounds Using Cyclic Voltammetry
The gold screen-printed electrode was applied in the detection of rutin and sinapic acid using cyclic voltammetry technique in 0.5 mol L −1 ABS with a pH value of 4.8, at a scan rate of 0.2 V/s in a reversible potential sweep range of −0.4 to +1.0 V. This analysis was performed as a rapid test for the detection of these phenolic compounds and to compare the results with the CPE and MCPE because of its reproducibility, sensitivity, accuracy, and avoidance of preparation and cleaning process. The voltammograms of the rutin and sinapic acid on the electrode surface ( Figure 8) indicate an overlay of the gold screen-printed electrode, CPE and MCPE. The sensitivity of the gold screen-printed electrode to the concentration of the phenolic compounds is compared to CPE and MCPE used, using the current density (J) = Current Intensity (A)/Cross-sectional Area (cm 2 ). The surface area was taken using π r 2 and divided by the value of current response for sinapic and rutin from the three electrodes used. For sinapic acid, the current density obtained is 0.1012 × 10 −3 A/cm 2 for gold screen-printed electrode, 0.1541 × 10 −3 A/cm 2 for CPE and 0.2466 × 10 −3 A/cm 2 for MCPE, for rutin the current density obtained is 0.0499 × 10 −3 A/cm 2 for gold screen-printed electrode, 0.0538 × 10 −3 A/cm 2 for CPE, and 0.0801 × 10 −3 A/cm 2 for MCPE. The result could also be suggested that the gold screen-printed active mass surface area is smaller than that of the CPE and MCPE. The use of the gold screen-printed electrode is important for future applications in the manufacture of electrochemical food sensor devices, which can detect different compounds present in food substances.
Micromachines 2021, 12, x FOR PEER REVIEW 12 of 18 screen-printed electrode, 0.0538 × 10 −3 A/cm 2 for CPE, and 0.0801 × 10 −3 A/cm 2 for MCPE. The result could also be suggested that the gold screen-printed active mass surface area is smaller than that of the CPE and MCPE. The use of the gold screen-printed electrode is important for future applications in the manufacture of electrochemical food sensor devices, which can detect different compounds present in food substances.

Effect of Concentration on the Phenolic Compounds Oxidation at CPE and MCPE
The effect of increasing the concentration of the phenolic compounds on their oxidation signals at MCPE was studied using DPV. The results were used to determine the limit of detection (LOD) and limit of quantification (LOQ) of the voltammetric method optimized for the quantification of the phenolic compounds on the modified carbon paste electrode surface. The phenolic compounds used in this work were investigated in a range of concentration from 0.3 × 10 −6 -13.0 × 10 −6 M. The Equations (1) and (2) were used to calculate the limit of detection and limit of quantification of the phenolic compounds using the peak currents, respectively.
where "b" is the slope of our calibration curve, and "Sa" represents the standard deviation.
The recorded oxidation signals of the phenolic compounds increased with a gradual increase in the concentration of the phenolic compounds ranging from 0.3 × 10 −6 -13.0 × 10 −6 M. The results obtained showed a linear relationship between peak currents and the change in concentration of the phenolic compounds. The following are the linear regression equations of the phenolic compounds: Ip = 1.3982 C + 1.2362 (Ip: µ A, C: mol L −1 and R 2 = 0.9865) for sinapic acid, Ip = 0.1457 C + 0.7410 (Ip: µ A, C: mol L −1 and R 2 = 0.9851) for syringic acid, and Ip = 0.7163 C + 0.6859 (Ip: µ A, C: mol L −1 and R 2 = 0.9860) for rutin. The proposed method for the phenolic compounds detection limit is compared with the maximum levels of antioxidants, within a range of 20 to 1000 ppm (20 to 1000 mg L −1 ) that are permitted within the guidelines for food taken within the EU and North America [44]. The detection limits of the developed DPV method for the phenolic compounds were calculated. The values were compared with other data reported by other research groups (Table  4), where rutin was reported with other acids, but less or no work has been reported on sinapic acid and syringic acid, respectively.

Effect of Concentration on the Phenolic Compounds Oxidation at CPE and MCPE
The effect of increasing the concentration of the phenolic compounds on their oxidation signals at MCPE was studied using DPV. The results were used to determine the limit of detection (LOD) and limit of quantification (LOQ) of the voltammetric method optimized for the quantification of the phenolic compounds on the modified carbon paste electrode surface. The phenolic compounds used in this work were investigated in a range of concentration from 0.3 × 10 −6 -13.0 × 10 −6 M. The Equations (1) and (2) were used to calculate the limit of detection and limit of quantification of the phenolic compounds using the peak currents, respectively.
where "b" is the slope of our calibration curve, and "Sa" represents the standard deviation. The recorded oxidation signals of the phenolic compounds increased with a gradual increase in the concentration of the phenolic compounds ranging from 0.3 × 10 −6 -13.0 × 10 −6 M. The results obtained showed a linear relationship between peak currents and the change in concentration of the phenolic compounds. The following are the linear regression equations of the phenolic compounds: Ip = 1.3982 C + 1.2362 (Ip: µA, C: mol L −1 and R 2 = 0.9865) for sinapic acid, Ip = 0.1457 C + 0.7410 (Ip: µA, C: mol L −1 and R 2 = 0.9851) for syringic acid, and Ip = 0.7163 C + 0.6859 (Ip: µA, C: mol L −1 and R 2 = 0.9860) for rutin. The proposed method for the phenolic compounds detection limit is compared with the maximum levels of antioxidants, within a range of 20 to 1000 ppm (20 to 1000 mg L −1 ) that are permitted within the guidelines for food taken within the EU and North America [44]. The detection limits of the developed DPV method for the phenolic compounds were calculated. The values were compared with other data reported by other research groups (Table 4), where rutin was reported with other acids, but less or no work has been reported on sinapic acid and syringic acid, respectively. Table 4. Limits of detection (LOD) and limit of quantification (LOQ) reported for the differential pulse voltammetry method employed in detecting phenolic compounds compared to other methods used.

Reproducibility, Repeatability, and Stability
The sensor's reproducibility was investigated by using the MCPE for the determination of 0.9 × 10 −3 M for sinapic acid, 1.0 × 10 −3 M for syringic acid, and 0.3 × 10 −3 M for rutin, respectively, using DPV in 0.5 M ABS pH 4.8. Seven independent electrodes were used to determine each analyte. The relative standard deviations (RSD) were found to be 4.2% for sinapic acid, 3.6% for syringic acid, and 4.6% for rutin ( Figure S7), hence showing good reproducibility. The repeatability was also investigated using seven prepared modified electrodes in seven prepared samples for each analyte, and the relative standard deviations of the peak currents were found to be 3.1% for sinapic acid, 4.2% for syringic acid, and 4.2% for rutin, hence indicating good repeatability ( Figure S7). Three modified electrodes were prepared for the determination of the stability of the sensor. A potential of 0.6 V was applied using the chronoamperometry method for each analyte with the above concentration at the modified electrode for 30 min, respectively ( Figure S8). These potentials are comparable to the phenolic compounds' oxidation potentials of the CV analysis results done previously from this study. The amperometric response observed remained constant throughout the experiment. The surface of the electrodes did not undergo any fouling; hence, this attests to the proposed sensor's stability [43]. The stability was again analyzed using the above concentrations of the analytes using DPV on the first day using two modified electrodes, which was then stored for 10 days at room temperature in the laboratory. The electrodes were then used to determine the same concentration of the phenolic compounds after 10 days. The respective voltammograms on the 1st day and 10th day of the modified electrodes ( Figure S8) demonstrated good stability with relative standard deviation values of 3.85% for sinapic acid, 4.54% for syringic acid, and 7.17% for rutin, respectively.

Selectivity of the Electrode
To evaluate the selectivity of the fabricated sensor in identifying the analytes of interest, the effects of possible interferences were investigated by analyzing a standard solution of 0.9 × 10 −3 M for sinapic, 1.0 × 10 −3 M for syringic, and 0.3 × 10 −3 M for rutin respectively in 0.5 M ABS pH 4.8. Common inorganic ions such as K + , Cl − , Fe +3 , and Ca +2 , had no significant interference in determining the phenolic compounds with an RSD of the oxidation peaks obtained to be less than 5% (Table 5). Other potential electroactive organic interferences, such as caffeic acid and 4-hydroxybenzoic acid, which may co-exist with the analytes, were also examined. These organic interferences with their concentration increased about 500-fold excess did not meaningfully change the oxidation peak currents of the analytes of interest, and the RSD values obtained were less than 5% (Table 5). Therefore, MCPE can be used for the selective determination of sinapic acid, syringic acid, and rutin. DPV was used to study all the three phenolic compounds (sinapic acid, syringic acid, and rutin) simultaneously to observe their electro-oxidation behavior at MCPE (Figure 9). The analytes presented oxidation potentials similar and within the same potential with the oxidation potentials of other results in this study, where the analytes were analyzed individually. This result showed that the modified electrode has an ability to detect the presence of all the three phenolic acids simultaneously in the solution.

Simultaneous Detection of the Phenolic Compounds at MCPE
DPV was used to study all the three phenolic compounds (sinapic acid, syringic acid, and rutin) simultaneously to observe their electro-oxidation behavior at MCPE (Figure 9). The analytes presented oxidation potentials similar and within the same potential with the oxidation potentials of other results in this study, where the analytes were analyzed individually. This result showed that the modified electrode has an ability to detect the presence of all the three phenolic acids simultaneously in the solution.

Application of CPE and MCPE on Phenolic Compounds Detection in Red and White Wines
The determination of the presence of the phenolic compounds in red and white samples, respectively, was done using MCPE. This study was carried out with diluted 10 mL samples of red and white wine, which served as blanks. CV and DPV of the red and white wine samples were analyzed ( Figure S9) using the modified electrode without the presence of the standard phenolic compounds to observe if these particular commercial wine samples that were purchased contained sinapic, syringic, and rutin. The sensor used detected that the wine samples did not contain the presence of the phenolic compounds of interest; instead, they contained other antioxidants or sulphites, indicating oxidation potentials that seem to appear as the analyte of interest. The CV of the wine samples showed little anodic peaks, while the DPV was able to indicate the presence of some other compounds, which were not the standard phenolic compounds used except sinapic acid in red wine, which has a very low concentration. Aliquots of the known concentration of phenolic compounds (sinapic, syringic, and rutin) through the standard addition method were then added to the wine samples to observe the modified electrode's detection ability.

Application of CPE and MCPE on Phenolic Compounds Detection in Red and White Wines
The determination of the presence of the phenolic compounds in red and white samples, respectively, was done using MCPE. This study was carried out with diluted 10 mL samples of red and white wine, which served as blanks. CV and DPV of the red and white wine samples were analyzed ( Figure S9) using the modified electrode without the presence of the standard phenolic compounds to observe if these particular commercial wine samples that were purchased contained sinapic, syringic, and rutin. The sensor used detected that the wine samples did not contain the presence of the phenolic compounds of interest; instead, they contained other antioxidants or sulphites, indicating oxidation potentials that seem to appear as the analyte of interest. The CV of the wine samples showed little anodic peaks, while the DPV was able to indicate the presence of some other compounds, which were not the standard phenolic compounds used except sinapic acid in red wine, which has a very low concentration. Aliquots of the known concentration of phenolic compounds (sinapic, syringic, and rutin) through the standard addition method were then added to the wine samples to observe the modified electrode's detection ability.
The wine samples analysis was carried out with DPV (Table 6), and the results of the analysis of wine samples suggest the activity of oxidation that occurred. However, as the spiked wine concentrations were increased, there was an increase in the oxidation peaks of the phenolic compounds in the red and white wine. The modified electrode detected the presence of the added standard phenolic compounds in both white and red wine samples with recoveries at almost 100%.

Conclusions
The fabrication of this electrochemical sensor for detecting the selected phenolic compounds (sinapic acid, syringic acid, and rutin) using Fe 3 O 4 nanoparticles to modify CPE is first reported in this study after a careful check of other reported articles. The results obtained by CV, DPV, and EIS showed that the CPE modified with Fe 3 O 4 nanoparticles increased the peak current, leading to increased sensitivity to bare CPE. EIS analysis confirmed that MCPE exhibited increased electro-conductibility, thus enhancing the electron transfer between the electrode surface and the redox solution. The MCPE showed high sensitivity, selectivity, reproducibility, repeatability, and stability towards the determination of the phenolic compounds. The CPE and MCPE were used to compare commercial gold screen-printed electrodes for rapid detection of the phenolic compounds, confirmed by their current density that the MCPE had higher current density than CPE and gold screen-printed electrodes. The fabrication of this electrochemical sensor was simple and costand time-effective. The LOD and LOQ results were compared to other literature' sensors; syringic acid is first reported in this work. The electrochemical sensor was applied for real sample analysis to determine phenolic compounds in red and white wine samples. The results found are within the maximum concentrations of 20 to 1000 ppm (20 to 1000 mg L −1 ) antioxidant levels permitted for phenolic compounds in food samples within the EU and North America.
Supplementary Materials: The following are available online at https://www.mdpi.com/2072-6 66X/12/3/312/s1. Figure S1: The chemical reaction process showing the oxidation of phenolic compounds having one peak in the reaction path (A) and the oxidation of phenolic compounds having two peaks in the reaction path (B). Figure Figure S6: The linear relationship plots of the logarithm of peak current and logarithm of scan rate (log I pa versus log ν) for phenolic compounds. The plots represent peak current and logarithm of scan rate at CPE and MCPE. Figure S7: Differential voltammograms showing the reproducibility and repeatability of MCPE for the determination of the phenolic compounds, respectively. Figure S8: A steady amperometric current response for determination of stability of the sensor for phenolic compounds detection in (0.5 M ABS of pH 4.8), for 30 min, at MCPE using 0.6 V potential. Differential pulse voltammograms show the determination of 0.9 × 10 −3 M for sinapic, 1.0 × 10 −3 M for syringic, and 0.3 × 10 −3 M for rutin, respectively, in 0.5 M ABS pH 4.8 for 1st day and 10th day to show the stability of the modified electrode. Figure