Electrochemical Deposition of Gold Nanoparticles on Reduced Graphene Oxide by Fast Scan Cyclic Voltammetry for the Sensitive Determination of As(III)

In this study, a stable, sensitive electrochemical sensor was fabricated by the electrochemical codeposition of reduced graphene oxide (rGO) and gold nanoparticles on a glassy carbon electrode (rGO-Aunano/GCE) using cyclic voltammetry (CV), which enabled a simple and controllable electrode modification strategy for the determination of trace As(III) by square wave anodic stripping voltammetry (SWASV). SWASV, CV, electrochemical impedance spectroscopy (EIS), X-ray diffraction (XRD) and scanning electron microscopy (SEM) were used to characterize the electrochemical properties and morphology of the proposed sensing platform. The number of sweep segments, the deposition potential and the deposition time were optimized to obtain ideal sensitivity. The presence of rGO from the electroreduction of graphene oxide on the sensing interface effectively enlarged the specific surface area and consequently improved the preconcentration capacity for As(III). The rGO-Aunano/GCE sensor exhibited outstanding detection performance for As(III) due to the combined effect of Aunano and rGO formed during the electroreduction process. Under the optimized conditions, a linear range from 13.375 × 10−9 to 668.75 × 10−9 mol/L (1.0 to 50.0 μg/L) was obtained with a detection limit of 1.07 × 10−9 mol/L (0.08 μg/L) (S/N = 3). The reproducibility and reliability of the rGO-Aunano/GCE sensor were also verified by performing 8 repetitive measurements. Finally, the rGO-Aunano/GCE sensor was used for the analysis of real samples with satisfactory results.


Introduction
While As(III) is one of the most toxic forms of arsenic, even at low concentrations, it is widespread in natural environments [1,2]. As(III) contamination in agricultural soil is a serious problem because the presence of As in the food chain can cause many health problems [3,4], such as bladder cancer, lung cancer, keratosis and skin lesions. Therefore, the development of a simple, fast and sensitive method for the determination of As(III) in soil is urgently needed. Generally, there are two types of analytical strategies for the detection of As(III): spectroscopic and electrochemical methods. Spectroscopic methods, such as inductively coupled plasma mass spectrometry [5], atomic absorption spectroscopy [6] and hydride generation atomic fluorescence spectrometry [7], have very high detection accuracy. However, these methods require expensive instruments, long analysis times, and laboratory conditions. They are not suitable for on-site analyses and routine monitoring, especially for large

Materials
HAuCl 4 was purchased from Sinopharm Chemical Reagent Co., Ltd. (c). Arsenic trioxide (As 2 O 3 ) was purchased from Aldrich (Sigma-Aldrich, Saint Louis, MO, USA) and then dissolved in 1.0 M aqueous NaOH to create a 1000 mg/L solution. GO was obtained from Nanjing JCNANO Materials Tech Co., Ltd. (Nanjing, China). A nitric acid solution (1.0 M) was used as the supporting electrolyte during the stripping voltammetry analysis of As(III). Millipore-Q (18.2 MΩ) water was used for all experiments.

Instrument
Scanning electron microscopy (SEM) was performed on a NovaNanoSEM 450 scanning electron microscope. X-ray diffraction (XRD) patterns of Au nano were obtained with a PANalytical Empyrean Series 2 X-ray diffractometer in the 2θ range from 30 • to 90 • . The electrochemical measurements and analyses, such as electrochemical impedance spectroscopy (EIS), CV and SWASV, were carried out using a CHI660D electrochemical workstation (Shanghai CH Instruments, Shanghai, China). A Pt wire electrode, Ag/AgCl electrode and GCE (ø3 mm) were used as the counter, reference and working electrodes, respectively, to construct a three-electrode system. A 25 mL beaker was used as the electrolytic cell for all electrochemical measurements. A magnetic stirrer was used to stir the test solution during the deposition step.

Synthesis and Modification of the rGO-Au nano Nanocomposite
Prior to modification, the GCE surface was polished using 0.05 mm alumina powder, and the electrode was then sonicated in 1:1 HNO 3 , absolute ethanol and water in sequence to remove alumina residue. GO was dispersed in a PBS solution (0.1 M, pH 9.0) by ultrasonication to form a GO dispersion with a concentration of 0.1 mg/mL. HAuCl 4 was added to the GO suspension to form 0.1 mM Au 3+ , and then nitrogen purging was performed for 1 h to deoxygenate the solution and exfoliate the stacked GO. CV electrodeposition was carried out at a scan rate of 50 mV/s in the potential range from −1.4 to 0.6 V for 10 sweep segments (i.e., 5 cycles). After electrodeposition, the rGO-Au nano /GCE was gently washed with ultrapure water. The same procedures described above were used to prepare the other electrodes.

Electrochemical Detection of As(III)
The SWASV measurements for As(III) detection were carried out based on a three-electrode system that used the rGO-Au nano /GCE as the working electrode under the optimized conditions. A deposition potential of −0.3 V vs. Ag/AgCl was applied to the as-prepared rGO-Au nano /GCE for 250 s in a nitric acid solution (0.1 M) with stirring during the deposition step. After equilibration for 10 s, the stripping voltammogram was recorded from −0.2 to 0.4 V vs. Ag/AgCl with a step potential of 5 mV, a pulse amplitude of 25 mV and a frequency of 25 Hz. An activation process was initiated in the nitric acid solution (0.1 M) to clean the electrode surface and remove previous residual As(0) using a constant positive potential of 0.6 V for 120 s. For comparison, the same process and conditions were used for the other modified electrodes.

Soil Sample Preparation
The soil samples were obtained from farmland in China. The detailed process is as follows: first, 40 mL of 0.1 M nitric acid was mixed with a 10 g soil sample using strong shaking. Then, a portable ultrasonic extractor was used to sonicate the above mixture for 10 min. After that, the obtained mixed solution was centrifuged (2000 rpm) using a portable centrifuge, and then, 20 mL of supernatant was loaded into a 30 mL electrolyte cell for the SWASV measurements. Krasnodêbska-Ostrêga and Kowalska [24] verified that there is no significant difference between the ultrasonic-assisted extraction method and the method proposed by the European Communities Bureau of Reference.

Electrochemical Deposition of the rGO-Au nano Composite
The CV electroanalysis curves of GO on a GCE can be seen in Figure 1A, and one oxidation peak (a) and two reduction peaks (b and c) can be observed. The peak currents increased with successive potential scans, demonstrating that GO was successfully electrochemically reduced on the electrode surface. The current peak (c) was due to the reduction of the oxygen-containing groups on GO [25][26][27]. The peak currents that appeared at positions "a" and "b" were due to the redox reaction of some redox pairs on the graphene plane, and peak "c" was attributed to the irreversible electrochemical reduction of GO [28].
The CV curve for the electroanalysis of 0.1 mM HAuCl 4 is shown in Figure 1B. A significant reduction peak is present at approximately 0.18 V, indicating that the gold salt was reduced from Au 3+ to Au 0 . The electrochemical reduction of Au nano on the electrode surface would be an outstanding active site for the reduction of Au 3+ to Au 0 . The CV electroanalysis results for the mixture of GO and HAuCl 4 are shown in Figure 1C, and these results were completely different from those observed for the GO electroanalysis. One significant difference is that the currents from scanning at a positive potential to scanning at a negative potential in the range of −1.2 V to −1.4 V for the mixture of GO and HAuCl 4 were dramatically larger than those obtained the GO electroanalysis, indicating the continuous electrodeposition of Au nano , which is more conductive than rGO, onto the electrode surface. ultrasonic-assisted extraction method and the method proposed by the European Communities Bureau of Reference.

Electrochemical Deposition of the rGO-Aunano Composite
The CV electroanalysis curves of GO on a GCE can be seen in Figure 1A, and one oxidation peak (a) and two reduction peaks (b and c) can be observed. The peak currents increased with successive potential scans, demonstrating that GO was successfully electrochemically reduced on the electrode surface. The current peak (c) was due to the reduction of the oxygen-containing groups on GO [25][26][27]. The peak currents that appeared at positions "a" and "b" were due to the redox reaction of some redox pairs on the graphene plane, and peak "c" was attributed to the irreversible electrochemical reduction of GO [28].
The CV curve for the electroanalysis of 0.1 mM HAuCl4 is shown in Figure 1B. A significant reduction peak is present at approximately 0.18 V, indicating that the gold salt was reduced from Au 3+ to Au 0 . The electrochemical reduction of Aunano on the electrode surface would be an outstanding active site for the reduction of Au 3+ to Au 0 . The CV electroanalysis results for the mixture of GO and HAuCl4 are shown in Figure 1C, and these results were completely different from those observed for the GO electroanalysis. One significant difference is that the currents from scanning at a positive potential to scanning at a negative potential in the range of −1.2 V to −1.4 V for the mixture of GO and HAuCl4 were dramatically larger than those obtained the GO electroanalysis, indicating the continuous electrodeposition of Aunano, which is more conductive than rGO, onto the electrode surface.

Characteristics of the Modified Electrodes
As shown in Figure 2, SEM was used to characterize the morphology of the rGO-Aunano sensing interface formed by electrochemical reduction. SEM imaging of the rGO-Aunano deposited on a GCE showed a large amount of Aunano evenly distributed on the substrate, which had a wrinkled texture associated with the presence of rGO sheets. The uniform distribution of Aunano with an average diameter of ~30 nm indicated that rGO was formed with Aunano by the electrochemical deposition, which effectively promoted the specific surface area of the sensing interface and allowed a large amount of Aunano to be deposited on the electrode surface.

Characteristics of the Modified Electrodes
As shown in Figure 2, SEM was used to characterize the morphology of the rGO-Au nano sensing interface formed by electrochemical reduction. SEM imaging of the rGO-Au nano deposited on a GCE showed a large amount of Au nano evenly distributed on the substrate, which had a wrinkled texture associated with the presence of rGO sheets. The uniform distribution of Au nano with an average diameter of~30 nm indicated that rGO was formed with Au nano by the electrochemical deposition, which effectively promoted the specific surface area of the sensing interface and allowed a large amount of Au nano to be deposited on the electrode surface.

Characteristics of the Modified Electrodes
As shown in Figure 2, SEM was used to characterize the morphology of the rGO-Aunano sensing interface formed by electrochemical reduction. SEM imaging of the rGO-Aunano deposited on a GCE showed a large amount of Aunano evenly distributed on the substrate, which had a wrinkled texture associated with the presence of rGO sheets. The uniform distribution of Aunano with an average diameter of ~30 nm indicated that rGO was formed with Aunano by the electrochemical deposition, which effectively promoted the specific surface area of the sensing interface and allowed a large amount of Aunano to be deposited on the electrode surface.   Figure 3, XRD analysis was carried out, and the XRD pattern was analyzed to investigate the monocrystalline nature of the nanoparticles formed by the electrochemical reduction of the gold salt. Four diffraction peaks appeared at 2θ = 38.2 • , 44.5 • , 65.6 • and 78.6 • in the XRD pattern, which were consistent with the (111), (200), (220) and (311) planes, respectively, of gold metal (International Center for Diffraction Data, ICDD No. 4-0783), suggesting that the product synthesized by electrodeposition contains pure crystalline gold. The Bragg reflection peaks of Au nano clearly indicate a face-centered cubic monocrystalline structure, which is consistent with the crystalline structure of gold. The XRD pattern shows a very high Bragg reflection peak corresponding to the (111) lattice plane, indicating that Au nano related to the (111) lattice plane was flat [29] on the planar surface. clearly indicate a face-centered cubic monocrystalline structure, which is consistent with the crystalline structure of gold. The XRD pattern shows a very high Bragg reflection peak corresponding to the (111) lattice plane, indicating that Aunano related to the (111) lattice plane was flat [29] on the planar surface. The electron transfer characteristics of the surfaces of different electrodes were studied by CV using [Fe(CN)6] 3−/4− redox probes, as shown in Figure 4A. The [Fe(CN)6] 3−/4− redox currents on rGO/GCE were larger than those on the bare GCE. Additionally, the potential difference between the oxidation peak and reduction peak of the [Fe(CN)6] 3−/4− redox probe decreased because of the good electron transfer capability of rGO. The redox currents of the redox probes were further enhanced on rGO-Aunano/GCE compared with those on bare GCE and rGO/GCE, which demonstrated that the presence of Aunano in the rGO-Aunano composite effectively improves the electron transfer capability.

As shown in
In this study, EIS, a common electrochemical characterization method that consists of low-and high-frequency regions corresponding to semicircular and linear portions, respectively, was also used to further investigate the electron transfer kinetics of the different electrodes, as shown in Figure 4B. Additionally, an equivalent circuit was used to fit the EIS data, as shown in Figure 4B, where Rct, Rs and CDL represent the charge transfer resistance, resistance of solution and double layer capacitance, respectively. Among the three parameters, Rct is the most important key factor and is widely used for the characterization of electron transfer kinetics on the electrode surface. The semicircular and linear portions represent the Rct and the surface diffusion process, respectively, which reflects the electron transfer kinetics of [Fe(CN)6] 3−/4− at the electrode interface [30]. A smaller, well-defined semicircle was observed in the high-frequency region with rGO/GCE than in the region of bare GCE, which suggested that the presence of rGO reduced the impedance of the electrode interface [31]. The smallest Rct value was observed with rGO-Aunano/GCE because the corresponding diameter of the semicircle was the smallest, which suggested that the presence of Aunano in the rGO layer can effectively improve the electron transfer kinetics of the electrode surface. The first EIS point of all the electrodes in the low-frequency region was almost the same, which indicated that Rs was the same for all the electrodes. These results were in good agreement with the CV results.

Optimization of the Experimental Conditions
In this study, the number of sweep segments, deposition time and deposition potential were investigated to determine the optimal experimental conditions for rGO-Aunano/GCE that result in the highest detection sensitivity of As(III). The effect of deposition time on the stripping response was investigated in a nitric acid solution (0.1 M). The thickness of the rGO-Aunano sensing composite, which could have a significant effect on the sensitivity, can be controlled by changing the number of electrodeposition cycles, i.e., the number of sweep segments. In this work, 4 to 18 sweep segments were tested to optimize the thickness of the sensing film, and the results can be seen in Figure 5A, which shows the As(III) stripping response with respect to the number of electrodeposition sweep segments. The stripping response of As(III) significantly increased with the number of sweep segments in the range from 4 to 10. This increase may be because the GCE surface was not saturated with rGO-Aunano until the number of sweep segments reached 10. As the number of sweep segments was increased above 10, the stripping response of As(III) remained nearly constant. Thus, 10 sweep segments were used for further experiments.
As shown in Figure 5B, the deposition potential was optimized for the As(III) stripping response in the potential range from −0.1 to −0.6 V with a deposition time of 250 s. At a deposition potential of −0.3 V, the highest peak currents were obtained; thus, a deposition potential of −0.3 V was used for further analyses.
Additionally, the effect of the deposition time on the stripping response was investigated in the range from 10 to 450 s, as shown in Figure 5C. The stripping signals of As(III) sharply increased as the deposition time increased. Finally, a deposition time of 250 s was chosen for the following measurements as a compromise between sensitivity and analysis time. In this study, EIS, a common electrochemical characterization method that consists of low-and high-frequency regions corresponding to semicircular and linear portions, respectively, was also used to further investigate the electron transfer kinetics of the different electrodes, as shown in Figure 4B. Additionally, an equivalent circuit was used to fit the EIS data, as shown in Figure 4B, where Rct, Rs and C DL represent the charge transfer resistance, resistance of solution and double layer capacitance, respectively. Among the three parameters, Rct is the most important key factor and is widely used for the characterization of electron transfer kinetics on the electrode surface. The semicircular and linear portions represent the Rct and the surface diffusion process, respectively, which reflects the electron transfer kinetics of [Fe(CN) 6 ] 3−/4− at the electrode interface [30]. A smaller, well-defined semicircle was observed in the high-frequency region with rGO/GCE than in the region of bare GCE, which suggested that the presence of rGO reduced the impedance of the electrode interface [31]. The smallest R ct value was observed with rGO-Au nano /GCE because the corresponding diameter of the semicircle was the smallest, which suggested that the presence of Au nano in the rGO layer can effectively improve the electron transfer kinetics of the electrode surface. The first EIS point of all the electrodes in the low-frequency region was almost the same, which indicated that Rs was the same for all the electrodes. These results were in good agreement with the CV results.

Optimization of the Experimental Conditions
In this study, the number of sweep segments, deposition time and deposition potential were investigated to determine the optimal experimental conditions for rGO-Au nano /GCE that result in the highest detection sensitivity of As(III). The effect of deposition time on the stripping response was investigated in a nitric acid solution (0.1 M). The thickness of the rGO-Au nano sensing composite, which could have a significant effect on the sensitivity, can be controlled by changing the number of electrodeposition cycles, i.e., the number of sweep segments. In this work, 4 to 18 sweep segments were tested to optimize the thickness of the sensing film, and the results can be seen in Figure 5A, which shows the As(III) stripping response with respect to the number of electrodeposition sweep segments. The stripping response of As(III) significantly increased with the number of sweep segments in the range from 4 to 10. This increase may be because the GCE surface was not saturated with rGO-Au nano until the number of sweep segments reached 10. As the number of sweep segments was increased above 10, the stripping response of As(III) remained nearly constant. Thus, 10 sweep segments were used for further experiments.
In this study, the number of sweep segments, deposition time and deposition potential were investigated to determine the optimal experimental conditions for rGO-Aunano/GCE that result in the highest detection sensitivity of As(III). The effect of deposition time on the stripping response was investigated in a nitric acid solution (0.1 M). The thickness of the rGO-Aunano sensing composite, which could have a significant effect on the sensitivity, can be controlled by changing the number of electrodeposition cycles, i.e., the number of sweep segments. In this work, 4 to 18 sweep segments were tested to optimize the thickness of the sensing film, and the results can be seen in Figure 5A, which shows the As(III) stripping response with respect to the number of electrodeposition sweep segments. The stripping response of As(III) significantly increased with the number of sweep segments in the range from 4 to 10. This increase may be because the GCE surface was not saturated with rGO-Aunano until the number of sweep segments reached 10. As the number of sweep segments was increased above 10, the stripping response of As(III) remained nearly constant. Thus, 10 sweep segments were used for further experiments.
As shown in Figure 5B, the deposition potential was optimized for the As(III) stripping response in the potential range from −0.1 to −0.6 V with a deposition time of 250 s. At a deposition potential of −0.3 V, the highest peak currents were obtained; thus, a deposition potential of −0.3 V was used for further analyses.
Additionally, the effect of the deposition time on the stripping response was investigated in the range from 10 to 450 s, as shown in Figure 5C. The stripping signals of As(III) sharply increased as the deposition time increased. Finally, a deposition time of 250 s was chosen for the following measurements as a compromise between sensitivity and analysis time.  As shown in Figure 5B, the deposition potential was optimized for the As(III) stripping response in the potential range from −0.1 to −0.6 V with a deposition time of 250 s. At a deposition potential of −0.3 V, the highest peak currents were obtained; thus, a deposition potential of −0.3 V was used for further analyses.
Additionally, the effect of the deposition time on the stripping response was investigated in the range from 10 to 450 s, as shown in Figure 5C. The stripping signals of As(III) sharply increased as the deposition time increased. Finally, a deposition time of 250 s was chosen for the following measurements as a compromise between sensitivity and analysis time.

Stripping Responses of Different Electrodes
The stripping responses of different electrodes, i.e., a bare GCE, rGO/GCE, Au nano /GCE and rGO-Au nano /GCE, were investigated under the optimum conditions. As shown in Figure 6, As(III) produced no obvious response on bare GCE and rGO/GCE. However, a significant As(III) response can be observed on Au nano /GCE. A higher and sharper As(III) stripping response was obtained on rGO-Au nano /GCE. The enhanced stripping response of rGO-Au nano /GCE can be attributed to the following: (1) the electrodeposition of solution-state As(III) was enhanced due to the electrogeneration of H 2 near Au nano by chemical reduction [11]; (2) rGO increased the specific surface area and conductivity of the sensing interface; and (3) the combined effect of Au nano and rGO synthesized by electrochemical reduction imparted a very high sensitivity for As(III) detection [14]. The results show that the rGO-Au nano sensing interface-modified GCE has outstanding sensitivity and is highly promising for application.
following: (1) the electrodeposition of solution-state As(III) was enhanced due to the electrogeneration of H2 near Aunano by chemical reduction [11]; (2) rGO increased the specific surface area and conductivity of the sensing interface; and (3) the combined effect of Aunano and rGO synthesized by electrochemical reduction imparted a very high sensitivity for As(III) detection [14]. The results show that the rGO-Aunano sensing interface-modified GCE has outstanding sensitivity and is highly promising for application.

Analytical Performance of rGO-Aunano/GCE
Under the optimal conditions, calibration curves for the stripping response of As(III) at both common and trace level concentrations were established. A series of SWASV As(III) responses and the corresponding calibration plots in the concentration range from 1 to 50 μg/L are shown in Figure  7A,B, respectively. The calibration curve and correlation coefficient were y = 1.521x + 7.89 (x: μg/L, y: μA) and 0.992, respectively. A significant and good linear relationship between the stripping peak current and the concentration of As(III) was obtained between 1.0 and 50.0 μg/L with a sensitivity of 1.521 μAμg/L. The SWASV As(III) responses in a lower concentration range, i.e., 0.1-1.2 μg/L, are shown in Figure 8A. The calibration curve and correlation coefficient were y = 10.597x − 0.401 (x: μg/L, y: μA) and 0.985, respectively. A detection limit of 0.08 μg/L was obtained according to the formula LOD=3 σ/S, where σ is the standard deviation of a blank sample tested 10 times and S is the slope of the corresponding linear calibration equation of y = 1.521x + 7.89. The responses indicate that a higher sensitivity and lower LOD can be obtained by increasing the deposition time from 250 s to 800 s ( Figure 8B). The specific stripping potential of As(III) changed as the As concentration changed from low to high. The peak shifting may be explained by the oxidation of additional As deposited on the electrode surface at a low positive potential with a high concentration of As(III). With a low concentration of As(III), only a trace amount of As was deposited on the electrode surface and was difficult to oxidize at a low positive potential; thus, the peak appeared at a more

Analytical Performance of rGO-Au nano /GCE
Under the optimal conditions, calibration curves for the stripping response of As(III) at both common and trace level concentrations were established. A series of SWASV As(III) responses and the corresponding calibration plots in the concentration range from 1 to 50 µg/L are shown in Figure 7A,B, respectively. The calibration curve and correlation coefficient were y = 1.521x + 7.89 (x: µg/L, y: µA) and 0.992, respectively. A significant and good linear relationship between the stripping peak current and the concentration of As(III) was obtained between 1.0 and 50.0 µg/L with a sensitivity of 1.521 µAµg/L. The SWASV As(III) responses in a lower concentration range, i.e., 0.1-1.2 µg/L, are shown in Figure 8A. The calibration curve and correlation coefficient were y = 10.597x − 0.401 (x: µg/L, y: µA) and 0.985, respectively. A detection limit of 0.08 µg/L was obtained according to the formula LOD = 3 σ/S, where σ is the standard deviation of a blank sample tested 10 times and S is the slope of the corresponding linear calibration equation of y = 1.521x + 7.89. The responses indicate that a higher sensitivity and lower LOD can be obtained by increasing the deposition time from 250 s to 800 s ( Figure 8B). The specific stripping potential of As(III) changed as the As concentration changed from low to high. The peak shifting may be explained by the oxidation of additional As deposited on the electrode surface at a low positive potential with a high concentration of As(III). With a low concentration of As(III), only a trace amount of As was deposited on the electrode surface and was difficult to oxidize at a low positive potential; thus, the peak appeared at a more positive potential [32]. A comparison of the major analytical properties obtained in this work and those from previous work is shown in Table 1. Table 1 shows that the analytical performance of rGO-Au nano /GCE is comparable and even better than that in previous reports, and this electrode is more controllable and easier to prepare and use and offers a lower limit of detection than other previously reported electrodes.
positive potential [32]. A comparison of the major analytical properties obtained in this work and those from previous work is shown in Table 1. Table 1 shows that the analytical performance of rGO-Aunano/GCE is comparable and even better than that in previous reports, and this electrode is more controllable and easier to prepare and use and offers a lower limit of detection than other previously reported electrodes.    positive potential [32]. A comparison of the major analytical properties obtained in this work and those from previous work is shown in Table 1. Table 1 shows that the analytical performance of rGO-Aunano/GCE is comparable and even better than that in previous reports, and this electrode is more controllable and easier to prepare and use and offers a lower limit of detection than other previously reported electrodes.

Stability Measurements
As shown in Figure 9, to verify the suitability and reproducibility of rGO-Au nano /GCE for As(III) determination, eight SWASV responses to 5 µg/L As(III) were obtained as continuous repetitive measurements. The times for each step were 250 s, 10 s, 4.8 s and 120 s for the deposition, standing, stripping and cleaning steps, respectively. Thus, the total working time of the electrode in the solution for each measurement should be 384.8 s. The relative standard deviation of the eight repetitive measurements of As(III) was 1.178%, which is less than 20%, demonstrating the good reusability and stability of rGO-Au nano /GCE for repeated SWASV measurements of As(III) under optimal conditions. butanethiol, PANI: polyaniline.

Stability Measurements
As shown in Figure 9, to verify the suitability and reproducibility of rGO-Aunano/GCE for As(III) determination, eight SWASV responses to 5 μg/L As(III) were obtained as continuous repetitive measurements. The times for each step were 250 s, 10 s, 4.8 s and 120 s for the deposition, standing, stripping and cleaning steps, respectively. Thus, the total working time of the electrode in the solution for each measurement should be 384.8 s. The relative standard deviation of the eight repetitive measurements of As(III) was 1.178%, which is less than 20%, demonstrating the good reusability and stability of rGO-Aunano/GCE for repeated SWASV measurements of As(III) under optimal conditions. Figure 9. Eight repeated SWASV measurements of 5 μg/L As(III) with rGO-Aunano/GCE. The insets correspond to the data collected from every SWASV curve for a total of eight measurements. RSD refers to the relative standard deviation.

Interference Studies
The influence of other ions, such as Na + , K + , Fe 2+ , Mn 2+ , Zn 2+ , Mg 2+ , Ca 2+ , Pb 2+ and Cd 2+ , was also investigated in a synthetic solution containing 100-fold higher concentrations of the above nontarget ions than of As to estimate the possible influence of interference on the stripping response of trace As(III) under optimal conditions, as shown in Figure 10. The results demonstrated that the addition of nontarget ions had no significant effect on the As(III) stripping responses of the proposed electrode because the current mean values obtained in the presence of nontarget ions were within the ±2 × standard deviation range of the current measured in the absence of nontarget ions. Figure 9. Eight repeated SWASV measurements of 5 µg/L As(III) with rGO-Au nano /GCE. The insets correspond to the data collected from every SWASV curve for a total of eight measurements. RSD refers to the relative standard deviation.

Interference Studies
The influence of other ions, such as Na + , K + , Fe 2+ , Mn 2+ , Zn 2+ , Mg 2+ , Ca 2+ , Pb 2+ and Cd 2+ , was also investigated in a synthetic solution containing 100-fold higher concentrations of the above nontarget ions than of As to estimate the possible influence of interference on the stripping response of trace As(III) under optimal conditions, as shown in Figure 10. The results demonstrated that the addition of nontarget ions had no significant effect on the As(III) stripping responses of the proposed electrode because the current mean values obtained in the presence of nontarget ions were within the ±2 × standard deviation range of the current measured in the absence of nontarget ions.

Application to Real Sample Analysis
To investigate the practical application of the proposed sensor for the determination of As(III), several soil samples were analyzed under optimal conditions. Hydride generation atomic fluorescence spectrometry (HG-AFS) was used to verify the accuracy of the prepared electrochemical platform. The pretreatment process for the soil samples is described in Section 2.4, and the analysis was carried out based on a standard addition method. The average recovery of As(III) was 98.97%, which demonstrated that the detection results obtained using the proposed platform were consistent with those obtained by HG-AFS (Table 2). To further verify the significant difference in accuracy between the proposed platform and HG-AFS, a paired t-test at the 95% Figure 10. Interference study on the stripping peak currents of 15 µg/L As(III) in the presence of 10-fold higher concentrations of different ions.

Application to Real Sample Analysis
To investigate the practical application of the proposed sensor for the determination of As(III), several soil samples were analyzed under optimal conditions. Hydride generation atomic fluorescence spectrometry (HG-AFS) was used to verify the accuracy of the prepared electrochemical platform. The pretreatment process for the soil samples is described in Section 2.4, and the analysis was carried out based on a standard addition method. The average recovery of As(III) was 98.97%, which demonstrated that the detection results obtained using the proposed platform were consistent with those obtained by HG-AFS (Table 2). To further verify the significant difference in accuracy between the proposed platform and HG-AFS, a paired t-test at the 95% confidence level was performed, and the results are presented in Table 2. Based on the statistical analysis results, we easily concluded that there was no significant difference between the two techniques because the t calculated of As(III) was below t critical (4.30 at 2 degrees of freedom). Therefore, the proposed modified electrode can be used for the analysis of As(III) at trace levels in real samples.

Conclusions
In this study, rGO-Au nano /GCE was created by a one-step electrodeposition process with CV and used for the sensitive detection of As(III). The rGO-Au nano /GCE sensor exhibits a better limit of detection than other Au nano -based modified electrodes and is more controllable and stable. Additionally, the morphology, physical and chemical properties of the fabricated rGO-Au nano /GCE were characterized by SEM, XRD, SWASV, EIS and CV. Parameters such as the deposition time, number of sweep cycles and deposition potential were optimized. The proposed rGO-Au nano /GCE sensor shows very high sensitivity for the determination of As(III) based on the combined effects of rGO and Au nano. The rGO produced during electrodeposition effectively increases the specific surface area and electron transfer capability of the electrode. Additionally, rGO prevents the agglomeration of Au nano , allowing good dispersal of Au nano on the electrode surface and further improving the catalytic performance. A satisfactory recovery of As(III) in a soil sample analysis was obtained, which demonstrated that the rGO-Au nano /GCE sensor may have potential promising applications for As(III) monitoring of environmental and food samples.