Electrodeposited rGO/AuNP/MnO2 Nanocomposite-Modified Screen-Printed Carbon Electrode for Sensitive Electrochemical Sensing of Arsenic(III) in Water

Herein, a simple and portable electrochemical sensor based on a reduced graphene oxide/gold nanoparticle/manganese dioxide (rGO/AuNP/MnO2) nanocomposite-modified screen-printed carbon electrode (SPCE) was constructed by the facile stepwise electrodeposition method and used for electrochemical detection of As(III). The resultant electrode was characterized for its morphological, structural, and electrochemical properties using scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), energy dispersive X-ray spectroscopy (EDX), cyclic voltammetry (CV), and electrochemical impedance spectroscopy (EIS). From the morphologic structure, it can be clearly observed that the AuNPs and MnO2 alone or their hybrid were densely deposited or entrapped in thin rGO sheets on the porous carbon surface, which may favor the electro-adsorption of As(III) on the modified SPCE. It is interesting that the nanohybrid modification endows the electrode with a significant decrease in charge transfer resistance and an increase in electroactive specific surface area, which dramatically increases the electro-oxidation current of As(III). This improved sensing ability was ascribed to the synergistic effect of gold nanoparticles with excellent electrocatalytic property and reduced graphene oxide with good electrical conductivity, as well as the involvement of manganese dioxide with a strong adsorption property in the electrochemical reduction of As(III). Under optimized conditions, the sensor can detect As(III) via square wave anodic stripping voltammetry (SWASV) with a low limit of detection of 2.4 μg L−1 and a linear range of 25–200 μg L−1. The proposed portable sensor shows the advantages of a simple preparation procedure, low cost, good repeatability, and long-term stability. The feasibility of rGO/AuNPs/MnO2/SPCE for detecting As(III) in real water was further verified.


Introduction
Arsenic (As) is a highly toxic metalloid and is widely distributed throughout the environment in water and soil. Due to the widespread use of arsenic-containing compounds in agricultural and industrial processes, elevated levels of inorganic arsenic including As(III) and As(V) are found in contaminated groundwater and drinking water [1]. Especially, as the most stable and toxic form, As(III), has a stronger capacity for cellular uptake and accumulation than As(V) in the process of arsenic poisoning. In particular, its trivalent metabolites have a high affinity for sulfhydryl groups of proteins during the methylation detoxification of arsenic, leading to severe damage to cells and tissues [2]. Long-term ingestion of arsenic, mainly through contaminated drinking water or food, will lead to serious skin diseases, organ, and nervous system damage, and even carcinogenesis, thus posing a great threat to public health. The International Agency for Research on Cancer (IARC) has officially classified arsenic as a first-class carcinogen to humans. The current doping of graphitic defects, the selectivity of SPCE can be improved since the electrode substrate becomes less interactive and the electrocatalytic response toward analytes is ignorable [31]. Thereby, the fabrication of a sensitive and stable surface modification layer on the SPCE is highly required for the wide application of a rapid detection method for As(III).
In view of the need for a sensitive, stable, and low-cost sensor for arsenite monitoring in waters, herein we reported an in situ electrodeposition strategy of rGO/AuNP/MnO 2 nanocomposites on the surface of SPCE for As(III) detection by square wave anodic stripping voltammetry (SWASV). The morphology and electrochemical behavior of the fabricated electrode were investigated. The proposed sensor was thoroughly examined to optimize the preparation and voltametric analytical parameters, as well as to evaluate the interference and reproducibility. In addition, the applicability of the sensor in the detection of As(III) in water samples was studied.

Fabrication of the Nanocomposite-Modified Electrode
As illustrated in Figure 1, the rGO/AuNP/MnO 2 nanocomposite-modified SPCE was fabricated via a step-by-step electrochemical deposition. Firstly, rGO was electrodeposited onto SPCE in 1.0 mg mL −1 graphene oxide dispersion (0.1 mol L −1 CBS (pH 9.0)) for eight cycles at a scan rate of 100 mV s −1 in the potential window of −1.5 V to 0.5 V. Then, the rGO/SPCE was immersed in 1.0 mmol L −1 HAuCl 4 (0.2 mol L −1 Na 2 SO 4 ) solution and deposited at a constant potential of −0.2 V for 90 s to obtain rGO/AuNPs/SPCE. Further, MnO 2 was deposited onto rGO/AuNPs/SPCE in 0.05 mol L −1 Mn(CH 3 COO) 2 (0.1 mol L −1 Na 2 SO 4 ) at different deposition potentials (from −1.5 V to 0 V) and a scan rate of 50 mV s −1 for two cycles. Finally, the as-prepared rGO/AuNP/MnO 2 -modified SPCE was cleaned with ultrapure water and dried in N 2 flow. For comparison, rGO/SPCE and rGO/AuNPs/SPCE were prepared, respectively, according to the above-mentioned electrode modification procedure.
The electrochemical sensing behavior of As(III) was studied with SWASV under optimized conditions. In SWASV, As(III) species were initially reduced to As(0) on the electrode surface at −1.4 V for 180 s. The anodic stripping of electrodeposited As(0) was performed in 0.01 mol L −1 H2SO4 by a potential scan from −0.7 to 0.2 V at a potential step of 4 mV, an amplitude of 25 mV, and a frequency of 25 Hz. After each stripping step, the rGO/AuNPs/MnO2/SPCE was renewed in 0.01 mol L −1 H2SO4 by applying a preconditioning potential of 0 V for 15 s to remove adsorbed As(0). In addition, three replicate experiments were carried out for each SWASV determination, and all data were collected and used to estimate the error, taking into account the differences between the experiment.

Detection of Water Samples
To verify the practicability of proposed sensor, various drinking water samples, including tap water, mountain spring water, and mineral water, were obtained from Fuzhou University and Wanjia Supermarket (Fuzhou University, Fuzhou, China), respectively, and their basic chemical composition is shown in Table S1. The water samples were filtered through a 0.22 µm filter membrane. These samples were mixed with 0.01 mol L −1 H2SO4 in a volume ratio of 10:1 and determined by SWASV, respectively. Water samples
The electrochemical sensing behavior of As(III) was studied with SWASV under optimized conditions. In SWASV, As(III) species were initially reduced to As(0) on the electrode surface at −1.4 V for 180 s. The anodic stripping of electrodeposited As(0) was performed in 0.01 mol L −1 H 2 SO 4 by a potential scan from −0.7 to 0.2 V at a potential step of 4 mV, an amplitude of 25 mV, and a frequency of 25 Hz. After each stripping step, the rGO/AuNPs/MnO 2 /SPCE was renewed in 0.01 mol L −1 H 2 SO 4 by applying a preconditioning potential of 0 V for 15 s to remove adsorbed As(0). In addition, three replicate experiments were carried out for each SWASV determination, and all data were collected and used to estimate the error, taking into account the differences between the experiment.

Detection of Water Samples
To verify the practicability of proposed sensor, various drinking water samples, including tap water, mountain spring water, and mineral water, were obtained from Fuzhou University and Wanjia Supermarket (Fuzhou University, Fuzhou, China), respectively, and their basic chemical composition is shown in Table S1. The water samples were filtered through a 0.22 µm filter membrane. These samples were mixed with 0.01 mol L −1 H 2 SO 4 in a volume ratio of 10:1 and determined by SWASV, respectively. Water samples spiked by three concentration levels of As(III) (50.0, 100.0, 150.0 µg L −1 ) were used for recovery experiments.

Preparation and Characterization of Nanocomposite-Modified Electrodes
Figure S1A-C show the cyclic voltammograms associated with the preparation of rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO 2 /SPCE by stepwise electrodeposition, respectively. For rGO/SPCE, when the potential scanned from 0.5 to −1.5 V, a large reduction peak was observed at −1.1 V with a starting potential of −0.6 V, which can be ascribed to the reduction of oxygen-containing functional groups in GO (-OH, -COOH, and epoxides) [32]. In the reverse scanning, a small oxidation peak appears at −0.8 V, referring to the re-oxidation of reduced graphene. When the rGO/SPCE was potentiostated at −0.2 V in 0.2 mol L −1 Na 2 SO 4 containing 1.0 mmol L −1 HAuCl 4 , the deposition of AuNPs on the electrode surface occurred and was completed at 90 s. As shown in Figure S1C, an enhancement of MnO 2 deposition on the rGO/AuNPs/SPCE through a series of reactions was observed. In this case, an oxidation peak at −0.2 V can be attributed to the electrochemical oxidative deposition of Mn 2+ to MnO 2 [33,34], while the reduction peaks at −0.5 V and −1.3 V correspond to the re-reduction of MnO 2 to Mn(OH) 2 and the further conversion of Mn(OH) 2 to Mn 2+ (Equations (1)-(4)).
The morphologies of bare SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/ MnO 2 /SPCE were investigated using SEM. As shown in Figure 2A, a porous structure with the flake-like morphology of carbon was observed on the bare SPCE. Figure 2B depicts the appearance of smooth and flexible ultra-thin sheets with slightly folded wrinkles on the porous carbon surface, indicating the reduction of GO to rGO. After deposition, uniformly and intensely dispersed AuNPs with an average diameter of~50 nm can be clearly observed on the surface of rGO sheets, as shown in Figure 2C, demonstrating an improvement in the dispersion and high loading density of AuNPs. The SEM image of further electrochemically deposited MnO 2 on the rGO/AuNPs/SPCE reveals a wall network-like structure and randomly dispersed larger nanoparticles (100~150 nm), sometimes appearing to be rods on the rGO sheet, as well as some particles in flower-like shapes, which is attributed to their combination with AuNPs ( Figure 2D). The grain analysis of the SEM micrographs was further performed through statistical parameters, using the SEM image processing software Image-Pro Plus. The results show that the mean size of AuNPs ranging from 40 to 65 nm, and MnO 2 ranging from 110 to 140 nm, are indicative of fine to medium size. A high-resolution SEM image of the electrodeposit-modified SPCE (enlarged partial view in the Figure 2D) also indicates that both AuNPs and MnO 2 nanoparticles, as well as their flower-like composites, are well anchored onto flaky rGO on the porous carbon surface, which is similar to the morphologies observed in related Au/rGO or MnO x /rGO-modified electrodes [35,36]. In addition, Figure S2 portrays the energy dispersive X-ray spectrum (EDS) for rGO/AuNPs/MnO 2 /SPCE. The results reveal that the nanocomposite material deposited on the SPCE mainly consists of C, O, Au, and Mn elements, which further confirms the successful fabrication of rGO/AuNPs/MnO 2 /SPCE. Moreover, XPS was used to investigate the composition of the rGO/AuNP/MnO 2 nanocomposites. As shown in Figure S3, the XPS full spectra showed peaks at 638. 31  CV and EIS were employed to characterize the electrochemical performance of the bare and nanocomposite-modified electrodes in a 5 mmol L −1 [Fe (CN)6] 3−/4− solution containing 0.1 mol L −1 KCl. As shown in Figure 3A, [Fe (CN)6] 3−/4− as the electrochemical marker, generated a pair of oxidation/reduction peaks at all electrodes with peak potential separation (ΔEP) larger than 120 mV and the peak current ratio (Ipa/Ipc) of about one, indicating a quasi-reversible oxidation/reduction process of the Fe 2+ /Fe 3+ species. In addition, the peak potential differences in bare SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO2/SPCE gradually decreased from 227 to 136 mV, demonstrating that the deposition of rGO/AuNP/MnO2 nanocomposites on SPCE did facilitate a faster electron transfer than other nanomaterials toward the electrode in solution. Furthermore, it was found that the peak current of Fe 2+ /Fe 3+ species obtained from rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO2/SPCE increased successively, as compared to that of bare SPCE, which can be attributed to a larger active surface area offered by the modified nanomaterials on the SPCE. The results also show a drastic acceleration of the electron transfer rate between the nanocomposite electrode surface and the solution.  Figure 3A, [Fe (CN) 6 ] 3−/4− as the electrochemical marker, generated a pair of oxidation/reduction peaks at all electrodes with peak potential separation (∆E P ) larger than 120 mV and the peak current ratio (I pa /I pc ) of about one, indicating a quasi-reversible oxidation/reduction process of the Fe 2+ /Fe 3+ species. In addition, the peak potential differences in bare SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO 2 /SPCE gradually decreased from 227 to 136 mV, demonstrating that the deposition of rGO/AuNP/MnO 2 nanocomposites on SPCE did facilitate a faster electron transfer than other nanomaterials toward the electrode in solution. Furthermore, it was found that the peak current of Fe 2+ /Fe 3+ species obtained from rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO 2 /SPCE increased successively, as compared to that of bare SPCE, which can be attributed to a larger active surface area offered by the modified nanomaterials on the SPCE. The results also show a drastic acceleration of the electron transfer rate between the nanocomposite electrode surface and the solution.  CV and EIS were employed to characterize the electrochemical performance of the bare and nanocomposite-modified electrodes in a 5 mmol L −1 [Fe (CN)6] 3−/4− solution containing 0.1 mol L −1 KCl. As shown in Figure 3A, [Fe (CN)6] 3−/4− as the electrochemical marker, generated a pair of oxidation/reduction peaks at all electrodes with peak potential separation (ΔEP) larger than 120 mV and the peak current ratio (Ipa/Ipc) of about one, indicating a quasi-reversible oxidation/reduction process of the Fe 2+ /Fe 3+ species. In addition, the peak potential differences in bare SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO2/SPCE gradually decreased from 227 to 136 mV, demonstrating that the deposition of rGO/AuNP/MnO2 nanocomposites on SPCE did facilitate a faster electron transfer than other nanomaterials toward the electrode in solution. Furthermore, it was found that the peak current of Fe 2+ /Fe 3+ species obtained from rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO2/SPCE increased successively, as compared to that of bare SPCE, which can be attributed to a larger active surface area offered by the modified nanomaterials on the SPCE. The results also show a drastic acceleration of the electron transfer rate between the nanocomposite electrode surface and the solution.  Figure S4 shows the relationship between the CV response of rGO/AuNPs/MnO 2 /SPCE and the scan rate. It can be observed that both the oxidation and reduction peak currents of [Fe (CN) 6 ] 3−/4− obtained at the electrode are proportional to the square root of the scan rate, indicating that the electrode reaction was controlled by the diffusion process. According to the Randles-Sevčik equation (Equation (5)): where I p represents the peak current (A), n is the electron transfer number, D is the diffusion coefficient of the redox probe in the solution (cm 2 s −1 ), v is the scan rate (V s −1 ), A stands for the electroactive surface area of the electrode (cm 2 ), and c is the concentration of redox probe (mol L −1 ). The electroactive surface area of rGO/AuNPs/MnO 2 /SPCE calculated from this equation is 0.11 cm 2 , which is significantly larger than that of bare SPCE (0.06 cm 2 ). These results indicate that the electrochemical performance of SPCE can be greatly enhanced by the electrodeposition of rGO/AuNPs/MnO 2 nanocomposites. The charge transfer resistance (R cti ) of bare SPCE and different nanomaterial-modified electrodes were also measured by EIS, depending on their linearity with the semicircle diameter in the Nyquist diagrams ( Figure 3B). The R ct of bare SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO 2 /SPCE is found to be 823.6 Ω, 543.6 Ω, 397.8 Ω, and 99.54 Ω, respectively, showing the most remarkable synergistic effect of rGO, AuNPs, and MnO 2 for promoting the electron transfer process.

Electrochemical Responses of rGO/AuNPs/MnO 2 /SPCE toward As(III)
The electrochemical behavior of As(III) on rGO/AuNPs/MnO 2 /SPCE in 0.01 mol L −1 H 2 SO 4 solution was examined using CV. As shown in Figure 4A, with the addition of As(III), an insignificant reduction peak was observed at −0.35 V (vs. Ag/AgCl), and a notable oxidation peak appeared at −0.05 V, which might be attributed to the electroreduction of As(III) to As(0) and the redissolution of As(0) to As(III) [37]. The obvious stripping response of As(0) at the nanocomposite electrode suggests the possible high-sensing sensitivity of As(III).   Figure S4 shows the relationship between the CV response of rGO/AuNPs/MnO2/SPCE and the scan rate. It can be observed that both the oxidation and reduction peak currents of [Fe (CN)6] 3−/4− obtained at the electrode are proportional to the square root of the scan rate, indicating that the electrode reaction was controlled by the diffusion process. According to the Randles-Sevčik equation (Equation (5)): Ip = (2.69 × 10 5 ) n 3/2 D 1/2 v 1/2 Ac (5) where Ip represents the peak current (A), n is the electron transfer number, D is the diffusion coefficient of the redox probe in the solution (cm 2 s −1 ), v is the scan rate (V s −1 ), A stands for the electroactive surface area of the electrode (cm 2 ), and c is the concentration of redox probe (mol L −1 ). The electroactive surface area of rGO/AuNPs/MnO2/SPCE calculated from this equation is 0.11 cm 2 , which is significantly larger than that of bare SPCE (0.06 cm 2 ). These results indicate that the electrochemical performance of SPCE can be greatly enhanced by the electrodeposition of rGO/AuNPs/MnO2 nanocomposites. The charge transfer resistance (Rcti) of bare SPCE and different nanomaterial-modified electrodes were also measured by EIS, depending on their linearity with the semicircle diameter in the Nyquist diagrams ( Figure 3B). The Rct of bare SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO2/SPCE is found to be 823.6 Ω, 543.6 Ω, 397.8 Ω, and 99.54 Ω, respectively, showing the most remarkable synergistic effect of rGO, AuNPs, and MnO2 for promoting the electron transfer process.

Electrochemical Responses of rGO/AuNPs/MnO2/SPCE toward As(III)
The electrochemical behavior of As(III) on rGO/AuNPs/MnO2/SPCE in 0.01 mol L −1 H2SO4 solution was examined using CV. As shown in Figure 4A, with the addition of As(III), an insignificant reduction peak was observed at −0.35 V (vs. Ag/AgCl), and a notable oxidation peak appeared at −0.05 V, which might be attributed to the electroreduction of As(III) to As(0) and the redissolution of As(0) to As(III) [37]. The obvious stripping response of As(0) at the nanocomposite electrode suggests the possible high-sensing sensitivity of As(III). Owing to its high sensitivity over CV, SWASV was employed to examine the analytical performance of rGO/AuNPs/MnO2/SPCE toward As(III). Figure 4B shows the SWASV diagrams of rGO/AuNPs/MnO2/SPCE for blank electrolyte solution (curve a) and 100 ppb As(III) (curve b) in a 0.01 mol L −1 H2SO4 solution. It is obvious that a dissolution peak appeared at near −0.05 V with the addition of As(III), which is assigned to the electrooxidation of deposited As (0) to As(III) species in the stripping step. Owing to its high sensitivity over CV, SWASV was employed to examine the analytical performance of rGO/AuNPs/MnO 2 /SPCE toward As(III). Figure 4B shows the SWASV diagrams of rGO/AuNPs/MnO 2 /SPCE for blank electrolyte solution (curve a) and 100 ppb As(III) (curve b) in a 0.01 mol L −1 H 2 SO 4 solution. It is obvious that a dissolution peak appeared at near −0.05 V with the addition of As(III), which is assigned to the electrooxidation of deposited As (0) to As(III) species in the stripping step.
To further evaluate the detection performance of rGO/AuNPs/MnO 2 /SPCE, SWASV responses of 100 ppb As(III) on different modified electrodes (SPCE, rGO/SPCE, rGO/ AuNPs/SPCE, and rGO/AuNPs/MnO 2 /SPCE) were recorded and compared. As shown in Figure S5, the observed stripping current increased in order by the type of composite (I SPCE < I rGO/SPCE < I rGO/AuNPs/SPCE < I rGO/AuNPs/MnO2/SPCE ), demonstrating the excellent response of rGO/AuNPs/MnO 2 /SPCE electrodes in detecting As(III). The remarkable enhancement of the rGO/AuNPs/MnO 2 nanocomposites toward As (III) may be attributed to the electrocatalytic properties of nanocomposites that derive from the synergetic effect of AuNPs, MnO 2 NPs, and rGO. It can be noticed that the reduction of GO to rGO on SPCE not only enlarged the surface area of the electrode, but also provided residual oxygenous active groups that can form a complex with As(III) and enhance detection sensitivity [38]. Further, the electrodeposition of AuNPs also leads to the formation of Au-As intermetallic compounds on the surface of nanocomposite-modified SPCE, thus notably promoting the redox process of As(III) [39]. Although the conductivity of the deposited MnO 2 layer is somewhat limited, its high adsorption capability and stability did contribute to the higher sensing performance of the nanocomposite electrode [40,41].
In addition, XPS was used to analyze the changes in the nanocomposite-modified electrode before and after the in situ electro-adsorption of As(III). As shown in Figure 5, the Mn2p peak almost completely disappeared after the electro-adsorption of As(III) at the rGO/AuNPs/MnO 2 /SPCE, while the Mn3s multiplet splitting peaks increased significantly, indicating a corresponding change in the valence state of the Mn element. These phenomena suggest that MnO 2 is directly involved in the electrochemical reduction step of As(III) and also acts as the center of electrocatalytic reaction rather than simple physical adsorption.
responses of 100 ppb As(III) on different modified electrodes (SPCE, rGO/SPCE, rGO/AuNPs/SPCE, and rGO/AuNPs/MnO2/SPCE) were recorded and compared. As shown in Figure S5, the observed stripping current increased in order by the type of composite (ISPCE < IrGO/SPCE < IrGO/AuNPs/SPCE < IrGO/AuNPs/MnO2/SPCE), demonstrating the excellent response of rGO/AuNPs/MnO2/SPCE electrodes in detecting As(III). The remarkable enhancement of the rGO/AuNPs/MnO2 nanocomposites toward As (III) may be attributed to the electrocatalytic properties of nanocomposites that derive from the synergetic effect of AuNPs, MnO2 NPs, and rGO. It can be noticed that the reduction of GO to rGO on SPCE not only enlarged the surface area of the electrode, but also provided residual oxygenous active groups that can form a complex with As(III) and enhance detection sensitivity [38]. Further, the electrodeposition of AuNPs also leads to the formation of Au-As intermetallic compounds on the surface of nanocomposite-modified SPCE, thus notably promoting the redox process of As(III) [39]. Although the conductivity of the deposited MnO2 layer is somewhat limited, its high adsorption capability and stability did contribute to the higher sensing performance of the nanocomposite electrode [40,41].
In addition, XPS was used to analyze the changes in the nanocomposite-modified electrode before and after the in situ electro-adsorption of As(III). As shown in Figure 5, the Mn2p peak almost completely disappeared after the electro-adsorption of As(III) at the rGO/AuNPs/MnO2/SPCE, while the Mn3s multiplet splitting peaks increased significantly, indicating a corresponding change in the valence state of the Mn element. These phenomena suggest that MnO2 is directly involved in the electrochemical reduction step of As(III) and also acts as the center of electrocatalytic reaction rather than simple physical adsorption. Therefore, the mechanism for electrochemical sensing of As(III) by rGO/AuNPs/MnO2/SPCE can be further speculated. Firstly, the integration of rGO as a sensitive layer matrix not only increases the surface area and conductivity of the electrode, but also significantly promotes the dispersion of AuNPs and MnO2. Secondly, the free As(III) in the solution can be enriched on the electrode surface either in the form of Au-As intermetallic compounds (via AuNPs) or through physical and chemical adsorption (via MnO2). Subsequently, the electrochemical reduction of As(III) to As(0), as well as the Therefore, the mechanism for electrochemical sensing of As(III) by rGO/AuNPs/MnO 2 / SPCE can be further speculated. Firstly, the integration of rGO as a sensitive layer matrix not only increases the surface area and conductivity of the electrode, but also significantly promotes the dispersion of AuNPs and MnO 2 . Secondly, the free As(III) in the solution can be enriched on the electrode surface either in the form of Au-As intermetallic compounds (via AuNPs) or through physical and chemical adsorption (via MnO 2 ). Subsequently, the electrochemical reduction of As(III) to As(0), as well as the anodic stripping of As(0), can be largely accelerated by the involvement of MnO 2 and AuNPs with excellent electrocatalytic properties on the nanocomposite electrode. In this case, the stripping peak current of As(III) will be significantly enhanced.

Optimization of Sensing Conditions toward As(III)
To obtain the maximum electrochemical response for As(III) detection with rGO/AuNPs/ MnO 2 /SPCE, several experimental parameters were optimized in 0.01 mol L −1 H 2 SO 4 using SWASV, including the cycles for electrodeposition of MnO 2 , the supporting electrolyte, deposition potential, and deposition time.
The influence of the cycles for electrodeposition of MnO 2 on the stripping current of arsenic species was studied. As shown in Figure S6A, the peak current was enhanced when the number of deposition cycles increased from 1 to 2, indicating that the initial deposition of MnO 2 NPs on rGO/AuNPs/SPCE increased the number of adsorption sites. However, a further increase in the number of deposition cycles resulted in an excessive deposition of MnO 2 on the surface of the nanocomposite electrodes, which will affect the conductivity of the electrode and decrease the peak current. Therefore, two cycles were selected for the deposition of MnO 2 . Figure S6B depicts the influence of H 2 SO 4 concentration on the stripping peak current of As(III) detection. It is clear that the magnitude of the peak current is determined by the concentration of the supporting electrolyte, and the maximal response was achieved in 0.01 mol L −1 H 2 SO 4 , due to the fact that the reduction of H 3 AsO 3 to As(0) can be promoted at an appropriate H + concentration. Whereas too high a concentration of H 2 SO 4 will lead to slower electron transfer kinetics, thereby reducing the peak current [42].
As shown in Figure S6C, the stripping current increased obviously when the deposition potential negative shifts from −1.2 to −1.4 V, which is attributed to the improved reduction of arsenite ions at a more negative potential. By a further negative shift in deposition potential, a sharp decrease in stripping current was found, suggesting the occurrence of hydrogen reduction, which will create bubbles and reduce the effective electrode surface area. In this case, an unstable response with larger background currents was observed. Therefore, −1.4 V was chosen as the deposition potential for further experiments. Figure S6D demonstrates the influence of deposition time ranging from 60 to 240 s on the peak current of As(III), which shows a gradual increase in current signal at 180 s. When the deposition time exceeded 180 s, the current was observed to be decreased, which is assigned to the surface saturation of the nanocomposite electrode by the As(0). Meanwhile, an obvious decrease in electrode stability and reproducibility was observed at higher deposition times. Therefore, 180 s was selected as the best deposition time to avoid interference. It is clear that the fabrication of the proposed nanocomposite electrode required a shorter deposition time than its other competing materials based on AuNPs and rGO/Fe 3 O 4 [38,43].

Analytical Performance of rGO/AuNPs/MnO 2 /SPCE for As(III) Detection
Under optimized SWASV sensing conditions, a calibration experiment was conducted, and the results are presented in Figure 6. As shown in Figure 6A, the stripping peak current of As(III) gradually increased with the increase in As(III) concentration. Figure 6B presents a good linear relationship between peak current and As(III) concentration in the range of 25-200 µg L −1 . The detection limit is calculated to be 2.4 µg L −1 (RSD 4.7%), based on the signal-to-noise of three, and the sensitivity of the proposed sensor is 0.03015 µA·µg −1 L (RSD 4.7%). As listed in Table 1, the rGO/AuNPs/MnO 2 /SPCE electrode exhibits good durability and rapid preparation in comparison with the other reported modified electrodes, and a satisfactory detection limit that was well below the WHO guidelines for drinking water. Most notable is that the proposed sensor provides a fast response, stable, low-cost, and field-deployable platform for real-time monitoring of As(III) in water.

Selectivity, Reproducibility, and Stability Measurement
To evaluate the selectivity of the proposed rGO/AuNPs/MnO 2 /SPCE sensor, 100 ppb As(III) was detected by SWASV in 0.01 mol L −1 H 2 SO 4 containing 10-foldother metal ions that might interfere with detection in the water sample, including Fe(III), Mn(II), Cr(III), Ni(II), Hg(II), Cd(II), Pb(II), Zn(II), and Cu(II). As can be seen from Figure 7, a not significant change in peak current was observed with the addition of Fe(III), Mn(II), and Cr(III). Although the presence of equivalent other interfering ions affected the peak current of As (III) due to peak overlapping or suppression; these interferences can be eliminated by adding a dose of EDTA or DTPA before water analysis.

Selectivity, Reproducibility, and Stability Measurement
To evaluate the selectivity of the proposed rGO/AuNPs/MnO2/SPCE sensor, 100 ppb As(III) was detected by SWASV in 0.01 mol L −1 H2SO4 containing 10-foldother metal ions that might interfere with detection in the water sample, including Fe(III), Mn(II), Cr(III), Ni(II), Hg(II), Cd(II), Pb(II), Zn(II), and Cu(II). As can be seen from Figure 7, a not significant change in peak current was observed with the addition of Fe(III), Mn(II), and Cr(III).
Although the presence of equivalent other interfering ions affected the peak current of As (III) due to peak overlapping or suppression; these interferences can be eliminated by adding a dose of EDTA or DTPA before water analysis.  Due to the inkjet printing preparation and modification of SPCE, it is necessary to evaluate the reproducibility (different batches) and repeatability (repetitively measuring on the same sensor) of the proposed SPCE-based sensors. As shown in Table S2, good repeatability and reproducibility (RSD < 5%) of the sensor were obtained at three As(III) concentration levels. Furthermore, the sensor can be regenerated with an electrochemical cleaning procedure by applying the corresponding potential. In addition, the sensor maintains the detection signal at 90% of its initial value after storage at 4 • C for 42 days, indicating excellent long-term stability.

Sample Analysis
To evaluate the practical applicability of the proposed sensor, various drinking water samples were utilized for the SWASV measurement of As(III) under optimal conditions. The results showed that no As(III) was detected in blank water samples. Three water samples were then spiked with As(III) at three concentration levels and used for analysis in the recovery experiments. As listed in Table 2, the recoveries obtained from the spiked water samples ranged from 92.8% to 114.6%, demonstrating the accuracy and reliability of As(III) detection by this sensing method.  Due to the inkjet printing preparation and modification of SPCE, it is necessary to evaluate the reproducibility (different batches) and repeatability (repetitively measuring on the same sensor) of the proposed SPCE-based sensors. As shown in Table S2, good repeatability and reproducibility (RSD < 5%) of the sensor were obtained at three As(III) concentration levels. Furthermore, the sensor can be regenerated with an electrochemical cleaning procedure by applying the corresponding potential. In addition, the sensor maintains the detection signal at 90% of its initial value after storage at 4 °C for 42 days, indicating excellent long-term stability.

Sample Analysis
To evaluate the practical applicability of the proposed sensor, various drinking water samples were utilized for the SWASV measurement of As(III) under optimal conditions. The results showed that no As(III) was detected in blank water samples. Three water samples were then spiked with As(III) at three concentration levels and used for analysis in the recovery experiments. As listed in Table 2, the recoveries obtained from the spiked water samples ranged from 92.8% to 114.6%, demonstrating the accuracy and reliability of As(III) detection by this sensing method.

Conclusions
In summary, a facile in situ electrodeposition method was developed to modify rGO/AuNP/MnO 2 nanocomposites onto SPCE for rapid As(III) sensing by SWASV in water samples. The nanocomposite electrode shows attractive sensing performance toward As(III) in terms of good sensitivity with LOD within the WHO guidelines, low-cost, and stability. In addition, as compared to the other arsenic sensor, the largely reduced preenrichment time (180 s), portability, and reinforced long-term stability (42 days) represent another figure of merit of the as-prepared sensor, which may favor in-field detection. The satisfied analytical capability can be attributed to a synergistic effect of the electrocatalytic properties of AuNPs and MnO 2 NPs, strong adsorption capabilities, and the stability of MnO 2 , as well as the large surface area of the rGO-based composite. The proposed sensor was applied to the analysis of trace As(III) in real water samples with satisfactory recoveries. These findings offer a promising application potential for the field-deployable and cost-effective monitoring of As(III) in water.