Electrochemical Sensing toward Trace As(III) Based on Mesoporous MnFe2O4/Au Hybrid Nanospheres Modified Glass Carbon Electrode

Au nanoparticles decorated mesoporous MnFe2O4 nanocrystal clusters (MnFe2O4/Au hybrid nanospheres) were used for the electrochemical sensing of As(III) by square wave anodic stripping voltammetry (SWASV). Modified on a cheap glass carbon electrode, these MnFe2O4/Au hybrid nanospheres show favorable sensitivity (0.315 μA/ppb) and limit of detection (LOD) (3.37 ppb) toward As(III) under the optimized conditions in 0.1 M NaAc-HAc (pH 5.0) by depositing for 150 s at the deposition potential of −0.9 V. No obvious interference from Cd(II) and Hg(II) was recognized during the detection of As(III). Additionally, the developed electrode displayed good reproducibility, stability, and repeatability, and offered potential practical applicability for electrochemical detection of As(III) in real water samples. The present work provides a potential method for the design of new and cheap sensors in the application of electrochemical determination toward trace As(III) and other toxic metal ions.


Introduction
Inorganic arsenic contamination in drinking water has become a serious worldwide threat to human health due to arsenic's high toxicity [1]. Such pollutants in drinking water may lead to many health problems, such as skin lesions, keratosis, lung cancer, and bladder cancer [2]. According to reports for twenty countries, the arsenic levels in drinking water is higher than the World Health Organization (WHO)'s arsenic guideline value of 10 µg¨L´1 (i.e., 10 ppb) [3,4]. Thus, it is rather important to have an accurate, rapid, and sensitive method to detect and monitor the environmental pollution of drinking water. Although various spectroscopic methods are excellent for arsenic testing in labs, they are very expensive and not suitable for in situ analyses due to the lengthy and complex instruments required [5,6]. However, electrochemical methods, particularly stripping voltammetry analysis, have provided promising techniques that are available for the sensitive detection and quantification of arsenic due to their low cost, portability, and suitability for on-site analysis [7][8][9]. The electrochemical application and performance of an electrode depends to a large extent on the materials from which it has been made. For the electrochemical detection of arsenic, a large effort has been poured into electrode materials and the modification of electrode surfaces in an attempt to improve their analytical performance [7,10]. So far, Au has been shown to be a promising choice in the sensitive determination of arsenic due to its excellent electrocatalytic ability [11,12]. However, considering the potential obstacles associated with using a solid gold macroelectrode, such as high costs, rigorous control, and care of the quality of its surface, Au nanoparticle modified

Apparatus
The Electrochemical measurements were performed on a CHI 660E computer-controlled potentiostat (ChenHua Instruments Co., Shanghai, China). A conventional three-electrode configuration was employed, consisting of a bare or modified glass carbon electrode (GCE) as a working electrode, an Ag/AgCl/saturated KCl electrode as a reference electrode, and a platinum wire as a counter electrode. Scanning electron microscopy (SEM) observation and energy-dispersive X-ray spectrum (EDS) analyses were carried out on a Quanta 200 FEG scanning electron microscope (FEI Company, Hillsboro, OR, USA). The High Resolution Transmission Electron Microscopy (HRTEM) and Transmission Electron Microscopy (TEM) images analyses were carried out on JEM-2010 and JEM-1400 microscopes (JEOL, Tokyo, Japan), respectively.

Preparation of Mesoporous MnFe 2 O 4 /Au Hybrid Nanospheres
Firstly, the mesoporous MnFe 2 O 4 nanocrystal clusters (MnFe 2 O 4 NCs (nanocrystal clusters)) were prepared according to a previous report [28]. Then they were APTMS-functionalized according to the previous report with some modification showing as follow [29]: MnFe 2 O 4 NCs (20 mg) were added to a solution containing ethanol (30 mL) and water (2 mL), followed by the addition of ammonium hydroxide (25%; 2 mL) and APTMS (200 µL). Then, the resulting solution was sonicated continuously for about 8 h. After four-step separation by means of an external magnetic field, the resulting APTMS-functionalized MnFe 2 O 4 NCs was dissolved in water (10 mL). After being selected with a magnet and rinsed with water four times, the resulting APTMS-functionalized MnFe 2 O 4 NCs were dissolved in water (10 mL) for further use.
Subsequently, the solution of APTMS-functionalized MnFe 2 O 4 NCs (1 mL) was mixed with 20 mL solution of Au nanoparticles via shaking for 24 h. The solution of Au nanoparticles was synthesized according to previously described methods [30]. The resulting product of mesoporous MnFe 2 O 4 /Au hybrid nanospheres was collected through an external magnet. An illustration of the preparation procedure of mesoporous MnFe 2 O 4 /Au hybrid nanospheres can be followed in Scheme 1.

Apparatus
The Electrochemical measurements were performed on a CHI 660E computer-controlled potentiostat (ChenHua Instruments Co., Shanghai, China). A conventional three-electrode configuration was employed, consisting of a bare or modified glass carbon electrode (GCE) as a working electrode, an Ag/AgCl/saturated KCl electrode as a reference electrode, and a platinum wire as a counter electrode. Scanning electron microscopy (SEM) observation and energy-dispersive X-ray spectrum (EDS) analyses were carried out on a Quanta 200 FEG scanning electron microscope (FEI Company, Hillsboro, OR, USA). The High Resolution Transmission Electron Microscopy (HRTEM) and Transmission Electron Microscopy (TEM) images analyses were carried out on JEM-2010 and JEM-1400 microscopes (JEOL, Tokyo, Japan), respectively.

Preparation of Mesoporous MnFe2O4/Au Hybrid Nanospheres
Firstly, the mesoporous MnFe2O4 nanocrystal clusters (MnFe2O4 NCs (nanocrystal clusters)) were prepared according to a previous report [28]. Then they were APTMS-functionalized according to the previous report with some modification showing as follow [29]: MnFe2O4 NCs (20 mg) were added to a solution containing ethanol (30 mL) and water (2 mL), followed by the addition of ammonium hydroxide (25%; 2 mL) and APTMS (200 μL). Then, the resulting solution was sonicated continuously for about 8 h. After four-step separation by means of an external magnetic field, the resulting APTMS-functionalized MnFe2O4 NCs was dissolved in water (10 mL). After being selected with a magnet and rinsed with water four times, the resulting APTMS-functionalized MnFe2O4 NCs were dissolved in water (10 mL) for further use.
Subsequently, the solution of APTMS-functionalized MnFe2O4 NCs (1 mL) was mixed with 20 mL solution of Au nanoparticles via shaking for 24 h. The solution of Au nanoparticles was synthesized according to previously described methods [30]. The resulting product of mesoporous MnFe2O4/Au hybrid nanospheres was collected through an external magnet. An illustration of the preparation procedure of mesoporous MnFe2O4/Au hybrid nanospheres can be followed in Scheme 1.

Preparation of Mesoporous MnFe2O4/Au Hybrid Nanospheres Modified GCE
Firstly, the GCE was polished with 0.3 μm and 0.05 μm alumina power slurries sequentially to form a mirror-shiny surface. Next, it was sonicated with 1:1 HNO3 solution, absolute ethanol, and water for 1 min, respectively. Subsequently, the mesoporous MnFe2O4/Au hybrid nanospheres film on the surface of the GCE was produced as follows: Mesoporous MnFe2O4/Au hybrid nanospheres (2 mg) were dispersed into purified water (2 mL) to obtain a uniform dispersion by 3 min of sonication. A drop of the above MnFe2O4/Au hybrid nanospheres solution was pipetted onto the fresh surface of the GCE and dried in air. After evaporation, a thin mesoporous MnFe2O4/Au hybrid nanospheres film was formed on the surface of the GCE.

Electrochemical Measurements
The electrochemical behavior for the MnFe2O4/Au hybrid nanospheres modified GCE was observed through square wave anodic stripping voltammetry (SWASV) under the optimized conditions. Under the potential of −0.9 V for 150 s, As(0) was deposited by the reduction of As(III) in

Preparation of Mesoporous MnFe 2 O 4 /Au Hybrid Nanospheres Modified GCE
Firstly, the GCE was polished with 0.3 µm and 0.05 µm alumina power slurries sequentially to form a mirror-shiny surface. Next, it was sonicated with 1:1 HNO 3 solution, absolute ethanol, and water for 1 min, respectively. Subsequently, the mesoporous MnFe 2 O 4 /Au hybrid nanospheres film on the surface of the GCE was produced as follows: Mesoporous MnFe 2 O 4 /Au hybrid nanospheres (2 mg) were dispersed into purified water (2 mL) to obtain a uniform dispersion by 3 min of sonication. A drop of the above MnFe 2 O 4 /Au hybrid nanospheres solution was pipetted onto the fresh surface of the GCE and dried in air. After evaporation, a thin mesoporous MnFe 2 O 4 /Au hybrid nanospheres film was formed on the surface of the GCE.

Electrochemical Measurements
The electrochemical behavior for the MnFe 2 O 4 /Au hybrid nanospheres modified GCE was observed through square wave anodic stripping voltammetry (SWASV) under the optimized conditions. Under the potential of´0.9 V for 150 s, As(0) was deposited by the reduction of As(III) in 0.1 M HAc-NaAc (pH 5.0). Subsequently, the electrodeposited As(0) was anodic stripped to As(III) which was performed in the potential range of´0.4 to 0.4 V under the following conditions: frequency, 25 Hz; amplitude, 25 mV; increment potential, 4 mV; vs. Ag/AgCl. Thereafter, desorption potential of´0.9 V for 150 s was carried out to remove the residual As(0) under stirring conditions. The same experimental conditions were applied in the interference, reproducibility, stability, repeatability, and real sample analysis studies. Cyclic voltammograms (CV) and electrochemical impedance spectra (EIS) were performed in the mixing solution containing 5 mM Fe(CN) 6 3´/4´a nd 0.1M KCl with the scanning rate of 100 mV¨s´1.  (Figure 1b). This mesoporous architecture might provide the microspheres with large surface areas in favor of adsorbing guest molecules [31][32][33]. The selected-area electron diffraction (SAED) pattern indicates that the diffraction spots are widened into narrow arcs (insets of Figure 1b), also signifying the clusters are composed of many misaligned ferrite nanocrystals [31,34]. The d value of the crystal plane (311) shown in Figure 1c Figure 2a displays the anodic and cathodic peak currents for Au, MnFe 2 O 4 NCs, and MnFe 2 O 4 /Au hybrid nanospheres modified GCE which all decline apparently, which thereby indicates that the bare GCE has been modified successfully by the above nanomaterials. Among all the modified GCE, it can be found that the anodic and cathodic peak currents of the MnFe 2 O 4 NCs modified GCE is the lowest, which is due to the poor conductivity of the metal oxide on the surface of the bare GCE which can slow the electron transfer [35]. However, after being decorated by Au nanoparticles, the anodic and cathodic peak currents of MnFe 2 O 4 /Au hybrid nanospheres modified GCE observably increases and the peak-to-peak separation ( Ep) decreases from 190 mV to 120 mV, indicating the electron transfer of MnFe 2 O 4 NCs were improved observably by the hybrid of the Au nanoparticles on the surface of MnFe 2 O 4 NCs.

Characterization of Mesoporous MnFe 2 O 4 /Au Hybrid Nanospheres
real sample analysis studies. Cyclic voltammograms (CV) and electrochemical impedance spectra (EIS) were performed in the mixing solution containing 5 mM Fe(CN)6 3−/4− and 0.1M KCl with the scanning rate of 100 mV·s −1 .  decorating of Au nanoparticles on the surface of MnFe 2 O 4 NCs could improve the electron transfer of MnFe 2 O 4 NCs on the surface of the GCE, which is consistent with the above CV results. Figure 2c presents the SWASV analytical results of the MnFe 2 O 4 NCs, Au nanoparticles, and MnFe 2 O 4 /Au hybrid nanospheres modified GCE. When the accumulation process was performed for 150 s at´0.9 V in 50 ppb As(III) solution containing 0.1 M acetate buffer (pH 5.0), it shows no peak appears in the curve of the MnFe 2 O 4 NCs modified GCE, and a weak peak at´0.9 V can be observed for the Au nanoparticles modified GCE. For the MnFe 2 O 4 /Au hybrid nanospheres modified GCE, the peak current increases much higher than the single MnFe 2 O 4 NCs or Au nanoparticles modified GCE. It shows that the hybrid that consists of Au nanoparticles on the surface of MnFe 2 O 4 NCs could provide much more remarkable electrochemical performances towards As(III) detection. It might be attributed to the catalytic effect and high electron transfer of Au nanoparticles would provide much more active sites for the electrochemical reactions in the detection process of As(III). Thus, based on the unique mesoporous structure of the MnFe 2 O 4 NCs responses to As(III) [25], the hybrid of Au nanoparticles on the surface of MnFe 2 O 4 NCs indeed provides a synergistic effect for the detection of As(III) through SWASV. for the Au nanoparticles modified GCE. For the MnFe2O4/Au hybrid nanospheres modified GCE, the peak current increases much higher than the single MnFe2O4 NCs or Au nanoparticles modified GCE. It shows that the hybrid that consists of Au nanoparticles on the surface of MnFe2O4 NCs could provide much more remarkable electrochemical performances towards As(III) detection. It might be attributed to the catalytic effect and high electron transfer of Au nanoparticles would provide much more active sites for the electrochemical reactions in the detection process of As(III). Thus, based on the unique mesoporous structure of the MnFe2O4 NCs responses to As(III) [25], the hybrid of Au nanoparticles on the surface of MnFe2O4 NCs indeed provides a synergistic effect for the detection of As(III) through SWASV.

Optimum Experimental Conditions of Electrochemical Detection of As(III)
The optimization of experimental conditions (supporting electrolytes, pH, deposition potential, and deposition time) was taken into account using SWASV. The effect of supporting electrolytes at pH 5.0 was tested and NaAc-HAc was finally chosen as the best electrolyte in comparison with PBS and NH4Cl-NH4OH (Figure 3a). The effect of pH on the SWASV response was investigated in the pH range of 3.0-8.0 in 0.1 M NaAc-HAc solutions, as shown in Figure 3b. The maximum peak current was achieved when the solution was maintained at pH 5.0. The decline of the pH value would provide a more acidic environment resulting in the destruction of mesoporous MnFe2O4 NCs, thus the stripping current became poor. The stripping current also observably declined when the pH value increased to a high level, such as 7 or 8, which is consistent with the report that the adsorption capacity of As(III) decreased under higher pH values [36]. In a higher pH solution, the electrostatic attraction will weaken between the negatively charged As(III) species and positively charged surface sites of MnFe2O4 NCs. Thus, pH 5.0 was chosen for the preconcentration solution in this experiment. Different deposition potentials in the range of −1.2 to −0.4 V were also experimented with via SWASV by standard additions of 50 ppb As(III) in 0.1 M NaAc-HAc (pH 5.0), as shown in Figure 3c. It was

Optimum Experimental Conditions of Electrochemical Detection of As(III)
The optimization of experimental conditions (supporting electrolytes, pH, deposition potential, and deposition time) was taken into account using SWASV. The effect of supporting electrolytes at pH 5.0 was tested and NaAc-HAc was finally chosen as the best electrolyte in comparison with PBS and NH 4 Cl-NH 4 OH (Figure 3a). The effect of pH on the SWASV response was investigated in the pH range of 3.0-8.0 in 0.1 M NaAc-HAc solutions, as shown in Figure 3b. The maximum peak current was achieved when the solution was maintained at pH 5.0. The decline of the pH value would provide a more acidic environment resulting in the destruction of mesoporous MnFe 2 O 4 NCs, thus the stripping current became poor. The stripping current also observably declined when the pH value increased to a high level, such as 7 or 8, which is consistent with the report that the adsorption capacity of As(III) decreased under higher pH values [36]. In a higher pH solution, the electrostatic attraction will weaken between the negatively charged As(III) species and positively charged surface sites of MnFe 2 O 4 NCs. Thus, pH 5.0 was chosen for the preconcentration solution in this experiment. Different deposition potentials in the range of´1.2 to´0.4 V were also experimented with via SWASV by standard additions of 50 ppb As(III) in 0.1 M NaAc-HAc (pH 5.0), as shown in Figure 3c. It was found that the deposition potential negative shifts from´0.4 V to´0.9 V (vs. Ag/AgCl) can obviously improve the peak current of reduction of As(III). Generally, the peak current would reach a plateau, settling to a constant value which should be the optimum deposition potential. However, when the potentials beyond´0.9 V were applied, a decrease response of the peak current were observed, as shown in Figure 3c, due to the reduction of hydrogen, which would lead to the formation of micro gas bubbles and thus reduce the effective electrode area and weaken the response of As(III) [35]. Therefore, the deposition potential of 0.9 V was selected for further studies. The dependence of the stripping peak current of As(III) at the deposition time on the sensing response was studied under 50 ppb As(III) (pH 5.0). Figure 3d shows the stripping peak current improved with the deposition time increase from 30 s to 240 s. Although the sensitivity was improved under a longer deposition time, it also reduced the upper detection limit due to the surface saturation in the high metal ion concentration [37]. When the deposition time exceeded 150 s, the increased velocity of the stripping current become slower, suggesting that the electrode surface was approximately saturated by the As(III). Considering the time consumed, the optimized deposition time of 150 s was used throughout. At last, the optimum experimental conditions for the electrochemical detection of As(III) was determined in 0.1 M NaAc-HAc (pH 5.0) by depositing it for 150 s under the deposition potential of´0.9 V. settling to a constant value which should be the optimum deposition potential. However, when the potentials beyond −0.9 V were applied, a decrease response of the peak current were observed, as shown in Figure 3c, due to the reduction of hydrogen, which would lead to the formation of micro gas bubbles and thus reduce the effective electrode area and weaken the response of As(III) [35]. Therefore, the deposition potential of -0.9 V was selected for further studies. The dependence of the stripping peak current of As(III) at the deposition time on the sensing response was studied under 50 ppb As(III) (pH 5.0). Figure 3d shows the stripping peak current improved with the deposition time increase from 30 s to 240 s. Although the sensitivity was improved under a longer deposition time, it also reduced the upper detection limit due to the surface saturation in the high metal ion concentration [37]. When the deposition time exceeded 150 s, the increased velocity of the stripping current become slower, suggesting that the electrode surface was approximately saturated by the As(III). Considering the time consumed, the optimized deposition time of 150 s was used throughout. At last, the optimum experimental conditions for the electrochemical detection of As(III) was determined in 0.1 M NaAc-HAc (pH 5.0) by depositing it for 150 s under the deposition potential of −0.9 V.

Electrochemical Detection of As(III) with MnFe2O4/Au Hybrid Nanospheres Modified GCE
Under the optimal experimental conditions, As(III) was determined on the MnFe2O4/Au hybrid nanospheres modified GCE using SWASV. Figure 4a presents the SWASV response toward As(III) over the concentration ranges of 10 to 110 ppb. The linearization equation was i/μA = 18.6 + 0.315 c/ppb, with the correlation coefficient of 0.996 (Figure 4b). Based on a signal-to-noise ratio equal to 3 (3σ method), the theoretical limit of detection (LOD) was estimated to be 3.37 ppb. The sensitivity and LOD of this study were compared with other previously reported electrodes listed in Table 1. In contrast to other Au electrode systems, it can be observed that the proposed electrode can obtain preferable sensitivity (0.315 μA/ppb) while remaining cheap at the same time. Although the LOD of 3.37 ppb is high, it is still within the maximum permissible limits for As(III) in drinking water issued by the World Health Organization (10 ppb). However, compared with the Fe3O4-RTIL modified SPCE

Electrochemical Detection of As(III) with MnFe 2 O 4 /Au Hybrid Nanospheres Modified GCE
Under the optimal experimental conditions, As(III) was determined on the MnFe 2 O 4 /Au hybrid nanospheres modified GCE using SWASV. Figure 4a presents the SWASV response toward As(III) over the concentration ranges of 10 to 110 ppb. The linearization equation was i/µA = 18.6 + 0.315 c/ppb, with the correlation coefficient of 0.996 (Figure 4b). Based on a signal-to-noise ratio equal to 3 (3σ Sensors 2016, 16, 935 8 of 13 method), the theoretical limit of detection (LOD) was estimated to be 3.37 ppb. The sensitivity and LOD of this study were compared with other previously reported electrodes listed in Table 1. In contrast to other Au electrode systems, it can be observed that the proposed electrode can obtain preferable sensitivity (0.315 µA/ppb) while remaining cheap at the same time. Although the LOD of 3.37 ppb is high, it is still within the maximum permissible limits for As(III) in drinking water issued by the World Health Organization (10 ppb). However, compared with the Fe 3 O 4 -RTIL modified SPCE [18], the sensitivity and LOD of this proposed sensor was at a disadvantage. It was considered that the Fe 3 O 4 -RTIL composite could provide a specific interface for arsenic to accumulate and exchange electrons, thus future studies focusing on the MnFe 2 O 4 -Au-RTIL composite may obtain exciting sensitivity and LOD for the electrochemical sensing of As(III).

Interference Measurements
Commonly, the other metal ions can coprecipitate and strip off under the experimental conditions for determination of As(III), thus sensitive detection of As(III) in the real sample without interference is a challenging task. In order to investigate the interference of other heavy metal ions to the electrochemical determination of As(III), a series of interference measurements were studied. Figure 5 presents the SWASV response of MnFe2O4/Au hybrid nanospheres modified GCE in 0.1 M HAc-NaAc (pH 5.0) containing 50 ppb As(III) in the presence of Cd(II), Hg(II), Pb(II), and Cu(II) and over a concentration range of 0 to 500 ppb, respectively. Figure 5a,b show the peak stripping currents of As(III) almost stays the same as the concentrations of Cd(II) and Hg(II) increase linearly. When the concentrations of Cd(II) and Hg(II) are up to 10-fold the concentration of As(III), the peak current of As(III) still changes a little. It means Cd(II) and Hg(II) do not observably interfere with the detection of As(III) for the MnFe2O4/Au hybrid nanospheres modified GCE when sensing the two target metal ions simultaneously. As to the interference of Cu(II), Figure 5c indicates the peak currents of As(III) also remain the same when the concentration of Cu(II) is less than 300 ppb. However, the peak current of As(III) drops observably when the concentration of Cu(II) increases further. Therefore, it means that Cu(II) would interfere with the detection of As(III) for the MnFe2O4/Au hybrid nanospheres modified GCE when the concentration of Cu(II) is beyond a certain value when the Cu(II) and As(III)

Interference Measurements
Commonly, the other metal ions can coprecipitate and strip off under the experimental conditions for determination of As(III), thus sensitive detection of As(III) in the real sample without interference is a challenging task. In order to investigate the interference of other heavy metal ions to the electrochemical determination of As(III), a series of interference measurements were studied. Figure 5 presents the SWASV response of MnFe 2 O 4 /Au hybrid nanospheres modified GCE in 0.1 M HAc-NaAc (pH 5.0) containing 50 ppb As(III) in the presence of Cd(II), Hg(II), Pb(II), and Cu(II) and over a concentration range of 0 to 500 ppb, respectively. Figure 5a,b show the peak stripping currents of As(III) almost stays the same as the concentrations of Cd(II) and Hg(II) increase linearly. When the concentrations of Cd(II) and Hg(II) are up to 10-fold the concentration of As(III), the peak current of As(III) still changes a little. It means Cd(II) and Hg(II) do not observably interfere with the detection of As(III) for the MnFe 2 O 4 /Au hybrid nanospheres modified GCE when sensing the two target metal ions simultaneously. As to the interference of Cu(II), Figure 5c indicates the peak currents of As(III) also remain the same when the concentration of Cu(II) is less than 300 ppb. However, the peak current of As(III) drops observably when the concentration of Cu(II) increases further. Therefore, it means that Cu(II) would interfere with the detection of As(III) for the MnFe 2 O 4 /Au hybrid nanospheres modified GCE when the concentration of Cu(II) is beyond a certain value when the Cu(II) and As(III) coexist within a solution. On the other hand, Figure 5d shows that the peak current of As(III) almost rises linearly as the concentration of Pb(II) increases from 0 to 500 ppb. These results indicate that the existence of Pb(II) would enhance the detection signal of As(III) to some extent. This can be reasoned by the mutual promotion of adsorption sites between Pb(II) and As(III) before the adsorbing capacity of heavy metal ions reaches saturation. by the mutual promotion of adsorption sites between Pb(II) and As(III) before the adsorbing capacity of heavy metal ions reaches saturation.

Reproducibility, Stability and Repeatability
To evaluate the reproducibility of the modified electrode, six MnFe2O4/Au hybrid nanospheres modified GCEs were prepared following the same procedure and were applied to detect 50 ppb As(III) in 0.1 M HAc-NaAc (pH 5.0) under the optimized conditions. The relative standard deviation (RSD) derived from the peak currents from six tests was 3.9%, indicating that the developed method has good reproducibility. Additionally, the stability of the modified electrodes were investigated by storing six prepared electrode at 4 °C for ten days, and then testing the SWASV response toward 50 ppb As(III) under the optimized conditions. The result showed that the stripping peak current only decreased by 7.8%, 7.3%, 3.7%, 4.6%, 7.0%, and 4.8% of the first value, thereby showing a long-term stability of the fabricated electrode. Moreover, the repeatability of the MnFe2O4/Au hybrid nanospheres modified GCE were also evaluated by conducting repetitive experiments for 20 times to detect 50 ppb As(III) on the modified electrode under the optimized conditions. Figure 6 shows the stripping current was nearly constant after continuous cycling for 20 times, and no obvious changes in the peak currents were observed with the RSD of 1.9%. The good reproducibility, long-term stability, and favorable repeatability of MnFe2O4/Au hybrid nanospheres modified GCE make them attractive for the preparation of a electrochemical sensor toward As(III).

Reproducibility, Stability and Repeatability
To evaluate the reproducibility of the modified electrode, six MnFe 2 O 4 /Au hybrid nanospheres modified GCEs were prepared following the same procedure and were applied to detect 50 ppb As(III) in 0.1 M HAc-NaAc (pH 5.0) under the optimized conditions. The relative standard deviation (RSD) derived from the peak currents from six tests was 3.9%, indicating that the developed method has good reproducibility. Additionally, the stability of the modified electrodes were investigated by storing six prepared electrode at 4˝C for ten days, and then testing the SWASV response toward 50 ppb As(III) under the optimized conditions. The result showed that the stripping peak current only decreased by 7.8%, 7.3%, 3.7%, 4.6%, 7.0%, and 4.8% of the first value, thereby showing a long-term stability of the fabricated electrode. Moreover, the repeatability of the MnFe 2 O 4 /Au hybrid nanospheres modified GCE were also evaluated by conducting repetitive experiments for 20 times to detect 50 ppb As(III) on the modified electrode under the optimized conditions. Figure 6 shows the stripping current was nearly constant after continuous cycling for 20 times, and no obvious changes in the peak currents were observed with the RSD of 1.9%. The good reproducibility, long-term stability, and favorable repeatability of MnFe 2 O 4 /Au hybrid nanospheres modified GCE make them attractive for the preparation of a electrochemical sensor toward As(III).

Real Sample Analysis
In order to evaluate the practical application of the MnFe2O4/Au hybrid nanospheres modified GCE, real water sample analyses were taken from laboratory tap water [26,42]. The real sample was diluted with a 0.1 M HAc-NaAc buffer solution (pH 5.0) in a ratio of 1:9 without any further treatment. The standard addition of 10 ppb As(III) was performed in the diluted sample and the recovery studies were carried out by further standard additions of As(III) into the diluted sample with a known concentration of As(III). The SWASV response and the corresponding calibration plot of peak currents against As(III) concentrations are presented in Figure 7. No obvious signals for As(III) were observed in the real samples (dash dotted line in Figure 7), indicating no As(III) was detected in the laboratory tap water. Furthermore, the recovery obtained is calculated to be 103% ± 8.1%, which reveals that the proposed detection method for As(III) has the potential for practical application.

Conclusions
In summary, Au nanoparticles decorated mesoporous MnFe2O4 nanocrystal clusters (MnFe2O4/Au hybrid nanospheres) with high electrochemical performance were used for the electrochemical

Real Sample Analysis
In order to evaluate the practical application of the MnFe 2 O 4 /Au hybrid nanospheres modified GCE, real water sample analyses were taken from laboratory tap water [26,42]. The real sample was diluted with a 0.1 M HAc-NaAc buffer solution (pH 5.0) in a ratio of 1:9 without any further treatment. The standard addition of 10 ppb As(III) was performed in the diluted sample and the recovery studies were carried out by further standard additions of As(III) into the diluted sample with a known concentration of As(III). The SWASV response and the corresponding calibration plot of peak currents against As(III) concentrations are presented in Figure 7. No obvious signals for As(III) were observed in the real samples (dash dotted line in Figure 7), indicating no As(III) was detected in the laboratory tap water. Furthermore, the recovery obtained is calculated to be 103%˘8.1%, which reveals that the proposed detection method for As(III) has the potential for practical application.

Real Sample Analysis
In order to evaluate the practical application of the MnFe2O4/Au hybrid nanospheres modified GCE, real water sample analyses were taken from laboratory tap water [26,42]. The real sample was diluted with a 0.1 M HAc-NaAc buffer solution (pH 5.0) in a ratio of 1:9 without any further treatment. The standard addition of 10 ppb As(III) was performed in the diluted sample and the recovery studies were carried out by further standard additions of As(III) into the diluted sample with a known concentration of As(III). The SWASV response and the corresponding calibration plot of peak currents against As(III) concentrations are presented in Figure 7. No obvious signals for As(III) were observed in the real samples (dash dotted line in Figure 7), indicating no As(III) was detected in the laboratory tap water. Furthermore, the recovery obtained is calculated to be 103% ± 8.1%, which reveals that the proposed detection method for As(III) has the potential for practical application.

Conclusions
In summary, Au nanoparticles decorated mesoporous MnFe2O4 nanocrystal clusters (MnFe2O4/Au hybrid nanospheres) with high electrochemical performance were used for the electrochemical

Conclusions
In summary, Au nanoparticles decorated mesoporous MnFe 2 O 4 nanocrystal clusters (MnFe 2 O 4 /Au hybrid nanospheres) with high electrochemical performance were used for the electrochemical determination of As(III) by square wave anodic stripping voltammetry (SWASV). The as-prepared product was characterized by SEM, HRTEM, TEM, and EDS. The results suggest that 13 nm Au nanoparticles were indeed supported on the surface of mesoporous MnFe 2 O 4 microspheres with the diameter of about 350 nm. Electrochemical behavior of the MnFe 2 O 4 /Au hybrid nanospheres modified GCE toward As(III) showed favorable sensitivity (0.315 µA/ppb) and LOD (3.37 ppb) for As(III) under the optimized conditions in 0.1 M NaAc-HAc (pH 5.0) by depositing for 150 s at the deposition potential of´0.9 V. No obvious interference from Cd(II) and Hg(II) was recognized during the detection of As(III). In addition, the excellent reproducibility, stability, and repeatability of MnFe 2 O 4 /Au hybrid nanospheres made it a promising electrode material for the electrochemical determination of As(III). Furthermore, the MnFe 2 O 4 /Au hybrid nanospheres modified GCE offered potential practical applicability in the electrochemical detection of As(III) in real water samples. Ultimately, the present work may provide a potential method for the design of new and cheap sensors in the application of electrochemical detection toward trace As(III) and other toxic metal ions.