Electrochemical Detection of Hydrazine Using Poly(dopamine)-Modified Electrodes

We have developed a simple and selective method for the electrochemical detection of hydrazine (HZ) using poly(dopamine) (pDA)-modified indium tin oxide (ITO) electrodes. Modification with pDA was easily achieved by submerging the ITO electrode in a DA solution for 30 min. The electrocatalytic oxidation of HZ on the pDA-modified ITO electrode was measured by cyclic voltammetry. In buffer solution, the concentration range for linear HZ detection was 100 µM–10 mM, and the detection limit was 1 µM. The proposed method was finally used to determine HZ in tap water to simulate the analysis of real samples. This method showed good recovery (94%–115%) and was not affected by the other species present in the tap water samples.


Introduction
Hydrazine (HZ, N 2 H 4 ) is an inorganic compound that exists as a volatile and flammable colorless liquid. HZ and its derivatives are mainly used in fuel-cell technology [1] and as propellants in rocket fuel, pesticides for agriculture, intermediates in photoprinting and pharmaceuticals, and deoxidizers in boilers. HZ is considered hazardous and is classified as a carcinogen by the US Environmental Protection Agency (EPA) [2] because it can cause chromosome aberrations and negatively affect the lungs, liver, spleen, thyroid gland, and central nervous system. Therefore, methods for the detection and monitoring systems for HZ should be developed to meet the demands of both industries and environmental groups. To date, various analytical methods such as titrimetry [3], chromatography [4], spectrophotometry [5], flow injection analysis with chemiluminescence [6], potentiometry [7], and amperometry [8][9][10][11] have been used for HZ detection. Among these techniques, electrochemical methods have advantages of simplicity, high sensitivity, rapid response, and low cost. To detect HZ using electrochemical methods, electrodes should be modified using special materials with electrocatalytic activities for the electrochemical oxidation of HZ to overcome the limitations associated with the high overpotential and sluggish kinetics of HZ in electrochemical reactions [12]. To achieve this, many studies have suggested the modification of electrodes by employing nanoparticles (e.g., Au, Ti, Pd, ZnO, Co, Sn, ZrO 2 , Bi, and Ce) [13][14][15][16][17][18][19][20][21], CoPC nanoparticles [22], manganese hexacyanoferrate [9], CoOOH nanosheets [12], zinc oxide nanowires [23], 4-pyridyl hydroquinone [24], biomolecules [25], and branched hierarchical ZnO nanorod arrays [26]. However, in many cases, preparation steps, which not only increases the cost of implementing the technique but also decreases its reproducibility. Therefore, simple and reproducible modification methods are urgently needed. To meet this requirement, in this study we adopted poly(dopamine) (pDA) to modify electrodes to detect HZ. We used pDA as a modifier or an electrocatalyst [19,[27][28][29][30][31], because pDA can be simply and quickly formed on many substrates in basic solutions [32][33][34][35].
Indium tin oxide (ITO) electrodes are mainly used in display and solar cell applications because of their good electrical conductivity, optical transparency, and low cost. These electrodes are particularly applicable to electrochemical sensors because of their low background current and high reproducibility [36,37]. Another characteristic property of ITO electrodes is that they have lower catalytic activity than other metal (e.g., Pt, Au, and Pd) electrodes, which can suppress interfering electrochemical reactions during measurements. Although the low catalytic activity of ITO electrodes can also decrease the target (e.g., HZ) current signal, this problem can be resolved by modifying the electrodes with specific materials.
Herein, we propose a pDA-modified ITO electrode for sensitive detection of HZ in solution. Because the formation of pDA films from DA solution is spontaneous and rapid, pDA-modified ITO electrodes were very easily obtained by submerging the electrodes in a DA solution for 30 min [27,33,38]. It is already known that DA polymerizes spontaneously to pDA in basic solution; in this process, DA is first oxidized to dopaminequinone, followed by intramolecular cyclization and rearrangement. Next, the product, leukodopaminechrome, is oxidized to dopaminechrome. Finally, dopaminechrome polymerizes to pDA on an ITO electrode. We used the resulting pDA-ITO electrode to detect HZ. As shown in Figure 1, the electrochemical reaction occurred on the pDA-ITO electrode when an appropriate potential was applied. In this process, pDA on the ITO electrode is first oxidized to poly(dopamine-o-quinone) (pDQ), generating an anodic current, and it is then reduced back to pDA by HZ, which is a strong reducing agent. Because of this redox cycling reaction, very high anodic currents can be obtained.

Electrochemical Measurement
A standard three-electrode cell with an ITO working electrode, a Au wire counter electrode, and a Ag/AgCl reference electrode (3 M KCl) was used. The area of the ITO working electrode was 0.28 cm 2 . The ITO working electrodes were sequentially washed in acetone, ethanol, and ultrapure water with ultrasonication for 15 min. The pDA-ITO working electrodes were obtained by

Electrochemical Measurement
A standard three-electrode cell with an ITO working electrode, a Au wire counter electrode, and a Ag/AgCl reference electrode (3 M KCl) was used. The area of the ITO working electrode was 0.28 cm 2 . The ITO working electrodes were sequentially washed in acetone, ethanol, and ultrapure water with ultrasonication for 15 min. The pDA-ITO working electrodes were obtained by submerging the cleaned ITO electrodes in Tris buffer containing 1 mM DA for 30 min. Subsequently, the pDA-ITO electrodes were washed with distilled water. All HZ-detection experiments were conducted in Tris buffer using pDA-ITO electrodes. All experiments were carried out at room temperature. Air-saturated solutions were used without any further preparation step (e.g., Ar bubbling).

Electrochemical Detection of HZ
The electrochemical oxidation properties of HZ were examined on a bare ITO electrode using commercially available Tris buffer (pH 8) (Figure 2). In the absence of HZ, only the capacitive current of the bare ITO electrode was measured (Figure 2A). In the presence of 1 mM¨HZ, a sluggish oxidation peak was observed because of the slow electron transfer between HZ and the bare ITO electrode ( Figure 2B). This slow electron transfer leads to a low anodic current and prevents sensitive detection. Thus, for HZ sensing applications, the HZ oxidation current must be increased. submerging the cleaned ITO electrodes in Tris buffer containing 1 mM DA for 30 min. Subsequently, the pDA-ITO electrodes were washed with distilled water. All HZ-detection experiments were conducted in Tris buffer using pDA-ITO electrodes. All experiments were carried out at room temperature. Airsaturated solutions were used without any further preparation step (e.g., Ar bubbling).

Electrochemical Detection of HZ
The electrochemical oxidation properties of HZ were examined on a bare ITO electrode using commercially available Tris buffer (pH 8) (Figure 2). In the absence of HZ, only the capacitive current of the bare ITO electrode was measured (Figure 2A). In the presence of 1 mM·HZ, a sluggish oxidation peak was observed because of the slow electron transfer between HZ and the bare ITO electrode ( Figure 2B). This slow electron transfer leads to a low anodic current and prevents sensitive detection. Thus, for HZ sensing applications, the HZ oxidation current must be increased. Spontaneous formation of a pDA film was achieved on a bare ITO electrode by simply immersing the electrode in Tris buffer containing 1 mM DA for 30 min. The resulting pDA film can be easily oxidized at the ITO electrode [34]. Previously, we suggested a DA-detection method using pDA films on ITO electrodes in the presence of HZ, wherein the oxidation current of pDA increased proportionally to the concentration of DA [35]. By inverting the concept underlying the previously described method, pDA films can be used as electrocatalytic mediators for sensitive detection of HZ oxidation. For this purpose, HZ oxidation was performed on pDA film-modified ITO electrodes in Tris buffer containing 1 mM·HZ. Unlike the case with the bare ITO electrode, we observed a large peak-shaped HZ oxidation current of approximately 60 µA at roughly 0.3 V in the modified electrode ( Figure 3).  Spontaneous formation of a pDA film was achieved on a bare ITO electrode by simply immersing the electrode in Tris buffer containing 1 mM DA for 30 min. The resulting pDA film can be easily oxidized at the ITO electrode [34]. Previously, we suggested a DA-detection method using pDA films on ITO electrodes in the presence of HZ, wherein the oxidation current of pDA increased proportionally to the concentration of DA [35]. By inverting the concept underlying the previously described method, pDA films can be used as electrocatalytic mediators for sensitive detection of HZ oxidation. For this purpose, HZ oxidation was performed on pDA film-modified ITO electrodes in Tris buffer containing 1 mM¨HZ. Unlike the case with the bare ITO electrode, we observed a large peak-shaped HZ oxidation current of approximately 60 µA at roughly 0.3 V in the modified electrode ( Figure 3). submerging the cleaned ITO electrodes in Tris buffer containing 1 mM DA for 30 min. Subsequently, the pDA-ITO electrodes were washed with distilled water. All HZ-detection experiments were conducted in Tris buffer using pDA-ITO electrodes. All experiments were carried out at room temperature. Airsaturated solutions were used without any further preparation step (e.g., Ar bubbling).

Electrochemical Detection of HZ
The electrochemical oxidation properties of HZ were examined on a bare ITO electrode using commercially available Tris buffer (pH 8) (Figure 2). In the absence of HZ, only the capacitive current of the bare ITO electrode was measured (Figure 2A). In the presence of 1 mM·HZ, a sluggish oxidation peak was observed because of the slow electron transfer between HZ and the bare ITO electrode ( Figure 2B). This slow electron transfer leads to a low anodic current and prevents sensitive detection. Thus, for HZ sensing applications, the HZ oxidation current must be increased. Spontaneous formation of a pDA film was achieved on a bare ITO electrode by simply immersing the electrode in Tris buffer containing 1 mM DA for 30 min. The resulting pDA film can be easily oxidized at the ITO electrode [34]. Previously, we suggested a DA-detection method using pDA films on ITO electrodes in the presence of HZ, wherein the oxidation current of pDA increased proportionally to the concentration of DA [35]. By inverting the concept underlying the previously described method, pDA films can be used as electrocatalytic mediators for sensitive detection of HZ oxidation. For this purpose, HZ oxidation was performed on pDA film-modified ITO electrodes in Tris buffer containing 1 mM·HZ. Unlike the case with the bare ITO electrode, we observed a large peak-shaped HZ oxidation current of approximately 60 µA at roughly 0.3 V in the modified electrode ( Figure 3).  This 60 µA current represents an increase of nearly 40-fold compared with that with the bare ITO electrode (ca. 0.15 µA). These results indicate that HZ oxidation by the pDA film could be applied for sensitive detection of HZ. Two reactions occurred simultaneously at the pDA-ITO electrode: the first is electrochemical oxidation of pDA to pDQ on the ITO electrode surface, and the second is pDQ reduction by HZ oxidation. We believe that the high anodic current shown in Figure 3 (red line) is governed by the diffusion process of HZ in solution. This is confirmed by the plot of the anodic peak current versus square root of the scan rate ( Figure 4). As shown in Figure 4B, the anodic peak currents showed good linearity with the square root of scan rate. Therefore, it is clear that the anodic current is certainly generated by the diffusion process of HZ in solution at the pDA-ITO electrode. This 60 µA current represents an increase of nearly 40-fold compared with that with the bare ITO electrode (ca. 0.15 µA). These results indicate that HZ oxidation by the pDA film could be applied for sensitive detection of HZ. Two reactions occurred simultaneously at the pDA-ITO electrode: the first is electrochemical oxidation of pDA to pDQ on the ITO electrode surface, and the second is pDQ reduction by HZ oxidation. We believe that the high anodic current shown in Figure 3 (red line) is governed by the diffusion process of HZ in solution. This is confirmed by the plot of the anodic peak current versus square root of the scan rate ( Figure 4). As shown in Figure 4B, the anodic peak currents showed good linearity with the square root of scan rate. Therefore, it is clear that the anodic current is certainly generated by the diffusion process of HZ in solution at the pDA-ITO electrode.

Optimization of the Experimental Conditions for HZ Detection
To maximize the HZ oxidation current, we varied the amount of DA and the pH of the Tris buffer containing HZ. First, the oxidation currents in 1 mM·HZ were measured with different amounts of DA on the ITO electrodes. The amount of DA can be controlled by varying the duration for which ITO electrodes are submerged in Tris buffer containing 1 mM DA. Figure 5 shows that the HZ oxidation current linearly increases up to a submersion time of 20 min and reaches saturation at 30 min. Based on this result, we selected 30 min of submersion as the optimal value. Next, the pH of the Tris buffer used for the electrochemical measurement was investigated because the oxidation current of pDA to pDQ can be altered based on the pKa value (8.8) of DA [39]. ITO electrodes submerged in 1 mM DA for 30 min were tested in Tris buffer containing 1 mM·HZ with various pH values. The

Optimization of the Experimental Conditions for HZ Detection
To maximize the HZ oxidation current, we varied the amount of DA and the pH of the Tris buffer containing HZ. First, the oxidation currents in 1 mM¨HZ were measured with different amounts of DA on the ITO electrodes. The amount of DA can be controlled by varying the duration for which ITO electrodes are submerged in Tris buffer containing 1 mM DA. Figure 5 shows that the HZ oxidation current linearly increases up to a submersion time of 20 min and reaches saturation at 30 min. This 60 µA current represents an increase of nearly 40-fold compared with that with the bare ITO electrode (ca. 0.15 µA). These results indicate that HZ oxidation by the pDA film could be applied for sensitive detection of HZ. Two reactions occurred simultaneously at the pDA-ITO electrode: the first is electrochemical oxidation of pDA to pDQ on the ITO electrode surface, and the second is pDQ reduction by HZ oxidation. We believe that the high anodic current shown in Figure 3 (red line) is governed by the diffusion process of HZ in solution. This is confirmed by the plot of the anodic peak current versus square root of the scan rate ( Figure 4). As shown in Figure 4B, the anodic peak currents showed good linearity with the square root of scan rate. Therefore, it is clear that the anodic current is certainly generated by the diffusion process of HZ in solution at the pDA-ITO electrode.

Optimization of the Experimental Conditions for HZ Detection
To maximize the HZ oxidation current, we varied the amount of DA and the pH of the Tris buffer containing HZ. First, the oxidation currents in 1 mM·HZ were measured with different amounts of DA on the ITO electrodes. The amount of DA can be controlled by varying the duration for which ITO electrodes are submerged in Tris buffer containing 1 mM DA. Figure 5 shows that the HZ oxidation current linearly increases up to a submersion time of 20 min and reaches saturation at 30 min. Based on this result, we selected 30 min of submersion as the optimal value. Next, the pH of the Tris buffer used for the electrochemical measurement was investigated because the oxidation current of pDA to pDQ can be altered based on the pKa value (8.8) of DA [39]. ITO electrodes submerged in 1 mM DA for 30 min were tested in Tris buffer containing 1 mM·HZ with various pH values. The Based on this result, we selected 30 min of submersion as the optimal value. Next, the pH of the Tris buffer used for the electrochemical measurement was investigated because the oxidation current of pDA to pDQ can be altered based on the pKa value (8.8) of DA [39]. ITO electrodes submerged in 1 mM DA for 30 min were tested in Tris buffer containing 1 mM¨HZ with various pH values. The tested pH values ranged from 8 to 10, and the highest oxidation current was observed at pH 9 ( Figure 6). Therefore, Tris buffer with pH 9 was chosen as the optimal test solution for HZ detection.
Sensors 2016, 16, 647 5 of 10 tested pH values ranged from 8 to 10, and the highest oxidation current was observed at pH 9 ( Figure 6). Therefore, Tris buffer with pH 9 was chosen as the optimal test solution for HZ detection. The subsequent HZ-detection experiments were performed under the optimal conditions of 30 min of ITO electrode submersion in DA solution and measurement in Tris buffer with pH 9.

HZ-Sensing Performances
Cyclic voltammetry was conducted to determine the concentrations of HZ on the pDA film-modified ITO electrodes at a scan rate of 50 mV/s. The oxidation peak currents at approximately 0.3 V were linearly proportional to the HZ concentration in the range of 100 µM-10 mM ( Figure 7A).

HZ-Sensing Performances
Cyclic voltammetry was conducted to determine the concentrations of HZ on the pDA film-modified ITO electrodes at a scan rate of 50 mV/s. The oxidation peak currents at approximately 0.3 V were linearly proportional to the HZ concentration in the range of 100 µM-10 mM ( Figure 7A). tested pH values ranged from 8 to 10, and the highest oxidation current was observed at pH 9 ( Figure 6). Therefore, Tris buffer with pH 9 was chosen as the optimal test solution for HZ detection. The subsequent HZ-detection experiments were performed under the optimal conditions of 30 min of ITO electrode submersion in DA solution and measurement in Tris buffer with pH 9.

HZ-Sensing Performances
Cyclic voltammetry was conducted to determine the concentrations of HZ on the pDA film-modified ITO electrodes at a scan rate of 50 mV/s. The oxidation peak currents at approximately 0.3 V were linearly proportional to the HZ concentration in the range of 100 µM-10 mM ( Figure 7A).  The linear relationship between the oxidation currents and HZ concentrations in this range can be described by the following equation ( Figure 7C): y " 60.5 log pxq p˘8.37%q´101 p˘14.9%q; pR 2 " 0.9070, 10 2 ď x ď 10 4 q Unlike the case of high concentrations (ě100 µM), another oxidation peak with an ambiguous shape appeared at approximately 0.1 V at low concentrations (ď10 µM) ( Figure 7B). We believe that this phenomenon occurred owing to the oxidation of the pDA film when the HZ concentration was lower than the DA concentration [30]. Therefore, we chose to measure the HZ concentration using the oxidation peak current at 0.3 V.
Although the oxidation peak currents increased with the HZ concentration (Figure 7), Figure 7C shows that this relationship was not linear when the concentration was lower than 10 µM. Indeed, the oxidation current of 1 µM HZ shown in the inset of Figure 7C is higher than the 3SD line, indicating that the limit of detection of this method is 1 µM HZ. Therefore, the linear concentration range for HZ detection is 100 µM-10 mM, and the detection limit is 1 µM. These values are comparable to those of other previously reported methods using, for example, nanoparticles [12,40], oxide materials [25,41], and polymers [42] (Table 1).

Influence of Interference Molecules for Detecting HZ and the Stability of pDA-ITO Electrodes
To apply this method to real samples, the pDA-ITO electrodes were used to detect various HZ concentrations in tap water solution with additional ions (containing 50 mM Tris, 300 µM Na + , 100 µM Ca 2+ , 100 µM Zn 2+ , 100 µM Mg 2+ , 100 µM Co 2+ , 100 µM Fe 2+ , 600 µM Cl´, 100 µM PO 4 3´, 100 µM SO 4 2´, and 100 µM CO 3 2´) (see the Supporting Information, Figure S1). As shown in Table 2, eight samples of tap water, four containing 100 µM and the other four containing 1000 µM HZ, could be detected using the pDA-modified ITO electrode with good recovery (94%-115%). Sixty-four electrodes were prepared on the same day, and four electrodes were used to measure the same amount of HZ each day (tap water solution with additional ions). The prepared electrodes were stored in a refrigerator at 4˝C. As shown in Figure 8, the normalized currents indicated that the detection currents decrease with time. When we used pDA-ITO electrodes 1 day later, 93.9% (˘7.96%) and 90.1% (˘10.5%) currents were observed when detection experiments were carried out at 1 mM and 0.1 mM¨HZ, respectively. After 3 days, the normalized currents decreased to 69.5% (˘4.34%) and 67.6% (˘1.2%), respectively, for 1 mM and 0.1 mM¨HZ. We observed similar decreasing trends in both 1 mM and 0.1 mM¨HZ detection experiments. This result means that the pDA-ITO electrodes are not suitable for long-term usage. However, more experiments are needed to extend the stability of pDA-ITO electrodes.
Sensors 2016, 16, 647 7 of 10 samples of tap water, four containing 100 µM and the other four containing 1000 µM HZ, could be detected using the pDA-modified ITO electrode with good recovery (94%-115%). Stabilities of pDA-ITO electrodes were measured with respect to time in a tap water solution containing additional ions (containing 50 mM Tris, 300 µM Na + , 100 µM Ca 2+ , 100 µM Zn 2+ , 100 µM Mg 2+ , 100 µM Co 2+ , 100 µM Fe 2+ , 600 µM Cl⁻, 100 µM PO4 3 ⁻, 100 µM SO4 2 ⁻, and 100 µM CO3 2 ⁻). Sixty-four electrodes were prepared on the same day, and four electrodes were used to measure the same amount of HZ each day (tap water solution with additional ions). The prepared electrodes were stored in a refrigerator at 4 °C. As shown in Figure 8, the normalized currents indicated that the detection currents decrease with time. When we used pDA-ITO electrodes 1 day later, 93.9% (±7.96%) and 90.1% (±10.5%) currents were observed when detection experiments were carried out at 1 mM and 0.1 mM·HZ, respectively. After 3 days, the normalized currents decreased to 69.5% (±4.34%) and 67.6% (±1.2%), respectively, for 1 mM and 0.1 mM·HZ. We observed similar decreasing trends in both 1 mM and 0.1 mM·HZ detection experiments. This result means that the pDA-ITO electrodes are not suitable for long-term usage. However, more experiments are needed to extend the stability of pDA-ITO electrodes.

Conclusions
We have developed a simple and fast EC-based HZ-detection method using pDA film-modified ITO electrodes. The modification of ITO electrodes with pDA films was easily achieved by submerging the electrodes in a DA solution for 30 min. The pDA film acts as a mediator for the EC reaction and shows good performance in the detection of HZ. This HZ detection method has a linear dynamic range of 100 µM-10 mM and has a detection limit of 1 µM. When it is used to analyze tap

Conclusions
We have developed a simple and fast EC-based HZ-detection method using pDA film-modified ITO electrodes. The modification of ITO electrodes with pDA films was easily achieved by submerging the electrodes in a DA solution for 30 min. The pDA film acts as a mediator for the EC reaction and shows good performance in the detection of HZ. This HZ detection method has a linear dynamic range of 100 µM-10 mM and has a detection limit of 1 µM. When it is used to analyze tap water containing various concentrations of HZ, it showed good recovery and was not affected by other ionic species present in the samples.