A sensitive hydrazine hydrate sensor based on a mercaptomethyl-terminated trinuclear Ni(II) complex modified gold electrode

Abstract

In this study, a novel mercapto-terminated trinuclear Ni(II) complex (Ni₃) was synthesized and used as an electrocatalyst for the detection of hydrazine hydrate in real water samples. The as-prepared Ni₃ molecule possesses six thiomethyl groups at its periphery and these SCH₃ groups can react with Au electrodes to immobilize the Ni₃ molecules on their surface through the formation of a self-assembled monolayer. The Ni₃-modified Au electrode (Ni₃/Au) demonstrates excellent electrocatalytic activity for the oxidation of hydrazine hydrate through a significant decrease in overpotential. The chronoamperometry study shows a diffusion coefficient (D) of 5.82 × 10⁻⁵ cm² s⁻¹ and a catalytic rate constant of 8.57 × 10³ M⁻¹ s⁻¹. Using the square wave voltammetry (SWV) technique, this Ni₃/Au electrode based hydrazine hydrate sensor exhibits a high sensitivity in quantitative analysis, and its detection limit could be as low as ∼0.07 μM with linearity ranging from 0.2 to 50 μM. In addition, due its good reproducibility, anti-interference performance, and long-term stability, the proposed sensor is capable of detecting trace levels of hydrazine hydrate in real water samples.

---

1. Introduction

Hydrazine, which is a strong reducing agent, has been widely used in numerous industrial fields since the 1950s, including pharmaceuticals, pesticides, foaming agents, fuel cells, manufacturing high purity metals, and propellant fuel for spacecrafts.^1,2 However, hydrazine is highly toxic for human health, and causes damage to the liver, kidneys, and central nervous system.³ Moreover, it has been classified as a human carcinogen by the US Environmental Protection Agency (EPA),⁴ and the high solubility of hydrazine (existing as hydrazine hydrate) in water increases its concentrations for ground water contamination.^5,6 Various methods, such as spectrophotometric determination,^4,6–9 electrochemiluminescence,⁵ gas chromatography (GC) or GC tandem mass spectrometry (GC-MS),¹⁰ and high performance liquid chromatography (HPLC) or LC-MS methods,^11,12 have been established for the detection of hydrazine. Although these methods demonstrate remarkable sensitivity and repeatability, they suffer from high running costs, time-consuming derivations, and complex separation or enrichment processes. Therefore, it is urgent to develop a simple but reliable and sensitive method for the detection of hydrazine hydrate at trace levels in an aqueous solution.

Due to the electroactive nature of hydrazine hydrate, electrochemical determination has proven to be a feasible and effective approach. Actually, as compared to the aforementioned methods, electrochemical techniques have several advantages such as easy to operate, rapid response, and high sensitivity. Previous literature have demonstrated that the electrooxidation of hydrazine is a four-electron irreversible process.¹³ However, the electron transfer rates of hydrazine oxidation at conventional electrodes (such as Au, Pt, or glass carbon electrodes) are relatively slow; therefore, direct electroanalysis is thus hindered by the accompanied large overpotentials. The common approach for lowering the over potential is through the introduction of electrocatalysts to fabricate chemically modified electrodes (CME). Various inorganic and organic materials with specific electrocatalytic properties have been reported, including metal nanoparticles,^14–16 semiconductor nanoparticles,^17,18 metallophthalocyanine complexes,^13,19,20 hexacyanoferrate salt,²¹ and organic compounds with a hydroquinone structure.^22,23 Sometimes, carbon nanotubes (CNTs) and graphene are used as the supporting substrate for electrocatalysts due to their high surface areas, excellent electric conductivity, and good chemical inertness.^16,24–26 It should be noted that most of these modified electrodes are fabricated by physical strategies, e.g. dip-dry, drop-dry, mixing, and electrostatic adsorption. However, physically anchored films are likely to be eluted from the surface of CMEs during electrochemical measurements, and it is difficult to control the concentration of the deposited electrocatalyst, thus reducing the reproducibility of the modified electrode and repeatability of the result.

Self-assembled monolayer (SAM) techniques, in contrast, have distinct advantages over traditionally modified electrodes, including high stability, well-arranged modified layers, and convenient operation steps.¹³ More importantly, the ultrathin solid films of modifiers can significantly prompt direct electron transfer (DET) between the immobilized electroactive centers and the surface of electrodes. These advantages of SAM techniques have aroused increasing interests in CME construction, particularly in the modification of coinage-metal based electrodes. As is known, molecules containing mercapto groups can self-assemble on the surface of Au or Ag electrodes through the formation of Au–S or Ag–S covalent bonds. In this respect, if we could functionalize such electrocatalysts with SH or SCH₃ groups, excellent catalytic ability and faster electron transfer between analytes and the electrode would be effectively integrated.

In our previous study, we reported the synthesis and usage of a mercapto-terminated hexanuclear Fe(III) complex,^27,28 in which the Fe₆ complex was chemically adsorbed on the surface of an Au electrode via self-assembly and a sensitive dopamine electrochemical sensor was constructed. Herein, we report the new mercapto-terminated Ni(II) complex, [Ni₃(C₃H₆N₄S)₆(OH)₆](Ni₃), for the first time. The as-prepared Ni₃ complex was immobilized on the surface of Au electrodes by self-assembly. Cyclic voltammetry, chronoamperometry, and electrochemical impedance spectroscopy methods were used to characterize and delineate the electrochemical behaviour of the Ni₃ modified Au electrode (Ni₃/Au). Moreover, the application of the Ni₃/Au electrode in the electrocatalytic oxidation of hydrazine hydrate was investigated. Square wave voltammetry was used for the quantitative analysis of hydrazine hydrate in real water samples.

2. Experimental section

2.1. Materials and apparatus

Hydrazine hydrate, HAuCl₄, and 3-amino-5-methylthio-1H-1,2,4-trizole (denoted as L) were purchased from Sigma-Aldrich. Ni(NO₃)₂·6H₂O and 2,2′-bipyridine (bpy) were purchased from Sinopharm Chemical Reagent Co., Ltd (China). The phosphate buffer solution (PBS, pH 7.0) used as the supporting electrolyte in this study was prepared with 0.1 M KH₂PO₄/K₂HPO₄. All reagents used were of analytical reagent grade and used as received. All solutions were prepared with ultrapure water produced from a Millipore-Q system.

High resolution transmission electron microscopy (HRTEM) images of the Ni₃/Au nanoparticles were taken on an FEI-Tecnai F20 (FEI, USA). To do this, colloidal gold was first synthesized according to the protocol of previous literature.²⁹ Then, an aliquot of 200 μL Ni₃ complex solution in ethanol (5 mg mL⁻¹) was added to 2.0 mL as-prepared gold sol. The mixture was stirred for 10 h to form SAM of Ni₃ molecules on the surface of the Au nanoparticles.

Chromatography detection of hydrazine derivation was performed on a Shimadzu LC-20A (Shimadzu, Japan) equipped with a diode array detector (DAD).

Cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and square wave voltammetry (SWV) were recorded on a CHI 660E electrochemical station (Shanghai ChenHua Instruments CO. LTD., China). A conventional three-electrode system was used. A bare or modified Au electrode was employed as the working electrode, a platinum plate as the auxiliary and Ag|AgCl (3 M KCl) as the reference electrode. All electrochemical measurements were conducted in an argon-purged (at least 15 min and maintained under an argon atmosphere during the entire experiment) solution at room temperature.

2.2. Electrode modification

The synthetic procedure for the Ni₃ complex is briefly described as follows. In a 50 mL double-necked, round-bottom flask equipped with a magnetic stirring bar, Ni(NO₃)₂·6H₂O (0.298 g, 1.0 mmol) in H₂O (5 mL) was added to a stirred solution of 3-amino-5-methylthio-1H-1,2,4-triazole (L) (0.065 g, 0.5 mmol) and 2,2-bpy (0.008 g 0.5 mmol) in acetone and methanol (v/v = 2 : 1) to obtain a green solution. The reaction mixture was refluxed for 4–5 hours and a blue solution was obtained. The solution was then filtered, and the solvent was allowed to evaporate slowly to form the final blue crystals of the mercaptomethyl-terminated trinuclear Ni(II) complex in several days (see ESI† for the synthetic scheme, H NMR, FTIR, and crystallographic data of the Ni₃ complex).

Prior to the modification, a bare gold disk electrode (2 mm diameter) was polished using aqueous slurries of alumina (1.0 down to 0.05 μm) and then washed ultrasonically in water/ethanol/water for 5 min to remove the physically adsorbed alumina. Furthermore, the electrode was electrochemically cleaned by cycling the electrode potential between 0.00 and +1.50 V in 0.5 M sulfuric acid (∼40 cycles, 50 mV s⁻¹), until reproducible voltammograms were observed. The cleaned Au electrodes were then immersed into 2 mL of 5 mM Ni₃ DMF solution for 24 h to ensure that saturated adsorption of the Ni₃ molecules the surface of Au electrodes occurred through Au–S bonds. The electrodes were thoroughly rinsed with ethanol and water to remove physically adsorbed Ni₃ molecules before electrochemical measurement.

2.3. Electrochemical measurements

EIS was performed in 0.1 M KNO₃ solution containing 5 mM Fe(CN)₆^3−/4−, and a sine wave potential with 5 mV amplitude superimposed on the formal potential of Fe(CN)₆^3−/4− (+0.20 V), was applied. The frequency range was 100 kHz to 0.01 Hz. For the SWV determination of hydrazine hydrate in this study, a sweep potential between −0.20 and +1.0 V was used, at an increasing potential of 4 mV, pulse amplitude of 25 mV, and frequency of 15 Hz.

2.4. Detection of hydrazine hydrate with HPLC method

As control experiments, HPLC was used to detect the concentration of hydrazine hydrate in the standard stock solution and sample solutions. First, ∼50 mg of hydrazine hydrate was accurately weighed and transferred into a 100 mL volumetric flask, and diluted to the volume with diluent (1 M NaOH in 1 : 1 water/methanol solution), which was used as the standard stock solution. This solution was then diluted quantitatively with diluent to obtain a series of standard solutions having a known concentration. Then, 0.5 mL of standard solution reacted with 4 mL 40 mg mL⁻¹ benzaldehyde to form benzalazine, which was extracted by hexane and used for HPLC analysis (benzalazine reference material was used for qualitative analysis and the external standard method for quantitative analysis). The chromatographic conditions were as follows: hexane and isopropyl alcohol (95 : 5) as the mobile phase, UV detector wavelength 310 nm, column temperature 30 °C, and the flow rate was set as 0.8 mL per minute.

3. Results and discussion

3.1. Characterization of the Ni₃ modified electrode

The crystal structure of the Ni₃ complex is shown as a ball-and-stick model in Fig. 1a. As can be observed, the structure consists of a centrosymmetric linear trinuclear molecule, with the central Ni^II (2) ion on a crystallographic inversion centre. Two terminal Ni^II (1) ions are coordinated by three nitrogen atoms from ligand L and three hydroxyl oxygen atoms to form a distorted octahedral environment. The central Ni^II (2) ion occupies a slightly distorted octahedron of N atoms from ligand L. Moreover, the Ni₃ complex features six peripheral SCH₃ from the 3-amino-5-methylthio-1H-1,2,4-trizole ligand. As mentioned above, mercapto groups can react with coinage metal to produce a stable covalent bond. In the present case, these SCH₃ groups enable the Ni₃ molecules to self-assemble on the surface of Au electrodes and generate a self-assembled monolayer, which may significantly improve the DET between the electroactive centers of Ni₃ and the Au electrode (as discussed below).

Fig. 1 (a) Structure of the Ni₃ complex, (b) HRTEM images of the Ni₃/Au nanoparticles, CV response at bare Au and Ni₃ complex modified Au electrodes (c) in 0.1 M PBS (pH 7.0) and (d) in 5 mM Fe(CN)₆³⁻. The scan rate is 50 mV s⁻¹.

The highlight of this study is the usage of the electrocatalytic property of the Ni₃ complex to lower the overpotential and enhance the sensitivity in hydrazine hydrate detection, thus it is first necessary to verify that the modified electrode was successfully built. In this study, although we cannot visually observe the surface of the Ni₃ modified electrode using a simple method, we managed to immobilize the Ni₃ complex on Au nanoparticles and obtained HRTEM images of the Ni₃/Au nanoparticles (Fig. 1b). Because Ni₃ molecules can self-assemble on the surface of Au nanoparticles, they certainly can be modified on the surface of Au electrodes by the same chemical reaction. As shown in Fig. 1b, several core–shell structures consisting of dark centers and light-colored edges are observed, which can be more clearly observed in the inset of Fig. 1b. The dark centers are Au nanoparticles with an average diameter of ∼12 nm. All the nanoparticles were wrapped by complex shells with a thickness of ∼1 nm (in good agreement with the size of Ni₃ measured using the SHELXS-97 program), which thus could be assigned to Ni₃ molecules. These images may be used as indirect evidence for the successful immobilization of the Ni₃ complex on the surface of the Au electrodes.

Fig. 1c compares the CVs of theNi₃/Au and bare Au electrodes in 0.1 M PBS (pH 7.0). As can be observed, the bare Au electrode did not show any noticeable redox process because there was no redox species in the electrolyte. In contrast, after modification of the Ni₃ complex, a couple of well-defined redox peaks were clearly observed, which could be attributed to the Ni(III)/Ni(II) redox couple. The formal potential (E_1/2) was ∼0.261 V, and the cathodic and anodic peaks were at +0.215 V and +0.307 V with ΔE_p = E_pa − E_pc = 92 mV, which is slightly greater than the theoretical 59 mV for fast one-electron transport. In addition, the ratio of the anodic peak current to the cathodic current I_pa/I_pc is approximately 0.94, which indicates a quasi-reversible electrochemical reaction for the immobilized Ni complex. The direct electrochemical signal of the Ni₃ complex further demonstrated the successful modification of the Au electrodes. In this study, Ni₃ was immobilized on the surface of Au electrodes via self-assembly. The formation of this monolayer could establish a direct electron transfer between the electroactive center of Ni₃ and the Au electrodes, which benefits the fabrication of an electrochemical sensor with high sensitivity and rapid response. Moreover, the CV responses of these two different electrodes in 5 mM Fe(CN)₆³⁻ (Fig. 1d) may also be evidence for the successful construction of the Ni₃/Au electrode. Compared with the Ni₃/Au electrode, the redox probe showed more reversible behavior at the bare Au electrode, thus indicating that the Ni₃ monolayer blocked electron transfer between the probe and the surface of the Au electrode.

3.2. Estimation of surface coverage

The reduction currents of gold oxide on the Ni₃/Au and bare Au electrodes in 0.5 M H₂SO₄ were used to roughly estimate the surface coverage of Ni₃ on the Au electrodes (θ). This was done because the Ni₃ complex on the surface of the Au electrode could insulate Au oxidation. Apparently, the larger the surface coverage of Ni₃ complexes, the smaller the reduction current will be. Therefore, θ was estimated according to the current values obtained from Fig. 2a, wherein the gold oxidation/reduction current was suppressed at the Ni₃/Au electrode (7.35 μA) in comparison with the bare Au electrode (34.64 μA) at 0.91 V. The value of θ was then calculated as 1 − 7.35/34.64 = 0.79.

Fig. 2 CVs representing (a) electrochemical oxidation/reduction of Au atoms at bare Au and Ni₃ complex modified Au electrodes in 0.5 M H₂SO₄, scan rate: 50 mV s⁻¹. (b) EIS, Nyquist plots of bare Au and Ni₃/Au electrodes in the presence of 0.1 M KNO₃ solution with 5 mM Fe(CN)₆^3−/4−.

EIS of the Ni₃/Au and bare Au electrodes were also measured to investigate the interface properties of the modified electrode; the Nyquist plots for both electrodes are shown in Fig. 2b. As observed, after modification of the Ni₃ complex on bare Au electrodes, the plot exhibited a much larger semicircle in the high frequency region. These results indicate that the modification of Ni₃ can increase the electron transfer resistance (R_et), thus confirming the successful immobilization of Ni₃ on the surface of the Au electrodes. By fitting the data with the Randles equivalent circuit, the values of electron transfer resistance, R_et, at the bare Au electrode and Ni₃/Au electrode, R′_et, were estimated to be 74 and 428 Ω, respectively. Using these data, we calculated the surface coverage of the Ni₃ molecules on the Au electrodes (θ′) according to θ′ = 1 − R_et/R′_et = 1 − 78/428 = 0.82, which was in good agreement with the calculated result using the gold oxide reduction current.

The concentration of the electroactive substance was evaluated according to the Laviron eqn (1) and (2) as follows:³⁰

where I_p represents the anodic or cathodic peak current, Γ is the concentration of the electroactive substance (mol cm⁻²), v is the scan rate (V s⁻¹), A is the electrode area (cm², in this case, it is πr² = 3.14 × (0.2/2)²), Q is the quantity of charge (C) calculated from the peak area of the voltammograms, and n is the number of electrons transferred. F, R, and T have their normal definitions. As obtained from Fig. 1c, which is the CV curve of the Ni₃/Au electrode (red solid), the I_p at 50 mV s⁻¹ scan rate was about 0.12 μA, and the quantity of charge transferred, Q, was 0.92 × 10⁻⁷ C, thus n was calculated to be 2.73 according to eqn (1), which is consistent with the single electron redox process of the Ni(III)/Ni(II) complex (note that each Ni₃ molecule has three Ni central atoms). Then, defining n = 3, the value of Γ was calculated to be 9.02 × 10⁻¹² mol cm⁻².

3.3. Electrochemical behavior of hydrazine at Ni₃/Au electrodes

The black dash curve in Fig. 3 represents a typical CV of 0.2 mM hydrazine hydrate at a bare Au electrode. A relatively strong oxidation peak at 0.72 V is observed, whereas no reverse current wave can be found. This electrochemical behavior suggests that the oxidation of hydrazine hydrate on the bare Au electrode is an irreversible electrochemical process. For a totally irreversible system controlled by an electrochemical step, because the inverse kinetic rate is very slow, the total current is thus similar to the positive scan current, i.e., no negative scan current will be found. In contrast, for the situation of the Ni₃/Au electrode, the oxidation peak potential negatively shifted to 0.42 V with an obviously increased peak current (note that the concentration of hydrazine hydrate was 50 μM in this case). Moreover, the oxidation peak of hydrazine occurred approximately at the same potential as the redox couple of the Ni₃ complex, whereas the current response increased by ∼10 times.

Fig. 3 CVs of 0.2 mM hydrazine hydrate at the bare Au electrode (black dash) and 50 μM hydrazine hydrate at the Ni₃/Au electrode (red solid). The electrolyte is 0.1 M Na₂SO₄ solution, and the scan rate is 50 mV s⁻¹.

The decrease in overpotential and increase in anodic peak current indicate a typical EC process, wherein the voltammetric response of hydrazine is catalytically mediated by the redox processes of the Ni₃ complex. Previous literatures have proven that Ni(III) can catalyze the oxidation of hydrazine, which starts with the formation of a Ni(III)–hydrazine complex, using the lone pair electrons on the N atoms and the unoccupied orbit on the central atom Ni(III).^31,32 In this study, we propose the following general mechanism for the oxidation of hydrazine at the Ni₃/Au electrode:

Ni(II) complex − e → Ni(III) complex (3)

Ni(III) complex + N₂H₄ → [Ni(III)N₂H₄ complex]_adduct (4)

[Ni(III)N₂H₄ complex]_adduct → [Ni(II)N₂H₃˙]_adduct + H⁺ (5)

[Ni(II)N₂H₃˙]_adduct → Ni(II) complex + N₂H₃˙ (6)

N₂H₃˙ + 3H₂O → N₂ + 3H₃O⁺ + 3e (7)

Overall reaction: N₂H₄ + 4H₂O → N₂ + 4H₃O⁺ + 4e (8)

Eqn (3) depicts the redox process of the nickel species confined in the Ni₃ molecules. The interaction of the Ni(III) center atom with N₂H₄ results in the formation of an adduct (eqn (4)). The adduct undergoes an internal reaction leading to the formation of the intermediate Ni–N₂H₃˙ complex (eqn (5)), which is also the rate-determining step. Eqn (6) and (7) represent the regeneration of the original Ni(II) complex and the subsequent chemical process, respectively. Eqn (8) is the chemical equation of the overall reaction, which can occur on a bare Au electrode surface.

To further characterize the electrode/solution interface behavior towards hydrazine hydrate oxidation, CVs of the Ni₃/Au electrode in 0.1 M Na₂SO₄ containing 50 μM hydrazine hydrate at various scan rates were recorded and the results are shown in Fig. 4a. As can be observed, the oxidation peak currents increase as the scan rate increases. Fig. 4b shows the linear dependence of the peak currents on scan rate from 25 to 800 mV s⁻¹. The linear regression equations is y = 5.681x + 0.924 (R = 0.9975), which indicates that the electrocatalytic oxidation of hydrazine hydrate is a surface-confined process. This result is in good agreement with the aforementioned rate-determining step rate (eqn (5)).

Fig. 4 (a) CV evolutions of Ni₃/Au obtained in 0.1 M Na₂SO₄ solution containing 50 μM hydrazine hydrate at scan rates ranging from 25 mV s⁻¹ to 0.8 V s⁻¹ (from inner to outer curves: 0.025, 0.05, 0.075, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.6, 0.7, and 0.8 V s⁻¹), (b) plot of current (I_p) vs. scan rate (from 25 to 800 mV s⁻¹).

The CVs at different scan rates further evidence the proposed EC mechanism. As shown in Fig. 4a, new cathode current peaks appear in the reverse scans. The potential difference between the cathode current peak and the oxidation peak of hydrazine hydrate is only about 40 mV, which is obviously less than that of the charge transfer process (59 mV/n). In an actual experimental operation, these cathode current peaks might not be readily observed at a relatively slow scan rate, due to no obvious intermolecular force acting on the adsorption layer. However, these peaks become increasingly clear with an increase in the scanning rate. The new peaks could be attributed to the desorption behavior of the [Ni(II)N₂H₃˙]_adduct from the electrode surface (eqn (6)).

Moreover, only a slight increase in the oxidation peak potentials is observed, which suggests a fast electron transfer rate. This could be ascribed to the Ni₃ SAM that immobilized on the surface of the Au electrodes, through which the considerable electrocatalytic activity of the Ni₃ complex and the DET between hydrazine and the electrode were achieved (see below for the calculated value of the electrocatalytic rate constant, k_cat). The fast response can also be verified via the chronoamperometry (CA) method (Fig. 5), wherein the oxidation currents stabilize rapidly within about 10 s. In addition, the facilitation for the diffusion of hydrazine molecules to the catalytic sites might be another factor for the fast response (see below for the detailed calculation of the diffusion coefficient, D).

Fig. 5 (A) Chronoamperograms obtained at the Ni₃/Au electrode in the absence (a) and in the presence of (b) 10 μM, (c) 20 μM, (d) 30 μM, and (e) 40 μM hydrazine hydrate in 0.1 M Na₂SO₄ solution. Inset: the dependence of current on the time^−1/2. (B) The dependence of I_cat/I_buff on time^1/2, the data were derived from the chronoamperograms.

The advantageous electrocatalytic reactivity of the modified electrode in the oxidation of hydrazine was tested using the CA method, which may offer informative and quantitative thermodynamic data employing the Cottrell effective constant. To determine the diffusion coefficient (D) and catalytic rate constant (k_cat) for the oxidation of hydrazine on the Ni₃ modified electrode, CA was carried out at different concentrations (10–40 μM) of hydrazine under a potential of 0.4 V and the results are shown in Fig. 5. The Cottrell equation eqn (9) was used to calculate the diffusion coefficient for hydrazine in this study given as follows:²⁸

where I_p is the catalytic current of the Ni₃ modified electrode in the presence of hydrazine, A is the geometric surface area of the electrode (0.0314 cm²), D is the diffusion coefficient (cm⁻² s⁻¹), C is the concentration of DA (mol mL⁻¹), and t is the time elapsed (s). From the plots of I vs. t^−1/2 at different concentrations (inset of Fig. 5), the average value for the diffusion coefficient of DA was calculated as 5.82 × 10⁻⁵ cm⁻² s⁻¹. This D value is relatively high as compared to some other electrodes,^21,22,33 and almost the same as that reported at a ZnO/CNTs/HPDB modified carbon paste electrode, which suggests a faster hydrazine diffusion at this Ni₃ modified electrode.

The catalytic rate constant (k_cat, M⁻¹ s⁻¹) for the electrochemical reaction between hydrazine and the redox sites of Ni₃ molecules can also be evaluated by chronoamperometry through eqn (10) as follows:^17,21

where I_cat is the catalytic current of the Ni₃ complex modified Au electrode in the presence of hydrazine, I_buff is the limiting current in the absence of hydrazine, and C is the concentration of hydrazine (M). Based on the slope of I_cat/I_buff vs. t^1/2 plots, k_cat can be obtained from a given DA concentration (Fig. 5B). From the values of the slopes, an average k_cat value was obtained for the oxidation of hydrazine as 8.57 × 10³ M⁻¹ s⁻¹, which is higher than the k_cat values reported for a zirconium hexacyanoferrate film (ZrHCFc) and bimetallic Au–Pt nanocomposite modified GCE,²¹ and single-walled carbon nanotubes–oxides of iron series elements (Fe, Co, and Ni) at a pyrolytic graphite electrode, wherein both were prepared via the coating/electrodeposition method.²⁶ It was also found that the present electrode system showed a higher reactivity (∼8.6 times) compared to the bare electrode for the oxidation of 50 μM hydrazine. These observations suggest that the formation of Ni₃ SAMs at the electrode surface provides a clear electrocatalytic effect on the oxidation of hydrazine, which offers the proposed modified electrode as a promising application in hydrazine detection.

3.4. Quantitative determination of hydrazine hydrate

In this study, the SWV technique was employed for the quantitative determination of a low concentration hydrazine, due to the minimum capacitive current and real improvement in sensitivity and LODs it offers. Fig. 6a shows the typical SWV curves of hydrazine with different concentrations ranging from 0.2 to 120 μM (for clarity, the inset of Fig. 6a shows the evolution of the SWV curves from 0.2 to 10 μM). As observed, the oxidation peak current (I_pa) increases with an increase in the concentration of hydrazine. Moreover, curve fitting shows a good linear relationship (R = 0.9985) between I_pa and hydrazine concentrations in the range of 0.2–50 μM with the linear regression equation expressed as y = 0.1362x + 0.07263 (inset of Fig. 6b). The limit of detection was calculated to be 0.07 μM, using the IUPAC definition LOD = 3s_b/q, where s_b is the standard deviation of the blank signal and q is the slope of the calibration curve. Table 1 compares the fabrication methods, the range of linearity, and the detection limit of this method with other reports for hydrazine detection. Compared to the analytical data in other literatures, the self-assembled electrochemical sensor proposed in this study provides a relatively low limit of detection.

Fig. 6 Typical SWVs of the Ni₃/Au electrode in 0.1 M Na₂SO₄ solution containing increasing concentrations of hydrazine hydrate (0.2, 0.4, 0.6, 0.8, 1.0, 2.0, 4.0, 6.0, 8.0, 10.0, 20.0, 30.0, 70.0, 100.0, and 120 μM), (b) the dependence of current on the concentration of hydrazine hydrate. The inset of (a) is a blowup of the SWV curves from 0.2 μM to 10 μM, and (b) is the linear relationship between peak current and concentration (from 0.2 to 50 μM).

Table 1 Comparison of hydrazine hydrate detection using various modified electrodes

Electrode	Fabrication method	Detection method	Linear range (μM)	LOD (μM)	Sensitivity (nA μM^−1)	Ref.
PdNPs/MWCNT modified GCE	Coating/electro-deposition	LSV	56–157	10	—	34
Curcumin/MWCNT modified GCE	Drop-dry	Chronoamperometry (I–t)	2.0–44	1.4	22.9	22
Cu–Pd/SPE	Electrodeposition	Flow injection amperometry	2–100	0.27	210	35
Au@Pd core–shell/rGO–GCE	Drop-dry	RDE (500 rpm)	2–40	0.08	270.2	14
GNPs/Ch/GCE	Eletrodeposition	LSV	0.5–500	0.1	84.3	33
ZrHCFc/Au–PtNPs/nanofibers modified GCE	Coating	Amperometry at RDE	0.15–112	0.09	∼52	21
PEG-CdS NPs modified Au	Coating-dry	Chronoamperometry (I–t)	0.1–1	0.06	890	18
ZnO/CNTs/HPDB/CPE	Mixing	SWV	0.02–0.7	0.009	184.8	17
FePc-linked-mPy-modified Au	LBL self-assembly	SWV	13–92	5	16.2	13
EPPGE–SWCNT–Ni	Coating/electro-decoration	LSV	—	5.3	517	26
NiHCF@TiO2 NPs modified GCE	Coating-dry	DPV	0.2–1	0.11	158.2	36
MWCNT/Ni nanocomposites GCE	Drop-dry	Amperometry	—	0.3	104.6	31
This work, Ni3 modified Au	Self-assembly	SWV	0.2–50	0.07	136.2	—

---

However, when the concentration of hydrazine was larger than 50 μM, the response of SWV gradually deviated from the calibration curve. Furthermore, when the concentration exceeded 0.2 mM, the SWV curves showed apparent differences with those of smaller concentrations (Fig. S3, ESI†). The oxidation peek potential positively shifted and new peaks corresponding to the oxidation of hydrazine at the bare Au electrodes appeared. This observation could be attributed to the saturation of the active sites in the Ni₃ complex molecules, which implies that the proposed sensor is particularly suitable for the detection of hydrazine in real samples with extremely low concentrations.

To verify the reproducibility and stability of this electrode, the SWV curves of a 50 μM hydrazine solution were recorded for twenty consecutive detections (Fig. S4, ESI†). The result showed a relative standard deviation (RSD) of 4.7%. As for six parallel fabricated electrodes, the RSD was 8.6%. The modified electrode was then stored in Ar at 4 °C over 15 days and 94% of its original current value was maintained. These results indicate the excellent stability and reliability of the sensor towards hydrazine determination. The good reproducibility of this electrode might be associated with a combination of following factors: (i) the self-assembly method allows the Ni₃ complex to functionalize the electrode surface via Au–S covalent bonds, and the amount of the electroactive substance on the electrode surface is relatively constant under a saturated adsorption, thus achieving a more stable signal than the physically fixed methods reported in some other literature; (ii) the closely arranged Ni₃ molecules occupy most of the Au surface, thus making it difficult for the adsorption of unwanted interferences; and (iii) the electrocatalytic reaction occurs at the center of the Ni₃ molecules, thus even if there was any adsorption on the electrode surface, the influence on the measured signal should be negligible.

The influence of various coexisting ions on the determination of hydrazine was investigated by adding various foreign species into 5 μM hydrazine solution. An approximately 100% ± 7.3% recovery caused by the addition of foreign substances is acceptable in trace analysis. The results reveal that K⁺, Na⁺, Ca²⁺, Mg²⁺, Fe²⁺, Cu²⁺, Zn²⁺, NO₂⁻, NO₃⁻, Cl⁻, Br⁻, SO₄²⁻, CO₃²⁻, and PO₄³⁻ (>200 fold) and sucrose, glucose, fructose, tartaric acid, and citric acid (>100 fold) have no obvious interference with hydrazine detection. However, the addition of 20-fold quantities of ascorbic acid and an equivalent concentration of NH₂OH led to a noticeable variation of peak current. This is reasonable because the oxidation potentials of ascorbic acid and NH₂OH are close to that of hydrazine.

3.5. Real sample determination

To evaluate the validity of the developed sensor for N₂H₄ determination, we extended our study to the detection of low concentrations of N₂H₄ in real samples. In this study, various environmental water samples (drinking water, tap water, and river water) were first spiked with different concentrations of hydrazine, and then analyzed by our proposed electrochemical assay and the HPLC method as a control comparison. As shown in Table 2, the results obtained from the Ni₃ sensor are in good agreement with those from the HPLC method, and the recovery values are in the range of 96–107%, which suggest that this Ni₃/Au electrode based hydrazine sensor has a promising application in real sample determination.

Table 2 Determination of hydrazine in various water samples (n = 4)

Samples	Original (μM)	Added (μM)	Found at Ni3/Au (μM)	Found in HPLC (mM)	RSD (%)	Recovery (%)
Drinking water	0	2	1.97	1.95	4.2	98.5
	0	5	4.86	4.92	2.2	97.2
	0	10	10.25	10.10	3.6	102.5
Tap water	0	2	2.12	1.93	4.1	106.0
	0	5	4.83	5.06	4.9	96.6
	0	10	9.91	9.95	3.5	99.1
River water	0	2	1.95	1.88	6.7	97.5
	0	5	5.34	5.21	5.3	106.8
	0	10	9.87	10.07	3.8	98.7

---

4. Conclusions

In conclusion, we have synthesized a new mercapto-terminated trinuclear Ni(II) complex [Ni₃(C₃H₆N₄S)₆(OH)₆](Ni₃) for the first time, which can self-assemble on the surface of Au electrodes due to the formation of Au–S bonds. Different electrochemical methods were then performed to investigate the behavior of the proposed Ni₃/Au electrode. The results demonstrate that the Ni₃/Au electrode could provide a faster electron transfer rate and better electrocatalytic oxidation abilities towards hydrazine than that of bare Au electrodes. The diffusion coefficient of DA and the catalytic rate constant were calculated to be 5.82 × 10⁻⁵ cm⁻² s⁻¹ and 8.57 × 10³ M⁻¹ s⁻¹, respectively. Using SWV as a quantitative analysis method, the detection limit of hydrazine could be as low as 0.07 μM, with a linear range from 0.2 to 50 μM. Moreover, the method was applied to the selective and precise analysis of hydrazine in various water samples. The good reproducibility and long-term stability of this Ni₃/Au electrode not only offer an opportunity for the detection of hydrazine in low concentrations, but also provide a platform to construct various sensors based on the well-designed mercapto-terminated structure.

Acknowledgements

Financial support from the Nature Science Foundation of China (No. 21173122, 21177067, 21505079), the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (14KJB150018), the Applied research program of Nantong city (BK2014071), and the Innovative Training Program for undergraduates (201410304015Z, 201510304082X) are gratefully acknowledged.

References

1. J. E. Troyan, Ind. Eng. Chem., 1953, 45, 2608–2612.
2. U. Ragnarsson, Chem. Soc. Rev., 2001, 30, 205–213.
3. E. Vernot, J. MacEwen, R. Bruner, C. Haun, E. Kinkead, D. Prentice, A. Hall, R. Schmidt, R. Eason and G. Hubbard, Toxicol. Sci., 1985, 5, 1050–1064.
4. M. Sun, J. Guo, Q. Yang, N. Xiao and Y. Li, J. Mater. Chem. B, 2014, 2, 1846–1851.
5. F. Liu, W. Li, F. Li and S. Sun, Environ. Monit. Assess., 2013, 185, 4153–4158.
6. K. Vijay, C. Nandi and S. D. Samant, RSC Adv., 2014, 4, 30712–30717.
7. A. D. Arulraj, M. Vijayan and V. S. Vasantha, Spectrochim. Acta, Part A, 2015, 148, 355–361.
8. H. T. Poh, T. S. Chwee and W. Y. Fan, RSC Adv., 2015, 5, 15159–15163.
9. X. Gu and J. P. Camden, Anal. Chem., 2015, 87, 6460–6464.
10. S. Subramanian, S. Narayanasastri and A. R. K. Reddy, Analyst, 2015, 140, 330–339.
11. J.-A. Oh and H.-S. Shin, J. Chromatogr. A, 2015, 1395, 73–78.
12. G. Li, S. Liu, Z. Sun, L. Xia, G. Chen and J. You, Food Chem., 2015, 170, 123–130.
13. K. I. Ozoemena and T. Nyokong, Talanta, 2005, 67, 162–168.
14. T. Pal, S. Dutta, C. Ray, S. Mallick, S. Sarkar and A. Roy, RSC Adv., 2015, 5, 51690.
15. J. Panchompoo, L. Aldous, C. Downing, A. Crossley and R. G. Compton, Electroanalysis, 2011, 23, 1568–1578.
16. A. Krittayavathananon, P. Srimuk, S. Luanwuthi and M. Sawangphruk, Anal. Chem., 2014, 86, 12272–12278.
17. H. Karimi-Maleh, M. Moazampour, A. A. Ensafi, S. Mallakpour and M. Hatami, Environ. Sci. Pollut. Res., 2014, 21, 5879–5888.
18. A. Umar, S. K. Kansal and S. K. Mehta, Sens. Actuators, B, 2013, 188, 372–377.
19. C. Karuppiah, R. Devasenathipathy, S. M. Chen, D. Arulraj, S. Palanisamy, V. Mani and V. Vasantha, Electroanalysis, 2015, 27, 485–493.
20. A. Maringa, E. Antunes and T. Nyokong, Electrochim. Acta, 2014, 121, 93–101.
21. M. B. Gholivand and A. Azadbakht, Electrochim. Acta, 2011, 56, 10044–10054.
22. L. Zheng and J.-F. Song, Sens. Actuators, B, 2009, 135, 650–655.
23. A. R. Fakhari, H. Ahmar, H. Hosseini and S. K. Movahed, Sens. Actuators, B, 2015, 213, 82–91.
24. J. Zhang, H. Liu, M. Dou, F. Wang, J. Liu, Z. Li and J. Ji, Electroanalysis, 2015, 27, 1188–1194.
25. Y. Liang, K. Wu, C. Ge, Y. Zhou, Y. Chen, Y. Tang and T. Lu, Fuel Cells, 2012, 12, 946–955.
26. A. S. Adekunle and K. I. Ozoemena, J. Solid State Electrochem., 2008, 12, 1325–1336.
27. G. Jiang, M. Wang, X. Gu, T. Chen, Y. Shang, Y. Tang, G. Jiang and Y. Shi, CrystEngComm, 2014, 16, 472–478.
28. X. Gu, G. Jiang, G. Jiang, T. Chen, W. Zhan, X. Li, S. Wu and S. Tian, Talanta, 2015, 137, 189–196.
29. K. C. Grabar, R. G. Freeman, M. B. Hommer and M. J. Natan, Anal. Chem., 1995, 67, 735–743.
30. E. Laviron, J. Electroanal. Chem. Interfacial Electrochem., 1974, 52, 355–393.
31. A. S. Adekunle and K. I. Ozoemena, J. Electroanal. Chem., 2010, 645, 41–49.
32. R. Ojani, J.-B. Raoof and S. Zamani, Bioelectrochemistry, 2012, 85, 44–49.
33. J. Li, H. Xie and L. Chen, Sens. Actuators, B, 2011, 153, 239–245.
34. X. Ji, C. E. Banks, A. F. Holloway, K. Jurkschat, C. A. Thorogood, G. G. Wildgoose and R. G. Compton, Electroanalysis, 2006, 18, 2481–2485.
35. C.-C. Yang, A. S. Kumar, M.-C. Kuo, S.-H. Chien and J.-M. Zen, Anal. Chim. Acta, 2005, 554, 66–73.
36. S. J. Sophia, S. Devi and K. Pandian, Int. J. Electrochem. Sci., 2012, 7, 6580–6598.

---

Footnote

† Electronic supplementary information (ESI) available: The crystallographic data of the Ni₃ complex, Fig. S1 and S2. CCDC 1435763. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra23809a

---