Well-defined gold nanoparticle@N-doped porous carbon prepared from metal nanoparticle@metal–organic frameworks for electrochemical sensing of hydrazine

Abstract

In this work, a composite material of Au nanoparticles (Au NPs) encapsulated in N-doped porous carbon (Au@NPC) was prepared through a one-pot thermolysis of Au NPs@zeolitic imidazolate framework (Au@ZIF-8) precursor. The obtained Au@NPC possessed a high specific surface area as well as superior thermal and chemical stability. The NPC shell functioned as a barrier to effectively prevent Au NPs from dissolution, migration and aggregation during the carbonization process and electrochemical testing, while it allowed the transit of electrolyte to the Au NPs surface. The core–shell structure and Au content were related to the preparation conditions. When the concentration of Au NPs was 0.5 mg mL⁻¹, the resulting Au@ZIF-8 could be carbonized to form a core–shell structure. The size of the encapsulated multi-core Au NPs slightly increased with enhancing carbonizing temperature (e.g. from 600 to 800 °C), while the Au content and surface area of the obtained composite material also increased. On this basis, a sensitive and stable electrochemical sensor was constructed for the detection of hydrazine. Under the optimized conditions, a linear dynamic range of 80 nM to 466.28 μM, with a satisfactory sensitivity of 2035.4 μA mM⁻¹ cm⁻² and a comparable detection limit of 8 nM (S/N = 3), was obtained. Moreover, it only lost 4.7% activity after one-month storage. This strategy could also be used to prepare other metal–carbon composite materials for constructing sensors.

---

1. Introduction

The preparation of composites with porous shells and metal cores has triggered much research interest because of their potential applications in catalysis, drug delivery and energy storage.¹ To obtain such core–shell structures, surface modifying and coating of metal cores are often performed with polymer,^2–4 metallic oxide,^5,6 and carbon matrices.^7,8 Owing to the inherent electrical conductivity, thermal stability and wide pH resistance, carbon materials are the preferred modifying/coating materials among them. The composites of metal nanoparticles encapsulated in porous carbon (MNPs@PC) are very beneficial for electrochemical applications because the encapsulated MNPs can keep their catalytic ability while the small active molecules diffuse in and out of the PC. On the other hand, the congregation of MNPs can be prevented by the PC shell even under hash conditions. For example, Pt@PC and Ru@PC materials showed excellent performance in catalyzing oxygen reduction and lithium battery, respectively.^9,10

Several approaches have been developed for the preparation of MNPs@PC. The template-assisted method is more widely used to achieve this goal. It generally involves multistep such as presynthesis of MNPs, coating of metal cores with carbon precursor, and carbonization. Liu et al. used resorcinol-formaldehyde resin as precursor to coat Au cores, and then through carbonization to synthesize Au@PC nanostructures.⁷ Besides resorcinol-formaldehyde resin, glucose¹¹ and dopamine¹² are also widely used as carbon source, while urea, sulfur power and sodium borate are often introduced as dopants to further improve the catalytic ability. However, the roasted products generally show single metal nuclear structure, the heterogeneous atoms can not uniformly dope in the carbon layer and the obtained PC may suppress the catalytic activity of MNPs due to its relatively low surface area.

Metal–organic frameworks (MOFs) are well-ordered crystalline solids with metal clusters bridged by organic ligands, and have tunable pore sizes, large surface area, extra high porosity and chemistry tenability.¹³ They have been recognized as ideal sacrificial templates to prepare metallic oxide, PC and MNPs/PC composites through controlling thermolytic conditions.^14,15 Owing to the confinement effect of MOFs, uniform MNPs and high quality nanoporous carbon materials can be derived from the in situ transformation of regularly arranged metal nodes and breakage of abundant organic bridging ligands.^16–18 The advantages of MOFs-derived porous materials lie in their special structural features such as large specific surface area, high porosity with inter-connected pores, good structural integrity and heterogeneous atoms uniformly doped in the carbon layer. For example, uniformly dispersed Co and Cu NPs within carbon matrix were produced by thermal decomposition of ZIF-67 (i.e. Co(MIM)₂, MIM = 2-methylimidazole) and HKUST-1 (i.e. Cu₃(BTC)₂, BTC = 1,3,5-benzene tricarboxylate) in N₂ gas atmosphere, respectively. The Brunauer–Emmett–Teller (BET) surface area of Cu NPs@PC was up to 285.03 m² g⁻¹ (ref. 18) and the BET surface area of Co NPs@PC could be adjusted through changing calcination temperatures, which ranged from 32.98 to 303.5 m² g⁻¹.¹⁹ Even though MNPs@PC core–shell structures can be generated via MOFs-templated strategy, the reported products are still limited and no noble metals are concerned. It is necessary to extend this method to preparing multiple metal composites.

Recently, hybrid materials composed of MNPs and MOFs were widely investigated, and a large number of MNPs@MOF hybrid materials were reported, including Pd@MOF,²⁰ Au@MOF,²¹ PtPd@MOF²² and Pt–ZnO@MOF.²³ The pyrolysis of MNPs@MOF could lead to the formation of PC, while the MNPs were encapsulated. Thus different MNPs@PC materials could be prepared.⁹ For example, magnetic Fe₃O₄@PC/Cu nanocomposite was successfully prepared via directly carbonizing Fe₃O₄@HKUST-1, which showed powerful photocatalytic activity for the degradation of methylene blue.²⁴ Zhang et al. prepared magnetic FeCo alloy NPs embedded in nanoporous carbon by the thermal decomposition of Fe₃O₄@ZIF-67.²⁵ Hitherto such reports are still few. To our knowledge, there is no report on the investigation of MNPs@MOF-derived composites as electrocatalyst to date.

Nitrogen-doped carbon (NPC) showed high sensitivity in sensing applications. The carbonized product of ZIF-8 (i.e. Zn(MIM)₂) is NPC, and it is widely used in electrochemical field because of its simple preparation, low cost, high surface area, favorable hydrophilicity and uniformly distributed nitrogen content in carbon layer.²⁶ Chen's group reported the electrocatalytic oxidation of ascorbic acid (AA), dopamine (DA) and uric acid (UA) on NPC modified electrode,²⁷ and they also loaded PtRu on NPC support for methanol oxidation.²⁸ Ling et al. fabricated an electrochemical glucose biosensor by using NPC as substrate material.²⁹ We expect that MNPs@NPC materials are promising electrode material.

Hydrazine is widely applied in industry, agriculture, pharmacology and so on.^30,31 However, it is highly toxic and is classified as a human carcinogen in group B2 with a low threshold limit value (TLV) of 10 ppb by the United States Environmental Protection Agency (USEPA).^32,33 Animal experiments have shown that hydrazine is mutagenic and carcinogenic, which can severely injury lung, liver, kidney, brain, and spinal cord.^34,35 Hence, the accurate determination of trace hydrazine discharged to the environment is rather essential. Many studies have shown that Au-based modified electrodes have unique catalysis for hydrazine oxidation and suit for its detection.^36–38

Herein, a core–shell structured Au@NPC was prepared by carbonizing Au@ZIF-8 precursor. The high dispersive Au NPs provided plentiful catalytic active sites, while the porous carbon shell functioned as protective layer, which could significantly improve the stability of Au NPs in electrocatalytic applications. The composite showed excellent performance in the electrocatalytic oxidation and amperometric determination of hydrazine. To the best of our knowledge, this is the first time to use an in situ approach to synthesize such a unique electrode material for electrochemical detection.

2. Experimental

2.1. Reagents and apparatus

Zn(NO₃)₂·6H₂O, Mg(NO₃)₂·6H₂O, HAuCl₄·4H₂O, sucrose, glucose, Na₂HPO₄·12H₂O, NaH₂PO₄·2H₂O, hydrazine, sodium citrate dehydrate, hydrochloric acid, 2-nitrophenol, 4-nitrophenol, 3-nitrotoluene, 2-nitrophenyl and methanol (MeOH) were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China); 2-methylimidazole and poly(vinylpyrrolidone) (PVP, M_w = 55 000) were obtained from Aladdin Chemistry Co., Ltd. (Shanghai, China). The support electrolyte was 0.1 M phosphate buffer solution (PBS, pH = 6.0), and it was prepared with Na₂HPO₄·12H₂O and NaH₂PO₄·2H₂O.

Cyclic voltammetry and chronoamperometric experiments were performed with a CHI 832C electrochemical workstation (CH Instrument Company, Shanghai, China). A conventional three-electrode system was employed. The working electrode was a modified glassy carbon electrode (GCE, diameter: 3 mm), and the auxiliary and reference electrodes were a platinum wire and a saturated calomel electrode (SCE), respectively. The scanning electron microscope (SEM) image was obtained by using field emission SEM (ZEISS, Germany). Transmission electron microscopy (TEM) image was obtained by using a JEM-2100 transmission electron microscope with an accelerating voltage of 200 kV. X-ray diffraction data (XRD) were recorded with a PANalytical X'Pert Pro diffractometer (Holland) using Cu Kα radiation (40 kV, 40 mA) with a Ni filter. X-ray photoelectron spectroscopy (XPS) analysis was carried out on ESCALAB 250Xi (Thermo Fisher) X-ray photoelectron spectrometer with Al Kα X-ray radiation for excitation. Energy dispersive X-ray spectroscopy (EDX) was performed by using a Hitachi X-650 SEM (Hitachi Co., Japan). N₂ sorption analysis was carried out on a Quantachrome Autosorb-iQ gas sorption analyzer. Before gas analysis, the samples were evacuated for 12 h at 100 °C under vacuum. The pore size distribution (PSD) was calculated by using the Horvath–Kawazoe method.

2.2. Preparation of PVP-stabilized Au NPs

The Au NPs were prepared by reducing HAuCl₄ with sodium citrate.³⁹ In a typical procedure for the synthesis of 13 nm Au NPs, sodium citrate solution (1%, 4.5 mL) was rapidly injected into a vigorous boiling aqueous solution of HAuCl₄ (0.01%, 150 mL) under stirring. After heating for 20 min, the resulting deep red suspension was removed from the heat and cooled to room temperature.

An aqueous solution of PVP (0.5 g, 20 mL) was added to the Au NPs solution under stirring, and the mixture was further stirred for 24 hours at room temperature. The PVP-stabilized Au NPs were collected by centrifugation at 14 000 rpm for 30 min, washed with MeOH for three times, and finally dispersed in MeOH to obtain suspensions with different concentrations (about 0.1 mg mL⁻¹, 0.5 mg mL⁻¹ and 0.75 mg mL⁻¹).

2.3. Synthesis of Au@ZIF-8, ZIF-8 and Au/ZIF-8

The Au@ZIF-8 composite was synthesized according to previous report with minor modification.⁴⁰ Briefly, 10 mL PVP-stabilized Au NPs suspension (containing 0.1 mg mL⁻¹, 0.5 mg mL⁻¹ or 0.75 mg mL⁻¹ Au NPs), 50 mL MeOH solution of Zn(NO₃)₂·6H₂O (25 mM), and 50 mL MeOH solution of 2-methylimidazole (25 mM) were mixed and then allowed to react at room temperature for 24 h without stirring. The product was collected by centrifugation, washed with MeOH, and dried overnight in vacuum. The obtained samples were denoted as Au@ZIF-8-1, Au@ZIF-8-2, and Au@ZIF-8-3, respectively. Pure ZIF-8 was prepared by the same procedure except not adding PVP-stabilized Au NPs suspension.

The Au/ZIF-8 composite was prepared by directly mixing ZIF-8 and PVP-stabilized Au NPs suspension at room temperature for 24 h under stirring condition.

2.4. Synthesis of Au@NPC and NPC

The obtained Au@ZIF-8-2 (0.1 g) was transferred into an alumina boat, and then put in a tube furnace. The air in the tube was expelled with N₂ stream for about 1 h. Afterward, the furnace was heated to the required temperature (i.e. 600 °C, 700 °C and 800 °C) at a heating rate of 3 °C min⁻¹. It was maintained at the temperature for 1 h and then let it cool to room temperature. The resultant materials were denoted as Au@ZnO/NPC-600, Au@ZnO/NPC-700 and Au@ZnO/NPC-800, where the numbers represented carbonization temperature of 600 °C, 700 °C and 800 °C, respectively.

The samples of Au@ZnO/NPC-600, Au@ZnO/NPC-700 and Au@ZnO/NPC-800 were dispersed into 2 M HCl with the aid of ultrasonication and stirred overnight. The resultant materials were collected by centrifugation, washed with distilled water until pH 7.0, and then dried in an oven. These obtained samples were labeled as Au@NPC-600, Au@NPC-700 and Au@NPC-800, respectively (Scheme 1).

Scheme 1 Schematic diagram of the preparation of Au@NPC-700.

Similarly, ZIF-8 and Au/ZIF-8 were directly carbonized at 700 °C and etched by 2 M HCl under stirring overnight. The products were labeled as NPC-700 and Au/NPC-700, respectively.

2.5. Preparation of modified electrodes

Prior to modification, the bare GCE was polished with slurry alumina (1.0, 0.3 and 0.05 μm, respectively), and then washed thoroughly with ultra-pure water with the aid of ultrasonication. To prepare the modified electrodes, 1 mg of the as-obtained sample was dispersed into 1 mL water to give homogeneous suspension. Then 5 μL suspension was coated on the cleaned GCE and the electrode was then dried at room temperature. Thus, the Au@NPC-600/GCE, Au@NPC-700/GCE, Au@NPC-800/GCE, PVP-stabilized Au NPs/GCE and NPC-700/GCE modified electrodes (or sensors) were obtained, respectively.

2.6. Electrochemical measurements

Cyclic voltammograms (CVs) of the sensors were recorded in 0.1 M PBS (pH 6.0) containing 0.1 mM hydrazine. The potential range was −0.2 to 0.9 V (vs. SCE), and the scan rate was 50 mV s⁻¹. For the chronoamperometric measurement the applied potential was 0.4 V (vs. SCE). The response currents were measured after successively adding hydrazine under magnetic stirring.

3. Results and discussion

3.1. Morphological and structural characterization

The Au@ZIF-8 precursor was synthesized by a facile and mature method, which involved the functionalization of 13 nm Au NPs with suitable polymer and the crystallization of ZIF-8. Here poly(vinylpyrrolidone) (PVP) was used to stabilize Au NPs in methanol as well as a capping agent to control the size and shape of Au NPs.⁴¹ The adsorption of amphiphilic PVP on Au NPs surface provided the NPs with enhanced affinity to coordination spheres. Then crystallization was performed. The crystalline nature and composition of the products were characterized by XRD (Fig. 1A). The diffraction patterns of the hybrid material were in line with the patterns of ZIF-8 and Au,²² indicating that it was composed of Au and ZIF-8 and the crystallinity of ZIF-8 was not disrupted by the incorporation of Au NPs.

Fig. 1 XRD patterns of Au, ZIF-8 and Au@ZIF-8-2 (A); XRD patterns of NPC-700, Au@NPC-800, Au@NPC-700 and Au@NPC-600 ((B) from bottom to top).

The morphology of ZIF-8 and Au@ZIF-8-2 was shown in Fig. 2 and S1A.† It could be seen that a large number of Au NPs located at the centre of the rhombic dodecahedral ZIF-8 after being encapsulated. The diameter of Au@ZIF-8 crystals became small and the crystals aggregated with each other. The permanent porosity of evacuated ZIF-8 and Au@ZIF-8-2 composites was confirmed by nitrogen-sorption measurements (Fig. S2A†). The composites displayed type I isotherm and the adsorption–desorption isotherm branches were reversible, confirming that they were predominantly microporous materials. The Brunauer–Emmett–Teller (BET) surface areas of pure ZIF-8 and Au@ZIF-8-2 were calculated to be 1975.35 and 1471.01 m² g⁻¹, respectively. The Au@ZIF-8-2 showed smaller specific surface area because of the effect of non-porous Au NPs and PVP on the mass of the composite. Moreover, the incorporation of Au NPs did not alter the pore-size distribution of the ZIF-8 matrix (Fig. S2B†), consistent with the fact that the introduced nanoparticles were too large to occupy the cavities (11.6 Å) of the framework.⁴² All these indicated that core–shell structured Au@ZIF-8-2 was successfully synthesized.

Fig. 2 TEM images of ZIF-8 (A), Au@ZIF-8-1 (B), Au@ZIF-8-2 (C), Au@ZIF-8-3 (D), NPC-700 (E), Au@NPC-600 (F), Au@NPC-700 (G) and Au@NPC-800 (H). The insets were the SEM images of ZIF-8 (A) and Au@ZIF-8-3 (D).

In general, the coating thickness is important for application, thin coating and high density of MNPs are favourable.⁴³ Here, increasing the concentration of Au NPs, the diffraction peak intensity of Au in the Au@ZIF-8-x increased (Fig. S2C†) and the color of the resultant material became deep (Fig. S2D†). At the same time, the particle diameter became small and more crystals aggregated (Fig. 2A–C). A possible reason was that the large number of PVP-stabilized Au NPs could provide superabundant nucleation centres for the formation of many small ZIF-8, which encapsulated a few Au NPs. And there was no enough mother solution for the small crystals to grow into large monodispersed dodecahedron crystals and finally aggregated together because of their high surface energy.⁴³ Nevertheless, when the concentration of Au NPs suspension was further increased to 0.75 mg mL⁻¹, the diameter of resulting Au@ZIF-8-3 did not further decrease and presented the same size as pure ZIF-8. The composite showed broken shell and could not maintain a core–shell structure (Fig. 2D). So Au@ZIF-8-2 was applied in the following experiments because it contained more Au NPs and maintained well core–shell structure.

After pyrolysis at temperature ranging from 600 to 800 °C for 1 h in N₂ atmosphere, the zeolite-type sodalite structure of ZIF-8 collapsed and ZnO formed, which was homogenously confined in the porous carbon.⁴⁴ The characteristic peak intensity of ZnO in Au@ZnO/NPC-x correlated well with the elevated pyrolysis temperature (Fig. S3A†).

Subsequently, ZnO in Au@ZnO/NPC-x was removed by acid leaching and the remained Au@NPC-x was recovered by filtration and thoroughly washed. The resultants were characterized using powder XRD (Fig. 1B). The reflected peaks around 38.2°, 43.3°, 64.6°, 77.9° and 81.8° were assigned to Au (111), (200), (220), (311) and (222), respectively. They were consistent with the standard XRD pattern of Au (JCPDS 04-0784). The diffraction peak of NPC-700 (at 25°) corresponded to carbon diffraction.¹⁸ Additionally, no peaks of impurity were observed in the XRD pattern, indicating that ZnO had been removed thoroughly. This was further supported by the analytical result of EDS of Au@NPC-700 composite (Fig. S3B†).

Fig. 2F–H showed the particle sizes and the morphology of Au@NPC-x. As could be seen, NPC-700 (Fig. 2E) displayed vague rhombic dodecahedral shape with amorphous carbon structure. The images of Au@NPC-600 (Fig. 2F), Au@NPC-700 (Fig. 2G) and Au@NPC-800 (Fig. 2H) clearly demonstrated the formation of carbon shell. The composites showed porous structure due to the volatilization of produced gas and removing of ZnO.⁴⁵ The Au NPs were encapsulated in carbon shell and their size increased from 15 to 25 nm, 25 to 40 nm and 30 to 80 nm with increasing temperature, respectively (Table S1†). The Au NPs were highly dispersed, which could be attributed to the perfect core–shell structure of the precursors. To prove this point a contrast experiment was performed. Au/ZIF-8 composite was prepared by directly mixing ZIF-8 and PVP-stabilized Au NPs suspension under stirring condition. In this case the Au NPs were mainly adsorbed on the surface of ZIF-8. They did not form core–shell structure and some Au NPs piled up together (Fig. S1B†). After pyrolysis they became irregular and seriously agglomerated (Fig. S1C†). This demonstrated that the carbon shell functioned as a barrier to effectively prevent Au NPs from migration and aggregation during calcination process.

The pore structure is another pivotal factor determining the catalytic performance. As shown in Fig. 3A, the BET surface area of as-obtained Au@NPC-x was much smaller compared with that of primary Au@ZIF-8-2 due to the decomposing of the well-defined microporous of ZIF-8 (Fig. S2A†). All of them showed typical type IV isotherm with a capillary condensation step and a hysteresis loop during the desorption branch, indicating the presence of both micropores and mesopores. The presence of porous structure could allow electrolyte to reach the active sites and simultaneously promoted the catalytic process.⁴⁶ In addition, with the x increasing, the value of BET and total pore volume of Au@NPC-x gradually increased (Table S1†). Besides, X-ray photoelectron spectroscopy (XPS) was used to explore the surface information of the calcined product. As shown in Fig. 3B, the Au 4f spectrum contained a doublet at binding energy of 84.1 and 87.7 eV, assigned to Au 4f_7/2 and 4f_5/2 lines, respectively. These values unambiguously suggested that Au NPs were still in metallic form, indicating the formation of Au NPs encapsulated in carbon shell.⁴⁷ Moreover, all the samples showed similar Au 4f peaks, but the peak intensity increased with elevating carbonization temperature, suggesting that the Au species exhibited similar chemical status except different Au contents (Table S1†). It was possibly due to the decreased N content because of the higher volatility of N species at higher temperature.⁴⁴ The XPS survey data of Au@NPC-700 was shown in Fig. 3C, indicating the binding energy (BE) of different atoms (i.e. Au, C, N and O). The peaks fitting the C 1s and N 1s XPS spectra were shown in Fig. 3D and E. The peaks at 284.6, 285.2, 285.8, 288.1 and 289.3 eV in the C 1s XPS spectrum corresponded to the sp²-hybridized graphitic carbon, C–N, C–O, C=O and O–C=O, respectively; and the N heteroatoms in NPC was made up of pyridinic N (at a BE of 398.5 eV), pyrrolic N (at a BE of 399.9 eV), graphitic N (at a BE of 401.1 eV) and oxidized N (at a BE of 402.2 eV). The status of C and N in the porous carbon was similar to previous reported results.^44,48

Fig. 3 Nitrogen-sorption isotherms (A) and high resolution Au 4f XPS data (B) of Au@NPC-600, Au@NPC-700 and Au@NPC-800; the survey XPS data (C), high resolution C 1s data (D) and high resolution N 1s data (E) of Au@NPC-700.

3.2. Electrocatalytic activity toward hydrazine oxidation

The electrocatalytic activity of different electrodes was compared. As shown in Fig. 4A, the blank signals almost did not affect the detection of hydrazine. At bare GCE hydrazine did not produce oxidation peak in the potential range probably due to its high overpotential (curve a). The NPC-700/GCE (curve b) exhibited poor electrocatalysis toward the oxidation of hydrazine, while hydrazine produced an obvious oxidation peak on Au@NPC-x modified electrodes due to the catalysis of Au NPs.⁴⁹ At the Au@NPC-600/GCE (curve c) the oxidation of hydrazine occurred at higher potential and a smaller oxidation peak was observed. This was due to the small surface area (Fig. S4†) as well as the low pore volume (Table S1†) of Au@NPC-600. The Au@NPC-700/GCE and Au@NPC-800/GCE had larger surface area (Fig. S4†) and exhibited more sensitive response to hydrazine. Among them the Au@NPC-700/GCE (Fig. 4A, curve d) was more sensitive than Au@NPC-800/GCE (Fig. 4A, curve e). The different current responses were related to the size of the encapsulated Au NPs. The agglomeration of Au NPs in Au@NPC-800 led to the decrease of active sites. Interestingly, the oxidation peak of hydrazine on Au@NPC-700/GCE located at 0.3 V, which was more positive than that of conventional Au NPs modified electrode.⁵⁰ This indicated that the electrochemical process was related to the small pores to some extent, which could affect the mass transfer.

Fig. 4 (A) CVs of bare GCE (a), NPC (b), Au@NPC-600 (c), Au@NPC-700 (d) and Au@NPC-800 (e) in 0.1 M PBS (pH 6.0) containing 0 (black lines) and 0.1 mM (red lines) hydrazine. (B) Influence of operating potential on the response current of 5 μM hydrazine. Influence of the amount of Au@NPC-700 composite (C) and solution pH (D) on the response current of 0.1 mM hydrazine.

The electrochemical oxidation of hydrazine in aqueous solution was studied extensively, and the established electrochemical reaction was a four-electron process with the final product N₂:^51,52

N₂H₅⁺ → N₂ + 5H⁺ + 4e⁻

In this case, the catalysis of Au NPs may be related to its adsorption or complexing action, since Au could not directly oxidize hydrazine and no oxidant gold oxide occurred at lower potential. But the detailed mechanism is still unclear so far.

In addition, the CVs of the Au@NPC-700 and PVP-stabilized Au NPs modified electrodes at different scan rates were recorded (Fig. S5A and B†). The oxidation peak current (I_pa) of hydrazine at Au@NPC-700/GCE was proportional to the square root of scan rate in the range of 10–200 mV s⁻¹. The linear equation could be expressed as I_pa (μA) = 0.6394v^1/2 (mV^1/2 s^−1/2) + 1.989, with a correlation coefficient of 0.9994, indicating that the electrochemical process was controlled by diffusion. However, further increasing the scan rate, the peak current did not increase linearly. Nevertheless, the peak current of hydrazine at the PVP-stabilized Au NPs/GCE was linear to the square root of scan rate in the studied range of 10–600 mV s⁻¹. The reason for this phenomenon may be that the channels of PC hindered the rapid pass of hydrazine molecule and the oxidation product. The influence of detection potential was evaluated through amperometric experiment (Fig. 4B). As a result, the response current increased with the applied potential changing from 0.1 V to 0.4 V. But further increasing potential, the respond current kept almost unchanged, while the noise increased. In this case, 0.4 V was chosen.

Furthermore, other experimental conditions such as the modifying amount of Au@NPC-700 and solution pH were also investigated, respectively (Fig. 4C and D). As can be seen, the peak current of hydrazine increased with increasing the volume of Au@NPC-700 suspension up to 5.0 μL, then it decreased (Fig. 4C). This was related to the change of electrode surface area and electron transfer resistance. When the amount of Au@NPC-700 was too much, the electrode area kept almost unchanged but the resistance increased due to the coating thickness growing. Moreover, the response current increased with the solution pH changing from 4.5 to 6.0, but further increasing solution pH to 8.0, the respond current decreased (Fig. 4D). This was related to the protonization (pK_a = 8.1) of hydrazine and proton transfer in the electrochemical process, which must show opposite effect. In addition, the catalysis of Au NPs also changed with solution pH. Hence, the influence of solution pH was very complicated, and pH 6 could be considered a good balance point in this case. Therefore, 5.0 μL of Au@NPC-700 suspension was adopted for preparing modified electrode and solution pH of 6.0 was selected.

3.3. Analytical performance

Under the optimized conditions, the chronoamperometric curves of hydrazine at Au@NPC-700/GCE were recorded. Fig. 5A showed the typical current response of Au@NPC-700/GCE to successive addition of hydrazine in 0.1 M PBS (pH 6.0). The current increased upon the increase of hydrazine concentration (Fig. 5B), and the linear range was 0.08–466.28 μM. The linear regression equation was I_pa (μA) = 0.1438c (μM) + 0.6787 (R² = 0.9972). The sensitivity was calculated to be 2035.4 μA mM⁻¹ cm⁻², based on the geometric area of GCE. The limit of detection (LOD, S/N = 3) was 8 nM. Compared with other hydrazine sensors, the Au@NPC-700/GCE offered wide linear range and high sensitivity (Table 1).

Fig. 5 (A) Amperometric response curves of Au@NPC-700/GCE upon successive addition of hydrazine at 0.4 V. Left inset: the amperometric response of hydrazine at low concentration level. (B) The relationship between hydrazine concentration and current signal for Au@NPC-700/GCE. Right inset: the relationship between hydrazine concentration and current signal at low concentration level.

Table 1 Comparison of the performance of Au@NPC-700/GCE with previously reported hydrazine sensor

Electrodes	Method	Sensitivity (μA mM^−1)	Linear range (μM)	Detection limit (μM)	Stability ^e (%)	Ref.
a DPV: differential pulse voltammetry.b PDTYB: poly 4,5-dihydro-1,3-thiazol-2-ylsulfanyl-1,2-benzenediol.c CV: cyclic voltammetry.d LSV: linear sweep voltammetry.e Ratio of the response current after storing for some time to its initial response value.
Au-SH-SiO2@Cu-MOF/GCE	DPV^a	100	0.04–500	0.01	—	51
Au/ZnO-MWCNTs/GCE	Amperometry	42.8	0.5–1800	0.15	—	53
Au/NH2-MIL-125(Ti)/GCE	Amperometry	—	0.01–10	0.0005	90.6 (a week)	54
		—	10–100
Au/PDTYB/MWCNTs/GCE^b	CV^c	235	2–130	0.6	92 (10 days)	55
		137	130–350
Au/activated carbon/GCE	Amperometry	394	0.44–208	0.0063	92.27 (a week)	52
AuCu-EGN-IL/GCE	Amperometry	56.7	0.2–110	0.1	71 (a month)	50
Nafion-TiO2-CNT/GCE	LSV^d	58	0.35–163	0.02	91.7 (a month)	56
Au@NPC-700/GCE	Amperometry	143.8	0.08–466.28	0.008	95.3 (a month)	This work

---

3.4. Selectivity, reproducibility and stability

In order to evaluate the selectivity of Au@NPC-700/GCE, 20-fold of 2-nitrophenol, 4-nitrophenol, 100-fold 2-nitrobenzene, 3-nitrobenzene, 0.08-fold ascorbic acid, uric acid, dopamine and 200-fold glucose, sucrose, Mg²⁺, Cu²⁺, Mn²⁺, NO₃⁻, Cl⁻ were tested under the optimal conditions in presence of 5 μM hydrazine. No obvious change of current signal was observed.

The repeatability and reproducibility of the sensor were evaluated by amperometric determination of 5 μM hydrazine. Eight Au@NPC-700/GCEs were prepared independently by the same way. As a result, the relative standard deviation (RSD) of the response current was 3.8% (n = 8). Furthermore, eight successive measurements using an electrode gave an RSD of 2.3% (n = 8). This indicated that the electrode had good reproducibility and repeatability for the detection of hydrazine.

The stability was confirmed by comparing the corresponding TEM images for before and after electrochemical measurement. As shown in Fig. S6,† after five cycles of consecutive runs, the Au@NPC-700 still exhibited intact morphology (compared with Fig. 2G), and the highly dispersed Au NPs were coated by PC without agglomeration or deformation.

The storage stability of the sensor was also examined. When the Au@NPC-700/GCE was stored in air for 15 days, the response current retained 98.1% of its initial value; after a month of storage, it still retained 95.3% (Table 1). These reflected the long-term stability of Au@NPC-700/GCE.

The superior electrochemical performance of Au@NPC-700/GCE should be attributed to following reasons: (i) NPC had large specific surface area; (ii) the pores in the carbon shell acted as channel for the easy diffusive analytes and the carbon shell prevented Au NPs from dissolution and aggregation during electrochemical testing; (iii) the multi-core Au NPs provided plentiful catalytic active sites.

3.5. Sample analysis

In order to evaluate the practical feasibility of the proposed method for the determination of hydrazine, three water samples from different sources were analyzed under optimized conditions. But no hydrazine was detected in the samples. Then, they were spiked with different concentrations of hydrazine and the recovery was measured. The results were shown in Table 2 and the recovery was acceptable.

Table 2 Results of the recovery test

Samples	Added (μM)	Found (μM)	Recovery (%)	RSD (%) (n = 3)
1	0	0	—	—
	8.0	7.2	90	2.5
	80.0	82.8	103	1.4
	200	203	102	0.9
2	0	0	—	—
	8.0	7.4	93	2.0
	80.0	78.6	98.3	1.6
	200	213	107	1.3
3	0	0	—	—
	8.0	8.1	100	1.0
	80.0	75.4	94.3	0.4
	200	199	99.5	0.8

---

4. Conclusions

Core–shell structured Au@NPC composite was firstly prepared by directly carbonizing Au@ZIF-8. The core–shell structure and Au content was related to the preparation conditions. When the concentration of Au NPs was 0.5 mg mL⁻¹, the resulting Au@ZIF-8 could be carbonized at 700 °C to form well core–shell structure. In the composite Au NPs showed high dispersion. Meanwhile, the pore structure of the shell was favorable for the penetration of hydrazine to Au NPs surface. The resulting electrochemical sensor Au@NPC-700/GCE thus showed excellent electrochemical response. The sensitivity of the sensor was up to 2035.4 μA mM⁻¹ cm⁻² and it only lost 4.7% activity after one month storage, which was quite good in comparison with previous reports. The strategy could also be used to prepare other metal–carbon composite materials for constructing sensors.

Acknowledgements

The authors appreciate the financial support of the National Natural Science Foundation of China (Grant No. 21075092, 21277105).

References

1. L. Tan, D. Chen, H. Liu and F. Tang, Adv. Mater., 2010, 22, 4885–4889.
2. Y. Mao, W. Jiang, S. Xuan, Q. Fang, K. C. F. Leung, B. S. Ong, S. Wang and X. Gong, Dalton Trans., 2015, 44, 9538–9544.
3. R. L. Zhang, S. Xu, J. Luo, D. J. Shi, C. Liu and X. Y. Liu, RSC Adv., 2014, 4, 25106.
4. M. E. Lynge, R. van der Westen, A. Postma and B. Stadler, Nanoscale, 2011, 3, 4916–4928.
5. C. Yu, K. Yang, Y. Xie, Q. Fan, J. C. Yu, Q. Shu and C. Wang, Nanoscale, 2013, 5, 2142.
6. J. Xi, Y. Zhang, N. Wang, L. Wang, Z. Zhang, F. Xiao and S. Wang, ACS Appl. Mater. Interfaces, 2015, 7, 5583–5590.
7. R. Liu, F. Qu, Y. Guo, N. Yao and R. D. Priestley, Chem. Commun., 2014, 50, 478–480.
8. T. Xu, N. Liu, J. Yuan and Z. Ma, Biosens. Bioelectron., 2015, 70, 161–166.
9. G. Li, H. Kobayashi, K. Kusada, J. M. Taylor, Y. Kubota, K. Kato, M. Takata, T. Yamamoto, S. Matsumura and H. Kitagawa, Chem. Commun., 2014, 50, 13750–13753.
10. L. Zhu, Y. Jiang, J. Zheng, N. Zhang, C. Yu, Y. Li, C. W. Pao, J. L. Chen, C. Jin, J. F. Lee, C. J. Zhong and B. H. Chen, Small, 2015, 11, 4385–4393.
11. Z. Z. Jiang, Z. B. Wang, D. M. Gu and E. S. Smotkin, Chem. Commun., 2010, 46, 6998–7000.
12. R. Liu, S. M. Mahurin, C. Li, R. R. Unocic, J. C. Idrobo, H. Gao, S. J. Pennycook and S. Dai, Angew. Chem., Int. Ed., 2011, 50, 6799–6802.
13. H. Li, M. Eddaoudi, M. O'Keeffe and O. M. Yaghi, Nature, 1999, 402, 276–279.
14. B. Liu, H. Shioyama, T. Akita and Q. Xu, J. Am. Chem. Soc., 2008, 130, 5390–5391.
15. A. Aijaz, N. Fujiwara and Q. Xu, J. Am. Chem. Soc., 2014, 136, 6790–6793.
16. J. K. Sun and Q. Xu, Energy Environ. Sci., 2014, 7, 2071.
17. M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O'Keeffe and O. M. Yaghi, Science, 2002, 295, 469–472.
18. H. Tan, C. Ma, L. Gao, Q. Li, Y. Song, F. Xu, T. Wang and L. Wang, Chem.–Eur. J., 2014, 20, 16377–16383.
19. X. Wang, J. Zhou, H. Fu, W. Li, X. Fan, G. Xin, J. Zheng and X. Li, J. Mater. Chem. A, 2014, 2, 14064.
20. M. Zhao, K. Deng, L. He, Y. Liu, G. Li, H. Zhao and Z. Tang, J. Am. Chem. Soc., 2014, 136, 1738–1741.
21. L. He, Y. Liu, J. Liu, Y. Xiong, J. Zheng, Y. Liu and Z. Tang, Angew. Chem., Int. Ed., 2013, 52, 3741–3745.
22. Y. Huang, Y. Zhang, X. Chen, D. Wu, Z. Yi and R. Cao, Chem. Commun., 2014, 50, 10115.
23. L. Lin, T. Zhang, H. Liu, J. Qiu and X. Zhang, Nanoscale, 2015, 7, 7615–7623.
24. Y. F. Zhang, L. G. Qiu, Y. P. Yuan, Y. J. Zhu, X. Jiang and J. D. Xiao, Appl. Catal., B, 2014, 144, 863–869.
25. X. Zhang, G. Ji, W. Liu, B. Quan, X. Liang, C. Shang, Y. Cheng and Y. Du, Nanoscale, 2015, 7, 12932–12942.
26. J. Tang, R. R. Salunkhe, J. Liu, N. L. Torad, M. Imura, S. Furukawa and Y. Yamauchi, J. Am. Chem. Soc., 2015, 137, 1572–1580.
27. P. Gai, H. Zhang, Y. Zhang, W. Liu, G. Zhu, X. Zhang and J. Chen, J. Mater. Chem. B, 2013, 1, 2742.
28. Y. Zhang, R. Zhu, Y. Cui, J. Zhong, X. Zhang and J. Chen, Nanotechnology, 2014, 25, 135607.
29. P. Ling, Q. Hao, J. Lei and H. Ju, J. Mater. Chem. B, 2015, 3, 1335–1341.
30. Y. He, W. Huang, Y. Liang and H. Yu, Sens. Actuators, B, 2015, 220, 927–931.
31. R. Sivasubramanian and M. V. Sangaranarayanan, Sens. Actuators, B, 2015, 213, 92–101.
32. L. Cui, Z. Peng, C. Ji, J. Huang, D. Huang, J. Ma, S. Zhang, X. Qian and Y. Xu, Chem. Commun., 2014, 50, 1485–1487.
33. A. Krittayavathananon, P. Srimuk, S. Luanwuthi and M. Sawangphruk, Anal. Chem., 2014, 86, 12272–12278.
34. C. A. Reilly and S. D. Aust, Chem. Res. Toxicol., 1997, 10, 328–334.
35. S. Garrod, M. E. Bollard, A. W. Nicholls, S. C. Connor, J. Connelly, J. K. Nicholson and E. Holmes, Chem. Res. Toxicol., 2005, 18, 115–122.
36. R. Devasenathipathy, V. Mani, S. M. Chen, D. Arulraj and V. S. Vasantha, Electrochim. Acta, 2014, 135, 260–269.
37. A. J. Jeevagan and S. A. John, RSC Adv., 2013, 3, 2256–2264.
38. W. Sultana, S. Ghosh and B. Eraiah, Electroanalysis, 2012, 24, 1869–1877.
39. S. H. Liu and M. Y. Han, Adv. Funct. Mater., 2005, 15, 961–967.
40. S. Li and F. Huo, Small, 2014, 10, 4371.
41. Z. Li and Y. Zhang, Angew. Chem., Int. Ed., 2006, 45, 7732–7735.
42. G. Lu, S. Li, Z. Guo, O. K. Farha, B. G. Hauser, X. Qi, Y. Wang, X. Wang, S. Han, X. Liu, J. S. DuChene, H. Zhang, Q. Zhang, X. Chen, J. Ma, S. C. J. Loo, W. D. Wei, Y. Yang, J. T. Hupp and F. Huo, Nat. Chem., 2012, 4, 310–316.
43. Z. Xu, L. Yang and C. Xu, Anal. Chem., 2015, 87, 3438–3444.
44. F. Bai, Y. Xia, B. Chen, H. Su and Y. Zhu, Carbon, 2014, 79, 213–226.
45. J. Ma, H. Wang, X. Yang, Y. Chai and R. Yuan, J. Mater. Chem. A, 2015, 3, 12038–12043.
46. C. Mao, A. Kong, Y. Wang, X. Bu and P. Feng, Nanoscale, 2015, 7, 10817–10822.
47. Y. Zhang, X. Bo, A. Nsabimana, A. Munyentwali, C. Han, M. Li and L. Guo, Biosens. Bioelectron., 2015, 66, 191–197.
48. Z. Yang, Y. Xia, X. Sun and R. Mokaya, J. Phys. Chem. B, 2006, 110, 18424–18431.
49. S. Koçak and B. Aslışen, Sens. Actuators, B, 2014, 196, 610–618.
50. L. Shang, F. Zhao and B. Zeng, Electroanalysis, 2012, 24, 2380–2386.
51. H. Hosseini, H. Ahmar, A. Dehghani, A. Bagheri, A. R. Fakhari and M. M. Amini, Electrochim. Acta, 2013, 88, 301–309.
52. R. Madhu, V. Veeramani and S. M. Chen, Sens. Actuators, B, 2014, 204, 382–387.
53. C. Zhang, G. Wang, Y. Ji, M. Liu, Y. Feng, Z. Zhang and B. Fang, Sens. Actuators, B, 2010, 150, 247–253.
54. Y. Han, L. Han, L. Zhang and S. Dong, J. Mater. Chem. A, 2015, 3, 14669–14674.
55. A. R. Fakhari, H. Ahmar, H. Hosseini and S. Kazemi Movahed, Sens. Actuators, B, 2015, 213, 82–91.
56. S. P. Kim and H. C. Choi, Sens. Actuators, B, 2015, 207, 424–429.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra00096g

---