Development of highly-sensitive hydrazine sensor based on facile CoS₂–CNT nanocomposites

Abstract

Cobalt pyrite-decorated carbon nanotube nanocomposites (CoS₂–CNT NCs) were prepared by a simple wet-chemistry method in an alkaline medium. The characterization of the resulting CoS₂–CNT NCs was performed in detail by field-emission scanning electron microscopy (FESEM), energy-dispersive spectroscopy (EDS), X-ray photoelectron spectroscopy (XPS), UV-vis spectroscopy, FT-IR spectroscopy, and X-ray diffraction patterns (XRD). A glassy carbon electrode (GCE) was fabricated from the CoS₂–CNT NCs and developed into a chemical sensor for hydrazine (HZ) using a simple and reliable I–V method. The poisonous chemical HZ was selected as the target analyte in a selectivity study, which demonstrated the fast response of the GCE sensor fabricated from CoS₂–CNT NCs using the I–V method. It also displayed an excellent sensitivity, a very low detection limit, long-term stability, and reproducibility. In a diagnostic study, a linear calibration plot (r² = 0.9992) was obtained for a 0.1 nM to 1.0 mM aqueous solution of HZ, with a sensitivity of 4.430 μA nM⁻¹ m⁻² and low detection limit (LOD) of 0.1 nM. The potential of CoS₂–CNT NCs in terms of chemical sensing is also discussed in this report. This approach is emerging as an effective technique for developing efficient chemical sensors for the detection of environmental pollutants in a large scale.

---

1. Introduction

Hydrazine (HZ) is a toxic chemical and is often considered to be a carcinogenic, poisonous, hazardous, cyanogenetic and nephrotoxic substance.¹ The various uses of HZ include pesticides, plant growth regulators, uses in the dye and photographic industry, the pharmaceutical and polymer industry, and industries related to agriculture, rocket fuel, fuel for spacecraft, explosives, etc.² Symptoms of severe exposure to HZ include burning in the eyes and nose, short-term loss of sight, fainting, vomiting, respiratory edema and unconsciousness. Liver and kidney functions may also be badly affected by long-term exposure to HZ.³ The central nervous system is also affected by HZ, which sometimes leads to unconsciousness. When HZ is absorbed through the skin, it produces alkali-like burning and also interrupts the production of blood.⁴ Hence, a superior analytical technique has become essential for the detection and quantification of HZ. Recently, Wei et al. have proposed a hydrazine sensor based on a GCE coated with sulfur-doped γ-MnOOH microrods⁵ and Ganesan et al. have studied a guar gum-based composite coated with palladium nanoparticles for the electrochemical detection of HZ,⁶ but in both approaches the linear dynamic ranges are very low and there are limitations in detecting nano-level concentrations of HZ.

Electrochemical sensors always offer fast, powerful and cost-effective methods for the detection and quantification of HZ. However, the electrochemical oxidation of HZ at a bare electrode is kinetically slow and is associated with a high overpotential. Consequently, the search for new materials for the modification of electrodes to increase the rate of electron transfer and reduce the overpotential in the oxidation of HZ is necessary.^7–11 Several redox mediators such as metal nanoparticles,^7,8,11–13 metal oxides,¹⁴ hexacyanoferrate salts,⁹ organic mediators,¹⁵ etc., have so far been developed for this purpose. Moreover, several methods for the detection of HZ including spectrophotometric methods,^16,17 chromatographic methods,^18,19 titrimetric methods,²⁰ chemiluminescence,^21,22 coulometry,²³ electrochemistry,²⁴ fluorescence,^25,26 etc., have also been reported, but all these techniques are either expensive or time-consuming or need sophisticated instruments and even sometimes fail to detect nano level concentrations of HZ. During the efficient detection of ultra-trace amounts of environmental toxins, nanomaterials exhibit better properties than the bulk substance, for example, mechanical strength, heat tolerance, electrocatalytic properties, electrical conductance, electromagnetic properties, and photocatalytic properties.²⁷ Moreover, owing to their low cost, rapid response and higher sensitivity, electrochemical sensors are often more useful than any other methods for the detection of HZ.

Lately, nanostructured transition metal sulfides have become important in the research field owing to their excellent electrochemical properties.^28,29 Cobalt sulfides play important roles as electrocatalysts among all other transition metal sulfides. Because of the importance of cobalt sulfides, thus far amorphous cobalt sulfides, cobalt sulfide/graphene nanocomposites, and cobalt sulfide layers on nickel foam, etc., have been prepared. Duan et al. have synthesized hierarchical arrays of CoS nanowire@NiCo₂S₄ nanosheets for high-performance supercapacitors.³⁰ Kim et al. have studied N-doped molybdenum sulfide–CNT nanocomposites for a high-performance hydrogen evolution reaction.³¹ Yuana et al. have investigated FeS₂/CNT nanocomposites for lithium-ion batteries,³² whereas Dai et al. have studied cobalt-doped FeS₂-CNT nanocomposites for the hydrogen evolution reaction.³³ However, these efforts have achieved little success in improving the electrochemical properties of cobalt sulfides. This may be due to the ineffective external interaction between cobalt sulfides and carbonaceous substances. These ineffective interactions between redox active entities and charge transfer materials will allow only a fraction of cobalt sulfide molecules to join in the electron transfer process. Hence, unexpected electrochemical properties of cobalt sulfides were observed and slower charge transfer than that of carbonaceous materials also occurred, etc. To overcome these complications, here we propose a facile wet chemistry method of preparing CoS₂–CNT NCs in the form of interconnected three-dimensional conductive linkages, in which CoS₂ nanostructures are attached in and around a CNT. A dispersed CNT solution plays a role as a precursor in these attachments, in which nanostructured CoS₂ is deposited in and around a CNT. An important benefit of this nanocomposite is that nanostructured cobalt sulphides are firmly attached to a CNT and these two different entities are rigidly joined together. The as-grown CoS₂–CNT NCs nanocomposite exhibits unprecedented electrochemical performance.

Because the improvement of the electrochemical properties of CNTs by incorporating them into nanocomposites is cost-effective with respect to other methods specified in the literature,³⁴ here we have developed a sensitive electrochemical sensor from a GCE fabricated from CoS₂–CNT NCs to detect and quantify HZ in aqueous solution by a simple I–V method. Among all the existing electrochemical detection methods, the two-electrode system I–V method provides an opportunity for the convenient, cost-effective and fast detection of HZ. To the best of our knowledge, a GCE sensor fabricated with CoS₂–CNT NCs has not yet been reported.

2. Experimental section

2.1. Materials and methods

Cobalt(II) chloride hexahydrate (CoCl₂·6H₂O), hydrated sodium sulphide (Na₂S·9H₂O), ethanol, disodium phosphate, monosodium phosphate, CNT, 4-aminophenol, 2-nitrophenol, 4-methoxyphenol, 3-methoxyphenol, chloroform, Nafion (5% ethanolic solution), n-hexane, tetrahydrofuran (THF), ammonium hydroxide, pyridine, acetone, benzaldehyde, and methanol used in this present work were used without any further purification and were purchased from Sigma-Aldrich Company. CoS₂–CNT NCs were investigated using UV/vis spectroscopy (Evolution 300 UV/visible spectrophotometer, Thermo Scientific). The FT-IR spectrum of CoS₂–CNT NCs was recorded with a spectrophotometer (Nicolet iS50 FTIR spectrometer, Thermo Scientific) in the range of 400 to 4000 cm⁻¹. XPS measurements were conducted to estimate the binding energies in eV of C, Co and S using a Mg Kα₁ spectrometer (Thermo Scientific, Kα 1066, USA) with an excitation radiation source (Al Kα, beam spot size: 300.0 μm, pass energy: 200.0 eV, pressure: ∼10⁻⁸ Torr). XRD patterns were recorded with an X-ray diffractometer (XRD, Thermo Scientific, ARL X'TRA) using Cu Kα₁ radiation (λ: 1.5406 nm; generator voltage: 40 kV; applied current: 35 mA). Morphological evaluation of CoS₂–CNT NCs was performed using FESEM (JEOL, JSM-7600F, Japan). Elemental analysis of CoS₂–CNT NCs was carried out by EDS (JEOL, Japan) connected to FESEM. I–V measurements were conducted using a Keithley electrometer at room temperature.

2.2. Synthesis of CoS₂–CNT NCs

A simple wet chemistry method was employed to synthesize CoS₂–CNT NCs in an alkaline medium using cobalt(II) chloride hexahydrate (CoCl₂·6H₂O), sodium sulphide (Na₂S·9H₂O) and CNT. In the synthesis process, 50.0 mL CoCl₂ solution and 50.0 mL Na₂S solution (both 0.1 M) were prepared using deionized (DI) water separately. Then 1.0 mg CNT was added to the 50.0 mL cobalt(II) chloride solution in a conical flask and heated to 85.0 °C for 30 min with continuous stirring. Then 50.0 mL of the Na₂S solution was added to the mixture dropwise with constant stirring. After 6 h of continuous stirring at 85.0 °C, the reaction mixture was cooled to room temperature (25.0 °C) and thus a black precipitate of CoS₂–CNT NCs was produced. The as-grown CoS₂–CNT NCs were cleaned with DI water and alcohol to remove any impurities that were present and then allowed to dry at room temperature. Later, the black powder sample that was collected was dried at 65.0 °C using an oven for 6 h and the as-grown CoS₂–CNT NCs were finally obtained.

The reaction scheme of the formation of CoS₂–CNT NCs depends firstly on the slow release of Co²⁺ and S²⁻ ions within solutions and secondly on the precipitation of these ions in the form of CoS₂ in and around a CNT. The formation of CoS₂ depends on the fact that the ionic product of Co²⁺ and S²⁻ ions is greater than the solubility product of CoS₂. Co²⁺ and S²⁻ ions are provided by hydrolysis reactions of CoCl₂ and Na₂S, respectively. The proposed mechanism may be described as follows:

CoCl₂ becomes ionized in water (eqn (i)) and produces Co²⁺ ions, which form complex ions and become dispersed in and around a CNT. Na₂S also becomes ionized (eqn (iii)), and S²⁻ ions will also diffuse into the solution.

CoCl₂ → Co²⁺_(aq) + 2Cl⁻_(aq) (i)

Co²⁺_(aq) + 4H₂O_(I) → [Co(H₂O)₄]²⁺_(aq) (ii)

Na₂S → 2Na⁺_(aq) + S²⁻_(aq) (iii)

Effective collisions between Co²⁺ and S²⁻ ions cause nucleation, followed by aggregation and finally the formation of CoS₂–CNT NCs in the presence of dispersed CNTs (eqn (iv) and (v)).

CNT_(Disp) + [Co(H₂O)₄]²⁺_(aq) + S²⁻_(aq) → CoS–CNT(s)↓ + 4H₂O_(l) (iv)

CoS–CNT(s) + S_x²⁻_(aq) → CoS₂–CNT(s)↓ + S²⁻_x−1 (v)

The overall reaction can be written as follows (eqn (vi)):

CoCl_2(aq) + 2Na₂S_(aq) + CNT_(disp) → CoS₂–CNT_(s)↓ + 4Na⁺_(aq) + 2Cl⁻_(aq) (vi)

Finally, the as-grown CoS₂–CNT NCs were cleaned with DI water and alcohol to remove any impurities that were present and dried at 65.0 °C using an oven. The as-obtained CoS₂–CNT NCs were characterized in detail for their crystallinity, morphology, structure, electrochemical properties, etc., and later used as a material for the fabrication of a GCE for use as a HZ sensor using a simple I–V technique. In the growth mechanism of CoS₂–CNT NCs, in the beginning the growth of the CoS₂ nucleus occurs by mutual aggregation of Co²⁺ and S²⁻. As in the Ostwald ripening method, nanocrystals later aggregated to form combined CoS₂ nanocrystals. In the presence of dispersed CNTs, CoS₂ nanocrystals recrystallized and reaggregated with one another and were deposited on and around CNTs via van der Waals forces, which resulted in the porous morphology of CoS₂–CNT NCs, which is shown in Scheme 1.

Scheme 1 Probable growth mechanism of CoS₂–CNT nanocomposites prepared by a facile wet chemistry method.

2.3. Fabrication of GCE using CoS₂–CNT NCs

The fabrication of a GCE was carried out from the as-prepared CoS₂–CNT NCs using a 5% ethanolic solution of Nafion as a conducting binder to obtain a film with a thickness of 0.5 mm. This was then heated in an oven at 65.0 °C for 2 h to produce a dry film for the GCE. In the electrochemical cell, the GCE coated with CoS₂–CNT NCs was used as the working electrode (WE), platinum wire was used as the counter electrode (CE), and an aqueous solution of HZ in 0.1 M phosphate buffer solution (PBS; pH 7.0) was used as the electrolyte. For use as a target analyte, the aqueous solution of HZ (0.1 M) was diluted to different concentrations (0.1 M to 0.1 nM) using DI water. All the I–V measurements were carried out in 5.0 mL PBS (pH = 7.0). From the slope of a calibration plot, the sensitivity of the proposed HZ sensor was estimated by considering the active surface area of the GCE. By using an electrometer (Keithley 6517A electrometer, USA), a simple I–V method was applied to solutions of HZ taking CoS₂–CNT NCs/GCE as the WE.

3. Results and discussion

3.1. Determination of binding energy

XPS was used for further investigation of the purity and nanostructure of the CoS₂–CNT NCs. From the full-scan spectrum (Fig. 1(a)), it can be deduced that the surface of the CoS₂–CNT NCs consists of cobalt, sulfur, carbon, and oxygen atoms. Oxygen present in the CoS₂–CNT NCs may be attached to carbon by a carbon-oxygen single or double covalent bond or in the form of C=O.³⁵ The asymmetric peak (Fig. 1(b)) of C1s centered at 284.3 eV and the broad peak (Fig. 1(c)) of O1s at 531.8 eV clearly demonstrate the presence of different bonds such as C=O and C–O and the adsorption of water. The Co2p spectrum consists of two well-resolved peaks (Fig. 1(d)) located at 781.6 and 797.2 eV, which correspond to Co2p_3/2 and Co2p_1/2, respectively, and are also in good agreement with the binding energies of Co2p in CoS₂. As shown in Fig. 1(e), the two electron signals at 164.0 and 169.5 eV can be attributed to S2p_3/2 and S2p_1/2, respectively, whereas Fig. 1(f) shows that the electron signal at 199.6 eV can be attributed to S2s. The main peaks in the XPS spectra (Fig. 1(a–f)) are consistent with those of CoS₂ reported previously.^36–40

Fig. 1 Determination of binding energy. XPS spectra of (a) as-prepared CoS₂–CNT NCs, (b) C1s, (c) O1s, (d) Co2p_, (e) S2p, and (f) S2s orbitals acquired with Mg Kα₁ radiation.

3.2. Morphological and elemental evaluation

The morphology and structure of the CoS₂–CNT NCs were investigated by FESEM. Typical morphological information for the as-grown CoS₂–CNT NCs is presented in Fig. 2(a) and (b). The as-prepared CoS₂–CNT NCs contain interconnected networks of carbon nanotubes in which CoS₂ nanoparticles are adsorbed onto CNTs (white spots). This exceptional structure with adsorbed pyrite CoS₂ nanoparticles provides a large surface area and increases the extent of electron transport.

Fig. 2 Morphological and elemental evaluation. (a and b) Low- and high-resolution FESEM images and (c and d) EDS spectrum of the as-grown CoS₂–CNT NCs.

The chemical composition of the as-prepared CoS₂–CNT NCs was determined by EDS analysis (Fig. 2(c) and (d)), which confirmed the presence of C, S and Co. The contents (wt.%) of carbon, sulphur, and cobalt in the as-grown CoS₂–CNT NCs were 85.56%, 2.79%, and 11.65%, respectively.

3.3. Evaluation of optical and structural properties

The UV/vis spectrum (300–600 nm) of the as-prepared CoS₂–CNT NCs was studied to estimate the band gap energy (E_bg). The broad peak at 501 nm was attributed to the characteristic peak of CoS₂–CNT NCs, as shown in Fig. 3(a). By entering UV/vis spectral data into Tauc's equation (vii), the value of E_bg of the CoS₂–CNT NCs was calculated.^41,42 Tauc's equation (eqn (vii)) can be represented as:

αhν = A(hν − E_bg)ⁿ (vii)

where n = ¹/₂ or 2 for a direct or indirect electronic transition, respectively. From a plot of (αhν)² vs. (hν) (Fig. 3(b)), the value of E_bg of the as-grown CoS₂–CNT NCs was estimated to be ∼2.07 eV, which is consistent with that of CoS₂–CNT NCs.

Fig. 3 Optical and structural evaluation. (a) UV/vis spectrum, (b) Tauc's plot for the band gap energy, (c) FTIR spectrum, and (d) XRD pattern of the as-grown CoS₂–CNT NCs.

The FTIR spectrum of the as-grown CoS₂–CNT NCs is shown in Fig. 3(c). The broad peaks at 3548 and 1474 cm⁻¹ are due to stretching and bending modes of vibrations of H₂O, respectively.⁴³ Stretching vibrations of Co–S bonds appear at 550 cm⁻¹. The peaks at 1207 and 980 cm⁻¹ are assigned to vibrations of C–O single bonds. Vibrational peaks of C=S and C–S are observed at 892 cm⁻¹ and 652 cm⁻¹, respectively.⁴⁴

The XRD pattern, as shown in Fig. 3(d), matched JCPDS card no. 41-1471 for the cubic pyrite phase of CoS₂. The XRD peaks at 2θ = 27.9, 32.1, 36.3, 39.8, 46.2, 54.9, 60.1 and 62.7° can be assigned to the specific (111), (200), (210), (211), (220), (311), (230), and (321) planes, respectively, which is obviously consistent with the cubic pyrite structure of CoS₂. Several additional peaks appeared, which were possibly due to the presence of CNT and noise. The greater intensity of the peak at 36.3° implies that the (210) plane of the cubic pyrite structure was inclined towards the experimental system. In general, the entire XRD pattern is well matched with the Bragg reflections of the standard cubic pyrite phase structure of CoS₂, which are consistent with the values given in the standard card (JCPDS no. 41-1471).^36,45–48

3.4. Applications

3.4.1. Detection of HZ using CoS₂–CNT NCs by I–V method. Toxic HZ in aqueous solution was detected and measured using a GCE fabricated from CoS₂–CNT NCs as a chemical sensor. Their nontoxic nature, chemical stability and electrochemical activity make the CoS₂–CNT NCs one of the best sensing materials for HZ. Upon contact with CoS₂–CNT NCs, HZ gives a remarkable response in a simple I–V method. Fig. 4(a) shows the fabricated surface of a CoS₂–CNT NCs sensor prepared in a 5% ethanolic solution of Nafion. A possible scheme of oxidation at the CoS₂–CNT NCs/GCE is generalized in Fig. 4(b), in which HZ is oxidized to N₃⁻, which releases free electrons on the sensor surface during I–V measurements. In the presence of HZ, electrons are also released from adsorbed reduced oxygen species on the surface of the CoS₂–CNT NCs, which further increases the current intensity with an increase in voltage at room temperature.^49–51 A practical I–V response both with and without HZ at the CoS₂–CNT NCs/GCE working electrode is illustrated in Fig. 4(c) at a delay time of 1.0 second in the electrometer, where a higher current response to an increase in voltage is clearly demonstrated.

Fig. 4 Scheme representing (a) rod-shaped round GCE electrode coated with CoS₂–CNT NCs with conducting coating binder of Nafion (5% in ethanol), (b) proposed detection mechanism of HZ, in which HZ is oxidized to N₃⁻ by releasing electrons onto the CoS₂–CNT NCs/GCE, and (c) I–V response observed using the CoS₂–CNT NCs/GCE. Surface area of GCE: 0.0316 cm²; method: I–V; delay time: 1.0 s.

The current intensities in PBS (pH = 7.0) without HZ for a bare GCE (blue dots) and the GCE fabricated from CoS₂–CNT NCs (red dots) are given in Fig. 5(a). The current intensity for the CoS₂–CNT NCs/GCE is much higher than that for the bare GCE, which demonstrates the excellent electrochemical properties of the CoS₂–CNT NCs. Fig. 5(b) shows the current responses to several toxic chemicals (thirteen analytes), in which a solution of HZ (red dots) gave the best response with the GCE surface coated with CoS₂–CNT NCs. In this selectivity study, we added 25.0 μL of a 1.0 μM solution of each analyte in 5.0 mL PBS (pH = 7.0). Fig. 5(c) shows the I–V responses of the GCE coated with CoS₂–CNT NCs without HZ (blue dots) and with HZ (red dots; 1.0 μM; 25.0 μL) in 5.0 mL PBS solution. In the presence of HZ in PBS, the noticeable increase in the current response implies the ability to sense HZ of the developed CoS₂–CNT NCs/GCE sensor. The I–V responses to HZ (1.0 μM; 25.0 μL) in 5.0 mL PBS solution using the GCE coated with CoS₂ nanostructures (blue dots) and GCE coated with CoS₂–CNT NCs (red dots) as working electrodes are given in Fig. 5(d). In the presence of HZ, the recognizable increase in the current response from the CoS₂–CNT NCs/GCE compared with the CoS₂/GCE implies the ability to sense HZ of the developed CoS₂–CNT NCs/GCE sensor. This large increase in the current may be because the surface area of the CoS₂–CNT NCs is larger compared with that of nanostructured CoS₂.

Fig. 5 I–V responses: (a) bare and CoS₂–CNT-coated GCE, (b) selectivity study with various interferences (thirteen analytes), (c) in the absence and presence of HZ (1.0 μM; 25.0 μL), and (d) control experiments conducted with CoS₂/GCE and CoS₂–CNT NCs/GCE separately.

HZ solutions (25.0 μL) of low (0.1 nM) to high (0.1 M) concentrations were injected sequentially into 5.0 mL PBS and variations in surface current were investigated after every injection of HZ into PBS by I–V method. The current responses of the GCE surface coated with as-prepared CoS₂–CNT NCs were estimated using aqueous HZ solutions of different concentrations (0.1 nM to 0.1 M) and are shown in Fig. 6(a), which clearly demonstrates that with an increase in potential, the current response of the GCE sensor fabricated from CoS₂–CNT NCs increased with an increase in the concentration of HZ at room temperature (25.0 °C). It was also observed that from a dilute (0.1 nM) to a concentrated (0.1 M) solution of HZ the current response increased gradually. Aqueous solutions of HZ (0.1 nM to 0.1 M) were selected to determine the detection limit of the proposed sensor. A calibration plot (at +0.6 V) of current vs. concentration for the entire concentration range is shown in Fig. 6(b). A very high sensitivity value (4.430 μA nM⁻¹ m⁻²) was calculated from the calibration plot. The LDR of the proposed sensor, which was obtained from the calibration plot, is 0.1 nM to 1.0 mM (r² = 0.9992), and the LOD obtained from Fig. 6(e) is 0.1 nM. Fig. 6(c) shows the repeatability of the I–V responses of the GCE coated with as-grown CoS₂–CNT NCs to HZ solutions (25.0 μL; 0.1 μM), where seven different working electrodes were used in runs R1–R7 under identical conditions. Almost the same current responses were observed in seven repeated experiments, which confirms the excellent repeatability of the sensor (RSD = 3.52%, n = 7). This small percentage RSD may be due to the variation in the mass of coating material (CoS₂–CNT NCs) in the seven working electrodes. Fig. 6(d) shows the repeatability using the same working electrode in runs R1–R7. Very similar current responses were observed in seven repeated experiments, which again demonstrates the excellent repeatability of the sensor (RSD = 3.35%; n = 7). When a single working electrode was used in different solutions, even under identical conditions, the current response decreased slightly. This may be due to a decrease in the number of active sites in the CoS₂–CNT NCs after each run.

Fig. 6 (a) Variations in current for different concentrations (0.1 nM to 0.1 M) of aqueous HZ solution in the voltage range of 0.0 to +1.5 V, (b) calibration plot for GCE surfaces fabricated from CoS₂–CNT NCs at +0.6 V (inset: a magnified form of LDR on current responses against the logarithm of analyte concentrations), (c) repeatability using different working electrodes, and (d) repeatability using the same working electrode. Potential range: 0 to +1.5 V; analyte concentration: 0.1 μM; analyte volume: 25.0 μL; 0.1 M PBS at pH 7. (e) 0.1 nM HZ vs. blank sample in 5.0 mL PBS (pH 7) using the CoS₂–CNT NCs/GCE.

3.4.2. Real sample analysis using CoS₂–CNT NCs/GCE. To confirm the validity of the developed sensor, the CoS₂–CNT NCs/GCE was used as the WE to quantify HZ in industrial effluent water (collected from an industrial effluent treatment plant, Jeddah, Saudi Arabia) and Red Sea water (from the Red Sea coastline, Jeddah, Saudi Arabia). For this purpose, we used the standard addition method to investigate the precision of the detection of HZ in aqueous samples. A fixed amount (∼25.0 μL) of aqueous samples of different concentrations, along with the same amount of real samples, were mixed and analyzed in PBS (5.0 mL) using CoS₂–CNT NCs/GCE working electrodes (Table 1).

Table 1 Real sample analyses with CoS₂–CNT NCs/GCE sensor at room conditions^a

Real sample	HZ concentration added^b	HZ concentration^b determined^c by the CoS2–CNT NCs/GCE	Recovery^c (%)	RSD^d (%) (n = 3)
a S1 and S2: real water samples collected from an industrial effluent treatment plant and the Red Sea coastline, respectively, Jeddah, Saudi Arabia.b Mean of three repeated determinations (S/N = 3) with the CoS2–CNT NCs/GCE.c Concentration of HZ determined/concentration of HZ used.d The relative standard deviation indicates the precision among three repeated determinations.
S1	2.000 nM	2.106 nM	105.3	3.4
S1	2.000 μM	2.052 μM	102.6	2.7
S1	2.000 mM	2.022 mM	101.1	1.8
S2	2.000 nM	2.016 nM	100.8	2.6
S2	2.000 μM	2.004 μM	100.2	2.7
S2	2.000 mM	1.994 mM	99.7	1.9

---

Table 1 shows the results, which demonstrate that the CoS₂–CNT NCs/GCE modified sensor provided quantitative (∼100%) recovery of HZ. Based on the results, therefore, it is concluded that the I–V method is suitable, consistent, and appropriate for real sample analysis with the CoS₂–CNT NCs/GCE system.

The interior resistance of the fabricated nanocomposite sensor decreases with increasing electron transfer characteristics, which is an important feature of nanomaterials at RTP.^33,52–55 The adsorption of oxygen (Fig. 1(a)) regulates a significant feature of the exceptional electrochemical properties of the CoS₂–CNT NCs/GCE. The adsorption of O₂⁻ and/or O⁻ ions decreases the number of electrons in the conduction band and hence increases the resistance of the CoS₂–CNT NCs. Dissolved O₂ from an aqueous solution or surrounding atmospheric air is adsorbed on the surface of the CoS₂–CNT NCs and ionized according to eqn (viii)–(x).

O_2(dissolved) → O_2(adsorbed) (viii)

O_2(adsorbed) + e⁻(NCs) → O₂⁻_(adsorbed) (ix)

O₂⁻_(adsorbed) + e⁻(NCs) → 2O⁻_(adsorbed) (x)

The oxidation of HZ to N₂ or N₃⁻ at the surface of the CoS₂–CNT NCs is the main phenomenon involved in this proposed HZ sensor. At room temperature, reactive oxygen species (O₂⁻ and O⁻) are chemisorbed on the mesoporous CoS₂–CNT NCs, where the extent of such adsorption is mainly dependent on the surface area of the CoS₂–CNT NCs. Mesoporous structures provide a very large surface area in the CoS₂–CNT NCs, which is ultimately responsible for such extensive adsorption. The extremely high sensitivity of the CoS₂–CNT NCs/GCE to HZ could be attributed to the oxidation of HZ by adsorbed oxygen species (O₂⁻ and O⁻). The greater the extent of adsorption of oxygen species on the CoS₂–CNT NCs, the higher will be the oxidizing power, which will oxidize HZ quickly, resulting in a very short response time (10 s) of the sensor. The rate of oxidation of HZ by the CoS₂–CNT NCs was higher than that of other chemicals, even under similar conditions, as shown in Fig. 5(b). When HZ is oxidized by adsorbed O₂⁻ or O⁻ on the surface of the CoS₂–CNT NCs, it is converted into N₂ or N₃⁻ and leaves electrons in the conduction band,⁵⁶ as given in the following eqn (xi)–(xiv).

N₂H₄ + O₂⁻ → N₂ + 2H₂O + e⁻ (xi)

N₂H₄ + 2O⁻ → N₂ + 2H₂O + 2e⁻ (xii)

4N₂H₄ + 5O⁻ → 2N₃⁻ + 2NH₃ + 5H₂O + 3e⁻ (xiii)

8N₂H₄ + 5O₂⁻ → 4N₃⁻ + 4NH₃ + 10H₂O + e⁻ (xiv)

The current response in the I–V method during the detection of HZ largely depends on the dimensions, morphology, and nanoporosity of the nanocomposites. When the surface of the CoS₂–CNT NCs is exposed to reducing HZ, a surface-mediated oxidation reaction takes place. The removal of adsorbed oxygen species (O₂⁻ and O⁻) increases the number of electrons in the conduction band and hence the surface conductance of the electrode increases. The oxidation of HZ also supplies electrons to the surface of the CoS₂–CNT NCs, which further increases the conductance of the WE.^57–59 Therefore, the current response increases with an increase in potential. These supplies of electrons increase the conductance of the CoS₂–CNT NCs coating quickly. The extensive deposition of CoS₂ NPs onto CNTs (white spots) with porous morphology increases the adsorption ability of the CoS₂–CNT NCs. The CoS₂–CNT NCs/GCE sensor requires approximately 10 s to achieve a constant current. This excellent sensitivity and high electrochemical performance of the CoS₂–CNT NCs are due to their mesoporous surface, which enhances the adsorption and absorption of oxygen species. The CoS₂–CNT NCs/GCE sensor is highly sensitive to HZ and has a lower detection limit than that of other sensors already reported for the detection of HZ,^{3,9,10,60–65} as stated in Table 2. As they have a large surface area, the CoS₂–CNT NCs provided a suitable nano environment during the detection and quantification of HZ. The proposed CoS₂–CNT NCs/GCE sensor has also displayed better reliability and stability. Despite these developments, there are still numerous important concerns that must be investigated further before the commercial production of this sensor.

Table 2 Comparison of analytical performance in the detection of HZ using various nanocomposites or nanomaterials with conventional electrodes

Electrode fabrication	Technique/method	Linear dynamic range (LDR)	Limit of detection (LOD)	Sensitivity	pH	Ref.
CNT/GCE catechol derivatives	CV	0.50 μM–1.0 mM	50 nM	1.834 nA M^−1 m^−2	7.0	3
Mn-hexacyanoferrate/graphite–wax composites	CV	3.33 × 10^−5–8.18 × 10^−3 M	6.65 μM	0.4753 μA mM^−1 m^−2	7.0	9
Pd NPs-polyaniline	CV	10–300 μM	0.06 μM	0.5 μA μM^−1 m^−2	7.0	10
Ni-hexacyanoferrate/graphite	Amperometry	2.4 × 10^−6–8.2 × 10^−3 M	1.0 μM	0.0113 μA μM^−1 m^−2	7.0	60
Terpyridine diacetonitrile triphenylphosphine Ru(II) tetrafluoroborate complex/f-MWCNT	CV	5 × 10^−6–6.5 × 10^−3 M	0.37 μM	0.01031 μA μM^−1 m^−2	7.4	61
Ruthenium/GCE	CV	10^−5–10^−2 M	8.5 μM	—	5.0	62
Monodispersed PEG-ZnS NPs		1 μM–3mM	1.073 μM	8.93 μA cm^−2 M^−1 m^−2	7.0	63
Au NPs/graphite pencil	Amperometry	—	3.07 μM	—	5.0	64
Au NPs/colorimetric probe	Spectrometry	6.0–40.0 × 10^−6 M	1.1 μM	—	7.0	65
CoS2/CNT NCs/Nafion/GCE	I–V	0.1 nM–1.0 mM	0.1 nM	4.430 μA nM^−1 m^−2	7.0	This work

---

4. Conclusions

A CoS₂–CNT nanocomposite has been prepared at a low temperature using a facile wet chemistry technique in an alkaline medium. A sensor for toxic chemicals was developed based on a flat GCE electrode fabricated from CoS₂–CNT NCs with conducting coating binders, which displayed higher sensitivity and selectivity in sensing applications for HZ among various interfering chemicals. This CoS₂–CNT NCs/GCE electrode was effectively employed as a chemical sensor for the detection and quantification of HZ. The proposed CoS₂–CNT NCs/GCE sensor for HZ exhibited higher sensitivity (4.430 μA nM⁻¹ m⁻²) and a very low detection limit (LOD = 0.1 nM) with an excellent linear response (r² = 0.9992) for a wide range of concentrations in a short response time (10 s). This method is emerging as an effective technique for developing efficient chemical sensors for the detection of environmental pollutants on a large scale.

Glossary

CoS2-CNT NCs	Cobalt pyrite-decorated carbon nanotube nanocomposites
GCE	Glassy carbon electrode
CNT	Carbon nanotube
NCs	Nanocomposites
NPs	Nanoparticles
HZ	Hydrazine
LDR	Linear dynamic range
LOD	Limit of detection
WE	Working electrode
CE	Counter electrode
DI	Deionized water
PBS	Phosphate buffer solution
Ebg	Band-gap energy
RSD	Relative standard deviation

Acknowledgements

Center of Excellence for Advanced Materials Research (CEAMR), King Abdulaziz University, Jeddah, Saudi Arabia is highly acknowledged for their lab facilities and instrumental supports. M. A. H thanks MOE, Bangladesh for providing supports (2015-16).

References

1. M. B. Gholivand and A. Azadbakht, Electrochim. Acta, 2011, 56, 10044.
2. I. G. Casella, M. R. Guascito, A. M. Salvi and E. Desimoni, Anal. Chim. Acta, 1997, 354, 333.
3. A. Salimi, L. Miranzadeh and R. Hallaj, Talanta, 2008, 75, 147.
4. U. P. Azad and V. Ganesan, Electrochim. Acta, 2011, 56, 5766.
5. J. Ding, T. Liu, W. Xu, H. Liao, J. Li, G. Wei and Z. Su, RSC Adv., 2015, 5, 80719.
6. P. K. Rastogi, V. Ganesan and S. Krishnamoorthi, Electrochim. Acta, 2014, 125, 593.
7. F. Li, B. Zhang, S. Dong and E. Wang, Electrochim. Acta, 1997, 42, 2563.
8. J. Li and X. Lin, Sens. Actuators, B, 2007, 126, 527.
9. D. Jayasri and S. S. Narayanan, J. Hazard. Mater., 2007, 144, 348.
10. S. Ivanov, U. Lange, V. Tsakova and V. M. Mirsky, Sens. Actuators, B, 2010, 150, 271.
11. H. Zhang, J. Huang, H. Hou and T. You, Electroanalysis, 2009, 21, 1869.
12. C. B. McAuley, C. E. Banks, A. O. Simm, T. G. J. Jones and R. G. Compton, Analyst, 2006, 131, 106.
13. B. Haghighi, H. Hamidi and S. Bozorgzadeh, Anal. Bioanal. Chem., 2010, 398, 1411.
14. B. Sljukic, C. E. Banks, A. Crossley and R. G. Compton, Electroanalysis, 2006, 18, 1757.
15. S. M. Golabi, H. R. Zare and M. Hamzehloo, Microchem. J., 2001, 69, 111.
16. A. Safavi and A. A. Ensafi, Anal. Chim. Acta, 1995, 300, 307.
17. S. Ganesh, F. Khan, M. K. Ahmed, P. Velavendan, N. K. Pandey and U. Kamachi Mudali, J. Anal. Sci., Methods Instrum., 2012, 2, 98.
18. A. D. Smolenkov, Rev. J. Chem., 2012, 2(4), 329.
19. G. Elias and W. F. Bauer, J. Sep. Sci., 2006, 29, 460.
20. K. R. Rao, R. Rambabu and K. R. Babu, Int. j. innov. res. sci. eng. technol., 2013, 2(11), 6378.
21. A. Safavi and M. R. Baezzat, Anal. Chim. Acta, 1998, 358, 121.
22. J. Lv, Y. Huang and Z. Zhang, Anal. Lett., 2001, 34(8), 1323.
23. E. C. Olson, Anal. Chem., 1960, 32, 1545.
24. L. Lauko, R. Hudec, K. Lenghartova, A. Manova, F. Cacho and E. Beinrohr, Pol. J. Environ. Stud., 2015, 24(4), 1659.
25. G. E. Collins and S. L. R. Pehrsson, Analyst, August 1994, 119, 1907.
26. M. Sun, J. Guo, Q. Yang, N. Xiao and Y. Li, J. Mater. Chem. B, 2014, 2, 1846.
27. M. M. Rahman, H. B. Balkhoyora and A. M. Asiri, RSC Adv., 2016, 6, 29342.
28. L. Shen, L. Yu, H. B. Wu, X. Y. Yu, X. Zhang and X. W. D. Lou, Nat. Commun., 2015, 6, 6694.
29. J. C. Xing, Y. L. Zhu, Q. W. Zhou, X. D. Zheng and Q. J. Jiao, Electrochim. Acta, 2014, 136, 550.
30. W. Zeng, G. Zhang, X. Wu, K. Zhang, H. Zhang, S. Hou, C. Li, T. Wang and H. Duan, J. Mater. Chem. A, 2015, 3, 24033.
31. D. J. Li, U. N. Maiti, J. Lim, D. S. Choi, W. J. Lee, Y. Oh, G. Y. Lee and S. O. Kim, Nano Lett., 2014, 14, 1228.
32. L. Liua, Z. Yuana, C. Qiua and J. Liub, Solid State Ionics, 2013, 241, 25.
33. D. Y. Wang, M. Gong, H. L. Chou, C. J. Pan, H. A. Chen, Y. Wu, M. C. Lin, M. Guan, J. Yang, C. W. Chen, Y. L. Wang, B. J. Hwang, C. C. Chen and H. Dai, J. Am. Chem. Soc., 2015, 137, 1587.
34. C. Karuppiaha, S. Palanisamya, S. M. Chena, S. K. Ramarajb and P. Periakaruppan, Electrochim. Acta, 2014, 139, 157.
35. E. Bi, H. Chen, X. Yang, W. Peng, M. Gratzel and L. Han, Energy Environ. Sci., 2014, 7, 2637.
36. Z. Shadike, M. H. Cao, F. Ding, L. Sang and Z. W. Fu, Chem. Commun., 2015, 51, 10486.
37. P. K. Chu and L. Li, Chem. Phys., 2006, 96, 253.
38. Z. Li, W. Li, H. Xue, W. Kang, X. Yang, M. Sun, Y. Tang and C. S. Lee, RSC Adv., 2014, 4, 37180.
39. Y. Wang, J. Tang, B. Kong, D. Jia, Y. Wang, T. An, L. Zhang and G. Zheng, RSC Adv., 2015, 5, 6886.
40. H. Chen, J. Jiang, Y. Zhao, L. Zhang, D. Guo and D. Xia, J. Mater. Chem. A, 2015, 3, 428.
41. M. M. Rahman, A. Jamal, S. B. Khan and M. Faisal, Biosens. Bioelectron., 2011, 28, 127.
42. Y. Bai, J. Yeom, M. Yang, S. H. Cha, K. Sun and N. A. Kotov, J. Phys. Chem. C, 2013, 117, 2567.
43. R. Akbarzadeh, H. Dehghani and F. Behnoudnia, Dalton Trans., 2014, 43, 16745.
44. S. Pan, J. Zhu and X. Liu, New J. Chem., 2013, 37, 654.
45. C. Zhao, D. Li and Y. Feng, J. Mater. Chem. A, 2013, 1, 5741.
46. L. Zhang, H. B. Wu and X. W. Lou, Chem. Commun., 2012, 48, 6912.
47. S. Amaresh, K. Karthikeyan, I. C. Jang and Y. S. Lee, J. Mater. Chem. A, 2014, 2, 11099.
48. F. Fu, Y. Chen, P. Li, J. He, Z. Wang, W. Lin and W. Zhang, RSC Adv., 2015, 5, 4328.
49. M. Zhang, B. Chen, H. Tang, G. Tang, C. Li, L. Chen, H. Zhang and Q. Zhang, RSC Adv., 2015, 5, 1417.
50. M. M. Rahman, S. B. Khan, M. Faisal, A. M. Asiri and M. A. Tariq, Electrochim. Acta, 2012, 75, 164.
51. X. Chen, Z. Wang, X. Wang, J. Wan, J. Liu and Y. Qian, Inorg. Chem., 2005, 44, 951.
52. M. M. Rahman and A. M. Asiri, RSC Adv., 2015, 5, 63252.
53. M. M. Rahman, A. Jamal, S. B. Khan, M. Faisal and A. M. Asiri, Microchim. Acta, 2012, 178, 99.
54. D. A. Rice, S. J. Hibble, M. J. Almond, K. A. H. Mohammad and S. P. Pearse, J. Mater. Chem., 1992, 2, 895.
55. M. M. Rahman, S. B. Khan, A. M. Asiri and A. G. Al-Sehemi, Electrochim. Acta, 2013, 112, 422.
56. L. M. F. Dantas, A. P. dos Reis, S. M. C. N. Tanaka, J. H. Zagal, Y. Y. Chen and A. A. Tanaka, J. Braz. Chem. Soc., 2008, 19, 720.
57. J. Wu, T. Zhou, Q. Wang and A. Umar, Sens. Actuators, B, 2016, 224, 878.
58. S. Nazir and U. Schwingenschlögl, RSC Adv., 2013, 3, 4518.
59. M. M. Rahman, A. Jamal, S. B. Khan and M. Faisal, ACS Appl. Mater. Interfaces, 2011, 3, 1346.
60. S. J. R. Prabakar and S. S. Narayanan, J. Electroanal. Chem., 2008, 617, 111.
61. I. Tiwari, M. Gupta, P. Sinha and C. E. Banks, Mater. Res. Bull., 2014, 60, 166.
62. J. S. Pinter, K. L. Brown, P. A. DeYoung and G. F. Peaslee, Talanta, 2007, 71, 1219.
63. S. K. Mehta, Khushboo and A. Umar, Talanta, 2011, 85, 2411.
64. M. A. Aziz and A. N. Kawde, Talanta, 2013, 115, 214.
65. B. Zargar and A. Hatamie, Sens. Actuators, B, 2013, 182, 706.

---