ZnO-CNT/Nano-Au modified electrodes for the detection of trace Hg(II) in coastal seawater

Mercury (Hg) in seawater enters the body through the food chain, causing damage to organs and the nervous system. Thus, there is an urgent need to explore a rapid and convenient sensor for the detection and monitoring of Hg(II) in seawater. Herein, a ZnO-CNTs/Nano-Au modified glassy carbon electrode was prepared by the dropping method. The structure of the composite membrane is mainly observed by scanning electron microscopy (SEM), and the results show that the composite has a larger specific surface area. Moreover, the composite can increase the ion adsorption of the surface electrode and enhance the conductivity. Differential pulse voltammetry (DPV) was applied to determine trace amounts of Hg(II) in seawater. The optimized conditions were as follows: accumulation potential, accumulation time, pH value, film thickness and concentration. Under the optimal experimental conditions, the linear relationship between the values of the oxidation peak current and concentration was kept in the range of 1.49 ∼ 5.97 μM, with a linear correlation coefficient R2 = 0.991 and a detection limit of Hg(II) of 0.0118 μM. The proposed method was applied to the analysis of coastal water of the Maowei Sea, giving values of recovery in the range of 94.2%∼98.4%. The ZnO-CNTs/Nano-Au-modified electrode has high sensitivity, convenient operation and good practical application value.


Introduction
Hg(II) is a natural pollutant that exists in air, soil and water [1]. Owing to land-sea interactions, coastal seawater is easily polluted by Hg(II) [2]. Its pollution mainly includes some industrial and chemical activities, the treatment of waste electronic batteries and domestic sewage [3]. As one of the most toxic heavy metal ions, Hg(II) is more likely to cause fatal harm to organisms [4], even at the micro level, because it prefers to combine with the sulfhydryl group of protein, leading to cytotoxicity [5] and endangering the central nervous system, digestive system and kidney [6]. The World Health Organization (WHO) stipulates that the maximum Mercury content is 30 nM [7]. Therefore, the detection of trace Hg(II) in coastal areas is of great practical significance.
Electrochemical stripping voltammetry has attracted extensive attention because of its advantages of simple operation, high sensitivity and rapidity [8,9]. Among them, differential pulse voltammetry (DPV) has become a main detection and analytical procedure because of its fast response speed and low background current [10,11]. At present, some electrochemical biosensors based on DNA and enzymes have been used to detect Hg(II) [12].
In a previous study, Some scholars have described the effect of photophysics on the fluorescence of metal nanomaterials, and provide an updated guide on fluorescent metallic nano-systems used as optical sensors of heavy metal ions and pesticides in water [13]. Rapid Mercury determination with graphene-based sensors by Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. some authors, research directions of Mercury sensors are discussed, which will guide future research of Hg(II) sensors in biomedical and environmental monitoring applications [14]. The electronic and molecular structure, as well as the morphology of innovative nanostructured materials whose optical properties respond to the presence of heavy metals in water samples was investigated. The sensitivity of the proposed nanosensor to Hg(II) in 1−5 ppm range was ascertained by optical tests [15]. However, strict operating conditions limit its application in seawater environments. In addition, electrochemical analysis methods are usually based on redox reactions, and substances with similar redox properties will affect electrochemical detection [16]. It is also a major challenge to eliminate such interference models.
In recent years, nano-Au has become a research hotspot due to its small size, high area-volume ratio [17], unique optical, electrical, thermodynamic properties [18], high catalytic activity and unique bioaffinity [19]. This demonstrates the potential application value in electrochemical sensors and other fields [20]. MWCNTs can be used as high-conductivity materials [21] and are widely used to enhance electron transfer between electrode surfaces and analytes [22] because of their high specific surface area and excellent mechanical strength. Nanoscale zinc oxide (ZnO) is an inorganic nanomaterial with many advantages, including excellent magnetic, optical and electrical properties [23], and it has promising applications in electrochemical sensors. Herein, as an effective substrate to fix gold nanoparticles, ZnO-CNTs enhance the stability of the electrode and the electron transfer ability [24], and the rapid and sensitive detection of trace Hg(II) in seawater can be realized effectively.

Instruments:
Electrochemical experiments were carried out using a CHI 830D (Shanghai Chenhua) computer-controlled potentiostat. A standard three-electrode system was utilized for the electrochemical measurements. A model PHS-25 pH meter was used to detect pH. The morphology of the modified electrode was photographed by Zeiss SigmaHD SEM. The LC-DMS-H magnetic agitator and FA124 external school style electronic balance were purchased from Lichen Technology Co., Ltd. (Shanghai, China). Ultra-pure water was used throughout the experiment (Nanjing Yipu Yida).
Nitric acid, sulfuric acid, bismuth nitrate, absolute ethanol, acetate, anhydrous sodium acetate, potassium chloride and potassium dihydrogen phosphate were provided by Sinopharm Chemical Reagent Co., Ltd. (China). All reagents were analytical reagent grade.

Preparation of modification solution
The purification process of MWCNTs was realized by chemical acidification and oxidation. Approximately 0.5 g MWCNTs were ultrasonically treated in 200 ml H 2 O 2 and HNO 3 mixed solution at a ratio of 1:3 (V/V). After 3 h of ultrasonic treatment, the acid-treated MWNTS were diluted with 200 ml distilled water and filtered through filter paper with a porosity of 3 μm [25,26]. Then, the acid-treated MWCNTs were thoroughly washed with distilled water until a neutral pH value was reached and dried at 100 ± 0.5°C for 3 h [27].
The ZnO-CNT-modified solution can be obtained by ultrasonically dispersing 6 mg MWCNTs and 3 mg ZnO in 10 ml DMF for more than 30 min to present a uniform black mixture.
To compare the experiment, 6 mg of MWCNTs was placed in 10 ml DMF for ultrasonic dispersion for more than 30 min to present a uniform black mixture, and the MWCNT-modified solution was obtained.

Preparation of modified electrode
The bare GCE (3 mm in diameter) was polished to the mirror surface with 1.0 μm, 0.3 μm and 5 nm Al 2 O 3 powder in turn. After washing, the surface powder was sequentially removed by ultrasonic cleaning with ultrapure water and absolute ethanol for 3 min. Then, the samples were scanned in a 0.5 mol l −1 H 2 SO 4 solution with cyclic voltammetry (CV) from −1.0 V to +1.0 V until it is stable. Finally, the bare GCE was rinsed with water and used for further measurements. After accurately pipetting, 4 μl of ZnO-CNT-modified solution was accurately transferred with a pipette, uniformly dropped onto a bare GCE and placed in a dryer to obtain a ZnO-CNT/GCE electrode.
Then, the chlorauric acid solution was potentiostatically deposited at an accumulation potential of −0.25 V for an accumulation time of 20 s. After removal, the ZnO-CNTs/Nano-Au/GCE was washed and dried with deionized water.

Experimental measurement
The ZnO-CNTs/Nano-Au/GCE was put into 0.1 M sodium acetate buffer solution (HAc-NaAc) containing a certain concentration of Hg(II), and the Hg(II) was enriched by electrodeposition under stirring conditions. The accumulation potential was −1.1 V, and the accumulation time was 720 s. After the accumulation, the stirring was stopped. Then, preelectrolysis was carried out at −0.8 V potential with a potential increment of 50 mV, frequency of 25 Hz, amplitude of 50 mV, and rest time of 4 s. The stripping voltammetry diagram corresponding to the Hg(II) concentration was recorded, and a linear standard curve was made to calculate the detection limit (figure 1).
To ensure the rigor of the experiment, the residual Hg(II) on the reaction surface of the modified electrode was removed by potentiostatic techniques (potential −0.6 V, time 30 s) after the above experimental process was completed, and all of the electrochemical experiments were conducted at room temperature. Figures 2(a) and (b) depict SEM images of MWCNTs and ZnO-CNTs treated by shock mixing in DMF. As seen in figure a, MWCNTs were observed to form an interconnected network on the bare GCE surface under a highmagnification SEM image. There was uneven mixing and a small amount of agglomeration. Figure 2(b) show that large quantities of ZnO-CNTs uniformly covered the surface of the bare GCE. ZnO is embedded in CNTs to reduce agglomeration and entangle each other, indicating that the two composites have enough micropores, which can provide active spots for the REDOX reaction of materials and improve the contact effect between electrolyte solution. At the same time, the high electrical conductivity and specific surface area of CNTs enhance the electrochemical performance of the composites.

Morphology analysis of MWCNTs and ZnO/CNTs
A comparison of figures 2(a) and (b) shows that MWCNTs are widely distributed on the surface at the same magnification. Figure 2(b) form a dense structure that is intertwined with each other and the morphology distribution is more uniform. Figure 3(a) shows the analysis diagram of carbon nanotubes with strong reflection. Figure 3(b) shows the XPS spectra of Zn and O. Figure 3(c) shows the XPS spectra of C, O and Zn elements on the surface of MWCNTS and ZnO composite nanoparticles.
The composite material has three obvious peaks at the binding energy of 285.0, 532.9 and 1021.7, corresponding to C1S, O1S, Zn2P1 and Zn2P3, and therefore contains C, O and Zn elements. It can be seen from the position height of each peak in the figure that the peaks of C, O and Zn are obvious. Therefore, it can be concluded that despite the interference of noise signals, A state in which the presence of precious metals can still be determined to be primarily in metal atoms. Figure 4(a) compares the influence of electrolyte types on the peak current of dissolution. Under the same experimental conditions, the response peak current of 0.1 M HAc-NaAc buffer solution is larger than that of the phosphate buffer solution. Therefore, 0.1 M HAc-NaAc was selected as the buffer solution.The reason is that acetate can be complexed with metal ions to form very stable new ions and promote the electron transfer rate. Therefore, 0.1M HAc-NaAc was selected as the buffer solution. Figure 4(b) displays the current response of Hg(II) in a 0.1 M HAc-NaAc solution, wherein the pH values varied from 3.0 to 6.0. When the pH value is from 3.0 to 4.0, the response peak current increases, and when the  pH value is over 4.0, the response current decreases. The reason is that when the PH value is too low, the hydrogen evolution phenomenon leads to the decrease of Mercury deposition, and when the PH value is too high, the complex will be generated and affect the dissolution signal. Thus, based on the aforementioned results, pH 4.0 was employed for further measurements. Figure 5(a) shows the peak current densities of Hg(II) on the ZnO-CNTs/Nano-Au GCE by varying the accumulation potential between −1.30 and −0.40 V. As the potential moves forward, from −1.30 V to −1.10 V, the current densities increased significantly, which was due to the enhanced kinetics; that is, Hg(II) could be easily reduced at more negative accumulation potentials. The current densities reached a maximum at an accumulation potential of −1.10 However, when the accumulation potential continued to shift forward, from −1.10 V to −0.40 V, the current densities decreased, which was attributed to the interference of hydrogen evolution. Thus, the accumulation potential of −1.10 V was selected as optimal.

Effects of accumulation potential and accumulation time
Second, as shown in figure 3(b), the peak current densities increased with longer accumulation times, from 60 to 900 s, and then slowly increased with accumulation times beyond 720 s. The reason is that there are a limited number of sites to modify the electrode, and over time, the supply of sites is limited, Therefore, an accumulation time of 720 s was selected to reach a lower detection limit and wider range of response.

Effect of concentration
The effects of concentration mainly include MWCNT concentration, ZnO concentration and film thickness. First, the influence of the MWCNT concentration on the peak current was investigated. As shown in figure 6(a), it can be seen from the figure that the peak current of Hg(II) is the largest when the concentration of MWCNTs is 0.6 mg ml −1 because the adsorption capacity is insufficient when the concentration is too low of MWCNTs, and the background current will be too large when the concentration is too high. Therefore, the concentration of MWCNTs was 0.6 mg ml −1 .
Second, the influence of ZnO concentration was investigated. As shown in figure 4(b), during the change in ZnO concentration from 0.1 mg ml −1 to 0.7 mg ml −1 , the peak current appeared at 0.3 mg ml −1 . Therefore, the concentration of ZnO was 0.3 mg ml −1 .
Finally, the influence of film thickness on the dissolution peak current of Hg(II) was investigated, as shown in figure 4(c). The volume of the modified film was changed from 1 μl to 6 μl. The peak current reached the maximum value when the volume of the modified film was 4 μl because the film was too thick to reduce the transmission of electrons, and the film was too thin to achieve the best adsorption capacity. Therefore, the film thickness selected in this paper is 4 μl.

Comparison of different modified electrodes
To compare the adsorption capacity of ZnO-CNTs/Nano-Au modified GCE. Under the optimum conditions, the dissolution curves of the 4.47 μM Hg(II) standard solution were tested by the bare GCE, MWCNT-modified GCE, ZnO/CNT-modified GCE and ZnO-CNT/Nano-Au-modified GCE. As shown in figure 7, the bare GCE and other modified GCE have obvious dissolution peaks of Hg(II) at approximately 0.191 V, the dissolution peak current of the bare GCE is the smallest, and the dissolution peak current of ZnO-CNTs/Nano-Au-modified GCE is the largest, which indicates that the modified electrode of this composite material has good sensitivity and selectivity.  Figure 8 presents DPV curves for the electrochemical detection of Hg(II) at the ZnO-CNTs/Nano-Au GCE in HAc-NaAc solution (0.1 M, pH 5.0) based on optimized conditions. The well-defined peak for Hg(II) may be clearly observed at +0.191 V with higher concentrations of Hg(II).

Standard curve
The linear regression curve of the ZnO-CNTs/Nano-Au GCE for Hg(II) detection is presented in figure 9, which gave a correlation coefficient of 0.991 over a linear range from 1.49 to 5.97 μM. The calculated detection limit was 0.0118 μM. Based on the listed results, it was indicated that the ZnO-CNTs/Nano-Au GCE exhibited favorable electrochemical performance for Hg(II) detection with high sensitivity and a low detection limit.

Interference experiment
Interference from common ions on the electrochemical analysis of Hg(II) was further investigated. Figure 10 depicts the voltammetric responses of 1.49 μM Hg(II) in the absence or presence of ions, including K + , Ba 2+ , Cd 2+ , Pb 2+ and Na + . The experimental results show that the peak current change rate of different ions to Mercury ions is about 5% These results indicated that the ZnO-CNTs/Nano-Au-modified GCE possessed an efficient anti-interference capacity.

Stability and reproducibility experiment
The stability and reproducibility of the GCE are important indicators to characterize its practicality. To verify the stability of the modified GCE, the same GCE was used to measure the 2.48 μM Hg(II) solution 6 times. As shown in figure 11(A), the relative standard deviation (RSD) was 3.007%, which indicated that the modified GCE had good stability. At the same time, under the same experimental conditions, six modified GCEs of this composite material were prepared, DPV of a 1.49 μM Hg(II) solution was performed, and the reproducibility of the GCE was tested. As shown in figure 11(B), the relative standard deviation (RSD) is 2.886%, which indicates that the composite-modified GCE has good stability.

Sample analysis
Qinzhou Bay area in Guangxi is an important support for the national construction of Beibu Gulf urban agglomeration. Qinzhou Bay is divided into inner bay and outer bay, which is called Maowei Sea. Maowei Sea belongs to the intersection of rivers entering the sea and the outer bay of Qinzhou Bay, and is a key area for the exchange of material and energy between rivers and oceans. Hg(II) in Maowei Sea is at a moderate ecological threat level. Water samples from three different places in Maowei Sea were collected, left standing for 3 days, filtered by a 0.45 μl microporous membrane and pretreated by ultraviolet rays, and the concentration of Hg(II) in the above water samples was determined by the standard addition method. At the same time, the spiked recovery experiment was performed. The experimental results are shown in table 1, and the spiked recovery is between 94.2% and 98.4%, which indicates that the modified electrode can be applied to the detection of trace Hg(II) concentrations in actual samples.

Conclusion
CNTs have high conductivity, high mechanical strength and good electron transport performance. ZnO has the surface effect and the small size effect, has excellent physical, chemical and surface properties of gold nanoparticles have special biocompatibility and affinity function has been widely applied in the detection of heavy metals. In this paper, CNTs and ZnO composite base seat, effectively reduce the reunion phenomenon, improve the activity of the points, to facilitate rapid and sensitive detection of Mercury ion in the water.
The preliminary preparation work of this paper is as follows: the acidified carbon nanotubes are first prepared, and the carbon nanotubes and zinc oxide are mixed with dimethylformamide in a certain proportion. After full mixing, ultrasonic shock more than 30 min, sealed reserve. Then the experimental conditions were optimized from the aspects of cumulative potential, cumulative time and film thickness to seek the best combination conditions, which effectively reduced agglomeration and improved the active point.The material was tested and analyzed by SEM and XPS scanning images, which showed that the composite had a good application effect for detecting Mercury ions in seawater.
Finally, the actual seawater is sampled to detect the Mercury content in seawater. Compared with the data changes of actual seawater and added standard solution, the recovery rate meets the range of 90%∼110%.
The rise and development of nano materials have already had a huge application prospect in environmental detection. Common heavy metal detection equipment is expensive and cumbersome to operate. Composite materials as a substrate combine with nano particle materials to provide a new research idea for the analysis and detection of heavy metal ions.