1. Introduction
Excessive heavy metal ions from industrial sewage and emissions cause a serious threat to marine ecosystems due to scientific and technological developments. Their carcinogenicity, teratogenicity, and mutagenicity [
1] are increasing hazards to public health and safety. Mercury, chromium, cadmium (Cd), lead, and zinc (Zn) are particularly relevant. This study focuses on Cd and Zn ions in seawater. The heavy metals Cd and Zn tend to coexist in the aquatic environment; further, because of their similar atomic structure, ionic radius, and electronegativity, they may interact and interfere with each other in the same biological system.
Zn is an essential trace element for aquaculture animals. Zn deficiency or excess in feed has serious negative effects on the growth, development, and nutritional metabolism of aquatic animals; therefore, it is particularly relevant to investigate the mechanism of Zn homeostasis regulation in organisms [
2]. Cd is one of the non-essential trace harmful elements in organisms. In recent years, the increase in industrial pollution has caused Cd to exceed the standard for aquatic animals. Cd has a long residual time, can accumulate and can be transferred along the food chain, thus posing a threat to human health [
3].
The joint toxicity test showed that high concentrations of Cd(II) and low concentrations of Zn(II) on the clam have a synergistic effect, and with a decrease in the concentration of Cd(II), the synergistic effect becomes weaker; high concentrations of Zn(II) and low concentrations of Cd(II) on the clam have an antagonistic effect, but with an increase in the concentration of Zn(II) this effect shifts toward each acting as an independent agent [
4]. According to Hou Liping et al., through the combined toxicity study of Cd and Zn in grass carp, it was found that the coexistence of metals increased the permeability of the cell membrane, and thus toxicity increased synergistically. Moreover, the enrichment of Cd in fish can change the permeability of the cell membrane [
5]. Therefore, it is particularly relevant to establish a rapid, convenient, and low-cost assay for the mixed detection of Cd and Zn ions.
Currently, apart from electrochemical analysis, the main methods most commonly used for heavy metal detection include spectrophotometry, chemiluminescence, high-performance liquid chromatography, ion chromatography, atomic spectrometry, and mass spectrometry [
6,
7,
8,
9,
10]. Although these methods are accurate and effective, they are complicated to operate and costly. In the current study, we selected the electrochemical method. Electrochemical analysis has become a relevant method for the detection of heavy metals because of its environmental protection, easy and convenient operation, simple instrumentation, ease of integration, and miniaturisation [
11,
12,
13]. In electrochemical detection, it is necessary to convert complex and difficult-to-measure chemical parameters into easily accessible electrical parameters and use the changes in electrical parameters to achieve rapid detection of chemical reaction processes [
14]. The traditional three-electrode detection system is composed of an electrolyte solution, working electrode, reference electrode, auxiliary electrode, electrochemical workstation, and host computer. The electrodes are separate and work together, forming a “three electrodes, two circuits” structure with the power supply, voltmeter, and ammeter in the electrochemical workstation [
15]. In this experiment, the working electrode selected is a glassy carbon electrode. Glassy carbon electrode has the advantages of good electrical conductivity, high chemical stability, small thermal expansion coefficient, hard texture, good airtightness, and a wide range of polarisation. It can be made into a cylinder, disc, and other electrode shapes and can be used as an inert electrode for anodic dissolution, and cathodic and voltammetric determination of valence ions. It can also be used as a matrix to create a mercury film glassy carbon electrode, chemically modified electrodes, and so on. It has been widely used in electrochemical experiments or electroanalytical chemistry. In the electrochemical method, the modification of the surface of the working electrode is particularly relevant, and in this thesis, graphene oxide and multi-walled carbon nanotubes were selected.
Graphene, a new type of carbon nanomaterial, has been widely used in recent years; further, this quasi-two-dimensional nanomaterial [
16,
17] has the advantages of a rapid electron transfer rate and high electrocatalytic activity [
18]. After the oxidation of ordinary graphene, graphene oxide is obtained, and a large number of oxygen-containing functional groups are added to its surface, which makes it more active and catalytic [
19]. The modification of graphene oxide on the surface of glassy carbon electrodes can be reduced electrochemically, and the specific surface area of the electrode and the detected signals are significantly improved after the reduction [
20]. Another type of nanomaterial, multi-walled carbon nanotubes [
21], has been widely used in the fields of environmental, food, and pharmaceutical analyses since its discovery due to its good mechanical strength, unique surface effect, electrical conductivity, and strong catalytic properties [
22].
In the current study, for the detection of heavy metal contaminants in the marine environment, graphene oxide and multi-walled carbon nanotubes were used to prepare enhanced, modified membranes. A rapid, wide-detection-range, high-sensitivity microfluidic detection platform was studied to achieve real-time, rapid detection of zinc and cadmium ions in water, effectively preventing zinc and cadmium ions from causing serious harm to marine ecosystems and human beings.
2. Experimental
2.1. Instruments and Reagents
The CHI830D electrochemical workstation (Shanghai, China), FA224 electronic analytical balance (Shanghai Hai Lichen Instrument Technology, Shanghai, China), JP-040s ultra-sonic cleaner (Shenzhen Jianmeng Ultra-sonic Instrument, Shenzhen, China), LC-LX-H165A high-speed centrifuge (Shanghai Hai Lichen Instrument Technology), DHG-101 electric blast drying oven (Shanghai Changyi Instrument Equipment, Shanghai, China), TESCAN MIRA LMS high-resolution scanning electron microscope (TESCAN Limited, Brno, Czech Republic), and PLUSE2-5TH ultra-pure water system (Nanjing Easy Pure Tech Development, Nanjing, China) were used in this study.
Carboxylated multi-walled carbon nanotubes (purity ≥ 99.9%), graphene (purity ≥ 98%), and dimethylformamide (DMF) (purity ≥ 99.8%), were from Shanghai Aladdin Biochemical Technology, Shanghai, China; cadmium ion standard solution (1 mg/mL) and zinc ion standard solution (1 mg/mL) were from the National Nonferrous Metals and Electronic Materials Analysis and Testing Centre. The purities used were either analytical pure or good-grade pure. Ultra-pure water with a resistivity of 18.0 MΩ·cm was used in the experiment. The seawater samples for the spiked recovery experiments were collected from lakes near Beibu Gulf University, originating from waters close to the Maowei Sea.
2.2. Preparation of a Modified Electrode
The surface of the glassy carbon electrode was polished with alumina paste powder with diameters of 1.0 μm, 0.3 μm, and 50 nm and washed in ultra-pure water and anhydrous ethanol by ultra-sonic washing for 3 min. The cleaned electrode was placed into a 0.1 mol/L potassium ferrocyanide solution for cyclic voltammetry testing.
Appropriate amounts of GO and MWCNT-COOH powder were weighed, added to DMF, and ultra-sonically dispersed for 8 h to obtain stable dispersion. Briefly, 10 μL of the suspension was applied vertically dropwise to a clean GCE surface. GO/MWCNT-COOH/GCE is produced by drying in an electric blast drying oven. The electrically reduced graphene oxide/carboxylated multi-walled carbon nanotube composite-modified electrode (rGO/MWCNT-COOH/GCE) was obtained by a constant potential reduction for 600 s.The schematic diagram of detection mechanism is shown in
Figure 1.
2.3. Experimental Measurement
First, the standard solutions of Zn and Cd ions were diluted to different concentrations by using 0.1 mol/L pH 5.0 sodium acetate buffer (HAc-NaAc). Subsequently, the rGO/MWCNT-COOH/GCE, platinum wire electrode, and saturated mercuric chloride electrode were placed into the solution for constant potential enrichment of Cd(II) and Zn(II), with an enrichment potential of −1.3 V and an enrichment time of 500 s. Finally, the differential pulse voltammetry (DPV) detection was performed after the enrichment was completed; further, the scanning potential interval was set at −1.3 to −0.6 V, the potential increment at 50 mV, the frequency at 25 Hz, the amplitude at 5 mV, and the resting time at 30 s.
3. Results and Discussion
3.1. Electrochemical Characterisation of Modified Materials
CV scanning was performed by placing GCE, GO/GCE, MWCNT-COOH/GCE, and rGO/MWCNT-COOH/GCE in a 1 mmol/L potassium ferricyanide solution, and the experimental results are shown in
Figure 2a–c. The effective surface area was calculated to be 0.296 cm
2 for GO/GCE, 0.519 cm
2 for MWCNT-COOH/GCE, and 1.09 cm
2 for rGO/MWCNT-COOH/GCE. rGO/MWCNT-COOH/GCE showed a significant increase in redox peak current compared with GCE, suggesting that the rGO/MWCNT-COOH composite modification material can accelerate the electrode surface electron transfer rate, enhance the reversibility of the electrode reaction, and have excellent electrical conductivity [
23]. The ion-selective enhancement modification film was successfully modified on the surface of the microelectrode. It is also shown that the zinc ion selectively enhanced modified film can increase the specific surface area of the microelectrode and improve electrochemical detection of microelectrodes with increased reversibility of the electrode reaction. From the voltammogram, it can be seen that the oxidation peak current appears to increase rapidly when the scanning voltage reaches a certain value. This is because the voltage at this time makes the electrode sufficient to capture the electrons of the ions in the surrounding solution. As the ions are continuously consumed, the current begins to decrease until it is almost 0; later, when a reverse voltage is applied, the ions that had lost their electrons from the electrode gain electrons to return to a reduced state. This process is just the opposite of the oxidation peak, and the combination of the two is the reversibility of the electrode process.
Figure 3a,b depict scanning electron microscopy (SEM) images of rGO-MWCNT-COOH treated by shock mixing in DMF. After a magnification of 5000×, it is evident that graphene oxide forms a multi-layered lamellar uneven structure after reduction, and a large amount of dispersed MWCNT-COOH is attached to the electrode surfaces and between the crevices, which makes the specific surface area of the material increase significantly. Concurrently, the number of active sites on the surface of the material is also increased substantially. This is extremely favourable for the enhancement of the physical adsorption and electrochemical properties of the material.
3.2. Detection Performance of Modified Electrodes
The strength of three different electrodes, rGO/MWCNT-COOH/GCE, MWCNT-COOH/GCE, and GCE, in response to Cd(II) and Zn(II) signals, was compared by DPV. As shown in
Figure 4, all three electrodes show apparent dissolution peaks of Zn(II) at −1.2 V and Cd(II) at −0.8 V. Because of the combination of rGO and MWCNT-COOH, the dissolution peak current of rGO/MWCNT-COOH/GCE is larger than that of the other two working electrodes; further, the specific surface area of the working electrode is effectively enlarged, which provides better electrocatalytic performance and improves the detection signals for these two heavy metal ions. This is because rGO and MWCNT-COOH have good conductivity and large specific surface area for Cd(II) and Zn(II) adsorption, and MWCNT-COOH adsorbed on the surface of rGO increases the specific surface area, and also provides a large number of active sites, which is conducive to physical adsorption of the material and improves the electrochemical detection performance. This indicates that the microelectrodes modified by the zinc ion selectivity-enhanced composite membrane selected in this paper have good sensitivity and selectivity for zinc ion detection.
3.3. Optimisation of Measurement Conditions
3.3.1. Effects of Buffer Solution and pH Value
Buffer solutions have a direct impact on the stability, anti-interference, and peak size of zinc ion detection due to the different electrolytes in different buffer solutions, resulting in different effects on heavy metal ion detection. The DPV assay was performed in potassium chloride buffer solution (KCl), phosphate buffer solution (PBS), and acetate buffer solution (HAc-NaAc) that all contained 500 μg/mL of mixed ions; the concentration of all three buffer solutions was 0.1 mol/L and the pH was 5.0. The measurements indicated that the peak ionic currents measured in the HAc-NaAc buffer solution were greater than in the remaining two buffer solutions, and therefore the HAc-NaAc solution was chosen as the buffer solution in this paper.
Figure 5a,b show the current response of Cd(II) and Zn(II) in 1 mmol/L HAc-NaAc solution. The peak response current increases at pH values from 3.5 to 5.0 and decreases at pH values above 5.0. This is due to the phenomenon of hydrogen evolution that leads to a decrease in Cd and Zn deposition at low pH, whereas at high pH, complexes are created and affect the dissolution signal.
3.3.2. Effects of Accumulation Potential and Accumulation Time
Figure 6a show the peak current densities of Cd(II) on rGO/MWCNT-COOH GCE as the accumulation potential varies from −1.40 to −1.00 V. When the detection potential is shifted forward from −1.40 V to −1.30 V, the current density of Cd(II) increases significantly due to the enhanced kinetics. The current density of Cd(II) reaches its maximum value at an accumulation potential of −1.30 V; as the accumulation potential continues to move from −1.30 V to −1.00 V, the current density decreases due to the interference of hydrogen evolution.
In order to maintain consistency of experimental conditions,
Figure 6b show the peak current densities of Zn(II) on rGO/MWCNT-COOH GCE as the accumulation potential varies from −1.40 to −1.00 V. The peak current does differ much between −1.4 V and −1.3 V. This is due to the fact that Zn(II) accumulates more readily at more negative potentials, while beyond −1.3 V, the peak current drops significantly due to hydrogen evolution.
As shown in
Figure 7a, the longer the accumulation time, the higher the peak current density, from 180 to 500 s, and then it slowly decreases as the accumulation time exceeds 500 s for Cd(II). As shown in
Figure 7b, the longer the accumulation time, the higher the peak current density, from 180 to 550 s, and then it slowly decreases as the accumulation time exceeds 550 s for Zn(II). The reason is that during accumulation, if the diffusion rate of the analyte on the electrode surface is limited and the diffusion rate of the analyte is far lower than the accumulation rate on the electrode surface, a long accumulation time may lead to a thicker diffusion layer of the analyte, and the rate of increase of the analyte concentration on the electrode surface slows down, which leads to a decrease in the potential.
3.3.3. Effect of Film Thickness
The effect of film thickness on the peak current of Cd(II) and Zn(II) dissolution was investigated, as shown in
Figure 8a,b. The film thickness referred to in this article refers to the amount of drop coating of the modified material. In the experiment, the volume of the modified film was selected from 5 to 9 μL. And the peak current reaches the maximum value when the volume is 9 μL, which was attributed to the fact that too thick a film reduces the electron transmittance, while too thin a film does not have the optimum adsorption capacity.
3.4. Orthogonal Experiments with Optimised Conditions
In this paper, two metal ions are detected at the same time. In order to be able to determine the optimal optimisation conditions while keeping the experimental conditions consistent, orthogonal tests were designed to validate the above-selected optimal experimental conditions. The orthogonal table of type L
9(3
4) was selected to establish a four-factor, three-level experimental condition, as shown in
Table 1. According to the four factors of different pH, accumulation potential, accumulation time, and film thickness, the optimal levels of the previous experiments were taken with the left and right two neighbouring two levels, respectively, to construct a four-factor, three-level orthogonal table. That is, L
9(3
4)-type orthogonal table, where L is the table code, 9 is the number of experimental programmes, 3 is the number of levels, and 4 is the number of factors.
After selecting the orthogonal test factor level table, the orthogonal test assistant software “Microsoft Excel 2016” can be used to fill in the corresponding factors and levels to generate the orthogonal test protocol table. This is because the orthogonal test protocol table has already disrupted the factor level arrangement and combinations so that the optimal conditions can be verified in the minimum number of experiments. In the experiment, each level was calculated three times, calculating the mean and extreme deviation of the results of three experiments at the same level and establishing effect curves for trends in the levels of each factor based on mean values. It was determined by a calculation that 5.0 was the optimal value of the three levels of the factor pH, −1.3 V was the optimal value of the three levels of the factor potential, 500 s was the optimal value of the three levels of the accumulation time, and 9 μL was the optimal value of the three levels of the film thickness.
3.5. Mixed Determination of Heavy Metal Ions
Under the optimal conditions, the DPV detection was conducted on the mixed solution samples with different concentrations of Cd(II) and Zn(II), and the results of the DPV curves in the range of 5–400 μg/L are shown in
Figure 9.
Figure 10a shows the relationship between the oxidation peak current of Cd(II) and its concentration, which showed a good linear relationship in the range of 5–400 μg/L. The calculated linear regression equation was Y = 0.1546X + 8.6955, with a linear correlation coefficient of R
2 = 0.955, and the limit of detection was 0.81 μg/L, which was lower than the permissible limit of Cd(II) concentration in drinking water stipulated by the WHO and China [
24,
25].
Figure 10b shows the relationship between the oxidation peak current and concentration of Zn(II), which showed a good linear relationship at 5–400 μg/L. The calculated linear regression equation was Y = 0.1158X + 3.453, with a linear correlation coefficient of R
2 = 0.983 and a detection limit of 0.98 μg/L. The experiments showed that rGO/MWCNT-COOH/GCE had good electrochemical performance, high sensitivity, and a low detection limit in the mixed solution of Cd(II) and Zn(II).
3.6. Stability and Reproducibility Experiment
In practical applications, stability is a relevant indicator for improving the usefulness of sensors. To evaluate the stability of rGO/MWCNT-COOH/GCE-modified materials in multiple experiments, we performed eight consecutive DPV experiments in a 0.1 mol/L HAc-NaAc buffer with a pH of 5. The concentrations of Cd(II) and Zn(II) were all 100 μg/L.
Figure 11a shows the experimental results for Cd(II), with an RSD of 4.1% for the peak dissolved current in eight DPV experiments.
Figure 11b shows the experimental results for Zn(II), with an RSD of 3.3% for the peak dissolution current in eight DPV experiments, which indicated that the rGO/MWCNT-COOH/GCE modification material had good stability.
Reproducibility is also an important indicator of sensor performance when it comes to the actual detection of heavy metal ions in the ocean. There are several key aspects to this. Stable signal output: The modified electrode should be able to produce a stable and reproducible electrochemical signal over multiple uses. This means that the electrode should be able to produce a similar response when multiple measurements are made under the same conditions, ensuring that the measured current or potential change in relation to analyte concentration is predictable and repeatable. Reduced measurement errors: Good reproducibility reduces measurement errors due to variations in electrode performance. If the response of the electrode varies greatly at different times or under different conditions, it may lead to inconsistency in the analytical results, which will affect the reliability and interpretability of the experimental results. Validation of the analytical method: In scientific research and industrial applications, there is a need to ensure the validity and credibility of the electrochemical analytical method. The good reproducibility of modified electrodes is one of the important bases for verifying the feasibility and validity of analytical methods. Cost-effectiveness and resource-saving: modified electrodes with good reproducibility can effectively reduce the number of experimental repetitions and save experimental costs and time. In research and application, it is often necessary to carry out multiple repetitions of experiments to ensure the reliability of the results, and a stable modified electrode can significantly reduce these costs.
According to
Figure 12a,b, five different batches of GCE electrodes with this composite material were prepared in a HAc-NaAc buffer solution with a concentration of 150 μg/L of Cd(II) and Zn(II). Based on the peak dissolution currents obtained from the five experiments, the relative standard errors (RSDs) were calculated to be 4.8% and 4.9%.
3.7. Disturbance Experiments
In order to evaluate the immunity of the microfluidic system and to ensure the detection performance of the system in the seawater base, the immunity of the system can be tested by adding interfering ions. When 50-fold Cl−, Na+, and K+ were added as interfering ions to a mixed buffer solution containing 300 μg/L of zinc–cadmium ions, the rates of change of the peak currents were all within 5%, indicating that the interfering ions at the above concentrations had no significant effect on the system. The interferences of three heavy metal ions, Pb(II), Hg(II), and Cu(II), on the detection of the system were also investigated because the heavy metal ions would occupy the binding sites of Zn(II) ion-selective enhancement modification membranes on the surface of microelectrodes with Zn(II) ions, which would affect the redox process of Zn(II) ions, and thus interfere with the system’s detection of Cd(II) and Zn(II). The experimental results showed that the variation rates of the peak dissolution currents of Cd(II) and Zn(II) under the conditions of 10-fold concentration of Pb(II), Hg(II), and Cu(II) were within 10%, which indicated that the microfluidic system studied in this paper has a strong ability to resist interference and can be used for zinc ion concentration detection in seawater bases.
3.8. Sensitivity Analysis
In order to further study the detection performance of the microfluidic system discussed in this paper on zinc ions, the detection effect of different electrodes on zinc ions in recent years will be compared this year, as shown in
Table 2 and
Table 3. The comparison results with other studies show that the rGO/MWCNT-COOH/GCE studied in this paper has a smaller gap with the results of other studies in terms of detection range and detection limit, with a wider detection range and lower detection limit, so the system can be applied to the detection of seawater-based Cd(II) and Zn(II).
3.9. Sample Analysis
Water samples were obtained from Beibu Gulf University’s internal lake water, originating from near-shore seawater in the Maowei Sea area, with a mixed composition.
Table 4 shows the results of the water quality analyses at the three sites. The water samples were collected from three different locations at large intervals in the university lake, and the samples were left for 48 h to fully precipitate the impurities. The upper layer of the sample was filtered using a 0.45 μm membrane, and 20 mL of the filtered sample was used for spiking and recovery experiments. From
Table 5 and
Table 6, it is evident that the concentration of Cd(II) and Zn(II) in the actual water samples is far lower than the maximum concentration of Cd(II) and Zn(II) in the drinking water specified by the state. The spiked recovery for Cd(II) is between 96.4% and 98.2%, and for Zn(II), it is between 96.7% and 99.4%; this modified electrode has good performance in detecting trace Cd(II) and Zn(II) concentrations in real samples.