Abstract
Microcystins with leucine arginine (MC-LR) is a virulent hepatotoxin, which is commonly present in polluted water with its demethylated derivatives [Dha7] MC-LR. This study reported a low-cost molecularly imprinted polymer network-based electrochemical sensor for detecting MC-LR. The sensor was based on a three-dimensional conductive network composed of multi-walled carbon nanotubes (MWCNTs), graphene quantum dots (GQDs), and gold nanoparticles (AuNPs). The molecularly imprinted polymer was engineered by quantum chemical computation utilizing p-aminothiophenol (p-ATP) and methacrylic acid (MAA) as dual functional monomers and L-arginine as a segment template. The electrochemical reaction mechanism of MC-LR on the sensor was studied for the first time, which is an irreversible electrochemical oxidation reaction involving an electron and two protons, and is controlled by a mixed adsorption–diffusion mechanism. The sensor exhibited a great detection response to MC-LR in the linear range of 0.08–2 μg/L, and the limit of detection (LOD) is 0.0027 μg/L (S/N = 3). In addition, the recoveries of the total amount of MC-LR and [Dha7] MC-LR in the actual sample by the obtained sensor were in the range from 91.4 to 116.7%, which indicated its great potential for environmental detection.
HIGHLIGHTS
A molecularly imprinted electrochemical sensor was proposed for detecting MC-LR.
The sensor was based on a 3D conductive network composed of MWCNTs, GQDs and AuNPs.
The molecularly imprinted polymer was engineered by quantum chemical computation.
The reaction mechanism of MC-LR on the sensor was studied.
The total amount of MC-LR in actual samples was detected successfully.
INTRODUCTION
Cyanobacterial blooms caused by eutrophication have become global environmental concerns (Huisman et al. 2018). Large numbers of freshwater lakes are infested with cyanobacteria, which pollute water sources by releasing microcystins (MCs) (Zhang et al. 2020a). MC with leucine arginine (MC-LR) is the most toxic and abundant MCs homolog (Li et al. 2016). Microcystis aeruginosa is considered to be the most prevalent microcystins producing organism, which produces both MC-LR and its demethylated derivatives [Dha7] MC-LR (Geada et al. 2017). Therefore, MC-LR in natural water is usually the mixture of these two isomers. The results of epidemiological experiments showed that MC-LR not only had significant toxic effects on the liver (Xu et al. 2021) but also had certain toxicity to the kidney, heart, reproductive system, and nervous system (Ma et al. 2021). In order to reduce the risk of MC-LR intake, the World Health Organization (WHO) recommended reference threshold for MC-LR in drinking water is 1 μg/L (WHO 1998).
Many analytical methods have been reported for detecting MCs, such as high-performance liquid chromatography tandem mass spectrometry (HPLC-MS) (Ríos et al. 2013), enzyme-linked immunosorbent assay (ELISA) (Reverté et al. 2013), colorimetric inhibition assay (CIA) (Sassolas et al. 2011), biochemical determination (Ikehara et al. 2008), and various sensing methods (Kordashta et al. 2020). Among these methods, HPLC and ELISA are the most widely used. Though both these methods have the advantages of good selectivity and low detection limits for the detection of MCs, they are difficult to apply on-site owing to the requirement of expensive equipment and long sample pretreatment time (Rasheed et al. 2019). However, a time-effective and low-cost electrochemical sensor based on molecular imprinted polymers (MIPs) can solve the above problems and has been widely used in field environmental detection, such as bisphenol A (Zhao et al. 2017), glyphosate (Uka et al. 2021), methyl parathion (Xue et al. 2014), estradiol (Duan et al. 2019), and so many. In recent years, some scholars have also tried to combine MIP with fluorescence (Qi et al. 2021), quartz crystal microbalance (QCM) (He et al. 2015), photoelectricity (Chen et al. 2012) and solid contact electrode (Queirós et al. 2012) for the detection of MC-LR. However, a molecularly imprinted electrochemical sensor still faces many challenges in detecting MC-LR, such as the large size and complex structure of MC molecules, the single imprinted recognition, and the poor electrical conductivity of typical molecularly imprinted polymers.
The recognition element and signal conversion element are two main components of molecularly imprinted electrochemical sensors, among which MIP is the recognition element that impacts the selectivity of the sensor, while various conductive materials are the signal conversion element that impact the sensitivity of the sensor. Currently, various materials have been developed to improve the sensitivity of molecularly imprinted electrochemical sensors, such as carbon nanomaterials (Kurbanoglu et al. 2019), metal nanocomposite (Lu et al. 2019), and magnetic substance (Yang et al. 2020). Multi-walled carbon nanotubes (MWCNTs) have been widely used as electrode modification materials due to their good electrical conductivity, large specific surface area, and high mechanical strength (Gui et al. 2018). And even, MWCNTs have been directly used as monomers and carriers to prepare molecularly imprinted electrochemical sensors for bisphenol A because it can form π bonds with aromatic compounds (Zhang et al. 2020b). However, due to the poor dispersibility and agglomeration tendency, MWCNTs must be surface modified to be used as electrode modification materials. Graphene quantum dots (GQDs) are a novel quasi-zero-dimensional carbon nanomaterial, whose size is less than 100 nm. The base surface of GQDs is oil-wetting sp2 carbon ring structures, and the edge contains abundant hydrophilic oxygen-containing functional groups. Therefore, GQDs can be used as a dispersant for surface modification of MWCNTs (Żołek et al. 2022), which not only solves the problem that MWCNTs are difficult to disperse but also significantly improves their conductivity and electron mobility. In the field of electrochemical sensing, gold nanoparticles (AuNPs) are considered good materials for modifying molecularly imprinted electrochemical sensors due to their easy preparation, good biocompatibility, and excellent electron transfer capability (Atta et al. 2019).
On the other hand, to improve the selectivity of the sensor, it is necessary to strengthen the interaction between templates and monomers and increases the number and category of imprints. Studies have shown that compared with the traditional single monomer, the carefully designed dual monomers cannot only enrich the diversity of imprinted dot types in the MIP architecture but also enhance the binding and affinity between templates and monomers through the synergistic effect of different monomers, which is conducive to improving the selectivity of MIP (Wu et al. 2016; Li et al. 2019).
In addition, the template molecules used in the construction of molecularly imprinted polymers also have a significant impact on the performance of the sensor. Due to its high price, high toxicity, and leakage risk, it is difficult to directly use MC-LR as a template molecule (Qi et al. 2021). Fortunately, Mbukwa et al. (2013) have successfully prepared MIP using L-arginine, which is a part of microcystin, as the segment template for selective recognition and solid-phase extraction of microcystins. Besides, it was shown that the guanidine terminus of L-arginine is geometrically stable, positively charged and contains five potentially active hydrogen bonding sites (Shreder et al. 2004; Schug & Lindner 2005), which are advantageous to enhance the affinity between templates and targets of the same category.
In this work, a novel molecularly imprinted electrochemical sensor based on MWCNTs, GQDs, and AuNPs modified glassy carbon electrode (GCE), and MIP with binary functional monomers was prepared for detecting MC-LR. L-arginine was selected as the segment template, p-aminothiophenol (p-ATP) and methacrylic acid (MAA) were selected as dual functional monomers through density functional theory (DFT) calculation, and MIP was prepared by in situ electrochemical polymerization. The recognition reaction mechanism and detection properties to MC-LR by the sensor were studied.
EXPERIMENTAL PARTS
Reagents and materials
MWCNTs were produced by Shandong Dazhan Nanomaterials Co. (Shandong, China). The aqueous solution of GQDs were purchased from Qingdao Haixi New Materials Co. (Shandong, China). Nafion solution (5.0% wt) was purchased from Dupont (Delaware, USA). The MC-LR standard sample (E-L-R-c100 microcystin LR ≥ 95%) was purchased from Puhuashi Technology Development Co. (Beijing, China). L-arginine, MAA, acrylamide (AM), and HAuCl4p-ATP were purchased from Macklin Biochemical Co. (Shanghai, China). All solutions were prepared by using double distilled water. Tap water samples were collected from municipal tap, Qingdao, China. The actual sample containing MC-LR and [Dha7] MC-LR was cultured and purified as described in previous work (Zhao et al. 2013), and the brief process description can be found in the Supplementary Material.
Instrument
All electrochemical experiments, including cyclic voltammetry (CV), differential pulse voltammetry (DPV), and electrochemical impedance spectrum (EIS), were performed on Model CHI 1140 and Gamry Reference 600 electrochemical analyzer, using a platinum plate (S = 5 cm2) as the counter electrode and a saturated calomel electrode (SCE) as the reference electrode. The surface morphology of the modified electrodes was characterized by SEM S-4800 (Hitachi, Japan). The structure of the modified materials was characterized by a Tensor-27 Fourier Transform InfraRed (FT-IR, Bruker Spectroscopy Instruments, Germany). Chromatographic analysis was implemented by HPLC-MS (Ultimate 3000 HPLC, Germany; AB-Sciex Qtrap 4500 mass spectrometer, Singapore). All experiments were done at room temperature (25 ± 1 °C).
Quantum chemical simulation
3D chemical structure was generated by Gaussian View 5.0. Computer simulations were performed by Gaussian 09W software. The geometries of the segment template molecule complexed with the functional monomers were optimized and calculated by the DFT at B3LYP/6-31G (d, p) algorithm. The binding energies (BEs) of the template molecules and functional monomers were simulated and optimized. The basis set superposition error (BSSE) was corrected by the counterpoise method.
Auto Dock Vina was used for the molecular conformational evaluation of GQDs, MWCNTs, and p-ATP, optimization for effective conformations of molecules, docking simulations, intermolecular binding sites, and interaction calculations.
Preparation of MIP-based electrochemical sensors
- (i)
At first, 8 mg of MWCNTs were added into 8 mL 1 g/mL GQDs aqueous solution, and the resulting mixture was ultrasonicated for 30 min. Then 0.25 mL of Nafion solution (5% wt) was added to the dispersion and ultrasonicated for 20 min to produce MWCNTs/GQDs dispersion.
- (ii)
The GCE was polished with aluminum oxide polishing powder on sandpaper, and then ultrasonic cleaning was performed in aqueous nitric acid solution, aqueous ethanol solution, and secondary distilled water in sequence. Then, the GCE was electrochemically activated by CV at 50 mV/s for 10 cycles in 0.1 M H2SO4 solution with a potential range from −1.0 to +1.0 V. After that, the MWCNTs/GQDs/GCE was obtained by applying a certain amount of MWCNTs/GQDs dispersion dropwise on the GCE surface and drying in a desiccator for 30 min.
- (iii)
To prepare AuNPs@MWCNTs/GQDs/GCE, AuNPs were electrodeposited on the MWCNTs/GQDs/GCE by applying cyclic potential from −1.2 to +0.6 V in 0.1 M KCl solution containing 0.2 g/L HAuCl4.
- (iv)
The AuNPs@MWCNTs/GQDs/GCE was immersed in 15 mL of ethanol solution containing 20 mM p-ATP for 3 h to form Au–S bonds between AuNPs and the thiol group (–SH) of the p-ATP molecule.
- (v)
Subsequently, MIP/AuNPs@MWCNTs/GQDs/GCE (with L-arginine) was produced by electrochemical polymerization at the potential range of −0.4 to +0.9 V at 50 mV/s in 20 mL phosphate buffer solution (PBS, 0.2 M, pH 7.0) with a 20% volume of ethanol solution to ensure the solubility, containing 3 mM L-arginine, 3 mM p-ATP, and 6 mM MAA.
- (vi)
Finally, the obtained MIP/AuNPs@MWCNTs/GQDs/GCE (with L-arginine) was eluted with a methanol/acetic acid (9:1, v/v) mixture for 15 min to remove the template molecules off. So far, the sensor has been obtained, which was MIP/AuNPs@MWCNTs/GQDs/GCE. To compare, the non-imprinted polymer modified electrode (called NIP) was also prepared in the same way, except for the absence of the template molecule (L-arginine).
The preparation, properties, and mechanism determination process of the sensors.
The preparation, properties, and mechanism determination process of the sensors.
Electrochemical procedures
The electrochemical behavior of modified electrodes by CV was carried out in 0.1 M KCl containing 1.0 mM [Fe(CN)6]3−/4− with a scan rate of 50 mV/s between −0.2 and 0.6 V. The electrochemical impedance spectroscopy (EIS) was carried out in 0.1 M KCl containing 1.0 mM [Fe (CN)6]3−/4− at the open circuit potential (with a frequency range of 10−2–105 Hz and amplitude of 5 mV). The electrode reaction mechanism by CV was carried out in 0.2 M phosphate buffer (PB) solution (by mixing 0.2 M KH2PO4 and 0.2 M K2HPO4, and its pH was adjusted using 1.0 M H3PO4 solution) containing 1 μM L-arginine or MC-LR between −1.0 and 1.3 V. The optimization of experimental conditions and properties study were performed by DPV in 0.1 M KCl containing 1.0 mM [Fe (CN)6]3−/4− with the pulse period of 0.2 sand the pulse width of 0.05 s. All peak current values measured in this study need to be deducted for baseline effects.
RESULTS AND DISCUSSION
Molecular modeling and docking calculations
(a) Interactions in the pre-assembled complex. The dashed lines represent the presence of a hydrogen bond. The length and angle of the hydrogen bonds are indicated by the numbers next to the dashed lines. (b) π–π interaction between MWCNTs/GQDs and p-ATP.
(a) Interactions in the pre-assembled complex. The dashed lines represent the presence of a hydrogen bond. The length and angle of the hydrogen bonds are indicated by the numbers next to the dashed lines. (b) π–π interaction between MWCNTs/GQDs and p-ATP.
Molecular docking studies help calculate intermolecular electrostatic interactions with the energy minimized poses (Liu et al. 2018). Intermolecular electrostatic interactions between MWCNTs, GQDs, and p-ATP were studied computationally by molecular docking. Auto Dock Vina was used to calculate the interaction forces of 16 possible models, and the results are shown in Supplementary Table S2. The optimal docking bonding energy of MWCNTs and p-ATP was −8.9 kcal/mol, and the optimal docking model is shown in Figure 2(b). It can be predicted that p-ATP not only forms Au–S bond with AuNPs but also plays an important role in the synthesis of imprinted polymers through π–π superposition with the MWCNTT/GQDs complex.
Characterization of different modified electrodes
SEM images of (a) raw MWCNTs powder dispersed in water, (b) MWCNTs after dispersion in GQDs solution, (c) AuNPs@MWCNTs/GQDs, (d) MIP/AuNPs@MWCNTs/GQDs/GCE (with L-arginine), (e) MIP/AuNPs@MWCNTs/GQDs/GCE (after elution), and (f) FT-IR spectra of MIP (with L-arginine), MIP, and NIP.
SEM images of (a) raw MWCNTs powder dispersed in water, (b) MWCNTs after dispersion in GQDs solution, (c) AuNPs@MWCNTs/GQDs, (d) MIP/AuNPs@MWCNTs/GQDs/GCE (with L-arginine), (e) MIP/AuNPs@MWCNTs/GQDs/GCE (after elution), and (f) FT-IR spectra of MIP (with L-arginine), MIP, and NIP.
The surface morphology of AuNPs@MWCNTs/GQDs and MIP/AuNPs@MWCNTs/GQDs before and after elution is shown in Figure 3(c)–3(e), respectively. From Figure 3(c), it can be seen that AuNPs were electrodeposited on the surface of MWCNTs/GQDs/GCE successfully. After the electro-polymerization of L-arginine and MAA, a smooth and homogeneous film was formed on the MIP AuNPs@MWCNTs/GQDs (with L-arginine) (Figure 3(d)). Many cavities with special shapes appeared on the surface of MIP/AuNPs@MWCNTs/GQDs (Figure 3(e)), which were molecularly imprinted cavities formed due to the elution of L-arginine.
FT-IR was used to analyze the functional groups of NIP and MIP before and after elution (Figure 3(f)). The strong band at 3,443 cm−1 may be ascribed to stretching the vibration of the –OH, and the absorption peaks at 2,360–2,370 cm−1 may be attributed to the Au–S bond stretching vibration, which also indicated the successful self-assembly of p-ATP with AuNPs. The absorption peaks at 1,280 and 1,139 cm−1 were attributed to the stretching vibrations of C–N and N–H, respectively. The moderate band at 1,620 cm−1 may be corresponding to the stretching vibration of C = C, indicating the presence of MAA. In addition, the weak shoulder peak at 1,540 and 788 cm−1, which was recognized as the C = N and –CNH groups among the L-arginine, respectively (Kumar & Rai 2010). For the MIP and NIP, the absence of the same peaks manifested that L-arginine did not exist, which was the fundamental difference among MIP, NIP, and MIP (with L-argigine). These results explicated the successful synthesis of MIP (with L-argigine), together with MIP and NIP.
Electrochemical characterization of different modified electrodes
(A) CVs of the bare GCE (a), MWCNTs/GQDs/GCE (b), AuNPs@MWCNTs/GQDs/GCE (c), MIP/AUNPS@MWCNTS/GQDS/GCE before (d) and after elution (e) and NIP (f). (B) Nyquist plot of the bare GCE, NIP, MIP/AUNPS@MWCNTS/GQDS/GCE before and after elution in 0.1 M KCl containing 1.0 mM [Fe (CN)6]3−/4−.
(A) CVs of the bare GCE (a), MWCNTs/GQDs/GCE (b), AuNPs@MWCNTs/GQDs/GCE (c), MIP/AUNPS@MWCNTS/GQDS/GCE before (d) and after elution (e) and NIP (f). (B) Nyquist plot of the bare GCE, NIP, MIP/AUNPS@MWCNTS/GQDS/GCE before and after elution in 0.1 M KCl containing 1.0 mM [Fe (CN)6]3−/4−.
Nyquist plots of different modified electrodes are shown in Figure 4(b). The EIS was analyzed with the Randles equivalent circuit model, and the Rct value was obtained by calculating the semicircle diameter of the high-frequency region of the EIS. The Rct value of NIP is 85.20 Ω, and the Rct values of MIP/AuNPs@MWCNTs/GQDs/GCE before and after elution are 16.21 and 62.46 Ω, respectively. These phenomena and results coincided with the results of CVs.
Study on the electrochemical behavior of MC-LR on the sensor
Determination of electrode reaction types
(a) CV curves of GCE and MIP in PBS (pH 6.0) containing 10−6 M MC-LR or L-arginine; (b) CV curves of MIP sensor in PBS (pH 6.0) containing 10−6 M MC-LR at sweep rates of 25–150 mV/s; (c) linear plots of peak current (Ip) and υ; (d) linear plots of peak current (Ip) and υ1/2; (e) linear plots of log Ip and log υ; (f) linear plots of Ep and lnv; (g) CV curves of MIP/AuNPs@MWCNTs/GQDs/GCE in different pH PBS containing 10−6 M MC-LR; (H) linear plots of Ep1 and Ep2 vs. pH.
(a) CV curves of GCE and MIP in PBS (pH 6.0) containing 10−6 M MC-LR or L-arginine; (b) CV curves of MIP sensor in PBS (pH 6.0) containing 10−6 M MC-LR at sweep rates of 25–150 mV/s; (c) linear plots of peak current (Ip) and υ; (d) linear plots of peak current (Ip) and υ1/2; (e) linear plots of log Ip and log υ; (f) linear plots of Ep and lnv; (g) CV curves of MIP/AuNPs@MWCNTs/GQDs/GCE in different pH PBS containing 10−6 M MC-LR; (H) linear plots of Ep1 and Ep2 vs. pH.
The response mechanism of MC-LR on the sensor
Optimization of experimental parameters
Some important fabrication parameters affecting the fabrication process were optimized such as the molar ratio of template to monomer, the concentration of MWCNTs/GQDs, the AuNPs loading, and the thickness of the MIP. As shown in Supplementary Fig. S1(A), the optimum molar ratio of L-arginine to p-ATP and MAA is 1:1:2, which is consistent with the optimized configuration visualized by the quantum chemical calculation as shown in Figure 2(a). As shown in Supplementary Fig. S1(B, C, and D), the other optimum conditions for the fabrication process are 50 μL of MWCNTs/GQDs, 10 scan cycles in the AuNP electrodeposition step, and 15 scan cycles in the MIP electro-polymerization step.
The analytical procedure steps including the elution time, the rebinding time, and the pH of the PBS were also optimized. As shown in Supplementary Fig. S2, the optimum elution time is 15 min, the optimum rebinding time is 20 min, and the optimum pH value is 7.0.
Performance analysis of MIP/AuNPs@MWCNTs/GQDs/GCE
Linearity range and limit of detection
(A) DPV of the MIP/AuNPs@MWCNTs/GQDs/GCE toward different concentrations of MC-LR (a → g correspond to MC-LR concentrations of (a) 0.08, (b) 0.1, (c) 0.2, (d) 0.4, (e) 0.8, (f) 1, (g) 2 μg/L, respectively; (B) the linear relationship between the peak currents and the logarithm of concentrations.
(A) DPV of the MIP/AuNPs@MWCNTs/GQDs/GCE toward different concentrations of MC-LR (a → g correspond to MC-LR concentrations of (a) 0.08, (b) 0.1, (c) 0.2, (d) 0.4, (e) 0.8, (f) 1, (g) 2 μg/L, respectively; (B) the linear relationship between the peak currents and the logarithm of concentrations.
Comparison of analytical methods or products for the analysis of MC-LR in water samples
Method . | Linear range (μg/L) . | LOD (μg/L) . | References . |
---|---|---|---|
Colorimetric sensor | 1–100 | – | Hu et al. (2009) |
Fluoroimmunoassay | 0–500 | 0.93 | Liu et al. (2012) |
Optical planar waveguide platform | 0.21–5.9 | 0.19 | Murphy et al. (2015) |
MIP-coated CQDs fluorescent sensor | 1–1,000 | 0.0093 | Qi et al. (2021) |
QCM sensor | 0.0995–995 | 0.0398 | He et al. (2015) |
MIP photoelectrochemical sensor | 0.5–100 | 0.1 | Chen et al. (2012) |
Sol–gel imprinted polymers on solid contact electrodes | 0.77–2.00 | 0.75 | Queirós et al. (2012) |
Molecularly imprinted electrochemical sensor | 0.08–2 | 0.0027 | Our work |
Method . | Linear range (μg/L) . | LOD (μg/L) . | References . |
---|---|---|---|
Colorimetric sensor | 1–100 | – | Hu et al. (2009) |
Fluoroimmunoassay | 0–500 | 0.93 | Liu et al. (2012) |
Optical planar waveguide platform | 0.21–5.9 | 0.19 | Murphy et al. (2015) |
MIP-coated CQDs fluorescent sensor | 1–1,000 | 0.0093 | Qi et al. (2021) |
QCM sensor | 0.0995–995 | 0.0398 | He et al. (2015) |
MIP photoelectrochemical sensor | 0.5–100 | 0.1 | Chen et al. (2012) |
Sol–gel imprinted polymers on solid contact electrodes | 0.77–2.00 | 0.75 | Queirós et al. (2012) |
Molecularly imprinted electrochemical sensor | 0.08–2 | 0.0027 | Our work |
Reproducibility and stability of MIP/AuNPs@MWCNTs/GQDs/GCE
The relative standard deviation (RSD) of the measurements of the six MIP/AuNPs@MWCNTs/GQDs/GCE prepared in the same concentration of MC-LR solution under the same conditions was 6.30%, and the RSD was 3.18% for six parallel measurements with the same sensor (Supplementary Fig. S3). This indicated that the sensor has good reproducibility. The same sensor was stored in PBS at 4 °C for 2 weeks and the peak current was 91.20% of the initial peak current when the same concentration of MC-LR was assayed (Supplementary Fig. S4). This indicated the long-term stability of the sensor prepared by this method.
Actual sample analysis
To further verify the practicability of the sensor, we not only investigated the recovery of MC-LR in tap water samples but also compared the detection results of the proposed sensor with the HPLC-MS method for the detection of total concentrations of MC-LR and [Dha7] MC-LR in actual water samples. As shown in Table S3, the recovery studies of tap water spiked samples have resulted in the range of 89.0–103.6%. Table 2 shows that the detection recoveries of MC-LR and [Dha7] MC-LR by MIP/AuNPs@MWCNTs/GQDs/GCE were in the range from 91.4 to 116.7%. It was shown that the prepared sensor could detect the total amount of MC-LR and its demethylated derivatives, which indicated that the prepared sensor had large environmental detection potential in practical application.
Recovery of studies of MC-LR and [Dha7] MC-LR from samples determined by the proposed sensor and HPLC-MS
Sample . | HPLC-MS (μg/L) . | MIP/AuNPs@MWCNTs/GQDs/GCE (μg/L) . | Recovery (%) . | ||
---|---|---|---|---|---|
MC-LR . | [Dha7] MC-LR . | Total . | |||
Sample 1 | 0.54 | 0.63 | 1.17 | 1.07 | 91.4 |
Sample 2 | 0.11 | 0.13 | 0.23 | 0.25 | 108.7 |
Sample 3 | 0.05 | 0.06 | 0.12 | 0.14 | 116.7 |
Sample . | HPLC-MS (μg/L) . | MIP/AuNPs@MWCNTs/GQDs/GCE (μg/L) . | Recovery (%) . | ||
---|---|---|---|---|---|
MC-LR . | [Dha7] MC-LR . | Total . | |||
Sample 1 | 0.54 | 0.63 | 1.17 | 1.07 | 91.4 |
Sample 2 | 0.11 | 0.13 | 0.23 | 0.25 | 108.7 |
Sample 3 | 0.05 | 0.06 | 0.12 | 0.14 | 116.7 |
CONCLUSION
In this study, a novel molecularly imprinted electrochemical sensor was proposed for the detection of MC-LR. MWCNTs, GQDs, and AuNPs were used as sensor modification materials to construct a three-dimensional conductive network, which enhanced the sensitivity of the sensor. The synergy between the selected dual functional monomers enriched the recognition sites of the sensor and enhanced the affinity for the targets, which improved the selectivity of sensor. Employment of the segment template ensures the sensor's recognition characteristic of target and cuts down the cost. The recognition reaction of MC-LR on the developed sensor is an irreversible electrochemical oxidation reaction involving an electron and two protons, which is controlled by a mixed adsorption − diffusion mechanism. In addition, the developed sensor could detect the total amount of MC-LR and its demethylated derivatives ([Dha7] MC-LR) in actual samples, and the results could be corrected by the HPLC-MS method. Therefore, this approach can be utilized in the detection of MC-LR in real environment monitoring for further development.
ACKNOWLEDGEMENTS
This work was financially supported by the National Key R&D Program of China (2019YFC0312602), the Open Project Program for Fujian Universities and Colleges Engineering Research Center of Modern Facility Agriculture, Fujian Polytechnic Normal University (G2-KF2008).
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.
REFERENCES
Author notes
These two authors contributed equally to this paper.