A Novel Paper-Based Electrochemical Biosensor Based on N,O-Rich Covalent Organic Frameworks for Carbaryl Detection

A new N,O-rich covalent organic framework (COFDHNDA-BTH) was synthesized by an amine-aldehyde condensation reaction between 2,6-dialdehyde-1,5-dihydroxynaphthalene (DHNDA) and 1,3,5-phenyltriformylhydrazine (BTH) for carbaryl detection. The free NH, OH, and C=O groups of COFDHNDA-BTH not only covalently couples with acetylcholinesterase (AChE) into the pores of COFDHNDA-BTH, but also greatly improves the catalytic activity of AChE in the constrained environment of COFDHNDA-BTH’s pore. Under the catalysis of AChE, the acetylthiocholine (ATCl) was decomposed into positively charged thiocholine (TCl), which was captured on the COFDHNDA-BTH modified electrode. The positive charges of TCl can attract anionic probe [Fe(CN)6]3−/4− on the COFDHNDA-BTH-modified electrode to show a good oxidation peak at 0.25 V (versus a saturated calomel electrode). The carbaryl detection can inhibit the activity of AChE, resulting in the decrease in the oxidation peak. Therefore, a turn-off electrochemical carbaryl biosensor based on a flexible carbon paper electrode loaded with COFDHNDA-BTH and AChE was constructed using the oxidation peak of an anionic probe [Fe(CN)6]3−/4− as the detection signal. The detection limit was 0.16 μM (S/N = 3), and the linear range was 0.48~35.0 μM. The sensor has good selectivity, repeatability, and stability, and has a good application prospect in pesticide detection.


Introduction
Pesticide residues and their degradation products are the main pollutants that pose great potential risks to the ecosystem and human health. They can destroy the ecological balance in an ecosystem and the main systems of the human body, such as the immune, nerve, and endocrine systems, and cause various diseases [1][2][3][4]. Carbaryl is one of the most common carbamate insecticides, named 1-naphthalene carbamate. Carbaryl enters the ecological food chain and endangers human health due to the biological accumulation and soil accumulation of carbaryl [5][6][7][8]. Moreover, carbaryl interferes with the hydrolysis of neurotransmitter acetylcholine (ATCl) and affects the normal secretion of the neurotransmitter. Therefore, it is very important to rapidly, reliably, and quantitatively detect carbaryl. At present, the commonly used methods of detecting carbaryl include chromatography [9], spectrophotometry [10], and immunoassay [11]. These methods are sensitive and reliable. However, their instruments are expensive, the sample pretreatments are cumbersome and time-consuming, and they require a trained person for operation and complex sample pretreatment. Electrochemical sensors are rapid, simple, sensitive, reliable, and low cost, which make up for the shortcomings of the above technologies and meet the requirements of portable detection [12][13][14][15][16][17][18][19][20]. Electrochemical sensors have broad development space and application prospects in the rapid detection of pesticide residues [21,22].
At present, electrochemical biosensors based on acetylcholinesterase (AChE) have attracted extensive attention in the detection of carbamate pesticides such as carbaryl Flexible carbon papers were purchased from Beichuan Co., Ltd. (Shanghai, China). White nail polish was purchased from Clio Co., Ltd. (Guangzhou, China).

Instruments
Transmission electron microscope (TEM) images were obtained with a JEM-2010 (HR) (Japan Electronics Co., Ltd., Tokyo, Japan). Scanning electron microscope (SEM) images were obtained with a HITACHI S-3400N SEM at an acceleration voltage of 15 KV (Hitachi Limited, Tokyo, Japan). Fourier transform infrared spectroscopy (FTIR) analysis was performed using a Perkin-Elmer Spectro100 from Perkin-Elmer Corporation (Perkin-Elmer, Waltham, MA, USA). Brunauer-Emmett-Teller (BET) surface areas and pore size distributions were measured using N 2 adsorption/desorption isotherms, which were tested with an Autosorb-iQ under 77 K (Quantachrome, Boynton Beach, FL, USA). X-ray diffraction (XRD) analysis was performed on a BRUCKER D8 ADVANCE X-ray powder diffractometer with scanning steps of 3 • /min in a scanning range of 0-50 • (Rigaku Corporation, Tokyo, Japan). All electrochemical characterization was performed with a CHI 760D (Shanghai, China).

Preparation of COF DHNDA-BTH
We dissolved 50 mg (0.2 mM) of BTH and 42 mg (0.3 mM) of DHNDA in 5 mL of mesitylene by ultrasound. Then, the solution was transferred to a reaction kettle with 3 mL of 6 M AcOH, and the solution was heated to 120 • C for 72 h. The obtained solid powder was washed with DMF, acetone, methylene chloride, and anhydrous ethanol, separately. Finally, the powder was dried in the oven at 60 • C to obtain COF DHNDA-BTH . The prepared procedures are shown in Scheme 1. Na2HPO4, NaH2PO4, ascorbic acid (AA), glucose, KCl, NaNO3, CuCl2, HgSO4, CdCl2, and PbCl2 were purchased from Inokai Co., Ltd. (Beijing, China). Carbaryl was purchased from Jiangxi Mackenech Technology Co., Ltd. (Nanchang, China). All solutions are prepared using ultra-pure water purified by a Millipore-Q purification system (18.2 MΩ cm −1 ). Flexible carbon papers were purchased from Beichuan Co. Ltd. (Shanghai, China). White nail polish was purchased from Clio Co. Ltd. (Guangzhou, China).

Instruments
Transmission electron microscope (TEM) images were obtained with a JEM-2010 (HR) (Japan Electronics Co., Ltd., Tokyo, Japan). Scanning electron microscope (SEM) images were obtained with a HITACHI S-3400N SEM at an acceleration voltage of 15 KV (Hitachi Limited, Tokyo, Japan). Fourier transform infrared spectroscopy (FTIR) analysis was performed using a Perkin-Elmer Spectro100 from Perkin-Elmer Corporation (Perkin-Elmer, Waltham, MA, USA). Brunauer-Emmett-Teller (BET) surface areas and pore size distributions were measured using N2 adsorption/desorption isotherms, which were tested with an Autosorb-iQ under 77 K (Quantachrome, Boynton Beach, Florida, USA). Xray diffraction (XRD) analysis was performed on a BRUCKER D8 ADVANCE X-ray powder diffractometer with scanning steps of 3°/min in a scanning range of 0-50° (Rigaku Corporation, Tokyo, Japan). All electrochemical characterization was performed with a CHI 760D (Shanghai, China).

Preparation of COFDHNDA-BTH
We dissolved 50 mg (0.2 mM) of BTH and 42 mg (0.3 mM) of DHNDA in 5 mL of mesitylene by ultrasound. Then, the solution was transferred to a reaction kettle with 3 mL of 6 M AcOH, and the solution was heated to 120 °C for 72 h. The obtained solid powder was washed with DMF, acetone, methylene chloride, and anhydrous ethanol, separately. Finally, the powder was dried in the oven at 60 °C to obtain COFDHNDA-BTH. The prepared procedures are shown in Scheme 1.

Preparation of AChE/COFDHNDA-BTH/Glassy Carbon Electrode (GCE)
Firstly, the surface of GCE was polished with Al2O3 powder and washed with ethanol and ultra-pure water until the surface of GCE was smooth and clean. Then, 2 mg was Scheme 1. Schematic diagram of preparation process of COF DHNDA-BTH and detection principle of electrochemical sensor based on AChE/COF DHNDA-BTH /GCE.

Preparation of AChE/COF DHNDA-BTH /Glassy Carbon Electrode (GCE)
Firstly, the surface of GCE was polished with Al 2 O 3 powder and washed with ethanol and ultra-pure water until the surface of GCE was smooth and clean. Then, 2 mg was dispersed into 1 mL of ultra-pure water to obtain 2 mg/mL of aqueous suspension. Next, 10 µL of 2 mg/mL COF DHNDA-BTH was dropped on the polished GCE surface. After 5 h, the COF DHNDA-BTH /GCE was rinsed with ultra-pure water. After that, 4 µL of 0.2 mg/mL AChE was dropped on the surface of the COF DHNDA-BTH /GCE. It was stored in a refrigerator at 4 • C. Finally, the electrode was rinsed with ultra-pure water to obtain AChE/COF DHNDA-BTH /GCE. The subsequent detection mechanism is shown in Scheme 1.

Preparation of Paper-Based Electrode
To prepare a portable AChE biosensor, a portable paper-based electrode was constructed as follows: First, the commercial carbon paper was cut into three long strips 3 mm in width and 3 cm in length. Next, both sides of a piece of white cardboard, 2 cm in width and 3 cm in length, were painted with white nail polish. Next, the three long carbon paper strips were pasted onto one side of the white cardboard in intervals of about 0.5 cm. The middle sections of the long carbon paper strips (about one-third of the long strips) were painted with white nail polish. The bottom section of long carbon paper strips was peeled off with acrylic transparent tape to obtain a new surface with graphene-like foam as a working electrode (WE), reference electrode (RE), and counter electrode (CE). The active areas of WE, RE, and CE were all 1 × 3 mm. The prepared procedures and a picture of the obtained paper-based electrode are shown in Scheme 2.
Biosensors 2022, 12, x FOR PEER REVIEW 4 of 13 dispersed into 1 mL of ultra-pure water to obtain 2 mg/mL of aqueous suspension. Next, 10 μL of 2 mg/mL COFDHNDA-BTH was dropped on the polished GCE surface. After 5 h, the COFDHNDA-BTH/GCE was rinsed with ultra-pure water. After that, 4 μL of 0.2 mg/mL AChE was dropped on the surface of the COFDHNDA-BTH/GCE. It was stored in a refrigerator at 4 °C. Finally, the electrode was rinsed with ultra-pure water to obtain AChE/COFDHNDA-BTH/GCE. The subsequent detection mechanism is shown in Scheme 1.

Preparation of Paper-Based Electrode
To prepare a portable AChE biosensor, a portable paper-based electrode was constructed as follows: First, the commercial carbon paper was cut into three long strips 3 mm in width and 3 cm in length. Next, both sides of a piece of white cardboard, 2 cm in width and 3 cm in length, were painted with white nail polish. Next, the three long carbon paper strips were pasted onto one side of the white cardboard in intervals of about 0.5 cm. The middle sections of the long carbon paper strips (about one-third of the long strips) were painted with white nail polish. The bottom section of long carbon paper strips was peeled off with acrylic transparent tape to obtain a new surface with graphene-like foam as a working electrode (WE), reference electrode (RE), and counter electrode (CE). The active areas of WE, RE, and CE were all 1 × 3 mm. The prepared procedures and a picture of the obtained paper-based electrode are shown in Scheme 2. Scheme 2. Preparation process of paper-based electrode and the picture of obtained paper-based electrode.

Preparation of AChE/COFDHNDA-BTH/Portable Paper-Based Electrode
The preparation procedures of the AChE/COFDHNDA-BTH/portable paper-based electrode were similar to that of AChE/COFDHNDA-BTH/GCE. Briefly, 10 μL of 2 mg/mL COFDHNDA-BTH was dropped on the surface of WE on portable paper-based electrode. Then, 4 μL of 0.2 mg/mL AChE was dropped on the COFDHNDA-BTH/portable paper-based electrode to obtain an AChE/COFDHNDA-BTH/portable paper-based electrode.

Characterization of COFDHNDA-BTH
The SEM ( Figure 1a) and TEM (Figure 1b) images showed that the as-prepared COFDHNDA-BTH presented a nanoflower-like morphology that was composited of many twodimensional (2D) nanoribbons with good flexibility based on π-π stacking. As shown in Scheme 1, these nanoribbons were formed by the amine-aldehyde condensation reaction between BTH and DHNDA. With good flexibility, the thickness of nearly a single molecule, as well as free NH, OH, and C=O groups, the COFDHNDA-BTH can better load enzymes and assemble on the electrode surface, which is good for constructing biosensors based on COFDHNDA-BTH. The FTIR spectrum (Figure 1c) showed the O=C-H peaks of Scheme 2. Preparation process of paper-based electrode and the picture of obtained paper-based electrode.

Preparation of AChE/COF DHNDA-BTH /Portable Paper-Based Electrode
The preparation procedures of the AChE/COF DHNDA-BTH /portable paper-based electrode were similar to that of AChE/COF DHNDA-BTH /GCE. Briefly, 10 µL of 2 mg/mL COF DHNDA-BTH was dropped on the surface of WE on portable paper-based electrode. Then, 4 µL of 0.2 mg/mL AChE was dropped on the COF DHNDA-BTH /portable paper-based electrode to obtain an AChE/COF DHNDA-BTH /portable paper-based electrode.

Characterization of COF DHNDA-BTH
The SEM ( Figure 1a) and TEM ( Figure 1b) images showed that the as-prepared COF DHNDA-BTH presented a nanoflower-like morphology that was composited of many two-dimensional (2D) nanoribbons with good flexibility based on π-π stacking. As shown in Scheme 1, these nanoribbons were formed by the amine-aldehyde condensation reaction between BTH and DHNDA. With good flexibility, the thickness of nearly a single molecule, as well as free NH, OH, and C=O groups, the COF DHNDA-BTH can better load enzymes and assemble on the electrode surface, which is good for constructing biosensors based on COF DHNDA-BTH . The FTIR spectrum (Figure 1c) showed the O=C-H peaks of DHNDA at 2744 and 2836 cm −1 , the C=O peak at 1692 cm −1 (green curve), and N-H peaks of BTH at 3058 and 3156 cm −1 (red curve). In the FTIR spectrum of COF DHNDA-BTH (black curve), a peak of -C=N-appeared at 1641 cm −1 , indicating the successful occurrence of an amine-aldehyde condensation reaction, which further proved the successful synthesis of COF DHNDA-BTH . The XRD patterns of COF DHNDA-BTH showed many peaks at 5.95 • , 12.27 • , 19.67 • , and 27.06 • , corresponding to (200), (121), (111), and (102) crystal planes, respectively, indicating that the as-prepared COF DHNDA-BTH had good crystallinity ( Figure 1d). The AA and AB model obtained by simulation were compared with the experimental result, in which the XRD pattern agreed with the simulated AA stacking pattern ( Figure S1, Supporting Information) and showed the layered structure with a layer distance of 3.58 Å ( Figures S2 and S3, Supporting Information). The N 2 adsorption/desorption isotherm of COF DHNDA-BTH showed that the BET specific surface area was 46.91 m 2 g −1 ( Figure S4, Supporting Information). The aperture distribution of COF DHNDA-BTH showed that the aperture was 1.379 nm ( Figure S5, Supporting Information).
DHNDA at 2744 and 2836 cm −1 , the C=O peak at 1692 cm −1 (green curve), and N-H peaks of BTH at 3058 and 3156 cm −1 (red curve). In the FTIR spectrum of COFDHNDA-BTH (black curve), a peak of -C=N-appeared at 1641 cm −1 , indicating the successful occurrence of an amine-aldehyde condensation reaction, which further proved the successful synthesis of COFDHNDA-BTH. The XRD patterns of COFDHNDA-BTH showed many peaks at 5.95°, 12.27°, 19.67°, and 27.06°, corresponding to (200), (121), (111), and (102) crystal planes, respectively, indicating that the as-prepared COFDHNDA-BTH had good crystallinity ( Figure  1d). The AA and AB model obtained by simulation were compared with the experimental result, in which the XRD pattern agreed with the simulated AA stacking pattern ( Figure  S1, Supporting Information) and showed the layered structure with a layer distance of 3.58 Å ( Figures S2 and S3, Supporting Information). The N2 adsorption/desorption isotherm of COFDHNDA-BTH showed that the BET specific surface area was 46.91 m 2 g −1 ( Figure S4, Supporting Information). The aperture distribution of COFDHNDA-BTH showed that the aperture was 1.379 nm ( Figure S5, Supporting Information).  Figure 2a,b show the morphology and structure of commercial carbon paper with high and low magnification, respectively. The figures reveal a rough surface with graphene-like foam, which was more conducive to loading active materials and to transferring electrons and electrolyte ions.  Figure 2a,b show the morphology and structure of commercial carbon paper with high and low magnification, respectively. The figures reveal a rough surface with graphene-like foam, which was more conducive to loading active materials and to transferring electrons and electrolyte ions.  significantly decreased, while the peak-to-peak potential difference increased. This might have occurred because the COFDHNDA-BTH material hindered the transmission of electrons. After the AChE was further loaded, the peak current of the [Fe (CN)6] 3−/4− further decreased, and the peak-to-peak potential difference became wider, because the electrical conductivity of the modified electrode was further reduced by the enzyme loading, which also indirectly proved the successful fixation of AChE on COFDHNDA-BTH/GCE. The atom force microscopy image also proved that the enzyme was modified on the COF ( Figure S6, Supporting Information). The charge transfer resistance (Rct) values of bare GCE, COFDHNDA-BTH/GCE, and AChE/ COFDHNDA-BTH/GCE were 40, 1293, and 2744 Ω, respectively (the error values of the resistance values were 6.22%, 3.9%, and 6.56%, respectively) ( Figure 3b). The result was consistent with that of CVs, further indicating the successful preparation of the AChE/COFDHNDA-BTH/GCE.

Electrochemical Detection of Carbaryl Based on AChE/COFDHNDA-BTH/GCE
Firstly, the loading amount of COFDHNDA-BTH on the AChE/COFDHNDA-BTH/GCE was optimized by dropping different volumes of 2 mg/mL COFDHNDA-BTH onto the GCE. As   6 ] 3−/4− significantly decreased, while the peak-to-peak potential difference increased. This might have occurred because the COF DHNDA-BTH material hindered the transmission of electrons. After the AChE was further loaded, the peak current of the [Fe(CN) 6 ] 3−/4− further decreased, and the peak-to-peak potential difference became wider, because the electrical conductivity of the modified electrode was further reduced by the enzyme loading, which also indirectly proved the successful fixation of AChE on COF DHNDA-BTH /GCE. The atom force microscopy image also proved that the enzyme was modified on the COF ( Figure S6, Supporting Information). The charge transfer resistance (Rct) values of bare GCE, COF DHNDA-BTH /GCE, and AChE/ COF DHNDA-BTH /GCE were 40, 1293, and 2744 Ω, respectively (the error values of the resistance values were 6.22%, 3.9%, and 6.56%, respectively) (Figure 3b). The result was consistent with that of CVs, further indicating the successful preparation of the AChE/COF DHNDA-BTH /GCE.

Electrochemical Detection of Carbaryl Based on AChE/COFDHNDA-BTH/GCE
Firstly, the loading amount of COFDHNDA-BTH on the AChE/COFDHNDA-BTH/GCE was optimized by dropping different volumes of 2 mg/mL COFDHNDA-BTH onto the GCE. As

Electrochemical Detection of Carbaryl Based on AChE/COF DHNDA-BTH /GCE
Firstly, the loading amount of COF DHNDA-BTH on the AChE/COF DHNDA-BTH /GCE was optimized by dropping different volumes of 2 mg/mL COF DHNDA-BTH onto the GCE. As shown in Figure 4a, with the amount of COF DHNDA-BTH dropped onto the COF DHNDA-BTH increased, its peak current also increased, because the more COF DHNDA-BTH was present, the more active loading sites were used to bind AChE. The maximum volume value was 7.5 µL. When the amount of COF DHNDA-BTH further increased, the peak current gradually decreased because too much COF DHNDA-BTH was loaded on the electrode, which led to an accumulation on the electrode surface, which hindered the conductivity of the electrode. As shown in Figure 4b, with the increase in soaking time, the inhibition rate gradually increased and the inhibition rate reached the maximum value after 12 min, and then remained stable. Therefore, 12 min was selected as the final incubation time. In the experiment, the dosage of AChE and ATCl and their reaction times affected the detection performance. Therefore, the loaded amount of AChE on COF DHNDA-BTH /GCE was also optimized, as shown in Figure 4c. When the amount of AChE was less than 0.6 mM, the corresponding peak current density increased with the increase in AChE. The amount of AChE was greater than 0.6 mM, and the peak current density gradually decreased because excess AChE accumulated on the COF DHNDA-BTH /GCE surface to block electron transport. Therefore, 0.6 mM AChE was selected as the final drop amount. As shown in Figure 4d, the content of TCl was related to the amount of ATCl and, accordingly, the amount of ATCl also needs to be optimized. When the amount of ATCl was less than 0.6 mM, the current density increased with the increase in ATCl. When the amount of ATCl exceeded 0.6 mM, the current density slightly decreased. Therefore, 0.6 mM ATCl was selected as the final detection concentration. shown in Figure 4a, with the amount of COFDHNDA-BTH dropped onto the COFDHNDA-BTH increased, its peak current also increased, because the more COFDHNDA-BTH was present, the more active loading sites were used to bind AChE. The maximum volume value was 7.5 μL. When the amount of COFDHNDA-BTH further increased, the peak current gradually decreased because too much COFDHNDA-BTH was loaded on the electrode, which led to an accumulation on the electrode surface, which hindered the conductivity of the electrode. As shown in Figure 4b, with the increase in soaking time, the inhibition rate gradually increased and the inhibition rate reached the maximum value after 12 min, and then remained stable. Therefore, 12 min was selected as the final incubation time. In the experiment, the dosage of AChE and ATCl and their reaction times affected the detection performance. Therefore, the loaded amount of AChE on COFDHNDA-BTH/GCE was also optimized, as shown in Figure 4c. When the amount of AChE was less than 0.6 mM, the corresponding peak current density increased with the increase in AChE. The amount of AChE was greater than 0.6 mM, and the peak current density gradually decreased because excess AChE accumulated on the COFDHNDA-BTH/GCE surface to block electron transport. Therefore, 0.6 mM AChE was selected as the final drop amount. As shown in Figure 4d, the content of TCl was related to the amount of ATCl and, accordingly, the amount of ATCl also needs to be optimized. When the amount of ATCl was less than 0.6 mM, the current density increased with the increase in ATCl. When the amount of ATCl exceeded 0.6 mM, the current density slightly decreased. Therefore, 0.6 mM ATCl was selected as the final detection concentration.  Figure 5a, the peak current density gradually decreased with the increase in carbaryl concentration. This occurred because the higher the concentration of carbaryl, the more obvious the inhibition  Figure 5a, the peak current density gradually decreased with the increase in carbaryl concentration. This occurred because the higher the concentration of carbaryl, the more obvious the inhibition effect on AChE activity, resulting in a decrease in the amount of TCl produced; thus, the content of [Fe(CN) 6 ] 3−/4− attracted on the electrode surface also decreased. Figure 5b shows the linear relationship between the peak current density and the carbaryl concentration. The detection limit of the biosensor was 0.23 µM (S/N = 3), and the linear range was 0.69-35 µM. As shown in Figure 5c, with the increase in carbaryl concentration, due to the inhibiting effect of carbaryl on AChE, less TCl was produced, and the corresponding impedance value was larger. Figure 5d shows the linear fitting diagram of the corresponding impedance value and carbaryl concentration. The corresponding detection limit was 0.13 µM (S/N = 3), and the linear range was 0.39-35 µM. Finally, we compared the performance of our biosensor with that of other biosensors (Table 1), and the results showed that our biosensor had a wider linear range and better sensitivity. Its good performance might be ascribed to the good catalytic activity of AChE in the constrained environment of COF DHNDA-BTH 's pore.  Figure 5b sh the linear relationship between the peak current density and the carbaryl concentra The detection limit of the biosensor was 0.23 μM (S/N = 3), and the linear range was 35 μM. As shown in Figure 5c, with the increase in carbaryl concentration, due t inhibiting effect of carbaryl on AChE, less TCl was produced, and the correspon impedance value was larger. Figure 5d shows the linear fitting diagram of corresponding impedance value and carbaryl concentration. The corresponding dete limit was 0.13 μM (S/N = 3), and the linear range was 0.39-35 μM. Finally, we comp the performance of our biosensor with that of other biosensors (Table 1), and the re showed that our biosensor had a wider linear range and better sensitivity. Its performance might be ascribed to the good catalytic activity of AChE in the constra environment of COFDHNDA-BTH's pore.   Finally, the current response of the AChE/COF DHNDA-BTH /portable paper-based electrode was measured with the accumulation of carbaryl concentration. Figure 6a shows the results of the peak current response increase with the increase in carbaryl concentration, and the linear range was 0.48~35 µM. Figure 6b shows the corresponding linear relationship between peak current density and carbaryl concentration. The linear equation was j p = -0.031c +3.342 (R 2 = 0.98), where j p and c are peak current density and carbaryl concentration, respectively. The detection limit of the paper-based electrochemical biosensor was 0.16 µM (S/N = 3). The results showed that the AChE/COF DHNDA-BTH /portable paper-based electrode biosensor has good performance. In general, compared with the three-electrode system composed of glassy carbon electrodes, the experimental results of paper-based electrodes were similar, which proves that the preparation of paper-based electrodes was successful. Finally, the current response of the AChE/COFDHNDA-BTH/portable paper-b electrode was measured with the accumulation of carbaryl concentration. Figure 6a s the results of the peak current response increase with the increase in car concentration, and the linear range was 0.48~35 μM. Figure 6b shows the correspon linear relationship between peak current density and carbaryl concentration. The l equation was jp = -0.031c +3.342 (R 2 = 0.98), where jp and c are peak current density carbaryl concentration, respectively. The detection limit of the paper-b electrochemical biosensor was 0.16 μM (S/N = 3). The results showed tha AChE/COFDHNDA-BTH/portable paper-based electrode biosensor has good performan general, compared with the three-electrode system composed of glassy carbon electr the experimental results of paper-based electrodes were similar, which proves tha preparation of paper-based electrodes was successful.

Selectivity, Stability, and Repeatability of Electrochemical Carbaryl Biosensors
As shown in Figure 7a, the selectivity of the AChE/COFDHNDA-BTH/GCE biosenso studied with an interfering substance concentration five times the concentratio carbaryl. It was observed that the interfering substances including AA, glucose, Na Cu 2+ , Hg 2+ , Cd 2+ , and Pb 2+ did not produce obvious interference. The good selectiv the AChE/COFDHNDA-BTH/GCE biosensor could be ascribed to the low oxidation potential of [Fe(CN)6] 3−/4− . Although the oxidation potential of AA was lower (abou V versus that of a saturated calomel electrode), the oxidation peak of AA was obvi different from that of [Fe(CN)6] 3−/4− at 0.25 V. Therefore, the AA showed ob interference for the detection of carbaryl. The electrochemical carbaryl biosensor stored in a refrigerator at 4 °C, and the peak current change (Figure 7b) at 0, 5, 10, 1 25, and 30 days was tested. It was observed that the peak current still maintained 9 of the initial value after 30 days. Next, six AChE/COFDHNDA-BTH/GCE were prepared fo detection of 5 μM carbaryl (Figure 7c), and the relative standard deviations (RSD) o final result was only 3.4%. In conclusion, the electrochemical carbaryl biosensor has selectivity, repeatability, and stability, and has a good application prospects fo detection of carbaryl.

Selectivity, Stability, and Repeatability of Electrochemical Carbaryl Biosensors
As shown in Figure 7a, the selectivity of the AChE/COF DHNDA-BTH /GCE biosensor was studied with an interfering substance concentration five times the concentration of carbaryl. It was observed that the interfering substances including AA, glucose, NaNO 3 , Cu 2+ , Hg 2+ , Cd 2+ , and Pb 2+ did not produce obvious interference. The good selectivity of the AChE/COF DHNDA-BTH /GCE biosensor could be ascribed to the low oxidation peak potential of [Fe(CN)6] 3−/4− . Although the oxidation potential of AA was lower (about 0.15 V versus that of a saturated calomel electrode), the oxidation peak of AA was obviously different from that of [Fe(CN)6] 3−/4− at 0.25 V. Therefore, the AA showed obvious interference for the detection of carbaryl. The electrochemical carbaryl biosensor was stored in a refrigerator at 4 • C, and the peak current change (Figure 7b) at 0, 5,10,15,20,25, and 30 days was tested. It was observed that the peak current still maintained 93.3% of the initial value after 30 days. Next, six AChE/COF DHNDA-BTH /GCE were prepared for the detection of 5 µM carbaryl (Figure 7c), and the relative standard deviations (RSD) of the final result was only 3.4%. In conclusion, the electrochemical carbaryl biosensor has good selectivity, repeatability, and stability, and has a good application prospects for the detection of carbaryl.

Actual Sample Testing
The content of carbaryl in lettuce was determined through a spiked experiment. The specific procedures were as follows: We chopped 100 g of lettuce (lettuce was purchased at Changsheng Market in Nanchang, Jiangxi province, on 21 September 2021). We obtained 50 μL of lettuce juice by flittering, which we then put into 5 mL of 0.1 mM KCl solution containing 5.0 mM [Fe (CN)6] 3−/4− . Then, different concentrations of carbaryl standard solution were added to the above actual samples, and the concentration of carbaryl was determined by the electrochemical biosensor. The obtained recovery rates and relative standard deviation (RSD) are shown in Table 2. It was found that the spiked recoveries of three different concentrations of carbaryl ranged from 94.4% to 106% with an RSD of less than 10%. These results indicated that the electrochemical carbaryl biosensor has the potential to detect pesticide residues in real samples.

Conclusions
A new COFDHNDA-BTH with abundant NH, OH, and C=O groups was synthesized. The COFDHNDA-BTH presented a nanoflower-like morphology, which was composited of many 2D nanoribbons. The COFDHNDA-BTH had high porosity, excellent flexibility, and good biocompatibility. Furthermore, a portable paper-based electrode with a surface of graphene-like foam was developed. The COFDHNDA-BTH could be firmly assembled on the portable paper-based electrode to load AchE, whose bioactivity was greatly enhanced in the confined COFDHNDA-BTH hole. The COFDHNDA-BTH could better capture TCl through hydrogen bonding. The positively charged TCl attracted the anion probe [Fe (CN)6] 3−/4− , which could detect pesticides at low potential (0.25 V). A turn-off electrochemical biosensor was constructed using the oxidation peak of the anionic probe [Fe (CN)6] 3−/4− as the detection signal. The linear range of the AChE/COFDHNDA-BTH/portable paper-based electrode was 0.48~35 μM, and the detection limit was 0.16 μM (S/N = 3). The paper-based electrode has broad application prospects.

Actual Sample Testing
The content of carbaryl in lettuce was determined through a spiked experiment. The specific procedures were as follows: We chopped 100 g of lettuce (lettuce was purchased at Changsheng Market in Nanchang, Jiangxi province, on 21 September 2021). We obtained 50 µL of lettuce juice by flittering, which we then put into 5 mL of 0.1 mM KCl solution containing 5.0 mM [Fe(CN) 6 ] 3−/4− . Then, different concentrations of carbaryl standard solution were added to the above actual samples, and the concentration of carbaryl was determined by the electrochemical biosensor. The obtained recovery rates and relative standard deviation (RSD) are shown in Table 2. It was found that the spiked recoveries of three different concentrations of carbaryl ranged from 94.4% to 106% with an RSD of less than 10%. These results indicated that the electrochemical carbaryl biosensor has the potential to detect pesticide residues in real samples.

Conclusions
A new COF DHNDA-BTH with abundant NH, OH, and C=O groups was synthesized. The COF DHNDA-BTH presented a nanoflower-like morphology, which was composited of many 2D nanoribbons. The COF DHNDA-BTH had high porosity, excellent flexibility, and good biocompatibility. Furthermore, a portable paper-based electrode with a surface of graphenelike foam was developed. The COF DHNDA-BTH could be firmly assembled on the portable paper-based electrode to load AchE, whose bioactivity was greatly enhanced in the confined COF DHNDA-BTH hole. The COF DHNDA-BTH could better capture TCl through hydrogen bonding. The positively charged TCl attracted the anion probe [Fe(CN) 6 ] 3−/4− , which could detect pesticides at low potential (0.25 V). A turn-off electrochemical biosensor was constructed using the oxidation peak of the anionic probe [Fe(CN) 6 ] 3−/4− as the detection signal. The linear range of the AChE/COF DHNDA-BTH /portable paper-based electrode was 0.48~35 µM, and the detection limit was 0.16 µM (S/N = 3). The paper-based electrode has broad application prospects.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/bios12100899/s1. Figure S1: Image of experimental and simulated XRD pattern for AA stacking and AB stacking structure of COF DHNDA-BTH . Figure S2: Image of pore size distribution of COF DHNDA-BTH . Figure S3. Image of π-π stacking distance of the layers of COF DHNDA-BTH . Figure S4: Image of BET pattern of COF DHNDA-BTH . Figure S5: Image of pore size distribution of COF DHNDA-BTH . Figure S6: Image of AFM of AChE/COF DHNDA-BTH .
Funding: This work was financially supported by the National Natural Science Foundation of China (21964010).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data is available under the request to the correspondence.