A High-Performance Self-Supporting Electrochemical Biosensor to Detect Aflatoxin B1

High-performance electrochemical biosensors for the rapid detection of aflatoxin B1 (AFB1) are urgently required in the food industry. Herein, a multi-scaled electrochemical biosensor was fabricated by assembling carboxylated polystyrene nanospheres, an aptamer and horseradish peroxidase into a free-standing carbon nanofiber/carbon felt support. The resulting electrochemical biosensor possessed an exceptional performance, owing to the unique structures as well as the synergistic effects of the components. The 3D porous carbon nanofiber/carbon felt support served as an ideal substrate, owing to the excellent conductivity and facile diffusion of the reactants. The integration of carboxylated polystyrene nanospheres with horseradish peroxidase was employed as a signal amplification probe to enhance the electrochemical responses via catalyzing the decomposition of hydrogen peroxide. With the aid of the aptamer, the prepared sensors could quantitatively detect AFB1 in wine and soy sauce samples via differential pulse voltammetry. The recovery rates of AFB1 in the samples were between 87.53% and 106.71%. The limit of detection of the biosensors was 0.016 pg mL−1. The electrochemical biosensors also had excellent sensitivity, reproducibility, specificity and stability. The synthetic strategy reported in this work could pave a new route to fabricate high-performance electrochemical biosensors for the detection of mycotoxins.


Introduction
As one of the secondary metabolites produced by Aspergillus flavus and parasiticus, aflatoxin B1 (AFB 1 ) is classified as a Group I carcinogen by the World Health Organization's Cancer Research Institute [1][2][3][4]. The tolerance of AFB 1 in food should be less than 20 ppb, as reported by the Food and Agriculture Organization of the United Nations [5,6]. Due to its dramatic poisoning effects on the human body, the rapid and accurate detection of AFB 1 is in high demand in food processing [7,8]. So far, various analytical methods such as high-performance liquid chromatography, thin-layer chromatography and enzymelinked immunosorbent assays have been developed to detect AFB 1 [9][10][11][12][13]. However, most of these methods involve costly equipment and complicated sample processing and suffer from several drawbacks such as a low efficiency and/or false-positive results, which greatly restrict their widespread use [14][15][16]. Alternatively, electrochemical sensors have drawn increasing attention in detecting mycotoxins, owing to their high efficiency and simplicity [15,[17][18][19]. For instance, Shi et al. developed a novel electrochemical immunosensor for AFB 1 detection based on Au nanoparticles supported by graphene, which showed a limit of detection (LOD) of 0.001 ng mL −1 [20]. In addition, Wang et al.

Preparation of the SSE
The SSE was prepared according to the method reported in our previous work [40]. First, the CF (3.0 × 3.0 × 0.5 cm) was thermally activated at 420 • C for 10 h using a tube furnace under an air atmosphere. A total of 1.11 g of oxalic acid was dissolved in 30 mL of ethanol and 1.7 g of nickel (II) nitrate hexahydrate and 0.7 g of copper (II) nitrate trihydrate was dissolved in 30 mL of ethanol. The metal precursor solution was titrated with the oxalic acid solution under magnetic stirring. Subsequently, the mixture was heated in an autoclave at 120 • C for 8 h and cooled to 25 • C afterward. The sample was washed with deionized water three times and then dried at 80 • C for 8 h. The powder sample was ultrasonically dispersed in ethanol to form a suspension with a concentration of 5 mg mL −1 . The activated CF was repeatedly immersed in the Ni-Cu oxalate suspension until the load was about 320~390 mg. The CF loaded with Ni-Cu oxalates was calcined at 600 • C for 2 h with a heating rate of 10 • C min −1 under a methane and nitrogen flow (volumetric ratio: 3:1).

Synthesis of the Gold-Modified SSE (Au/SSE)
The Au/SSE was prepared via a displacement reaction. The details of the experimental procedures were as follows: three pieces of SSE (1.5 × 0.6 × 0.25 cm) were immersed in 0.5 mM of HAuCl 4 at 50 • C for 24 h. After the reaction, the SSE was washed thoroughly with deionized water and then dried at 50 • C for 8 h.

Preparation of the BSA-Apt-Au/SSE
For the synthesis of Apt-Au/SSE, the PBS containing 1 µM Apt was introduced dropwise to the Au/SSE. After reacting for 10 min, the electrode was washed with deionized water to remove the residual solution. After that, 5 mg mL −1 of the BSA solution was added dropwise to the Apt-Au/SSE to passivate the non-specific active sites that were not bound to the Apt. The as-prepared BSA-Apt-Au/SSE was then washed three times and dried for further use.

Preparation of the Spiked Samples
A total of 5 g of wine and soy sauce was added to a methanol solution with a concentration of 0.2 g mL −1 . The mixed solution was ultrasonically dispersed and then centrifuged at 10,000 rpm for 10 min. The supernatant was then obtained for the preparation of the AFB 1 spiked samples. A solution of methanol and chloroform (v/v = 1:1) was used to dissolve AFB 1 .

Biosensor Fabrication
Carboxylated polystyrene (PS-COOH) nanospheres were prepared according to our previous work [28]. Before the biosensor fabrication, the PS-COOH was activated using a carboxyl activation solution that contained 20 mM of HEPES (pH = 6.8), 5 mM of EDC and 5 mM of NHS. The BSA-Apt-Au/SSE electrodes were immersed in the AFB 1 solution at room temperature for 2 h. The obtained electrode was denoted as AFB 1 -BSA-Apt-Au/SSE. The AFB 1 -BSA-Apt-Au/SSE was then immersed in the PS-COOH suspension for 30 min, thoroughly washed with deionized water and dried at room temperature. Finally, the modified electrode, denoted as PS-BSA-Apt-Au/SSE, was immediately immersed in 2 mg mL −1 of HRP at 4 • C for 30 min. The prepared electrodes are shown in Figure S1. The biosensor preparation process is shown in Scheme 1.

Structural Characterization and Electrochemical Analysis
For the electrochemical analysis, a CHI760E electrochemical workstation was used. The electrochemical measurements were carried out on a conventional three-electrode system. The biosensor, carbon rod and Ag/AgCl were used as the working, counter and reference electrodes, respectively. PBS containing 10 mM of H 2 O 2 was used as the electrolyte for the differential pulse voltammetry (DPV) measurements. The DPV curves were recorded at potentials ranging from −0.7 to −0.2 V vs. Ag/AgCl with a scan rate of 1 mV s −1 . The peak current (I p ) of H 2 O 2 electro-oxidation was recorded. Amperometric i-t curves were recorded at a potential of −0.35 V. The PBS containing the 10 mM H 2 O 2 solution was used as the electrolyte. Cyclic voltammetry (CV) curves were recorded at potentials of −0.2~+0.6 V with a scan rate of 10 mV s −1 . All the electrochemical measurements were conducted at an ambient temperature and pressure. All the potentials reported in this work were referenced to Ag/AgCl. The morphology of the samples was observed by field emission electron scanning microscopy (FESEM).

Structural Characterization and Electrochemical Analysis
For the electrochemical analysis, a CHI760E electrochemical workstation was used. The electrochemical measurements were carried out on a conventional three-electrode system. The biosensor, carbon rod and Ag/AgCl were used as the working, counter and reference electrodes, respectively. PBS containing 10 mM of H2O2 was used as the electrolyte for the differential pulse voltammetry (DPV) measurements. The DPV curves were recorded at potentials ranging from −0.7 to −0.2 V vs. Ag/AgCl with a scan rate of 1 mV s −1 . The peak current (Ip) of H2O2 electro-oxidation was recorded. Amperometric i-t curves were recorded at a potential of −0.35 V. The PBS containing the 10 mM H2O2 solution was used as the electrolyte. Cyclic voltammetry (CV) curves were recorded at potentials of −0.2 ~ +0.6 V with a scan rate of 10 mV s −1 . All the electrochemical measurements were conducted at an ambient temperature and pressure. All the potentials reported in this work were referenced to Ag/AgCl. The morphology of the samples was observed by field emission electron scanning microscopy (FESEM).

Structural and Electrochemical Characterization
Before the electrode fabrication, the pristine CF was activated via a thermal treatment to improve its hydrophilicity. The improved hydrophilicity of the CF after the thermal treatment could be verified from the hydrophilic experiment shown in Figure 1a,b. The water droplet could quickly permeate into the activated CF whereas it remained on the surface of the pristine CF. In addition, the activated CF could quickly sink to the bottom whereas the pristine CF floated on the water. The CVs of the CF before and after the thermal treatment are shown in Figure 1c. The activated CF showed an enhanced current intensity compared with that of the pristine CF, which may have been due to the better hydrophilicity of the activated CF.

Scheme 1.
Stepwise illustration of fabrication procedure of the electrochemical biosensor based on the self-supported electrode.

Structural and Electrochemical Characterization
Before the electrode fabrication, the pristine CF was activated via a thermal treatment to improve its hydrophilicity. The improved hydrophilicity of the CF after the thermal treatment could be verified from the hydrophilic experiment shown in Figure 1a,b. The water droplet could quickly permeate into the activated CF whereas it remained on the surface of the pristine CF. In addition, the activated CF could quickly sink to the bottom whereas the pristine CF floated on the water. The CVs of the CF before and after the thermal treatment are shown in Figure 1c. The activated CF showed an enhanced current intensity compared with that of the pristine CF, which may have been due to the better hydrophilicity of the activated CF.  Figure S2a,b are the FESEM micrographs of the pristine CF, showing that the pristine CF was of a porous structure consisting of intricately interlacing carbon fibers. The diameter of the carbon fiber was about 10 µ m. Figure 2a,b are the images of the CF loaded with Ni-Cu oxalates. It could be seen that the Ni-Cu oxalates were evenly distributed on the  Figure S2a,b are the FESEM micrographs of the pristine CF, showing that the pristine CF was of a porous structure consisting of intricately interlacing carbon fibers. The diameter of the carbon fiber was about 10 µm. Figure 2a,b are the images of the CF loaded with Ni-Cu oxalates. It could be seen that the Ni-Cu oxalates were evenly distributed on the CF. Figures S3 and 2a,b show that the SSE electrode maintained a well-defined hierarchical structure after the in situ growth of the carbon nanofibers. After the catalytic methane decomposition reaction, many dense carbon nanotubes were present on the surface of the CF. Notably, the Ni-Cu nanoparticles were deposited on the tip of the nanotubes. The macropores of the original CF still existed, regardless of the growth of the carbon nanofibers. The FESEM image of the as-prepared PS-COOH is shown in Figure S3b. The PS-COOH showed a uniform spherical structure with an average diameter of ca. 100 nm.

Fabrication of the Electrochemical Biosensor
Scheme 1 illustrates the procedures for the fabrication of the electrochemical biosensor. The electrochemical biosensor was fabricated based on the Au/SSE electrode. The Au/SSE electrode had a porous structure with a large surface area, which served as an ideal platform for the electrochemical sensors, owing to the facile substance transfer. In addition, it had a three-dimensional structure that could improve the performance of the biosensor. The fabrication procedures of the electrochemical biosensor were as follows. First, Au atoms were introduced into the Ni-Cu nanoparticles by replacement reactions. The electrodes before and after the replacement reaction were characterized by CV in the 0.1 M NaOH electrolyte; the results are shown in Figure 3a. For the SSE electrode, the characteristic peaks at +0.1 V and +0.63 V may have been related to the oxidation of Cu and Ni, respectively [41][42][43][44]. For the Au/SSE electrode, the peaks related to Ni and Cu oxidation became less distinctive because Au partially replaced the Ni/Cu atoms on the surface of the nanoparticles. The peaks at +0.7 V and −0.1 V could be ascribed to the Au redox, which suggested that Au had been successfully introduced into the electrode [45,46].

Fabrication of the Electrochemical Biosensor
Scheme 1 illustrates the procedures for the fabrication of the electrochemical biosensor. The electrochemical biosensor was fabricated based on the Au/SSE electrode. The Au/SSE electrode had a porous structure with a large surface area, which served as an ideal platform for the electrochemical sensors, owing to the facile substance transfer. In addition, it had a three-dimensional structure that could improve the performance of the biosensor. The fabrication procedures of the electrochemical biosensor were as follows. First, Au atoms were introduced into the Ni-Cu nanoparticles by replacement reactions. The functionalized Apt was then immobilized on the electrode through the Au-S bonds. Before the AFB 1 addition, the non-specific active sites were passivated by BSA. The Apt on the electrode specifically bound to AFB 1 through van der Waals forces, the hydrogen bond and/or hydrophobic forces. The free Apt bound to the activated PS-COOH through the amido bond. As there were many activated carboxyl groups in the PS-COOH, HRP with amino groups could combine with them through amido bonds. Meanwhile, the iron porphyrin ring in the HRP catalyzed the redox of H 2 O 2 in the electrolyte, which served as the basis for the quantitative detection of AFB 1 via the electrochemical signal of H 2 O 2 .

Fabrication of the Electrochemical Biosensor
Scheme 1 illustrates the procedures for the fabrication of the electrochemical biosensor. The electrochemical biosensor was fabricated based on the Au/SSE electrode. The Au/SSE electrode had a porous structure with a large surface area, which served as an ideal platform for the electrochemical sensors, owing to the facile substance transfer. In addition, it had a three-dimensional structure that could improve the performance of the biosensor. The fabrication procedures of the electrochemical biosensor were as follows. First, Au atoms were introduced into the Ni-Cu nanoparticles by replacement reactions. The functionalized Apt was then immobilized on the electrode through the Au-S bonds. Before the AFB1 addition, the non-specific active sites were passivated by BSA. The Apt on the electrode specifically bound to AFB1 through van der Waals forces, the hydrogen bond and/or hydrophobic forces. The free Apt bound to the activated PS-COOH through the amido bond. As there were many activated carboxyl groups in the PS-COOH, HRP with amino groups could combine with them through amido bonds. Meanwhile, the iron porphyrin ring in the HRP catalyzed the redox of H2O2 in the electrolyte, which served as the basis for the quantitative detection of AFB1 via the electrochemical signal of H2O2.   [47][48][49]. After Apt binding, the I p value decreased and the peak potential positively shifted, which could be ascribed to the fact that the Apt as a bio-macromolecule does not have electrochemically active sites. This result also confirmed that the Apt was successfully assembled on the electrode. After passivating the non-specific active sites on the electrode surface with BSA, most of the active sites of the electrode material were covered by BSA and Apt, thus leading to a minimum value of the I p . After the BSA-Apt-Au/SSE electrode was modified by the PS-COOH, the I p value increased due to the electrochemically active sites on the PS-COOH. However, the I p value of the PS-BSA-Apt-Au/SSE electrode decreased after the AFB 1 addition due to the combination of AFB 1 and Apt, which gave rise to fewer PS-COOH nanospheres on the electrodes. The steric hindrance of AFB 1 may also have hindered the charge transfer in the electrolyte. When the PS-BSA-Apt-Au/SSE electrode was modified by HRP, a relatively higher I p value was noted, indicating that the HRP was successfully immobilized on the electrode. In addition, the integration of the PS-COOH with HRP, serving as a signal amplification probe, greatly enhanced the current of the electrode.

Detection of AFB 1
The relationship between the I p and the concentration of AFB 1 is shown in Figure 4a. It indicated that the biosensors exhibited distinct responses to different concentrations of AFB 1 because of the varying amounts of HRP immobilized on the electrode. It was found that there was a linear relationships between the I p and the concentrations of AFB 1 in the low concentration ranges. In all cases, the I p gradually decreased with an increasing concentration of AFB 1 . Figure S4 and Figure 4b are the DPV curves of AFB 1 within the high and low concentration ranges, respectively. Figure 4c demonstrates that the I p was inversely proportional to the AFB 1 concentration from 0.1 to 10 pg mL −1 with a square of the correlation coefficient R 2 of 0.9943. The linear relationship could be described as I p = −1.8627 C AFB1 + 44.966, where C AFB1 denotes the concentration of AFB 1 . According to the method reported by Shrivastava and Gupta, the calculation formula of the LOD is LOD = 3L/b, where 3L is the standard deviation based on three times the blank sample measurement and b is the slope of the calibration curve [50]. The LOD of the electrochemical biosensor was 0.016 pg mL −1 . For three independent electrodes, the relative standard deviations (RSDs) of the I p values at different concentrations of AFB 1 were less than 3%. The LOD of the biosensor was closely related to the three-dimensional structure of the selfsupporting electrodes. The results showed that the prepared biosensors were successfully applied to the quantitative analysis of AFB 1 .

Effects of H2O2 Concentration
To evaluate the effects of the H2O2 concentration on the electrolyte, i-t was applied. Figure 4d shows the electrochemical response of the electrochemical biosensor with different concentrations of H2O2. For each time, 5 mM of H2O2 was added to the electrolyte and the current change was observed. It could be seen that the current gradually increased with an increasing H2O2 concentration. However, the magnitude of the increase became smaller after 10 mM of H2O2 was added. Therefore, 10 mM of H2O2 was considered to be the optimal concentration in the electrolyte solution for the detection of AFB1.

Stability, Reproducibility and Specificity
For the practicability of an electrochemical biosensor, the stability, reproducibility and specificity of the biosensor are important. To evaluate the reproducibility of the biosensors, six electrode samples (S1-S6) were prepared at the same time. The test was repeated three times on each electrode and the RSD was calculated. Figure 5a shows the results of the reproducibility. The recovery rates were in the range of 93.8~105.9%. In addition, the RSDs of the recovery rate values from each electrode of the six samples were less than 3%. The results indicated that the developed electrochemical biosensors had a good reproducibility. For the stability of the biosensors, the prepared BSA-Apt-Au/SSE electrodes were prepared and stored in a sealed container at 4 °C for four weeks. They were used to detect 10 pg mL −1 of AFB1 per week. The corresponding results are shown in Figure 5b. The current responses slightly decreased with the increasing storage time. After four weeks, the recovery rates of the electrochemical biosensors reduced to 89.6%. The

Effects of H 2 O 2 Concentration
To evaluate the effects of the H 2 O 2 concentration on the electrolyte, i-t was applied. Figure 4d shows the electrochemical response of the electrochemical biosensor with different concentrations of H 2 O 2 . For each time, 5 mM of H 2 O 2 was added to the electrolyte and the current change was observed. It could be seen that the current gradually increased with an increasing H 2 O 2 concentration. However, the magnitude of the increase became smaller after 10 mM of H 2 O 2 was added. Therefore, 10 mM of H 2 O 2 was considered to be the optimal concentration in the electrolyte solution for the detection of AFB 1 .

Stability, Reproducibility and Specificity
For the practicability of an electrochemical biosensor, the stability, reproducibility and specificity of the biosensor are important. To evaluate the reproducibility of the biosensors, six electrode samples (S1-S6) were prepared at the same time. The test was repeated three times on each electrode and the RSD was calculated. Figure 5a shows the results of the reproducibility. The recovery rates were in the range of 93.8~105.9%. In addition, the RSDs of the recovery rate values from each electrode of the six samples were less than 3%. The results indicated that the developed electrochemical biosensors had a good reproducibility. For the stability of the biosensors, the prepared BSA-Apt-Au/SSE electrodes were prepared and stored in a sealed container at 4 • C for four weeks. They were used to detect 10 pg mL −1 of AFB 1 per week. The corresponding results are shown in Figure 5b. The current responses slightly decreased with the increasing storage time. After four weeks, the recovery rates of the electrochemical biosensors reduced to 89.6%. The excellent stability of the biosensor could be attributed to the three-dimensional structure of the SSE and the robustness of the Apt.  As a few impurities may have existed in the actual detection process, different toxins, ions and macromolecular organics were used as interference to test the specificity of the biosensors. Figure 6 shows the results of the interfering toxin OTA, ions and macromolecules on the electrode. When only OTA (1 pg mL −1 ) was present, the recovery rate was 2.1%, which may have been due to the absorption of the PS-COOH in/on the electrode via non-chemical forces. When both AFB1 (1 pg mL −1 ) and OTA (1 pg mL −1 ) were present, the recovery rate was similar to that of AFB1 (1 pg mL −1 ) alone. In addition, tests of other interferential substances were performed by i-t. The details are as follows. The same concentration (5 mM) of the interfering substances and H2O2 were added to the electrolyte at regular time intervals (60 s). It was found that the current did not change after adding the interfering substances whereas it significantly increased after adding H2O2. Therefore, the results confirmed that the electrode material could specifically recognize AFB1 without being interfered with by OTA, Na + , K + , Mg 2+ , AA, CA, starch and G.

Real Sample Analyses
Due to the portability of materials, electrochemical biosensors based on self-supporting materials have unique advantages in applications. To verify the operability of the electrochemical biosensor in actual samples, the concentrations of AFB1 in different linear ranges of two real samples of wine and soy sauce were detected. A known amount of AFB1 As a few impurities may have existed in the actual detection process, different toxins, ions and macromolecular organics were used as interference to test the specificity of the biosensors. Figure 6 shows the results of the interfering toxin OTA, ions and macromolecules on the electrode. When only OTA (1 pg mL −1 ) was present, the recovery rate was 2.1%, which may have been due to the absorption of the PS-COOH in/on the electrode via non-chemical forces. When both AFB 1 (1 pg mL −1 ) and OTA (1 pg mL −1 ) were present, the recovery rate was similar to that of AFB 1 (1 pg mL −1 ) alone. In addition, tests of other interferential substances were performed by i-t. The details are as follows. The same concentration (5 mM) of the interfering substances and H 2 O 2 were added to the electrolyte at regular time intervals (60 s). It was found that the current did not change after adding the interfering substances whereas it significantly increased after adding H 2 O 2 . Therefore, the results confirmed that the electrode material could specifically recognize AFB 1 without being interfered with by OTA, Na + , K + , Mg 2+ , AA, CA, starch and G.  As a few impurities may have existed in the actual detection process, different toxins, ions and macromolecular organics were used as interference to test the specificity of the biosensors. Figure 6 shows the results of the interfering toxin OTA, ions and macromolecules on the electrode. When only OTA (1 pg mL −1 ) was present, the recovery rate was 2.1%, which may have been due to the absorption of the PS-COOH in/on the electrode via non-chemical forces. When both AFB1 (1 pg mL −1 ) and OTA (1 pg mL −1 ) were present, the recovery rate was similar to that of AFB1 (1 pg mL −1 ) alone. In addition, tests of other interferential substances were performed by i-t. The details are as follows. The same concentration (5 mM) of the interfering substances and H2O2 were added to the electrolyte at regular time intervals (60 s). It was found that the current did not change after adding the interfering substances whereas it significantly increased after adding H2O2. Therefore, the results confirmed that the electrode material could specifically recognize AFB1 without being interfered with by OTA, Na + , K + , Mg 2+ , AA, CA, starch and G.

Real Sample Analyses
Due to the portability of materials, electrochemical biosensors based on self-supporting materials have unique advantages in applications. To verify the operability of the electrochemical biosensor in actual samples, the concentrations of AFB1 in different linear ranges of two real samples of wine and soy sauce were detected. A known amount of AFB1

Real Sample Analyses
Due to the portability of materials, electrochemical biosensors based on self-supporting materials have unique advantages in applications. To verify the operability of the electrochemical biosensor in actual samples, the concentrations of AFB 1 in different linear ranges of two real samples of wine and soy sauce were detected. A known amount of AFB 1 (1 pg mL −1 , 5 pg mL −1 , 10 pg mL −1 , 1 ng mL −1 , 3 ng mL −1 and 5 ng mL −1 ) was added to the two samples. Tables 1 and 2 show the recovery rates of the low and high concentration ranges of AFB 1 in wine and soy sauce, respectively. The recovery rates of the low and high concentration ranges of AFB 1 in wine were between 95.5~106.7% and 89.0~101.3%, respectively. In soy sauce, the recovery rates were between 94.1~101.6% and 87.5~92.3%, respectively. The RSDs of all detections were less than 2%. It showed that the biosensors had an outstanding accuracy, reliability and operability. These results indicated that the developed biosensor could be used to detect AFB 1 in complex samples. To further highlight the advantages of the electrochemical biosensors prepared in this work, a comparison of the performance of the biosensor with that of other electrochemical sensors reported in the literature is shown in Table 3. The existing methods for AFB 1 detection have many limitations such as a complicated operation, low accuracy, high detection cost and long detection time. Despite the complicated preparation process of the preliminary electrodes, the detection accuracy of ABF 1 was high and the detection time was relatively short. To be consistent, the performance tests were immediately conducted once the electrodes had been prepared.

Conclusions
In summary, a free-standing electrochemical biosensor with multi-scaled nanostructures was manufactured for the detection of AFB 1 on the basis of oxidation of H 2 O 2 catalyzed by HRP. A three-dimensional CNF/CF composite was utilized as a platform for the fabrication of the electrochemical sensors. Au was introduced into the electrode via a replacement reaction with Ni-Cu nanoparticles and further employed as a binder with Apt whilst the PS-COOH nanospheres were assembled into the support to afford abundant sites for immobilizing the HRP, thereby amplifying the electrochemical signals. The electrochemical results indicated that the current signal was closely related to the concentration of AFB 1 . The linear calibration ranges of AFB 1 were 0.1~10 pg mL −1 and the LOD was 0.016 pg mL −1 . The recovery rates of the detection of AFB 1 in wine and soy sauce were between 87.5% and 106.7%, respectively. The resulting electrochemical biosensors possessed prominent advantages of outstanding sensitivity and stability in the detection of AFB 1 , owing to the synergistic effects of the components. The free-standing feature of the electrochemical sensor afforded a facile operation during practical applications. The fabrication strategy reported in this work could be extended to prepare other electrochemical sensors by varying the Apts of target substrates.