The porous hollow cobalt-based oxides encapsulated with bimetallic PdAu Nanoparticles of electrochemical biosensor for highly sensitive pesticides detection

Efficient and portable electrochemical biosensors are received to evaluation of pesticides in the environment, which can make great significance for food safety. In this study, the Co-based oxides with a kind of hierarchical porous hollow and nanocages were constructed, in which the materials (Co3O4-NC) were encapsulated with PdAu nanoparticles (NPs). Due to the unique porous structure, the changeable valence state of cobalt and the synergistic effect of bimetallic PdAuNPs, PdAu@Co3O4-NC possessed excellent electron pathways, and showed more exposed active sites. Accordingly, the porous Co-based oxides have been applied to construct an acetylcholinesterase (AChE) electrochemical biosensor, which showed good performance for organophosphorus pesticides (OPs) detection. The optimum biosensing platform based on nanocomposites was applied to exhibit highly sensitive determination of omethoate and chlorpyrifos, with the relative low detection limit of 6.125 × 10−15 M and 5.10 × 10−13 M, respectively. And a wide detection range of 6.125 × 10−15 ∼ 6.125 × 10−6 M and 5.10 × 10−13 ∼ 5.10 × 10−6 M for these two pesticides were achieved. Therefore, the PdAu@Co3O4-NC may represent a powerful tool for ultrasensitive sensing of OPs, and have great potential application.


Introduction
Organophosphorus pesticides (OPs) were a common chemical agent used in agriculture to protect crops from worms and insects [1]. Over the last few decades, since the growing population and rapid urbanization, the use of OPs has increased globally, which may affect ecosystems, agricultural product safety and human health [2]. These OPs can destroy the biological function of acetylcholinesterase (AChE) in nervous system that regulates the acetylcholine neurotransmitter in the cholinergic system [3,4]. The AChE was inhibited that it led to the acetylcholine accumulation in the neuro-synapse resulting in prolonged neuro-excitation, which may cause muscular dystrophy, neurological disorders, reproductive disorders, even lead to death [5,6]. As a result, it is crucial to develop effective and sensitive pesticide detection methods. Currently, traditional pesticide detection methods included mass spectrometry (MS) [7], high performance liquid chromatography [8] and gas chromatography-mass spectrometry [9]. These methods can achieve qualitative and quantitative analysis, but cannot meet the point-of-care testing detection of pesticides [10][11][12]. Notably, AChE inhibitionbased biosensors have gotten attractive attention due to their simplicity, high sensitivity, and relatively low cost [13]. Enzyme inhibition-based method depended on that OPs can inactivate the AChE by phosphorylation of the serine residue from the catalytic triad in the AChE active center. And this would result in a decrease in the production of thiocholine (TCl), which was generated from the hydrolysis of acetylthiocholine chloride (ATCl) by AChE. TCl was an electroactive substance and can be oxidized to produce an oxidation current at a certain voltage, which can be used as a label for the detection of OPs [14].
New nanomaterials have been used in electrochemical sensors for more than a decade. Many of the research efforts have been directed toward applying the adjustable materials including metal-organic frameworks (MOFs) [15,16], graphene [17,18], or MXene [19]. The MOFs with unique properties have been used as precursors with possibility of post synthetic modifications and stability. Liu et al [20] prepared a robust ratiometric fluorescent nanosensor base on NH 2 -MIL-101(Fe), exhibiting unusually high performance to assay carbaryl. Notably, MOF-derived transition metal oxides may create better catalytic performance due to valence changes. Variable metal between two valence states had suitable redox properties and could speed up electron transfer. For instance, Ce 4+ /Ce 3+ redox pairs can be used as charge transfer media to promote electron transfer and demonstrate a sensitive detection of paraoxon [21,22]. Moreover, certain Co-MOF derivatives (e.g. Co 3 O 4 ), have gained prominence for better electrochemical performance, owing to a redox reaction could occur from Co 3+ to Co 2+ and/or accelerated catalysis versus Co-MOF [23,24]. Surprisingly, in these systems, the attractive features of Co 3 O 4 suggested tremendous promise for the higher oxytropism increased the affinity and adsorption of ATCl [25]. Notably, noble metal particles, such as Pd nanoparticles (PdNPs), PtNPs et al were suitable as modified electrode materials, because of their high conductivity and catalytic activity. In addition, -NH 2 on the framework of MOF can offer effective adsorption sites for Pd NPs, benefitting the dispersion and stabilization of activated Pd metal particles [26]. Uniform dispersion of nanoparticles is of great significance for the improvement of electrical conductivity and catalytic performance [27]. And gold nanoparticles (AuNPs) have been widely used in biosensing field due to their excellent biocompatibility and electrical conductivity [28,29]. Additionally, bimetallic nanomaterials can integrate the characteristics of two metals and have higher specific surface area and synergy, better catalysis and conductivity than single metal nanoparticles [30]. Bimetallic nanoparticles could be used as a multifunctional platform because their properties depended on their composition, size and shape, so their synthesis methods and technical applications attract many researchers. And bio-signal amplification based on bimetallic nanomaterials has achieved remarkable achievements in nanotechnology-based biosensor platforms and has great potential to improve the sensitivity and selectivity of electrochemical biosensors [31]. Lu et al [32] synthesized one-dimensional bimetallic Pd@Au core-shell nanorods and constructed a highly sensitive AChE biosensor for the detection of OPs. Therefore, we immobilized the PdNPs by amino-modified Co-MOF to keep them well dispersed. Besides, the biocompatibility of the material was improved by the formation of Pd-Au bimetals, which in turn enhance its electrical conductivity and catalytic activity.
Herein, PdAu@CoMOF-NH 2 nanocomposites were synthesized and treated at annealing to form a Co-based oxide with a hollow structure for the detection of OPs. The PdAu@Co 3 O 4 -NC have preferable biocompatibility, excellent catalytic activity. The conversion of ATCl to TCl was promoted by using the synergistic effect of AChE immobilized on the electrode with the double-shell amorphous Co 3 O 4 . Meanwhile, the Co 3 O 4 -NC were modified by PdAuNPs, which promoted the catalysis of TCl (with functional group of -S-H-) to generate dithio-bis-choline (-S-S-), thereby generating an oxidation current [33,34]. At the same time, PdAu@Co 3 O 4 -NC shortened the electron transfer route, which further effectively improved the amplification of the electrocatalytic signal. Overall, this work not only provided a new and innovative method to design double-shelled and porous Co 3 O 4 -NC loaded bimetallic PdAuNPs to constructed AChE biosensor, but also showed a sensitive and efficient electrochemical strategy for OPs detection.

Synthesis of CoMOF-NH 2
The synthesized method of ZIF-67 crystal was reported by Sara Rafiei et al [35] and the synthesis step has been slightly improved. In details, 2 mmol Co(NO 3 ) 2 ·6H 2 O and 6 mmol H-MeIM were dissolved into methanol (30 ml) solution respectively. And then the Co(NO 3 ) 2 ·6H 2 O mixing of solutions was slowly added to the H-MeIM solution with continuously stirring to evenly mix the solution with the condition of room temperature for 8 h. After centrifugation, it was washed with methanol and water five times. The collected centrifuged material was dried in the vacuum drying oven at 60°C for 10 h, and the product of ZIF-67 was obtained.
To prepare the CoMOF-NH 2 , 18.8 mg ZIF-67 and 0.030 24 g NH 2 -BDC with a solvothermal method in DMF. NH 2 -BDC was dissolved in methanol, and then 20 ml of ZIF-67 was added dropwise to the NH 2 -BDC mixture every five minutes and left to stir at room temperature for 3 h. The stirred mixed solution is transferred to the autoclave and the solvent heat reaction is carried out at 100°C for 10 h. When the reaction was finished, the precipitate was repeatedly washed with distilled water, collected by centrifugation and dried at 60°C for 10 h.

Synthesis of PdAu@CoMOF-NH 2 and PdAu@Co 3 O 4 -NC
PdAu@CoMOF-NH 2 was synthesized based on other literatures [36,37]. And the experimental parameters including temperature, pressure, reaction time and environment were modified to obtain the desired nanocomposite. The final sample preparation method was determined as follows: first of all, 50 mg CoMOF-NH 2 and a certain amount of K 2 PdCl 4 (500 μl, 2.5 mM) aqueous solution were put in a small beaker. Subsequently, added 2 ml of methanol to the above mixture, and stirred to make it evenly mixed. It was then transferred to a porcelain crucible and heated at a temperature of 70°C to evaporate the solution. To give Pd 2+ /CoMOF-NH 2 , this experimental procedure was repeated several times. The Pd 2+ /CoMOF-NH 2 sample was then dispersed into 5 ml of MeOH by sonication, and NaBH 4 (500 μl, 5 mM)was added while stirring. The NaBH 4 solution was then slowly added dropwise to the above mixture, allowing it to react completely. The obtained product was washed with methanol, heated to 60°in a water bath, and HAuCl 4 aqueous solution (500 μl, 5 mM) was slowly added to it and reacted for 8 h. The prepared product was washed several times with MeOH, and then centrifuged until the supernatant liquid became colorless. The product was dried in a vacuum oven at 60°C for 12 h. Finally, the sample (PdAu@Co 3 O 4 -NC) of 100 mg was placed in a muffle furnace and treated at 300°C for 2 h in the air to obtain.

Preparation of AChE-CS/PdAu@Co 3 O 4 -NC/GCE
Before modifying the electrode, the bare GCE was polished using alumina powder (50 nm to 0.3 μm), washed with acetone, anhydrous ethanol and ultra-pure water, then dry under nitrogen. After the GCE was cleaned, 5 μl of PdAu@Co 3 O 4 -NC aqueous solution (diluted to 0.25 mg·ml −1 with deionized water) was mixed with naphthol and dropped onto the electrode surface prepared in advance. The treated electrodes are left to dry in natural conditions. The CS had great biocompatible, hydrophilicity, film forming ability and non-toxicity properties. And it provided a natural microenvironment for AChE and also offered sufficient access for electrons to move between AChE and electrodes [38]. Then, 1 μl of AChE (0.2 U·μl −1 ) and 4 μl of CS (1:4) solution were added dropwise on the surface to fix the electrode surface (AChE-CS/PdAu@Co 3 O 4 -NC/GCE), and placed in a refrigerator at 4°C.

AChE sensing experiment
The AChE-CS/PdAu@Co 3 O 4 -NC/GCE biosensors were used for OPs detection. The differential pulse voltammetry (DPV) measurements were employed for detecting the pesticides (omethoate and chlorpyrifos) with a potential from 0 to 1 V (pulse amplitude: 50 mV, potential increment: 5 mV, pulse width: Omethoate and chlorpyrifos were considered as inhibitors of AChE, which could reduce the output of the corresponding hydrolyzed TCl, resulting in a weaker oxidative current. The inhibition of OPs was estimated as follows: inhibition (%) = (I 0 -I 1 )/I 0 × 100%, where I 0 and I 1 were the peak currents before and after incubation with pesticides.

Characterization of PdAu@Co 3 O 4 -NC
The XRD pattern was used to characterize the crystal structure, and the lattice parameters were showed in figure S1. The XRD pattern of the synthesized products can be indexed to the spinel Co 3 O 4 (JCPDS # 43-1003) [39]. All the sharp and prominent peaks between 5°and 80°match well with that of previous reported x-ray data, clearly indicate that spinel Co 3 O 4 is obtained.
The structure and morphology of ZIF-67 and PdAu@Co 3 O 4 -NC were studied by SEM and TEM. The SEM images in figure 1(a) and the TEM images in figure 1(c) showed that the ZIF-67 has a typical rhombic dodecahedron crystal morphology with a particle diameter of about 500 nm. The TEM in figure 1(d) and the SEM in figure 1(b) showed the size and surface morphology of ZIF-67-derived CoMOF-NH 2 . After the reaction, the contours of the rhombic dodecahedron material became blurred (from the figure 1(b) can be seen). The morphology and the size of the CoMOF-NH 2 was similar to the ZIF-67. From the figure 1(d), these CoMOF-NH 2 exhibited a double-shell structure with no apparent change in shape as compared to ZIF-67. This was because ZIF-67 could react with diamino isophthalic acid ligand to form CoMOF-NH 2 [40]. The TEM images in figure 1(g)   visually illustrate the similar structural profiles of PdAu@CoMOF-NH 2 and CoMOF-NH 2 . It can be seen that the PdAuNPs were uniformly distributed on the surface of CoMOF-NH 2 , and the high-quality NPs were homogeneous morphology and with the size distribution of about 4−8 nm ( figure 1(g)). And in the case of good dispersion, smaller nanoparticle sizes have higher catalytic activity, greater electrochemical activity and electrical conductivity [31,41]. This may be attributed to the fact that smaller diameter nanoparticles have larger specific surface area, more surface defects and greater surface activity energy [42][43][44]. After the precursor was heat-treated in air at 300°C (figures 1(f) and (h)), the polyhedron size changed a little. In the figure 1(f), the co-based oxides were hollow and the surface becomes rough with holes, showing a unique two-dimensional layered morphology. The TEM image in figure 1(h) further revealed the rough surface of the PdAu@Co 3 O 4 -NC and the morphology has significantly changed, and the PdAu@CoMOF-NH 2 framework partly collapsed to form PdAu@Co 3 O 4 -NC. The reasons for the size collapse were the decomposition of organic ligands and the shedding of some functional groups on the surface caused by heat treatment. From the TEM image in figure 1(h) and the figure 1(i) can be seen that the PdAuNPs dispersed on the surface of CoMOF-NH 2 with distinct lattice fringes. The interplanar distance was 0.195 nm, which could be attributed to the Pd (200) plane, and the interplanar distance was 0.235 nm, which could be matched to the plane of Au (111). This further confirmed that the bimetallic PdAuNPs were successfully encapsulated on the precursor surface. In addition, the homogeneous distribution of Pd, Au, N, and O elements in the TEM-mapping images further confirmed the successful synthesis of the target PdAu@Co 3 O 4 -NC ( figure 1(j)).

Electrochemical characterization based on biosensing platform
CV and EIS are used to measure different performance characteristics of modified electrodes in 5 mM [Fe(CN) 6 ] 3− / 4− solution containing 0.1 M KCl. The CV was performed from −0.2 to 0.6 V, and the scan rate ranged from 0.01 to 0.2 V s −1 . This process was used to investigate the step-by-step fabrication process of PdAu@Co 3 O 4 -NC electrodes. In figure 2(a), the PdAu@Co 3 O 4 -NC and PdAu@CoMOF-NH 2 modified GCE exhibited excellent enhancement of peak currents compared to the bare GCE. Due to the conductive properties of PdAu@Co 3 O 4 -NC, the peak current was significantly enhanced, accelerating the charge transfer rate. Furthermore, in the AChE/PdAu@Co 3 O 4 -NC/GCE electrode, a significant decrease in the peak current value was observed due to the non-conductive nature of the AChE, and it was also demonstrated that AChE was successfully immobilized on the surface of the electrode.
The EIS was performed in the frequency ranged from 0.01 Hz to 100 kHz with an amplitude of 5 mV. In general, the semicircle in the Nyquist diagram represents the electron transfer resistance (Rct) [45][46][47]. And the value of the resistance depends on the diameter of the semicircle. In figure 2(b) we can see that the charge transfer resistances decrease in the sequence of PdAu@Co 3 O 4 -NC/ GCE > PdAu@CoMOF-NH 2 /GCE > Bare GCE > CoMOF-NH 2 /GCE > AChE/PdAu@Co 3 O 4 -NC/GCE. The addition of AuPdNPs increases the electrical conductivity of the sample, and also reduces its electrical impedance. In addition, according to the EIS measurement results, the modification of AuPdNPs also enhances the capacitive response of the composite [48]. It also showed that the electrochemical performance of PdAu@Co 3 O 4 -NC was excellent.

Optimization of AChE/PdAu@Co 3 O 4 -NC/GCE platforms
The effect of different conditions was investigated to achieve a better performance of the biosensor. The current response of the enzyme electrode was tested in a 5 mM ATCl. The current responses of sensors constructed through different ratios of HAuCl 4 to K 2 PdCl 6 were explored. The nanomaterials prepared when the ratio of HAuCl 4 to K 2 PdCl 6 was 0:1, 1:5, 1:2, 1:1, 2:1, 5:1 were used to construct different AChE biosensors. The biosensors were placed in 5 ml PBS containing ATCl (5 mM, 1 ml), and the current response was recorded (n = 3). The results were shown in figure S2. The current response increased with the increasing ratio of HAuCl 4 , and the maximum was reached when the ratio of HAuCl 4 to K 2 PdCl 6 was 1:1 and as the ratio of HAuCl 4 to K 2 PdCl 6 further increased, the current tended to be flat accordingly. This may be attributed to the better electrical conductivity and excellent catalytic activity of PdAu bimetals. When HAuCl 4 solution was added to Pd@CoMOF-NH 2 , the PdAu bimetals were formed by the in situ replacement reaction. And the number of bimetals increased with the addition of HAuCl 4 solution and eventually reached saturation.
The synthesis of the materials would also be influenced by the reaction time of HAuCl 4 and Pd@CoMOF-NH 2 . Biosensors based on the nanomaterials prepared with different reaction times were assembled and the current responses of them to ATCl were recorded though DPV. The results in figure S3 showed that after a reaction time of 8 h, the current response reached its maximum. This phenomenon exhibited that the final nanocomposites prepared were almost PdAu bimetallic nanomaterials.
The pH of the PBS could seriously affect the catalytic activity of immobilized AChE. The pH of the PBS (6.5-8.5) was adjusted to test for optimal conditions. From figure 3(a), it can be seen that the current reaches a maximum at the pH of 7.4. Therefore, the optimum pH of the experiment was 7.4 in the subsequent tests.
In addition, the dosage of AChE was also the critical parameter for biosensors. In figure 3(b), with the increase of AChE dosage, the electrochemical response gradually increased. When the AChE concentration was further increased to 0.2 U, the electrochemical response reached the optimum value, and when the AChE concentration exceeded 0.2 U, the DPV response decreased relatively. This may be too little AChE will lead to insufficient enzyme catalysis reaction. However, when too much AChE was used to modify the electrode, the modified layer will be too thick, which will hinder the speed of electron transfer and reduce the sensitivity of the biosensor. Thus, in the following experiments, 0.2 U was used as the optimal dosage of AChE.
Furthermore, the incubation time also had an important impact on the analytical performance of the enzyme inhibition biosensor. Figure 3(c) showed the effect of incubation time on the inhibition. The optimum incubation time was determined with omethoate. The electrodes were immersed in omethoate solution (6.125 × 10 −12 M) and set for different times. It can be seen that the inhibition gradually increased from 5 to 25 min, and then remained at a similar level, indicating that the binding of OPs to AChE reached saturation. Therefore, the optimal inhibition time for pesticides was set to 20 min.

AChE/PdAu@Co 3 O 4 -NC/GCE platforms amperometric response
In order to evaluate the electrochemical response of the constructed AChE biosensor for ATCl, the time-current (i−t)  figure 3(f)). The Km was inversely proportional to the affinity of the AChE to the substrate. In this study, the Km value of the constructed AChE-CS/PdAu@Co 3 O 4 -NC/GCE biosensor was 67.618 μM, less than 1.56 mM [50], 126 μM [51], 142 μM [52]. As a result, because of the conversion of Co 3+ to Co 2+ , the material has good catalytic ability and the porosity of the material can effectively accelerate the electrochemical reaction. This confirmed that the AChE-CS/PdAu@Co 3 O 4 -NC/GCE had higher affinity and catalytic activity for ATCl.

Determination of omethoate and chlorpyrifos
The AChE-CS/PdAu@Co 3 O 4 -NC/GCE electrochemical biosensor was prepared and DPV was used to detect pesticides under optimized conditions. The OPs can covalently bind the activity center of AChE, which reduce peak current from electrochemical oxidation of TCl.  Table 1 summarized the performance comparison of AChE-CS/PdAu@Co 3 O 4 -NC/GCE biosensor with other AChE biosensors, and the results showed that the as-prepared biosensor had comparable or even lower LOD and wider detection range. There are several possible reasons for the good analytical performance of AChE-CS/PdAu@Co 3 O 4 -NC/GCE. The porous hollow structure provided additional sites for the immobilisation of AChE. In addition, the introduction of PdAuNPs offered satisfactory conductivity for Co 3 O 4 based materials. Furthermore, the AChE-CS/PdAu@Co 3 O 4 -NC/GCE had excellent affinity and catalytic activity for ATCl.
The prepared biosensor was applied to the detection of pesticides in real samples, which further evidenced the practicability. The tap water was used as the sample for testing, and the standard addition method was used for the recovery experiment. After calculation, it can be concluded that the relative standard deviation (RSD) was 1.81% ∼ 2.52% and the acceptable recovery was 93.17% ∼ 108.25%, indicating that the developed biosensor was feasible and sensitive for monitoring pesticides in real samples (see table 2).

Reproducibility, anti-interferences and stability
To investigate the reproducibility and repeatability of the electrochemical sensor, five independent DPV experiments were carried out in the presence of ATCl for the AChE-CS/PdAu@Co 3 O 4 -NC/GCE. Based on the electrochemical test result, the RSD of the five independent tests were 1.37%, 2.06%, 2.16%, 1.96% and 1.78% (each electrode was used for five replicate DPV experiments), indicating that the biosensor we constructed had good repeatability. The RSD of five different electrodes was 1.26%, which further indicated that the AChE-CS/PdAu@Co 3 O 4 -NC/GCE has good reproducibility.
The negligible current changes after the addition of some representative biological species and electrolytes indicate that the sensor has an outstanding anti-interference property for the detection of OPs. Hence, to evaluate the anti-interference ability of AChE-CS/PdAu@Co 3 O 4 -NC/GCE for the detection of OPs, experiments were carried out via the addition of some common interfering species such as Zn 2+ , Mg 2+ , Hg 2+ , NO 3

Conclusion
In summary, we have developed a PdAu@Co 3 O 4 -NC based AChE biosensor and used it for the highly sensitive determination of pesticides. The electron transfer capability and corresponding electrochemical properties were further improved due to the synergistic effect of Co 3 O 4 -NC and PdAuNPs. Meanwhile, the hollow and microporous structure of PdAu@Co 3 O 4 -NC greatly accelerated the electron transfer and effectively improved the catalytic performance of AChE. Through optimized conditions, using omethoate and chlorpyrifos as model pesticides, the AChE-CS/PdAu@ Co 3 O 4 -NC/GCE biosensor achieved a wide linear detection range and a low detection limit. In addition, the developed biosensor possessed excellent storage stability, anti-interference and favorable detection effect in water samples. The results showed that the biosensor had great potential in the detection of OPs as well as the field analysis.