Efficient Preparation of a Nonenzymatic Nanoassembly Based on Cobalt-Substituted Polyoxometalate and Polyethylene Imine-Capped Silver Nanoparticles for the Electrochemical Sensing of Carbofuran.

The ever-growing exploitation of pesticides and their lethal effects on living beings have made it a dire need of the day to develop an accurate and reliable approach for their monitoring at trace levels. The designing of an enzyme-free electrocatalyst to electrochemically detect the pesticide residues is currently gaining much importance. In this study, a novel redox-sensing film was constructed successfully based on cobalt-substituted Dawson-type polyoxometalate [P2W17O61 (Co2+·OH2)]7- (Co-POM) and polyethylene imine (PEI)-capped silver nanoparticles (AgNPs). A nanohybrid assembly was fabricated on a glassy carbon electrode's surface by alternately depositing Co-POM and PEI-AgNPs using the layer-by-layer self-assembly method. The surface morphology of the immobilized CoPOM/AgNP multilayer nanoassembly was analyzed through scanning electron microscopy along with energy-dispersive spectroscopy for elemental analysis. The redox properties and surface morphologies of fabricated assemblies were evaluated by cyclic voltammetry and electrochemical impedance spectroscopy. The practicability and feasibility of the proposed sensing layer was tested for the detection of a highly toxic insecticide, that is, carbofuran. The fabricated sensor exhibited a limit of detection of 0.1 mM with a sensitivity of 13.11 μA mM-1 for carbofuran. The results depicted that the fabricated nonenzymatic hybrid film showed excellent electrocatalytic efficiency for the carbofuran oxidation. Furthermore, the obtained value of "apparent Km", that is, 0.4 mM, illustrates a good electro-oxidation activity of the sensor for the detection of carbofuran. The exceptionally stable redox activity of Co-POM, high surface area and greater conductivity of AgNPs, and the synergistic effect of all components of the film resulted in an excellent analytical performance of the proposed sensing assembly. This work provides a new direction to the progress and designing of nonenzymatic electrochemical sensors for pesticide determination in real samples.

Introduction

The problem of environmental pollution is alarmingly increasing despite the advancement in science and technology. The accumulation of pesticide residues in the environment due to the unrestrained use of organophosphate (OP) and carbamate (CB) pesticides imparts serious hazards in human and nontarget organisms as well as the sustainability of the environment is compromised.[1] Therefore, the development of precise and rapid techniques for the sensitive detection of pesticide residues from air, food, or water is an important focus of modern research to ensure food safety. To date, various electrochemical sensing techniques have been designed for the detection of toxic amounts of pesticide residues from the environment or agricultural products.[2−5] These methods have emerged as potential alternates for traditional detection techniques which are considered as time-consuming, expensive, laborious, requiring highly trained persons to operate the highly sophisticated instruments, and are limited in centralized laboratories such as mass spectrometry (MS), liquid chromatography–MS, gas chromatography–MS, HPLC, electrophoresis, and so forth owing to their inexpensiveness, rapidity, lack of sample pretreatment steps, simple design, easy handling, and feasibility of on-site detection of pesticides.[6,7] The electrochemical methods have usually based on targeting a nervous system enzyme acetylcholinesterase (AChE), which is inhibited by the action of OP and CB pesticides.[8−11] The enzyme-based electrochemical biosensors can be reliable tools for on-site monitoring of pesticides if they could be miniaturized in the form of a portable measuring device. However, the use of enzymes is critical because of its low stability, storage requirements, and optimum conditions such as pH, temperature, and so forth. The requirements of a controlled environment for enzyme and its low reproducibility in turn reduce the life span of these biosensors and thus limit their practical applications.[12,13] To overcome these problems, the nonenzymatic electrochemical methods have attracted much attention over the past few years, as rapid, inexpensive, and sensitive tools for the detection of various organic and inorganic substrates.[7,14−18] In this context, the modification of a transducer surface with a suitable nanomaterial that provides excellent electrocatalytic ability to detect the oxidation of pesticides on the electrode surface is a useful strategy to fabricate a nanozyme electrochemical sensor having good analytical performance. Various nanomaterials, nanotubes, and nanocomposites have been reported in the literature for nonenzymatic electrochemical monitoring of pesticides, harnessing their distinctive structure, high surface area, and excellent electrical conductivity, which makes them an ideal candidate for this purpose.[19−21]

Polyoxometalates (POMs) is a significant class of inorganic anionic metal oxides displaying a huge diversity in their structural framework and chemical composition.[22,23] Due to their diversified and unparalleled intrinsic features, they have been utilized in indispensable research areas, such as photocatalysis, biotechnology, environmental monitoring, sensing, electrochromism, and medicine.[24,25] Specifically, POMs exhibit numerous advantages as catalysts, such as multielectron reversible redox features which can be fine-tuned by modifying their compositions and chemical structures, which makes them a good redox electrocatalyst for electrochemical procedures. Moreover, they are not very liable to oxidative and thermal degradation and their active sites can be modified by a combination of multiple transition metal ions as inorganic ligands. The POM-centered redox sensors have been reported for the detection of a variety of analytes such as NO–2, IO–3, BrO–3, hydrazine sulfate, H2O2, and so forth.[26−29] However, the realization of POM-based devices requires the surface anchoring of these building blocks onto some transducer, for example, electrode surfaces. The morphology of nanostructures greatly influences the signal response of the designed electrochemical sensor, and their structural reproducibility depends on the method of their immobilization on the electrode surface. In this regard, various immobilization strategies have been employed by researchers such as layer-by-layer (LBL) methodology, high vacuum evaporation, Langmuir–Blodgett technique, or electrochemical approach, where the LBL self-assembly method has been known to build thin films with desired thickness, morphology, and functional properties at the molecular level.[30−32]

Metal nanoparticles have a huge potential in nanotechnology, especially noble metal nanoparticles possess fascinating physical and chemical properties. The shape and size of metal nanoparticles allow efficient transfer of electrons, which explains their exceptional catalytic activity. They have been widely used in biochemistry, catalysis, and electrochemical sensors.[33,34] The incorporation of noble metal nanoparticles into the POM films not only improves the intrinsic features to the original POM films but also introduces new remarkable properties by the virtue of synergistic effect of both components. Various POM films integrated with noble metallic NPs have been reported recently.[26,27]

In this work, we have combined the features of cobalt-substituted polyoxometalate (Co-POM) and polyethylenimines (PEIs) wrapped cationic silver nanoparticles (AgNPs). AgNPs have been selected as they have been largely studied for their potential as effective electrocatalysts for the oxidation and detection of pesticides. They were capped with PEIs which are well-known synthetic polymers containing amine groups. These are highly basic and positively charged polymers, which can interact strongly with metal nanoparticles and prevent their aggregation as well as stabilize their structure. The LBL approach was exploited to form alternate multiple layers of Co-POM and AgNPs on the glassy carbon electrode (GCE). The fabricated assembly has been employed for the sensing of a potential CB pesticide, carbofuran, which is highly toxic if accumulated in the environment. The as-prepared Co-POM/AgNPs/GCE sensor exhibited excellent electrocatalytic activity for the detection of carbofuran oxidation. Regarding the materials used in this research work, POMs and AgNPs are very easy to synthesize from inexpensive starting compounds by adopting very simple synthetic procedures. Hence, the devised sensor platform provides a new opportunity for the designing of inexpensive, fast nonenzymatic electrochemical sensors with high sensitivity and low detection limits that would provide possible applications in the area of electroanalytical chemistry and agriculture (Scheme ).

Scheme 1

Schematic Illustration of Construction of an Enzyme-Free Electrochemical Sensor by LBL Immobilization of Polyanionic Co-POM and Cationic PEI-Wrapped AgNPs on Substrate GCE for the Electro-Oxidation of Carbofuran Pesticide

Results and Discussion

Fourier Transform Infrared Spectroscopy Analysis of Co-POM

Fourier transform infrared spectroscopy (FT-IR) spectra displayed characteristic vibration bands of Dawson-type POMs present in the range of 4000–500 cm–1. Figure A demonstrated the absorption bands between 700 and 1100 cm–1, proving the existence of the Dawson structure. The peak value at 789 cm–1 can be allotted to the bridging vibrations of tungstate-oxygen-tungstate ν(W–Oc–W), band at 908 is attributed to ν(W–Ob–W), and another band at 967 is attributed to ν(W–Ot), while bands at 1015 and 1082 represented the existence of ν(P–O) vibrations. The presence of water of crystallization was confirmed by the band at 3562 (cm–1).

Figure 1

FT-IR spectra (A) and X-ray diffraction pattern (B) of K8 [CoP2W17O61·H2O]·16H2O.

X-ray Diffraction Analysis of Co-POM

Figure B explained the X-ray diffraction pattern of K8 [CoP2W17O61·H2O]·16H2O. It showed the intense peaks at 7.85, 8.9, and 9.4 characteristic for the Dawson structure.[35]

UV–Visible Spectra of AgNPs

The synthesis of the PEI-stabilized AgNPs was confirmed by UV–visible spectroscopy. The UV–vis spectrum (Figure S1) of AgNPs (1 × 10–3 M) showed a sharp and strong plasma resonance band at 408 nm which is distinctive of spherical-shaped AgNPs and an indication of uniform size distribution of as-prepared nanoparticles.[36]

Surface Characterization of the Co-POM/AgNP Nanoassembly

The surface morphology of immobilized CoPOM/AgNP multilayer nanoassembly was investigated through scanning electron microscopy (SEM) images. Figure demonstrates the SEM micrographs of the multilayer assembly with the outer most layer of AgNP (A) and of CoPOM (B). A uniform distribution of spherical aggregates can be observed after the fabrication of the multilayer assembly. The film with CoPOM as the outermost layer was distinguished by spherical aggregates of diameter from 20 to 65 nm. However, for the multilayer assembly having AgNP as an exposed layer, a bit smaller sized particles or compact masses/aggregates with diameters 10–15 nm have been observed probably due to the nanoparticles. Furthermore, the chemical composition of the Co-POM/AgNP composite film was analyzed by energy-dispersive spectroscopy (EDS). Figure C indicates the occurrence of seven elements, that is, C, O, Cl, K, Ag, W, and Co, with varied percentage compositions presented in the inset of Figure C. These results further verify the presence of AgNPs and Co-POM in the fabricated multilayer nanoassembly.

Figure 2

SEM micrographs of the CoPOM/AgNP nanoassembly comprising eight bilayers having outer a cationic AgNP layer (A) and outer anionic POM layer (B). Energy-dispersive spectrum of Co-POM/AgNP-based multilayer nanoassembly: inset shows percentage composition of each element present in the material (C).

Solution Electrochemistry of [P2W17O61(Co2+·OH2)]7–

The solution-phase electrochemical features of Co-substituted Dawson POM were evaluated by cyclic voltammetry at pH 4.5, as described in Figure S2. The voltammogram shows two reversible, stable, and well-defined redox couple process of tungsten-oxo (W–O) framework i–e W(I)/W(I′) and W(II)/W(II′) of POM associated with tungsten-oxo electrochemical activity with formal potential values of 57 and 35 mV (vs Ag/AgCl reference system). Another prominent redox couple can be seen at much higher positive potential with E1/2 of +125 mV, which is related to the redox activity of cobalt transition metal embedded within the framework of POM.

UV–Vis Spectroscopy of Co-POM/AgNP-Based Multilayer

Fabrication of the multilayers of Co-POM and AgNPs was also scrutinized by employing UV–vis spectroscopy. Figure S3 demonstrated the overlay of UV–vis spectra recorded after each deposition step. One can see an absorption band at 350–400 nm, due to the presence of AgNPs and a sharp absorption peak at 290 nm, which is attributed to the O → W transition of the cobalt-substituted POM. A shift in the absorption band of the AgNPs (as compared to Figure S1) is probably due to their interaction with the POM present in the film. A continuous escalation in the absorption intensity of the peaks indicated the subsequent deposition of the multilayer assembly onto the electrode surface.[37]

Electrochemical Response of the Multilayer Nanoassembly

The stepwise buildup of the LBL nanoassembly of cobalt-substituted Dawson-type POM and AgNPs onto the surface of GCA was assessed through cyclic voltammetry (Figure A). The redox behavior of the fabricated assembly was recorded in an acetate buffer of pH 3.5. A regular intensification in the cathodic and anodic currents for all the redox processes represents a continuous and facile deposition of POM and AgNP layers onto the electrode during the construction of the multilayer assembly. The continuous growth of the redox peak currents (Ip) with the increase in number of deposited layers represents the absence of the terminal layer effect, which depicts that the outermost layer does not influence the transfer of ions through the multilayer assembly/electrolyte interface during the redox processes. Figure B demonstrates a linear relationship between the measured charges entrapped in the second W–O redox activity with the number of layers deposited onto the working electrode surface.

Figure 3

(A) Cyclic voltammograms taken after the stepwise accumulation of the multilayer film comprising 16 monolayers (outermost layer of Co-POM) on GCE using pH 3.5 buffer as the electrolyte (0.1 M Na2SO4/20 mM CH3COOH) at 100 mV s–1 scan rate with negative scan direction, (B) relationship between the number of layers deposited on the surface of GCE and the corresponding charge accumulated for the second tungsten-oxo (W–O) redox process.

The surface coverage of the Co-POM/AgNP multilayer assembly was calculated by integrating the charge (Q) entrapped under the cathodic peak of the first tungsten-oxo process by using the equation: [Γ = Q/nFA], where, Q′ is the charge in coulomb, A′ is the electrode surface area in cm2, n′ is the no. of electrons involved in specific electrochemical reaction, and F′ is Faraday’s constant (96485 C mol–1). A steady increase in surface coverage “Γ” was observed from 0.001 nmol·cm–2 for the 2nd layer to 0.099 nmol·cm–2 for the 16th layer.

A representative voltammogram of the fabricated multilayer assembly can be seen in Figure . Two well-behaved reversible redox couples were observed and labeled as W(I)/W(I′) and W(II)/W(II′) having E1/2 values of 55 and 31 mV, respectively. These couples are linked to the redox behavior of the tungsten oxo centers of the POM framework. The peak-to-peak separation values (ΔE) of the two processes were calculated to be 21 and 45 mV, which are typical for the surface confined redox processes. Another electrochemical process occurring at some positive potential is attributed to the redox chemistry of AgNPs with the oxidation peak current comparatively higher as compared to the reduction peak current. An insignificant variation has been reflected in the electrochemical behavior of the cobalt-substituted POM in the solution phase and immobilized phase (Figures and 8). In the solution phase, a well-defined reversible redox process was observed at quite positive potential owing to the electrochemical activity of cobalt metal incorporated in the POM framework; however, this redox process was not evident in the case of the surface immobilized state. The reason of this peak suppression is still unknown.

Figure 4

Redox behavior of Co-POM/AgNP-based multilayer assembly with a surface coverage 0.1 nmol·cm2 of in pH 3.5 buffer (0.1 M Na2SO4/20 mM CH3COOH) on GCE (at scan speed 100 mV s–1, negative scan direction).

Figure 5

(A) Overlay of cyclic voltammograms recorded for the multilayer nanoassembly in pH 3.5 buffer at various scan rates [sweep rates: 10 (innermost), 20, 30, 40, 50, 60, 70, 80, 90, 100 125, 150, and 175 mV s–1]. (B) Plot of oxidation and reduction peak currents (Ipa and Ipc) against scan rate for multilayer assembly for the second W–O electrochemical process.

Figure 8

(A) Cyclic voltammograms of GCE coated with Co-POM/AgNP multilayer nanoassembly comprising 16 layers in the absence and presence of carbofuran up to 2 mM in Britton–Robinson buffer pH 4.0 (a) bare GCE, (b) 0.0, (c) 0.2, (d) 0.4, (e) 0.6, and (f) 0.8 mM carbofuran at a scan rate of 10 mV s–1; inset represents the calibration curve between the catalytic current (Icat) and concentration of carbofuran. (B) Lineweaver Burk plot between inverse of the concentration of carbofuran (1/CBF) and steady-state current response (1/Icat) of Co-POM/AgNP/GCE.

Effect of Scan Rate

The effect of the sweeping scans on the redox chemistry of the film showed electrochemical peak potentials corresponding to the tungsten-oxo electrochemical reactions were not influenced by the scan speed up to 400 mV s–1; however, the redox currents revealed a direct relationship to the scan rate as could be seen in Figure A,B. This feature is a characteristic of the surface immobilized process.[38] The redox behavior of the second W–O redox reaction is precisely described in Table .

Table 1

Electrochemical Data Obtained for the Co-POM/AgNP Multilayer Nanoassembly Comprising Eight Bilayers with an Outermost Anionic Co-POM Layer Depicting Different Voltammetric Parameters in pH 3.5 Buffer on the Surface of GCE (Area = 0.0707 cm2)

		W–O (I)				W–O (II)
surface coverage Γ (nmol cm–2)	scan rate (mV s–1)	Epc (mV) (Ag/AgCl)	Epa (mV) (Ag/AgCl)	E1/2 (mV) (Ag/AgCl)	ΔE (mV) (Ag/AgCl)	Epc (mV) (Ag/AgCl)	Epa (mV) (Ag/AgCl)	E1/2 (mV) (Ag/AgCl)	ΔE (mV) (Ag/AgCl)
0.1	10	–393.1	–322.9	–358	–70.2	–518.3	–501.1	–509.7	–17.2
	20	–154.9	–285.1	–220	130.2	–526.9	–485.7	–506.3	–41.2
	30	–189.6	–280	–234.8	90.4	–528.6	–489.1	–508.85	–39.5
	40	–357.18	–281.7	–319.44	–75.48	–528.6	–487.4	–508.00	–41.2
	50	–362.3	–283.4	–322.85	–78.9	–530.3	–485.7	–508.	–44.6
	60	–370.9	–280	–325.45	–90.9	–532	–489.1	–510.55	–42.9
	70	–352	–276.6	–314.3	–75.4	–537.1	–487.4	–512.25	–49.7
	80	–388	–278.3	–333.15	–109.7	–540.6	–489.1	–514.85	–51.5
	90	–377.7	–281.7	–329.7	–96	–530.3	–485.7	–508	–44.6
	100	–379.4	–273.1	–326.25	–106.3	–538.9	–485.7	–512.3	–53.2
	125	–285.1	–273.1	–279.1	–12	–537.1	–487.4	–512.25	–49.7
	150	–285.1	–271.4	–278.25	–13.7	–545.7	–487.4	–516.55	–58.3
	175	–280	–269.7	–274.85	–10.3	–547.4	–472	–509.7	–75.4
	200	–278.3	–271.4	–274.85	–6.9	–550.9	–487.4	–519.15	–63.5
	225	–278.3	–269.7	–274	–8.6	–552.6	–487.4	–520	–65.2
	250	–280	–271.4	–275.7	–8.6	–552.6	–487.4	–520	–65.2
	275	–280	–268	–274	–12	–552.6	–485.7	–519.15	–66.9
	300	–285.1	–268	–276.55	–17.1	–561.1	–484	–522.55	–77.1
	350	–285.1	–268	–276.55	–17.1	–559.4	–485.7	–522.55	–73.7
	400	–283.4	–266.3	–274.85	–17.1	–561	–482.3	–521.65	–78.7

Effect of pH of the Electrolyte

The electrochemical activity of the W–O framework of POMs has been reported to depend on the availability of protons.[25]Figure demonstrates the variation in the redox properties of the POM/AgNP-based multilayer assembly when the pH of contacting electrolyte solution was varied. A cathodic shift was observed in the peak potentials linked to the tungsten-oxo redox activity with the increase in alkaline behavior of the electrolyte (Figure A). Slope of the graphs from Figure B for two W–O redox processes are 81 and 85 mV per pH, respectively. Therefore, the number of protons can be calculated by using the following relationshipwhere ΔE/ΔpH represents the slope of the curve, m is the no. of protons to be calculated, and n is the no. of electrons participating in the respective electrochemical process.

Figure 6

(A) Cyclic voltammograms obtained from the multilayer assembly comprising 16 layers with the outermost layer of POM contacting aqueous buffer solutions of pH varying from 2 to 5. Scan rate was 100 mV s–1. (B) Graph of pH vs E1/2 for the second W–O electrochemical reaction.

From this equation, m′ was found to be 2′ for both redox processes, showing that each redox process requires not only two electrons but also two protons.

Permeability of the Fabricated Assembly toward Redox Probes

The permeability and porosity of the fabricated film was explored by evaluating the effect of the absence or presence of the film anchored onto the electrode’s surface on the electrochemical performance of a monoelectronic negatively charged redox probe, that is, [Fe(CN)6]3–. The CV response for [Fe (CN)6]3– exhibits a reversible, well-defined redox couple with an E1/2 value of 231 mV (vs Ag/AgCl reference electrode). Figure S4 demonstrates the impact of the total number of layers assembled on the electrode and the nature of the outer layer on the feasibility of the redox reaction of the reacting probe. It can be seen that with the electrodes modified with four bilayers, Fe(CN)6 can still percolate to the underlying electrode and displays some distorted electrochemical activity with higher peak to peak separation values (Figure S4A,B, curve iii) as compared to the bare electrode (curve i). However, this peak-to-peak separation value is high when the outermost layer is an anionic probe (Figure S4A), that is, 478 mV as compared to the film with the outer most layer of cationic AgNP where this value is 336 mV, indicating that the probe might face some electrostatic repulsive interaction while penetrating the multilayer assembly with the outer POM layer. However, the multilayer film with eight bilayers did not allow the Fe (CN)6 probe to diffuse through it as shown in Figure S4C where curve iii only shows the redox processes linked to the tungsten-oxo building blocks of the POM but no electrochemical reaction/activity of Fe (CN)6 probe could be detected.

Electrochemical Impedance Spectroscopy

The diffusional and the kinetic parameters of the redox active surface confined electrochemical systems can be investigated through impedance spectroscopy, which gives an indication about the degree of hindrance created by the redox active species to the diffusion of some electroactive species, that is, [Fe(CN)6]3–/4– toward the electrode. Figure shows the Nyquist graphs of impedance spectra of the Fe2+/Fe3+ redox process at the bare GCE as well as at various stages of the fabrication of Co-POM/AgNP nanoassembly with an applied potential of 230 mV (E1/2 of the ferri/ferro redox couple). The Randles equivalent circuit was exploited to interpret the Nyquist electrochemical impedance spectroscopy (EIS) plots. This circuit comprises double layer capacitance (Cdl) in parallel with diffusional parameters, also known as Warburg impedance (W) and charge transfer resistance (RCT) while solution resistance is in series. A constant phase element replaces Cdl in the circuit because of the unevenly distributed interfacial features at the surface of the electrode. What is perceived from the plots is the increase in the RCT value with the layers of poly(diallyl dimethyl ammonium chloride) (PDDA) and POM. However, an anomalous behavior can be observed with the anchoring of the AgNP layer with the decline in the RCT value. This decrease can probably be owing to various factors: (i) higher conductivity of the metallic nanoparticles facilitating the electron transportation between the electrode and redox probe, (ii) electrostatic attraction is the other most probable factor between the anionic probe and cationic nanoparticles. In contrast to this, RCT values show a continuous increase with each POM layer deposited owing to the repulsive interaction between the outermost anionic POM layer and anionic probe. Furthermore, a shifting of the redox process from purely diffusion controlled to a kinetically controlled process was observed as the number of layers increased.

Figure 7

Nyquist plots of the Co-POM/AgNP multilayer assembly with the increase in number of layers (a) cleaned GCE, (b) GCE/PDDA, (c) GCE/PDDA/Co-POM, (d) GCE/PDDA/POM/AgNP, and (e) GCE/PDDA/POM/AgNP/POM (10–1 to 102 KHz; signal amplitude = 5 mV; E = +230 mV; electrolyte: K4[Fe(CN)6]/K3[Fe(CN)6]/0.1 M KCl).

Electrocatalysis and Sensing of Carbofuran

The proposed platform was tested to assess its electrocatalytic and sensing capability for the detection of an extremely toxic CB pesticide, carbofuran. Owing to its high insecticidal activity, carbofuran has been widely used in agriculture; however, its residual products accumulate in the environment and impose serious threats to living beings. Therefore, a rapid, cost-effective, and sensitive method for an efficient on-site monitoring of these pesticides is the need of present times. The proposed Co-POM/AgNP/GCE sensor platform was found highly effective to detect the electro-oxidation of carbofuran. Figure A represents the electrocatalytic efficiency of the Co-POM/AgNP/GCE-based multilayer nanoassembly that consists of eight bilayers toward the oxidation of carbofuran. The amperometric response of the fabricated sensor was recorded in the absence and presence of varied concentrations of carbofuran using Britton–Robinson buffer pH 4.0 saturated with N2. It is evident from the figure that the current associated to the oxidation peak subsequently increased in intensity with the addition of high amounts of analyte (Figure A, curves c–e). However, the bare GCE and AgNP/GCE-modified electrode did not show any response toward the catalytic oxidation of the analyte under similar conditions (Figure A, curves a and b), which assured that CoPOM is responsible for the electro-oxidation of carbofuran. The quantitative catalytic performance of the fabricated assembly was evaluated by employing the following relationshipwhere I(POM + CBF) is the current of the oxidation peak generated in the presence of carbofuran and I(POM) is the current obtained without the addition of analyte. Therefore, the calculated catalytic competence was augmented from 49% for 0.1 mM carbofuran to 380% for 2 mM analyte. The inset represents a linear relation between the concentration of carbofuran and the corresponding catalytic oxidation current up to 2 mM and a correlation coefficient of 0.957. The detection limit (LOD) was 0.1 mM based on the slope and SD of lowest response and sensitivity was 13.11 μA mM–1. To further monitor whether the proposed nanoassembly follows the Michaelis–Menten behavior, the Lineweaver–Burk plot was formed between inverse of the concentration of carbofuran (1/Icat) and steady-state current response (1/CBF) (Figure B). The Kmapp was calculated from the slope and intercept of the 1/Icat vs 1/CBF double-reciprocal plot which was found to be 0.4 mM. The low Kmapp value suggests high catalytic affinity of Co-POM/AgNP/GCE toward carbofuran. The high analytical performance and ease of assembly of the proposed sensor accentuate its potential in practical applications for CB detection.

Interferences

The selectivity of the Co-POM/AgNP/GCE toward the carbofuran in the presence of various different interfering species was evaluated by employing the amperometric analysis. Figure explained that no oxidative catalytic response could be observed in the presence of 100 times more quantities of Mg2+, Na+, Ca2+, Al3+, SO3–, and NO3– and 10-fold of catechol. However, the system showed a noticeable amperometric response for each 10 μM addition of carbofuran. These results clearly explained that our sensing assembly possesses a distinct selectivity toward the detection of carbofuran.

Figure 9

Amperometric response of Co-POM/AgNP/GCE in 0.1 M PBS (pH 7.0) with the addition of (a) Mg2+, (b) Na+, (c) Ca2+, (d) SO3–, (e) NO3–, and (f) catechol.

Real Sample Analysis

Co-POM/AgNP-modified electrode was also tested for its real-time catalytic efficiency toward the detection of carbofuran in vegetables and fruits. Grapes, tomatoes, and spinach were used for this purpose. The sample spiking was done with definite/known concentrations of analyte by the standard addition procedure and recovery values were then calculated and are shown in Table . The calculated recovery values were ranging between 99.2 and 102%, showing that Co-POM/AgNP-based nanoassembly is a good sensor platform for the detection of carbofuran in real samples. Validity of this sensor was also confirmed by comparing the recovery values with the values obtained by the HPLC analysis. Table explains that the recovery values for all the three real samples obtained by exploiting the two methods, that is, HPLC and the fabricated sensor, are comparable, ranging from 99 to 102%. Table S1 describes the comparison of the average recoveries of the carbofuran in real samples with that of the values obtained by previously reported methods in the literature. It showed that the fabricated sensor exhibits very fast response toward carbofuran and shows better recovery values as compared to other techniques.

Table 2

Real-Time Detection of Carbofuran in Fruit and Vegetable Samples

	sensor method			HPLC method
sample	added (μM)	found (μM)	recovery %	added (μM)	found (μM)	recovery %
tomato	150	153	102	150	152	101
spinach	200	198	99	200	201	100.5
grapes	180	178.6	99.2	180	179	99.4

Repeatability and Reproducibility

To confirm the reliability and practicality of the proposed sensor interface, the operational reproducibility of fabricated electrodes was investigated under the same analytical parameters by taking five individual CV measurements of integrated electrodes in response to carbofuran. The relative standard deviation was less than 5.5% which indicates good reproducibility of the developed sensor. Moreover, the repeatability of the offered sensor was analyzed after the recurrent usage of the same electrode (n = 5), and the relative standard deviation was <3%, complying good reproducibility of the developed sensor.

Conclusions

In the present study, a highly efficient electrochemical nonenzymatic sensing film has been constructed utilizing Co-POM and PEI-capped AgNPs deposited on the surface of GCE through the LBL self-assembly technique. Cyclic voltammetry and AC-impedance techniques indicated a continuous growth of the hybrid multilayer nanoassembly. The fabricated multilayers were found to be reproducible and well organized. Furthermore, the nature of the outermost layer and the total number of layers deposited seemed to affect the porosity of the film. In addition, the surface morphological analysis assessed by SEM explained the globular features of the film. The presence of POM and AgNPs in the hybrid film was also confirmed by EDS. The fabricated surface exhibited enhanced electrocatalytic performance and acted as an efficient tool for the sensing of carbofuran electro-oxidation. This high-performance fabricated sensor exhibits novelty in terms of its design endowed by the choice of its individual components. Based on its outstanding performance, the presented design would take on promising possibilities in the real-time applications of electrochemical pesticide sensors for the monitoring of trace amounts of CBs in fields and agriculture.

Experimental Section

Chemicals and Reagents

PEI (MW 72,000), potassium ferrocyanide (K4[Fe(CN)6]·3H2O), silver nitrate (AgNO3), sodium tungstate (Na2WO4), PDDA (MW approx. 20,000), and potassium ferricyanide (K3[Fe(CN)6]) were purchased from Sigma Aldrich Corp. (USA). The chemicals were of analytical grade and used as it is. All the buffer solutions/electrolytes were made from highly purified water from ELGA PURELAB. The following electrolyte solutions were utilized for the electrochemical studies: 0.1 M Na2SO4 (pH 2.0, 2.5, and 3.0), 0.1 M Na2SO4/20 mM CH3COOH (pH 3.5, 4.0, 4.5, and 5.0), and 0.1 M Na2SO4/20 mM NaH2PO4 (pH 5.5 and 6.0). 0.1 M NaOH or 0.1 M H2SO4 was used to adjust the pH of respective buffer solutions. In addition, 1 mM solution of potassium ferrocyanide and potassium ferricyanide [1:1]/0.1 M KCl was used for electrochemical impedance experiments.

Instruments

The whole redox experimentation was carried out using potentiostat by Gamry interface 1010E electrochemical workstation by employing a tri-electrode scheme containing a glassy carbon working electrode (3 mm diameter) platinum wire as a counter and Ag/AgCl with 3 M KCl solution as a reference electrode. The working electrode was cleaned prior to use with a slurry of 1.0, 0.3, and 0.05 μm sized alumina powder, respectively, and further sonicated in deionized water for almost 5–10 min. Finally, the electrodes were dried with pure nitrogen stream before use. Impedance experiments were carried out on same instrument using 0.01 M potassium ferro/ferricyanide in 0.1 M KCl solutions (amplitude of 0.01 V, frequency 0.01 to 1 kHz, applied potential +230 mV). X-ray diffraction analysis was performed by using a Rigaku D/max-2550 diffractometer with a Cu Kα radiation source and intensities were measured at a diffraction angle ranged from 10 to 90° by 2θ angle. UV–visible characterization of the synthesized compounds and also the monitoring of multilayer assembly were performed by employing Lambda-35 spectrophotometer by PerkinElmer (America) with a range of 250–800 nm. FT-IR was carried out on a Nicolet 6700 spectrometer by Thermo Fischer Scientific and the scan range was set at 500–4000 cm–1. The surface features of the assembled films were evaluated by using SEM–EDS analysis. SEM was performed on a Nova Nano scanning electron microscope at an accelerating voltage of 3 keV. HPLC analysis was performed by using an Agilent Infinity 1220 HPLC system with C18 reverse phase column and a UV detector with manual injector and dual pump. The mobile phase was prepared from phosphate-buffered saline (PBS with pH 7.0) and acetonitrile in 40:60 ratio v/v. The flow rate was set up at 1 mL/min. Different concentrations of carbofuran standard were injected to the HPLC system. The calibration curve was then plotted between concentration and peak area. Recovery values were calculated for real samples, that is, spinach, grapes, and tomatoes by spiking them with 100 μM of carbofuran standard. These recovery values were then compared with the values obtained by using the POM/AgNP-based sensor.

Synthesis of Cobalt-Substituted Dawson-Type Polyoxometalate and PEI-Stabilized AgNPs

A cobalt-substituted Dawson-type polyoxometalate (POM), [P2W17O61 (Co2+·OH2)]7– was successfully synthesized according to the methods as described in the literature and characterized by electrochemical techniques.[39] Briefly explained; 25.5 g of K10P2W17O61·15H2O, synthesized by following the literature,[39] was dissolved in 100 mL of water (95 °C). Then, added a solution of 1.68 Co(NO3)2·6H2O in 20 mL of H2O while constant stirring to the abovementioned clear solution. After 15–20 min, added 15 g of KCl to the dark red solution. After cooling to ambient temperature, red crystals were obtained and re-crystallized twice from boiling water.

PEI-stabilized AgNPs were prepared by adopting the previously reported method.[36] Briefly, PEI-stabilized AgNPs were prepared by adding 50 mL of aqueous solution of 0.01 M AgNO3 in 1.5 mL of 2% polyethyleneimine (w/w) solution with constant stirring, followed by heating at 50 °C for 15–20 min. The formation of a brown colloidal suspension indicates the preparation of nanoparticles. This reaction is considered as a typical redox reaction, where PEI served as a reductive template to reduce the Ag+ ions. Polyethyleneimine is considered as a hyper-branched polymeric compound having haphazardly branched network topology and exhibits primary, secondary, and tertiary amine groups. In this reaction, numerous amino functional groups along different branches of polymer get oxidized by losing an electron present on their nitrogen atoms. These electrons are then picked up by the Ag+ ions to produce zero valent silver atoms, that is, Ag0. Here, PEI also acts as a stabilizer and controls the growth of AgNPs.

Fabrication of LBL Self-Assembly of Co-POM and PEI-AgNPs on GCE

First, the precleaned glassy carbon working electrode was modified with a base layer of polyelectrolyte (PDDA) to produce a homogeneously charged surface and to increase the charge density on the surface. It was done by dipping the substrate in 8% PDDA solution for half an hour. The electrode was then rinsed with DI H2O. Then, the PDDA modified electrode was immersed in 1 mM solution of POM for 20 min. After that, a multilayer was assembled by alternating dipping of the electrode in AgNPs and POM solution until the required number of coatings was deposited. The modified electrode was rinsed and dried carefully with a slow stream of nitrogen after every single deposition step. All the buffer solutions were purged with pure N2 gas to remove the atmospheric oxygen. The building up of the multilayer assembly was constantly monitored by cyclic voltammetry as well as electrochemical impedance spectroscopy.

Sample Preparation

The reliability of the Co-POM/AgNP/GCE was scrutinized by applying it toward the detection of carbofuran in real samples such as vegetables and fruits, for example, grapes, spinach, and tomatoes. The grapes, tomatoes, and spinach (obtained from the local market) were chopped into small pieces. 5 g of each of vegetable and fruit were mixed with 25 mL of 0.1 M PBS buffer with pH 7.0. The mixtures were blended completely to get homogenized. The homogenized pulp mixtures were then sonicated for 1 h and centrifuged. The supernatants were collected and diluted. Assessment of the carbofuran presence was executed by employing a standard addition method.