Introduction

Organophosphate pesticides (OPs), a group of essential agricultural chemicals, are widely used to control pests and to increase harvest. As toxic compounds, they can disrupt the cholinesterase enzyme activity with prejudice to nervous system, leading to serious paralysis and even to death [1, 2]. Owing to their persistence, bioaccumulation, and toxicity, they may cause long-term damage to the environment and living species even if their concentration is very low [3]. The OPs present in the natural aquatic system or remaining in food potentially threaten the health of both animals and human beings. Therefore, it is of great significance to develop rapid, accurate and sensitive detection of these toxic reagents in the environment in order to cater for the increasing demand on environmental security, food safety and health protection. Several methods have been developed over the past few years for the determination of OPs, such as gas chromatography (GC) [4, 5], high performance liquid chromatography (HPLC) [6] and capillary electrophoresis (CE) [7]. However, these methods are always performed at centralized laboratories, requiring expensive instruments and/or long analysis time, which limit their application for on-site determination. Moreover, GC is inadequate to analyze polar, thermo-labile and low-volatility compounds, and most of HPLC require multi-step sample preparation and complicated procedures. Nevertheless, the electrochemical analysis possesses high sensitivity, good stability and low-cost instrumentation. In recent years, electroanalysis has gained an important place in the detection of pesticide residues due to its simplicity, high sensitivity and versatility. Enzyme-based biosensors together with electrochemical measurement have emerged as a promising alternative to detect pesticides. In order to give appropriate results within a short time under field conditions or on-line monitoring, various inhibition and non-inhibition biosensor systems, based on the immobilization of acetylcholinesterase or organophosphates pesticide hydrolyse onto various electrochemical transducers, have been extensively developed for pesticides analysis [8–12].

Recently, nanomaterials have been widely used in the fabrication of electrochemical sensors for their large surface area, nano porous structure, and high loading capacity [13–15]. OP compounds, including methyl parathion (MP), paraoxon, and fenitrothion, exhibit good redox activities at the electrodes based on some nanomaterials [16]. These favorable characteristics make the detection of different organophosphate pesticides possible, and several parathion electrochemical sensors based on nanoparticles have been applied to detect OPs [17–19]. As a group of representative one-dimensional nanomaterials, carbon nanotubes (CNTs) are extremely promising electrode materials because of their significant mechanical strength, excellent electrical conductivity, high surface area and good chemical stability [20, 21]. Many CNTs-modified electrodes have been widely used for electroanalysis of glucose [22–24], dopamine [25–27] and DNA [28].

In this study, Pd/MWCNTs nanocomposite-modified glassy carbon electrode was prepared for the detection of methyl parathion, which is one of the most widely used pesticides. The Pd/MWCNTs nanocomposite sensor shows high sensitivity for the determination of MP, which is promising for practical applications.

Experimental

Chemicals and apparatus

Methyl parathion was purchased from Dr. Ehrenstorfer GmbH Company (http://www.ehrenstorfer.com) and stock solution of 1.0 mg mL−1 was prepared by dissolving an appropriate amount of MP in acetone. 5% (w/w) Nafion was purchased from Arfa Aesar (http://www.alfa.com). MWCNTs were customer-made by catalytic chemical vapor deposition [29]. The 0.10 M phosphate buffer solutions of different pH values were prepared for use. All other reagents were of analytical grade and used without further purification. All solutions were prepared with high-quality deionized (DI) water (18.4 MΩ cm−1).

The characterization of Pd/MWCNTs was investigated by transmission electron microscope (Philips CM 300 FEG) and X-ray diffractometer (Shimadzu XD-3A). Electrochemical measurements were performed on a CHI-660C electrochemical workstation (Shanghai Chenhua, China) with a conventional three-electrode system. A glassy carbon electrode (3.0 mm in diameter) modified by MWCNTs or Pd/MWCNTs nanocomposite served as working electrode, with a platinum wire and an Ag/AgCl (3.0 M KCl) electrode as auxiliary and reference electrode, respectively. All measurements were performed at ambient temperature and the solutions were deoxygenated by purging with high-purity nitrogen.

Preparation of Pd/MWCNTs nanocomposite

MWCNTs were pretreated as previous report for functionalization [30]. The Pd/MWCNTs nanocomposite was prepared in an ethylene glycol (EG) solution by a reduction method. The detailed preparation process was as follows: 0.0520 g sodium citrate was added into 46 mL ethylene glycol with continuous stirring until the citrate was completely dissolved. After that, 2.45 mL PdCl2 solution (0.0250 M) (the mole ratio of PdCl2 and sodium citrate is 1:2) was added into the mixture and then stirred for 0.5 h. Then, 100 mg MWCNTs (metal loading: 10.0 wt%) was added and dispersed into the mixture under ultrasonic vibration for half an hour. The pH of the system was adjusted to 9.0 by dropwise addition of a 1.0 M KOH/EG solution with vigorous stirring. After 6 h of refluxing at 160°C in an oil bath, the resulting material was then centrifuged and the obtained slurry was washed with deionized water and ethanol, followed by drying at 80°C. Thus, the MWCNTs supported Pd nanoparticles were prepared.

Fabrication of Pd/MWCNTs electrode

Prior to modification, the basal GC electrode was polished to a mirror finish using 1.0 μm, 0.3 μm and 0.05 μm alumina slurries. After each polishing, the electrode was sonicated in ethanol and DI water for 5 min, successively, in order to remove any adsorbed substances on the electrode surface. Finally, it was dried under nitrogen atmosphere. A suspension was prepared by adding 5.0 mg Pd/MWCNTs to 2.0 mL 0.25% Nafion solution (prepared by diluting 5% Nafion solution with ethanol), and dispersed homogeneously by ultrasonication. Then, 8.0 μL of the ink-like suspension was dropped on a well-polished GC electrode and then dried under an infrared lamp.

Electrochemical experiments

The preliminary studies on the electrochemical behavior of the Pd/MWCNTs-modified electrode were performed by cyclic voltammetry (CV) in 0.10 M phosphate buffer solution. DPV measurements were performed to optimize the experimental conditions and determine the MP at room temperature in the potential range from 0.2 V to −0.3 V. The supporting electrolyte used in the experiments was thoroughly deoxygenated by bubbling high-purity nitrogen before each experiment.

Results and discussion

Characterization of the Pd/MWCNTs composite

The structure and phase analysis of the catalyst sample were performed by XRD. Figure 1a shows the XRD pattern of the Pd/MWCNTs catalyst with a metal loading of 10.0 wt%. The peak at 2θ values of 25.96° is the characteristic peak of MWCNTs. The other peaks are characteristic of face-centered cubic (fcc) crystalline Pd corresponding to the planes (1 1 1), (2 0 0), (2 2 0) at 2θ values of 40.1°, 46.7°, 68.1°, respectively. The micromorphology of the Pd/MWCNTs composite was investigated by TEM, as depicted in Fig. 1b. It can be seen that a large number of nanoparticles were loaded on MWCNTs with homogeneous distribution.

Fig. 1
figure 1

a XRD pattern and b TEM image of Pd/MWCNTs

Full size image

Electrochemical impedance spectroscopy can provide the information of capacitance and resistance of the electrode materials. It is an effective approach for investigating electron transfer across the electrolyte and the surface of the electrode. Figure 2 shows the impedance spectra of the MWCNTs and Pd/MWCNTs electrodes in 0.10 M KCl solution containing equimolar [Fe(CN)6]3−/4− at an ac frequency varying from 100 kHz to 0.01 Hz. The impedance spectra of the two electrodes are very similar. The Nyquist complex plane plot of the Pd/MWCNTs electrode exhibits an almost straight line which is the characteristic of a diffusion limiting step of the electrochemical reaction at the electrode. The result also reveals that the charge transfer resistance of the Pd/MWCNTs electrode is a little bit smaller than the MWCNTs-modified electrode. The result indicates that the Pd/MWCNTs electrode is highly conductive and the electron transfer at the interface of Pd/MWCNTs electrode is very fast.

Fig. 2
figure 2

Electrochemical impedance spectroscopy of the MWCNTs and Pd/MWCNTs electrodes in 0.10 M KCl solution containing equimolar [Fe(CN)6]3−/4− (0.01 M/0.01 M)

Full size image

Electrochemical behavior of MP

Figure 3 shows the cyclic voltammogram of MP on the Pd/MWCNTs electrode in 0.10 M phosphate buffer solution (pH 7.0) (curve A). The characteristic peaks in the region from −0.8 to −0.2 V are ascribed to the reversible adsorption–desorption of hydrogen on palladium. A pair of rather well-defined redox peaks (Epa1 = −0.038 V and Epc1 = −0.114 V) and an irreversible reduction peak (Epc2 = −0.642 V) are observed in the potential range from 0.4 to −0.8 V (curve A). The irreversible reduction peak corresponds to the reduction of nitro group to hydroxylamine group (Fig. 4, reaction 1), the reversible redox peaks are attributed to a two-electron transfer process (Fig. 4, reaction 2). The mechanism for this process has been widely reported [31, 32]. Another experiment was performed under the same conditions in the absence of methyl parathion; in which no redox peak is observed (Fig. 3, curve B). It can be concluded that the redox peaks correspond to the reaction of the methyl parathion. These results also indicate that MP exhibits high current response on the Pd/MWCNTs electrode in phosphate buffer solution. Several factors may attribute to these results. The high surface area of Pd/MWCNTs is certainly favorable for the adsorption and reduction of methyl parathion. Moreover, the excellent conductivity of Pd/MWCNTs also plays an important role for the fast electron transfer. Additionally, palladium nanoparticles have the characteristic feature of adsorption–desorption of hydrogen and the absorbed hydrogen on palladium is beneficial for reduction reaction since H+ is involved in the reduction process (Fig. 4).

Fig. 3
figure 3

Cyclic voltammograms of Pd/MWCNTs in 0.10 M phosphate buffer solution with (A) or without (B) 4.0 μg mL−1 methyl parathion. Scan rate: 100 mV s−1

Full size image
Fig. 4
figure 4

Mechanism of the electrochemical reaction of methyl parathion at the Pd/MWCNTs-modified electrode

Full size image

Effect of pH on the response of MP

The influence of pH on the peak potential and peak current of MP is displayed in Fig. 5. With pH rising, the reduction peak potential decreases, while stripping signal of the peak increases. A slope of 55.6 mV pH−1 suggests that the numbers of proton and electron involved in the reaction are equal, which is consistent well with the redox mechanism in Fig. 4 (reaction 2). This result also testifies a two-electron transfer in reaction 2. For the electrochemical reaction involves proton transfer, increasing concentration of H+ is favorable for the positive-shift of potential for the reduction of MP. This is consistent with the mechanism shown in Fig. 4. While H+ present in the phosphate buffer solution plays a dual role during the reduction steps, the decrease in peak current with lower pH could be due to the adsorption of H+ on the Pd, which competes for surface adsorption sites with MP and in turn inhibits the reduction of MP as well. Since basic media would result in the degradation of OP compounds and the excessively low pH such as 3.0 may lead to the analyzing hydrogen side reaction, which may affect the adsorption of MP at the Pd nanoparticles and cause the decrease of electrochemical signal. Based on the experimental results, pH 7.0 was selected for DPV measurements in this study.

Fig. 5
figure 5

Effect of pH on the electrochemical response of the sensor. The concentration of methyl parathion is 8.0 μg mL−1. DPV scanning potential range, 0.2 to −0.3 V

Full size image

Analytical performance for the detection of methyl parathion

Under the above optimal parameters, Pd/MWCNTs-modified electrode was used to detect methyl parathion by DPV. Figure 6 shows the DPV responses recorded in phosphate buffer solution (pH 7.0) containing different concentration of methyl parathion on the modified electrode. As can be seen, there is a good reduction peak at −0.068 V and the response currents enhance gradually with the increase of the methyl parathion concentration. The current responses are linear with methyl parathion ranging from 0.10 μg mL−1 to 14 μg mL−1. The linear regression equation I (μA) = 5.07 + 18.30c (μg mL−1) (r2 = 0.9996). A detection limit of 0.05 μg mL−1 was obtained with the calculation based on signal/noise of 3.

Fig. 6
figure 6

Variation of peak current with MP concentration. Concentration of MP: 0 μg mL−1, 0.10 μg mL−1, 0.20 μg mL−1, 2.0 μg mL−1, 4.0 μg mL−1, 6.0 μg mL−1, 8.0 μg mL−1, 10 μg mL−1, 12 μg mL−1 and 14 μg mL−1 (from bottom to top)

Full size image

To study the role of palladium, a contrast measurement was conducted using a MWCNTs-modified glassy carbon electrode to determine MP under the same conditions. As to the MWCNTs-modified electrode, the linear regression equation is I (μA) = 3.66 + 5.99 c (μg mL−1) (r2 = 0.9984). Compared to the MWCNTs, the Pd/MWCNTs-modified electrode exhibits higher current response to the same concentration of methyl parathion with a higher sensitivity of 18.30 μA μg−1 mL, which is about three times that of the MWCNTs-modified electrode. It demonstrates that palladium deposited on MWCNTs actually can enhance the response of MP on the modified electrode. This can be attributed to the excellent synergistic effect of palladium nanoparticles and MWCNTs.

Anti-interference property, stability and reproducibility of the sensor

Inorganic ions’ interferences on the detection of methyl parathion were also evaluated. The experiments were performed with 2.0 μg mL−1 methyl parathion in phosphate buffer solution (pH 7.0) in the absence or presence of 0.10 M NO −3 , 0.10 M SO 2−4 , 0.10 M PO 3−4 or 0.10 M Cl−, respectively. No obvious interferences were observed from the inorganic ions (Cl−, PO 3−4 , SO 2−4 and NO −3 ) with the peak currents of MP varied slightly. Ninety-two to one hundred four percent current responses for MP were obtained in the presence of the above anions. Furthermore, the relative standard deviation was 4.6% for ten replicate determinations of 2.0 μg mL−1 MP, indicating acceptable reproducibility. The stability of the Pd/MWCNTs-modified electrode can be maintained by being stored at 4°C in a dry condition. No obvious decrease in the response of MP was observed in the first 10-day storage. After a 30-day storage period, the sensor retained 86% of its initial current response.

Conclusions

An electrochemical sensor for organophosphates is introduced based on carbon nanotubes supported palladium nanoparticles-modified glassy carbon electrode. The method is convenient and reliable, and the electrode is promising for developing sensors for determination of organophosphates with high sensitivity. This study successfully demonstrates a simple and efficient strategy for stripping analysis of OPs at the Pd/MWCNTs-modified GCE. The Pd/MWCNTs electrode is able to extract nitroaromatic MP from the solution, and facilitate the preconcentration of MP onto the surface with the stripping current response greatly enhanced. The fast, sensitive and selective determination of MP was realized. The resulting sensor shows both good reproducibility and ideal stability.