1. Introduction
Pesticides and herbicides are extensively used in agriculture, forestry, and domestic activities for controlling pests. This rapid increase in their use can cause a real threat to the environment and human health. So, a highly restricted control must be followed to avoid unacceptable levels of these contaminants from entering the water environment, thus influencing the food chain of humans and animals [
1].
Quats, a group of quaternary ammonium salts, is considered a particularly uneasy type of herbicide [
2], attributed to their physico-chemical properties, which lags the known multi-residue methods from their quantification. Diquat (1,1′-dimethyl-4,4′-bipyridilium dibromide) is one of the most widely used herbicides. It is used for controlling aquatic weed and pre-harvesting desiccation of potato vines, carrots, onions, vines etc., and seed crops (including rice, peas, clover, rape, beans, maize, etc.) [
3]. In addition, it holds the largest share of the global herbicide market until recently overtaken by glyphosate. This chemical type of herbicide—a bipyridyl—is shared with few other pesticides. Due to its high solubility (about 620 g/L at 25 °C), diquat (DQ) in fact a potential contaminant of waters [
4]. It also sticks tightly via its doubly charged cation to the mineral anions present in the soil sediments for long periods without leaching to the groundwater [
5]. The maximum contamination levels (MCL) for diquat in drinking water should not exceed 20 μg/L as stated by the United States Environmental Protection Agency (USEPA) [
6]. The repeated exposures to DQ may result in skin irritation, sensitization, or ulcerations [
7]. Chronic exposure may lead to cataract formation [
8]. To be mentioned, diquat products contain the carcinogen ethylene dibromide (EDB) as a trace impurity. The occupational exposure limit for EDB should not exceed 0.13 ppm during any 15-minute sampling period as recommended by The National Institute for Occupational Safety and Health (NIOSH) [
9]. So, in order to minimize risks associated with DQ use, there is an uprising need for a fast and reliable method for its quantification.
Different analytical methodologies have been reported in the literature for the determination of DQ different in real samples. These methods included voltammetric techniques [
10,
11,
12,
13,
14,
15,
16], spectrophotometry [
17,
18], spectrofluorimetry [
19,
20,
21], capillary electrophoresis (CE) [
22,
23], mass spectrometry coupled to either liquid chromatography [
24,
25,
26,
27] or HPLC [
28], and gas chromatography [
29]. Despite the fact of high sensitivity and selectivity of the above-mentioned analytical methods, the inconsistent results obtained suggested inconvenience with the chromatographic and/or mass spectrometric procedure for the quantification of DQ. To be noted, also is the sophisticated above-mentioned procedures, starting from extraction solvent composition, appropriate temperature, sample extract filtration for its pre-treatment for laboratory use, so as to obtain acceptable results. The ultimate goal is designing a simple, affordable, and easy to manipulate tool for DQ trace detection in real samples. In this context, potentiometric ISEs can be considered a good alternative because of their fast response, ease of automation, and applicability to turbid and colored matrices. As far, very few potentiometric sensors were reported for diquat monitoring [
30,
31,
32,
33].
Finding new designs for potentiometric ion-selective electrodes is the focus of attention of researchers working in this field. These new designs have the advantages of cost-effectiveness, ease of miniaturization and modification, robust and simple to be automated. Screen-printed platforms are examples of these new architectures. They are extensively used due their features such as no regeneration of the surface is required and highly reproducible geometric area for all electrodes. These interesting advantages can enhance the selectivity of the electrode and minimize any poisoning that can occur for the electrode surface [
15].
Solid contact screen printed potentiometric sensors are stepping forward at a remarkable pace for trace level detection. They have attracted great attention over the past decades due to their simple planar design (no internal filling solution), low cost, and low equipment requirements. Carbon nanomaterials as ion-to-electron transducers were used for the fabrication of a high-performance and long-life solid state ISEs [
34,
35,
36]. They are characterized by their high surface area, enhanced conductivity, and high ability to play the role of ion-to-electron transducer when mixed with ion sensing membrane or used as an electron conductor in solid state ISEs [
37].
Molecularly-imprinted polymers (MIPs) are selective artificial receptors for a wide range of different templated molecules. Integration of MIPs in the fabrication of potentiometric ISEs exhibit a great attention to shift the view of using non-affordable ionophores, which are limited by their high cost, or using ion exchangers, which afford poor selectivity. Man-tailored polymers are characterized by many advantages such as high thermal stability, ease of preparation, and cost- effectiveness [
38]. Recently, potentiometric ISEs based on MIPs as sensory elements have been fabricated for different templated organic molecules [
39,
40,
41].
In this work, we present miniaturized planar potentiometric ISEs modified with poly(3,4-ethylenedioxythiophene) (PEDOT) as solid-contact material for selective detection of diquat (DQ) herbicide. The addition of PEDOT/PSS into the diquat-selective membrane enhanced the hydrophobicity and capacitance with considerable potential stability, which was tested by electrochemical impedance spectroscopy (EIS) and constant-current chronopotentiometry techniques. For comparison, liquid-contact ISEs were also prepared and characterized, then compared with the solid-contact ISEs. The proposed ISEs revealed a high sensitivity and selectivity for potentiometric monitoring of diquat in soil samples and determination in commercial herbicidal formulations.
2. Materials and Methods
2.1. Chemicals and Reagents
Diquat dibromide monohydrate, paraquat dichloride dehydrate, mepiquat chloride, chlormequat chloride, ethylene glycol dimethacrylate (EGDMA), methacrylic acid (MAA), benzoyl peroxide (BPO), poly (3,4-ethylenedioxythiophene)/poly(styrenesulfonate) (PEDOT/PSS), sodium tetrakis [3,5 bis (trifluoromethyl) phenyl] borate (NaTFPB), and acetonitrile were obtained from (Sigma, St. Louis, MO, USA) and used as received. Dioctyl phthalate (DOP), high molecular weight poly (vinyl chloride) (PVC), were obtained from Fluka AG (Buchs, Switzerland). Tetrahydrofuran (THF) was freshly distilled prior to use. All chemicals were of analytical grade and were used without further purification. Bi-distilled de-ionized water (BDW) was used throughout the work. Reglone 200 SL used for pesticide technical formulation was purchased from Syngenta Company (Cairo, Egypt) to make a 31.8 w/w% soluble liquid (SL) of diquat dibromide.
A (10−2 M) stock solution of DQ was prepared by dissolving 0.362 g of pure diquat dibromide monohydrate in 100 mL distilled water. Diluents (10−2–10−8 M) of DQ were preserved in brown bottles.
2.2. Apparatus
All potentiometric measurements were carried out at ±25 °C using an Orion pH/mV meter (model SA 720, Cambridge, MA, USA). Selectivity measurements were carried out using the so called “modified separate solution method (MSSM)” [
42]. Fourier-transform infrared spectroscopy (FT-IR) measurements were carried out using FT-IR spectrometer (Alpha II, Bruker ABCO, Cairo, Egypt) using the attenuated total reflection (ATR) technique. Chronopotentiometry measurements of the screen-printed electrodes (SPE) were measured using Metrohom potentiostat/galvanostat (Autolab, model 204) purchased from Metrohom Instruments (Herisau, Switzerland). These tests were carried out in 10
−4 M diquat solution using a conventional three electrode system including an ISE working electrode, Ag/AgCl (3 M) as the reference electrode, and a Pt wire as the counter electrode.
2.3. Man-Taillored MIPs Synthesis
In a 25 mL glass capped vial, diquat (DQ) (temblate), methacrylic acid (MAA) (monomer), and ethylene glycol dimethacrylate (EGDMA) (cross-linker) (in the ratio of 0.5:3:3 mmol) were dissolved in the porogenic solvent acetonitrile (15 mL). Free radical initiator benzoyl peroxide (BPO, 60 mg) was added, followed by passing a flow of N2 gas in to the mixture for 10 min to remove any dissolved oxygen. The solution is then sonicated for further 10 min to ensure solution homogeneity. The glass-capped vial was then immersed in a constant temperature oil bath for 18 h, preset at 75 °C. The control non-imprinted polymer (NIP) was prepared in a similar way as mentioned above, without involvement of the template molecule. The MIP was rendered void of the template by means of soxhlet extraction using methanol/acetic acid (8/2, v/v) and methanol, ascertained by the zero absorption of diaquat using a Shimadzu UV/VIS spectrophotometer (Model UV-1601), and by the Fourier-transform infrared spectroscopy (FT-IR). A 24 h period at ambient temperature was a sufficient time for the polymer to dry.
2.4. Screen-Printed Design and Sensor Fabrication
The screen-printed platforms (5 × 30 mm) were designed and fabricated manually using a polyester sheet (~200 μm thick) as a substrate for electrode printing. Firstly, silver conductive ink was used to print the conducting track for the working electrode. Secondly, printing of carbon conductive ink at the end of the conducting track is done to form the sensing area of the working electrode. The polyester film, after the printing step, was heat cured at 150 °C for 15 min in a pre-heated oven. Finally, the electrodes were covered with an insulating tape leaving a rectangular area (3 × 3 mm) for defining the electrode sensing area as well as the connecter leads, forming a protective layer over the electrode tracking during analysis.
Figure 1 shows the final fabricated screen-printed platforms which were further used for electrochemical analysis. To the carbon screen-printed platform (C/SPE), 10 μL of PEDOT/PSS was added to the conductive carbon using drop casting method and left to dry. This acts as the solid contact between the ion-sensing membrane (ISM) and the carbon conductor. The sensing membrane was prepared by dissolving 100 mg of the components in 2 mL THF as follow: (6.0 wt %) MIP or NIP, (1.0 wt %) (NaTFPB), (31.2 wt %) PVC, and (61.3 wt %) DOP. Then, 15 µL of the membrane cockatiel was added via drop-casting over the PEDOT/PSS layer pertain the sensing membranes. The sensors were left to dry for at least 6 h. Prior to usage, a 2 h soaking in 10
−3 M DQ is convenient.
2.5. Diquat Determination in Real Samples
Analysis of real samples using the pre-designed screen-printed electrodes was checked towards their applicability as a diagnostic device for diquat monitoring in different commercial formulations and potato samples. Then, 10 g of a homogenized potato samples was placed into a 50 mL Teflon centrifuge tube. Fortified samples were prepared by spiking aqueous standards into the pre-weighed sample followed by 30 min equilibration. Then, 10 mL of 50:50 methanol/0.1 M HCl in water as an extraction solution was added to the sample. The mixture was shaken vigorously for 2 min by hand and heated in a water bath at 80 °C for 15 min. The supernatant was then analyzed using the proposed potentiometric method. Prior to HPLC analysis, the supernatant was filtered using a 45 micron PTFE syringe filter. A 500 μL portion of the filtered sample was diluted to 1.0 mL with acetonitrile prior to HPLC analysis.
In addition, DQ was analyzed in commercial agricultural formulations. Then, 0.5–1.0 mL of (Reglone 200 SL, Syngenta Company (Cairo, Egypt), 31.8 w/v% soluble liquid (SL) of diquat dibromide) formulations were diluted to 100-mL and their potential readings were used to look for their corresponding concentration along a calibration plot prepared from (10−2 to 10−7 M) standard diquat dibromide solutions.