“Metal-Free” Fluorescent Supramolecular Assemblies for Distinct Detection of Organophosphate/Organochlorine Pesticides

Abstract

The “metal-free”, easy-to-prepare fluorescent supramolecular assemblies based on anthracene/perylene bisamide (PBI) derivatives have been developed for the distinct detection of organophosphate (CPF) and organochlorine (DCN) pesticides in aqueous media. The supramolecular assemblies of anthracene derivative show rapid and highly selective “on–on” response toward organophosphate (CPF), which is attributed to the formation of CPF-induced formation of “closely packed” assemblies. A detection limit in the nanomolar range is observed for CPF. On the other hand, the inner filter effect is proposed as the mechanism for the “on–off” detection of DCN using supramolecular assemblies of the anthracene derivative. This is the first report on the development of fluorescent materials having the potential to differentiate between organophosphate and organochlorine pesticides. The assemblies of anthracene derivative 2 also act as “enzyme mimic” as organophosphate pesticide show a preferential affinity for assemblies of derivative 2 over acetylcholinesterase enzyme. Further, the real-time applications of supramolecular assemblies have also been explored for the detection of CPF and DCN in spiked water and in agricultural products such as grapes and apples.

Introduction

Pesticides are mostly employed for the control of pests to increase crop production. The unawareness about the requirement of the actual amount of pesticides needed as well as greed to increase crop production rapidly has led to an uncontrolled spray of pesticides. Despite the huge contribution of pesticides in increasing crop production, their slow degradation causes long time persistence under environmental conditions. The prolonged stay of pesticides in environment increases their chances to enter the food chain also. Furthermore, amounts of around 1% of applied pesticides are practically carried by pests, while the remaining impacts the soil or the water system. These pesticides can directly affect the living organisms through the soil, aquatic system, and food material.¹ Pesticides belonging to some categories such as organophosphates and organochlorines are especially very harmful. Organophosphates are also regarded as “nerve agents” as they irreversibly bind the catalytic site of acetylcholinesterase (AChE) and inhibit the catalytic activity of the enzyme to hydrolyze acetylcholine.² The increased concentration of acetylcholine causes harm to the central nervous system, thus, leading to dangerous diseases like paralysis, organ failure, etc. On the other hand, organochlorine pesticides are also known for their high toxicity and bioaccumulation due to the presence of chlorinated hydrocarbons. Hence, it is very important to develop a sensitive and rapid method to detect organochlorine and organophosphate pesticides in an aqueous medium, soil, and food items. In literature, a variety of approaches have been developed for the detection of pesticides such as gas chromatography, surface-enhanced Raman spectroscopy (SERS), enzyme-linked immunosorbent assay (ELISA), and high-performance liquid chromatography (HPLC), but all these methods require costly instrumentations and additional procedures.³⁻⁵ Recently, fluorescence-based simple probes are being explored for the detection of pesticides.^6,7 Due to their simplicity, facile applications, and low detection limits, these probes could overcome the problems associated with methods dependent on costly instrumentations. Very recently, metal–organic frameworks (MOFs)/carbon dots have been reported for the detection of pesticides in organic media.⁸ The sensing event is centered around interactions between metallic species and pesticides; however, due to moisture sensitivity of metallic centers, the methods are not very effective in an aqueous medium. In literature, a few “metal-free” materials have also been reported for the “on–on” detection of organophosphates, but these probes are reaction-based and sensing event is operative in organic/mixed aqueous media.^1,6,9 Since pesticides can very easily enter the water cycle, the development of materials for the detection of pesticides in an aqueous medium is necessary. Very recently, a reversible polymeric film has been reported for the colorimetric detection of organophosphates in aqueous media.¹⁰ Although the probe exhibited a naked-eye response toward pesticides, the detection response was not very sensitive. Thus, the development of an efficient, sensitive, and easy-to-handle probe for the detection of pesticides in aqueous media is a challenge. This area of research is relatively less explored, and great efforts are needed to understand various aspects of design principles of the development of fluorescent probes for the detection of pesticides. Our experience in the field of development of fluorescent assemblies for the detection of various biologically important analytes encouraged us to develop efficient probes for the detection of pesticides.^11,12 Keeping in mind the harmful effect of pesticides in general, for initial studies, we focused on the detection of chlorpyrifos (CPF, organophosphate) and 2,6-dichloro-4-nitroaniline (DCN, organochlorine pesticide) as the target analytes. We chose CPF and DCN because of their real-time utilization for the cultivation of crops such as grapes, peas, apples, etc. We envisaged that an electron-rich scaffold may be advantageous for designing a sensitive probe for electron-deficient CPF and DCN. To start with, we planned to utilize commercially available anthracene derivative 1 for the studies. Unfortunately, derivative 1 showed poor water compatibility. To make it water compatible, we planned to functionalize it by introducing formyl phenyl groups at the periphery.

We expected that the presence of aldehyde groups besides providing a balance between polar and nonpolar groups may also provide additional sites for the interaction of probe molecules with pesticides through noncovalent interactions such as H-bonding. We further expected that due to the hydrophobic effect in an aqueous medium, the anthracene derivative may undergo self-assembly to generate fluorescent aggregates. Amazingly, the supramolecular fluorescent assemblies of anthracene derivative 2 exhibit an on–on response toward CPF and an “on–off” response toward DCN (Figure 1). Unprecedented, the supramolecular assemblies of derivative 2 exhibit a CPF-induced emission enhancement due to the formation of closely packed aggregates, while the combined influence of the “inner filter effect” and energy-transfer processes is responsible for sensing response in the presence of DCN. To the best of our knowledge, this is the first report on the differential detection of organophosphate/organochlorine pesticides. To understand the role of different structural features to be encoded in the designed building blocks, we also developed assemblies of derivatives 3, 4, and 5 based on an electron-rich scaffold (hexaphenylbenzene derivative) and electron-deficient scaffold (perylene bisamide derivatives) for the detection of DCN and CPF. Interestingly, all of the derivatives show sensitive and rapid response toward CPF in an aqueous medium; however, perylene bisamide (PBI) derivatives exhibit a weak response toward DCN. In this study, an effort has been made to understand essential structural features to be encoded in the designed molecules for sensitive detection of pesticides in aqueous media (Figure 2). Additionally, the present work demonstrates the real-time applications of assemblies of derivative 2 to detect different concentrations of pesticides in various agricultural products such as grapes and apples very efficiently. Furthermore, the potential of assemblies of derivative 2 to act as “AChE mimic” has also been demonstrated. In a competitive experiment, CPF exhibits preferential affinity toward assemblies of derivative 2 over the AChE. In comparison to other reported fluorescent probes in nature, the supramolecular assemblies reported in this manuscript show a better “turn-on” response (Table S1, Supporting Information).

Figure 1.

Schematic representation highlighting the advantages of supramolecular assemblies for the detection of organophosphate/organochlorine pesticides.

Figure 2.

Structures of derivatives used for the detection of pesticides.

Results and Discussion

Derivative 2 was prepared by the Suzuki–Miyaura coupling between 9,10-dibromoanthrane and 4-formyl-phenyl boronic acid using the standard procedure¹³ (Scheme S1, Supporting Information). The absorption and emission behavior of derivative 2 was examined in tetrahydrofuran (THF) and in different fractions of THF/water solvent mixtures. The absorption spectrum of derivative 2 (10 μM) in THF shows three absorption bands at 395, 375, and 358 nm, which shift slightly toward large wavelength (∼7 nm) upon adding ∼90% water fraction and are broadened with the decrease in their absorption intensity. Further, the absorption spectra rose from the axis and the tail appeared in the red region, which indicates the formation of aggregates of derivative 2 (Figure S5, Supporting Information). The THF solution of probe 2 (10 μM) exhibits an emission band at 451 nm in the fluorescence spectrum when excited at 380 nm. Upon the addition of ∼90% water, the emission band is red-shifted to 475 nm with the quenching of the emission band (∼96%) (Figure S6, Supporting Information). The concentration-dependent ¹H nuclear magnetic resonance (NMR) studies of derivative 2 show a slight upfield shift of aromatic signals (Figure S7, Supporting Information). On the basis of UV–vis, fluorescence, and concentration-dependent NMR studies, we believe that derivative 2 undergoes self-assembly to form J-aggregates in aqueous media.

Next, we investigated the affinity of assemblies of derivative 2 (10 μM) toward different metal ions as their perchlorate salts (Fe²⁺, Cu²⁺, Co²⁺, Ni²⁺, Zn²⁺, Ag⁺, and Al³⁺ ions), biomolecules (spermine, spermidine, glutathione, cysteine, homocysteine, and hydrazine), reactive oxygen species (H₂O₂ and ClO^–), and amines (aniline, triethylamine, diethylamine, and p-nitroaniline), etc. Among various metal ions, biomolecules, and reactive oxygen species examined, derivative 2 shows an affinity for Cu²⁺ ions in an aqueous medium. In the UV–vis spectra, with the addition of Cu²⁺ ions (0–150 equiv) to the 90% aqueous solution of derivative 2, bands corresponding to copper oxide nanoparticles appeared at 280 and 757 nm (Figure S8, Supporting Information). The formation of copper oxide nanoparticles was further confirmed from X-ray photoelectron spectroscopy (XPS) studies, which shows peaks at 934, 942, 955, and 962 eV corresponding to mixed copper oxide nanoparticles^14,15 (Figure S9, Supporting Information). In the fluorescence spectra, upon the addition of Cu²⁺ ions (0–150 equiv), the emission band is slightly red-shifted from 475 to 478 nm with a decrease in its intensity (∼59%) (Figure S10, Supporting Information). However, under the same set of conditions as used for Cu²⁺ ions, no significant change in the photophysical behavior of assemblies of derivative 2 was observed in the presence of other metal ions, biomolecules, amines, and reactive oxygen species (Figures S11–S13, Supporting Information). These studies indicate that the assemblies of derivative 2 are selective for Cu²⁺ ions.

Afterward, we examined the absorption and emission behavior of assemblies of derivative 2 toward different pesticides like chlorpyrifos (CPF), 2,6-dichloro-4-nitroaniline (DCN), glyphosate, dichlorvos, bisphenol A (BPA), and 2,4-dichlorophenol (DCP) (Figure 3).

Figure 3.

Chemical structures of different pesticides investigated in our study.

Upon increased addition of an aqueous solution of CPF (0–100 equiv) to the 90% aqueous solution of derivative 2, the UV–vis spectra are broadened with an increase in the intensity of the bands. Further, a tail was observed in the red region and the spectra rose from the axis (Figure S14, Supporting Information), whereas in the fluorescence studies, upon addition of CPF (0–100 equiv) to the solution of derivative 2 (10 μM), the emission band is blue-shifted to 451 nm with ∼2.5-fold increase in the fluorescence intensity (Figure 4a). The CIE coordinates for the aqueous solution of assemblies of derivative 2 are found to be 0.182, 0.314, which correspond to greenish blue color, while with CPF the values change to 0.165, 0.166, corresponding to light blue color (Figure S15, Supporting Information); hence, the emission change is clearly visible to the naked eye. Anthracene derivatives are known to exhibit aggregation-induced quenching;¹⁶ however, the emission studies indicate CPF-induced emission enhancement. We believe that in the presence of CPF, molecules of derivative 2 are organized to generate closely packed fluorescent aggregates. The detection limit of assemblies of derivative 2 for CPF was found to be 0.33 nM (Figure S16, Supporting Information). Since literature reports suggest a potential for metal-based materials for the detection of pesticides,¹⁷ we examined the affinity of supramolecular copper oxide ensemble (as-obtained solution of derivative 2 and Cu²⁺ ions in the fluorescence studies) toward CPF. Upon the addition of CPF (0–225 equiv) to 90% aqueous solution of a supramolecular copper ensemble of derivative 2, the emission band is blue-shifted to 459 nm with ∼2.5-fold increase in its intensity (Figure S17, Supporting Information). In the case of a supramolecular copper ensemble, the detection limit for CPF was found to be 3.85 μM (Figure S18, Supporting Information). We believe that a high detection limit observed in the case of the supramolecular copper ensemble could be attributed to the inability of CPF molecules to organize in a closely packed arrangement. Thus, metal-free assemblies of derivative 2 show a better response toward CPF.

Figure 4.

Fluorescence studies of assemblies of derivative 2 on the addition of (a) CPF (100 equiv) and (b) DCN (150 equiv) in 90% water (λ_ex = 380 nm and slit width = 3–3).

Among various pesticides examined, assemblies of derivative 2 also show a response toward DCN. In the absorption studies, with increased addition of DCN (10^–2 M in ethanol, 0–150 equiv) to 90% aqueous solution of derivative 2, a broad band at 365 nm corresponding to the absorption of DCN is observed (Figure S19, Supporting Information). In the fluorescence studies, upon adding a solution of DCN (0–150 equiv, 10^–2 M solution in ethanol) to the solution of derivative 2, the emission band is blue-shifted from 475 to 460 nm and emission intensity is fully quenched (Figure 4b). The masking of absorption spectra of derivative 2 with the absorption of DCN prompted us to verify the possibility of an inner filter effect as the sensing mechanism. To examine the inner filter effect, the solution of derivative 2 and DCN was irradiated at different excitation wavelengths (ranging from 310, 320, 330, 340, 350, 360, 370, 380, 390, to 400 nm). On changing excitation wavelength ranging from 310 to 400 nm, a change in the emission intensity was observed; however, λ_max (475 nm) remained the same, which supports the inner filter effect in the system (Figure S20, Supporting Information). The detection limit of the assemblies of derivative 2 for DCN was found to be 0.33 × 10^–11 M (Figure S21, Supporting Information). Under the same experimental conditions as used for DCN, we examined the changes in photophysical behavior of derivative 2 in the presence of other pesticides such as glyphosate, dichlorvos, bisphenol A (BPA), and 2,4-dichlorophenol (DCP). In the fluorescence studies, upon introducing a solution of DCP (0–150 equiv, 10^–2 M solution in water) to the solution of derivative 2, a 81% decrease in emission intensity is observed, which may be attributed to the electron-deficient nature of the DCP scaffold. However, no significant change is observed in all of the studies (Figure S22, Supporting Information).

Next, we investigated the decay time of the assemblies of derivative 2 in the absence and in the presence of CPF and DCN in aqueous medium using time-resolved fluorescence spectroscopy. The 90% aqueous solution of assemblies of derivative 2 show an average lifetime of 0.421 ns (29%, τ₁ = 1.27 ns, 54.4%, τ₂ = 0.25 ns, 15.9%, τ₃ = 5.23 ns). However, upon the addition of CPF (100 equiv), the average lifetime of the assemblies of derivative 2 increases to 2.10 ns (8.17%, τ₁ = 0.91 ns, 89.4%, τ₂ = 3 ns, 2.4%, τ₃ = 0.27 ns), which suggests the formation of more fluorescent aggregates in the presence of CPF (Figure S23, Supporting Information).

To understand the mechanism of the interaction of derivative 2 with CPF, we carried out ¹H NMR studies of derivative 2 in dimethyl sulfoxide (DMSO)-d₆-D₂O in the presence of CPF. Upon the addition of 10 μL solution of CPF (in DMSO-d₆ + D₂O, in 9:1 respectively) to a solution of derivative 2 (DMSO-d₆ + D₂O), a slight upfield shift in the aromatic protons as well as aldehyde protons is observed; however, a downfield shift in the position of signals of CPF is observed (Figure S24, Supporting Information). Unfortunately, upon increasing the D₂O content in DMSO-d₆ beyond this ratio (DMSO-d₆/D₂O, 9:1), the solution becomes cloudy and a well-resolved ¹H NMR spectrum could not be recorded. We believe that the percentage of water as a cosolvent in the solvent mixture is not sufficient to get the desired information. To understand the importance of water as a cosolvent, we also recorded the ¹H NMR spectrum of derivative 2 in the presence of CPF in CDCl₃. As expected, no change in the chemical shifts of aromatic protons as well as aldehyde protons is observed, which confirms our assumption. The powder X-ray diffraction (XRD) analysis of the assemblies of derivative 2 shows sharp peaks, indicating the crystalline nature of the molecule; however, in the presence of CPF, crystallinity was moderately disrupted and therefore a semicrystalline morphology was observed^18,19 (Figure S25, Supporting Information). The transmission electron microscopy (TEM) analysis of derivative 2 in the presence of CPF confirms the transformation of assemblies from a sheetlike structure to random-shaped aggregates (Figure 5). Further, dynamic light scattering (DLS) studies of derivative 2 in the presence of CPF confirmed the decrease in the size of aggregates from 365 to 231 nm (Figures S26 and S27, Supporting Information), which support our assumption regarding the formation of closely packed assemblies. On the basis of all of the above experimental studies, we believe that CPF molecules interact with the molecules of derivative 2 through π–π stacking, resulting in the generation of closely packed fluorescent aggregates of smaller size (Figure 5b).

Figure 5.

(a) TEM image of a solution of derivative 2 in aqueous media. (b) Schematic representation of the assemblies of derivative 2 in the presence of CPF. (c) TEM image of a solution of derivative 2 in the presence of CPF.

On the other hand, in time-resolved fluorescence studies, in the presence of DCN (150 equiv), no significant change in the average lifetime of assemblies of derivative 2 is observed, which suggests the static quenching of molecules of derivative 2 in the presence of DCN (Figure S28, Supporting Information). In powder XRD studies, the crystalline nature of derivative 2 is retained in the presence of DCN (Figure S29, Supporting Information). An increase in the size of aggregates of derivative 2 is observed in the presence of DCN from 365 to 542 nm (Figure S30, Supporting Information). Further, the efficient spectral overlap is observed between the absorption spectrum of DCN and the emission spectrum of the assemblies of derivative 2, which supports the possibility of an energy-transfer mechanism between the two (Figure S31, Supporting Information). On the basis of various absorption and fluorescent studies, we believe that the assemblies of derivative 2 show distinct responses toward DCN due to the inner filter effect (vide supra) and energy-transfer pathway. Further, to confirm the inner filter effect as the mechanism, we synthesized derivative 3 using perylene bisamide (PBI) as the scaffold. As PBI derivatives are known to show absorption in the region of 510–570 nm and emission in the region of 580–670 nm,²⁰ due to huge differences between the absorption of DCN (365 nm) and absorption/emission wavelengths of PBI derivative (vide infra), the possibility of operation of the inner filter effect as a mechanistic pathway is ruled out. The assemblies of the PBI derivative show absorption at 562 nm and emission at 668 nm (Figure S32, Supporting Information). As per expectations, PBI derivative 3 shows slight/no response toward DCN under the same set of experimental conditions, which clearly supports our assumption regarding the inner filter effect as the possible detection mechanism in case of supramolecular assemblies of anthracene derivative 2 (Figure S33, Supporting Information). Interestingly, assemblies of the PBI derivative show a sensitive response toward CPF (Figure S34, Supporting Information) with a detection limit in the range of 0.49 nM (Figure S35, Supporting Information). To understand the role of formyl phenyl groups in the detection process, we also synthesized derivative 4. Due to a balance between polar/nonpolar groups in the molecule, derivative 4 shows good water compatibility. The assemblies of derivative 4 show a sensitive response toward CPF (Figure S36, Supporting Information) with a detection limit of 0.11 nM, which is almost similar to that of derivative 3 (Figure S37, Supporting Information). On the basis of these results, we believe that formyl phenyl groups are not playing any significant role in facilitating the CPF-induced closely packed arrangement of molecules. To further understand the role of electronic nature and rigidity of the scaffold in the molecular recognition event, we also prepared derivative 5 based on a flexible hexaphenylbenzene scaffold. Derivative 5 forms fluorescent aggregates in aqueous medium and shows response toward CPF and DCN with a detection limit of 0.10 × 10^–7 M and 0.18 nM, respectively (Figures S39–S42, Supporting Information). Very clearly, the response toward DCN was more efficient in comparison to that of CPF. For DCN detection, we propose an inner filter effect and energy-transfer pathway as the mechanistic route as an efficient overlap between the absorption spectra of DCN and the emission spectrum of derivative 5 was observed. Despite the presence of an electron-rich HPB scaffold, the detection limit of assemblies of derivative 5 for CPF was high in comparison to those of derivatives 2–4. On the basis of these experimental results, we believe that rigid, polymeric scaffolds are more suitable platforms for the detection of CPF due to the possibility of attaining pronounced π–π stacking between host assemblies and analyte (CPF) molecules.

Organophosphates are known to inhibit the catalytic activity of acetylcholinesterase enzyme (AChE) to hydrolyze acetylcholine chloride to choline and acetate, which results in the accumulation of acetylcholine in living systems.²¹ To protect the living organisms from harmful effects of accumulated acetylcholine, the development of “enzyme mimic”, which could bind strongly with organophosphates in the presence of enzyme is highly needed. To understand the binding preference of CPF (an organophosphate pesticide) between assemblies of derivative 2 and AChE, we performed a competitive experiment. In the fluorescence studies, an aqueous solution of AChE (solution in phosphate-buffered saline (PBS) buffer, 0–20 equiv) was introduced to the solution of derivative 2 (300 μL from 10 μM solution of derivative) in the PBS buffer. The fluorescence intensity of the emission band at 475 nm was slightly decreased. Upon the addition of aqueous solution of CPF (0–100 equiv) to the same solution, a blue shift from 475 to 459 nm is observed and the emission intensity of the band at 475 nm is increased (Figure 6).

Figure 6.

(a) Fluorescence studies of derivative 2 in the presence of AChE and CPF in PBS buffer (λ_ex = 380 nm, slit = 3–3). (b) Schematic representation showing competitive affinity of CPF toward assemblies of derivative 2.

Thus, even in the presence of acetylcholinesterase, the CPF-induced emission behavior of assemblies of derivative 2 remains unchanged. These studies clearly show the preferred affinity of organophosphates toward the assemblies of derivative 2 over acetylcholinesterase.

For real-time applications of derivative 2 for the detection of CPF, we monitored the concentration of CPF in apple using assemblies of derivative 2. We prepared two sets of samples. In set A, we spiked the tap water samples with different concentrations of CPF (0, 10, 30, 50, 70, and 100 μL) as the control. Afterward, we prepared different samples of set B containing juices of apple (Section S1.4, Supporting Information) and spiked them with different concentrations (0, 10, 30, 50, 70, and 100 μL) of an aqueous solution of CPF. The photophysical changes observed in the inhibition plots of CPF in set A (tap water) were comparable to those in set B (apples) (Figure S43a, Supporting Information). We also examined the potential of assemblies of derivative 2 to detect DCN under real-time conditions using grapes as an agricultural product by following the same procedure. Similar fluorescence results were obtained upon comparing the inhibition plots of DCN in distilled water with those in spiked grapes juice (Figure S43b, Supporting Information). These results indicate that derivative 2 also has the potential to detect pesticides in real samples. Due to the high absorption capacity of pesticides on the skins of agricultural products, their residues in fruits/vegetables also cause serious health problems. The potential of assemblies of derivative 2 is also used to monitor the residues of DCN in grapes juice. For this, a standard solution of DCN was sprayed on the skins of four units of grapes. After washing the grape skin (single unit each day) with water, DCN residues collected in water samples were examined daily and the fluorescence studies of the as-obtained aqueous solution were performed for 4 consecutive days. From these studies, it was observed that DCN residues in grapes remained for at least 4 days and assemblies of derivative 2 show excellent potential to detect the presence of residues of DCN in grapes (Figure S44, Supporting Information).

Conclusions

Metal-free fluorescent supramolecular assemblies using electron-rich and electron-deficient scaffolds have been developed for the detection of organophosphate/organochlorine pesticides. These assemblies show an on–on response toward CPF and an on–off response toward DCN in an aqueous medium with different sensing pathways. Very interestingly, organophosphate (CPF) shows a great affinity toward assemblies of anthracene derivative even in the presence of acetylcholinesterase enzyme (AChE), thus confirming the potential of assemblies of derivative 2 to act as an AChE mimic. The potential of assemblies of derivative 2 to detect CPF and DCN in agriculture products and their residues in a real sample has also been demonstrated. The studies presented in this manuscript highlight the importance of rigid polyaromatic scaffold, electronic nature, and water compatibility of host assemblies for the detection of organophosphates via efficient stacking of aromatic rings of host as well as guest molecules in aqueous media. For designing probes for DCN detection, the present study demonstrates the efficiency of the inner filter effect as a simple and easy-to-attain mechanistic approach.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c02315.

• ¹H NMR, ¹³C NMR, and ESI-MS of derivatives 2 and 3; UV–vis and fluorescence studies; powder XRD; DLS; and XPS studies: CIE coordinates and tables of comparison of present manuscript with previous reports (PDF)

The authors declare no competing financial interest.