Facile destruction of formulated chlorpyrifos through green oxidation catalysis

Abstract

The organophosphorus (OP) insecticide, chlorpyrifos (CP, O,O-diethyl-O-3,5,6- trichloro-2-pyridyl phosphorothioate) in an emulsifiable concentrate formulation (CP-EC) is totally degraded in water by hydrogen peroxide catalytically activated by the TAML activator (1), to a combination of small aliphatic acids and minerals. CP-EC rapidly forms an oil-in-water emulsion when added to water. The CP in this emulsion is more resistant to oxidation than pure CP in aqueous solution. A one-pot, two-step process consisting of perhydrolysis followed by 1/H₂O₂ oxidation achieved total degradation of CP in the emulsion. In the first step, emulsified CP was hydrolyzed by H₂O₂ at high pH to induce the release of 3,5,6-trichloropyridin-2-ol (TCPy). The cationic surfactants CTAB or CTAC accelerated this hydrolysis. Addition of tert-butanol or ethanol also enhanced the hydrolysis rate. Xylenes serving as the solvent in CP-EC were shown to be the cause of the impeded hydrolysis. In the second step, the CP-EC hydrolysate was treated with 1/H₂O₂ to readily degrade the TCPy. In water, TCPy exists in the enol and not the keto form, which was found to facilitate its rapid oxidation. This degradation procedure produced a 72.5-fold reduction in toxicity (Microtox® assays) of the treated reaction mixture, compared to the untreated CP-EC emulsion. This ambient catalytic process establishes a promising line of investigation for alleviating the decades-old problem of obsolete thiophosphate pesticides.

---

Introduction

Worldwide, organophosphorus insecticides (OPIs) account for an estimated 38% of insecticide sales¹ and contribute significantly to about 500 000 tonnes of unusable obsolete pesticides.² Of this, 120 000 tonnes reside in Africa in millions of containers threatening the health of both rural and urban populations.² In resource poor circumstances, any real-world degradation system must be safe, inexpensive, simple to use, and adaptable for formulations as well as for the free active ingredients and the toxic byproducts that form upon aging or leaking into water or soil. Excellent noncombustion studies of formulated OPI destruction involve photolytic,^3,4 photo-Fenton,⁵ enzymatic,⁶ and electrochemical⁷ approaches, but the process details of each are not suitable for concentrated obsolete formulations or for the infrastructural conditions typically found in underdeveloped areas.

TAML activator catalysts (1, Fig. 1) are small molecule mimics of peroxidase and short-circuited P450 enzymes.^8,9 These catalysts function effectively at concentrations in the range of parts-per-billion to low parts-per-million at room temperature over a wide pH range (∼3 to 14). They have shown considerable promise in degrading hazardous environmental pollutants including toxic polychlorophenols,¹⁰ thiophosphate pesticides,¹¹ azo dyes,^12,13 dibenzothiophenes,¹⁴ natural and synthetic estrogens,¹⁵ the active pharmaceutical agent sertraline,¹⁶ and bacterial spores.¹⁷

Fig. 1 TAML activator (1) used in this work, chlorpyrifos (CP) and 3,5,6-trichloropyridin-2-ol (TCPy).

We have previously reported that TAML activators catalyze the rapid, total degradation of pure OPIs in water by hydrogen peroxide.¹¹ However, pesticides are primarily marketed and applied as formulations to improve the efficacy on the target, the safety of handling, and to protect the active ingredient (AI) from oxidation and hydrolysis.^18,19 Thus, any attempt to solve the decades-old problem of obsolete OP pesticides requires that pesticide formulations be studied. Here we present a case study demonstrating highly effective degradation of a representative formulated pesticide by examining chlorpyrifos, O,O-diethyl-O-3,5,6-trichloro-2-pyridyl phosphorothioate (CP, Fig. 1), in an emulsifiable concentrate (EC).

This CP-EC test system was chosen for two reasons. First, CP is a particularly significant OPI degradation target with 10 million pounds of the active ingredient applied annually in US agriculture.²⁰ It is associated with DNA damage in human sperm,²¹ with adverse developmental effects,²² and it is implicated in ADHD.²³ The residential use of CP was discontinued in the US in 2001, but CP-EC is still widely applied to corn and other crops.²⁴ Second, emulsifiable concentrate formulations have been very popular for many years and represent the biggest volume of all pesticide formulations deployed world-wide.¹⁸ In this report, we describe a systematic study of CP-EC degradation by 1/H₂O₂ that includes an evaluation of the impacts of the formulation adjuvants on the degradation process. Under ambient conditions, the following one-pot, two-step process results in the total degradation of the CP and its hydrolysate 3,5,6-trichloropyridin-2-ol (TCPy) leading to a significant reduction in toxicity.

Materials and methods

Chemicals

Pure CP [O,O-diethyl-O-3,5,6-trichloro-2-pyridyl phosphorothioate], Chlorpyrifos oxon (CPO, O,O-diethyl-O-3,5,6-trichloro-2-pyridyl phosphate), and TCPy (3,5,6-trichloropyridin-2-ol) were purchased from Chem Services Inc. (West Chester, PA). CP-EC was a “Pestban” brand formulation produced by Marman USA, Inc. containing 42.8% w/w chlorpyrifos active ingredient and unspecified formulating ingredients. Chloromaleic acid was synthesized by hydrolyzing chloromaleic anhydride (Wako Chemicals). Catalase (from Aspergillus niger, 2350 U mg⁻¹) was obtained from Sigma-Aldrich. Hydrogen peroxide (30% w/w) was purchased from Fluka. All other reagents and solvents (at least ACS reagent grade) were obtained from commercial sources (Aldrich, Fisher, Acros, and Fluka). The TAML activator (1) was synthesized following the published method.²⁵ Hydrogen peroxide stock solutions were standardized spectrophotometrically daily (Extinction Coefficient at 230 nm = 72.4 M⁻¹ cm⁻¹).²⁶ Phosphate buffer (0.1 M) was used as the reaction medium. The pH of the reaction mixtures were adjusted by addition of potassium hydroxide (∼10 M) or hydrochloric acid (∼6 M) solutions. Stock solutions of CP (10 mM), CPO (10 mM), TCPy (10 mM), xylenes (10 mM), pyridin-2-ol (5 mM), and pyridin-3-ol (5 mM) were prepared in tert-butanol. Stock solutions of the surfactants, cetyltrimethylammonium chloride (CTAC), cetyltrimethylammonium bromide (CTAB), sodium dodecyl sulfate (SDS), and Triton X-100 were prepared in HPLC grade water and were sonicated for ∼30 min prior to use in a Bransonic® tabletop ultrasonic bath (Model 1510). The concentration of CP in CP-EC (42.8% w/w) was determined to be 1.25 M by HPLC, based on a calibration curve produced using pure CP. The stock solution of 1 (1 mM) was prepared in HPLC grade water. Freshly prepared stock solutions of catalase (23500 U mL⁻¹) in HPLC grade water were used to quench the reactions by decomposing unused H₂O₂ in the reaction mixture.

Instrumentation and analysis

A Waters HPLC system consisting of a Waters 600 controller, 2996 Photodiode Array Detector and 717 plus autosampler was used for monitoring both the hydrolysis of CP-EC and the oxidation of TCPy. Chromatographic separation of CP, CPO, TCPy and xylenes was achieved using a Varian microsorb-MV 100-5 CN (150 mm × 4.6 mm, 5 μM) column. The mobile phase consisted of acetonitrile (40%) and water containing 0.2% acetic acid (60%), which was first filtered through a Millicup vacuum-driven bottle-top filter unit. The HPLC analyses parameters were as follows: injection volume, 20 μL, column temperature, 30 °C, and flow rate, 1 mL min⁻¹. The retention times of CP, CPO, TCPy and xylenes under these conditions were 10.5, 5, 3.7, and 4.5 min, respectively. The limit of quantification (LOQ) for CP under these conditions (300 nm) was 0.5 μM. The hydrolysate degradation mixture was analyzed by HPLC for small acids after quenching with catalase (235 U mL⁻¹) to decompose the remaining H₂O₂ and adjusting the pH to < 2 using H₃PO₄. A Varian C18 Omnispher column was employed with eluting solvents in gradient mode—100% water (0.1% H₃PO₄) to 90% water (0.1% H₃PO₄) and 10% methanol in 5 min. By setting the UV detector at 215 nm, qualitative detection of formic, oxalic and chloromaleic acids was achieved. The presence of chloromaleic acids was also supported by ESI-MS where the reaction was conducted in carbonate buffer (0.01 M) using a Finnigan LCQ MS ion trap with ESI detection. GC-MS analyses were performed using a Thermo-Finnigan Trace DSQ. A Bruker 300 MHz NMR instrument was used for ³¹P NMR studies. Small organic degradation products and mineral ions were analyzed by ion chromatography (IC) performed by the R. J. Lee Group, Monroeville, PA, using samples where the reaction had been quenched with catalase. Total organic carbon (TOC) analyses were performed by Analytical Laboratory Services, Inc. Middletown, PA to determine the extent of mineralization during degradation. The reaction mixture was adjusted to pH<2 by addition of HCl (∼6 M). All kinetic measurements were performed at 25 °C using an Agilent spectrophotometer (model 8453) equipped with a thermostated cell holder and an automatic 8-cell positioner. The pH measurements were performed using a Corning 220 pH meter calibrated with standard buffer solutions at pHs 4, 7, and 10. Aquatic toxicity studies of the starting CP-EC and final degradation mixture employed Vibrio fisheri Microtox® assays and were performed at Coastal Bioanalysts Inc., Gloucester, VA. The degradation reaction mixture was quenched with catalase and the pH of the solution was adjusted to neutral (pH ∼7) for these studies.

Reactions

Reaction of CP and CP-EC with 1/H₂O₂ at pH 10. Stock solutions of pure CP (1 mM) in tert-butanol and 1 (10 μM) in water were combined with 0.1 M phosphate buffer solution at pH 10, so that final composition was tert-butanol (10% v/v) and buffer solution (90% v/v). To this solution, H₂O₂ (0.2 M) was added and stirred under ambient conditions (2 h). The reaction was quenched by catalase addition and analyzed by HPLC. The reactions with CP-EC were carried out under the same conditions and without using tert-butanol as a co-solvent.

Hydrolysis of chlorpyrifos EC assisted by peroxy anions. CP-EC (1 mM) was added to an aqueous buffered solution (pH 12, 0.1 M phosphate), which produced a thick milky white oil-in-water emulsion. The reaction mixture was treated with H₂O₂ (0.1 M) in a total volume of 5 mL. Addition of H₂O₂ resulted in lowering of the pH of the reaction medium, which was readjusted to 12 by addition of KOH solution. The reaction mixture was stirred vigorously, well beyond the clarification of the emulsion (45 min) and was monitored by HPLC following the disappearance of the CP (UV detection, 300 nm) having established concomitant formation of TCPy (UV detection, 300 nm) at 15 min intervals until > 99.95% of the CP was hydrolyzed. Simultaneously, loss of xylenes by evaporation from the emulsion was monitored by HPLC (UV detection, 265 nm).

Influence of mixed xylenes on the hydrolysis of CP. A stock solution was prepared by dissolving pure CP (300 mg, 1.1 M) in tert-butanol (0.3 mL) and mixed xylenes (0.5 mL, 5.1 M, giving ∼50% w/w). An aliquot of this stock solution (4.5 μL) was added to a pH 12 buffered solution (0.1 M phosphate) and sonicated to form a pseudoemulsion. The reaction mixture was treated with H₂O₂ (60 μL, 8.5 M) taking the total volume of the reaction mixture to 5 mL. The starting concentrations of CP, xylenes and H₂O₂ were 1 mM, 4.6 mM, and 0.1 M, respectively. The reaction mixture was stirred at room temperature and aliquots were collected every 15 min and analyzed by HPLC as above.

Influence of anionic and nonionic emulsifying agents on the hydrolysis of CP. A common anionic surfactant, sodium dodecyl sulfate (SDS, also called sodium lauryl sulfate), and a non-ionic surfactant, Triton X-100 0.05% (w/w) in aqueous solution, were tested separately by addition to the stirred pure CP (1 mM) in 0.1 M phosphate buffered aqueous reaction medium with H₂O₂ (0.1 M) at pH 12 (attained by addition of KOH solution). The hydrolysis of CP was monitored by HPLC as above.

Effect of CTAC on the hydrolysis of CP-EC. CP-EC (1 mM CP) was added to an aqueous buffered solution (pH 12, 0.1 M phosphate) to produce a thick milky white oil-in-water emulsion. Appropriate volumes of CTAC stock solution were added so that the final concentrations of CTAC in the reaction mixtures were 0.15, 0.31, 1.9, and 3.8 mM. The reaction mixtures were treated with H₂O₂ (0.1 M) in the total volume of 5 mL. The reaction mixtures were stirred vigorously and disappearance of CP and formation of TCPy was monitored by HPLC (UV 300 nm) at 15 min intervals. The effect of CTAB (0.01% w/w, 0.27 mM) on the hydrolysis of CP-EC was determined similarly. The reactions were monitored until > 99.95% of the CP was hydrolyzed.

Effect of Alcohol co-solvents on the hydrolysis of CP-EC. The effect of alcohol co-solvents on CP-EC hydrolysis was tested by addition of tert-butanol (10% v/v) and ethanol (10%, 5%, and 1% v/v) separately into the buffered reaction medium (0.1 M phosphate, pH 12). The reaction mixture consisting of CP-EC (1 mM) in the blend of alcohol-buffer was treated with H₂O₂ (0.1 M) and the progress of the reaction was monitored by HPLC (UV 300 nm) at 15 min intervals. The reactions were monitored until > 99.95% of the CP was hydrolyzed.

Kinetics of catalytic oxidation of TCPy by 1/H₂O₂ at different pH. Optimal conditions for the oxidation of TCPy by 1/H₂O₂ were identified by kinetic measurements at pH 7–12, using UV-vis spectroscopy to follow the decrease in absorbance at 319 nm (λ_max TCPy). A datum was collected every 20 s. In a typical kinetics run, phosphate buffer was added to a cuvette, followed by appropriate amounts of the stock solutions of TCPy (0.18 mM) and 1 (3.6 μM). The reaction was initiated by the addition of an aliquot of H₂O₂ (17.8 mM) and the cuvette was capped.

Catalytic oxidation of the hydrolysate containing TCPy with 1/H₂O₂. The hydrolyzed CP-EC solution was adjusted to pH 8 (by adding HCl) and 1 (20 μM) was added followed by H₂O₂ (0.1 M). The reaction mixture was stirred. The progress of the degradation was monitored by HPLC (UV detection, 300 nm) at 5 min intervals until the TCPy was no longer detectable. A stock solution in tert-butanol of pure TCPy (1 mM) was similarly treated with 1 (10 μM) and H₂O₂ (0.1 M) in a pH 8 buffer (0.1 M phosphate) and the reaction was monitored by HPLC (UV detection, 300 nm).

Differentiation of 1/H₂O₂ oxidation of TCPy, with pyridin-2-ol and pyridin-3-ol. Oxidation of TCPy (0.1 mM, λ_max 319 nm), pyridin-2-ol (0.1 mM, λ_max 294 nm), and pyridin-3-ol (0.1 mM, λ_max 308 nm) by 1 (2 μM)/H₂O₂ (0.01 M) at pH 8 (0.1 M phosphate) was followed by UV-vis spectroscopy. A datum was collected every 4 s. In a typical kinetic run, phosphate buffer was added to a cuvette, followed by appropriate amounts of the stock solutions of substrates (0.1 mM) and 1 (2 μM). The reaction was initiated by the addition of an aliquot of H₂O₂ (0.01 M) and the cuvette was capped.

Results and discussion

Oxidative degradation of CP-EC by 1/H₂O₂ at pH 10

Previous kinetic studies show that TAML activator 1 is most reactive at pH 10.²⁷ Thus, solutions of pure CP (1 mM) and CP-EC (1 mM) in buffered aqueous media (pH 10, 0.1 M phosphate) were treated separately with 1 (10 μM)/H₂O₂ (0.2 M) at room temperature. Over 2 h, this resulted in hydrolysis of the pure CP (99%) and oxidative degradation (70%) of the TCPy hydrolysis product; addition of a second aliquot of 1 resulted in the complete removal of TCPy. On the other hand, for CP-EC this same procedure resulted in only 8% hydrolysis of CP and 60% oxidative degradation of the TCPy that was formed by hydrolysis. This indicates that CP in the oil-in-water emulsion is protected by components of the formulation and the reaction conditions that were optimized for degradation of pure OPIs¹¹ are not satisfactory for formulated OPIs. Therefore, the reaction parameter space was broadly explored in an attempt to identify the source(s) of this protection and to improve the performance to achieve a simple and thoroughly effective protocol for CP-EC degradation. A one-pot, two-step process consisting of perhydrolysis (see text) followed by 1/H₂O₂ oxidation (see text) easily achieved the total degradation goal for CP-EC.

Perhydrolysis of CP-EC

H₂O₂ (pK_a ≈ 11.2–11.6²⁸) forms the hydroperoxide anion (HO₂⁻) upon deprotonation at pH 12 which is an α-effect nucleophile. This accelerates the hydrolysis of OP triesters in a process called perhydrolysis.²⁹ Therefore, in an attempt to accelerate the CP hydrolysis, H₂O₂ was added to the reaction mixture at pH 12. CP-EC perhydrolysis (> 99.95%) to TCPy was completed in 140 min under the prescribed conditions. Addition of 1 did not alter the CP hydrolysis rate. CP-EC emulsion is clearly resistant to base hydrolysis as no TCPy was formed after stirring for 6 h in pH 12 buffer. In comparison, pure CP was completely hydrolyzed by H₂O₂ to TCPy in just 15 min under similar conditions. This again demonstrates the increased degradation resistance of OPI formulations compared to pure active ingredients. The potent cholinesterase inhibitor CPO,³⁰ which is produced by oxidative desulfuration of CP,^30,31 was not detected under these conditions as it sometimes is in other decontamination processes.^4,7 This was an advantage of this process that CPO formation was avoided. However, it was observed in small quantities (< 1%) if 1 was added before the completion of perhydrolysis.

Impact of the solvent and emulsifier on perhydrolysis rate of CP-EC

The matrices of emulsifiable concentrate formulations are primarily non-polar hydrocarbon solvents, xylenes, C9–C10 solvents, kerosene, and other proprietary hydrocarbon solvents.¹⁸ Mixed xylenes (ca. 50% w/w) were identified in the CP-EC used here by HPLC and GC-MS. The impact of xylenes on CP-EC perhydrolysis was investigated by adding a comparable xylenes mixture (4.6 mM) to a pure CP (1 mM) solution in pH 12 buffer (0.1 M, phosphate). The two-phase solution was sonicated to form a pseudo-emulsion and subjected to perhydrolysis by the addition of H₂O₂ (0.1 M). The stirred mixture revealed that CP perhydrolysis proceeded to completion in 60 min. The rate of perhydrolysis in this pseudo-emulsion was four times slower (60 min) than for pure CP (15 min). It appears that the peroxy anions localized in the aqueous phase are hindered from interacting with the CP because it partitions preferentially to the xylene droplets.

EC pesticide formulations also contain small amounts (typically ca. 5% by weight) of anionic and non-ionic surfactant emulsifiers and are likely to be present in the proprietary CP-EC used here.³² Because the surfactant mixtures are usually complex, the identification of specific compounds was not attempted. The influence of anionic and non-ionic emulsifiers was assessed using a common anionic surfactant, sodium dodecyl sulfate (SDS) and a common non-ionic surfactant, Triton X-100. Addition of each surfactant to pure CP had no discernable effect on the rate of CP perhydrolysis. However, this observation comes with a caveat that some other set of surfactants/emulsifiers in CP-EC might influence the degradation process.

Accelerated Perhydrolysis of CP-EC by cationic surfactants

The rate of hydrolytic/oxidative degradation of phosphoesters,^33,34 including OP and carbamate pesticides,^35–37 in the presence of α-effect nucleophiles is greatly enhanced by cationic surfactants. The rate enhancements have been attributed to the ability of the micellar pseudophase to create higher localized reactant concentrations.^37,38 Reactions of non-ionic substrates (e.g. CP) with ionic reagents (e.g. HO₂⁻) are accelerated by counterionic micelles (cationic surfactants in this case).³⁹ Thus, the quaternary ammonium cationic surfactants, cetyl trimethylammonium bromide (CTAB) and cetyl trimethylammonium chloride (CTAC) were added separately to CP-EC perhydrolysis reaction media to test for acceleration of the perhydrolysis process.

While CP-EC perhydrolysis took 140 min to complete (>99.95%) in the absence of cationic surfactants, it was completed in 110 min with added CTAB (0.01% w/w, 0.27 mM, 21% faster) and in 90 min with CTAC (0.01% w/w, 0.31 mM, 35% faster). The results suggest a differential Br⁻versus Cl⁻ counter ion effect. These reproducible modest rate differences are in line with similar examples where an explanation has been postulated.³⁹ In CTAB, the more polarisable Br⁻ probably displaces the reactive HO₂⁻ from the cationic head groups in the Stern layer of the micelle, resulting in a lower rate enhancement than found for Cl⁻ in CTAC. A correlation between CTAC concentrations and CP-EC perhydrolysis times [0.15 mM, 105 min.; 0.31 mM, 90 min.; 1.9 mM, 40 min. and 3.8 mM, 20 min] (Fig. 2) shows that CP perhydrolysis is significantly enhanced by CTAC.

Fig. 2 Dependence of time for complete (> 99.95%) CP-EC perhydrolysis on CTAC concentration. Conditions: [CP-EC] = 1 mM, [H₂O₂] = 0.1 M, pH 12 (0.1 M, phosphate).

While CTAC improves the rate of perhydrolysis of CP-EC it would both add to the overall cost and reduce the environmental compatibility of a decontamination technology based on these findings. As shown below, CTAC also inhibits the oxidation of the hydrolysate TCPy by 1/H₂O₂ in the next step (see text). Thus, we looked for alternative methods for enhancing the rate of perhydrolysis of CP-EC to TCPy.

Effect of alcohol addition on perhydrolysis of CP-EC

When a CP-EC emulsion in pH 10 buffer (10% v/v tert-butanol) was treated with 1/H₂O₂, 20% CP hydrolyzed to TCPy in 2 h, an improvement over the 8% found without tert-butanol (see text). Inspired by this observation, we explored the effect of tert-butanol in perhydrolysis of CP-EC at pH 12. In presence of 10% v/v tert-butanol, the perhydrolysis of CP-EC to TCPy was indeed accelerated and reached completion (>99.95%) in 90 min. Toxicity concerns with tert-butanol⁴⁰ led us to explore other alternative alcohol sources e.g. ethanol (Table 1). While ethanol is oxidizable, it is relatively benign, inexpensive and readily available—we hoped it would speed up the perhydrolysis without unduly suppressing the oxidation of TCPy. Indeed, ethanol was found to be as effective as tert-butanol in promoting CP-EC perhydrolysis. At 10%, 5%, and 1% v/v, perhydrolysis took 90, 105, and 130 min, respectively, and the TCPy oxidation in the second step was not impacted (Table 1).

Table 1 Degradation times of pure CP and CP-EC (>99.95%) in the presence of different additives

Substrate	Medium	Time (min)
		Hydrolysis of CP^b	Oxidation of TCPy^c	Total
a Pure CP stock solution in tert-BuOH. b CP (1 mM), H2O2 (0.1 M), pH 12 (0.1 M phosphate). c TCPy (from 1 mM CP perhydrolysis), 1 (20 μM), H2O2 (0.1 M), pH 8 (0.1 M phosphate), except for pure CP where 1 (10 μM). HPLC measurements were performed at 15 min interval for hydrolysis of CP and 5 min interval for oxidation of TCPy—times are reported at which the species was first non-detectable. Each degradation process was performed at least in triplicate.
CP	Buffer + 10% t-butanol^a	15	15	30
CP-EC	Buffer	140	15	155
CP-EC	Buffer + 0.01% CTAB	110	30	140
CP-EC	Buffer + 0.01% CTAC	90	30	120
CP-EC	Buffer + 10% t-butanol	90	15	105
CP-EC	Buffer + 10% ethanol	90	15	105
CP-EC	Buffer + 5% ethanol	105	15	120
CP-EC	Buffer + 1% ethanol	130	15	145

---

Catalytic oxidation of TCPy

TCPy is known to persist longer in the environment than CP and exhibits low to moderate toxicity to aquatic and terrestrial biota.⁴¹ It has been found to be more toxic than CP to the soil bacterium Vibrio fisheri in Microtox® assays.⁴² When TCPy was repeatedly applied to soil, the CP biodegradation rate was reduced.⁴² TCPy has been shown to be a developmental neurotoxicant in PC12 and SH-SY5Y cell assays.⁴³ Sinclair et al. have assigned TCPy a high risk index, that incorporates human exposure and health effects.⁴⁴ In certain media, CP and TCPy act synergistically or additively, thereby increasing the toxicity of the mixture compared to the individual chemicals.⁴⁵ Thus, the fate of TCPy is of considerable importance in evaluating the greenness of any CP-EC decontamination technology.

At pH 8, 1/H₂O₂ is effective in oxidatively degrading TCPy (see text). Hydroxypyridines are subject to keto-enol tautomerism and can exist as equilibrium mixtures of hydroxy and oxo-forms (Fig. 3).⁴⁶ To assess the significance of tautomerism on the oxidative vulnerability of TCPy, 1/H₂O₂ oxidations of pyridin-2-ol (2-OH-Py) and pyridin-3-ol (3-OH-Py) were compared. In polar solvents, the equilibrium of hydroxypyridines usually shifts to the more polar oxo form. Several experimental and theoretical studies have concluded that 2-OH-Py exists predominantly in the pyrid-2-one (oxo) form in aqueous medium (A, Fig. 3).^47,48 The presence of an electron-withdrawing substituent alpha to the N atom, the equilibrium shifts in favor of the less polar hydroxy form, even in protic solvents such as water.⁴⁶ For example, the hydroxy form is strongly favored (≥95%) in aqueous solutions of 3,4,5,6-tetrachloro-2-hydroxypyridine, 2,6-dichloro-4-hydroxypyridine and 2,3,5,6-tetrachloro-4-hydroxypyridine.⁴⁹ Based on these facts, it has been suggested that 3,5,6-TCPy exists primarily in the hydroxy form in water (B, Fig. 3).⁴⁶

Fig. 3 Dominant species in the tautomeric equilibria of pertinent hydroxypyridines in polar solvents.

When 2-pyridone was treated with 1/H₂O₂ at pH 8 for 15 min, it remained largely unchanged (Fig. 4) probably because it exists as the cyclic amide tautomer (A, Fig. 3). Under the same reaction conditions, 3-OH-Py reacted much faster (Fig. 4). Unsubstituted 3-OH-Py in aqueous solution has been shown to exist as an equilibrium mixture of the neutral hydroxy form and the zwitterionic species (C and D, Fig. 3, respectively),⁵⁰ where the structures are electron rich compared to the A form. TCPy favors the hydroxy form (B). Moreover, at pH 8 TCPy is fully deprotonated (pK_a 4.5),⁵¹ and this fact is probably dominant in leading to the observed high susceptibility to 1/H₂O₂ oxidation. Nevertheless, the three electron-withdrawing chlorine substituents are expected to slow the oxidation of TCPy compared with that of 3-OH-Py and indeed the latter is oxidized more rapidly (Fig. 4).

Fig. 4 Kinetics of oxidation of 2-hydroxypyridine (294 nm), 3-hydroxypyridine (308 nm) and TCPy (319 nm) by 1/H₂O₂: Conditions: [Substrate] = 0.1 mM, [1] = 2 μM, [H₂O₂] = 0.01 M, pH 8 0.1 M phosphate, 25 °C.

pH dependence of oxidation reaction of TCPy

The oxidation of TCPy with 1/H₂O₂ was found to be highly dependent on the pH of the reaction medium. Kinetic studies reveal that, while the initial reaction rate is highest at pH 10, the most effective pH range for TCPy degradation is 8–9. The oxidative degradation of pure TCPy is complete at pH 8 in 20 min and at pH 9 in about 10 min as determined by UV-vis spectroscopy (Fig. 5a). Catalyst 1 exhibits its highest reactivity measured by the initial rate in decomposing TCPy at pH 10. However, the half-life of 1 is shorter at pH 10 than at pH 9 and catalyst degradation is always in competition with productive oxidation.⁵² For this reason, the pH region of 8–9 was found to be more efficacious for TCPy degradation. In addition, pH 8 is closer to environmental pH, giving another reason for choosing it as the optimal pH for TCPy degradation. Control reactions involving treatment of TCPy with either 1 or H₂O₂ produced no degradation over 2 h. In another set of control experiments, we followed the oxidation of TCPy by hydrogen peroxide catalysed by the generic iron salt, ferric chloride, and compared it with the 1/H₂O₂ system. Under the optimized reaction conditions for TCPy degradation by 1/H₂O₂, FeCl₃ at equimolar or at much higher concentrations compared to 1 did not oxidize TCPy (Fig. 5b).

Fig. 5 (a) Kinetics of oxidation of TCPy with 1/H₂O₂ at different pHs monitored by UV-vis spectroscopy at 319 nm: Conditions: [TCPy] = 0.18 mM, [1] = 3.6 μM, [H₂O₂] = 17.8 mM ([1]:[TCPy]:[H₂O₂] = 1 : 50 : 4944), 0.1 M phosphate, 25 °C. (b) Comparative kinetics of oxidation of TCPy with 1/H₂O₂ and FeCl₃/H₂O₂ monitored by UV-vis spectroscopy at 319 nm: Conditions: [TCPy] = 0.18 mM, [1] = 3.6 μM, [FeCl₃] = 3.6 μM (black), 36 μM (red), 360 μM (blue), [H₂O₂] = 17.8 mM, 0.1 M phosphate buffer, 25 °C. FeCl₃ absorbs light at 319 nm (λ_max ∼295 nm). Consequently, the starting absorbance for the experiment with [FeCl₃] = 360 μM was higher (Abs = 2.16) than that of all the rest (Abs = 1.56). The absolute value of the absorbance at 319 nm remained unchanged for experiments with ferric chloride.

The hydrolyzed CP-EC (1 mM CP) solution was adjusted to pH 8 and 1 (20 μM) and H₂O₂ (0.1 M) were added (1:TCPy:H₂O₂ = 1 : 50 : 5000, this [H₂O₂] is ∼10 × TCPy mineralization requirement). At pH 8, the total degradation of TCPy in process solutions containing ethanol (10%, 5%, and 1%) required only 15 min (Table 1), where EtOH addition improves performance in the prior step. Increasing EtOH concentration does not slow the rate of oxidation of TCPy noticeably, which illustrates the remarkable selectivity of 1/H₂O₂. The reaction mixtures without EtOH also required 15 min for TCPy oxidation so that the EtOH has negligible influence on this second step. Significant H₂O₂ decomposition occurs at the high pH⁵³ of the CP-EC perhydrolysis process which required the addition of another equivalent (0.1 M) of H₂O₂ during the oxidation of TCPy.

Oxidative degradation of TCPy in presence of CTAC

CTAC, added to accelerate the CP-EC perhydrolysis, inhibits the degradation of TCPy in the second stage of the process. Oxidation of TCPy resulting from CP-EC hydrolysis containing CTAC (0.01% w/w) or CTAB (0.01% w/w) required 30 min for completion, instead of 15 min for reaction mixtures devoid of cationic surfactants. In pH 8 buffer (0.1 M phosphate), TCPy exhibits a λ_max at 319 nm. The addition of seven aliquots of CTAC (10 μL of 5 mM stock solution) caused the λ_max to shift isosbestically to 328 nm (Fig. 6). As noted above, TCPy (pK_a 4.5) at pH 8 exists in the deprotonated state. TCPy⁻ could ion pair with CTA⁺. The isosbestic point in Fig. 6 shows that two light absorbing species are present in the reaction medium, suggesting that TCPy⁻ does indeed associate with CTA⁺ to give a spectroscopically distinct species. The formation of ion pairs between cationic surfactants and anionic substrates has been proposed for other systems.^54–56 Oxidation of the TCPy⁻CTA⁺ ion pair by 1/H₂O₂ was followed at 328 nm by UV-visible spectroscopy. The initial rate with CTAC was found to be considerably slower (1.1 × 10⁻⁷ M s⁻¹) than that for pure TCPy (1.6 × 10⁻⁶ M s⁻¹). Moreover, as shown in Fig. 7, 1 undergoes deactivation more rapidly in the presence of CTAC giving another reason why these surfactants are counterproductive overall. The total degradation of TCPy (0.18 mM) was achieved with 1 (3.6 μM)/H₂O₂ (17.8 mM) added as single aliquots. However in the presence of CTAC (0.43 mM), three aliquots of 1 (3.6 μM each) were required to complete the degradation process (Fig. 7).

Fig. 6 Spectral changes of TCPy upon addition of aliquots of CTAC stock solution. Conditions: [TCPy] = 0.18 mM, [CTAC] after each aliquot addition = 0.05, 0.10, 0.15, 0.20, 0.24, 0.28, 0.33 mM, 0.1 M phosphate pH 8, 25 °C.

Fig. 7 Kinetics of oxidative degradation of TCPy by 1/H₂O₂. Conditions: [TCPy] = 0.18 mM, [H₂O₂] = 17.8 mM, 0.1 M phosphate pH 8, 25 °C. (No CTAC) [1] = 3.6 μM, (With CTAC) [1] = 10.8 μM (in three equal aliquots), [CTAC] = 0.43 mM.

Ion pairing has been implicated as the source of increased resistance to oxidation in other reactions.⁵⁷ TAML activators are functional analogues of catalase-peroxidase enzymes where the catalase activity becomes increasingly dominant as the peroxidase substrate become increasingly oxidatively resistant.⁵⁸ Catalyst deactivation also competes with the peroxidase-like activity.⁵² It is possible that TCPy⁻ forms an ion pair with the catalyst that leads to catalyst deactivation—unfortunately, TCPy⁻ and 1 have overlapping absorptions in the UV-vis spectra which prevents a straightforward experimental analysis. If ion pairing occurs with other components of the media, such as the catalyst, any such process does not exhibit a discernable impact on the oxidative degradation. The reduced initial rate of oxidation of TCPy and the increased catalyst deactivation in the presence of CTAC can be reasonably attributed to ion pairing of TCPy⁻ and CTA⁺.

Thus, the complete degradation of CP-EC can be achieved in 155 min in a two-step process involving (i) perhydrolysis by HO₂⁻ of the CP in CP-EC at pH 12 followed by (ii) oxidation of TCPy in the resulting hydrolysate by 1/H₂O₂ at pH 8 (Scheme 1). If necessary, ethanol can be added to reduce the total time to as little as 105 min.

Scheme 1 Overall reaction pathway for the degradation of pure CP (and CP-EC) by tandem perhydrolysis and 1/H₂O₂ treatment.

Analysis of the final degradation mixture

The degradation reaction mixtures were analyzed for end products. After the perhydrolysis step, ion chromatographic analysis (IC) indicated the release of 47 ± 3% of the sulfur as SO₄²⁻.With the second stage treatment employing 1/ H₂O₂ at pH 8, the release of sulfur as SO₄²⁻ reached an almost quantitative 95 ± 5%.

Analysis by ³¹P NMR showed phosphorus quantitatively as diethylphosphate. This analysis required that we not use phosphate buffer as phosphate might be the ultimate mineral product. Thus, all reactions and samples for the ³¹P NMR experiments were prepared with carbonate buffer. CP exhibits a ³¹P singlet at 60.7 ppm under the solvent conditions prescribed (Fig. 8). The hydrolysis of CP (1 mM) by H₂O₂ (0.1 M) at pH 12 (0.01 M carbonate) resulted in the disappearance of the ³¹P signal from CP, indicating complete hydrolysis. A new peak appeared at 0.66 ppm (Fig. 8). Addition of 0.5 mM of diethylphosphate (DEP) to this sample did not result a new peak in the ³¹P NMR. Instead, the peak at 0.66 ppm increased significantly against noise confirming that it arose from DEP. At pH 12, DEP (1 mM) was treated with H₂O₂ (0.1 M) and even after 24 h only one peak at 0.66 ppm was detected in ³¹P NMR indicating that DEP is exceptionally resistant to perhydrolysis under the reaction conditions. The TCPy from CP hydrolysis was oxidized by 1 (20 μM)/H₂O₂ (0.1 M) at pH 8 (0.01 M carbonate). Analysis of the reaction mixture by ³¹P NMR displayed one signal at 0.66 ppm indicating that DEP is persistent in the final reaction mixture (Fig. 8). Treatment of DEP (0.5 mM) with 1 (5 μM)/ H₂O₂ (0.025 M) at pH 8 for 2 h did not convert DEP to other phosphorus compounds.

Fig. 8 ³¹P NMR of pure CP and diethylphosphate formed after the hydrolysis and oxidation reactions. (A) Pure CP, (Solvent comprised of 45% 0.01 M carbonate buffer, pH 12, 45% CH₃OH, 10% D₂O). (B) Hydrolysis of CP by H₂O₂ at pH 12, (Solvent comprised of 45% 0.01 M carbonate buffer, pH 12, 45% CH₃OH, 10% D₂O). (C) Oxidation of the hydrolysate TCPy by 1/H₂O₂ at pH 8, (Solvent comprised of 45% 0.01 M carbonate buffer, pH 8, 45% CH₃OH, 10% D₂O).

After the tandem process, organic carbon (TOC) analysis revealed 30% carbon mineralization and HPLC and IC showed qualitatively that chloromaleic, oxalic, and formic acids were formed indicating deep oxidation overall. IC analysis showed 89 ± 1% chlorine released as Cl⁻ with 2% of the nitrogen being converted to NO₂⁻ and NO₃⁻. Ammonia was detected (IC) at 32% of the total nitrogen. Microtox® aquatic toxicity (Vibrio fisheri) testing was used as an elementary probe of the environmental acceptability. A very high acute toxicity EC50 (15 min, 0.78%) was found for the CP-EC emulsion with (CP 1 mM). The treated reaction mixture exhibited significantly lowered acute toxicity, EC50 (15 min, 56.6%), a 72.5-fold reduction from the starting value. Possible sources of the residual toxicity are diethylphosphate (Vibrio fisheri EC50 = 19 mg L⁻¹)⁵⁹ and/or the formulation surfactants.

Conclusions

The present study establishes that both CP in a commercial emulsifiable concentrate and its toxic hydrolysis product TCPy can be easily eliminated via a tandem process employing a TAML activator with H₂O₂. The degradations of a range of OP pesticides by 1/H₂O₂ process reported earlier¹¹ suggests that the approach described herein could have broader utility for degrading many OP pesticide formulations. The process has promise for the onsite safe disposal of unused and obsolete OP pesticides in closed stirred tank reactors. The process is versatile, relatively easy-to-use, and should be effective for detoxifying pesticide-laden effluents associated with spills and for the washing of agricultural machinery. The CP-EC used for this study is an important commercial formulation and the simple tandem process we have developed provides a novel starting approach for examining any commercial OPI formulation.

Acknowledgements

T.J.C. thanks the Heinz Endowments for support and Ms Cynthia Persad, Director, Central Experiment Station, Centeno, Trinidad and Tobago, for a gift of CP-EC. S.K. thanks the R. K. Mellon Foundation for a Presidential Fellowship in the Life Sciences (Carnegie Mellon University). The authors thank Dr Alexander D. Ryabov for helpful discussions. A.C. thanks the John and Nancy Harrison for a Legacy Dissertation Fellowship. We thank Dr Roberto R. Gil and Dr Gayathri C. Withers for their help with NMR. NMR instrumentation at CMU was partially supported by NSF(CHE-0130903 and CHE-1039870).

References

1. B. K. Singh and A. Walker, FEMS Microbiol. Rev., 2006, 30, 428–471.
2. FAO, 2002. To be found under http://www.fao.org/english/newsroom/news/2002/9109-en.html.
3. A. García-Ripoll, A. Arques, R. Vicente, A. Domenech and A. M. Amat, J. Sol. Energ. Eng., 2008, 130, 041011-1–041011-5.
4. J. M. Herrmann, C. Guillard, M. Arguello, A. Agüera, A. Tejedor, L. Piedra and A. Fernández-Alba, Catal. Today, 1999, 54, 353–367.
5. P. L. Huston and J. J. Pignatello, Water Res., 1999, 33, 1238–1246.
6. D. M. Munnecke, J. Agric. Food Chem., 1980, 28, 105–111.
7. Q. Q. Wang and A. T. Lemley, Water Res., 2002, 36, 3237–3244.
8. T. J. Collins and C. Walter, Sci. Am., 2006, 294, 82–90.
9. A. Ghosh, A. D. Ryabov, S. M. Mayer, D. C. Horner, D. E. Prasuhn, S. S. Gupta, L. Vuocolo, C. Culver, M. P. Hendrich, C. E. F. Rickard, R. E. Norman, C. P. Horwitz and T. J. Collins, J. Am. Chem. Soc., 2003, 125, 12378–12379.
10. S. S. Gupta, M. Stadler, C. A. Noser, A. Ghosh, B. Steinhoff, D. Lenoir, C. P. Horwitz, K. W. Schramm and T. J. Collins, Science, 2002, 296, 326–328.
11. A. Chanda, S. K. Khetan, D. Banerjee, A. Ghosh and T. J. Collins, J. Am. Chem. Soc., 2006, 128, 12058–12059.
12. E. S. Beach, R. T. Malecky, R. R. Gil, C. P. Horwitz and T. J. Collins, Catal. Sci. Technol., 2011, 1, 437–443.
13. N. Chahbane, D. L. Popescu, D. A. Mitchell, A. Chanda, D. Lenoir, A. D. Ryabov, K. W. Schramm and T. J. Collins, Green Chem., 2007, 9, 49–57.
14. S. Mondal, Y. Hangun-Balkir, L. Alexandrova, D. Link, B. Howard, P. Zandhuis, A. Cugini, C. P. Horwitz and T. J. Collins, Catal. Today, 2006, 116, 554–561.
15. N. W. Shappell, M. A. Vrabel, P. J. Madsen, G. Harrington, L. O. Billey, H. Hakk, G. L. Larsen, E. S. Beach, C. P. Horwitz, K. Ro, P. G. Hunt and T. J. Collins, Environ. Sci. Technol., 2008, 42, 1296–1300.
16. L. Q. Shen, E. S. Beach, Y. Xiang, D. J. Tshudy, N. Khanina, C. P. Horwitz, M. E. Bier and T. J. Collins, Environ. Sci. Technol., 2011, 45, 7882–7887.
17. D. Banerjee, A. L. Markley, T. Yano, A. Ghosh, P. B. Berget, E. G. Minkley, S. K. Khetan and T. J. Collins, Angew. Chem., Int. Ed., 2006, 45, 3974–3977.
18. A. Knowles, Environmentalist, 2008, 28, 35–44.
19. R. A. Chapman and C. Harris, J. Environ. Sci. Health, Part B, 1984, 19, 397–407.
20. U. S. EPA, 2002.
21. J. D. Meeker, N. P. Singh, L. Ryan, S. M. Duty, D. B. Barr, R. F. Herrick, D. H. Bennett and R. Hauser, Hum. Reprod., 2004, 19, 2573–2580.
22. T. A. Slotkin, Toxicology of Organophosphate and Carbamate Compounds, Elsevier Academic Press, Boston, 2006.
23. V. A. Rauh, R. Garfinkel, F. P. Perera, H. F. Andrews, L. Hoepner, D. B. Barr, R. Whitehead, D. Tang and R. W. Whyatt, Pediatrics, 2006, 118, e1845–e1859.
24. T. Kiely, D. Donaldson and A. Grube, E.P.A., Washington, D.C., 2004, To be found under http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey = 3000659P.txt.
25. C. P. Horwitz and A. Ghosh, Carnegie Mellon University, PA, USA, 2006, To be found under http://www.chem.cmu.edu/groups/collins/patents/.
26. P. George, Biochem J., 1953, 55, 220–230.
27. A. Ghosh, D. A. Mitchell, A. Chanda, A. D. Ryabov, D. L. Popescu, E. C. Upham, G. J. Collins and T. J. Collins, J. Am. Chem. Soc., 2008, 130, 15116–15126.
28. C. W. Jones, Applications of hydrogen peroxide and derivatives, The Royal Society of Chemistry, Cambridge, 1999.
29. G. W. Wagner and Y. C. Yang, Ind. Eng. Chem. Res., 2002, 41, 1925–1928.
30. J. G. Wu and D. A. Laird, Environ. Toxicol. Chem., 2003, 22, 261–264.
31. H. W. Chambers, Organophosphates: Chemistry, Fate, and Effects, San Diego, CA, 1992, pp. 3–17.
32. T. Katagi, Rev. Environ. Contam. Toxicol., 2008, 194, 71–177.
33. S. Couderc and J. Toullec, Langmuir, 2001, 17, 3819–3828.
34. C. A. Bunton and H. J. Foroudian, Langmuir, 1993, 9, 2832–2835.
35. M. Arias, L. Garcia-Rio, J. C. Mejuto, P. Rodriguez-Dafonte and J. Simal-Gandara, J. Agric. Food Chem., 2005, 53, 7172–7178.
36. V. K. Balakrishnan, X. Han, G. W. VanLoon, J. M. Dust, J. Toullec and E. Buncel, Langmuir, 2004, 20, 6586–6593.
37. D. K. Dubey, A. K. Gupta, M. Sharma, S. Prabha and R. Vaidyanathaswamy, Langmuir, 2002, 18, 10489–10492.
38. D. Banerjee, F. M. Apollo, A. D. Ryabov and T. J. Collins, Chem.–Eur. J., 2009, 15, 10199–10209.
39. C. A. Bunton, Adv. Colloid Interface Sci., 2006, 123–126, 333–343.
40. S. J. Borghoff, J. E. Murphy and M. A. Medinsky, Fundam. Appl. Toxicol., 1996, 30, 264–275.
41. H. Shemer, C. M. Sharpless and K. G. Linden, Water, Air, Soil Pollut., 2005, 168, 145–155.
42. L. Somasundaram, J. R. Coats, K. D. Racke and H. M. Stahr, Bull. Environ. Contam. Toxicol., 1990, 44, 254–259.
43. D. Qiao, F. J. Seidler and T. A. Slotkin, Toxicol. Appl. Pharmacol., 2005, 206, 17–26.
44. C. J. Sinclair, A. B. A. Boxall, S. A. Parsons and M. R. Thomas, Environ. Sci. Technol., 2006, 40, 7283–7289.
45. T. Cáceres, W. He, R. Naidu and M. Meharaj, Water Res., 2007, 41, 4497–4503.
46. D. Vasudevan, P. J. Dorley and X. Zhuang, Environ. Sci. Technol., 2001, 35, 2006–2013.
47. M. M. Karelson, A. R. Katritsky, M. Szafran and M. C. Zerner, J. Org. Chem., 1989, 54, 6030–6034.
48. P. Beak, J. B. Covington and J. M. White, J. Org. Chem., 1980, 45, 1347–1353.
49. A. R. Katritzky, J. D. Rowe and S. K. Roy, J. Chem. Soc. B, 1967, 758–761.
50. J. Lior and S. B. Asensio, J. Solution Chem., 1995, 24, 1293–1305.
51. K. D. Racke, Rev. Environ. Contam. Toxicol., 1993, 131, 1–150.
52. A. Chanda, A. D. Ryabov, S. Mondal, L. Alexandrova, A. Ghosh, Y. Hangun-Balkir, C. P. Horwitz and T. J. Collins, Chem.–Eur. J., 2006, 12, 9336–9345.
53. W. Chu, Chemosphere, 2001, 44, 935–941.
54. M. L. Magnuson and C. A. Kelty, Anal. Chem., 2000, 72, 2308–2312.
55. K. K. Karukstis, D. A. Savin, C. T. Loftus and N. D. D'Angelo, J. Colloid Interface Sci., 1998, 203, 157–163.
56. R. K. Dutta and S. N. Bhat, Bull. Chem. Soc. Jpn., 1993, 66, 2457–2460.
57. R. Shukla and S. K. Upadhyay, Int. J. Chem. Kinet., 2007, 39, 595–604.
58. A. Ghosh, D. A. Mitchell, A. Chanda, A. D. Ryabov, D. L. Popescu, E. C. Upham, G. J. Collins and T. J. Collins, J. Am. Chem. Soc., 2008, 130, 15116–15126.
59. R. Gälli, H. W. Rich and R. Scholtz, Aquat. Toxicol., 1994, 30, 259–269.

---