Electrochemical Behavior and Determination of the Insecticide

Electrochemistry of Benzophenanthridine Alkaloids. Formation and Characterization of Redox Active Films from Products of Sanguinarine and Chelerythrin...
0 downloads 0 Views 250KB Size
Anal. Chem. 1997, 69, 898-903

Electrochemical Behavior and Determination of the Insecticide Synergist Piperonyl Butoxide Darren C. Coomber† and Daryl J. Tucker

School of Biological and Chemical Sciences, Deakin University, Geelong, Victoria 3217, Australia Alan M. Bond*

Department of Chemistry, Monash University, Clayton, Victoria 3168, Australia

Piperonyl butoxide may be reversibly oxidized in acetonitrile at a glassy carbon electrode to a cation radical under short time scale voltammetric conditions, e.g., cyclic voltammetry when the potential scan rate is above 500 mV s-1. During longer time domain experiments, the cation radical decays in a rate-limiting heterolytic bond cleavage step and subsequent transfer of a second electron at the potential of the first process. Additionally, a second oxidation process develops at more positive potentials. One product isolated from the initial oxidation process in an almost quantitative yield, under controlled potential electrolysis conditions, is 6-n-propyl-1,3-benzodioxole-5-carboxaldehyde. This carboxaldehyde is oxidized at the same positive applied potential as the second oxidation process observed in long time domain voltammetric experiments with piperonyl butoxide. The limit of detection for piperonyl butoxide in acetonitrile, using differential pulse voltammetry at a glassy carbon electrode, is 1.6 × 10-6 M (3σ), with a limit of determination of 4.1 × 10-6 M (10σ). Piperonyl butoxide was selectively determined using differential pulse voltammetry with a concentration of 5.11 ( 0.02 g L-1 in a commercial insecticide formulation containing pyrethrins. This result is in good agreement with the manufacturer’s stated concentration of 5.07 g L-1. The sample preparation requires only simple dilution of the formulation in an acetonitrile/dichloromethane (95:5) solvent mixture. Piperonyl butoxide, 5-[[2-(2-butoxyethoxyl)ethoxy]methyl]-6propyl-1,3-benzodioxole (I), is a commercially important insecticide synergist which can be used to increase the activity of the natural insecticide rotenone and the pyrethroid, carbamate, organophosphorus, and chlorinated hydrocarbon classes of insecticides.1

I † Present address: Institute of Biotechnology, University of Cambridge, Tennis Court Road, Cambridge CB2 1QT, England. (1) Yamamoto, I. In Pyrethrum. The Natural Insecticide; Casida, J. E., Ed.; Academic Press: New York, 1973; pp 195-210.

898 Analytical Chemistry, Vol. 69, No. 5, March 1, 1997

The electrochemical behavior of piperonyl butoxide has not been reported, although the compound has been described as possessing no intrinsic electrochemical activity.2 This paper describes the electrochemical oxidation of piperonyl butoxide in acetonitrile. Additionally, a method is developed for the determination of piperonyl butoxide in a commercial formulation using differential pulse voltammetry. EXPERIMENTAL SECTION Piperonyl butoxide (95.9%) was donated by Reckitt and Colman Products Pty. Ltd. (Ermington, NSW, Australia). Piperonal (reagent grade) was purchased from Tokyo Chemical Industry Co. Ltd. (Tokyo, Japan). An insecticide formulation containing piperonyl butoxide (5.07 g L-1) and pyrethrins (0.70 g L-1) in a solvent of liquid hydrocarbons (736.0 g L-1) was purchased commercially (Mortein Liquid Plus, Samuel Taylor Pty. Ltd., Ermington, NSW, Australia). The solvents used in voltammetric and chromatographic experiments were ChromAR HPLC grade (Mallinckrodt, Clayton, VIC, Australia). Voltammetric instrumentation, electrodes, electrolyte, and procedures are the same as those described elsewhere.3 For controlled potential electrolyses (CPE), the working electrode was a piece of glassy carbon (area of 18 cm2), the reference electrode was the same as that used in voltammetric experiments, and two electrically coupled platinum mesh auxiliary electrodes, immersed in salt bridges (0.10 M Et4NBF4 in acetonitrile), were positioned on either side of the working electrode. The solution was stirred using a magnetic stirring bar and purged with solvent-saturated nitrogen throughout the electrolysis. Excess water was added to the postelectrolysis solutions prior to extraction with diethyl ether (3 × 20 mL). The combined ether extracts were dried over magnesium sulfate and filtered, and the solvent was removed to give a dark brown-orange residue. Electrolysis solutions and residues were examined using highperformance liquid chromatography (HPLC) and thin-layer chromatography (TLC). The HPLC system consisted of a Model 6000A solvent delivery system, a Model 441 UV detector operated at 214 nm (Waters Associates, Milford, MA), a 150 mm × 4.6 mm Zorbax ODS column (DuPont Instruments, Wilmington, DE), a Rheodyne (Cotati, CA) injector with a 10 µL sample loop, and a Chromatopac C-R3A integrator (Shimadzu Corp., Kyoto, Japan). (2) Wintersteiger, R.; Ofner, B.; Juan, H.; Windisch, M. J. Chromatogr. 1994, 660, 205-10. (3) Coomber, D. C.; Tucker, D. J.; Bond, A. M. Anal. Chem. 1996, 68, 126771. S0003-2700(96)00846-3 CCC: $14.00

© 1997 American Chemical Society

The acetonitrile/water (55:45) mobile phase had a flow rate of 1.0 mL min-1. TLC was performed on silica gel plates (Kieselgel 60 F254, Merck, Darmstadt, Germany) with heptane/acetone mobile phases and monitored under a UV lamp (254 nm, Oliphant Pty. Ltd., Adelaide, SA, Australia). Residues of CPE solutions were chromatographed on a silica gel column (15 cm × 2.5 cm i.d.), prepared by washing with heptane (50 mL). The residue was stirred with heptane (1 mL), and the soluble portion was added to the top of the column and washed with heptane (50 mL). A product was eluted with heptane/acetone (5:1) by discarding the first 50 mL of solvent and collecting the next 20 mL, which was taken to dryness and rechromatographed, using the same procedure, on a fresh silica gel column. The composition of the resulting light-orange compound was elucidated by mass and 1H nuclear magnetic resonance (NMR) spectrometry. Mass spectra were obtained using a Hewlett Packard 5988A GC/MS system (Palo Alto, CA), operated under electron impact (EI) or positive-ion chemical ionization (PCI) conditions with an ionization potential of 70 eV. Sample introduction was from a direct insertion probe, and the reactant gas for PCI mass spectrometry was methane. 1H NMR spectra of samples in chloroform-d (99.8%, Cambridge Isotope Laboratories, Woburn, MA) were obtained using a Bruker AM300 spectrometer (300 MHz, Rheinstetten, Germany) controlled by an ASPECT 3000 computer. Chemical shift values are reported in ppm and referenced against the chloroform resonance (7.24 ppm for 1H). The determination of piperonyl butoxide in an insecticide formulation was achieved via the differential pulse voltammetric technique at a glassy carbon electrode using the 646VA Processor (Metrohm, Herisau, Switzerland) and the general technique described previously.3 An aliquot of the formulation (400 µL) was added to a solution (20.0 mL) of acetonitrile/dichloromethane (95: 5). Dichloromethane was necessary to ensure solubility of the liquid hydrocarbon solvent of the formulation. The potential was scanned at a rate of 20 mV s-1 between +500 and +1500 mV vs Ag/Ag+, a pulse amplitude of +50 mV was applied to the dc ramp, and the peak current was measured at +960 mV vs Ag/Ag+. Quantitation was by the technique of standard addition, with six additions (each of 20.0 µL) of a piperonyl butoxide (35.7 g L-1) stock solution in acetonitrile. RESULTS AND DISCUSSION Voltammetry. Piperonyl butoxide is electrochemically oxidized, but not reduced, in acetonitrile. Cyclic voltammograms (corrected for background currents) obtained at a 3 mm diameter glassy carbon electrode are shown in Figure 1a-c, and the data are summarized in Table 1. At a scan rate of 20 mV s-1, two chemically irreversible oxidation processes are observed (Figure 1a). An increase in scan rate to 100 mV s-1 results in the appearance of a small reduction current on the reverse scan direction of the initial oxidation process (Table 1). Additionally, at this higher scan rate, the peak current of the second oxidation process diminishes relative to that of the first oxidation process, with respect to the relative values observed at lower scan rates. With an increase in scan rate to 500 mV s-1, the peak current of the second oxidation process has further diminished, with respect to the first process (Figure 1b), and the peak current of the reduction process coupled to the initial oxidation process has increased relative to the oxidation peak current. At a scan rate of 2007 mV s-1, only a response due to the first oxidation process

Figure 1. Voltammograms obtained for the oxidation of piperonyl butoxide (5 × 10-4 M) in acetonitrile (0.10 M Et4NBF4) at a glassy carbon electrode. Curves a, b, and c are cyclic voltammograms at scan rates (a) 20, (b) 500, (c) 2007 mV s-1, while curve d was obtained at an RDE using a scan rate of 20 mV s-1 and an electrode rotation rate of 3000 rpm.

is observed (Figure 1c), and the ratio of oxidation to reduction peak current of the initial electron transfer process approaches unity at a scan rate of 5120 mV s-1 (Table 1). Under fast scan rate conditions, the first electron-transfer process is, therefore, observed to be chemically reversible. For cyclic voltammetric scan rates above 200 mV s-1, the halfwave potential (E1/2), calculated as the average value for the oxidation and reduction peak potentials, has a value of +985 ( 5 mV vs Ag/Ag+ (Table 1). This value is believed to be close to the formal reversible potential or E°′ value. Additionally, since red the ratio of iox was independent of the concentration of p /ip piperonyl butoxide, it follows that the reaction causing the chemical irreversibility is a first-order process. At a 3 mm diameter glassy carbon rotating disk electrode, only a single sigmoidal-shaped oxidation wave with an E1/2 value of +970 mV vs Ag/Ag+ was obtained at a rotation rate of 3000 rpm (Figure 1d). That is, the second oxidation step is absent, as is the case when using fast scan rates under conditions of cyclic voltammetry. The E3/4 - E1/4 value was determined to be 65 ( 5 mV, which, for this rotating disk steady state experiment, is Analytical Chemistry, Vol. 69, No. 5, March 1, 1997

899

Table 1. Cyclic Voltammetric Data Obtained for the Oxidation of Piperonyl Butoxide (5.4 × 10-4 M) in Acetonitrile (0.10 M Et4NBF4) at a Glassy Carbon Electrodea first process scan rate (mV 20 50 100 200 500 1003 2007 5120 a

s-1)

Eox p (mV) 1020 1030 1025 1025 1030 1030 1040 1070

iox p

(µA)

Ered p

7.3 10.2 13.2 17.3 26.5 35.4 49.0 68.9

(mV)

885 935 940 940 930 915

ired p

second process (µA)

4.0 7.4 15.3 27.9 41.3 65.9

∆Ep (mV)

135 85 90 90 110 155

E°′ (mV)

950 980 985 985 985 990

Eox p (mV)

iox p (µA)

1290 1310 1320 1340

4.6 6.2 6.8 6.5

The peak current data have been corrected for the background current.

Figure 2. Cyclic voltammograms obtained at a glassy carbon electrode with a scan rate of 100 mVs-1 after oxidative CPE of piperonyl butoxide (5.0 × 10-4 M) in acetonitrile (0.10 M Et4NBF4) at a glassy carbon electrode held at a potential of +1150 mV vs Ag/ Ag+.

close to the value expected for a reversible one-electron charge transfer process. At a 100 µm diameter platinum microdisk electrode and with a scan rate of 50 mV s-1, the voltammogram was sigmoidal shaped. An E1/2 value of +950 mV vs Ag/Ag+ was obtained, with a calculated E3/4 - E1/4 value of 75 mV. Again, the second process was absent under these near steady state conditions, where radial rather than linear diffusion is the dominant mode of mass transport.4 Furthermore, at a microelectrode, the limiting current per unit concentration (ignoring sign) was identical to that of the known one-electron reduction of tetramethrin3 under the same near steady state conditions, confirming, on the assumption of equal diffusion coefficients, that the process simplifies to a one-electron charge transfer process at short time domains. Controlled Potential Electrolysis and Coulometry. CPE of piperonyl butoxide (+1150 mV vs Ag/Ag+) involved the transfer of 2.2 ( 0.2 electron/molecule. Cyclic voltammetric scans of the postelectrolysis solution shows a major irreversible oxidation process with a peak potential of +1350 mV vs Ag/Ag+ (Figure 2). Additionally, small and drawn out reduction processes are found at -1.50 and -2.12 V vs Ag/Ag+, which implies that a range of products are formed. The peak potential and the wave shape for the oxidation process are similar to those observed for the second oxidation process of piperonyl butoxide under voltammetric conditions when slow scan rates are employed (Table 1). Thus, under conditions where the first oxidation response is irreversible (slow scan rate voltammetry or CPE), the process involves the transfer of two electrons (ECE-type process) to generate a new compound which can be oxidized at +1350 mV vs Ag/Ag+, while at short time domains (fast scan rates, fast rotation rates or steady state conditions) a reversible one-electron process is observed that produces a cation radical. (4) Aoki, K.; Osteryoung, J. J. Electroanal. Chem. 1981, 122, 19-35.

900 Analytical Chemistry, Vol. 69, No. 5, March 1, 1997

Figure 3. Chromatograms of (a) piperonyl butoxide (5.0 × 10-4 M) and (b) postelectrolysis solution of piperonyl butoxide (5.0 × 10-4 M, CPE at a glassy carbon electrode held at a potential of +1150 mV vs Ag/Ag+) in acetonitrile (0.10 M Et4NBF4). Zorbax ODS column, acetonitrile/water (55:45), flow rate of 1.0 mL min-1, 10 µL injection, UV absorbance at 214 nm.

Under conditions of reversed-phase HPLC, piperonyl butoxide eluted in a broad peak with a retention time of 12.10 min (Figure 3a). Chromatography of the postelectrolysis solution showed the complete removal of piperonyl butoxide and the presence of a major product eluting with a retention time of 4.25 min, together with a range of very polar compounds eluting close to or with the solvent front (Figure 3b). Normal-phase TLC (heptane/acetone 5:1) confirmed the presence of a major species as well as a range of other products in the postelectrolysis solution. A broad spot with a retention value, Rf, of 0.40 was observed, together with unresolved species extending as a continuous band from the origin to an Rf value of 0.13. Extraction of the postelectrolysis solution with diethyl ether yielded a residue which was shown by TLC to contain the band eluting at an Rf value of 0.40. This band was separated from other components on a silica gel column. HPLC showed that the major species isolated from the residue eluted with a retention time of 4.25 min, which is identical to that of the major species observed in the chromatogram of the postelectrolysis solution. Mass spectrometry of the species obtained from the postoxidation extract indicates a molecular ion of m/z 192 [EI m/z (relative abundance) of 192 (M+, 100), 191 (46), 177 (36), 163 (72), 149 (42), 135 (38), 77 (54)]. In PCI studies, an ion of m/z 193 (M + 1) was the only major ion observed [PCI m/z (relative abundance) of 233 (M + 41, 4), 221 (M + 29, 11), 193 (M + 1, 100)].

The structure giving rise to the 1H NMR spectrum of the isolated component of the postoxidation residue was assigned to that of 6-n-propyl-1,3-benzodioxole-5-carboxaldehyde (II). The

Scheme 1. Mechanism Proposed To Explain the Electrochemical Oxidation of Piperonyl Butoxide in Acetonitrile

II

molecular mass of this compound is 192 Da, which is consistent with the value obtained from mass spectrometry. The spectrum obtained consisted of four singlet (s), two triplet (t), and a quartet of triplet (qt) resonances. 1H NMR (300 MHz, CDCl3): δ 10.15 (s, 1H), 7.30 (s, 1H), 6.68 (s, 1H), 6.00 (s, 2H), 2.90 (t, 2H, J ) 7.6 Hz), 1.62 (qt, 2H, J ) 7.4 Hz), and 0.95 ppm (t, 3H, J ) 7.4 Hz). The spectrum was assigned with reference to spectra obtained for both piperonyl butoxide and piperonal (1,3-benzodioxole-5carboxaldehyde, results not shown). The 1H NMR spectrum obtained for piperonal butoxide consisted of four singlet, four triplet, and three multiplet (m) resonances. 1H NMR (300 MHz, CDCl3): δ 6.83 (s, 1H), 6.64 (s, 1H), 5.88 (s, 2H), 4.45 (s, 2H), 3.5 (m, 8H), 3.44 (t, 2H, J ) 6.7 Hz), 2.51ppm (t, 2H, J ) 7.8 Hz), 1.54 (m, 4H, unresolved coupling), 1.36 (m, 2H, J ) 7.7 Hz), 0.93 (t, 3H, J ) 7.3 Hz), and 0.88 ppm (t, 3H, J ) 7.3 Hz). The most striking feature of the 1H NMR spectrum of the post-electrolysis product, relative to that obtained with piperonyl butoxide, was the presence of a resonance assigned as an aldehyde (10.15 ppm). Importantly, the resonance due to the methylenedioxy protons is observed (6.00 ppm), as well as that of protons coupling to an aromatic ring system, indicating that the 1,3-benzodioxole system is unchanged by the oxidation process. Resonances which were assigned to the propyl chain are observed at 2.90, 1.62, and 0.95 ppm. As piperonal has a structure similar to that of the proposed oxidation product, the two compounds would be expected to possess not only similar spectrometric behavior but also similar voltammetric behavior. In acetonitrile, piperonal may be oxidized (peak potential of +1415 mV vs Ag/Ag+) and reduced (-2270 mV vs Ag/Ag+) in chemically irreversible processes. The peak potential (+1350 mV vs Ag/Ag+) and current obtained for the voltammetric oxidation of the piperonyl butoxide postoxidation product (Figure 2) are consistent with the analogous data obtained for the oxidation of piperonal. It is difficult to compare peak potentials for irreversible processes since they have no direct thermodynamic significance. However, since piperonal can be reduced, it is possible that one of the reduction processes in the postoxidation solution (-1.50 or -2.12 V vs Ag/Ag+) is due to reduction of 6-n-propyl-1,3-benzodioxole-5-carboxaldehyde and the other is due to reduction of a compound(s) formed from the polyether chain which was(were) not characterized. Comparison of the chromatographic peak areas of both piperonyl butoxide and piperonal with that of 6-n-propyl-1,3-benzodioxole-5-carboxaldehyde indicate that the product was formed in an essentially quantitative yield. A similar conclusion is reached upon comparison of voltammetric peak heights for oxidation of piperonal and the electrolysis product. Mechanism. Voltammetric data indicate that piperonyl butoxide is initially oxidized with the loss of one electron to form the piperonyl butoxide cation radical, which rapidly decays via

cleavage of the oxygen-carbon bond between the benzylic ether oxygen and the alkyl ether chain to form a stable compound, 6-npropyl-1,3-benzodioxole-5-carboxaldehyde as well as other products, which can be oxidized at more positive potentials. The independence of the ratio of the oxidation and reduction peak currents upon piperonyl butoxide concentration indicates that the chemical step is first order, which is consistent with heterolytic cleavage of the ether chain. As well as forming 6-n-propyl-1,3-benzodioxole-5-carboxaldehyde, cleavage will yield a proton and an alkyl ether chain radical cation, which apparently is further oxidized at the initial applied potential to a carbocation which can undergo a variety of rearrangement reactions, resulting in the formation of alkene groups and subsequent polymerization, as suggested by the fact that components of the postelectrolysis residue elute as a broad retention band on silica gel. However, no species arising from the alkyl ether chain were characterized by mass or NMR spectrometry. A proposed oxidation mechanism is shown in Scheme 1. Although no previous studies of the electrochemistry of piperonyl butoxide have been reported, a number of authors have described the electrochemical behavior of 1,3-benzodioxole and some related compounds. 1,3-Benzodioxole has been oxidized in acetonitrile and fluorosulfuric acid.5 The initial, irreversible oxidation of 1,3-benzodioxole, in fluorosulfuric acid, led to the formation of a species which was reversibly oxidized at a less positive applied potential.5 The reduction of a range of 1,3benzodioxole compounds, using a dropping mercury electrode in dimethylformamide, has been reported.6 In all cases, the site of the reduction response was in a substituent side chain. Previous studies of the electrochemistry of 1,3-benzodioxole compounds (5) Rudenko, A. P.; Zarubin, M. Ja.; Pragst, F. J. Electroanal. Chem. 1983, 151, 89-100. (6) Lurik, B. B.; Marinova, R. I.; Volkov, Y. P. Zh. Oschch. Khim. 1975, 45, 2287-91.

Analytical Chemistry, Vol. 69, No. 5, March 1, 1997

901

in acetonitrile have shown that oxidation leads to the formation of polymeric species coupled through the benzene ring systems.7 In addition, 1,3-benzodioxole is a structural feature of methylenedioxyamphetamine drugs with this moiety shown, for a series of these compounds, to be the site of irreversible electrochemical oxidation in buffered aqueous media.8 The authors stated that a radical cation, stabilized in the benzodioxole system, was involved in the oxidation process, although no products of the oxidation were identified.8 Electrochemical oxidations of 1,3-benzodioxole compounds in methanol resulted in electrode passification and low yields of expected products.9,10 The starting material was oxidized to high molecular weight compounds through the aromatic nucleus and methylenedioxy bridge, leading to unstable species, which further polymerized.9 2-Methoxy-1,3-benzodioxoles have been obtained as products from the electrochemical oxidation of 1,3-benzodioxoles in methanol (saturated with carbon dioxide and containing sodium methoxide).10,11 Determination of Piperonyl Butoxide in a Formulation. The determination of the detection limit of piperonyl butoxide and the concentration in a formulation was achieved using the differential pulse voltammetric response obtained for the initial electron transfer process at a glassy carbon electrode. The calculated limit of detection in acetonitrile with a pulse amplitude of 50 mV was 1.6 × 10-6 M (3σ), with a limit of determination of 4.1 × 10-6 M (10σ). Using the technique of standard addition, a concentration of piperonyl butoxide of 5.11 ( 0.02 g L-1 (average of 10 formulation sample determinations) was determined, which is in excellent agreement with the stated concentration of 5.07 g L-1. The average correlation coefficient of the standard addition curves was 0.998, and average errors associated with the slope and intercept values were less than 2%. Voltammograms obtained for the sample with three standard additions are shown in Figure 4a. The sample volume (400 µL) was chosen to provide piperonyl butoxide at a concentration (6.39 × 10-4 M), which minimized errors associated with determining peak current from the sloping baseline. The electrochemical procedure developed in this study, therefore, is concluded to provide a simple and selective method for the determination of piperonyl butoxide in the insectide formulation without the requirement for sample pretreatment or a separation technique. Previous literature studies have been directed toward the determination of piperonyl butoxide in the presence of pyrethroid and other insecticides. For example, direct derivative UV12,13 and fluorescence14 spectrophotometry, gas chromatography with flame ionization detection,15 and liquid chromatography with UV2,16 or fluorescence17 detection methods have been successfully developed. Haddad et al.16 have calculated (7) Fleischhauer, J.; Ma, S.; Schleker, W.; Gersonde, K.; Twilfer, H.; Dallacker, F. Z. Naturforsch. A 1982, 37, 680-7. (8) Squella, J. A.; Cassels, B. K.; Arata, M.; Bavestrello, M. P.; Nun ˜ez-Vergara, L. J. Talanta 1993, 40, 1379-84. (9) Bornewasser, U.; Steckhan, E. In Electroorganic Synthesis; Little, R. D., Weinberg, N. L., Eds.; Marcel Dekker, Inc.: New York, 1991; pp 205-15. (10) Barba, I.; Chinchilla, R.; Go´mez, C. J. Org. Chem. 1990, 55, 3270-2. (11) Thomas, H. G.; Schmitz, A. Synthesis 1985, 31-3. (12) Sharaf El Din, M.; El-Brashy, A. Spectrosc. Lett. 1990, 23, 899-909. (13) Jimena Garcı´a, J. A.; Gime´nez Plaza, J.; Cano Pavo´n, J. M. Anal. Chim. Acta 1992, 268, 152-7. (14) Bowman, M. C.; Beroza, M. Residue Rev. 1967, 17, 1-22. (15) Sakaue, S.; Kitajima, M.; Horiba, M.; Yamamoto, S. Agric. Biol. Chem. 1981, 45, 1135-40. (16) Haddad, P. R.; Brayan, J. G.; Sharp, G. J.; Dilli, S.; Desmarchelier, J. M. J. Chromatogr. 1989, 461, 337-46.

902 Analytical Chemistry, Vol. 69, No. 5, March 1, 1997

Figure 4. Differential pulse voltammograms obtained for piperonyl butoxide at a glassy carbon electrode. Voltammogram a is the standard addition determination (three additions) of piperonyl butoxide in a formulation (400 µL added to 20.0 mL). This determination was performed in 95:5 acetonitrile/dichloromethane (0.10 M Et4NBF4) using a pulse amplitude of +50 mV, a scan rate of 20 mVs-1, standard addition of 20.0 µL of a 35.7 g L-1 piperonyl butoxide solution in acetonitrile, and peak current measurement at +960 mV vs Ag/Ag+. Voltammograms b and c are the oxidation of piperonyl butoxide (5.0 × 10-4 M) in acetonitrile (0.10 M Et4NBF4) at (b) a stationary electrode and (c) a RDE, using a scan rate of 5 mV s-1 and a pulse amplitude of 50 mV for both experiments and a rotation rate of 3000 rpm for the RDE experiment.

a detection limit of 3.0 × 10-6 M for piperonyl butoxide on rice using HPLC with UV detection. On the dc time scale associated with the very slow scan rates commonly used in differential pulse voltammetry (e.g., 5-20 mV s-1) at a stationary electrode, the oxidative cleavage of piperonyl butoxide to the carboxaldehyde occurs. Consequently, two oxidation processes are observed when using this form of the differential pulse technique (Figure 4a and b), and the process used in the determination is irreversible. As the formation of the oxidizable carboxaldehyde is dependent on a chemical step, the second oxidation processes may be eliminated by coupling a shorter time domain steady state technique such as rotating disk voltammetry with differential pulse voltammetry (Figure 4c). This form of response is inherently simpler and represents an ideally reversible process, which provides an advantage with respect to sensitivity and resolution. CONCLUSIONS This paper reports that the electrochemical oxidation of the synergist piperonyl butoxide may be used to provide a selective and accurate method for the determination of this compound in a formulation. The results are in direct contrast to the work of Wintersteiger et al., who have stated that piperonyl butoxide has no intrinsic electrochemical activity.2 (17) Isshiki, K.; Tsumura, S.; Watanabe, T. Bull. Environm. Contam. Toxicol. 1978, 19, 518-23.

It has been suggested that the synergistic action of some 1,3benzodioxole compounds occurs by the compound acting as an alternative oxidizable substrate, with the formation of benzodioxolium ions and homolytic free radicals implicated.18 In this way, it has been proposed that the synergists can react with, and inhibit, both enzyme and nonenzyme free radical generating systems.18 This study demonstrates that a short-lived cation radical of piperonyl butoxide can be formed electrochemically.

ACKNOWLEDGMENT This work was financially supported (D.C.C.) by the Australian Wool Research and Development Corp. through the award of postgraduate scholarship UDG002. Piperonyl butoxide was donated by Reckitt and Colman Products Pty. Ltd. Mass spectra were obtained by Mr. G. Franklin of Deakin University. Received for review August 20, 1996. Accepted December 3, 1996.X AC960846E

(18) Casida, J. E. J. Agric. Food Chem. 1970, 18, 753-72.

X

Abstract published in Advance ACS Abstracts, January 15, 1997.

Analytical Chemistry, Vol. 69, No. 5, March 1, 1997

903