Voltammetric Determination of the Synthetic Pyrethroid Insecticide

Apr 1, 1996 - Darren C. Coomber andDaryl J. Tucker. School of Biological and Chemical Sciences, Deakin University, Geelong, Victoria 3217, Australia...
0 downloads 0 Views 182KB Size
Anal. Chem. 1996, 68, 1267-1271

Voltammetric Determination of the Synthetic Pyrethroid Insecticide Tetramethrin in Acetonitrile Darren C. Coomber† and Daryl J. Tucker

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

School of Chemistry, La Trobe University, Bundoora, Victoria 3083, Australia

The synthetic pyrethroid insecticide tetramethrin may be reduced reversibly (E°′ ) -1.650 V vs Ag/Ag+) in acetonitrile at hanging mercury drop electrodes (HMDE) and glassy carbon electrodes. On the voltammetric time scale, the initial electron-transfer process involves the reversible formation of a radical anion. Data obtained from electron paramagnetic resonance spectroscopy indicate that the unpaired electron of the radical is located within the phthalimide system of the molecule. The radical anion may be further reduced at very negative applied potentials with the number of processes being dependent on the nature of the voltammetric technique. The detection limit (3σ) for the determination of tetramethrin in acetonitrile at a glassy carbon electrode, using differential pulse voltammetry, was found to be 2.1 × 10-6 M. At a HMDE the detection limit is lower, having a value of 9.6 × 10-7 M. The limit of determination (10σ) at a glassy carbon electrode is 3.5 × 10-6 M and at a HMDE is 3.0 × 10-6 M. Tetramethrin was selectively determined in an insecticide formulation, at a glassy carbon electrode using differential pulse voltammetry, at a concentration (w/v) of 0.34 ( 0.02%. The determined concentration is in good agreement with the stated value of 0.350 ( 0.018% (w/v). Tetramethrin [cyclohex-1-ene-1,2-dicarboximidomethyl (1RS)cis,trans-2,2-dimethyl-3-(2-methyl-prop-1-enyl)cyclopropanecarboxylate, 1 is a synthetic pyrethroid insecticide1 commonly used in

1

domestic insecticide formulations. The electrochemical reduction of tetramethrin in buffered aqueous solutions has been previously reported and applied to the determination of the compound using † Present address: Joint Protein Structure Laboratory, Ludwig Institute for Cancer Research and The Walter and Eliza Hall Institute of Medical Research, PO 2008 Royal Melbourne Hospital, Victoria, 3050, Australia. ‡ Present address: School of Chemistry, Monash University, Clayton, Victoria, 3168, Australia. (1) Kato, T.; Ueda, K. Agric. Biol. Chem. 1964, 28, 914-5.

0003-2700/96/0368-1267$12.00/0

© 1996 American Chemical Society

adsorptive stripping voltammetric methods.2,3 In one method, tetramethrin was adsorbed, under open-circuit conditions, at a modified carbon paste electrode and determined using differential pulse voltammetry.2 The second method involves the adsorption of tetramethrin, under potential control, at a hanging mercury drop electrode (HMDE) with subsequent reduction and determination using square wave voltammetry.3 This paper extends studies on the electrochemical reduction of tetramethrin to the aprotic nonaqueous solvent acetonitrile. The reversible reduction process has been applied to the direct determination of tetramethrin in an insecticide formulation. Significantly, insecticide formulations containing tetramethrin are readily soluble in organic solvents so that no pretreatment of the sample is required. Furthermore, other formulation components are not electroactive in the potential range of interest so an inherently simple analytical method, void of chromatographic separations, was readily developed. EXPERIMENTAL SECTION Reagents. Tetramethrin (92.0%) and a simulated aerosol formulation [containing 0.350% (w/v) tetramethrin and 0.177% (w/v) permethrin] were donated by Reckitt and Colman Pty. Ltd. (Ermington, NSW, Australia). The supporting electrolyte was electrometric-grade tetraethylammonium tetrafluoroborate (Et4NBF4; Southwestern Analytical Chemicals Inc., Austin, TX). Acetonitrile (ChromAR HPLC grade, Mallinckrodt, Clayton, Victoria, Australia) was passed down a column of activated neutral alumina prior to voltammetric experiments. Voltammetry. Voltammetic experiments were performed using a BAS 100A electrochemical analyzer (Bioanalytical Systems, West Lafayette, IN). The electrochemical cell was operated in a three-electrode configuration at 20 ( 2 °C. Working electrodes used in cyclic and differential pulse voltammetry consisted of either a 3.0 mm diameter glassy carbon disk (Metrohm, Herisau, Switzerland) or a hanging mercury drop electrode (HMDE, Metrohm E450) with a surface area of 0.015 cm2. Rotating disk voltammetry was performed at a glassy carbon disk (3.0 mm diameter, Metrohm) with a Metrohm 628-50 drive unit and a 62810 rotation speed controller. The solid electrodes were polished with alumina on a polishing cloth (Microcloth, Buehler, Lake Bluff, IL); the surface was rinsed with water and dried under a stream (2) Herna´ndez, P.; Vicente, J.; Herna´ndez, L. Fresenius Z. Anal. Chem. 1989, 334, 550-3. (3) Herna´ndez, P.; Gala´n-Estella, F.; Herna´ndez, L. Electroanalysis 1992, 4, 45-9.

Analytical Chemistry, Vol. 68, No. 7, April 1, 1996 1267

of air before each voltammetric experiment. A platinum microdisk electrode (100 µm diameter) was obtained from Microglass Instruments (Greensborough, Victoria, Australia). The microdisk experiments were performed in a Faraday cage, and the current was measured with the aid of a preamplifier (PA-1, Bioanalytical Systems Inc.). The auxiliary electrode was a platinum wire. The reference electrode consisted of a silver wire immersed in a solution of AgNO3 (0.010 M) and Et4NBF4 (0.10 M) in acetonitrile and was separated from the analyte solution by a salt bridge containing the electrolyte solution. The ferrocene/ferricenium couple has a reversible potential of 87 ( 2 mV vs the Ag/Ag+ reference electrode with a peak-to-peak separation of 68 ( 1 mV, as obtained at a glassy carbon electrode at a scan rate of 100 mV s-1 for a solution of 5.0 × 10-4 M ferrocene in acetonitrile. Solutions used in all electrochemical experiments were purged with solventsaturated nitrogen (high purity, CIG, St. Leonards, NSW, Australia). A nitrogen pressure was maintained over the solution during all experiments. Both a mercury pool (area of 11 cm2) and a piece of glassy carbon (area of 18 cm2) were used as working electrodes in controlled-potential electrolyses (CPE). Electron Paramagnetic Resonance Spectroscopy. Electron paramagnetic resonance (EPR) spectra were obtained with a Bruker ECS 106 spectrometer (Rheinstetten, Germany). Paramagnetic species were generated using the electrochemical cell described by Fiedler et al.4 and detected in situ in the cavity of the spectrometer. A solution of tetramethrin (5.0 × 10-3 M) in acetonitrile (0.10 M Et4NBF4) was purged with argon and added to the evacuated cell under an argon atmosphere. The radical species was generated at a constant applied potential (-1.7 V vs Ag/AgCl) at room temperature using a BAS CV27 potentiostat (Bioanalytical Systems). Determination of Tetramethrin in an Insecticide Formulation. A Metrohm 646 VA processor controlling a 647 VA stand was used for the voltammetric determination of tetramethrin. The three electrodes consisted of glassy carbon working and auxiliary electrodes and the Ag/Ag+ reference electrode described previously. A 20.0 µL aliquot of the liquid formulation was added to an electrochemical cell containing 20.0 mL of acetonitrile (0.10 M Et4NBF4). The solution was purged with solvent-saturated nitrogen for 5 min prior to the application of a differential pulse scan (pulse amplitude of -50 mV, scan rate of 20 mV s-1) over the potential range of -1400 to -2000 mV vs Ag/Ag+. Under these experimental conditions, the peak current for the reduction of tetramethrin was measured at -1695 mV vs Ag/Ag+. Six separate additions of a tetramethrin solution (20.0 µL of 3.44 g L-1 tetramethrin in the electrolyte solution) were made, with nitrogen purging for 90 s after each addition. The solution was purged and the electrode rotated for 15 s after each scan. The concentration of tetramethrin was determined from a plot of the average peak current and peak currents from added standards, using the method of standard addition to minimize any matrix effects from other compounds contained in the formulation. RESULTS AND DISCUSSION Voltammetry. In acetonitrile (0.10 M Et4NBF4) at a HMDE, under conditions of cyclic voltammetry, tetramethrin exhibits three (4) Fiedler, D. A.; Bond, A. M.; Koppenol, M. J. Electrochem. Soc. 1995, 142, 862-7.

1268 Analytical Chemistry, Vol. 68, No. 7, April 1, 1996

Figure 1. Voltammograms obtained at a scan rate of 100 mV s-1 for reduction of tetramethrin (5.0 × 10-4 M) in acetonitrile (0.10 M Et4NBF4): (a) cyclic voltammogram at a HMDE with a switching potential of -2500 mV vs Ag/Ag+; (b) cyclic voltammogram at a HMDE with a switching potential of -1900 mV vs Ag/Ag+; (c) linear sweep voltammogram at a glassy carbon rotating disk electrode with an electrode rotation rate of 2000 rpm.

reduction processes with peak potentials of -1685, -2145, and -2305 mV vs Ag/Ag+ at a scan rate of 100 mV s-1 (Figure 1a). Under identical experimental conditions, tetramethrin is reduced at a glassy carbon electrode with peak potentials of -1690, -2220, and -2315 mV vs Ag/Ag+. The reverse scan shows the presence of a single oxidation process with a peak potential of -1610 mV vs Ag/Ag+ at the HMDE and -1615 mV vs Ag/Ag+ at the glassy carbon electrode. No other oxidation peaks are observed prior to the oxidation of tetramethrin5 at +1200 mV vs Ag/Ag+. Cyclic voltammetric data obtained at a HMDE at a range of scan rates and switching potentials are summarized in Table 1. A linear relationship is obtained between the peak current of the first electron-transfer process and the square root of the scan rate. The relationship is consistent with a diffusion-controlled process. To study the electrochemical and chemical reversibility of this first reduction process, cyclic voltammetric experiments were performed over a range of scan rates with a switching potential of -1900 mV vs Ag/Ag+. The voltammogram obtained at the HMDE, at a scan rate of 100 mV s-1, is shown in Figure 1b. When the potential is switched prior to the onset of the second process,

Table 1. Cyclic Voltammetric Data Obtained for the Reduction of Tetramethrin (5.0 × 10-4 M) in Acetonitrile (0.10 M Et4NBF4) at the HMDE with Switching Potentials of -2500 and -1900 mV vs Ag/Ag+a switching potential of -2500 mV vs Ag/Ag+ scan rate (mV s-1) 20 50 100 200 500 1003 2007 a

reduction process 1 Ered ired p p -1685 -1685 -1680 -1680 -1685 -1680 -1685

1.2 1.7 2.4 3.5 5.1 6.9 9.9

reduction process 2 Ered ired p p -2125 -2140 -2145 -2160 -2180 -2190 -2200

2.1 3.1 4.5 6.7 9.9 13 19

reduction process 3 Ered ired p p -2280 -2300 -2305 -2310 -2330 -2340 -2345

0.20 0.37 0.63 0.92 1.6 2.5 4.1

switching potential of -1900 mV vs Ag/Ag+

oxidation process on reverse scan Eox iox p p -1610 -1615 -1610 -1610 -1605 -1610 -1610

0.57 0.82 0.92 0.88 1.1 1.3 1.7

reduction process 1 Ered ired p p -1685 -1690 -1685 -1685 -1685 -1680 -1685

1.2 1.9 2.5 3.5 5.2 7.4 10

oxidation process on reverse scan Eox iox p p -1615 -1615 -1620 -1620 -1615 -1615 -1605

1.2 1.8 2.3 3.2 4.8 6.9 9.5

E°′

∆Ep

ox ired p /ip

-1650 -1650 -1650 -1650 -1650 -1650 -1645

70 75 65 70 70 65 80

1.0 1.0 1.1 1.1 1.1 1.1 1.1

The Ep, E°′, and ∆Ep potential values are expressed in units of millivolts and the peak current values in units of microamperes.

the reduction and oxidation peak currents measured for the first electron-transfer process are equivalent within experimental error. An average peak-to-peak potential separation of 70 mV was obtained over the range of scan rates, which is essentially the same as for the known reversible one-electron oxidation of ferrocene. All results are consistent with the behavior expected for a diffusion-controlled, one-electron electrochemically reversible reduction. A value for the reversible potential, E°′, of -1650 mV vs Ag/Ag+, which is independent of the scan rate, was calculated for the first reduction process at a HMDE from the average of the reduction and oxidation peak potentials. At a glassy carbon electrode, the E°′ value was calculated to be -1655 mV vs Ag/Ag+. For cyclic voltammetry with a switching potential of -2500 mV vs Ag/Ag+ (Figure 1a), the ratio of the reduction to oxidation peak current values of the initial reduction response was not unity. The product of the initial electron transfer therefore can be concluded to be consumed in the more negative reduction processes. Steady-state voltammetric experiments were performed at both a rotating disk electrode (RDE) and a microdisk electrode. Only two reduction waves were obtained at a glassy carbon RDE (Figure 1c). The waves had E1/2 values of -1660 mV vs Ag/Ag+ (consistent with the E°′ value obtained from cyclic voltammetry) and -2200 mV vs Ag/Ag+, respectively. The E3/4-E1/4 value obtained for the first reduction process was 55 mV (again consistent with the reversible transfer of one electron) and for the second reduction process was 88 mV. Under the conditions of Figure 1c, the limiting current for the initial process was 41 ( 2 µA and for the second process was 79 ( 2 µA. As the initial process has been proven to involve the transfer of one electron, the limiting current observed for the second reduction wave involves the transfer of two electrons. Thus, at the RDE the observed voltammetry is consistent with an overall three-electron reduction process. However, the third reduction process observed with a stationary electrode (Figure 1a) is absent when the electrode is rotated rapidly. Presumably this reflects the shorter effective time scale of the RDE experiment. Voltammetry at a platinum microdisk electrode yielded an E1/2 value of -1670 mV vs Ag/Ag+ and an E3/4-E1/4 value of 53 mV for the initial reduction process. Again, this result confirms that the first process is a well-defined reversible one-electron reduction at a range of electrode materials, techniques, and voltammetric time domains. From the microdisk electrode experiment and using the relationship il ) 4nFDCor [where il is the limiting current

(A), n is the number of electrons, F is Faraday’s constant (96485 C mol-1), Co is the bulk concentration of the electroactive species (mol cm-3), r is the radius of the microdisk electrode (cm), and D is the diffusion coefficient (cm2 s-1)]6, we calculated a diffusion coefficient of 2.8 × 10-5 cm2 s-1 for tetramethrin in acetonitrile. CPE of solutions of a high tetramethrin concentration (5 × 10-3 M) resulted in the formation of a yellow-green species at the electrode surface which was observed in the bulk solution for 20-30 min after the potential was switched off. In the presence of the colored species, a single wave having the same E1/2 value as prior to the electrolysis, but now exhibiting an oxidation current, is observed by means of rotating disk voltammetry. The value of the anodic limiting current obtained from rotating disk voltammetry diminishes with the fading of the colored species. This result demonstrates that the same species is formed initially in synthetic and voltammetric time-scale experiments and that the one-electron reduction product is yellow-green in color but only moderately stable. Exhaustive CPE of tetramethrin (-1900 mV vs Ag/Ag+), at either a mercury pool or glassy carbon electrode, involved the transfer of 1.7 ( 0.2 electrons per molecule as determined by coulometry data. Clearly, this is more than the one-electron process observed under voltammetric conditions. For electrolysis of solutions of low concentrations of tetramethrin (5 × 10-4 M) the yellow-green color of the bulk solution can no longer be observed after the transfer of approximately 1 electron per molecule, as monitored by coulometry. Additionally, the solution color fades rapidly if the electrolysis cell is opened to the atmosphere, indicating that the product is air sensitive as might be expected for a species produced at very negative potentials. The fact that an n value of >1 is obtained in bulk electrolyses implies that a decomposition product of the yellowgreen material also is reduced at an applied potential of -1900 mV vs Ag/Ag+. One product isolated and characterized from the reduction process was the chrysanthemate [(1RS)-cis,trans-2,2dimethyl-3-(2-methyl-prop-1-enyl)cyclopropanecarboxylate] anion.5 However, no products resulting from the tetrahydrophthalimide moiety of tetramethrin could be characterized.5 On the basis of the electrochemical studies the initial reduction process of tetramethrin in acetonitrile may be summarised as Reaction Scheme 1. Electron Paramagnetic Resonance Spectrometry. The voltammetric results indicate that the initial electron-transfer (5) Coomber, D. C. The Electrochemical Behaviour of the Pyrethroid Insecticides. Ph.D. Thesis, Deakin University, Geelong, Australia, 1995. (6) Aoki, K.; Osteryoung, J. J. Electroanal. Chem. 1981, 122, 19-35.

Analytical Chemistry, Vol. 68, No. 7, April 1, 1996

1269

Scheme 1. Mechanism for Reduction of Tetramethrin in Acetonitrile

Figure 2. Electron paramagnetic resonance spectrum obtained after electrolysis of tetramethrin (5.0 × 10-3 M) in acetonitrile (0.10 M Et4NBF4) at a platinum working electrode at an applied potential of -1.7 V v Ag/AgCl.

process causes the formation of a moderately stable radical anion. The in situ EPR spectrum obtained for the initial reduction process of tetramethrin in acetonitrile, when the platinum electrode potential is held at -1.7 V vs Ag/AgCl, is shown in Figure 2. The yellow-green solution color observed during previously described CPE experiments was clearly visible at the platinum working electrode in the electrochemical EPR cell. The spectrum obtained consists of 30 lines with a g value of 2.0058. The spectrum appears as a quintet of triplets with each peak showing further hyperfine coupling. The triplets are in a ratio of 1:1:1, which is consistent with hyperfine coupling caused by the nitrogen atom. The value of the coupling constant of the triplet was 2.47 G. The quintet is in the ratio of 1:2:3:2:1 which is consistent with coupling to four equivalent atoms of I ) 1/2. This coupling can be assigned to the four protons at positions 3 and 6 of the phthalimide ring. The coupling constant has a value of 7.37 G. Further coupling, with a total value of 0.38 G may be due to a combination of the CH2 group bonded to the nitrogen atom and the protons on the phthalimide ring at positions 4 and 5. The EPR spectra of electrochemically and chemically generated radical anions of phthalimide and substituted phthalimides have been previously reported.7-12 The formation of the radical anion of some N-alkyl-9 and N-phenylphthalimides10 is associated with the formation of a yellow-green color in the electrolysis solution. In general, hyperfine couplings with the nitrogen atom, protons within the phthalimide ring, and protons on the N substituent are observed.7-9,11,12 For example, Hirayama9 reported the coupling constants obtained for the electrochemically generated radical of methylphthalimide to be 2.62 G for the nitrogen, 0.25 G for the protons at positions 3 and 6, 2.41 G for the protons at positions 4 and 5, and 0.91 G for the methyl protons. In contrast to this present study, only a small coupling constant is observed for the protons in positions 3 and 6 in the fully delocalized system. For tetramethrin, the coupling constant for these protons is large, 7.37 G, most likely as a consequence of the radical not being delocalized throughout the entire phthalimide ring system. The EPR experiment implies that the initial electron transfer introduces the electron into the phthalimide system where it exists (7) Nelsen, S. F. J. Am. Chem. Soc. 1967, 89, 5256-9. (8) Sioda, R. E.; Koski, W. S. J. Am. Chem. Soc. 1967, 89, 475-81. (9) Hirayama, M. Bull. Chem. Soc. Jpn. 1967, 40, 1557-62. (10) Horner, L.; Singer, R. J. Tetrahedron Lett. 1969, 20, 1545-7. (11) Farnia, G.; Romanin, A.; Capobianco, G.; Torza, F. J. Electroanal. Chem. 1971, 33, 31-44. (12) Shimozato, Y.; Shimada, K.; Szwarc, M. J. Am. Chem. Soc. 1975, 97, 58313.

1270 Analytical Chemistry, Vol. 68, No. 7, April 1, 1996

Figure 3. Differential pulse voltammograms used for the determination (three additions) of tetramethrin in an insecticide formulation [20.0 µL added to 20.0 mL of acetonitrile (0.10 M Et4NBF4)] at a glassy carbon electrode: pulse amplitude of -50 mV, scan rate of 20 mV s-1, and standard addition of 20.0 µL of a 3.44 g L-1 tetramethrin solution in acetonitrile. The peak current was measured at -1695 mV vs Ag/Ag+.

in a delocalized form. Apparently, neither the chrysanthemate moiety nor the ester group are the sites where reduction takes place. The electrochemical behavior of phthalimide13 and phthalimide derivatives11,14 in nonaqueous solutions is well documented. The phthalimide derivatives are reduced in two discrete one-electron steps to the stable radical anion and dianion.11,14 Limit of Detection. The detection limit (3σ) for tetramethrin in acetonitrile, calculated using the technique of differential pulse voltammetry (pulse amplitude of 50 mV, scan rate of 5 mV s-1, sample width of 20 ms, pulse width of 60 ms, pulse period of 1 s), was 2.1 × 10-6 M at a glassy carbon electrode and 9.6 × 10-7 M at a HMDE. The limit of determination (10σ) was 3.5 × 10-6 M at a glassy carbon electrode and 3.0 × 10-6 M at a HMDE. The initial reduction process of tetramethrin does not cause fouling of the electrode surface or a loss of precision for repetitive current measurements. Additionally, being a reversible process in acetonitrile, it should be ideally suited for determination by transient voltammetric techniques such as differential pulse voltammetry. Determination of Tetramethrin in an Insecticide Formulation. Tetramethrin was determined in a formulation, at a glassy carbon electrode, by means of differential pulse voltammetry. The voltammograms obtained for the sample and the first three standard additions are shown in Figure 3. From the plot of peak current against concentration of added tetramethrin, a concentration (w/v) of tetramethrin of 0.34 ( 0.02% was obtained. The (13) Lasia, A. J. Electroanal. Chem. 1974, 52, 229-36. (14) Leedy, D. W.; Muck, D. L. J. Am. Chem. Soc. 1971, 93, 4264-70.

results are in good agreement with the stated tetramethrin concentration (w/v) of 0.350 ( 0.018% as determined using chromatographic methods. The voltammetric determination at the glassy carbon electrode is inherently simple and only requires direct dissolution of the formulation into the acetonitrile. Importantly, at the glassy carbon electrode, no interference to the initial reduction process is observed from matrix constituents. Mercury has been shown to be a useful electrode for the reduction of pure tetramethrin solutions. Unfortunately, at a mercury electrode the tetramethrin reduction peak is obscured by a peak arising from the presence of the formulation matrix. However, the reduction response arising from the matrix is sufficiently removed from the reduction peak for tetramethrin at a glassy carbon electrode to enable direct determination without the need for chromatographic or other forms of separation. In summary, tetramethrin may be electrochemically reduced in the aprotic, nonaqueous solvent acetonitrile. The initial reduction process involves the formation of a radical anion on the voltammetric time scale. This reduction response was successfully

applied to the selective determination of tetramethrin in an insecticide formulation. The determination method recommended requires only simple dissolution of the sample in acetonitrile without the requirement for a separation technique or other sample preparation and the use of differential pulse voltammetry at a glassy carbon electrode. ACKNOWLEDGMENT Financial support (D.C.C.) was provided by the Australian Wool Research and Development Corp. through the award of postgraduate scholarship UDG002. The donation of tetramethrin and the sample formulation by Reckitt and Colman Products Pty. Ltd. is gratefully acknowledged. The EPR spectrum was obtained by Dr. Dirk Fiedler of La Trobe University. Received for review July 6, 1995. Accepted January 2, 1996.X AC9506701 X

Abstract published in Advance ACS Abstracts, February 1, 1996.

Analytical Chemistry, Vol. 68, No. 7, April 1, 1996

1271