ANALYST. APRIL 1989, VOL. 114 425 Simultaneous Determination of Fenamiphos, its Sulphoxide and Sulphone in Water by High-performance Liquid Chromatography Rai Singh* Soil Science and Plant Nutrition, University of Western Australia, Nedlands, Western Australia 6009, Australia A method for the simultaneous determination of fenamiphos and its two metabolites, fenamiphos sulphoxide and fenamiphos sulphone, in water is described. The proposed method is based on the separation of the compounds on a silica gel column with ultraviolet detection and overcomes the problems of inadequate separation and possible decomposition associated with the gas-chromatographic determination of these compounds. The application of the method to groundwaters showed that a minimum detection level of the order of 10 pg 1-1 could be achieved easily with pre-concentration of the samples on Sep-Pak C18 disposable cartridges.Keywords: Fenamiphos analysis; fenamiphos metabolites; organophosphorus pesticides; high-perfor- mance liquid chromatography Fenamiphos [ethyl 3-methyl-4-(methylthio)phenyl isopropyl- phosporamidate, Nemacur] is a broad-spectrum. non-volatile, neniaticide insecticide and is commonly used for nematode control in horticultural crops. In soils fenamiphos is oxidised, both microbiologically and chemically, to fenamiphos sul- phoxide (f. sulphoxide) and then to fenamiphos sulphone (f. sulphone). The structural formulae of the three compounds are illustrated in Fig. 1. Both metabolites have nematicidal activity' and are toxic to man.2 The oral LDSo (rats) of the sum of fenamiphos, its sulphoxide and sulphone, expressed as fenamiphos, is 15-19 mg kg-1 as reported by Krause el a / .; The oxidation products are more polar and have more mobility in soil than the parent compound.4 With some gas-chromatographic methods reported in the literature. either the parent compound only,5 or the sulphone as the total residue after permanganate oxidation of both the parent compound and the sulphoxide,6 are determined. Brown' reported a gas-chromatographic method for the determination of fenamiphos and its sulphoxide and sulphone using column chromatography with silica gel to separate the sulphoxide and sulphone. Satisfactory separation of the sulphoxide and sulphone has not been possible on both packed' and capillary8 GC columns. Some aspects of the gas - liquid chromatographic analysis of the sulphone have been discussed by Krause.9 Sulphoxide and sulphone species of organophosphorus pesticides are difficult to determine because of their low volatility, decomposition at elevated temperatures10 and difficult separation.8 High-performance liquid chromatography (HPLC) would appear to be a more suitable alternative to gas - liquid chromatography (GLC).To my knowledge, no HPLC method for the simultaneous determination of fenamiphos and its metabolites has yet been reported. This paper describes the simultaneous determination of fenamiphos, f. sulphoxide and f. sulphone in water by HPLC. Experimental Reagents Fenamiphos, f . sulphoxide and f . sulphone (99% pure) were obtained from Bayer Australia (Botany, NSW, Australia).Liquid-chromatographic grade methanol, acetonitrile and phosphoric acid were obtained from BDH (Sydney, Aus- tralia). Water used to prepare HPLC eluents was distilled twice using a silica still (TQS, Wiesbaden. FRG). On leave from Haryana Agricultural University. Hisar. India. 0 CH3 II / - P - N H -CH I \ O-CC2H5 CH3 CH, Fig. 1. f. sulphone (111) Structural formulae of fenamiphos (I); f. sulphoxide (11) and Sample Pre-concentration Water samples were passed through a 0.45-pm membrane filter (type SM 11306, Sartorius, Gottingen, FRG) and spiked with fenamiphos, f. sulphoxide and f. sulphone to concentra- tions ranging from 2 to 100 pg 1-1. The samples were then passed through Sep-Pak CI8 disposable cartridges (Millipore, NSW, Australia) at a rate of 3-5 ml min-1.The cartridges were pre-conditioned with 5 ml of methanol followed by 5 ml of water. The samples thus concentrated were extracted with 1 ml of acetonitrile and were ready for analysis. Liquid Chromatography Samples were analysed using a high-performance liquid chromatograph (ETP-Kortec, Sydney, Australia) consisting of an HPLC pump (K 35M), an automatic sampler (K 65) and a variable wavelength UV detector (K 95). Data were processed with an LDC/Milton Roy integrator. The compounds were separated on a Spherisorb silica gel column (Phase Separation, Clwyd, UK) (250 X 4.6 mm i.d., particle size 5 pm). The following chromatographic conditionsANALYST. APRIL 1989, VOL. 114 were used: eluent, acetonitrile - water (20 + 80, V / V ) ; pH, 2.2 (adjucted with phosphoric acid): flow-rate, 0.5 ml min-1; detector wavelength, 200 mm: injection volume, 200 p1; temperature.ambient = 20 & 2°C. A Sphericorb ODS (C,,) column (250 X 4.6 mm i.d., particle size 10 pm) (Phase Separation) was also used to separate the compounds. The following conditions were used: eluent, methanol - water (65 + 35, WV); flow-rate, 1.5 ml min-I. Other conditions were as used for the silica gel column. Gas Chromatography - Mass Spectrometry The identities of peaks of fenamiphos, t. sulphoxide and f. sulphone were verified on a gas chromatograph equipped with a mass-selective detector and an H-P 5 column (25 m x 0.31 mm i.d.)? by collecting the individual compounds as they eluted from the HPLC column. The conditions were as follows: injection port temperature, 250 "C: initial column temperature, 175 "C (programmed to 250 "C at 25 "C min-1); carrier gas, helium (44 cm s-1).Detection was carried out with a single-ion monitoring mass spectrometer (Hewlett-Packard 5970) at rti'z 303.2, 303.2 and 320.2 for fenamiphos, f. sulphoxide and f. sulphone, respectively, on the basis of maximum abundance of ions. Results and Discussion A spectrophotometric scan of fenamiphos, f. sulphoxide and f. sulphone at wavelengths between 190 and 400 nm showed absorption maxima for all three compounds at 200 nm. Fenamiphos and f. sulphone gave a further peak of much lower intensity at 248 and 226 nm. respectively. For the simultaneous determination of the three compounds and considering the peak maxima, a wavelength of 200 nm was selected. Liquid Chromatography The three compounds could not be separated simultaneously on an ODS (Clx) column under isocratic conditions because f.sulphone and f. sulphoxide showed similar retention times. Fig. 2 shows the chromatograms of ( a ) f. sulphone and tenamiphos and of ( h ) f. sulphoxide. The sulphone and sulphoxide could be separated by changing the composition of the mobile phase and the flow-rate but the parent compound did not elute from the column. Gradient elution could not be used because of the problem of base-line drift during analysis. al 0 C m +! 2 1) a a) A 0.01 A T B I I I I 0 5 10 15 20 +- a, .- i\l 0.01 A ! -c I 1 I I 0 5 10 15 tim in Fig. 2. Chromatograms of ( u ) A , f. sulphone and B. fenamiphos.t:ach 5 big nil-' i n water; ( h ) C, f . sulphoxide, 5 b~g mlV1 in water. Conditions: eluent, 65% MeOH in water; column, ClS ODS (particle size. 10 pin); flow-rate. 1.5 ml min I ; detector wavelength. 200 nm; detector sensitivity, 0.08 a.u.f.s.; injection volume, 200 1.11 All three compounds could be separated, isocratically, on a silica gel column. A typical chromatogram is shown in Fig. 3. Base-line resolution between the three compounds was achieved in less than 15 min. Calibration graphs with good linearity were obtained for all compounds in the concentration range studied (0.2-4 ug ml-1). Reproducibility The reproducibility of the proposed method was determined by estimating the variation in four replicate runs of the three compounds at a concentration of 1 ug ml-1.The relative standard deviations ("/o) obtained for f. sulphone, f. sulphox- ide and fenamiphos were 3.0, 4.3 and 3.9, respectively. Long-term reproducibility was also determined by comparing the relative response in peak heights of the three compounds at a concentration of 1 pg ml-1 for a number of assays (12 = 5 ) , over a 4 week period. The relative standard deviations (Yo) obtained were 4.0, 4.3 and 5.6 for f. sulphone, f. sulphoxide and fenamiphos, respectively. No significant change in the retention times of the compounds was observed during this period. Based on these observations, no unfavourable effect of the mobile phase on the long-term performance of the column is expected. The low pH (2.2) of the mobile phase should not affect the performance of the chromatographic column as silica is only soluble significantly at a pH above 9.11 The column, connective tubing and pump in contact with the eluent were made of marine-grade stainless steel (SS 316) and were considered to be resistant to phosphoric acid at pH 2.2.However, it must be noted that a satisfactory separation of the three compounds is possible even in the absence of phosphoric acid but the detector response for the compounds is affected? particularly during analysis of groundwater samples. Application to Groundwater Samples The environmental applicability of the normal phase HPLC method was tested with two groundwater samples from horticultural areas of Western Australia. Groundwater from Carnarvon represents a banana plantation area and Wanne- roo groundwater a vegetable growing area around Perth.In the Carnarvon area fenamiphos is used for bananas at 10 1 of active ingredient (a.i.) per hectare twice a year with irrigation water. In the Wanneroo area the recommended amount for various vegetable crops varies from 5 to 10 1 of a.i. per hectare, once or twice per crop growing season (2-3 crops per year). Total organic carbon contents of Carnarvon and Wanneroo waters were found to be 13.9 and 21.1 pg ml-1, respectively. During analysis the water sample with the lower content of organic matter (Carnarvon) did not give any interfering peaks. al V c m e 2 a n 10.002 A l,i, 0 4 8 12 16 [0.002 A L I I L I I 0 4 8 12 16 timin Fig. 3. Chromatograms o f (a) A. f . sulphone; B. f . sulphoxide; and C. fenamiphos, each 1 pg ml-1 in water; and ( h ) unspiked distilled water.Conditions: eluent, 20% MeCN in water (pH 2.2); column. silica gel (particle size, 5 pin); flow-rate, 0.5 ml min-I; detector wavelength, 200 nm; detector sensitivity, 0.02 a.u.f.s.; injection volume, 200 plANALYST, APRIL 1989. VOL. 113 427 1 I I I I 0 4 8 12 16 0 4 8 '12 16 timin Fig. 4. Chromatograms of Wanneroo groundwater ( a ) spiked with A. f . sulphone: B, f . sulphoxide; and C. fenamiphos, each at a level of 1.0 big 1-1; ( h ) spiked with A, 1 pg 1-1 olf. sulphone; B, 1 pg 1 1 off. sulphoxide: and C, 2 big 1 - 1 of fenamiphos. Peak D is unidentified. Dashed line indicates the chromatogram of an unspiked groundwater sample. Conditions: as for Fig. 3, except detector sensitivity ( a ) 0.04 o r ( h ) 0.08 i1.u.f.s.The other sample, however, produced a significant peak, tailing up to the retention time of fenamiphos. Sample clean-up Cleaning of the samples was attempted with Sep-Pak cart- ridges of florisil and silica with limited success. The extraction of the compounds retained on the cartridges by acetone (as mentioned by Peterson and Winterlins) resulted in a highlq turbid extract due to precipitation. and low recoveries after filtration. Clean-up of the compounds is difficult and there is also the possibility of breakdown of fenamiphos during the clean-up procedure .s Hence camples were processed without clean up. In Fig. 3 , chromatograms of water with a high organic carbon content with and without spiking the three cornpounds after pre-concentrating ( a ) 100 and ( b ) 1000 times are shown.The higher level of pre-concentration resulted in increased interference [Fig. 4(6)]. No fenamiphos was detec- ted in unspiked samples. Recoveries The recoveries of the three compounds from three different surface u aters and groundwaters were quantified using a silica gel column. The average percentage recoveries from the water samples spiked at various levels with the three compounds are given in Table 1. Better than 73% recoveries for all three ;ompounds were obtained with different quality waters. Generally, low recovery values with higher variability were obtained in groundwaters at the lowest spike level (2 pg I-'). The recoirery was sufficiently reproducible and for ground- water samples a correction factor for 100% recovery can be applied.The detection limit of the method will depend upon the interference by organic matter. Based on these results however, detection levels of at least 10 pg 1-1 can be achieved from groundwaters. Table 1. Recovery of fenamiphos, f . sulphoxide and f . sulphone from water samples spiked at three levels Recovery, " '% Spike/ Distilled Carnarvon Wanneroo Compound Fig 1 water water water F.sulphone . . 100 87.2(2.3) 82.6(3.7) 94.4(3.5) 10 99.0 (2.7) 107.9 (5.3) X2.0 (3.7) F.sulphoxide . . 100 8X.5(2.9) 96.0(5.6) 92.6(2.1) 10 (17.0 (3.9) 79.2 (3.6) 86.8 (4.2) Fenamiphos . . 100 90.4 (2.4) 92.9 (5.X) 102.6 (8.9) 2 87.9 (4.4) x3.3 (9.9) 72.9 (6.0) 2 X1.2(3.5) 81.8(11.4) XX.S(2.0) 2 84.0 (3.3) 74.7 (2.0) 76.1 ( 1 3) 10 85.8 (3.0) 91.9 (6.2) 101.1 (3.4) * Values in parentheses are the standard errors ((YO) based on three replicate measurements. Conclusion With high - pe r f o r m an ce 1 i q ui d chromatograph y ten am i p h o s and its two major metabolites can be determined, simul- taneoucly. in water samples at low levels.Detection level\ of the order of 10 ug 1-1 can be achieved easily by pre- concentrating the samples on Sep-Pak Cls cartridges. All experiments were conducted in the laboratories of the CSIRO Division of Water Resources, Wembley , Western Australia. The author is grateful to Robert Gerritse and John Adeney (CSIRO) and Graham Aylmore (University of Western Australia) for their assistance and advice. The Australian International Development Assistance Bureau has supported the research by granting a fellowship to the author. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Waggoner, T. B.. and Khasawinah, A. M.. XesidiieRev., 1974, 53, 79. Waggoner, '1'. B., J . Agric. Food Chenz., 1972, 20. 157. Krause, M.. Loubser, J . T.. and de Beer, P. R . , J . Agvic. Food Chem., 1986, 34, 717. Bilkert. J . N., and Rao. P. S. C.. J . Eizviron. Sci. Health.. 19x5, H20, 1. Sagredos, A . N., and Eckert, W. R . , Bcitr. T(ihukfbr.wh.. 1977, 9, 107. Thornton, J . S., J . Agric. Food Chenz., 1971, 19, 890. Brown. M. J . , J. Agric. Food Chem., 1981. 29. 1129. Peterson, D., and Winterlin. W., J . Agric. Food Cheni.. I9X6, 34, 153. Krause, M.. Analysr, 1985. 110, 673. Hill, A . R . C.. Wilkins. J . P. G.. Findlay, N. R. I.. and Lontay, K. E. M., Analyst, 1984, 109, 48.3. Unger, K. K . . "Porous Silica. Its Properties and Use as Support in Column Liquid Chromatography." Elsevier, Amsterdam. 1979. p. 13. Pciper 81032 70J RectGiwi Aiigirst lOth, 1988 Accepred December Sth, 1988