August, 19581 TOBACCO AND TOBACCO SMOKE. PART I 447 Determination of DDT and Chlorobenzilate Occurring Together in Spray Deposits BY E. A. BAKER AND E. JOHN SKERRETT (Research Station, Long Ashton, Bristol) Mixtures of DDT and chlorobenzilate from spray deposits on plant surfaces are separated on alumina columns, and the two constituents are determined by procedures based on the colours developed from their nitration products. CHLOROBENZILATE (ethyl 4 : 4’-dichlorobenzilate, I) is a non-phytocidal spray material developed for the control of mite infestations of plants. It has been used successfully for controlling the citrus-bud mite and may prove to be of value against other species, such as the big-bud mite of blackcurrants. For this latter purpose it would be convenient to use DDT [l : 1 : 1-trichloro-2 : 2-di(9-chlorophenyl)ethane, 111 in conjunction with the acaricide for the simultaneous control of the currant-leaf midge.In order to assess the distribution and the levels of the deposits, it is necessary to have available methods for the determination of DDT and chlorobenzilate when used together in such sprays. Several methods have been described for the determination of DDT1s2J and chlorobenz- ilate4J occurring singly in spray deposits on plants, but Harris4 has been the only worker to consider mixtures of the two. He separated chlorobenzilate from DDT by treatment with ethanolic potassium hydroxide solution. This reagent hydrolysed the former to the potassium salt of the acid I11 and dehydrohalogenated the DDT to DDX (IV).OH H OH KO2 NO, I I cc1, After dilution with water, the DDX was removed with light petroleum, and, after The acid was then nitrated to acidification, the acid I11 was extracted with diethyl ether.448 BAKER AND SKERRETT: DETERMINATION OF DDT AND [Vol. 83 a tetranitro compound (V), which was determined colorimetrically. The method did not measure the DDT present, but our preliminary experiments showed that the DDX removed in the light petroleum washes could have been used to determine the original level of DDT. However, chromatographic methods of separation were investigated in order to reduce both time and apparatus requirements. It was thought that such a separation was feasible, as chlorobenzilate, with its tertiary carbinol hydroxyl group, might be strongly adsorbed on a hydrophilic column through which DDT would pass.EXPERIMENTAL Initial experiments with silica and alumina columns indicated a quantitative separation of chlorobenzilate from DDT in carbon tetrachloride solution, the final traces of DDT being eluted from the column with small portions of carbon tetrachloride. Unfortunately, in later experiments with plant extracts, silica columns failed to retain materials that interfered with the colorimetric determination of DDT, and the use of this chromatographic medium had to be abandoned. Incomplete recoveries of chlorobenzilate from alumina columns in early stages of the work were attributed to the formation of the insoluble sodium salt of the acid, 111, with traces of alkali in the alumina. This trouble was overcome by carefully neutralising, washing, drying and conditioning the alumina before use.6 During the preliminary experiments, it was found to be desirable to modify slightly the established methods for the individual toxicants.In Harris's method4 for the deter- mination of chlorobenzilate, the nitration mixture is slowly warmed over a period of 30 minutes to 85" C and then heated in a steam-bath for 1 hour. It was found that the initial heating period could be omitted and the flask containing the reaction mixture placed directly into boiling water. However, under these conditions a minimum heating period of 1 hour was necessary. Saturated sodium sulphate solution was used to dilute the reaction mixture ; this gave sharper and much more rapid separations during the ether extractions. Difficulties were caused by the growth of Fusarium spp.in the sodium sulphate solution, which was prevented by bubbling sulphur dioxide through the solution for a short time. This development of mould was rather surprising, although a few analogous instances'$* have been reported. 2-Ethoxyethanol (ethyl Celllosolve) was preferred to benezene as solvent for the nitration product, which has been showns to be a tetranitrobenzophenone derivative (V). Methanolic potassium hydroxide solution was preferred for the colour development, as it gave less trouble from carbonate precipitation than did the ethanolic solution. The maximum colour developed after 30 minutes and was stable for a further 2 hours. Determination of DDT was based on the nitration method of Schechter, Soloway, Hayes and Haller.' When solutions of plant extract were evaporated, it was found to be desirable to add a little oxalic acid to prevent loss of insecticide. (The stearic acid used by Harris for a similar purpose in the determination of chlorobenzilate was found to cause interference in the final colour development.) The time of nitration was reduced to 10 minutes,l0 and the cooled acid mixture was diluted with saturated sodium sulphate solution.2-Ethoxyethanol was again used to dissolve the nitration product, the final blue colour being developed with 5 per cent. ethanolic potassium hydroxide solution and measured at 390 mp with a Unicam SP600 spectrophotometer. In both procedures it was found that traces of rubber, silicone and Apiezon stopcock lubricants interfered with the final dissolution of the nitro compounds.However, a stiff paste of bentonite and analytical-reagent grade glycerol proved to be satisfactory, and the addition of a glass bead appeared to be a further aid to the dissolution of residues. METHOD APPARATUS- All-glass apparatus is used throughout. Chromatographic columns are prepared in tubes, as shown in Fig. 1. The small glass projections are of use when the tubes are positioned in the Bolton extractors during the extraction stage.August, 19581 CHLOROBENZILATE OCCURRING TOGETHER I N SPRAY DEPOSITS 449 I I REAGENTS- Use a Widmer column with a proportionating head for all fractionations. Diethyl ether-Purify technical grade ether by fractionation over sodium.Methanol-Purify B.P. grade methanol by fractionation after the addition of sodium. Ethanol-Purify technical grade absolute ethanol by fractionation after the addition of sodium. 2-Ethoxyethanol-Purify technical grade ethyl Cellosolve by distill- ation after the addition of sodium. Carbon tetrachloride-Purify the analytical-reagent grade material by fractionation. Nitration mixture-Cautiously add an equal volume of analytical- reagent grade concentrated sulphuric acid to well stirred fuming nitric acid, also of analytical-reagent grade. Stearic acid solution-Twice recrystallise stearic acid from ethanol, and prepare a 0.5 per cent. solution in light petroleum (boiling range 40" to 60" C). Oxalic acid solution-Prepare a 0.5 per cent. solution of analytical- reagent grade oxalic acid in acetone.- i m m im - Fig. 1. Modified Sodzum sulphate solution, saturated-Saturate distilled water with chromatographic tube sodium sulphate, and pass sulphur dioxide through the solution for a few minutes. Methanolic potassium hydroxide solution, 5 per cent.-Freshly prepare this solution in the cold, and filter before use. Ethanolic totassium hydroxide solution, 5 t e r cent.-Freshly prepare this solution in the cold, and filter before use. Alumina-Stir 500 g of H-grade alumina with 300 ml of 0.2 per cent. v/v hydrochloric acid for 2 hours. Allow the mixture to settle, and then decant the supernatant liquid. Repeat this procedure with a further 300 ml of 0.2 per cent. hydrochloric acid. Wash the alumina twice with 300-ml portions of distilled water, stir it with 300 ml of distilled water and then with 300ml of 0.3 per cent.v/v acetic acid for 1 hour. Filter the alumina on a Buchner funnel, and dry at 120" C. Break up any lumps, spread in a thin layer on a silica tray and heat a t 600" C for 3 hours. Allow to cool from approximately 200" C in a closed container, and condition the alumina to Brockmann activity IIP by placing a beaker containing 15 ml of water in the container for 3 days. Stopcock lubricant-Add sufficient bentonite to analytical-reagent grade glycerol to form a viscous paste after being stirred. PROCEDURE FOR SEPARATING THE MIXTURE- Extract 1 g of the plant material for 15 minutes in a Soxhlet extractor with 25 ml of carbon tetrachloride. Lightly plug the bottom of a chromatographic tube with cotton-wool that has been extracted with hot ethanol, add 5 g of alumina and tap the column to ensure that the packing is uniform. Use another small plug of cotton-wool to prevent displacement of the top of the column.Pour the cooled plant extract on to the column, and wash with five 5-ml portions of carbon tetrachloride. Collect the eluate, and use it for the determination of DDT. Place the column in a Bolton extractor, and extract for 2 hours with 35 ml of hot ethanol. Use this extract for the determination of chlorobenzilate. PROCEDURE FOR DETERMINING DDT- Add 2 ml of oxalic acid solution and a glass bead to the eluate, and evaporate to dryness at approximately 55" C and a pressure of 20 cm uf mercury. To the cooled residue add 2 ml of nitration mixture, and carefully rotate the flask to wet any particles of solid.Im- merse the flask in a boiling-water bath for 10 minutes, occasionally swirling the contents. Cool the flask in an ice - water mixture, and add 75 ml of saturated sodium sulphate solution. Transfer the contents of the flask to a separating funnel with 70 ml of diethyl ether. Shake, discard the aqueous layer, and wash the ethereal layer successively with 25 and 1Oml of 5 per cent. potassium hydroxide solution and 15 ml of saturated sodium sulphate solution.450 BAKER AND SKERRETT [Vol. 83 Run the ethereal solution through a 15-mm layer of anhydrous sodium sulphate into a 100-ml round-bottomed flask containing a glass bead of diameter 5 mm. Evaporate the solution to dryness in a water bath at 55" C, and add 0.5 ml of 2-ethoxyethanol.Rotate the flask so that the bead assists the dissolution of the residue, cool, and then add 10.0 ml of 5 per cent. ethanolic potassium hydroxide solution. After 5 minutes, measure the colour at 390 mp with a Unicam SP600 spectrophotometer, and interpolate the result on a curve prepared from standard mixtures of plant extracts and :DDT (m.p. 107.5" C). PROCEDURE FOR DETERMINING CHLOROBENZILATE- Add 2ml of stearic acid solution to the extract obtained from the separation, and evaporate the mixture to dryness at approximately 55' C and a pressure of 20 cm of mercury. To the cooled residue add 5 ml of nitration mixture, and immerse the flask in a boiling-water bath for 1 hour. Cool the flask in an ice - water mixture, and add 75 ml of saturated sodium sulphate solution.Transfer the contents of the flask to a separating funnel with 70ml of diethyl ether, and shake. Discard the aqueous layer, and wash the ethereal layer successively with 25 and 10 ml of 5 per cent. potassium hydroxide solution and 15 ml of saturated sodium sulphate solution. Run the ethereal solution through a 15-mm layer of anhydrous sodium sulphate into a 100-ml round-bottomed flask containing a gla!js bead of diameter 5 mm. Evaporate the solution to dryness at 55" C, and add 0.5 ml of 2-ethoxyethanol. After swirling the bead round the flask to assist dissolution, add 10.0 ml of 5 per cent. methanolic potassium hydroxide solution. After 30 minutes, measure the colour developed at 538 mp with a Unicam SP600 spectrophotometer, and determine the chlorobenzilate by using a curve prepared from standard mixtures of plant extracts and chlorobenzilate (m.p.37" to 38.5" C). RESULTS Mixtures prepared by adding known amounts of DDT and chlorobenzilate to extracts of 1 g of untreated leaves were analysed by the proposed method. The results shown in Table I indicate good recoveries. TABLE I RECOVERIES OF DDT AND CHLOROBENZILATE FROM SYNTHETIC MIXTURES Amount of Amount of Amount of Amount of DDT added, chlorobenzilate added, DDT found, chlorobenzilate found, Crg r g r g Crg 10 10 10.5, 10.0 10.0, 9.8 20 20 21.0, 20.5 19.9. 19.7 50 50 49.7, 49.5 49.7, 49.5 75 75 73.5, 72.5 72.5, 72.0 100 100 98.0. 97.0 98.0, 95.0 200 200 193, 192 197, 196 We thank Professor H. G. H. Kearns and Dr. J. T. Martin for their interest in the work, and Messrs. Geigy Ltd. for the gift of a sample of chlorobenzilate. REFERENCES 1. 2. 3. 4. 6 . 6. 7. 8. 9. 10. Schechter, M. S., Soloway, S. B., Hayes, R. A., and Haller, H. L., Ind. Eng. Chsm., Anal. Ed., Butterfield, D. E., Parkin, E. A., and Gale, M. M., J . Soc. Chew. Ind., 1949, 68, 310. Amsden, R. C., and Walbridge, D. J., J . Agric. Food Chem., 1954, 2, 1323. Harris, H. J., Ibid., 1956, 3, 939. Blinn, R. C., Gunther, F. A., and Kilbezen, M. J., Ibid., 1954,2, 1080. Brockmann, H., and Schodder, H., Ber., 1941, 74, 73. Starkey, R. L., and Waksman, S. A., J . Bact., 1943, 45, 509. Pulst, C., Jahr. wiss. Botan., 1902, 37, 205. Skerrett, E. J., and Baker, E. A., Ann. Rep. Agric. Hort. Res. Sta., Bristol, 1958, 130. Martin, J. T., and Batt, R. F., Ibid., 1953, 121. 1945, 17, 704. Received January loth, 196.8