June, 19581 SERGEANT 335 The Determination of Chlorinated Hydrocarbon Pesticide Residues in Plant Material BY G. A. SERGEANT (Department of the Government Chemist, Clement's I n n Passage, Straizd, Loitdon, W.C.2) h method is described for the determination of a number of chlorinated hydrocarbon pesticide residues in plant material. After extraction and partial purification, the residues are analysed for total organic chlorine by a modified Stepanow procedure, followed, after the removal of organic matter and excess of sodium ions, by a volumetric determination of chloride by the mercuric oxycyanide method. PROCEDURES for the determination of organic chlorine that involve reduction with sodium by modifications of the original Stepanow methodl have frequently been published, and that described in a recent report of the Analytical Methods Committee of the Society for Analytical Chemistry2 is a typical application.isoPropyl alcohol is now the usual medium for the reduction, although isobutyP and isoamy14 alcohols and a mixture of dioxan and ethanol- amine5 have also been used. In this paper a further modification of the Stepanow method is described, together with its application on the semi-micro scale to the determination of chlorinated hydrocarbon pesticide residues in plant material. The rapid decomposition of organic halogen compounds by reaction with aromatic sodium compounds has been utilised by Liggett6 and others,'*s and is especially advantageous in the analysis of volatile materials, as the reaction proceeds rapidly even in the cold.Reagents such as sodium diphenyl and sodium naphthalene are, however, troublesome to prepare and store, and, in the proposed application of the reaction, sodium naphthalene is generated directly in the reaction flask instead of being added as a prepared reagent. The resulting technique resembles that of Stepanow, but is much more rapid, partly because the end-point of the reaction is self-indicating. From the limited range of solvents in which the formation of aromatic sodium compounds readily takes place, tetrahydrofuran was selected in preference to 1 : 2-dimethoxyethane. Two other solvents of interest noted in the course of the work were isopropylamine and rt-butylamine. The velocity of the reaction increases markedly with the concentration of the aromatic compound, in this instance naphthalene, and it is therefore desirable to keep the volume of solvent as small as possible.The modified Stepanow procedure described later is satisfactory for the semi-micro determination of amounts of organic chlorine up to at least 10mg. The end-point of the reaction, which is indicated by the intense green colour of excess of sodium naphthalene, may become less apparent with the larger amounts of chlorine, as the reacting solution often becomes deeply coloured. Gravimetric determination of chlorine as silver chloride was preceded by part neutralisation of the reaction solution, evaporation to dryness in a platinum basin and gentle ignition to remove all traces of organic matter. The part removal of organic matter by means of an ether separation gave very low results in the analysis of a chlorinated nitro compound, owing to incomplete precipitation of silver chloride.Application of the proposed method to pesticide residues was intended to attain a sensitivity of 10 pg or less of chlorine, and a volumetric finish by the mercuric oxycyanide methodgJO was adopted, as this is simple and, on the semi-micro scale, sufficiently sensitive. To achieve a reasonably sharp end-point in the titration of chloride at high dilutions, it is necessary to remove both the excess of sodium ions, by means of a cation-exchange resin, and, by careful ignition, all traces of organic matter. The extraction with light petroleum of chlorinated hydrocarbon pesticides, added in known amounts to plant material, gave low yields and was abandoned in favour of extraction with a mixture of equal volumes of light petroleum and acetone.Benzene, which is often recommended, has the disadvantage of not being readily removed in a Danish - Kuderna evaporator, The excellent penetrating power for plant tissues possessed by the solvent mixture gives rise to extracts containing considerable amounts of plant extractives, and some degree of chromatographic separation is generally desirable. Activated alumina of suitably controlled water content can be used to effect a part separation of plant colouring336 SERGEANT : THE DETERMINATION OF CHLORINATED HYDROCARBON [Vol. 83 matter without retention, after elution with light petroleum, of any of a number of chlorinated hydrocarbon pesticides tested, with the exception of Kelthane, which is rather strongly adsorbed.The method is not intended for the analysis of materials that have an appreciable fat content. Although it does not appear to be practicable to separate a wide range of the chlorinated pesticides from large amounts of fat by a single method, techniques of limited application have been described,ll~12~13~14 and a combustion method of analysis has recently been recommended in this connection.lt' APPARATUS- Machine for swb-dividing the sample material-A domestic mincing machine of large capacity. Extraction vessels-Flat-bottomed 500-ml or larger flasks with standard ground-glass stoppers held in position by spring clamps. A wide socket, such as a B34, is desirable to facilitate the introduction of the sample material.Before insertion of the stopper, the ground-glass surfaces should be carefully cleaned and the stopper wetted with water to improve the seal. Shaking machine-This should rotate the extraction vessels with an end-over-end motion at a rate of about 1 revolution per second. REAGENTS- METHOD OF EXTRACTING THE PESTICIDE Light petroleum, redistilled-The fraction boiling over the range 60" to 80" C. Acetone, redistilled. PROCEDURE- Introduce 100 to 300g of the minced or shredded sample into the extraction vessel and, for each gram of material, add 1 ml of light petroleum, accurately measured, and 1 ml of acetone. For fibrous absorbent materials, such as grass, these volumes need to be doubled. Insert the stopper, rotate the flask for 1 hour and then pour the liquid through a fast- filtering paper into a separating funnel.Run off and reject any lower aqueous layer and to the remainder add an equal volume of distilled water. Wash the extract free from acetone and water-soluble chloride by shaking first with this and then with two further similar portions of distilled water. Run off the final water washings, add a few grams of anhydrous sodium sulphate and then shake to absorb the remaining water. Run off the liquid into a measuring cylinder and note the volume of light petroleum recovered, or remove by pipette an aliquot calculated to contain not more than 1.5 mg of chlorine. The pesticide is assumed to be distributed uniformly in the total original volume of light petroleum. APPARATUS- Chyomatographic tube-A filter tube, 30mm x 200mm, with a No.1 sintered-glass plate. Evaporator-A simple form of Danish - Kuderna evaporator.ls This consists of a 500-ml round-bottomed flask, the neck of which is fitted with a B24 ground-glass socket, and to the bottom of which is sealed a B14 ground-glass cone. In use, a splash head joined to a condenser for solvent recovery is fitted t o the socket, and a 25-ml round-bottomed flask with a B14 ground-glass socket is attached to the cone and held in place by a spring clamp. The ground-glass joints are moistened with liquid paraffin before assembly and the apparatus is heated by supporting the larger flask on an open-ended cylinder, e.g., made from a tin can, that stands on a steam-bath in such a way that the 25-ml flask and the lower part of the larger flask are surrounded by steam.Some device to prevent bumping of the liquid is necessary, and for this purpose a narrow strip of polytetrafluoroethylene sheeting was found to be effective. REAGENTS- Activated alumina-Heat dry aluminium hydroxide to 800" C and maintain at this temperature for not less than 1 hour; when cool, transfer the ignited material to a stoppered bottle, add 5 per cent. w/w of water and set aside overnight. Shake to mix the prepared alumina before use. METHOD OF TREATING THE EXTRACTJune, 19581 PESTICIDE RESIDUES IN PLANT MATERIAL 337 Sodium sulphate, anhydrous. Light petroleum, redistilled-The fraction boiling over the range 60" to 80" C. PROCEDURE- Transfer 10 g of activated alumina to the chromatographic tube and tap down to ensure that the column is uniformly packed. Cover the alumina with a 1-cm layer of sodium sulphate and arrange to collect the effluent from the column in the evaporator.Introduce the extract carefully, so as not to disturb the column, and, when it has passed completely through the column, add light petroleum rinsings; allow these to drain, and complete the operation by washing the column with 100 ml of light petroleum. Place the evaporator on a steam-bath after adding to the contents 2 drops of liquid paraffin and, when the volume of liquid has been reduced to a few millilitres, remove the evaporator and allow it to cool. Take off the splash head and rinse the apparatus with light petroleum. Detach the 25-ml flask and continue the evaporation on a water bath at 40" C with a stream of dry air until the remaining solvent has been removed.APPARATU s- MODIFIED STEPANOW METHOD FOR THE DETERMINATION OF ORGANIC CHLORINE Re$ux condenser-A small Liebig condenser fitted with a B14 ground-glass cone. Agla micrometer-syringe pipette-A right-angled glass jet is required. A suitable jet can be readily made from drawn-out glass tubing and attached to the syringe by a short piece of capillary plastic or rubber tubing, the flexibility of which protects the jet against accidental breakage or dislocation. Magnetic stirrer. Silica basins-Between determinations these should be kept filled with chromic acid cleaning mixture. REAGENTS- Naphthalene-Analytical-reagent grade. Tetrahydrofuran-Purify by placing 20 ml in a 200-ml Aask fitted with a reflux condenser by means of a ground-glass joint, and add 4 g of naphthalene and 4 g of clean sodium cut into small pieces.Boil the liquid until it becomes dark green, and then for a further 5 minutes. Continue boiling, and, by way of the condenser, add from a dropping funnel a further 100 ml of tetrahydrofuran and then 20 ml of xylene, the additions being made at such a rate that the colour of the solution is not discharged. Finally, replace the reflux condenser by an ordinary condenser with the addition of a fractionating column and distil off the purified tetrahydrofuran, which should be stored away from light. Sodium-To prepare the sodium in a convenient form for use, clean a number of pellets in turn by immersion in isopropyl alcohol and warming on a steam-bath until a vigorous reaction is proceeding.Quickly rinse the pellets with light petroleum and transfer them to a nickel crucible containing liquid paraffin. Heat the pellets until they melt, and then stir gently to assist coagulation into a large globule. Draw up the molten metal into clean dry glass tubes of about 5 mm internal diameter and 300 mm long, wetted internally with liquid paraffin. Allow the tubes to cool thoroughly, and extrude the sodium as required with the help of a glass rod. isoPropyl alcohol, diluted (1 + 1)-Add 1 volume of analytical-reagent grade isopropyl alcohol to 1 volume of distilled water. Screened methyl red indicator solution-Dissolve 0.08 g of water-soluble methyl red in 25 ml of distilled water, and 0.02 g of methylene blue in 25 ml of ethanol.Mix equal volumes of the two solutions for use. Sulphuric acid, 0.1 N, and approximately 0.04 N-A 0.1 N solution of sulphamic acid can alternatively be used; prepare an aqueous solution containing 0-971 g of the pure recrystallised acid per 100 ml. Potassium hydroxide solution-An approximately 0.04 N solution of the analytical-reagent grade material. Mercuric oxycyanide reagent solution-Dissolve 4 g of mercuric oxycyanide in 100 ml of distilled water, stirring on a steam-bath to assist dissolution. Cool and filter the solution, which is slightly alkaline and must be neutralised. For this purpose, carry out the titration procedure described later for chloride determination on 15 ml of distilled water, and thereby338 SERGEANT : THE DETERMINA.TIOX OF CHLORINATED HYDROCARBOS [Vol.83 determine the volume of 0.1 N acid required to neutralise 2 ml of the mercuric oxycyanide reagent under the conditions of the titration. Hence calculate the volume of acid to be added to neutralise the remaining bulk #of the reagent. Ion-exchange resin-Introduce 10 ml (of dry Zeo-Karb 225, (40 to 60-mesh) in the hydrogen form, into a glass tube of 16 mm internal d!iameter and 150 mm long, which tapers at the lower end to a jet and is fitted with a retaining plug of cotton-wool. Regenerate the resin before use and after each determination by running through the column, successively, 15ml of acetone, 15 ml of diethyl ether, 15 ml of 10 per cent. v/v nitric acid and, finally, not less than 40 ml of distilled water.If distilled water from the regular supply is found to contain significant amounts of chloride, it should be made alkaline and redistilled. The resin can be used many times. PROCEDURE- To the residue in the 25-ml flask add approximately 0.4 g of naphthalene, 1.5 to 2 ml of purified tetrahydrofuran and 0.2g of sodium cut into small pieces. Attach the reflux condenser, which should be quite dry, and boil the liquid over the flame of a microburner, e g . , the jet of a bunsen burner, until the appearance of an intense dark green colour indicates the end of the reaction, usually after from 2 to 5 minutes in the absence of appreciable amounts of moisture. Continue heating and add 2 ml of diluted isopropyl alcohol drop by drop. When all the residual sodium has dissolved, remove the flask and add distilled water to a total volume of about 15ml.Cool the flask in ice, then pour the contents into the resin column and collect the effluent in a silica basin. When all the liquid has entered the resin, rinse the flask with about 20 ml of distilled water and add the washings to the column. When the column has drained wash the resin with a further 20ml of distilled water. To the combined effluent and washings add phenolphthalein indicator and 0.04 N potassium hydroxide until just alkaline. Evaporate the liquid to dryness under an infra-red lamp, and then heat the basin, held in tongs over the flame of a bunsen burner, to a temperature just short of redness. It is essential to carbonise organic matter completely, but neither necessary nor desirable to attempt to burn off all the residual carbon.Cool the basin and add about 5 ml of distilled water and 1 drop of screened methyl red indicator solution. Stir the solution and make it just acid with 0.04 N sulphuric acid, then place the basin on a steam-bath for 1 to 2 minutes, with occasional stirring and further additions of acid if the solution turns alkaline. Transfer the solution to a 25-ml beaker, rinse the basin with distilled water and add the washings to the beaker until the total volume is approximately 16 ml. Cool the solution to room temperature and place the beaker on a magnetic stirrer. Add 4 drops of screened methyl red indicator solution and 0.04 N potassium hydroxide until the solution is just alkaline, and then standard 0.1 N acid from the micrometer-syringe pipette to give a persistent pinkish grey tint.Add 2ml of mercuric oxycyanide reagent solution and titrate with the acid to the same end-point colour. Determinatioiz of titration factor-Prepare an aqueous solution of potassium chloride containing 100 pg of chlorine per ml. Place aliquots up to 10 ml of this solution in a series of 25-nil beakers and dilute each to 15 ml with distilled water. Carry out the titration on each aliquot, and on 15 ml of distilled water, in the way described above. Plot graphically micrograms of chlorine taken against micrometer-reading differences less that for the distilled water titration. The relationship is not strictly stoicheiometric, but should be substantially linear over this range.Hence calculate the titration factor in micrograms of chlorine per millimetre on the micrometer scale. Determination of the blank value-Carry out a blank determination to ascertain the correction to be applied for chlorine present in the reagents. With reagents of reasonable purity and a laboratory atmosphere free from chloride fumes, this correction should not be more than 1Opg. R.ESULTS As sample material known to be uncontaminated with pesticide was not conveniently obtainable, and the concentration of any natural organic chlorine was uncertain, cabbage and tomato samples were minced and the minced material was well mixed and divided into halves. One half of each sample was treated as a control and a known weight of pesticideJune, 19581 PESTICIDE RESIDUES IN PLANT MATERIAL 339 was added to the other half in the form of a solution in light petroleum, which was allowed to evaporate.The results in Table I represent over-all recoveries of the various pesticides after deduc- tion of the equivalent of any organic chlorine found in the control samples, and are calculated on the theoretical chlorine contents of the pure crystalline substances and on the chlorine content of the technical materials as determined by analysis. TABLE I RECOVERY OF PESTICIDES ADDED TO PLANT MATERIAL Pesticides were added in the pure crystalline forms, except chlordane and toxaphene, which were added as the technical products Pesticide Pesticide found, Organic chlorine in control, p.p.m. p.p.m. With 1 mg of pesticide added to 100 g of minced cabbage, i.e., 10 +.p.m.of pesticide- y-Hexachlorocyclohexane . . 9.8, 9.4 0.05, 0.05 pp’-DDT . . . . . . . . 9.8, 9.8 0.05, Nil Methoxychlor . . . . . . 9.0, 8.8 0.35, Nil Aldrin . . .. . . . . 9.6, 9.6 0.05, Nil Dieldrin . . .. . . . . 9.2, 9.8 Nil, Nil Endrin . . . . .. . . 9.4, 9.9 Nil, 0.2 Chlorbenside . . . . . . 9.9, 9.4, 9.5 Nil, 0.15, 0.05 Chlordane . . .. . . . . 8.7, 9.2 Xil, 0.05 Toxaphene . . . . . . 9.6, 10.3 0.05, Kil With 1 mg of pesticide added to 100 g of minced tomato, i.e., 10 p.9.m. ofpesticide- y-Hexachlorocydohexane . . 10.0, 10.1 0.05, 0.05 pp’-DDT . . . . . . . . 10.7, 9.9 0.05, 0.05 Chlorbenside , . . . . . 10.0. 10.4, 10.2 0.i. 0.2, 0.05 Toxaphene .. .. . . 10.1, 10.3 0.25, 0.05 With 250 pg of pesticide added to 250 g of minced cabbage, i.e., 1 p.p.m. of pesticide- Aldrin . . .. * . . . 0.91, 0.91, 0.95 0.12, 0.08, 0.08 Dieldrin . . .. .. . . 0.97, 0.85, 0.90 0.08, 0.14, 0.14 Endrin . . .. .. . . 1.09, 0.83, 0.67 0.14, 0.02, Nil I thank the Government Chemist for permission to publish this paper, and also Mr. P. B. Thompson for technical assistance. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. REFERENCES Stepanow, A., Ber., 1906, 39, 4056. Analytical Methods Committee, “The Determination of Small Amounts of Total Organic Chlorine Phillips, W. F., and DeBenedictis, &I. E., J . Agric. Food Chem., 1954, 2 , 1226. Quist, W., and Holmstrom, T., Finska Kemistsanzfundets Medd., 1955, 64, 68. Rauscher, W. R., Ind. Eng. Chem., A?zal. Ed., 1937, 9, 296. Liggett, L. M., Anal. Chem., 1954, 26, 748. Benton, F. L., and Hamill, W. H., Ibid., 1948, 20, 269. Pecherer, B., Gambrill, C. M., and Wilcox, G. W., Ibid., 1950, 22, 311. Viebock, F., B e y . , 1932, 65, 496. Belcher, R., Macdonald, A. Ill. G., and Nutten, A. J., Mikrochim. Acta, 1954, 104. Davidow, B., J . Ass. Og. Agric. Chem., 1950, 33, 130. - , Ibid., 1950, 33, 886. Claborn. H. V.. and Beckman. H. F.. AIzal. Chem.. 1952. 24. 220. in Solvent Extracts of Vegetable Material,” Analyst, 1957, 82, 378. I . Hornstein, I., J. Agric. Food Chem., 1957, 5 , 446. ’ Hudy, J. A,, and Dunn, C. L., Ibid., 1957, 5 , 351. Gunther, F. A., and Blinn, R. C., “Analysis of Insecticides and Acaracides,” Interscience Publishers Inc.. New York and London, 1955, p. 231. Received November 18th, 1957