June, 19581 STEPAKEK AND CERNA A Method for the Detection of Traces of Acrylonitrile BY J. 114. STEPANEK AND V. M. CERNA (The Institute of Hygiene, 48 .%obdrova, Prague 12, Czechoslovakia) A chromatographic method is described for the detection and approximate determination of traces of acrylonitrile. The method is based on a reaction between acrylonitrile, thiourea and hydrochloric acid to yield 2-(2-cyano- ethy1)isothiuronium chloride, which is detected by paper chromatography with an ammoniacal silver nitrate spray reagent. The method is widely applicable, specific and highly sensitive; it makes possible the detection of acrylonitrile in amounts of the order of 1 pg. 345 IN the last 10 years, acrylonitrile has achieved many technological applications, e.g., as the raw material for artificial fibres and synthetic rubber, for various organic synthetic products and also as a fumigant insecticide for crops.The need for a suitable method for its detection has consequently extended into various industries. As acrylonitrile is a highly toxic sub- stance, a specific and sensitive method is required for its qualitative detection as well as for the determination of residual amounts of it in industrial raw materials and products, air, industrial waste waters, crops, food-stuffs and also in various physiological systems, in order to follow its metabolism, e.g., urine, blood and blood serum. Of the numerous chemical reactions of acrylonitrile only three have been suggested for its qualitative detection. The first of these makes use of its reaction with piperidine to form P-piperidinopropionitrile,l the picrate of which melts at 161" to 162" C ; this reaction can be used to detect concentrations of acrylonitrile as low as 1 per cent.The second method, a colorimetric test for amides and nitriles, is based on the reaction of acrylonitrile with hydroxyammonium chloride in propylene glycol. In the presence of potassium hydroxide, after being heated and subsequently cooled, a red to violet colour is formed with ferric chloride.2 Finally, a method was suggested for the detection of hydrogen cyanide or acrylonitrile in air or other gases, based on contact with a reagent consisting of o-tolidine and cupric sulphate in glycerol on silica gel. The presence of hydrogen cyanide or acrylo- nitrile causes a colour change to blue or blue-green.s The first of these reactions is not sufficiently sensitive and the others are not specific.Established quantitative chemical methods for the determination of acrylonitrile are based either on titration of the ammonia liberated when acrylonitrile is hydrolysed with concentrated alkali4 or sodium and ethanol5 or else on the oxidation of acrylonitrile with potassium permanganate637ys or a mixture of chromic and sulphuric acids.9 None of these methods is sufficiently specific for the reliable detection of residual traces of acrylonitrile. Beesing, Tyler, Kurtz and HarrisonlO have suggested a method based on the cyanoethylation of dodecyl mercaptan in the presence of potassium hydroxide as catalyst. The reaction is carried out in isopropyl alcoholll and the excess of mercaptan is titrated in an acidified medium with potassium bromate and potassium iodide.If the solution is coloured, the silver salt of the mercaptan can be precipitated and then determined potentiometrically.12 The cyanoethylation reaction between acrylonitrile and dodecyl mercaptan has also been applied to the determination of concentrations of acrylonitrile in air up to 150 mg per cubic metre.13 The ability of acrylonitrile to cyanoethylate isopropyl alcohol and the mercaptan simultaneously in an alkaline medium has been used in a modified method for the simultaneous determination of acrylonitrile and 1-cyano-1 : 3-butadiene.14 A similar quantitative method is based on the reaction between acrylonitrile and sodium or potassium glycocholate.15 For the determination of acrylonitrile, several quantitative methods have recently been suggested.One of these is based on a general reaction of ap-unsaturated compounds, viz., the addition of sodium hydrogen sulphite to acrylonitrile to form a substituted sodium sulphonate; the optimum pH for this reaction is attained by the addition of a measured excess of N sulphuric acid.16 Another method involves the reaction of acrylonitrile (and of @-unsaturated com- pounds in general) with morpholine in the presence of acetic acid to form a tertiary amine, which can be titrated with methanolic hydrochloric acid after the excess of morpholine has been converted to the neutral amide with acetic anhydride.17 Other methods are based on346 STEPANEK AND CERNA: A, METHOD FOR THE DETECTION OF [Vol.83 the addition of sodium sulphite to acrylonitrile and subsequent titration of the sodium hydroxide liberated1*; and also on the addition of sodium hydrogen sulphite, the excess of which is titrated with an alkali solution.1g Physical methods can also be used for the detection and determination of acrylonitrile; spectrophotometric,20 infra-red absorption21@9 and polarographic pr0cedures2~ to 29 have been reported. PRINCIPLE OF THE METHOD It has been reported in a review of the chemical properties of acrylonitrilell that it reacts with thiourea, in the presence oi a basic catalyst, to form 2-(2-cyanoethyl)~seudo- t h i o ~ r e a , ~ ~ j ~ l according to the equation-- NH, OH- I H,N*CS.NH, + CH2 = CH CN --+ NH = C-S-CH,*CH,.CK.According to another report, however, thiourea and acrylonitrile do not react at 100" C in the presence of alkali.s1p32 In the course of our study of acrylonitrile,33 we found that a similar cyanoethylation reaction occurs between acrylonitrile and thiourea in the presence of halogen acids in an aqueous or alcoholic medium. The halogen acid takes part in the reaction, the main product of which is the corresponding isothiuroniiim salt, according to the equation- p N H ) + GN*CH = CH, + H2N*CS*NH2 + HX ----+ CN.CH2.CH2-S-C H / x-, I \NH, J where X represents C1, Br or I. The addition product of acrylonitrile and hydrochloric acid, P-chloropropionitrile, reacts in the same way and forms 2-(2-cyanoethyl)isothiuronium chloride directly with thiourea.s4 It is hoped that details of the probable intermediate formation of P-chloropropionitrile from acrylonitrile and aqueous or alcoholic hydrochloric acid will be dealt with in a separate paper.The homologous methacrylonitrile, when allowed to react with thiourea under the same conditions, does not form a corresponding isothiuronium salt. In the presence of concentrated hydrochloric acid the reaction occurs with quantitative consumption of acrylonitrile. The resulting 2-(2-cyanoethyl)isothiuronium chloride can be detected by paper chromatography with a,n ammoniacal silver nitrate spray reagent. This method is both specific and simple; it permits the detection of amounts of acrylonitrile as small as 1 pg. The corresponding bromide (m.p. 177" C) and iodide (m.p. 131" to 132" C) can also be detected by paper chromatography, as they have different RF values, but the procedure is less sensitive and considerable complications can arise, particularly with the iodide.EXPERIMENTAL To verify the conditions of the reaction between acrylonitrile, thiourea and hydrochloric acid, the optimum ratio of reactants was first studied. It was found that the reaction occurred almost stoicheiometrically when an excess of hydrochloric acid was present. In order to increase the sensitivity of detection of minute amounts of acrylonitrile, e.g., from 1 to 5 pg, it was necessary to increase proportionately the amounts of thiourea and hydro- chloric acid in the reaction solution, up to 35 moles and 50 moles, respectively, per mole of acrylonitrile. Either water or a polar organic solvent can be used as the reaction medium.Ethanol, Iz-propyl alcohol and isopropyl alcohol proved to be the most suitable solvents; isobutyl alcohol, tert.-butyl alcohol and acetone caused the formation of interfering by-pro- ducts, especially at the lower concentration of acrylonitrile. The reaction takes place at ordinary temperatures, but can be accelerated by heating, e.g., under reflux. When different amounts of acrylonitrile are allowed to react with thiourea in the absence of hydrochloric acid, or in an alkaline medium, the isothiuronium salt cannot be detected; this is in accordance with the findings of Hurd and Ger~hbein.~~ It was also found that, after the completion of the reaction, the solution can be diluted with water or an organic polar solvent or, alternatively, concentrated to a certain extent by evaporation, without any apparent effect on the detection of acrylonitrile. Dilution canJune, 19581 TRACES OF ACRYLONITRILE 347 be practically unlimited, but evaporation, which can only be carried out with alcoholic solu- tions, must not be continued after an approximate concentration of 0.1 mg of acrylonitrile per 2 ml of solution has been reached.The chromatographic spots in this instance, however, are less marked, being about half their normal intensity. It is preferable, therefore, to con- centrate the test solution by distilling the acrylonitrile from the original sample, either in a current of pre-heated air or azeotropically with isopropyl alcohol. By using this procedure, it is possible to determine minute amounts of acrylonitrile, when working with specified volumes, down to a minimum of 1 pg.The colour intensity of the isothiuronium spot produced by 1 pg of acrylonitrile corresponds to that produced by 0-3 to 0.5 pg of standard isothiuronium chloride solution. Impurities in commercial acrylonitrile do not affect its detection or determination. METHOD REAGENTS- All reagents should be of recognised analytical grade. Hydrochloric acid, concentrated. Ammonia solution, 10 per cent. aqueous. Sodium thiosulphate solution, 2 per cent.-A 2 per cent. aqueous solution of sodium Thiourea. Ethanol or isopropyl alcohol. n-Butyl alcohol-Saturate with distilled water before use. Ammoniacal silver nitrate solution (Tollens’s reagent)-Prepare by treating a 0.1 S aqueous solution of silver nitrate with an equimolecular amount of 5 per cent.aqueous sodium hydroxide. Add 5 N aqueous ammonia to the precipitated silver oxide, with constant agitation, until the solid just dissolves, and then make the solution slightly ammoniacal by the further addition of aqueous ammonia. This reagent must be freshly prepared before use. 2-(2-Cyanoethyl)isothiuronium chloride, ethanolic standard solution-Dissolve 2 g of the isothiuronium salt, prepared by the procedure described on p. 348, in 1 litre of ethanol. thiosulphate, Na,S20,.5H20. PREPARATION O F THE SAMPLE FOR CHROMATOGRAPHY- By absorption, distillation or extraction from the system containing acrylonitrile, prepare in ethanol or isopropyl alcohol a solution containing 5 to 10 mg of acrylonitrile per 10 ml.Place 10 ml of this solution in a 50-ml flask, add 0.5 g of thiourea and then 0.75 ml of con- centrated hydrochloric acid. Boil the reaction mixture under reflux on a water bath for 30 minutes. When the acrylonitrile is present in the form of an aqueous solution, follow the same procedure, but heat under reflux at a correspondingly higher temperature. When 10 ml of the alcoholic solution contain less than 5 mg of acrylonitrile, follow the same procedure, but make correspondingly smaller additions of thiourea and hydrochloric acid, eg., 5 and 10 times diminished at acrylonitrile contents of 2 and 1 mg per 10 ml of solution, respectively. After 15 to 30 minutes’ boiling under reflux, evaporate the reaction solution on a water bath to a specified volume, not less than 2 ml, and use this for the chromatographic procedure.If the original solution contains more than 10 mg of acrylonitrile per 10m1, dilute it with the solvent until the isothiuronium spot that corresponds to the minimum amount of 1 pg of acrylonitrile is achieved by paper chromatography. In this instance, it is not necessary to alter the additions of thiourea and hydrochloric acid. It is always preferable to estimate the amount of acrylonitrile present in the sample by a pre- liminary experiment, in order to establish the precise conditions for the final procedure. PROCEDURE FOR DEVELOPING THE CHROMATOGRAM- On a strip of Whatman No. 4 filter-paper, 150mm x 300mm, place, by means of a micropipette, at least 0.001 ml of the reaction solution along a line 8 cm from the upper edge of the strip.Allow the solvent to evaporate and then repeat the procedure several times, making any number of additions up to ten and allowing the solvent to evaporate after each addition. Chromatograms are thus prepared from 0.001 to 0.01 ml of the reaction solution, and these can be used to achieve an approximate evaluation of the amount of acrylonitrile present. When cool, use this solution for the paper-chromatographic procedure.348 STEPANEK AND CERNA: .4 METHOD FOR THE DETECTION OF [Vol. 83 To compare the position and character of the isothiuronium spots detected with that of a standard solution, apply simultaneously 0.001 ml of the latter by means of a micropipette. Suspend the strip of filter-paper in i2 chromatographic tank for 2 hours to attain equili- brium with the atmosphere of n-butyl alcohol and water vapours.Dip the upper end of the paper strip in n-butyl alcohol that has been saturated with water and allow the solvent front to travel 25 to 28cm down the paper over a period of about 4 hours. Remove the strip from the tank, mark the position of the solvent front and then allow the paper to dry at room temperature until all the solvent has evaporated (12 to 24 hours). PROCEDURE FOX COLOUR DEVELOPMENT- Spray the dry strip with Tollens’s ammoniacal silver nitrate reagent. The dark brown spot of thiourea, which is the most substantial one on the chromatogram, R, 0.43, appears immediately, owing to the presence of a considerable excess of this substance.As spraying proceeds, a typical yellow spot of 2-(2-c:yanoethyl)isothiuronium chloride appears, RF about 0.25, depending on the concentration of acrylonitrile present. This spot soon darkens until it is dark grey, with a yellow centre at higher concentrations of acrylonitrile. The develop- ment of the spot, particularly at minute concentrations of acrylonitrile, can be accelerated by heating for 5 minutes at 100” C and subsequently washing the chromatogram with a 10 per cent. aqueous solution of ammoni,s and then with distilled water. A faint brownish spot, R, 0.32, which gradually darkens, also appears; this spot corresponds to the reaction of thiourea and hydrochloric acid with the alcohol and appears even in the absence of acrylonitrile. When the standard isothiuroniuin chloride solution is used, only the spot RF 0.25 appears.The spots of isothiuronium chloride have a tendency to “tail” somewhat as the concentra- tion of acrylonitrile increases from 1 pg up to approximately 10 pg per 10 ml, but the position of the solvent front remains constant. The position, character and colour of these spots correspond closely to those obtained from standard solutions at the respective concentrations. When the colours of the spots have developed to maximum intensity, wash the chromato- gram thoroughly with distilled water and allow it to dry at room temperature. The removal of the excess of ammoniacal silver nitrate solution and the fixation of the chromatogram are finally accomplished by washing with a 2 per cent.solution of sodium thiosulphate and water. The presence of any proportion of methacrylonitrile does not interfere with the detection of acrylonitrile. When the chromatograin is treated with ammonia solution after it has been sprayed with ammoniacal silver nitrate solution, methacrylonitrile slowly forms a faint yellowish to brownish spot just above that of thiourea, thus producing an apparent “tailing” of the thiourea spot. To determine approximately the acrylonitrile content of the solution, record the iso- thiuronium spot at which a faint durable grey colour just appears. This spot corresponds to 1 pg of acrylonitrile. From the concentration of the solution examined and the amount of it necessary to produce that spot, calculate the approximate amount of acrylonitrile present in 1 kg of the original sample.REAGENTS- PREPARATION OF 2-(2-CYANOETHYL)iSOTHIURONIUM CHLORIDE Acrylonitrile-Technical grade. Thiowea-Analytical-reagent grade. Hydrochloric acid, concentrated-Analy tical-reagent grade. isoPropyl alcohol-Technical grade. EthanodTechnical grade. PROCEDURE- Dissolve 25 g of acrylonitrile in 200 ml of isopropyl alcohol contained in a 500-ml flask. Add 65 ml of concentrated hydrochloric acid and 40 g of thiourea. Boil the reaction mix- ture under reflux on a water bath for 30 minutes. After it has been cooled and shaken thoroughly, the solution deposits fine needles of 2-(2-cyanoethyl)isothiuronium chloride, which are separated and dried in air. By part evaporation and crystallisation of the mother-liquor, further portions of isothiuronium chloride can be obtained.Purify the product by recrystallisation from ethanol until it melts between 165-5” and 166’ C. This method of preparation gives an approximately 60 per cent. yield of the pure product.June, 19581 TRACES OF ACRYLONITRILE 349 Purity test-The purity of the 2-(2-~yanoethyl)isothiuroniurn chloride can be tested by Only subjecting the material to the above-described paper-chromatographic procedure. one spot, R, 0.25, should be produced. APPLICATION OF THE METHOD The proposed method can be applied for the detection and approximate determination of trace amounts of acrylonitrile in a wide range of materials, such as industrial products, food-stuffs, crops, air and waste waters, and in physiological systems such as urine, blood and blood serum.It is hoped that the respective analytical procedures, together with any interfering effects that arise from the character of the material being examined-crops, urine, blood and so on-will be dealt with in separate publications. The proposed method may find valuable application in food technology by specifying the amount of material under examination that should not produce a coloured isothiuronium chloride spot by the method, and so be in accordance with the official tolerances for acrylonitrile in the material. We thank M.U. Dr. I<. Symon, Cniversity Docent, Director of the Institute of Hygiene, Prague, for his kind permission to publish this paper. REFERENCES Brockway, C. E., Anal. Chem., 1949, 21, 1207. Soloway, S., and Lipschitz, A., Ibid., 1952, 24, 898.McConnaughey, P. W., U.S. Patent No. 2,728,639, 1955; Chew Abstr., 1956, 50, 5470. Peterson, G. W., and Radke, H. H., I n d . Eng. Chem., Anal. E d . , 1944, 16, 63. David, H. S., and Wiedeman, 0. F., I n d . Eng. Chem., 1945, 37, 482. “Deutsche Gesellschaft f u r Schadlingsbekanzpfufung,” Degesch, Frankfurt, 1943, “Ventox,” p. 7. Kroger, E. P., and Schuler, W. A., Dtsch. LebensmittRdsch., 1950, 46, 129. Dal Nogare, S., Perkins, L. R., and Hale, A. H., Anal. Chem., 1952, 24, 512. Kobayashi, Y., YzZki GBsei Kagaku KyBkai Shi, 1956, 14, 673; Chem. Abstr., 1957, 51, 7240. Beesing, D. W., Tyler, W. P., Kurtz, D. M., and Harrison, S. A., Anal. Chem., 1949, 21, 1073. American Cyanamid Co., New York, Cyawamid’s Nitrogen Chemicals Digest, 1951, “The Chemistry Janz, G.J., and Duncan, N. E., Anal. Chem., 1953,25, 1410. Haslam, J., and Newlands, G., Analyst, 1955, 80, 50. de MaldB, M., Ann. Chim. Roma, 1952, 42, 437. Terent’ Bv. A. P., Butskus, P. F., and YaShunsky, V. G., Zhur. Anal. Khim., 1954, 9, 162. Critchfield, F. E., and Johnson, J. B., Anal. Chem., 1956, 28, 73. Critchfield, F. E., Funk, G. L., and Johnson, J. B., Ibid., 1956, 28, 76 Terent’ Bv, A. P., and Obtemperanskaya, S. I., Zhur. Anal. Khim., 1956, 11, 638. Kostin, L. D., and Vidanova, V. A,, U.S.S.R. Patent No. 104,643, 1957; Chem. Abstr., 1957,51, 7246. Brieger, H., Rieders, F., and Hodes, W. A., Arch. I n d . Hyg., 1952, 6, 128. Barnes, R. B., Liddel, M., and Williams, V. Z., Ind. Eng. Chem., Anal. Ed., 1943, 15, 83. Bentley, F. F., and Rappaport, G., Anal. Chem., 1954, 26, 1980. DuprB, E. F., Armstrong, A. C., Klein, E., and O’Connor, R. T., Ibid., 1955, 27, 1878. Bird, W. L., and Hale, C. H., Ibid., 1952, 24, 586. Matyska, B., and Kbssler, I., Congress of Practical Polarography, Bratislava, 1952: Anal. Chem., Ryabov, A. V., and Panova, G. D., Dokl. Akad. Nauk SSSR, 1954, 99, 547. Strause, S. F., and Dyer, E., Anal. Chem., 1955, 27, 1906. Platonova, M. N., Zhur. Anal. Khim., 1956, 11, 310 and 317. Danes, G. W., and Hamner, W. F., Anal. Chem., 1957, 29, 1035. Harman, M. W., U.S. Patent No. 2,413,917, 1947. Bruson, H. A,, “Organic Reactions,” John Wiley & Sons Inc., New York, 1948, Volume V, p. 96. Hurd, C. D., and Gershbein, L. L., J . Amer. Chem. Soc., 1947, 69, 2328. StBpknek, J. M., and CernP, V. M., CSR Patent Applicn. No. PV-3174, 1957. Miller, E., Sprague, J . M., Kissinger, L. W., and McBurney, L. F., J . Amer. Chem. Soc., 1940, Received October 7th, 1057 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. of Acrylonitrile.” 1952, 24, 916. 62, 2099.