544 Analyst, September, 1974, Vol. 99, $9. 544-546 Determination of Cyanide Ion in Cyano-complexes BY S. K. TOBIA, Y. A. GAWARGIOUS AND M. F. EL-SHAHAT (Chemistry Department, A in Shams University, and Micro-analytical Chemistry Laboratory, National Research Centre, Dokki, Cairo, Egypt) A method is described for the determination of the CN- ion in a variety of cyano-complexes. The method is based upon mixing and grinding the sample with flowers of sulphur and potassium chloride in the ratio of 1 : 2 : 2; the addition of potassium carbonate is also essential with cyano-acid complexes. After mixing, a micro-amount is fused in a test-tube a t about 300 "C for 2 minutes and the tube is then broken in a few millilitres of water. In this way the CN- ion is converted quantitatively into CNS-, which is determine? colorimetrically.The average recovery is 98-54 per cent. THE determination of the cyanide ion in cyano-complexes and, in particular, in relatively stable compounds, constitutes one of the important problems in analytical chemistry because of a lack of methods for the analysis of such complexes. This is the reason why cyanide in cyano- complexes is usually determined with the help of micro-scale elemental analysis. However, such methods, in addition to requiring the use of sophisticated apparatus, usually yield low results for carbon and nitrogen contents, probably owing to the formation of very stable carbides and nitrides, respectively, with the central metal ion of the complex. Other possible methods are based upon decompositionl9 of the complex, with the liberation of hydrogen cyanide gas.The latter can either be determined as such, using a gas-chromatographic method3** (although this technique has been attempted only with pure hydrogen cyanide) or, after conversion into the cyanide of an alkali metal, by a variety of techniques, e.g., titri- metry, K s * polarography7 9 or colorimetry. lo In the present work a method has been developed for the determination of cyanide ion in different cyano-complexes. The method is based upon gradual heating, to 300 "C, of the sample after admixture with potassium chloride and flowers of sulphur. After breaking the ignition tube in a small volume of water and the colorimetric determination of the thiocyanate formed, the cyanide content is calculated.EXPERIMENTAL REAGENTS- All reagents used were of analytical-reagent grade unless otherwise specified. Potassium thiocyanate solzttion, 2227 9.p.m. of CNS- (=lo00 p.6.m. of CN-)-Dissolve about 3.73 g of dried potassium thiocyanate in double-distilled water, dilute the solution to 1 litre and standardise it against standard silver nitrate solution. The concentration of the solution is such that 1 ml contains 1 mg of CN-. Solutions containing the required microgram amounts were prepared by suitable dilution. Iron(1lI) chloride, 10 per cent. solution-Dissolve 10 g of the hydrated salt, FeCl,.GH,O, in 100 ml of nitric acid (1 + 1). Potassium chloride. Potassium carbonate. Flowers of sulphur. PROCEDURE- Mix the sample under test thoroughly in an agate mortar with flowers of sulphur and potassium chloride in the ratio 1 : 2 : 2.The complex cyanides analysed were potassium hexac y ano f errat e (I I), potassium hexac yano f errat e (I I I), sodium t e t rac y anonickelat e (I I) , potassium pentacyanonitrosylchromate( 111) and zinc cyanide. Weigh accurately on a micro- balance a few milligrams (about 12-6 to 34-3 mg) of the sample mixture, i.e., about 2.5 to 7.5 Ing of cyanide complex, and transfer them into a micro-scale test-tube. Heat the tube gradually in a micro-flame until the temperature reaches about 300 "C and continue heating it at this temperature for about 2 minutes. Break the tube in a small beaker containing about @ SAC and the authors.TOBIA, GAWARGIOUS AND EL-SHRHAT 545 10 ml of double-distilled water, bring the solution to the boil, cool and filter it, collect the filtrate in a measuring flask and dilute it to 25 ml.Transfer 5 ml of the filtrate into a 100-ml calibrated flask, dilute it with double-distilled water, add 5 ml of the iron(II1) chloride solution and dilute the mixture to the mark. Mix well and measure the absorbance of the solution against a blank by use of a Hilger Biochemical Colorimeter, a l-cm cell and a greenish blue filter (470 nm). Calculate the cyanide content of the sample with the aid of a c,alibration graph. The same procedure was applied to the determination of the cyanide ion in hexacyano- ferric(I1) acid, hexacyanoferric(II1) acid and potassium hexacyanocobaltate(III), with which complexes potassium carbonate (one sixth of the amount of potassium chloride) should be well admixed with the flowers of sulphur and potassium chloride.RESULTS AND DISCUSSION The cyanide-ion content of solutions of simple cyanides can be determined colorimetrically after their conversion into thiocyanate with sodium tetrathionatell or yellow ammonium sulphide.12 However, simple cyanides are usually determined by a Konig-type reaction.13914 No attempt to determine cyanide ion in complexed cyanides has so far been reported. Deter- mination of these complexes with S,062- or S2- ions was found to be unsatisfactory. The drawbacks of these methods are a low recovery (about 70 per cent.) and also that they cannot be used for insoluble compounds. The low results obtained were attributed to incomplete conversion of cyanide into thiocyanate; it was therefore necessary to modify the method so that almost complete conversion into thiocyanate occurs.When micro-scale amounts (about 5 to 10 mg) of the cyano-complexes were heated in small test-tubes with various amounts of flowers of sulphur, this stage being followed by colorimetric measurement of the thiocyanate produced, consistent, but low, cyanide recoveries of about 66 per cent. were obtained. This result indicates that in all instances the percentage conversion into thiocyanate is only two thirds of that of the original, irrespective of the amount of sulphur added. The loss was found to be caused by an evolution of gas, which was identi- fied as being a mixture of hydrogen cyanide and dicyanogen. These gases were detected, in turn, by applying the copper - benzidine15 and 8-hydroxyquinoline16 tests.The evolution of both gases can be attributed to the high temperature used. In attempting to overcome this difficulty, it was thought that the gas loss would be prevented if the reaction were allowed to take place at a lower temperature. This temperature reduction has been achieved by the addition of a salt to act as a flux. The use of potassium chloride in an amount twice the expected amount of cyanide ion gave rise to satisfactory cyanide recoveries. In the presence of this salt, the melting-point dropped from 242 to 165 "C when potassium hexacyanoferrate(II1) was under test. Under these conditions no hydrogen cyanide or dicyanogen gas could be detected on heating. Quantitative recoveries (Table I) were obtained with a potassium chloride to cyanide ion ratio of 2 : 1 for the following cyano-complexes : potassium hexacyanoferrate(II1) and hexacyano- ferrate(I1) ; potassium pentacyanonitrosylchromate(II1) ; and sodium tetracyanonickelate(I1).In addition, results were obtained for simple zinc cyanide. Amounts of potassium chloride greater than that indicated by the specified ratio also proved efficient. The maximum absolute error is h4.9 per cent. and the over-all average cyanide recovery is ,t98.58 per cent. TABLE I POTASSIUM CHLORIDE (RATIO 1 : 2 : 2) DETERMINATION OF CYANIDE IN CYANO-COMPLEXES BY USING FLOWERS OF SULPHUR AND Amount of cyanidelmg (- p1 Recovery, Complex * Expected Found per cent. 1.378 1.362 98.7 1 1.490 1.467 98.7 1 1.802 1.780 98.58 I< 3[Cr (CX) ,NO] Na JNi (CN) 4] 1.336 1.296 97.14 * The results given are an average of five experiments for each complex.K*[Fe(CN) 61 K,[Fe(CN),I Zn (W?. 1.801 1.799 99.94546 TOBIA, GAWARGIOUS AND EL-SHAHAT However, the addition of potassium chloride did not prove successful with hesacyano- ferric(I1) and hexacyanoferric(II1) acids, nor with potassium hexacyanocobaltate(III), which may still be attributed to the loss of some CN- ion as hydrogen cyanide gas, tlie results obtained being low by about 25 per cent. The addition of a substance that would prevent the loss of hydrogen cyanide, such as the carbonate of an alkali metal, was next tried. For a potassium carbonate to potassium chloride ratio of 1 : 5, satisfactory results (Table 11) were obtained.This procedure was iound to be necessary for complex cyanides that lose hydrogen cyanide on heating, such as the cyano-acids or their ammonium salts, and when dicyanogen is evolved, especially if the central atom of the complex has oxidising properties, e.g., iron(II1) or cobalt(II1). In these instances, very gentle and gradual heating is required until a temperature of about 300 “C is reached; the temperature should then be kept constant so as to prevent the possible formation of dicyanogen gas. This modification afforded a maximum absolute error of h3.2 per cent. and a total average recovery of h98.46 per cent. TABLE I1 DETERMINATION OF CYANIDE I N CYANO-COMPLEXES BY USING FLOWEKS OF SULPHUR, POTASSIUM CHLORIDE -4ND POTASSIUM CARBONATE (RATIO 3 : 6 : 6 : 1) Cyanide content/mg F-----~-, Recovery, Complex’ Expected Found per cent. K,[Co(CN) sj 1.356 1.338 98-95 H,[Fe(CN) a1 1-369 1.350 98.61 H,[Fe(CN) 1.751 1.713 97.82 * The results given are an average of four experiments for each complex. Mixing micro-scale amounts of the cyano-complex, sulphur and potassium chloride (in addition to potassium carbonate in certain instances) in a test-tube, followed by fusion a t 300 “C, gave rise to unreliable results owing to inhomogeneity.In order to obtain a homogeneous mixture it was found that preliminary grinding was necessary. The grinding of such micro-amounts in an agate mortar introduced a serious source of error because of unavoidable sample losses. Therefore, the mixing and grinding of macro-amounts was adopted in order to obtain a representative mixture.A micro-amount of this mixture was then weighed for analysis. Following this grinding technique satisfactory results were obtained in all instances. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. REFERENCES M-illiams, H. E., “Cyanogen Compounds, Their Chemistry, Detection and Estimation,” Second Kruse, J . M., and Mellon, N. G., Sewage l n d . Wclstes Engng, 1951, 23, 1402; Chem. Absly., 1052, 46, Woolmington, K. G., J . Appl. Chem.., Lond., 1961, 11, 114. Schneider, C. R., and Freund, H., Avialyf. Chem., 1962, 34, 69. Laszlo, L., Magy. KLnz. FoZy., 1968, 74, 61; Chem. Abstr., 1968, 68, 84042e. Nomura, T., Takeuchi, K., and Komatsu, S., Nippon Kagaku Zasski, 1968, 89, 291; Cliem. Abstr., Jura, W. H., Analyt. Chem., 1954, 26, 1121. Hetman, J., J . A$#. Chem., Lond., 1960, 10, 16. Johnson, M. 0.. J . Amer. Chem. SOC., 1916, 38, 1230. Guilbault, G. G., and McQueen, R. J., Analytica Chim. Acta, 1968, 40, 251. Kolthoff, I. M., 2. analyt. Chew., 1924, 63, 188. Rozina, A. M., Dankova, N. M., Amitina, N. I., and Rutshtein, E. M., h’ohs Khim., 1957, 5, 45; Chem. Abstr., 1957, 51, 13359e. Konig, W., J . prakt. Chem., 1904, 69, 105. -, 2. angew. Chem., 1905, 115. Moir, J., Chem. News, Lond., 1910, 102, 17. Feigl, F., and Hainberger, L., Analyst, 1955, 80, 807. Edition, Edward Arnold (Publishers) Ltd., London, 1948, p. 168. 36858. 1968, 68, 119227t. Received July 301h, 1973 Accepted Februwy 21st, 1974