24 WILLMOTT AND RAYMOND: THE DETERMINATION OF SMALL [Vol. 75 The Determination of Small Quantities of Copper in Lead and Lead Alloys BY P. L. WILLMOTT AND F. J. RAYMOND SYNoPsIs-The method is based on the formation of the copper complex with sodium diethyldithiocarbamate in strongly ammoniacal solution, its extraction therefrom by ether and photometric reading of the ether extract. A 6-g. sample is dissolved in nitric acid and evaporated and, after addition of sulphuric acid, the major part of the tin, antimony and lead is filtered off. Interference by nickel and cobalt are avoided by addition of dimethylglyoxime to the neutralised filtrate and filtration. For complex formation the filtrate is treated with ammonia, ammonium citrate and the reagent. The ether extract is read in a Spekker absorptiometer.Bismuth interferes, producing a similar but much feebler colour. A correction for it is made by carrying out a similar determination in which the formation of the copper complex is prevented by prior addition of cyanide. Copper contents up to 0.01 per cent. are determinable to within 0.00005 per cent. By suitable choice of sample weights and dilutions the method may be adapted to copper contents up to 10 per cent. DURING the past year there has been an increasing demand for chemical lead and also lead pipe made to B.S. 1085. In order to meet the heavy demand, it was essential that the works refining processes were not delayed, and that results of chemical tests on the partially refined metal were reported as quickly as possible.For determining small quantities of copper there were two methods available. The method given by the British Standards Institution is a very lengthy process, and not suitable for routine works control. An alternative method is to electrolyse the copper from solution, so that it may be redissolved in a small volume and determined colorimetrically with ammonia.Jan., 19501 QUANTITIES OF COPPER IN LEAD AND LEAD ALLOYS 26 This method is also very lengthy, since it requires taking large quantities of the sample because of the rather insensitive colour reaction. Among the most sensitive reagents for copper is sodium diethyldithiocarbamate. A brown precipitate of the very slightly soluble copper carbamate is formed when an aqueous solution of sodium diethyldithiocarbamate is added to a very slightly acid, neutral or alkaline (ammoniacal) solution of a cupric salt.In very dilute solutions a colloidal suspension is formed which is suitable for colour comparison. Gum arabic, gum acacia, gum tragacanth or gelatin may be used as protective colloid to prevent the coagulation of the precipitate. The copper carbamate complex is soluble in a number of organic solvents, such as amyl alcohol, amyl acetate, bromo-benzene, chloroform or ethyl ether. The resulting brown solution may be used for the colorimetric determination of copper, either by comparison with standards similarly prepared or by measuring its light absorption. With many of the heavy metals sodium diethyldithiocarbamate forms precipitates, most of which are soluble to a greater or less extent in the organic soltents named.Zinc, cadmium, mercuric mercury, silver, lead and tin salts give white precipitates, whilst ferric iron yields a brown-black precipitate in neutral or acid solutions. However, in alkaline citrate solutions there is no reaction if the pH is maintained above 9. Bismuth yields a similar colour to that given by copper, although the colour produced is only about one-thirtieth to one-fiftieth as intense. Of the metals that interfere with the copper determination, nickel, cobalt and bismuth are the worst offenders. Zinc, cadmium, mercuric mercury, silver, lead and tin salts cause no interference after extraction because their solutions have 100 per cent. transmittancy. Iron is not likely to be found in large quantities in lead.The interference from nickeland cobalt is avoided by adding 1 ml. of 0-5 per cent. dimethyl glyoxime solution to thealkaline solution. The cobalt remains in the aqueous layer as an orange complex, which is not extracted with ether. Copper diethyl- dithiocarbamate is destroyed by dilute solutions of potassium cyanide, whilst the bismuth compound is not affected. Strong solutions of potassium cyanide, however, make the extrac- tion of the bismuth complex difficult, and may cause bismuth to revert to the aqueous phase if an ethereal solution of the complex is shaken with very concentrated potassium cyanide solution. Investigation by the authors has shown that less potassium cyanide is required to prevent the formation of the copper complex than is required to destroy the complex once it has been formed.Solutions of copper diethyldithiocarbamate in ether are very stable and undergo no change in intensity over a period of many months if kept in stoppered bottles to prevent evaporation of the ether. The procedure described is very accurate and detects as little as 1 pg. of copper. The total time taken in a determination is about 20 minutes, made up as follows: preparation of the sample, 1 rnin.; weighing the sample, 1 min.; solution of sample, 4 rnin.; precipitation and cooling, 4 min. ; filtering, 5 min. ; extraction and comparing colour, 5 min. In order to check the accuracy of the method, a solution of pure lead nitrate was prepared and different amounts of pure copper nitrate were added. The following table shows the results obtained.The nickel is precipitated and removed by centrifuging or filtering. The interference from bismuth is best overcome by a difference method. Copper added, Copper found, % on amount of lead present yo on amount of lead present 0.00005 0~00010 0.00015 0.00050 0.00 100 0.00350 0.00440 0.0 1000 0.03000 0.00005 0~00010 0-000 15 0-00050 0-00095 0-00355 0-00440 0-01000 0-03050 The synthetic samples contained lead nitrate equivalent to 5 g. of lead. A further series of tests was performed on synthetic samples containing different amounts of bismuth, and in every test the result found was within 0.00005 per cent. of the true figure when calculated on the amount of lead present.26 WILLMOTT AND RAYMOND: THE DETERMINATION OF SMALL METHOD SOLUTIONS REQUIRED- (1) Sulphuric acid, 50 per cent. v/v.[vol. 75 (2) Ammonium citrate solution. (3) Potassium cyanide solution in water, 0.5 g. per 100 ml. (4) Sodium diethyldithiocarbamate solution in water, 0.5 g. per 100 ml. Dissolve 250g. of citric acid in 250ml. of water. Cool and add 250ml. of aqueous ammonia, sp.gr. 04380. PROCEDURE- Weigh two separate 5-g. portions of the sample and place each in a 300-ml. tall-form beaker. Dissolve egch in 15 ml. of water and 7 ml. of nitric acid, sp.gr. 1.42. The sample will dissolve rapidly if in the form of thin rollings. When it has dissolved, boil down to small volume and then take up in the minimum quantity of water. This treatment will eliminate the bulk of the tin and antimony from solution. Add 10 ml. of 50 per cent.sulphuric acid and cool thoroughly. Filter off the lead sulphate through a fairly fine paper; a Whatman No. 530 is recommended. Wash well with cold water until free from acid. (If the presence of nickel or cobalt be suspected, add also 1 ml. of arnmoniacal 0-5 per cent. dimethylglyoxime solution. Allow to stand 5 minutes and then filter through a Whatman No. 530 paper. In absence of nickel and cobalt this treatment can be omitted.) To one of the two test solutions add a 10-ml. excess of aqueous ammonia (sp.gr. 0.880), 10 ml. of ammonium citrate solution (Reagent 2) and 10 ml. of sodium diethyldithiocarbamate solution (Reagent 4). Rinse out the beaker with 25 ml. of ether and add the rinsings to the solution in the funnel. Shake vigorously under running water for about 1 minute and allow the two layers to separate.(The lower, aqueous, layer should be colourless, and should remain so when a few ml. of the sodium diet hyldit hiocarbamat e solution are added.) Collect the coloured ethereal layer in a 100-ml. graduated measuring flask. Rinse out the funnel with ether saturated with reagents and add the rinsings to the main solution in the flask. Shake to mix the extract and determine the transmittancy as a Drum Reading 1, on the Spekker absorptiometer, with a water-to-ether setting of 1.00. For copper contents below 0.003 per cent. (calculated on sample) use an Ilford violet filter No. 601 and the large 4-cm. cells. For copper contents between 0.002 and 0.010 per cent. use the I-cm. cells and the same filter. The 1-cm. cells with an Ilford blue-green filter No.603 should be used for copper contents between 0.01 and 0.03 per cent. The Drum Reading 1 is a measure of copper plus bismuth, and in order to obtain the true copper figure a difference method is employed. To the other test, which has been left standing after the filtrate has been neutralised with ammonia as described above, add 1 ml. of 0.5 per cent. potassium cyanide (Reagent 3) for each 0.01 per cent. or part of 0-01 per cent. of copper plus bismuth found from the first uncorrected test. Add a 10-ml. excess of aqueous ammonia (sp.gr. 0*880), 10 ml. of the ammonium citrate solution and 10 ml. of the sodium diethyldithiocarbamate and continue as before. This is due to bismuth alone. The drum difference due to copper is equal to Drum Reading 1 minus Drum Reading 2.Standard graphs are prepared from pure standard copper nitrate solution and drum differences plotted against per cent. of copper (on 5 g. sample). Neutralise the filtrate with ammonia, using litmus as indicator. Cool and transfer to a 200-ml. separating funnel. Dilute to the 100-ml. mark with ether saturated with reagents. Determine the transmittancy as Drum Reading 2. ABBREVIATED METHOD- For routine testing the test may be modified. Weigh only one sample of 5 g. and proceed Determine the trans- Pour the ethereal solution back into the separating funnel Shake Allow the two layers to separate and discard the aqueous layer. Determine The difference in drum reading as in the method given above to determine copper plus bismuth. mittancy as Drum Reading 1.and add about 30 ml. of water and 5 ml. of 0.5 per cent. potassium cyanide solution. for 1 minute. the transmittancy of the ethereal layer as Drum Reading 2. is that due to copper, and the percentage may be read off the appropriate graph.Jan., 19501 NOTES- When the solution is neutralised with ammonia a precipitate may form. This is due to antimony and is caused by insufficient boiling down. The precipitate should be filtered off through a Whatman No. 541 paper. The quantities of the ammonia, citrate and carbamate prescribed should be adhered to. Excess of sodium diethyldithiocarbamate causes a white precipitate to form which does not dissolve in the ether. It does not effect the test except by making separation of the two layers difficult. If very small quantities of copper are present, an initial weight of 10 g. should be taken and the extract made up to only 25 ml. In this case, if the graphs used are based on 5-g. samples, the result must be divided by 8. High copper contents can be determined by the above procedure if smaller weights of sample are taken. For a 10 per cent. copper-lead master alloy it is recommended that 0.5 g. be taken and dissolved, the lead precipitated and filtered off, and the filtrate made up to 500 ml. and 10 ml. (equivalent to 0.01 g. of sample) taken for the determination. QUANTITIES OF COPPER IN LEAD AND LEAD ALLOYS 27 The authors thank the management of the Millwall Branch of the Associated Lead Manufacturers, in whose laboratories this work was carried out, for permission to publish this paper. 136 SIBLEY GROVE MANOR PARK LONDON, E.12 March, 1949