June, 19611 WITH ETHYLENEDIAMINETETRA-ACETIC ACID 407 Colorimetric Determination of Copper with Zinc OO-Di-isopropyl Phosphorodithioate BY W. A. FORSTER," MRS. P. BRAZENALL? AND J. BRIDGE (Albright & Wilson (Mfg.) Ltd., P.O. Box hTo. 3, Oldbury, Birmingham) Zinc 00-di-isopropyl phosphorodithioate is recommended for the colori- metric determination of copper in the range 4 to 40 pg. Beer's law is obeyed, and both the solid reagent and its solution in carbon tetrachloride are stable. The method is rapid and precise, and large excesses of aluminium, antimony, arsenic, barium, cadmium, calcium, cobalt, chromium, ferrous and ferric iron, lead, magnesium, manganese, mercury, nickel, potassium, sodium and zinc do not interfere. Results are high when more than 0.1 mg of bismuth is present and erratic in presence of more than 5 mg of silver.OF the many reagents recommended for determining copper colorimetrically, cuproinel and zinc dibenzyldithiocarbamate2 are perhaps of most general use. Cuproine is specific, but requires careful control of pH ; zinc dibenzyldithiocarbamate (like zinc 00-di-isopropyl phosphorodithioate) is non-specific, but applicable over a wide range of acidity. Zinc 00-di-isopropyl phosphorodithioate is less subject to interference from antimony, mercury, nickel and silver than is zinc dibenzyldithiocarbamate, but more subject to interference from arsenic and chromium. Bismuth interferes seriously when either of the zinc compounds is used. Busev and Ivanyutin,3 in a short paper published in 1957, described the use of potassium 00-diethyl phosphorodithioate for the colorimetric determination of copper.We were interested in their paper, inasmuch as zinc 00-di-isopropyl phosphorodithioate had been used for several years in this laboratory as a reference substance for determining phosphorus in organic compounds and for the colorimetric determination of dialkyl phosphorodithioates. We therefore decided to investigate its suitability as a colorimetric reagent for copper. EXPERIMENTAL Acid solutions containing copper were extracted with a 0-04 or 0-004 per cent. solution The extracts obtained (;) Beer's law is obeyed over the range 4 to 40 pg of copper when a mercury-vapour The yellow complex with copper of zinc 00-di-isopropyl phosphorodithioate in carbon tetrachloride. were examined with a Spekker absorptiometer, and the results showed that- lamp and filters transmitting a t 4047 A are used.has an absorption maximum at approximately 4200 A. * Present address: B.I.P. Chemicals Ltd., Oldbury, Birmingham. t Present address : 58 Darnick Road, Sutton Coldfield, Warwickshire.408 FORSTER, BRAZENALL AND BRIDGE COLORIMETRIC DETERMINATION OF [VOl. 86 (ii) Bright sunlight causes the colour to fade slowly; in one experiment, the optical (iii) A single extraction for 30 seconds removes copper completely. (iv) Wide variation in acidity is permissible. density decreased by 36 per cent. in 3 hours. Experiments were carried out in which 50-ml portions of solutions of different acidity and containing 24pg of copper were extracted with 25-ml portions of a 0-004 per cent.solution of zinc 00-di-isopropyl phosphorodithioate. Twenty-two such experiments under the conditions of acidity listed below gave a standard deviation of only +_0-002 unit of optical density. Acid . . . . .. . . Sulphuric Hydrochloric Perchloric Orthophosphoric Nitric Acidity of extracted solution 0.1 to 5 N 0.1 t o 5 N 0.1 to 5 N 0.1 to 1.5 M 0.2 to 2 1 \ ~ The colour of the extract from a solution 3 N in nitric acid fades rapidly; for the other acids, no experiments were made at concentrations higher than the maximum values indicated above. (v) Wide variation in the concentration of the zinc 00-di-isopropyl phosphoro- dithioate solution is permissible. In most of the preliminary experiments we used a 0.004 per cent. solution, but studies on interfering ions showed that a 0.04 per cent.solution is better. (vi) Delay between the addition of the zinc 00-di-isopropyl phosphorodithioate solution and shaking should be avoided. Experiments were made in which intervals of 0, 2, 5 and 10 minutes elapsed, and the optical-density readings were 0.232, 0.224, 0.216 and 0.191, respectively. The probable explanation for this decrease in optical density is that dialkyldithiophosphoric acid slowly passes into the aqueous layer and there forms the cupric complex. It is known that the latter, in aqueous solution, rapidly changes to a colourless cuprous compound. Both 0.004 and 0.04 per cent. solutions of zinc 00-di-isopropyl phosphorodithioate were stored for 10 weeks in plain- glass bottles on the laboratory window-sill without deterioration. (vii) The stability of the reagent is excellent.METHOD REAGENTS- Carbon tetrachEoride, pwrified-The use of technical-grade carbon tetrachloride occasionally leads to low results, owing, we suspect, to the presence of traces of phosgene. Purify the solvent by shaking 600 ml of it successively with 100 ml of dilute ammonia solution (1 + 4) and two 100-ml portions of water. Dry the carbon tetrachloride over anhydrous sodium sulphate, and store in a dark-glass bottle. Zinc 00-di-isopropyl phosphorodithioate solution, 0.04 per cent. w/v-Dissolve the neces- sary amount of solid reagent, prepared as described below, in purified carbon tetrachloride. Hydrochloric acid, sp .gr. 1.18-Analyt ical-reagent grade. Standard copper sohution-Dissolve 0.1964 g of hydrated copper sulphate, CuS0,.5H20, in water in a 250-ml calibrated flask, acidify with a little hydrochloric acid, and dilute to the mark with water.Dilute 10 ml of this solution, again acidified with hydrochloric acid, to 250 ml; prepare this solution freshly each day, as required. PREPARATION OF ZINC 00-DI-ISOPROPYL PHOSPHORODITHIOATE- Weigh 10 g of phosphorus pentasulphide into a Quickfit & Quartz 250-ml conical flask, attach a reflux condenser, add 100ml of isopropyl alcohol, and heat for 5 minutes or more on a water bath, with frequent agitation, to give a clear solution. Cool to about 40" C, and expel hydrogen sulphide by passing a stream of nitrogen through the liquid for 3 minutes. Add 4 g of analytical-reagent grade zinc oxide, heat on the water bath for 10 minutes, and filter the hot mixture through a sintered-glass funnel (porosity No.4). Chill the filtrate, separate the crystals that form, and wash them with two small portions of cold isopropyl alcohol. Recrystallise twice from hot isopropyl alcohol, and dry the white crystalline product in air ; this procedure yields about 6 g of zinc 00-di-isopropyl phosphorodithioate. Each melted at 145" "I 2" C and contained proportions of carbon, hydrogen and phosphorus that agreed well with the values calculated from the formula [(C,H,O),PSS],Zn. All four samples gave satisfactory results in the colorimetric determination of copper. 1 ml E 8 pg of copper. Four samples were prepared in this way.June, 19611 COPPER WITH ZINC 00-DI-ISOPKOPYL PHOSPHORODITHIOATE 409 PROCEDURE- Place an aliquot of solution containing 4 to 40 pg of copper in a Pyrex-glass 125-ml separating funnel, and dilute with water to approximately 50 ml, adding sufficient hydro- chloric acid, sp.gr.1.18, to make the diluted solution approximately 4~ in this acid. By pipette, add 25 ml of 0.04 per cent. zinc 00-di-isopropyl phosphorodithioate solution to the contents of the separating funnel; fill the pipette by means of suction from a water pump. Without delay, shake the separating funnel vigorously for 30 seconds, and allow the layers to separate. Dry the stem of the separating funnel with a roll of filter-paper, and insert a loose plug of cotton-wool. Measure the optical density of the extract in a 2-cm cell with a Spekker absorptiometer; use a mercury-vapour lamp and a filter transmitting at 4047 A.,\lternatively, use any system permitting measurements to be made in the vicinity of 4200 A. TABLE I EFFECTS OF VARIOUS CATIONS The amount of copper taken for determinations by the proposed method was 24 pg; amounts from 10 to 40 pg were used in the zinc dibenzyldithiocarbamate method, as recommended by Johnson.2 The figures for each cation are the maximum amounts that may be present if the error is not to exceed &2 pg of copper Maximum amount permissible in- I A 1 zinc dibenzyldithio- Cation added proposed method, carbamate method, g !? ~ 1 3 + 0.5 0.5 Sb3 + 0.05" <0.01 -\s3+ 0.17 0.5 Ra2L 0.5 - Ri3f < 0-000 1 < 0.002 Cd2+ 0.3 0.5 Ca2 + 0.5 0.5 co2+ 0.01 0.0 1 0.5 Cr3 + - CrsL 0.01 - Fe2'- 0.5 0.5 Fe3 + 0.01 < 0.5 Pb2+ 0.5 0.5 Mg2+ 0.5 - Mn2+ 0.5 0.5 Hg2+ 0.05 0~0001 Ni2 + 0.5 0.05 K" 0.5 - Ag+ 0.005 0~0001 Na+ 0-5 - Zn2+ 0.5 0.5 * Colour fades in the presence of antimony, and the optical t Solution and reagent shaken for S minutes.density must be read with dispatch. CALIBRATION- Transfer portions of the standard copper solution ranging from 0 to 5 m l to separate 125-ml separating funnels, and add to each portion, from a measuring cylinder, 17.5 ml of hydrochloric acid, sp.gr. 1-18. Dilute each solution with water to approximately 50 ml, add reagent, extract, and measure the colour as described above. INTERFERENCES CATIONS- The effects of twenty cations were studied, and the amounts that may be present when determining 24 pg of copper are listed in Table I, which also shows comparative results for zinc dibenzyldithiocarbamate.Bismuth interferes seriously with the proposed method by forming a complex that absorbs strongly in the near-ultra-violet region ; this interference could perhaps be overcome by using a prism or grating spectrophotometer. Silver must not be present in amounts greater than 5mg.410 FORSTER, BRAZENALL AND BRIDGE [Vol. 86 The effects of the twenty cations were also investigated by means of a modified method, the concentration of hydrochloric acid being decreased from 4 to 0.1 N and that of the zinc 00-di-isopropyl phosphorodithioate solution from 0.04 to 0.004 per cent. Several cations interfered to a greater extent when the modified method was used, as shown by the results below.Cation . . . . . . . . . . . . Sb3+ Fez+ Fe3+ Hg2+ Ni2+ Maximum amount permissible for error of 2 2 pg of copper, g . . . . .. . . . . 0.0005 0.05 t0.005 0.0001 0.01 Only arsenic interfered less, the permissible amount being 0.5 g. Interference from bismuth-As much as 0.01 g of bismuth may be present if the acidity of the test solution is kept at 4~ in hydrochloric acid and the concentration of the zinc 00-di-isopropyl phosphorodithioate solution is decreased to 0.004 per cent. Interference from ferric iron also disappears under these conditions, and as much as 0.5 g of this ion may be present. AWONS- are summarised in Table 11; only molybdate and iodide interfered to a marked extent. The results of experiments carried out in presence of nine sodium and potassium salts TABLE I1 EFFECTS OF VARIOUS SALTS The figure for each salt is the maximum amount that may be present if the error in determining 24 pg of copper is not to exceed +2 pg Maximum amount permissible g Sodium acetate .. . . .. .. .. 0.5 Potassium bromide . . .. .. .. 0.5 Trisodium citrate . . .. .. . . .. 0.5 Potassium iodide . . .. . . .. 0.0 1 Sodium molybdate dihydrate . . . . .. 0.05 Sodium nitrite . . .. .. .. . . 0*25* Potassium oxaIate monohydrate . . .. 0.5 Potassium permanganate . . .. .. 0.5t Salt added in proposed method, Potassium sodium tartrate tetrahydrate . . 0.5 * Add 0.5 g of potassium permanganate, boil until pale brown in colour, t Boil until pale brown in colour, and cool before extraction. and cool before extraction with reagent solution. CONCLGSION In developing this method, we had in mind its eventual application to calcium phos- phates and decided on that account t o use hydrochloric acid solutions. It seems probable from the information under “Experimental” that other acids could equally well be used in determining copper colorimetrically. We thank Mr. S. F. Holder for suggesting this investigation and for his interest in its progress and the Directors of Albright &Wilson (Mfg) Ltd. for permission to publish this paper. REFERENCES 1. 2. 3. Diehl, H., and Smith, G. F., “The Copper Reagents: Cuproine, Neocuproine, Bathocuproine,” Johnson, W. C., Editor, “Organic Reagents for Metals, and Other Reagent Monographs,” Fifth Busev, A. I., and Ivanyutin, M. I., Vestn. Moskov. Univ., 1957 (5), 157; Chem. Abstr., 1958, 52, Received December 30th, 1960 G. Frederick Smith Chemical Co., Columbus, Ohio, 1958. Edition, Hopkin & Williams Ltd., Chadwell Heath, Essex, 1955, Volume I, p. 188. 19,676.