224 Analyst, April, 1968, Vol. 93, @@. 224-227 A Selective Amplification - Titration Procedure for the Determination of Microgram Amounts of Phosphate BY G. F. KIRKBRIGHT, A. M. SMITH AND T. S. WEST (Chemistry Department, Imperial College, London, S . W. 7) Phosphate is converted into phosphomolybdic acid, which is separated from excess of molybdate by extraction. The phosphomolybdate is back- extracted into aqueous solution, and the twelve molybdate ions accompanying each phosphate ion are reduced on a silver reductor column and titrated with 1 0 - S M cerium(1V) by use of a 50-ml burette. Other heteropoly acid- forming elements, e.g., arsenic, antimony, germanium and silicon do not interfere, and there is no interference from a wide range of other ions. The method is both precise and rapid, and has been applied to the submicro determination of phosphorus in a standard organic compound.IN earlier papers from this Department indirect methods for the sensitive and selective determination of phosphate1 and phosphate and silicate2 by molecular-absorption and atomic- absorption spectroscopy, respectively, have been reported. These amplification procedures are based on the determination of the twelve molybdenum atoms associated with each phosphorus or silicon atom after selective solvent extraction of phosphomolybdic or silico- molybdic acid from excess of reagent into isobutyl acetate or butanol. This paper reports a rapid alternative titrimetric procedure of considerably higher precision for the selective determination of microgram amounts of phosphorus that is based on the same amplification procedure.Phosphomolybdic acid is selectively extracted from excess of molybdate into isobutyl acetate, back-extracted and degraded into alkaline solution, and the molybdenum(V1) reduced to molybdenum(V) by a silver reductor column. The determination is then completed by the direct visual or photometric titration of the molybdenum(V) with standard cerium(1V) sulphate solution, with ferroin indicator and a 50-ml grade A b ~ r e t t e . ~ EXPERIMENTAL APPARATUS- An E.E.L. photoelectric titrator and galvanometer (Evans Electroselenium Ltd., Halstead, Essex) was used. Reductor column-Silver for the reductor column was prepared by the method of Kolthoff and Belcher.4 The column used was a laboratory 10-ml burette of internal diameter 1 cm; a column length of 7 cm was used.This column is smaller than that used by Birnbaum and Walde~~,~ so that a smaller total volume of eluate can be obtained. The column was maintained in 2 N hydrochloric acid at all times, and was equilibrated with hot (60" to 80" C) 2 N hydrochloric acid immediately before use. REAGENTS- dihydrogen orthophosphate in distilled water and dilute to 1 litre. 1 ml of solution = 25 yg of phosphorus. Standard ortho@hosphate solzttion-Dissolve 0.1098 g of analytical-reagent grade potassium Molybdate reagent solution-Dissolve 10.69 g of analytical-reagent grade ammonium 0 SAC and the authors. molybdate tetrahydrate, (NH4),Mo,024.4H20, in distilled water and dilute to 1 litre.KIRKBRIGHT, SMITH AND WEST 225 Cerium(IV) sulphate solution-Prepare by diluting British Drug Houses, Ltd.concen- M with 2 N sulphuric trated volumetric solution. acid as required. Dilute the stock 0.05 M solution to 1 ml of 10-3 M cerium(1V) solution = 2.58 pg of phosphorus. Ferroin indicator solutiolz-Use a 5 x 10-3 M aqueous solution of ferroin, Fe(C,,H,N,),SO,. Hydrochloric acid-Analytical-reagent grade. Ammonia solution, sp.gr. 0.88-Analytical-reagent grade. Isobutyl acetate-General-purpose reagent grade. PROCEDURE- The optimum conditions for the selective extraction of phosphomolybdic acid from excess of molybdate reagent into isobutyl acetate have already been describedJ2s6 and were used here. Transfer the sample solution (up to 5 ml), containing between 3 and 30 pg of phosphorus, into a 100-ml separating funnel containing 10ml of molybdate stock solution and 10ml of water.Add sufficient concentrated hydrochloric acid to make the solution 0-96 M with respect to hydrochloric acid. Allow the solution to stand for 5 minutes, add 10 ml of isobutyl acetate and shake the funnel for 1 minute. Allow the phases to separate, discard the aqueous phase, and wash the organic phase with 10 ml of 2 N hydrochloric acid. Shake the organic phase with 5 ml of 4 M ammonia solution, discard the isobutyl acetate and retain the alkaline solution containing the molybdenum(V1). Two drops of indicator solution were used for each titration. REDUCTION AND TITRATION- Transfer the molybdenum solution from the extraction funnel into a 50-ml beaker, add sufficient 6 N hydrochloric acid to neutralise the ammonia and make the solution 2 N with respect to hydrochloric acid.Transfer the solution quantitatively into the top of the reductor column by using 2~ hydrochloric acid as wash liquid. Reduce the molybdenum(V1) to molybdenum(V) on this column by using the method described by Birnbaum and Walden,3 eluting the sample three times with 5-ml portions of hot 2~ hydrochloric acid and three times with 5-ml portions of cold 2 N hydrochloric acid. Collect the eluate (35 to 40 ml) in the titration cell, heat rapidly to boiling, cool the solution to near room temperature, and titrate the molybdenum(V) with standard M cerium(1V) sulphate, with ferroin indicator and E.E.L. filter No. 603. Alternatively, after some experience, the end-point can be detected visually by the disappearance of the last trace of pink colour.We find the visual method to be entirely satisfactory. RESULTS AND DISCUSSION Before the determination of phosphate was attempted via the selective solvent-extraction procedure, pure aqueous molybdenum(V1) solutions were reduced and titrated with cerium(1V) sulphate to ensure that quantitative reduction and recovery of molybdenum was obtained on the reductor column. The molybdenum concentrations were chosen in these experiments to correspond to those which would be obtained after the amplification of the initial phosphorus content (3 to 30 pg) of samples. It was found that a significant blank [1=8 ml of 10-8 M cerium(IV)] was obtained in the absence of molybdenum in the reduction - titration procedure.At the ferroin indicator concentration used, the indicator blank accounts for about one third of this blank (0.65ml) when the titration is carried out with 1 0 - 3 ~ cerium(1V) sulphate. It is our opinion that the remainder of the blank was caused by the formation of hydrogen peroxide on the silver reductol"' and can be eliminated by momentarily boiling the eluate before titration. The indicator blank is reproducible and can safely be subtracted from the sample titration volume. Quantitative recovery and titration of 12 moles of molybdenum is obtained for each mole of phosphorus in the sample. As reported previously, the isobutyl acetate extraction separates phosphomolybdic acid efficiently from the excess of molybdate reagent. The procedure has selectivity equal to those previously reported, and no interference is caused by the presence of large excesses (100-fold by weight) of silicate, arsenate, antimony(V) and germanium(1V).As shown in Table I, phosphorus can be determined in the presence of several milligrams of each of a wide range of cations without interference. Although we have not established it experimentally, we see no reason why a, similar method should not be devised for the titrimetric determination of silicate by modification226 [Analyst, Vol. 93 of the amplification procedure previously described in earlier papers from this Department. There is no reason to suppose that the presence of silicate ions should interfere in any way with the functioning of the Birnbaum and Walden reductor column.KIRKBRIGHT et al. : A SELECTIVE AMPLIFICATION - TITRATION TABLE I DETERMINATION OF PHOSPHORUS IN SYNTHETIC MIXTURES TREATED AS UNKNOWNS Phosphorus 10.0 pg 12.5 20.0 2090 6.0 26.0 20.0 25.0 26.0 Theoretical titre, ml 3.87 4-84 7-74 7.74 1.94 9.68 7.74 9.68 9.68 Titre obtained, ml 3.87 4-64 7-78 7.88 2.05 9.68 7.93 9-94 10.06 Phosphorus found, pg 10.0 12.0 20.1 20.3 8.3 25.0 20.5 26.7 26.0 Foreign ions present, mg Selenium(1V) 169 Tellurium(1V) 1.28 Barium 6.8’7 Manganese 6.6 Silicon 2.0 Aluminium 1.0 Lead 2.07 Cobalt 2-95 Nickel 245 Selenium(1V) 1.69 Tellurium(1V) 1.28 Arsenic(V) 2.0 Germanium(1V) 2.0 Silicon 4-0 Calcium 2.5 Iron( 111) 3.0 zinc 6.5 Barium 6.87 Manganese 6-6 Silicon 2-0 Zinc 6.6 Cadmium 6.6 Copper 6.4 The precision of the reduction - titration procedure was established by repetitive analysis of a pure molybdenum(V1) solution.The standard deviation in the titration value for the determination of 929 pg of molybdenum (equivalent to 25 pg of phosphorus) was 0.05 ml, or 0-55 per cent. The precision of the whole procedure was established by the repetitive determination of 25 pg of phosphorus by the solvent-extraction, reduction and titration procedures. The standard deviation in the titration volume for the determination of 25 pg of phosphorus, repeated seven times, was 0.135 ml, or 1.3 per cent. A measure of the accuracy of the procedure was obtained by determining phosphorus in synthetic solutions containing foreign ions. The results obtained for the determination of phosphorus in microanalytical-reagent grade triphenylphosphine are shown in Table 11.Different sample weights (in the range 60 to 200 pg) were digested with a The results of these analyses are shown in Table I. TABLE I1 DETERMINATION OF PHOSPHORUS IN TRIPHENYLPHOSPHINE SAMPLES Phosphorus in sample taken, CCg 12.11 11.27 7.53 16.77 9.10 16.65 20.20 21.63 10.76 11.72 Phosphorus found, Pg 12.12 11.25 7-50 16.60 9.00 18-12 21.00 21-26 10.66 11.87 Error, per cent. +0.1 - 0.2 - 0.3 - 1.6 - 1.0 - 3.2 + 3.9 - 1.7 - 2.0 + 1.3 Phosphorus in triphenylphosphine, (theory 11.81 per cent.) 11.82 11-79 11.76 11-62 11.68 11-44 12.27 11.76 11.69 11.96April, 19681 PROCEDURE FOR THE DETERMINATION OF PHOSPHATE 227 sulphuric acid - perchloric acid mixture, and the perchloric acid was removed by evaporation. The residual sulphuric acid smear was then treated directly by the recommended procedure. The determination of lO+-g amounts of phosphorus, by using an ordinary burette with titration volumes of several millilitres of a M titrant and visual detection of the end-point, furnishes a good illustration of the analytical value of amplification reactions. We are grateful to the Ministry of Technology for supporting this work, and to Mr. R. W. Fennell of the Royal Aircraft Establishment for the supply of digested triphenylphosphine samples. REFERENCES 1. 2. 3. 4. 5. Djurkin, V., Kirkbright, G. F., and West, T. S., Analyst, 1966, 91, 89. Kirkbright, G. F., Smith, A. M., and West, T. S., Ibid., 1967, 92, 411. Birnbaum, N., and Walden, G. H., jun., J . Amer. Chern. SOC., 1938, 60, 64. Kolthoff, I. M., and Belcher, R., “Volumetric Analysis,’’ Volume 111, Interscience Publishers, a division of John Wiley & Sons Inc., New York and London, 1957, p. 15. Paul, J., Mikrochim. Actu, 1966, 830. Received May 25th, 1967