406 FOGG AND WILKINSON THE COLORIMETRIC [Vol. 83 The Colorimetric Determination of Phosphorus BY D. N. FOGG AND N. T. WILKINSON (Impevial Chemical Industries Ltd.. Research Department, Alkali Division, Winnington, Northwich, Cheshire) In the colorimetric determination of phosphate by Denigks’s method the molybdophosphate is reduced by the addition of stannous chloride. In the method described, stannous chloride has been replaced by ascorbic acid. Colour development is rapid a t the boiling-point of the solution, and, once developed, the molybdenum blue colour :!s extremely stable a t room tempera- ture. The stability of the molybdenum blue allows the colour intensity to be measured either visually or instrummtally. Instrumental measurement has made it possible to apply the method to phosphate concentrations in the range 1 to 600 pg (as P,O,).The ascorbic acid is added in solid form, since its solution is unstable. Reactions with arsenic and silica have been investigated, and application of the method to the determination of phclsphate in boiler water and effluents is described. As a result of our work on the determination of selenium,l it occurred to us that ascorbic acid might be a suitable reagent to replace stannous chloride for the reduction of molybdo- phosphate in the colorimetric determination of phosphate by Denighs’s method.2 Holman and Pollard3 have modified Denigh’s method and use a colour disc to match the colours in a Lovibond Nessleriser. Interfering elements have been investigated by the previously mentioned workers, and it is of significant importance to us that ferric iron must not be present in amounts greater than 1 p.p.rri.in the final solution. Although this may cause no concern in water analysis, since most waters contain less than this amount of iron, it will undoubtedly be of significance in the analysis of effluents. Another point of importance is that the stannous chloride solution must be freshly prepared, but it is doubtful whether each batch of solution prepared contains exactly the same concentration of the stannous salt. In our experience, different calibration curves are obtained with standard concentrations of phosphate when ammonium molybdate from different sources is used in Holman and Pollard’s3 modification of Denigks’s method. In a literature survey we found that Ammori and Hinsberg4 were the first to use ascorbic acid for the reduction of molybdophosphate.Lowry, Roberts, Leiner, Wu and Farrs later modified the procedure of Ammon and Hinsber;: by using a more concentrated solution of ascorbic acid and heating for a longer time at 37” C. Chen, Toribara and Warnere have applied the method modified by Lowry, Roberts, Leiner, Wu and Farr to the determinationJuly, 19581 DETERMIN.4TION OF PHOSPHORUS 407 of phosphorus in blood, plasma, serum and urine, and have compared the method with that of Fiske and Subbarow,' in which the reduction is carried out with sodium sulphite and aminonaphtholsulphonic acid. We have not investigated the method described by Chen, Toribara and Warner,6 since it has the disadvantage of a mixed reagent containing sulphuric acid, ammonium molybdate and ascorbic acid, which must be prepared freshly each day.Also, in their procedure, the solution must be heated for 2 hours to achieve full colour development and the range is limited to 20 pg of phosphate as P,O,. We have found that, by using the sulphuric acid - ammonium molybdate solution recommended by Holman and Pollards and adding solid ascorbic acid before heating the solution to boiling, the blue colour formed in the presence of phosphate is fully developed after the solution has been boiled for 1 minute and the colour is stable at room temperature for several weeks. The colours are reproducible with a reagent prepared from different batches of ammonium molybdate. No great accuracy is required in weighing the 0.1 g of ascorbic acid used in each determination, since the same result is given if 0.2 g is used.Amounts of iron up to 0.02 g, chloride equivalent to 2 g of sodium chloride, nitrate equivalent to 0.05 g of sodium nitrate and sulphate and perchlorate equivalent to at least 5 g of sodium sulphate and 5 g of sodium perchlorate have no effect on the proposed method. The presence of 0.05 g of soluble silica can also be tolerated. Arsenic present as arsenate gives the same blue colour as phosphate, but, if the arsenate is reduced to arsenite, 0.01 g of arsenic has no effect. We have reduced the arsenate with sodium metabisulphite, using a volume of sulphuric acid equivalent to the weight of meta- bisulphite used. It was necessary to set the solution aside for 4 hours in order to reduce the arsenate completely.Organic matter can be destroyed by wet oxidation, and, if a sample contains sodium chloride or sodium nitrate in amounts greater than those mentioned, chloride and nitrate can be removed by evaporation with a slight excess of sulphuric acid. The solution is then neutralised before the method is applied. Meta- and pyrophosphates are partly hydrolysed during the method and should therefore be completely hydrolysed before application of the method to a sample containing these phosphates. EXPERIMENTAL The reagents used for the experimental work are described on p. 413. In our first series of experiments, 0, 5.0, 10.0, 15.0, 20.0 and 25.0-ml portions of the dilute phosphate solution were placed in six 100-ml beakers and each solution was diluted to 40 ml with distilled water.One millilitre of ammonium molybdate - sulphuric acid solution was added to each, and then 0.1 g of ascorbic acid, after which the solutions were stirred until the ascorbic acid had dissolved. No colour developed after the solutions had been standing for several minutes. Each solution was heated to the boiling-point, when it appeared that the ascorbic acid had reduced the molybdophosphate completely. Boiling was continued for 1 minute. The solutions were cooled, transferred to 50-ml calibrated flasks and then each was diluted to the mark. The optical density of each solution was measured in a 1-cm cell with a Spekker absorptiometer, Ilford No. 608 red filters being used. Distilled water was used in the comparison cell.The results, corrected for a blank value of 0.007, were as follows- -4mount of phosphate, as P,O,, pg . . 25 50 100 150 200 250 Corrected indicator-drum reading. . . . 0.065 0.130 0.265 0.398 0.528 0.660 It can be seen that, when the weight of phosphate is plotted against the corrected indicator- drum reading, the graph is linear. A further series of tests with amounts of phosphate between 0 and 50 pg was carried out, the final colour measurement being made in 4-cm cells. The results, corrected for a blank value of 0.017, were as follows- .\mount of phosphate, as P,O,, pg . . 5 10 20 30 40 50 Corrected indicator-drum reading. . . . 0.053 0.108 0.213 0.325 0.428 0.533 EFFECT OF HIGHER CONCENTRATION OF AMMONIUM MOLYBDATE - SULPHURIC ACID SOLUTION - Two series of experiments were carried out.In six 100-ml beakers, 0, 5.0, 10.0, 15.0, 20.0 and 25.0-ml portions of the dilute phosphate solution were placed and each was diluted408 FOGG AND WILKINSOX : THE COLORIMETRIC [Vol. 83 to 40 nil. Three millilitres of the ammonium molybdate - sulphuric acid solution were added to each and then 0.1 g of ascorbic acid. The solutions were stirred until the ascorbic acid had dissolved. Each solution was heated to the boiling-point and boiled for 1 minute. The solutions were cooled and each was transferred to a 50-ml calibrated flask and then diluted to the mark, The optical densities of the solutions were identical with those for the first series of results shown above. iz similar series of tests was carried out in which 5.0 ml of the ammonium molybdate - sulphuric acid solution were added.The results were again identical with those for the first series shown above. REACTIOS WITH SILICA- It is well known that, under certain conditions of acidity and concentration ofmolybdate, silica will produce a yellow molybdosilicate complex, which can be reduced to form a molybdenum blue - silica complex. It was therefore important to know how silica behaved under the conditions of the method. Since the samples we were likely to analyse for phosphate would contain a relatively high proportion of silica, we investigated the (effect of silica in fairly high concentration. A solution of sodium silicate was prepared, 1 ml of which contained 0.0025 g of silica. 111 six 100-ml beakers, 0, 4.0, 8.0, 12.0, 16.0 and 2O.O-ml portions of this solution were placed.Each solution was diluted to 40 ml and then treated further with 1 ml of ammonium molyb- date - sulphuric acid solution and 0.1 g of ascorbic acid as described for the phosphate test. The optical densities were measured in 1-cm cells; the results, corrected for a blank value of 0.005, were as follows- .\mount of silicate, as SiO,, g . . . . 0.01 0.02 0.03 0.04 0.05 Corrected indicator-drum reading. . . . 0.060 0.070 0.105 0.125 0.156 It can be seen from these results that silica reacts under the conditions of the method, but the relationship between increasing silica content and indicator-drum reading is not linear. lye have shown that different amounts of ammonium molybdate - sulphuric acid solution produce the same colour intensity with equal amounts of phosphate.We decided, therefore, to ascertain the effect of different amounts of ammonium molybdate - sulphuric acid solution on its reaction with silicate. Portions of the solution each containing 0.05 g of silica were measured :into five beakers, each solution was diluted to 40 ml and different volumes of the ammonium molybdate - sulphuric acid solution and 0.1 g of ascorbic acid were added to each. The solutions were heated to the boiling-point and boiled for 1 minute, after which they were cooled, transferred to 50-ml calibrated flasks and diluted to the mark. The optical densities were measured in 1-cm cells; the results were as follows- ‘4 series of tests was made with the solution of sodium silicate..Imonnt of ammonium molybdate - sulphuric Indicator-drum reading . . . . . . . . 0.150 0.060 0.025 0.010 0.010 acid solution added, ml . . . . . . 1.0 2.0 3.0 4-0 5.0 The method was then applied to solutions containing both silicate and phosphate, 4 ml The results are shov;n of the ammonium molybdate - sulphuric acid solution being used. in Table I. TABLE I OPTICAL DENSITY OF SOLUTIONS CONTAINING PHOSPHATE AND SILICATE The optical densities were measured in 1-cm cells .Amount of phosphate Amount of silicate Indicator-drum reading, present, as P,Oj, pg added, as SiO,, g Indicator-drum reading corrected for blank value 0 0.05 0.015 - 0 0.03 0,012 - 0 0.02 0.012 - 0 0.01 0.010 - 20 0.05 0.070 0.055 50 0.05 0,150 0.135 100 0.02 0.278 0.268 150 0.01 0405 0.395 250 0.05 0.675 0.660 10 0.03 0.040 0.028July, 19581 DETERMISATION OF PHOSPHORUS 409 By comparing the results in Table I with those obtained with phosphate alone in presence of 4 ml of ammonium molybdate - sulphuric acid solution, it can be seen that interference from between 0.01 and 0.05 g of silicate, as SO,, is negligible.The optical densities of the solutions of phosphate alone were measured in I-cm cells; the results, corrected for a blank value of 0.007, were as follows- .Imount of phosphate, as P,O,, pg 10 20 30 40 50 100 150 250 Corrected indicator-drum reading 0.025 0.048 0.073 0.100 0.130 0.265 0,398 0.660 Since we had now established the volume of ammonium molybdate - sulphuric acid solution that was suitable for the determination of phosphate without interference from silicate, all subsequent experiments were carried out with 4ml of this reagent.EFFECT OF ISCREASE IS AMOUNT OF ASCORBIC ACID- A series of experiments was carried out exactly as described under “Method” on p. 413, except that different amounts of ascorbic acid were used. The results are shown in Table 11. TABLE I1 The optical densities were measured in 1-cm cells EFFECT OF ASCORBIC ACID OX OPTICAL DEKSITY OF PHOSPHATE SOLUTIOSS .\mount of phosphate Indicator-drum reading, present, as P,O,, pg Indicator-drum reading corrected for blank value Colour of final solution 0 0.005 - - 10 0.033 0.028 Blue 20 0.060 0.055 Blue 50 0.135 0.130 Blue 100 0.270 0.265 Blue 150 0.403 0.398 Blue 200 0.533 0,528 Blue 250 0.665 0.660 Blue I n presence of 0.2 g of ascorbic acid- 112 presence of 0.5 g of’ ascorbic acid- 0 0.010 - Yellow 10 0.040 0.030 Yellowish green 20 0.065 0.055 Yellowish green 50 0.140 0.130 Green 100 0.276 0,266 Blue-green 150 0.408 0.398 Blue 200 0.536 0.526 Blue 250 0.668 0.658 Blue ..Although the indicator-drum readings with these higher amounts of ascorbic acid agree with those found when 0.1 g is used, the colours obtained with 0.5 g of ascorbic acid were less satisfactory visually.Since excellent colours and results were obtained with 0.1 g of ascorbic acid, we adhered t o this amount in all subsequent work. The ascorbic acid was added in solid form and not as a solution, since the latter slowly deteriorates. STABILITY OF THE MOLYBDEWX BLUE COLOUR- on p. 413. the results were as follows- A series of standards was prepared as described under “Preparation of Calibration Curves” The optical density of each solution was measured immediately in 1-cm cells; -4mount of phosphate, as P,O,, pg .. 0 10 50 100 250 Indicator-drum reading . . .. . , 0.007 0.033 0.137 0.272 0.667 These solutions were again measured after 1 day, 1 week, 1 month and 3 months. period there was no change in the optical density of any of the solutions. RANGE OF THE METHOD- The method can be modified to cover a wider range of phosphate concentration than shown previously by diluting the final solution to 100 ml and using 1-cm cells in the measure- ment of optical density. Typical results, corrected for a blank value of 0.005, were as follows- Over this Amount of phosphate, as P,O,, pg .. 250 300 400 500 600 Corrected indicator-drum reading. . . . 0.335 0.405 0.540 0.675 0.810410 FOGG AND WILKIKSOK : THE COLORIMETRIC [Vol. 83 Similarly, the method can be made to cover ,I very low range of phosphate concentration by diluting the final solution to 50 ml and measuring the optical density in 4-cm cells or larger. Solutions were prepared each containing 2 g of sodium chloride dissolved in 10 ml of Different amounts of phosphate were added to each and the method was The results obtained were exactly the same as when phosphate alone was present. A similar series of experiments was then carried out in the presence of 3 g of sodium EFFECT O F SODIUM CHLORIDE- distilled water. applied. chloride, the results of which are shown in Table 111.TABLE I11 The optical densities were measured in 1-cm cells Amount of Amount of phosphate added, Indicator-drum phosphate found, as PaOm Indicator-drum reading, corrected as PaO,, EFFECT OF PRESEYCE OF 3 g OF SODIUM CHLORIDE 0s RECOVERY OF PHOSPH.4TE Pg reading for blank value Pg - - 0 0.005 10 0.030 0.025 10 50 0.120 0.115 43 100 0.237 0.232 87 150 0.327 0.322 122 200 0.465 0.460 172 250 0.530 0,525 198 EFFECT OF SODIUM SULPHATE- Several solutions were prepared containing 2, 3 and 5 g of sodium sulphate and different The results were exactly amounts of phosphate, and the method was applied as described. the same as when phosphate alone was present. EFFECT OF SODIUM NITRATE- When the method was applied to solutions containing 1 g of sodium nitrate and different amounts of phosphate, no blue colour 'was formed.Further experiments showed that only 0.05 g of added sodium nitrate could be tolerated. The molybdenum blue colour was not stable and faded within 12 hours. Since sodium chloride in amounts above 2 g per 50 ml of final solution and sodium nitrate above 0.05 g interfere with the method, but sodium sulphate up to 5 g has no effect, it is obvious that interference from chloride and nitrate can be overcome by evaporation with sulphuric acid and subsequent neutralisalion of the acid with sodium hydroxide. Similarly, organic matter interferes with the test, but can be destroyed with nitric and sulphuric acids, the test being applied on the rewlting solution after neutralisation. EFFECT OF SODIUM PERCHLORATE- Since the destruction of organic matter in a sample is often accomplished more effectively by using perchloric acid as well as nitric and sulphuric acids, the effect of sodium perchlorate on the method was investigated. Several solutions were prepared containing 1 and 5 g of sodium perchlorate and difterent amounts of phosphate.The method was applied and the results were exactly the same as when phosphate alone was present. EFFECT OF FERRIC IRON- When the method was applied to solutions containing known amounts of phosphate and 0.02 g of iron, added as ferric chloride, excellent results were obtained, but with increased amounts of iron the results were low. In the presence of 0.03 g of iron, the recoveries of added phosphate varied from approximately 60 to 80 per cent.with increasing amounts of phosphate. EFFECT OF ARSENIC- Under the conditions of the method, arsenate gives the molybdenum blue reaction. In a series of experiments we found that, if the arsenzte were first reduced to arsenite, relatively large amounts of arsenic gave no reaction.July, 19581 DETERMINATION OF PHOSPHORUS 41 1 Solutions were prepared containing different amounts of phosphate and to each was added 0.01 g of arsenic as sodium arsenate. The solutions were diluted to 30 ml and t o each was added 1 g of sodium metabisulphite and then 12 ml of AT sulphuric acid. The solutions were set aside for 4 hours, after which 4.0 ml of ammonium molybdate - sulphuric acid solution were added and then 0.1 g of ascorbic acid. The solutions were heated to boiling- point and boiled for 1 minute.They were then cooled, diluted to the mark in 50-ml calibrated flasks and the optical density of each was determined. The results are shown in Table IV. TABLE IV EFFECT OF ARSEKIC ON THE RECOVERY OF PHOSPHATE Each sample contained 0.01 g of arsenic, as sodium arsenate The optical densities were measured in 1-cm cells Amount of phosphate added, as P,Os, r g 0 10 50 100 1 A0 _. . 200 250 Indicator-drum reading 0.015 0.040 0.148 0.277 0.410 0.545 0.675 Indicator-drum reading, corrected for blank value - 0.025 0.130 0.262 0.395 0.530 0.660 Amount of phosphate found, as P A Pg 10 49 99 149 200 250 - APPLICATION TO SOLUTIONS CONTAIXING CHLORIDE, NITRATE AKD ORGANIC MATTER- Determinations of phosphate were carried out on solutions containing sodium chloride, sodium nitrate, sucrose and phosphate.Organic matter was destroyed and the determina- tion carried out exactly as described under “Determination of Phosphate in Effluents” on p. 414. The results are shown in Table V. TABLE v RECOVERY OF PHOSPHATE FROM SOLUTIONS CONTAIKING CHLORIDE, Each sample contained 0.2 g of sucrose, 3 g of sodium nitrate and 2 g of sodium chloride The optical densities were measured in 1-cm cells KITRATE AND ORGANIC MATTER Amount of phosphate added, as P,05, r g 0 20 50 100 Amount of Indicator-drum phosphate found, Indicator-drum reading, corrected as P,05. reading for blank value Pg 0.020 - - 0.075 0.055 21 0.150 0.130 50 0-285 0,265 100 SENSITIVITY OF THE METHOD- Comparison of the sensitivity of the method with Holman and Pollard’s modification of Denig6s’s method was made, and, as the results in Table VI show, the sensitivity is about half that of Denig6s’s method.TABLE VI SENSITIVITY OF THE METHOD The optical densities were measured in 4-cm cells Indicator-drum reading Amount of phosphate Pollard’s method when the proposed added, as P,O,, pg was used method was used when Holman and Indicator-drum reading 5 0.130 0.053 10 0.258 0.108 15 0.388 0.160 20 0.615 0.215 25 0.636 0.270[Vol. 83 We are of the opinion, however, that this seduction in sensitivity is well compensated for by the wide range of phosphate concentration that the method will cover, its relative freedom from interference and the stability of the molybdenum blue colour. APPLICATION TO THE DETERMINATION OF PHOSF’HATE IX BOILER WATER- 412 FOGG AND WILKINSON : THE COLORIMETRIC A sample of boiler water with the following composition was used for the experiments.Sodium carbonate . . . . . . 689 p.p.m. Sodium hydroxide . . . . . . 2000 p.p.m. Sodium sulphate . . .. . . 7029 p.p.m. Silica . . . . . . . . . . 37 p.p.m. Sodium sulphite . . .. . . 143p.p.m. Sodium chloride . . * . . . 1958 p.p.m. Five-millilitre portions of the boiler water were measured into 100-ml beakers and sufficient N sulphuric acid was added to neutralise the alkalinity. The amount of N sulphuric acid required was determined on a separate portion of the boiler water. Known volumes of standard phosphate solution were added to t.he neutralised solutions, each solution was diluted to 40 ml and the phosphate was determined as described under “Procedure,” beginning a t “Add 4.0 ml of the ammonium molybdate - sulphuric acid solution.” The results are shown in Table VII. Amount of phosphate added, as P,O,, rLg 0 20 50 100 200 500* TABLE VII RECOVERY OF PHOSPHATE FROM BOILER WATER Indicator-drum Indicator -drum reading reading (I-cm cell) (4-cm cell) 0,008 0*03!2 - 0.24’7 - 0.555 0,280 - 0.542 - 0.680 - Total amount of phosphate found, as P,O,, Pg 3 23 52 I05 205 503 Corrected amount of phosphate found, as P,O,, Pg 20 49 102 202 500 - * Final solution diluted to 100 ml.APPLICATIOS TO THE DETERMISATION OF PHOSPHATE IN EFFLUESTS- A sample of effluent with the following composition was used for the experiments, Calcium bicarbonate . . , . 235 p.p.m. Calcium sulphate .. , . . . 196 p.p.m. Calcium chloride . . . . . . 610p.p.m. Sodium chloride . . . . , . 1400 p.p.m. Magnesium chloride . . . . 66 p.p.m. Organic matter . . . . . . 100p.p.m. To a series of 50-ml portions of the effluent, known volumes of standard phosphate solution were added and the determination was carried out as described under “Determination of Phosphate in Effluents.” The results are shown in Table VIII. TABLE VIII RECOVERY OF PHOSPHATE FROM EFFLUENT Amount of phosphate added, as pz05, r g 0 10 20 50 100 200 500* Total amount of Indicator-drum Indicator-drum phosphate found, reading reading as P,O,, 0.025 0.100 9.5 - 0.2111 20 - 0.320 29.5 60 0,160 - 113 0.300 - 0.563 - 212.5 0.690 - 512 * Final solution diluted to 100 ml. (I-cm cell) (4-cm cell) rLg Corrected amount of phosphate found, as P,O,, /G 10.5 20 50.5 103.5 203 502.5 -July, 19581 DETERMINATION OF PHOSPHORUS 413 It can be seen that, in addition to the known composition, the effiuent contained a small amount of phosphorus.REAGENTS- Ammonium molybdate - sulphuric acid solution-Dissolve 10.0 g of crystalline ammonium molybdate in about 70 ml of distilled water and dilute the solution to 100 ml. Carefully add 150 ml of sulphuric acid, sp.gr. 1.84, to 150 ml of distilled water, mixing the solution during the addition. Cool the solution, add the ammonium molybdate solution carefully and with mixing to the diluted sulphuric acid and allow the mixture to cool. METHOD Ascorbic acid. Standard Phosphate solution-Dissolve 0.7669 g of potassium dihydrogen orthophosphate For use, dilute 25 ml of this solution in distilled water and dilute the solution to 1 litre.to 1 litre. PROCED u RE- Measure a suitable volume of the sample solution containing the phosphate present as orthophosphate, neutralise it and adjust the volume to 40 ml either by dilution or evaporation. Add 4-0 ml of the ammonium molybdate - sulphuric acid solution and mix. Add 0.1 g of ascorbic acid, heat the solution to boiling-point and boil for 1 minute. Cool the solution, transfer it to a 50-ml calibrated flask (or a 100-ml calibrated flask for phosphate contents between 250 and 600pg) and dilute to the mark with distilled water. Carry out a blank test on the reagents with distilled water in place of the sample. Measure the optical densities of the blank and test solutions in either 1-cm or 4-cm From a previously prepared 1 ml = 10 pg of P,O,.cells in an absorptiometer with Ilford KO. 608 red filters. calibration curve, read the amounts of phosphate present in the two solutions. PREPARATION OF CALIBRATION CURVES- For the range 0 to 50 pg ofphosphate-In seven 100-ml beakers place 0.0, 0.5, 1.0, 2.0, 3.0, 4.0 and 5.0-ml portions of standard phosphate solution. Dilute each solution to 40 ml and continue as described under “Procedure.” For the range 0 to 250 pg of phosphate-In a series of beakers place 0.0, 5.0, 7.5, 10.0, 12.5, 15.0, 17.5, 20.0, 22.5 and 25.0-ml portions of standard phosphate solution. Dilute each solution to 40 ml and continue as described under “Procedure.” Measure the optical densities in 1-cm cells.For the range 250 to 600 pg of phosphate-In a series of beakers place 0.0 and from 25.0 to 60.0-ml portions, in steps of 5 ml, of standard phosphate solution. Adjust the volumes of the solutions to 40 ml either by dilution or evaporation and continue as described under “Procedure,” but dilute the final solutions to 100 ml. Neasure the optical densities in 1-cm cells. Measure the optical densities in 4-cm cells. APPLICATIOKS OF THE METHOD DETERMIXATION OF PHOSPHATE IN BOILER LVATER- Calgon is frequently used in the treatment of boiler-feed water to prevent the formation of scale and it is often required to know the phosphate content of the boiler blow-down water. During the time the water is in the boiler, the Calgon is usually completely hydrolysed to orthophosphate.Direct application of the method can therefore be carried out on a suitable volume of the sample that has been neutralised with Ar sulphuric acid. The recom- mended procedure is as follows. Measure 20 ml of the sample into a 100-ml calibrated flask. Add sufficient N sulphuric acid to neutralise the alkalinity and dilute to the mark. Measure 25 ml of the solution into a 100-ml beaker, dilute to 40 ml and continue as described under “Procedure.” For the determination of phosphate present in boiler-feed water as hexametaphosphate we recommend hydrolysis of the hexametaphosphate by neutralisation of the sample, addition of 1 ml of hydrochloric acid, spgr. 1.18, and evaporation of the solution to dryness. Complete hydrolysis of hexametaphosphate is not usually achieved simply by boiling an acidified solution of the sample.414 STEELE : A DIFFERENCE PHOTOMETRIC hIETHOD FOR [Vol.83 DETERMISATION OF PHOSPHATE IN EFFLUENTS-. Neutralise the sample by the addition of either dilute sulphuric acid or dilute sodium hydroxide solution. Add 3 ml of sulphuric acid, spgr. 1.84, and evaporate the solution on a sand-bath until white fumes of sulphur trioxide appear. If organic matter is present, remove the beaker from the sand-bath, allow it to cool somewhat, add 1 ml of nitric acid, sp.gr. 1.42, and again heat until fumes of sulphur trioxide appear. Again allow to cool somewhat, add 1 ml of nitric acid, 1 ml of 60 per cent. perchloric acid and heat on a sand-bath until fumes of sulphur trioxide appear; organic matter should then have been destroyed. Allow the solution to cool, add 5 ml of distilled water and again evaporate until fumes appear. Cool the solution, add 25 ml of distilled water, boil for 5 minutes and then neutralise by adding 2.5 N sodium hydroxide. Dilute the solution to 40 ml and continue as described under “Procedure,” beginning at the addition of 4.0 ml of the ammonium molybdate - sulphuric acid solution. Carry out a blank test on the reagents. DETERMIXATION OF PHOSPHATE IN SALT DEPOSITS Take a suitable weight, not exceeding 5 g, of the sample and place it in a 250-ml beaker. Add 15 ml of distilled water and carefully add 7 ml of sulphuric acid, sp.gr. 1.84. Evaporate the solution on a sand-bath until fumes of sulphur trioxide appear. Cool the solution, add 25 ml of distilled water, boil for 5 minutes and then neutralise by adding 2-5 N sodium hydroxide, Dilute the solution to 40 ml and continue as described under “Procedure,” commencing with the addition of 4.0 ml of the ammonium molybdate - sulphuric acid solution. REFEREWES Measure a suitable volume, e g . , 50 ml, of the sample into a 250-ml beaker. Carry out a blank test on the reagents. 1. 2. 3. 4. 5. 6. 7. Fogg, D. N., and lvilkinson, N. T., Analyst, 1956, 81, 525. DenigBs, G., Cornpt. Rend., 1920, 171, 802. Holman, W. M., and Pollard, A. G., J . SOC. C h w z . Ind., 1937, 56, 3391. Ammon, R., and Hinsberg, K., 2. physiol. Chern., 1936, 239, 207. Lowry, 0. H., Roberts, N. R., Leiner, K. Y., Wu, M. L., and Farr, A. L., J . Biol. Chem., 1954, Chen, P. S., Toribara, T. Y., and Warner, H., Anal. Chem., 1956, 28, 1756. Fiske, C. H., and Subbarow, Y., J . Biol. C h e m , 1925, 66, 375. 207, 1. Received December 23vd, 1957