802 Afialyst, December, 1968, Vol. 93, $9. 802-804 The Determination of Zirconium in Mineral Rutile with Alizarin Red S BY N. B. STANTON ( A ssociated Minerals Consolidated Limited, Southport, Queensland, A ustralia) A colorimetric procedure has been developed for the determination of the zirconium oxide content of rutile. A fusion with potassium hydrogen difluoride engbles the rutile to be brought into solution rapidly and eliminates the task of filtering off silica. The zirconium is separated from the titanium by a phosphate precipitation, but the interference of tin inherent in the method involving phosphate precipitation followed by ignition to oxide, is eliminated by dissolving the zirconium phosphate and applying a colorimetric finish, with Alizarin red S. Calibration is effected by making additions of zircon to titanium oxide and carrying each through the entire procedure, thus eliminating reagent-blank difficulties. It has been used in the range of 0 to 1 per cent.of zirconium oxide. Good agreement with independent determinations by X-ray fluorescence has been obtained. THE main constituents of the mineral sands of Australian east coast beaches are zircon, rutile and ilmenite. The guarantees under which these products are sold include the maximum amounts of each present as impurities in the others. A need, therefore, exists for a method for determining the zirconium oxide content of rutile. From this content the percentage of zircon, which is accepted as the sole contributor to the zirconium oxide content, can be inferred.The guarantee at the present time requires a maximum of 1 per cent. of zirconium oxide. Methods previously used for this purpose were variants of two main procedures. The first was the precipitation of zirconium with diammonium hydrogen orthophosphate. As pointed out by Wood and McKenna,l tin interferes in this method. Rutile commonly contains about 0.1 per cent. of tin oxide, and it has been shown in this laboratory that tin oxide added to rutile can be recovered by phosphate precipitation. A cassiterite concentrate that had previously been analysed for tin oxide was used for additions to a rutile sample, which was then fused with potassium hydrogen difluoride and subjected to a phosphate precipitation technique to be described, followed by ignition of the phosphate precipitate. The weight of contained tin oxide added was 6.2 mg.The difference in weights between the oxide obtained from the rutile and from the “spiked” rutile was 5.2 mg. The method is, therefore, not satisfactory for this purpose. The second method involves precipitation with mandelic acid. This approach is tedious and time consuming and, further, is not suited to the determination of amounts at the lower end of the range encountered, e.g., 0.3 per cent. of zirconium oxide. It appears that a certain amount must be present before the precipitate can be collected. Mayer and Bradshaw2 first put forward the use of Alizarin red S as a reagent for the determination of zirconium, with reference to magnesium alloys. Snell3 states that titanium interferes in the determination of zirconium with Alizarin red S, and Wood and McKennal correct for a small amount of interference by their method of calibration.An attempt by the author to determine zirconium in rutile directly was unsuccessful because of the difficulty of controlling the hydrochloric acid concentration of the dissolved melt, and the added disadvantage of the presence of sulphate ions. The approach adopted was that suggested by Vinogradov and Ryabchikov.* The titanium was first separated by precipitation of the zirconium with phosphate, the precipitate re-dissolved and a colorimetric finish applied. However, the technique used by these authors for re-dissolving the precipitate on the paper with 5 per cent. oxalic acid was not successful under the conditions experienced, and this approach was modified.The method used for colour formation was that recommended by Wood and McKennal and the U.S. Atomic Energy Commission.6 0 SAC and the author.STANTON 803 EXPERIMENTAL Calibration was carried out by weighing about 1-5,4*5, 7-5,lO and 15 mg of finely ground zircon consecutively into five platinum crucibles, each containing 6 g of potassium hydrogen difluoride and 1 g of Johnson, Matthey “Specpure” titanium dioxide. These additions correspond to 0.10, 0.30, 0.50, 0.675 and 1.0 per cent. of zirconium dioxide, respectively. Each was then taken through the full procedure to be described. This approach was adopted for two reasons, Firstly, Wood and McKennal did not use zirconium nitrate solution because the reagent contains more than the theoretical amount of zirconium, resulting from gradual decomposition of the zirconium nitrate; instead, they used metallic zirconium.It was, however, possible to obtain a pure grade of zircon of known zirconium content for this work. Secondly, the problem of adjusting for the reagent blank is overcome as it is automatically incorporated in the calibration. Any approach in which an attempt is made to subtract a blank reading, obtained by taking the reagents alone through the procedure, from an observed reading is not considered valid. Anomalous enhancement and depression effects on colour development can occur. Wood and McKennal refer to the enhancement effect of titanium on zirconium solutions, as compared with the effect on the reagent blank.The author has made a comparison in some instances of the colour development obtained by adding a solution of the element, and the colour reagent, to distilled water with that obtained by taking the element through the entire procedure and subtracting a reagent blank. An enhancement effect occurs with the former. For these reasons it is desirable to simulate the actual method in its entirety when carrying out the calibration. COLOUR DEVELOPMENT- Wood and McKenna heated the solution for 5 minutes a t 70” to 80°C. Vinogradov and Ryabchikov advocate bringing it to the boil. It was found that neither approach, as applied to our problem, guaranteed full colour development, and prolonged boiling, as recommended by the U.S. Atomic Energy Commission,5 was adopted successfully.Double filtration of the Alizarin red S solution, as recommended by Wood and McKenna, is considered essential. REAGENTS- Alizarin red S solution, 0-1 per cent.l-Dissolve 1.0 g of Alizarin red S in about 300 ml of hot water, boil and then filter the solution through a pad of paper pulp. Dilute the solution to 1 litre and again filter through a pad of paper pulp. PROCEDURE- Fuse 1 g of previously ground sample with 6 g of potassium hydrogen difluoride in a platinum crucible. Cool the melt, transfer it to a platinum basin and add 60 ml of sulphuric acid (1 + 1). Heat gently until copious fumes of sulphuric acid are evolved. This removes the silica, and all of the titanium and zirconium is brought into solution. Cool the solution, dilute to 450 ml, and transfer it to a 600-ml beaker.Add 25 ml of 30 per cent. hydrogen peroxide and 25ml of 20 per cent. dibasic ammonium hydrogen orthophosphate solution; allow to stand overnight in a warm place. Filter the solution through a Whatman No. 31 paper and wash with 5 per cent. sulphuric acid, to which a little hydrogen peroxide has been added, until no colour remains on the paper. Continue washing (6 or 7 times) with 5 per cent. ammonium nitrate solution. Wash the precipitate from the paper into a 250-ml beaker with hot water, then add 10 ml of 10 per cent. sodium hydroxide solution. Bring the mixture to the boil and allow the precipitate to coagulate for 20 minutes in a steam-bath. Filter on a Whatman No. 41 filter-paper and wash the precipitate with 5 per cent. ammonium nitrate solution.Wash the precipitate from the paper into the same 250-ml beaker with hot distilled water and add 25 ml of 5 N hydrochloric acid; the volume should now be about 100 ml. Boil until the volume is reduced to 40 ml, cool, make up to 100 ml and filter through a Whatman No. 41 filter-paper. To a 10-ml aliquot of the filtrate, add 2.5 ml of 5 N hydrochloric acid. Add 5 ml of 0.1 per cent. Alizarin red S and make the solution up to 50ml. After allowing it to stand for 1 hour, measure the absorption a t 560nm. The instrument used was a Bausch and Lomb Spec- tronic 20. METHOD804 STANTON COMPARISON WITH X-RAY FLUORESCENCE DETERMINATION- The accuracy of the method has been demonstrated by comparison of the results with those obtained by X-ray fluorescence determination, in which the method of additions, as described by Birks,6 was used, background correction being allowed for.Results are shown in Table I. TABLE I DETERMINATION OF ZIRCONIUM OXIDE IN RUTILE BY CHEMICAL AND X-RAY FLUORESCENCE METHODS Zirconium oxide (ZrO,), per cent., by- Sample No. Chimica1 method X-ray fluorescence mithod 1 0.27 0-28 2 0-36 0-37 3 0.38 0-40 4 0-46 0.43 5 0.5 1 0.50 6 1.10 1.06 CONCLUSION With this method a single determination can be carried out in 24 hours. It requires a slightly longer time than the phosphate precipitation method, but eliminates the error caused by tin oxide inherent in the latter. The calibration method initially is time consuming, but no further calibration is required for a given set of reagents. The assistance of Miss Rose Thomas in carrying out some of the chemical determinations is acknowledged with thanks. Acknowledgment is also accorded to the Directors and Management of Associated Minerals Consolidated Limited for permission to publish this paper. REFERENCES 1. 2. 3. 4. 5. 6. Wood, D. F., and McKenna, R. H., Analyst, 1962, 87, 880. Mayer, A., and Bradshaw, G., Ibid., 1952, 77, 476. Snell, F. D., and Snell, C. T., “Colorimetric Methods of Analysis,” Volume IV, Organic, 11, Third Edition, D. Van Nostrand Co. Inc., New York; Macmillan & Co. Ltd., London, 1954, p. 447. Vinogradov, A. P., and Ryabchikov, D. I., Edztors, “Detection and Analysis of Rare Elements,” National Science Foundation, Washington, 1962, p. 366. “Manual of Special Materials Analytical Laboratory Procedures,” U. S. Atomic Energy Commission Report ANL-5410, 1955, p. 49. Birks, L. S., “X-Ray Spectrochemical Analysis, ” Interscience Publishers Inc., New York, 1959, p. 67. Received March 25th, 1965 Amended J d y loth, 1967