Analyst, June, 1968, Vol. 93, pp. 383-387 383 The Spectrophotometric Determination of Titanium in Iron and Steel with Diantipyrylmethane BY J. A. CORBETT (Physical Metallurgy Section, Commonwealth Scientific 6. Industrial Research Organization, Baillieu Labmatovy, University of Melbourne, Australia) The use of diantipyrylmethane its a spectrophotometric reagent for the determination of titanium in iron and steel has been investigated. A rapid method is suggested in which the reagent, which permits the determination of titanium in a wide range of iron and steels, is used. Results are given for the analysis of stainless, high speed and mild steels, and cast irons. THE British Standards Institution has recently issued B.S. 1121 : Part 47 : 1966, which gives a method for the spectrophotometric determination of titanium in iron and steel, in which hydrogen peroxide is used to form a coloured complex.This method is similar to the American Society for Testing Materials method and, while quite satisfactory, is slow and tedious, particularly when molybdenum and vanadium are present. Several atomic- absorption methods1 s2 have been suggested but they have limited sensitivity. In seeking a more rapid method for general application, a study was made of various reagents, and diantipyrylmethane appeared to offer many advantages. The optimum conditions given in the literature under which diantipyrylmethane is used vary considerably. Polyak3 states that the optimum acid concentration, with respect to hydrochloric acid, lies within the range 1 to 4 M, and gives a development time of 40 to 50 minutes before measuring the absorbance. Later, Polyak4 introduced tartaric acid to keep niobium in solution when determining titanium in the presence of niobium, and allowed a development time of 2 hours.Jeffery and Greg~ry,~ when using 1 M hydrochloric acid, allow 3 hours for development. They also use tartaric acid when niobium is present and again develop for 3 hours. Lazareva and Lazarev6 conclude that the rate of formation of the coloured complex decreases as the hydrochloric acid concentration increases, but the complex formed is always the same. They consider that the optimum concentration, with respect to hydrochloric acid, is 0.25 M. They further show that when sufficient tartaric acid is present, a different complex is formed, with an absorption maximum at 326 to 330 nm, instead of 380 to 400 nm without tartaric acid.In methods of analysis given by the above workers, sulphuric acid is used in the prepara- tion of sample and standard solutions. The concentrations of sulphuric acid vary considerably and in some instances the amounts used are not specified at all. Two reactions are involved: first, the breakdown of titanium polyions to TiO+, and then the formation of the titanium oxide (TiO) - diantipyrylmethane complex. The de-poly- mensation rate of the titanium ions is slow and can be expected to be affected by the acid concentration and the presence of complexing agents, such as tartaric acid. The large excess of diantipyryImethane recommended (300 times the titanium concentration) suggests a low equilibrium constant for the titanium oxide - diantipyrylmethane reaction.The effects of the presence of other cations on the reaction have been studied by Pol~ak.~s* Nickel, chromium, copper, manganese and cobalt (11) do not react with diantipyrylmethane. Niobium, if present in high concentrations, suppresses the titanium - diantipyrylmethane reaction. Iron and vanadium do not interfere if reduced to their bi- and quadrivalent forms. Tungsten, in amounts exceeding 10 mg per 100 ml, precipitates with diantipyrylmethane, 0 SAC; Crown Copyright Reserved384 CORBETT : SPECTROPHOTOMETRIC DETERMINATION OF [ArtdJU!, VOl. 93 but smaller amounts do not interfere. Molybdenum, if reduced to the quinquivalent form, does not precipitate, but causes an increase in absorbance necessitating the addition of molybdenum to the calibration standards.’ Ascorbic acid is added to reduce iron, vanadium and molybdenum; the yellow titanium - ascorbic acid complex is destroyed by adding hydrochloric acid.3 EXPERIMENTAL Diantipyrylmethane (obtainable from Aldrich Chemical Co., Milwaukee, U.S.A.) was dissolved in various aqueous solutions containing acid; a molar solution of hydrochloric acid was found to be suitable.A stock titanium solution was prepared by dissolving high purity titanium metal in dilute sulphuric acid (1 + 3) and oxidising it with the minimum amount of dilute nitric acid (1 + 1). This solution was made up to volume so that the final sulphuric acid concen- tration was 2 M. To control the acid concentration, and also to avoid hydrolysis, dilutions were made of the stock solution with water immediately before use and the solutions were afterwards discarded.Solutions of diverse ions were prepared from high purity metals or analytical-reagent grade chemicals. RESULTS Tests were carried out with the conditions and reagents as recommended by other workers, viz., with 0 to 100 pg of titanium per 100 ml, 1 M hydrochloric acid, 0.006 M dianti- pyrylmethane and varying concentrations of sulphuric and tartaric acids. The following results were obtained. Beer’s law was not obeyed with amounts of titanium above 50 pg, suggesting that even a 300-fold excess of the reagent was not sufficient. Varying concentrations of sulphuric, hydrochloric and tartaric acids all had an influence on the development time before colour measurement and on the final absorbance.The wavelength - absorption plot was the same in all of the tests. The departure from Beer’s law was not serious, but, to achieve a reasonable precision, the variations in development time and absorbance would make it necessary to ensure that the calibration standards were treated in exactly the same manner as the test solution. Tests were carried out in which both the titanium and diantipyrylmethane concentrations were increased with the aim of increasing the breakdown rate of the titanium polyions. This was effected by reducing the final volume from 100 to 50 ml. To conserve the diantipyryl- methane, the excess was reduced to about 250-fold at the 1OO-pg titanium level.The results obtained, as shown in Fig. 1, indicate a considerable improvement in tolerance to variations in acid concentration and the presence of tartaric acid. Concentrations of hydrochloric acid up to 1 M give the same absorbance after 1 hour. Sulphuric acid in the concentrations tested, while not affecting the development time, caused a lower absorption. As the concentration of tartaric acid increases, the development time increases until, at 2.6 M, equilibrium has not been reached after 5 hours, but lower concentrations (0.01 M) can be accommodated. An improvement was also obtained in the plot for Beer’s law, although with amounts of titanium above 100pg Beer’s law was not obeyed. The addition of sodium or potassium sulphate in amounts up to 0.5 g per 50 ml in 0.5 M hydrochloric acid has no effect. Apparently, it is the hydrogen-ion concentration that causes the lower absorbance in the presence of sulphuric acid.Evidence of the formation of a titanium oxide - diantipyrylmethane - tartaric acid com- plex was obtained by using the conditions and reagents as described by Lazareva and Lazarev (viz., 0.25 M hydrochloric acid, 0.05 M diantipyrylmethane and 0.3 to 0.5 M tartaric acid). An absorption maximum was obtained at 326 to 330nm, but the complex slowly changed with time to the titanium oxide - diantipyrylmethane complex, with an absorption maximum at 380 to 400nm. The addition of ascorbic acid has no effect on the development time - absorbance curves shown in Fig. 1. No difficulty was experienced as a result of the formation of a titanium - ascorbic acid complex, probably because of the high acid concentration present when the ascorbic acid is added; the aliquot is later diluted with water to give the required acid concentration.June, 19681 TITANIUM IN IRON AND STEEL WITH DIANTIPYRYLMETHANE 385 The presence of 100mg of iron, 20mg of nickel, 20mg of chromium, 5 mg of copper, 5 mg of vanadium, 2 mg of manganese and 2 mg of tungsten ions per 50 ml of solution has no effect when ascorbic acid is present, and the absorption measurements are made with a reference solution containing the same amounts of these cations and ascorbic acid.Molybdenum, after reduction with ascorbic acid, was found to form a weak yellow complex with diantipyrylmethane (1 mg of molybdenum is equivalent to 1-5 pg of titanium).This interference was prevented by forming a complex of the molybdenum with tartaric acid. Tests showed that tartaric acid at a concentration of 0.013 M overcame interference from amounts of molybdenum up to 1Omg. Tungsten precipitates with diantipyrylmethane under the conditions used in these tests when more than 2 mg is present in the 50-ml aliquot. Most of the tungsten in steels can be rendered insoluble by treatment with a suitable acid, the amount remaining in solution in the final aliquot being less than 2mg. 0.3 E i f a . .. . . .. J L I I I I I Niobium is retained in solution by forming a complex with tartaric acid. However, it is preferable to restrict the amount of tartaric acid used, as shown in Fig. 1. As a result of these tests, it appeared possible to adopt a set of conditions that would permit the determination of titanium in a wide range of iron and steels. The reagents and conditions adopted were: 0.013 M tartaric acid, 0.01 M diantipyryl- methane, 0.5 M hydrochloric acid and 0.06 M ascorbic acid in a 50-ml volume, with a develop- ment time of 90 minutes.The complex is stable over the temperature range 18" to 30" C, and the extinction coefficient at 390 nm is 14,500. The standard deviation of the absorbance is 0.0025. When molybdenum and niobium are known to be absent the tartaric acid can be excluded and the development time reduced to 1 hour. The recovery of titanium from the acid-insoluble material may not be necessary with some samples.386 CORBETT : SPECTROPHOTOMETRIC DETERMINATION OF [Analyst, Vol.93 Table I shows the results obtained by using the method with standard samples of various steels. TABLE I RESULTS OBTAINED WITH THE METHOD DESCRIBED COMPARED WITH THE AVERAGE B.S. 1121 : PART 47 RESULT OBTAINED BY FIVE DIFFERENT LABORATORIES WITH THE STANDARD METHOD B.S. 1121 :Part 47 method Sample Type per cent. N.B.S. 170A B.O.H. 0.282 S.A.A. 101 High speed steel (chromium 4, molyb- 0.006 S.A.A. 106 Stainless steel (18/8) 0.81 S.A.A. 106 Austenitic iron (nickel 4, chromium 3 0.063 S.A.A. 108 Stainless steel (18/8) (niobium 1 per 0.014 S.A.A. 112 Stainless steel (18/8) (molybdenum 3 0.424 S.A.A. 114 Cast iron (vanadium 0-2 per cent.) 0.032 B.C.S. 322 Mild steel 0.043 S.A.A. 101 + 100 pg of titanium S.A.A. 112 denum 6 and tungsten 6 per cent.) and copper 6 per cent.) cent.) per cent.) S.A.A.106 + 10 mg of molybdenum - Recovery 106 pg = 100 per cent. Mean of 7 results = 0.426 per cent.; standard deviation = 0.003. (S.A.A. refers to Standards Association of Australia.) Diantipyrylmethane, method, per cent. 0.282, 0.282 0.006, 0.006 0-828, 0.824 0.060, 0.061 0.018, 0.017 0.424, 0.428 0-036, 0.037 0.046, 0-046 0.051 METHOD REAGENTS AND SOLUTIONS- Titanium sohtion-Dissolve O-lOOO g of high purity titanium metal in 50 ml of dilute sulphuric acid (1 + 3) and oxidise the titanium by adding nitric acid, dropwise. Cool, transfer the solution into a 1-litre calibrated flask, dilute to the mark with dilute sulphuric acid (1 + 9) and mix. Transfer 10 ml of this solution into a 100-ml calibrated flask, add 10 ml of tartaric acid solution (10 per cent.), make up to the mark with water and mix.This solution must be prepared freshly each day. Tartaric acid solution, 10 per cent. w/v-Dissolve 100g of tartaric acid in water and dilute to 1 litre. Ascorbic acid solution, 10 per cent. w/v-Dissolve 100 g of ascorbic acid in water and dilute to 1 litre. Diantipyrylmethane solution, 2 fier cent. w/v-Dissolve 2 g of diantipyrylmethane in dilute hydrochloric acid (1 + 9) and make up to 100 ml with dilute hydrochloric acid (1 + 9). PROCEDURE- Weigh 0.5 or 1 g of sample (Note 1) and transfer it into a 250-ml beaker. Add 20 ml of hydrochloric acid (sp.gr. 1-16 to 1*18), cover the beaker and digest until solvent action ceases, add 5ml of nitric acid (sp.gr.1-42) and evaporate the solution to dryness (Note 2). Add 20 ml of dilute hydrochloric acid (1 + 3) and heat to dissolve the iron salts. Filter the solution through an ll-cm No. 41 filter-paper and wash the paper with hot water. Remove the iron salts by washing the paper with 10 ml of dilute hydrochloric acid (1 + 1) and water (Note 3). Transfer the filter-paper into a platinum crucible, dry and ignite it at a temperature of 700" C. Cool, add several drops of dilute sulphuric acid (1 + 1) and 2 ml of hydrofluoric acid (40 per cent.), evaporate to dryness, ignite at 700" C, and cool. Fuse the residue with 1 g of potassium hydrogen sulphate and cool. Dissolve the melt by heating it with 10 ml of tartaric acid solution (10 per cent.), and add the solution to the original filtrate.Transfer two aliquots (Note 1) into 60-ml calibrated flasks, add dilute hydrochloric acid (1 + l), to make the final concentration 0.5 M with respect to hydrochloric acid (Note I), 1 ml of solution = 10 pg of titanium.June, 19681 TITANIUM IN IRON AND STEEL WITH DIANTIPYRYLMETHANE 387 and 5 ml of ascorbic acid (10 per cent.), mix, and allow the solution to stand for 10 minutes. To one flask, add 10 ml of diantipyrylmethane solution (2 per cent.), and make the solutions in both flasks up to the mark with water and mix. Allow the solutions to stand for 90 minutes and measure the absorbance of the coloured complex at 390 nm, with the blank as a reference solution, in cells of a suitable size (Note 1). Determine the blank value of the reagents concurrently with the test determination. CALIBRATION- Titanium in the range 0 to 0.07 per cent.-Transfer 0, 1.0, 2.0, 3.0, 4.0, 5-0, 6-0 and 7.0 ml of titanium solution (1 ml of solution = 10 pug of titanium) into 50-ml calibrated flasks, add 3 ml of dilute hydrochloric acid (1 + 1) and 5 ml of ascorbic acid solution (10 per cent.), mix, and allow to stand for 10 minutes.Add 10ml of diantipyrylmethane solution (2 per cent.), make up to the mark with water and allow the solutions to stand for 90 minutes. Measure the absorbance of the solutions at 390 nm in 2-cm cells, with the solution containing no titanium as a reference solution. Titanium in the range 0 to 1.20 #er cent.-Transfer 0, 2.0, 4.0, 6.0, 8-0, 10.0, 12-0, 14.0 and 150ml of the same titanium solution into 50-ml calibrated flasks and treat them as before, but measure the absorbance in l-cm cells. NOTES- 1. Appropriate weights and dilutions are listed below. Titanium range, per cent. 0 to 0.07 0 to 0.15 0 to 0-30 0 to 0.60 0 to 1.20 Weight of sample, Dilution, Aliquot, g ml ml 1.0 100 10 1.0 100 10 0.5 100 10 0.5 200 10 0.5 200 6 Hydrochloric acid (1 + 1) added, Cell size, mI cm 1.0 2 1.0 1 1.0 1 2.0 1 2.6 1 2. Additional nitric acid may have to be added to dissolve some steels. When this is necessary, evaporate the solution to dryness, dissolve the residue in 10 ml of hydrochloric acid (sp.gr. 1-16 to 1.18) and evaporate to dryness again. Continue as in Procedure. 3. If tungsten is present, wash the precipitate with ammonia solution (1 + 1). REFERENCES 1. 2. 3. 4. 6. 6. 7. Bowman, J. A., and Willis, J. B., Analyt. Chem., 1967, 39, 1210. Headridge, J. B., and Hubbard, D., Analytica Chim. Acta, 1967, 37, 161. Polyak, L. Ya., Zh. Analit. Khim., 1962, 17, 206. - , Ibid., 1964, 19, 1468. Jeffery, P. G., and Gregory, G. R. E. C., Analyst, 1965, 90, 177. Lazareva, V. I., and Lazarev, A. I., Zh. Analit. Khim., 1966, 21, 172. Polyak, L. Ya., Ibid., 1963, 18, 966. Received Junztury loth, 1968