January, 19583 NOTES 53 Notes THE 1)ETERMINATION OF MANGANESE IN TITANIUM THE persulphate oxidation - sodium arsenite titration method is recommended1 for determining manganese in titanium. This is a well established method used in steel analysis, phosphoric acid being added to extend the range of the method and to improve the end-point. Phosphate cannot be tolerated when a large amount of titanium is present, because a titanium phosphate is pre- cipitated under the conditions of the method. The behaviour of anions, other than orthophosphate, was therefore investigated. It was shown that fluoride could be used in place of phosphate. EXPERIMENTAL The method recommended by the Panel on Methods of Analysis,l which is based on oxidation of a solution of the sample in sulphuric acid with ammonium persulphate in the presence of silver nitrate, was investigated. It was found that, if more than 10 mg of manganese are present, manganese dioxide is pre- cipitated when the ammonium persulphate is added.The end-point of the titration becomes difficult t o determine as the amount of titanium present is increased. The method is empirical, and the effective normality of the sodium arsenite and the colour of the end-point depend upon the amount of titanium present and the manganese concentration. Sodium fluoride was added after the metal had dissolved and the procedure was continued as described in the standard method. A visual improvement was noted in the colour change, which was from pink to yellow-green, rather than pink to yellow-brown. When no titanium is present, the presence of fluoride causes the end-point to be from pink to colourless.The greenish shade appears when titanium is present (1 g) and the yellow becomes more visible as the man- ganese content increases. POTENTIOMETRIC TITRATIONS- The end-points obtained with and without sodium fluoride were compared by potentiometric titration. One gram of titanium was dissolved as in the Panel Method1 and various amounts of manganese were added as manganese sulphate. The manganese was oxidised with 15 ml of 25 per cent. ammonium persulphate solution in the presence of 10 ml of 0-8 per cent. silver nitrate solution, The solutions with and without sodium fluoride present were titrated with sodium arsenite. Platinum and calomel electrodes were used.Because of the silver ions present in the solution, ;t. sodium sulphate salt bridge was used between the calomel and the solution. The break is sharper when fluoride is present, and a t the end-point without fluoride present there is a drift that is not detected in the presence of fluoride. RANGE OF METHOD- When fluoride is present, up to 25 mg of manganese can be oxidised without any precipitation of manganese, provided the conditions are controlled correctly. It was found that, if the ammon- ium persulphate was added at the boiling-point and the solution was then boiled for 1 minute, no precipitation occurred, and reproducible end-points were obtained. The results obtained on titrat- ing various amounts of manganese in the presence of 0.5 g of titanium with sodium arsenite solution (3.85 g per litre) after the addition of 5 g of sodium fluoride, and then silver nitrate and ammonium persulphate in the usual manner, were as follows- Manganese added, mg .. . . 3.48 6.96 13-92 17-40 20-88 85-00 Volume of sodium arsenite solution used, ml (average of five results) . . . . 4.10 8.18 15.35 19-20 23-25 27-50 Standard deviation . . . . 0.01 0.02 0.03 0-03 0.04 0.05 The standard deviation increases with the amount of manganese present. The method is empirical, the effective normality of the sodium arsenite depending upon the amount of manganese present, TITANIUM CONTENT- Experiments were performed in which the titanium content was varied, but a constant amount of manganese was present; the amount of manganese was 5.02 mg in each solution and the results on titrating with sodium arsenite solution (3.85 g per litre) were- Titanium added, mg .. .. .. 100 500 1000 Volume of sodium arsenite solution used, ml . . .. .. . . 6.22 5-96 5-6054 NOTES [Vol. 83 It can be seen from the results given that the sodium arsenite solution must be standardised under conditions such that the manganese and titanium concentrations are similar to those in the sample. RESULTS Table I shows the results obtained in the analysis of titanium - manganese alloys, prepared The method as described by the by arc-melting sintered titanium powder and manganese metal. Panel1 p h s the addition of 5 g of sodium fluoride, was used. TABLE. I ANALYSIS OF TITANIUM .. MANGANESE ALLOYS composition : Sample five results for Standard Range of 2.00 0.5 2-02 0.002 0.01 4-00 0-5 4-03 0-006 0.02 6.00 0.25 6-04 0.01 0.04 8.00 0.25 8.08 0.02 0.04 Nominal Average of manganese, % weight, g manganese, ”/o deviation results REFERENCE of Analysis,” Information Bulletin No.T 8, Part I, Ordnance Corps, U.S.A., 1955. 1. Metallurgical Advisory Committee on Titanium, “Recommended Methods of the Panel on Methods COMMONWEALTH SCIENTIFIC AND INDUSTRIAL RESEARCH ORGANISATION PHYSICAL METALLURGY SECTION BAILLIEU LABORATORY J. A. CORBETT UNIVERSITY OF MELBOURNE Received M a d 19th, 1957 THE APPLICATION OF THE POLARISED PLATINUM ELECTRODE TO THE DETERMINATION OF ASCOREIIC ACID IN FRUIT PRODUCTS (Presented at the meeting of the Western Section on Saturday, November loth, 1966) THE “dead-stop” method of Foulk and Bawdenl, in which polarised platinum electrodes are used, as adapted by Liebmann and Ayres2 for the determination of ascorbic acid by titration with 2 : 6-dichlorophenolindophenol, has been used in these laboratories for many years.The end- point is determined graphically from plots of galvanometer deflections read 25 seconds after each 042-ml addition of dye solution. The additions of the dye solution are made at 30-second intervals. It was found that for low concentrations of ascorbic acid the slope of the plots after the end- point was reduced and so a loss of precision resulted. In addition, the galvanometer deflection, on occasions, would be still altering when the time came for its value to be noted. The effect was especially noticeable when the frequency of the additions of dye solution was increased in an effort to improve the precision.The modifications described below were the outcome of attempts to improve the precision of the method. The method described by Sully3 was used, in which the e.m.f. is applied to the electrodes through a very high resistance of 2 to 3 megohms, and the potential across them is measured by means of an electronic millivoltmeter. When the system is polarised, a high e.m.f. is observed, but it is reduced when depolarisation takes place. It was found that the system, when partly depolarised by a small temporary excess of oxidant, became progressively slower to recover as the end-point was approached. By reducing the size of the electrodes to that of pin heads, by cutting them off flush with the glass surface, the sluggish response was largely overcome.This effect was presumed to be due to the smaller surface area of the electrodes, which would require less reductant to achieve polarisation. The method finally adopted made use of electrodes of small surface area as described above. A Mullard millivoltmeter as supplied with a well known electrometric titration outfit was used for measuring the e.m.f. The circuit of the millivoltmeter was modified so that the existing polarising voltage could be applied through a 2-2-megohm resistor and the voltage was reduced to approxi- mately 260 mV by means of a suitable tapping off the bias resistor network of the first valve in the instrument. Two millilitres of extract were taken and mixing was carried out by means of a stream of oxygen-free nitrogen.The dye solution was added in 0-01-ml increments at 20-second in1:ervals near the end-point. The e.m.f. was The titration was performed in a small flat-bottomed tube.January, 19581 NOTES 55 measured 15 seconds after each addition of dye solution. The concentration of the dye was approximately equivalent to 0.4 mg of ascorbic acid per ml and was standardised against 0.005 N potassium iodate. RESULTS AND DISCUSSION The coefficient of variation for quadruplicate determinations on a solution of blackcurrant syrup was 0.6 per cent., compared with 1-4 per cent. by the original method. A similar test on a solution of ascorbic acid gave 0.33 and 1.0 per cent. It is thought that the improvement may be due to the reduction of the current through the system at the end of the titration from 0.4 to 0.05 PA.Three assays on pure ascorbic acid gave 100.0, 99.7 and 100.3 per cent., which indicate good accuracy. Evans and Simmonds4 have discussed the mechanism of polarisation end-points with special reference to dilute solutions of iodine and sodium thiosulphate. They state that under certain conditions there may be insufficient anions to keep the cathode depolarised. This effect was shown by a very slow fall in e.m.f. when dye solution was added to 2 per cent. oxalic acid solution in the original method, When the modified circuit was used, a sharp fall in e.m.f. was observed. These observations support the view that, the smaller the size of the electrode, the lower the concentration of anions necessary to depolarise it.In order to test the effect of anions on the results obtained by the method, a solution of black- currant syrup and a dilute solution of ascorbic acid were both assayed with and without the addition of 2 mg of chloride as potassium chloride. TABLE I The modifications described resulted in an increase of precision. The results are shown in Table I. EFFECT OF CHLORIDE ON THE DETERMINATION OF ASCORBIC ACID Ascorbic acid found in 2 ml of sample with 2 mg of chloride added, Ascorbic acid found in 2 ml of sample, Sample mg mg Ascorbic acid solution . . .. 0.042 (8), 0.042 0.042, 0-041 (6) Blackcurrant syrup solution A . . 0.392, 0-387 0.382, 0.383 Blackcurrant syrup solution B . . 0.426, 0.430 0.430, 0.429 In all tests the differences were within the experimental error of the method.The modified method has, therefore, the advantage of increased precision with no loss in accuracy, and it is not affected by concentrations of chloride up to 1 mg per ml of titrating solution. REFERENCES 1. 2. 3. 4. H. W. CARTER & Co. LTD. Foulk, C. W., and Bawden, A. T., J. Amev. Chem. Soc., 1926,48, 2044. Liebmann, H., and Ayres, A. D., Analyst, 1945, 70, 411. Sully, B. D., Chem. 6. Ind., 1955, 1146. Evans, D. P., and Simmonds, N. T., J. SOC. Chem. Ind., 1944,63, 29. THE ANALYTICAL LABORATORY THE ROYAL FOREST FACTORY R. C. CURTIS COLEFORD, GLOS. Received December 1 Ith, 1966 A NOTE ON SILVER NITRATE TITRATIONS WHEN solutions of sodium or potassium chloride are titrated with silver nitrate, with dichloro- fluorescein as indicator, a more accurate and clear-cut end-point is obtained by observing the character of the precipitate, At the end-point the precipitate breaks down and becomes finely divided and granular in appearance. The colour changes in the supernatant liquid and of the precipitate are useful associated indications of the end-point, but are difficult to describe by them- selves. With 0.1 M solutions, this end-point gives reproducible results in agreement with those obtained by the Mohr method corrected for a blank. No comparable reproducible change in the character of the precipitate occurs in titrating solutions of sodium or potassium bromide, iodide or thiocyanate with silver nitrate. DEPARTMENT OF CHEMISTRY THE QUEEN’S UNIVERSITY OF BELFAST N. IRELAND W. F. N. LEITCH F. A. LEWIS Received August 2nd, 1957