Nov., 19511 HARRIS AND LINDSEY 647 Vibrating Electrodes in Amperometric Titrations Part I The Determination of Thiosulphate, Iodine and Tervalent Arsenic J ~ Y E. D. HARRIS AND A. J. LINDSEY (Preseutcd at the meeting of the Physical Methods Grot@ on Tuesday, April loth, 1951) The use of vibrating electrodes is extended to amperometric titrations. The apparatus consists of a simple polarising unit with a galvanometer to measure the diffusion currents and a vibrating micro-electrode in conjunction with a saturated calomel electrode. Amperometric titrations of 0.001 N solutions of iodine, sodium thiosulphate and tervalent arsenic give results that, in suitable supporting electrolytes, are comparable in accuracy with titrations in which starch is used as indicator. IK earlier publications’ j 2 the behaviour of vibrating platinum micro-electrodes was described and essential conditions for their use in polarography were established. These electrodes can be used as substitutes for the dropping-mercury electrode, which cannot be used for anodic reactions occurring at potentials more positive than +0.4 volt with reference to the saturated calomel electrode because mercury is attacked in such circumstances.Platinum electrodes, however, can be used up to a potential of +1.1 volts without attack3 and they have been used in this potential range by a number of e x p e r i r n e n t e r ~ . * ~ ~ ~ ~ , ~ In addition they can be used for electrode reactions where the presence of mercury is undesirable, such as the polarographic reduction of alkyl peroxides.’ Permanent polarisation and electrode “poisoning,” which are disadvantages of the platinum micro-electrode in quantitative polarography, are less troublesome in amperometric titrations where the electrode is used as an indicator for an end-point.If no current flows until after the end-point is reached, this difficulty does not occur. As the vibrating electrode is insensitive to external vibration, it has considerable advantage over a stationary electrode or a dropping-mercury electrode. Another advantage is that the vibrating electrode gives a much greater diffusion current than a stationary electrode of the same area. The use of vibrating electrodes does not differ greatly from the well-established techniques for mercury electrodes or rotating platinum micro-electrodes. APPARATUS- The electrodes for this work were constructed as described else~here2,~ and were driven by means of an electromagnetic massage vibrator from the alternating-current mains.They were made of platinum wire sealed into glass tubing of 4 mm diameter, and were 0.5 mm in diameter and 2 mm long. The electrodes were vibrated at constant amplitude greater than 3 mm, which conforms to the conditions recommended elsewhere.2 The electrical circuit is shown in Fig. 1, and is similar to that previously described. This simple polarising circuit can be replaced by the unit described by Stock.8 PROCEDURE- substance being titrated. a change in the direction of the diffusion current - titre curve. was best found by plotting current readings against those of the burette. The polarising voltage was set to correspond to the limiting diffusion current of the The titrating solution was then run in, with stirring, until there was The location of the end-point It is advisable648 HARRIS AND LINDSEY: VIBRATIKG ELECTRODES IN [Vol.76 always to plot a current - voltage curve as a preliminary to any new type of determination in order to determine the characteristics of the electrode process under the conditions of the experiment. Although the diffusion current changes with temperature and thermostatic control is necessary for quantitative polarography, there is seldom any need to control or n il Fig. 1. Circuit for polarising vibrating electrodes A 2-volt accumulator. and two potentiometers provide a variable voltage that is measured by the voltmeter, V, and applied to the cell, which consists of a calomel half-element, a salt bridge and the vibrating electrode.A galvanometer, G, in series with the cell measures the current measure the temperature in amperometric titrations. unnecessary in the titrations described in this paper. Oxygen removal was shown to be DETERMINATION OF IODINE IN ACID SOLUTION A study of the current - voltage curve of 0.0001 N iodine in N hydrochloric acid con- taining 0.004 N potassium iodide as supporting electrolyte showed that a limiting diffusion current due to reduction of iodine is attained at from +0-05 to +040 volt with respect to the saturated calomel electrode, The vibrating electrode was therefore held at this potential and various amounts of iodine were added to fiOml of supporting electrolyte.In Fig. 2 is shown the diffusion current plotted against the concentration of iodine after allowing for volume changes. The range over which diffusion current was directly proportional to iodine concentration was from 2.5 x N, but if this part of the curve is extrapolated to zero current it does not pass through zero volume. The error is small, however, (equivalent to 0.3 per cent. of the full titre) and may be neglected for many analytical purposes. If more accurate results are required the figure may be used as a calibration curve. At the chosen potential the products of a sodium thiosulphate - iodine titration gave no diffusion current when added to the electrolyte even at a concentration corresponding to 0.01 N.As this titration is carried out at zero potential the electrical circuit can be modified considerably and the galvanometer only is needed in addition to the electrolytic cell. Fig. 3, which is representative of manj. experiments, shows the relationship between current and burette readings obtained when 100 ml of 0.001 Ai iodine, N with respect to hvdrochloric acid and 0.004N with respect to potassium iodide, were titrated with 0.01 N sodium thiosulphate solution from a micro-burette. l h e micro-electrode was held at the potential of the saturated-calomel reference electrode and current readings were taken over the whole titration, although they are only necessary during the last part. The end-point was taken as the extrapolation of the line to zero current, and corresponds to 10.22 ml of thiosulphate solution.Check titrations on the iodine and thiosulphate with starch as indicator gave an end-point of 10.24 ml, after correcting for an indicator blank. DETERMISATIONS IN NEUTRAL SOLUTIONS When 0.1 N potassium chloride and 0.004 N potassium iodide was used as supporting electrolyte, the current - voltage curve with iodine at a concentration of 0-0001 N was similar to that obtained in acid solution, so that a titration with 0.01 N iodine of 100 ml of approxi- mately 0.001 N sodium thiosulphate in this supporting electrolyte can be carried out under the same electrical conditions. In these circumstances no current is observed until the end- point is reached, after which the diffusion current is proportional to the iodine concentration over the same range as previously found.To obtain a rcproducible end-point it is recom- mended that a number of current readings be taken and the plotted values be extrapolated to 7.5 xNov., 19511 AMPEROMETRIC TITRATIONS. PART I 649 back to zero current. In a series of determinations with 10ml of iodine solution the end-points differed by 0.01 ml from those in which starch indicator was used. DETERMINATIONS IN SUPPORTING ELECTROLYTE CONTAINING SODIUM BICARBONATE- By the same procedure, 100 ml of approximately 0.001 N sodium thiosulphate, 0.1 N with respect to potassium chloride and sodium bicarbonate and 0.004 N with respect to potassium iodide, were titrated. A diffusion current was obtained for the end-point and its . . Iodine.ml Iodine. ml Fig. 2. Diffusion current - concentration curve Fig. 3. Diffusion currents measured when 100 ml of 0.001 N iodine in supporting electrolyte is titrated with 0.01 N sodium thiosulphate for iodine in a supporting electrolyte of N hydro- chloric acid and 0.004 N potassium iodide magnitude was proportional to the amount of iodine added for the range 2.5 x lW5 to 7.5 x In a series of determinations with 10 ml of iodine the end-points differed by 0.02 ml from those in which starch indicator was used. N . DETERMINATION OF TERVALENT ARSESIC BY IODIYE- By the same procedure, 100 ml of 0.001 N sodium arsenite, 0.1 N with respect to both potassium chloride and sodium bicarbonate and 0.004 N with respect to potassium iodide, were titrated with similar results.In a series of determinations the end-points differed by 0-03 ml from those in which starch indicator was used. These experiments show that the vibrating micro-electrode is satisfactory as an indicator electrode in amperometric titrations with iodine as the electro-active material and that results are within 0.3 per cent. of the values obtained when starch is used as an indicator. The authors wish to state that the experimental work was approved as part of a thesis for the degree of Master of Science in the University of London.9 They also wish to record their thanks to Imperial Chemical Industries Ltd., for a grant used for the purchase of chemicals. REFERENCES 1 . Harris, E. D., and Lindsey, A. J., Natzcve, 1948, 162, 413. 2. Lindsey, A. J., J . Phys. Coll. Cheni.. in the press. 3. Morris, C. J . 0. I<., Analyst, 1947, 72, 298. 4. Randle, J. E. B., Ibid., 1947, 72. 301. 5. Airey, L., Ibid., 1947, 72, 304. 6. Miller, S. D., Trans. All-Union Conf. Anal. Chem., 1943, 2, 551. 7. Roberts, E. R.. and Meek, J. S., Analyst, in the press. 8. Stock, J. T.. Ibid., 1946, 71, 585. 9. Harris, E. D., M.Sc. Thesis, University of London, 1949. SIR JOHN CASS COLLEGE LONDON, E.C.3