56 Analyst, January, 1968, Vol. 93, $@. 56-58 Separation of Low Concentrations of Halogen from Some Luminescent Materials and Elemental Sulphur by a Modified Oxygen-f lask Method BY F. J. DE BOER AND J. VISSER (Philips Research Laboratories, N . V . Philips’ GloeilamfienfaArieken, Eindhoven, The Netherlands) A separation method for low concentrations of halogen from some luminescent materials and elemental sulphur is described, which is based on a modified oxygen-flask method. Relatively large samples can be burned and, in this way, halogen at the p.p.m. level can be determined by any suitable method. The combustion of sulphur can serve as a concentration method for determining other impurities. THE determination of a trace of halogen in luminescent materials and elemental sulphur is a difficult problem, especially in the p.p.m.range. The main elements present in the compound interfere with the analysis1 and it is necessary to remove them. In this paper we report on a modification of the well known oxygen-flask method,2J with which we have, for several years, obtained good results. The procedure is simple, free from contamination, and the blank is low because oxygen is the main reagent. The modification to the apparatus made it possible to burn a large amount of cadmium sulphide, zinc sulphide or selenide and elemental sulphur in the oxygen atmosphere and to determine halogen in this way at the p.p.m. level. EXPERIMENTAL APPARATUS AND METHOD- The apparatus consists of a wide-mouthed 500-ml conical flask with a ground-glass stopper through which pass two platinum electrodes of I-mm diameter, the ends of which are bent upwards.The flask is fitted with an oxygen inlet tube of 2-mm bore, with a stopcock. Fig. 1. Details of the quartz vessel The pressure of the oxygen is controlled by a bubbler device filled with mercury. A platinum wire of 500-mm length and 0.5-mm diameter is wound round the outside wall and the bottom of a cylindrical quartz vessel (30-mm diameter, 15 mm in height), as shown in Fig. 1. This wire is held in place by ten grooves in the vessel, and each end is formed into a loop. The 0 SAC and the authors.DE BOER AND VISSER 57 vessel can be suspended from the electrodes, thus making electrical contact. The length of these electrodes should be adjusted so that the vessel is situated at, or near, the centre of the flask.The current to the electrodes is supplied from a variable low-voltage transformer (20 volts, 30 amp&res), with an ammeter connected in series (see Fig. 2). P tat i n u m electrodes Fig. 2. Details of the apparatus The stopper, with the quartz vessel suspended from the electrodes, is placed in a clamp, and the vessel is heated electrically at 800" to 900" C in air for about 5 minutes. The weighed sample is placed in the quartz vessel after cooling. The vessel must be handled only with tweezers. A platinum wire (about 50 mm in length, 0.1 mm diameter), making contact with both electrodes, is carefully buried in the sample. The flask is charged with a suitable halogen-free absorption solution3s4 and flushed with oxygen for a few minutes.The stopcock is closed, the stopper, with the quartz vessel in place, is inserted into the flask and secured with metal springs, and the flask immersed up to the neck in a cooled water-bath. By passing a small current through the platinum wires, the thin wire starts glowing and the sample is ignited. There is a large consumption of oxygen which lowers the oxygen pressure in the flask. Initially, the current is too low to heat the quartz vessel significantly, but, by raising the current gradually to about 12 amperes, the thin platinum wire fuses and the vessel is heated to a temperature of 900" C. This heating is necessary to ensure complete conversion of the sample and to release all of the halogen. After the vapours have been absorbed, careful addition of oxygen is required with continued heating.It can be made by turning the stopcock. This addition must be repeated once or twice, depending on the amount of sample, which may vary from 0.1 g to several grams. After the combustion is complete, the flask is left to cool for about half an hour. The pressure in the flask is brought up to atmospheric pressure, and the stopper, with the quartz vessel, is removed. The halogen present in the absorption solution can be analysed by any suitable method. NEPHELOMETRIC DETERMINATION OF CHLORIDE- The absorption solution used is 10 ml of water with a few drops of formic acid to convert chlorine into chloride. As some batches of hydrogen peroxide and sodium forrnate2 contained traces of chloride, formic acid was preferred.To obtain chloride-free formic acid, a mixture of formic acid and water (1 + 1) is distilled in a quartz apparatus with a small volume of silver nitrate. PROCEDURES58 DE BOER AND VISSER After combustion of sulphide samples, the absorption solution is heated with nitric acid and silver nitrate by Lamb's method.6 Silver chloride is determined with the E.E.L. nephelometer . If large amounts of sulphide are burned, the considerable amount of sulphuric acid thus formed interferes with the nephelometric method. Chloride can easily be separated by heating the absorption solution, until it nearly fumes, in a 50-ml conical flask fitted with a distillation head. The latter consists of a ground-glass stopper with side-arm and inlet tube. The chloride is distilled as hydrochloric acid in a current of nitrogen and received in a test-tube containing 1 ml of water.After a few combustions, a platinum film may be found to have formed on the surface of the absorptioi? solution and the walls of the flask. Samples containing selenide give red selenium in the combustion flask. In both cases, the nephelometric determination of chloride is interfered with. All the chloride can be distilled, as previously described. In the case of selenide samples, however, some chloride-free sulphuric acid must be added. In the combustion of elemental sulphur, sulphur dioxide is formed. This can be easily expelled by boiling. Sensitivity of the nephelometric chloride determination is about 0.2 pg in 10 ml. PHOTOMETRIC DETERMINATION OF BROMIDE- The absorption solution is about 50ml of water.By adding organic material to the sample (we prefer three drops of isopropyl alcohol), the total bromine is converted into hydrobromic acid during combustion. Sulphur dioxide can be expelled by boiling, and hydro- bromic acid is determined by using a modified method by Pohl,6 in which bromide is converted into bromine with chloramine-T. The bromine reacts with fluorescein to give eosin (tetra- bromofluorescein), and the optical density of the red eosin is measured photometrically at 525mp. Chloride and traces of chlorine do not interfere. Sensitivity of the bromide determination is about 0.1 pg. PHOTOMETRIC DETERMINATION OF IODIDE- The absorption solution is 10 ml of water. If chloride is expected in the sample, a few drops of formic acid are added.If necessary, sulphur dioxide is expelled by boiling and the iodide converted into iodate with bromine. The iodine liberated from the iodate solution gives six times the amount of iodine in the original sample when excess of iodide is added, according to the following reaction'- 10,- + 51- + 6H+ - 31, + 3H,O. The iodine is extracted into chloroform and the optical density measured at 510mp. Sensitivity of the iodide determination is about 0.3 pg. DISCUSSION The following determinations have been carried out on the original materials: chloride 0.1 to 300 p.p.m. in zinc sulphide, cadmium sulphide, zinc selenide and sulphur; bromide, 10 to 30 p.p.m. in doped cadmium sulphide; and iodide, 10 to 30 p.p.m. in doped cadmium sulphide. It was not necessary to determine the halogens in the presence of other halogens. The sulphides were prepared with chloride-free hydrogen sulphide. We have determined 002pg of chloride in samples of zinc sulphide weighing 2g. The amounts brought to combustion varied from 0.1 to 2 g, depending on the halogen content, but it is possible to bum larger amounts. Luminescent materials, such as zinc or cadmium telluride, do not ignite at all. It is possible that this combustion method for sulphur can serve as a concentra- tion method for the determination of other impurities. REFERENCES 1. 2. Schbniger, W., Mikrochim. Actu, 1955, 123; 1956, 869. 3. 4. 5. 6. 7. Malur, J., 2. analyt. Chem., 1965, 211, 324. Macdonald, A. M. G., AnaZyst, 1961, 86, 3. Childs, C. E., Meyers, E. E., Cheng, J., Laframboise, E., and Balodis, R. B., Microchem. J., 1963, Lamb, A. B., Carleton, P. W., and Meldrum, W. B., J. Amer. Chem. Soc., 1920, 42, 251. Pohl, F. A., 2. analyt. Chem., 1966, 68, 149. Crouch, W. H., jun., AnaZyt. Chem., 1962, 34, 1698. 7, 266. Received July loth, 1967