NATURE 



461 



THURSDAY, MARCH 18, li 



DISSOCIATION OF CHLORINE, BROMINE 

 AND IODINE 



IX NATURE, vol. xx. p. 357, I gave an account of Prof. 

 V. and Herr C. Meyer's remarkable observations 

 on the density of chlorine at high temperatures, which 

 showed that the chlorine evolved from platinous chloride 

 at temperatures of 1,200° and above had a density only 

 two-thirds of that corresponding to the molecular formula 

 CL. I also mentioned that the Meyers had stated that 

 iodine exhibited a similar behaviour. 



These observations, tending as they did to show that 

 chlorine was not the simple substance it had hitherto 

 been supposed to be, naturally excited great interest 

 among chemists, and further information has been most 

 anxiously looked for ; it must be admitted, however, that 

 they were received with considerable scepticism, more 

 especially because the statement relating to iodine was 

 in direct contradiction with a most careful series of ex- 

 periments on the comparative behaviour of air and this 

 substance made by Deville and Troost, who, after assur- 

 ing themselves that iodine vapour underwent a normal 

 expansion, made use of iodine as a pyrometer in many 

 determinations in the course of their celebrated investi- 

 gation of the density of a variety of inorganic bodies at 

 furnace temperatures. 



This scepticism was considerably strengthened by the 

 appearance, in a recent number of the Comptes Rendus 

 of a paper by a well-known American chemist, Prof. 

 Crafts, describing a quasi-repetition of the Meyers' experi- 

 ment with chlorine. The method adopted by Crafts was 

 a slight modification of that introduced by V. Meyer 

 Two graduated and calibrated U-tubes, maintained at 

 constant temperature by a bath of cold water, were con- 

 nected with V. Meyer's apparatus in such a manner that 

 a known volume of gas could be transferred from the one 

 to the heated bulb of the density apparatus through a 

 very fine tube, the volume of gas displaced by it being 

 collected and measured in the second U-tube. In two 

 experiments made in this manner at the highest tempera- 

 ture of the furnace, the density apparatus being filled 

 with air, 10 c.c. of chlorine displaced 10-37 c.c. and 1024 

 c.c. of air ; the apparatus being filled with chlorine, 10 c.c. 

 of air were found to displace 9/98 and 10 c.c. of this gas. 

 These experiments were made with a porcelain apparatus ; 

 using a platinum apparatus, 10 c.c. of chlorine were found 

 to displace 1043 c.c. and 10-50 c.c. of air. If the expan- 

 sion observed by the Meyers had taken place, the 

 quantities of air and of chlorine collected should have 

 been 15 c.c. and 6'6 c.c. respectively, so that operating 

 with free chlorine, Crafts failed to verify the observation 

 of the German chemists. 



With iodine, however, he obtained results confirmatory 

 of their statement, the observed density being 6 - oi and 

 5 -93, instead or 879, the theoretical number corresponding 

 to the formula I 2 . Bromine was found to be intermediate in 

 its behaviour, the numbers obtained being 439 and 44S, 

 instead of 5-57, indicating a reduction in density of one-fifth 

 in place of the reduction of one-third observed in the case 

 of iodine. Hydrogen chloride and carbon dioxide gave 

 Vol. x.xi.— No. 542 



normal results, showing that there was no fault inherent 

 in the method ; Crafts, however, noticed that the glaze of 

 the Bayeux porcelain vessels used was much attacked by 

 the coal-gas flame, and that at the high temperatures 

 employed they were slightly porous to hydrogen and water 

 gas, but not to other gases, although not to an extent to 

 vitiate the experiments, only 001 — "002 gramme of water 

 passing through in the course of an hour. 



The announcement of these results has led Meyer to 

 give an account of experiments he has made in conjunc- 

 tion with Herr Ziiblin since the publication of the paper 

 by C. Meyer and himself, but prior to the publication of 

 the paper of Crafts. Meyer and Ziiblin confirm the 

 accuracy of Crafts's observations. Using chlorine gas 

 prepared in the ordinary way, and carefully purified and 

 dried by passing it through water and sulphuric acid and 

 over phosphoric anhydride, they found in three experi- 

 ments at a yellow heat, 2-57, 2-63, 2-64; in mean 2'6l, 

 instead of 2-45, which is the density corresponding to the 

 formula Cl„. 



We have then the astonishing result that whereas ready- 

 prepared free chlorine is stable at high temperatures, 

 nascent chlorine, i.e., chlorine at the moment of liberation 

 from the compound platinous chloride, is unstable, and 

 undergoes dissociation : for there can now be little doubt 

 that such is the nature of the phenomenon involved in 

 the reduction of its density observed by the Meyers, the 

 argument that this may be due to a great difference in 

 the rate of expansion of chlorine as compared with gases 

 such as oxygen and nitrogen at high temperatures being 

 disposed of by the fact that free chlorine does exhibit a 

 normal behaviour in this respect. 



Meyer also publishes the results of a long series o: 

 experiments on the density of iodine. In all of these, 

 purified solid iodine was employed and not an iodine 

 compound. The first series of observations, made in a 

 porcelain vessel, are summarised in the following table : — 



On comparing these results with those for chlorine from 

 platinous chloride, it will be observed that the dissociation 

 of iodine is complete at a considerably lower temperature 

 (about 1,000°) than that of chlorine (at about 1,200 . 



These results being so at variance with those obtained 

 by Deville and Troost at a temperature of 1,040°, Meyer 

 subsequently made further experiments with entirely new 

 apparatus and fresh iodine, but without observing any 

 departure from them. A determination at about 1,052° 

 in a porcelain apparatus gave 5-88 ; and the density of 

 mercury at the same temperature being simultaneously 

 determined to control this result, the number 698 was 

 obtained in place of the theoretical number, 6-91. Two 

 experiments with iodine in a platinum vessel at about 

 1,567° gave 5-71 and 5 - Si as the density. 



