Van Name and Brown — Tri-lodide and Tri- Bromide. 105 



Art. XI. — On Tri-lodide and Tri-Bromide Eqioilihria^ 

 especially in Cadmium Solutions ; by R. G. Van ]^[ame 

 and W. G. Brown. 



[Contributions from the Kent Claemical Laboratory of Yale Univ. — ccxc] 



The generally accepted view that iodine dissolved in a water 

 solution of a metallic iodide is present chiefly in the form of 

 tri-iodide, is based to a large extent upon evidence of the kind 

 first furnished by the work of Jakowkin, published in 1894 

 and 1896.* The principle involved consists in calculating the 

 concentration of the iodine which remains uncombined in the 

 aqueous iodine-iodide mixture from the observed concentration 

 of iodine in a layer of carbon bisulphide in equilibrium with 

 the water phase, by means of the previously determined dis- 

 tribution coefficient of iodine between water and carbon 

 bisulphide. Jakowkin showed that if the addition compound 

 which is formed is assumed to be tri-iodide (the simplest pos- 

 sibility) the data obtained as above give a good constant, within 

 fairly wide limits of concentration, for the thermal dissociation 

 of the tri-iodide according to the equation : 



RT3:^Z±:RI 4- I^, or ^ /JtV — constant. 



This equilibrium constant, which we shall hereafter designate 

 by Kj, proved to have approximately the same value in dilute 

 solutions of each of the five iodides studied by Jakowkin, 

 which means, of course, that equivalent amounts of these 

 iodides combined wdth almost exactly the same amount of 

 iodine under like conditions. The iodides tested were those of 

 potassium, sodium, lithium, hydrogen and barium. At 25°, 

 K, = 0*0014, approximately, the low value showing that the 

 tri-iodides of these I^lvq elements are only to a very small extent 

 dissociated into iodide and iodine. f 



Although Jakowkin's work has been corroborated and ex- 

 tended by other investigators, these later researches have in- 

 cluded very few iodides not studied by him. We have been 

 able to find in the literature the data necessary for the calcula- 

 tion of K, for only three such, namely, calcium and strontium 

 iodides studied by Herz and Eulla,J and cupric iodide covered 



*Zeitscbr. phys. Chem., xiii, 539, 1894, and xx, 19, 1896. 



f A sligbt difference was noted by Jakowkin in the case of hydrogen 

 iodide, which in the more dilute solutions gave Ki = 0*00]34. This dis- 

 crepancy, which merits further investigation, probably tends to disappear 

 with increasing dilution. We have confined our study of the case to a single 

 series of determinations in 0*0605 molar hydrogen iodide, which gave 

 Ki = 0*00133, in close agreement with Jakowkin's result at this concentra- 

 tion. 



j. Zeitschr. anorg. Chem., Ixxi, 254, 1911. 



