RATE OF DIFFUSION OF IODINE IN KI 483 



rate. In the earlier investigations a like result was obtained with the five metals, 

 Hg, Cu, Ag, Zn, and Cd. Eight metals in all have, therefore, been shown to possess 

 the same rate of solution in iodine, a result for which there seems to be no satisfactory 

 explanation other than that furnished by the diffusion theory. 



Work of Van Name and Hill. Recently Van Name and Hill (14) 

 have measured the effect of added non-electrolytes on the reaction velocity^ 

 using metalUc cadmium and iodine in potassium iodide solution as the 

 reacting substances. The results obtained by them were not in accord 

 with the formula deduced by Arrhenius (15); this they attribute to the- 

 increase in the thickness of the diffusion layer caused by greater viscosity 

 of the solution. For our purpose the most important point in the work 

 of Van Name and Hill is that they have shown that the reaction velocity 

 varies approximately as the fluidity varies when the fluidity change is 

 caused by the addition of a non-electrolyte. This, as we shall see later, 

 certainly does not hold for the case where the decrease in fluidity is caused 

 by increasing the concentration of potassimn iodide beyond a certain 

 limit. 



Attention has been called to the fact that the reaction velocity increases 

 as the concentration of the KI increases. Nernst deduces the following 

 theoretical expression for the rate of diffusion of any electrolyte: 



D = 2 '-'''-. g.RT.10.-' 

 u + V 



Where D is the diffusion constant, u the velocity of migration for the 

 cation, v that of the anion, and g is a constant. According to this simple 

 formula we should not expect the addition of a common ion (K) to have 

 any effect on the rate of diffusion. (Or on the reaction velocity.) This 

 apparently unexplained deviation from the results naturally expected 

 from the diffusion theory of reaction velocity, becomes of great importance 

 when we note that it is the only such deviation observed in the very care- 

 ful and convincing work of the investigators quoted above. In fact it 

 seemed that if this exception could be cleared up in a satisfactory manner 

 we would be in a position to definitely look upon the chffusion theory of 

 reaction velocity as established in those cases — typical of reaction in heter- 

 geneous systems — studied by Van Name, Edgar, and Bosworth. 



At the suggestion of Dr. Graham Edgar and in every way aided and 

 guided by him, the MTiter undertook to measure directly the rate of diffusion 

 of iodine in solutions of KI of various concentrations. 



