The Halogen Hydrides as Conducting Solvents. 321 



"The Halogen Hydrides as Conducting Solvents. Part IV. 

 Preliminary Notice." By B. D. STEELE, D. M ! !NTOSH, and 

 E. H. ARCHIBALD. Communicated by SIR WILLIAM EAMSAY, 

 K.C.B., F.R.S. Received November 29, Read December 15, 

 1904. 



Recent investigations have shown that the power of forming con- 

 ducting solutions is manifested to a greater or less extent by a large 

 number of inorganic and organic solvents. Some of the resulting 

 solutions behave in a similar manner to those in water, as regards the 

 variation with dilution of the molecular conductivity jtx, and of the 

 average molecular weight of the dissolved substance. Others show a 

 variation of /x which appears to be inconsistent with Arrhenius' theory 

 of ionic dissociation. 



Conducting solutions in which the halogen hydrides and sulphuretted 

 hydrogen act as solvents, are strikingly abnormal, and show, as has 

 been pointed out in Parts 1 and 2,* an enormous diminution of p with 

 dilution, instead of the steady slow increase required by the theories of 

 Arrhenius, Van't Hoff and Ostwald. 



These abnormal results can, we believe, be simply explained on the 

 assumption that the dissolved substance enters into combination with 

 the solvent, and that the resulting compound undergoes ionic dissocia- 

 tion. It can be shown from the law of mass action that, if n molecules 

 of solute combine with m molecules of the solvent to form a single 

 molecule of the electrolyte, the concentration of the latter will be 

 proportional to v~ n , where v 1 = c is the concentration of the original 

 dissolved substance. But the specific conductivity K is proportional to 

 the number of ions, and this is proportional to the concentration of 

 the electrolyte multiplied by a, where a is its coefficient of ionisation. 

 Thus 



K = aKc n = aK/v 2 , if n = 2, 

 or 



*v 2 = aK. 



If a varies but slightly over a given range this equation becomes 



KV 2 = K'. 



It is found that the whole of our results may be thus explained. In 

 some cases xv 2 is approximately constant over a certain range of 

 dilution; and that this approximate constancy does not hold at greater 

 dilutions is to be assigned to the variation of a, which may become 

 greater in more dilute solutions on account of a secondary dissociation 

 intervening. 



* ' Boy. Soc. Proc.,' vol. 73, pp. 450 and 455. 



