156 Roberts — Action of Reducing Agents on Iodic Acid. 



an easy matter to determine the iodic acid which has con- 

 tributed to the formation of the iodine chloride. 

 The following results were obtained : 



HI0 3 : KI ) 



1 : 1-92 [-Average = 1 : 1*97 

 1 : 2-02 ) 



The three methods, then, agree fairly well in indicating two 

 molecules of potassium iodide to one of iodic acid, a reaction 

 which would be expressed by the following equation : 



HI0 3 + 2KI + 5 HC1 = 3H 2 + 2KC1 + 3IC1 



The formation of iodine monochloride has thus been proved 

 by three methods, two of which would be sufficient to estab- 

 lish the reaction. The first two methods are independent of 

 the reaction between iodine monochloride and carbonated 

 alkalies, while the third depends directly upon that reaction. 

 Assuming the correctness of Gay-Lussac's equation, this third 

 method may be taken simply as an additional way of proving 

 what has been already sufficiently established in two different 

 ways. On the other hand, considering the formation of iodine 

 chloride sufficiently established by the two other methods, the 

 last set of experiments may be used to discriminate between 

 Gay-Lussac's and Griineberg's reaction. The latter* states, 

 without describing his work in detail, that he has found by 

 many carefully repeated experiments that the action of alka- 

 line carbonates on iodine monochloride gives potassium chlo- 

 ride #ncl potassium chlorate, and all of the iodine is set free. 

 My results, however, show the correctness of Gay-Lussac's 

 equation, since it is only on the assumption that four-fifths of 

 the iodine is set free that my third method of work gives 

 results corresponding with those obtained in two other wa} r s. 

 Moreover, some iodine chloride was prepared by putting to- 

 gether potassium iodide, and an excess of iodic and hydro- 

 chloric acids, and extracting with ether, specially purified by 

 standing over sodium bisulphite and distilling from sodium 

 hydroxide. Portions of this ethereal solution were drawn off 

 from a burette and treated with potassium iodide, and the 

 iodine estimated by sodium thiosulphate. Similar portions 

 were then treated with hydrogen potassium carbonate, and the 

 iodine estimated by arsenious acid. Griineberg's reaction 

 would demand that one-half as much iodine should be set free 

 in the latter case as in the former, instead of which only four- 

 fifths of that amount was found, which is in direct agreement 

 with Gay-Lussac's equation. 



*Jour. fur prakt. Chemie, lx, 172. 



