Gooch and Read — Determination vf Chlorine. 551 



those of Series A and show errors of the same sort, though 

 the soluble residue was considerably less. Apparently material 

 included in the silver plating of the anode was not wholly 

 extracted by the process of washing and boiling. 



The next experiments were made with the rotating gauze 

 anode plated with silver from a solution of silver oxalate in 

 ammonium hydroxide to avoid all possible contamination of 

 the silver deposit by nonvolatile material. The course of the 

 electrolysis was similar to that of the experiments previously 

 described, nearly to the finish ; but, near the end, when neu- 

 trality to litmus indicated the exhaustion of hydrochloric acid, 

 the solution suddenly became opalescent and soon afterward the 

 current practically ceased to flow. Upon standing, the liquid, 

 which at the end of the electrolysis had slowly bleached blue 

 litmus paper without reddening it, developed distinct acidity, 

 and when tested in separate portions gave further opalescence 

 with silver nitrate and set free iodine from potassium iodide. 

 All these phenomena point to the formation of hypochlorous 

 acid in the process of electrolysis and its attack upon the 

 anode to form silver hypochlorite and derived silver salts. 

 It appears further, that in the absence of alkali salt, soluble silver 

 hypochlorite, apparently formed chiefly when the hydrochloric 

 acid approaches the point of exhaustion, is thrown into solu- 

 tion, to be partially decomposed with production of opalescent 

 silver chloride. In Table II are given the details of experi- 

 ments made with the silver anode plated in the oxalate solution. 

 In the first experiment the electrolysis was continued until the 

 current practical!}" ceased to pass. In the other experiments 

 the operation was ended when the diffusion of the opalescent 

 silver chloride indicated that the silver anode was being attacked, 

 dissolved, and partially reprecipitated in the liquid. In all the 

 liquid was neutral at the end of the electrolysis. 



The results of these experiments, in which no alkali salt 

 derived from the plated anode could be present, confirm those 

 of the preceding series in respect to the fixing of oxygen as 

 well as chlorine upon the anode, the removal of silver from 

 anode to cathode, and the formation of hypochlorous acid. 



It is plain, therefore, that the electrolytic determination of 

 the chlorine of hydrochloric acid with the use of the silver 

 anode is by no means an exact process. Even when precau- 

 tions are taken to plate the anode with a silver solution con- 

 taining salts which, if included in the deposit, will leave 

 nothing but silver, and to ignite the anode with the electrolytic 

 deposit to the point of decomposition of attached silver oxide, 

 the results are very irregular and always low. 



