546 Gooch and Head — Determination of Chlorine. 



that when the rotating anode was used in the double cell 

 nothing but pure water remained in the inner compartment 

 after the salt under examination had been decomposed ; that 

 the falling of the current to 0*01 ampere or less indicated 

 the end of the process ; and that no harm could be done by 

 running longer, as further increase in the weight of the 

 anode was not possible. The results obtained by weighing 

 the anode were concordant with one another, close to the 

 theory, and in agreement with the figures obtained by titration 

 of the alkali formed in the liquid. 



McCutcheon,* Lukens and Smith,f and Lukens and 

 McCutcheon,^: have studied still further the behavior of 

 various chlorides, and other salts, in electrolysis with the use of 

 the rotary silver anode and the mercury cathode. 



Throughout all the later elaborate experimentation no 

 reference is made to the points emphasized by Vortmann as 

 necessary in the electrolytic determination of iodine, viz., the 

 ignition of the silver anode to break away fixed oxygen and 

 the determination of silver dissolved or carried to the cathode. 



From a consideration of the apparently very exact results 

 obtained in the treatment of various chlorides it would seem 

 that nothing could be simpler than the accurate determination 

 of chlorine in hydrochloric acid by similar means. That such 

 is not the case, however, will be seen from the following 

 account of experimentation upon the electrolytic determination 

 of chlorine in hydrochloric acid with the use of the silver 

 anode. In a preliminary experiment a large silver crucible 

 was used as the anode with a smaller platinum cathode, and 

 a current of 1*5 to 0*09 amperes under a potential of 3-5 to 4 

 volts passed for an hour through a solution containing origi- 

 nally 0*2184: grin, of hydrogen chloride. It was found in this 

 experiment that the combined weights of anode, cathode, and 

 suspended silver compound collected on asbestos failed by 

 several milligrams to make up the weight of chlorine contained 

 in the hydrogen chloride ; and testing of the residue left after 

 evaporation of the clear filtrate showed distinctly the presence 

 of a chlorate. § 



In another experiment, similar excepting that the voltage of 

 the current was raised as the operation progressed, a current 

 of 0*5 to 0*24 ampere under a potential of 4 to 80 volts was 

 passed for thirty minutes through a solution containing at the 

 outset 0*2184 grm. of hydrogen chloride. The combined 



*Jour. Amer. Chem. Soc, xxix, 1445. f Ibid., xxix, 1455. 



t Ibid., xxix, 1460. 



§ The test for a chlorate was made by treating a portion of the residue 

 with a saturated solution of manganous chloride in hydrochloric acid (Gooch 

 and Gruener, this Journal, xliv, 118, 1892). 



