56 Gooch and Blumenthal — Use of Selenic Acid. 



The accompanying illustration, fig. 1, shows the apparatus 

 used in the preliminary tests. The reaction flask was grad- 

 uated at 5 cmS intervals to 50 cm3 , the position of the meniscus 

 being marked on a narrow slip of paper placed upon the flask 

 or by graduation marks etched in with hydrofluoric acid. The 

 alkali halides and 5 cm3 of selenic acid were introduced into the 

 graduated flask and the total volume of liquid was made up to 

 50 cm3 . After connecting the receiver, containing 3 to 4 grms. 

 of potassium iodide in 200 cm3 to 250 cm3 of water, carbon dioxide 

 was slowly passed through the apparatus and the liquid in the 

 flask, gradually heated to boiling, was kept in a state of gentle 

 ebullition until sufficiently concentrated. In an experiment 

 made under the above conditions, in which the bromide was 

 omitted, 0*2000 grams of sodium chloride caused distinct color- 

 ation of the iodide in the receiver when the volume in the 

 flask had decreased to about 20 cm3 , while 0*1000 grm. caused 

 practically no color until the volume had decreased to about 

 10 cm \ In the following experiments in which the bromide 

 was associated with chloride, therefore, the amount of sodium 

 chloride was restricted and the final concentration was kept 

 above the limit at which the attack upon the chloride begins. 

 The flask was cooled and disconnected, and the free iodine in 

 the receiver was titrated and taken as the measure of the 

 bromine. Afterward, when the chlorine was to be estimated, 

 connection was made with the receiver freshly charged with 

 iodide and the distillation was continued to secure the evolu- 

 tion of the chlorine, which was estimated by titration of the 

 iodine set free in the second distillation. The results follow 

 in Table I. 



The figures for bromine are rather low and somewhat 

 irregular, while those for chlorine, although the concentrations 

 are carried below the point at which the selenic acid itself 

 begins (as was shown in blank tests) to liberate iodine in the 

 receiver, are very irregular and generally low. It does not 

 seem to be feasible to adjust the process so that the chlorine 

 may be estimated by a second distillation, after the removal of 

 the bromine. In the following experiments, therefore, atten- 

 tion was given exclusively to the determination of bromine. 



The potassium bromide used in the following experiments 

 was prepared for use from the purest available bromate by 

 ignition, thorough fusion, pouring out upon platinum, cooling, 

 powdering, drying at 105°, boiling and keeping in a sulphuric 

 acid desiccator. A part of this material when dissolved in hot 

 water recrystallized, dried between filters and afterward at 

 100° and analyzed by the gravimetric silver nitrate process, 

 showed no variation in composition in consequence of this 

 treatment. The results obtained with the bromide alone, and 

 with that substance associated with sodium chloride, are given 

 in Table II. 



