Peirce — Gravimetric Determination of Selenium. 417 



showed that for small amounts the process gave accurate results, 

 but that for large amounts the errors came far too high : 





Se taken 



Se found 



KI 



Volume 



Error 





grm. 



grm. 



grm. 



cm 3 



grm. 



(1) 



0-0355 



0-0356 



1 



100 



0-0001 + 



(2) 



0-0355 



0-0355 



1 



a 



o-oooo 



(3) 



0-0355 



0-0356 



1 



it 



o-oooi + 



(*) 



0-2968 



0-3883 



4 



75 



0-0915 + 



(5) 



0-2033 



0-2475 



4 



100 



0-0442 + 



(6) 



0-3058 



0-3495 



10 



100 



0-0437 + 



When the selenium amounted to less than a tenth of a gram 

 the results came out well. When the amount was larger the 

 selenium assumed on boiling a pasty molten condition which 

 made filtering and washing impossible. This condition was 

 observed in the work already referred to, and the mass seemed 

 to consist of selenium with included iodine, as it gave up 

 iodine slowly to water on standing and more rapidly to a solu- 

 tion of potassium iodide. 



It has been found in the work to be described that if an 

 excess of potassium iodide be used considerably above the 

 amount necessary for precipitation, the pasty condition of the 

 selenium does not occur, the iodine evidently being held in 

 solution by the excess of the potassium iodide. This would 

 seem to indicate the total release of the iodine, and would 

 make possible the determination of quantities larger than the 

 two-tenths of a gram set in the former work as a limit for the 

 range of the process. Several determinations, which resulted 

 very satisfactorily, were made according to the volumetric 

 method thus modified, and at the same time the selenium 

 which was precipitated was weighed. By the former method, 

 in which the iodine evolved was estimated, the results were too 

 low, and by the latter method, in which the residue was 

 weighed, too high ; but there appeared to be no definite rela- 

 tion between the errors of the two processes, as there would 

 be if the retention of iodine were the only difficulty. Diges- 

 tion of the selenium in the crucible with hot , water removed 

 a small amount of potassium iodide and reduced the error 

 considerably ; so that it was apparent that the increase in 

 weight was due to the fact that the precipitated selenium 

 included potassium iodide from the concentrated solution. 





Se0 2 



Se0 2 





Se taken 









taken 



found KI 



Volume Error 



as Se0 2 



Se found 



Error 





grm. 



grm. grm. 



cm 3 grm. 



grm. 



grm. 



grm. 



(») 



0*4870 



0-4868 7 



60 0-0002 — 



0-3467 



0-3507 



0-0040 + 



(8) 



0-4980 



0-4971 10 



" 0*0009 — 



0-3545 



0-3575 



0-0030 + 



(9) 



0-7323 



0*7310 10 



75 0-0013 — 



0-5214 



0-5312 



0-0098 + 



