250 Browning-— Detection of Silicates, Fluorides, etc. 



The detection of fluosilicates by this method is of course 

 even simpler than that of either the fluorides or the silicates. 

 The results follow in the tables : 



I 



Silicon Tests. 



Name and amount Approxi- V:\V-, 



of silicate used mate present 



grm. % of SiO a grm. Result 



(1) 0-1000 SiO. -- 100 none nothing 



(2) 0-1000 " " 0-1000 verv good 



(8) 0-0100 " " " very good 



(4) 0-0050 " " " very good 



(5) 0-0010 " " " trace 



(6) 0-0100 Kaolinite 46 " very good 



(I) 0050 " " " very good 



(8) 0-0010 " " " " trace 



(9) 0-0100 Gadolinite 24 " very good 



(10) 0-0050 " " " trace 



(II) 0-0100 Lepidolite 50 0-2000 good 



II 



Fluorine Tests. 

 Name and amount Approxi- Si0 2 



of fluoride used mate present 



grm. % of F. grm. Result 



(1) 01000 CaF 49 none nothing 



(2) 0-0100 " " 0-0500 very good 



(3) 0-0050 " " " apparent 



(4) 0-0010 " " " trace 



(5) 0-0100 Na 3 AlF 6 54 " very good 



(6) 0-0050 " '■' " distinct 



III 



Fluosilicate Tests. 

 Na 2 SiF 6 present 



grm. Result 



(1) 0-0050 _. very good 



(2) 0-0010 very good 



From these results it is apparent that one milligram of 

 silicon or fluorine may be detected by the method described. 



When very small amounts of material are to be tested it is 

 especially desirable to have the material very finely pulverized, 

 and the delicacy of the test seems to be also somewhat 

 furthered by warming the sulphuric acid before adding it to the 

 mixture of the fluoride and silicate. When fluosilicates are to 

 be examined these precautions are not necessary, as the reaction 

 with the sulphuric acid takes place cpiite readily in the cold. 



July, 1911. 



