150 Pratt — Determination of Ferrous Iron in Silicates. 



In the second series, a mixture of hydrofluoric and sulphuric 

 acids was taken, such as would be used in dissolving silicates. 

 The results indicate that when hydrofluoric acid is present and 

 the heating long continued, there is a slight oxidation. It is 

 evident, however, from the results of the last five experiments 

 that very accurate determinations of ferrous iron can be made 

 by this method. 



Time. 



Method of heating. 



FeO taken. 



FeO found. 



Error. 



4 hours 



water bath 



•1214 



•1202 



0012 — 



3 hours 



water bath 



•1054 



•1047 



0007 — 



1 hour 



water bath 



•0866 



•0862 



•0004 — 



3 hours 



water bath 



•0459 



■0458 



0001- 



7 minutes 



flame 



•1527 



•1525 



0002 — 



1 minutes 



flame 



'0445 



•0445 



■oooo 



8 minutes 



flame 



•0345 



•0345 



•oooo 



8 minutes 



flame 



•0421 



•0424 



0003 + 



8 minutes 



flame 



•0472 



•0479 



•0007 + 



Fairly accurate determinations may also be made without 

 the use of carbon dioxide, the contents of the crucible being 

 cooled as quickly as possible after boiling, and titrated as usual. 

 This is the method commonly employed for determining ferrous 

 iron in silicates. To test the extent of the oxidation the fol- 

 lowing series of experiments was made, using a mixture of 

 hydrofluoric and sulphuric acids as a solvent. 



Time. 



Method of heating. 



FeO taken. 



FeO found. 



Error. 



10 minutes 



flame 



•0843 



•0825 



•0018 — 



10 minutes 



flame 



•0594 



•0578 



•0016 — 



10 minutes 



flame 



•0739 



•0728 



•0011 — 



It is evident from these results that the errors are somewhat 

 greater than in the cases where C0 2 was used. 



In experiments with very refractory minerals, where only 

 a portion is dissolved by the acids, the insoluble mineral can 

 be filtered off after titration, and the filtrate evaporated in a 

 platinum dish until all the hydrofluoric acid is expelled. After 

 dissolving the residual sulphates, the iron is reduced by hydro- 

 gen sulphide, the excess of the latter removed by boiling and 

 the total iron determined by titration. By these two determi- 

 nations, the ratio of the ferrous to ferric iron is established, and 

 as the total iron can be determined in a separate portion fused 

 with sodium carbonate, the percentages of ferrous and ferric 

 iron can be readily calculated. 



In making these experiments it has been found best to use 

 water that has been previously boiled to expel air and subse- 

 quently cooled. As an additional precaution the carbon di- 

 oxide was generated from cleavage pieces of calcite which 



