[archibald-fulton] 



ESTIMATION OF IRON 



245 



A corresponding solution of manganese chloride of approximately 

 twice the strength of the iron solution was prepared. This solution 

 was standardized by weighing the manganese as pyrophosphate 

 thrown down as phosphate from a measured volume of the solution. 

 The average weight of pyrophosphate obtained from different 25 cc. 

 portions of solution was 0-3522 gram. The solution therefore contain- 

 ed 12-496 grams of MnCl2 per litre. 



A series of determinations was now made to test the influence of 

 of the volume of the mother liquor, and therefore the dilution of the 

 acid upon the weight of precipitate obtained and the completeness of 

 the separation of the manganese. For this purpose 25 cc. of the iron 

 solution were measured off and to this was added 25 cc. of manganese 

 solution and 20 cc. of concentrated hydrochloric acid. The solutions 

 were then made up to the volumes set forth in the table below. For 

 each precipitation 2}/i grams of cupferron were weighed out and 

 dissolved in sufficient water to yield a 6% solution. These solutions 

 were cooled and filtered in each case before using, and a fresh solution 

 was prepared for each precipitation. The solutions of the chlorides 

 were now cooled in water at 7°C. and the cupferron solution added 

 slowly with constant stirring. The precipitate was reddish-brown, 

 crystalline, rather bulky, but could be easily crushed with the stirring 

 rod. The solutions were filtered almost immediately, the beakers 

 meantime being immersed in the cold water. The precipitates were 

 in each case washed with 75 cc. of cold water, twice with water con- 

 'taining ammonium hydroxide to remove the nitroso-phenylhydroxyla- 

 mine, and twice with water. After the filter paper had been burned 

 and all organic matter destroyed, the iron oxide was brought to con- 

 stant weight with the Meker burner. In a number of cases the pre- 

 cipitate had been weighed, and in order to be sure that no iron had been 

 reduced, it was treated with nitric acid, the excess of acid expelled, 

 the precipitate again ignited and weighed. This treatment never 

 made any difference in the weight. 



Below are the results of experiments at four different dilutions. 

 The weight of iron oxide found is the average, in each case, of two 

 closely agreeing values. 



Table I 



