i86 Sanitary Studies of Baking Powders 



presence of ammonium acetate and acetic acid after reduction by 

 thiosulfate. 



The method is carried out in the following manner: The Solution 

 in which the aluminium is to be determined should be, after the removal 

 of silica, about 300 cc. in vol. and contain about 2.5 cc. of conc. hydro- 

 chloric acid. Add di-ammonium hydrogen phosphate to the sol. (0.5 

 gm. for each 100 mg. of aluminium phosphate present), heat and add, 

 while hot, 5 gm. of ammonium thiosulfate^ in sol., and after several 

 minutes, 6 to 8 gm. of ammonium acetate in sol., and 4 cc. of strong 

 acetic acid sol. (80 percent was used). Continue heating the sol. for 

 about 30 min. completely to expel sulfur dioxid, allow the precipitate 

 to settle, filter, and wash once by decantation. Dissolve the precipitate 

 in 2 to 2.5 cc. of conc. hydrochloric acid, dilute the sol. to about 300 cc, 

 add 0.5 gm. of ammonium phosphate for each 100 mg. of aluminium 

 phosphate present, and again precipitate the aluminium as described 

 above. Filter and wash several times with hot water to remove 

 Chlorides, dry and ignite in a transparent silica crucible until constant 

 weight is obtained. 



The authors found it necessary, when there were large amounts of 

 iron present in the sol. containing the aluminium, to resort to a third 

 or even fourth precipitation of the aluminium phosphate, in order to 

 obtain a precipitate free from iron. This process may be avoided by 

 using larger amounts of thiosulfate and heating for longer periods of 

 time. 



Analytic results. The analytic data are recorded in the accom- 

 panying tables (i and 2). 



Discussion. An examination of the results obtained by the 

 method of Steel (and Kahn), as summarized in Table i, shows, with 

 the exception of Det'ns 25 and 26, aluminium values uniformly 

 lower than the theoretical amounts added. In the case of Det'ns 38, 

 41, and 42, the amounts of ferric phosphate alone, calculated from 

 the permanganate titration values, are greater than the respective 

 amounts of ferric phosphate plus aluminium phosphate, as obtained 

 gravimetrically. In Det'n 37 the total weight of ferric phosphate 



3 Owing to the impossibility of obtaining ammonium thiosulfate, the authors 

 used sodiiim thiosulfate in these investigations. This necessitates a little more 

 care in the final washing of the precipitate of aluminium phosphate than when 

 the ammonium salt is used, as any excess of the latter can be readily removed 

 during ignition. 



