26 BATES AND JACKSON: QUARTZ-WEDGE SACCHARIMETER 



For the present investigation the purest granulated sugar of 

 commerce was dissolved in its own weight of distilled water, clari- 

 fied with washed alumina cream and filtered. The clear solution 

 was boiled in a glass vacuum boiling apparatus at about 35°C. 

 until it reached a concentration of about 80 per cent, sugar. The 

 supersaturated syrup was poured out and allowed to crystallize 

 while in continuous motion. The crystals were separated from 

 the mother liquor by a powerful centrifugal machine and washed 

 with pure alcohol. They were then redissolved and recrystallized 

 repeatedly until deemed of sufficient purity to test. 



The progress of the purification was studied by testing for such 

 impurities as could be detected. By weighing the ash left after 

 ignition, inorganic impurities were found to be satisfactorily 

 removed after two crystallizations. 



The test for reducing substance was complicated by the fact 

 that sucrose itself possessed a slight reducing power. Neverthe- 

 less a diminution of reducing power indicated the elimination of 

 foreign reducing substances. By boiling a 10 gram sample of the 

 original granulated sugar with 50 cc. of the Striegler 4 reagent, 

 a precipitate of 20 mg. of Cu 2 was obtained. After one re- 

 crystallization, this precipitate diminished to 9 mg., and after 

 two crystallizations it became 7 mg. This latter quantity 

 proved to be a minimum. To show that this precipitate was 

 caused mainly by sucrose itself, analyses were made with a modi- 

 fication of the Soldaini reagent 5 consisting of about 300 grams 

 KHC0 3 and 1 gram CuS0 4 5H 2 in a liter of solution. By 

 boiling a 10 gram sample with 50 cc. of this solution for two 

 minutes a precipitate of 1.1 mg. Cu 2 was obtained. To an- 

 other sample 0.01 per cent or 1 mg. of invert sugar was added 

 and the mixture boiled with the copper solution. The invert 

 sugar caused an increased precipitation of 1.9 mg. Hence even 

 if the entire precipitate of 1.1 mg. Cu 2 were due to impurities, 



the latter could amount to but -^ X 0.01 per cent = 0.006 per 



i.y 



cent. 



4 von Lippmann, Die Chemie der Zuckerarten, 1: 606. 1904. 



5 Ibid. 



