102 ATOMIC WEIGHT DETERMINATIONS. 



RUBIDIUM. 



Kopp determined the specific heat of rubidium chloride. 

 It corresponds to an atomic weight of about 85. [Gmelin- 

 Kraut, I. c.) 



KiRCHHOFF and Bunsen : 85.36 (O = 16). 



Determined from the mean of four experiments on the 

 precipitation of the chloride with argentic nitrate. The 

 extreme difference was 0.24. Ag = 107.94; CI = 35.46. 

 An impure mixture of rubidium and potassium chlorides, 

 nearly free from lithium and the earths, was partially pre- 

 cipitated with platinum chloride and the precipitate freed 

 from KOI by repeated boiling with water. The residue 

 was reduced in a current of hydrogen, the rubidium chlo- 

 ride extracted with water, and reprecipitated with platinum 

 chloride. This process was repeated until the potassium 

 lines in the spectrum disappeared. The rubidium was then 

 converted into a mixture of carbonate and oxide, and the 

 caesium separated by extraction with alcohol. The amount 

 of silver precipitated was also tested from time to time and 

 the purification continued till this became constant. {Pog- 

 gend. Ann., 113, 1861, 339.) 



J. PiccARD : 85.41 (0 = 16). 



Determined by four analyses of rubidium chloride with 

 argentic nitrate. The number is the mean ; extreme differ- 

 ence, 0.09. The separation of potassium from rubidium 

 was effected for the different analyses by 6, 7, and 8 succes- 

 sive partial precipitations with platinum chloride, and the 

 separation of caesium by thirty successive extractions of 

 the anhydrous carbonates with warm absolute alcohol. The 

 salt analysed was spectroscopically pure. Ag = 107.94 ; 

 01 = 35.46. The experiments were made with Bunsen's 

 assistance. [Erdmann's Journ. fur Prak. Chem., 86, 1862, 

 449.) 



L. Grandeau, who is sometimes credited with making a 

 determination of Rb, expressly disclaims doing so. He 

 mentions Bunsen's value as the true atomic weight and says 

 that his analyses of the sulphate, undertaken to test its 

 purity, led him to adopt the number 85.4 ; apparently for 

 brevity's sake. {Annal. de Chim. et de Phys., (3,) 67, 1863, 

 227.) 



