364 PROCEEDINGS OF THE AMERICAN ACADEMY 



somewhat smaller than the probable error of the individual determina- 

 tions of the atomic weight.* 



In the research of M. Dumas,t a known weight of chloride of anti- 



* Fogg. Ann., Bd. C. S. 577. An abstract of the memoir referred to is appended. 

 Determination of the Atomic Weight of Antimony. By W. P. Dexter. 



(Translated and abridged from the original Memoir in Poggendorff's Annals.) 



The antimony used in the following determinations was prepared by igniting 

 commercial tartrate of antimony and potash, which had been previously purified 

 by recrystallization, with an equal weight of nitre, adding to the mass after it had 

 become perfectly white a quantity of hydrate of potash equal to half that of the 

 nitre employed, and keeping the mixture in fusion at a dull red heat for about half 

 an hour. The fused mass was then poured into a clean iron vessel : it dissolved 

 easily and completely in hot water. The solution was filtered, and precipitated as 

 metantimoniate of soda by a solution of common salt, from which the magnesia 

 which it always contains had been separated by means of carbonate of soda. The 

 metantimoniate of soda was washed first by decantation, and then upon a filter. 

 It was then digested with nitric acid to remove the soda, and finally repeatedly 

 washed, by decantation, with dilute nitric acid. 



The hydrated antimonic acid was dried and reduced to the metallic state by ig- 

 nition in a porcelain crucible thickly coated with charcoal. To make sure that the 

 metal should take up no impurities from the charcoal, the latter was used in the 

 state of lampblack, deposited by a lamp fed with oil of turpentine and burning 

 under a large porcelain vessel filled with water. The coal thus obtained is abso- 

 lutely free from all inorganic impurities, and burns without leaving the slightest 

 trace of ash. 



The antimony procured by this process contained sodium, reduced with it from 

 some of the soda which had resisted the action of the nitric acid. It was finely 

 powdered, mixed intimately with a fresh portion of antimonic acid, and again ig- 

 nited in a porcelain crucible. The metal collected in a button at the bottom cov- 

 ered with a layer of fused oxide of antimony. 



The atomic weight having been determined upon one portion of this antimony, 

 the residue was again fused in the same way with more antimonic acid, and 

 another determination made upon the product. The agreement of the two de- 

 terminations showed that the sodium had been completely removed by the first 

 fusion. 



As it was possible that the metal might retain oxide of antimony in combination, 

 a third determination was made upon a portion which had been kept in fusion for 

 half an hour in a current of dry hydrogen. The result, however, was not sensibly 

 altered by this treatment. 



The determination of the atomic weight was effected by oxidizing the metal with 

 nitric acid, and converting the residue, by ignition, into the compound of antimonic 

 acid and oxide of antimony expressed by the formula SbOi or SbaOg. This pro- 



t Ann. Chim. Phys., S""' Se'rie, Tom. LV. p. 175. 



