BAXTER AND CIIAPIN. — ATOMIC WEIGHT OF NEODYMIUM. 217 



to decomposition with loss of sulphuric acid at higher temperatures. 

 The investigations of Bailey,^ Kruss,^ Jones,^ Brauner,* Wild,^ Brill,^ 

 and Holmberg 7 leave this matter still in a somewhat unsettled state. 



The dilhculty in obtaining the anhydrous sulphate in a neutral con- 

 dition applies also to the method of ignition of sulphate to oxide. 

 Furthermore it is probable that the oxide prepared by ignition from the 

 sulphate retains traces of sulphuric acid.^ 



A third method depending upon the use of the sulphates involves the 

 determination of the water in the octahydrated sulphate. Here the 

 difficulty in obtaining crystals free from included cells of mother liquor 

 is superimposed upon that of insuring complete dehydration. 



The Analysis of the Chloride. The analysis of the chlorides has been 

 little used on account of the tendency for the formation of insoluble 

 basic compounds during drying. As Matignon ^ has pointed out, if the 

 dehydration of the chloride is conducted slowly and thoroughly enough 

 in an atmosphere of dry hydrochloric acid gas, this difficulty disappears 

 and a perfectly soluble product can be obtained, even after fusion. 

 Hence on account of the accuracy with which halogen salts may be ana- 

 lyzed by precipitation with silver, this method is one of the best for 

 atomic weight purposes. In testing the applicability of this method, 

 we chose neodymium chloride for investigation, since neodymium com- 

 pounds are comparatively easy to secure in an approximately pure state 

 in large quantities. Furthermore, unlike its close relation praseodym- 

 ium, neodymium does not readily change from the trivalent state. 



Historical. The atomic weight of neodymium has already been de- 

 termined many times, with discrepant results. This element and 

 jiraseodymium, the other component of the old didymium, were first 

 separated through the fractional crystallization of the double ammonium 

 nitrates by Auer von Welsbach^^ in 1885. He determined their atomic 

 weights by Bunsen's method of converting oxide to sulphate, but ap- 

 parently interchanged his results in publication, as first suggested by 

 Brauner, as he gave the value 140.8 for neodymium and 143.6 for 

 praseodymium. 



1 Trans. Chem. Soc, 51, 683 (1887). 



2 Zeit. anorg. Chem., 3, 46 (1893). 



3 Amer. Chem. Jour., 28, 31 (1902). 



* Trans. Chem. Soc, 81, 1248 (1902). 

 6 Ibid., 38, 195 (1904). 



6 Ibid., 47, 464 (1905). 



' Zeit. anorg. Chem., 53, 124 (1907). 



* Brauner, loc. cit., 1263. 



9 Compt. Rend., 133, 289 (1901); 140, 1182 (1905). 

 " Sitzungsber. Akad. Wiss. Wien., 92, 317 (1885). 



