1889.] 



Zirconium and its Atomic Weight 



81 



It is evident, therefore, that with due precaution an oxide agreeing 

 very definitely with the formula Zr0 3 , is obtainable by precipitation 

 with hydrogen peroxide, either in alkaline or acid solution. The 

 addition of ammonia, especially in excess, is, however, not to be 

 recommended, since it tends to bring down the ordinary oxide, and to 

 decompose the hydrogen peroxide added, with, the production of 

 nitrous and nitric acid. Dried for several weeks over phosphorus 

 pentoxide, the oxide became constant in weight and showed the com- 

 position Zr0 3 .3H 2 0, the hydrate of zirconia being Zr0 2 .2H 2 0. The 

 salt, however, lost about 2 per cent, of oxygen during the process of 

 drying, and if this was carried out at 100° it had the composition 

 Zr 2 5 . 



Zirconium Tetrachloride. 



About 30 grams of this salt were prepared in the usual way, by 

 passing dry chlorine over a mixture of charcoal and zirconia. If a 

 determination was to be made from this salt some process of purifica- 

 tion was necessary, and a criterion that a body of constant composi- 

 tion had been obtained, especially as it is known that volatile 

 oxychlorides exist. 



For this purpose a combustion-tube, in which the chloride was pre- 

 pared, was drawn out into a series of bulbs, separated by constrictions. 

 The chloride was then sublimed (a lower temperature being used for each 

 successive bulb) in a current of dry chlorine into these bulbs, which 

 were separately sealed off and their contents analysed. It was hoped in 

 this way that two or more bulbs would be obtained showing con- 

 cordant results, but notwithstanding that every precaution was taken 

 to exclude air and moisture, no such result could be achieved, and 

 decomposition was clearly indicated to have taken place with the for- 

 mation of traces of a reddish-brown product, the equivalent of 

 chlorine in relation to zirconium being in some cases higher and in 

 others lower than four to one. Furthermore, in no case was it found 

 possible to prepare such a product perfectly free from silica or iron. 

 It seems, indeed, that zirconium tetrachloride is even more liable 

 to dissociation and decomposition than the corresponding silicon 

 compound. 



Zirconium Oxychloride. 



This salt is prepared, as already stated, by crystallisation from con- 

 centrated hydrochloric acid. The difficulties attending its preparation 

 in the pure state arise from the presence of free hydrochloric acid 

 and from the readiness with which the salt passes into oxychlorides 

 of a more basic character. Berzelius attempted to remove the excess 

 of free acid by heating the salt to 60°, but was not able to obtain a 

 definite product, his numbers being — 



VOL. xlvi. G 



