273 On the Manufacture of Aluminium^ 



posing a mixture of alumina or of kaoline with excess of fluoride 

 of calcium, in a tube of carbon properly protected, to a white 

 heat, and passing through it gaseous hydrochloric acid. By a 

 modification of this method Deville has obtained the sesquifluoride 

 of zirconium. 



The sesquifluoride of iron is formed by treating calcined oxide 

 of iron with hydrofluoric acid and heating strongly. It is ob- 

 tained in the form of small transparent cubic crystals, very 

 slightly coloured. 



. By a modification of this method sesquifluoride of chromium 

 is produced. It is obtained in the form of octahedra of the 

 colour of oxide of chromium, and whose faces have a very bril- 

 liant lustre. 



M. Ch.Tissier* obtained by treating cryolite, APFl^SNaFI, 

 with quicklime, an aluminate of soda. The peculiarity of the 

 action was, that however the proportion of the lime was varied, 

 only about one-third of the aluminium was obtained in the 

 form of aluminate of soda. By prolonging the action, there was 

 only obtained very pure hydrate of soda. 



M. Tissier states, that in numerous experiments on the reduc- 

 tion of aluminium from cryolite by sodium, only a certain fraction 

 of the aluminium contained in the mineral is obtained. This 

 result is in accordance with the observations of many other che- 

 mists who have worked on this subject. 



Brunnerf, and also Deville J, have given methods for preparing 

 artificial cryolite. Deville's method consists in mixing calcined 

 alumina with dry carbonate of soda and drenching the mixture 

 with hydrofluoric acid, then drying and heating it to redness. 

 An insoluble mass is obtained which has the composition of cry- 

 olite. From this cryolite and from natural cryolite, Ueville and 

 Brunuer, and more recently W6hler§, have given methods by 

 which aluminium may be prepared by the action of sodium. 



From the labours of Deville, Rousseau, and Morinjl, the ma- 

 nufacture of aluminium has received great improvements. The 

 method used is still essentially the same ; chloride of aluminium 

 is decomposed by sodium ; it is in tlic manufacture of these 

 substances, and in the apparatus by which the action is efi'ected, 

 that the improvements have been made. 



Instead of chloride of aluminium alone, the prejjaration of 

 which on the large scale is very difficult, a double chloride of 



* Comptes Rendus, July 14, 1856. 



t Poggendorff's Annalen, March 185G. J Loc. cit. 



§ Liebig's Annalen, August 1856. 



II Comptes Rendus, October 13, 1856. 



