ON  PREPARATION  OF  ALUMINIUM   FROM  CRYOLITE.  237 
of  potash  forms  the  residue.  The  prussic  acid  which  separates 
is  from  the  cyanide  of  potassium,  and  the  carbonate  of  ammonia 
from  the  decomposition  of  the  cyanate  of  potash  : 
1  at.  K0+C2N0  and  3  at.  HO,  form 
1  at.  KO+0O2  and  i  at.  NH3-rC02. 
In  closed  vessels  the  aqueous  solution  behaves  like  that  of  the 
pure  cyanide,  i.  it  acquires  a  brown  color  and  deposits  a  car- 
bonaceous matter.  Heated  wTith  dilute  sulphuric  acid  it  evolves 
hydrocyanic  and  carbonic  acids,  (consequently,  when  passed  into 
lime  water  these  gases  cause  a  turbidness,)  and  the  residue  will 
consist  of  sulphate  of  potash  and  sulphate  of  ammonia  : 
1  at.  7KOy+3  (KO+C2NO)  13  at.  S03  and  19  at.  HO,  form 
10  at.  KO+S03,  3  at.  NH40+S03,  7  at.  HCy,  and  G  at,  G02. 
Black  specks  in  the  mass  will  be  metallic  iron,  from  want  of  care 
in  pouring  it  from  the  crucible,  in  which  case  its  solution  will 
generally  have  a  slight  yellow  tinge  (contains  ferrocyanide  of 
potassium)  and  gives  a  similar  reaction,  with  sulphate  of  copper 
as  the  previous  preparation  (when  this  is  contaminated  with  iron.) 
Prac.  Pharm.  Chemistry. 
PREPARATION  OF  ALUMINIUM  FROM  CRYOLITE. 
By  Heinrich  Rose. 
Since  the  publication  of  Deville's  method  of  extracting  alumi- 
nium, I  have  endeavored  to  obtain  the  metal,  by  means  of  sodium, 
from  the  double  chloride  of  aluminium  and  sodium  ;  not,  however, 
following  exactly  the  directions  of  Deville,  but  heating  these 
substances  placed  in  alternate  layers.  The  result  was  unsatis- 
factory. Rammelsberg,  likewise,  operating  exactly  in  accordance 
with  Deville's  directions,  obtained  but  a  very  inconsiderable 
product,  and  at  the  same  time  was  seldom  able  to  prevent  the 
bursting  of  the  glass  tubes,  in  which  the  sodium  vapor  reacted 
upon  the  chloride  of  aluminium. 
The  use  of  chloride  of  aluminium  and  its  compounds  with  the 
alkaline  chlorides  is  particularly  inconvenient,  owing  to  their 
volatility  and  deliquescent  character,  which  necessitate  the  exclu- 
sion of  air  during  the  reaction  with  sodium. 
It  therefore  occurred  to  me  to  employ  instead  of  chloride  of 
aluminium,  the  fluoride  of  aluminium,  or  rather  its  compounds, 
