234 ANNUAL OF SCIENTIFIC DISCOVERY. 



is then in a double chloride of aluminum and sodium, a very fusible and 

 volatile compound. The boats are removed from the glass tube, introduced 

 into a large tube of porcelain fitted to an adapter, and traversed by a cur- 

 rent of hydrogen, quite dry and free from air. It is heated to bright red- 

 ness : the chloride of aluminum and sodium distils without decomposi- 

 tion : it is collected in the adapter, and after the operation all the aluminum 

 will be found in each boat collected in one, or at most two large globules. 

 They are washed in water, which again removes a little of a salt with an 

 acid reaction and brown silicium. To form one button of all these globules, 

 after having cleaned and dried them, they are introduced into a porcelain 

 capsule, into which is put a little of the product distilled in the previous 

 operation ; viz., some of the double chloride of aluminum and sodium. 

 The capsule being heated in a mufne to a temperature approaching the 

 point at which silver fuses, all these globules will be seen to reunite into 

 one button, which is allowed to cool, and washed. The melted metal must 

 remain in a covered porcelain crucible until the vapors of chloride of alu- 

 minum and sodium, with which the metal is always impregnated, have 

 entirely disappeared. The metallic button is found enveloped with a thin. 

 pellicle of alumina, proceeding from the partial decomposition of the 

 small quantity of distillate. It will be understood that sodium may be 

 replaced by its vapor, which is produced easily, and the aluminum will 

 be obtained economically even by employing an alkaline reducing agent. 



2d. By the Battery. It appeared to me impossible to obtain aluminum 

 by the battery in aqueous liquids. I should believe this to be an. absolute 

 impossibility if the brilliant experiments of M. Bunsen on the production 

 of barium did not shake my conviction. Still, I may say that all processes 

 of this species which have recently been published for the preparation of 

 aluminum have failed to give me good results. 



It is by means of the double chloride of aluminum and sodium, (Al 2 

 Cl 3 Na C1,V* of which I have already spoken, that this decomposition is 

 affected. The bath of aluminum is prepared with two parts, by weight 

 of chloride of aluminum, with the addition of one part of dry and pul- 

 verized common salt. The whole is mixed in a porcelain capsule heated to 

 about 392 F. The combination is effected with disengagement of heat, 

 and a liquid is obtained which is very fluid at 392 F., and fixes at thot 

 temperature. It is introduced into a tube of polished porcelain, which is to 

 be kept at a temperature of about 392 F. The negative electrode is a plate 

 of platinum, on which the aluminum mixed with common salt is deposited, 

 under the form of a grayish crust. The positive electrode is formed by a 

 perfectly dry, porous vessel, containing melted chloride of aluminum and 

 sodium, into which is placed a cylinder of charcoal, f which conducts elec- 



*This interesting substance, which represents spinel ruby with a base of soda, in 

 chlorine replaces the oxvgen, is the typo of a giv.'it number of analogous bodies which I 

 am now studying, for the purpose of comparing them with the mineral oxides, from 

 which they only differ in chlorine being substituted for oxygen. 



i The densest charcoal is rapidly dissolved in the bath, and becomes pulverulent ; hence 

 the necessity for the porous vessel. 



