NATURAL SCIENCES OF PHILADELPHIA. 13 



Mr. Monnier in his experiments met with the same diCBculties in the reduction. 

 He found, however, by careful investigation and analysis, that the effect pro- 

 duced was also perfectly in accordance with the theory. When, after great dif- 

 ficulties, sufficient quantities of pure aluminium were obtained, and its proper- 

 ties as a metal carefully studied ; it was found that it was not in the least 

 degree oxidized by fusion with the nitrate of potash. This peculiar property 

 causes a strong contrast between it and any well known metal, and this won- 

 derful fact produced a new phase in the manufacture of aluminium. Owing to 

 this discovery, the efforts which had been made to produce aluminium from 

 perfectly pure material were found unnecessary. 



By using pure materials for its manufacture, it was necessary to employ ap- 

 paratus which was very costly, as it required that it should be free from any 

 injurious substances. 



By the facility with which aluminium can be refined, owing to its peculiar 

 properties to resist oxidation, it can be manufactured from impure and crude 

 materials, in apparatus which can be cheaply made of still cheaper material, and 

 without the great care and watchfulness necessary in the manipulations, where 

 pure materials are employed. 



Since July, 1855, Mr. Monnier has made the double chloride by mixing alumina 

 with salt (chloride of Sv/dium) and coal, and bypassing over this mass (ignited) 

 a current of dry chlorine gas, (kaolin or common clay can be used instead of the 

 alumina.) The double chloride runs out from the condenser in a stream, and 

 is collected in a receiver; it becomes solid when cold. 



The reaction is so complete that no chlorine is lost. It has already been 

 manufactured at a cost of eight cents per pound, but the operations were car- 

 ried on to a very limited extent. It is, however, clearly (to be) demonstrated that, 

 with works of sufficient size and a proportional economy in manufacture, the 

 double chloride of aluminium and sodium, can be produced at a cost not ex- 

 ceeding four and a half cents per pound. 



In the manufacture of sodium Mr. Monnier has made considerable improve- 

 ment; it has already been produced at a cost of twenty-five cents per pound. 



In the manufacture of zinc (by the Belgian process) one retort produces about 

 thirteen pounds of metal in 24 hours. In the manufacture of sodium the reduction 

 is so quick that 52 pounds of this metal can be produced in a retort of the same 

 size in the same space of time. 



The reduction of zinc costs from two to two and a quarter cents per pound. 

 The amount of ore necessary, being from a half to three-quarter cents per pound. 

 Giving for the cost of one pound of metal about three to three and a quarter 

 cents. 



The reduction of sodium costs about ... 4 cents per pound. 



The carbonate of soda, 2^ lbs. at 4 cents, . . 10 " •' 



Giving for total cost of one pound of sodium, . . 14 " 



Sodium can be manufactured on as large a scale as zinc, and when the work- 

 men have the same practical experience in the manipulations, the price of reduc- 

 tion will be still farther reduced, so that the difference between it and zinc will 

 be that of the first cost of carbonate of soda and the zinc ore. 



It has already been shown that the double chloride of aluminium and sodium 

 can be produced at a very low price ; metallic sodium can also be made very 

 cheaply ; the reduction is readily effected in accordance with the theory, but the 

 difficulty in uniting the metallic globules of aluminium when formed, is obviated 

 by stirring at this point with a rod, and the destruction of the metal by the 

 alkaline scoria is prevented by adding to the charge of the double chloride of 

 aluminium and sodium, chloride of sodium and metallic sodium, either fluoride 

 of calcium, (fluor spar,) or cryolite. 



By these means the greater portion of the metal is united in one mass, the 

 other portion of the metal in small globules remain with the slag, which can be 

 removed mechanically, or by first digesting in water ; there will however remain 



1857.] 



