1889/ on Aluimntuj.:. 453 



Annexed is a sketch plan of the works, winch now cover a 

 space of nearly five acres. They are divided into five separate 

 departments, viz.. 1st. soditnn, A : 2nd, chlorine. B ; 3rd. chloride, C ; 

 •ith. alnminium, D ; and 5th, fonndry, rolling, wire mills, itc. E. 



In each department an accnrate acconnt is kept of the production 

 each day. the amount of material used, the diiferent furnaces and 

 apparatus in operation. &c. In this manner it has teen found possible 

 to ascertain each day exactly how the dinerent processes are progress- 

 ing, and what efiect any modification has, either on cost, quantity, or 

 quiility of product. By this means a complicated chemical process 

 is reduced to a series of very simple operations, so that whilst tie 

 processes are apparently complicated and difficidt to c-arry cut 

 successfully, this is not the case now that the details connected with 

 the manufacture Lave been perfected, and each operation carried on 

 quite independently until the final materials are brought together for 

 the production of the aluminium. 



Manufacture of Sodium. 



The first improvement occurs in the manufacture of sodium by 

 what is known as the •* Castner Process." The succ-essful working of 

 this process marks an era in the production of s«3diuni, as it not 

 nly has greatly cheapened the metal, but has enabled the manufacture 

 10 be carrieii out upon a very large sc-ale with little or no danger. 

 Practically, the process consists in heating fused c-austic soda in contact 

 with carbon whilst the former substance is in a perfectly liquid con- 

 dition. By the process in vogue before the introduction of this 

 method, it was always deemed necessary that special means should be 

 taken to guard against actual fnsion of the mixed charges, which, 

 if it were to take place, would to a large extent allow the alkali and 

 reducing material to separate. Thus having an infusible charge to 

 heat, requiring the employment of a very high temperature for its 

 decomposition, the iron vessels must be of small circimiference to 

 allow the penetration of the heat to the centre of the charge without 

 actually melting the vessel in which the materials are heated. By 

 the new process, owing to the alkali tteing in a fused or perfectly 

 liquid condition in contact directly with carbon, the necessity of this 

 is avoided, and consequently, the reduction can be carried on in large 

 vessels at a comparatively low temperature. The reaction taking 

 place may be expressed as follows : — 



BXaHO - C = X^:C03 - 3H - Xi^. 



The vessels in which the charges of alkali and reducing material 

 are heated are of egg-shap'ed pattern, abc>ut IS inches in width at 

 their widest part and abour 3 feet high, and are made in two portions, 

 the lower one t»eing actually in the form of a crucible, while the 

 -pper one is provided with an upright stem and a protruduig hollow 

 ,rm. This part of the apparatus is known as the cover. In com- 



