April 5, 1889.] 



SCIENCR 



261 



duction of the latter metal, it will be seen that the cheapness and 

 abundance of the aluminium depends very much on the cost of the 

 sodium, and the quantity in which it can be produced. Till quite 

 recently, the price of sodium was as high as 5 shillings a pound ; 

 and the process of manufacture was so difficult, and even danger- 

 ous, that very large quantities could not be obtained. The im- 

 provements effected by Mr. Hamilton Y. Castner in the manufac- 

 ture of sodium, by which it can be made in any quantity without 

 the slightest risk at about i shilling per pound, has rendered it 

 possible to produce aluminium of about 98 per cent purity which 

 can be sold profitably at 20 shillings per pound. 



The object of a paper by William Anderson, in a recent number 

 of the Journal of the Society of Arts, London, is to describe the 

 process of manufacture adopted by the Aluminium Company, at 

 the works, which have just been started, at Oldbury, near Birming- 

 ham. 



We will first take the manufacture of the double chloride of 

 aluminium and sodium. The raw material is hydrate of alumina 

 (AlaOj + water), which is the only oxide of aluminium. It can be 

 prepared in a variety of ways, and from various materials, such as 

 common alum, which is the double sulphate of aluminium and 

 potassium, Al K (SO4) i2HaO; bau.xite, which, as already stated, 

 is a ferruginous hydrate ; and other substances ; the price being 

 about ^13 per ton when it is sufficiently pure for the purpose. 



The hydrate of alumina, in a finely divided state, is mixed, on a 

 suitable floor, with lamp-black, charcoal, and common salt ; mois- 

 tened with water ; the mass is thrown into a pug-mill, and, after 

 being thoroughly mixed and incorporated, is forced through dies 

 constructed exactly as in a drain-pipe machine, the issuing cylin- 

 ders of the compound being cut off by wires into pieces about three 

 inches long, which are carried to the tops of the chloride furnaces, 

 and spread out there to dry thoroughly. 



The next process is to expose the mixture of hydrate of alumina, 

 carbon, and salt, to a high temperature in presence of chlorine gas, 

 in order to obtain the vapor of the double chloride, which is dis- 

 tilled over, and condensed in the form of a deliquescent, light- 

 yellow substance of very pungent odor. This operation is per- 

 formed in regenerative furnaces, constructed very like banks of 

 ordinary earthenware gas-retorts. The gas from the producers 

 plays round groups of five retorts, set in ovens, in which their 

 temperature is raised to a bright-red heat, the exact intensity of 

 which is a matter of much importance, and requires an experienced 

 €ye to regulate. 



The retorts are connected at their mouths — that is, their open- 

 ing ends — by means of earthenware pipes to gas-holders contain- 

 ing chlorine gas, special means being taken to regulate the pressure 

 of the gas and the rate at which it is allowed to flow. The op- 

 posite ends of the retorts are fitted with pipes, which convey the 

 fumes of double chloride to cast-iron condensers, and thence to 

 brick chests or' boxes, the outsides or ends of which are closed by 

 means of wooden doors. Convenient openings are arranged for 

 clearing out the passages, because the double chloride condenses 

 very quickly. The greater portion of it liquefies, and trickles down 

 into the brick chambers ; while a portion sublimes, and comes over 

 in the form of a yellow powder. The brick chambers are emptied 

 from time to time, and the contents packed away in air-tight 

 wooden chests, — a precaution rendered necessary on account of 

 the deliquescent properties of the substance. The re-action which 

 takes place is as follows : — 



2 AlsOa -f 3 Cs -1- 4 Na CI -1- 6 CI3 = 2 (Al^Cle 2 Na CI) -f 6 CO. 

 That is, two molecules of alumina, three molecules of carbon, four 

 molecules of salt, and six molecules of chlorine, give two molecules 

 of double chloride and six molecules of carbonic oxide. 



The chlorine gas used in the retorts is manufactured on an enor- 

 mous scale, being prepared in the usual way, by the action of 

 hydrochloric acid on manganese dioxide, at a moderately high 

 temperature. One molecule of the manganese dioxide combining 

 with four molecules of hydrochloric acid, produces one molecule of 

 manganous chloride, two molecules of water, and one molecule of 

 gaseous chlorine : — 



Mn O, -f 4 H CI = Mn CI3 -)- 2 H2O -I- CU. 



The manganous chloride is soluble, and forms the " spent still 

 liquor," which is reconverted into manganese dioxide by Weldon's 



method ; that is, by first neutralizing all free hydrochloric acid by 

 means of powdered limestone, and then adding milk of lime to the 

 neutral solution, when manganous oxide and calcic chloride are 

 formed ; — 



Mn CI5 -I- Ca O = Mn O -I- Ca Cla. 



By exposing the manganous oxide to a strong current of air, it 

 takes up another atom of oxygen, and becomes again MnOg, or 

 manganese dioxide. 



The chlorine plant forms a very imposing part of the factory. 

 The hydrochloric acid is conveyed by a 2-inch gutta-percha pipe a 

 distance of some 700 feet across the canal, from Messrs. Chance 

 Brothers' Alkali Works. It is received into six large stone storage- 

 tanks, each capable of containing 10 tons ; and from these it is run, 

 as it is wanted, into two large stone stills made up of huge slabs 

 of sandstone cramped together in an ingenious manner by iron 

 bolts and cast-iron angle saddles, the joints being made by means 

 of solid India-rubber cord. In these stills the manganese dioxide 

 and the acid are mixed ; and, being warmed by injected steam to 

 the proper temperature, the chlorine gas is at first given off rapidly 

 and with effervescence. The rate gradually decreases, and at 

 length the disengagement of gas ceases altogether. The chlorine 

 is carried off by means of lead and earthenware pipes to four large 

 lead-lined gas-holders, capable of containing several thousand 

 cubic feet of gas ; and from them it is led away to the double 

 chloride retorts, various ingenious devices having been introduced 

 for indicating the pressure of the gas, and measuring the quantity 

 passed into each retort. Chlorine, besides being valuable, is a 

 very disagreeable gas when it gets out of its proper place : hence 

 great care and method are required in manipulating it. 



The " spent still liquor," the solution of manganous chloride, is 

 run into a large neutralizing well, 20 feet diameter, and 20 feet 

 deep, built of stone, and fitted with agitators. It is there neutral- 

 ized by intimate mixture with powdered limestone, and is allowed 

 to settle, after having been pumped up to a system of tanks ele- 

 vated above the oxidizing-tower, during which process iron and 

 some other impurities are carried down. The clear solution, which 

 has a pinkish color, is then run into the oxidizing-tower, which is a 

 wrought-iron cylinder standing on end, about 12 feet diameter and 

 30 feet higft, where it is warmed by injected steam. Milk of Hme 

 is added, and the whole violently agitated by a powerful current of 

 air, pumped in at the bottom of the tower by an 80-horse-power 

 horizontal engine driving a large double-acting air-pump. In two 

 or three hours the manganous oxide has absorbed as much oxygen 

 from the air-current as it had at first given up to the hydrogen of 

 the hydrochloric acid, and thus reverts to its original state. 



The contents of the tower, now a thick black turbid liquid, are 

 run into a second system of settling-tanks, five in number, erected 

 below the level of the tower. The tanks are each 18 feet square 

 by 7 feet deep, and are used alternately for settling the charges as 

 they are withdrawn from the tower. The recovered manganese 

 dioxide settles out, leaving a clear solution of chloride of calcium, 

 which is drawn off by overflow pipes ; and the recovery process is 

 then complete, the manganese mud being thus used over and over 

 again, and re-recovered, suffering but an inconsiderable amount of 

 loss in the process. 



We next come to the manufacture of sodium. Previous to the 

 year 1886, sodium was produced by reducing it from the hydrate 

 or carbonate of soda by heating it to a very high temperature, with 

 an excess of carbon, great care being taken to avoid fusion of the 

 mass, to which end lime was added. Fusion was prejudicial to 

 the process, because, when fused, the carbon separated from the 

 alkali, and only a small return was obtained : hence the tempera- 

 ture had to be carried sufficiently high for the alkalies to be vola- 

 tilized, because only in that form could the soda compounds come 

 into sufficiently intimate contact with the carbon for its combina- 

 tion with the oxygen of the alkali to take place, and set the me- 

 tallic sodium free. The high temperature required caused great 

 wear and tear of ihe iron retorts in which the process had to be 

 carried on, and dangerous explosions were not uncommon ; the 

 practical effect being that the production of sodium was very lim- 

 ited in quantity, and the price, as already stated, ranged as high as 

 5 shillings per pound. 



Mr. Castner, a chemical engineer of New York, became pes- 



