The liquor that is drawn off to the settlers should be of a yellowish brown colour and perfectly clear. It should be tested two or thrie times daily for sulphide of sodium, to make sure that tho tanks are not being overheated or the liquor allowed to stand too long before being drawn off. The amount shown should never exceed 0.75 per cent., though where the weak liquors are pumped back upon the tanks, and used over and over again in place of water—a piece of poor economy—as much sulphide as 2 per cent. will be often registered. As this is simply converted into sulphate in the after processes, it is sheer loss of sada. An average tank liquor, not the best, will show ubout the followiug composition:— Sodium carbonate 69.0 Sodium cyanide trace „ hydrate .. 15.0 „ ferroeyanide trace „ sulphide 1 0 „ silicate .. 0• 5 •• sulphite .. .. 2 0 „ aluminate 0.5 „ sulphate .. .. 7 . 0 Iron and alumina .. 0.5 „ chloride 3.0 I n solu bit) 0 5 Many processes have been proposed to purify the tank liquors from cyanides, dm., and carbonate or peroxidize the sodium compounds. The method proposed by Gossage is perhaps the most extensively adopted. An iron tower, usually a largo pipe, is loosely packed with coke, and the tank liq.uor run down. During its passage it is n3et with a stream of air, or impure carbon dioxide from a lime-kiln, or the waste gases from some furnace containing large quantities of both oxygen and carbon dioxide. The caustic soda is thereby converted into carbonate, the sulphide oxidized, and the aluminates decomposed with precipitation of alumina. Another process is to pass a steam blast into the liquor, under a perforated false bottom which ensures the division of the steam and its action upon every portion of the liquors. Upon the whole tlaese processes do not repay for the trouble and expense ineurred in adopting them, the best plan being to get as good liquor as possible, and then use it without further manipulation. The eyanogen compounds of the liquor have attracted a good deal of atteution, but no practical result has ever been attained in the way of getting rid of them. Stevenson has proposed to heat the liquors under pressure to 150°. Probably the best plan is to use a mixing coal containing the least possible amount of nitrogen. The smell of ammonia whieh proceeds from well-worked balls when cooling, is the result of decom position of certain of the eyanogen compound.s.
The weak liquor, standing about 1° Tw., is, as has been stated, sometimes used over again in the place of water, but is usually run to waste. It contains very small proportions of sodium carbonate, hydrate, sulphide, hypesulphite, sulphate, chloride, silicate, and aluminate.
Various methods for, utilizing the tank waste will be detailed hereafter. ITsually it is removed as soon as thrown out of the tanks, and either carried out to sea or deposited upon waste land. Some use is made of it in building walls and laying foundations, the sulphate of calcium, or gypsum, which is formed by the action of the air causing it to set very hard. If allowed to remain in heaps, as loosely thrown out of the tanks, the mass speedily becomes hot, even red hot. The oxygen of the air, and the moisture present, cause the formation of soluble calcium hydrosulphide, bisulphide, and hyposulphite, &c., and the presence of carbon dioxide causes an evolution of sul phuretted hydrogen, which is most offensive and injurious. Much of this evil can be prevented by spreading the waste over the ground, or building it promptly into whatever shape may be required, keeping out all ashes or substances that would tend to porosity, and beating it down carefully with shovels so as to keep out the air as much as possible. The most abiding mischief is caused by the drainage from all "tank heaps." The sulphide becomes soluble, and is washed out by rain, &c., forming a yellow liquid, which gives off a well-known nauseous odour. The yellow coating that appears upon the surface of a heap of tank after exposure to the air consists mostly of free sulphur from the oxidation of iron sulphide.
The next operation is the boiling down of the soda liquors from the tanks. When they are drawn off from the balls, the liquors are run into a series of settlers, slightly warmed by waste beat, or a coil of steam piping, to prevent any deposit of crystals. Here any insoluble matter, mechani cally carried over, subsides, the tbllowing being about the composition of the residue : Calcium sulphide Iron compounds .. 6.00 Silica .. .. 26 . 00 Sodium, ditto .. Alumina .. .. 19 00 The settlers require clearing out about every five or six weeks, the residue lying upon the bottom being removed to the tanks, and mixed with fresh balls, so as to lose none of the soda. The clear, warm liquor is next run into the boiling-down pans, which have already been described, and exposed to the waste heat of the ball furnace. After a few hours' concentration, a barrowful or two of sawdust is thrown io, and as the body of liquor contracts a further quantity is run in, to keep the sides of the pan from burning. After about twenty-three hours the liquor is boiled down to a pasty mass. The doors are then removed, and the whole is raked out into the drainer placed in front, an operation requiring about a quarter of an hour. The doors are then fitted on and luted with moist clay, and a fresh charge of liquor is run in. The tanks, settlers, and boiling-down pans should be so arranged that the liquor runs from one to another without any pumping, &c., expense. As the salt crystallizes out and forms a crust over the liquor, a system of careful stirring should be commenced, and continued from time to time during the whole of the boiling-down process. As the mass gets more solid, the salts should be worked back from the fire, that no portion of them may become burned. If the supply of liquor runs short, one or more of the boiling-dowu pans must be filled with water, and the liquor reserved for the rest. As a rule, a band ball furnace will boil down rather more than the liquor from its own balls. The salts are allowed to remain in the drainer for-about twelve hours, during which time the greater part of the mother liquor drains out, runs along underneath the false bottom (see Figs. 243 and 245), and collects in the well at the low end. From here it is pumped from time to time back into the pan, or into a separate cistern if a particularly pure product be required. These mother liquors .contain, nearly all the absolute impurities from the liquor, the sodium salts consisting of earbonate, hydrate, sulphide, and sulphate. After being thoroughly drained, the contents of the drainer are wheeled away to a depot or to the finishing furnace. These "black salts " should present a mixed appearance of black and white salt, glistening with small crystals. If very fine in grain, and white or yellow in colour, the liquor has been impure, or the heat in the tanks too great. The salt that is taken from the furnace end is called " strong "; that from the further end " weak," and for the finishing operation a judicious mixture should be made. If the workman takes all weak salt, it will not "stand the fire ; "if all strong, it is found impossible to " clean " it. Upon this point more will be said hereafter. When the mother—technically termed "red "—liquors are separated from the salts—not pumped back into the pan—they are utilized in making caustic soda, or are worked up into an inferior greyish coloured carbonate. Separation of the red liquors is always adopted by manufacturers who work with revolvers, and by the majority of those who use hand furnaces. It is not, however, necessary, and the writer's experience is that a thoroughly good carbonate, testing 53 per cent., and pure white, may be made, even when the red liquors are pumped back.