The still is firet charged with hydrochloric acid, and the settled mud run in through the funnel. Tllie mud should contain 5 lb. of peroxide of manganese to the cubic foot, the chloride of calcium being drawn off as closely as possible. It dissolves in the acid far more reaclily than native manganese ore, and requires comparatively little steam. About four hours is sufficient to work off a charge. The liquors are then run off through the large tap to the neutralizing well and treated with chalk in the manner before described. The amount of free acid, if the operation in the still hoe been carefully conducted, will not exceed 5 lb. per cubic foot of liquor. The more acid liquors from the " native" stills, if they are in operation, are now run into the mud stills, and so on to the neutralizing well. In this way the gre,ater portion of the free acid contained iB utilized in the treatment of the mud.
Although theoretically the whole of the manganese is recovered by this process, in practice there is a loss of about 5 per cent. The greater part of this loss iB incurred in the liquor settlers. There is always an appreciable amount of mud escaping with the liquors from the stills into the neu tralizing well, and this is pumped up with a quantity of chalk and sulphate of lime into the settlers. The deposit from here is regularly removed and washed in a special tank (omitted in the drawing), the resulting liquors being returned to the neutralizing wells ; but nevertheless a considerable loss of both mud and chloride of manganese is incurred, and it would not pay to carry this part of the process out more delicately.
Two tests must be regularly made of the mud of the Weldon process—for peroxide and for bases." In the former a sample is taken as the mud is run from the oxidizer, and to 1 cubic in. is added a solution, in alight excess, of a known quantity of iron protosulphate in hydrochloric acid. The mud dissolves promptly to a brownish yellow liquid, and the amount of iron salt left unoxidized is then determined by e. standard solution of bichromate of potash, which is added until a drop of the liquid taken out upon a glass rod, ceases to strike blue with ferticyanide of potassium. If after adding the iron solution the mud does not completely dissolve, it is because an excess of protosul phate is not present. The original weight of iron salt being known, and the amount left unoxidized ascertained, the difference gives the amount acted upon, and this quantity divided by 25.88, gives the pounds of peroxide in 1 oubic foot of the mud.
After ascertaining the amount of ferrous sulphate peroxidized by the Mn02 in 1 cubic in. of the mud, the amount dime is readily obtained. To the same amount of the nand sample a solution of a known weight of oxalic acid is added—in excess. The mixture is then heated gently, and the excess of oxalic acid determined by a standard solution of carbonate of soda. The difference between this and the original weight of oxalic acid gives the amount decomposed and neutralized by the mud. Then, as the number of grains of Fe2SO4 + 7 11.20 (crystallized ferrous sulphate) oxidized is to the number of grains of oxalic acid decomposed, eo is 100 to a figure in column A of the subjoined table, against which in column B is the equivalent of base per equivalent of manganese peroxide.
Besides these two tests, that must be made from every batch of rand in order to carry the process on well, it is usual from time to time to test for total manganese, by oxidizing the whole of the manganese present with bleaching powder, and ascertaining, in the manner above described, the amount of ferrous sulphate oxidized by the sample. A convenient quantity of mud is the 1 cubic inch
already spoken of. This is dissolved in hydrochloric acid, in the least possible excess, which excess is neutralized with soda. The solution of mud iB then heated and oxidized with bleaching-powder solution, complete conversion of all the manganese into peroxide being obtained when a slight purple coloration betrays the formation of a permanganate. After filtration and washing, the precipitate is added to the solution of a known quantity of ferrous sulphate in hydrochloric acid, and the total manganese calculated from the amount of ferrous salt acted upon.
There can be no doubt about the very great success of the Weldon process. Not only is the manganese recovered at a low cost, but a better yield of bleach is obtainable than by the old process.
One great reason of this is the facility for accurately gauging the quantity of acid necessary for the operation in the stills.
Many plans have been proposed for utilizing the chlorine wasted as chloride of calcium—two thirds of the amount originally contained in the hydrochloric acid. Foremost among these is Weldon's own magnesia process. The liquor from the stills is neutralized with " Greek stone," a native carbonate of magnesium—preferably calcined and powdered—and the mixed chlorides of manganese and magnesium pumped into settlers, where the oxide of iron, alumina, and sulphate of lime are deposited. From here the clear liquor is run into an iron pot and evaporated by underneath heat until a temperature of 320° F., or thereabouts, is obtained. It is then drawn off into a double bedded furnace, the two beds being separated by a sliding door in a manner similar to a " pot " and " roaster " arrangement. Here evaporation is continued to dryness, gaseous hydrochloric acid being driven off, together with a considerable amount of chlorine. The mixed gases are passed up a tower packed with coke, similar in arrangement to the ordinary hydrochloric acid condenser, and the resulting liquid acid utilized in the afterpart of the process for the evolution of chlorine. The residue left in the furnace after being evaporated, is broken up by a slice into cakes and trans ferred to the second bed. Hero it is heated gently, so that the mass may not fuse but remain open, and a carefully regulated amount of air introduced. Oxidation of the manganese protoxide—formed when the chlorine is driven off from the mixed chlorides of manganese and magnesium—proceeds! regularly, and a peculiar compound which has been termed manganite of magnesium (Mglin0g) finally obtained. This is put into the stills and treated with the hydrochloric acid from the con denaer, as in the ordinary bleach process. By carefully regulating the propOrtions of manganite and acid, an almost neutral still liquor is obtained, which is treated with a little magnesia, settled, evaporated, and worked over again in the rnanner deacribed. The gaseous product of the first part of the furnacing—the evaporation—consiste chiefly of hydrochloric acid, a sufficient amount of water being present. Afterwards a more or less dilute chlorine is evolved. This may be conveniently passed into milk of lime to form a bh. aching liquor, or may be mixed with the "maiden " gas from the stills.