480 MANGANESE tion of the fluoride with sodium in a clay crucible) is not manganese at all, but a silicide of the metal. Hugo Tamm, who endeavoured to work out a process for the manu facture of the metal, gives the following process : 11 parts of good pyrolusite is mixed with 1 part of lamp-black and 6 parts of a flux consisting of 20 parts of lead-free bottle-glass, 7 parts of quick lime, and 7 of fluor-spar; and the mixture is strongly heated, in a graphite crucible coated over with a mixture of 3 parts of graphite and 1 part of clay, by means of a blast-furnace. There is formed a regulus covered by a green silicious slag containing much protoxide of mapganese. The " raw " manganese thus pro duced is contaminated with about 1 per cent, each of iron, silicon, and carbon, and traces of sulphur, phosphorus, calcium, and aluminium. The green slag in subsequent operations is substituted for part of the white flux. The raw metal, when re-fused with about one-third of its weight of manganous carbonate, yielded a regulus which contained 99 9 per cent, of the metal, the remaining ^ per cent, consisting of carbon, silicon, and iron. Manganese metal is grey, like cast iron (Deville s had a reddish hue like bismuth) ; its specific gravity is about 8 ; it is hard and brittle, and about as difficult to fuse as wrought iron. It readily tarnishes in ordinary air ; even pure water, and much more dilute acid, attack it with evolution of hydrogen and formation of manganous (MnO) hydrate or salt. It is worth stating that neither MnO nor MnCl 2 is reducible at a red heat by hydrogen gis ; yet Bunsen succeeded in obtaining metallic manganese by the electrolysis of a concentrated solution of the chloride, using a strong current and a negative electrode of very small area. Oxides. Pure peroxide can be obtained artificially by keeping the pure nitrate at 200 C. But really pure nitrate is hard to procure. Perhaps the only method for obtain ing a really pure preparation is Volhard s : 10 grammes of " pure " (iron- and cobalt-free) manganous sulphate is dissolved in half a litre of water and 100 cm. of nitric acid of 1 "2 specific gravity; the solution is heated to boiling, and strong solution of permanganate of potash added until the MnO is nearly but not quite down, and the mixture kept for a while on a water-bath. The precipitate of binoxide formed (according to equation Mn 2 O 7 + 3MnO = 5MnO.,) is washed, first with dilute nitric acid, then with water, and dried (when it retains some combined water). When binoxide of manganese is heated to redness in vacuum, air, oxygen, nitrogen it loses oxygen with formation of lower oxides. This phenomenon was investigated by W. Dittmar, who found that, when the binoxide is heated in a constantly renewed atmosphere, the result, for a given temperature, depends only on the partial tension of the oxygen in that atmosphere. 1 ure (brown-red) Mn 3 4 remains when this tension is less, while (black) Mn. 2 0.j remains when the tension is greater than a certain limit value p. In Dittmar s experiments (which were all made at a temperature somewhat above the melting point of sterling silver), the value p was found to lie close to 26 cm. of mercury. An exact determina tion of this critical point was not possible, because the temperature was not perfectly constant, and an increase in temperature is equi valent to a diminution in the partial tension of the-oxygen. Hence, supposing the oxide Mn 2 3 to be heated, say in vacuum and within a close apparatus, it will give off oxygen at any temperature greater than a certain minimum < , and at any temperature t Q + At the gas- evolution will come to a stop as soon as the tension of the gas has come up to the critical value p corresponding to this t^ + At, p increasing with At. The protoxide, MnO, is most readily obtained by heating any higher oxide to redness in a current of hydrogen gas, as a dull green powder which gets readily discoloured by oxidation in ordinary air. It is not acted on by water, but readily dissolves in aqueous acids, with formation of manganous salts. The sulphate, MnS0 4 , is prepared by making pyrolusite into a paste with concentrated sulphuric acid and then heating in a crucible to dull redness until vapours of the acid cease to come off. The ferric and aluminic sulphates (originally present) are now, at least mostly, decomposed and reduced to insoluble basic salts, so that the residue when treated with water and filtered may yield a solution free of these impurities, and, of course, of baryta. Should any iron or alumina be left it is easily eliminated by digestion with a little carbonate of manganese (prepared from a small portion of the solution by precipitation with carbonate of soda) and filtration. Cobalt and nickel, if present, can be removed by fractional precipitation with sulphide of sodium (or H 2 S in the presence of MnC0 3 ?) ; the black sulphide of Co or Ni comes down flrst, the (flesh-coloured) MnS afterwards. But lime, magnesia, and alkalies (which are frequently present) arc very difficult to get rid of. Compare the section on binoxide. The salt, according to the temperature at which it crystallizes, takes tip 7 or 5 or 4 or even 3 or 1 H 2 0. Crystallized sulphate of manganese generally exhibits a rose-red tint ; but this is owing to the presence of a tract; of manganic salt (if not to cobalt salt). The pure salt is colourless. The chloride, MnCl 2 . The crude chloride contained in the preparation of chlorine from the binoxide and muri atic acid is purified by methods analogous to those explained for the sulphate. This (very soluble) salt crystallizes at 15-20 C., with 4H 2 O. To obtain real MnClo, the salt must be dehydrated in a current of dry hydrochloric acid gas. The carbonate, MnC0. 5 , is obtained by precipitating the solution of the sulphate or chloride by excess of carbonate of soda on boiling. It is a white precipitate, soluble in 8000 parts of water, which, when dried in the air, gets slightly oxidized with discoloration. Far more oxidizable is the hydrate, Mn(OH) 2 , as ob tained by precipitation of manganous solution by caustic alkalies. In the presence of an excess of alkali or other strong soluble base (such as lime, for instance) the oxidation progresses very rapidly, with formation, ultimately, of a black manganite MnO 2 .RO (e.g., MnOoCaO). Tim is the rationale of the famous " Weldon Process " for the recovery of the " manganese " from chlorine liquors (see BLEACH ING POWDER). A mixed solution of chloride of manganese and sal-ammoniac, when mixed with ammonia, gives no pre cipitate ; but the alkaline liquor readily absorbs oxygen from the atmosphere with formation of a brown precipitate of a higher oxide. If, immediately after addition of the ammonia, the excess of volatile alkali is chased away by boiling, the resulting (neutral or slightly acid) liquor remains clear, even in air. Hereupon is founded a method for the separation of ferric iron and alumina from man ganese. Manganic salts, i.e., salts of Mn 2 3 , arc produced only under very special conditions. Solutions containing the sulphate and a phosphate respectively are obtained by heating finely divided pyro lusite with strong sulphuric or phosphoric acid. Both products dissolve in water with formation of intensely purple solutions, which, however, are very unstable, showing a great tendency to pass into the manganous condition. Any manganic salt when boiled with hydrochloric acid gives manganous salt with evolution of chlorine. This tendency separates them sharply from the corre sponding compounds of iron. Manganates and permanganates (compounds with bases of the hypothetical oxides MnO 3 and Mn 2 O 7 ). The most important of these is the manganate of potash, K 2 MuO 4 . Four parts of very finely divided binoxide of manganese and 3 1 parts of chlorate of potash are evaporated to dry- ness with the solution of 5 parts of caustic potash, and the residue ignited (not fused) in platinum crucibles until all the chlorate is decomposed. The intensely green mass, containing large excess of caustic alkali, dissolves in water into an intensely green solution from which crystals of the salt K 2 MnO 4 canbeobtained; but when the alkali is neutral ized by an acid, the liquor turns intensely purple with for mation of permanganate and a precipitate of alkaliferous binoxide : 2K,MnO 4 + 2H O = 2KMnO 4 + 2KHO + 211 and 2H + K 2 M"nO 4 = H 2 O + K 2 MnO 3 . The purple solution when alkalinized with potash contaminated with organic matter reassumes its original green colour (whence its old name of " chameleon minerale "). For the preparation of permanganate of potash it is best to pass chlorine into the green solution, when the whole of the manganese assumes
the permanganate form: K 2 Mn0 4 + Cl = KC1 + KMnO 4 .