592 POTASSIUM The salt has been recommended as a substitute for chlorate in pyrotechnic mixtures, because it contains more oxygen, and yet, on account of its greater stability, is a less dangerous ingredient. Bromide, KBr. This salt is formed when bromine is dissolved in caustic-potash ley. The reaction is quite analogous to that go ing on in the case of chlorine ; only the hypobromite (KBrO) first produced is far less stable than hypochlorite, and vanishes after short heating. The addition of bromine is continued until the liquid is permanently yellow and retains its colour after short heat ing. The solution is then evaporated to dryness and the bromatc decomposed by cautious heating. A small portion of the bromatc breaks up into K.,0 + Br 2 + 50 ; hence the residual bromide is con taminated with a little free alkali ; but this is easily set right by neutralizing its solution with hydrobromic acid. The salt crystal lizes in colourless transparent cubes, easily soluble in water. It is used in medicine for quieting the nerves, to cure sleeplessness, for instance ; also (internally) as a local anaesthetic preparatory to operations on the larynx or the eye. The dose of the pure (KI free) salt for adults can safely be raised to 2 grammes (about 30 grains). It is also used in photography. Iodide, KI. Of the very numerous methods which have been recommended for the preparation of this important salt the simplest (and probably the best) is to dissolve in a caustic-potash ley (which is dilute enough to hold the rather difficultly soluble iodate KI0 3 in solution) enough iodine to produce a permanent yellow colour (the iodine passes at once into 6KI + KI0 3 ; the hypo body KIO has no existence practically) and to deoxidize the iodate, which is done most conveniently by adding a sufficiency of powdered char coal to the solution, evaporating to dryness in an iron vessel, and heating the residue. The oxygen goes off as CO., at a lower tem perature than that which would be needed for its expulsion as oxygen gas. The residue is dissolved, and the solution filtered and evaporated to crystallization. The salt comes out in colourless transparent cubes, very easily soluble in even cold water. The commercial salt forms opaque milk-white crystals, which, as a matter of habit, are preferred to the clear salt, although they are produced by causing the salt to crystallize from a strongly alkaline solution and by drying the crystals (finally) in a stream of hot air, and although through the former operation they are at least liable to contain carbonate. Iodide of potassium acts far more powerfully on the human system than bromide, and therefore is administered in smaller doses. It is used against skin-diseases, and also for eliminating the mercury which settles in the system after long- continued administration of mercurial medicines. It is also used, far "more largely than the bromide, in photography. See PHOTO GRAPHY, passim. Sulphate (K 2 S0 4 ) used to be extracted from kainite, but the process is now given up because the salt can be produced cheaply enough from the muriate by decomposing it with its exact equi valent of oil of vitriol and calcining the residue. To purify the crude product it is dissolved in hot water and the solution filtered and allowed to cool, when the bulk of the dissolved salt crystallizes out with characteristic promptitude.- The very beautiful (anhydrous) crystals have as a rule the habitus of a double six-sided pyramid, but really belong to the rhombic system. They are transparent, very hard, and absolutely permanent in the air. They have a bitter salty taste. 100 parts of water dissolve at 12 100 C 8 36 10 26 parts of the salt. Sulphate of potash fuses at a strong red heat, and at this temperature volatilizes, for an alkaline salt, rather slowly. The chloride, weight for weight, volatilizes at ten times the rate (Bunsen). Sulphate of potash used to be employed in medicine, but is now obsolete. The crude salt is used occasionally in the manufacture of glass. Bisulphate (KHS0 4 ) is readily produced by fusing thirteen parts of the powdered normal salt with eight parts of oil of vitriol. It dissolves in three parts of water of C. The solution behaves pretty much as if its two congeners, K 2 S0 4 and H 2 S0 4 , were present side by side of each other uncombined. An excess of alcohol, in fact, precipitates normal sulphate (with little bisulphate) and free acid remains in solution. Similar is the behaviour of the fused dry salt at a dull red heat; it acts on silicates, titanates, &c., as if it were sulphuric acid raised beyond its natural boiling point. Hence its frequent application in analysis as a disintegrating agent. For the following potash salts we refer to the articles named : CJiromates, see CHROMIUM; Cyanide and Fcrrocyanidc, PRUSSIC ACID; Chloroplatinate, PLATINUM (supra, p. 192); Nitrate, NITRO GEN (vol. xvii. p. 518) ; Phosphates, PHOSPHORUS (vol. xviii. pp. 818-19); Oxnlates, OXALIC ACID; Sulphides &nd Sulphites, SULPHUR; Silicates, GLASS (vol. x. p. 655 sq. } and SILICA ; Tartrates, TA RTAUIC ACID. For potash salts not named, see the handbooks of chemistry. Rutndium and Caesium. When Bunsen and Kirchhoff in 1860 applied their method of spectrum analysis to the alkali salts which they had extracted analytically from Diirkheim mineral water, they obtained a spectrum which, in addition to the lines characteristic for sodium, potassium, and lithium, exhibited two blue lines which were foreign to any other spectrum they had ever seen. They accord ingly concluded that these lines must be owing to the presence of a new alkali metal, which they called "caesium." Bunsen at once resumed the preparation of the mixed alka line salt with 44,000 litres of Diirkheim water, with the view of isolating the caesium in the form of a pure salt ; and he was more than successful for the new alkali salt, after elimination of all the ordinary alkali metals, proved to be a mixture of the salts of two new alkali metals, which he succeeded in separating from each other. For one he retained the name already chosen ; the other he called "rubidium," on account of the presence in his spectrum of certain characteristic red lines. Since Bunsen s time these two metals have been discovered in a great many native potassiferous materials minerals, mineral waters, plant ashes, &c. but in all cases they form only a small fraction of the alkali, the caesium in general amounting to only a fraction of even the rubidium. One solitary exception to both rules is afforded by a rare mineral called "pollux," which is found only on the island of Elba. Plattner analysed this mineral in 1846 and recognized it as a compound silicate of alumina, oxide of iron, soda, potash, and water ; but his quantitative analysis came up to only 92 75 per cent., and he could not account for the 7 25 per cent, of loss. After Bunsen s discovery Pisani analysed the mineral again, and he found that it contained no potash at all, but, instead of it, a large percentage (34 1) of cassia. Recalculating Plattner s analy sis on the assumption that the presumed chloroplatinatc of potassium was really chloroplatinate of caesium, he found that the corrected numbers did add up to near 100 and agreed with his own. Rubidium, singularly, is absent from this mineral. That both rubidium and ca?sium are contained in sea water might well be taken for granted ; but it is worth while to state that Schmidt of Dorpat actually proved the presence of rubidium, and even determined it quantita tively. For the preparation of rubidium compounds one of the best materials is a mixture of alkaline salts, which falls as a bye-product in the industrial preparation of carbonate of lithia from lepidolite. A supply of this salt mixture which Bunsen worked up contained 20 per cent, of chloride of rubidium, 33 of chloride of potassium, and 36 of common salt, but very little cresium ; his supply came from the Saxon or Bohemian mineral. The lepidolite of Hebron, Maine, United States, on the other hand, is rich in caesium. Another practically available source for crcsium is the mother- liquor salt of Nauheim in Germany. It yielded to Bb ttcher 1 per cent, of its weight of the chloroplatinate PtCl 6 Cs.,. Bunsen s method for the extraction of the two rare potassium metals from a given mixture of alkaline salts is founded upon the different solubility of the several alkaline chloroplatinates. Accord ing to him 100 parts of water dissolve at C. 20 C. ,, 100 C. Potassium 0-74 1-12 5-13 Rubidium 0-13 0-14 0-03 Cxsium 0-024 0-079 0-377 parts of the several salts. The chloroplatinates of sodium and lithium are easily soluble even in cold water, so that chloride of platinum does not precipitate these two metals at all. Hence, supposing we boil a given mixture of chloroplatinates of potassium and (say) rubidium with a quantity of water insufficient to dissolve the whole, part of both salts will dissolve ; but the residual chloro platinate will be richer in rubidium than the dissolved part. And supposing, on the other hand, we add to a mixed solution of the two chlorides a quantity of chloroplatinic-acid solution insufficient to bring down the whole of both metals, the rubidium will accumu late in the precipitate and the potassium in the solution. It is also easily understood that, if the amount of reagent added falls short even of that which would be needed by the rubidium if present alone, a very nearly pure PtCl fi ftb 2 may be expected to come down. Any dry chloroplatinate is easily reduced to a mixture of metallic platinum and alkaline chloride by the simple operation of heat ing in hydrogen to about 300 C. The chloride can be dissolved
out, and thus again made amenable to fractional precipitation by