298

ALKALIMETRY.

[appendix.

The weight of 2 atoms of green sulphate of iron is 278 = (139 X 2), consisting of 2atoms of protoxide = 72, X 2 of sulphuric acid = 80, X 14. of water = 126; in all = 278;and this weight is equivalent to 36 of chlorine, to 8 of oxygen, and to 44 of peroxide ofmanganese.* Therefore, if we take a solution of copperas, containing 278 grains in1,000 water grain measures, that volume of liquid will represent, by the conversion ofits protoxide into peroxide, exactly one atom, either of peroxide of manganese =44grains, or 1 atom of chlorine = 36. Hence the following plan of research :—

Into the flask or phial c of my chlorometric apparatus (fig. 188), put 100 grains of the

manganese to be tested, and into the globes a, b, pour outan alkalimetrical tube charged with 1,000 grain measuresthe above equivalent copperas solution, from 200 to 500grain measures, according to the supposed quality of themanganese; then introduce through the funnel d , some hydro-chloric acid of known specific gravity (suppose 1*1), contain-ing nearly 20 per cent, of chlorine, also from a chargedalkalimetrical tube, and apply gentle heat to the bottom ofthe flask by placing it in a capsule of water standing over

a spirit-lamp. The chlorine evolved will rise up throughthe tube /, which passes merely beyond the cork, and willenter into the solution in b and A, converting it into redsulphate. Have ready some dry paper imbued with solutionof red ferrocyanide of potassium (red prussiale of iron).Dip a slip of whalebone into the liquor in the globe a,through the funnel e (represented in the figure rather toohigh above the globe), and touch the paper with its point.As long as it forms a blue spot, some of the iron still existsas black oxide, and the process is to be urged by the ad-dition of a little more hydrochloric acid to the manganese,as long as chlorine gas continues to be disengaged, and whileit maintains the level of the liquor in a above that in b.Whenever the liquor, by the reaction of the chlorine, ceasesto stain the test-paper blue, more of the solution from thegraduated tube must be added till it begins to do so. Bythe cautious administration of the hydrochloric acid on theone hand, and of the copperas liquor on the other, the termof saturation will be arrived at in a few minutes. Themanganese has then produced all the chlorine which it canyield. The number of water grain measures, of the liquor,or degrees of its alkalimeter scale being multiplied by 44,will give a product denoting the per-centage of pure manga-

nese present in the sample; or being multiplied by 36, a product which will denotethe quantity of chlorine by weight which 100 grains of it can serve to generate.

Since one atom of pure manganese (44 grains), in producing 36 grains of chlorine,consumes 2 atoms = 74 grains of hydrochloric acid, the quantity of this acid expendedfrom the graduated tubes, beyond the due proportion of chlorine obtained, will showhow much of the acid is unprofitably consumed by foreign substances in the manganese.In fact, every grain of chlorine should, with pyrolusite, be generated by an expenditureof little more than 2 grains of real muriatic acid, or 10 grains weight of the dilute acid, =about 9 grain measures of the graduated tube. Liquid hydrochloric acid of spec. grav.1*093 contains in 1,000 grain measures exactly 200 grains of real acid. Hence 100 grainsof pure pyrosulite should produce about 82 grains of chlorine, and consume about 169of real muriatic acid = 845 grain measures of liquid acid, spec. grav. 1*093. Insteadof taking 100 grains of manganese as the testing dose, 10 or 20 grains may be taken,according to the dimensions of the apparatus and the exactness of the operator.

But if it be wished to obtain direct per-centage of manganese by the graduated tubeswithout the trouble of reduction, then for a dose of 10 grains take a solution of fresh greencopperas (free from adhering moisture), containing 632 grains in 10,000 grain measures.Proceed as above directed. If the manganese be a pure peroxide, 10 grains of

♦Berzelius , in tho 4th edition of his Lehrburh , rates the atom of the green sulphate of iron (ferroussulphate) at 12iM3, hydrogen = 1, and considers it. after-Mitscherlich, to contain only 6 atoms of water-I have ascertained, by the most careful experiments, that it contains 7 atoms of water ; and that }■*grains of it, or 138-44 (Berzelius ) are equivalent to 1 atom of chlorbariunt, and to very nearly 40 grainof peroxide of iron. . . ^

This remarkable error has probably arisen from an attempt to measure the proportion of water in tnsalt from its loss of weight by desiccation. But I have found it impossible by this means to expel mothan 6 atoms of water without causing partial decomposition of the salt by disengagement of sulpnyacid. The copperas so dried acquires such an affinity for water, that it absorbs fully one tenth oiweight of moisture from the atmosphere in the course of an hour.