chalk, as this would not give a constant stream of gas). B is a smaller flask containing concentrated sulphuric acid. The flask A is closed with a double-perforated cork, into the one aperture of which is inserted a funnel-tube (a), which reaches nearly to the bottom of the flask; into the other perforation is fitted a tube (b), which serves to conduct the evolved gas into the sulphuric acid in B, where it is thoroughly freed from moisture. The tube c conducts the dried gas into the reductiontube C, of which Fig. 28 gives a representation on the scale of one-third of the actual length. The tubes which I employ for the purpose in own experiments, have an inner diameter of eight millimètres.

my

When the apparatus is fully prepared for use, triturate the perfectly dry sulphide of arsenic, or arsenite in a slightly heated mortar with about twelve parts of a well-dried mixture consisting of three parts of carbonate of soda and one part of cyanide of potassium. Put the powder upon a narrow slip of card-paper bent into the shape of a gutter, and push this into the reduction-tube down to e; turn the tube now halfway round its axis, which will cause the mixture to drop into the tube between e and d, every other part remaining perfectly clean. Connect the tube now with the gas-evolution apparatus, and evolve a moderate stream of carbonic acid, by pouring some hydrochloric acid into the flask A. Heat the tube C in its whole length very gently with a spirit-lamp, until the mixture in it is quite dry; when every trace of water is expelled, and the gas-stream has become so slow that the single bubbles pass through the sulphuric acid in B at intervals of one second, heat the reduction-tube to redness at c, by means of a spirit or gas-lamp ; when c is red-hot, apply the flame of a second gas or larger spirit-lamp to the mixture, proceeding from d to e, until the whole of the arsenic is expelled. The far greater portion of the volatilized arsenic recondenses ath, whilst a small portion only escapes through i, imparting to the surrounding air the peculiar odor of garlic. Advance the flame of the second lamp slowly and gradually up to c, by which means the whole of the arsenic which may have condensed in the wide part of the tube is driven to h. When you have effected this, close the tube at the point i by fusion, and apply heat, proceeding from i towards h, by which means the extent of the mirror is narrowed, whilst its beauty and lustre are correspondingly increased. In this manner perfectly distinct mirrors of arsenic may be produced from as little as the th part of a grain of tersulphide of arsenic. No mirrors are obtained by this process from tersulphide of antimony, nor from any other compound of antimony.

13. If arsenious acid or one of its compounds is exposed on a charcoal support to the reducing flame of the blowpipe, a highly characteristic garlic odor is emitted, more especially if some carbonate of soda is added to the examined sample. This odor has its origin in the reduction and re-oxidation of the arsenic, and enables us to detect very minute quantities. This test, however, like all others that are based upon the mere indications of the sense of smell, cannot be implicitly relied on.

§ 132.

e. ARSENIC ACID (AS 0.).

1. Arsenic acid is a transparent or white mass, which gradually deliquesces in the air, and dissolves slowly in water. It fuses at a gentle red heat without suffering decomposition; but at a higher temperature it is resolved into oxygen, and arsenious acid, which volatilizes. It is highly poisonous.

2. Most of the arsenates are insoluble in water. Of the so-called neutral arsenates those with alkaline bases alone are soluble in water. Most of the neutral and basic arsenates can bear a strong red heat without suffering decomposition. The acid arsenates lose their excess of acid upon ignition, the free acid being resolved into arsenious acid and

oxygen.

3. Hydrosulphuric acid fails to precipitate alkaline and neutral solutions of arsenates; but in acidified solutions it produces a yellow precipitate of PENTASULPHIDE OF ARSENIC (AS S). This precipitate never forms instantaneously, and in dilute solutions frequently only after the lapse of a considerable time (twenty-four hours, for instance). Heat promotes its separation. The pentasulphide of arsenic manifests the same deportment as the tersulphide with the various solvents and decomposing agents mentioned in the preceding paragraph. If a solution of arsenic acid, or of an arsenate, is mixed with sulphurous acid, or with sulphite of soda and some hydrochloric acid, the sulphurous acid is converted into sulphuric acid, and the arsenic acid reduced to arsenious acid; application of heat promotes the change. If hydrosulphuric acid is now added, the whole of the arsenic is thrown down as tersulphide.

4. Sulphide of ammonium converts the arsenic acid in neutral and alkaline solutions of arsenates into pentasulphide of arsenic, which remains in solution as ammonio-pentasulphide of arsenic (pentasulphide of arsenic and sulphide of ammonium). Upon the addition of an acid to the solution, this double sulphide is decomposed, and pentasulphide of arsenic precipitates. The separation of this precipitate proceeds more rapidly than is the case when acid solutions of arsenates are precipitated with hydrosulphuric acid. It is promoted by heat.

5. Nitrate of silver produces under the circumstances stated § 131, 6, a highly characteristic reddish-brown precipitate of ARSENATE OF SILVER (3 Ag 0, As 0), which is readily soluble in dilute nitric acid and in ammonia, and dissolves also slightly in nitrate of ammonia. Accordingly, if a little of the precipitate is dissolved in a large proportion of nitric acid, neutralization with ammonia often fails to reproduce the precipitate.

6. Sulphate of copper produces under the circumstances stated § 131, 7, a greenish-blue precipitate of ARSENATE OF COPPER (2 Cu O, H 0, As 0,).

7. With zinc in presence of sulphuric acid, with copper, with cyanide of potassium, and before the blowpipe, the compounds of arsenic acid comport themselves in the same way as those of arsenious acid. If the reduction of arsenic acid by zinc is effected in a platinum dish, the platinum does not turn black, as is the case in the reduction of antimony by zinc (§ 130, 8).

8. If a solution of arsenic acid, or of an arsenate soluble in water, is

added to a clear mixture of sulphate of magnesia, chloride of ammonium, and a sufficient quantity of ammonia, a crystalline precipitate of ARSENATE OF AMMONIA AND MAGNESIA (2 Mg O, N H ̧ 0, As O̟ ̧ + 12 aq.) separates; from concentrated solutions immediately, from dilute solutions after some time.

§ 133.

4

Recapitulation and remarks.--I have again made the separation and positive identification of the oxides belonging to the second division of the sixth group the object of a most careful study, and my endeavors have been crowned with complete success. I will here describe first the different ways best adapted to effect the detection or separation of tin, antimony, and arsenic, when present together in the same compound or mixture, and afterwards the most reliable means of distinguishing between the several oxides of each of the three metals.

1. If you have a mixture of sulphide of tin, sulphide of antimony, and sulphide of arsenic, triturate 1 part of it, together with 1 part of dry carbonate of soda, and 1 part of nitrate of soda, and transfer the mixed powder gradually to a small porcelain crucible containing 2 parts of nitrate of soda kept in a state of fusion at a not over-strong heat; oxidation of the sulphides ensues, attended with slight deflagration. The fused mass contains binoxide of tin, arsenate and antimonate of soda, with sulphate, carbonate, nitrate, and nitrite of soda. You must take

care not to raise the heat to such a degree, nor continue the fusion so long, as to lead to a reduction of the nitrite of soda to the caustic state. Treat the fused mass, poured out upon a piece of porcelain, with cold water until it is completely softened; then filter the fluid off from the undissolved residue, which contains the binoxide of tin and antimonate of soda nearly unacted on. Mix the filtrate, which contains the arsenate of soda and the other salts, with nitric acid to distinctly acid reaction, then with a sufficient proportion of solution of nitrate of silver; a precipitate of chloride of silver forms (if the reagents employed or the precipitated sulphides contained a chlorine compound) and some nitrite of silver. Filter, and carefully add to the filtrate dilute solution of ammonia, whereupon the characteristic reddish-brown precipitate of arsenate of oxide of silver will make its appearance, first in the uppermost stratum of the fluid where the solution of ammonia comes first into contact with it, but subsequently, upon complete neutralization of the free acid, in every part of the fluid.

Wash now the filter containing the residuary binoxide of tin and antimonate of soda once with water, then three times with a mixture of equal parts of water and spirit of wine, dry, incinerate,* and put the ash into a tube of difficultly fusible glass, sealed at one end, measuring some eight or ten centimètres in length, and having an inner diameter of from five to seven millimètres; add to the ash in the tube four times the quantity of cyanide of potassium, and heat over a Berzelius or gas-lamp. This effects the reduction of the binoxide of tin and

* With small quantities of substance this may be done most conveniently by twisting the little filter together, inserting it into a spiral coil of platinum wire, and holding this in the outer mantle of a flame.

Had not the nitrate of soda been removed by washing, as directed, this part of the process would be attended with an explosion.

antimonic acid, and a fused mass is produced which appears gray from the minutely divided reduced metals in it. If the tube is now dipped whilst still red-hot into a test-tube filled with cold water, the part containing the fused mass will crack off, and the broken fragments sink to the bottom of the water. Upon application of heat, the fused mass will readily dissolve, leaving the metals behind, which may now by repeated boiling with water and decantation be freed from the last traces of other matters still adhering to them. Heat the residuary metallic mass with hydrochloric acid just to the boiling point, when the tin or a portion of it will dissolve to protochloride, with evolution of hydrogen gas; by means of chloride of mercury or a mixture of ferricyanide of potassium and sesquichloride of iron you may now easily, and with positive certainty, detect the protochloride of tin in the solution, even though only a minute trace of the metal be present. Upon heating the residue remaining in the tube repeatedly with hydrochloric acid, the whole, or nearly the whole of the tin present is dissolved out, and the black-colored antimony alone is left undissolved. Upon heating this residue with hydrochloric acid, with addition of one or two drops of nitric acid, it dissolves, and if you now add to the solution hydrosulphuric acid, an orange-colored precipitate of tersulphide of antimony is obtained. If only slight traces of antimony are present, the color of this precipitate is not very distinct, owing to imperfect separation from the tin. Let, therefore, the precipitate of the impure tersulphide of antimony settle, decant the clear fluid, dissolve the precipitate in a little boiling hydrochloric acid, concentrate the solution somewhat, and then test in the lid of a platiuum crucible with zinc, which will at once remove all doubt, showing even the minutest trace of antimony.

2. Instead of incinerating the filter with the binoxide of tin and antimonate of soda, you may also dissolve the precipitate on it in a little hot hydrochloric acid, wash the filter with water, concentrate the fluid somewhat, and test with zinc in a small platinum dish, when the antimony will at once reveal its presence by staining the platinum brownish-black; to detect the tin, wait until the little slip of zinc is almost entirely dissolved, then decant the fluid, heat the residue with some hydrochloric acid, and test the solution for protochloride of tin as in 1.

3. If the quantity of binoxide of tin and antimonate of soda is somewhat larger, the two bodies may be separated also by boiling with concentrated solution of soda, which dissolves the binoxide of tin, leaving the antimonate of soda unaffected. It is advisable to add, after the process of boiling, some spirit of wine, say about by volume of the entire fluid, to guard against the antimonate of soda being dissolved; for the same reason the precipitate must be washed with equal volumes of water and spirit of wine. If the alkaline solution is now evaporated to drive off the spirit of wine, hydrochloric acid then added, and lastly, hydrosulphuric acid, the tin is precipitated as sulphide; the antimony may in like manner be precipitated from the solution of the antimonate of soda in hydrochloric acid.

4. If the mixed sulphides are treated with fuming hydrochloric acid, the sulphide of antimony and sulphide of tin dissolve, whilst the sulphide of arsenic is left nearly altogether undissolved. If the residuary sulphide of arsenic is treated with ammonia, and the solution obtained evaporated, after adding to it a very little carbonate of soda, an arsenical mirror may readily be produced from the residue with cyanide of potassium and

carbonate of soda in a stream of carbonic acid gas. The solution, which contains the tin and antimony, may be treated as directed in 2. Or, if there is a considerable excess of antimony, the solution may be mixed with sesquicarbonate of ammonia in excess, and boiled. A considerable proportion of the antimony present dissolves in this process, whilst binoxide of tin mixed with a small quantity of teroxide of antimony remains undissolved. The presence of the tin may now be the more readily proved by the method given in 1 (Bloxam).

5. If sulphide of antimony, sulphide of tin, and sulphide of arsenic are dissolved in sulphide of potassium, a large excess of a concentrated solution of sulphurous acid added, the mixture digested for some time in the water-bath, then boiled until all sulphurous acid is expelled, and filtered, the filtrate contains all the arsenic as arsenious acid (which may be precipitated from it by hydrosulphuric acid), whilst tersulphide of antimony and bisulphide of tin are left behind undissolved (Bunsen). After dissolving the residue in hydrochloric acid, the antimony and tin may be detected as directed in 2.

6. In the analysis of alloys, binoxide of tin and teroxide of antimony are often obtained together as a residue insoluble in nitric acid. If the quantities are considerable, the teroxide of antimony may be extracted with tartaric acid, and the solution tested for it with hydrosulphuric acid; the binoxide of tin may be reduced with cyanide of potassium. For minute quantities I recommend the method described in 1. An absolute separation of the two oxides may be effected by fusing them with hydrate of soda in a silver crucible, treating the mass with water, and adding one-third (by volume) of spirit of wine. The binoxide of tin is by this means obtained in solution as a compound of binoxide of tin and soda, whilst the antimonate of soda is left undissolved (H. Rose). 7. For the most accurate way of separating antimony and arsenic, and distinguishing between the two metals, viz., by treating with hydrosulphuric acid the mirror produced by Marsh's method, and separating the resulting sulphides by means of hydrochloric acid gas, I refer to § 131, 10. Antimony and arsenic may, however, when mixed together in form of hydrides, be separated also in the following way: Conduct the gases mixed with an excess of hydrogen, first through a tube containing glass splinters moistened with solution of acetate of lead, to retain the hydrochloric and hydrosulphuric gas, then in a slow stream into a solution of nitrate of silver. All the antimony in the gas falls down as black antimonide of silver, whilst the arsenic remains in solution as arsenious acid, and may, after precipitation of the excess of silver by hydrochloric acid, be readily detected in the fluid.

Protoxide and binoxide of tin may be detected and identified in presence of each other, by testing one portion of the solution containing both oxides for the protoxide with chloride of mercury, terchloride of gold or a mixture of ferricyanide of potassium and sesquichloride of iron, and another portion for the binoxide, by pouring it into a concentrated solution of soda.

Teroxide of antimony in presence of antimonic acid may be identified by the reaction described in § 130, 9. Antimonic acid in presence of teroxide of antimony, by heating the teroxide suspected to contain an admixture of the acid, but without any other admixture, with hydrochloric acid free from chlorine, and iodide of potassium free from iodate of potassa. If antimonic acid is present, iodine will separate, which, in