The determination of silver in alloys may be effected with great rapidity and accuracy by using the standard solution of chloride of sodium, that is, a solution; a certain number of measures of which will precipitate, as chloride, a knowu quantity of silver.

The solution to be analyzed is mixed with some diluted nitric acid, gently heated, and the standard solution of chloride of sodium gradually added, with frequent agitation, until a drop fails to produce a fresh precipitate of chloride of silver; the number of measures of solution which have been used is then read off, and the quantity of silver calculated.

SEPARATION OF GOLD, SILVER, AND COPPER.

ANALYSIS OF STANDARD GOLD.

§ 428. The analysis of alloys of gold, silver, and copper, in the moist way, is attended with some difficulty.

The general method consists in laminating the alloy, and boiling with nitrohydrochloric acid; if the metal becomes coated with chloride of silver, it must be carefully removed, and treated with ammonia, so as to expose a fresh metallic surface, the ammoniacal solution being afterwards neutralized with hydrochlorie acid, and added to the solution in aqua regia; the residual chloride of silver is collected, and treated as at p. 576.

The solution containing the gold and copper is evaporated to dryness on a water-bath, the residue dissolved in hydrochloric acid, and again evaporated, to expel all nitric acid; it is then redissolved in dilute hydrochloric acid, and the solution boiled for some time with oxalic acid; the supernatant liquid is decanted from the precipitated gold, the latter washed with a little ammonia, to remove any oxalate of copper, then, with water, transferred to a weighed capsule, dried in an air-bath, ignited, and weighed.

The copper may be determined in the solution by precipitating with sulphuretted hydrogen, and subsequently converting the sulphide into oxide (p. 579).

AMALGAMS.

SEPARATION of Mercury and Zinc.

§ 429. The amalgam used for electrical machines may be analyzed for practice. Fifteen grains of the amalgam are dissolved in nitric acid, the solution evaporated with hydrochloric acid, on a water-bath, till all free nitric acid is expelled, and the mercury precipitated by sulphuretted hydrogen (p. 577).

The filtrate from the sulphide of mercury is evaporated to a small bulk, to expel excess of acid, and the zinc precipitated as basic carbonate (p. 588).

SEPARATION OF MERCURY AND TIN.

§ 430. Amalgam of tin may be analyzed by a process similar to the above, the tin being determined as binoxide (p. 582), and the mercury precipitated as sulphide (p. 577), or the mercury may be expelled by heat, and the tin converted into binoxide by roasting.

SEPARATION OF ARSENIC, COBALT, NICKEL, AND IRON.

SPEISS-COBALT.

§ 431. This substance contains arsenic, cobalt, nickel, iron, sulphur, and silica. Since the arsenic exists in very large proportion, it is well to determine it in a separate quantity.

Determination of Arsenic.-About 10 grs. of the finely powdered substance

are boiled in a large flask, with concentrated nitric acid, till no further action takes place. The solution is diluted with water, filtered, mixed with excess of ammonia and digested with colorless sulphide of ammonium, at a gentle heat, for a considerable time. The digestion should be conducted in a flask. The solution is filtered off, and the residue washed with water containing colorless sulphide of ammonium. (If yellow sulphide of ammonium were employed, some sulphide of nickel might be dissolved.) The filtrate is decomposed with a slight excess of acetic acid, and the precipitate of pentasulphide of arsenic treated as at p. 584.

If this precipitate should contain any sulphide of nickel or of copper, it may be dissolved in warm ammonia, and reprecipitated with acetic acid.

Another method of determining the arsenic consists in expelling the excess of acid from the nitric solution by evaporation; diluting largely with water, reducing the arsenic acid by sulphurous acid, evaporating the excess of the latter, and determining the arsenic as tersulphide (p. 584).

The treatment with sulphurous and hydrosulphuric acids must, however, be repeated several times, until no more arsenic is separated.

Determination of Sulphur.-About 20 grains of the substance are boiled with the strongest nitric acid, until the sulphur is either completely oxidized, or till the excess is separated in clear yellow globules; the solution is then diluted with water and passed through a weighed filter.

The sulphuric acid in the solution is determined as sulphate of baryta, from which the amount of sulphur is calculated.

The undissolved residue (sulphur and silica) is dried at 212° F. and weighed; it is then ignited in the usual manner, when the sulphur is volatilized and may be estimated from the loss.

Determination of Iron, Nickel, and Cobalt.-About 15 or 20 grs. of the ore are carefully roasted in a porcelain crucible, to expel as much as possible of the sulphur and arsenic. The roasted ore is then treated as before, with nitric acid, the solution evaporated to dryness, the residue digested with concentrated hydrochloric acid, water added, and the liquid filtered. The filtrate is saturated with sulphurous acid, digested for some time at a gentle heat, evaporated to expel excess of sulphurous acid, saturated with sulphuretted hydrogen, and allowed to stand for some time in a warm place; this treatment with sulphuretted hydrogen is repeated, until the odor no longer disappears after digestion for a short time. The precipitate is filtered off and washed. The filtrate is again treated, in the same way, with sulphurous and hydrosulphuric acids, as long as any arsenic is separated. The filtrate and washings are then concentrated by evaporation, and the iron separated as succinate (or benzoate) as directed p. 607.

For the separation of the cobalt and nickel, the solution (free from iron, arsenic, &c.), slightly acidified with hydrochloric acid, is mixed with a dilute solution of chloride of lime, to which a slight excess of sulphuric acid has been added, by which the chloride of cobalt is entirely converted into sesquichloride. A thin cream of pure carbonate of lime is then added in excess, and the mixture digested, in the cold, with frequent agitation, for at least 24 hours. The nickel remains in solution as chloride, while the cobalt is precipitated in the form of sesquioxide, mixed with the excess of carbonate of lime.

The precipitate is collected upon a filter and thoroughly washed; the filter is then placed in a capacious beaker, and covered with water, to which hydrochloric acid must be added from time to time, until the precipitate is entirely dissolved, which may be promoted by gently heating; the solution is separated from the filter paper (which must be very thoroughly washed), concentrated by evapora tion, mixed with ammonia in slight excess, and saturated with sulphuretted hydrogen; the precipitated sulphide of cobalt is filtered off and treated as at p. 587.

The solution containing the chloride of nickel is concentrated by evaporation, mixed with a slight excess of ammonia and saturated with sulphuretted hydrogen; the precipitated sulphide of nickel being dissolved in nitro-hydrochloric acid, and the nickel determined as at p. 587.

Another process for separating nickel and cobalt, known as Liebig's method, is executed as follows. The solution is mixed with a considerable quantity of hydrocyanic acid, and afterwards nearly neutralized with potassa; heat is then applied until the solution becomes clear and the excess of hydrocyanic acid is expelled. The cobalt is thus converted into cobalticyanide of potassium, and the nickel into the double cyanide of nickel and potassium. An excess of freshly precipitated well-washed oxide of mercury is now added to the hot solution, when the whole of the nickel is precipitated, partly as oxide, partly as cyanide. The solution is boiled for a short time, to convert the nickel entirely into oxide, which is collected upon a filter, washed, dried, and ignited, when the excess of oxide of mercury is expelled, and oxide of nickel alone remains.

The filtrate containing the cobalt is mixed with excess of acetic acid, and precipitated, while boiling, with sulphate of copper; the mixture is boiled until the precipitated cobalticyanide of copper has become somewhat granular, and filtered. The precipitate is washed, dried, ignited, dissolved in hydrochloric acid with a little nitric, the solution largely diluted, and the copper precipitated by sulphuretted hydrogen; the solution filtered from the sulphide of copper is concentrated by evaporation, the cobalt precipitated as oxide, by boiling with potassa, and determined as usual.

DETERMINATION OF THE Value of ManGANESE-ORES.

§ 432. Since the natural oxides of manganese are used chiefly for the preparation of chlorine in bleach-works, it is important that we should possess some ready method of ascertaining the quantity of chlorine which a given amount of the ore is capable of eliminating, as well as the quantity of hydrochloric acid consumed, which will depend upon the nature of the foreign matters contained in the ore.

In order to determine how much chlorine may be liberated by a certain amount of ore, about 100 grains of the latter, in a state of very fine powder, are heated with hydrochloric acid, in a flask provided with a bent tube, which conducts the chlorine into a weak solution of potassa contained in another flask; care is taken to evolve the whole of the chlorine, and the solution of hypochlorite of potassa and chloride of potassium is then tested by a chlorimetric process (p. 615).

To ascertain how much hydrochloric acid is consumed in the evolution of the chlorine, about 50 grains of the finely powdered ore are dissolved at a gentle heat in a measured quantity of dilute acid (of known strength), and the excess of acid remaining is then determined, after the complete expulsion of the chlorine, by adding a standard solution of carbonate of soda until a permanent precipitate begins to be formed.

A very neat method of testing the ores of manganese is that of Fresenius and Will, which consists in treating the ore (previously freed from earthy carbonates by washing with dilute nitric acid) with oxalate of potassa and sulphuric acid, when the oxalic acid (C,O) is converted into 2 eqs. of carbonic acid, the weight of which is ascertained from the loss suffered by the apparatus (so constructed that no aqueous vapor shall be carried off).

1 Wöhler recommends the precipitation of the solution, nearly neutralized with nitric acid, by a solution of nitrate of suboxide of mercury, which precipitates the mercury as cobalticyanide; the latter, when washed, dried, and ignited, leaves the black intermediate oxide of cobalt.

Since the action of oxalic acid upon binoxide of manganese, in the presence of sulphuric acid, is represented by the equation

2

MnO2+C2O, HO+HO.SO,=MnO.SO,+2CO,+3HO, every 44 parts (2 eqs.) of carbonic acid represent 8 parts (1 eq.) of available oxygen, and consequently 35.5 parts (1 eq.) of chlorine, which may be eliminated by the specimen.

The operation is conducted exactly as the determination of carbonic acid in an alkaline carbonate (p. 616), except that, instead of the carbonate, about 20 grains of very finely-powdered binoxide of manganese, and twice as much oxalate of potassa, are placed in the generating flask. The operation is continued until no more black particles of binoxide of manganese are visible, and at the conclusion, air is sucked through the flasks in the usual manner.

If the ore contain any carbonate of lime, a weighed portion must be washed with very dilute nitric acid, and subsequently with water, dried, and the available oxygen determined as above.

CHLORIMETRY.

§ 433. This name is given to the various methods of determining the amount of available chlorine contained in any specimen of the chloride of lime of commerce. The oldest of these methods consists in ascertaining what weight of the specimen is required to decolorize a given quantity of a standard solution of indigo (in sulphuric acid), previously graduated by means of a solution of potassa, which has absorbed a known volume of chlorine. This method has been, however, for the most part, abandoned, since the standard solution of indigo is changed by keeping.

A better chlorimetric process consists in determining the amount of bleach necessary to convert a known quantity of arsenious acid (ASO) into arsenic acid (AsO,).

A standard solution of arsenious acid is prepared by dissolving about 140 grains of the pure acid in a little dilute hydrochloric acid, with the aid of heat, and adding as inuch distilled water as will bring the volume to 10,000 grain measures; if the operator be competent to determine the amount of arsenious acid in a given volume of this solution, it will be found the best course; but otherwise, the weight of the arsenious acid originally employed, and the volume of the solution ultimately prepared from it, should be accurately determined, the object being to obtain a solution of arsenious acid of known strength.

About 50 grains of a fair specimen of the bleaching-powder are triturated in a mortar with a small quantity of water, a larger quantity being afterwards added; the solution is then rapidly filtered, the mortar being carefully rinsed, and the filter washed with cold water till the washings do not bleach solution of indigo. The volume of the filtered solution is then accurately determined.

About 1000 grains of the standard solution of arsenious acid are measured into a beaker, mixed with a moderate quantity of dilute hydrochloric acid, and colored with a little solution of indigo; a burette is then filled up with the solution of bleach prepared as above, and this solution added to that of arsenious acid, with constant stirring, until the color of the indigo-solution disappears, showing that an excess of chlorine has been added. The number of volumes of solution of chloride of lime necessary to effect this is then observed, and the amount of the original bleaching-powder to which they correspond calculated by a proportion; the amount of arsenious acid employed being likewise known,

the quantity of available chlorine present in the bleaching-powder is calculated according to the equation

AsO2+2HO+Cl,=AsO,+2HCl,

by which it will be seen that 99 parts, or one equivalent, of arsenious acid correspond to 71 parts, or two equivalents, of available chlorine.1

99 71 Arsenious acid employed: x

x= Available chlorine.

The method most commonly used, however, for determining the value of spe cimens of bleaching powder, is to ascertain the quantity of the latter which is required to peroxidize a known weight of the green sulphate of iron.

Pure crystals of the sulphate are powdered and dried, by pressure between folds of blotting paper; about 50 grains of the powder are accurately weighed, and dissolved in about 1000 grain-measures of cold water; the solution is then acidified with sulphuric acid.

The solution of bleaching-powder, prepared as in the last method, is then poured from a burette into the liquid, until the latter ceases to give a blue precipitate in a drop of solution of ferricyanide of potassium (placed on a white plate), showing that all the oxide of iron has been converted into sesquioxide. The amount of bleaching-powder employed is then calculated, and the available chlorine deduced according to the equation

2(FeO.SO3)+HO.SO,+Cl=Fe2O, 3SO+HCl,

by which we see that 278 parts, or 2 eqs., of crystallized sulphate of iron (FeO. SO, HO+6Aq) correspond to 35.5 parts, or 1 eq., of chlorine."

Fig. 78.

10

30

40

170

30

90

100

278 35.5: Sulphate of iron employed: x

x= Available chlorine.

ALKALIMETRY.

§ 434. The methods of determining the amount of available alkali contained in various specimens of commercial potash and soda are known by the name of alkalimetry. We shall first consider the valuation of potash, and subsequently the modifications necessary in the case of soda.

The older alkalimetrical process consists in ascertaining how many measures of dilute sulphuric acid of known strength are required to neutralize a given quantity of carbonate of potassa. For this purpose a test-acid is first prepared by mixing about 1000 grs. of pure oil of vitriol with about 10,000 grs. of water; this acid is graduated in the following manner; about 20 grs. of perfectly pure and dry carbonate of soda are accurately weighed and dissolved in water; the solution is colored blue with a few drops of tincture of litmus, and the test-acid slowly added from a graduated burette, the liquid being constantly stirred. The first addition of acid merely converts the carbonate of soda into bicarbonate; carbonic acid is afterwards liberated, and colors the litmus wine-red, but when an excess of acid is added, a bright red tint is produced, which indicates that the operation is completed. Having now observed the number of measures of dilute acid employed, we have only to calculate the amount of real acid which they contain.

1 Penot has modified this process. He dissolves 4.44 parts of arsenious acid, and 13 parts of crystallized carbonate of soda in water, and adds this solution, from a burette, to the solution of a known weight of chloride of lime, until the liquid no longer produces a blue color upon a test-paper impregnated with 1 part of iodine, 7 parts of crystallized carbonate of soda, and 3 parts of starch, heated with water until the blue color has disappeared.

2 Müller has proposed a new method for the valuation of chloride of lime. A standard