Page:Encyclopædia Britannica, Ninth Edition, v. 2.djvu/793

approximate standard of the alloy with sufficient accuracy. The sample of gold alloy is either cut from a corner, or drilled out of the centre of an ingot, care being taken to secure uniformity of composition with the whole mass. The amount of lead to be added to the alloy varies with the proportion of base metal present; as a rule, from 10 to

20 parts are required to 1 of alloy.

The amount of lead having been determined, the alloy is wrapped up in a known quantity (say one-half of that required for its purification), formed into a case somewhat resembling a thimble, great care being taken to make the joints firm and close so that no gold shall escape. The re quisite quantity of silver is added at the same time. When a number of assays are made at the same time, they are arranged, enveloped in their lead cases, on a board divided into compartments corresponding in number and position with the cupels into which they are intended to be charged. As the assayer generally makes two or more trials of the same piece, so that great accuracy may be secured, it is his practice to give one assay a side place in the muffle, and the second a middle one, in order to check any irregularity in the result. When a sufficient number of assays are weighed and arranged upon the board in the manner referred to, and the furnace as well as the cupels raised to the neces sary point of heat, the charging tongs are then taken, and the rest of the lead and silver apportioned to each assay placed individually upon the cupels, beginning at the back of the muffle. The lead added in this case is not flattened, but is a piece of known weight various sizes of which, as well as cases, are kept in stock by the assayer. The lead so placed in the furnace rapidly melts, and becomes covered with a gray oxide, but soon after appears fluid and bright. At this point the assays are added by means of a pair of tongs (fig. 13), great care being taken that no part overhangs or touches the edge of the cupel. The assays are thus drawn into the mass of molten lead, and any particles of gold are in this manner prevented from adhering to the sides of the cupels in charging, sufficient despatch being used to obviate the fusion of the assay in its transition. The assays being charged in order on their respective cupels, and the furnace previ ously filled with fuel, the door of the muffle is partly closed, and the progress of the cupellation watched. Too much air must not enter, or the j muffle will be chilled and the progress retarded, whilst if too little enters the operation will be too slow.

Fig. 13.

At first dense fumes will be observed to rise from the melted metal, indicating the oxidation and subsequent volatilisation of the lead. These after continuing some time are followed by the appearance of small luminous points on the surface, which increase in size and brilliancy as the operation progresses. Then a minute stream of red fused oxide of lead is seen to flow from the top of the metal globule and circulate round, when it is carried down and absorbed by the cupel. This is caused by the oxida tion of the lead by the air, which at the same time oxidises the other metals, except silver, which accompany the gold. As the cupellation advances the fumes gradually lessen in density till they disappear altogether. The melted button at this stage is observed to become more convex and round, and as the last vestiges of the lead and alloy are being carried off, it assumes a cloudy appearance on the surface, changing to large bright points of the fused oxide, till at length it is nearly freed from all impurity. At this point the gold-silver alloy displays some singular and beautiful characteristics. Deprived of all the base alloy save the last minute portion that tarnished its lustre, it has become bright and pure, and finally it gives forth from its surface iridescent circulating bands of light, which indicate the suc cessful completion of the operation. The globules after being cooled, are removed from the cupels with a pair of pincers and carefully cleaned. They are then placed in their com partments and weighed with the greatest nicety. Finally, they are submitted to the operation of parting. This is effected by boiling with strong nitric acid, which dissolves the silver and leaves the gold as a sponge. The gold-silver globule is first passed through a flattening mill, and reduced to long thin strips, which are annealed and then rolled up into a corkscrew spiral, so that the acid may penetrate between each fold. The spirals are now transferred in order to small platinum cups arranged on a frame, so that they can be simultaneously lowered into and removed from the nitric acid. They are kept in the hot acid for the requisite time, then washed, and the residual gold sponge, which possesses considerable coherence, and retains the shape of the original spiral, is carefully dried and finally weighed with the greatest possible accuracy.

We shall now proceed to describe briefly the process of silver assay. The cupellation process does not differ except in details from that of gold, the outline already given will, therefore, be a sufficient description. Like gold, silver occurs in two classes of combinations, mineralised, as in red silver ore, chloride of silver, argentiferous galena, &c., or as metallic silver and its alloys. In metallurgi cal establishments, where silver occurs in small propor tion with metallic sulphides, such as those of lead or copper, the process of cupellation is generally adopted not ouly for quantitative estimation of the amount of silver present in the ore, but also for its extraction on the large scale. In the assay of silver bullion, however, the process of cupellation is now almost entirely superseded by a volumetric process, devised by the distinguished French chemist, Gay-Lussac, by whose influence it was introduced into the Paris Mint. The process consists in determining the fineness of silver bullion by the quantity of a standard solution of common salt necessary to precipitate fully and exactly the silver contained in a known weight of alloy. This process is based on the following principles:—

The alloy previously dissolved in nitric acid is mixed

with a standard-solution of common salt, which precipitates the silver as chloride, a compound perfectly insoluble in water and even in acids. The quantity of chloride of silver precipitated is determined not by its weight, which would be less exact, and occupy too much time, but by the weight or volume of the standard solution of common salt neces sary to precipitate exactly the silver previously dissolved in nitric acid. The term of complete precipitation of the silver can be readily recognised by the cessation of all cloudiness when the salt solution is gradually poured into that of the nitrate of silver. One milligramme of that metal is readily detected in 150 grammes of liquid, and even a half or a quarter of a milligramme may be detected if the liquid be perfectly bright before the addition of the salt solution. By violent agitation during a minute or two, the liquid, rendered milky by the precipitation of chloride of silver, becomes sufficiently bright after a few moments repose to allow of the effect of the addition of half a milligramme of silver to be perceptible. Filtration of the liquid is more efficacious than agitation ; but the latter, which is much more rapid, generally suffices. The presence of copper, lead, or any other metal, with the exception of mercury in the silver solution, has no sensible influence on the quantity of salt required for precipitation ; in other words, the same quantity of silver, pure or alloyed, requires for its precipitation a constant quantity of the standard salt solution. Supposing that 1 gramme of pure silver be the quantity operated on, the solution of salt required to

precipitate exactly the whole of the silver ought to be of