SILVER.—IIdmid Assay

the presence of this metal can be at once ascer-tained in the manner subsequently described (seepage 762), and any error which it would occasionmay also be avoided.

The process might be conducted in either of thethree following ways : — (1) By precipitating thesilver by an excess of the chloride, and calculatingthe weight of metal from that of the chloride pro-duced. (2) By determining the weight of a standardsolution of salt which is required to precipitate thesilver present. (3) By determining the volume of astandard solution of salt which is required for thatpurpose.

Of these the first is that employed in the Indianmints, which has been already referred to; thesecond is tedious, requiring considerable care, andis therefore rarely practised; and the third is thatrecommended by Gay-Lussac , which we now pro-ceed to describe.

If the approximate assay of the alloy under ex-amination be not known, it must be ascertained byone of the methods explained at page 758; or, ifit be preferred, by the volumetric method describedat the end of this article (see page 763). Such aportion of the silver alloy is taken as that it shallcontain rather more than 1 gram, of pure silver;this amount is ascertained by the simple proportion,

A : 1000 : : 1-0 : x,

where A is the approximate assay expressed inthousandths, and x the amount of alloy required,

. •. x = L° a 0 - 0 - grams.

The metal is dissolved in about 1 oz. of nitric acid,of specific gravity 1T90, by the aid of a gentle heat,and 100 c.c. of a standard solution of salt are added;this solution being of such a strength that 100 c.c.exactly precipitate 1 gram, of pure silver. Thebottle is now shaken, either by hand or in the“agitator” shown in Fig. 14, in order to agglomeratethe precipitated chloride of silver, and leave thesupernatant liquid clear. Now, if the additionof a drop of salt solution to this liquid producesa precipitate, it will of course indicate that nitrateof silver still exists in the solution; but if no pre-cipitate is formed, and a drop of a solution ofnitrate of silver produces a precipitate, it is evidentthat too much salt has been added. Two dropsof the above “normal” solution of salt suffice todeposit 1 milligram, of silver; hence, for delicateassays, it is usual to prepare a solution of one-tenththe strength, and to continue adding this, in c.c.’s ata time, until no further precipitate is obtained. Theamount of salt so added (making an appropriatededuction where it has been found necessary to addsilver, which is done by the aid of a titrated solutionof the nitrate containing 1 gram, of pure silver in 1litre of water) gives the requisite data for ascertainingthe actual amount of silver present in the portion(x) of alloy taken for assay; whence the amount per1000 parts may be ascertained by means of tablesprepared for the purpose, or by the following simpleproportion:—

x 1000 : : M : S.

VOL. II.

Volumetric Methods. 761

where x means the same as in the last case, Mthe amount of silver it contains, measured directlyby the amount of salt added, and S the standard ofthe alloy, expressed in thousandths, . •. S = —25-? 1 .

The method, so far, is only capable of indicatingthe amount of silver present to the xiro utli part, buta skilled assayer can attain to even greater accuracy,for he will know by practice, from the amount ofthe precipitate obtained on adding a c.c. of thedecimal solution of salt, whether, after shaking theliquid, a precipitate will be produced when a freshaddition is made; and therefore, knowing theamount produced when the solution is just saturated,he can estimate with the eye the fractional part of athousandth, which any less precipitate proves to bepresent. The composition can thus be ascertainedwith considerable certainty to the xn,'W(l^ P ar *-From the above brief account of the process itwill be seen that three solutions are necessary, aswell as suitable pipettes for the exact admeasure-ment of them. The solutions are—

(a) A standard or normal solution of salt.

( h ) A decimal solution of salt.

(c) A decimal solution of nitrate of silver.

The preparation of these will now be described.(a) The standard solution may be made eitherby adding a known weight of pure, thoroughly drysalt to a definite bulk of water, or, what is moreconvenient, especially where a large quantity of thesolution has to be prepared, by diluting a saturatedsolution of salt to the requisite strength. To obtainpure chloride of sodium when the first method isadopted, a concentrated solution of the best tablesalt is treated with caustic baryta, to remove sul-phuric acid and magnesia; then with a slight excessof carbonate of soda, to remove baryta and lime.The liquid having been warmed, any precipitateformed is allowed to subside. The supernatantliquid is evaporated, and the resulting crystalswashed slightly, and carefully heated to dull redness.They may be kept in a glass-stoppered bottle. Ifattainable, chemically pure rock salt may be used inplace of the above.

The atomic weights of silver and chlorine being,according to Stas, 107-929 and 35*457, it will beevident that 107-929 grams, of pure silver requirefor their complete precipitation as chloride, 35-457grams, of chlorine; hence each gram, of silverrequires 0-3285 gram, of chlorine, an amount whichis contained in 0-54162 gram, of pure salt. If there-fore it be required to prepare M litres of standard-solution (an amount which will suffice for 10 X Massays), it will only be necessary to dissolve 5-4162X M grams, of the pure salt in M litres of distilledwater, care being taken that the solution is complete.

It is, however, more convenient to employ asaturated solution of salt, and common water, which,of course, itself contains a small quantity of salt,may be added to this until the requisite strength isattained. It is impossible to give the exact pro-portions necessary; but if the solution be thoroughlysaturated, by being left to stand for a few days at atemperature of 15° C., the addition of 3 troy ozs. of96