This section is from the "Henley's Twentieth Century Formulas Recipes Processes" encyclopedia, by Norman W. Henley and others.
Add to every 100 parts, by weight, 5 to 8 parts, by weight, of cupric chloride and melt until the oily layer which forms has disappeared. Then pour out, and in most cases a perfectly pliable gold will have been obtained. If this should not be the case after the first fusion, repeat the operation with the same quantity of cupric chloride. The cupric chloride must be kept in a well-closed bottle, made tight with paraffine, and in a dry place.
Pass chlorine gas through the molten gold, by which treatment most of the gold which has otherwise been set aside as unfit for certain kinds of work may be redeemed.
To determine the presence of gold in ores, etc., mix a small quantity of the finely powdered ore in a flask with an equal volume of tincture of iodine, shake repeatedly and well, and leave in contact about 1 hour, with repeated shaking. Next allow the mixture to deposit and dip a narrow strip of filtering paper into the solution. Allow the paper to absorb, next to dry; then dip it again into the solution, repeating this 5 to 6 times, so that the filtering paper is well saturated and impregnated. The strip is now calcined, as it were, and the ashes, if gold is present, show a purple color. The coloring disappears immediately if the ashes are moistened with bromine water. The same test may also be modified as follows: Cover the finely pulverized ore with bromine water, shake well and repeatedly during about 1 hour of the contact, and filter. Now add to the solution stannic protochloride in solution, whereby, in case gold is present, a purple color (gold purple of Cassius) will at once appear. In case the ore to be assayed contains sulphides, it is well to roast the ore previously, and should it contain lime carbonate, it is advisable to calcine the ore before in the presence of ammonium carbonate.