The Etruscans made granulated gold jewelry with granules, tiny gold spheres as small as 1/300th of an inch, attached to a gold base. It is hard to see how that could be done using solder without flooding the join and converting a granule to a bump. The technique they are believed to have used was described, not for granulation but for delicate filigree, by Theophilus in the 11th century and Cellini in the 16th, rediscovered and patented as “colloidal hard soldering” by Littlechild in the 20th century. The idea is to glue the granules down using a mix of a copper salt or oxide, a flux, and, for Littlechild but not for Cellini, an organic glue. The evidence that that is how it was done in classical antiquity is that “chrysocolla,” the name of a semiprecious gemstone that is a copper mineral, is Greek for “gold glue.” Pliny uses its name
Heat in a reducing flame and the copper oxide reduces to a film of metallic copper between granule and surface. At a temperature a little below the melting point of the gold or silver the copper soaks into the surface of the metal of the granule and the sheet, turning it into a film of copper-gold or copper-silver alloy, aka gold or silver solder, which fuses together. At a slightly hotter temperature the whole piece melts, so you have to remove the torch flame in time. Cellini says to watch for the skin of the gold to move, which is what I try to do with my silver.
You need to use fine silver rather than sterling, or high carat gold. For doing it with gold, which I have not tried, a modern source recommends depletion gilding the surface to get it closer to pure gold.
To make the granules you cut your silver or gold in small pieces and melt them with your torch on a charcoal block — they will pull up into spheres. Herbert Maryon suggests doing it on hemispherical dents in the charcoal so as to avoid slight flats at the bottom of the granule.
Cellini’s formula, which I use, is six parts of verdigris (copper carbonate) to one each of borax and salts of ammonia (sal ammoniac). I have only experimented a little with adding glue. My modern source uses hide glue diluted one to twelve in water to glue the granules down.
The most frustrating part of the process is when, in either pickling or polishing the piece, some of the granules that you thought were fused turn out to be only stuck down by the melted borax and you have to redo them.
Metropolitan Museum Simple Granulation, probably
11th century granulation
Etruscan pin and pin head, Metropolitan Museum
Etruscan Earring, Metropolitan Museum
Simple Granulation, probably
One Silver Granule and Video of attaching a granule.
A Granulated Cross: Original My Recreation