Ok, had several people ask out side of thread.
Before precipitating AgCl, here's what I would do:
Preparing the Ag solution for precipitation.
Typically, before filtering the nitrate solution, I would remove any lead and/or metallic gold with coprecipitation/entrainment with a small portion of the silver as its sulfate (this combined precipitate can go off and get cupeled with small loss of silver and full recovery of the gold, after parting). It is best to do this with a more dilute sulfuric acid solution, say 1-2 N (0.5-1 M). This step also helps deal with filtering any metastannic acid that may have been formed if the bullion was improperly blown in the furnace and has residual high tin/lead. Silver sulfate filters super nice and it holds up PbSO4/Au/SnO2.H2Ox nicely.
Precipitating the AgCl
1. To the remaining silver solution which may contain base metal nitrates in large quantities (Zn, Cu, Pd, Ni, etc.) I would add HCl such that there is an excess in solution and the silver is thrown down entirely as flocculant AgCl. On big batches, this can be done with a titration and the acid figured ahead of time. As an aside: if there's mostly silver in solution, I don't even bother precipitating all of the silver, just 99% of it and then I will reuse the nitric acid over again to dissolve more silver. Any occluded AgNO3 in the AgCl filter cake gets converted with step 2 below.
Filtering the AgCl, removal of primary solution done here
2.I will filter off the AgCl and suck it down until the cake cracks on me. This is where the waste volume is saved. I dig most of it out carefully or flip the funnel and load up the AgCl in a drum and reslurry the AgCl with 5% or so room HCl solution. AgCl is at its least soluble in dilute HCl solutions.
Filtering the repulped AgCl-last "rinse" done here
3. I then filter this pulp acid off which goes with the primary liquor that was precipitated from. That goes to recovery of any trace PGMs. If there are any PM's like Pd/Pt, I rinse more at this stage and check with stannous/dmg. If it's base metals, I don't even bother.
Converting the silver chloride to metallic state
4. This AgCl filter cake gets converted to Ag metal in a variety of fashions but not by the Kundu method. The aldose sugar method many people use here with Karo syrup or whatever reducing sugar; some use sulfuric and iron (which I myself prefer). If the base + sugar method is used, copper and other base metals form hydroxides. Reductant must be added because Ag2O is in equilibrium with AgOH so just adding NaOH to AgCl isn't good enough as the filtrate will reprecipitate AgCl on addition to chloride wastes.
Washing with the borax / nitre solution
5. The metallic silver so obtained is filtered from the liquids, given a rinse with water to remove most of the salt and then is rinsed with a solution containing borax, nitre, and optionally, soda ash and sucked dry. I never really had a critical amount of either or a ratio, usually a heaping scoop of each (maybe 50-100 g/L) into hot water. The solution can be used over and over again.
@samuel-a : The water savings is in filtering the AgCl and the silver cement. It is hard to get everything off of it and it is worthless to rinse it unless it has a bunch of Pd/Pt in the solution along for the ride. One wash cycle for the AgCl, then convert, then a few washes to get the NaCl off. The nitre is a workhorse at getting base metals oxidized up. Leaving it on the damp cake and/or drying it on the cake is much better than trying to flux it in the melt. Nitre just decomposes when adding to the melt and wrecks crucibles..it's not around long enough to meaningfully burn out impurities. Adding it to the cake (and the borax) stops the silver from getting contaminated at the microscopic level and if there is any Pd in there, it keeps it from getting the silver out of spec for Pd because nitre and Pd get along quite well.
The only issues I've seen with nitre/borax washing the cement silver is
1) needs exhaust because usually some Cl2/NOx from decomposing nitre and oxidizing NaCl. In practice, this is much less than that done
2.) hard to recover Pd from slag if it wasn't rinsed off the filter cake
3.) the silver coming out is high in oxygen so it needs the slag tapped off. I usually will treat the silver with a deoxidant to pull out dissolved oxygen so the bars don't stand up and crab.
As to its effectiveness:
Even garbage AgCl converted to metal will be cleaned up by this process, particularly if the crucibles are silicon carbide not clay graphite. 999 and change every time. Graphite can act as a reducer to the metal oxides produced by the nitre and stabilized/dissolved by the borax glass and put them back in the silver.
something what is quite the only "analytical" thing amateur refiner could use to tell if the conversion to silver oxide is done.
