Flux mystery

[amesametrita]

Hi, dear forum members!
I refine karat gold using inquartation with silver and parting in nitric for a long time.
The result is consistent 996+ which is enough for my goals.
Contaminant is only silver.
Sometimes bleaching with KNO3
Melting is in a gas kiln.
I tryed fluxing using the forum recepies to achieve 999

The first one is by 4metals.
2 pounds Anhydrous borax glass (no waters)
1 pound Soda Ash
1 pound Diatomaceous earth (silica)
1 pound Manganese Dioxide
1/4 pound Flurospar (calcium fluoride)
1/4 pound Calcium oxide (CaO, Lime, Type S, Unslaked)
No results.
Fire assay - no improvement. XFR - no improvement.

The second one is by Rick. "The Rock Man".
3 parts Sodium Carbonate. (Soda Ash)
1/2 part Borax.
1 part Silica.
1 part Manganese Dioxide.
1/4 part Fluorspar. (CaF2, Calcium Fluoride)
1/4 part Calcium Oxide. (CaO, Lime, Type S, Unslaked)
No results.
Fire assay - no improvement. XFR - no improvement.

Premixing, no moisture.
Slow heating, fast heating, prolonged boiling...
Nothing works.

Any advices?

Best regards!

 

[etack]

You could try AR and no flux to see better results. what is the end use of your gold?

Eric

 

[amesametrita]

It is a king of scientific experiment.
To understand how stupid I am
So I want to achieve exactly fluxing result.

No idea about end use of the gold.
Just selling it to a dealer.

AR purification is not economically feasible in the place I live.

 

[etack]

If you get .996 the you are alright to sell to a dealer. Just make sure they can test that high.

experiment away goldsilverpro might be able to help with cleaning gold with flux. that's an OLD argument on this board. Sorry don't want to stir the pudding. :lol:

Eric

 

[MysticColby]

I wouldn't expect any flux to be able to remove silver impurities from gold. copper, maybe. but not silver.
flux collects solid parts from melts (such as metal oxides, which generally have a higher melting point than the pure metal).
silver oxide converts to elemental silver when heated, so it does not form an oxide.
I say maybe copper because that does form oxides.

you need to refine if you want higher purity. melting does not generally refine.

 

[Westerngs]

As MysticColby indicated, you will have a hard time achieving 0.999 purity fluxing to remove silver.

About the only thing I can think of that you might try is adding salt to your flux to see if you can slag off the silver as a chloride. I guess you could call it a variation on the Miller process. Again, I think you will have a hard time getting 0.999 gold

 

[butcher]

I agree with Mystic, silver is not a metal that would oxidize easily due to its position it the reactivity of metals, and any flux that would oxidize silver I believe would also oxidize the gold and cause loss such as chloride in the flux.

I do not quite understand, you can in-quarter, and part with Nitric, but you say you cannot finish and refine with aqua regia, or another solvent for the gold, or that is not economically feasible, Here for me the nitric is more expensive, so the part of the process you are doing is what cost me the most.

 

[Harold_V]

I'm of the opinion that it will be difficult to improve the quality if the contamination is silver. It has long been recognized that traces of silver behave much like gold, and is difficult to separate from gold, even when dissolving the gold.

I am not a chemist, nor do I profess to understand well the function of soda ash, but I am a firm believer that including soda ash in a flux that is intended to purify is a mistake if the end product is to be pure gold. That would be particularly true in view of the fact that soda ash is advised for reducing silver chloride in the furnace. To add to that, my years of experience in cleaning melting dishes with soda ash, to me, is evidence enough that any traces of metal that may be present in the material will simply be melted and still report in the material, as your experience has shown.

If you care to experiment with flux as a method to improve gold quality, two things are important. One of them is an oxidizing agent, so traces of base metals are reduced to an oxide. The other is borax, which absorbs the oxides. Fluorspar can be used to thin the flux, but most likely would not be a requirement for this particular application.

I would suggest that you try melting with JUST borax, and an oxidizing agent (salt peter has been suggested in the past). If you find the flux discolors upon melting, pretty good sign you're picking up some contaminants, and your quality may be improved.

Personally, I found the key to high quality gold, assuming you're chemical refining, was to dissolve the gold, filter, precipitate, wash well, then repeat the process. A second refining has the potential to yield very high grade (better than 9995) gold. My routine was to double refine all of my gold, which I did as a matter of pride. I wanted to be recognized as a competent refiner.

Harold

 

[nickvc]

I'm take a slightly different attitude to many on the forum and I'm a great believer in only doing the minimum work needed to reach the goal I require. I guess that your gold dealer doesn't care what your gold assays at so long as you get 995 or better and won't pay anymore for 9999 so there's no point to refine further in my opinion, if this is your business then definitely not, but maybe the way the Asian jewellers refine their metals might work, they use copper to inquart which is much easier to remove and if it does still report in your gold then the fluxes might do the job. The downside is it requires much more nitric and any silver in the material will need to be cemented out, the copper could be recovered and reused by cementation with steel but I'm not sure the extra work and acid costs would be worth the effort. K.A.Driver did an experiment using only nitric and achieved excellent purity but again it's takes more time and probably more acids which for you might again be a waste of effort and acid costs if you can't achieve a higher price.

 

[amesametrita]

... and I'm a great believer in only doing the minimum work needed to reach the goal I require. I guess that your gold dealer doesn't care what your gold assays at so long as you get 995 or better and won't pay anymore for 9999 so there's no point to refine further in my opinion...

Exactly. It is economically unfeasible to refine from 996 up.

I'm of the opinion that it will be difficult to improve the quality if the contamination is silver. It has long been recognized that traces of silver behave much like gold, and is difficult to separate from gold, even when dissolving the gold.

This was my point of view as well.
Until I found a lot of mentions of 4metals' recipe here.
The point is to use Manganese(IV) oxide.
As super strong oxidizer.

After I found Rick's composition, again with MnO2

And tried...
And failed...

 

[HAuCl4]

amesa:

Did you melt your inquarted/leached with nitric, 996 gold, and THEN used the 4metals flux?.

The flux is supposed to be mixed well and then combined well with the powdery gold, then heated slowly (otherwise it will spill over the crucible). Melting only at the end of all reactions. I pour out the (reacted) flux liquid before the gold melts (about 700 Centigrade), then put the crucible back in the furnace to melt the gold.

When I have used the flux, it has worked well (i.e. 9995+ result).

I used 77/23 inquartation instead of 75/25 many times, and my metal of choice has been copper or zinc or a combination of both. I doubt any of what is said in this last paragraph has any relevance on the results.

 

[amesametrita]

Did you melt your inquarted/leached with nitric, 996 gold, and THEN used the 4metals flux?.

Correct.

The flux is supposed to be mixed well and then combined well with the powdery gold, then heated slowly (otherwise it will spill over the crucible). Melting only at the end of all reactions. I pour out the (reacted) flux liquid before the gold melts (about 700 Centigrade), then put the crucible back in the furnace to melt the gold.

I thought that contaminant survived nitric is entrapped behind gold shield and should be melted out before flux can play.
Ok, will try without melting.

and my metal of choice has been copper or zinc or a combination of both.

Looks like this is the main difference.
It's much easier to slug out copper and zinc.
And you don't need Manganese Dioxide for this task.

I must use silver for inquartation to remove PGM.

 

[HAuCl4]

I must use silver for inquartation to remove PGM.

That's what 4metals used, actually sterling silver. I believe silver is what the original guys that discovered the method by mistake used too.

"The trick" is to mix well the flux with the gold powder after inquart. I like to get rid of the "dirty flux" before melting too. Silver and also base metals are scavenged into some complex oxides by the MnO2. Of course this happens better at maximum surface area (i.e. before melting the gold).

If you have some tin in your inquarted gold, a 15 minute boil in sulphuric acid, after the nitric acid treatment, will help with the purity.

Also a second, 15 minute boil in concentrated nitric acid under pressure (2-3 bar) will likely bring you to 997, 998 purity. This is unrelated to the fluxing method (also don't melt the gold before the second nitric boil... :lol: ).

 

[samuel-a]

Silver will not form oxide over 450C.
Even if it is trapped in the slag (which i doubt) it could just as easily drain down. Flux is definatly not a bullet proof way to refine gold to higher purity 999+, no matter how you twist it.

Even if you pump Cl to the charge, most likely you would not remove this last bit of silver.
On this subject, i've seen a mention above about NaCl as chlorinating agent, That is not the case.
Personally, I have melted silver with NaCl many times (from AgCl method) without any loss of silver.

 

[HAuCl4]

Yet it works...

 

[amesametrita]

4metals original post mentions 30 minute @ 2000F:

http://goldrefiningforum.com/~goldrefi/phpBB3/viewtopic.php?f=49&t=6187#p54099

 

[kadriver]

This is an old post, but I have some experience to share. I have had consistent success producing 3 nines (99.9%) pure gold using only inquartation and dilute nitric treatments.

When calculating for inquartation, I use 75/25, but I multiply the calculated amount of silver by 1.1 to increase the amount of sterling I add by 10%. This usually makes some nice pieces that cling together through the whole process.

I then do an initial dilute nitric treatment in a COVERED container for about 1/2 hour with medium heat, drain the liquid into my silver jar and rinse well with distilled water. The liquid I drain off is dark blue.

The second nitric treatment is the same except I turn the heat way up and boil until there are no more brown fumes, and the liquid is at a rolling boil, I remove from the heat, drain and rinse again - this take about 60 to 90 minutes.

The third treatment is the same, but this time I turn the heat up and just let it boil away for several hours - maybe three. By now the acid is only pale blue as most of the silver and base metals were removed in the first two nitric treatments. I pour off the liquid into my silver jar - a heavy Pyrex coffee pot with a glass lid.

The fourth treatment is the same as the third, except I let it boil for up to 6 hours with fresh clear dilute nitric acid. At this point the dilute nitric sometimes turns a pale yellow color, but stannous test turns up negative. I pour the hot clear acid in my silver jar along with the wash water and use it to dissolve more silver later on for my silver cell.

The pieces of inquarted gold are in their original shape and are an orange brown color. I can tell that they are really pure by the color, especially after rinsing and drying.

This process takes lots more time and electricity, but I can count on getting 999 on the assay just about every time.

I've never (yet anyway) had a problem with tin or lead.

I've used this process so regularly that I have lost some proficiency in my refining technique. I used AR to do a refining the other day and left out a step (adding H2SO4) before filtering.

kadriver

 

[kannang338]

Hi folks,

Currently we are trying to reduce zinc, indium, iron and partly copper in the melting stage with the help of oxidizing flux and oxygen purging.

But we can't able to select the perfect flux and their combinations to oxidize the input material. (4 to 5kg of 22kt gold which is having about 300grams of alloy materials as mentioned above)

Is there any better flux available for effective oxidation to occur.

We recommend the final output have less than 100ppm of the zinc and iron.

 

[Lou]

Heat will cook out the zinc but so too will borax. Silica is a good trap, along with borax, for Fe.

You might consider Miller process.

HAuCl4 is no where to be seen (hope he is well) but he has small batch experience with Miller.

 

[houseofsas]

Does anyone know about the old school assay and the new assay methodology ? I have a complex ore that is supposed to be around 2% gold per tonne got 1 test on it all the other tests show basically nothing, is there a difference in testing for monoatomic and diatomic gold ? Also when taking the raw complex ore, i microwave the ultra fine powder first, what would be a good formula for smelting in a induction furnace to get the best results? Anyone have some going tips at all ?