Soda Ash for Cleaning Gold?

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

goldsilverpro

In Remembrance
Moderator Emeritus
Joined
Mar 30, 2007
Messages
7,961
Location
Benton, Arkansas
To my friend, nemesis, and, a damned good refiner, Harold,

A friendly discussion.

I disagree with your idea that soda ash will reduce base metal oxides to metal. It doesn't work out chemically, as I will prove.

You are aware that we disagree on the use of soda ash for fluxing pure gold. Your contention, as I understand, is that soda ash reduces the base metal oxides to metals, thus re-contaminating the gold. A question is, Do you assume this because soda ash reduces silver chloride or, did you read it somewhere? Hoke? I sure don't remember it in Hoke. Easiest reading technical book I've ever read, BTW. I've probably read and studied Hoke from 10 - 20 times, cover to cover. If fact, the first time I ever heard it was from you. You did make me think about it. It kinda made sense, seeing that it reduces silver chloride to silver metal or, does it?

Soda ash is used widely to help clean gold. I have met about 100 pro refiners that use it. I have never seen the reduction thing in print and, I have read a lot of stuff.

I have found that the silver chloride is not reduced by the soda ash. The soda ash does not act as a reducing agent in this melt system. The soda ash (Na2CO3) converts the silver chloride to silver carbonate. The HEAT reduces the Ag2CO3 to silver metal, not the soda ash. In the 1st equation, below, the silver valence is +1 on both sides of the equation. If reduction had occurred, the valence would have gone from +1 to Zero, like in (3), below.

This is a quote from the silver chloride chapter in the yellow IPMI notebook that both of us own, page 9. Verbatim quote.

This high temperature technique proceeds as follows:

(1) 2AgCl + Na2CO3 = Ag2CO3 + NaCl

(2) Ag2CO3 = Ag2O + CO2

(3) 2Ag2O = 4Ag + O2

Reaction (2) and (3) occur by thermal composition. The silver carbonate of reaction (2) completely decomposes at 225 deg C. The silver oxide of reaction (3) completely decomposes at 340 deg C.

I conclude that soda ash does not act as a reducing agent in the melt

What if copper, nickel, etc., went through the same 3 equations? In the CRC Handbook, it gives the decomposition temp. of silver oxide (equation 3) as 644 deg F. There is none given for either of the val.+1 or val.+2 copper oxides. Valence +1 copper oxide has the lowest melting point of the two, 2255 deg F. Maybe, it decomposes at 3500 deg F. It doesn't say. You find very little copper metal in nature, BTW. For nickel, etc., their positions are even higher, than copper, on the electromotive series. They want to be metals even less than copper. Nickel oxide melts at 3614 deg F. And, the metal oxides must melt before they can become reduced.

The soda ash could easily create carbonates with the base metals, which decompose to oxides at low temps. However, the reduction temperatures of the base metal oxides (equation 3) are much higher than that of the gold melt. The metal oxides will stay in the slag and will not contaminate the gold. The use of soda ash will not contaminate the gold, period.

I conclude, without a doubt, that soda ash will definitely not reduce any base metal oxides in the gold melt.

Gold is more efficiently cleaned using soda ash, along with borax and niter.
 
goldsilverpro said:
To my friend, nemesis, and, a damned good refiner, Harold,

Chuckle! Some talk, coming from an arch enemy! (very large grin!)

A friendly discussion.

I can deal with that! :)

I disagree with your idea that soda ash will reduce base metal oxides to metal. It doesn't work out chemically, as I will prove.

Clip of some very interesting information, none of which I contest.

I conclude that soda ash does not act as a reducing agent in the melt.

Clip of more information.

I conclude, without a doubt, that soda ash will definitely not reduce any base metal oxides in the gold melt.

I'm the first person to admit that I am woefully lacking in a chemical education. I am not in any position to prove, or disprove, any of the equations.

Gold is more efficiently cleaned using soda ash, along with borax and niter.

My practical experience forces me to disagree, and I'll give you the reason why.

I posted some time ago on cleaning melting dishes. A melting dish that is virtually black from contaminants, barren of all visible metal, can be heated several times, with the addition of more borax, with little change. The flux remains thick and sticky, and changes color only because it is thinned with a greater volume of borax.

Something magical happens, however, when you introduce soda ash. The dark sticky flux begins to change color, thinning in the process. From this heated flux there magically begins to appear beads of metal that were previously not visible. The flux gets lighter in color, metal that was not recognized appears, yet the only source could be the dirty flux----for you've introduced nothing else. That this happens is irrefutable. I did that process time and again over a period of several years, and even offered a cleaning service to my customers for their centrifuge crucibles. Armed with my comments, are you telling me that you'd be happy to use a flux that introduces something to the heat that otherwise was held in suspension? That's sure what you're doing when you use soda ash when melting pure gold, in spite of the fact that I do not understand the nature of the chemistry involved.

Melting under a flux cover for doré can be useful, I openly admit. Fluxing pure gold? With borax, maybe-----even with a little niter (to acceleraate oxidization, which would be absorbed by the borax), but soda ash? Not in my opinion, for what it's worth. My objective, being an end product of pure gold, was to remove oxides, not cause them to join the gold.

For the benefit of others, gold that is contaminated can be melted under a flux cover of almost any composition and yield bright, shiny gold---so long as it's cast with the flux, and the gold never sees the atmosphere. Gold so cast will readily oxidize when re-melted without the protective cover of flux. Fact is, a button of gold will give up the oxide skin if it is sprinkled with borax the moment it solidifies, but is still red hot. That in no way should be construed as the gold having been cleaned, or being pure. Heat it again, and the skin returns, assuming it isn't flux covered. Gold must be able to withstand melting in the atmosphere before it can be considered pure, if even then.

All I can say is my experience in the use of soda ash yielded excellent performance for the purpose for which it was used. If you don't feel it acts as a reducer, I'm unable to discuss the issue. My books on refining are lost in our huge stack of storage, likely never to be seen until I finish the house. An hour of searching some time ago yielded only a few books, nothing that might help me in this discussion. The point is, when I finally ran my waste materials after storing them for years, I did enough research on fluxing to insure that I would get a good recovery. Soda ash was very much a part of the flux formula----along with borax, silica, and fluorspar. The results were staggering----over 200 ounces of gold recovered, along with more than 200 ounces of silver. The flux that resulted from the process was assayed and found to be too low in value to receive any further treatment (melting with litharge) to warrant the operation, so it was discarded. If the soda ash wasn't reducing, I'm at a loss to explain what was. It will take one hell of a lot of evidence to convince me that I wasn't using good and proper procedures.

I didn't flux** my pure gold aside from keeping the melting dish lightly covered with borax, so the gold wouldn't stick. Remember----the gold below was cast without the use of flux, and didn't need pickling! I left the use of soda ash to those that couldn't get their gold pure, but wanted to hide the fact. :wink:

Harold

edit: changed spelling to correct for proper tense
 

Attachments

  • Gold shot.jpg
    Gold shot.jpg
    73.3 KB · Views: 1,395
When brazing the copper will get shiny long before it melts, so if you are melting the metals, I think the copper will reduce anyways.
 
GSP, I do not know, only my observation when brazing copper together, the oxide will form then the copper will convert back to shiny copper then melt, so I think the heat is maybe braking it down or the gases from the torch flame.

Jim
 
Probably are using a reducing flame. Also, oxygen scavengers (germanium, molybdenum, etc.) are added to jewelry alloys to help prevent oxidation when casting. Maybe similar things are added to brazing alloys.
 
looking at the debate i remember of a chemist(siro arribas jimeno) who wrote a book called (qualitative analisis without using h2s)---he uses a hot solution of na2co3 to precipitate gold and pt from solutions --he them filter this precipitates and add to the precipitate hno3 to leave pure gold or pt and pass to the solution any base metals and pd
it might have nothing with your debate but it shows that soda ash reduces precious metal in solution to its elemental form------i have done this with solutions having rh mixed with base metals to obtain purer rh
 
Since a couple of months I can't get this subject out of my head.

Fact is, sodium carbonate is not reducing, nor are the end products NaO or CO2.

The gas flame has a reducing zone and like already correctly mentioned, this is obvious when brazing copper. As soon as you take the flame away, the copper oxidizes again.

But a cover of inert gas and inert melted salt combined with the reducing flame might be an explanation for the fact that you can find reduced base metals in the melt.


Today I came to this conclusion when I smelted some dirty and highly AgCl contaminated Ag/[stt]AgO2[/stt] Ag2O powders that I still had around from former times. I used 1:1 soda ash and it worked well to recover the silver. Now on one of those buttons you can see pieces or areas of red copper.

If my conclusion is right, soda ash will not harm pure gold, but it will not help either, if there should be some base metal oxides flying around. Borax does help by binding chemically to the base metal oxides, both in the reducing and in the oxidizing flame - that is the purpous of borax perls.

If there is no reducing flame, but you are heating by other means than an open flame in direct contact to the melt, you can use soda ash if you want, or you can refuse - doesn't matter at all.

So both are right.
 
Gold and silver both are fine in contact with alkali.

What happens between carbonate and silver chloride is driven by the production of CO2.
The silver carbonate is thermodynamically unstable and likely decomposes to Ag2O which finally does its own self reduction during disproportioning. Looks like one step but there are several.

To answer the question at hand, sodium carbonate would only instigate the reduction to metal through elements that form thermodynamically unstable carbonates (a good portion of them do easily decompose to give CO2) AND have oxides which further decompose to the elements. So, that's things like gold, silver, mercury, palladium.
 
But in the case of gold this can only make a difference if there are chlorides of those metals left in the gold powder due to precipitation from dirty solution and/or incorrect washing, right? On the other hand this shows again: in the best case soda ash is doing no good but also no harm, in the worst case it can be the backdoor to some impurities which would have been vaporized without soda ash.

Still I have those copper parts in one of my silver buttons. I think I should try to melt copper oxide with soda ash and see if I can get it reduced by the flame, only to see, if the soda ash was the reason for the red copper areas in that silver.
 
I do not think so. I think that is a thin film effect and not representative of you seeing copper reduced into your silver from carbonate.

Give it a shot though, I'd be curious to see the result. I don't imagine it to be possible under neutral or oxidizing atmospheric conditions. I know you can reduce molten copper oxide by methane or CO/H2 introduction.

I think we all should be in agreement that melting gold with sodium chloride or free chlorides in the flux in an oxidizing environment is not good for gold accountability. I have personally witnessed 0.5-0.75 wt % losses of gold that had been inadequately rinsed (still had aqua regia, still had silver chloride in it) at a refiner I consulted for--it was all going to his bag house or making his slag purple.
 
The harsh reality is that anyone who tries to clean a melting dish as I've suggested should achieve the same results---prills of metal forming where there was no visible metal. I may not understand the chemistry involved, but the resulting alloy can't be denied.

Any question that it happens?

Try it, then attempt to explain the phenomenon.

The only thing I know is that it worked every time I cleaned a dish, and I cleaned a large number of them over the years, including my own.

Harold
 
That is the lowering of viscosity (in the slag) and the consolidation of gold vaporized by the torch.
 
I have no clue, although I find this very interesting, and worth further investigation.
It is my thought, that it may not be the soda ash itself that is the reducing agent of the base metal oxides, but the gases from the torch (carbon) and the metal oxides in the fluxing environment of soda ash, (I am not sure if the sodium carbonate would lower the metals melting points or not ??), like how Na2CO3 soda ash can lower the melting point of silica to form glass.

We have the flux here, all we need is an oxidizer or reducing agent and base metals can be oxidized, or metal oxides can be reduced to elemental metal, fairly easily, most of our torches with the fuel we commonly use and oxygen may act as that oxidizing or reducing agent...

Soda ash can be used to reduce some metal salts to elemental metal, (like the sulfides), to either oxides or elemental metal (much depending I think on the reactivity of that metal?) as in the case of lead sulfide to elemental lead when heated in soda ash, also as we know reduced metals can act as collectors of other metals, which can also lead to contamination of our gold...
 
Lou:
I think that is a thin film effect

Correct, scraped it with a knife and what look like massive copper pieces, forged together with the silver, is only a thin film.

But I`ll try with CuO. Since I once had made a 2mm cobalt metal perl from pink cobalt hydroxide only by heating in the reducing flame without fluxing, I think it will work.

Btw. CuO under H2 only needs to glow to form red elemental copper powder at the surface. This reduction is not complete, but enough to show pupils the reduction effect. My bunsen burners at school weren't hot enough to melt it in a glass tube. Today I know, I always used to much CuO and didn't use anything to reflect and isolate the heat.
 
Lou said:
That is the lowering of viscosity (in the slag) and the consolidation of gold vaporized by the torch.
That might be the explanation if the gathering of metal was just gold, but it isn't. It appears to be an average of all the metals that have been melted in the dish. Much like the inquarted stuff I'd create on a daily basis.

With rare exception, I did all my melting for inquartation in a large melting dish.* For that reason, I cleaned it often, as the material introduced certainly hadn't been cleaned, so some junk got included regularly. When the dish got discolored and sticky, I'd clean it by adding fresh borax, then when it was fluid I'd add soda ash. The conversion was quite apparent---you could see the color of the slag lighten as small bits of metal become visible.

If you pursue this operation, you can clean a dish until it's white again. Problem is, a good portion of the dish gets dissolved in the process, so my practice was to clean it until I could see the dish, then stop. I could clean a dish maybe four times before it was too thin to use.

Rather than speculate, anyone with an interest should use a melting dish until it's quite dirty, then give this process a try. Perhaps when you witness the results you might figure out what the hell is transpiring. In my case, when the borax covering became sticky, it was time to clean the dish.

For the record, I examined the dish on several occasions, to see if I could detect any prills. None were obvious, and none were obvious when I introduced new borax and started the heating process. Only when soda ash was introduced and the color changed did small prills start to appear.

*It was faster, by far, to melt with a torch, even if I had to melt more than one dish full, than to fire my melting furnace. I could have the metal melted, gathered and in nitric faster than the furnace could heat up enough to start melting. Only if I had more than 20 ounces of gold would I consider using the furnace. Most of my customers didn't submit that much at one time.

Harold
 
If this would be part of what happens, it would basically be an equilibrium reaction:

Na2CO3 + Cu(BO2)2 <=> 2NaBO2 + CuCO3

but as we all know, CuCO3 is not stable under those high temperatures and will decompose:

CuCO3 => CuO + CO2

and by developing CO2 which disapears from this system, the equilibrium will completely be on the right side:

Na2CO3 + Cu(BO2)2 => 2NaBO2 + CuO + CO2

At some point all borax, Na2B4O7, will be converted to sodium metaborate, NaBO2, and will no longer be able to bind base metal oxides. Now, the reducing flame can do what a reducing flame has got to do: reduce the basemetal oxides, which are partly covered and prevented oxidation by the molten inert salts and maybe the developed CO2.

So, what Harold says, makes sense: using an old dirty melting dish. Nevertheless, we need simplified versions of reality to have a chance to prove the speculations. I think I should prove the equations above with stoichiometric amounts, so all borax is converted.

Feel free to criticize my theory where you think it is weak - my chemistry skills are not much above 13th grade (if at all^^), since my main business is didactics and quite not scientificity.
 
While I really do appreciate debating this issue from a chemistry standpoint, I can verify Harold's findings from a hands on practical experience perspective.

Years ago, in the days of big time jewelry manufacturing in the USA, I received in drums of polishing sweeps for incineration, crushing and sifting. You would not believe some of the things that were found in those drums. As a result, the drums were dumped into a burn tray and all of the items we considered not worthwhile processing were removed. This included melting dishes. We looked at the dish and if there were no visible beads we threw them into a separate drum. (Every barrel had a few dishes from shops with multiple bench jewelers.) Most refineries have a drum like that, the "Yeah we will deal with that when we have to drum!" If the dish had beads it stayed in the lot because the beads always came out in the metallics fraction which was melted into a bar if there was enough so the customer was credited.

So fast forward a few years and I was sitting on a few full drums of clean looking melt dishes, some with a green tint to the glaze and some with a purple tint to the glaze. By this time I had also acquired a jaw crusher which made breaking them up for milling quite easy. So they were broken up in the jaw crusher and ball milled to a -60 mesh powder. I do remember there were no oversize beads so it all went through the sifter easily to make an easy sample-able drum of prepared material. I really did not expect to find much if anything of value in there but I have always lived by the refiners rule that anything that touches gold is worthy of consideration for recovery. I was shocked to see the fire assay result indicated just a slight bit less than 1/4 percent. Now this is a low grade sweeps product for sure but it meant that the drum and a half of material we had considered to be worthless was holding 20 ounces of gold. That translated to a hair over 1 gram of gold per pound of ground up dishes. These are dishes that I personally saw and they appeared to be clean.

So I offer this as practical first hand proof of what I thought to be clean, bead free melting dishes, actually contained valuable gold which was not apparent on inspection.
 
I've cleaned my share of dishes/crucibles and do it every quarter at our premises. On one consulting trip I found 2/3 of gold in less than 10 dishes just from poor melt hygiene. I can also tell you why you note the appearance of tiny BBs when using the torch, specifically Harold.

As we've both cleaned our share of dishes, in virtually the same fashion, I'm able to comment on what I've seen:

When I use the torch flame, be it rosebud, or be it a cutting tip (which I actually prefer for small dishes and small Pt amounts), I have a wide distribution in temperature zone.

At the hottest part of the C2H2/O2 flame, the flame is 3000 C. Also, the oxidizing zone of the flame is different from tip to tip, fuel gas to fuel gas, and may have portions that are neutral/reducing. I prefer H2/O2.
When I heat the dish, I'm often are able to develop/aggregate little prills of PM and base metal alloy in to larger ones just by pushing them around with the force of the flame.

Where that hot spot of the flame hits a part of the dish, and silica being a poor conductor of heat, it gets locally very hot and the reaction quite aggressive with the soda ash--there I can temporarily exceed the breakdown temperature for some metal oxides, turning them back into metal and they agglomerate easy due to the turbulence of the boiling flux mixture. That's my guess as to what's happening, a local phenomena.


You will notice that the dirty flux in the dish will yield BBs of metal, but it will never come out perfectly clear. Some base metal oxides always remain in the slag.
 
Back
Top