Smelting

[kurtak]

Lou

Thank you for clarifying the AgCl smelting --- I have only done it once & that was quite some time ago - I followed these instructions by samual-a (wherein Sam says to bring it up to the decomposition temp & then when decomposed take it up to melt temp) simple question.

This was back before I had a silver cell set up so I was still making AgCl deliberately & doing the lye sugar conversion so tried the smelt method in the link above - shortly after I got my cell set up & quit making AgCl deliberately.

I do have some that has been accumulated from testing solutions etc. so may give it another try based on the info you just posted just for the learning experience.

Kurt

 

[kurtak]

when you are blowing gasses into a melt ,what do you use for the part that goes inside the crucible?

usually a quartz glass tube

Or these Homepage | Morgan Molten Metals Systems

Or Silicon Carbide Degassing Rotors | Morgan Molten Metals Systems

Kurt

 

[kurtak]

When I first started posting about smelting on RPM it was because a guy was trying to smelt some boards by simply putting them in his furnace & bringing it up to melt temp - after posting the info I have copied to this thread he then asked --- why can't you just put the boards in the crucible & add the flux - why do you have to incinerate & mill first?

Though the answer to that (for most) may be obvious - it was actually a good question prompting my following answer -----------

There are "several" reasons why you incinerate & mill the circuit boards first.

First & foremost is that you "need" to break down the chemical composition of the epoxy resin - epoxy resin & its hardeners are just that - "compositions of chemicals" - including acids --- when you heat epoxy (to the temps we are talking about with smelting) chemical reactions of the epoxy itself takes place (reduction) producing a wide range of (chemicals) gasses &/or vapors (some of the vapors being acidic) --- as these gasses & vapors are released in the higher smelting temps they are going to produce chemical reactions of their very own thereby having a direct effect &/or altering of the "desired" smelting flux chemistry --- in other words if you don't first incinerate it will cause unknown & most likely undesired reactions to occur.

Incineration of CBs is the same principal that is often used with ores by first (before smelting) doing a reduction &/or oxidation roasting to them to drive off (for lack of better term) chemical components that can cause undesirable reactions in your smelt & its flux chemistry.

Incineration drives off the chemicals that make up the epoxy as gasses & vapors thereby "reducing" it to ash - ash being comprised of sodium, potassium & calcium all of which are readily handled by your basic flux composition (borax/soda ash).

Also - because incineration reduces the epoxy to ash you can reduce its particle size with milling --- just like with wet chemistry particle size plays a role in both the time & how hard the process has to work to do its job --- it takes more time to dissolve a chunk of metal with acid than dissolving fine powders of metal & the acid has to work harder on the chunk than on the powders (which is why we usually add heat when dissolving solid metals).

The same is true with smelting - the finer the ash, silica (fiberglass, etc.) ceramic (ceramic capacitors) etc. is - the less hard the flux has to work to slag off the impurities & that means less time for the smelt to do its job & that means the less time it is in the crucible which in turn gives the flux less time to work at destroying the crucible.

Also - having the flux "well mixed" with the material being smelted is important & again can be compared to wet chemistry - if you have say a 1/2 inch of metal powders covering the bottom of a beaker - when you first pour your acid in you get a real good reaction (lots of foaming etc.) but then it settles down & slows down working at dissolving just the metal on top that the acid is in contact with but not the metals in the middle &/or bottom - now give the metal powders a good stir so they lift off the bottom up into the acid & the reaction will take off with more aggression again because the acid is now making better & more complete contact with all of the fine metal particles (its more homogenized).

So pre-mixing the flux & material to get it homogenized to start with is just more practical than trying to do it during the smelting - which for one thing would require un-necessary opening of the furnace.

And finally - by incinerating & milling you can now concentrate the material you are smelting down to a higher concentration of metal & lower concentration of organics which in turn reduces the amount of flux needed to slag the organics off in order to melt & collet the metals you are after - in other words you not only reduce the volume of material to be smelted - but the amount of flux as well which in turn reduces the amount of slag produced.

So as you can see - incineration & milling play a BIG & important role in the process before smelting.

Kurt

 

[alexxx]

How carbon can affect your melt ?
Let's say that your incineration is incomplete and lots of carbon remains in your ash.

 

[butcher]

How carbon can affect your melt ?
Let's say that your incineration is incomplete and lots of carbon remains in your ash.

Carbon is a reducing agent in a melt.

With carbon or materials in a flux, like sugar, sawdust, or flour that will break down into carbon with the high temperatures, the carbon acts as a reducing agent, let's look at reducing copper oxide in the melt.
Copper oxide + carbon
CuO + C --> Cu (s) + CO (g)
2CuO + C --> 2Cu (s) + CO₂ (g)

Above we see where the oxidized copper salts of copper oxide (CuO) is reduced to metal copper {Cu (s)},in the melt, with carbon (C), in this reaction this results in (-->), the carbon (C) removes the oxygen (O) from the copper oxide (CuO) to form carbon oxide gas {CuO(g)}, or carbon dioxide gas {CO₂(g)},and the metal copper {Cu(s)}, depending on the ratio of the carbon and the oxides or oxygen of the melt.

Many metals salts or oxides above copper in the reactivity series can be reduced to metal by carbon in a melt.
Some metals are so reactive that carbon will have a hard time reducing them.
Then we also have some of the very reactive metals which cannot be reduced at all by carbon in a melt.
This is why in many reactivity series of metals we see carbon added in the list of reactivity.

Carbon can also take oxygen from other sources like the oxidizing nature of the flame of your torch, say you are trying to oxidize a metal with the torch's oxidizing flame, and you have carbon involved in your melt (or flux), the carbon will take the oxygen so that you may not be able to use the torches flame as an oxidizer, carbon in the melt may also take oxygen from other ingredients in your melt, like an oxidizing agent or flux (PbO, KNO₃...).

Let's take lead oxide (litharge) for example.
Say we have added litharge (PbO) to a melt for two reasons, for an oxidizer to oxidize base metals, and in doing so becomes itself a metal of molten lead to collect the values in the melt.

Having carbon in the melt or an added ingredient of the flux, can counter-react with the oxidizing nature of our oxide in the lead oxide (litharge), where we may not be able to oxidize the base metals in this melt, with too much carbon all of the metals that can be reduced by carbon could be reduced with the lead, and remain with our values, defeating one of our purposes for choosing litharge (lead oxide) in our flux, instead of the choice of using metal lead in the melt.

Everything that can react in the melt can, or may react, some metal oxides can act as oxidizers for other metals, as well as oxygen from the air, or even from your heating source, like air from the atmosphere, or oxygen from your torch or burner.
Metals can act as reducing agents in the melt, or oxides of metals can act as oxidizing agents in the melt.

Salts of metals like silver chloride can act as an oxidizing agent for many metals, even gold (which normally is very hard to oxidize), with AgCl in the melt the gold can more easily be oxidized, where some of our gold can vaporize off as gold fumes causing loss of values from the melt as a chemical reaction of the melt. A flux like sodium carbonate and controlling the temperature during the melt can help to chemically react with the silver chloride converting the silver to metal and the chloride to a salt or fumes of chlorine gas, so that the silver chloride does not react with our gold as easily.
Flux ingredients like carbon, can be chosen to change these chemical reactions.

For example we may wish to add a strong oxidizing agent like KNO₃, to a melt high in carbon to chemically react with the carbon, so that we can oxidize the base metals in that melt...

 

[4metals]

How carbon can affect your melt ?

Smelting is a process not unlike a chemical reaction you have going on in a beaker in your hood. Except it happens at a higher temperature, much higher. But let me phrase the question a little differently, because more of us can relate to the beaker.

So why don't you just keep shoveling in the sodium metabisulfite when you drop the gold until no more goes into solution? Unfortunately we do get questions on the forum like this and every time, they are asking because they made a mess. But why a mess? Because the sodium metabisulfite is selective to reduce the gold, to a point. Too much brings on unintended consequences.

Well carbon is also a powerful reducing agent, and if we knew how much was in there to start with we could possibly use it to our advantage. But in most cases there is too much. So just like the beaker example of adding too much metabisulfite, not knowing how much carbon is in the ash to start can lead to problems. Refiners, and smelters, like to work under controlled conditions when possible. By eliminating carbon by incineration, we can now add flux components in a way to better control our results. We can add some carbon, but we know how much and can measure its effect.

Whenever I had large smelt lots, I always fused small samples to determine the best flux combinations and applied that to the entire batch. In sweeps fusions they call that determining the reducing power of the flux. By using this procedure I knew what proportions would give the best results and that little bit of analytics could be applied to the entire lot.

So to summarize, we like to get rid of as much of the carbon as possible so we can selectively add flux components to attain predictable results.

 

[Lou]

A similar process can be done for deoxidation of silver (or copper) by sparging with methane, town gas, or hydrogen.

Now that's a horse of a different color.

Care to go into a little more detail Lou?

Yes, but it's a horse worth betting on if you seek to produce a high purity product on more than a metallics basis. This method I have done on pure silver to remove the oxygen from the silver. Many large refiners that we both know have oxygen issues with their silver product. I attribute it to two things:

1. Poor atmosphere control during granulation and casting (I always shotted under an H₂ flame with Ar over the top to which I fed the silver crystal).
2. The direct remelting of pure silver solids that had been melted once or more before in uncontrolled conditions or "clean up silver" done upon feeds like silver/Ag₂S flake that is done with nitre and borax but specs at 3N. This is to say that some large volume refiners may not refine every feed material.
The oxygen is undesirable for more than cosmetics: oxygen saturated solid silver does not have the physical or electrical properties.

For the lowest O content (and impurities in general), one can have no better material than large crystals out of the silver cell. Silver sponge made with formate always carries more O content, similarly with cement silver. I do not know why. Maybe it has to do with atmospheric absorption. In any event it may be remedied in several fashions, but the easiest is:

by sparging the melt with an iron pipe (driven to a half inch of the bottom of the crucible) and a reducing gas through the silver for a period of time sufficient to deoxidize it. In some instances the silver will react violently to the treatment at first. When the melt quiets down the pure silver may be cast into finished bars with no eruption of oxygen from the cooled bar top surface. If conducting this treatment on copper, a silica pipe is used.


Excess carbon that survives the incineration process is usually detrimental to the coalescence of the melt because it forms gas pockets. One will observe popping at the top of the melt/flux as the carbon gets oxidized into CO then CO₂.

So much with smelting is situational, and I echo 4metals' advice that small samples should be tested first to optimize the process. This is to ensure no "prill holdup" in the slag. Slag chemistry and slag temperature dictate slag viscosity which is what determines if all the values pool or get held up.

Lou

 

[Barren Realms 007]

The oxygen is undesirable for more than cosmetics: oxygen saturated solid silver does not have the physical or electrical properties.

Lou can you point me to some information on the fact that the silver lacks the electrical properties because of the oxygen content? I'm not saying you are wrong in this I would just like to read up on this if you have a link to an article on this. Is this true for just impure silver that needs to be run through a cell or is it true for silver that has been run through a cell also?

Thanks

 

[butcher]

I cannot point to specific papers, and I cannot say about oxygen in silver (silver oxide) crystals?
But in electrical contact points silver oxide can be a problem silver points can burn up easier or hold the arc longer while the points open burning the points up faster or being unreliable, example a higher resistance across the points and a higher amperage of the circuit...

Many contacts, or switches are made to mechanically "wipe" as they open or close to clear them of an oxide film, which can degrade the continuity.

Many times the silver has other metal or even metal oxides added to change the properties of the silver to better suit their use in different electrical or mechanical conditions.

It makes sense that the oxides or salts of metals would not be as good of a conductor as the metal, and that if the metal has an oxide film or oxide crystals in its structure it would not be as reliable of a conductor as a more purer metal.

Electrical copper (wire bus bars...) is normally a fairly pure copper metal, if the copper had more other metals involved its resistance or mechanical property could change to the point of not being reliable or uniform for the use in electrical application...

I suspect similar to gold in the jewelry industry, or the metals electronic industry, the purity of the metal becomes very important, especially in electronics as we are running lower current circuits and faster frequencies in the digital switching or pulses...

 

[alexxx]

Smelting the magnetic fractions

It has been discussed how to smelt the non magnetic fractions of circuit boards.

After sieving and magnetic separation, how do we process the magnetic oversize pieces that remains ?

Lead smelting ? Is there another way to achieve a decent recovery ?
Chemical leaching ?

Any metals worth retrieving besides precious metals ? Nickel maybe ?

 

[Barren Realms 007]

Yes Butcher, I know all of that because I am a master electrician. I was more curious about the following statment :

"The oxygen is undesirable for more than cosmetics: oxygen saturated solid silver does not have the physical or electrical properties".

And how it effected the electrical properties.

 

[4metals]

I know that years ago I ran a large electroplating shop, we did a lot of silver cyanide plating. The anode stubs that come from the undissolved pieces which remain above the solution level as the anodes dissolve were always melted down and cast into bookmolds to make new anodes from the residues. Because of the way we melted them, in a gas furnace with no gas cover to prevent oxygen absorption, the recast anodes were not oxygen free.

There was a huge difference in the performance of oxygen free anodes vs. the recast anodes we made. The recast anodes dissolved in a way that produced scales which flaked off into the anode bags while the oxygen free anodes dissolved beautifully and left a nice crystal pattern on their surface. To the point where the oxygen free anodes did not need to be bagged but the recast anodes did. So there was definitely something electrical going on because it was essentially the same silver just recast in a way that did not exclude oxygen absorption.

 

[Barren Realms 007]

Ok now that train of thought makes sense.

 

[Lou]

Frank, I was told that by a company I sold very high purity silver to, they said the same of gold. I guess interstitial impurities affect the resistivity(?). I'm no physicist, I just figure out how to get a material to fit an application or specification.

I guess that silver oxide disrupts the flow of electrons to some extent. Maybe like it does for copper:

https://en.m.wikipedia.org/wiki/Oxygen-free_copper
It (O) also seems to be bad for many other metals' physical properties. C, H, N, and O are usually on the spec list for me.

 

[Barren Realms 007]

Yea I understand. Theoretically it it would make sense to me with what 4metals said it does. I was thinking how it would effect it along the line of use in electrical components in which case it probably would not come into play because the manufacturer should know this. But with running silver through a cell I can see now where it would come into play and might pop up in a question from a forum member as an answer to a problem that might be encountered.

 

[4metals]

What Barren just said gave me a thought. Everyone casting anode material for silver cells does it by melting and casting without protection from the oxygen in the air surrounding us. So in theory all silver cell anodes have a degree of adsorbed oxygen in them. I have noticed that the copper in sterling makes it behave beautifully and it never seems, I guess seems is the operative word here, to behave as pure silver melted under the same conditions reacts.

However silver anodes in silver cells (Moebius cells) do slough off and behave in a similar fashion to those recast pure silver anodes I mentioned. And the solids remaining in the anode bag as slimes is always a high percentage silver. I wonder if an anode was cast to be oxygen free if it would dissolve better, faster, and with less silver scale in the anode bag. That could eliminate a large pain in the butt factor in cleaning up anode slimes! Worthy of playing around with I think.

 

[alexxx]

oxygen sparging to oxidize base metals during smelting.

It was mentioned that O₂ could be slowly injected into molten metal during the smelting in order to oxidize base metals and have them driven into the slag.

A few questions here;

Any additional stirring needed during that process or the sparging is enough to move the pool of metal around and have the O₂ in contact with a maximum of metal?

4Metals has mentioned the use of a quartz glass tube to inject the oxygen into the melt. Any other material could be used to handle the high temperature and the oxidizing nature of the melt? An alumina tube maybe? What about a stainless steel tube?

I understand that O₂ sparging might be quite dangerous. What type of apparatus is usually used to control the debit of the injected gas? A simple regulator? What debit / pressure would you guys use for a pool of 100 lbs. of molten metal and a 10mm diameter tube (material being again ashes & oversize non ferrous from circuit boards)?

Would you suggest to simply use ambient air or an O₂ concentrator is preferred for the task?

How long would you keep injecting O₂ into the melt & slagging off?

cheers !

Alex

 

[4metals]

I prefer quartz glass but Morgan materials makes gassing tubes as well

https://www.morganmms.com/media/5619/10-dg-tube-and-plunger.pdf
When gassing I have used room air pressurized and pumped into the melt, and I have used compressed oxygen. Whichever you choose, start slowly and only begin with a slow bubbling of gas reaching the surface of the melt. Depending on what you are starting with, gassing can take from a half hour up to hours.

The compressed oxygen obviously delivers more oxygen into the melt which can result in shorter gassing times but I have had good success with compressed air. I will always defer to simplicity.

I like to periodically ladle off slag and pour it into a cone mold. This way you know if you are pulling out metal as it settles quickly in the molten slag and can be collected at the bottom of the mold.

a pool of 100 lbs of molten metal, material being again ashes & oversize non ferrous from circuit boards

Typical PCB metallics are very high in copper to begin with, you are choosing the wrong path if you expect to oxidize all of the copper into the slag, copper is the last to go and usually remains with the PM's. This material is usually shipped to a copper refiner as it produces a homogeneous bar which can be sampled and quantified easily. The copper refiner will upgrade this material to high 90's% copper and refine the copper out electrolytically and segregate the PM's in the slimes. Or you can build a copper cell and refine it yourself.

 

[alexxx]

Typical PCB metallics are very high in copper to begin with, you are choosing the wrong path if you expect to oxidize all of the copper into the slag, copper is the last to go and usually remains with the PM's.

The goal is only to remove other base metals besides copper to obtain a 95%+ copper anode quality alloy at the end for a copper cell. 99% prefered, even if it implies a lost of 5% in the copper content.

It was pointed out that copper will react in a different way than other base metals with this procedure, a small % will be oxidized and lost to the slag while a great % of other contaminants such as Zinc, Nickel, aluminum, tin would be removed.

I believe I could live with a sacrifice of a few % in the copper content to save a lot of troubles in prematurely fouling the electrolyte from the copper cell.

 

[4metals]

For what you just suggested the oxygen sparging is fine and actually the preferred treatment. I just don't want you to think all of the copper base metal is going into the flux.