Smelting

[4metals]

Alexx,

No you do not have to reach the oxide melting temperatures for them to form. It is posted to show they will exist in the oxide form and not decompose or volatilize until you exceed those temperatures. Most are quite high and not attainable in a gas furnace.

In cases where you will exceed the oxide melt temperatures, like lead, checking the boiling point for that oxide shows a temperature not likely to be seen (2961F) so it will remain fluid and not gas off significantly until it is caught up in a slag. Antimony oxide also melts at a low temperature but doesn't boil until it gets to 2597F.

 

[4metals]

Kurt,

The circuit board refining processing I set up were all out of the US where labor is cheap so boards are depopulated by many hands. Without ever realizing its necessity (which you have pointed out) the stripped boards were always sheared to thin strips either manually or by a machine. So all I ever saw was a rather well burnt ash. The fact that the smaller pieces aided this makes perfect sense.

And I welcome your posts from the Refinement of Precious Metals forum.

 

[4metals]

I have always believed that a successful refinery has good strong analytical support. To that end, all of the materials I collected as shot from the molten pool discussed above were analysed for metals other than copper.

Please note that the process of oxygen or air sparging will cause the less noble metals to oxidize first and when the base metals are mostly gone and a majority copper remains, some copper will begin to oxidize as well. For this reason, good analytics and experience with gassing time and gas flow vs. the relative starting concentrations of less noble metals than copper allowed us to learn to control the gassing without too much copper oxidation. I do have concerns with this process (O₂ sparging) being used without a proper lab to support it.

I would like to see this thread evolve to the more realistic approach as the analytics will not exist for smaller producers. The more realistic approach involves obviously better pre treating by removing components and dealing with the accumulation in the slimes of some of the undesirable metals. Some metals will go into the electrolyte and cause issues with copper purity necessitating electrolyte change but I think that will be easier to deal with for a medium sized producer without a decent analytical lab.

 

[kurtak]

Here is are a couple pics of a hammer mill I picked up last year (just before the snow started to fly here) that I hope to incorporate into my processing - now I have to find a motor to run it - it's going to have to be a gas motor as it's going to take "at least" 20 horse power & when you go over 10 horse (electric) you need three phase which I don't have --- paid $350 for the hammer mill.

Also a pic of a roller mill I picked as free scrap - not sure if, how or where I will incorporate it - but it's not going to the scrap yard any time soon.

I am using a cement mixer for my ball mill - it works "very well" but it is a small one & may have to up grade to a bigger one.

Kurt

Edit; - oops - forgot to add the pics

 

[kurtak]

Something that is important to understand about smelting (which 4metals has sort of pointed to but not really explained) is that when smelting you are actually preforming chemistry - in other words there are actually chemical reactions taking place in the crucible - things are not just melting - but rather when things reach certain temps - reactions start to take place - some reactions start to take place even before things start to become molten while other reactions take place once things become molten.

Here is another post I made on RPM that "in part" explains this - it was in response to a question asking if baking soda could be used as a flux & also if glass (silica) could be used as a thinning agent instead of fluorspar --------

There needs to be a few corrections to the above (question asked on RPM) - first baking soda is sodium Bi - carbonate not sodium carbonate --- soda ash (or washing soda) is sodium carbonate.

Baking soda produces most of the same results as soda ash - however it (baking soda) liberates a lot of carbon dioxide at molten temps which has some "reducing" effects in the smelt (part of the chemistry) BUT - the escaping gases also tend to carry off precious metals as a mechanical loss & therefore is NOT recommended for use in smelting flux --- it is "sometimes" used in assay flux of "lead assays" to help reduce lead complexes to lead.

Glass (silica) is not a thinning agent - at least not to any degree to make your flux fluid enough to achieve the good circulation you are looking for.

Soda ash (sodium carbonate) will help make the flux run "a bit" thinner - but likely not thin enough --- which is why I said earlier that if your initial flux/material load is not running fluid enough when it all gets molten you may need to add some fluorspar along with some soda ash - the fluorspar is doing the real thinning & the soda ash is bringing the melt temp down on the fluorspar so it can do its job of thinning.

Fluorspar - like silica has a "high" melt point & like silica needs the other flux components to liquefy it (the chemistry again) --- also it is not the fluorspar in & of its self that is hard on the crucibles.

Flux chemistry ---- Fluorspar as a flux by its self is a "neutral" agent - being neutral it has no "corrosive" properties in & of its self.

On the other hand - other flux components have ether acidic or basic properties - both of which are corrosive - though bases & acids work to cancel each other out (make neutral) you are not going to get a neutral flux nor do you want one - you need chemistry to happen --- so you are going to have the corrosive effects caused by ether basic or acidic flux --- what happens is that when you add the fluorspar - due to the thinning - the effect of ether the base or the acid corrosion becomes more effective ------ handling the slag (after it cools of course) will cause chemical burns just like getting acid or sodium hydroxide on your hands - wear gloves.

Because of the acid or base properties of different flux components you get the chemistry happening that then preforms tasks in the smelt - those tasks are oxidation &/or reduction &/or complexing (complexing being both oxidation & reduction at the same time by the same flux component).

Some flux components (in fact most) are complexing fluxes whether they be acidic, basic --- what that means is they do both oxidation & reduction in the smelt - dependent on what they are reacting with in the material being smelted.

Soda ash is a "basic" (chemically speaking) flux - but also a complexing flux.

Borax & silica are both acidic fluxes (in fact silica is a "powerful" acid flux) - they are both also complexing fluxes.

There are other components that can be added as well - all dependent on the material being smelted & the tasks you need preformed in the smelt.

I will make a more complete list along with their tasks later.

Kurt

 

[kurtak]

Oops - I thought I had posted this "before" my last post (which was copied from RPM) but see now that I didn't.


The discussion on RPM started with someone asking if he could smelt boards by simply putting them in the crucible, then in the furnace & heating it up to a melt temp --- the following was my reply to that question (& should have been posted before my last post) ------------

You "first" need to incinerate your material - you then need to mill it - then you need to make up a flux of 50 % borax & 50% soda ash - this is the basic "starting" make up for your flux which you may want to alter a bit one way or another depending on your material composition.

You will "most likely" want to add some fluorspar to that borax/soda ash mix - the fluorspar is a thinning agent that helps your slag (flux after it becomes molten) run more fluid thereby providing better circulation in the crucible which in turn improves "metal collecting" during the smelt as well as allowing the metals to settle to the bottom of your cone mold when you make the pour (it doesn't take a lot).

Also - if ceramics are involved (like ceramic capacitors) you will need to add some cryolite to the flux.

It is best if you get anhydrous borax to make your flux rather than using regular borax (like 20 mule team) in order to reduce the problem of excessive foaming up.

Once your starting material has been incinerated & milled you then want to mix your flux in with it using 2 - 3 time flux to the material to be smelted - just how much flux to material depends on the material - that depends on how much & what you are trying to "slag off" in order to collect your metals.

In other words how much ash, glass (from fiberglass) silicon (from chips) metals oxides, ceramics, etc. needs to be slagged off from the metals - so it's a question of organics/nonmetals to metals content question.

Your metal content also needs to be high enough so that you get "good" collection of the metal otherwise you will have a problem with "small" balls/beads of metal hanging up in the slag - that means you "may" have to add more copper if the copper content is not high enough in your starting material.

Smelting copper requires both a relatively high metal (copper) content & a very "fluid" molten flux/slag (which is why I say you will most likely need fluorspar in your flux) to get good collection of the metal.

Your flux needs to be "well mixed" with your material - I mix mine by running it back though my ball mill before going to smelt.

Some things you need to know

1) When smelting you are not just melting things by bringing them to a high enough temp to get everything molten - chemistry is also happening meaning chemical reactions of oxidation &/or reduction are also taking place - that means "fumes" (not smoke) are being produced so fume control needs to be considered.

2) "Some" fumes (not all but some) that are of concern are beryllium, cadmium, & lead if those metals are present - these metals vaporize at the high smelting temps.

3) The flux is hard (very hard) on crucibles - so you want "good quality" crucibles like these - Homepage | Morgan Molten Metals Systems or these http://www.morganmms.com/products/crucibles/ultramelt

4) You will have an initial foaming up as things get molten & reactions start to happen so you want to start with your crucible only about 1/3 full & no more then 1/2 full - the foaming will eventually go down & you can then make small additions of more flux/material watching the foam up so they don't over flow the crucible.

5) You will have a greater problem with foaming if you use regular borax instead of anhydrous borax.

6) After the foaming settles down & everything is molten you want to see the flux/slag rolling & fluid - not sputtering - this indicates you have good circulation & therefore good collecting (of the metal) going on - if you are not getting good circulation add some more fluorspar (it doesn't take much) & some more soda ash.

7) Be sure your mold is HOT when you pour to it or you may have an explosion.

(8) Your mold needs to have a layer of carbon lining it or your pour won't release from the mold - I do this by putting a little cooking oil in the bottom of the mold & brushing it up on the sides & let it get "smoking" hot --- the excess oil will flame up when making the pour so watch out for that.

Kurt

 

[kurtak]

4metals has already posted a list of flux ingredients - here is another one - it has a few ingredients that where not in 4metals list.

Flux used in smelting &/or fire assay along with their chemical properties & the tasks they preform

1) Lead Oxide (Litharge) chemical = basic - task = complexing

2) Sodium Carbonate (Soda Ash) chem = basic - task = complexing

3) Sodium Bicarbonate (Baking Soda) chem = basic - task = complexing

4) Potassium Carbonate (Pot Ash) chem = basic - task = complexing

5) Silica (Quartz, Sand, Glass) chem = acidic - task = complexing

6) Borax (decahydrate - like 20 mule team) chem = acidic - task = complexing

7) Borax (Anhydrous = no H₂O = glass) chem = acidic - task = complexing

(8) Calcium Fluoride (Fluorspar) chem = neutral - task = thinning agent

9) Calcium Oxide (Lime) chem = basic - task = complexing

10) Ferric Oxide (Hematite) chem = basic - task = oxidizing

11) Potassium Bitartrate (Tartar) chem = basic - task = reducing

12) Charcoal (Carbon) chem = neutral - task = reducing

13) Flour or Sugar (Carbon) chem = neutral - task = reducing

14) Potassium Cyanide (Cyanide) chem = neutral - task = reducing

15) Potassium Nitrate (Saltpeter) chem = basic - task = oxidizing

16) Iron (Iron) Chem = basic - task = reducing

17) Sodium Chloride (salt) chem = neutral - task = used as an assay cover

Edit to add; - (18) Cryolite --- used to dissolve ceramic (alumina/aluminum oxide) in smelting & I am not sure of the chem properties & or the actual task but assume it is a base due to being a sodium complex (Na₃AIF₆) & believe it would be considered a complexing task as it works by breaking the oxygen bond in the alumina (reduction) but does not completely reduce the alumina to aluminum thereby allowing the remaining (now dissolved) aluminum complex to be carried off in the slag.

Kurt

 

[kurtak]

Collector metal in smelting.

The discussion about collector metal so far has been focused on copper & copper is a good collector especially if your starting material is already high in copper content such as the case with smelting whole circuit boards - silver is another option as a collector & in my opinion a better choice *IF* the material being smelted has "little or no" copper to start with --- also - lead can be used as your collector (which kjavanb123 has posted a fair amount about in some of his threads).

Here is a bit about copper vs. silver as collectors that I posted on RPM.

In assaying, lead is used as a collector metal - in smelting copper &/or silver are the common metals used for collecting.

Silver is a much better collector then copper for two reasons - first it does a better job of collecting & second it is easier to part the gold from &/or PGMs with the exception of palladium due to solubility of Pd in nitric (whether direct dissolving in nitric or running a silver cell).

To better understand what I mean by silver doing a better job of "collecting" you can do an experiment - melt some silver "powder" in a melting dish - you will get some silver that forms balls/beads on the walls of the dish - now - once you have a nice molten pool of silver in the bottom of your dish rotate the dish around & watch the pool of silver "suck" up the balls/beads with an almost vacuum like effect as it makes contact with the balls/beads.

Now - do the same thing with some copper powder (the powder helps to cause the beading you want for the experiment) now try rotating the molten copper pool to collect the beads - you will see that the copper pool does not "suck" the beads up like the silver did & that you actually need to use some jiggling of the dish in order to create some force in the copper pool hitting the beads to get the copper pool to collect all the beads.

What that means is when using copper as your collector - you need your flux/slag to run "more" fluid so that you get better "circulation" in the crucible in order for the copper to do its job of collecting --- so - thinning becomes more important with copper smelt collecting & the prevention of metals hanging up in the slag.

Kurt

 

[bmgold2]

Are there any other "thinning agents" besides Calcium Fluoride?

I have been playing around with trying to cupel a tiny piece of sterling silver with some store bought flux. I have been using a MAPP gas torch running on propane from a 20# tank. That did get the lead beads to form from the flux but getting the lead to reduce and soak into the tiny homemade cupels is taking a long time. Since I used silver and not gold for this experiment, it is hard to tell when all the lead is gone. The remaining bead has not "sprouted" when it cools so it either isn't absorbing oxygen from the torch flame (probably not an oxidizing flame) or there is still copper or lead in the bead and not pure silver.

The whole idea was from the Butler torch assay manual. I was just trying to use a more portable torch than oxy/acetylene. I'm convinced that a chemical test will be better for what I was looking for but the torch assay and smelting has always interested me.

 

[4metals]

The list that Kurt has posted is a rather complete list which can be found in some of the older assaying texts. The list I posted contains a some but not all of those ingredients. The reason I left some out is because they are obscure and rarely used and the reason I left lead and lead oxide out is, well because it's lead.

Lead is used as a collector in fusion assays and it is very effective. It is also added in ratios greatly in excess of what we will add in a smelt, so it is effective but you need a lot. However fusion assays are followed up with cupellation which removes the lead and absorbs it into the cupel along with base metals. If we were wanting to refine large lots in 3 to 5 gram increments, that would be wonderful but not practical. In general for smelting you don't want to add anything that will be an issue to remove later on.

Silver is a better collector and I encourage readers to try the little experiment (that Kurt proposed above) involving the beads collecting in a melt dish just to see it. The silver is a better collector in the absence of flux because, being noble, the little balls of silver don't oxidize while the little balls of copper get an ever so slight oxide surface which makes them not as efficient at coming together. However, add a few grains of flux to the little test and you will see, as Kurt mentioned, that slight resistance is gone.

 

[kurtak]

Silver is a better collector and I encourage readers to try the little experiment (that Kurt proposed above) involving the beads collecting in a melt dish just to see it. The silver is a better collector in the absence of flux because, being noble, the little balls of silver don't oxidize while the little balls of copper get an ever so slight oxide surface which makes them not as efficient at coming together. However, add a few grains of flux to the little test and you will see, as Kurt mentioned, that slight resistance is gone.

Correct - When smelting - the copper gets the surface oxidation when the smelt load is heating up & before things become molten - this surface oxide does not just "suddenly" go away as the load becomes molten - like any chemical reaction it takes time for the reaction to work (in this case for the copper oxide to dissolve & go off in the slag). As the load becomes molten it starts forming beads/balls of molten metal & in the case of copper the beads have the surface oxide on them - the more fluid the flux is when the whole load becomes fully molten the more effective the chemistry of the flux is at dissolving the oxide away so that when the beads/balls contact the pool of metal forming at the bottom of the crucible the pool will collect them.

It is important to have a fluid flux whether smelting with silver as your collector or copper as your collector - it's just more important if copper is your collector.

I was made aware of this when reading my book on fire assay (Fire Assay Home Study Course) from Action Mining - which states --- "When using copper as a collector, the proper constitution (thinning) "must" be produced that will allow circulation within the crucible to cause the ore to come into contact with the pool of copper at the bottom".

And by the way - I *highly recommend* that anyone wanting to get into smelting to get a book(s) on fire assay - smelting is a part of fire assay & so a good book on fire assay is like reading Hokes for a clear understanding of the wet chemistry - whereas a book on fire assay will give you a clear understanding of smelting.

Kurt

 

[kurtak]

General Flux Chemistry &/or purpose.

As already mentioned, flux preforms different tasks for the purpose of allowing certain impurities to be carried off in the slag leaving you with molten metals that can then be separated from the slag when poured to a cone mold.

The separation of the impurities is made possible due to chemistry taking place in the crucible when the flux becomes molten --- or in some cases at different temps such as in the case of converting silver chloride to silver wherein soda ash is used as the flux but you need to bring the Na₃CO₂/AgCl mix up to a temp just below the melting point of the Na₃CO₂ (I believe 800°F - will have to double check that) & hold it there for a time for the chemical reaction (reduction) of the AgCl to Ag & then after the reduction has taken place raise the temp to finish the flux/Ag molten stage of the smelt.

The amount of impurity separation you can achieve depends on the make up of the material you are smelting as well as on the make up of the flux ingredient - organics for the most part are not a problem - however when it comes to metals & their complexes it's another story because some metals can "at least to a degree" be oxidized & slagged off & on the other hand some metal complexes will under undergo reduction & end up in the metal as a part of the metal alloy.

So the separation of impurities has limitations based on the material being smelted along with the ingredients in the flux make up & the chemical reactions they are preforming.

The reactions (tasks &/or flux chemistry) taking place in the crucible are oxidation - reduction - complexing --- the problem being that generally these reactions are in combination with each other - meaning that while some things in the smelt are being oxidized - some things are also being reduced & yet other things are being complexed - "depending on the material being smelted" - you can "influence" the task you want preformed in the smelt - whether it be oxidation, reduction or complexing by adjusting &/or adding different ingredients to the flux make up based on the material being smelted --- the less complex the material the better you can control your smelt outcome with your flux composition (ingredients used in your flux make up).

Examples - AgCl (as mentioned above) a less complex material (you just need to break the chloride bond to achieve the reduction from AgCl to Ag) & you only need soda ash as your flux to achieve reduction with smelting. Now on the other hand - incinerated circuit boards which is a more complex material because of organics, metals & most likely metal oxides so your flux ingredients also become more complex - starting with your basic flux of borax/soda ash which as a complexing flux will slag off the organics - but it will also cause some oxidation of some metals but also some reduction of some metal oxides depending on what metals &/or oxides are involved - so besides upping the soda ash &/or adding fluorspar for thinning you may want to add cryolite if the ceramic capacitors are left on the boards &/or you may want to add charcoal (carbon) to help reduction (prevent oxidation) of base metals acting as your collector &/or add some potassium nitrate to effect better oxidation of things in the smelt --- or you can add silver as your collector along with litharge to oxidize the base metals leaving you with PMs collected in silver instead of copper --- in other words you can "influence" to a point the outcome of your smelt with adjustment of flux ingredients.

In the list of flux ingredients you see that some are acidic some bases & some neutral & they preform tasks of oxidation or reduction all of which is kind of self explained but then there is also the complexing fluxes - I have kind of explained this but will try to do a better job here.

Complexing fluxes are fluxes that preform both oxidation & reduction in the crucible - what that means is that they (complexing flux) can perform ether oxidation - or - reduction --- depending on what it is they are reacting with - in other words they may oxidize one component of the material while also reducing another component of the material being smelted - or - they cause more oxidation but also some reduction or vice versa.

In other words - a complexing reaction in smelting is like a redox reaction in wet chemistry - redox being both reduction & oxidation taking place in the same reaction.

A good example of a redox reaction in wet chemistry is cementing silver with copper wherein the copper is being oxidized (dissolving & thereby going into a copper plus state of copper nitrate) & at the same time in the same reaction the silver is being reduced from its silver plus state (silver nitrate) to elemental silver.

A good example of a complexing reaction in smelting is using iron to smelt silver sulfide to silver with soda ash as the flux - the sulfur in the silver sulfide oxidizes the iron (putting it in an iron plus state) which in turn reduces the silver sulfide to silver with the iron (plus) then being carried off by the soda ash flux to form the slag.

Kurt

 

[alexxx]

A few more questions...

- neutral flux & neutral flame
Not sure if I understand it the right way. But a neutral flux/flame will accomplish nothing on the "chemistry" taking place during fusion. No reduction nor oxidation. A neutral setup would be used in a scenario where one would simply want to melt and recast. Like when 4metals suggests making an anode from previously poured shots. In this scenario, I have no oxides to remove, no base metals to oxidize further and I dont want to lose copper.

- when to stop the melt
for anyone wanting to introduce forced air/oxygen into a molten pool to drive oxidation of base metals. Or simply in a scenario where I want to oxidize my base metals and want to preserve as much copper as possible. What signs are telling me that it's enough? When do I know that most of my base metals are gone and I need to stop the reaction taking place to save the copper? When do I know that I have removed enough slag and I dont need to add more flux to clean the melt further?
Any specific signs ? Color of the slag ? Color of the molten pool? Color of the cold alloy?

-Lead as a collector
Kevin posted many procedures on recovery/refining using lead /PbO as a collector (+ silver also).
How viable is this route for the magnetic fraction removed from the fines & oversizes?
I prefer the approach of cupellation & parting instead of skimming the pool of oxidized lead on the final stage in order to drive the lead.

edit : 1 more here

- Is it possible to change the type of flux during the same melt ?
Let's say I first want to get rid of a maximum of base metals by using a oxidizing flux & flame, then, after skimming a few times my slag full of base metal oxides I want to introduce a reducing agent in the melt, is it practical ? Is it something that is done ? Like a 2 stage smelting to get the benefits from both oxidizing & reducing agents after the complexing fluxes have worked to their full potential ?

 

[Lou]

Great information!

One quip on the AgCl conversion via carbonate:

"The separation of the impurities is made possible due to chemistry taking place in the crucible when the flux becomes molten --- or in some cases at different temps such as in the case of converting silver chloride to silver wherein soda ash is used as the flux but you need to bring the Na₂CO₃/AgCl mix up to a temp just below the melting point of the Na₂CO₃(I believe 800°F - will have to double check that) & hold it there for a time for the chemical reaction (reduction) of the AgCl to Ag & then after the reduction has taken place raise the temp to finish the flux/Ag molten stage of the smelt"

The AgCl is actually the solvent for the carbonate. I should emphasize that is imperative NOT to "finish" it off in that fashion though Kurt. You will have silver losses, greatly damaged crucible and quite a bit of unnecessary and unhealthy smoke.

This is what is occurring:
(1) 2AgCl + Na₂CO₃ → Ag₂CO₃ + NaCl
(2) Ag₂CO₃ → Ag₂O + CO₂
(3) 2Ag₂O → 4Ag + O₂
Reactions (2) and (3) occur by thermal decomposition (disproportionation to the elements). The silver carbonate of reaction (2) will completely decompose by 225°C. The Ag₂O of reaction (3) completely decomposes at 340°C.

Wasyl Kunda optimized this procedure and found that the ideal holding range for conversion is 500-600°C. Too high and sintering of the silver cake occurs and it is difficult to leach out the NaCl, too low and the reaction is either incomplete or very slow. Time is dependent mostly on how long it takes the middle portion of the AgCl to melt and react with the solid carbonate.

In any event, if using this method, the silver should be leached in hot water before subsequent melting to remove the NaCl.


As for books to recommend--anything to do with fire assaying.

As for Alexx's question about when to stop the melt, that's kind of difficult to do--there is a lot of visual feed back but not as much as with the Miller chlorination process. When I used to do this, I would test it out in a #16 SiC crucible with SiC pipe to supply the compressed air. When melting 5K oz heats for anodes, you can pin tube a sample, quench and XRF to follow the progress of the removal of the Sn(basic flux)/Cu(borax)/Pb(borax/silica). Rocking reverberatory furnaces are of course ideal for this, but silver does report to the bag house and refractory erosion is always an issue. It is possible to take a bunch of "low" grade (say 70-85% Ag basis) and upgrade to the high 90s and preserve electrolyte life. Little silver is lost and usually paid upon when the copper/tin slags go off for refining.

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

 

[richard2013]

Just a clarification calcium fluoride or calcium fluorite? Am having trouble where to buy these please

 

[4metals]

It is calcium fluoride also called Fluorspar.

 

[4metals]

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?

 

[g_axelsson]

I was going to say it sounded like the classical method of poling to remove oxide from copper by sticking fresh tree poles into the melt. When I googled it first and found the wikipedia article Poling (metallurgy) - Wikipedia I realized it was the same process that Lou is talking about.

Göran

 

[ericrm]

when you are blowing gasses into a melt, what do you use for the part that goes inside the crucible? Does anyone have a picture to help me visualize it?

 

[4metals]

usually a quartz glass tube

http://www.technicalglass.com/product_pages/fused_quartz_tubing/fused_quartz_tubing.html