Smelting losses?

[Liquidau]

Just started a smelting trial with a 100 gram powder sample (with 200 grams flux) of 1.5% gold by weight derived from incinerated BGA chips. The problem is I’m getting no gold button (or any metal) at the bottom of the inverted cone mold.
We pyrolyzed and then incinerated the chips in the same device as we’re now smelting, and the temperature hits 1100C. Is it possible the gold (and silver) were lost to vaporization, or could they have melted and bonded to the copper heat sinks? Any other ideas?
I realize we should only heat to 800C maximum in the future…
Thanks forum.

 

[4metals]

What was your flux made up from? If you closely examine the cone shaped flux you ended up with are any beads visible? What was the make-up of the 100 gram starting powder and what was the particle size? Did the flux flow like water or was it thick?

 

[kurtak]

Is it possible the gold (and silver) were lost to vaporization,

NO !!!

Just started a smelting trial with a 100 gram powder sample (with 200 grams flux) of 1.5% gold by weight derived from incinerated BGA chips.

Per the bold print - what type of BGAs

Examples; - North/South Bridge BGAs - BGAs from RAM - other random BGAs from circuit boards - mix of BGAs - etc. etc.

Also to add to 4metals questions --- after incineration did you mill the BGAs to powder

If you milled them to fine powder did you put them through a concentration process

If milled to powder &/or concentrated did you mix the flux in with the milled BGAs - or just put the flux on top

Did you add a collector metal

Did you remove the solder balls &/or other base metals before smelting

If N/S bridge BGAs where they whole BGAs or did you separate the green fiber from the black epoxy top

And as 4metals asked - what was your flux make up

Kurt

 

[Liquidau]

What was your flux made up from? If you closely examine the cone shaped flux you ended up with are any beads visible? What was the make-up of the 100 gram starting powder and what was the particle size? Did the flux flow like water or was it thick?

 

[Yggdrasil]

I see no benefits or reason for these "retweets" of other people questions to you.
If you want help, you need to answer them in stead.

 

[Liquidau]

Thanks guys for the responses…here are some answers to the questions:
The slag composition is very close to the original powder:
SLAG:
Si 45
Cu 38
Sn 10
Fe 7
Al 3
Au 1.43
It was a fairly liquid pour. Flux was sodium carbonate/borax/silica sand 10/4/1.
BGAs were half Broadcom, half another type 1” square with a gold tip on one corner. I did not separate the base from the black top.

 

[Liquidau]

What was your flux made up from? If you closely examine the cone shaped flux you ended up with are any beads visible? What was the make-up of the 100 gram starting powder and what was the particle size? Did the flux flow like water or was it thick?

No beads visible. Sieved to 80 mesh. Quite fluid, but not runny. Flux glass was very brittle, crystalline, reddish-brown colour (Cu?)

 

[Liquidau]

NO !!!

Per the bold print - what type of BGAs

Examples; - North/South Bridge BGAs - BGAs from RAM - other random BGAs from circuit boards - mix of BGAs - etc. etc.

Also to add to 4metals questions --- after incineration did you mill the BGAs to powder

If you milled them to fine powder did you put them through a concentration process

If milled to powder &/or concentrated did you mix the flux in with the milled BGAs - or just put the flux on top

Did you add a collector metal

Did you remove the solder balls &/or other base metals before smelting

If N/S bridge BGAs where they whole BGAs or did you separate the green fiber from the black epoxy top

And as 4metals asked - what was your flux make up

Kurt

All of the BGAs came off what I think were network boards. Not RAM. After incineration, milled in my cement mixer for an hour with different sized steel balls and other steel forms, changing the angle every few minutes. Then used a Gold Cube. Mixed the flux in thoroughly. Used no collector metal. Flux was Na2CO3 10 parts/Borax anhydrous 4 parts/Silica sand 2 parts. Did not separate green from black parts.

 

[4metals]

If you did not have any metallic fraction in the bottom of the cone my suggestion would be to add a collector metal and add about 1/2 part of fluorspar to thin it further.

 

[orvi]

That low ammount of metal would likely disperse through slag and never collect properly on it´s own - sticking to the walls of the crucible etc. Would be better to use some collector metal as 4metals suggested, what metal - sepending on what is your next intention with it. You can add silver (3x the ammount of gold) to obtain bead ready to inquart. Or you can add lead/bismuth and cupel subsequently. Or simply use some reasonable ammount of copper and then refine with AR as usual.

I had this problem with low silver smelts - MLCCs residues smelts, where there were like 2-3% of AgPd alloy left, with traces of many base metals. I smelted in graphite in induction, and despite the slag was runny and liquidy, mini beads of metal were simply refused to coalesce and attached to the phase borderline of slag/crucible, or simply sticked to the graphite wall and that was it. No ammount of stirring, cranking the temperature or other things were helpful. I was forced to add some silver as collector, and at around 3%+ metal concentration, one larger bead started to form and slowly "swallowed" all other small beads when stirred for prolonged time at fairly high temperature (past 1350 °C).

Maybe it has something to do with graphite, I am not completely sure, as I did not tried with some simple clay crucible.

 

[Liquidau]

NO !!!

Per the bold print - what type of BGAs

Examples; - North/South Bridge BGAs - BGAs from RAM - other random BGAs from circuit boards - mix of BGAs - etc. etc.

Also to add to 4metals questions --- after incineration did you mill the BGAs to powder

If you milled them to fine powder did you put them through a concentration process

If milled to powder &/or concentrated did you mix the flux in with the milled BGAs - or just put the flux on top

Did you add a collector metal

Did you remove the solder balls &/or other base metals before smelting

If N/S bridge BGAs where they whole BGAs or did you separate the green fiber from the black epoxy top

And as 4metals asked - what was your flux make up

Kurt

Did not remove solder balls or base metals. I’ve read that others boil the chips in HCl before anything else in order to remove tin, aluminum, nickel, iron, zinc?

 

[Ultrax]

That low ammount of metal would likely disperse through slag and never collect properly on it´s own - sticking to the walls of the crucible etc. Would be better to use some collector metal as 4metals suggested, what metal - sepending on what is your next intention with it. You can add silver (3x the ammount of gold) to obtain bead ready to inquart. Or you can add lead/bismuth and cupel subsequently. Or simply use some reasonable ammount of copper and then refine with AR as usual.

In the presence of a large number of impurities of base metals, lead as a collector is not very convenient because impurities cause liquation of the gold crust and delamination of the melt.
This happens in Au-Pb alloy if the material contains bismuth, zinc, or tin.
If there is a small amount of gold, you will lose it in the oxides' crust when the lead drop cools.

Of the inexpensive metals in dirty conditions, the best collector is copper, since gold will pass into copper as completely as possible without liquation due to impurities of base metals. Some of the copper will go into matte and slag, but the gold will remain in the copper bead.

 

[Liquidau]

Thanks Ultrax, very useful info. Can you just please explain why copper should be added, if it’s already at 38% of the powder mass? Does it need to be so much more?

 

[Ultrax]

If you have melted your material where 38% copper was already present in powder form and did not see a metal(!) bead of copper after melting, there are two options:

(A) You have insufficient temperature during melting (copper in the presence of impurities requires overheating to 1200C to obtain the required fluidity and protect the surface of the melt (for example, with charcoal).

(B) You have the incorrect structure of the charge or flux, which led to the complete oxidization of copper into matte or slag.

This situation often happens with copper powder - small particles oxidize faster than the crystallization centers begin to melt. In this case, it is worth adding to melt compact copper in the form of thick wire or pieces of metal. The best slags for smelting copper are calc-iron silicates with acidity K = 1 - 1.3.

 

[Ultrax]

The slag composition is very close to the original powder:
SLAG:
Si 45
Cu 38
Sn 10
Fe 7
Al 3
Au 1.43
It was a fairly liquid pour. Flux was sodium carbonate/borax/silica sand 10/4/1.

No beads visible. Sieved to 80 mesh. Quite fluid, but not runny. Flux glass was very brittle, crystalline, reddish-brown colour (Cu?)

Try adjusting the slag from basic to at least neutral, this will reduce the fluidity to some extent, but will greatly reduce the oxidized (reddish-brown color) copper content of the slag.

 

[Ultrax]

Just started a smelting trial with a 100 gram powder sample (with 200 grams flux) of 1.5% gold by weight derived from incinerated BGA chips. The problem is I’m getting no gold button (or any metal) at the bottom of the inverted cone mold.

Also, chip substrates are often not made of silicon oxide (SiO2), but of ceramics (Al2O3), which greatly increases the infusibility of the material in total (and the flux with sodium carbonate may well be fluid visually)

 

[kurtak]

1.5% gold by weight derived from incinerated BGA chips.

The slag composition is very close to the original powder:
SLAG:
Si 45
Cu 38
Sn 10
Fe 7
Al 3
Au 1.43

I assume these are XRF read outs ?

Kurt

 

[kurtak]

Did not remove solder balls

The solder balls should always be removed as the very first step in the process (HCl leach of whole chips) if for no other reason then tin in the smelt dore metal will cause problems when you go to refine the dore metal - it also causes problems in the smelt as pointed out by Ultrex

I’ve read that others boil the chips in HCl before anything

Per the bold print - so - yes - removing the solder should always be the very first thing you do - but - you most certainly do NOT need to boil the chips in HCl - warm is all you need - chips in a 5 gallon bucket covered with HCl set out in the sun works fine

There is gold in that solder & we can talk about recovery of that gold later as that is not the issue currently at hand in this thread

Kurt

 

[kurtak]

Just started a smelting trial with a 100 gram powder sample

So this was a "test run" on a larger batch

Then used a Gold Cube

So the 100 grams came from concentrates of the larger batch - Edit to add - I would like to see a pic of the remaining concentrates

After milling did you sift the milled material through an 80 mesh screen or just run it through the cube (this is more a question about your process other then the smelting process)

No beads visible. Sieved to 80 mesh

Flux was sodium carbonate/borax/silica sand 10/4/1.

With that much soda ash in the flux it should dissolve most of the slag in HOT water (simmer boil) --- try that & lets see what we get - post pic

Kurt

 

[minero74]

I Know what I will said is not about computer parts,but maybe help in any way

Many problem in the smelting process
is that many time precious metal
stay in the slag
because the collector metal was wasn’t enough to collect all the precious metals
or the Heat It wasn't enough to melting all the material

When I was smelting ore
I collect more metals,gold,silver,ect
with Copper as collector than with lead

and for me it was more easy and faster to get the gold out by dissolving copper with Nitric acid
than with lead
Lead is pain to dealing with
and Lead is way toxic,those lead fumes in the cupellation process can poisoning You.