Old MLCC Capacitors
- Thread starter mmzhr
- Start date
Help Support Gold Refining Forum:
Are they magnetic? The magnetic ones will have less, to no PMs at all. They are highly variable.
You need to sort them out with different strength magnets. Those that stick even to the weak magnet are nothing but nickel electrodes. The intermediate ones can have some silver and palladium, at least in a good enough fraction of them to process.
The NON-magnetic ones will be 100% silver/palladium electrodes. Those are the ones for which an estimate can be made.
You need to sort them out with different strength magnets. Those that stick even to the weak magnet are nothing but nickel electrodes. The intermediate ones can have some silver and palladium, at least in a good enough fraction of them to process.
The NON-magnetic ones will be 100% silver/palladium electrodes. Those are the ones for which an estimate can be made.
If you happen to posses them, then I wouldn´t do much sorting at all. All in all, some magnetic ones actually do contain some silver, maybe some even some Pd. Some just have the nickel in the place where it solders to the board, not actually in the "sandwich".Hi Guys
Happy New Year 2023
I Found 1.5KG Old MLCC Capacitors
How Much Palladium and Silver Extract From These Capacitors ?
Thanks
I wouldn´t bother with it too much. When you start fractioning the stuff down - it simply means you need to do multiple runs in the future, aside of just one batch, mixed tho.
If you already have it, you need to deal with it anyway... So let´s go smelting
Nice and clean material (at least it look clean, not too much resistors inside). Put them into graphite crucible in induction furnance and smelt them with silica flux. Regular silica sand or pulver. Around 20wt.% to lower the melting point of the smelt down to less than 1300°C. Or you can add boric acid instead of silica to lower the melting point of the slag below 1100°C, but this slag is relatively viscous and you will need to either heat it more to lower viscosity, or prolong the smelting time for metal beads to better coalesce.
No collector metal is needed. Do not overshoot the ammounts of fluxes, as these are eutectics of titanate capacitor ceramics - if you add less or more, mp will go up sharply. In addition, too much boric acid not only rise the mp, but also make the slag very very thick and viscous.
I disclosed these approaches some time ago in one thread in detail:
https://goldrefiningforum.com/threads/processing-old-mlccs-disclosed-process.31895/#post-338124
Do not go wet route, as you will probably loose very significant portion of values and burn so many time working around it. Tedious, headaching process, liters of waste to process, washing the ceramic mud is utopy-like... Etc
Together with kurtak, we know very close how it end up when somebody put MLCCs in acid To the yield question - this is impossible to correctly guesstimate - if you happen to have only Pd types of old manufacture, Pd yield is somewhere in between 1-4 wt% most probably. Soviet era SMD MLCCs could go up to 7wt% PGMs.
You now can see - if your material is diluted with just nickel containing ones, your yield is gonna fall sharply
Good luck.
mmzhr
Mohammad Mazaheri
Thank You For Your Help and AdviceIf you happen to posses them, then I wouldn´t do much sorting at all. All in all, some magnetic ones actually do contain some silver, maybe some even some Pd. Some just have the nickel in the place where it solders to the board, not actually in the "sandwich".
I wouldn´t bother with it too much. When you start fractioning the stuff down - it simply means you need to do multiple runs in the future, aside of just one batch, mixed tho.
If you already have it, you need to deal with it anyway... So let´s go smelting
Nice and clean material (at least it look clean, not too much resistors inside). Put them into graphite crucible in induction furnance and smelt them with silica flux. Regular silica sand or pulver. Around 20wt.% to lower the melting point of the smelt down to less than 1300°C. Or you can add boric acid instead of silica to lower the melting point of the slag below 1100°C, but this slag is relatively viscous and you will need to either heat it more to lower viscosity, or prolong the smelting time for metal beads to better coalesce.
No collector metal is needed. Do not overshoot the ammounts of fluxes, as these are eutectics of titanate capacitor ceramics - if you add less or more, mp will go up sharply. In addition, too much boric acid not only rise the mp, but also make the slag very very thick and viscous.
I disclosed these approaches some time ago in one thread in detail:
https://goldrefiningforum.com/threads/processing-old-mlccs-disclosed-process.31895/#post-338124
Do not go wet route, as you will probably loose very significant portion of values and burn so many time working around it. Tedious, headaching process, liters of waste to process, washing the ceramic mud is utopy-like... Etc
To the yield question - this is impossible to correctly guesstimate - if you happen to have only Pd types of old manufacture, Pd yield is somewhere in between 1-4 wt% most probably. Soviet era SMD MLCCs could go up to 7wt% PGMs.
You now can see - if your material is diluted with just nickel containing ones, your yield is gonna fall sharply
Good luck.
mmzhr
Mohammad Mazaheri
How does a smelt of palladium work at such a low temperature? Pd's melting point is 1555C.If you happen to posses them, then I wouldn´t do much sorting at all. All in all, some magnetic ones actually do contain some silver, maybe some even some Pd. Some just have the nickel in the place where it solders to the board, not actually in the "sandwich".
I wouldn´t bother with it too much. When you start fractioning the stuff down - it simply means you need to do multiple runs in the future, aside of just one batch, mixed tho.
If you already have it, you need to deal with it anyway... So let´s go smelting
Nice and clean material (at least it look clean, not too much resistors inside). Put them into graphite crucible in induction furnance and smelt them with silica flux. Regular silica sand or pulver. Around 20wt.% to lower the melting point of the smelt down to less than 1300°C. Or you can add boric acid instead of silica to lower the melting point of the slag below 1100°C, but this slag is relatively viscous and you will need to either heat it more to lower viscosity, or prolong the smelting time for metal beads to better coalesce.
No collector metal is needed. Do not overshoot the ammounts of fluxes, as these are eutectics of titanate capacitor ceramics - if you add less or more, mp will go up sharply. In addition, too much boric acid not only rise the mp, but also make the slag very very thick and viscous.
I disclosed these approaches some time ago in one thread in detail:
https://goldrefiningforum.com/threads/processing-old-mlccs-disclosed-process.31895/#post-338124
Do not go wet route, as you will probably loose very significant portion of values and burn so many time working around it. Tedious, headaching process, liters of waste to process, washing the ceramic mud is utopy-like... Etc
To the yield question - this is impossible to correctly guesstimate - if you happen to have only Pd types of old manufacture, Pd yield is somewhere in between 1-4 wt% most probably. Soviet era SMD MLCCs could go up to 7wt% PGMs.
You now can see - if your material is diluted with just nickel containing ones, your yield is gonna fall sharply
Good luck.
kurtak
Well-known member
Because your not melting pure Pd &/or only Pd
The other metals involved in the smelt (silver, copper etc.) that have a lower melt point then the Pd act as (high temp) solvents on the Pd & thereby bring down the melt point of the Pd
Kurt
Palladium is just a small fraction of the metals contained. There is usually considerably more tin, lead, silver and copper - which alloy together with Pd and lower melting point
Hmm, so the button you're looking to get is actually a copper-silver-palladium alloy. If they're clean MLCCs, primarily an Ag-Pd button. Interesting. I'll have to try this when I get my kiln built in spring.
Depends on how much of legs/solder you are able to remove from them. But some of it will be still present. With slow incineration you can pretty much oxidize the solder to corresponding crumbly oxide residue, and be left with just ceramics - this is the case I like when processing THT MLCCs, not SMD. Then with slight "grinding" and stirring action legs could be easily detached from sides of ceramics, leaving practically no solder residue.
Copper and silver does not bother me that much. Worst is the tin - as it ruins possibility to nicely dissolve that alloy in nitric acid. So instead, you need to either somehow cupel/scorify the alloy from tin, or oxidatively clean it with basic NaOH flux and suitable oxidant (nitrate or oxygen).
I soaked my test batch from weakly-magnetic ones in HCl for a week. That took care of all the solder and any nickel (since the nickel was so thin, it dissolved away quite well), and left the Ag and Pd alone. After several good washes in hot distilled water, I could very nicely dissolve the Ag and Pd in nitric.
Now, I assume I could do the initial HCl dissolve, then wash and dry the MLCCs before crushing and smelting. That would leave me with a pretty pure PM button.
kurtak
Well-known member
Per the bold print - did you mean to say - "when I get my furnace built in spring" ?
Kurt
It's kindof a furnace-forge-kiln thing. It can be rearranged into different configurations by moving iron grates and bars around inside it, channeling the air flow in different ways, and so on. This one will be an upgraded design from one I built as a kid, which lasted 10 years and some 50 melts. A couple times I got it hot enough to melt copper with only wood. And it was just made mostly of cement patio blocks and regular bricks I found lying around.
If you left the MLCCs in HCL, and there is copper present, it start to dissolve Pd too. Minor, but real. Good to consider it in the future.
Be aware of liquified barium in the solution from the ceramics and always treat waste from these with any sulfate or sulfuric acid to precipitate very insoluble barium sulfate, which is practically harmless.
Good to know about the barium. Barium sulfate is used as a contrast agent for intestinal scans. :]
Major elements in the ceramics (as far as I found in old ones) are barium, titanium and bismuth in the form of their oxides, to the less extent some alumina and silica. If you once disintegrate ceramics, mostly insoluble TiO2 mud is left behind and you leach Ba and other soluble elements into the juice.
Thankfully, up to now I've only been doing very small tests. I won't be doing large batches until all my tubs of circuit boards are depopulated, and I can do everything at once.
