Is there gold in left over 486 CPU ceramic?

[NobleMetalsRecovery]

I recently found about 30 pounds of broken up ceramic CPU's. Mostly 486's. I processed them to remove the gold about 20 years ago. I used HCL and sodium nitrate. I did not break them up very much.

I have wondered for years if they is still gold to be found in these left overs? The main thing I would really like to know is if there is gold in the layers of ceramic that make up the CPU?

In regards to where the pins connect to the bottom of the CPU, what conducts the electric signal from the pins to the gold wires that attach to the silicon die.

There must be a conductive circuit, but what is that circuit made of? Does anyone reading this have first hand knowledge of this subject matter?

Regards, Steve

 

[anachronism]

The solution is pretty simple Steve.

Break them up more and try a sample in AR. If you get gold, the answer to whether you left any gold there is yes. As to whether there is gold between ceramic layers I don't believe there is as such in the form of a layer. There's a lot of speculation as to the constituent metal in the wires between the die and pins and the results I've had from reprocessing after further grinding are inconclusive. I've always had more gold, but sometimes the amounts are very small and not worth the time spent to recover it.

We recently did this with 240Kg of spent ceramics and recovered 45g of gold. Sadly there was no way to tell which ceramics produced it as it was a completely mixed sample. Was it worth it overall when taking into account time and waste? No.

I think that some do have gold wires but I think that the majority are not. Hope that helps Steve.

Jon

 

[NobleMetalsRecovery]

Thank you Jon.

Although you couldn't answer this with certainty, at least you could shed some light on the subject. It sounds like more often than not it will be worth processing them again after grinding them fine. How are you doing your grinding? Ball mill?

I would have to think that someone, somewhere was involved in the design and manufacturing of these ceramic CPU's. There must have existed somewhere the instructions for making the CPU's, and in those instructions it would have to state with certainty the materials used for the traces inside the ceramic chip.

With the size and scope of this forum, perhaps my question will eventually make it as far as needed to reach the right person.

If someone took the time to do an extensive search of older patents related to CPU manufacturing the info may be found buried somewhere within those documents.

 

[nickvc]

You are going to have a problem if you finely grind the material as it will be hard to fully remove any pregnant AR from it, I’d be tempted to do two samples of say a kilo each one crushed the other as is and compare the returns to determine if the crushing is worth the effort, I’d also do this in small quantities which ever route you take and use heat and time to fully digest any values and reuse the same solution, if you go for the larger pieces route you can rinse the pieces of in water which can be heated to reduce the volume and added back to the main solution for precipitation.

 

[NobleMetalsRecovery]

Thank you for your input. You seem to be in agreement with the other person that responded so far.
I will probably do what both of you have suggested, but I would still hope to hear from someone that actually was involved in the manufacture of ceramic CPU's in the late 80's or early 90's.

I am convinced, that somewhere in world there are written instructions for the manufacture of 486 CPU's.
My view is that which ever company was making them they had to have an established procedure that spelled out EXACTLY how to make the conductive traces in-between the layers of ceramic.

The information exists, somewhere, it's just a matter of enough people seeing this until one of them has the correct answer.

But, on the other hand, I can see your point, just reprocess them, if I get more gold, then there's was gold left behind, and it doesn't matter too much where is was hidden.

 

[philddreamer]

I like to break the ceramic CPU's as small as possible... I don't grind them, though.
That way, the acids can reach better in between the layers. And also, I separate the silicon chip by hand, with heat.
https://goldrefiningforum.com/phpBB3/viewtopic.php?f=52&t=12845&p=128183&hilit=processing+Pentium+Pro#p128293

Take care!
Phil

 

[anachronism]

I will probably do what both of you have suggested, but I would still hope to hear from someone that actually was involved in the manufacture of ceramic CPU's in the late 80's or early 90's.

I am convinced, that somewhere in world there are written instructions for the manufacture of 486 CPU's.
My view is that which ever company was making them they had to have an established procedure that spelled out EXACTLY how to make the conductive traces in-between the layers of ceramic.

The BOMs for these processors are available somewhere. That given a few have been posted on the forum over the years but I cannot recall where or how to search for them. Speaking for myself Ive not come across anyone here yet (since 2011 when I joined) who was involved directly in the manufacture however you're certainly in the right place to speak to people who have refined large quantities of them. 8) 8)

Jon

 

[g_axelsson]

The technology used to make ceramic CPU bodies are called HTCC (High Temperature Co-fired Ceramic). Wires inside the ceramic CPU body are made of screen printed tungsten, molybdenum or manganese in various combinations. It is printed on thin layers of unfired ceramics (green tape), stacked and then fired with the ceramics to form the internal traces. Next step is to cover the exposed traces with a metal, often nickel and then plated with gold. Then the die is brazed in place in the cavity. Among the last steps is wire bonding where bond wires are used to connect the chip to the internal traces. Bond wires are never used inside the ceramic body.

I'm on the firm conviction that properly leached ceramic CPU bodies does not have any amount of gold to talk about. Maybe traces of absorbed gold chloride but that would be a result of the leach process and not from the manufacturing. Gold is used sparingly and only where there is a real technical need for it.

If the IC chip is still attached to the ceramic body after leaching, then there is probably some gold braze hiding beneath it. Undissolved parts of heat spreaders and lids will also hold back some of the gold by cementing it.

There are some special applications that uses LTCC (Low Temperature Co-fired Ceramic) with internal wires of one or several different precious metals, but for a mass product as ceramic CPU bodies (386, 486, Pentium, PPro, Motorola, Sparc....), HTCC is the only technology used.

I'm willing to sell any left over ceramics from my refining, shipping plus something to cover the work to pack it basically.

https://en.wikipedia.org/wiki/Co-fired_ceramic
http://www.ecrimpower.com/products/high-temperature-co-fired-ceramic-htcc.html
http://www.ecrimpower.com/products/low-temperature-co-fired-ceramic-ltcc.html

Göran

 

[kurtak]

The technology used to make ceramic CPU bodies are called HTCC (High Temperature Co-fired Ceramic). Wires inside the ceramic CPU body are made of screen printed tungsten, molybdenum or manganese in various combinations. It is printed on thin layers of unfired ceramics (green tape), stacked and then fired with the ceramics to form the internal traces. Next step is to cover the exposed traces with a metal, often nickel and then plated with gold. Then the die is brazed in place in the cavity. Among the last steps is wire bonding where bond wires are used to connect the chip to the internal traces. Bond wires are never used inside the ceramic body.

I'm on the firm conviction that properly leached ceramic CPU bodies does not have any amount of gold to talk about. Maybe traces of absorbed gold chloride but that would be a result of the leach process and not from the manufacturing. Gold is used sparingly and only where there is a real technical need for it.

If the IC chip is still attached to the ceramic body after leaching, then there is probably some gold braze hiding beneath it. Undissolved parts of heat spreaders and lids will also hold back some of the gold by cementing it.

There are some special applications that uses LTCC (Low Temperature Co-fired Ceramic) with internal wires of one or several different precious metals, but for a mass product as ceramic CPU bodies (386, 486, Pentium, PPro, Motorola, Sparc....), HTCC is the only technology used.

I'm willing to sell any left over ceramics from my refining, shipping plus something to cover the work to pack it basically.

https://en.wikipedia.org/wiki/Co-fired_ceramic
http://www.ecrimpower.com/products/high-temperature-co-fired-ceramic-htcc.html
http://www.ecrimpower.com/products/low-temperature-co-fired-ceramic-ltcc.html

Göran

Interesting & informitive Goran --- thanks for posting

Kurt

 

[Johnny5]

Maybe traces of absorbed gold chloride but that would be a result of the leach process and not from the manufacturing.
Göran

Many years ago, I found a good way to limit this, is by soaking the ceramics in boiling water for about 10 minutes, or run them in a vacuum chamber for a few minutes, while completely submerged in water. Then process as normal.

 

[anachronism]

Maybe traces of absorbed gold chloride but that would be a result of the leach process and not from the manufacturing.
Göran

Many years ago, I found a good way to limit this, is by soaking the ceramics in boiling water for about 10 minutes, or run them in a vacuum chamber for a few minutes, while completely submerged in water. Then process as normal.

Can you give more information?

 

[snoman701]

Its loading the pores of the alumina up with water, thus making it harder for the surface to adsorb gold chloride.

I don’t really know how absorbent the surface is....I would think minimally.


Sent from my iPhone using Tapatalk

 

[aft_lizard01]

I don't know if it's been mentioned but if the die is still attached to the ceramic you will likely have gold trapped underneath them. I always desolder CPU dies and then break the cpu package up. Run the dies with the broken packages because they will contain some small measure of gold still attached to them, at least I work with the principle there is still some attached. As far as interstitial gold wires I think if any is present it's small and not worth it unless you have a massive amount to process and a cheap method.

 

[g_axelsson]

I don't know if it's been mentioned but if the die is still attached to the ceramic you will likely have gold trapped underneath them. I always desolder CPU dies and then break the cpu package up. Run the dies with the broken packages because they will contain some small measure of gold still attached to them, at least I work with the principle there is still some attached. As far as interstitial gold wires I think if any is present it's small and not worth it unless you have a massive amount to process and a cheap method.

Yeah, it's been mentioned... http://goldrefiningforum.com/phpBB3/viewtopic.php?p=297637#p294906

Some 486 dies seems to be glued to the body though... IBM did this sometimes as I've found some 486 CPU:s where the die still is attached with some soft plastic.

Here is an example, top and bottom with respective dies. Both dies were still attached to the body and on the right you can see the remnants of the gold braze. It's mostly gone but some remains.




Göran

 

[NobleMetalsRecovery]

The technology used to make ceramic CPU bodies are called HTCC (High Temperature Co-fired Ceramic). Wires inside the ceramic CPU body are made of screen printed tungsten, molybdenum or manganese in various combinations. It is printed on thin layers of unfired ceramics (green tape), stacked and then fired with the ceramics to form the internal traces. Next step is to cover the exposed traces with a metal, often nickel and then plated with gold. Then the die is brazed in place in the cavity. Among the last steps is wire bonding where bond wires are used to connect the chip to the internal traces. Bond wires are never used inside the ceramic body.

I'm on the firm conviction that properly leached ceramic CPU bodies does not have any amount of gold to talk about. Maybe traces of absorbed gold chloride but that would be a result of the leach process and not from the manufacturing. Gold is used sparingly and only where there is a real technical need for it.

If the IC chip is still attached to the ceramic body after leaching, then there is probably some gold braze hiding beneath it. Undissolved parts of heat spreaders and lids will also hold back some of the gold by cementing it.

There are some special applications that uses LTCC (Low Temperature Co-fired Ceramic) with internal wires of one or several different precious metals, but for a mass product as ceramic CPU bodies (386, 486, Pentium, PPro, Motorola, Sparc....), HTCC is the only technology used.

I'm willing to sell any left over ceramics from my refining, shipping plus something to cover the work to pack it basically.

https://en.wikipedia.org/wiki/Co-fired_ceramichttp://www.ecrimpower.com/products/high-temperature-co-fired-ceramic-htcc.htmlhttp://www.ecrimpower.com/products/low-temperature-co-fired-ceramic-ltcc.html
Göran

Forgive me for having never replied to your response. You definitively answered my question. Your info and that of the others that also responded filled in all the gaps I had on this topic. It is much appreciated.