Removing silicon die ceramic ic
- Thread starter Slochteren
- Start date
Help Support Gold Refining Forum:
I have a handful of these as well, I'm not sure if they're worthy of reprocessing just for that little amount. You could either leave them in very hot AR for longer or I just throw them in the next batch of ceramics.
Refinerjohn
Member
I usually heat the chip to release the cap. Once the cap is removed I continue with heating in the chip until that die can be picked right off with pliers or tweezers.
Thats how I remove them.
In a commercial refinery, one which see's a lot of these processors sent in as a separate scrap type, would granulate this material before processing them in aqua regia. The theory being the smaller pieces allow better penetration of the acids and digest more of the values.
Also note that the refiners processing this material in aqua regia also save the residues because they can never be totally stripped of all of their values in AR. (If properly granulated to small enough pieces you will get most of the gold but some remains.) The residues are melted in a furnace with copper and cryolite which will dissolve the ceramic. It is a slow process requiring a long heat to melt the ceramics. And it eats up crucibles as well.
Also note that the refiners processing this material in aqua regia also save the residues because they can never be totally stripped of all of their values in AR. (If properly granulated to small enough pieces you will get most of the gold but some remains.) The residues are melted in a furnace with copper and cryolite which will dissolve the ceramic. It is a slow process requiring a long heat to melt the ceramics. And it eats up crucibles as well.
Slochteren
Well-known member
I tried heat and works fine, thanks.
Slochteren
Well-known member
From the legs to the die inside the ceramic is that also gold plated? That would be a reason to granulated I guess...
If you think for one minute a commercial refiner has the chip by chip specifics about these circuits that a lot of members here have you are mistaken. Commercial refiners will not pick through the chips they simply toss tbe lot in the granulator.
Slochteren
Well-known member
I have no idea how commercial refiners do things, they will have their reasons how to process things.
But I was wondering in general how the connection runs from the legs thru the ceramic to the silicon die, I broke some but cannot see any connection wire... The normal dip ceramics are made of 2 parts "glued" together and breaking the parts you can see the connection. But for the ons that look as 1 piece I don't know.
But I was wondering in general how the connection runs from the legs thru the ceramic to the silicon die, I broke some but cannot see any connection wire... The normal dip ceramics are made of 2 parts "glued" together and breaking the parts you can see the connection. But for the ons that look as 1 piece I don't know.
But I was wondering in general how the connection runs from the legs thru the ceramic to the silicon die?
Alu or solid goldwire so thin only good eyes can see it.
Henrik
Alu or solid goldwire so thin only good eyes can see it.
Henrik
- Joined
- Mar 21, 2018
- Messages
- 140
This may be the answer.
Post in thread 'Is there gold in left over 486 CPU ceramic?' https://goldrefiningforum.com/threads/is-there-gold-in-left-over-486-cpu-ceramic.27905/post-294906
Janie
Post in thread 'Is there gold in left over 486 CPU ceramic?' https://goldrefiningforum.com/threads/is-there-gold-in-left-over-486-cpu-ceramic.27905/post-294906
Janie
I always broke the ceramic CPU chips into peices to remove the gold. I still have several of the ceramic CPU and the large Pentium chips.
- Joined
- Jun 22, 2023
- Messages
- 115
you aren't losing anything. There is no gold underneath the die - just bonding glue. The gold is in the wires that come off the top of the die, a trace from the legs and a trace from the "paddle" inside the package.
I usually break mine in half to get the stupid metal cap off (has a lot of nickel in it and makes a mess) but other than that, the die just get tossed with the trash at the end, stuck to the package or not.
I usually break mine in half to get the stupid metal cap off (has a lot of nickel in it and makes a mess) but other than that, the die just get tossed with the trash at the end, stuck to the package or not.
- Joined
- May 31, 2016
- Messages
- 3,509
Agreed. I usually make sure the die is cracked and then process in AR. The gold plated lids are pretty awesome though.
I disagree strongly!
A classical way to attach the die in the cavity of a ceramic body is to use a gold silicon braze. It's not the only way to attach the die, other ways are silver or palladium based braze or even glue (possible silver filled to make it conductive). Gold braze is expensive so naturally people have tried to come up with cheaper alternatives.
As a rule of thumb, pale yellow metal around the die is gold silicon braze.
I always run my ceramic dies with gold braze until the dies falls off or can be pushed off with a glass or plastic rod. Any dies that I missed goes back in the next batch so I'll get it eventually.
Here is a patent that discuss different braze and solder alloys. :
https://patents.google.com/patent/US4214904A/en
Here is a manufacturer of silver filled epoxy and other bonding materials. :
https://tanaka-preciousmetals.com/en/products/detail/silver-adhesive/?nav=use
And finally a pdf with examples of gold - tin alloys used for mounting dies.
https://www.indium.com/media/documents/0424/au-based-precision-die-attach-preforms-pds-99990-r2.pdf
There is a huge amount of other information sources on the net so you can spend hours just digging in on this topic.
Göran
Slochteren
Well-known member
Just revovered 0,6 gr from +/- 40 die's wich I seprrated with heat. Also worth the few minutes to remove them.
- Joined
- May 31, 2016
- Messages
- 3,509
How hard do you drive the reaction to get this to happen Goran?
The glass dyes?
Well, I'm a slow refiner so I just put it in a beaker with aqua regia and let it sit until ready. No heat, just time.
Granted, if you are running material for a customer and need the result within a day then crushing the material will speed up the process, but if you really want the last of the gold then re-running any chips that still have the die attached will give you more gold in the end.
... and I'm in for the gold, not the money.
Here is a picture from two 486 cpu:s, one with a glued die and one brazed with some gold left.
Göran
If I may ask a question. So you place the entire chip into AR?
Are not the gold pins on these chips magnetic? Contains some type of ferrous metal?
Cause I have a bunch of these ceramic CPU. And a bunch that have the gold pins. Also the CPU that has only the gold dots on them.
Are not the gold pins on these chips magnetic? Contains some type of ferrous metal?
Cause I have a bunch of these ceramic CPU. And a bunch that have the gold pins. Also the CPU that has only the gold dots on them.
Yes, I'll place the chip whole in a beaker and add aqua regia to cover the lot. If the acid runs out of dissolving power (which it will several times) the gold cements back on the remaining metal. I then pour off, test and discard the liquid, replacing it with new acid. This way I get rid of most of the base metals and the final batch contains all the dissolved gold.
Sometimes I also add more chips to the beaker as earlier ones are fully dissolved and the dies falls off.
This can take several months, mostly because I just put the beaker aside and forget about it for a week or two. I've been very busy with other stuff lately and it's been a year or two since I last ran any ceramic chips.
Last year I ran large batches of BGA chips as a toll refining and that took up most of my refining time. But I recently came over scrap with some ceramic chips and a lot of gold so it's about time to do a batch again...
Göran
Sometimes I also add more chips to the beaker as earlier ones are fully dissolved and the dies falls off.
This can take several months, mostly because I just put the beaker aside and forget about it for a week or two. I've been very busy with other stuff lately and it's been a year or two since I last ran any ceramic chips.
Last year I ran large batches of BGA chips as a toll refining and that took up most of my refining time. But I recently came over scrap with some ceramic chips and a lot of gold so it's about time to do a batch again...
Göran
