particle size of "crushed" ceramics in AR-leaching

[frank-20011]

hello,

often ceramic cpu's which should leached by the AR-process are crushed with a hammer to a few big parts only.

is it maybe better to crush them in a ball mill to a finer "powder"... letzt say 2 or 1mm in particle size or is finer crushing unnecessary?

when i've opend a lid of an older ceramic cpu i see a lot of bonding wires, some made of Al, some made of Au but all of them are attached to the silicone die on one end and to an electrical conductor made of gold on it's other end.
these golden areas are arranged in 2...3...4 horizontal layers and a lot of them in every layer. between these layers there is the ceramic "substrate"

my thoughts are the AR can't reach every golden conducter if the cpu is only crushed in very large chunks but much better if the cpu's are powdered?!

what du you think abaout that?

best regards!

 

[resabed01]

I think you want to stay away from anything that will turn the ceramic into powder. It will make leaching and filtering a nightmare.

 

[frank-20011]

hello,

not really a powder but rather smaller particles than let's say 5mm-2cm (like these from cracking the cpu's with a hammer only).

regards!

 

[solar_plasma]

You can be quite sure the first time, let's say a year or more, in your learning progress there is not any thought, that others haven't been thinking before. There are lots of discussion about how to treat the different ceramics. For sure common 486-686-like consumer CPUs do not have any gold within the ceramics.

Since this is said not to be true for all kinds of CPUs, you could easily make your own test, hammering a piece to powder and test it.


A sintered tungsten top is a problem of its own though - some gold might cement under the leaching process within the porous heat spreader and can only be recovered then by milling the heat spreader to powder, as far as I read in other threads. I can't find it by a simple search, but someone was able to show that this is true and a reason for smaller losses.

 

[Wingedcloud]

A teacher of mine often said to us students:

"Don't make easy things complicated".

I think this applies to your situation

As far as my experience goes, a simple hammering is enough to remove the heat spreader if you want to save it for any other occasion and to expose the gold in the very center. Besides, milling to a powder is only going to make you lose some values, which I suppose you don't want.

Winged

 

[frank-20011]

hello,

and thanks...i will belive you ;-)

and about the heat spreaders, the little lid (sometimes made of ceramic, sometimes of metal which is gold plated) above the silicon piece with the bonding wires: i remove them with a torch.
the result is
- less base metals (ceramic lid goes in the litter box, gold plated lids in sulfuric cell)
- less soft solder and
- bonding wires are ready for AR without hammer-crushing

thanks and regards, frank!

 

[patnor1011]

Gold plated lid from ceramic CPU is not good material for sulfuric cell.
Majority of gold is in braze so it better be dissolved whole.

 

[frank-20011]

hello,

o.k. but only in cpu's which have hard soldered lids of gold plated material. (or is there gold present in soft solder too)

in conclusion of that i will desolder only the soft soldered ceramic lids and discard them.

main intention of desoldering al types of soft soldered lids was to avaoid as much base matals as possible.


about the next step: time is not a critical thing and i hate hot acids. i'll let stand a beaker outdoors with a lid on it and Hcl/tapwater half and half (so HCl should have something arround 10 percent or should i use less water? the crude HCl come with 21%), put the ceramic cpu's, other ceramic eproms with gold braze and a little "bonding wire concentrate" in it and add 5ml HNO3 (36%) at first and repeat the last step as often as required.
do you think the process is finished without any extra heating (it's summer now) in one week or is heating advisable?

best regards!