My yield for ceramic IC's

My first lot of 17 kg of old ceramic chips. Started with 17 kg of chips, all IC's are with gold under the silicon chip. Process description:
Step 1: al IC's are submerged in hno3. After one month hno3 is replaced with fresh acid. Acid will disolve legs and silicon layer between two halfs of ceramic. This process will last for 2-3 months.

Step 2: after 3 months in hno3 all chips are disolved and splited on two halfs. Wash it with tap water until water becomes clean. Dry leftovers and use strong magnet to clean all magnetic base metals wich are not disolved by nitric acid. Hand pick up halfs with gold base and throw away the rest.

Step 3: with mortar brake the ceramic halfs. Broken ceramic is submerged in HCl. Add nitric in pinches. Every day a pinch off nitric until all gold is disolved. Sifon AR and filter it. Wash the rest of ceramic with water and filter it in AR solution. Now drop gold with SMB, wash it and melt it.

My yield for 17 kg of ceramic IC 's with gold base is 24,75g of :G per 17 kg of start material.

From your description I guess you have processed EPROM:s with windows.

3 months, that's quite some time. I have also run some EPROM:s, but I cracked them in two along the legs, then just ripping off the legs with a pliers as the bond wires were all aluminum in my case. The glass frit holding the two ceramic halves dissolved and created a purple mud in just a few days. The increased surface let the nitric acid attack the frit so much faster and I didn't need to dissolve any legs.

To crack the EPROM:s into halves I used a hydraulic press with chisels (as I have one), but I think it can be done just as easily with a hammer and chisel.

I don't have any data on yield as I ran the EPROM:s with other ceramic IC:s.

Edit : Here is another thread mentioning the frit. GSP used heat to split the halves.

Göran

This was my way. If you have big lot like me thousands of chips it is time consuming but les job to split all by hands. I didn't process eproms, those was regular ceramic chips from hundreds of same boards. I smashed those to see core of chip. bonding wires where of aluminum.

This is for ceramic IC's without gold bonding wires?

Is it just me or is this a really high return? 0.6 g / lb or 1.4 g / kg.

snoman701 said:

This is for ceramic IC's without gold bonding wires?

Is it just me or is this a really high return? 0.6 g / lb or 1.4 g / kg.

Click to expand...

No, sometimes the braze is quite thick and you get a great return, even with aluminium bond wires -
https://www.youtube.com/watch?v=TL20U6HflQo&t=169s

Braze was thick. Selected chips with gold base only.