Gold inside chips (black, flatpacks - not CPU)

[patnor1011]

I have finally started a small test run with RAM chips: Just 22 single and double sided SDRAM and DDR sticks carrying 130g of chips.

Using the coffee tin method with a few holes punched above the bottom and placed into the coals of a wood fire, only about half of them burned to white. I am now considering to make one of those soup-can furnaces. [Edit: I may mean forge. I'm not sure.]

This is not a question, but I'm not sure I entirely understand bond wires. Google / wiki don't say much. Inside my RAM it looks like the thick leg connecting wires, made of "whatever" spider into the centre and than have a short (1 - 1.5mm long) golden tip attached to them. They seem fairly firmly attached. But the main point is that they're quite thick and visible, whereas I was expecting something I could barely see.

That tip you see is where bonding wire is attached-soldered to pin.

With the large black bits picked out for reprocessing, I've had a go at panning anyway. I own one of those turbopans, with the spiralling ridges. They're great for processing large amounts of gravel quickly, but I'm not sure it's the best for small amounts with dust; with so many ridges in which to collect, a regular pan might be better. I am also not about to win any prizes for World's Greatest Panner*. Nevertheless I am pleased to start to see a little colour. But this is just fine dust, much finer than what I thought I could see on the legs.

It is dust or more wires it all depend on type of IC some do have bit longer wires some shorter. It all depend also on method of crushing of incinerated material. The more crushing the finer the gold wires - dust. If you incinerate them thoroughly to white dust more wires tend stay not broken. If you magnify that dust you will see it is actually wires. So small and thin they appear as dust. Also with 100+grams of IC you will see about 0.1 g of gold which is not much of dust to be seen. But you found it, you can see it and you now know what to look for.

Am I right in understanding that the silicon wafers with the shiny yellow appearance (one placed into shot) contain NO gold at all?
* I googled that term and found a pile of recipes for Indian cuisine.

Well they might. Mainly on sides where wires were attached to them. If you collect enough of them just give them bath in warm AR to be sure there is no more of gold on them. I do keep some with wires still attached, one day when I will get some small microscope I will do some cool pictures.

 

[jason_recliner]

Thanks Pat. As you say, if one can gather about a gazillion of them they'd be worth a soak. Otherwise... all that glitters is not gold.

 

[patnor1011]

You will be surprised how fast they add up. I just checked and I do have close to 2 kilograms of smallest one, those with 6-8 legs (3-4 on each side).
Just toss them to a bucket and in no time you will have it full. Then you can process them.

 

[g_axelsson]

Hi Jason, there is a lot of black stuff in the pan. Did you incinerate until the outside was white or were the chips white all the way through?

I always run the chips until the carbon is lost, any chips with black residues goes back again. With properly incinerated chips you could crush the chips between your fingers.

That tip you see is where bonding wire is attached-soldered to pin.

Actually, the wires are welded with heat, pressure and ultrasonic sound.
https://en.wikipedia.org/wiki/Wire_bonding
Göran

 

[jason_recliner]

Yes mate. I know it's still very black. I just wanted to try while I had the Easter break. First thing I did was hold a single chip in the gas stove with pliers so I would have a target. It disintegrated like a Hollywood marriage.

So I filled a washing machine tub brazier with fire (wood, actually) and once the flame died down a bit, added the can to the coals. Some pretty flaming colours came from the chips, I stirred a little, and left it a couple of hours. About a third of them burned out to white - as soon as the flame was all gone it had lost adequate heat. Blowing air through a tube into/under the can didn't help much either.

I was going to try a little "chip-apartheid", but it was messy. I need something just a little hotter, or with better airflow, and will ultimately reprocess the lot.

 

[patnor1011]

I learned that less is sometimes more. That mean I try to incinerate only one layer of IC, they are white in no time. If you put more layers IC in the middle or on bottom will never get to that fluffy white stage when they are crumbling while you try to take them out.

 

[Smack]

Been working on chips the last few days, trying to get everything that needs incinerated done before it gets real warm out.
Not going to go over everything I've done as it's all been covered here, I just might do things a little different. If someone (not new) really want's to know the steps I took, ask and I will lay it out for you. The SST bucket is 7 gallon and has 43 lbs. of powder ready for panning. The chips are all mixed black IC type off memory, mobo's and so on, gold is gold. Some of the weights are on the post "Show Your Work Area" for individual chip type but I added quite a bit to those amounts. I will say this about the first picture; The chips are pre-cooked in my friends wood burner prior to incineration to get rid of most of the smoke.

 

[mls26cwru]

Am I right in understanding that the silicon wafers with the shiny yellow appearance (one placed into shot) contain NO gold at all?

This is a wrong assumption...There will be some gold wire still attached to the chips even if they are ground up. The braze they use to attach the wire is mostly left on the Si chip along with some trace amounts of gold wire. it is most deffinitly worth processing this. I just processed a good amount of these Silcon chips and pulled out a fair amount of silver and more than enough gold to make it worth while.

 

[kurtak]

Hi Jason, there is a lot of black stuff in the pan. Did you incinerate until the outside was white or were the chips white all the way through?

I always run the chips until the carbon is lost, any chips with black residues goes back again. With properly incinerated chips you could crush the chips between your fingers.

Göran

You "do not" need to incinerate chips to complete ash in the first part of processing them - it is "only" important that the carbon gets reduced to white ash at the very end just before they go to being leached :!:

Here is a (copy/paste) post I posted on Ken's forum - it was a reply to Geo posting a link to this video

My post from Ken's forum ------------------

That was a batch of my chips :mrgreen: please take note of how black the ash is - that is because that bucket of ash is actually about 75 - 80 percent carbon --- now you might be thinking :shock: OMG he didn't do complete incineration to his chips - & you are right.

complete incineration of chips - at this point in the processes "is not important" !!!

There are only two things that are important at this stage in the game.

(1) is that they are at least "completely" carbonized - meaning that ALL the volatiles (epoxy resins) are burnt/roasted out so that they are nothing but carbon to the core (with a surface layer that has gone white ash).

As long as they are complete carbon to the core they will crush/mill as fine as you wish to mill (80 mesh & finer).

(2) crush/mill to "at least" minus 80 mesh --- that means 80 mesh is the largest that passes through the screen - but - 85 - 90 % is going to be MUCH smaller.

Anything that does not go through the 80 mesh screen gets re-milled & re-sifted till all carbon has passed though the screen leaving only ground silicon & base metals (from legs).

The fine carbon is "plenty lite" & will wash off just fine when working on getting it down to your concentrates for processing.

Once you have washed it down to your concentrate - that is when you want to do a final incineration to turn the carbon still in the concentrate to complete ash which you can then do a final wash on before processing.

The only real concern for carbon is in the leaching process because the carbon will absorb & hold metal ions - meaning you will lose gold to this absorption if there is carbon in the leaching process.

Other than that carbon is not a problem during the washing/concentrating part of the process !!!

I do about 200 - 300 pounds of "mixed chips" per year (in 30 - 50 pound batches) & have no problem doing it this way & get recovery of .5 grams/lb (if there is a lot of proms in the mix) to as high .648 grams/lb (not a lot of proms).

When I say mixed chips that is RAM, quads & proms - no BGAs - BGAs are always processed alone as BGAs.

Kurt

 

[patnor1011]

I just finished processing 5,1 kg BGA s/n bridge for one friend. That 5.1 kg BGA (very sloppy harvesting with big blobs of solder and sometimes even screws and capacitors lodged in that solder) yield was 26.25g (5.147g/kg).

 

[Juan Manuel Arcos Frank]

Well, here I am trying to process ICs and BGA.

Here is a pic of some ICs I have found. Do they worth?. How much gold can I recover from 20 kg?

Kindest regards.

Manuel

 

[g_axelsson]

Well, here I am trying to process ICs and BGA.

Here is a pic of some ICs I have found. Do they worth?. How much gold can I recover from 20 kg?

Kindest regards.

Manuel

The first three are ICs, more specific DIL capsules (Dual In Line pin) and the yellow one is a resistor array. Don't know about the exact content of the resistor but at least there are no gold in bond wires there.

For the ICs I would expect 0.6-1.2 g/kg.
https://goldrefiningforum.com/threads/pyrolysis-reactor.22581/#p237261
Göran

 

[patnor1011]

Manuel, I just break that IC in 3 equal parts. I do process only middle part where bonding wires are.

 

[Juan Manuel Arcos Frank]

Thanks, Göran!!!!. I do appreciate your info.
Kindest regards
Manuel

 

[Juan Manuel Arcos Frank]

Ok, Pat. I am starting breaking them. Here are some pics from cicuit boards they come from.
Thanks a lot!!!!
Have a nice day
Manuel

 

[patnor1011]

Here is what I meant. I break them only because I do have time to do it and because I am saving on fuel and time when incinerating, milling and sieving material. That center piece is where gold is so I do not bother with the rest. But if you do not pay too much for incineration or do not have time to cut them just go ahead and incinerate whole IC.

On one picture there is one IC with center broken and silicone die is exposed. I marked it with red circle. That is where goodies are.

 

[Juan Manuel Arcos Frank]

Well, I am ready for ICs and black packs pyrolysis but I have found some ceramic EPROMS, here are the pics. What to do with them?

Thanks to all

Kindest regards.

Manuel

 

[patnor1011]

Some of them will have aluminium bonding wires some will have gold. You can see it through that small window. Some people just smash them with hammer in small pieces. It is possible to put them in fire and when heated they tend to easy split in 2 part along the middle where they are stuck together. I do not have any yield data as I always sold them on ebay as a part of lot of items I did not wanted to process.

 

[patnor1011]

There was post recently somewhere which caught my attention however I cant locate it and I decided to post my findings here. It will perhaps reach more people when posted here. It was post where Dan was looking for some answers about potential yield from RAM and there was link to some seller selling some sort of guide for buying RAM. That came when I was half way done with my RAM stash, I went through 8109 grams/591pcs of mixed RAM taking as much notes as possible - funnily enough I was thinking of creating guide with the same purpose.
I decided to share some of my data with you so here it goes:

I sorted RAM in 4 categories. (Idealy I wanted to divide each one to two more DDR2 and DDR3 but decided not to due to time constraints, I just needed to have this done as soon as possible) All of them were gold fingered RAM. I will add some pictures later on.

1. Long RAM stick with regular legged IC
Average weight of one RAM stick: 17.82 gram
Average weight for 1 RAM stick of:
MLCC: 0.35 gram
IC: 4.31 gram
Fingers: 1.23 gram

2. Long RAM stick with BGA IC (servers)
Average weight of one RAM stick: 18.47 gram
Average weight for 1 RAM stick of:
MLCC: 0.40 gram
IC: 5.52 gram
Fingers: 1.32 gram

3. Short laptop RAM stick with regular legged IC
Average weight of one RAM stick: 9.48 gram
Average weight for 1 RAM stick of:
MLCC: 0.12 gram
IC: 3.70 gram
Fingers: 0.55 gram

4. Short laptop RAM stick with BGA IC
Average weight of one RAM stick: 7.67 gram
Average weight for 1 RAM stick of:
MLCC: 0.19 gram
IC: 1.79 gram
Fingers: 0.62 gram


Average weight of paper stickers on RAM stick is 0,1gram per stick. While it looks like small number it should be taken in consideration when buying large amounts of RAM - you do not want to pay for paper (63.5g in just this lot) :mrgreen:

 

[FrugalRefiner]

The moderators would like to thank all of the members who contributed to the original thread that this post was based on, as well as those who asked questions showing what was missing. Because of the extra length and interest in this thread we have created the above, consolidated version making for an easier read. We encourage all members to read, comment, and ask questions in the original thread, Gold inside chips (black, flatpacks - not CPU).

The moderators have edited some posts in this thread that made reference to posts that were deleted. We strive to make these Library threads as concise as possible by deleting posts that do not add information to the original topic. We were particulary ruthless in editing this thread. We removed many posts that dealt with related subjects, such as how to remove ICs from circuit boards, alternate processing methods, IC nomenclature, electrolytic capacitors, rare earth elements, the composition of silicon dies, solder composition of BGA packages, etc. Much of it is interesting and informative, but not on the specific topic of this process. All of the original posts are still preserved in the thread listed above.

The Library threads should not be considered to constitute a complete education. Instead, they're more like reading a single book on the subject of recovery and refining. There is so much more information on the forum, and it is impossible to include it all in these condensed threads. Members are strongly encouraged to read the rest of the forum to round out their education.