Hi, I hope all are well. I have been given a job to process these chips, there is 1kg of them and it looks like they have already been stripped. I have included a photo of a small sample, all the chips are the same as ones in the photo. To me it doesnt look like the gold left would be worth the time or expense to recover. Does anybody have an opinion on this?
Would these chips be worth trying to recover the gold from?
- Thread starter jord
- Start date
Help Support Gold Refining Forum:
I think you are right.
If there is any left at all it may be out of reach, so you can of course test one or two, to see if there is anything left, but it may cost you more than you can extract.
I agree with you as well. Are you a hobby refiner or do you do toll refining? When I refined commercially it was amazing how much absolute garbage people would bring in for me to evaluate and refine. The thing is if it has no value, or not enough value to cover your costs, you lose. Not the guy who asked you to do it.
I learned the hard way to have a minimum refining charge and if you were a customer I did not know, the minimum was paid up front.
That solved the problem for me. Unless he or she is a friend and you are doing a friend a favor, don't put much effort or chemistry into it.
I learned the hard way to have a minimum refining charge and if you were a customer I did not know, the minimum was paid up front.
That solved the problem for me. Unless he or she is a friend and you are doing a friend a favor, don't put much effort or chemistry into it.
Thank you, I will run a small test batch to confirm, but i think we already know the answer.
I
I am between both at the moment, i do batches for customers but not on a huge Commercial scale, there are 500 kg of these in total, the 1 KG was a starter batch. Yes I wanted to keep acid use to a minimum on this one so will try seperate mechanically and maybe run a small test batch incinerating just for the practice.
These appear to be the black plastic tops of what some folks call "gold cornered BGA chips", with the fiberglass PCB base removed. I'm not sure the best way to tell if they've been stripped already or not. One clue would be to see if any copper that is present/visible is corroded and green. This would indicate that someone used some kind of acid which is able to dissolve copper. But, if they aren't covered in salt deposits, you might can try to look more closely at the part which should show the ends of the bond wires. I don't know how common it is to strip BGA tops before trying to resell them, but I would bet it's possible somehow.
You can see the square outline of the silicon die in the middle. This is the backside of the silicon die, the side which has no circuitry on it. This part would have been soldered to the fiberglass PCB base, to transfer heat through the fiberglass layer and into the main board.
Outside of that are little tiny pads arranged in rows of curved arcs. These pads were on the fiberglass base, and bonding wires are used to connect those pads to connections on the topside of the silicon die (the side which is embedded in plastic). If you look closely at those pads with some magnification, you may be able to see the end of the bond wire, and see if it's golden colored or aluminum colored. From what I understand, the majority of the gold in these packages is accounted for in the bond wires, which are embedded in that black plastic. But only if the bond wires are actually gold and not aluminum.
Outside of that are the remains of the traces pulled up off of the fiberglass base. These traces are only within the fiberglass PCB base, and connect those tiny little pads on the top of the fiberglass PCB base to a grid of solderballs on the bottom of the fiberglass base. Those solderballs are what connect the chip package to the board.
You can see the square outline of the silicon die in the middle. This is the backside of the silicon die, the side which has no circuitry on it. This part would have been soldered to the fiberglass PCB base, to transfer heat through the fiberglass layer and into the main board.
Outside of that are little tiny pads arranged in rows of curved arcs. These pads were on the fiberglass base, and bonding wires are used to connect those pads to connections on the topside of the silicon die (the side which is embedded in plastic). If you look closely at those pads with some magnification, you may be able to see the end of the bond wire, and see if it's golden colored or aluminum colored. From what I understand, the majority of the gold in these packages is accounted for in the bond wires, which are embedded in that black plastic. But only if the bond wires are actually gold and not aluminum.
Outside of that are the remains of the traces pulled up off of the fiberglass base. These traces are only within the fiberglass PCB base, and connect those tiny little pads on the top of the fiberglass PCB base to a grid of solderballs on the bottom of the fiberglass base. Those solderballs are what connect the chip package to the board.
Thank you for taking the time to write that very detailed explanation. I believe you are correct in what you are saying, from what i understand if I can see the bonded wires it might be worth doing as that's where most of the gold is accounted for in these packages?
Martijn
Well-known member
Be sure to rinse them well to pH neutral wash water before you heat treat them in any way.
With acids you will have enough nasty stuff in the air, but uf it was cyanide leached, it can go really bad.
Maybe test for cyanide somehow?
Anyway, be carefull. You don't know what's been done to it.
Martijn.
I will take the precautions beforehand like mentioned, I have proper chemical fumehoods/extractors fitted with correct filters to deal with most and always use PPE. For the cyanide do you think it would be possible to soak them in water using the method you stated and then maybe test the water after for cyanide?
I have something like this to test for cyanide
Attachments
Martijn
Well-known member
Someone else should advise on that. I have no experience with cyanide.
Martijn
Well-known member
And for the rinsing part, if negative for cyanide, test a bit of the wash water with soda or lye to see if there are metal salts left. Metal hydroxides will form. Blue for copper. Any color means more rinsing.
This is my main uncertainty, and I don't have anywhere near the experience of the rest of the responses here to tell you how to rule that out. I am a hobbyist; all of my recoveries are a financial loss. I like Martijn's recommendations, and your followup, to test for pH and for cyanide. The more you can find out about your material, the more you can anticipate and keep from making costly (both safety and money) discoveries in the middle of a process.
I did not anticipate encountering anything leached with cyanide and so I do not have a very good understanding of its process, but I do recall that keeping the pH at a "correct" level was very important for cyanide leaches not only for it to work right, but also for safely using cyanide. Basically, it sounded finicky enough that I didn't even give it a second thought, and just filed it under "don't use this method".
Signs of acid treatment
I know that if I don't separate out stuff and just put test items in acid, it makes a mess and is pretty easy to tell that it's been in acid and not washed well. There's colorful metal salts on everything, and it soaks into fiberglass too and takes quite a few rinses to get rid of. Even when the water runs clear, my crappy pH test strips still show slightly acidic compared to water.
Physically, any exposed copper has signs of being dissolved (sharp corners get rounded, a rougher texture appears). From your pics, I'm not seeing anything that jumps out at me to say these were soaked in HCL for a long time; the copper does not look like it was attacked much, if at all. There is one of the samples which shows no signs of any copper at all, but sometimes you get lucky and those BGAs just peel that way.
Signs of cyanide leach
But cyanide leach is more selective, and maybe could leach gold without leaving a noticeable mark on the exposed copper. Maybe someone else more familiar with the cyanide leach process can chime in on some telltale signs that something like IC chips has already been run through a cyanide leach.
Ideally, you would do these chemical tests to make sure that your samples are clean, because to magnify you'll need to hold those samples very close to your face, and it would be nice to know that you don't risk breathing in some powdered metal salts or cyanide while you are looking through the magnifier. Even if you can see that there IS gold to be recovered, you still will want to know if someone tried to do something shady before these made it to you, because then you have even less of an idea of whether or not it's economically viable to recover gold from these.
I think for creating a solution to test, you should try to put the chips in distilled water, as little water as possible to cover the sample, and let it sit for a long time with some agitation from time to time. Then, you can do some pH tests, and cyanide tests. (read the instructions, looks like the cyanide test may require the solution to be slightly acidic)
Once you have found your chips are clean and do have gold bonding wires, it's still worth to refine a sample if you are able to, before committing to the lot. At that point, you will be refining materials very similar to typical plastic IC chips, but your materials will have much less base metals (especially if you get rid of those round copper chunks on some of the chips)
Looking at these in person it looks more like they have been split tops from bottoms manually, but as saftey is a big concern the precautionary test will be carried out so there is no guessing involved. I think going forward we will process some the 1 KG thats if the tests come back negative, see the results and then make a decision on wether to go forward with the rest. Either way if the recovery is not viable it never hurts to brush up on things, the advice given is massively appreciated and a great reason why people should use this forum.
I belive there is a lot of gold , look after our member "patnor" and "bga".
Henrik
Henrik
patnor1011
Well-known member
These are top parts from BGA IC. I do not know why would they be contaminated with cyanide and even if that would be the case then incineration with proper precaution will take care of it. They need to be incinerated first then proceed as usual when processing IC chips. Bonding wires are in substrate hence they cant be visible untill incinerated and milled. All wires and traces you do see are mostly copper and should disappear when you wash concentrate with nitric prior to dissolving bonding wires.
Some BGA IC do not contain gold or very little. I once ran batch of BGA from TV boards where there was next to no gold recovered.
Some BGA IC do not contain gold or very little. I once ran batch of BGA from TV boards where there was next to no gold recovered.
Hi, can you do a clear close up photo from those flat top BGA`s. I fear those have already been leached from gold, and what you have are just the carbon epoxy shells with very little to no gold at all.
Pete.
kurtak
Well-known member
Guys - these chips are the "black epoxy" part of north/south bridge BGA chips (the green fiber part has been removed) the black epoxy part is where the gold is as that is where the gold bond wires are - that part (the black epoxy) runs about 4 grams per pound - so this IS the part you want to process - incinerate & mill to liberate the bond wires for processing just as you would any other chips
The green fiber part of N/S bridge BGA chips (which has been removed from the black epoxy in the pic) has some gold but VERY little (so in the pic he has the GOOD part) the green fiber part has a "little bit" of gold plating on the copper traces so takes MANY pounds of the green fiber part to get a gram of gold
If you take a pound of the whole N/S bridge BGAs & separate the green fiber from the black epoxy you will get right at a half pound of green fiber & a half pound of black epoxy - with the black epoxy being where the real gold value is
So what you have there is (about) 9 grams in gold value
This black epoxy part of N/S bridge chips is the second highest paying chip you can process (at 4 grams per pound) - the highest paying chips are BGA chips off of RAM which pay (about) 5 grams per pound
Kurt
kurtak
Well-known member
There is NO WAY that the gold has been "striped" from these chips yet - the gold bond wires are "incased" in the black epoxy (so acid CAN NOT get to them) - to recover (strip) the gold from them they NEED to be incinerated (turn epoxy to ash) in order to "liberate" the bond wires for processing
Kurt
