silicon wafer scrap

[usernameguy]

OK. I've obtained a power supply for the weekend, and was told I need to run the process at 2mA/cm2. I guess I'll see what I can come up with. I'll reread what you wrote earlier, but basically, you think I can just plate using a couple pieces of stainless for the electrodes, ehh? I guess it's decided.

Thanks so much

 

[usernameguy]

usernameguy,

I'm thinking. Do you have access to a small rectifier? If so, the simplest and safest way to get the gold may be to plate it out onto a stainless sheet. If you tape the edges and back with plastic tape, you should be able to peel the gold off the stainless and then simply melt it. The gold is pure to start with and will stay that way unless you contaminate it. You won't get 100% of the gold but you should get at least 90-95%. This should work no matter what the bath is composed of. Since the bath is almost new, it should plate great. If you want to try this, let me know and I'll give you more details. Also, on the operating instructions, it should give the operating temperature and the current density (amps per sq.ft.). This info will help if you want to plate the gold out.

Prepare for a series of dumb newbie questions. 1.)Is a rectifier just the power supply, or are you referring to the entire set up (electrodes, glass tank, hotplate, etc.?)? How improtant is the operating temp? I found the current density to be 2mA/cm2. Lastly, are there other metal sheets I can use for the electrodes? Does distance between them have an affect? Oh yes, I can grab some capton tape used in vac systems, but does the type of tape I use matter while I'm asking?

Thanks again. You've been extremely helpful to me. Apparently that way for a while from the other post of yours I just brought up from 07. Hopefully I can be useful on here sometime.

 

[goldsilverpro]

usernameguy,

That's .013 A/sq.in. or about 2 A/sq.ft. These current densities are based on the area of the cathode that is being plated. At 100% efficiency, about 7 grams of gold will plate per amp per hour. Towards the end, the efficiency will probably decrease and you may want to then decrease the current to about half. I would make the anode and cathode about the same size or make the anode a little larger. Stainless is not very conductive, so don't use material that is too thin. I would make the electrode sheets at least 1/16" thick. Use non-magnetic stainless if possible. For the size of your bath, 6" x 6" (.5 sq.ft) electrodes should work fine

The most important thing is the cathode. You want the plating to barely stick enough to stay on the stainless but not stick so tight that you can't peel the gold off. This is done by preparing the cathode. Ideally, soak the clean cathode in nitric acid for a couple of hours and rinse well before using it. This will passivate the stainless and coat it with an oxide layer. If you can't do this, don't over clean the stainless. Use no abrasives that might remove the oxide layer already on the stainless. Just let it soak in a weak dishwashing soap solution for a minute or so, rinse well, and let it completely air dry. I would then wrap plastic tape around all edges and cover the back of the cathode with tape. This will prevent the gold plating from plating in those areas and make the gold hard to peel off. In other words, seal off everything except a square on the face of the cathode. The best tape is 3M plater's tape but you probably don't have that. About the closest thing I can think of is plastic electrician tape.

Ideally, operate the bath at the recommended temperature. If you can't, it will probably still work. The plating might not look as good, but who cares.

I would hang the electrodes facing each other at least 3" or 4" apart. Hang them so the top 1" is out of the solution and that's where you can use alligator clips to clip on from the rectifier. Keep the clips out of the solution. A very clean 1 or 2 gallon bucket, depending on the amount of solution, should work fine. A couple of rods laid across the top of the bucket can be used to secure the wires. Don't allow the anode and cathode to short together or you'll blow the fuse on the rectifier.

Very carefully stir the solution every 15 minutes, or so, and keep an eye on things. You can't just set it and forget it.

This is all much simpler than I've described it and I have assumed that you haven't plated before. Just use common sense

Good luck

 

[goldsilverpro]

I was writing my last post before I saw your last post.

Prepare for a series of dumb newbie questions. 1.)Is a rectifier just the power supply, or are you referring to the entire set up (electrodes, glass tank, hotplate, etc.?)? How improtant is the operating temp? I found the current density to be 2mA/cm2. Lastly, are there other metal sheets I can use for the electrodes? Does distance between them have an affect? Oh yes, I can grab some capton tape used in vac systems, but does the type of tape I use matter while I'm asking?

1) The rectifier is the power supply. I'm almost positive you'll be able to plate at room temp but I'm not 100% sure. Try it at room temp and see. If it stays on the cathode, it's OK, no matter what it looks like.
2) stainless is about the only metal that you'll be able to peel off the gold
3) Any tape that won't come loose in the solution

 

[usernameguy]

I couldn't get the rectifier this weekend, so I'm gonna try again this coming weekend. The stainless I got a hold of is the right thicnkess you mentioned, unfortunately, it's the magnetic variety. Do you think this'll pose a significant problem? I'm hoping to not threat it with nitric if it's avoidable. It sounds like you think the native oxide may be sufficient anyway. The concern would just be it may be hard to remove, correct? I could always chemically strip the Au also if it turns to be a problem.

 

[usernameguy]

I'm hoping someone can let me know (preferably before this weekend whether it matters if I use magnetic versus non-magnetic stainless for this plating process GSP has outlined above. I hate to pester him. I'm very grateful for all the information he's provided thus far. VERY generous. I guess it is kind of a labor of love, huh? I kind of get the same "gold fever" from this as from prospecting. With my situation, it's also very similar. Depending on use, I never know hoe much gold I"ll drop from solution, kind of like seeing what you have at the bottom of that pan after you've been working on concentrating it for potentially hours. Never enough for retirement(usually not even a days wages), really just fun to do. successfully. Anyway, about the stainless, anyone know? As I said if I don't hear back, I'll give it a whirl and just etch off if I can't peel. I'd just like to get as much recovery as possible.

Thanks!

 

[goldsilverpro]

I think the magnetic stainless will be OK. The main thing is to not damage the oxide coating. The oxide coating is what makes the stainless, "stainless." If you do remove it, the gold will stick tight to those areas. Abrasion or acids will remove the oxides. You don't want it to stick so tight that it won't peel off. The tape is important. Not only will it prevent the gold from wrapping around the edge or plating on the back, it gives you an place to start the peeling. If you can flex the stainless, it might help pop the gold loose and make the peeling easier. If it does stick, you have nothing to dissolve the gold in without dissolving the stainless.

Good luck!

 

[yvonbug]

I've got a bunch of these silicon wafers that are already cut up for IC's. I got 'em out of a dumpster back in the 80's when dumpster diveing was still OK. I've never processed them because I thought that silicon was going to be a bitch to deal with. I'm not sure, still, how to deal with them.

 

[Harold_V]

To my knowledge, silicon is a non-issue. It isn't a problem with acids, so the rule about not using AR for extraction can well be ignored, assuming there is not a large amount of base metal included. If so, treat the base metal with the appropriate process to eliminate all that can be eliminated, then dissolve the values with AR. Silicon should be easily eliminated in filtration.

Bottom line, near as I can tell, is the real problem is not silicon, but base metals. I stand to be corrected, and welcome comments from those that understand that which I may not.

Harold

 

[Smitty]

Yvonbug, finished wafers are comprised of many material, but there is no gold inside the actual production of the wafer. If there is, i've never heard of it. The only gold I know about is when the wafer is ready for mounting, certain chips need gold backing or on bonding pads for the fine leads to connect to. From the picture I can't tell if they have the gold film on the back or if they are just golden in color? Semiconductor fabs are typically using copper and aluminum for there metal steps internally. I would agree with harold for using AR. The amount of metal in the wafer is microns thick, typically measured in angstroms.

 

[yvonbug]

I have a few pics of them close up.View attachment 1 I forgot to add that these are from the 80's. And they are gold on the back. I'm not looking forward to scraping them off the sticky plastic, though. A lot of them are like those little stadiums, but alot are smooth across too.

 

[Smitty]

Very nice pics. I've learned from the guys on this forum not to include base metals in my solutions during refining, but I think I would have to break the rules on this one. Your solution may be tainted with some base metals during recovery and hopefully will be removed on the subsequent refining you will do to improve the quality. I would use 50/50 nitric/water and cook until you think all the base metals can be removed from the batch, wash with hot water until you think it's good enough to dissolve the gold using AR.

let me know how it goes, this is the second time i've dealt with Semiconductor Gold.

 

[qst42know]

This is a complete know nothing question. And my not be safe at all.

Could this solution be evaporated to dryness and incinerated to recover the gold? Or would this release the cyanide?

 

[goldsilverpro]

gst,

I can't grasp what you mean. If you're talking about incinerating cyanide chemicals or solutions, read this.

You might die. The last paragraph on page 10 of this link indicates that the incineration of cyanide salts will release hydrogen cyanide.
http://74.125.95.132/search?q=cache:NlxUFDJdFkcJ:www.atsdr.cdc.gov/toxprofiles/tp8-c5.pdf+cyanide+incineration&cd=8&hl=en&ct=clnk&gl=us

 

[qst42know]

I don't know anything of cyanide that's why I was just asking.

So hydrogen cyanide gas is liberated by the heat of incineration or by acids.

Good to know.

Thanks GSP. Qst

 

[netseeker]

I was employed as an equipment technician in a wafer fab until recently so I'm quite familiar with the manufacturing processes used. Scrap silicon wafers are gathered by manufacturing fabs for resale to solar cell manufacturers. This is a recent (the last 3 years) occurrence when solar cell manufacturers really ramped up production and couldn't get the silicon they required. I had worked the last 9 years in the PVD (physical vapor deposition) and we repaired equipment used to sputter a variety of metals including Ta, W, Al, Ti, Co and Cu. We actually did deposit Ni/Pt on wafers although the layer was very thin. The targets used were 10% Pt/90% Ni. Our research fab actually experimented with depositing gold as layer. The PM targets used are very valuable and are carefully monitored. The backside gold layer on wafers actually uses the same type of equipment. If you find whole Si wafers they are the ones the solar cell guys want. I don't think the wafer size matters either.

Robert

 

[u0uils08]

I just bought 30 5" gold silicon wafers, made by AT&T. I paid $57 for all them. One wafer is 7g. I readed that the gold that is on them is 24k, hoping that it is a good buy. I see two different views about these wafers on this post. After I'm done with them, I will post my findings. It may be a little bit because my wife is in labor.

 

[Smack]

Your new, never processed anything for PM's and want to make money right? Sell them. Look them up on Ebay.

 

[GOLDbuyerCA]

i bought a load, suv full, saved em from being turned into solar cells, sold quite a few , for collectors, teachers, educators, technical people in the biz. at Seaside Oregon swap meet. " i paid to much, i am told " as there is very little gold on the 8 inch wafers, Yet. i have seen assay reports out of Nevada, showing that the wafers, of 2 to 5, to 8 inch, are decimal 58 percent, that is 58/100 of 1 percent by weight.
some wafers, have two or three layers, of gold, used as heat sinking the circuits etched , by heat removal. and there is chrome, nickle, and titanium, and more.