jpneuby81 - 28-9-2015 at 21:26
So I've got a fair supply of various printed circuit boards from computers and other electronics. And like so many committed to this fool's errand, I
am trying to find an easier way to extract the various metals in the boards.
The experiment consists of a cell with somewhat isolated electrodes. Using a graphite anode in a smaller chamber for confinement and a stainless
steel cathode. The chambers are created by a wall obstructing any direct pathways between electrodes except for a small hole (less than 1/8") in the
wall.
My PCB is placed in the anode chamber close to, but not touching the anode. The electrolyte consists of an aqueous 700 mL solution, approx. .33 M
NaCl, .33 M NH4NO3 and .33 M NH4Cl. I set my power supply at a 3 volt limit, and limited to 100 mA. pH measured at
6.5-6.6 initially.
The theory was to somewhat concentrate the Cl2 around the PCB allowing it to oxidize it's various metals, while any plating on the cathode
be easily removed due to poor adhesion to the stainless.
After a 6 hr test run, the PCB showed plenty of signs of etching. Most solder and exposed metal was being dissolved, the bath's pH dropped to
2.6-2.8, and an insoluble white precipitate was being formed (presumed PbCl2 and AgCl). Only the slightest film was present on the
cathode, which easily wiped away. There was no change in color of the solution indicating no copper, nickel, iron, etc.
Electrodes were removed to rest the bath (PCB remaining), and the white precipitate settled with most changing to a metallic gray. The cycle was
repeated several times with the same results. Most of the silver-colored metal is now removed from the PCB and several of the monolithic capacitors
are beginning to fall off the board.
When last checked, the pH was the same 2.6-2.8, still no change in the solution's color, and the metal powder has accumulated a fair amount. Also no
real plating occurred at the cathode.
What I did notice was the surface tension of the solution went through the roof. When using plastic forceps to move the PCB, the liquid submerged
nearly 1/8" before wetting the forceps and the bubbles from the electrodes began to look suspended. Also the solution's resistance dropped
significantly as the current limiter kicked in on my power supply and began dropping the voltage.
There was a faint chlorine odor present at all times during the process, but never a smell of ammonia or any signs of NO/NOx gas. Oddly
enough, no copper seems to have etched from the PCB. There was very minimal errosion of the graphite anode (taken from a D cell zinc/carbon battery).
I have not tested any of the products so far, just looking for ideas at this point as to what exactly happened.
My theory was that through various half reactions, the ammonium ion would be partially consumed releasing nitrogen and H+ with some of it
being used in a metal-ammine complex. The NO3- --> NO2- and Cl2 --> Cl-
should have been able to dissolve most metals (although concentrations likely weren't high enough for more noble metals) forming appropriate Chlorides
and Nitrites, many of which should be insoluble.
The increased surface tension is clearly signs of increased ions in solution, and likely the formation of some kind of complex. The decrease in pH
also indicating an increase of metal ions acting as Lewis Acids and/or HCl/HOCl formation. Tin should have been the only abundant soluble metal,
forming a supply of Sn2+, which when the bath was at rest perhaps acted as a reducing agent on the insoluble white salts to drop metal
powder. The resulting Sn4+ could readily form many complexes increasing the surface tension, only to be reduced back to Sn2+ at
the cathode, thus acting as a sort of catalyst.
I hope to begin testing the products this weekend. Any thoughts on what to test for and how to test for it would be appreciated.
Thanks,
Neuby
hyfalcon - 29-9-2015 at 03:05
First thing you want to do is give everything a hydroxide soak to take the solder mask off and expose all the metal. I don't see going to the trouble
you are. Just dump them in a plastic bucket and cover with muratic acid and blow air through it with an aquarium air pump. Eventually, it will strip
all metal from the boards. Any gold will be flakes in the bottom of the bucket.
jpneuby81 - 29-9-2015 at 03:37
I suppose the key word in my original post was "experiment". I'm not using a PCB's with visible gold, btw. But if one could make an electrochemical
cell that simply needed a chloride refreshing every so often, then why not?
[Edited on 29-9-2015 by jpneuby81]
[Edited on 29-9-2015 by jpneuby81]
hyfalcon - 29-9-2015 at 06:01
Standard etchent once the solder mask is removed is a solution of ferric chloride. That will remove the copper. Any gold foil overlay will float.
Or I should say will be agitated off. Air bubbling helps this process also, if for no other reason the a little mechanical action.
This isn't exactly what you were asking for but it's an interesting read so here:
http://www.google.com.ar/patents/US4462879
Also:
http://www.google.com/patents/US4182671
and so on...
http://www.google.com.ar/patents/US4606797
[Edited on 29-9-2015 by hyfalcon]
[Edited on 29-9-2015 by hyfalcon]
[Edited on 29-9-2015 by hyfalcon]
[Edited on 29-9-2015 by hyfalcon]
pesco - 22-11-2015 at 03:24
Hi,
you can strip soldermask of boards using NaOH, however to accomplish it in reasonable time the solution should be rather warm - at least above 40
DegC. Working with warm (or especially HOT!) hydroxide is VERY DANGEROUS.
Added bonus is that you'll have all your Sn, Pb, Al and few others in a solution. Metals left are Cu, Ni, Fe.
If you leave it too long in hot NaOH the board itself will start to dissolve making mess.
So that way you can expose top and bottom layer of Cu, but what about multilayered boards ?
The only reasonable way to extract ALL the metals "locked" in PCB's is via pyrolysis.
The way I used to do:
- take empty propane/butane bottle and fill it with water purging any gas
- cut the top of with a grinder
- weld 6 nuts (size 10 or 12) to the bottom part, around perimeter just below the top edge
- weld 6 identical nuts but with drilled out thread to the top part. Without thread the bolt should go through the nut easily without having to turn
it.
- make sure both line up nicely. they will be your tight locking mechanism - just bolt both cut sections together after filling up with PCB's and a
bit of WATER
- glue glass rope (the type used in sealing fireplaces) on top of the edge
Now you have vessel in which you will place your boards. It can be easily opened, filled with stuff to pyrolyze and then sealed nicely using 6 bolts.
- modify the bottle top fitting to accomodate 15mm copper pipe
- out of empty steel paint can make a bubbler by cutting two holes in a lid and fit brass tank connectors
- now connecting, the pipe coming out of your gas bottle should go at least a meter up in the air (I found 2m works better), then you make a bend and
plumb it into you bubbler. In the bubbler it should go nearly to the bottom - leave around 1 cm. Through second hole mount another pipe and run it
into the burner.
The easiest way is to make fireplace out of bricks, stick your modified gas bottle in it. Fill the bottle with your boards, ADD 0.5-1L OF
WATER. The water is crucial. Once your fire starts going the water will turn into a steam expelling all the oxigen from your vessel. It is
important for two reasons : 1) you want to perform pyrolysis so no O2 allowed and 2) more important, O2 present in your vessel once the boards starts
to decompose could potentially cause expolosion.
Heat the vessel in your fireplace for several hours. How long is hard to say as you can't see the inside of the vessel.
Let the vessel cool down and open it. If you bards are all crumbly whiteish ash your good. If some are still not completely pyrolysed then start your
fireplace for few more hours. Once you do batch or two you'll get the "feeling" for time needed.
Fill your bubbler with water leaving few cm space below the cover and then seal it. If the cover is loose then clamp it with couple of threaded rods
and steel bars. Its generally good idea to secure the lid that way anyway.
Once you have your "ashes", brake them into a sand like consistency. Finer (dust) is OK.
This stuff you can easily leach with NaOH first (Sn, Pb, Al) then H2SO4 or HCl.
The boards do not have to be depopulated, however it is VERY good idea to remove some of the components:
- any "springy" strips - they might contain Be (usually as alloy Berylium Copper) and you REALLY don't want to mess around with it
...
- any larger inductors (check google how they look like) - they are essentially Fe core with Cu wire. Easier to separate mechanically and if removed
you'll use much less reagents
- any Al or Cu heatsinks, same reason as above
- any steel fittings, bolts, screws etc, same reason as above
Edit: cosmetics
[Edited on 22-11-2015 by pesco]
[Edited on 22-11-2015 by pesco]