Electrochemistry Copper electrorefining question

[solar_plasma]

That being said, let us come to the calculation of amortizing. If this facility will cost 1 million € (which won't be enough) and if he will raise his profits 10% (which he won't), it will take copper and pm's for 10 million just to pay for building the plant. Only calculating copper (because 90% of your material has to be something pure like millberry, this will be about 1500 tons,if he got them for free. Let us say,he gets them at 50% (which he won't -didI say this before?), he needs to process about 3000 tons. Let us say, he can process 20 tons per month (which will be hard to achieve), it will take 12,5 years just to pay for building the plant.

Very,very simplified with numbers far more positive, than they will be, not calculating the running costs, wages, taxes, repair, transport, assaying etc.

300 tons will be copper from incinerated and smeltet e-scrap,which contains gold for about 700 000 €. so,for the running costs, he has 56000€ per year left. If it is only one man who runs this, nothing has to be repaired and everything runs very lucky, - this size might let him live acceptable, - at least if he gets electricity for free.

Sincerely, you will have to have a long wind.


used sources:
http://psrcentre.org/images/extraimages/312513.pdf
Cu 7€
Au 30€

Now try to do this calculation without investing 1 million and just sell the bars to a big boy. Don't do anything your neighbor does better and cheaper than you, but find something, you can do cheaper and let him do the rest.

Just my two cents. I don't know, if I am right. Just wanted to give you some impulses to think of.

 

[g_axelsson]

Interesting... I had to bring out my copy of Davenport again and read up on the details. 8)

What I get is that one of the important things with low impurity (low oxygen among other) is to get a flat cast of the anode. Uneven surface would affect the dissolving rate over the surface and force the change of anode earlier. They even include thinner top parts of the anode to minimize the amount of recycled copper in the anodes. (pg. 255)
Some refineries even have a machine to flatten out the anodes to improve efficiency.

But still, I have just read the full chapter 16, "Electrolytic Refining" and there was a lot about keeping the electrolyte fresh and loaded with copper, how the electrolyte was cleaned from impurities, how to find and deal with shorted cells.... and so on. A lot of information of how it is done but no reasons why it would be impossible to run anodes of lower quality.
I get it, if the goal is to have a large smooth running operation with copper as main product then there is no reason to run impure anodes. The impurities would cause the casting of the anodes to be of bad quality, the amount of scrap anodes to recycle would increase, the risk of shorted cells would increase.
Short answer would be that any refinery in such an evolved market that copper refining is would be crazy to run the process on anything but peak efficiency. It would eat up any slim margin they operate on.

The 98,4-99,8% purity of the anodes mentioned in Davenport is for a primary smelter / refinery. But on page 360 (chapter 21, Chemical Metallurgy of Copper recycling) it is mentioned "The impurity levels in secondary anodes is higher than that in most anodes from primary smelting operations." No numbers are mentioned until in the summary where the process is simplified into "Molten black copper" (80% Cu) -> converted into "rough copper" (96% Cu) -> fire refined and cast into anodes (98,5% Cu).

But what if the goal is to recover the gold, silver and palladium in the refiners bar and copper is just a secondary product maybe the trick is to do a quick refining where most of the impurities are removed and then do a second refining in a pure cell.
To be able to run larger amount of impure copper you would need a larger amount of electrolyte in the tank, maybe a separate tank to pump it into and out of or replace a large part daily. You would need to increase the distance between the electrodes to minimize the risk of shorts but not too far, the efficiency would drop off.

Now when I have read up on the process I saw some things that might affect the surface.
- Temperature, cells are heated to 60-65 C
- Electrolyte circulation - removes impurities, gives smoother concentration across the cathode surface which gives better surfaces. In commercial cells the electrolyte is replaced in a couple of hours.
- Current density, usually below 300 A/m2 Depending on the impurities in the anode! (to keep down risk of passivisation of the anode)
- Circulation of the electrolyte should not bring any slime to the cathode
- There should be a tiny part chloride ions in the electrolyte to improve grain size of the deposit. But with stainless steel cathodes the level must be kept below 30 ppm to avoid Cl corrosion and the copper layer to stick hard to the surface. (pg. 334)

I've spent far too much time on this now... but it is so fun doing research. :mrgreen:

Göran

 

[solar_plasma]

A big problem with suboptimal conditions is, that the copper will not cling to the cathode, fall down in parts and contaminate and shorten the whole cell. In a 5 liter cell you can remove it, in a large tank this will be nearly impossible.

I would like to think those problems could be solved, but obviously no one has actually solved it, yet. Which makes me believe, a solution must be hard to find or quite does not exist...at least not for running tons of copper economically.

On the other side, my only experiences are from plated german silver with much higher percentages of Ni and Zn. But still, investing lots of money in a process, that might not work (most probably not), he is walking on very thin ice.

Björn

 

[solar_plasma]

Another problem comes to mind. Silver dissolves in that electrolyte, leaves the anode bag and cements everywhere. Further, it will a) consume some of the chloride (AgCl-solubulity 1,88 ppm at 25C), which is b) needed for the thiourea reations and c) has to be kept below 30ppm in order to avoid corrosion of the steel cathode, if you use steel. I can't see any other way, than to reduce the silver content in the anode material to the nominal 0,01-0,6% (Davenport), while the chloride concentration has to be monitored and depending on that, chloride has to be added continously. Maybe titan or pure copper cathodes would make the chloride concentration less critical, so all AgCl could be trapped within the anode bag. Maybe! Silver chloride can form very small particles, as we all know.

The more you think about it, the more complex and fragile this process is getting. No room for deviating from experienced, working parameters. Though if you should find out, I am pretty sure, you could write a doctoral dissertation on that.

Since I don't know EdwinEL's chemical background, I would like to add, that thiourea, nickel and many other compounds in this process are carcinogen and will enter the air together with acid particles as aerosols, which to some extent can be avoided by correct voltage and addition of surfactants, but still have to be monitored in such a professional scale setup. Protective wear is mandatory.

 

[EdwinEL]

Solar_Plasma, I have read the paper from N.T. Beukes and J. Badenhorst, but it is about electrowinning not electrorefining so it does not have anything to do with what we are trying to acomplish.

Now, we are not focussing on reaching the highest purity copper (it will be nice if we reach it though). The figures in the ICCEE document are way higher then we calculated on precious metal content.
The copper content is about right. Now as you can see by the values of the materials copper accounts of only 10 to 20% of the revenue. So it is rediculious to say that we are mainly a copper refiner.
We do process copper but it is a side product of the actual value in the PCB's.

I do enjoy the research you guys do.

Behind the electrorefining cell there will be a tank with about the same volume of electrolyte that is in the electrorefining cell. We do not use Anode bags we will let the anode slime drop to the bottom of the tank. That's why I do not like it that my copper does not perfectly plate on. Otherwise it would be no problem, we would just dry the dropped copper and melt it. As it does pass trough the electrolyte and the anode bags do capture the precious metals. "

While I do understand your point of just melt the PCB's and sell the "refined" metals to let's say Aurubis. This is possibly but quite stupid, because the process of smelting is the most expensive of all.

So basicly Göran underlines what we try to do here.

Basicly (in the very basis of electrorefining) there are the following elements in a cell;
The cell it self (plastic or concrete)
A power source (we have a variable magna power power supply, we will do electrorefining at around 0.5V with 280 A/M2 (lower production better quality))
The cathode (in my case they are made of 316 L SS with a copper hanger bar) I have produced them at 80 by 100 cm
The anode (from the former smelting process high purity anodes are the only way to go)
The electrolyte (variables are; temperature, amount of h2so4 and copper in the solution, amount of impurities)
Circulation pump ( we have bought circulation pumps which circulate the electrolyte every 5 hours)
Bonus items; Thiourea, Hcl and Bone glue.

So basicly without talking about any economics. The basis for and electrorefining cell is there.
It is just (please do not see this sentence of me taking refining to lightly) a matter of getting all the variables in the cell correct.

So without taking economic values in account would you agree, disagree or comment about the above statement made?

 

[solar_plasma]

So it is rediculious to say that we are mainly a copper refiner. We do process copper but it is a side product of the actual value in the PCB's.

And that is the way it won't work. Everybody who uses this material in this process successfully and economically is mainly a copper smelter and refiner, who has precious metal as a byproduct. A llot of people here have tried, what you want to do. All those threads are still waiting on the holy grail, as far as I have seen.

It is just (please do not see this sentence of me taking refining to lightly) a matter of getting all the variables in the cell correct.

just? just?? :lol: the correct variables are the holy grail

So without taking economic values in account would you agree, disagree or comment about the above statement made?

From my deepest heart, I disagree, but I enjoy reading about and learning from your experiments, since it is not my money. I would really love to read, that you have made it work! Chances are going against zero, though, but I can be wrong. Good luck!

 

[solar_plasma]

The figures in the ICCEE document are way higher then we calculated on precious metal content.
The copper content is about right. Now as you can see by the values of the materials copper accounts of only 10 to 20% of the revenue.

TABLE II
PCB
FEED COMPOSITIONS
PCB Feed Compositions
(wt %)
Gold 0.039
Silver 0.156
Palladium 0.009
Copper 18.448
Other metals 9.35
Non-metals

That is the composition of the feed stock before incinerating. So after incineration and melting you have about 68% copper, about 30% other metals (theoretically mainly tin, lead, nickel, iron, cobalt, alu) and about 2% Au/Ag/Pd.

 

[Harold_V]

Contaminants in copper are generally slagged off (oxidized), which raises the copper content. As Göran has suggested, it makes little sense to add pure copper, as, in the end, you'll still have processed only the same amount of waste alloy, so nothing would have been gained.

Whether slagging would be suitable in this case, or not, I can not say. I also expect that it may not be as easy with an induction furnace (you alluded to a large kw furnace, which I assume to be induction) as it would be in the typical furnace used in copper smelting. If you would be satisfied with sending some elements to waste instead of making a recovery, that would be a solution for you. The vast majority of the values would report in the copper, with base metals (aside from copper) lost to the slag.

My chief concern with 316 stainless (or any other alloy of stainless) is that it will most likely react with the sulfuric acid in the electrolyte. I have no experience in that regard, I'm simply relating what I learned as a machinist (in regards to titanium being used for reusable cathode starter sheets). Were nitric acid be the constituent of the electrolyte, I'd give stainless my full approval.

If you have evidence that stainless is being used with success, go for it. I may be wrong.

Harold

 

[g_axelsson]

According to Davenport (in 2000), most refineries have already converted to 316L stainless steel as cathode blanks. Complete with a polymer edge strip and a V groove at the end for easier removal of the anode copper.

The steel should be protected by the electric potential of the cathode. The main danger seems to be higher chloride levels that would induce pit corrosion.

Göran

 

[nickvc]

Guys the way I see this is that without constant monitoring, testing and changing of the electrolyte there is no way forward that will yield the results you want, a good copper deposit and the values left with a lot of other contaminants in the slimes to process or recover later. That means large amounts of acids and metals needing recovery and treatment which all cost time and money.
I wish someone could crack this particular nut as it would bring lower grade scrap into the well set up home refiners scope but i fear that's not going to happen any time soon, the amount of copper needs reducing to leave a more concentrated mix of values and the only way is by electrolytic refining due to the cost and amount of acids needed to remove it any other way and the problem with electolytic refining is it needs high grade feedstock to work efficiently.
The only possible way I can think of is using a ceramic membrane or membranes that will trap everything but the copper or vice versa, I doubt such a thing exists unfortunately as that could be the answer we are all looking for.

 

[goldsilverpro]

Edwin,

Unless I missed it, no copper sulfate was mentioned in your solution makeup. Also, the sulfuric acid was quite high.

Here is a typical solution makeup and operation parameters. These can vary from tankhouse to tankhouse.

CuSO4.5H2O = 100-160 g/l
H2SO4 = 180-250 g/l (I calculate this as about 10-14% conc. H2SO4, by volume)
[Cl-] (added as HCl) = .02-.05 g/l
Thiourea = .0001-.0005 g/l
Bone glue = .0001-.0010 g/l
Temp. = 55-65 C
Cathode Current Density = 130-170 A/m2

The 90% copper you mentioned seems high to me. In most of the installations I worked at that produced refiner's bars, it was lower, depending on the boards. Sometimes as low as 65%.

No matter what you do, you still have all those impurities poisoning the solution to deal with. All of the literature spoken of deals with anodes made from blister copper, which is typically 99%. You are dealing with something totally different. Unless you can find a way to first upgrade the anodes to 99% or continuously remove the impurities from the solution with something like ion exchange or solvent extraction, I think you have a losing proposition. If you want to make money from this operation, ship the refiner's bars to a primary copper facility. Forget processing them on your own.

 

[solar_plasma]

Though GSP just says in better words, what I tried to say, and what I have learned mostly here, I can't stop the flow of mind, when it first has been started, maybe just like Göran. So, I'd like to add my ideas as they come, if this is ok. Don't get me wrong, I still say, it will lead nowhere.


I'd say, back to the roots. Hoke tells to separate as much as possible the mechanical way. Hoke tells to leach all accessible base metals then.

The bare PCBs are pretty pure copper. Some high grade boards might have tiny platings of nickel and gold, though not that much, that it will disturb the process.

If you find a way to separate the bare boards from the rest economically and without generating tons of diluted heavy metal loaded solutions, you would have something, that would work well in your process.

The rest will at least have a lot more pm's and will weigh far less.

The capacitor caps are mostly aluminium and manganese dioxide. Those could be collected and parted. Aluminium can be melted. Manganese dioxide can be smelted.

Then you have ferrite core inductors. Easily parted into iron and copper.

The rest might be milled and separated from large magnetic parts, leached for tin (the pros do know, which chemical is used, can't remember that, since it is nothing for me), incinerated, milled, separated from the ashes by gravity. The heavy powders might be a candidate for HCl-leaching of base metals, the leftover solids go to, for example, AR, then the solids can be leached for silver chloride.

Not cheaper or less work than your idea, but based on working processes and probably not more expensive.

By the way, Kevin alias kjavanb123 has started experiments in that direction. You can find his thread about tin recovery and mlcc processing.

 

[Harold_V]

According to Davenport (in 2000), most refineries have already converted to 316L stainless steel as cathode blanks. Complete with a polymer edge strip and a V groove at the end for easier removal of the anode copper.

The steel should be protected by the electric potential of the cathode. The main danger seems to be higher chloride levels that would induce pit corrosion.

Göran

Thanks, Göran. I stand corrected!
My years away from refining are beginning to show.

Harold

 

[richoc]

Just to put 2 pennies in here.
The copper will never plate out of the electro winning system smothe and even.
Copper likes to make square crystal shapes that is what your seeing.
That is how they want to grow on each other here on earth.
As more pure crystals grow they start to grow out to the sides growing on each other.
Scrape them and put the retrevile plates back in.
You are recovering copper not refining it at this ponit in the game.


You are not copper plating a object with a Micro coating that lays on as the object moves through the tank at a rate time to lay it on just so thick.

 

[skippy]

I don't know if it was mentioned here or not, but I do recall reading you can get around the problem with the poor plating by stopping the plating part of the process. Basically what it means is you use electricity to dissolve the anode, and a porous membrane prevents the dissolved metal ions from getting to the cathode and plating onto it. You need to use more acid in this process.

 

[solar_plasma]

I wonder how far the OP did get. Especially if he has learned, how fast the electrolyte gets fowled by nickel. When the electrolyte has turned from copper II sulphate blue to nickel green the cell doesn't work satisfying anymore and with nasty anodes like german silver, this takes only a day.

 

[solar_plasma]

Solar_Plasma, I have read the paper from N.T. Beukes and J. Badenhorst, but it is about electrowinning not electrorefining so it does not have anything to do with what we are trying to acomplish.

I hoped the OP would discover the mistake in this post by himself, as soon as he has read Davenport. Since there didn't come anymore until now, I would like to add:
Actually, electrowinning IS part of the rejuvenating process of the spent (contaminated) electrolyte (Davenport, chapter 19). Though another option mentioned is vaporizing water and collecting CuSO4 crystals. Also the other contaminants are removed by electrowinning and crystallization (NiSO4). The leftover H2SO4 will be used in the cells again. As we know, this is not a perfectly selective process and will not give all needed components quantitatively back to the process. So, the rejuvenated electrolyte has to be monitored for concentration of needed components as well for contaminants.

so it does not have anything to do with what we are trying to acomplish.

Maybe not, though I am not sure you have any clue what you are trying to accomplish.

 

[goldsilverpro]

Just to put 2 pennies in here.
The copper will never plate out of the electro winning system smothe and even.
Copper likes to make square crystal shapes that is what your seeing.
That is how they want to grow on each other here on earth.
As more pure crystals grow they start to grow out to the sides growing on each other.
Scrape them and put the retrevile plates back in.
You are recovering copper not refining it at this ponit in the game.


You are not copper plating a object with a Micro coating that lays on as the object moves through the tank at a rate time to lay it on just so thick.

richoc,

I skimmed through your other posts and have come to the conclusion that you constantly make incorrect statements and state them as facts. This is not acceptable on this forum.

First of all, this isn't an electrowinning system, since the anodes dissolve. The correct term is electrorefining, but that's already been covered in this thread.

Secondly, you say that copper will never plate out smooth and even. I admit that with 10% impurities, it won't, but I can't see how this has anything to do with the basic crystal structure of copper. If things were done properly, the copper will plate out very smooth and even. Proof of this is seen on this link at the bottom right of the page. You can't get much smoother and even than that and it appears to be at least 1/4" thick.
http://www.isaprocess.com/EN/AboutIsaKiddTechnology/Pages/AboutIsaKiddTechnology.aspx

The reason the copper in the photos on this thread is lumpy and non-adherent is due to the inclusion of the co-deposited impurities into the crystal lattice structure of the copper. So, actually, the smooth even copper is spoiled by the impurities.

I would suggest that you study the subject before making these false statements. Also, I would suggest that you either learn how to spell or use spell check and correct your mistakes.

 

[Evan2468WDWA]

I've been thinking about this problem, and I'm curious what other people think of my thoughts. What if you had a separate cathodes for each type of metal that can contaminate the solution. Each of them can be used to remove that particular metal from solution. In order to control this set-up shine a halogen light through the solution, and monitor the results with a spectrometer. Different metal ions have distinctive colors so this seems feasible. Based off the amount of the different contaminants in solution the computer can adjust the current flow to the different cathodes.

 

[Harold_V]

I'm far from being a chemist, but I don't think that would work, and if it did, it would be only somewhat selective. Reason? The cathode plays no real role in which metal would be deposited. If it did, plating of elements would be impossible, as most plating is done for the purpose of coating a base metal with an element that displays particular characteristics and is unlike the element (or alloy) being plated.

Some of you guys with the knowledge will chime in here, I'm sure, but it seems to me that if a person was careful to select the proper voltage for each element, you might achieve success to some degree. You'd most likely have to start with the element that responds to the lowest voltage, then step up accordingly. That, of course, is assuming that the element in question can be plated, or otherwise recovered, electrolytically from the given solution. If the objective was to recover the elements for further use, or recycling, then the cathode might require changing for each element.

A good example of this might be a nickel plating solution that is contaminated with copper. Could the copper be removed, restoring the nickel solution to a useable condition?

Comments?

Harold