Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Extracting Cobalt from cemented carbide tools.
I've been playing about with some broken carbide end-mills, trying to extract the cobalt. The "carbide" is basically wolfram carbide in a metal
matrix, usually cobalt.
Experiences so far:
Due to the nature of the material grinding is out of the question. Even breaking it down into coarse chunks is out of the question. Just too much
brute force and flying shrapnel flying about.
Acids are slow and wasteful. Concentrated or diluted, warm or cold, only a limited concentration is achieved, not enough to freeze out the salts.
Precipitation as insoluble salts requires low pH, so a lot of acid must be neutralized in the end.
Next I'll try roasting the tools, I need to heat treat some tools so I'll toss in the mills in for a run or three. Don't know how fast it oxidizes or
if the oxides will form a protective coating or not. But hopefully the cobalt oxide should be easier to dissolve in acid.
Processing by precipitation as the hydroxide or carbonate seems like the obvious route, but I also want to explore precipitation from ammonia
complexes. Basic copper carbonate can be produced by boiling off ammonia from an ammonium carbonate leeching solution. This process generates a dense
precipitate that is far easier to work with than the directly precipitated compounds.
Any other suggestions for what I should try. I like to work with OTC chemicals, so nothing to elaborate please.
Edit: Clarified the topic (added "cemented").
[Edited on 16-4-17 by Fulmen]
We're not banging rocks together here. We know how to put a man back together.
|
|
|
Melgar
Anti-Spam Agent
Posts: 2004
Registered: 23-2-2010
Location: Connecticut
Member Is Offline
Mood: Estrified
|
|
By "wolfram", do you mean "tungsten"? If so, tungsten can be dissolved by H2O2, whereas cobalt cannot. That's really slow though. Try electrolyzing
in NaOH or KOH with the tool as the anode. I'm not sure what cobalt would do here, but tungsten dissolves. Also, if you add NaNO2 to the
electrolyte, there will be a massive increase in current and reaction speed, if it's anything like when I did it.
|
|
|
karlos³
International Hazard
Posts: 1520
Registered: 10-1-2011
Location: yes!
Member Is Offline
Mood: oxazolidinic 8)
|
|
Very likely tungsten, since it´s chemical short is "W" for wolfram.
|
|
|
Melgar
Anti-Spam Agent
Posts: 2004
Registered: 23-2-2010
Location: Connecticut
Member Is Offline
Mood: Estrified
|
|
I kinda figured, but it seemed more polite to ask a question than to correct the OP, and besides, I wanted to make sure anyone else reading it
knew what he was talking about.
|
|
|
clearly_not_atara
International Hazard
Posts: 2692
Registered: 3-11-2013
Member Is Offline
Mood: Big
|
|
Tungsten metal will dissolve in H2O2, but he's clearly stated he's working with tungsten carbide in a cobalt matrix, so that's out of the question.
Cobalt might dissolve in aqueous ammonia containing ammonium acetate (2:1 NH3:HOAc):
4 Co3O4 + 36 NH3 + 36 NH4+ + 3 O2 >> 12 Co(NH3)6(3+) + 18 H2O
Upon addition of a source of Br-, hexamminecobalt tribromide will precipitate from aqueous solution. It may be possible to dissolve the cobalt
galvanically by adding a suitable cathode e.g. copper.
[Edited on 16-4-2017 by clearly_not_atara]
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
Interesting. If I didn't already have some cobalt salts, I'd probably whip up a batch of ammonium acetate and try dissolving some drill bits.
Tungstates look like fun.
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
I'm not really interested in the tungsten carbide now, just the cobalt. Ammonium acetate sounds like an interesting route, I'll give that a try. I
assume air is needed to oxidize the Co? It sound similar to the ammonium carbonate leech for copper...
We're not banging rocks together here. We know how to put a man back together.
|
|
|
diddi
National Hazard
Posts: 723
Registered: 23-9-2014
Location: Victoria, Australia
Member Is Offline
Mood: Fluorescent
|
|
the OP did not make it very clear (but has since given more detail) that the actual material he is trying to work with is not WC - tungsten carbide
but actually a cobalt dosed tool steel. usually 5% Co
so depending on what is in the steel (there is a cocktail of refractory metals used to improve hardness or heat disipation etc which includes W, Mo
and others as well as a like dose of Cr) it may be a long slow process with H2O2 i believe. I cannot imagine any way of selectively dissolving Co in
this situation. I use this method for dissolving TIG electrodes to obtain Th or other goodies. If you can get HF that will help things along.
Beginning construction of periodic table display
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Diddi: Not tool steel, cemented carbides. The composition can vary, but it's usually tungsten carbide (and/or other carbides like TiC or TaC) in a
cobalt (or nickel) matrix. Perhaps 10% Co from what I can gather.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Clearly_not:
You might be onto something there, it is certainly producing a medium concentration solution in a few hours. Hard to judge from color, I'll probably
let it sit for a few days to be sure before precipitation (need to get more sodium carbonate).
I wonder if the ammonia/ammonium-carbonate trick might work? That has the added benefit of producing dense carbonates directly from the leech.
I like to work with OTC chems as long as possible, but that list is getting shorter year for year. The goal is of course to find the minimum-effort
route, simple chems and simple equipment. Getting a crude product is fine as long as it has some use, either as a catalyst or a convenient precursor
to other or purer products.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
clearly_not_atara
International Hazard
Posts: 2692
Registered: 3-11-2013
Member Is Offline
Mood: Big
|
|
Solubility of hexamminecobalt acetate is 1.9M so that's 110g of cobalt per liter of solution in the best case. Actual solubility may be somewhat less.
I think carbonates are generally less soluble than acetates unfortunately but it could work I suppose.
Edit: amusingly the post in which I suggested that was my 420th.
[Edited on 17-4-2017 by clearly_not_atara]
|
|
|
Melgar
Anti-Spam Agent
Posts: 2004
Registered: 23-2-2010
Location: Connecticut
Member Is Offline
Mood: Estrified
|
|
I don't think you can or should attempt to only dissolve the cobalt. You want OTC? Get an old USB cable and cut the end off that would plug into
whatever device it formerly went with. The red and black wires are +5V and ground, attach alligator clips to those, or wait and bend copper wire into
clips that'll fit on your electrodes, once the rest is set up, and solder the wires to the clips. Now find one of those phone charger things that USB
charger cables plug into. If you have more than one, find the one rated at the highest amperage. Use chlorine bleach as an electrolyte. Attach the
red wire to the thing you want to dissolve, and the black one to... a different one made out of the same material that you want to dissolve later.
That way there's necessarily no contamination, and you don't even have to worry about hooking the wires up right.
Now plug the whole thing in and leave it overnight, somewhere with good ventilation. If you have a voltmeter, check the voltage across the
electrodes, and add electrolyte if it's 4V or more. Keep adding until it's below 4V. Phone chargers are current-limited, so it won't short out. It
should look like a big mess the next day, because it's being stirred by evolving hydrogen. Anyway, neutralize the chlorine however you want to, but
try to minimize the number of new elements you add. I recommend hydrogen peroxide. You'll get lots of oxygen bubbles and hypochlorite will become
chloride. Stir it up good, remove the electrodes, and let it stand for like 30 minutes, then decant if the water is colorless. Do this a few times
to get rid of anything water-soluble.
Now you're in a much better position to separate the metals. I'd recommend separating iron and cobalt by the difference in solubility of their
sulfate salts in methanol.
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Heat definitively works. Tossed the parts in the furnace @850°C for a few hours, came out with a 3-5mm coat of oxides that peeled off with little
effort. The parts were about 1mm smaller afterwards. Next is a few days in NH3/HOAc.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
yobbo II
National Hazard
Posts: 709
Registered: 28-3-2016
Member Is Offline
Mood: No Mood
|
|
If you added steam to the furnace it would convert all of the Co + carbide to oxides IWI
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Huh, guess I need to learn more about Co and W chemistry. The oxides does not dissolve appreciably in either HCl or NH3/HOAc, so next up is sodium
hydroxide. This should produce soluble sodium tungstate which can be converted into WO3 with acid. Don't know if it will help me get any soluble
Co-compounds, but at least I'll have a more concentrated material to work with.
Update: This is working far better than expected. The tungsten dissolved fairly easily in NaOH, precipitation with HCl produced the expected
voluminous white precipitate of tungstic acid.
Cobalt was extracted from the remainder and precipitated with sodium carbonate. It was then partially complexed with ammonia and boiled to produce a
more filter-friendly product.
[Edited on 20-4-17 by Fulmen]
We're not banging rocks together here. We know how to put a man back together.
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Tungsten chemistry is actually quite interesting as well. It appears I made ammonium paratungstate (NH4)10(H2W12O42)·4H2O rather than tungstic acid,
but that's OK. It should decompose to WO3 at 600°C, from there I'm guessing hydrogen reduction is the only route to metal.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
AJKOER
Radically Dubious
Posts: 3026
Registered: 7-5-2011
Member Is Offline
Mood: No Mood
|
|
Actually, my only need for cobalt compounds are in trace amounts as apparently it is a highly effective catalyst in varoius redox reactions.
This is a good thing as toxicity is generally cited as to why it should not be employed even in small amounts (see, for example, https://www.environmental-expert.com/articles/successful-deg... ).
This being said, one needs, in my opinion, to only prepare very dilute salt solutions from the cobalt metal itself.
[Edited on 28-4-2017 by AJKOER]
|
|
|
clearly_not_atara
International Hazard
Posts: 2692
Registered: 3-11-2013
Member Is Offline
Mood: Big
|
|
CoBr2 catalyzes the addition of aryl halides to zinc in acetonitrile. This can be used for certain nucleophilic aryl substitution reactions
(particularly Michael additions) although the arylzinc reagents are less reactive than the corresponding Grignards.
In particular, phenylzinc bromide reacts with phosphorus trichloride to dichlorophenylphosphine:
https://dx.doi.org/10.1002/hlca.19520350443
and apparently dichlorophenylphosphine will react again:
http://onlinelibrary.wiley.com/doi/10.1002/ange.201404019/fu...
It's not clear from the abstract whether this reaction can be continued to triphenylphosphine, since probably the goal of the experimenters was to
produce the more functionalized compounds which are the subject of the paper, and triphenylphosphine is a commodity chemical. However, it appears
promising, and generally I would expect a phosphorus halide to react with an arylzinc.
Diphenylchlorophosphine is also interesting since it can be converted to bis(diphenylphosphino)ethane by reduction and reaction with ethylene iodide.
[Edited on 28-4-2017 by clearly_not_atara]
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
I'm having some odd results with the tungsten oxide/acid. Extracted with NaOH and ppt with HCl I got a fine, white precipitate. Both WO3 and H2WO4 are
supposed to be yellow, right?
After filtration the filter was left on a luke warm place to dry out, after a few days all that was left was a dark glass-like substance. Ground down
I got a tan dense powder. Any ideas as to what I have?
We're not banging rocks together here. We know how to put a man back together.
|
|
|
Fulmen
International Hazard
Posts: 1693
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
I'll try to get some pics of this on the next batch (assuming it happens again).
Leaching the oxides turned out to be slow work, so in the end I collected all the insoluble material (filter paper and all) and fused it with NaOH.
This seems to be a very efficient approach, more than 90% dissolved easily in water after heating to a dull orange with a propane torch.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
