Sciencemadness Discussion Board

Platinum-Tungsten Separation and Other Mystery Metals

kilowatt - 5-3-2008 at 14:47

I have a couple large pieces of tungsten/platinum alloy supposedly containing about 2.5% platinum. I have been trying to dissolve them down by anodizing them in concentrated sodium hydroxide, but this has proven to be a very slow process. The initial tests I did with this process on pure tungsten rods proceeded to dissolve them very quickly. However after weeks of continuous electrolysis, only mild pitting can be seen in the alloy metal pieces, while the graphite anodes have eroded away, been replaced, and eroded again into graphite sludge and CO2.

I need a better way to dissolve the bulk of the metal down, either leaving the platinum behind or leaving it as an easily separable compound. The chemistry of these metals is difficult though and I can find pretty much no useful information regarding how quickly various chemicals attack them.

I have considered refluxing the pieces in bromine, as perhaps that will dissolve away the tungsten more quickly, but I'm not sure what oxidation state it will go into. I am not sure if bromine will attack platinum at all under these conditions, and if so what oxidation state it will go into. I have also considered manually oxidizing the pieces by heating them in the air with an arc and tungsten cathode, and then reacting that with concentrated alkali, dissolving it layer by layer. The two remaining options I have considered are total dissolution in aqua regia (but I don't know anything the chemistry of tungsten in aqua regia) and anodizing in hydrofluoric acid doped with a conducting fluoride to boil off tungsten hexafluoride gas and be left with some sort of platinum fluoride (not entirely sure this would work as there are again other oxidation states).

I have another supposed source of platinum which is in the form of some metal foil that was rolled onto tungsten plates. The foil is in two layers, with the outer layer (about 0.001" thick) supposedly being platinum. I have weighed rectangular pieces of the foil which I have peeled off, and the density, roughly as I can measure it, does appear to be upwards of 20g/mL. The bottom layer of foil is a different metal which tarnishes when heated strongly unlike the platinum layer. I tried to dissolve the ground off foil in boiling aqua regia a couple times, and even fine particles have persisted after many hours of boiling (and releasing noxious fumes) even though the reaction was violent at first, and it appears that the vast majority of the metal has not dissolved. Even pieces of the bottom layer metal have not dissolved (though that could be due to passive oxide) so I'm not sure what that is. The solution has remained deep red over time but I am not sure if this is just left from the aqua regia's color (it is close to the original color) or the red color of hexachloroplatinic acid (which has a similar color). I have not done any tests on the solution but it has since been contaminated as I have had it around for some time and flies went into it and died. The resistance of the metal to boiling aqua regia seems to suggest it is either a metal which is not platinum (perhaps worthless iridium or tantalum or something :() or alloyed with another metal, making it more corrosion resistant. Again I have considered other methods of attack such as refluxing in bromine, or smelting with sodium nitrate in my little arc furnace.

[Edited on 5-3-2008 by kilowatt]

12AX7 - 5-3-2008 at 14:56

Have you tried my friend Molten Sodium Chlorate? :D No idea if it'd react with tungsten metal, but if it does, it's sure to produce HCl, sodium tungstate and sodium whatever-else-ate, plus plenty of heat.

Tim

kilowatt - 5-3-2008 at 15:18

Quote:

No idea if it'd react with tungsten metal, but if it does, it's sure to produce HCl, sodium tungstate and sodium whatever-else-ate, plus plenty of heat.

HCl? There is no hydrogen in sodium chlorate or tungsten metal.

Polverone - 5-3-2008 at 15:38

Tungsten is pretty resistant to attack from most reagents, but it dissolves tolerably fast in hydrogen peroxide. I'd try using 30% peroxide before anything really exotic. I don't know if the platinum content is going to catalyze peroxide breakdown to an extent that it won't have a chance to dissolve the tungsten, but I think it's worth trying.

not_important - 5-3-2008 at 16:56

seconding Polverone. You can use dilute NaOH or Na2CO3 solution, and slowly add the H2O2 to it which will reduce any loss from Pt catalysing its decomposition. If you have a soluble fluoride, a _small_ amount can help speed things up.

Tungsten is rather resistant to aqua regia, HNO3 plus HF will dissolve it. Fusion with NaOH + NaNO3 is another, as is passing chlorine over it at 400 C or so to get a mixture of the hexachloride and several oxychlorides.

I'd really advise against "boiling off" WF6, unless you're doing this somewhere way away from anything important including yourself.

I'll try to get back with more.


BTW - I'll take any worthless iridium you have off your hands, rather than just throwing it in the dustbin.



[Edited on 6-3-2008 by not_important]

Fleaker - 5-3-2008 at 17:09

Likewise on the iridium.


I'm seconding the use of peroxide.

Reason other acids (save HF) don't work well is that you get a tough tungsten oxide layer prevent further attack. Same thing happens with molybdenum.

12AX7 - 5-3-2008 at 18:45

Quote:
Originally posted by kilowatt
Quote:

No idea if it'd react with tungsten metal, but if it does, it's sure to produce HCl, sodium tungstate and sodium whatever-else-ate, plus plenty of heat.

HCl? There is no hydrogen in sodium chlorate or tungsten metal.


Eheh... I meant to say Cl2!

Might go faster with NaOH in the mix, too. Idunno, most things that make acidic oxides go fine under "acidic" fusion conditions (producing Cl2).

Tim

kilowatt - 5-3-2008 at 20:32

Thanks for the advice; I will try some H2O2 first. I probably won't be able to get much past 15% though. I have no source of 30% stuff and am not set up to distill it safely. I can concentrate 3% stuff to a degree by freezing and have done so before with some success.

Anyhow my understanding is that NaOH dissolves WO3 quite readily to form sodium tungstate. In fact as I mentioned when I was testing out the NaOH anodization process on pure tungsten rods it worked pretty darn fast with no other additives. The platinum seems to be holding it up.

Quote:

Tungsten is rather resistant to aqua regia, HNO3 plus HF will dissolve it. Fusion with NaOH + NaNO3 is another, as is passing chlorine over it at 400 C or so to get a mixture of the hexachloride and several oxychlorides.

Any idea how quickly HNO3 plus HF will work? Admittedly that is a very nasty combination, and I would have to do it in a teflon vessel, but I do have the means as a last resort. My idea of refluxing it in bromine was sort of an alternative to the idea of passing chlorine over it continually, but I'm not sure it will act very quickly at such low temperature. Using any gas would be very slow. That NaOH + NaNO3 melt you describe sounds like it would attack almost any vessel imaginable, including glass and pretty much all metals; sounds difficult. Think graphite, alumina, or graphite bonded silicon carbide would hold it?

Quote:

BTW - I'll take any worthless iridium you have off your hands, rather than just throwing it in the dustbin.

Well at only $400/troy oz instead of $2200, it is relatively worthless compared to platinum, or even gold. :P Not to mention it is much harder to work with chemically and physically. It might still be a good investment though since its price has not gone a lot up in some time, and may shoot up at any time like platinum did.

Fleaker - 6-3-2008 at 06:15

Iridium is much rarer than platinum, however, it is much less useful and much more difficult to process because of the attributes you just mentioned.

I would think nickel would hold up decently to the NaOH+NaNO3 fusion.

50% HF alone attacks tungsten rather quickly.

12AX7 - 6-3-2008 at 10:11

Nickel, iron, stainless steel and so on seem okay for holding such a reaction. Molten sodium chlorate bubbles only slowly in a steel container.

Tim

garage chemist - 6-3-2008 at 11:25

Silver crucibles are perfect for any kind of caustic melt!
They are not attacked by plain NaOH or NaOH/oxidiser mixtures even at red heat.

kilowatt - 6-3-2008 at 12:54

I don't have enough silver to make a very large crucible for the caustic melt method, would a sintered alumina one work (I have one of those)? I am certainly not going to be using a metal that is attacked whatsoever, even slowly.

If 50% HF alone will work it looks like that may be the best bet, as there is little confusion over the general composition of the products except for which oxidation state they will be in. After separation I should be able to reduce the heavy metal salts with molten aluminum. I am not sure whether HF on tungsten or platinum results in a fluoride or an acid fluoride, nor how to determine that, but I guess it doesn't really matter as they will all reduce easily. What would I even get if I used HNO3 + HF? What is the point of adding HNO3?

Edit: I just randomly happened upon a patent describing a process where a tungsten alloy anode is broken down to ammonium paratungstate and the other metals by electrolysis in ammonium nitrate and ammonia solution.

[Edited on 6-3-2008 by kilowatt]

not_important - 6-3-2008 at 16:53

Quote:
Originally posted by kilowatt
I don't have enough silver to make a very large crucible for the caustic melt method, would a sintered alumina one work (I have one of those)? I am certainly not going to be using a metal that is attacked whatsoever, even slowly.


No, it will dissolve in the flux, which is also used to break up minerals for analysis.
Quote:

If 50% HF alone will work it looks like that may be the best bet, as there is little confusion over the general composition of the products except for which oxidation state they will be in. ...(snip)... I am not sure whether HF on tungsten or platinum results in a fluoride or an acid fluoride, nor how to determine that, but I guess it doesn't really matter as they will all reduce easily. What would I even get if I used HNO3 + HF? What is the point of adding HNO3?


Massive tungsten and alloys that are mostly tungsten usually dissolve very slowly in straight aqueous HF. The references I saw that used it all were dissolving turnings, chips, or powder; the dissolution may be getting an assist from atmospheric oxygen.

The HNO3 is the oxidiser, the HF supplies fluoride ions to complex the W.
Quote:
After separation I should be able to reduce the heavy metal salts with molten aluminum.


You like to do things the difficult way, don't you? This would leave you with a aluminium-platinum alloy that you would then have to dissolve. If you dissolved only the Al, the platinum would be in the form of a fine powder saturated with hydrogen; if you read lab procedures for using platinum catalysts you'll note that finely divided platinum that has been saturated with hydrogen can ignite on exposure to air. A zinc-platinum alloy treated with hydrochloric acid leaves "explosive platinum", which on warming deflagrates or even explodes.

Most of the methods we've given will likely leave the platinum as the metal, dropping zinc, iron, or aluminum in the solution will make sure no platinum remains dissolved. If the solution is alkaline the tungsten will be in solution as tungstates which would allow the platinum to be isolated by filtration. Dissolve the platinum solids in aqua regia, fume down to remove free chlorine and NOCl, extract with 5 to 10 N HCl; any tungsten will remain undissolved as the oxide. Follow one of the standard procedures to precipitate ammonium chloroplatinate, which can be ignited to give platinum sponge.
Quote:
Edit: I just randomly happened upon a patent describing a process where a tungsten alloy anode is broken down to ammonium paratungstate and the other metals by electrolysis in ammonium nitrate and ammonia solution.


Ah - US patent 4283258 should leave the platinum as anode slime.



[Edited on 7-3-2008 by not_important]

kilowatt - 6-3-2008 at 17:14

Quote:

Most of the methods we've given will likely leave the platinum as the metal, dropping zinc, iron, or aluminum in the solution will make sure no platinum remains dissolved.

Ah yes, often times anhydrous reactions like reduction with molten metals are the first to pop into my head. It did not occur to me that an aqueous treatment could do the same, while being both easier and safer.

Quote:

The HNO3 is the oxidiser, the HF supplies fluoride ions to complex the W.

Ah, so like aqua regia, basically. But what is the resultant complex? Is it just a simple tungsten fluoride? Acid fluoride? Or an acid like some fluorometallic acid?

I'm gonna try the electrolytic process in that patent first though and see how that goes since it doesn't require any HF or even HNO3.

not_important - 7-3-2008 at 00:09

Quote:
Originally posted by kilowatt
...
Ah, so like aqua regia, basically. But what is the resultant complex? Is it just a simple tungsten fluoride? Acid fluoride? Or an acid like some fluorometallic acid?

I'm gonna try the electrolytic process in that patent first though and see how that goes since it doesn't require any HF or even HNO3.


Not very closely like aqua reia, which contains Cl2 and NOCl. In this case I think the HNO3 forms a thin oxide layer, the HF dissolves the oxide, repeat until something runs out.

Not sure on what form the tungsten is in. WOF4, WO2F3, [WO2F4]2- (all W(vi)) , [MnOF5]2-[Mn3O3F14]5- (W(v)), and others. Generally not important, fuming with H2SO4 and B(OH)3 removes the fluorine, on dilution with water tungstic acid precipitates.

The electrolytic reaction sounds interesting and is a good first test. A small amount of added metal after electrolysis should insure the platinum is left as a solid; taking that up in aqua regia and following one of the stadard workups will give the pure metal.

JohnWW - 7-3-2008 at 00:34

Quote:
Originally posted by kilowatt
I have considered refluxing the pieces in bromine, as perhaps that will dissolve away the tungsten more quickly, but I'm not sure what oxidation state it will go into. I am not sure if bromine will attack platinum at all under these conditions, and if so what oxidation state it will go into. I have also considered manually oxidizing the pieces by heating them in the air with an arc and tungsten cathode, and then reacting that with concentrated alkali, dissolving it layer by layer. The two remaining options I have considered are total dissolution in aqua regia (but I don't know anything the chemistry of tungsten in aqua regia) and anodizing in hydrofluoric acid doped with a conducting fluoride to boil off tungsten hexafluoride gas and be left with some sort of platinum fluoride (not entirely sure this would work as there are again other oxidation states).
(cut)

Heating in air would result in solid WO3, but the Pt and probably the Ir would be unchanged, although any Os would come off as volatile and toxic OsO4. Use of HF would not result in WF6, though. WF6 would be formed by F2 gas in the absence of water vapor. F2 gas, in the absence of water vapor, O2, N2, or Xe, would also form PtF6, which is so strongly oxidizing that it can abstract electrons from O2, N2, and Xe to form ionic salts of the PtF6- anion.

[Edited on 7-3-08 by JohnWW]

MxMeac - 27-8-2015 at 07:41

Hello
I am new here and new to chemistry in general. I am running into a problem that I hope fits in this forum area.
I have a high Pt alloy(90+%) with high W(5+%) that I cannot dissolve fully. My first attempt to digest the alloy was as follows: Used HCL, HNO3 and HF with very tiny pieces of the alloy. 210F for about 2.5 hrs using microwave digestion. I have green pieces remaining, 80% of the alloy is gone. However I am unable to fully digest. Any tips?

gdflp - 27-8-2015 at 11:24

Reflux the alloy with the acid mixture for longer periods of time. Platinum is extremely resistant to dissolution, even in aqua regia. Dissolution in boiling aqua regia can take days depending on the size of the pieces.