Sciencemadness Discussion Board - Arsenic from Skutterudite

On another note, is a 'thermite' of arsenic oxide possible (again, steel tube + blowtorch reagent mix As₂O₃ + Al powder, arsenic sublimes out due to volatility (SP = 615 C)?)

[Edited on 3-9-2013 by elementcollector1]

Aluminothermy of As₂O₃ would be a profoundly BAD idea. 'Thermite' mixtures routinely develop temperatures in excess of 2,000 C and higher (depending on target oxide and other factors) and with a sublimation point of only 615 C that's a recipe for arsenic flying all over the place, because such reactions are hard to control. I'm sure you understand that would be a very bad thing.

Like antimony it can probably be obtained from roasting its oxide with carbon at high temperature, the element then sublimes off.

My advice with regards to arsenic preparation? Unless you are well equipped and well versed in its chemistry, leave well alone! Far too toxic.

[Edited on 3-9-2013 by blogfast25]

Pulverulescent - 3-9-2013 at 06:38

Skutterudite ─ a highly learned mineral . . .

elementcollector1 - 3-9-2013 at 11:16

Quote: Originally posted by Pulverulescent

Skutterudite ─ a highly learned mineral . . .

I am just not getting these references today... First a robot, and now a scholar?

Anyway, what I meant was that the mixture would be in a *closed* steel tube, so that any arsenic that sublimed off would recondense on one of the caps (similar to the decomposition of arsenic sulfide mentioned in Magpie's linked thread). Unfortunately, that thread mentions that the pressure would be too great for even a quartz ampoule, so that's out of the question. In that case, I guess the way to go would be to produce arsenic using 'wet' chemistry, and then sublime it into crystals so long-term storage is safer.

A wet-chem method I keep hearing about is SnCl₂, HCl and arsenic chloride. However, I'm confused, as arsenic trichloride reacts with water to form arsenous acid (As(OH)₃) and chlorine. Maybe this is what the extra HCl is for?

woelen - 3-9-2013 at 22:34

AsCl3 does not give chlorine with water, but it gives HCl and arsenious acid. This is an equilibrium reaction and in very concentrated HCl the reaction is more on the side of AsCl3. That is the reason of adding extra HCl. Without that, nearly all AsCl3 would hydrolyse.

If SnCl2 is used, then another reason for using lots of HCl is to keep this and its oxidation product dissolved. SnCl2 is notoriously sensitive to hydrolysis (formation of HCl and Sn(OH)2) and solutions in water always are somewhat turbid. The oxidation product, containg tin in oxidation state +4 is even more sensitive to hydrolysis.

elementcollector1 - 4-9-2013 at 08:17

If I dissolved skutterudite in HCl, would a product be arsine? It seems likely, as the same reaction occurs for sulfides (producing hydrogen sulfide). If so, would a way to avoid this be passing dry Cl₂ over the powdered mineral?

Pok - 4-9-2013 at 10:04

Yes! I think arsine would be evolved! You shouldn't try this!

Why don't you simply buy arsenic? It can be found in mineral stores for very low prices. You can easily make very pure arsenic from this just by subliming it in a test tube. But in most cases the "native arsenic" that you can buy is pure enough for an element collection. I heated a small piece in a test tube and a mirror of arsenic was produced with crystals of arsenic (in absence of air and in a fume hood!).

Bad photo quality, looks better in reality.

This is by far the easiest and safest way to get some arsenic (no toxic wastes, no difficult/dangerous procedure)

BTW: arsenic trioxide can also be made from arsenic and it shows a remarkable property: crystalloluminescence , the strongest known. And also very strong white triboluminescence (same link)! (translation with google if required)

[Edited on 4-9-2013 by Pok]

12AX7 - 4-9-2013 at 11:35

Fascinating; As dissolves in sulfuric acid, reducing some of the acid to SO2? Looks like a great yield (of As2O3 crystals). Cool!

As I recall, arsenic can be reduced from solution, similar to Se and Te? Or is it just those...

I would also suppose, once you isolate As2O3, a ground-up mixture with charcoal should distill (well, sublimate) elemental As under similar conditions to the picture above, and accordingly, in high purity. Probably under inert conditions (CO2, N2 or etc.), but not vacuum. The active ingredient in carbon reduction is often CO; the same reaction might be done under a stream of CO, with more expensive equipment of course (CO generator, reaction tube, connectors). The reduction could also be done with hydrogen, with the obvious danger of AsH3, but that decomposes over 200C so it should be a fine method as well. I think you'd want to stick with nonmetals and gasses for reduction, as metallothermic reduction will produce arsenides as impurities (compare to Mg + SiO2 thermite making silicide), besides the drawback of the low boiling point.

That said, metallothermic reduction in a bomb might be interesting. Like the sodium reaction, it should produce a lump of fused arsenic, which would be an unusual sample for the collector.

Tim

watson.fawkes - 4-9-2013 at 13:00

Quote: Originally posted by 12AX7

Probably under inert conditions (CO2, N2 or etc.), but not vacuum. The active ingredient in carbon reduction is often CO; the same reaction might be done under a stream of CO, with more expensive equipment of course (CO generator, reaction tube, connectors).

You want a reduction with CO under CO2 cover with extra N2? That's what you get from a forced air fire with insufficient oxygen, in other words, a reducing flame. I'll admit the reaction vessel etc. is indeed extra gear, but the CO generator isn't difficult at all.

bfesser - 4-9-2013 at 13:42

Quote: Originally posted by Pok

BTW: arsenic trioxide can also be made from arsenic and it shows a remarkable property: crystalloluminescence , the strongest known. And also very strong white triboluminescence (same link)! (translation with google if required)

Interesting. I had never heard of <a href="https://en.wikipedia.org/wiki/Crystalloluminescence" target="_blank">crystalloluminescence</a> <img src="../scipics/_wiki.png" /> before now. Thanks for sharing that link.

elementcollector1 - 4-9-2013 at 15:15

Quote: Originally posted by Pok

Yes! I think arsine would be evolved! You shouldn't try this!

Why don't you simply buy arsenic? It can be found in mineral stores for very low prices. You can easily make very pure arsenic from this just by subliming it in a test tube. But in most cases the "native arsenic" that you can buy is pure enough for an element collection. I heated a small piece in a test tube and a mirror of arsenic was produced with crystals of arsenic (in absence of air and in a fume hood!).

Bad photo quality, looks better in reality.

[Edited on 4-9-2013 by Pok]

That should belong in Pretty Pictures...

I've yet to find native arsenic, try though I may... You're right that this is dangerous, but my budget is very low.

What about burning the powdered skutterudite in a closed area - this should produce Fe₂O₃, NiO, and CoO as well as sublimed As₂O₃, correct? This would avoid the hazardous arsine, and if the container were well sealed, be almost safe.

[Edited on 4-9-2013 by elementcollector1]

bfesser - 4-9-2013 at 15:56

A few years ago, I picked up a sample of skutterudite ((Co,Ni,Fe)As₃).

A sample of powdered ore or a mineral specimen?

elementcollector1 - 4-9-2013 at 16:16

A few years ago, I picked up a sample of skutterudite ((Co,Ni,Fe)As₃).

A sample of powdered ore or a mineral specimen?

Mineral specimen - about a 2"- by- 3"- by- 2" lump of metallic substance.

blogfast25 - 5-9-2013 at 06:04

EC1:

It could be worth grinding a pinch of it down and trying to sublime it in a test tube. The As might just sublime off, who knows? It definitely sounds more like an alloy than a compound...

elementcollector1 - 5-9-2013 at 08:06

EC1:

It could be worth grinding a pinch of it down and trying to sublime it in a test tube. The As might just sublime off, who knows? It definitely sounds more like an alloy than a compound...

According to the Wiki, it has a definite structure and arsenides are well known. But I'll give it a shot, see what I come up with. The test tube will be sealed first, obviously - I'll crack it open after the experiment.

12AX7 - 5-9-2013 at 10:46

I don't have phase diagrams for most arsenides, but the phosphides and antimonides that I do have seem to suggest somewhere between compound and intermetallic behavior with the Fe-Ni-Co crowd. I think they'll be reasonably stable. At the very least, there will be high (solid or liquid state) affinity or solubility of As, so the partial pressure will be small. It is certainly possible to liberate P, As or Sb from any of these compounds with enough heating, vacuum and time, but that doesn't mean you want to bother.

If, instead of only heating the arsenides to drive off arsenic, they were slowly oxidized (let some oxygen leak into the vacuum?), it would push the equilibrium significantly. The result would probably be a combination of metal oxides and As (otherwise, As2O3 and metals). Of course, slow oxidation of the whole thing would give separation of the oxides. (You'd also want to check on the stability of the arsenites and arsenates, but it's likely these decompose with heating, just as the sulfates do.)

Tim

elementcollector1 - 5-9-2013 at 15:50

As2O3 boils at just 465 degrees C, so this could well separate the oxide from the rest. What to do with the oxide, however, is another thing entirely.

bfesser - 5-9-2013 at 17:29

Has anyone bothered to do a lit. search on possible use of this mineral as an ore?

elementcollector1 - 6-9-2013 at 07:11

Has anyone bothered to do a lit. search on possible use of this mineral as an ore?

All I found from a cursory search was that it finds a use as a cobalt and sometimes-nickel ore, with arsenic as the byproduct. I did find a source for purification by sublimation of arsenic:

Quote:

Wikipedia:
On roasting in air of arsenopyrite, arsenic sublimes as arsenic(III) oxide leaving iron oxides,[18] while roasting without air results in the production of metallic arsenic. Further purification from sulfur and other chalcogens is achieved by sublimation in vacuum or in a hydrogen atmosphere or by distillation from molten lead-arsenic mixture

[Edited on 6-9-2013 by elementcollector1]

elementcollector1 - 20-12-2013 at 21:42

Quote: Originally posted by 12AX7

Fascinating; As dissolves in sulfuric acid, reducing some of the acid to SO2? Looks like a great yield (of As2O3 crystals). Cool!

As I recall, arsenic can be reduced from solution, similar to Se and Te? Or is it just those...

I would also suppose, once you isolate As2O3, a ground-up mixture with charcoal should distill (well, sublimate) elemental As under similar conditions to the picture above, and accordingly, in high purity. Probably under inert conditions (CO2, N2 or etc.), but not vacuum. The active ingredient in carbon reduction is often CO; the same reaction might be done under a stream of CO, with more expensive equipment of course (CO generator, reaction tube, connectors). The reduction could also be done with hydrogen, with the obvious danger of AsH3, but that decomposes over 200C so it should be a fine method as well. I think you'd want to stick with nonmetals and gasses for reduction, as metallothermic reduction will produce arsenides as impurities (compare to Mg + SiO2 thermite making silicide), besides the drawback of the low boiling point.

That said, metallothermic reduction in a bomb might be interesting. Like the sodium reaction, it should produce a lump of fused arsenic, which would be an unusual sample for the collector.

Tim

Bringing this back because I missed this post.
Does this apply to arsenides? I feel like it wouldn't, it would just make arsine due to the H⁺ ions and the negative arsenide ions.

In that case, is there any wet way to produce arsenic oxide from an arsenide? I think it might be possible to roast the mineral with molten sodium hydroxide to form sodium arsenate, and acidify this to precipitate arsenic (hydr)oxide. This would not affect the cobalt, iron or nickel to my knowledge.

I still have yet to try sublimating this material in a closed test tube (nerves more than anything), or do any wet chemistry (due to that arsine hazard).

It appears at this point that isolating As₂O₃ will be the tricky part of this. The reduction with HCl and SnCl₂ seems fairly straightforward, but I have one question for this step: Presuming I filter and wash the arsenic precipitate, as well as all arsenic-containing wastes from earlier, how do I dispose of them safely? As far as I know, there is no 'safe' form of arsenic.

blogfast25 - 21-12-2013 at 06:58

Safest way is probably to recover all As as some solid, insoluble compound. Then mix with cement and after hardening dispose at your local tip as 'non-recyclable'.

elementcollector1 - 21-12-2013 at 13:18

Sounds good. What about all the filters and such that will be undoubtedly contaminated?
Found that when precipitated arsenic oxide is mixed with sodium hydroxide in solution and boiled, it goes back into solution as sodium arsenite. This would then be filtered off from the precipitated metal hydroxides (mostly cobalt), since none are amphoteric in this case. The solution would then be acidified, and this would precipitate arsenic trioxide again, correct? If so, this is a convenient and simple wet-chemistry route to arsenic trioxide.

However, the question remains as to precipitating arsenic trioxide in the first place, while avoiding arsine. Acids cannot be used, as arsenides react with any strong acid to give arsine. Bases might be possible due to the arsenite double salt. In this case, what would be the products of the reaction?

Additional note: Formula of skutterudite is actually CoAs₃, with occasional substitutions of Ni and Fe.

??CoAs₃ + ??NaOH -> ??Na₃AsO₃ + ???(Co)

[Edited on 12-21-2013 by elementcollector1]

blogfast25 - 21-12-2013 at 14:09

Just include any kind of solid waste too.

It may not be ideal but it's better than to get the 'smurf suited people' in at high cost and with suspicion falling on yourself ('what's a non chemist private person messing about with a mega poison for like this?') and the search for bodies to start!

plante1999 - 21-12-2013 at 14:11

Gas all the equipment with H2S, and filter paper, washes etc is put into the ground. After all, realgar and orpiment are sulphides...

elementcollector1 - 21-12-2013 at 17:04

Makes sense. What about the reaction products from the leach with NaOH? I can't find any balanced equation for either sodium arsenite (NaAsO₂ or Na₃AsO₃). I would assume the products would be the arsenite, some oxide or hydroxide of cobalt, and possibly water or hydrogen.

EDIT: I have found one balanceable reaction.
11 O₂ + 36 NaOH + 4 CoAs₃ -> 12 Na₃AsO₃ + 18 H₂O + 4 CoO

Does this seem at all feasible? It would take some time for the oxygen in the air to react with the fused mixture of the ore and the sodium hydroxide, and I wouldn't be able to tell when it was done (except that the ore would have turned into a black powder?)

[Edited on 12-22-2013 by elementcollector1]

woelen - 22-12-2013 at 05:19

I have tried the reaction between As2O3 (dissolved in excess HCl) with SnCl2. This reaction is fairly facile, but it is not clean. You get very impure arsenic (dirty brown stuff). I also tried making As with NaH2PO2 instead of SnCl2, but this is not working very well. You get very pure fine crystals of As, but the yield is pathetic. From 1.5 grams of As2O3 I only get 100 mg or so of arsenic after hours and hours of heating, it simply won't react easily.

I did not yet find an aqueous method of making As from its compounds (I have some As2O3 and I have some orpiment from eBay sellers). All well-known methods require heating and deposition of vapors. This is something I certainly will not do at home, I do not want to poison myself with As-vapors.

blogfast25 - 22-12-2013 at 06:17

There's an old forensic test which consists of plating out the As on copper wire, from an AsCl3 solution. See Reinsch test:

http://en.wikipedia.org/wiki/Reinsch_test

[Edited on 22-12-2013 by blogfast25]

elementcollector1 - 22-12-2013 at 10:13

There's an old forensic test which consists of plating out the As on copper wire, from an AsCl3 solution. See Reinsch test:

http://en.wikipedia.org/wiki/Reinsch_test

[Edited on 22-12-2013 by blogfast25]

In the much higher concentrations expected of this experiment, do you think it would still be a 'plating' or more akin to the copper 'tree' from CuSO₄ and Fe?

@woelen: I don't mind pathetic yields in this case, as I only have to make a pure, visible sample - even 100mg is acceptable (I ampouled this amount of palladium, and it looked fine). I agree that I'd really rather not handle any vaporous compound of arsenic, hence the query above about leaching with molten NaOH and air.

blogfast25 - 22-12-2013 at 14:16

In the much higher concentrations expected of this experiment, do you think it would still be a 'plating' or more akin to the copper 'tree' from CuSO₄ and Fe?

I've no idea, it's conditions dependent, I would think...

elementcollector1 - 22-12-2013 at 15:42

Hmm. Something to try, but I should probably go with the hypophosphite method (which will unfortunately involve distilling white phosphorus and reacting with NaOH, due to lack of availability).

EDIT: Found in an earlier thread that apparently tartaric acid can take care of the tin contamination, but I need more clarification on this:

Quote:

Other people seem to be happy with reduction with As+3. Here they use warm very conc. HCl with their SnCl2, using tartaric acid to produce a precipitate free of tin. The actual analysis using iodine titration seems typical.
http://books.google.com/books?id=40QPAAAAIAAJ&pg=PA316
But a couple different authors say that As+5 is not so fast to be reduced.
Really there are very many articles over the years using either hypophosphite or SnCl2 in some way for quantitative analysis of arsenic, but this does not necessarily apply here.

Quote is from S.C.Wack, in this thread.

[Edited on 12-23-2013 by elementcollector1]

elementcollector1 - 23-12-2013 at 10:36

Found what I think is the reaction mechanisms for the full process I intend to follow, based on balanceable reactions and all products being water-soluble except for As at the end.

First, CoAs₃ is powdered and fused with NaOH and air for a long period of time:
4 CoAs₃ + 36 NaOH + 11 O₂ -> 12 Na₃AsO₃ + 18 H₂O + 4CoO

The sodium arsenite is dissolved in water, and the insoluble cobalt oxide is filtered off.

In solution, the sodium arsenite is acidified. This precipitates As₂O₃.
12 Na₃AsO₃ + 36 HCl -> 6 As₂O₃ + 36 NaCl + 18 H₂O

The supernatant solution is decanted, and the arsenic oxide is dried. It is then dissolved in an excess of fresh HCl.
6 As₂O₃ + 36 HCl -> 12 AsCl₃ + 18 H₂O

Meanwhile, sodium hypophosphite is reacted with hot, concentrated HCl.
NaH₂PO₂ + HCl -> NaCl + H₃PO₂

The phosphinic acid is added to the arsenic trichloride, and with much heating, some small amount of arsenic precipitates.
12 AsCl₃ + 12 H₃PO₂ -> 12 As + 12 POCl₃ + 12 H₂O + 6 H₂

The POCl₃ immediately reacts with the water.
POCl₃ + 3 H₂O -> 3 HCl + H₃PO₄

There are a few caveats to this reaction pathway:
-As must not be reactive with HCl. I cannot find a reference that says it does react, nor can I find one that says it doesn't.
-As must also not be reactive with phosphoric acid.
-H₂ must be released - but no one has commented on any gas evolution?

For 100g of CoAs₃ at the beginning, the theoretical yield of pure arsenic is 80g. Given woelen's comment on an earlier arsenic thread about a yield of less than 7% using the hypophosphite method, the expected yield is about 5.6g As, which is almost exactly enough to fill a 1mL ampoule.

blogfast25 - 24-12-2013 at 05:18

-As must not be reactive with HCl. I cannot find a reference that says it does react, nor can I find one that says it doesn't.
-As must also not be reactive with phosphoric acid.
-H₂ must be released - but no one has commented on any gas evolution?

As won't react with acids to any appreciable degree, I think. The element is still too non-metallic to react directly with non-oxidising acids. It probably responds to HNO3.

That Reinsch test is also carried out in acid conditions.

[Edited on 24-12-2013 by blogfast25]

elementcollector1 - 24-12-2013 at 09:04

Ah. That works out perfectly then.

AJKOER - 24-12-2013 at 17:47

Hee is a comment from Atomistry.com on isolating Arsenic (link: http://arsenic.atomistry.com/production.html ):

"Metallic arsenic is sometimes obtained from arsenious oxide. Thus in Altenberg (Silesia) the oxide is heated at 650° to 700° C. with charcoal in an earthenware crucible covered with a conical iron cap which acts as receiver, while in Chicago retorts composed of steel pipes large enough to take a charge of 450 lbs. were at one time used for the same purpose, the arsenic being collected in water-cooled pipe condensers. This reduction method is not satisfactory, however, as the product is largely amorphous and is not so desirable as the crystalline form, being suitable only for making arsenic compounds. Moreover it always contains arsenious oxide. A Japanese method consists in fixing the vapour of arsenious oxide by means of ferric oxide or alumina at a temperature above 218° C. and then reducing the product with water-gas, Mond gas or producer gas above 100° C.; the arsenic thus freed is then sublimed. "

The Japanese method sounds interestingly.

elementcollector1 - 25-12-2013 at 09:12

It sounds interesting, but high-temperature reduction is not something I'd like to try if it can be avoided.

elementcollector1 - 28-12-2013 at 11:15

Attempting the first step (ore and molten sodium hydroxide). Pics forthcoming.

BobD1001 - 28-12-2013 at 12:01

Attempting the first step (ore and molten sodium hydroxide). Pics forthcoming.

Looking forward to the results!

elementcollector1 - 28-12-2013 at 12:24

Aqueous analysis is not started yet, but here are some photos of the reaction!
First, a pic of the mineral itself. It weighs 120g, and looks extremely metallic in nature. For something that behaves and looks like a metal or alloy, this stuff is remarkably fragile - likely due to the high arsenic content? This was from one hit with a hammer.

Next, the mix: 2.5g of skutterudite and 3.5g of NaOH (adjusted upward from stoichiometry due to hydration).

The setup (the crucible on a hot plate, turned to max):

During the reaction, noticeable gas evolution was observed. According to the reaction schema I stated in a previous post, this should be H₂O boiling off, but could also be As₂O₃ sublimating - this is discussed farther down. Additionally, significant amounts of skutterudite were still present after the hot plate was turned off - with more heating, these may have disappeared.

And finally, after pretty much one hour of heating, the post-reaction mix (some more NaOH was added because I thought it was somehow boiling to dryness):

I think that brown stuff is CoO? I'm not sure.

A new worry I thought of is that the produced Na₃AsO₃ might have just decomposed, and the As₂O₃ sublimated away. Well, I guess we'll see if there's any soluble arsenic in the aqueous extraction. If not, there's really no point to the molten procedure - it would be the same in terms of what happens to the arsenic when the ore is heated in air.
In that case, would an extraction with concentrated aqueous NaOH and somewhat (~15%) concentrated H₂O₂ do the trick?

Random - 28-12-2013 at 13:07

If As2O3 is sublimating away it can't be a good thing. Try to catch at least some of this stuff by having something chilled on top of the crucible.

blogfast25 - 28-12-2013 at 13:42

EC1:

Was the ore powdered prior to fusion?

Usually these fusions prescribe twice the weight of alkali to ore, sometimes more.

Using finely ground powders, I've 'cracked' materials like Beryl, Zircon and food grade TiO2 in a matter of about 20 minutes, by fusion with KOH. Once steam evolution is over, the reaction is essentially over, heating further won't do much if anything.

I must warn you again of the dangers of arsenic, especially as you now seem not to know what's become of it!

Arsenic oxide's BP is 465 C, that sounds a little high for a hot plate.

[Edited on 28-12-2013 by blogfast25]

elementcollector1 - 28-12-2013 at 15:00

EC1:

Was the ore powdered prior to fusion?

Usually these fusions prescribe twice the weight of alkali to ore, sometimes more.

Using finely ground powders, I've 'cracked' materials like Beryl, Zircon and food grade TiO2 in a matter of about 20 minutes, by fusion with KOH. Once steam evolution is over, the reaction is essentially over, heating further won't do much if anything.

I must warn you again of the dangers of arsenic, especially as you now seem not to know what's become of it!

Arsenic oxide's BP is 465 C, that sounds a little high for a hot plate.

[Edited on 28-12-2013 by blogfast25]

The ore was in the form of 3-5mm irregular granules (which are now somehow welded onto the crucible...). Not the best for fusion, but my original intent was to see if I could measure an appreciable loss of mass, and given the expected CoO powder, granules would be better separated from the end mixture. Judging from the granules present, there was no appreciable loss of mass - not even half a gram. This indicates either a longer melt was needed, or the reaction does not proceed under these conditions. Noticeably, the ore granules were significantly darkened afterwards, indicating surface oxidation to black CoO.

Steam evolution did seem to occur, though I was unsure what gas was being produced at the time. But considering that the heating element was red-hot (faintest of glows in daylight, much stronger in shadow) and this thing barely reaches 200 C with a sandbath, I doubt that it can sublimate arsenic under any conditions.

Well, at least I know what's become of most of it. It's probably worth testing the aqueous extraction for any arsenic, but any oxide precipitation will be on the order of milligrams at most, meaning the solution needs to be heavily concentrated.

The thing is, in your cases, the metal/element in question was a cation. Here, arsenic is an anion - so how does that play differently?

EDIT: I took a look at the stoichiometry, and according to the reaction, if it were to proceed to completion, about 0.88 liters of steam should have been driven off (about 0.7 mL of liquid). There was nowhere near this much steam, and given that some reaction clearly occurred (the brown powder, blackening of the ore particles, some gas evolution) I would say that the reaction is indeed happening. Just very slowly, and very inefficiently due to the particle size. Perhaps another test with very fine powder is in order...

EDIT.2: The pertinent reaction for an aqueous extraction would be
11 H₂O₂ + 18 NaOH + 2 CoAs₃ -> 20 H₂O + 2 CoO + 6 Na₃AsO₃
I'm imagining that this might proceed, but the molten process would be more efficient.

[Edited on 12-29-2013 by elementcollector1]

blogfast25 - 30-12-2013 at 10:36

The thing is, in your cases, the metal/element in question was a cation. Here, arsenic is an anion - so how does that play differently?

Firstly I'm somewhat surprised your ore didn't appreciably react with molten NaOH. Pure arsenic should do that quite vigorously, I should think.

But the arsenic isn't necessarily present as an anion: the ore appears to be an alloy or intermetallic compound (even if As doesn't really qualify as a metal). Have a look at its structure:

http://en.wikipedia.org/wiki/Skutterudite

I wonder if this stuff reacts appreciably with hot, conc. sulphuric acid, which might get a handle on the cobalt.

Aqua Regia (even the 'poor man's version') might also dissolve it through the combined oxidising action of the nitrosyl chloride and the solvating action of the hydrogen chloride (AsCl4- complex anions exist).

Re. hot plates, I think many household types are calibrated for about 200 C surface temperature at high setting, which is more than enough for basic cooking requirements. Yours obviously ran hotter because it melted NaOH.

[Edited on 30-12-2013 by blogfast25]

elementcollector1 - 31-12-2013 at 11:57

I did wonder about the structure.
I heard elemental arsenic reacts with concentrated sulfuric acid to make arsenic oxide and SO₂. If this occurs even for skutterudite, this would be very convenient!
In that case, another test with sulfuric acid is in order.

plante1999 - 31-12-2013 at 14:28

Nitric acid and sulphuric acid should in theory be able to attack the powdered ore.

elementcollector1 - 31-12-2013 at 19:07

Wait... wouldn't this produce arsine?

plante1999 - 31-12-2013 at 19:40

The oxidizing condition should oxidize it if formed at all. Arsine is a very strong reducing agent.

bfesser - 31-12-2013 at 19:46

Numbers, please.

elementcollector1 - 31-12-2013 at 21:30

Numbers, please.

For the reaction?

EDIT: Picture of the skutterudite powder:

Stuff got a lot darker when in powder form - as a lump it was very bright and shiny, like polished steel. I should have about 120g here.

[Edited on 1-1-2014 by elementcollector1]

blogfast25 - 1-1-2014 at 05:30

Like plante I'm convinced no arsine will evolve. But as a precaution it is easily destroyed (look at the Marsh test, for instance), if it did evolve.

bfesser - 1-1-2014 at 08:23

No, I meant redox potentials or other data instead of this useless should–shouldn't back and forth.

blogfast25 - 1-1-2014 at 12:41

Redox potentials for reactions in concentrated acids?

On Arsine Wiki mentions that it "will react violently in presence of strong oxidizing agents, such as potassium permanganate, sodium hypochlorite or nitric acid". No chance of free arsine with strong acids.

Interestingly, the Wiki entry on BiH3 mentions:

"The methodology used for detection of arsenic ("Marsh test") can also be used to detect BiH3. This test relies on the thermal decomposition of these trihydrides to the metallic mirrors of metallic As, Sb, and Bi. These deposits can be further distinguished by their distinctive solubility characteristics: As dissolves in NaOCl [my emph.], [...]"

So this CoAs3 may well dissolve in hypochlorite, presumably to arsenite.

[Edited on 1-1-2014 by blogfast25]

elementcollector1 - 1-1-2014 at 12:54

And what would it leave behind? CoO? Co(OH)₂? I would presume the cobalt would not go into solution.

blogfast25 - 1-1-2014 at 13:03

And what would it leave behind? CoO? Co(OH)₂? I would presume the cobalt would not go into solution.

Co(II) hydroxide, possibly (depending on oxidation potentials) a III or IV hydroxide. It would not dissolve.

[Edited on 1-1-2014 by blogfast25]

elementcollector1 - 1-1-2014 at 15:19

Put one gram of ore in a test tube with about 15 mL of bleach. So far, no results. I'm guessing this will take a while, but if I attempt to heat the mixture, the NaOCl will immediately disproportionate - which might ruin the dissolution entirely.

plante1999 - 1-1-2014 at 15:56

I would go for the strong inorganic acid.

elementcollector1 - 1-1-2014 at 16:29

Update, 1 hour later: Slow bubbling is observed. There is some sort of darker layer on top of the ore - this is likely CoO. So, I think this reaction is occuring - just very slowly.

bfesser - 1-1-2014 at 16:46

Any chance you could take some photos of the process?

elementcollector1 - 1-1-2014 at 17:10

Let it be known that I tried. The bubbling is almost invisible, and the bottom layer is only visible with flash on.

elementcollector1 - 1-1-2014 at 19:44

Update, 3 hours later: The black layer mentioned previously has grown larger, and more noticeable. This substance is noticeably different from the skutterudite. As you can see from the photo above, the ore sample is bright and silvery. This stuff is a black, flocculent powder that very easily forms a suspension, leading me to suspect Co(OH)₂. The same slow rate of bubbling is observed.

DraconicAcid - 1-1-2014 at 19:56

I suspect cobalt arsenate would also be insoluble.

elementcollector1 - 1-1-2014 at 20:01

Yeah, but apparently that's sort of a pinkish-purple (see here: http://en.wikipedia.org/wiki/Erythrite), so I feel pretty safe about discounting that.

blogfast25 - 2-1-2014 at 06:04

That cobalt arsenate [Erythrite] is an octahydrate so it must be fairly soluble. It's described also as 'cobalt bloom', which also suggests solubility. Pinkish/purplish/red is of course the colour of the Co2+ hydrated ion, see CoCl2, for instance.

But we're not sure whether As(III) or As(IV) is being formed, must check the potentials for that...

And we're in alkaline conditions, which would precipitate the cobalt as a hydroxide.

[Edited on 2-1-2014 by blogfast25]

blogfast25 - 2-1-2014 at 10:53

A few relevant reduction potentials (CRC Ed. 86 values):

Hypochlorite to chloride = + 0.81 V
HAsO2 to As(0) = + 0.248 V
H3AsO4 to HAsO2 = + 0.56 V
Co2+ to Co(0) = - 0.28 V
Co3+ to Co2+ = + 1.92 V

This would indicate that hypochlorite can oxidise As all the way up to H3AsO4 and Co to Co2+ (but not Co3+

.

If so, with an excess of alkaline sodium hypochlorite, sodium orthoarsenate (Na3AsO4) would be formed and Co(OH)2 would precipitate. Filtering would then allow separation of the Co and the arsenic.

According to my Holleman, solid H3AsO4.H2O can be obtained by concentrating an acidic solution of this acid and cooling to below 15 C. Here I would suggest acidifying the sodium arsenate solution with H2SO4 (sodium sulphate being very soluble), boiling to further destroy any excess hypochlorite and to concentrate the solution. Chilling should see ortho arsenic acid hydrate crystallise, more or less free from Na, sulphate and chloride.

[Edited on 2-1-2014 by blogfast25]

DraconicAcid - 2-1-2014 at 12:46

A few relevant reduction potentials (CRC Ed. 86 values):

Hypochlorite to chloride = + 0.81 V
HAsO2 to As(0) = + 0.248 V
H3AsO4 to HAsO2 = + 0.56 V
Co2+ to Co(0) = - 0.28 V
Co3+ to Co2+ = + 1.92 V

This would indicate that hypochlorite can oxidise As all the way up to H3AsO4 and Co to Co2+ (but not Co3+

I don't have my CRC handy, but what's the reduction potential for Co₂O₃ to Co(OH)₂? I suspect the oxidation of cobalt will be more favourable in basic solution.

elementcollector1 - 2-1-2014 at 12:52

Judging from this info, would this be a suitable reaction?
3 Co + 2 As + 8 NaOCl + 6 NaOH -> 8 NaCl + 2 Na₃AsO₄ + 3 Co(OH)₂

From this and a general formula of CoAs₃ for skutterudite, this suggests that 7 As get 'ignored' by this reaction for every 3 Co. If I try with the formula for skutterudite, the reaction is not balanceable (presumably because it's assumed to be a compound).

Update on the test tube: The ore is now all darkened and intermixed with the brown-black powder. I'm going to add some more bleach, fill the test tube up (should be about 30 mL, or 1.94g NaOCl).

blogfast25 - 2-1-2014 at 13:24

DA: I'll have a look tomorrow.

EC1: for an overall reaction equation with the ore, just write one for 1 Co (eq. 1), then one for 1 As (eq. 2). Then combine for CoAs3, as 1 x (eq. 1) + 3 x (eq.2) = overall reaction equation (theor. stoichiometry). What you've got there seems correct, so just add one of eq. 2 and you're done.

It's rather a lot of NaCl, which may make obtaining the pure H3AsO4.H2O harder.

[Edited on 2-1-2014 by blogfast25]

DraconicAcid - 2-1-2014 at 13:33

I get:

2 CoAs₃ + 12 OH^- + 11 OCl^- --> 2 Co(OH)₂ + 6 HAsO₃^2- + H₂O + 11 Cl^-.

blogfast25 - 2-1-2014 at 14:05

DA:

Balances in everything, including charge, so it must be fine. Just add the Na+ to get rid of the charges.

With commercial bleach being so weak you'll have a lot of water and a lot of sodium chloride!

I'll see if CRC make any mention of the solubility values of ortho arsenic acid.

:cool:

Another approach could be fusion of the ore with KOH + KClO3 or KNO3. These oxidisers may also get As to ortho arsenate (or maybe arsenite (III)?) but with the advantage of temperature and much higher concentrations of leachates.

[Edited on 2-1-2014 by blogfast25]

elementcollector1 - 2-1-2014 at 15:27

For that matter, I could add NaOH and heat in a water bath: The bleach would disproportionate into chlorate, and the reaction would be faster.
Update on the test tube, 24 hours in: In addition to the previous update, about 1mm of dark brown powder has settled. Bubbling rate seems to have slightly increased? It's so small I can't properly tell, but the bubbles do look bigger.

EDIT: I can't believe this didn't occur to me earlier, but... what is the bubbling? None of the reactions outlined mention any sort of gas, and it obviously can't be steam, 24 hours in and room temperature. Is it air? I did shake this thing a few times to ensure mixing, but I usually leave it for a few hours and it still bubbles when I get back...

[Edited on 1-2-2014 by elementcollector1]

plante1999 - 2-1-2014 at 15:38

Cobalt catalyse hypochlorite decomposition to oxygen and chloride.

For bfesser:

Reference from my own work.

elementcollector1 - 2-1-2014 at 15:48

Oh. Hmm. So, this reaction will actually lose potency over time.
That is quite the problem - I might have to try this again with some different reagents. I have some sodium nitrate I can experiment with.

blogfast25 - 3-1-2014 at 06:02

Try fusing with NaNO3 (MP = 308 C).

For overall reaction equation, assume 4 NaNO3 === > 2 Na2O + 4 NO + 3 O2

and use the O2 for Co + 1/2 O2 === > CoO

and 6 As + 9 Na2O + 5 O2 === > 6 Na3AsO4

For balancing, you may be a bit short of Na: supplement with NaOH.

[Edited on 3-1-2014 by blogfast25]

As a starting point I would use 1 mole ore (= 3 moles As) + 5 moles NaNO3 + 4 moles NaOH, perhaps adding 10 - 20 % excess NaNO3.

[Edited on 3-1-2014 by blogfast25]

BobD1001 - 6-1-2014 at 00:01

Ive been following this thread quite closely, and actually just purchased a rather large and beautiful Skutterudite specimen. I was a bit concerned however about the handling of the Skutterudite itself, how dangerous is a large 'rock' of this stuff? I have been handling it only with gloves just in case, but I haven't been able to find much of anything regarding its safety online, just a few vague references such as this: http://csmsgeologypost.blogspot.com/2013/04/skutterudite-cobalt-nickel-arsenide.html, stating "It is interesting to note that the local population (Berbers) knew about the toxicity of the outcropping arsenates long before the onset of commercial mining—they used it for insect control and rat poison.".

blogfast25 - 6-1-2014 at 06:07

I would read up on the dangers of As and As2O3 and would keep the rock under lock and key (display cabinet), so no humans or other animals can get their paws on it.

[Edited on 6-1-2014 by blogfast25]

elementcollector1 - 6-1-2014 at 12:00

I've gotten powder of the stuff on my fingers due to carelessness, and experienced no symptoms. Still, with arsenic, you can never be too careful.

BobD1001 - 6-1-2014 at 13:48

I've gotten powder of the stuff on my fingers due to carelessness, and experienced no symptoms. Still, with arsenic, you can never be too careful.

I must say, I would probably be defecating myself after such an experience

I would think however, that if Skutterudite was largely toxic there would be a considerable amount more literature to found online regarding such toxicity and relating concerns. Perhaps the minerals crystal structure itself helps to reduce the Arsenic toxicity by 'isolating' it within its crystal structure.

Wikipedia states "The unit cell can be considered to consist of eight smaller cubes made up of the Co atoms. Six of these cubes are filled with (almost) square planar rings of As, each of which is oriented parallel to one of the unit cell edges.... The As atoms then form octahedra with Co in the centre." So perhaps the outside structure of Co, Fe, or Ni helps isolate the As. I added some 'commentary' to the Wikipedia picture, as I think it may help to explain the structure of Skutterudite somewhat better, see below.

Please note that this diagram is just my interpretation of the above statement, and please correct me if it is wrong, as I am not a mineralogist by any means.

DraconicAcid - 6-1-2014 at 13:53

I'm sure its low solubility keeps its toxicity low (although I wouldn't want to leave it sitting on my hands for any length of time).

bfesser - 6-1-2014 at 14:10

Quote: Originally posted by BobD1001

I would think however, that if Skutterudite was largely toxic there would be a considerable amount more literature to found online regarding such toxicity and relating concerns.

There is a huge amount of literature online regarding this toxicity, but most is not specific to just one arsenic-bearing mineral. See: <a href="https://en.wikipedia.org/wiki/Arsenic_contamination_of_groundwater" target="_blank">Arsenic contamination of groundwater</a> <img src="../scipics/_wiki.png" />

blogfast25 - 6-1-2014 at 14:28

I've gotten powder of the stuff on my fingers due to carelessness, and experienced no symptoms. Still, with arsenic, you can never be too careful.

Wow, EC1, do wear latex gloves at a minimum!

BobD1001 - 6-1-2014 at 15:24

Bfesser, I fail to see where in this link it describes the toxicity of Skutterudite. Of course Arsenic's toxic properties are hugely documented, however, I was specifically referring to this mineral, for which there appears to be a lack of toxicological information.

bfesser - 6-1-2014 at 18:04

It doesn't describe the toxicity of skutterudite. As I said, most of the published information is not specific to a particular arsenic-bearing mineral, because in the environment, it rarely matters much which mineral species provide the source for the groundwater contamination, only that they do. Why do you require toxicological information for one particular mineral species? For most minerals, it doesn't make sense to gather such data. Also, a mineral species can contain various impurities and can vary slightly in composition between localities and even specimens.

elementcollector1 - 20-2-2014 at 20:44

Update: Took another gram or so and put it in a test tube, filling the test tube up with sulfuric acid and a little water. So far, it's exhibiting the exact same reaction as with the bleach - minor bubbling, some black powder that easily forms a suspension. I am so confused. This shouldn't be cobalt oxide - that should be in solution as cobalt sulfate.

Maybe I do have to try 'cracking' this with sodium nitrate and sodium hydroxide?

Update for the bleach tube: The cobalt oxide is now much larger-grained, coming out of suspension very quickly. All but the coarsest of the skutterudite particles seem to have dissolved - it's hard to tell by color which is CoO and which is skutterudite, but I'm fairly certain.

elementcollector1 - 17-4-2014 at 21:49

2 months later, and not much has changed on the test tube scale - in either case. I'm going to see if the ore responds in any way to a blowtorch (in really small quantities!). If yes, I guess I should try sublimation under vacuum...

blogfast25 - 18-4-2014 at 04:31

No joy with the bleach extraction?

elementcollector1 - 18-4-2014 at 06:47

Nope. No change at all...

bismuthate - 18-4-2014 at 14:54

Ok, so I thought up a few possible reactions. Sorry if they don't work.
1A skutterudite+HCl+Na2S2O3
1B skutterudite+H2O2+Na2S2O3
One of these, If they react as I predict, would yield As2S3.
2 skutterudite+HCl+I2
should yield arsenous acid
3 skutterudite+H2O2+HCl
Again this should yield arsenous acid.
Again sorry if I mde mistakes.

elementcollector1 - 18-4-2014 at 18:48

H2O2 and HCl is something to try, but keep in mind that this stuff was nigh-immune to concentrated sulfuric acid. Also, what happens to the cobalt in each of these cases?

bismuthate - 19-4-2014 at 02:52

It theoreticaly would form cobalt chloride in solution.

[Edited on 19-4-2014 by bismuthate]

blogfast25 - 19-4-2014 at 05:03

I wonder if fusing and quenching may help. It's often used on minerals to render them responsive to strong acids, based or other reagents.

By strongly heating (melting, if possible), followed by quenching in ice cold water, the crystalline structure is often disrupted and becomes more 'glassy'. The glassy structure can then be ground up and tends to be more responsive to chemical attack.

Worth trying on a tiny piece, IMHO...

elementcollector1 - 19-4-2014 at 07:15

I wonder if fusing and quenching may help. It's often used on minerals to render them responsive to strong acids, based or other reagents.

By strongly heating (melting, if possible), followed by quenching in ice cold water, the crystalline structure is often disrupted and becomes more 'glassy'. The glassy structure can then be ground up and tends to be more responsive to chemical attack.

Worth trying on a tiny piece, IMHO...

Unfortunately, all of my stock is a fairly fine powder at this point. Also, wouldn't we want to avoid formation of gaseous arsenic or arsenic oxide under uncontrolled conditions?

Per the cobalt: Is there some complexing agent that can separate it and arsenic?

blogfast25 - 19-4-2014 at 11:52

Per the cobalt: Is there some complexing agent that can separate it and arsenic?

The mystery is that cobalt ad arsenic would normally be sooo easy to separate. So there's something about the ore's unusual structure that's making things hard.

elementcollector1 - 19-4-2014 at 13:43

Found this tidbit from... answers.com?

Quote:

Skutterudite fuses on charcoal, giving an arsenic (garlic) odor and forming magnetic ball. A cobalt-rich skutterudite ball dissolves in nitric acid to form a pink solution, though iron and nickel are usually present in sufficient abundance to mask the color. Tests on analyzed samples showed that a nitric acid solution of (cobalt) skutterudite neutralized by ammonia becomes red-violet and a red-violet precipitate settles out. Nickel skutterudite under the same conditions gave a blue-violet solution and a pale green precipitate. Rarer ferroan skutterudite gives a strong brown iron-hydroxide precipitate that will mask either of the other elements.

So... I guess I need some nitric acid then.

elementcollector1 - 2-5-2014 at 20:14

Per the post above, I wonder what happened to the arsenic?
Ah well, time to test with a few drops of nitric acid. Going to dilute it, so I don't blow my face off...

EDIT: I was so very wrong about how much to dilute it. The test tube overflowed almost immediately! (I quickly washed my hands, hope for the best!) On the plus side, the supernatant liquid is now a ruddy brown (indicating both presence of iron and a healthy reaction), and there is a fluffly gray precipitate that is so far insoluble (though more nitric acid should determine this). If this precipitate is elemental arsenic, I will be a very happy man.

[Edited on 5-3-2014 by elementcollector1]

blogfast25 - 3-5-2014 at 04:43

EC1:

Your insoluble residue is probably either arsenic or arsenic trioxide.

There's a really simple test for As in solution. It used to be used as a forensic screening test. To test the supernatant liquid for As, dip a piece of thick gauge copper conductor wire in it. Allow to stand overnight. Copper reduces As compounds and the wire will turn black (plated As) if As is present. It does this also with Sb for which I tried it and it works.

If the solution is very strongly acidic you may have to partly neutralise it for this test, or your copper will dissolve.

[Edited on 3-5-2014 by blogfast25]

elementcollector1 - 3-5-2014 at 11:57

First, I added a bunch more skutterudite, which once again reacted (proving that there was still acid present). This gave another ruddy orange-brown solution, indicating ferric nitrate. Once this had died down, I attempted to dilute and filter this out, obtaining a yellow, cloudy solution (iron hydroxide precipitating from lower pH?). I added the copper wire whilst filtering, and more of this strange precipitate appeared. The color appears to be anything from off-white to peach, due to the color of the solution. Going to do a control test with no copper wire to determine if that's the cause...

On a side note, added the rest of my nitric acid to the leftover skutterudite particles, where it vigorously reacted (releasing lots of colorful fumes of NO₂). The same red-brown solution was obtained, and the insoluble residue on the bottom was dark gray in color. Some insoluble residue appears to be floating, and white in color.

What an odd experiment!

blogfast25 - 3-5-2014 at 12:40