Neodymium Chloride Solution

[elementcollector1]

When I've dissolved NIB magnets in HCl before, I ended up with a dark purple solution, as expected. But on this run, my solution turned a fairly dark green. Is this due to Fe²⁺ ions in solution? If so, what should I do with it to remove everything but the NdCl₃? Some things I've seen are:

 

-Solvent extraction by isopropyl and methyl-ethyl ketol

I can do this, but I'd have to order the MEK.

-Convert to sulfate, then solvent extraction in acetone

I'm not sure this works, I read it on another science forum.

-Precipitate out Nd(OH)₃ by adding solution of NaOH, then reconvert to chloride

This might precipitate out the Fe³⁺ ion as well, so not sure what to do here.

 

Can anyone give me the best course of action for an amateur chemist?

[John Cuthber]

I'd oxidise the solution to convert all the Fe to Fe+++ then add ammonia solution slowly to ppt all the Fe as Fe(OH)3 which is much less soluble than the Nd(OH)3.

An alternative would be to extract Fe(+++) from a solution of the salts in excess HCl with ether- but that only works if you have ether.

[elementcollector1]

Right. So, oxidise with 3% H2O2 solution, and add household ammonia until the brown Fe(OH)₃ precipitates out (keep adding until no further precipitation).

This should leave only NdCl₃ and possibly some leftover Fe(OH)₃ in solution, correct?

 

EDIT: Some online sites say a solution of NaOH could also be used. Is this true?

EDIT #2: Tried the process with NaOH instead of ammonia. Results:

On addition of H2O2, the green solution bubbled, and turned brown-orange. I did not add enough H2O2 to finish whatever reaction was taking place, as the solution still bubbled no matter how much I put in.

Next, I added a small amount of NaOH granules (about the size of a quarter) to 2 oz. This turned the solution opaque brown, and presumably an orange precipitate is forming.

Edited December 11, 2011 by elementcollector1

[elementcollector1]

Right. So, oxidise with 3% H2O2 solution, and add household ammonia until the brown Fe(OH)₃ precipitates out (keep adding until no further precipitation).

This should leave only NdCl₃ and possibly some leftover Fe(OH)₃ in solution, correct?

 

EDIT: Some online sites say a solution of NaOH could also be used. Is this true?

EDIT #2: Tried the process with NaOH instead of ammonia. Results:

On addition of H2O2, the green solution bubbled, and turned brown-orange. I did not add enough H2O2 to finish whatever reaction was taking place, as the solution still bubbled no matter how much I put in.

Next, I added a small amount of NaOH granules (about the size of a quarter) to 2 oz. This turned the solution opaque brown, and presumably an orange precipitate is forming.

 

Well, that didn't work. I learned through surfing the Web that I probably have some Praseodymium in my magnet. How would I get rid of this? (Also, I'm planning to use sulfuric acid instead of hydrochloric, it's supposed to be purer or something.)

I'm planning to follow TheChemlife's video on YouTube "turning magnets into chemicals", so presumably I would get rid of the Pr, try to eliminate some of the impurities, and then follow his procedure.

[John Cuthber]

I don't think it's practical for a home chemist to try to separate Pr from Nd.

[elementcollector1]

I don't think it's practical for a home chemist to try to separate Pr from Nd.

 

Well, I found an old purplish-pink solution of "magnet chloride," which did not have Pr in it, and evaporated it into purple crystals. I washed it with 99% isopropyl alcohol, and the alcohol came out dark green, leaving transparent crystals behind. I suspect the Nd is soluble in alcohol, but what should I do now? Should I attempt to evaporate the green alcohol solution? Should I just start over? This is all very confusing.

[John Cuthber]

Do you realise how difficult it is to separate the rare earths?

[StringJunky]

Wouldn't it be easier to buy?...it's $16 for 5g on this site. I don't do this sort of thing so forgive me if I've missed the point of the exercise.

 

http://elementsales.com/pl_element.htm#nd

[elementcollector1]

...I'm getting the distinct vibe I should start over.

To StringJunky, technically yes, that would be easier and maybe even cheaper... but it wouldn't be any fun to make neodymium that way! I'll only buy the element if I have absolutely no way to make it.

[StringJunky]

...I'm getting the distinct vibe I should start over.

To StringJunky, technically yes, that would be easier and maybe even cheaper... but it wouldn't be any fun to make neodymium that way! I'll only buy the element if I have absolutely no way to make it.

 

Gotcha...wouldn't be chemistry otherwise would it?

[elementcollector1]

Yup. I heard oxalate was a better way to go for lanthanide separation (from non-lanthanides), because Ln₂(C₂O₄)₃ seems to be insoluble in acidic solutions. So, dissolve magnet in hydrochloric acid, and add oxalic acid solution to precipitate out Nd₂(C₂O₄)₃, a light purple powder, which will be mostly free of impurities. Dry this, and put it back into hydrochloric acid to make a pure NdCl₃ solution... and so on. First, is there any way to make oxalic acid, or is it available in some form in stores? Wikipedia quotes 'Bar Keeper's Friend' as a cleaner containing oxalic acid, but I don't think the brand name would check out well at my age, and there's probably a whole lot of impurities. It also says it's used in baking powder, but that's probably for production.

[John Cuthber]

Your determination to do things the hard way is commendable (within limits).

Bar keeper's friend is for sale as a cleaning product. If you are young enough to worry about bars then the only concern people will have is that people that young don't usually do a lot of cleaning up.

It's probably horribly impure, but I have not checked.*

Practically pure oxalic acid is for sale on ebay.

You might want to add sodium oxalate to ppt the Ln salts, rather than adding the free acid but that's no problem as long as you have sodium bicarbonate, carbonate or hydroxide.

If you plan to buy the acid, you might just as well buy the salt. I'm not sure so check on this.

 

 

*

I checked: it's not worth it.

http://www.barkeepersfriend.com/files/file/Bar-Keepers-Friend_Powder-MSDS-2011.pdf

[elementcollector1]

Hmm. So, sodium oxalate is a toxic salt of sodium. I think I'll try to avoid that unless I have to.

Well, I do have baking soda, so I could make some if I order the acid. I'll just have to resist the temptation to eat it

[John Cuthber]

The free acid is at least as toxic as the salt.

A quick lookont google tells me that the stuff isn't very toxic.

Sodium oxalate has an LD50 (measured in mice) of about 5g/kg. If you weigh 70Kg that suggests it would take 350g to have a 50:50 chance of killing you.

It's silly to try to extrapolate like that because you are not the same as your bodyweight in mice, but it suggests that you are unlikely to kill yourself with this stuff by accident.There's some evidence of someone being harmed (but not killed) at about a tenth of that. Are you likely to accidentally eat an ounce of the stuff?

 

(Incidentally, lanthanum chloride is roughly as toxic)

[elementcollector1]

I see. Anyway, should I just order the oxalic acid online, or is sodium oxalate available somewhere?

[Chemistoftheelements]

Elementcollector1,

 

This is an interesting but difficult task you’ve set for yourself. From your previous experiments, during which you obtained a purple solution after dissolving your magnet, I suspect that the reason you had a purple solution was that you started with an alloy which consisted of cobalt and neodymium, at least in the main part. Many of the salts of both these elements are purple. The dark green ingredient to which you allude is likely iron (probably a mixture of FeII and FeIII) rather than Praseodymium, as the latter is, from my experience, definitely light green, although you could possibly have praseodymium in there as well as iron.

 

As for separation using the diffential solubilities of Fe and Nd hydroxides, I’m afraid that’s impractical, as they’re both extremely insoluble and the quantities you’d likely get would be so small as to make recovery very difficult if not impossibe for the amateur chemist using the resources we have available to us.

 

I looked into producing some simple representitive compounds of neodymium and praseodymium a while back, as I’m a keen element collector and amateur metallurgist, and decided that the chlorides would be way to hygroscopic to be useful. I settled on sulphates/ sulfates, and succeeded in producing some decent crystals stable in humidities of at least 90%. To produce large crystals (~ 1mm)relatively free of basic sulfate, ie Nd/Pr(OH)SO4, however, you need to begin with more metal than will dissolve in the H2SO4 and remain in solution; in other words, you have to exceed the solubility of (Nd/Pr)2(SO4)3 in sulfuric acid, wait intil you have a decent amount of crystals but still have a very acidic solution, then decant off the solution, leaving the crystals, which you will have to dry by tissue paper. You will then need to neutralise any acid you have on your tissue and anywhere else, of course. In the case of neodymium, if you rely on evaporation from a solution of weak acid, then your precipitate will turn orange, because you have produced Nd(OH)SO4.

 

I started with Neodymium and Praseodymium metals, and I know you’re actually aiming to end up with what I started with, but some of the things I learned may be of use to you. You have almost certainly encountered some of the compounds I did, but for different reasons. Turning these into Nd metal will be very difficult, though, and you may have to settle with a representitive compound. All this needs to be done, of course, with gloves, specs and other protective equipment, an outdoors if you’re going to do any heating, and always have plenty of sodium bicarbonate on hand to soak up and neutralise any acid spills.

 

 

[elementcollector1]

Hah, speaking of sodium bicarbonate, I just lost a large beaker to the stuff. Poor thing has a massive chunk missing.

Anyway, I know oxalic acid is used as wood bleach, so I'll look into that.

I'd like to go all the way to Nd and maybe Pr metals if I could, but we'll see.

Process of getting to Nd from neomagnets, in short:

1) Dissolve in acid.

2) Add oxalic acid or salt to precipitate insoluble oxalates.

3) Ignite in air to produce oxides.

 

(for separation of Pr)

4) Somehow convert the Pr₆O₁₁ that formed into PrO₂.

5) (for separation of Pr) Place in 5% acetic acid? I read below on Sciencemadness that PrO₂ is insoluble in 5% acetic acid, and apparently Nd₂O₃ is.

 

 

kmno4

International Hazard

 

Posts: 984

Registered: 1-6-2005

Location: Silly, stupid country

Member Is Offline

 

There is some old article* about "schwarzen Oxyde des Praseodyms", unfortunately in Übermenschen language.

As far I understand, PrO2 is not soluble in 5% acetic acid and Pr6O11 may be converted to Pr(3+) and PrO2. This procedure is also mentioned in another article** and its authors make pure PrO2 from Pr6O11. After a few steps I should be able to separate Pr and Nd from Pr-Nd mixture. Currently it is under my investigation. It looks simple on paper, I do not know if it works for mixture of oxides at all.

* DOI: 10.1002/zaac.19251490118

** DOI: 10.1021/ic50016a030

Edited December 31, 2011 by elementcollector1

[Chemistoftheelements]

Elementcollector1, it's likely that the free acid, H2C2O4 would be better than the sodium salt, as using the latter, you may get a double- salt or a Na2C2O4-Nd complex, and removing the sodium may prove difficult. If you need to raise the pH at any point, try adding NH3, but try this in a small scale first, so you don't ruin the whole lot if it doesn't work!

[elementcollector1]

My main problem with this is how to convert the Pr₆O₁₁ to PrO₂. Would hydrogen peroxide work? Excess heating? From the same forum as earlier, members have mentioned that on ignition, praseodymium oxalate turns brown, but I'm not sure if this is PrO₂ (in which case, yay!) or something else. Nd (III) cannot be oxidized any further by igniting in air, so that's safe.

[Chemistoftheelements]

Hah, speaking of sodium bicarbonate, I just lost a large beaker to the stuff. Poor thing has a massive chunk missing.

Anyway, I know oxalic acid is used as wood bleach, so I'll look into that.

I'd like to go all the way to Nd and maybe Pr metals if I could, but we'll see.

Process of getting to Nd from neomagnets, in short:

1) Dissolve in acid.

2) Add oxalic acid or salt to precipitate insoluble oxalates.

3) Ignite in air to produce oxides.

 

(for separation of Pr)

4) Somehow convert the Pr₆O₁₁ that formed into PrO₂.

5) (for separation of Pr) Place in 5% acetic acid? I read below on Sciencemadness that PrO₂ is insoluble in 5% acetic acid, and apparently Nd₂O₃ is.

 

 

 

Elementcollector1,

 

1) Happy new year!

 

2) To answer this and your latest question:

There’s some good logic to this, but I’m not sure that this would work, for several reasons:

 

1) At the oxide stage, the Pr would be in solid solution in the Nd2O3, and probably being only a minor constituent , would be prevented from oxidising to higher than Pr(III). In other words, the Nd(III) would stabilise the Pr(III), in the same way that Th(IV) stabilises U( IV) in high ( and low) U thorianites.

 

2) Oxidation states of Pr higher than 3 would be unstable in acid solution- think CeO2, so adding acetic acid would probably cause what little Pr in a >III oxidation state to revert pretty quickly to PrIII and go into solution.

 

3) Lanthanoid (and related oxides) don’t easily dissolve in acid once they’ve been ignited. Working on a theory that whether a metal oxide dissolves or not depends on the ionic radius, the coordination number of the metal, and possibly it’s affinity for oxigen, I produced some Yb2O3, Sc2O3 and Sm2O3 and found that all 3 were very reluctant to dissolve; after ignition at 1000 degrees C (and then cooling), only the Sm2O3 dissolved, but very slowly. If, however you heat them only minimally ( just enough to ignite the oxalate in your case), then you will probably succeed.

 

4) By using H2O 2, you may end up with a product containing O-O bonds, which would be bad, because such a compound may prove thermally dangerously unstable. You would have to work on very small quantities to find this out, or alternatively oxidise with something else.

 

What I would suggest is the following, which circumvents problem 1, but not the others, so it’s not guaranteed to work:

 

1) Dissolve in acid

 

2) Add H2C2O4 to precipitate Nd and Pr and separate them from Fe, etc. Add NH3 to raise pH to increase precipitation if necessary, or dilute with water.

 

3) Gently heat precipitate to create Nd/Pr oxides.

 

4) Dissolve oxides in H2SO4, NOT HCl, as this is reducing.

 

5) Oxidise this resulting solution to produce Pr2 O(SO4)3(?), a higher basic sulfate of praseodymium which will be insoluble.

 

6) Filter. You have now recovered Pr.

 

7) Ignite your P2 O(SO4)3

 

7)To remaining solution, add NH3 to precipitate Nd(OH)3.

 

8) Ignite the Nd(OH)3.

 

From here, you’ll have to figure out a way to reduce the Pr and Nd oxides. Good luck, and let us know how you get on. In the mean time, I’ll try to figure out how to create subscripts!

[elementcollector1]

Elementcollector1,

 

1) Happy new year!

 

2) To answer this and your latest question:

There's some good logic to this, but I'm not sure that this would work, for several reasons:

 

1) At the oxide stage, the Pr would be in solid solution in the Nd2O3, and probably being only a minor constituent , would be prevented from oxidising to higher than Pr(III). In other words, the Nd(III) would stabilise the Pr(III), in the same way that Th(IV) stabilises U( IV) in high ( and low) U thorianites.

 

2) Oxidation states of Pr higher than 3 would be unstable in acid solution- think CeO2, so adding acetic acid would probably cause what little Pr in a >III oxidation state to revert pretty quickly to PrIII and go into solution.

 

3) Lanthanoid (and related oxides) don't easily dissolve in acid once they've been ignited. Working on a theory that whether a metal oxide dissolves or not depends on the ionic radius, the coordination number of the metal, and possibly it's affinity for oxigen, I produced some Yb2O3, Sc2O3 and Sm2O3 and found that all 3 were very reluctant to dissolve; after ignition at 1000 degrees C (and then cooling), only the Sm2O3 dissolved, but very slowly. If, however you heat them only minimally ( just enough to ignite the oxalate in your case), then you will probably succeed.

 

4) By using H2O 2, you may end up with a product containing O-O bonds, which would be bad, because such a compound may prove thermally dangerously unstable. You would have to work on very small quantities to find this out, or alternatively oxidise with something else.

 

What I would suggest is the following, which circumvents problem 1, but not the others, so it's not guaranteed to work:

 

1) Dissolve in acid

 

2) Add H2C2O4 to precipitate Nd and Pr and separate them from Fe, etc. Add NH3 to raise pH to increase precipitation if necessary, or dilute with water.

 

3) Gently heat precipitate to create Nd/Pr oxides.

 

4) Dissolve oxides in H2SO4, NOT HCl, as this is reducing.

 

5) Oxidise this resulting solution to produce Pr2 O(SO4)3(?), a higher basic sulfate of praseodymium which will be insoluble.

 

6) Filter. You have now recovered Pr.

 

7) Ignite your P2 O(SO4)3

 

7)To remaining solution, add NH3 to precipitate Nd(OH)3.

 

8) Ignite the Nd(OH)3.

 

From here, you'll have to figure out a way to reduce the Pr and Nd oxides. Good luck, and let us know how you get on. In the mean time, I'll try to figure out how to create subscripts!

 

Excellent! I assume they could be reduced with calcium metal? I got a large amount from United Nuclear, so it would be more than enough to reduce whatever I produce.

So I can learn this for later, why does raising pH increase precipitation rate?

As for ignition, a blowtorch and a clay crucible would work, right?

Thanks for the solubility tip. I was trying to make CeCl₃ from CeO₂ earlier, with little success.

Happy New Year to you as well!

 

 

[Chemistoftheelements]

Excellent! I assume they could be reduced with calcium metal? I got a large amount from United Nuclear, so it would be more than enough to reduce whatever I produce.

So I can learn this for later, why does raising pH increase precipitation rate?

As for ignition, a blowtorch and a clay crucible would work, right?

Thanks for the solubility tip. I was trying to make CeCl₃ from CeO₂ earlier, with little success.

Happy New Year to you as well!

 

 

It’s not the rate of precipitation, but the amount of precipitate. The pH at which precipitation occurs depends on the species in question. In electopositive metals like Pr, which tend to form oxides and hydroxides with no acidic properties, precipitation will be promoted by alkalinity ( i.e. higher pH).

 

In the case of the stage where you’re trying to precipitate the Pr₂O(SO4)3, the same principle applies. But because the Pr will be in a higher oxidation state, it will be less basic, and will precitate at a lower pH, i.e. the solid will be more stable against acid. Think of the example of Sn- it has 2 common oxidation states, with Sn(II) predomination, except when it comes to oxides, when Sn(IV) is easily produced.; Sn(IV)O₂ easily precipitates in acid conditions, whereas Sn(II)O requires more alkaline conditions, i.e. a higher pH.

 

As Nd(III) is more basic than Pr(IV), your Pr will precipitate from your Nd/Pr in H2SO4 solution at a lower pH, i.e. in more acidic conditions, than your Nd will. You may find, though, that your solution is so acidic that even the Pr won’t precipitate, hence the requirement to raise the pH by adding NH3 (and it must be NH3, not, for example NaOH). You don’t want to raise the pH (make your solution more alkaline) so much, though, that your Nd precipitates out as well. I’m not sure what colour/color Pr(IV) is (this will depend anyway on the individual compound due to differences in coordination number), but from my own experiments, as you raise the pH, your Nd will begin to precipitate not as Nd(OH)₃ but as Nd₂O(SO₄)₂, which is orange (I began with 99.9% Nd metal, so there’s no possibility of contamination), so this may, and I stress may, serve as an indication of when to stop.

 

A crucible and a blowtorch will be a good way to ignite your precipitates. When you heat your Pr₂O(SO4)3, be sure to do this outdoors, as you will produce SO₃! You may find you need to heat strongly here- to 1000⁰C at least. Perhaps using small pieces of sheet metal would be better for this one, and heat underneath (I’ve used Ti on a number of occasions, but with any metal, you have to be carefull it doesn’t flake and contaminate the material you’re trying to ignite) as the crucible may conduct too much heat away.

 

The Ca method of reducing your oxides was the way I had in mind, too. I’ve not done this myself, and have often wondered how you avoid contaminating the metal you’re trying to produce with Ca. I guess it depends on getting your stoichiometry exactly right, and performing in a inert atmosphere.

 

Good luck!

 

 

 

P.S. The bit about the greater stability of higher oxides towards acids seems to apply only to those which are not strongly oxidising. For example, when you treat Pb in HNO3, you might think

 

that you’d produce a passivating layer of Pb(IV)O₂ which would prevent the Pb from dissolving, but this doesn’t happen- the Pb dissolves, and you can produce Pb(NO₃)₂ this way. However, it’spossible that Pb(IV)O₂ forms at least temporarily, as this reaction takes ages. Pb(IV) is very strongly oxidising. After doing a bit of research, Ce(IV) seems to be a lot more stable, (albeit still oxidising), and this quite possible applies to Pr(IV) as well, which is encouaging for the sake of your endevour.

 

 

[elementcollector1]

Right. Would household ammonia work for this project, or is there another source for a purer product?

As far as I know, Pr(IV) tends to be brown, and Pr(III) dark green.

So, I'll have to borrow some sheet metal. Could a stainless steel shot-glass work as well? Assuming I could find one, that is.

Thermites can be difficult to ignite in inert-atmosphere, but once you get them going they have their own source of oxygen, so it might work.

Household ammonia seems to be 5-10% concentration. Is this enough to raise pH significantly?

[Chemistoftheelements]

Yes, and this is even desirable, as it makes for a gentler means of raising pH. It’s what I’ve always used. I have a feeling that you may not need to use the NH3, as I think your praseodymium will precipitate at a fairly low pH. Just make sure you dissolve all your mixed Nd/Pr oxides before you oxidise, as then you’ll know that then you have a solution which is too acidic to precipitate Nd, but not necessarily Pr when the solution is oxidised to crease Pr(IV).

 

As for sheet metal, check out steel beer cans, or food cans, although the latter are a little thicker and would be a bit more difficult to cut. All you’re looking to do is to ignite your precipitate to a bright red heat, so you may have what you need already; do a few experiments, and with steel cans if this doesn’t work, noting whether the steel oxidises and flakes too much, in which case you may have to use stainless.

 

I have only college- level chemistry, and all this advice comes from my work, current study, and hobby experience, so in a sense, my practice is ahead of my theory, so don’t take my (or anyone’s for that matter) word as fact- be cautious and always experiment on a small scale first. The lanthanoids are a fascinating group of elements which deserve more attention, which are as least equally as frustrating, so good luck, keep us informed and don’t give up!

 

P.S. The colour of your Pr(IV) precipitate may not be the same as that of pure Pr(IV) in solution, so that it may not be brown. Think of NiCl₂, which is grass- green as a solid and in solution, but when NH3 is added to an aqueuous solution containing this species, you get a very purple precipitate of NiCl_2­- NH3 complex! This is not due to the NH3 as such, but to the resultant coordination number of the Ni atom. What I'm saying is that it may be better to look for colour CHANGE as you adjust the pH, rather than colour per say.

 

 

 

 

P.S. 2, when you add H₂C₂O_4, to precipitate your Nd and Pr, if you use any NH_3­, then do so very cautiously- you don’t want to raise the pH enough to precipitate Fe or any other contaminents. H₂C₂O_{4 ­}should give you the advantage here, as it should precititate Nd and Pr in quite low pH (acidic) solution, so NH_{3 ­}may not be needed.

[elementcollector1]

Most appreciated. And you said Pr and Nd were difficult to separate, John.

As soon as I get back from Canada, I'll get right on this.