Sciencemadness Discussion Board

General Discussion of Terbium Compounds

 Pages:  1  

Brain&Force - 13-11-2013 at 17:12

Hello,
I have a question for you all. I'm making hexakis(antipyrine)terbium(III) iodide for my science fair project. The idea for this project came from this thread: Green Smash-Glow Crystals
I just got terbium and antipyrine, and I have access to iodine. Pok stated that he dissolved the terbium in HCl and added antipyrine and potassium iodide. However, I'm going to make both hexakis(antipyrine)terbium(III) iodide and hexakis(antipyrine)terbium(III) chloride for comparison (my project is quite complex).
I want to make the compound without going this route - rather, I want to make pure TbI3 from direct union of the elements by dissolving terbium in an aqueous iodine solution and slowly adding more iodine. I would expect the iodine to complex with the iodide created by the reaction to form triiodide ions, and the terbium would react with the triiodide to form more iodide. Is this mechanism feasible? Would it work for other metals?
Thanks for all of your replies; this is my first time here.

<!-- bfesser_edit_tag -->[<a href="u2u.php?action=send&username=bfesser">bfesser</a>: changed subject from "Synthesis of terbium(III) iodide" per OP request]

[Edited on 10.1.14 by bfesser]

blogfast25 - 14-11-2013 at 06:19

The lanthanides are very electropositive, react vigorously with dilute strong acids etc so direct union of Tb and I2 (in water) might be possible. But reactivity towards water may ruin that plan. Have you considered direct union in the absence of water? It's harder to do of course.

Do you have any solubility data on TbI3? Several RE chlorides are extremely hygroscopic and hard to crystallise. Not sure about the REI3s...


[Edited on 14-11-2013 by blogfast25]

Pok - 14-11-2013 at 06:35

TbI3 is well soluble in water. According to this site TbI3 can be made by direct reaction of terbium and iodine. Iodine is not very soluble in water so it may take longer to react than pure Tb + pure I2.

Maybe heating Tb metal with iodine will make the whole reaction proceed much better. In this case I would recommend to use a massive piece of Tb an not powder. Because powder will probably react much too violent.

But I don't understand why you don't want to use the method with potassium iodide. Do you only have pure iodine and not KI ?



[Edited on 14-11-2013 by Pok]

Wizzard - 14-11-2013 at 08:46

I would follow the pre-conceived procedure, and separate by crystallization.

Looks like you have enough terbium!

DraconicAcid - 14-11-2013 at 09:19

I've never worked with terbium, but my students make zinc iodide by heating zinc metal with iodine in dry methanol. If you're worried about it being too reactive, I'd use a small amount of iodine, and just add more iodine as the colour disappears from solution.

blogfast25 - 14-11-2013 at 10:30

I like DA's idea a lot. For want of a better term, 'reactivities' of Tb and Zn are probably quite comparable. This has the potential of being highly exothermic, especially if the TbI<sub>3</sub> turns out to be insoluble in the solvent.

[Edited on 14-11-2013 by blogfast25]

Brain&Force - 14-11-2013 at 16:11

Thank you all for your replies.

blogfast25, direct union is quite favorable, but terbium is nearly as reactive as magnesium (-2.28V for Tb3+ + 3e- → Tb) and I don't want to have a fire or explosion risk. If the aqueous iodine solution is acidic enough, my idea should work.

Pok, I'm trying to make absolutely pure terbium chloride and terbium iodide as well as their antipyrine complexes. My goal is to figure out whether swapping the anion has an effect on the intensity of the triboluminescence. Also, I need to determine the hydration number of pure terbium iodide (I can't find a reference to this anywhere).

DraconicAcid: I might try two batches, one with methanol and one with water. I have access to a vacuum dessicator, so that'll help if terbium iodide is soluble in methanol. Also I won't have to worry about unknown hydration numbers this route.

blogfast25 - 15-11-2013 at 05:23

Quote: Originally posted by Brain&Force  
If the aqueous iodine solution is acidic enough, my idea should work.



Do compare the half potentials for:

H3O+ + e- === > H2O + 1/2 H2

And:

1/2 I2 + e- === > I-

... to ensure your precious terbium doesn't get chewed up by the acid.

Also, methanol should be bone dry: dry with lime (CaO) for instance.

I'll be having a go at ZnI2 (and maybe some other iodides in anhydrous denaturated ethanol).

Brain&Force - 15-11-2013 at 15:32

Quote:

Do compare the half potentials for:

H3O+ + e- === > H2O + 1/2 H2

And:

1/2 I2 + e- === > I-

... to ensure your precious terbium doesn't get chewed up by the acid.


Acidification of the solution shouldn't be necessary. I thought I2 would disassociate slightly into HI and HIO, and that the produced H3O+ and IO- would oxidize the terbium.

However, I'm pretty sure I'll be doing this in methanol or ethanol - it sounds a whole lot simpler, and I may be able to get data for the solubility of terbium iodide in these solvents - data which is very sparse. Nobody investigates lanthanide compounds. :(

deltaH - 15-11-2013 at 15:41

Quote:
However, I'm pretty sure I'll be doing this in methanol or ethanol - it sounds a whole lot simpler, and I may be able to get data for the solubility of terbium iodide in these solvents - data which is very sparse. Nobody investigates lanthanide compounds. :(

In the absence of data, I would think in this respect that RE halides behave a little like aluminium halides and would thus be highly soluble in polar solvents like alcohols.

Brain&Force - 15-11-2013 at 16:10

Both aluminum and lanthanide halides hydrolyze easily and act as Lewis acids, so you might be right on that. Antipyrine is extremely soluble in ethanol, so I might be able to make [Tb(C11H12N2O)6]I3 in it.

I wonder if terbium halides show fluorescence in alcohols - they don't show it in water solution.

blogfast25 - 16-11-2013 at 06:05

B&F:

There is actually quite a bit of research going on, on REs, now it appears they're not so rare anymore and with so many so useful. And there was a lot of it during the quest for the transuranium elements due to the expected similarities of lanthanides and actinides.

I think several of us here would appreciate a report on the TbI<sub>3</sub> synth (in MeOH or EtOH).

Brain&Force - 16-11-2013 at 10:47

Quote:

I think several of us here would appreciate a report on the TbI3 synth (in MeOH or EtOH).


I'll certainly keep you posted with pictures and hopefully videos.

What I'm particularly interested in is the formation of a terbium(II) species forming during the reaction - dysprosium(II) and samarium(II) iodides exist, but I can't find any references to terbium(II) iodide. WebElements lists a +2 oxidation state for terbium, but doesn't give a reduction potential for it.

blogfast25 - 17-11-2013 at 06:06

A Tb(II) species must transiently exist, as the ionisation must proceed in steps: I, II, III. If enough oxidant is present it will run all the way to III. Perhaps just using enough iodine for Tb + I<sub>2</sub> === > TbI<sub>2</sub> should do the trick?

watson.fawkes - 17-11-2013 at 07:37

Quote: Originally posted by blogfast25  
A Tb(II) species must transiently exist, as the ionisation must proceed in steps: I, II, III. If enough oxidant is present it will run all the way to III.
Must it? This a claim that needs evidence. Particularly it should exclude the hypothesis that Tb+ + I2 --> Tb3+ + 2 I- is an elementary reaction. I'm not saying this is likely, but it's just not right to assert an ionization process as a necessity without supporting evidence for it.

blogfast25 - 17-11-2013 at 11:43

Quote: Originally posted by watson.fawkes  
[...] Tb+ + I2 --> Tb3+ + 2 I- is an elementary reaction


Two ionisations in one collision: I just don't think it's likely.

Tb<sup>+</sup> + I<sub>2</sub> === > Tb<sup>2+</sup> + I<sup>-</sup> + I*

or

Tb<sup>+</sup> + I* === > Tb<sup>2+</sup> + I<sup>-</sup>

... seem more likely to me. But I shouldn't have used to word 'must'.

[Edited on 17-11-2013 by blogfast25]

Brain&Force - 17-11-2013 at 14:41

Wikipedia has a reference to terbium(I) chloride and terbium(I,II) chloride. Perhaps in MeOH/EtOH Tb+, Tb2+ and Tb3+ can at least coexist (if a stepwise oxidation doesn't occur).

My plan is to put the whole piece of terbium in the alcohol and add only the amount of I2 needed, then extract the leftover terbium once the reaction is finished. That way no iodine remains in solution. But if I have leftover iodine dissolved in MeOH/EtOH after making TbI3, will vacuum desiccation cause the iodine to evaporate as well? Or do I have to remove the iodine through some other method?

blogfast25 - 18-11-2013 at 05:30

There's no reason why you shouldn't use an excess of Tb powder to react away the I<sub>2</sub>: with lump metal things will be much, much slower. I<sub>2</sub> is quite volatile, so leftovers should evaporate on removal of the solvent.

Coexistence of multiple metal oxidation states in water is quite rare: the equilibria tend to settle on the ion with the highest (well, most negative!) Free Energy of Formation. Tb(I) and Tb(II) would probably result in Tb(0) and Tb(III).


[Edited on 18-11-2013 by blogfast25]

Eddygp - 18-11-2013 at 11:55

This certainly is a great synthesis. I would love to have those lanthanides now, but they are too expensive. Anyway, careful because it might start to oxidise if exposed to air, even if it's not among the most reactive.

Trivia: Yttrium, ytterbium, terbium and erbium are ALL named after the same city in Sweden: Ytterby.

[Edited on 18-11-2013 by Eddygp]

DraconicAcid - 18-11-2013 at 12:54

Ytterby isn't even a city- it's a village.

Eddygp - 18-11-2013 at 13:29

Quote: Originally posted by DraconicAcid  
Ytterby isn't even a city- it's a village.

Yes... well, I meant town or something hahaha.

Brain&Force - 19-11-2013 at 21:01

Eddygp, lanthanides aren't that expensive - that is, if you know where to look. Sigma and Alfa Aesar charge ridiculous prices for gram amounts.

Terbium doesn't oxidize quickly in air. It takes at least several years; check here. But the piece is dark from Tb4O7 contamination.

Blogfast: I don't have equipment to powder the terbium. Hopefully the metal is reactive enough for the reaction to proceed quickly.

I'm going to try the synthesis this Thursday. Hopefully, I'll have nice fluorescent crystals of TbI3 soon.

blogfast25 - 20-11-2013 at 04:34

Quote: Originally posted by Brain&Force  
Blogfast: I don't have equipment to powder the terbium. Hopefully the metal is reactive enough for the reaction to proceed quickly.



Then I must have misread, I thought you had powder. Lump could be very slow.

Lab report, day 1 - Funny story

Brain&Force - 20-11-2013 at 16:36

Even though I can't get powder, I did discuss the idea with my mentor for the fair and we decided to to the next best thing - break it into small pieces. So he got a hammer. Our plan was to wrap it in paper, to tap it lightly and hope we'd get small pieces.
Tapping the terbium didn't work too well. The rough edges fractured, but not the whole thing. So he decided to give it a bit more force. That also didn't work, but the terbium did emit sparks.
And then came the second hit.
The terbium didn't fracture. It flew all the way from one end of the classroom to the other, and landed in the hallway (it took us a few minutes to figure that out) On retrieval, it was still in one piece.
We tried pretty much everything possible. A biotech student even suggested wrapping the terbium in parafilm. At the end of the whole ordeal, the bulk of the terbium had survived, only slightly flattened by the multiple, highly damaging impacts. The metal is so soft that pieces of paper had actually embedded themselves in the surface.
Terbium is VERY impact resistant, just not scratch resistant - the coating on a neodymium magnet can scratch the metal. We're hoping we can make some turnings with a drill.
(By the way, neodymium magnet really comes in handy for picking up small pieces.)
More updates coming up tomorrow.

blogfast25 - 21-11-2013 at 06:28

Hacksaw? Won't be easy either...

On the subject of immortality

Brain&Force - 22-11-2013 at 17:19

Terbium will not die.

My mentor for the fair took the terbium to his house and tried to break it. He used a saw, a knife, and all sorts of other tools. The only thing that worked was a chisel. We now have small pieces, as well as a large chunk of terbium that made it.

I attempted the synthesis yesterday. 0.48 grams of terbium was added to 1.14 grams of iodine dissolved in 20 mL of methanol. Not all of the iodine was added at once because I wanted to monitor the reaction to see any signs. I added about 20 mg initially, but that was enough to turn the solution a very dark brownish-purple. I added in all of the terbium, but it was very difficult to see with the amount of iodine in solution. A few bubbles were escaping from solution. In order to accelerate the reaction, I added the rest of the iodine. The reaction did not appear to speed up at all.

In a test tube, I decided to perform a test with small flakes of iodine and terbium (exact amounts were unknown) and distilled water (about 2 mL) was added. The solution became yellowish brown and bubbling began. Not all of the iodine dissolved. The two solutions were left to sit for the next day.

The next day, no change appeared in the methanol beaker. The solution remained extremely dark, and terbium was visible at the base of the flask.

Something completely different happened in the test tube with water. The water had mostly evaporated, leaving behind a froth. On the addition of water, it became apparent that all of the terbium had reacted with the iodine solution. No oxide powders appeared to be in the solution. I added in more terbium and water to react with the iodine. After an hour or so, the terbium flakes were bubbling, and a very fine white powder appeared to be suspended in the solution. I forgot to check for fluorescence (terbium oxide exhibits green fluorescence), so I can't be sure if that was the oxide or something else.

I won't be experimenting next week, but I'll update soon.

Pok - 23-11-2013 at 06:10

Quote: Originally posted by Brain&Force  
Something completely different happened in the test tube with water. The water had mostly evaporated, leaving behind a froth. On the addition of water, it became apparent that all of the terbium had reacted with the iodine solution. No oxide powders appeared to be in the solution.

So what did you get? A clear solution dark coloured due to a KI-I like iodine-complex?
Quote: Originally posted by Brain&Force  
I added in more terbium and water to react with the iodine. After an hour or so, the terbium flakes were bubbling, and a very fine white powder appeared to be suspended in the solution.

This sounds as if no free iodine was present anymore. Is this true? If this product doesn't dissolve then you should have something different than in the first case described above!
Quote: Originally posted by Brain&Force  
I forgot to check for fluorescence (terbium oxide exhibits green fluorescence), so I can't be sure if that was the oxide or something else.

This isn't a good way to check your product. TbCl3 also shows green fluorescence. TbI3 should dissolve in water extremely well but the oxide does not. So this is the easiest way to check your product. But: some rare earth halides form oxy or hydroxy halides when heated in water. This is what you may have produced in the second case (white precipitate)

Here is a paper with solubility data for the rare earth halides (except fluorides). I don't understand Fig. 1 there, but this may be the basis to calculate the solubility of TbI3. Does anyone understand this figure and can give an example for the calculation of the solubility of one salt?

Morgan - 23-11-2013 at 06:19

I wonder if terbium would become brittle, harder, or unaffected if placed in liquid nitrogen?

blogfast25 - 23-11-2013 at 06:24

@ Pok:

Great find [that paper]! I believe, gathered from the text, that 'x<sub>LnX3</sub>' stands for 'mole fraction of LnX<sub>3</sub> in the solution'. A very thermodynamics type of solubility measuring unit. It can be easily converted to the more traditional 'g / 100 solvent'.


@ B&F:

In either case, did you notice any heat being generated?

"No oxide powders appeared to be in the solution." Does this mean the solution was clear?

If the bubbling wasn't due to boiling then it's likely to be hydrogen (you can check that of course) which would indicate you're getting the water to oxidise the Tb. Possibly to the hydroxide Tb(OH)<sub>3</sub>, which may then react with the iodine to form iodide + iodate:

3 I<sub>2</sub> + 3 OH<sup>-</sup> === > 5 I<sup>-</sup> + IO<sub>3</sub><sup>-</sup> + 3 H<sup>+</sup>

... according to the classical preparation of iodide from iodine and a strong alkali. As a hypothesis it would at least explain why you get bubbles AND the iodine reacts away.

When it comes to the direct union of Tb and I<sub>2</sub> in a solvent and at RT, this may not necessarily start without heat: 'activation energy' may be needed.

@ Morgan: it will become more brittle, of course. But a practical way of size reducing this ain't, at the home level.


[Edited on 23-11-2013 by blogfast25]

Pok - 23-11-2013 at 10:26

Quote: Originally posted by blogfast25  
I believe, gathered from the text, that 'x<sub>LnX3</sub>' stands for 'mole fraction of LnX<sub>3</sub> in the solution'. A very thermodynamics type of solubility measuring unit. It can be easily converted to the more traditional 'g / 100 solvent'.

And how? Please tell me if the following calculation is correct:

Example: NdCl3

Solublity of NdCl3 = 967 g/1000 ml at 13 °C.

967 g NdCl3 = 3,85 moles
1000 ml water = 55.5 moles

Sum of moles = 59.35

moles NdCl3 / total moles = 3,85 / 59.35 = 0.0649 = the "x" ?

According to Fig. 1 (see below) "x" would be ca. 0.063 at 25 °C.

This sounds correct, right?

test.png - 24kB
Solubilities of rare earth halides

[Edited on 23-11-2013 by Pok]

blogfast25 - 23-11-2013 at 10:51

Ok. Suppose a binary system of A and water, with x<sub>A</sub> the mole fraction of A. Call n<sub>A</sub> the amount of A in moles and n<sub>H2O</sub> the amount H2O in moles . Then:

x<sub>A</sub> = n<sub>A</sub> / (n<sub>A</sub> + n<sub>H2O</sub>;)


and:

x<sub>H2O</sub> = n<sub>H2O</sub> / (n<sub>A</sub> + n<sub>H2O</sub>;)


It follows that x<sub>A</sub> / x<sub>H2O</sub> = n<sub>A</sub> / n<sub>H2O</sub>

Also: mass balance: x<sub>A</sub> + x<sub>H2O</sub> = 1

Assume you have 100 moles of mixture (or any number, for that matter), that contains 100 . x<sub>A</sub> moles of A and 100 . (1 - x<sub>A</sub>;) moles of water.

Multiply both numbers of moles by molar mass to obtain weight in g (say A g and H2O g). Divide A g by H2O g and multiply by 100 to obtain gram of A per 100 g of water.


The hydrates of TbI<sub>3</sub> could be a serious hurdle for the OP's project: dehydrating them w/o hydrolysis is going to be nigh impossible I feel. It's hard enough for the chlorides, never mind the more fragile iodides.



[Edited on 23-11-2013 by blogfast25]

Pok - 23-11-2013 at 10:55

Ok. This sounds as if my last post is right. Correct me if this is wrong.

blogfast25 - 23-11-2013 at 10:58

Quote: Originally posted by Pok  
Ok. This sounds as if my last post is right. Correct me if this is wrong.


No, you are right. It's easier to do it from first principles though, IMO.

deltaH - 23-11-2013 at 11:54

Excuse my previous post, utter nonsense, completely misread that you got the reaction with the water version, thought it was all methanol. :mad:

Brain&Force - 23-11-2013 at 13:51

I need to clarify something else. The white powder was like suspended dust, very sparse and hard to see. I can't be sure if that was Tb(OH)3 or something else, which is why I wanted to use the UV light to check. If it was Tb(OH)3 the stuff would have glown green.
Quote:

So what did you get? A clear solution dark coloured due to a KI-I like iodine-complex?


That's what I'd guess. The solution was dark yellowish brown because of iodine, but there was still some undissolved iodine at the bottom.
Quote:

This sounds as if no free iodine was present anymore. Is this true?


There was still dissolved and undissolved iodine in the test tube. Sorry, I should have clarified that also.
Quote:

Here is a paper with solubility data...


Thanks for the paper; this might be useful for my write-up for the fair.
Quote:

In either case, did you notice any heat being generated?


I didn't feel any heat in the methanol solution; I just saw bubbles being generated. The iodine solution in water had a lot more bubbling activity.
Quote:

"No oxide powders appeared to be in the solution." Does this mean the solution was clear?


Correct.
Quote:

If the bubbling wasn't due to boiling then it's likely to be hydrogen (you can check that of course) which would indicate you're getting the water to oxidise the Tb. Possibly to the hydroxide Tb(OH)3, which may then react with the iodine to form iodide + iodate:


3 I2 + 3 OH- === > 5 I- + IO3- + 3 H+

... according to the classical preparation of iodide from iodine and a strong alkali. As a hypothesis it would at least explain why you get bubbles AND the iodine reacts away.


True, but I thought this equilibrium exists between iodine and water:

I2 + H2O ↔ HI + HIO

and that this would acidify the solution so that the Tb would be able to dissolve in the solution rather than precipitate as Tb(OH)3. As for the formed hypoiodous acid, it would either reduce the Tb or disproportionate to I- and IO3-. If I have iodate, how do I reduce it to iodide?
Quote:

When it comes to the direct union of Tb and I2 in a solvent and at RT, this may not necessarily start without heat: 'activation energy' may be needed.


I thought the reaction would be immediate because of terbium's extremely negative reduction potential. But for a supposedly reactive metal, terbium is quite inert! I'll try direct union (heating Tb and I2 in a test tube on a VERY small scale) and I'll see what happens.
Quote:

Excuse my previous post, utter nonsense, completely misread that you got the reaction with the water version, thought it was all methanol. :mad:


I did get bubbling in the methanol solution, but it wasn't much. I know the methanol is dry (it says absolute methanol on the label). I'd be surprised if terbium formed a methoxide. I'll leave a piece of terbium in methanol and see what happens.

deltaH - 23-11-2013 at 14:27

If all else fails, perhaps a round about experimental route:

Tb pieces are reacted in/with excess glacial acetic acid to form terbium acetate solution in glacial acetic acid. The rationale here is that terbium acetate is probably soluble enough in acetic acid, the system is anhydrous and the acetic acid should make quick work of any surface oxides on the Tb that may otherwise have passivated it in anhydrous conditions.

Then bubble through a slight excess of HI and finally filter hopefully insoluble TbI3 that forms.

Finally evaporate under prolonged vacuum to recover acetic acid free TbI3 and get rid of everything else (acetic acid and HI).

The reason I suggest this route and not using a preprepared HI/acetic acid solution is because I expect TbI3 not to be particularly soluble in glacial acetic acid.

However, EVEN if soluble, prolonged vacuum will strip off the acetic acid anyhow, it will just take longer.

Just remember to be kind to your vacuum pump and use an inline cold trap! These vapours would be corrosive as hell! Liquid nitrogen is best but even dry ice should work well enough in this case.

[Edited on 23-11-2013 by deltaH]

blogfast25 - 24-11-2013 at 05:52

HI is difficult to synth/obtain. Might as well start planning for direct union: Tb + I<sub>2</sub>. Lead I<sub>2</sub> vapour/Ar mix over size-reduced Tb metal in quartz tube.

[Edited on 24-11-2013 by blogfast25]

@ B&F:

In the synthesis of iodide/iodate from KOH and I2, the iodate is converted to iodide by heating. But there we start from an anhydrous product...

Re. the electropositivity of Tb and activation energy: it doesn't automatically follow that an electropositive element will combine spontaneously with an electronegative one without increasing temperature. The Tb doesn't spontaneously combine with oxygen, for instance. Clearly the Tb/I<sub>2</sub> system is different from Draconic students' Zn/I<sub>2</sub> system.


[Edited on 24-11-2013 by blogfast25]

deltaH - 24-11-2013 at 06:18

Quote: Originally posted by blogfast25  
HI is difficult to synth/obtain. Might as well start planning for direct union: Tb + I<sub>2</sub>...


Metal halogenations in furnaces are no joke. While the furnace is relatively 'easy' to get a hold of, making the innards for it, typically of fused quartz, is not.

If HI is cumbersome, what about a simple ion metathesis using sodium iodide added to the terbium acetate/acetic acid solution. NaI is pretty soluble in many organic solvents. One would need to look up all the solubilities preferably if they are available to design the experiment properly. Here's one:

From wiki's article on NaI:

Solubility of NaI in various solvents
(g NaI / 100g of solvent at 25°C)

H2O 184
Liquid ammonia 162
Liquid sulfur dioxide 15
Methanol 62.5 - 83.0
Formic acid 61.8
Acetonitrile 24.9
Acetone 28.0
Formamide 57 - 85
Acetamide 32.3
Dimethylformamide 3.7 - 6.4
Dichloromethane 0.009

Solubilities from:

Burgess, J. "Metal Ions in Solution" (Ellis Horwood, New York, 1978) ISBN 0-85312-027-7
Danil de Namor, A.F.; J. Chem. Soc., Faraday Trans. 1, 1989,85, 2705-2712 DOI: 10.1039/F19898502705

I could not find data for the solubility of sodium acetate in acetic acid but surely it exists somewhere.

Still also need to know the solubility of TbI3 and acetates in acetic acid (preferably).

Anyhow, trying to dissolve some terbium in glacial acetic acid should be trivially easy. Then adding a saturated solution of NaI in glacial acetic acid will either yield a coloured precipitate or not. Simple enough to try, no?

[Edited on 24-11-2013 by deltaH]

blogfast25 - 24-11-2013 at 06:33

@ Delta:

I think it gets complicated rather quickly. Your GAA would have to be truly anhydrous for instance. The stuff I've got is 95 +, IIRW.

Direct union of iodine vapour and terbium isn't for the faint of heart either but marginally easier (I think) than what you propose. I guess it depends: some are much easier than others.


[Edited on 24-11-2013 by blogfast25]

deltaH - 24-11-2013 at 06:41

Yeah ok... the op's got some options then and we've done our job, but I don't see why the average run of the mill GAA wouldn't be just fine in this case. Is it not ordinarily pretty anhydrous?

Just checked sig-al's GAA's spec sheet, it says water <= 1500ppm, surely that's fine, or is CP grade GAA normally much worse than this? This is your area of expertise :)

Ah ok, saw you edited yours as well :P 95%... that's not very good, is the 5% all water? I wouldn't exactly call that 'glacial'?

[Edited on 24-11-2013 by deltaH]

blogfast25 - 24-11-2013 at 09:07

Actually, I told an inadvertent lie: mine says 99.85 % (AA assay). Assuming the rest is water that would be about 1500 ppm, pretty good.

Re. Sig Al: good luck buying from them. I run a chem business and they still won't well to me. I'm going to appeal just to rattle their cage a bit.

As far as stabs in the dark go, your idea would be worth testing.


[Edited on 24-11-2013 by blogfast25]

woelen - 24-11-2013 at 11:01

If you have acetic anhydride, then you can make your glacial acetic acid really free of water. I have 250 ml of glacial acetic acid and I added 1 ml of acetic anhydride to be sure that it is absolutely free of water.

deltaH - 24-11-2013 at 11:04

Blogfast
Yeah same specs then, I knew you run a chem business ;) I also expected the sig. al. comment lol I'm no fan either for the record!

Nice trick woelen!

[Edited on 24-11-2013 by deltaH]

Brain&Force - 25-11-2013 at 15:51

The GAA idea is a good one, but there's a problem. I'm working in a school, and as far as I know we don't have glacial acetic acid. (Also, acetic anhydride is prohibited, so I can't dry it.) So I might try direct union as blogfast25 proposed (if the iodine water/terbium mix didn't work.) Now I'll have to calculate hydration numbers and things (and I can't find a source for the hydration of TbI3, but I have found terbium chloride hexahydrate and terbium bromide hexahydrate mentioned in the abyss of the interwebs).

I was considering HI, but that stuff is hard to get or make and it raises eyebrows everywhere...

This is getting really complicated really fast.

On a different note, I found this short paper on a terbium-cobalt redox system in aqueous solution. Apparently terbium can be plated out of aqueous solution if cobalt is present. Quite unusual for such a reactive metal.

blogfast25 - 30-11-2013 at 07:12

B&F:

The GAA is easy to obtain and cheap. Buy it yourself from eBay.

The degree of hydration and solubility of TbI3 are presented in the paper linked to by ‘pok’, higher up. The REs form octahydrate iodides, so e.g. TbI<sub>3</sub>.8H<sub>2</sub>O. Fig. 1 tells us that for that salt, at saturation the mole fraction of TbI<sub>3</sub> in the saturated salt-water system (at 25 C) is about 0.065. You can then convert that to the more traditional g TbI<sub>3</sub> per 100 g of water, as outlined above.

The real problem with such high hydrates then is to get rid of the crystal water. For an iodide that would probably mean heating in vacuum to reduce the temperature at which the water leaves the crystal lattice and thus to prevent hydrolysis to a mess of hydroxy iodides. Oxygen needs to be avoided to avoid oxidation of the iodide to iodine.

Dehydration of the REX3 compounds is generally not easy. Of course your project might not require anhydrous TbI<sub>3</sub>, I don’t know that.

[Edited on 30-11-2013 by blogfast25]

Some success! But...

Brain&Force - 4-12-2013 at 16:15

My whole post (quite a long one) got deleted when I previewed it. I'm not going to bother writing it up again (unless you want me to).

In short, I made some terbium iodide through direct union and heating. The product was confirmed by hydrolyzing a solution of it, precipitating terbium(III) oxide and some higher oxides. I didn't expect this to happen on hydration.

Enjoy the picture of terbium ions fluorescing green in solution.

terbium fluorescence.png - 1.5MB

I didn't expect this to occur. After watching mrhomescientist's video on making terbium nitrate, I found a comment stating that the salt didn't fluoresce in solution. Perhaps the anion can make a difference (there was iodide and triiodide in solution.) I'll try terbium chloride and/or the acetate.

Better pics to come soon.

[edit] Blogfast, thanks for the note on LnI3 hydration. But I cannot access the paper.

[Edited on 5-12-2013 by Brain&Force]

deltaH - 5-12-2013 at 06:03

Fascinating, congratulations on carrying out a direct union. When you feel up to it, I am sure we would love to hear how you did this! Sorry about the deletion :(

blogfast25 - 5-12-2013 at 11:35

Quote: Originally posted by Brain&Force  
My whole post (quite a long one) got deleted when I previewed it. I'm not going to bother writing it up again (unless you want me to).



Can I insist on a short version of a write up? Tip: always save something longish in your favourite word processor before submitting here.

I'm surprised it seems to hydrolyse so easily and am wondering if you're perhaps seeing something else. If there's a next time, try acidifying the solution a little with acetic acid (I'm assuming the acetate is very soluble, so won't crystallise when you want to try and recrystallize the TbI<sub>3</sub>;).


TbCl<sub>3</sub> hydrate:

For TbCl<sub>3</sub>, you can dissolve the Tb directly in 37 % HCl.

At RT the mole fraction of TbCl3 in a saturated solution is about 0.06 (acc. that paper). That works out at 94.1 g TbCl3 / 100 g water (the MM of TbCl3 is 265.28 g/mole).

Dissolving stoichiometric amounts of Tb in HCl 37 % would give you about 142.3 g TbCl3 / 100 g water (assuming no loss of acid or water), so already above the saturation point. On cooling quite some TbCl3 hydrate should form, because much water will be tied up in the hydrate.

To dissolve the Tb in the acid, add small lumps of the metal to the acid, expect much effervescence and heat. Cool between additions if needed (probably, at least at first!)

And take pictures! You don't get to see that kind of thing everyday!



[Edited on 5-12-2013 by blogfast25]

MrHomeScientist - 5-12-2013 at 11:56

Quote: Originally posted by Brain&Force  
In short, I made some terbium iodide through direct union and heating. The product was confirmed by hydrolyzing a solution of it, precipitating terbium(III) oxide and some higher oxides. I didn't expect this to happen on hydration.
...
I didn't expect this to occur. After watching mrhomescientist's video on making terbium nitrate, I found a comment stating that the salt didn't fluoresce in solution. Perhaps the anion can make a difference (there was iodide and triiodide in solution.) I'll try terbium chloride and/or the acetate.


Awesome work! I'm glad you saw my video, too. When I made the nitrate, it definitely did not fluoresce while in solution - only the crystals. To make sure I understand your process (I'd love to see a writeup too): you heated elemental iodine and terbium together to form the triiodide, then added this to water to precipitate terbium oxides. Are these somewhat soluble? I.e. how did terbium ions get into solution to enable it to glow?

blogfast25 - 5-12-2013 at 13:23

Quote: Originally posted by MrHomeScientist  
To make sure I understand your process (I'd love to see a writeup too): you heated elemental iodine and terbium together to form the triiodide, then added this to water to precipitate terbium oxides. Are these somewhat soluble? I.e. how did terbium ions get into solution to enable it to glow?


The hydrolysis, if that is what really took place, is only partial. So some Tb hydroxychloride/Tb hydroxide may have formed but there will be plenty solvated Tb<sup>3+</sup> left.

Unless oxygen got to the Tb during the direct union, of course. W/o vacuum/argon that cannot be excluded a priory.


[Edited on 5-12-2013 by blogfast25]

blogfast25 - 6-12-2013 at 13:19

Checking my calcs on the dissolution of Tb in HCl 37 w%, it works out (as above) for one mole of Tb you need 296 g of HCl 37 w% (stoichio amount, in reality you need a bit more, of course), giving 1 mole of TbCl3 dissolved in 186.4 g water, or 10.3 mole of water. The mole fraction X<sub>T,S</sub> of TbCl3 in solution is thus 1 / (1 + 10.3) = 0.088, post reaction.

From here can also be estimated how much of the TbCl3.6H2O will crystallise out on cooling to RT, where X<sub>T,S</sub> = 0.06, from:

n<sub>T,C</sub> = [(n<sub>T</sub> + n<sub>W</sub>;) . X<sub>T,S</sub> - 1] / [(1 + 6) . X<sub>T,S</sub> - 1 ]

With:

n<sub>T,C</sub> number of moles of TbCl3 in the solid (crystals) phase

n<sub>T</sub> total number of moles of TbCl3 in the system (i.e. 1)

n<sub>W</sub> total number of moles of water in the system (i.e. 10.3)

X<sub>T,S</sub> = 0.06, mole fraction of TbCl3 in solution

This works out at n<sub>T,C</sub> = 0.555

So even without chilling, at about RT, more or less half of the TbCl3 should crystallise out as TbCl3.6H2O. Not bad at all...



[Edited on 6-12-2013 by blogfast25]

An interesting finding

Brain&Force - 6-12-2013 at 18:18

Warning: long post.
Quote: Originally posted by blogfast25  

Can I insist on a short version of a write up? Tip: always save something longish in your favourite word processor before submitting here.

I'm surprised it seems to hydrolyse so easily and am wondering if you're perhaps seeing something else. If there's a next time, try acidifying the solution a little with acetic acid (I'm assuming the acetate is very soluble, so won't crystallise when you want to try and recrystallize the TbI<sub>3</sub>;).

...

To dissolve the Tb in the acid, add small lumps of the metal to the acid, expect much effervescence and heat. Cool between additions if needed (probably, at least at first!)

And take pictures! You don't get to see that kind of thing everyday!



[Edited on 5-12-2013 by blogfast25]


Tip taken. And the write-up will be included.

Terbium halides (and most other lanthanide salts) hydrolyze very easily on heating. The thing is, I boiled the solution very quickly, and a lot of terbium iodide remained in solution.

Well, you read my mind about the TbCl3. I've already done the reaction and taken pictures. You won't see this in the picture, but at high concentrations of Tb3+, the solution becomes yellow; it's the color seen in Pok's post in versuchschemie. I only used 1M HCl and got the result. 37% HCl should show an even stronger yellow. I don't speak German, but from what I understood from a machine translation, the color was thought to be a contamination. It seems to be inherent to the terbium(III) ion at high concentration. I need to test this solution - it may have some of the absorption band effects of other rare-earth salts.

So here's what I did:

A small test was done with a few small bits of terbium added to a test tube. A large excess of iodine was added in the expectation that most of it would vaporize. The test tube was heated, and a huge puff of iodine vapor formed, with what appeared to be iodine droplets rolling down the test tube.

Note to blogfast25: I don't have any fancy vacuum stuff or argon, so I just heated them together. There was oxygen in the tube, but terbium seems unusually reluctant to react with oxygen outside of a flame (I just heated terbium in a test tube and there was no visible reaction, but it burns readily, similarly to burning cerium).

At the bottom of the test tube, some stuff had accumulated. It looked like a salt, but had a metallic sheen. I don't know if this was terbium iodide or just iodine - the color was not clear. It may have been terbium triiodide (which refers to Tb(I3)3, just to clear up any confusion). Unreacted iodine coated the test tube.

In order to test if the terbium had actually reacted to produce terbium iodide, I dissolved the solution in water. I forgot to test this solution for fluorescence. No insoluble particles were suspended (so likely no oxides formed), but the solution became very dark, likely due to triiodide formation. The solution was heated in the test tube, and it bubbled violently (a test with iodine aqueous solution only showed light bubbling and iodine vapor emission).

After removal, the solution had become lighter and mixed terbium oxides precipitated. This solution is the one that fluoresced.

In the fluorescence trials, I've discovered something very interesting.

A test tube with only iodine solution in water shows no fluorescence and no green color in the light. A test tube with terbium chloride also shows no fluorescence - it looks absolutely clear.

The test tube that fluoresced contained a mixture of terbium iodides and triiodides, with precipitated terbium(III) and terbium(IV) oxides. In order to rule out suspended oxides and hydroxides as the source of the fluorescence, I added 1M HCl to a different test tube containing the solution. The oxides dissolved and the fluorescence was still observed, and looked much clearer due to the lack of mixed oxides.

So this implies that the fluorescence is due to an iodine species - either iodide, triiodide, or possibly iodate. I'll report back soon. My next test will be to add KI or NaI to TbCl3 and check for any fluorescence.

Some non-experimentation notes:

MrHomeScientist - Thanks for replying! Your posts and video inspired my project.
I don't have a YouTube account, so I'll tell you this here - the black stuff that formed in the terbium nitrate synthesis is a mixed terbium(III,IV) oxide, similar to manganese dioxide. The stuff appears to be a good catalyst for the ignition of coal gas and the contact process for sulfuric acid production. I hope you kept it - it's not a waste product by any means.

blogfast25 - I made a mistake when I said I cannot access the paper; I meant that the link is broken.
And thanks for the calculations, they are pretty comprehensive.

tl;dr: Stuff worked.

blogfast25 - 7-12-2013 at 07:21

B&F:

Firstly, if I ever mentioned vacuum in the context of Tb/I<sub>2</sub> then I must have been foolish: only Argon would work here because the iodine is too volatile.

Secondly, acc. wiki, solid terbium sulphate strongly UV fluoresces green. All RE sulphates are basically very poorly soluble in water, so adding sulphuric acid to any RE solution will precipitate them. Worth doing if the fluorescence is what you want to study.

Also, the double salts RE<sub>2</sub>(SO<sub>4</sub>;)<sub>3</sub>.K<sub>2</sub>SO<sub>4</sub>.nH</sub>2</sub>O (n is 2 or 3, can't remember right now) are very poorly soluble and can be used to separate REs from 'stuff' or just to recover odds and sods of RE solutions. This is done by saturating the RE solution with potassium sulphate.

Both the sulphate and the double sulphates can be turned back into hydroxide by treatment with strong NH3. The sulphate, ammonia/ammonium and potassium can then be washed out easily and the RE(OH)3 redissolved in the acid of your choice. I've done this quite a few times by now with Nd. Dissolving fairly dry RE(OH)3 (suction dried, for instance) in strong HCl is another way of preparing a strong solution of RECl3.

Lastly, it's possible your terbium is contaminated: it's not supposed to display the mechanical properties you described. A test for iron (III) with thiocyanate could be useful, as Fe is a common fellow traveller of REs. Incidentally, the yellowish colour of Fe(III) might even explain the yellow you mentioned (although it would take rather a lot of Fe and it would first form as Fe(II)). So it would be strongly recommended to rule out Fe with K or NH4 thiocyanate solution, for peace of mind.


[Edited on 7-12-2013 by blogfast25]

Brain&Force - 7-12-2013 at 20:45

The vacuum thing was my creation, sorry about that.

What I found interesting is that the fluorescence of chlorides (and according to MrHomeScientist, nitrates) doesn't show up in solution. The problem I'm starting to have with my theory of iodide complexing with terbium to form some other ion in solution is that iodide is not a very good complexing agent, and would likely have a higher free energy of formation than, say, chloride, or even water itself.

I'll do the thiocyanate test for iron when I can experiment. I don't know if I can get ahold of thiocyanate, though. Metallium gives a minimum purity of 99.5% for terbium, so I doubt that there's a significant amount of Fe in the sample.

I've noticed that my project has strayed significantly from its original intent - studying triboluminescence of hexakis(antipyrine)terbium(III) iodide and substituted derivatives. Part of the difficulty (and the interesting results) have arisen solely from the synthesis of terbium iodide (and I still have to find a way to get rid of excess iodine without hydrolyzing the product).

blogfast25 - 8-12-2013 at 06:39

Quote: Originally posted by Brain&Force  
[...], and would likely have a higher free energy of formation than, say, chloride, or even water itself.

[...]

Metallium gives a minimum purity of 99.5% for terbium, so I doubt that there's a significant amount of Fe in the sample.



The free energy of formation values of most simple ions in water are tabled, so just search. But I'm not sure how this is to impact complexing? TbI3 in water shouldn't be complexed I think...

0.5 % impurity could be enough to make the metal harder than the pure substance. Iron is quite a common contaminant, that's why I suggest testing for it.

More findings

Brain&Force - 12-12-2013 at 20:20

More fun results...
The solution that "fluoresced" actually didn't. It was caused by triiodide ions in solution. As the concentration of triiodide in solution increases, the UV part of the spectrum fades, and the remaining visible spectrum becomes apparent. I'm using an LED diode.
blogfast25, I tried the thiocyanate test and I got a negative result, with the solution remaining almost clear (barely yellow from Tb). A very dilute solution of ferric sulfate (I don't know exactly how much, it was just a few grains) gave a blood red color. Now I know what I'm doing for Halloween next year...:D
The solution has a very slightly greenish yellow tint with no turbidity. Addition of a few drops of HCl solution causes the complex to become perfectly transparent. It's very odd and has nothing to do with dilution. My best guess is that a colorless complex [TbCln]3-n exists in solution, and with high concentrations of Tb3+, the cations form yellow some-number-aquaterbium(III) ions. (Terbium can vary its coordination number from 6 to 9) The concentration of terbium in solution can't be greater than 0.33 molar (I started with 1M HCl). How could I determine the composition of a complex ion in solution?
Of course, I could be wrong, feel free to suggest your theories.
This complex MAY be fluorescent. I can't be sure; I'm attempting to make more TbCl3 and getting a better UV light.

Note: I made a larger amount of the yellow stuff, but I unintentionally destroyed it with HCl.


Tb-in-HCl.jpg - 415kB


yellow-Tb.jpg - 273kB

blogfast25 - 13-12-2013 at 06:11

B&F:

I've never really heard of coordination complexes of the f-block elements. Most d-block elements that form them are amphoteric to some extent and the REs aren't amphoteric. But I could be wrong on complexes. Seems to me the f-orbitals don't lend themselves to bonding ligands very well. Again, I could be completely off the mark here. Will google a little later on.

What's happening in pic 1? Tb in HCl?

[Edited on 13-12-2013 by blogfast25]

DraconicAcid - 13-12-2013 at 09:28

Quote: Originally posted by blogfast25  
B&F:

I've never really heard of coordination complexes of the f-block elements. Most d-block elements that form them are amphoteric to some extent and the REs aren't amphoteric. But I could be wrong on complexes.


Most transition metals will form complexes- you just have to pick the right ligand. Right off the top of my head, copper, nickel, silver and iron aren't amphoteric (I've heard that copper is, but the solution must be screamingly basic for it to show any sign of being so), but readily form complexes with ammonia, ammonia, ammonia and halide ions (respectively).

For the lanthanides, Cotton & Wilkinson tell me that the most stable complexes are those with oxygen-binding ligands (EDTA, citrate, tartrate, carbonate, acac, etc.), and the most common coordination numbers are 7, 8, and 9.

Relevant to this discussion is the following tidbit (C&W, Advanced Inorganic Chemistry, 4th ed., p 992):

"Halogeno complexes. In aqueous solutions rather weak complexes MF2+(aq) are formed with fluoride ion, but there is little evidence for complex anion formation; this is a distinction as a group from the actinide elements, which do form complexes in strong HCl solutions. However by the use of nonaqueous media such as ethanol or acetonitrile, salts of the weak complexes MX63- can be prepared. The iodo complexes are exceedingly weak, dissociating in nonaqueous solvents even in the presence of an excess of I-, and they are attacked by moisture and oxygen."

blogfast25 - 13-12-2013 at 10:07

DA:

Thanks. Cu is definitely amphoteric (can be shown quite easily at home level) and Fe forms (instable) ferrate (VI). But I agree it's not a very strong correlation.

In your quotation, does M stand for a lanthanide? I had certainly forgotten about EDTA as a complexing agent for REs. EDTA complexometry is one way of determining RE solutions.

But RE chloride complexes: probably not. And certainly not iodo complexes.

DraconicAcid - 13-12-2013 at 11:45

Quote: Originally posted by blogfast25  
In your quotation, does M stand for a lanthanide? I had certainly forgotten about EDTA as a complexing agent for REs. EDTA complexometry is one way of determining RE solutions.

But RE chloride complexes: probably not. And certainly not iodo complexes.

Yes- that's from the section on lanthanides. Chloride complexes- only in nonaqueous solvents, iodo complexes, even then, just barely. They will also form complexes with chelating N-donors (en, etc.), and thiocyanate.

Brain&Force - 17-12-2013 at 21:06

I was able to dry some TbCl3 in a vacuum dessicator, but only partially. It displayed a strong yellow color, similar to that of a fluorescent highlighter. It looks the same in sunlight and tube lighting. (I'll need to try CFL lighting as well because that produces the most dramatic color shift.) However, no fluorescence could be observed. I have a feeling water may be able to quench the fluorescence of terbium compounds - after all, terbium chloride is supposed to fluoresce. I'll test nitrate as well.

blogfast25: Pic 1 is Tb in HCl. The complex isn't visible until after the HCl solution is depleted.

Is it possible that this yellow color is TbOCl (terbium oxychloride) forming? This may be possible, given the complex appears to turn clear in even the slight amount of acid. If this were a chloro complex of terbium, I would expect the color to change only slightly as progressively larger amounts of HCl are added. I haven't tried redissolving in pure water - maybe that will confirm the result.

From woelen's site, praseodymium oxychloride appears to be only slightly, if at all, soluble in water. This complex is soluble in water - so either TbOCl is soluble or there's something else going on.

Perhaps the answer may lie in this thread? Perhaps different preparations can produce different colors.

Pic 1: complex after drying. It was still wet, I'm going to vacuum it some more.
Pic 2: I decided to dissolve it in a 0.3 molar TbCl3 solution with some excess HCl, originally clear, and the solution turns very pale yellow.

wet terbium chloride.jpg - 435kB Solution to dry.jpg - 408kB

Brain&Force - 21-12-2013 at 10:12

I dried the solution further and ended up getting yellow flakes of what appears to be terbium(III) chloride. Again, it should be white, but I have no clue as to what's causing the yellow color. There was iodine in the dessicator after a spill occured, but that can be ruled out because the color of the solution was yellow to begin with. A test for fluorescence came up mostly negative, but there were a few flakes that glowed greenish. The rest took on a very dim yellow color. (I've heard dysprosium has yellow fluorescence, but I have no idea if there really is any.) As expected, the flakes are paramagnetic, but only slightly.
In CFL light, the color does not change at all. It's still yellow.

Oddly enough, it smells like oranges. This is really unusual - the compound shouldn't be volatile at all. The only volatile compound I could fathom exists would be terbium(IV) chloride, but I doubt HCl can produce this. I can't be sure if it's the compound itself or some volatile substance that was in the vial earlier.

Now I need to figure out what this yellow stuff is...

Here are two photos of the compound - one it normal light, the other in UV. During the vacuuming, some glass in the dessicator chipped off, and there are fragments on the watch glass.

dry terbium chloride.jpg - 773kB tb in uv.jpg - 528kB

blogfast25 - 21-12-2013 at 14:16

B&F:

I doubt if TbOCl2 can form in these conditions. These halides are just not that prone to hydrolysis.

How pure is your HCl? Iron is very common in garden variety strong HCl. That picture looks so much like some white chlorides I've prepared with hardware store 'muriatic acid'!

Try also to add H2SO4 to the solution to see if the sulphate (hydrate) precipitates and whether it fluoresces (strongly green under UV, acc. Wiki).

Can I borrow (with credit to you) that picture of Tb dissolving in HCl?


[Edited on 21-12-2013 by blogfast25]

Brain&Force - 22-12-2013 at 21:58

You can use the picture. Where is it going to be used? If you need a larger one, I can send you the full size version.
The hydrochloric acid is not hardware store acid - it's 1.0 molar acid, source unknown, but definitely from one of those large chemical suppliers. I made Tb(NO3)3 and it still showed the same color. I also added hydrogen peroxide and tested for thiocyanate so that any Fe2+ would be oxidized to Fe3+, but there is still no red color.

[edit] I've ruled out the following as contaminants:
Iron, lanthanum, praseodymium, neodymium, holmium, erbium, thulium, ytterbium, lutetium.

These are possibly contaminants:
Cerium, samarium, europium, gadolinium, dysprosium (they all form salts that have some sort of yellow tint).
I think cerium(IV) may be a contaminant in my sample - cerium(IV) sulfate is yellow, so other cerium salts may be mixed. However, I highly doubt CeCl4 is stable.

Any suggestions?

[Edited on 23-12-2013 by Brain&Force]

blogfast25 - 23-12-2013 at 05:46

Ce(IV) oxidises Cl<sup>-</sup>, IIRW.

I'll use the photo on my blog.

Brain&Force - 25-12-2013 at 19:06

There's an interesting finding I forgot to mention in this thread. In the solution that appeared to fluoresce but didn't, there were terbium(III) and triiodide ions, and terbium(III) oxide had precipitated out. Over the next few days, the brown color faded to yellow, and eventually the color became clear. Apparently iodine is strong enough to oxidize terbium(III) oxide to terbium(III,IV) oxides.

I'll make some Tb2O3 and see if I can repeat it.

blogfast25 - 26-12-2013 at 06:57

Quote: Originally posted by Brain&Force  
Apparently iodine is strong enough to oxidize terbium(III) oxide to terbium(III,IV) oxides.



That is of course only one possible explanation but not very plausible IMHO.

From my latest gadget, the CRC Ed. 86:

Tb<sup>3+</sup> === > Tb<sup>4+</sup> + e, E<sub>Ox</sub> = - 3.1 V
I<sub>2</sub> + 2 e === > 2 I<sup>-</sup>, E<sub>Red</sub> = + 0.5355 V

This makes oxidation of Tb(III) to Tb(IV) by iodine very unlikely.

Is it possible that some I<sub>2</sub> or I<sub>3</sub><sup>-</sup> got chemisorbed onto the surface of the oxide/hydroxide and that the iodine later simply oxidised by air (to hypoiodate/iodate)?


[Edited on 26-12-2013 by blogfast25]

DraconicAcid - 26-12-2013 at 12:07

Quote: Originally posted by blogfast25  

From my latest gadget, the CRC Ed. 86:

Tb<sup>3+</sup> === > Tb<sup>4+</sup> + e, E<sub>Ox</sub> = - 3.1 V
I<sub>2</sub> + 2 e === > 2 I<sup>-</sup>, E<sub>Red</sub> = + 0.5355 V

This makes oxidation of Tb(III) to Tb(IV) by iodine very unlikely.


You can't apply solution potentials to oxides.

A half reaction such as Tb(OH)3(s) + OH- -> TbO2(s) + 2 H2O + e- will have a potential completely unrelated to the one you gave above.

[Edited on 26-12-2013 by DraconicAcid]

blogfast25 - 26-12-2013 at 14:19

DA:

I wouldn’t say ‘completely unrelated’. Break it down:

Tb(OH)<sub>3</sub> < == > Tb<sup>3+</sup> + 3 OH<sup>-</sup>…. (1)

Tb<sup>3+</sup> === > Tb<sup>4+</sup> + e …. (2)

Tb<sup>4+</sup> + 4 OH<sup>-</sup> === > TbO<sub>2</sub> + 2 H<sub>2</sub>O …. (3)

To the value for (2) will have to be added terms to account for (1) and (3).


[Edited on 26-12-2013 by blogfast25]

Brain&Force - 26-12-2013 at 20:55

The fact that terbium forms mixed oxides (Tb4O7 and Tb6O11) may complicate matters further. However it's not known whether the compound is stoichiometric (like aluminum oxide) or interstitial (like manganese dioxide, which this oxide strongly resembles). Pure TbO2 can't be prepared easily - you have to either expose it to atomic oxygen or reflux it with a solution of HCl in acetic acid.

Also, WebElements gives potential of Tb(OH)3 + OH- → TbO2 + e- as -0.9. Reduction is possible, but probably not with I2. Interestingly, WebElements gives potentials for DyO2 and HoO2 (I can't find any references to these compounds).

However, the fact that no soluble terbium(IV) compounds form may drive equilibrium to the right in this case.

blogfast25 - 27-12-2013 at 05:20

Quote: Originally posted by Brain&Force  
However, the fact that no soluble terbium(IV) compounds form may drive equilibrium to the right in this case.


Actually, the formation of ionic solids would drive it to the right, because of the release of lattice energy (although it also decreases entropy). And if no solids are formed (Tb(IV) in solution) then surely the value for (2) stands?

Whether iodine can or cannot oxidise Tb(III) to whatever will need to be determined experimentally, as you suggested. Not easy, since as iodine is itself prone to oxidation (in alkaline medium). I would mix a sub-stoichio amount of KI<sub>3</sub> solution with Tb(OH)<sub>3</sub> and see...

Intuitively though, the more I think about it the more the oxidation of Tb(III) with iodine sounds implausible.

[Edited on 27-12-2013 by blogfast25]

Brain&Force - 27-12-2013 at 16:25

I found an article, written in 1905, that details several terbium compounds - namely, the chloride, bromate, perchlorate, carbonate, citrate, meconate, formate, acetate, and carbide. They didn't have ion exchange technology, so I'm guessing there were some significant impurities (likely LREEs) in the samples. Nevertheless, it describes terbium chloride as producing a clear and colorless solution. I'll try to find some cleaner HCl to continue experimentation.

I got some permanganate and citric acid for Christmas, so I'll test out two things:
Whether permanganate can oxidize Tb2O3. If iodine can, then permanganate should be able to as well.
Whether terbium citrate is fluorescent - and determine its color.

Is it possible for iodine to have contaminated the TbCl3? It looks similar to some polyhalides. I have some in a vial, so I'll leave it uncapped and see what happens.

And apparently neodymium can form a complex with HCl (scroll to the bottom).

blogfast25 - 28-12-2013 at 06:17

B&F:

If by Tb<sub>2</sub>O<sub>3</sub> you mean Tb(OH)<sub>3</sub> , then permanganate should not be able to oxidise it. For one, the hydroxide can only exist in neutral/alkaline conditions and in these conditions the reduction of permanganate leads to the dreaded MnO<sub>2</sub>(s).

In acid conditions Tb would exist as Tb<sup>3+</sup> but the reduction potential of Mn(VII) === > Mn(II) is then + 1.51 V, not enough if my value of – 3.1 V is correct. But it’s worth trying because its colourimetry in action! But you have to make sure you've got the quantities right because permanganate is so intensely coloured it can obscure everything. Excess, unreacted permanganate could lead to erroneous conclusions

If your HCl is Fe(III) free then it should be good enough for your purposes. You can ‘recondition’ any contaminated ods and sods of Tb solutions by first filtering them, then precipitate as hydroxide from dilute solution with ammonia solution, filter off hydroxide, wash filter cake profusely, then redissolve in minimum amount of HCl.

Re, Nd/HCl complexes, where does he write that? All I can find is the fairly vague:

”Although the lanthanide elements are said to have much less pronounced color shifts due to complex formation, when compared with the transition metal elements, the effect of a small shift is much stronger.”

Re, polyhalides, that's a good find. I didn't know they were so easy to prepare. So it can't be excluded out of hand that some ICl<sub>4</sub><sup>-</sup> was formed...


[Edited on 28-12-2013 by blogfast25]

Brain&Force - 28-12-2013 at 10:04

Hmm...I forgot MnO2 would form. Considering the similarity of Tb4O7, it wouldn't be easy to tell if something else had oxidized the permanganate.

But DraconicAcid has a point about oxide reduction potentials. The potential Fe2+ + 2e- → Fe is -0.44 V, but the potential for Fe(OH)2 + 2e- → Fe + 2OH- is -0.89 V. These should seem related, but they're quite different. Terbium(III,IV) oxide is a lot more stable than the reduction potential suggests. I believe Tb4O7 forms when terbium is burned, but I'll have to test again.

The Nd-Cl- complex is mentioned on the bottom; there is a preparation of NdCl3 with dilute HCl and concentrated HCl.

I came up with this reaction to produce ICl4-:

I2 + 4Cl- → ICl4- + 3I-

so likely not much ICl4- has been formed. But this appears to be sufficient to color the crystals yellow.

The polyhalide theory explains a lot:
Why the compound smells. Strangely, the smell does not resemble iodine or chlorine - it's similar to tangerines. I'm thinking the large terbium cation can stabilize this ion to the point that it does not decompose much. That would also explain why there is no change in color after leaving the bottle open overnight.
The unexpected yellow color with relatively pure HCl.

I got the idea from this thread. I had been storing iodine in the fume hood with the test tube rack. Won't be making that mistake again.

blogfast25 - 28-12-2013 at 11:24

Your point about the differing reduction potentials boils down to what I wrote above:
Quote: Originally posted by blogfast25  

Tb(OH)<sub>3</sub> < == > Tb<sup>3+</sup> + 3 OH<sup>-</sup>…. (1)

Tb<sup>3+</sup> === > Tb<sup>4+</sup> + e …. (2)

Tb<sup>4+</sup> + 4 OH<sup>-</sup> === > TbO<sub>2</sub> + 2 H<sub>2</sub>O …. (3)

To the value for (2) will have to be added terms to account for (1) and (3).


The oxidation potential corresponds to a change in Gibbs Free Energy, ΔG. In the case of oxidation of Tb(III) to TbO<sub>2</sub> (or a mixed oxide), Hess’ Law allows to split it into three terms:

ΔG = ΔG<sub>1</sub> + ΔG<sub>2</sub> + ΔG<sub>3</sub>

ΔG<sub>1</sub> will be positive, because it’s the precipitation of the hydroxide that is spontaneous, not dissolution.

ΔG<sub>2</sub> corresponds to the oxidation of ‘naked’ Tb(III).

ΔG<sub>3</sub> is likely to be strongly negative because it involves the spontaneous precipitation of an ionic solid.

On balance, ΔG is probably more negative than ΔG<sub>2</sub> alone, making the oxidation thermodynamically more favourable, done that way.

It would be worth trying simple thin bleach: hypochlorite is a very powerful oxidising agent, in alkaline conditions. And no MnO2 in sight…

A better analogy would be Fe2+/Fe3+ and Fe(OH)2/Fe(OH)3. Or even Fe2+/Fe(OH)3. Ferrous stuff does oxidise much faster in neutral to alkaline conditions, I think...

[Edited on 28-12-2013 by blogfast25]

Eddygp - 29-12-2013 at 04:30

This should really be in bfesser's Topical Compendium, under terbium. Very interesting topic. In fact, I like the lanthanides' chemistry as it is fascinating. I'm going to try a similar synthesis soon, with ytterbium (hopefully).

blogfast25 - 29-12-2013 at 06:19

Quote: Originally posted by Eddygp  
I'm going to try a similar synthesis soon, with ytterbium (hopefully).


That would be very interesting. Don't forget to include some good quality photos!

We should take advantage of the increasing availability of the REs: when I was a student the lanthanide block was like a beach you could only dream of, only the rich and powerful could visit it.

Brain&Force - 30-12-2013 at 10:28

Speaking of bfesser, can a mod change this thread's title to something like "General discussion of terbium compounds?" This is a whole lot more than just the synthesis of terbium iodide.

Eddygp, if you are going to work with ytterbium, tell us what color the pure metal is. There was an edit war on Wikipedia on whether ytterbium is silvery or slightly golden, and it's still not clear as to what color it is. As one of the users states, it's like the debate over the color of cesium - almost all references before 1984 list it as silver.

And don't forget to try a pyro mixture with it - ytterbium burns very bright green, purer than copper (bluish-green) or barium (slightly yellowish-green). If you have equipment for it you can also see the strong IR emission of burning Yb metal. Militaries are investigating its use in decoy flares because it can emit more IR than a Mg based flare (and the formed Yb2O3 is more emissive in the IR spectrum than MgO).

Ytterbium has a lot of really interesting applications.

siegfried - 30-12-2013 at 10:55

All RE elements can be ordered from Metallium, Inc A 20 gm lump of Tb is $135 and a 5 gm piece is $32. Most oxides can be ordered at Smart Elements is Austria. Some RE compounds can be ordered on Ebay.

Eddygp - 30-12-2013 at 11:03

Quote: Originally posted by Brain&Force  
Speaking of bfesser, can a mod change this thread's title to something like "General discussion of terbium compounds?" This is a whole lot more than just the synthesis of terbium iodide.

Eddygp, if you are going to work with ytterbium, tell us what color the pure metal is. There was an edit war on Wikipedia on whether ytterbium is silvery or slightly golden, and it's still not clear as to what color it is. As one of the users states, it's like the debate over the color of cesium - almost all references before 1984 list it as silver.

And don't forget to try a pyro mixture with it - ytterbium burns very bright green, purer than copper (bluish-green) or barium (slightly yellowish-green). If you have equipment for it you can also see the strong IR emission of burning Yb metal. Militaries are investigating its use in decoy flares because it can emit more IR than a Mg based flare (and the formed Yb2O3 is more emissive in the IR spectrum than MgO).

Ytterbium has a lot of really interesting applications.

Ytterbium is silvery with a tinge of cream. Definitely not golden or brass.

[Edited on 30-12-2013 by Eddygp]

Brain&Force - 30-12-2013 at 12:07




Comparing thulium to ytterbium shows ytterbium is definitely yellowish. My friend calls it a champagne color. It reminds me of the color of that new iPhone that came out.

Back on topic...

Exposing the yellow terbium compound to air has led to no change - it's still as yellow as it was before. Perhaps I need to reduce the ICl4- to prove its existence.

blogfast25 - 30-12-2013 at 13:09

Quote: Originally posted by Brain&Force  

Exposing the yellow terbium compound to air has led to no change - it's still as yellow as it was before. Perhaps I need to reduce the ICl4- to prove its existence.


Try treating a bit of it with strong sulphite solution?

Eddygp - 31-12-2013 at 02:15

A bit on topic, are any lanthanides amphoteric?

blogfast25 - 31-12-2013 at 05:53

Quote: Originally posted by Eddygp  
A bit on topic, are any lanthanides amphoteric?


No, not remotely.

Brain&Force - 31-12-2013 at 15:12

I have a better idea regarding removal of ICl4- - just precipitate terbium as Tb(OH)3 and any ICl4- will be removed. The remaining Tb(OH)3 will then be reused to make pure TbCl3.

blogfast25 - 1-1-2014 at 05:33

Quote: Originally posted by Brain&Force  
I have a better idea regarding removal of ICl4- - just precipitate terbium as Tb(OH)3 and any ICl4- will be removed. The remaining Tb(OH)3 will then be reused to make pure TbCl3.


Sure but I was only referring to testing the ICl<sub>4</sub><sup>-</sup> hypothesis, not recovering the Tb(III).

Brain&Force - 8-1-2014 at 17:11

So break ended. I began to search for sulfite or thiosulfate, but nothing turned up. But I did find a vial of an unusual chemical - ammonium heptamolybdate. I had previously looked upon it as a curiosity, but then I realized I could prepare terbium molybdate, a compound that is poorly characterized.
I attempted to prepare Tb-molybdate through metathesis. Some solid ammonium molybdate was added to a solution of terbium chloride (a different solution that did not have any noticeable ICl4- contamination) and added the ammonium molybdate. The solution became milky white, but quickly became perfectly clear. I'm wondering if I made a complex between Tb and either molybdate or ammonia.
Also, I'm trying to make Tb-citrate to hopefully make sodium or potassium tris(citrato)terbate (is that even correct? I'm dead sure I'm wrong here). More to come soon.

Brain&Force - 9-1-2014 at 16:19

blogfast25, you were right to begin with.
There's iron in the terbium - what seems to be a large amount. I knew that could be the only cause because the terbium citrate I had made was as yellow as a school bus. I tried a thiocyanate test, this time with a different source (our school has some very old, degraded chemicals, and I'm sure I got an old batch), and the solution turned reddish-black. I did this test with all terbium sources and got the exact same result. How wonderful.
I'm attempting a very roundabout way to recover the remaining terbium in solution. Manganese pieces are reacted with the solution, replacing Fe, which precipitates as either Fe metal or Fe2O3. In solution are Tb3+ and Mn2+. Mn is removed by oxidation by acidified H2O2, and Tb is precipitated as hydroxide.
I'll try to get an accurate measurement on the amount of iron in solution soon.

blogfast25 - 10-1-2014 at 05:08

Quote: Originally posted by Brain&Force  
I did this test with all terbium sources and got the exact same result. How wonderful.
I'm attempting a very roundabout way to recover the remaining terbium in solution.


No, no. Use the sulphate method! RE sulphates are very poorly soluble but Fe (III) sulphate is very soluble. So add quite a bit of sulphuric acid to your contaminated Tb solution, make sure pH < 3. Tb (and other REs) sulphate precipitates, ferric sulphate remains in solution. Then filter and rinse filter cake with cold H2SO4 (say about 25 %).

Works: been there, done that, have T-shirt...

MrHomeScientist - 10-1-2014 at 12:16

Well, in the case of neodymium it's been a bit more complex than that. Regarding our work in processing neodymium magnets, the Nd-sulfate remains quite soluble along with Fe(III) sulfate. The way I've been separating is by heating to boiling, whereupon some Nd-sulfate precipitates out while the iron remains very soluble. Filter the pink crystals and wash with boiling water. After cooling down to room temp, usually lots of iron(III) has crystallized. This is removed, and the process repeated several times. I know you did a lot of work in separating the two, certainly beyond just adding a lot of sulfuric acid. So I don't think the general statement of "RE sulphates are very poorly soluble" is accurate. For this particular case with terbium, it may well be.

blogfast25 - 10-1-2014 at 12:40

Quote: Originally posted by MrHomeScientist  
So I don't think the general statement of "RE sulphates are very poorly soluble" is accurate. For this particular case with terbium, it may well be.


Mr HS, you're right, I'm not being very accurate (haste makes waste). The low solubility of Re sulphates is at 100 C. For Ce(III), Tb, Ce, La (and others) Wiki's solubility table shows that clearly.

I actually prefer to use the double sulphates with potassium, which are almost insoluble, also at low temperature.

Here a strongly acidic (to prevent Fe(OH)3 precipitating) RE solution is saturated with K2SO4. The double sulphates precipitate, the iron remains in solution. Wash filter cake with acidic K2SO4 solution. This is one industrial route.

Potassium is released by treatment of filter cake with NH3 solution, which then yields RE(OH)3.

Brain&Force - 11-1-2014 at 21:51

At this point, it's too late to continue experimentation - my project is supposed to be complete by the 24th. I already have a lot of new information (although some of it is incorrect now that I know that the Tb was contaminated).
However, I am attempting to determine the amount of iron in the terbium using the terbium potassium alum method. The reason I had decided to use my convoluted method was because I already had added KSCN to some of the mixtures, and from what you had stated earlier, blogfast25, I thought terbium potassium alum was highly insoluble and could not be precipitated as hydroxide with NH3. These recent posts say otherwise.
I read somewhere that NH3 should be used to precipitate Ln3+ and not NaOH/KOH. Am I correct? And if so why?
I may have stumbled upon another method of removing iron from terbium. More on that later.
Quote: Originally posted by blogfast25  
For Ce(III), Tb, Ce, La (and others) Wiki's solubility table shows that clearly.


Do you have a link to this table? I could use it for my project.

[edit] Correction: I will continue experimentation, it just won't be included in the writeup and other materials for the project. I still have a bit of Tb left.

[Edited on 12-1-2014 by Brain&Force]

blogfast25 - 12-1-2014 at 01:57

Quote: Originally posted by Brain&Force  
I read somewhere that NH3 should be used to precipitate Ln3+ and not NaOH/KOH. Am I correct? And if so why?
I may have stumbled upon another method of removing iron from terbium. More on that later.

Quote: Originally posted by blogfast25  
For Ce(III), Tb, Ce, La (and others) Wiki's solubility table shows that clearly.


Do you have a link to this table? I could use it for my project.



The REs don't from complexes with NH3, so I can't see an immediate impediment to using it. Plenty references cite its use with REs. Ammonium is easier to wash out of precipitates, or so I've read. It also leaves more easily on calcining, if the actual oxide is what one wants.

The double sulphates convert to hydroxides on treatment with NH3 easily because the Ksp of the hydroxides is much lower than that of the double sulphates.

Solubility table:

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

Glad to hear you'll carry on experimenting with your terbium/iron!

Brain&Force - 12-1-2014 at 17:21

The solubility table only lists the solubility of terbium at 20 C. By solubility table I thought you had a graphic, I had used that page before.

I just realized that the manganese extraction route has a flaw. The solution currently contains these ions: H+, Mn(II), Tb(III), K+, likely a small amount of Fe(III), complemented with chloride, thiocyanate, and a bit of triiodide. By adding H2O2, I may generate chlorine gas (through the formation of MnO2), oxidize triiodide to iodine or iodate, or have a whole host of other side reactions take place. However, it doesn't seem too late to just precipitate the terbium as terbium potassium alum, then produce terbium hydroxide with ammonia.

I figured out what was causing the smell of oranges. Several of the vials at my school smell like assorted fruits, and appear to have been used for some sort of demo. It has nothing to do with the Tb salt.

And about the new extraction method: From reacting terbium (actually ferroterbium) with citric acid, I discovered that a yellow precipitate formed. I can't find much information on ferric citrate (regarding solubility and color), but that's what i think the precipitate is. Terbium citrate seems to remain in solution. I'll continue looking into this. I'm surprised that terbium citrate did not precipitate.

And this paper indicates that terbium metal can be precipitated out of solution in the presence of cobalt and boric acid. If I can find a suitable power source I'll try to reform terbium metal. This is very unusual for an element of such electropositivity.

Wait, what? Strange result

Brain&Force - 13-1-2014 at 20:03

I've gotten an odd result out of the separation attempts. After allowing terbium to dissolve in HCl, a yellowish solution formed (of course, the Fe contamination). I had used too much terbium and the HCl I have isn't concentrated enough, so I decanted some of the solution into another test tube. However, after leaving the solutions to sit for a few hours, ferric hydroxide precipitated out - but not in the form I expected. There was powder at the bottom, but fluffy clumps of red powder were also suspended in the solution.
I think the terbium ion acted as a flocculating agent similarly to aluminum sulfate, and that caused the ferric hydroxide to precipitate rather than remain suspended. I'll re-acidify the solution and continue separation. At what concentration does Fe hydroxide become visible in suspension? That may explain the yellow stuff and why it clarifies so clearly when only a few drops of HCl are added. There may not be as much Fe contamination as I thought.

suspended fe(oh)3.png - 1.1MB

blogfast25 - 14-1-2014 at 05:59

Quote: Originally posted by Brain&Force  
I've gotten an odd result out of the separation attempts. After allowing terbium to dissolve in HCl, a yellowish solution formed (of course, the Fe contamination). I had used too much terbium and the HCl I have isn't concentrated enough, so I decanted some of the solution into another test tube. However, after leaving the solutions to sit for a few hours, ferric hydroxide precipitated out - but not in the form I expected. There was powder at the bottom, but fluffy clumps of red powder were also suspended in the solution.



Most likely explanation: your acid became depleted and wasn't capable of keeping the ferric chloride in solution. The latter would need a pH of less than 4, less than 3 even if there's quite a lot of Fe<sup>3+</sup>.

The K<sub>sp</sub> of Fe(OH)<sub>3</sub> is very, very low, which causes ferric ions to precipitate from about said pH values.

Ferric hydroxide has a tendency to peptise: go colloidal when ionic strength of the supernatant solution is low. I've seen such precipitates run right through a filter as if they were proper solutes!

As I wrote above, if you use the double sulphates method for separation of iron and terbium, make sure the solution and any filter cake washing solution is pH < 3, to avoid the iron precipitating...

[Edited on 14-1-2014 by blogfast25]
Quote: Originally posted by Brain&Force  

And this paper indicates that terbium metal can be precipitated out of solution in the presence of cobalt and boric acid. If I can find a suitable power source I'll try to reform terbium metal. This is very unusual for an element of such electropositivity.


It talks about Co/Tb alloys mainly, if I'm not mistaken.

[Edited on 14-1-2014 by blogfast25]

No precipitate?

Brain&Force - 14-1-2014 at 17:06

I reacidified the solution, the precipitate dissolved, and the yellow color disappeared, as expected. Then I added in K2SO4. That's when things started to get weird. Nothing precipitated out of solution. I thought I didn't add in enough potassium sulfate, so I dropped in several crystals, and allowed them to dissolve. They dissolved slowly (is it just me, or are all sulfates slow to dissolve?), but nothing precipitated out. I heated the solution to boiling and only a few crystals dropped out of solution. Considering I only added half a gram of terbium to the solution, and about a whole gram of potassium sulfate, I'm surprised only that tiny amount precipitated.

Perhaps the solution is not concentrated enough? It's probably only .3 molar terbium ion in solution. I couldn't find any 6M HCl. I'm afraid to boil the solution in case hydrolysis occurs.

blogfast25 - 15-1-2014 at 05:29

Quote: Originally posted by Brain&Force  
Considering I only added half a gram of terbium to the solution, and about a whole gram of potassium sulfate, I'm surprised only that tiny amount precipitated.

Perhaps the solution is not concentrated enough? It's probably only .3 molar terbium ion in solution. I couldn't find any 6M HCl. I'm afraid to boil the solution in case hydrolysis occurs.


0.3 M is quite low. What was the volume of solution? Hydrolysis won't occur if there's enough acid present.

You really need more concentrated HCl to carry on. Try to get 'Patio cleaner'? That's 15 - 20 %, I think.

But while searching for the original reference for the separation of iron from REs using double sulphates, I came across this reference:

http://www.kth.se/en/che/divisions/transport_phenomena/resea...

“One of the most widely used precipitation method to separate rare earth elements (REE) from acidic solutions is by precipitation of sodium double sulphate hydrates (NaRE(SO4)2.xH2O) through the addition of sodium sulphate (Gupta 1992). These salts are only slightly soluble in acidic solutions. Double sulphate precipitation result in separations in two fractions one enriched in the light-REE group rare earths and the other enriched in the heavy-REE group rare earths (Gupta, 1992). The different REE can then be separated from each other by converting the double sulphate into a highly soluble compound, such as RE-hydroxide. Although monazite leach liquor is usually purified by precipitation of REE as sodium double sulphate hydrates the influence of the process variables on the double sulphate precipitation and on its conversion into RE-hydroxide is rarely mentioned in literature (Abreu and Morais, 2010).”

… “in two fractions one enriched in the light-REE group rare earths and the other enriched in the heavy-REE group rare earths”, this maybe the snake in the grass here: perhaps Tb doesn’t precipitate as eagerly or fully as e.g. Nd (for which I used the method)?

One side point: my original reference advised K2SO4 because that double sulphate is even less soluble, so the use of K2SO4 shouldn’t be the cause of the problem.

Oxalates: I described elsewhere on this forum practical experimentation using oxalates to separate Fe and RE (all RE oxalates are exceedingly insoluble, with very low Ksp). It involves adding an excess oxalic acid, than potassium hydroxide or potassium oxalate. The RE drops out as RE2Ox3, the iron remains soluble as K3FeOx3 (potassium trisoxalato ferrate (III), a grass green, slightly fluorescent complex). Drawback: the RE oxalate is harder to convert back into something soluble. I'll try and find that work now.


Here's the oxalate method at work (by MrHS) on neodymium/iron from some large magnets:

http://www.sciencemadness.org/talk/viewthread.php?tid=14145&...

[Edited on 15-1-2014 by blogfast25]

Brain&Force - 15-1-2014 at 18:01

I finally got my hands on 6M HCl (iron free and perfectly colorless) and added about 10 mL to the solution. The solution turned greenish-yellow, almost the color of highlighter ink. Evaporating the solution to 50 mL (was about 75 mL initially) resulted in no precipitation or hydrolysis.

I can't get my hands on oxalates (someone is taking all the chemicals out of our classroom), and they're not a good idea with terbium. Tb oxalate converts to a higher oxide when it is heated, and the product cannot be dissolved easily - it takes a week and generates large amounts of chlorine gas.

Right now I think the best course of action is to boil the solution dry and redissolve in 6 molar HCl. That should keep the Fe(III) salts in solution and leave the Tb sulfate or terbium potassium alum out.

Here is the acid after adding extra HCl. Note that the color just suddenly appeared after a few seconds after adding the acid.

yellow solution.png - 1.1MB

[edit] Terbium sulfate is only slightly soluble in water: http://www.youtube.com/watch?v=DJcObFauFIc

[Edited on 16-1-2014 by Brain&Force]

blogfast25 - 16-1-2014 at 05:30

Quote: Originally posted by Brain&Force  

[edit] Terbium sulfate is only slightly soluble in water: http://www.youtube.com/watch?v=DJcObFauFIc

[Edited on 16-1-2014 by Brain&Force]


3.56 g / 100 g at 20 C, acc. Wiki. This value should further decrease at higher temperature (going by the other REs), so boiling the solution with H2SO4 should precipitate most of the terbium sulphate. If pH is low enough iron should remain in solution.

If you dissolve a fresh batch of your terbium in the 6 M HCl, the iron will be present as Fe<sup>2+</sup>, which is much less prone to precipitating as hydroxide than Fe<sup>3+</sup>.

Oxalic acid is very, very OTC. The higher Tb oxide could probably be 'cracked' by fusion with NaHSO<sub>4</sub>.


[Edited on 16-1-2014 by blogfast25]
 Pages:  1