Thoughts an ideas on Rh separation from PtPd

[orvi]

Me again with same old question. But this time I take it a bit differently.

To start with, I must admit process we developed with great help of many members of this forum, most notably Lou, 4metals and goldandsilver123, does work in separation of Rh from bulk PtPd with average efficiency of 60-70% by one turnover, from feed containing ca 1-2% Rh.
Dissolve in AR -> cement on Fe to get rid of bulk BM -> denox with HCL boil-down, with addition of NaCl to prevent complete hydrolysis -> dilution with water -> simultaneous drop of PtPd by addition of NH4Cl and chlorine sparging.

Firstly to say, there are many many problems with first stages and also with precipitation PtPd stage. Results can vary wildly from feed to feed and are sometimes irreplicable in sane margins of efficiencies. Despite the protocol followed is the same.

This is basic chemistry, most expensive reagent for us is chlorine... And we know there are much more opportunities how to do this.

As we know nobody will give us procedure of how to do it, we try to think - sometimes maybe bit outside the box - about what we can try/do differently. But more brains, more knowledge, more power

Many members could have great ideas, but they cannot make them real thing - maybe they lack analytical equipment, general equipment, resources, time or chemicals.

So, my question is - what are your thoughts on this ? How we can fine-tune the process to make it more efficient, or completely re-build that process based on some interesting observations or thoughts that only need to be tested out real life ?

The positive thing for this forum, which could arise from this effort is - we post our findings here, wheather positive or negative.

Thank you for your help

 

[orvi]

So far, we vaguely/quickly tryied or thought about these pathways/ideas:

1. for NH4Cl precipitation, crucial factor for efficiency is particle size of the drop. As with crystallization of NH4 Pt salt, I could get essentially 999Pt, forming nice orange crystals. So this could be the first lead - make crystals of the drop somehow to grow larger in size. Diffusion of NH4Cl solution through some membrane into the bulk ?
Or taking advantage of greater NH4 Pt salt solubility when hot - so heat to boiling, add enough NH4Cl to make saturated solution of NH4Pt salt, then let it cool, siphon the liquid into another container, harvest pure NH4 salt Pt crystals and repeat this until no more Pt drops...

2. Elaborate on cementation a bit more, as sometimes trend could be seen that one of three metals tend to precipitate faster than the others... We tried various metals for precipitation aside from copper, zinc and iron, but results are irreplicable for now. But sometimes striking resolution is seen, greater than 50% for Rh, but another time with replicated conditions and slightly altered PGM ratios it have no selectivity at all.

3. Flotation/centrifugation of carefully and slowly cemented metal powder ? Pt is practically twice as dense as Pd or Rh... I see it as clear difference but to utilize it may be impossible. But I do not know much about it to say it for sure.

4. selective dissolution of metal sponge/solid metal - this is also possibility where we failed previously, but trends were here, for sure.

5. other possible precipitant than NH4Cl ? We tried to use DMG in our beginnings, but it isn´t easily recyclable. It also offers the advantage of possible Pt reduction to Pt(II) - and this can be also precipitated to some extent with DMG. But work with DMG is tedious. Possibly some other dioxime or beta-oxime ? Pyridil aldoximes ? Ethylene diamine as complexing agent ? (I heard quite much about it´s use, but never find any data regarding it´s properties as precipitant and proper conditions used) As I consider myself quite capable organic chemist, preparation of some expensive hard-to-get precipitants is also an option to rational degree.

And then there are things that surely work - SX and resins. Is there any commercial product with appreciable loading capacity and high parting coefficient for couple Pd-Rh ? Usable for some reasonable concentrations, say 50+g/L PGMs ? It should be fully or at least partly recyclable to economically work well for us. But if you only know about single use SX solvent or resin, I am also more than willing to hear about it So far, I didn´t quite find any suitable for us while researching scientific papers.

From what I read about SX or resins, Rh is usually adsorbed/scavenged last. I well understand reasons and chemistry behind it. So first two metals - Pd and Pt - should be removed fairly selectively in order to beat our cheap clandestine-looking procedure that gives Rh concentrate in more than 60% for single pass. Purity greater than 15% is required for final Rh fraction (as %Rh from overall PGMs present).

 

[orvi]

As it is probably clear, separation of Pt from PdRh is doable with known accessible methods like hydrolysis, tuned NH4Cl drop+crystallization etc.

In current state of my knowledge, PdRh separation one from other is the thing that is difficult.

I admit that it works to appreciable degree by conventional known means - but I know for sure it can be tuned even more. Lots of variables, lots of opportunities to add some % to the recovered portion
Also ammonia/HCL swing for Pd works quite OK to get enriched Rh, but this method once gave us very nice result, but other times failed to be efficient. Lots of strange things happened - formation of pink salt of mixed PdRh of unknown composition, Pd refusing to precipitate as yellow salt, insane dragdown of other PGMs... We cannot elaborate why this happen. Methodology isn´t robust in our hands

 

[Lou]

If you are indeed competent organic chemist, make yourself some n-dodecylmethylsulfide and dissolve that in paraffin oil. That compound is very good at removing Pd from Pt and Rh to the tune of virtually quantitative. You strip / regen the organic sulfide with ammonia and then go straight to reduction with formic acid, or else precipitate the diammine with HCl and off to calcination. In my practice with it, Pd recovery is always quantitative and purity is always near 5N. Just be sure gold and silver are gone.

DMG is a pain in the ass. Only useful analytical in my opinion.

 

[orvi]

If you are indeed competent organic chemist, make yourself some n-dodecylmethylsulfide and dissolve that in paraffin oil. That compound is very good at removing Pd from Pt and Rh to the tune of virtually quantitative. You strip / regen the organic sulfide with ammonia and then go straight to reduction with formic acid, or else precipitate the diammine with HCl and off to calcination. In my practice with it, Pd recovery is always quantitative and purity is always near 5N. Just be sure gold and silver are gone.

DMG is a pain in the ass. Only useful analytical in my opinion.

I am gonna look that synthesis up. Sounds relatively simple, but I worked with sulfides a lot, and it could become tricky very quickly. Thank you for the suggestion

 

[Lou]

It can be come stinky really quickly.

The actual dialkylsulfide isn't bad.

 

[orvi]

It can be come stinky really quickly.

The actual dialkylsulfide isn't bad.

Price is quite high, regarding it is relatively simple chemical. But I think I will purchase one small bottle of 25g to test it, and if it goes nicely, then I will focus on the synthesis. Building blocks - dodecylbromide and dimethyl disulfide - are relatively cheap also on kg level, so if it goes nicely with it, I will purchase raw materials and optimize the synthesis and synthesize my own in bulk. Should be nicely vacuum distilled from concentrated reaction mixture (bp 90°C/ 5 torr). At least in theory.

Yeah, low molecular weight sulfides are hell to work with (performed few Swern oxidations and worked with tiophenol and thioethanol). And oxidize very very quickly in air, if there is free SH group.

 

[Lou]

Nothing to test, I’ll vouch 100% it works!!

Synth it up

 

[orvi]

Nothing to test, I’ll vouch 100% it works!!

Synth it up

I completely trust you in terms it works, but I want to know how it would perform in terms of maximum loading, composition of the PGM solution also with base metals included, suitable pH/free HCL concentration for stripping etc. There are things we still cannot achieve in consistent and efficent way, and we always deal with quite complicated mixtures of PdPtRh combined also with wide variety of BM (Co, Ni, Cr, Cu, Pb - most often all together). Luckily, no Ag nor Au in it.

Does it tolerate base metals in solution ? Or there are some that binds irreversibly, or are a problem in general ?

And if it brings some consistency and robustness into our procedure, then I am all for it I will need to clean the dust off the 10L reactor vessel and some organic technology engineering skills from the past, as nowdays I rarely do reactions on bigger than 5g scale - research is very different in organics...

 

[Lou]

It would be a mistake not to use it in your circumstance. It is far and away superior to anything I’ve used. Remove the lead with sulfuric too and it tolerates base metals exceptionally well, Ni, Co, Cr, Fe, and of course other PGMs.

In fact, I personally would handle the separation you are doing with solvent extraction.

What would you like to know? This reagent and separation is among my favorites in that it gives virtually quantitative recovery and extreme purity. Those are usually trade offs!

I always diluted 1:1 with IsoParM but I’m sure kerosene would work too.
Loading, 2 mol extractant per mole Pd. In practice 80-90 g/L Pd(II) if cut in half. Too much Pd and phases invert.

Try not to extract from aqua regia. Scrub with 1 N HCl before stripping. Strip with ammonia water.

 

[orvi]

It would be a mistake not to use it in your circumstance. It is far and away superior to anything I’ve used. Remove the lead with sulfuric too and it tolerates base metals exceptionally well, Ni, Co, Cr, Fe, and of course other PGMs.

In fact, I personally would handle the separation you are doing with solvent extraction.

What would you like to know? This reagent and separation is among my favorites in that it gives virtually quantitative recovery and extreme purity. Those are usually trade offs!

I always diluted 1:1 with IsoParM but I’m sure kerosene would work too.
Loading, 2 mol extractant per mole Pd. In practice 80-90 g/L Pd(II) if cut in half. Too much Pd and phases invert.

Try not to extract from aqua regia. Scrub with 1 N HCl before stripping. Strip with ammonia water.

I am in stage of purchasing bulk dodecyl bromide and dimethyl disulfide now I would first try to synthesize around 100-200g to see if whole firefighters departement doesn´t come to my lab because methylmercaptan could potentially be evolved from the reaction - which is used as odoriferous gas in natural gas altough I doubt quantity of it will escape, as reaction is two-phase, second phase being 6% aq. NaOH, but threshold smell limit is extremely low.

Sweet parameters. Now I understand. This will do exceptionally well in our case based on your description
Yeah, I would de-NOx it firstly - as I know sulfides are pretty easily oxidized to sulfoxides and sulfones by also relatively mild oxidants, such as H2O2. So NOx gasses and possible chlorine in mixture will easily destroy precious extractant.
Our problem was to strip the feed of Pd somewhat, as we can relatively efficiently part Pt and Rh by hydrolysis, since Rh is rarely over 2% in our solutions, It also works fine in terms of purity of obtained Pt and recovery % of Rh.

What you mean by saying scrub with 1N HCL ? I am sorry, but I have a bit of hard understanding probably due to my language barrier. You mean de-NOxing it with 1N HCL or dilute feed/wash the extractant/condition extractant with 1N HCL ?

 

[Lou]

Oddly, it seems like dilute nitric is ok, even 10-15%. It’s NOCl that seems to cause the issues.

I know what you mean about the whole fire department. Back in school I was using some mercaptans with no hypochlorite trap. Perhaps 5-10 mL in the hood. They thought there was a gas leak at the local newspaper and that was a mile away. Whoops!

I was NOT a competent organic chemist.

 

[Lou]

By scrub I meant stir the loaded extractant with some dilute hydrochloric after loading with Pd. To remove entrained aqueous phase. I usually do this twice for five minutes. Then once with water and then once with acid after the ammonia.
Both of those I add to the ammonia strip.

 

[goldandsilver123]

It would be a mistake not to use it in your circumstance. It is far and away superior to anything I’ve used. Remove the lead with sulfuric too and it tolerates base metals exceptionally well, Ni, Co, Cr, Fe, and of course other PGMs.

In fact, I personally would handle the separation you are doing with solvent extraction.

What would you like to know? This reagent and separation is among my favorites in that it gives virtually quantitative recovery and extreme purity. Those are usually trade offs!

I always diluted 1:1 with IsoParM but I’m sure kerosene would work too.
Loading, 2 mol extractant per mole Pd. In practice 80-90 g/L Pd(II) if cut in half. Too much Pd and phases invert.

Try not to extract from aqua regia. Scrub with 1 N HCl before stripping. Strip with ammonia water.

Would you extract the platinum with solvent as well? There's a solvent for rhodium?

 

[Lou]

Yes, with solvent for Pt. The Rh I’d just evap down and reconstitute in c. HCl and do as I’ve already told you privately.

Now everyone on me for the keys to the kingdom on S/X. Heck, I’m about to turn my keys in here.

 

[orvi]

Oddly, it seems like dilute nitric is ok, even 10-15%. It’s NOCl that seems to cause the issues.

I know what you mean about the whole fire department. Back in school I was using some mercaptans with no hypochlorite trap. Perhaps 5-10 mL in the hood. They thought there was a gas leak at the local newspaper and that was a mile away. Whoops!

I was NOT a competent organic chemist.

Yup. I will need to be diligent in work with dimethyldisulfide... Traps, waste treatment with hypochlorite, using some electric cheap membrane pump for distillation as it probably stay stinky forever... And work up the procedure for myself, as original papers are using high dilutions, which will never work for me in bulk. But it is a challenge for me, I have done some sulfur chemistry before when it was necessary to use, but only in small quantities. And as I am doing the maths, it will be very nice to have something about 1L of the stuff in hand I need to figure it out somehow, how to manage everything, but I stay positive and think it will be done OK.

 

[orvi]

Yes, with solvent for Pt. The Rh I’d just evap down and reconstitute in c. HCl and do as I’ve already told you privately.

Now everyone on me for the keys to the kingdom on S/X. Heck, I’m about to turn my keys in here.

If I could just spend all of my work time by researching this very interesting toppic... I would like to.
Unfortunately, I am doing completely different chemistry as my primary work. But also nice. And day still has only 24 hors
I am very greatful for your suggestions and insights. With your help, my understanding in this field was greatly enriched. Now, also with my own intended procedures, I can implement those valuable bits of information and proceed further in developing our own methods. When I complete the sulfide extractant, I will let you know how it went but it will probably took quite a bit of time. For now, we are on the cheap side with currently used process, but lots of problems arise from its principal setup, which cannot be overcomed by basic chemistry, where most expensive chemical used is nitric acid and it works, but nowhere efficiently and nicely. So I stay motivated by future changes for the better processes

 

[Abdoulapapatte]

hello friends it's always interesting to read you're reaction
rhodium is very interesting