Pd Pt Rh mix industrial cat refining tips ?

[orvi]

hi orvi

I would like to recommend these videos, they are very high quality and illustrate your situation perfectly.

You start with plain water, so before the first step you either concentrate your solution and boil it.

You evacuate the water in vapor and you concentrate your concentrations of pgms. then when the solution is ultra concentrated, and quite dark, you dilute by adding hcl to eradicate hno3, you evaporate, you concentrate and you put back hcl a second time.

,1) if now you have 100 liters, add zinc, added with zinc (or cu) and precipitate the pgms

hno3+hcl from dissolving pgms------> boil-----> evaporate ( pgms concentration)-----> add hcl

-----> boil and insert the solution again then you get to this stage, note all the tips in the video.



2) after applying the platinum sponge ----> reduction in its metallic form by ammonium formate



3)Then separation of palladium:



3.1) add ammonium chloride again in small quantities, add also

kclo3 (potassium chlorate or NaClO3 (sodium chlorate works better). -----> boil

the red concentrate-----> red precipitate will be palladium , the rest of the highly concentrated liquid will be rhodium .

3.2 rinse with HCL or with a 10% ammonium chloride solution (pd salt will not dissolve)

4 turn the palladium sponge into metal:



dilute palladium precipitate with water ------> boil ------> ammonium formate----> Pd metal

5) precipitation of rhodium in its metal form :



THE OLD ZINC TECHNIQUE ------> will not dissolve in aqua regia

6) another observation

to refine rhodium easily it is necessary to prepare it (make it etchable):

two videos not to be underestimated :

https://www.youtube.com/watch?v=Bd8Ou4WGNXk&t=486s
How to detect Rhodium (precipitation with alcohol)



I hope that it helped you, I wonder if the stoichiometry is really important, I know well that Nh4cl can precipitate platinum and palladium at different precipitation times

1)platinum -----> precipitates-----------time-----> palladium precipitates

bye

Yeah, I know and we tried these easy and straightforward approaches in the beginning, but we have limited time, and we cannot afford to scavenge PGMs from all feeds back together and re-refine all fractions to get values to the two piles.
I am still working on the subject, with little pauses.

 

[orvi]

So, after long time two new things were tested, with partial success in one way.

I had an idea... Wacker oxidation (org. chemistry) uses PdCl2 as catalyst together with CuCl2 as co-catalyst. In the reaction, CuCl2 is able to oxidize Pd to PdCl2 and it reduce itself to CuCl. So I give it a go with my PGM mix cemented on zinc powder.
Composition roughly 50% Pd 43% Pt 2% Rh. For 2 grams of the PGM powder, I used 1 g of CuCl2x2H2O and 2 x the ammount of HCL needed to make chlorocomplexes from all PGMs. Stirred for 2 days. Found that CuCl2 etch all PGMs in cemented form, Pd mostly, Rh is going also quite fast, with Pt as least reactive to these conditions.
Solution transitioned from blue-green (from copper) to dirty-green-yellow to nearly black-brown. More than half of the PGM charge dissolved to the solution, but I didn´t measured the remaining powder exactly - I just calculated it from XRF readings ( I knew how much CuCl2 and PGMs I used, so I know the composition of the solution).

It may be a valuable info for anyone trying to cement or dissolve PGMs easy way from various sources. I must point out, that I do not know how PGMs would behave in form of plating or wire etc...

-------------

Optimization of Pt drop with NH4Cl

After previous failure, I decided to follow instincts rather than scientific papers without any exact procedures given - this particular thing can get me quite upset. Lots of papers without exact procedure given, limited numbers and unclear experiments. So I planned my own one, tuned by experience from the first time drop.
I used roughly 2,17g PGMs in solution, 50%Pd42%Pt2+%Rh with just a 1% base metals, mainly zinc and touch of copper.
To this solution (7,5 mL) I added NaCl (1,79 g) corresponding to the stoichiometric ammount needed to form Na-PGM salts on heating - denoxxing. As salt do not dissolve well in solutions full of chlorides, I added just enough water to completely dissolve the salt at boiling point. This is important, because if salt remain undissolved, chlorocomplexes start to liberate HCL and on redissolution, opalescent liquid with precipitate is obtained, which need additional HCL to redissolve = more chlorides = less selective drop.

After complete evaporation to crispy solid, I added water (20 mL) and dissolve the solid to semi-clear solution. Some HCL was used to correct the pH and allow the solids to enter the solution (less than 0,5 mL). Solution was acidic after this operation (pH 2-3). I measured just stoichiometric ammount of NH4Cl needed for Pt drop (0,501 g) and dissolved in minimal water.

I added NH4Cl solution in one portion to the clear, stirred solution of the PGMs at room temperature. I wash the small beaker (used for dissolving NH4Cl) with little water to quantitatively transfer NH4Cl to the reaction. Stirring was continued for 40 minutes. Total volume of the solution was 32 mL, so after adding all ingredients, PGMs concentration in solution was 70g/L - so Pt content was 29,4g/L.

Filtering the precipitate and washing the cake 2x10 mL of 5% NH4Cl resulted in precipitate of composition:

98,6% Pt 0,80 % Pd and 0,56% Rh (XRF). Much much better than previous attempt. Cake is also much nicer in color, no orange tinge, no green specs. From that highly "polluted" solution with other PGMs, I think it is not a bad result - correct me if it can be done much better, I do not know since these are my best numbers

Effectivity of the drop was roughly 80-85% - I correct the number after weighing the ppt in the liquid.
After filtering the remaining precipitate in filtering flask (due to NH4Cl wash) and XRF measurement of the liquid, effectivity of the drop was 87%.
Remaining NH4 platinate precipitated in the filtering flask after washing the cake with NH4Cl. It will be wise to change the filtering flasks before washing to eliminate this problem and allow to re-drop the Pt from mother liquor in controlled way. Since Pt drop isn´t 100% any time, it is easy to calculate NH4Cl left in the solution and adjust the ammount of second addition of NH4Cl to precipitate more Pt from juice, altough with low Pt content, It would probably be quite impure. Or with prolonged stirring in large batch, it would be easy to measure liquid and solid phase to find comfortable point between purity and yield. I will see.
Effectivity is not very good for one-step operation. We need to get at least 90% of Pt out in this operation. I will see if it is possible - to balance purity and recovery somwhere acceptable. If this cannot be accomplished... we need to consider two stage refining of NH4Pt salt :/

Two more experiments are planned - one with stoichiometric NH4Cl and higher concentration of PGMs (120-150g/L, from this Pt 50-63g/L), second with slightly over-stoichiometric ammount of NH4Cl (1,2 x NH4Cl needed) with retained concentration of 70g/L PGMs.

 

[orvi]

Another failed experiment, but with some interesting things happening:

Theory:

If Rh is insoluble in nitric acid, and palladium sure is, there could be possibility to resolve these two by making PGM powders and adding HNO3 to them. Then filter the undissolved Rh.

Real life:

I have taken the enriched fraction of denoxed AR solution of PdRh - around 90% Pd 4-5% Rh and ca 3% Pt - mother liquor after Pt precipitation with NH4Cl - total weight of the metals was around 1,0 g. While stirring vigorously, I added zinc powder until all PGMs were dropped. Then added HCl and heated to ca 60°C to redissolve any excess zinc that could be present. After half hour, settled cement (nicely flocculated by the action of ZnCl2) was decanted and repeatedly washed with water (around 4 times). It was clearly visible, that ZnCl2 was slowly washed from liquid - because the precipitate started to fall appart from nice flocculated particles to nearly dust. Last two washes were with distilled water. AgNO3 test on wash water proved that there are no chlorides in solution.

Then to the wet powder (in ca 2-3 mL dH2O), I added 2 mL p.a. grade nitric (65%). Then strange observations started. When heated to ca 60-80°C, palladium started to go to the solution - but no gas, no NOx fumes were evolved. Just solution taken more and more brown orange colour and that was all. No single bubble...

And, as you can guess, everything passed to the solution, leaving small ammount of some black dust in the beaker. After filtering on Hirsch funnel with pre-weighed filter paper, ppt weighed less than 1 mg...

So, apparently Rh is quite soluble in nitric acid - if finely divided of course.

Time to move on to the more elaborate, but working procedures

 

[orvi]

So, finally we have some methodology which we can actually use. Not perfectly efficient, but it slowly gets us where we want to be

Theory:

NH4Cl can drop Pt and also Pd oxidized to +4 oxidation state from their chlorocomplexes. Since solubility of (NH4)2PdCl6 and (NH4)2PtCl6 is greatly dependent from overall chloride content of the solution, NH4+ ion concentration and also HCl content = acidity, we made few control experiments on selectivity of Pt drop in various conditions. We found (no surprise), that Pd and Rh coprecipitate more in more concentrated solutions. NH4Cl has dramatic effect on selectivity towards Rh. If overstoichiometric NH4Cl ammount was used on our mixture, percentage of Rh was rising more rapidly than Pd content of the precipitate.
SO we designed experiment like this - add stechiometric ammount of NH4Cl for Pt and also Pd, and then with stirring, oxidize Pd to the +4. This will cause sharp drop in overall chloride content of the solution (since chlorometallates will precipitate out), which could help keep Rh soluble.

Reality:

2,000 g of PdPtRh cement mixture (ca 50%Pd 42%Pt 2,5%Rh) was dissolved in AR. After complete dissolution, stechiometric NaCl was added and solution was evaporated to dryness (at 175°C). Great care was taken to have enough water in the solution to dissolve all the NaCl - if some remain undissolved, portion of the PGM-acids will release HCl and become insoluble after evaporation.
After evaporation to crispy powder, residue was redissolved in water to form clear dark liquid. Volume was adjusted to 20 mL. Then, stechiometric ammount (for Pt and Pd) of NH4Cl was dissolved in 4 mL of water and poured into the stirring PGM solution. 1 mL of water was used to rinse the beaker from NH4Cl.

Then, I set up the chlorine generator from the round-bottomed flask with side tube, connecting to PVC tubing and Pasteur pipette used as bubbler. I filled the flask with two spoons of KMnO4, and attached addition (pressure equalized) funnel filled with 18-20 % HCL (diluted to 20% because of less HCL drag with generated chlorine).

Chlorine was then steadily generated and bubbled to the PGM solution. Whole operation took about 30 minutes, during which it was clearly evident that Pd is creating precipitate, as color of the solution faded. This was a very good lead to the endpoint of the precipitation - when the solution does not discolor more for some time, it was over. I let the reaction stirr for another 5 minutes with very low bubbling rate of chlorine, to keep ORP of the solution high.

Then, the suspension was filtered on Buchner funnel, sucked as much liquid as possible (very light red-orange solution), and washed the cake with 2 x 10 mL of 1 M NH4Cl (ca 5% w/w). After second addition, filtrate become colorless.

Filtrate was collected, PGMs were cemented with powdered zinc to obtain 120 mg of fluffy, flocculated black powder. XRF measurement revealed that it contain 28% Rh, 35% Pd and 15% Pt. Filter cake measurement revealed 59% Pd 39% Pt and around 1% Rh, which roughly correspond with weight comparisons of feed and zinc-ed filtrate. So we managed to concentrate 67 % of Rh to the enriched fraction, which we can sell.

If the selectivity remain the same when we redissolve the NH4 salts precipitates and do the second drop of PtPd, we could squeeze 89% of Rh to the enriched fraction. We will see how this second dissolution-redrop sequence will go.

But for now, I am buying a bottle of champagne and I pop it when we get the Rh to the vial

 

[orvi]

Some photos of setup and cake+filtrate

 

[Swissgoldrefiner]

Some photos of setup and cake+filtrate

Nice job...but what's the aim of increasing the % of Rh? As you mention, you dont want get 99.9% of each metal...So that mean you will end up with giving the concentrate to a refinery or i miss understood?
Otherwise it's because you like the scientific challenge?
I mean if i was you...i will probably send it like this to refinery...except if that's to have fun and/or scientific challenge.

 

[Golddigger76]

I assume he's trying to come up with a way to recover enough rhodium % to meet a minimum to be able to sell without very expensive and complicated reactor and supporting equipment plus it's a highly guarded secret by most of those who are currently doing it now.. we know it can be done and they will throw you a bone here and there but they had to learn the same way Orvi experiments. I don't think that I would be so quick to hand out a method that took me a year of working on it to figure out either.
Obviously rhodium is a very expensive metal, complicated to refine, and if you can get 67% rhodium out of your pgm solution without that equipment the Pt and Pd are the bonus metals. Could be having to keep a balance of one metal to hold another in solution ?
I am no expert on PGM'S, still learning palladium and platinum but Rhodium is still way over my head. He's refining catalytic converters and along with that comes the Rhodium.
I have a nice pile of converters waiting for me to get more educated before I try and that's mainly because my waste is still going to have the Rh left in it if I only can recover the palladium and platinum. That would be a damn shame since rhodium is around $15,500 per ounce right now and I don't desire to keep the waist in storage until I get around to figuring it out so for now those converters are a safe form for keeping them.

 

[orvi]

Nice job...but what's the aim of increasing the % of Rh? As you mention, you dont want get 99.9% of each metal...So that mean you will end up with giving the concentrate to a refinery or i miss understood?
Otherwise it's because you like the scientific challenge?
I mean if i was you...i will probably send it like this to refinery...except if that's to have fun and/or scientific challenge.

I assume he's trying to come up with a way to recover enough rhodium % to meet a minimum to be able to sell without very expensive and complicated reactor and supporting equipment plus it's a highly guarded secret by most of those who are currently doing it now.. we know it can be done and they will throw you a bone here and there but they had to learn the same way Orvi experiments. I don't think that I would be so quick to hand out a method that took me a year of working on it to figure out either.
Obviously rhodium is a very expensive metal, complicated to refine, and if you can get 67% rhodium out of your pgm solution without that equipment the Pt and Pd are the bonus metals. Could be having to keep a balance of one metal to hold another in solution ?
I am no expert on PGM'S, still learning palladium and platinum but Rhodium is still way over my head. He's refining catalytic converters and along with that comes the Rhodium.
I have a nice pile of converters waiting for me to get more educated before I try and that's mainly because my waste is still going to have the Rh left in it if I only can recover the palladium and platinum. That would be a damn shame since rhodium is around $15,500 per ounce right now and I don't desire to keep the waist in storage until I get around to figuring it out so for now those converters are a safe form for keeping them.

This issue started way long ago. And yes, it seems irrational to just enrich, not refine completely. But as Golddigger76 said, our buyer of PMs have minimum ammount requirement of 15% for Rh to be sellable. From the point that there are thousands of other buyers in the world that account for Rh also in 0,6-2,5% range (what we have in the feed) - yep, it is true. But my boss have long term relationship with his PM buyer, so I was given a task to enrich fraction with Rh to squeeze some more bucks out of the PGM melt. He probably has his own reasons to stick to that specific buyer despite all the trouble. This I do not know, and it is completely up to him.

This whole secrecy, hidden information... Whole another world of immense knowledge is hidden completely from our eyes and made secret just because money. I understand the logic behind it, but personally I do not agree with this. I learned huge ammount of things here on the forum, and I am now maybe giving a little bit of it back. Everywhere is only stated use NH4Cl, oxidize the Pd with chlorate. Nowhere is enough informations for the methods hundreds of years old. So I shared my experiences, which can be valuable as lead for other beginners to avoid some mistakes that I made.

As to the method and sharing it with others... This is a very specific problem, I faced. I do not think there are many refiners who enrich catalytic PGMs that way the Rh is past 15%. Also, I do not feel that I would be able to refine the Rh efficiently to 99+% without few years of trying. We get the smelting right, we get every percent of PGM out of matrix very efficiently. And that took eternity to do right way with limited equipment and furnance temperature.

I am just an organic chemist doing my PhD, I do not want to make a refining career. It is just a little bit more than a hobby for me. Very exciting field. So I play with expensive toys

 

[Golddigger76]

The information you've provided us with is 100% priceless, hopefully someday l will be able to have half of the incite of you.
We are still learning our lessons based on Hoke's books and in all of the time that has passed since it was written I have a feeling that a few someone's have found new ways that would blow our minds. Either way I enjoy learning new techniques regardless of if I will ever need them or not so thank you for sharing what has and hasn't worked for you !!

 

[Golddigger76]

Orvi
If you don't mind me asking are the industrial cats by chance from big diesel engines ?

 

[orvi]

Orvi
If you don't mind me asking are the industrial cats by chance from big diesel engines ?

I doubt it, but more probably these are catalytic beds used in industry for various reactions. Composition of the ceramic do not correspond to the catalytic converter like matrix.

 

[Mdctr]

This issue started way long ago. And yes, it seems irrational to just enrich, not refine completely. But as Golddigger76 said, our buyer of PMs have minimum ammount requirement of 15% for Rh to be sellable. From the point that there are thousands of other buyers in the world that account for Rh also in 0,6-2,5% range (what we have in the feed) - yep, it is true. But my boss have long term relationship with his PM buyer, so I was given a task to enrich fraction with Rh to squeeze some more bucks out of the PGM melt. He probably has his own reasons to stick to that specific buyer despite all the trouble. This I do not know, and it is completely up to him.

This whole secrecy, hidden information... Whole another world of immense knowledge is hidden completely from our eyes and made secret just because money. I understand the logic behind it, but personally I do not agree with this. I learned huge ammount of things here on the forum, and I am now maybe giving a little bit of it back. Everywhere is only stated use NH4Cl, oxidize the Pd with chlorate. Nowhere is enough informations for the methods hundreds of years old. So I shared my experiences, which can be valuable as lead for other beginners to avoid some mistakes that I made.

As to the method and sharing it with others... This is a very specific problem, I faced. I do not think there are many refiners who enrich catalytic PGMs that way the Rh is past 15%. Also, I do not feel that I would be able to refine the Rh efficiently to 99+% without few years of trying. We get the smelting right, we get every percent of PGM out of matrix very efficiently. And that took eternity to do right way with limited equipment and furnance temperature.

I am just an organic chemist doing my PhD, I do not want to make a refining career. It is just a little bit more than a hobby for me. Very exciting field. So I play with expensive toys

Hi

 

[Mdctr]

Hi Orvi,
It's good to see there's still some good people in this world!
Id like to take a minute to give you recognition and appreciate your post. I read ALOT I'm not sure the ratio but its probably up there with odds of hitting the lottery to read honest positivity. It's people like you that give some light at the end of the tunnel for beginners like myself to continue to educate put forth the effort to achieve our own.
Let someone mention pgms from catalytic converters and o man peoples jaw drops. Yes i just said the c word sorry if I offend anyone. Lol
I will look through what you have posted and look forward to learning what youre willing to share. Thanks

 

[orvi]

After initial findings of heavy coprecipitation of Rh and Pd together with Pt, I started to think about why this happen. Solubility of Pd and Rh ammonium chlorocomplexes should be high enough to keep everything in solution, and with prolonged stirring, equilibrium should be established and theoretically, pure (NH4)2PtCl6 should be obtained.
But there is one thing which I didn´t realized - cocrystalization and adsorption. These compounds are heavily charged complex particles - and they would have tendency to adsorb themselves onto the surface of the precipitate. And since the precipitate of platinum ammonium salt have vast surface area - it will have plenty of opportunities to do so. Also, in the ppt, there can be locked Pd and Rh, as the formation of precipitate is immediate.

So, I designed an experiment to prove or disprove this hypothesis. I took (NH4)2PtCl6, of composition roughly 93,6 Pt 2,9Pd and 2,11 Rh - which was dropped earlier in the "research". I divided the cake into 2 portions, weighing 1,00 g each. Then, I prepared 0,25 M solution of NH4Cl (roughly 13g/L, very diluted in comparison to more than 5M saturated solution) and added 11 mL to each of the precipitates in the small beakers. One was stirred 15 hours at room temperature, and second was heated to 80 °C for 5 hours, and then cooled undisturbed to r.t. overnight.

Results were clear - this trituration shown that Pd and Rh ARE leached to the liquid phase from precipitate, but only to some extent. Room temp experiment yielded 96,7 Pt 2,0 Pd 1,2 Rh precipitate, and 80 °C experiment yielded 97,8 Pt 1,3 Pd 0,9 Rh precipitate. It was clear, that even in practically very diluted solutions, PGMs tend to stay adsorbed onto the solid phase, possibly due to some electronic effects or affinity. Experiment at 80 °C yielded more orange tinted precipitate, despite the fact precipitate washed at 80 °C was purer in XRF reading - crystals were bigger and probably partially hydrolyzed Pt species, not full hexachloroanions...
This is due to the fact that at 80 °C, Pt salt has much greater solubility in water, and thus it was crystallizing from the solution in bigger crystals = lowering the surface area of the precipitate = minimizing possible adsorption.

On the photo, right Petri dish and beaker correspond to 80°C experiment, on the left, room temp. experiment.

As this operation has some potential in cleaning the precipitated Pt salt from some limited ammount of PdRh, it shown very clear where lies the problem with NH4Cl precipitations of platinum from HCL solutions.

 

[orvi]

So next, I thought about how to minimize the adsorption of Pd and Rh chlorocomplexes to the Pt precipitate. Logically, there are two obvious things that could be done - change or alter the surface charge of particles or solution, so the other ions do not want to adhere on the surface of the particles. Or, simply lower the surface area available for this to happen. Since I am not very into surface chemistry mambo-jambo, I tryied the classic way - crystallization.

It should be said that ammonium platinate does not have great solubility in water, even at boiling point (tables state 33,6g/L), but gradient for crystallization is very pleasing - depending on source - 1-5g/L at room temperature. So effectivity of crystallization should be good.

I picked 1,65 g of impure (NH4)2PtCl6 containing 96,2 Pt 2,58 Pd 1,19 Rh, and certainly traces of NH4Cl from washing steps.

This was brought to boiling in water, and overall 55 mL were required to fully dissolve the cake to very nice, deep orange liquid. I suspect traces of NH4Cl in the precipitate could lower the solubility, because in theory less than 50 mL should be enough to completely digest the precipitate at boiling point. Then I left the beaker on hot hotplate to cool down slowly to room temperature overnight. In the morning, nice crop of small very brightly orange crystals appeared, which were collected on Buchner funnel - weighing 1,11 g. XRF shown essentially pure Pt (99,95% reading, confidently more than 99,8% due to fact it is measured by XRF handgun).

I attempted to collect remaining platinum from still quite deep-orange colored mother liquor by adding saturated NH4Cl - but without the success. Hexachloroplatinate anions must have hydrolysed in boiling conditions to mixed hydroxo-chlorocomplexes, thus largely enhancing the solubility in water. So I also added few mL of HCL to the solution in effort to re-chlorate the platinum. And it worked.
Shifting the equilibrium towards the precipitate of fully chlorinated Pt species is slow, it took few hours, but this is advantageous - slow release of fully chlorinated Pt salt tend to produce bigger crystals = lowering surface area for adsorption. And by this procedure, I managed to get additional 0,298 g of yellow crystals with slight green tinge (Rh). XRF reading was 99,4 % Pt 0,6% PdRh.

First crop of nice orange crystals don´t resemble hexachlorocomplex, but more probably partially hydrated aquachloro complex. So for weight, there will be probably slightly more Pt (since H2O has lower molar mass than Cl) than expected.

Zinc-ing out the mother liquor (very faint coloured solution) gave only 30-60 mg of PGM precipitate with 18% Rh content. Which is spectacular result for me.

With careful addition of just a little NH4Cl and HCL to the hydrolysed PtCl6 mother liquor, it would be possible to go for even better purity than 99,4 % for the next precipitation.

Clearly, this isn´t industry-friendly process. 33,6g/L resembles just around 15g of platinum metal per liter of water needed. Working in 5L beaker safely can only accomodate 60g of platinum. But it works. And could be helpful for small scale refiners to achieve good separations



Satrurated solution on the left and crystals from first crystallization on the right


XRF of the first crop of crystals. Our XRF is confused and measure Rh and Pd in very small concentrations as Zn, when measuring chloride salts.

So, I am happy finally, something is working.

 

[Mdctr]

So next, I thought about how to minimize the adsorption of Pd and Rh chlorocomplexes to the Pt precipitate. Logically, there are two obvious things that could be done - change or alter the surface charge of particles or solution, so the other ions do not want to adhere on the surface of the particles. Or, simply lower the surface area available for this to happen. Since I am not very into surface chemistry mambo-jambo, I tryied the classic way - crystallization.

It should be said that ammonium platinate does not have great solubility in water, even at boiling point (tables state 33,6g/L), but gradient for crystallization is very pleasing - depending on source - 1-5g/L at room temperature. So effectivity of crystallization should be good.

I picked 1,65 g of impure (NH4)2PtCl6 containing 96,2 Pt 2,58 Pd 1,19 Rh, and certainly traces of NH4Cl from washing steps.

This was brought to boiling in water, and overall 55 mL were required to fully dissolve the cake to very nice, deep orange liquid. I suspect traces of NH4Cl in the precipitate could lower the solubility, because in theory less than 50 mL should be enough to completely digest the precipitate at boiling point. Then I left the beaker on hot hotplate to cool down slowly to room temperature overnight. In the morning, nice crop of small very brightly orange crystals appeared, which were collected on Buchner funnel - weighing 1,11 g. XRF shown essentially pure Pt (99,95% reading, confidently more than 99,8% due to fact it is measured by XRF handgun).

I attempted to collect remaining platinum from still quite deep-orange colored mother liquor by adding saturated NH4Cl - but without the success. Hexachloroplatinate anions must have hydrolysed in boiling conditions to mixed hydroxo-chlorocomplexes, thus largely enhancing the solubility in water. So I also added few mL of HCL to the solution in effort to re-chlorate the platinum. And it worked.
Shifting the equilibrium towards the precipitate of fully chlorinated Pt species is slow, it took few hours, but this is advantageous - slow release of fully chlorinated Pt salt tend to produce bigger crystals = lowering surface area for adsorption. And by this procedure, I managed to get additional 0,298 g of yellow crystals with slight green tinge (Rh). XRF reading was 99,4 % Pt 0,6% PdRh.

First crop of nice orange crystals don´t resemble hexachlorocomplex, but more probably partially hydrated aquachloro complex. So for weight, there will be probably slightly more Pt (since H2O has lower molar mass than Cl) than expected.

Zinc-ing out the mother liquor (very faint coloured solution) gave only 30-60 mg of PGM precipitate with 18% Rh content. Which is spectacular result for me.

With careful addition of just a little NH4Cl and HCL to the hydrolysed PtCl6 mother liquor, it would be possible to go for even better purity than 99,4 % for the next precipitation.

Clearly, this isn´t industry-friendly process. 33,6g/L resembles just around 15g of platinum metal per liter of water needed. Working in 5L beaker safely can only accomodate 60g of platinum. But it works. And could be helpful for small scale refiners to achieve good separations

View attachment 50756View attachment 50757

Satrurated solution on the left and crystals from first crystallization on the right

View attachment 50758
XRF of the first crop of crystals. Our XRF is confused and measure Rh and Pd in very small concentrations as Zn, when measuring chloride salts.

So, I am happy finally, something is working.

Have you ever tried melting NH4CL or NaCL03 by itself and adding to solution at room temperature?

 

[Yggdrasil]

Have you ever tried melting NH4CL or NaCL03 by itself and adding to solution at room temperature?

Since NH4Cl melts at 338C that is not possible.

 

[orvi]

Have you ever tried melting NH4CL or NaCL03 by itself and adding to solution at room temperature?

What would be the intention of such melting and addition ? I don´t understand.
Molten sodium chlorate is insanely dangerous oxidant. And ammonium chloride rather sublime than melt.

 

[Yggdrasil]

If time is not of essence, won’t the different salts separate as they crystallize?
At least to some degree.

 

[orvi]

If time is not of essence, won’t the different salts separate as they crystallize?
At least to some degree.

Yes, they would, and quite sharply. But you need to get conditions right. Not too much HCL not too much NH4Cl in second crystallization, slowly and steadily. You obtain two crops of crystals and one mother liquor. If you get it right, you get two crops of xx you can combine and process. With mother liquor, you get what you get. In my case from 2% Rh to 18% Rh as cemented powders.