questions about using sodium formate to drop palladium

[mls26cwru]

so I have a filtered nitric solution with mostly silver and palladium in it. I have been looking around for information about using sodium formate to drop the palladium. from what I read, it seems easy and straight forward, but I wanted to check with anyone here that could help.

I have a stash of scrap that I used hot nitric to dissolve palladium and silver from (Pd positive stannous test & DMG test, and hcl test for silver). The material had a prior HCl bath to remove most tin solder already. Possible contaminates to consider 'might be' lead and copper in small quantities... the material seems pretty clean of those though. After dissolving I let cool and filtered the resulting solution. It is currently waiting in flasks for my next step.

my thought is this:
- neutralize the nitric with sulphamic acid, let settle, decant/filter
- add HCl to drop silver as silver chloride, let settle, decant/filter
- adjust pH to 1.5 with NaOH
- prepare solution of saturated sodium formate
- add to palladium bearing (now) chloride solution until pH is about 4.5
- wait for palladium to drop, decant/filter

is it really that easy?

other question I have are as follows:
- Can I just neutralize the nitric and add the sodium formate if I am not concerned about silver contamination?
- What other metals would sodium formate possible precipitate? And are they related to the pH?
- Are there any concerns about adding sodium formate to my general acid waste stream?
- Is a saturated sodium fromate solution good enough, or should I be aiming for a specific concentration?
- any other pointers from those who know would be helpful... even if its just material to read/study

I do have zinc and could precipitate material this way if needed. I was originally going to try this, and then melt the resulting precipitate and xrf the bead to get the data. but If i can eliminate the zinc with an easy method, I would be interested in trying do do that.

Thanks,
Mike

 

[FrugalRefiner]

I'm somewhat simple minded, so I like to keep things simple.

There is no need to neutralize your nitric with sulphamic acid. As you planned, I would add HCl to drop the silver as AgCl. Since you "might" have lead, most of the lead would also drop as an insoluble chloride. It's easy to eliminate from the AgCl by washing with very hot water.

I have no experience with formate reduction, so after dropping all the silver with HCl, I would simply cement the Pd out on copper. I wouldn't use zinc as it will drop almost anything you have in solution, including any remaining lead and the copper. Copper will only drop precious metals and mercury if it happens to be there.

A simple process my simple mind can understand.

Dave

 

[Lino1406]

Sodium or ammonium formate with 15-30% concentration. No need for NaOH (pH 1.5...)
as formate itself does the buffering. Both solutions should be heated to almost boiling
lastly work under hood

 

[Palladium]

I do have zinc and could precipitate material this way if needed. I was originally going to try this, and then melt the resulting precipitate and xrf the bead to get the data. but If i can eliminate the zinc with an easy method, I would be interested in trying do do that.

Thanks,
Mike

If all your looking for is data then drop it all with zinc, wash three times with hot water, boil in hcl, wash three times with hot water, melt, weigh, xrf! That should give you the silver, gold, and any pgm's present.

 

[mls26cwru]

That was going to be original plan, but when I got to reading about the sodium formate reduction, it got me thinking.

If it is as easy as what I read, I may try that route. As long as the waste doesn't need to be treated differently than my general acid waste, and if there are no other pitfalls to look out for, I may consider it.

For a test batch, I can work around the impurities and such and still get the data. For the job overall, it would be more advantageous to drop the palladium in a more pure form, so i was just looking to understand the process better.

Mike

 

[Lou]

It is that easy. If you do not add a gross excess of formate, waste treatment is the same.

Remove silver first. Wash free of Pd.

Adjust pH to 2. Add in sodium formate and heat well until no Pd color remaining.

Filter Pd. Refine if desired.

Waste liquor Cu recovery, then precipitation with bleach.

 

[mls26cwru]

Thank for the info Lou... Especially about the waste treatment part!

I was hoping to try to get around to this last weekend, but the time got away from me... Hopefully this weekend though. I will keep you all posted!

Mike

 

[mls26cwru]

Well, so far so good I guess!

 

[g_axelsson]

Since you decided to skip the denoxing stage (wise decision) this is a bit moot observation. I'll just add it for anyone reading this thread later on with that idea in mind.

Denoxing a silver containing nitrate solution with sulfamic acid would result in silver sulfate. That is only slightly soluble so most of the silver would precipitate out as silver sulfate.

I might be wrong as I've never tested it myself just because of that reason.

Göran

 

[mls26cwru]

Well, here is the result... a bit late, and I haven't tested it for purity yet.

The top is how it came out of the crucible. The bottom i polished up a bit to see what kind of shine it would take on. It was tough to get it to buff up, and definitely took a good bit more effort to melt then gold.

 

[kurtak]

Well, here is the result... a bit late, and I haven't tested it for purity yet.

Nice - thanks for up dating your results :G :mrgreen:

Kurt

 

[Geo]

It is that easy. If you do not add a gross excess of formate, waste treatment is the same.

Remove silver first. Wash free of Pd.

Adjust pH to 2. Add in sodium formate and heat well until no Pd color remaining.

Filter Pd. Refine if desired.

Waste liquor Cu recovery, then precipitation with bleach.

So the solution can be contaminated with copper? I wonder how I missed this. I've been separating the Pd from the contaminated solution with ammonium chloride and chlorine first just to have a clean solution to work with. Looks like I have been working too hard for it.

 

[mls26cwru]

my particular solution had no copper in it so I don't know if there is drag down or a partial reduction of the copper... and from what I read in my research, it seems that sodium formate is not particularly selective in regards to silver, platinum, palladium, rhodium... while the reduction rates are highest for palladium (95+), the others are reduced 40-60%. I am not particularly interested in refining the palladium past this point so the method suites my situation.

My starting solution contained silver and palladium almost exclusively... silver was easy enough to separate, leaving me just palladium. I was going to use zinc if this wasn't as easy as it was.

The final caveat is that I do not have purity data yet... hopefully I can get an xrf test done soon-ish and I will post the results.

M

 

[mls26cwru]

okay, so I got a preliminary result back and it is a bit perplexing... the read out was this:

element : (%) : (+/-)
LE : 93.3% : 0.33
Pd : 2.78% : 0.12
Ti : 2.48% : 0.19
Pb : 1.00% : 0.05
Cu : 0.38% : 0.02
Mo : 0.0068% : 0.0009

Sample was the 3 gram button above and was from an Olympus family xrf gun.

Now, I can understand why all the other trace elements would be there, even the titanium... but I really cant think what light elements (LE) would be considering the process and steps I went through, and certainly not that would account for 93% of the weight... even my stannous tests were very strong... much more than ~0.1 grams of Pd in a liter+ of leech would be.

My thoughts, and hopefully one of you guru's would be able to chime in here.... is the LE an artifact of melting the palladium sponge. I used an Oxy/propane torch. after it was melted, I dialed back the oxygen flow to the flame and let it cool. upon solidification, it had a matte finish but there was no discoloration to the fresh melting dish. the button is hard and not very malleable. Is the LE reading because of absorbed oxygen on the surface, or maybe throughout the sample?

PROCESS (material assumed to have Tin, Lead, copper, titanium dioxide, solder mask, palladium, silver. material base is a ceramic)
original material goes into an initial HCl leach. (24hrs)
material rinsed in DI water
misc. attached copper is removed
copper free material rinsed in DI water
Copper free material soaked in HCl to leach material from newly exposed surfaces (24hrs)
material removed from HCl leach, rinsed and dried in oven (24hrs)
material is placed in hot nitric leach
material is inspected to be barren of PM's
Nitric leech is collected and combined with rinses and solution allowed to settle to decant.
Nitric leech is decanted and filtered with no additional DI water added.
HCl added to clear nitric solution to drop Silver as AgCl. Decanted/filtered.
Nitric/HCl solution is denoxed with sulphamic
Denoxed Nitric/HCl solution is decanted/filtered
Denoxed Nitric/HCl Ph is adjusted to Ph 2 with small additions of NaOH
Sodium Formate is added at this point to precipitate Pd.
Precipitate is rinsed with DI water 3 times.
Precipitate melted

 

[mls26cwru]

Any thoughts from anyone regarding the xrf reading?

*fingers crossed*

 

[Jiraiya]

Hmmm.

Extremely unlikely that you would have light metals or Ti in a precipitate, not to mention entering an alloy during torch melting. Those results look like a scan done on a slag, rather than a PM button.

XRF scan depth is very low, so you may have just hit a slag layer, in stead of the button?

There are lots of variables here, so can't say for sure what is wrong ... but something is wrong ;-)

Regards

 

[Geo]

Light elements that does not ignite when exposed to oxygen starts at magnesium until calcium. Includes Aluminum, silicon, phosphorus, sulfur, with chlorine and argon as a gas and then potassium and calcium. Below that is beryllium, boron and carbon. Could there have been any of these metals present during the melt?

 

[snoman701]

Impossible to know without seeing the actual spectra from the xrf read.

Can you get the bead back, or is the deal done?

 

[snoman701]

Sample prep is extremely important for XRF. I'm wondering if you have a refractory surface of Pd alloyed with silica/ti/etc as well as the PdO that is making up the full counting depth of beam penetration.

 

[Lou]

The Ti and light elements might just be slag from the dish. Without seeing the melted button, it's hard for us to describe. I would probably say the XRF is wrong, as torch melting anything with titanium in it inevitably leads to the titanium exiting the stage as TiO2. Lead probably also would've been long gone.


Usually formate Pd is really only contaminated with Ag or other PGMs.