This is my first forray into taking platinum further than cementing it out of solution. I've AR'ed a couple dozen automotive oxygen sensors to strip platinum. I concentrated it down, dropped a nice quantity of yellow platinum salt with ammonium chloride, and filtered it. Now I'm left with a still reddish colored platinum solution diluted with ammonium chloride. It tests strongly for PT with the stannous test but adding more ammonium chloride doesn't yield any more PT salt. I'd like advice on possibly concentrating this solution down to attempt another drop. I'm hesitant to just start heating it given (residual) nitric and ammonium are both involved at this point. At least not unless you guys give me the all clear. The stannous test makes me think there is still enough value there to not just toss it in the stock pot just yet.
Precipitating platinum questions
- Thread starter Duxthe1
- Start date
Help Support Gold Refining Forum:
- Joined
- Feb 25, 2007
- Messages
- 8,360
The color you describe does not sound like platinum. That's not to suggest it isn't, but it may not be. The reddish tint you mentioned is troubling to me, as platinum in solution is not red--not even when concentrated. I can't imagine it would have a red tint after recovery.Duxthe1 said:
Are you familiar with the color reactions of palladium when tested with stannous chloride? Do you happen to have any DMG, so you can test a second way? That would remove all doubt, that it is or is not palladium.
One of the possible inclusions could be iron, which could account for the color. Hoke describes a test solution that can be used to make that determination.
It is my opinion that if you do, indeed, have platinum, you most likely will struggle recovering any more of it chemically, as it rarely precipitates fully. That's the nature of the beast. My advice would be to send it to the stock pot, or use some zinc flour and cement it down, so it can be included in your next batch for extraction and recovery by ammonium chloride.
Harold
jimdoc
Well-known member
Automotive oxygen sensors are supposed to be platinum only,
no palladium like in cats.
Jim
no palladium like in cats.
Jim
Thanks for the replies. The extraction was done with just the ceramic probes from the sensors so I was expecting only PT. After contemplating your replies concerning contamination I am now recalling that they also contain Zirconium. If ZR is causing the red coloration that would explain why the solution didn't much change color after dropping the yellow PT salt. It would also mean my PT salt will have some amount of ZR contamination requiring further purification.
Assuming I am right about the ZR the next question is how to further refine my PT. Calcine and HCL bath(s)? I've read about using ammonia to dissolve the salt and form base metal hydroxides that can be filtered. Would that approach be more appropriate. My goal is metallic PT as chemically pure as is possible in the home lab.
Assuming I am right about the ZR the next question is how to further refine my PT. Calcine and HCL bath(s)? I've read about using ammonia to dissolve the salt and form base metal hydroxides that can be filtered. Would that approach be more appropriate. My goal is metallic PT as chemically pure as is possible in the home lab.
- Joined
- Feb 25, 2007
- Messages
- 8,360
My knowledge of the platinum metals is very limited. If you're fortunate enough to get Lou's attention, he may have some advice to help you with purity. If not, there's nothing wrong with the idea of calcining the recovered platinum, then giving it a good wash in hot HCl. I'd then suggest you re-dissolve and re-precipitate. Each time you process, you should leave more of the contaminants behind. A second refining for gold does wonders, so it surely couldn't hurt with platinum. Base metals shouldn't be all that hard to eliminate.Duxthe1 said:
Thanks for the comments on the platinum only, jimdoc. I never attempted to process any of them, so I was totally clueless.
Harold
Some pictures would be helpful, and some idea on the quantity. Usually those sensors are with zirconia, which isn't really soluble in aqua regia. Stannous chloride is a very sensitive test, so you may be confusing iron's strong coloration for platinum, and only missing a small amount.
Please use the very best hygiene when handling platinum salts--treat them with great care!
FYI, perhaps consider zinc cementation on the remaining solution, decanting, rinsing, and simmering in HCl.
Take that de-zinc'd material and dissolve in aqua regia, along with your ammonium hexachloroplatinate (slowly add in the nitric acid), evaporate to a syrup several times, each time adding a small amount of hydrochloric acid until it stops producing an off gas. Not driving off nitrates can prevent a complete precipitation. You'll want it to be a syrup. The zinc-treated solutions that test barren for Pt should be neutralized with caustic to pH 10.5, the zinc hydroxide filtered off, and the pH adjusted to neutral.
Then try again by adding the ammonium chloride solution to a 60-80oC chloroplatinic acid syrup, allowing the solution to cool to room temp, then filtering on an ashless paper. Rinse the yellow salt with cold 10% ammonium chloride solution which has been acidified with a small amount of HCl. Your filtrate should run clear.
Please use the very best hygiene when handling platinum salts--treat them with great care!
FYI, perhaps consider zinc cementation on the remaining solution, decanting, rinsing, and simmering in HCl.
Take that de-zinc'd material and dissolve in aqua regia, along with your ammonium hexachloroplatinate (slowly add in the nitric acid), evaporate to a syrup several times, each time adding a small amount of hydrochloric acid until it stops producing an off gas. Not driving off nitrates can prevent a complete precipitation. You'll want it to be a syrup. The zinc-treated solutions that test barren for Pt should be neutralized with caustic to pH 10.5, the zinc hydroxide filtered off, and the pH adjusted to neutral.
Then try again by adding the ammonium chloride solution to a 60-80oC chloroplatinic acid syrup, allowing the solution to cool to room temp, then filtering on an ashless paper. Rinse the yellow salt with cold 10% ammonium chloride solution which has been acidified with a small amount of HCl. Your filtrate should run clear.
The original solution looked slightly more orange than the "after drop" solution which is now more red. It was approximately 300ML that was concentrated down to less than 100ML after which the ammonium chloride was added. The salt appeared immediately and settled quickly. The salt is fine and there is noticeable amount that made it through the first filtering.
If you recall from another thread about palladium I had doubts about my DMG being in alcohol solution. So I have slowly been evaporating it down with the intention to re-hydrate with water. On your advice I went ahead and took a sample of the original solution and added a little of my DMG. Sure enough there came the fluffy yellow tell-tale of PD. I've never heard of PD in oxygen sensors but you can't argue with the chemistry. FWIW, to those in posterity that read of my saga, it may be worth noting that the sensors all came from Bosch and were installed on high end German automobiles.
- Joined
- Feb 25, 2007
- Messages
- 8,360
While I never processed a lot of palladium or platinum (about 75 ounces of each in my many years of refining), I was very familiar with the solutions, as I processed a great deal of dental gold, so I saw traces on a regular basis. What you had described fit my experience, so it was difficult for me to not suspect the presence of palladium, in spite of the fact that the O sensors are reputed to be just platinum.
Something of value for many readers should have been learned here. Testing is extremely important, as you can never be sure what you may be processing. I had an ongoing curiosity, so I tested my solutions regularly and routinely. I wanted to know and understand why a solution was a given color.
Hoke's book provides excellent guidance in the preparation of standard solutions, as well as testing solutions. I highly recommend a person study that part of the book until it is VERY well understood.
Harold
Something of value for many readers should have been learned here. Testing is extremely important, as you can never be sure what you may be processing. I had an ongoing curiosity, so I tested my solutions regularly and routinely. I wanted to know and understand why a solution was a given color.
Hoke's book provides excellent guidance in the preparation of standard solutions, as well as testing solutions. I highly recommend a person study that part of the book until it is VERY well understood.
Harold
niteliteone
Well-known member
Wouldn't the oxygen sensor need to be "bi-metallic" to create the piezo electric effect that is read by the vehicles computer ???
According to http://www.platinummetalsreview.com/resources/view-questions-answers/pgms-in-oxygen-sensors-and-spark-plugs-2/
My guess on how an oxygen sensor based on PGM works is that oxygen is adsorbed on the surface of the metal and affects the electrical potential. If two pieces of metals with different amount of oxygen on the surface comes in contact with each other there will be an electrical potential differential that can be measured. So in effect, we don't measure the oxygen level directly, we measure the difference between air and exhaust. But that's only my guess.
While trying to find out exactly how oxygen sensors work I found this patent
http://www.google.com/patents/US4303490
In it they talk about the amount of platinum or palladium needed for the sensor. 5-10 mg palladium or 7-13 mg platinum.
This gives an indication of how much palladium or platinum there is in a single oxygen sensor and possible maximum yield.
Göran
niteliteone, the piezoelectric effect is when a deformation of a crystal (often quartz) creates an electrical field, or the other way around, an electrical signal creates a mechanical deformation of a crystal. It has nothing to do with oxygen sensors as far as I know.
My guess on how an oxygen sensor based on PGM works is that oxygen is adsorbed on the surface of the metal and affects the electrical potential. If two pieces of metals with different amount of oxygen on the surface comes in contact with each other there will be an electrical potential differential that can be measured. So in effect, we don't measure the oxygen level directly, we measure the difference between air and exhaust. But that's only my guess.
While trying to find out exactly how oxygen sensors work I found this patent
http://www.google.com/patents/US4303490
In it they talk about the amount of platinum or palladium needed for the sensor. 5-10 mg palladium or 7-13 mg platinum.
This gives an indication of how much palladium or platinum there is in a single oxygen sensor and possible maximum yield.
Göran
niteliteone
Well-known member
My bad. I used word from memory to describe a somewhat similar process. I should have looked it up and I would have found a much better description of what I was trying to get to.g_axelsson said:
I found this on wiki;
http://en.wikipedia.org/wiki/Oxygen_sensor
Thanks for the link, I had a faulty memory that there wasn't so much information on the wikipedia page so I never looked at that yesterday. I wasn't so far off with my guess on how it worked. 
Göran
Göran
Tub Buster
Well-known member
- Joined
- Sep 5, 2013
- Messages
- 53
Duxthe1 wrote:
Assuming the solution was denoxxed, the precipitated Pt will be contaminated with Pd. This might be a good occasion to employ a technique referenced by Hoke on page 157 of her book:
The original technique, by Whitely and Dietz, employs chlorine, but this should not be necessary in the present case, as no gold was precipitated. It has the virtue of simplicity - a common, safe reagent, ethyl alcohol, is all you need.
Assuming the solution was denoxxed, the precipitated Pt will be contaminated with Pd. This might be a good occasion to employ a technique referenced by Hoke on page 157 of her book:
The original technique, by Whitely and Dietz, employs chlorine, but this should not be necessary in the present case, as no gold was precipitated. It has the virtue of simplicity - a common, safe reagent, ethyl alcohol, is all you need.
samuel-a
Well-known member
- Joined
- Oct 7, 2009
- Messages
- 2,190
Duxthe1 ,
Your SnCl2 test sure looks like Palladium to me.
http://www.goldnscrap.com/index.php...test-for-gold-platinum-and-palladium-presence
Your SnCl2 test sure looks like Palladium to me.
http://www.goldnscrap.com/index.php...test-for-gold-platinum-and-palladium-presence
Correct me if I'm wrong but wouldn't iridium be red in his solutions. But it really looks like an sn04 PT solution with peroxide in it
Harold has not been here this year.
Welcome to us by the way.
