How to determine purity of silver 3N, 3N5?, 4N?

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

snoman701

Well-known member
Joined
Oct 8, 2016
Messages
2,108
Location
SE MI
Essentially, I need a litmus test to determine if it actually is 3N silver, but I don't think there's an easy answer for this based upon past discussions.
 
Fire assay isn't for high grade refined silver, nor is the Volhard titration or the Gay Lussac titration. Usually high grade refined metals are run by difference where the impurities are quantified by AA or ICP and subtracted from 100%.
 
4metals said:
Fire assay isn't for high grade refined silver, nor is the Volhard titration or the Gay Lussac titration. Usually high grade refined metals are run by difference where the impurities are quantified by AA or ICP and subtracted from 100%.

So when high grade metals were produced prior to the technology of AA or ICP was it just "as pure as we can do"? AA is relatively new...at least in the last 50 years. I know that 3N bullion really wasn't a public offering until Engelhard in the 80s, and I don't know of any products where 3N purity is a requirement.
 
There is a test that can give you an indication of purity but it is not quantitative.

Dissolve a sample of the silver in reagent Nitric acid. There should be no color to this solution, if there is, it is because of the impurities.

Evaporate this silver nitrate in a porcelain dish still looking for color. When it gets dry enough not to spit and close to crystallizing, raise the temperature to 825ºF and hold it there for a while. At this temperature the dish will have a faint barely visible glow.

Let the crystallized silver nitrate cool and re-dissolve it in hot distilled water. Now you are looking for sediment on the bottom. Sediment indicates metal oxides that form when the silver nitrate fuses at 825ºF.

All of the impurities in your silver will form the sediment with this test. So if you have no coloration and no sediment, you are over .9999 pure silver.

But if you do have coloration or sediment, like I said this isn't quantitative.
 
That's a really good method, 4metals. I wonder if there is a possibility that an alternative method might work too.

When reading the first post I started to think about the problem and before you posted the procedure with fusing the silver nitrate I came up with a similar procedure. I haven't tested it so it's purely speculative that it would work.
There are a couple of problems with my method, but the main advantage is there is no need for a controlled temperature fusing of the silver nitrate.

1. Dissolve the silver in nitric acid. Collect any solid remnants if any.
2. Add diluted HCl to form silver chloride. Use diluted so all the silver chloride precipitate.
3. Filter off the silver chloride. Wash well with diluted HCl.
4. Evaporate the nitric and hydrochloric acid, any solids left will be nitrates or chlorides of contaminants.

Drawbacks that I see is:
- If there is a lot of lead contaminating the silver, some will be trapped as lead chloride with the silver chloride.
- Any mercury will be trapped with the silver chloride.
- Chloride or nitrate salts are heavier than corresponding oxides so it will be more inaccurate than silver nitrate fusing.

4metals, have you ever seen a method like this used? Is there a reason it wouldn't work?

A drawback with both methods are that it only gives a hint about the purity, oxides, chlorides and nitrates are all heavier than the metals contained. If we get 0.1g leftover per 100 gram of metal then we know that the purity is at least 999/1000 but it can be better, depending on the contaminant and method some of that weight is oxygen (or chloride or nitrate in my speculative method).

Göran
 
4metals said:
Fire assay isn't for high grade refined silver, nor is the Volhard titration or the Gay Lussac titration. Usually high grade refined metals are run by difference where the impurities are quantified by AA or ICP and subtracted from 100%.


There is an ISO method for the detrmination of impurities in silver by difference with ICP-OES. Likewise for potentiometric determination of silver, which I still maintain to be the most accurate method for silver determination in silver nitrate. I would trust the accounting on that better than any other method as the RSDs are 10-30 less than the best ICP analyses.
 
Goran,

Any gravimetric method where you are collecting and weighing such miniscule quantities is going to introduce errors. The method I proposed does not weigh the residue it merely puts them out there for you to observe and infer purity. (or not)

I agree with Lou as a potentiometric titration can be exceptionally accurate because there are very precise burettes made that can minimize the volumetric errors.

But the OP asked for a litmus test, in my mind that is an easy visual test that allows you to make conclusions in a timely manner. If this is the bar, AA, ICP, and Potentiometric titrations are all out because they're complicated and time consuming. Not to mention their availability to a small scale refining operation.
 
Ok, so the biggest reason I'd like to know, is because until I know that my silver is 3N, I can't stamp it.

Potentiometric Titration

What kind of cost am I looking at to get set up to run this?
 
The theory behind the Silver Voltameter may be found here and related publications. The Silver Cell was used by the National Bureau of Standards to determine the value of the Ampere, thus the amount of Silver deposited with a known Current over time may be used to calculate the purity of the Silver. With an accurate current meter and scales, it is currently a simple process.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1091173/pdf/pnas01946-0074.pdf
 
snoman701 said:
Ok, so the biggest reason I'd like to know, is because until I know that my silver is 3N, I can't stamp it.

Potentiometric Titration

What kind of cost am I looking at to get set up to run this?

You can do that in the US? Here in the UK your material has to go to an approved site and be assayed before a legally binding stamp can go on it.
 
I can stamp steel 24k gold as long as I don't get caught.

There are gov't guidelines for fineness...but it is up to the individual mint or refiner to ensure that their product maintains quality control, to maintain that fineness. I am not sure how serious the FBI and FTC are about enforcement, but I'd rather not find out.

But yes, it is legal to stamp our own. You must do so with a trademark or legal name that is registered to you. That is, I can't just stamp .999 silver, without stamping .999 fine silver, snoman mint...and have that trademark registered in my name.
 
snoman701 said:
I can stamp steel 24k gold as long as I don't get caught.

There are gov't guidelines for fineness...but it is up to the individual mint or refiner to ensure that their product maintains quality control, to maintain that fineness. I am not sure how serious the FBI and FTC are about enforcement, but I'd rather not find out.

But yes, it is legal to stamp our own. You must do so with a trademark or legal name that is registered to you. That is, I can't just stamp .999 silver, without stamping .999 fine silver, snoman mint...and have that trademark registered in my name.

Wouldn't it be easier to just sell the silver to me? :mrgreen:
 
Iggy-poo said:
The theory behind the Silver Voltameter may be found here and related publications. The Silver Cell was used by the National Bureau of Standards to determine the value of the Ampere, thus the amount of Silver deposited with a known Current over time may be used to calculate the purity of the Silver. With an accurate current meter and scales, it is currently a simple process.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1091173/pdf/pnas01946-0074.pdf
Are you sure? I can't see any way to get purity based on that cell. It is a clever way to relate ampere to mass and my guess is that they are using quite pure silver nitrate in the cell to get a reproducible standard.

Can you make a list of the steps needed to measure fineness?

Göran
 
g_axelsson said:
Iggy-poo said:
The theory behind the Silver Voltameter may be found here and related publications. The Silver Cell was used by the National Bureau of Standards to determine the value of the Ampere, thus the amount of Silver deposited with a known Current over time may be used to calculate the purity of the Silver. With an accurate current meter and scales, it is currently a simple process.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1091173/pdf/pnas01946-0074.pdf
Are you sure? I can't see any way to get purity based on that cell. It is a clever way to relate ampere to mass and my guess is that they are using quite pure silver nitrate in the cell to get a reproducible standard.

Can you make a list of the steps needed to measure fineness?

Göran

Here's a good starting point. The references provided should give you all the information you need. Additional references are available with a simple search at NIST.gov:
http://nvlpubs.nist.gov/nistpubs/jres/84/jresv84n2p157_A1b.pdf
 
Topher_osAUrus said:
snoman701 said:
I can stamp steel 24k gold as long as I don't get caught.

There are gov't guidelines for fineness...but it is up to the individual mint or refiner to ensure that their product maintains quality control, to maintain that fineness. I am not sure how serious the FBI and FTC are about enforcement, but I'd rather not find out.

But yes, it is legal to stamp our own. You must do so with a trademark or legal name that is registered to you. That is, I can't just stamp .999 silver, without stamping .999 fine silver, snoman mint...and have that trademark registered in my name.

Wouldn't it be easier to just sell the silver to me? :mrgreen:

TONS...but I want to coin.
 
Iggy-poo said:
Here's a good starting point. The references provided should give you all the information you need. Additional references are available with a simple search at NIST.gov:
http://nvlpubs.nist.gov/nistpubs/jres/84/jresv84n2p157_A1b.pdf
In that text they started with "Silver shot, said to be 99.999% pure by the manufacturer", how would unknown contaminants affect the procedure?
If I get you right you would dissolve silver then by electrolysis plate it back onto a platinum cathode? Are you integrating the current or weighing the cathode afterwards?

Göran
 
g_axelsson said:
Iggy-poo said:
Here's a good starting point. The references provided should give you all the information you need. Additional references are available with a simple search at NIST.gov:
http://nvlpubs.nist.gov/nistpubs/jres/84/jresv84n2p157_A1b.pdf
In that text they started with "Silver shot, said to be 99.999% pure by the manufacturer", how would unknown contaminants affect the procedure?
If I get you right you would dissolve silver then by electrolysis plate it back onto a platinum cathode? Are you integrating the current or weighing the cathode afterwards?

Göran

The silver shot was used as proof of concept. You would do the same with your unknown to calculate the Silver recovered, which will give you your purity.
I built a similar apparatus in 1980-81 and it passed a stringent QC. It was a bit simpler and just used current and time to determine the amount of Silver. The voltage across the cell will rise at the endpoint, stopping a counter that is keeping track of time. If you know the current and the time, it is simple to calculate the amount of Silver plated on the Cathode. No need to weigh the Cathode. The Cathode I used was Silver plated with Gold. You can backplate the Silver by reversing the current. It's a reversible system, so you can check your value by plating it back and recalculating.
As far as unknown contaminants go: You could get side reactions if the Silver was dirty, but I would only use the cell to check supposedly pure Silver as a Quality Control check, not for incoming, potentially dirty Silver.
 
So what you suggests is to basically use a modified voltammetry system.
https://en.wikipedia.org/wiki/Voltammetry

Göran
 
Back
Top