Reducing Silver nitrate to metallic silver

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Just an update on the sodium formate method, recently I went back to a refiner who I set up using the formate method. He's been running it for 2 years now and has settled down to refining 500 oz/day of sterling silver. He recently got in some silver odds and ends and processed them the same way as usual. The resultant silver was seriously contaminated with lead. Now I had set them up to test for lead while assaying the bars (qualitatively) but that procedure somehow got swept under the rug.

We ran a few batches, which I pre screened for lead, and they came out well. To be sure we sent it out to LeDoux for an ICP and it came back .9996, the second days lot came out .9997. Considering silver has to be .999 for sale they're still in good shape.

The thing I'm most surprised at is the low silver content of the scrap they are getting which is supposedly sterling silver. The best assay is 90%, down to 85 on some bars. Unfortunately that level of copper eliminates a silver cell as an alternative.

We cleaned up the lead contaminated silver by digesting it, dropping it as a chloride, hot water leached the lead, and reduced the chloride with sugar. The bar now had no traces of lead and it was processed by the formate method for sale.
 
I'm testing the method as we speak using analysis grade sodium formate and silver nitrate. Seems to give a very pleasant light grey powder.
 
Funny enough, I still haven't gotten around to the formate reduction.

I really don't like the Karo Syrup/aldose method all that much despite the inherent advantage that it is arguably one of the most selective ways to get pure silver.
Something about dissolving, Cu cementing, and the Thum cell is very attractive... ah yes, no silver chloride!

4metals, ever considered alloying up and putting it through the cell?
 
Lou,

Alloying up and running in a cell is only an option if you have a good quantity of almost fine silver to alloy up with. Lately the guys I've worked for are getting in sterling and coin silver.

I get clients on both ends of the refining spectrum, some have the capital to finance the operation and the support lab properly, lately I've seen the low budget type. If I set up a guy who can put in an AA or an ICP I'd consider a silver cell because the instrumentation makes it easy to monitor the copper levels which climb quickly. It involves quite a bit of electrolyte changing but it is still do-able with less waste acid than the formate method. The low budget guys make management harder. I have yet to find a volumetric analysis for the copper levels in the cell to aid the more frugal refiner. Any suggestions?
 
4Metals,

I'm like Lou in regards to the fact that I've been wanting to try the formate reduction of silver, but I haven't gotten around to it yet.

The day you posted the reaction I bought a few pounds of sodium formate for the reaction, but I haven't got around to trying it out. I only process silver when I don't have any other PM's to run, and lately I've not had a shortage of gold and platinum to process, so the silver will have to wait.

When I get around to trying the process, i likely film it and post it to my website to go along side my other silver reduction videos.

Thanks for the great tip again!

Steve
 
I have yet to find a volumetric analysis for the copper levels in the cell to aid the more frugal refiner. Any suggestions?
I, also, don't know of any good methods of titrating copper in this matrix. I controlled 12, 30 gallon cells by first analyzing the silver with a Volhard titration and then measuring the specific gravity of the the solution with a hydrometer (of which I had a series of). Before then, I had made up about two dozen different silver/copper solutions, compared the silver concentratration verses the made-up copper concentration verses the S.G., and made up a few graphs. From all this, if I knew the silver concentration, I could determine the approximate copper concentration from a graph. I don't think I explained it very well, but it's pretty simple stuff. About the only things that contribute to the S.G. are silver and copper. This isn't rocket science. All you usually need to control an industrial system are close, quicky, approximations.

This is way off the beaten track. One time, I worked for a huge plating shop that had a job to electroless nickel plate a bunch of stainless screens that were used as the sides in a huge centrifuge to filter out sugar crystals. The screens were made up of very tiny hexagonal holes. The holes were too big. We had to close them up a bit to an exact width with electroless nickel. The quality control problem was to measure the final hole size after we plated them. I discovered that, by placing a screen of known hole size over a plated screen on a light table and then rotating one of the screens, the holes visibly increased in size. This is due to the moire' pattern that was produced. As you turned one of the the screens to a certain angle, the illusion of the average hole size on the light table might increase from, say, 1/64" to 1". That way, you could easily measure with a ruler the ratio of the average hole size to the spacing between the holes and, therefore, estimate the actual hole diameter. This is a very difficult thing to explain, but it was very simple to do.

What I'm saying is that, with ingenuity, difficult quality control problems can be made very simple. Comparing the silver analysis with the S.G. to get the copper analysis is one of those things.
 
GSP,

Sounds like an interesting procedure, the silver analysis is simple and accurate. How close to actual levels will a baume reading get you, within 1 oz/gallon? Sounds like you put a lot of effort into making standards and graphs.

I now use the silver content of the anodes going in and the gold content if any, the rest I assume is copper. Using a spreadsheet I can track theoretical levels of copper, or at least what it will be when all of the anodes are dissolved. Analysis is much better!!!!!
 
Sounds like an interesting procedure, the silver analysis is simple and accurate. How close to actual levels will a baume reading get you, within 1 oz/gallon? Sounds like you put a lot of effort into making standards and graphs.
I would say that the copper estimate was accurate within 0.1 - 0.2 oz/gal. It didn't take long to make the standards and graphs, maybe 3 or 4 hours. I just started on the low end and then made additions - with mathematical adjustments, of course. Temperature, of course, had to be considered. I surely used S.G. hydrometers, since baume' is a pain. I think each chart had a single silver content - 1 oz/gal, 2 oz/gal, etc. The intermediates could be estimated (can't think of the mathematical word I really want to use here - interpolated?).

Probably, when all was said and done, my true calling was that of a Q.C. guy - or a math teacher. I never did either.
 
Although it is less than scientific I watch the color of my electrolyte and do my best to monitor my feedstock to keep track of copper concentrations. Maybe I reverse engineer it a bit but I take some of the produced silver crystal near the end of the electrolytes life and after washing dissolve a small sample in nitric. It takes very little copper contamination to change the color of the solution. My greater concern in the cell is palladium contamination.
 
goldsilverpro said:
Probably, when all was said and done, my true calling was that of a Q.C. guy - or a math teacher. I never did either.

I would say your both. And dam good at it to. 8)
 
Oz said:
Although it is less than scientific I watch the color of my electrolyte and do my best to monitor my feedstock to keep track of copper concentrations. Maybe I reverse engineer it a bit but I take some of the produced silver crystal near the end of the electrolytes life and after washing dissolve a small sample in nitric. It takes very little copper contamination to change the color of the solution. My greater concern in the cell is palladium contamination.
It was my routine to make the new batch of electrolyte from crystals that were removed from the cell at the time of operation, and always that which represented the tail end of the session. I ran my cell as a batch cell, it was not continuous. I could almost always part a 200 ounce anode and get pure silver.

As Oz stated, it took almost nothing in the way of contamination to create an indication of purity of the silver. In preparing my electrolyte, the solution was boiled, with a small amount less of nitric than would be expected to dissolve the allotted amount of silver (30 ounces). I'd balance pH by adding a few drops of nitric to the operational cell. By boiling, all traces of the big red cloud were removed from the solution, yielding a clear color. A tint of any description was an indicator of contamination. Running an anode that was too heavily concentrated in platinum group metals was occasionally a problem, causing a trace of palladium to co-deposit, as Oz suggested.

The best indicator I had was the nature of the crystals that were developed. I always started my cell sans copper. It accumulated fast enough that I didn't wish to lose any production. Initially, the deposit of silver was strongly attached to the cell, and often was a solid sheet. It would strip, but required a little effort. I used to use a fiber glass scarper (shop built, with a stainless handle) to remove the deposit. As the copper level climbed, the deposit turned more towards crystals, and were attached much looser.

When the copper level approached being too high, the nature of the deposit changed drastically. Instead of growing crystals, my particular setup would start growing thin strands of silver, much like strands of steel wool. That was a sure sign it was time to change electrolyte, and usually came at the end of the anode's life.

To insure that the production was pure, I always dissolved the fine strands, along with crystal. The strands, being the last to be grown, were an excellent measure of the over-all quality, usually yielding a clear solution. Any crystal harvested from a batch that made an off color electrolyte was placed back in the basket, with an anode placed on top. The crystal was parted a second time to insure purity. There was no need to melt.

I believe I have discussed this previously, but old age is taking a toll on my memory.

Harold

P.S.
Four scotches?

Damn, that sounds good! :)
 
Harold_V said:
By boiling, all traces of the big red cloud were removed from the solution, yielding a clear color
This is important in that the color of the solution while driving out the gases and hot will resemble that of a slight palladium contamination. Until cool it can also have a slight bluish green hue that looks like a slight copper contamination.

Harold_V said:
Running an anode that was too heavily concentrated in platinum group metals was occasionally a problem, causing a trace of palladium to co-deposit
I have not found to date any PGM other than palladium in my electrolyte. When one is concerned about palladium due to feedstock, DMG will tell the tale. Nickel can be a problem as well but is seldom seen unless running material from white gold inquartations in my experience.

I owe a huge thank you to Harold, GSP, and 4metals for most of what I know on this topic.

Thank you
 
Harold,

What you say is great with one small cell. However, I was controlling 12 cells, each of which started with 200 oz of silver. I preferred a more quantitative approach rather than flying by the seat of my pants. I could analyze the 12 cells in less than an hour. Through experience and a little experimentation, the analysis results were easily related to the silver quality.
 
There seems to be lots of coin silver being refined
What would be the best and most econmical method for refining coins
Would it be to use half half nitric and water
then drop the silver with suger then drop the copper
or this method is not that great of an idea
 
Much depends on the purity of the coins.

I would keep sugar away from nitric as it could have unintended entertainment value.
 
4metals,

One could easily analyze for copper concentration using simple colorometric titration with a cheapo Spec 20 spectrophotometer. Ammonia would be the titrant. Of course, a standard curve would need to be constructed.

Chris's idea of using specific gravity might be better. Back in the day it was baume that used to be on the sulfuric acid bottles, indicating their strength. I think the number may of been 67. Way back in the day.


Scotch sounds good. I just got a bottle as a present the other day: The Distiller's Edition Isle of Skye Talisker. It's a double matured single malt, distilled 1990.
 

Latest posts

Back
Top