Electrochemistry Silver cells in series trial run

[Palladium]

Looking good!

 

[goldsilverpro]

Ralph,

Assume you set up a single cell and determine that you need 10A @4V.

For 4 identical cells in series, the PS would be set at 10A and 16V (4V/cell) - Constant amperage, additive voltage.

For 4 cells in parallel, it would be set at 40A (10A/cell) and 4V - Additive amperage, constant voltage.

In all 3 cases, you would have A X V = 10 X 4 = 40 watts per cell. If all is perfect (it never is!!), the same amount of silver would be produced in each cell in the same amount of time.

 

[mike6]

GSP,
You wrote,

"You could probably increase the efficiency by lowering the voltage to about 3.0-3.5V per cell, thus lowering the amperage a bit, or by adding more silver to the solution."

Please help me understand. By lowering voltage, and thus amperage, would deposition rate over a given time not decrease, thus lowering efficiency?
Mike

 

[g_axelsson]

Lower amperage would decrease the deposition rate but the efficiency is measured as the deposition rate divided by the theoretical maximum at that current. For the silver cell it is 1 silver atom per electron, but other parasitic effects, for example splitting water into oxygen and hydrogen uses up some of the electrons so the practical efficiency is less than 1.

Look at the close up of the anode basket, the bubbles around it is probably from splitting water.

Göran

 

[mike6]

Thanks Goran,
It seems I am confusing process speed with electrical efficiency, but the plus side is, maybe I can generate enough oxyhydrogen (knallgas) to power my melting torch! :lol:
Mike

 

[mike6]

Gentlemen,
Building on the advice offered on the 4 cells in series exercise, I have just completed a 2 cells in series run. I paid much more attention to capturing input data, and feel I am able to present a more comprehensive picture for your comment.

A new power supply unit was tested, rated for a maximum 16V, and 30A output.

The same electrolyte formula as discussed earlier was used.

Total run time, 26 hours.

Cells were harvested, and electrolyte mixed and filtered 3 times during the run, every 8 hours or so.

New anode electrodes were cast, using .999 silver, weighing 250 grams each.

Copper carrier wires were trimmed to equal lengths.

Readings were taken every 90 minutes, or so, to record voltage and amperage values.

The following values are based on averages over 16 readings.

Power supply output voltage, 8.03V, with a standard deviation of 0.107.
Power supply output amperage,17.64A, with a standard deviation of 3.339

Cell A: 3.90V, with a standard deviation of 0.10.
Cell B: 3.81V, with a standard deviation of 0.11.

The standard deviation values on voltage are close to zero, indicating confidence in the mean voltage readings, however, the relatively high standard deviation value on amperage has confused my efforts in determining how close I am to 100% of possible theoretical deposition.

If I am correct, at 100% of theoretical, a 26 hr run of 2 cells in series at an average input amperage of 17.64A should yield 3692.67grams of silver.

My total actual silver yield after 26 hrs was 3596.52grams. I calculate this to 97.3% of theoretical.

It is in my nature to strive for perfection, but I have forced myself to take cognisance of the law of diminishing returns. For now I would be happy with 95%, I'm just not sure that this is what has been achieved.

your inputs, as always, would be greatly appreciated.
Mike.

 

[g_axelsson]

That's a nice setup.

I'm not surprised in the readings you get, the voltage is regulated so it shouldn't change much.
The current will vary depending on the varying conditions in the cell.
- The silver concentration changes during a run unless you run pure silver.
- The distance between anode and cathode changes as the silver builds up.
- Changing electrode environment as the pollutants adds up, for example clogging the anode bags.

You will never reach 100% theoretical max and there is no reason to aim for it, the real goal is to get a pure product. Do not sacrifice purity to gain process efficiency.

You could probably get a lot closer to 100% if you dropped the voltage way down, but that would take a long time to run instead.

Göran

 

[FrugalRefiner]

Mike, I wouldn't lose any sleep over the efficiency. There will always be some loss due to the resistance of electrical flow through the electrolyte.

I'm not great at electrical calculations, but I'll try this and I'm sure someone else will correct me if I get it wrong. We pay for electrical usage in watts or kilowatts used. Volts x Amps = Watts. Using your figures, 8.03V x 17.64A = 141.65 Watts. If you run at that rate for an hour, you'll use 141.65 watt hours, or 0.14165 kilowatt hours. Running for 26 hours, you'll consume 3.68 kilowatt hours. I don't know what electricuty costs where you are, but I pay a little under $0.06 (6 cents) per kilowatt hour. The electricity you used would cost me around 22 cents. The 5% loss in efficiency amounts to about 1 cent of loss.

You produced 3596.52grams of crystal. At $16.76 per ounce, the silver is worth $1,938.18. Don't worry about the penny. 8)

Dave

 

[METLMASHER]

Well done, Mike6. :mrgreen:

 

[FrugalRefiner]

Cells were harvested, and electrolyte mixed and filtered 3 times during the run, every 8 hours or so.

If you really want to get serious about keeping the electrolyte the same, you can set it up like the commercial systems I've read about. The cells are set up on a stair step type of arrangement. An outlet near the top of the highest cell leads to an inlet at the bottom of the next one down, and so on to the last cell in the series. At the last cell, there is a pump that pumps electrolyte back up to the highest cell, or to a separate operation where copper is removed before returning to the highest cell. The circulation also helps prevent silver depletion of the electrolyte near the cathode, which can lead to codeposition of copper, especially when you're running hard.

Or you can just build a bigger cell. :lol:

Dave

 

[goldsilverpro]

To calculate the efficiency, you also have to know the silver content of the bath before and after the run. It most always changes. In this case, I would bet that the silver content has decreased and the efficiency will be less than what you figured. 97% always sounded too high to me.

When the voltage is too high (i.e., higher than it should be - more than double in your case), it is most always an indication that something is wrong.

1st possibility:

In the traditional Thum or Moebius silver cells, the anode area and the effective cathode area are about the same. In yours, a bowl type arrangement, the cathode area is as much as 10 times larger (and maybe more) than the anode area. On the anode, you can only use the outer surface of the basket to calculate the area. From your amperage number and, by looking at the cell, I would guess that the cathode current density is about 25 A/ft2 (ASF), which is on the low side but is fine. It should be between 25-50 ASF. However, the anode current density might be as much as 250 ASF, which is way too high. As you dissolve silver, it combines with nitrate ion at the anode surface and, if the anode current density is way too high, the population of nitrate ion in contact with the anode decreases. If this happens, the anode efficiency will decrease, since there's not enough nitrate ion in the immediate vicinity to go around. The result is that you will plate out at a higher rate than you dissolve and, therefore, the silver in the solution will decrease. Another result of this is that the resistance at the anode will increase and, therefore, the voltage needed to pull x number of amps will be higher.

2nd possibility:

If the anode bag mesh is too fine, solution exchange will be hindered and the result will be a slow depletion of silver in the solution. This will also increase the resistance and, therefore, the voltage.

Maybe a combination of the two.

 

[Palladium]

In my video's i teach complete exhausting of the nitric acid during cell solution make up. I have come to learn to add free nitric acid to the cell several times during a run. It will help with residues in the filter.

 

[mike6]

Gentlemen,
Thank you for your comments to date.

How is it that GSP is right 99% of the time, with a standard deviation of 0.011? 8)

The Ag content in the electrolyte has indeed been depleted, quite significantly.

I have just completed the end of the run cleanup, and have the following figures for your perusal.

4100.54g. Total weight of cemented silver shot introduced into anode baskets.

506.49g. Cemented silver shot, (rinsed free of slimes), left over in anode baskets. (Will be introduced to next run's anode shot). Without taking the slimes into account, this means that only 3594.05g of shot was actually processed.

1347.20g. Total weight of Ag used to make electrolyte. At the end of the 26 hour run, the Ag was precipitated out of the electrolyte solution using copper. Only 588.47g remain. Although the "missing" 758.73g has been deposited at the cathode, it did not originate from the shot in the anode basket, thus throwing my "97.3%" efficiency figure in the garbage, , and proving GSP's acute observations. Using the previous, as well as the following figures, GSP, would you assist me in calculating a more realistic efficiency figure, which I hope will still allow me to retain some dignity. . (...Talk about the forum adage, " If it sounds too good to be true..."...) :lol:

All rinse water was processed with copper, to yield 87.02g of silver cement.

The 500g .999 silver anode electrodes "lost" 47g in weight and now weigh 453g.

86.81g of Slimes were filtered, and saved for later processing.

All weights quoted are rinsed, filtered and dry.

Your perusal, and resultant commentary is, as always, greatly appreciated.
Mike

 

[MarcoP]

I always thought the silver nitrite wouldn't lose any silver but you say some is cementing at the cathode. I'm still trying to figure out how process silver for a client, for example if he give me 1Kg of sterling 800 what's the route if I want to return him ~800gr of pure silver from his material.

 

[goldsilverpro]

Before attempting the math, I see a grave error I made (the other 1%, I guess). For the voltage, I assumed there was only 1 cell, whereas there were 2. Since voltage is additive in a series circuit, this would be a little over 4V per cell. That's a little high (for the 25ASF on the cathode, I would guess an ideal cell would run at about 3.3V), but maybe not high enough to severely affect the efficiency and contribute much to silver depletion.

Looking again at the close up of the working anode, that cloth sure looks heavy to me and could very well be the main culprit. What type cloth are you using? Too fine a cloth can partially plug up the system and deplete the silver quickly. The cathode will keep churning away at full speed while the anode will be quite sluggish. The only cloth I have used is a double thickness of thin unbleached muslin (light tan color). According to the books, the muslin should have 100 threads per inch. Some people have used other types of cloth successfully. Maybe they will chime in. A warning about muslin - all that I have used contained a sizing which, at first, blocked all of the current - the ammeter read zero. Over time, the size started dissolving and the current progressively increased. After about 4 hours under power, all the size was gone and the amperage was what it should have been. The sizing could possible be removed by pre-washing. By frequent adjustments, I kept the voltage at a fairly constant level (usually about 3.5V for my setup) during the 4 hours.

Another thing you must consider is the purity of the 3600g of cemented shot that was consumed. Assuming it was 99% silver, you added 36g of copper, which removed 36 x 3.4 = 122g of Ag from the solution. This could be less with excellent rinsing of the cement. It's not that easy to get much better than 99%, though.

Everybody's silver cell design and setup will be, at least, a bit different, and each will have its own unique set of problems to adapt to. In your case, it can be fairly simple. First off, I would try different filter cloths. You need it to be (1) impervious to the solution; (2) fine enough to retain the slime, yet, (3) coarse enough to freely allow solution exchange through it.

Were I you, I would definitely have a way to analyze the silver in solution. The simplest and fastest would be the Volhard titration method, which I would probably have to spend about $100 to set up. A cheaper but more laborious and slower way would be to extract an adequate sized solution sample of known volume, precipitate the silver as silver chloride, rinse thoroughly, dry slowly in darkness to a constant weight, and weigh. AgCl is 75.2% silver. With some ingenuity, which you obviously possess, you could probably now perform the AgCl method with what equipment you already have.

With analysis capabilities, you could then build up the silver content, periodically, with freshly made silver nitrate solution. If the silver nitrate solution is hot and saturated, it can probably be added without having to bail out any of the existing cell solution to make room for it. If you thus maintain the silver concentration, and don't contaminate the solution in any way, the solution should work until the copper level reaches a level where it co-deposits with the silver. Some say about 65-70g/l and some say about 100g/l. Analysis of the Cu is more difficult than with silver. Instead, I often keep a running estimate of the Cu I put into the cell. In any case, there is a trend: the higher the silver content maintained in the solution, the more tolerant the system is towards the co-deposition of copper.

Back to the math. Like the actual silver, scattered hither and yon in your system, information on the whereabouts and quantities of the silver, before and after, is scattered throughout your posts. I've read them several times and am still unsure in certain areas. Can you summarize the numbers?
(1) A concise list of the various quantities and locations of all the silver put into the system, from start to finish.
(2) A concise list of the various locations and quantities of all the silver at the very end.

 

[Palladium]

Some say about 65-70g/l and some say about 100g/l. Analysis of the Cu is more difficult than with silver. Instead, I often keep a running estimate of the Cu I put into the cell. In any case, there is a trend: the higher the silver content maintained in the solution, the more tolerant the system is towards the co-deposition of copper.

Here is an interesting article! :arrow:

 

[Palladium]

In the traditional Thum or Moebius silver cells, the anode area and the effective cathode area are about the same. In yours, a bowl type arrangement, the cathode area is as much as 10 times larger (and maybe more) than the anode area. On the anode, you can only use the outer surface of the basket to calculate the area. From your amperage number and, by looking at the cell, I would guess that the cathode current density is about 25 A/ft2 (ASF), which is on the low side but is fine. It should be between 25-50 ASF. However, the anode current density might be as much as 250 ASF, which is way too high. As you dissolve silver, it combines with nitrate ion at the anode surface and, if the anode current density is way too high, the population of nitrate ion in contact with the anode decreases. If this happens, the anode efficiency will decrease, since there's not enough nitrate ion in the immediate vicinity to go around. The result is that you will plate out at a higher rate than you dissolve and, therefore, the silver in the solution will decrease. Another result of this is that the resistance at the anode will increase and, therefore, the voltage needed to pull x number of amps will be higher.

I think with the rotating cathode cell that i have designed that it may solve some of those problems. The anode basket will set almost directly in contact with the cathode at all times with fresh solution being circulated between them by the rotation of the cathode. Single cells are difficult enough for me. Chris what is the added benefit if any of the cathode area being smaller than the anode area and being in such close proximity? Would that have the opposite effect and lead to an excess in nitrate ions. More being dissolved than deposited? Is this one of the reasons for excess free nitric in the solution? School me please! :mrgreen:

 

[goldsilverpro]

In the traditional Thum or Moebius silver cells, the anode area and the effective cathode area are about the same. In yours, a bowl type arrangement, the cathode area is as much as 10 times larger (and maybe more) than the anode area. On the anode, you can only use the outer surface of the basket to calculate the area. From your amperage number and, by looking at the cell, I would guess that the cathode current density is about 25 A/ft2 (ASF), which is on the low side but is fine. It should be between 25-50 ASF. However, the anode current density might be as much as 250 ASF, which is way too high. As you dissolve silver, it combines with nitrate ion at the anode surface and, if the anode current density is way too high, the population of nitrate ion in contact with the anode decreases. If this happens, the anode efficiency will decrease, since there's not enough nitrate ion in the immediate vicinity to go around. The result is that you will plate out at a higher rate than you dissolve and, therefore, the silver in the solution will decrease. Another result of this is that the resistance at the anode will increase and, therefore, the voltage needed to pull x number of amps will be higher.

I think with the rotating cathode cell that i have designed that it may solve some of those problems. The anode basket will set almost directly in contact with the cathode at all times with fresh solution being circulated between them by the rotation of the cathode. Single cells are difficult enough for me. Chris what is the added benefit if any of the cathode area being smaller than the anode area and being in such close proximity? Would that have the opposite effect and lead to an excess in nitrate ions. More being dissolved than deposited? Is this one of the reasons for excess free nitric in the solution? School me please! :mrgreen:

The amount of nitrate ions in the solution doesn't change unless you add more nitric. They are negative and migrate somewhat slowly towards the positive anode. Also, when silver plates out on the cathode, it releases more nitrate ions. With normal amperage, they replace those nitrate ions at the anode surface that have combined with dissolved silver. If the amperage is too high, they can't do this fast enough and this can result in a lower efficiency.

The rotating cathode will tend to stir the solution and this will replace the nitrate ions at the anode surface at a faster rate. Of course, the closer the anode to the cathode, the faster this would happen. With a closer proximity, there is more danger of shorting by the crystal. The moving cathode just might dislodge all the crystals. I would design a scraper, though, just in case it didn't. The scraper could be as simple as a horizontal rigid plastic rod clamped to the tank and maybe held 1/4" away from the cathode. You don't want even a partial short. You could also use a steel rod insulated with heat shrink tubing (about any plastic except nylon).

 

[mike6]

Back to the math. Like the actual silver, scattered hither and yon in your system, information on the whereabouts and quantities of the silver, before and after, is scattered throughout your posts. I've read them several times and am still unsure in certain areas. Can you summarize the numbers?
(1) A concise list of the various quantities and locations of all the silver put into the system, from start to finish.
(2) A concise list of the various locations and quantities of all the silver at the very end.

Sir,
My first attempt at running cells in series was with a 4 cell setup.That run, I realize contains insufficient data to include in the calculation pertaining to the latest run, which had only two cells in series, and thus, I feel that the two experiments should be viewed in isolation.

I have scaled the individual cells to produce 2kg to 2.2kg of fine silver crystals per run. Each new run starts "fresh", with a fully charged batch of electrolyte. At the end of the run, all remaining silver is precipitated from the spent electrolyte, as well as from the rinse water used to clean the crystals during harvest, for re incorporation into the next run, weather as a constituent of fresh electrolyte, or as unprocessed shot for introduction into the anode basket.

Total weights and locations of Ag in 2x cells at start of run.

1347.20g. Total weight of Ag used to make electrolyte.
4100.54g. Total weight of cemented silver shot introduced into anode baskets.
500.00g. Total weight of 2x .999 anode electrodes.

Total weights and locations of remaining Ag at the end of 26 hour run.

588.47g. Total weight of Ag recovered from spent electrolyte.
506.49g. Total weight of Cemented silver shot, remaining unprocessed in anode baskets.
453.00g. Total weight of 2x .999 anode electrodes.
87.02g. Total weight of silver cement recovered from crystal rinse water.
86.81g. Total weight of Slimes, rinsed, filtered, and saved for later processing.

3596.52g. Total weight of electrolytically refined silver crystals.

miscellaneous.

The cloth being used for the anode bag is indeed bleached muslin, however, two layers of it! Tomorrow I will pick apart a sample and count the threads, (warp and weft). Micro surgery! :lol:

Today I ordered a burette, with which to try the Volhard titration, and although I grasp the concept, the method and indicators used, currently elude me. The forum's information cupboard on this subject, seems bare! The internet seems obsessed with the titration of cheddar cheese! Put me out of my misery, please!
Mike

 

[5ixb]

Today I ordered a burette, with which to try the Volhard titration, and although I grasp the concept, the method and indicators used, currently elude me. The forum's information cupboard on this subject, seems bare! The internet seems obsessed with the titration of cheddar cheese! Put me out of my misery, please! Mike

Order GSP's e-book.
He puts it in lay man's terms