- Joined
- Apr 17, 2014
- Messages
- 63
consolidation of Loss Gold thread
- Thread starter 4metals
- Start date
Help Support Gold Refining Forum:
- Joined
- Apr 17, 2014
- Messages
- 63
all from polypropylen
Geo
Well-known member
It is not an instantaneous reaction decomposing acid with heat. It takes quite a long time to decompose nitric acid with just heat. Sulfamic acid needs the heat to drive the reaction forward. That is the temp the reaction is at it's most violent. In practice, you do not want to heat the solution that hot to add sulfamic acid.
- Joined
- Apr 17, 2014
- Messages
- 63
thanks, I have one more;
we provide diamond removal service and we run titanium reactors. I always keep the AR temperature around 75c thinking of avoiding boiling. with the information you provided above, what temperature would you recommend? we have 5 hours turnaround time start to finish. I always look for ways to speed it up.
So, if I remember correctly, we pointed out few culprits, where gold could be hiding:
1. retention of gold solution in the system, mainly in some filter parts, which were pointed to be badly designed for the purpose, they should serve
2. incomplete precipitation of gold, which could be found in the second tank sludge (with XRF) after cementation
3. gold being locked up in silver chloride sediment after dissolution in AR
Gold.refinery, what was done with these issues ? Did we improve the yield ? or gold is still escaping. Were some parts of instruments reorganized/altered to better serve their purpose ?
I only want to know, if we are going somewhere.
Geo
Well-known member
When I am working, I don't usually have my IR thermometer handy so I don't have an exact temp. I normally go by touch. I heat the solution until the beaker is slightly too hot to touch the beaker. If I had to guess, I would say between 46°C and 50°C. Just too hot to touch the beaker. Any higher than that and you risk a boil over.
I was under the impression that the touch method was a bit warmer than you have experienced. This chart I found indicates time for a 3rd degree burn.
48ºC is 118ºF for reference. I shower with water warmer than that.
Maybe a thermometer can give more reliable data? 94ºC is 201ºF!
Water Temperature | Adults (skin thickness of 2.5 mm) | Children 0-5 Years (skin thickness of 0.56 mm) |
|---|---|---|
<160°F | 1 second | – |
<149°F | 2 seconds | 0.5 seconds |
<140°F | 5 seconds | 1 second |
135°F | 10 seconds | 4 seconds |
133°F | 16 seconds | – |
<130°F | 35 seconds | 10 seconds |
127°F | 1 minute | – |
125°F | 2 minutes | – |
124°F | 3 minutes | – |
120°F | 10 minutes | – |
Last edited:
- Joined
- Apr 17, 2014
- Messages
- 63
I control the reactions with dipstick thermometer. I keep it around 75c. I thought I would loose the nitric above that. Touch is the way to go too. 45c is outside temperature in summer in some counties. too low for our work in my opinion.. My original idea was, if I heat AR to its maximum limit, would it penetrate through Agcl crust on the jewelry and dissolve better or faster?
Your reactors are static and silver chloride build-up is worse in static reactions. The entire benefit of the tumbler reactors is the tumbling of the vessel breaks away the chloride crust. Not sure how it would affect the speed of the reaction (it should help) but the manufacturers of tumbler plants claim you can process alloy with higher Silver due to the mechanical advantage in breaking up those chlorides. In stone removal, where you cannot alloy lots together to get silver lower, this advantage may be substantial in minimizing undissolved chloride encrusted chunks.
Geo
Well-known member
Maybe I'm just a big softie. 
I do increase the heat as I go. That way I ensure the nitric has been depleted. If you stop heating while still getting any kind of reaction, even a very subdued reaction means there's still free nitric acid. The reason I don't start adding while hotter is the real possibility of a boil over.
I generally follow the same process I use for formic acid conversions. Gradual increases in heat the whole time.
I do increase the heat as I go. That way I ensure the nitric has been depleted. If you stop heating while still getting any kind of reaction, even a very subdued reaction means there's still free nitric acid. The reason I don't start adding while hotter is the real possibility of a boil over. I generally follow the same process I use for formic acid conversions. Gradual increases in heat the whole time.
My professor on university said good organic chemist must hold 100°C vessel in hand 
Yeah, that´s right. Any kind or reactionin AR when warm or hot means the active species are there. It is completely possible that after 1/2hour or so, there is no nitric acid in chloride solution, if it is hot and diluted enough. But dissolved chlorine, NOx-gasses and NOCl like species persist. That is why when you take the pot off the heat in the evening and let it cool overnight to room temperature, it is usually enough to add 2-3 spoonfuls of sulfamic to kill all active species, even for 1,5-2 liters of pregnant AR. AR deplete it´s oxidizing capacity at elevated temperature even if no metal is added to it. To the point of solubility of chlorine, NOx gasses and NOCl-like species at that temperature.
That is the reason why it is wise not to boil things in AR (if good reagent economy is desired), because loss of oxidizing power in Cl2 evaporation is very big.
Yeah, that´s right. Any kind or reactionin AR when warm or hot means the active species are there. It is completely possible that after 1/2hour or so, there is no nitric acid in chloride solution, if it is hot and diluted enough. But dissolved chlorine, NOx-gasses and NOCl like species persist. That is why when you take the pot off the heat in the evening and let it cool overnight to room temperature, it is usually enough to add 2-3 spoonfuls of sulfamic to kill all active species, even for 1,5-2 liters of pregnant AR. AR deplete it´s oxidizing capacity at elevated temperature even if no metal is added to it. To the point of solubility of chlorine, NOx gasses and NOCl-like species at that temperature.
That is the reason why it is wise not to boil things in AR (if good reagent economy is desired), because loss of oxidizing power in Cl2 evaporation is very big.
Gold.refinery
Well-known member
We performed a new refining operation and applied some of the recommendations mentioned in this forum.
Obviously we could not implement all the recommendations due to device problems (until they are fixed soon)
I try to explain the process step by step, with the changes applied in each step
We prepared a reliable feed with the following analysis:
Au:755.4
Ag:57.4
Cu:164
Zn:11.2
Fe:0.6
Ni:4.8
Cd:6.4
Feed weight 2859.8 gr. equivalent to 750 carat.
Melting and Flaking :
We melted all the scrap in one step (for the homogeneity of the percentage of silver in the gold beads), then we emptied the melt into water at 2 ° C and flake it well. Then we dried and weighed. 2.2 grams of gold was not obtained after smelting and flaking operations. I must say that we used boric acid before melting to cover the graphite crusible.
Scrap digestion , Gas purification with scrubbers:
According to the calculations, we should add 2.4 lit nitric acid and 9.8 lit hydrochloric acid.
First we added 8 liters of HCl and 1 liter of HNo3. After about an hour, we added another 1 liter of nitric. After one hour, we added 0.5 liters of HNo3 and 1.8 liters of HCl.
We did not use hydrogen peroxide in this reaction, but in the past we added it using a dosing pump.
After an hour of reaction, we turned on the Tumbler rotating.
The reaction subsided after about 4 hours and 30 minutes. The amount of undissolved seeds was 23.15 gr.
Scrubbers:
Fortunately, we have removed ammonia from the scrubbers. It was great.
We have four scrubber towers that pump water from two tanks to purify gases. We set the pH in the first tank to 10.5 with a dosing pump and a concentrated caustic soda solution. We also manually set the second tank of scrubbers to ph 10.5.
At the beginning of the reaction we had some fume at the end of the chimney, but by adjusting the airfan flow with driver, all the fumes was removed completely. When the airfan motor frequency was between 9 and 15, we had no odor or fumes in the chimney.
When the frequency was set above 25, we had some odor and steam at the end of the chimney.
When the frequency was less than 8, gas was released into the room.
We set the frequency at 18 at the peak of the reaction and 10 when the reaction calmed down.
Diluting, Cooling and Filtering AgCl:
After the reaction subsided, we added 3 liters of ice and allowed the solution to cool. We did not add extra water at this stage. (In the past we used to add water).
Unfortunately we could not modify the previous filters here. But we used a fabric filter and two paper filters to cover the gaps.
(In the case of gold filtration, we did better, we filtered with the italimpianti filter, I will explain later)
We filtered the solution and the silver chloride remained. We washed the Tumbler and valves completely.
Rinse the silver chloride with cold water until the soluble stannous chloride test is negative.
We have not removed the cylindrical filter yet, the stannous test was negative, the color of the filter was not gray, ie the silver chloride was filtered.
We also collected unresolved gold beads.
23,15 gr.
We wash the silver chloride filter and store it in gallons of chloride with some water.
DeNOxx:
(we could not heat the solution, so we could not use sulfamic acid)
After the clear golden green solution was transferred to the first tank, we added a little water. Then we slowly added the urea and stir it.
We added about 600 gr.of urea.
At first it was completely reactive and solvable. Finally, the urea seeds reached the bottom of the tank, so that it felt unresolved in the bottom of the tank. (But we still have doubts about the amount of urea seen by the reaction)
Precipitation:
We know the best precipitator is so2 gas, and the next option is smb. We prepared ourselves to use SMB, unfortunately we did not use it in this process for two reasons, first, our tanks do not have a chimney for 24-hour ventilation, our first tank does not have a mixer with shaft and propeller. After making the changes, we will definitely use smb.
Unfortunately we had to use hydrazine. But this time we used much less. We dissolved one liter of hydrazine in one liter of distilled water.
First, we added 500 cc to the solution and mixed it well. After a few minutes, we added another 500 cc and mixed well. We took a sample from the solution and performed a stannous test. The sample became completely dark. Then we added another 400 cc and mixed, even moving the settled gold particles. After 15 minutes and the particles settle. We tested the solution again, this time it darkened less but it was positive again. The color of the solution had changed from golden green to blue-green, but we were not sure. We added some precipitate again, it did not react when added, but reacted by stirring a small amount. We did the stannous test again, it was as dark as the previous test but like a semi-dark cloud, not black and purple.
(We have doubts about the diagnosis of Stanus chloride test. Because the sample is blue and green. When we add a drop of Stanus chloride, a semi-dark cloud forms after a few seconds, but the blue color is still visible)
Add a little flocculant, mix well and leave for 30 minutes.
Gold Filtering :
At this point we did not use incomplete filters. We applied the italimpianti filter according to the photo. We used two paper filters and closed the filter with a rubber band around it.
First we drained the salt solution of the base metals. When the gold sponge appeared, we started rinsing with hot pressed water in the tank. The gold sponge was drained slowly and hot water was passed over the gold several times. We kept the diaphragm pump on for about 30 minutes. The gold sponge dried a bit. (According to photo)
XRF Pure Gold:
We took some samples of gold sponge to test with Xrf .
The result of the analysis:
Au:998.8
Pt:1.1
Ag:0.0
Cu:0.0
(We did not have platinum in the initial feed analysis.)
Drying and melting:
This time we dried the gold sponge with tow induction furnace. Fill the tow graphite crucible , after drying, turn off the furnace, add again, turn on the furnace, dry , and so on.
We did not add any flux.
One of the crucible was new. Another used.
We performed several XRF tests after melting.
1) XRF sample of new crucible:
Au:999.4
Ag:0.6
2)XRF sample of used crucible:
Au:999.3
Ag:0.4
Cu:0.1
Fe:0.23
3) When I focused with XRF on the surface contaminated areas of the molten sample:
Au:968
Fe:29.6 !!!!!!
Cr: 1.9
W:4.9
Ru: 0.3
Zn:0.4
The discovery of iron surprised me. We did not dry the sponge with a steel container. Pure gold contact before melting was only with a steel spoon to fill the crucible.
Is iron used in the production of graphite crucible?
We melted the gold again with some silver, to the 999 carat. A very small amount of impurity was observed on the molten surface during melting. It was shiny but not perfect. When draining the melt, add as much pinch of potassium nitrate and borax.
Accounting:
Total gold obtained without gold trapped in silver chloride and soluble cement:
Finally we have 2122.65 gr (999)
We have undissolved 23.15 gr (755)
Gold recovered : 2850.67 gr (750)
Starting weight : 2859.8 gr (750)
Gold Not obtained: 9.13
Yield without AgCl and cementation:99.68
Cementation:
We are preparing the second and third tanks for solution cementing with copper plates and air flow (according to 4metal recommendations). After finishing a few refining operations, we want to use them.
Before the hydroxide tank, we had two cylindrical tanks and two waste water tanks. We want to make cement with copper plates in the cylindrical tanks.
Also cement with atomize copper powder in tanks with mixers and air pipes.
We are equipping them.
Obviously, the cement report is completed after a few operations to increase the solution.
Our solution has not yet reached the filter press stage, but we have adjusted the pressure of the hydraulic filter jack until no liquid escapes.
We connected the filter press to the air for drying the hydroxides.
The waste water report will be completed.
Obviously we could not implement all the recommendations due to device problems (until they are fixed soon)
I try to explain the process step by step, with the changes applied in each step
We prepared a reliable feed with the following analysis:
Au:755.4
Ag:57.4
Cu:164
Zn:11.2
Fe:0.6
Ni:4.8
Cd:6.4
Feed weight 2859.8 gr. equivalent to 750 carat.
Melting and Flaking :
We melted all the scrap in one step (for the homogeneity of the percentage of silver in the gold beads), then we emptied the melt into water at 2 ° C and flake it well. Then we dried and weighed. 2.2 grams of gold was not obtained after smelting and flaking operations. I must say that we used boric acid before melting to cover the graphite crusible.
Scrap digestion , Gas purification with scrubbers:
According to the calculations, we should add 2.4 lit nitric acid and 9.8 lit hydrochloric acid.
First we added 8 liters of HCl and 1 liter of HNo3. After about an hour, we added another 1 liter of nitric. After one hour, we added 0.5 liters of HNo3 and 1.8 liters of HCl.
We did not use hydrogen peroxide in this reaction, but in the past we added it using a dosing pump.
After an hour of reaction, we turned on the Tumbler rotating.
The reaction subsided after about 4 hours and 30 minutes. The amount of undissolved seeds was 23.15 gr.
Scrubbers:
Fortunately, we have removed ammonia from the scrubbers. It was great.
We have four scrubber towers that pump water from two tanks to purify gases. We set the pH in the first tank to 10.5 with a dosing pump and a concentrated caustic soda solution. We also manually set the second tank of scrubbers to ph 10.5.
At the beginning of the reaction we had some fume at the end of the chimney, but by adjusting the airfan flow with driver, all the fumes was removed completely. When the airfan motor frequency was between 9 and 15, we had no odor or fumes in the chimney.
When the frequency was set above 25, we had some odor and steam at the end of the chimney.
When the frequency was less than 8, gas was released into the room.
We set the frequency at 18 at the peak of the reaction and 10 when the reaction calmed down.
Diluting, Cooling and Filtering AgCl:
After the reaction subsided, we added 3 liters of ice and allowed the solution to cool. We did not add extra water at this stage. (In the past we used to add water).
Unfortunately we could not modify the previous filters here. But we used a fabric filter and two paper filters to cover the gaps.
(In the case of gold filtration, we did better, we filtered with the italimpianti filter, I will explain later)
We filtered the solution and the silver chloride remained. We washed the Tumbler and valves completely.
Rinse the silver chloride with cold water until the soluble stannous chloride test is negative.
We have not removed the cylindrical filter yet, the stannous test was negative, the color of the filter was not gray, ie the silver chloride was filtered.
We also collected unresolved gold beads.
23,15 gr.
We wash the silver chloride filter and store it in gallons of chloride with some water.
DeNOxx:
(we could not heat the solution, so we could not use sulfamic acid)
After the clear golden green solution was transferred to the first tank, we added a little water. Then we slowly added the urea and stir it.
We added about 600 gr.of urea.
At first it was completely reactive and solvable. Finally, the urea seeds reached the bottom of the tank, so that it felt unresolved in the bottom of the tank. (But we still have doubts about the amount of urea seen by the reaction)
Precipitation:
We know the best precipitator is so2 gas, and the next option is smb. We prepared ourselves to use SMB, unfortunately we did not use it in this process for two reasons, first, our tanks do not have a chimney for 24-hour ventilation, our first tank does not have a mixer with shaft and propeller. After making the changes, we will definitely use smb.
Unfortunately we had to use hydrazine. But this time we used much less. We dissolved one liter of hydrazine in one liter of distilled water.
First, we added 500 cc to the solution and mixed it well. After a few minutes, we added another 500 cc and mixed well. We took a sample from the solution and performed a stannous test. The sample became completely dark. Then we added another 400 cc and mixed, even moving the settled gold particles. After 15 minutes and the particles settle. We tested the solution again, this time it darkened less but it was positive again. The color of the solution had changed from golden green to blue-green, but we were not sure. We added some precipitate again, it did not react when added, but reacted by stirring a small amount. We did the stannous test again, it was as dark as the previous test but like a semi-dark cloud, not black and purple.
(We have doubts about the diagnosis of Stanus chloride test. Because the sample is blue and green. When we add a drop of Stanus chloride, a semi-dark cloud forms after a few seconds, but the blue color is still visible)
Add a little flocculant, mix well and leave for 30 minutes.
Gold Filtering :
At this point we did not use incomplete filters. We applied the italimpianti filter according to the photo. We used two paper filters and closed the filter with a rubber band around it.
First we drained the salt solution of the base metals. When the gold sponge appeared, we started rinsing with hot pressed water in the tank. The gold sponge was drained slowly and hot water was passed over the gold several times. We kept the diaphragm pump on for about 30 minutes. The gold sponge dried a bit. (According to photo)
XRF Pure Gold:
We took some samples of gold sponge to test with Xrf .
The result of the analysis:
Au:998.8
Pt:1.1
Ag:0.0
Cu:0.0
(We did not have platinum in the initial feed analysis.)
Drying and melting:
This time we dried the gold sponge with tow induction furnace. Fill the tow graphite crucible , after drying, turn off the furnace, add again, turn on the furnace, dry , and so on.
We did not add any flux.
One of the crucible was new. Another used.
We performed several XRF tests after melting.
1) XRF sample of new crucible:
Au:999.4
Ag:0.6
2)XRF sample of used crucible:
Au:999.3
Ag:0.4
Cu:0.1
Fe:0.23
3) When I focused with XRF on the surface contaminated areas of the molten sample:
Au:968
Fe:29.6 !!!!!!
Cr: 1.9
W:4.9
Ru: 0.3
Zn:0.4
The discovery of iron surprised me. We did not dry the sponge with a steel container. Pure gold contact before melting was only with a steel spoon to fill the crucible.
Is iron used in the production of graphite crucible?
We melted the gold again with some silver, to the 999 carat. A very small amount of impurity was observed on the molten surface during melting. It was shiny but not perfect. When draining the melt, add as much pinch of potassium nitrate and borax.
Accounting:
Total gold obtained without gold trapped in silver chloride and soluble cement:
Finally we have 2122.65 gr (999)
We have undissolved 23.15 gr (755)
Gold recovered : 2850.67 gr (750)
Starting weight : 2859.8 gr (750)
Gold Not obtained: 9.13
Yield without AgCl and cementation:99.68
Cementation:
We are preparing the second and third tanks for solution cementing with copper plates and air flow (according to 4metal recommendations). After finishing a few refining operations, we want to use them.
Before the hydroxide tank, we had two cylindrical tanks and two waste water tanks. We want to make cement with copper plates in the cylindrical tanks.
Also cement with atomize copper powder in tanks with mixers and air pipes.
We are equipping them.
Obviously, the cement report is completed after a few operations to increase the solution.
Our solution has not yet reached the filter press stage, but we have adjusted the pressure of the hydraulic filter jack until no liquid escapes.
We connected the filter press to the air for drying the hydroxides.
The waste water report will be completed.
Attachments
The tumbler should be running whenever the acid is reacting with metal in the tumbler. It is what provides the mechanical forces to break up encrusting silver chloride. If you ran the tumbler all through the reaction the undissolved alloy would be less.
You also will have gold in the silver chlorides in addition to the chunks that you removed. Have you reduced the silver chlorides to metal and melted them?
The XRF result on the reduced Silver Chlorides that you melt into a bar will determine the actual quantity of gold that is missing.
I am not really understanding what all of your numbers in parentheses mean
I am traveling tomorrow morning but I will make you an excel spreadsheet to figure this stuff out in a manner that makes more sense. (at least to me!)
One question I forgot, and Orvi asked this but it was not answered. You refer to this material as karat and the XRF results do approximate 18 karat gold, but I am getting the distinct impression that this is not karat gold jewelry because of the cadmium. And the silver is very low for karat as well. Is there something here you are not telling us?
You have never given us details about the flocculant you are using but you are now having iron issues in your fine gold. You do know there are iron based flocculants used for settling. After hearing some of the other advice the manufacturer of the equipment gave you, is it possible you are using an iron floc?
The iron-based flocculants include ferric chloride, ferric sulfate, ferrous sulfate, and ferric chloride sulfate.
What type of floc are you using?
A few things about XRF.
An average energy dispersive benchtop XRF cannot determine impurities and distinguish readily between 99.5 and 99.95 purity.
So counting on your XRF to tell you it's .995 or .999 so you can add silver to cut it to .995 is not exactly a number I'd take to the bank.
Is this like a thing????? Using XRF on molten metals in a crucible? How many of our members do this?
An average energy dispersive benchtop XRF cannot determine impurities and distinguish readily between 99.5 and 99.95 purity.
So counting on your XRF to tell you it's .995 or .999 so you can add silver to cut it to .995 is not exactly a number I'd take to the bank.
I have never, or don't know if anyone else has ever, used an XRF to shoot a molten pool of gold or any other molten metal for that matter. But it seems to me, that a device that measures secondary fluorescent energy that is emitted from an X-ray source just possibly gets it's numbers mixed up when it's looking at a sample that's glowing white and around 2000ºF above ambient temperature?
Is this like a thing????? Using XRF on molten metals in a crucible? How many of our members do this?
Last edited:
- Joined
- Apr 17, 2014
- Messages
- 63
Please no one should even think about doing that!!!! XRF needs to be less than 1 inch away from the (solid) object you are trying to measure.. Heat would burn the device...
I think he meant "Poured bar"
I absolutely concur with this. Damage to the unit will more than likely occur. It could even damage the shutter and lead to leakage of radiation (for XRF's containing a source).
Geo
Well-known member
Have you tried magnesium sulfate as a flocculent?
