Refining of Gold from Mines

[4metals]

For some reason, refiners have been a closed mouthed group for a very long time. The concept of refiners helping refiners that is exhibited in the spirit of this forum is rather unique.

A degree in chemistry did not teach me refining. It prepared me to learn the science of refining, and as I said it is not something often spoken about. Successful refiners don't want competition. It comes as no surprise that an experienced chemist could miss some of the points we have discussed, refining is a science combined with a hands on experience where an operator learns by seeing, and doing. Not necessarily from reading. In my mind a good experienced refiner is also an artist.

As Goran pointed out, sulfamic acid is available in India and it is worth finding a source. Sulfur dioxide gas would also be good to find. The difference in your end results will make the search and acquisition of both of these chemicals worth the effort.

I have been thinking why Ashapura would have as many 200 liter vessels for holding the solution after precipitation and I believe it is to maintain batch accountability through to waste cementation. The cost of the individual vessels and the support stands and connecting plumbing is considerable. By keeping every lot separate they have the ability to re-check the waste liquors for values. I feel that implementing the methods we already discussed along with assays of outturn, the long term holding of individual waste batches will not be necessary.

I also believe in batch accountability, but after assay of the outturn of fine gold, and the assay of the silver bar for gold, an allowable loss as high as .15% for a small refiner is acceptable and these values are usually recoverable in the residues at a later date. So I believe that after a preliminary short term holding of waste liquors, they can be combined with waste liquors of previous lots all of which have met the accountability standard. (99.85% of assay) Then the wastes can be cemented and harvested in economical sized batches.

One issue in maintaining tight accountability standards is the ability to assay to the required precision with proof corrected fire assays. This is critical, as some alloys have high proof correction factors and if they are not determined and applied properly you may be looking for gold that was never there. We can go over this if you have questions. I like to see labs that I work with participate in the ASTM round robin fire assay program to learn, and monitor, how precise your assayers are. In addition the LMB runs a proactive monitoring program (PAM) for all refiners certified to produce good gold delivery bars. Good analytics is the foundation of good accountability.

I would like to hear the waste treatment process used at Ashapura to figure out how the metals already lost to waste may be recovered. Do you have pictures of your neutralization process and filter press for retaining the solid hydroxides from waste treatment?

 

[Palladium]

Can we sticky this thread or add it to the best of threads.

 

[Lou]

I've made a simple flow chart to demonstrate the process steps. Note that it is not fully complete, but is a general conceptual example of what needs done and where material must flow. The Devil remains in the details.


Note that much of this is similar to if the Miller chlorination process were used. The silver chloride slag bail off is dumped into water, leached with HCl/NaClO3, centrifuged, and then converted and run through cells. I'm not a fan of Miller for the most part.


Lou

 

[4metals]

Excellent flow chart Lou.

One thing Ashapura should note. They are stamping their bars .995 from the photo early on in the thread. The procedures we are outlining will produce gold in excess of .999 and if he does the sulfuric pulping process he will attain .9999. (Barring the presence of Selenium or Tellurium in the feedstock) That means for every kilo bar they produce, they will be giving away 0.5% of gold.

0.005 X 32.15 = .160 ounces of gold at $1200 an ounce thats $192 extra value with every kilo out the door.

They have to be careful to either alloy down their .9999 product or get a new stamp and charge for .9999.

Considering large refineries don't even charge 0.5% for refining these days, that can be a make or break deal!

 

[ashapura refinery]

We are definitely very fortunate for having such great people out there helping us through this entire process. We sat down yesterday and took note of all the points listed out.

To begin with:
This is a completely new setup. And we had a chemical engineer who had previously worked in one of the gold refineries in the Middle East. And the practise of using Soda ash after the silver chloride removal is very common there. What they do is after the aqua regia process(ratio being kept at 3:1), they remove the silver chloride, add the soda ash to the solution and Sodium Metabisulfite to precipitate the gold. Once this is done they further add caustic soda inorder to recover any left over material. Now this is being followed in refineries who refine 200 Kgs to 800 Kgs of gold on a daily basis and our chemists who work here had been previously working in one of these refineries.
Our chemical engineer who helped do the setup was the one who had suggested that we use stannous chloride for the recovery part. Anyway now due to some reasons we no more contact him.

But now we understand the following(Hope we have understood it well enough):

1) Soda ash will only neutralize but give a problem of creating the salts as mentioned by Butcher.
2) We have to remove the excess nitric acid from the solution as the most important process before we go ahead for the precipitation.
3) Stannous chloride is not a method of recovery as it forms colloidal form of gold and this makes recovery difficult.

Keeping these points in mind we have done the following:

1) Sulfamic Acid : Finally we did track down a dealer and got it in the powder form. The place which Goran had mentioned is actually another state all together. That state is pretty famous for chemical manufacturing. And the problem with things being in other states is that you will get the goods but nothing less than 10 working days. But it is okay we did manage to find a dealer locally and have got it with us now.

2) Condenser : We have a batch of fresh stannous chloride coming in today so today we will be testing the condensers and other places for the presence of the metal as suggested in the thread.

3)The storage of Nitric Acid and Hydrochloric Acid - We generally fill the tank with Hydrochloric Acid first and transfer that to the mixing reactor and then fill it up with Nitric and transfer to the Mixing reactor. But we now understand as per your advice and the other advice by others is that Nitric acid should be added lot by lot so that we do not add excess of nirtic.

4) The Storage Tanks - We are considering getting small conical tanks now instead of using these glass cylinders. Hopefully we can use these glass cylinders for some other purpose.

Apart from this:

1) Sodium Metabisulfite - Earlier in one of our batches, we had a problem of precipitating gold using SMB, it was just not settling down and we contacted our engineer he asked us to get Food Grade SMB and use that. As per his advice we did that and are now using food grade SMB. We generally dilute the powder in water and then add it to the solution.

2) The Gold Bars - Sir (4metals) When we complete the refining and get the pure gold in hand we do an assay test. For the last batch we had got the purity of 999.22. We generally get a purity of 999 in all the batches. But the local market where we sell our bars require the product to be of purity 995.0. So for this we do the following conversion:

995.80 gms Pure Gold of purity 999.22 + 4.20 gms Silver = 1000 gms weight of Gold Kilobar of purity 995.0
In this case we sell the gold at the price of 995.

3) The filters used - We have been using 2 layers of Polypropylene filter cloths - first layer is of 1-2 microns and the layer below this is 4-5 microns. (We are not 100% sure of the microns as this is India and we learned a lot of things after starting this setup that you will get all kinds of stuff but guarantee of the quality is given by very few).

4) Filtration Plant - We transfer the solution from the cylindrical vessels to two tanks of 2000 Ltrs each, we store the solution in these tanks for few days then transfer the solution to the Filter press and from the filter press we transfer to another tank of 4000 Ltrs. Picture attached

(Lou) Sir thank you very much for the flow chart, this now gives a much more better clarity in what we should be doing (definitely better than the way i have scribbled all over my book )

We will get back with more information on our past workings which might help you all in guiding us further.

 

[Palladium]

I love air assisted double diaphragm pumps. I have one myself and I've seen two in your pictures. I use filter socks though and not a filter press.

 

[4metals]

Let me start out by saying that a process which is good for one feedstock is not necessarily good for another. When I evaluate a process, the first thing I consider is what equipment is available for the refining. This can eliminate some processes immediately. No one likes to spend more money than they have to and no refiner likes to turn away work because they are not set up to process it. For this reason I favor versatile equipment. Once a refinery starts to see a consistent quantity of a certain material then it is fine to set up to maximize efficiency and yield with dedicated setups.

I would never build a refinery without a process hood where individual lots of varied material could be processed if required.

Now the material at hand. The preliminary assays indicate 99.5% precious metals to start with. Wow, pinch me! What an ideal feedstock to start with! You have not indicated if there are any traces of Platinum Group metals in the material so I have assumed not. This becomes important because the flow chart Lou provided and his descriptions include a resin system to catch these values before waste treatment. In your original post you stated 0.04% other metals. Even if that was all platinum, in a 10 kilo lot you are talking about $4.50 worth of platinum (less recovery cost). But even at that level, your refinery is set up to process 100 Kg a day and $45 a day can add up to real money over time. Plus lots of higher PGM value can sneak in. Anyway, it is good to track its presence. A good fire assayer, and some good fresh stannous chloride can determine its presence qualitatively and allow you to separate your waste streams, one holding tank set aside for PGM trace recovery, and another for Gold and silver traces.

This is a resin system which solutions are pumped through to recover values before waste treatment.




If you have been reading about refining on this forum you have surely noticed that in the best case you want to clean up your material as much as possible before using aqua regia, the reason being you do not want to affect your final purity by dragging down other metals in solution. That is why some members suggested inquartation for this material. In reality, the quantity of base metals you will have in solution will be exceptionally low for this material almost as if you had done an inquartation with the exception being an inquartation and treatment will have removed the silver as well. In this case the cost of acids and extra handling does not justify inquartation.

So what does this mean about the refineries in the Mid East where your methods came from? We do not know what the purity of the material they are adding to the aqua regia is. If they pre treat by inquartation or other processes before aqua regia then the relatively high purity of the material going into aqua regia will allow for the chemistry you have outlined. I am not saying I like the method, I am saying it can work (obviously it does). But the reason it works is how clean the material is going into the aqua regia. The picture you posted of your digestion vessel full of beautiful red acid tells me your feedstock has very little base metals without even looking at the preliminary assay. (That could be a centerfold in a refining magazine!)

Lets for a moment say you were refining karat gold in your system. No pretreatment, no inquartation, just straight melted and granulated karat. In the industry, karat gold can be refined on a large scale by melting all of the metal for a particular lot together and making sure the cumulative silver content is under 9%. With that silver content or less, a straight aqua regia dissolve will give you a good dissolve and leave you with silver chloride as an insoluble. The acid however will be a green color, not even close to the pretty red acid you have shown us. If you were to proceed to neutralize to pH 7 the excess nitric with soda ash as you described, you would be bringing down metals as hydroxides which would mix with your gold dropped with metabisulfite. In short you would have a mess.

The soda ash can have an unintended consequence adversely effecting your gold purity. Remember your primary goal is to eliminate excess nitric acid from the dissolve process. You can add soda ash and have problems with some of the other metals dissolved in the acid, or you can add sulfamic acid and not cause any unwanted precipitation save the dropping of the lead as lead sulfate. In the karat gold refining scenario, the sulfamic choice will allow you to drop only the gold leaving the base metals in solution. Note you add the sulfamic before filtering out the silver chlorides.

Continuing on in your process following the Mid East method. After filtering out your precipitated gold, you continue to add Caustic soda to recover any other materials. Here we call that waste treatment. The problem being you have not attempted to cement out remaining values, you have dropped all of the metals as hydroxides and co-mingled your values with any base metals. Not to mention the fact that any Platinum will remain in solution and go out with your waste water.

On to waste treatment. Classic waste treatment is done after you have recovered all of the values by cementation or resin column. The remaining metals will be base metals and we raise the pH to drop them as metal hydroxides. The base metal hydroxides are collected in a filter press and the water from the filter press is neutralized and checked for metals to meet discharge standards and if it passes, discharged to sewer. The photo you attached shows the filter press feeding off the tank where normally the pH is raised with good agitation so the hydroxides remain suspended in solution and can be pumped through the press. I do not see any mixing motors on the tank. Is this where you raise the pH or does the liquid come into these tanks alkaline as per the Mid East recipe?

This is what a typical mixing tank feeding a filter press would look like.


And this is the hydroxides that come out. These particular hydroxides are very high in copper due to the cementation process used. They are dried and shipped to a copper smelter who pays on copper and any precious metals contained.


I notice your filter press, and the one in the photo I attached are both European design. The difference is American made presses pass the liquid through the plates and discharge out the end of the press into a pipe all using the power of the double diaphragm pump to move the liquid. European models allow each plate to drain into a trough which has to be collected and transferred to the next process. I have used both and both work well.

3) The filters used - We have been using 2 layers of Polypropylene filter cloths - first layer is of 1-2 microns and the layer below this is 4-5 microns. (We are not 100% sure of the microns as this is India and we learned a lot of things after starting this setup that you will get all kinds of stuff but guarantee of the quality is given by very few).

The paper of the most coarse porosity, in this case the 4-5 micron, should be the first layer to contact the liquid being filtered, followed by the least coarse 1-2 micron. Your goal is to filter out the larger particles first and allow the smaller particles to pass and be collected by the finer paper. It will not clog as fast if you do coarser first.

2) The Gold Bars - Sir (4metals) When we complete the refining and get the pure gold in hand we do an assay test. For the last batch we had got the purity of 999.22. We generally get a purity of 999 in all the batches. But the local market where we sell our bars require the product to be of purity 995.0. So for this we do the following conversion:

995.80 gms Pure Gold of purity 999.22 + 4.20 gms Silver = 1000 gms weight of Gold Kilobar of purity 995.0
In this case we sell the gold at the price of 995.

Excellent, no sense giving gold away!

 

[4metals]

I am still a little confused as to where you recovered some gold after the losses. If you can detail the steps you followed and what and where in the process you found the gold that was missing along with how much remains missing. Maybe we can piece together the whereabouts of the missing gold.

 

[ashapura refinery]

@4metals:

We do not have any platinum group metals present in the material we receive. The main contents in the material is gold, silver, copper, iron.

As for the recovery we have done so far:
1) We had used cpvc pipes earlier and we found fine gold particles in the pipes, we then removed these pipes and cleaned them out.

2) We had added caustic soda and stannous chloride to the left over solutions and we recovered a small amount of gold from the material that settled. It was not a 100% recovery but a small part.

3) Apart from this the used gloves, filter cloths, etc was washed out carefully as well.

 

[ashapura refinery]

Hello. We noted down the points mentioned in the previous posts. And started a new fresh batch of refining.

The weight of the Gold Dore Bar - 7.8 Kgs (Actual weight - 7812.20 Gms)

We did the assay of the Dore Bar to know the exact contents.
The content of the dore bar are:
1) Gold - 92.96%
2) Silver - 06.55
3) Copper - 00.20
4) Iron - 00.24

We started off with the aqua regia process:

1) We added HCL - 24 Ltrs.
2) Then started adding nitric acid - total added - 7.5 Ltrs
3) This time very few grains remained undissolved.
4) Then we added about 1.8 Kgs of Sulfamic Acid powder in the solution ( lot by lot).
5) Then we filtered the aqua regia solution and removed the silver chloride.
6) After the filtration of the silver chloride, we were left with total solution of 45Ltrs including the water used for the washing.
7) We then took the first batch of 20 Ltrs of the solution into the other mixing reactor and added about 3 Kgs of Sodium Metabisulfite diluted in water for the precipitation. The reaction happened immediately. The gold started settling but after a few minutes the gold got dissolved in the solution again i. e it turned back to the aqua regia solution.

The Color of the solution is exactly same as the Aqua Regia process. I have attached two pictures of this solution.

can someone please explain where have we gone wrong this time?

 

[Barren Realms 007]

You said you added 1.8 Kgs of sulfamic acid.

Was this as a crystal dry form of dissolved in water?

When using sulfamic acid I have had best results with dissolving it in hot water and then adding it to the solution. You need to be able to stir your solution when adding the sulfamic acid to get to mix well with your solution. when the violent reaction stops from adding the sulfamic acid I usually add 15%-20% more of the solution to make sure I have removed all of the nitric.

 

[ashapura refinery]

We have sulfamic acid in the powder form. We mixed this in water (at normal temperature) and added it to the solution of aqua regia before the filtration of the silver chloride. While adding the sulfamic acid liquid the stirring was also going on.

We checked for the PH level and it is 3 right now.

You said you added 1.8 Kgs of sulfamic acid.

Was this as a crystal dry form of dissolved in water?

When using sulfamic acid I have had best results with dissolving it in hot water and then adding it to the solution. You need to be able to stir your solution when adding the sulfamic acid to get to mix well with your solution. when the violent reaction stops from adding the sulfamic acid I usually add 15%-20% more of the solution to make sure I have removed all of the nitric.

Does it mean that we still have some nitric acid left over in the solution? Is this why the entire solution is reacting back and forming aqua regia? I thought about this as a possibility but I am not very sure if this is the reason. Also what is the best way to know if Nitric acid is still present in the solution or not? Is there any test for this?

 

[Lou]

Barren is right!

If you want the sulfamic to work correctly, here's the gig:

Take boiling hot water and dissolve the sulfamic. The dissolution is endothermic and the solution will get cold. Hence the use of boiling water. Do not boil it on a hot plate--the heat from the hot plate on the crystals on the bottom will decompose the sulfamic acid in ammonium cation amongst other things (more important if you deNOX PGM solutions).

Add the hot sulfamic solution to the HOT gold solution, and stir. Add until the production of nitrous oxide gas stops. Add slowly to avoid foam over.

If you use an ORP meter (get a titanium probe!) you can see the ORP fall from typical >950 mV+ to around 650-700 mV+. <--this tells you the speciation of the gold and how much nitrate is available. Every ORP meter is different--make up an aqua regia solution and check that ORP. Then dissolve gold, and check again. This is calibration. You will soon be able to tell. ORP varies with temperature too, so make sure to keep meticulous notes. You are almost there.

Regarding incomplete precipitation--obviously all the gold was not out of the solution. Keep treating with sulfite and you'll eventually get it all. If you don't want to deal with that and want it all out pretty much instantly, use hydrazine hydrochloride salt solution. Produces nitrogen gas as the by product.

Do NOT overuse the sulfamic or you will get plating upon the exposed surfaces in the reduction kettle.

@ 45 L total gold solution with 7.8 kg (say 12 g grains remained) X 0.9296 = 161.13 g/L, over 5 oz/L. That's a good concentration.
24L HCl means total chloride content (you should have rinsed chloride with 1M HCl! telling you it helps to get the gold off the AgCl and prevents peptization of the AgCl), assuming 12 M conc HCl is 6.4 M in [Cl-]. This is a good strength to precip. Maybe a bit high on the chloride content.

It's important to get the chloride out of the gold sponge with ammonia rinse b/c if you don't, and you melt the gold sponge with AgCl, some of the AgCl decomposes to chlorine and the gold volatilizes.

just saw your post: 3 kg of metabisulfite should NOT bring the pH up high enough to make it pH 3. Adjust the pH down to 0 with fresh HCl. If the pH isn't low enough, the metabisulfite will not form the SO2 in situ which is what's doing the reduction. Add 10 L concentrated HCl and try again

Lou

 

[nickvc]

I'd say you still have nitric left to redissolve your gold.
The one good use of urea prills is for testing whether you have nitric left in your solutions, simply add a couple of pieces and see the reaction, if it fizzes and dissolves you have nitric left.

 

[ashapura refinery]

Okay understood. So our first step should be to remove the excess nitric acid. Now in this batch of refining we have used 7.5 Ltrs of Nitric Acid and have added 1.8 Kg Sulfamic Acid.

Can anyone suggest how much more should we add - like 3 Kgs more or 5 Kgs more?

 

[Lou]

Don't add anymore sulfamic. If anything heat the solution and it'll get it.

Failing that, add more muriatic acid and try the sulfite again. I think that'll get it.

EDIT:

You all need to start doing your maths. The math is less critical on the digestion side because aqua regia decomposes in use so the stoichiometry is never right.

If your assay is
1) Gold - 92.96%
2) Silver - 06.55
3) Copper - 00.20
4) Iron - 00.24


Now, you should also do your math on the metabisulfite. I'm not going to run through the stoichiometry because this is something that your chemist, if he's worth anything, should be able to do.

Take the weight of the charge in grams X purity of gold and divide by molar mass of gold 196.97 g/mol, then use the stoichiometric coefficients to get the minimum amount of acids to do the job. In practice, you'll probably end up using a 25-30% excess of HCl, and a 5-10% excess of nitric acid because of side reactions of decomposing nitroysl chloride and the small amounts of copper.

Given the grade of your feed, this is an ideal candidate for Miller chlorination--you should hit 995 no problem, and that's what would be made into anodes to go into the Wohwill cell, or else put into aqua regia. Cell is the cheapest but requires extensive gold inventory and accounting, something which your refinery might struggle with at this point.

Lou

 

[ashapura refinery]

@ Lou:

We started this process today morning. We did the assay, the granulating and then went on with the chemical process. This time we told our chemist that we are not going to use the Soda Ash procedure and will be using Sulfamic Acid and then precipitate by using only SMB. Apparently he was not very happy with this idea. Later in the evening he gave up saying I cannot do this process anymore and left. (This is India employees leave whenever they want :roll: ) So right now we are on our own and working with the help of our refinery department helpers who generally help around. So calculating the stoichiometry part would be difficult for now.

So as per your suggestion we went ahead with the Sulfamic Acid addition and also added some HCL and the PH level right now is near 1. We have now added SMB and are checking for the precipitation reaction.

A question on a general basis, is there any method/test on testing the presence of Nitric Acid in any solution?

 

[4metals]

Part of the challenge in refining is the chemical equations for all of the reactions can be done on paper but then they have to be applied to the real world.

For example sulfamic acids reaction with Nitric acid is pretty straightforward. For every 3 1/3 pounds (1.5 Kg) of sulfamic acid you add, you will convert one liter of nitric acid. That's the easy part.

You said you ran the reaction by adding a total of 7.5 liters of nitric acid. (With only 24 liters of Hydrochloric it is only a 3.2:1 ratio, still too much nitric.) The variable is how much of the nitric acid remained after the reaction. Every system is different, heat, no heat, agitation, pressure, condensate return, and a few other factors all play in determining this. And the type of material is critical, your nitric consumption for a high gold alloy like you are running will be different with karat gold for example.

Lou said,

so make sure to keep meticulous notes. You are almost there.

Once you determine the proper quantities of reagent for each scrap type write it down and keep copies of the notebook. (You never know when an employee decides to open up his own refinery down the street and takes all of his knowledge, both written and learned, with him. Don't ask me how I know this!)

So from what you said you can see you did not add enough sulfamic to kill the excess nitric, and the sulfamic that you did add, 1.8 Kg, is enough to kill 1.2 liters. The presence of free nitric acid in your solution after you dropped the gold is what caused the gold to re-dissolve. The thing to take from this is LESS NITRIC.

Doing the math thus far 7.5 liters to start, less 1.2 liters neutralized is 6.3 liters. Had you started with 6.3 liters of nitric and used the same 24 liters of Hydrochloric you now are at a 3.8:1 ratio. Since you had enough nitric to re-dissolve the gold there was obviously more in there. I would suggest at a minimum (a starting point which can likely be decreased further) start with 24 liters HCl and 6 liters of Nitric. DO NOT add all of the nitric at once, be patient and you will learn to observe when the reaction is complete and stop adding nitric.

Lou's suggestion to use ORP to determine when the nitric has been converted is sound advice. However it has been my experience that the analytical capabilities of intermediate sized refiners would never live up to Lou's expectations. Another trick an analytically poor refiner can use use was mentioned by NickVC. A few, one or two prills, not a handfull, dropped into the solution after you have added your first dose of nitric will fizz and dissolve quickly if there is free nitric in solution and float up and barely react when you are close.

From here on out, start out with a 4:1 nitric ratio but only add as much nitric as needed. I am thinking you will find a ratio of 1:5 will work. Keep notes!
Test for free nitric with urea prills
Add 1.5 Kg of sulfamic acid dissolved in HOT water. An excess of sulfamic acid will not hurt but do not get carried away.
RE test with urea prills
Add ever decreasing quantities of sulfamic acid.
Never add more than 4.5Kg for a 6 liter nitric addition, this quantity allows for 50% of the nitric to be un-consumed and you can do better than that by adding the nitric conservatively.

If I were running this process, I would add ice to the reactor after the sulfamic additions, the ice lowers the temperature of the solution as well as the solubility of silver chloride so it makes for less silver carry over into the next step.

If you get the technique down of not using too much nitric it is common to use SO2 gas (I didn't give up on you finding that yet) which will drop the gold only to have the gold re-dissolve and emit a red fume until all of the free nitric is consumed. If you can have a minimal quantity of free nitric you can skip sulfamic acid altogether and use SO2 to burn off the excess free nitric acid and then the SO2 will continue to drop the gold until the solution is barren of values. Sodium metabisulfite will also drop the gold and consume the free nitric just like SO2 but I have always preferred the gas for this.

 

[Lou]

Given that sulfamic is about $2/kg in bulk (at least for me it is), they need to be cognizant about chemical costs.

Take home message is PATIENCE and LESS NITRIC ACID. Either make smaller grain so it dissolves faster or else wait longer on the material to get full digestion of the gold. Get a flashlight to look into the reactor.

Another thing, if the digestion starts to stall, you might consider adding some more muriatic acid FIRST then small additions of nitric acid.

You're lucky to have such a great material to refine.


As for your chemist, if he is unhappy about expert advice and is giving up despite having worked at a refinery in the Middle East (Kuloti?), find a new chemist. This process isn't hard. Gold refining should be one of the simple pleasures of life, and if your process/equipment is right, your feed material consistent, it should run the same every time all the time and print you money so long as you can feed the machine. One thing I did not see, is how you are heating your digestion reactor?

I'd tell him to get on here and come talk to us, he'll either start making you money, or start looking for a job. Did you show him the flow chart I made?


To recap, step by step for this chemist of yours:

1. Weigh incoming dore bars after visual inspection.
2. Melt the gold at 1115C, stir it well and take a vacuum pin tube sample. Have your melt operator clip off 2 g or so for fire assay (centimeter length is plenty!); return the rest of the sample to the molten gold. De slag the gold with a graphite rod. If your slag is too thin to stick to the rod, add some sand or broken glass so it doesn't report in the granules. Alternatively, cast a bar at this stage, peen off the slag, and melt the rest of the inbound refining lots together after sampling individual heats.
3. Granulate the gold as small as possible into cold, circulating water.
4. Pump off the water from the granulation tank and put the granules into a flat stainless steel pan that is tared and dry them until no more steam comes off and no water drips off of a cold piece of glass held over it. Or until they weigh the same (constant mass).
5. Mass up the charge, doing the math based off of preliminary X-ray, add in all of the 12M muriatic acid, then 90% of theoretical 16M nitric acid to get the reaction going. Stir very well. Bring to 80C. Add small 100 mL amounts of nitric acid followed by 400 mL amounts of muriatic until all dissolves and no more red NOx gas is seen. Use flashlight to check the chlorides for granules and fizzing.
6. When all the gold is dissolved, add in hot saturated sulfamic acid to destroy nitric acid. When fizzing with sulfamic stops, test with a few prills of urea to be sure it is all gone; if not, urea will leave trail of bubbles. Check the ORP with ORP probe before and after to see the decrease in potential indicating nitric acid is gone.
7. Add clean ice to reactor to dilute the acids down to 4-5 M concentration and help precipitate the silver chloride by lowering its solubility. Sulfamic makes sulfuric when reacting with nitric acid, so any lead is going to precipitate as its sulfate.
8. Discharge reactor carefully, wash the silver chloride in a dedicated Buchner funnel with 1 M HCl until the washings are colorless. Go convert the AgCl into metal in a large plastic concrete mixer with clean scrap iron and 10% sulfuric, pull test samples and take to lab, rinse well with water til not acidic and melt up in cupel with small amount of borax and carbonate. If good, melt all of it and make anodes. Take sample for parting to determine gold content and find the silver content via titration. Silver should be 995 or so. Run the anodes through the silver cell to make crystal silver that is 9995 or better. Black residue is gold mixed with lead and other insolubles.
9. Return to your gold solution that should be cool due to the ice. If it was filtered right, should be a clear red solution no haze. If there is haze, no big deal, it's just silver chloride that you can remove later. ORP should be as mentioned above. This is then treated with either sulfur dioxide gas, or metabisulfite (do the math, you'll need probably 5-10% excess). Again, hydrazine hydrochloride is terrifically good at this operation and gives much nicer precipitate.
10. Stir the gold solution well as it is reducing, and when done, test with stannous chloride on Q-tip to be sure no gold remains.
11. Allow to settle, siphon off the barren gold solution, discharge reactor into an 18" buchner funnel with a 1 um polypropylene felt filter. Wash the gold with 6M acid that you can re-use until no color leaves, then with cold water, then ammonia to remove any silver chloride (keep this ammonia separate from other waste liquids). When rinsing with ammonia, rinse until no blue color is seen (tetrammine copper complex) and the silver chloride is surely removed. The final rinse is with hot DI water.
12. Dry and melt the gold sponge, weigh it, sample, and it should assay at 9999 given your feed. Alloy down to 99.5 with copper (cheaper than silver).








Some points to be made:
1. test your acids for their strength via titration and have the manufacturer/supplier send you the analysis of the batch
2. If you don't yield the right gold following our explicit instructions, then the problem is your assayer.

 

[4metals]

Lou,

What I see more often than not is refiners wanting to kick off a reaction and let it work over night. These guys dose up the nitric acid based on their best guess calculations and deal with any excess nitric in the morning. For them sulfamic is a necessary evil and they find a 20% excess does no harm.

As far as heat, I believe they do not heat the reaction. I don't see a mantle on the vessel so unless they have an internal heat probe it is running on whatever heat the reaction provides.