Further things which may be of interest to members

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Judging by the queries from forum members there is a need for basic gold analytical information regarding ores.

It should be noted that there are always specialised exceptions to the general rules but these exceptions are unlikely to be encountered by the average prospector.

Most queries start along the lines of how much gold is in the ore sample.

An experienced person can estimate the grade of visible free gold in an ore sample. They know what gold in ore looks like and from having previous samples assayed they can estimate the grade.

If gold is not obviously present as coarse free particles then it is necessary to call in a professional assayer.

There are three types of assay for gold which are applied according to what the prospector wants to know.

Keep in mind that most labs charge a fairly hefty batch fee to cover the cost of doing paperwork so it pays to look at pricing to see if it is near the same cost overall to do more than one assay type on an ore sample.

The first assay type is the standard fire assay. This will tell you how much gold is present in your ore and is the basic assay format. Unfortunately it does not tell you anything more than the gold grade. An experienced assayer should be able to look at an ore sample and identify the presence of sulfides and anything else obvious which may affect the fire assay and use the appropriate assay flux accordingly.

The modern trend is to use only two flux types covering ores with and without sulfides present, this has been brought about by the shortage of assayers able to formulate fluxes and the high cost of doing these formulations compared with just taking a scoop of premixed flux from a commercially purchased container of such.

That said, the standard flux formulations will give you a go/no go result for possible further work on a sample.

Always keep in mind that human nature is to take the best looking samples and have them assayed so your first samples assayed should be treated as indicative only, the results are not necessarily applicable to the whole deposit.

So now we have done the basic fire assay and registered gold in viable quantities. The next thing we want to know is where in the ore sample is the gold; is it in the quartz matrix for example or is it associated with sulfides or is it all as free gold.

Most labs will offer you a range of acid type assays along with reasons for having them done on your samples.

For the average prospector just wanting some basic information these assays are bewildering and expensive.

What is required is a basic aqua regia digest assay. This assay type will show gold which is either available as free or exposed gold particles or is in a matrix such as sulfides which can be dissolved by the aqua regia.

The aqua regia values are usually slightly less than fire assay values due to fire assay being able to access the gold locked in ore particles which the aqua regia cannot access.

If we have a large difference in values between fire assay and aqua regia digest then the usual reason is that the bulk of the gold is encapsulated in a matrix which aqua regia cannot dissolve such as quartz.

In the above case we now have the information that there is a strong possibility that fine milling of the ore may expose enough gold values to allow viable treatment of the ore.

What the above tests do not tell you is if the gold is free or is in a matrix which aqua regia can dissolve, both the fire assay and the aqua regia assay will show you the above gold values.

Often gold in sulfides is in the form where it is as small particles down to atoms locked inside the sulphide crystal matrix. You are not going to recover much of this gold without either smelting the sulfide in the presence of a collector metal or by chemically destroying the sulfide matrix.

The third form of assay has two uses; it tells you what gold recovery can be expected with cyanide and it tells you what level of the gold is locked in sulfides or similar by comparing the results with those of an aqua regia assay.

This assay form is a cyanidation assay, it is usually offered in scale from a few grams to several kilograms.

The smallest scale is often a few grams of ore stirred in a beaker but it is usually run as a bottle roll of 50 grams plus of ore run for 24 hours.

Because the bottle roll is an alkaline leach it has little dissolution effect on most but not all sulfides.

If you have highish gold values from fire and aqua regia assays but low gold values from cyanide then it is a fair assumption that the gold is locked in sulfides and that recovering the sulfides alone will recover most of the gold. This will give you a concentrate which can be sold or treated in a separate stage.

None of the above is totally definitive but is the basis for test work to set out an operating plan.

Deano

It will at least let you know if doing further work on a deposit is warranted and will do it at the lowest cost.
 
Thanks for the run-through of ore assay.

I wonder where in this scheme does tellurides and other sulfosalts of gold end up? Is the aqua regia solution strong enough to break down tellurides?

We have an interesting mine close by (about 150 km, suitable for a day trip).
- Kankbergsgruvan (Kankberg Mine) where most of the gold is locked up in 10-15 um large grains of sylvanite and calaverite in a matrix of mostly topaz and quartz, the ore is grayish white to pale blue from topaz. Visible gold is really rare. Boliden have a secret process to process this ore that also recovers the tellurium. It was a great thing a few years back and they increased the world tellurium production with 10% but since then the tellurium price have crashed.

Göran
 
Somehow when I wrote about recovering gold from slag I left out the treatment of the table tails.

By jaw crushing and tabling the minus 1mm jaw product you will recover all of the coarse free gold prills.

There will still be substantial quantities of both fine gold prills and superfine gold prills in the table tailings.

The fine gold prills can be recovered by ball milling the table tails to 100% minus 100 microns and then running this product slowly over the table with minimal wash water.

Depending on the slag used, the recovery of fine prills usually runs between 300g to 1 kg of prills per ton of slag tabled.

The tailings from the second tabling are really only worth leaching as a gold recovery method, any further milling will require running the table as a slimes table with the very slow feed rates that come with that format.

When looking at what leach type you are going to use on the final table tails you must always keep in mind that the reason the prills are present in the slag in the first place is that the gold was dirty when smelted.

Certainly the slag will remove substantial quantities of metallic impurities from the gold during the smelt and thus allow the cleaned gold particles to coalesce into a pool.

There will, however, be a substantial quantity of gold which becomes coated with the metallic impurities and cannot coalesce into the gold pool.

Acting under a range of physical restraints, this coated gold forms prills which are distributed throughout the slag.

The size of these prills ranges from several millimetres to well under 1 micron.

This means that in order to recover all of the gold present in the slag you are looking to mill the slag down to single micron size.

This will require the use of either a stirred mill or an air mill or similar.

This type of equipment is both expensive to buy and use so you are looking at how you can get the best recovery for the cheapest cost.

Jaw crushing is the cheapest method of size reduction with minimal amounts of shape deformation, thus allowing fast tabling with good recoveries of any coarse prills.

Ball milling allows you to reduce the slag size to the point where you can still table liberated fine sized prills because the deformation of these fine sized prills is minimal.

The prills present in the slag after ball milling usually run a gold grade of 200 to 500 grams per ton of slag.

You can carry out milling and leaching tests to see if it is worth carrying out the more exotic milling methods before leaching.

Unless you have access to large quantities of slag, the further milling is usually not worthwhile.

It must be kept in mind that the prills are only present because they are coated with base metals, most of these base metal coatings are resistant to cyanide leaching.

Leaching is usually only carried out on the slag after all gravity recoverable gold has been recovered.

As a general rule the coatings can be dissolved in chloride liquors.

This allows for the dissolution of the coatings in HCl, a filtration step with a water wash, pulping the residue with lime or caustic and then carrying out a cyanide leach.

This sequence is going to be cheaper than carrying out an aqua regia or chlorine leach and safer than both aqua regia or chlorine leaches.

Deano
 
Another option could be smelting the slags with copper in a rotary furnace and producing copper anodes for electrolytic recovery. This eliminates milling but often employs equipment not available on site. (Electrolytic copper cells)

With the quantities of gold you listed, 300g to 1 kg per ton, it may be easier to install a rotary furnace and produce assayable copper bars to be refined elsewhere. As with all gold containing feeds, the ability to quantify the gold content on site by producing the copper based bullion will yield the maximum payout.
 
A lot of fairly smart metallurgists have attempted gold recovery from slag using molten copper in various configurations.

Basically they all failed, gold recovery was minimal.

It appeared that the coatings on the gold which prevented the gold from coalescing into the original gold pool also prevented the prills from being adsorbed into molten copper.

Thus the use of physical methods to get a concentrate which can be attacked by a chloride matrix in a cleaning step before further smelting, the residue from fine milling and HCl attack and then leaching still contains substantial gold values.

Deano
 
027.JPGFurther to the treatment of slags I was recently involved in the setting up of a small vat leach system for treating the slag after the ball milling step mentioned in a previous post.

We could not get enough further values by air milling or stirred milling down to single micron size to justify anything bar a ball milling to 100% minus 35 microns.

The milled product was leached in a small vat 1.5 metres diameter and 0.5 metres high. This vat size took a charge of 500 kg milled slag.

A 2.5 metre length of 19mm grey water dispersion hose was used as the collection hosing, this was covered with polypropylene filter fabric which was retained with cable ties.

After tying off, the ties were cut neatly so that the hose could lie flat on the base of the vat.

The blind end of the hose had a poly plug siliconed in and a 90 degree 13mm poly bend was siliconed in at the other suction end.

The hose was positioned on the floor of the vat and a length of 13mm poly hose was attached to the poly bend, no clamps were used as the pressure on the join is negative.

Note that no metallic components were installed in the wetted area of the vat, this allowed the use of a hypochlorite leach if necessary.

The hosing on the floor of the vat was then covered with 150mm of concrete sand and a coarse filter fabric was laid on top of this sand before the milled slag was placed in the vat.

The function of the filter fabric on top of the sand is to act as a demarcation lining so that when the slag was being emptied from the vat after leaching it was easy to stop shovelling when the fabric was reached and no disturbance of the sand and collection hose occurred.

The 13mm poly hose suction line was fed to a speed controlled peristaltic pump which then fed the liquor into a vertical carbon column.

A tapped off take point was installed in the line after the pump and immediately before the carbon column to allow for sampling to monitor the gold tenor of the liquor before carbon adsorption.

By having a loose coil of the poly pipe between the pump and the carbon column it was possible to minimise the pulsation of the pump flow, only a slight movement of the carbon was observed.

The liquor overflow from the column was returned to the leach vat via a 20mm poly pipe, the flow was plunged into the vat liquor.

Provided a bubble trail extended from the return stream entry point as shown in the photo, the dissolved oxygen level in the vat liquor was kept high for efficient leaching.

The carbon column had 7 kg of washed carbon placed in it after the circuit had only water pumped through it for 24 hours as a leak check.

The flow rate of the liquor up the carbon column was set at 70 litres / hour and the first few minutes of flow was run to waste to get rid of any carbon fines generated by the placement of the carbon in the column.

A cyanide level of 1 gram/litre sodium cyanide was used and the vat was run for 4 weeks. At this stage the gold in liquor levels had dropped from an initial 60ppm to 1.2ppm and the daily return was approaching break even stage.

The liquor was then pumped into a drum for further use and after two rinse runs each of three bed volumes a hypochlorite leach in 20% salt was introduced into the vat. The calcium hypochlorite level was 10 grams per litre at pH 6.5.

The carbon was changed between the two runs to prevent reagent wastage.

After 1 hours pumping the gold tenor in the leach liquor had raised to 7 ppm and it maintained around this level for two weeks before tapering off to just under 1ppm after 3 weeks.

The Eh of the solution remained around 980 mv for the entire time, pH was adjusted with HCl

The liquor was then pumped to storage, the vat subject to a rinse run, the vat solids shovelled out and a new batch of material placed inside and the cycle was started again.

The method of leaching is capable of being scaled up or down depending on what application is proposed.

By keeping the wetted surfaces non-metallic either form of leach could be run with no problems.

Deano

141.JPG
Suction hose placed on floor of vat with pump suction line attached.

154.JPG
Suction hose covered with sand and filter fabric sheet in place immediately prior to filling vat with slag.

161.JPG
Vat partially filled with suction line to pump

165.JPG
Peristaltic pump with speed controller
 
Further vat leach photos

167.JPG
Carbon column being leak tested


169.JPG
View of circuit, note that the pump and speed controller kept weatherproof in shed while vat and carbon column are in open air.


170.JPG
Sampling point at base of carbon column

168.JPG
Return liquor line from carbon column, note bubble trail showing high dissolved oxygen levels in vat liquor

Deano
 
Thank you for posting these descriptive photo's. A picture always clarifies things!

I assume the milled slag has been screened. Was the gold recovery from the oversize substantial or is the lion's share of the value small enough to pass the screen?
 
Recovery per ton of slag.

Oversize beads and strange shapes >1mm. = 1.3kg

Tabled beads < 1mm = 1.8 kg

Leached in vat, milled to 100% < 100 microns, cyanide leach = 280g, followed by CaOCl leach = 180g.

Other slags have similar grades of gold, much depends on the ore type and the processing at the mine including the smelting flux formulation.

Deano
 
Very interesting setup, that's on a scale that even I could handle. :)

Last week I binge-watched the TV-series "Aussie Gold Hunters" season 2. In episode 10 (originally aired a couple of months ago) one team constructed a heap leaching pad and it was shown in quite good detail. One of the problems they encountered was (according to them) the water quality, a lot of crystals formed in the pipes, clogging the system. The water source was a deep well at the site (about 40 km outside of Kalgoorlie) and they suspected that it contained too much dissolved minerals and that interfered with the cyanide.
At that point they shut down the system, ashed the carbon and smelted the ash to recover the gold.

Now to my question... How does the water quality affect the cyanide leaching as long as the pH is controlled? Does hardness of the water affect it? What substances in the water can interfere with the cyanide leach.

Of what I remember from my brief visit in Kalgoorlie 15 years ago was that the deep ground water in the area was heavily contaminated by arsenic and salt. All fresh water was brought in via a pipeline or by road.
Maybe their problems was only evaporation and the crystals formed was from ordinary salt.

Göran
 
Kalgoorlie was one of the highlights of my trip to Australia last year Goran. What an incredible place it is, although I'm sure it may have changed somewhat in the 15 years since you went.

ps I've actually leant against that exact shed wall with a cig in one hand and a bottle of beer in the other discussing how little I actually know about refining with Dean. 8)

Something I will never ever forget.

Jon
 
OK I can imagine members drooling on their screens from these numbers. At $1320 gold we are talking about a recovery worth $151,000 from a ton of slag!

So to put this into perspective, how many tons of ore are processed (and approximate assay of unprocessed ore) to generate one ton of slag. From a refiners standpoint I'd be saying screw the raw ore let me process slags for a living!

Surely the slag generating 114 ounces of gold as in this example was from an ore body representing a much larger quantity of gold. Do you have any idea of the quantity of gold generated from the original ore resulted in slag of this quality?
 
The ratio of slag to bar metal is very operator and mine dependent, but a very general approximation is 2:1 without any allowance for metals transferred into the slag such as iron and copper.

So look at needing slag from a high producing mine to generate appreciable quantities of slag.

If you accept that the gold grade into the plant approximately equals the amount of recovered metal then a 20kg bar will be recovered from processing 4,000 tons of 5 gram ore.

At a through rate of 100 tons/hour you are looking at 40 hours processing, call it 2 days worth.

So 10 days per 100 kg metal or 50 days for 500 kg metal = 1 ton slag.

There is very little romance in the treatment of slag, it is a dirty, dusty process you go through only because the rewards are good.

This pretty well sums up gold mining in general.

Deano
 
One of the main problems in gold mining areas, notably Kalgoorlie, is that the ground water is often contaminated with very high levels of salts, especially sulphate ions.

This means that using lime for pH control in such areas will result in the generation of gypsum and precipitation of same.

Some clever systems of lower pH cyanide leaching were developed to overcome this problem, many mines will shandy the poor quality water with just enough higher quality, read more expensive, water to allow efficient processing.

Smelting the ash from carbon used in ore processing will lead to large losses of metal values, the most efficient method is leaching the ash.

Deano
 
I have only seen it once, but perhaps you are familiar with a device, that I shall call "air-sluice", a series of 3 meter plexiglass pipes, of increasing diameters, with small riffles inside. It concentrated better than 20 to 1. with 98% recovery super fine, super dry dust of < 100 microns. Where water-based gravity recovery failed, vacuum or pressured air worked perfectly. I imagine it would work even better with a tight sized feed. :G
 
cuchugold said:
I have only seen it once, but perhaps you are familiar with a device, that I shall call "air-sluice", a series of 3 meter plexiglass pipes, of increasing diameters, with small riffles inside. It concentrated better than 20 to 1. with 98% recovery super fine, super dry dust of < 100 microns. Where water-based gravity recovery failed, vacuum or pressured air worked perfectly. I imagine it would work even better with a tight sized feed. :G
You would not happen to have a photo of that "air sluice" would you? Ive been doing some research on dry concentrating methods and this is one that I have not come into contact with. Very interesting concept.
 
Subverted said:
cuchugold said:
I have only seen it once, but perhaps you are familiar with a device, that I shall call "air-sluice", a series of 3 meter plexiglass pipes, of increasing diameters, with small riffles inside. It concentrated better than 20 to 1. with 98% recovery super fine, super dry dust of < 100 microns. Where water-based gravity recovery failed, vacuum or pressured air worked perfectly. I imagine it would work even better with a tight sized feed. :G
You would not happen to have a photo of that "air sluice" would you? Ive been doing some research on dry concentrating methods and this is one that I have not come into contact with. Very interesting concept.

If you keep the sizing tight, i.e. -200 mesh/+400 mesh, -400/+800, etc. You only need one diameter of pipe, and an adjustable speed vaccuum cleaner with a 55 gallon drum for the receiving chamber (or blower for production applications). At high enough speed everything gets airborne, at zero speed everything settles, the right speed is that which retains all heavies in the pipe, and zero gold flakes in the receiving chamber. A transparent pipe is almost a necessity, IMO.
 

Attachments

  • airsluice.jpg
    airsluice.jpg
    24.2 KB · Views: 539
Deano said:
Recovery per ton of slag.

Oversize beads and strange shapes >1mm. = 1.3kg

Tabled beads < 1mm = 1.8 kg

Leached in vat, milled to 100% < 100 microns, cyanide leach = 280g, followed by CaOCl leach = 180g.

Other slags have similar grades of gold, much depends on the ore type and the processing at the mine including the smelting flux formulation.

Deano
Hi Deano: What is the rationale for doing a cyanide leach followed by a chlorine leach?. And how much gold was left after all of that?. TIA.
 
A cyanide leach is the simplest and easiest to run in a vat format so if you are time poor you do this first.

The hypochlorite leach is more efficient than the cyanide leach at allowing the leach access to the gold through the coatings on the gold particles. It does , however, require more supervision than the cyanide leach as the hypochlorite will react with more metals than will a cyanide leach.

Usually there is a lot of iron in the slag from the electrowinning stage of processing. This iron will not interfere with a cyanide leach but will consume highish levels of hypochlorite.

By removing what gold grades you can by using a cyanide leach you minimise the contact time required for the hypochlorite leach and thus lessen the supervision time needed.

Assaying the residual slag is difficult, it is not fire assay's finest moment.

Aqua regia digest will often (read usually depending on the flux formulation) form a gel and be unreliable in the values.

The best method I have found is to cover the slag sample with 10% HCl and let it sit for several months. Keep topping the liquid level up with water.

After leaving the brew to stew for as long as your impatience will let you, you can filter off the residual acid solution and do any of the standard assays.

This generally shows a gold level in the residual slag of 100 - 400 ppm.

It all comes down top how much slag you can get hold of and how long you can leave it in the HCl soak as to what level of recovery you get.

You will certainly recover more gold than the plant operators who just throw the slag in the mill expecting total recovery.

Deano
 
Back
Top