When a gold ore leaches poorly, the unrecovered gold is either inaccessible to cyanide, for example encapsulated in pyrite, arsenopyrite or quartz, or it dissolves and is then re-adsorbed by natural carbon in the ore before the plant can recover it. The diagnostic leach addresses the first case by dissolving mineral groups in sequence and leaching the gold released at each stage, so that the gold's location is inferred from the stage at which it reports. The preg-robbing test addresses the second case by measuring whether the ore itself removes dissolved gold from solution.
These tests are the first steps when an ore is refractory or when a plant's recovery has fallen for reasons a bottle roll test cannot explain. This article sets out a typical test sequence, the calculations and the interpretation of the results. Both tests form part of the investigation method we describe for plant problems.
A typical diagnostic leach sequence
- Direct cyanidation of the ground sample, in which the gold that dissolves is free-milling.
- Hydrochloric acid followed by cyanide, which dissolves carbonates, pyrrhotite and some oxides; the gold released is reported as locked in those minerals.
- Nitric acid followed by cyanide, which oxidises sulphides such as pyrite and arsenopyrite; the gold released was locked in sulphides.
- Final residue, in which the remaining gold is mostly locked in silicates and is determined by fire assay.
Laboratories add or modify stages to suit the ore, for example a sulphuric acid stage, a roasting stage, or a stage to destroy carbonaceous matter. Carbon is often added to the cyanide stages so that the gold released is captured before natural carbon in the ore can re-adsorb it. The detailed design of the test is important, and Lorenzen's guidelines are the standard reference.
Calculating gold deportment
where
- gold reporting to stage k, including the final residue (g/t of original ore)
- a preg-robbing index: the percentage of a gold spike removed from solution by the ore
- ,
- gold in the spiked solution at the start and after contact with the ore (mg/L)
| Stage | Gold reporting (g/t of original ore) | Share (%) | Interpretation |
|---|---|---|---|
| Cyanide | 1.80 | 61.2 | Free-milling |
| HCl, then cyanide | 0.15 | 5.1 | In carbonates, pyrrhotite |
| HNO₃, then cyanide | 0.75 | 25.5 | In sulphides |
| Residue | 0.24 | 8.2 | In silicates |
| Calculated head | 2.94 | 100 |
The gold from each stage is expressed per tonne of the original ore, allowing for the mass lost as minerals dissolve, so that the stages sum to a calculated head. This should be compared with the assayed head, since a large difference indicates that gold was lost or a stage was miscounted.
Interpreting the result
A high free-milling share combined with low plant recovery indicates that the problem lies in the plant rather than the ore: grind, leach time, oxygen, cyanide or preg-robbing.
A large sulphide-locked share, such as the 26% in the example, calls for testing of finer grinding, flotation of the sulphides, and oxidation of the concentrate by pressure oxidation, bio-oxidation or roasting.
A significant share in carbonates or pyrrhotite also deserves attention, because pyrrhotite consumes cyanide and oxygen; pre-aeration and an oxygen uptake test should be considered.
Gold locked in silicates can usually be released only by finer grinding, often at a cost that the gold does not justify.
The preg-robbing test
Some ores contain natural carbon, graphitic or organic, or certain clays that adsorb gold cyanide and so remove gold from the pregnant solution. A simple test spikes a solution of known gold concentration, contacts it with the ore under leach conditions, and measures the gold lost from solution.
Worked example: a solution spiked at 5.0 mg/L gold is contacted with the ore for the test period, after which 3.4 mg/L remains. The ore has removed (5.0 − 3.4) ÷ 5.0 = 32% of the spike, indicating a strongly preg-robbing ore under these test conditions. Laboratories use different contact times and ratios, so results should only be compared within one method.
The test should be supported by assays for total organic carbon and graphitic carbon, and by a comparison of a standard bottle roll with a carbon-in-leach bottle roll on the same ore. If the carbon-in-leach test recovers substantially more gold, preg-robbing is reducing recovery.
Managing preg-robbing ore
Carbon-in-leach allows plant carbon to compete with the natural carbon from the start of the leach (see CIL and CIP). Blanking agents such as kerosene or diesel coat the natural carbon, but must be tested carefully because they can also foul plant carbon. Preg-robbing ore can be blended with non-preg-robbing ore within limits established by testing, and carbonaceous material can be removed by flotation or destroyed by oxidation.
Glossary
- Diagnostic leach
- A sequence of leaches and acid treatments that shows which minerals hold the gold.
- Deportment
- How the gold is distributed between minerals or forms in an ore.
- Free-milling gold
- Gold that dissolves directly in cyanide after normal grinding.
- Refractory ore
- Ore from which cyanide alone recovers gold poorly.
- Preg-robbing
- Re-adsorption of dissolved gold by natural carbon or clays in the ore.
- Blanking agent
- A chemical that coats natural carbon to reduce preg-robbing.
- Total organic carbon
- An assay of carbon present as organic matter, a warning sign for preg-robbing.
Frequently asked questions
What does a diagnostic leach show?
Where the gold that cyanide cannot reach is located: in carbonates or pyrrhotite, in sulphides, or in silicates. That indicates the treatment that could recover it.
How can preg-robbing ore be identified?
Spike a gold solution onto the ore and measure the gold lost, compare a standard bottle roll with a carbon-in-leach bottle roll, and assay for organic and graphitic carbon.
Why is carbon added to the cyanide stages of a diagnostic leach?
So that gold dissolved at each stage is captured before natural carbon in the ore can adsorb it, which would otherwise make the gold appear locked when it is not.
Is a diagnostic leach a design test?
No. It is a diagnostic tool that directs further work, such as grind, flotation and oxidation tests, from which the design figures are obtained.
Discuss your plant with a metallurgist
Tiegel is a Perth metallurgy company. We help with testwork, plant optimisation, metal accounting and investigations, and we build the Tiegel Suite software.
Talk to a metallurgistReferences and further reading
- Lorenzen, L. (1995). Some guidelines to the design of a diagnostic leaching experiment. Minerals Engineering, 8(3), 247–256.
- Marsden, J.O. and House, C.I. (2006). The Chemistry of Gold Extraction, 2nd edition. Society for Mining, Metallurgy and Exploration.
Worked examples and charts marked illustrative use invented figures, not data from any operation.