In a carbon-in-leach (CIL) plant, ground ore flows as slurry through a train of agitated tanks, where cyanide dissolves the gold and activated carbon in the same tanks adsorbs it. Carbon is pumped upstream against the slurry and removed loaded from the first tank. A circuit model predicts the gold remaining in the residue, the gold adsorbed onto carbon and the gold lost in solution, so that changes such as additional tanks, longer residence time, higher throughput or a different carbon movement rate can be evaluated before they are made.
This article builds a simple CIL model from three components: the leach kinetics, the effect of agitated tanks in series, and the gold and carbon balances. More detailed adsorption models, which track gold on carbon tank by tank, are built on the same balances. The differences between CIL and CIP are covered in CIL and CIP circuits.
Leach kinetics and tanks in series
Bottle roll tests usually show rapid initial dissolution that levels off at a residue that does not leach. A first-order model with an unleachable residue fits many ores, with the leachable portion decreasing exponentially with time. In plug flow, with no back-mixing, all of the slurry would spend exactly the design time in the leach. In agitated tanks, some slurry leaves early and some late, and the early fraction carries unleached gold. The standard result for equal, well-mixed tanks in series is:
where
- first-order leach rate constant for the leachable gold (h⁻¹), fitted to bottle roll or pilot data
- total residence time of the circuit (h)
- ,
- number of equal tanks and residence time of each tank (h)
- head grade (g/t)
- gold that will not leach however long it stays, for example locked in sulphides (g/t)
- predicted residue grade (g/t)
Worked example: head grade 2.0 g/t, unleachable gold 0.15 g/t, rate constant 0.25 h⁻¹ (fitted to bottle roll results) and a total residence time of 24 hours.
| Tanks | Residence per tank (h) | Unleached fraction | Residue (g/t) |
|---|---|---|---|
| 1 | 24 | 0.143 | 0.41 |
| 2 | 12 | 0.063 | 0.27 |
| 4 | 6 | 0.026 | 0.20 |
| 6 | 4 | 0.016 | 0.18 |
| 8 | 3 | 0.011 | 0.17 |
| Plug flow | 24 | 0.002 | 0.155 |
Short-circuiting reduces recovery, which is why CIL circuits use six to eight or more tanks rather than a single large tank. The same model shows why a tank out of service, or a sanded tank with a reduced live volume, increases the residue grade.
Gold adsorption, soluble loss and carbon movement
where
- dry solids rate (t/h)
- solution (water) flow leaving with the tailings (m³/h)
- gold in solution in the last tank (mg/L, the same as g/m³)
- carbon that must be moved forward and eluted each day (t/d)
- overall recovery (fraction)
Continuing the six-tank example at 300 t/h and 45% solids by mass, gold leached is 300 × (2.0 − 0.179) = 546 g/h. The solution leaving with the tailings is 300 × 55 ÷ 45 = 367 m³/h, and at 0.01 mg/L gold in the last tank the soluble loss is 3.7 g/h, or 0.012 g/t.
Gold adsorbed onto carbon is therefore 546 − 4 = 543 g/h, or 13.0 kg per day. For a loaded carbon grade of 3,000 g/t, the carbon movement to elution is 13.0 ÷ 3.0 = 4.3 t/d, and overall recovery is (2.0 − 0.179 − 0.012) ÷ 2.0 = 90.4%.
The carbon inventory follows from the carbon concentration and the tank volumes. The slurry flow is 300 ÷ 2.7 + 367 = 478 m³/h, so each 4-hour tank holds about 1,900 m³ of slurry. At 15 g/L of carbon in all six tanks, the circuit holds about 172 t of carbon, and each granule spends about 172 ÷ 4.3 = 40 days in the circuit. Carbon movement rate and loaded carbon grade are linked: moving carbon faster lowers the loading, but reduces the gold inventory held in the circuit and the gold carried by the last tank, so soluble loss falls.
Applying the model
The rate constant and unleachable residue should be fitted to bottle roll results at plant grind and cyanide strength, and the model checked against a plant leach survey before it is used to support decisions.
For throughput changes, the model quantifies the residue penalty of a shorter residence time, which can be weighed against the additional tonnes treated.
Soluble loss is governed by the solution grade in the last tank. If it rises, carbon activity, the carbon concentration in the last tanks and the carbon movement rate should be checked.
The carbon inventory multiplied by its average loading is gold that has been recovered but not yet poured, and forms part of gold in circuit; see gold reconciliation and gold in circuit.
Glossary
- CIL
- Carbon in leach: leaching and adsorption onto carbon happen in the same tanks.
- First-order kinetics
- A rate that is proportional to what is left to react, giving an exponential decline.
- Plug flow
- Flow where every element of slurry spends exactly the same time in the vessel.
- Short-circuiting
- Slurry leaving a stirred tank sooner than the average residence time.
- Unleachable residue
- Gold that does not dissolve however long it is leached, for example gold locked in sulphides.
- Soluble loss
- Dissolved gold that leaves with the tailings solution.
- Carbon movement
- The rate at which carbon is advanced through the circuit and removed for elution (t/d).
- Loaded carbon grade
- Gold on carbon leaving the first tank (g/t of carbon).
Frequently asked questions
Why do CIL plants use many tanks rather than one large tank?
Because of short-circuiting. In a single large agitated tank, part of the slurry leaves almost immediately with its gold still undissolved. Several tanks in series behave more like plug flow.
How is the leach rate constant obtained?
By fitting the first-order model, with an unleachable residue, to bottle roll results at plant grind size and cyanide strength, then checking it against a plant leach survey.
How much carbon must be moved each day?
Gold adsorbed onto carbon per day divided by the target loaded carbon grade. For 13 kg/d at 3,000 g/t, the result is 4.3 t/d.
What determines soluble gold loss?
The gold in solution in the last tank multiplied by the tailings solution flow. Low carbon activity or insufficient carbon in the last tanks increases it.
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
- Marsden, J.O. and House, C.I. (2006). The Chemistry of Gold Extraction, 2nd edition. Society for Mining, Metallurgy and Exploration.
- Levenspiel, O. (1999). Chemical Reaction Engineering, 3rd edition. Wiley. Chapters on ideal reactors and reactors in series.
- Nicol, M.J., Fleming, C.A. and Cromberge, G. (1984). The absorption of gold cyanide onto activated carbon. II. Application of the kinetic model to multistage absorption circuits. Journal of the South African Institute of Mining and Metallurgy, 84(3), 70–78.
Worked examples and charts marked illustrative use invented figures, not data from any operation.