Cyanide Consumers: Sulfide, Copper, Iron, and What Steals Your Leach
Only a small fraction of the cyanide charged to a gold plant ever dissolves gold. The rest is consumed by other minerals, oxidised in the circuit, or destroyed during effluent…
Only a small fraction of the cyanide charged to a gold plant ever dissolves gold. The rest is consumed by other minerals, oxidised in the circuit, or destroyed during effluent treatment. When a plant responds to a high tailings grade by adding more cyanide, it is often feeding the very reactions that were stealing the leach in the first place. This article reviews the main consumers and how to recognise them.
Start with the stream you cannot account for. The main gold path, carbon loop, and loss path should close as a mass balance.
Where the cyanide goes
Sodium cyanide dissociates readily:
NaCN → Na⁺ + CN⁻
In water, cyanide participates in an equilibrium with hydrogen cyanide. At about pH 9.3, half of the total cyanide exists as HCN. Raising pH shifts the balance toward free cyanide ions, which is one reason lime is dosed before the leach — both to protect against HCN loss and to keep cyanide available for dissolution.
But not all remaining cyanide is available for gold. Both free cyanide and HCN can be oxidised by oxygen to cyanate, which does not dissolve gold:
4HCN + 3O₂ → 4CNO⁻ + 2H₂O
3CN⁻ + 2O₂ + H₂O → 3CNO⁻ + 2OH⁻
A plant-wide mass balance may show that only a small fraction of the charged cyanide actually dissolves gold. The balance goes to other metals — the cyanide consumers — and to oxidation in the plant and in waste treatment. Knowing this reframes the dosing question: the plant is not only buying dissolution, it is also paying for every consumer in the ore.
Sulfide minerals
Sulfide minerals are among the most consequential consumers. They decompose in the leach and follow two damaging paths:
They form thiocyanate, or combine with oxygen to form sulfide and sulfate species — consuming cyanide and oxygen that gold dissolution needs.
They can form a film of gold sulfide on the gold surface, passivating the particle and slowing dissolution even when cyanide and oxygen are adequate.
How serious the effect is depends on the specific mineral and the available oxygen. The practical lesson is that a sulfide-rich ore can show adequate free-cyanide readings and adequate reagent addition while still leaching slowly. The barrier is at the surface, not in the bulk solution.
Copper
Copper minerals dissolve in cyanide to form a range of copper-cyanide complexes, and the speciation depends on cyanide concentration and pH. The damage is twofold: dissolving copper consumes both cyanide and oxygen that were intended for gold.
A useful teaching rule is to maintain a cyanide-to-copper mass ratio of approximately 3 for good gold extraction. If the ratio falls to around 2.5, extraction may slow. Even moderate levels of copper-cyanide in solution can exert a slight drag on gold dissolution; confirm the ratio with site-specific tests.
This is why copper ores are diagnosed on a ratio, not on a dose. Raising cyanide to cover copper consumption also raises reagent cost and the cyanide load in the effluent; the ratio tells you when you have added enough and when you are simply paying the consumer.
Diagnose consumption, not only concentration
A free-cyanide measurement is a concentration. It does not tell you how much cyanide the ore consumed since the last addition. The useful operating record pairs free-cyanide readings with a reagent balance:
consumption = cyanide added − free cyanide remaining − cyanide destroyed or removed in the measured streams
The exact accounting depends on the plant boundary, but the principle is important. A target of 300 ppm free cyanide can mean a healthy leach in one ore and a severe overfeed in another. If addition is high, free cyanide is high, and gold recovery is still poor, the problem is unlikely to be solved by raising the setpoint.
Read the chemistry in this order:
Is the reagent addition reaching the intended stream?
Is the measured value free cyanide or total cyanide?
Is pH stable enough to keep cyanide in the useful form?
Is dissolved oxygen available at the gold surface?
Which mineral or carbonaceous species can consume or adsorb the complex?
That sequence prevents a plant from treating a concentration target as a diagnosis.
Iron minerals
Iron behaviour is more varied, and the mineral form decides the outcome:
Iron mineral form
Behaviour in cyanide leach
Oxides (hematite, magnetite), silicates
Generally do not dissolve; little direct cyanide demand
Carbonates (siderite)
Slightly decompose below about pH 10; mostly unreactive at high pH
Sulfides (pyrrhotite)
Readily break down; consume cyanide and oxygen; form iron-cyanide complexes and sulfur compounds
Pyrrhotite is the aggressive case. It drives formation of the stable ferrocyanide ion and can deposit an iron-hydroxide layer on gold surfaces, hindering extraction. Interestingly, high ferrocyanide concentrations themselves can be tolerated — the problem is the consumption and the surface film, not the complex in isolation.
Preg-robbing: a competitor, not a consumer
A distinct loss mechanism deserves its own category. Carbonaceous matter, clay minerals, and graphite in the ore can adsorb the gold-cyanide complex directly from solution. This is preg-robbing: the ore itself behaves like activated carbon and steals dissolved gold before the circuit can capture it.
Preg-robbing cannot be solved by cyanide dosage. It is addressed by circuit design — CIL places activated carbon in the tanks while gold is still dissolving, giving the engineered carbon an early advantage — and by ore-blending or treatment decisions upstream.
Diagnosis before dosage
An illustrative ore-change case is useful. As feed shifts from an oxide-like feed to a sulfide-bearing feed, recovery can fall even when reagent consumption is adjusted:
Illustrative period
Feed grade (g/t)
Tailings grade (g/t)
Recovery
Oxide-like feed
2.80
0.34
88%
Sulfide-bearing feed
3.09
0.65
79%
After grind and cyanide changes
2.17
0.51
76.5%
In the example, cyanide concentration was raised from 800 to 1200 ppm and the SAG feed size reduced, yet the tailings grade remained roughly double the illustrative 0.3 g/t target and cyanide consumption rose. The ore change altered the consumer population; more reagent mostly paid the consumer bill. These values are teaching examples, not a plant record.
The disciplined sequence is:
Reconcile feed, recovery, and tailings assays on one basis.
Characterise the feed and residue — XRF for sulfur, arsenic, copper and related elements guides the mineralogical hypothesis.
Test particle size and cyanide concentration in a designed bottle-roll study while holding pH, density, and time at representative values.
Decide from the response surface, not from a single dosage step.
Tailings diagnostic tree
Start with a mass-balance check, then branch into physical, chemical, and sampling causes.
A diagnostic campaign is a decision tree, not a shopping list of extra cyanide.
~0.5%
share of plant cyanide that typically dissolves gold
CN:Cu ≈ 3
illustrative cyanide-to-copper ratio for good extraction
pH ≥ 10.3
illustrative floor to limit HCN formation
Check yourself
Check yourselfFree cyanide is within spec, dissolved oxygen is adequate, yet extraction has fallen after an ore change. What should you investigate?
Characterise the new ore for sulfide, copper, and iron minerals, and check for preg-robbing carbonaceous matter. A new consumer population or a passivating surface film explains slow leaching even when bulk-solution chemistry looks correct.
Check yourselfWhy is the cyanide-to-copper ratio more useful than a fixed cyanide set-point for a copper-bearing ore?
Copper consumption scales with the copper dissolved. A fixed dose can be excessive for a low-copper feed and insufficient for a high-copper feed. The ratio tracks the actual competitive demand and bounds both reagent cost and extraction risk.
Check yourselfWhy will more cyanide not fix a preg-robbing loss?
Preg-robbing is adsorption of the gold-cyanide complex onto carbonaceous matter in the ore. Extra cyanide makes more complex available for the ore carbon to steal. The remedies are circuit design (CIL timing), carbon inventory, and ore management.
Cyanide is acutely toxic and HCN can form under unsuitable chemistry. This article supports engineering study and test planning; any reagent work must follow the site chemical-management plan and qualified supervision.