Fine gold is the fraction that connects almost every page in this recovery methods set back to a single physical fact: gold below roughly 75 microns behaves differently in a gravity circuit than coarse, liberated gold does, and understanding why is what determines whether a tailings dump needs specialized equipment or standard gravity recovery covered on gravity recovery of gold from tailings. The 75-micron threshold itself is a practical rule of thumb rather than a hard physical boundary, since the exact point at which recovery efficiency starts falling off shifts somewhat with gangue density, water viscosity, and the specific equipment in use, but it remains a useful reference point across the industry.
Why does gold below 75 microns behave differently in a gravity circuit?
Gold below roughly 75 microns settles too slowly, relative to the surrounding gangue, for its density advantage to dominate the way it does at coarser sizes, because settling velocity in a fluid falls off sharply as particle diameter decreases. At coarse sizes, gold's roughly sevenfold density advantage over silicate gangue translates directly into a large settling-velocity advantage, which is what any gravity separator exploits. As particle size drops, drag and turbulence start to compete with that density advantage, and below a certain size, the two gold and gangue particles settle at similar enough rates that a standard gravity separator can no longer tell them apart.
The underlying physics here is Stokes' law, which describes settling velocity as proportional to the square of particle diameter but only linearly proportional to the density difference between particle and fluid. That squared relationship to size is the crux of the problem: halving a particle's diameter cuts its settling velocity far more than halving the density advantage would, which is why gold's density edge, substantial as it is, eventually loses the argument to sheer smallness. No amount of operator skill changes this relationship on standard gravity equipment; it can only be overcome by adding an external force, which is exactly what a centrifugal concentrator's spinning bowl does.
Is all fine gold in tailings free, or is some of it locked?
Not all fine gold in a tailings dump is free-milling; a meaningful portion can still be locked inside gangue mineral particles, encapsulated in sulfides, or otherwise shielded from direct exposure, and no amount of gravity or centrifugal force recovers gold that a gravity separator's fluid flow never actually contacts as an exposed surface. This is the liberation half of the fine gold problem, distinct from but closely related to the pure settling-velocity problem, and it is why particle size distribution testwork is always paired with a liberation assessment rather than treated as sufficient on its own. A QEMSCAN or equivalent mineralogical analysis identifies not just how small the gold particles are but how much of that gold is actually exposed at the current grind, which determines whether finer grinding, rather than better gravity equipment, is the more relevant investment for a specific dump.
Which equipment actually recovers this fraction?
Recovering fine gold takes equipment purpose-built for it: a Falcon UF unit specifically, fine-tuned centrifugal models generally, and ultimately flotation or leaching for whatever fraction even the best gravity equipment still misses. The practical reality on most Ghanaian tailings dumps is that gravity, however well specified, only ever recovers part of the fine fraction, which is exactly why the hybrid gravity-leach flowsheets covered on tailings gold recovery methods exist rather than a gravity-only approach.
Flotation is worth naming specifically here as a complementary, non-gravity route for the fine fraction. Where fine gold is associated with sulfide minerals, pyrite in particular, flotation can recover a gold-bearing sulfide concentrate regardless of particle settling velocity, since flotation separates by surface chemistry rather than by density and size. A tailings stream carrying meaningful sulfide-associated fine gold may therefore route part of its fine fraction through flotation ahead of, or instead of, further gravity effort, with the resulting sulfide concentrate then treated separately, often by further leaching, to release the gold it carries.
Cyanide leaching remains the backstop for whatever fine gold neither gravity nor flotation recovers, since leaching dissolves gold at the particle surface regardless of the particle's physical size or settling behavior, as long as the gold surface is actually exposed to the leach solution. This is the practical reason a gravity-only flowsheet is rarely the complete answer on tailings with a significant fine fraction, and why the carbon-in-leach and carbon-in-pulp circuits covered elsewhere in this recovery methods set exist as the fallback stage rather than an alternative to gravity.
How is the fine fraction identified before design, not assumed?
The fine fraction in a specific dump is identified through particle size distribution analysis rather than assumed from a general industry pattern, because two dumps with the same head grade can have very different size distributions depending on the original plant's grinding history. That analysis is what a flowsheet designer actually uses to decide how much capacity to build into fine-gold-specific equipment versus standard gravity duty.
A representative sample program for this purpose typically screens material across a full range of size fractions and assays each fraction separately, producing a size-by-size gold deportment picture rather than a single bulk grade number. That deportment curve is what actually tells a project what share of contained value sits above and below a given cut point, information a single composite assay cannot provide on its own.
What role does classification play upstream?
Classification, typically cyclones, determines what particle size range actually reaches fine-gold-targeted equipment in the first place, since a poorly set cyclone cut point can either starve a UF-class concentrator of the fine material it's built for or overload it with material coarse enough for a standard unit to handle instead. Getting the upstream cut point right is as much a part of solving the fine gold problem as the recovery equipment itself.
Cyclone performance itself depends on feed pressure, pulp density, and the specific cyclone's geometry, and a unit tuned for one tailings stream's particle size distribution will not automatically perform correctly on a different stream with a different profile. Periodic cyclone survey work, checking the actual cut point achieved against the design intent, is standard good practice on any circuit relying on classification to feed size-sensitive equipment downstream.
Why is the fine gold problem especially significant on Ghanaian legacy tailings?
Ghana's legacy tailings dumps span decades of processing history, and many originate from plants operating with the equipment and metallurgical understanding of their era, gravity circuits and amalgamation among the earliest, which had essentially no ability to address the fine fraction at all. Layered on top of that history, artisanal and small-scale processing of already-milled material has, in some cases, further concentrated the fine fraction in what remains, since coarser gold is the easiest target for informal recovery methods and is disproportionately removed first. The practical result is that a meaningful share of a legacy Ghanaian dump's remaining contained value often sits precisely in the size range conventional equipment cannot reach, which is exactly why fine gold recovery capability is not an optional refinement on these projects but frequently central to the investment case itself.
Is fine gold recovery ever not worth pursuing?
Below a certain contained-ounce threshold, the extra equipment and testwork cost of chasing the fine fraction specifically may not be justified, the same cut-off-grade economics already established on tailings vs waste rock: the decision is a comparison against processing cost, not an assumption that finer recovery is always worth the investment.
Why this is the historical reason tailings exist as a resource at all
The fine gold problem is not a modern quirk; it is the direct, physical reason old tailings still contain gold, since historic gravity-only and amalgamation-era plants had no answer to it at all.
How particle size testwork settles the question for a specific dump
Quantifying exactly how much of a dump's gold sits in the fine fraction, and how liberated it is at the current grind, is a testwork question in its own right, covered on particle size, liberation and recovery.
Why regrinding sometimes solves the problem better than better equipment
Where testwork shows the fine fraction is dominated by locked rather than free gold, adding the regrind stage described on regrind milling: when tailings need finer grinding ahead of recovery can outperform any amount of additional gravity or flotation capacity, since regrinding exposes gold surfaces that were never physically accessible to a separator in the first place. Whether regrinding, better recovery equipment, or some combination is the right answer depends entirely on that testwork result, not on a default assumption about which fix applies.
Where the practical recovery limit sits even with the right equipment
Even correctly specified fine-gold equipment eventually hits a genuine recovery ceiling as particle size drops further still, a limit worked through in full on ultrafine gold: the limits of recovery.