Tiegel home Talk to a metallurgist

Gravity recoverable gold: interpreting the GRG test for circuit design and metal accounting

How the GRG test is run, how to interpret the result, why plant gravity recovery is lower than GRG, and the implications for circuit design and the month-end.

· 5 minute read · By

Gold has a density about 19 times that of water, compared with roughly 2.7 for typical rock, so once liberated by grinding it can be recovered by gravity. Coarse free gold is often better recovered this way than left to the leach: it dissolves slowly, can pass through a CIL train only partly leached, and tends to accumulate in the grinding circuit. Before investing in a gravity circuit, however, it is necessary to know how much of the gold could be recovered by gravity. That is what the gravity-recoverable gold (GRG) test measures.

Test method

The standard test was developed by André Laplante and co-workers at McGill University. As commercial laboratories run it today, a sample of 50 to 100 kg is processed through a laboratory centrifugal concentrator, typically a Knelson unit, at three successive grind sizes, from a P80 of about 850 µm down to about 75 µm. After each stage the concentrate is removed and the tailing is ground finer for the next stage. Each concentrate and the final tailing are assayed by size fraction. Smaller, simplified versions of the test exist, but a large sample is important because coarse gold is unevenly distributed through ore.

The result is the cumulative percentage of gold recovered at each stage, together with the size distribution of that gold. Gold recovered at the coarse stage is liberated early; gold that appears only after fine grinding must be liberated before any gravity device can recover it.

The three-stage GRG testGrind, concentrate, regrind the tailing finer, concentrate again, three times.Stage 1: grindP80 ≈ 850 µmCentrifugalconcentratorConcentrate 1assay by sizeStage 2: grindfiner grindCentrifugalconcentratorConcentrate 2assay by sizeStage 3: grindP80 ≈ 75 µmCentrifugalconcentratorConcentrate 3assay by sizetailingtailing
The GRG test passes one sample through a laboratory centrifugal concentrator at three successively finer grinds. Each concentrate, and the final tailing, is assayed by size.

Interpreting the result

The headline GRG figure is less informative than three other aspects of the result.

The first is the size at which the gold is recoverable. Gold reporting at the coarse stage can be recovered from the mill discharge or cyclone underflow before it is overground. Gold that reports only at 75 µm will be considerably harder to recover in a plant.

The second is variability. GRG can vary widely between ore types and between samples of the same ore type, so a single test on one composite is a weak basis for a capital decision. Each geometallurgical domain expected in the plant feed should be tested.

The third is the match between the gold size distribution and the plant grind. Comparing the GRG size distribution with the cyclone underflow sizing shows what a gravity unit in the circuit would actually receive.

Plant gravity recovery compared with GRG

GRG represents the gold that could be recovered by gravity under near-ideal laboratory conditions. It is an upper limit rather than a forecast, and laboratories note that plant results are lower for several reasons. A laboratory bowl recovers a much larger proportion of the feed mass as concentrate than a plant unit, in which the concentrate is a fraction of a percent of the feed. The plant treats only a bleed of the circulating load, so some gravity gold bypasses the unit on each pass. A batch test also cannot reproduce a continuous closed circuit, in which gold circulates many times before it is either recovered or ground fine.

Designers therefore use scaling factors or circuit simulation to estimate plant gravity recovery from GRG. Any single conversion factor should be treated with caution, and its derivation should be understood.

Implications for circuit design

Gravity circuit in a grinding circuitMill discharge to cyclones; part of the underflow bled to a centrifugal concentrator; concentrate to intensive leach; overflow to leach.Ball millCyclonesGravityconcentratorIntensive leachto gold roomLeach / CILdischargeoverflowbleed of underflowrest of underflowgravity tailing
Dense gold collects in the cyclone underflow, so gravity units usually treat a bleed of that stream and return their tailing to the mill.

A high GRG with coarse gold generally justifies a gravity circuit; a low GRG concentrated at fine sizes may not.

Gravity gold accumulates in the cyclone underflow, because dense particles report there, and for that reason the gravity feed is commonly taken from the circulating load.

Removing coarse gold early also reduces the risk of slow-leaching gold reaching the tailings.

Gravity gold in metal accounting

Gravity gold presents its own accounting challenges. It is recovered from the grinding circuit, often treated by intensive leaching, and poured separately. Gold also accumulates in the mill circuit as an unmeasured inventory that varies with throughput and ore type, and a change in that inventory between month-ends appears as an apparent loss or gain in the balance. Recording gravity recovery separately, and estimating the circuit inventory consistently each month, keeps the month-end defensible.

Ajax includes gravity concentrators among its unit models, so a gravity bleed can be evaluated on a plant model before it is built. In Saldora, gravity gold and leach gold can be set up as separate measured streams in the daily figures.

Glossary

Gravity-recoverable gold (GRG)
The share of the gold in an ore that a laboratory gravity concentrator can recover under the standard test.
Centrifugal concentrator
A spinning bowl that uses many times the force of gravity to trap dense particles such as gold.
Liberation
Freeing a valuable mineral from the rock around it by crushing and grinding.
P80
The size that 80% of the material passes through, used as a single number for the fineness of a grind.
Circulating load
Material that a grinding mill sends back to itself through a classifier before it is fine enough to leave.
Cyclone underflow
The coarse, dense stream from a hydrocyclone, which returns to the mill. Gravity gold collects here.
Intensive leaching
A high-strength cyanide leach used to dissolve gold from a small gravity concentrate.

Frequently asked questions

Is GRG the same as the gravity recovery a plant will achieve?

No. GRG is an upper limit measured under near-ideal laboratory conditions. Plant gravity recovery is lower, and designers use scaling factors or simulation to estimate it.

How much sample does the GRG test need?

The full test uses 50 to 100 kg. Smaller, simplified versions exist, but a large sample is important because coarse gold is unevenly distributed through ore.

Why does gold accumulate in the grinding circuit?

Gold is so dense that the cyclone returns it to the mill repeatedly. Without a gravity circuit, it accumulates until it is ground fine enough to leave.

Should every ore type be tested?

Yes. GRG can vary widely between ore types, so each domain that will be fed to the plant should be tested.

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 metallurgist

References and further reading

  1. ALS. Gravity gold recovery: the GRG test and frequently asked questions. Link
  2. Laplante, A.R. and Clarke, J. (2006). A simple gravity-recoverable gold test. Canadian Institute of Mining, Metallurgy and Petroleum. Link
  3. Koppalkar, S. (2010). Evolution and optimization of the gravity recoverable gold test. Canadian Institute of Mining, Metallurgy and Petroleum. Link

Worked examples and charts marked illustrative use invented figures, not data from any operation.