Geometallurgy for mine data - explaining the advantages

Materials World magazine
,
1 Jul 2009

Mike Hallewell of SGS Mineral Services’ UK-site believes in the merits of geometallurgy. Michael Forrest reports

It may seem that once you have found your mineral deposit it is just a matter of buying yellow equipment to mine and process the ore. However, many factors can weaken or destroy the economics of the project.

In the past, the feasibility of metal recovery was tested with a large sample, perhaps 100t or more, and the recovery proved in a pilot plant. This significantly reduces metallurgical risk since the pilot plant is a scaled-down replica of the mine processing path. Pilot plants typically cost US$300,000-$1m and require 100-1,000t of ore.

At a recent IOM3 local society meeting, Mike Hallewell of SGS Mineral Services’ UK branch explained that this method, although sound, has a fatal flaw due to the ‘sampling density’. When piloting, the traditional approach is to select several large composite samples from various zones within the deposit. They are then subjected to grinding, gravity separation and flotation. The difficulty is that the sample, although large in mass, is from one location within the deposit. As the ore is mined, it will vary over time.

Quantifying the variability is the job of the geology team. Using techniques including geometallurgy and 3D modelling, it determines ore zones or domains using the geochemical, mineralogical and textural data obtained from diamond drill cores. If the ore is complex, there will be many geometallurgical domains, all of which will have to be sampled to document variability in the deposit.

According to Hallewell, ‘Chemical analysis of the core gives an absolute grade, but this value may have no relationship to the recovery, physical competence or metallurgical variability within the ore body. These parameters must also be tested. The geometallurgical framework provides the solution’.

Daily grind

The grinding circuit of a metallurgical plant contains mine equipment that is among the most costly to buy and operate. Autogenous and semi-autogenous grinding mills (AG/SAG) allow mines to treat high tonnage, low grade ores quickly and cost effectively. However, AG/SAG mill operation is susceptible to the physical competence (partly controlled by hardness) of the ore. Furthermore, once this equipment is installed, it is difficult and expensive to change its operating configuration.

Mineral domains or zones within an orebody are geometallurgically defined by combining data from chemical analyses, high definition mineralogy, rock textures, and other core log and geological data. Traditionally, the mineralogy of the ore and gangue (waste) was identified using optical microscopy. Core containing different minerals was evaluated to determine what grinding size (P80) yielded the best separation of valuable and waste minerals. This would then be applied across the whole orebody, resulting in a very simplistic model. By increasing the number of samples selected for grindability or metallurgical testing, a better understanding of the variability can be obtained.

As an exploration programme moves to pre-feasibility and feasibility, the objectives change, along with the testing required to meet them. For instance, early in an exploration programme, when the core is only used for geochemical analysis, testing consumes between half and two grammes of material per test and a one to two kilogramme sample is needed to be representative. However, analysis later in the project needs more core volume. Metallurgical testing consumes 2.5-25kg/test and typically 10-75kg is requested. Analytical rejects are seldom useful. Because they are pre-crushed they cannot be used for grinding tests, and since they may be oxidised, they are no use for flotation tests.

The diamond drill core must be managed carefully to ensure that, while the samples selected for testing are representative, sufficient material is conserved for later work. In many instances, ‘metallurgical holes’ must be drilled to provide samples for metallurgical testing because core usage has not been well planned or the test programme requires more weight than is available.

‘SGS is continually reducing the amount of core and the topsize (mm) required in testwork programmes to increase the use of clients’ core,’ says Hallewell. ‘For example, tests such as the SAG Power Index and Static Pressure Test use small sample volumes – as low as two kilogrammes in weight. This means that we can use core for variability samples. So we can map the grindability of the domains throughout a deposit and properly specify the grinding mills.’

In addition, SGS has developed standardised protocols to clearly quantify variability in flotation circuits (the MFT test), and is researching the use of the technology in gravity and leaching processes.

Mapping minerals

The most promising development within geometallurgy to date is mapping the variability of minerals and mineral textures within an ore deposit using high definition mineralogy. It is considered ‘high definition’ because of the greatly increased number of samples that can be analysed as a result of faster modern instrumentation.

The advent of the QEMSCAN technology, developed by the Commonwealth Scientific and Industrial Research Organization of Australia, facilitates quantitative mineralogical examination. Core or rock samples are cut and made into polished blocks. These are scanned by the QEMSCAN, a combination scanning electron microscope and energy dispersive X-ray fluorescence instrument, which identifies and sizes all the minerals in the sample.

The process is automated so that many samples can be analysed without operator intervention. Hallewell notes that every ore domain has its own signature and high definition mineralogy exploits this to the best effect. ‘Geometallurgy significantly improves the understanding of ore variability and lessens the spatial uncertainty in mine planning. By identifying the variation across the ore body, we can map grindability and metallurgical performance. It provides a firm platform to build production forecasting models. It will generate throughput, product yield/grade and recovery data throughout the mine life.’ He adds, ‘The data from geometallurgy can also be used to construct more accurate net present values (financial valuations) that have direct implications in bankable feasibility studies’.

Although the use of piloting is still invaluable to assess certain flowsheets, multiple small scale tests with the latest modeling technology has made the interpolation of metallurgical performance on a block-by-block basis far more accurate. It reduces the spatial risk by applying 200 or 300 metallugical performance tests across a 100Mt ore body instead of the two or three tests used previously. With or without a pilot plant, it reduces the technical risks involved, and has been facilitated over the past 20 years as computing power and storage space have improved and modelling techniques have become more reliable.

Hallewell says geometallurgy is a holistic approach that can be applied across the board, but care is required in application. SGS has used such techniques on deposits and mines including Los Pelambres in Chile, Voisey’s Bay in Labrador, Canada, Bulyanhulu in Tanzania, and Chelopech in Bulgaria.

He estimates that there is growing recognition of the value of the technology in the industry and believes that in the 1980s, few feasibility studies examined the spatial ore variability, whereas today such testing is virtually universal. The majority of unsuccessful mining projects fail because of the lack of understanding of how geology and mineralogy can affect metallurgy and thus cash flow, he adds.

Further information: SGS Mineral Services