Adding structure analysis into iron-ore classification
Traditional analysis of iron ore samples provides information about their composition, but it is a blunt instrument.
Two samples with apparently similar mineralogical and chemical properties may behave quite differently during processing, because their structures are not the same.
Structural differences may include hardness, attrition resistance and moisture carrying capacity.
Geometallurgy reveals true process performance
These differences in structure are not revealed by standard geological analysis. It is only when the two disciplines of geology and metallurgy are combined – geometallurgy – that the true picture is revealed. And it is a vital step in understanding process performance and the parameters required to maintain product quality.
"Geometallurgy is a hybrid discipline comprising geological and metallurgical expertise and helps us understand how the geological characteristics of an ore or a material affect the metallurgical downstream processes," CSIRO research group leader, Keith Vining, says.
"One of the first examples of geometallurgy in an iron ore context, was when Australian iron ore exports transitioned from hematitic Brockman to more goethitic Marra Mamba ores.
"Customers were understandably wary. They wanted to be reassured that the new product would process efficiently, and understand how it should be handled.
"Geometallurgy solved these problems and gained market acceptance for the new ore types."
Textural classification
Today, geometallurgy is the basis for a new way of classifying Australian ore, known as textural classification.
Measuring the porosity of a sample provides an indication of the hardness, which allows it to be assigned to a particular textural group. This value leads to a greater understanding of how the material will behave in beneficiation processes, as far down the value chain as processing at the customer end.
Geometallurgy and textural classification is typically performed on ore that has been mined, but according to Dr Vining, that could well be about to change.
"At the moment we're very good at using geometallurgy to add efficiencies from mining onwards, but we're not as good earlier in the value chain," Dr Vining says.
"We think we could have a greater impact, by relating what we see in the textures to a physical signature we read from a sensor at the mine before blasting, or perhaps even at the exploration or resource definition stages.
"We're trying to expand the value of the textural classification scheme backwards towards the source."
Mineral4/ Recognition4 Optical Image Analysis software
Key to the expansion of textural classification is the Mineral4/ Recognition4 Optical Image Analysis software, developed by CSIRO and recently licensed to a Chinese company to distribute in the Asian market. The software allows for on-or off-site classification of ores and can provide an accurate breakdown of ore components.
Dr Vining also sees opportunities for textural classification in other markets and other commodities, such as bauxite because it is a high volume product with similar characteristics to iron ore.
"The opportunities are global, and as far as I know, no one else in the world has this capability," Dr Vining says.
"The main impediment to its wider use at the moment are the operational silos along the value chain in which a series of process-driven performance indicators are not interrelated.
"If people start viewing the value chain more holistically, I think textural classification will become widely adopted on an international scale."