By Keith Vining
The ability to accurately predict downstream processing performance of any given ore body is critical to resource evaluation, development decision-making, and maximising output.
A powerful iron ore classification system is a critical prerequisite. Without a sufficiently and appropriately characterised resource, modelling will be inaccurate reducing an organisations ability to implement cost reductions and productivity gains across the iron ore value chain.
Without accurate information, the cost and productivity losses that occur throughout the iron ore value chain can be detrimental to your organisation.
Chemical composition data is primarily used for resource evaluation, planning and product quality control. However, this approach fails to account for the significant business impact the textural composition of an ore body can have.
Ore bodies contain a wide range of ore textural types in differing proportions which can vary significantly even at quite local scale within the ore body.
The two sets of images above show different types of hematite and goethite at macro and microscopic scale. The different textures shown all have very distinct properties and behaviours that will impact grade, crushing, processing, bulk handling and sintering (fine ore) and blast furnace (lump ore) behaviour.
Of course, the variety of ore types is much more varied than is represented here, but can be practically subdivided on the basis of geometallurgical characteristics.
Failure to consider textural classification of iron ore can impact your organisation in the following :
With a texture-based iron ore classification, you are able to understand the porosity, physical properties, mineral proportions and mineral associations of ores, increasing the efficiency of downstream processing and allowing proactive response to changing feed type.
Textural classification and an understanding of gangue deportment also allows a more accurate prediction of the grind size required for beneficiation and maximising the energy efficiency of comminution circuits.
An example of how characterising the texture of an ore feed to a difficult-to-optimise hydrocyclone process significantly improves the predictability of the product and waste streams is shown above. Example 1 shows a simplistic visualisation of how a feed material can be characterised on the basis of higher density (black) and lower density (white) ore particles with different sizes. A more advanced classification scheme may account for association of the two minerals in a single particle (Example 2). Textural classification (Example 3) allows a more comprehensive classification of mineral association and introduces the dimension of porosity.
By measuring the textural characteristics (porosity, mineral proportions and associations) of the particles within different size fractions their density can be calculated and the probability of them reporting to the product or waste stream can be modelled. With this information grade and recovery values can be calculated for a certain set of operating parameters.
The graph below shows excellent agreement between predicted and experimental recoveries from a hydrocyclone underflow using this approach.
The variability of ore texture must be considered when determining handling and transport options. Knowing which components of the ore are associated with clay minerals or are inherently ‘sticky’ ensures that screens, chutes, transfer points and rail cars are not blocked, holding up your supply chain and creating additional costs.
The additional costs of transporting ore with unpredictable moisture carrying capacity can also be mitigated by employing effective textural classification. With this information you are able to develop blending strategies and adjust handling systems to proactively deal with these moisture issues, instead of reactively adjusting to minimise output losses.
At CSIRO, the core of our geometallurgical philosophy is founded upon the classification of iron ore on the basis of texture. Our classification methods allow organisations to understand how different ore bodies will process downstream, improving the predictability of their value chain.
Implementing a textural classification scheme will reduce your overall production and energy costs by improving decision-making; decisions which can be based on more accurate predictions of texture-based processing and handling modelling.
The scheme also create a new means of communication for the various stakeholders in your business. Currently, language used to discuss resource extraction is focused on the chemical composition. However, this does not adequately describe the likely processing behaviour that stakeholders further down the value chain can expect. By reframing the conversation around texture and downstream utilisation, you can shift the language to break down any communication barriers.
If you are interested in understanding more about the significant benefits a textural classification system contact the Carbon Steel Futures’ Geometallurgy team on +61 07 3327 4761.