Improve lump/fines prediction by understanding porosity

April 16th, 2018

Keith ViningBy Keith Vining

Optimise your planning and product predictions

Environmental drivers in China have increased the attractiveness of directly feeding lump ore into the blast furnace without the need for sintering or other agglomeration.

This has helped lump ore to attract a premium with increased focus on ensuring consistent lump ore quality and improving lump/fines prediction. While the fundamental nature of a deposit can’t be changed, improved understanding of the ore textural components within a product can help in optimising product quality and consistency.

Previously, we’ve discussed the value of characterising ore by texture. A key element defining ore texture — and the focus of this month’s blog — is pore structure.

Porosity has significant effects on physical and metallurgical properties and understanding this can help you understand the true value of your lump ore.

The impacts of variations in ore porosity

Pore structure, including the level of porosity, as well as the size and spatial distribution of pores, strongly influences key ore processing characteristics.

Porosity strongly affects, among other things, the physical properties of ore and consequently affects lump ore performance during handling and in the blast furnace burden. For instance, whether the primary breakage mechanism is brittle fracture (e.g. Groups 1, 10, Table 1) or abrasion (Groups 2, 9).

Variations in pore structure can affect:

  • resistance to breakdown during transport
  • decrepitation index (DI)
  • reduction degradation index (RDI)
  • reducibility index (RI)
  • lump/fines ratio prediction.

Each ore group in Table 1 has a distinct set of characteristics measurable by industry-standard lump ore metallurgical tests.  The key to understanding these characteristics is through ore group mineralogy and pore structure.

How to evaluate pore structure

Physical characterisation of porosity is typically carried out by pycnometry, although only open porosity can be measured directly. Optical Image Analysis can be used to characterise individual lump ore particles at high (micron-scale) resolution, providing data not only on overall porosity, but size distribution and interconnectivity of pores, all of which influence processing characteristics.

An automated procedure in CSIRO’s software Mineral4/Recognition4 has been developed for this purpose, stitching together a matrix of high magnification individual images to enable analysis of fine-scale microscopic structure variation within a macroscopic particle, as well as identifying both pore structure and mineralogy.


Mineral characterisation

Figure 1 — Lump ore particle with main mineral phases identified in false colour showing the high-resolution porosity analysis possible with CSIRO’s Mineral4/Recognition4 software. Porosity is yellow. (Poliakov et al, 2017)

Visualisation of pore structure in three dimensions is also possible thanks to the increasing use of sophisticated techniques — for example, CT-scanning of ore/sinter particles.

Knowing the proportions of ore groups in an ore and the values of key metallurgical test parameters enables the prediction of processing performance. Ore group values can be determined for a specific deposit/ore type, enabling tracking of blend characteristics and variation due to changing ore group proportions; for instance, when moving from near-surface ore with a higher proportion of dehydrated and/or hydrated ore groups to the primary orebody at greater depth.

An ore classification based on metallurgical properties is a powerful tool providing a common reference frame across mines/deposits.

Ultimately, a better understanding of ore porosity enables better mine planning and scheduling decisions, improved prediction of crushing behaviour, as well as providing data which supports technical marketing.

Without an appropriately characterised resource, modelling will be less accurate, reducing an organisation’s ability to implement cost reductions and productivity gains across the iron ore value chain.

With increased understanding of porosity, you can enhance your lump/fines prediction and improve your revenue forecasting.

Partner with CSIRO for a comprehensive study into your product

At CSIRO, we have extensively developed our ore classification approach and have the ability to classify ores according to mineralogical/textural classes that take the nature of porosity into consideration.

We have further strategic research planned for the coming year to improve lump/fines prediction and are seeking interested partners to contribute to and shape our research direction.

Through our research and breakthrough science, we are uniquely positioned to tackle this challenge. We are also able to provide analytical support, training, product development and full in-house support across the iron production value chain from iron ore mining to iron-making, to help you better understand your product.

Want to know more about our strategic research or how we can help you with classification and ore porosity? Give me a call on +61 07 3327 4761 or email me directly on


Subscribe and receive the latest updates on our iron-ore and carbon steel research.


Clout, JMF, 2003.  Upgrading processes in BIF-derived iron ore deposits: implications for ore genesis and downstream mineral processing.  Applied Earth Science (Trans. IMM: Section B), 112(1), 89-95.

Poliakov, A, Donskoi, E, Hapugoda, S and Lu, L, 2017.  Optical image analysis of iron ore pellets and lumps using CSIRO software Mineral4/Recognition4.  Proceedings, Iron Ore 2017, p. 583-592 (AUSIMM, Melbourne).