Pioneering work to 3D image the inside of rocks will transform the mining industry by removing the guesswork from mineral exploration and processing. JOHN MILLER reports
Article from resourceful: Issue 12
Almost as good as rocks having the ability to talk and tell us whether or not they host valuable minerals, could be x-ray-like vision to see what’s inside.
A 3D imaging project, known as “digital rock”, led by CSIRO senior research scientist, Belinda Godel, aims to do just that and is gaining the attention of mining companies and universities in Australia and abroad.
Dr Godel says digital rock technology has been used in oil and gas for 20 years, but its use in minerals is limited due mostly to the variability and complexity of the mineral systems involved, with each commodity presenting unique challenges.
'Cat' scans to reveal the characteristics inside rocks
Dr Godel's team is developing a world-leading methodology that uses high-resolution x-ray computed tomography (MicroCT) combined with image analysis to create a 3D image from the inside of rocks and compute characteristics without the need for extensive sample preparation or slicing of the rock.
MicroCT works similarly to medical CT (Cat) scans, but more powerful x-rays are used to better penetrate complex and dense materials containing valuable resources. For the past five years, CSIRO has been developing its application for the minerals industry across the value chain.
"The MicroCT technology forms an important component of the work, but the strength in the CSIRO methodology relies on its integration with other technologies to provide an accurate 3D characterisation," Dr Godel says.
These technologies include optical microscopy, scanning electron microscopy, x-ray fluorescence mapping, laser-ablation inductively couple mass spectrometry (LA-ICP-MS) and electron back-scattered diffraction analysis.
"With many variables, one of the challenges is setting up the instruments to obtain the required data and another is processing the data so that the numerical results provided can be used by customers to address their problems. Our development philosophy is whatever the parameters required; we will find a way to calculate them," Dr Godel says.
Mineral analysis in 3D
Traditionally, mineral characterisation has been two-dimensional – the geologist would gather the sample, slice it, put it under the microscope or another technique to observe it in 2D and sometimes quantify it.
"MicroCT introduces the third dimension, eliminating the assumptions made looking at 2D by producing “real” statistics. It provides the entire picture – such as shape, size, real association of minerals or pores and their degree of connectivity," Dr Godel says.
3D data of mineral relationships is a significant improvement over traditional 2D microscopic analysis.
CSIRO's digital rocks program initially focused on nickel and platinum group metals.
"With platinum group metals the technology was developed to determine where the platinum was in the rocks, how much was there, how small it was, was it associated with particular mineral sulphide, silicate minerals, or chromium oxide. The results had implications for ore genesis but also for mineral processing.
"Then, we looked at using MicroCT to infer physical processes leading to the formation of nickel sulphide deposits looking at sulphide distribution, size and morphology, and their degree of connectivity in the rock.
"Knowing the sulphide characteristics, we then analysed the palladium content using LA-ICP-MS and merged all datasets to help us understand how the deposit formed. This methodology can potentially be used to target high grade ores.
"We have since applied the concept to other commodities such as graphite, copper and gold, and owing to iron ore industry interest, are now focusing mainly on this area. We are looking at a range of iron ore products across a range of scales – full exploration cores, lumps, fines, pellets or sinters."
CSIRO research group leader, Keith Vining, says that for the first time porosity can be measured with some degree of accuracy, with benefits for iron ore processing and metal production.
"Understanding porosity is important for predicting downstream processing behaviour – from crushing and handling to blast furnace performance, and the strength and reactivity of agglomerates put into the furnace," Dr Vining says.
"Previously we could only look at things in 2D. Essentially, you had to cut the rock or sinter and try to measure porosity looking at one face, which is time-consuming and doesn’t provide an accurate picture. Now we can identify the open pores from the closed pores, which is important to iron ore's behaviour in the iron making process.
"We can understand what products are doing in processing and get a feel for what a product's value is to the consumer.
"For iron ore we are trying to understand how quickly is it melting; which parts aren’t melting; which parts are remaining stable; how quickly is the pore structure developing; and how it is behaving in the blast furnace."
Improving productivity and processing
Dr Godel says the mining industry is becoming interested in the work because it has to find new and better ways of doing things to improve productivity, as deposits get harder to find and more complicated to process.
"Our work is trying to put new context into exploration and processing – the 'Holy Grail' would be the ability to use data obtained during exploration to predict how the ore will behave downstream during processing. It is where geology meets data analytics and large-scale simulation.
"The speed of MicroCT processing has improved drastically over the past few years, but the most important aspect is the quality of the data. Even if we can access the fastest computer and the smartest algorithms on Earth, if we feed in poor-quality data, we will most probably end up with a poor prediction."