Stellenbosch University (SU) will be the first university in South Africa to house a hyperspectral core imaging system, thanks to a donation from mining company AngloGold Ashanti.

The core scanner will be set up and operationalised under the auspices of the African Rainbow Minerals Geometallurgy research chair, held jointly by Professor Bjorn von der Heyden in the Department of Earth Sciences, and Dr Margreth Tadie from the Department of Chemical Engineering.

Von der Heyden says the university will use the scanner for research and student training.

The donation is viewed as especially valuable as systems such as these carry a steep price tag and are, therefore, usually found only at mining companies and consulting firms.

From a research perspective, there is excitement about the continuous nature of mineralogical data that the core scanner provides.

“The machine can scan up to 200 m of rock core on a daily basis, providing short-wave infrared hyperspectral data (1 000 to 2 500 nm spectral range) and red, green and blue (RGB) colour data at a pixel resolution of 0.1 to 2 mm, depending on the selected field of view,” explains Von der Heyden.

The resulting spectra provide insight into the relative abundances of minerals such as amphibole, tourmaline, muscovite and gibbsite, among others.

Several of these minerals are regarded as alteration minerals, which are important exploration vectors that point geology teams towards the mineralisation of critical mineral resource endowments.

Von der Heyden says the data can be used in exploration geology and soil science, and the research chair will be placing emphasis on developing workflows for use in geometallurgy.

Because the large datasets are amenable to the application of machine learning algorithms, the ARM research chair is also advancing its relationship with the School for Data Science at SU.

“Because of the scales at which the hyperspectral core scanner operates, it is a great complement to our existing geometallurgy toolkit, which comprises of micro-analytical characterisation, bulk sample empirical testwork and spatial- and process modelling,” notes Tadie.

“It is an incredibly robust machine that offers opportunities for students to work at remote field sites.

“These students will be trained to conduct multidimensional systems-type thinking, and will need to think of the mineralogy, chemistry and beneficiation response of the rocks in terms of their diagnostic spectral signatures, which change over the spatial scales evaluated.”