Calibrated X-ray micro-tomography for mineral ore quantification

• Published in: Minerals engineering Vol. 110; pp. 122 - 130
• Main Authors: Reyes, F., Lin, Q., Udoudo, O., Dodds, C., Lee, P.D., Neethling, S.J.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.08.2017
• Subjects: Mineral characterisation; Mineral liberation analysis; X-ray micro-CT; Mineral liberation analysis; X-ray micro-CT; Mineral characterisation
• ISSN: 0892-6875; 1872-9444
• DOI: 10.1016/j.mineng.2017.04.015

•Calibration of X-ray micro tomography data based on SEM/EDS.•Able to distinguish Pyrite and Chalcopyrite in XMT images.•Production of simplified 3D mineral maps with known confidence bounds. Scanning Electron Microscopy (SEM) based assessments are the most widely used and trusted imaging technique for mineral ore quantification. X-ray micro tomography (XMT) is a more recent addition to the mineralogy toolbox, but with the potential to extend the measurement capabilities into the three dimensional (3D) assessment of properties such as mineral liberation, grain size and textural characteristics. In addition, unlike SEM based assessments which require the samples to be sectioned, XMT is non-invasive and non-destructive. The disadvantage of XMT, is that the mineralogy must be inferred from the X-ray attenuation measurements, which can make it hard to distinguish from one another, whereas SEM when coupled with Energy-Dispersive X-ray Spectroscopy (EDX) provides elemental compositions and thus a more direct method for distinguishing different minerals. A new methodology that combines both methods at the mineral grain level is presented. The rock particles used to test the method were initially imaged in 3D using XMT followed by sectioning and the 2D imaging of the slices using SEM-EDX. An algorithm was developed that allowed the mineral grains in the 2D slice to be matched with their 3D equivalents in the XMT based images. As the mineralogy of the grains from the SEM images can be matched to a range of X-ray attenuations, this allows minerals which have similar attenuations to one another to be distinguished, with the level of uncertainty in the classification quantified. In addition, the methodology allowed for the estimation of the level of uncertainty in the quantification of grain size by XMT, the assessment of stereological effects in SEM 2D images and ultimately obtaining a simplified 3D mineral map from low energy XMT images. Copper sulphide ore fragments, with chalcopyrite and pyrite as the main sulphide minerals, were used to demonstrate the effectiveness of this procedure.