Characterisation of samples of ore particles using x-ray micro-tomography

The degree of mineral liberation is important for the efficiency of subsequent physical separation processes such as froth flotation. Mineral liberation studies involve determining the volumetric abundance or volumetric grade distribution of a specific mineralogical phase in a particular mineral. Currently, methodologies for assessing mineral liberation are laborious regarding sample preparation, analysis time (from weeks to months), and the need for stereological correction. These constraints can be eliminated by using X-ray CT which gives the cross-sections directly from three-dimensional data in shorter time (from ten minutes to hours) with minimal sample preparation. X-ray computed tomography (CT) is a non-destructive technique which allows three-dimensional visualisation of inner structures of an object based on the variations in density and atomic composition. Initially, it was developed as a medical tool for imaging soft tissue and bone. During the last decade, the number of X-ray CT applications in engineering and geology has steadily increased, with the improvements in performance and imaging capabilities. The aim of the present work is to apply X-ray CT technique for finely divided ore samples and to study the relationship between mineral liberation and CT results. Four different ore types were used in this study: Northparkes ore (Australia), Ernest Henry ore (Australia), Keetac ore (USA) and Cannington ore (Australia). Different settings of the desktop X-ray CT technique were applied for each particular ore sample for several ore liberation (particle size distribution) properties. Two dimensional CT images were reconstructed from the three-dimensional X-ray CT data. It was found that the settings for CT technique were a function of the ore type. Particularly in the case of Cannington (high density ore) the best setting conditions split from the rest of the ores tested. The appearance of different artifacts occurring during the analysis were studied and kept to the minimum. A functionality between mineral liberation and CT results was found. The variables affecting the most the results were the Voltage and Minimum Intensity Percentage. Contrary to the expected trends, variables having a negligible effect on the results were found to be exposure time / equivalent Al filter thickness