Correlation of computed tomography (µCT) and MLA (mineral liberation analyzer) data and investigation of its use in mineral characterization

In this thesis, it is aimed to investigate the use of µCT (Computed Micro Tomography) in mineral characterization and to compare the data obtained with conventional MLA (Mineral Liberation Analyzer) findings. In this context, feldspar, boron, lead-zinc and rare earth ore samples with industrial applications were obtained and chemical and mineralogical analysis of these samples were carried out. In the continuation of the study, three-dimensional imaging studies were carried out using µCT with the same sample sets and image analysis studies were carried out in order to use the images obtained for the purpose of the thesis. Before the comparison of the three-dimensional µCT images with MLA, the problems occurring in the images such as x-ray beam-hardening, ring artifact and x-ray reflections and their solutions are also discussed. ImageJ and CTan software were used to process the images obtained with µCT, and three-dimensional data analysis was performed after the image analysis studies performed on the raw images. The obtained data were analyzed and the existence of a significant correlation between MLA and µCT data was discussed. Within the scope of the results, a significant correlation was observed between MLA and µCT data due to the non-complex structure of feldspar ore, while no significant correlation was observed between MLA and µCT data of other ores due to image impurities and complex structures of ores. It was observed that there was a significant relationship between the densities of the minerals identified in the MLA analysis and the µCT data, and the regression method was used to determine the mineral density with µCT data at high confidence intervals.