3D Petrography of a Cu-Ba Mineralization by Combining μCT and μXRF

Abstract

Petrographic descriptions of geological samples is usually done by optical microscopy on thin sections or polished blocks combined with other, more advanced, techniques such as scanning electron microscopy (SEM), electron microprobe analysis (EMPA), X-ray diffraction (XRD) and X-ray fluorescence (XRF). These petrographic techniques provide detailed information about properties (e.g. the mineral phases, elemental distribution, textures, porosity …) on the surface or on a two dimensional plane of the rock sample. Although many interpolation and simulation techniques exist to generate a 3D model of the rock by combining different 2D surface analyses, exact information about the rock properties in 3D is hard to obtain, especially for complex geological materials. In this study X-ray computed micro-tomography (µCT) is used to visualize in 3D the internal structure of a complex Cu-Ba ore mineralization from the Hartz mountains (Germany). Small cylindrical samples were drilled from the ore mineralization and scanned at resolutions ranging from 18 µm to 13 µm. Because the 3D image obtained with µCT depends on the linear attenuation coefficient, which is a function of the density and the atomic number of the elements of the constituent phases, quantitative chemical identification is not available for these mineral phases. The 3D µCT data is therefore combined with µXRF mappings providing chemical information of the surface of the scanned sample. The different phases apparent in the 3D µCT data could be coupled to the chemical µXRF information, resulting in five major distinguishable mineral phases in the sample. Three of these phases (quartz SiO2 , Fe-rich sphalerite (Zn,Fe)S and barite BaSO4) could easily be separated based on the linear attenuation coefficient. The other two phases, copper-bearing minerals chalcopyrite CuFeS2 and malachite Cu2CO3(OH)2, could not immediately be separated due to the similar linear attenuation coefficients of these minerals. The mineral habit of the Cu-phases, however, was quite different: more spherical for the chalcopyrite and more vein-like for the malachite. This difference in shape and a slight difference in the mean grey value, made it possible to discriminate both mineral phases from one other and determine their 3D distribution using the in-house developed 3D analysis software Morpho+. In this study the different mineral phases in an ore sample are analysed and quantified by using X-ray µCT and XRF. The analysis of the internal distribution of the different phases in 3D is combined with an elemental surface analysis in order to obtain a 3D elemental mapping of the major mineral phases and thus allowing 3D petrography of an ore mineralization.