Simultaneous combined XRF-XRD analysis of geological sample: New methodological approach for on-site analysis on New-Caledonian Ni-rich harzburgite

International audience; There is a growing interest in on-site, real-time analytical solutions for mining and environmental projects to characterize large areas and/or volumes of raw materials that are sometimes highly heterogeneous in terms of elemental distribution and mineralogy. Several fast and cost-effective methods are used for rapid on-site screening and real-time chemical and mineralogical characterization, such as portable X-ray fluorescence (pXRF) and X-ray diffraction (pXRD). However, these methods are not always applicable due to limitations in the detection and quantification of light elements (Mg, Al, Si) for pXRF or complex or minor minerals for pXRD, whose results need to be supported by laboratory analysis.This study presents a new methodological approach for in situ rapid chemical and mineralogical characterization of samples, based on the use of a transportable instrument (called ID2B) that allows, in a single acquisition step, a combined XRD-XRF analysis to identify and quantify the chemical elements and their associated minerals. The HI0 harzburgite sample from New Caledonia used to evaluate the data was analyzed in the laboratory (SEM-EDS, EPMA, XRF and XRD) and with the ID2B instrument to highlight the potential of our new methodology. In order to demonstrate the interest of using the ID2B combined XRF-XRD analysis approach directly in the field, where sample preparation is not always easy to implement, this comparison was made on the same sample (HI0), prepared in two different ways, either as a powderized (optimal preparation) or as-sawn (unprepared) sample. After automated processing of the combined XRF-XRD datasets acquired with the ID2B instrument, the chemical elements and mineralogical phases identified on both the powder and as-sawn samples are identical to the laboratory analyses.The chemical proportions calculated from the combined XRF-XRD data sets are also close to the laboratory XRF analysis with relative errors