inclusion trails surround the inclusions. In Cr-pyropes, the inclusions additionally contain Al-spinel, clinopyroxene, and dolomite. The major and trace element compositions of the inclusion phases are generally consistent with the early stages of kimberlite differentiation trends. Extensive chemical exchange between the host phases and the inclusions is indicated by enrichment of the inclusions in major components of the host crystals, such as Cr 2 O 3 and Al 2 O 3 . This chemical evidence, along with phase equilibria constraints, supports the proposal that the inclusions within Cr-diopside record the decarbonation reaction: dolomitic melt + diopside → forsterite + calcite + CO 2 , yielding the observed inclusion mineralogy and producing associated (CO 2 -rich) fluid inclusions. Our study of polymineralic inclusions in megacrysts provides clear mineralogical and chemical evidence for an origin of kimberlite that involves the reaction of high-pressure dolomitic melt with diopside-bearing mantle assemblages producing a lower-pressure melt that crystallizes a calcite-dominated assemblage in the crust.

and then clinopyroxene, evolving carbonatitic melts to kimberlite and causing CO 2 exsolution that drives rapid ascent.

No abstract

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

This study evaluates the suitability of portable (handheld) X-Ray fluorescence spectrometry (pXRF) in the exploration for Aley-type ‘hard-rock’ (primary) carbonatite-hosted Nb deposits. The assessment consists of comparisons between: (1) results of pXRF analyses on selected pulp samples and results of analyses of the same pulps using traditional laboratory methods; (2) results of averaged, multiple pXRF spot field analyses performed directly on 10 to 15 cm long pieces of core (before pulverization) compared with those of traditional laboratory analyses of the same pieces of core after pulverization; and (3) results of a manual core scanning method compared with the results of conventional analytical methods of the pulps of the corresponding scanned sections. A strong correlation exists between pXRF measurements on pulps and laboratory methods for most specialty metals, such as Nb (r 2 = 0.99), La (r 2 = 0.97), Ce (r 2 = 0.67), Y (r 2 = 0.93), and P (r 2 = 0.89); however, the values of r 2 for Pr and Nd are 0.19 and 0.38, respectively. As expected, textural heterogeneities within sample intervals reduced the quality of pXRF results when multiple spot readings were taken directly on the core. Nevertheless, the data can still be used to identify carbonatite-related Nb (± other specialty metal mineralization) and delimitate potentially economically significant zones within it. The core scanning reduced the degree of variation associated with spot analyses. Scanning is useful during the early exploration stages, but provides data limited by the inability of the operator to maintain constant scanning speed. The scanning results correlate with laboratory methods for Nb (r 2 = 0.88), Th (r 2 = 0.80), Fe (r 2 = 0.84), Sr (r 2 = 0.74), Ba (r 2 = 0.73), Y (r 2 = 0.59), and Zn (r 2 = 0.75). The values of r 2 for La, Ce, Pr, and Nd were only 0.31, 0.26, 0.01 and 0.03, respectively, suggesting that concentrations of these elements were too low, and/or that the light rare earth elements (LREEs) were present not only in the crystal structure of fersmite, pyrochlore and apatite, but also in minor or accessory minerals such as REE-bearing fluorocarbonates or zircon erratically distributed throughout the core. Portable XRF is a robust tool facilitating exploration-related decision-making in the field, assuming that elements of interest such as Nb are present in concentrations within the analytical range of the instrument. The pXRF core scanning reduces the need for sample preparation (no pulps) and can be done directly on the drill-site, but the precision and accuracy of the data are reduced relative to laboratory and pXRF pulp analyses. The multiple spot analyses (no pulps) approach is good for instant verification of unknown, potentially ore-bearing minerals and for analysing discrete homogeneous features, layers, veins, etc; however, under normal circumstances this method is inferior to pulp analyses in precision and accuracy, and to scanning for determining average grade of core intervals.