Concentration of cobalt in IOCG systems: Exemplified by the Proterozoic Kangdian IOCG metallogenic province, SW China

Cobalt was defined as an important strategic metal by major industrial countries. The clan of iron oxide copper–gold (IOCG) deposits has been documented to be a potentially important Co provider (predicted to be accounting for > 10 % of continental Co resources), but the mechanism for Co concentration in IOCG systems is still poorly constrained. In this study, we conducted a combined geological, mineralogical and geochemical investigation on the Co enrichments/mineralization in the Lala, Dahongshan and Yinachang deposits from the Proterozoic Kangdian IOCG metallogenic province, SW China (with an estimated total Co resource of > 70,000 t). These deposits have a similar paragenesis consisting of early Fe-oxides and late Cu-sulfides mineralization stages. This study revealed that Co in these deposits is mainly hosted in pyrite, with the way by substituting for Fe²⁺. The Co-rich pyrite grains in the Lala deposit have formed in two generations. The early generation (Py I; 5432 to 27,883 ppm Co), clearly predating the Fe-Cu mineralization, is characterized by massive pyrite bands concordant with the hosting strata, and thus were suggested to be syn-sedimentary in origin. These grains were commonly overprinted by late Fe-Cu mineralization through a fluid-aided dissolution-reprecipitation process during which the primary Co in Py I were leached. The second generation of Co-rich pyrite (Py II; 2152 to 8329 ppm Co) was closely associated with Cu-sulfides of the Cu mineralization. On the other hand, only one generation of Co-rich pyrite was present in the Dahongshan and Yinachang deposits, and are synchronous with the Cu mineralization.

Chemical and S isotopic compositions show that all the Co-rich pyrite grains from the Lala (both Py I and Py II), Yinachang and Dahongshan deposits have a notable S isotopic peak value around at ∼ 3 ‰ with similar high Co/Ni (mostly > 1), but low Mo/Ni (mostly < 0.001) ratios, indicating strong affinities of mafic provenances for the Co metals. The Co in the sedimentary Py I grains of Lala was constrained to be likely related to the coeval mafic volcanic activities (as potential Co sources). On the other hand, the Co of the Cu stage in the Lala (i.e. Py II), Dahongshan, and Yinachang deposits were suggested to be sourced from coeval mafic magmas or hosting mafic rocks through fluid-rock interaction. However, the former source could be insignificant, as the potential Co contents of the mafic magma-derived fluids were estimated to be lower than 12.0 ppm. In contrast, our mass balance calculation strongly confirmed that large amounts of Co in the mafic hosting rocks were mobilized or leached during fluid-rock interaction. Considering that such kind of Fe or Cu mineralizing fluids are comparable in various IOCG deposits (e.g., high salinity, NaCl-rich), we further proposed that the Co budgets of hosting rocks would be the first key factor controlling variable Co enrichments/mineralization in IOCG systems. This study highlights that Co-rich hosting/country rocks (e.g., ultramafic–mafic rocks or Co-rich black shales) presenting in specific IOCG deposits could be potentially important indicators for Co enrichments/mineralization in the IOCG systems.