Trace elements in magnetite from massive iron oxide-apatite deposits indicate a combined formation by igneous and magmatic-hydrothermal processes

Abstract

Iron oxide-apatite (IOA) deposits are an important source of iron and other elements (e.g., REE, P, U, Ag and Co) vital to modern society. However, their formation, including the namesake Kiruna-type IOA deposit (Sweden), remains controversial. Working hypotheses include a purely magmatic origin involving separation of an Fe-, P-rich, volatile-rich oxide melt from a Si-rich silicate melt, and precipitation of magnetite from an aqueous ore fluid, which is either of magmatic-hydrothermal or non-magmatic surface or metamorphic origin. In this study, we focus on the geochemistry of magnetite from the Cretaceous Kiruna-type Los Colorados IOA deposit (∼350Mt Fe) located in the northern Chilean Iron Belt. Los Colorados has experienced minimal hydrothermal alteration that commonly obscures primary features in IOA deposits. Laser ablation-inductively coupled plasma-mass spectroscopy (LA-ICP-MS) transects and electron probe micro-analyzer (EPMA) wavelength-dispersive X-ray (WDX) spectrometry mapping demonstrate distinct chemical zoning in magnetite grains, wherein cores are enriched in Ti, Al, Mn and Mg. The concentrations of these trace elements in magnetite cores are consistent with igneous magnetite crystallized from a silicate melt, whereas magnetite rims show a pronounced depletion in these elements, consistent with magnetite grown from an Fe-rich magmatic-hydrothermal aqueous fluid. Further, magnetite grains contain polycrystalline inclusions that re-homogenize at magmatic temperatures (>850°C). Smaller inclusions (<5μm) contain halite crystals indicating a saline environment during magnetite growth. The combination of these observations are consistent with a formation model for IOA deposits in northern Chile that involves crystallization of magnetite microlites from a silicate melt, nucleation of aqueous fluid bubbles on magnetite surfaces, and formation and ascent of buoyant fluid bubble-magnetite aggregates. Decompression of the fluid-magnetite aggregate during ascent along regional-scale transcurrent faults promotes continued growth of the magmatic magnetite microlites from the Fe-rich magmatic-hydrothermal fluid, which manifests in magnetite rims that have trace element abundances consistent with growth from a magmatic-hydrothermal fluid. Mass balance calculations indicate that this process can leach and transport sufficient Fe from a magmatic source to form large IOA deposits such as Los Colorados. Furthermore, published experimental data demonstrate that a saline magmatic-hydrothermal ore fluid will scavenge significant quantities of metals such as Cu and Au from a silicate melt, and when combined with solubility data for Fe, Cu and Au, it is plausible that the magmatic-hydrothermal ore fluid that continues to ascend from the IOA depositional environment can retain sufficient concentrations of these metals to form iron oxide copper–gold (IOCG) deposits at lateral and/or stratigraphically higher levels in the crust. Notably, this study provides a new discrimination diagram to identify magnetite from Kiruna-type deposits and to distinguish them from IOCG, porphyry and Fe–Ti–V/P deposits, based on low Cr (<100ppm) and high V (>500ppm) concentrations.