Abstract

Kiruna-type iron oxide-apatite (IOA) deposits constitute an important source of iron and phosphorus, and potentially of rare earth elements (REE). However, the origin of IOA deposits is still a matter of debate with models that range from a purely magmatic origin by liquid immiscibility to replacement of host rocks by hydrothermal fluids from different sources. In order to better constrain the origin of Andean IOA deposits, we focused on the Cretaceous Cerro Negro Norte deposit located in the Chilean Iron Belt, northern Chile. The Cerro Negro Norte magnetite ore is hosted in andesitic rocks and is spatially and genetically associated with a diorite intrusion. Our results show that the deposit is characterized by three main mineralization/alteration episodes: an early Fe-oxide event with magnetite and actinolite followed by four stages that comprise the main hydrothermal event (hydrothermal magnetite + actinolite; calcic-sodic alteration + sulfides; quartz-tourmaline and propylitic alteration) and a minor supergene event. Based on textural and chemical characteristics, four different types of magnetite are recognized at Cerro Negro Norte: type I, represented by high-temperature (~ 500 degrees C) magnetite cores with amphibole, pyroxene, and minor Ti-Fe oxide inclusions; type II, an inclusion-free magnetite, usually surrounding type I magnetite cores; type III corresponds to an inclusion-free magnetite with chemical zoning formed under moderate temperatures; and type IV magnetite contains abundant inclusions and is related to low-temperature (~ 250 degrees C) hydrothermal veinlets. Electron probe and laser ablation ICP-MS analyses of the four magnetite types show that the incorporation of Al, Mn, Ti, and V into the magnetite structure is controlled by temperature. Vanadium and Ga concentrations are relatively constant within each magnetite type, but are statistically different among magnetite types, suggesting that both elements could be used to discriminate between magmatic and hydrothermal magnetite. However, our results show that the use of elemental discrimination diagrams should be coupled with detailed textural studies in order to identify superimposed metasomatic events and evaluate the impact of inclusions on the interpretation of microanalytical data. The presence of a distinct textural and chemical variation between magnetite types in Cerro Negro Norte is explained by a transition from high- to low-temperature magmatic-hydrothermal conditions. The microanalytical data of magnetite presented here, coupled with new delta S-34 data for pyrite (- 0.5 to + 4.3 parts per thousand) and U-Pb ages of the diorite (129.6 +/- 1.0 Ma), are indicative of a genetic connection between the diorite intrusion and the magnetite mineralization, supporting a magmatic-hydrothermal flotation model to explain the origin of Kiruna-type deposits in the Coastal Cordillera of northern Chile.