256 MINERAL DEPOSIT RESEARCH FOR A HIGH-TECH WORLD   ▪   12th SGA Biennial Meeting 2013. Proceedings, Volume 1 KEYNOTE SPEAKER Abstract. In order to use trace element composition of magnetite as an indication of its origin it is necessary to understand the processes that control the trace element concentrations in magnetite. We have characterised trace element distribution in magnetite, using laser ablation ICP-MS, from magmatic ore deposits (Fe-Ti-V-P and Ni- Cu-PGE) where the paragenetic sequences are well constrained. Changes in composition of the liquid, driven by crystal fractionation, are recorded by magnetite in both silicate and sulfide melts. The composition of magnetite is sensitive to co-crystallizing phases, with marked depletion in Ti when ilmenite crystallizes before magnetite and in Cu when a sulfide liquid segregates. Multi-element variation diagrams show that magmatic magnetites have chemical signatures distinct from hydrothermal magnetite due to differences in fluid composition and different conditions of formation (e.g., competing phases, redox and temperature). Chemical fingerprinting of magnetite from the magnetite ‘lava flows’ of El Laco, northern Chile, provides new evidence to support the hydrothermal alteration model rather than a magmatic origin. Keywords. Magnetite, trace elements, laser ablation- ICP-MS, provenance, El Laco 1 Introduction Magnetite forms under a wide variety of conditions, crystallizing at high temperature from silicate, sulfide and carbonatite magmas or it can precipitate at lower temperatures from hydrothermal fluids. These different conditions may lead to distinctive trace element signatures for the magnetite. Recent analytical developments make it possible to determine a much wider range of trace elements than previously. Thus it may become possible to use the trace element signature of magnetite as petrogenetic and provenance indicators. One problem is how to establish which trace element variations are significant. Modern laser ablation-ICP-MS systems provide results for ~ 20-25 elements. Some studies have been carried out using a statistical approach (e.g., Dupuis and Beaudoin 2011; Nadoll et al. 2012). Other studies have concentrated on petrogenesis (e.g., Reguir et al. 2008; Dare et al. 2012). Our current studies consider well characterized samples from different settings to establish which elements are the most diagnostic and which processes have the most effect. These studies could then be applied to less well understood examples or provenance studies. For example, we show that the difference observed in the behaviour of trace elements between magnetite of magmatic and hydrothermal origin permits a re- evaluation of the nature of the fluid (Fe-rich melt or hydrothermal) involved in forming the enigmatic magnetite ‘lava flows’ of El Laco, northern Chile (Nyström and Henriquez 1994; Sillitoe and Burrows 2000). 2 Methodology To characterise magnetite formed from high temperature silicate melts we have analysed magnetite in Fe-Ti-V-P deposits from both layered intrusions, such as the Bushveld (South Africa) and Sept-Iles (Quebec, Canada) Complexes, and massif-type anorthosites, such as Lac St. Jean (Quebec, Canada). Magnetite from Sudbury and Voisey’s Bay Ni-Cu-Platinum-Group-Element (PGE) sulfide deposits (Canada) represent magnetite formed from sulfide liquids. Samples from hydrothermal and low temperature environments include examples of Fe- oxide-Copper-Gold (IOCG) from Ernest Henry (Australia), porphyry-Cu from Morococha (Peru) and banded iron formation from Thompson Ni Belt (Manitoba, Canada). Samples from the El Laco area represent our test case. The LA-ICP-MS system used at LabMaTer, UQAC, is a Resonetics M-50 193nm laser coupled with an Agilent ICP-MS. An international reference material (GSE-1g) was used for calibration and Fe used as the internal standard. A beam size of 55 – 80 μm was used so that any fine-grained exsolution lamellae (e.g., ilmenite, spinel) were incorporated into the analysis to better represent the initial composition of the Fe-Ti oxide (Dare et al. 2012). The data are presented in order of increasing compatibility with magnetite and normalized to average crust (Fig. 1). We find that these types of diagrams help to investigate the behaviour of trace elements during crystallization of magnetite from different melt compositions (silicate and sulfide) and anomalies due to the competition for an element among co-crystallizing phases (e.g., magnetite, ilmenite and sulfide) are more readily observed. 3 Magnetite records changing melt composition The concentration of an element in magnetite depends The use of trace elements in Fe-oxides as provenance and petrogenetic indicators in magmatic and hydrothermal environments Sarah A.S. Dare, Sarah-Jane Barnes, Julien Méric, Alexandre Néron Université du Québec à Chicoutimi (UQAC), Chicoutimi, Québec, Canada, G7H 2B1 Georges Beaudoin, Emilie Boutroy Université Laval, Québec, Québec, Canada, G1V 0A6