A multi-technique evaluation of hydrothermal hematite U[sbnd]Pb isotope systematics: Implications for ore deposit geochronology

Abstract

Hematite (α-Fe2O3) is a common iron-oxide mineral known to incorporate U into its crystal lattice at up to wt% concentrations and has been previously used to date ore formation within iron-oxide copper gold and banded iron formation deposits. However, there has been no detailed evaluation of the potential challenges this novel mineral geochronometer may present for accurate temporal interpretation. We report a multi-technique UPb geochronological study comprising laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS), sensitive high-resolution ion microprobe (SHRIMP) and the first isotope dilution-thermal ionization mass spectrometry (ID-TIMS) procedure and analyses using an in-situ grain micro-sampling technique. We assess intra- and inter-sample data reproducibility, and examine the compatibility of spatial and analytical resolutions to texturally and compositionally diverse samples.Five samples of hydrothermal, primary, oscillatory-zoned hematite were analysed from distinct lithologies throughout the ~1.59 Ga Olympic Dam Cu-U-Au-Ag deposit, with typical U concentrations ranging between ~20 and 1000 ppm. LA-ICP-MS mapping of isotope distributions across mm-sized grains allows pinpointing of U-rich and relatively U/Pb isotopically ‘homogeneous’ domains. Micro-domains were extracted via laser-cut square shaped trenches (100 × 100 μm). UPb datum range in complexity, with concordant age domains measured by ID-TIMS indicating that hematite can retain a near-closed UPb system over ~1.6 Ga, while also demonstrating that younger Pb-loss events may be recorded. Significant reverse discordance is common in the data, likely to be a real feature due to both internal decoupling and mobility of Pb within a single grain and UPb fractionation during microbeam analysis. These effects will obscure the interpretation of real ages for Proterozoic samples where non-zero age Pb-loss events occur, and restricts the evaluation of common Pb mixing components. As a result, microbeam analyses may be biased younger or older in terms of 207Pb/206Pb dates in samples with complex histories. Microbeam datum compare favourably with high precision ID-TIMS ages. Although reproducible ID-TIMS UPb dates were generally not obtained, in the best-case example, a 207Pb/206Pb weighted mean age of 1589.91 ± 0.91 Ma (MSWD = 1.3, n = 5), within 0.2% of the Olympic Dam granite host rock age (1593.87 ± 0.21 Ma) was generated. All five hematite samples dated by LA-ICP-MS and SHRIMP yield weighted mean 207Pb/206Pb dates within 2% of the host granite age after small degrees of data rejection. The results demonstrate that with careful sample petrography, screening and data interpretation, hematite can be considered a very useful UPb mineral geochronometer, with potential application to all U-bearing Fe-oxide rich mineral systems.