Chlorine isotope fractionation recorded in atacamite during supergene copper oxidation

Abstract

In the Atacama Desert of northern Chile, large amounts of the copper hydroxy-chloride mineral atacamite (Cu2Cl(OH)3) are formed in the supergene oxidation zone of Cu deposits. Since atacamite requires saline water to form and is commonly preserved under hyperarid conditions, it has been proposed as a climate-sensitive mineral marker that can also provide relevant geochemical information regarding halogen (in particular chlorine) fluid sources during supergene Cu oxidation. However, chlorine stable isotope data for atacamite in Cu deposits are scarce and no experimental data for chlorine isotope fractionation between atacamite and water are currently available that could provide constraints on the possible mechanisms of fractionation. In this study we report δ37Cl values of atacamite along a thick (~100 m) and well-developed supergene enrichment profile at the Barreal Seco iron oxide-copper-gold (IOCG) deposit in the Atacama Desert. The δ37Cl values of atacamite along this profile range from +2.1‰ to −0.6‰ (mean = +0.4‰ ± 0.7‰; median = +0.3‰), and show a distinct trend with depth, characterized by higher δ37Cl values towards the top of the profile, with lower values at the bottom. In addition, the chlorine isotope compositions of experimentally synthetized atacamite crystals and coexisting CuCl2 solutions were determined at room temperature (23 °C). The per mil fractionation factor (Δ37Clatacamite-Cl) was determined at +0.75‰, showing that the heavy isotope (37Cl) is preferentially incorporated into atacamite. We used numerical models to test two possible scenarios that might explain the observed δ37Cl profile at Barreal Seco. A first scenario involves lowering of the water table with concomitant precipitation of atacamite, which removes Cl from solution following a Rayleigh fractionation process. This model qualitatively fits our observations but does not explain the full magnitude of the observed δ37Cl shift along the supergene profile. In contrast, a model that simulates Cl- diffusion following the upward injection of a deep brine that overlies fresh groundwater reproduces well the observed δ37Cl data, suggesting that seismic pumping of basinal brines, followed by Cl diffusion, is a feasible mechanism to explain the formation of atacamite at Barreal Seco. These data are a first step towards interpreting δ37Cl profiles of atacamite in Cu deposits in the Atacama Desert and exploring the potential use of stable chlorine isotopes to monitor Cu weathering and enrichment in supergene environments.