A critical evaluation of copper isotopes in Precambrian Iron Formations as a paleoceanographic proxy. (1st November 2019)

• Record Type: Journal Article
• Authors: Thibon, Fanny; Blichert-Toft, Janne; Albarede, Francis; Foden, John; Tsikos, Harilaos
• Is Part Of: Geochimica et cosmochimica acta. Volume 264(2019)
• Page Start: 130
• Page End: 140
• Publication Date: 2019-11-01
• Subjects: Iron formation -- Great Oxidation Event -- Copper isotopes -- Ocean chemistry -- Transvaal Supergroup -- Hamersley Supergroup; Geochemistry -- Periodicals; Meteorites -- Periodicals; Géochimie -- Périodiques; Météorites -- Périodiques; Geochemie; Astrochemie; Electronic journals; 551.905
• DOI: 10.1016/j.gca.2019.08.020
• Languages: English
• ISSNs: 0016-7037

Abstract

Abstract: Trace metals in Iron Formations (IF) have been widely used as proxies for oceanic redox processes and oxygen evolution leading to the Great Oxidation Event (GOE). Copper has hitherto received comparatively little attention, with a single study reporting variation in δ 65 Cu between pre- and post-GOE black shales. This is attributed to postulated isotope fractionation effects in the pre-GOE ocean during widespread iron oxide deposition in IF. Here we focus on the application of Cu isotopes in two classic IF-containing sequences of the Neoarchean-Paleoproterozoic, namely the Hamersley (Australia) and Transvaal (South Africa) Supergroups. We specifically targeted the oxide-rich Joffre and Kuruman IF, the carbonate-rich Griquatown IF, and the Mn-rich Hotazel IF, which collectively record over 100 Ma of sustained IF deposition in the pre-GOE ocean. The aim was to assess the utility of Cu isotopes in IF as a paleoceanographic proxy in view of existing oxygen evolution models. Iron formation Cu concentrations are low compared to average crust and modern oceanic Fe-Mn oxide deposits, with average values between 1 and 5 ppm for all four IF data sets. Copper concentrations show no systematic variability with mineralogy, no statistical correlation with bulk Fe and Mn contents, good correlations with Ti (rCu-Ti = 0.73) for the Joffre data set and with Cr, Ni, and V for the South African data sets, and shale-like Cu/Ti ratios. Isotopic results show statistically invariant average δ 65 Cu values very close to 0‰ for all but the Joffre IF which has a marginally negative average δ 65 Cu value (−0.24 ± 0.22‰). Combined with other trace transition metal systematics, the Cu isotope data point to a Cu source that was controlled largely by inputs of fine volcanic-derived particles (ash), thus placing limitations on its utility as a paleoceanographic redox proxy during IF genesis.