Abstract:
Tungsten belongs to the group 6 of the Periodic Table of the elements and shares numerous properties with Mo, such as a long residence time in the modern ocean and a seawater stable W isotopic composition (d186/184W value) of +0.55 ‰ controlled by adsorption of tungstate (WO42-) onto Mn-Fe oxides with a well characterized associated isotopic fractionation of +0.59 to +0.51 ‰, respectively. Tungsten exists as soluble WO42- over a very large range of Eh-pH values and as such is considered as a promising proxy for the redox state of ancient and reducing oceans. In this thesis, the d186/184W values of ancient and modern sediments as well as weathered profiles are investigated in order to test the potential of W for tracking the earliest changes in the redox state of the ocean.
Following a general introduction in chapter 1, chapter 2 of this thesis focuses on establishing the foundations of the use of W isotopes in Precambrian marine samples. First, the stable W isotopic composition of igneous rocks representative of the Precambrian continental crust (Precambrian Igneous Inventory, PII) was investigated and revealed a narrow range in
d186/184W values from -0.007 to +0.097 ‰, similar to that of modern igneous rocks (-0.010 to +0.110 ‰). Subsequently, well characterized euxinic sapropel samples deposited during the Holocene in the Black Sea revealed crustal-like W concentrations and d186/184W values, in accordance with models predicting little to no W authigenic enrichments in euxinic depositional
settings. Similarly, marine shales deposited around 2.32 Ga revealed a very narrow d186/184W range, and suggest porewater or bottom water euxinic conditions during deposition. The study of open ocean ferruginous marine shales deposited between 3.47 to 2.5 Ga showed a large spread in d186/184W values from the PII range towards heavier d186/184W values, reaching up to +0.246 ‰ at 2.5 Ga. The isotopically heavy W component in these shales was interpreted as originating from an authigenic marine W source with a heavy d186/184W value. These findings contrast to the crustal-like stable Mo isotopic compositions (d98/85Mo values) of these samples, which were interpreted as indicating low redox potential disabling Mo as a soluble oxyanion until 2.6 Ga. This study concludes that W was present in the ocean as early as 3.47 Ga and can authigenically accumulate in ferruginous ancient marine sediments.
Chapter 3 of this thesis investigated combined Mo and W isotopes in organic rich-marine shales of the Zaonega Fm (Karelian Craton, Fennoscandian Shield), deposited during the Shunga Event ( 2.0 Ga), representing the termination of the Great Oxidation Event (GOE). Many studies previously interpreted the chemical signals of the Zaonega Fm. as evidence for a global collapse in molecular dioxygen, or changes in the depositional settings. Using new Mo and W isotopes data, we reveal opposed d98/85Mo and d186/184W trends associated with increasing concentrations in V, Mo and W upwards in the sedimentary sequence. The evolution in these chemical signals in the Zaonega Fm. are related to the opening of the Onega paleobasin to the open ocean, inducing changes in depositional environments from restricted-euxinic to an oxygen minimum zone-like environment. The d98/85Mo and d186/184W values of these sediments are characteristic of the isotopic fractionation measured for Mo and W adsorption onto Feoxides, indicating that the cycling of these elements was controlled by Fe-oxides, rather than Mn-oxides. This observation implies that the termination of the GOE did not enable Mn-oxides persistence in the global ocean.
Chapter 4 focuses on W concentrations and d186/184W values in two weathered profiles of the 2.77 Ga Mt Roe Basalt (Pilbara Craton, Australia). These two profiles were weathered under an anoxic atmosphere resulting in W loss. The association between d186/184W values and plagioclase alteration proxies revealed that isotopically light W was likely retained in weathering products such as clays and aluminum oxides, while the residual dissolved W was removed from the profile with meteoritic waters. The d186/184W and associated loss of W values follow an equilibrium isotopic fractionation in a closed system with an isotopic fractionation factor e186/184Wdissolved-adsorbed of +0.126 ‰. This study reveals that W, unlike U and Mo, was very efficiently removed from exposed surfaces before the atmosphere became oxidized. The riverine inputs of W to the oceans likely drove the ocean W isotopic composition towards heavier values, and likely can explain the earliest fractionated d186/184W values from the 3.47 Ga to 2.77 Ga marine shales reported in the chapter 2.
Chapter 5 reports method development for W isotopes determination in carbonates. First, the trace element concentrations of four carbonate reference materials leached with 3 M HCl, 5 %v HNO3, 5 M acetic acid were compared with the aim of selecting the most efficient leaching method for high authigenic W concentrations in the leachates. Then, the efficiency of the equilibration of sample and double spike-sourced W was investigated in both HNO3 and HCl leachates. Based on W concentrations and d186/184W values of the leachates, the method of choice for chemical separation of W from carbonate matrix involve a HCl leaching step, followed by double spiking in the same media before column chemistry, in order to reduce the use of different acids before column chemistry. This chapter reports d186/184W values of carbonates from various depositional environments, from modern to 2.39 Ga, and reveal the potential of W isotopes in open marine carbonates as a proxy for the evolution of the ocean's redox state through time.
In conclusion, this study confirms that W stable isotopes in ancient ferruginous marine sediments has very high potential to fill the knowledge gap of the evolution of the redox states of the Paleoarchean to Paleoproterozoic oceans. The W isotopes proxy is based on a well constrained detrital background (PII d186/184W values), on early W mobility, either from anoxic weathering (d186/184W values up to +0.174 ‰) or from hydrothermal fluids (source-like d186/184W values), and on large isotopic fractionation during WO42- adsorption onto Fe-oxides (+0.51 ‰) resulting in isotopically heavy residual WO42- . As a result, this thesis proposes a three step increase in redox potential during the Archean: (1) at 3.47 Ga, the low W isotopic
composition (<0.132 ‰) of marine shales reveal that the W inputs to the ocean were mostly controlled by hydrothermal fluids leaching igneous W. (2) Then, from 2.94 to 2.77 Ga, marine shales reveal d186/184W values reaching up to +0.177 ‰, interpreted as resulting from the rise of continental masses and associated increasing inputs of W from anoxic subaerial weathering, rather than W cycling onto oxides in proximal-shallow environments. (3) Finally, carbonates and ferruginous shales deposited from 2.5 to 2.0 Ga recorded increasing d186/184W values reaching up to modern-like seawater, indicating a global increase of redox potential with intensive Fe-(Mn)-oxides cycling.
Print-on-Demand