| dc.description.abstracteng | Triple oxygen isotope systematics (Δ'17O) are an established tool to trace and classify extraterrestrial material. This tool is used in the field of cosmochemistry to study the genetic link between different Solar System bodies (accretion/planet formation processes) as well as the material flux in the Solar System in the past and at present. To decipher the accretion history of the Earth-Moon system it is crucial to gather information on the nature and provenance of formerly impacted bodies. Triple oxygen isotope systematics can be used to identify the chemical trace of an impacted projectile (impactor signature) in the created impact rock. This approach was so far only applied to constrain the giant impactor, which initiated the Moon formation. A few studies compared the Δ'17O of silicate Earth and Moon for this purpose with contrasting results. It is also not yet proven, if the same approach can be applied to late accreted smaller impacts on e.g. the lunar surface. The reconstruction of impactor material based on the Δ'17O is challenging. The effect of the impactor material on the Δ'17O of the produced impact rock is usually very small, hence it requires high-precision measurements. The definition of a representative Δ'17O of the Earth and the Moon is complicated by possible isotopic heterogeneities associated with different Earth mantle materials and lunar lithologies. This dissertation combines three experimental studies, which aim at (I) setting up a measurement protocol, (II) assessing the possibility of a triple oxygen isotope anomaly affecting the results of feldspar-bearing rocks, constraining the triple oxygen isotope composition of the Earth (III) and the Moon (IV) as well as evaluating if triple oxygen isotopes can be used to reconstruct the impactor material which accreted to the terrestrial mantle (V) and the lunar crust (VI). We are interested in studying the Moon-forming impactor as well as the material delivered during the late accretion (after the Moon formation). If this material was volatile-rich, then it might have been a crucial contribution to the water content of the Earth.
In the first study, we investigated the oxygen isotope fractionation in terrestrial feldspar-bearing rocks as a conceptual model of common lunar feldspar-rich lithologies (highland rocks). We characterised their petrography, chemistry and performed oxygen isotope measurements on mineral separates. We studied the equilibrium fractionation behaviour between anorthite-rich plagioclase and other co-genetic minerals. The study revealed that there is no direct indication for an oxygen isotope anomaly connected to plagioclase, which would lead to a systematically lower Δ'17O in feldspar-bearing terrestrial or lunar rocks. We evaluated different approaches to accurately analyse feldspar using the laser fluorination technique and we discussed alternative explanations for the low Δ'17O values of lunar feldspar-rich rocks as reported in the literature.
In the second study, we reassessed the triple oxygen isotope composition of the Earth’s post-Archean mantle, and the Archean mantle prior to the late accretion. We determined the Δ'17O of the post-Archean mantle with improved precision based on a comprehensive sample set of subcontinental and suboceanic lithospheric mantle peridotites: -51.6 ± 1.1 ppm (1σ, λ = 0.528, relative to San Carlos olivine: Δ'17O= -51.8 ppm. We compared this value with the Δ'17O of ultramafic rocks from the Eoarchean Itsaq Gneiss Complex (IGC) and the Mesoarchean Fiskefjord region in southwest Greenland, whose mantle source did not incorporate the full amount of Earth’s late accreted material based on ruthenium (Ru) isotope data. We found no significant Δ'17O offset between the post-Archean and the Archean pre-late accretion mantle. This restricts the late accretion contribution to the terrestrial mantle to ≤ 0.12% of Earth’s mass assuming material similar to most carbonaceous chondrites (CM, CV, CO, CK, CR, CH, CB, excluding CI), to ≤ 0.11% of Earth’s mass assuming CM-like material, or to ≤ 1.2% of Earth’s mass assuming CI-like material. A CI-like late accretion is the only option which is in agreement with a larger late accretion contribution, because CI chondrites resemble the Earth in Δ'17O.
In the third study, we reassessed the triple oxygen isotope composition of the Moon. We determined the Δ'17O of the pristine Moon with improved precision based on feldspar-rich highland rocks, mare basalts and pyroclastic glass: -51.2 ± 0.5 ppm (1σ, λ = 0.528, relative to San Carlos olivine: Δ'17O= -51.8 ppm). We found that Earth and Moon are identical within 1 ppm in their Δ'17O. We studied various lunar rock types and found no indication for the discussed lithology-dependent Δ'17O variations (e.g. no systematically lower Δ'17O values of the highland rocks). However, the Δ'17O of impact-influenced lunar rocks (soils, impact melt rocks and breccias) deviates from the pristine lunar Δ'17O. We conclude that impactor signatures can be traced using triple oxygen isotope systematics. The oxygen isotope data of Apollo 16 and Apollo 17 impact rocks is resolvable with a carbonaceous chondrite (CC, exception CI)-like or primitive chondrite-like late accretion contribution to the lunar crust.
Our studies reveal no trace of the Moon-forming impactor by means of triple oxygen isotopes, which implicates either a similar isotopic composition of the impactor and the proto-Earth, or particular impact conditions that suppressed an initial heterogeneity (e.g. high-energy impact). We found no indication for lithology-dependent Δ'17O variations in lunar rocks, which would affect this conclusion (e.g. a feldspar-specific isotope anomaly). Our findings concerning late accreted material to the terrestrial mantle and the lunar crust support the concept of a heterogeneous accretion of the Earth and a late delivery of volatile-rich material. We resolved, for the first time, impactor signatures in the triple oxygen isotope composition of lunar impact rocks. | de |