THE CONCENTRATION OF HYDROGEN IN INCOMPLETELY AND WHOLLY MELTED TERRESTRIAL BUILDING BLOCKS

Abstract

Hydrogen (H) is the most abundant element in our solar system and exerts a primary control on the habitability, and geochemical and geodynamic evolution of rocky bodies. Therefore, constraining the source(s), timing of accretion, and abundance of H in the Earth and other bodies is of fundamental importance for understanding how planets evolve. Direct constraints on the source(s) of H and other highly volatile elements (HVEs; e.g., H, C, F, Cl, and S) to the bulk Earth can be provided by analyzing meteorites, which are the remnants of early-formed rocky bodies that were present during the accretion of the terrestrial planets. Such samples either directly sample or provide analogs for terrestrial precursor materials.Rocky solar system materials can be subdivided based upon their nucleosynthetic isotopic compositions (“genetic” tracers; e.g., 50Ti, 54Cr) into two groups, which are thought to correspond to the inner- and outer- solar system. Materials may be further subdivided by their extent of thermal processing (i.e., unmelted, incompletely melted, and wholly melted). Earths H budget is commonly accounted for by addition of unmelted (i.e., chondritic) materials, namely carbonaceous chondrite-like (CC-like) materials, thought to be derived from the outer solar system, which have high H concentrations (up to ~14 wt. % H2O; total H as H2O equivalents) and similar H isotopic compositions to the bulk Earth. Furthermore, chondrites derived from the inner solar system (e.g., ordinary and enstatite chondrites) are H-poor relative to carbonaceous chondrites. Similarly, all melted planetesimals are commonly assumed to be anhydrous. However, recent analyses of enstatite chondrites (ECs), which are formed in the inner solar system and are the closest match to the nucleosynthetic isotopic composition of the bulk Earth, suggest that ECs have a similar H isotopic composition to the bulk Earth and can account for its entire H budget. Furthermore, recent analyses suggest that melted (i.e., achondritic) bodies may retain considerable amounts of H, potentially enough to account for Earth’s H budget in the case of the enstatite achondrites (i.e., aubrites). However, achondritic materials are predominantly highly H-poor relative to chondritic materials, and it is unclear if the aubrites are an anomaly, and at which stage of planetesimal evolution H and other HVEs are lost. In chapter 2, I re-examine prior bulk analyses of H in aubrites, and by extension ECs, using in situ methods and suggest that nearly all H measured in aubrites by bulk methods reflects pervasive terrestrial contamination and alteration, a result which may extend to concurrent bulk H analyses of ECs. In chapters 3 and 4, I examine the H content of incompletely melted (i.e., primitive achondritic) planetesimals to constrain at what stage of planetesimal evolution H is lost. Chapter 3 characterizes the H contents of the ureilites, a group of C-rich primitive achondrites, and chapter 4 characterizes the H contents of the acapulcoite-lodranite clan which represents the “prototypical” primitive achondritic parent body. I find that primitive achondritic parent bodies are highly H-depleted relative to chondrites, requiring that H is efficiently lost prior to or at the onset of planetesimal melting, and that Earth’s H budget is accounted for by accretion of thermally primitive materials (e.g., chondrites). Within my primitive achondrite data sets, I observe apparent disequilibrium with respect to H between olivine, pyroxene, and feldspar. In chapter 5, I explore whether this apparent disequilibrium is the result of extrapolating high pressure experimental data to low pressures. I conduct olivine–melt H partitioning experiments at low pressures (10 – 200 MPa) and find that the olivine-melt H partition coefficient increases at low pressures, contrary to extrapolation from high pressure data. This observation is best explained by a control of H speciation in the melt on the partitioning of H between olivine and melt.