Search

The aim of this thesis is to understand the chemical behaviour of halogens in silicate melts by experimentally determining their thermodynamic activities. This includes defining how halogens become incorporated into a melt and what influences chlorine's solubility and bonding behaviour in silicate melts. A halogen buffer has been developed using Ag/AgCl, in which Ag (as oxide) is virtually insoluble in silicate melt under the conditions of the experiment. The buffer controls the fugacity of chlorine. Experiments were performed at 5-20 kbar and 1200-1500 °C in a piston-cylinder apparatus. The effect of oxygen fugacity on Cl solubility was determined using Re-ReO2 and C-CO2 oxygen buffers. Experiments show (1) that chlorine solubility in haplobasalt at 15 kbar/1400 °C can reach 5 wt%, even at Cl2 fugacities as low as 0.0035 bar. 2) Cl concentration increases linearly with the square root of chlorine fugacity at fixed f(O2), obeying Henry's Law up to ~3% Cl in the melt of a haplobasalt. 3) Cl solubility decreases with the fourth root of oxygen fugacity at fixed √f(Cl2). This implies that Cl- replaces O2- in the silicate framework as follows: Cl2(gas)+[O2-](melt) = 2[Cl-](melt)+ ½O2(gas) [I] This relationship is used to correct for the effect of oxygen and chlorine fugacity on pressure and temperature, to determine more accurately how chlorine solubility changes with these intensive variables. Results show that Cl increases with temperature and decreases with pressure at fixed conditions. The chlorine buffering method was then applied to 44 chemically different compositions, all performed at 1400 °C, 15 kbar, and a fixed oxygen and chlorine fugacity. Results show that chlorine solubility increases with Ca and Fe and decreases with Si content. Knowing the compositional effects at fixed conditions allowed for the quantification of a chloride capacity (CCl), defined as: CCl=Cl(wt%)√f(Cl2)x√f(O2)4[ii]. The chlorine capacity allows for the comparison of literature data (including those performed with water) if the temperature, pressure, oxygen fugacity and chemical composition are known. Literature data were combined with the 44 experiments performed across a wide compositional range and the pressure and temperature series of experiments. Cl content in wt% can then be expressed as a function of the different variables as: logCCl=0.571+(4873XCa-3803XSi+2724XFe+1891-943P)/T [iii]. Water was found to be insignificant in controlling chlorine solubility at concentrations less than 4.3 wt%. [iii] can then be used to determine the conditions that HCl would degas. Results indicate that, at fixed Cl2 fugacity, pressure, and temperature most HCl loss occurs at pressures below 2 MPa. Additionally, the fraction lost as the surface is approached varies dramatically with composition as rhyolite > dacite > andesite > basalt. Constraining the thermodynamic activities also allows for determining the conditions under which Cl-rich minerals such as sodalite would form and when brines would begin to segregate from the melt. The bonding behaviour of chlorine in silicate melts was investigated by X-ray absorption fine structure. Results indicate that Cl predominantly bonds with 2+ cations (Ca, Mg and Fe) but can also covalently bond with Si in melts that exhibit low 2+ cation contents.

The Earth and the Moon are observed to be almost indistinguishable over a range of refractory isotopes systems. This thesis aims to test the hypothesis that the lunar magma ocean (LMO) possessed the same elemental composition as that of the terrestrial mantle. In particular, I explore whether the LMO's differentiation would result in liquid and crystal phases that can reproduce the Apollo observations of the lunar surface. To achieve this, I have conducted a detailed series of experiments which track the evolution of a hypothetical lunar magma ocean of terrestrial mantle composition. I have broken the differentiation of the LMO into discrete steps, with the bulk composition of the next step taken as the melt composition of the previous step. The final experimental liquid, representing ∼98% crystallisation of the LMO, is highly enriched in Fe (>30 wt.%) and Ti (∼3 wt.%). To explore the petrogenesis of the mare basalts and the Ferroan Anorthosite Suite (FAS), I have simulated the remelting of the lunar cumulate mantle using the MELTS program. By recombining experimental phases and melts in different proportions, I am able to prove that the Apollo observations are compatible with a LMO of terrestrial mantle composition. The mafic crystals present in the ferroan anorthosite suite are best matched to experimentally produced crystals rather than melt, indicating inefficient flotation rather than trapped melt has contributed to the ferroan anorthosite suite. Mineral/melt Rare Earth element (REE) partitioning occurring throughout the evolution of the lunar magma ocean has been experimentally determined, and predict heavy REE depletion of final liquids, due to high-temperature pyroxene crystallisation. Consequently, the 'bow shaped' trend witnessed in the mare basalts results from a lunar mantle possessing pyroxene enriched in heavy REE and a light REE enriched KREEP basalt. I explore the implications of different compositions on the lunar moment of inertia, finding that an overturn of the lunar mantle is more significant than any conceivable change in composition.

The loss rates of volatile elements from silicate melts are of considerable interest to a number of branches of volcanology, petrology, and geochemistry. In the volcanic context emanation coefficients are used as empirical measures of the extent of loss of individual elements in the gas phase. In geochemistry and cosmochemistry relative loss rates control the pattern of element depletions during melting and core segregation on asteroids and proto planets. Incorporation of such differentiated, pre-melted bodies probably influenced the well-known pattern of volatile element depletion in the Silicate Earth. This study makes two independent measurements of the volatilities of the elements Ag, Bi, Cd, Cr, Cu, Ga, Ge, In, Mo, Pb, Sn, Tl, V, W and Zn, in molten silicate. The first includes a series of high pressure (1.5GPa) metal-silicate partitioning experiments used to determine the activities of trace elements in silicate melts. The second involves direct measurement of volatile loss from a representative basaltic melt at oxygen fugacities applicable to planetary formation. Results from this study show that some elements (e.g. indium) are less volatile than previously thought, while others (e.g. copper) are more so. Applying these results to volatile depletion in the Silicate Earth reconciles the supposed overabundance of indium, as suggested by condensation temperatures. Comparing depletion of the Silicate Earth relative to CI meteorites shows a remarkably consistent trend with the measured volatility factors, suggesting that evaporative volatile loss occurred during planetary formation, most likely from planetesimals. The volatile loss data from this study also support the observation of volatile depletion in lunar basalts. Such agreement provides additional evidence that the Moon experienced some degree of volatile loss during its accretion after the giant impact by Theia.

A series of experiments related to igneous processes within Mars are presented, with a particular emphasis on the behaviour of sulfur during Martian magmatism. Of most significance are a series of one-atmosphere experiments conducted on a broad range of silicate melt compositions, under oxygen fugacity (fO2) conditions ranging from FMQ-1.67 to FMQ+1.6: a range which brackets much of the variation observed in Terrestrial and Martian igneous rocks. Quenched glasses from these experiments have been analysed by X-ray absorption Near Edge Spectroscopy (XANES) and Secondary Ion Mass Spectrometry (SIMS) to determine the speciation of their dissolved sulfur, as well as its overall concentration. The XANES spectra collected indicate the presence of the sulfide and sulfate species, and the progressive replacement of the former by the latter as fO2 increases. Clear differences are observed between the S6+/ΣS ratios of different silicate melts equilibrated under the same fO2 conditions, demonstrating that the major-element composition of a silicate melt has some measureable influence over sulfur speciation. The component most closely correlated with this change is FeO, whose presence in the silicate melt shifts the exchange between sulfide and sulfate towards higher fO2 conditions. This is interpreted as an outcome of coupling between sulfur and iron in the melts, and on this basis a quantitative relationship between log S6+/ΣS and log (Fe2O3/FeO2) is proposed. Such a relationship potentially provides an alternative to direct measurement of S6+/ΣS, but warrants further study to establish its generality. SIMS analyses of the most oxidized melts indicate a strong inverse correlation between their sulfate concentration and the abundance of the tetrahedrally coordinated cations Si and Ti, suggesting that sulfate solubility is governed by competition for tetrahedral sites in silicate melts. Additional experiments performed at 1.5GPa establish the sulfur concentration at sulfide saturation (SCSS) of a putative Martian primary melt at 1400°C. Experiments performed at the same pressure but under more oxidizing conditions constrain the sulfide-sulfate transition to between -1.5 and +0.8 log units relative to the Fayalite-Magnetite-Quartz (FMQ) buffer. This inference is supported by XANES spectra for these experiments, which show no evidence for mixed speciation. In addition to sulfur, the elements Sc, Y and a selection of REEs were also investigated at 1.5GPa. Partitioning coefficients were measured for these elements between a likely primary Martian melt, and three phases believed to be present in the Martian mantle: high- and low- calcium pyroxene and Fo76 olivine. The values obtained closely approximate those predicted by the lattice-strain model, suggesting that these elements partitioning behaviour is unaffected by the unusually Ferich chemistry of this melt. REE profiles generated by a simple partial melting model demonstrate the complexity of Martian meteorites' petrogenesis, including the need for chemically separate reservoirs within Mars' mantle. To assist in further investigations of Martian magmatism, in particular those involving sulfur, a method is presented for the manufacture of polycrystalline olivine capsules. These are suitable as containers for olivine-saturated melts in one-atmosphere experiments, and permit studies involving a wider range of melt compositions than those allowed by natural olivines (the presently used alternative). The equipment required for manufacture is readily available in most experimental petrology laboratories. Use of these capsules avoids common problems associated with using wire loops, such as their propensity to absorb iron, or to react with the sulfur in the experimental charge.

The Howardite-Eucrite-Diogenite (HED) meteorites are a suite of basalts, cumulates and breccias which originate from one differentiated parent body, and are linked to the asteroid Vesta. The HEDs are petrologically diverse with a range of major, minor and trace element compositions. Early crystallisation ages are recorded and so the HEDs provide us with a unique snapshot into the early solar system. The aim of this thesis is to investigate the petrogenesis of the eucrites and diogenites by addressing two questions. What is the Bulk Silicate Vesta (BSV) composition? What differentiation and magmatic processes have occurred? Putative BSV compositions were derived from the geochemistry of the meteorites and geophysical observations of Vesta. Series of one-atmosphere experiments and thermodynamic models investigated the BSV phase relations. Olivine crystallised at ~1625 °C, followed by orthopyroxene at ~1350 °C and feldspar at ~1125 °C. Low-Ca pyroxene-melt partition coefficients for the minor and trace elements were measured. The compatibility of the REEs and HFSEs in low- Ca pyroxene increased by a factor of three, as temperature decreased from 1300-1125 °C and calcium content increased from Wo0.5-Wo8. These partition coefficients were combined with the observed phase relations to perform geochemical trace element calculations of differentiation and magmatic processes. My results suggest that BSV had an Mg#(100*(Mg/(Mg+Fe2+)) between 75-80, > 43 wt. % SiO2, 2.5 x CI refractory lithophile elements, 0.5 wt. % MnO and 0.75 wt. % Cr2O3. A three stage model for Vesta's evolution is suggested. Firstly, extensive if not global partial melting of BSV. Then, equilibrium crystallisation of the mantle and fractional crystallisation of mantle-derived melts produced diogenitic cumulates and eucrite liquids, accounting for the range in major and trace element abundances. The re-equilibration of trapped melt in cumulates is also thought to have occurred. Finally, crustal anatexis produced the range in trace element fractionations seen.

The accretion of the Earth was marked by the high-pressure segregation of most of its core, accompanied by dissolution of about 10% of one or more “light” elements into the metallic phase. Various light elements have been proposed including S, Si, C and O, with each having an effect on the partitioning behaviour of the trace elements. Metallurgical data indicate that dissolution of even small amounts of light elements in liquid Fe can have profound effects on the activities of some trace components. For instance, significant partitioning of Si into the core of the growing Earth should have affected the observed Mo content of the mantle (Ono-Nakazato et al., 2007). Here, I use the epsilon model of non-ideal interactions in Fe liquids (ε. I present interaction parameters (ε, derived from 1.5GPa, 1550-1650oC metal-silicate equilibration experiments, for W, Ni, Co and Mo in liquid Fe alloyed with C, and W, Ni, Co, Mo, Cu, Mn, Ag, Sb, Cd, In, Tl, Pb, Ga, Ge, Cr, V and Zn in liquid Fe alloyed with S. Additional epsilon values were taken from the steelmaking sourcebook when necessary. I used this new data in conjunction with published metal–silicate partitioning results to develop a model of continuous accretion and core segregation taking explicit account of the partitioning of Si (this study) and O (from Ozawa et al., 2008) between metal and silicate and their effects on metal–silicate partitioning of siderophile elements. The best model for explaining the Earth’s formation including the Mo:W ratio of the silicate Earth posits that the Earth’s oxidation state increased in steps from 1 to 6.26% FeO, increasing Si in the Earth’s core and the light elements C and S being added to the planet in the last ~20% of accretion material.