An experimental investigation of F, Cl and H2O mineral-melt partitioning in a reduced, model lunar system
Introduction
The presence of ‘water’ or related H-bearing species, i.e. H2O, OH−, H2, etc., in the interior of the Earth has a controlling influence on a range of mantle properties from melting behaviour to rheology (Peslier, 2010, and references therein). The additional role of water in sustaining life means that the origin of Earth’s volatiles and the timing of their delivery are key areas of study in Earth and planetary sciences. However, many geological processes have modified the Earth since its initial formation, redistributing these volatiles and masking geochemical signatures of their origin. If it is accepted that the Earth and Moon share some common history, the question is when did their compositions diverge and to what extent (Albarède, 2009). According to the canonical “Giant Impact” hypothesis, the Moon formed following collision of a Mars-sized planetesimal with the early Earth (Canup and Agnor, 2000, Canup, 2004). This catastrophic heating event was previously believed to be consistent with an anhydrous lunar interior and lunar sample suite (Lucey et al., 2006, Taylor et al., 2006). The subsequent detection of water and other volatiles in volcanic lunar glasses (Saal et al., 2008) has heralded a new era of lunar research and reinvestigation of the lunar sample suite. This has revealed the presence of ‘water’ in volcanic lunar glasses (Saal et al., 2008, Saal et al., 2013, Greenwood, 2018), in apatites in mare basalts (Boyce et al., 2010, McCubbin Francis et al., 2010, McCubbin et al., 2010, Greenwood et al., 2011, Barnes et al., 2013, Greenwood, 2018), in olivine-hosted melt inclusions in mare basalts (Chen et al., 2015, Hauri et al., 2011, Hauri et al., 2015, Ni et al., 2017, Ni et al., 2019), in plagioclase from lunar anorthosites (Hui et al., 2013, Hui et al., 2017), as well as halogens in mesostasis in lunar basalts (Greenwood et al., 2020). It is now clear that various lunar mantle source regions contain appreciable H, in addition to Cl, F, C and S. As the Moon shares a common origin with the Earth, the logical conclusion is that a similar process may have been involved in volatile delivery to both components of the early Earth-Moon system. Volatiles were either delivered to the Earth prior to the Moon-forming event or delivered to the Earth-Moon system shortly after the Moon-forming event, whilst the newly-formed Moon was still in a largely molten state. As the Moon has remained less affected by large-scale geological processes since its formation, the lunar volatile budget may provide much needed insight into the early Earth-Moon system, and ultimately, provide constraints on the origin of Earth’s hydrosphere.
The lunar sample suite is consistent with a high temperature origin for the Moon. For example, the lunar anorthositic crust probably represents a plagioclase floatation crust formed during lunar magma ocean (LMO) solidification (Warren, 1985). The later mare basalts most likely formed by partial melting of deeper mantle cumulates, also produced by the same LMO solidification process, following late-stage cumulate overturn (Ryder, 1991). As a result, the volatile content of mare basalts and associated volcanic glasses provides insight into the volatile budget of the lunar mantle melts and the LMO solidification products (e.g. Hauri et al., 2011, Saal et al., 2008). With the detection of ‘water’ in lunar materials and inferred ‘water’ contents in parental lunar magmas, it is tempting to make direct comparison with terrestrial magmas. For example, Hauri et al. (2011) noted that volatile contents of melt inclusions in lunar olivine are similar to those in terrestrial mid-ocean ridge basalts, implying similar volatile budgets in terrestrial and lunar mantle source regions. However, any such comparison can be misleading because of differences in magmatic conditions between evolved terrestrial and lunar systems. Mare basalts and associated volcanic glasses are reduced compared to terrestrial volcanic materials, with mineral assemblages and direct oxygen fugacity (fO2) measurements indicating fO2 below that of the iron-wüstite (IW, i.e. Fe-FeO) solid buffer, and either at or near Fe-metal saturation (Longhi, 1992, Shearer et al., 2006). The reduced nature of lunar magmas implies melting of a similarly reduced lunar interior (Longhi, 1992), with the oxidation state of the source region for mare basalts and volcanic glasses generally assumed to be at least 2 log units lower than Earth’s upper mantle (Rutherford and Papale, 2009, Sato, 1976). Oxygen fugacity has a fundamental control on ‘water’ or H incorporation in silicate materials. In terrestrial systems at high mantle temperatures and pressures ‘water’ is incorporated into nominally anhydrous minerals such as olivine and pyroxenes as interstitial H associated with underbonded O atom sites (e.g. Skogby et al., 1990, Smyth et al., 1991, Ingrin and Skogby, 2000, Stalder and Skogby, 2003, Bromiley et al., 2004a, Bromiley and Bromiley, 2006, Smyth et al., 2006), forming various O–H defects (Lemaire et al., 2004, Matveev et al., 2005, Grant et al., 2007). In terrestrial silicate melts, water is incorporated initially as OH−, with molecular H2O becoming more dominant at higher total water contents, at least under more oxidising conditions (e.g. Stolper, 1982, Pandya et al., 1992, Dixon et al., 1995). This has recently been contested by Cody et al. (2020) who suggest that H2O dominates at low H abundances, at least in highly polymerised silicate melts, although whether this is true in less polymerised, basaltic melts remains unclear. Under reducing conditions H2O becomes unstable as a species, and other H-related defects such as H2 and CH4 are observed in spectroscopic investigations of melts at the expense of OH− (Kadik et al., 2006, Mysen et al., 2011, Hirschmann et al., 2012, Ardia et al., 2013). Magmas are commonly used as probes of planetary interiors as they represent escaped partial melts that transport information about source regions. The ‘water’ content of magmas is controlled in part by mineral-melt partitioning during partial melting of the mantle source. Any change in speciation of ‘water’ as a function of fO2 could result in significant changes in partitioning. The ‘water’ content of a magma will likely depend on both the H concentration and the fO2 in its mantle source region, and any comparison between lunar and terrestrial magmas must take this into account. Kadik et al. (2006) noted that at high pressure in silicate melts at fO2 below IW, H2 becomes increasingly important at the expense of more oxidised species. Hirschmann et al. (2012) noted that the proportion of H2 in lunar magmas is low under the low total water contents of the lunar interior. However, in reduced lunar systems over a range of inferred bulk ‘water’ contents (Elkins-Tanton and Grove, 2011) the proportion of H2/H2 + H2O could range up to 20–50%. In solid phases under similarly reducing conditions, Yang et al. (2016) demonstrated that mantle silicates can also incorporate some molecular H2, in addition to OH defects. The effects of any change in speciation on volatile behaviour remain unclear. Newcombe et al. (2017) demonstrated that degassing mechanisms in lunar melts over a range of highly reducing conditions (below the IW buffer) are strongly controlled by H speciation, although they noted that diffusion is dominated by flux of OH− rather than H2. Recently, Lin et al. (2019) demonstrated that fO2 has a strong control on H2O partitioning between plagioclase and silicate melt, but aside from that study there is an absence of data on the effect of fO2 on volatile partitioning in lunar systems. Here we present the first experimental data on partitioning of H2O, F and Cl in a reduced, model lunar mantle lithology system. F and Cl are important tracers in lunar materials (e.g. Hauri et al., 2011, Chen et al., 2015), and are increasingly used to provide insight into volatile redistribution during terrestrial geological processes (e.g. Urann et al., 2017, Klemme and Stalder, 2018). Furthermore, isotopic variations of Cl in lunar materials provide significant additional insight into lunar magmatic processes (McCubbin et al., 2011, Potts et al., 2018, Boyce et al., 2018). Therefore, combined study of F, Cl and H2O distribution during magmatic processes provides a powerful tool for reinvestigating the importance of measured volatile concentrations in lunar samples, and for drawing inferences on volatile delivery to the early Earth-Moon system.
Section snippets
System and bulk composition
To constrain volatile mineral-melt partitioning during lunar mantle melting, a synthetic starting material was used with a composition similar to lunar low-Ti, picritic, green volcanic glass beads. These pyroclastic glasses are believed to be amongst the most primitive lunar samples (Delano, 1979) and suggested to represent partial melts from lunar mantle cumulates formed at around 1.5–2.5 GPa (Shearer et al., 2006). A composition, corresponding to an average green lunar glass from Saal et al.
Phase and melt relations in recovered samples.
Most experiments were conducted at 2 GPa as this is an approximate pressure for multiple mineral saturation in low-Ti picritic lunar systems, and within the range of estimates for the pressure of formation of mare basalts and associated picritic glasses (Elkins-Tanton et al., 2003, Shearer et al., 2006). Run products consisted of quenched glass with some minor growth of quench crystals, and coexisting crystals of olivine, orthopyroxene, and pigeonite. There was no evidence of a fluid phase in
Controls on H2O, F, and Cl mineral/melt partitioning in reduced lunar systems.
Fig. 4 shows F, Cl and H2O partition coefficients from this and previous studies as a function of temperature. This new data for reduced lunar conditions confirms all three volatiles are incompatible, consistent with previous studies. There is a general trend DF > DH2O≫DCl, except for limited olivine data, which suggest DCl > DH2O. H2O is slightly more incompatible than F for all phases, but Cl is much more incompatible than F. This would be predicted based on the significantly larger ionic
Implications for volatiles in the early Earth-Moon system.
New partitioning data for H2O, Cl and F presented here demonstrate that volatile incorporation in a reduced, model lunar system is comparable to more oxidised, terrestrial systems. At fO2 down to approximately IW-2, H remains dominantly incorporated in nominally anhydrous mantle minerals as OH−, at least at relatively low pressures of the lunar interior. Comparison of limited F, Cl and H2O mineral-melt partitioning data demonstrates that all three volatiles are highly incompatible, although F
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
Work was supported by the UK National Environmental Research Council via NE/M000346/1 (to Bromiley) and award of instrument time at the Edinburgh Ion Microprobe facility, IMF597/0516. Brooker was funded by the NERC Thematic Grant consortium NE/M000419/1. Dr Amrei Baasner is thanked for preparing starting mixes, and Dr Chris Hayward and Dr Richard Hinton for assistance in conducted EMPA and SIMS analyses, respectively. The authors thank Prof. Youxue Zhang and two anonymous reviewers whose
References (95)
- et al.
Solubility of CH4 in a synthetic basaltic melt, with applications to atmosphere-magma ocean-core partitioning of volatiles and to the evolution of the Martian atmosphere
Geochim. Cosmochim. Acta
(2013) - et al.
Speciation and solubility of reduced C-O-H-N volatiles in mafic melt: Implications for volcanism, atmospheric evolution, and deep volatile cycles in the terrestrial planets
Geochim. Cosmochim. Acta
(2015) - et al.
Accurate and precise measurements of the D/H ratio and hydroxyl content in lunar apatites using NanoSIMS
Chem. Geol.
(2013) - et al.
Experimental petrology of the Apollo 15 group A green glasses: Melting primordial lunar mantle and magma ocean cumulate assimilation
Geochim. Cosmochim. Acta
(2013) - et al.
Fluorine in nominally fluorine-free mantle minerals: Experimental partitioning of F between olivine, orthopyroxene and silicate melts with implications for magmatic processes
Earth Planet. Sci. Lett.
(2012) - et al.
Early loss, fractionation, and redistribution of chlorine in the Moon as revealed by the low-Ti lunar mare basalt suite
Earth Planet. Sci. Lett.
(2018) - et al.
Origin of the Apollo 14, 15, and 17 yellow ultramafic glasses by mixing of deep cumulate remelts
Geochim. Cosmochim. Acta
(2015) - et al.
Water, fluorine, and sulfur concentrations in the lunar mantle
Earth Planet. Sci. Lett.
(2015) - et al.
Revisiting water speciation in hydrous alumino-silicate glasses: A discrepancy between solid-state 1H NMR and NIR spectroscopy in the determination of X-OH and H2O
Geochim. Cosmochim. Acta
(2020) - et al.
Clumped fluoride-hydroxyl defects in forsterite: Implications for the upper-mantle
Earth Planet. Sci. Lett.
(2014)
Petrogenesis of primitive and evolved basalts in a cooling Moon: Experimental constraints from the youngest known lunar magmas
Earth Planet. Sci. Lett.
Water (hydrogen) in the lunar mantle: Results from petrology and magma ocean modeling
Earth Planet. Sci. Lett.
The effect of oxygen fugacity on hydroxyl concentrations and speciation in olivine: Implications for water solubility in the upper mantle
Earth Plant. Sci. Letts.
Hydrogen and D/H analysis of apatite by Elemental Analyzer-Chromium/High-Temperature Conversion-Isotope Ratio Mass Spectrometry (EA-Cr/HTC-IRMS)
Chem. Geol.
Partitioning of water during melting of the Earth’s upper mantle at H2O-undersaturated conditions
Earth Planet. Sci. Lett.
Water in the Moon’s interior: Truth and consequences
Earth Planet. Sci. Lett.
Solubility of molecular hydrogen in silicate melts and consequences for volatile evolution of terrestrial planets
Earth Planet. Sci. Lett.
The Cr-Cr2O3 oxygen buffer and the free-energy of formation of Cr2O3 from high-temperature electrochemical measurements
Geochim. Cosmochim. Acta
A heterogeneous lunar interior for hydrogen isotopes as revealed by the lunar highlands samples
Earth Planet. Sci. Lett.
Experimental partitioning of F and Cl between olivine, orthopyroxene and silicate melt at Earth’s mantle conditions
Chem. Geol.
Experimental investigation of the influence of oxygen fugacity on the source depths for high titanium lunar ultramafic magmas
Geochim. Cosmochim. Acta
Experimental petrology and petrogenesis of mare volcanics
Geochim. Cosmochim. Acta
Fluorine and chlorine abundances in lunar apatite: Implications for heterogeneous distributions of magmatic volatiles in the lunar interior
Geochim. Cosmochim. Acta
Solubility and solution mechanisms of C-O–H volatiles in silicate melt with variable redox conditions and melt composition at upper mantle temperatures and pressures
Geochim et Cosmochim Acta
Solubility of water in lunar basalt at low pH2O
Geochim. Cosmochim. Acta
A melt inclusion study on volatile abundances in the lunar mantle
Geochim. Cosmochim. Acta
Volatile loss during homogenization of lunar melt inclusions
Earth Planet. Sci. Lett.
The effect of tetrahedral Al3+ on the partitining of water between clinopyroxene and silicate melt
Earth Planet. Sci. Lett.
The effect of bulk composition on the speciation of water in submarine volcanic glasses
Geochim. Cosmochim. Acta
A review of water contents of nominally anhydrous natural minerals in the mantles of Earth, Mars and the Moon
J. Volcanol. Geotherm. Res.
Chlorine isotopic compositions of apatite in Apollo 14 rocks: Evidence for widespread vapor-phase metasomatism on the lunar nearside similar to 4 billion years ago
Geochim. Cosmochim. Acta
Experimental determination of C, F, and H partitioning between mantle minerals and carbonated basalt, CO2/Ba and CO2/Nb systematics of partial melting, and the CO2 contents of basaltic source regions
Earth Planet. Sci. Lett.
Solubility of H2O- and CO2-bearing fluids in tholeiitic basalts at pressures up to 500 MPa
Chem. Geol.
The effect of melt composition and oxygen fugacity on manganese partitioning between apatite and silicate melt
Chem. Geol.
The speciation of water in silicate melts
Geochim. Cosmochim. Acta
Hydrogen partitioning between nominally anhydrous upper mantle minerals and melt between 3 and 5 GPa and applications to hydrous peridotite partial melting
Chem. Geol.
Core formation and the oxidation state of the Earth
Earth Planet. Sci. Lett.
Origin of titaniferous lunar basalts
Geochim. Cosmochim. Acta
Molecular hydrogen in mantle minerals
Geochemical Perspect. Lett.
Volatile accretion history of the terrestrial planets and dynamic implications
Nature
Lunar apatite with terrestrial volatile abundances
Nature
The chlorine isotope fingerprint of the lunar magma ocean
Science Advances
Hydrogen and minor element incorporation in synthetic rutile
Min. Mag.
High-pressure phase transitions and hydrogen incorporation in MgSiO3 enstatite
Am. Mineral.
Hydrogen and deuterium diffusion in non-stoichiometric spinel
High Press. Res.
Hydrogen solubility and speciation in natural, gem-quality chromian diopside
Am. Mineral.
Cited by (8)
The H-poor nature of incompletely melted planetesimals: The view from acapulcoites and lodranites
2024, Geochimica et Cosmochimica ActaFluorine abundance of the lunar magma ocean constrained by experimentally determined mineral-melt F partitioning
2024, Geochimica et Cosmochimica ActaNew interpretations of lunar mare basalt flow emplacement from XCT analysis of Apollo samples
2022, IcarusCitation Excerpt :In 2008, Saal et al. showed that the cores of low-Ti lunar volcanic glass beads contain up to 46 ppm H2O, 40 ppm, F, and 576 ppm S, with cores of high-Ti beads containing up to 15 ppm H2O, 15 ppm F, and ~ 400 ppm S, and the cores of very-low-Ti volcanic glasses containing up to 30 ppm H2O, 10 ppm F, and 270 ppm S (Saal et al., 2008). Researchers have since argued that the sources of these volatile species in the lunar mantle include residual liquid that remained trapped between lunar mantle cumulates after the solidification of the mantle, and nominally anhydrous minerals that contain trace amounts of volatiles (e.g., Liu et al., 2012; Hui et al., 2013; Mills et al., 2014; McCubbin et al., 2015; Potts et al., 2021). The urKREEP reservoir which formed towards the end of the crystallization of the Lunar Magma Ocean and which is enriched in K, rare earth elements, and P, likely also played a significant role in provide volatile species to lunar magmas.
High precision measurement of trace F and Cl in olivine by electron probe microanalysis
2023, Progress in Earth and Planetary ScienceOn the origin of fluorine-poor apatite in chondrite parent bodies
2023, American Mineralogist