Lunar Magma Ocean Crystallization Research Articles

Sound velocities in lunar mantle aggregates at simultaneous high pressures and temperatures: Implications for the presence of garnet in the deep lunar interior

Recent experimental and theoretical studies on lunar magma ocean crystallisation have suggested the presence of significant proportions of garnet in the deep lunar interior. While phase relation studies indicate a deep lunar mantle consisting of olivine, pyroxene, and garnet, the compatibility of such an assemblage with seismic models of the lunar interior is yet untested. In this study we report compressional and shear wave velocities in an iron-rich assemblage consisting of olivine, orthopyroxene, clinopyroxene, and garnet up to ∼8 GPa and 1300 K, by means of ultrasonic interferometry measurements combined with synchrotron techniques using the multi-anvil press apparatus. Sound velocity and density models of lunar mantle rocks along a selenotherm based on our experimental results find good agreement with the seismic and density profiles at lunar interior depths of 740–1260 km. Further models are constructed, allowing for the variation of chemical composition, phase proportion, and temperature; these suggest that a garnet-rich deep lunar mantle is compatible with present-day lower lunar mantle temperatures of between 1400–1800 K. Our results show that lunar mantle rocks with up to 33 wt.% garnet may provide an explanation for the observed high velocities of the lower lunar mantle. The presence of garnet in the lowermost part of the Moon's mantle has significant implications for the depth and temperature of the Moon's magma ocean as well as the composition, structure and internal dynamics of the solid Moon.

Fractionation of iron and titanium isotopes by ilmenite and the isotopic compositions of lunar magma ocean cumulates

Basaltic volcanism on the Moon produced low- and high-Ti mare basalt suites that are also distinct with respect to their iron, titanium, and magnesium isotopic compositions. Here, the equilibrium fractionation of Fe and Ti isotopes between ilmenite and melt was experimentally investigated in order to evaluate the role of ilmenite in generating the isotopic compositional variability among the lunar mare basalts. Ilmenite crystallization experiments were conducted using two bulk compositions: an ilmenite-saturated basaltic andesite and an ilmenite-saturated Apollo 14 black glass, and the Fe and Ti isotopic compositions of the experimental ilmenites and glass (quenched melt) were analyzed using solution MC-ICPMS after hand-picking. Additionally, Nuclear Resonant Inelastic X-ray Scattering (NRIXS) measurements on synthetic ilmenite were conducted and compared to previous NRIXS measurements on synthetic lunar glasses in order to derive temperature-dependent equilibrium ilmenite-melt Fe isotopic fractionations. Experimentally determined ilmenite-melt fractionations were then incorporated into a lunar magma ocean crystallization model that tracks the major element and isotopic compositional evolution of lunar magma ocean cumulates and residual liquid. There is good agreement between the Fe equilibrium isotopic fractionation measured by NRIXS and the laboratory equilibration experiments, and we find that the isotopic fractionation is sensitive to ilmenite compositional differences (0 vs. 10% Fe3+). Further, the light Ti isotopic composition of ilmenite relative to the melt (Δ49Ti=ilmenite-melt−0.09±0.03‰ at 1100∘C) is consistent with the higher coordination of Ti in ilmenite relative to melts and results of previous studies. The modeled Ti isotopic compositions for lunar magma ocean cumulates display Ti isotopic variability sufficient to explain the low- and high-Ti mare basalt sources. However, the difference in Fe isotopic composition between the low- and high-Ti mare basalts cannot be attributed solely to ilmenite fractionation. Instead, Fe isotopic fractionation by additional products of lunar magma ocean crystallization, such as clinopyroxene, is required to generate the inferred Fe and Mg isotopic variability in the lunar mantle. Alternatively, the Fe and Mg isotopic compositions of the lunar mare basalts may indicate Fe-Mg interdiffusion has occurred in the Ti-rich component of the mare basalt source regions via reaction between ilmenite cumulates and the olivine- and pyroxene-rich lunar mantle.

Methodologies for 176Lu-176Hf Analysis of Zircon Grains from the Moon and Beyond.

Zircons are found in extraterrestrial rocks from the Moon, Mars, and some differentiated meteorite parent-bodies. These zircons are rare, often of small size, and have been affected by neutron capture induced by cosmic ray exposure. The application of the 176Lu-176Hf decay system to zircons from planetary bodies such as the Moon can help establish the chronology of large-scale differentiation processes such as the crystallization of the lunar magma ocean. Here, we present methods to measure the isotopic composition of Hf of extraterrestrial zircons dated using ID-TIMS U-Pb after chemical abrasion. We introduce a 2-stage elution scheme to separate Hf from Zr while preserving the unused Zr fraction for future isotopic analysis. The effect of neutron capture is also re-examined using the latest thermal neutron capture cross sections and epithermal resonance integrals. Our tests show that the precision of Hf isotopic analyses is close to what is theoretically attainable. We have tested this method to a limited set of zircon grains from lunar rocks returned by the Apollo missions (lunar soil 14163, fragmental polymict breccia 72275, and clast-rich breccia 14321). The model ages align with previously reported values, but further work is needed to assess the chronology of lunar magma ocean crystallization as only a handful of small zircons (5 zircons from 3 samples) were analyzed, and the precision of the analyses can be improved by measuring more and larger lunar zircon grains.

Fluorine abundance of the lunar magma ocean constrained by experimentally determined mineral-melt F partitioning

To quantify fluorine (F) evolution during lunar magma ocean (LMO) crystallization, high-pressure, high-temperature experiments have been conducted to determine mineral-melt partitioning of F for lunar minerals (plagioclase, orthopyroxene and ilmenite). Results constrain the F abundance in the magma ocean to 21–41 ppm at the time crust-forming plagioclase started crystallizing. Forward modeling shows that 352–703 ppm F would remain in the final 1 % of magma toward the end of magma ocean solidification. This range overlaps that inferred for the urKREEP reservoir (660 ppm). Taking into account model uncertainties, from the perspective of F abundances the urKREEP reservoir can be formed at 98.9–99.5% LMO solidification, with negligible loss of F from the Moon since the onset of crust formation. Backward modeling from initial crust-forming plagioclase, an initial LMO would contain 4.2–8.5 ppm F, which is consistent with estimates of the lunar primitive mantle F content derived from melt inclusions in Apollo samples. This finding is consistent with previous suggestions that the bulk silicate Moon is depleted in F relative to the bulk silicate Earth (which contains ∼25 ppm F). A BSE-like initial LMO would yield a magma containing 122 ppm F at the onset of crust formation, significantly higher than our calculated 21–41 ppm F. Fluorine depletion could have occurred by degassing during the early LMO stages (between the onset of LMO crystallization and first crust formation), and/or prior to the LMO stage (e.g., depletion during the giant impact or vapor drainage in the protolunar disk), but seems to have ended by the time the crust started forming.

Moon’s high-energy giant-impact origin and differentiation timeline inferred from Ca and Mg stable isotopes

Mass-dependent stable isotopic variations recorded in lunar samples provide novel resolution to the formation and differentiation history of the Moon. In this study, we report new high-precision Ca-isotope measurements for lunar rocks and minerals. Ca-isotope data and modeling of the lunar magma ocean together demonstrate indistinguishable mass-dependent Ca isotopic compositions of the bulk silicate Earth and Moon. This implied Earth-Moon isotope equilibration is consistent with the Moon’s high-energy giant-impact (Synestia) origin and not readily compatible with the traditional giant-impact models. Moreover, a cross-comparison between Ca and Mg isotopic data for an important anorthosite sample (60025) consistently clarifies its formation near the completion of the lunar magma ocean crystallization. Therefore, the various existing radiometric dating for 60025 sets the lunar magma ocean to have fully solidified by either 4.51 or 4.38 billion years ago, constraining the two respective lunar differentiation timescales to <30 (short-lived) or ~130–150 (long-lived) million years.

Controls on determining the bulk water content of the Moon

AbstractThe Moon is much wetter than previously thought. The estimated bulk H2O concentrations based on the analyses of H2O in lunar materials show a wide range from 5 to 1650 ppm. To better constrain bulk H2O in the lunar magma ocean (LMO), we model LMO crystallization and vary DH (concentration of H2O in LMO mineral/concentration of H2O in melt), interstitial melt fraction, and initial LMO depth. We take the highest and lowest values of DH reported in the literature for the LMO minerals. We assess the bulk H2O content required in the initial magma ocean to satisfy two observational constraints: (1) H2O measured in plagioclase grains from ferroan anorthosites and (2) crustal mass from GRAIL. We find that the initial bulk LMO H2O that best explains the H2O content in crustal plagioclase is strongly dependent on DH rather than interstitial melt fractions or initial LMO depths, with a drier magma ocean (10 ppm H2O) being favored with higher DH and a wetter magma ocean (100–1000 ppm H2O) with lower DH. This underscores the importance of constraining DH specific to lunar conditions in future studies. We also demonstrate that crustal mass is not an effective hygrometer.

Experimental constraints on the long-lived radiogenic isotope evolution of the Moon

This study presents the results of high pressure and temperature experiments to investigate the mineral–melt trace element partitioning behaviour for minerals predicted to have formed during the crystallisation of the Lunar Magma Ocean (LMO). The focus of this work has been particularly on determining partition coefficients for parent–daughter pairs of radiogenic elements, for LMO-relevant temperatures, pressures and compositions. The new experimental data are compared with previous studies for the same minerals and elements in order to establish best estimates for the partition coefficient of each element for evolving compositions of minerals as predicted in recent studies modelling LMO crystallisation. These estimates are used to calculate evolving parent–daughter ratios in the LMO residual melt and crystallising minerals for the four main long-lived radiogenic isotope systems that have been studied in lunar samples (Rb–Sr, Sm–Nd, Lu–Hf and U–Pb). The calculated 87Rb/86Sr, 147Sm/144Nd, and 176Lu/177Hf ratios are consistent with predictions for the mantle sources of lunar basalts and evolved lithologies. In contrast, it is difficult to explain the wide range of 238U/204Pb source ratios predicted from the Pb isotopic compositions of basaltic lunar samples. Potential explanations for this observation are discussed, with the conclusion that the Moon most likely experienced a significant loss of volatiles (including Pb), towards the end of LMO crystallisation, resulting in the dramatic U–Pb fractionation evidenced by recent sample analyses.

Garnet stability in the deep lunar mantle: Constraints on the physics and chemistry of the interior of the Moon

High-pressure, high-temperature experiments have been conducted at deep lunar mantle conditions to constrain the garnet stability field. Using the Taylor Whole Moon composition, garnet is found to be stable at pressures above 3 GPa and temperatures below 1700°C, yielding a smaller stability field than previously suggested on the basis of thermodynamic calculations. Experimental data are used to model equilibrium crystallization in a ‘two-stage’ model of lunar magma ocean (LMO) crystallization starting from a fully molten Moon. In the first stage, isothermal (1600°C) equilibrium crystallization of the LMO would produce garnet-bearing lherzolite cumulates (containing up to ∼20 wt.% garnet) in the lowermost lunar mantle. Garnet in the deep lunar mantle would significantly decrease the Al2O3 content of the residual LMO and impact HREE/MREE fractionation. Numerical modeling of the second stage (residual LMO fractional crystallization) shows a delay in plagioclase saturation compared to models of single-stage fractional crystallization of a whole-Moon LMO of the Taylor Whole Moon composition, thinning the anorthositic crust from 95 km to 75 km. To reach the upper limit of current estimates of the average lunar crustal thickness (∼45 km), the two-stage scenario needs to be accompanied by a total of 10% liquid trapped in cumulates and 70% efficiency of plagioclase flotation. We also conduct trace element evolution modeling and reproduce a REE pattern identical to high K, REE, and P (KREEP) compositions after 99.8% solidification, when starting with a CI chondritic REE abundance. The density of a garnet-bearing deep lunar mantle is significantly higher than the density of olivine/orthopyroxene mixtures without garnet. The present-day lowest mantle in the Moon could therefore be characterized by chemical interactions between the earliest (garnet-bearing) and latest (ilmenite-bearing) products of LMO crystallization.

Conditions and extent of volatile loss from the Moon during formation of the Procellarum basin

Rocks from the lunar interior are depleted in moderately volatile elements (MVEs) compared to terrestrial rocks. Most MVEs are also enriched in their heavier isotopes compared to those in terrestrial rocks. Such elemental depletion and heavy isotope enrichments have been attributed to liquid-vapor exchange and vapor loss from the protolunar disk, incomplete accretion of MVEs during condensation of the Moon, and degassing of MVEs during lunar magma ocean crystallization. New Monte Carlo simulation results suggest that the lunar MVE depletion is consistent with evaporative loss at 1,670 ± 129 K and an oxygen fugacity +2.3 ± 2.1 log units above the fayalite-magnetite-quartz buffer. Here, we propose that these chemical and isotopic features could have resulted from the formation of the putative Procellarum basin early in the Moon's history, during which nearside magma ocean melts would have been exposed at the surface, allowing equilibration with any primitive atmosphere together with MVE loss and isotopic fractionation.

Effects of initial velocity on the sinking of ilmenite-bearing cumulates in the early lunar mantle: Insight from the theory of Rayleigh-Taylor instability

The early lunar mantle overturn, associated with the sinking of the dense ilmenite-bearing cumulate (IBC) crystallized at the shallow lunar mantle, provides satisfactory explanations for the origination of high-Ti basalt, the abnormally strong magnetic field between ∼ 3.9 and ∼ 3.6 Ga and the low-viscosity zone in the deep lunar mantle, but still poses a debate regarding the initial state of IBC in the early lunar mantle. If the sinking of IBC initiated before the end of lunar magma ocean crystallization, the solidified IBC can acquire a greater thickness and a higher initial velocity at the IBC-mantle boundary. The variation of initial velocity can affect the strain rate of IBC and, correspondingly, the dislocation creep components at the shallow lunar mantle. In this work, we analyze the effects of initial velocity on the dynamics of early lunar mantle by using the theory of Rayleigh-Taylor instability. To couple the effects of diffusion creep and dislocation creep for all major minerals in the lunar mantle, we exploit an improved Minimized Power Geometric (IMPG) model and isostress mixing model to characterize the upper limit and lower limit for the viscosity of the lunar mantle comprising four major minerals, i.e. olivine, orthopyroxene, clinopyroxene and ilmenite. The modeling results suggest that a high initial velocity, in any case, can shorten the onset time, tending to promote the early lunar mantle overturn even in a rheologically-strong lunar mantle. The effect of initial velocity on the overturn wavelength shows a strong dependence on the rheological mixing model. For the isostress mixing model, the increase of initial velocity tends to elongate the overturn wavelength. For the IMPG mixing model, the overturn wavelength is insensitive to the variation of initial velocity. As the actual lunar mantle rheology sandwiches between the rheologies predicted by isostress mixing model and IMPG model, it can be anticipated that the increase of initial velocity tends to elongate the overturn wavelength. In consideration of the importance of the initial velocity on the dynamics of early lunar mantle, future investigations should focus on the dynamics of the solid IBC in the solidifying lunar magma ocean.

Constraints on early Earth's water budget from the evolution of the lunar hydrogen cycle

During the Hadean, Earth recovered from the Moon-forming giant impact, became covered with liquid water oceans, and witnessed the onset of plate tectonics and life. Quantifying the abundances, distribution, and chemical states of water in the atmosphere, on the surface, and in the interior of the early Earth is essential to constrain the early evolution of System Earth. Assessing these parameters is hampered by the general dearth of early Earth samples, the difficulty of distinguishing primary signatures from later alteration processes in such samples, leading to large uncertainties on the influx and outflux of water to and from the early Earth. Given the close proximity of Earth and Moon, constraints on the early hydrogen cycle in the Moon may reflect coeval aspects of the water cycle on early Earth. Here, we assess constraints on the lunar water cycle from the time the Moon formed until the end of late accretion at ~3.8 Ga, and implications of these constraints for the early Earth water budget. Dynamic accretion models suggest the Moon initially contained ~455 ppm of water. Recent experimental studies of lunar magma ocean crystallization suggest similarly substantial initial lunar water contents. Hydrogen concentration measurements in lunar plagioclase crystals derived from the magma ocean illustrate that the Moon experienced significant degassing during the solidification of the lunar magma ocean (thought to have occurred between 4.5 and ~ 4.3 Ga). Hydrogen and chlorine systematics in lunar magmatic apatite grains formed between ~4.1 Ga and ~ 3 Ga indicate that lunar hydrogen reservoirs were replenished by volatile delivery during late accretion (~4.1–3.8 Ga), after which the water abundance of the Moon stabilized. Using this knowledge of the lunar water cycle to model Earth's early water budget leads to two scenarios that are consistent with the observed present-day terrestrial water content of 1000–3000 ppm: (1) Earth contained significantly more water than the Moon-forming material immediately after the giant impact, suggesting hydrogen heterogeneity in the initial Earth-Moon system; (2) Earth did not experience significant degassing in the aftermath of the giant impact, and the late accretion mass added to Earth was large and water-rich

The Fate of Sulfur and Chalcophile Elements During Crystallization of the Lunar Magma Ocean

AbstractTo assess the viability of sulfide liquid saturation during crystallization of the lunar magma ocean (LMO), we present a new data set describing both the sulfur (S) concentration at sulfide liquid saturation (SCSS) and sulfide liquid‐silicate melt partition coefficients of many trace elements for various differentiated lunar magmas at lunar‐relevant conditions. Using these parameterizations, we model the SCSS and the distribution of the most chalcophile elements with progressive LMO crystallization in the absence and presence of sulfide liquids. Modeling results for different modes of LMO crystallization show that for proposed lunar mantle S abundances FeS sulfide liquid saturation is expected to occur between 96% and 98% of LMO crystallization. This is decreased to &gt;91% for Fe‐S liquids with 30% Ni or Cu. Saturation of S‐poor sulfide liquids can occur at &gt;75% of LMO crystallization. The timing of sulfide liquid saturation depends most strongly on the assumed S content of the lunar mantle following formation of the lunar core and on the sulfide liquid composition. Modeled abundances of chalcophile elements indicate that sulfide‐liquid saturation during late‐stage LMO crystallization would yield much lower abundances of Ni and Cu than observed in KREEP basalts and estimated for the urKREEP reservoir, as well as lower Ni/Co than observed in the latter. Sulfide liquids therefore did not affect moderately siderophile and chalcophile element fractionation within the LMO, supporting the hypothesis that the nonvolatile, siderophile element abundances of the lunar mantle reflect a phase of core formation and/or the addition of a meteoritic late veneer.

The chlorine-isotopic composition of lunar KREEP from magnesian-suite troctolite 76535

AbstractWe conducted in situ Cl isotopic measurements of apatite within intercumulus regions and within a holocrystalline olivine-hosted melt inclusion in magnesian-suite troctolite 76535 from Apollo 17. These data were collected to place constraints on the Cl-isotopic composition of the last liquid to crystallize from the lunar magma ocean (i.e., urKREEP, named after its enrichments in incompatible lithophile trace elements like potassium, rare earth elements, and phosphorus). The apatite in the olivine-hosted melt inclusion and within the intercumulus regions of the sample yielded Cl-isotopic compositions of 28.3 ± 0.9‰ (2σ) and 30.3 ± 1.1‰ (2σ), respectively. The concordance of these values from both textural regimes we analyzed indicates that the Cl-isotopic composition of apatites in 76535 likely represents the Cl-isotopic composition of the KREEP-rich magnesian-suite magmas. Based on the age of 76535, these results imply that the KREEP reservoir attained a Cl-isotopic composition of 28–30‰ by at least 4.31 Ga, consistent with the onset of Cl-isotopic fractionation at the time of lunar magma ocean crystallization or shortly thereafter. Moreover, lunar samples that yield Cl-isotopic compositions higher than the value for KREEP are likely affected by secondary processes such as impacts and/or magmatic degassing. The presence of KREEP-rich olivine-hosted melt inclusions within one of the most pristine and ancient KREEP-rich rocks from the Moon provides a new opportunity to characterize the geochemistry of KREEP. In particular, a broader analysis of stable isotopic compositions of highly and moderately volatile elements could provide an unprecedented advancement in our characterization of the geochemical composition of the KREEP reservoir and of volatile-depletion processes during magma ocean crystallization, more broadly.

The chlorine isotopic composition of the Moon: Insights from melt inclusions

The Moon exhibits a heavier chlorine (Cl) isotopic composition compared to the Earth. Several hypotheses have been put forward to explain this difference, based mostly on analyses of apatite in lunar samples complemented by bulk-rock data. The earliest hypothesis argued for Cl isotope fractionation during the degassing of anhydrous basaltic magmas on the Moon. Subsequently, other hypotheses emerged linking Cl isotope fractionation on the Moon with the degassing during the crystallization of the Lunar Magma Ocean (LMO). Currently, a variant of the LMO degassing model involving mixing between two end-member components, defined by early-formed cumulates, from which mare magmas were subsequently derived, and a KREEP component, which formed towards the end of the LMO crystallization, seems to reconcile some existing Cl isotope data on lunar samples. To further ascertain the history of Cl in the Moon and to investigate any evolution of Cl during magma crystallization and emplacement events, which could help resolve the chlorine isotopic variation between the Earth and the Moon, we analysed the Cl abundance and its isotopic composition in 36 olivine- and pyroxene-hosted melt inclusions (MI) in five Apollo basalts (10020, 12004, 12040, 14072 and 15016). Olivine-hosted MI have an average of 3.3±1.4ppm Cl. Higher Cl abundances (11.9 ppm on average) are measured for pyroxene-hosted MI, consistent with their formation at later stages in the crystallization of their parental melt compared to olivines. Chlorine isotopic composition (δ37Cl) of MI in the five Apollo basalts have weighted averages of +12.8±2.4‰ and +10.1±3.2‰ for olivine- and pyroxene-hosted MI, respectively, which are statistically indistinguishable. These isotopic compositions are also similar to those measured in apatite in these lunar basalts, with the exception of sample 14072, which is known to have a distinct petrogenetic history compared to other mare basalts. Based on our dataset, we conclude that, post-MI-entrapment, no significant Cl isotopic fractionation occurred during the crystallization and subsequent eruption of the parent magma and that Cl isotopic composition of MI and apatite primarily reflect the signature of the source region of these lunar basalts. Our findings are compatible with the hypothesis that in the majority of the cases the heavy Cl isotopic signature of the Moon was acquired during the earliest stages of LMO evolution. Interestingly, MI data from 14072 suggests that Apollo 14 lunar basalts might be an exception and may have experienced post-crystallization processes, possibly metasomatism, resulting in additional Cl isotopic fractionation recorded by apatite but not melt inclusions.

A lunar hygrometer based on plagioclase-melt partitioning of water

The Moon was initially covered by a magma ocean. Hydrogen detected in plagioclase of ferroan anorthosites, the only available samples directly crystallised from the lunar magma ocean (LMO), can be used to quantify LMO hydrogen content. We performed experiments to determine plagioclase-melt partition coefficients of water under LMO conditions with water contents of co-existing plagioclase and melt quantified using Fourier-Transform Infrared Spectroscopy. Results indicate lunar plagioclase can incorporate approximately one order of magnitude more water than previously assumed. Using measured water contents of lunar plagioclase, this suggests that ~100 μg/g H2O equivalent was present in the residual magma when 95 % of the initial LMO had crystallised. Our results constrain initial LMO water contents to ~ 5 μg/g H2O equivalent if water was conserved throughout LMO evolution. If on the other hand the initial LMO contained >1000 μg/g water as suggested by experiments on LMO crystallisation, >99 % hydrogen degassing occurred during the evolution of the LMO.

Overturn of Ilmenite‐Bearing Cumulates in a Rheologically Weak Lunar Mantle

AbstractThe crystallization of the lunar magma ocean (LMO) determines the initial structure of the solid Moon. Near the end of the LMO crystallization, ilmenite‐bearing cumulates (IBC) form beneath the plagioclase crust. Being denser than the underlying mantle, IBC are prone to overturn, a hypothesis that explains several aspects of the Moon's evolution. Yet the formation of stagnant lid due to the temperature dependence of viscosity can easily prevent IBC from sinking. To infer the rheological conditions allowing IBC to sink, we calculated the LMO crystallization sequence and performed high‐resolution numerical simulations of the overturn dynamics. We assumed a diffusion creep rheology and tested the effects of reference viscosity, activation energy, and compositional viscosity contrast between IBC and mantle. The overturn strongly depends on reference viscosity and activation energy and is facilitated by a low IBC viscosity. For a reference viscosity of 1021 Pa s, characteristic of a dry rheology, IBC overturn cannot take place. For a reference viscosity of 1020 Pa s, the overturn is possible if the activation energy is a factor of 2–3 lower than the values typically assumed for dry olivine. These low activation energies suggest a role for dislocation creep. For lower‐reference viscosities associated with the presence of water or trapped melt, more than 95% IBC can sink regardless of the activation energy. Scaling laws for Rayleigh‐Taylor instability confirmed these results but also showed the need of numerical simulations to accurately quantify the overturn dynamics. Whenever IBC sink, the overturn occurs via small‐scale diapirs.

A melt inclusion study on volatile abundances in the lunar mantle

Abstract Earth’s Moon was thought to be highly depleted in volatiles due to its formation by a giant impact. Over the last decade, however, evidence has been found in apatites, lunar volcanic glass beads, nominally anhydrous minerals and olivine-hosted melt inclusions, to support a relatively “wet” Moon. In particular, based on H2O/Ce, F/Nd, and S/Dy ratios, recent melt inclusion (MI) work estimated volatile (H2O, F, and S) abundances in lunar rocks to be similar to or slightly lower than the terrestrial depleted mantle. Uncertainties still occur, however, in whether the limited numbers of lunar samples studied are representative of the primitive lunar mantle, and whether the high H2O/Ce ratio for pyroclastic sample 74220 is due to local heterogeneity. In this paper, we report major element, trace element, volatile, and transition metal data in MIs for 5 mare basalt samples (10020, 12040, 15016, 15647 and 74235) and a pyroclastic deposit (74220). With our new lunar MI data, H2O/Ce ratios are still found to vary significantly among different lunar samples, from ∼50 for 74220, to ∼9 for 10020, ∼3 for 74235, 1.7 to 0.9 for 12008, 15016, and 15647, and 0.5 for 12040. H2O/Ce ratios for these samples show positive correlation with their cooling rates, indicating a possible effect of post-eruptive loss of H on their H2O/Ce variations. It is evident that most other lab and lunar processes, including loss of H2O during homogenization, mantle partial melting, magma evolution, and ingassing during or post eruption are unlikely the causes of high H2O/Ce variations among different lunar samples. By comparing ratios of F/Nd, S/Dy, Zn/Fe, Pb/Ce, Cs/Rb, Rb/Ba, Cl/K, Na/Sr, Ga/Lu, K/Ba, and Li/Yb between 74220 and other lunar samples, the possibility of 74220 originating from a volatile-enriched heterogeneity in the lunar mantle can also be excluded. With all the above considerations, we think that the H2O/Ce ratio for 74220 best represents the pre-degassing lunar basaltic melt and primitive lunar mantle, either because it was formed by a rapid eruption process, or it was sourced from a deeper part of the lunar mantle that experienced less degassing H2O loss during lunar magma ocean crystallization. With an H2O/Ce ratio of ∼50, the primitive lunar mantle is estimated to contain ∼84 ppm H2O. Comparing volatile abundances in melt inclusions, glassy embayments, and glass beads in 74220 yields the following volatility trend for volcanic eruptions on the lunar surface: H2O ≫ Cl ≫ Zn ≈ Cu ≈ F > S ≈ Ga ≈ Cs > Rb ≈ Pb > Na > K ≈ Li. Using the melt inclusion data obtained thus far, the volatile depletion trend for the Moon from a MI perspective is estimated. Our results show that most of the volatile elements in the lunar mantle are depleted relative to the bulk silicate Earth by a factor of 2 to 20, however, a good correlation between half condensation temperature and the volatile depletion trend is not observed. The relatively flat pattern for the lunar volatile depletion trend requires a lunar formation model that can reconcile the abundances of these volatiles in the lunar mantle.

Crystallization of the lunar magma ocean and the primordial mantle-crust differentiation of the Moon

We present crystallization experiments on silicate melt compositions related to the lunar magma ocean (LMO) and its evolution with cooling. Our approach aims at constraining the primordial internal differentiation of the Moon into mantle and crust. We used graphite capsules in piston cylinder (1.35–0.80 GPa) and internally-heated pressure vessels (<0.50 GPa), over 1580–1020 °C, and produced melt compositions using a stepwise approach that reproduces fractional crystallization. Using our new experimental dataset, we define phase equilibria and equations predicting the saturation of liquidus phases, magma temperature, and crystal/melt partitioning for major elements relevant for the crystallization of the LMO. These empirical expressions are then used in a forward model that predicts the liquid line of descent and crystallization products of a 600 km-thick magma ocean. Our results show that the effects of changes in the bulk composition on the sequence of crystallization are minor. Our experiments also show the crystallization of a silica phase at ca. 1080 °C and we suggest that this phase might have contributed to the building of the lower anorthositic crust. Calculation of crustal thickness clearly shows that a thin crust similar to that revealed by GRAIL cannot have been generated through solidification of whole Moon magma ocean. We discuss the role of magma ocean depth, trapped liquid fraction (with implication for the alumina budget in the mantle and the crust), and the efficiency of plagioclase flotation in producing the thin crust. We also constrain the potential range of pyroxene compositions that could be incorporated into the crust and show that delayed crustal building during ca. 4% LMO crystallization on the nearside of the Moon may explain the dichotomy for Mg-number. Finally, we show that the LMO can produce magnesian anorthosites during the first stages of plagioclase crystallization.

Fractional crystallization of the lunar magma ocean: Updating the dominant paradigm

AbstractWe report results of systematic experimental simulation of fractional crystallization of a lunar magma ocean (LMO) with the Lunar Primitive Upper Mantle bulk composition. These results complement prior work that simulated equilibrium crystallization. In contrast to previous numerical models for investigating magma ocean solidification processes and implications, our combined program simulates these processes directly using petrologic experimentation. Our experiments mimic LMO crystallization that is fractional throughout the process, rather than switching from initially equilibrium to fractional crystallization partway through. To do this, we adopted an iterative approach in which the starting material for each run is synthesized using the composition of the melt phase from the prior run. We compare our results to those from long‐standing numerical models of LMO crystallization and show that although some features of those models are broadly reproduced, there are key differences in liquid lines of descent and the cumulate lithologies generated. Our results can be used to estimate the possible thickness of a primordial lunar crust formed from flotation of plagioclase during magma ocean solidification. Our estimate is greater than that from the recent Gravity Recovery and Interior Laboratory (GRAIL) mission, but consistent with the criteria on which the starting bulk composition was originally calculated. It assumes perfectly efficient separation of all plagioclase formed from the crystallizing magma ocean, which is likely not the case. We also demonstrate that a non‐chondritic bulk composition, with respect to trace elements, is not required in order to generate a KREEP (potassium, rare earth elements, and phosphorus) signature from magma ocean crystallization.

Volatile element loss during planetary magma ocean phases

Moderately volatile elements (MVE) are key tracers of volatile depletion in planetary bodies. Zinc is an especially useful MVE because of its generally elevated abundances in planetary basalts, relative to other MVE, and limited evidence for mass-dependent isotopic fractionation under high-temperature igneous processes. Compared with terrestrial basalts, which have δ66Zn values (per mille deviation of the 66Zn/64Zn ratio from the JMC-Lyon standard) similar to some chondrite meteorites (∼+0.3‰), lunar mare basalts yield a mean δ66Zn value of +1.4 ± 0.5‰ (2 st. dev.). Furthermore, mare basalts have average Zn concentrations ∼50 times lower than in typical terrestrial basaltic rocks. Late-stage lunar magmatic products, including ferroan anorthosite, Mg- and Alkali-suite rocks have even higher δ66Zn values (+3 to +6‰). Differences in Zn abundance and isotopic compositions between lunar and terrestrial rocks have previously been interpreted to reflect evaporative loss of Zn, either during the Earth–Moon forming Giant Impact, or in a lunar magma ocean (LMO) phase. To explore the mechanisms and processes under which volatile element loss may have occurred during a LMO phase, we developed models of Zn isotopic fractionation that are generally applicable to planetary magma oceans. Our objective was to identify conditions that would yield a δ66Zn signature of ∼+1.4‰ within the lunar mantle. For the sake of simplicity, we neglect possible Zn isotopic fractionation during the Giant Impact, and assumed a starting composition equal to the composition of the present-day terrestrial mantle, assuming both the Earth and Moon had zinc ‘consanguinity’ following their formation. We developed two models: the first simulates evaporative fractionation of Zn only prior to LMO mixing and crystallization; the second simulates continued evaporative fractionation of Zn that persists until ∼75% LMO crystallization. The first model yields a relatively homogenous bulk solid LMO δ66Zn value, while the second results in a stratification of δ66Zn values within the LMO sequence. Loss and/or isolation mechanisms for volatiles are critical to these models; hydrodynamic escape was not a dominant process, but loss of a nascent lunar atmosphere or separation of condensates into a proto-lunar crust are possible mechanisms by which volatiles could be separated from the lunar interior. The results do not preclude models that suggest a lunar volatile depletion episode related to the Giant Impact. Conversely, LMO models for volatile loss do not require loss of volatiles prior to lunar formation. Outgassing during planetary magma ocean phases likely played a profound role in setting the volatile inventories of planets, particularly for low mass bodies that experienced the greatest volatile loss. In turn, our results suggest that the initial compositions of planets that accreted from smaller, highly differentiated planetesimals were likely to be severely volatile depleted.