Bulk Partition Coefficients Research Articles

Fluorine behavior during experimental muscovite dehydration melting and natural partitioning between micas: Implications for the petrogenesis of peraluminous leucogranites and pegmatites

Abstract Fluorine behavior during the partial melting of two mica-bearing protoliths has been experimentally investigated at 700 to 930 °C and 0.4 and 0.6 GPa. Muscovite dehydration and H2O-HF fluid-assisted partial-melting experiments were carried out using both a natural and synthetic two-mica schist made of natural micas. The mineral composition of the experiments was assessed by BSE imaging and EDS analyses. The F, Cl, and major elements contents of the glass and micas were determined by EPMA. The muscovite dehydration melting reaction is muscovite + quartz + plagioclase = peraluminous melt + biotite + sillimanite + potassic feldspar ± hercynite. The starting biotite stays largely stable, showing only minor melt + ilmenite and trace magnetite formation in the cleavages. The newly formed biotite shows similar F contents and a slightly higher XSid component when compared to the starting biotite. HF-added experiments yield F-rich newly formed biotite. The experimentally produced melts were of a peraluminous leucogranitic composition with F contents increasing with F-rich protoliths. The bulk partition coefficient DFschist/melt increases from 0.5 to 3.0 when the F content of the protolith rises from 0.05 to 1.2 wt%. The partition coefficient, DFBt/melt, increases from 2.0 to 6.0 where the biotite MgO content increases from 5 to 18 wt%. The natural partition coefficient DFBt/Ms, measured for a set of rocks with a varied lithology from the Seridó Belt, northeastern Brazil, was 2.7 ± 0.5. The F partition coefficients measured in this study, along with published F partition coefficients between biotite and melt, biotite and muscovite, and fluid and melt, allow for the modeling of F behavior during muscovite dehydration and fluid-present melting. F-rich, two-mica protoliths will increase F partitioning in favor of the micaceous anatectic residue compared to the peraluminous melt. Furthermore, the model indicates that the more Fe-rich the schist and its residual biotite are, the higher the F content of the melt and the fluid. Fluorine-rich peraluminous leucogranites and related fluids may be generated by the anatexis of F- and Fe-rich, two-mica protoliths. As F can be a complexing ligand for Li, Be, Cs, Nb, Ta, W, Sn, and U, muscovite dehydration could potentially be associated with metallic occurrences associated with peraluminous melts.

RETRACTED: Fe3+ partitioning between clinopyroxene and silicate melt at 1–2.5 GPa: Implications for Fe3+ content of MORB and OIB source mantle

Pyroxene is the chief reservoir of Fe 3+ in upper mantle peridotite, but experiments exploring pyroxene/melt Fe 3+ partitioning ( D Fe3+ pyx/liq ) have been restricted to 100 kPa and pyroxene with low alumina. Here we present Fe 3+ partitioning experiments between clinopyroxenes (cpx) and mafic melt at elevated pressures (1–2.5 GPa). Experiments were conducted with f O2 buffered and modulated by Ru + RuO 2 and Fe-Pt alloy capsules, respectively, between ΔQFM −2.7 and +5.1. Fe 3+ /Fe T of both cpx and melt were determined by Fe K -edge X-ray absorption near edge structure spectroscopy. The experimentally synthesized cpx compositions (Al 2 O 3 = 2.36–6.01 wt.%, CaO = 19.33–22.21 wt.%) approximate those expected in basalt source regions. We find that Fe 3+ is moderately incompatible in cpx and D Fe3+ cpx/liq correlates with cpx Al 2 O 3 content, increasing from 0.05 ± 0.09 to 0.81 ± 0.04. D Fe3+ cpx/liq also increases with increasing f O2 . Comparison between experimentally synthesized cpx with those from natural peridotites indicates influences of both temperature and composition on Fe 3+ /Fe T for cpx in spinel and garnet peridotites. The combined effects of decreased pyroxene Al 2 O 3 concentration and pyroxene mode with progressive partial melting of peridotite diminishes the bulk partition coefficients of Fe 3+ , leading to greater Fe 2 O 3 contents in high degree partial melts, and this accounts for an inverse relationship between Na 2 O and Fe 2 O 3 observed in mid-ocean ridge basalts (MORB). Comparison to numerical experiments with pMELTS and the model of Jennings and Holland (2015) show that these models overpredict D Fe3+ cpx/liq for partial melting of the mantle, and so they do not accurately determine the relationship between the f O2 and Fe 2 O 3 of peridotite in basalt source regions. To estimate the Fe 3+ /Fe T ratio of the mantle source of MORB, we modeled liquid Fe 2 O 3 during isentropic batch melting of peridotite at three potential temperatures (1320 °C, 1400 °C, and 1440 °C) for peridotitic sources with Fe 3+ /Fe T ratios between 0.02–0.06. A source with an Fe 3+ /Fe T ratio of 0.038 ± 0.007 matches most of the span of natural MORB. This ratio is similar to that typical of continental lithospheric mantle sampled by xenoliths, but lower than that surmised by several recent experimental and thermodynamic studies. Considering this source Fe 3+ /Fe T but extending the partial melting calculations to higher pressures (2.5–4 GPa) reveals that bulk D Fe3+ perid/liq significantly decreases for garnet peridotite relative to spinel peridotite because the cpx become significantly less aluminous with increasing pressure. This results in high pressure partial melts with greater liquid Fe 3+ /Fe T ratios. Therefore, elevated Fe 3+ /Fe T ratios observed from some oceanic island basalts (OIB), such as those from Hawaii and Iceland, reflect in part the differences in conditions of melting and may not require mantle source regions more oxidized than those that produce MORB.

Origin of Compositional Gradients with Temperature in the High-SiO2 Rhyolite Portion of the Bishop Tuff: Constraints on Mineral–Melt–Fluid Reactions in the Parental Mush

Abstract The Bishop Tuff (BT), erupted from the Long Valley caldera in California, displays two types of geochemical gradients with temperature: one is related to magma mixing, whereas the other is found in the high-SiO2 rhyolite portion of the Bishop Tuff and is characterized by twofold or lower concentration variations in minor and trace elements that are strongly correlated with temperature. It is proposed that the latter zonation, which preceded phenocryst growth, developed as a result of mineral–melt partitioning between interstitial melt and surrounding crystals in a parental mush, from which variable melt fractions were segregated. To test this hypothesis, trends of increasing vs decreasing element concentrations with temperature (as a proxy for melt fraction), obtained from published data on single-clast pumice samples from the high-SiO2 rhyolite portion of the Bishop Tuff, were used to infer their relative degrees of incompatibility vs compatibility between crystals and melt in the parental mush. Relative compatibility values (RCVi) for all elements i, defined as the concentration slope with temperature divided by average concentration, are shown to be linearly correlated with their respective bulk partition coefficients (bulk Di). Mineral–melt partition coefficients from the literature were used to constrain the average stoichiometry of the crystallization/melting reaction in the parental mush: 32 % quartz + 34 % plagioclase + 31 % K-feldspar + 1·60 % biotite + 0·42 % titanomagnetite + 0·34 % ilmenite + 0·093 % allanite + 0·024 % zircon + 0·025 % apatite = 100 % liquid. The proportions of tectosilicates in the reaction (i.e. location of eutectic) are consistent with depths of melt segregation of ~400–550 MPa and an activity of H2O of ~0·4–0·6. Temperatures of <770–780 °C are constrained by allanite in the reaction. Evidence that a fluid phase was present in the parental mush is seen in the decreasing versus increasing H2O and CO2 contents with temperature in the segregated interstitial melt that formed the high-SiO2 rhyolite portion of the Bishop Tuff. The presence of an excess fluid phase, which strongly partitions CO2 relative to the melt, is required to explain the compatible behavior of CO2, whereas the fluid abundance must have been low to explain the incompatible behavior of H2O. Calculated degassing paths for interstitial melts, which segregated from the parental mush and ascended to shallower depths to grow phenocrysts, match published volatile analyses in quartz-hosted melt inclusions and constrain fluid abundances in the mush to be ≤1 wt%. The source of volatiles in the parental mush, irrespective of whether it formed by crystallization or partial melting, must have been primarily from associated basalts, as granitoid crust is too volatile poor. Approximately twice as much basalt as rhyolite is needed to provide the requisite volatiles. The determination of bulk Di for several elements gives the bulk composition of the parental leucogranitic mush and shows that it is distinct from Mesozoic Sierran arc granitoids, as expected. Collectively, the results from this study provide new constraints for models of the complex, multi-stage processes throughout the Plio-Quaternary, involving both mantle-derived basalt and pre-existing crust, that led to the origin of the parental body to the Bishop Tuff.

Simultaneous Partition Experiment of Divalent Metal Ions between Sphalerite and 1 mol/L (Ni, Mg, Co, Fe, Mn)Cl2 Aqueous Solution under Supercritical Conditions

A simultaneous partition experiment of divalent metal ions was performed between sphalerite and 1 mol/L (Ni, Mg, Co, Fe, Mn)Cl2 aqueous solution under supercritical hydrothermal conditions of 500–800 °C and 100 MPa. The bulk partition coefficient that was defined by KPB(ZnS) = (xMeS/xZnS)/(mMeaq/mZnaq) followed the order of Zn ≑ Co ≑ Ni > Fe > Mn > Mg at all temperatures. In the partition coefficient versus ionic radius (PC–IR) diagrams with the logarithmic value of the bulk partition coefficient (log KPB(ZnS)) on the vertical axis, and the ionic radius of the six-fold coordinated site on the horizontal axis, Ni shows a positive partition anomaly, and the other elements were almost on the PC–IR curve. Based on the PC–IR curve, the optimum ionic radius for sphalerite existed where the ionic radius was slightly larger than Zn (~0.76 Å). A Ni positive partition anomaly may result from its large electronegativity.

Trace element fractionation and isotope ratio variation during melting of a spatially distributed and lithologically heterogeneous mantle

The source region of basalts in the mantle is chemically and lithologically heterogeneous. During decompression melting of a spatially distributed and lithologically heterogeneous mantle, mineral modes of distinct mantle sources vary continuously, resulting in spatial and temporal variations in the bulk partition coefficient of a trace element in different lithologies in the melting column, which in turn affects the fractionation of the trace element in partial melt and residual solid. This problem can be quantified by following the motion of solid in the melting column. This study presents a new melting model that can be used to keep track of spatial and temporal variations of mineral mode, melting reaction, bulk partition coefficient, and trace element concentration in the lithologically heterogeneous melting column. Simple analytical solutions for a time-dependent perfect fractional melting model are obtained. Essential features of the new model are elucidated through case studies of melting a two-lithology mantle that consists of blobs of orthopyroxene-rich lithology in the upwelling lherzolitic mantle; and an application to Sr-Nd-Hf isotope ratio variations in basalts from the Mid-Atlantic Ridge is presented. Fractional melting of the two-lithology mantle results in large temporal variations in incompatible trace element concentrations and Sr-Nd-Hf isotope ratios in the pooled melt. Mixing of fractional melts derived from different lithologies in the melting column produces enriched and depleted melts that form mixing loops in Sr-Nd-Hf isotope ratio correlation diagrams. These mixing loops rotate away from mixing lines defined by the binary mixing model and are a unique feature of melting a spatially heterogeneous mantle. Formation of the mixing loop can be traced to the location and spacing of the enriched lithological units in the melting column. The role of lithological heterogeneity is to change the bulk partition coefficient of a trace element from its values in the lherzolitic mantle to new values in the pyroxenitic mantle, which alters the extent of depletion of the trace element in the melting column. Changing bulk partition coefficient with time and space through lithological heterogeneity can result in greater variabilities in Sr-Nd-Hf isotope ratios and highly incompatible trace element concentrations in the pooled melt. Results from this study establish a framework for systematic studies of trace element fractionation and isotope ratio variation during decompression melting of a spatially distributed and lithologically heterogeneous mantle.

Molecular Dynamics Simulations of Aqueous Nonionic Surfactants on a Carbonate Surface

The interactions and structure of secondary alcohol ethoxylates with 15 and 40 ethoxylate units in water near a calcite surface are studied. It is found that water binds preferentially to the calcite surface. Prediction of the free-energy landscape for surfactant molecules shows that single-surfactant molecules do not adsorb because they cannot get close enough to the surface because of the water layer for attractive ethoxylate-calcite or dispersion interactions to be significant. Micelles can adsorb onto the surface even with the intervening water layer because of the integrative effect of the attractive interactions of all the surfactant molecules. Adsorption is found to increase because of the closer proximity of the micelles to the surface due to a weakened water layer at higher temperatures. The free-energy well and barrier values are used to estimate surface to bulk partition coefficients for different surfactants and temperatures, and qualitative agreement is found with experimental observations. The combined effect of surfactant-water and surfactant-solid interactions is found to be responsible for an increased adsorption for nonionic surfactants as the system approaches the cloud point.

Simultaneous Partitioning of Divalent Metal Ions between Alabandite and 1 mol/L (Ni, Mg, Co, Zn, Fe)Cl2 Aqueous Solutions under Supercritical Conditions

To clarify the element partitioning behavior between minerals and aqueous chloride solutions, we conducted experiments to investigate simultaneous partitioning of Ni2+, Mg2+, Co2+, Zn2+, Fe2+, and Mn2+ ions between alabandite (MnS) and 1 mol/L (Ni, Mg, Co, Zn, Fe)Cl2 aqueous solutions at 500–800 °C and 100 MPa. The bulk partition coefficients calculated using the following equation were in the order of Fe2+ > Co2+ > Ni2+ ≈ Zn2+ > Mn2+ >> Mg2+; KPN = (xMeS/mMeaq)/(xMnS/mMnaq). A partition coefficient-ionic radius (PC-IR) curve was plotted with the logarithmic value of the partition coefficient on the y-axis and the ionic radius at the six-fold coordinated site on the x-axis. The peak of this curve was located near the ionic radius of Fe2+ and not near the ionic radius of Mn2+. Zn2+ showed a slight negative partitioning anomaly, which increased in the order of sulfide minerals < arsenic sulfide minerals < arsenide minerals as the covalent bond became stronger. Ni2+ showed a positive partitioning anomaly, which indicated that it preferred an octahedral structure. The width of the PC-IR curve decreased in the order of sulfide minerals > arsenic sulfide minerals > arsenide minerals as the covalent bond became stronger, that is, the ion selectivity became stronger.

The almost lithophile character of nitrogen during core formation

Nitrogen is a key constituent of our atmosphere and forms the basis of life, but its early distribution between Earth reservoirs is not well constrained. We investigate nitrogen partitioning between metal and silicate melts over a wide range of conditions relevant for core segregation during Earth accretion, i.e. 1250–2000°C, 1.5–5.5 GPa and oxygen fugacities of ΔIW-5.9 to ΔIW-1.4 (in log units relative to the iron–wüstite buffer).At 1250°C, 1.5 GPa, DNmetal melt/silicate melt ranges from 14 ± 0.1 at ΔIW-1.4 to 2.0 ± 0.2 at ΔIW-5, N partitioning into the core forming metal. Increasing pressure has no effect on DNmetal melt/silicate melt, while increasing temperature dramatically lowers DNmetal melt/silicate melt to 0.5 ± 0.15 at ΔIW-4. During early core formation N was hence mildly incompatible in the metal. The partitioning data are then parameterised as a function of temperature and oxygen fugacity and used to model the evolution of N within the two early prevailing reservoirs: the silicate magma ocean and the core. Depending on the oxidation state during accretion, N either behaves lithophile or siderophile. For the most widely favoured initially reduced Earth accretion scenario, N behaves lithophile with a bulk partition coefficient of 0.17 to 1.4, leading to 500–700 ppm N in closed-system core formation models. However, core formation from a magma ocean is very likely accompanied by magma ocean degassing, the core would thus contain ≤100 ppm of N, and hence, does not constitute the missing N reservoir. Bulk Earth N would thus be 34–180 ppm in the absence of other suitable reservoirs, >98% N of the chondritic N have hence been lost during accretion.

In situ Raman spectroscopic study of nitrogen speciation in aqueous fluids under pressure

Nitrogen speciation in aqueous fluids is important for understanding the partitioning of N between fluid and coexisting melt or mineral phases and the storage and recycling of nitrogen in Earth's interior. Previously N speciation in aqueous fluids has only been investigated using the approach of quenched fluid inclusions or thermodynamic modeling. Here we present the results from in situ Raman spectroscopic measurements of N speciation in aqueous fluids held in hydrothermal diamond anvil cell, at P-T conditions up to above 800 °C and 2 GPa. We identify the presence of N in aqueous fluids as N2 and NH3 (replaced by NH4+ at low pH) with N2 being the favored species toward higher temperature. The fugacity equilibrium constant for reaction N2 + 3 H2O = 2 NH3 + 3/2 O2 is determined to be lnKf = −16.15–23,489/T with T being temperature in Kelvin. Our equilibrium constant is significantly lower than that from quenched fluid inclusions, but is in good agreement with that calculated from the Deep Earth Water (DEW) model. We suggest that the thermodynamic stability of N2 relative to NH3 in aqueous fluids has been underestimated by quench experiments. Because N is stored in silicate minerals and melts mainly as NH4+, the bulk partition coefficient of N between fluid and mineral or melt should be greater than previously thought if one assumes a fixed partition coefficient for trivalent N. For subduction zones this means that a higher fraction of N is recycled by slab-derived fluids, and concomitantly less N is carried by subducting slabs to Earth's deeper interior.

Parental Magma Composition of the Main Zone of the Bushveld Complex: Evidence fromin situLA-ICP-MS Trace Element Analysis of Silicate Minerals in the Cumulate Rocks

In situ trace element analysis of cumulus minerals may provide a clue to the parental magma from which the minerals crystallized. However, this is hampered by effects of the trapped liquid shift (TLS). In the Main Zone (MZ) of the Bushveld Complex, the Ti content in plagioclase grains shows a clear increase from core to rim, whereas most other elements [e.g. rare earth elements (REE), Zr, Hf, Pb] do not. This is different from the prominent intra-grain variation of all trace elements in silicate minerals in mafic dikes, which have a faster cooling rate. We suggest that crystal fractionation of trapped liquid occurred in the MZ of Bushveld and the TLS may have modified the original composition of the cumulus minerals for most trace elements except Ti during slow cooling. Quantitative model calculations suggest that the influence of the TLS depends on the bulk partition coefficient of the element. The effect on highly incompatible elements is clearly more prominent than that on moderately incompatible and compatible elements because of different concentration gradients between cores and rims of cumulate minerals. This is supported by the following observations in the MZ of Bushveld: (1) positive correlation between Cr, Ni and Mg# of clinopyroxene and orthopyroxene; (2) negative correlation between moderately incompatible elements (e.g. Mn and Sc in clinopyroxene and orthopyroxene; Sr, Ba and Eu in plagioclase); but (3) poor correlation between highly incompatible elements and Mg# of clinopyroxene and orthopyroxene or An# of plagioclase. Modeling suggests that the extent of the TLS for a trace element is also dependent on the initial fraction of the primary trapped liquid, with strong TLS occurring if the primary trapped liquid fraction is high. This is supported by the positive correlation between highly incompatible trace element abundances in cumulus minerals and whole-rock Zr contents. We have calculated the composition of the parental magma of the MZ of the Bushveld Complex. The compatible and moderately incompatible element contents of the calculated parental liquid are generally similar to those of the B3 marginal rocks, but different from those of the B1 and B2 marginal rocks. For the highly incompatible elements, we suggest that the use of the sample with the lowest whole-rock Zr content and the least degree of TLS is the best approach to obtain the parental magma composition. The heavy REE contents of the magma calculated from orthopyroxene are similar to those of B3 rocks and lower than those of B2 rocks. The calculated REE contents from clinopyroxene are generally significantly higher than for B2 or B3 rocks, and those from plagioclase are in the lower level of B2, but slightly higher than for B3. However, the calculated REE patterns for both clinopyroxene and plagioclase show strong negative Eu anomalies, which are at the lower level of the B2 field and within the B3 field, respectively. We suggest that Eu may be less affected by TLS than other REE owing to its higher bulk compatibility. Based on this and the fact that the calculated REE contents of the parental magma should be higher than the real magma composition owing to some degree of crystal fractionation and TLS, even for the sample with the lowest amount of trapped liquid, we propose that a B3 type liquid is the most likely parental magma to the MZ of the Bushveld Complex. In the lowermost part of the MZ, there is involvement of the Upper Critical Zone (UCZ) magma.

Molybdenum partitioning behavior and content in the depleted mantle: Insights from Balmuccia and Baldissero mantle tectonites (Ivrea Zone, Italian Alps)

Constant concentration ratios of trace elements of similar incompatibility in oceanic basalts have been used to determine the abundances and ratios of incompatible elements in mantle sources. However, the Mo content in the depleted mantle estimated from Mo/Ce of oceanic basalts is much lower than the estimates based on Mo contents of mantle peridotites (25 ± 7 ng/g versus 113–180 ng/g). This discrepancy partly reflects uncertainties about the geochemical behavior of Mo in the upper mantle. New Mo concentration data obtained on unserpentinized spinel-facies lherzolites, harzburgites, dunites and pyroxenites (n = 47) from the Balmuccia and Baldissero mantle tectonites (Ivrea-Verbano Zone, Italian Alps) provide new insights into the magmatic behavior of Mo and its content in the depleted mantle.The peridotites and pyroxenites display variable and very low Mo contents (4–16 and 3–11 ng/g, respectively). A slightly serpentinized peridotite shows the highest Mo content of 37 ng/g. The Mo contents show no systematic variation with other incompatible or compatible elements. Chromium spinel in the mantle rocks likely does not control the behavior of Mo because complete or incomplete digestion of spinel did not lead to a noticeable change in Mo contents. The data indicate incorporation of little Mo into olivine, pyroxene and spinel, consistent with the predominant occurrence of Mo6+ and the experimentally-determined very low partition coefficients at typical upper mantle conditions. Bulk rock Mo contents from these mantle rocks, upper limits of Mo contents in sulfides and the lack of variations of Mo concentrations with varying sulfide fractions indicate a minor influence of mantle sulfides on the bulk Mo budget. The incorporation of little Mo into pyroxenes of the mantle rocks contrasts with Ce which is mainly controlled by clinopyroxene. This contrasting behavior suggests a higher incompatibility of Mo than Ce during mantle melting. Modelling polybaric melting of mantle peridotites indicates that melts formed at low degrees of partial melting (e.g., <2–3%) would not retain the Mo/Ce of the mantle sources. In contrast, >5% melting, as in tholeiites and komatiites, leads to extraction of nearly the complete inventory of Mo and Ce into the melts and thus Mo/Ce of the melts should be representative of mean mantle source compositions, even if Mo and Ce show different bulk partition coefficients. Molybdenum contents of mantle rocks from the Ivrea Zone and the data on oceanic basalts and komatiites indicate that the depleted upper mantle has a low Mo content of <30 ng/g.

Calc-alkaline magmatism associated with salt diapirs in the Shurab and Garmsar back-arc areas (Central Basin, Iran): magma genesis and tectonic implications

Medium- and high-K calc-alkaline magmatism of the Shurab (southeast Qom city) and Garmsar (northwest Garmsar city) areas occurred within the Lower Red Formation in the Central Basin behind the Urumieh-Dokthar Magmatic Arc. In terms of whole-rock geochemical analyses and in agreement with the petrographic features, all representative samples of the Shurab area are classified into three groups: group 1 with mainly intergranular texture comprises basalt/trachybasalt, while groups 2 and 3 with trachytic and porphyritic textures, respectively, have basaltic trachyandesite composition. The overall major constituents are plagioclase with composition in the range from An 49 Ab 23 to An 75 Ab 47 , clinopyroxene with composition in the range Wo 43-45 En 39-45 Fs 9-15 , and olivine with composition in the range from Fo 58 Fa 31 to Fo 67 Fa 40 . Minor minerals consist of opaque minerals and K-feldspar in the range Or 41-65 Ab 33-50 An 0.69-8 . The characteristic accessories are apatite and sphene. In the Garmsar area, rocks are seen as subvolcanic with mafic (basalt) and intermediate (trachybasalt/basaltic trachyandesite) compositions. The Garmsar area rocks represent intergranular, granular, ophitic, and subophitic textures. In these rocks, major mineral phases are plagioclase and clinopyroxene. The minor constituents are olivine, opaque minerals, amphibole, biotite, and quartz. Apatite is the most important accessory mineral. The rocks of both areas display REE patterns characterized by LREE-enriched and HREE-depleted segments typical of arc lavas. Primitive mantle-normalized trace element patterns for samples of both areas exhibit high ratios of strongly incompatible elements with similar bulk partition coefficients (e.g., Th/ Ta and Th/Ce), enrichment in large-ion lithophile elements (LILEs: Cs, Ba, Rb, Th) relative to the high field-strength elements (HFSEs: Ti, Hf, Zr, and REEs), and troughs for Nb, Ta, Ti, and Zr and peaks for Cs, Th, K, and Sr, all of which are indicators for subductionrelated magmatism. Subduction of the Neo-Tethys beneath the Eurasian margin led to upper mantle deformation and metasomatism. Once the Arabian plate collided with the Eurasian margin, subduction ended through a slab breakoff process, and thermal flux of asthenospheric origin uprising through the slab tear induced the thermal erosion of the mantle metasomatized during the previous subduction event and triggered its partial melting. Also, the late Eocene-early Oligocene collision of Eurasian with Arabian plates led to the subsidence and formation of faults and extensions in the Central Basin (i.e. the Shurab and Garmsar areas) such that eruption of medium- and high-K metasomatic magmatism along these faults and extensions caused postcollision volcanism in the Central Basin.

The behaviour of ferric iron during partial melting of peridotite

In order to better understand the behaviour of Fe3+ during partial melting of a mid-ocean-ridge-basalt (MORB) mantle source, we performed partial melting experiments of a spinel peridotite at 1.5 GPa, between 1320 and 1450 °C, and over a range of oxygen fugacities (fO2) varying from FMQ-3.5 to FMQ+6 (log-units relative to FMQ = Fayalite-Magnetite-Quartz oxygen buffer). Changes in fO2 were achieved using different capsule materials and by varying the oxidation state of the starting material. Fe3+ contents of spinels and glasses were examined using Fe K-edge X-ray Absorption Near Edge Structure (XANES) spectroscopy in full-field mode. Our results show that this method allows Fe3+/ƩFe measurements to be made of spinels in our experimental charges but is not suitable for quantifying Fe3+/ƩFe of glasses, particularly at relatively low fO2 where melt Fe2O3 contents are small. We therefore employed several methods (Fe3+/ƩFe ratios of spinels measured by XANES and calculated by stoichiometry, and Fe-alloy sliding redox sensors) to assess the fO2 of the samples. These were then used to calculate the Fe3+/ƩFe ratios of glasses using published models, which were further constrained by Mössbauer spectroscopy measurements on an additional set of partial melting experiments.For a particular initial redox state, the Fe3+/ΣFe ratio in melts remains approximately constant with partial melting degree. This is explained by a slight decrease of the bulk partition coefficient of Fe2O3 (DFe2O3peridotite-melt) as the degree of partial melting increases. The Fe3+/ΣFe ratios of resulting melts are thus restricted to a relatively narrow range over large degrees of melting, as observed for global MORB glass analyses. This observation is also confirmed by pMELTS calculations employing different bulk Fe3+/ΣFe ratios. We conclude that the Fe2O3 content in MORB can be accounted for through partial melting of a mantle source containing 0.1–0.5 wt% Fe2O3 and a bulk DFe2O3peridotite-melt ranging from 0.3 to 0.1.

Generation of continental adakitic rocks: Crystallization modeling with variable bulk partition coefficients

The geochemical signatures (i.e., high Sr/Y and La/Yb ratios) of adakitic rocks in continental settings, which are derived from the continental lower crust rather than from subducted slabs, may reflect high-pressure melting in the lower crust or may be inherited from their sources. The North China Craton (NCC) is an ideal place for investigation of this type of adakites due to its ubiquitous distribution. As an example, we explore the petrogenesis of the Jurassic (~163Ma) adakitic rocks in western Liaoning, in the NE part of the NCC, using elemental and Sr–Nd isotopic analysis and crystallization modeling based on Rhyolite-MELTS. The modeling demonstrates that adakitic signatures can be generated by fractional crystallization of magmas within crust of normal thickness (i.e., 33km). Partial-melting modeling based on the composition of the lower continental crust shows that only the adakitic rocks from orogenic belts require a thickened crust (i.e., 45km). We suggest that continental adakitic rocks are not necessarily linked to high-pressure processes and their use as an indicator of thickened/delaminated continental crust should be regarded with caution.

Mobility of major and trace elements in the eclogite-fluid system and element fluxes upon slab dehydration

The equilibrium between aqueous fluids and allanite-bearing eclogite has been investigated to constrain the effect of temperature (T) and fluid composition on the stability of allanite and on the mobility of major and trace elements during the dehydration of eclogites. The experiments were performed at 590–800°C and 2.4–2.6GPa, and fluids were sampled as synthetic fluid inclusions in quartz using an improved entrapment technique. The concentrations and bulk partition coefficients were determined for a range of major (Mg, Ca, Na, Fe, Al, Ti) and 16 trace elements as a function of T and fluid composition. The results reveal a significant effect of T on element partitioning between the fluids and the solid mineral assemblage. The partition coefficients increase by more than an order of magnitude for most of the major and trace elements, and several orders of magnitude for light rare-earth elements (LREE) from 590 to 800°C. The addition of various ligand species into the fluid at 700°C results in distinctive trends on element partitioning. The concentrations and corresponding partition coefficients of most of the elements are enhanced upon addition of NaF to the fluid. In contrast, NaCl displays a nearly opposite effect by suppressing the solubilities of major elements and consequently affecting the mobility of trace elements that form stable complexes with alkali-(alumino)-silicate clusters in the fluid, e.g. high field strength elements (HFSE). The results further suggest that fluids in equilibrium with orthopyroxene and/or diopsidic clinopyroxene are peralkaline (ASI ∼0.1–0.7), whereas fluids in equilibrium with omphacitic pyroxene are more peraluminous (ASI ∼1.15). Therefore, natural aqueous fluids in equilibrium with eclogite at about 90km depth will be slightly peraluminous in composition. Another important finding of this study is the relatively high capacity of aqueous fluids to mobilize LREE, which may be even higher than that of hydrous melts.

Source components and magmatic processes in the genesis of Miocene to Quaternary lavas in western Turkey: constraints from HSE distribution and Hf–Pb–Os isotopes

Hf–Pb–Os isotope compositions and highly siderophile element (HSE) abundance variations are used to evaluate the mantle source characteristics and possible effects of differentiation processes in lavas from western Turkey, where the eruption of Late Miocene to Quaternary OIB-type intraplate mafic alkaline lavas followed pre-Middle Miocene convergent margin-type volcanism. Concentrations of Os, Ir, and Ru (IPGE) in the OIB-type intraplate lavas decrease with fractionation for primitive melts (MgO > 10 wt%), suggesting that these elements reside predominantly in olivine and associated HSE retaining trace phases and behave compatibly during olivine-dominated fractionation. Fractional crystallization trends indicate distinctly lower bulk partition coefficients for IPGE in more evolved lavas, possibly reflecting a change in the fractionating assemblages. Pd and Re in the primitive melts display negative correlations with MgO, demonstrating moderately incompatible behavior of these elements during fractionation, while the significantly scattered variation in Pt against MgO may indicate the effects of micronuggets of a Pt-rich alloy. Os-rich alkaline primary lavas (>50 ppt Os) exhibit a limited range of 187Os/188Os (0.1361–0.1404), with some xenolith-bearing lavas displaying depletions in 187Os/188Os (0.1131–0.1232), suggesting slight compositional modification of primitive melts through contamination with highly depleted, Os-rich mantle lithosphere. More radiogenic Os isotope ratios (187Os/188Os > 0.1954) in the evolved lavas reflect contamination of the magmas by high187Os/188Os crustal material during shallow differentiation. The OIB-type lavas show limited variations in Hf and Pb isotopes with 176Hf/177Hf = 0.282941–0.283051, 206Pb/204Pb = 18.683–19.091, 207Pb/204Pb = 15.579–15.646, 208Pb/204Pb = 38.550–38.993; 176Hf/177Hf ratios correlate negatively with 208Pb*/206Pb*, suggesting the effects of similar mantle processes on the evolution of time-integrated Th/U and Lu/Hf. These lavas have distinctly higher 176Hf/177Hf and lower 208Pb*/206Pb* than the Early–Middle Miocene lavas of the region, which are interpreted as melts of enriched mantle with an overprint by sediment-derived subduction component. The source region for the OIB-type alkaline melts is interpreted to be a sub-lithospheric reservoir enriched in Hf and Pb isotopes with respect to depleted MORB mantle. Combined evaluation of Hf, Pb, and Os isotopes suggests that the relative enrichment in this domain is related to mixing of ancient oceanic crust with the ambient mantle through long-term plate recycling processes.

Highly siderophile element depletion in the Moon

Coupled 187Os/188Os and highly siderophile element (HSE: Os, Ir, Ru, Pt, Pd, Re) abundance data are reported for Apollo 12 (12005, 12009, 12019, 12022, 12038, 12039, 12040), Apollo 15 (15555) and Apollo 17 (70135) mare basalts, along with mare basalt meteorites La Paz icefield (LAP) 04841 and Miller Range (MIL) 05035. These mare basalts have consistently low HSE abundances, at ∼2×10−5 to 2×10−7 the chondritic abundance. The most magnesian samples have broadly chondrite-relative HSE abundances and chondritic measured and calculated initial 187Os/188Os. The lower abundances and fractionated HSE compositions of more evolved mare basalts can be reproduced by modeling crystal–liquid fractionation using rock/melt bulk-partition coefficients of ∼2 for Os, Ir, Ru, Pt and Pd and ∼1.5 for Re. Lunar mare basalt bulk-partition coefficients are probably higher than for terrestrial melts as a result of more reducing conditions, leading to increased HSE compatibility. The chondritic-relative abundances and chondritic 187Os/188Os of the most primitive high-MgO mare basalts cannot readily be explained through regolith contamination during emplacement at the lunar surface. Mare basalt compositions are best modeled as representing ∼5–11% partial melting of metal-free sources with low Os, Ir, Ru, Pd (∼0.1 ng g−1), Pt (∼0.2 ng g−1), Re (∼0.01 ng g−1) and S (∼75 μg g−1), with sulphide-melt partitioning between 1000 and 10,000.Apollo 12 olivine-, pigeonite- and ilmenite normative mare basalts define an imprecise 187Re–187Os age of 3.0±0.9 Ga with an initial 187Os/188Os of 0.107±0.010. This age is within uncertainty of 147Sm–143Nd ages for the samples. The initial Os isotopic composition of Apollo 12 samples indicates that the source of these rocks evolved with Re/Os within ∼10% of chondrite meteorites, from the time that the mantle source became a system closed to siderophile additions, to the time that the basalts erupted. Similarity in absolute HSE abundances between mare basalts from the Apollo 12, 15 and 17 sites, and from unknown regions of the Moon (La Paz mare basalts, MIL 05035), indicates relatively homogeneous and low HSE abundances within the lunar interior. Low absolute HSE abundances and chondritic Re/Os of mare basalts are consistent with a late accretion addition of ∼0.02 wt.% of the Moon's mass to the mantle, prior to the formation of the lunar crust. Late accretion must also have occurred significantly prior to cessation of lunar mantle differentiation (>4.4 Ga), to enable efficient mixing and homogenization within the mantle. Low lunar HSE abundances are consistent with proportionally 40 times more late accretion to Earth than the Moon. Disproportional late accretion to the two bodies is consistent with the small 182W excess (∼21–28 ppm) measured in lunar rocks, compared to the silicate Earth.

Comparative geochemistry of rhenium in oxidized arc magmas and MORB and rhenium partitioning during magmatic differentiation

The geochemical behavior of Re in oxidized arc magmas (Eastern Manus basin) and MORB has been reexamined. In the early differentiation stage of oxidized arc magmas, Yb, Cu, Au, Ag, and Re have a bulk partition coefficient (D) between the crystallized mineral assemblage (olivine, clinopyroxene, and plagioclase) and the residual melt on the order of 0.33=DYb~DAu>DRe>DAg>DCu, while in the late stage they have a bulk D between the crystallized mineral assemblage (clinopyroxene, plagioclase, magnetite, and monosulfide solid solution — MSS) and the residual melt on the order DRe>DCu>DAu≥1>DYb=0.29>DAg. In oxidized arc magmas, Cu, Au, and Ag are primarily controlled by crystalline MSS, whereas Re is primarily controlled by magnetite and it has a MSS/silicate melt partition coefficient significantly lower than 20. Unlike oxidized arc magmas, most MORB are saturated with sulfide liquid. Based on the positive correlation of log (DCu/DRe) and ∆FMQ regardless of fS2, log (DCu/DRe)=0.997ΔFMQ+0.181 (R2=0.99) (D=partition coefficient between sulfide liquid and silicate melt; FMQ=fayalite–magnetite–quartz oxygen buffer), the partition coefficient of Re between sulfide liquid and silicate melt for MORB is estimated to be in the range of 600–10,000 at fO2 of FMQ to FMQ−1, increasing with decreasing fO2. These partitioning data constrain that during MORB differentiation Re may be as compatible as Cu or Ag at relatively oxidized conditions (FMQ−0.5<fO2<FMQ) or as Au at relatively reduced conditions (FMQ−1<fO2<FMQ−0.5), and its concentration in MORB maybe expected to decrease with decreasing MgO in a similar manner as Cu, Ag, or Au. Yb is an incompatible element during MORB differentiation. Consequently, during MORB differentiation Yb/Re ratio may be expected to increase with decreasing MgO. However, the previously observed nearly constant Yb/Re ratio in a suite of MORB indicates similar incompatibility of Yb and Re and thus a minor role of sulfide liquid in controlling the geochemical behavior of Re during MORB differentiation. More work is required to resolve this paradox.

Contamination of MORB by anatexis of magma chamber roof rocks: Constraints from a geochemical study of experimental melts and associated residues

Mid-ocean ridge basalts (MORBs) are the most abundant magmas produced on Earth. They are widely studied to infer mantle compositions and melting processes. However, MORB liquids are also the complex end-product of a variety of intra-crustal processes such as partial or fractional crystallization, melt–rock interaction, and contamination. Deciphering the relative contribution of these different processes is of first-order importance. Contamination at ocean crustal levels is likely, and may occur at magma chamber margins where fresh magmas can interact with previously hydrothermally altered rocks. Characterizing the composition of this crustal contaminant component is critical if we are to understand the relative importance of each component in the resulting MORB liquid. Here we present the results of experiments designed to reproduce the processes occurring at oceanic magma chamber roofs, where crustal contamination should be most extensive, by melting a representative sample of the sheeted dike complex. Anatectic melts thus produced are likely to represent the principal crustal contaminant in MORB. These melts were characterized for major and trace elements, showing B, Zr, Hf, and U enrichment, and Sr, Ti, and V depletion relative to original MORB liquids. In comparison to the starting material, relative element fractionations are observed in the anatectic melts, with enrichments of: U relative to Ba, Nb, and Th; LREE and MREE relative to Sr; and Zr–Hf relative to LREE. Bulk partition coefficients for element partitioning during magma chamber roof anatexis are derived and proposed as valuable tools for tracking MORB contamination. Comparison with natural samples from the East Pacific Rise and the Oman ophiolite shows that anatectic melts can crystallize in situ to form oceanic plagiogranite intrusions, and that residual assemblages associated with the hydrous partial melting stage are represented by hornfelsic dikes and enclaves (also named granoblastic basalts). We now recognize these as commonplace at the root of the sheeted dike complex both at present-day and fossil oceanic spreading centers.

Partition coefficients of chalcophile elements between sulfide and silicate melts and the early crystallization history of sulfide liquid: LA-ICP-MS analysis of MORB sulfide droplets

Chalcophile element partitioning among base-metal sulfide, oxide and silicate phases during magmatic processes is poorly constrained in part because there are very few studies of reliable sulfide melt–silicate melt partition coefficients (Ds) and because the crystallization history of the sulfide liquid is ignored in most studies. Here we present laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) analyses of sulfide droplets and their host fresh mid-ocean-ridge basalt (MORB) glasses. The compositions of the sulfide droplets suggest that they formed in magma chambers beneath the mid-ocean ridges and equilibrated with their host silicate melt, enabling us to calculate new Ds for chalcophile elements. These are Co (45±4.5), Ni (776±98), Cu (1334±210), Zn (3.5±0.9), Se (345±37), Ag (1138±245), Cd (107±47), Sn (11±1.6), Te (4478±1146), Pb (57±10) and Bi (316±38). The contents of the highly siderophile elements in the glasses were too low to be determined using LA-ICP-MS. The whole rock values were used as a proxy for the glass and these allow an estimate of minimum Ds in the 10000 to 40000 range for the platinum-group elements (PGE) and 870 for Re. Partition coefficient values are affected by oxygen fugacity; comparison of our values with those from experimental studies suggests that oxygen fugacities of the MORBs considered here were between FMQ and FMQ-1. From the determined D values we calculate the contribution of the sulfides to bulk partition coefficient during mantle melting. Considering these values in combination with what is known about the behavior of the chalcophile elements during mantle partial melting, we suggest that Co, Ni, Zn, Te, PGE and Au behave as compatible elements during mantle melting, with Ni, Co and Zn being controlled mainly by the silicate and oxide minerals and the PGE, Au and Te being controlled mainly by the sulfides (or other discrete metallic phases). Copper, Se, Ag, Cd, Sn, Re, Pb and Bi are slightly to strongly incompatible during mantle melting. MORB sulfide droplets consist mainly of monosulfide solid solution (MSS), which is the first mineral to crystallize from a sulfide liquid, and intermediate solid solution (ISS), which crystallizes from the remaining liquid. The distribution of the chalcophile elements within the sulfide droplets shows that Co and Re have a slight preference for MSS, whereas the Cu, Zn, Ag, Cd, Sn, Te, Au, Bi and Pb partition into ISS. Selenium is present in approximately equal amounts in both MSS and ISS, and Pt and Pd are also present in both phases. Previous experimental and empirical studies have shown that Os, Ir, Ru and Rh partition into MSS and a portion of these elements exsolve from the sulfides as platinum-group minerals (PGM) during cooling. The same studies show that most of the Pt and Pd partition into the liquid and eventually crystallize from the late fractionated liquid as PGM. We examined our MORB sulfide droplets closely for platinum group-minerals (PGM) but none were found. We suggest that because the rocks sustained rapid cooling PGM were unable to exsolve from the sulfide minerals and the liquid did not fractionate sufficiently to permit the crystallization of the Pt or Pd minerals. Thus the PGE are present in MSS and ISS.