We investigated geothermal gases from Homa Hills, a carbonatitic complex situated along an adjacent branch of the Kenyan rift system, using neon, argon, krypton, xenon and nitrogen isotopes. Large quantities of gas were sampled in Giggenbach-type bottles (Giggenbach, 1975) and analyzed by dynamic mass spectrometry to resolve isotopic variations at high precision (0.01-0.1 parts per thousand; Seltzer and Bekaert, 2022; Bekaert et al., 2023; 2024). Neon and nitrogen isotope compositions are consistent with parental magmas being derived from the convecting mantle. Xenon isotopic data present ubiquitous enrichments (relative to air) of 129Xe from the decay of extinct 129I (T1/2 = 15.7 Myr) and 131-136Xef from fissions of 238U (T1/2 = 4.468 Myr) and/or 244Pu (T1/2 = 82 Myr). We also find slight excesses of 128Xe (relative to 130Xe and air), which could be due to subsurface isotopic fractionation during e.g., diffusive transport fractionation (DTF) and gravitational settling. However, the 128Xe excesses are not accompanied by correlated Kr isotope excesses and plot off the empirical fractionation line defined from several other locations worldwide (Bekaert et al., 2023). Instead, a detailed isotope deconvolution suggests the occurrence of either chondritic Xe (with mantle 130Xe consisting of up to 22 % of chondritic 130Xe) or recycled Xe from the Archean atmosphere could explain the observed Xe isotope signatures. The latter possibility would have profound implications for models of mantle-surface exchange throughout Earth history. The fission spectra indicate a predominantly 238U origin for fissiogenic Xe, with contribution of 244Pu-derived Xe being negligible within uncertainties, implying extensive mantle degassing during the Hadean and Archean eons. The 129Xe*/136Xe* ratio (where * indicates non-atmospheric excesses of Xe isotopes) of Homa Hills samples correlates with other tracers of mantle/crust contributions such as He, Ar and N isotopes. Variations in 129Xe*/136Xe* among the different gases sampled at Homa Hills is mainly the result of contribution from fissiogenic Xe produced in uranium-rich crustal material. Therefore, this ratio may constitute a robust tracer of mantle-crust interactions. Given available high precision data (Bekaert et al., 2023; 2024; this work) together with mantle-derived rock data, 129Xe*/136Xe* appears homogenous in the convecting mantle, and comparable to values observed at mantle plumes. Such homogeneity is in sharp contrast with light noble gas systematics and may call for whole mantle convection and a core origin for He and Ne..

Under modern oxidising Earth surface conditions, dehydrated subducted slabs show Mo isotope compositions as low as 898/95Mo = -1.5 degrees/00, compared to the depleted mantle 898/95Mo = -0.2 degrees/00. Such light Mo isotope compositions reflect the redox-dependent aqueous mobility of isotopically heavy Mo associated with slab dehydra-tion. Here we analysed basaltic glasses from the South-Mid Atlantic Ridge, whose parental melts are influenced by the enriched Discovery and Shona mantle plumes. We report increasingly higher 898/95Mo of up to -0.1 degrees/00 from the most depleted sam-ples towards those tapping more enriched mantle sources. 898/95Mo values correlate with radiogenic Sr and Nd isotopes, which indicates the recycling of Proterozoic sedi-ments with a Mo isotopic composition that was not affected by subduction-related, oxic dehydration. We propose that the Mo isotope signatures were retained during subduction and reflect anoxic conditions during deep sea sedimentation in the mid-Proterozoic. Finally, Mo isotope fractionation between different terrestrial reservoirs likely depends on the slab redox budget, and therefore on the timing of subduction with regard to Earth's surface oxygenation.

How much nitrogen and light noble gases are recycled in modern subduction zones is unclear. Fumaroles act as a means for passive degassing in arcs. They receive variable contributions of volatiles from arc magmas, themselves sourced from the mantle wedge. The gas compositions reflect the extent of volatile enrichment in sub-arc mantle sources and constrain slab dehydration. However, contributions from atmospheric components in fumaroles are unavoidable. For N-2, neon and argon, the atmospheric components are challenging to discern from slab-derived components. Here, we report (NN)-N-15-N-15 measurements from eight fumaroles and seven bubbling springs, along the Central American arc. Our new (NN)-N-15-N-15 data are coupled with noble gases measurements and show that air-derived components in volcanic gas discharges can easily be underestimated, in both fumaroles and springs, using conventional stable isotope or noble gases methods. We show that, in the absence of (NN)-N-15-N-15 data, previously used tracers for air (e.g., delta N-15, N-2/Ar, N-2/He, among others) may lead to erroneous conclusions regarding the origin of volatiles in mixed gases. In contrast, (NN)-N-15-N-15 data provide quantitative constraints on the nature and contributions of both atmospheric and magmatic components. Most springs are heavily dominated by air-derived N-2, while fumaroles show substantial contributions of volcanic endmembers. Based on the fumarole data, we show that magma sources beneath the central American arc are enriched in all volatiles relative to He-3, by two to three orders of magnitude compared to the MORB source. We use new (NN)-N-15-N-15 data to obtain source N-2/He-3, He-3/Ar-36 and He-3/Ne-22 ratios which we then use to compute volcanic N-2, Ar and Ne degassing fluxes. Using this approach, we show that outgassing fluxes appear to match subduction fluxes in the Central America subduction zone. We determine an N-2 outgassing flux of between 4.0 x 10(8) and 1.0 x 10(9) mol N-2/y, comparable to the subduction flux of 5.7 x 10(8) mol N-2/yr determined previously. We obtain a similar conclusion for Ne-22 and Ar-36. Overall, the volatile fluxes in the central American subduction zone no longer seem to require net transfer of N-2, Ar, and Ne, to the deep mantle. (C) 2021 Elsevier B.V. All rights reserved.

The measurement of methane clumped isotopologues (Delta(CH3D)-C-13 and Delta(CH2D2)-C-12) allows exploring isotope bond ordering within methane molecules, and may reveal equilibrium temperatures. Whether such temperature reflects the formation or re-equilibration temperature of the methane is not well understood, but would have critical implications for the use of methane clumped isotopologues as geo-thermometers. Here we investigate gas bubbles from vigorous emissions at cold seeps (n = 14) in the Sea of Marmara, Turkey. These cold seeps are sourced from deeper sedimentary reservoirs. Conventional geochemical tracers such as carbon and hydrogen bulk isotopic ratios (C-13/C-12 and D/H) or n-alkane molecular ratios, suggest these gases reflect various degrees of mixing between thermogenic and microbial sources. Some samples would generally be considered purely microbial in origin (C-1/C2+ > 1500; delta C-13 < -60 parts per thousand) We report measurements of Delta(CH3D)-C-13 and Delta(CH2D2)-C-12 showing that a fraction of those gases are in internal thermodynamic equilibrium, with the abundances of the two mass-18 isotopologues indicating concordant temperatures of similar to 90 degrees C and similar to 130 degrees C. These concordant temperatures are recorded by gases of putative microbial and thermogenic origin; the temperatures of equilibration are irrespective of the formation mechanism of the gases. We conclude that the two high-temperatures recorded by Delta(CH3D)-C-13 and Delta(CH2D2)-C-12 are best explained by non-enzymatic re-equilibration at two local subsurface temperatures. First principles suggest that unequal rates of exchange are possible. Disequilibrium signatures where the two isotopologues yield discordant apparent temperatures are exhibited by other samples. In those cases the data define a trend of variable Delta(CH3D)-C-13 at nearly constant Delta(CH2D2)-C-12. These signatures are enigmatic, and we investigate and reject multiple possible explanations including mixing, diffusion or Anaerobic Oxidation of Methane. Different rates of re-equilibration between the two rare isotopologues are implied, although lacks experimental foundation at present. In general, all of these data point towards re-equilibration of the mass-18 methane isotopologues as an important process. (C) 2020 Elsevier B.V. All rights reserved.

The development of high-resolution gas source mass spectrometry has permitted entirely new types of measurements of multiply-substituted isotopologues in gas species of geochemical significance. Here, we present recent advances afforded by measurements of 15N15N in natural samples, together with 14N14N and 15N14N. We show that the abundance of the doubly-substituted 15N15N isotopologue in hydrothermal gases, often mixtures of volatiles of widely different origins, allows tracing the provenance of nitrogen. The approach is based on the recent finding that atmospheric N2 has a substantial enrichment in 15N15N of nearly 20‰ relative to any other source of N2. This is particularly useful for the study of hydrothermal gases, where characterizing the isotopic composition and provenance of volcanic N2 is important for a wide range of applications in high-temperature geochemistry, but where air-derived N2 is unavoidable. In this review, we summarize the evidence that 15N15N is an unambiguous tracer of air contamination. We compare two sets of published 15N15N data acquired on gases from plume and arc volcanoes. We show how different sources of volcanic N2 may be in plume versus arc environments, and discuss the first-order constraints on the deep N cycle that are provided by the new 15N15N data. Important findings include that the δ15N tracer, used alone or in conjunction with N2/Ar and N2/He ratios, can be surprisingly deceiving. Isotope fractionation of atmospheric nitrogen occurs within hydrothermal systems, resulting in negative δ15N values similar to estimates for mantle values, yet with 15N15N values that preclude a mantle origin. The 15N15N data show that the true δ15N of volcanic components is positive in arcs but near-zero at the Yellowstone plume. In other words, atmospheric N2 can mimic mantle δ15N, and mantle δ15N can look like the value of air. Without 15N15N, the apportioning of mantle and atmospheric N2 in mixed gases can easily be wrong. With 15N15N, we also determine the true N2/3He and N2/36Ar ratios of volcanic components in hydrothermal systems. Results inform our understanding of the deep nitrogen cycle. Plume and arc volcanic endmembers show distinct isotope and elemental ratios, consistent with sub-arc sources being overwhelmed by near-quantitative slab devolatilization, while the Yellowstone plume source is not reflecting volatile subduction.

Abstract Mangaia, an ocean island in the Cook‐Austral volcanic chain, is the type locality for the HIMU mantle reservoir and has also been shown to exhibit evidence for recycled sulfur with anomalous δ34S and Δ33S that has been attributed an Archean origin. Here we report bulk S‐isotope data from sulfide inclusions in olivine and pyroxene phenocrysts from one of the previously analyzed and four additional Mangaia basalts to further test for the prevalence of anomalous S in the HIMU mantle source feeding Mangaia. We document compositions that range from −5.13‰ to +0.21‰ (±0.3 2σ), +0.006‰ to +0.049‰ (±0.016 2σ), −0.81‰ to +0.69‰ (±0.3 2σ) for δ34S, Δ33S, and Δ36S, respectively. These data extend the range of measured compositions and suggest S‐isotope heterogeneity in the HIMU mantle source at Mangaia. We show that S‐isotope compositions of bulk sulfide in olivine is not in isotopic equilibrium with bulk sulfide in pyroxene from the same samples and that samples from a confined area (M4, M10, M12, and M13) in the northern central part of the island show a distinct covariation for δ34S and Δ33S. This isotopic variation (forming an array) suggests mixing of sulfur from two sources that were captured at different stages of crystallization by phenocrysts in the Mangaia HIMU sulfur endmember.

Deep-sea hydrothermal fluids are often enriched in carbon dioxide, methane, and hydrogen. Methane effuses from metal-rich black smokers such as the Rainbow hydrothermal field, at temperatures higher than 200 degrees C. At the Lost City field, CH4 emanates from alkaline fluids at <100 degrees C. The abundance of the rare, mass-18 CH4 isotopologues, (CH3D)-C-13 and (CH2D2)-C-12, can mitigate degeneracies in the conventional isotopic signatures of methane. We studied the isotopologue compositions of methane from the Rainbow, Lucky Strike, Von Damm, and Lost City hydrothermal fields. At Rainbow, where the vented fluids are at similar to 360 degrees C, our coupled Delta(CH2D2)-C-12 - Delta(CH3D)-C-13 data establish that methane is in internal equilibrium at 343(-35)(+41)degrees C. This may track the formation temperature of abiotic methane, or it may be the result of equilibration of methane isotopologues within the carrier fluid. Lucky Strike and Von Damm have fluid temperatures <300 degrees C and although Delta(CH3D)-C-13 values are indistinguishable from those at Rainbow, (CH2D2)-C-12 abundances are marginally higher. At Lost City, Delta(CH3D)-C-13 data show a range of values, which at face value correspond to apparent temperatures of between 265(-24)(+28) degrees C and 158(-14)(+16) degrees C, far hotter than fluid temperatures. A unique aspect of the Lost City data is the range of large (CH2D2)-C-12 excesses. The Delta(CH2D2)-C-12 data correspond to temperatures of between 101(-8)(+9) degrees C and 69(-4)(+4) degrees C, showing a near-perfect match with fluid temperatures. We find that mixing scenarios involving microbial methane may not account for all of the isotope data. We suggest that Delta(CH2D2)-C-12 values, unlike Delta(CH3D)-C-13 values, are prone to near-complete re-equilibration at host fluid temperatures. We suggest that Delta(CH3D)-C-13 isotopologue data are consistent with abiotic methane being synthesized at similar to 350 degrees C. On the other hand, (CH2D2)-C-12 isotopologue ordering records post formation residence temperatures. We explore a possible mechanism decoupling the reequilibration systematics of the doubly-substituted isotopologues. (C) 2020 Elsevier Ltd. All rights reserved.

Volatile elements (water, carbon, nitrogen, sulfur, halogens, and noble gases) played an essential role in the secular evolution of the solid Earth and emergence of life. Here we provide an overview of Earth's volatile inventories and describe the mechanisms by which volatiles are conveyed between Earth's surface and mantle reservoirs, via subduction and volcanism. Using literature data, we compute volatile concentration and flux estimates for Earth's major volatile reservoirs and provide an internally balanced assessment of modern global volatile recycling. Using a nitrogen isotope box model, we show that recycling of N (and possibly C and S) likely began before 2 Ga and that ingassing fluxes have remained roughly constant since this time. In contrast, our model indicates recycling of H2O(and most likely noble gases) was less efficient in the past. This suggests a decoupling of major volatile species during subduction through time, which we attribute to the evolving thermal regime of subduction zones and the different stabilities of the carrier phases hosting each volatile. ▪ This review provides an overview of Earth's volatile inventory and the mechanisms by which volatiles are transferred between Earth reservoirs via subduction. ▪ The review frames the current thinking regarding how Earth acquired its original volatile inventory and subsequently evolved through subduction processes and volcanism.

Silicon and Mg in differentiated rocky bodies exhibit heavy isotope enrichments that have been attributed to evaporation of partially or entirely molten planetesimals. We evaluate the mechanisms of planetesimal evaporation in the early solar system and the conditions that controled attendant isotope fractionations. Energy balance at the surface of a body accreted within ~1 Myr of CAI formation and heated from within by 26Al decay results in internal temperatures exceeding the silicate solidus, producing a transient magma ocean with a thin surface boundary layer of order < 1 meter that would be subject to foundering. Bodies that are massive enough to form magma oceans by radioisotope decay (≥ 0.1% M ⊕) can retain hot rock vapor even in the absence of ambient nebular gas. We find that a steady-state rock vapor forms within minutes to hours and results from a balance between rates of magma evaporation and atmospheric escape. Vapor pressure buildup adjacent to the surfaces of the evaporating magmas would have inevitably led to an approach to equilibrium isotope partitioning between the vapor phase and the silicate melt. Numerical simulations of this near-equilibrium evaporation process for a body with a radius of ~ 700 km yield a steady-state far-field vapor pressure of 10-8 bar and a vapor pressure at the surface of 10-4 bar, corresponding to 95% saturation. Approaches to equilibrium isotope fractionation between vapor and melt should have been the norm during planet formation due to the formation of steady-state rock vapor atmospheres and/or the presence of protostellar gas. We model the Si and Mg isotopic composition of bulk Earth as a consequence of accretion of planetesimals that evaporated subject to the conditions described above. The results show that the best fit to bulk Earth is for a carbonaceous chondrite-like source material with about 12% loss of Mg and 15% loss of Si resulting from near-equilibrium evaporation into the solar protostellar disk of H2 on timescales of 104 to 105 years.

We have investigated the quadruple sulfur isotopic composition of inorganic sulfur-bearing phases from 13 carbonaceous chondrites of CM type. Our samples include 4 falls and 9 Antarctic finds. We extracted sulfur from sulfides, sulfates, and elemental sulfur (S0) from all samples. On average, we recover a bulk sulfur (S) content of 2.11±0.39wt.% S (1σ). The recovered sulfate, S0 and sulfide contents represent 25±12%, 10±7% and 65±15% of the bulk S, respectively (all 1σ). There is no evidence for differences in the bulk S content between falls and finds, and there is no correlation between the S speciation and the extent of aqueous alteration. We report ranges of Δ33S and Δ36S values in CMs that are significantly larger than previously observed. The largest variations are exhibited by S0, with Δ33S values ranging between −0.104±0.012‰ and +0.256±0.018‰ (2σ). The Δ36S/33S ratios of S0 are on average −3.1±1.0 (2σ). Two CMs show distinct Δ36S/33S ratios, of +1.3±0.1 and +0.9±0.1. We suggest that these mass independent S isotopic compositions record H2S photodissociation in the nebula. The varying Δ36S/Δ33S ratios are interpreted to reflect photodissociation that occurred at different UV wavelengths. The preservation of these isotopic features requires that the S-bearing phases were heterogeneously accreted to the CM parent body. Non-zero Δ33S values are also preserved in sulfide and sulfate, and are positively correlated with S0 values. This indicates a genetic relationship between the S-bearing phases: We argue that sulfates were produced by the direct oxidation of S0 (not sulfide) in the parent body. We describe two types of models that, although imperfect, can explain the major features of the CM S isotope compositions, and can be tested in future studies. Sulfide and S0 could both be condensates from the nebula, as the residue and product, respectively, of incomplete H2S photodissociation by UV light (wavelength <150nm). This idea requires that FeS formation and the S0 condensation co-occur. As an alternative, ice accretion to the CM parent body could allow the delivery of S-MIF in CMs. In that case, sulfides would have been the only S-bearing condensate in CM precursors, and S0 would have been derived from the oxidation of H2S trapped in ices, after its photodissociation at low temperature (<500K) in the nebula. In our models, the observations of H2S UV photodissociation is required to occur at the disk surface, and allowed in nebular environments with canonical C/O ratios. Vertical motions in the disk would redistribute phases that condensed at high altitude to the midplane, where they accreted in the phases that make up the chondritic matrix.

J.-P. Lorand, Labidi J., Thomassot E., Rollion-Bard C., Bellucci, J., Whitehouse, M., Nemchin, A., Hewins, R.H., Humayun, M., Farquhar, J., Zanda, B., Remusat, L., Pont, S. LPG, University of Nantes,44322 Nantes, France (jean-pierre.lorand@univ-nantes.fr); Earth Science Dept. University of Tuebingen, 1272074, Tuebingen, Germany jabrane.labidi@uni-tuebingen.de); CRPG-CNRS, Nancy, France (emilie.@crpg.cnrs-nancy.fr); IPGP 75238, Paris France (rollion@ipgp.fr); Dept. of Applied Geology, Curtin University, Perth, WA 6845, Australia (jeremy.belluci@gmail.com); A. Nemchin@curtin.edu.au); Laboratory for Isotope Geology, Swedish Mus. of Nat History, Stockholm SE-104 05, Sweden (martin.whitehouse@nrm.se).IMPMC,MNHN, 75005 Paris, France (hewins@rci.rutgers.edu; zanda@mnhn.fr ; remusat@mnhn.fr Rutgers University, Piscataway, NJ 08854, USA; Florida State University, Tallahassee, FL 32310, USA (humayun@magnet.fsu.edu) Dept of Geology and ESSIC, University of Maryland, 20742, USA.

We report the quadruple sulfur isotope compositions, sulfur contents and speciation major and trace elements (including copper and chlorine abundances) of eleven basalts collected in the Garrett transform fault. We combine these data to discuss the absence of S isotopic fractionation along both partial melting and low-pressure fractional crystallization.The variations of K2O/TiO2 and La/Sm-N-ratios (respectively between 0.017 and 0.067, and between 0.31 and 0.59) suggest a range of depletion in Garrett lavas that includes ultra depleted samples (K2O/TiO2 < 0.03). The remarkable level of incompatible element depletion is consistent with re-melting of a depleted source. Contrasting with incompatible element depletion, all samples display similar S and Cu abundance (at a given major-element composition) to mid-ocean ridge basalts (MORB). This indicates that Garrett Intra Transform Lavas (ITL) are sulfide saturated as MORB are. Copper content for Garrett parental melts (MgO >8%) are similar to 80 ppm, indistinguishable from MORBs. This requires their mantle sources, variably depleted in incompatible element, to host residual sulfide buffering the Cu content of all erupted melts. We calculate a minimum S content for the source of ultra-depleted Garrett lavas of 100 +/- 40 ppmS, i.e. roughly a factor of 2 below the MORB mantle source.After exclusion of a single sample with CI/K ratio >0.1 that likely experienced hydrothermal sulfide assimilation, Garrett ITLs display homogeneous delta S-34, Delta S-33 and Delta S-36 values with averages of -0.68 +/- 0.08 parts per thousand, +0.010 +/- 0.005 parts per thousand and -0.04 +/- 0.04 parts per thousand, respectively (all 1 sigma, n = 10). The delta S-34 values display no relationship with either K2O/TiO2 variations or extent sulfide fractionation. From these observations, we derive a S-34/S-32 fractionation factor between exsolved sulfides and sulfide dissolved in silicate melts of 1.0000 +/- 0.0003. The S isotopic fractionation during partial melting can thus be considered as negligible, and both MORBs and ITLs record the S-34/S-32 ratio of their mantle source.The concept of sulfide melts segregating from the mantle, sinking and being added to the core during planetary differentiation was termed the 'Hadean Matte'. The segregation of sulfides from the mantle to the core during planetary differentiation could account for various geochemical features of the Earth's mantle. Based on S isotopic mass balance, we derive a lower and upper limit for the hadean matte. While the lower bound corresponds to a virtually negligible hadean matte, the upper limit is 3.36 x 10(24) gS (i.e. similar to 10% of the bulk terrestrial S), which remains 5 to 10 times lower than previous estimates. This upper bound nonetheless requires high mantle S content >1000 ppm S before the extraction of the hadean matte. This suggestion would have chronological requirements, requiring any sulfide melt to have formed after the core extraction but before late accretion of the highly siderophile elements. (C) 2016 Elsevier B.V. All rights reserved.

The Earth's mantle displays a subchondritic S-34/S-32 ratio. Sulfur is a moderately siderophile element (i.e. iron-loving), and its partitioning into the Earth's core may have left such a distinctive isotope composition on the terrestrial mantle. In order to constrain the sulfur isotope fractionation occurring during core-mantle differentiation, high-pressure and temperature experiments were conducted with synthetic mixtures of metal and silicate melts. With the purpose to identify the mechanism(s) responsible for the S isotope fractionations, we performed our experiments in different capsules - namely, graphite and boron nitride capsules - and thus at different fO(2), with varying major element chemistry of the silicate and metal fractions.The S isotope fractionations Delta S-34(metal-silicate) of equilibrated metal alloys versus silicate melts is + 0.2 perpendicular to 0.1% in a boron-free and aluminum-poor system quenched at 1-1.5 GPa and 1650 degrees C. The isotope fractionation increases linearly with increasing boron and aluminum content, up to +1.4 + 0.2%, and is observed to be independent of the silicon abundance as well as of the fO(2) over similar to 3.5 log units of variations explored here. The isotope fractionations are also independent of the graphite or nitride saturation of the metal. Only the melt structural changes associated with aluminum and boron concentration in silicate melts have been observed to affect the strength of sulfur bonding. These results establish that the structure of silicate melts has a direct influence on the S2- average bonding strengths.These results can be interpreted in the context of planetary differentiation. Indeed, the structural environments of silicate evolve strongly with pressure. For example, the aluminum, iron or silicon coordination numbers increase under the effect of pressure. Consequently, based on our observations, the sulfur-bonding environment is likely to be affected. In this scheme, we tentatively hypothesize that S isotope fractionations between the silicate mantle and metallic core of terrestrial planetary bodies would depend on the average pressure at which their core-mantle differentiation occurred. (C) 2015 Elsevier Ltd. All rights reserved.

To better address how subducted protoliths drive the Earth's mantle sulfur isotope heterogeneity, we report new data for sulfur (S) and copper (Cu) abundances, S speciation and multiple S isotopic compositions ((32)s, (33)s, (34)s, (36)s) in 15 fresh submarine basaltic glasses from the Samoan archipelago, which defines the enriched-mantle-2 (EM2) endmember.Bulk S abundances vary between 835 and 2279 ppm. About 17 +/- 11% of sulfur is oxidized (S6+) but displays no consistent trend with bulk S abundance or any other geochemical tracer. The S isotope composition of both dissolved sulfide and sulfate yield homogeneous Delta S-33 and Delta S-36 values, within error of Canyon Diablo Troilite (CDT). In contrast, delta S-34 values are variable, ranging between +0.11 and +2.79 parts per thousand (+/- 0.12 parts per thousand 1 sigma) for reduced sulfur, whereas oxidized sulfur values vary between +4.19 and +9.71 parts per thousand (+/- 0.80 parts per thousand, 1 sigma). Importantly, delta S-34 of the reduced S pool correlates with the Sr-87/Sr-86 ratios of the glasses, in a manner similar to that previously reported for South-Atlantic MORB, extending the trend to delta S-34 values up to +2.79 +/- 0.04 parts per thousand, the highest value reported for undegassed oceanic basalts.As for EM-1 basalts from the South Atlantic ridge, the linear delta S-34-Sr-87/Sr-86 trend requires the EM-2 endmember to be relatively S-rich, and only sediments can account for these isotopic characteristics. While many authors argue that both the EM-1 and EM-2 mantle components record subduction of various protoliths (e.g. upper or lower continental crust, lithospheric mantle versus intra-metasomatized mantle, or others), it is proposed here that they primarily reflect sediment recycling. Their distinct Pb isotope variation can be accounted for by varying the proportion of S-poor recycled oceanic crust in the source of mantle plumes. (C) 2015 Elsevier B.V. All rights reserved.

To better address how Mid-Ocean Ridge Basalt (MORB) sulfur isotope composition can be modified by assimilation and/or by immiscible sulfide fractionation, we report sulfur (S), chlorine (Cl) and copper (Cu) abundances together with multiple sulfur isotope composition for 38 fresh basaltic glasses collected on the Pacific-Antarctic ridge. All the studied glasses - with the exception of 8 off-axis samples - exhibit relatively high Cl/K, as the result of pervasive Cl-rich fluid assimilation. This sample set hence offers an opportunity to document both the upper mantle S isotope composition and the effect of hydrothermal fluids assimilation on the S isotope composition of erupted basalts along segments that are devoid of plume influence.Delta S-33 and Delta S-36 yield homogenous values within error of Canyon Diablo Troilite (CDT), whereas delta S-34 are variable, ranging between -1.57 +/- 0.11 parts per thousand and +0.60 +/- 0.10 parts per thousand with a mean value of -0.64 +/- 0.40 parts per thousand (1 sigma, versus V-CDT). The geographic distribution of delta S-34 follows a spike-like pattern, with local S-34-enrichments by up to +1.30 parts per thousand compared to a low-delta S-34 baseline. As hydrothermal massive sulfides are characterized by relative S-34-enrichments, such first-order variability can be accounted for by hydrothermal sulfide assimilation, a process that would occur for a subset of samples (n = 10). Excluding these particular samples, the mean delta S-34 is significantly less variable, averaging at -0.89 +/- 0.11 parts per thousand (1 sigma, n = 28), a value that we suggest to be representative of the average MORB source value for Pacific-Antarctic basalts. Weak trends between delta S-34 and Pb-206/Pb-204 are displayed by such uncontaminated samples suggesting the recycled oceanic crust to have a modest impact on the S budget of the mantle. Their positive signs, however, suggest the depleted mantle to have a delta S-34 of -1.40 +/- 0.50 parts per thousand. The sub-chondritic S-34/S-32 value that was previously observed for the South-Atlantic mantle is here extended to the Pacific-Antarctic domain. Such a feature cannot originate from oceanic crust recycling and substantiates the concept of a core-mantle fractionation relict. (C) 2014 Elsevier Ltd. All rights reserved.

Earth’s mantle is shown to display heterogeneous sulphur isotope ratios, with a depleted end-member that is not chondritic as has been thought; the mantle’s inferred composition can be accounted for by fractionation during core–mantle differentiation.

We report on improvements to the sulfur extraction method out of silicate glasses by the use of HF + CrCl2 solution for the precise determination of S multi-isotope compositions. This protocol has been designed to be sulfide specific and if needed, can be easily extended to oxidized sulfur. The complete method was validated on a synthetic powder and three laboratory mid-ocean ridge glassy internal standards (CH98 DR12, ED DR46 1-6 and ED DR26 type 1). S extraction yields of the three basalt glass standards were 100 ± 4% (1σ, n = 12), 102 ± 6% (1σ, n = 4) and 97 ± 6% (1σ, n = 4) respectively. Their δ34S show little variation, at − 1.20 ± 0.08‰, − 1.09 ± 0.14‰ and − 1.25 ± 0.12‰ (all 1σ) with respect to V-CDT. Δ33S and Δ36S are both negative with respect to our SF6 tank, between − 0.018 and − 0.021‰ (± 0.012‰ maximum 1σ) for Δ33S and between − 0.216 and − 0.282 (± 0.106 maximum 1σ) for Δ36S. The method was then applied to sixteen additional glasses from worldwide mid-ocean ridges, including three samples for which δ34S was reported previously. Our results show that previous data are affected by a systematic δ34S shift toward positive values, of ≃ 0.4 to 1.5‰. We infer that this shift originates from an incomplete S recovery when the Kiba extraction protocol is used. δ34S values range between − 1.80 and 0.02‰ with a mean value of − 0.91 ± 0.50‰ (1σ n = 19). Such negative values contrast with the positive range previously reported leading us to suggest a revised mantle δ34S, mainly dominated by negative values. Δ33S and Δ36S are strikingly homogeneous with mean respective values of − 0.019 ± 0.005‰ and − 0.193 ± 0.093‰ (1σ n = 19) versus our SF6 tank. These estimates are indistinguishable from our CDT measurements and define the best present-day upper mantle estimate.