Partitioning Of Os Research Articles

Lead and osmium isotopic constraints on the oceanic mantle from single abyssal peridotite sulfides

Single sulfides from abyssal peridotites have been analyzed for Pb and Re-Os to constrain the evolution of oceanic mantle composition. These represent the first analyses of Pb and Os isotopic compositions in the same sulfide grain. The sulfides are from Gakkel and Southwest Indian ridge peridotites, occur at <0.1% modal abundances, and contain 0.001–0.4ppm Re, 0.003–5ppm Os, and 0.12–12ppm Pb. Sulfide Pb isotopic compositions extend from depleted (e.g., 206Pb/204Pb=17.0) to enriched (19.6), covering a larger range than associated basalts. The Os isotopic range of sulfides is more restricted, but extends from depleted (187Os/188Os=0.116) to enriched (0.150). Pb and Os concentrations and isotopic compositions co-vary, with correlation coefficients of 0.76–0.94. Both Pb and Re–Os isotopic data follow ∼2Ga isochrons, with isotopic compositions varying down to small (⪡1km) length-scales and some sulfides containing supra-chondritic 187Re/188Os and 187Os/188Os. These observations are best explained by long-term recycling of oceanic lithosphere combined with melt extraction and refertilization at ancient ocean ridges, rather than a specific event at 2Ga.The concentration of Pb in sulfides indicates that they host <4% of the mantle Pb budget. A re-evaluation of the mass balance of Pb in peridotites indicates that most mantle Pb is stored in silicate phases. The Pb partition coefficient between sulfide melt and silicate melt is estimated to be ∼3, based on the correlation of Pb–Os concentrations in this study and measured Os partition coefficients from the literature. These observations indicate that sulfides do not exert a strong control on the fractionation of Pb during mantle melting, but they can be used to constrain mantle Pb isotopic composition.Sulfides in this study, combined with literature data for Pb isotopes in peridotite whole rocks and pyroxene separates, provide evidence for ultra-depleted mantle, as 24% of peridotites are unradiogenic (i.e., 207Pb/204Pb <15.440 and 206Pb/204Pb <17.726), compared to only 3% of ridge basalts. This suggests that the mantle contains volumetrically significant reservoirs of ultra-depleted material, probably derived from recycled oceanic lithospheric mantle. These depleted reservoirs contribute in only small amounts to oceanic crust generation, both due to a limited ability to melt and to the dilution of any melt by more enriched melts during crust formation.

Software challenges in extreme scale systems

Computer systems anticipated in the 2015 – 2020 timeframe are referred to as Extreme Scale because they will be built using massive multi-core processors with 100's of cores per chip. The largest capability Extreme Scale system is expected to deliver Exascale performance of the order of 1018 operations per second. These systems pose new critical challenges for software in the areas of concurrency, energy efficiency and resiliency. In this paper, we discuss the implications of the concurrency and energy efficiency challenges on future software for Extreme Scale Systems. From an application viewpoint, the concurrency and energy challenges boil down to the ability to express and manage parallelism and locality by exploring a range of strong scaling and new-era weak scaling techniques. For expressing parallelism and locality, the key challenges are the ability to expose all of the intrinsic parallelism and locality in a programming model, while ensuring that this expression of parallelism and locality is portable across a range of systems. For managing parallelism and locality, the OS-related challenges include parallel scalability, spatial partitioning of OS and application functionality, direct hardware access for inter-processor communication, and asynchronous rather than interrupt-driven events, which are accompanied by runtime system challenges for scheduling, synchronization, memory management, communication, performance monitoring, and power management. We conclude by discussing the importance of software-hardware co-design in addressing the fundamental challenges for application enablement on Extreme Scale systems.

High-pressure solid/liquid partitioning of Os, Re and Pt in the Fe–S system

Coupled enrichments in 186Os/ 188Os and 187Os/ 188Os in some ocean island basalts have been interpreted to reflect material transfer from Earth's outer core to the base of the mantle. The outer core is a viable source for these coupled enrichments if it is able to maintain sufficiently high Pt/Os and Re/Os ratios for sufficient time. Iron meteorite studies and some experimental data indicate that progressive crystallization of the inner core is a plausible mechanism for generating the required high Pt/Os and Re/Os ratios in the outer core. However, the conditions of the experiments and meteorite parent bodies are far different from those in Earth's core, particularly in terms of pressure. Here we present experimental results on the partitioning of Os, Re and Pt between solid iron and liquid iron sulfide solutions over a wide range of pressures, from 3.3 to 22 GPa. The solid/liquid partition coefficients for these elements decrease, and become more similar to each other, as pressure increases. This behavior is consistent with an increasing misfit of each element in the iron lattice with increasing pressure, due to the small compressibility of Os, Re and Pt relative to iron. The partition coefficients, and differences among them, are too small at the conditions of these experiments to generate the radiogenic Os isotope anomalies observed in some plume-derived lavas. At 22 GPa the partition coefficients are such that inner core crystallization elevates the Pt/Os ratio of the outer core by only 11%, and the Re/Os ratio by only 6%; after 4.5 Gyr, the calculated 186Os/ 188Os ratio of the outer core is only 0.0012% higher than it would be in the absence of inner core crystallization, and the 187Os/ 188Os ratio is only 1.4% higher. If the pressure effect is due to lattice strain, the partition coefficients are expected to decrease further at the pressure and temperature conditions relevant to Earth's core. Thus, the present data set does not support the idea that radiogenic Os isotope anomalies originate in the outer core.

High pressure solid-metal/liquid-metal partitioning of Os, Re and Pt in the Fe–S system

High pressure solid-metal/liquid-metal partitioning of Os, Re and Pt in the Fe–S system

An experimental study of the solubility and partitioning of iridium, osmium and gold between olivine and silicate melt

We have performed metal solubility and olivine–melt partitioning experiments to understand the behaviour of iridium, osmium and gold during crust–mantle differentiation. All experiments were performed metal-saturated, with molten gold added to suppress nugget formation. Coexisting olivine and iron-bearing basalt glass produced in experiments at 1 bar (10 5 Pa), 1260–1350 °C and fO 2 of FMQ + 0.6 to + 5.4 were analysed by laser ablation ICPMS. Olivine–melt partition coefficients (Ds) for Ir increase from ∼0.4 at FMQ + 5.4 to ∼2 at FMQ + 2.5, with the latter value being consistent with the behaviour of Ir in lavas that have evolved by olivine fractionation. The increase in D Ir with decreasing fO 2 is consistent with an increase in the relative proportion of Ir 2+ in the melt, as it has an estimated ionic radius close to the “strain-free” value for octahedral coordination in olivine. Gold is highly incompatible in olivine ( D = 1 − 2 × 10 − 3 or less), consistent with its relatively large ionic radius in the Au 1+ oxidation state. This, together with our previous measurements of olivine–melt partitioning for Pt, Pd, Ru and Rh, indicates that olivine-equilibrated melts will have depletions in Ir, Rh and Ru relative to Pt, Pd and Au, consistent with that measured in primitive, sulfur-poor magmas. Our data provide an upper bound on Os solubility in silicate melt of 10 ppb at FMQ + 0.6, even though Os partition coefficients could not be measured. The solubility of Os, Ru and Ir in silicate melt is comparable to the concentrations of these elements in primitive, sulfur-poor magmas, suggesting they could be metal-saturated.

Decoupling of the Sm-Nd and Re-Os Isotopic Systems in Sulphide-Saturated Magmas in the Halls Creek Orogen, Western Australia

We have observed apparent decoupling of the Re–Os and Sm–Nd isotopic systems in sulphide-saturated magmas that suggests that bulk two-component or assimilation–fractional crystallization (AFC) mixing modelling, based on Re–Os isotopic data, is inappropriate for chalcophile isotopic systems in turbulent sulphide-saturated magmas. This behaviour is observed in three Palaeoproterozic layered mafic–ultramafic intrusions in the Halls Creek Orogen of Western Australia. All intrusions clearly have a basaltic parental magma based on primitive olivine and spinel compositions. The intrusions are light rare earth element enriched and define a narrow range of initial Nd-isotopic signatures (εNd = −0·9 to +0·70), being derived from an enriched mantle source or involving small degrees of crustal contamination. However, a wide variation in Os-isotopic compositions is observed, ranging from (1) slightly enriched for the Panton intrusion (γOs = +11), requiring an enriched-mantle source or a crustally contaminated mantle-derived magma, through (2) enriched for the McIntosh intrusion (γOs= +38 to +87), to (3) strongly enriched for the Sally Malay intrusion (γOs= +69 to +1300), requiring crustal contamination of a mantle-derived magma. AFC and two-component mixing models can explain the Os- and Nd-isotopic compositions, and in these models an upper crust or radiogenic Lewisian-type lower crust provides a good analogue for the contaminant. Furthermore, the Os isotopic data suggest a picritic parental magma for the Panton and McIntosh intrusions, whereas a basaltic parental magma is indicated for the Sally Malay intrusion. This is in contradiction with the clearly basaltic parental magma implied for all intrusions by olivine and spinel compositions. The inability of two-component mixing and AFC modelling to precisely simulate this system is the result of (1) the magma(s) attaining sulphide saturation, (2) the preferential partitioning of Os into immiscible sulphide and (3) the lowering of Os-isotopic ratios during equilibration between the host magma and immiscible sulphide at high R-factor (high mass ratio of silicate to sulphide melt). Such processes would not affect the lithophile Sm–Nd isotopic system, hence decoupling of the Sm–Nd and chalcophile Re–Os isotopic systems is implied.

Os solubility in silicate melts: New efforts and results

New experimental results are reported on the partitioning of Os between Ni 94 Os 6 (at) alloys and silicate melt at 1400 °C at one atmosphere total pressure and oxygen fugacities ranging from 10 -6 to 10 -8 atm. Specially designed experiments allowed us to avoid nugget formation, which was responsible for the anomalously high Os solubilities reported by Borisov and Palme (1998). Experimental glasses were analyzed by instrumental neutron activation analysis and by isotope dilution analysis. Results reveal a strong dependence of Os solubility on f O₂ , with Os contents decreasing from about 100 ppb to about 5 ppb as f O₂ decreases. The slope of log (Os solubility) vs. log f O₂ implies that Os 3+ is the dominant Os valence state in silicate melts. The relative geochemical behavior of Os and Pt is also compared. Results suggest that Pt is more siderophile, and may also be more chalcophile than Os under the conditions considered.

Nature and Causes of Os Isotope Variations un Late Triassic and Jurassic Seawater

Marine sediments which accumulate under suboxic or anoxic conditions contain certain elements, such as U, V, and Mo, at levels which greatly exceed average crustal abundances as a result of their concentration from seawater by scavenging, precipitation or redox reactionsl. Thus the chemical and isotopic compositions of well-preserved organic-rich mudrocks (ORM) can, in principle, reflect the composition of the seawater in which the sediments were deposited. At times during the Jurassic, much of Europe was covered by extensive areas of epicontinental sea with open-marine connections to the Tethyan ocean; the finely-laminated Jurassic ORM indicate that sub- or anoxic conditions pertained close to the sediment-water interface. On burial and heating, ORM mature to produce hydrocarbons into which some of the organophile elements of the ORM, notably Ni and V, may be partitioned. It is well established that Re and the platinum group elements (PGE) are also present at high concentration (compared with average crustal abundances) in ORM, while a recent preliminary study has shown that both Re and Os partition strongly into crude oil z. Thus maturation and loss of a hydrocarbon phase is likely to perturb the Re-Os system in ORM.

Experimental partitioning of osmium, iridium and gold between basalt melt and sulphide liquid at 1300°C

The partitioning of Os, Ir and Au between FeNi sulphide liquid and basalt melt was determined at 1300°C and low pressure using internally buffered charges in alumina containers sealed in silica tubes. Six runs were at the oxygen fugacity of the iron‐silica phase‐fayalite (IQF) buffer and one at iron‐wustite (IW), with Os, Ir and Au concentrations in the sulphide ranging from approximately 45 to 400 ppm for each metal. The monosulphide liquid (FeNi) varied from 0.02 to 8.9 at % Ni. The major element composition of silicate glass and sulphide run products was quantitatively determined by electron microprobe analysis. The noble metal content of the sulphide phase was calculated from the initial platinum‐group element (PGE) content added and corrected for dilution by constituents derived from the silicate melt, which was mainly iron. Noble metals in silicate glasses were determined by radiochemical neutron activation. Average distribution coefficients (D) for the IQF buffered runs were D(Os) = 3720 ± 1320, D(...