Search

Calcium carbonate and cadmium-rich fluid interactions have been studied at the nano and microscale with fluid flow and static fluid conditions for three forms of CaCO3: calcite in single crystals of Iceland Spar, calcite in a polycrystalline Carrara marble, and aragonite single crystals. Atomic Force Microscopy (AFM) showed the nanoscale effect of cadmium on CaCO3 dissolution and growth under flow-through conditions at ambient temperature, with the modification of calcite dissolution behaviour and simultaneous precipitation of a Cd-rich phase on all the different samples. Hydrothermal experiments at 200°C revealed that the reactivity of single calcite crystals is passivated by epitaxial growth of the less soluble Cd-rich endmember of the (Ca,Cd)CO3 solid-solution on the sample surface due to the similar crystallographic structures of calcite and otavite (CdCO3). Conversely, the presence of grain boundaries in Carrara marble or the change of crystallographic structure and reaction-induced fracturing in aragonite allowed, to some extent, the pseudomorphic replacement of Carrara marble and aragonite samples by a porous (Ca,Cd)CO3 solid-solution phase of variable composition. These phenomena have been observed in solutions undersaturated with respect to all solid phases and are the result of an interface-coupled dissolution-precipitation mechanism where the dissolving CaCO3 provides ions to supersaturate the mineral-fluid interfacial layer, leading to the precipitation of a Cd-containing phase on the samples’ surfaces. This coupled dissolution-precipitation mechanism could potentially be used as a remediation process to sequester cadmium from contaminated effluents.

Silicon isotopes of chemical sediments have received growing attention, given their applicability in the search for properties of ancient seawater. An important target is the reconstruction of secular changes in surface temperature of the Precambrian Earth, but interpretations are problematic since controls of the isotopic signals are potentially manifold. Here we report the existence of significant silicon isotope variability in chemically precipitated chert layers covering a continuous stratigraphic section across the ~ 3.42 Ga Buck Reef Chert (BRC) in the Barberton Greenstone Belt (South Africa). Black chert bands are variably but consistently higher in δ30Si (up to ca. 0.9‰) than co-existing translucent counterparts at the same stratigraphic level, which is interpreted as a primary feature acquired during deposition upon interaction between submarine discharging hydrothermal water and a stratified water body. Averaged δ30Si values range from + 0.3‰ (maximum + 0.5‰) for shallow-marine sediments in the lower part of the section to − 0.5‰ (minimum − 1.3‰) for deeper water deposits in the upper part when results from translucent and black cherts are taken together. An average shift of ca. 0.8‰ (maximum 1.8‰) thus accompanied the formation of chemical cherts in a single Archean basin. Cherts at the base of the section represent silicified felsic volcanics and volcaniclastics and have exclusively positive δ30Si values between + 0.1‰ and + 0.7‰. Rare-earth-elements-and-yttrium patterns confirm the marine origin of the cherts and support the presumed basin evolution. From the combined evidence we infer that the δ30Si variations in the BRC reflect changes in the predominant origin of the silica, with terrigenous input supplying positive δ30Si to shallow waters and seafloor hydrothermal sources negative δ30Si to deeper levels. Our findings demonstrate the viability of silicon isotopes in cherts for reconstructing the evolution of ancient marine basins.

Sr and Nd isotopes were determined using new thermal ionisation mass spectrometry (TIMS) techniques for a suite of 21 olivine-hosted (85-92mol% Fo) melt inclusions selected from potassic and ultra-potassic lavas from the Italian peninsula. Sr isotopes were measured using default 1011Ω resistors, whereas Nd isotope compositions were determined using new 1013Ω resistors mounted in the amplifiers' feedback loop. Compared to default 1011Ω resistors, use of 1013Ω resistors results in a 10-fold improvement of the signal-to-noise ratio and more precise data when analysing small ion beams (

Multiple sulfur isotope variability in Archean sedimentary rocks provides constraints on the composition of the Earth’s earliest atmosphere. The magnitude and sign of mass-independent anomalies reflect not only atmospheric processes, but also transformations due to the Archean marine sulfur cycle prior to preservation into sedimentary pyrite. The processes affecting the Archean marine sulfur cycle and the role of microbial or abiotic redox reactions during pyrite formation remain unclear. Here we combine iron (Fe) and multiple sulfur (S) isotope data in individual pyrite grains with petrographic information and a one-dimensional reactive transport model, to investigate the sources of Fe and S in pyrite formed in a Paleoarchean sedimentary basin. Pyrites were selected from mudstones, sandstones and chert obtained from a drill core in the ca. 3.2 Ga Mapepe and Mendon Formations of the Fig Tree and Onverwacht Groups, respectively, in the Barberton Greenstone Belt, Kaapvaal Craton, South Africa. Pyrite textures and δ 56 Fe distinguish early-diagenetic pyrite formed with pore-water ferrous iron (disseminated grains with average δ 56 Fe pyrite = 0‰) from late-diagenetic pyrite formed through sulfidation of iron oxide minerals (layered and aggregate forms with average δ 56 Fe pyrite = + 1‰). Mass dependent S isotope variability in pyrite was small (δ 34 S pyrite ranged from − 1.1 to + 3.3‰) with a correspondingly minor spread in Δ 33 S pyrite (ranging from + 0.3 to + 2.1‰) and Δ 36 S pyrite (ranging from − 3.08 to + 0.27‰) that indicates a lack of post-depositional re-working with other distinct sulfur sources. Our combined Fe and S isotope data are most readily explained with pyrite sulfide derived from microbial-reworking of solid elemental S. Iron oxide minerals were necessary to buffer sulfide concentrations and provide favorable conditions for microbial sulfur disproportionation to proceed. The lack of a negative Δ 33 S signal indicates that pyrite from relatively deep marine diagenetic environments only partially records the products of atmospheric photolysis, consistent with low sulfate concentrations in the Paleoarchean ocean.

The natural abundance of stable Se isotopes in methylselenides reflects sources and formation conditions of methylselenides. We tested the effects of (i) different inorganic Se species spiked to soils and (ii) different soil samples on the extent of fungal biomethylation of Se and the Se isotope ratios (δ82/76Se) in methylselenides. Furthermore, we assessed the decrease of dissolved, bioavailable Se during three days of equilibration of the soils with Se-enriched solutions. We conducted closed microcosm experiments containing soil spiked with Se(IV) or Se(VI), a growth medium, and the fungus species Alternaria alternata for 11 d. The concentrations and isotope ratios of Se were determined in all components of the microcosm with multicollector ICP-MS. The equilibration of the spiked Se(IV) and Se(VI) for 3 d resulted in a decrease of dissolved, bioavailable Se concentrations by 32 to 44% and 8 to 14%, respectively. Very little isotope fractionation occurred during this phase, and it can be attributed to mixing of the added Se with the pre-existing Se in the soils and minor Se(IV) reduction in one experiment. In two of the incubated soils – moderately acidic roadside and garden soils – between 9.1 and 30% of the supplied Se(IV) and 1.7% of the supplied Se(VI) were methylated while in a strongly acidic forest soil no Se methylation occurred. The methylselenides derived from Se(IV) were strongly depleted in 82Se (δ82/76Se = − 3.3 to − 4.5‰) compared with the soil (0.16–0.45‰) and the added Se(IV) (0.20‰). The methylselenide yield of the incubations with Se(VI) was too small for isotope measurements. Our results demonstrate that Se source species and soil properties influence the extent of Se biomethylation and that the produced methylselenides contain isotopically light Se.

The sequential extraction (SEP) of selenium to determine the isotopic composition of different Se pools in soils is an important tool for tracking biogeochemical cycling of Se in soils. An appropriate Se extraction method for the stable Se isotope analysis was established and tested for different Se pools in soils in terms of reproducibility and recovery. Soil samples (n = 4) from a seleniferous field site in Punjab, India and two standard reference materials were chosen because of their relatively high total Se concentration of 2 to 4.5 mg kg− 1. Additionally, a soil sample with low Se content was utilized to verify the procedure and reproducibility. The isotopic mass-balance calculation gives consistent δ82Se values for the sum of the extracted pools, relative to δ82Se of the bulk soils. For instance, in one studied soil (soil-1), the sum of δ82Se of all individual Se-pools (δ82Semass-balance = 3.35‰) is in good agreement with the analysis of δ82Se of the bulk soil (δ82Sebulk of 3.56‰). We observed up to a 2‰ range in δ82Se among the different Se soil-pools (soil-3). Our Se isotopic results of Se soil pools provide more detailed observations of the Se linked processes in soils compared to bulk δ82Se. Thus, quantitative Se isotopic analysis of different Se pools can be an important tool for predicting the environmental cycling of selenium in soils or sediments.

The widespread use of monazite (LREEPO4) in U-Pb geochronology is underpinned by the assumption that it incorporates negligible amounts of Pb during initial growth, and that radiogenic Pb remains immobile after formation. We have investigated the nanoscale distribution of Pb in monazite from granulite facies rocks of the Sandmata Metamorphic Complex (Rajasthan, India) by atom probe microscopy to further understand the utility of monazite as a geochronometer. The studied monazite contains distinct 10 nm clusters, enriched in Ca and with a bulk composition consistent with them being apatite (Ca5(PO4)3(OH)), that are also enriched in Si and Pb relative to the monazite host. The 208Pb/232Th ratios of the clusters ranged from 1.1 ± 0.1 to 1.4 ± 0.2 (2σ), indicating that the clusters hold unsupported Pb. The 208Pb/232Th ratios of the whole specimen (including clusters) and the matrix alone are similar (

Orthopyroxene (Opx) was present during a significant portion of crystallisation of the lunar magma ocean, but its influence on co-existing melt trace element contents is not well quantified. We performed high-pressure ( P, 1.1 to 3.2 GPa), high-temperature ( T, 1400 to 1600 °C) experiments on synthetic Fe-rich compositions at reducing conditions relevant to the lunar mantle to constrain trace element partitioning between Opx and anhydrous silicate melts. Opx–melt partition coefficients ( D Opx-melt ) for a wide range of trace elements (LILE: Li, Ba; REE: La, Ce, Nd, Sm, Dy, Eu, Er, Tm Y, Yb, Lu; HFSE: Zr, Nb, Hf, Ta, Th, U); and transition metals (Sc, V, Mn, Co, Mo) show only very minor variations across the considered P – T range. REE partition coefficients increase from D La opx-melt = 0.0014 ± 0.0008 to D Lu opx-melt = 0.051 ± 0.007. D values for highly charged elements vary from D Th opx-melt = 0.0013 ± 0.0008 through D Nb opx-melt = 0.0018 ± 0.0006 and D U opx-melt = 0.0015 ± 0.0006 to D Ti opx-melt = 0.068 ± 0.0010. D Lu opx-melt / D Hf opx-melt values of 6.3 ± 2.4 are at the high end of reported values for minerals that played a role during crystallisation of the lunar magma ocean, and higher than previously reported for Opx under identical oxygen fugacity conditions, implying Opx-rich cumulates in the lunar mantle have highly superchondritic Lu–Hf ratios. Lattice strain modelling of our REE partitioning data suggests that varying the concentration of divalent Fe in Opx very slightly decreases the ideal cation radius for M 3+ elements entering the M2 site, r 0 M2 , whereas the partitioning of M 3+ elements entering the M1 site is unaffected. A subtle increase in the maximum partition coefficient for M 3+ elements entering both the M1 and M2 sites with decreasing T is identified, when experiments carried out at similar reducing oxygen fugacities are considered. The presence of Li at concentrations of up to ~ 350 ppm does not have a measureable effect on the Opx–melt partitioning behaviour of REE or any other element, showing that charge-balancing of M 3+ , M 4+ and M 5+ elements in Opx is likely dominated by a vacancy mechanism.

We present numerical experiments on structure development in grain-boundaries during dissolution–precipitation creep. Two solids that are represented by an elastic spring configuration are pressed together with a compressible fluid in the grain-boundary. The solid can dissolve or precipitate depending on elastic and surface energy as well as fluid pressure and concentration of dissolved material in the fluid. We perform a number of numerical experiments with different starting configurations that represent a large-scale island-channel interface with solid–solid contacts across the islands, a rough grain-boundary interface with a fluid along the whole interface and a smooth thin-film interface. The simulations suggest that the solid–solid islands become unstable by necking and anti-cracking so that the island-channel interface will develop into a grain-boundary interface. The rough interface of the grain-boundary will coarsen with time so that in the extreme case a smooth thin-film interface will emerge. However the thin-film interface is also not necessarily stable but may develop into a rough grain-boundary where the roughness is initiated by heterogeneous dissolution of the solid. Our numerical experiments suggest that interface structures during dissolution–precipitation creep are transient and that the dominance of one interface structure depends on material properties, the displacement rate, purity of the reacting solid and the scale of observation.

Sr isotopic compositions and Rb/Sr ratios of three USGS glasses (BHVO-2G, BIR-1G, BCR-2G) are identical to those of the original USGS reference materials. NKT-1G and TB-1G give values of 0.70351 and 0.70558, respectively. Pb isotopic ratios were measured by the standard-sample bracketing technique on an MC-ICP-MS, which give results that are comparable in accuracy and reproducibility to double spike analyses. However, assessment of the reproducibility of the technique is hampered by inhomogeneous contamination of all USGS reference materials analysed. This contamination is likely to be the reason why the USGS glasses do not all have the same Pb isotopic composition as their unfused originals. Powdered glasses, distributed for characterisation of the glasses by bulk analytical techniques, do not all have the same Pb isotopic compositions as the solid glass material, and can therefore not be used for this purpose. © 2005 Elsevier B.V. All rights reserved.

We have studied the sedimentary and basaltic inputs of lithium to subduction zones. Various sediments from DSDP and ODP drill cores in front of the Mariana, South Sandwich, Banda, East Sunda and Lesser Antilles island arcs have been analysed and show highly variable Li contents and δ7Li values. The sediment piles in front of the Mariana and South Sandwich arcs largely consist of pelagic sediments (clays and oozes). The pelagic clays have high Li contents (up to 57.3 ppm) and Li isotope compositions ranging from +1.3‰ to +4.1‰. The oozes have lower Li contents (7.3–16 ppm) with δ7Li values of the diatom oozes from the South Sandwich lower (+2.8‰ to +3.2‰) than those of the radiolarian oozes from the Mariana arc (+8.1‰ to +14.5‰). Mariana sediment also contains a significant portion of volcanogenic material, which is characterised by a moderate Li content (14 ppm) and a relatively heavy isotope composition (+6.4‰). Sediments from the Banda and Lesser Antilles contain considerable amounts of continental detritus, and have high Li contents (up to 74.3 ppm) and low δ7Li values (around 0‰), caused by weathering of continental bedrock. East Sunda sediments largely consist of calcareous oozes. These carbonate sediments display intermediate to high Li contents (2.4–41.9 ppm) and highly variable δ7Li values (−1.6‰ to +12.8‰). Basaltic oceanic crust samples from worldwide DSDP and ODP drill cores are characterised by enrichment of Li compared to fresh MORB (6.6–33.1 vs. 3.6–7.5 ppm, respectively), and show a large range in Li isotope compositions (+1.7‰ to +11.8‰). The elemental and isotopic enrichment of Li in altered basalts is due to the uptake of isotopically heavy seawater Li during weathering. However, old oceanic crust samples from Sites 417/418 exhibit lighter Li isotope compositions compared to young basaltic crust samples from Sites 332B and 504B. This lighter Li isotope signature in old crust is unexpected and further research is needed to explore this issue.

Volcanism in the Canary Islands is notable for its highly alkalic character even in the shield building lavas of the currently most active island, La Palma. In order to understand better the processes responsible for this alkalic end of the compositional spectrum of ocean island basalts (OIB), we have studied primary magmatic inclusions in olivine and clinopyroxene from La Palma. The investigated primary melt and fluid inclusions are hosted by olivine and clinopyroxene from lavas of two ∼700-ka, continuous stratigraphic sections in the Fagundo and Izcagua barrancos of the main shield volcano Taburiente. Lavas from the two barrancos show contrasting whole-rock compositions: Izcaqua lavas are highly alkalic or “basanitic”, typical of La Palma lavas in general, whilst the Fagundo lavas are characterised by more silica-rich, “transitional”, compositions. Although whole-rock compositions of the lavas from the two barrancos show different degrees of silica undersaturation, they show a similar range of elemental abundances such as MgO. In contrast, the Fagundo melt inclusions show a wider range, to more mafic compositions, than those from the Izcagua samples. This is in keeping with more variable partial homogenisation temperatures and host mineral compositions in the Fagundo samples, which are also generally higher (1130–1260 °C) and more primitive (Fo 76–87) than in the Izcagua suite. Indeed, the Izcagua melt inclusions are hosted in relatively evolved liquidus mineral assemblages (Fo

Melt, fluid and crystal inclusions have been studied in olivine phenocrysts from a suite of picritic basalts dredged during the “Marion Dufresne” MD 109 cruise (Dredge 6) from a seamount located between the Kerguelen Archipelago and Heard Island (Southern Indian Ocean). A two-stage polybaric crystallisation of the trapped melt is indicated by compositions of the host olivine+spinel assemblage and by the density of coexisting fluid inclusions. The melt trapped in early olivine phenocrysts (Fo 82–85.5 ) crystallised under pressures from 2 to 4 kbar, whereas the melt trapped in late olivine phenocrysts (Fo 78.7–80 ) crystallised at shallower depths ( 2 O 5 =0.4–1.6 wt.%, K 2 O=1.4–2.4 wt.%) at a relatively restricted MgO content (7.0–9.6 wt.%). The small amounts of P 2 O 5 -rich melt trapped in the early olivine phenocrysts are remarkably similar to the contemporaneous basalt suite (Dredge 5) from the same seamount. The chemistry of the host dredge 6 basalts requires a significant contribution (more than 80%) of richer in K 2 O and poorer in P 2 O 5 subalkaline melt (K 2 O>2.5 wt.%, P 2 O 5 206 Pb/ 204 Pb– 143 Nd/ 144 Nd trachybasaltic suite of the Aphanasey Nikitin Rise, may explain the striking Nb–Ta depletion, Pb enrichment and isotopic compositions of the dredge 6 basalts. The petrogenesis of the picritic basalts from dredge 6 is controlled by two-component mixing between a P 2 O 5 -rich Kerguelen plume-derived melt and a subalkaline melt derived from the Kerguelen Plateau lithosphere and provides more evidence for the involvement of low 206 Pb/ 204 Pb– 143 Nd/ 144 Nd lithospheric material into young Kerguelen plume-derived basaltic magmas.

Both concentration and isotope composition of bulk carbon in apatite and host eclogites from the Dabie–Sulu ultrahigh pressure (UHP) terranes in China have been determined along with their oxygen isotope composition. The results show significant 13 C-depletion in the apatite (δ 13 C=−27.7‰ to −20.8‰) with the carbon concentrations of 0.59 to 1.65 wt.% CO 2 despite a large variation in δ 18 O (−6.5‰ to +9.5‰). The bulk carbon in 21 of the 24 eclogites has low δ 13 C values of −26.1‰ to −17.9‰ with low carbon content of 500 to 1000 ppm, whereas the other three samples show high δ 13 C values of −7.1‰ to −2.8‰ with high carbon contents of 2400 to 4300 ppm. Noncarbonate carbon was measured by treating the all eclogites with 5 N HCl solution, yielding uniformly low δ 13 C values of −27.9‰ to −24.2‰ and carbon contents of 200 to 300 ppm. Carbonate carbon is thus calculated by mass balance to have also low δ 13 C values of −25.6‰ to −15.1‰ for the 21 samples but high δ 13 C values of −4.3‰ to −1.2‰ for the remaining three samples. Secondary carbonate was identified in the three eclogites that are also depleted in 13 C primarily, but subjected to overprint of 13 C-rich CO 2 -bearing fluid subsequent to the UHP metamorphism. The isotopically light carbon in both eclogite and apatite is interpreted to represent the isotope composition of carbon in eclogite precursors before plate subduction, and thus has an origin of organic carbon from the Earth's surface. The uniformly low δ 13 C values of apatite suggest that the CO 2 of metamorphic fluid in equilibrium with the host-eclogites would be derived from the oxidation of organic carbon rather than the decarbonation of underthrust carbonates during progressive metamorphism. Protoliths of the eclogites are inferred to be of igneous origin, which underwent more extensive interaction with organic matter than with meteoric-hydrothermal fluid on the subsurface of the continental crust. The break off of the subducted plate containing the isotopically light carbon and subsequent interactions with the surrounding mantle could produce the mafic and/or silicic magmas that are significantly depleted in 13 C relative to the primary mantle carbon. This may provide evidence for a linkage of the 13 C-depleted mantle carbon to a surficial source via plate subduction.