Jochen Hoefs

The nature of O18/O16 and C13/C12 secular trends in sedimentary carbonate rocks.

Abstract Measurements on 170 carbonate rocks show decreasing δO 18 of similar magnitude for both limestones and dolomites over a time span of ~ 2800 m.y. The Proterozoic dolomites are on average heavier by ~5%. in δO 18 than their coeval limestones. The data may indicate displacement in δC 13 of about 3%. at approximately 570 m.y. ago, with Precambrian carbonates being heavier in δC 13 at a given δO 18 level than their Phanerozoic counterparts. The compilation of the previously published data is consistent with the above described features.

Oxygen isotope heterogeneity of the mantle deduced from global 18O systematics of basalts from different geotectonic settings

Based upon a compilation and analysis of O-isotope data for Neogene volcanic rocks worldwide, the δ18O variation for 743 basalts (historic lavas, submarine glasses, and lavas with <0.75 wt% H2O) is +2.9 to +11.4‰. Mid-ocean-ridge basalt (MORB) has a uniform O-isotope composition with δ180=+5.7±0.2‰. Basalts erupted in different tectonic settings have mean 18O/16O ratios that are both lower and higher than MORB, with continental basalts enriched in 18O by ca. 1‰ over oceanic basalts. The δ18O range for the subset of 88 basalts with Mg# [100·Mg(Mg+Fe2+)] 75–68, considered to be unmodified primary mantle partial melts, is +3.6 to +8.7‰. These features are a clear indication that: (1) the Earths upper mantle is heterogeneous with respect to its O-isotope composition; (2) that both low-18O and high-18O reservoirs have contributed to basalt petrogenesis. Large-ion lithophile element-enriched basalts associated with subduction at convergent plate margins are slightly enriched in 18O, a characteristic that is considered to be an intrinsic feature of the subduction process. Intraplate oceanic and continental basalts have highly variable 18O/16O ratios, with individual localities displaying δ18O ranges in excess of 1.5 to 2‰. Systematic co-variations between O-, Sr-, Nd-, and Pb-isotope ratios reflect the same principal intramantle end-member isotopic components (DMM, HIMU, EM-I, EM-II) deduced from radiogenic isotope considerations and, therefore, imply that a common process is responsible for the origin of upper mantle stable and radiogenic isotope heterogeneity, namely the recycling of lithospheric material into the mantle.

Kinetically controlled variations of major components and carbon and oxygen isotopes in a calcite-precipitating spring

Abstract A calcite-depositing stream in the Pyrenees, south France, enabled us to study the equilibration of major components and carbon and oxygen isotopes during the formation of calcite and the degassing of CO 2 , and to examine the possible influence of aquatic flora on these processes. Calcite is not precipitated from the solution until a supersaturation of ∼ 10 is attained. In order to start the reaction the activity of CO 2− 3 has to be increased from ∼ 5 · 10 −6 to 22 · 10 −6 . This is caused by an increase of pH due to the loss of CO 2 to the atmosphere. The excess free energy provided by CO 2 degassing is ∼ −5 kJ mol −1 . If formed under equilibrium conditions, the calcite should differ isotopically by Δ 13 C = + 2.3‰ from the dissolved carbonate. However, the observed difference is virtually zero and demonstrates substantial disequilibrium with respect to the stable carbon isotopes. Isotopic disequilibrium is also indicated by oxygen. 13 O-temperatures differ on the average by ∼ + 5°C from the observed temperatures of calcite precipitation. In contrast to calcite formation carbon-isotope equilibrium is attained between the dissolved carbonate and the atmosphere on the degassing of CO 2 . The observed changes of δ 13 C are in agreement with theoretical expectations, unless the surface area of the water is large. The variations of the major components and the carbon and oxygen isotopes observed during diurnal cycles are small and demonstrate that metabolic effects are negligible in a system with a high supply rate of ∼ 25 g HCO − 3 s −1 .

Relationship between13C and18O fractionation and changes in major element composition in a recent calcite-depositing spring — A model of chemical variations with inorganic CaCO3 precipitation

Abstract A theoretical model is derived in which isotopic fractionations can be calculated as a function of variations in dissolved carbonate species on CO 2 degassing and calcite precipitation. This model is tested by application to a calcite-depositing spring system near Westerhof, Germany. In agreement with the model, 13 C of the dissolved carbonate species changes systematically along the flow path. The difference in δ values between the upper and lower part of the stream is about 1‰. The 13 C content of the precipitated calcite is different from that expected from the theoretical partitioning. The isotopic composition of the solid CaCO 3 is similar to that of the dissolved carbonate, though in theory it should be isotopically heavier by about 2.4‰. The 18 O composition of dissolved carbonate and H 2 O is constant along the stream. Calculated calcite-water temperatures differ by about +5°C from the observed temperatures demonstrating isotopic disequilibrium between the water and precipitated solid. This is attributed to kinetic effects during CaCO 3 deposition from a highly supersaturated solution, in which precipitation is faster than equilibration with respect to isotopes. Plant populations in the water have virtually no influence on CO 2 degassing, calcite saturation and isotopic fractionation. Measurements of P CO 2 , S C and 13 C within a diurnal cycle demonstrate that metabolic effects are below the detection limit in a system with a high supply-rate of dissolved carbonate species. The observed variations are due to differences in CO 2 degassing and calcite precipitation, caused by continuously changing hydrodynamic conditions and carbonate nucleation rates.

Mantle buffering of the early oceans

Archean calcites and dolomites, if compared to their Phanerozoic counterparts, are enriched in Sr2+, Ba2+, Mn2+, Fe2+, depleted in 18O, (Na+), and contain mantle-like 87Sr/86Sr and, in associated S phases, mantle-like 34S/32S. This may be a consequence of massive seawater pumping through, and equilibration with, the coeval basaltic oceanic crust. The exponential decline of oceanic geothermal gradient in the course of terrestrial evolution led to a waning of this “mantle” flux and to the enhancement of the continental river discharge as the controlling factor of seawater composition; the major transition occurring probably during the late Archean — early Proterozoic time interval. Such evolution is consistent with the observed tectonic, sedimentological, geochemical and metallogenic secular patterns and may also provide an alternative, or complementary, inorganic explanation for the development of the post-Archean oxygenic atmosphere.

Content and isotopic composition of sulphur in ultramafic xenoliths from central Asia

Sulphur contents and isotopic compositions have been determined in 90 fresh mantle-derived garnet and spinel peridotite and pyroxenite xenoliths from six regions of Cenozoic alkali basaltic volcanism in southern Siberia and Mongolia. Sulphur contents in most of the peridotite nodules fall in the range 6–25 ppm, with largely positive δ34S values clustering between −1 and +7‰ (average +2‰). By contrast, the pyroxenites are richer in sulphur (25–140 ppm), and their δ34S values are close to 0‰ (−1.5 to +1.4‰). In the peridotite nodules, sulphur content correlates negatively with MgO, while their δ34S values correlate positively with MgO. Most of the fertile lherzolites (MgO= 37–39%, CaO= 3–4%) have δ34S values close to 0‰ (−1 to +2‰), similar to meteorite values, while in the harzburgites these values reach +3 to +5‰. These features apparently reflect sulphur depletion during partial melting of peridotite mantle accompanied by sulphur isotopic fractionation between residual peridotites and generated basaltic melts. Clinopyroxene-poor xenoliths from southern Mongolia yielded highest δ34S values of +5 to +7‰ accompanied by high Ba and K contents and high 87Sr/86Sr ratios; these may be a result of interaction with fluids derived from subducted oceanic crust which followed the partial melting and magmatic fractionation episodes. Low average sulphur concentrations ( < 50 ppm) and largely positive δ34S values may predominate in the continental lithospheric mantle worldwide. Sulphur isotope compositions typical of tholeiitic basalts and MORB (0 to +1‰ [1,2]) may be produced by melting of moderately depleted lherzolites. Primary melts with positive δ34S values may be generated from mantle peridotites with larger degrees of depletion and/or from rocks metasomatised by subduction-related fluids.

Carbon and oxygen isotopic covariations in hydrothermal calcites

Isotopic covariations of carbon and oxygen in hydrothermal calcites are quantitatively modeled in terms of the following three mixing processes: (1) mixing between two different fluids which leads to the precipitation of calcite; (2) mixing between fluid and rock: (a) calcite precipitation due to fluid/rock interaction, (b) secondary alteration of primary calcite by interaction with a subsequent fluid. The models are derived from mass balance equations. A distinction among the three mixing processes can be made on a δ13C vs δ18O diagram, which places important constraints on the genesis of hydrothermal mineralization. The variables which control the ultimate isotopic composition of hydrothermal calcites include the composition of the initial fluid and the wallrock, temperature, and dissolved carbon species. Owing to significant temperature-dependent fractionation effects during equilibrium precipitation of calcite from a hydrothermal fluid, the mixing processes may be distinguished by telltale patterns of isotopic data in δ13C vs δ18O space. In particular, caution must be exercised in postulating the fluid mixing as the cause for mineral deposition. This is demonstrated for hydrothermal Pb-Zn deposits in the western Harz Mountains, Germany. A positive correlation between δ13C and δ18O values is observed for calcites from the Bad Grund deposit in the Upper Harz. Two sample profiles through calcite veins show similar correlations with the lowest δ-values at the center of the veins and the highest δ-values at the vein margins. Because the correlation array has a greater slope than for calcite precipitation at equilibrium in a closed system and because fluid mixing may not proceed perpendicular to the vein strike, it is assumed that a fluid/rock interaction is responsible for the observed correlation and thus for the precipitation of calcite. A deep-seated fluid is inferred with a δ13C value of — 7% and a δ18O value of +10%., as well as H2CO3 as the dominant dissolved carbon species; precipitation temperatures of the calcites are estimated to be about 280 ∼ 170°C. Quite different isotopic distributions are observed for calcites from the St. Andreasberg deposit in the Middle Harz. An alteration model is suggested based mainly on the isotopic distribution through a calcite vein. In addition to a primary fluid which has the same isotopic composition as that in the Bad Grund deposit and thus seems to be responsible for the precipitation of calcite associated with sulfides, an evolved, HCO3--dominant subsurface fluid with δ13C about -20 ∼ — 15% and δ18O ≤ 0% is deduced to alter the primary calcite at low temperatures of 70 ∼40°C.

Geochemistry of Precambrian carbonates: I. Archean hydrothermal systems

Abstract Carbonate rocks from the Superior and Slave Provinces of Canada, Kaapvaal Craton of South Africa and the Pilbara Block of Australia, considered of hydrothermal origin from field criteria, have been characterized mineralogically, isotopically (Sr, O, C) and chemically (major and trace elements). In agreement with previous studies, the bulk chemical composition suggests that the carbonate rocks originated by massive carbonatization, silicification and K (±Na) metasomatism of intermediate to ultramafic silicate precursors. The carbonate component itself defines two partially overlapping populations. The ferroan dolomite + breunnerite assemblage is confined mostly to shear zones, conduits and stockworks, while the calcite + ferroan dolomite (ankerite) ± siderite mineral assemblage represents a wider regional halo. This dichotomy may result from a multistory—not necessarily coeval—plumbing system of the volcanosedimentary piles, with the “regional halo” population being a product of shallow circulation cells, while the “conduit” assemblage probably formed from deeper focussed flows. Trace element chemistry of carbonates is consistent with their precipitation from relatively low salinity solutions (⩽2× sea water) and with derivation of solutes from the contemporaneous volcanosedimentary piles ( 87 Sr 86 Sr 0.7020 ± 0.0008) . The hydrothermal waters were derived either from metamorphic dewatering at the base of the piles, or from a magmatic source with δ18O of about +6 ± 7%. SMOW. In contrast, carbon dioxide in the “regional” and the “conduit” assemblages had dissimilar principal sources. In the former it was derived from exogenic sources (decarbonation of marine carbonates and thermal decomposition and oxidation of organic matter), while in the latter the CO2 was of “mantle” origin (δ13C of +2 to −10 and −4 ± 2%. PDB, respectively).

Geochemistry of Precambrian carbonates: 3-shelf seas and non-marine environments of the Archean

A comprehensive whole-rock study of mineralogical, chemical, and isotopic attributes of Archean carbonates suggests that their lithologies and facies have been controlled by tectonic setting. In the first two papers of this series we have shown that the dominant lithology of sedimentary carbonates in greenstone belt settings is limestone ( Fe-rich dolostone of ankeritic composition in older examples ). In this paper we suggest that the Archean shelf sequences are mostly dolostone, and the contemporaneous lacustrine playa lakes are characterized by limestone facies. The present study is of the shelf environments of the Archean, represented by the Pongola Supergroup of South Africa and the Hamersley Group (Wittenoom and Carawine Dolomite ) of Australia. The lacustrine playa examples have been sampled from the Ventersdorp Supergroup of South Africa and the Fortescue Group of Australia. Geological, trace element, and oxygen isotope considerations of the shelf carbonates suggest that their original mineralogy may have been aragonite and that the Pongola dolostones probably represent a direct dolomitization product of this precursor. In contrast, the stabilization of the Hamersley carbonates may have involved an additional step of transformation of a metastable precursor into limestone prior to dolomitization. Subsequently, both sequences, but the Pongola Supergroup much more so, were subjected to pervasive thermal events which overprinted their chemical and isotopic signatures. Consideration of the “best preserved” samples yields +0.9 ± 2.1 and −6 ± l%. PDB for the δ13C and δ18O, respectively, for the initial Archean signature. If compared to Phanerozoic carbonates, the Archean δ13C values are comparable, but the oxygen isotope measurements again support the primary nature of 18O depletion in ancient marine sediments, regardless of its ultimate cause. The 87Sr86Sr of ~2.5 Ga old sea water was not more than 0.7033. If this value reflects the contemporaneous sea water, it would indicate the commencement of a significant 87Sr input from the growing and aging continents. It is, however, entirely possible that the slight enrichment in radiogenic 87Sr—relative to the contemporaneous mantle—represents only an unresolved artefact of post-depositional alteration. If so, the 87Sr86Sr of the 2.5 Ga old sea water may still have been within a mantle range. The lacustrine limestones retain a strong facies control in their trace element attributes. They are also ~4%. depleted in 13C compared to marine carbonates and strongly depleted in 18O (−19 ± l%. PDB). These light δ18O values are somewhat enigmatic, unless the lakes were of the high altitude/latitude variety. The 87Sr86Sr of 0.711 for the Ventersdorp lakes suggests that their drainage basin incorporated not only volcanic rocks but also aged continental crust. Archean carbonates of all lithologies and facies are enriched in Mn and Fe compared to their Phanerozoic counterparts.

Open-system O-isotope behaviour and trace element enrichment in the sub-Eifel mantle

Abstract Spinel peridotites from the West Eifel, western Germany exhibit a wide range of δ 18 O SMOW values for olivine ( +5.1 to +6.3‰), smaller ranges for pyroxenes ( +5.7 to +6.3‰ for orthopyroxene; +5.6 to +6.4‰ for clinopyroxene) and variable 18 O/ 16 O fractionations between olivine and pyroxenes ( Δ clinopyroxene-olivine = +0.6 to −0.3‰ and Δ orthopyroxene-olivine = +0.7 to −0.1‰). These features are characteristic of a disequilibrium distribution of O-isotopes between coexising olivine and pyroxenes. Enrichment in 18 O is observed primarily in olivine from amphibole and/or phlogopite-bearing peridotites and correlates with degree of LREE/HREE enrichment, indicating that the processes responsible for trace element enrichment (i.e. mantle metasomatism) and O-isotope disequilibrium are linked. The 18 O- and incompatible trace element-enriched fluid(s) cannot be generated by intra-mantle fractionation processes as presently understood, and are attributed to recycling of altered oceanic lithosphere and/or pelagic sediments into the mantle via subduction. This is supported by δD SMOW values for associated Eifel amphibole and mica ( −50 to −45‰ respectively) which fall midway between δD values for upper mantle ( −80‰) and seawater (0‰). Coexisting amphibole and mica also have distinctly different δ 18 O values ( +7.0 and +5.9‰, respectively) which suggests that they did not originate from the same fluids/melts. This implies that the sub-Eifel mantle has experienced at least two distinct metasomatic episodes.