Trace/minor element:calcium ratios in cultured benthic foraminifera. Part II: Ontogenetic variation

Ontogenetic (developmental stage) measurements of Mg/Ca and Sr/Ca were made on the benthic foraminifer Bulimina aculeata, which were cultured under controlled physicochemical conditions of temperature, pH, alkalinity, salinity, and trace- and minor-element concentrations. We utilized two methods of ontogenetic sampling—whole specimens progressively increasing in length and laser microdissection of a single specimen with subsequent analysis of dissected portions. A novel high-resolution laser-microdissection (HRLM) method allowed for precise (10 μm) cuts of the foraminiferal tests (shells) along the geometrically complex sutures distinguishing individual chambers. This new microdissection method limited sample loss and cross-contamination between foraminiferal chambers. Little or no variation in D_Sr was observed at different foraminiferal developmental stages. Conversely, D_Mg was enriched during a mid-developmental stage of whole-specimen samples (150-225 μm D_Mg = 1.6 × 10⁻³) compared to earlier and later stages (<150 μm, >225 μm D_Mg = 8.3 × 10⁻⁴). Further analysis of HRLM ontogenetic samples showed a larger, age-dependent D_Mg signature variation. This increase in shell Mg/Ca may contribute substantially to the measured inter-individual variability in Mg/Ca temperature prediction for cultured B. aculeata. Due to relatively large Mg/Ca inter- and intra-individual variability, measuring similar-size foraminiferal samples may improve the precision of paleotemperature prediction. Additionally, partial dissolution of the highest ontogenetically Mg-enriched calcite (D_Mg = 1.3 × 10⁻²-1.6 × 10⁻²) may occur in undersaturated bottom-water environments or during reductive cleaning procedures. Thus, the calcite phases remaining after partial dissolution by either natural or laboratory cleaning processes may not accurately represent the calcification environment.     

Adsorption of Sr and Ba at the cleaved mica-water interface: Free energy profiles and interfacial structure

Monte Carlo simulations show that the adsorption position of the Sr²⁺ or Ba²⁺ ion on the cleaved muscovite surface is determined by the radius of the ion’s first hydration shell, hydrogen bonding of the first shell water molecules with the basal oxygens of muscovite as well as a requirement of minimization of the number of muscovite’s lattice cations in the ion’s first coordination shell. Accordingly, Sr²⁺ or Ba²⁺ adsorbs in ditrigonal cavities at a distance of 1.12 Å or 1.35 Å, respectively, from the basal surface on dehydrated mica and above tetrahedral substitutions at a height of 1.93 ± 0.02 Å or 2.15 ± 0.03 Å, respectively, at the highest simulated water coverage of 28 H₂O per ion. The ion’s displacement from a ditrigonal cavity occurs upon adsorption of 2 H₂O per ion for Sr²⁺ and 3 H₂O per ion for Ba²⁺. At a coverage of 28 H₂O per ion, outer-sphere adsorption of Sr²⁺ or Ba²⁺ at a height of 3.9 ± 0.2 Å or 4.17 ± 0.07 Å, respectively, is possible albeit unfavorable on the free energy scale by 107 ± 7 kJ/mol or 89 ± 13 kJ/mol, respectively, as compared to inner-sphere adsorption. Activation energies for the transformation between inner-sphere and outer-sphere adsorptions are calculated to be 121 ± 3 kJ/mol for Sr²⁺ and 99 ± 10 kJ/mol for Ba²⁺. A comparison of these values with those reported recently for Mg²⁺ and Ca²⁺ results in an adsorption affinity sequence Mg²⁺ > Ca²⁺ > Sr²⁺ > Ba²⁺ in agreement with the sequence predicted recently for low dielectric constant solids (which include mica) (Sverjensky, 2006). A recent resonant anomalous X-ray reflectivity study of Sr²⁺ adsorption on muscovite (Park et al., 2006) has questioned the common assumption (Stumm, 1992), which is supported by the present simulation results, that inner-sphere adsorption is stronger than outer-sphere adsorption. A modification of the cleaved muscovite surface as a result of Sr²⁺ adsorption in muscovite’s ditrigonal cavities and related destabilization of muscovite’s hydroxyl groups is proposed as a possible reason for this controversy.     

The impact of salinity on the Mg/Ca and Sr/Ca ratio in the benthic foraminifera Ammonia tepida: Results from culture experiments

Over the last decade, sea surface temperature (SST) reconstructed from the Mg/Ca ratio of foraminiferal calcite has increasingly been used, in combination with the δ¹⁸O signal measured on the same material, to calculate the δ¹⁸O_w, a proxy for sea surface salinity (SSS). A number of studies, however, have shown that the Mg/Ca ratio is also sensitive to other parameters, such as pH or , and salinity. To increase the reliability of foraminiferal Mg/Ca ratios as temperature proxies, these effects should be quantified in isolation. Individuals of the benthic foraminifera Ammonia tepida were cultured at three different salinities (20, 33 and 40 psu) and two temperatures (10-15 °C). The Mg/Ca and Sr/Ca ratios of newly formed calcite were analyzed by Laser Ablation ICP-MS and demonstrate that the Mg concentration in A. tepida is overall relatively low (mean value per experimental condition between 0.5 and 1.3 mmol/mol) when compared to other foraminiferal species, Sr being similar to other foraminiferal species. The Mg and Sr incorporation are both enhanced with increasing temperatures. However, the temperature dependency for Sr disappears when the distribution factor D_Sr is plotted as a function of calcite saturation state (Ω). This suggests that a kinetic process related to Ω is responsible for the observed dependency of Sr incorporation on sea water temperature. The inferred relative increase in D_Mg per unit salinity is 2.8% at 10 °C and 3.3% at 15 °C, for the salinity interval 20-40 psu. This implies that a salinity increase of 2 psu results in enhanced Mg incorporation equivalent to 1 °C temperature increase. The D_Sr increase per unit salinity is 0.8% at 10 °C and 1.3% at 15 °C, for the salinity interval 20-40 psu.     

Determination of Sr and Ba partition coefficients between apatite from fish (Sparus aurata) and seawater: The influence of temperature

The Sr/Ca and Ba/Ca ratios in inorganic apatite are strongly dependent on the temperature of the aqueous medium during precipitation. If valid in biogenic apatite, these thermometers would offer the advantage of being more resistant to diagenesis than those calibrated on biogenic calcite and aragonite. We have reared seabreams (Sparus aurata) in tanks with controlled conditions during experiments lasting for more than 2 years at 13, 17, 23 and 27 °C, in order to determine the variations in Sr and Ba partitioning relative to Ca (D^Sr and D^Ba, respectively) between seawater and fish apatitic hard tissues (i.e. teeth and bones), as a function of temperature. The sensitivity of the Sr and Ba thermometers (i.e. ∂D^Sr/∂T and ∂D^Ba/∂T, respectively), are similar in bone (/∂T = 0.0036 ± 0.0003 and /∂T = 0.0134 ± 0.0026, respectively) and enamel (/∂T = 0.0037 ± 0.0005 and /∂T = 0.0107 ± 0.0026, respectively). The positive values of ∂D^Sr/∂T and ∂D^Ba/∂T in bone and enamel indicate that D^Sr and D^Ba increase with increasing temperature, a pattern opposite to that observed for inorganic apatite. This distinct thermodependent trace element partitioning between inorganic and organic apatite and water highlights the contradictory effects of the crystal-chemical and biological controls on the partitioning of Ca, Sr and Ba in vertebrate organisms. Taking into account the diet Sr/Ca and Ba/Ca values, it is shown that the bone Ba/Ca signature of fish can be explained by Ca-biopurification and inorganic apatite precipitation, whereas both of these processes fail to predict the bone Sr/Ca values. Therefore, the metabolism of Ca as a function of temperature still needs to be fully understood. However, the biogenic Sr thermometer is used to calculate an average seawater temperature of 30.6 °C using the Sr/Ca compositions of fossil shark teeth at the Cretaceous/Tertiary boundary, and a typical seawater Sr/Ca ratio of 0.02. Finally, while the present work should be completed with data obtained in natural contexts, it is clear that Sr/Ca and Ba/Ca ratios in fossil biogenic apatite already constitute attractive thermometers for marine paleoenvironments.     

Seawater temperature proxies based on DSr, DMg, and DU from culture experiments using the branching coral Porites cylindrica

In order to investigate the incorporation of Sr, Mg, and U into coral skeletons and its temperature dependency, we performed a culture experiment in which specimens of the branching coral (Porites cylindrica) were grown for 1 month at three seawater temperatures (22, 26, and 30 °C). The results of this study showed that the linear extension rate of P. cylindrica has little effect on the skeletal Sr/Ca, Mg/Ca, and U/Ca ratios. The following temperature equations were derived: Sr/Ca (mmol/mol) = 10.214(±0.229) − 0.0642(±0.00897) × T (°C) (r² = 0.59, p < 0.05); Mg/Ca (mmol/mol) = 1.973(±0.302) + 0.1002(±0.0118) × T (°C) (r² = 0.67, p < 0.05); and U/Ca (μmol/mol) = 1.488(±0.0484) − 0.0212(±0.00189) × T (°C) (r² = 0.78, p < 0.05). We calculated the distribution coefficient (D) of Sr, Mg, and U relative to seawater temperature and compared the results with previous data from massive Porites corals. The seawater temperature proxies based on D calibrations of P. cylindrica established in this study are generally similar to those for massive Porites corals, despite a difference in the slope of D_U calibration. The calibration sensitivity of D_Sr, D_Mg, and D_U to seawater temperature change during the experiment was 0.64%/°C, 1.93%/°C, and 1.97%/°C, respectively. These results suggest that the skeletal Sr/Ca ratio (and possibly the Mg/Ca and/or U/Ca ratio) of the branching coral P. cylindrica can be used as a potential paleothermometer.     

Experimentally determined biomediated Sr partition coefficient for dolomite: Significance and implication for natural dolomite

Two strains of moderately halophilic bacteria were grown in aerobic culture experiments containing gel medium to determine the Sr partition coefficient between dolomite and the medium from which it precipitates at 15 to 45 °C. The results demonstrate that Sr incorporation in dolomite does occur not by the substitution of Ca, but rather by Mg. They also suggest that Sr partitioning between the culture medium and the minerals is better described by the Nernst equation (D_Sr^dol = Sr_dol/Sr_bmi), instead of the Henderson and Kracek equation (D_Sr^dol = (Sr/Ca)_dol/(Sr/Ca)_solution. The maximum value for D_Sr^dol occurs at 15 °C in cultures with and without sulfate, while the minimum values occur at 35 °C, where the bacteria exhibit optimal growth. For experiments at 25, 35 and 45 °C, we observed that D_Sr^dol values are greater in cultures with sulfate than in cultures without sulfate, whereas D_Sr^dol values are smaller in cultures with sulfate than in cultures without sulfate at 15 °C.Together, our observations suggest that D_Sr^dol is apparently related to microbial activity, temperature and sulfate concentration, regardless of the convention used to assess the D_Sr^dol. These results have implications for the interpretation of depositional environments of ancient dolomite. The results of our culture experiments show that higher Sr concentrations in ancient dolomite could reflect microbial mediated primary precipitation. In contrast, previous interpretations concluded that high Sr concentrations in ancient dolomites are an indication of secondary replacement of aragonite, which incorporates high Sr concentrations in its crystal lattice, reflecting a diagenetic process.     

Partitioning of Nb and Ta between rutile and felsic melt and the fractionation of Nb/Ta during partial melting of hydrous metabasalt

In order to fully assess the role of rutile in fractionation of Nb/Ta during partial melting of hydrous metabasalt, we have measured rutile - felsic melt partition coefficients (D values) for Nb and Ta with tonalitic to trondhjemitic compositions at 1.5-3.5 GPa, 900-1350 °C and ∼5.0-20 wt% H₂O. D_Nb, D_Ta and D_Nb/D_Ta range from 17 ± 1 to 246 ± 13, 34 ± 2 to 232 ± 25 and 0.51 ± 0.04 to 1.06 ± 0.13, respectively. For the compositions investigated, melt composition appears to have no observable effect on the partitioning; the effect of pressure is also slight; whereas temperature and H₂O have marked effects. D_Nb, D_Ta and D_Nb/D_Ta increase with decreasing temperature and H₂O content, showing a reversal of D_Nb/D_Ta from <1.0 to >1.0. Using the data that approached equilibrium and obeyed Henry’s law, expressions describing the dependences of D_Nb, D_Ta and D_Nb/D_Ta on temperature, pressure and melt H₂O content were obtained:

(1)     

Barium uptake into the shells of the common mussel (Mytilus edulis) and the potential for estuarine paleo-chemistry reconstruction

In this study we test if calcite shells of the common mussel, Mytilus edulis, contain barium in proportion to the water in which they grew. Similar to all bivalves analyzed to date, the [Ba/Ca]_shell profiles are characterized by a relatively flat background [Ba/Ca]_shell, interrupted by sharp [Ba/Ca]_shell peaks. Previous studies have focused on these [Ba/Ca]_shell peaks, but not on the background [Ba/Ca]_shell. We show that in both laboratory and field experiments, there is a direct relationship between the background [Ba/Ca]_shell and [Ba/Ca]_water in M. edulis shells. The laboratory and field data provided background Ba/Ca partition coefficients (D_Ba) of 0.10 ± 0.02 and 0.071 ± 0.001, respectively. This range is slightly higher than the D_Ba previously determined for inorganic calcite, and slightly lower than foraminiferal calcite. These data suggest that M. edulis shells can be used as an indicator of [Ba/Ca]_water, and therefore, fossil or archaeological M. edulis shells could be used to extend knowledge of estuarine dissolved Ba throughputs back in time. Moreover, considering the inverse relationship between [Ba/Ca]_water and salinity, background [Ba/Ca]_shell data could be used as an estuary specific indicator of salinity. The cause of the [Ba/Ca]_shell peaks is more confusing, both the laboratory and field experiments indicate that they cannot be used as a direct proxy of [Ba/Ca]_water or phytoplankton production, but may possibly be caused by barite ingestion.     

Tracer diffusion of Mg, Ca, Sr, and Ba in Na-aluminosilicate melts

We employed the thin source technique to investigate tracer diffusion of Mg, Ca, Sr, and Ba in glasses and supercooled melts of albite (NaAlSi₃O₈) and jadeite (NaAlSi₂O₆) compositions. The experiments were conducted at 1 bar and at temperatures between 645 and 1025°C. Typical run durations ranged between 30 min and 35 days. The analysis of the diffusion profiles was performed with the electron microprobe. Diffusivities of Ca, Sr, and Ba were found to be independent of either duration t of the experiment or tracer concentration M, initially introduced into the sample. Mg exhibits a diffusivity depending on run time and concentration and tracer diffusivity is derived by extrapolation to M/√t = 0. Temperature dependence of the diffusivity D can be represented by an Arrhenius equation D = D_o exp(−E_a/RT), yielding the following least-squares fit parameters (with D in m²/s and E_a in kJ/mol): D_Mg = 1.8 · 10⁻⁵ exp(−234 ± 20/RT), D_Ca = 3.5 · 10⁻⁶ exp(−159 ± 6/RT), D_Sr = 3.6 · 10⁻⁶ exp(−160 ± 6/RT), and D_Ba = 6.0 · 10⁻⁶ exp(−188 ± 12/RT) for albite; and D_Mg = 8.3 · 10⁻⁶ exp(−207 ± 18/RT), D_Ca = 3.8 · 10⁻⁶ exp(−153 ± 4/RT), D_Sr = 2.3 · 10⁻⁶ exp(−150 ± 4/RT), and D_Ba = 3.7 · 10⁻⁵ exp(−198 ± 4/RT) for jadeite composition. Ca and Sr diffusivities agree within error in both compositions and exhibit the fastest diffusivities, whereas Mg reveals the lowest diffusivity. The relationship between activation energy and radius shows a minimum at Ca and Sr for albite and jadeite compositions extending the relationship already observed elsewhere for alkalies. With increasing substitution of Si by (Na + Al), diffusivities increase, whereas activation energies decrease. Furthermore, a simple model modified from that of Anderson and Stuart (Anderson O. L. and Stuart D. A., “Calculation of activation energy of ionic conductivity in silica glasses by classical methods,” J. Am. Ceram. Soc.37, 573-580, 1954) is discussed for calculating the activation energies.     

High-precision high field strength element partitioning between garnet, amphibole and alkaline melt from Kakanui, New Zealand

The high field strength elements (HFSE: Zr, Hf, Nb, Ta, and W) are an important group of chemical tracers that are increasingly used to investigate magmatic differentiation processes. Successful modeling of these processes requires the availability of accurate mineral-melt partition coefficients (D). To date, these have largely been determined by ion microprobe or laser ablation-ICP-MS analyses of the run products of high-pressure, high-temperature experiments. Since HFSE are (highly) incompatible, relatively immobile, high-charge, and difficult to ionize, these experiments and their analysis are challenging. Here we explore whether high-precision analyses of natural mineral-melt systems can provide additional constraints on HFSE partitioning.The HFSE concentrations in natural garnet and amphibole and their alkaline host melt from Kakanui, New Zealand are determined with high precision isotope dilution on a multi-collector-ICP-MS. Major and trace element compositions combined with Lu-Hf isotopic systematics and detailed petrographic sample analysis are used to assess mineral-melt equilibrium and to provide context for the HFSE D measurements. The whole-rock nephelinite, ∼1 mm sized amphiboles in the nephelinite, and garnet megacrysts have similar initial Hf isotope ratios with a mean initial ¹⁷⁶Hf/¹⁷⁷Hf_{(34 Ma)} = 0.282900 ± 0.000026 (2σ). In contrast, the amphibole megacrysts are isotopically distinct (¹⁷⁶Hf/¹⁷⁷Hf_{(34 Ma)} = 0.282830 ± 0.000011). Rare earth element D values for garnet megacryst-nephelinite melt and ∼1 mm amphibole-nephelinite melt plotted as a function of ionic radii show classic near-parabolic trends that are in excellent agreement with crystal lattice-strain models. These observations are consistent with equilibrium between the whole-rock nephelinite, the ∼1 mm amphibole grains within the nephelinite and the garnet megacrysts.High-precision isotope dilution results for Zr and Hf in garnet (D_Zr = 0.220 ± 0.007 and D_Hf = 0.216 ± 0.005 [2σ]), and for all HFSE in amphibole are consistent with previous experimental findings. However, our measurements for Nb and Ta in garnet (D_Nb = 0.0007 ± 0.0001 and D_Ta = 0.0011 ± 0.0006 [2σ]) show that conventional methods may overestimate Nb and Ta concentrations, thereby overestimating both Nb and Ta absolute D values for garnet by up to 3 orders of magnitude and underestimating D_Nb/D_Ta by greater than a factor of 100. As a consequence, the role of residual garnet in imposing Nb/Ta fractionation may be less important than previously thought. Moreover, garnet D_Hf/D_W = 17 and D_Nb/D_Zr = 0.003 imply fractionation of Hf from W and Nb from Zr upon garnet crystallization, which may have influenced short-lived ¹⁸²Hf-¹⁸²W and ⁹²Nb-⁹²Zr isotopic systems in Hadean time.     

Adsorption of Rb and Sr at the orthoclase (0 0 1)-solution interface

Adsorption of Rb⁺ and Sr²⁺ at the orthoclase (0 0 1)-solution interface is probed with high-resolution X-ray reflectivity and resonant anomalous X-ray reflectivity. Specular X-ray reflectivity data for orthoclase in contact with 0.01 m RbCl solution at pH 5.5 reveal a systematic increase in electron density adjacent to the mineral surface with respect to that observed in contact with de-ionized water (DIW). Quantitative analysis indicates that Rb⁺ adsorbs at a height of 0.83 ± 0.03 Å with respect to the bulk K⁺ site with a nominal coverage of 0.72 ± 0.10 ions per surface unit mesh (55.7 Å²). These results are consistent with an ion-exchange reaction in which Rb⁺ occupies an inner-sphere adsorption (IS) site. In contrast, X-ray reflectivity data for orthoclase in contact with 0.01 m Sr(NO₃)₂ solution at pH 5.3 reveal few significant changes with respect to DIW. Resonant anomalous X-ray reflectivity was used to probe Sr²⁺ adsorption and to image its vertical distribution. This element-specific measurement reveals that Sr²⁺ adsorbs with a total coverage of 0.37 ± 0.02 ions per surface unit mesh, at a substantially larger height (3.28 ± 0.05 Å) than found for Rb⁺, and with a relatively broad density distribution (having a root-mean-square width of 1.88 ± 0.08 Å for a single-peak model), implying that Sr²⁺ adsorbs primarily as a fully-hydrated outer-sphere (OS), species. Comparison to a two-height model suggests that 13 ± 5% of the adsorbed Sr²⁺ may be present as an IS species. This partitioning implies a ∼5 kJ/mol difference in free energy between the IS and OS Sr²⁺ on orthoclase. Differences in the partitioning of Sr²⁺ between IS and OS species for orthoclase (0 0 1) and muscovite (0 0 1) suggest control by the geometry of the IS adsorption site. Results for the OS distribution are compared to predictions of the Poisson-Boltzmann equation in the strong coupling regime, which predicts an intrinsically narrow vertical diffuse ion distribution; the OS distribution might thus be thought of as the diffuse ion profile in the limit of high surface charge.     

Determination of fluid/melt partition coefficients by LA-ICPMS analysis of co-existing fluid and silicate melt inclusions: Controls on element partitioning

Analyses of co-existing silicate melt and fluid inclusions, entrapped in quartz crystals in volatile saturated magmatic systems, allowed direct quantitative determination of fluid/melt partition coefficients. Investigations of various granitic systems (peralkaline to peraluminous in composition, log fO₂ = NNO−1.7 to NNO+4.5) exsolving fluids with various chlorinities (1-14 mol/kg) allowed us to assess the effect of these variables on the fluid/melt partition coefficients (D). Partition coefficients for Pb, Zn, Ag and Fe show a nearly linear increase with the chlorinity of these fluid (D_Pb ∼ 6 ∗ m_Cl, D_Zn ∼ 8 ∗ m_Cl, D_Ag ∼ 4 ∗ m_Cl, D_Fe ∼ 1.4 ∗ m_Cl, where m_Cl is the molinity of Cl). This suggests that these metals are dissolved primarily as Cl-complexes and neither oxygen fugacity nor the composition of the melt affects significantly their fluid/melt partitioning. By contrast, partition coefficients for Mo, B, As, Sb and Bi are highest in low salinity (1-2 mol/kg Cl) fluids with maximum values of D_Mo ∼ 20, D_B ∼ 15, D_As ∼ 13, D_Sb ∼ 8, D_Bi ∼ 15 indicating dissolution as non-chloride (e.g., hydroxy) complexes. Fluid/melt partition coefficients of copper are highly variable, but highest between vapor like fluids and silicate melt (D_Cu ? 2700), indicating an important role for ligands other than Cl. Partition coefficients for W generally increase with increasing chlorinity, but are exceptionally low in some of the studied brines which may indicate an effect of other parameters. Fluid/melt partition coefficients of Sn show a high variability but likely increase with the chlorinity of the fluid (D_Sn = 0.3-42, D_W = 0.8-60), and decrease with decreasing oxygen fugacity or melt peraluminosity.     

Monomethylmercury sources in a tropical artificial reservoir

The distribution and speciation of mercury (Hg) in the water column, the inputs (wet deposition and tributaries) and the outputs (atmospheric evasion and outlet) of an artificial partially anoxic tropical lake (Petit-Saut reservoir, French Guiana) were investigated on a seasonal basis in order to appraise the cycling and transformations of this metal. The total mercury (HgT) concentrations in the oxygenated epilimnetic waters averaged 5 ± 3 pmol L⁻¹ in the unfiltered samples (HgT_UNF) and 4 ± 2 pmol L⁻¹ in the dissolved (HgT_D) phase (<0.45 μm). On average, the monomethylmercury (MMHg) constituted 8%, 40% and 18% of the HgT in the dissolved phase, the particulate suspended matter and in the unfiltered samples, respectively. Covariant elevated concentrations of particulate MMHg and chlorophyll a in the epilimnion suggest that phytoplankton is an active component for the MMHg transfer in the lake. In the anoxic hypolimnion the HgT_UNF averages 13 ± 6 pmol L⁻¹ and the HgT_D 8 ± 4 pmol L⁻¹. The averages of MMHg_P and MMHg_D in hypolimnetic waters were two and three times the corresponding values of the epilimnion, 170 ± 90 pmol g⁻¹ and 0.9 ± 0.5 pmol L⁻¹, respectively. In the long dry and wet seasons, at the flooded forest and upstream dam sampling stations, the vertical profiles of MMHg_D concentrations accounted for two distinct maxima: one just below the oxycline and the other near the benthic interface. Direct wet atmospheric deposition accounted for 14 moles yr⁻¹ HgT_UNF, with 0.7 moles yr⁻¹ as MMHg_UNF, while circa 76 moles yr⁻¹ of HgT_UNF, with 4.7 moles yr⁻¹ as MMHg_UNF, coming from tributaries. Circa 78 moles (∼17% as MMHg) are annually exported through the dam, while 23 moles yr⁻¹ of Hg⁰ evolve in the atmosphere. A mass balance calculation suggests that the endogenic production of MMHg_UNF attained 8.1 moles yr⁻¹, corresponding to a methylation rate of 0.06% d⁻¹. As a result, the Petit-Saut reservoir is a large man-made reactor that has extensively altered mercury speciation in favor of methylated species.     

Hydrogen isotope fractionation by Methanothermobacter thermoautotrophicus in coculture and pure culture conditions

We grew a hydrogen-utilizing methanogen, Methanothermobacter thermoautotrophicus strain ΔH, in coculture and pure culture conditions to evaluate the hydrogen isotope fractionation associated with carbonate reduction under low (< several tens of μM; coculture) and high (>6 mM; pure culture) concentrations of H₂ in the headspace. In the cocultures, which were grown at 55 °C with a thermophilic butyrate-oxidizing syntroph, the hydrogen isotopic relationship between methane and water was well represented by the following equation:

δD_CH4=0.725(±0.003)·δD_H2O-275(±3),     

The effect of aqueous diffusion on the fractionation of chlorine and bromine stable isotopes

Diffusive isotopic fractionation factors are important in order to understand natural processes and have practical application in radioactive waste storage and carbon dioxide sequestration. We determined the isotope fractionation factors and the effective diffusion coefficients of chloride and bromide ions during aqueous diffusion in polyacrylamide gel. Diffusion was determined as functions of temperature, time and concentration. The effect of temperature is relatively large on the diffusion coefficient (D) but only small on isotope fractionation. For chlorine, the ratio, D_³⁵Cl/D_³⁷Cl varied from 1.00128 ± 0.00017 (1σ) at 2 °C to 1.00192 ± 0.00015 at 80 °C. For bromine, D_⁷⁹Br/D_⁸¹Br varied from 1.00098 ± 0.00009 at 2 °C to 1.0064 ± 0.00013 at 21 °C and 1.00078 ± 0.00018 (1σ) at 80 °C. There were no significant effects on the isotope fractionation due to concentration. The lack of sensitivity of the diffusive isotope fractionation to anything at the most common temperatures (0 to 30 °C) makes it particularly valuable for application to understanding processes in geological environments and an important natural tracer in order to understand fluid transport processes.     

Trace element partitioning between ilmenite, armalcolite and anhydrous silicate melt: Implications for the formation of lunar high-Ti mare basalts

We performed a series of experiments at high pressures and temperatures to determine the partitioning of a wide range of trace elements between ilmenite (Ilm), armalcolite (Arm) and anhydrous lunar silicate melt, to constrain geochemical models of the formation of titanium-rich melts in the Moon. Experiments were performed in graphite-lined platinum capsules at pressures and temperatures ranging from 1.1 to 2.3 GPa and 1300-1400 °C using a synthetic Ti-enriched Apollo ‘black glass’ composition in the CaO-FeO-MgO-Al₂O₃-TiO₂-SiO₂ system. Ilmenite-melt and armalcolite-melt partition coefficients (D) show highly incompatible values for the rare earth elements (REE) with the light REE more incompatible compared to the heavy REE ( 0.0020 ± 0.0010 to 0.069 ± 0.010 for ilmenite; 0.0048 ± 0.0023 to 0.041 ± 0.008 for armalcolite). D values for the high field strength elements vary from highly incompatible for Th, U and to a lesser extent W (for ilmenite: 0.0013 ± 0.0008, 0.0035 ± 0.0015 and 0.039 ± 0.005, and for armalcolite 0.008 ± 0.003, 0.0048 ± 0.0022 and 0.062 ± 0.03), to mildly incompatible for Nb, Ta, Zr, and Hf (e.g. 0.28 ± 0.05 and : 0.76 ± 0.07). Both minerals fractionate the high field strength elements with D_Ta/D_Nb and D_Hf/D_Zr between 1.3 and 1.6 for ilmenite and 1.3 and 1.4 for armalcolite. Armalcolite is slightly more efficient at fractionating Hf from W during lunar magma ocean crystallisation, with D_Hf/D_W = 12-13 compared to 6.7-7.5 for ilmenite. The transition metals vary from mildly incompatible to compatible, with the highest compatibilities for Cr in ilmenite (D ∼ 7.5) and V in armalcolite (D ∼ 8.1). D values show no clear variation with pressure in the small range covered.Crystal lattice strain modelling of D values for di-, tri- and tetravalent trace elements shows that in ilmenite, divalent elements prefer to substitute for Fe while armalcolite data suggest REE replacing Mg. Tetravalent cations appear to preferentially substitute for Ti in both minerals, with the exception of Th and U that likely substitute for the larger Fe or Mg cations. Crystal lattice strain modelling is also used to identify and correct for very small (∼0.3 wt.%) melt contamination of trace element concentration determinations in crystals.Our results are used to model the Lu-Hf-Ti concentrations of lunar high-Ti mare basalts. The combination of their subchondritic Lu/Hf ratios and high TiO₂ contents requires preferential dissolution of ilmenite or armalcolite from late-stage, lunar magma ocean cumulates into low-Ti partial melts of deeper pyroxene-rich cumulates.     

Trace element partitioning between apatite and silicate melts

We present new experimental apatite/melt trace element partition coefficients for a large number of trace elements (Cs, Rb, Ba, La, Ce, Pr, Sm, Gd, Lu, Y, Sr, Zr, Hf, Nb, Ta, U, Pb, and Th). The experiments were conducted at pressures of 1.0 GPa and temperatures of 1250 °C. The rare earth elements (La, Ce, Pr, Sm, Gd, and Lu), Y, and Sr are compatible in apatite, whereas the larger lithophile elements (Cs, Rb, and Ba) are strongly incompatible. Other trace elements such as U, Th, and Pb have partition coefficients close to unity. In all experiments we found D_Hf > D_Zr, D_Ta ≈ D_Nb, and D_Ba > D_Rb > D_Cs. The experiments reveal a strong influence of melt composition on REE partition coefficients. With increasing polymerisation of the melt, apatite/melt partition coefficients for the rare earth elements increase for about an order of magnitude. We also present some results in fluorine-rich and water-rich systems, respectively, but no significant influence of either H₂O or F on the partitioning was found. Furthermore, we also present experimentally determined partition coefficients in close-to natural compositions which should be directly applicable to magmatic processes.     

Oxygen isotope fractionation between synthetic aragonite and water: Influence of temperature and Mg concentration

Aragonite was precipitated in the laboratory at 0, 5, 10, 25, and 40 °C to determine the temperature dependence of the equilibrium oxygen isotope fractionation between aragonite and water. Forced CO₂ degassing, passive CO₂ degassing, and constant addition methods were employed to precipitate aragonite from supersaturated solutions, but the resulting aragonite-water oxygen isotope fractionation was independent of the precipitation method. In addition, under the experimental conditions of this study, the effect of precipitation rate on the oxygen isotope fractionation between aragonite and water was almost within the analytical error of ±∼0.13‰ and thus insignificant. Because the presence of Mg²⁺ ions is required to nucleate and precipitate aragonite from Na-Ca-Cl-HCO₃ solutions under these experimental conditions, the influence of the total Mg²⁺ concentration (up to ∼0.9 molal) on the aragonite-water oxygen isotope fractionation was examined at 25 °C. No significant Mg²⁺ ion effect, or oxygen isotope salt effect, was detected up to 100 mmolal total Mg²⁺ but a noticeable isotope salt effect was observed at ∼0.9 molal total Mg²⁺.On the basis of results of the laboratory synthesis experiments, a new expression for the aragonite-water fractionation is proposed over the temperature range of 0-40 °C:

1000lnα_{aragonite-water}=17.88±0.13(10³/T)-31.14±0.46     

Li,Sr, Mg,and Na in foraminiferal calcite shells from laboratory culture,sediment traps,and sediment cores

Constant-temperature laboratory culture experiments of the planktonic foraminiferal species Globigerinoides sacculifer (Brady) suggest that the ratios of Li and Sr to Ca in the shells are a function of these ratios in the culture solutions. $\text{Mg}\text{Ca}$ and $\text{Na}\text{Ca}$ in the shells did not vary with changes of these ratios in the culture solution. These are the first direct determinations of the relationship between foraminiferal shell chemistry and solution composition.The possibility of temperature dependence for the minor elemental composition of foraminiferal shells was also investigated in the laboratory and by analysis of several planktonic and one benthic foraminiferal species from sediment trap and sediment core samples. The $\text{Sr}\text{Ca}$, $\text{Mg}\text{Ca}$, and $\text{Na}\text{Ca}$ ratios in the natural samples roughly correlate with calcification temperature, whereas differences in the Li/Ca ratios are small and not systematically related to temperature. However, laboratory culture experiments at 20°C and 30°C showed no variation in the $\text{Li}\text{Ca}$, $\text{Sr}\text{Ca}$, $\text{Mg}\text{Ca}$, and $\text{Na}\text{Ca}$ ratios with calcification temperature for the planktonic foraminifera G. sacculifer and Orbulina universa. Therefore, observed differences in the $\text{Sr}\text{Ca}$, $\text{Mg}\text{Ca}$, and $\text{Na}\text{Ca}$ ratios for the sediment trap and core foraminiferal samples cannot be ascribed to direct effects of calcification temperature, but may be due to some other environmental factor which is correlated with temperature.