Silicon Isotope Fractionation Research Articles

Silicon isotope fractionation during abiotic silica precipitation at low temperatures: Inferences from flow-through experiments

Silicon isotopes have considerable potential as proxy for (near-) surface processes and environmental conditions. However, unambiguous interpretations of isotope signatures in natural silica deposits are often hampered by a lack of independent quantitative information on isotopic fractionations operating under the environmental conditions of interest. We performed seeded silica precipitation experiments using flow-through reactors in the 10–60°C temperature range to alleviate this problem. The principal objective was to quantify the silicon isotope fractionations during controlled precipitation of amorphous silica from a flowing aqueous solution. The experiments were designed to simulate silica deposition induced by a temperature drop, with particular relevance for (near-) surface hydrothermal systems associated with steep temperature gradients.Monitored differences in silicon isotope ratios (30Si/28Si and 29Si/28Si) between input and output solutions demonstrated a systematic sequence in behavior. During an initial time interval, that is, before the reaction system reached steady state, the observed isotope shifts were influenced by dissolution of the seed material, the saturation state of the solution and the specific surface area of the seeds. After reaching steady state, the selective incorporation of silicon isotopes by the solid phase exhibited an explicit temperature dependency: the lighter isotopes were preferentially incorporated, and apparent fractionation magnitudes increased with decreasing temperature.Calculated magnitudes of silicon isotope fractionations between precipitated and dissolved silica (Δ30Si=δ30Siprecipitate (calculated)−δ30Siinput solution) were −2.1‰ at 10°C, −1.2‰ at 20°C, −1.0‰ at 30°C, −0.5‰ at 40°C, 0.1‰ at 50°C, and 0.2‰ at 60°C (s.d.⩽0.6‰, based on replicate experiments). Hence, fractionation was nearly insignificant at temperatures ⩾50°C. Apart from this relationship with temperature, our results indicate that the effective Si isotope fractionation during precipitation from a solution is subject to changes in the saturation state, reactive surface area and flow regime. We therefore infer that, to a significant extent, solid–fluid fractionation in natural (near-) surface environments is system dependent.

First-principles calculations of equilibrium silicon isotope fractionation among mantle minerals

Silicon isotope fractionation factors for mantle silicate minerals, including olivine, wadsleyite, ringwoodite, pyroxenes, garnet (pyrope), majorite, and Mg-perovskite, are calculated using density functional theory. Our results show that equilibrium fractionations of Si isotopes are negligible among pyroxenes, olivine, and pyrope, but are significant between olivine and its polymorphs (wadsleyite and ringwoodite). There is also significant Si isotope fractionations between mantle minerals with different Si coordination numbers (CN), such as Mg-perovskite (CN=6) and olivine polymorphs (CN=4). When in equilibrium with each other, 30Si/28Si decreases in the order of olivine>pyroxenes>wadsleyite>majorite>ringwoodite>Mg-perovskite.Our calculation predicts significant Si isotope fractionation between mantle minerals, e.g., perovskite vs. ringwoodite, majorite vs. pyroxene, and olivine vs. its polymorphs even at high pressure and temperature conditions of deep mantle. The Si CN in silicate melt increases with increasing pressure, implying that Si isotope fractionation between silicate and metal could be a function of pressure. Our results suggest that Si isotopic fractionation factor between silicate and metal may decrease with increasing pressure; consequently, Si isotopic fractionation factor obtained from low pressure experiments may not be applicable to Si isotope fraction during core formation which occurred at high pressure. Finally, Si isotopes could also be fractionated between perovskite-rich mantle and residual melt during magma-ocean cooling in the lower mantle because of their different Si CNs. If such primordial signature is not destroyed and partially preserved through the Earth’s history, significant Si isotope heterogeneity could still exist between the upper and lower mantle.

Silicon and oxygen isotopic composition of igneous rocks from the eastern Manus Basin

This paper reports silicon and oxygen isotopes of 20 kinds of igneous rocks and their major elements from the eastern Manus Basin. Combining silicon and oxygen isotopic data from other studies, we suppose that both delta Si-30 and delta O-18 values increase with the increasing of SiO2 content. It means that the fractionation of silicon and oxygen isotopes are affected by the silica content. The positive correlation between CaO/Al2O3 ratios and MgO and that between Si/Al and SiO2 content indicate that clinopyroxene is the predominant mineral phase in our samples. We suppose that the fractionation of silicon and oxygen isotopes are influenced by mineral fractional crystallization. Probably, it is due to their different silicon and oxygen bridges. In this study, the delta S-30(imean value) = -0.17aEuro degrees +/- 0.17aEuro degrees and delta O-18(mean value) = +6.07aEuro degrees +/- 0.57aEuro degrees are higher than normal delta Si-30 and delta O-18 values of mantle, and we propose that these igneous rocks in the eastern Manus Basin are affected by hydrothermal alteration.

Silicon isotope fractionation in silicate minerals: Insights from first-principles models of phyllosilicates, albite and pyrope

Isotopic fractionation factors for oxygen and silicon in phyllosilicates (pyrophyllite, talc), albite and pyrope have been calculated using first-principles methods based on density functional theory. Based on exhaustive analysis of numerical convergence, we also update our previous calculations on enstatite and forsterite silicon fractionation properties. Calculated oxygen isotope fractionations agree well with existing estimates for talc and albite. In the case of silicon, qualitative agreement is found with natural data. For phyllosilicates (kaolinite, lizardite, pyrophyllite, talc), Si isotope fractionation properties appear to be correlated with stoichiometry:(1)1000lnα30Siphyllosilicate–quartz=aMg(T)·Mgeq.+aAl(T)·Aleq.Sieq.,where Sieq.=#Si, Aleq.=34#Al and Mgeq.=12#Mg (cation equivalents) are the charge-weighed stoichiometric coefficients of each cation, normalized to the charge of the silicon atom, and aX(T) are proportionality coefficients depending on temperature. It is suggested that the effect of cation X on Si isotope fractionation (i.e. aX(T)) will increase with decreasing electronegativity of X. Si isotope fractionation is further correlated with Si–O distances, suggesting a crystal chemical explanation for relation (1) in terms of electron donation effects. This relationship appears valid for quartz, pyrope and enstatite (R2=0.99,n=7), but forsterite is strongly anomalous (error of 0.7‰ at 600°C). These models indicate that attention should be given to chemical compositions in Si isotope studies. Relation (1) would explain the enrichment in heavy silicon isotopes accompanying magmatic differentiation.

What controls silicon isotope fractionation during dissolution of diatom opal?

The silicon isotope composition of opal frustules from photosynthesising diatoms is a promising tool for studying past changes in the marine silicon cycle, and indirectly that of carbon. Dissolution of this opal may be accompanied by silicon isotope fractionation that could disturb the pristine silicon isotope composition of diatom opal acquired in the surface ocean. It has previously been shown that dissolution of fresh and sediment trap diatom opal in seawater does fractionate silicon isotopes. However, as the mechanism of silicon isotope fractionation remained elusive, it is uncertain whether opal dissolution in general is associated with silicon isotope fractionation considering that opal chemistry and surface properties are spatially and temporally (i.e. opal of different age) diverse.In this study we dissolved sediment core diatom opal in 5mM NaOH and found that this process is not associated with significant silicon isotope fractionation. Since no variability of the isotope effect was observed over a wide range of dissolution rates, we can rule out the suggestion that back-reactions had a significant influence on the net isotope effect. Similarly, we did not observe an impact of temperature, specific surface area, or degree of undersaturation on silicon isotope partitioning during dissolution, such that these can most likely also be ruled out as controlling factors. We discuss the potential impacts of the chemical composition of the dissolution medium and age of diatom opal on silicon isotope fractionation during dissolution. It appears most likely that the controlling mechanism of silicon isotope fractionation during dissolution is related to the reactivity, or potentially, aluminium content of the opal. Such a dependency would imply that silicon isotope fractionation during dissolution of diatom opal is spatially and temporally variable. However, since the isotope effects during dissolution are small, the silicon isotope composition of diatom opal appears to be robust against dissolution in the deep sea sedimentary environment.

Effects of growth and dissolution on the fractionation of silicon isotopes by estuarine diatoms

Studies of silicon (Si) isotope fractionation during diatom growth in open ocean systems have documented lower Si isotopic values (δ30Si) in the biogenic silica of diatom frustules compared to dissolved silicon. Recent findings also indicate that Si isotope fractionation occurs during dissolution of diatom frustules, producing higher δ30Si values in the remaining biogenic silica. This study focuses on diatoms from high production areas in estuarine and coastal areas that represent approximately 30–50% of the global marine primary production. Two species of diatoms, Thalassiosira baltica and Skeletonema marinoi, were isolated from the brackish Baltic Sea, one of the largest estuarine systems in the world. These species were used for laboratory investigations of Si isotope fractionation during diatom growth and the subsequent dissolution of the diatom frustules. Both species of diatoms give an identical Si isotope fractionation factor during growth of −1.50±0.36‰ (2σ) for 30Si, which falls in the range of −2.09‰ to −0.55‰ of published data. Our results also suggest a dissolution-induced Si isotope fractionation factor of −0.86‰ at early stage of dissolution, but this effect was observed only in DSi and no significant Si isotope change was observed for BSi. The growth and dissolution results are applied to a Baltic Sea sediment core to reconstruct DSi utilization by diatoms, and found to be in agreement with the observed DSi uptake rates in the overlying water column during diatom growth.

Chemical and Isotopic Characteristics of the Water and Suspended Particulate Materials in the Yangtze River and Their Geological and Environmental Implications

Chemical and Isotopic Characteristics of the Water and Suspended Particulate Materials in the Yangtze River and Their Geological and Environmental Implications

Experimental determination of the Si isotope fractionation factor between liquid metal and liquid silicate

The conditions of core formation and the abundances of the light elements in Earthʼs core remain debated. Silicon isotope fractionation provides a tool contributing to this subject. We present experimentally determined Si isotope fractionation factors between liquid metal and liquid silicate at 1450 °C and 1750 °C, which allow calibrating the temperature dependence of Si isotope fractionation. Experiments were performed in a centrifuging piston cylinder at 1 GPa, employing both graphite and MgO capsules. Tin was used to lower the melting temperature of the metal alloys for experiments performed at 1450 °C. Tests reveal that neither Sn nor C significantly affects Si isotope fractionation. An alkaline fusion technique was employed to dissolve silicate as well as metal phases prior to ion exchange chemistry and mass spectrometric analysis. The results show that metal is consistently enriched in light isotopes relative to the silicate, yielding average metal–silicate fractionation factors of −1.48±0.08‰ and −1.11±0.14‰ at 1450 °C and 1750 °C, respectively. The temperature dependence of equilibrium Si isotope fractionation between metal and silicate can thus be described as Δ30SiMetal-Silicate=−4.42(±0.05)×106/T2. The Si isotope equilibrium fractionation is thus about 1.7 times smaller than previously proposed on the basis of experiments. A consequence of this smaller fractionation is that the calculated difference between the Si isotope composition of the bulk Earth and that of the bulk silicate Earth generated by core formation is smaller than previously thought. It is therefore increasingly difficult to match the Si isotope composition of the bulk silicate Earth with that of chondrites for metal–silicate equilibration temperatures above ∼2500 K. This suggests that Si isotopes were more sensitive to the early stages of core formation when low oxygen fugacities allowed significant incorporation of Si into metal.

Characteristics of silicon and oxygen isotopic compositions of basalts near East Pacific Rise 13°N

In this study, 13 groups of silicon and oxygen isotopes and major elements of the basalts near the East Pacific Rise 13°N are used to study the fractionation of silicon and oxygen isotopes. Among these data, Δ30Si values of basalts vary from −0.4‰ to 0.2‰ with a mean value of Δ30Si of (−0.18±0.22)‰. The δ18O values range from 4.1‰ to 6.4‰ with a mean Δ18O value of (+5.35±0.73)‰. Since the δ30Si values increase in the series of basalt-basaltic andesite-andesite, and δ18O values display a positive correlation with the SiO2 content, we propose that the fractionation of silicon and oxygen isotopes is influenced by the SiO2 content in igneous rocks. Compared with the igneous rocks from Manus Basin with clinopyroxene as their dominant mineral phase, MORBs in this study containing olivine and plagioclase as primary minerals have lower Δ18O and Δ30Si values, indicating that the fractionation of silicon and oxygen isotopes is also affected by different Si-O bridges in silicate minerals. Furthermore, our samples from the EPR are defined as E-MORB based on K/Ti ratios. Probably, the difference in Δ30Si and Δ18O between our samples and a normal MORB are cause by the enriched components in E-MORBs.

Silicon isotope variations in Central Siberian rivers during basalt weathering in permafrost-dominated larch forests

This work is devoted to the characterization of natural mechanisms of silicon isotope fractionation within Siberian watersheds and predicting the climate warming effect on Si fluxes from the land to the Arctic Ocean. To unravel the different sources of silica generated by basalt weathering in Central Siberia under permafrost and larch deciduous forest conditions, we measured the Si isotopic composition of large and small rivers, surface flow, interstitial soil solutions, plant litter and soils. The average annual discharge-weighted δ30Si values of the second largest tributary of the Yenissei River, Nyzhnaya Tunguska and its main northern tributary (Kochechum) are equal to 1.08±0.10‰ and 1.67±0.15‰, respectively, while their average annual Si concentrations are very similar (3.46 and 3.50mg/L, respectively). During summer baseflow, the dissolved Si isotope composition of both large rivers and a small stream ranges between 1.5 and 2.5‰. This is much heavier compared to the source basaltic rocks but similar to the fresh litter of Larix gmelinii, the dominating tree species in this region. It could be consistent with litter degradation in the uppermost soil horizons being the dominant source of solutes annually exported by Central Siberian rivers. During spring flood, accounting for 60–80% of annual Si flux, the δ30Si of the large rivers' dissolved load decreases by 1–1.5‰, thus approaching the value of the bedrock and the silicate suspended matter of the rivers (RSM). This may reflect the dissolution of the silicate suspended load at high water/mineral ratio. The winter δ30Si values of the large river dissolved load range between 1.0 and 2.5‰. During this period, contributing to ≤10% of the annual Si chemical flux, the interaction between bedrock (porous tuffs) and deep ground waters occurs at a very high solid/solution ratio, leading to the precipitation of isotopically light secondary minerals and enrichment of 30Si in the fluids that feed the river through the unfrozen flowpaths. Results of this study imply that more than a half of the silica transported by Siberian rivers may transit through the biogenic pool and that, like in other stable basaltic regions, bedrock–water interactions account for a lesser fraction of the silica flux. As a result of projected future climate warming and weathering increases in boreal regions, the δ30Si isotopic composition of large Siberian rivers is likely to shift towards less positive values.

Species-dependent silicon isotope fractionation by marine diatoms

Fractionation of silicon (Si) isotopes was measured in seven species (nine strains) of polar and sub-polar marine diatoms grown in semi-continuous unialgal cultures under optimal irradiance and temperature for each diatom strain. Results from this work provide the first evidence that Si isotope fractionation by diatoms is species-dependent. The greatest difference in the Si isotope fractionation factor (ε) was observed between two Southern Ocean diatoms, Fragilariopsis kerguelensis (−0.54‰, average for two strains) and Chaetoceros brevis (−2.09‰). The ε for the other species, both polar and sub-polar, ranged from −0.72‰ to −1.21‰. The two remaining polar diatoms had ε values of −0.74±0.05‰ for Thalassiosira antarctica, and −1.21±0.04‰ for Thalassiosira nordenskioeldii, while the sub-polar species had ε values of −0.72±0.04‰ for Thalassiosira weissflogii, −0.88±0.06‰ for Thalassiosira pseudonana (CCCM58), −0.97±0.14‰ for Thalassiosira pseudonana (CCMP1014), and −1.15±0.03‰ for Porosira glacialis. The range in ε for the diatoms evaluated in this study may be large enough to significantly impact the Si isotope composition measured in diatom opal (δ30Si-bSiO2) from marine sediments and its subsequent interpretation. To test the influence of diatom taxonomic composition on δ30Si-bSiO2, we developed a model that considered the relative abundance of diatom species and the ε values (from this study) for each species present within the sediment core (i.e. weighted-average ε). The model was applied to records from a Southern Ocean sediment core (TN057-13) where both diatom abundance and δ30Si-bSiO2 data were available. The analysis indicated that 67% of the variation in δ30Si-bSiO2 could be explained by species-dependent Si isotope fractionation. We suggest that future work should assess phytoplankton taxonomic composition when using δ30Si-bSiO2 as a proxy for Si utilization.

The relationship between silicon isotope fractionation in sponges and silicic acid concentration: Modern and core-top studies of biogenic opal

Recent work has shown the silicon isotope composition, denoted by δ30Si, of deep-sea sponges reflects the concentration of ambient silicic acid (Si(OH)4) in seawater. However, existing calibrations are based predominantly on living sponges collected from the Southern Ocean. These data cannot, however, be used to determine whether other parameters that correlate with silicic acid in the Southern Ocean, such as temperature and salinity, influence δ30Si of sponges. Furthermore, the published data do not demonstrate whether disaggregated core-top sedimentary spicules preserve the primary δ30Si signal recorded in living sponges. Here, we address both of these issues. We refine and widen the existing calibration by including a global distribution of modern sponges. In addition, we provide the first systematic calibration from spicules picked from core-top sediments that covers sites from different ocean basins. The relationship between Si(OH)4 and δ30Si in sponge spicules is the same in different ocean basins, between specimens that grew in different temperature and salinity conditions. Our core-top data agree well with the modern sponge calibration indicating there are no significant post-depositional effects or early diagenetic overprints. These two new datasets support the assertion that sponge δ30Si can be used as a proxy for silicic acid concentrations in the past.

High-temperature Si isotope fractionation between iron metal and silicate

The Si stable isotope fractionation between metal and silicate has been investigated experimentally at 1800, 2000, and 2200 °C. We find that there is a significant silicon stable isotope fractionation at high temperature between metal and silicate in agreement with Shahar et al. (2009). Further we find that this fractionation is insensitive to the structure and composition of the silicate as the fractionation between silicate melt and olivine is insignificant within the error of the analyses. The temperature-dependent silicon isotope fractionation is Δ 30Si silicate–metal = 7.45 ± 0.41 × 10 6/ T 2. We also demonstrate the viability of using laser ablation MC-ICPMS as a tool for measuring silicon isotope ratios in high pressure and temperature experiments.

Silicon isotope fractionation during magmatic differentiation

The Si isotopic composition of Earth’s mantle is thought to be homogeneous (δ 30Si = −0.29 ± 0.08‰, 2 s.d.) and not greatly affected by partial melting and recycling. Previous analyses of evolved igneous material indicate that such rocks are isotopically heavy relative to the mantle. To understand this variation, it is necessary to investigate the degree of Si isotopic fractionation that takes place during magmatic differentiation. Here we report Si isotopic compositions of lavas from Hekla volcano, Iceland, which has formed in a region devoid of old, geochemically diverse crust. We show that Si isotopic composition varies linearly as a function of silica content, with more differentiated rocks possessing heavier isotopic compositions. Data for samples from the Afar Rift Zone, as well as various igneous USGS standards are collinear with the Hekla trend, providing evidence of a fundamental relationship between magmatic differentiation and Si isotopes. The effect of fractionation has been tested by studying cumulates from the Skaergaard Complex, which show that olivine and pyroxene are isotopically light, and plagioclase heavy, relative to the Si isotopic composition of the Earth’s mantle. Therefore, Si isotopes can be utilised to model the competing effects of mafic and felsic mineral fractionation in evolving silicate liquids and cumulates. At an average SiO 2 content of ∼60 wt.%, the predicted δ 30Si value of the continental crust that should result from magmatic fractionation alone is −0.23 ± 0.05‰ (2 s.e.), barely heavier than the mantle. This is, at most, a maximum estimate, as this does not take into account weathered material whose formation drives the products toward lighter δ 30Si values. Mass balance calculations suggest that removal of continental crust of this composition from the upper mantle will not affect the Si isotopic composition of the mantle.

Silicon isotopic fractionation in marine sponges: A new model for understanding silicon isotopic variations in sponges

The silicon (Si) isotope ( δ 30Si) composition of deep-sea sponges from near Antarctica, subantarctic waters (Tasmania Seamounts) and subtropical waters north of New Zealand vary widely between + 0.87‰ and − 3.40‰ (vs. NBS28). Depth profiles show that sponge δ 30Si compositions trend to lower values with increasing depth. This is exemplified by sponges from the Tasmania Seamounts where δ 30Si varies from + 0.87‰ to − 3.13‰ over a depth range from 100 m to 1200 m. These changes in δ 30Si of sponges are inconsistent with a Rayleigh type isotope fractionation model requiring constant δ 30Si fractionation between sponge and seawater. We conclude that overall Si isotope fractionation Δ 30Si ( δ 30Si sponge – δ 30Si seawater) is influenced by seawater Si concentration, with more fractionated (lower) isotope values being associated with sponges collected from waters high in Si. We invoke and fit a model whereby the Δ 30Si fractionation varies as a function Si influx and efflux. Using this model it appears that Δ 30Si fractionation during transport into the sponge is constant at − 1.34‰. The model also shows asymptotic behaviour with Δ 30Si trending towards a maximum of − 6.02‰ at very high Si concentrations. These results suggest that the δ 30Si composition of fossil spicules may be useful for reconstruction paleo-Si concentrations during the past.

Impact of soil weathering degree on silicon isotopic fractionation during adsorption onto iron oxides in basaltic ash soils, Cameroon

The sequestration of silicon in soil clay-sized iron oxides may affect the terrestrial cycle of Si. Iron oxides indeed specifically adsorb aqueous monosilicic acid (H 4SiO 4 0), thereby influencing Si concentration in soil solution. Here we study the impact of H 4SiO 4 0 adsorption on the fractionation of Si isotopes in basaltic ash soils differing in weathering degree (from two weathering sequences, Cameroon), hence in clay and Fe-oxide contents, and evaluate the potential isotopic impact on dissolved Si in surrounding Cameroon rivers. Adsorption was measured in batch experiment series designed as function of time (0–72 h) and initial concentration (ic) of Si in solution (0.61–1.18 mM) at 20 °C, constant pH (5.5) and ionic strength (1 mM). After various soil-solution contact times, the δ 30Si vs. NBS28 compositions were determined in selected solutions by MC-ICP-MS (Nu Plasma) in medium resolution, operating in dry plasma with Mg doping with an average precision of ±0.15‰ (±2σ SEM). The quantitative adsorption of H 4SiO 4 0 by soil Fe-oxides left a solution depleted in light Si isotopes, which confirms previous study on synthetic Fe-oxides. Measured against its initial composition (δ 30Si = +0.02 ± 0.07‰ (±2σ SD)), the solutions were systematically enriched in 30Si reaching maximum δ 30Si values ranging between +0.16‰ and +0.95‰ after 72 h contact time. The enrichment of the solution in heavy isotopes increased with increasing values of three parameters: soil weathering degree, iron oxide content, and proportion of short-range ordered Fe-oxide. The Si-isotopic signature of the solution was partly influenced by Si release, possibly through mineral dissolution and Si desorption from oxide surfaces, depending on soil type, highlighting the complex pattern of natural soils. Surrounding Cameroon rivers displayed a mean Si-isotopic signature of +1.19‰. Our data imply that in natural environments, H 4SiO 4 0 adsorption by soil clay-sized Fe-oxides at least partly impacts the Si-isotopic signature of the soil solution exported to water streams.

Silicon isotopic fractionation of CAI-like vacuum evaporation residues

Calcium-, aluminum-rich inclusions (CAIs) are often enriched in the heavy isotopes of magnesium and silicon relative to bulk solar system materials. It is likely that these isotopic enrichments resulted from evaporative mass loss of magnesium and silicon from early solar system condensates while they were molten during one or more high-temperature reheating events. Quantitative interpretation of these enrichments requires laboratory determinations of the evaporation kinetics and associated isotopic fractionation effects for these elements. The experimental data for the kinetics of evaporation of magnesium and silicon and the evaporative isotopic fractionation of magnesium is reasonably complete for Type B CAI liquids (Richter F. M., Davis A. M., Ebel D. S., and Hashimoto A. (2002) Elemental and isotopic fractionation of Type B CAIs: experiments, theoretical considerations, and constraints on their thermal evolution. Geochim. Cosmochim. Acta 66, 521–540; Richter F. M., Janney P. E., Mendybaev R. A., Davis A. M., and Wadhwa M. (2007a) Elemental and isotopic fractionation of Type B CAI-like liquids by evaporation. Geochim. Cosmochim. Acta 71, 5544–5564.). However, the isotopic fractionation factor for silicon evaporating from such liquids has not been as extensively studied. Here we report new ion microprobe silicon isotopic measurements of residual glass from partial evaporation of Type B CAI liquids into vacuum. The silicon isotopic fractionation is reported as a kinetic fractionation factor, α Si, corresponding to the ratio of the silicon isotopic composition of the evaporation flux to that of the residual silicate liquid. For CAI-like melts, we find that α Si = 0.98985 ± 0.00044 (2 σ) for 29Si/ 28Si with no resolvable variation with temperature over the temperature range of the experiments, 1600–1900 °C. This value is different from what has been reported for evaporation of liquid Mg 2SiO 4 (Davis A. M., Hashimoto A., Clayton R. N., and Mayeda T. K. (1990) Isotope mass fractionation during evaporation of Mg 2SiO 4. Nature 347, 655–658.) and of a melt with CI chondritic proportions of the major elements (Wang J., Davis A. M., Clayton R. N., Mayeda T. K., and Hashimoto A. (2001) Chemical and isotopic fractionation during the evaporation of the FeO–MgO–SiO 2–CaO–Al 2O 3–TiO 2–REE melt system. Geochim. Cosmochim. Acta 65, 479–494.). There appears to be some compositional control on α Si, whereas no compositional effects have been reported for α Mg. We use the values of α Si and α Mg, to calculate the chemical compositions of the unevaporated precursors of a number of isotopically fractionated CAIs from CV chondrites whose chemical compositions and magnesium and silicon isotopic compositions have been previously measured.

Fractionation of silicon isotopes during biogenic silica dissolution

Silicon isotopes have been investigated for their potential to reveal both past and present patterns of silicic acid utilization, primarily by diatoms, in surface waters of the ocean. Interpretation of this proxy has thus far relied on characteristic trends in the isotope composition of the dissolved and particulate silicon pools in the upper ocean, as driven by biological fractionation during the production of biogenic silica (bSiO 2, or opal) by diatoms. However, other factors which may influence the silicon isotope composition of diatom opal, particularly post-formational aging and maturation processes, remain largely uninvestigated. Here, we report a consistent fractionation of silicon isotopes during the physicochemical dissolution of diatom bSiO 2 suspended in seawater under closed conditions. This fractionation acts counter to that occurring during bSiO 2 production and at about half its absolute magnitude, with dissolution discriminating against the release of the heavier isotopes of silicon at an enrichment factor ε DSi–BSi of −0.55‰, corresponding to a fractionation factor α 30/28 of 0.99945. The enrichment factor did not vary with source material, indicating the lack of a significant species effect, or with temperature from 3 to 20 °C. Thus, the dissolution of bSiO 2 produces dissolved silicon with a δ 30Si value that is 0.55‰ more negative than its parent bSiO 2, an effect that must be accounted for when interpreting oceanic δ 30Si distributions. The δ 30Si values of both the dissolved and particulate silicon pools increased linearly as dissolution progressed, implying a measurable (±0.1‰) change in the relative δ 30Si of opal samples whenever the difference in preservation efficiency between them is >20%. This effect could account for ∼10–30% of the difference in diatom δ 30Si values observed between glacial and interglacial conditions. It is unlikely, however, that the inferred maximum possible change in δb 30SiO 2 of +0.55‰ would be manifested in situ, as a high mean percentage of dissolution would include complete loss of the more soluble members of the diatom assemblage.

Silicon isotopic fractionation during adsorption of aqueous monosilicic acid onto iron oxide

The quantification of silicon isotopic fractionation by biotic and abiotic processes contributes to the understanding of the Si continental cycle. In soils, light Si isotopes are selectively taken up by plants, and concentrate in secondary clay-sized minerals. Si can readily be retrieved from soil solution through the specific adsorption of monosilicic acid ( H 4 Si O 4 0 ) by iron oxides. Here, we report on the Si-isotopic fractionation during H 4 Si O 4 0 adsorption on synthesized ferrihydrite and goethite in batch experiment series designed as function of time (0–504 h) and initial concentration (ic) of Si in solution (0.21–1.80 mM), at 20 °C, constant pH (5.5) and ionic strength (1 mM). At various contact times, the δ 29Si vs. NBS28 compositions were determined in selected solutions (ic = 0.64 and 1.06 mM Si) by MC–ICP–MS in dry plasma mode with external Mg doping with an average precision of ±0.08‰ (±2 σ SEM). Per oxide mass, ferrihydrite (74–86% of initial Si loading) adsorbed more Si than goethite (37–69%) after 504 h of contact over the range of initial Si concentration 0.42–1.80 mM. Measured against its initial composition (δ 29Si = +0.01 ± 0.04‰ (±2 σ SD)), the remaining solution was systematically enriched in 29Si, reaching maximum δ 29Si values of +0.70 ± 0.07‰ for ferrihydrite and +0.50 ± 0.08‰ for goethite for ic 1.06 mM. The progressive 29Si enrichment of the solution fitted better a Rayleigh distillation path than a steady state model. The fractionation factor 29ε (±1 σ SD) was estimated at −0.54 ± 0.03‰ for ferrihydrite and −0.81 ± 0.12‰ for goethite. Our data imply that the sorption of H 4 Si O 4 0 onto synthetic iron oxides produced a distinct Si-isotopic fractionation for the two types of oxide but in the same order than that generated by Si uptake by plants and diatoms. They further suggest that the concentration of light Si isotopes in the clay fraction of soils is partly due to H 4 Si O 4 0 sorption onto secondary clay-sized iron oxides.

Silicon isotope fractionation between rice plants and nutrient solution and its significance to the study of the silicon cycle

The silicon isotope fractionation between rice plant and nutrient solution was studied experimentally. Rice plants were grown to maturity with the hydroponic culture in a naturally lit glasshouse. The nutrient solution was sampled for 14 times during the whole rice growth period. The rice plants were collected at various growth stages and different parts of the plants were sampled separately. The silica contents of the samples were determined by the gravimetric method and the silicon isotope compositions were measured using the SiF 4 method. In the growth process, the silicon content in the nutrient solution decreased gradually from 16 mM at starting stage to 0.1–0.2 mM at harvest and the amount of silica in single rice plant increased gradually from 0.00013 g at start to 4.329 g at harvest. Within rice plant the SiO 2 fraction in roots reduced continuously from 0.23 at the seedling stage, through 0.12 at the tiller stage, 0.05 at the jointing stage, 0.023 at the heading stage, to 0.009 at the maturity stage. Accordingly, the fraction of SiO 2 in aerial parts increased from 0.77, through 0.88, 0.95, 0.977, to 0.991 for the same stages. The silicon content in roots decreased from the jointing stage, through the heading stage, to the maturity stage, parallel to the decrease of silicon content in the nutrient solution. At the maturity stage, the silicon content increased from roots, through stem and leaves, to husks, but decreased drastically from husks to grains. These observations show that transpiration and evaporation may play an important role in silica transportation and precipitation within rice plants. It was observed that the δ 30Si of the nutrient solution increased gradually from −0.1‰ at start to 1.5‰ at harvest, and the δ 30Si of silicon absorbed by bulk rice plant increased gradually from −1.72‰ at start to −0.08‰ at harvest, reflecting the effect of the kinetic silicon isotope fractionation during silicon absorption by rice plants from nutrient solutions. The calculated silicon isotope fractionation factor between the silicon instantaneously absorbed by rice roots and the silicon in nutrient solution vary from 0.9983 at start to 0.9995 at harvest, similar to those reported for bamboo, banana and diatoms in direction and extent. In the maturity stage, the δ 30Si value of rice organs decreased from −1.33‰ in roots to −1.98‰ in stem, and then increased through −0.16‰ in leaves and 1.24‰ in husks, to 2.21‰ in grains. This trend is similar to those observed in the field grown rice and bamboo. These quantitative data provide us a solid base for understanding the mechanisms of silicon absorption, transportation and precipitation in rice plants and the role of rice growth in the continental Si cycle.