Eizo Nakamura

Possible eastward extension of Chinese collision belt in South Korea: The Imjingang belt

Structural, petrological, and geochronological data from the middle Korean peninsula indicate that the Qinling-Dabie-Sulu collisional belt of east-central China crosses the Yellow Sea and extends into the Imjingang belt. The Yeoncheon complex, first identified as the western Imjingang belt, comprises primarily north-dipping metamorphic sequences: (1) the northern Jingok unit, consisting of Barrovian-type metapelites, and (2) the southern Samgot unit, consisting of calc-silicate and amphibolitic rocks. South-vergent structures with reverse-sense shearing are dominant in the Jingok unit, whereas late normal-sense shearing is pervasive in the Samgot unit and the deformed granitoid to the south. These structural patterns are interpreted to correspond to extensional deformation associated with uplift following compression in a collisional belt. Pressure-temperature ( P-T ) estimates from the amphibolites suggest a high-P amphibolite-facies metamorphism (8–13 kbar and 630–790 °C), possibly evolving from eclogite facies conditions along a clockwise P-T path. Sm-Nd and Rb-Sr geochronological data suggest that the amphibolites emplaced in Late Proterozoic time were metamorphosed during Permian-Triassic time.

Boron isotope systematics of marine sediments

Abstract Boron contents and boron isotopic compositions were determined for modern and ancient (Permian to Miocene) marine sediments, including pelagic clay, calcareous ooze, siliceous ooze and neritic clay sediments. δ11B values of modern marine sediments range from −6.6 to +4.8‰. Isotopic variation is controlled by the simple mixing of four major constituents, detritus of continental origin, marine smectite, biogenic carbonates and biogenic silica. Detritus of continental origin, with an average δ11B value of −13 to −8‰, is the low-δ11B end-member constituent of marine sediments and its boron is largely controlled by the concentration of illite which originates from wind or fluvial transport. Marine smectite, biogenic carbonate and biogenic silica, on the other hand, represent the high-δ11B end members, with δ11B values of +2.3 to +9.2‰, +8.0 to +26.2‰ and +2.1 to +4.5‰, respectively. These high δ11B values are the result of the equilibrium uptake of boron from seawater. Spatial variations in boron isotopes in the Pacific sediments are essentially due to the distribution of the above four constituents. Although ancient argillaceous sediments (shale and slate) have boron contents that are identical with those of modern equivalents, boron contents of limestone and chert are distinctly lower than those of modern calcareous and siliceous oozes. Ancient marine sediments have systematically lower δ11B values (−17.0 to −5.6‰) than those of the modern sediments. The lower δ11B values can be caused by diagenesis, which induces (1) preferential removal of high-δ11B boron in calcium carbonate and silica during recrystallization and (2) boron isotopic exchange in the course of the smectite/illite transition. The observed boron isotopic compositions of ancient argillaceous sediments are distinctly different from those of fresh and altered MORB. Therefore, boron isotope systematics will be useful in identifying components from the descending oceanic slab involved in the formation of island arc magma and in investigating mantle-crust recycling through subduction processes.

Across-arc variation of Li isotopes in lavas and implications for crust/mantle recycling at subduction zones

Abstract Li isotope was analyzed in island arc volcanics from the Izu arc, Japan, to investigate geochemical processes in subduction zones. Li isotope ratios (δ7Li) and Li/Y of the arc lavas show clear across-arc variations, decreasing (δ7Li: +7.6 to +1.1‰, Li/Y: 0.36 to 0.25) with increasing depth to the Wadati–Benioff zone (150 to 210 km). This suggests that the amount of subduction component as a fluid added to the source region decreases with depth. δ7Li–Y/Li systematics of the arc lavas clearly indicate a simple mixing between two distinctive chemically homogenous endmembers, a slab-derived fluid and the mantle wedge. Furthermore, Li–B–Pb isotope systematics allow clear discrimination between the relative contribution of altered oceanic crust (AOC), oceanic sediment and mantle wedge to arc lavas, and suggests that AOC is the dominant subduction component, whereas the contribution of oceanic sediment is extremely small (AOC/oceanic sediment = 97/3). The contrasting physicochemical properties for Li and B in mineral structures imply that Li may be less likely to migrate from the slab into the overlying mantle wedge than B. Thus the Li isotopic composition in the Earths surface material evolved under near-surface condition, could be more efficiently introduced into the deep mantle through subduction zones than the B isotopic signature, making Li isotopes a powerful geochemical tracer for better understanding of crust/mantle recycling.

Evaluation of the coprecipitation of incompatible trace elements with fluoride during silicate rock dissolution by acid digestion

Insoluble fluoride precipitates which form during HF digestion of mafic silicate rocks coprecipitate in their structures the trace elements such as Rb, Sr, Y, Cs, Ba, REE, Pb, Th and U, thus hindering their accurate determination. We have estimated quantitatively the coprecipitation of trace elements into such fluorides, and suggest an effective method of digestion that can suppress completely fluoride precipitation. Conventional acid digestion of three samples of mafic and ultramafic silicate rocks resulted in the precipitation of sticky material and very poor yields of certain trace element in the resultant solution. XRD analysis indicated that the precipitates were composed of fluorides such as CaAlF5, CaMg2Al2F12, Na0.88Mg0.88Al1.12(F,OH)6·H2O and MgF2, the formation of which depended on the major element composition of the rock sample. Coprecipitation of trace elements appeared to be strongly controlled by both ionic radius and valency of the elements as well as the species of the host fluoride precipitate, resulting in selective losses of the elements into these fluorides. On the other hand, almost 100% of the trace elements were recovered using larger amounts of HClO4 than is conventionally used and evaporating the sample to dryness in a step-wise fashion. Using this method, white precipitates were formed as oxides of high field strength elements after decomposition of the sample. Coprecipitation of trace elements of interest in this study with the oxides is negligible except for Th for which 0.5–3.2% by weight was coprecipitated probably as the insoluble oxide. As our method also results in negligible blank values, it can be used for both the accurate determination of trace element using ICP-MS as well as isotope analysis using TIMS.

Across-arc variations of isotope and trace element compositions from Quaternary basaltic volcanic rocks in northeastern Japan : implications for interaction between subducted oceanic slab and mantle wedge

Isotopic compositions of Pb, Sr, and Nd and concentrations of trace elements were determined for Quaternary island arc basaltic rocks from northeastern Japan. Sr and Pb isotopic ratios decrease, and Nd isotopic ratios increase from the volcanic front toward the back arc. The isotopic compositions nearest the back arc side are nearly identical to those of mid-ocean ridge basalt (MORB). The high field strength elements and heavy rare earth elements show homogeneous and MORB-like characteristics. These observations indicate that the mantle wedge beneath northeastern Japan originally had a MORB-type mantle composition that was homogeneous across the arc. Pb isotope compositions show a mixing relationship between mantle wedge and oceanic sediments reflecting the introduction of subduction component into the mantle wedge, Across-arc isotopic variations were caused by interaction between MORB-type mantle wedge and the subducting slab, and the amount of subduction component correlates with the depth to the slab. The isotopic compositions of subduction component are expressed by bulk mixing of 15 wt % of oceanic sediment and 85 wt % of altered MORB. Inversion analyses of isotopic compositions using two-component mixing relationships show that the Sr/Nd and Pb/Nd ratios in subduction component decrease with increasing depth to the slab, while the Sr/Pb ratio is nearly constant. These changes can be explained only by a preferential discharge of the elements into the wedge mantle associated with continuous dehydration of the subducting slab. The present study further demonstrates that a very wide range of isotopic and elemental compositions in island arc magmas is a consequence of the interaction between subducting slab and mantle wedge without the involvement of an oceanic island basalts component, and the slab can carry water and supply a subduction component as a fluid to the overlying mantle wedge to depths exceeding 150 km.

Fate of the subducted Farallon plate inferred from eclogite xenoliths in the Colorado Plateau

We present the first finding of the high-pressure mineral coesite in lawsonite-bearing eclogite xenoliths from the Colorado Plateau, United States. The presence of coesite in these xenoliths supports the hypothesis that the eclogite formed in a low-temperature–high-pressure environment such as envisaged inside subducted oceanic lithosphere. Ion-microprobe U-Pb dating of micrometer-scale zircons in the eclogites yields ages ranging from 81 Ma to 33 Ma, the two extremes in age likely indicating the age of crystallization during subduction-related metamorphism and the age of recrystallization by the host magmatic event, respectively. These observations conclusively demonstrate that certain eclogite xenoliths from the Colorado Plateau originated as fragments of the subducted Farallon plate, which had been residing in the upper mantle since the Late Cretaceous. This is the first conclusive evidence that any eclogite xenoliths can be directly linked to a known subducted plate.

High-yield lithium separation and the precise isotopic analysis for natural rock and aqueous samples

A high-yield lithium separation technique for rock and aqueous samples has been established together with precise Li isotope analysis by thermal ionization mass spectrometry. Four separate stages of ion-exchange chromatography were carried out using organic ion-exchange resin. An ethanol-HCl solution was used for complete separation of Li from Na at the third column state. Total reagent volume for the entire chemical process was reduced to 42 ml and 33.3 ml for rock samples and seawater, respectively. The recovery yield and total procedural blank are 99.2–99.3% and 11 pg, respectively. Li3PO4 was used as an ion-source material in the mass spectrometric analysis. The in-run precision and reproducibility of measured 7Li/6Li ratios were ±0.04–0.07‰ (2σmean) and 0.37‰ (relative standard deviation; RSD) for rock and ±0.05-0.08‰ (2σmean) and 0.35‰ (RSD) for seawater. In this method, Rb, Sr, Sm, Nd, La and Ce can be collected after Li elution in the first column chromatography, then separated by the following specific procedures for these elements. Therefore, this method makes possible multi-isotope analysis for Li-poor and restricted small amounts of samples such as meteorites and mantle materials, extending to Li isotope geochemistry and cosmochemistry.

Precise boron isotopic analysis of natural rock samples using a boron-mannitol complex

Newly developed techniques for boron chemical separation and isotopic analysis in natural silicate rocks and waters are described. Sample dissolution and the subsequent ion-exchange chromatography were conducted using hydrofluoric and hydrochloric acids in the presence of mannitol which suppresses boron volatilization and isotopic fractionation by the formation of a boron-mannitol complex. Thermal ionization mass spectrometry using the Cs2BO2+-graphite method was employed for the determination of boron isotopic composition. No boron isotopic fractionation was observed in the course of chemical separation and mass spectrometry. In the whole analytical procedure, procedural blank and recovery yield of boron were 3–4 ng and 99 ± 1%, respectively. The analytical precision and reproducibility of measured 11B/10B ratios were ±0.1−0.1% and ±0.2‰ for the measurements of basalt and seawater, respectively. The present method enables us to determine the isotopic composition of < 1 μg B in silicate samples and in natural fluids with the above-mentioned analytical errors. This method also provides a remarkable improvement in the measurement of boron concentration by isotope dilution mass spectrometry because of the achievement of complete mixing between sample and spike during sample decomposition.

Boron isotope geochemistry of metasedimentary rocks and tourmalines in a subduction zone metamorphic suite

Abstract In order to understand the behavior of boron (B) and its isotope fractionation during subduction zone metamorphism, B contents and isotopic compositions together with major element compositions were determined for metasedimentary rocks and tourmalines from the Sambagawa Metamorphic Belt, central Shikoku, Japan. No systematic changes in whole-rock B content and isotope composition of the metasediments were observed among the different metamorphic grades, indicating the lack of a bulk fluid-rock B isotope fractionation as a result of devolatilization. Both modal abundance and grain size of tourmaline increase with increasing metamorphic grade. In contrast, B contents in muscovite and chlorite decrease with increasing metamorphic grade. These observations combined with mass balance calculations of B suggest the formation of tourmaline during progressive metamorphism from metamorphic fluids containing B mainly derived from muscovite and subordinately from chlorite without allowing significant net removal of B from the metasedimentary rocks. Tourmalines in the higher-grade metasedimentary rocks have zonal structure of B isotope and major element composition with decreasing δ 11 B and increasing Mg/(Mg+Fe) from the inner rim (core) to the outer rim. The change of Mg/(Mg+Fe) in the tourmalines with increasing grade is paralleled by similar variation in chlorite. These observations suggest that the growing tourmalines record the progressive evolution of the B isotopic composition of the metamorphic fluid, in the outermost rims preserving the isotope signature of peak metamorphic P–T-fluid conditions. Based on the above observations, the δ 11 B of the tourmaline is thought to have been nearly identical to that of the metamorphic fluid resulting in the “apparent” B isotopic fractionation factor between metamorphic fluid and whole-rock ( α=( 11 B / 10 B ) fluid /( 11 B / 10 B ) whole - rock ) which decreases from 1.007±0.003 to 1.001±0.003 from chlorite to biotite zone metamorphism. Such results together with the formation of tourmaline from (and sequestering of) B in metamorphic fluids may lead to less B isotopic fractionation as a result of subduction zone devolatilization than noted in suites containing less tourmaline. This, therefore, makes it possible to transport B isotopic signatures, which ultimately reflect Earth’s surface materials, to the deep mantle, perhaps resulting in mantle B isotope anomalies near convergent margins.

Trace element diffusion in jadeite and diopside melts at high pressures and its geochemical implication

Abstract Diffusivities of geochemically important trace elements (eleven rare earth elements (REE), Rb, Sr, Ba, and Y) in jadeite and diopside melts and those of Zr, Nb, Th, and U in jadeite melt have been measured at pressures between 7.5 and 20 kbar and at temperatures 50–200° above the liquidus, using diffusion couples and ion microprobe analysis. The concentrations of these elements in the experimental charges are close to those in natural igneous rocks. In the jadeite melt which is nearly fully polymerized (NBO/T ∼ 0), (1) diffusivities of REE increase with increasing ionic radii, (2) diffusivities of tri- and tetravalent elements increase with increasing pressure at constant temperature, whereas those of mono- and divalent elements do not change significantly with pressure, and (3) diffusivities of these elements decrease with increasing their ionic charge at constant pressure and temperature. In the diopside melt which is considerably less polymerized (NBO/T ∼ 2), (1) diffusivities of these trace elements depend mainly on their ionic radii rather than ionic charges; the diffusivities of mono- and divalent ions decrease with increasing ionic radii at constant pressure and temperature, and (2) diffusivities of REE are nearly the same as that of Ca and decrease with increasing pressure at constant temperature. The behavior of these trace elements is correlated with that of major elements; in the jadeite melt, tri- and tetravalent elements behave similarly to network-forming cations Al and Si, whereas mono- and divalent elements behave as network-modifying cations similarly to Na. In the diopside melt, the diffusion behavior of all these trace elements is similar to that of network-modifying cations Ca and Mg. The results of the present experiments suggest that the abundance of some trace elements in igneous rocks may have been affected by diffusion process at the magmatic stage. In the case of REE, for example, if two different magmas with high and low REE concentrations become in contact one another by multiple intrusion, and a zoned magma chamber is formed, diffusion begins to take place between them, and near the interface, the REE-enriched magma will become more light REE-depleted, whereas REE-depleted magma will become more light REE-enriched, and in addition, if magmas are reduced, the former will show a negative Eu anomaly, whereas the latter will show a positive Eu anomaly.