Eiichi Takahashi

Major and trace element geochemistry of eclogitic mineral inclusions from Chinese diamonds are reported in this study, for the first time. Bulk major element compositions of mantle eclogite, estimated from diamond inclusions, are very close to that of MORB. All the analyzed samples exhibit evident positive Eu anomalies. Estimated bulk trace element compositions of mantle eclogite are generally parallel to that of MORB, but with deviations like enrichment in LILE and depletion in HFSE. It is proposed that the formation of mantle eclogite could be closely related to recycling of ancient oceanic crust. Other processes like (1) metasomatism by incompatible trace element rich melts; or (2) remelting and interaction with mantle peridotite, may also be involved. Coexisting of olivine with eclogitic mineral inclusions in a same diamond host, and evident trace element variations in some mineral inclusions show that some diamonds were formed by disequilibrium growth.

This study presents a forward model to quantify the P‐wave velocity (V_P), S‐wave velocity (V_S), and electrical conductivity (σ) of the solid‐liquid mixtures for a given set of pressure, temperature, lithology, liquid phase (aqueous fluid or melt), liquid fraction, and geometrical parameters in relation to the aspect ratio and connectivity of the liquid phase. This is based on previous experimental and theoretical studies on seismic velocity and electrical conductivity of solid rocks and liquid phases. A total of 78 lithologies, an aqueous fluid with NaCl (∼0–10 wt.%), and mafic to felsic melt appropriate for the crust and the uppermost mantle conditions were described in terms of V_P, V_S, and σ, as per previous experimental measurements and molecular dynamics simulation. This forward model is provided as a Windows executable program, and generates synthetic V_P, V_S, and σ, referring to the seismic velocities and electrical conductivity observed in the northeast Japan arc. After generation of the synthetic V_P, V_S, and σ, the original lithology and liquid parameters (phase, fraction, aspect ratio, and connectivity) were searched by implementing the grid search algorithm to map the misfit over the broad parameter space. The mapping shows the presence of a global misfit minimum around the optimized solution and the possibility of resolving the lithology and the liquid phase parameters based on the observed V_P, V_S, and σ by using the forward model presented in this study.

The “wet” silicate solidus of mantle peridotite defines the initial melting temperature of Earth's mantle under water‐saturated conditions and the second critical endpoint (SCEP) marks the high P‐T end of the wet solidus. However, the location of the wet solidus has remained an outstanding issue for over 50 years and the position of the SCEP is hotly debated. Published wet solidi show a difference of 200–600°C at a given pressure while reported SCEPs range from 6 GPa. Using a large‐volume multianvil apparatus, we investigated the water‐saturated melting behavior of a fertile peridotite at 3–6 GPa, 950–1200°C, and obtained well‐preserved quenched materials. On the basis of textures and compositions of the quenched materials, we bracket the wet solidus to between 950°C and 1000°C at 3 GPa and the SCEP between 3 and 4 GPa. Combining our experimental results with seismologic and petrologic observations, we propose that the lithosphere‐asthenosphere boundary in subduction zones should be constrained by the wet solidus and emphasize the role of a deep hydrous partial‐melting zone (DHPMZ) on magma genesis within the mantle wedge. We suggest that the DHPMZ is a source of hydrous melts to the primary melting zone in the mantle wedge and that the position of the volcanic front and its magma production rate may largely be controlled by melting and melt segregation processes within the DHPMZ. Our experimental results also suggest that high‐magnesian magmas (e.g., boninite, picrite, and komatiite) could be formed at conditions representative of subduction zones.

To constrain the preeruptive physical and chemical conditions (pressure, temperature, H₂O content, and fO₂) of the most recent and largest eruption (Aso‐4) of Aso volcano, Japan, we compared phase assemblages and chemical compositions of erupted material with those of experimentally determined run charges of the silicic end‐member magma of Aso‐4 (the KJ5665 pumice and MY5774 welded lens from the Aso‐4 pyroclastic flow deposit; SiO₂ = 70.5 wt %). Experimental conditions were as follows: starting material = KJ5665 pumice, pressure = 100–700 MPa, temperature range = 810–950 °C, and injected H₂O content = 2.9–6.7 wt %. The starting material comprised mainly rhyolitic glass with plagioclase, orthopyroxene, magnetite, and ilmenite phenocrysts. Hornblende phenocrysts are not found on cross sections of the KJ5665 pumice. Core compositions of these phenocrysts are unimodal. Using the magnetite‐ilmenite thermometer, the preeruptive magmatic temperature and fO₂ of KJ5665 were estimated to be ~900 °C and Nickel‐Nickel oxide (NNO) +1.1 log units, respectively. Run products contain plagioclase (An_20–48), orthopyroxene (Mg# = 53–72), magnetite, ilmenite, biotite, and K‐feldspar. Hornblende was not observed. Preeruptive conditions of the silicic end‐member magma of the Aso‐4 eruption are estimated to be ~900 °C, ≤ 300 MPa, and ≤ 3 wt % H₂O.

The melting of peridotite plays a key role in the chemical differentiation of planetary bodies. Water-saturated (‘wet’) solidus of mantle peridotite defines initial melting temperature of Earth’s mantle under water-saturated conditions and the second critical endpoint (SCEP) marks the end of the wet solidus. However, the location of the wet solidus remains an outstanding issue for over 50 years and the position of the SCEP is hotly debated. Published wet solidus shows a difference of 200−600 °C at given pressures, meanwhile, reported SCEP ranges from < 4 to > 6 GPa. Using a large-volume multi-anvil apparatus, we investigated the water-saturated melting behavior of a fertile peridotite (at 3−6 GPa, 950−1200 °C) and obtained well-preserved quenched liquid. Based on the texture and chemistry of the quenched liquid, we successfully determined the wet solidus and the SCEP of peridotite (Fig 1 A). The quenched fluids exhibit fragile fibres at 950 ℃ and spherule–fibre mixtures at temperatures above 1000 ℃ (Fig 1B). At 3 GPa, the quenched hydrous melt appears as a felt-like mass or as dendritic crystallites and coexist with the quenched fluids (Fig. 1B). We interpreted the presence of spherule–fibre mixtures as an evidence for aqueous fluid above the solidus and fragile fibres as evidence for aqueous fluid below the solidus. Thus, the occurrence of quenched melt and spherule–fibre mixtures indicates that the wet solidus lies between 950 and 1000 ℃ at 3 GPa and that 3 GPa is lower than the critical pressure (Pc). The most important textural difference between the run products at 3 GPa and those at other pressures (4 and 6 GPa)o is the presence of aqueous fluids in the former (Fig. 1B) and the absence of which in the latter (Figs. 1C–D).The spherule–fibre mixtures were not found in the 4 and 6 GPa run products. Liquids quenched from 4 and 6 GPa run products are homogeneous (SCF supercritical fluid), suggesting that Pc is lower than 4 GPa. Compositions of the liquids were analysed by EDS. In combine with previous studies, we find that with increasing pressures, the liquid compositions become more deficient in quartz and richer in olivine components. The compositions of silicate melts or SCFs change consistently with respect to pressure: andesitic at 1 GPa, boninite-like at 3 GPa, picritic at 4 GPa, and kimberlite-like at pressures > 5 GPa.

Whether arc mantle is more oxidized than oceanic mantle is persistently debated. The behavior of multivalent vanadium (V) is oxygen fugacity (fO₂) sensitive, and the ratios of V to a homovalent element (e.g., Sc, Ti, or Yb) in basalts were commonly used as fO₂ proxies. Similar ratios, such as V/Sc, between arc basalts and mid‐ocean ridge basalts were previously taken as evidence for similar fO₂s in their mantle sources. However, this claim may be problematic because elemental ratios are primarily controlled by partition coefficients (D values), which are further affected by various factors. Here we determined D values of V and other transition elements between mantle minerals and basaltic melts at typical arc T–P‐H₂O conditions and variable fO₂s. Combining experimental results with published data, the effects of fO₂, T, P, and phase compositions on D_V, D_Sc, and D_Ti for olivine, orthopyroxene (opx), clinopyroxene (cpx), and spinel were evaluated using multiple linear regressions. The results show that D_V values for these four minerals all increase with decreasing fO₂ and temperature, leading to higher D_V/D_Sc and D_V/D_Ti ratios at low temperatures than those at high temperatures given a certain fO₂. Thus, similar V/Sc and V/Ti ratios between arc basalts and mid‐ocean ridge basalts reflect a relatively oxidized arc mantle due to its lower melting temperatures. In light of the highly incompatible behavior of Ti during mantle melting, V‐Ti systematics are regarded to be more superior than V‐Sc systematics in the fO₂ estimation. Partial melting modelling results using V‐Ti systematics reveal that arc mantle is, on average, ~0.9 log units higher in fO₂ than oceanic mantle.

We present halogen, noble gas, and major and trace element compositions of mantle xenoliths from intraplate settings (Eifel, Kilbourne Hole, San Carlos, and Hawaii). The xenoliths show a wide range of halogen elemental ratios, which form two arrays centered on the halogen composition of mid‐ocean ridge basalts. The samples on the array toward high I/Cl value have relatively low Cl concentration and low ratios of highly incompatible elements relative to heavy rare earth elements, whereas the samples on the array toward low Br/Cl value have higher Cl concentration and trace elements ratios. The detailed mechanisms to account for these signatures are equivocal at present. However, they are most likely to be related to secondary processes of volatile loss during partial melting and secondary phase formation during interaction with melts. The common primary mid‐ocean ridge basalt‐like halogen ratios in mantle xenoliths from different parts of the globe indicate that the mantle itself must have a relatively uniform composition over a wide scale. The mantle has maintained its halogen composition over billion year timescales without being affected by I‐rich halogens being transported into the mantle. Mass balance calculations suggest that, in order to maintain the I/Cl ratio of the convecting mantle over 2 Gyr, the I/Cl ratio of the subducted halogens must be no more than several times higher than the present‐day mantle value.

The evolution of the magma plumbing system of Miyakejima volcano, an island arc basaltic volcano located on the volcanic front of the Izu arc, Japan, was investigated by combining (i) melting experiments of hydrous basaltic magma, (ii) analysis of melt inclusions in the volcanic rocks, and (iii) thermodynamic modeling using the thermodynamic program (MELTS) to understand the temporal evolution of the magma plumbing system. Basalts of the Ofunato Stage (10,000–4,000 year BP) show limited compositional variation without evidence of magma mixing, and seem to be parental to volcanic rocks of later stages. Hydrous melting experiments were conducted using the Ofunato Scoria (OFS), which represents the most undifferentiated basaltic magma during the past 10,000 years. The magma chamber conditions required to crystallize the phenocryst assemblage of the OFS were constrained to ~150 MPa, ~1100 °C, and ~3 wt% H₂O in melt. The constrained H₂O content is consistent with H₂O content of olivine‐hosted melt inclusions in the OFS. Here we propose that a simple magma plumbing system existed beneath Miyakejima volcano during the Ofunato Stage, comprising a single magma chamber. The MELTS calculations show that andesites of the Tsubota Stage (4,000–2,500 year BP; representing the later part of the Ofunato Stage after a period of dormancy of 3,000 years), can be derived from the Ofunato basalts, provided that H₂O content in magma is <1 wt%. The results of modeling indicate that a new magma chamber formed at a shallower depth after the Ofunato Stage, in which the differentiation of degassed basaltic magmas occurred.