The electrical conductivity of mantle rocks during phase transformation from ringwoodite to silicate perovskite and ferro-periclase was measured at 25 GPa and various temperatures ranging from 1300 ...to 1900 K. The electrical conductivity was high at the initial stage of annealing, suggesting that ferro-periclase forms interconnected layers in aggregates of silicate perovskite and ferro-periclase that are representative of lower mantle rock. At 1900 K the electrical conductivity quickly decreased and reached that of silicate perovskite, suggesting the cut-off of the interconnected ferro-periclase because of rounding of crystals. Below 1700 K, the high conductivity values were maintained for experimental duration. The interconnection of ferro-periclase, which has a lower viscosity than silicate perovskite, can be maintained in a cold descending slab over geological time scales (∼1 My), indicating that a colder slab is less viscous than the warmer mantle surrounding it. The low-viscosity slab can be prevented from penetrating into the deeper part of the lower mantle by the high viscosities encountered at a depth of ∼1000 km, referred to as the “viscosity hill”, that cause stagnation at this depth as observed by seismic tomography.
•An interconnectivity of ferro-periclase in post-spinel rock were examined.•The interconnection can be maintained in subducted slab for geological time scale.•The viscosity reduction of the slab due to the interconnection is expected.
The electrical conductivity of olivine and its high‐pressure polymorphs with various iron contents XFe = Fe/(Fe + Mg) = 0.1, 0.2, 0.3, 0.5, 0.7 and 1.0 was measured over a wide range of pressure (P) ...and temperature (T) conditions covering the stability field of olivine, wadsleyite and ringwoodite in a Kawai‐type multianvil apparatus. The pressure was determined using in situ X‐ray diffraction of MgO as a pressure marker in SPring 8. Molybdenum electrodes were used so that oxygen fugacity is similar to that for the iron‐wüstite buffer. The transition from low‐pressure phase to high‐pressure phase led to an increase of conductivity. In the stability field of each phase, the electrical conductivity slightly increased with increasing pressure at a constant temperature, suggesting a negative activation volume. The conductivity increased with increasing total iron content for each phase. All electrical conductivity data fit the formula for electrical conductivityσ = σ0 XFeexp{−ΔE0 − αXFe1/3 + P(ΔV0 − βXFe)/kT}, where σ0is the pre‐exponential term, ΔE0 and ΔV0 are the activation energy and the activation volume at very low total iron concentration, respectively, and k is the Boltzmann constant. The activation energy decreased with increasing total Fe content in olivine and ringwoodite. Dependence of the activation energy on the total Fe content suggests that the dominant mechanism of charge transport is Fe2+‐Fe3+hopping (small polaron). The activation volume for small polaron conduction in olivine and its high‐pressure polymorphs tends to decrease with total Fe content. For olivine with low Fe content, the activation volume for small polaron conduction still is negative and very small. Assuming constant Fe content (XFe = 0.1) and oxygen buffer condition, the conductivity will increase with depth mainly due to the increase of the temperature along the mantle adiabat.
Key Points
Effect of pressure on electrical conductivity of olivine
Low to high‐pressure phase transition led to an increase of conductivity
Negative activation volume was observed in hopping conduction
► We investigated electrical conductivity of partially molten carbonate peridotite. ► On the solidus, the conductivity was markedly higher than that of carbonate-free peridotite. ► Electrical ...conductivity is not markedly increased by higher melting degree. ► The moderate increase is attributed to a decrease in carbonate content in the partial melt. ► The conductivity in the upper mantle is enhanced by very small degree of melting.
In order to investigate the effect of carbonate-content in partial melt on bulk conductivity under high pressure, electrical conductivity measurements were performed on carbonate melt-bearing peridotites using a Kawai-type multi-anvil apparatus. The starting materials were composed of spinel lherzolite (KLB1) with small amounts of dolomite (1 and 3wt.%). To obtain various melt fractions, annealing experiments were performed at different temperatures above 1400K at 3GPa. At low temperatures (⩽1500K), the conductivity was distinctly higher than that of carbonate-free peridotite and close to that of the carbonatite melt-bearing olivine aggregates. Although the sample conductivity increased with increasing temperature, the rate at which the conductivity increases was small and the conductivity approached that of silicate melt-bearing peridotite. CO2 concentration in the partial melt decreased with increasing annealing temperature. Thus, the small increase of the conductivity with annealing temperature is attributed to a decrease of the melt conductivity due to a decrease in carbonate content in the partial melt. As the carbonate concentration in the melt decreases, the estimated melt conductivity approaches that of the basaltic melt. Therefore, conductivity enhancement by the carbonate-bearing melt is very effective at temperature just above that of the carbonate peridotite solidus.
The Fe-bearing wadsleyite-ringwoodite phase transition loop under dry conditions in a temperature range of 1473 and 1873 K was determined by in situ X-ray diffraction experiments at the synchrotron ...facility SPring-8. Pressure at high temperature was precisely determined within a 0.23 GPa error using in situ X-ray diffraction of MgO as a pressure standard. Under dry conditions, assuming an equilibrium chemical composition of wadsleyite and ringwoodite coexisting with garnet in a pyrolite model and an adiabatic temperature gradient with a potential temperature of 1550-1650 K, the phase transition depth and effective width of the seismic discontinuity were found to be 500-514 and 20-22 km, respectively. This effective width, which is three times greater than that of the olivine-wadsleyite phase boundary, can reflect a seismic wave of approximately 0.25 Hz. The wider transition loop between wadsleyite and ringwoodite could create a broad seismic discontinuity. Considering wet and oxidized conditions, the depth of the wadsleyite-ringwoodite phase boundary could be greater than 520 km assuming the small temperature dependency on water and oxygen fugacity effects. Variation in the depth of seismic anomaly may be attributed to water content or oxygen fugacity of the transition zone.
► Electrical conductivity of fluid-bering quartzite follows Arrhenius law. ► Electrical conductivity of fluid-bering quartzite increases with fluid contents. ► Fluid-bearing quartzite is unable to ...account for high-conductivity anomaly region. ► Large amount of other ionic species in fluid is required to increase conductivity.
The electrical conductivity of fluid-bearing quartzite was determined as function of temperature and fluid fraction at 1GPa in order to assess the origin of the high conductivity anomalies observed in the middle to lower crustal levels. Dihedral angles of quartz-fluid-quartz determined from recovered samples were below 60°, suggesting that fluid forms an interconnected network through the quartz aggregate. The electrical conductivity of quartzite increases with increasing temperature, which can be approximately expressed by Arrhenius equation. The apparent activation enthalpy decreases from 0.70 to 0.25eV with increasing fluid fraction in volume from 0.00043 to 0.32. The electrical conductivity (σ) of the fluid-bearing quartzite increased with fluid fraction (ϕ) proportionally to a power law (σ∝ϕ0.56–0.71) within the temperature range of 900–1000K. The electrical conductivity of the aqueous fluid-bearing quartzite with the maximum fluid fraction (0.32) was found to be about three orders of magnitude higher than that of dry quartzite at 1000K. However, its electrical conductivity was definitely lower than the geophysically observed values of high-conductivity anomalies, even if the quartzite contained large fluid fractions (0.32). The present results suggest that fluid-bearing quartzite is unable to account for the high-conductivity anomalies in terms of fluid fraction. A significant amount of other ionic species, such as Na, Cl, and Al in aqueous fluid, in addition to silica phases dissolved in fluid, is required to increase conductivity.
Redox input by subducted slab into mantle is important for deep cycle and isotopic evolution of volatile elements, whose stable forms are controlled by redox state. Given reduced condition in lower ...part of the upper mantle, taking redox budget from lithospheric mantle into consideration is crucial in redefining redox state there. To constrain to which extent subducted slab modified redox state of the uppermost mantle and how much oxygen budget slab carried into deep Earth, we investigated redox kinetics of olivine adopting diffusion couple method at 1 GPa and 1,373–1573 K in a piston cylinder apparatus. It is found that redox process in olivine is diffusion‐controlled, and diffusing on the order of 10−12 m2/s at 1473 K. Oxidation process in reduced part is oxygen fugacity (fO2)‐independent with activation enthalpy of 235 ± 56 kJ/mol, while reduction process in oxidized part is fO2‐dependent with an fO2 exponent of 2/5. Diffusion profile analysis reveals that for magnetite‐free couple, redox process is controlled by oxygen grain boundary diffusion (GBD) below ΔFMQ + 1, and rate‐limited by faster species which might be hydrogen related Mg vacancy above ΔFMQ + 1. However, for magnetite‐bearing couple, oxygen GBD dominates redox process across wide fO2 range. The extremely slow rate limits the homogenization of the slab with surrounding mantle so that redox state of the uppermost mantle remains unchanged in the past 3.5 Gyrs. A highly underestimated oxygen reservoir may have formed in deep Earth, when subducted slab transports oxidized components to region deeper than the mantle transition zone.
Plain Language SummaryAs oxidized slabs continue subducting into mantle, redox exchange proceeds between slabs and the surrounding mantle. Knowledge of redox kinetics of olivine is essential for understanding redox evolution of the uppermost mantle in the Earth's history and unraveling redox budget in slab residues that brought into deeper Earth. In this study, we conducted a series of diffusion experiments to determine rates of redox processes in olivine aggregates under high pressure and high temperature. Our results show that diffusion‐controlled redox processes in olivine aggregates are extremely slow. The extent of surrounding upper mantle which can be oxidized by slabs is very limited. The unchanging redox state of the uppermost mantle is not caused by its infinite redox capacity as previously supposed but due to its inability to digest oxidized components efficiently. Being not fully absorbed by the uppermost mantle, the oxidized slabs tend to transport a considerable amount of oxidized components into deeper mantle with further subducting.
Key PointsRedox kinetics of olivine were investigated by diffusion couple method at 1 GPa and 1,373–1,573 KBelow ΔFMQ + 1, redox process is controlled by O grain boundary diffusion, while above that, by H diffusion related with Mg vacancySlow redox rate limits homogenization of subducted slab with mantle. A highly underestimated oxygen reservoir may be present in deep Earth
We describe paper models which represent the cortical shell structures of radiolarian genus Pantanellium using the unit origami method: the construction of a polyhedral frame using origami units. ...Each unit corresponds to the edge of a polyhedron. The models are constructed based on planar graphs representing the cortical shell structures of real specimens. The resulting models reproduce the cortical shell structures appropriately. This means that the origami model is useful for both naked-eye practical observation of shell structures and group educational/art activities.
This paper discusses the geometrical properties of a radiolarian skeletal structure, namely, that of genus Didymocyrtis. We characterized the evolution of skeletal structures and analyzed the ...structures using geometry. We defined two ratios in order to quantify the geometrical properties of Didymocyrtis and verified that the two ratios changed with their phylogenic evolution. We also used the 3D skeletal data of a specimen of species D. tetrathalamus, which was obtained through micro X-ray CT. The cortical shell obtained in the 3D data was projected onto a spherical surface, and we determined the centers of the pores. Our analysis revealed that the number of pores is approximately 200 and their distribution is not regular. We also determined that the column-like parts of the skeleton, which connect the inner and upper parts of the specimen, do not lie on a plane and their intervals are not equal.
Al-rich phases (NAL: new hexagonal aluminous phase and CF: calcium–ferrite phase) are believed to constitute 10∼30 wt% of subducted mid-ocean ridge basalt (MORB) in the Earth's lower mantle. In order ...to understand the effects of iron on compressibility and elastic properties of the NAL phase, we have studied two single-crystal samples (Fe-free Na1.14Mg1.83Al4.74Si1.23O12 and Fe-bearing Na0.71Mg2.05Al4.62Si1.16Fe2+0.09Fe3+0.17O12) using synchrotron nuclear forward scattering (NFS) and X-ray diffraction (XRD) combined with diamond anvil cells up to 86 GPa at room temperature. A pressure-induced high-spin (HS) to low-spin (LS) transition of the octahedral Fe3+ in the Fe-bearing NAL is observed at approximately 30 GPa by NFS. Compared to the Fe-free NAL, the Fe-bearing NAL undergoes a volume reduction of 1.0% (∼1.2 Å3) at 33∼47 GPa as supported by XRD, which is associated with the spin transition of the octahedral Fe3+. The fits of Birch–Murnaghan equation of state (B–M EoS) to P–V data yield unit-cell volume at zero pressure V0=183.1(1) Å3 and isothermal bulk modulus KT0=233(6) GPa with a pressure derivative KT0′=3.7(2) for the Fe-free NAL; V0-HS=184.76(6) Å3 and KT0-HS=238(1) GPa with KT0-HS′=4 (fixed) for the Fe-bearing NAL. The bulk sound velocities (VΦ) of the Fe-free and Fe-bearing NAL phase are approximately 6% larger than those of Al, Fe-bearing bridgmanite and calcium silicate perovskite in the lower mantle, except for the spin transition region where a notable softening of VΦ with a maximum reduction of 9.4% occurs in the Fe-bearing NAL at 41 GPa. Considering the high volume proportion of the NAL phase in subducted MORB, the distinct elastic properties of the Fe-bearing NAL phase across the spin transition reported here may provide an alternative plausible explanation for the observed seismic heterogeneities of subducted slabs in the lower mantle at depths below 1200 km.
•A spin transition of the octahedral Fe3+ in the NAL phase occurs at 30∼47 GPa.•The spin transition is associated with a volume reduction of the NAL phase.•Significant softening in KT and VΦ are observed within the spin transition region.•The NAL phase may help to explain the seismic heterogeneities in the lower mantle.
PDF
