th
th
Joint 20 AIRAPT – 43 EHPRG, June 27 – July 1, Karlsruhe/Germany 2005
Solubility of silicon and oxygen in liquid iron coexisting with
(Mg,Fe)SiO3-perovskite and implications for core formation
T. Kawazoe* and E. Ohtani Tohoku University, Sendai, Japan,
Email: [email protected]
Summary
Solubilities of silicon and oxygen in liquid iron coexisting with Mg-perovskite were
investigated at 27 GPa and 2320-3040 K to discuss a core formation process and light
elements in earth’s core. The earth’s core contains light elements and these light elements
were dissolved into liquid iron to form the core during core formation process. A reaction
between the liquid iron and Mg-perovskite must have occurred at a base of a deep magma
ocean and could have provided silicon and oxygen as the light elements into the liquid iron.
In this study, high pressure and temperature experiments were conducted with a Kawai-type
multi-anvil apparatus. The liquid iron reacted with Mg-perovskie to form the magnesiowustite
and silicon and oxygen dissolved into the liquid iron at temperatures above 2640 K. The
solubility of silicon in the liquid iron decreases, whereas that of oxygen increases with
increasing oxygen fugacity, and both increase with increasing temperature. This implies a
possibility that silicon and oxygen are the light elements in the earth’s core.
Introduction
The light elements were dissolved into liquid iron to form the Earth’s core during the core
formation process. A terrestrial magma ocean is likely to have extended to the lower-mantle
depth in the core formation stage (Ohtani et al., 1997; Li and Agee, 2001). Liquid iron
separated in the magma ocean ponded at the base, and segregated to form the core. A
reaction between the liquid iron and (Mg,Fe)SiO3-perovskite (hereafter called Mg-perovskite),
the most dominant mineral in the lower mantle, may have occurred at the base of the deep
magma ocean and it may have provided silicon and oxygen for the light elements in the core.
The dissolution of silicon also makes it possible to explain a depletion of silicon in the
primitive mantle compared with CI chondrite (McDonough and Sun, 1995; Allegre et al.,
2001) without enrichment of silicon in the lower mantle. Another importance of this reaction is
that it may occur at the core-mantle boundary throughout the core formation stage. Results
of previous studies on this reaction using a laser-heated diamond anvil cell (LH-DAC) (Knittle
and Jeanloz, 1989, 1991; Hillgren and Boehler, 2000) were controversial perhaps because of
a large temperature gradient in the sample and possible difference in the fO2 conditions. In
this study, equilibrium experiments between liquid iron and Mg-perovskite have been
conducted with a Kawai-type multi-anvil apparatus at 27 GPa and up to 3040 K in order to
obtain more reliable results on this reaction and solubilities of silicon and oxygen in the liquid
iron.
Experimental Method
Experiments were performed with a Kawai-type multi-anvil apparatus at Tohoku University,
using WC anvils with 2.0-mm truncated edges and a semi-sintered (Mg,Co)O pressure
medium. Pure iron (99.99 wt% purity) rod was packed into a synthesized MgSiO3 or
(Mg0.9,Fe0.1)SiO3 capsule. The sample was heated at a high temperature by a Re cylindrical
heater and a LaCrO3 thermal insulator system. Generated temperature was calibrated with a
relation between an input power to the heater and a melting temperature of Mg-perovsktie
(Zerr and Boehler, 1993) and a eutectic temperature in the Fe-FeO system (Boehler, 1993).
Temperatures monitored outside the heater with a W 97Re3-W 75Re25 thermocouple were also
taken into account. Especially in the run 83 quenched at 3040 K and 27 GPa, Mg-perovskite
th
th
Joint 20 AIRAPT – 43 EHPRG, June 27 – July 1, Karlsruhe/Germany 2005
was molten in the outer part of the capsule contacting with the Re heater. Generated
pressure was calibrated with the pressure of the phase transformation boundary from
ringwoodite to Mg-perovskite and periclase in Mg2SiO4 (Fei et al., 2004) and that from pyrope
garnet to alminous Mg-perovskite and corundum in Mg3Al2Si3O12 (Hirose et al., 2001)
(corrected to an equation of state of MgO (Speziale et al., 2001) by Fei et al. (2004)) at 1873
K. Quenched liquid iron was analyzed with an electron probe micro-analyzer (EPMA) with the
wave length dispersive mode (JEOL JXA-8200, 8800). The chemical compositions of Mgperovskite and magnesiowüstite were analyzed with an EPMA with the energy dispersive
mode (JEOL JSM-5410). An oxygen fugacity (fO2) relative to the Iron-Wüstite buffer (ΔlogfO2
(IW)) was estimated by thermodynamic calculations (Hillgren et al., 1994) and partitioning of
FeO between Mg-perovskite and magnesiowüstite.
Results
The quench texture of liquid iron was similar to that of melting experiments in the Fe-FeO
system at 16 GPa (Ringwood and Hibberson, 1990), except for a precipitation of stishovite
and blobs composed of SiO2 and FeO. At lower oxygen contents in the liquid iron, quench
products were iron crystal, sub-micron wüstite and blobs composed of SiO2 and FeO. At
higher oxygen contents in the liquid iron, quench products are wüstite dendrite, sub-micron
wüstite and blobs composed of FeO and Fe. The quench textures of the liquid iron in runs 83
and 102 are shown in Figures 1 and 2, respectively. These quench products are interpreted
to be dissolved in liquid iron at high temperatures and pressures and formed by precipitation
from the liquid iron during rapid quenching (Ringwood and Hibberson, 1990, 1991). A
reaction between liquid iron and Mg-perovskite to form magnesiowüsite was clearly observed
in runs 83 and 86 quenched at 3040 K and 2640 K at 27 GPa, respectively. We could not
observe magnesiowüsite and confirm the reaction in the other runs, although dissolution of Si
into liquid iron was observed. Back-scattered electron images of run 83 quenched at 3040 K
at 27 GPa are shown in Figure 1.
Figure 1. Back-scattered electron images of run 83 quenched at 3040 K and 27 GPa. (a)
Mg-perovskite (dark gray) reacted with liquid iron (white) to form magnesiowüstite
(light gray) at a boundary between Mg-perovskite and the liquid iron. This reaction
provides silicon and oxygen as “ light elements” in liquid iron. A black crack between
the liquid iron and a reaction zone was formed during decompression. (b) Needle-like
stishovite (black) was grown over iron crystals (white) and many sub-micron wüstite
(gray area) were formed between the iron crystals. Silicon and oxygen contents in the
liquid iron were 1.70 wt% and 2.3 wt%, respectively.
th
th
Joint 20 AIRAPT – 43 EHPRG, June 27 – July 1, Karlsruhe/Germany 2005
Figure 2. Back-scattered electron image of run 102 quenched at 2830 K and 27 GPa.
Ionic blob (gray, spherule) and dendritic wüstite (gray) were observed homogeneously
in liquid iron. Black grains were Mg-perovskite. Silicon and oxygen contents in the
liquid iron were 0.18 and 7.52 wt%, respectively.
Discussion
In a core formation stage, liquid iron descended in the magma ocean, then ponded at its
bottom and subsequently descended through the solid mantle composed of Mg-perovskite
and magnesiowüstite to the center of the proto-Earth. As the depth of the magma ocean
increased, pressure and temperature at the bottom of magma ocean increased along the
mantle solidus (Zerr et al., 1998). According to recent studies on partitioning of siderophile
elements between liquid iron and magma or magnesiowüstite (e.g., Gessmann and Rubie,
2000; Li and Agee, 2001), the magma ocean could have extended to the upper part of the
lower mantle and ΔlogfO2 (IW) of the magma ocean was in the range between -2.5 and -2.1.
In the case that the magma ocean had extended to the pressure of 27 GPa in this study, the
temperature at the bottom of the magma ocean was between 2430 and 2840 K (Zerr et al.,
1998). The liquid iron ponded at the bottom could contain several wt% silicon and about 1
wt% oxygen in such conditions of oxygen fugacity and temperatures. Rubie et al. (2004)
suggested that oxygen solubility in liquid iron could increase with increasing pressure at the
bottom of the magma ocean beyond 27 GPa because a positive effect of temperature is
larger than a negative effect of pressure in this region. The liquid iron could contain several
wt% silicon when the depth of the magma ocean increased, taking into account of a positive
temperature effect found in this study and an estimated negative pressure effect (Gessmann
et al., 2001). Thus silicon and oxygen explaining the density deficit of the present core
(Anderson and Isaak, 2002; Hirao et al., 2004) could be transported by liquid iron into the
center of the Earth during the core formation stage.
Conclusion
Solubilities of silicon and oxygen in liquid iron coexisting with Mg-perovskite were
investigated at 27 GPa and 2320-3040 K to discuss the core formation process and light
elements in earth’s core. The liquid iron reacted with Mg-perovskie and silicon and oxygen
dissolved into the liquid iron at temperatures above 2640 K. Silicon and oxygen contents in
liquid iron were 1.70 wt% and 2.3 wt%, respectively, in the run made at 3040 K and 1.70 log
units below IW buffer. They were 0.18 wt% and 7.5 wt%, respectively, in the run made at
2800 K and -0.79 log units above IW buffer. The solubility of silicon in the liquid iron
decreases, whereas that of oxygen increases with increasing oxygen fugacity, and both
increase with increasing temperature. This implies a possibility that silicon and oxygen are
the light elements in the earth’s core.
th
th
Joint 20 AIRAPT – 43 EHPRG, June 27 – July 1, Karlsruhe/Germany 2005
References
Allegre C., Manhes G., Lewin E, 2001. Chemical composition of the Earth and the volatility
control on planetary genetics, Earth Planet Sci Lett (185), 49-69.
Anderson O.L., Isaak D.G.,2002. Another look at the core density deficit of Earth’s outer core,
Phys Earth Planet Inter (131), 19-27.
Boehler R., 1993. Temperatures in the Earth’s core from melting-point measurements of iron
at high static pressures, Nature (363), 534-536.
Fei Y., Van Orman J., Li J., van Westrenen W., Sanloup C., Minarik W., Hirose K.,
Komabayashi T., Walter M., Funakoshi K., 2004. Experimentally determined
postspinel transformation boundary in Mg2SiO4 using MgO as an internal pressure
standard and its geophysical implications, J Geophys Res (109), B02305.
Gessmann C.K., Rubie D.C., 2000. The origin of the depletions of V, Cr and Mn in the
mantles of the Earth and Moon, Earth Planet Sci Lett (184), 95-107.
Gessmann C.K., Wood B.J., Rubie D.C., Kilburn M.R., 2001. Solubility of silicon in liquid
metal at high pressure: implications for the composition of the Earth’s core, Earth
Planet Sci Lett (184), 367-376.
Hillgren V.J., Boehler R., 2000. High pressure geochemistry in the diamond cell to 100 GPa
and 3300 K, In: Manghnani MH, Nellis WJ, Nicol MF (eds) Proceedings of AIRAPT-17.
Universities Press, Hyderabad, India, pp 609-611.
Hillgren V.J., Drake M.J., Rubie D.C., 1994. High-pressure and high-temperature
experiments on core-mantle segregation in the accreting earth, Science (264), 14421445.
Hirao N., Ohtani E., Kondo T., Kikegawa T., 2004. Equation of state of iron-silicon alloys to
megabar pressure, Phys Chem Minerals (31), 329-336.
Hirose K., Fei Y., Ono S., Yagi T., Funakoshi K., 2001. In situ measurements of the phase
transition boundary in Mg3Al2Si3O12: implications for the nature of the seismic
discontinuities in the Earth’s mantle, Earth Planet Sci Lett (184), 567-573.
Knittle E., Jeanloz R., 1989. Simulating the core-mantle boundary: An experimental study of
high-pressure reactions between silicates and liquid iron, Geophys Res Lett (16),
609-612.
Knittle E., Jeanloz R., 1991. Earth’s core-mantle boundary: Results of experiments at high
pressures and temperatures, Science (251), 1438-1443.
Li J., Agee C.B., 2001. The effect of pressure, temperature, oxygen fugacity and composition
on partitioning of nickel and cobalt between liquid Fe-Ni-S alloy and liquid silicates:
Implications for the Earth’s core formation, Geochim Cosmochim Acta (65), 18211832.
McDonough W.F., Sun S.S., 1995. The composition of the Earth, Chem Geol (120), 223-253.
Ohtani E., Yurimoto H., Seto S., 1997. Element partitioning between metallic liquid, silicate
iquid, and lower mantle minerals: Implication for core formation of the Earth, Phys
Earth Planet Inter (100), 97-114.
Ringwood A.E., Hibberson W., 1990. The System Fe-FeO Revisited, Phys Chem Minerals
(17), 313-319.
Ringwood A.E., Hibberson W., 1991. Solubilities of mantle oxides in molten iron at high
pressures and temperatures: implications for the composition and formation of Earth’s
core, Earth Planet Sci Lett (102), 235-251.
Rubie D.C., Gessmann C.K., Frost D.J., 2004. Partitioning of oxygen during core formation
on the Earth and Mars, Nature (429), 58-61.
Speziale S., Zha C., Duffy T.S., Hemley R.J., Mao H.K., 2001. Quasi-hydrostatic
compression of magnesium oxide to 52 GPa: Implications for the pressure-volumetemperature equation of state, J Geophys Res (106), 515-528.
Zerr A., Boehler R., 1993. Melting of (Mg,Fe)SiO3-perovskite to 625 kilobars: Indication of a
high melting temperature in the lower mantle, Science (262), 553-555.
Zerr A., Diegeler A., Boehler R., 1998. Solidus of earth’s deep mantle, Science (281), 243246.