Geoscience Reference
In-Depth Information
suggestion is confirmed, it will provide a tight
constraint on the temperature-depth profile near
the CMB (Figure 8.1). However, recent experi-
ments on determination of the post-perovskite
phase boundary suggest absence of the double
crossing at the base of the lower mantle because
of a wide coexisting region of perovskite and post-
perovskite (Catalli
et al
., 2009). Further detailed
studies combined with establishment of the pres-
sure scale at high temperature are needed to reach
a definitive conclusion on the existence of the
double crossing at the base of the lower mantle.
such as ferropericlase or Mg-perovskite/post-
perovskite have been conducted by several
authors (Takafuji
et al
., 2005; Sakai
et al
., 2006;
Asahara
et al
., 2007; Ozawa
et al
., 2008, 2009).
These experiments have revealed that O and
Si can be dissolved in molten iron at high
pressure and temperature as shown in Figure 8.2,
suggesting both O and Si are candidates for the
light elements in the outer core. Ferropericlase
and perovskite/post-perovskite coexisting with
molten metallic iron contain very small amounts
of FeO, suggesting very low O fugacity during
these experiments. These experiments indicate
that if we assume O fugacity at CMB is similar
to that of the experiments, the composition of
the silicate portion of the CMB could be FeO
depleted in composition, as was suggested by
Asahara
et al
. (2007) and Ozawa
et al
. (2008).
Mao
et al
. (2006) studied the stability of
post-perovskite at high pressure and observed an
FeO-enriched post-perovskite phase at pressures
corresponding to the CMB. They proposed a
model of FeO enrichment at the CMB based on
their high-pressure experiments. Ferrous/ferric
iron enrichment at the base of the lower mantle,
as suggested by Mao
et al
. (2006), i.e., iron
enrichment in perovskite/post-perovskite and
ferropericlase,may be inconsistent with the parti-
tioning experiments (Asahara
et al
., 2007; Ozawa
et al
., 2008). Iron enrichment in perovskite or
post-perovskite requires relatively high O fugac-
ity at the base of the lower mantle and the top of
the core. In these studies, mechanisms for iron-
enrichment have not been discussed; therefore,
assessing the validity of these models is difficult.
A layered structure at the top of the outer core
has been suggested by some authors (Tanaka,
2007; Helffrich & Kaneshima, 2010). Helffrich
and Kaneshima (2010) proposed a model of ac-
cumulation of light elements such as O at the
top of the outer core. Dissolution of O and Si
into molten iron was inevitable because of high
temperature and high pressure conditions during
the core formation stage of the Earth (e.g., Sakai
et al
., 2006). High concentration of O and Si in
molten iron in the high-temperature stage during
core formation could produce a higher solubility
8.2.2 Existence of dense melts at the CMB
The existence of a dense melt at the base of the
lower mantle has been proposed by Ohtani (1983)
and has been argued repeatedly based on static
experiments (Ohtani & Maeda, 2001; Murakami
& Bass, 2011),
ab initio
calculations (e.g. Stixrude
& Karki, 2005) and shock compression experi-
ments (e.g., Mosenfelder
et al
., 2009). Tsuno
et al
.
(2007) found that oxide liquid and metallic liq-
uids inthe Fe-O and Fe-O-S systems become
mutually miscible at high pressures and tempera-
tures based on in situ X-ray radiographic imaging
experiments. The enhancement of the mutual
solubility of ionic and metallic liquids at high
pressures and temperatures suggests that the dis-
solution of silicate magmas into the molten outer
core of the metallic component into silicate mag-
mas can also occur in silicate magma systems
at the core-mantle boundary (Ohtani, 2009). The
magmas dissolving the metallic iron component
should be denser than the surrounding crystalline
mantle, suggesting the existence of a dense melt
as the origin of the ULVZ, at the base of the lower
mantle (Ohtani, 2009). If the dense magmas at the
core mantle boundary contain the metallic iron
component, they could coexist with the FeO de-
pleted lower mantle because of lower FeO activity
in the dense magmas.
8.2.3 FeO-enrichment at the CMB
Partitioning experiments of O and Si be-
tween molten iron and lower-mantle minerals
