Geoscience Reference
In-Depth Information
high-spin state of iron. The
G
0
of w ustite (Fe
0.94
O)
is smaller (
in the pressure range where spin transition occurs
(Figure 6.9). Although the pressure dependence
of shear wave velocity under pressure regime at
low-spin state of iron is still poorly constrained,
these studies suggested that after the completion
of spin transition, the shear velocity profile of
ferropericlase will not be so different from the
one extrapolated from the low-pressure high-spin
state (Jackson
et al
., 2006). In such a case, the
iron in ferropericlase may not play a crucial role
in the pressure dependence of shear wave veloc-
ity or modulus except in the pressure range where
the spin-paring transition occurs. Since the shear
velocity profiles presented by Crowhurst
et al
.
(2008) and Marquardt
et al
. (2009) at high-spin
state of iron are basically consistent with the pre-
vious low pressure works (Jackson
et al
., 2006),
and because there is no data on the shear velocity
trend with pressure at low-spin state of iron, I
use the
G
0
values by Kung
et al
. (2002) and Jack-
son
et al
. (2006) obtained for high spin state to
evaluate the effect of iron on
G
0
including both
high and low spin states. The dependence of
G
0
of
(Mg,Fe)O on iron content is shown in Figure 6.10.
This figure suggests that there is no clear correla-
tion between iron concentration and
G
0
at least
for low iron concentration of ferroperoclase under
0.71) than that of MgO according
to the previous lower pressure experiments to
∼
∼
3 GPa (Jackson
et al
., 1990). However, many
properties of w ustite (FeO) so highly anomalous
(e.g., McCammon
et al
., 1985; Cohen
et al
., 1997;
Badro
et al
., 1999; Fang
et al
., 1999; Kantor
et al
.,
2004; Yagi
et al
., 1985; Fei & Mao, 1994; Mazin
et al
., 1998; Mao, 1996; Murakami, 2004; Kondo,
2004; Ozawa, 2010) that I will not include the
properties of FeO in considering the dependence
of elastic properties of (Mg,Fe)O (McCammon,
1985; Cohen, 1997; Badro, 1999; Fang, 1999; Kan-
tor, 2004; Yagi, 1985; Fei & Mao, 1994; Mazin,
1998; Mao
et al
., 1996; Murakami
et al
., 2004;
Kondo
et al
., 2004; Ozawa
et al
., 2010).
6.4
Applications to the Lower Mantle
Mineralogy
Due to the recent direct experimental studies on
the elastic properties of two dominant phases of
the lower mantle under the lower mantle con-
ditions as summarized above, it is now possible
to discuss the composition of the lower man-
tle with stronger constraints than before. In this
section, I first outline the modeling procedures
that include model compositions, lower mantle
goetherms, formalism and elasticity data set to
apply, then I will discuss the plausible composi-
tion of the lower mantle.
3.5
Murakami
et al
. (2009)
Jackson
et al
. (2006)
Kung
et al
. (2002)
3.0
2.5
6.4.1 Model compositions
2.0
Based on the representative new shear wave
velocity data presented above, the shear wave
velocity profile appropriate for the representative
lower mantle geotherms in the three components
system MgO-SiO
2
-FeO were calculated. For the
purposes, a simplified lower mantle composition
was adopted: it consists of a two phase mixtures
of (Mg,Fe)SiO
3
perovskite (pv) and (Mg,Fe)O
ferropericlase (fp) with representative values of
X
Mg
(
1.5
Zha
et al
., 2000
Chopelas
et al
., 2000
Jackson & Niesler, 2000
Spetzler, 1970
Yoneda, 1990
Sinogeikin & Bass, 2000
1.0
0.5
0
8
(100*Fe/(Fe
4
12
16
20
+
Mg))
mol
Fig. 6.10
The pressure derivatives of the adiabatic
shear moduli at ambient condition of (Mg
1
−
x
,Fe
x
)O
plotted as a function of iron content after (Murakami
et al
., 2009). Reproduced with permission of Elsevier.
94 in pv
and 79 in fp, respectively (Jackson & Rigden,
1998). A constant Mg-Fe partitioning coefficient
=
100
∗
MgO
/
(MgO
+
FeO)
mole
)
=
