Geoscience Reference
In-Depth Information
study the lower mantle mineralogy. However, all
those measurements were in fact carried out well
outside the thermodynamic stability
P-T
field of
MgSiO
3
perovskite. There were uncertainties in
the shear properties determined by these studies
due to the possible effects of meta-stability of
the MgSiO
3
perovskite sample. To address such a
problem, Murakami
et al
. (2007a) first conducted
the high-pressure Brillouin scattering measure-
ments of polycrystalline MgSiO
3
perovskite up to
pressures of 96 GPa, which extended the pressure
condition by a factor of
90 GPa
−
10
−
5
0
5
10
10 over the previous
works. In this study, perovskite samples were
synthesized in situ in a DAC from an MgSiO
3
gel under thermodynamically stable pressure
condition of pv by infrared laser heating. Repre-
sentative Brillouin spectra at high-pressure are
shown in Figure 6.4a, exhibiting very sharp peaks
from the shear acoustic mode (
V
S
)ofMgSiO
3
perovskite. Sharp and well-resolved Raman bands
from MgSiO
3
perovskite at high pressure as
shown in Figure 6.4b strongly suggest that the
laser annealing technique in a DAC efficiently
promotes the synthesis of well-crystalized
polycrystalline MgSiO
3
perovskite and reduces
deviatoric stresses on the sample, thus substan-
tially improving the quality of the Brillouin
spectra at high pressure. On the other hand, the
peaks from longitudinal acoustic modes (
V
P
)of
MgSiO
3
perovskite were masked by the shear
acoustic modes of diamond and consequently,
V
P
of perovskite cannot be determined by this
technique.
Figure 6.5 shows the aggregate shear wave
velocity profile of MgSiO
3
perovskite as a
function of pressure by (Murakami
et al
.,
2007a). Extrapolation of the high-pressure shear
velocities to ambient pressure is in fact highly
consistent with the previous results determined
at lower pressure within the experimental errors
(Table 6.1). However, the pressure derivative of
the shear modulus (
G
0
)is1
.
56
∼
Velocity (km/s)
(a)
Pv
90 GPa
Pv
Pv
Pv
Pv
Pv
Pv
Pv
300
400
500 600
Raman shift (cm
−
1
)
700
800
(b)
Fig. 6.4
Brillouin (a) and Raman (b) spectrum of
MgSiO
3
perovskite (Pv) at 90 GPa after Murakami
et al
. (2007a). N, NaCl pressure medium; Dia,
diamond; (S), shear acoustic mode; (P), longitudinal
mode. Reproduced with permission of Elsevier.
derived from the different
G
0
values are shown
in Figure 6.5 showing a large difference from the
extrapolated values from the previous studies
(up to 8% differences in
V
S
at lowermost mantle
pressure). These discrepancies are likely due to
the limited pressure range of the experiments and
such a comparison demonstrates the importance
of measurements of velocities at pressure and
temperature relevant to the deep mantle of the
Earth. In contrast, the results by (Murakami
et al
., 2007a) appear to be consistent with the
recent ab initio calculations both at static (0K)
±
0.04, which is
distinctively (15
28%) lower than that of pre-
vious lower-pressure experiments below 9 GPa
determined by the ultrasonic interferometry
(Sinelnikov
et al
., 1998; Li & Zhang, 2005).
For comparison, the shear wave velocity profile
−
