Geoscience Reference
In-Depth Information
the oceanic crust, and the upper peridotite layer,
when melting in the mantle wedge occurs along
almost any subduction PT-profiles except the
hottest ones (Figures 2.3 and 2.7). The existence of
carbonatite melt in the deep mantle is confirmed
by studies of super-deep diamonds and their inclu-
sions. Some carbonate inclusions were found in
close association with transition zone and lower
mantle minerals (Stachel
et al
., 2000; Brenker
et al
., 2007; Kaminsky
et al
., 2009) Moreover,
indirect comparison of Ca-perovskite inclusions
with experimentally synthesized phases also re-
veals their likely equilibrium with carbonatite
melt in the lower mantle (Walter
et al
., 2008).
These findings suggest that subducted plates re-
main oxidized to the level of the lower mantle
and their buffering capacity is not exhausted by
the surrounding reduced mantle. This may be
related to the limited buffering capacity of the
mantle (i.e. low contents of metallic Fe) or the
slow rate of solid-state redox reactions (as in-
ferred from the low diffusion rates of oxygen in
silicates) (Dohmen
et al
., 2002; Dobson
et al
.,
2008). Comparison of solidi for Na- and K- bearing
carbonatite with subduction PT-profiles indicates
that if the oxygen fugacity of the subducting slab
remains sufficiently high to stabilize carbonate,
melting of alkali carbonate may occur at a range
of depths depending on the PT-profile of subduc-
tion. In some cases, they may survive to lower
mantle depths, but in most cases a likely region
for melting of subducted alkali-bearing carbon-
ated eclogite is the transition zone (Figures 2.7,
2.8) as the solidi of carbonate-bearing systems
become flatter above about 10GPa (Figure 2.7).
Staudigel
et al
., 1989). According to recent esti-
mations by Johnston
et al
. (2011) decarbonation
efficiency at the island arc may be 20-80%,
whichmeans that in the coldest subduction zones
most carbonates can be transported down beyond
150 km. At the conditions of the transition zone
these concentration of CO
2
will lead to the forma-
tion of carbonatite melt with a carbonate content
of about 83 wt % (Shatskiy
et al
., 2013) and a
volume fraction (
) of up to 0.1, which is two
times lower than the equilibrium (maximum)
value (Laporte & Watson, 1995; Hammouda &
Laporte, 2000), but 5-10 times higher than the
limit of carbonatite melt interconnectivity in
silicate mantle (Hunter & McKenzie, 1989; Mi-
narik &Watson, 1995). The density of carbonatite
melt (
ϕ
ρ
Melt
) at the transition zone conditions is
3.0 g/ cm
3
based on the partial molar volume
of CO
2
and the thermal expansion of carbonate
melt (Liu & Lange, 2003; Ghosh
et al
., 2007;
Sakamaki
et al
., 2011). This density is signifi-
cantly lower than for mantle rocks at transition
zone conditions (3.6-3.8 g/cm
3
). Consequently,
buoyancy-driven porous flow should result in
carbonatite melt segregation at the slab-mantle
interface (Figure 2.14a).
The compaction-driven flow of a low-viscosity
melt through a creeping matrix has two limiting
regimes: (i) a hydraulically limited regime, where
the rate of fluid flow is controlled by the ma-
trix permeability, and (ii) a rheologically limited
regime, where the rate of melt flow is controlled
by the matrix rheology (Connolly
et al
., 2009). In
the first regime the melt ascent velocity (
v
) can
be expressed as:
k
g
ϕ
·
η
Melt
·
ρ
·
v
Porous
=
,
(2.1)
2.9.2 The mechanism of melt segregation and
movement
where
is the density contrast between melt
and residual silicate matrix (
ρ
600 kg/m
3
),
g
is
∼
the gravity (9.8 m/s
2
),
is the viscosity of melt
(0.015-0.005 Pa s) (Treiman & Schedl, 1983;
Genge
et al
., 1995, Dobson
et al
., 1996), and
k
is
the permeability, which depends on melt fraction
(
η
Since the principle of segregation and the move-
ment of hydrous and carbonatite melts are gen-
erally similar, we will consider the details of the
proposed model using a carbonatite melt with
some simple approximations. The average CO
2
content of the upper 500m of the oceanic crust
is estimated to be 3 wt % (
ϕ
), grain size (
d
), and the geometrical factor (
C
):
ϕ
n
d
2
C
k
=
.
(2.2)
∼
7 wt % carbonate;
