Earth’s Mantle Melting in the Presence of C–O–H–Bearing Fluid - Physics and Chemistry of the Deep Earth - page 38

volatile components, is involved in melting,
island-arc volcanism, and the chemical alteration
of the mantle wedge above subducting plates. In
turn, mantle wedge material becomes entrained
in mantle convection. Incompatible and volatile
components can recycle and escape from the
mantle in mid-ocean ridges and hotspots as well
as during continental magmatism. Indeed, these
materials are also involved in the formation of
mantle heterogeneities in the form of depleted
peridotites, pyroxenites and eclogites produced
by deep-level melt fractionation and reactions
with the peridotites (Hofmann & White, 1982;
Sobolev et al ., 2007). A potential source of
mantle heterogeneity is the Earth's core, but
the scale of core-mantle exchange remains
uncertain. According to theoretical modeling,
mantle heterogeneities can persist over geological
time, that is, for billions of years. They can be
preserved both in the rigid lithosphere and in the
convecting mantle. Evidences for lithospheric
heterogeneities come from ancient pyroxenite
xenoliths in a lherzolite matrix. Heterogeneities
in the convecting mantle can be detected
by subduction-related seismic anomalies and
different seismic reflectors (e.g., Stixrude, 2007).
Mantle heterogeneities are closely related
to the distribution of volatile components and
volatile cycles in the Earth's interior. Deter-
mining the role of volatiles, especially in the
C-O-H-S system, in deep mantle processes is
one of the key issues in modern mantle petrology.
One of the major tools for solving these issues is
the experimental studies at high pressures. The
abundance and behavior of hydrogen, carbon,
and their compounds in the mantle have been
widely discussed in a number of papers, which
are based on experimental studies of melting
phase relations in systems with volatiles at
pressures up to 6-10GPa and some limited
data for higher pressures (e.g., Poli & Schmidt,
2002; Dasgupta et al ., 2004; Kawamoto, 2004;
Dasgupta & Hirschmann, 2006; Dasgupta &
Hirschmann, 2010). Other important problems
related to the behavior of volatiles in the mantle
are hydrogen and carbon solubility in silicates,
which were studied at pressures up to 25-27GPa

(Bolfan-Casanova, 2005; Shcheka et al ., 2006;
Litasov & Ohtani, 2007; Hirschmann et al .,
2009), the influence of hydrogen and carbon
on the rheological and transport properties of
silicates (Mei & Kohlstedt, 2000a,b; Karato &
Jung, 2003; Hayden & Watson, 2008; Yoshino
et al ., 2010), and the formation of diamond in
fluid-bearing systems (Palyanov et al ., 2005;
Palyanov & Sokol, 2009).
In recent years we have published several
studies on volatile-bearing systems at pressures
of up to 20-30GPa (Litasov & Ohtani, 2002;
Litasov & Ohtani, 2005; Ghosh et al ., 2009;
Litasov & Ohtani, 2009a, 2010; Litasov, 2011).
In this chapter, we analyse these and other
experimental data on peridotite and eclogite
solidi and on melting in the presence of C-O-H
volatiles. We consider the systems with H 2 O,
CO 2 ,H 2 O

CO 2 , as well as those with a reduced
C-O-H fluid (presumably CH 4 +

+

H 2 O). On the
basis of this analysis, the importance of eclogites
in mantle melting in the presence of volatiles is
emphasized along with the key role of carbonates
in the deep-level melting of subducted plates.

2.2 High-Pressure Experimental Techiques
for Fluid-Bearing Systems

Simple systems with volatiles can be studied
in conventional high pressure experiments.
Descriptions of the methods can be found in
(Litasov & Ohtani, 2002; Litasov & Ohtani,
2009a,b; Shatskiy et al ., 2011). However, buffer-
ing techniques for controlling redox conditions
are required in many systems. Unbuffered
experiments, for example with H 2 OorCO 2 , yield
fO 2 conditions, which are determined either
by starting composition (i.e. mineral equilibria
with a fluid phase) or by the fO 2 imposed by
the cell assemblage (e.g., BN, LaCrO 3 , graphite).
Hydrogen may migrate in or out of the sample
capsule, which is particularly dangerous for
systems with CO 2 (Brooker et al ., 1998). In order
to protect sample from undesirable hydrogen
gain or loss different techniques and capsule
materials can be used. Among different metal