Chemical and Physical Properties and Thermal State of the Core - Physics and Chemistry of the Deep Earth - page 258

these conditions. Kamada et al . (2010) showed
that hcp-Fe is stable up to 220 GPa and 3000 K,
and that it can contain S up to 7.5 at % coexist-
ing with the Fe 3 S phase at 126 GPa and 2370 K.
Tateno et al . (2010) reported the stability of hcp-
Fe under the conditions of the inner core of 330
GPs and 4000 K, although their data contained
iron carbide, Fe 3 C, as a reaction product of the Fe
sample and the diamond anvil surface. Sakai et al .
(2011b) confirmed that hcp-phase is stable in pure
Fe up to 273 GPa and 4490 K and no other phases
of iron alloys were observed under these condi-
tions. Phase relations of Fe at high pressure and
temperature are shown in Figure 8.20. Asanuma
et al . (2008) showed that the hcp phase is stable
at least up to 3600 K at 242 GPa and up to 2400 K
at 257 GPa in Fe 3.4 wt% Si (Fe 0.93 Si 0.07 ) alloy. All
these experimental results suggest that the inner
core is likely composed of iron alloy with an hcp
structure. The phase relations of the Fe-Si alloy
(Kuwayama et al ., 2009) are shown in Figure 8.21.
Assuming that the light element is only Si, its

10.4
10.2

hcp-Fe
Nguyen &
Holmes (2004)

10.0
9.8
9.6
9.4
9.2
9.0

fcc (or bcc)-Fe
Brown & McQueen (1986)

Liquid iron

150 200 250 300 350

Pressure, GPa

Fig. 8.19 Hugoniots of iron under the core conditions
by Brown and McQueen (1986) and Nguyen and
Holmes (2004). Brown and McQueen (1986) suggested
existence of fcc phase below the melting temperature
at around 200 GPa, whereas Nguyen and Holmes
(2004) denied existence of the phase.

There are several studies suggesting that the
inner core possesses the hcp structure. They in-
clude shock experiments, laser-heated diamond
anvil cell experiments, and ab initio calculations.
Nguyen and Holmes (2004) performed shock ex-
periments on pure iron and reported the transition
into the liquid state at around 260 GPa. They
observed no evidence for a phase transition cor-
responding to the bcc or fcc-phase before melting
of iron, suggesting absence of bcc-iron and di-
rect transformation of hcp-iron into the liquid
state. The possible existence of hcp-iron in the
inner core has also been argued theoretically
based on ab initio calculations. Steinle-Neumann
et al . (2001), Vo cadlo (2007), Vo cadlo et al . (2008),
Tsuchiya and Fujibuchi (2009), and Sha and Co-
hen (2010a,b) showed the stability of hcp-iron
and calculated its elasticity under the inner core
conditions.
Intensive efforts have been made to clarify
the stable phase of iron and iron-nickel, and
iron-light-element alloys in the inner core ex-
perimentally using a laser heated diamond anvil
cell. Kuwayama et al . (2008) studied the phase
relations and stable phases in Fe and Fe-Ni alloys
up to 300 GPa and 2000 K, and reported stability of
the hcp-phase and absence of the bcc-phase under

5000

HP-bcc

4000

Melt

3000

2000

fcc

1000

hcp

bcc

0

0 50 100 150 200 250 300 350

Pressure, GPa

Fig. 8.20 Phase relations of iron at high pressure and
temperature. Experimental conditions made by
Kuwayama et al . (2008), gray squares; Tateno et al .
(2010), open squares, and Sakai et al . (2011b), solid
squares, are given in this figure. An open triangle; a bcc
phase in Fe 0.8 Ni 0.2 observed by Dubrovinsky et al .
(2006) at 220 GPa and 3500 K. A dashed curve: the
melting curve of hcp-Fe by Shen et al . (2004); a dotted
curve: the melting curve by Boehler (1993).