Physics of the Earth and Planetary Interiors 215 (2012) 12–20

Contents lists available at SciVerse ScienceDirect

Physics of the Earth and Planetary Interiors
journal homepage: www.elsevier.com/locate/pepi

A comparison of ice VII formed in the H2O, NaCl–H2O, and CH3OH–H2O
systems: Implications for H2O-rich planets
Mark R. Frank a,⇑, Elizabeth Aarestad a, Henry P. Scott b, Vitali B. Prakapenka c
a

Department of Geology and Environmental Geosciences, Northern Illinois University, DeKalb, IL 60115, United States
Department of Physics and Astronomy, Indiana University South Bend, South Bend, IN 46634, United States
c
Center for Advanced Radiation Sources, University of Chicago, Chicago, IL 60637, United States
b

a r t i c l e

i n f o

Article history:
Received 23 February 2012
Received in revised form 2 October 2012
Accepted 20 October 2012
Available online 8 November 2012
Edited by Kei Hirose
Keywords:
Ice VII
Impurities
High-pressure H2O
Ice planets

a b s t r a c t
High-pressure H2O polymorphs, namely ice VI, ice VII, and ice X, are hypothesized to make up a considerable portion of the interiors of large icy satellites and select extra-solar planets. The incorporation of
foreign ions or molecules into these high-pressure phases is possible through ocean–ice interaction,
rock–ice interaction at depth, or processes that occurred during accretion. Recent research concerning
the effects charged ions have on ice VII has shown that these ions notably affect the structure of ice
VII (Frank et al., 2006; Klotz et al., 2009). This study was designed to determine the effects of a molecular
impurity on ice VII and compare those effects to both pure H2O ice and ice with an ionic impurity. Ice
samples were formed in this study via compression in a diamond anvil cell from either H2O, a
1.60 mol% NaCl aqueous solution, a 1.60 mol% CH3OH aqueous solution, or a 5.00 mol% CH3OH aqueous
solution and were compressed up to 71 GPa at room temperature. Ice formed from pure H2O had no
impurities whereas ices formed in the NaCl–H2O and CH3OH–H2O systems contained the impurities
Na+ and Cl and CH3OH, respectively. Pressure–volume relations were observed in situ by using synchrotron based X-ray diffraction and were used to determine the equations of state for ices formed in the H2O,
NaCl–H2O and CH3OH–H2O systems. The data illustrate that ice VII formed from a NaCl-bearing aqueous
solution exhibited a depressed volume when compared to that of H2O-only ice VII at any given pressure,
whereas ice VII formed from CH3OH-bearing aqueous solutions showed an opposite trend, with an
increase in volume relative to that of pure ice VII. The ices within planetary bodies will most likely have
both ionic and molecular impurities and the trends outlined in this study can be used to improve density
proﬁles of H2O-rich planetary bodies.
Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction
As exploration reaches farther out into the solar system and beyond, the understanding of planetary bodies that have been
encountered has expanded as well. Before exploratory missions
were sent to the outer reaches of the solar system, very little was
known of the ice-rich bodies encountered there and many were assumed to be geologically inactive worlds (Consolmagno and Lewis,
1976; Reynolds and Cassen, 1979). The Voyager, Galileo, and Cassini-Huygens missions have greatly increased our understanding
of the icy bodies within the solar system and illustrated that
H2O-rich planetary bodies can be extremely active and complex
(Anderson et al., 1998; Gurnett et al., 1996; Khurana et al., 1998;
McCord et al., 2001; Schubert et al., 1996; Sohl et al., 2002). Additionally, recent discoveries of extra-solar planetary bodies, such as
Gliese 436b (or GJ 436b), an extra-solar planet hypothesized to
⇑ Corresponding author.
E-mail address: mfrank@niu.edu (M.R. Frank).
0031-9201/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.pepi.2012.10.010

have a signiﬁcant proportion of solidiﬁed H2O (Gillon et al.,
2007), illustrates the importance of high-pressure H2O-rich phases
in planetary physics beyond the solar system. High-pressure
experimental studies have provided much needed pressure–volume–temperature (PVT) equations of state (EoS) for solid phases
of H2O that may comprise the interiors of planets, such as GJ
436b (e.g., Hemley et al., 1987; Wolanin et al., 1997; Loubeyre
et al., 1999; Asahara et al., 2010; Fortes et al., 2011). It is likely,
however, that the high-pressure H2O ice present within these
bodies is not pure and would have trace to minor concentrations
of included impurities (Frank et al., 2006, 2008). These impurities
may have been incorporated into the ice structure through interactions with subsurface oceans allowing for the inclusion of either ionic or molecular impurities. Further, in a differentiated planetary
body the interaction between the high-pressure H2O phases and
the surrounding silicate or metallic materials at depth may also
provide a source of impurities through solid state diffusion at the
elevated temperatures of planetary interiors. Therefore, a systematic study of the effects of both ionic and molecular impurities

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20

on the structure and properties of high-pressure phases of H2O
would allow for additional constraints to be placed on the interiors
of H2O-rich planets and also provide much needed data on the
incorporation of trace elements into high-pressure phases.

1.1. Impurities in high-pressure ice
Most research on the high-pressure phases of H2O has previously been limited to the one-component H2O system. The onecomponent H2O system at high pressures is well known and has
been characterized through studies using various in situ techniques, such as neutron and X-ray diffraction, Brillouin spectroscopy and Raman spectroscopy. Numerous experimental studies
of this system have revealed that ice VI forms from liquid water
at room temperature (Choukrouna and Grasset, 2007) and with
increasing pressure from about 1 GPa to 2.2 GPa and has a hexagonal structure whereas ice VII is stable from 2.2 GPa to approximately 60 GPa and has a body-centered cubic structure with
partially ordered hydrogen atoms (e.g., Loubeyre et al., 1999). Ice
X is hypothesized to be stable at pressures greater than 60 GPa
with a structure similar to that of ice VII but ordered hydrogen
throughout the lattice centered between oxygen atoms (Aoki
et al., 1996; Goncharov et al., 1999; Song et al., 2003; Marx,
2006; Asahara et al., 2010). Many H2O-rich bodies are hypothesized to have these high-pressure polymorphs of H2O at depth,
possibly in the form of mixed phases (Anderson et al., 1996,
1997; Hubbard and Marley, 1989; Khurana et al., 1998; Kuskov
and Kronrod, 2005; McCord et al., 2001; Mueller and McKinnon,
1988; Nellis et al., 1988; Schubert et al., 1996; Sohl et al., 2002;
Spohn and Schubert, 2003; Zimmer et al., 2000). However,
there is a lack of data in geologically reasonable multi-component
systems, and this can inhibit proper modeling of planetary
interiors.
Only a few studies have explored the effects of impurities on ice
VII. Frank et al. (2006) studied ice formed from NaCl–H2O solutions
by using X-ray diffraction and Raman spectroscopy to determine
the effect that the incorporation of Na+ and Cl had on the volumetric properties and crystal chemical bonding of ice VII. It was
found that there was a systematic depression of the PV curve at
298 K relative to ice VII formed from the one-component H2O system. This, at ﬁrst glance, may seem counterintuitive as the added
ions should increase the volume of the ice simply by the addition
of mass. However, since NaCl dissociates into Na+ and Cl ions in
liquid H2O at room temperature and ambient pressure, it was
hypothesized that, upon increasing pressure, the ions would go
into the open, face-centered sites of the body-centered cubic structure of ice VII. Frank et al. (2006) suggested that the charged ions in
the face-centered sites attracted or repulsed the polar H2O molecule, causing the O–H bond to rotate and become ordered within
the lattice structure. This ordering of the hydrogen atoms was presumed to be similar to the ordering found in the ice X structure.
This study will examine the effect that a molecular or ionic
impurity has on the lattice of the ice VII structure in an attempt
to determine if the systematic depression of the unit cell volume
found in the Frank et al. (2006) study was due to the charged nature of the impurity or if it was due to the presence of an impurity
in the ice VII lattice, ionic or otherwise. This hypothesis was tested
in this study by the compression of aqueous solutions with a
molecular impurity in a diamond anvil cell to form ice VII. X-ray
diffraction data taken in situ were used to calculate the PVT relations of the sample and were directly compared to the PVT relations of the ice VII formed in the Frank et al. (2006) study to see
if there is a similar systematic variation of the unit cell volume
at the same temperatures and pressures.

13

1.2. The ice VII-ice X phase transition
The study of the effects that may occur with the presence of
impurities in high-pressure ice has mainly been limited to the
high-pressure phase ice VII. This phase has a very large temperature and pressure stability ﬁeld, indicating its likely relevance in
the interior of large icy bodies. At pressures beyond the stability
of ice VII, ice X is stable. Similar to the ice VII structure, ice X is a
BCC structure, but the hydrogen–oxygen bond distance is the same
between neighboring molecules with equal bonding between oxygen atoms (e.g., Aoki et al., 1996; Pruzan et al., 1997; Benoit et al.,
1998; Goncharov et al., 1999; Sugimura et al., 2008). The transition
from ice VII to ice X is likely complex and includes intermediate
steps as a result of a double well potential (Benoit et al., 1998;
Marx, 2006). It has been observed in previous studies examining
the ice VII-ice X phase transition that there is a decrease in the unit
cell volume of the structure (Loubeyre et al., 1999; Sugimura et al.,
2008; Asahara et al., 2010). As Frank et al. (2006) found a similar
decrease to the volume of the ice X structure due to the presence
of an ionic impurity, the effect that impurities would have on the
ice X structure, if any, merits further elucidation. This study will
also compare the ice VII-ice X phase transition in pure H2O with
two different two-component systems; 1.60 mol% NaCl–H2O and
1.60 mol% CH3OH–H2O to see if a transition can be observed and
if impurities, either ionic or molecular, affect the transition and/
or ice X structure.
2. Experimental and analytical methods
2.1. Solutions
Ice VII samples were formed from four solutions: (1) pure H2O,
(2) 1.60 mol% NaCl in H2O, (3) 1.60 mol% CH3OH in H2O and (4)
5.00 mol% CH3OH in H2O. Aqueous solutions were prepared to
examine various mole percentages of the solute, NaCl or CH3OH,
in the solvent, H2O. The 1.60 mol% NaCl-H2O aqueous solutions is
equivalent, in mole percent, to the 5.0 wt% NaCl aqueous solution
from Frank et al. (2006, 2008) and provides data on ionic impurities in ice VII and X. Frank et al. (2006) noted that ice VII forming
from a NaCl–H2O solution would incorporate Na+ and Cl because,
at low concentrations, the NaCl would be completely dissociated in
the liquid H2O and trapped as ionic impurities in the high pressure
ice. Aqueous solutions of CH3OH (1.60 and 5.00 mol%) were chosen
because CH3OH does not dissociate in water and therefore would
exist as molecular species during the formation of ice VII. These
solutions were chosen to provide a base PV curve (H2O) and to provide additional PV data on the effects of ionic (1.60 mol% NaCl) and
molecular impurities (1.60 and 5.00 mol% CH3OH aqueous solutions) on the properties of ice VII and ice X.
2.2. Symmetric diamond anvil cell
High-pressure phases of H2O with incorporated impurities were
produced by using a symmetric diamond anvil cell (DAC). The diamond anvils were brilliant cut, had 250 lm diameter culets, and
ranged from 2.26 to 2.45 mm in length from the table to the culet.
Rhenium gaskets were pre-indented to approximately 30 lm and
used to contain the sample between the diamonds. A 100-lm
diameter hole was drilled into the indented portion of the gasket
to serve as a sample chamber. Gold ﬂakes are the pressure indicator and were added to the chamber with one of the four solutions.
The gold was positioned in the chamber and then a large drop of
the solution was put over the pre-indented area and quickly sealed
by closing the DAC and compressing to >2 GPa to seal the chamber
and form ice VII. The bubble points for the 1.6 mol% NaCl, 1.6 mol%

14

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20

CH3OH, and 5.0 mol% CH3OH aqueous solutions are not depressed
substantially below that of pure H2O so the compositions of the
solutions added during loadings were not modiﬁed by evaporation.
The samples were compressed up to approximately 70, 71, 69, and
48 GPa for ices formed from H2O, 1.60 mol% NaCl, 1.60 mol% CH3OH, and 5.00 mol% CH3OH aqueous solutions, respectively.
2.3. Powder X-ray diffraction
Experiments were conducted at the GSECARS 13-BM-D beam
line of the Advanced Photon Source at Argonne National Laboratory. Synchrotron X-ray radiation with a monochromatic wavelength of 0.3344 Å and overall beam size of 6  15 lm was used
for in situ powder diffraction analysis. The geometry of the setup was calibrated by using CeO2 with data processed using FIT2D
software (Hammersley, 1997). The contents of the sample chamber
were analyzed by using an angle dispersive powder X-ray diffraction technique with 5-min exposure times. Multiple patterns were
taken over the course of 30 min to evaluate the stress of the sample. Both high-pressure ice VII (1 1 0, 2 0 0, and 2 1 1) and gold diffraction lines (1 1 1, 2 0 0, 2 2 0, and 3 1 1) were present in most of
the diffraction patterns and had to be distinguished from one another, but there was little overlap between the diffraction lines
of both materials throughout majority of the pressure range studied. At lower pressure, diffraction patterns of ice VII only could be
obtained to index the (1 1 0) peak which overlaps with gold’s (1 1 1)
peak in a mixed pattern, however, the compaction of the sample
chamber at higher pressures made obtaining an ice-only pattern
difﬁcult and compression was stopped once the ice (1 1 0) peak began to overlap with gold’s (2 0 0) peak. Gold was used as the internal pressure indicator and its unit cell parameter (a) was
determined at each exposure and, using the Anderson et al.
(1989) EoS, a pressure was calculated. Additionally, the (2 0 0) diffraction line of Au was closely monitored to provide an indication
of the stress state of the sample. The calculation of the uncertainty
of the pressure determination relied on the standard deviation obtained from the mean lattice parameter calculated from the gold
diffraction lines. The uncertainty of the volume of the high pressure ice was similarly determined by using all available ice diffraction lines.
3. Results
The pure H2O, NaCl-bearing, and CH3OH-bearing aqueous solutions were loaded into separate diamond anvil cells and pressurized to >2 GPa. The high-pressure polymorphs of H2O and CH3OH
are transparent in the visible range and thus it is easy to observe
optically any phase separation in the DAC. In general, all ice phases
have different refractive indices and would exhibit relief if those
two phases (or NaCl) were present in the sample chamber. Optical
observations at pressures greater than 2 GPa noted only a single
transparent H2O-rich ice phase and opaque ﬂakes of gold; no separate NaCl-rich or CH3OH-rich phase appeared to be present in the
sample chamber.
3.1. Powder X-ray diffraction
The H2O, 1.60 and 5.00 mol% CH3OH–H2O, and 1.60 mol% NaCl–
H2O solutions were compressed to pressures up to 71 GPa at 298 K
(Table 1). Powder diffraction data indicated that ice VII and gold
were the only phases present in the sample chamber. Diffraction
lines were indexed with FCC gold, Miller indices (1 1 1), (2 0 0),
(2 2 0), and (3 1 1), and BCC ice VII, (1 1 0), (2 0 0), and (2 1 1) (Figs. 1–
4). There were no additional peaks that suggested the presence of
CH3OH-rich or NaCl-bearing ice separates from the H2O-rich ice,

and no extra amorphous phases were observed optically in the
sample chamber for experiments utilizing a methanol-bearing
solution. Multiple samples were loaded and compressed to demonstrate the reproducibility of the method and to illustrate that the
CH3OH-bearing solutions did not change in composition due to
preferential vaporization of methanol.
The ice VII (2 0 0) and gold (2 0 0) lines were sometimes inconsistent relative to the other diffractions lines for a given phase
when a pattern was taken after a pressure increase. This suggests
the sample was under some deviatoric stress (Meng et al.,1993),
however, diffraction patterns collected tens of minutes after the
pressure increase did not exhibit stresses beyond those related to
the uncertainty in the pressure measurement. Thus, the retrieved
lattice parameter and cell volumes of all ice VII samples were obtained from diffraction patterns collected after the deviatoric stress
lessened and, as a result, produced a mostly smooth PV curve with
only slight deviations from the projected trends. A noted exception
to this is an apparent dip in the volumes of all ice samples occurring at approximately 50–55 GPa and becoming more pronounced
at 60 GPa. The decrease in volume at these pressures ﬂattened out a
bit as the pressure was increased to 70 GPa, but no substantial and
unambiguous change in volume was noticed within the precision
of our experiments at the supposed transition of ice VII to ice X
at approximately 60 GPa.
3.2. Isothermal compression at 298 K
The ice VII diffraction lines (1 1 0), (2 0 0), and, where present,
(2 1 1) were used to calculate its unit cell parameter (a). The unit
cell parameter was then used to calculate the volume of ice VII
in Å3. The ice VII PV data from the various solutions were each ﬁt
to a third-order Birch–Murnaghan EoS (Birch, 1978):

" 7  5 #
V0 3
V0 3

V
V
!#
 2
 V0 3
3
1
 1  4  K 0T0
4
V

3
PðGPaÞ ¼ K T0
2
"

where KT0, K 0T0 , and V0 are, respectively, the bulk modulus, its pressure derivative, and the zero-pressure volume of ice VII. As ice VII is
not a quenchable phase, the zero-pressure volume has to be either
assumed or calculated along with the other two variables. The compression data for ice VII formed from H2O are broadly consistent
with previous studies and give KT0, K 0T0 , and V0 of 20.14 ± 1.1 GPa,
4.62 ± 0.07, and 40.8 ± 0.33 Å3, respectively (Fig. 5) with a Q-factor
of 0.71. The addition of 1.60 mol% Na+ and Cl to ice VII resulted
in a decrease of V0 to 38.96 ± 0.08 Å3 with KT0 of 26.47 ± 0.46 GPa
and K 0T0 of 4.27 ± 0.03 and an overall Q-ﬁt of 0.02 (Fig. 6). Ice VII
samples formed from CH3OH–H2O solutions had an opposite trend
in volume with an observed increase relative to pure H2O at any given pressure (Fig. 7). The V0, KT0, and K 0T0 were found to be
41.0 ± 0.19 Å3, 21.53 ± 0.68 GPa, and 4.37 ± 0.04 for ice VII formed
from a 1.60 mol% CH3OH aqueous solution; Q was 0.965. Ice VII
formed from a 5.00 mol% CH3OH aqueous solution had an even
greater increase in V0 to 41.0 ± 0.19 Å3 with a KT0 of
19.58 ± 1.08 GPa and K 0T0 of 4.62 ± 0.04; Q was 0.01 (Fig. 8).
The most readily observable trend in the compilation of all of
the ice VII data is that the addition of Na+ and Cl resulted in a decrease in the volume of ice VII relative to that formed from H2O
(Fig. 9). The data from Frank et al. (2006) are consistent with the
data obtained in this study, although their data has greater scatter.
In contrast, the addition of CH3OH (either 1.60 or 5.00 mol%) produced ice VII with a noticeably larger volume relative to pure
H2O systems. These opposite effects suggest the incorporation of
Na+ and Cl into ice VII does not produce the same structural ef-

15

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20
Table 1
Pressure–volume data for ice VII collected at 295 K. The uncertainties are 2r standard deviation from the mean.
Ice VII formed from H2O

Ice VII formed from 1.60 mol% NaCl–H2O
3

3

Ice VII formed from 1.60 mol% CH3OH–H2O

Ice VII formed from 5.00 mol% CH3OH–H2O

Pressure (GPa)

Volume (Å )

Pressure (GPa)

Volume (Å )

Pressure (GPa)

Volume (Å )

Pressure (GPa)

Volume (Å3)

2.30 ± 0.29
3.89 ± 0.35
4.35 ± 0.22
5.03 ± 0.27
5.79 ± 0.37
7.86 ± 0.17
9.33 ± 0.39
11.63 ± 0.23
14.58 ± 0.12
16.98 ± 0.37
18.47 ± 0.63
19.24 ± 0.06
20.91 ± 0.54
22.64 ± 0.51
24.41 ± 0.88
26.43 ± 0.57
29.73 ± 0.34
31.99 ± 0.76
33.35 ± 0.71
34.63 ± 0.72
36.29 ± 0.45
38.24 ± 0.64
38.87 ± 0.65
40.51 ± 0.89
41.83 ± 0.57
44.89 ± 0.62
48.08 ± 0.38
49.18 ± 0.97
51.00 ± 0.91
52.47 ± 0.81
53.06 ± 0.85
55.24 ± 0.42
56.25 ± 0.60
59.46 ± 0.81
61.30 ± 0.62
63.70 ± 0.76
66.94 ± 0.68
68.65 ± 0.24
70.05 ± 0.99

37.17 ± 0.30
35.66 ± 0.03
35.10 ± 0.01
34.64 ± 0.02
34.00 ± 0.04
32.49 ± 0.01
31.82 ± 0.14
30.30 ± 0.11
29.25 ± 0.01
28.44 ± 0.02
27.97 ± 0.02
27.70 ± 0.02
27.26 ± 0.02
26.83 ± 0.02
26.38 ± 0.03
25.96 ± 0.10
25.34 ± 0.10
24.89 ± 0.13
24.77 ± 0.18
24.53 ± 0.12
24.07 ± 0.08
23.95 ± 0.10
23.83 ± 0.07
23.57 ± 0.10
23.38 ± 0.12
23.03 ± 0.10
22.70 ± 0.15
22.47 ± 0.18
22.25 ± 0.14
22.14 ± 0.14
22.07 ± 0.09
21.81 ± 0.09
21.70 ± 0.07
21.38 ± 0.08
21.16 ± 0.06
20.89 ± 0.06
20.74 ± 0.10
20.63 ± 0.05
20.53 ± 0.10

6.58 ± 0.37
7.73 ± 0.41
9.16 ± 0.58
9.76 ± 0.21
11.27 ± 0.29
13.6 ± 0.10
17.37 ± 0.57
19.99 ± 0.48
24.78 ± 0.10
26.76 ± 0.33
28.65 ± 0.87
31.35 ± 0.51
34.39 ± 0.12
37.66 ± 0.21
38.65 ± 0.47
39.38 ± 0.44
40.27 ± 0.75
42.03 ± 0.24
44.61 ± 0.50
48.38 ± 0.60
50.17 ± 0.41
53.73 ± 0.36
55.36 ± 0.51
57.86 ± 0.54
61.76 ± 0.49
62.99 ± 0.35
63.49 ± 0.79
64.57 ± 0.47
66.19 ± 0.45
69.73 ± 0.20
70.86 ± 0.32

32.72 ± 0.05
32.03 ± 0.03
31.48 ± 0.10
31.21 ± 0.04
30.38 ± 0.01
29.36 ± 0.03
28.15 ± 0.03
27.36 ± 0.03
26.22 ± 0.10
25.70 ± 0.10
25.37 ± 0.05
24.89 ± 0.06
24.38 ± 0.04
23.90 ± 0.08
23.71 ± 0.06
23.59 ± 0.03
23.48 ± 0.01
23.25 ± 0.01
22.91 ± 0.01
22.46 ± 0.02
22.13 ± 0.10
21.80 ± 0.05
21.69 ± 0.10
21.44 ± 0.02
21.10 ± 0.08
20.98 ± 0.05
20.93 ± 0.02
20.87 ± 0.07
20.67 ± 0.06
20.37 ± 0.08
20.31 ± 0.05

2.56 ± 0.08
4.57 ± 0.37
6.54 ± 0.14
7.56 ± 0.23
7.73 ± 0.38
8.52 ± 0.41
8.64 ± 0.52
9.40 ± 0.26
10.52 ± 0.13
11.65 ± 0.49
12.02 ± 0.49
12.37 ± 0.43
13.07 ± 0.34
13.38 ± 0.10
14.04 ± 0.48
14.70 ± 0.31
16.24 ± 0.16
16.35 ± 0.16
17.10 ± 0.67
18.02 ± 0.07
19.68 ± 0.20
20.06 ± 0.24
21.56 ± 0.22
22.11 ± 0.48
23.37 ± 0.23
24.86 ± 0.21
25.42 ± 0.23
26.90 ± 0.35
29.13 ± 0.21
30.51 ± 0.51
31.28 ± 0.32
33.23 ± 0.44
33.58 ± 0.57
34.28 ± 0.42
35.61 ± 0.53
41.93 ± 0.67
44.30 ± 0.27
45.30 ± 0.76
46.83 ± 0.79
49.01 ± 0.61
50.06 ± 0.73
51.83 ± 0.30
53.18 ± 0.61
54.85 ± 0.87
58.37 ± 0.05
63.47 ± 0.48
66.08 ± 0.46
68.01 ± 0.97
69.37 ± 0.87

37.29 ± 0.09
35.05 ± 0.07
33.59 ± 0.01
32.91 ± 0.10
32.85 ± 0.02
32.40 ± 0.04
32.30 ± 0.03
31.85 ± 0.02
31.28 ± 0.01
30.85 ± 0.03
30.49 ± 0.03
30.37 ± 0.02
30.14 ± 0.01
29.94 ± 0.05
29.71 ± 0.02
29.49 ± 0.01
28.92 ± 0.02
28.86 ± 0.04
28.59 ± 0.03
28.38 ± 0.10
27.99 ± 0.15
27.67 ± 0.01
27.44 ± 0.08
27.15 ± 0.04
26.83 ± 0.02
26.38 ± 0.02
26.29 ± 0.04
25.94 ± 0.04
25.55 ± 0.10
25.26 ± 0.06
25.08 ± 0.03
24.72 ± 0.05
24.68 ± 0.03
24.54 ± 0.02
24.31 ± 0.07
23.34 ± 0.08
23.06 ± 0.05
22.90 ± 0.07
22.72 ± 0.08
22.47 ± 0.03
22.33 ± 0.07
22.14 ± 0.02
21.99 ± 0.05
21.85 ± 0.09
21.52 ± 0.09
21.03 ± 0.10
20.80 ± 0.05
20.69 ± 0.02
20.55 ± 0.06

3.22 ± 0.18
5.59 ± 0.50
5.75 ± 0.30
6.01 ± 0.36
7.78 ± 0.03
8.98 ± 0.45
11.90 ± 0.12
13.12 ± 0.52
15.92 ± 0.60
16.77 ± 0.37
18.28 ± 0.10
18.79 ± 0.16
19.62 ± 0.59
19.99 ± 0.17
21.76 ± 0.23
23.28 ± 0.77
24.15 ± 0.65
29.37 ± 0.19
31.30 ± 0.36
32.33 ± 0.78
33.70 ± 0.05
35.85 ± 0.10
37.83 ± 0.08
41.24 ± 0.67
42.70 ± 0.23
43.75 ± 0.81
44.97 ± 0.59
45.55 ± 0.59
46.21 ± 0.82
47.56 ± 0.20

36.55 ± 0.02
34.64 ± 0.04
34.21 ± 0.02
34.07 ± 0.01
32.64±0.06
32.07 ± 0.05
30.71 ± 0.03
30.11 ± 0.06
29.02 ± 0.03
28.76 ± 0.06
28.30 ± 0.03
27.99 ± 0.06
27.76 ± 0.03
27.71 ± 0.06
27.23 ± 0.01
26.84 ± 0.04
26.73 ± 0.02
25.60 ± 0.04
25.17 ± 0.01
24.97 ± 0.03
24.67 ± 0.01
24.24 ± 0.03
24.05 ± 0.01
23.73 ± 0.06
23.40 ± 0.05
23.24 ± 0.04
23.08 ± 0.06
22.95 ± 0.02
22.91 ± 0.08
22.75 ± 0.05

fects as the incorporation of molecular CH3OH. The difference in ice
volumes is greatest at low pressures and decreases with increasing
pressure up to approximately 45 GPa where the volumetric
differences are within the uncertainty of the measurement. Finally,
a decrease in all ice volumes was observed at approximately
50 GPa and again at 60 GPa relative to the EoS for ice VII formed
from H2O.

4. Discussion
4.1. Isothermal compression of H2O ice VII
The isothermal compression data of the molecular study were
compared to several pure ice VII compression studies (Fig. 10
and Table 2). The V0 determined in this study, 40.8 ± 0.33 Å3, is
broadly consistent with all of the previous studies except that of
Loubeyre et al. (1999), they reported a value of 48.2 Å3. The KT0

3

(GPa) determined in this study for pure ice VII is approximately
15% less than those of Liu (1982), Hemley et al. (1987) and Fei
et al. (1993), but is approximately 30% greater than the KT0 of Wolanin et al. (1997). Additionally, this study’s reported K 0T0 is 10%
greater than those of Hemley et al. (1987) and Fei et al. (1993)
and 25% less than Wolanin et al. (1997). A comparison of the model
PV curves from all studies (Fig. 10) illustrates that the data from
this study are at generally lower volumes for any given pressure
when compared to the Hemley et al. (1987), Fei et al. (1993) and
Frank et al. (2004) studies, but at greater volumes than those suggested by the EoS from Wolanin et al. (1997) and Loubeyre et al.
(1999). The observed deviations in the published PVT models is
produced by the lack of a control on the V0, a result of the nonquenchable nature of ice VII, and the compressibility of ice VII in
the 10–40 GPa range showing the greatest variability. Generally,
these factors produce a group of models suggesting a less compressible ice VII, including Liu (1982), Hemley et al. (1987) and
Fei et al. (1993), and a more compressible one, such as Wolanin

16

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20

Fig. 1. Diffraction pattern of ice VII formed from H2O. A pressure of
14.57 ± 0.12 GPa was estimated from the gold diffraction lines (1 1 1), (2 0 0),
(2 2 0), (3 1 1), and (2 2 2). Ice VII diffraction lines (1 1 0), (2 0 0), and (2 1 1) were used
to calculate a volume corresponding to this pressure.

Fig. 2. Diffraction pattern of ice VII at 9.76 ± 0.21 GPa formed from a 1.60 mol%
NaCl–H2O aqueous solution. All diffractions lines can be attributed to gold, the
pressure marker, or ice VII. No diffraction lines indicative of NaCl were present. Ice
VII diffraction lines from (1 1 0) and (2 0 0) were detected in the experimental runs
and used to determine a for the ice VII formed from the NaCl–H2O aqueous solution.

et al. (1997) and Loubeyre et al. (1999), with the data from this
study and Frank et al. (2004) falling in between these end
members.
Recent studies have suggested that the ice VII diffraction lines
may split indicating a modulated structure and suggesting a lowering of symmetry at high pressure (Somayazulu et al., 2008). However, neutron diffraction data are consistent with a cubic structure
(e.g., Kuhs et al., 1984; Nelmes et al., 1993). Although we did observe minor peak splitting in some of our patterns, the splitting
was spurious and not consistent and may be due to large shear
stresses (Ahart et al., 2011). Furthermore, the unit cell volume calculated from the three observable ice VII diffraction lines were
consistent within the established uncertainty and indicate that
analyzing the diffraction patterns assuming a BCC ice VII phase is
appropriate for comparing PV data and calculating KT0 and K 0T0 .

Fig. 3. Diffraction pattern of ice VII formed from a 1.60 mol% CH3OH–H2O aqueous
solution and at 11.65 ± 0.49 GPa. The gold diffraction lines (1 1 1), (2 0 0), and (2 2 0)
and the ice VII diffraction lines (1 1 0) and (2 0 0) were the only diffraction lines
present in the pattern. There was no evidence of CH3OH existing as a separate
phase.

Fig. 4. Ice VII formed from a 5.00 mol% CH3OH–H2O aqueous solution. A pressure of
11.65 ± 0.49 GPa was estimated from the gold diffraction lines. No diffraction lines
indicative of CH3OH were noted in any X-ray diffraction pattern, suggesting that all
of the CH3OH was incorporated into the ice VII structure. Ice VII diffraction lines
from (1 1 0) and (2 0 0) were used to determine a for the ice VII formed from the
CH3OH–H2O aqueous solution.

4.2. Isothermal compression of ice VII formed with NaCl
Frank et al. (2006) noticed a systematic decrease in the unit cell
volume of ice VII when ionic impurities, Na+ and Cl, were present.
Furthermore, they suggested that the incorporation of ionic impurities into the BCC structure of ice VII would result in a partial
ordering of hydrogen which is akin to the ice X structure. This contention is supported by their PV data, which project up to a more
dense structure of ice X, as well as Raman spectroscopic measurements showing a decrease in the OH bond length with the addition of Na+ and Cl. They suggested this resulted in an ice X-like
structure, where the protons are equidistant between the oxygen

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20

Fig. 5. X-ray diffraction data were used to calculate an isothermal equation of state
at 295 K for ice VII formed from H2O. The solid circles represent individual PV data
whereas the error bars are calculated following the methods outlined in the text
and are 2r. A third-order Birch–Murnaghan equation of state (EoS) was ﬁtted to the
data and produced a zero-pressure volume (V0), bulk modulus (KT0), and its
pressure derivative (K 0T0 ). The EoS from Frank et al. (2004) is drawn for comparison.

Fig. 6. Pressure-volume data for ice VII formed from a 1.60 mol% NaCl aqueous
solution at 295 K. At any given pressure, the data have consistently lower volumes
relative to ice VII formed from pure H2O. Frank et al. (2006) conducted a study in
the same system, but their data were limited to less than 28 GPa whereas we
collected data up to approximately 70 GPa. Both data sets compare favorably and a
third-order Birch–Murnaghan equation of state (EoS) was ﬁtted to all of the PV data.

atoms, at much lower pressures than found in pure H2O systems.
The PV data from this study for ice VII formed from a 1.60 mol%
NaCl aqueous solution were found to be consistent with Frank
et al. (2006) (Fig. 6). The data presented here extended the pressure
range up to approximately 71 GPa from the previous maximum

17

Fig. 7. The pressure–volume relations of ice VII formed from a 1.60 mol% CH3OH
aqueous solution at 295 K are compared to the EoS for pure ice VII. The data for
CH3OH-bearing ice at pressures up to 40 GPa are at systematically elevated volumes
when compared to those expected for pure ice, but, at pressures great than 40 GPa,
the CH3OH-bearing ice is indistinguishable from pure ice VII.

Fig. 8. X-ray diffraction data for ice VII formed from a 5.00 mol% CH3OH aqueous
solution are compared to the EoS for pure ice and ice with 1.60 mol% CH3OH. The
data show an increase in volume compared to pure ice and a slight increase relative
to ice with 1.60 mol% CH3OH.

pressure of 27.34 GPa and allowed for a more precise determination of V0, KT0, and K 0T0 (Table 2). The KT0 of ice VII doped with
Na+ and Cl was greater than that of pure ice whereas the K 0T0
was found to decrease illustrating the Na+ and Cl stiffened the
ice’s structure. The KT0 and K 0T0 values reported by Frank et al.
(2006) were 25.7 ± 0.4 and 4.15 which are consistent with the values found in this study. The data do not show a substantial change

18

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20
Table 2
This is a comparison of the zero-pressure volume V0), bulk modulus (KT0), and
pressure derivative ðK 0T0 Þ of pure ice VII (Fei et al., 1993; Frank et al., 2004; Hemley
et al., 1987; Liu, 1982; Loubeyre et al., 1999; Wolanin et al., 1997) and ice VII formed
from a 1.60 mol% NaCl–H2O aqueous solution (Frank et al., 2006) to ice VII formed
from the H2O, NaCl–H2O, and CH3OH–H2O solutions of this study.

Ice VII compression EoS
One-component, H2O
1.60 mol% NaCl–H2O
1.60 mol% CH3OH–H2O
5.00 mol% CH3OH–H2O

Fig. 9. The pressure–volume data for all ice VII samples with either NaCl or CH3OH
as an impurity are presented as function of pressure. The VIce VII – VPure Ice VII term is
the calculated difference between the measured volume and that predicted for pure
ice VII at the same pressure. Ice formed from a NaCl-bearing solution was found to
decrease volumetrically compared to pure ice whereas the incorporation of CH3OH
into ice VII produced an overall opposite effect and expanded the volume relative to
pure ice. Samples formed with 1.60 mol% CH3OH became indistinguishable from
pure ice at pressures greater than 40 GPa while higher concentrations of CH3OH
produced a greater difference at those same pressures.

V0 (Å3)

KT0 (GPa)

K 0T0

40.8 ± 0.33
38.96 ± 0.08
41.0 ± 0.19
41.3 ± 0.29

20.14 ± 1.11
26.47 ± 0.46
21.53 ± 0.68
19.58 ± 1.08

4.62 ± 0.07
4.27 ± 0.03
4.37 ± 0.04
4.62 ± 0.04

Previous ionic impurity study – 1.60 mol% NaCl–H2O
Frank et al. (2006)
39.1 ± 0.1
25.7 ± 0.4

4.15

Select previous one component (H2O) studies
Frank et al. (2004)
41.2 ± 0.1
Fei et al. (1993)
40.9 ± 0.7
Hemley et al. (1987)
40.9 ± 0.9
Loubeyre et al. (1999)a
48.2
Liu (1982)
42.2 ± 0.3
b
Wolanin et al. (1997)
41.1 ± 0.3
c
40.2 ± 0.3
Wolanin et al. (1997)

4.4 ± 0.1
4.2 ± 0.5
4.15 ± 0.07
7.75
4.4
5.4 ± 0.1
6.2 ± 0.1

21.3 ± 1.3
23.9 ± 0.7
23.7 ± 0.9
4.26
24.6 ± 0.9
14.9 ± 0.8
14.9 ± 0.8

a
Data were obtained from a single crystal and were ﬁt to the Vinet equation of
state (Vinet et al., 1987).
b
Data were ﬁt to the Birch–Murnaghan equation of state (Birch, 1978).
c
Data were ﬁt to the Vinet equation of state (Vinet et al., 1987).

sis of an ice X-like structure as proposed by Frank et al. (2006) and
the intermediate transitional steps suggested by Benoit et al.
(1998) and in the modeling of Marx (2006).
4.3. Isothermal compression of ice VII formed with CH3OH
The isothermal compression data on ice VII formed from two
CH3OH aqueous solutions (1.60 and 5.00 mol%) data were compared to those for pure ice VII from this study, Hemley et al.
(1987), Fei et al. (1993), Wolanin et al. (1997), Loubeyre et al.
(1999) and Frank et al. (2004) as well as the data from NaCl-doped
ice VII (Frank et al., 2006; and Figs. 7 and 8 of this study). While a
systematic decrease in volume was noted in NaCl doped ice VII,
the presence of CH3OH resulted in a measureable increase in the
unit cell volume for any given pressure up to 40 GPa when compared to ice VII formed from pure H2O and when doped with Na+
and Cl. At pressures greater than 40 GPa, the calculated volume
of ice VII formed from the 1.60 mol% CH3OH solution was indistinguishable from that of pure ice. The PV data for 5.00 mol% CH3OH
ice VII are slightly elevated relative to that of the 1.60 mol% CH3OH
data, but, unlike those data, remain elevated relative to pure ice at
pressures greater than 40 GPa (Fig. 9). These data illustrate that the
addition of CH3OH produces an expanded unit cell for ice VII and
that the extent of this expansion is directly related to the concentration of CH3OH.
4.4. The ice VII to ice X transition

Fig. 10. A comparison of select pressure–volume equations of state (EoS) for ice VII
formed from pure H2O at room temperature. The EoS presented in this study
implies a slight depressed volume relative to Hemley et al. (1987) and Frank et al.
(2004), but an elevated volume relative to Wolanin et al. (1997) and Loubeyre et al.
(1999). The differences in these models underline the inherent uncertainty
associated with high pressures studies and are likely the result of variable pressure
markers and their equations of state as well as the difﬁculty of working with highly
compressible sample.

in volume at the reported ice VII-ice X phase transition pressure,
but a slight decrease in volume was observed at approximately
50 and again at 62 GPa and thus, tentatively support the hypothe-

Previous studies of the ice VII-ice X phase transition reported a
small volume decrease across the transition (e.g., the data in Hemley et al., 1987). This study tried to observe this decrease in volume
with the change of phase as well as determine what, if any, effect
the presence of impurities would have on this phase transition. The
data at pressures near 60 GPa, the transition of ice VII to ice X, do
show a slight volumetric decrease which is consistent with studies
hypothesizing the existence of ice X (Aoki et al., 1996; Goncharov
et al., 1999; Song et al., 2003; Asahara et al., 2010), but the data do
not show a decrease in volume that can be conclusively linked to
ice X and instead suggest possible intermediate stages of the transition due to a double well potential (Benoit et al., 1998; Marx,

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20

2006; Sugimura et al., 2008). The two-component system of
1.60 mol% NaCl–H2O was found to be consistent with the previous
study of Frank et al. (2006) with no discernible volume change at
approximately 60 GPa. Hence, at the suggested ice VII to X phase
transition, the data do not show any readily apparent change in
volume for the NaCl–H2O system. The other two-component system, 1.60 mol% CH3OH–H2O, was broadly consistent with the PV
curve of H2O and also showed a decrease in the unit cell volume
at 60 GPa. Taken together, these data cannot be used to disprove
the existence of ice X and that the phase transition occurs at
approximately 60 GPa, but the data do suggest that volumetric
change for ice formed in the NaCl–H2O system was less pronounced than in the H2O and CH3OH–H2O systems.
4.5. Structural controls on impurities in ice VII
The BCC structure of ice VII has open, face-centered sites where
interstitial impurities are most likely to reside. Depending on the
impurity in the face-centered sites, there may or may not be
noticeable impact on the unit cell of ice VII. Our work supports
the conclusions of Frank et al. (2006) that that Na+ and Cl ﬁt into
the open, face-centered sites of ice VII and, due to the charges on
the ions and the polar nature of the H2O molecule, an inter-ionic,
or electrostatic, attraction between the H2O molecule and the ionic
impurity caused the hydrogen of the O–H bond in the H2O molecule to rotate towards the Cl ion and away from the Na+ ion. They
suggest that this caused the hydrogen to become more ordered,
giving an ice X-like structure at pressures less than 60 GPa. The
result of this change to an ice X-like structure is reﬂected in the decrease in the unit cell volume for ice VII formed from NaCl-bearing
aqueous solutions.
The incorporation of a molecular impurity, such as CH3OH, is
also hypothesized to ﬁt into the large voids found in the face-centered sites of the ice VII BCC structure. However, this impurity does
not exist as charged ions and instead remains as an associated,
electrically neutral molecule. There is a small polarity to CH3OH
due to the structure of the molecule which might have caused a
slight rotation of the molecule within the lattice, but the potential
effect of this is assumed to be small or nonexistent. From the data
presented in this study, it is hypothesized that there is little to no
inter-ionic attraction between the polar H2O molecule and the CH3OH impurity and that the main impact is an expansion of the unit
cell relative to that found in pure ice VII. Thus, the high-pressure
ice with the CH3OH impurity maintains PV properties closely allied
to pure ice VII but the unit cell became progressively larger, especially at lower pressures, when compared to pure ice VII.
With the addition of impurities into the high-pressure ice
phases, there must be a limit to how much of an impurity can be
present in the ice VII structure before it forms an impurity-rich
phase separate of the ice within the sample chamber. Frank et al.
(2006) determined that a maximum amount of 2.4 ± 0.8 mol% NaCl
could be incorporated into the ice VII structure. The presence of a
new impurity-rich phase was conﬁrmed by the appearance of NaCl
diffraction lines in the powder X-ray diffraction data as the NaCl
could not be incorporated into the unit cell of ice VII and formed
a separate phase. This study was designed to test if a similar
‘‘impurity limit’’ could be observed in the CH3OH–H2O system.
The presence of a second, CH3OH-rich phase in the sample chamber would support the hypothesis that the solution was of a concentration beyond which can be stabilized in the ice VII
structure. In that case, diffraction lines consistent with CH3OH
would be detected in the analyses of the powder X-ray diffraction
data. No diffraction lines indicative of CH3OH were observed in
experiments utilizing aqueous solutions of 1.60 and 5.00 mol%
CH3OH; thus, there is no diffraction-based evidence for a second
CH3OH-rich phase. However, CH3OH at high pressure has been

19

determined to be amorphous in select instances and would be difﬁcult to detect with X-ray diffraction techniques (Allan et al.,
1998). Thus, the sample chamber was monitored constantly for
any appearance of an optically distinct phase. None of the solutions
of CH3OH–H2O displayed any optical evidence of a separate CH3OH-rich phase and thus it was presumed that ice VII could hold a
minimum of 5.00 mol% CH3OH as an impurity. Due to absence of
any data suggesting a second phase was present, it was concluded
that the CH3OH was incorporated fully into the ice VII structure.
4.6. Planetary implications: H2O-rich planets
Ice VI and ice VII are likely to be present in the icy satellites of the
solar system (e.g., Anderson et al., 2001; Scott et al., 2002) whereas
ice VII and ice X might be stable in larger H2O-rich bodies such as
the extra-solar planet Gliese 436b (Gillon et al., 2007). It is probable
that the ice phases in these H2O-rich bodies are not pure and may
have various impurities incorporated into those phases. The differences in the volumes of ice VII with an ionic impurity or ice VII with
a molecular impurity clearly affect the densities of high-pressure
ices that exist in the interiors of icy bodies and hence would inﬂuence the internal proﬁle of any given planet inferred from modeled
density depth proﬁles. Difﬁculties in applying a general correction
to impure ice VII come when dealing with the possibility of the
presence of both ionic and molecular impurities in the high-pressure ice. If the ice-rich region in the interior of these bodies has both
molecular and ionic impurities present, then it could be assumed
that the density of this layer would be a value between the less
dense ice VII with a molecular impurity and the more dense ice
VII with an ionic impurity, depending, of course, on the supposed
proportions of the impurities. The opposite effects imposed by
these impurities means that models employing pure ice VII would
be a good ﬁrst order approximation. However, the presence of shallow magnetic ﬁelds below the outer ice layers of the icy satellites
(e.g., Khurana et al., 1998; Zimmer et al., 2000; Kivelson et al.,
2000) suggest that ionic impurities may, in fact, be the dominant
impurity incorporated into high pressure ices and thus the more
dense ice volume should be assumed in these instances.
5. Conclusion
Pressure–volume data were collected up to approximately
71 GPa and at room temperature for ice samples formed from four
aqueous solutions. The solutions were H2O, 1.60 mol% NaCl in H2O,
1.60 mol% CH3OH in H2O, and 5.00 mol% CH3OH in H2O. The H2O
ice was used as a baseline so that an internally consistent analysis
could be conducted on the effects of NaCl and CH3OH incorporation
into high-pressure ice VII. It was hypothesized that the incorporation of any impurity into the structure of ice VII would notably affect its unit cell volume and density. Since NaCl dissociates into
Na+ and Cl in H2O and CH3OH does not dissociate, the two chosen
impurities represent the end members of ionic and molecular species that could be incorporated in high pressure ice. Ice VII was
formed from each of the solutions and compressed to approximately 70 GPa for all but the 5.00 mol% CH3OH samples, which
were compressed to 48 GPa. The PV data show that the introduction of Na+ and Cl resulted in a volumetric decrease at any given
pressure when compared to pure ice VII. Conversely, the presence
of CH3OH had the opposite effect and caused an increase in the
ice’s unit cell volume relative to pure ice VII. Frank et al. (2006)
suggested that the presence of Na+ and Cl would produce an inter-ionic, or electrostatic, attraction between the H2O molecule
and the ionic impurity caused the hydrogen to rotate towards
the Cl ion and away from the Na+ ion. Our data support their
study and also suggest that a molecular impurity, which does not

20

M.R. Frank et al. / Physics of the Earth and Planetary Interiors 215 (2012) 12–20

exert a signiﬁcant electrical pull on the hydrogen in the structure,
produced an enlarged unit cell. Thus, these data indicate that
charged and non-charged ions/molecules that are incorporated
into the ice VII structure produce an opposite effect. With both
impurity varieties present, a range of acceptable densities can be
established for use in modeling the density proﬁles of H2O-rich
bodies with one end deﬁned by ice VII incorporating ionic impurities and the other by ice with large molecular impurities. The density of the ice VII present in a natural system would most likely fall
within the range deﬁned by these different ices.
Acknowledgements
This research was supported by the Research and Artistry Program at Northern Illinois University (MRF). Portions of this work
were performed at GeoSoilEnviroCARS (Sector 13), Advanced Photon Source (APS), Argonne National Laboratory. GeoSoilEnviroCARS
is supported by the National Science Foundation – Earth Sciences
(EAR-0217473), Department of Energy – Geosciences (DE-FG0294ER14466) and the State of Illinois. Use of the APS was supported
by the US Department of Energy, Basic Energy Sciences, Ofﬁce of
Science, under Contract No. W-31-109-Eng-38.
References
Ahart, M., Somayazulu, M., Gramsch, S.A., Boehler, R., Mao, H.-K., Hemley, R.J., 2011.
Brillouin scattering of H2O ice to megabar pressures. Journal of Chemical
Physics 134, 124517.
Allan, D., Clark, S., Brugmans, M., Ackland, G., Vos, W., 1998. Structure of crystalline
methanol at high pressure. Physical Review B 58 (18), 809–812.
Anderson, J.D., Lau, E.L., Sjogren, W.L., Schubert, G., Moore, W.B., 1996. Gravitational
constraints on the internal structure of Ganymede. Nature 384, 541–543.
Anderson, J.D., Lau, E.L., Sjogren, W.L., Schubert, G., Moore, W.B., 1997. Gravitational
evidence for an undifferentiated Callisto. Nature 387, 264–266.
Anderson, J.D., Schubert, G., Jacobson, R.A., Lau, E.L., Moore, W.B., Sjogren, W.L.,
1998. Distribution of rock, metals, and ices in Callisto. Science 280, 1573–1576.
Anderson, J.D., Jacobson, R.A., McElrath, T.P., Moore, W.B., Schubert, G., Thomas, P.C.,
2001. Shape, mean radius, gravity ﬁeld, and interior structure of Callisto. Icarus
153 (1), 157–161.
Anderson, O.L., Isaak, D.G., Yamamoto, S., 1989. Anharmonicity and the equation of
state for gold. Journal of Applied Physics 65, 1534–1543.
Aoki, K., Yamawaki, H., Sakashita, M., Fujihisa, H., 1996. Infrared absorption study of
the hydrogen-bong symmetrization in ice to 110 GPa. Physical Review B 54,
15673–15677.
Asahara, Y., Hirose, K., Ohishi, Y., Hirao, N., Murakami, M., 2010. Thermoelastic
properties of ice VII and its high-pressure polymorphs: implications for
dynamics of cold slab subduction in the lower mantle. Earth and Planetary
Science Letters 299, 474–482.
Benoit, M., Bernasconi, M., Marx, D., 1998. Tunnelling and zero-point motion in
high-pressure ice. Nature 392, 258–261.
Birch, F., 1978. Finite strain isotherm and velocities for single-crystal and
polycrystalline NaCl at high pressures and 300 K. Journal of Geophysical
Research A 83, 1257–1268.
Choukrouna, M., Grasset, O., 2007. Thermodynamic model for water and highpressure ices up to 2.2 GPa and down to the metastable domain. Journal of
Chemical Physics 127, 124506.
Consolmagno, G.J., Lewis, J.S., 1976. The evolution of icy satellite interiors and
surfaces. Icarus 34 (2), 280–293.
Fei, Y., Mao, H.K., Hemley, R.J., 1993. Thermal expansivity, bulk modulus, and melting
curve of H2O-Ice VII to 20 GPa. Journal of Chemical Physics 99, 5369–5373.
Fortes, A.D., Wood, I.G., Tucker, M.G., Marshall, W.G., 2011. The P-V-T equation of
state of D2O ice VI determined by neutron powder diffraction in the range 0 < P
< 2.6 GPa and 120 < T < 330 K, and the isothermal equation of state of D2O ice VII
from 2 to 7 GPa at room temperature. Journal of Applied Crystallography 45,
523–534.
Frank, M.R., Fei, Y., Hu, J., 2004. Constraining the equation of state of the ﬂuid H2O to
80 GPa using the melting curve, bulk modulus and thermal expansivity of Ice
VII. Geochimica et Cosmochimica Acta 68 (13), 2781–2790.
Frank, M.R., Runge, C.E., Scott, H.P., Maglio, S.J., Olson, J., Prakapenka, V.B., Shen, G., 2006.
Experimental study of the NaCl–H2O system up to 28 GPa: implications for ice-rich
planetary bodies. Physics of the Earth and Planetary Interiors 155, 152–162.
Frank, M.R., Scott, H.P., Maglio, S.J., Prakapenka, V.B., Shen, G., 2008. Temperature
induced immiscibility on the NaCl–H2O system at high pressure. Physics of the
Earth and Planetary Interiors 170, 107–114.

Gillon, M., Pont, F., Demory, B.-O., Mallmann, F., Mayor, M., Mazeh, T., Queloz, D.,
Shporer, A., Udry, S., Vuissoz, C., 2007. Detection of transits of the nearby hot
Neptune GJ 436 b. Astronomy and Astrophysics 472, L13–L16.
Goncharov, A.F., Struzhkin, V.V., Mao, H.K., Hemley, R.J., 1999. Raman spectroscopy
of dense H2O and the transition to symmetric hydrogen bonds. Physical Review
Letters 83, 1998–2001.
Gurnett, D.A., Kurth, W.S., Roux, A., Bolton, S.J., Kennel, C.F., 1996. Evidence for a
magnetosphere at Ganymede from plasma-wave observations by the Galileo
spacecraft. Nature 384, 535–537.
Hammersley, A.P., 1997. FIT2D: An introduction and overview. ESRF Internal Report
ESRF97HA02T.
Hemley, R.J., Jephcoat, A.P., Mao, H.K., Zha, C.S., Finger, L.W., Cox, D.E., 1987. Static
compression of H2O-Ice to 128 GPa (1.28 Mbar). Nature 330, 737–740.
Hubbard, W.B., Marley, M.S., 1989. Optimized Jupiter, Saturn, and Uranus interior
models. Icarus 78, 102–118.
Khurana, K.K., Kivelson, M.G., Stevenson, D.J., Schubert, G., Russell, C.T., Walker, R.J.,
Polansky, C., 1998. Induced magnetic ﬁelds as evidence for subsurface oceans in
Europa and Callisto. Nature 395, 777–780.
Kivelson, M.G., Khurana, K.K., Russell, C.T., Volwerk, M., Walker, R.J., Zimmer, C.,
2000. Galileo magnetometer measurements: a stronger case for a subsurface
ocean at Europa. Science 289, 1340–1343.
Klotz, S., Bove, L., Strassle, T., Hansen, T., Saitta, A., 2009. The preparation and
structure of salty ice VII under pressure. Nature Materials 8, 405–409.
Kuhs, W.F., Finney, J.L., Vettier, C., Bliss, D.V., 1984. Structure and hydrogen ordering
in ices VI, VII, and VIII by neutron powder diffraction. Journal of Chemical
Physics 81, 3612.
Kuskov, O.L., Kronrod, V.A., 2005. Models of the internal structure of Callisto. Solar
System Research 39 (4), 283–301.
Liu, L.-G., 1982. Compression of ice VII to 500 kbar. Earth and Planetary Science
Letters 61 (2), 359–364.
Loubeyre, P., LeToullec, R., Wolanin, E., Hanﬂand, M., Häusermann, D., 1999.
Modulated phases and proton centering in ice observed by X-ray diffraction up
to 170 GPa. Nature 397, 503–506.
Marx, D., 2006. Proton transfer 200 Years after von Grotthuss: insights fromAb initio
simulations. ChemPhysChem 7, 1848–1870.
McCord, T.B., Hansen, G.B., Hibbitts, C.A., 2001. Hydrated salt minerals on
Ganymede’s surface. Evidence of an ocean below. Science 292, 1523–1525.
Mueller, S., McKinnon, W.B., 1988. Three-layered models of Ganymede and Callisto:
compositions, structures, and aspects of evolution. Icarus 76, 437–464.
Meng, Y., Weidner, D.J., Fei, Y., 1993. Deviatoric stress in a quasihydrostatic
diamond anvil cell: effect on the volume-based pressure calibration.
Geophysical Research Letters 20, 1147–1150.
Nellis, W.J., Hamilton, D.C., Holmes, N.C., Radousky, H.B., Rhee, F.H., Mitchell, A.C.,
Nicol, M., 1988. The nature of the interior of Uranus based on studies of
planetary ices at high dynamic pressure. Science 240, 779–781.
Nelmes, R.J., Loveday, J.S., Wilson, R.M., Besson, J.M., Pruzan, P., Klotz, S., Hamel, G.,
Hull, S., 1993. Neutron diffraction study of the structure of deuterated ice VIII to
l0 GPa. Physical Review Letters 71, 1192.
Pruzan, P., Wolanin, E., Gauthier, M., Chervin, J.C., Canny, B., 1997. Raman scattering
and X-ray diffraction of ice in the megabar range occurrence of a symmetric
disordered solid above 62 GPa. Journal of Physical Chemistry B 101 (32), 6230–
6233.
Reynolds, R.T., Cassen, P.M., 1979. On the internal structure of the major satellites of
the outer planets. Geophysical Research Letters 6, 121–124.
Schubert, G., Zhang, K., Kivelson, M.G., Anderson, J.D., 1996. The magnetic ﬁeld and
internal structure of Ganymede. Nature 384 (6609), 544–545.
Scott, H.P., Williams, Q., Ryerson, F.J., 2002. Experimental constraints on the
chemical evolution of icy satellites. Earth and Planetary Science Letters 203,
399–412.
Sugimura, E., Iitaka, T., Hirose, K., Kawamura, K., Sata, N., Ohishi, Y., 2008.
Compression of H2O ice to 126 GPa and implications for hydrogen-bond
symmetrization: synchrotron X-ray diffraction measurements and densityfunctional calculations. Physical Review B 77, 214103.
Sohl, F., Spohn, T., Breuer, D., Nagel, K., 2002. Implications from Galileo observations
on the interior structure and chemistry of the Galilean satellites. Icarus 161,
104–119.
Somayazulu, M., Shu, J., Chang-sheng, Z., Goncharov, A.F., 2008. In situ high-pressure
X-ray diffraction study of H2O ice VII. Journal of Chemical Physics 128, 064510.
Song, M., Yamawaki, H., Fujihisa, M., Sakashita, M., Aoki, K., 2003. Infrared
investigation on ice VIII and the phase diagram of dense ices. Physical Review
B 68, 014106.
Spohn, T., Schubert, G., 2003. Oceans in the icy Galilean satellites of Jupiter? Icarus
161, 456–467.
Vinet, P., Smith, J.R., Ferrante, J., Rose, J.H., 1987. Temperature effect on the universal
equation of state of solids. Physical Review B 35, 1945–1953.
Wolanin, E., Pruzan, P., Chervin, J.C., Canny, B., Gauthier, M., Häusermann, D.,
Hanﬂand, M., 1997. Equation of state of ice VII up to 106 GPa. Physical Review B
56, 5781–5785.
Zimmer, C., Khurana, K.K., Kivelson, M.G., 2000. Subsurface oceans on Europa and
Callisto: constraints from Galileo magnetometer observations. Icarus 147, 329–
347.