Hydrogen at extreme pressures (Review Article)
Here we review recent experimental and theoretical studies of hydrogen approaching metallization regime. Ex-perimental techniques have made great advances over the last several years making it possible to reach previously unachievable conditions of pressure and temperature and to probe hydrogen at t...
Gespeichert in:
| Datum: | 2013 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
2013
|
| Schriftenreihe: | Физика низких температур |
| Schlagworte: | |
| Online Zugang: | https://nasplib.isofts.kiev.ua/handle/123456789/118453 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Zitieren: | Hydrogen at extreme pressures (Review Article) / Alexander F. Goncharov, Ross T. Howie, Eugene Gregoryanz, Ross T. Howie, Eugene Gregoryanz // Физика низких температур. — 2013. — Т. 39, № 5. — С. 523–530. — Бібліогр.: 94 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
nasplib_isofts_kiev_ua-123456789-118453 |
|---|---|
| record_format |
dspace |
| spelling |
nasplib_isofts_kiev_ua-123456789-1184532025-02-09T16:50:10Z Hydrogen at extreme pressures (Review Article) Goncharov, Alexander F. Howie, Ross T. Gregoryanz, Eugene 9th International Conference on Cryocrystals and Quantum Crystals Here we review recent experimental and theoretical studies of hydrogen approaching metallization regime. Ex-perimental techniques have made great advances over the last several years making it possible to reach previously unachievable conditions of pressure and temperature and to probe hydrogen at these conditions. Theoretical me-thods have also greatly improved; exemplified through the prediction of new structural and ordered quantum states. Recently, a new solid phase of hydrogen, phase IV, has been discovered in a high-pressure high-temperature do-main. This phase is quite unusual structurally and chemically as it represents an intermediate state between common molecular and monatomic configurations. Moreover, it shows remarkable fluxional characteristics related to its quantum nature, which makes it unique among the solid phases, even of light elements. However, phase IV shows the presence of a band gap and exhibits distinct phonon and libron characteristic of classical solids. The quantum behavior of hydrogen in the limit of very high pressure remains an open question. Prospects of studying hydrogen at more extreme conditions by static and combined static-dynamic methods are also presented. A.F.G. acknowledges support from the NSF, Army Re-search Office, NAI, and EFRee. E.G. and R. T. H acknowledge support from the U.K. Engineering and Physical Sciences Research Council and Institute of the Shock Physics, Imperial College. 2013 Article Hydrogen at extreme pressures (Review Article) / Alexander F. Goncharov, Ross T. Howie, Eugene Gregoryanz, Ross T. Howie, Eugene Gregoryanz // Физика низких температур. — 2013. — Т. 39, № 5. — С. 523–530. — Бібліогр.: 94 назв. — англ. 0132-6414 https://nasplib.isofts.kiev.ua/handle/123456789/118453 en Физика низких температур application/pdf Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
9th International Conference on Cryocrystals and Quantum Crystals 9th International Conference on Cryocrystals and Quantum Crystals |
| spellingShingle |
9th International Conference on Cryocrystals and Quantum Crystals 9th International Conference on Cryocrystals and Quantum Crystals Goncharov, Alexander F. Howie, Ross T. Gregoryanz, Eugene Hydrogen at extreme pressures (Review Article) Физика низких температур |
| description |
Here we review recent experimental and theoretical studies of hydrogen approaching metallization regime. Ex-perimental techniques have made great advances over the last several years making it possible to reach previously unachievable conditions of pressure and temperature and to probe hydrogen at these conditions. Theoretical me-thods have also greatly improved; exemplified through the prediction of new structural and ordered quantum states. Recently, a new solid phase of hydrogen, phase IV, has been discovered in a high-pressure high-temperature do-main. This phase is quite unusual structurally and chemically as it represents an intermediate state between common molecular and monatomic configurations. Moreover, it shows remarkable fluxional characteristics related to its quantum nature, which makes it unique among the solid phases, even of light elements. However, phase IV shows the presence of a band gap and exhibits distinct phonon and libron characteristic of classical solids. The quantum behavior of hydrogen in the limit of very high pressure remains an open question. Prospects of studying hydrogen at more extreme conditions by static and combined static-dynamic methods are also presented. |
| format |
Article |
| author |
Goncharov, Alexander F. Howie, Ross T. Gregoryanz, Eugene |
| author_facet |
Goncharov, Alexander F. Howie, Ross T. Gregoryanz, Eugene |
| author_sort |
Goncharov, Alexander F. |
| title |
Hydrogen at extreme pressures (Review Article) |
| title_short |
Hydrogen at extreme pressures (Review Article) |
| title_full |
Hydrogen at extreme pressures (Review Article) |
| title_fullStr |
Hydrogen at extreme pressures (Review Article) |
| title_full_unstemmed |
Hydrogen at extreme pressures (Review Article) |
| title_sort |
hydrogen at extreme pressures (review article) |
| publisher |
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України |
| publishDate |
2013 |
| topic_facet |
9th International Conference on Cryocrystals and Quantum Crystals |
| url |
https://nasplib.isofts.kiev.ua/handle/123456789/118453 |
| citation_txt |
Hydrogen at extreme pressures (Review Article) / Alexander F. Goncharov, Ross T. Howie, Eugene Gregoryanz, Ross T. Howie, Eugene Gregoryanz // Физика низких температур. — 2013. — Т. 39, № 5. — С. 523–530. — Бібліогр.: 94 назв. — англ. |
| series |
Физика низких температур |
| work_keys_str_mv |
AT goncharovalexanderf hydrogenatextremepressuresreviewarticle AT howierosst hydrogenatextremepressuresreviewarticle AT gregoryanzeugene hydrogenatextremepressuresreviewarticle |
| first_indexed |
2025-11-28T03:33:56Z |
| last_indexed |
2025-11-28T03:33:56Z |
| _version_ |
1850003522781708288 |
| fulltext |
© Alexander F. Goncharov, Ross T. Howie, and Eugene Gregoryanz, 2013
Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5, pp. 523–530
Hydrogen at extreme pressures
(Review Article)
Alexander F. Goncharov
1,2
1
Geophysical Laboratory, Carnegie Institution of Washington, Washington, D.C. 20015, USA
2
Key Laboratory of Materials Physics, Institute of Solid State Physics, Chinese Academy of Sciences,
Hefei 230031, China
E-mail: agoncharov@ciw.edu
Ross T. Howie and Eugene Gregoryanz
Centre for Science at Extreme Conditions and School of Physic and Astronomy, University of Edinburgh, Mayfield
Road, Edinburgh EH9 3JZ, United Kingdom
Received February 25, 2013
Here we review recent experimental and theoretical studies of hydrogen approaching metallization regime. Ex-
perimental techniques have made great advances over the last several years making it possible to reach previously
unachievable conditions of pressure and temperature and to probe hydrogen at these conditions. Theoretical me-
thods have also greatly improved; exemplified through the prediction of new structural and ordered quantum states.
Recently, a new solid phase of hydrogen, phase IV, has been discovered in a high-pressure high-temperature do-
main. This phase is quite unusual structurally and chemically as it represents an intermediate state between common
molecular and monatomic configurations. Moreover, it shows remarkable fluxional characteristics related to its
quantum nature, which makes it unique among the solid phases, even of light elements. However, phase IV shows
the presence of a band gap and exhibits distinct phonon and libron characteristic of classical solids. The quantum
behavior of hydrogen in the limit of very high pressure remains an open question. Prospects of studying hydrogen at
more extreme conditions by static and combined static-dynamic methods are also presented.
PACS: 64.30.Jk Equations of state of nonmetals;
67.80.F– Solids of hydrogen and isotopes.
Keywords: hydrogen, extreme pressures, solid phase of hydrogen.
Contents
1. Introduction ......................................................................................................................................... 523
2. Phase relations ..................................................................................................................................... 524
3. Melting and fluid behavior .................................................................................................................. 525
4. Phase II ................................................................................................................................................ 526
5. Phase III .............................................................................................................................................. 526
6. Phase IV .............................................................................................................................................. 528
7. Conclusions ......................................................................................................................................... 529
References ............................................................................................................................................... 529
1. Introduction
Hydrogen has a special interest for many fields of re-
search as it represents the perfect model object due to its
seeming simplicity and abundance in the cosmos [1–4].
One of the objectives of studying hydrogen at extreme
pressures is to rationalize the notion of metallic hydrogen
as a future energy carrier. There are three major technical
drivers in this pursuit: theoretical calculations and dynamic
and static compressions. Each has its own pressure —
temperature — time-scale domain, which largely do not
intersect and this poses a serious difficulty in unifying and
comparing results. This issue is now being addressed by
improving and modifying these techniques and by creating
new combined static-dynamic experimental methods.
With regard to theoretical and dynamic experimental
studies, we refer readers to the recent review on mainly the
theoretical approach to study hydrogen under extreme con-
ditions [5], which also contains a brief review of experi-
mental works. Study of hydrogen using dynamic compres-
sion (see the review papers [1,6,7]) is progressing very
rapidly now with a development of laser driven technique
compression of statically pre-compressed samples [8,9].
Alexander F. Goncharov, Ross T. Howie, and Eugene Gregoryanz
524 Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5
The purpose of this review is to critically look at the
experimental studies of the past two decades, which have
been performed using diamond anvil cell (DAC) tech-
niques and combined DAC heating experiments, covering
all known solid phases of hydrogen and its melting curve.
Static compression of hydrogen to very high pressure is
technically very challenging. Hydrogen is very compressi-
ble, while the materials commonly used to form the sample
chamber around it are not. Generating high-pressure on
hydrogen requires larger compression of the gasket materi-
al than with less compressible samples due to the limitation
of plastic flow. Secondly, hydrogen is very diffusive; it
tends to penetrate and rupture any small crack in both the
diamond and gasket. In the case of diamond this always
results in premature anvil failures. These effects accelerate
with temperature: rarely occurring below 100 K, but in-
creasing substantially above 200 K. Owing to this, until
2011, there were no reliable reports on static compression
of hydrogen or deuterium above 180 GPa at room temper-
ature [10]. Improved sample loading techniques, which
include diamond protective coating, focused electron beam
(FIB) gasket drilling, and better optimized anvil geometry
have recently allowed achieving static compression of hy-
drogen well above 300 GPa at 300 K [11].
These technical breakthroughs resulted in extending the
achievable pressure range for hydrogen research up to 320
GPa at 300 K [11] and up to 360 GPa at 80 K [12]. New
semiconducting (or semimetallic) solid phase hydrogen has
been discovered above 220 GPa at 300 K by combined ex-
perimental (Raman and optical spectroscopy) and theoretical
efforts [11,13]. A conflicting report claiming electrically
conducting hydrogen in the fluid state above 260–270 GPa
has been earlier published by Eremets and Troyan [14] infra-
red measurements in phase III to 360 GPa [12] also did not
report metallic conductivity. However, one should note, that
pressure metrology remains a problem as measurements of
the diamond Raman edge as pressure calibrant [15] are
somewhat uncertain, and, moreover, some experiments re-
lied on higher pressure extrapolations [12]. Here, we will
review the recent works and present prospects of new tech-
nical advances, which can enable next major breakthroughs.
2. Phase relations
Until recently, only three solid states of hydrogen have
been known (Fig. 1). Phase I is a plastic phase of freely ro-
tating molecules forming an hcp lattice whilst phases II and
III are partially (or completely) ordered phases, which ap-
pear at lower temperatures and/or higher pressures (see Refs.
2, 16, 17 for review). The symmetries and orientation order
types of phases II and III have been extensively discussed in
the literature based on experimental spectroscopy observa-
tions [2,16,17,19,33–37] and theoretical calculations [20,38–
45], however the available x-ray diffraction data are still not
conclusive [46–48]. The important issue of ortho–para dis-
tinction, and its effect on both the structure and phase transi-
tions, has also been discussed extensively. The available
data remain fragmentary due to difficulties in performing
experiments on materials with pure ortho–para composition.
Nonetheless, the current consensus is that the ortho–para
distinction does not affect the transition to phase III, which
is suggested to be classically orientationally ordered [18,49].
Due to technical difficulties, the extension of the phase line
between phases I and III to room temperature could not have
been reached until recently. It has been proposed [23] based
on the crystal symmetry arguments that this line should have
a termination at a critical point with finite P–T conditions,
and phase I’, with the same symmetry as phase III, merges
with phase I in the triple point, giving rise to the I–I’ phase
line (Fig. 1). Nevertheless, suggestions about the existence
of phase I’ based on these symmetry considerations, theoret-
ical calculations [20] or experimental observations of subtle
changes in vibrational frequencies [21] have yet to be con-
firmed (see Ref. 17 for more information). Instead, recently
it has been found that the I–III phase line does extend to
room temperature, and perhaps even beyond, where it meets
a new phase line with solid phase IV (Fig. 1). At room tem-
perature the transition sequence is I–III–IV, and the corres-
ponding transitions occur at 200 and 230 GPa (in H2) [11].
Fig. 1. (Color online) Phase diagram of hydrogen. The I–II and I–
III phase line for normal H2 are from Ref. 18; the I–III phase (solid
line) has been corrected as proposed in Ref. 17. The filed circle is
room temperature data from Ref. 11; the dashed line is the pro-
posed I–III phase line at high T. The dotted gray line shows a
schematic location of the I–I’ phase line inferred in Refs. 19–23.
The melting measurements are from Refs. 24–29: thick gray line
(Ref. 24), open circles (Ref. 25), crosses (Ref. 27), vertical gray
bars (Ref. 28), open squares (Ref. 26), dashed line (Ref. 29). Stars
correspond to the III–IV transition [11] (see also Ref. 30). Open
triangles and gray dashed-dotted lines (from DFT and QMC calcu-
lations) are theoretical results for a liquid-liquid transition [31,32]
associated with the molecular dissociation. Thick dotted gray and
blue lines are suggested I–IV and IV–liquid lines, respectively.
Hydrogen at extreme pressures
Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5 525
3. Melting and fluid behavior
Determination of the melting line of hydrogen, espe-
cially in the limit of high pressures, remains one of the
most challenging experimental tasks. Theoretical two-
phase simulations up to 200 GPa suggest that there is a
decline in the melting temperature above 90 GPa related to
softening of the intermolecular interactions, which occur at
a faster rate in the liquid than in the solid as a function of
pressure [29]. First principles calculations performed on
this and other works also suggest the presence of another
high-temperature boundary above the melt line related to
the molecular dissociation. This transformation is often
called the plasma line but can be also considered as a first-
order liquid-liquid transition [32,50–53]. Extrapolations of
the melt line and the liquid–liquid phase transition [29]
determined in theoretical calculations suggest the presence
of a triple point at 300 GPa and 400 K. Above this pres-
sure, the solid is expected to melt into a metallic liquid.
Two major experimental techniques have been used to
detect melting: visual observations, which include detec-
tion of the laser speckle pattern [24,27,28], and Raman
spectroscopy measurements [25,26]. Generally, the results
of visual observations should be considered quite reliable
at relatively low pressures as the optical contrast between
solid and fluid is sufficiently large due to the difference in
the refractive indices. The results of two available experi-
mental studies [24,54] are in agreement within the P–T
range of overlap. The study by Datchi et al. [24] extended
the melting line up 15.2 GPa and 530 K, but experienced
difficulties in reaching more extreme conditions because
the metallic gasket materials used could not contain the
hydrogen sample. These visual observation experiments
required substantial time as very slow temperature change
is required to stabilize fluid and solid materials in equili-
brium. Gregoryanz et al. [25] used cubic boron nitride and
alumina insets in rhenium gaskets and employed express
Raman observations to detect melting. At melting, they
observed a small Raman vibron discontinuity up to 44
GPa, but no further discontinuities have been detected
above this pressure. They also reported a large increase in
the negative temperature shift of the Raman vibron with
pressure. Combined melting temperature data to 44 GPa
obtained in resistive heating experiments [24,25,54] sug-
gest a possible melting line maximum near 100 GPa and
1000 K in qualitative agreement with the theoretical calcu-
lations of Ref. 29.
Experiments on the melting of hydrogen to higher pres-
sures have been performed using laser heating techniques
[26–28] including pulsed laser heating. The results of these
very challenging experiments remain largely controversial,
as there are a number of inconsistent observations. In par-
ticular the results of Deemyad, and Silvera [27], which uti-
lized visual observations, are standing alone, as they suggest
a very narrow maximum at the melting line, inconsistent
with the theoretical predictions and the results of other mea-
surements. Notably, Deemyad, and Silvera have reported
four pressure points obtained in one single experimental
run; they have not been able to provide any experimental
evidence of presence of hydrogen in the high-pressure cav-
ity after the initial laser heating experiments. The results of
this study were not reproduced in subsequent investiga-
tions [26,28], which presents results of multiple loads, and
clear Raman evidence of hydrogen present in the sample
cavity. Both studies [26,28] suggest that the melt line has a
broad maximum near 100 GPa, in a qualitative agreement
with the theoretical calculations of Ref. 29. However, the
diagnostics of melting in Refs. 26, 28 is somewhat contro-
versial. Eremets and Trojan [28] report changes in the laser
speckle pattern and a large reversible drop in resistivity of
a Pt foil which probe the sample cavity. These observations
may be related to melting but could, in principle, be due to
chemical reactions, or other phenomena unrelated to melt-
ing. A drop in the resistance of the Pt foil, claimed by Ere-
mets and Trojan to be an indication of melting, was pro-
posed by them to be due to a shunting by conducting fluid
hydrogen. Instead, we suggest that the thermal flux, out of
the laser heated Pt foil, increases rapidly through the convec-
tion in molten hydrogen, causing the foil to drop the temper-
ature, and hence the electrical resistance. Subramanian et
al. [26] reported on a large discontinuity of the Raman vi-
bron at melting and attributed this to a change in chemical
bonding in fluid hydrogen. However, this observation see-
mingly contradicts Raman measurements in resistively
heated DACs, where a very small, or even no discontinuity
was observed [25]. The reason for such discrepancy may be
due to difficulties of containing, and hence measuring Ra-
man spectra of fluid hydrogen in resistively heated DACs.
Alternatively very large temperature gradients across the
sample can give rise to bimodal Raman spectra observed in
the laser heating experiments [26] as the Raman vibron
shows a very steep temperature dependence. The available
experimental melting data of hydrogen provide definitive
prove of a maximum in the melting line.
Conventionally, it is assumed that fluid hydrogen is mo-
lecular at moderate pressures below the triple point with
solid and dissociated fluid > 200 GPa, < 1000 K. Raman
measurements of fluid hydrogen [26,55] however show a
continuous change with pressure in intramolecular bonding
in the fluid state. Goncharov and Crowhurst [55] also
found a large increase in the vibron bandwidth accompa-
nied by a decreased vibron anharmonicity deduced from
the spacings between excited vibrational states. Subrama-
nian et al. [26] show that the roton modes essentially dis-
appear in the fluid state above 30 GPa. These observations
can be understood due to the drastic decrease in lifetime of
molecular states in fluid hydrogen with pressure. The life-
time of the molecular states become comparable with the
vibrational period, but are shorter than the rotational pe-
riod, making the latter unobservable.
Alexander F. Goncharov, Ross T. Howie, and Eugene Gregoryanz
526 Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5
Until recently, experimental observations of conduct-
ing states in dense hydrogen could only be performed in
shock wave experiments [56–59] and static DAC experi-
ments on hydrogen exceeding temperatures of 3000 K
were inaccessible. Recently, Goncharov et al. [60] devel-
oped a new optical spectroscopy technique in pulsed laser
heated DAC which allow to measure optical spectra in the
visible spectral range. The sample is heated by 1–5 s
pulses of electrically modulated Yb fiber laser at 1070 nm.
The optical spectra are measured using a supercontinuum
generated in a photonic crystal fiber and are recorded as a
function of time using a streak camera in a single two-
dimensional CCD image along with the radiation spectra to
measure the temperature spectroradiometrically. Such
technique has opened a window of opportunity to probe
hot dense hydrogen at P–T conditions thought to be una-
chievable through static compressions.
4. Phase II
The transition to phase II has been originally described as
the one from spherically symmetric rotational states of pure
para H2 or ortho D2 to a broken symmetry phase in which
these symmetric states deform and material transforms an
orientationally ordered state [34]. It has been shown that
mixed ortho-para materials (for example with a normal
composition corresponding to the high-T limit [61]) also
transform to phase II (which reveals different rotational dy-
namics [37] and perhaps even a different crystal symmetry)
at lower pressures. A very large isotope effect has been ob-
served for the transition to phase II [34,62,63]. The large
isotope effect on the transition pressure to BSP phase sug-
gests that the transition is related to ordering of the quantum
rotational degrees of freedom [18,49] as the rotational con-
stants
2/4 ,B h cI where I is the rotational moment of
inertia, governing the rotational energies are very different
for H2 and D2. On the microscopic level, at the entry to
phase II, free molecular rotations are expected to transform
to wide-angle librations for some of the rotational coordi-
nates, which can be largely incoherent [39]. The first-
principles path-integral molecular dynamic calculations re-
vealed the quantum character of these molecular motions,
however, these experience a ―quantum localization‖ (or
―quantum confinement‖) as molecular rotations become
hindered in some rotation directions [38]. In contrast, recent
ab initio path integral molecular dynamics (PIMD) of Li et
al. [49] do not support the ―quantum confinement‖ and in-
stead suggest that the transition is governed by a competition
between anisotropic inter-molecular interactions, and the
thermal and quantum nuclear fluctuations.
Raman spectra of phase II reveal a combination of free
molecular rotation excitations and libron like vibrations
characteristic of the orientationally ordered molecules [35].
Raman and IR spectra of vibron modes have been used to
map the II–I phase line. Below approximately 140 GPa, the
transition can be traced by observing a small vibron dis-
continuity [16,18,19,34,37]. Above 140 GPa, the vibron
frequency has a strong temperature dependence in phase II
prior to the transition to phase I [17,33], suggesting that the
orientational ordering develops gradually with pressure
within phase II.
The determination of the structure of orientationally or-
dered hydrogen phases is a very challenging topic. Theo-
retical structure search is difficult because phase II retains
a large amount of orientational disorder. Thus, a single
theoretical approach (e.g., density functional theory, DFT)
does not work well. Recently, Li et al. [49] suggested us-
ing PIMD technique for the most stable static molecular
configuration to account for quantum nuclear motion at
finite temperatures. However, the validity of these results
needs to be verified against the experimental observations.
The experimental data are also very limited [46–48,64].
Normally, only 1 or 2 of the strongest reflections originat-
ing from 100 and 101 major peaks of hcp phase I of hy-
drogen could be observed. However, Goncharenko and
Loubeyre [47] additionally reported one extra reflection
observed in single crystal x-ray and neutron diffraction of
D2. They interpreted this as due to an incommensurate
long-range order. In contrast, a Raman study [37] sug-
gested 3x5 Brillouin zone folding. Moreover, the modula-
tion appears at a lower pressure than that reported for the
I–II transition in Raman measurements [37].
5. Phase III
Phase III has been discovered in Raman observations at
77 K: the Raman vibron revealed an astonishing 100 cm
–1
discontinuity at 155 GPa, and observations showed a two-
phase coexistence in the pressure range of about 20 GPa,
which is characteristic of the first-order transition [65].
Subsequent infrared absorption (IR) measurements showed
a two order of magnitude increase in the vibron mode ac-
tivity in phase III [36,66–68]. These observations initiated
a number of suggestions about a new chemical bonding
type in phase III related to a large intermolecular charge
transfer [69]. However, direct reflectivity measurements
[68] showed that the dipole moment associated to the IR
vibron is very small (0.04e at 210 GPa), so the charge
transfer may be of dynamic nature and be restricted within
the molecule. However, density functional theory does
predict a small structural distortion of the parent hexagonal
closed-packed lattice of phase I [39,44].
For a long time vibrational spectroscopy served as the
sole source of information on properties of phase III. Ra-
man spectroscopy measurements of phase III revealed a
number of observations, which shed light on the structural
and dynamical properties of phase III. In addition to the
vibron discontinuity, the II–III transition is characterized
by a total alteration of the low-frequency spectra: the roton
spectra (or their remnants) disappear and a number of new
Hydrogen at extreme pressures
Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5 527
Fig. 2. (Color online) Raman spectra of hydrogen through transi-
tions to phases II and III [35].
peaks appear at the transition to phase III (Fig. 2). These
show a very strong pressure dependence, which identify
them as the lattice modes (translational and librational)
unlike the rotational modes (rotons) in phases I and II
which are very weakly pressure dependent [34,70]. The
frequencies of the Raman modes increase strongly with
pressure and the modes become sharper (Fig. 2) [35]. Ra-
man and IR spectra of phase III are also strongly tempera-
ture dependent. The Raman and IR vibron frequencies in-
crease with temperature continuously in a wide
temperature range which was determined in quasi-isobaric
experimental scans [17–19]. There is a discontinuity in the
vibron frequency at the II–III and I–III transitions , which
quickly decreases with pressure and was reported to disap-
pear above 235 K (in D2) [37] even though two vibron
peaks were observed near the transition. This was inter-
preted as a (tri)critical point, where either the transition
becomes second order or terminates, so there is no distinc-
tion between phases I and III at higher pressures (and tem-
peratures). The IR intensity was also found to decrease in
intensity in the temperature runs [18,33] similar to that of
the Raman and IR frequencies. This was described by a
Maier–Saupe model [71], which characterizes the orienta-
tional ordering of classical rotors and initially was derived
for liquid crystals. Within this model, the IR frequency and
intensity and Raman frequency of the vibron can be treated
as scalar order parameters characterizing the orientational
ordering in phase III [18,33]. The conclusion about the
nature of orientational ordering in low-temperature phase
III is also supported by a relatively weak isotope effect (cf.
transition pressures of transitions to phase II for H2 and
D2), the insensitivity of the transition pressure to the ortho-
para concentration [18,35] and the observation of the total
disappearance of the roton Raman bands (Fig. 2).
As in the case of phase II, the determination of the struc-
ture of orientationally ordered phase III of hydrogen is a
very challenging topic and the experimental data are very
limited [46]. Moreover, only 1 or 2 strongest reflections
originated from 100 and 101 major peaks of hcp phase I of
hydrogen could be observed. Recently, x-ray diffraction
studies have been performed in the P–T range of stability of
phase III (>155 GPa below 120 K) [46]. The results suggest
that an hcp lattice remains a structural basis of phase III.
Theoretical structural search for high-pressure phases of
hydrogen has a long history [39–44,72–74]. Here we brief-
ly review the most relevant works for the high-pressure
(>100 GPa) range, where the effects of quantum rotations
and ortho-para distinctions is substantially diminished. In
this regime the (DFT) should be well applicable. However,
these results should also be treated carefully as the quan-
tum effect related to large zero point energy make substan-
tial contributions into the free energy.
The results of an extensive theoretical DFT structural
search [40,42] suggested a monoclinic C2/c structure as the
primary candidate for phase III. A number of structures are
very competitive in enthalpy in the pressure range of inter-
est; the results depend on the level of DFT theory, form of
pseudopotentials used, and treatment of proton zero point
motion [40]. It is interesting that none of these structures
agree well with the x-ray diffraction data (Fig. 3), although
some level of agreement has been achieved with the Ra-
man and IR data [35,67,75], especially with the presence
of a strong IR vibron absorption mode. It is interesting that
hybrid DFT calculations [76] find the P63/m structure
Fig. 3. (Color online) X-ray diffraction of phase III of hydrogen.
Gray line: C2/c structure from Ref. 49 and pink line is an hcp of
molecular centers with the lattice parameters from the experimen-
tal study of Akahama et al. [46].
Alexander F. Goncharov, Ross T. Howie, and Eugene Gregoryanz
528 Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5
(which would yield the x-ray pattern that nicely agrees
with the x-ray experiment) the most stable, although the
previous study found that this structure is dynamically un-
stable above 120 GPa [40]. However, this structure seems
inconsistent with the IR observations. For the sake of com-
pleteness, we would like to mention that the Cmc21 struc-
ture proposed by Toledano et al. [23] based on group
theory is somewhat higher in DFT enthalpy, although Ra-
man and IR activity and x-ray diffraction patterns broadly
agree with the observations.
It is interesting that in spite of a large number of ener-
getically competing structures determined in theoretical
calculations, experimental observations show the stability
of only one classically oriented solid phase in a very broad
pressure-temperature range [12,77]. The pressure and tem-
perature dependencies of vibron and phonon frequencies
suggest that phase III becomes more stable at higher pres-
sures and lower temperatures. A rather strong softening of
molecular vibron Raman mode (above 35 GPa) has been
interpreted as a ―harbinger‖ of molecular dissociation, but
later it was understood (e.g., Ref. 78) that a substantial part
of this softening is coming from the increase of the intramo-
lecular coupling [79,80]. The IR vibron, which contains
much less contribution of this coupling starts softening only
above 120 GPa [79]. However, unlike the situation with the
classical soft modes related to the displacive phase transi-
tions, there is no acceleration of the softening with pressure,
making predictions of molecular dissociation with pressure
rather uncertain [75]. Extrapolation of the optical data sug-
gests that the optical closure in phase II should occur near
450 GPa [75,77]. The effect of temperature was recognized
to be very essential for metallization of hydrogen in static
high-pressure conditions [11,14].
6. Phase IV
Until 2011 only the high-pressure room-temperature
studies of hydrogen up to 180 Gpa [10] and to the claimed
340 GPa have been reported 81,82. The latter results are
very controversial mainly due to the fact that no positive
diagnostics of hydrogen was offered. In Fig. 4 we show the
compilation of the recently obtained Raman data on the
molecular vibron up to 320 GPa compared to that reported
previously by Ruoff [81]. The obvious conclusion is that
either the pressure metrology in these early experiments
was not reliable or other factors (e.g., lack of hydrogen in
the sample chamber) are responsible for apparent discre-
pancy with the current results. The diamond Raman edge is
the currently adopted method of pressure measurements in
ultra-high compression experiments. The Raman frequency
of the diamond edge (e.g., Ref. 15) has been calibrated
with respect to other sensors (mostly ruby) and is reliable
in situations when the experiments are performed in simi-
lar geometrical conditions. However the results of Ruoff
[81] obviously stand alone (Fig. 4) making the claim of
transparent hydrogen at 342 GPa in the subsequent paper
[82], which also does not present any positive diagnostics
of hydrogen, highly questionable.
Two independent experiments have recently succeeded
in reaching pressures in excess of 300 GPa at 300 K
[11,14]. Similar Raman observations have been reported
that show remarkable changes in Raman spectra above 200
GPa; firstly: the gradient of the vibron frequency versus
pressure slope changes dramatically and a broad low-
frequency peaks appear, and secondly: another system of
low-frequency high intensity peaks emerge and the vibron
splits in two. Eremets and Troyan [14] did not notice the
appearance of new low-frequency peaks and interpreted
this change as due to a transition to the Cmca-12 phase
[40]. They also reported a change in optical properties and
a total disappearance of Raman signal above 260–270 GPa,
which was suggested to be due to transformation to metal-
lic monatomic fluid.
On the contrary, Howie et al. [11] observed Raman sig-
nal to the highest pressure reached in the experiment — 320
GPa. They noticed the appearance of a second Raman vi-
bron with very different pressure behavior of both the fre-
quency and linewidth. Based on these observations and
theoretical predictions [40], they suggested a Pbcn structure
for phase IV of hydrogen. This structure matches much bet-
ter with the experimental observations, as the appearance of
two distinct vibron modes and a strong low-frequency libron
mode can be naturally explained based on the unique fea-
tures of phase IV. Indeed, Pbcn hydrogen consists of mole-
cular layers of two kinds: weakly bounded hexagonal, and
strongly bounded graphene-like [40], which differ by the
Fig. 4. (Color online) Raman vibron frequencies of hydrogen
though the transition to phases III and IV at 300 K [10,11,14,81].
Hydrogen at extreme pressures
Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5 529
intramolecular distances that are substantially larger in the
graphene-like layer. It is interesting that the hexagonal con-
figuration of molecules in the graphene-like layer is some-
what reminiscent to the prediction of LeSar and Herschbach
(Ref. 83, see also Ref. 84), who suggested that termolecular
complexes [(H2)3] could form before the transition to the
atomic phase. This structure has been further examined theo-
retically in a number of recent publications, which suggest
slightly different crystal symmetries [13,85] and fluxional
behavior of graphene-like layers [86] related to large atomic
tunneling quantum effects, and even suggest quantum liquid
behavior for these layers [87]. Experimental and theoretical
studies clearly indicate that phase IV is insulating or semi-
metallic as the optical spectra show the presence of the opti-
cal gap [11,30].
7. Conclusions
Key questions still remain about the higher pressure be-
havior. Predictions propose that phase IV will transform to
a metallic molecular phase with Cmca-4 structure above
360 GPa [86]. However, monatomic phases [88–90] may
compete at these compressions. We believe that experi-
mental static compression studies which will verify these
predictions are down the road [91]. Such studies will also
address the issue of the predicted ground state fluid atomic
metallic hydrogen [92–94]. The central problem is the
treatment of the quantum effects at such regimes, which
needs to be solved for such fundamentally important sys-
tem as the element number one.
Acknowledgment
A.F.G. acknowledges support from the NSF, Army Re-
search Office, NAI, and EFRee.
E.G. and R. T. H acknowledge support from the U.K.
Engineering and Physical Sciences Research Council and
Institute of the Shock Physics, Imperial College.
1. W.J. Nellis, Rep. Prog. Phys. 69, 1479 (2006).
2. A.F. Goncharov and R.J. Hemley, Chem. Soc. Rev. 35 (10),
899 (2006).
3. A.F. Goncharov and J. Crowhurst, Phase Transitions 80,
1051 (2007).
4. E.G. Maksimov and Y.I. Shilov, Usp. Fiz. Nauk. 169, 1223
(1999).
5. J.M. McMahon, M.A. Morales, C. Pierleoni, and D.M.
Ceperley, Rev. Mod. Phys. 84, 1607 (2012).
6. R. Jeanloz, P.M. Celliers, G.W. Collins, J.H. Eggert, K.K.M.
Lee, R.S. McWilliams, S. Brygoo, and P. Loubeyre, P. Natl.
Acad. Sci. USA 104, 9172 (2007).
7. R.F. Trunin, V.D. Urlin, and A.B. Medvedev, Phys. Usp. 53
577 (2010).
8. D.G. Hicks, T.R. Boehly, P.M. Celliers, J.H. Eggert, S.J.
Moon, D.D. Meyerhofer, and G.W. Collins, Phys. Rev. B,
79, 014112 (2009).
9. P. Loubeyre, S. Brygoo, J. Eggert, P.M. Celliers, D.K.
Spaulding, J.R. Rygg, T.R. Boehly, G.W. Collins, and R.
Jeanloz, Phys. Rev. B 86, 144115 (2012).
10. B.J. Baer, M.E. Chang, and W.J. Evans, J. Appl. Phys. 104
(2008).
11. R.T. Howie, C.L. Guillaume, T. Scheler, A.F. Goncharov,
and E. Gregoryanz, Phys. Rev. Lett. 108, 125501 (2012).
12. C.S. Zha, Z.X. Liu, and R.J. Hemley, Phys. Rev. Lett. 108
146402 (2012).
13. C.J. Pickard, M. Martinez-Canales, and R.J. Needs, Phys.
Rev. B 85, 214114 (2012).
14. M.I. Eremets and I.A. Troyan, Nat. Mater. 10, 927 (2011).
15. Y. Akahama and H. Kawamura, J. Appl. Phys. 100, 043516
(2006).
16. H.K. Mao and R.J. Hemley, Rev. Mod. Phys. 66, 671 (1994).
17. A.F. Goncharov, R.J. Hemley, and H.K. Mao, J. Chem.
Phys. 134, 174501 (2011).
18. I.I. Mazin, R.J. Hemley, A.F. Goncharov, M. Hanfland, and
H.K. Mao, Phys. Rev. Lett. 78, 1066 (1997).
19. A.F. Goncharov, I.I. Mazin, J.H. Eggert, R.J. Hemley, and
H.K. Mao, Phys. Rev. Lett. 75, 2514 (1995).
20. M.P. Surh, K.J. Runge, T.W. Barbee, E.L. Pollock, and C.
Mailhiot, Phys. Rev. B 55, 11330 (1997).
21. B.J. Baer, W.J. Evans, and C.S. Yoo, Phys. Rev. Lett. 98,
235503 (2007).
22. B.J. Baer, W.J. Evans and C.S. Yoo, Phys. Rev. Lett. 102,
235503 (2009).
23. P. Toledano, H. Katzke, A.F. Goncharov, and R.J. Hemley,
Phys. Rev. Lett. 103, 105301 (2009).
24. F. Datchi, P. Loubeyre, and R. LeToullec, Phys. Rev. B 61,
6535 (2000).
25. E. Gregoryanz, A.F. Goncharov, K. Matsuishi, H. Mao and
R.J. Hemley, Phys. Rev. Lett. 90, 175701 (2003).
26. N. Subramanian, A.F. Goncharov, V.V. Struzhkin, M.
Somayazulu, and R.J. Hemley, P. Natl. Acad. Sci. USA 108,
6014 (2011).
27. S. Deemyad and I.F. Silvera, Phys. Rev. Lett. 100, 155701
(2008).
28. M.I. Eremets and I.A. Trojan, J. Lett. 89, 174 (2009).
29. S.A. Bonev, E. Schwegler, T. Ogitsu, and G. Galli, Nature
431, 669 (2004).
30. R.T. Howie, T. Scheler, C.L. Guillaume, and E. Gregoryanz,
Phys. Rev. B 86, 214104 (2012).
31. I. Tamblyn and S.A. Bonev, Phys. Rev. Lett. 104, 065702
(2010).
32. M.A. Morales, C. Pierleoni, E. Schwegler, and D.M.
Ceperley, P. Natl. Acad. Sci. USA 107, 12799 (2010).
33. L.J. Cui, N.H. Chen, and I.F. Silvera, Phys. Rev. B 51, 14987
(1995).
34. I.F. Silvera and R.J. Wijngaarden, Phys. Rev. Lett. 47, 39
(1981).
35. A.F. Goncharov, R.J. Hemley, H.K. Mao, and J.F. Shu,
Phys. Rev. Lett. 80, 101 (1998).
36. M. Hanfland, R.J. Hemley, and H.K. Mao, Phys. Rev. Lett.
70, 3760 (1993).
Alexander F. Goncharov, Ross T. Howie, and Eugene Gregoryanz
530 Low Temperature Physics/Fizika Nizkikh Temperatur, 2013, v. 39, No. 5
37. A.F. Goncharov, J.H. Eggert, I.I. Mazin, R.J. Hemley, and
H.K. Mao, Phys. Rev. B 54, 15590 (1996).
38. H. Kitamura, S. Tsuneyuki, T. Ogitsu, and T. Miyake,
Nature 404, 259 (2000).
39. K.A. Johnson and N.W. Ashcroft, Nature 403, 632 (2000).
40. C.J. Pickard and R.J. Needs, Nat. Phys. 3, 473 (2007).
41. J.S. Tse and D.D. Klug, Nature 378, 595 (1995).
42. Rodgers, Solid State Commun. 145, 5 (2008).
43. J. Kohanoff, S. Scandolo, G.L. Chiarotti, and E. Tosatti,
Phys. Rev. Lett. 78, 2783 (1997).
44. J. Kohanoff, S. Scandolo, S. de Gironcoli, and E. Tosatti,
Phys. Rev. Lett. 83, 4097 (1999).
45. K.J. Runge, M.P. Surh, C. Mailhiot, and E.L. Pollock, Phys.
Rev. Lett. 69, 3527 (1992).
46. Y. Akahama, M. Nishimura, H. Kawamura, N. Hirao, Y.
Ohishi, and K. Takemura, Phys. Rev. B 82, 060101(R) (2010).
47. I. Goncharenko and P. Loubeyre, Nature 435, 1206 (2005).
48. H. Kawamura, Y. Akahama, S. Umemoto, K. Takemura, Y.
Ohishi, and O. Shimomura, J. Phys.: Condens. Matter. 14,
10407 (2002).
49. X.-Z. Li, B. Walker, M.I.J. Probert, C.J. Pickard, R.J. Needs,
and A. Michaelides, J. Phys.: Condens. Matter. 25, 085402
(2013).
50. W.R. Magro, D.M. Ceperley, C. Pierleoni, and B. Bernu,
Phys. Rev. Lett. 76, 1240 (1996).
51. D. Saumon and G. Chabrier, Phys. Rev. A 46, 2084 (1992).
52. S. Scandolo, P. Natl. Acad. Sci. USA 100, 3051 (2003).
53. B. Boates and S.A. Bonev, Phys. Rev. Lett. 102, 015701
(2009).
54. V. Diatschenko, C.W. Chu, D.H. Liebenberg, D.A. Young,
M. Ross, and R.L. Mills, Phys. Rev. B 32, 381 (1985).
55. A.F. Goncharov and J.C. Crowhurst, Phys. Rev. Lett. 96,
055504 (2006).
56. S.T. Weir, A.C. Mitchell, and W.J. Nellis, Phys. Rev. Lett.
76, 1860 (1996).
57. W.J. Nellis, S.T. Weir, and A.C. Mitchell, Phys. Rev. B 59,
3434 (1999).
58. P. Loubeyre, P.M. Celliers, D.G. Hicks, E. Henry, A. Dewaele,
J. Pasley, J. Eggert, M. Koenig, F. Occelli, K.M. Lee, R.
Jeanloz, D. Neely, A. Benuzzi-Mounaix, D. Bradley, M. Bastea,
S. Moon, and G.W. Collins, High Pressure Res 24, 25 (2004).
59. P.M. Celliers, G.W. Collins, L.B. Da Silva, D.M. Gold, R.
Cauble, R.J. Wallace, M.E. Foord, and B.A. Hammel, Phys.
Rev. Lett. 84, 5564 (2000).
60. A.F. Goncharov, D.A. Dalton, R.S. McWilliams, and
M.F. Mahmood, Mater. Res. Soc.
Symp.DOI:10.1557/opl.2012.1560 (2012) Proc. 1405
DOI:10.1557/opl.2012.1560 (2012).
61. I.F. Silvera, Rev. Mod. Phys. 52, 393 (1980).
62. H.E. Lorenzana, I.F. Silvera, and K.A. Goettel, Phys. Rev.
Lett. 64, 1939 (1990).
63. F. Moshary, N.H. Chen, and I.F. Silvera, Phys. Rev. Lett. 71,
3814 (1993).
64. H. Kawamura, Y. Akahama, S. Umemoto, K. Takemura, Y.
Ohishi, and O. Shimomura, Solid State Commun. 119, 29
(2001).
65. R.J. Hemley and H.K. Mao, Phys. Rev. Lett. 61, 857 (1988).
66. L.J. Cui, N.H. Chen, and I.F. Silvera, Phys. Rev. Lett. 74,
4011 (1995).
67. N.H. Chen, E. Sterer, and I.F. Silvera, Phys. Rev. Lett. 76,
1663 (1996).
68. R.J. Hemley, I.I. Mazin, A.F. Goncharov, and H.K. Mao,
Europhys. Lett. 37, 403 (1997).
69. R.J. Hemley, Z.G. Soos, M. Hanfland, and H.K. Mao,
Nature 369, 384 (1994).
70. A.F. Goncharov, M.A. Strzhemechny, H.K. Mao, and R.J.
Hemley, Phys. Rev. B 63, 064304 (2001).
71. M. Plischke and B. Bergersen, in Equilibrium Statistical
Physics, Prentice Hall, Englewood Cliffs, NJ, (1989), p. 74.
72. C.J. Pickard and R.J. Needs, Phys. Rev. Lett. 102 (2009).
73. E. Kaxiras and J. Broughton, Europhys. Lett. 17, 151 (1992).
74. E. Kaxiras, J. Broughton, and R.J. Hemley, Phys. Rev. Lett.
67, 1138 (1991).
75. A.F. Goncharov, E. Gregoryanz, R.J. Hemley, and H.K.
Mao, P. Natl. Acad. Sci. USA 98, 14234 (2001).
76. S. Azadi and T.D. Kuhne, Pis'ma v ZhETF 95, 509 (2012).
77. P. Loubeyre, F. Occelli, and R. LeToullec, Nature 416, 613
(2002).
78. N.W. Ashcroft, Phys. Rev. B 41, 10963 (1990).
79. M. Hanfland, R.J. Hemley, H.K. Mao, and G.P. Williams,
Phys. Rev. Lett. 69, 1129 (1992).
80. F. Moshary, N.H. Chen, and I.F. Silvera, Phys. Rev. B 48,
12613 (1993).
81. A.L. Ruoff, in: High Pressure Science and Technology, W.
Trzeciakowski (ed.), World Scientific, Singapore, (1996).
82. C. Narayana, H. Luo, J. Orloff, and A.L. Ruoff, Nature 393,
46 (1998).
83. R. Lesar and D.R. Herschbach, J. Phys. Chem. 85, 3787 (1981).
84. V. Labet, R. Hoffmann, and N.W. Ashcroft, J. Chem. Phys.
136, 074502 (2012).
85. H.Y. Liu, L. Zhu, W.W. Cui, and Y.M. Ma, J. Chem. Phys.
137, 074501 (2012).
86. A.F. Goncharov, J.S. Tse, H. Wang, J.H. Yang, V.V.
Struzhkin, R.T. Howie, and E. Gregoryanz, Phys. Rev. B 87,
024101 (2013).
87. H. Liu and Y. Ma, Phys. Rev. Lett. 110, 025903 (2013).
88. J.M. McMahon and D.M. Ceperley, Phys. Rev. Lett. 106,
165302 (2011).
89. V. Labet, R. Hoffmann, and N.W. Ashcroft, J. Chem. Phys.
136, 074504 (2012).
90. H.Y. Liu, H. Wang, and Y.M. Ma, J. Phys. Chem. C 116,
9221 (2012).
91. L. Dubrovinsky, N. Dubrovinskaia, V.B. Prakapenka, and
A.M. Abakumov, Nat. Commun. 3, 1163 (2012).
92. E. Babaev, A. Sudbo, and N.W. Ashcroft, Nature 431, 666
(2004).
93. E. Babaev, A. Sudbo, and N.W. Ashcroft, Phys. Rev. Lett.
95, 105301 (2005).
94. J. Chen, X.-Z. Li, Q. Zhang, M.I.J. Probert, C.J. Pickard, R.J.
Needs, A. Michaelides, and E. Wang, arXiv:1212.4554v1
[cond-mat.mtrl-sci] (2012).
doi:10.1557/opl.2012.1560%20(2012)
|