High pressure ices

Abstract

H 2 O will be more resistant to metallization than previously thought. From computational evolutionary structure searches, we find a sequence of new stable and meta-stable structures for the ground state of ice in the 1-5 TPa (10 to 50 Mbar) regime, in the static approximation. The previously proposed Pbcm structure is superseded by a Pmc2 1 phase at p ¼ 930 GPa, followed by a predicted transition to a P2 1 crystal structure at p ¼ 1.3 TPa. This phase, featuring higher coordination at O and H, is stable over a wide pressure range, reaching 4.8 TPa. We analyze carefully the geometrical changes in the calculated structures, especially the buckling at the H in O-H-O motifs. All structures are insulatingchemistry burns a deep and (with pressure increase) lasting hole in the density of states near the highest occupied electronic levels of what might be component metallic lattices. Metallization of ice in our calculations occurs only near 4.8 TPa, where the metallic C2∕m phase becomes most stable. In this regime, zero-point energies much larger than typical enthalpy differences suggest possible melting of the H sublattice, or even the entire crystal.hydrogen bonds | compressed water

The accurate calculation of the band gap of liquid water by means of GW corrections applied to plane-wave density functional theory molecular dynamics simulations

“…The structure at 5 K was determined by Leadbetter and co-workers using neutron diffraction (Table 3) 69 and investigated theoretically by several groups using different approaches. 11,26,27,32,41,42 We performed full structural optimization of this ice phase using the van der Waals density functional optB88-vdW. The results are shown in Table 3.…”

“…49,50,52,53,70 For example, Hermann and co-workers performed a non-self-consistent G0W0 correction for ice. 31,32 The requirements for computational capability limit us to the Ice-XI phase. Fig.…”

Isotopic differentiation and sublattice melting in dense dynamic ice

“…1 Recent studies on stable high pressure phases of ice have uncovered a variety of new phases that become stable at pressures beyond P = 1 TPa (1000 GPa, or 10 Mbar). [2][3][4][5][6] There is now an understanding that the electronic band gap of ice is not likely to close until pressures reach around P = 5 TPa (corresponding to a compression of roughly V 0 /V = 11.5), where metallic ice phases actually first become more stable than all known insulating phases. The stable ice phases at these pressures are notably more complex than, e.g., the highly symmetric phase ice X, which becomes stable around P = 100 GPa and which features symmetric and linear O-H-O bridging bonds.…”

“…1 Hydrogen (atomic mass 1) and oxygen (mass 16) are both comparatively light elements, and the zero point energies (ZPEs) at megabar pressures are exceedingly large, on the order of 1-1.5 eV per H 2 O unit in the harmonic approximation. 6 The question thus arises as to whether the associated motions in the ground state of ice under these conditions could lead to a diffusional, a superionic, or even a fully fluid phase. These states of matter of ice are of great astrophysical interest because of the presence of H 2 O, one of the thermodynamic sinks of all worlds, 7 in the interior of giant gas planets.…”

“…We start by using the structures presented in our recent work 6 to study the changes in the phase boundaries that might occur if all H atoms were to be replaced by D or T atoms. To do this, we need to estimate the zero point vibrational energies and corresponding average displacements associated with each structure over the pressure range of interest.…”