Anomalous hydrogen dynamics of the ice VII–VIII transition revealed by high-pressure neutron diffraction

Abstract

Above 2 GPa the phase diagram of water simplifies considerably and exhibits only two solid phases up to 60 GPa, ice VII and ice VIII. The two phases are related to each other by hydrogen ordering, with the oxygen sublattice being essentially the same. Here we present neutron diffraction data to 15 GPa which reveal that the rate of hydrogen ordering at the ice VII–VIII transition decreases strongly with pressure to reach timescales of minutes at 10 GPa. Surprisingly, the ordering process becomes more rapid again upon further compression. We show that such an unusual change in transition rate can be explained by a slowing down of the rotational dynamics of water molecules with a simultaneous increase of translational motion of hydrogen under pressure, as previously suspected. The observed cross-over in the hydrogen dynamics in ice is likely the origin of various hitherto unexplained anomalies of ice VII in the 10–15 GPa range reported by Raman spectroscopy, X-ray diffraction, and proton conductivity.

Some Aspects of the Liquid Water Thermodynamic Behavior: From The Stable to the Deep Supercooled Regime

“…The best known of these is the density maximum ( ) at 277 K [ 2 ]. Another surprising feature is the polymorphism that is evident in the crystalline [ 3 , 4 , 5 ], the glass phases [ 6 , 7 , 8 , 9 ], and especially for the liquid phase [ 10 ]; these too accompanied by uncovered intriguing phenomena occurring upon changes in the thermodynamic conditions, in particular, pressure.…”

High-temperature and high-pressure plastic phase of ice at the boundary of liquid water and ice VII

“…An interesting deduction can be made from the above observations in which it can be hypothesized that the relaxation time for the rotational diffusion does not change in the plastic crystal phase (similar to the condition for 4 normalkm normals−1 shock-compressed water) but it increases (relative to liquid water) in the proper ice VII phase observed at 300 K and 10 GPa. This observation can also be associated with the anomalous behaviour of ice VII at around 10 GPa reported in previous literature [33,35,60]. A word of caution should be noted here, in that the temporal values reported have been taken such that the value of Iω2 is less than 10−9 (because of numerical tolerance), whereas ideally the temporal values should report the time at which Iω2=0.…”

“…Bulk liquid water contains water molecules in a completely disordered orientation, whereas ice VII consists of a periodic arrangement of water molecules. The structures of ice VII at different temperature and pressure ranges have also been probed in literature using NMR and XRD/Rietveld studies [30][31][32][33][34]. Molecular dynamic studies demonstrating formation of this eightcoordinated ice VII structure [14] also demonstrate a close match with experimentally observed XRD peaks and/or structure factors.…”

Dynamic phase transition theory

“…Palmer called such a region the "component" [3], and we note that conventional equilibrium statistics are exactly designed to handle such intracomponent motions. However, the temporal DOF becomes nontrivial when T is moderate: there are rare but significant inter-component motions for particles moving farther away beyond thermal fluctuations [18][19][20], which contributes to transport properties such as the diffusion coefficients and ionic mobility [21]. Instead of treating those motions as typical nonequilibrium phenomena, we found the dynamics for inter-component motions gradually become homogeneous towards the long time limit [17].…”